From: sandmann AT clio DOT rice DOT edu (Charles Sandmann)
Message-Id: <10004261328.AA14717@clio.rice.edu>
Subject: Re: The new cwsdpmi
To: eliz AT is DOT elta DOT co DOT il
Date: Wed, 26 Apr 2000 08:28:26 -0500 (CDT)
Cc: ams AT ludd DOT luth DOT se, djgpp-workers AT delorie DOT com
In-Reply-To: <200004261104.HAA23662@indy.delorie.com> from "Eli Zaretskii" at Apr 26, 2000 07:04:20 AM
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

> This ``something used up the disk space in the meantime'' scenario is
> quite possible in the case at hand, since its TMPDIR points to the
> disk (as the amount of physical RAM is very low).

But r5 and r4 should use the same amount of RAM and disk, which means
there is probably some wiggly left - probably in the paging stuff 
especially when nested.  I also changed compilers - so the possibility
of a compiler bug in 32-bit arithmetic is also there (seen it before...)

> FWIW, CWSDPMI r5 is working for me with no visible problems (P166,
> 64MB of RAM, pleanty of disk space, DOS 5.0, QEMM 8).

It could be a tight memory issue.  A 386 issue. A Himem vs vcpi issue.
Hard to tell at this point.