From mboxrd@z Thu Jan 1 00:00:00 1970 From: Peter Zijlstra Subject: Re: [PATCH 3/7] idle, thermal, acpi: Remove home grown idle implementations Date: Fri, 22 Nov 2013 12:33:00 +0100 Message-ID: <20131122113300.GM10022@twins.programming.kicks-ass.net> References: <20131120162736.508462614@infradead.org> <20131120165406.14fa0f09@ultegra> <20131121082151.GU10022@twins.programming.kicks-ass.net> <20131121160716.GT4138@linux.vnet.ibm.com> <528E32EF.2050300@linux.intel.com> <20131121191916.GZ4138@linux.vnet.ibm.com> <528E62D0.3010003@linux.intel.com> <20131121200717.GB4138@linux.vnet.ibm.com> <20131121161005.34150ab2@ultegra> <20131122042036.GL4138@linux.vnet.ibm.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Received: from merlin.infradead.org ([205.233.59.134]:43236 "EHLO merlin.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753064Ab3KVLdK (ORCPT ); Fri, 22 Nov 2013 06:33:10 -0500 Content-Disposition: inline In-Reply-To: <20131122042036.GL4138@linux.vnet.ibm.com> Sender: linux-pm-owner@vger.kernel.org List-Id: linux-pm@vger.kernel.org To: "Paul E. McKenney" Cc: Jacob Pan , Arjan van de Ven , lenb@kernel.org, rjw@rjwysocki.net, Eliezer Tamir , Chris Leech , David Miller , rui.zhang@intel.com, Mike Galbraith , Ingo Molnar , hpa@zytor.com, Thomas Gleixner , linux-kernel@vger.kernel.org, linux-pm@vger.kernel.org, "Rafael J. Wysocki" On Thu, Nov 21, 2013 at 08:20:36PM -0800, Paul E. McKenney wrote: > The 6ms to 25ms range should be just fine as far as normal RCU grace > periods are concerned. However, it does mean that expedited grace > periods could be delayed: They normally take a few tens of microseconds, > but if they were unlucky enough to show up during an idle injection, > they would be magnified by two to three orders of magnitude, which is > not pretty. > > Hence my suggestion of hooking into RCU on idle-injection start and end > so that RCU considers that time period to be idle. Just like it does > for user-mode execution on NO_HZ_FULL kernels, so I still don't see this > approach to be a problem. I must confess that I still don't understand > what Arjan doesn't like about it. Using these patches it would indeed use the RCU idle machinery as per the normal idle path. If you can I can add more WARN_ON()s in play_idle() to ensure we're not called while holding any RCU locks.