On Mon, Aug 06, 2018 at 12:37:38PM +1000, Dave Chinner wrote: > IOWs, I can't see any reliable access serialisation being provided > by holding the ILOCK while modifying the value. Perhaps this should > use WRITE_ONCE/READ_ONCE to ensure the correct memory barriers are > used to serialise tthe sequence count against itself rather than the > wider inode extent modification process? Yes, I think we need a barrier before the later actual modification of the extent tree and thus should use WRITE_ONCE/READ_ONCE. I've got a patch that is undergoing testing at the moment. -- To unsubscribe from this list: send the line "unsubscribe linux-xfs" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
