From owner-freebsd-hackers  Mon Dec  4 13:43:07 1995
Return-Path: owner-hackers
Received: (from root@localhost)
          by freefall.freebsd.org (8.6.12/8.6.6) id NAA02666
          for hackers-outgoing; Mon, 4 Dec 1995 13:43:07 -0800
Received: from phaeton.artisoft.com (phaeton.Artisoft.COM [198.17.250.211])
          by freefall.freebsd.org (8.6.12/8.6.6) with ESMTP id NAA02656
          ; Mon, 4 Dec 1995 13:42:49 -0800
Received: (from terry@localhost) by phaeton.artisoft.com (8.6.11/8.6.9) id OAA05965; Mon, 4 Dec 1995 14:35:49 -0700
From: Terry Lambert <terry@lambert.org>
Message-Id: <199512042135.OAA05965@phaeton.artisoft.com>
Subject: Re: ipx on 802.3
To: wollman@lcs.mit.edu (Garrett A. Wollman)
Date: Mon, 4 Dec 1995 14:35:49 -0700 (MST)
Cc: terry@lambert.org, wollman@lcs.mit.edu, jhay@mikom.csir.co.za,
        hackers@FreeBSD.org, wollman@FreeBSD.org
In-Reply-To: <9512042128.AA01138@halloran-eldar.lcs.mit.edu> from "Garrett A. Wollman" at Dec 4, 95 04:28:26 pm
X-Mailer: ELM [version 2.4 PL24]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Content-Length: 1260      
Sender: owner-hackers@FreeBSD.org
Precedence: bulk

> >> >      ifconfig interface address_family [address [dest_address]] [parameters]
> >> >                         ^^^^^^^^^^^^^^
> >> 
> >> > How about that?
> >> 
> >> Utterly irrelevant.
> 
> > Are you sure "AF_IPX" won't work?
> 
> DID YOU EVBEN BOTHER TO READ THE MESSAGE WHICH YOU REPLIED TO???!!!
> 
> Obviously not, since, as I said, your reply was utterly irrelevant to
> the question being asked.
> 
> The question, in case you've forgotten, is: how should one indicate to
> a network interface that it is to use 802.3 encapsulation rather than
> Ethernet v2?
> 
> The answer, as I gave it, was: use the IFF_LINK* flags; that's what
> they're there for.
> 
> Now would you please READ the messages before you respond to them?

Excuse me.

The AF_INET (default) address family causes the link flags to mean
interface type currently.

If one were to switch the behaviour based on address family to do the
"correct" thing in this regard, it would resolve the problem without
having to rewrite all the hardware selection code (not that the fix
you want isn't the correct thing to do in the long run).


					Regards,
					Terry Lambert
					terry@lambert.org
---
Any opinions in this posting are my own and not those of my present
or previous employers.