From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755469Ab0AZXze (ORCPT ); Tue, 26 Jan 2010 18:55:34 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1754473Ab0AZXz3 (ORCPT ); Tue, 26 Jan 2010 18:55:29 -0500 Received: from tomts5-srv.bellnexxia.net ([209.226.175.25]:33657 "EHLO tomts5-srv.bellnexxia.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755470Ab0AZXzY (ORCPT ); Tue, 26 Jan 2010 18:55:24 -0500 X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: AsoEAGsLX0uuWOiG/2dsb2JhbACBO9gvhDcE Date: Tue, 26 Jan 2010 18:55:16 -0500 From: Mathieu Desnoyers To: Andi Kleen Cc: paulmck@linux.vnet.ibm.com, linux-kernel@vger.kernel.org, mingo@elte.hu, laijs@cn.fujitsu.com, dipankar@in.ibm.com, akpm@linux-foundation.org, josh@joshtriplett.org, dvhltc@us.ibm.com, niv@us.ibm.com, tglx@linutronix.de, peterz@infradead.org, rostedt@goodmis.org, Valdis.Kletnieks@vt.edu, dhowells@redhat.com Subject: Re: [PATCH RFC tip/core/rcu] accelerate grace period if last non-dynticked CPU Message-ID: <20100126235516.GA15855@Krystal> References: <20100125034816.GA14043@linux.vnet.ibm.com> <873a1sft9q.fsf@basil.nowhere.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline In-Reply-To: <873a1sft9q.fsf@basil.nowhere.org> X-Editor: vi X-Info: http://krystal.dyndns.org:8080 X-Operating-System: Linux/2.6.27.31-grsec (i686) X-Uptime: 18:49:22 up 41 days, 8:07, 4 users, load average: 0.18, 0.24, 0.15 User-Agent: Mutt/1.5.18 (2008-05-17) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org * Andi Kleen (andi@firstfloor.org) wrote: > "Paul E. McKenney" writes: > > Kind of offtopic to the original patch, but I couldn't > resist... > > > +config RCU_FAST_NO_HZ > > + bool "Accelerate last non-dyntick-idle CPU's grace periods" > > + depends on TREE_RCU && NO_HZ && SMP > > Having such a thing as a config option doesn't really make > any sense to me. Who would want to recompile their kernel > to enable/disable this? If anything it should be runtime, or better > just unconditionally on. > > In general RCU could probably reduce its number of "weird" > Kconfig options. > > While I think I have a better understanding of RCU than a lot > of normal users I often have no clue what to set there when > building a kernel. Maybe we could keep them under a CONFIG_DEBUG_RCU umbrella. Compiling out parts of the rcu options can be useful for debugging purposes, but I agree with you that end users should not be bothered with that much options when some of them are "obviously" wanted. OTOH, I understand that Paul seems to want to introduce new RCU features gradually, without hitting all kernel users with bugs in newer features. That's a sane approach to keep things generally stable, but maybe it is time to set some of the stabilized RCU options to default Y and move their config to the debug menu. Let's see what Paul has to say about this... Thanks, Mathieu > > -Andi > > -- > ak@linux.intel.com -- Speaking for myself only. -- Mathieu Desnoyers OpenPGP key fingerprint: 8CD5 52C3 8E3C 4140 715F BA06 3F25 A8FE 3BAE 9A68