Support -mnop-mcount and -mrecord-mcount on x86_64 elf targets.

clang/lib/Driver/ToolChains/Clang.cpp

@@ -6884,14 +6884,16 @@ void Clang::ConstructJob(Compilation &C, const JobAction &JA,
             << A->getAsString(Args) << TripleStr;
     }
     if (Arg *A = Args.getLastArg(options::OPT_mnop_mcount)) {
-      if (Arch == llvm::Triple::systemz)
+      if (Arch == llvm::Triple::systemz ||
+          (TC.getTriple().isX86() && TC.getTriple().isOSBinFormatELF()))
         A->render(Args, CmdArgs);
       else
         D.Diag(diag::err_drv_unsupported_opt_for_target)
             << A->getAsString(Args) << TripleStr;
     }
     if (Arg *A = Args.getLastArg(options::OPT_mrecord_mcount)) {
-      if (Arch == llvm::Triple::systemz)
+      if (Arch == llvm::Triple::systemz ||
+          (TC.getTriple().isX86() && TC.getTriple().isOSBinFormatELF()))
         A->render(Args, CmdArgs);
       else
         D.Diag(diag::err_drv_unsupported_opt_for_target)

clang/test/CodeGen/mnop-mcount.c

@@ -6,6 +6,22 @@
 // RUN:   %s 2>&1 | FileCheck -check-prefix=NOPG %s
 // RUN: %clang_cc1 -mnop-mcount -triple s390x-ibm-linux -emit-llvm -o - %s \
 // RUN:   2>&1 | FileCheck -check-prefix=NOPG %s
+// RUN: %clang_cc1 -pg -mfentry -mnop-mcount -triple x86_64-unknown-linux -emit-llvm \
+// RUN:   -o - %s 2>&1 | FileCheck  %s
+// RUN: not %clang_cc1 -pg -mnop-mcount -triple x86_64-unknown-linux -emit-llvm -o - \
+// RUN:   %s 2>&1 | FileCheck -check-prefix=NOMFENTRY %s
+// RUN: %clang_cc1 -mfentry -mnop-mcount -triple x86_64-unknown-linux -emit-llvm -o - \
+// RUN:   %s 2>&1 | FileCheck -check-prefix=NOPG %s
+// RUN: %clang_cc1 -mnop-mcount -triple x86_64-unknown-linux -emit-llvm -o - %s \
+// RUN:   2>&1 | FileCheck -check-prefix=NOPG %s
+// RUN: %clang_cc1 -pg -mfentry -mnop-mcount -triple i686-unknown-linux -emit-llvm \
+// RUN:   -o - %s 2>&1 | FileCheck  %s
+// RUN: not %clang_cc1 -pg -mnop-mcount -triple i686-unknown-linux -emit-llvm -o - \
+// RUN:   %s 2>&1 | FileCheck -check-prefix=NOMFENTRY %s
+// RUN: %clang_cc1 -mfentry -mnop-mcount -triple i686-unknown-linux -emit-llvm -o - \
+// RUN:   %s 2>&1 | FileCheck -check-prefix=NOPG %s
+// RUN: %clang_cc1 -mnop-mcount -triple i686-unknown-linux -emit-llvm -o - %s \
+// RUN:   2>&1 | FileCheck -check-prefix=NOPG %s
 
 int foo(void) {
   return 0;

clang/test/CodeGen/mrecord-mcount.c

@@ -6,6 +6,22 @@
 // RUN:   %s 2>&1 | FileCheck -check-prefix=NOPG %s
 // RUN: %clang_cc1 -mrecord-mcount -triple s390x-ibm-linux -emit-llvm -o - %s \
 // RUN:   2>&1 | FileCheck -check-prefix=NOPG %s
+// RUN: %clang_cc1 -pg -mfentry -mrecord-mcount -triple x86_64-unknown-linux -emit-llvm \
+// RUN:   -o - %s 2>&1 | FileCheck  %s
+// RUN: not %clang_cc1 -pg -mrecord-mcount -triple x86_64-unknown-linux -emit-llvm -o - \
+// RUN:   %s 2>&1 | FileCheck -check-prefix=NOMFENTRY %s
+// RUN: %clang_cc1 -mfentry -mrecord-mcount -triple x86_64-unknown-linux -emit-llvm -o - \
+// RUN:   %s 2>&1 | FileCheck -check-prefix=NOPG %s
+// RUN: %clang_cc1 -mrecord-mcount -triple x86_64-unknown-linux -emit-llvm -o - %s \
+// RUN:   2>&1 | FileCheck -check-prefix=NOPG %s
+// RUN: %clang_cc1 -pg -mfentry -mrecord-mcount -triple i686-unknown-linux -emit-llvm \
+// RUN:   -o - %s 2>&1 | FileCheck  %s
+// RUN: not %clang_cc1 -pg -mrecord-mcount -triple i686-unknown-linux -emit-llvm -o - \
+// RUN:   %s 2>&1 | FileCheck -check-prefix=NOMFENTRY %s
+// RUN: %clang_cc1 -mfentry -mrecord-mcount -triple i686-unknown-linux -emit-llvm -o - \
+// RUN:   %s 2>&1 | FileCheck -check-prefix=NOPG %s
+// RUN: %clang_cc1 -mrecord-mcount -triple i686-unknown-linux -emit-llvm -o - %s \
+// RUN:   2>&1 | FileCheck -check-prefix=NOPG %s
 
 int foo(void) {
   return 0;

clang/test/Driver/mcount.c

@@ -1,12 +1,11 @@
 // RUN: %clang --target=s390x -c -### %s -mnop-mcount -mrecord-mcount 2>&1 | FileCheck %s
+// RUN: %clang --target=x86_64 -c -### %s -mnop-mcount -mrecord-mcount 2>&1 | FileCheck %s
+// RUN: %clang --target=i686 -c -### %s -mnop-mcount -mrecord-mcount 2>&1 | FileCheck %s
 
 // CHECK: "-mnop-mcount"
 // CHECK: "-mrecord-mcount"
 
-// RUN: not %clang --target=x86_64 -c -### %s -mnop-mcount -mrecord-mcount 2>&1 | FileCheck --check-prefix=ERR1 %s
 // RUN: not %clang --target=aarch64 -c -### %s -mnop-mcount -mrecord-mcount 2>&1 | FileCheck --check-prefix=ERR2 %s
 
-// ERR1: error: unsupported option '-mnop-mcount' for target 'x86_64'
-// ERR1: error: unsupported option '-mrecord-mcount' for target 'x86_64'
 // ERR2: error: unsupported option '-mnop-mcount' for target 'aarch64'
 // ERR2: error: unsupported option '-mrecord-mcount' for target 'aarch64'

llvm/lib/Target/X86/X86ISelDAGToDAG.cpp

@@ -179,11 +179,19 @@ namespace {
     bool runOnMachineFunction(MachineFunction &MF) override {
       // Reset the subtarget each time through.
       Subtarget = &MF.getSubtarget<X86Subtarget>();
-      IndirectTlsSegRefs = MF.getFunction().hasFnAttribute(
-                             "indirect-tls-seg-refs");
+      const Function &F = MF.getFunction();
+
+      IndirectTlsSegRefs = F.hasFnAttribute("indirect-tls-seg-refs");
+
+      if (F.getFnAttribute("fentry-call").getValueAsString() != "true") {
+        if (F.hasFnAttribute("mnop-mcount"))
+          report_fatal_error("mnop-mcount only supported with fentry-call");
+        if (F.hasFnAttribute("mrecord-mcount"))
+          report_fatal_error("mrecord-mcount only supported with fentry-call");
+      }
 
       // OptFor[Min]Size are used in pattern predicates that isel is matching.
-      OptForMinSize = MF.getFunction().hasMinSize();
+      OptForMinSize = F.hasMinSize();
       return SelectionDAGISel::runOnMachineFunction(MF);
     }
 

llvm/lib/Target/X86/X86MCInstLower.cpp

@@ -26,6 +26,7 @@
 #include "llvm/ADT/STLExtras.h"
 #include "llvm/ADT/SmallString.h"
 #include "llvm/ADT/StringExtras.h"
+#include "llvm/BinaryFormat/ELF.h"
 #include "llvm/CodeGen/MachineBranchProbabilityInfo.h"
 #include "llvm/CodeGen/MachineConstantPool.h"
 #include "llvm/CodeGen/MachineFunction.h"
@@ -44,6 +45,7 @@
 #include "llvm/MC/MCInst.h"
 #include "llvm/MC/MCInstBuilder.h"
 #include "llvm/MC/MCSection.h"
+#include "llvm/MC/MCSectionELF.h"
 #include "llvm/MC/MCStreamer.h"
 #include "llvm/MC/MCSymbol.h"
 #include "llvm/MC/TargetRegistry.h"
@@ -662,9 +664,8 @@ static unsigned emitNop(MCStreamer &OS, unsigned NumBytes,
   // target cpu.  15-bytes is the longest single NOP instruction, but some
   // platforms can't decode the longest forms efficiently.
   unsigned MaxNopLength = 1;
+  unsigned BaseReg = X86::RAX;
   if (Subtarget->is64Bit()) {
-    // FIXME: We can use NOOPL on 32-bit targets with FeatureNOPL, but the
-    // IndexReg/BaseReg below need to be updated.
     if (Subtarget->hasFeature(X86::TuningFast7ByteNOP))
       MaxNopLength = 7;
     else if (Subtarget->hasFeature(X86::TuningFast15ByteNOP))
@@ -673,16 +674,19 @@ static unsigned emitNop(MCStreamer &OS, unsigned NumBytes,
       MaxNopLength = 11;
     else
       MaxNopLength = 10;
-  } if (Subtarget->is32Bit())
+  } else if (Subtarget->is32Bit() &&
+             Subtarget->getFeatureBits()[X86::FeatureNOPL]) {
+    BaseReg = X86::EAX;
+    MaxNopLength = 10;
+  } else if (Subtarget->is32Bit())
     MaxNopLength = 2;
 
   // Cap a single nop emission at the profitable value for the target
   NumBytes = std::min(NumBytes, MaxNopLength);
 
   unsigned NopSize;
-  unsigned Opc, BaseReg, ScaleVal, IndexReg, Displacement, SegmentReg;
+  unsigned Opc, ScaleVal, IndexReg, Displacement, SegmentReg;
   IndexReg = Displacement = SegmentReg = 0;
-  BaseReg = X86::RAX;
   ScaleVal = 1;
   switch (NumBytes) {
   case 0:
@@ -709,13 +713,13 @@ static unsigned emitNop(MCStreamer &OS, unsigned NumBytes,
     NopSize = 5;
     Opc = X86::NOOPL;
     Displacement = 8;
-    IndexReg = X86::RAX;
+    IndexReg = BaseReg;
     break;
   case 6:
     NopSize = 6;
     Opc = X86::NOOPW;
     Displacement = 8;
-    IndexReg = X86::RAX;
+    IndexReg = BaseReg;
     break;
   case 7:
     NopSize = 7;
@@ -726,19 +730,19 @@ static unsigned emitNop(MCStreamer &OS, unsigned NumBytes,
     NopSize = 8;
     Opc = X86::NOOPL;
     Displacement = 512;
-    IndexReg = X86::RAX;
+    IndexReg = BaseReg;
     break;
   case 9:
     NopSize = 9;
     Opc = X86::NOOPW;
     Displacement = 512;
-    IndexReg = X86::RAX;
+    IndexReg = BaseReg;
     break;
   default:
     NopSize = 10;
     Opc = X86::NOOPW;
     Displacement = 512;
-    IndexReg = X86::RAX;
+    IndexReg = BaseReg;
     SegmentReg = X86::CS;
     break;
   }
@@ -885,6 +889,22 @@ void X86AsmPrinter::LowerFENTRY_CALL(const MachineInstr &MI,
                                      X86MCInstLower &MCIL) {
   bool Is64Bits = Subtarget->is64Bit();
   MCContext &Ctx = OutStreamer->getContext();
+
+  if (MF->getFunction().hasFnAttribute("mrecord-mcount")) {
+    MCSymbol *DotSym = OutContext.createTempSymbol();
+    OutStreamer->pushSection();
+    OutStreamer->switchSection(
+        Ctx.getELFSection("__mcount_loc", ELF::SHT_PROGBITS, ELF::SHF_ALLOC));
+    OutStreamer->emitSymbolValue(DotSym, Is64Bits ? 8 : 4);
+    OutStreamer->popSection();
+    OutStreamer->emitLabel(DotSym);
+  }
+
+  if (MF->getFunction().hasFnAttribute("mnop-mcount")) {
+    emitX86Nops(*OutStreamer, 5, Subtarget);
+    return;
+  }
+
   MCSymbol *fentry = Ctx.getOrCreateSymbol("__fentry__");
   const MCSymbolRefExpr *Op = MCSymbolRefExpr::create(fentry, Ctx);
 

llvm/test/CodeGen/X86/mnop-mcount-01.ll

@@ -0,0 +1,45 @@
+; RUN: llc %s -mtriple=x86_64-linux-gnu -o - -verify-machineinstrs \
+; RUN:   | FileCheck --check-prefixes=X64 %s
+; RUN: llc %s -mtriple=i686-linux-gnu -o - -verify-machineinstrs \
+; RUN:   | FileCheck --check-prefix=X86 %s
+; RUN: llc %s -mtriple=i686-linux-gnu -mcpu=pentium_pro -o - -verify-machineinstrs \
+; RUN:   | FileCheck --check-prefix=PPRO %s
+
+define void @test1() #0 {
+entry:
+  ret void
+
+; X64-LABEL: @test1
+; X64: callq __fentry__
+; X64: retq
+; X86-LABEL: @test1
+; X86: calll __fentry__
+; X86: retl
+; PPRO-LABEL: @test1
+; PPRO: calll __fentry__
+; PPRO: retl
+}
+
+define void @test2() #1 {
+entry:
+  ret void
+
+; X64-LABEL: @test2
+; X64-NOT: callq __fentry__
+; X64: nopl 8(%rax,%rax)
+; X64: retq
+; X86-LABEL: @test2
+; X86-NOT: calll __fentry__
+; X86: xchgw %ax, %ax
+; X86: xchgw %ax, %ax
+; X86: nop
+; X86: retl
+; PPRO-LABEL: @test2
+; PPRO: nopl 8(%eax,%eax)
+; PPRO-NOT: calll __fentry__
+; PPRO: retl
+}
+
+attributes #0 = { "fentry-call"="true" }
+attributes #1 = { "fentry-call"="true" "mnop-mcount" }
+

llvm/test/CodeGen/X86/mnop-mcount-02.ll

@@ -0,0 +1,11 @@
+; RUN: not --crash llc %s -mtriple=x86_64-linux-gnu -o - 2>&1 | FileCheck %s
+; RUN: not --crash llc %s -mtriple=i686-linux-gnu -o - 2>&1 | FileCheck %s
+;
+; CHECK: LLVM ERROR: mnop-mcount only supported with fentry-call
+
+define void @test1() #0 {
+entry:
+  ret void
+}
+
+attributes #0 = { "instrument-function-entry-inlined"="mcount" "mnop-mcount" }

llvm/test/CodeGen/X86/mrecord-mcount-01.ll

@@ -0,0 +1,68 @@
+; RUN: llc %s -mtriple=x86_64-linux-gnu -o - -verify-machineinstrs \
+; RUN:   | FileCheck --check-prefix=X64 %s
+; RUN: llc %s -mtriple=i686-linux-gnu -o - -verify-machineinstrs \
+; RUN:   | FileCheck --check-prefix=X86 %s
+; RUN: llc %s -mtriple=i686-linux-gnu -mcpu=pentium_pro -o - -verify-machineinstrs \
+; RUN:   | FileCheck --check-prefix=PPRO %s
+
+define void @test1() #0 {
+entry:
+  ret void
+
+; X64-LABEL: test1:
+; X64: .section __mcount_loc,"a",@progbits
+; X64: .quad .Ltmp0
+; X64: .text
+; X64: .Ltmp0:
+; X64: callq __fentry__
+; X64: retq
+; X86-LABEL: test1:
+; X86: .section __mcount_loc,"a",@progbits
+; X86: .long .Ltmp0
+; X86: .text
+; X86: .Ltmp0:
+; X86: calll __fentry__
+; X86: retl
+; PPRO-LABEL: test1:
+; PPRO: .section __mcount_loc,"a",@progbits
+; PPRO: .long .Ltmp0
+; PPRO: .text
+; PPRO: .Ltmp0:
+; PPRO: calll __fentry__
+; PPRO: retl
+}
+
+define void @test2() #1 {
+entry:
+  ret void
+
+; X64-LABEL: test2:
+; X64: .section __mcount_loc,"a",@progbits
+; X64: .quad .Ltmp1
+; X64: .text
+; X64: .Ltmp1:
+; X64: nopl 8(%rax,%rax)
+; X64-NOT: callq __fentry__
+; X64: retq
+; X86-LABEL: test2:
+; X86: .section __mcount_loc,"a",@progbits
+; X86: .long .Ltmp1
+; X86: .text
+; X86: .Ltmp1:
+; X86: xchgw %ax, %ax
+; X86: xchgw %ax, %ax
+; X86: nop
+; X86-NOT: calll __fentry__
+; X86: retl
+; PPRO-LABEL: test2:
+; PPRO: .section __mcount_loc,"a",@progbits
+; PPRO: .long .Ltmp1
+; PPRO: .text
+; PPRO: .Ltmp1:
+; PPRO: nopl 8(%eax,%eax)
+; PPRO-NOT: calll __fentry__
+; PPRO: retl
+}
+
+attributes #0 = { "fentry-call"="true" "mrecord-mcount" }
+attributes #1 = { "fentry-call"="true" "mnop-mcount" "mrecord-mcount" }

llvm/test/CodeGen/X86/mrecord-mcount-02.ll

@@ -0,0 +1,11 @@
+; RUN: not --crash llc %s -mtriple=x86_64-linux-gnu -o - 2>&1 | FileCheck %s
+; RUN: not --crash llc %s -mtriple=i686-linux-gnu -o - 2>&1 | FileCheck %s
+;
+; CHECK: LLVM ERROR: mrecord-mcount only supported with fentry-call
+
+define void @test1() #0 {
+entry:
+  ret void
+}
+
+attributes #0 = { "instrument-function-entry-inlined"="mcount" "mrecord-mcount" }