This is the mail archive of the cygwin mailing list for the Cygwin project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: STC for libapr1 failure


On Aug 26 13:15, Corinna Vinschen wrote:
> On Aug 25 17:39, David Rothenberger wrote:
> > For a while now, the test cases that come with libapr1 have been
> > bombing with this message:
> > 
> >   *** fatal error - NtCreateEvent(lock): 0xC0000035
> > 
> > I finally took some time to investigate and have extracted a STC
> > that demonstrates the problem.
> 
> Thanks a lot for the testcase.  In theory, the NtCreateEvent call should
> not have happened at all, since it's called under lock, and the code
> around that should have made sure that the object doesn't exist at the
> time.
> 
> After a few hours of extrem puzzlement, I now finally know what happens.
> It's kinda hard to explain.
> 
> A lock on a file is represented by an event object.  Process A holds the
> lock corresponding with event a.  Process B tries to lock, but the lock
> of process A blocks that.  So B now waits for event a, until it gets
> signalled.  Now A unlocks, thus signalling event a and closing the handle
> afterwards.  But A's time slice isn't up yet, so it tries again to lock
> the file, before B returned from the wait for a.  And here a wrong
> condition fails to recognize the situation.  It finds the event object,
> but since it's recognized as "that's me", it doesn't treat the event as
> a blocking factor.  This in turn is the allowance to create its own lock
> event object.  However, the object still exists, since b has still an
> open handle to it.  So creating the event fails, and rightfully so.
> 
> What I don't have is an idea how to fix this problem correctly.  I have
> to think about that.  Stay tuned.

Please test the latest snapshot.  It should fix this problem, as well as
a starvation problem with signals (and, fwiw, thread cancel events) in
flock, lockf, and POSIX fcntl locks.

Thanks again for the testcase.  It was very helpful to test both problems.


Corinna

-- 
Corinna Vinschen                  Please, send mails regarding Cygwin to
Cygwin Project Co-Leader          cygwin AT cygwin DOT com
Red Hat

--
Problem reports:       http://cygwin.com/problems.html
FAQ:                   http://cygwin.com/faq/
Documentation:         http://cygwin.com/docs.html
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]