X-Authentication-Warning:	delorie.com: mailnull set sender to djgpp-workers-bounces using -f
From:	sandmann AT clio DOT rice DOT edu (Charles Sandmann)
Message-Id:	<10112231831.AA18609@clio.rice.edu>
Subject:	Re: RFC - Dynamic loading
To:	djgpp-workers AT delorie DOT com
Date:	Sun, 23 Dec 2001 12:31:17 -0600 (CST)
In-Reply-To:	<250B3114DA16D511B82C00E0094005F8055AC7F1@MSGWAW11> from "=?iso-8859-2?Q?Wojciech_Ga=B3=B1zka?=" at Dec 23, 2001 01:00:50 PM
X-Mailer:	ELM [version 2.5 PL2]
Mime-Version:	1.0
Reply-To:	djgpp-workers AT delorie DOT com
Errors-To:	nobody AT delorie DOT com
X-Mailing-List:	djgpp-workers AT delorie DOT com
X-Unsubscribes-To:	listserv AT delorie DOT com

> > Today there is no communications center between loaded DJGPP 
> > applications, so if the first one in memory loads it, the next one has no 
> > idea another  DJGPP image is even running.

> That's not an issue - we can develop such a interface

A protected mode interrupt hook by the parent is probably the way to do
this (I've thought about the problem).  But this doesn't have anything
to do with DXEs or DLLs - we could do this today.  The first program
loaded has control, if a child loads he just doesn't use his code 
but instead passes requests to the parent code base.

> Actually I'm not so concerned of turning libc into a dynamic library but
> rather to have the ability of writing a dynamic library which is loaded by 
> a first djgpp program that uses it while another programs would use the 
> first copy instead of loading their own ones.

We can do that today ... Noone has ever been interested enough to do it :-)

> Why is this for? 
> I'm hoping I'll be finished with my Microsoft exams by March and I'm
> planning to work on TCPIP library under Windows NT
> Having a single instance of a TCPIP routines makes the whole design much
> simpler ...

Okay, I'd be happy to help you get that working.  I'd be even happier if
loading the TCP/IP library would just find the right stack under DOS, NT
or W9x and use it :-)

- Raw text -