From:	Mihai Moise <moisemih AT ift DOT ulaval DOT ca>
Newsgroups:	comp.os.msdos.djgpp
Subject:	Re: allegro author?
Date:	Tue, 10 Sep 1996 18:46:00 -0400
Organization:	Universite Laval
Lines:	28
Message-ID:	<3235EFA8.167EB0E7@ift.ulaval.ca>
References:	<Pine DOT NEB DOT 3 DOT 95 DOT 960910132501 DOT 2429A-100000 AT ananke DOT amu DOT edu DOT pl>
NNTP-Posting-Host:	britten.ift.ulaval.ca
Mime-Version:	1.0
To:	djgpp AT delorie DOT com
DJ-Gateway:	from newsgroup comp.os.msdos.djgpp

Mark Habersack wrote:
> 

> Well... I think it doesn't make sense to write such an interface. Allegro is a
> self-contained library in every aspect and writing an interface to SVGAlib
> would be just creating a new front-end. Allegro has completely different model
> of driver structure. What you should do is to isolate the parts tightly M$-DOG
> related and re-write them Linux-like.
>

Not really, Mark. Allegro relies on a VESA UniVBE host for mode
switching, if I have read the sources correctly. And it does so using
real-mode interrupts. Real-mode interrupts don't look too good in Linux,
notice I am not saying its impossible. Besides, to take advantage of
linear framebuffers, a Linux user would have to buy $citech's DD.
   
>  >2) rewrite keyboard functions
> An easy one - you might use libkb. Or even it might not be necessary to change
> the keyboard handling routines of allegro. AFAIR, Allegro installs hw int
> handler for keyboard? If so, then, as long as PC platform is concerned, you
> are in home with what you have in Allegro. If, however, you want to use
> Allegro also on other hardware platforms then you have to create an
> independent interface to UNIX kernel's kb handler.

Hardware interrupt handler? Ouuch!... Linux won't like that... ( and you
know the fate of processes which Linux doesn't like! )

Mihai

- Raw text -