diff options
| author | Matt Arsenault <Matthew.Arsenault@amd.com> | 2020-01-17 15:40:15 -0500 |
|---|---|---|
| committer | Matt Arsenault <arsenm2@gmail.com> | 2020-01-22 09:26:17 -0500 |
| commit | 9c928649a085646c4c779bac095643b50b464d83 (patch) | |
| tree | 9dd51c9e5fab735e82617a682c6c56351b59c19d /llvm/lib/CodeGen/MachineScheduler.cpp | |
| parent | 4481eefbe8425c63289186dd13319aaa7043e67f (diff) | |
| download | llvm-9c928649a085646c4c779bac095643b50b464d83.zip llvm-9c928649a085646c4c779bac095643b50b464d83.tar.gz llvm-9c928649a085646c4c779bac095643b50b464d83.tar.bz2 | |
AMDGPU: Fix interaction of tfe and d16
This using the wrong result register, and dropping the result entirely
for v2f16. This would fail to select on the scalar case. I believe it
was also mishandling packed/unpacked subtargets.
Diffstat (limited to 'llvm/lib/CodeGen/MachineScheduler.cpp')
0 files changed, 0 insertions, 0 deletions