Date: Sun, 12 May 1996 10:16:43 +0200 (IST)
From: Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>
To: Stephen Johin Leung <sjleung AT acs DOT ucalgary DOT ca>
Cc: djgpp AT delorie DOT com
Subject: Re: Q: Anyone tried GUS SDK?
In-Reply-To: <31915F4E.2070@acs.ucalgary.ca>
Message-Id: <Pine.SUN.3.91.960512101211.25343I-100000@is>
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII


On Wed, 8 May 1996, Stephen Johin Leung wrote:

> 	Ah.  Sorry.  I guess the big, BIG problem is that I have no idea how to compile in the 
> libraries that come with it.  The docs say there are three sets coded for Borland, Microsoft, 
> and Watcom compilers.  They also make some mention of Metaware HighC.  Anyhow, since full 
> source is not included, I find that that kind of sucks.  In examining the libraries 
> themselves, they do seem to be quite different from whatever format DJGPP uses.

Then read chapter 17 of the DJGPP FAQ list (available as v2/faq200b.zip 
from the same place you get DJGPP).  It explains this issue and points 
you to a converter that might help you.

> 	Oh, and on a kind of a side note (but also maybe a reason why nothing would work), do 
> far pointers really exist in DJGPP?  It's all protected mode and 32-bit addressing, so is 
> there any distinction there?  I've noticed that function prototypes which declare anything as 
> far * generate a parse error when compiling.

DJGPP produces 32-bit protected-mode code that uses flat address space, 
so you can safely define away the ``far'' keyword.  In DJGPP ``far 
pointers'' are used when you have to access memory outside your usual 
address space (like when peeking at DOS memory) without triggering memory 
protection violation.