From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Miller Subject: Re: [PATCH net-next] net: sched: consolidate tc_classify{,_compat} Date: Thu, 27 Aug 2015 14:19:39 -0700 (PDT) Message-ID: <20150827.141939.681226506664674587.davem@davemloft.net> References: <3dfe133299d033dfa52bcf63d846f3f91b56d30c.1440620622.git.daniel@iogearbox.net> Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Cc: ast@plumgrid.com, netdev@vger.kernel.org To: daniel@iogearbox.net Return-path: Received: from shards.monkeyblade.net ([149.20.54.216]:47121 "EHLO shards.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752010AbbH0VTk (ORCPT ); Thu, 27 Aug 2015 17:19:40 -0400 In-Reply-To: <3dfe133299d033dfa52bcf63d846f3f91b56d30c.1440620622.git.daniel@iogearbox.net> Sender: netdev-owner@vger.kernel.org List-ID: From: Daniel Borkmann Date: Wed, 26 Aug 2015 23:00:06 +0200 > For classifiers getting invoked via tc_classify(), we always need an > extra function call into tc_classify_compat(), as both are being > exported as symbols and tc_classify() itself doesn't do much except > handling of reclassifications when tp->classify() returned with > TC_ACT_RECLASSIFY. > > CBQ and ATM are the only qdiscs that directly call into tc_classify_compat(), > all others use tc_classify(). When tc actions are being configured > out in the kernel, tc_classify() effectively does nothing besides > delegating. > > We could spare this layer and consolidate both functions. pktgen on > single CPU constantly pushing skbs directly into the netif_receive_skb() > path with a dummy classifier on ingress qdisc attached, improves > slightly from 22.3Mpps to 23.1Mpps. > > Signed-off-by: Daniel Borkmann Applied, thanks Daniel.