[PATCH] D52944: AMDGPU: Add llvm.amdgcn.ds.ordered.add & swap

[Marek Olšák via Phabricator via llvm-commits]

mareko added a comment.

I might change the intrinsic to add the option to insert "s_and_saveexec s[N:M], 1" and "s_mov_b64 exec, s[N:M]" around the intrinsic to get an optimal single-lane block.


Repository:
  rL LLVM

https://reviews.llvm.org/D52944