RPCS3/llvm-mirror
1
0
Fork 0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2024-11-22 18:54:02 +01:00
Code Issues Projects Releases Wiki Activity

[InstCombine] limit extracting shuffle transform based on uses

Browse Source 
As discussed in D53037, this can lead to worse codegen, and we
don't generally expect the backend to be able to optimize
arbitrary shuffles. If there's only one use of the 1st shuffle,
that means it's getting removed, so that should always be
safe.

llvm-svn: 353235
This commit is contained in:
Sanjay Patel 2019-02-05 22:58:45 +00:00
parent cb321a6905
commit f15483dfff
2 changed files with 7 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
lib/Transforms/InstCombine/InstCombineVectorOps.cpp
View File

@ -1498,6 +1498,11 @@ static Instruction *foldIdentityExtractShuffle(ShuffleVectorInst &Shuf) {
if (!match(Op0, m_ShuffleVector(m_Value(X), m_Value(Y), m_Constant(Mask)))) if (!match(Op0, m_ShuffleVector(m_Value(X), m_Value(Y), m_Constant(Mask))))
return nullptr; return nullptr;
// Be conservative with shuffle transforms. If we can't kill the 1st shuffle,
// then combining may result in worse codegen.
if (!Op0->hasOneUse())
return nullptr;
// We are extracting a subvector from a shuffle. Remove excess elements from // We are extracting a subvector from a shuffle. Remove excess elements from
// the 1st shuffle mask to eliminate the extract. // the 1st shuffle mask to eliminate the extract.
// //

4
test/Transforms/InstCombine/vec_shuffle.ll
View File

@ -195,7 +195,7 @@ define <4 x i8> @extract_subvector_of_shuffle_undefs_types(<2 x i8> %x, <2 x i8>
ret <4 x i8> %extract_subv ret <4 x i8> %extract_subv
} }
; FIXME: Extra uses are not ok - we should only do the transform when we can eliminate an instruction. ; Extra uses are not ok - we only do the transform when we can eliminate an instruction.
declare void @use_v5i8(<5 x i8>) declare void @use_v5i8(<5 x i8>)
@ -203,7 +203,7 @@ define <4 x i8> @extract_subvector_of_shuffle_extra_use(<2 x i8> %x, <2 x i8> %y
; CHECK-LABEL: @extract_subvector_of_shuffle_extra_use( ; CHECK-LABEL: @extract_subvector_of_shuffle_extra_use(
; CHECK-NEXT: [[SHUF:%.*]] = shufflevector <2 x i8> [[X:%.*]], <2 x i8> [[Y:%.*]], <5 x i32> <i32 undef, i32 2, i32 0, i32 1, i32 0> ; CHECK-NEXT: [[SHUF:%.*]] = shufflevector <2 x i8> [[X:%.*]], <2 x i8> [[Y:%.*]], <5 x i32> <i32 undef, i32 2, i32 0, i32 1, i32 0>
; CHECK-NEXT: call void @use_v5i8(<5 x i8> [[SHUF]]) ; CHECK-NEXT: call void @use_v5i8(<5 x i8> [[SHUF]])
; CHECK-NEXT: [[EXTRACT_SUBV:%.*]] = shufflevector <2 x i8> [[X]], <2 x i8> [[Y]], <4 x i32> <i32 undef, i32 2, i32 0, i32 undef> ; CHECK-NEXT: [[EXTRACT_SUBV:%.*]] = shufflevector <5 x i8> [[SHUF]], <5 x i8> undef, <4 x i32> <i32 0, i32 1, i32 2, i32 undef>
; CHECK-NEXT: ret <4 x i8> [[EXTRACT_SUBV]] ; CHECK-NEXT: ret <4 x i8> [[EXTRACT_SUBV]]
; ;
%shuf = shufflevector <2 x i8> %x, <2 x i8> %y, <5 x i32> <i32 undef, i32 2, i32 0, i32 1, i32 0> %shuf = shufflevector <2 x i8> %x, <2 x i8> %y, <5 x i32> <i32 undef, i32 2, i32 0, i32 1, i32 0>