[CodeExtractor] Update function's assumption cache after extracting blocks from it
Summary: Assumption cache's self-updating mechanism does not correctly handle the case when blocks are extracted from the function by the CodeExtractor. As a result function's assumption cache may have stale references to the llvm.assume calls that were moved to the outlined function. This patch fixes this problem by removing extracted llvm.assume calls from the function’s assumption cache.
Reviewers: hfinkel, vsk, fhahn, davidxl, sanjoy
Reviewed By: hfinkel, vsk
Subscribers: llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D57215
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@353500 91177308-0d34-0410-b5e6-96231b3b80d8
diff --git a/include/llvm/Analysis/AssumptionCache.h b/include/llvm/Analysis/AssumptionCache.h
index 93dbae8..b428464 100644
--- a/include/llvm/Analysis/AssumptionCache.h
+++ b/include/llvm/Analysis/AssumptionCache.h
@@ -103,6 +103,10 @@
/// not already be in the cache.
void registerAssumption(CallInst *CI);
+ /// Remove an \@llvm.assume intrinsic from this function's cache if it has
+ /// been added to the cache earlier.
+ void unregisterAssumption(CallInst *CI);
+
/// Update the cache of values being affected by this assumption (i.e.
/// the values about which this assumption provides information).
void updateAffectedValues(CallInst *CI);
@@ -208,6 +212,10 @@
/// existing cache will be returned.
AssumptionCache &getAssumptionCache(Function &F);
+ /// Return the cached assumptions for a function if it has already been
+ /// scanned. Otherwise return nullptr.
+ AssumptionCache *lookupAssumptionCache(Function &F);
+
AssumptionCacheTracker();
~AssumptionCacheTracker() override;
diff --git a/include/llvm/Transforms/Utils/CodeExtractor.h b/include/llvm/Transforms/Utils/CodeExtractor.h
index e13940b..becbf0e 100644
--- a/include/llvm/Transforms/Utils/CodeExtractor.h
+++ b/include/llvm/Transforms/Utils/CodeExtractor.h
@@ -26,6 +26,7 @@
class BlockFrequency;
class BlockFrequencyInfo;
class BranchProbabilityInfo;
+class AssumptionCache;
class CallInst;
class DominatorTree;
class Function;
@@ -56,6 +57,7 @@
const bool AggregateArgs;
BlockFrequencyInfo *BFI;
BranchProbabilityInfo *BPI;
+ AssumptionCache *AC;
// If true, varargs functions can be extracted.
bool AllowVarArgs;
@@ -84,6 +86,7 @@
CodeExtractor(ArrayRef<BasicBlock *> BBs, DominatorTree *DT = nullptr,
bool AggregateArgs = false, BlockFrequencyInfo *BFI = nullptr,
BranchProbabilityInfo *BPI = nullptr,
+ AssumptionCache *AC = nullptr,
bool AllowVarArgs = false, bool AllowAlloca = false,
std::string Suffix = "");
@@ -94,6 +97,7 @@
CodeExtractor(DominatorTree &DT, Loop &L, bool AggregateArgs = false,
BlockFrequencyInfo *BFI = nullptr,
BranchProbabilityInfo *BPI = nullptr,
+ AssumptionCache *AC = nullptr,
std::string Suffix = "");
/// Perform the extraction, returning the new function.
diff --git a/lib/Analysis/AssumptionCache.cpp b/lib/Analysis/AssumptionCache.cpp
index bc7b093..cf2f845 100644
--- a/lib/Analysis/AssumptionCache.cpp
+++ b/lib/Analysis/AssumptionCache.cpp
@@ -53,11 +53,11 @@
return AVIP.first->second;
}
-void AssumptionCache::updateAffectedValues(CallInst *CI) {
+static void findAffectedValues(CallInst *CI,
+ SmallVectorImpl<Value *> &Affected) {
// Note: This code must be kept in-sync with the code in
// computeKnownBitsFromAssume in ValueTracking.
- SmallVector<Value *, 16> Affected;
auto AddAffected = [&Affected](Value *V) {
if (isa<Argument>(V)) {
Affected.push_back(V);
@@ -108,6 +108,11 @@
AddAffectedFromEq(B);
}
}
+}
+
+void AssumptionCache::updateAffectedValues(CallInst *CI) {
+ SmallVector<Value *, 16> Affected;
+ findAffectedValues(CI, Affected);
for (auto &AV : Affected) {
auto &AVV = getOrInsertAffectedValues(AV);
@@ -116,6 +121,18 @@
}
}
+void AssumptionCache::unregisterAssumption(CallInst *CI) {
+ SmallVector<Value *, 16> Affected;
+ findAffectedValues(CI, Affected);
+
+ for (auto &AV : Affected) {
+ auto AVI = AffectedValues.find_as(AV);
+ if (AVI != AffectedValues.end())
+ AffectedValues.erase(AVI);
+ }
+ remove_if(AssumeHandles, [CI](WeakTrackingVH &VH) { return CI == VH; });
+}
+
void AssumptionCache::AffectedValueCallbackVH::deleted() {
auto AVI = AC->AffectedValues.find(getValPtr());
if (AVI != AC->AffectedValues.end())
@@ -240,6 +257,13 @@
return *IP.first->second;
}
+AssumptionCache *AssumptionCacheTracker::lookupAssumptionCache(Function &F) {
+ auto I = AssumptionCaches.find_as(&F);
+ if (I != AssumptionCaches.end())
+ return I->second.get();
+ return nullptr;
+}
+
void AssumptionCacheTracker::verifyAnalysis() const {
// FIXME: In the long term the verifier should not be controllable with a
// flag. We should either fix all passes to correctly update the assumption
diff --git a/lib/Transforms/IPO/HotColdSplitting.cpp b/lib/Transforms/IPO/HotColdSplitting.cpp
index 648e2ae..b8def7a 100644
--- a/lib/Transforms/IPO/HotColdSplitting.cpp
+++ b/lib/Transforms/IPO/HotColdSplitting.cpp
@@ -173,8 +173,9 @@
HotColdSplitting(ProfileSummaryInfo *ProfSI,
function_ref<BlockFrequencyInfo *(Function &)> GBFI,
function_ref<TargetTransformInfo &(Function &)> GTTI,
- std::function<OptimizationRemarkEmitter &(Function &)> *GORE)
- : PSI(ProfSI), GetBFI(GBFI), GetTTI(GTTI), GetORE(GORE) {}
+ std::function<OptimizationRemarkEmitter &(Function &)> *GORE,
+ function_ref<AssumptionCache *(Function &)> LAC)
+ : PSI(ProfSI), GetBFI(GBFI), GetTTI(GTTI), GetORE(GORE), LookupAC(LAC) {}
bool run(Module &M);
private:
@@ -183,11 +184,13 @@
bool outlineColdRegions(Function &F, bool HasProfileSummary);
Function *extractColdRegion(const BlockSequence &Region, DominatorTree &DT,
BlockFrequencyInfo *BFI, TargetTransformInfo &TTI,
- OptimizationRemarkEmitter &ORE, unsigned Count);
+ OptimizationRemarkEmitter &ORE,
+ AssumptionCache *AC, unsigned Count);
ProfileSummaryInfo *PSI;
function_ref<BlockFrequencyInfo *(Function &)> GetBFI;
function_ref<TargetTransformInfo &(Function &)> GetTTI;
std::function<OptimizationRemarkEmitter &(Function &)> *GetORE;
+ function_ref<AssumptionCache *(Function &)> LookupAC;
};
class HotColdSplittingLegacyPass : public ModulePass {
@@ -198,10 +201,10 @@
}
void getAnalysisUsage(AnalysisUsage &AU) const override {
- AU.addRequired<AssumptionCacheTracker>();
AU.addRequired<BlockFrequencyInfoWrapperPass>();
AU.addRequired<ProfileSummaryInfoWrapperPass>();
AU.addRequired<TargetTransformInfoWrapperPass>();
+ AU.addUsedIfAvailable<AssumptionCacheTracker>();
}
bool runOnModule(Module &M) override;
@@ -316,12 +319,13 @@
BlockFrequencyInfo *BFI,
TargetTransformInfo &TTI,
OptimizationRemarkEmitter &ORE,
+ AssumptionCache *AC,
unsigned Count) {
assert(!Region.empty());
// TODO: Pass BFI and BPI to update profile information.
CodeExtractor CE(Region, &DT, /* AggregateArgs */ false, /* BFI */ nullptr,
- /* BPI */ nullptr, /* AllowVarArgs */ false,
+ /* BPI */ nullptr, AC, /* AllowVarArgs */ false,
/* AllowAlloca */ false,
/* Suffix */ "cold." + std::to_string(Count));
@@ -577,6 +581,7 @@
TargetTransformInfo &TTI = GetTTI(F);
OptimizationRemarkEmitter &ORE = (*GetORE)(F);
+ AssumptionCache *AC = LookupAC(F);
// Find all cold regions.
for (BasicBlock *BB : RPOT) {
@@ -638,8 +643,8 @@
BB->dump();
});
- Function *Outlined =
- extractColdRegion(SubRegion, *DT, BFI, TTI, ORE, OutlinedFunctionID);
+ Function *Outlined = extractColdRegion(SubRegion, *DT, BFI, TTI, ORE, AC,
+ OutlinedFunctionID);
if (Outlined) {
++OutlinedFunctionID;
Changed = true;
@@ -698,17 +703,21 @@
ORE.reset(new OptimizationRemarkEmitter(&F));
return *ORE.get();
};
+ auto LookupAC = [this](Function &F) -> AssumptionCache * {
+ if (auto *ACT = getAnalysisIfAvailable<AssumptionCacheTracker>())
+ return ACT->lookupAssumptionCache(F);
+ return nullptr;
+ };
- return HotColdSplitting(PSI, GBFI, GTTI, &GetORE).run(M);
+ return HotColdSplitting(PSI, GBFI, GTTI, &GetORE, LookupAC).run(M);
}
PreservedAnalyses
HotColdSplittingPass::run(Module &M, ModuleAnalysisManager &AM) {
auto &FAM = AM.getResult<FunctionAnalysisManagerModuleProxy>(M).getManager();
- std::function<AssumptionCache &(Function &)> GetAssumptionCache =
- [&FAM](Function &F) -> AssumptionCache & {
- return FAM.getResult<AssumptionAnalysis>(F);
+ auto LookupAC = [&FAM](Function &F) -> AssumptionCache * {
+ return FAM.getCachedResult<AssumptionAnalysis>(F);
};
auto GBFI = [&FAM](Function &F) {
@@ -729,7 +738,7 @@
ProfileSummaryInfo *PSI = &AM.getResult<ProfileSummaryAnalysis>(M);
- if (HotColdSplitting(PSI, GBFI, GTTI, &GetORE).run(M))
+ if (HotColdSplitting(PSI, GBFI, GTTI, &GetORE, LookupAC).run(M))
return PreservedAnalyses::none();
return PreservedAnalyses::all();
}
diff --git a/lib/Transforms/IPO/LoopExtractor.cpp b/lib/Transforms/IPO/LoopExtractor.cpp
index 6e7e59a..91c7b5f 100644
--- a/lib/Transforms/IPO/LoopExtractor.cpp
+++ b/lib/Transforms/IPO/LoopExtractor.cpp
@@ -14,6 +14,7 @@
//===----------------------------------------------------------------------===//
#include "llvm/ADT/Statistic.h"
+#include "llvm/Analysis/AssumptionCache.h"
#include "llvm/Analysis/LoopPass.h"
#include "llvm/IR/Dominators.h"
#include "llvm/IR/Instructions.h"
@@ -50,6 +51,7 @@
AU.addRequiredID(LoopSimplifyID);
AU.addRequired<DominatorTreeWrapperPass>();
AU.addRequired<LoopInfoWrapperPass>();
+ AU.addUsedIfAvailable<AssumptionCacheTracker>();
}
};
}
@@ -138,7 +140,10 @@
if (ShouldExtractLoop) {
if (NumLoops == 0) return Changed;
--NumLoops;
- CodeExtractor Extractor(DT, *L);
+ AssumptionCache *AC = nullptr;
+ if (auto *ACT = getAnalysisIfAvailable<AssumptionCacheTracker>())
+ AC = ACT->lookupAssumptionCache(*L->getHeader()->getParent());
+ CodeExtractor Extractor(DT, *L, false, nullptr, nullptr, AC);
if (Extractor.extractCodeRegion() != nullptr) {
Changed = true;
// After extraction, the loop is replaced by a function call, so
diff --git a/lib/Transforms/IPO/PartialInlining.cpp b/lib/Transforms/IPO/PartialInlining.cpp
index f971cee..8339eb4 100644
--- a/lib/Transforms/IPO/PartialInlining.cpp
+++ b/lib/Transforms/IPO/PartialInlining.cpp
@@ -199,10 +199,12 @@
PartialInlinerImpl(
std::function<AssumptionCache &(Function &)> *GetAC,
+ function_ref<AssumptionCache *(Function &)> LookupAC,
std::function<TargetTransformInfo &(Function &)> *GTTI,
Optional<function_ref<BlockFrequencyInfo &(Function &)>> GBFI,
ProfileSummaryInfo *ProfSI)
- : GetAssumptionCache(GetAC), GetTTI(GTTI), GetBFI(GBFI), PSI(ProfSI) {}
+ : GetAssumptionCache(GetAC), LookupAssumptionCache(LookupAC),
+ GetTTI(GTTI), GetBFI(GBFI), PSI(ProfSI) {}
bool run(Module &M);
// Main part of the transformation that calls helper functions to find
@@ -222,9 +224,11 @@
// Two constructors, one for single region outlining, the other for
// multi-region outlining.
FunctionCloner(Function *F, FunctionOutliningInfo *OI,
- OptimizationRemarkEmitter &ORE);
+ OptimizationRemarkEmitter &ORE,
+ function_ref<AssumptionCache *(Function &)> LookupAC);
FunctionCloner(Function *F, FunctionOutliningMultiRegionInfo *OMRI,
- OptimizationRemarkEmitter &ORE);
+ OptimizationRemarkEmitter &ORE,
+ function_ref<AssumptionCache *(Function &)> LookupAC);
~FunctionCloner();
// Prepare for function outlining: making sure there is only
@@ -260,11 +264,13 @@
std::unique_ptr<FunctionOutliningMultiRegionInfo> ClonedOMRI = nullptr;
std::unique_ptr<BlockFrequencyInfo> ClonedFuncBFI = nullptr;
OptimizationRemarkEmitter &ORE;
+ function_ref<AssumptionCache *(Function &)> LookupAC;
};
private:
int NumPartialInlining = 0;
std::function<AssumptionCache &(Function &)> *GetAssumptionCache;
+ function_ref<AssumptionCache *(Function &)> LookupAssumptionCache;
std::function<TargetTransformInfo &(Function &)> *GetTTI;
Optional<function_ref<BlockFrequencyInfo &(Function &)>> GetBFI;
ProfileSummaryInfo *PSI;
@@ -365,12 +371,17 @@
return ACT->getAssumptionCache(F);
};
+ auto LookupAssumptionCache = [ACT](Function &F) -> AssumptionCache * {
+ return ACT->lookupAssumptionCache(F);
+ };
+
std::function<TargetTransformInfo &(Function &)> GetTTI =
[&TTIWP](Function &F) -> TargetTransformInfo & {
return TTIWP->getTTI(F);
};
- return PartialInlinerImpl(&GetAssumptionCache, &GetTTI, NoneType::None, PSI)
+ return PartialInlinerImpl(&GetAssumptionCache, LookupAssumptionCache,
+ &GetTTI, NoneType::None, PSI)
.run(M);
}
};
@@ -948,8 +959,9 @@
}
PartialInlinerImpl::FunctionCloner::FunctionCloner(
- Function *F, FunctionOutliningInfo *OI, OptimizationRemarkEmitter &ORE)
- : OrigFunc(F), ORE(ORE) {
+ Function *F, FunctionOutliningInfo *OI, OptimizationRemarkEmitter &ORE,
+ function_ref<AssumptionCache *(Function &)> LookupAC)
+ : OrigFunc(F), ORE(ORE), LookupAC(LookupAC) {
ClonedOI = llvm::make_unique<FunctionOutliningInfo>();
// Clone the function, so that we can hack away on it.
@@ -972,8 +984,9 @@
PartialInlinerImpl::FunctionCloner::FunctionCloner(
Function *F, FunctionOutliningMultiRegionInfo *OI,
- OptimizationRemarkEmitter &ORE)
- : OrigFunc(F), ORE(ORE) {
+ OptimizationRemarkEmitter &ORE,
+ function_ref<AssumptionCache *(Function &)> LookupAC)
+ : OrigFunc(F), ORE(ORE), LookupAC(LookupAC) {
ClonedOMRI = llvm::make_unique<FunctionOutliningMultiRegionInfo>();
// Clone the function, so that we can hack away on it.
@@ -1111,7 +1124,9 @@
int CurrentOutlinedRegionCost = ComputeRegionCost(RegionInfo.Region);
CodeExtractor CE(RegionInfo.Region, &DT, /*AggregateArgs*/ false,
- ClonedFuncBFI.get(), &BPI, /* AllowVarargs */ false);
+ ClonedFuncBFI.get(), &BPI,
+ LookupAC(*RegionInfo.EntryBlock->getParent()),
+ /* AllowVarargs */ false);
CE.findInputsOutputs(Inputs, Outputs, Sinks);
@@ -1193,7 +1208,7 @@
// Extract the body of the if.
Function *OutlinedFunc =
CodeExtractor(ToExtract, &DT, /*AggregateArgs*/ false,
- ClonedFuncBFI.get(), &BPI,
+ ClonedFuncBFI.get(), &BPI, LookupAC(*ClonedFunc),
/* AllowVarargs */ true)
.extractCodeRegion();
@@ -1257,7 +1272,7 @@
std::unique_ptr<FunctionOutliningMultiRegionInfo> OMRI =
computeOutliningColdRegionsInfo(F, ORE);
if (OMRI) {
- FunctionCloner Cloner(F, OMRI.get(), ORE);
+ FunctionCloner Cloner(F, OMRI.get(), ORE, LookupAssumptionCache);
#ifndef NDEBUG
if (TracePartialInlining) {
@@ -1290,7 +1305,7 @@
if (!OI)
return {false, nullptr};
- FunctionCloner Cloner(F, OI.get(), ORE);
+ FunctionCloner Cloner(F, OI.get(), ORE, LookupAssumptionCache);
Cloner.NormalizeReturnBlock();
Function *OutlinedFunction = Cloner.doSingleRegionFunctionOutlining();
@@ -1484,6 +1499,10 @@
return FAM.getResult<AssumptionAnalysis>(F);
};
+ auto LookupAssumptionCache = [&FAM](Function &F) -> AssumptionCache * {
+ return FAM.getCachedResult<AssumptionAnalysis>(F);
+ };
+
std::function<BlockFrequencyInfo &(Function &)> GetBFI =
[&FAM](Function &F) -> BlockFrequencyInfo & {
return FAM.getResult<BlockFrequencyAnalysis>(F);
@@ -1496,7 +1515,8 @@
ProfileSummaryInfo *PSI = &AM.getResult<ProfileSummaryAnalysis>(M);
- if (PartialInlinerImpl(&GetAssumptionCache, &GetTTI, {GetBFI}, PSI)
+ if (PartialInlinerImpl(&GetAssumptionCache, LookupAssumptionCache, &GetTTI,
+ {GetBFI}, PSI)
.run(M))
return PreservedAnalyses::none();
return PreservedAnalyses::all();
diff --git a/lib/Transforms/Utils/CodeExtractor.cpp b/lib/Transforms/Utils/CodeExtractor.cpp
index 9148c11..e941de2 100644
--- a/lib/Transforms/Utils/CodeExtractor.cpp
+++ b/lib/Transforms/Utils/CodeExtractor.cpp
@@ -20,6 +20,7 @@
#include "llvm/ADT/SetVector.h"
#include "llvm/ADT/SmallPtrSet.h"
#include "llvm/ADT/SmallVector.h"
+#include "llvm/Analysis/AssumptionCache.h"
#include "llvm/Analysis/BlockFrequencyInfo.h"
#include "llvm/Analysis/BlockFrequencyInfoImpl.h"
#include "llvm/Analysis/BranchProbabilityInfo.h"
@@ -43,6 +44,7 @@
#include "llvm/IR/LLVMContext.h"
#include "llvm/IR/MDBuilder.h"
#include "llvm/IR/Module.h"
+#include "llvm/IR/PatternMatch.h"
#include "llvm/IR/Type.h"
#include "llvm/IR/User.h"
#include "llvm/IR/Value.h"
@@ -66,6 +68,7 @@
#include <vector>
using namespace llvm;
+using namespace llvm::PatternMatch;
using ProfileCount = Function::ProfileCount;
#define DEBUG_TYPE "code-extractor"
@@ -235,18 +238,20 @@
CodeExtractor::CodeExtractor(ArrayRef<BasicBlock *> BBs, DominatorTree *DT,
bool AggregateArgs, BlockFrequencyInfo *BFI,
- BranchProbabilityInfo *BPI, bool AllowVarArgs,
- bool AllowAlloca, std::string Suffix)
+ BranchProbabilityInfo *BPI, AssumptionCache *AC,
+ bool AllowVarArgs, bool AllowAlloca,
+ std::string Suffix)
: DT(DT), AggregateArgs(AggregateArgs || AggregateArgsOpt), BFI(BFI),
- BPI(BPI), AllowVarArgs(AllowVarArgs),
+ BPI(BPI), AC(AC), AllowVarArgs(AllowVarArgs),
Blocks(buildExtractionBlockSet(BBs, DT, AllowVarArgs, AllowAlloca)),
Suffix(Suffix) {}
CodeExtractor::CodeExtractor(DominatorTree &DT, Loop &L, bool AggregateArgs,
BlockFrequencyInfo *BFI,
- BranchProbabilityInfo *BPI, std::string Suffix)
+ BranchProbabilityInfo *BPI, AssumptionCache *AC,
+ std::string Suffix)
: DT(&DT), AggregateArgs(AggregateArgs || AggregateArgsOpt), BFI(BFI),
- BPI(BPI), AllowVarArgs(false),
+ BPI(BPI), AC(AC), AllowVarArgs(false),
Blocks(buildExtractionBlockSet(L.getBlocks(), &DT,
/* AllowVarArgs */ false,
/* AllowAlloca */ false)),
@@ -1216,6 +1221,13 @@
// Insert this basic block into the new function
newBlocks.push_back(Block);
+
+ // Remove @llvm.assume calls that were moved to the new function from the
+ // old function's assumption cache.
+ if (AC)
+ for (auto &I : *Block)
+ if (match(&I, m_Intrinsic<Intrinsic::assume>()))
+ AC->unregisterAssumption(cast<CallInst>(&I));
}
}
diff --git a/test/Transforms/CodeExtractor/extract-assume.ll b/test/Transforms/CodeExtractor/extract-assume.ll
new file mode 100644
index 0000000..b79c6a6
--- /dev/null
+++ b/test/Transforms/CodeExtractor/extract-assume.ll
@@ -0,0 +1,29 @@
+; RUN: opt -passes="function(slp-vectorizer),module(hotcoldsplit),function(slp-vectorizer,print<assumptions>)" -disable-output %s 2>&1 | FileCheck %s
+;
+; Make sure this compiles. Check that function assumption cache is refreshed
+; after extracting blocks with assume calls from the function.
+
+; CHECK: Cached assumptions for function: fun
+; CHECK-NEXT: Cached assumptions for function: fun.cold
+; CHECK-NEXT: %cmp = icmp uge i32 %x, 64
+
+declare void @fun2(i32) #0
+
+define void @fun(i32 %x) {
+entry:
+ br i1 undef, label %if.then, label %if.else
+
+if.then:
+ ret void
+
+if.else:
+ %cmp = icmp uge i32 %x, 64
+ call void @llvm.assume(i1 %cmp)
+ call void @fun2(i32 %x)
+ unreachable
+}
+
+declare void @llvm.assume(i1) #1
+
+attributes #0 = { alwaysinline }
+attributes #1 = { nounwind }
