Date: Mon, 3 Mar 1997 13:21:52 +0200 (IST)
From: Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>
To: shedd timothy allen <shedd AT ehsn8 DOT cen DOT uiuc DOT edu>
cc: djgpp AT delorie DOT com
Subject: Re: Device driver needs segment:offset info
In-Reply-To: <Pine.Sola.3.91.970302175553.2191C-100000@ehsn8.cen.uiuc.edu>
Message-ID: <Pine.SUN.3.91.970303132053.9009K-100000@is>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII


On Sun, 2 Mar 1997, shedd timothy allen wrote:

> make old equipment work with new research projects.)  As a result, if the 
> answer is in the documentation, I have missed it and I apologize in 
> advance.  
> 
> Would anyone be able to help with the following scenario?

With all due respect, it is not a good idea to treat this news group
as a free hot line.  Generally, you can *not* expect people to sit
down, write code and debug it for you.  People *will* read *your* code
if you will post it (especially if it is short enough to be grasped in
a couple of minutes), together with a description of how does it fail,
and will try to help you make it work by pointing out which parts of
it are incorrect or unsafe.

Therefore, I suggest that you try to rewrite the code as DJGPP
requires, then ask here specific questions if it doesn't work.  The
issues pertinent to your problem are described in the FAQ list:
section 18.2 explains how to pass segment:offset addresses to
real-mode interrupts, and section 18.4 explains how to put data to, or
get data from a buffer in conventional memory where the real-mode
interrupts can get at it.  If you need working examples of code that
actually works using the techniques described in the FAQ, download the
DJGPP source distribution (v2/djlsr201.zip) and look at low-level I/O
functions there, such as src/libc/dos/io/putpath.c,
src/libc/dos/io/_read.c, src/libc/dos/io/_write.c, etc.