[llvm-branch-commits] [llvm] [AMDGPU] Support one immediate folding for global load (PR #178608)

[via llvm-branch-commits]

ruiling wrote:

> Needs negative tests for the known bits handling.

added a test for this.
>Though I would hope this would set the overwrap flags which you use instead?

Did not quite follow this. Do you mean possibly attaching nsw/nuw to the add instruction during early transformation?



https://github.com/llvm/llvm-project/pull/178608