Date:	Mon, 6 Mar 2000 09:54:09 +0200 (IST)
From:	Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>
X-Sender:	eliz AT is
To:	Robert van der Boon <rjvdboon AT europe DOT com>
cc:	djgpp-workers AT delorie DOT com, Richard Dawe <richdawe AT bigfoot DOT com>
Subject:	Re: PMODE/DJ (was Re: ANNOUNCE: Bison 1.28 ported to DJGPP)
In-Reply-To:	<01a9c2228200530NET088S@hetnet.nl>
Message-ID:	<Pine.SUN.3.91.1000306095337.9921K-100000@is>
MIME-Version:	1.0
Reply-To:	djgpp-workers AT delorie DOT com
Errors-To:	dj-admin AT delorie DOT com
X-Mailing-List:	djgpp-workers AT delorie DOT com
X-Unsubscribes-To:	listserv AT delorie DOT com

On Sun, 5 Mar 2000, Robert van der Boon wrote:

> On the subject of PMODE/DJ: did anyone ever notice that it doesn't work 
> if one start plain msdos, with:
> device=emm386.exe NOEMS NOVCPI
> and then tries to execute a PMODE/DJ executable? I never succeeded to do 
> this.

Why do you need NOVCPI?  Does it work if you remove it?

> For example the new unzip32 as found on simtel/djgpp doesn't work in 
> this case.

I never saw any problems like that, but I do know that unzip32 wedges
the machine cold if it is invoked when CWSDPMI is already in memory
(e.g., if you shell out to DOS from a DJGPP program, then invoke
unzip32).  I reported the problem to Charles Sandmann; hopefully, he
will be able to debug it some day...

In general, use of PMODE/DJ-stubified programs is only advisable when
you *must* make sure the program runs on a bare DOS machine and it is
not reasonable to expect people to install CWSDPMI first.  Unzip32 is
just such a case, since it's supposed to be used to unzip other DJGPP
packages, including CWSDPMI.

In all other cases, my advice is to stay away of PMODE/DJ, since it is
much less debugged and is not actively maintained.

- Raw text -