Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!sun-barr!olivea!spool.mu.edu!mips!sdd.hp.com!think.com!snorkelwacker.mit.edu!thunder.mcrcim.mcgill.edu!mouse
From: mouse@thunder.mcrcim.mcgill.edu (der Mouse)
Newsgroups: comp.unix.internals
Subject: Re: SLIP without kernel modification
Message-ID: <1991Jun22.131633.25667@thunder.mcrcim.mcgill.edu>
Date: 22 Jun 91 13:16:33 GMT
References: <128234@tyrell.stgt.sub.org> <800@minya.UUCP>
Organization: McGill Research Centre for Intelligent Machines
Lines: 56

In article <800@minya.UUCP>, jc@minya.UUCP (John Chambers) writes:
> In article <128234@tyrell.stgt.sub.org>, rodney@tyrell.stgt.sub.org (Rodney Volz) writes:
>> Is there a way to speak SLIP to a remote machine, if you do not have
>> the possibility to rebuild the kernel with a slip-driver in it?
> Well, in general, *something* needs to be added, because there is
> the problem  of how you persuade the IP package in the kernel to hand
> over the packets that need to go out the SLIP port.  But that
> something  is not necessarily a SLIP driver.

> Recently, I got my hands on a driver for Ultrix that is something
> that I've wished was in the IP package from the  start.   It's
> called  the "PNI" driver, which stands for Pseudo-Network Interface.
> What it is is a network equivalent to a pty:

I built something similar to this a few years back.  Very useful thing.
In our case, though, the idea was to shove IP packets through a network
of DECnet-only routers.  So I had the psi (my name for it) interface
talk to a program that sent the packets out over DECnet connections,
and the process on the other end did the reverse transformation....
Later, I built a user-level program to run it as a SLIP link.  I
considered using it to circumvent a brain-damaged administrative
nightmare of a router (they'd implemented a packet fence that dropped
most useful services, like mail and finger, on the floor, but allowed
connections on high-numbered ports.  (There is something very appealing
about using a TCP connection as an IP packet transport :-)  Never did,
because the packet fence came down before the irritation level got
sufficient to prod me into setting it up.

> I have no idea whether such a pseudo-interface is available for
> other systems.

My driver works on a couple of old versions of Ultrix and on 4.3.  I'm
told it works on relatively recent Ultrices.  It probably wouldn't be
much effort to port it to anything that provides a Berkeleyish driver
interface.

> Dealing with the kernel for things like SLIP is a real pain, since
> TFM is of little benefit and you invariably have to wade through
> piles of insults to ferret out the facts concerning things like
> TCP/IP...

UTSL...and if you don't have TS, scrap what you have and get a real OS!
(Only half a :-), and that half only because there seem to be no sane
hardware manufacturers these days....)

I've put a copy of if_psi.c up for anonymous ftp on 132.206.1.1;
there's also if_psi.README, which is the README from the IP-over-DECnet
stuff I mentioned above; I included it because it describes installing
if_psi into the kernel.  (It also describes setting up the DECnet-IP
stuff too; just ignore that.)  I can also mail copies if necessary....

					der Mouse

			old: mcgill-vision!mouse
			new: mouse@larry.mcrcim.mcgill.edu