[llvm] [AMDGPU] Support arbitrary types in amdgcn.dead (PR #134841)

[Jay Foad via llvm-commits]

================
@@ -6629,6 +6629,11 @@ void SITargetLowering::ReplaceNodeResults(SDNode *N,
       Results.push_back(LoadVal);
       return;
     }
+    case Intrinsic::amdgcn_dead: {
----------------
jayfoad wrote:

Why do you need to handle this both here in `ReplaceNodeResults` and in `LowerINTRINSIC_WO_CHAIN` below?

https://github.com/llvm/llvm-project/pull/134841