Skip to content

[X86] lowerV8F32Shuffle - use lowerShufflePairAsUNPCKAndPermute on AVX1 targets #140881

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
RKSimon merged 1 commit into from
May 21, 2025
Merged

[X86] lowerV8F32Shuffle - use lowerShufflePairAsUNPCKAndPermute on AVX1 targets #140881

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
9 changes: 5 additions & 4 deletions llvm/lib/Target/X86/X86ISelLowering.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -16705,10 +16705,11 @@ static SDValue lowerV8F32Shuffle(const SDLoc &DL, ArrayRef<int> Mask,

// Try to match an interleave of two v8f32s and lower them as unpck and
// permutes using ymms. This needs to go before we try to split the vectors.
//
// TODO: Expand this to AVX1. Currently v8i32 is casted to v8f32 and hits
// this path inadvertently.
if (Subtarget.hasAVX2() && !Subtarget.hasAVX512())
// Don't attempt on AVX1 if we're likely to split vectors anyway.
if ((Subtarget.hasAVX2() ||
!(isFreeToSplitVector(peekThroughBitcasts(V1), DAG) ||
isFreeToSplitVector(peekThroughBitcasts(V2), DAG))) &&
!Subtarget.hasAVX512())
if (SDValue V = lowerShufflePairAsUNPCKAndPermute(DL, MVT::v8f32, V1, V2,
Mask, DAG))
return V;
Expand Down
13 changes: 4 additions & 9 deletions llvm/test/CodeGen/X86/vector-interleave.ll
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -265,15 +265,10 @@ define <16 x i32> @interleave2x8i32(<8 x i32> %a, <8 x i32> %b) {
;
; AVX1-LABEL: interleave2x8i32:
; AVX1: # %bb.0:
; AVX1-NEXT: vunpckhps {{.*#+}} xmm2 = xmm0[2],xmm1[2],xmm0[3],xmm1[3]
; AVX1-NEXT: vunpcklps {{.*#+}} xmm3 = xmm0[0],xmm1[0],xmm0[1],xmm1[1]
; AVX1-NEXT: vinsertf128 $1, %xmm2, %ymm3, %ymm2
; AVX1-NEXT: vextractf128 $1, %ymm1, %xmm1
; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm0
; AVX1-NEXT: vunpckhps {{.*#+}} xmm3 = xmm0[2],xmm1[2],xmm0[3],xmm1[3]
; AVX1-NEXT: vunpcklps {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1]
; AVX1-NEXT: vinsertf128 $1, %xmm3, %ymm0, %ymm1
; AVX1-NEXT: vmovaps %ymm2, %ymm0
; AVX1-NEXT: vunpckhps {{.*#+}} ymm2 = ymm0[2],ymm1[2],ymm0[3],ymm1[3],ymm0[6],ymm1[6],ymm0[7],ymm1[7]
; AVX1-NEXT: vunpcklps {{.*#+}} ymm0 = ymm0[0],ymm1[0],ymm0[1],ymm1[1],ymm0[4],ymm1[4],ymm0[5],ymm1[5]
; AVX1-NEXT: vperm2f128 {{.*#+}} ymm1 = ymm0[2,3],ymm2[2,3]
; AVX1-NEXT: vinsertf128 $1, %xmm2, %ymm0, %ymm0
; AVX1-NEXT: retq
;
; AVX2-LABEL: interleave2x8i32:
Expand Down
Loading