Github user CodingCat commented on a diff in the pull request:
https://github.com/apache/spark/pull/1313#discussion_r15436880
--- Diff:
core/src/main/scala/org/apache/spark/scheduler/TaskSetManager.scala ---
@@ -738,6 +771,8 @@ private[spark] class TaskSetManager(
/**
* Compute the locality levels used in this TaskSet. Assumes that all
tasks have already been
* added to queues using addPendingTask.
+ *
+ * NOTE: don't need to handle NOPREF here, because NOPREF is scheduled
as PROCESS_LOCAL
--- End diff --
En, if we simply take NOPREF as a valid locality, it means that we need to
wait for some time to schedule noPref tasks....now we want to try NOPREF after
we failed to find task on PROCESS and NODE level, and NOPREF task will only be
scheduled after all PROCESS and NODE_LOCAL tasks have been scheduled
---
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.
---
