[PATCH] D155699: [AMDGPU] Allow vector access types in PromoteAllocaToVector

[Pierre van Houtryve via Phabricator via llvm-commits]

Pierre-vh added inline comments.


================
Comment at: llvm/lib/Target/AMDGPU/AMDGPUPromoteAlloca.cpp:474
+
+      // insert_vector is only legal if (Index % NumElts == 0)
+      // If we can't use it, use insertelements.
----------------
arsenm wrote:
> Is there a benefit to using it here? I'd assume we end up breaking it down anyway during legalization
Probably not, it's indeed broken into insert/extracts anyway. I've simplified it.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D155699/new/

https://reviews.llvm.org/D155699