[
https://issues.apache.org/jira/browse/HTTPCLIENT-726?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Roland Weber resolved HTTPCLIENT-726.
-------------------------------------
Resolution: Fixed
The only (a)wait is in WaitingThread, the spurious wakeups are handled in
ConnPoolByRoute. I've implemented a testcase, and the Clover counts indicated
that I even caught one real spurious wakeup in addition to the simulated ones
:-)
For the curious: in ConnPoolByRoute.getEntry(...) is the following code block:
if (waitingThread == null) {
waitingThread =
newWaitingThread(poolLock.newCondition(), rospl);
}
The condition is 'false' only if a spurious wakeup is being handled. The
testcase generates three simulated ones. If the count indicates that the block
has been skipped more often, you know that a real spurious wakeup has occurred.
cheers,
Roland
> review TSCCM for spurious wakeups
> ---------------------------------
>
> Key: HTTPCLIENT-726
> URL: https://issues.apache.org/jira/browse/HTTPCLIENT-726
> Project: HttpComponents HttpClient
> Issue Type: Improvement
> Components: HttpConn
> Reporter: Roland Weber
> Assignee: Roland Weber
> Fix For: 4.0 Alpha 3
>
>
> Review the code of the TSCCM/ConnPoolByRoute for places where spurious
> wakeups may happen.
> Verify that this case is dealt with correctly. Unit test by giving invalid
> wakeup signals?
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
