Date: Mon, 20 Mar 2000 11:47:08 +0200 (IST)
From: Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>
X-Sender: eliz AT is
To: AndrewJ <luminous-is AT home DOT com>
cc: djgpp AT delorie DOT com
Subject: Re: DPMI 1.0 is supported?
In-Reply-To: <6p4B4.41332$Hq3.883787@news2.rdc1.on.home.com>
Message-ID: <Pine.SUN.3.91.1000320114645.24837D-100000@is>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Reply-To: djgpp AT delorie DOT com
Errors-To: dj-admin AT delorie DOT com
X-Mailing-List: djgpp AT delorie DOT com
X-Unsubscribes-To: listserv AT delorie DOT com
Precedence: bulk


On Sun, 19 Mar 2000, AndrewJ wrote:

> It should also be noted that 3.x didn't *fully* (or is that properly?)
> support DPMI 0.9. If you check Ralf Brown's Interrupt List, it shows all the
> calls that 3.x dislikes.

Almost all of the unsupported functions are only relevant for Windows
3.0 in standard mode--hardly an interesting environment for DJGPP.
Windows 3.1 and 3.10 has problems with only 1 function.

> Isn't it kind of amusing that, in the 200+ MB of Windows '98 (and the
> minimum 100 MB I got '95 down to), Microsoft didn't bother to squeeze in 1.0
> support?

It's the usual Microsoft way: introduce a new API, get the entire
industry on the run to support it, then drop it a year or two later.

Microsoft needed DPMI because the Windows kernel loads itself as a
DPMI client.  But once the kernel is loaded, its needs of the DPMI API
are quite limited, so they didn't bother to support more than is
necessary.