From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1757024Ab3FKUKS (ORCPT ); Tue, 11 Jun 2013 16:10:18 -0400 Received: from mx1.redhat.com ([209.132.183.28]:48948 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756486Ab3FKUKM (ORCPT ); Tue, 11 Jun 2013 16:10:12 -0400 Date: Tue, 11 Jun 2013 22:06:00 +0200 From: Oleg Nesterov To: John Stultz Cc: Andrew Morton , "Eric W. Biederman" , Tomas Janousek , Tomas Smetana , Linux Kernel Mailing List Subject: Re: [PATCH 0/3] de_thread() should update ->real_start_time Message-ID: <20130611200600.GA15168@redhat.com> References: <20130610183300.GA14379@redhat.com> <20130611171300.GA7920@redhat.com> <51B7690C.8090805@linaro.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <51B7690C.8090805@linaro.org> User-Agent: Mutt/1.5.18 (2008-05-17) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 06/11, John Stultz wrote: > > On 06/11/2013 10:13 AM, Oleg Nesterov wrote: > >> But perhaps this is fine and even more correct? > > So I think it probably *makes more sense* to include suspend_time in the > elapsed runtime value being exported via bacct_add_tsk() and > do_acct_process(), but I unfortunately worry now any such change would > risk breaking userland expectations. > > The *actual* risk may be quite minor, so this could be one of those: > "Let the tree fall and if no one is there to hear it, fine" interface > breaks, but I'm not sure I'm eager enough to be the one proposing it. :) Yes, same thoughts here ;) Still it is ugly imho to keep task->start_time just for taskstats, and _I think_ nobody really cares. Perhaps I'll try to send the patch later... And look. It seems that ->ac_btime (Process Creation Time) in bacct_add_tsk() is obviously wrong anyway? So perhaps we can fix this and in this case we can also change the meaning of start_time. Oleg.