Sender: nate AT cartsys DOT com
Message-ID: <36BF8A49.9DCCFC21@cartsys.com>
Date: Mon, 08 Feb 1999 17:07:21 -0800
From: Nate Eldredge <nate AT cartsys DOT com>
X-Mailer: Mozilla 4.08 [en] (X11; I; Linux 2.0.36 i586)
MIME-Version: 1.0
To: djgpp AT delorie DOT com
Subject: Re: Is that bug ?
References: <Pine DOT SUN DOT 3 DOT 91 DOT 990207114524 DOT 15342L-100000 AT is>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Reply-To: djgpp AT delorie DOT com

Eli Zaretskii wrote:
> 
> On Sat, 6 Feb 1999, Paul Derbyshire wrote:
> 
> > >Under DOS, Unix that program is crashed, BUT
> > >under Windows that program won't crash.
> > >What is it ? Is that bug ?
> >
> > Yep. It is. It's been known about for quite some time. Unfortunately, Bill
> > Gates seems disinclined to take a few minutes to fix it even while he's
> > giving away 4 billion dollars to various charitable organizations.
> > :-)
> 
> Somebody whose opinions I trust told me that if Windows included null
> page protection, gobs of Windows own code would immediately begin
> crashing.  So there might be some reason why MS doesn't fix that.

But, if I understand correctly, the only thing they need to supply is
the DPMI Set Page Attributes call (or something of the sort).  It's up
to the process to actually call it to unmap the null page, and so the
kernel wouldn't have to.

But who ever said Microsoft thought rationally?
 
> (In case you didn't know: when Windows starts up, its kernel loads into
> the system VM as a DPMI client.  Imagine what would happen with null page
> protection...)

I'm half terrified and half amused, but not at all surprised. :-o
-- 

Nate Eldredge
nate AT cartsys DOT com