diff options
| author | Sanjay Patel <spatel@rotateright.com> | 2020-11-02 12:01:49 -0500 |
|---|---|---|
| committer | Sanjay Patel <spatel@rotateright.com> | 2020-11-02 12:20:17 -0500 |
| commit | 35fa3c474f4e6685da0d883bb6707ac5aa4d37cc (patch) | |
| tree | 4db524e2b25ac44b85d8bb047f98d2e9e4d969a6 /clang/lib/Frontend/CompilerInvocation.cpp | |
| parent | 9586656f4e3cf53a2755ad965773909b52728332 (diff) | |
| download | llvm-35fa3c474f4e6685da0d883bb6707ac5aa4d37cc.zip llvm-35fa3c474f4e6685da0d883bb6707ac5aa4d37cc.tar.gz llvm-35fa3c474f4e6685da0d883bb6707ac5aa4d37cc.tar.bz2 | |
[x86] add AVX2 cost model entries for maxnum of 256-bit vectors
As noticed in D90554 ,
the AVX2 costs for 256-bit vectors did not include FMAXNUM entries,
so we fell back to AVX1 which assumes those ops will be split into
128-bit halves or something close to that.
Differential Revision: https://reviews.llvm.org/D90613
Diffstat (limited to 'clang/lib/Frontend/CompilerInvocation.cpp')
0 files changed, 0 insertions, 0 deletions