diff options
| author | Simon Pilgrim <llvm-dev@redking.me.uk> | 2022-01-13 17:29:41 +0000 |
|---|---|---|
| committer | Simon Pilgrim <llvm-dev@redking.me.uk> | 2022-01-13 17:44:10 +0000 |
| commit | 08212dbc449bbf10652aa735791280266d473076 (patch) | |
| tree | b063b27e89bb903cc6edc9ba74ddb0a2bd355ba1 /llvm/lib/TextAPI/Architecture.cpp | |
| parent | e2c78f99c41921927bc5a1ca8adbbc722d3c2a01 (diff) | |
| download | llvm-08212dbc449bbf10652aa735791280266d473076.zip llvm-08212dbc449bbf10652aa735791280266d473076.tar.gz llvm-08212dbc449bbf10652aa735791280266d473076.tar.bz2 | |
[X86] Add xop/avx2 shifts to X86TargetLowering::isBinOp
Allows shuffle combining through per-element shift nodes
This exposed a number of issues with shuffle combining with target intrinsics that are lowered to nodes later during legalization - in particular shuffle combining and SimplifyDemandedVectorElts were being called after canonicalizeShuffleWithBinOps, meaning that shuffles didn't have a chance to be combined away before the shuffle(binop(x,y)) -> binop(shuffle(x),shuffle(y)) fold.
Diffstat (limited to 'llvm/lib/TextAPI/Architecture.cpp')
0 files changed, 0 insertions, 0 deletions