diff options
| author | Carl Ritson <carl.ritson@amd.com> | 2022-03-28 17:30:09 +0900 |
|---|---|---|
| committer | Carl Ritson <carl.ritson@amd.com> | 2022-03-28 17:44:54 +0900 |
| commit | 1f52d02cebf1fd7a8777fcc925837ffabcf73681 (patch) | |
| tree | d5526c8b4702bb19d12d9b9fc37b5f54bb5a26b2 /llvm/lib/IR/Module.cpp | |
| parent | 12f0802c93f5073b1fe02bcae118160ea169cb29 (diff) | |
| download | llvm-1f52d02cebf1fd7a8777fcc925837ffabcf73681.zip llvm-1f52d02cebf1fd7a8777fcc925837ffabcf73681.tar.gz llvm-1f52d02cebf1fd7a8777fcc925837ffabcf73681.tar.bz2 | |
[AMDGPU] Split waterfall loop exec manipulation
Split waterfall loops into multiple blocks so that exec mask
manipulation (s_and_saveexec) does not occur in the middle of
a block.
VGPR live range optimizer is updated to handle waterfall loops
spanning multiple blocks.
Reviewed By: ruiling
Differential Revision: https://reviews.llvm.org/D122200
Diffstat (limited to 'llvm/lib/IR/Module.cpp')
0 files changed, 0 insertions, 0 deletions