From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jarek Poplawski Subject: Re: [RFC] [PATCH] Don't run __qdisc_run() on a stopped TX queue Date: Sat, 25 Jul 2009 13:53:09 +0200 Message-ID: <20090725115309.GB3086@ami.dom.local> References: <20090720074607.32463.23013.sendpatchset@localhost.localdomain> <20090724083006.GA5698@ff.dom.local> <20090724083749.GA23521@gondor.apana.org.au> <20090725032436.GA30741@gondor.apana.org.au> <20090725110409.GA3086@ami.dom.local> <20090725111207.GA518@gondor.apana.org.au> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Krishna Kumar2 , davem@davemloft.net, netdev@vger.kernel.org To: Herbert Xu Return-path: Received: from mail-fx0-f218.google.com ([209.85.220.218]:41382 "EHLO mail-fx0-f218.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751628AbZGYLxb (ORCPT ); Sat, 25 Jul 2009 07:53:31 -0400 Received: by fxm18 with SMTP id 18so1846835fxm.37 for ; Sat, 25 Jul 2009 04:53:30 -0700 (PDT) Content-Disposition: inline In-Reply-To: <20090725111207.GA518@gondor.apana.org.au> Sender: netdev-owner@vger.kernel.org List-ID: On Sat, Jul 25, 2009 at 07:12:07PM +0800, Herbert Xu wrote: > On Sat, Jul 25, 2009 at 01:04:09PM +0200, Jarek Poplawski wrote: > > > > It would be nice to establish what difference is made by each part of > > this patch. The check for stopped queue before each skb xmit was done > > earlier too, so it's not exactly multiqueue penalization. > > No, but getting into qdisc_restart when we know that the queue is > stopped is just silly. The only reason that is done right now is > because of the pathological case where a packet on qdisc X maps to > hardware queue Y. This is something which almost nobody uses and > yet it penalises everybody. Hmm... I agree with you 100%! (I'd only like to see more digits... ;-) Cheers, Jarek P.