[all-commits] [llvm/llvm-project] 006423: [mlir][SPIR-V] Lower arith.maxnumf/minnumf to spir...

[Arseniy Obolenskiy via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 006423944b8efdb115d8fd001cffca888605cd7f
      https://github.com/llvm/llvm-project/commit/006423944b8efdb115d8fd001cffca888605cd7f
  Author: Arseniy Obolenskiy <arseniy.obolenskiy at amd.com>
  Date:   2026-07-08 (Wed, 08 Jul 2026)

  Changed paths:
    M mlir/lib/Conversion/ArithToSPIRV/ArithToSPIRV.cpp
    M mlir/test/Conversion/ArithToSPIRV/arith-to-spirv.mlir
    M mlir/test/Conversion/ArithToSPIRV/fast-math.mlir

  Log Message:
  -----------
  [mlir][SPIR-V] Lower arith.maxnumf/minnumf to spirv.GL.NMax/NMin (#205975)

The previous lowering targeted spirv.GL.FMax/FMin + NaN guards

spirv.GL.NMax/NMin natively treat NaN as missing, matching
arith.maxnumf/minnumf exactly (no guards required)



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