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 S1751528AbdECFoJ (ORCPT <rfc822;w@1wt.eu>);
        Wed, 3 May 2017 01:44:09 -0400
Received: from s3.sipsolutions.net ([5.9.151.49]:60860 "EHLO sipsolutions.net"
        rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP
        id S1750966AbdECFoB (ORCPT <rfc822;linux-kernel@vger.kernel.org>);
        Wed, 3 May 2017 01:44:01 -0400
Message-ID: <1493790238.2602.3.camel@sipsolutions.net>
Subject: Re: [RFC PATCH 8/9] debugfs: defer debugfs_fsdata allocation to
 first usage
From: Johannes Berg <johannes@sipsolutions.net>
To: Nicolai Stange <nicstange@gmail.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
        "Paul E.McKenney" <paulmck@linux.vnet.ibm.com>,
        linux-kernel@vger.kernel.org
Date: Wed, 03 May 2017 07:43:58 +0200
In-Reply-To: <87y3ufvxaz.fsf@gmail.com> (sfid-20170502_220600_227096_F45CC85C)
References: <871stdyg0u.fsf@gmail.com>
         <20170416095137.2784-1-nicstange@gmail.com>
         <20170416095137.2784-9-nicstange@gmail.com>
         <1492508204.2472.7.camel@sipsolutions.net> <87y3ufvxaz.fsf@gmail.com>
         (sfid-20170502_220600_227096_F45CC85C)
Content-Type: text/plain; charset="UTF-8"
X-Mailer: Evolution 3.22.6-1 
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 Tue, 2017-05-02 at 22:05 +0200, Nicolai Stange wrote:

> > So either you could return some valid ops (perhaps
> > debugfs_noop_file_operations although those don't have .name or
> > .poll, so it doesn't cover everything), or you can just BUG_ON()
> > here directly, saving the incomprehensible crash later.
> 
> The purpose of that WARN_ON() there was to turn a potentially
> incomprehensible crash into a comprehensible one ;)
> 
> In order to avoid a new BUG_ON(), what about keeping the WARN_ON() as
> is and returning NULL instead of the garbage? That would crash
> current on first access and the earlier warning would hopefully give
> some clue?

Yeah, I guess that might work. Probably less harmful to the system than
a BUG_ON(), but I still operate under the assumption that there might
actually be something mapped at NULL - not sure if that's still true.

johannes