Summary: SampleProfile pass needs to be performed after InstructionCombiningPass, which helps eliminate un-inlinable function calls.
Reviewers: davidxl, dnovillo
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D17742
llvm-svn: 262419
PreservedAnalyses run(Function &F, AnalysisManager<Function> *AM);
};
+/// \brief The legacy pass manager's instcombine pass.
+///
+/// This is a basic whole-function wrapper around the instcombine utility. It
+/// will try to combine all instructions in the function.
+class InstructionCombiningPass : public FunctionPass {
+ InstCombineWorklist Worklist;
+
+public:
+ static char ID; // Pass identification, replacement for typeid
+
+ InstructionCombiningPass() : FunctionPass(ID) {
+ initializeInstructionCombiningPassPass(*PassRegistry::getPassRegistry());
+ }
+
+ void getAnalysisUsage(AnalysisUsage &AU) const override;
+ bool runOnFunction(Function &F) override;
+};
}
#endif
#include "llvm/Support/raw_ostream.h"
#include "llvm/Transforms/IPO.h"
#include "llvm/Transforms/Utils/Cloning.h"
+#include "llvm/Transforms/InstCombine/InstCombine.h"
#include <cctype>
using namespace llvm;
bool runOnModule(Module &M) override;
void getAnalysisUsage(AnalysisUsage &AU) const override {
- AU.setPreservesCFG();
+ AU.addRequired<InstructionCombiningPass>();
}
protected:
INITIALIZE_PASS_BEGIN(SampleProfileLoader, "sample-profile",
"Sample Profile loader", false, false)
INITIALIZE_PASS_DEPENDENCY(AddDiscriminators)
+INITIALIZE_PASS_DEPENDENCY(InstructionCombiningPass)
INITIALIZE_PASS_END(SampleProfileLoader, "sample-profile",
"Sample Profile loader", false, false)
bool SampleProfileLoader::runOnFunction(Function &F) {
F.setEntryCount(0);
+ getAnalysis<InstructionCombiningPass>(F);
Samples = Reader->getSamplesFor(F);
if (!Samples->empty())
return emitAnnotations(F);
return PA;
}
-namespace {
-/// \brief The legacy pass manager's instcombine pass.
-///
-/// This is a basic whole-function wrapper around the instcombine utility. It
-/// will try to combine all instructions in the function.
-class InstructionCombiningPass : public FunctionPass {
- InstCombineWorklist Worklist;
-
-public:
- static char ID; // Pass identification, replacement for typeid
-
- InstructionCombiningPass() : FunctionPass(ID) {
- initializeInstructionCombiningPassPass(*PassRegistry::getPassRegistry());
- }
-
- void getAnalysisUsage(AnalysisUsage &AU) const override;
- bool runOnFunction(Function &F) override;
-};
-}
-
void InstructionCombiningPass::getAnalysisUsage(AnalysisUsage &AU) const {
AU.setPreservesCFG();
AU.addRequired<AAResultsWrapperPass>();
--- /dev/null
+foo:1000:1000
+ 1: bar:1000
; both branches out of while.body had the same weight. In reality,
; the edge while.body->if.then is taken most of the time.
;
-; CHECK: edge while.body -> if.then probability is 0x80000000 / 0x80000000 = 100.00% [HOT edge]
; CHECK: edge while.body -> if.else probability is 0x00000000 / 0x80000000 = 0.00%
+; CHECK: edge while.body -> if.then probability is 0x80000000 / 0x80000000 = 100.00% [HOT edge]
if.then: ; preds = %while.body
; RUN: opt < %s -sample-profile -sample-profile-file=%S/Inputs/cov-zero-samples.prof -sample-profile-check-record-coverage=100 -pass-remarks=sample-profile -o /dev/null 2>&1 | FileCheck %s
;
-; CHECK: remark: cov-zero-samples.cc:9:25: Applied 404065 samples from profile (offset: 2.1)
+; CHECK: remark: cov-zero-samples.cc:9:29: Applied 404065 samples from profile (offset: 2.1)
; CHECK: remark: cov-zero-samples.cc:10:9: Applied 443089 samples from profile (offset: 3)
; CHECK: remark: cov-zero-samples.cc:10:36: Applied 0 samples from profile (offset: 3.1)
; CHECK: remark: cov-zero-samples.cc:11:12: Applied 404066 samples from profile (offset: 4)
--- /dev/null
+; RUN: opt < %s -sample-profile -sample-profile-file=%S/Inputs/inline-combine.prof -S | FileCheck %s
+
+%"class.llvm::FoldingSetNodeID" = type { %"class.llvm::SmallVector" }
+%"class.llvm::SmallVector" = type { %"class.llvm::SmallVectorImpl.base", %"struct.llvm::SmallVectorStorage" }
+%"class.llvm::SmallVectorImpl.base" = type { %"class.llvm::SmallVectorTemplateBase.base" }
+%"class.llvm::SmallVectorTemplateBase.base" = type { %"class.llvm::SmallVectorTemplateCommon.base" }
+%"class.llvm::SmallVectorTemplateCommon.base" = type <{ %"class.llvm::SmallVectorBase", %"struct.llvm::AlignedCharArrayUnion" }>
+%"class.llvm::SmallVectorBase" = type { i8*, i8*, i8* }
+%"struct.llvm::AlignedCharArrayUnion" = type { %"struct.llvm::AlignedCharArray" }
+%"struct.llvm::AlignedCharArray" = type { [4 x i8] }
+%"struct.llvm::SmallVectorStorage" = type { [31 x %"struct.llvm::AlignedCharArrayUnion"] }
+%"class.llvm::SmallVectorImpl" = type { %"class.llvm::SmallVectorTemplateBase.base", [4 x i8] }
+
+$foo = comdat any
+
+$bar = comdat any
+
+define void @foo(%"class.llvm::FoldingSetNodeID"* %this) comdat align 2 !dbg !3 {
+ %1 = alloca %"class.llvm::FoldingSetNodeID"*, align 8
+ store %"class.llvm::FoldingSetNodeID"* %this, %"class.llvm::FoldingSetNodeID"** %1, align 8
+ %2 = load %"class.llvm::FoldingSetNodeID"*, %"class.llvm::FoldingSetNodeID"** %1, align 8
+ %3 = getelementptr inbounds %"class.llvm::FoldingSetNodeID", %"class.llvm::FoldingSetNodeID"* %2, i32 0, i32 0
+; the call should have been inlined after sample-profile pass
+; CHECK-NOT: call
+ call void bitcast (void (%"class.llvm::SmallVectorImpl"*)* @bar to void (%"class.llvm::SmallVector"*)*)(%"class.llvm::SmallVector"* %3), !dbg !7
+ ret void
+}
+
+define void @bar(%"class.llvm::SmallVectorImpl"* %this) comdat align 2 !dbg !8 {
+ ret void
+}
+
+!llvm.module.flags = !{!0, !1}
+!llvm.ident = !{!2}
+
+!0 = !{i32 2, !"Dwarf Version", i32 4}
+!1 = !{i32 1, !"Debug Info Version", i32 3}
+!2 = !{!"clang version 3.5 "}
+!3 = distinct !DISubprogram(name: "foo", scope: !4, file: !4, line: 3, type: !5, isLocal: false, isDefinition: true, scopeLine: 3, virtualIndex: 6, flags: DIFlagPrototyped, isOptimized: false, variables: !6)
+!4 = !DIFile(filename: "test.cc", directory: ".")
+!5 = !DISubroutineType(types: !6)
+!6 = !{}
+!7 = !DILocation(line: 4, scope: !3)
+!8 = distinct !DISubprogram(name: "bar", scope: !4, file: !4, line: 7, type: !5, isLocal: false, isDefinition: true, scopeLine: 7, virtualIndex: 6, flags: DIFlagPrototyped, isOptimized: false, variables: !6)
; 12 }
;
; CHECK: remark: coverage.cc:10:12: inlined hot callee '_Z3fool' with 172746 samples into 'main'
-; CHECK: remark: coverage.cc:9:19: Applied 23478 samples from profile (offset: 2.1)
+; CHECK: remark: coverage.cc:9:21: Applied 23478 samples from profile (offset: 2.1)
; CHECK: remark: coverage.cc:10:16: Applied 23478 samples from profile (offset: 3)
; CHECK: remark: coverage.cc:4:10: Applied 31878 samples from profile (offset: 1)
; CHECK: remark: coverage.cc:11:10: Applied 0 samples from profile (offset: 4)