[all-commits] [llvm/llvm-project] 08fb46: [AVX-512] Fix for disjoint-or-fold (VGF2P8AFFINEQB...

[Kartik Ohlan via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 08fb46468e9b03d8f13e8c3d9e40f7e9adff1a1d
      https://github.com/llvm/llvm-project/commit/08fb46468e9b03d8f13e8c3d9e40f7e9adff1a1d
  Author: Kartik Ohlan <kartik7ohlan at gmail.com>
  Date:   2026-04-14 (Tue, 14 Apr 2026)

  Changed paths:
    M llvm/lib/Target/X86/X86ISelLowering.cpp
    A llvm/test/CodeGen/X86/gfni-or-fold.ll

  Log Message:
  -----------
  [AVX-512] Fix for disjoint-or-fold (VGF2P8AFFINEQB) (#190896)

Fixes #190502

Added implementation of helper combineOrWithGF2P8AFFINEQB and wired the logic with combineOrXorWithSETCC:

Fold: (GF2P8AFFINEQB(X, Y, Imm) or_disjoint SplatVal) -> GF2P8AFFINEQB(X, Y, Imm ^ SplatVal)

When OR is disjoint (no common bits), the splat constant can be folded directly into the GF2P8AFFINEQB immediate via XOR.



To unsubscribe from these emails, change your notification settings at https://github.com/llvm/llvm-project/settings/notifications