#include "llvm/ADT/SetVector.h"
#include "llvm/ADT/Statistic.h"
#include "llvm/Analysis/AliasAnalysis.h"
+#include "llvm/Analysis/AssumptionCache.h"
+#include "llvm/Analysis/CodeMetrics.h"
#include "llvm/Analysis/LoopInfo.h"
#include "llvm/Analysis/ScalarEvolution.h"
#include "llvm/Analysis/ScalarEvolutionExpressions.h"
return Opcode;
}
+/// Get the intersection (logical and) of all of the potential IR flags
+/// of each scalar operation (VL) that will be converted into a vector (I).
+/// Flag set: NSW, NUW, exact, and all of fast-math.
+static void propagateIRFlags(Value *I, ArrayRef<Value *> VL) {
+ if (auto *VecOp = dyn_cast<BinaryOperator>(I)) {
+ if (auto *Intersection = dyn_cast<BinaryOperator>(VL[0])) {
+ // Intersection is initialized to the 0th scalar,
+ // so start counting from index '1'.
+ for (int i = 1, e = VL.size(); i < e; ++i) {
+ if (auto *Scalar = dyn_cast<BinaryOperator>(VL[i]))
+ Intersection->andIRFlags(Scalar);
+ }
+ VecOp->copyIRFlags(Intersection);
+ }
+ }
+}
+
/// \returns \p I after propagating metadata from \p VL.
static Instruction *propagateMetadata(Instruction *I, ArrayRef<Value *> VL) {
Instruction *I0 = cast<Instruction>(VL[0]);
}
}
+/// \returns True if in-tree use also needs extract. This refers to
+/// possible scalar operand in vectorized instruction.
+static bool InTreeUserNeedToExtract(Value *Scalar, Instruction *UserInst,
+ TargetLibraryInfo *TLI) {
+
+ unsigned Opcode = UserInst->getOpcode();
+ switch (Opcode) {
+ case Instruction::Load: {
+ LoadInst *LI = cast<LoadInst>(UserInst);
+ return (LI->getPointerOperand() == Scalar);
+ }
+ case Instruction::Store: {
+ StoreInst *SI = cast<StoreInst>(UserInst);
+ return (SI->getPointerOperand() == Scalar);
+ }
+ case Instruction::Call: {
+ CallInst *CI = cast<CallInst>(UserInst);
+ Intrinsic::ID ID = getIntrinsicIDForCall(CI, TLI);
+ if (hasVectorInstrinsicScalarOpd(ID, 1)) {
+ return (CI->getArgOperand(1) == Scalar);
+ }
+ }
+ default:
+ return false;
+ }
+}
+
/// Bottom Up SLP Vectorizer.
class BoUpSLP {
public:
BoUpSLP(Function *Func, ScalarEvolution *Se, const DataLayout *Dl,
TargetTransformInfo *Tti, TargetLibraryInfo *TLi, AliasAnalysis *Aa,
- LoopInfo *Li, DominatorTree *Dt)
- : NumLoadsWantToKeepOrder(0), NumLoadsWantToChangeOrder(0),
- F(Func), SE(Se), DL(Dl), TTI(Tti), TLI(TLi), AA(Aa), LI(Li), DT(Dt),
- Builder(Se->getContext()) {}
+ LoopInfo *Li, DominatorTree *Dt, AssumptionCache *AC)
+ : NumLoadsWantToKeepOrder(0), NumLoadsWantToChangeOrder(0), F(Func),
+ SE(Se), DL(Dl), TTI(Tti), TLI(TLi), AA(Aa), LI(Li), DT(Dt),
+ Builder(Se->getContext()) {
+ CodeMetrics::collectEphemeralValues(F, AC, EphValues);
+ }
/// \brief Vectorize the tree that starts with the elements in \p VL.
/// Returns the vectorized root.
/// This list holds pairs of (Internal Scalar : External User).
UserList ExternalUses;
+ /// Values used only by @llvm.assume calls.
+ SmallPtrSet<const Value *, 32> EphValues;
+
/// Holds all of the instructions that we gathered.
SetVector<Instruction *> GatherSeq;
/// A list of blocks that we are going to CSE.
for (User *U : Scalar->users()) {
DEBUG(dbgs() << "SLP: Checking user:" << *U << ".\n");
- // Skip in-tree scalars that become vectors.
- if (ScalarToTreeEntry.count(U)) {
- DEBUG(dbgs() << "SLP: \tInternal user will be removed:" <<
- *U << ".\n");
- int Idx = ScalarToTreeEntry[U]; (void) Idx;
- assert(!VectorizableTree[Idx].NeedToGather && "Bad state");
- continue;
- }
Instruction *UserInst = dyn_cast<Instruction>(U);
if (!UserInst)
continue;
+ // Skip in-tree scalars that become vectors
+ if (ScalarToTreeEntry.count(U)) {
+ int Idx = ScalarToTreeEntry[U];
+ TreeEntry *UseEntry = &VectorizableTree[Idx];
+ Value *UseScalar = UseEntry->Scalars[0];
+ // Some in-tree scalars will remain as scalar in vectorized
+ // instructions. If that is the case, the one in Lane 0 will
+ // be used.
+ if (UseScalar != U ||
+ !InTreeUserNeedToExtract(Scalar, UserInst, TLI)) {
+ DEBUG(dbgs() << "SLP: \tInternal user will be removed:" << *U
+ << ".\n");
+ assert(!VectorizableTree[Idx].NeedToGather && "Bad state");
+ continue;
+ }
+ }
+
// Ignore users in the user ignore list.
if (std::find(UserIgnoreList.begin(), UserIgnoreList.end(), UserInst) !=
UserIgnoreList.end())
// We now know that this is a vector of instructions of the same type from
// the same block.
+ // Don't vectorize ephemeral values.
+ for (unsigned i = 0, e = VL.size(); i != e; ++i) {
+ if (EphValues.count(VL[i])) {
+ DEBUG(dbgs() << "SLP: The instruction (" << *VL[i] <<
+ ") is ephemeral.\n");
+ newTreeEntry(VL, false);
+ return;
+ }
+ }
+
// Check if this is a duplicate of another entry.
if (ScalarToTreeEntry.count(VL[0])) {
int Idx = ScalarToTreeEntry[VL[0]];
}
}
- // If any of the scalars appears in the table OR it is marked as a value that
- // needs to stat scalar then we need to gather the scalars.
+ // If any of the scalars is marked as a value that needs to stay scalar then
+ // we need to gather the scalars.
for (unsigned i = 0, e = VL.size(); i != e; ++i) {
- if (ScalarToTreeEntry.count(VL[i]) || MustGather.count(VL[i])) {
- DEBUG(dbgs() << "SLP: Gathering due to gathered scalar. \n");
+ if (MustGather.count(VL[i])) {
+ DEBUG(dbgs() << "SLP: Gathering due to gathered scalar.\n");
newTreeEntry(VL, false);
return;
}
TargetTransformInfo::OK_AnyValue;
TargetTransformInfo::OperandValueKind Op2VK =
TargetTransformInfo::OK_UniformConstantValue;
+ TargetTransformInfo::OperandValueProperties Op1VP =
+ TargetTransformInfo::OP_None;
+ TargetTransformInfo::OperandValueProperties Op2VP =
+ TargetTransformInfo::OP_None;
// If all operands are exactly the same ConstantInt then set the
// operand kind to OK_UniformConstantValue.
CInt != cast<ConstantInt>(I->getOperand(1)))
Op2VK = TargetTransformInfo::OK_NonUniformConstantValue;
}
+ // FIXME: Currently cost of model modification for division by
+ // power of 2 is handled only for X86. Add support for other targets.
+ if (Op2VK == TargetTransformInfo::OK_UniformConstantValue && CInt &&
+ CInt->getValue().isPowerOf2())
+ Op2VP = TargetTransformInfo::OP_PowerOf2;
- ScalarCost =
- VecTy->getNumElements() *
- TTI->getArithmeticInstrCost(Opcode, ScalarTy, Op1VK, Op2VK);
- VecCost = TTI->getArithmeticInstrCost(Opcode, VecTy, Op1VK, Op2VK);
+ ScalarCost = VecTy->getNumElements() *
+ TTI->getArithmeticInstrCost(Opcode, ScalarTy, Op1VK, Op2VK,
+ Op1VP, Op2VP);
+ VecCost = TTI->getArithmeticInstrCost(Opcode, VecTy, Op1VK, Op2VK,
+ Op1VP, Op2VP);
}
return VecCost - ScalarCost;
}
for (UserList::iterator I = ExternalUses.begin(), E = ExternalUses.end();
I != E; ++I) {
// We only add extract cost once for the same scalar.
- if (!ExtractCostCalculated.insert(I->Scalar))
+ if (!ExtractCostCalculated.insert(I->Scalar).second)
+ continue;
+
+ // Uses by ephemeral values are free (because the ephemeral value will be
+ // removed prior to code generation, and so the extraction will be
+ // removed as well).
+ if (EphValues.count(I->User))
continue;
VectorType *VecTy = VectorType::get(I->Scalar->getType(), BundleWidth);
ValueList Operands;
BasicBlock *IBB = PH->getIncomingBlock(i);
- if (!VisitedBBs.insert(IBB)) {
+ if (!VisitedBBs.insert(IBB).second) {
NewPhi->addIncoming(NewPhi->getIncomingValueForBlock(IBB), IBB);
continue;
}
BinaryOperator *BinOp = cast<BinaryOperator>(VL0);
Value *V = Builder.CreateBinOp(BinOp->getOpcode(), LHS, RHS);
E->VectorizedValue = V;
+ propagateIRFlags(E->VectorizedValue, E->Scalars);
++NumVectorInstructions;
if (Instruction *I = dyn_cast<Instruction>(V))
Value *VecPtr = Builder.CreateBitCast(LI->getPointerOperand(),
VecTy->getPointerTo(AS));
+
+ // The pointer operand uses an in-tree scalar so we add the new BitCast to
+ // ExternalUses list to make sure that an extract will be generated in the
+ // future.
+ if (ScalarToTreeEntry.count(LI->getPointerOperand()))
+ ExternalUses.push_back(
+ ExternalUser(LI->getPointerOperand(), cast<User>(VecPtr), 0));
+
unsigned Alignment = LI->getAlignment();
LI = Builder.CreateLoad(VecPtr);
if (!Alignment)
Value *VecPtr = Builder.CreateBitCast(SI->getPointerOperand(),
VecTy->getPointerTo(AS));
StoreInst *S = Builder.CreateStore(VecValue, VecPtr);
+
+ // The pointer operand uses an in-tree scalar so we add the new BitCast to
+ // ExternalUses list to make sure that an extract will be generated in the
+ // future.
+ if (ScalarToTreeEntry.count(SI->getPointerOperand()))
+ ExternalUses.push_back(
+ ExternalUser(SI->getPointerOperand(), cast<User>(VecPtr), 0));
+
if (!Alignment)
Alignment = DL->getABITypeAlignment(SI->getValueOperand()->getType());
S->setAlignment(Alignment);
setInsertPointAfterBundle(E->Scalars);
Function *FI;
Intrinsic::ID IID = Intrinsic::not_intrinsic;
+ Value *ScalarArg = nullptr;
if (CI && (FI = CI->getCalledFunction())) {
IID = (Intrinsic::ID) FI->getIntrinsicID();
}
// a scalar. This argument should not be vectorized.
if (hasVectorInstrinsicScalarOpd(IID, 1) && j == 1) {
CallInst *CEI = cast<CallInst>(E->Scalars[0]);
+ ScalarArg = CEI->getArgOperand(j);
OpVecs.push_back(CEI->getArgOperand(j));
continue;
}
Type *Tys[] = { VectorType::get(CI->getType(), E->Scalars.size()) };
Function *CF = Intrinsic::getDeclaration(M, ID, Tys);
Value *V = Builder.CreateCall(CF, OpVecs);
+
+ // The scalar argument uses an in-tree scalar so we add the new vectorized
+ // call to ExternalUses list to make sure that an extract will be
+ // generated in the future.
+ if (ScalarArg && ScalarToTreeEntry.count(ScalarArg))
+ ExternalUses.push_back(ExternalUser(ScalarArg, cast<User>(V), 0));
+
E->VectorizedValue = V;
++NumVectorInstructions;
return V;
BinaryOperator *BinOp1 = cast<BinaryOperator>(VL1);
Value *V1 = Builder.CreateBinOp(BinOp1->getOpcode(), LHS, RHS);
- // Create appropriate shuffle to take alternative operations from
- // the vector.
- std::vector<Constant *> Mask(E->Scalars.size());
+ // Create shuffle to take alternate operations from the vector.
+ // Also, gather up odd and even scalar ops to propagate IR flags to
+ // each vector operation.
+ ValueList OddScalars, EvenScalars;
unsigned e = E->Scalars.size();
+ SmallVector<Constant *, 8> Mask(e);
for (unsigned i = 0; i < e; ++i) {
- if (i & 1)
+ if (i & 1) {
Mask[i] = Builder.getInt32(e + i);
- else
+ OddScalars.push_back(E->Scalars[i]);
+ } else {
Mask[i] = Builder.getInt32(i);
+ EvenScalars.push_back(E->Scalars[i]);
+ }
}
Value *ShuffleMask = ConstantVector::get(Mask);
+ propagateIRFlags(V0, EvenScalars);
+ propagateIRFlags(V1, OddScalars);
Value *V = Builder.CreateShuffleVector(V0, V1, ShuffleMask);
E->VectorizedValue = V;
AliasAnalysis *AA;
LoopInfo *LI;
DominatorTree *DT;
+ AssumptionCache *AC;
bool runOnFunction(Function &F) override {
if (skipOptnoneFunction(F))
AA = &getAnalysis<AliasAnalysis>();
LI = &getAnalysis<LoopInfo>();
DT = &getAnalysis<DominatorTreeWrapperPass>().getDomTree();
+ AC = &getAnalysis<AssumptionCacheTracker>().getAssumptionCache(F);
StoreRefs.clear();
bool Changed = false;
// Use the bottom up slp vectorizer to construct chains that start with
// store instructions.
- BoUpSLP R(&F, SE, DL, TTI, TLI, AA, LI, DT);
+ BoUpSLP R(&F, SE, DL, TTI, TLI, AA, LI, DT, AC);
// Scan the blocks in the function in post order.
for (po_iterator<BasicBlock*> it = po_begin(&F.getEntryBlock()),
void getAnalysisUsage(AnalysisUsage &AU) const override {
FunctionPass::getAnalysisUsage(AU);
+ AU.addRequired<AssumptionCacheTracker>();
AU.addRequired<ScalarEvolution>();
AU.addRequired<AliasAnalysis>();
AU.addRequired<TargetTransformInfo>();
BinaryOperator *B0 = dyn_cast<BinaryOperator>(B->getOperand(0));
BinaryOperator *B1 = dyn_cast<BinaryOperator>(B->getOperand(1));
if (tryToVectorizePair(A, B0, R)) {
- B->moveBefore(V);
return true;
}
if (tryToVectorizePair(A, B1, R)) {
- B->moveBefore(V);
return true;
}
}
BinaryOperator *A0 = dyn_cast<BinaryOperator>(A->getOperand(0));
BinaryOperator *A1 = dyn_cast<BinaryOperator>(A->getOperand(1));
if (tryToVectorizePair(A0, B, R)) {
- A->moveBefore(V);
return true;
}
if (tryToVectorizePair(A1, B, R)) {
- A->moveBefore(V);
return true;
}
}
unsigned i = 0;
for (; i < NumReducedVals - ReduxWidth + 1; i += ReduxWidth) {
- ArrayRef<Value *> ValsToReduce(&ReducedVals[i], ReduxWidth);
- V.buildTree(ValsToReduce, ReductionOps);
+ V.buildTree(makeArrayRef(&ReducedVals[i], ReduxWidth), ReductionOps);
// Estimate cost.
int Cost = V.getTreeCost() + getReductionCost(TTI, ReducedVals[i]);
/// \brief Emit a horizontal reduction of the vectorized value.
Value *emitReduction(Value *VectorizedValue, IRBuilder<> &Builder) {
assert(VectorizedValue && "Need to have a vectorized tree node");
- Instruction *ValToReduce = dyn_cast<Instruction>(VectorizedValue);
assert(isPowerOf2_32(ReduxWidth) &&
"We only handle power-of-two reductions for now");
- Value *TmpVec = ValToReduce;
+ Value *TmpVec = VectorizedValue;
for (unsigned i = ReduxWidth / 2; i != 0; i >>= 1) {
if (IsPairwiseReduction) {
Value *LeftMask =
// Try to vectorize them.
unsigned NumElts = (SameTypeIt - IncIt);
DEBUG(errs() << "SLP: Trying to vectorize starting at PHIs (" << NumElts << ")\n");
- if (NumElts > 1 &&
- tryToVectorizeList(ArrayRef<Value *>(IncIt, NumElts), R)) {
+ if (NumElts > 1 && tryToVectorizeList(makeArrayRef(IncIt, NumElts), R)) {
// Success start over because instructions might have been changed.
HaveVectorizedPhiNodes = true;
Changed = true;
for (BasicBlock::iterator it = BB->begin(), e = BB->end(); it != e; it++) {
// We may go through BB multiple times so skip the one we have checked.
- if (!VisitedInstrs.insert(it))
+ if (!VisitedInstrs.insert(it).second)
continue;
if (isa<DbgInfoIntrinsic>(it))
}
}
+ // Try to vectorize horizontal reductions feeding into a return.
+ if (ReturnInst *RI = dyn_cast<ReturnInst>(it))
+ if (RI->getNumOperands() != 0)
+ if (BinaryOperator *BinOp =
+ dyn_cast<BinaryOperator>(RI->getOperand(0))) {
+ DEBUG(dbgs() << "SLP: Found a return to vectorize.\n");
+ if (tryToVectorizePair(BinOp->getOperand(0),
+ BinOp->getOperand(1), R)) {
+ Changed = true;
+ it = BB->begin();
+ e = BB->end();
+ continue;
+ }
+ }
+
// Try to vectorize trees that start at compare instructions.
if (CmpInst *CI = dyn_cast<CmpInst>(it)) {
if (tryToVectorizePair(CI->getOperand(0), CI->getOperand(1), R)) {
// Process the stores in chunks of 16.
for (unsigned CI = 0, CE = it->second.size(); CI < CE; CI+=16) {
unsigned Len = std::min<unsigned>(CE - CI, 16);
- ArrayRef<StoreInst *> Chunk(&it->second[CI], Len);
- Changed |= vectorizeStores(Chunk, -SLPCostThreshold, R);
+ Changed |= vectorizeStores(makeArrayRef(&it->second[CI], Len),
+ -SLPCostThreshold, R);
}
}
return Changed;
INITIALIZE_PASS_BEGIN(SLPVectorizer, SV_NAME, lv_name, false, false)
INITIALIZE_AG_DEPENDENCY(AliasAnalysis)
INITIALIZE_AG_DEPENDENCY(TargetTransformInfo)
+INITIALIZE_PASS_DEPENDENCY(AssumptionCacheTracker)
INITIALIZE_PASS_DEPENDENCY(ScalarEvolution)
INITIALIZE_PASS_DEPENDENCY(LoopSimplify)
INITIALIZE_PASS_END(SLPVectorizer, SV_NAME, lv_name, false, false)
projects / oota-llvm.git / blobdiff