On Sun, Apr 25, 2010 at 08:32:40PM -0400, Theodore Ts'o wrote: > Any thoughts or suggestions would be greatly appreciated. I've looked > at the xfs and btrfs code for some ideas, but dealing with current > writeback and truncate is nasty, especially if there's a subsequent > delalloc write happening in parallel with the writeback and immediately > after the truncate. After studying the code quite extensively over the > weekend, I'm still not entirely sure that XFS and btrfs gets this case > right (I know ext4 currently doesn't). Of course, it's not clear > whether users will trip against this in practice, but it's nevertheless > still a botch, and I'm wondering if it's simpler to avoid the concurrent > vmtruncate/writeback case entirely. What case are you concerned that is XFS not getting right? Cheers, Dave. -- Dave Chinner david@xxxxxxxxxxxxx -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@xxxxxxxxxx For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>