From mboxrd@z Thu Jan 1 00:00:00 1970 From: tytso@mit.edu Subject: Re: [PATCH] Rearrange i_flags to be consistent with FS_IOC_GETFLAGS Date: Mon, 5 Jul 2010 13:27:18 -0400 Message-ID: <20100705172718.GE25518@thunk.org> References: <20100705154319.31193.56706.stgit@warthog.procyon.org.uk> <4C32001F.8060906@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: David Howells , viro@ZenIV.linux.org.uk, linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org, linux-ext4@vger.kernel.org To: Eric Sandeen Return-path: Content-Disposition: inline In-Reply-To: <4C32001F.8060906@redhat.com> Sender: linux-kernel-owner@vger.kernel.org List-Id: linux-fsdevel.vger.kernel.org On Mon, Jul 05, 2010 at 10:54:07AM -0500, Eric Sandeen wrote: > > ... a stern comment here about not re-ordering since other fs/*/* > code depends strongly on this order? A stern warning is needed there, but it's also needed a few lines further down in the inode flags section where the FS_*_FL flags are defined. These were originally ext2-specific inode flags, and it's become generalized to a fs-independent set of bit fields, but what's nasty/important to remember is that these flags are also used as on-disk flags for ext2/3/4, and so extreme care is needed before new flags are for FS_IOC_GETFLAGS/FS_IOC_SETFLAGS are allocated.... In fact, I'd argue it's much more strongly needed for the FS_*_FL flags. - Ted