[all-commits] [llvm/llvm-project] 28afae: [InstCombine] fold sub of min/max intrinsics with ...

[RotateRight via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 28afaed691a0a7ca46bb9f64fac1153e03961fc1
      https://github.com/llvm/llvm-project/commit/28afaed691a0a7ca46bb9f64fac1153e03961fc1
  Author: Sanjay Patel <spatel at rotateright.com>
  Date:   2021-09-11 (Sat, 11 Sep 2021)

  Changed paths:
    M llvm/lib/Transforms/InstCombine/InstCombineAddSub.cpp
    M llvm/test/Transforms/InstCombine/minmax-intrinsics.ll
    M llvm/test/Transforms/InstCombine/sub-minmax.ll

  Log Message:
  -----------
  [InstCombine] fold sub of min/max intrinsics with invertible ops

This is a translation of the existing code to handle the intrinsics
and another step towards D98152.

https://alive2.llvm.org/ce/z/jA7eBC

This pattern is already handled by underlying folds if there are
less uses, so the minimal tests in this case have extra uses.

The larger cmyk tests show the motivation - when combined with
other folds, we invert a larger sequence and eliminate 'not' ops.