Github user viirya commented on a diff in the pull request:
https://github.com/apache/spark/pull/1418#discussion_r15592346
--- Diff: core/src/main/scala/org/apache/spark/scheduler/DAGScheduler.scala
---
@@ -195,11 +195,21 @@ class DAGScheduler(
shuffleToMapStage.get(shuffleDep.shuffleId) match {
case Some(stage) => stage
case None =>
+ val parentsWithNoMapStage =
getParentShuffleDependencies(shuffleDep.rdd)
+ while (!parentsWithNoMapStage.isEmpty) {
+ val currentShufDep = parentsWithNoMapStage.pop()
+ val stage =
+ newOrUsedStage(
+ currentShufDep.rdd, currentShufDep.rdd.partitions.size,
currentShufDep, jobId,
+ currentShufDep.rdd.creationSite)
+ shuffleToMapStage(currentShufDep.shuffleId) = stage
+ }
--- End diff --
Yes. There is a recursive calling chain formed by `getShuffleMapStage`
calling `newOrUsedStage` that calls `newStage` that calls `getParentStages`
that calls `getShuffleMapStage` again. The calling chain is going to fill
`shuffleToMapStage` for parent map stages. Here I use an iterative approach
instead.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
