Re: crash on latest cygwin snapshot

[Corinna Vinschen <corinna-cygwin@cygwin.com>]

On Aug  2 10:53, Ken Brown wrote:
> On 8/1/2012 8:58 AM, Corinna Vinschen wrote:
> >On Aug  1 08:46, Ken Brown wrote:
> >>I never had a reliable way of reproducing the crash (see
> >>http://cygwin.com/ml/cygwin/2012-06/msg00464.html).  It happened
> >>seemingly at random, and very sporadically.  But I have the snapshot
> >>installed and will exercise it as much as I can.
> >
> >Thanks!
> 
> The good news: I haven't seen a repeat of that old crash so far.

That's really good news, thank you.

> Unfortunately, I'm finding that emacs is unstable:  The emacs window
> (running under X) simply disappears after 12-24 hours.  This may not
> have anything to do with the most recent changes.  I haven't yet
> tested any earlier snapshots.
> 
> Testing this is a very slow process, since I don't know how to
> produce the problem; I just have to wait and see if emacs will die.
> 
> I tried to get some information by running emacs under gdb the most
> recent time I started it, but all I got was this:
> 
> (gdb) c
> Continuing.
> [Inferior 1 (process 8196) exited with code 05400]
> (gdb) bt
> No stack
> 
> Does that exit code mean anything to you?  I couldn't find anything
> by googling.

05400 oct is 0xb00 hex.  As a wait(2) compatible exit code, it would
mean a normal exit, not signal-induced, with an exit value of 11.  Which
is kind of suspicious, since 11 is also the signal number of a SEGV.  Of
course, not every SEGV must be related to the problem at hand, but
still...

Does emacs have a signal handler which handles SEGVs?  If so, you might
consider to set a breakpoint on this signal handler in GDB.


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