Re: [ANNOUNCEMENT] Updated: emacs-24.3.93-3 [TEST]

[Ken Brown <kbrown at cornell dot edu>]

On 10/17/2014 2:45 PM, Jon TURNEY wrote:
> On 17/10/2014 14:53, Ken Brown wrote:
> > If this turns out not to be the issue, then I think you would need an
> > unoptimized build for further debugging.  I can provide one for you.
>
> If you could, that would be great.

OK, I've built emacs-24.4-0 (based on a release candidate for 
emacs-24.4), which you can get from my personal Cygwin repository:

  http://sanibeltranquility.com/cygwin/

There are instructions at that URL.  This build has no optimization and 
also has some extra checking enabled: There are assertions sprinkled 
throughout the code that will trigger a call to emacs_abort if they fail.

The release includes a (slightly modified) .gdbinit file in 
/usr/share/emacs/24.4/etc, so you can source that in gdb.  (The 
modification consists of disabling the feature that automatically 
produces a lisp backtrace whenever you ask for a backtrace, because that 
feature sometimes doesn't work well on Cygwin.  But you can still use 
the 'xbacktrace' command to get a lisp backtrace.)  See 
/usr/share/emacs/24.4/etc/DEBUG for further information about the 
commands defined in .gdbinit.

The next time you get a crash, I suggest you report it as an emacs bug 
(M-x report-emacs-bug), send a backtrace, and keep the gdb session 
running.  There are people who read the bug reports, especially Eli 
Zaretskii, who are much better at debugging emacs than I am.  And I 
subscribe to the bug list also, so I can chime in if there's something 
Cygwin-specific that I can add.

Thanks for offering to help debug this.

Ken

--
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