[all-commits] [llvm/llvm-project] 9f0194: [ConstantRange] Add getEquivalentICmp() variant wi...

[Nikita Popov via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 9f0194be4570469ce830e6d581d21c75650562c7
      https://github.com/llvm/llvm-project/commit/9f0194be4570469ce830e6d581d21c75650562c7
  Author: Nikita Popov <nikita.ppv at gmail.com>
  Date:   2021-11-06 (Sat, 06 Nov 2021)

  Changed paths:
    M llvm/include/llvm/IR/ConstantRange.h
    M llvm/lib/IR/ConstantRange.cpp
    M llvm/lib/Transforms/InstCombine/InstructionCombining.cpp
    M llvm/unittests/IR/ConstantRangeTest.cpp

  Log Message:
  -----------
  [ConstantRange] Add getEquivalentICmp() variant with offset (NFCI)

Add a variant of getEquivalentICmp() that produces an optional
offset. This allows us to create an equivalent icmp for all ranges.

Use this in the with.overflow folding code, which was doing this
adjustment separately -- this clarifies that the fold will indeed
always apply.