[PATCH] D62627: [NFC] Do not run CGProfilePass when not using integrated assembler

[Zhizhou Yang via Phabricator via cfe-commits]

zhizhouy updated this revision to Diff 252143.
zhizhouy marked 4 inline comments as done.
zhizhouy retitled this revision from "[NFC] Do not run CGProfilePass when 
-fno-integrated-as is on" to "[NFC] Do not run CGProfilePass when not using 
integrated assembler".
zhizhouy added a comment.

Thanks for the comments.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D62627/new/

https://reviews.llvm.org/D62627

Files:
  clang/include/clang/Basic/CodeGenOptions.def
  clang/lib/CodeGen/BackendUtil.cpp
  clang/lib/Frontend/CompilerInvocation.cpp
  llvm/include/llvm/Passes/PassBuilder.h
  llvm/lib/Passes/PassBuilder.cpp
  llvm/test/Other/new-pm-cgprofile.ll


Index: llvm/test/Other/new-pm-cgprofile.ll
===================================================================
--- /dev/null
+++ llvm/test/Other/new-pm-cgprofile.ll
@@ -0,0 +1,11 @@
+; RUN: opt -debug-pass-manager -passes='default<O2>' %s 2>&1 |FileCheck %s --check-prefixes=DEFAULT
+; RUN: opt -debug-pass-manager -passes='default<O2>' -enable-npm-call-graph-profile=0 %s 2>&1 |FileCheck %s --check-prefixes=OFF
+; RUN: opt -debug-pass-manager -passes='default<O2>' -enable-npm-call-graph-profile=1 %s 2>&1 |FileCheck %s --check-prefixes=ON
+;
+; DEFAULT: Running pass: CGProfilePass
+; OFF-NOT: Running pass: CGProfilePass
+; ON: Running pass: CGProfilePass
+
+define void @foo() {
+  ret void
+}
Index: llvm/lib/Passes/PassBuilder.cpp
===================================================================
--- llvm/lib/Passes/PassBuilder.cpp
+++ llvm/lib/Passes/PassBuilder.cpp
@@ -237,6 +237,10 @@
     EnableCHR("enable-chr-npm", cl::init(true), cl::Hidden,
               cl::desc("Enable control height reduction optimization (CHR)"));
 
+static cl::opt<bool> EnableCallGraphProfile(
+    "enable-npm-call-graph-profile", cl::init(true), cl::Hidden,
+    cl::desc("Enable call graph profile pass for the new PM (default = on)"));
+
 PipelineTuningOptions::PipelineTuningOptions() {
   LoopInterleaving = EnableLoopInterleaving;
   LoopVectorization = EnableLoopVectorization;
@@ -246,6 +250,7 @@
   Coroutines = false;
   LicmMssaOptCap = SetLicmMssaOptCap;
   LicmMssaNoAccForPromotionCap = SetLicmMssaNoAccForPromotionCap;
+  CallGraphProfile = EnableCallGraphProfile;
 }
 
 extern cl::opt<bool> EnableHotColdSplit;
@@ -1060,7 +1065,8 @@
   // Add the core optimizing pipeline.
   MPM.addPass(createModuleToFunctionPassAdaptor(std::move(OptimizePM)));
 
-  MPM.addPass(CGProfilePass());
+  if (PTO.CallGraphProfile)
+    MPM.addPass(CGProfilePass());
 
   // Now we need to do some global optimization transforms.
   // FIXME: It would seem like these should come first in the optimization
Index: llvm/include/llvm/Passes/PassBuilder.h
===================================================================
--- llvm/include/llvm/Passes/PassBuilder.h
+++ llvm/include/llvm/Passes/PassBuilder.h
@@ -105,6 +105,10 @@
   /// Tuning option to disable promotion to scalars in LICM with MemorySSA, if
   /// the number of access is too large.
   unsigned LicmMssaNoAccForPromotionCap;
+
+  /// Tuning option to enable/disable call graph profile. Its default value is
+  /// that of the flag: `-enable-npm-call-graph-profile`.
+  bool CallGraphProfile;
 };
 
 /// This class provides access to building LLVM's passes.
Index: clang/lib/Frontend/CompilerInvocation.cpp
===================================================================
--- clang/lib/Frontend/CompilerInvocation.cpp
+++ clang/lib/Frontend/CompilerInvocation.cpp
@@ -823,6 +823,7 @@
   Opts.RerollLoops = Args.hasArg(OPT_freroll_loops);
 
   Opts.DisableIntegratedAS = Args.hasArg(OPT_fno_integrated_as);
+  Opts.CallGraphProfile = !Opts.DisableIntegratedAS;
   Opts.Autolink = !Args.hasArg(OPT_fno_autolink);
   Opts.SampleProfileFile =
       std::string(Args.getLastArgValue(OPT_fprofile_sample_use_EQ));
Index: clang/lib/CodeGen/BackendUtil.cpp
===================================================================
--- clang/lib/CodeGen/BackendUtil.cpp
+++ clang/lib/CodeGen/BackendUtil.cpp
@@ -1107,6 +1107,7 @@
   PTO.LoopInterleaving = CodeGenOpts.UnrollLoops;
   PTO.LoopVectorization = CodeGenOpts.VectorizeLoop;
   PTO.SLPVectorization = CodeGenOpts.VectorizeSLP;
+  PTO.CallGraphProfile = CodeGenOpts.CallGraphProfile;
   PTO.Coroutines = LangOpts.Coroutines;
 
   PassInstrumentationCallbacks PIC;
@@ -1514,6 +1515,7 @@
   Conf.PTO.LoopInterleaving = CGOpts.UnrollLoops;
   Conf.PTO.LoopVectorization = CGOpts.VectorizeLoop;
   Conf.PTO.SLPVectorization = CGOpts.VectorizeSLP;
+  Conf.PTO.CallGraphProfile = CGOpts.CallGraphProfile;
 
   // Context sensitive profile.
   if (CGOpts.hasProfileCSIRInstr()) {
Index: clang/include/clang/Basic/CodeGenOptions.def
===================================================================
--- clang/include/clang/Basic/CodeGenOptions.def
+++ clang/include/clang/Basic/CodeGenOptions.def
@@ -254,6 +254,7 @@
 CODEGENOPT(VectorizeLoop     , 1, 0) ///< Run loop vectorizer.
 CODEGENOPT(VectorizeSLP      , 1, 0) ///< Run SLP vectorizer.
 CODEGENOPT(ProfileSampleAccurate, 1, 0) ///< Sample profile is accurate.
+CODEGENOPT(CallGraphProfile  , 1, 0) ///< Run call graph profile.
 
   /// Attempt to use register sized accesses to bit-fields in structures, when
   /// possible.

_______________________________________________
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits