From: sandmann AT clio DOT rice DOT edu (Charles Sandmann)
Message-Id: <9610041353.AA12917@clio.rice.edu>
Subject: Re: `abort' in nested programs
To: eliz AT is DOT elta DOT co DOT il (Eli Zaretskii)
Date: Fri, 4 Oct 1996 08:53:28 -0600 (CDT)
Cc: dj AT delorie DOT com, djgpp-workers AT delorie DOT com
In-Reply-To: <Pine.SUN.3.91.961004094230.29098H-100000@is> from "Eli Zaretskii" at Oct 4, 96 09:45:46 am
Content-Type: text
Content-Length: 743       

> > > I understand that dpmiexcp is always linked in, so that shouldn't be a 
> > > problem.  Am I right on this?
> > Actually, this will break things like the TSR code, but I'm sort of used
> > to things breaking between releases anymore.
> 
> Is there any way to do it without breaking that code?

Probably not without modifying that code, since it depends on not needing
to link in the exception handling code.  As the code becomes less and less
modular, these things happen.  I'm not so sure the right idea wasn't
the original one, which was to toggle exceptions/interrupts before/after
you call system/spawn/exec/etc - making the parent defensive to these
things instead of being dependent on the child to clean up after itself.
Oh well.