Merge branch 'af_unix-minor-fixes-for-msg_oob-and-msg_peek'

Kuniyuki Iwashima says:

====================
af_unix: Minor fixes for MSG_OOB and MSG_PEEK.

Fahad Alharbi reported blocking recv(MSG_PEEK) could hog CPU
due to OOB skb.

Patch 1 and 2 fixes the issues and Patch 3 adds tests.
====================

Link: https://patch.msgid.link/20260902202202.892676-1-kuniyu@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

2 files changed, 75 insertions, 4 deletions

diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c
index 13f9926bf205..2da1017f8873 100644
--- a/net/unix/af_unix.c
+++ b/net/unix/af_unix.c
@@ -2812,8 +2812,8 @@ static int unix_stream_recv_urg(struct unix_stream_read_state *state)
 	return 1;
 }
 
-static struct sk_buff *manage_oob(struct sk_buff *skb, struct sock *sk,
-				  int flags, int copied)
+static struct sk_buff *manage_oob(struct sk_buff *skb, struct sk_buff **last,
+				  struct sock *sk, int flags, int copied)
 {
 	struct sk_buff *read_skb = NULL, *unread_skb = NULL;
 	struct unix_sock *u = unix_sk(sk);
@@ -2827,11 +2827,13 @@ static struct sk_buff *manage_oob(struct sk_buff *skb, struct sock *sk,
 		if (copied && (!u->oob_skb || skb == u->oob_skb)) {
 			skb = NULL;
 		} else if (flags & MSG_PEEK) {
+			*last = skb;
 			skb = skb_peek_next(skb, &sk->sk_receive_queue);
 		} else {
 			read_skb = skb;
 			skb = skb_peek_next(skb, &sk->sk_receive_queue);
 			__skb_unlink(read_skb, &sk->sk_receive_queue);
+			*last = skb;
 		}
 
 		if (!skb)
@@ -2850,8 +2852,10 @@ static struct sk_buff *manage_oob(struct sk_buff *skb, struct sock *sk,
 			__skb_unlink(skb, &sk->sk_receive_queue);
 			unread_skb = skb;
 			skb = skb_peek(&sk->sk_receive_queue);
+			*last = skb;
 		}
 	} else if (!sock_flag(sk, SOCK_URGINLINE)) {
+		*last = skb;
 		skb = skb_peek_next(skb, &sk->sk_receive_queue);
 	}
 
@@ -2971,8 +2975,8 @@ redo:
 again:
 #if IS_ENABLED(CONFIG_AF_UNIX_OOB)
 		if (skb) {
-			skb = manage_oob(skb, sk, flags, copied);
-			if (!skb && copied) {
+			skb = manage_oob(skb, &last, sk, flags, copied);
+			if (!skb && (copied || !state->size)) {
 				unix_state_unlock(sk);
 				break;
 			}
diff --git a/tools/testing/selftests/net/af_unix/msg_oob.c b/tools/testing/selftests/net/af_unix/msg_oob.c
index 1b499d56656c..f051d79f7a8e 100644
--- a/tools/testing/selftests/net/af_unix/msg_oob.c
+++ b/tools/testing/selftests/net/af_unix/msg_oob.c
@@ -290,6 +290,25 @@ static void __setinlinepair(struct __test_metadata *_metadata,
 	}
 }
 
+static void __setblockingpair(struct __test_metadata *_metadata,
+			      FIXTURE_DATA(msg_oob) *self)
+{
+	int i;
+
+	for (i = 0; i < 2; i++) {
+		int ret, old_flags, flags;
+
+		old_flags = fcntl(self->fd[i * 2 + 1], F_GETFL, 0);
+		ASSERT_NE(-1, old_flags);
+
+		ret = fcntl(self->fd[i * 2 + 1], F_SETFL, old_flags & ~O_NONBLOCK);
+		ASSERT_EQ(0, ret);
+
+		flags = fcntl(self->fd[i * 2 + 1], F_GETFL, 0);
+		ASSERT_EQ(old_flags & ~O_NONBLOCK, flags);
+	}
+}
+
 static void __siocatmarkpair(struct __test_metadata *_metadata,
 			     FIXTURE_DATA(msg_oob) *self,
 			     bool oob_head)
@@ -347,6 +366,9 @@ static void __resetpair(struct __test_metadata *_metadata,
 #define setinlinepair()							\
 	__setinlinepair(_metadata, self)
 
+#define setblockingpair()						\
+	__setblockingpair(_metadata, self)
+
 #define resetpair(reset)						\
 	__resetpair(_metadata, self, variant, reset)
 
@@ -888,4 +910,49 @@ TEST_F(msg_oob, inline_ex_oob_siocatmark)
 	resetpair(true);
 }
 
+TEST_F(msg_oob, zero_buf_oob)
+{
+	sendpair("a", 1, MSG_OOB);
+	recvpair("", 0, 0, 0);
+}
+
+TEST_F(msg_oob, zero_buf_oob_blocking)
+{
+	sendpair("a", 1, MSG_OOB);
+	setblockingpair();
+	recvpair("", 0, 0, 0);
+}
+
+TEST_F(msg_oob, zero_buf_non_oob_oob)
+{
+	sendpair("ab", 2, MSG_OOB);
+	recvpair("", 0, 0, 0);
+}
+
+TEST_F(msg_oob, zero_buf_non_oob_oob_blocking)
+{
+	sendpair("ab", 2, MSG_OOB);
+	setblockingpair();
+	recvpair("", 0, 0, 0);
+}
+
+TEST_F(msg_oob, zero_buf_ex_oob_oob)
+{
+	sendpair("a", 1, MSG_OOB);
+	recvpair("a", 1, 1, MSG_OOB);
+
+	sendpair("b", 1, MSG_OOB);
+	recvpair("", 0, 0, 0);
+}
+
+TEST_F(msg_oob, zero_buf_ex_oob_oob_blocking)
+{
+	sendpair("a", 1, MSG_OOB);
+	recvpair("a", 1, 1, MSG_OOB);
+
+	sendpair("b", 1, MSG_OOB);
+	setblockingpair();
+	recvpair("", 0, 0, 0);
+}
+
 TEST_HARNESS_MAIN