From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S267503AbUBSToq (ORCPT ); Thu, 19 Feb 2004 14:44:46 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S267509AbUBSTop (ORCPT ); Thu, 19 Feb 2004 14:44:45 -0500 Received: from dbl.q-ag.de ([213.172.117.3]:39913 "EHLO dbl.q-ag.de") by vger.kernel.org with ESMTP id S267503AbUBSTon (ORCPT ); Thu, 19 Feb 2004 14:44:43 -0500 Message-ID: <40351211.1030200@colorfullife.com> Date: Thu, 19 Feb 2004 20:44:17 +0100 From: Manfred Spraul User-Agent: Mozilla/5.0 (X11; U; Linux i686; fr-FR; rv:1.4.1) Gecko/20031030 X-Accept-Language: en-us, en MIME-Version: 1.0 To: Sam Ravnborg CC: Christoph Hellwig , Krzysztof Benedyczak , linux-kernel@vger.kernel.org, Michal Wronski Subject: Re: [RFC][PATCH] 2/6 POSIX message queues References: <20040219145331.B23685@infradead.org> <20040219190720.GA2421@mars.ravnborg.org> In-Reply-To: <20040219190720.GA2421@mars.ravnborg.org> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Sam Ravnborg wrote: >Maybe something like: >mqueue.h for kernel-only >mqueue_abi.h for kernel+user > > I don't think that this is necessary. Everything in is kernel only. user space should copy the headers and remove the kernel only parts. kernel+user files mean that it's not possible rename structures or move them around. Perhaps someone wants to move all 16-bit uid structures into a directly - shared headers make that impossible. I agree that the placement of the #ifdef is a bit arbitrary - it's a hint that the structures outside are visible to user space and must remain unchanged. Christoph: I'm sure there will be users that must call the message queue functions directly from C code. E.g. the 32-bit emulation layers on 64/32 bit archs. And they need the prototypes. -- Manfred