diff options
| author | Matt Arsenault <Matthew.Arsenault@amd.com> | 2015-12-19 01:46:41 +0000 |
|---|---|---|
| committer | Matt Arsenault <Matthew.Arsenault@amd.com> | 2015-12-19 01:46:41 +0000 |
| commit | 2aed6ca1d33468f7e6f8509213a1b9dcd8026ec1 (patch) | |
| tree | a4575401ecb619e067cc2b4fed2a83056b80163c /llvm/lib/IR/Module.cpp | |
| parent | 10a509292cda51aabf3630bf30ceb990d3bf3108 (diff) | |
| download | llvm-2aed6ca1d33468f7e6f8509213a1b9dcd8026ec1.zip llvm-2aed6ca1d33468f7e6f8509213a1b9dcd8026ec1.tar.gz llvm-2aed6ca1d33468f7e6f8509213a1b9dcd8026ec1.tar.bz2 | |
AMDGPU: Switch barrier intrinsics to using convergent
noduplicate prevents unrolling of small loops that happen to have
barriers in them. If a loop has a barrier in it, it is OK to duplicate
it for the unroll.
llvm-svn: 256075
Diffstat (limited to 'llvm/lib/IR/Module.cpp')
0 files changed, 0 insertions, 0 deletions