From owner-freebsd-net@FreeBSD.ORG Wed Aug 6 06:42:54 2008 Return-Path: Delivered-To: freebsd-net@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 5D4501065679 for ; Wed, 6 Aug 2008 06:42:54 +0000 (UTC) (envelope-from mav@FreeBSD.org) Received: from cmail.optima.ua (cmail.optima.ua [195.248.191.121]) by mx1.freebsd.org (Postfix) with ESMTP id 999B18FC1E for ; Wed, 6 Aug 2008 06:42:53 +0000 (UTC) (envelope-from mav@FreeBSD.org) X-Spam-Flag: SKIP X-Spam-Yversion: Spamooborona-2.1.0 Received: from orphanage.alkar.net (account mav@alkar.net [212.86.226.11] verified) by cmail.optima.ua (CommuniGate Pro SMTP 5.1.14) with ESMTPA id 187095364; Wed, 06 Aug 2008 09:42:52 +0300 Message-ID: <489947EB.4010606@FreeBSD.org> Date: Wed, 06 Aug 2008 09:42:51 +0300 From: Alexander Motin User-Agent: Thunderbird 2.0.0.14 (X11/20080612) MIME-Version: 1.0 To: Sten Daniel Soersdal References: <48992D64.1020502@gmail.com> In-Reply-To: <48992D64.1020502@gmail.com> X-Enigmail-Version: 0.95.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit Cc: "freebsd-net@FreeBSD.org" , Archie Cobbs Subject: Re: ng_ppp: question about "bandwidth" parameter. X-BeenThere: freebsd-net@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Networking and TCP/IP with FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 06 Aug 2008 06:42:54 -0000 Sten Daniel Soersdal wrote: > Is there a reason NG_PPP_MAX_BANDWIDTH is limited to a total of 10 Mbit? AFAIR the only reason is 32bit integer math overflow protection. Probably it is the time now to think about 64 bits. > If any of the links are faster than 10 Mbit, will i achieve same end > result by dividing Bandwidth by say 10? That is, for every 10 Mbit > physical link i configure ng_ppp to 1 Mbit? > Or could i change the constant to the equivalent of 100 Mbit/s without > the risk of breaking anything? I am not sure. You should first check value ranges in all math operations. > The reason i ask is I noticed in ng_ppp.c (1.70.2.3) in function > ng_ppp_mp_strategy(..) that ng_ppp also takes into account queued > packets and the additional latency that would implicate. Yes. If you divide all speeds on 10 it will break link queue length calculation. So until traffic will be below specified constants it will work more or less correct. Above it node will think that queues are constantly overflowed and link balancing algorithm will become ineffective. -- Alexander Motin