[llvm] cec24f0 - [VPlan] Update stale test after 9536a6286, fix formatting.
Florian Hahn via llvm-commits
llvm-commits at lists.llvm.org
Wed Jan 31 05:45:53 PST 2024
Author: Florian Hahn
Date: 2024-01-31T13:45:38Z
New Revision: cec24f0d7efc206e329890821e92a4b25c550885
URL: https://github.com/llvm/llvm-project/commit/cec24f0d7efc206e329890821e92a4b25c550885
DIFF: https://github.com/llvm/llvm-project/commit/cec24f0d7efc206e329890821e92a4b25c550885.diff
LOG: [VPlan] Update stale test after 9536a6286, fix formatting.
Added:
Modified:
llvm/lib/Transforms/Vectorize/VPlanTransforms.cpp
llvm/test/Transforms/LoopVectorize/PowerPC/optimal-epilog-vectorization.ll
Removed:
################################################################################
diff --git a/llvm/lib/Transforms/Vectorize/VPlanTransforms.cpp b/llvm/lib/Transforms/Vectorize/VPlanTransforms.cpp
index 26cb87184c7a8..71f5285f90236 100644
--- a/llvm/lib/Transforms/Vectorize/VPlanTransforms.cpp
+++ b/llvm/lib/Transforms/Vectorize/VPlanTransforms.cpp
@@ -492,7 +492,7 @@ void VPlanTransforms::removeDeadRecipes(VPlan &Plan) {
static VPValue *createScalarIVSteps(VPlan &Plan, const InductionDescriptor &ID,
ScalarEvolution &SE, Instruction *TruncI,
VPValue *StartV, VPValue *Step,
- VPBasicBlock::iterator IP ) {
+ VPBasicBlock::iterator IP) {
VPBasicBlock *HeaderVPBB = Plan.getVectorLoopRegion()->getEntryBasicBlock();
VPCanonicalIVPHIRecipe *CanonicalIV = Plan.getCanonicalIV();
VPSingleDefRecipe *BaseIV = CanonicalIV;
@@ -546,9 +546,9 @@ void VPlanTransforms::optimizeInductions(VPlan &Plan, ScalarEvolution &SE) {
continue;
const InductionDescriptor &ID = WideIV->getInductionDescriptor();
- VPValue *Steps =
- createScalarIVSteps(Plan, ID, SE, WideIV->getTruncInst(),
- WideIV->getStartValue(), WideIV->getStepValue(), InsertPt);
+ VPValue *Steps = createScalarIVSteps(Plan, ID, SE, WideIV->getTruncInst(),
+ WideIV->getStartValue(),
+ WideIV->getStepValue(), InsertPt);
// Update scalar users of IV to use Step instead.
if (!HasOnlyVectorVFs)
diff --git a/llvm/test/Transforms/LoopVectorize/PowerPC/optimal-epilog-vectorization.ll b/llvm/test/Transforms/LoopVectorize/PowerPC/optimal-epilog-vectorization.ll
index 0bee21a481d06..b4ee4249109d5 100644
--- a/llvm/test/Transforms/LoopVectorize/PowerPC/optimal-epilog-vectorization.ll
+++ b/llvm/test/Transforms/LoopVectorize/PowerPC/optimal-epilog-vectorization.ll
@@ -389,31 +389,31 @@ define dso_local signext i32 @f2(ptr noalias %A, ptr noalias %B, i32 signext %n)
; VF-TWO-CHECK-NEXT: br label [[VECTOR_BODY:%.*]]
; VF-TWO-CHECK: vector.body:
; VF-TWO-CHECK-NEXT: [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; VF-TWO-CHECK-NEXT: [[TMP8:%.*]] = add i64 [[INDEX]], 0
+; VF-TWO-CHECK-NEXT: [[TMP9:%.*]] = add i64 [[INDEX]], 4
+; VF-TWO-CHECK-NEXT: [[TMP10:%.*]] = add i64 [[INDEX]], 8
+; VF-TWO-CHECK-NEXT: [[TMP11:%.*]] = add i64 [[INDEX]], 12
+; VF-TWO-CHECK-NEXT: [[TMP12:%.*]] = add i64 [[INDEX]], 16
+; VF-TWO-CHECK-NEXT: [[TMP13:%.*]] = add i64 [[INDEX]], 20
+; VF-TWO-CHECK-NEXT: [[TMP14:%.*]] = add i64 [[INDEX]], 24
+; VF-TWO-CHECK-NEXT: [[TMP15:%.*]] = add i64 [[INDEX]], 28
; VF-TWO-CHECK-NEXT: [[OFFSET_IDX:%.*]] = trunc i64 [[INDEX]] to i32
-; VF-TWO-CHECK-NEXT: [[TMP8:%.*]] = add i32 [[OFFSET_IDX]], 0
-; VF-TWO-CHECK-NEXT: [[TMP9:%.*]] = add i32 [[OFFSET_IDX]], 4
-; VF-TWO-CHECK-NEXT: [[TMP10:%.*]] = add i32 [[OFFSET_IDX]], 8
-; VF-TWO-CHECK-NEXT: [[TMP11:%.*]] = add i32 [[OFFSET_IDX]], 12
-; VF-TWO-CHECK-NEXT: [[TMP12:%.*]] = add i32 [[OFFSET_IDX]], 16
-; VF-TWO-CHECK-NEXT: [[TMP13:%.*]] = add i32 [[OFFSET_IDX]], 20
-; VF-TWO-CHECK-NEXT: [[TMP14:%.*]] = add i32 [[OFFSET_IDX]], 24
-; VF-TWO-CHECK-NEXT: [[TMP15:%.*]] = add i32 [[OFFSET_IDX]], 28
-; VF-TWO-CHECK-NEXT: [[TMP16:%.*]] = add i64 [[INDEX]], 0
-; VF-TWO-CHECK-NEXT: [[TMP17:%.*]] = add i64 [[INDEX]], 4
-; VF-TWO-CHECK-NEXT: [[TMP18:%.*]] = add i64 [[INDEX]], 8
-; VF-TWO-CHECK-NEXT: [[TMP19:%.*]] = add i64 [[INDEX]], 12
-; VF-TWO-CHECK-NEXT: [[TMP20:%.*]] = add i64 [[INDEX]], 16
-; VF-TWO-CHECK-NEXT: [[TMP21:%.*]] = add i64 [[INDEX]], 20
-; VF-TWO-CHECK-NEXT: [[TMP22:%.*]] = add i64 [[INDEX]], 24
-; VF-TWO-CHECK-NEXT: [[TMP23:%.*]] = add i64 [[INDEX]], 28
-; VF-TWO-CHECK-NEXT: [[TMP24:%.*]] = xor i32 [[TMP8]], -1
-; VF-TWO-CHECK-NEXT: [[TMP25:%.*]] = xor i32 [[TMP9]], -1
-; VF-TWO-CHECK-NEXT: [[TMP26:%.*]] = xor i32 [[TMP10]], -1
-; VF-TWO-CHECK-NEXT: [[TMP27:%.*]] = xor i32 [[TMP11]], -1
-; VF-TWO-CHECK-NEXT: [[TMP28:%.*]] = xor i32 [[TMP12]], -1
-; VF-TWO-CHECK-NEXT: [[TMP29:%.*]] = xor i32 [[TMP13]], -1
-; VF-TWO-CHECK-NEXT: [[TMP30:%.*]] = xor i32 [[TMP14]], -1
-; VF-TWO-CHECK-NEXT: [[TMP31:%.*]] = xor i32 [[TMP15]], -1
+; VF-TWO-CHECK-NEXT: [[TMP16:%.*]] = add i32 [[OFFSET_IDX]], 0
+; VF-TWO-CHECK-NEXT: [[TMP17:%.*]] = add i32 [[OFFSET_IDX]], 4
+; VF-TWO-CHECK-NEXT: [[TMP18:%.*]] = add i32 [[OFFSET_IDX]], 8
+; VF-TWO-CHECK-NEXT: [[TMP19:%.*]] = add i32 [[OFFSET_IDX]], 12
+; VF-TWO-CHECK-NEXT: [[TMP20:%.*]] = add i32 [[OFFSET_IDX]], 16
+; VF-TWO-CHECK-NEXT: [[TMP21:%.*]] = add i32 [[OFFSET_IDX]], 20
+; VF-TWO-CHECK-NEXT: [[TMP22:%.*]] = add i32 [[OFFSET_IDX]], 24
+; VF-TWO-CHECK-NEXT: [[TMP23:%.*]] = add i32 [[OFFSET_IDX]], 28
+; VF-TWO-CHECK-NEXT: [[TMP24:%.*]] = xor i32 [[TMP16]], -1
+; VF-TWO-CHECK-NEXT: [[TMP25:%.*]] = xor i32 [[TMP17]], -1
+; VF-TWO-CHECK-NEXT: [[TMP26:%.*]] = xor i32 [[TMP18]], -1
+; VF-TWO-CHECK-NEXT: [[TMP27:%.*]] = xor i32 [[TMP19]], -1
+; VF-TWO-CHECK-NEXT: [[TMP28:%.*]] = xor i32 [[TMP20]], -1
+; VF-TWO-CHECK-NEXT: [[TMP29:%.*]] = xor i32 [[TMP21]], -1
+; VF-TWO-CHECK-NEXT: [[TMP30:%.*]] = xor i32 [[TMP22]], -1
+; VF-TWO-CHECK-NEXT: [[TMP31:%.*]] = xor i32 [[TMP23]], -1
; VF-TWO-CHECK-NEXT: [[TMP32:%.*]] = add i32 [[TMP24]], [[N]]
; VF-TWO-CHECK-NEXT: [[TMP33:%.*]] = add i32 [[TMP25]], [[N]]
; VF-TWO-CHECK-NEXT: [[TMP34:%.*]] = add i32 [[TMP26]], [[N]]
@@ -478,14 +478,14 @@ define dso_local signext i32 @f2(ptr noalias %A, ptr noalias %B, i32 signext %n)
; VF-TWO-CHECK-NEXT: [[TMP77:%.*]] = fadd fast <4 x float> [[REVERSE11]], <float 1.000000e+00, float 1.000000e+00, float 1.000000e+00, float 1.000000e+00>
; VF-TWO-CHECK-NEXT: [[TMP78:%.*]] = fadd fast <4 x float> [[REVERSE13]], <float 1.000000e+00, float 1.000000e+00, float 1.000000e+00, float 1.000000e+00>
; VF-TWO-CHECK-NEXT: [[TMP79:%.*]] = fadd fast <4 x float> [[REVERSE15]], <float 1.000000e+00, float 1.000000e+00, float 1.000000e+00, float 1.000000e+00>
-; VF-TWO-CHECK-NEXT: [[TMP80:%.*]] = getelementptr inbounds float, ptr [[A]], i64 [[TMP16]]
-; VF-TWO-CHECK-NEXT: [[TMP81:%.*]] = getelementptr inbounds float, ptr [[A]], i64 [[TMP17]]
-; VF-TWO-CHECK-NEXT: [[TMP82:%.*]] = getelementptr inbounds float, ptr [[A]], i64 [[TMP18]]
-; VF-TWO-CHECK-NEXT: [[TMP83:%.*]] = getelementptr inbounds float, ptr [[A]], i64 [[TMP19]]
-; VF-TWO-CHECK-NEXT: [[TMP84:%.*]] = getelementptr inbounds float, ptr [[A]], i64 [[TMP20]]
-; VF-TWO-CHECK-NEXT: [[TMP85:%.*]] = getelementptr inbounds float, ptr [[A]], i64 [[TMP21]]
-; VF-TWO-CHECK-NEXT: [[TMP86:%.*]] = getelementptr inbounds float, ptr [[A]], i64 [[TMP22]]
-; VF-TWO-CHECK-NEXT: [[TMP87:%.*]] = getelementptr inbounds float, ptr [[A]], i64 [[TMP23]]
+; VF-TWO-CHECK-NEXT: [[TMP80:%.*]] = getelementptr inbounds float, ptr [[A]], i64 [[TMP8]]
+; VF-TWO-CHECK-NEXT: [[TMP81:%.*]] = getelementptr inbounds float, ptr [[A]], i64 [[TMP9]]
+; VF-TWO-CHECK-NEXT: [[TMP82:%.*]] = getelementptr inbounds float, ptr [[A]], i64 [[TMP10]]
+; VF-TWO-CHECK-NEXT: [[TMP83:%.*]] = getelementptr inbounds float, ptr [[A]], i64 [[TMP11]]
+; VF-TWO-CHECK-NEXT: [[TMP84:%.*]] = getelementptr inbounds float, ptr [[A]], i64 [[TMP12]]
+; VF-TWO-CHECK-NEXT: [[TMP85:%.*]] = getelementptr inbounds float, ptr [[A]], i64 [[TMP13]]
+; VF-TWO-CHECK-NEXT: [[TMP86:%.*]] = getelementptr inbounds float, ptr [[A]], i64 [[TMP14]]
+; VF-TWO-CHECK-NEXT: [[TMP87:%.*]] = getelementptr inbounds float, ptr [[A]], i64 [[TMP15]]
; VF-TWO-CHECK-NEXT: [[TMP88:%.*]] = getelementptr inbounds float, ptr [[TMP80]], i32 0
; VF-TWO-CHECK-NEXT: [[TMP89:%.*]] = getelementptr inbounds float, ptr [[TMP80]], i32 4
; VF-TWO-CHECK-NEXT: [[TMP90:%.*]] = getelementptr inbounds float, ptr [[TMP80]], i32 8
@@ -521,10 +521,10 @@ define dso_local signext i32 @f2(ptr noalias %A, ptr noalias %B, i32 signext %n)
; VF-TWO-CHECK-NEXT: br label [[VEC_EPILOG_VECTOR_BODY:%.*]]
; VF-TWO-CHECK: vec.epilog.vector.body:
; VF-TWO-CHECK-NEXT: [[INDEX21:%.*]] = phi i64 [ [[VEC_EPILOG_RESUME_VAL]], [[VEC_EPILOG_PH]] ], [ [[INDEX_NEXT25:%.*]], [[VEC_EPILOG_VECTOR_BODY]] ]
+; VF-TWO-CHECK-NEXT: [[TMP97:%.*]] = add i64 [[INDEX21]], 0
; VF-TWO-CHECK-NEXT: [[OFFSET_IDX22:%.*]] = trunc i64 [[INDEX21]] to i32
-; VF-TWO-CHECK-NEXT: [[TMP97:%.*]] = add i32 [[OFFSET_IDX22]], 0
-; VF-TWO-CHECK-NEXT: [[TMP98:%.*]] = add i64 [[INDEX21]], 0
-; VF-TWO-CHECK-NEXT: [[TMP99:%.*]] = xor i32 [[TMP97]], -1
+; VF-TWO-CHECK-NEXT: [[TMP98:%.*]] = add i32 [[OFFSET_IDX22]], 0
+; VF-TWO-CHECK-NEXT: [[TMP99:%.*]] = xor i32 [[TMP98]], -1
; VF-TWO-CHECK-NEXT: [[TMP100:%.*]] = add i32 [[TMP99]], [[N]]
; VF-TWO-CHECK-NEXT: [[TMP101:%.*]] = sext i32 [[TMP100]] to i64
; VF-TWO-CHECK-NEXT: [[TMP102:%.*]] = getelementptr inbounds float, ptr [[B]], i64 [[TMP101]]
@@ -533,7 +533,7 @@ define dso_local signext i32 @f2(ptr noalias %A, ptr noalias %B, i32 signext %n)
; VF-TWO-CHECK-NEXT: [[WIDE_LOAD23:%.*]] = load <2 x float>, ptr [[TMP104]], align 4
; VF-TWO-CHECK-NEXT: [[REVERSE24:%.*]] = shufflevector <2 x float> [[WIDE_LOAD23]], <2 x float> poison, <2 x i32> <i32 1, i32 0>
; VF-TWO-CHECK-NEXT: [[TMP105:%.*]] = fadd fast <2 x float> [[REVERSE24]], <float 1.000000e+00, float 1.000000e+00>
-; VF-TWO-CHECK-NEXT: [[TMP106:%.*]] = getelementptr inbounds float, ptr [[A]], i64 [[TMP98]]
+; VF-TWO-CHECK-NEXT: [[TMP106:%.*]] = getelementptr inbounds float, ptr [[A]], i64 [[TMP97]]
; VF-TWO-CHECK-NEXT: [[TMP107:%.*]] = getelementptr inbounds float, ptr [[TMP106]], i32 0
; VF-TWO-CHECK-NEXT: store <2 x float> [[TMP105]], ptr [[TMP107]], align 4
; VF-TWO-CHECK-NEXT: [[INDEX_NEXT25]] = add nuw i64 [[INDEX21]], 2
@@ -597,31 +597,31 @@ define dso_local signext i32 @f2(ptr noalias %A, ptr noalias %B, i32 signext %n)
; VF-FOUR-CHECK-NEXT: br label [[VECTOR_BODY:%.*]]
; VF-FOUR-CHECK: vector.body:
; VF-FOUR-CHECK-NEXT: [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; VF-FOUR-CHECK-NEXT: [[TMP8:%.*]] = add i64 [[INDEX]], 0
+; VF-FOUR-CHECK-NEXT: [[TMP9:%.*]] = add i64 [[INDEX]], 4
+; VF-FOUR-CHECK-NEXT: [[TMP10:%.*]] = add i64 [[INDEX]], 8
+; VF-FOUR-CHECK-NEXT: [[TMP11:%.*]] = add i64 [[INDEX]], 12
+; VF-FOUR-CHECK-NEXT: [[TMP12:%.*]] = add i64 [[INDEX]], 16
+; VF-FOUR-CHECK-NEXT: [[TMP13:%.*]] = add i64 [[INDEX]], 20
+; VF-FOUR-CHECK-NEXT: [[TMP14:%.*]] = add i64 [[INDEX]], 24
+; VF-FOUR-CHECK-NEXT: [[TMP15:%.*]] = add i64 [[INDEX]], 28
; VF-FOUR-CHECK-NEXT: [[OFFSET_IDX:%.*]] = trunc i64 [[INDEX]] to i32
-; VF-FOUR-CHECK-NEXT: [[TMP8:%.*]] = add i32 [[OFFSET_IDX]], 0
-; VF-FOUR-CHECK-NEXT: [[TMP9:%.*]] = add i32 [[OFFSET_IDX]], 4
-; VF-FOUR-CHECK-NEXT: [[TMP10:%.*]] = add i32 [[OFFSET_IDX]], 8
-; VF-FOUR-CHECK-NEXT: [[TMP11:%.*]] = add i32 [[OFFSET_IDX]], 12
-; VF-FOUR-CHECK-NEXT: [[TMP12:%.*]] = add i32 [[OFFSET_IDX]], 16
-; VF-FOUR-CHECK-NEXT: [[TMP13:%.*]] = add i32 [[OFFSET_IDX]], 20
-; VF-FOUR-CHECK-NEXT: [[TMP14:%.*]] = add i32 [[OFFSET_IDX]], 24
-; VF-FOUR-CHECK-NEXT: [[TMP15:%.*]] = add i32 [[OFFSET_IDX]], 28
-; VF-FOUR-CHECK-NEXT: [[TMP16:%.*]] = add i64 [[INDEX]], 0
-; VF-FOUR-CHECK-NEXT: [[TMP17:%.*]] = add i64 [[INDEX]], 4
-; VF-FOUR-CHECK-NEXT: [[TMP18:%.*]] = add i64 [[INDEX]], 8
-; VF-FOUR-CHECK-NEXT: [[TMP19:%.*]] = add i64 [[INDEX]], 12
-; VF-FOUR-CHECK-NEXT: [[TMP20:%.*]] = add i64 [[INDEX]], 16
-; VF-FOUR-CHECK-NEXT: [[TMP21:%.*]] = add i64 [[INDEX]], 20
-; VF-FOUR-CHECK-NEXT: [[TMP22:%.*]] = add i64 [[INDEX]], 24
-; VF-FOUR-CHECK-NEXT: [[TMP23:%.*]] = add i64 [[INDEX]], 28
-; VF-FOUR-CHECK-NEXT: [[TMP24:%.*]] = xor i32 [[TMP8]], -1
-; VF-FOUR-CHECK-NEXT: [[TMP25:%.*]] = xor i32 [[TMP9]], -1
-; VF-FOUR-CHECK-NEXT: [[TMP26:%.*]] = xor i32 [[TMP10]], -1
-; VF-FOUR-CHECK-NEXT: [[TMP27:%.*]] = xor i32 [[TMP11]], -1
-; VF-FOUR-CHECK-NEXT: [[TMP28:%.*]] = xor i32 [[TMP12]], -1
-; VF-FOUR-CHECK-NEXT: [[TMP29:%.*]] = xor i32 [[TMP13]], -1
-; VF-FOUR-CHECK-NEXT: [[TMP30:%.*]] = xor i32 [[TMP14]], -1
-; VF-FOUR-CHECK-NEXT: [[TMP31:%.*]] = xor i32 [[TMP15]], -1
+; VF-FOUR-CHECK-NEXT: [[TMP16:%.*]] = add i32 [[OFFSET_IDX]], 0
+; VF-FOUR-CHECK-NEXT: [[TMP17:%.*]] = add i32 [[OFFSET_IDX]], 4
+; VF-FOUR-CHECK-NEXT: [[TMP18:%.*]] = add i32 [[OFFSET_IDX]], 8
+; VF-FOUR-CHECK-NEXT: [[TMP19:%.*]] = add i32 [[OFFSET_IDX]], 12
+; VF-FOUR-CHECK-NEXT: [[TMP20:%.*]] = add i32 [[OFFSET_IDX]], 16
+; VF-FOUR-CHECK-NEXT: [[TMP21:%.*]] = add i32 [[OFFSET_IDX]], 20
+; VF-FOUR-CHECK-NEXT: [[TMP22:%.*]] = add i32 [[OFFSET_IDX]], 24
+; VF-FOUR-CHECK-NEXT: [[TMP23:%.*]] = add i32 [[OFFSET_IDX]], 28
+; VF-FOUR-CHECK-NEXT: [[TMP24:%.*]] = xor i32 [[TMP16]], -1
+; VF-FOUR-CHECK-NEXT: [[TMP25:%.*]] = xor i32 [[TMP17]], -1
+; VF-FOUR-CHECK-NEXT: [[TMP26:%.*]] = xor i32 [[TMP18]], -1
+; VF-FOUR-CHECK-NEXT: [[TMP27:%.*]] = xor i32 [[TMP19]], -1
+; VF-FOUR-CHECK-NEXT: [[TMP28:%.*]] = xor i32 [[TMP20]], -1
+; VF-FOUR-CHECK-NEXT: [[TMP29:%.*]] = xor i32 [[TMP21]], -1
+; VF-FOUR-CHECK-NEXT: [[TMP30:%.*]] = xor i32 [[TMP22]], -1
+; VF-FOUR-CHECK-NEXT: [[TMP31:%.*]] = xor i32 [[TMP23]], -1
; VF-FOUR-CHECK-NEXT: [[TMP32:%.*]] = add i32 [[TMP24]], [[N]]
; VF-FOUR-CHECK-NEXT: [[TMP33:%.*]] = add i32 [[TMP25]], [[N]]
; VF-FOUR-CHECK-NEXT: [[TMP34:%.*]] = add i32 [[TMP26]], [[N]]
@@ -686,14 +686,14 @@ define dso_local signext i32 @f2(ptr noalias %A, ptr noalias %B, i32 signext %n)
; VF-FOUR-CHECK-NEXT: [[TMP77:%.*]] = fadd fast <4 x float> [[REVERSE11]], <float 1.000000e+00, float 1.000000e+00, float 1.000000e+00, float 1.000000e+00>
; VF-FOUR-CHECK-NEXT: [[TMP78:%.*]] = fadd fast <4 x float> [[REVERSE13]], <float 1.000000e+00, float 1.000000e+00, float 1.000000e+00, float 1.000000e+00>
; VF-FOUR-CHECK-NEXT: [[TMP79:%.*]] = fadd fast <4 x float> [[REVERSE15]], <float 1.000000e+00, float 1.000000e+00, float 1.000000e+00, float 1.000000e+00>
-; VF-FOUR-CHECK-NEXT: [[TMP80:%.*]] = getelementptr inbounds float, ptr [[A]], i64 [[TMP16]]
-; VF-FOUR-CHECK-NEXT: [[TMP81:%.*]] = getelementptr inbounds float, ptr [[A]], i64 [[TMP17]]
-; VF-FOUR-CHECK-NEXT: [[TMP82:%.*]] = getelementptr inbounds float, ptr [[A]], i64 [[TMP18]]
-; VF-FOUR-CHECK-NEXT: [[TMP83:%.*]] = getelementptr inbounds float, ptr [[A]], i64 [[TMP19]]
-; VF-FOUR-CHECK-NEXT: [[TMP84:%.*]] = getelementptr inbounds float, ptr [[A]], i64 [[TMP20]]
-; VF-FOUR-CHECK-NEXT: [[TMP85:%.*]] = getelementptr inbounds float, ptr [[A]], i64 [[TMP21]]
-; VF-FOUR-CHECK-NEXT: [[TMP86:%.*]] = getelementptr inbounds float, ptr [[A]], i64 [[TMP22]]
-; VF-FOUR-CHECK-NEXT: [[TMP87:%.*]] = getelementptr inbounds float, ptr [[A]], i64 [[TMP23]]
+; VF-FOUR-CHECK-NEXT: [[TMP80:%.*]] = getelementptr inbounds float, ptr [[A]], i64 [[TMP8]]
+; VF-FOUR-CHECK-NEXT: [[TMP81:%.*]] = getelementptr inbounds float, ptr [[A]], i64 [[TMP9]]
+; VF-FOUR-CHECK-NEXT: [[TMP82:%.*]] = getelementptr inbounds float, ptr [[A]], i64 [[TMP10]]
+; VF-FOUR-CHECK-NEXT: [[TMP83:%.*]] = getelementptr inbounds float, ptr [[A]], i64 [[TMP11]]
+; VF-FOUR-CHECK-NEXT: [[TMP84:%.*]] = getelementptr inbounds float, ptr [[A]], i64 [[TMP12]]
+; VF-FOUR-CHECK-NEXT: [[TMP85:%.*]] = getelementptr inbounds float, ptr [[A]], i64 [[TMP13]]
+; VF-FOUR-CHECK-NEXT: [[TMP86:%.*]] = getelementptr inbounds float, ptr [[A]], i64 [[TMP14]]
+; VF-FOUR-CHECK-NEXT: [[TMP87:%.*]] = getelementptr inbounds float, ptr [[A]], i64 [[TMP15]]
; VF-FOUR-CHECK-NEXT: [[TMP88:%.*]] = getelementptr inbounds float, ptr [[TMP80]], i32 0
; VF-FOUR-CHECK-NEXT: [[TMP89:%.*]] = getelementptr inbounds float, ptr [[TMP80]], i32 4
; VF-FOUR-CHECK-NEXT: [[TMP90:%.*]] = getelementptr inbounds float, ptr [[TMP80]], i32 8
@@ -729,10 +729,10 @@ define dso_local signext i32 @f2(ptr noalias %A, ptr noalias %B, i32 signext %n)
; VF-FOUR-CHECK-NEXT: br label [[VEC_EPILOG_VECTOR_BODY:%.*]]
; VF-FOUR-CHECK: vec.epilog.vector.body:
; VF-FOUR-CHECK-NEXT: [[INDEX21:%.*]] = phi i64 [ [[VEC_EPILOG_RESUME_VAL]], [[VEC_EPILOG_PH]] ], [ [[INDEX_NEXT25:%.*]], [[VEC_EPILOG_VECTOR_BODY]] ]
+; VF-FOUR-CHECK-NEXT: [[TMP97:%.*]] = add i64 [[INDEX21]], 0
; VF-FOUR-CHECK-NEXT: [[OFFSET_IDX22:%.*]] = trunc i64 [[INDEX21]] to i32
-; VF-FOUR-CHECK-NEXT: [[TMP97:%.*]] = add i32 [[OFFSET_IDX22]], 0
-; VF-FOUR-CHECK-NEXT: [[TMP98:%.*]] = add i64 [[INDEX21]], 0
-; VF-FOUR-CHECK-NEXT: [[TMP99:%.*]] = xor i32 [[TMP97]], -1
+; VF-FOUR-CHECK-NEXT: [[TMP98:%.*]] = add i32 [[OFFSET_IDX22]], 0
+; VF-FOUR-CHECK-NEXT: [[TMP99:%.*]] = xor i32 [[TMP98]], -1
; VF-FOUR-CHECK-NEXT: [[TMP100:%.*]] = add i32 [[TMP99]], [[N]]
; VF-FOUR-CHECK-NEXT: [[TMP101:%.*]] = sext i32 [[TMP100]] to i64
; VF-FOUR-CHECK-NEXT: [[TMP102:%.*]] = getelementptr inbounds float, ptr [[B]], i64 [[TMP101]]
@@ -741,7 +741,7 @@ define dso_local signext i32 @f2(ptr noalias %A, ptr noalias %B, i32 signext %n)
; VF-FOUR-CHECK-NEXT: [[WIDE_LOAD23:%.*]] = load <4 x float>, ptr [[TMP104]], align 4
; VF-FOUR-CHECK-NEXT: [[REVERSE24:%.*]] = shufflevector <4 x float> [[WIDE_LOAD23]], <4 x float> poison, <4 x i32> <i32 3, i32 2, i32 1, i32 0>
; VF-FOUR-CHECK-NEXT: [[TMP105:%.*]] = fadd fast <4 x float> [[REVERSE24]], <float 1.000000e+00, float 1.000000e+00, float 1.000000e+00, float 1.000000e+00>
-; VF-FOUR-CHECK-NEXT: [[TMP106:%.*]] = getelementptr inbounds float, ptr [[A]], i64 [[TMP98]]
+; VF-FOUR-CHECK-NEXT: [[TMP106:%.*]] = getelementptr inbounds float, ptr [[A]], i64 [[TMP97]]
; VF-FOUR-CHECK-NEXT: [[TMP107:%.*]] = getelementptr inbounds float, ptr [[TMP106]], i32 0
; VF-FOUR-CHECK-NEXT: store <4 x float> [[TMP105]], ptr [[TMP107]], align 4
; VF-FOUR-CHECK-NEXT: [[INDEX_NEXT25]] = add nuw i64 [[INDEX21]], 4
More information about the llvm-commits
mailing list