From: sandmann AT clio DOT rice DOT edu (Charles Sandmann)
Message-Id: <10301060315.AA16609@clio.rice.edu>
Subject: Re: your mail
To: djgpp-workers AT delorie DOT com
Date: Sun, 5 Jan 2003 21:15:08 -0600 (CST)
Cc: bdeck AT lycos DOT co DOT uk
In-Reply-To: <1041802018016510@lycos-europe.com> from "Ben Decker" at Jan 05, 2003 10:26:58 PM
X-Mailer: ELM [version 2.5 PL2]
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
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

> > > But (at least this version) needs a second stack (probably for keeping track
> > > of mutex, i don't know...). DJGPP has no sigstack or sigaltstack.
> > > 
> > > Is this something doable, or should I just stop now and not waste my time.

Take a look at go32/dpmiexcp.c and go32/exceptn.S

The documentation I have for sigstack looks like it would be fairly 
straightforward to implement - the current exception stack is declared
in exceptn.S 

> I would like to give it a shot, that is if no one is in too big of a hurry.
> I downloaded the 2.0.4 sources (for the first time) and had a look.

Via ZIP or CVS?  

> Where should I send my work?

Infrequent contributors typically send patches or modules to the list
for review, then someone will commit them to CVS for them.

Frequent contributors end up with CVS write access after a while, get
code review from the list but commit themselves.

By the way, if you do some restructuring in exceptn.S, you can move the
data symbols out of the .text section to make the DXEs work better :-)