From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932865Ab0LTQva (ORCPT ); Mon, 20 Dec 2010 11:51:30 -0500 Received: from e5.ny.us.ibm.com ([32.97.182.145]:47895 "EHLO e5.ny.us.ibm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932292Ab0LTQv3 (ORCPT ); Mon, 20 Dec 2010 11:51:29 -0500 Date: Mon, 20 Dec 2010 08:51:18 -0800 From: "Paul E. McKenney" To: Lai Jiangshan Cc: linux-kernel@vger.kernel.org, mingo@elte.hu, dipankar@in.ibm.com, akpm@linux-foundation.org, mathieu.desnoyers@polymtl.ca, josh@joshtriplett.org, niv@us.ibm.com, tglx@linutronix.de, peterz@infradead.org, rostedt@goodmis.org, Valdis.Kletnieks@vt.edu, dhowells@redhat.com, eric.dumazet@gmail.com, darren@dvhart.com, Frederic Weisbecker , Peter Zijlstra Subject: Re: [PATCH RFC tip/core/rcu 15/20] rcu: Keep gpnum and completed fields synchronized Message-ID: <20101220165118.GI2143@linux.vnet.ibm.com> Reply-To: paulmck@linux.vnet.ibm.com References: <20101217205433.GA10199@linux.vnet.ibm.com> <1292619291-2468-15-git-send-email-paulmck@linux.vnet.ibm.com> <4D0EBBCF.2070305@cn.fujitsu.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <4D0EBBCF.2070305@cn.fujitsu.com> User-Agent: Mutt/1.5.20 (2009-06-14) X-Content-Scanned: Fidelis XPS MAILER Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Mon, Dec 20, 2010 at 10:13:35AM +0800, Lai Jiangshan wrote: > On 12/18/2010 04:54 AM, Paul E. McKenney wrote: > > From: Frederic Weisbecker > > > > When a CPU that was in an extended quiescent state wakes > > up and catches up with grace periods that remote CPUs > > completed on its behalf, we update the completed field > > but not the gpnum that keeps a stale value of a backward > > grace period ID. > > > > Later, note_new_gpnum() will interpret the shift between > > the local CPU and the node grace period ID as some new grace > > period to handle and will then start to hunt quiescent state. > > > > But if every grace periods have already been completed, this > > interpretation becomes broken. And we'll be stuck in clusters > > of spurious softirqs because rcu_report_qs_rdp() will make > > this broken state run into infinite loop. > > > > The solution, as suggested by Lai Jiangshan, is to ensure that > > the gpnum and completed fields are well synchronized when we catch > > up with completed grace periods on their behalf by other cpus. > > This way we won't start noting spurious new grace periods. > > > > Suggested-by: Lai Jiangshan > > Signed-off-by: Frederic Weisbecker > > Cc: Paul E. McKenney > > Cc: Ingo Molnar > > Cc: Thomas Gleixner > > Cc: Peter Zijlstra > > Cc: Steven Rostedt > Signed-off-by: Paul E. McKenney > > --- > > kernel/rcutree.c | 9 +++++++++ > > 1 files changed, 9 insertions(+), 0 deletions(-) > > > > diff --git a/kernel/rcutree.c b/kernel/rcutree.c > > index 916f42b..8105271 100644 > > --- a/kernel/rcutree.c > > +++ b/kernel/rcutree.c > > @@ -680,6 +680,15 @@ __rcu_process_gp_end(struct rcu_state *rsp, struct rcu_node *rnp, struct rcu_dat > > rdp->completed = rnp->completed; > > > > /* > > + * If we were in an extended quiescent state, we may have > > + * missed some grace periods that others CPUs took care on > > + * our behalf. Catch up with this state to avoid noting > > + * spurious new grace periods. > > + */ > > + if (rdp->completed > rdp->gpnum) > > + rdp->gpnum = rdp->completed; > > Need to use ULONG_CMP_LT(rdp->gpnum, rdp->completed) instead. You are quite correct! And the next patch in this series made exactly that change. Thanx, Paul > > + > > + /* > > * If another CPU handled our extended quiescent states and > > * we have no more grace period to complete yet, then stop > > * chasing quiescent states. >