diff options
| author | George Rokos <grokos@us.ibm.com> | 2017-12-07 20:27:31 +0000 |
|---|---|---|
| committer | George Rokos <grokos@us.ibm.com> | 2017-12-07 20:27:31 +0000 |
| commit | 145c54721b6e91a4fcf0d352d7e396cce3939cb1 (patch) | |
| tree | b18124cf16c739b8533c4ecf21b6d39ee1bc5fd4 /llvm/lib/Target/AMDGPU/GCNHazardRecognizer.cpp | |
| parent | dfc79c7c333b1764140c9da01ffd37222ba4939c (diff) | |
| download | bcm5719-llvm-145c54721b6e91a4fcf0d352d7e396cce3939cb1.tar.gz bcm5719-llvm-145c54721b6e91a4fcf0d352d7e396cce3939cb1.zip | |
[OpenMP] NVPTX: Set default/minimum compute capability to sm_35
The current implementation of the nvptx runtime (to be upstreamed shortly) uses the atomicMax operation on 64-bit integers.
This is only supported in compute capabilities 3.5 and later. I've changed the clang default to sm_35.
Differential Revision: https://reviews.llvm.org/D40977
llvm-svn: 320082
Diffstat (limited to 'llvm/lib/Target/AMDGPU/GCNHazardRecognizer.cpp')
0 files changed, 0 insertions, 0 deletions
