From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752366Ab1GNF02 (ORCPT ); Thu, 14 Jul 2011 01:26:28 -0400 Received: from e6.ny.us.ibm.com ([32.97.182.146]:56560 "EHLO e6.ny.us.ibm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752069Ab1GNF01 (ORCPT ); Thu, 14 Jul 2011 01:26:27 -0400 Date: Wed, 13 Jul 2011 22:26:22 -0700 From: "Paul E. McKenney" To: Peter Zijlstra Cc: Dave Jones , Linux Kernel , Ingo Molnar Subject: Re: lockdep circular locking error (rcu_node_level_0 vs rq->lock) Message-ID: <20110714052622.GA5366@linux.vnet.ibm.com> Reply-To: paulmck@linux.vnet.ibm.com References: <20110711231932.GB20367@redhat.com> <1310502014.2309.7.camel@laptop> <20110712205126.GM2326@linux.vnet.ibm.com> <1310507047.2309.10.camel@laptop> <20110712225443.GS2326@linux.vnet.ibm.com> <1310545448.14978.40.camel@twins> <20110713163309.GC2355@linux.vnet.ibm.com> <20110714042932.GA2932@linux.vnet.ibm.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20110714042932.GA2932@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 Wed, Jul 13, 2011 at 09:29:32PM -0700, Paul E. McKenney wrote: > On Wed, Jul 13, 2011 at 09:33:09AM -0700, Paul E. McKenney wrote: [ . . . ] > Hmmm... Something is confused. I get boot time hangs with the > occasional stack overflow, whether or not I add my patch on top. > Left to myself, I would try applying your patch incrementally, and > also disabling irqs before rcu_read_lock() and enabling them after > rcu_read_unlock(), as this prevents rcu_read_unlock() from ever getting > to the rcu_read_unlock_special() slowpath. And the real question... Why did the rcu_read_unlock() called from cpuacct_charge() enter rcu_read_unlock_special() in the first place? If the runqueue lock is held, irqs should be disabled, so how did the task get blocked or interrupted? I need to get some sleep then figure out what is going on here. Thanx, Paul