From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1759547Ab0KQC4b (ORCPT ); Tue, 16 Nov 2010 21:56:31 -0500 Received: from smtp1.linux-foundation.org ([140.211.169.13]:33055 "EHLO smtp1.linux-foundation.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1759431Ab0KQC4a (ORCPT ); Tue, 16 Nov 2010 21:56:30 -0500 MIME-Version: 1.0 In-Reply-To: <20101116170005.2646f4fb.randy.dunlap@oracle.com> References: <20101116170005.2646f4fb.randy.dunlap@oracle.com> From: Linus Torvalds Date: Tue, 16 Nov 2010 18:56:08 -0800 Message-ID: Subject: Re: [PATCH] cciss: fix build for PROC_FS disabled To: Randy Dunlap , Jens Axboe Cc: Linux Kernel Mailing List Content-Type: text/plain; charset=ISO-8859-1 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Tue, Nov 16, 2010 at 5:00 PM, Randy Dunlap wrote: > > Fix cciss driver for CONFIG_PROC_FS not enabled: > > drivers/block/cciss.c:4929: error: 'proc_cciss' undeclared (first use in this function) Hmm. I would almost prefer to not have this #ifdef inside code, and solve it by simply moving the definition of "proc_cciss" to outside the other #ifdef CONFIG_PROC_FS. That way it allways stays NULL, and without CONFIG_PROC_FS the code in question will all be empty inline function calls anyway, so the compiler will compile it all away (at the cost of just keeping that one unnecessary variable around). I'd rather waste one useless pointer than have ugly source code. Comments? Linus