otterc commented on a change in pull request #30691: URL: https://github.com/apache/spark/pull/30691#discussion_r626227795
########## File path: core/src/main/scala/org/apache/spark/scheduler/DAGScheduler.scala ########## @@ -689,7 +716,7 @@ private[spark] class DAGScheduler( dep match { case shufDep: ShuffleDependency[_, _, _] => val mapStage = getOrCreateShuffleMapStage(shufDep, stage.firstJobId) - if (!mapStage.isAvailable) { + if (!mapStage.isAvailable || !mapStage.isMergeFinalized) { Review comment: The comment for `isMergeFinalized` is about what the method is about right. My suggestion is to add a clarifying comment here that why we are add a non merge-finalized stage to missing set -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org