Return-path: Received: from rhun.apana.org.au ([64.62.148.172]:53406 "EHLO arnor.apana.org.au" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1753912AbYGUNTa (ORCPT ); Mon, 21 Jul 2008 09:19:30 -0400 Date: Mon, 21 Jul 2008 21:19:18 +0800 From: Herbert Xu To: jamal Cc: davem@davemloft.net, kaber@trash.net, netdev@vger.kernel.org, johannes@sipsolutions.net, linux-wireless@vger.kernel.org Subject: Re: [PATCH 20/31]: pkt_sched: Perform bulk of qdisc destruction in RCU. Message-ID: <20080721131918.GA11114@gondor.apana.org.au> (sfid-20080721_151934_190090_5A227B8A) References: <1216645724.4847.275.camel@localhost> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii In-Reply-To: <1216645724.4847.275.camel@localhost> Sender: linux-wireless-owner@vger.kernel.org List-ID: On Mon, Jul 21, 2008 at 09:08:44AM -0400, jamal wrote: > > The one thing i am unsure of still: > I think it would be cleaner to just stop a single queue (instead of all) > when one hardware queue fills up. i.e if there is no congestion on the > other hardware queues, packets should continue to be fed to their > hardware queues and not be buffered at qdisc level. Right you don't necessarily have to stop all queues but you do need to direct all packets into the qdisc. > Parallelization would work if you can get X CPUs to send to X hardware > queues concurently. Feasible in static host setup like virtualization > environment where you can tie a vm to a cpu. Not very feasible in > routing where you are driven to a random hardware tx queue by arriving > packets. It should work just fine for routing assuming your card does multi-queue RX. Cheers, -- Visit Openswan at http://www.openswan.org/ Email: Herbert Xu ~{PmV>HI~} Home Page: http://gondor.apana.org.au/~herbert/ PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt