From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1759481Ab1KWAmO (ORCPT ); Tue, 22 Nov 2011 19:42:14 -0500 Received: from cantor2.suse.de ([195.135.220.15]:43021 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1759469Ab1KWAmJ (ORCPT ); Tue, 22 Nov 2011 19:42:09 -0500 X-Mailbox-Line: From gregkh@clark.kroah.org Tue Nov 22 16:24:08 2011 Message-Id: <20111123002408.725504127@clark.kroah.org> User-Agent: quilt/0.48-20.1.2 Date: Tue, 22 Nov 2011 16:23:51 -0800 From: Greg KH To: linux-kernel@vger.kernel.org, stable@vger.kernel.org Cc: torvalds@linux-foundation.org, akpm@linux-foundation.org, alan@lxorguk.ukuu.org.uk, xfs@oss.sgi.com, Alex Elder , Christoph Hellwig Subject: [44/53] xfs: fix xfs_mark_inode_dirty during umount In-Reply-To: <20111123002419.GA8531@kroah.com> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org 3.0-stable review patch. If anyone has any objections, please let me know. ------------------ From: Christoph Hellwig commit 866e4ed77448a0c311e1b055eb72ea05423fd799 upstream. During umount we do not add a dirty inode to the lru and wait for it to become clean first, but force writeback of data and metadata with I_WILL_FREE set. Currently there is no way for XFS to detect that the inode has been redirtied for metadata operations, as we skip the mark_inode_dirty call during teardown. Fix this by setting i_update_core nanually in that case, so that the inode gets flushed during inode reclaim. Alternatively we could enable calling mark_inode_dirty for inodes in I_WILL_FREE state, and let the VFS dirty tracking handle this. I decided against this as we will get better I/O patterns from reclaim compared to the synchronous writeout in write_inode_now, and always marking the inode dirty in some way from xfs_mark_inode_dirty is a better safetly net in either case. Signed-off-by: Christoph Hellwig Reviewed-by: Dave Chinner Signed-off-by: Alex Elder Signed-off-by: Greg Kroah-Hartman --- fs/xfs/linux-2.6/xfs_iops.c | 14 +++++++++++--- 1 file changed, 11 insertions(+), 3 deletions(-) --- a/fs/xfs/linux-2.6/xfs_iops.c +++ b/fs/xfs/linux-2.6/xfs_iops.c @@ -69,9 +69,8 @@ xfs_synchronize_times( } /* - * If the linux inode is valid, mark it dirty. - * Used when committing a dirty inode into a transaction so that - * the inode will get written back by the linux code + * If the linux inode is valid, mark it dirty, else mark the dirty state + * in the XFS inode to make sure we pick it up when reclaiming the inode. */ void xfs_mark_inode_dirty_sync( @@ -81,6 +80,10 @@ xfs_mark_inode_dirty_sync( if (!(inode->i_state & (I_WILL_FREE|I_FREEING))) mark_inode_dirty_sync(inode); + else { + barrier(); + ip->i_update_core = 1; + } } void @@ -91,6 +94,11 @@ xfs_mark_inode_dirty( if (!(inode->i_state & (I_WILL_FREE|I_FREEING))) mark_inode_dirty(inode); + else { + barrier(); + ip->i_update_core = 1; + } + } /*