Return-Path: <linux-kernel-owner@vger.kernel.org>
Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand
        id S1754474AbdDLO6G (ORCPT <rfc822;w@1wt.eu>);
        Wed, 12 Apr 2017 10:58:06 -0400
Received: from Galois.linutronix.de ([146.0.238.70]:33961 "EHLO
        Galois.linutronix.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
        with ESMTP id S1752412AbdDLO6B (ORCPT
        <rfc822;linux-kernel@vger.kernel.org>);
        Wed, 12 Apr 2017 10:58:01 -0400
Date: Wed, 12 Apr 2017 16:57:58 +0200 (CEST)
From: Thomas Gleixner <tglx@linutronix.de>
To: Frederic Weisbecker <fweisbec@gmail.com>
cc: Wanpeng Li <kernellwp@gmail.com>, Mike Galbraith <efault@gmx.de>,
        Rik van Riel <riel@redhat.com>,
        Luiz Capitulino <lcapitulino@redhat.com>,
        "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
        Peter Zijlstra <peterz@infradead.org>
Subject: Re: [BUG nohz]: wrong user and system time accounting
In-Reply-To: <20170412131818.GB21309@lerouge>
Message-ID: <alpine.DEB.2.20.1704121654150.1874@nanos>
References: <20170323165512.60945ac6@redhat.com> <CANRm+CxcgSP2-x+A822DmHLvFLzFmTptS6oYwYtwVdErTpiB=Q@mail.gmail.com> <1490636129.8850.76.camel@redhat.com> <20170328132406.7d23579c@redhat.com> <20170329131656.1d6cb743@redhat.com> <1490818125.28917.11.camel@redhat.com>
 <1490848051.4167.57.camel@gmx.de> <CANRm+CzfqkU6iV1BFk3PmWvy7MOsYH1mSwejJXMkvWW9C5ngwg@mail.gmail.com> <alpine.DEB.2.20.1704111523100.1772@nanos> <20170412131818.GB21309@lerouge>
User-Agent: Alpine 2.20 (DEB 67 2015-01-07)
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Sender: linux-kernel-owner@vger.kernel.org
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org
Content-Length: 577
Lines: 17

On Wed, 12 Apr 2017, Frederic Weisbecker wrote:
> On Tue, Apr 11, 2017 at 04:22:48PM +0200, Thomas Gleixner wrote:
> > It's not different from the current jiffies based stuff at all. Same
> > failure mode.
> 
> Yes you're right, I got confused again. So to fix this we could do our snapshots
> at a frequency lower than HZ but still high enough to avoid overhead.
> 
> Something like TICK_NSEC / 2 ?

If you are using TSC anyway then you can do proper accumulation for both
system and user and only account the data when the accumulation is more
than a jiffie.

Thanks,

	tglx