[llvm-branch-commits] [llvm] [AMDGPU] Add wave reduce intrinsics for double types - 2 (PR #170812)

[Juan Manuel Martinez Caamaño via llvm-branch-commits]

================
@@ -5959,7 +6008,9 @@ static MachineBasicBlock *lowerWaveReduce(MachineInstr &MI,
         break;
       }
       case AMDGPU::V_MIN_F64_e64:
-      case AMDGPU::V_MAX_F64_e64: {
+      case AMDGPU::V_MAX_F64_e64:
+      case AMDGPU::V_ADD_F64_e64:
----------------
jmmartinez wrote:

Can we use `V_ADD_F64_pseudo_e64` instead and let it get lowered to `V_ADD_F64_e64/V_ADD_F64_e64_gfx12` later ?

https://github.com/llvm/llvm-project/pull/170812