[all-commits] [llvm/llvm-project] fa46fe: [AMDGPU] Use V_FMA_MIX* more often

[Mateja Marjanovic via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: fa46feb3148133f4fd540c6d8f912d7980a27515
      https://github.com/llvm/llvm-project/commit/fa46feb3148133f4fd540c6d8f912d7980a27515
  Author: Mateja Marjanovic <mmarjano at amd.com>
  Date:   2023-07-13 (Thu, 13 Jul 2023)

  Changed paths:
    M llvm/lib/Target/AMDGPU/VOP3PInstructions.td
    M llvm/test/CodeGen/AMDGPU/GlobalISel/fdiv.f16.ll
    M llvm/test/CodeGen/AMDGPU/fdiv.f16.ll
    M llvm/test/CodeGen/AMDGPU/frem.ll
    M llvm/test/CodeGen/AMDGPU/mad-mix-lo.ll

  Log Message:
  -----------
  [AMDGPU] Use V_FMA_MIX* more often

Combine mul (f32) + fptrunc (f32->f16) to "v_fma_mixlo_f16 mulSrc1, mulSrc2, 0".

Differential Revision: https://reviews.llvm.org/D153544
Reviewers: arsenm, foad