[
https://issues.apache.org/jira/browse/FLINK-7456?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16320520#comment-16320520
]
ASF GitHub Bot commented on FLINK-7456:
---------------------------------------
Github user NicoK commented on a diff in the pull request:
https://github.com/apache/flink/pull/4552#discussion_r160714424
--- Diff:
flink-runtime/src/main/java/org/apache/flink/runtime/io/network/netty/PartitionRequestQueue.java
---
@@ -250,10 +304,12 @@ private void handleException(Channel channel,
Throwable cause) throws IOExceptio
private void releaseAllResources() throws IOException {
SequenceNumberingViewReader reader;
- while ((reader = nonEmptyReader.poll()) != null) {
+ while ((reader = availableReaders.poll()) != null) {
--- End diff --
Previously, we did not have access to all views but only those with data,
but shouldn't we release all views instead? Especially now since there are
views with data but with not enough credit.
> Implement Netty sender incoming pipeline for credit-based
> ---------------------------------------------------------
>
> Key: FLINK-7456
> URL: https://issues.apache.org/jira/browse/FLINK-7456
> Project: Flink
> Issue Type: Sub-task
> Components: Network
> Reporter: zhijiang
> Assignee: zhijiang
> Fix For: 1.5.0
>
>
> This is a part of work for credit-based network flow control.
> On sender side, each subpartition view maintains an atomic integer
> {{currentCredit}} from receiver. Once receiving the messages of
> {{PartitionRequest}} and {{AddCredit}}, the {{currentCredit}} is added by
> deltas.
> Each view also maintains an atomic boolean field to mark it as registered
> available for transfer to make sure it is enqueued in handler only once. If
> the {{currentCredit}} increases from zero and there are available buffers in
> the subpartition, the corresponding view will be enqueued for transferring
> data.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
