forked from OSchip/llvm-project
[LoopVectorize] Don't vectorize loops when everything will be scalarized
This change prevents the loop vectorizer from vectorizing when all of the vector types it generates will be scalarized. I've run into this problem on the PPC's QPX vector ISA, which only holds floating-point vector types. The loop vectorizer will, however, happily vectorize loops with purely integer computation. Here's an example: LV: The Smallest and Widest types: 32 / 32 bits. LV: The Widest register is: 256 bits. LV: Found an estimated cost of 0 for VF 1 For instruction: %indvars.iv25 = phi i64 [ 0, %entry ], [ %indvars.iv.next26, %for.body ] LV: Found an estimated cost of 0 for VF 1 For instruction: %arrayidx = getelementptr inbounds [1600 x i32], [1600 x i32]* %a, i64 0, i64 %indvars.iv25 LV: Found an estimated cost of 0 for VF 1 For instruction: %2 = trunc i64 %indvars.iv25 to i32 LV: Found an estimated cost of 1 for VF 1 For instruction: store i32 %2, i32* %arrayidx, align 4 LV: Found an estimated cost of 1 for VF 1 For instruction: %indvars.iv.next26 = add nuw nsw i64 %indvars.iv25, 1 LV: Found an estimated cost of 1 for VF 1 For instruction: %exitcond27 = icmp eq i64 %indvars.iv.next26, 1600 LV: Found an estimated cost of 0 for VF 1 For instruction: br i1 %exitcond27, label %for.cond.cleanup, label %for.body LV: Scalar loop costs: 3. LV: Found an estimated cost of 0 for VF 2 For instruction: %indvars.iv25 = phi i64 [ 0, %entry ], [ %indvars.iv.next26, %for.body ] LV: Found an estimated cost of 0 for VF 2 For instruction: %arrayidx = getelementptr inbounds [1600 x i32], [1600 x i32]* %a, i64 0, i64 %indvars.iv25 LV: Found an estimated cost of 0 for VF 2 For instruction: %2 = trunc i64 %indvars.iv25 to i32 LV: Found an estimated cost of 2 for VF 2 For instruction: store i32 %2, i32* %arrayidx, align 4 LV: Found an estimated cost of 1 for VF 2 For instruction: %indvars.iv.next26 = add nuw nsw i64 %indvars.iv25, 1 LV: Found an estimated cost of 1 for VF 2 For instruction: %exitcond27 = icmp eq i64 %indvars.iv.next26, 1600 LV: Found an estimated cost of 0 for VF 2 For instruction: br i1 %exitcond27, label %for.cond.cleanup, label %for.body LV: Vector loop of width 2 costs: 2. LV: Found an estimated cost of 0 for VF 4 For instruction: %indvars.iv25 = phi i64 [ 0, %entry ], [ %indvars.iv.next26, %for.body ] LV: Found an estimated cost of 0 for VF 4 For instruction: %arrayidx = getelementptr inbounds [1600 x i32], [1600 x i32]* %a, i64 0, i64 %indvars.iv25 LV: Found an estimated cost of 0 for VF 4 For instruction: %2 = trunc i64 %indvars.iv25 to i32 LV: Found an estimated cost of 4 for VF 4 For instruction: store i32 %2, i32* %arrayidx, align 4 LV: Found an estimated cost of 1 for VF 4 For instruction: %indvars.iv.next26 = add nuw nsw i64 %indvars.iv25, 1 LV: Found an estimated cost of 1 for VF 4 For instruction: %exitcond27 = icmp eq i64 %indvars.iv.next26, 1600 LV: Found an estimated cost of 0 for VF 4 For instruction: br i1 %exitcond27, label %for.cond.cleanup, label %for.body LV: Vector loop of width 4 costs: 1. ... LV: Selecting VF: 8. LV: The target has 32 registers LV(REG): Calculating max register usage: LV(REG): At #0 Interval # 0 LV(REG): At #1 Interval # 1 LV(REG): At #2 Interval # 2 LV(REG): At #4 Interval # 1 LV(REG): At #5 Interval # 1 LV(REG): VF = 8 The problem is that the cost model here is not wrong, exactly. Since all of these operations are scalarized, their cost (aside from the uniform ones) are indeed VF*(scalar cost), just as the model suggests. In fact, the larger the VF picked, the lower the relative overhead from the loop itself (and the induction-variable update and check), and so in a sense, picking the largest VF here is the right thing to do. The problem is that vectorizing like this, where all of the vectors will be scalarized in the backend, isn't really vectorizing, but rather interleaving. By itself, this would be okay, but then the vectorizer itself also interleaves, and that's where the problem manifests itself. There's aren't actually enough scalar registers to support the normal interleave factor multiplied by a factor of VF (8 in this example). In other words, the problem with this is that our register-pressure heuristic does not account for scalarization. While we might want to improve our register-pressure heuristic, I don't think this is the right motivating case for that work. Here we have a more-basic problem: The job of the vectorizer is to vectorize things (interleaving aside), and if the IR it generates won't generate any actual vector code, then something is wrong. Thus, if every type looks like it will be scalarized (i.e. will be split into VF or more parts), then don't consider that VF. This is not a problem specific to PPC/QPX, however. The problem comes up under SSE on x86 too, and as such, this change fixes PR26837 too. I've added Sanjay's reduced test case from PR26837 to this commit. Differential Revision: http://reviews.llvm.org/D18537 llvm-svn: 264904
This commit is contained in:
Hal Finkel
parent
69b29b2b7c
commit
2e0ff2b244
|
|
@ -1532,15 +1532,26 @@ public:
|
|||
calculateRegisterUsage(const SmallVector<unsigned, 8> &VFs);
|
||||
|
||||
private:
|
||||
/// The vectorization cost is a combination of the cost itself and a boolean
|
||||
/// indicating whether any of the contributing operations will actually operate on
|
||||
/// vector values after type legalization in the backend. If this latter value is
|
||||
/// false, then all operations will be scalarized (i.e. no vectorization has
|
||||
/// actually taken place).
|
||||
typedef std::pair<unsigned, bool> VectorizationCostTy;
|
||||
|
||||
/// Returns the expected execution cost. The unit of the cost does
|
||||
/// not matter because we use the 'cost' units to compare different
|
||||
/// vector widths. The cost that is returned is *not* normalized by
|
||||
/// the factor width.
|
||||
unsigned expectedCost(unsigned VF);
|
||||
VectorizationCostTy expectedCost(unsigned VF);
|
||||
|
||||
/// Returns the execution time cost of an instruction for a given vector
|
||||
/// width. Vector width of one means scalar.
|
||||
unsigned getInstructionCost(Instruction *I, unsigned VF);
|
||||
VectorizationCostTy getInstructionCost(Instruction *I, unsigned VF);
|
||||
|
||||
/// The cost-computation logic from getInstructionCost which provides
|
||||
/// the vector type as an output parameter.
|
||||
unsigned getInstructionCost(Instruction *I, unsigned VF, Type *&VectorTy);
|
||||
|
||||
/// Returns whether the instruction is a load or store and will be a emitted
|
||||
/// as a vector operation.
|
||||
|
|
@ -5145,7 +5156,7 @@ LoopVectorizationCostModel::selectVectorizationFactor(bool OptForSize) {
|
|||
return Factor;
|
||||
}
|
||||
|
||||
float Cost = expectedCost(1);
|
||||
float Cost = expectedCost(1).first;
|
||||
#ifndef NDEBUG
|
||||
const float ScalarCost = Cost;
|
||||
#endif /* NDEBUG */
|
||||
|
|
@ -5156,16 +5167,22 @@ LoopVectorizationCostModel::selectVectorizationFactor(bool OptForSize) {
|
|||
// Ignore scalar width, because the user explicitly wants vectorization.
|
||||
if (ForceVectorization && VF > 1) {
|
||||
Width = 2;
|
||||
Cost = expectedCost(Width) / (float)Width;
|
||||
Cost = expectedCost(Width).first / (float)Width;
|
||||
}
|
||||
|
||||
for (unsigned i=2; i <= VF; i*=2) {
|
||||
// Notice that the vector loop needs to be executed less times, so
|
||||
// we need to divide the cost of the vector loops by the width of
|
||||
// the vector elements.
|
||||
float VectorCost = expectedCost(i) / (float)i;
|
||||
VectorizationCostTy C = expectedCost(i);
|
||||
float VectorCost = C.first / (float)i;
|
||||
DEBUG(dbgs() << "LV: Vector loop of width " << i << " costs: " <<
|
||||
(int)VectorCost << ".\n");
|
||||
if (!C.second && !ForceVectorization) {
|
||||
DEBUG(dbgs() << "LV: Not considering vector loop of width " << i <<
|
||||
" because it will not generate any vector instructions.\n");
|
||||
continue;
|
||||
}
|
||||
if (VectorCost < Cost) {
|
||||
Cost = VectorCost;
|
||||
Width = i;
|
||||
|
|
@ -5313,7 +5330,7 @@ unsigned LoopVectorizationCostModel::selectInterleaveCount(bool OptForSize,
|
|||
// If we did not calculate the cost for VF (because the user selected the VF)
|
||||
// then we calculate the cost of VF here.
|
||||
if (LoopCost == 0)
|
||||
LoopCost = expectedCost(VF);
|
||||
LoopCost = expectedCost(VF).first;
|
||||
|
||||
// Clamp the calculated IC to be between the 1 and the max interleave count
|
||||
// that the target allows.
|
||||
|
|
@ -5540,13 +5557,14 @@ LoopVectorizationCostModel::calculateRegisterUsage(
|
|||
return RUs;
|
||||
}
|
||||
|
||||
unsigned LoopVectorizationCostModel::expectedCost(unsigned VF) {
|
||||
unsigned Cost = 0;
|
||||
LoopVectorizationCostModel::VectorizationCostTy
|
||||
LoopVectorizationCostModel::expectedCost(unsigned VF) {
|
||||
VectorizationCostTy Cost;
|
||||
|
||||
// For each block.
|
||||
for (Loop::block_iterator bb = TheLoop->block_begin(),
|
||||
be = TheLoop->block_end(); bb != be; ++bb) {
|
||||
unsigned BlockCost = 0;
|
||||
VectorizationCostTy BlockCost;
|
||||
BasicBlock *BB = *bb;
|
||||
|
||||
// For each instruction in the old loop.
|
||||
|
|
@ -5559,24 +5577,26 @@ unsigned LoopVectorizationCostModel::expectedCost(unsigned VF) {
|
|||
if (ValuesToIgnore.count(&*it))
|
||||
continue;
|
||||
|
||||
unsigned C = getInstructionCost(&*it, VF);
|
||||
VectorizationCostTy C = getInstructionCost(&*it, VF);
|
||||
|
||||
// Check if we should override the cost.
|
||||
if (ForceTargetInstructionCost.getNumOccurrences() > 0)
|
||||
C = ForceTargetInstructionCost;
|
||||
C.first = ForceTargetInstructionCost;
|
||||
|
||||
BlockCost += C;
|
||||
DEBUG(dbgs() << "LV: Found an estimated cost of " << C << " for VF " <<
|
||||
VF << " For instruction: " << *it << '\n');
|
||||
BlockCost.first += C.first;
|
||||
BlockCost.second |= C.second;
|
||||
DEBUG(dbgs() << "LV: Found an estimated cost of " << C.first <<
|
||||
" for VF " << VF << " For instruction: " << *it << '\n');
|
||||
}
|
||||
|
||||
// We assume that if-converted blocks have a 50% chance of being executed.
|
||||
// When the code is scalar then some of the blocks are avoided due to CF.
|
||||
// When the code is vectorized we execute all code paths.
|
||||
if (VF == 1 && Legal->blockNeedsPredication(*bb))
|
||||
BlockCost /= 2;
|
||||
BlockCost.first /= 2;
|
||||
|
||||
Cost += BlockCost;
|
||||
Cost.first += BlockCost.first;
|
||||
Cost.second |= BlockCost.second;
|
||||
}
|
||||
|
||||
return Cost;
|
||||
|
|
@ -5653,17 +5673,28 @@ static bool isStrideMul(Instruction *I, LoopVectorizationLegality *Legal) {
|
|||
Legal->hasStride(I->getOperand(1));
|
||||
}
|
||||
|
||||
unsigned
|
||||
LoopVectorizationCostModel::VectorizationCostTy
|
||||
LoopVectorizationCostModel::getInstructionCost(Instruction *I, unsigned VF) {
|
||||
// If we know that this instruction will remain uniform, check the cost of
|
||||
// the scalar version.
|
||||
if (Legal->isUniformAfterVectorization(I))
|
||||
VF = 1;
|
||||
|
||||
Type *VectorTy;
|
||||
unsigned C = getInstructionCost(I, VF, VectorTy);
|
||||
|
||||
bool TypeNotScalarized = VF > 1 && !VectorTy->isVoidTy() &&
|
||||
TTI.getNumberOfParts(VectorTy) < VF;
|
||||
return VectorizationCostTy(C, TypeNotScalarized);
|
||||
}
|
||||
|
||||
unsigned
|
||||
LoopVectorizationCostModel::getInstructionCost(Instruction *I, unsigned VF,
|
||||
Type *&VectorTy) {
|
||||
Type *RetTy = I->getType();
|
||||
if (VF > 1 && MinBWs.count(I))
|
||||
RetTy = IntegerType::get(RetTy->getContext(), MinBWs[I]);
|
||||
Type *VectorTy = ToVectorTy(RetTy, VF);
|
||||
VectorTy = ToVectorTy(RetTy, VF);
|
||||
|
||||
// TODO: We need to estimate the cost of intrinsic calls.
|
||||
switch (I->getOpcode()) {
|
||||
|
|
|
|||
|
|
@ -0,0 +1,62 @@
|
|||
; RUN: opt -S -loop-vectorize < %s | FileCheck %s
|
||||
target datalayout = "E-m:e-i64:64-n32:64"
|
||||
target triple = "powerpc64-bgq-linux"
|
||||
|
||||
; Function Attrs: nounwind
|
||||
define zeroext i32 @test() #0 {
|
||||
; CHECK-LABEL: @test
|
||||
; CHECK-NOT: x i32>
|
||||
|
||||
entry:
|
||||
%a = alloca [1600 x i32], align 4
|
||||
%c = alloca [1600 x i32], align 4
|
||||
%0 = bitcast [1600 x i32]* %a to i8*
|
||||
call void @llvm.lifetime.start(i64 6400, i8* %0) #3
|
||||
br label %for.body
|
||||
|
||||
for.cond.cleanup: ; preds = %for.body
|
||||
%1 = bitcast [1600 x i32]* %c to i8*
|
||||
call void @llvm.lifetime.start(i64 6400, i8* %1) #3
|
||||
%arraydecay = getelementptr inbounds [1600 x i32], [1600 x i32]* %a, i64 0, i64 0
|
||||
%arraydecay1 = getelementptr inbounds [1600 x i32], [1600 x i32]* %c, i64 0, i64 0
|
||||
%call = call signext i32 @bar(i32* %arraydecay, i32* %arraydecay1) #3
|
||||
br label %for.body6
|
||||
|
||||
for.body: ; preds = %for.body, %entry
|
||||
%indvars.iv25 = phi i64 [ 0, %entry ], [ %indvars.iv.next26, %for.body ]
|
||||
%arrayidx = getelementptr inbounds [1600 x i32], [1600 x i32]* %a, i64 0, i64 %indvars.iv25
|
||||
%2 = trunc i64 %indvars.iv25 to i32
|
||||
store i32 %2, i32* %arrayidx, align 4
|
||||
%indvars.iv.next26 = add nuw nsw i64 %indvars.iv25, 1
|
||||
%exitcond27 = icmp eq i64 %indvars.iv.next26, 1600
|
||||
br i1 %exitcond27, label %for.cond.cleanup, label %for.body
|
||||
|
||||
for.cond.cleanup5: ; preds = %for.body6
|
||||
call void @llvm.lifetime.end(i64 6400, i8* nonnull %1) #3
|
||||
call void @llvm.lifetime.end(i64 6400, i8* %0) #3
|
||||
ret i32 %add
|
||||
|
||||
for.body6: ; preds = %for.body6, %for.cond.cleanup
|
||||
%indvars.iv = phi i64 [ 0, %for.cond.cleanup ], [ %indvars.iv.next, %for.body6 ]
|
||||
%s.022 = phi i32 [ 0, %for.cond.cleanup ], [ %add, %for.body6 ]
|
||||
%arrayidx8 = getelementptr inbounds [1600 x i32], [1600 x i32]* %c, i64 0, i64 %indvars.iv
|
||||
%3 = load i32, i32* %arrayidx8, align 4
|
||||
%add = add i32 %3, %s.022
|
||||
%indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
|
||||
%exitcond = icmp eq i64 %indvars.iv.next, 1600
|
||||
br i1 %exitcond, label %for.cond.cleanup5, label %for.body6
|
||||
}
|
||||
|
||||
; Function Attrs: argmemonly nounwind
|
||||
declare void @llvm.lifetime.start(i64, i8* nocapture) #1
|
||||
|
||||
; Function Attrs: argmemonly nounwind
|
||||
declare void @llvm.lifetime.end(i64, i8* nocapture) #1
|
||||
|
||||
declare signext i32 @bar(i32*, i32*) #2
|
||||
|
||||
attributes #0 = { nounwind "target-cpu"="a2q" "target-features"="+qpx,-altivec,-bpermd,-crypto,-direct-move,-extdiv,-power8-vector,-vsx" }
|
||||
attributes #1 = { argmemonly nounwind }
|
||||
attributes #2 = { "target-cpu"="a2q" "target-features"="+qpx,-altivec,-bpermd,-crypto,-direct-move,-extdiv,-power8-vector,-vsx" }
|
||||
attributes #3 = { nounwind }
|
||||
|
||||
|
|
@ -0,0 +1,39 @@
|
|||
; RUN: opt -S -basicaa -loop-vectorize < %s | FileCheck %s
|
||||
target datalayout = "e-m:o-i64:64-f80:128-n8:16:32:64-S128"
|
||||
target triple = "x86_64-apple-macosx10.11.0"
|
||||
|
||||
define i32 @accum(i32* nocapture readonly %x, i32 %N) #0 {
|
||||
entry:
|
||||
; CHECK-LABEL: @accum
|
||||
; CHECK-NOT: x i32>
|
||||
|
||||
%cmp1 = icmp sgt i32 %N, 0
|
||||
br i1 %cmp1, label %for.inc.preheader, label %for.end
|
||||
|
||||
for.inc.preheader:
|
||||
br label %for.inc
|
||||
|
||||
for.inc:
|
||||
%indvars.iv = phi i64 [ %indvars.iv.next, %for.inc ], [ 0, %for.inc.preheader ]
|
||||
%sum.02 = phi i32 [ %add, %for.inc ], [ 0, %for.inc.preheader ]
|
||||
%arrayidx = getelementptr inbounds i32, i32* %x, i64 %indvars.iv
|
||||
%0 = load i32, i32* %arrayidx, align 4
|
||||
%add = add nsw i32 %0, %sum.02
|
||||
%indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
|
||||
%lftr.wideiv = trunc i64 %indvars.iv.next to i32
|
||||
%exitcond = icmp eq i32 %lftr.wideiv, %N
|
||||
br i1 %exitcond, label %for.end.loopexit, label %for.inc
|
||||
|
||||
for.end.loopexit:
|
||||
%add.lcssa = phi i32 [ %add, %for.inc ]
|
||||
br label %for.end
|
||||
|
||||
for.end:
|
||||
%sum.0.lcssa = phi i32 [ 0, %entry ], [ %add.lcssa, %for.end.loopexit ]
|
||||
ret i32 %sum.0.lcssa
|
||||
|
||||
; CHECK: ret i32
|
||||
}
|
||||
|
||||
attributes #0 = { "target-cpu"="core2" "target-features"="+sse,-avx,-avx2,-sse2" }
|
||||
|
||||
Loading…
Reference in New Issue