From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756144Ab3LSUQn (ORCPT ); Thu, 19 Dec 2013 15:16:43 -0500 Received: from merlin.infradead.org ([205.233.59.134]:36149 "EHLO merlin.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751357Ab3LSUQm (ORCPT ); Thu, 19 Dec 2013 15:16:42 -0500 Date: Thu, 19 Dec 2013 21:16:34 +0100 From: Peter Zijlstra To: "H. Peter Anvin" Cc: mingo@kernel.org, linux-kernel@vger.kernel.org, arjan@linux.intel.com, rui.zhang@intel.com, bitbucket@online.de, rafael.j.wysocki@intel.com, lenb@kernel.org, tglx@linutronix.de, jacob.jun.pan@linux.intel.com, hpa@linux.intel.com, linux-tip-commits@vger.kernel.org Subject: Re: [tip:x86/idle] x86, acpi, idle: Restructure the mwait idle routines Message-ID: <20131219201634.GR16438@laptop.programming.kicks-ass.net> References: <20131212141654.616820819@infradead.org> <52B35369.6070208@zytor.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <52B35369.6070208@zytor.com> User-Agent: Mutt/1.5.21 (2012-12-30) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Thu, Dec 19, 2013 at 12:13:29PM -0800, H. Peter Anvin wrote: > On 12/19/2013 12:09 PM, tip-bot for Peter Zijlstra wrote: > > diff --git a/drivers/thermal/intel_powerclamp.c b/drivers/thermal/intel_powerclamp.c > > index 8f181b3..e8275f2 100644 > > --- a/drivers/thermal/intel_powerclamp.c > > +++ b/drivers/thermal/intel_powerclamp.c > > @@ -438,9 +438,7 @@ static int clamp_thread(void *arg) > > */ > > local_touch_nmi(); > > stop_critical_timings(); > > - __monitor((void *)¤t_thread_info()->flags, 0, 0); > > - cpu_relax(); /* allow HT sibling to run */ > > - __mwait(eax, ecx); > > + mwait_idle_with_hints(eax, ecx); > > start_critical_timings(); > > atomic_inc(&idle_wakeup_counter); > > } > > > > Should this cpu_relax() be in the common code as well? I don't think so; it seems weird to allow a sibling some time between monitor and wait because the moment wait stalls the cpu the sibling gets all the time it wants.