From mboxrd@z Thu Jan 1 00:00:00 1970 From: Al Viro Subject: Re: [PATCH 43/58] switch ocfs2 to ->evict_inode() Date: Wed, 9 Jun 2010 02:49:19 +0100 Message-ID: <20100609014919.GV31073@ZenIV.linux.org.uk> References: <20100609001710.GA6795@mail.oracle.com> <20100609011959.GR31073@ZenIV.linux.org.uk> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii To: linux-fsdevel@vger.kernel.org Return-path: Received: from zeniv.linux.org.uk ([195.92.253.2]:56753 "EHLO ZenIV.linux.org.uk" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750885Ab0FIBtU (ORCPT ); Tue, 8 Jun 2010 21:49:20 -0400 Received: from viro by ZenIV.linux.org.uk with local (Exim 4.69 #1 (Red Hat Linux)) id 1OMAPv-0007J4-Jg for linux-fsdevel@vger.kernel.org; Wed, 09 Jun 2010 01:49:19 +0000 Content-Disposition: inline In-Reply-To: <20100609011959.GR31073@ZenIV.linux.org.uk> Sender: linux-fsdevel-owner@vger.kernel.org List-ID: On Wed, Jun 09, 2010 at 02:19:59AM +0100, Al Viro wrote: > On Tue, Jun 08, 2010 at 05:17:11PM -0700, Joel Becker wrote: > > I do think it works, though, if you check for MAYBE_ORPHANED. > > > > if (inode->i_nlink && !(oi->ip_falgs & OCFS2_INODE_MAYBE_ORPHANED)) { > > ocfs2_cleanup_delete_inode(inode, 0); > > goto bail; > > } > > Damn, that probably means we have other problems like that. Other filesystems > that have extra "delete" cases (and do not play with i_nlink like gfs2 does) > and have non-trivial ->delete_inode(). > > Ho-hum... OK, not too many. btrfs seems to be the only one besides ocfs2. > Will fix... > > I'll post replacement patches in a few. Posted and pushed to vfs-2.6.git...