Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756628Ab2B1NHM (ORCPT ); Tue, 28 Feb 2012 08:07:12 -0500 Received: from e33.co.us.ibm.com ([32.97.110.151]:40757 "EHLO e33.co.us.ibm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752966Ab2B1NHK (ORCPT ); Tue, 28 Feb 2012 08:07:10 -0500 Date: Tue, 28 Feb 2012 05:05:52 -0800 From: "Paul E. McKenney" To: Yong Zhang Cc: Venki Pallipadi , Peter Zijlstra , Thomas Gleixner , Ingo Molnar , "H. Peter Anvin" , Suresh Siddha , Aaron Durbin , Paul Turner , linux-kernel@vger.kernel.org Subject: Re: [PATCH] Extend mwait idle to optimize away CAL and RES interrupts to an idle CPU -v1 Message-ID: <20120228130548.GA2465@linux.vnet.ibm.com> Reply-To: paulmck@linux.vnet.ibm.com References: <1329957415-15239-1-git-send-email-venki@google.com> <1329989454.24994.57.camel@twins> <20120224054150.GA17543@zhy> <20120226013253.GN2393@linux.vnet.ibm.com> <20120227090646.GA1112@zhy> <20120227170527.GD2463@linux.vnet.ibm.com> <20120228071255.GD1112@zhy> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20120228071255.GD1112@zhy> User-Agent: Mutt/1.5.21 (2010-09-15) X-Content-Scanned: Fidelis XPS MAILER x-cbid: 12022813-2398-0000-0000-0000049B5EB6 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 2686 Lines: 64 On Tue, Feb 28, 2012 at 03:12:55PM +0800, Yong Zhang wrote: > On Mon, Feb 27, 2012 at 09:05:27AM -0800, Paul E. McKenney wrote: > > On Mon, Feb 27, 2012 at 05:06:46PM +0800, Yong Zhang wrote: > > > On Sat, Feb 25, 2012 at 05:32:53PM -0800, Paul E. McKenney wrote: > > > > On Fri, Feb 24, 2012 at 01:41:50PM +0800, Yong Zhang wrote: > > > > > On Thu, Feb 23, 2012 at 11:34:11AM -0800, Venki Pallipadi wrote: > > > > > > On Thu, Feb 23, 2012 at 1:30 AM, Peter Zijlstra wrote: > > > > > > > Why not scheduler_ipi()? > > > > > > > > > > > > Was trying to avoid irq_enter/exit. As the work here is done in idle > > > > > > thread context, I though we could avoid enter/exit. > > > > > > > > > > It seems we could not. > > > > > At least RCU need it, see commit c5d753a55, otherwise we will get > > > > > warning like 'RCU used illegally from extended quiescent state!' > > > > > > > > If the use is tracing, then Steven Rostedt's patchset plus use of his > > > > _rcuidle() tracing variants handles this: > > > > > > > > https://lkml.org/lkml/2012/2/7/231 > > > > > > > > If this is instead algorithmic use of RCU, a set of patches I have queued > > > > up for 3.4 will be required. > > > > > > scheduler_ipi() doing more than tracing. Will look at your patches :) > > > > Ah! The key question is whether or not the code in question is called > > both from idle and from non-idle. > > In fact before this patch from Venki, the only call site of scheduler_ipi() > is resched irq handler. Then Venki introduce __scheduler_ipi()(which avoid > irq_enter()/irq_exit()) into cpu_idle(). So the answer is yes. Ah, that explains why I didn't see it in my testing. ;-) > But when I was testing this patch, I didn't see explicit warning on > illegal rcu usage. The reason maybe 1) there are no much rcu dereference > in scheduler_ipi(), but we indeed do tracing in it; 2) rq->lock provide > some kind of protection. > Maybe I'm overstraining, but it is potential danger. Did you have CONFIG_PROVE_RCU=y when testing? > But anyway, it's not an issue anymore since Venki removed __scheduler_ipi() > in his latest version. OK. > > This will be easiest if the code is > > called only from idle, in which case you should only need this one: > > > > https://lkml.org/lkml/2012/2/3/498 > > Hmm... Yeah, RCU_NONIDLE() could survive IMHO :) Seems like it might be needed sooner rather than later... Thanx, Paul -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/