On Thu, Dec 22, 2016 at 5:16 PM, Jan Kara <jack@xxxxxxx> wrote: > On Thu 22-12-16 14:59:35, Amir Goldstein wrote: >> On Thu, Dec 22, 2016 at 11:15 AM, Jan Kara <jack@xxxxxxx> wrote: >> > show_fdinfo() iterates group's list of marks. All marks found there are >> > guaranteed to be alive and they stay so until we release >> > group->mark_mutex. So remove uncecessary tests whether mark is alive. >> > >> >> The statement above is true for fanotify. I don't think it holds for inotify. >> >> SYS_inotify_rm_watch() >> fsnotify_destroy_mark() >> fsnotify_destroy_mark(mark, group) >> mutex_lock_nested(&group->mark_mutex) /* not really nested for >> inotify */ >> fsnotify_detach_mark(mark) >> mutex_unlock(&group->mark_mutex); >> fsnotify_free_mark(mark) >> mark->flags &= ~FSNOTIFY_MARK_FLAG_ALIVE; >> /* !!! mark is not alive and on the group's list. group->mark_mutex is >> not held !!! */ > > How come it is on the group's list? fsnotify_detach_mark() will remove it > from that list... The destroy_list is just a private list used for mark > destruction, not a list of any group. > I stand corrected. This is bound to happen a few more times ;-) Reviewed-by: Amir Goldstein <amir73il@xxxxxxxxx> -- To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
