[PATCH] D133787: [AMDGPU][SILoadStoreOptimizer] Merge SGPR_IMM scalar buffer loads.

[Matt Arsenault via Phabricator via llvm-commits]

arsenm added a comment.

In D133787#3788908 <https://reviews.llvm.org/D133787#3788908>, @foad wrote:

> I can't resist pointing out that this patch would be simpler if we never used the _SGPR form on subtargets where the _SGPR_IMM form is available.

It would be even simpler if we did this in the IR to begin with. It's only after codegen you have to worry so much about addressing mode minutia


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D133787/new/

https://reviews.llvm.org/D133787