[PATCH] D84403: [AMDGPU] Use ds_read/write_b96/b128 when possible for SDag

[Matt Arsenault via Phabricator via llvm-commits]

arsenm added a comment.

In D84403#2248536 <https://reviews.llvm.org/D84403#2248536>, @mareko wrote:

> This breaks LDS. LLVMSetAlignment(inst, 4) on loads and stores has no effect. The IR says "align 4", yet the backend still selects b128.
>
> Please revert ASAP. @nhaehnle

As far as I can tell, b64/b96/b128 do work correctly with unaligned access, and it's potentially a bug we don't use them more aggressively


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D84403/new/

https://reviews.llvm.org/D84403