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 S1752148AbbBKKXK (ORCPT <rfc822;w@1wt.eu>);
	Wed, 11 Feb 2015 05:23:10 -0500
Received: from userp1040.oracle.com ([156.151.31.81]:32720 "EHLO
	userp1040.oracle.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
	with ESMTP id S1751313AbbBKKXI (ORCPT
	<rfc822;linux-kernel@vger.kernel.org>);
	Wed, 11 Feb 2015 05:23:08 -0500
Date: Wed, 11 Feb 2015 13:23:07 +0300
From: Dan Carpenter <dan.carpenter@oracle.com>
To: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Quentin Lambert <lambert.quentin@gmail.com>, devel@driverdev.osuosl.org,
        sparmaintainer@unisys.com, kernel-janitors@vger.kernel.org,
        linux-kernel@vger.kernel.org,
        Benjamin Romer <benjamin.romer@unisys.com>,
        David Kershner <david.kershner@unisys.com>
Subject: Re: [PATCH 1/1] staging: unisys: Remove allocation from declaration
 line
Message-ID: <20150211102306.GB26542@mwanda>
References: <20150210130214.GA18429@sloth>
 <20150210222627.GC27072@kroah.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20150210222627.GC27072@kroah.com>
User-Agent: Mutt/1.5.21 (2010-09-15)
X-Source-IP: ucsinet21.oracle.com [156.151.31.93]
Sender: linux-kernel-owner@vger.kernel.org
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org

On Wed, Feb 11, 2015 at 06:26:27AM +0800, Greg Kroah-Hartman wrote:
> On Tue, Feb 10, 2015 at 02:02:14PM +0100, Quentin Lambert wrote:
> > This patch removes allocation from declaration line because
> > people are known to gloss over declarations.
> 
> Again, who are these lazy people, and why are they reading kernel code?
> 

>>From my work with smatch:
1) Probably 70-80% of inconsistent NULL checking is when done in the
   initializer.  I'm sending a patch for one of these today.
2) If there is an allocation in the initializer then it's more likely
   that the NULL check will be missing.
Initializers are a blind spot that people do not read.  It's not just
one maintainer, it's consistent across the board.

Also if you put an allocation in the initializer then it almost always
has to be mangled to fit in 80 characters and it looks ugly.  But after
these patches then all the allocations fit naturally.

Plus you have to have that blank line to separate the initialization
paragraph from the paragraph with the check for allocation failure.

Really, it is fairly uncommon to put an allocation in the initalizer.

regards,
dan carpenter