[llvm] [VPlan] Use InstSimplifyFolder in tryToConstantFold (PR #141222)
via llvm-commits
llvm-commits at lists.llvm.org
Fri May 23 03:17:11 PDT 2025
llvmbot wrote:
<!--LLVM PR SUMMARY COMMENT-->
@llvm/pr-subscribers-llvm-transforms
Author: Ramkumar Ramachandra (artagnon)
<details>
<summary>Changes</summary>
For more powerful folding with operands that are not necessarily all-constant, use InstSimplifyFolder instead of TargetFolder in tryToConstantFold.
---
Patch is 109.56 KiB, truncated to 20.00 KiB below, full version: https://github.com/llvm/llvm-project/pull/141222.diff
23 Files Affected:
- (modified) llvm/lib/Transforms/Vectorize/VPlanTransforms.cpp (+5-4)
- (modified) llvm/test/Transforms/LoopVectorize/AArch64/call-costs.ll (+2-3)
- (modified) llvm/test/Transforms/LoopVectorize/AArch64/deterministic-type-shrinkage.ll (+24-24)
- (modified) llvm/test/Transforms/LoopVectorize/AArch64/loop-vectorization-factors.ll (+204-212)
- (modified) llvm/test/Transforms/LoopVectorize/AArch64/optsize_minsize.ll (+6-9)
- (modified) llvm/test/Transforms/LoopVectorize/AArch64/pr60831-sve-inv-store-crash.ll (+2-3)
- (modified) llvm/test/Transforms/LoopVectorize/AArch64/transform-narrow-interleave-to-widen-memory-remove-loop-region.ll (+5-9)
- (modified) llvm/test/Transforms/LoopVectorize/AArch64/vector-loop-backedge-elimination-epilogue.ll (+4-5)
- (modified) llvm/test/Transforms/LoopVectorize/ARM/optsize_minsize.ll (+6-9)
- (modified) llvm/test/Transforms/LoopVectorize/RISCV/low-trip-count.ll (+12-18)
- (modified) llvm/test/Transforms/LoopVectorize/RISCV/short-trip-count.ll (+4-6)
- (modified) llvm/test/Transforms/LoopVectorize/X86/pr34438.ll (+4-6)
- (modified) llvm/test/Transforms/LoopVectorize/X86/replicate-recipe-with-only-first-lane-used.ll (+1-2)
- (modified) llvm/test/Transforms/LoopVectorize/X86/replicate-uniform-call.ll (+2-8)
- (modified) llvm/test/Transforms/LoopVectorize/X86/scev-checks-unprofitable.ll (+1-2)
- (modified) llvm/test/Transforms/LoopVectorize/constantfolder.ll (+2-4)
- (modified) llvm/test/Transforms/LoopVectorize/skeleton-lcssa-crash.ll (+1-2)
- (modified) llvm/test/Transforms/LoopVectorize/vector-loop-backedge-elimination-early-exit.ll (+4-8)
- (modified) llvm/test/Transforms/LoopVectorize/vector-loop-backedge-elimination-outside-iv-users.ll (+8-12)
- (modified) llvm/test/Transforms/LoopVectorize/vector-loop-backedge-elimination.ll (+15-21)
- (modified) llvm/test/Transforms/LoopVectorize/version-stride-with-integer-casts.ll (+1-2)
- (modified) llvm/test/Transforms/LoopVectorize/vplan-printing-before-execute.ll (+4-5)
- (modified) llvm/test/Transforms/LoopVectorize/vplan-sink-scalars-and-merge.ll (+1-2)
``````````diff
diff --git a/llvm/lib/Transforms/Vectorize/VPlanTransforms.cpp b/llvm/lib/Transforms/Vectorize/VPlanTransforms.cpp
index 34633cd748eb1..2c20c4a95fd96 100644
--- a/llvm/lib/Transforms/Vectorize/VPlanTransforms.cpp
+++ b/llvm/lib/Transforms/Vectorize/VPlanTransforms.cpp
@@ -27,8 +27,8 @@
#include "llvm/ADT/SetVector.h"
#include "llvm/ADT/TypeSwitch.h"
#include "llvm/Analysis/IVDescriptors.h"
+#include "llvm/Analysis/InstSimplifyFolder.h"
#include "llvm/Analysis/LoopInfo.h"
-#include "llvm/Analysis/TargetFolder.h"
#include "llvm/Analysis/VectorUtils.h"
#include "llvm/IR/Intrinsics.h"
#include "llvm/IR/PatternMatch.h"
@@ -939,8 +939,9 @@ static void recursivelyDeleteDeadRecipes(VPValue *V) {
}
}
-/// Try to fold \p R using TargetFolder to a constant. Will succeed and return a
-/// non-nullptr Value for a handled \p Opcode if all \p Operands are constant.
+/// Try to fold \p R using InstSimplifyFolder to a constant. Will succeed and
+/// return a non-nullptr Value for a handled \p Opcode if \p Operands are
+/// foldable.
static Value *tryToConstantFold(const VPRecipeBase &R, unsigned Opcode,
ArrayRef<VPValue *> Operands,
const DataLayout &DL,
@@ -952,7 +953,7 @@ static Value *tryToConstantFold(const VPRecipeBase &R, unsigned Opcode,
Ops.push_back(Op->getLiveInIRValue());
}
- TargetFolder Folder(DL);
+ InstSimplifyFolder Folder(DL);
if (Instruction::isBinaryOp(Opcode))
return Folder.FoldBinOp(static_cast<Instruction::BinaryOps>(Opcode), Ops[0],
Ops[1]);
diff --git a/llvm/test/Transforms/LoopVectorize/AArch64/call-costs.ll b/llvm/test/Transforms/LoopVectorize/AArch64/call-costs.ll
index 1c91b4956b309..8c2a48aa38695 100644
--- a/llvm/test/Transforms/LoopVectorize/AArch64/call-costs.ll
+++ b/llvm/test/Transforms/LoopVectorize/AArch64/call-costs.ll
@@ -79,11 +79,10 @@ define void @powi_call(ptr %P) {
; CHECK: [[VECTOR_PH]]:
; CHECK-NEXT: br label %[[VECTOR_BODY:.*]]
; CHECK: [[VECTOR_BODY]]:
-; CHECK-NEXT: [[TMP1:%.*]] = getelementptr inbounds double, ptr [[P]], i64 0
-; CHECK-NEXT: [[TMP2:%.*]] = getelementptr inbounds double, ptr [[TMP1]], i32 0
+; CHECK-NEXT: [[TMP2:%.*]] = getelementptr inbounds double, ptr [[P]], i32 0
; CHECK-NEXT: [[WIDE_LOAD:%.*]] = load <2 x double>, ptr [[TMP2]], align 8
; CHECK-NEXT: [[TMP3:%.*]] = call <2 x double> @llvm.powi.v2f64.i32(<2 x double> [[WIDE_LOAD]], i32 3)
-; CHECK-NEXT: [[TMP4:%.*]] = getelementptr inbounds double, ptr [[TMP1]], i32 0
+; CHECK-NEXT: [[TMP4:%.*]] = getelementptr inbounds double, ptr [[P]], i32 0
; CHECK-NEXT: store <2 x double> [[TMP3]], ptr [[TMP4]], align 8
; CHECK-NEXT: br label %[[MIDDLE_BLOCK:.*]]
; CHECK: [[MIDDLE_BLOCK]]:
diff --git a/llvm/test/Transforms/LoopVectorize/AArch64/deterministic-type-shrinkage.ll b/llvm/test/Transforms/LoopVectorize/AArch64/deterministic-type-shrinkage.ll
index c2502aac5b61d..4c046cd7da0f6 100644
--- a/llvm/test/Transforms/LoopVectorize/AArch64/deterministic-type-shrinkage.ll
+++ b/llvm/test/Transforms/LoopVectorize/AArch64/deterministic-type-shrinkage.ll
@@ -160,45 +160,45 @@ define void @test_shrink_zext_in_preheader(ptr noalias %src, ptr noalias %dst, i
; CHECK: vector.main.loop.iter.check:
; CHECK-NEXT: br i1 false, label [[VEC_EPILOG_PH:%.*]], label [[VECTOR_PH:%.*]]
; CHECK: vector.ph:
+; CHECK-NEXT: [[BROADCAST_SPLATINSERT:%.*]] = insertelement <16 x i16> poison, i16 [[B]], i64 0
; CHECK-NEXT: [[TMP0:%.*]] = trunc i32 [[A]] to i16
; CHECK-NEXT: [[TMP1:%.*]] = insertelement <16 x i16> poison, i16 [[TMP0]], i64 0
-; CHECK-NEXT: [[TMP2:%.*]] = insertelement <16 x i16> poison, i16 [[B]], i64 0
-; CHECK-NEXT: [[TMP3:%.*]] = mul <16 x i16> [[TMP1]], [[TMP2]]
-; CHECK-NEXT: [[TMP4:%.*]] = lshr <16 x i16> [[TMP3]], <i16 8, i16 0, i16 0, i16 0, i16 0, i16 0, i16 0, i16 0, i16 0, i16 0, i16 0, i16 0, i16 0, i16 0, i16 0, i16 0>
-; CHECK-NEXT: [[TMP5:%.*]] = trunc <16 x i16> [[TMP4]] to <16 x i8>
-; CHECK-NEXT: [[TMP6:%.*]] = shufflevector <16 x i8> [[TMP5]], <16 x i8> poison, <16 x i32> zeroinitializer
+; CHECK-NEXT: [[TMP2:%.*]] = mul <16 x i16> [[TMP1]], [[BROADCAST_SPLATINSERT]]
+; CHECK-NEXT: [[TMP3:%.*]] = lshr <16 x i16> [[TMP2]], <i16 8, i16 0, i16 0, i16 0, i16 0, i16 0, i16 0, i16 0, i16 0, i16 0, i16 0, i16 0, i16 0, i16 0, i16 0, i16 0>
+; CHECK-NEXT: [[TMP4:%.*]] = trunc <16 x i16> [[TMP3]] to <16 x i8>
+; CHECK-NEXT: [[TMP5:%.*]] = shufflevector <16 x i8> [[TMP4]], <16 x i8> poison, <16 x i32> zeroinitializer
; CHECK-NEXT: br label [[VECTOR_BODY:%.*]]
; CHECK: vector.body:
; CHECK-NEXT: [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
-; CHECK-NEXT: [[TMP7:%.*]] = sext i32 [[INDEX]] to i64
-; CHECK-NEXT: [[TMP8:%.*]] = getelementptr inbounds i8, ptr [[DST]], i64 [[TMP7]]
-; CHECK-NEXT: [[TMP9:%.*]] = getelementptr inbounds nuw i8, ptr [[TMP8]], i64 16
-; CHECK-NEXT: store <16 x i8> [[TMP6]], ptr [[TMP8]], align 1
-; CHECK-NEXT: store <16 x i8> [[TMP6]], ptr [[TMP9]], align 1
+; CHECK-NEXT: [[TMP6:%.*]] = sext i32 [[INDEX]] to i64
+; CHECK-NEXT: [[TMP7:%.*]] = getelementptr inbounds i8, ptr [[DST]], i64 [[TMP6]]
+; CHECK-NEXT: [[TMP8:%.*]] = getelementptr inbounds nuw i8, ptr [[TMP7]], i64 16
+; CHECK-NEXT: store <16 x i8> [[TMP5]], ptr [[TMP7]], align 1
+; CHECK-NEXT: store <16 x i8> [[TMP5]], ptr [[TMP8]], align 1
; CHECK-NEXT: [[INDEX_NEXT]] = add nuw i32 [[INDEX]], 32
-; CHECK-NEXT: [[TMP10:%.*]] = icmp eq i32 [[INDEX_NEXT]], 992
-; CHECK-NEXT: br i1 [[TMP10]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP5:![0-9]+]]
+; CHECK-NEXT: [[TMP9:%.*]] = icmp eq i32 [[INDEX_NEXT]], 992
+; CHECK-NEXT: br i1 [[TMP9]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP5:![0-9]+]]
; CHECK: middle.block:
; CHECK-NEXT: br i1 false, label [[EXIT:%.*]], label [[VEC_EPILOG_ITER_CHECK:%.*]]
; CHECK: vec.epilog.iter.check:
; CHECK-NEXT: br i1 false, label [[VEC_EPILOG_SCALAR_PH]], label [[VEC_EPILOG_PH]]
; CHECK: vec.epilog.ph:
-; CHECK-NEXT: [[TMP11:%.*]] = trunc i32 [[A]] to i16
-; CHECK-NEXT: [[TMP12:%.*]] = insertelement <8 x i16> poison, i16 [[TMP11]], i64 0
-; CHECK-NEXT: [[TMP13:%.*]] = insertelement <8 x i16> poison, i16 [[B]], i64 0
-; CHECK-NEXT: [[TMP14:%.*]] = mul <8 x i16> [[TMP12]], [[TMP13]]
-; CHECK-NEXT: [[TMP15:%.*]] = lshr <8 x i16> [[TMP14]], <i16 8, i16 0, i16 0, i16 0, i16 0, i16 0, i16 0, i16 0>
-; CHECK-NEXT: [[TMP16:%.*]] = trunc <8 x i16> [[TMP15]] to <8 x i8>
-; CHECK-NEXT: [[TMP17:%.*]] = shufflevector <8 x i8> [[TMP16]], <8 x i8> poison, <8 x i32> zeroinitializer
+; CHECK-NEXT: [[BROADCAST_SPLATINSERT3:%.*]] = insertelement <8 x i16> poison, i16 [[B]], i64 0
+; CHECK-NEXT: [[TMP10:%.*]] = trunc i32 [[A]] to i16
+; CHECK-NEXT: [[TMP11:%.*]] = insertelement <8 x i16> poison, i16 [[TMP10]], i64 0
+; CHECK-NEXT: [[TMP12:%.*]] = mul <8 x i16> [[TMP11]], [[BROADCAST_SPLATINSERT3]]
+; CHECK-NEXT: [[TMP13:%.*]] = lshr <8 x i16> [[TMP12]], <i16 8, i16 0, i16 0, i16 0, i16 0, i16 0, i16 0, i16 0>
+; CHECK-NEXT: [[TMP14:%.*]] = trunc <8 x i16> [[TMP13]] to <8 x i8>
+; CHECK-NEXT: [[TMP15:%.*]] = shufflevector <8 x i8> [[TMP14]], <8 x i8> poison, <8 x i32> zeroinitializer
; CHECK-NEXT: br label [[VEC_EPILOG_VECTOR_BODY:%.*]]
; CHECK: vec.epilog.vector.body:
; CHECK-NEXT: [[INDEX7:%.*]] = phi i32 [ 992, [[VEC_EPILOG_PH]] ], [ [[INDEX_NEXT8:%.*]], [[VEC_EPILOG_VECTOR_BODY]] ]
-; CHECK-NEXT: [[TMP18:%.*]] = sext i32 [[INDEX7]] to i64
-; CHECK-NEXT: [[TMP19:%.*]] = getelementptr inbounds i8, ptr [[DST]], i64 [[TMP18]]
-; CHECK-NEXT: store <8 x i8> [[TMP17]], ptr [[TMP19]], align 1
+; CHECK-NEXT: [[TMP16:%.*]] = sext i32 [[INDEX7]] to i64
+; CHECK-NEXT: [[TMP17:%.*]] = getelementptr inbounds i8, ptr [[DST]], i64 [[TMP16]]
+; CHECK-NEXT: store <8 x i8> [[TMP15]], ptr [[TMP17]], align 1
; CHECK-NEXT: [[INDEX_NEXT8]] = add nuw i32 [[INDEX7]], 8
-; CHECK-NEXT: [[TMP20:%.*]] = icmp eq i32 [[INDEX_NEXT8]], 1000
-; CHECK-NEXT: br i1 [[TMP20]], label [[VEC_EPILOG_MIDDLE_BLOCK:%.*]], label [[VEC_EPILOG_VECTOR_BODY]], !llvm.loop [[LOOP6:![0-9]+]]
+; CHECK-NEXT: [[TMP18:%.*]] = icmp eq i32 [[INDEX_NEXT8]], 1000
+; CHECK-NEXT: br i1 [[TMP18]], label [[VEC_EPILOG_MIDDLE_BLOCK:%.*]], label [[VEC_EPILOG_VECTOR_BODY]], !llvm.loop [[LOOP6:![0-9]+]]
; CHECK: vec.epilog.middle.block:
; CHECK-NEXT: br i1 true, label [[EXIT]], label [[VEC_EPILOG_SCALAR_PH]]
; CHECK: vec.epilog.scalar.ph:
diff --git a/llvm/test/Transforms/LoopVectorize/AArch64/loop-vectorization-factors.ll b/llvm/test/Transforms/LoopVectorize/AArch64/loop-vectorization-factors.ll
index cc972d3040adb..e97bb857fdba3 100644
--- a/llvm/test/Transforms/LoopVectorize/AArch64/loop-vectorization-factors.ll
+++ b/llvm/test/Transforms/LoopVectorize/AArch64/loop-vectorization-factors.ll
@@ -23,16 +23,16 @@ define void @add_a(ptr noalias nocapture readonly %p, ptr noalias nocapture %q,
; CHECK-NEXT: br label [[VECTOR_BODY:%.*]]
; CHECK: vector.body:
; CHECK-NEXT: [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
-; CHECK-NEXT: [[TMP2:%.*]] = getelementptr inbounds i8, ptr [[P]], i64 [[INDEX]]
-; CHECK-NEXT: [[TMP3:%.*]] = getelementptr inbounds i8, ptr [[TMP2]], i32 0
-; CHECK-NEXT: [[WIDE_LOAD:%.*]] = load <16 x i8>, ptr [[TMP3]], align 1
-; CHECK-NEXT: [[TMP4:%.*]] = add <16 x i8> [[WIDE_LOAD]], splat (i8 2)
-; CHECK-NEXT: [[TMP5:%.*]] = getelementptr inbounds i8, ptr [[Q]], i64 [[INDEX]]
-; CHECK-NEXT: [[TMP6:%.*]] = getelementptr inbounds i8, ptr [[TMP5]], i32 0
-; CHECK-NEXT: store <16 x i8> [[TMP4]], ptr [[TMP6]], align 1
+; CHECK-NEXT: [[TMP1:%.*]] = getelementptr inbounds i8, ptr [[P]], i64 [[INDEX]]
+; CHECK-NEXT: [[TMP2:%.*]] = getelementptr inbounds i8, ptr [[TMP1]], i32 0
+; CHECK-NEXT: [[WIDE_LOAD:%.*]] = load <16 x i8>, ptr [[TMP2]], align 1
+; CHECK-NEXT: [[TMP3:%.*]] = add <16 x i8> [[WIDE_LOAD]], splat (i8 2)
+; CHECK-NEXT: [[TMP4:%.*]] = getelementptr inbounds i8, ptr [[Q]], i64 [[INDEX]]
+; CHECK-NEXT: [[TMP5:%.*]] = getelementptr inbounds i8, ptr [[TMP4]], i32 0
+; CHECK-NEXT: store <16 x i8> [[TMP3]], ptr [[TMP5]], align 1
; CHECK-NEXT: [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 16
-; CHECK-NEXT: [[TMP7:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
-; CHECK-NEXT: br i1 [[TMP7]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
+; CHECK-NEXT: [[TMP6:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
+; CHECK-NEXT: br i1 [[TMP6]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
; CHECK: middle.block:
; CHECK-NEXT: [[CMP_N:%.*]] = icmp eq i64 [[TMP0]], [[N_VEC]]
; CHECK-NEXT: br i1 [[CMP_N]], label [[FOR_COND_CLEANUP_LOOPEXIT:%.*]], label [[VEC_EPILOG_ITER_CHECK:%.*]]
@@ -47,16 +47,16 @@ define void @add_a(ptr noalias nocapture readonly %p, ptr noalias nocapture %q,
; CHECK-NEXT: br label [[VEC_EPILOG_VECTOR_BODY:%.*]]
; CHECK: vec.epilog.vector.body:
; CHECK-NEXT: [[INDEX4:%.*]] = phi i64 [ [[VEC_EPILOG_RESUME_VAL]], [[VEC_EPILOG_PH]] ], [ [[INDEX_NEXT6:%.*]], [[VEC_EPILOG_VECTOR_BODY]] ]
-; CHECK-NEXT: [[TMP9:%.*]] = getelementptr inbounds i8, ptr [[P]], i64 [[INDEX4]]
-; CHECK-NEXT: [[TMP10:%.*]] = getelementptr inbounds i8, ptr [[TMP9]], i32 0
-; CHECK-NEXT: [[WIDE_LOAD5:%.*]] = load <4 x i8>, ptr [[TMP10]], align 1
-; CHECK-NEXT: [[TMP11:%.*]] = add <4 x i8> [[WIDE_LOAD5]], splat (i8 2)
-; CHECK-NEXT: [[TMP12:%.*]] = getelementptr inbounds i8, ptr [[Q]], i64 [[INDEX4]]
-; CHECK-NEXT: [[TMP13:%.*]] = getelementptr inbounds i8, ptr [[TMP12]], i32 0
-; CHECK-NEXT: store <4 x i8> [[TMP11]], ptr [[TMP13]], align 1
+; CHECK-NEXT: [[TMP7:%.*]] = getelementptr inbounds i8, ptr [[P]], i64 [[INDEX4]]
+; CHECK-NEXT: [[TMP8:%.*]] = getelementptr inbounds i8, ptr [[TMP7]], i32 0
+; CHECK-NEXT: [[WIDE_LOAD5:%.*]] = load <4 x i8>, ptr [[TMP8]], align 1
+; CHECK-NEXT: [[TMP9:%.*]] = add <4 x i8> [[WIDE_LOAD5]], splat (i8 2)
+; CHECK-NEXT: [[TMP10:%.*]] = getelementptr inbounds i8, ptr [[Q]], i64 [[INDEX4]]
+; CHECK-NEXT: [[TMP11:%.*]] = getelementptr inbounds i8, ptr [[TMP10]], i32 0
+; CHECK-NEXT: store <4 x i8> [[TMP9]], ptr [[TMP11]], align 1
; CHECK-NEXT: [[INDEX_NEXT6]] = add nuw i64 [[INDEX4]], 4
-; CHECK-NEXT: [[TMP14:%.*]] = icmp eq i64 [[INDEX_NEXT6]], [[N_VEC3]]
-; CHECK-NEXT: br i1 [[TMP14]], label [[VEC_EPILOG_MIDDLE_BLOCK:%.*]], label [[VEC_EPILOG_VECTOR_BODY]], !llvm.loop [[LOOP3:![0-9]+]]
+; CHECK-NEXT: [[TMP12:%.*]] = icmp eq i64 [[INDEX_NEXT6]], [[N_VEC3]]
+; CHECK-NEXT: br i1 [[TMP12]], label [[VEC_EPILOG_MIDDLE_BLOCK:%.*]], label [[VEC_EPILOG_VECTOR_BODY]], !llvm.loop [[LOOP3:![0-9]+]]
; CHECK: vec.epilog.middle.block:
; CHECK-NEXT: [[CMP_N7:%.*]] = icmp eq i64 [[TMP0]], [[N_VEC3]]
; CHECK-NEXT: br i1 [[CMP_N7]], label [[FOR_COND_CLEANUP_LOOPEXIT]], label [[VEC_EPILOG_SCALAR_PH]]
@@ -70,8 +70,8 @@ define void @add_a(ptr noalias nocapture readonly %p, ptr noalias nocapture %q,
; CHECK: for.body:
; CHECK-NEXT: [[INDVARS_IV:%.*]] = phi i64 [ [[INDVARS_IV_NEXT:%.*]], [[FOR_BODY]] ], [ [[BC_RESUME_VAL]], [[VEC_EPILOG_SCALAR_PH]] ]
; CHECK-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds i8, ptr [[P]], i64 [[INDVARS_IV]]
-; CHECK-NEXT: [[TMP15:%.*]] = load i8, ptr [[ARRAYIDX]], align 1
-; CHECK-NEXT: [[CONV:%.*]] = zext i8 [[TMP15]] to i32
+; CHECK-NEXT: [[TMP13:%.*]] = load i8, ptr [[ARRAYIDX]], align 1
+; CHECK-NEXT: [[CONV:%.*]] = zext i8 [[TMP13]] to i32
; CHECK-NEXT: [[ADD:%.*]] = add nuw nsw i32 [[CONV]], 2
; CHECK-NEXT: [[CONV1:%.*]] = trunc i32 [[ADD]] to i8
; CHECK-NEXT: [[ARRAYIDX3:%.*]] = getelementptr inbounds i8, ptr [[Q]], i64 [[INDVARS_IV]]
@@ -124,16 +124,16 @@ define void @add_a1(ptr noalias nocapture readonly %p, ptr noalias nocapture %q,
; CHECK-NEXT: br label [[VECTOR_BODY:%.*]]
; CHECK: vector.body:
; CHECK-NEXT: [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
-; CHECK-NEXT: [[TMP2:%.*]] = getelementptr inbounds i8, ptr [[P]], i64 [[INDEX]]
-; CHECK-NEXT: [[TMP3:%.*]] = getelementptr inbounds i8, ptr [[TMP2]], i32 0
-; CHECK-NEXT: [[WIDE_LOAD:%.*]] = load <16 x i8>, ptr [[TMP3]], align 1
-; CHECK-NEXT: [[TMP4:%.*]] = add nuw nsw <16 x i8> [[WIDE_LOAD]], splat (i8 2)
-; CHECK-NEXT: [[TMP5:%.*]] = getelementptr inbounds i8, ptr [[Q]], i64 [[INDEX]]
-; CHECK-NEXT: [[TMP6:%.*]] = getelementptr inbounds i8, ptr [[TMP5]], i32 0
-; CHECK-NEXT: store <16 x i8> [[TMP4]], ptr [[TMP6]], align 1
+; CHECK-NEXT: [[TMP1:%.*]] = getelementptr inbounds i8, ptr [[P]], i64 [[INDEX]]
+; CHECK-NEXT: [[TMP2:%.*]] = getelementptr inbounds i8, ptr [[TMP1]], i32 0
+; CHECK-NEXT: [[WIDE_LOAD:%.*]] = load <16 x i8>, ptr [[TMP2]], align 1
+; CHECK-NEXT: [[TMP3:%.*]] = add nuw nsw <16 x i8> [[WIDE_LOAD]], splat (i8 2)
+; CHECK-NEXT: [[TMP4:%.*]] = getelementptr inbounds i8, ptr [[Q]], i64 [[INDEX]]
+; CHECK-NEXT: [[TMP5:%.*]] = getelementptr inbounds i8, ptr [[TMP4]], i32 0
+; CHECK-NEXT: store <16 x i8> [[TMP3]], ptr [[TMP5]], align 1
; CHECK-NEXT: [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 16
-; CHECK-NEXT: [[TMP7:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
-; CHECK-NEXT: br i1 [[TMP7]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP5:![0-9]+]]
+; CHECK-NEXT: [[TMP6:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
+; CHECK-NEXT: br i1 [[TMP6]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP5:![0-9]+]]
; CHECK: middle.block:
; CHECK-NEXT: [[CMP_N:%.*]] = icmp eq i64 [[TMP0]], [[N_VEC]]
; CHECK-NEXT: br i1 [[CMP_N]], label [[FOR_COND_CLEANUP_LOOPEXIT:%.*]], label [[VEC_EPILOG_ITER_CHECK:%.*]]
@@ -148,16 +148,16 @@ define void @add_a1(ptr noalias nocapture readonly %p, ptr noalias nocapture %q,
; CHECK-NEXT: br label [[VEC_EPILOG_VECTOR_BODY:%.*]]
; CHECK: vec.epilog.vector.body:
; CHECK-NEXT: [[INDEX4:%.*]] = phi i64 [ [[VEC_EPILOG_RESUME_VAL]], [[VEC_EPILOG_PH]] ], [ [[INDEX_NEXT6:%.*]], [[VEC_EPILOG_VECTOR_BODY]] ]
-; CHECK-NEXT: [[TMP9:%.*]] = getelementptr inbounds i8, ptr [[P]], i64 [[INDEX4]]
-; CHECK-NEXT: [[TMP10:%.*]] = getelementptr inbounds i8, ptr [[TMP9]], i32 0
-; CHECK-NEXT: [[WIDE_LOAD5:%.*]] = load <4 x i8>, ptr [[TMP10]], align 1
-; CHECK-NEXT: [[TMP11:%.*]] = add nuw nsw <4 x i8> [[WIDE_LOAD5]], splat (i8 2)
-; CHECK-NEXT: [[TMP12:%.*]] = getelementptr inbounds i8, ptr [[Q]], i64 [[INDEX4]]
-; CHECK-NEXT: [[TMP13:%.*]] = getelementptr inbounds i8, ptr [[TMP12]], i32 0
-; CHECK-NEXT: store <4 x i8> [[TMP11]], ptr [[TMP13]], align 1
+; CHECK-NEXT: [[TMP7:%.*]] = getelementptr inbounds i8, ptr [[P]], i64 [[INDEX4]]
+; CHECK-NEXT: [[TMP8:%.*]] = getelementptr inbounds i8, ptr [[TMP7]], i32 0
+; CHECK-NEXT: [[WIDE_LOAD5:%.*]] = load <4 x i8>, ptr [[TMP8]], align 1
+; CHECK-NEXT: [[TMP9:%.*]] = add nuw nsw <4 x i8> [[WIDE_LOAD5]], splat (i8 2)
+; CHECK-NEXT: [[TMP10:%.*]] = getelementptr inbounds i8, ptr [[Q]], i64 [[INDEX4]]
+; CHECK-NEXT: [[TMP11:%.*]] = getelementptr inbounds i8, ptr [[TMP10]], i32 0
+; CHECK-NEXT: store <4 x i8> [[TMP9]], ptr [[TMP11]], align 1
; CHECK-NEXT: [[INDEX_NEXT6]] = add nuw i64 [[INDEX4]], 4
-; CHECK-NEXT: [[TMP14:%.*]] = icmp eq i64 [[INDEX_NEXT6]], [[N_VEC3]]
-; CHECK-NEXT: br i1 [[TMP14]], label [[VEC_EPILOG_MIDDLE_BLOCK:%.*]], label [[VEC_EPILOG_VECTOR_BODY]], !llvm.loop [[LOOP6:![0-9]+]]
+; CHECK-NEXT: [[TMP12:%.*]] = icmp eq i64 [[INDEX_NEXT6]], [[N_VEC3]]
+; CHECK-NEXT: br i1 [[TMP12]], label [[VEC_EPILOG_MIDDLE_BLOCK:%.*]], label [[VEC_EPILOG_VECTOR_BODY]], !llvm.loop [[LOOP6:![0-9]+]]
; CHECK: vec.epilog.middle.block:
; CHECK-NEXT: [[CMP_N7:%.*]] = icmp eq i64 [[TMP0]], [[N_VEC3]]
; CHECK-NEXT: br i1 [[CMP_N7]], label [[FOR_COND_CLEANUP_LOOPEXIT]], label [[VEC_EPILOG_SCALAR_PH]]
@@ -171,8 +171,8 @@ define void @add_a1(ptr noalias nocapture readonly %p, ptr noalias nocapture %q,
; CHECK: for.body:
; CHECK-NEXT: [[INDVARS_IV:%.*]] = phi i64 [ [[INDVARS_IV_NEXT:%.*]], [[FOR_BODY]] ], [ [[BC_RESUME_VAL]], [[VEC_EPILOG_SCALAR_PH]] ]
; CHECK-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds i8, ptr [[P]], i64 [[INDVARS_IV]]
-; CHECK-NEXT: [[TMP15:%.*]] = load i8, ptr [[ARRAYIDX]], align 1
-; CHECK-NEXT: [[ADD:%.*]] = add nuw nsw i8 [[TMP15]], 2
+; CHECK-NEXT: [[TMP13:%.*]] = load i8, ptr [[ARRAYIDX]], align 1
+; CHECK-NEXT: [[ADD:%.*]] = add nuw nsw i8 [[TMP13]], 2
; CHECK-NEXT: [[ARRAYIDX3:%.*]] = getelementptr inbounds i8, ptr [[Q]], i64 [[INDVARS_IV]]
; CHECK-NEXT: store i8 [[ADD]], ptr [[ARRAYIDX3]], align 1
; CHECK-NEXT: [[INDVARS_IV_NEXT]] = add nuw nsw i64 [[INDVARS_IV]], 1
@@ -216,16 +216,16 @@ define void @add_b(ptr noalias nocapture readonly %p, ptr noalias nocapture %q,
; CHECK-NEXT: br label [[VECTOR_BODY:%.*]]
; CHECK: vector.body:
; CHECK-NEXT: [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
-; CHECK-NEXT: [[TMP2:%.*]] = getelementptr inbounds i16, ptr [[P]], i64 [[INDEX]]
-; CHECK-NEXT: [[TMP3:%.*]] = getelementptr inbounds i16, ptr [[TMP2]], i32 0
-; CHECK-NEXT: [[WIDE_LOAD:%.*]] = load <8 x i16>, ptr [[TMP3]], align 2
-; CHECK-NEXT: [[TMP4:%.*]] = add <8 x i16> [[WIDE_LOAD]], splat (i16 2)
-; CHECK-NEXT: [[TMP5:%.*]] = getelementptr inbounds i16, ptr [[Q]], i64 [[INDEX]]
-; CHECK-NEXT: [[TMP6:%.*]] = getelementptr inbounds i16, ptr [[TMP5]], i32 0
-; CHECK-NEXT: store <8 x i16> [[TMP4]], ptr [[TMP6]], align 2
+; CHECK-NEXT: [[TMP1:%.*]] = getelementptr inbounds i16, ptr [[P]], i64 [[INDEX]]
+; CHECK-NEXT: [[TMP2:%.*]] = getelementptr inbounds i16, ptr [[TMP1]], i32 0
+; CHECK-NEXT: [[WIDE_LOAD:%.*]] = load <8 x i16>, ptr [[TMP2]], align 2
+; CHECK-NEXT: [[TMP3:%.*]] = add <8 x i16> [[WIDE_LOAD]], splat (i16 2)
+; CHECK-NEXT: [[TMP4:%.*]] = getelementptr inbounds i16, ptr [[Q]], i64 [[INDEX]]
+; CHECK-NEXT: [[TMP5:%.*]] = getelementptr inbounds i16, ptr [[TMP4]], i32 0
+; CHECK-NEXT: store <8 x i16> [...
[truncated]
``````````
</details>
https://github.com/llvm/llvm-project/pull/141222
More information about the llvm-commits
mailing list