Mailing-List: contact cygwin-help AT sourceware DOT cygnus DOT com; run by ezmlm
List-Subscribe: <mailto:cygwin-subscribe AT sourceware DOT cygnus DOT com>
List-Archive: <http://sourceware.cygnus.com/ml/cygwin/>
List-Post: <mailto:cygwin AT sourceware DOT cygnus DOT com>
List-Help: <mailto:cygwin-help AT sourceware DOT cygnus DOT com>, <http://sourceware.cygnus.com/ml/#faqs>
Sender: cygwin-owner AT sourceware DOT cygnus DOT com
Delivered-To: mailing list cygwin AT sourceware DOT cygnus DOT com
From: Chris Faylor <cgf AT cygnus DOT com>
Date: Thu, 18 May 2000 00:04:15 -0400
To: cygwin AT sourceware DOT cygnus DOT com
Cc: kcoleman AT yahoo DOT com
Subject: Re: Crash in 1.1.1
Message-ID: <20000518000415.A8166@cygnus.com>
Reply-To: cygwin AT sourceware DOT cygnus DOT com
Mail-Followup-To: cgf AT cygnus DOT com, cygwin AT sourceware DOT cygnus DOT com,
	kcoleman AT yahoo DOT com
References: <20000518035308 DOT 22166 DOT qmail AT web218 DOT mail DOT yahoo DOT com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.1.12i
In-Reply-To: <20000518035308.22166.qmail@web218.mail.yahoo.com>; from kcoleman@yahoo.com on Wed, May 17, 2000 at 08:53:08PM -0700

On Wed, May 17, 2000 at 08:53:08PM -0700, Ken Coleman wrote:
>Ok, I've got some potentially useful information.  Hopefully it will
>help
>
>What I discovered is that the new 1.1.2 candidate is not crashing per
>se, it's page faulting due to stack overflow.  I got a valid stackdump
>file, and by using it and GDB, I was able to see what's going on
>somewhat.  It seems to be infinitely recurring among the following
>three functions:
>
>sigframe::set(sigthread &, int)
>muto::acquire(unsigned long)
>WSFO

Yup.  I just checked in a fix for this.  Hopefully it will show up in
tonight's snapshot, assuming I didn't get it in too late.

I appreciate your debugging of the problem, though.  I couldn't
duplicate this on my Windows 95 system, so, using the addresses that you
supplied in your last email, I had to do a comparison of the assembly
language output from the snapshot versus a version of the DLL with
debugging symbols.  Then, I found that I could easily duplicate the
problem on my NT system...

I should have just waited a little bit and your analysis would have
made everything clear.

If you have access to CVS you can check out my changes and build
a new version of cygwin.  It should work better.

Thanks,
cgf

--
Want to unsubscribe from this list?
Send a message to cygwin-unsubscribe AT sourceware DOT cygnus DOT com