X-Authentication-Warning: delorie.com: mailnull set sender to djgpp-workers-bounces using -f
Date: Wed, 15 May 2002 00:45:58 +0300
From: "Eli Zaretskii" <eliz AT is DOT elta DOT co DOT il>
Sender: halo1 AT zahav DOT net DOT il
To: lauras AT softhome DOT net
Message-Id: <8296-Wed15May2002004558+0300-eliz@is.elta.co.il>
X-Mailer: emacs 21.2.50 (via feedmail 8 I) and Blat ver 1.8.9
CC: djgpp-workers AT delorie DOT com
In-reply-to: <18721621770.20020514230156@softhome.net> (message from Laurynas
	Biveinis on Tue, 14 May 2002 23:01:56 +0100)
Subject: Re: emacs under w2k
References: <Pine DOT SUN DOT 3 DOT 91 DOT 1020512080610 DOT 329D-100000 AT is>
 <1997943472 DOT 20020514191357 AT softhome DOT net>
 <6480-Tue14May2002211843+0300-eliz AT is DOT elta DOT co DOT il> <18721621770 DOT 20020514230156 AT softhome DOT net>
Reply-To: djgpp-workers AT delorie DOT com
Errors-To: nobody AT delorie DOT com
X-Mailing-List: djgpp-workers AT delorie DOT com
X-Unsubscribes-To: listserv AT delorie DOT com
Precedence: bulk

> Date: Tue, 14 May 2002 23:01:56 +0100
> From: Laurynas Biveinis <lauras AT softhome DOT net>
> 
> Below is the backtrace at the entry point of map_char_table where it
> causes abort later somewhere in that function. I've dumped Lisp_Objects
> there too, but I don't understand anything there. Maybe that says
> something to you?

It says a lot (keymaps are large data structures, since they bind
many keys), but it doesn't say the most important part: where does it
crash, exactly?

AFAICS, map_char_table is a straightforward code, albeit slightly
recursive one (but the recursion depth seems to be limited at 3, so
it shouldn't be a problem).  Sorry, I still don't see the light.

> Where should I look now?

Try to establish what instruction causes Emacs to be killed.

Once you find the source line that crashes, it might be a good idea to
put a call to abort() before and after it, and run the same sequence
of command outside GDB.  That way, you will know that the crashes you
see under GDB are the same ones you experience in normal usage.
Tracking such bugs could get tricky, so it's good to know up front we
aren't chasing a wild goose...

> I wish I could understand what's going on, but I'm not even an average
> Emacs user, and now I feel something like debugging an OS kernel :)

Well, I see that you've studied etc/DEBUG and use its techniques
successfully, so you have nothing to be ashamed of.  Thanks for
working on this.