From mboxrd@z Thu Jan 1 00:00:00 1970 From: Chris Ball Subject: Re: [PATCH 2/2] mmc: fix integer assignments to pointer Date: Tue, 23 Aug 2011 14:28:42 -0400 Message-ID: References: <1314114363-6767-1-git-send-email-svenkatr@ti.com> <1314114363-6767-3-git-send-email-svenkatr@ti.com> <20110823172851.GB1784@leaf> <874o18m2cn.fsf@blp.benpfaff.org> <20110823181359.GB3418@leaf> Mime-Version: 1.0 Content-Type: text/plain Return-path: Received: from void.printf.net ([89.145.121.20]:55161 "EHLO void.printf.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754685Ab1HWS25 (ORCPT ); Tue, 23 Aug 2011 14:28:57 -0400 In-Reply-To: <20110823181359.GB3418@leaf> (Josh Triplett's message of "Tue, 23 Aug 2011 11:13:59 -0700") Sender: linux-mmc-owner@vger.kernel.org List-Id: linux-mmc@vger.kernel.org To: Josh Triplett Cc: blp@cs.stanford.edu, Venkatraman S , linux-mmc@vger.kernel.org, linux-sparse@vger.kernel.org Hi, On Tue, Aug 23 2011, Josh Triplett wrote: > {} produces the same effect, as far as I know. Yeah. I prefer {0}, because {} is a gcc-ism (the ANSI grammar demands initializer-lists be non-empty) and is less readable for people who haven't seen the idiom before and are wondering what's going on. I'm still a little confused -- the {0} or memset(0, struct ..); formations are used often in the kernel, even with pointers involved. Is the warning (Wnon_pointer_null) run against the kernel by default, or did Venkatraman add it manually? If default, is it catching bugs? Thanks, - Chris. -- Chris Ball One Laptop Per Child