[llvm] [AMDGPU] Convert ds_bpermute/wave_shuffle XOR patterns to DPP row_xmask and permlanex16 (PR #194565)

[Jay Foad via llvm-commits]

================

----------------
jayfoad wrote:

You could handle all XOR values 16..31 if you emit the right constants for src1 and src2.

https://github.com/llvm/llvm-project/pull/194565