[clang] [llvm] [AMDGPU] Implement AsyncMark Stages (PR #220442)

[Matt Arsenault via llvm-commits]

================
@@ -1096,9 +1133,8 @@ void WaitcntBrackets::updateByEvent(HWEvents E, MachineInstr &Inst) {
         setVMemScore(LDSDMA_BEGIN + Slot, T, CurrScore);
     }
 
-    if (Context->shouldUpdateAsyncMark(Inst, T)) {
-      AsyncScore[T] = CurrScore;
-    }
+    if (auto Stage = Context->getAsyncStageToUpdate(Inst, T))
----------------
arsenm wrote:

No, it's to only use auto if the type already appears, like in `dyn_cast<Type>`

https://github.com/llvm/llvm-project/pull/220442