[
https://issues.apache.org/jira/browse/FLINK-25653?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17487007#comment-17487007
]
Yingjie Cao commented on FLINK-25653:
-------------------------------------
Fixed via:
b562070e025c9448facd1f53ff1b5d34d5c8a14f on master
fdbf60f52e2b2eaf3fdb474b4624d0857f892824 on release-1.14
7012130d49a9305d996d3afe83ef383d9d9c85c4 on release-1.13
> Move buffer recycle in SortMergeSubpartitionReader out of lock to avoid
> deadlock
> --------------------------------------------------------------------------------
>
> Key: FLINK-25653
> URL: https://issues.apache.org/jira/browse/FLINK-25653
> Project: Flink
> Issue Type: Sub-task
> Components: Runtime / Network
> Affects Versions: 1.13.5, 1.14.3
> Reporter: Yingjie Cao
> Assignee: Yuxin Tan
> Priority: Major
> Labels: pull-request-available
> Fix For: 1.15.0, 1.14.4, 1.13.7
>
>
> For the current sort-shuffle implementation, the different lock order in
> SortMergeSubpartitionReader and SortMergeResultPartitionReadScheduler can
> cause deadlock. To solve the problem, we can move buffer recycle in
> SortMergeSubpartitionReader out of the lock.
--
This message was sent by Atlassian Jira
(v8.20.1#820001)
