[PATCH] D124843: AMDGPU: Add G_AMDGPU_MAD_64_32 instructions

[Matt Arsenault via Phabricator via llvm-commits]

arsenm added inline comments.


================
Comment at: llvm/lib/Target/AMDGPU/AMDGPUInstructionSelector.cpp:3168
+  case AMDGPU::G_AMDGPU_MAD_I64_I32:
+    return selectG_AMDGPU_MAD_64_32(I);
   case TargetOpcode::G_INTTOPTR:
----------------
Don't the selector patterns work? This one should work without much fuss


================
Comment at: llvm/lib/Target/AMDGPU/AMDGPURegisterBankInfo.cpp:1624
 
+bool AMDGPURegisterBankInfo::applyMappingMAD_64_32(
+    const OperandsMapper &OpdMapper) const {
----------------
I would assume we form this after regbank select and don't need to legalize it based on the regbank


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D124843/new/

https://reviews.llvm.org/D124843