[llvm] [AMDGPU] narrow only on store to pow of 2 mem location (PR #150093)

Tiger Ding via llvm-commits llvm-commits at lists.llvm.org
Thu Jul 24 13:19:03 PDT 2025


zerogtiger wrote:

I was aware that most random types were used to test the robustness. As to whether we should support arbitrary integer types, this is not a question I can answer -- I don't have enough context or knowledge.

However, what I know is that given GlobalISel supports these weird types, it feels like a natural next step to support the legalization of these types, especially since GlobalISel is meant to replace SelectionDAG and fastISel, which already has support for arbitrary integer sizes. 

https://github.com/llvm/llvm-project/pull/150093


More information about the llvm-commits mailing list