#include "llvm/Analysis/AssumptionCache.h"
#include "llvm/Analysis/CFG.h"
#include "llvm/Analysis/ConstantFolding.h"
+#include "llvm/Analysis/GlobalsModRef.h"
#include "llvm/Analysis/InstructionSimplify.h"
#include "llvm/Analysis/Loads.h"
#include "llvm/Analysis/MemoryBuiltins.h"
DenseMap<uint32_t, LeaderTableEntry> LeaderTable;
BumpPtrAllocator TableAllocator;
+ // Block-local map of equivalent values to their leader, does not
+ // propagate to any successors. Entries added mid-block are applied
+ // to the remaining instructions in the block.
+ SmallMapVector<llvm::Value *, llvm::Constant *, 4> ReplaceWithConstMap;
SmallVector<Instruction*, 8> InstrsToErase;
typedef SmallVector<NonLocalDepResult, 64> LoadDepVect;
AU.addRequired<TargetLibraryInfoWrapperPass>();
if (!NoLoads)
AU.addRequired<MemoryDependenceAnalysis>();
- AU.addRequired<AliasAnalysis>();
+ AU.addRequired<AAResultsWrapperPass>();
AU.addPreserved<DominatorTreeWrapperPass>();
- AU.addPreserved<AliasAnalysis>();
+ AU.addPreserved<GlobalsAAWrapperPass>();
}
// Helper functions of redundant load elimination
bool processLoad(LoadInst *L);
bool processNonLocalLoad(LoadInst *L);
+ bool processAssumeIntrinsic(IntrinsicInst *II);
void AnalyzeLoadAvailability(LoadInst *LI, LoadDepVect &Deps,
AvailValInBlkVect &ValuesPerBlock,
UnavailBlkVect &UnavailableBlocks);
void verifyRemoved(const Instruction *I) const;
bool splitCriticalEdges();
BasicBlock *splitCriticalEdges(BasicBlock *Pred, BasicBlock *Succ);
- bool propagateEquality(Value *LHS, Value *RHS, const BasicBlockEdge &Root);
+ bool replaceOperandsWithConsts(Instruction *I) const;
+ bool propagateEquality(Value *LHS, Value *RHS, const BasicBlockEdge &Root,
+ bool DominatesByEdge);
bool processFoldableCondBr(BranchInst *BI);
void addDeadBlock(BasicBlock *BB);
void assignValNumForDeadCode();
INITIALIZE_PASS_DEPENDENCY(MemoryDependenceAnalysis)
INITIALIZE_PASS_DEPENDENCY(DominatorTreeWrapperPass)
INITIALIZE_PASS_DEPENDENCY(TargetLibraryInfoWrapperPass)
-INITIALIZE_AG_DEPENDENCY(AliasAnalysis)
+INITIALIZE_PASS_DEPENDENCY(AAResultsWrapperPass)
+INITIALIZE_PASS_DEPENDENCY(GlobalsAAWrapperPass)
INITIALIZE_PASS_END(GVN, "gvn", "Global Value Numbering", false, false)
#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
if (Tags)
NewLoad->setAAMetadata(Tags);
+ if (auto *InvGroupMD = LI->getMetadata(LLVMContext::MD_invariant_group))
+ NewLoad->setMetadata(LLVMContext::MD_invariant_group, InvGroupMD);
+
// Transfer DebugLoc.
NewLoad->setDebugLoc(LI->getDebugLoc());
if (isa<PHINode>(V))
V->takeName(LI);
if (Instruction *I = dyn_cast<Instruction>(V))
- I->setDebugLoc(LI->getDebugLoc());
+ if (LI->getDebugLoc())
+ I->setDebugLoc(LI->getDebugLoc());
if (V->getType()->getScalarType()->isPointerTy())
MD->invalidateCachedPointerInfo(V);
markInstructionForDeletion(LI);
return PerformLoadPRE(LI, ValuesPerBlock, UnavailableBlocks);
}
+bool GVN::processAssumeIntrinsic(IntrinsicInst *IntrinsicI) {
+ assert(IntrinsicI->getIntrinsicID() == Intrinsic::assume &&
+ "This function can only be called with llvm.assume intrinsic");
+ Value *V = IntrinsicI->getArgOperand(0);
+
+ if (ConstantInt *Cond = dyn_cast<ConstantInt>(V)) {
+ if (Cond->isZero()) {
+ Type *Int8Ty = Type::getInt8Ty(V->getContext());
+ // Insert a new store to null instruction before the load to indicate that
+ // this code is not reachable. FIXME: We could insert unreachable
+ // instruction directly because we can modify the CFG.
+ new StoreInst(UndefValue::get(Int8Ty),
+ Constant::getNullValue(Int8Ty->getPointerTo()),
+ IntrinsicI);
+ }
+ markInstructionForDeletion(IntrinsicI);
+ return false;
+ }
+
+ Constant *True = ConstantInt::getTrue(V->getContext());
+ bool Changed = false;
+
+ for (BasicBlock *Successor : successors(IntrinsicI->getParent())) {
+ BasicBlockEdge Edge(IntrinsicI->getParent(), Successor);
+
+ // This property is only true in dominated successors, propagateEquality
+ // will check dominance for us.
+ Changed |= propagateEquality(V, True, Edge, false);
+ }
+
+ // We can replace assume value with true, which covers cases like this:
+ // call void @llvm.assume(i1 %cmp)
+ // br i1 %cmp, label %bb1, label %bb2 ; will change %cmp to true
+ ReplaceWithConstMap[V] = True;
+
+ // If one of *cmp *eq operand is const, adding it to map will cover this:
+ // %cmp = fcmp oeq float 3.000000e+00, %0 ; const on lhs could happen
+ // call void @llvm.assume(i1 %cmp)
+ // ret float %0 ; will change it to ret float 3.000000e+00
+ if (auto *CmpI = dyn_cast<CmpInst>(V)) {
+ if (CmpI->getPredicate() == CmpInst::Predicate::ICMP_EQ ||
+ CmpI->getPredicate() == CmpInst::Predicate::FCMP_OEQ ||
+ (CmpI->getPredicate() == CmpInst::Predicate::FCMP_UEQ &&
+ CmpI->getFastMathFlags().noNaNs())) {
+ Value *CmpLHS = CmpI->getOperand(0);
+ Value *CmpRHS = CmpI->getOperand(1);
+ if (isa<Constant>(CmpLHS))
+ std::swap(CmpLHS, CmpRHS);
+ auto *RHSConst = dyn_cast<Constant>(CmpRHS);
+
+ // If only one operand is constant.
+ if (RHSConst != nullptr && !isa<Constant>(CmpLHS))
+ ReplaceWithConstMap[CmpLHS] = RHSConst;
+ }
+ }
+ return Changed;
+}
static void patchReplacementInstruction(Instruction *I, Value *Repl) {
// Patch the replacement so that it is not more restrictive than the value
if (Instruction *ReplInst = dyn_cast<Instruction>(Repl)) {
// FIXME: If both the original and replacement value are part of the
// same control-flow region (meaning that the execution of one
- // guarentees the executation of the other), then we can combine the
+ // guarantees the execution of the other), then we can combine the
// noalias scopes here and do better than the general conservative
// answer used in combineMetadata().
// regions, and so we need a conservative combination of the noalias
// scopes.
static const unsigned KnownIDs[] = {
- LLVMContext::MD_tbaa,
- LLVMContext::MD_alias_scope,
- LLVMContext::MD_noalias,
- LLVMContext::MD_range,
- LLVMContext::MD_fpmath,
- LLVMContext::MD_invariant_load,
- };
+ LLVMContext::MD_tbaa, LLVMContext::MD_alias_scope,
+ LLVMContext::MD_noalias, LLVMContext::MD_range,
+ LLVMContext::MD_fpmath, LLVMContext::MD_invariant_load,
+ LLVMContext::MD_invariant_group};
combineMetadata(ReplInst, I, KnownIDs);
}
}
++NumGVNLoad;
return true;
}
- }
- // If the value isn't available, don't do anything!
- if (Dep.isClobber()) {
+ // If the value isn't available, don't do anything!
DEBUG(
// fast print dep, using operator<< on instruction is too slow.
dbgs() << "GVN: load ";
return Pred != nullptr;
}
+// Tries to replace instruction with const, using information from
+// ReplaceWithConstMap.
+bool GVN::replaceOperandsWithConsts(Instruction *Instr) const {
+ bool Changed = false;
+ for (unsigned OpNum = 0; OpNum < Instr->getNumOperands(); ++OpNum) {
+ Value *Operand = Instr->getOperand(OpNum);
+ auto it = ReplaceWithConstMap.find(Operand);
+ if (it != ReplaceWithConstMap.end()) {
+ assert(!isa<Constant>(Operand) &&
+ "Replacing constants with constants is invalid");
+ DEBUG(dbgs() << "GVN replacing: " << *Operand << " with " << *it->second
+ << " in instruction " << *Instr << '\n');
+ Instr->setOperand(OpNum, it->second);
+ Changed = true;
+ }
+ }
+ return Changed;
+}
+
/// The given values are known to be equal in every block
/// dominated by 'Root'. Exploit this, for example by replacing 'LHS' with
/// 'RHS' everywhere in the scope. Returns whether a change was made.
-bool GVN::propagateEquality(Value *LHS, Value *RHS,
- const BasicBlockEdge &Root) {
+/// If DominatesByEdge is false, then it means that it is dominated by Root.End.
+bool GVN::propagateEquality(Value *LHS, Value *RHS, const BasicBlockEdge &Root,
+ bool DominatesByEdge) {
SmallVector<std::pair<Value*, Value*>, 4> Worklist;
Worklist.push_back(std::make_pair(LHS, RHS));
bool Changed = false;
std::pair<Value*, Value*> Item = Worklist.pop_back_val();
LHS = Item.first; RHS = Item.second;
- if (LHS == RHS) continue;
+ if (LHS == RHS)
+ continue;
assert(LHS->getType() == RHS->getType() && "Equality but unequal types!");
// Don't try to propagate equalities between constants.
- if (isa<Constant>(LHS) && isa<Constant>(RHS)) continue;
+ if (isa<Constant>(LHS) && isa<Constant>(RHS))
+ continue;
// Prefer a constant on the right-hand side, or an Argument if no constants.
if (isa<Constant>(LHS) || (isa<Argument>(LHS) && !isa<Constant>(RHS)))
// LHS always has at least one use that is not dominated by Root, this will
// never do anything if LHS has only one use.
if (!LHS->hasOneUse()) {
- unsigned NumReplacements = replaceDominatedUsesWith(LHS, RHS, *DT, Root);
+ unsigned NumReplacements =
+ DominatesByEdge
+ ? replaceDominatedUsesWith(LHS, RHS, *DT, Root)
+ : replaceDominatedUsesWith(LHS, RHS, *DT, Root.getEnd());
+
Changed |= NumReplacements > 0;
NumGVNEqProp += NumReplacements;
}
Value *NotCmp = findLeader(Root.getEnd(), Num);
if (NotCmp && isa<Instruction>(NotCmp)) {
unsigned NumReplacements =
- replaceDominatedUsesWith(NotCmp, NotVal, *DT, Root);
+ DominatesByEdge
+ ? replaceDominatedUsesWith(NotCmp, NotVal, *DT, Root)
+ : replaceDominatedUsesWith(NotCmp, NotVal, *DT,
+ Root.getEnd());
Changed |= NumReplacements > 0;
NumGVNEqProp += NumReplacements;
}
return true;
}
+ if (IntrinsicInst *IntrinsicI = dyn_cast<IntrinsicInst>(I))
+ if (IntrinsicI->getIntrinsicID() == Intrinsic::assume)
+ return processAssumeIntrinsic(IntrinsicI);
+
if (LoadInst *LI = dyn_cast<LoadInst>(I)) {
if (processLoad(LI))
return true;
Value *TrueVal = ConstantInt::getTrue(TrueSucc->getContext());
BasicBlockEdge TrueE(Parent, TrueSucc);
- Changed |= propagateEquality(BranchCond, TrueVal, TrueE);
+ Changed |= propagateEquality(BranchCond, TrueVal, TrueE, true);
Value *FalseVal = ConstantInt::getFalse(FalseSucc->getContext());
BasicBlockEdge FalseE(Parent, FalseSucc);
- Changed |= propagateEquality(BranchCond, FalseVal, FalseE);
+ Changed |= propagateEquality(BranchCond, FalseVal, FalseE, true);
return Changed;
}
// If there is only a single edge, propagate the case value into it.
if (SwitchEdges.lookup(Dst) == 1) {
BasicBlockEdge E(Parent, Dst);
- Changed |= propagateEquality(SwitchCond, i.getCaseValue(), E);
+ Changed |= propagateEquality(SwitchCond, i.getCaseValue(), E, true);
}
}
return Changed;
// Instructions with void type don't return a value, so there's
// no point in trying to find redundancies in them.
- if (I->getType()->isVoidTy()) return false;
+ if (I->getType()->isVoidTy())
+ return false;
uint32_t NextNum = VN.getNextUnusedValueNumber();
unsigned Num = VN.lookup_or_add(I);
DT = &getAnalysis<DominatorTreeWrapperPass>().getDomTree();
AC = &getAnalysis<AssumptionCacheTracker>().getAssumptionCache(F);
TLI = &getAnalysis<TargetLibraryInfoWrapperPass>().getTLI();
- VN.setAliasAnalysis(&getAnalysis<AliasAnalysis>());
+ VN.setAliasAnalysis(&getAnalysis<AAResultsWrapperPass>().getAAResults());
VN.setMemDep(MD);
VN.setDomTree(DT);
// Merge unconditional branches, allowing PRE to catch more
// optimization opportunities.
for (Function::iterator FI = F.begin(), FE = F.end(); FI != FE; ) {
- BasicBlock *BB = FI++;
+ BasicBlock *BB = &*FI++;
bool removedBlock =
MergeBlockIntoPredecessor(BB, DT, /* LoopInfo */ nullptr, MD);
return Changed;
}
-
bool GVN::processBlock(BasicBlock *BB) {
// FIXME: Kill off InstrsToErase by doing erasing eagerly in a helper function
// (and incrementing BI before processing an instruction).
if (DeadBlocks.count(BB))
return false;
+ // Clearing map before every BB because it can be used only for single BB.
+ ReplaceWithConstMap.clear();
bool ChangedFunction = false;
for (BasicBlock::iterator BI = BB->begin(), BE = BB->end();
BI != BE;) {
- ChangedFunction |= processInstruction(BI);
+ if (!ReplaceWithConstMap.empty())
+ ChangedFunction |= replaceOperandsWithConsts(&*BI);
+ ChangedFunction |= processInstruction(&*BI);
+
if (InstrsToErase.empty()) {
++BI;
continue;
// Create a PHI to make the value available in this block.
PHINode *Phi =
PHINode::Create(CurInst->getType(), predMap.size(),
- CurInst->getName() + ".pre-phi", CurrentBlock->begin());
+ CurInst->getName() + ".pre-phi", &CurrentBlock->front());
for (unsigned i = 0, e = predMap.size(); i != e; ++i) {
if (Value *V = predMap[i].first)
Phi->addIncoming(V, predMap[i].second);
for (BasicBlock::iterator BI = CurrentBlock->begin(),
BE = CurrentBlock->end();
BI != BE;) {
- Instruction *CurInst = BI++;
+ Instruction *CurInst = &*BI++;
Changed = performScalarPRE(CurInst);
}
}
// R be the target of the dead out-coming edge.
// 1) Identify the set of dead blocks implied by the branch's dead outcoming
// edge. The result of this step will be {X| X is dominated by R}
-// 2) Identify those blocks which haves at least one dead prodecessor. The
+// 2) Identify those blocks which haves at least one dead predecessor. The
// result of this step will be dominance-frontier(R).
// 3) Update the PHIs in DF(R) by replacing the operands corresponding to
// dead blocks with "UndefVal" in an hope these PHIs will optimized away.
projects / oota-llvm.git / blobdiff