From mboxrd@z Thu Jan 1 00:00:00 1970 From: jamal Subject: Re: [WIP][PATCHES] Network xmit batching Date: Fri, 08 Jun 2007 07:31:07 -0400 Message-ID: <1181302267.4063.33.camel@localhost> References: <1181216629.4064.22.camel@localhost> <20070607161335.GA4987@2ka.mipt.ru> <1181254996.4071.27.camel@localhost> <20070608083807.GB11488@2ka.mipt.ru> Reply-To: hadi@cyberus.ca Mime-Version: 1.0 Content-Type: text/plain Content-Transfer-Encoding: 7bit Cc: Krishna Kumar2 , Gagan Arneja , netdev@vger.kernel.org, Rick Jones , Sridhar Samudrala , David Miller , Robert Olsson To: Evgeniy Polyakov Return-path: Received: from wx-out-0506.google.com ([66.249.82.226]:58760 "EHLO wx-out-0506.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S966955AbXFHLbM (ORCPT ); Fri, 8 Jun 2007 07:31:12 -0400 Received: by wx-out-0506.google.com with SMTP id t15so700931wxc for ; Fri, 08 Jun 2007 04:31:11 -0700 (PDT) In-Reply-To: <20070608083807.GB11488@2ka.mipt.ru> Sender: netdev-owner@vger.kernel.org List-Id: netdev.vger.kernel.org On Fri, 2007-08-06 at 12:38 +0400, Evgeniy Polyakov wrote: > On Thu, Jun 07, 2007 at 06:23:16PM -0400, jamal (hadi@cyberus.ca) wrote: > > I believe both are called with no lock. The idea is to avoid the lock > > entirely when unneeded. That code may end up finding that the packet [..] > + netif_tx_lock_bh(odev); > + if (!netif_queue_stopped(odev)) { > + > + idle_start = getCurUs(); > + pkt_dev->tx_entered++; > + ret = odev->hard_batch_xmit(&odev->blist, odev); [..] > The same applies to *_gso case. > You missed an important piece which is grabbing of __LINK_STATE_QDISC_RUNNING > Without lock that would be wrong - it accesses hardware. We are achieving the goal of only a single CPU entering that path. Are you saying that is not good enough? > I only saw results Krishna posted, Ok, sorry - i thought you saw the git log or earlier results where other things are captured. > and i also do not know, what service demand is :) >>From the explanation seems to be how much cpu was used while sending. Do you have any suggestions for computing cpu use? in pktgen i added code to count how many microsecs were used in transmitting. > Result looks good, but I still do not understand how it appeared, that > is why I'm not that excited about idea - I just do not know it in > details. To add to KKs explanation on other email: Essentially the value is in amortizing the cost of barriers and IO per packet. For example the queue lock is held/released only once per X packets. DMA kicking which includes both a PCI IO write and mbs is done only once per X packets. There are still a lot of room for improvement of such IO; cheers, jamal