From mboxrd@z Thu Jan 1 00:00:00 1970 From: Peter Zijlstra Subject: Re: 50 Watt idle power regression bisected to Linux-3.10 Date: Thu, 12 Dec 2013 16:51:53 +0100 Message-ID: <20131212155153.GK13532@twins.programming.kicks-ass.net> References: <1386764955.12005.60.camel@marge.simpson.net> <20131211124352.GB21999@twins.programming.kicks-ass.net> <20131211134048.GH21683@pd.tnic> <20131211145655.GB4510@gmail.com> <20131211164318.GA2480@laptop.programming.kicks-ass.net> <20131211175036.GC12431@gmail.com> <52A8F073.9040500@zytor.com> <20131212085143.GC21999@twins.programming.kicks-ass.net> <20131212132846.GA16750@gmail.com> <52A9D111.3090409@zytor.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Received: from merlin.infradead.org ([205.233.59.134]:39338 "EHLO merlin.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751452Ab3LLPwP (ORCPT ); Thu, 12 Dec 2013 10:52:15 -0500 Content-Disposition: inline In-Reply-To: <52A9D111.3090409@zytor.com> Sender: linux-pm-owner@vger.kernel.org List-Id: linux-pm@vger.kernel.org To: "H. Peter Anvin" Cc: Ingo Molnar , Borislav Petkov , Mike Galbraith , Thomas Gleixner , Len Brown , Linux PM list , "linux-kernel@vger.kernel.org" , Jeremy Eder , x86@kernel.org On Thu, Dec 12, 2013 at 07:06:57AM -0800, H. Peter Anvin wrote: > On 12/12/2013 05:28 AM, Ingo Molnar wrote: > >> > >> The reason we use thread_info::flags is because we need to write > >> TIF_NEED_RESCHED into it to wake up anyhow. > >> > >> Using another cacheline would mean the wakeup path would need to write a > >> second cross cpu cacheline -- that is badness too. > >> > >> So no, I don't think we want to listen to another line. > > > > Right, okay, so that's the implicit wakeup. However, I would think the > CLFLUSH would hurt a lot more. Maybe, but still, who cares? Its only a few broken cpus that actually need the clflush, normal cpus do not. We should not optimize for the broken case.