[PATCH] D140798: [InstCombine] Fold zero check followed by decrement to usub.sat

[Alexander via Phabricator via llvm-commits]

alex-t added a comment.

This change caused a regression in AMDGPU backend.
In case optimization is done before inline, it covers opportunities for other inst-combines that may have higher precedence.
As a result, we get suboptimal code.

The example is in the https://reviews.llvm.org/D147146 which I have created to propose a possible fix.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D140798/new/

https://reviews.llvm.org/D140798