[concurrency-interest] thread queueing qn

Dhanji R. Prasanna dhanji at gmail.com
Mon Jan 21 22:17:33 EST 2008


On 1/22/08, Bob Lee <crazybob at crazybob.org> wrote:
>
> On Jan 21, 2008 6:44 PM, Dhanji R. Prasanna <dhanji at gmail.com> wrote:
>
> > Does this mean longest-waiting threads with be notified first? Given
> > that a request thread is unlikely to reenter a held lock, this should
> > provide me with FIFO semantics should it not?
> >
>
> This may not be an issue for you, but the thread could re-enter if you use
> a request dispatcher.
>

Indeed that has been keeping me up at night, but I'm ignoring it for two
reasons:

- currently, my framework does not honor dispatcher (if I can't see it it's
not there, jk =)
- the request is already bound a conv, so reentering the conv filter will
dilute the lock's queue but should not affect the FIFO semantic with regard
to unique threads (right?)

Thanks everyone for your insights!

Dhanji.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20080122/b877fd1f/attachment.html 


More information about the Concurrency-interest mailing list