From mboxrd@z Thu Jan 1 00:00:00 1970 From: Eric Dumazet Subject: Re: Scalability of interface creation and deletion Date: Sat, 07 May 2011 18:26:29 +0200 Message-ID: <1304785589.3207.5.camel@edumazet-laptop> References: <891B02256A0667292521A4BF@Ximines.local> <1304770926.2821.1157.camel@edumazet-laptop> <0F4A638C2A523577CDBC295E@Ximines.local> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: netdev@vger.kernel.org To: Alex Bligh Return-path: Received: from mail-wy0-f174.google.com ([74.125.82.174]:52712 "EHLO mail-wy0-f174.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755783Ab1EGQ0d (ORCPT ); Sat, 7 May 2011 12:26:33 -0400 Received: by wya21 with SMTP id 21so3030287wya.19 for ; Sat, 07 May 2011 09:26:32 -0700 (PDT) In-Reply-To: <0F4A638C2A523577CDBC295E@Ximines.local> Sender: netdev-owner@vger.kernel.org List-ID: Le samedi 07 mai 2011 =C3=A0 16:26 +0100, Alex Bligh a =C3=A9crit : > On the current 8 core box I am testing, I see 280ms per interface > delete **even with only 10 interfaces**. I see 260ms with one > interface. I know doing lots of rcu sync stuff can be slow, but > 260ms to remove one veth pair sounds like more than rcu sync going > on. It sounds like a sleep (though I may not have found the > right one). I see no CPU load. >=20 Here, on 2.6.38 kernel (Ubuntu 11.04 provided, on my 2 core laptop) # time rmmod dummy real 0m0.111s user 0m0.000s sys 0m0.000s This removed my two dummy0/dummy1 devices. On another machine with a very recent kernel : $ modprobe dummy numdummies=3D1 $ ifconfig dummy0 192.168.46.46 up $ time rmmod dummy real 0m0.032s user 0m0.000s sys 0m0.001s $ uname -a Linux svivoipvnx001 2.6.39-rc6-00097-g6ac1576-dirty #550 SMP Sat May 7 00:12:26 CEST 2011 i686 i686 i386 GNU/Linux So 260ms is a bit too much, maybe you hit yet another bug.