Date:	Tue, 19 Jun 2001 12:32:03 +0300 (IDT)
From:	Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>
X-Sender:	eliz AT is
To:	djgpp AT delorie DOT com
Subject:	Re: Peculiar behavior of program.
In-Reply-To:	<3b2efaa3.496489@news.primus.ca>
Message-ID:	<Pine.SUN.3.91.1010619122203.7390Z-100000@is>
MIME-Version:	1.0
Reply-To:	djgpp AT delorie DOT com
Errors-To:	nobody AT delorie DOT com
X-Mailing-List:	djgpp AT delorie DOT com
X-Unsubscribes-To:	listserv AT delorie DOT com

On Tue, 19 Jun 2001, Graaagh the Mighty wrote:

> I'm trying to debug some kind of memory management problem.
> Problem is, it refuses to give proper traceback information. In
> Windows, it causes a reboot, which SHOULD NEVER HAPPEN in protected
> mode.

Blame Windows: it doesn't protect itself as well as it could have.

> In DOS, it prints a general protection fault message, but no
> "call frame traceback EIPs" which is decidedly odd.

It would be much better to post the crash message here, instead of 
describing it.

I can only guess that either your stack is badly smashed, which makes 
traceback impossible, or the crash message comes from CWSDPMI.

See section 12.2, it shows what the CWSDPMI crash message looks like.

> FSDB segfaults
> itself on running the program; RHIDE simply silently exits to DOS on
> attempting to so much as step into the program. When I modified the
> code so that at certain checkpoints it would open a file, append some
> info to the file, and close it again (so if it died between
> checkpoints the file wouldn't be left truncated, but would be
> readable), and of course the program began to work...*sigh*

Is it possible that you have very large automatic variables?

Anyway, looking at the crash message would probably give a few ideas, so 
please post it.

- Raw text -