[tor-bugs] #16644 [Tor]: Unpredictable scheduling behavior of threads

["Tor Bug Tracker & Wiki" <blackhole@xxxxxxxxxxxxxx>]

#16644: Unpredictable scheduling behavior of threads
-------------------------+------------------------------------
 Reporter:  cypherpunks  |          Owner:
     Type:  defect       |         Status:  new
 Priority:  minor        |      Milestone:
Component:  Tor          |        Version:  Tor: 0.2.6.3-alpha
 Keywords:               |  Actual Points:
Parent ID:               |         Points:
-------------------------+------------------------------------
 The
 [http://pubs.opengroup.org/onlinepubs/9699919799/functions/pthread_cond_broadcast.html
 POSIX standard] states
 >The pthread_cond_broadcast() or pthread_cond_signal() functions may be
 called by a thread whether or not it currently owns the mutex that threads
 calling pthread_cond_wait() or pthread_cond_timedwait() have associated
 with the condition variable during their waits; '''''however, if
 predictable scheduling behavior is required, then that mutex shall be
 locked by the thread calling pthread_cond_broadcast() or
 pthread_cond_signal()'''''.

 The highlighted recommendation isn't followed in by some thread pool
 functions which release the mutex before unblocking the threads. The issue
 was found using Helgrind (a Valgrind tool) which gave the warning `Thread
 #1: pthread_cond_{signal,broadcast}: dubious: associated lock is not held
 by any thread`

 It is worth mentioning that the thread tests do follow the recommendation.

--
Ticket URL: <https://trac.torproject.org/projects/tor/ticket/16644>
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