From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752737AbaHXOqG (ORCPT ); Sun, 24 Aug 2014 10:46:06 -0400 Received: from mail-wi0-f173.google.com ([209.85.212.173]:43957 "EHLO mail-wi0-f173.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752510AbaHXOqF (ORCPT ); Sun, 24 Aug 2014 10:46:05 -0400 Date: Sun, 24 Aug 2014 16:45:59 +0200 From: Ingo Molnar To: Frederic Weisbecker Cc: LKML , Viresh Kumar , Thomas Gleixner , Preeti U Murthy Subject: Re: [GIT PULL] nohz: Fix double clock write on idle ticks Message-ID: <20140824144559.GA9455@gmail.com> References: <1408727373-8166-1-git-send-email-frederic@kernel.org> <20140824100734.GA1309@gmail.com> MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Disposition: inline Content-Transfer-Encoding: 8bit In-Reply-To: User-Agent: Mutt/1.5.23 (2014-03-12) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org * Frederic Weisbecker wrote: > Le dimanche 24 août 2014, Ingo Molnar a écrit : > > > * Frederic Weisbecker > wrote: > > > > > From: Frederic Weisbecker > > > > > > > Ingo, > > > > > > Please pull the nohz/drop-double-write-v3 branch that can be found at: > > > > > > git:// > > git.kernel.org/pub/scm/linux/kernel/git/frederic/linux-dynticks.git > > > nohz/drop-double-write-v3 > > > > > > Although the 1st patch fixes a bug, it was a longstanding issue so > > > that branch doesn't need to be applied in 3.17. The next merge window > > > looks better. > > > > > > -- > > > The tick reschedules itself unconditionally. It's relevant in periodic > > > mode but not in dynticks mode where it results in spurious double clock > > > writes and even spurious periodic behaviour for low-res case. > > > > > > This set fixes that: > > > > > > * 1st patch removes low-res periodic tick rescheduling in nohz mode. > > > This fixes spurious periodic behaviour. > > > > > > * 2nd patch does the same for high-res mode. Here there is no such > > > spurious periodic behaviour but it still spares a double clock write > > > in some cases. > > > --- > > > > > > > > > Thanks, > > > Frederic > > > --- > > > > > > Viresh Kumar (2): > > > nohz: Fix spurious periodic tick behaviour in low-res dynticks mode > > > nohz: Avoid tick's double reprogramming in highres mode > > > > > > > > > kernel/time/tick-sched.c | 8 ++++++++ > > > 1 file changed, 8 insertions(+) > > > > > These fixes are pretty small - any objections > > > against putting it into > > > timers/urgent and including it in v3.17? > > > OK sounds good to me. They are easily reverted if any problem arises. On a second thought, 'spurious periodic behavior' is really just sub-optimal full-nohz behavior - it's not an outright user-noticeable bug, correct? Thanks, Ingo