1 files changed, 65 insertions, 39 deletions

diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 4dce1b418acc..00648a478c6a 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -436,7 +436,7 @@ void tcp_init_sock(struct sock *sk)
 	tp->snd_cwnd_clamp = ~0;
 	tp->mss_cache = TCP_MSS_DEFAULT;
 
-	tp->reordering = sock_net(sk)->ipv4.sysctl_tcp_reordering;
+	tp->reordering = READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_reordering);
 	tcp_assign_congestion_control(sk);
 
 	tp->tsoffset = 0;
@@ -515,6 +515,7 @@ __poll_t tcp_poll(struct file *file, struct socket *sock, poll_table *wait)
 	__poll_t mask;
 	struct sock *sk = sock->sk;
 	const struct tcp_sock *tp = tcp_sk(sk);
+	u8 shutdown;
 	int state;
 
 	sock_poll_wait(file, sock, wait);
@@ -557,9 +558,10 @@ __poll_t tcp_poll(struct file *file, struct socket *sock, poll_table *wait)
 	 * NOTE. Check for TCP_CLOSE is added. The goal is to prevent
 	 * blocking on fresh not-connected or disconnected socket. --ANK
 	 */
-	if (sk->sk_shutdown == SHUTDOWN_MASK || state == TCP_CLOSE)
+	shutdown = READ_ONCE(sk->sk_shutdown);
+	if (shutdown == SHUTDOWN_MASK || state == TCP_CLOSE)
 		mask |= EPOLLHUP;
-	if (sk->sk_shutdown & RCV_SHUTDOWN)
+	if (shutdown & RCV_SHUTDOWN)
 		mask |= EPOLLIN | EPOLLRDNORM | EPOLLRDHUP;
 
 	/* Connected or passive Fast Open socket? */
@@ -575,8 +577,8 @@ __poll_t tcp_poll(struct file *file, struct socket *sock, poll_table *wait)
 		if (tcp_stream_is_readable(tp, target, sk))
 			mask |= EPOLLIN | EPOLLRDNORM;
 
-		if (!(sk->sk_shutdown & SEND_SHUTDOWN)) {
-			if (sk_stream_is_writeable(sk)) {
+		if (!(shutdown & SEND_SHUTDOWN)) {
+			if (__sk_stream_is_writeable(sk, 1)) {
 				mask |= EPOLLOUT | EPOLLWRNORM;
 			} else {  /* send SIGIO later */
 				sk_set_bit(SOCKWQ_ASYNC_NOSPACE, sk);
@@ -588,7 +590,7 @@ __poll_t tcp_poll(struct file *file, struct socket *sock, poll_table *wait)
 				 * pairs with the input side.
 				 */
 				smp_mb__after_atomic();
-				if (sk_stream_is_writeable(sk))
+				if (__sk_stream_is_writeable(sk, 1))
 					mask |= EPOLLOUT | EPOLLWRNORM;
 			}
 		} else
@@ -706,7 +708,7 @@ static bool tcp_should_autocork(struct sock *sk, struct sk_buff *skb,
 				int size_goal)
 {
 	return skb->len < size_goal &&
-	       sock_net(sk)->ipv4.sysctl_tcp_autocorking &&
+	       READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_autocorking) &&
 	       !tcp_rtx_queue_empty(sk) &&
 	       refcount_read(&sk->sk_wmem_alloc) > skb->truesize;
 }
@@ -1160,7 +1162,8 @@ static int tcp_sendmsg_fastopen(struct sock *sk, struct msghdr *msg,
 	struct sockaddr *uaddr = msg->msg_name;
 	int err, flags;
 
-	if (!(sock_net(sk)->ipv4.sysctl_tcp_fastopen & TFO_CLIENT_ENABLE) ||
+	if (!(READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_fastopen) &
+	      TFO_CLIENT_ENABLE) ||
 	    (uaddr && msg->msg_namelen >= sizeof(uaddr->sa_family) &&
 	     uaddr->sa_family == AF_UNSPEC))
 		return -EOPNOTSUPP;
@@ -1669,11 +1672,13 @@ int tcp_read_sock(struct sock *sk, read_descriptor_t *desc,
 				if (!copied)
 					copied = used;
 				break;
-			} else if (used <= len) {
-				seq += used;
-				copied += used;
-				offset += used;
 			}
+			if (WARN_ON_ONCE(used > len))
+				used = len;
+			seq += used;
+			copied += used;
+			offset += used;
+
 			/* If recv_actor drops the lock (e.g. TCP splice
 			 * receive) the skb pointer might be invalid when
 			 * getting here: tcp_collapse might have deleted it
@@ -2335,14 +2340,13 @@ bool tcp_check_oom(struct sock *sk, int shift)
 	return too_many_orphans || out_of_socket_memory;
 }
 
-void tcp_close(struct sock *sk, long timeout)
+void __tcp_close(struct sock *sk, long timeout)
 {
 	struct sk_buff *skb;
 	int data_was_unread = 0;
 	int state;
 
-	lock_sock(sk);
-	sk->sk_shutdown = SHUTDOWN_MASK;
+	WRITE_ONCE(sk->sk_shutdown, SHUTDOWN_MASK);
 
 	if (sk->sk_state == TCP_LISTEN) {
 		tcp_set_state(sk, TCP_CLOSE);
@@ -2502,6 +2506,12 @@ adjudge_to_death:
 out:
 	bh_unlock_sock(sk);
 	local_bh_enable();
+}
+
+void tcp_close(struct sock *sk, long timeout)
+{
+	lock_sock(sk);
+	__tcp_close(sk, timeout);
 	release_sock(sk);
 	sock_put(sk);
 }
@@ -2558,6 +2568,12 @@ int tcp_disconnect(struct sock *sk, int flags)
 	int old_state = sk->sk_state;
 	u32 seq;
 
+	/* Deny disconnect if other threads are blocked in sk_wait_event()
+	 * or inet_wait_for_connect().
+	 */
+	if (sk->sk_wait_pending)
+		return -EBUSY;
+
 	if (old_state != TCP_CLOSE)
 		tcp_set_state(sk, TCP_CLOSE);
 
@@ -2590,7 +2606,7 @@ int tcp_disconnect(struct sock *sk, int flags)
 	if (!(sk->sk_userlocks & SOCK_BINDADDR_LOCK))
 		inet_reset_saddr(sk);
 
-	sk->sk_shutdown = 0;
+	WRITE_ONCE(sk->sk_shutdown, 0);
 	sock_reset_flag(sk, SOCK_DONE);
 	tp->srtt_us = 0;
 	tp->rcv_rtt_last_tsecr = 0;
@@ -2604,6 +2620,8 @@ int tcp_disconnect(struct sock *sk, int flags)
 	icsk->icsk_probes_out = 0;
 	tp->snd_ssthresh = TCP_INFINITE_SSTHRESH;
 	tp->snd_cwnd_cnt = 0;
+	tp->is_cwnd_limited = 0;
+	tp->max_packets_out = 0;
 	tp->window_clamp = 0;
 	tp->delivered = 0;
 	tp->delivered_ce = 0;
@@ -2621,8 +2639,7 @@ int tcp_disconnect(struct sock *sk, int flags)
 	icsk->icsk_ack.rcv_mss = TCP_MIN_MSS;
 	memset(&tp->rx_opt, 0, sizeof(tp->rx_opt));
 	__sk_dst_reset(sk);
-	dst_release(sk->sk_rx_dst);
-	sk->sk_rx_dst = NULL;
+	dst_release(xchg((__force struct dst_entry **)&sk->sk_rx_dst, NULL));
 	tcp_saved_syn_free(tp);
 	tp->compressed_ack = 0;
 	tp->segs_in = 0;
@@ -2983,18 +3000,18 @@ static int do_tcp_setsockopt(struct sock *sk, int level,
 
 	case TCP_LINGER2:
 		if (val < 0)
-			tp->linger2 = -1;
-		else if (val > net->ipv4.sysctl_tcp_fin_timeout / HZ)
-			tp->linger2 = 0;
+			WRITE_ONCE(tp->linger2, -1);
+		else if (val > TCP_FIN_TIMEOUT_MAX / HZ)
+			WRITE_ONCE(tp->linger2, TCP_FIN_TIMEOUT_MAX);
 		else
-			tp->linger2 = val * HZ;
+			WRITE_ONCE(tp->linger2, val * HZ);
 		break;
 
 	case TCP_DEFER_ACCEPT:
 		/* Translate value in seconds to number of retransmits */
-		icsk->icsk_accept_queue.rskq_defer_accept =
-			secs_to_retrans(val, TCP_TIMEOUT_INIT / HZ,
-					TCP_RTO_MAX / HZ);
+		WRITE_ONCE(icsk->icsk_accept_queue.rskq_defer_accept,
+			   secs_to_retrans(val, TCP_TIMEOUT_INIT / HZ,
+					   TCP_RTO_MAX / HZ));
 		break;
 
 	case TCP_WINDOW_CLAMP:
@@ -3054,7 +3071,8 @@ static int do_tcp_setsockopt(struct sock *sk, int level,
 	case TCP_FASTOPEN_CONNECT:
 		if (val > 1 || val < 0) {
 			err = -EINVAL;
-		} else if (net->ipv4.sysctl_tcp_fastopen & TFO_CLIENT_ENABLE) {
+		} else if (READ_ONCE(net->ipv4.sysctl_tcp_fastopen) &
+			   TFO_CLIENT_ENABLE) {
 			if (sk->sk_state == TCP_CLOSE)
 				tp->fastopen_connect = val;
 			else
@@ -3081,7 +3099,7 @@ static int do_tcp_setsockopt(struct sock *sk, int level,
 		err = tcp_repair_set_window(tp, optval, optlen);
 		break;
 	case TCP_NOTSENT_LOWAT:
-		tp->notsent_lowat = val;
+		WRITE_ONCE(tp->notsent_lowat, val);
 		sk->sk_write_space(sk);
 		break;
 	case TCP_INQ:
@@ -3358,7 +3376,8 @@ static int do_tcp_getsockopt(struct sock *sk, int level,
 	switch (optname) {
 	case TCP_MAXSEG:
 		val = tp->mss_cache;
-		if (!val && ((1 << sk->sk_state) & (TCPF_CLOSE | TCPF_LISTEN)))
+		if (tp->rx_opt.user_mss &&
+		    ((1 << sk->sk_state) & (TCPF_CLOSE | TCPF_LISTEN)))
 			val = tp->rx_opt.user_mss;
 		if (tp->repair)
 			val = tp->rx_opt.mss_clamp;
@@ -3382,13 +3401,14 @@ static int do_tcp_getsockopt(struct sock *sk, int level,
 		val = icsk->icsk_syn_retries ? : net->ipv4.sysctl_tcp_syn_retries;
 		break;
 	case TCP_LINGER2:
-		val = tp->linger2;
+		val = READ_ONCE(tp->linger2);
 		if (val >= 0)
-			val = (val ? : net->ipv4.sysctl_tcp_fin_timeout) / HZ;
+			val = (val ? : READ_ONCE(net->ipv4.sysctl_tcp_fin_timeout)) / HZ;
 		break;
 	case TCP_DEFER_ACCEPT:
-		val = retrans_to_secs(icsk->icsk_accept_queue.rskq_defer_accept,
-				      TCP_TIMEOUT_INIT / HZ, TCP_RTO_MAX / HZ);
+		val = READ_ONCE(icsk->icsk_accept_queue.rskq_defer_accept);
+		val = retrans_to_secs(val, TCP_TIMEOUT_INIT / HZ,
+				      TCP_RTO_MAX / HZ);
 		break;
 	case TCP_WINDOW_CLAMP:
 		val = tp->window_clamp;
@@ -3534,7 +3554,7 @@ static int do_tcp_getsockopt(struct sock *sk, int level,
 		break;
 
 	case TCP_FASTOPEN:
-		val = icsk->icsk_accept_queue.fastopenq.max_qlen;
+		val = READ_ONCE(icsk->icsk_accept_queue.fastopenq.max_qlen);
 		break;
 
 	case TCP_FASTOPEN_CONNECT:
@@ -3549,7 +3569,7 @@ static int do_tcp_getsockopt(struct sock *sk, int level,
 		val = tcp_time_stamp_raw() + tp->tsoffset;
 		break;
 	case TCP_NOTSENT_LOWAT:
-		val = tp->notsent_lowat;
+		val = READ_ONCE(tp->notsent_lowat);
 		break;
 	case TCP_INQ:
 		val = tp->recvmsg_inq;
@@ -3686,12 +3706,16 @@ static void __tcp_alloc_md5sig_pool(void)
 	 * to memory. See smp_rmb() in tcp_get_md5sig_pool()
 	 */
 	smp_wmb();
-	tcp_md5sig_pool_populated = true;
+	/* Paired with READ_ONCE() from tcp_alloc_md5sig_pool()
+	 * and tcp_get_md5sig_pool().
+	*/
+	WRITE_ONCE(tcp_md5sig_pool_populated, true);
 }
 
 bool tcp_alloc_md5sig_pool(void)
 {
-	if (unlikely(!tcp_md5sig_pool_populated)) {
+	/* Paired with WRITE_ONCE() from __tcp_alloc_md5sig_pool() */
+	if (unlikely(!READ_ONCE(tcp_md5sig_pool_populated))) {
 		mutex_lock(&tcp_md5sig_mutex);
 
 		if (!tcp_md5sig_pool_populated)
@@ -3699,7 +3723,8 @@ bool tcp_alloc_md5sig_pool(void)
 
 		mutex_unlock(&tcp_md5sig_mutex);
 	}
-	return tcp_md5sig_pool_populated;
+	/* Paired with WRITE_ONCE() from __tcp_alloc_md5sig_pool() */
+	return READ_ONCE(tcp_md5sig_pool_populated);
 }
 EXPORT_SYMBOL(tcp_alloc_md5sig_pool);
 
@@ -3715,7 +3740,8 @@ struct tcp_md5sig_pool *tcp_get_md5sig_pool(void)
 {
 	local_bh_disable();
 
-	if (tcp_md5sig_pool_populated) {
+	/* Paired with WRITE_ONCE() from __tcp_alloc_md5sig_pool() */
+	if (READ_ONCE(tcp_md5sig_pool_populated)) {
 		/* coupled with smp_wmb() in __tcp_alloc_md5sig_pool() */
 		smp_rmb();
 		return this_cpu_ptr(&tcp_md5sig_pool);
@@ -3791,7 +3817,7 @@ void tcp_done(struct sock *sk)
 	if (req)
 		reqsk_fastopen_remove(sk, req, false);
 
-	sk->sk_shutdown = SHUTDOWN_MASK;
+	WRITE_ONCE(sk->sk_shutdown, SHUTDOWN_MASK);
 
 	if (!sock_flag(sk, SOCK_DEAD))
 		sk->sk_state_change(sk);