diff options
| author | Sanjay Patel <spatel@rotateright.com> | 2019-11-13 15:41:26 -0500 |
|---|---|---|
| committer | Sanjay Patel <spatel@rotateright.com> | 2019-11-13 15:57:35 -0500 |
| commit | a3e61946c5bd7bdfab15af76b292e52d6ffa27f7 (patch) | |
| tree | e774948420156e303bbf01091ce5984b38dec136 /llvm/test/Transforms/SLPVectorizer | |
| parent | 52980576ee4a5683aa9a428b76335ed617f142ab (diff) | |
| download | bcm5719-llvm-a3e61946c5bd7bdfab15af76b292e52d6ffa27f7.tar.gz bcm5719-llvm-a3e61946c5bd7bdfab15af76b292e52d6ffa27f7.zip | |
[SLP] fix miscompile on min/max reductions with extra uses (PR43948)
The bug manifests as replacing a reduction operand with an undef
value.
The problem appears to be limited to cases where a min/max reduction
has extra uses of the compare operand to the select.
In the general case, we are tracking "ExternallyUsedValues" and
an "IgnoreList" of the reduction operations, but those may not apply
to the final compare+select in a min/max reduction.
For that, we use replaceAllUsesWith (RAUW) to ensure that the new
vectorized reduction values are transferred to all subsequent users.
Differential Revision: https://reviews.llvm.org/D70148
Diffstat (limited to 'llvm/test/Transforms/SLPVectorizer')
| -rw-r--r-- | llvm/test/Transforms/SLPVectorizer/X86/reduction.ll | 2 | ||||
| -rw-r--r-- | llvm/test/Transforms/SLPVectorizer/X86/used-reduced-op.ll | 2 |
2 files changed, 2 insertions, 2 deletions
diff --git a/llvm/test/Transforms/SLPVectorizer/X86/reduction.ll b/llvm/test/Transforms/SLPVectorizer/X86/reduction.ll index 40cb4f0ad4a..5ef66dbb57b 100644 --- a/llvm/test/Transforms/SLPVectorizer/X86/reduction.ll +++ b/llvm/test/Transforms/SLPVectorizer/X86/reduction.ll @@ -91,7 +91,7 @@ define i32 @horiz_max_multiple_uses([32 x i32]* %x, i32* %p) { ; CHECK-NEXT: [[TMP5:%.*]] = select i1 [[TMP4]], i32 [[TMP3]], i32 [[T4]] ; CHECK-NEXT: [[C012345:%.*]] = icmp sgt i32 [[TMP5]], [[T5]] ; CHECK-NEXT: [[T17:%.*]] = select i1 [[C012345]], i32 [[TMP5]], i32 [[T5]] -; CHECK-NEXT: [[THREE_OR_FOUR:%.*]] = select i1 undef, i32 3, i32 4 +; CHECK-NEXT: [[THREE_OR_FOUR:%.*]] = select i1 [[TMP4]], i32 3, i32 4 ; CHECK-NEXT: store i32 [[THREE_OR_FOUR]], i32* [[P:%.*]], align 8 ; CHECK-NEXT: ret i32 [[T17]] ; diff --git a/llvm/test/Transforms/SLPVectorizer/X86/used-reduced-op.ll b/llvm/test/Transforms/SLPVectorizer/X86/used-reduced-op.ll index a8752bcff3e..468821130e7 100644 --- a/llvm/test/Transforms/SLPVectorizer/X86/used-reduced-op.ll +++ b/llvm/test/Transforms/SLPVectorizer/X86/used-reduced-op.ll @@ -71,7 +71,7 @@ define void @n() local_unnamed_addr #0 { ; CHECK-NEXT: [[NEG_1_1:%.*]] = sub nsw i32 0, [[SUB_1_1]] ; CHECK-NEXT: [[TMP46:%.*]] = select i1 [[TMP45]], i32 [[NEG_1_1]], i32 [[SUB_1_1]] ; CHECK-NEXT: [[CMP12_1_1:%.*]] = icmp slt i32 [[TMP46]], [[OP_EXTRA]] -; CHECK-NEXT: [[NARROW:%.*]] = or i1 [[CMP12_1_1]], undef +; CHECK-NEXT: [[NARROW:%.*]] = or i1 [[CMP12_1_1]], [[TMP44]] ; CHECK-NEXT: [[SPEC_SELECT8_1_1:%.*]] = select i1 [[CMP12_1_1]], i32 [[TMP46]], i32 [[OP_EXTRA]] ; CHECK-NEXT: [[SUB_2_1:%.*]] = sub i32 [[TMP30]], [[TMP3]] ; CHECK-NEXT: [[TMP47:%.*]] = icmp slt i32 [[SUB_2_1]], 0 |
