[llvm] [MachineScheduler][AMDGPU] Allow scheduling of single-MI regions (PR #128739)

[Jay Foad via llvm-commits]

================
@@ -759,7 +759,10 @@ GCNScheduleDAGMILive::GCNScheduleDAGMILive(
       MFI(*MF.getInfo<SIMachineFunctionInfo>()),
       StartingOccupancy(MFI.getOccupancy()), MinOccupancy(StartingOccupancy),
       RegionLiveOuts(this, /*IsLiveOut=*/true) {
-
+  // We want regions with a single MI to be scheduled so that we can reason on
+  // them correctlt during scheduling stages that move MIs between regions (e.g.
----------------
jayfoad wrote:

```suggestion
  // them correctly during scheduling stages that move MIs between regions (e.g.
```

https://github.com/llvm/llvm-project/pull/128739