From mboxrd@z Thu Jan 1 00:00:00 1970 From: Patrick McHardy Subject: Re: [RFC NET_SCHED 00/02]: Flexible SFQ flow classification Date: Wed, 30 May 2007 17:32:08 +0200 Message-ID: <465D98F8.9000102@trash.net> References: <20070530094020.24073.84277.sendpatchset@localhost.localdomain> <465D5D7A.8090200@andyfurniss.entadsl.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-15 Content-Transfer-Encoding: 7bit Cc: netdev@vger.kernel.org, hadi@cyberus.ca To: lists@andyfurniss.entadsl.com Return-path: Received: from stinky.trash.net ([213.144.137.162]:42085 "EHLO stinky.trash.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752120AbXE3PcL (ORCPT ); Wed, 30 May 2007 11:32:11 -0400 In-Reply-To: <465D5D7A.8090200@andyfurniss.entadsl.com> Sender: netdev-owner@vger.kernel.org List-Id: netdev.vger.kernel.org Andy Furniss wrote: > Patrick McHardy wrote: > >> It currently does not support perturbation, I didn't want to move this >> into >> the classifier, so I need to think about a way to handle it within SFQ. >> > > Cool, but isn't this going to show the same collision problems that the > pre jhash esfq saw? Perturbation doesn't prevent collisions, it just distributes them (hopefully evenly). My classifier uses jhash, but that won't prevent collisions either. Anyways, I'm going to change SFQ so perturbation can also be used with external classifiers.