1 files changed, 125 insertions, 75 deletions

diff --git a/net/unix/garbage.c b/net/unix/garbage.c
index 2a81880dac7b..fa39b6265238 100644
--- a/net/unix/garbage.c
+++ b/net/unix/garbage.c
@@ -81,12 +81,80 @@
 #include <net/scm.h>
 #include <net/tcp_states.h>
 
-#include "scm.h"
+struct unix_sock *unix_get_socket(struct file *filp)
+{
+	struct inode *inode = file_inode(filp);
+
+	/* Socket ? */
+	if (S_ISSOCK(inode->i_mode) && !(filp->f_mode & FMODE_PATH)) {
+		struct socket *sock = SOCKET_I(inode);
+		const struct proto_ops *ops;
+		struct sock *sk = sock->sk;
+
+		ops = READ_ONCE(sock->ops);
 
-/* Internal data structures and random procedures: */
+		/* PF_UNIX ? */
+		if (sk && ops && ops->family == PF_UNIX)
+			return unix_sk(sk);
+	}
+
+	return NULL;
+}
 
+DEFINE_SPINLOCK(unix_gc_lock);
+unsigned int unix_tot_inflight;
 static LIST_HEAD(gc_candidates);
-static DECLARE_WAIT_QUEUE_HEAD(unix_gc_wait);
+static LIST_HEAD(gc_inflight_list);
+
+/* Keep the number of times in flight count for the file
+ * descriptor if it is for an AF_UNIX socket.
+ */
+void unix_inflight(struct user_struct *user, struct file *filp)
+{
+	struct unix_sock *u = unix_get_socket(filp);
+
+	spin_lock(&unix_gc_lock);
+
+	if (u) {
+		if (!u->inflight) {
+			WARN_ON_ONCE(!list_empty(&u->link));
+			list_add_tail(&u->link, &gc_inflight_list);
+		} else {
+			WARN_ON_ONCE(list_empty(&u->link));
+		}
+		u->inflight++;
+
+		/* Paired with READ_ONCE() in wait_for_unix_gc() */
+		WRITE_ONCE(unix_tot_inflight, unix_tot_inflight + 1);
+	}
+
+	WRITE_ONCE(user->unix_inflight, user->unix_inflight + 1);
+
+	spin_unlock(&unix_gc_lock);
+}
+
+void unix_notinflight(struct user_struct *user, struct file *filp)
+{
+	struct unix_sock *u = unix_get_socket(filp);
+
+	spin_lock(&unix_gc_lock);
+
+	if (u) {
+		WARN_ON_ONCE(!u->inflight);
+		WARN_ON_ONCE(list_empty(&u->link));
+
+		u->inflight--;
+		if (!u->inflight)
+			list_del_init(&u->link);
+
+		/* Paired with READ_ONCE() in wait_for_unix_gc() */
+		WRITE_ONCE(unix_tot_inflight, unix_tot_inflight - 1);
+	}
+
+	WRITE_ONCE(user->unix_inflight, user->unix_inflight - 1);
+
+	spin_unlock(&unix_gc_lock);
+}
 
 static void scan_inflight(struct sock *x, void (*func)(struct unix_sock *),
 			  struct sk_buff_head *hitlist)
@@ -105,20 +173,15 @@ static void scan_inflight(struct sock *x, void (*func)(struct unix_sock *),
 
 			while (nfd--) {
 				/* Get the socket the fd matches if it indeed does so */
-				struct sock *sk = unix_get_socket(*fp++);
-
-				if (sk) {
-					struct unix_sock *u = unix_sk(sk);
+				struct unix_sock *u = unix_get_socket(*fp++);
 
-					/* Ignore non-candidates, they could
-					 * have been added to the queues after
-					 * starting the garbage collection
-					 */
-					if (test_bit(UNIX_GC_CANDIDATE, &u->gc_flags)) {
-						hit = true;
+				/* Ignore non-candidates, they could have been added
+				 * to the queues after starting the garbage collection
+				 */
+				if (u && test_bit(UNIX_GC_CANDIDATE, &u->gc_flags)) {
+					hit = true;
 
-						func(u);
-					}
+					func(u);
 				}
 			}
 			if (hit && hitlist != NULL) {
@@ -151,7 +214,7 @@ static void scan_children(struct sock *x, void (*func)(struct unix_sock *),
 			/* An embryo cannot be in-flight, so it's safe
 			 * to use the list link.
 			 */
-			BUG_ON(!list_empty(&u->link));
+			WARN_ON_ONCE(!list_empty(&u->link));
 			list_add_tail(&u->link, &embryos);
 		}
 		spin_unlock(&x->sk_receive_queue.lock);
@@ -166,17 +229,18 @@ static void scan_children(struct sock *x, void (*func)(struct unix_sock *),
 
 static void dec_inflight(struct unix_sock *usk)
 {
-	atomic_long_dec(&usk->inflight);
+	usk->inflight--;
 }
 
 static void inc_inflight(struct unix_sock *usk)
 {
-	atomic_long_inc(&usk->inflight);
+	usk->inflight++;
 }
 
 static void inc_inflight_move_tail(struct unix_sock *u)
 {
-	atomic_long_inc(&u->inflight);
+	u->inflight++;
+
 	/* If this still might be part of a cycle, move it to the end
 	 * of the list, so that it's checked even if it was already
 	 * passed over
@@ -186,40 +250,16 @@ static void inc_inflight_move_tail(struct unix_sock *u)
 }
 
 static bool gc_in_progress;
-#define UNIX_INFLIGHT_TRIGGER_GC 16000
-
-void wait_for_unix_gc(void)
-{
-	/* If number of inflight sockets is insane,
-	 * force a garbage collect right now.
-	 * Paired with the WRITE_ONCE() in unix_inflight(),
-	 * unix_notinflight() and gc_in_progress().
-	 */
-	if (READ_ONCE(unix_tot_inflight) > UNIX_INFLIGHT_TRIGGER_GC &&
-	    !READ_ONCE(gc_in_progress))
-		unix_gc();
-	wait_event(unix_gc_wait, gc_in_progress == false);
-}
 
-/* The external entry point: unix_gc() */
-void unix_gc(void)
+static void __unix_gc(struct work_struct *work)
 {
-	struct sk_buff *next_skb, *skb;
-	struct unix_sock *u;
-	struct unix_sock *next;
 	struct sk_buff_head hitlist;
-	struct list_head cursor;
+	struct unix_sock *u, *next;
 	LIST_HEAD(not_cycle_list);
+	struct list_head cursor;
 
 	spin_lock(&unix_gc_lock);
 
-	/* Avoid a recursive GC. */
-	if (gc_in_progress)
-		goto out;
-
-	/* Paired with READ_ONCE() in wait_for_unix_gc(). */
-	WRITE_ONCE(gc_in_progress, true);
-
 	/* First, select candidates for garbage collection.  Only
 	 * in-flight sockets are considered, and from those only ones
 	 * which don't have any external reference.
@@ -237,14 +277,12 @@ void unix_gc(void)
 	 */
 	list_for_each_entry_safe(u, next, &gc_inflight_list, link) {
 		long total_refs;
-		long inflight_refs;
 
 		total_refs = file_count(u->sk.sk_socket->file);
-		inflight_refs = atomic_long_read(&u->inflight);
 
-		BUG_ON(inflight_refs < 1);
-		BUG_ON(total_refs < inflight_refs);
-		if (total_refs == inflight_refs) {
+		WARN_ON_ONCE(!u->inflight);
+		WARN_ON_ONCE(total_refs < u->inflight);
+		if (total_refs == u->inflight) {
 			list_move_tail(&u->link, &gc_candidates);
 			__set_bit(UNIX_GC_CANDIDATE, &u->gc_flags);
 			__set_bit(UNIX_GC_MAYBE_CYCLE, &u->gc_flags);
@@ -271,7 +309,7 @@ void unix_gc(void)
 		/* Move cursor to after the current position. */
 		list_move(&cursor, &u->link);
 
-		if (atomic_long_read(&u->inflight) > 0) {
+		if (u->inflight) {
 			list_move_tail(&u->link, &not_cycle_list);
 			__clear_bit(UNIX_GC_MAYBE_CYCLE, &u->gc_flags);
 			scan_children(&u->sk, inc_inflight_move_tail, NULL);
@@ -306,38 +344,50 @@ void unix_gc(void)
 
 	spin_unlock(&unix_gc_lock);
 
-	/* We need io_uring to clean its registered files, ignore all io_uring
-	 * originated skbs. It's fine as io_uring doesn't keep references to
-	 * other io_uring instances and so killing all other files in the cycle
-	 * will put all io_uring references forcing it to go through normal
-	 * release.path eventually putting registered files.
-	 */
-	skb_queue_walk_safe(&hitlist, skb, next_skb) {
-		if (skb->destructor == io_uring_destruct_scm) {
-			__skb_unlink(skb, &hitlist);
-			skb_queue_tail(&skb->sk->sk_receive_queue, skb);
-		}
-	}
-
 	/* Here we are. Hitlist is filled. Die. */
 	__skb_queue_purge(&hitlist);
 
 	spin_lock(&unix_gc_lock);
 
-	/* There could be io_uring registered files, just push them back to
-	 * the inflight list
-	 */
-	list_for_each_entry_safe(u, next, &gc_candidates, link)
-		list_move_tail(&u->link, &gc_inflight_list);
-
 	/* All candidates should have been detached by now. */
-	BUG_ON(!list_empty(&gc_candidates));
+	WARN_ON_ONCE(!list_empty(&gc_candidates));
 
 	/* Paired with READ_ONCE() in wait_for_unix_gc(). */
 	WRITE_ONCE(gc_in_progress, false);
 
-	wake_up(&unix_gc_wait);
-
- out:
 	spin_unlock(&unix_gc_lock);
 }
+
+static DECLARE_WORK(unix_gc_work, __unix_gc);
+
+void unix_gc(void)
+{
+	WRITE_ONCE(gc_in_progress, true);
+	queue_work(system_unbound_wq, &unix_gc_work);
+}
+
+#define UNIX_INFLIGHT_TRIGGER_GC 16000
+#define UNIX_INFLIGHT_SANE_USER (SCM_MAX_FD * 8)
+
+void wait_for_unix_gc(struct scm_fp_list *fpl)
+{
+	/* If number of inflight sockets is insane,
+	 * force a garbage collect right now.
+	 *
+	 * Paired with the WRITE_ONCE() in unix_inflight(),
+	 * unix_notinflight(), and __unix_gc().
+	 */
+	if (READ_ONCE(unix_tot_inflight) > UNIX_INFLIGHT_TRIGGER_GC &&
+	    !READ_ONCE(gc_in_progress))
+		unix_gc();
+
+	/* Penalise users who want to send AF_UNIX sockets
+	 * but whose sockets have not been received yet.
+	 */
+	if (!fpl || !fpl->count_unix ||
+	    READ_ONCE(fpl->user->unix_inflight) < UNIX_INFLIGHT_SANE_USER)
+		return;
+
+	if (READ_ONCE(gc_in_progress))
+		flush_work(&unix_gc_work);
+}