From:	sandmann AT clio DOT rice DOT edu (Charles Sandmann)
Message-Id:	<10301121803.AA21102@clio.rice.edu>
Subject:	Re: Quirk with command.com shell on XP
To:	eliz AT is DOT elta DOT co DOT il (Eli Zaretskii)
Date:	Sun, 12 Jan 2003 12:03:33 -0600 (CST)
Cc:	djgpp-workers AT delorie DOT com
In-Reply-To:	<Pine.SUN.3.91.1030112074010.29737A-100000@is> from "Eli Zaretskii" at Jan 12, 2003 07:41:13 AM
X-Mailer:	ELM [version 2.5 PL2]
Mime-Version:	1.0
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

> > I use CMD.EXE exclusively (whenever I've had to use command.com I've found
> > some problem that needed to be fixed anyway) - and I've never seen any 
> > problems.  Given the enhanced capabilities of CMD.EXE, why not recommend
> > using it instead?
> 
> Doesn't Windows 2000/XP invoke command.com when a DOS program is 
> launched, even if your default shell is cmd.exe?

For any DOS type application, Win NT/2K/XP create a separate process
NTVDM which contains the virtual DOS machine - and inside that virtual
machine it always has command.com loaded (in the DOS machine).  This
is what would be expected under DOS - command.com is loaded at "boot"
time.  The first time you run a DOS application there is the overhead
of creating the VDM - but then it stays present in the background as
long as CMD.EXE (or whatever created it) hangs around.  This allows
the second DOS application to run more quickly.

Now, if you execute command.com, what happens is you get a second
command.com started in the VDM (similar to what happens if you 
execute command.com under DOS - when it's already loaded).

- Raw text -