From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Miller Subject: Re: [NET-NEXT PATCH 2/3][UPDATED] pkt_sched: Add multiqueue scheduler support Date: Thu, 11 Sep 2008 20:19:40 -0700 (PDT) Message-ID: <20080911.201940.04283963.davem@davemloft.net> References: <20080912030733.6982.49986.stgit@jtkirshe-mobile.jf.intel.com> <20080912030750.6982.26892.stgit@jtkirshe-mobile.jf.intel.com> Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Cc: jeff@garzik.org, netdev@vger.kernel.org, akpm@linux-foundation.org, alexander.h.duyck@intel.com To: jeffrey.t.kirsher@intel.com Return-path: Received: from 74-93-104-97-Washington.hfc.comcastbusiness.net ([74.93.104.97]:54932 "EHLO sunset.davemloft.net" rhost-flags-OK-FAIL-OK-OK) by vger.kernel.org with ESMTP id S1751852AbYILDTq (ORCPT ); Thu, 11 Sep 2008 23:19:46 -0400 In-Reply-To: <20080912030750.6982.26892.stgit@jtkirshe-mobile.jf.intel.com> Sender: netdev-owner@vger.kernel.org List-ID: From: Jeff Kirsher Date: Thu, 11 Sep 2008 20:08:31 -0700 > From: Alexander Duyck > > This patch is intended to add a qdisc to support the new tx multiqueue > architecture by providing a band for each hardware queue. By doing > this it is possible to support a different qdisc per physical hardware > queue. > > This qdisc uses the skb->queue_mapping to select which band to place > the traffic onto. It then uses a round robin w/ a check to see if the > subqueue is stopped to determine which band to dequeue the packet from. > > Signed-off-by: Alexander Duyck > Signed-off-by: Jeff Kirsher I'm more and more agreeing with Jarek that the root-only restriction is rediculious. What if I want to do PRIO multiqueue for certain classes of traffic? If I want to shoot myself in the foot trying to use a setup like that, this thing damn well better let me. You don't unilaterally know better than the user.