The patch titled Subject: fanotify: fix double free of pending permission events has been removed from the -mm tree. Its filename was fanotify-fix-double-free-of-pending-permission-events.patch This patch was dropped because it was merged into mainline or a subsystem tree ------------------------------------------------------ From: Jan Kara <jack@xxxxxxx> Subject: fanotify: fix double free of pending permission events Commit 85816794240b ("fanotify: Fix use after free for permission events") introduced a double free issue for permission events which are pending in group's notification queue while group is being destroyed. These events are freed from fanotify_handle_event() but they are not removed from groups notification queue and thus they get freed again from fsnotify_flush_notify(). Fix the problem by removing permission events from notification queue before freeing them if we skip processing access response. Also expand comments in fanotify_release() to explain group shutdown in detail. Fixes: 85816794240b9659e66e4d9b0df7c6e814e5f603 Signed-off-by: Jan Kara <jack@xxxxxxx> Reported-by: Douglas Leeder <douglas.leeder@xxxxxxxxxx> Tested-by: Douglas Leeder <douglas.leeder@xxxxxxxxxx> Reported-by: Heinrich Schuchard <xypron.glpk@xxxxxx> Cc: <stable@xxxxxxxxxxxxxxx> Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> --- fs/notify/fanotify/fanotify.c | 9 ++++++++- fs/notify/fanotify/fanotify_user.c | 12 ++++++++++++ fs/notify/notification.c | 18 +++++++++++++++++- include/linux/fsnotify_backend.h | 2 ++ 4 files changed, 39 insertions(+), 2 deletions(-) diff -puN fs/notify/fanotify/fanotify.c~fanotify-fix-double-free-of-pending-permission-events fs/notify/fanotify/fanotify.c --- a/fs/notify/fanotify/fanotify.c~fanotify-fix-double-free-of-pending-permission-events +++ a/fs/notify/fanotify/fanotify.c @@ -70,8 +70,15 @@ static int fanotify_get_response(struct wait_event(group->fanotify_data.access_waitq, event->response || atomic_read(&group->fanotify_data.bypass_perm)); - if (!event->response) /* bypass_perm set */ + if (!event->response) { /* bypass_perm set */ + /* + * Event was canceled because group is being destroyed. Remove + * it from group's event list because we are responsible for + * freeing the permission event. + */ + fsnotify_remove_event(group, &event->fae.fse); return 0; + } /* userspace responded, convert to something usable */ switch (event->response) { diff -puN fs/notify/fanotify/fanotify_user.c~fanotify-fix-double-free-of-pending-permission-events fs/notify/fanotify/fanotify_user.c --- a/fs/notify/fanotify/fanotify_user.c~fanotify-fix-double-free-of-pending-permission-events +++ a/fs/notify/fanotify/fanotify_user.c @@ -359,6 +359,11 @@ static int fanotify_release(struct inode #ifdef CONFIG_FANOTIFY_ACCESS_PERMISSIONS struct fanotify_perm_event_info *event, *next; + /* + * There may be still new events arriving in the notification queue + * but since userspace cannot use fanotify fd anymore, no event can + * enter or leave access_list by now. + */ spin_lock(&group->fanotify_data.access_lock); atomic_inc(&group->fanotify_data.bypass_perm); @@ -373,6 +378,13 @@ static int fanotify_release(struct inode } spin_unlock(&group->fanotify_data.access_lock); + /* + * Since bypass_perm is set, newly queued events will not wait for + * access response. Wake up the already sleeping ones now. + * synchronize_srcu() in fsnotify_destroy_group() will wait for all + * processes sleeping in fanotify_handle_event() waiting for access + * response and thus also for all permission events to be freed. + */ wake_up(&group->fanotify_data.access_waitq); #endif diff -puN fs/notify/notification.c~fanotify-fix-double-free-of-pending-permission-events fs/notify/notification.c --- a/fs/notify/notification.c~fanotify-fix-double-free-of-pending-permission-events +++ a/fs/notify/notification.c @@ -73,7 +73,8 @@ void fsnotify_destroy_event(struct fsnot /* Overflow events are per-group and we don't want to free them */ if (!event || event->mask == FS_Q_OVERFLOW) return; - + /* If the event is still queued, we have a problem... */ + WARN_ON(!list_empty(&event->list)); group->ops->free_event(event); } @@ -125,6 +126,21 @@ queue: } /* + * Remove @event from group's notification queue. It is the responsibility of + * the caller to destroy the event. + */ +void fsnotify_remove_event(struct fsnotify_group *group, + struct fsnotify_event *event) +{ + mutex_lock(&group->notification_mutex); + if (!list_empty(&event->list)) { + list_del_init(&event->list); + group->q_len--; + } + mutex_unlock(&group->notification_mutex); +} + +/* * Remove and return the first event from the notification list. It is the * responsibility of the caller to destroy the obtained event */ diff -puN include/linux/fsnotify_backend.h~fanotify-fix-double-free-of-pending-permission-events include/linux/fsnotify_backend.h --- a/include/linux/fsnotify_backend.h~fanotify-fix-double-free-of-pending-permission-events +++ a/include/linux/fsnotify_backend.h @@ -326,6 +326,8 @@ extern int fsnotify_add_event(struct fsn struct fsnotify_event *event, int (*merge)(struct list_head *, struct fsnotify_event *)); +/* Remove passed event from groups notification queue */ +extern void fsnotify_remove_event(struct fsnotify_group *group, struct fsnotify_event *event); /* true if the group notification queue is empty */ extern bool fsnotify_notify_queue_is_empty(struct fsnotify_group *group); /* return, but do not dequeue the first event on the notification queue */ _ Patches currently in -mm which might be from jack@xxxxxxx are origin.patch fs-ext4-fsyncc-generic_file_fsync-call-based-on-barrier-flag.patch fs-mpagec-forgotten-write_sync-in-case-of-data-integrity-write.patch fs-isofs-logging-clean-up.patch linux-next.patch mm-add-strictlimit-knob-v2.patch -- To unsubscribe from this list: send the line "unsubscribe mm-commits" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html