[llvm] r271410 - [LV] For some IVs, use vector phis instead of widening in the loop body
Michael Kuperstein via llvm-commits
llvm-commits at lists.llvm.org
Fri Jun 3 08:04:00 PDT 2016
Sure, I'll take a look, thanks!
Michael
On Fri, Jun 3, 2016 at 4:12 AM, Arnaud De Grandmaison <
Arnaud.DeGrandmaison at arm.com> wrote:
> Hi Michael,
>
> We have seen significant performance regressions on some widely used
> industry benchmark.
>
> I've filed https://llvm.org/bugs/show_bug.cgi?id=27988 with a reproducer.
>
> As described in the Bugzilla ticket, the generated code looks not great at
> all on x86_64 or aarch64: none of the backends are able to do a proper job
> with the IV.
>
> Could you have a look into it ?
>
> Kind regards,
> Arnaud
>
> > -----Original Message-----
> > From: llvm-commits [mailto:llvm-commits-bounces at lists.llvm.org] On
> Behalf
> > Of Michael Kuperstein via llvm-commits
> > Sent: 01 June 2016 19:17
> > To: llvm-commits at lists.llvm.org
> > Subject: [llvm] r271410 - [LV] For some IVs, use vector phis instead of
> > widening in the loop body
> >
> > Author: mkuper
> > Date: Wed Jun 1 12:16:46 2016
> > New Revision: 271410
> >
> > URL: http://llvm.org/viewvc/llvm-project?rev=271410&view=rev
> > Log:
> > [LV] For some IVs, use vector phis instead of widening in the loop body
> >
> > Previously, whenever we needed a vector IV, we would create it on the
> fly,
> > by splatting the scalar IV and adding a step vector. Instead, we can
> create a
> > real vector IV. This tends to save a couple of instructions per
> iteration.
> >
> > This only changes the behavior for the most basic case - integer primary
> IVs
> > with a constant step.
> >
> > Differential Revision: http://reviews.llvm.org/D20315
> >
> > Modified:
> > llvm/trunk/lib/Transforms/Vectorize/LoopVectorize.cpp
> > llvm/trunk/test/Transforms/LoopVectorize/PowerPC/vsx-tsvc-s173.ll
> > llvm/trunk/test/Transforms/LoopVectorize/X86/gather_scatter.ll
> > llvm/trunk/test/Transforms/LoopVectorize/cast-induction.ll
> > llvm/trunk/test/Transforms/LoopVectorize/gcc-examples.ll
> > llvm/trunk/test/Transforms/LoopVectorize/gep_with_bitcast.ll
> > llvm/trunk/test/Transforms/LoopVectorize/global_alias.ll
> > llvm/trunk/test/Transforms/LoopVectorize/induction.ll
> > llvm/trunk/test/Transforms/LoopVectorize/induction_plus.ll
> >
> > Modified: llvm/trunk/lib/Transforms/Vectorize/LoopVectorize.cpp
> > URL: http://llvm.org/viewvc/llvm-
> > project/llvm/trunk/lib/Transforms/Vectorize/LoopVectorize.cpp?rev=271410
> > &r1=271409&r2=271410&view=diff
> > ==========================================================
> > ====================
> > --- llvm/trunk/lib/Transforms/Vectorize/LoopVectorize.cpp (original)
> > +++ llvm/trunk/lib/Transforms/Vectorize/LoopVectorize.cpp Wed Jun 1
> > +++ 12:16:46 2016
> > @@ -422,6 +422,14 @@ protected:
> > /// from SCEV or creates a new using SCEVExpander.
> > virtual Value *getStepVector(Value *Val, int StartIdx, const SCEV
> *Step);
> >
> > + /// Create a vector induction variable based on an existing scalar
> one.
> > + /// Currently only works for integer primary induction variables with
> > + /// a constant step.
> > + /// If TruncType is provided, instead of widening the original IV, we
> > + /// widen a version of the IV truncated to TruncType.
> > + void widenInductionVariable(const InductionDescriptor &II, VectorParts
> > &Entry,
> > + IntegerType *TruncType = nullptr);
> > +
> > /// When we go over instructions in the basic block we rely on
> previous
> > /// values within the current basic block or on loop invariant values.
> > /// When we widen (vectorize) values we place them in the map. If the
> > values @@ -2099,6 +2107,40 @@ Value *InnerLoopVectorizer::getStepVecto
> > return getStepVector(Val, StartIdx, StepValue); }
> >
> > +void InnerLoopVectorizer::widenInductionVariable(const
> > InductionDescriptor &II,
> > + VectorParts &Entry,
> > + IntegerType
> > +*TruncType) {
> > + Value *Start = II.getStartValue();
> > + ConstantInt *Step = II.getConstIntStepValue();
> > + assert(Step && "Can not widen an IV with a non-constant step");
> > +
> > + // Construct the initial value of the vector IV in the vector loop
> > + preheader auto CurrIP = Builder.saveIP();
> > + Builder.SetInsertPoint(LoopVectorPreHeader->getTerminator());
> > + if (TruncType) {
> > + Step = ConstantInt::getSigned(TruncType, Step->getSExtValue());
> > + Start = Builder.CreateCast(Instruction::Trunc, Start, TruncType);
> > + } Value *SplatStart = Builder.CreateVectorSplat(VF, Start); Value
> > + *SteppedStart = getStepVector(SplatStart, 0, Step);
> > + Builder.restoreIP(CurrIP);
> > +
> > + Value *SplatVF =
> > + ConstantVector::getSplat(VF, ConstantInt::get(Start->getType(),
> > + VF)); // We may need to add the step a number of times, depending on
> > + the unroll // factor. The last of those goes into the PHI.
> > + PHINode *VecInd = PHINode::Create(SteppedStart->getType(), 2,
> > "vec.ind",
> > +
> > + &*LoopVectorBody->getFirstInsertionPt());
> > + Value *LastInduction = VecInd;
> > + for (unsigned Part = 0; Part < UF; ++Part) {
> > + Entry[Part] = LastInduction;
> > + LastInduction = Builder.CreateAdd(LastInduction, SplatVF,
> > + "step.add"); }
> > +
> > + VecInd->addIncoming(SteppedStart, LoopVectorPreHeader);
> > + VecInd->addIncoming(LastInduction, LoopVectorBody); }
> > +
> > Value *InnerLoopVectorizer::getStepVector(Value *Val, int StartIdx,
> > Value *Step) {
> > assert(Val->getType()->isVectorTy() && "Must be a vector"); @@
> -4056,19
> > +4098,25 @@ void InnerLoopVectorizer::widenPHIInstru
> > llvm_unreachable("Unknown induction");
> > case InductionDescriptor::IK_IntInduction: {
> > assert(P->getType() == II.getStartValue()->getType() && "Types must
> > match");
> > - // Handle other induction variables that are now based on the
> > - // canonical one.
> > - Value *V = Induction;
> > - if (P != OldInduction) {
> > - V = Builder.CreateSExtOrTrunc(Induction, P->getType());
> > - V = II.transform(Builder, V, PSE.getSE(), DL);
> > - V->setName("offset.idx");
> > - }
> > - Value *Broadcasted = getBroadcastInstrs(V);
> > - // After broadcasting the induction variable we need to make the
> vector
> > - // consecutive by adding 0, 1, 2, etc.
> > - for (unsigned part = 0; part < UF; ++part)
> > - Entry[part] = getStepVector(Broadcasted, VF * part, II.getStep());
> > + if (P != OldInduction || VF == 1) {
> > + Value *V = Induction;
> > + // Handle other induction variables that are now based on the
> > + // canonical one.
> > + if (P != OldInduction) {
> > + V = Builder.CreateSExtOrTrunc(Induction, P->getType());
> > + V = II.transform(Builder, V, PSE.getSE(), DL);
> > + V->setName("offset.idx");
> > + }
> > + Value *Broadcasted = getBroadcastInstrs(V);
> > + // After broadcasting the induction variable we need to make the
> vector
> > + // consecutive by adding 0, 1, 2, etc.
> > + for (unsigned part = 0; part < UF; ++part)
> > + Entry[part] = getStepVector(Broadcasted, VF * part,
> II.getStep());
> > + } else {
> > + // Instead of re-creating the vector IV by splatting the scalar IV
> > + // in each iteration, we can make a new independent vector IV.
> > + widenInductionVariable(II, Entry);
> > + }
> > return;
> > }
> > case InductionDescriptor::IK_PtrInduction:
> > @@ -4239,15 +4287,23 @@ void InnerLoopVectorizer::vectorizeBlock
> > if (CI->getOperand(0) == OldInduction &&
> > it->getOpcode() == Instruction::Trunc) {
> > InductionDescriptor II =
> > - Legal->getInductionVars()->lookup(OldInduction);
> > + Legal->getInductionVars()->lookup(OldInduction);
> > if (auto StepValue = II.getConstIntStepValue()) {
> > - StepValue = ConstantInt::getSigned(cast<IntegerType>(CI-
> > >getType()),
> > - StepValue->getSExtValue());
> > - Value *ScalarCast = Builder.CreateCast(CI->getOpcode(),
> Induction,
> > - CI->getType());
> > - Value *Broadcasted = getBroadcastInstrs(ScalarCast);
> > - for (unsigned Part = 0; Part < UF; ++Part)
> > - Entry[Part] = getStepVector(Broadcasted, VF * Part,
> StepValue);
> > + IntegerType *TruncType = cast<IntegerType>(CI->getType());
> > + if (VF == 1) {
> > + StepValue =
> > + ConstantInt::getSigned(TruncType,
> StepValue->getSExtValue());
> > + Value *ScalarCast =
> > + Builder.CreateCast(CI->getOpcode(), Induction,
> CI->getType());
> > + Value *Broadcasted = getBroadcastInstrs(ScalarCast);
> > + for (unsigned Part = 0; Part < UF; ++Part)
> > + Entry[Part] = getStepVector(Broadcasted, VF * Part,
> StepValue);
> > + } else {
> > + // Truncating a vector induction variable on each iteration
> > + // may be expensive. Instead, truncate the initial value,
> and create
> > + // a new, truncated, vector IV based on that.
> > + widenInductionVariable(II, Entry, TruncType);
> > + }
> > addMetadata(Entry, &*it);
> > break;
> > }
> >
> > Modified: llvm/trunk/test/Transforms/LoopVectorize/PowerPC/vsx-tsvc-
> > s173.ll
> > URL: http://llvm.org/viewvc/llvm-
> > project/llvm/trunk/test/Transforms/LoopVectorize/PowerPC/vsx-tsvc-
> > s173.ll?rev=271410&r1=271409&r2=271410&view=diff
> > ==========================================================
> > ====================
> > --- llvm/trunk/test/Transforms/LoopVectorize/PowerPC/vsx-tsvc-s173.ll
> > (original)
> > +++ llvm/trunk/test/Transforms/LoopVectorize/PowerPC/vsx-tsvc-s173.ll
> > +++ Wed Jun 1 12:16:46 2016
> > @@ -43,7 +43,7 @@ for.end12:
> >
> > ; CHECK-LABEL: @s173
> > ; CHECK: load <4 x float>, <4 x float>* -; CHECK: add i64 %index, 16000
> > +; CHECK: add nsw i64 %.lhs, 16000
> > ; CHECK: ret i32 0
> > }
> >
> >
> > Modified: llvm/trunk/test/Transforms/LoopVectorize/X86/gather_scatter.ll
> > URL: http://llvm.org/viewvc/llvm-
> > project/llvm/trunk/test/Transforms/LoopVectorize/X86/gather_scatter.ll?re
> > v=271410&r1=271409&r2=271410&view=diff
> > ==========================================================
> > ====================
> > --- llvm/trunk/test/Transforms/LoopVectorize/X86/gather_scatter.ll
> > (original)
> > +++ llvm/trunk/test/Transforms/LoopVectorize/X86/gather_scatter.ll Wed
> > +++ Jun 1 12:16:46 2016
> > @@ -95,7 +95,7 @@ for.end:
> > %struct.In = type { float, float }
> >
> > ;AVX512-LABEL: @foo2
> > -;AVX512: getelementptr %struct.In, %struct.In* %in, <16 x i64>
> %induction,
> > i32 1
> > +;AVX512: getelementptr %struct.In, %struct.In* %in, <16 x i64> %{{.*}},
> > +i32 1
> > ;AVX512: llvm.masked.gather.v16f32
> > ;AVX512: llvm.masked.store.v16f32
> > ;AVX512: ret void
> > @@ -170,10 +170,10 @@ for.end:
> > ;}
> >
> > ;AVX512-LABEL: @foo3
> > -;AVX512: getelementptr %struct.In, %struct.In* %in, <16 x i64>
> %induction,
> > i32 1
> > +;AVX512: getelementptr %struct.In, %struct.In* %in, <16 x i64> %{{.*}},
> > +i32 1
> > ;AVX512: llvm.masked.gather.v16f32
> > ;AVX512: fadd <16 x float>
> > -;AVX512: getelementptr %struct.Out, %struct.Out* %out, <16 x i64>
> > %induction, i32 1
> > +;AVX512: getelementptr %struct.Out, %struct.Out* %out, <16 x i64>
> > +%{{.*}}, i32 1
> > ;AVX512: llvm.masked.scatter.v16f32
> > ;AVX512: ret void
> >
> >
> > Modified: llvm/trunk/test/Transforms/LoopVectorize/cast-induction.ll
> > URL: http://llvm.org/viewvc/llvm-
> > project/llvm/trunk/test/Transforms/LoopVectorize/cast-
> > induction.ll?rev=271410&r1=271409&r2=271410&view=diff
> > ==========================================================
> > ====================
> > --- llvm/trunk/test/Transforms/LoopVectorize/cast-induction.ll (original)
> > +++ llvm/trunk/test/Transforms/LoopVectorize/cast-induction.ll Wed Jun
> > +++ 1 12:16:46 2016
> > @@ -8,7 +8,7 @@ target triple = "x86_64-apple-macosx10.8 @a = common
> > global [2048 x i32] zeroinitializer, align 16
> >
> > ;CHECK-LABEL: @example12(
> > -;CHECK: trunc i64
> > +;CHECK: %vec.ind1 = phi <4 x i32>
> > ;CHECK: store <4 x i32>
> > ;CHECK: ret void
> > define void @example12() nounwind uwtable ssp {
> >
> > Modified: llvm/trunk/test/Transforms/LoopVectorize/gcc-examples.ll
> > URL: http://llvm.org/viewvc/llvm-
> > project/llvm/trunk/test/Transforms/LoopVectorize/gcc-
> > examples.ll?rev=271410&r1=271409&r2=271410&view=diff
> > ==========================================================
> > ====================
> > --- llvm/trunk/test/Transforms/LoopVectorize/gcc-examples.ll (original)
> > +++ llvm/trunk/test/Transforms/LoopVectorize/gcc-examples.ll Wed Jun 1
> > +++ 12:16:46 2016
> > @@ -368,7 +368,7 @@ define void @example11() nounwind uwtabl }
> >
> > ;CHECK-LABEL: @example12(
> > -;CHECK: trunc i64
> > +;CHECK: %vec.ind1 = phi <4 x i32>
> > ;CHECK: store <4 x i32>
> > ;CHECK: ret void
> > define void @example12() nounwind uwtable ssp {
> >
> > Modified: llvm/trunk/test/Transforms/LoopVectorize/gep_with_bitcast.ll
> > URL: http://llvm.org/viewvc/llvm-
> > project/llvm/trunk/test/Transforms/LoopVectorize/gep_with_bitcast.ll?rev=
> > 271410&r1=271409&r2=271410&view=diff
> > ==========================================================
> > ====================
> > --- llvm/trunk/test/Transforms/LoopVectorize/gep_with_bitcast.ll
> (original)
> > +++ llvm/trunk/test/Transforms/LoopVectorize/gep_with_bitcast.ll Wed Jun
> > +++ 1 12:16:46 2016
> > @@ -12,10 +12,11 @@ target datalayout = "e-m:e-i64:64-i128:1
> >
> > ; CHECK-LABEL: @foo
> > ; CHECK: vector.body
> > -; CHECK: %0 = getelementptr inbounds double*, double** %in, i64 %index
> -
> > ; CHECK: %1 = bitcast double** %0 to <4 x i64>* -; CHECK: %wide.load =
> load
> > <4 x i64>, <4 x i64>* %1, align 8 -; CHECK: %2 = icmp eq <4 x i64>
> %wide.load,
> > zeroinitializer
> > +; CHECK: %0 = phi
> > +; CHECK: %2 = getelementptr inbounds double*, double** %in, i64 %0 ;
> > +CHECK: %3 = bitcast double** %2 to <4 x i64>* ; CHECK: %wide.load =
> > +load <4 x i64>, <4 x i64>* %3, align 8 ; CHECK: %4 = icmp eq <4 x i64>
> > +%wide.load, zeroinitializer
> > ; CHECK: br i1
> >
> > define void @foo(double** noalias nocapture readonly %in, double**
> > noalias nocapture readnone %out, i8* noalias nocapture %res) #0 { @@
> -37,4
> > +38,4 @@ for.body:
> >
> > for.end:
> > ret void
> > -}
> > \ No newline at end of file
> > +}
> >
> > Modified: llvm/trunk/test/Transforms/LoopVectorize/global_alias.ll
> > URL: http://llvm.org/viewvc/llvm-
> >
> project/llvm/trunk/test/Transforms/LoopVectorize/global_alias.ll?rev=27141
> > 0&r1=271409&r2=271410&view=diff
> > ==========================================================
> > ====================
> > --- llvm/trunk/test/Transforms/LoopVectorize/global_alias.ll (original)
> > +++ llvm/trunk/test/Transforms/LoopVectorize/global_alias.ll Wed Jun 1
> > +++ 12:16:46 2016
> > @@ -12,7 +12,7 @@ target datalayout = "e-p:32:32:32-i1:8:8 @PA =
> external
> > global i32*
> >
> >
> > -;; === First, the tests that should always vectorize, wither statically
> or by
> > adding run-time checks ===
> > +;; === First, the tests that should always vectorize, whether
> > +statically or by adding run-time checks ===
> >
> >
> > ; /// Different objects, positive induction, constant distance @@ -387,7
> > +387,7 @@ for.end:
> > ; return Foo.A[a];
> > ; }
> > ; CHECK-LABEL: define i32 @noAlias08(
> > -; CHECK: sub <4 x i32>
> > +; CHECK: sub nuw nsw <4 x i32>
> > ; CHECK: ret
> >
> > define i32 @noAlias08(i32 %a) #0 {
> > @@ -439,7 +439,7 @@ for.end:
> > ; return Foo.A[a];
> > ; }
> > ; CHECK-LABEL: define i32 @noAlias09(
> > -; CHECK: sub <4 x i32>
> > +; CHECK: sub nuw nsw <4 x i32>
> > ; CHECK: ret
> >
> > define i32 @noAlias09(i32 %a) #0 {
> > @@ -721,7 +721,7 @@ for.end:
> > ; return Foo.A[a];
> > ; }
> > ; CHECK-LABEL: define i32 @noAlias14(
> > -; CHECK: sub <4 x i32>
> > +; CHECK: sub nuw nsw <4 x i32>
> > ; CHECK: ret
> >
> > define i32 @noAlias14(i32 %a) #0 {
> >
> > Modified: llvm/trunk/test/Transforms/LoopVectorize/induction.ll
> > URL: http://llvm.org/viewvc/llvm-
> > project/llvm/trunk/test/Transforms/LoopVectorize/induction.ll?rev=271410
> > &r1=271409&r2=271410&view=diff
> > ==========================================================
> > ====================
> > --- llvm/trunk/test/Transforms/LoopVectorize/induction.ll (original)
> > +++ llvm/trunk/test/Transforms/LoopVectorize/induction.ll Wed Jun 1
> > +++ 12:16:46 2016
> > @@ -1,4 +1,6 @@
> > ; RUN: opt < %s -loop-vectorize -force-vector-interleave=1
> -force-vector-
> > width=2 -S | FileCheck %s
> > +; RUN: opt < %s -loop-vectorize -force-vector-interleave=1
> > +-force-vector-width=2 -instcombine -S | FileCheck %s --check-prefix=IND
> > +; RUN: opt < %s -loop-vectorize -force-vector-interleave=2
> > +-force-vector-width=2 -instcombine -S | FileCheck %s
> > +--check-prefix=UNROLL
> >
> > target datalayout =
> "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-
> > f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-
> > n8:16:32:64-S128"
> >
> > @@ -27,8 +29,6 @@ for.end:
> > ret void
> > }
> >
> > -; RUN: opt < %s -loop-vectorize -force-vector-interleave=1
> -force-vector-
> > width=2 -instcombine -S | FileCheck %s --check-prefix=IND
> > -
> > ; Make sure we remove unneeded vectorization of induction variables.
> > ; In order for instcombine to cleanup the vectorized induction
> variables that
> > we ; create in the loop vectorizer we need to perform some form of
> > redundancy @@ -241,3 +241,64 @@ entry:
> > exit:
> > ret void
> > }
> > +
> > +; Check that we generate vectorized IVs in the pre-header ; instead of
> > +widening the scalar IV inside the loop, when ; we know how to do that.
> > +; IND-LABEL: veciv
> > +; IND: vector.body:
> > +; IND: %index = phi i32 [ 0, %vector.ph ], [ %index.next, %vector.body
> > +] ; IND: %vec.ind = phi <2 x i32> [ <i32 0, i32 1>, %vector.ph ], [
> > +%step.add, %vector.body ] ; IND: %step.add = add <2 x i32> %vec.ind,
> > +<i32 2, i32 2> ; IND: %index.next = add i32 %index, 2 ; IND:
> > +%[[CMP:.*]] = icmp eq i32 %index.next ; IND: br i1 %[[CMP]] ;
> > +UNROLL-LABEL: veciv ; UNROLL: vector.body:
> > +; UNROLL: %index = phi i32 [ 0, %vector.ph ], [ %index.next,
> > +%vector.body ] ; UNROLL: %vec.ind = phi <2 x i32> [ <i32 0, i32 1>,
> > +%vector.ph ], [ %step.add1, %vector.body ] ; UNROLL: %step.add = add <2
> > +x i32> %vec.ind, <i32 2, i32 2> ; UNROLL: %step.add1 = add <2 x i32>
> > +%vec.ind, <i32 4, i32 4> ; UNROLL: %index.next = add i32 %index, 4 ;
> > +UNROLL: %[[CMP:.*]] = icmp eq i32 %index.next ; UNROLL: br i1 %[[CMP]]
> > +define void @veciv(i32* nocapture %a, i32 %start, i32 %k) {
> > +for.body.preheader:
> > + br label %for.body
> > +
> > +for.body:
> > + %indvars.iv = phi i32 [ %indvars.iv.next, %for.body ], [ 0,
> > +%for.body.preheader ]
> > + %arrayidx = getelementptr inbounds i32, i32* %a, i32 %indvars.iv
> > + store i32 %indvars.iv, i32* %arrayidx, align 4
> > + %indvars.iv.next = add nuw nsw i32 %indvars.iv, 1
> > + %exitcond = icmp eq i32 %indvars.iv.next, %k
> > + br i1 %exitcond, label %exit, label %for.body
> > +
> > +exit:
> > + ret void
> > +}
> > +
> > +; IND-LABEL: trunciv
> > +; IND: vector.body:
> > +; IND: %index = phi i64 [ 0, %vector.ph ], [ %index.next, %vector.body
> > +] ; IND: %[[VECIND:.*]] = phi <2 x i32> [ <i32 0, i32 1>, %vector.ph ],
> > +[ %[[STEPADD:.*]], %vector.body ] ; IND: %[[STEPADD]] = add <2 x i32>
> > +%[[VECIND]], <i32 2, i32 2> ; IND: %index.next = add i64 %index, 2 ;
> > +IND: %[[CMP:.*]] = icmp eq i64 %index.next ; IND: br i1 %[[CMP]] define
> > +void @trunciv(i32* nocapture %a, i32 %start, i64 %k) {
> > +for.body.preheader:
> > + br label %for.body
> > +
> > +for.body:
> > + %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0,
> > +%for.body.preheader ]
> > + %trunc.iv = trunc i64 %indvars.iv to i32
> > + %arrayidx = getelementptr inbounds i32, i32* %a, i32 %trunc.iv
> > + store i32 %trunc.iv, i32* %arrayidx, align 4
> > + %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
> > + %exitcond = icmp eq i64 %indvars.iv.next, %k
> > + br i1 %exitcond, label %exit, label %for.body
> > +
> > +exit:
> > + ret void
> > +}
> >
> > Modified: llvm/trunk/test/Transforms/LoopVectorize/induction_plus.ll
> > URL: http://llvm.org/viewvc/llvm-
> > project/llvm/trunk/test/Transforms/LoopVectorize/induction_plus.ll?rev=27
> > 1410&r1=271409&r2=271410&view=diff
> > ==========================================================
> > ====================
> > --- llvm/trunk/test/Transforms/LoopVectorize/induction_plus.ll (original)
> > +++ llvm/trunk/test/Transforms/LoopVectorize/induction_plus.ll Wed Jun
> > +++ 1 12:16:46 2016
> > @@ -1,4 +1,4 @@
> > -; RUN: opt < %s -loop-vectorize -force-vector-interleave=1
> -force-vector-
> > width=4 -instcombine -S | FileCheck %s
> > +; RUN: opt < %s -loop-vectorize -force-vector-interleave=1
> > +-force-vector-width=4 -S | FileCheck %s
> >
> > target datalayout =
> "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-
> > f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-
> > n8:16:32:64-S128"
> > target triple = "x86_64-apple-macosx10.8.0"
> > @@ -6,8 +6,11 @@ target triple = "x86_64-apple-macosx10.8 @array =
> > common global [1024 x i32] zeroinitializer, align 16
> >
> > ;CHECK-LABEL: @array_at_plus_one(
> > -;CHECK: add i64 %index, 12
> > -;CHECK: trunc i64
> > +;CHECK: %index = phi i64 [ 0, %vector.ph ], [ %index.next, %vector.body
> > +]
> > +;CHECK: %vec.ind = phi <4 x i64> [ <i64 0, i64 1, i64 2, i64 3>,
> > +%vector.ph ], [ %step.add, %vector.body ]
> > +;CHECK: %vec.ind1 = phi <4 x i32> [ <i32 0, i32 1, i32 2, i32 3>,
> > +%vector.ph ], [ %step.add2, %vector.body ]
> > +;CHECK: add <4 x i64> %vec.ind, <i64 4, i64 4, i64 4, i64 4>
> > +;CHECK: add nsw <4 x i64> %vec.ind, <i64 12, i64 12, i64 12, i64 12>
> > ;CHECK: ret i32
> > define i32 @array_at_plus_one(i32 %n) nounwind uwtable ssp {
> > %1 = icmp sgt i32 %n, 0
> >
> >
> > _______________________________________________
> > llvm-commits mailing list
> > llvm-commits at lists.llvm.org
> > http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-commits
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20160603/c1e104a9/attachment.html>
More information about the llvm-commits
mailing list