[llvm] [AMDGPU] Model ordered XDL writes in expert scheduling (PR #218326)

[Austin Kerbow via llvm-commits]

================
@@ -884,14 +903,25 @@ void WaitcntBrackets::updateByEvent(HWEvents E, MachineInstr &Inst) {
     Increment = 2;
   }
   unsigned CurrScore = UB + Increment;
-  if (CurrScore == 0)
+  if (CurrScore <= UB)
----------------
kerbowa wrote:

It was an exiting potential bug since scaled WMMA can increment by 2

https://github.com/llvm/llvm-project/pull/218326