[all-commits] [llvm/llvm-project] f2fbdf: [InstCombine] do not exclude min/max from icmp wit...

[RotateRight via All-commits]

  Branch: refs/heads/master
  Home:   https://github.com/llvm/llvm-project
  Commit: f2fbdf76d8d07f6a0fbd97825cbc533660d64a37
      https://github.com/llvm/llvm-project/commit/f2fbdf76d8d07f6a0fbd97825cbc533660d64a37
  Author: Sanjay Patel <spatel at rotateright.com>
  Date:   2020-03-30 (Mon, 30 Mar 2020)

  Changed paths:
    M llvm/lib/Transforms/InstCombine/InstCombineCompares.cpp
    M llvm/test/Transforms/InstCombine/adjust-for-minmax.ll
    M llvm/test/Transforms/InstCombine/select_meta.ll

  Log Message:
  -----------
  [InstCombine] do not exclude min/max from icmp with casted operand fold

InstCombine has a mess of logic that tries to preserve min/max patterns,
but AFAICT, this one is not necessary because we can always narrow the
corresponding select in this sequence to match the narrow compare.

The biggest danger for this patch is inducing infinite looping or
assert from exceeding max iterations. If any bots hit that in the
vicinity of this commit, this is the likely patch to blame.