From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932451AbWG0GIz (ORCPT ); Thu, 27 Jul 2006 02:08:55 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S932526AbWG0GIy (ORCPT ); Thu, 27 Jul 2006 02:08:54 -0400 Received: from mtagate6.uk.ibm.com ([195.212.29.139]:33041 "EHLO mtagate6.uk.ibm.com") by vger.kernel.org with ESMTP id S932518AbWG0GIx (ORCPT ); Thu, 27 Jul 2006 02:08:53 -0400 Date: Thu, 27 Jul 2006 08:06:26 +0200 From: Heiko Carstens To: Arnd Bergmann Cc: linux-kernel@vger.kernel.org, Martin Schwidefsky Subject: Re: [patch/rfc] s390: get rid of own uid16 compat system calls Message-ID: <20060727060626.GA9594@osiris.boeblingen.de.ibm.com> References: <20060710085142.GB9440@osiris.boeblingen.de.ibm.com> <200607270303.42959.arnd.bergmann@de.ibm.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <200607270303.42959.arnd.bergmann@de.ibm.com> User-Agent: mutt-ng/devel-r804 (Linux) Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Thu, Jul 27, 2006 at 03:03:42AM +0200, Arnd Bergmann wrote: > On Monday 10 July 2006 10:51, Heiko Carstens wrote: > > "Only" thing is that we unfortunately have different sizes for > > __kernel_old_[uid|gid]_t (16 bit on s390, 32 on s390x). I was tempted to > > change these just to find out that there are other users as well: > > > > include/linux/ncp_fs.h: > > ?#define NCP_IOC_GETMOUNTUID _IOW('n', 2, __kernel_old_uid_t) > > include/linux/smb_fs.h: > > ?#define SMB_IOC_GETMOUNTUID _IOR('u', 1, __kernel_old_uid_t) > > > > So, this is no option. Would anybody know of something to get this work? > > Or is this just a stupid idea? > > Ok, I don't know exactly what you're talking about, but I have in the > past tried to hack that area as well. It's probably a good idea to > pick up my old patch and work from there, by making these two file systems > understand all possible ways: I dropped my initial patch and gave up on this. The idea of using kernel/uid16.c for compat code seems to be... not very good :)