tree 4fcdde0dd6d4c4baac04f1723fa15550201acd94
parent 6c801086a38924edbcd1928d252e69eb34332abd
author Matt Arsenault <Matthew.Arsenault@amd.com> 1544220737 +0000
committer Matt Arsenault <Matthew.Arsenault@amd.com> 1544220737 +0000

AMDGPU: Fix offsets for < 4-byte aggregate kernel arguments

We were still using the rounded down offset and alignment even though
they aren't handled because you can't trivially bitcast the loaded
value.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@348658 91177308-0d34-0410-b5e6-96231b3b80d8