}
SmallVector<RuntimePointerChecking::PointerCheck, 4>
-RuntimePointerChecking::generateChecks(
- const SmallVectorImpl<int> *PtrPartition) const {
+RuntimePointerChecking::generateChecks() const {
SmallVector<PointerCheck, 4> Checks;
for (unsigned I = 0; I < CheckingGroups.size(); ++I) {
const RuntimePointerChecking::CheckingPtrGroup &CGI = CheckingGroups[I];
const RuntimePointerChecking::CheckingPtrGroup &CGJ = CheckingGroups[J];
- if (needsChecking(CGI, CGJ, PtrPartition))
+ if (needsChecking(CGI, CGJ))
Checks.push_back(std::make_pair(&CGI, &CGJ));
}
}
return Checks;
}
-bool RuntimePointerChecking::needsChecking(
- const CheckingPtrGroup &M, const CheckingPtrGroup &N,
- const SmallVectorImpl<int> *PtrPartition) const {
+void RuntimePointerChecking::generateChecks(
+ MemoryDepChecker::DepCandidates &DepCands, bool UseDependencies) {
+ assert(Checks.empty() && "Checks is not empty");
+ groupChecks(DepCands, UseDependencies);
+ Checks = generateChecks();
+}
+
+bool RuntimePointerChecking::needsChecking(const CheckingPtrGroup &M,
+ const CheckingPtrGroup &N) const {
for (unsigned I = 0, EI = M.Members.size(); EI != I; ++I)
for (unsigned J = 0, EJ = N.Members.size(); EJ != J; ++J)
- if (needsChecking(M.Members[I], N.Members[J], PtrPartition))
+ if (needsChecking(M.Members[I], N.Members[J]))
return true;
return false;
}
CheckingGroups.clear();
+ // If we need to check two pointers to the same underlying object
+ // with a non-constant difference, we shouldn't perform any pointer
+ // grouping with those pointers. This is because we can easily get
+ // into cases where the resulting check would return false, even when
+ // the accesses are safe.
+ //
+ // The following example shows this:
+ // for (i = 0; i < 1000; ++i)
+ // a[5000 + i * m] = a[i] + a[i + 9000]
+ //
+ // Here grouping gives a check of (5000, 5000 + 1000 * m) against
+ // (0, 10000) which is always false. However, if m is 1, there is no
+ // dependence. Not grouping the checks for a[i] and a[i + 9000] allows
+ // us to perform an accurate check in this case.
+ //
+ // The above case requires that we have an UnknownDependence between
+ // accesses to the same underlying object. This cannot happen unless
+ // ShouldRetryWithRuntimeCheck is set, and therefore UseDependencies
+ // is also false. In this case we will use the fallback path and create
+ // separate checking groups for all pointers.
+
// If we don't have the dependency partitions, construct a new
- // checking pointer group for each pointer.
+ // checking pointer group for each pointer. This is also required
+ // for correctness, because in this case we can have checking between
+ // pointers to the same underlying object.
if (!UseDependencies) {
for (unsigned I = 0; I < Pointers.size(); ++I)
CheckingGroups.push_back(CheckingPtrGroup(I, *this));
PtrToPartition[PtrIdx1] == PtrToPartition[PtrIdx2]);
}
-bool RuntimePointerChecking::needsChecking(
- unsigned I, unsigned J, const SmallVectorImpl<int> *PtrPartition) const {
+bool RuntimePointerChecking::needsChecking(unsigned I, unsigned J) const {
const PointerInfo &PointerI = Pointers[I];
const PointerInfo &PointerJ = Pointers[J];
if (PointerI.AliasSetId != PointerJ.AliasSetId)
return false;
- // If PtrPartition is set omit checks between pointers of the same partition.
- if (PtrPartition && arePointersInSamePartition(*PtrPartition, I, J))
- return false;
-
return true;
}
}
}
-void RuntimePointerChecking::print(
- raw_ostream &OS, unsigned Depth,
- const SmallVectorImpl<int> *PtrPartition) const {
+void RuntimePointerChecking::print(raw_ostream &OS, unsigned Depth) const {
OS.indent(Depth) << "Run-time memory checks:\n";
- printChecks(OS, generateChecks(PtrPartition), Depth);
+ printChecks(OS, Checks, Depth);
OS.indent(Depth) << "Grouped accesses:\n";
for (unsigned I = 0; I < CheckingGroups.size(); ++I) {
}
}
-unsigned RuntimePointerChecking::getNumberOfChecks(
- const SmallVectorImpl<int> *PtrPartition) const {
-
- unsigned NumPartitions = CheckingGroups.size();
- unsigned CheckCount = 0;
-
- for (unsigned I = 0; I < NumPartitions; ++I)
- for (unsigned J = I + 1; J < NumPartitions; ++J)
- if (needsChecking(CheckingGroups[I], CheckingGroups[J], PtrPartition))
- CheckCount++;
- return CheckCount;
-}
-
-bool RuntimePointerChecking::needsAnyChecking(
- const SmallVectorImpl<int> *PtrPartition) const {
- unsigned NumPointers = Pointers.size();
-
- for (unsigned I = 0; I < NumPointers; ++I)
- for (unsigned J = I + 1; J < NumPointers; ++J)
- if (needsChecking(I, J, PtrPartition))
- return true;
- return false;
-}
-
namespace {
/// \brief Analyses memory accesses in a loop.
///
}
if (NeedRTCheck && CanDoRT)
- RtCheck.groupChecks(DepCands, IsDepCheckNeeded);
+ RtCheck.generateChecks(DepCands, IsDepCheckNeeded);
- DEBUG(dbgs() << "LAA: We need to do " << RtCheck.getNumberOfChecks(nullptr)
+ DEBUG(dbgs() << "LAA: We need to do " << RtCheck.getNumberOfChecks()
<< " pointer comparisons.\n");
RtCheck.Need = NeedRTCheck;
/// \brief Check whether the access through \p Ptr has a constant stride.
int llvm::isStridedPtr(ScalarEvolution *SE, Value *Ptr, const Loop *Lp,
const ValueToValueMap &StridesMap) {
- const Type *Ty = Ptr->getType();
+ Type *Ty = Ptr->getType();
assert(Ty->isPointerTy() && "Unexpected non-ptr");
// Make sure that the pointer does not point to aggregate types.
-
const PointerType *PtrTy = cast<PointerType>(Ty);
+
auto *PtrTy = cast<PointerType>(Ty);
if (PtrTy->getElementType()->isAggregateType()) {
DEBUG(dbgs() << "LAA: Bad stride - Not a pointer to a scalar type"
<< *Ptr << "\n");
return nullptr;
}
-/// \brief IR Values for the lower and upper bounds of a pointer evolution.
+/// \brief IR Values for the lower and upper bounds of a pointer evolution. We
+/// need to use value-handles because SCEV expansion can invalidate previously
+/// expanded values. Thus expansion of a pointer can invalidate the bounds for
+/// a previous one.
struct PointerBounds {
- Value *Start;
- Value *End;
+ TrackingVH<Value> Start;
+ TrackingVH<Value> End;
};
/// \brief Expand code for the lower and upper bound of the pointer group \p CG
return ChecksWithBounds;
}
-std::pair<Instruction *, Instruction *> LoopAccessInfo::addRuntimeCheck(
+std::pair<Instruction *, Instruction *> LoopAccessInfo::addRuntimeChecks(
Instruction *Loc,
const SmallVectorImpl<RuntimePointerChecking::PointerCheck> &PointerChecks)
const {
for (const auto &Check : ExpandedChecks) {
const PointerBounds &A = Check.first, &B = Check.second;
+ // Check if two pointers (A and B) conflict where conflict is computed as:
+ // start(A) <= end(B) && start(B) <= end(A)
unsigned AS0 = A.Start->getType()->getPointerAddressSpace();
unsigned AS1 = B.Start->getType()->getPointerAddressSpace();
return std::make_pair(FirstInst, Check);
}
-std::pair<Instruction *, Instruction *> LoopAccessInfo::addRuntimeCheck(
- Instruction *Loc, const SmallVectorImpl<int> *PtrPartition) const {
+std::pair<Instruction *, Instruction *>
+LoopAccessInfo::addRuntimeChecks(Instruction *Loc) const {
if (!PtrRtChecking.Need)
return std::make_pair(nullptr, nullptr);
- return addRuntimeCheck(Loc, PtrRtChecking.generateChecks(PtrPartition));
+ return addRuntimeChecks(Loc, PtrRtChecking.getChecks());
}
LoopAccessInfo::LoopAccessInfo(Loop *L, ScalarEvolution *SE,
}
bool LoopAccessAnalysis::runOnFunction(Function &F) {
- SE = &getAnalysis<ScalarEvolution>();
+ SE = &getAnalysis<ScalarEvolutionWrapperPass>().getSE();
auto *TLIP = getAnalysisIfAvailable<TargetLibraryInfoWrapperPass>();
TLI = TLIP ? &TLIP->getTLI() : nullptr;
- AA = &getAnalysis<AliasAnalysis>();
+ AA = &getAnalysis<AAResultsWrapperPass>().getAAResults();
DT = &getAnalysis<DominatorTreeWrapperPass>().getDomTree();
LI = &getAnalysis<LoopInfoWrapperPass>().getLoopInfo();
}
void LoopAccessAnalysis::getAnalysisUsage(AnalysisUsage &AU) const {
- AU.addRequired<ScalarEvolution>();
- AU.addRequired<AliasAnalysis>();
+ AU.addRequired<ScalarEvolutionWrapperPass>();
+ AU.addRequired<AAResultsWrapperPass>();
AU.addRequired<DominatorTreeWrapperPass>();
AU.addRequired<LoopInfoWrapperPass>();
#define LAA_NAME "loop-accesses"
INITIALIZE_PASS_BEGIN(LoopAccessAnalysis, LAA_NAME, laa_name, false, true)
-INITIALIZE_AG_DEPENDENCY(AliasAnalysis)
-INITIALIZE_PASS_DEPENDENCY(ScalarEvolution)
+INITIALIZE_PASS_DEPENDENCY(AAResultsWrapperPass)
+INITIALIZE_PASS_DEPENDENCY(ScalarEvolutionWrapperPass)
INITIALIZE_PASS_DEPENDENCY(DominatorTreeWrapperPass)
INITIALIZE_PASS_DEPENDENCY(LoopInfoWrapperPass)
INITIALIZE_PASS_END(LoopAccessAnalysis, LAA_NAME, laa_name, false, true)
projects / oota-llvm.git / blobdiff