[llvm] [AMDGPU] SelectionDAG support for vector type set 0 to multiple sgpr64 (PR #128017)

Janek van Oirschot via llvm-commits llvm-commits at lists.llvm.org
Thu Feb 20 08:06:02 PST 2025


JanekvO wrote:

> You also have a number of regressions in the VALU cases. Should do something about those (which may mean not doing this for VALU values). 

Looking into it
> Also should consider v_mov_b64 on gfx940

Was going to be a follow up patch to fold the sgpr64 set 0 cases into vgpr64 in SIFoldOperand (perhaps should be peephole opt?)



https://github.com/llvm/llvm-project/pull/128017


More information about the llvm-commits mailing list