From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752816AbaFKVOv (ORCPT ); Wed, 11 Jun 2014 17:14:51 -0400 Received: from youngberry.canonical.com ([91.189.89.112]:42584 "EHLO youngberry.canonical.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752001AbaFKVOu (ORCPT ); Wed, 11 Jun 2014 17:14:50 -0400 Message-ID: <5398C6C3.90405@canonical.com> Date: Wed, 11 Jun 2014 16:14:43 -0500 From: Dave Chiluk Reply-To: chiluk@canonical.com User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:24.0) Gecko/20100101 Thunderbird/24.5.0 MIME-Version: 1.0 To: "Eric W. Biederman" CC: paulmck@linux.vnet.ibm.com, Rafael Tinoco , linux-kernel@vger.kernel.org, davem@davemloft.net, Christopher Arges , Jay Vosburgh Subject: Re: Possible netns creation and execution performance/scalability regression since v3.8 due to rcu callbacks being offloaded to multiple cpus References: <20140611133919.GZ4581@linux.vnet.ibm.com> <539879B8.4010204@canonical.com> <20140611161857.GC4581@linux.vnet.ibm.com> <53989F7B.6000004@canonical.com> <874mzr41kf.fsf@x220.int.ebiederm.org> In-Reply-To: <874mzr41kf.fsf@x220.int.ebiederm.org> X-Enigmail-Version: 1.6 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 06/11/2014 03:46 PM, Eric W. Biederman wrote: > ip netns add also performs a bind mount so we get into all of the vfs > level locking as well. It's actually quite a bit worse than that as ip netns exec creates a new mount namespace as well. That being said, the vfs issues have been healthily mitigated by a number of recent patches by Al Viro to fs/namespace.c.