From mboxrd@z Thu Jan 1 00:00:00 1970 From: Theodore Tso Subject: Re: RFC, 32-bit compat handlers for EXT4_IOC_GROUP_ADD Date: Sat, 6 Dec 2008 15:24:09 -0500 Message-ID: <20081206202409.GI1323@mit.edu> References: <49385BC5.2070703@redhat.com> <20081206010204.GY3186@webber.adilger.int> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Eric Sandeen , ext4 development To: Andreas Dilger Return-path: Received: from www.church-of-our-saviour.org ([69.25.196.31]:41341 "EHLO thunker.thunk.org" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1752325AbYLFUYN (ORCPT ); Sat, 6 Dec 2008 15:24:13 -0500 Content-Disposition: inline In-Reply-To: <20081206010204.GY3186@webber.adilger.int> Sender: linux-ext4-owner@vger.kernel.org List-ID: On Fri, Dec 05, 2008 at 06:02:04PM -0700, Andreas Dilger wrote: > > How far out of "dev" are we? I'm leaning towards saying "oh well, would > > have been nicer the other way" but going ahead and just putting the > > compat handler into the kernel. > > I would be OK with changing to the "proper" struct layout. Not being able > to resize with an older e2fsprogs + newer kernel isn't going to cause any > serious problems (unlike e.g. not being able to mount or e2fsck "/"). > > If we are seriously worried about compatibility, we could add the compat > handler for 32-bit kernels (should have a different IOC number anyways > because of the struct size) and add some arbitrary check like: > > #ifdef LINUX_KERNEL_VERSION > KERNEL_VERSION(2,6,40) > #warning remove this old compat code > #endif Given that a bunch of distro's have shipped e2fsprogs 1.41.x which we advertised as being ext4 compatibility, I think we need to keep the compatibility code. If we want to add the complexity for the 32-bit side, with a 2-3 year timeout, that seems like a reasonable compromise. - Ted