From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756227Ab1GJRfv (ORCPT ); Sun, 10 Jul 2011 13:35:51 -0400 Received: from e3.ny.us.ibm.com ([32.97.182.143]:40800 "EHLO e3.ny.us.ibm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755491Ab1GJRft (ORCPT ); Sun, 10 Jul 2011 13:35:49 -0400 Date: Sun, 10 Jul 2011 10:35:30 -0700 From: "Paul E. McKenney" To: julie Sullivan Cc: linux-kernel@vger.kernel.org Subject: Re: PROBLEM: 3.0-rc kernels unbootable since -rc3 Message-ID: <20110710173530.GA16954@linux.vnet.ibm.com> Reply-To: paulmck@linux.vnet.ibm.com References: <201107072047.14429.kernelmail.jms@gmail.com> <20110707195845.GD6014@linux.vnet.ibm.com> <201107072147.12751.kernelmail.jms@gmail.com> <20110708002903.GI6014@linux.vnet.ibm.com> <20110709100900.GB6014@linux.vnet.ibm.com> <20110710032510.GG6014@linux.vnet.ibm.com> <20110710171626.GK6014@linux.vnet.ibm.com> MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Disposition: inline Content-Transfer-Encoding: 8bit In-Reply-To: <20110710171626.GK6014@linux.vnet.ibm.com> User-Agent: Mutt/1.5.20 (2009-06-14) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Sun, Jul 10, 2011 at 10:16:26AM -0700, Paul E. McKenney wrote: > On Sun, Jul 10, 2011 at 05:38:25PM +0100, julie Sullivan wrote: > > > I added another patch onto the jms.2011.07.07a in the -rcu git tree. > > > If you get a chance to test it, please let me know how it goes. > > > > > >                                                        Thanx, Paul > > > > > > Right, adding this commit to the other six: > > > > commit 8a153670878242b8a95213d986f4922111321743 > > > > makes us freeze on boot again. > > Thank you again for the testing! This one will take some thought... But maybe not that much thought. My current guess is that your configuration is somehow causing some RCU callbacks to be executed during early boot, and that these callbacks are attempting to use kernel features that are not yet initialized. If my guess is correct, then the reason that enabling RCU_BOOST makes things work is that it prevents RCU callbacks from executing until after the scheduler is up and running. So I added another commit to the jms.2011.07.07a branch in the -rcu git tree. If my guesses are correct, this will allow your system to boot, possibly emitting a warning on the way up. When you get a chance to try this out, please let me know how it goes. Thanx, Paul