Cap non-reduction unrolling at 4, updated dependency versions.

Author: chriselrod

Manifest.toml

@@ -61,15 +61,15 @@ uuid = "9a3f8284-a2c9-5f02-9a11-845980a1fd5c"
 
 [[SIMDPirates]]
 deps = ["MacroTools", "VectorizationBase"]
-git-tree-sha1 = "6c6a77a41c846c08c61a0e556183a9c33b53e3d1"
+git-tree-sha1 = "c0f42ddb2645c54b8620979df5dc979c4742db59"
 uuid = "21efa798-c60a-11e8-04d3-e1a92915a26a"
-version = "0.1.3"
+version = "0.1.4"
 
 [[SLEEFPirates]]
 deps = ["SIMDPirates", "VectorizationBase"]
-git-tree-sha1 = "1c5b6827da87a12bdb7a4c893f44c3adbce3389d"
+git-tree-sha1 = "547bcf7d30967d87d4c62b3fe5efdb0e57a6e436"
 uuid = "476501e8-09a2-5ece-8869-fb82de89a1fa"
-version = "0.1.1"
+version = "0.1.2"
 
 [[Serialization]]
 uuid = "9e88b42a-f829-5b0c-bbe9-9e923198166b"
@@ -83,6 +83,6 @@ uuid = "8dfed614-e22c-5e08-85e1-65c5234f0b40"
 
 [[VectorizationBase]]
 deps = ["CpuId", "LinearAlgebra"]
-git-tree-sha1 = "54f5ba672c7d684fb0312825721368e22354ecd5"
+git-tree-sha1 = "bb905673925d36d1bf4693a114687d56723d5991"
 uuid = "3d5dd08c-fd9d-11e8-17fa-ed2836048c2f"
-version = "0.1.5"
+version = "0.1.6"

Project.toml

@@ -14,9 +14,9 @@ VectorizationBase = "3d5dd08c-fd9d-11e8-17fa-ed2836048c2f"
 [compat]
 MacroTools = "0.5"
 Parameters = "0.12.0"
-SIMDPirates = "0.1.3"
-SLEEFPirates = "0.1.1"
-VectorizationBase = "0.1.5"
+SIMDPirates = "0.1.4"
+SLEEFPirates = "0.1.2"
+VectorizationBase = "0.1.6"
 julia = "1.3.0"
 
 [extras]

benchmark/benchmarks.jl

@@ -1,11 +1,24 @@
 
 using BenchmarkTools
 
-suite = BenchmarkGroup()
-suite["linalg"] = BenchmarkGroup(["matmul","dot"])
+const SUITE = BenchmarkGroup()
+SUITE["linalg"] = BenchmarkGroup(["matmul","dot"])
 
 include(joinpath(@__DIR__, "looptests.jl"))
 
-for n ∈ 1:256
-
+SUITE["linalg"]["matmul"] = BenchmarkGroup()
+SUITE["linalg"]["dot"] = BenchmarkGroup()
+for n ∈ 1:64
+    A = rand(n,n);
+    A′ = copy(A');
+    B = rand(n,n);
+    C = Matrix{Float64}(undef, n, n);
+    SUITE["linalg"]["matmul"]["AmulB", n] = @benchmarkable gemmavx!($C, $A, $B)
+    SUITE["linalg"]["matmul"]["A′mulB", n] = @benchmarkable jAtmulBavx!($C, $A′, $B)
+    x = rand(n); y = rand(n);
+    SUITE["linalg"]["dot"]["dot", n] = @benchmarkable jdotavx($x, $y)
+    SUITE["linalg"]["dot"]["selfdot", n] = @benchmarkable jselfdotavx($x)
+    SUITE["linalg"]["dot"]["dot3", n] = @benchmarkable jdot3avx($x, $A, $y)
 end
+
+

src/determinestrategy.jl

@@ -125,7 +125,7 @@ function unroll_no_reductions(ls, order, vectorized, Wshift, size_T)
     end
     # heuristic guess
     # @show compute_rt, load_rt
-    min(2, round(Int, (compute_rt + load_rt + 1) / compute_rt))
+    min(4, round(Int, (compute_rt + load_rt + 1) / compute_rt))
 end
 function determine_unroll_factor(
     ls::LoopSet, order::Vector{Symbol}, unrolled::Symbol, vectorized::Symbol = first(order)

test/runtests.jl

@@ -483,6 +483,12 @@ end
         fill!(c2, 99999.9);
         @avx @. c2 = a + bl;
         @test c1 ≈ c2
+        br = reshape(b, (1,100,100));
+        bl = LowDimArray{(false,true,true)}(br);
+        @. c1 = a + br;
+        fill!(c2, 99999.9);
+        @avx @. c2 = a + bl;
+        @test c1 ≈ c2
 
         a = rand(T, M); B = rand(T, M, N); c = rand(T, N); c′ = c';
         d1 =      @. a + B * c′;