Message-Id: <200005181129.NAA29628@cerbere.u-strasbg.fr>
To: djgpp AT delorie DOT com
Date: Thu, 18 May 2000 13:11:01 +0200
Newsgroups: comp.os.msdos.djgpp
Subject: Re: win2k
From: muller AT cerbere DOT u-strasbg DOT fr (Pierre Muller)
Organization: Institut Charles Sadron
X-Newsreader: WinVN 0.99.9 (Released Version) (x86 32bit)
References: <Pine DOT SUN DOT 3 DOT 91 DOT 1000518114246 DOT 12245R-100000 AT is>
MIME-Version: 1.0
Content-Type: Text/Plain; charset=US-ASCII
Reply-To: djgpp AT delorie DOT com

In article <Pine DOT SUN DOT 3 DOT 91 DOT 1000518114246 DOT 12245R-100000 AT is>, eliz AT is DOT elta DOT co DOT il 
says...
>
>
>On Wed, 17 May 2000, Adam Herzfeld wrote:
>
>> I am using win2k and have some troubels using rhide.
>> When I try to run or compile a c++ source file (with the extension .cc) it
>> almost always crahes with the windows message "ntvdm.exe has generated
>> errors and will be closed by windows...")
>
>Yep, a known problem.  I suggest to take it up with Microsoft, since this 
>problem doesn't exist in any other Windows version, inclduing NT 4.
>
>If you are willing to try to dig into this problem and perhaps find a way 
>to work around it, please search the archives of this news group (URL 
>below) for a thread about this where I suggested how to investigate this 
>problem.

  The problem might well be in the real mode hook of 
keyboard interrupt 0x1b, maybe disabling this hooking for testers
on win2000 would help to get a answer to that problem !
  Couldn't it be the same problem as the mouse under NT ?

  Reminder, the mouse problem under NT seems to come from the fact that
the call to the DPMI function 0x303 where the address
of the real mode call structure is wrong :
NT apparently does not consider the high part of EDI !

  As DJGPP also uses this function for hooking real mode interrupt
0x1b the problem could be a possible trouble point.