Merge branch 'work.epoll' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs

Pull epoll fixes from Al Viro:
 "Several race fixes in epoll"

* 'work.epoll' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
  ep_create_wakeup_source(): dentry name can change under you...
  epoll: EPOLL_CTL_ADD: close the race in decision to take fast path
  epoll: replace ->visited/visited_list with generation count
  epoll: do not insert into poll queues until all sanity checks are done

1 files changed, 31 insertions, 41 deletions

diff --git a/fs/eventpoll.c b/fs/eventpoll.c
index 8107e06d7f6f..4df61129566d 100644
--- a/fs/eventpoll.c
+++ b/fs/eventpoll.c
@@ -218,8 +218,7 @@ struct eventpoll {
 	struct file *file;
 
 	/* used to optimize loop detection check */
-	struct list_head visited_list_link;
-	int visited;
+	u64 gen;
 
 #ifdef CONFIG_NET_RX_BUSY_POLL
 	/* used to track busy poll napi_id */
@@ -274,6 +273,8 @@ static long max_user_watches __read_mostly;
  */
 static DEFINE_MUTEX(epmutex);
 
+static u64 loop_check_gen = 0;
+
 /* Used to check for epoll file descriptor inclusion loops */
 static struct nested_calls poll_loop_ncalls;
 
@@ -283,9 +284,6 @@ static struct kmem_cache *epi_cache __read_mostly;
 /* Slab cache used to allocate "struct eppoll_entry" */
 static struct kmem_cache *pwq_cache __read_mostly;
 
-/* Visited nodes during ep_loop_check(), so we can unset them when we finish */
-static LIST_HEAD(visited_list);
-
 /*
  * List of files with newly added links, where we may need to limit the number
  * of emanating paths. Protected by the epmutex.
@@ -1450,7 +1448,7 @@ static int reverse_path_check(void)
 
 static int ep_create_wakeup_source(struct epitem *epi)
 {
-	const char *name;
+	struct name_snapshot n;
 	struct wakeup_source *ws;
 
 	if (!epi->ep->ws) {
@@ -1459,8 +1457,9 @@ static int ep_create_wakeup_source(struct epitem *epi)
 			return -ENOMEM;
 	}
 
-	name = epi->ffd.file->f_path.dentry->d_name.name;
-	ws = wakeup_source_register(NULL, name);
+	take_dentry_name_snapshot(&n, epi->ffd.file->f_path.dentry);
+	ws = wakeup_source_register(NULL, n.name.name);
+	release_dentry_name_snapshot(&n);
 
 	if (!ws)
 		return -ENOMEM;
@@ -1522,6 +1521,22 @@ static int ep_insert(struct eventpoll *ep, const struct epoll_event *event,
 		RCU_INIT_POINTER(epi->ws, NULL);
 	}
 
+	/* Add the current item to the list of active epoll hook for this file */
+	spin_lock(&tfile->f_lock);
+	list_add_tail_rcu(&epi->fllink, &tfile->f_ep_links);
+	spin_unlock(&tfile->f_lock);
+
+	/*
+	 * Add the current item to the RB tree. All RB tree operations are
+	 * protected by "mtx", and ep_insert() is called with "mtx" held.
+	 */
+	ep_rbtree_insert(ep, epi);
+
+	/* now check if we've created too many backpaths */
+	error = -EINVAL;
+	if (full_check && reverse_path_check())
+		goto error_remove_epi;
+
 	/* Initialize the poll table using the queue callback */
 	epq.epi = epi;
 	init_poll_funcptr(&epq.pt, ep_ptable_queue_proc);
@@ -1544,22 +1559,6 @@ static int ep_insert(struct eventpoll *ep, const struct epoll_event *event,
 	if (epi->nwait < 0)
 		goto error_unregister;
 
-	/* Add the current item to the list of active epoll hook for this file */
-	spin_lock(&tfile->f_lock);
-	list_add_tail_rcu(&epi->fllink, &tfile->f_ep_links);
-	spin_unlock(&tfile->f_lock);
-
-	/*
-	 * Add the current item to the RB tree. All RB tree operations are
-	 * protected by "mtx", and ep_insert() is called with "mtx" held.
-	 */
-	ep_rbtree_insert(ep, epi);
-
-	/* now check if we've created too many backpaths */
-	error = -EINVAL;
-	if (full_check && reverse_path_check())
-		goto error_remove_epi;
-
 	/* We have to drop the new item inside our item list to keep track of it */
 	write_lock_irq(&ep->lock);
 
@@ -1588,6 +1587,8 @@ static int ep_insert(struct eventpoll *ep, const struct epoll_event *event,
 
 	return 0;
 
+error_unregister:
+	ep_unregister_pollwait(ep, epi);
 error_remove_epi:
 	spin_lock(&tfile->f_lock);
 	list_del_rcu(&epi->fllink);
@@ -1595,9 +1596,6 @@ error_remove_epi:
 
 	rb_erase_cached(&epi->rbn, &ep->rbr);
 
-error_unregister:
-	ep_unregister_pollwait(ep, epi);
-
 	/*
 	 * We need to do this because an event could have been arrived on some
 	 * allocated wait queue. Note that we don't care about the ep->ovflist
@@ -1972,13 +1970,12 @@ static int ep_loop_check_proc(void *priv, void *cookie, int call_nests)
 	struct epitem *epi;
 
 	mutex_lock_nested(&ep->mtx, call_nests + 1);
-	ep->visited = 1;
-	list_add(&ep->visited_list_link, &visited_list);
+	ep->gen = loop_check_gen;
 	for (rbp = rb_first_cached(&ep->rbr); rbp; rbp = rb_next(rbp)) {
 		epi = rb_entry(rbp, struct epitem, rbn);
 		if (unlikely(is_file_epoll(epi->ffd.file))) {
 			ep_tovisit = epi->ffd.file->private_data;
-			if (ep_tovisit->visited)
+			if (ep_tovisit->gen == loop_check_gen)
 				continue;
 			error = ep_call_nested(&poll_loop_ncalls,
 					ep_loop_check_proc, epi->ffd.file,
@@ -2019,18 +2016,8 @@ static int ep_loop_check_proc(void *priv, void *cookie, int call_nests)
  */
 static int ep_loop_check(struct eventpoll *ep, struct file *file)
 {
-	int ret;
-	struct eventpoll *ep_cur, *ep_next;
-
-	ret = ep_call_nested(&poll_loop_ncalls,
+	return ep_call_nested(&poll_loop_ncalls,
 			      ep_loop_check_proc, file, ep, current);
-	/* clear visited list */
-	list_for_each_entry_safe(ep_cur, ep_next, &visited_list,
-							visited_list_link) {
-		ep_cur->visited = 0;
-		list_del(&ep_cur->visited_list_link);
-	}
-	return ret;
 }
 
 static void clear_tfile_check_list(void)
@@ -2195,11 +2182,13 @@ int do_epoll_ctl(int epfd, int op, int fd, struct epoll_event *epds,
 		goto error_tgt_fput;
 	if (op == EPOLL_CTL_ADD) {
 		if (!list_empty(&f.file->f_ep_links) ||
+				ep->gen == loop_check_gen ||
 						is_file_epoll(tf.file)) {
 			mutex_unlock(&ep->mtx);
 			error = epoll_mutex_lock(&epmutex, 0, nonblock);
 			if (error)
 				goto error_tgt_fput;
+			loop_check_gen++;
 			full_check = 1;
 			if (is_file_epoll(tf.file)) {
 				error = -ELOOP;
@@ -2263,6 +2252,7 @@ int do_epoll_ctl(int epfd, int op, int fd, struct epoll_event *epds,
 error_tgt_fput:
 	if (full_check) {
 		clear_tfile_check_list();
+		loop_check_gen++;
 		mutex_unlock(&epmutex);
 	}