[llvm] [AMDGPU][MachineScheduler] Alternative way to control excess RP. (PR #68004)

[via llvm-commits]

================
@@ -894,10 +894,24 @@ void GCNSchedStage::setupNewBlock() {
 
 void GCNSchedStage::finalizeGCNRegion() {
   DAG.Regions[RegionIdx] = std::pair(DAG.RegionBegin, DAG.RegionEnd);
-  DAG.RescheduleRegions[RegionIdx] = false;
-  if (S.HasHighPressure)
+  PressureAfter = DAG.getRealRegPressure(RegionIdx);
+
+  unsigned NewVGPRRP = PressureAfter.getVGPRNum(false);
+  unsigned NewAGPRRP = PressureAfter.getAGPRNum();
+  unsigned NewSGPRRP = PressureAfter.getSGPRNum();
+
+  if ((NewVGPRRP >= S.VGPRCriticalLimit - S.VGPRExcessMargin) ||
+      (NewAGPRRP >= S.VGPRCriticalLimit - S.VGPRExcessMargin) ||
+      (NewSGPRRP >= S.SGPRCriticalLimit - S.SGPRExcessMargin))
     DAG.RegionsWithHighRP[RegionIdx] = true;
 
+  if ((NewVGPRRP >= S.VGPRExcessLimit - S.VGPRExcessMargin) ||
+      (NewAGPRRP >= S.VGPRExcessLimit - S.SGPRExcessMargin) ||
----------------
alex-t wrote:

Sure, it is just a typo. The whole code is an early draft. I used the pull request as a place for discussing the idea.


https://github.com/llvm/llvm-project/pull/68004