From owner-freebsd-hackers Sat Jul 8 14:06:27 1995 Return-Path: hackers-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id OAA23343 for hackers-outgoing; Sat, 8 Jul 1995 14:06:27 -0700 Received: from inet-gw-1.pa.dec.com (inet-gw-1.pa.dec.com [16.1.0.22]) by freefall.cdrom.com (8.6.10/8.6.6) with SMTP id OAA23328 for ; Sat, 8 Jul 1995 14:06:26 -0700 Received: from muggsy.lkg.dec.com by inet-gw-1.pa.dec.com (5.65/24Feb95) id AA17944; Sat, 8 Jul 95 14:00:27 -0700 Received: from whydos.lkg.dec.com by muggsy.lkg.dec.com (5.65/DEC-Ultrix/4.3) with SMTP id AA28035; Sat, 8 Jul 1995 17:00:25 -0400 Received: from localhost (localhost [127.0.0.1]) by whydos.lkg.dec.com (8.6.11/8.6.9) with SMTP id RAA06452; Sat, 8 Jul 1995 17:00:59 GMT Message-Id: <199507081700.RAA06452@whydos.lkg.dec.com> X-Authentication-Warning: whydos.lkg.dec.com: Host localhost didn't use HELO protocol To: Julian Howard Stacey Cc: hackers@freebsd.org, n1epo4tl@ibmmail.com Subject: Re: token ring anyone In-Reply-To: Your message of "Thu, 06 Jul 1995 16:35:57 +0200." <199507061435.QAA03684@vector.eikon.e-technik.tu-muenchen.de> X-Mailer: exmh version 1.5omega 10/6/94 Date: Sat, 08 Jul 1995 17:00:51 +0000 From: Matt Thomas Sender: hackers-owner@freebsd.org Precedence: bulk > a couple of weeks ago I asked (on behalf of n1epo4tl@ibmmail.com Stuart > Arnold) if anyone knew of any support for token ring cards with FreeBSD, > we heard nothing ... either there's no interest or maybe it was mail > failure ? Token Ring is a proverbial pain in the ass. Unlike writing a new Ethernet driver, Token Ring also involves writing the Token Ring support code, modifying IP Multicast to work over Token Ring (ie. map to the right funtional address), and, worst of all, modifying ARP so that it understand how to do source routing. Of course this ignores the fact that you really want fairly generic source routing so you could use it with IPX or OSI or whatever else is around. The major Token Ring chipsets are fairly painful. It also means that the driver writer have the infrastructure to test the driver (unlike Ethernet where you can plug two systems back-to-back or use some ThinWire Token Ring requires a MAU). If you don't need to test source routing, thin a single ring will be fine otherwise you'll need a bridge. You also need a router to make sure you are dealing with stupidity of bit-flipped ARP address properly. Instead of writing token ring drivers, I think it would be a far better investment to write NDIS3 miniport wrapper code for FreeBSD but I digress. So in essence, until the pain of not having Token Ring exceeds the pain threshold of implementation it's unlikely to be done. That's my take on it. Matt Thomas Internet: matt@lkg.dec.com 3am Software Foundry WWW URL: Westford, MA Disclaimer: Digital disavows all knowledge of this message