On Tue, Nov 01, 2011 at 04:02:24PM -0700, Andy Lutomirski wrote: > Hmm. Isn't it permitted to at least read from an fs while holding the > page lock? I thought that the page lock was held for the entire > duration of a read and at the beginning of writeback. > > I can push this down to the ->writepage implementations or to the > clear_page_dirty_for_io callers, but that will result in a bigger > patch. Besides the current way that seems to be the only reasonable place to do it. Pushing it into ->writepage also has the benefit that filesystems could piggy back the ctime update onto the transaction that updates the extent tree. -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@xxxxxxxxx. For more info on Linux MM, see: http://www.linux-mm.org/ . Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/ Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>