Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next

Pull networking updates from David Miller:

 1) BPF debugger and asm tool by Daniel Borkmann.

 2) Speed up create/bind in AF_PACKET, also from Daniel Borkmann.

 3) Correct reciprocal_divide and update users, from Hannes Frederic
    Sowa and Daniel Borkmann.

 4) Currently we only have a "set" operation for the hw timestamp socket
    ioctl, add a "get" operation to match.  From Ben Hutchings.

 5) Add better trace events for debugging driver datapath problems, also
    from Ben Hutchings.

 6) Implement auto corking in TCP, from Eric Dumazet.  Basically, if we
    have a small send and a previous packet is already in the qdisc or
    device queue, defer until TX completion or we get more data.

 7) Allow userspace to manage ipv6 temporary addresses, from Jiri Pirko.

 8) Add a qdisc bypass option for AF_PACKET sockets, from Daniel
    Borkmann.

 9) Share IP header compression code between Bluetooth and IEEE802154
    layers, from Jukka Rissanen.

10) Fix ipv6 router reachability probing, from Jiri Benc.

11) Allow packets to be captured on macvtap devices, from Vlad Yasevich.

12) Support tunneling in GRO layer, from Jerry Chu.

13) Allow bonding to be configured fully using netlink, from Scott
    Feldman.

14) Allow AF_PACKET users to obtain the VLAN TPID, just like they can
    already get the TCI.  From Atzm Watanabe.

15) New "Heavy Hitter" qdisc, from Terry Lam.

16) Significantly improve the IPSEC support in pktgen, from Fan Du.

17) Allow ipv4 tunnels to cache routes, just like sockets.  From Tom
    Herbert.

18) Add Proportional Integral Enhanced packet scheduler, from Vijay
    Subramanian.

19) Allow openvswitch to mmap'd netlink, from Thomas Graf.

20) Key TCP metrics blobs also by source address, not just destination
    address.  From Christoph Paasch.

21) Support 10G in generic phylib.  From Andy Fleming.

22) Try to short-circuit GRO flow compares using device provided RX
    hash, if provided.  From Tom Herbert.

The wireless and netfilter folks have been busy little bees too.

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next: (2064 commits)
  net/cxgb4: Fix referencing freed adapter
  ipv6: reallocate addrconf router for ipv6 address when lo device up
  fib_frontend: fix possible NULL pointer dereference
  rtnetlink: remove IFLA_BOND_SLAVE definition
  rtnetlink: remove check for fill_slave_info in rtnl_have_link_slave_info
  qlcnic: update version to 5.3.55
  qlcnic: Enhance logic to calculate msix vectors.
  qlcnic: Refactor interrupt coalescing code for all adapters.
  qlcnic: Update poll controller code path
  qlcnic: Interrupt code cleanup
  qlcnic: Enhance Tx timeout debugging.
  qlcnic: Use bool for rx_mac_learn.
  bonding: fix u64 division
  rtnetlink: add missing IFLA_BOND_AD_INFO_UNSPEC
  sfc: Use the correct maximum TX DMA ring size for SFC9100
  Add Shradha Shah as the sfc driver maintainer.
  net/vxlan: Share RX skb de-marking and checksum checks with ovs
  tulip: cleanup by using ARRAY_SIZE()
  ip_tunnel: clear IPCB in ip_tunnel_xmit() in case dst_link_failure() is called
  net/cxgb4: Don't retrieve stats during recovery
  ...

1 files changed, 197 insertions, 133 deletions

diff --git a/net/tipc/socket.c b/net/tipc/socket.c
index e741416d1d24..aab4948f0aff 100644
--- a/net/tipc/socket.c
+++ b/net/tipc/socket.c
@@ -55,9 +55,6 @@ struct tipc_sock {
 #define tipc_sk(sk) ((struct tipc_sock *)(sk))
 #define tipc_sk_port(sk) (tipc_sk(sk)->p)
 
-#define tipc_rx_ready(sock) (!skb_queue_empty(&sock->sk->sk_receive_queue) || \
-			(sock->state == SS_DISCONNECTING))
-
 static int backlog_rcv(struct sock *sk, struct sk_buff *skb);
 static u32 dispatch(struct tipc_port *tport, struct sk_buff *buf);
 static void wakeupdispatch(struct tipc_port *tport);
@@ -239,7 +236,6 @@ static int tipc_sk_create(struct net *net, struct socket *sock, int protocol,
 int tipc_sock_create_local(int type, struct socket **res)
 {
 	int rc;
-	struct sock *sk;
 
 	rc = sock_create_lite(AF_TIPC, type, 0, res);
 	if (rc < 0) {
@@ -248,8 +244,6 @@ int tipc_sock_create_local(int type, struct socket **res)
 	}
 	tipc_sk_create(&init_net, *res, 0, 1);
 
-	sk = (*res)->sk;
-
 	return 0;
 }
 
@@ -570,6 +564,31 @@ static int dest_name_check(struct sockaddr_tipc *dest, struct msghdr *m)
 	return 0;
 }
 
+static int tipc_wait_for_sndmsg(struct socket *sock, long *timeo_p)
+{
+	struct sock *sk = sock->sk;
+	struct tipc_port *tport = tipc_sk_port(sk);
+	DEFINE_WAIT(wait);
+	int done;
+
+	do {
+		int err = sock_error(sk);
+		if (err)
+			return err;
+		if (sock->state == SS_DISCONNECTING)
+			return -EPIPE;
+		if (!*timeo_p)
+			return -EAGAIN;
+		if (signal_pending(current))
+			return sock_intr_errno(*timeo_p);
+
+		prepare_to_wait(sk_sleep(sk), &wait, TASK_INTERRUPTIBLE);
+		done = sk_wait_event(sk, timeo_p, !tport->congested);
+		finish_wait(sk_sleep(sk), &wait);
+	} while (!done);
+	return 0;
+}
+
 /**
  * send_msg - send message in connectionless manner
  * @iocb: if NULL, indicates that socket lock is already held
@@ -589,9 +608,9 @@ static int send_msg(struct kiocb *iocb, struct socket *sock,
 {
 	struct sock *sk = sock->sk;
 	struct tipc_port *tport = tipc_sk_port(sk);
-	struct sockaddr_tipc *dest = (struct sockaddr_tipc *)m->msg_name;
+	DECLARE_SOCKADDR(struct sockaddr_tipc *, dest, m->msg_name);
 	int needs_conn;
-	long timeout_val;
+	long timeo;
 	int res = -EINVAL;
 
 	if (unlikely(!dest))
@@ -628,8 +647,7 @@ static int send_msg(struct kiocb *iocb, struct socket *sock,
 		reject_rx_queue(sk);
 	}
 
-	timeout_val = sock_sndtimeo(sk, m->msg_flags & MSG_DONTWAIT);
-
+	timeo = sock_sndtimeo(sk, m->msg_flags & MSG_DONTWAIT);
 	do {
 		if (dest->addrtype == TIPC_ADDR_NAME) {
 			res = dest_name_check(dest, m);
@@ -663,14 +681,9 @@ static int send_msg(struct kiocb *iocb, struct socket *sock,
 				sock->state = SS_CONNECTING;
 			break;
 		}
-		if (timeout_val <= 0L) {
-			res = timeout_val ? timeout_val : -EWOULDBLOCK;
+		res = tipc_wait_for_sndmsg(sock, &timeo);
+		if (res)
 			break;
-		}
-		release_sock(sk);
-		timeout_val = wait_event_interruptible_timeout(*sk_sleep(sk),
-					       !tport->congested, timeout_val);
-		lock_sock(sk);
 	} while (1);
 
 exit:
@@ -679,6 +692,34 @@ exit:
 	return res;
 }
 
+static int tipc_wait_for_sndpkt(struct socket *sock, long *timeo_p)
+{
+	struct sock *sk = sock->sk;
+	struct tipc_port *tport = tipc_sk_port(sk);
+	DEFINE_WAIT(wait);
+	int done;
+
+	do {
+		int err = sock_error(sk);
+		if (err)
+			return err;
+		if (sock->state == SS_DISCONNECTING)
+			return -EPIPE;
+		else if (sock->state != SS_CONNECTED)
+			return -ENOTCONN;
+		if (!*timeo_p)
+			return -EAGAIN;
+		if (signal_pending(current))
+			return sock_intr_errno(*timeo_p);
+
+		prepare_to_wait(sk_sleep(sk), &wait, TASK_INTERRUPTIBLE);
+		done = sk_wait_event(sk, timeo_p,
+				     (!tport->congested || !tport->connected));
+		finish_wait(sk_sleep(sk), &wait);
+	} while (!done);
+	return 0;
+}
+
 /**
  * send_packet - send a connection-oriented message
  * @iocb: if NULL, indicates that socket lock is already held
@@ -695,9 +736,9 @@ static int send_packet(struct kiocb *iocb, struct socket *sock,
 {
 	struct sock *sk = sock->sk;
 	struct tipc_port *tport = tipc_sk_port(sk);
-	struct sockaddr_tipc *dest = (struct sockaddr_tipc *)m->msg_name;
-	long timeout_val;
-	int res;
+	DECLARE_SOCKADDR(struct sockaddr_tipc *, dest, m->msg_name);
+	int res = -EINVAL;
+	long timeo;
 
 	/* Handle implied connection establishment */
 	if (unlikely(dest))
@@ -709,30 +750,24 @@ static int send_packet(struct kiocb *iocb, struct socket *sock,
 	if (iocb)
 		lock_sock(sk);
 
-	timeout_val = sock_sndtimeo(sk, m->msg_flags & MSG_DONTWAIT);
+	if (unlikely(sock->state != SS_CONNECTED)) {
+		if (sock->state == SS_DISCONNECTING)
+			res = -EPIPE;
+		else
+			res = -ENOTCONN;
+		goto exit;
+	}
 
+	timeo = sock_sndtimeo(sk, m->msg_flags & MSG_DONTWAIT);
 	do {
-		if (unlikely(sock->state != SS_CONNECTED)) {
-			if (sock->state == SS_DISCONNECTING)
-				res = -EPIPE;
-			else
-				res = -ENOTCONN;
-			break;
-		}
-
 		res = tipc_send(tport->ref, m->msg_iov, total_len);
 		if (likely(res != -ELINKCONG))
 			break;
-		if (timeout_val <= 0L) {
-			res = timeout_val ? timeout_val : -EWOULDBLOCK;
+		res = tipc_wait_for_sndpkt(sock, &timeo);
+		if (res)
 			break;
-		}
-		release_sock(sk);
-		timeout_val = wait_event_interruptible_timeout(*sk_sleep(sk),
-			(!tport->congested || !tport->connected), timeout_val);
-		lock_sock(sk);
 	} while (1);
-
+exit:
 	if (iocb)
 		release_sock(sk);
 	return res;
@@ -770,16 +805,11 @@ static int send_stream(struct kiocb *iocb, struct socket *sock,
 
 	/* Handle special cases where there is no connection */
 	if (unlikely(sock->state != SS_CONNECTED)) {
-		if (sock->state == SS_UNCONNECTED) {
+		if (sock->state == SS_UNCONNECTED)
 			res = send_packet(NULL, sock, m, total_len);
-			goto exit;
-		} else if (sock->state == SS_DISCONNECTING) {
-			res = -EPIPE;
-			goto exit;
-		} else {
-			res = -ENOTCONN;
-			goto exit;
-		}
+		else
+			res = sock->state == SS_DISCONNECTING ? -EPIPE : -ENOTCONN;
+		goto exit;
 	}
 
 	if (unlikely(m->msg_name)) {
@@ -876,7 +906,7 @@ static int auto_connect(struct socket *sock, struct tipc_msg *msg)
  */
 static void set_orig_addr(struct msghdr *m, struct tipc_msg *msg)
 {
-	struct sockaddr_tipc *addr = (struct sockaddr_tipc *)m->msg_name;
+	DECLARE_SOCKADDR(struct sockaddr_tipc *, addr, m->msg_name);
 
 	if (addr) {
 		addr->family = AF_TIPC;
@@ -961,6 +991,37 @@ static int anc_data_recv(struct msghdr *m, struct tipc_msg *msg,
 	return 0;
 }
 
+static int tipc_wait_for_rcvmsg(struct socket *sock, long timeo)
+{
+	struct sock *sk = sock->sk;
+	DEFINE_WAIT(wait);
+	int err;
+
+	for (;;) {
+		prepare_to_wait(sk_sleep(sk), &wait, TASK_INTERRUPTIBLE);
+		if (skb_queue_empty(&sk->sk_receive_queue)) {
+			if (sock->state == SS_DISCONNECTING) {
+				err = -ENOTCONN;
+				break;
+			}
+			release_sock(sk);
+			timeo = schedule_timeout(timeo);
+			lock_sock(sk);
+		}
+		err = 0;
+		if (!skb_queue_empty(&sk->sk_receive_queue))
+			break;
+		err = sock_intr_errno(timeo);
+		if (signal_pending(current))
+			break;
+		err = -EAGAIN;
+		if (!timeo)
+			break;
+	}
+	finish_wait(sk_sleep(sk), &wait);
+	return err;
+}
+
 /**
  * recv_msg - receive packet-oriented message
  * @iocb: (unused)
@@ -980,7 +1041,7 @@ static int recv_msg(struct kiocb *iocb, struct socket *sock,
 	struct tipc_port *tport = tipc_sk_port(sk);
 	struct sk_buff *buf;
 	struct tipc_msg *msg;
-	long timeout;
+	long timeo;
 	unsigned int sz;
 	u32 err;
 	int res;
@@ -996,25 +1057,13 @@ static int recv_msg(struct kiocb *iocb, struct socket *sock,
 		goto exit;
 	}
 
-	timeout = sock_rcvtimeo(sk, flags & MSG_DONTWAIT);
+	timeo = sock_rcvtimeo(sk, flags & MSG_DONTWAIT);
 restart:
 
 	/* Look for a message in receive queue; wait if necessary */
-	while (skb_queue_empty(&sk->sk_receive_queue)) {
-		if (sock->state == SS_DISCONNECTING) {
-			res = -ENOTCONN;
-			goto exit;
-		}
-		if (timeout <= 0L) {
-			res = timeout ? timeout : -EWOULDBLOCK;
-			goto exit;
-		}
-		release_sock(sk);
-		timeout = wait_event_interruptible_timeout(*sk_sleep(sk),
-							   tipc_rx_ready(sock),
-							   timeout);
-		lock_sock(sk);
-	}
+	res = tipc_wait_for_rcvmsg(sock, timeo);
+	if (res)
+		goto exit;
 
 	/* Look at first message in receive queue */
 	buf = skb_peek(&sk->sk_receive_queue);
@@ -1086,7 +1135,7 @@ static int recv_stream(struct kiocb *iocb, struct socket *sock,
 	struct tipc_port *tport = tipc_sk_port(sk);
 	struct sk_buff *buf;
 	struct tipc_msg *msg;
-	long timeout;
+	long timeo;
 	unsigned int sz;
 	int sz_to_copy, target, needed;
 	int sz_copied = 0;
@@ -1099,31 +1148,19 @@ static int recv_stream(struct kiocb *iocb, struct socket *sock,
 
 	lock_sock(sk);
 
-	if (unlikely((sock->state == SS_UNCONNECTED))) {
+	if (unlikely(sock->state == SS_UNCONNECTED)) {
 		res = -ENOTCONN;
 		goto exit;
 	}
 
 	target = sock_rcvlowat(sk, flags & MSG_WAITALL, buf_len);
-	timeout = sock_rcvtimeo(sk, flags & MSG_DONTWAIT);
+	timeo = sock_rcvtimeo(sk, flags & MSG_DONTWAIT);
 
 restart:
 	/* Look for a message in receive queue; wait if necessary */
-	while (skb_queue_empty(&sk->sk_receive_queue)) {
-		if (sock->state == SS_DISCONNECTING) {
-			res = -ENOTCONN;
-			goto exit;
-		}
-		if (timeout <= 0L) {
-			res = timeout ? timeout : -EWOULDBLOCK;
-			goto exit;
-		}
-		release_sock(sk);
-		timeout = wait_event_interruptible_timeout(*sk_sleep(sk),
-							   tipc_rx_ready(sock),
-							   timeout);
-		lock_sock(sk);
-	}
+	res = tipc_wait_for_rcvmsg(sock, timeo);
+	if (res)
+		goto exit;
 
 	/* Look at first message in receive queue */
 	buf = skb_peek(&sk->sk_receive_queue);
@@ -1327,14 +1364,12 @@ static u32 filter_connect(struct tipc_sock *tsock, struct sk_buff **buf)
 static unsigned int rcvbuf_limit(struct sock *sk, struct sk_buff *buf)
 {
 	struct tipc_msg *msg = buf_msg(buf);
-	unsigned int limit;
 
 	if (msg_connected(msg))
-		limit = sysctl_tipc_rmem[2];
-	else
-		limit = sk->sk_rcvbuf >> TIPC_CRITICAL_IMPORTANCE <<
-			msg_importance(msg);
-	return limit;
+		return sysctl_tipc_rmem[2];
+
+	return sk->sk_rcvbuf >> TIPC_CRITICAL_IMPORTANCE <<
+		msg_importance(msg);
 }
 
 /**
@@ -1448,6 +1483,28 @@ static void wakeupdispatch(struct tipc_port *tport)
 	sk->sk_write_space(sk);
 }
 
+static int tipc_wait_for_connect(struct socket *sock, long *timeo_p)
+{
+	struct sock *sk = sock->sk;
+	DEFINE_WAIT(wait);
+	int done;
+
+	do {
+		int err = sock_error(sk);
+		if (err)
+			return err;
+		if (!*timeo_p)
+			return -ETIMEDOUT;
+		if (signal_pending(current))
+			return sock_intr_errno(*timeo_p);
+
+		prepare_to_wait(sk_sleep(sk), &wait, TASK_INTERRUPTIBLE);
+		done = sk_wait_event(sk, timeo_p, sock->state != SS_CONNECTING);
+		finish_wait(sk_sleep(sk), &wait);
+	} while (!done);
+	return 0;
+}
+
 /**
  * connect - establish a connection to another TIPC port
  * @sock: socket structure
@@ -1463,7 +1520,8 @@ static int connect(struct socket *sock, struct sockaddr *dest, int destlen,
 	struct sock *sk = sock->sk;
 	struct sockaddr_tipc *dst = (struct sockaddr_tipc *)dest;
 	struct msghdr m = {NULL,};
-	unsigned int timeout;
+	long timeout = (flags & O_NONBLOCK) ? 0 : tipc_sk(sk)->conn_timeout;
+	socket_state previous;
 	int res;
 
 	lock_sock(sk);
@@ -1485,8 +1543,7 @@ static int connect(struct socket *sock, struct sockaddr *dest, int destlen,
 		goto exit;
 	}
 
-	timeout = (flags & O_NONBLOCK) ? 0 : tipc_sk(sk)->conn_timeout;
-
+	previous = sock->state;
 	switch (sock->state) {
 	case SS_UNCONNECTED:
 		/* Send a 'SYN-' to destination */
@@ -1508,43 +1565,22 @@ static int connect(struct socket *sock, struct sockaddr *dest, int destlen,
 		 * case is EINPROGRESS, rather than EALREADY.
 		 */
 		res = -EINPROGRESS;
-		break;
 	case SS_CONNECTING:
-		res = -EALREADY;
+		if (previous == SS_CONNECTING)
+			res = -EALREADY;
+		if (!timeout)
+			goto exit;
+		timeout = msecs_to_jiffies(timeout);
+		/* Wait until an 'ACK' or 'RST' arrives, or a timeout occurs */
+		res = tipc_wait_for_connect(sock, &timeout);
 		break;
 	case SS_CONNECTED:
 		res = -EISCONN;
 		break;
 	default:
 		res = -EINVAL;
-		goto exit;
-	}
-
-	if (sock->state == SS_CONNECTING) {
-		if (!timeout)
-			goto exit;
-
-		/* Wait until an 'ACK' or 'RST' arrives, or a timeout occurs */
-		release_sock(sk);
-		res = wait_event_interruptible_timeout(*sk_sleep(sk),
-				sock->state != SS_CONNECTING,
-				timeout ? (long)msecs_to_jiffies(timeout)
-					: MAX_SCHEDULE_TIMEOUT);
-		lock_sock(sk);
-		if (res <= 0) {
-			if (res == 0)
-				res = -ETIMEDOUT;
-			else
-				; /* leave "res" unchanged */
-			goto exit;
-		}
+		break;
 	}
-
-	if (unlikely(sock->state == SS_DISCONNECTING))
-		res = sock_error(sk);
-	else
-		res = 0;
-
 exit:
 	release_sock(sk);
 	return res;
@@ -1575,6 +1611,42 @@ static int listen(struct socket *sock, int len)
 	return res;
 }
 
+static int tipc_wait_for_accept(struct socket *sock, long timeo)
+{
+	struct sock *sk = sock->sk;
+	DEFINE_WAIT(wait);
+	int err;
+
+	/* True wake-one mechanism for incoming connections: only
+	 * one process gets woken up, not the 'whole herd'.
+	 * Since we do not 'race & poll' for established sockets
+	 * anymore, the common case will execute the loop only once.
+	*/
+	for (;;) {
+		prepare_to_wait_exclusive(sk_sleep(sk), &wait,
+					  TASK_INTERRUPTIBLE);
+		if (skb_queue_empty(&sk->sk_receive_queue)) {
+			release_sock(sk);
+			timeo = schedule_timeout(timeo);
+			lock_sock(sk);
+		}
+		err = 0;
+		if (!skb_queue_empty(&sk->sk_receive_queue))
+			break;
+		err = -EINVAL;
+		if (sock->state != SS_LISTENING)
+			break;
+		err = sock_intr_errno(timeo);
+		if (signal_pending(current))
+			break;
+		err = -EAGAIN;
+		if (!timeo)
+			break;
+	}
+	finish_wait(sk_sleep(sk), &wait);
+	return err;
+}
+
 /**
  * accept - wait for connection request
  * @sock: listening socket
@@ -1591,7 +1663,7 @@ static int accept(struct socket *sock, struct socket *new_sock, int flags)
 	struct tipc_port *new_tport;
 	struct tipc_msg *msg;
 	u32 new_ref;
-
+	long timeo;
 	int res;
 
 	lock_sock(sk);
@@ -1601,18 +1673,10 @@ static int accept(struct socket *sock, struct socket *new_sock, int flags)
 		goto exit;
 	}
 
-	while (skb_queue_empty(&sk->sk_receive_queue)) {
-		if (flags & O_NONBLOCK) {
-			res = -EWOULDBLOCK;
-			goto exit;
-		}
-		release_sock(sk);
-		res = wait_event_interruptible(*sk_sleep(sk),
-				(!skb_queue_empty(&sk->sk_receive_queue)));
-		lock_sock(sk);
-		if (res)
-			goto exit;
-	}
+	timeo = sock_rcvtimeo(sk, flags & O_NONBLOCK);
+	res = tipc_wait_for_accept(sock, timeo);
+	if (res)
+		goto exit;
 
 	buf = skb_peek(&sk->sk_receive_queue);