phosek created this revision.
phosek added reviewers: vsk, davidxl.
Herald added a subscriber: hiraditya.
phosek requested review of this revision.
Herald added projects: clang, LLVM.
Herald added subscribers: llvm-commits, cfe-commits.
When using -fprofile-list to selectively apply instrumentation only
to certain files or functions, we may end up with a binary that doesn't
have any counters in the case where no files were selected. However,
because on Linux and Fuchsia, we pass -u__llvm_profile_runtime, the
runtime would still be pulled in and incur some non-trivial overhead,
especially in the case when the continuous or runtime counter relocation
mode is being used. A better way would be to pull in the profile runtime
only when needed by declaring the __llvm_profile_runtime symbol in the
translation unit only when needed.
This approach was already used prior to 9a041a75221ca, but we changed it
to always generate the __llvm_profile_runtime due to a TAPI limitation.
Since TAPI is only used on Mach-O platforms, we could use the early
emission of __llvm_profile_runtime there, and on other platforms we
could change back to the earlier approach where the symbol is generated
later only when needed. We can stop passing -u__llvm_profile_runtime to
the linker on Linux and Fuchsia since the generated undefined symbol in
each translation unit that needed it serves the same purpose.
Repository:
rG LLVM Github Monorepo
https://reviews.llvm.org/D98061
Files:
clang/lib/Driver/ToolChains/Fuchsia.cpp
clang/lib/Driver/ToolChains/Fuchsia.h
clang/lib/Driver/ToolChains/Linux.cpp
clang/lib/Driver/ToolChains/Linux.h
clang/test/Driver/coverage-ld.c
clang/test/Driver/fuchsia.c
llvm/lib/Transforms/Instrumentation/InstrProfiling.cpp
llvm/test/Instrumentation/InstrProfiling/linkage.ll
llvm/test/Instrumentation/InstrProfiling/profiling.ll
Index: llvm/test/Instrumentation/InstrProfiling/profiling.ll
===================================================================
--- llvm/test/Instrumentation/InstrProfiling/profiling.ll
+++ llvm/test/Instrumentation/InstrProfiling/profiling.ll
@@ -5,10 +5,7 @@
; RUN: opt < %s -mtriple=x86_64-apple-macosx10.10.0 -instrprof -S | FileCheck %s
-; ELF_GENERIC: @__llvm_profile_runtime = external global i32
-; ELF_LINUX-NOT: @__llvm_profile_runtime
; MACHO: @__llvm_profile_runtime = external global i32
-; WIN: @__llvm_profile_runtime = external global i32
@__profn_foo = hidden constant [3 x i8] c"foo"
; CHECK-NOT: __profn_foo
@@ -54,9 +51,11 @@
declare void @llvm.instrprof.increment(i8*, i64, i32, i32)
-; ELF: @llvm.compiler.used = appending global {{.*}} @__llvm_profile_runtime_user {{.*}} @__profd_foo {{.*}} @__profd_bar {{.*}} @__profd_baz
+; ELF: @__llvm_profile_runtime = external global i32
+; ELF: @llvm.compiler.used = appending global {{.*}} @__profd_foo {{.*}} @__profd_bar {{.*}} @__profd_baz {{.*}} @__llvm_profile_runtime
; MACHO: @llvm.used = appending global {{.*}} @__llvm_profile_runtime_user {{.*}} @__profd_foo {{.*}} @__profd_bar {{.*}} @__profd_baz
-; WIN: @llvm.used = appending global {{.*}} @__llvm_profile_runtime_user {{.*}} @__profd_foo {{.*}} @__profd_bar {{.*}} @__profd_baz
+; WIN: @__llvm_profile_runtime = external global i32
+; WIN: @llvm.used = appending global {{.*}} @__profd_foo {{.*}} @__profd_bar {{.*}} @__profd_baz {{.*}} @__llvm_profile_runtime_user
; ELF_GENERIC: define internal void @__llvm_profile_register_functions() unnamed_addr {
; ELF_GENERIC-NEXT: call void @__llvm_profile_register_function(i8* bitcast ({ i64, i64, i64*, i8*, i8*, i32, [2 x i16] }* @__profd_foo to i8*))
Index: llvm/test/Instrumentation/InstrProfiling/linkage.ll
===================================================================
--- llvm/test/Instrumentation/InstrProfiling/linkage.ll
+++ llvm/test/Instrumentation/InstrProfiling/linkage.ll
@@ -10,7 +10,6 @@
; RUN: opt < %s -mtriple=x86_64-pc-win32-coff -passes=instrprof -S | FileCheck %s --check-prefixes=COFF
; MACHO: @__llvm_profile_runtime = external global i32
-; ELF-NOT: @__llvm_profile_runtime = external global i32
; ELF: $__profd_foo = comdat noduplicates
; ELF: $__profd_foo_weak = comdat noduplicates
@@ -87,5 +86,6 @@
; MACHO: ret i32 %[[REG]]
; MACHO: }
; COFF: define linkonce_odr hidden i32 @__llvm_profile_runtime_user() {{.*}} comdat {
+; ELF: @__llvm_profile_runtime = external global i32
; ELF-NOT: define linkonce_odr hidden i32 @__llvm_profile_runtime_user() {{.*}} {
; ELF-NOT: %[[REG:.*]] = load i32, i32* @__llvm_profile_runtime
Index: llvm/lib/Transforms/Instrumentation/InstrProfiling.cpp
===================================================================
--- llvm/lib/Transforms/Instrumentation/InstrProfiling.cpp
+++ llvm/lib/Transforms/Instrumentation/InstrProfiling.cpp
@@ -543,8 +543,10 @@
UsedVars.clear();
TT = Triple(M.getTargetTriple());
- // Emit the runtime hook even if no counters are present.
- bool MadeChange = emitRuntimeHook();
+ bool MadeChange = false;
+ // Emit the runtime hook even if no counters are present in Mach-O.
+ if (TT.isOSBinFormatMachO())
+ MadeChange = emitRuntimeHook();
// Improve compile time by avoiding linear scans when there is no work.
GlobalVariable *CoverageNamesVar =
@@ -584,6 +586,8 @@
emitVNodes();
emitNameData();
emitRegistration();
+ if (!TT.isOSBinFormatMachO())
+ emitRuntimeHook();
emitUses();
emitInitialization();
return true;
@@ -1058,11 +1062,6 @@
}
bool InstrProfiling::emitRuntimeHook() {
- // We expect the linker to be invoked with -u<hook_var> flag for Linux or
- // Fuchsia, in which case there is no need to emit the user function.
- if (TT.isOSLinux() || TT.isOSFuchsia())
- return false;
-
// If the module's provided its own runtime, we don't need to do anything.
if (M->getGlobalVariable(getInstrProfRuntimeHookVarName()))
return false;
@@ -1073,6 +1072,12 @@
new GlobalVariable(*M, Int32Ty, false, GlobalValue::ExternalLinkage,
nullptr, getInstrProfRuntimeHookVarName());
+ if (TT.isOSBinFormatELF()) {
+ // Mark the user variable as used so that it isn't stripped out.
+ CompilerUsedVars.push_back(Var);
+ return true;
+ }
+
// Make a function that uses it.
auto *User = Function::Create(FunctionType::get(Int32Ty, false),
GlobalValue::LinkOnceODRLinkage,
@@ -1088,7 +1093,7 @@
auto *Load = IRB.CreateLoad(Int32Ty, Var);
IRB.CreateRet(Load);
- // Mark the user variable as used so that it isn't stripped out.
+ // Mark the function as used so that it isn't stripped out.
CompilerUsedVars.push_back(User);
return true;
}
Index: clang/test/Driver/fuchsia.c
===================================================================
--- clang/test/Driver/fuchsia.c
+++ clang/test/Driver/fuchsia.c
@@ -249,7 +249,6 @@
// RUN: -fuse-ld=lld 2>&1 \
// RUN: | FileCheck %s -check-prefix=CHECK-PROFRT-AARCH64
// CHECK-PROFRT-AARCH64: "-resource-dir" "[[RESOURCE_DIR:[^"]+]]"
-// CHECK-PROFRT-AARCH64: "-u__llvm_profile_runtime"
// CHECK-PROFRT-AARCH64: "[[RESOURCE_DIR]]{{/|\\\\}}lib{{/|\\\\}}aarch64-fuchsia{{/|\\\\}}libclang_rt.profile.a"
// RUN: %clang %s -### --target=x86_64-fuchsia \
@@ -258,5 +257,4 @@
// RUN: -fuse-ld=lld 2>&1 \
// RUN: | FileCheck %s -check-prefix=CHECK-PROFRT-X86_64
// CHECK-PROFRT-X86_64: "-resource-dir" "[[RESOURCE_DIR:[^"]+]]"
-// CHECK-PROFRT-X86_64: "-u__llvm_profile_runtime"
// CHECK-PROFRT-X86_64: "[[RESOURCE_DIR]]{{/|\\\\}}lib{{/|\\\\}}x86_64-fuchsia{{/|\\\\}}libclang_rt.profile.a"
Index: clang/test/Driver/coverage-ld.c
===================================================================
--- clang/test/Driver/coverage-ld.c
+++ clang/test/Driver/coverage-ld.c
@@ -12,9 +12,7 @@
// RUN: --sysroot=%S/Inputs/basic_linux_tree \
// RUN: | FileCheck --check-prefix=CHECK-LINUX-I386 %s
//
-// CHECK-LINUX-I386-NOT: "-u__llvm_profile_runtime"
// CHECK-LINUX-I386: /Inputs/resource_dir{{/|\\\\}}lib{{/|\\\\}}linux{{/|\\\\}}libclang_rt.profile-i386.a"
-// CHECK-LINUX-I386-NOT: "-u__llvm_profile_runtime"
// CHECK-LINUX-I386: "-lc"
//
// RUN: %clang -no-canonical-prefixes %s -### -o %t.o 2>&1 \
Index: clang/lib/Driver/ToolChains/Linux.h
===================================================================
--- clang/lib/Driver/ToolChains/Linux.h
+++ clang/lib/Driver/ToolChains/Linux.h
@@ -42,8 +42,6 @@
bool isNoExecStackDefault() const override;
bool IsMathErrnoDefault() const override;
SanitizerMask getSupportedSanitizers() const override;
- void addProfileRTLibs(const llvm::opt::ArgList &Args,
- llvm::opt::ArgStringList &CmdArgs) const override;
std::string computeSysRoot() const override;
std::string getDynamicLinker(const llvm::opt::ArgList &Args) const override;
Index: clang/lib/Driver/ToolChains/Linux.cpp
===================================================================
--- clang/lib/Driver/ToolChains/Linux.cpp
+++ clang/lib/Driver/ToolChains/Linux.cpp
@@ -905,16 +905,6 @@
return Res;
}
-void Linux::addProfileRTLibs(const llvm::opt::ArgList &Args,
- llvm::opt::ArgStringList &CmdArgs) const {
- // Add linker option -u__llvm_profile_runtime to cause runtime
- // initialization module to be linked in.
- if (needsProfileRT(Args))
- CmdArgs.push_back(Args.MakeArgString(
- Twine("-u", llvm::getInstrProfRuntimeHookVarName())));
- ToolChain::addProfileRTLibs(Args, CmdArgs);
-}
-
llvm::DenormalMode
Linux::getDefaultDenormalModeForType(const llvm::opt::ArgList &DriverArgs,
const JobAction &JA,
Index: clang/lib/Driver/ToolChains/Fuchsia.h
===================================================================
--- clang/lib/Driver/ToolChains/Fuchsia.h
+++ clang/lib/Driver/ToolChains/Fuchsia.h
@@ -71,9 +71,6 @@
SanitizerMask getSupportedSanitizers() const override;
SanitizerMask getDefaultSanitizers() const override;
- void addProfileRTLibs(const llvm::opt::ArgList &Args,
- llvm::opt::ArgStringList &CmdArgs) const override;
-
RuntimeLibType
GetRuntimeLibType(const llvm::opt::ArgList &Args) const override;
CXXStdlibType
Index: clang/lib/Driver/ToolChains/Fuchsia.cpp
===================================================================
--- clang/lib/Driver/ToolChains/Fuchsia.cpp
+++ clang/lib/Driver/ToolChains/Fuchsia.cpp
@@ -394,13 +394,3 @@
}
return Res;
}
-
-void Fuchsia::addProfileRTLibs(const llvm::opt::ArgList &Args,
- llvm::opt::ArgStringList &CmdArgs) const {
- // Add linker option -u__llvm_profile_runtime to cause runtime
- // initialization module to be linked in.
- if (needsProfileRT(Args))
- CmdArgs.push_back(Args.MakeArgString(
- Twine("-u", llvm::getInstrProfRuntimeHookVarName())));
- ToolChain::addProfileRTLibs(Args, CmdArgs);
-}
_______________________________________________
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
