From mboxrd@z Thu Jan 1 00:00:00 1970 From: Patrick McHardy Subject: Re: [RFC VLAN 00/10]: VLAN netlink support try 2 Date: Tue, 05 Jun 2007 23:12:23 +0200 Message-ID: <4665D1B7.5040902@trash.net> References: <20070605143650.23717.91261.sendpatchset@localhost.localdomain> <46658E84.1090301@candelatech.com> <4665AB57.7030605@trash.net> <4665C2CE.4050504@candelatech.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-15 Content-Transfer-Encoding: 7bit Cc: netdev@vger.kernel.org To: Ben Greear Return-path: Received: from stinky.trash.net ([213.144.137.162]:37937 "EHLO stinky.trash.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1765478AbXFEVMc (ORCPT ); Tue, 5 Jun 2007 17:12:32 -0400 In-Reply-To: <4665C2CE.4050504@candelatech.com> Sender: netdev-owner@vger.kernel.org List-Id: netdev.vger.kernel.org Ben Greear wrote: > Patrick McHardy wrote: > >> Ben Greear wrote: >> >>> Patrick McHardy wrote: >>> >>>> The last VLAN patchset was outdated, sorry about the mixup. These >>>> are the >>>> correct patches. The iproute patch I posted was correct, so no >>>> repost of >>>> that one. >>>> >>>> >>> >>> Is there any significant performance penalty in creating VLANs using >>> these patches? >>> >>> If you have a test station handy, could you let us know what 'time' >>> shows for a script >>> creating 4000 VLANs with the old vconfig/ioctl method v/s this new >>> method? >> >> >> >> Not significantly more, for 1000 VLANs I get: >> >> ip link add: >> >> real 0m22.836s >> user 0m0.100s >> sys 0m5.850s >> >> vconfig add: >> >> real 0m19.739s >> user 0m0.090s >> sys 0m3.600s > > > Thanks for doing the tests. That small performance drop seems > fine to me. I'll try to speed it up a bit more, my initial version needed something like 10s for 1000 VLANs. I suspect the iproute RTM_NEWLINK probing done for every (non-batched) operation adds quite significantly to the overhead.