From owner-freebsd-arch Mon Jan 29 16:11:49 2001 Delivered-To: freebsd-arch@freebsd.org Received: from iguana.aciri.org (iguana.aciri.org [192.150.187.36]) by hub.freebsd.org (Postfix) with ESMTP id E220E37B404; Mon, 29 Jan 2001 16:11:26 -0800 (PST) Received: (from rizzo@localhost) by iguana.aciri.org (8.11.1/8.11.1) id f0U0BNx48599; Mon, 29 Jan 2001 16:11:23 -0800 (PST) (envelope-from rizzo) From: Luigi Rizzo Message-Id: <200101300011.f0U0BNx48599@iguana.aciri.org> Subject: Re: Is anybody working on bridging code & a question for -arch on userland/kernel In-Reply-To: <4.3.2.7.0.20010130002432.00b5f100@mail.drwilco.net> from "Rogier R. Mulhuijzen" at "Jan 30, 2001 0:31:38 am" To: drwilco@drwilco.net (Rogier R. Mulhuijzen) Date: Mon, 29 Jan 2001 16:11:13 -0800 (PST) Cc: freebsd-net@FreeBSD.ORG, freebsd-arch@FreeBSD.ORG X-Mailer: ELM [version 2.4ME+ PL43 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG > 1) Is anyone working on the bridging code? I'm going to extend the > ng_bridge node with Spanning Tree Protocol and I wouldn't want to be > duplicating work. I checked in -current, but I thought I'd check on -net as > well. (And -arch because of my next question) i am doing some minor fixes to the non-netgraph version of the bridging code. > 2) Where does one draw the line at handling stuff in the kernel or > userspace. The algorithms that are used in the Spanning Tree Protocol > aren't very complicated, and I'm pretty sure I could contain everything in > kernel space, but what is the Right Thing to do? Everything in kernelspace, > or running a userland daemon that does all the calculating, decision making > and time tracking? i'd rather do as much as possible in userspace, and issue appropriate calls (basically sysctl or similar to enable/disable forwarding on some of the bridged interfaces) when necessary. If you need a specific control interface i will be glad to implement it, and your spanningTreed will be much much easier to implement, test, maintain and reuse for different bridging things (not to mention that a single box can act as multiple independent bridges for which you want multiple spanning trees). let me know if you have some spanning tree code that you want to test/integrate. cheers luigi ----------------------------------+----------------------------------------- Luigi RIZZO, luigi@iet.unipi.it . ACIRI/ICSI (on leave from Univ. di Pisa) http://www.iet.unipi.it/~luigi/ . 1947 Center St, Berkeley CA 94704 Phone: (510) 666 2927 ----------------------------------+----------------------------------------- To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message