Move EntryExitInstrumentation pass location

[aeubanks]

Summary

This seems to be more of a Clang thing rather than a generic LLVM thing,
so this moves it out of LLVM pipelines and as Clang extension hooks into
LLVM pipelines.

Move the post-inline EEInstrumentation out of the backend pipeline and
into a late pass, similar to other sanitizer passes. It doesn't fit
into the codegen pipeline.

Also fix up EntryExitInstrumentation not running at -O0 under the new
PM. PR49143

[aeubanks]

missed non-x86 tests

clang/lib/CodeGen/BackendUtil.cpp

 static void addDataFlowSanitizerPass(const PassManagerBuilder &Builder,
                                      legacy::PassManagerBase &PM) {
   const PassManagerBuilderWrapper &BuilderWrapper =
       static_cast<const PassManagerBuilderWrapper&>(Builder);
   const LangOptions &LangOpts = BuilderWrapper.getLangOpts();
   PM.add(createDataFlowSanitizerLegacyPassPass(LangOpts.NoSanitizeFiles));
 }
 
+static void addEntryExitInstrumentationPass(const PassManagerBuilder &Builder,
+                                            legacy::PassManagerBase &PM) {
+  PM.add(createEntryExitInstrumenterPass());
+}
+
+static void
+addPostInlineEntryExitInstrumentationPass(const PassManagerBuilder &Builder,
+                                          legacy::PassManagerBase &PM) {
+  PM.add(createPostInlineEntryExitInstrumenterPass());
+}
+
 static TargetLibraryInfoImpl *createTLII(llvm::Triple &TargetTriple,
                                          const CodeGenOptions &CodeGenOpts) {
   TargetLibraryInfoImpl *TLII = new TargetLibraryInfoImpl(TargetTriple);
 
   switch (CodeGenOpts.getVecLib()) {
   case CodeGenOptions::Accelerate:
     TLII->addVectorizableFunctionsFromVecLib(TargetLibraryInfoImpl::Accelerate);
     break;
void EmitAssemblyHelper::CreatePasses(legacy::PassManager &MPM,
 
   if (LangOpts.Sanitize.has(SanitizerKind::DataFlow)) {
     PMBuilder.addExtension(PassManagerBuilder::EP_OptimizerLast,
                            addDataFlowSanitizerPass);
     PMBuilder.addExtension(PassManagerBuilder::EP_EnabledOnOptLevel0,
                            addDataFlowSanitizerPass);
   }
 
+  if (CodeGenOpts.InstrumentFunctions ||
+      CodeGenOpts.InstrumentFunctionEntryBare ||
+      CodeGenOpts.InstrumentFunctionsAfterInlining ||
+      CodeGenOpts.InstrumentForProfiling) {
+    PMBuilder.addExtension(PassManagerBuilder::EP_EarlyAsPossible,
+                           addEntryExitInstrumentationPass);
+    PMBuilder.addExtension(PassManagerBuilder::EP_EnabledOnOptLevel0,
+                           addEntryExitInstrumentationPass);
+    PMBuilder.addExtension(PassManagerBuilder::EP_OptimizerLast,
+                           addPostInlineEntryExitInstrumentationPass);
+    PMBuilder.addExtension(PassManagerBuilder::EP_EnabledOnOptLevel0,
+                           addPostInlineEntryExitInstrumentationPass);
+  }
+
   // Set up the per-function pass manager.
   FPM.add(new TargetLibraryInfoWrapperPass(*TLII));
   if (CodeGenOpts.VerifyModule)
     FPM.add(createVerifierPass());
 
   // Set up the per-module pass manager.
   if (!CodeGenOpts.RewriteMapFiles.empty())
     addSymbolRewriterPass(CodeGenOpts, &MPM);
if (!CodeGenOpts.DisableLLVMPasses) {
     if (IsThinLTOPostLink)
       PB.registerPipelineStartEPCallback(
           [](ModulePassManager &MPM, PassBuilder::OptimizationLevel Level) {
             MPM.addPass(LowerTypeTestsPass(/*ExportSummary=*/nullptr,
                                            /*ImportSummary=*/nullptr,
                                            /*DropTypeTests=*/true));
           });
 
-    if (Level != PassBuilder::OptimizationLevel::O0) {
+    if (CodeGenOpts.InstrumentFunctions ||
+        CodeGenOpts.InstrumentFunctionEntryBare ||
+        CodeGenOpts.InstrumentFunctionsAfterInlining ||
+        CodeGenOpts.InstrumentForProfiling) {
       PB.registerPipelineStartEPCallback(
           [](ModulePassManager &MPM, PassBuilder::OptimizationLevel Level) {
             MPM.addPass(createModuleToFunctionPassAdaptor(
                 EntryExitInstrumenterPass(/*PostInlining=*/false)));
           });
+      PB.registerOptimizerLastEPCallback(
+          [](ModulePassManager &MPM, PassBuilder::OptimizationLevel Level) {
+            MPM.addPass(createModuleToFunctionPassAdaptor(
+                EntryExitInstrumenterPass(/*PostInlining=*/true)));
+          });
     }
 
     // Register callbacks to schedule sanitizer passes at the appropriate part
     // of the pipeline.
     if (LangOpts.Sanitize.has(SanitizerKind::LocalBounds))
       PB.registerScalarOptimizerLateEPCallback(
           [](FunctionPassManager &FPM, PassBuilder::OptimizationLevel Level) {
             FPM.addPass(BoundsCheckingPass());

clang/test/CodeGen/X86/x86_64-instrument-functions.c

 // REQUIRES: x86-registered-target
-// RUN: %clang_cc1 -fno-experimental-new-pass-manager -triple x86_64-unknown-unknown -S -finstrument-functions -O2 -o - %s | FileCheck %s
-// RUN: %clang_cc1 -fno-experimental-new-pass-manager -triple x86_64-unknown-unknown -S -finstrument-functions-after-inlining -O2 -o - %s | FileCheck -check-prefix=NOINLINE %s
+// RUN: %clang_cc1 -fno-experimental-new-pass-manager -triple x86_64-unknown-unknown -S -finstrument-functions -O0 -emit-llvm -o - %s | FileCheck %s
+// RUN: %clang_cc1 -fno-experimental-new-pass-manager -triple x86_64-unknown-unknown -S -finstrument-functions -O2 -emit-llvm -o - %s | FileCheck %s
+// RUN: %clang_cc1 -fno-experimental-new-pass-manager -triple x86_64-unknown-unknown -S -finstrument-functions-after-inlining -O2 -o - -emit-llvm %s | FileCheck -check-prefix=NOINLINE %s
+
+// RUN: %clang_cc1 -fexperimental-new-pass-manager -triple x86_64-unknown-unknown -S -finstrument-functions -O0 -o - -emit-llvm %s | FileCheck %s
+// RUN: %clang_cc1 -fexperimental-new-pass-manager -triple x86_64-unknown-unknown -S -finstrument-functions -O2 -o - -emit-llvm %s | FileCheck %s
+// RUN: %clang_cc1 -fexperimental-new-pass-manager -triple x86_64-unknown-unknown -S -finstrument-functions-after-inlining -O2 -o - -emit-llvm %s | FileCheck -check-prefix=NOINLINE %s
 
-// RUN: %clang_cc1 -fexperimental-new-pass-manager -triple x86_64-unknown-unknown -S -finstrument-functions -O2 -o - %s | FileCheck %s
-// RUN: %clang_cc1 -fexperimental-new-pass-manager -triple x86_64-unknown-unknown -S -finstrument-functions-after-inlining -O2 -o - %s | FileCheck -check-prefix=NOINLINE %s
-
-// It's not so nice having asm tests in Clang, but we need to check that we set
-// up the pipeline correctly in order to have the instrumentation inserted.
-
-int leaf(int x) {
+__attribute__((always_inline)) int leaf(int x) {
   return x;
-// CHECK-LABEL: leaf:
-// CHECK: callq __cyg_profile_func_enter
+// CHECK-LABEL: define {{.*}} @leaf
+// CHECK: call void @__cyg_profile_func_enter
 // CHECK-NOT: cyg_profile
-// CHECK: callq __cyg_profile_func_exit
+// CHECK: call void @__cyg_profile_func_exit
 // CHECK-NOT: cyg_profile
 // CHECK: ret
 }
 
 int root(int x) {
   return leaf(x);
-// CHECK-LABEL: root:
-// CHECK: callq __cyg_profile_func_enter
+// CHECK-LABEL: define {{.*}} @root
+// CHECK: call void @__cyg_profile_func_enter
 // CHECK-NOT: cyg_profile
 
 // Inlined from leaf():
-// CHECK: callq __cyg_profile_func_enter
+// CHECK: call void @__cyg_profile_func_enter
 // CHECK-NOT: cyg_profile
-// CHECK: callq __cyg_profile_func_exit
-
+// CHECK: call void @__cyg_profile_func_exit
 // CHECK-NOT: cyg_profile
-// CHECK: callq __cyg_profile_func_exit
+
+// CHECK: call void @__cyg_profile_func_exit
 // CHECK: ret
 
-// NOINLINE-LABEL: root:
-// NOINLINE: callq __cyg_profile_func_enter
+// NOINLINE-LABEL: define {{.*}} @root
+// NOINLINE: call void @__cyg_profile_func_enter
+// NOINLINE-NOT: cyg_profile
+// NOINLINE: call void @__cyg_profile_func_exit
 // NOINLINE-NOT: cyg_profile
-// NOINLINE: callq __cyg_profile_func_exit
 // NOINLINE: ret
 }

clang/test/CodeGen/mcount.c

 int __attribute__((no_instrument_function)) no_instrument(void) {
   return foo();
 }
 
 int main(void) {
   return no_instrument();
 }
 
-// CHECK: attributes #0 = { {{.*}}"instrument-function-entry-inlined"="mcount"{{.*}} }
-// CHECK: attributes #1 = { {{.*}} }
-// CHECK-PREFIXED: attributes #0 = { {{.*}}"instrument-function-entry-inlined"="_mcount"{{.*}} }
-// CHECK-PREFIXED: attributes #1 = { {{.*}} }
-// CHECK-DOUBLE-PREFIXED: attributes #0 = { {{.*}}"instrument-function-entry-inlined"="__mcount"{{.*}} }
-// CHECK-DOUBLE-PREFIXED: attributes #1 = { {{.*}} }
-// NO-MCOUNT-NOT: attributes #{{[0-9]}} = { {{.*}}"instrument-function-entry-inlined"={{.*}} }
-// NO-MCOUNT1-NOT: attributes #1 = { {{.*}}"instrument-function-entry-inlined"={{.*}} }
+// CHECK: call void @mcount
+// CHECK: call void @mcount
+// CHECK: call void @mcount
+// CHECK-NOT: call void @mcount
+// CHECK-PREFIXED: call void @_mcount
+// CHECK-PREFIXED: call void @_mcount
+// CHECK-PREFIXED: call void @_mcount
+// CHECK-PREFIXED-NOT: call void @_mcount
+// CHECK-DOUBLE-PREFIXED: call void @__mcount
+// CHECK-DOUBLE-PREFIXED: call void @__mcount
+// CHECK-DOUBLE-PREFIXED: call void @__mcount
+// CHECK-DOUBLE-PREFIXED-NOT: call void @__mcount
+// NO-MCOUNT-NOT: call void @{{.*}}mcount
+// NO-MCOUNT1-NOT: call void @{{.*}}mcount

clang/test/Frontend/gnu-mcount.c

 // RUN: %clang -target aarch64-unknown-cloudabi-gnueabihf -meabi gnu -pg -S -emit-llvm -o - %s | FileCheck %s --check-prefixes=CHECK,MCOUNT
 
 int f() {
   return 0;
 }
 
 // CHECK-LABEL: f
 // TODO: add profiling support for arm-baremetal
-// UNSUPPORTED-NOT: "instrument-function-entry-inlined"=
-// CHECK-ARM-EABI: attributes #{{[0-9]+}} = { {{.*}}"instrument-function-entry-inlined"="\01mcount"{{.*}} }
-// MCOUNT: attributes #{{[0-9]+}} = { {{.*}}"instrument-function-entry-inlined"="mcount"
-// UNDER: attributes #{{[0-9]+}} = { {{.*}}"instrument-function-entry-inlined"="\01_mcount"
-// UNDER_UNDER: attributes #{{[0-9]+}} = { {{.*}}"instrument-function-entry-inlined"="__mcount"
-// CHECK-ARM64-EABI-FREEBSD: attributes #{{[0-9]+}} = { {{.*}}"instrument-function-entry-inlined"=".mcount"{{.*}} }
-// CHECK-ARM-EABI-MEABI-GNU: attributes #{{[0-9]+}} = { {{.*}}"instrument-function-entry-inlined"="llvm.arm.gnu.eabi.mcount"{{.*}} }
+// UNSUPPORTED-NOT: call void
+// CHECK-ARM-EABI: call void @"\01mcount"()
+// MCOUNT: call void @mcount()
+// UNDER: call void @"\01_mcount"()
+// UNDER_UNDER: call void @__mcount()
+// CHECK-ARM64-EABI-FREEBSD: call void @.mcount()
+// CHECK-ARM-EABI-MEABI-GNU: call void @llvm.arm.gnu.eabi.mcount()

llvm/include/llvm/Transforms/Utils/EntryExitInstrumenter.h

 
 struct EntryExitInstrumenterPass
     : public PassInfoMixin<EntryExitInstrumenterPass> {
   EntryExitInstrumenterPass(bool PostInlining) : PostInlining(PostInlining) {}
 
   PreservedAnalyses run(Function &F, FunctionAnalysisManager &AM);
 
   bool PostInlining;
+
+  static bool isRequired() { return true; }
 };
 
 } // namespace llvm
 
 #endif // LLVM_TRANSFORMS_UTILS_ENTRYEXITINSTRUMENTER_H

llvm/lib/CodeGen/TargetPassConfig.cpp

if (getOptLevel() != CodeGenOpt::None && !DisableConstantHoisting)
     addPass(createConstantHoistingPass());
 
   if (getOptLevel() != CodeGenOpt::None)
     addPass(createReplaceWithVeclibLegacyPass());
 
   if (getOptLevel() != CodeGenOpt::None && !DisablePartialLibcallInlining)
     addPass(createPartiallyInlineLibCallsPass());
 
-  // Instrument function entry and exit, e.g. with calls to mcount().
-  addPass(createPostInlineEntryExitInstrumenterPass());
-
   // Add scalarization of target's unsupported masked memory intrinsics pass.
   // the unsupported intrinsic will be replaced with a chain of basic blocks,
   // that stores/loads element one-by-one if the appropriate mask bit is set.
   addPass(createScalarizeMaskedMemIntrinLegacyPass());
 
   // Expand reduction intrinsics into shuffle sequences if the target wants to.
   // Allow disabling it for testing purposes.
   if (!DisableExpandReductions)

llvm/lib/Transforms/IPO/PassManagerBuilder.cpp

void PassManagerBuilder::addInitialAliasAnalysisPasses(
   // support "obvious" type-punning idioms.
   PM.add(createTypeBasedAAWrapperPass());
   PM.add(createScopedNoAliasAAWrapperPass());
 }
 
 void PassManagerBuilder::populateFunctionPassManager(
     legacy::FunctionPassManager &FPM) {
   addExtensionsToPM(EP_EarlyAsPossible, FPM);
-  FPM.add(createEntryExitInstrumenterPass());
 
   // Add LibraryInfo if we have some.
   if (LibraryInfo)
     FPM.add(new TargetLibraryInfoWrapperPass(*LibraryInfo));
 
   // The backends do not handle matrix intrinsics currently.
   // Make sure they are also lowered in O0.
   // FIXME: A lightweight version of the pass should run in the backend

llvm/test/CodeGen/AArch64/O0-pipeline.ll

 ; CHECK-NEXT:     Pre-ISel Intrinsic Lowering
 ; CHECK-NEXT:     FunctionPass Manager
 ; CHECK-NEXT:       Expand Atomic instructions
 ; CHECK-NEXT:       Module Verifier
 ; CHECK-NEXT:       Lower Garbage Collection Instructions
 ; CHECK-NEXT:       Shadow Stack GC Lowering
 ; CHECK-NEXT:       Lower constant intrinsics
 ; CHECK-NEXT:       Remove unreachable blocks from the CFG
-; CHECK-NEXT:       Instrument function entry/exit with calls to e.g. mcount() (post inlining)
 ; CHECK-NEXT:       Scalarize Masked Memory Intrinsics
 ; CHECK-NEXT:       Expand reduction intrinsics
 ; CHECK-NEXT:       AArch64 Stack Tagging
 ; CHECK-NEXT:     Rewrite Symbols
 ; CHECK-NEXT:     FunctionPass Manager
 ; CHECK-NEXT:       Exception handling preparation
 ; CHECK-NEXT:       Safe Stack instrumentation pass
 ; CHECK-NEXT:       Insert stack protectors

llvm/test/CodeGen/AArch64/O3-pipeline.ll

 ; CHECK-NEXT:       Remove unreachable blocks from the CFG
 ; CHECK-NEXT:       Natural Loop Information
 ; CHECK-NEXT:       Post-Dominator Tree Construction
 ; CHECK-NEXT:       Branch Probability Analysis
 ; CHECK-NEXT:       Block Frequency Analysis
 ; CHECK-NEXT:       Constant Hoisting
 ; CHECK-NEXT:       Replace intrinsics with calls to vector library
 ; CHECK-NEXT:       Partially inline calls to library functions
-; CHECK-NEXT:       Instrument function entry/exit with calls to e.g. mcount() (post inlining)
 ; CHECK-NEXT:       Scalarize Masked Memory Intrinsics
 ; CHECK-NEXT:       Expand reduction intrinsics
 ; CHECK-NEXT:     Stack Safety Analysis
 ; CHECK-NEXT:       FunctionPass Manager
 ; CHECK-NEXT:         Dominator Tree Construction
 ; CHECK-NEXT:         Natural Loop Information
 ; CHECK-NEXT:         Scalar Evolution Analysis
 ; CHECK-NEXT:         Stack Safety Local Analysis

llvm/test/CodeGen/AMDGPU/opt-pipeline.ll

 ; RUN: opt -O0 -mtriple=amdgcn--amdhsa -disable-output -disable-verify -debug-pass=Structure -enable-new-pm=0 %s 2>&1 | FileCheck -check-prefix=GCN-O0 %s
 ; RUN: opt -O1 -mtriple=amdgcn--amdhsa -disable-output -disable-verify -debug-pass=Structure -enable-new-pm=0 %s 2>&1 | FileCheck -check-prefix=GCN-O1 %s
 ; RUN: opt -O2 -mtriple=amdgcn--amdhsa -disable-output -disable-verify -debug-pass=Structure -enable-new-pm=0 %s 2>&1 | FileCheck -check-prefix=GCN-O2 %s
 ; RUN: opt -O3 -mtriple=amdgcn--amdhsa -disable-output -disable-verify -debug-pass=Structure -enable-new-pm=0 %s 2>&1 | FileCheck -check-prefix=GCN-O3 %s
 
 ; REQUIRES: asserts
 
 ; GCN-O0:      Pass Arguments:
 ; GCN-O0-NEXT: Target Transform Information
 ; GCN-O0-NEXT:   FunctionPass Manager
 ; GCN-O0-NEXT:     Early propagate attributes from kernels to functions
 ; GCN-O0-NEXT:     Replace builtin math calls with that native versions.
-; GCN-O0-NEXT:     Instrument function entry/exit with calls to e.g. mcount() (pre inlining)
 
 ; GCN-O0-NEXT: Pass Arguments:
 ; GCN-O0-NEXT: Target Library Information
 ; GCN-O0-NEXT: Target Transform Information
 ; GCN-O0-NEXT: Target Pass Configuration
 ; GCN-O0-NEXT: Assumption Cache Tracker
 ; GCN-O0-NEXT: Profile summary info
 ; GCN-O0-NEXT:   ModulePass Manager

 ; GCN-O1-NEXT: Scoped NoAlias Alias Analysis
 ; GCN-O1-NEXT:   FunctionPass Manager
 ; GCN-O1-NEXT:     Early propagate attributes from kernels to functions
 ; GCN-O1-NEXT:     Replace builtin math calls with that native versions.
 ; GCN-O1-NEXT:     Dominator Tree Construction
 ; GCN-O1-NEXT:     Basic Alias Analysis (stateless AA impl)
 ; GCN-O1-NEXT:     Function Alias Analysis Results
 ; GCN-O1-NEXT:     Simplify well-known AMD library calls
-; GCN-O1-NEXT:     Instrument function entry/exit with calls to e.g. mcount() (pre inlining)
 ; GCN-O1-NEXT:     Simplify the CFG
 ; GCN-O1-NEXT:     Dominator Tree Construction
 ; GCN-O1-NEXT:     SROA
 ; GCN-O1-NEXT:     Early CSE
 ; GCN-O1-NEXT:     Lower 'expect' Intrinsics
 
 ; GCN-O1-NEXT: Pass Arguments:
 ; GCN-O1-NEXT: Target Library Information

 ; GCN-O2-NEXT: Scoped NoAlias Alias Analysis
 ; GCN-O2-NEXT:   FunctionPass Manager
 ; GCN-O2-NEXT:     Early propagate attributes from kernels to functions
 ; GCN-O2-NEXT:     Replace builtin math calls with that native versions.
 ; GCN-O2-NEXT:     Dominator Tree Construction
 ; GCN-O2-NEXT:     Basic Alias Analysis (stateless AA impl)
 ; GCN-O2-NEXT:     Function Alias Analysis Results
 ; GCN-O2-NEXT:     Simplify well-known AMD library calls
-; GCN-O2-NEXT:     Instrument function entry/exit with calls to e.g. mcount() (pre inlining)
 ; GCN-O2-NEXT:     Simplify the CFG
 ; GCN-O2-NEXT:     Dominator Tree Construction
 ; GCN-O2-NEXT:     SROA
 ; GCN-O2-NEXT:     Early CSE
 ; GCN-O2-NEXT:     Lower 'expect' Intrinsics
 
 ; GCN-O2-NEXT: Pass Arguments:
 ; GCN-O2-NEXT: Target Library Information

 ; GCN-O3-NEXT: Scoped NoAlias Alias Analysis
 ; GCN-O3-NEXT:   FunctionPass Manager
 ; GCN-O3-NEXT:     Early propagate attributes from kernels to functions
 ; GCN-O3-NEXT:     Replace builtin math calls with that native versions.
 ; GCN-O3-NEXT:     Dominator Tree Construction
 ; GCN-O3-NEXT:     Basic Alias Analysis (stateless AA impl)
 ; GCN-O3-NEXT:     Function Alias Analysis Results
 ; GCN-O3-NEXT:     Simplify well-known AMD library calls
-; GCN-O3-NEXT:     Instrument function entry/exit with calls to e.g. mcount() (pre inlining)
 ; GCN-O3-NEXT:     Simplify the CFG
 ; GCN-O3-NEXT:     Dominator Tree Construction
 ; GCN-O3-NEXT:     SROA
 ; GCN-O3-NEXT:     Early CSE
 ; GCN-O3-NEXT:     Lower 'expect' Intrinsics
 
 ; GCN-O3-NEXT: Pass Arguments:
 ; GCN-O3-NEXT: Target Library Information

llvm/test/CodeGen/ARM/O3-pipeline.ll

 ; CHECK-NEXT:      Remove unreachable blocks from the CFG
 ; CHECK-NEXT:      Natural Loop Information
 ; CHECK-NEXT:      Post-Dominator Tree Construction
 ; CHECK-NEXT:      Branch Probability Analysis
 ; CHECK-NEXT:      Block Frequency Analysis
 ; CHECK-NEXT:      Constant Hoisting
 ; CHECK-NEXT:      Replace intrinsics with calls to vector library
 ; CHECK-NEXT:      Partially inline calls to library functions
-; CHECK-NEXT:      Instrument function entry/exit with calls to e.g. mcount() (post inlining)
 ; CHECK-NEXT:      Scalarize Masked Memory Intrinsics
 ; CHECK-NEXT:      Expand reduction intrinsics
 ; CHECK-NEXT:      Natural Loop Information
 ; CHECK-NEXT:      Scalar Evolution Analysis
 ; CHECK-NEXT:      Basic Alias Analysis (stateless AA impl)
 ; CHECK-NEXT:      Function Alias Analysis Results
 ; CHECK-NEXT:      Transform functions to use DSP intrinsics
 ; CHECK-NEXT:      Interleaved Access Pass

llvm/test/CodeGen/ARM/gnu_mcount_nc.ll

 ; CHECK-ARM-GLOBAL-ISEL:        stmdb   sp!, {lr}
 ; CHECK-ARM-GLOBAL-ISEL-NEXT:   bl      __gnu_mcount_nc
 ; CHECK-THUMB:                  push    {lr}
 ; CHECK-THUMB-NEXT:             bl      __gnu_mcount_nc
 ; CHECK-THUMB-FAST-ISEL:        push    {lr}
 ; CHECK-THUMB-FAST-ISEL-NEXT:   bl      __gnu_mcount_nc
 ; CHECK-THUMB-GLOBAL-ISEL:      push    {lr}
 ; CHECK-THUMB-GLOBAL-ISEL-NEXT: bl      __gnu_mcount_nc
+  call void @llvm.arm.gnu.eabi.mcount()
   ret void
 }
 
 define dso_local void @caller() #0 {
 ; CHECK-ARM:                    stmdb   sp!, {lr}
 ; CHECK-ARM-NEXT:               bl      __gnu_mcount_nc
 ; CHECK-ARM-FAST-ISEL:          stmdb   sp!, {lr}
 ; CHECK-ARM-FAST-ISEL-NEXT:     bl      __gnu_mcount_nc
 ; CHECK-ARM-GLOBAL-ISEL:        stmdb   sp!, {lr}
 ; CHECK-ARM-GLOBAL-ISEL-NEXT:   bl      __gnu_mcount_nc
 ; CHECK-THUMB:                  push    {lr}
 ; CHECK-THUMB-NEXT:             bl      __gnu_mcount_nc
 ; CHECK-THUMB-FAST-ISEL:        push    {lr}
 ; CHECK-THUMB-FAST-ISEL-NEXT:   bl      __gnu_mcount_nc
 ; CHECK-THUMB-GLOBAL-ISEL:      push    {lr}
 ; CHECK-THUMB-GLOBAL-ISEL-NEXT: bl      __gnu_mcount_nc
+  call void @llvm.arm.gnu.eabi.mcount()
   call void @callee()
   ret void
 }
 
-attributes #0 = { nofree nounwind "instrument-function-entry-inlined"="llvm.arm.gnu.eabi.mcount" }
+declare void @llvm.arm.gnu.eabi.mcount() #1
+
+attributes #0 = { nofree nounwind }
+attributes #1 = { nounwind }

llvm/test/CodeGen/Mips/long-call-mcount.ll

 ; Check call to mcount in case of long/short call options.
 ; RUN: llc -march=mips -target-abi o32 --mattr=+long-calls,+noabicalls < %s \
 ; RUN:   -mips-jalr-reloc=false | FileCheck -check-prefixes=CHECK,LONG %s
 ; RUN: llc -march=mips -target-abi o32 --mattr=-long-calls,+noabicalls < %s \
 ; RUN:   -mips-jalr-reloc=false | FileCheck -check-prefixes=CHECK,SHORT %s
 
-; Function Attrs: noinline nounwind optnone
-define void @foo() #0 {
+define void @foo() {
 entry:
+  call void @_mcount()
   ret void
 
 ; CHECK-LABEL: foo
 ; LONG:          lui     $1, %hi(_mcount)
 ; LONG-NEXT:     addiu   $25, $1, %lo(_mcount)
 ; LONG-NEXT:     jalr    $25
 ; SHORT:         jal     _mcount
 }
 
-attributes #0 = { "instrument-function-entry-inlined"="_mcount" }
+declare void @_mcount()

llvm/test/CodeGen/Mips/mcount.ll

 ; RUN:   < %s | FileCheck %s -check-prefix=MIPS64-PIC
 ; RUN: llc -march=mips -verify-machineinstrs -mattr=+micromips \
 ; RUN:   < %s | FileCheck %s -check-prefix=MIPS32-MM
 ; RUN: llc -march=mips -verify-machineinstrs -relocation-model=pic -mattr=+micromips \
 ; RUN:   < %s | FileCheck %s -check-prefix=MIPS32-MM-PIC
 
 ; Test that checks ABI for _mcount calls.
 
-; Function Attrs: noinline nounwind optnone
-define void @foo() #0 {
+define void @foo() {
 ; MIPS32-LABEL: foo:
 ; MIPS32:       # %bb.0: # %entry
 ; MIPS32-NEXT:    addiu $sp, $sp, -24
 ; MIPS32-NEXT:    .cfi_def_cfa_offset 24
 ; MIPS32-NEXT:    sw $ra, 20($sp) # 4-byte Folded Spill
 ; MIPS32-NEXT:    .cfi_offset 31, -4
 ; MIPS32-NEXT:    move $1, $ra
 ; MIPS32-NEXT:    jal _mcount

 ; MIPS32-MM-PIC-NEXT:    .reloc ($tmp0), R_MICROMIPS_JALR, _mcount
 ; MIPS32-MM-PIC-NEXT:  $tmp0:
 ; MIPS32-MM-PIC-NEXT:    jalr $25
 ; MIPS32-MM-PIC-NEXT:    addiu $sp, $sp, -8
 ; MIPS32-MM-PIC-NEXT:    lw $ra, 20($sp) # 4-byte Folded Reload
 ; MIPS32-MM-PIC-NEXT:    jr $ra
 ; MIPS32-MM-PIC-NEXT:    addiu $sp, $sp, 24
 entry:
+  call void @_mcount()
   ret void
 }
 
-attributes #0 = { "instrument-function-entry-inlined"="_mcount" }
+declare void @_mcount()

llvm/test/CodeGen/PowerPC/mcount-insertion.ll

-; RUN: opt -ee-instrument < %s | opt -inline | llc -mtriple=powerpc64-unknown-linux-gnu | FileCheck %s
+; RUN: opt -ee-instrument < %s | opt -inline | opt -post-inline-ee-instrument | llc -mtriple=powerpc64-unknown-linux-gnu | FileCheck %s
 
 ; The run-line mimics how Clang might run the instrumentation passes.
 
 target datalayout = "E-m:e-i64:64-n32:64"
 
 
 define void @leaf_function() #0 {
 entry:

llvm/test/CodeGen/X86/O0-pipeline.ll

 ; CHECK-NEXT:     FunctionPass Manager
 ; CHECK-NEXT:       Expand Atomic instructions
 ; CHECK-NEXT:       Lower AMX type for load/store
 ; CHECK-NEXT:       Module Verifier
 ; CHECK-NEXT:       Lower Garbage Collection Instructions
 ; CHECK-NEXT:       Shadow Stack GC Lowering
 ; CHECK-NEXT:       Lower constant intrinsics
 ; CHECK-NEXT:       Remove unreachable blocks from the CFG
-; CHECK-NEXT:       Instrument function entry/exit with calls to e.g. mcount() (post inlining)
 ; CHECK-NEXT:       Scalarize Masked Memory Intrinsics
 ; CHECK-NEXT:       Expand reduction intrinsics
 ; CHECK-NEXT:       Expand indirectbr instructions
 ; CHECK-NEXT:     Rewrite Symbols
 ; CHECK-NEXT:     FunctionPass Manager
 ; CHECK-NEXT:       Exception handling preparation
 ; CHECK-NEXT:       Safe Stack instrumentation pass
 ; CHECK-NEXT:       Insert stack protectors

llvm/test/CodeGen/X86/musttail-inalloca.ll

 ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
 ; RUN: llc %s -o - | FileCheck %s
 
 ; Previously, we would accidentally leave behind SP adjustments to setup a call
 ; frame for the musttail call target, and SP adjustments would end up
 ; unbalanced. Reported as https://crbug.com/1026882.
 
 target datalayout = "e-m:x-p:32:32-p270:32:32-p271:32:32-p272:64:64-i64:64-f80:32-n8:16:32-a:0:32-S32"
 target triple = "i386-pc-windows-msvc19.16.0"
 
 ; 20 bytes of memory.
 %struct.Args = type { i32, i32, i32, i32, i32 }
 
 declare dso_local x86_thiscallcc void @methodWithVtorDisp(i8* nocapture readonly, <{ %struct.Args }>* inalloca)
 
+; Function Attrs: nounwind optsize
 define dso_local x86_thiscallcc void @methodWithVtorDisp_thunk(i8* %0, <{ %struct.Args }>* inalloca %1) #0 {
 ; CHECK-LABEL: methodWithVtorDisp_thunk:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    pushl %esi
 ; CHECK-NEXT:    movl %ecx, %esi
 ; CHECK-NEXT:    subl -4(%ecx), %esi
 ; CHECK-NEXT:    pushl {{[0-9]+}}(%esp)
 ; CHECK-NEXT:    pushl $_methodWithVtorDisp_thunk
 ; CHECK-NEXT:    calll ___cyg_profile_func_exit
 ; CHECK-NEXT:    addl $8, %esp
 ; CHECK-NEXT:    movl %esi, %ecx
 ; CHECK-NEXT:    popl %esi
 ; CHECK-NEXT:    jmp _methodWithVtorDisp # TAILCALL
   %3 = getelementptr inbounds i8, i8* %0, i32 -4
   %4 = bitcast i8* %3 to i32*
   %5 = load i32, i32* %4, align 4
   %6 = sub i32 0, %5
   %7 = getelementptr i8, i8* %0, i32 %6
+  %8 = call i8* @llvm.returnaddress(i32 0)
+  call void @__cyg_profile_func_exit(i8* bitcast (void (i8*, <{ %struct.Args }>*)* @methodWithVtorDisp_thunk to i8*), i8* %8)
   musttail call x86_thiscallcc void @methodWithVtorDisp(i8* %7, <{ %struct.Args }>* inalloca nonnull %1)
   ret void
 }
 
-attributes #0 = { nounwind optsize "instrument-function-exit-inlined"="__cyg_profile_func_exit"  }
+declare void @__cyg_profile_func_exit(i8*, i8*)
+
+; Function Attrs: nofree nosync nounwind readnone willreturn
+declare i8* @llvm.returnaddress(i32 immarg) #1
+
+attributes #0 = { nounwind optsize }
+attributes #1 = { nofree nosync nounwind readnone willreturn }

llvm/test/CodeGen/X86/opt-pipeline.ll

 ; CHECK-NEXT:       Remove unreachable blocks from the CFG
 ; CHECK-NEXT:       Natural Loop Information
 ; CHECK-NEXT:       Post-Dominator Tree Construction
 ; CHECK-NEXT:       Branch Probability Analysis
 ; CHECK-NEXT:       Block Frequency Analysis
 ; CHECK-NEXT:       Constant Hoisting
 ; CHECK-NEXT:       Replace intrinsics with calls to vector library
 ; CHECK-NEXT:       Partially inline calls to library functions
-; CHECK-NEXT:       Instrument function entry/exit with calls to e.g. mcount() (post inlining)
 ; CHECK-NEXT:       Scalarize Masked Memory Intrinsics
 ; CHECK-NEXT:       Expand reduction intrinsics
 ; CHECK-NEXT:       Interleaved Access Pass
 ; CHECK-NEXT:       X86 Partial Reduction
 ; CHECK-NEXT:       Expand indirectbr instructions
 ; CHECK-NEXT:       Natural Loop Information
 ; CHECK-NEXT:       CodeGen Prepare
 ; CHECK-NEXT:     Rewrite Symbols

llvm/test/Other/opt-O0-pipeline-enable-matrix.ll

 ; RUN: opt -enable-new-pm=0 -O0 -enable-matrix -debug-pass=Structure < %s -o /dev/null 2>&1 | FileCheck %s
 
 ; REQUIRES: asserts
 
 ; CHECK:      Pass Arguments:
 ; CHECK-NEXT: Target Transform Information
 ; CHECK-NEXT:   FunctionPass Manager
 ; CHECK-NEXT:     Module Verifier
-; CHECK-NEXT:     Instrument function entry/exit with calls to e.g. mcount() (pre inlining)
 ; CHECK-NEXT:     Lower the matrix intrinsics (minimal)
 
 
 define void @f() {
   ret void
 }

llvm/test/Other/opt-O0-pipeline.ll

 ; RUN: opt -enable-new-pm=0 -mtriple=x86_64-- -O0 -debug-pass=Structure < %s -o /dev/null 2>&1 | FileCheck %s --check-prefixes=CHECK,%llvmcheckext
 
 ; REQUIRES: asserts
 
 ; Suppress FileCheck --allow-unused-prefixes=false diagnostics.
 ; CHECK-NOEXT: {{^}}
 
 ; CHECK-LABEL: Pass Arguments:
 ; CHECK-NEXT: Target Transform Information
 ; CHECK-NEXT:   FunctionPass Manager
 ; CHECK-NEXT:     Module Verifier
 ; CHECK-EXT:     Good Bye World Pass
-; CHECK-NEXT:     Instrument function entry/exit with calls to e.g. mcount() (pre inlining)
 ; CHECK-NEXT: Pass Arguments:
 ; CHECK-NEXT: Target Library Information
 ; CHECK-NEXT: Target Transform Information
 ;             Target Pass Configuration
 ; CHECK:      Assumption Cache Tracker
 ; CHECK-NEXT: Profile summary info
 ; CHECK-NEXT:   ModulePass Manager
 ; CHECK-NEXT:     Annotation2Metadata

llvm/test/Other/opt-O2-pipeline.ll

 ; RUN: opt -enable-new-pm=0 -mtriple=x86_64-- -O2 -debug-pass=Structure < %s -o /dev/null 2>&1 | FileCheck --check-prefixes=CHECK,%llvmcheckext %s
 
 ; REQUIRES: asserts
 
 ; CHECK-LABEL: Pass Arguments:
 ; CHECK-NEXT: Target Transform Information
 ; CHECK-NEXT: Type-Based Alias Analysis
 ; CHECK-NEXT: Scoped NoAlias Alias Analysis
 ; CHECK-NEXT: Assumption Cache Tracker
 ; CHECK-NEXT: Target Library Information
 ; CHECK-NEXT:   FunctionPass Manager
 ; CHECK-NEXT:     Module Verifier
 ; CHECK-EXT:      Good Bye World Pass
 ; CHECK-NOEXT-NOT:      Good Bye World Pass
-; CHECK-NEXT:     Instrument function entry/exit with calls to e.g. mcount() (pre inlining)
 ; CHECK-NEXT:     Simplify the CFG
 ; CHECK-NEXT:     Dominator Tree Construction
 ; CHECK-NEXT:     SROA
 ; CHECK-NEXT:     Early CSE
 ; CHECK-NEXT:     Lower 'expect' Intrinsics
 ; CHECK-NEXT: Pass Arguments:
 ; CHECK-NEXT: Target Library Information
 ; CHECK-NEXT: Target Transform Information

llvm/test/Other/opt-O3-pipeline-enable-matrix.ll

 ; RUN: opt -enable-new-pm=0 -O3 -enable-matrix -debug-pass=Structure < %s -o /dev/null 2>&1 | FileCheck %s
 
 ; REQUIRES: asserts
 
 ; CHECK-LABEL: Pass Arguments:
 ; CHECK-NEXT: Target Transform Information
 ; CHECK-NEXT: Type-Based Alias Analysis
 ; CHECK-NEXT: Scoped NoAlias Alias Analysis
 ; CHECK-NEXT: Assumption Cache Tracker
 ; CHECK-NEXT: Target Library Information
 ; CHECK-NEXT:   FunctionPass Manager
 ; CHECK-NEXT:     Module Verifier
 ; CHECK-EXT:      Good Bye World Pass
 ; CHECK-NOEXT-NOT:      Good Bye World Pass
-; CHECK-NEXT:     Instrument function entry/exit with calls to e.g. mcount() (pre inlining)
 ; CHECK-NEXT:     Simplify the CFG
 ; CHECK-NEXT:     Dominator Tree Construction
 ; CHECK-NEXT:     SROA
 ; CHECK-NEXT:     Early CSE
 ; CHECK-NEXT:     Lower 'expect' Intrinsics
 ; CHECK-NEXT: Pass Arguments:
 ; CHECK-NEXT: Target Library Information
 ; CHECK-NEXT: Target Transform Information

llvm/test/Other/opt-O3-pipeline.ll

 ; RUN: opt -enable-new-pm=0 -mtriple=x86_64-- -O3 -debug-pass=Structure < %s -o /dev/null 2>&1 | FileCheck --check-prefixes=CHECK,%llvmcheckext %s
 
 ; REQUIRES: asserts
 
 ; CHECK-LABEL: Pass Arguments:
 ; CHECK-NEXT: Target Transform Information
 ; CHECK-NEXT: Type-Based Alias Analysis
 ; CHECK-NEXT: Scoped NoAlias Alias Analysis
 ; CHECK-NEXT: Assumption Cache Tracker
 ; CHECK-NEXT: Target Library Information
 ; CHECK-NEXT:   FunctionPass Manager
 ; CHECK-NEXT:     Module Verifier
 ; CHECK-EXT:      Good Bye World Pass
 ; CHECK-NOEXT-NOT:      Good Bye World Pass
-; CHECK-NEXT:     Instrument function entry/exit with calls to e.g. mcount() (pre inlining)
 ; CHECK-NEXT:     Simplify the CFG
 ; CHECK-NEXT:     Dominator Tree Construction
 ; CHECK-NEXT:     SROA
 ; CHECK-NEXT:     Early CSE
 ; CHECK-NEXT:     Lower 'expect' Intrinsics
 ; CHECK-NEXT: Pass Arguments:
 ; CHECK-NEXT: Target Library Information
 ; CHECK-NEXT: Target Transform Information

llvm/test/Other/opt-Os-pipeline.ll

 ; RUN: opt -enable-new-pm=0 -mtriple=x86_64-- -Os -debug-pass=Structure < %s -o /dev/null 2>&1 | FileCheck --check-prefixes=CHECK,%llvmcheckext %s
 
 ; REQUIRES: asserts
 
 ; CHECK-LABEL: Pass Arguments:
 ; CHECK-NEXT: Target Transform Information
 ; CHECK-NEXT: Type-Based Alias Analysis
 ; CHECK-NEXT: Scoped NoAlias Alias Analysis
 ; CHECK-NEXT: Assumption Cache Tracker
 ; CHECK-NEXT: Target Library Information
 ; CHECK-NEXT:   FunctionPass Manager
 ; CHECK-NEXT:     Module Verifier
 ; CHECK-EXT:      Good Bye World Pass
 ; CHECK-NOEXT-NOT:      Good Bye World Pass
-; CHECK-NEXT:     Instrument function entry/exit with calls to e.g. mcount() (pre inlining)
 ; CHECK-NEXT:     Simplify the CFG
 ; CHECK-NEXT:     Dominator Tree Construction
 ; CHECK-NEXT:     SROA
 ; CHECK-NEXT:     Early CSE
 ; CHECK-NEXT:     Lower 'expect' Intrinsics
 ; CHECK-NEXT: Pass Arguments:
 ; CHECK-NEXT: Target Library Information
 ; CHECK-NEXT: Target Transform Information