From owner-freebsd-atm Fri Jul 5 13:54:03 1996 Return-Path: owner-atm Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id NAA10122 for atm-outgoing; Fri, 5 Jul 1996 13:54:03 -0700 (PDT) Received: from mail13.digital.com (mail13.digital.com [192.208.46.30]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id NAA10110 for ; Fri, 5 Jul 1996 13:53:56 -0700 (PDT) Received: from muggsy.lkg.dec.com by mail13.digital.com (8.7.5/UNX 1.2/1.0/WV) id QAA16798; Fri, 5 Jul 1996 16:44:09 -0400 (EDT) Received: from whydos.lkg.dec.com by muggsy.lkg.dec.com (5.65/DEC-Ultrix/4.3) with SMTP id AA07885; Fri, 5 Jul 1996 16:44:08 -0400 Received: from localhost (localhost [127.0.0.1]) by whydos.lkg.dec.com (8.6.12/8.6.12) with SMTP id QAA23702; Fri, 5 Jul 1996 16:47:13 GMT Message-Id: <199607051647.QAA23702@whydos.lkg.dec.com> X-Authentication-Warning: whydos.lkg.dec.com: Host localhost didn't use HELO protocol X-Mailer: exmh version 1.6.5 12/11/95 To: freebsd-atm@freebsd.org, tech-net@netbsd.org Cc: ewv@boom.bsdi.com Subject: My thoughts on ATM for *BSD. Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Fri, 05 Jul 1996 16:47:13 +0000 From: Matt Thomas Sender: owner-atm@freebsd.org X-Loop: FreeBSD.org Precedence: bulk While I away at the IETF, there was a discussion about ATM posted to the freebsd-hackers mailing list. However, since I am going to be discussing is not specific to FreeBSD, I'm also sending this to the tech-net mailing of NetBSD. For those of you who haven't seen it, you might want to check out the paper about IP Switching on http://www.ipsilon.com. Unlike Ethernet or FDDI, there is not a one-to-one correspondence between an ATM adapter and the number of interfaces for the system. Indeed, ATM might best be represented by a layered approach similar to what is done with SCSI. In a traditional system, you will have one or more Classic IP networks and possibly additional emulated LANs. However, it is unlikely that most ATM adapters will care about that. What they will need is a minimalist interface which ties in just below the signalling layer (possibly even not needing a singalling layer in some environments). This means that a MI signalling layer, classic IP, etc is needed. Hopefully that code can reuse existing software such as VINCE (ftp://ftp.cmf.nrl.navy.mil/pub/vince/). My current concern is the driver interface to signaling and data movement. FWIW, I've given up on ATMworks 350 because it uses a 21050 PPB and that will just lockups on Triton-based motherboards. I'm currently working on the next generation card after the 350. -- Matt Thomas Internet: matt@3am-software.com 3am Software Foundry WWW URL: http://www.3am-software.com/bio/matt.html Westford, MA Disclaimer: I disavow all knowledge of this message