From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from moutng.kundenserver.de ([212.227.126.177]:57689 "EHLO moutng.kundenserver.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751226AbZFEMbw (ORCPT ); Fri, 5 Jun 2009 08:31:52 -0400 From: Arnd Bergmann Subject: Re: [Patch] kbuild: fix headers_check.pl Date: Fri, 5 Jun 2009 13:31:14 +0100 References: <20090605021419.7905.89870.sendpatchset@localhost.localdomain> <200906051021.51726.arnd@arndb.de> <4A28F1CD.6040804@redhat.com> In-Reply-To: <4A28F1CD.6040804@redhat.com> MIME-Version: 1.0 Content-Type: Text/Plain; charset="iso-8859-15" Content-Transfer-Encoding: 7bit Message-Id: <200906051331.14978.arnd@arndb.de> Sender: linux-kbuild-owner@vger.kernel.org List-ID: To: Amerigo Wang Cc: linux-kernel@vger.kernel.org, akpm@linux-foundation.org, sam@ravnborg.org, linux-kbuild@vger.kernel.org On Friday 05 June 2009, Amerigo Wang wrote: > > > > > I don't think we really need that distinction here, the old > > text applies to both. > > Even for function declarations? Yes. Any use of 'extern' in a kernel header file by definition refers to a symbol that is defined in the kernel and therefore not accessibly in user space. It is the same problem for variables and functions, with the complication that leaving out the 'extern' statement on a function declaration will hide it from this check, while leaving it out on a variable declaration turns it into a definition. Arnd <><