[llvm-branch-commits] [clang] [llvm] [AMDGPU] Introduce asyncmark/wait intrinsics (PR #180467)

[Pierre van Houtryve via llvm-branch-commits]

================
@@ -710,6 +710,24 @@ class SIInsertWaitcnts {
   std::optional<WaitEventType>
   getExpertSchedulingEventType(const MachineInstr &Inst) const;
 
+  bool isAsync(const MachineInstr &MI) const {
----------------
Pierre-vh wrote:

nit: I think those 3 functions should go in `TII` directly

https://github.com/llvm/llvm-project/pull/180467