diff options
| author | Matt Arsenault <Matthew.Arsenault@amd.com> | 2016-05-21 01:53:33 +0000 |
|---|---|---|
| committer | Matt Arsenault <Matthew.Arsenault@amd.com> | 2016-05-21 01:53:33 +0000 |
| commit | 81a709503d378c188462465918df161d664c4205 (patch) | |
| tree | f7abd4dc6db616979961c93ca4dc06a11d852ee5 /llvm/lib/Transforms | |
| parent | f2723a2a91252c7cee5dff054ebfff1324dab733 (diff) | |
| download | bcm5719-llvm-81a709503d378c188462465918df161d664c4205.tar.gz bcm5719-llvm-81a709503d378c188462465918df161d664c4205.zip | |
AMDGPU: Fix high bits after division optimization
This is essentially doing a 24-bit signed division with FP.
We need to truncate to the N bit result.
llvm-svn: 270305
Diffstat (limited to 'llvm/lib/Transforms')
0 files changed, 0 insertions, 0 deletions
