From mboxrd@z Thu Jan 1 00:00:00 1970 From: Christoph Hellwig Subject: Re: [PATCH] autofs4 - fix comms packet struct size Date: Mon, 20 Feb 2006 16:00:54 +0000 Message-ID: <20060220160054.GA31797@infradead.org> References: <20060219141517.GA7942@infradead.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Andrew Morton , Kernel Mailing List , linux-fsdevel , autofs mailing list Return-path: Received: from pentafluge.infradead.org ([213.146.154.40]:57835 "EHLO pentafluge.infradead.org") by vger.kernel.org with ESMTP id S964940AbWBTQBJ (ORCPT ); Mon, 20 Feb 2006 11:01:09 -0500 To: Ian Kent Content-Disposition: inline In-Reply-To: Sender: linux-fsdevel-owner@vger.kernel.org List-Id: linux-fsdevel.vger.kernel.org On Mon, Feb 20, 2006 at 09:05:59AM +0800, Ian Kent wrote: > > > autofs_wqt_t wait_queue_token; > > > > Hiding types in user visible structures behind typedefs is bad. > > What type is behind this? If this is not an __u32 you have > > a padding issue. > > This has been an occassion problem for a long time. > Since it dates back to way before version 4 I have always been reluctant > to change it. I'd rather leave it as is unless you really can't accept it. So is this an __u32? ;-) > OK. But will they be 32 bit for the life of this struct? If we'd ever bump pid_t or uid_t to 64bits tons of kernel interface would need to change. And no one guarantees it'd be come an u64 then. So __u32 is the safest choice.