[all-commits] [llvm/llvm-project] 817cd7: [AMDGPU] Fix folding clamp into pseudo scalar inst...

[Mirko Brkušanin via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 817cd726454f01e990cd84e5e1d339b120b5ebaa
      https://github.com/llvm/llvm-project/commit/817cd726454f01e990cd84e5e1d339b120b5ebaa
  Author: Mirko Brkušanin <Mirko.Brkusanin at amd.com>
  Date:   2024-07-25 (Thu, 25 Jul 2024)

  Changed paths:
    M llvm/lib/Target/AMDGPU/SIFoldOperands.cpp
    A llvm/test/CodeGen/AMDGPU/si-fold-scalar-clamp.mir

  Log Message:
  -----------
  [AMDGPU] Fix folding clamp into pseudo scalar instructions (#100568)

Clamp is canonically a v_max* instruction with a VGPR dst. Folding clamp
into a pseudo scalar instruction can cause issues due to a change in
regbank. We fix this with a copy.



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