[PATCH] D84223: [AMDGPU] Don't combine memory intrs to v3i16

[Matt Arsenault via Phabricator via llvm-commits]

arsenm added a comment.

Is there any real obstacle to handling these? Even the DAG has v3i16/v3f16 types now


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D84223/new/

https://reviews.llvm.org/D84223