Date:	Wed, 13 Dec 2000 13:43:47 +0200 (IST)
From:	Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>
X-Sender:	eliz AT is
To:	djgpp AT delorie DOT com
Subject:	Re: Porting DJGPP DPMI Apps-->Borland C++5.x
In-Reply-To:	<3A36A472.7040003@home.com>
Message-ID:	<Pine.SUN.3.91.1001213134330.11254W-100000@is>
MIME-Version:	1.0
Reply-To:	djgpp AT delorie DOT com
Errors-To:	nobody AT delorie DOT com
X-Mailing-List:	djgpp AT delorie DOT com
X-Unsubscribes-To:	listserv AT delorie DOT com

On Tue, 12 Dec 2000, Russ Magee wrote:

> Our concern is whether it will be easy to convert the existing DJGPP
> C and GAS assembler source to Borland's environment; VESA video BIOS calls
> are also a big concern, as there are graphics assignments using it.

As written elsewhere in this thread, this is far from being easy.

> 1) If we use the 5.0 compiler and make 16-bit DOS apps, is it true we don't
> need DPMI at all, and can just call BIOS the traditional way, and bang
> on hardware directly?

Yes and yes.

> 2) If we use the newer compiler, and compile win32 console apps, can these
> access the BIOS and hit hardware directly as well?

I don't think so.  IIRC, a Win32 application cannot invoke real-mode
interrupts directly (because thunking interferes with this).  You will
probably have to use the low-level (and mostly undocumented) Win32 API
calls.

> 3) And can we use VESA BIOS calls with either of the above schemes, without
> DJGPP? (IE, what kind of support for VESA calls is there in TASM, Borland
> C/C++)?

I expect you to have to use the int86 function and absolute
addresses.  If there are protected-mode VESA calls, or absolute
addresses above 1MB mark, you will have to use XMS calls to get
there.

I don't really understand the reasons for switching, but it looks like
you are in for a bumpy ride.  Might as well reconsider the original
decision to switch.

- Raw text -