[PATCH] D155267: [DAGCombiner] Change foldAndOrOfSETCC() to optimize and/or patterns with floating points.

[Brooks Moses via Phabricator via llvm-commits]

brooksmoses added a comment.

As a heads-up, I'm seeing a crash in the AArch64 Instruction Selection pass when building x264's encoder/analyse.c with `--target=aarch64-grtev4-linux-gnu -mcpu=neoverse-n1` that root-causes to this revision.  It includes this error message:

  fatal error: error in backend: Cannot select: 0x4927f6931c0: f32 = fmaxnum_ieee 0x4927f67b070, 0x4927c7197e0
    0x4927f67b070: f32,ch = CopyFromReg 0x4927ee820e0, Register:f32 %9
      0x4927f68bb60: f32 = Register %9
    0x4927c7197e0: f32,ch = CopyFromReg 0x4927ee820e0, Register:f32 %10
      0x4927f3f0230: f32 = Register %10
  In function: x264_10_weights_analyse

We're working on getting a standalone reproducer.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D155267/new/

https://reviews.llvm.org/D155267