[
https://issues.apache.org/jira/browse/PROTON-1408?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15941825#comment-15941825
]
ASF GitHub Bot commented on PROTON-1408:
----------------------------------------
Github user mgoulish commented on the issue:
https://github.com/apache/qpid-proton/pull/99
I have updated to use Andrew's cleaner sequence_cmp() idea, and added a new
change: preventing sign-extension from pn_sequence_t to uintptr_t in the calls
to pn_hash_put() and pn_hash_get(). This sign extension appears to be
happening in one pathway but not the other, so that , when we wrap around to
negative p_sequence_t values, pn_hash_get() and pn_hash_put() are getting
different hash values.
I know how this sounds -- but with this change my test got through the 2^31
barrier for the first time ever.
It sounds like heresy -- but -- eppur si muove.
This is probably not the end of the story -- I had runaway memory
consumption later, but the test got to 2.25 billion messages before I had to
kill it. Well past the 2^32 barrier, which == 2.147 billion.
> long-lived connections suffer large performance hit after many messages
> -----------------------------------------------------------------------
>
> Key: PROTON-1408
> URL: https://issues.apache.org/jira/browse/PROTON-1408
> Project: Qpid Proton
> Issue Type: Bug
> Components: proton-c
> Reporter: michael goulish
> Assignee: Alan Conway
> Attachments: jira_proton_1408_reproducer.tar.gz
>
>
> In long-running soak tests, in which connections are never taken down, I am
> seeing a sudden & severe performance degradation when the number of messages
> over the connection reaches about 6.4 billion.
> This is happening in tests with two senders, two receivers & one router
> intermediating.
> I have tried C libUV clients as well as CPP clients. Behavior is not
> identical, but I see sudden performance drop, ie. 8x throughput decrease or
> worse, in both cases.
> Alan / Ted / Ken see an issue in use of improper comparison logic in
> pn_do_disposition(), in transport.c . I am trying to prove this now.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
