[PATCH] D88245: [AMDGPU] Make bfi patterns divergence-aware

[Jay Foad via Phabricator via llvm-commits]

foad added a comment.

In D88245#2293131 <https://reviews.llvm.org/D88245#2293131>, @arsenm wrote:

> LGTM (although I think readfirstlane is the same cost as a regular copy)

readfirstlane is bad news because there's a pretty big stall between a VALU instruction that writes an SGPR, and any following SALU instruction.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D88245/new/

https://reviews.llvm.org/D88245