From mboxrd@z Thu Jan 1 00:00:00 1970 From: Patrick McHardy Subject: Re: [NET] Add proc file to display the state of all qdiscs. Date: Thu, 03 Sep 2009 19:56:33 +0200 Message-ID: <4AA00351.4090601@trash.net> References: <20090902081429.GB4878@ff.dom.local> <4A9E2CC7.1010103@gmail.com> <4A9FD2DC.7070807@trash.net> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-15 Content-Transfer-Encoding: 7bit Cc: Christoph Lameter , Eric Dumazet , Jarek Poplawski , David Miller , netdev@vger.kernel.org To: Jesper Dangaard Brouer Return-path: Received: from stinky.trash.net ([213.144.137.162]:44485 "EHLO stinky.trash.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753911AbZICR4d (ORCPT ); Thu, 3 Sep 2009 13:56:33 -0400 In-Reply-To: Sender: netdev-owner@vger.kernel.org List-ID: Jesper Dangaard Brouer wrote: > > On Thu, 3 Sep 2009, Patrick McHardy wrote: > >> The approach I'm currently working on will present multiqueue root >> qdiscs as children of a dummy classful qdisc. This avoids handle >> clashes and the need for new identifiers and allows to address each >> qdisc seperately, similar to how it works with other classful qdiscs: > > I like your approach. Its well suited for the qdiscs :-) > > I especially like the possibility to access each qdisc seperately. Does > it then support having seperate qdisc per TX queue? (I'm toying with > the idea of transmitting our multicast traffic into/via a seperate TX > hardware queue, and making a special qdisc for IPTV MPEG2-TS shaping) Yes, you can attach qdiscs to the classes representing the queues. At least it should work :) It would probably also be possible to use TC classifiers for queue selection.