From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1758138Ab1EMInO (ORCPT ); Fri, 13 May 2011 04:43:14 -0400 Received: from mx2.mail.elte.hu ([157.181.151.9]:37978 "EHLO mx2.mail.elte.hu" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1758113Ab1EMInL (ORCPT ); Fri, 13 May 2011 04:43:11 -0400 Date: Fri, 13 May 2011 10:42:53 +0200 From: Ingo Molnar To: Yinghai Lu Cc: paulmck@linux.vnet.ibm.com, linux-kernel@vger.kernel.org Subject: Re: [GIT PULL rcu/next] rcu commits for 2.6.40 Message-ID: <20110513084253.GE13647@elte.hu> References: <20110511201852.GC2258@linux.vnet.ibm.com> <4DCAF894.7030707@kernel.org> <4DCAFFD8.2080605@kernel.org> <4DCB157F.20202@kernel.org> <20110512060344.GB3191@elte.hu> <4DCB8BCD.1080607@kernel.org> <4DCB8F7A.90603@kernel.org> <20110512092013.GJ2258@linux.vnet.ibm.com> <4DCC52FB.6030500@kernel.org> <4DCC894D.3070204@kernel.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <4DCC894D.3070204@kernel.org> User-Agent: Mutt/1.5.20 (2009-08-17) X-ELTE-SpamScore: -2.0 X-ELTE-SpamLevel: X-ELTE-SpamCheck: no X-ELTE-SpamVersion: ELTE 2.0 X-ELTE-SpamCheck-Details: score=-2.0 required=5.9 tests=BAYES_00 autolearn=no SpamAssassin version=3.3.1 -2.0 BAYES_00 BODY: Bayes spam probability is 0 to 1% [score: 0.0000] Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org * Yinghai Lu wrote: > more info: > if only reverting: e59fb3120becfb36b22ddb8bd27d065d3cdca499 > > when compiled with gcc from opensuse 11.3 > [ 32.579585] cpu_dev_init done > [ 81.203193] memory_dev_init done > > but when compile that with gcc from fedora 14, then will less delay. > [ 32.404140] cpu_dev_init done > [ 37.602986] memory_dev_init done > > gcc version on opensuse 11.3 > gcc (SUSE Linux) 4.5.0 20100604 [gcc-4_5-branch revision 160292] > > gcc version on fedora 14: > gcc (GCC) 4.5.1 20100924 (Red Hat 4.5.1-4) Would it be possible to narrow down it a bit which commit causes the boot slowdown? known bad : 11c476f31a0f: net,rcu: convert call_rcu(prl_entry_destroy_rcu) to kfree known good: 0ee5623f9a6e: Linux 2.6.39-rc6 There's 76 commits inbetween so an about 6-step bisection ought to be able to narrow the source of regression down quite a bit ... The complication with that bisection will be the need to revert the other broken commit you found: e59fb3120bec: rcu: Decrease memory-barrier usage based on semi-formal proof Which is unfortunately very early in the series, so you'd have to revert this at almost every step of the bisection ... Thanks, Ingo