From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752065AbdHARku (ORCPT ); Tue, 1 Aug 2017 13:40:50 -0400 Received: from mail-qk0-f169.google.com ([209.85.220.169]:37978 "EHLO mail-qk0-f169.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751832AbdHARks (ORCPT ); Tue, 1 Aug 2017 13:40:48 -0400 MIME-Version: 1.0 In-Reply-To: <882c304e-0677-1f08-cca7-78b1c38bcafe@redhat.com> References: <1501592128-11334-1-git-send-email-prarit@redhat.com> <882c304e-0677-1f08-cca7-78b1c38bcafe@redhat.com> From: John Stultz Date: Tue, 1 Aug 2017 10:40:47 -0700 Message-ID: Subject: Re: [PATCH v2] printk: Add boottime and real timestamps To: Prarit Bhargava Cc: lkml , Mark Salyzyn , Jonathan Corbet , Petr Mladek , Sergey Senozhatsky , Steven Rostedt , Thomas Gleixner , Stephen Boyd , Andrew Morton , Greg Kroah-Hartman , "Paul E. McKenney" , Christoffer Dall , Deepa Dinamani , Ingo Molnar , Joel Fernandes , Kees Cook , Peter Zijlstra , Geert Uytterhoeven , "Luis R. Rodriguez" , Nicholas Piggin , "Jason A. Donenfeld" , Olof Johansson , Josh Poimboeuf , linux-doc@vger.kernel.org Content-Type: text/plain; charset="UTF-8" Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Tue, Aug 1, 2017 at 10:35 AM, Prarit Bhargava wrote: > > > On 08/01/2017 01:00 PM, John Stultz wrote: >> Is there a reason your not also adding PRINTK_TIME_BOOT here (which to >> me would be more generally useful then REAL or MONO)? > > REAL has been useful to me in debug cases where events on the system were timed > to the wall clock (ex cron job running at 3AM). I hadn't really thought much > about using BOOT TBH because MONO seemed to work just fine. > > Mark Salyzyn, did you want BOOT or MONO? I know Mark has specific formatting needs, so I'm not going to speak for him, but BOOT is actually a nice improvement over MONO, since it includes suspend time, and avoids inconsistencies due to userspace tweaking the time, which REALTIME has. So for debugging suspend related issues and being able to understand how much time a system has spent in suspend vs not it would be quite useful. thanks -john