[all-commits] [llvm/llvm-project] 63fe80: [SeperateConstOffsetFromGEP] Handle `or disjoint` ...

[Krzysztof Drewniak via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 63fe80fb18c9660e678d24c184021841cb02d82f
      https://github.com/llvm/llvm-project/commit/63fe80fb18c9660e678d24c184021841cb02d82f
  Author: Krzysztof Drewniak <Krzysztof.Drewniak at amd.com>
  Date:   2024-01-26 (Fri, 26 Jan 2024)

  Changed paths:
    M llvm/lib/Transforms/Scalar/SeparateConstOffsetFromGEP.cpp
    M llvm/test/CodeGen/AMDGPU/GlobalISel/merge-buffer-stores.ll
    M llvm/test/CodeGen/AMDGPU/constant-address-space-32bit.ll
    M llvm/test/CodeGen/AMDGPU/extract-subvector-16bit.ll
    M llvm/test/CodeGen/AMDGPU/machine-sink-temporal-divergence-swdev407790.ll
    M llvm/test/CodeGen/NVPTX/vector-loads.ll
    M llvm/test/CodeGen/PowerPC/mma-intrinsics.ll
    M llvm/test/CodeGen/PowerPC/sched-addi.ll
    M llvm/test/Transforms/SeparateConstOffsetFromGEP/AMDGPU/split-gep-and-gvn-addrspace-addressing-modes.ll
    M llvm/test/Transforms/SeparateConstOffsetFromGEP/NVPTX/split-gep.ll
    M llvm/test/Transforms/SeparateConstOffsetFromGEP/split-gep-or-as-add.ll

  Log Message:
  -----------
  [SeperateConstOffsetFromGEP] Handle `or disjoint` flags (#76997)

This commit extends separate-const-offset-from-gep to look at the
newly-added `disjoint` flag on `or` instructions so as to preserve
additional opportunities for optimization.

The tests were pre-committed in #76972.