diff options
| author | Matt Arsenault <Matthew.Arsenault@amd.com> | 2017-01-19 06:35:27 +0000 |
|---|---|---|
| committer | Matt Arsenault <Matthew.Arsenault@amd.com> | 2017-01-19 06:35:27 +0000 |
| commit | 3e6f9b577368f20c39f7a4fd181b1a02f6911679 (patch) | |
| tree | 3bdb65a7680092ef2eea3970dabd35a43259d710 /llvm/utils/TableGen/TableGen.cpp | |
| parent | 3b99f12a4e6ff5f2f5292b8db8aa6904e1e26bc6 (diff) | |
| download | bcm5719-llvm-3e6f9b577368f20c39f7a4fd181b1a02f6911679.tar.gz bcm5719-llvm-3e6f9b577368f20c39f7a4fd181b1a02f6911679.zip | |
AMDGPU: Disable some fneg combines unless nsz
For -(x + y) -> (-x) + (-y), if x == -y, this would
change the result from -0.0 to 0.0. Since the fma/fmad
combine is an extension of this problem it also
applies there.
fmul should be fine, and I don't think any of the unary
operators or conversions should be a problem either.
llvm-svn: 292473
Diffstat (limited to 'llvm/utils/TableGen/TableGen.cpp')
0 files changed, 0 insertions, 0 deletions
