Skip to content

[VPlan] Handle canonical VPWidenIntOrFpInduction in branch-condition simplification #153539

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 5 commits into from
Aug 20, 2025
Merged

[VPlan] Handle canonical VPWidenIntOrFpInduction in branch-condition simplification #153539

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
e5374bd
[VPlan] optimize BranchCond for canonical WidenIntOrFpInduction recipe
arcbbb Aug 14, 2025
851cf77
Add TODO
arcbbb Aug 15, 2025
1260809
Update TODO
arcbbb Aug 19, 2025
a13305c
Merge remote-tracking branch 'pub/main' into refine-branch-on-true
arcbbb Aug 19, 2025
d90c9ec
Update tests after rebase
arcbbb Aug 19, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
20 changes: 17 additions & 3 deletions llvm/lib/Transforms/Vectorize/VPlanTransforms.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1492,11 +1492,25 @@ static bool simplifyBranchConditionForVFAndUF(VPlan &Plan, ElementCount BestVF,
// The vector loop region only executes once. If possible, completely remove
// the region, otherwise replace the terminator controlling the latch with
// (BranchOnCond true).
// TODO: VPWidenIntOrFpInductionRecipe is only partially supported; add
// support for other non-canonical widen induction recipes (e.g.,
// VPWidenPointerInductionRecipe).
auto *Header = cast<VPBasicBlock>(VectorRegion->getEntry());
if (all_of(Header->phis(),
IsaPred<VPCanonicalIVPHIRecipe, VPEVLBasedIVPHIRecipe,
VPFirstOrderRecurrencePHIRecipe, VPPhi>)) {
if (all_of(Header->phis(), [](VPRecipeBase &Phi) {
if (auto *R = dyn_cast<VPWidenIntOrFpInductionRecipe>(&Phi))
return R->isCanonical();
return isa<VPCanonicalIVPHIRecipe, VPEVLBasedIVPHIRecipe,
VPFirstOrderRecurrencePHIRecipe, VPPhi>(&Phi);
})) {
for (VPRecipeBase &HeaderR : make_early_inc_range(Header->phis())) {
if (auto *R = dyn_cast<VPWidenIntOrFpInductionRecipe>(&HeaderR)) {
VPBuilder Builder(Plan.getVectorPreheader());
VPValue *StepV = Builder.createNaryOp(VPInstruction::StepVector, {},
R->getScalarType());
HeaderR.getVPSingleValue()->replaceAllUsesWith(StepV);
HeaderR.eraseFromParent();
continue;
}
auto *Phi = cast<VPPhiAccessors>(&HeaderR);
HeaderR.getVPSingleValue()->replaceAllUsesWith(Phi->getIncomingValue(0));
HeaderR.eraseFromParent();
Expand Down
250 changes: 65 additions & 185 deletions llvm/test/Transforms/LoopVectorize/AArch64/conditional-branches-cost.ll
View file
Open in desktop

Large diffs are not rendered by default.

35 changes: 11 additions & 24 deletions ...s/LoopVectorize/AArch64/transform-narrow-interleave-to-widen-memory-remove-loop-region.ll
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -47,14 +47,9 @@ define void @load_store_interleave_group_tc_2(ptr noalias %data) {
; VF4: [[VECTOR_PH]]:
; VF4-NEXT: br label %[[VECTOR_BODY:.*]]
; VF4: [[VECTOR_BODY]]:
; VF4-NEXT: [[INDEX:%.*]] = phi i64 [ 0, %[[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], %[[PRED_STORE_CONTINUE6:.*]] ]
; VF4-NEXT: [[VEC_IND:%.*]] = phi <4 x i8> [ <i8 0, i8 1, i8 2, i8 3>, %[[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], %[[PRED_STORE_CONTINUE6]] ]
; VF4-NEXT: [[TMP0:%.*]] = icmp ule <4 x i8> [[VEC_IND]], splat (i8 1)
; VF4-NEXT: [[TMP1:%.*]] = extractelement <4 x i1> [[TMP0]], i32 0
; VF4-NEXT: br i1 [[TMP1]], label %[[PRED_STORE_IF:.*]], label %[[PRED_STORE_CONTINUE:.*]]
; VF4-NEXT: br i1 true, label %[[PRED_STORE_IF:.*]], label %[[PRED_STORE_CONTINUE:.*]]
; VF4: [[PRED_STORE_IF]]:
; VF4-NEXT: [[TMP2:%.*]] = add i64 [[INDEX]], 0
; VF4-NEXT: [[TMP3:%.*]] = shl nsw i64 [[TMP2]], 1
; VF4-NEXT: [[TMP3:%.*]] = shl nsw i64 0, 1
; VF4-NEXT: [[TMP4:%.*]] = getelementptr inbounds i64, ptr [[DATA]], i64 [[TMP3]]
; VF4-NEXT: [[TMP5:%.*]] = load i64, ptr [[TMP4]], align 8
; VF4-NEXT: store i64 [[TMP5]], ptr [[TMP4]], align 8
Expand All @@ -64,11 +59,9 @@ define void @load_store_interleave_group_tc_2(ptr noalias %data) {
; VF4-NEXT: store i64 [[TMP8]], ptr [[TMP7]], align 8
; VF4-NEXT: br label %[[PRED_STORE_CONTINUE]]
; VF4: [[PRED_STORE_CONTINUE]]:
; VF4-NEXT: [[TMP9:%.*]] = extractelement <4 x i1> [[TMP0]], i32 1
; VF4-NEXT: br i1 [[TMP9]], label %[[PRED_STORE_IF1:.*]], label %[[PRED_STORE_CONTINUE2:.*]]
; VF4-NEXT: br i1 true, label %[[PRED_STORE_IF1:.*]], label %[[PRED_STORE_CONTINUE2:.*]]
; VF4: [[PRED_STORE_IF1]]:
; VF4-NEXT: [[TMP10:%.*]] = add i64 [[INDEX]], 1
; VF4-NEXT: [[TMP11:%.*]] = shl nsw i64 [[TMP10]], 1
; VF4-NEXT: [[TMP11:%.*]] = shl nsw i64 1, 1
; VF4-NEXT: [[TMP12:%.*]] = getelementptr inbounds i64, ptr [[DATA]], i64 [[TMP11]]
; VF4-NEXT: [[TMP13:%.*]] = load i64, ptr [[TMP12]], align 8
; VF4-NEXT: store i64 [[TMP13]], ptr [[TMP12]], align 8
Expand All @@ -78,11 +71,9 @@ define void @load_store_interleave_group_tc_2(ptr noalias %data) {
; VF4-NEXT: store i64 [[TMP16]], ptr [[TMP15]], align 8
; VF4-NEXT: br label %[[PRED_STORE_CONTINUE2]]
; VF4: [[PRED_STORE_CONTINUE2]]:
; VF4-NEXT: [[TMP17:%.*]] = extractelement <4 x i1> [[TMP0]], i32 2
; VF4-NEXT: br i1 [[TMP17]], label %[[PRED_STORE_IF3:.*]], label %[[PRED_STORE_CONTINUE4:.*]]
; VF4-NEXT: br i1 false, label %[[PRED_STORE_IF3:.*]], label %[[PRED_STORE_CONTINUE4:.*]]
; VF4: [[PRED_STORE_IF3]]:
; VF4-NEXT: [[TMP18:%.*]] = add i64 [[INDEX]], 2
; VF4-NEXT: [[TMP19:%.*]] = shl nsw i64 [[TMP18]], 1
; VF4-NEXT: [[TMP19:%.*]] = shl nsw i64 2, 1
; VF4-NEXT: [[TMP20:%.*]] = getelementptr inbounds i64, ptr [[DATA]], i64 [[TMP19]]
; VF4-NEXT: [[TMP21:%.*]] = load i64, ptr [[TMP20]], align 8
; VF4-NEXT: store i64 [[TMP21]], ptr [[TMP20]], align 8
Expand All @@ -92,11 +83,9 @@ define void @load_store_interleave_group_tc_2(ptr noalias %data) {
; VF4-NEXT: store i64 [[TMP24]], ptr [[TMP23]], align 8
; VF4-NEXT: br label %[[PRED_STORE_CONTINUE4]]
; VF4: [[PRED_STORE_CONTINUE4]]:
; VF4-NEXT: [[TMP25:%.*]] = extractelement <4 x i1> [[TMP0]], i32 3
; VF4-NEXT: br i1 [[TMP25]], label %[[PRED_STORE_IF5:.*]], label %[[PRED_STORE_CONTINUE6]]
; VF4-NEXT: br i1 false, label %[[PRED_STORE_IF5:.*]], label %[[PRED_STORE_CONTINUE6:.*]]
; VF4: [[PRED_STORE_IF5]]:
; VF4-NEXT: [[TMP26:%.*]] = add i64 [[INDEX]], 3
; VF4-NEXT: [[TMP27:%.*]] = shl nsw i64 [[TMP26]], 1
; VF4-NEXT: [[TMP27:%.*]] = shl nsw i64 3, 1
; VF4-NEXT: [[TMP28:%.*]] = getelementptr inbounds i64, ptr [[DATA]], i64 [[TMP27]]
; VF4-NEXT: [[TMP29:%.*]] = load i64, ptr [[TMP28]], align 8
; VF4-NEXT: store i64 [[TMP29]], ptr [[TMP28]], align 8
Expand All @@ -106,9 +95,7 @@ define void @load_store_interleave_group_tc_2(ptr noalias %data) {
; VF4-NEXT: store i64 [[TMP32]], ptr [[TMP31]], align 8
; VF4-NEXT: br label %[[PRED_STORE_CONTINUE6]]
; VF4: [[PRED_STORE_CONTINUE6]]:
; VF4-NEXT: [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 4
; VF4-NEXT: [[VEC_IND_NEXT]] = add <4 x i8> [[VEC_IND]], splat (i8 4)
; VF4-NEXT: br i1 true, label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
; VF4-NEXT: br label %[[MIDDLE_BLOCK:.*]]
; VF4: [[MIDDLE_BLOCK]]:
; VF4-NEXT: br label %[[EXIT:.*]]
; VF4: [[SCALAR_PH]]:
Expand All @@ -125,7 +112,7 @@ define void @load_store_interleave_group_tc_2(ptr noalias %data) {
; VF4-NEXT: store i64 [[L_1]], ptr [[DATA_1]], align 8
; VF4-NEXT: [[IV_NEXT]] = add nuw nsw i64 [[IV]], 1
; VF4-NEXT: [[EC:%.*]] = icmp eq i64 [[IV_NEXT]], 2
; VF4-NEXT: br i1 [[EC]], label %[[EXIT]], label %[[LOOP]], !llvm.loop [[LOOP3:![0-9]+]]
; VF4-NEXT: br i1 [[EC]], label %[[EXIT]], label %[[LOOP]], !llvm.loop [[LOOP0:![0-9]+]]
; VF4: [[EXIT]]:
; VF4-NEXT: ret void
;
Expand Down Expand Up @@ -315,7 +302,7 @@ define void @test_complex_add_float_tc_4(ptr %res, ptr noalias %A, ptr noalias %
; VF4-NEXT: store float [[ADD_1]], ptr [[GEP_RES_1]], align 4
; VF4-NEXT: [[IV_NEXT]] = add nuw nsw i64 [[IV]], 1
; VF4-NEXT: [[EC:%.*]] = icmp eq i64 [[IV_NEXT]], 4
; VF4-NEXT: br i1 [[EC]], label %[[EXIT]], label %[[LOOP]], !llvm.loop [[LOOP4:![0-9]+]]
; VF4-NEXT: br i1 [[EC]], label %[[EXIT]], label %[[LOOP]], !llvm.loop [[LOOP3:![0-9]+]]
; VF4: [[EXIT]]:
; VF4-NEXT: ret void
;
Expand Down
50 changes: 16 additions & 34 deletions llvm/test/Transforms/LoopVectorize/X86/constant-fold.ll
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -70,46 +70,37 @@ define void @redundant_or_1(ptr %dst, i1 %c.0, i1 %c.1) {
; CHECK-NEXT: [[BROADCAST_SPLAT2:%.*]] = shufflevector <4 x i1> [[BROADCAST_SPLATINSERT1]], <4 x i1> poison, <4 x i32> zeroinitializer
; CHECK-NEXT: br label [[VECTOR_BODY:%.*]]
; CHECK: vector.body:
; CHECK-NEXT: [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[PRED_STORE_CONTINUE8:%.*]] ]
; CHECK-NEXT: [[VEC_IND:%.*]] = phi <4 x i8> [ <i8 0, i8 1, i8 2, i8 3>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[PRED_STORE_CONTINUE8]] ]
; CHECK-NEXT: [[TMP1:%.*]] = icmp ule <4 x i8> [[VEC_IND]], splat (i8 2)
; CHECK-NEXT: [[TMP2:%.*]] = select <4 x i1> [[TMP1]], <4 x i1> [[TMP0]], <4 x i1> zeroinitializer
; CHECK-NEXT: [[TMP2:%.*]] = select <4 x i1> <i1 true, i1 true, i1 true, i1 false>, <4 x i1> [[TMP0]], <4 x i1> zeroinitializer
; CHECK-NEXT: [[TMP5:%.*]] = select <4 x i1> [[TMP2]], <4 x i1> [[BROADCAST_SPLAT2]], <4 x i1> zeroinitializer
; CHECK-NEXT: [[TMP6:%.*]] = extractelement <4 x i1> [[TMP5]], i32 0
; CHECK-NEXT: br i1 [[TMP6]], label [[PRED_STORE_IF:%.*]], label [[PRED_STORE_CONTINUE:%.*]]
; CHECK: pred.store.if:
; CHECK-NEXT: [[TMP7:%.*]] = add i32 [[INDEX]], 0
; CHECK-NEXT: [[TMP8:%.*]] = getelementptr inbounds i32, ptr [[DST:%.*]], i32 [[TMP7]]
; CHECK-NEXT: [[TMP8:%.*]] = getelementptr inbounds i32, ptr [[DST:%.*]], i32 0
; CHECK-NEXT: store i32 0, ptr [[TMP8]], align 4
; CHECK-NEXT: br label [[PRED_STORE_CONTINUE]]
; CHECK: pred.store.continue:
; CHECK-NEXT: [[TMP9:%.*]] = extractelement <4 x i1> [[TMP5]], i32 1
; CHECK-NEXT: br i1 [[TMP9]], label [[PRED_STORE_IF3:%.*]], label [[PRED_STORE_CONTINUE4:%.*]]
; CHECK: pred.store.if3:
; CHECK-NEXT: [[TMP10:%.*]] = add i32 [[INDEX]], 1
; CHECK-NEXT: [[TMP11:%.*]] = getelementptr inbounds i32, ptr [[DST]], i32 [[TMP10]]
; CHECK-NEXT: [[TMP11:%.*]] = getelementptr inbounds i32, ptr [[DST]], i32 1
; CHECK-NEXT: store i32 0, ptr [[TMP11]], align 4
; CHECK-NEXT: br label [[PRED_STORE_CONTINUE4]]
; CHECK: pred.store.continue4:
; CHECK-NEXT: [[TMP12:%.*]] = extractelement <4 x i1> [[TMP5]], i32 2
; CHECK-NEXT: br i1 [[TMP12]], label [[PRED_STORE_IF5:%.*]], label [[PRED_STORE_CONTINUE6:%.*]]
; CHECK: pred.store.if5:
; CHECK-NEXT: [[TMP13:%.*]] = add i32 [[INDEX]], 2
; CHECK-NEXT: [[TMP14:%.*]] = getelementptr inbounds i32, ptr [[DST]], i32 [[TMP13]]
; CHECK-NEXT: [[TMP14:%.*]] = getelementptr inbounds i32, ptr [[DST]], i32 2
; CHECK-NEXT: store i32 0, ptr [[TMP14]], align 4
; CHECK-NEXT: br label [[PRED_STORE_CONTINUE6]]
; CHECK: pred.store.continue6:
; CHECK-NEXT: [[TMP15:%.*]] = extractelement <4 x i1> [[TMP5]], i32 3
; CHECK-NEXT: br i1 [[TMP15]], label [[PRED_STORE_IF7:%.*]], label [[PRED_STORE_CONTINUE8]]
; CHECK-NEXT: br i1 [[TMP15]], label [[PRED_STORE_IF7:%.*]], label [[PRED_STORE_CONTINUE8:%.*]]
; CHECK: pred.store.if7:
; CHECK-NEXT: [[TMP16:%.*]] = add i32 [[INDEX]], 3
; CHECK-NEXT: [[TMP17:%.*]] = getelementptr inbounds i32, ptr [[DST]], i32 [[TMP16]]
; CHECK-NEXT: [[TMP17:%.*]] = getelementptr inbounds i32, ptr [[DST]], i32 3
; CHECK-NEXT: store i32 0, ptr [[TMP17]], align 4
; CHECK-NEXT: br label [[PRED_STORE_CONTINUE8]]
; CHECK: pred.store.continue8:
; CHECK-NEXT: [[INDEX_NEXT]] = add nuw i32 [[INDEX]], 4
; CHECK-NEXT: [[VEC_IND_NEXT]] = add <4 x i8> [[VEC_IND]], splat (i8 4)
; CHECK-NEXT: br i1 true, label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP3:![0-9]+]]
; CHECK-NEXT: br label [[MIDDLE_BLOCK:%.*]]
; CHECK: middle.block:
; CHECK-NEXT: br label [[EXIT:%.*]]
; CHECK: scalar.ph:
Expand All @@ -129,7 +120,7 @@ define void @redundant_or_1(ptr %dst, i1 %c.0, i1 %c.1) {
; CHECK: loop.latch:
; CHECK-NEXT: [[IV_NEXT]] = add nuw nsw i32 [[IV]], 1
; CHECK-NEXT: [[EC:%.*]] = icmp eq i32 [[IV_NEXT]], 3
; CHECK-NEXT: br i1 [[EC]], label [[EXIT]], label [[LOOP_HEADER]], !llvm.loop [[LOOP4:![0-9]+]]
; CHECK-NEXT: br i1 [[EC]], label [[EXIT]], label [[LOOP_HEADER]], !llvm.loop [[LOOP3:![0-9]+]]
; CHECK: exit:
; CHECK-NEXT: ret void
;
Expand Down Expand Up @@ -172,46 +163,37 @@ define void @redundant_or_2(ptr %dst, i1 %c.0, i1 %c.1) {
; CHECK-NEXT: [[BROADCAST_SPLAT2:%.*]] = shufflevector <4 x i1> [[BROADCAST_SPLATINSERT1]], <4 x i1> poison, <4 x i32> zeroinitializer
; CHECK-NEXT: br label [[VECTOR_BODY:%.*]]
; CHECK: vector.body:
; CHECK-NEXT: [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[PRED_STORE_CONTINUE8:%.*]] ]
; CHECK-NEXT: [[VEC_IND:%.*]] = phi <4 x i8> [ <i8 0, i8 1, i8 2, i8 3>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[PRED_STORE_CONTINUE8]] ]
; CHECK-NEXT: [[TMP2:%.*]] = icmp ule <4 x i8> [[VEC_IND]], splat (i8 2)
; CHECK-NEXT: [[TMP3:%.*]] = select <4 x i1> [[TMP2]], <4 x i1> [[TMP0]], <4 x i1> zeroinitializer
; CHECK-NEXT: [[TMP3:%.*]] = select <4 x i1> <i1 true, i1 true, i1 true, i1 false>, <4 x i1> [[TMP0]], <4 x i1> zeroinitializer
; CHECK-NEXT: [[TMP4:%.*]] = select <4 x i1> [[TMP3]], <4 x i1> [[BROADCAST_SPLAT2]], <4 x i1> zeroinitializer
; CHECK-NEXT: [[TMP5:%.*]] = extractelement <4 x i1> [[TMP4]], i32 0
; CHECK-NEXT: br i1 [[TMP5]], label [[PRED_STORE_IF:%.*]], label [[PRED_STORE_CONTINUE:%.*]]
; CHECK: pred.store.if:
; CHECK-NEXT: [[TMP6:%.*]] = add i32 [[INDEX]], 0
; CHECK-NEXT: [[TMP7:%.*]] = getelementptr inbounds i32, ptr [[DST:%.*]], i32 [[TMP6]]
; CHECK-NEXT: [[TMP7:%.*]] = getelementptr inbounds i32, ptr [[DST:%.*]], i32 0
; CHECK-NEXT: store i32 0, ptr [[TMP7]], align 4
; CHECK-NEXT: br label [[PRED_STORE_CONTINUE]]
; CHECK: pred.store.continue:
; CHECK-NEXT: [[TMP8:%.*]] = extractelement <4 x i1> [[TMP4]], i32 1
; CHECK-NEXT: br i1 [[TMP8]], label [[PRED_STORE_IF3:%.*]], label [[PRED_STORE_CONTINUE4:%.*]]
; CHECK: pred.store.if3:
; CHECK-NEXT: [[TMP9:%.*]] = add i32 [[INDEX]], 1
; CHECK-NEXT: [[TMP10:%.*]] = getelementptr inbounds i32, ptr [[DST]], i32 [[TMP9]]
; CHECK-NEXT: [[TMP10:%.*]] = getelementptr inbounds i32, ptr [[DST]], i32 1
; CHECK-NEXT: store i32 0, ptr [[TMP10]], align 4
; CHECK-NEXT: br label [[PRED_STORE_CONTINUE4]]
; CHECK: pred.store.continue4:
; CHECK-NEXT: [[TMP11:%.*]] = extractelement <4 x i1> [[TMP4]], i32 2
; CHECK-NEXT: br i1 [[TMP11]], label [[PRED_STORE_IF5:%.*]], label [[PRED_STORE_CONTINUE6:%.*]]
; CHECK: pred.store.if5:
; CHECK-NEXT: [[TMP12:%.*]] = add i32 [[INDEX]], 2
; CHECK-NEXT: [[TMP13:%.*]] = getelementptr inbounds i32, ptr [[DST]], i32 [[TMP12]]
; CHECK-NEXT: [[TMP13:%.*]] = getelementptr inbounds i32, ptr [[DST]], i32 2
; CHECK-NEXT: store i32 0, ptr [[TMP13]], align 4
; CHECK-NEXT: br label [[PRED_STORE_CONTINUE6]]
; CHECK: pred.store.continue6:
; CHECK-NEXT: [[TMP14:%.*]] = extractelement <4 x i1> [[TMP4]], i32 3
; CHECK-NEXT: br i1 [[TMP14]], label [[PRED_STORE_IF7:%.*]], label [[PRED_STORE_CONTINUE8]]
; CHECK-NEXT: br i1 [[TMP14]], label [[PRED_STORE_IF7:%.*]], label [[PRED_STORE_CONTINUE8:%.*]]
; CHECK: pred.store.if7:
; CHECK-NEXT: [[TMP15:%.*]] = add i32 [[INDEX]], 3
; CHECK-NEXT: [[TMP16:%.*]] = getelementptr inbounds i32, ptr [[DST]], i32 [[TMP15]]
; CHECK-NEXT: [[TMP16:%.*]] = getelementptr inbounds i32, ptr [[DST]], i32 3
; CHECK-NEXT: store i32 0, ptr [[TMP16]], align 4
; CHECK-NEXT: br label [[PRED_STORE_CONTINUE8]]
; CHECK: pred.store.continue8:
; CHECK-NEXT: [[INDEX_NEXT]] = add nuw i32 [[INDEX]], 4
; CHECK-NEXT: [[VEC_IND_NEXT]] = add <4 x i8> [[VEC_IND]], splat (i8 4)
; CHECK-NEXT: br i1 true, label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP5:![0-9]+]]
; CHECK-NEXT: br label [[MIDDLE_BLOCK:%.*]]
; CHECK: middle.block:
; CHECK-NEXT: br label [[EXIT:%.*]]
; CHECK: scalar.ph:
Expand All @@ -231,7 +213,7 @@ define void @redundant_or_2(ptr %dst, i1 %c.0, i1 %c.1) {
; CHECK: loop.latch:
; CHECK-NEXT: [[IV_NEXT]] = add nuw nsw i32 [[IV]], 1
; CHECK-NEXT: [[EC:%.*]] = icmp eq i32 [[IV_NEXT]], 3
; CHECK-NEXT: br i1 [[EC]], label [[EXIT]], label [[LOOP_HEADER]], !llvm.loop [[LOOP6:![0-9]+]]
; CHECK-NEXT: br i1 [[EC]], label [[EXIT]], label [[LOOP_HEADER]], !llvm.loop [[LOOP4:![0-9]+]]
; CHECK: exit:
; CHECK-NEXT: ret void
;
Expand Down
Loading