From mboxrd@z Thu Jan 1 00:00:00 1970 From: =?ISO-8859-2?Q?Pawe=B3_Staszewski?= Subject: Re: rib_trie / Fix inflate_threshold_root. Now=15 size=11 bits Date: Mon, 29 Jun 2009 11:43:54 +0200 Message-ID: <4A488CDA.1000804@itcare.pl> References: <19012.37515.146191.198843@robur.slu.se> <20090627192057.GA5041@ami.dom.local> <19015.20051.432572.460207@robur.slu.se> <20090628143511.GA20541@ami.dom.local> <4A478D02.10805@itcare.pl> <4A4790C3.508@itcare.pl> <20090628213602.GA4584@ami.dom.local> <4A48766B.8020206@itcare.pl> <4A487FB0.7010903@itcare.pl> <20090629092703.GB4712@ff.dom.local> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-2; format=flowed Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: David Miller , Robert Olsson , Robert Olsson , "Jorge Boncompte [DTI2]" , Eric Dumazet , Robert Olsson , Linux Network Development list To: Jarek Poplawski Return-path: Received: from smtp.iq.pl ([86.111.241.19]:45319 "EHLO smtp.iq.pl" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1758727AbZF2Jny (ORCPT ); Mon, 29 Jun 2009 05:43:54 -0400 In-Reply-To: <20090629092703.GB4712@ff.dom.local> Sender: netdev-owner@vger.kernel.org List-ID: Jarek Poplawski pisze: > On Mon, Jun 29, 2009 at 10:47:44AM +0200, Pawe=B3 Staszewski wrote: > =20 >> But >> With all this patches i have the same problem with CPU load >> Every time when route cache entries are purged cpu load is increasin= g =20 >> from 1% to 40 / 80% it depends >> >> I see that on 64bit machine when route cache entries are going down = i =20 >> have almost 80% load on each cpu where ethernet card is binded by =20 >> smp_affinity >> But on 32bit machine cpu load reported by mpstat is half that on 64b= it =20 >> machine >> here is example from 32bit machine ( mpstat + rtstat -k entries ) >> >> Linux 2.6.29.5 (TM_02_C1) 06/29/09 _i686_ (2 CPU) >> >> 12:36:54 CPU %usr %nice %sys %iowait %irq %soft %s= teal =20 >> %guest %idle RT CACHE ENTRIES (from rtstat) >> 12:36:57 all 0.00 0.00 0.00 0.00 1.51 15.08 = 0.00 =20 >> 0.00 83.42 83346 >> =20 > > I guess Eric is thinking about this. Btw., two little suggestions: > it should be easier to track if these route cache reports stay in its > starting thread ("weird problem"?), and if you could send these > stats/logs as attachements or turn off line wrapping, please? ;-) > > Thanks, > Jarek P. > -- > To unsubscribe from this list: send the line "unsubscribe netdev" in > the body of a message to majordomo@vger.kernel.org > More majordomo info at http://vger.kernel.org/majordomo-info.html > > > =20 Sorry Jarek for combining problems :) And yes i will apply next stats in attachements :)