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

[Sameer Sahasrabuddhe via llvm-commits]

================
@@ -249,7 +376,7 @@ cases. These are just examples and not meant to be an exhaustive list.
    Here, `X` can be eliminated.
 
 2. A `wait.asyncmark` which sees an empty sequence of asyncmarks along every
-   path that reaches it.
+   path that reaches it, for every stage it covers.
----------------
ssahasra wrote:

Same comment. Stages are independent.

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