[
https://issues.apache.org/jira/browse/FLINK-4894?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15615213#comment-15615213
]
ASF GitHub Bot commented on FLINK-4894:
---------------------------------------
GitHub user uce opened a pull request:
https://github.com/apache/flink/pull/2716
Fix for [FLINK-4894] [network] Don't request buffer after writing to
partition
This re-opens #2690.
After merging, we noticed that the test were broken by the changed
`hasData` check. This is reverted here and the `clearCurrentBuffer` has been
updated to correctly reset the position and limit.
I'm mainly re-opening this to wait for Travis. In release-1.1 it's already
fixed. If someone has spare time and wants to look at the changes again, feel
free to do so.
You can merge this pull request into a Git repository by running:
$ git pull https://github.com/uce/flink fix_record_serializer
Alternatively you can review and apply these changes as the patch at:
https://github.com/apache/flink/pull/2716.patch
To close this pull request, make a commit to your master/trunk branch
with (at least) the following in the commit message:
This closes #2716
----
commit aaee4a9d889b76cd3d9e63912c5481ed1ce89140
Author: Ufuk Celebi <[email protected]>
Date: 2016-10-24T16:01:28Z
[FLINK-4894] [network] Don't request buffer after writing to partition
After emitting a record via the RecordWriter, we eagerly requested
a new buffer for the next emit on that channel (although it's not clear
that we will immediately need it). With this change, we request that
buffer lazily when an emit call requires it.
This closes #2690.
commit 124ab2caf467dca882d4924beba7586d5f8cfbee
Author: Ufuk Celebi <[email protected]>
Date: 2016-10-28T09:31:44Z
[FLINK-4894] [network] Fix hasData() and correctly clear buffer state
----
> Don't block on buffer request after broadcastEvent
> ---------------------------------------------------
>
> Key: FLINK-4894
> URL: https://issues.apache.org/jira/browse/FLINK-4894
> Project: Flink
> Issue Type: Improvement
> Reporter: Ufuk Celebi
> Assignee: Ufuk Celebi
> Fix For: 1.2.0, 1.1.4
>
>
> After broadcasting an event (like the checkpoint barrier), the record writer
> might block on a buffer request although that buffer will only be needed on
> the next write on that channel.
> Instead of assuming that each serializer has a buffer set, we can change the
> logic in the writer to request the buffer when it requires one.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
