diff options
| author | Craig Topper <craig.topper@gmail.com> | 2016-10-26 04:59:58 +0000 |
|---|---|---|
| committer | Craig Topper <craig.topper@gmail.com> | 2016-10-26 04:59:58 +0000 |
| commit | 812d3d30aede80b5f9e5b1fc0484700ecaf1afef (patch) | |
| tree | 029ecae74275eb078a716136ec5ad8ca1fa66b1e /clang/lib/CodeGen/CGDecl.cpp | |
| parent | a3b7527ccb4400be6978b6ed6283afc202ecd753 (diff) | |
| download | llvm-812d3d30aede80b5f9e5b1fc0484700ecaf1afef.zip llvm-812d3d30aede80b5f9e5b1fc0484700ecaf1afef.tar.gz llvm-812d3d30aede80b5f9e5b1fc0484700ecaf1afef.tar.bz2 | |
[AVX-512] Add scalar vfmsub/vfnmsub mask3 intrinsics
Summary: Clang's intrinsic header currently tries to negate the third operand of a vfmadd mask3 in order to create vfmsub, but this fails isel. This patch adds scalar vfmsub and vfnmsub mask3 that we can use instead to avoid the negate. This is consistent with the packed instructions.
Reviewers: igorb, delena
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D25933
llvm-svn: 285173
Diffstat (limited to 'clang/lib/CodeGen/CGDecl.cpp')
0 files changed, 0 insertions, 0 deletions