From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <linux-kernel-owner@vger.kernel.org>
Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand
	id S1754156AbbCaCEl (ORCPT <rfc822;w@1wt.eu>);
	Mon, 30 Mar 2015 22:04:41 -0400
Received: from mail-wg0-f43.google.com ([74.125.82.43]:35538 "EHLO
	mail-wg0-f43.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
	with ESMTP id S1753915AbbCaCEi (ORCPT
	<rfc822;linux-kernel@vger.kernel.org>);
	Mon, 30 Mar 2015 22:04:38 -0400
Message-ID: <1427767475.3147.32.camel@gmail.com>
Subject: Re: [PATCH] watchdog: nohz: don't run watchdog on nohz_full cores
From: Mike Galbraith <umgwanakikbuti@gmail.com>
To: Don Zickus <dzickus@redhat.com>
Cc: cmetcalf@ezchip.com, Andrew Morton <akpm@linux-foundation.org>,
        Andrew Jones <drjones@redhat.com>,
        chai wen <chaiw.fnst@cn.fujitsu.com>, Ingo Molnar <mingo@kernel.org>,
        Ulrich Obergfell <uobergfe@redhat.com>,
        Fabian Frederick <fabf@skynet.be>, Aaron Tomlin <atomlin@redhat.com>,
        Ben Zhang <benzh@chromium.org>, Christoph Lameter <cl@linux.com>,
        Frederic Weisbecker <fweisbec@gmail.com>,
        Gilad Ben-Yossef <gilad@benyossef.com>,
        Steven Rostedt <rostedt@goodmis.org>,
        open list <linux-kernel@vger.kernel.org>
Date: Tue, 31 Mar 2015 04:04:35 +0200
In-Reply-To: <20150330191245.GO162412@redhat.com>
References: <1427741465-15747-1-git-send-email-cmetcalf@ezchip.com>
	 <20150330191245.GO162412@redhat.com>
Content-Type: text/plain; charset="UTF-8"
X-Mailer: Evolution 3.12.11 
Mime-Version: 1.0
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org

On Mon, 2015-03-30 at 15:12 -0400, Don Zickus wrote:
> On Mon, Mar 30, 2015 at 02:51:05PM -0400, cmetcalf@ezchip.com wrote:
> > From: Chris Metcalf <cmetcalf@ezchip.com>
> > 
> > Running watchdog can be a helpful debugging feature on regular
> > cores, but it's incompatible with nohz_full, since it forces
> > regular scheduling events.  Accordingly, just exit out immediately
> > from any nohz_full core.
> > 
> > An alternate approach would be to add a flags field or function to
> > smp_hotplug_thread to control on which cores the percpu threads
> > are created, but it wasn't clear that much mechanism was useful.
> 
> Hi Chris,
> 
> It seems like the correct solution would be to hook into the idle_loop
> somehow.  If the cpu is idle, then it seems unlikely that a lockup could
> occur.
> 
> My fear with this apporach is a lockup would occur on the nohz cpu and it
> would go undetected because that cpu is disabled.  Further no printk is
> thrown out to even indicate a cpu is disabled making it more difficult to
> debug.

Hm, I don't see why this is needed, for debugging/testing you turn it
on, when you set up for critical operation, you turn it off.

A bigger deal is the clocksource watchdog methinks.  Measurement
inspired me to make it dead yesterday.

	-Mike