From mboxrd@z Thu Jan 1 00:00:00 1970 From: Eric Dumazet Subject: Re: [PATCH] ipv4: add DiffServ priority based routing Date: Thu, 14 Jan 2010 13:50:51 +0100 Message-ID: <4B4F132B.7020001@gmail.com> References: <201001121432.43301.schmto@hrz.tu-chemnitz.de> <20100112.121607.39835310.davem@davemloft.net> <201001141250.27287.schmto@hrz.tu-chemnitz.de> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: David Miller , netdev@vger.kernel.org To: Torsten Schmidt Return-path: Received: from gw1.cosmosbay.com ([212.99.114.194]:38807 "EHLO gw1.cosmosbay.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753152Ab0ANMu4 (ORCPT ); Thu, 14 Jan 2010 07:50:56 -0500 In-Reply-To: <201001141250.27287.schmto@hrz.tu-chemnitz.de> Sender: netdev-owner@vger.kernel.org List-ID: Le 14/01/2010 12:50, Torsten Schmidt a =E9crit : =2E > Our company is more interested in IP DiffServ traffic accounting. > So we simply need a fast mechanism which generates IP DiffServ traffi= c > statistics, without setting the network interface in promiscuous mode= , > (which softflowd ore tcpdump does). tcpdump can work in non promiscuous mode (-p). Anyway, switched network= s makes promiscuous a non issue nowaday. >=20 > So first idea was to implement a virtual file e.g. called /proc/net/i= p_dscp. > If the main line peoples are not interested in, i will no longer send= these > patches.=20 Problem is, you might need many sets of counters... 1) General (for each net namespace ...) 2) Per interface 3) Per protocol ? =46or performance reasons, SNMP counters are per cpu, so each set would= be very large. In any case, dont add yet another /proc/net file, this is considered as= a hacky way to do these things today. Thanks