diff options
| author | Simon Pilgrim <llvm-dev@redking.me.uk> | 2022-10-21 10:39:57 +0100 |
|---|---|---|
| committer | Simon Pilgrim <llvm-dev@redking.me.uk> | 2022-10-21 10:40:13 +0100 |
| commit | 5ca77541446d7040638b53e0ddff3f76ee005681 (patch) | |
| tree | dcb14b0b789f6d662985ae77445cd163ef103296 /llvm/lib/Bitcode/Writer/BitcodeWriter.cpp | |
| parent | daf067da04c98ce666f67ba2b3beed4cf4be7549 (diff) | |
| download | llvm-5ca77541446d7040638b53e0ddff3f76ee005681.zip llvm-5ca77541446d7040638b53e0ddff3f76ee005681.tar.gz llvm-5ca77541446d7040638b53e0ddff3f76ee005681.tar.bz2 | |
[X86] Fold scalar_to_vector(i64 zext(x)) -> bitcast(vzext_movl(scalar_to_vector(i32 x)))
Extends existing anyextend fold to make use of the implicit zero-extension of the movd instruction
This also helps replace some nasty xmm->gpr->xmm traffic with a shuffle pattern instead
Noticed while looking at D130953
Diffstat (limited to 'llvm/lib/Bitcode/Writer/BitcodeWriter.cpp')
0 files changed, 0 insertions, 0 deletions