[Author Prev][Author Next][Thread Prev][Thread Next][Author Index][Thread Index]

Re: [tor-bugs] #9682 [Tor]: Better work queue implementation for cpuworkers

#9682: Better work queue implementation for cpuworkers
-------------------------+-------------------------------------------------
     Reporter:  nickm    |      Owner:
         Type:           |     Status:  needs_review
  enhancement            |  Milestone:  Tor: 0.2.6.x-final
     Priority:  major    |    Version:
    Component:  Tor      |   Keywords:  tor-relay, performance, cpuworker,
   Resolution:           |  026-triaged-0 nickm-patch
Actual Points:           |  Parent ID:
       Points:           |
-------------------------+-------------------------------------------------

Comment (by dgoulet):

 Ok so as discussed on IRC, I've pushed to branch {{{better_workqueue_v3}}}
 in my repo the support for {{{pthread_cond_timedwait()}}} to use a
 monotonic clock so time adjustment will not impact the timeout value.

 There are also two commits, first one fixes Copyright years in the tests
 and second one is a comments/whitespaces fix. Please squash or ignore them
 at will.

 1) in cpuworker.c, {{{workqueue_entry_cancel()}}} calls tor_free on the
 workqueue entry but I think it should call {{{workqueue_entry_free}}} so
 it's properly cleaned.

 The rest lgtm! Big feature so I'll make sure to test it quite a bit in
 chutney.

--
Ticket URL: <https://trac.torproject.org/projects/tor/ticket/9682#comment:23>
Tor Bug Tracker & Wiki <https://trac.torproject.org/>
The Tor Project: anonymity online
_______________________________________________
tor-bugs mailing list
tor-bugs@xxxxxxxxxxxxxxxxxxxx
https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-bugs