[
https://issues.apache.org/jira/browse/PROTON-1771?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16439749#comment-16439749
]
ASF subversion and git services commented on PROTON-1771:
---------------------------------------------------------
Commit fc1df0551947c357e9fa0bf9da4b836c5a97c11f in qpid-proton's branch
refs/heads/master from [~aconway]
[ https://git-wip-us.apache.org/repos/asf?p=qpid-proton.git;h=fc1df05 ]
Revert "PROTON-1771: [c] locking around epoll_extended_t"
This reverts commit 188ce28066df8f5e965fb63593f419f49c950760.
The fix caused hangs due to mutex deadlocks.
> [c-proactor] multi-thread race test for proactor
> ------------------------------------------------
>
> Key: PROTON-1771
> URL: https://issues.apache.org/jira/browse/PROTON-1771
> Project: Qpid Proton
> Issue Type: Test
> Components: proton-c
> Affects Versions: proton-c-0.20.0
> Reporter: Alan Conway
> Assignee: Alan Conway
> Priority: Major
> Fix For: proton-c-0.23.0
>
>
> Crate a new test exe that runs for a (configurable, default short) period of
> time, with a single proactor acted on by multiple proactor and user threads.
> Run
> with helgrind or tsan to detect races.
> Exercise potentially racy APIs concurrently:
> - making, accepting and closing (from both ends) a connection.
> - pn_connection_wake
> - pn_proactor_release_connection
> - re-use of released pn_connection_t on a new connection
> - timeout
> - concurrent with some normal use: sending/receiving messages.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
