From mboxrd@z Thu Jan 1 00:00:00 1970 From: Eric Dumazet Subject: Re: [PATCH net-next] net_sched: red: split red_parms into parms and vars Date: Fri, 06 Jan 2012 06:47:02 +0100 Message-ID: <1325828822.2748.27.camel@edumazet-laptop> References: <1325766316.2415.18.camel@edumazet-HP-Compaq-6005-Pro-SFF-PC> <1325770777.2415.35.camel@edumazet-HP-Compaq-6005-Pro-SFF-PC> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: David Miller , netdev , Stephen Hemminger To: Dave Taht Return-path: Received: from mail-we0-f174.google.com ([74.125.82.174]:60791 "EHLO mail-we0-f174.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750906Ab2AFFrI (ORCPT ); Fri, 6 Jan 2012 00:47:08 -0500 Received: by werm1 with SMTP id m1so895371wer.19 for ; Thu, 05 Jan 2012 21:47:05 -0800 (PST) In-Reply-To: <1325770777.2415.35.camel@edumazet-HP-Compaq-6005-Pro-SFF-PC> Sender: netdev-owner@vger.kernel.org List-ID: Le jeudi 05 janvier 2012 =C3=A0 14:39 +0100, Eric Dumazet a =C3=A9crit = : > Problems of RED are : >=20 > 1) the 'average' on highly variable trafic : It means nothing > interesting at all. >=20 > 2) the idle time management requires to specify the link bandwidth, a= s a > fixed value. This is a problem. A third problem is that ECN marking is performed on the packet we add t= o the tail of the queue. It's very late... I am currently testing my SFQRED and I try to add ECN in the first packet in queue so that TCP can react faster.