Path: utzoo!attcan!uunet!bbs!karl
From: karl@naitc.naitc.com (Karl Denninger)
Newsgroups: comp.sys.ibm.pc.misc
Subject: Re: Need Int 14h redirector source
Summary: No, this is not true!
Message-ID: <1990Oct02.211126.14260@naitc.naitc.com>
Date: 2 Oct 90 21:11:26 GMT
References: <36568@cc.usu.edu>
Reply-To: karl@bbs.naitc.com (Karl Denninger)
Organization: A.C. Nielsen Co.
Lines: 34

In article <36568@cc.usu.edu> JRD@cc.usu.edu (Joe Doupnik) writes:
>
>	The Int 14h business seems to be heating up recently. Today a msg
>asked about source code for an Int 14h hooker to be applied to a commercial
>TCP/IP package. Alas, this is a much bigger problem than writing a driver.
>	As seen from the terminal emulator side, the Int 14h interface is
>called upon (invoke Int 14h) to get/send characters and status. The Int 14h
>interface does not invoke the terminal emulator this way.
>	Thus, the whole TCP/IP part needs to become a TSR and buffer incoming
>characters awaiting the terminal emulator's request to do work. Well, most
>Telnet packages call upon a resident kernel but are not TSRs themselves; far
>from it. The problem then devolves down to converting the Telnet part to
>be a TSR and that means redesigning it from the very beginning. The point of
>who invokes whom is central.
>	Joe D.

Not true!

The TCP/IP I am talking about, Beame and Whiteside's, provides a
socket-level interface which works via DOS opens, ioctls, and close calls.

Yes, you really do it that easily.  You open a device, ioctl your options
(like the port to bind to, the connection request, etc) and then read/write
data just like it was a file.

Thus, a TSR really is just a cheap and dirty translation filter.  Nothing
more.  Quite an elegant solution to the problem, if you ask me....

--
Karl Denninger	AC Nielsen
kdenning@ksun.naitc.com
(708) 317-3285
Disclaimer:  Contents represent opinions of the author; I do not speak for
	     AC Nielsen on Usenet.