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 S1758222AbcFANhU (ORCPT <rfc822;w@1wt.eu>);
	Wed, 1 Jun 2016 09:37:20 -0400
Received: from mail-qk0-f173.google.com ([209.85.220.173]:33532 "EHLO
	mail-qk0-f173.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
	with ESMTP id S1752956AbcFANhR (ORCPT
	<rfc822;linux-kernel@vger.kernel.org>);
	Wed, 1 Jun 2016 09:37:17 -0400
From: Daniel Bristot de Oliveira <daolivei@redhat.com>
X-Google-Original-From: Daniel Bristot de Oliveira <bristot@redhat.com>
Subject: Re: [RFC PATCH 2/2] sched: sysctl: Panic on scheduling while atomic
To: Peter Zijlstra <peterz@infradead.org>
References: <cover.1464652607.git.bristot@redhat.com>
 <654ac2ba100f609917b68d1c6a33331ec9df8da4.1464652607.git.bristot@redhat.com>
 <20160601094544.GP3190@twins.programming.kicks-ass.net>
Cc: linux-kernel@vger.kernel.org, Jonathan Corbet <corbet@lwn.net>,
        "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>,
        Josh Triplett <josh@joshtriplett.org>,
        Steven Rostedt <rostedt@goodmis.org>,
        Mathieu Desnoyers <mathieu.desnoyers@efficios.com>,
        Lai Jiangshan <jiangshanlai@gmail.com>, Ingo Molnar <mingo@redhat.com>,
        Arnaldo Carvalho de Melo <acme@kernel.org>,
        "Luis Claudio R. Goncalves" <lgoncalv@redhat.com>
Message-ID: <e38a7f99-b3d1-adaf-26b3-800a6628d98f@redhat.com>
Date: Wed, 1 Jun 2016 10:37:11 -0300
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:45.0) Gecko/20100101
 Thunderbird/45.1.0
MIME-Version: 1.0
In-Reply-To: <20160601094544.GP3190@twins.programming.kicks-ass.net>
Content-Type: text/plain; charset=windows-1252
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 06/01/2016 06:45 AM, Peter Zijlstra wrote:
> Do we really need more panic_on_* knobs? Can't we re-purpose
> panic_on_warn for this?

I think this case is very specific, specific enough to deserve its own
sysctl. But I see your point, and the possibilities I can see are:

1) convert the printk(KERN_ERR "BUG:...") to a WARN(1,"BUG:..."), but
this will hide all other information printed in the function, so -1 for
this option;

2) fire a WARN in the end of the function with a message like the
previous printk(); or

3) re-use the panic_on_warn variable to condition the panic,
or to force a WARN.

But, IMHO, I think these are not as polish as use a specific sysctl.

If you think that any of these options are better than what I propose,
I can cook a v2 with the best option :-).

-- Daniel