diff options
| author | Matt Arsenault <Matthew.Arsenault@amd.com> | 2019-06-14 21:52:26 +0000 |
|---|---|---|
| committer | Matt Arsenault <Matthew.Arsenault@amd.com> | 2019-06-14 21:52:26 +0000 |
| commit | aa41e92e17a34d5cddfc5cf4b42bea9c5b82a22d (patch) | |
| tree | a8e610d7ea94ae9da85fc9d43609ba037ab9f202 /clang/lib | |
| parent | af857b93df36a983a27a82455c0dea39f53da5b1 (diff) | |
| download | bcm5719-llvm-aa41e92e17a34d5cddfc5cf4b42bea9c5b82a22d.tar.gz bcm5719-llvm-aa41e92e17a34d5cddfc5cf4b42bea9c5b82a22d.zip | |
AMDGPU: Avoid most waitcnts before calls
Currently you get extra waits, because waits are inserted for the
register dependencies of the call, and the function prolog waits on
everything.
Currently waits are still inserted on returns. It may make sense to
not do this, and wait in the caller instead.
llvm-svn: 363465
Diffstat (limited to 'clang/lib')
0 files changed, 0 insertions, 0 deletions
