socket.c

/*
 * net/tipc/socket.c: TIPC socket API
 *
 * Copyright (c) 2001-2007, Ericsson AB
 * Copyright (c) 2004-2008, 2010-2011, Wind River Systems
 * All rights reserved.
 *
 * Redistribution and use in source and binary forms, with or without
 * modification, are permitted provided that the following conditions are met:
 *
 * 1. Redistributions of source code must retain the above copyright
 *    notice, this list of conditions and the following disclaimer.
 * 2. Redistributions in binary form must reproduce the above copyright
 *    notice, this list of conditions and the following disclaimer in the
 *    documentation and/or other materials provided with the distribution.
 * 3. Neither the names of the copyright holders nor the names of its
 *    contributors may be used to endorse or promote products derived from
 *    this software without specific prior written permission.
 *
 * Alternatively, this software may be distributed under the terms of the
 * GNU General Public License ("GPL") version 2 as published by the Free
 * Software Foundation.
 *
 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
 * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
 * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
 * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
 * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
 * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
 * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
 * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
 * POSSIBILITY OF SUCH DAMAGE.
 */

#include <linux/export.h>
#include <net/sock.h>

#include "core.h"
#include "port.h"

#define SS_LISTENING	-1	/* socket is listening */
#define SS_READY	-2	/* socket is connectionless */

#define OVERLOAD_LIMIT_BASE	5000
#define CONN_TIMEOUT_DEFAULT	8000	/* default connect timeout = 8s */

struct tipc_sock {
	struct sock sk;
	struct tipc_port *p;
	struct tipc_portid peer_name;
	unsigned int conn_timeout;
};

#define tipc_sk(sk) ((struct tipc_sock *)(sk))
#define tipc_sk_port(sk) ((struct tipc_port *)(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);

static const struct proto_ops packet_ops;
static const struct proto_ops stream_ops;
static const struct proto_ops msg_ops;

static struct proto tipc_proto;

static int sockets_enabled;

static atomic_t tipc_queue_size = ATOMIC_INIT(0);

/*
 * Revised TIPC socket locking policy:
 *
 * Most socket operations take the standard socket lock when they start
 * and hold it until they finish (or until they need to sleep).  Acquiring
 * this lock grants the owner exclusive access to the fields of the socket
 * data structures, with the exception of the backlog queue.  A few socket
 * operations can be done without taking the socket lock because they only
 * read socket information that never changes during the life of the socket.
 *
 * Socket operations may acquire the lock for the associated TIPC port if they
 * need to perform an operation on the port.  If any routine needs to acquire
 * both the socket lock and the port lock it must take the socket lock first
 * to avoid the risk of deadlock.
 *
 * The dispatcher handling incoming messages cannot grab the socket lock in
 * the standard fashion, since invoked it runs at the BH level and cannot block.
 * Instead, it checks to see if the socket lock is currently owned by someone,
 * and either handles the message itself or adds it to the socket's backlog
 * queue; in the latter case the queued message is processed once the process
 * owning the socket lock releases it.
 *
 * NOTE: Releasing the socket lock while an operation is sleeping overcomes
 * the problem of a blocked socket operation preventing any other operations
 * from occurring.  However, applications must be careful if they have
 * multiple threads trying to send (or receive) on the same socket, as these
 * operations might interfere with each other.  For example, doing a connect
 * and a receive at the same time might allow the receive to consume the
 * ACK message meant for the connect.  While additional work could be done
 * to try and overcome this, it doesn't seem to be worthwhile at the present.
 *
 * NOTE: Releasing the socket lock while an operation is sleeping also ensures
 * that another operation that must be performed in a non-blocking manner is
 * not delayed for very long because the lock has already been taken.
 *
 * NOTE: This code assumes that certain fields of a port/socket pair are
 * constant over its lifetime; such fields can be examined without taking
 * the socket lock and/or port lock, and do not need to be re-read even
 * after resuming processing after waiting.  These fields include:
 *   - socket type
 *   - pointer to socket sk structure (aka tipc_sock structure)
 *   - pointer to port structure
 *   - port reference
 */

/**
 * advance_rx_queue - discard first buffer in socket receive queue
 *
 * Caller must hold socket lock
 */

static void advance_rx_queue(struct sock *sk)
{
	buf_discard(__skb_dequeue(&sk->sk_receive_queue));
	atomic_dec(&tipc_queue_size);
}

/**
 * discard_rx_queue - discard all buffers in socket receive queue
 *
 * Caller must hold socket lock
 */

static void discard_rx_queue(struct sock *sk)
{
	struct sk_buff *buf;

	while ((buf = __skb_dequeue(&sk->sk_receive_queue))) {
		atomic_dec(&tipc_queue_size);
		buf_discard(buf);
	}
}

/**
 * reject_rx_queue - reject all buffers in socket receive queue
 *
 * Caller must hold socket lock
 */

static void reject_rx_queue(struct sock *sk)
{
	struct sk_buff *buf;

	while ((buf = __skb_dequeue(&sk->sk_receive_queue))) {
		tipc_reject_msg(buf, TIPC_ERR_NO_PORT);
		atomic_dec(&tipc_queue_size);
	}
}

/**
 * tipc_create - create a TIPC socket
 * @net: network namespace (must be default network)
 * @sock: pre-allocated socket structure
 * @protocol: protocol indicator (must be 0)
 * @kern: caused by kernel or by userspace?
 *
 * This routine creates additional data structures used by the TIPC socket,
 * initializes them, and links them together.
 *
 * Returns 0 on success, errno otherwise
 */

static int tipc_create(struct net *net, struct socket *sock, int protocol,
		       int kern)
{
	const struct proto_ops *ops;
	socket_state state;
	struct sock *sk;
	struct tipc_port *tp_ptr;

	/* Validate arguments */

	if (unlikely(protocol != 0))
		return -EPROTONOSUPPORT;

	switch (sock->type) {
	case SOCK_STREAM:
		ops = &stream_ops;
		state = SS_UNCONNECTED;
		break;
	case SOCK_SEQPACKET:
		ops = &packet_ops;
		state = SS_UNCONNECTED;
		break;
	case SOCK_DGRAM:
	case SOCK_RDM:
		ops = &msg_ops;
		state = SS_READY;
		break;
	default:
		return -EPROTOTYPE;
	}

	/* Allocate socket's protocol area */

	sk = sk_alloc(net, AF_TIPC, GFP_KERNEL, &tipc_proto);
	if (sk == NULL)
		return -ENOMEM;

	/* Allocate TIPC port for socket to use */

	tp_ptr = tipc_createport_raw(sk, &dispatch, &wakeupdispatch,
				     TIPC_LOW_IMPORTANCE);
	if (unlikely(!tp_ptr)) {
		sk_free(sk);
		return -ENOMEM;
	}

	/* Finish initializing socket data structures */

	sock->ops = ops;
	sock->state = state;

	sock_init_data(sock, sk);
	sk->sk_backlog_rcv = backlog_rcv;
	tipc_sk(sk)->p = tp_ptr;
	tipc_sk(sk)->conn_timeout = CONN_TIMEOUT_DEFAULT;

	spin_unlock_bh(tp_ptr->lock);

	if (sock->state == SS_READY) {
		tipc_set_portunreturnable(tp_ptr->ref, 1);
		if (sock->type == SOCK_DGRAM)
			tipc_set_portunreliable(tp_ptr->ref, 1);
	}

	return 0;
}

/**
 * release - destroy a TIPC socket
 * @sock: socket to destroy
 *
 * This routine cleans up any messages that are still queued on the socket.
 * For DGRAM and RDM socket types, all queued messages are rejected.
 * For SEQPACKET and STREAM socket types, the first message is rejected
 * and any others are discarded.  (If the first message on a STREAM socket
 * is partially-read, it is discarded and the next one is rejected instead.)
 *
 * NOTE: Rejected messages are not necessarily returned to the sender!  They
 * are returned or discarded according to the "destination droppable" setting
 * specified for the message by the sender.
 *
 * Returns 0 on success, errno otherwise
 */

static int release(struct socket *sock)
{
	struct sock *sk = sock->sk;
	struct tipc_port *tport;
	struct sk_buff *buf;
	int res;

	/*
	 * Exit if socket isn't fully initialized (occurs when a failed accept()
	 * releases a pre-allocated child socket that was never used)
	 */

	if (sk == NULL)
		return 0;

	tport = tipc_sk_port(sk);
	lock_sock(sk);

	/*
	 * Reject all unreceived messages, except on an active connection
	 * (which disconnects locally & sends a 'FIN+' to peer)
	 */

	while (sock->state != SS_DISCONNECTING) {
		buf = __skb_dequeue(&sk->sk_receive_queue);
		if (buf == NULL)
			break;
		atomic_dec(&tipc_queue_size);
		if (TIPC_SKB_CB(buf)->handle != 0)
			buf_discard(buf);
		else {
			if ((sock->state == SS_CONNECTING) ||
			    (sock->state == SS_CONNECTED)) {
				sock->state = SS_DISCONNECTING;
				tipc_disconnect(tport->ref);
			}
			tipc_reject_msg(buf, TIPC_ERR_NO_PORT);
		}
	}

	/*
	 * Delete TIPC port; this ensures no more messages are queued
	 * (also disconnects an active connection & sends a 'FIN-' to peer)
	 */

	res = tipc_deleteport(tport->ref);

	/* Discard any remaining (connection-based) messages in receive queue */

	discard_rx_queue(sk);

	/* Reject any messages that accumulated in backlog queue */

	sock->state = SS_DISCONNECTING;
	release_sock(sk);

	sock_put(sk);
	sock->sk = NULL;

	return res;
}

/**
 * bind - associate or disassocate TIPC name(s) with a socket
 * @sock: socket structure
 * @uaddr: socket address describing name(s) and desired operation
 * @uaddr_len: size of socket address data structure
 *
 * Name and name sequence binding is indicated using a positive scope value;
 * a negative scope value unbinds the specified name.  Specifying no name
 * (i.e. a socket address length of 0) unbinds all names from the socket.
 *
 * Returns 0 on success, errno otherwise
 *
 * NOTE: This routine doesn't need to take the socket lock since it doesn't
 *       access any non-constant socket information.
 */

static int bind(struct socket *sock, struct sockaddr *uaddr, int uaddr_len)
{
	struct sockaddr_tipc *addr = (struct sockaddr_tipc *)uaddr;
	u32 portref = tipc_sk_port(sock->sk)->ref;

	if (unlikely(!uaddr_len))
		return tipc_withdraw(portref, 0, NULL);

	if (uaddr_len < sizeof(struct sockaddr_tipc))
		return -EINVAL;
	if (addr->family != AF_TIPC)
		return -EAFNOSUPPORT;

	if (addr->addrtype == TIPC_ADDR_NAME)
		addr->addr.nameseq.upper = addr->addr.nameseq.lower;
	else if (addr->addrtype != TIPC_ADDR_NAMESEQ)
		return -EAFNOSUPPORT;

	if (addr->addr.nameseq.type < TIPC_RESERVED_TYPES)
		return -EACCES;

	return (addr->scope > 0) ?
		tipc_publish(portref, addr->scope, &addr->addr.nameseq) :
		tipc_withdraw(portref, -addr->scope, &addr->addr.nameseq);
}

/**
 * get_name - get port ID of socket or peer socket
 * @sock: socket structure
 * @uaddr: area for returned socket address
 * @uaddr_len: area for returned length of socket address
 * @peer: 0 = own ID, 1 = current peer ID, 2 = current/former peer ID
 *
 * Returns 0 on success, errno otherwise
 *
 * NOTE: This routine doesn't need to take the socket lock since it only
 *       accesses socket information that is unchanging (or which changes in
 *       a completely predictable manner).
 */

static int get_name(struct socket *sock, struct sockaddr *uaddr,
		    int *uaddr_len, int peer)
{
	struct sockaddr_tipc *addr = (struct sockaddr_tipc *)uaddr;
	struct tipc_sock *tsock = tipc_sk(sock->sk);

	memset(addr, 0, sizeof(*addr));
	if (peer) {
		if ((sock->state != SS_CONNECTED) &&
			((peer != 2) || (sock->state != SS_DISCONNECTING)))
			return -ENOTCONN;
		addr->addr.id.ref = tsock->peer_name.ref;
		addr->addr.id.node = tsock->peer_name.node;
	} else {
		addr->addr.id.ref = tsock->p->ref;
		addr->addr.id.node = tipc_own_addr;
	}

	*uaddr_len = sizeof(*addr);
	addr->addrtype = TIPC_ADDR_ID;
	addr->family = AF_TIPC;
	addr->scope = 0;
	addr->addr.name.domain = 0;

	return 0;
}

/**
 * poll - read and possibly block on pollmask
 * @file: file structure associated with the socket
 * @sock: socket for which to calculate the poll bits
 * @wait: ???
 *
 * Returns pollmask value
 *
 * COMMENTARY:
 * It appears that the usual socket locking mechanisms are not useful here
 * since the pollmask info is potentially out-of-date the moment this routine
 * exits.  TCP and other protocols seem to rely on higher level poll routines
 * to handle any preventable race conditions, so TIPC will do the same ...
 *
 * TIPC sets the returned events as follows:
 *
 * socket state		flags set
 * ------------		---------
 * unconnected		no read flags
 *			no write flags
 *
 * connecting		POLLIN/POLLRDNORM if ACK/NACK in rx queue
 *			no write flags
 *
 * connected		POLLIN/POLLRDNORM if data in rx queue
 *			POLLOUT if port is not congested
 *
 * disconnecting	POLLIN/POLLRDNORM/POLLHUP
 *			no write flags
 *
 * listening		POLLIN if SYN in rx queue
 *			no write flags
 *
 * ready		POLLIN/POLLRDNORM if data in rx queue
 * [connectionless]	POLLOUT (since port cannot be congested)
 *
 * IMPORTANT: The fact that a read or write operation is indicated does NOT
 * imply that the operation will succeed, merely that it should be performed
 * and will not block.
 */

static unsigned int poll(struct file *file, struct socket *sock,
			 poll_table *wait)
{
	struct sock *sk = sock->sk;
	u32 mask = 0;

	poll_wait(file, sk_sleep(sk), wait);

	switch ((int)sock->state) {
	case SS_READY:
	case SS_CONNECTED:
		if (!tipc_sk_port(sk)->congested)
			mask |= POLLOUT;
		/* fall thru' */
	case SS_CONNECTING:
	case SS_LISTENING:
		if (!skb_queue_empty(&sk->sk_receive_queue))
			mask |= (POLLIN | POLLRDNORM);
		break;
	case SS_DISCONNECTING:
		mask = (POLLIN | POLLRDNORM | POLLHUP);
		break;
	}

	return mask;
}

/**
 * dest_name_check - verify user is permitted to send to specified port name
 * @dest: destination address
 * @m: descriptor for message to be sent
 *
 * Prevents restricted configuration commands from being issued by
 * unauthorized users.
 *
 * Returns 0 if permission is granted, otherwise errno
 */

static int dest_name_check(struct sockaddr_tipc *dest, struct msghdr *m)
{
	struct tipc_cfg_msg_hdr hdr;

	if (likely(dest->addr.name.name.type >= TIPC_RESERVED_TYPES))
		return 0;
	if (likely(dest->addr.name.name.type == TIPC_TOP_SRV))
		return 0;
	if (likely(dest->addr.name.name.type != TIPC_CFG_SRV))
		return -EACCES;

	if (!m->msg_iovlen || (m->msg_iov[0].iov_len < sizeof(hdr)))
		return -EMSGSIZE;
	if (copy_from_user(&hdr, m->msg_iov[0].iov_base, sizeof(hdr)))
		return -EFAULT;
	if ((ntohs(hdr.tcm_type) & 0xC000) && (!capable(CAP_NET_ADMIN)))
		return -EACCES;

	return 0;
}

/**
 * send_msg - send message in connectionless manner
 * @iocb: if NULL, indicates that socket lock is already held
 * @sock: socket structure
 * @m: message to send
 * @total_len: length of message
 *
 * Message must have an destination specified explicitly.
 * Used for SOCK_RDM and SOCK_DGRAM messages,
 * and for 'SYN' messages on SOCK_SEQPACKET and SOCK_STREAM connections.
 * (Note: 'SYN+' is prohibited on SOCK_STREAM.)
 *
 * Returns the number of bytes sent on success, or errno otherwise
 */

static int send_msg(struct kiocb *iocb, struct socket *sock,
		    struct msghdr *m, size_t total_len)
{
	struct sock *sk = sock->sk;
	struct tipc_port *tport = tipc_sk_port(sk);
	struct sockaddr_tipc *dest = (struct sockaddr_tipc *)m->msg_name;
	int needs_conn;
	long timeout_val;
	int res = -EINVAL;

	if (unlikely(!dest))
		return -EDESTADDRREQ;
	if (unlikely((m->msg_namelen < sizeof(*dest)) ||
		     (dest->family != AF_TIPC)))
		return -EINVAL;
	if ((total_len > TIPC_MAX_USER_MSG_SIZE) ||
	    (m->msg_iovlen > (unsigned)INT_MAX))
		return -EMSGSIZE;

	if (iocb)
		lock_sock(sk);

	needs_conn = (sock->state != SS_READY);
	if (unlikely(needs_conn)) {
		if (sock->state == SS_LISTENING) {
			res = -EPIPE;
			goto exit;
		}
		if (sock->state != SS_UNCONNECTED) {
			res = -EISCONN;
			goto exit;
		}
		if ((tport->published) ||
		    ((sock->type == SOCK_STREAM) && (total_len != 0))) {
			res = -EOPNOTSUPP;
			goto exit;
		}
		if (dest->addrtype == TIPC_ADDR_NAME) {
			tport->conn_type = dest->addr.name.name.type;
			tport->conn_instance = dest->addr.name.name.instance;
		}

		/* Abort any pending connection attempts (very unlikely) */

		reject_rx_queue(sk);
	}

	timeout_val = sock_sndtimeo(sk, m->msg_flags & MSG_DONTWAIT);

	do {
		if (dest->addrtype == TIPC_ADDR_NAME) {
			res = dest_name_check(dest, m);
			if (res)
				break;
			res = tipc_send2name(tport->ref,
					     &dest->addr.name.name,
					     dest->addr.name.domain,
					     m->msg_iovlen,
					     m->msg_iov,
					     total_len);
		} else if (dest->addrtype == TIPC_ADDR_ID) {
			res = tipc_send2port(tport->ref,
					     &dest->addr.id,
					     m->msg_iovlen,
					     m->msg_iov,
					     total_len);
		} else if (dest->addrtype == TIPC_ADDR_MCAST) {
			if (needs_conn) {
				res = -EOPNOTSUPP;
				break;
			}
			res = dest_name_check(dest, m);
			if (res)
				break;
			res = tipc_multicast(tport->ref,
					     &dest->addr.nameseq,
					     m->msg_iovlen,
					     m->msg_iov,
					     total_len);
		}
		if (likely(res != -ELINKCONG)) {
			if (needs_conn && (res >= 0))
				sock->state = SS_CONNECTING;
			break;
		}
		if (timeout_val <= 0L) {
			res = timeout_val ? timeout_val : -EWOULDBLOCK;
			break;
		}
		release_sock(sk);
		timeout_val = wait_event_interruptible_timeout(*sk_sleep(sk),
					       !tport->congested, timeout_val);
		lock_sock(sk);
	} while (1);

exit:
	if (iocb)
		release_sock(sk);
	return res;
}

/**
 * send_packet - send a connection-oriented message
 * @iocb: if NULL, indicates that socket lock is already held
 * @sock: socket structure
 * @m: message to send
 * @total_len: length of message
 *
 * Used for SOCK_SEQPACKET messages and SOCK_STREAM data.
 *
 * Returns the number of bytes sent on success, or errno otherwise
 */

static int send_packet(struct kiocb *iocb, struct socket *sock,
		       struct msghdr *m, size_t total_len)
{
	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;

	/* Handle implied connection establishment */

	if (unlikely(dest))
		return send_msg(iocb, sock, m, total_len);

	if ((total_len > TIPC_MAX_USER_MSG_SIZE) ||
	    (m->msg_iovlen > (unsigned)INT_MAX))
		return -EMSGSIZE;

	if (iocb)
		lock_sock(sk);

	timeout_val = 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_iovlen, m->msg_iov,
				total_len);
		if (likely(res != -ELINKCONG))
			break;
		if (timeout_val <= 0L) {
			res = timeout_val ? timeout_val : -EWOULDBLOCK;
			break;
		}
		release_sock(sk);
		timeout_val = wait_event_interruptible_timeout(*sk_sleep(sk),
			(!tport->congested || !tport->connected), timeout_val);
		lock_sock(sk);
	} while (1);

	if (iocb)
		release_sock(sk);
	return res;
}

/**
 * send_stream - send stream-oriented data
 * @iocb: (unused)
 * @sock: socket structure
 * @m: data to send
 * @total_len: total length of data to be sent
 *
 * Used for SOCK_STREAM data.
 *
 * Returns the number of bytes sent on success (or partial success),
 * or errno if no data sent
 */

static int send_stream(struct kiocb *iocb, struct socket *sock,
		       struct msghdr *m, size_t total_len)
{
	struct sock *sk = sock->sk;
	struct tipc_port *tport = tipc_sk_port(sk);
	struct msghdr my_msg;
	struct iovec my_iov;
	struct iovec *curr_iov;
	int curr_iovlen;
	char __user *curr_start;
	u32 hdr_size;
	int curr_left;
	int bytes_to_send;
	int bytes_sent;
	int res;

	lock_sock(sk);

	/* Handle special cases where there is no connection */

	if (unlikely(sock->state != SS_CONNECTED)) {
		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;
		}
	}

	if (unlikely(m->msg_name)) {
		res = -EISCONN;
		goto exit;
	}

	if ((total_len > (unsigned)INT_MAX) ||
	    (m->msg_iovlen > (unsigned)INT_MAX)) {
		res = -EMSGSIZE;
		goto exit;
	}

	/*
	 * Send each iovec entry using one or more messages
	 *
	 * Note: This algorithm is good for the most likely case
	 * (i.e. one large iovec entry), but could be improved to pass sets
	 * of small iovec entries into send_packet().
	 */

	curr_iov = m->msg_iov;
	curr_iovlen = m->msg_iovlen;
	my_msg.msg_iov = &my_iov;
	my_msg.msg_iovlen = 1;
	my_msg.msg_flags = m->msg_flags;
	my_msg.msg_name = NULL;
	bytes_sent = 0;

	hdr_size = msg_hdr_sz(&tport->phdr);

	while (curr_iovlen--) {
		curr_start = curr_iov->iov_base;
		curr_left = curr_iov->iov_len;

		while (curr_left) {
			bytes_to_send = tport->max_pkt - hdr_size;
			if (bytes_to_send > TIPC_MAX_USER_MSG_SIZE)
				bytes_to_send = TIPC_MAX_USER_MSG_SIZE;
			if (curr_left < bytes_to_send)
				bytes_to_send = curr_left;
			my_iov.iov_base = curr_start;
			my_iov.iov_len = bytes_to_send;
			res = send_packet(NULL, sock, &my_msg, bytes_to_send);
			if (res < 0) {
				if (bytes_sent)
					res = bytes_sent;
				goto exit;
			}
			curr_left -= bytes_to_send;
			curr_start += bytes_to_send;
			bytes_sent += bytes_to_send;
		}

		curr_iov++;
	}
	res = bytes_sent;
exit:
	release_sock(sk);
	return res;
}

/**
 * auto_connect - complete connection setup to a remote port
 * @sock: socket structure
 * @msg: peer's response message
 *
 * Returns 0 on success, errno otherwise
 */

static int auto_connect(struct socket *sock, struct tipc_msg *msg)
{
	struct tipc_sock *tsock = tipc_sk(sock->sk);

	if (msg_errcode(msg)) {
		sock->state = SS_DISCONNECTING;
		return -ECONNREFUSED;
	}

	tsock->peer_name.ref = msg_origport(msg);
	tsock->peer_name.node = msg_orignode(msg);
	tipc_connect2port(tsock->p->ref, &tsock->peer_name);
	tipc_set_portimportance(tsock->p->ref, msg_importance(msg));
	sock->state = SS_CONNECTED;
	return 0;
}

/**
 * set_orig_addr - capture sender's address for received message
 * @m: descriptor for message info
 * @msg: received message header
 *
 * Note: Address is not captured if not requested by receiver.
 */

static void set_orig_addr(struct msghdr *m, struct tipc_msg *msg)
{
	struct sockaddr_tipc *addr = (struct sockaddr_tipc *)m->msg_name;

	if (addr) {
		addr->family = AF_TIPC;
		addr->addrtype = TIPC_ADDR_ID;
		addr->addr.id.ref = msg_origport(msg);
		addr->addr.id.node = msg_orignode(msg);
		addr->addr.name.domain = 0;	/* could leave uninitialized */
		addr->scope = 0;		/* could leave uninitialized */
		m->msg_namelen = sizeof(struct sockaddr_tipc);
	}
}

/**
 * anc_data_recv - optionally capture ancillary data for received message
 * @m: descriptor for message info
 * @msg: received message header
 * @tport: TIPC port associated with message
 *
 * Note: Ancillary data is not captured if not requested by receiver.
 *
 * Returns 0 if successful, otherwise errno
 */

static int anc_data_recv(struct msghdr *m, struct tipc_msg *msg,
				struct tipc_port *tport)
{
	u32 anc_data[3];
	u32 err;
	u32 dest_type;
	int has_name;
	int res;

	if (likely(m->msg_controllen == 0))
		return 0;

	/* Optionally capture errored message object(s) */

	err = msg ? msg_errcode(msg) : 0;
	if (unlikely(err)) {
		anc_data[0] = err;
		anc_data[1] = msg_data_sz(msg);
		res = put_cmsg(m, SOL_TIPC, TIPC_ERRINFO, 8, anc_data);
		if (res)
			return res;
		if (anc_data[1]) {
			res = put_cmsg(m, SOL_TIPC, TIPC_RETDATA, anc_data[1],
				       msg_data(msg));
			if (res)
				return res;
		}
	}

	/* Optionally capture message destination object */

	dest_type = msg ? msg_type(msg) : TIPC_DIRECT_MSG;
	switch (dest_type) {
	case TIPC_NAMED_MSG:
		has_name = 1;
		anc_data[0] = msg_nametype(msg);
		anc_data[1] = msg_namelower(msg);
		anc_data[2] = msg_namelower(msg);
		break;
	case TIPC_MCAST_MSG:
		has_name = 1;
		anc_data[0] = msg_nametype(msg);
		anc_data[1] = msg_namelower(msg);
		anc_data[2] = msg_nameupper(msg);
		break;
	case TIPC_CONN_MSG:
		has_name = (tport->conn_type != 0);
		anc_data[0] = tport->conn_type;
		anc_data[1] = tport->conn_instance;
		anc_data[2] = tport->conn_instance;
		break;
	default:
		has_name = 0;
	}
	if (has_name) {
		res = put_cmsg(m, SOL_TIPC, TIPC_DESTNAME, 12, anc_data);
		if (res)
			return res;
	}

	return 0;
}

/**
 * recv_msg - receive packet-oriented message
 * @iocb: (unused)
 * @m: descriptor for message info
 * @buf_len: total size of user buffer area
 * @flags: receive flags
 *
 * Used for SOCK_DGRAM, SOCK_RDM, and SOCK_SEQPACKET messages.
 * If the complete message doesn't fit in user area, truncate it.
 *
 * Returns size of returned message data, errno otherwise
 */

static int recv_msg(struct kiocb *iocb, struct socket *sock,
		    struct msghdr *m, size_t buf_len, int flags)
{
	struct sock *sk = sock->sk;
	struct tipc_port *tport = tipc_sk_port(sk);
	struct sk_buff *buf;
	struct tipc_msg *msg;
	long timeout;
	unsigned int sz;
	u32 err;
	int res;

	/* Catch invalid receive requests */

	if (unlikely(!buf_len))
		return -EINVAL;

	lock_sock(sk);

	if (unlikely(sock->state == SS_UNCONNECTED)) {
		res = -ENOTCONN;
		goto exit;
	}

	timeout = 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);
	}

	/* Look at first message in receive queue */

	buf = skb_peek(&sk->sk_receive_queue);
	msg = buf_msg(buf);
	sz = msg_data_sz(msg);
	err = msg_errcode(msg);

	/* Complete connection setup for an implied connect */

	if (unlikely(sock->state == SS_CONNECTING)) {
		res = auto_connect(sock, msg);
		if (res)
			goto exit;
	}

	/* Discard an empty non-errored message & try again */

	if ((!sz) && (!err)) {
		advance_rx_queue(sk);
		goto restart;
	}

	/* Capture sender's address (optional) */

	set_orig_addr(m, msg);

	/* Capture ancillary data (optional) */

	res = anc_data_recv(m, msg, tport);