Mailing-List: contact cygwin-help AT cygwin DOT com; run by ezmlm
List-Subscribe: <mailto:cygwin-subscribe AT cygwin DOT com>
List-Archive: <http://sourceware.org/ml/cygwin/>
List-Post: <mailto:cygwin AT cygwin DOT com>
List-Help: <mailto:cygwin-help AT cygwin DOT com>, <http://sourceware.org/ml/#faqs>
Sender: cygwin-owner AT cygwin DOT com
Mail-Followup-To: cygwin AT cygwin DOT com
Delivered-To: mailing list cygwin AT cygwin DOT com
To: cygwin AT cygwin DOT com
From: Larry Adams <larryjadams AT comcast DOT net>
Subject:  Re: Segfault in Cactid
Date: Fri, 2 Sep 2005 09:47:34 +0000 (UTC)
Lines: 52
Message-ID:  <loom.20050902T113935-922@post.gmane.org>
References:  <loom DOT 20050902T024037-588 AT post DOT gmane DOT org> <4317D378 DOT 236A02D9 AT dessent DOT net>
Mime-Version:  1.0
Content-Type:  text/plain; charset=us-ascii
Content-Transfer-Encoding:  7bit
User-Agent: Loom/3.14 (http://gmane.org/)
X-IsSubscribed: yes

Brian Dessent <brian <at> dessent.net> writes:

> 
> Larry Adams wrote:
> 
> > So my questions are: How do I trace the memory location above to a known 
system
> > call using gdb?  When, I attach to my hung process, what should I be doing 
to
> > determine the calling/offending function.
> 
> Build a cygwin1.dll with debug symbols and then use addr2line.  Or use
> strace.  At the very least you should try the latest snapshot to see if
> it fixes your problem.
> 

Brian, I am assuming that I can just download the latest cygwin of winsup and 
then configure/make.  What configure/make options should I be using to 
incorporate debug symbols?

> The fact that you get that popup is a little puzzling.  That popup is
> the result of an SEH exception not being handled by the process and
> falling through to the system default handler.  But Cygwin installs its
> own SEH handlers so that it can convert those sort of exceptions into
> "signal 11" without the popup.  Normally you would not get a popup and
> it would terminate as on unix with a segmentation fault message and
> stackdump.  But I guess if something internal to Cygwin is getting
> corrupted then all bets are off.

The previous version of Cactid did not have this problem.  The one major 
change in this version of Cactid is that I am using upto 100 process threads 
and each thread is allocating large amounts of memory using primarily 
the "calloc" function.  I was wondering if maybe Cygwin is having difficulty 
making memory available to the calloc function after several threads have 
chopped up the heap.

If that is the case, are there other alternatives?  I could use a linked list 
so that memory fragmentation would occur in smaller chunks, or maybe there are 
some house keeping functions I can utilize to free up some contiguous blocks 
of memory or some way to increase the amount of memory available for the heap.

Your further comments would be very appreciated.

Larry

> 
> Brian
> 
> 





--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Problem reports:       http://cygwin.com/problems.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/