#include "llvm/ADT/SmallVector.h"
#include "llvm/ADT/StringExtras.h"
#include "llvm/Analysis/AliasAnalysis.h"
-#include "llvm/Analysis/Dominators.h"
#include "llvm/Analysis/LoopInfo.h"
#include "llvm/Analysis/LoopIterator.h"
#include "llvm/Analysis/LoopPass.h"
#include "llvm/Analysis/ScalarEvolutionExpressions.h"
#include "llvm/Analysis/TargetTransformInfo.h"
#include "llvm/Analysis/ValueTracking.h"
-#include "llvm/Analysis/Verifier.h"
#include "llvm/IR/Constants.h"
#include "llvm/IR/DataLayout.h"
#include "llvm/IR/DerivedTypes.h"
+#include "llvm/IR/Dominators.h"
#include "llvm/IR/Function.h"
#include "llvm/IR/IRBuilder.h"
#include "llvm/IR/Instructions.h"
#include "llvm/IR/Module.h"
#include "llvm/IR/Type.h"
#include "llvm/IR/Value.h"
+#include "llvm/IR/Verifier.h"
#include "llvm/Pass.h"
#include "llvm/Support/CommandLine.h"
#include "llvm/Support/Debug.h"
#include "llvm/Support/PatternMatch.h"
-#include "llvm/Support/raw_ostream.h"
#include "llvm/Support/ValueHandle.h"
+#include "llvm/Support/raw_ostream.h"
#include "llvm/Target/TargetLibraryInfo.h"
#include "llvm/Transforms/Scalar.h"
#include "llvm/Transforms/Utils/BasicBlockUtils.h"
"trip count that is smaller than this "
"value."));
+/// This enables versioning on the strides of symbolically striding memory
+/// accesses in code like the following.
+/// for (i = 0; i < N; ++i)
+/// A[i * Stride1] += B[i * Stride2] ...
+///
+/// Will be roughly translated to
+/// if (Stride1 == 1 && Stride2 == 1) {
+/// for (i = 0; i < N; i+=4)
+/// A[i:i+3] += ...
+/// } else
+/// ...
+static cl::opt<bool> EnableMemAccessVersioning(
+ "enable-mem-access-versioning", cl::init(true), cl::Hidden,
+ cl::desc("Enable symblic stride memory access versioning"));
+
/// We don't unroll loops with a known constant trip count below this number.
static const unsigned TinyTripCountUnrollThreshold = 128;
unsigned UnrollFactor)
: OrigLoop(OrigLoop), SE(SE), LI(LI), DT(DT), DL(DL), TLI(TLI),
VF(VecWidth), UF(UnrollFactor), Builder(SE->getContext()), Induction(0),
- OldInduction(0), WidenMap(UnrollFactor) {}
+ OldInduction(0), WidenMap(UnrollFactor), Legal(0) {}
// Perform the actual loop widening (vectorization).
- void vectorize(LoopVectorizationLegality *Legal) {
+ void vectorize(LoopVectorizationLegality *L) {
+ Legal = L;
// Create a new empty loop. Unlink the old loop and connect the new one.
- createEmptyLoop(Legal);
+ createEmptyLoop();
// Widen each instruction in the old loop to a new one in the new loop.
// Use the Legality module to find the induction and reduction variables.
- vectorizeLoop(Legal);
+ vectorizeLoop();
// Register the new loop and update the analysis passes.
updateAnalysis();
}
typedef DenseMap<std::pair<BasicBlock*, BasicBlock*>,
VectorParts> EdgeMaskCache;
- /// Add code that checks at runtime if the accessed arrays overlap.
- /// Returns the comparator value or NULL if no check is needed.
- Instruction *addRuntimeCheck(LoopVectorizationLegality *Legal,
- Instruction *Loc);
+ /// \brief Add code that checks at runtime if the accessed arrays overlap.
+ ///
+ /// Returns a pair of instructions where the first element is the first
+ /// instruction generated in possibly a sequence of instructions and the
+ /// second value is the final comparator value or NULL if no check is needed.
+ std::pair<Instruction *, Instruction *> addRuntimeCheck(Instruction *Loc);
+
+ /// \brief Add checks for strides that where assumed to be 1.
+ ///
+ /// Returns the last check instruction and the first check instruction in the
+ /// pair as (first, last).
+ std::pair<Instruction *, Instruction *> addStrideCheck(Instruction *Loc);
+
/// Create an empty loop, based on the loop ranges of the old loop.
- void createEmptyLoop(LoopVectorizationLegality *Legal);
+ void createEmptyLoop();
/// Copy and widen the instructions from the old loop.
- virtual void vectorizeLoop(LoopVectorizationLegality *Legal);
+ virtual void vectorizeLoop();
/// \brief The Loop exit block may have single value PHI nodes where the
/// incoming value is 'Undef'. While vectorizing we only handled real values
VectorParts createEdgeMask(BasicBlock *Src, BasicBlock *Dst);
/// A helper function to vectorize a single BB within the innermost loop.
- void vectorizeBlockInLoop(LoopVectorizationLegality *Legal, BasicBlock *BB,
- PhiVector *PV);
+ void vectorizeBlockInLoop(BasicBlock *BB, PhiVector *PV);
/// Vectorize a single PHINode in a block. This method handles the induction
/// variable canonicalization. It supports both VF = 1 for unrolled loops and
/// arbitrary length vectors.
void widenPHIInstruction(Instruction *PN, VectorParts &Entry,
- LoopVectorizationLegality *Legal,
unsigned UF, unsigned VF, PhiVector *PV);
/// Insert the new loop to the loop hierarchy and pass manager
virtual void scalarizeInstruction(Instruction *Instr);
/// Vectorize Load and Store instructions,
- virtual void vectorizeMemoryInstruction(Instruction *Instr,
- LoopVectorizationLegality *Legal);
+ virtual void vectorizeMemoryInstruction(Instruction *Instr);
/// Create a broadcast instruction. This method generates a broadcast
/// instruction (shuffle) for loop invariant values and for the induction
/// Maps scalars to widened vectors.
ValueMap WidenMap;
EdgeMaskCache MaskCache;
+
+ LoopVectorizationLegality *Legal;
};
class InnerLoopUnroller : public InnerLoopVectorizer {
private:
virtual void scalarizeInstruction(Instruction *Instr);
- virtual void vectorizeMemoryInstruction(Instruction *Instr,
- LoopVectorizationLegality *Legal);
+ virtual void vectorizeMemoryInstruction(Instruction *Instr);
virtual Value *getBroadcastInstrs(Value *V);
virtual Value *getConsecutiveVector(Value* Val, int StartIdx, bool Negate);
virtual Value *reverseVector(Value *Vec);
/// Insert a pointer and calculate the start and end SCEVs.
void insert(ScalarEvolution *SE, Loop *Lp, Value *Ptr, bool WritePtr,
- unsigned DepSetId);
+ unsigned DepSetId, ValueToValueMap &Strides);
/// This flag indicates if we need to add the runtime check.
bool Need;
/// pointer itself is an induction variable.
/// This check allows us to vectorize A[idx] into a wide load/store.
/// Returns:
- /// 0 - Stride is unknown or non consecutive.
+ /// 0 - Stride is unknown or non-consecutive.
/// 1 - Address is consecutive.
/// -1 - Address is consecutive, and decreasing.
int isConsecutivePtr(Value *Ptr);
unsigned getMaxSafeDepDistBytes() { return MaxSafeDepDistBytes; }
+ bool hasStride(Value *V) { return StrideSet.count(V); }
+ bool mustCheckStrides() { return !StrideSet.empty(); }
+ SmallPtrSet<Value *, 8>::iterator strides_begin() {
+ return StrideSet.begin();
+ }
+ SmallPtrSet<Value *, 8>::iterator strides_end() { return StrideSet.end(); }
+
private:
/// Check if a single basic block loop is vectorizable.
/// At this point we know that this is a loop with a constant trip count
/// if the PHI is not an induction variable.
InductionKind isInductionVariable(PHINode *Phi);
+ /// \brief Collect memory access with loop invariant strides.
+ ///
+ /// Looks for accesses like "a[i * StrideA]" where "StrideA" is loop
+ /// invariant.
+ void collectStridedAcccess(Value *LoadOrStoreInst);
+
/// The loop that we evaluate.
Loop *TheLoop;
/// Scev analysis.
bool HasFunNoNaNAttr;
unsigned MaxSafeDepDistBytes;
+
+ ValueToValueMap Strides;
+ SmallPtrSet<Value *, 8> StrideSet;
};
/// LoopVectorizationCostModel - estimates the expected speedups due to
unsigned Width;
/// Vectorization unroll factor.
unsigned Unroll;
+ /// Vectorization forced (-1 not selected, 0 force disabled, 1 force enabled)
+ int Force;
LoopVectorizeHints(const Loop *L, bool DisableUnrolling)
: Width(VectorizationFactor)
, Unroll(DisableUnrolling ? 1 : VectorizationUnroll)
+ , Force(-1)
, LoopID(L->getLoopID()) {
getHints(L);
// The command line options override any loop metadata except for when
Unroll = Val;
else
DEBUG(dbgs() << "LV: ignoring invalid unroll hint metadata\n");
+ } else if (Hint == "enable") {
+ if (C->getBitWidth() == 1)
+ Force = Val;
+ else
+ DEBUG(dbgs() << "LV: ignoring invalid enable hint metadata\n");
} else {
DEBUG(dbgs() << "LV: ignoring unknown hint " << Hint << '\n');
}
/// Pass identification, replacement for typeid
static char ID;
- explicit LoopVectorize(bool NoUnrolling = false)
- : LoopPass(ID), DisableUnrolling(NoUnrolling) {
+ explicit LoopVectorize(bool NoUnrolling = false, bool AlwaysVectorize = true)
+ : LoopPass(ID),
+ DisableUnrolling(NoUnrolling),
+ AlwaysVectorize(AlwaysVectorize) {
initializeLoopVectorizePass(*PassRegistry::getPassRegistry());
}
DominatorTree *DT;
TargetLibraryInfo *TLI;
bool DisableUnrolling;
+ bool AlwaysVectorize;
virtual bool runOnLoop(Loop *L, LPPassManager &LPM) {
// We only vectorize innermost loops.
DL = getAnalysisIfAvailable<DataLayout>();
LI = &getAnalysis<LoopInfo>();
TTI = &getAnalysis<TargetTransformInfo>();
- DT = &getAnalysis<DominatorTree>();
+ DT = &getAnalysis<DominatorTreeWrapperPass>().getDomTree();
TLI = getAnalysisIfAvailable<TargetLibraryInfo>();
// If the target claims to have no vector registers don't attempt
return false;
if (DL == NULL) {
- DEBUG(dbgs() << "LV: Not vectorizing because of missing data layout\n");
+ DEBUG(dbgs() << "LV: Not vectorizing: Missing data layout\n");
return false;
}
LoopVectorizeHints Hints(L, DisableUnrolling);
+ if (Hints.Force == 0) {
+ DEBUG(dbgs() << "LV: Not vectorizing: #pragma vectorize disable.\n");
+ return false;
+ }
+
+ if (!AlwaysVectorize && Hints.Force != 1) {
+ DEBUG(dbgs() << "LV: Not vectorizing: No #pragma vectorize enable.\n");
+ return false;
+ }
+
if (Hints.Width == 1 && Hints.Unroll == 1) {
- DEBUG(dbgs() << "LV: Not vectorizing.\n");
+ DEBUG(dbgs() << "LV: Not vectorizing: Disabled/already vectorized.\n");
return false;
}
// Check if it is legal to vectorize the loop.
LoopVectorizationLegality LVL(L, SE, DL, DT, TLI);
if (!LVL.canVectorize()) {
- DEBUG(dbgs() << "LV: Not vectorizing.\n");
+ DEBUG(dbgs() << "LV: Not vectorizing: Cannot prove legality.\n");
return false;
}
Attribute::AttrKind SzAttr = Attribute::OptimizeForSize;
Attribute::AttrKind FlAttr = Attribute::NoImplicitFloat;
unsigned FnIndex = AttributeSet::FunctionIndex;
- bool OptForSize = F->getAttributes().hasAttribute(FnIndex, SzAttr);
+ bool OptForSize = Hints.Force != 1 &&
+ F->getAttributes().hasAttribute(FnIndex, SzAttr);
bool NoFloat = F->getAttributes().hasAttribute(FnIndex, FlAttr);
if (NoFloat) {
DEBUG(dbgs() << "LV: Vectorization is possible but not beneficial.\n");
if (UF == 1)
return false;
+ DEBUG(dbgs() << "LV: Trying to at least unroll the loops.\n");
// We decided not to vectorize, but we may want to unroll.
InnerLoopUnroller Unroller(L, SE, LI, DT, DL, TLI, UF);
Unroller.vectorize(&LVL);
LoopPass::getAnalysisUsage(AU);
AU.addRequiredID(LoopSimplifyID);
AU.addRequiredID(LCSSAID);
- AU.addRequired<DominatorTree>();
+ AU.addRequired<DominatorTreeWrapperPass>();
AU.addRequired<LoopInfo>();
AU.addRequired<ScalarEvolution>();
AU.addRequired<TargetTransformInfo>();
AU.addPreserved<LoopInfo>();
- AU.addPreserved<DominatorTree>();
+ AU.addPreserved<DominatorTreeWrapperPass>();
}
};
// LoopVectorizationCostModel.
//===----------------------------------------------------------------------===//
-void
-LoopVectorizationLegality::RuntimePointerCheck::insert(ScalarEvolution *SE,
- Loop *Lp, Value *Ptr,
- bool WritePtr,
- unsigned DepSetId) {
- const SCEV *Sc = SE->getSCEV(Ptr);
+static Value *stripCast(Value *V) {
+ if (CastInst *CI = dyn_cast<CastInst>(V))
+ return CI->getOperand(0);
+ return V;
+}
+
+///\brief Replaces the symbolic stride in a pointer SCEV expression by one.
+///
+/// If \p OrigPtr is not null, use it to look up the stride value instead of
+/// \p Ptr.
+static const SCEV *replaceSymbolicStrideSCEV(ScalarEvolution *SE,
+ ValueToValueMap &PtrToStride,
+ Value *Ptr, Value *OrigPtr = 0) {
+
+ const SCEV *OrigSCEV = SE->getSCEV(Ptr);
+
+ // If there is an entry in the map return the SCEV of the pointer with the
+ // symbolic stride replaced by one.
+ ValueToValueMap::iterator SI = PtrToStride.find(OrigPtr ? OrigPtr : Ptr);
+ if (SI != PtrToStride.end()) {
+ Value *StrideVal = SI->second;
+
+ // Strip casts.
+ StrideVal = stripCast(StrideVal);
+
+ // Replace symbolic stride by one.
+ Value *One = ConstantInt::get(StrideVal->getType(), 1);
+ ValueToValueMap RewriteMap;
+ RewriteMap[StrideVal] = One;
+
+ const SCEV *ByOne =
+ SCEVParameterRewriter::rewrite(OrigSCEV, *SE, RewriteMap, true);
+ DEBUG(dbgs() << "LV: Replacing SCEV: " << *OrigSCEV << " by: " << *ByOne
+ << "\n");
+ return ByOne;
+ }
+
+ // Otherwise, just return the SCEV of the original pointer.
+ return SE->getSCEV(Ptr);
+}
+
+void LoopVectorizationLegality::RuntimePointerCheck::insert(
+ ScalarEvolution *SE, Loop *Lp, Value *Ptr, bool WritePtr, unsigned DepSetId,
+ ValueToValueMap &Strides) {
+ // Get the stride replaced scev.
+ const SCEV *Sc = replaceSymbolicStrideSCEV(SE, Strides, Ptr);
const SCEVAddRecExpr *AR = dyn_cast<SCEVAddRecExpr>(Sc);
assert(AR && "Invalid addrec expression");
const SCEV *Ex = SE->getBackedgeTakenCount(Lp);
}
int LoopVectorizationLegality::isConsecutivePtr(Value *Ptr) {
- assert(Ptr->getType()->isPointerTy() && "Unexpected non ptr");
+ assert(Ptr->getType()->isPointerTy() && "Unexpected non-ptr");
// Make sure that the pointer does not point to structs.
if (Ptr->getType()->getPointerElementType()->isAggregateType())
return 0;
// We can emit wide load/stores only if the last non-zero index is the
// induction variable.
- const SCEV *Last = SE->getSCEV(Gep->getOperand(InductionOperand));
+ const SCEV *Last = 0;
+ if (!Strides.count(Gep))
+ Last = SE->getSCEV(Gep->getOperand(InductionOperand));
+ else {
+ // Because of the multiplication by a stride we can have a s/zext cast.
+ // We are going to replace this stride by 1 so the cast is safe to ignore.
+ //
+ // %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
+ // %0 = trunc i64 %indvars.iv to i32
+ // %mul = mul i32 %0, %Stride1
+ // %idxprom = zext i32 %mul to i64 << Safe cast.
+ // %arrayidx = getelementptr inbounds i32* %B, i64 %idxprom
+ //
+ Last = replaceSymbolicStrideSCEV(SE, Strides,
+ Gep->getOperand(InductionOperand), Gep);
+ if (const SCEVCastExpr *C = dyn_cast<SCEVCastExpr>(Last))
+ Last =
+ (C->getSCEVType() == scSignExtend || C->getSCEVType() == scZeroExtend)
+ ? C->getOperand()
+ : Last;
+ }
if (const SCEVAddRecExpr *AR = dyn_cast<SCEVAddRecExpr>(Last)) {
const SCEV *Step = AR->getStepRecurrence(*SE);
assert(V != Induction && "The new induction variable should not be used.");
assert(!V->getType()->isVectorTy() && "Can't widen a vector");
+ // If we have a stride that is replaced by one, do it here.
+ if (Legal->hasStride(V))
+ V = ConstantInt::get(V->getType(), 1);
+
// If we have this scalar in the map, return it.
if (WidenMap.has(V))
return WidenMap.get(V);
"reverse");
}
-
-void InnerLoopVectorizer::vectorizeMemoryInstruction(Instruction *Instr,
- LoopVectorizationLegality *Legal) {
+void InnerLoopVectorizer::vectorizeMemoryInstruction(Instruction *Instr) {
// Attempt to issue a wide load.
LoadInst *LI = dyn_cast<LoadInst>(Instr);
StoreInst *SI = dyn_cast<StoreInst>(Instr);
Type *DataTy = VectorType::get(ScalarDataTy, VF);
Value *Ptr = LI ? LI->getPointerOperand() : SI->getPointerOperand();
unsigned Alignment = LI ? LI->getAlignment() : SI->getAlignment();
+ // An alignment of 0 means target abi alignment. We need to use the scalar's
+ // target abi alignment in such a case.
+ if (!Alignment)
+ Alignment = DL->getABITypeAlignment(ScalarDataTy);
unsigned AddressSpace = Ptr->getType()->getPointerAddressSpace();
unsigned ScalarAllocatedSize = DL->getTypeAllocSize(ScalarDataTy);
unsigned VectorElementSize = DL->getTypeStoreSize(DataTy)/VF;
if (ScalarAllocatedSize != VectorElementSize)
return scalarizeInstruction(Instr);
- // If the pointer is loop invariant or if it is non consecutive,
+ // If the pointer is loop invariant or if it is non-consecutive,
// scalarize the load.
int ConsecutiveStride = Legal->isConsecutivePtr(Ptr);
bool Reverse = ConsecutiveStride < 0;
}
}
-Instruction *
-InnerLoopVectorizer::addRuntimeCheck(LoopVectorizationLegality *Legal,
- Instruction *Loc) {
+static Instruction *getFirstInst(Instruction *FirstInst, Value *V,
+ Instruction *Loc) {
+ if (FirstInst)
+ return FirstInst;
+ if (Instruction *I = dyn_cast<Instruction>(V))
+ return I->getParent() == Loc->getParent() ? I : 0;
+ return 0;
+}
+
+std::pair<Instruction *, Instruction *>
+InnerLoopVectorizer::addStrideCheck(Instruction *Loc) {
+ Instruction *tnullptr = 0;
+ if (!Legal->mustCheckStrides())
+ return std::pair<Instruction *, Instruction *>(tnullptr, tnullptr);
+
+ IRBuilder<> ChkBuilder(Loc);
+
+ // Emit checks.
+ Value *Check = 0;
+ Instruction *FirstInst = 0;
+ for (SmallPtrSet<Value *, 8>::iterator SI = Legal->strides_begin(),
+ SE = Legal->strides_end();
+ SI != SE; ++SI) {
+ Value *Ptr = stripCast(*SI);
+ Value *C = ChkBuilder.CreateICmpNE(Ptr, ConstantInt::get(Ptr->getType(), 1),
+ "stride.chk");
+ // Store the first instruction we create.
+ FirstInst = getFirstInst(FirstInst, C, Loc);
+ if (Check)
+ Check = ChkBuilder.CreateOr(Check, C);
+ else
+ Check = C;
+ }
+
+ // We have to do this trickery because the IRBuilder might fold the check to a
+ // constant expression in which case there is no Instruction anchored in a
+ // the block.
+ LLVMContext &Ctx = Loc->getContext();
+ Instruction *TheCheck =
+ BinaryOperator::CreateAnd(Check, ConstantInt::getTrue(Ctx));
+ ChkBuilder.Insert(TheCheck, "stride.not.one");
+ FirstInst = getFirstInst(FirstInst, TheCheck, Loc);
+
+ return std::make_pair(FirstInst, TheCheck);
+}
+
+std::pair<Instruction *, Instruction *>
+InnerLoopVectorizer::addRuntimeCheck(Instruction *Loc) {
LoopVectorizationLegality::RuntimePointerCheck *PtrRtCheck =
Legal->getRuntimePointerCheck();
+ Instruction *tnullptr = 0;
if (!PtrRtCheck->Need)
- return NULL;
+ return std::pair<Instruction *, Instruction *>(tnullptr, tnullptr);
unsigned NumPointers = PtrRtCheck->Pointers.size();
SmallVector<TrackingVH<Value> , 2> Starts;
LLVMContext &Ctx = Loc->getContext();
SCEVExpander Exp(*SE, "induction");
+ Instruction *FirstInst = 0;
for (unsigned i = 0; i < NumPointers; ++i) {
Value *Ptr = PtrRtCheck->Pointers[i];
Value *End1 = ChkBuilder.CreateBitCast(Ends[j], PtrArithTy0, "bc");
Value *Cmp0 = ChkBuilder.CreateICmpULE(Start0, End1, "bound0");
+ FirstInst = getFirstInst(FirstInst, Cmp0, Loc);
Value *Cmp1 = ChkBuilder.CreateICmpULE(Start1, End0, "bound1");
+ FirstInst = getFirstInst(FirstInst, Cmp1, Loc);
Value *IsConflict = ChkBuilder.CreateAnd(Cmp0, Cmp1, "found.conflict");
- if (MemoryRuntimeCheck)
+ FirstInst = getFirstInst(FirstInst, IsConflict, Loc);
+ if (MemoryRuntimeCheck) {
IsConflict = ChkBuilder.CreateOr(MemoryRuntimeCheck, IsConflict,
"conflict.rdx");
+ FirstInst = getFirstInst(FirstInst, IsConflict, Loc);
+ }
MemoryRuntimeCheck = IsConflict;
}
}
Instruction *Check = BinaryOperator::CreateAnd(MemoryRuntimeCheck,
ConstantInt::getTrue(Ctx));
ChkBuilder.Insert(Check, "memcheck.conflict");
- return Check;
+ FirstInst = getFirstInst(FirstInst, Check, Loc);
+ return std::make_pair(FirstInst, Check);
}
-void
-InnerLoopVectorizer::createEmptyLoop(LoopVectorizationLegality *Legal) {
+void InnerLoopVectorizer::createEmptyLoop() {
/*
In this function we generate a new loop. The new loop will contain
the vectorized instructions while the old loop will continue to run the
const SCEV *ExitCount = SE->getBackedgeTakenCount(OrigLoop);
assert(ExitCount != SE->getCouldNotCompute() && "Invalid loop count");
+ // The exit count might have the type of i64 while the phi is i32. This can
+ // happen if we have an induction variable that is sign extended before the
+ // compare. The only way that we get a backedge taken count is that the
+ // induction variable was signed and as such will not overflow. In such a case
+ // truncation is legal.
+ if (ExitCount->getType()->getPrimitiveSizeInBits() >
+ IdxTy->getPrimitiveSizeInBits())
+ ExitCount = SE->getTruncateOrNoop(ExitCount, IdxTy);
+
+ ExitCount = SE->getNoopOrZeroExtend(ExitCount, IdxTy);
// Get the total trip count from the count by adding 1.
ExitCount = SE->getAddExpr(ExitCount,
SE->getConstant(ExitCount->getType(), 1));
BasicBlock *LastBypassBlock = BypassBlock;
+ // Generate the code to check that the strides we assumed to be one are really
+ // one. We want the new basic block to start at the first instruction in a
+ // sequence of instructions that form a check.
+ Instruction *StrideCheck;
+ Instruction *FirstCheckInst;
+ tie(FirstCheckInst, StrideCheck) =
+ addStrideCheck(BypassBlock->getTerminator());
+ if (StrideCheck) {
+ // Create a new block containing the stride check.
+ BasicBlock *CheckBlock =
+ BypassBlock->splitBasicBlock(FirstCheckInst, "vector.stridecheck");
+ if (ParentLoop)
+ ParentLoop->addBasicBlockToLoop(CheckBlock, LI->getBase());
+ LoopBypassBlocks.push_back(CheckBlock);
+
+ // Replace the branch into the memory check block with a conditional branch
+ // for the "few elements case".
+ Instruction *OldTerm = BypassBlock->getTerminator();
+ BranchInst::Create(MiddleBlock, CheckBlock, Cmp, OldTerm);
+ OldTerm->eraseFromParent();
+
+ Cmp = StrideCheck;
+ LastBypassBlock = CheckBlock;
+ }
+
// Generate the code that checks in runtime if arrays overlap. We put the
// checks into a separate block to make the more common case of few elements
// faster.
- Instruction *MemRuntimeCheck = addRuntimeCheck(Legal,
- BypassBlock->getTerminator());
+ Instruction *MemRuntimeCheck;
+ tie(FirstCheckInst, MemRuntimeCheck) =
+ addRuntimeCheck(LastBypassBlock->getTerminator());
if (MemRuntimeCheck) {
// Create a new block containing the memory check.
- BasicBlock *CheckBlock = BypassBlock->splitBasicBlock(MemRuntimeCheck,
- "vector.memcheck");
+ BasicBlock *CheckBlock =
+ LastBypassBlock->splitBasicBlock(MemRuntimeCheck, "vector.memcheck");
if (ParentLoop)
ParentLoop->addBasicBlockToLoop(CheckBlock, LI->getBase());
LoopBypassBlocks.push_back(CheckBlock);
// Replace the branch into the memory check block with a conditional branch
// for the "few elements case".
- Instruction *OldTerm = BypassBlock->getTerminator();
+ Instruction *OldTerm = LastBypassBlock->getTerminator();
BranchInst::Create(MiddleBlock, CheckBlock, Cmp, OldTerm);
OldTerm->eraseFromParent();
}
}
-void
-InnerLoopVectorizer::vectorizeLoop(LoopVectorizationLegality *Legal) {
+void InnerLoopVectorizer::vectorizeLoop() {
//===------------------------------------------------===//
//
// Notice: any optimization or new instruction that go
// Vectorize all of the blocks in the original loop.
for (LoopBlocksDFS::RPOIterator bb = DFS.beginRPO(),
be = DFS.endRPO(); bb != be; ++bb)
- vectorizeBlockInLoop(Legal, *bb, &RdxPHIsToFix);
+ vectorizeBlockInLoop(*bb, &RdxPHIsToFix);
// At this point every instruction in the original loop is widened to
// a vector form. We are almost done. Now, we need to fix the PHI nodes
void InnerLoopVectorizer::widenPHIInstruction(Instruction *PN,
InnerLoopVectorizer::VectorParts &Entry,
- LoopVectorizationLegality *Legal,
unsigned UF, unsigned VF, PhiVector *PV) {
PHINode* P = cast<PHINode>(PN);
// Handle reduction variables:
setDebugLocFromInst(Builder, P);
// Check for PHI nodes that are lowered to vector selects.
if (P->getParent() != OrigLoop->getHeader()) {
- // We know that all PHIs in non header blocks are converted into
+ // We know that all PHIs in non-header blocks are converted into
// selects, so we don't have to worry about the insertion order and we
// can just use the builder.
// At this point we generate the predication tree. There may be
}
}
-void
-InnerLoopVectorizer::vectorizeBlockInLoop(LoopVectorizationLegality *Legal,
- BasicBlock *BB, PhiVector *PV) {
+void InnerLoopVectorizer::vectorizeBlockInLoop(BasicBlock *BB, PhiVector *PV) {
// For each instruction in the old loop.
for (BasicBlock::iterator it = BB->begin(), e = BB->end(); it != e; ++it) {
VectorParts &Entry = WidenMap.get(it);
continue;
case Instruction::PHI:{
// Vectorize PHINodes.
- widenPHIInstruction(it, Entry, Legal, UF, VF, PV);
+ widenPHIInstruction(it, Entry, UF, VF, PV);
continue;
}// End of PHI.
case Instruction::Store:
case Instruction::Load:
- vectorizeMemoryInstruction(it, Legal);
+ vectorizeMemoryInstruction(it);
break;
case Instruction::ZExt:
case Instruction::SExt:
DT->changeImmediateDominator(LoopScalarBody, LoopScalarPreHeader);
DT->changeImmediateDominator(LoopExitBlock, LoopMiddleBlock);
- DEBUG(DT->verifyAnalysis());
+ DEBUG(DT->verifyDomTree());
+}
+
+/// \brief Check whether it is safe to if-convert this phi node.
+///
+/// Phi nodes with constant expressions that can trap are not safe to if
+/// convert.
+static bool canIfConvertPHINodes(BasicBlock *BB) {
+ for (BasicBlock::iterator I = BB->begin(), E = BB->end(); I != E; ++I) {
+ PHINode *Phi = dyn_cast<PHINode>(I);
+ if (!Phi)
+ return true;
+ for (unsigned p = 0, e = Phi->getNumIncomingValues(); p != e; ++p)
+ if (Constant *C = dyn_cast<Constant>(Phi->getIncomingValue(p)))
+ if (C->canTrap())
+ return false;
+ }
+ return true;
}
bool LoopVectorizationLegality::canVectorizeWithIfConvert() {
}
// Collect the blocks that need predication.
+ BasicBlock *Header = TheLoop->getHeader();
for (Loop::block_iterator BI = TheLoop->block_begin(),
BE = TheLoop->block_end(); BI != BE; ++BI) {
BasicBlock *BB = *BI;
return false;
// We must be able to predicate all blocks that need to be predicated.
- if (blockNeedsPredication(BB) && !blockCanBePredicated(BB, SafePointes))
+ if (blockNeedsPredication(BB)) {
+ if (!blockCanBePredicated(BB, SafePointes))
+ return false;
+ } else if (BB != Header && !canIfConvertPHINodes(BB))
return false;
+
}
// We can if-convert this loop.
DEBUG(dbgs() << "LV: Found a loop: " <<
TheLoop->getHeader()->getName() << '\n');
- // Check if we can if-convert non single-bb loops.
+ // Check if we can if-convert non-single-bb loops.
unsigned NumBlocks = TheLoop->getNumBlocks();
if (NumBlocks != 1 && !canVectorizeWithIfConvert()) {
DEBUG(dbgs() << "LV: Can't if-convert the loop.\n");
if (Ty->isPointerTy())
return DL.getIntPtrType(Ty);
+ // It is possible that char's or short's overflow when we ask for the loop's
+ // trip count, work around this by changing the type size.
+ if (Ty->getScalarSizeInBits() < 32)
+ return Type::getInt32Ty(Ty->getContext());
+
return Ty;
}
Type *T = ST->getValueOperand()->getType();
if (!VectorType::isValidElementType(T))
return false;
+ if (EnableMemAccessVersioning)
+ collectStridedAcccess(ST);
}
+ if (EnableMemAccessVersioning)
+ if (LoadInst *LI = dyn_cast<LoadInst>(it))
+ collectStridedAcccess(LI);
+
// Reduction instructions are allowed to have exit users.
// All other instructions must not have external users.
if (hasOutsideLoopUser(TheLoop, it, AllowedExit))
return true;
}
+///\brief Remove GEPs whose indices but the last one are loop invariant and
+/// return the induction operand of the gep pointer.
+static Value *stripGetElementPtr(Value *Ptr, ScalarEvolution *SE,
+ DataLayout *DL, Loop *Lp) {
+ GetElementPtrInst *GEP = dyn_cast<GetElementPtrInst>(Ptr);
+ if (!GEP)
+ return Ptr;
+
+ unsigned InductionOperand = getGEPInductionOperand(DL, GEP);
+
+ // Check that all of the gep indices are uniform except for our induction
+ // operand.
+ for (unsigned i = 0, e = GEP->getNumOperands(); i != e; ++i)
+ if (i != InductionOperand &&
+ !SE->isLoopInvariant(SE->getSCEV(GEP->getOperand(i)), Lp))
+ return Ptr;
+ return GEP->getOperand(InductionOperand);
+}
+
+///\brief Look for a cast use of the passed value.
+static Value *getUniqueCastUse(Value *Ptr, Loop *Lp, Type *Ty) {
+ Value *UniqueCast = 0;
+ for (Value::use_iterator UI = Ptr->use_begin(), UE = Ptr->use_end(); UI != UE;
+ ++UI) {
+ CastInst *CI = dyn_cast<CastInst>(*UI);
+ if (CI && CI->getType() == Ty) {
+ if (!UniqueCast)
+ UniqueCast = CI;
+ else
+ return 0;
+ }
+ }
+ return UniqueCast;
+}
+
+///\brief Get the stride of a pointer access in a loop.
+/// Looks for symbolic strides "a[i*stride]". Returns the symbolic stride as a
+/// pointer to the Value, or null otherwise.
+static Value *getStrideFromPointer(Value *Ptr, ScalarEvolution *SE,
+ DataLayout *DL, Loop *Lp) {
+ const PointerType *PtrTy = dyn_cast<PointerType>(Ptr->getType());
+ if (!PtrTy || PtrTy->isAggregateType())
+ return 0;
+
+ // Try to remove a gep instruction to make the pointer (actually index at this
+ // point) easier analyzable. If OrigPtr is equal to Ptr we are analzying the
+ // pointer, otherwise, we are analyzing the index.
+ Value *OrigPtr = Ptr;
+
+ // The size of the pointer access.
+ int64_t PtrAccessSize = 1;
+
+ Ptr = stripGetElementPtr(Ptr, SE, DL, Lp);
+ const SCEV *V = SE->getSCEV(Ptr);
+
+ if (Ptr != OrigPtr)
+ // Strip off casts.
+ while (const SCEVCastExpr *C = dyn_cast<SCEVCastExpr>(V))
+ V = C->getOperand();
+
+ const SCEVAddRecExpr *S = dyn_cast<SCEVAddRecExpr>(V);
+ if (!S)
+ return 0;
+
+ V = S->getStepRecurrence(*SE);
+ if (!V)
+ return 0;
+
+ // Strip off the size of access multiplication if we are still analyzing the
+ // pointer.
+ if (OrigPtr == Ptr) {
+ DL->getTypeAllocSize(PtrTy->getElementType());
+ if (const SCEVMulExpr *M = dyn_cast<SCEVMulExpr>(V)) {
+ if (M->getOperand(0)->getSCEVType() != scConstant)
+ return 0;
+
+ const APInt &APStepVal =
+ cast<SCEVConstant>(M->getOperand(0))->getValue()->getValue();
+
+ // Huge step value - give up.
+ if (APStepVal.getBitWidth() > 64)
+ return 0;
+
+ int64_t StepVal = APStepVal.getSExtValue();
+ if (PtrAccessSize != StepVal)
+ return 0;
+ V = M->getOperand(1);
+ }
+ }
+
+ // Strip off casts.
+ Type *StripedOffRecurrenceCast = 0;
+ if (const SCEVCastExpr *C = dyn_cast<SCEVCastExpr>(V)) {
+ StripedOffRecurrenceCast = C->getType();
+ V = C->getOperand();
+ }
+
+ // Look for the loop invariant symbolic value.
+ const SCEVUnknown *U = dyn_cast<SCEVUnknown>(V);
+ if (!U)
+ return 0;
+
+ Value *Stride = U->getValue();
+ if (!Lp->isLoopInvariant(Stride))
+ return 0;
+
+ // If we have stripped off the recurrence cast we have to make sure that we
+ // return the value that is used in this loop so that we can replace it later.
+ if (StripedOffRecurrenceCast)
+ Stride = getUniqueCastUse(Stride, Lp, StripedOffRecurrenceCast);
+
+ return Stride;
+}
+
+void LoopVectorizationLegality::collectStridedAcccess(Value *MemAccess) {
+ Value *Ptr = 0;
+ if (LoadInst *LI = dyn_cast<LoadInst>(MemAccess))
+ Ptr = LI->getPointerOperand();
+ else if (StoreInst *SI = dyn_cast<StoreInst>(MemAccess))
+ Ptr = SI->getPointerOperand();
+ else
+ return;
+
+ Value *Stride = getStrideFromPointer(Ptr, SE, DL, TheLoop);
+ if (!Stride)
+ return;
+
+ DEBUG(dbgs() << "LV: Found a strided access that we can version");
+ DEBUG(dbgs() << " Ptr: " << *Ptr << " Stride: " << *Stride << "\n");
+ Strides[Ptr] = Stride;
+ StrideSet.insert(Stride);
+}
+
void LoopVectorizationLegality::collectLoopUniforms() {
// We now know that the loop is vectorizable!
// Collect variables that will remain uniform after vectorization.
/// non-intersection.
bool canCheckPtrAtRT(LoopVectorizationLegality::RuntimePointerCheck &RtCheck,
unsigned &NumComparisons, ScalarEvolution *SE,
- Loop *TheLoop, bool ShouldCheckStride = false);
+ Loop *TheLoop, ValueToValueMap &Strides,
+ bool ShouldCheckStride = false);
/// \brief Goes over all memory accesses, checks whether a RT check is needed
/// and builds sets of dependent accesses.
} // end anonymous namespace
/// \brief Check whether a pointer can participate in a runtime bounds check.
-static bool hasComputableBounds(ScalarEvolution *SE, Value *Ptr) {
- const SCEV *PtrScev = SE->getSCEV(Ptr);
+static bool hasComputableBounds(ScalarEvolution *SE, ValueToValueMap &Strides,
+ Value *Ptr) {
+ const SCEV *PtrScev = replaceSymbolicStrideSCEV(SE, Strides, Ptr);
const SCEVAddRecExpr *AR = dyn_cast<SCEVAddRecExpr>(PtrScev);
if (!AR)
return false;
/// \brief Check the stride of the pointer and ensure that it does not wrap in
/// the address space.
static int isStridedPtr(ScalarEvolution *SE, DataLayout *DL, Value *Ptr,
- const Loop *Lp);
+ const Loop *Lp, ValueToValueMap &StridesMap);
bool AccessAnalysis::canCheckPtrAtRT(
- LoopVectorizationLegality::RuntimePointerCheck &RtCheck,
- unsigned &NumComparisons, ScalarEvolution *SE,
- Loop *TheLoop, bool ShouldCheckStride) {
+ LoopVectorizationLegality::RuntimePointerCheck &RtCheck,
+ unsigned &NumComparisons, ScalarEvolution *SE, Loop *TheLoop,
+ ValueToValueMap &StridesMap, bool ShouldCheckStride) {
// Find pointers with computable bounds. We are going to use this information
// to place a runtime bound check.
unsigned NumReadPtrChecks = 0;
else
++NumReadPtrChecks;
- if (hasComputableBounds(SE, Ptr) &&
+ if (hasComputableBounds(SE, StridesMap, Ptr) &&
// When we run after a failing dependency check we have to make sure we
// don't have wrapping pointers.
- (!ShouldCheckStride || isStridedPtr(SE, DL, Ptr, TheLoop) == 1)) {
+ (!ShouldCheckStride ||
+ isStridedPtr(SE, DL, Ptr, TheLoop, StridesMap) == 1)) {
// The id of the dependence set.
unsigned DepId;
// Each access has its own dependence set.
DepId = RunningDepId++;
- RtCheck.insert(SE, TheLoop, Ptr, IsWrite, DepId);
+ RtCheck.insert(SE, TheLoop, Ptr, IsWrite, DepId, StridesMap);
DEBUG(dbgs() << "LV: Found a runtime check ptr:" << *Ptr << '\n');
} else {
///
/// Only checks sets with elements in \p CheckDeps.
bool areDepsSafe(AccessAnalysis::DepCandidates &AccessSets,
- MemAccessInfoSet &CheckDeps);
+ MemAccessInfoSet &CheckDeps, ValueToValueMap &Strides);
/// \brief The maximum number of bytes of a vector register we can vectorize
/// the accesses safely with.
// We can access this many bytes in parallel safely.
unsigned MaxSafeDepDistBytes;
- /// \brief If we see a non constant dependence distance we can still try to
+ /// \brief If we see a non-constant dependence distance we can still try to
/// vectorize this loop with runtime checks.
bool ShouldRetryWithRuntimeCheck;
/// distance is smaller than any other distance encountered so far).
/// Otherwise, this function returns true signaling a possible dependence.
bool isDependent(const MemAccessInfo &A, unsigned AIdx,
- const MemAccessInfo &B, unsigned BIdx);
+ const MemAccessInfo &B, unsigned BIdx,
+ ValueToValueMap &Strides);
/// \brief Check whether the data dependence could prevent store-load
/// forwarding.
/// \brief Check whether the access through \p Ptr has a constant stride.
static int isStridedPtr(ScalarEvolution *SE, DataLayout *DL, Value *Ptr,
- const Loop *Lp) {
+ const Loop *Lp, ValueToValueMap &StridesMap) {
const Type *Ty = Ptr->getType();
- assert(Ty->isPointerTy() && "Unexpected non ptr");
+ assert(Ty->isPointerTy() && "Unexpected non-ptr");
// Make sure that the pointer does not point to aggregate types.
const PointerType *PtrTy = cast<PointerType>(Ty);
return 0;
}
- const SCEV *PtrScev = SE->getSCEV(Ptr);
+ const SCEV *PtrScev = replaceSymbolicStrideSCEV(SE, StridesMap, Ptr);
+
const SCEVAddRecExpr *AR = dyn_cast<SCEVAddRecExpr>(PtrScev);
if (!AR) {
DEBUG(dbgs() << "LV: Bad stride - Not an AddRecExpr pointer "
}
bool MemoryDepChecker::isDependent(const MemAccessInfo &A, unsigned AIdx,
- const MemAccessInfo &B, unsigned BIdx) {
+ const MemAccessInfo &B, unsigned BIdx,
+ ValueToValueMap &Strides) {
assert (AIdx < BIdx && "Must pass arguments in program order");
Value *APtr = A.getPointer();
if (!AIsWrite && !BIsWrite)
return false;
- const SCEV *AScev = SE->getSCEV(APtr);
- const SCEV *BScev = SE->getSCEV(BPtr);
+ const SCEV *AScev = replaceSymbolicStrideSCEV(SE, Strides, APtr);
+ const SCEV *BScev = replaceSymbolicStrideSCEV(SE, Strides, BPtr);
- int StrideAPtr = isStridedPtr(SE, DL, APtr, InnermostLoop);
- int StrideBPtr = isStridedPtr(SE, DL, BPtr, InnermostLoop);
+ int StrideAPtr = isStridedPtr(SE, DL, APtr, InnermostLoop, Strides);
+ int StrideBPtr = isStridedPtr(SE, DL, BPtr, InnermostLoop, Strides);
const SCEV *Src = AScev;
const SCEV *Sink = BScev;
const SCEVConstant *C = dyn_cast<SCEVConstant>(Dist);
if (!C) {
- DEBUG(dbgs() << "LV: Dependence because of non constant distance\n");
+ DEBUG(dbgs() << "LV: Dependence because of non-constant distance\n");
ShouldRetryWithRuntimeCheck = true;
return true;
}
return false;
}
-bool
-MemoryDepChecker::areDepsSafe(AccessAnalysis::DepCandidates &AccessSets,
- MemAccessInfoSet &CheckDeps) {
+bool MemoryDepChecker::areDepsSafe(AccessAnalysis::DepCandidates &AccessSets,
+ MemAccessInfoSet &CheckDeps,
+ ValueToValueMap &Strides) {
MaxSafeDepDistBytes = -1U;
while (!CheckDeps.empty()) {
I1E = Accesses[*AI].end(); I1 != I1E; ++I1)
for (std::vector<unsigned>::iterator I2 = Accesses[*OI].begin(),
I2E = Accesses[*OI].end(); I2 != I2E; ++I2) {
- if (*I1 < *I2 && isDependent(*AI, *I1, *OI, *I2))
+ if (*I1 < *I2 && isDependent(*AI, *I1, *OI, *I2, Strides))
return false;
- if (*I2 < *I1 && isDependent(*OI, *I2, *AI, *I1))
+ if (*I2 < *I1 && isDependent(*OI, *I2, *AI, *I1, Strides))
return false;
}
++OI;
// read a few words, modify, and write a few words, and some of the
// words may be written to the same address.
bool IsReadOnlyPtr = false;
- if (Seen.insert(Ptr) || !isStridedPtr(SE, DL, Ptr, TheLoop)) {
+ if (Seen.insert(Ptr) || !isStridedPtr(SE, DL, Ptr, TheLoop, Strides)) {
++NumReads;
IsReadOnlyPtr = true;
}
unsigned NumComparisons = 0;
bool CanDoRT = false;
if (NeedRTCheck)
- CanDoRT = Accesses.canCheckPtrAtRT(PtrRtCheck, NumComparisons, SE, TheLoop);
-
+ CanDoRT = Accesses.canCheckPtrAtRT(PtrRtCheck, NumComparisons, SE, TheLoop,
+ Strides);
DEBUG(dbgs() << "LV: We need to do " << NumComparisons <<
" pointer comparisons.\n");
bool CanVecMem = true;
if (Accesses.isDependencyCheckNeeded()) {
DEBUG(dbgs() << "LV: Checking memory dependencies\n");
- CanVecMem = DepChecker.areDepsSafe(DependentAccesses,
- Accesses.getDependenciesToCheck());
+ CanVecMem = DepChecker.areDepsSafe(
+ DependentAccesses, Accesses.getDependenciesToCheck(), Strides);
MaxSafeDepDistBytes = DepChecker.getMaxSafeDepDistBytes();
if (!CanVecMem && DepChecker.shouldRetryWithRuntimeCheck()) {
PtrRtCheck.Need = true;
CanDoRT = Accesses.canCheckPtrAtRT(PtrRtCheck, NumComparisons, SE,
- TheLoop, true);
+ TheLoop, Strides, true);
// Check that we did not collect too many pointers or found an unsizeable
// pointer.
if (!CanDoRT || NumComparisons > RuntimeMemoryCheckThreshold) {
// Check whether we found a reduction operator.
FoundReduxOp |= !IsAPhi;
- // Process users of current instruction. Push non PHI nodes after PHI nodes
+ // Process users of current instruction. Push non-PHI nodes after PHI nodes
// onto the stack. This way we are going to have seen all inputs to PHI
// nodes once we get to them.
SmallVector<Instruction *, 8> NonPHIs;
if (it->mayWriteToMemory() || it->mayThrow())
return false;
+ // Check that we don't have a constant expression that can trap as operand.
+ for (Instruction::op_iterator OI = it->op_begin(), OE = it->op_end();
+ OI != OE; ++OI) {
+ if (Constant *C = dyn_cast<Constant>(*OI))
+ if (C->canTrap())
+ return false;
+ }
+
// The instructions below can trap.
switch (it->getOpcode()) {
default: continue;
return StepVal > MaxMergeDistance;
}
+static bool isStrideMul(Instruction *I, LoopVectorizationLegality *Legal) {
+ if (Legal->hasStride(I->getOperand(0)) || Legal->hasStride(I->getOperand(1)))
+ return true;
+ return false;
+}
+
unsigned
LoopVectorizationCostModel::getInstructionCost(Instruction *I, unsigned VF) {
// If we know that this instruction will remain uniform, check the cost of
case Instruction::And:
case Instruction::Or:
case Instruction::Xor: {
+ // Since we will replace the stride by 1 the multiplication should go away.
+ if (I->getOpcode() == Instruction::Mul && isStrideMul(I, Legal))
+ return 0;
// Certain instructions can be cheaper to vectorize if they have a constant
// second vector operand. One example of this are shifts on x86.
TargetTransformInfo::OperandValueKind Op1VK =
static const char lv_name[] = "Loop Vectorization";
INITIALIZE_PASS_BEGIN(LoopVectorize, LV_NAME, lv_name, false, false)
INITIALIZE_AG_DEPENDENCY(TargetTransformInfo)
-INITIALIZE_PASS_DEPENDENCY(DominatorTree)
+INITIALIZE_PASS_DEPENDENCY(DominatorTreeWrapperPass)
INITIALIZE_PASS_DEPENDENCY(ScalarEvolution)
INITIALIZE_PASS_DEPENDENCY(LCSSA)
INITIALIZE_PASS_DEPENDENCY(LoopInfo)
INITIALIZE_PASS_END(LoopVectorize, LV_NAME, lv_name, false, false)
namespace llvm {
- Pass *createLoopVectorizePass(bool NoUnrolling) {
- return new LoopVectorize(NoUnrolling);
+ Pass *createLoopVectorizePass(bool NoUnrolling, bool AlwaysVectorize) {
+ return new LoopVectorize(NoUnrolling, AlwaysVectorize);
}
}
}
}
-void
-InnerLoopUnroller::vectorizeMemoryInstruction(Instruction *Instr,
- LoopVectorizationLegality*) {
+void InnerLoopUnroller::vectorizeMemoryInstruction(Instruction *Instr) {
return scalarizeInstruction(Instr);
}
projects / oota-llvm.git / blobdiff