Xref: news2.mv.net comp.os.msdos.djgpp:4103
From: brien AT leland DOT Stanford DOT EDU (brien oberstein)
Newsgroups: comp.os.msdos.djgpp
Subject: Re: Making the nearptr hacks use longs.
Date: 21 May 1996 13:51:33 -0700
Organization: Stanford University
Lines: 21
Message-ID: <4ntacl$hqo@elaine10.Stanford.EDU>
References: <9605211249 DOT AA0047 AT pc084 DOT altvater DOT com> <4nrsoc$2n3 AT firebird13 DOT Stanford DOT EDU> <31a1efc1 DOT sandmann AT clio DOT rice DOT edu>
NNTP-Posting-Host: elaine10.stanford.edu
To: djgpp AT delorie DOT com
DJ-Gateway: from newsgroup comp.os.msdos.djgpp

i wrote:
>> I'm wondering if there is a simple way to _just_ map the vga frame buffer
>> into the djgpp address space ... This would only have to be done once at the 
>> beginning of the program and would be totally safe!  You get speed+safety.  
>> Isn't this THE solution to graphics under dpmi?

cws wrote:
>Sure, this is the solution, but it REQUIRES DPMI 1.0 features, which 
>Microsoft decided were not worth supporting, so most other DPMI providers
>followed their lead.  The DPMI 1.0 functions which support mapping devices
>into your own memory space are supported by CWSDPMI, but then your 
>program would only run under DOS (so I discourage using non 0.9 functions
>except as functional enhancements).

I think borland c++ comes with a vxd which provides some (all?) dpmi 1.0
functionality to windows.  How 'bout a freeware windows dpmi 0.9 -> 1.0
extender project???  I bet there are plenty of people who would be
willing to help.

-brien