From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752640Ab2GZM53 (ORCPT ); Thu, 26 Jul 2012 08:57:29 -0400 Received: from mail-bk0-f46.google.com ([209.85.214.46]:45313 "EHLO mail-bk0-f46.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752542Ab2GZM52 (ORCPT ); Thu, 26 Jul 2012 08:57:28 -0400 Date: Thu, 26 Jul 2012 14:57:21 +0200 From: Ingo Molnar To: John Stultz Cc: lkml , Peter Zijlstra , Richard Cochran , Prarit Bhargava , Thomas Gleixner Subject: Re: [PATCH 2/2] time: Cleanup offs_real/wall_to_mono and offs_boot/total_sleep_time updates Message-ID: <20120726125721.GE24672@gmail.com> References: <1342660753-10382-1-git-send-email-john.stultz@linaro.org> <1342660753-10382-3-git-send-email-john.stultz@linaro.org> <20120719093305.GA27086@gmail.com> <500DA6A4.1000009@linaro.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <500DA6A4.1000009@linaro.org> User-Agent: Mutt/1.5.21 (2010-09-15) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org * John Stultz wrote: > On 07/19/2012 02:33 AM, Ingo Molnar wrote: > >* John Stultz wrote: > > > >>+static void tk_set_sleep_time(struct timekeeper *tk, struct timespec t) > >>+{ > >>+ /* Verify consistency before modifying */ > >>+ WARN_ON_ONCE(tk->offs_boot.tv64 != > >>+ timespec_to_ktime(tk->total_sleep_time).tv64); > >asserts like this can be put into a single line - we rarely need > >to read them if they don't trigger - and making them > >non-intrusive oneliners is a bonus. > > Ack. > > >> @@ -456,8 +478,8 @@ int timekeeping_inject_offset(struct timespec *ts) > >> tk_xtime_add(&timekeeper, ts); > >>- timekeeper.wall_to_monotonic = > >>- timespec_sub(timekeeper.wall_to_monotonic, *ts); > >>+ tk_set_wall_to_mono(&timekeeper, > >>+ timespec_sub(timekeeper.wall_to_monotonic, *ts)); > >There's a *lot* of unnecessary ugliness here and in many other > >places in kernel/time/ due to repeating patterns of > >"timekeeper.", which gets repeated many times and blows up line > >length. > > > >Please add this to the highest level (system call, irq handler, > >etc.) code: > > > > struct timekeeper *tk = &timekeeper; > > > >and pass that down to lower levels. The tk-> pattern will be the > >obvious thing to type in typical time handling functions. > > > >This increases readability and clarifies the data flow and might > >bring other advantages in the future. > > Sounds good. Are you ok if this is done in a follow-on patch? Yeah, sure - the code is going in the right direction in general, so no hurry. Thanks, Ingo