Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S933187Ab1EMSaQ (ORCPT ); Fri, 13 May 2011 14:30:16 -0400 Received: from hrndva-omtalb.mail.rr.com ([71.74.56.124]:39011 "EHLO hrndva-omtalb.mail.rr.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S933056Ab1EMSaP (ORCPT ); Fri, 13 May 2011 14:30:15 -0400 X-Authority-Analysis: v=1.1 cv=ou1QuR4lBR9YeJgEH9ccYmbAdaWqVVq3lOvCKJtMpGM= c=1 sm=0 a=wom5GMh1gUkA:10 a=2uYajzE4MAAA:10 a=Rj1_iGo3bfgA:10 a=kj9zAlcOel0A:10 a=g3F5VGk0NOMZWSIEWMgijA==:17 a=Z4Rwk6OoAAAA:8 a=VV-r4apBFXeysI0C4VIA:9 a=CjuIK1q_8ugA:10 a=jbrJJM5MRmoA:10 a=g3F5VGk0NOMZWSIEWMgijA==:117 X-Cloudmark-Score: 0 X-Originating-IP: 70.123.158.191 Date: Fri, 13 May 2011 13:30:12 -0500 From: "Serge E. Hallyn" To: Linus Torvalds Cc: "Serge E. Hallyn" , "Serge E. Hallyn" , "Eric W. Biederman" , Daniel Lezcano , David Howells , James Morris , Andrew Morton , Linux Kernel Mailing List , containers@lists.linux-foundation.org, Al Viro Subject: Re: acl_permission_check: disgusting performance Message-ID: <20110513183012.GA31958@mail.hallyn.com> References: <20110513025013.GA13209@mail.hallyn.com> <20110513040214.GA25270@mail.hallyn.com> <20110513131904.GA2519@mail.hallyn.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.5.21 (2010-09-15) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1362 Lines: 41 Quoting Linus Torvalds (torvalds@linux-foundation.org): > On Fri, May 13, 2011 at 9:16 AM, Linus Torvalds > wrote: > > > > Looks ok to me. And generates good code for acl_permission_check > > without CONFIG_USER_NS. > > > > I'll see how much that function drops on the kernel profiles.. > > Yup, looking good. > > For my "kernel make with no changes" workload, it dropped from > > 1.28% make [kernel.kallsyms] [k] acl_permission_check > > to > > 0.88% make [kernel.kallsyms] [k] > acl_permission_check > > which is pretty much exactly the expected 30% drop from no longer > having that expensive load of user_ns. > > Of course, that 30% improvement is just a 0.4% performance improvement > in the big picture, but hey, almost half a percentage point on a real > load from just one single function in the kernel is definitely worth > doing. That's great, thanks for the help. > Do you want to carry this for 2.6.40, or should I just apply it? It makes no user-visible difference so I'd say just apply it. thanks, -serge -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/