From mboxrd@z Thu Jan 1 00:00:00 1970 From: Patrick McHardy Subject: Re: [PATCH] Stochastic Fair Blue queue discipline Date: Tue, 08 Apr 2008 02:56:49 +0200 Message-ID: <47FAC2D1.6010909@trash.net> References: <87skxxb8br.fsf@pirx.pps.jussieu.fr> <47FAB688.7010807@trash.net> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-15; format=flowed Content-Transfer-Encoding: 7bit Cc: netdev@vger.kernel.org To: Juliusz Chroboczek Return-path: Received: from stinky.trash.net ([213.144.137.162]:52635 "EHLO stinky.trash.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752586AbYDHA7n (ORCPT ); Mon, 7 Apr 2008 20:59:43 -0400 In-Reply-To: <47FAB688.7010807@trash.net> Sender: netdev-owner@vger.kernel.org List-ID: Patrick McHardy wrote: > Juliusz Chroboczek wrote: >> Hello, >> >> The attached is an implementation of the Stochastic Fair Blue queue >> discipline for Linux. >> >> I would like to request a review of the following code, and whether it >> is worthwile to add the Kconfig bits and submit it for inclusion into >> the mainline kernel. >> >> You will find a ready to build tarball on >> >> http://www.pps.jussieu.fr/~jch/software/files/sch_sfb-20080407.tar.gz >> >> http://www.pps.jussieu.fr/~jch/software/files/sch_sfb-20080407.tar.gz.asc >> >> >> and a short description on >> >> http://www.pps.jussieu.fr/~jch/software/sfb/ > > Do you have some graphs or something showing the effects? > I experienced with SFB myself some time ago and recall > getting huge unfairness towards non-ECN connections. > I can't remember exactly how good it behaved overall. I just found my old tree and it was just Blue, which might explain this :) Thanks for posting this, I'll have a closer look at your patch tommorrow.