[clang] [llvm] [AMDGPU] Introduce asyncmark/wait intrinsics (PR #173259)

[Krzysztof Drewniak via llvm-commits]

================
@@ -368,12 +368,15 @@ enum CPol {
   GLC = 1,
   SLC = 2,
   DLC = 4,
+  SWZ_pregfx12 = 8,
   SCC = 16,
+  ASYNC_pregfx12 = 32,
----------------
krzysz00 wrote:

... AutoIpgrade to add an argument to an intrinsic with a default value is a perfectly fine usage of that facility - we did it just recently for adding `clamp` to WMMA operations, for example 

And having buffer loads working is important, yes

https://github.com/llvm/llvm-project/pull/173259