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

via libc-commits libc-commits at lists.llvm.org
Tue Oct 24 05:35:32 PDT 2023


================
@@ -702,7 +702,7 @@ bool UnclusteredHighRPStage::initGCNSchedStage() {
   if (!GCNSchedStage::initGCNSchedStage())
     return false;
 
-  if (DAG.RegionsWithHighRP.none() && DAG.RegionsWithExcessRP.none())
+  if (DAG.RegionsWithExcessRP.none())
----------------
alex-t wrote:

This change was intended to be the preliminary for the balanced scheduling model.
Maybe it is better to join it with the large change or revert the line in question.
Nevertheless, I would like to see if this affects the performance.

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


More information about the libc-commits mailing list