[llvm] [AMDGPU] Prevent SDWA 'preserve' transformation for instructions in different basic blocks. (PR #82406)
Jay Foad via llvm-commits
llvm-commits at lists.llvm.org
Wed Feb 21 02:24:41 PST 2024
jayfoad wrote:
Why not allow the optimization, and just fix the "Move MI before v_or_b32" code in `convertToSDWA`?
https://github.com/llvm/llvm-project/pull/82406
More information about the llvm-commits
mailing list