[InstrProfiling] Emit the runtime hook when no counters are lowered

[vsk]

Summary

The API verification tool tapi has difficulty processing frameworks
which enable code coverage, but which have no code. The profile lowering
pass does not emit its usual set of symbols in this case because no
counters are lowered.

While the symbols are not needed for program correctness, they are
needed to allow tapi to validate the exported symbol set of instrumented
binaries.

It was not possible to add a workaround in tapi for empty binaries due
to an architectural issue: tapi generates its expected symbol set before
it inspects a binary. Changing that model has a higher cost than simply
forcing llvm to emit the usual set of profiling-related symbols even
when there no counters are present.

rdar://36076904

[davidxl]

Can this be done under an option?

More fundamentally, do you just need to create a reference to the runtime hook variable? Is it possible to just unconditionally emit that while keep others unchanged? Or even better, is it possible to do the same linker trick as on linux -- to force undef of the variable on the command line?

[vsk]

In D43794#1020233, @davidxl wrote:

Can this be done under an option?

The change doesn't affect many programs, so istm that it's simpler to make it unconditional.

More fundamentally, do you just need to create a reference to the runtime hook variable? Is it possible to just unconditionally emit that while keep others unchanged?

Currently all we need is a reference to the runtime hook. I can reduce the scope of this patch down to just this part, and then add a test that the exported symbol set for an empty program looks good.

Or even better, is it possible to do the same linker trick as on linux -- to force undef of the variable on the command line?

It's possible with an option like -force_load, but this is guaranteed to break Darwin programs which intentionally circumvent the RegisterRuntime static initializer.

[vsk]

• Simplify the patch per David's feedback.
• Add an end-to-end test: D43844

[davidxl]

lgtm

llvm/trunk/include/llvm/Transforms/InstrProfiling.h

private:
 
   /// Emit value nodes section for value profiling.
   void emitVNodes();
 
   /// Emit runtime registration functions for each profile data variable.
   void emitRegistration();
 
   /// Emit the necessary plumbing to pull in the runtime initialization.
-  void emitRuntimeHook();
+  /// Returns true if a change was made.
+  bool emitRuntimeHook();
 
   /// Add uses of our data variables and runtime hook.
   void emitUses();
 
   /// Create a static initializer for our data, on platforms that need it,
   /// and for any profile output file that was specified.
   void emitInitialization();
 };
 
 } // end namespace llvm
 
 #endif // LLVM_TRANSFORMS_INSTRPROFILING_H

llvm/trunk/lib/Transforms/Instrumentation/InstrProfiling.cpp

static bool containsProfilingIntrinsics(Module &M) {
   if (auto *F = M.getFunction(
           Intrinsic::getName(llvm::Intrinsic::instrprof_value_profile)))
     if (!F->use_empty())
       return true;
   return false;
 }
 
 bool InstrProfiling::run(Module &M, const TargetLibraryInfo &TLI) {
-  // Improve compile time by avoiding linear scans when there is no work.
-  GlobalVariable *CoverageNamesVar =
-      M.getNamedGlobal(getCoverageUnusedNamesVarName());
-  if (!containsProfilingIntrinsics(M) && !CoverageNamesVar)
-    return false;
-
-  bool MadeChange = false;
-
   this->M = &M;
   this->TLI = &TLI;
   NamesVar = nullptr;
   NamesSize = 0;
   ProfileDataMap.clear();
   UsedVars.clear();
   getMemOPSizeRangeFromOption(MemOPSizeRange, MemOPSizeRangeStart,
                               MemOPSizeRangeLast);
   TT = Triple(M.getTargetTriple());
 
+  // Emit the runtime hook even if no counters are present.
+  bool MadeChange = emitRuntimeHook();
+
+  // Improve compile time by avoiding linear scans when there is no work.
+  GlobalVariable *CoverageNamesVar =
+      M.getNamedGlobal(getCoverageUnusedNamesVarName());
+  if (!containsProfilingIntrinsics(M) && !CoverageNamesVar)
+    return MadeChange;
+
   // We did not know how many value sites there would be inside
   // the instrumented function. This is counting the number of instrumented
   // target value sites to enter it as field in the profile data variable.
   for (Function &F : M) {
     InstrProfIncrementInst *FirstProfIncInst = nullptr;
     for (BasicBlock &BB : F)
       for (auto I = BB.begin(), E = BB.end(); I != E; I++)
         if (auto *Ind = dyn_cast<InstrProfValueProfileInst>(I))
bool InstrProfiling::run(Module &M, const TargetLibraryInfo &TLI) {
   }
 
   if (!MadeChange)
     return false;
 
   emitVNodes();
   emitNameData();
   emitRegistration();
-  emitRuntimeHook();
   emitUses();
   emitInitialization();
   return true;
 }
 
 static Constant *getOrInsertValueProfilingCall(Module &M,
                                                const TargetLibraryInfo &TLI,
                                                bool IsRange = false) {
auto *NamesRegisterF =
                          getInstrProfNamesRegFuncName(), M);
     IRB.CreateCall(NamesRegisterF, {IRB.CreateBitCast(NamesVar, VoidPtrTy),
                                     IRB.getInt64(NamesSize)});
   }
 
   IRB.CreateRetVoid();
 }
 
-void InstrProfiling::emitRuntimeHook() {
+bool InstrProfiling::emitRuntimeHook() {
   // We expect the linker to be invoked with -u<hook_var> flag for linux,
   // for which case there is no need to emit the user function.
   if (Triple(M->getTargetTriple()).isOSLinux())
-    return;
+    return false;
 
   // If the module's provided its own runtime, we don't need to do anything.
   if (M->getGlobalVariable(getInstrProfRuntimeHookVarName()))
-    return;
+    return false;
 
   // Declare an external variable that will pull in the runtime initialization.
   auto *Int32Ty = Type::getInt32Ty(M->getContext());
   auto *Var =
       new GlobalVariable(*M, Int32Ty, false, GlobalValue::ExternalLinkage,
                          nullptr, getInstrProfRuntimeHookVarName());
 
   // Make a function that uses it.
   auto *User = Function::Create(FunctionType::get(Int32Ty, false),
                                 GlobalValue::LinkOnceODRLinkage,
                                 getInstrProfRuntimeHookVarUseFuncName(), M);
   User->addFnAttr(Attribute::NoInline);
   if (Options.NoRedZone)
     User->addFnAttr(Attribute::NoRedZone);
   User->setVisibility(GlobalValue::HiddenVisibility);
   if (Triple(M->getTargetTriple()).supportsCOMDAT())
     User->setComdat(M->getOrInsertComdat(User->getName()));
 
   IRBuilder<> IRB(BasicBlock::Create(M->getContext(), "", User));
   auto *Load = IRB.CreateLoad(Var);
   IRB.CreateRet(Load);
 
   // Mark the user variable as used so that it isn't stripped out.
   UsedVars.push_back(User);
+  return true;
 }
 
 void InstrProfiling::emitUses() {
   if (!UsedVars.empty())
     appendToUsed(*M, UsedVars);
 }
 
 void InstrProfiling::emitInitialization() {

llvm/trunk/test/Instrumentation/InstrProfiling/linkage.ll

 ;; Check that runtime symbols get appropriate linkage.
 
 ; RUN: opt < %s -mtriple=x86_64-apple-macosx10.10.0 -instrprof -S | FileCheck %s --check-prefix=OTHER --check-prefix=COMMON
 ; RUN: opt < %s -mtriple=x86_64-unknown-linux -instrprof -S | FileCheck %s --check-prefix=LINUX --check-prefix=COMMON
 ; RUN: opt < %s -mtriple=x86_64-apple-macosx10.10.0 -passes=instrprof -S | FileCheck %s --check-prefix=OTHER --check-prefix=COMMON
 ; RUN: opt < %s -mtriple=x86_64-unknown-linux -passes=instrprof -S | FileCheck %s --check-prefix=LINUX --check-prefix=COMMON
 ; RUN: opt < %s  -mtriple=x86_64-pc-win32-coff -instrprof -S | FileCheck %s --check-prefix=COFF
 ; RUN: opt < %s  -mtriple=x86_64-pc-win32-coff -passes=instrprof -S | FileCheck %s --check-prefix=COFF
 
+; OTHER: @__llvm_profile_runtime = external global i32
+; LINUX-NOT: @__llvm_profile_runtime = external global i32
+
 @__profn_foo = hidden constant [3 x i8] c"foo"
 @__profn_foo_weak = weak hidden constant [8 x i8] c"foo_weak"
 @"__profn_linkage.ll:foo_internal" = internal constant [23 x i8] c"linkage.ll:foo_internal"
 @__profn_foo_inline = linkonce_odr hidden constant [10 x i8] c"foo_inline"
 @__profn_foo_extern = linkonce_odr hidden constant [10 x i8] c"foo_extern"
 
 ; COMMON: @__profc_foo = hidden global
 ; COMMON: @__profd_foo = hidden global

 ; OTHER: @__profd_foo_extern = linkonce_odr hidden global
 define available_externally void @foo_extern() {
   call void @llvm.instrprof.increment(i8* getelementptr inbounds ([10 x i8], [10 x i8]* @__profn_foo_extern, i32 0, i32 0), i64 0, i32 1, i32 0)
   ret void
 }
 
 declare void @llvm.instrprof.increment(i8*, i64, i32, i32)
 
-; OTHER: @__llvm_profile_runtime = external global i32
-; LINUX-NOT: @__llvm_profile_runtime = external global i32
-
 ; OTHER: define linkonce_odr hidden i32 @__llvm_profile_runtime_user() {{.*}} {
 ; OTHER:   %[[REG:.*]] = load i32, i32* @__llvm_profile_runtime
 ; OTHER:   ret i32 %[[REG]]
 ; OTHER: }
 ; COFF: define linkonce_odr hidden i32 @__llvm_profile_runtime_user() {{.*}} comdat {
 ; LINUX-NOT: define linkonce_odr hidden i32 @__llvm_profile_runtime_user() {{.*}} {
 ; LINUX-NOT:   %[[REG:.*]] = load i32, i32* @__llvm_profile_runtime

llvm/trunk/test/Instrumentation/InstrProfiling/no-counters.ll

-;; No instrumentation should be emitted if there are no counter increments.
+;; Emit the runtime hook even if there are no counter increments.
 
-; RUN: opt < %s -instrprof -S | FileCheck %s
-; RUN: opt < %s -passes=instrprof -S | FileCheck %s
-; CHECK-NOT: @__profc
-; CHECK-NOT: @__profd
-; CHECK-NOT: @__llvm_profile_runtime
+; RUN: opt < %s -mtriple=x86_64-apple-macosx10.10.0 -instrprof -S | FileCheck %s -check-prefixes=ALL,DARWIN
+; RUN: opt < %s -mtriple=x86_64-apple-macosx10.10.0 -passes=instrprof -S | FileCheck %s -check-prefixes=ALL,DARWIN
+; RUN: opt < %s -mtriple=x86_64-linux-unknown -passes=instrprof -S | FileCheck %s -check-prefixes=ALL,LINUX
+; ALL-NOT: @__profc
+; ALL-NOT: @__profd
+; DARWIN: @__llvm_profile_runtime
+; LINUX-NOT: @__llvm_profile_runtime
 
 define void @foo() {
   ret void
 }

llvm/trunk/test/Instrumentation/InstrProfiling/profiling.ll

 ; RUN: opt < %s -instrprof -S | FileCheck %s
 ; RUN: opt < %s -passes=instrprof -S | FileCheck %s
 
 target triple = "x86_64-apple-macosx10.10.0"
 
+; CHECK: @__llvm_profile_runtime = external global i32
+
 @__profn_foo = hidden constant [3 x i8] c"foo"
 ; CHECK-NOT: __profn_foo
 @__profn_bar = hidden constant [4 x i8] c"bar\00"
 ; CHECK-NOT: __profn_bar
 @__profn_baz = hidden constant [3 x i8] c"baz"
 ; CHECK-NOT: __profn_baz
 
 ; CHECK: @__profc_foo = hidden global [1 x i64] zeroinitializer, section "__DATA,__llvm_prf_cnts", align 8
define void @baz() {
   call void @llvm.instrprof.increment(i8* getelementptr inbounds ([3 x i8], [3 x i8]* @__profn_baz, i32 0, i32 0), i64 0, i32 3, i32 0)
   call void @llvm.instrprof.increment(i8* getelementptr inbounds ([3 x i8], [3 x i8]* @__profn_baz, i32 0, i32 0), i64 0, i32 3, i32 1)
   call void @llvm.instrprof.increment(i8* getelementptr inbounds ([3 x i8], [3 x i8]* @__profn_baz, i32 0, i32 0), i64 0, i32 3, i32 2)
   ret void
 }
 
 declare void @llvm.instrprof.increment(i8*, i64, i32, i32)
 
-; CHECK: @__llvm_profile_runtime = external global i32
 ; CHECK: @llvm.used = appending global {{.*}} @__profd_foo {{.*}} @__profd_bar {{.*}} @__profd_baz {{.*}} section "llvm.metadata"