From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752039Ab3C0Qec (ORCPT ); Wed, 27 Mar 2013 12:34:32 -0400 Received: from 8bytes.org ([85.214.48.195]:47843 "EHLO mail.8bytes.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750905Ab3C0Qeb (ORCPT ); Wed, 27 Mar 2013 12:34:31 -0400 Date: Wed, 27 Mar 2013 17:34:27 +0100 From: Joerg Roedel To: "Paul E. McKenney" Cc: Peter Zijlstra , Borislav Petkov , Ingo Molnar , Arnaldo Carvalho de Melo , lkml , Stephane Eranian , Namhyung Kim , Jiri Olsa Subject: Re: BUG: using smp_processor_id() in preemptible [00000000] code: asm/8267 Message-ID: <20130327163427.GN30540@8bytes.org> References: <20130324115556.GA4866@pd.tnic> <20130324155924.GB4866@pd.tnic> <20130326183452.GC27518@pd.tnic> <20130327131532.GL30540@8bytes.org> <1364393852.5053.74.camel@laptop> <20130327143715.GY4379@linux.vnet.ibm.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20130327143715.GY4379@linux.vnet.ibm.com> User-Agent: Mutt/1.5.21 (2010-09-15) X-DSPAM-Result: Whitelisted X-DSPAM-Processed: Wed Mar 27 17:34:29 2013 X-DSPAM-Confidence: 0.9991 X-DSPAM-Probability: 0.0000 X-DSPAM-Signature: 51531f9523675553987564 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Wed, Mar 27, 2013 at 07:37:15AM -0700, Paul E. McKenney wrote: > On Wed, Mar 27, 2013 at 03:17:32PM +0100, Peter Zijlstra wrote: > > On Wed, 2013-03-27 at 14:15 +0100, Joerg Roedel wrote: > > > What makes me wonder here is that the code is preemptible in an > > > rcu_read_locked section. As far as I know preemption needs to be > > > disabled while holding the rcu_read_lock(). > > > > Nah, a long long time ago some -rt people complained to paulmck that > > keeping preemption disabled over all this RCU stuff was killing > > latencies. Paul liked the challenge and came up with some mind twisting > > stuff to make it work. > > What can I say? I was young and foolish. And I still am pretty > foolish. ;-) > > But yes, you are not required to disable preemption across > rcu_read_lock(), and rcu_read_lock() is not guaranteed to disable > preemption. So if you need preemption to be disabled, do it explicitly > with preempt_disable(), local_irq_save(), rcu_read_lock_sched(), > or whatever, because rcu_read_lock() isn't always going to disable > preemption. > > > If you're into that kind of pain, look at CONFIG_*_PREEMPT_RCU :-) > > Or just set CONFIG_PREEMPT=y, which will set CONFIG_TREE_PREEMPT_RCU=y > on CONFIG_SMP=y builds and will set CONFIG_TINY_PREEMPT_RCU=y otherwise. > (But please note that CONFIG_TINY_PREEMPT_RCU is going away, after which > CONFIG_PREEMPT=y will always set CONFIG_TREE_PREEMPT_RCU=y.) > > Thanx, Paul > > > But yeah, you need to have that stuff enabled before you can hit this > > particular snag. Interesting read, thanks guys. I think I should have a look into the tree-preempt implementation and try to understand it :) Joerg