From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Paul E. McKenney" Subject: Re: [PATCH] fix misplaced mb() in rcu_enter/exit_nohz() Date: Mon, 17 Mar 2008 12:06:05 -0700 Message-ID: <20080317190605.GG10955@linux.vnet.ibm.com> References: <20080317010821.GA29875@linux.vnet.ibm.com> <20080317183047.GA188@tv-sign.ru> Reply-To: paulmck@linux.vnet.ibm.com Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: linux-kernel@vger.kernel.org, rostedt@goodmis.org, linux-rt-users@vger.kernel.org, mingo@elte.hu, ego@in.ibm.com, dipankar@in.ibm.com, tytso@us.ibm.com, dvhltc@us.ibm.com, akpm@linux-foundation.org, josh@freedesktop.org, tglx@linutronix.de, niv@us.ibm.com To: Oleg Nesterov Return-path: Received: from e31.co.us.ibm.com ([32.97.110.149]:39524 "EHLO e31.co.us.ibm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751391AbYCQTGY (ORCPT ); Mon, 17 Mar 2008 15:06:24 -0400 Content-Disposition: inline In-Reply-To: <20080317183047.GA188@tv-sign.ru> Sender: linux-rt-users-owner@vger.kernel.org List-ID: On Mon, Mar 17, 2008 at 09:30:47PM +0300, Oleg Nesterov wrote: > On 03/16, Paul E. McKenney wrote: > > > > In the process of writing up the mechanical proof of correctness for the > > dynticks/preemptable-RCU interface, I noticed misplaced memory barriers > > in rcu_enter_nohz() and rcu_exit_nohz(). > > Can't comment this patch, there is no rcu_enter_nohz() in my rcupreempt.h ;) It is in 2.6.25-rc4 and later. ;-) > I'm not sure the code below is up to date, but what I have in > arch/s390/kernel/time.c is: > > stop_hz_timer: > > cpu_set(cpu, nohz_cpu_mask); > > if (rcu_needs_cpu(cpu) || local_softirq_pending()) { > cpu_clear(cpu, nohz_cpu_mask); > return; > } > > Don't we need smp_mb() after cpu_set() ? S390's memory model is quite strong, so it might not be needed. In any case, if needed, it goes -before- the cpu_set(), because the problems would arise if prior RCU read-side critical sections were to be reordered to follow this cpu_set(), right? Let's see... In S390, a store cannot be reordered to precede any prior load or store, so any preceding RCU read-side critical section would be seen by all CPUs as preceding the shift to nohz mode. Might be trouble for the opposite transition... But last I heard, the s390 guys were thinking in terms of moving to the generic dynticks model. If they really are doing so, then the above code goes away in any case. Thanx, Paul