summaryrefslogtreecommitdiff
path: root/llvm/lib/Analysis/LoopAccessAnalysis.cpp
diff options
context:
space:
mode:
Diffstat (limited to 'llvm/lib/Analysis/LoopAccessAnalysis.cpp')
-rw-r--r--llvm/lib/Analysis/LoopAccessAnalysis.cpp131
1 files changed, 50 insertions, 81 deletions
diff --git a/llvm/lib/Analysis/LoopAccessAnalysis.cpp b/llvm/lib/Analysis/LoopAccessAnalysis.cpp
index 93b8d28ef749..38bf6d8160aa 100644
--- a/llvm/lib/Analysis/LoopAccessAnalysis.cpp
+++ b/llvm/lib/Analysis/LoopAccessAnalysis.cpp
@@ -239,7 +239,7 @@ getStartAndEndForAccess(const Loop *Lp, const SCEV *PtrExpr, Type *AccessTy,
assert(SE->isLoopInvariant(ScEnd, Lp)&& "ScEnd needs to be invariant");
// Add the size of the pointed element to ScEnd.
- auto &DL = Lp->getHeader()->getModule()->getDataLayout();
+ auto &DL = Lp->getHeader()->getDataLayout();
Type *IdxTy = DL.getIndexType(PtrExpr->getType());
const SCEV *EltSizeSCEV = SE->getStoreSizeOfExpr(IdxTy, AccessTy);
ScEnd = SE->getAddExpr(ScEnd, EltSizeSCEV);
@@ -309,7 +309,7 @@ bool RuntimePointerChecking::tryToCreateDiffCheck(
return false;
const DataLayout &DL =
- SinkAR->getLoop()->getHeader()->getModule()->getDataLayout();
+ SinkAR->getLoop()->getHeader()->getDataLayout();
unsigned AllocSize =
std::max(DL.getTypeAllocSize(SrcTy), DL.getTypeAllocSize(DstTy));
@@ -612,12 +612,12 @@ void RuntimePointerChecking::printChecks(
OS.indent(Depth) << "Check " << N++ << ":\n";
OS.indent(Depth + 2) << "Comparing group (" << Check1 << "):\n";
- for (unsigned K = 0; K < First.size(); ++K)
- OS.indent(Depth + 2) << *Pointers[First[K]].PointerValue << "\n";
+ for (unsigned K : First)
+ OS.indent(Depth + 2) << *Pointers[K].PointerValue << "\n";
OS.indent(Depth + 2) << "Against group (" << Check2 << "):\n";
- for (unsigned K = 0; K < Second.size(); ++K)
- OS.indent(Depth + 2) << *Pointers[Second[K]].PointerValue << "\n";
+ for (unsigned K : Second)
+ OS.indent(Depth + 2) << *Pointers[K].PointerValue << "\n";
}
}
@@ -627,15 +627,12 @@ void RuntimePointerChecking::print(raw_ostream &OS, unsigned Depth) const {
printChecks(OS, Checks, Depth);
OS.indent(Depth) << "Grouped accesses:\n";
- for (unsigned I = 0; I < CheckingGroups.size(); ++I) {
- const auto &CG = CheckingGroups[I];
-
+ for (const auto &CG : CheckingGroups) {
OS.indent(Depth + 2) << "Group " << &CG << ":\n";
OS.indent(Depth + 4) << "(Low: " << *CG.Low << " High: " << *CG.High
<< ")\n";
- for (unsigned J = 0; J < CG.Members.size(); ++J) {
- OS.indent(Depth + 6) << "Member: " << *Pointers[CG.Members[J]].Expr
- << "\n";
+ for (unsigned Member : CG.Members) {
+ OS.indent(Depth + 6) << "Member: " << *Pointers[Member].Expr << "\n";
}
}
}
@@ -1497,7 +1494,7 @@ std::optional<int64_t> llvm::getPtrStride(PredicatedScalarEvolution &PSE,
return std::nullopt;
}
- auto &DL = Lp->getHeader()->getModule()->getDataLayout();
+ auto &DL = Lp->getHeader()->getDataLayout();
TypeSize AllocSize = DL.getTypeAllocSize(AccessTy);
int64_t Size = AllocSize.getFixedValue();
const APInt &APStepVal = C->getAPInt();
@@ -1910,7 +1907,7 @@ MemoryDepChecker::getDependenceDistanceStrideAndSize(
const AccessAnalysis::MemAccessInfo &B, Instruction *BInst,
const DenseMap<Value *, SmallVector<const Value *, 16>>
&UnderlyingObjects) {
- auto &DL = InnermostLoop->getHeader()->getModule()->getDataLayout();
+ auto &DL = InnermostLoop->getHeader()->getDataLayout();
auto &SE = *PSE.getSE();
auto [APtr, AIsWrite] = A;
auto [BPtr, BIsWrite] = B;
@@ -2030,7 +2027,7 @@ MemoryDepChecker::Dependence::DepType MemoryDepChecker::isDependent(
}
ScalarEvolution &SE = *PSE.getSE();
- auto &DL = InnermostLoop->getHeader()->getModule()->getDataLayout();
+ auto &DL = InnermostLoop->getHeader()->getDataLayout();
uint64_t MaxStride = std::max(StrideA, StrideB);
// If the distance between the acecsses is larger than their maximum absolute
@@ -2355,9 +2352,9 @@ void MemoryDepChecker::Dependence::print(
bool LoopAccessInfo::canAnalyzeLoop() {
// We need to have a loop header.
- LLVM_DEBUG(dbgs() << "LAA: Found a loop in "
- << TheLoop->getHeader()->getParent()->getName() << ": "
- << TheLoop->getHeader()->getName() << '\n');
+ LLVM_DEBUG(dbgs() << "\nLAA: Checking a loop in '"
+ << TheLoop->getHeader()->getParent()->getName() << "' from "
+ << TheLoop->getLocStr() << "\n");
// We can only analyze innermost loops.
if (!TheLoop->isInnermost()) {
@@ -2386,10 +2383,12 @@ bool LoopAccessInfo::canAnalyzeLoop() {
return false;
}
+ LLVM_DEBUG(dbgs() << "LAA: Found an analyzable loop: "
+ << TheLoop->getHeader()->getName() << "\n");
return true;
}
-void LoopAccessInfo::analyzeLoop(AAResults *AA, LoopInfo *LI,
+bool LoopAccessInfo::analyzeLoop(AAResults *AA, LoopInfo *LI,
const TargetLibraryInfo *TLI,
DominatorTree *DT) {
// Holds the Load and Store instructions.
@@ -2430,10 +2429,8 @@ void LoopAccessInfo::analyzeLoop(AAResults *AA, LoopInfo *LI,
// With both a non-vectorizable memory instruction and a convergent
// operation, found in this loop, no reason to continue the search.
- if (HasComplexMemInst && HasConvergentOp) {
- CanVecMem = false;
- return;
- }
+ if (HasComplexMemInst && HasConvergentOp)
+ return false;
// Avoid hitting recordAnalysis multiple times.
if (HasComplexMemInst)
@@ -2508,10 +2505,8 @@ void LoopAccessInfo::analyzeLoop(AAResults *AA, LoopInfo *LI,
} // Next instr.
} // Next block.
- if (HasComplexMemInst) {
- CanVecMem = false;
- return;
- }
+ if (HasComplexMemInst)
+ return false;
// Now we have two lists that hold the loads and the stores.
// Next, we find the pointers that they use.
@@ -2520,8 +2515,7 @@ void LoopAccessInfo::analyzeLoop(AAResults *AA, LoopInfo *LI,
// care if the pointers are *restrict*.
if (!Stores.size()) {
LLVM_DEBUG(dbgs() << "LAA: Found a read-only loop!\n");
- CanVecMem = true;
- return;
+ return true;
}
MemoryDepChecker::DepCandidates DependentAccesses;
@@ -2574,8 +2568,7 @@ void LoopAccessInfo::analyzeLoop(AAResults *AA, LoopInfo *LI,
LLVM_DEBUG(
dbgs() << "LAA: A loop annotated parallel, ignore memory dependency "
<< "checks.\n");
- CanVecMem = true;
- return;
+ return true;
}
for (LoadInst *LD : Loads) {
@@ -2622,8 +2615,7 @@ void LoopAccessInfo::analyzeLoop(AAResults *AA, LoopInfo *LI,
// other reads in this loop then is it safe to vectorize.
if (NumReadWrites == 1 && NumReads == 0) {
LLVM_DEBUG(dbgs() << "LAA: Found a write-only loop!\n");
- CanVecMem = true;
- return;
+ return true;
}
// Build dependence sets and check whether we need a runtime pointer bounds
@@ -2642,21 +2634,20 @@ void LoopAccessInfo::analyzeLoop(AAResults *AA, LoopInfo *LI,
<< "cannot identify array bounds";
LLVM_DEBUG(dbgs() << "LAA: We can't vectorize because we can't find "
<< "the array bounds.\n");
- CanVecMem = false;
- return;
+ return false;
}
LLVM_DEBUG(
dbgs() << "LAA: May be able to perform a memory runtime check if needed.\n");
- CanVecMem = true;
+ bool DepsAreSafe = true;
if (Accesses.isDependencyCheckNeeded()) {
LLVM_DEBUG(dbgs() << "LAA: Checking memory dependencies\n");
- CanVecMem = DepChecker->areDepsSafe(DependentAccesses,
- Accesses.getDependenciesToCheck(),
- Accesses.getUnderlyingObjects());
+ DepsAreSafe = DepChecker->areDepsSafe(DependentAccesses,
+ Accesses.getDependenciesToCheck(),
+ Accesses.getUnderlyingObjects());
- if (!CanVecMem && DepChecker->shouldRetryWithRuntimeCheck()) {
+ if (!DepsAreSafe && DepChecker->shouldRetryWithRuntimeCheck()) {
LLVM_DEBUG(dbgs() << "LAA: Retrying with memory checks\n");
// Clear the dependency checks. We assume they are not needed.
@@ -2676,30 +2667,30 @@ void LoopAccessInfo::analyzeLoop(AAResults *AA, LoopInfo *LI,
recordAnalysis("CantCheckMemDepsAtRunTime", I)
<< "cannot check memory dependencies at runtime";
LLVM_DEBUG(dbgs() << "LAA: Can't vectorize with memory checks\n");
- CanVecMem = false;
- return;
+ return false;
}
-
- CanVecMem = true;
+ DepsAreSafe = true;
}
}
if (HasConvergentOp) {
recordAnalysis("CantInsertRuntimeCheckWithConvergent")
- << "cannot add control dependency to convergent operation";
+ << "cannot add control dependency to convergent operation";
LLVM_DEBUG(dbgs() << "LAA: We can't vectorize because a runtime check "
"would be needed with a convergent operation\n");
- CanVecMem = false;
- return;
+ return false;
}
- if (CanVecMem)
+ if (DepsAreSafe) {
LLVM_DEBUG(
dbgs() << "LAA: No unsafe dependent memory operations in loop. We"
<< (PtrRtChecking->Need ? "" : " don't")
<< " need runtime memory checks.\n");
- else
- emitUnsafeDependenceRemark();
+ return true;
+ }
+
+ emitUnsafeDependenceRemark();
+ return false;
}
void LoopAccessInfo::emitUnsafeDependenceRemark() {
@@ -2814,7 +2805,7 @@ bool LoopAccessInfo::isInvariant(Value *V) const {
/// stores. This ignores trailing indices that have no effect on the final
/// pointer.
static unsigned getGEPInductionOperand(const GetElementPtrInst *Gep) {
- const DataLayout &DL = Gep->getModule()->getDataLayout();
+ const DataLayout &DL = Gep->getDataLayout();
unsigned LastOperand = Gep->getNumOperands() - 1;
TypeSize GEPAllocSize = DL.getTypeAllocSize(Gep->getResultElementType());
@@ -2856,21 +2847,6 @@ static Value *stripGetElementPtr(Value *Ptr, ScalarEvolution *SE, Loop *Lp) {
return GEP->getOperand(InductionOperand);
}
-/// If a value has only one user that is a CastInst, return it.
-static Value *getUniqueCastUse(Value *Ptr, Loop *Lp, Type *Ty) {
- Value *UniqueCast = nullptr;
- for (User *U : Ptr->users()) {
- CastInst *CI = dyn_cast<CastInst>(U);
- if (CI && CI->getType() == Ty) {
- if (!UniqueCast)
- UniqueCast = CI;
- else
- return nullptr;
- }
- }
- return UniqueCast;
-}
-
/// Get the stride of a pointer access in a loop. Looks for symbolic
/// strides "a[i*stride]". Returns the symbolic stride, or null otherwise.
static const SCEV *getStrideFromPointer(Value *Ptr, ScalarEvolution *SE, Loop *Lp) {
@@ -2933,21 +2909,14 @@ static const SCEV *getStrideFromPointer(Value *Ptr, ScalarEvolution *SE, Loop *L
return nullptr;
// Look for the loop invariant symbolic value.
- const SCEVUnknown *U = dyn_cast<SCEVUnknown>(V);
- if (!U) {
- const auto *C = dyn_cast<SCEVIntegralCastExpr>(V);
- if (!C)
- return nullptr;
- U = dyn_cast<SCEVUnknown>(C->getOperand());
- if (!U)
- return nullptr;
+ if (isa<SCEVUnknown>(V))
+ return V;
- // Match legacy behavior - this is not needed for correctness
- if (!getUniqueCastUse(U->getValue(), Lp, V->getType()))
- return nullptr;
- }
+ if (const auto *C = dyn_cast<SCEVIntegralCastExpr>(V))
+ if (isa<SCEVUnknown>(C->getOperand()))
+ return V;
- return V;
+ return nullptr;
}
void LoopAccessInfo::collectStridedAccess(Value *MemAccess) {
@@ -2992,7 +2961,7 @@ void LoopAccessInfo::collectStridedAccess(Value *MemAccess) {
// Match the types so we can compare the stride and the MaxBTC.
// The Stride can be positive/negative, so we sign extend Stride;
// The backedgeTakenCount is non-negative, so we zero extend MaxBTC.
- const DataLayout &DL = TheLoop->getHeader()->getModule()->getDataLayout();
+ const DataLayout &DL = TheLoop->getHeader()->getDataLayout();
uint64_t StrideTypeSizeBits = DL.getTypeSizeInBits(StrideExpr->getType());
uint64_t BETypeSizeBits = DL.getTypeSizeInBits(MaxBTC->getType());
const SCEV *CastedStride = StrideExpr;
@@ -3048,7 +3017,7 @@ LoopAccessInfo::LoopAccessInfo(Loop *L, ScalarEvolution *SE,
MaxTargetVectorWidthInBits);
PtrRtChecking = std::make_unique<RuntimePointerChecking>(*DepChecker, SE);
if (canAnalyzeLoop())
- analyzeLoop(AA, LI, TLI, DT);
+ CanVecMem = analyzeLoop(AA, LI, TLI, DT);
}
void LoopAccessInfo::print(raw_ostream &OS, unsigned Depth) const {
generated by cgit v1.2.3 (git 2.25.1) at 2026-05-31 11:33:47 +0000