[all-commits] [llvm/llvm-project] cf80de: [AMDGPU][GFX11] Do not rewrite V_FMA/FMAC_* to V_F...

[Ivan Kosarev via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: cf80defae20b47502fb0921e9358dc90477e9563
      https://github.com/llvm/llvm-project/commit/cf80defae20b47502fb0921e9358dc90477e9563
  Author: Ivan Kosarev <ivan.kosarev at amd.com>
  Date:   2023-10-04 (Wed, 04 Oct 2023)

  Changed paths:
    M llvm/lib/Target/AMDGPU/SIInstrInfo.cpp
    M llvm/test/CodeGen/AMDGPU/fma.f16.ll

  Log Message:
  -----------
  [AMDGPU][GFX11] Do not rewrite V_FMA/FMAC_* to V_FMAAK_F16_t16 on operand legalization. (#66202)

V_FMAAK_F16_t16 takes VGPR_32_Lo128 operands whereas the original
instructions would have VGPR_32 operands. Switching the opcodes without
updating operands' register classes leads to MachineVerifier complaining
about the classes not matching instruction definitions. The problem only
reveals itself of builds with expensive checks enabled because of
missing -verify-machineinstrs in the test.

This is the third attempt to update CodeGen/AMDGPU/fma.f16.ll to run for
GFX11, following the second attempt in a1e38e0b8e3e, partially reverted
in eaf737a4e004.