tipc: Fix race condition when creating socket or native port

[safe/jmp/linux-2.6] / net / tipc / socket.c

/*
 * net/tipc/socket.c: TIPC socket API
 *
 * Copyright (c) 2001-2007, Ericsson AB
 * Copyright (c) 2004-2007, 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/module.h>
#include <linux/types.h>
#include <linux/net.h>
#include <linux/socket.h>
#include <linux/errno.h>
#include <linux/mm.h>
#include <linux/slab.h>
#include <linux/poll.h>
#include <linux/fcntl.h>
#include <asm/string.h>
#include <asm/atomic.h>
#include <net/sock.h>

#include <linux/tipc.h>
#include <linux/tipc_config.h>
#include <net/tipc/tipc_msg.h>
#include <net/tipc/tipc_port.h>

#include "core.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;
};

#define tipc_sk(sk) ((struct tipc_sock *)(sk))
#define tipc_sk_port(sk) ((struct tipc_port *)(tipc_sk(sk)->p))

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 = 0;

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)
 *
 * 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)
{
        const struct proto_ops *ops;
        socket_state state;
        struct sock *sk;
        struct tipc_port *tp_ptr;
        u32 portref;

        /* Validate arguments */

        if (net != &init_net)
                return -EAFNOSUPPORT;

        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 */

        portref = tipc_createport_raw(sk, &dispatch, &wakeupdispatch,
                                      TIPC_LOW_IMPORTANCE, &tp_ptr);
        if (unlikely(portref == 0)) {
                sk_free(sk);
                return -ENOMEM;
        }

        /* Finish initializing socket data structures */

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

        sock_init_data(sock, sk);
        sk->sk_rcvtimeo = msecs_to_jiffies(CONN_TIMEOUT_DEFAULT);
        sk->sk_backlog_rcv = backlog_rcv;
        tipc_sk(sk)->p = tipc_get_port(portref);

        spin_unlock_bh(tp_ptr->lock);

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

        atomic_inc(&tipc_user_count);
        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 != msg_data(buf_msg(buf)))
                        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;

        atomic_dec(&tipc_user_count);
        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;

        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 to obtain socket name, 1 to obtain peer socket name
 *
 * 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 get_name(struct socket *sock, struct sockaddr *uaddr,
                    int *uaddr_len, int peer)
{
        struct sockaddr_tipc *addr = (struct sockaddr_tipc *)uaddr;
        u32 portref = tipc_sk_port(sock->sk)->ref;
        u32 res;

        if (peer) {
                res = tipc_peer(portref, &addr->addr.id);
                if (res)
                        return res;
        } else {
                tipc_ownidentity(portref, &addr->addr.id);
        }

        *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:
 * a) POLLRDNORM and POLLIN are set if the socket's receive queue is non-empty
 *    or if a connection-oriented socket is does not have an active connection
 *    (i.e. a read operation will not block).
 * b) POLLOUT is set except when a socket's connection has been terminated
 *    (i.e. a write operation will not block).
 * c) POLLHUP is set when a socket's connection has been terminated.
 *
 * IMPORTANT: The fact that a read or write operation will not block does NOT
 * imply that the operation will succeed!
 */

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

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

        if (!skb_queue_empty(&sk->sk_receive_queue) ||
            (sock->state == SS_UNCONNECTED) ||
            (sock->state == SS_DISCONNECTING))
                mask = (POLLRDNORM | POLLIN);
        else
                mask = 0;

        if (sock->state == SS_DISCONNECTING)
                mask |= POLLHUP;
        else
                mask |= POLLOUT;

        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 (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;
        int res = -EINVAL;

        if (unlikely(!dest))
                return -EDESTADDRREQ;
        if (unlikely((m->msg_namelen < sizeof(*dest)) ||
                     (dest->family != AF_TIPC)))
                return -EINVAL;

        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);
        }

        do {
                if (dest->addrtype == TIPC_ADDR_NAME) {
                        if ((res = dest_name_check(dest, m)))
                                break;
                        res = tipc_send2name(tport->ref,
                                             &dest->addr.name.name,
                                             dest->addr.name.domain,
                                             m->msg_iovlen,
                                             m->msg_iov);
                }
                else if (dest->addrtype == TIPC_ADDR_ID) {
                        res = tipc_send2port(tport->ref,
                                             &dest->addr.id,
                                             m->msg_iovlen,
                                             m->msg_iov);
                }
                else if (dest->addrtype == TIPC_ADDR_MCAST) {
                        if (needs_conn) {
                                res = -EOPNOTSUPP;
                                break;
                        }
                        if ((res = dest_name_check(dest, m)))
                                break;
                        res = tipc_multicast(tport->ref,
                                             &dest->addr.nameseq,
                                             0,
                                             m->msg_iovlen,
                                             m->msg_iov);
                }
                if (likely(res != -ELINKCONG)) {
                        if (needs_conn && (res >= 0)) {
                                sock->state = SS_CONNECTING;
                        }
                        break;
                }
                if (m->msg_flags & MSG_DONTWAIT) {
                        res = -EWOULDBLOCK;
                        break;
                }
                release_sock(sk);
                res = wait_event_interruptible(*sk->sk_sleep,
                                               !tport->congested);
                lock_sock(sk);
                if (res)
                        break;
        } 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;
        int res;

        /* Handle implied connection establishment */

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

        if (iocb)
                lock_sock(sk);

        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);
                if (likely(res != -ELINKCONG)) {
                        break;
                }
                if (m->msg_flags & MSG_DONTWAIT) {
                        res = -EWOULDBLOCK;
                        break;
                }
                release_sock(sk);
                res = wait_event_interruptible(*sk->sk_sleep,
                        (!tport->congested || !tport->connected));
                lock_sock(sk);
                if (res)
                        break;
        } 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;
        }

        /*
         * 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;
                        if ((res = send_packet(NULL, sock, &my_msg, 0)) < 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_port *tport = tipc_sk_port(sock->sk);
        struct tipc_portid peer;

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

        peer.ref = msg_origport(msg);
        peer.node = msg_orignode(msg);
        tipc_connect2port(tport->ref, &peer);
        tipc_set_portimportance(tport->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);
                if ((res = put_cmsg(m, SOL_TIPC, TIPC_ERRINFO, 8, anc_data)))
                        return res;
                if (anc_data[1] &&
                    (res = put_cmsg(m, SOL_TIPC, TIPC_RETDATA, anc_data[1],
                                    msg_data(msg))))
                        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)))
                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;
        unsigned int sz;
        u32 err;
        int res;

        /* Catch invalid receive requests */

        if (m->msg_iovlen != 1)
                return -EOPNOTSUPP;   /* Don't do multiple iovec entries yet */

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

        lock_sock(sk);

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

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 (flags & MSG_DONTWAIT) {
                        res = -EWOULDBLOCK;
                        goto exit;
                }
                release_sock(sk);
                res = wait_event_interruptible(*sk->sk_sleep,
                        (!skb_queue_empty(&sk->sk_receive_queue) ||
                         (sock->state == SS_DISCONNECTING)));
                lock_sock(sk);
                if (res)
                        goto exit;
        }

        /* 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);
        if (res)
                goto exit;

        /* Capture message data (if valid) & compute return value (always) */

        if (!err) {
                if (unlikely(buf_len < sz)) {
                        sz = buf_len;
                        m->msg_flags |= MSG_TRUNC;
                }
                if (unlikely(copy_to_user(m->msg_iov->iov_base, msg_data(msg),
                                          sz))) {
                        res = -EFAULT;
                        goto exit;
                }
                res = sz;
        } else {
                if ((sock->state == SS_READY) ||
                    ((err == TIPC_CONN_SHUTDOWN) || m->msg_control))
                        res = 0;
                else
                        res = -ECONNRESET;
        }

        /* Consume received message (optional) */

        if (likely(!(flags & MSG_PEEK))) {
                if ((sock->state != SS_READY) &&
                    (++tport->conn_unacked >= TIPC_FLOW_CONTROL_WIN))
                        tipc_acknowledge(tport->ref, tport->conn_unacked);
                advance_rx_queue(sk);
        }
exit:
        release_sock(sk);
        return res;
}

/**
 * recv_stream - receive stream-oriented data
 * @iocb: (unused)
 * @m: descriptor for message info
 * @buf_len: total size of user buffer area
 * @flags: receive flags
 *
 * Used for SOCK_STREAM messages only.  If not enough data is available
 * will optionally wait for more; never truncates data.
 *
 * Returns size of returned message data, errno otherwise
 */

static int recv_stream(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;
        unsigned int sz;
        int sz_to_copy;
        int sz_copied = 0;
        int needed;
        char __user *crs = m->msg_iov->iov_base;
        unsigned char *buf_crs;
        u32 err;
        int res = 0;

        /* Catch invalid receive attempts */

        if (m->msg_iovlen != 1)
                return -EOPNOTSUPP;   /* Don't do multiple iovec entries yet */

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

        lock_sock(sk);

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

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 (flags & MSG_DONTWAIT) {
                        res = -EWOULDBLOCK;
                        goto exit;
                }
                release_sock(sk);
                res = wait_event_interruptible(*sk->sk_sleep,
                        (!skb_queue_empty(&sk->sk_receive_queue) ||
                         (sock->state == SS_DISCONNECTING)));
                lock_sock(sk);
                if (res)
                        goto exit;
        }

        /* 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);

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

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

        /* Optionally capture sender's address & ancillary data of first msg */

        if (sz_copied == 0) {
                set_orig_addr(m, msg);
                res = anc_data_recv(m, msg, tport);
                if (res)
                        goto exit;
        }

        /* Capture message data (if valid) & compute return value (always) */

        if (!err) {
                buf_crs = (unsigned char *)(TIPC_SKB_CB(buf)->handle);
                sz = (unsigned char *)msg + msg_size(msg) - buf_crs;

                needed = (buf_len - sz_copied);
                sz_to_copy = (sz <= needed) ? sz : needed;
                if (unlikely(copy_to_user(crs, buf_crs, sz_to_copy))) {
                        res = -EFAULT;
                        goto exit;
                }
                sz_copied += sz_to_copy;

                if (sz_to_copy < sz) {
                        if (!(flags & MSG_PEEK))
                                TIPC_SKB_CB(buf)->handle = buf_crs + sz_to_copy;
                        goto exit;
                }

                crs += sz_to_copy;
        } else {
                if (sz_copied != 0)
                        goto exit; /* can't add error msg to valid data */

                if ((err == TIPC_CONN_SHUTDOWN) || m->msg_control)
                        res = 0;
                else
                        res = -ECONNRESET;
        }

        /* Consume received message (optional) */

        if (likely(!(flags & MSG_PEEK))) {
                if (unlikely(++tport->conn_unacked >= TIPC_FLOW_CONTROL_WIN))
                        tipc_acknowledge(tport->ref, tport->conn_unacked);
                advance_rx_queue(sk);
        }

        /* Loop around if more data is required */

        if ((sz_copied < buf_len)    /* didn't get all requested data */
            && (!skb_queue_empty(&sock->sk->sk_receive_queue) ||
                (flags & MSG_WAITALL))
                                     /* ... and more is ready or required */
            && (!(flags & MSG_PEEK)) /* ... and aren't just peeking at data */
            && (!err)                /* ... and haven't reached a FIN */
            )
                goto restart;

exit:
        release_sock(sk);
        return sz_copied ? sz_copied : res;
}

/**
 * rx_queue_full - determine if receive queue can accept another message
 * @msg: message to be added to queue
 * @queue_size: current size of queue
 * @base: nominal maximum size of queue
 *
 * Returns 1 if queue is unable to accept message, 0 otherwise
 */

static int rx_queue_full(struct tipc_msg *msg, u32 queue_size, u32 base)
{
        u32 threshold;
        u32 imp = msg_importance(msg);

        if (imp == TIPC_LOW_IMPORTANCE)
                threshold = base;
        else if (imp == TIPC_MEDIUM_IMPORTANCE)
                threshold = base * 2;
        else if (imp == TIPC_HIGH_IMPORTANCE)
                threshold = base * 100;
        else
                return 0;

        if (msg_connected(msg))
                threshold *= 4;

        return (queue_size >= threshold);
}

/**
 * filter_rcv - validate incoming message
 * @sk: socket
 * @buf: message
 *
 * Enqueues message on receive queue if acceptable; optionally handles
 * disconnect indication for a connected socket.
 *
 * Called with socket lock already taken; port lock may also be taken.
 *
 * Returns TIPC error status code (TIPC_OK if message is not to be rejected)
 */

static u32 filter_rcv(struct sock *sk, struct sk_buff *buf)
{
        struct socket *sock = sk->sk_socket;
        struct tipc_msg *msg = buf_msg(buf);
        u32 recv_q_len;

        /* Reject message if it is wrong sort of message for socket */

        /*
         * WOULD IT BE BETTER TO JUST DISCARD THESE MESSAGES INSTEAD?
         * "NO PORT" ISN'T REALLY THE RIGHT ERROR CODE, AND THERE MAY
         * BE SECURITY IMPLICATIONS INHERENT IN REJECTING INVALID TRAFFIC
         */

        if (sock->state == SS_READY) {
                if (msg_connected(msg)) {
                        msg_dbg(msg, "dispatch filter 1\n");
                        return TIPC_ERR_NO_PORT;
                }
        } else {
                if (msg_mcast(msg)) {
                        msg_dbg(msg, "dispatch filter 2\n");
                        return TIPC_ERR_NO_PORT;
                }
                if (sock->state == SS_CONNECTED) {
                        if (!msg_connected(msg)) {
                                msg_dbg(msg, "dispatch filter 3\n");
                                return TIPC_ERR_NO_PORT;
                        }
                }
                else if (sock->state == SS_CONNECTING) {
                        if (!msg_connected(msg) && (msg_errcode(msg) == 0)) {
                                msg_dbg(msg, "dispatch filter 4\n");
                                return TIPC_ERR_NO_PORT;
                        }
                }
                else if (sock->state == SS_LISTENING) {
                        if (msg_connected(msg) || msg_errcode(msg)) {
                                msg_dbg(msg, "dispatch filter 5\n");
                                return TIPC_ERR_NO_PORT;
                        }
                }
                else if (sock->state == SS_DISCONNECTING) {
                        msg_dbg(msg, "dispatch filter 6\n");
                        return TIPC_ERR_NO_PORT;
                }
                else /* (sock->state == SS_UNCONNECTED) */ {
                        if (msg_connected(msg) || msg_errcode(msg)) {
                                msg_dbg(msg, "dispatch filter 7\n");
                                return TIPC_ERR_NO_PORT;
                        }
                }
        }

        /* Reject message if there isn't room to queue it */

        recv_q_len = (u32)atomic_read(&tipc_queue_size);
        if (unlikely(recv_q_len >= OVERLOAD_LIMIT_BASE)) {
                if (rx_queue_full(msg, recv_q_len, OVERLOAD_LIMIT_BASE))
                        return TIPC_ERR_OVERLOAD;
        }
        recv_q_len = skb_queue_len(&sk->sk_receive_queue);
        if (unlikely(recv_q_len >= (OVERLOAD_LIMIT_BASE / 2))) {
                if (rx_queue_full(msg, recv_q_len, OVERLOAD_LIMIT_BASE / 2))
                        return TIPC_ERR_OVERLOAD;
        }

        /* Enqueue message (finally!) */

        msg_dbg(msg, "<DISP<: ");
        TIPC_SKB_CB(buf)->handle = msg_data(msg);
        atomic_inc(&tipc_queue_size);
        __skb_queue_tail(&sk->sk_receive_queue, buf);

        /* Initiate connection termination for an incoming 'FIN' */

        if (unlikely(msg_errcode(msg) && (sock->state == SS_CONNECTED))) {
                sock->state = SS_DISCONNECTING;
                tipc_disconnect_port(tipc_sk_port(sk));
        }

        if (waitqueue_active(sk->sk_sleep))
                wake_up_interruptible(sk->sk_sleep);
        return TIPC_OK;
}

/**
 * backlog_rcv - handle incoming message from backlog queue
 * @sk: socket
 * @buf: message
 *
 * Caller must hold socket lock, but not port lock.
 *
 * Returns 0
 */

static int backlog_rcv(struct sock *sk, struct sk_buff *buf)
{
        u32 res;

        res = filter_rcv(sk, buf);
        if (res)
                tipc_reject_msg(buf, res);
        return 0;
}

/**
 * dispatch - handle incoming message
 * @tport: TIPC port that received message
 * @buf: message
 *
 * Called with port lock already taken.
 *
 * Returns TIPC error status code (TIPC_OK if message is not to be rejected)
 */

static u32 dispatch(struct tipc_port *tport, struct sk_buff *buf)
{
        struct sock *sk = (struct sock *)tport->usr_handle;
        u32 res;

        /*
         * Process message if socket is unlocked; otherwise add to backlog queue
         *
         * This code is based on sk_receive_skb(), but must be distinct from it
         * since a TIPC-specific filter/reject mechanism is utilized
         */

        bh_lock_sock(sk);
        if (!sock_owned_by_user(sk)) {
                res = filter_rcv(sk, buf);
        } else {
                sk_add_backlog(sk, buf);
                res = TIPC_OK;
        }
        bh_unlock_sock(sk);

        return res;
}

/**
 * wakeupdispatch - wake up port after congestion
 * @tport: port to wakeup
 *
 * Called with port lock already taken.
 */

static void wakeupdispatch(struct tipc_port *tport)
{
        struct sock *sk = (struct sock *)tport->usr_handle;

        if (waitqueue_active(sk->sk_sleep))
                wake_up_interruptible(sk->sk_sleep);
}

/**
 * connect - establish a connection to another TIPC port
 * @sock: socket structure
 * @dest: socket address for destination port
 * @destlen: size of socket address data structure
 * @flags: file-related flags associated with socket
 *
 * Returns 0 on success, errno otherwise
 */

static int connect(struct socket *sock, struct sockaddr *dest, int destlen,
                   int flags)
{
        struct sock *sk = sock->sk;
        struct sockaddr_tipc *dst = (struct sockaddr_tipc *)dest;
        struct msghdr m = {NULL,};
        struct sk_buff *buf;
        struct tipc_msg *msg;
        int res;

        lock_sock(sk);

        /* For now, TIPC does not allow use of connect() with DGRAM/RDM types */

        if (sock->state == SS_READY) {
                res = -EOPNOTSUPP;
                goto exit;
        }

        /* For now, TIPC does not support the non-blocking form of connect() */

        if (flags & O_NONBLOCK) {
                res = -EWOULDBLOCK;
                goto exit;
        }

        /* Issue Posix-compliant error code if socket is in the wrong state */

        if (sock->state == SS_LISTENING) {
                res = -EOPNOTSUPP;
                goto exit;
        }
        if (sock->state == SS_CONNECTING) {
                res = -EALREADY;
                goto exit;
        }
        if (sock->state != SS_UNCONNECTED) {
                res = -EISCONN;
                goto exit;
        }

        /*
         * Reject connection attempt using multicast address
         *
         * Note: send_msg() validates the rest of the address fields,
         *       so there's no need to do it here
         */

        if (dst->addrtype == TIPC_ADDR_MCAST) {
                res = -EINVAL;
                goto exit;
        }

        /* Reject any messages already in receive queue (very unlikely) */

        reject_rx_queue(sk);

        /* Send a 'SYN-' to destination */

        m.msg_name = dest;
        m.msg_namelen = destlen;
        res = send_msg(NULL, sock, &m, 0);
        if (res < 0) {
                goto exit;
        }

        /* Wait until an 'ACK' or 'RST' arrives, or a timeout occurs */

        release_sock(sk);
        res = wait_event_interruptible_timeout(*sk->sk_sleep,
                        (!skb_queue_empty(&sk->sk_receive_queue) ||
                        (sock->state != SS_CONNECTING)),
                        sk->sk_rcvtimeo);
        lock_sock(sk);

        if (res > 0) {
                buf = skb_peek(&sk->sk_receive_queue);
                if (buf != NULL) {
                        msg = buf_msg(buf);
                        res = auto_connect(sock, msg);
                        if (!res) {
                                if (!msg_data_sz(msg))
                                        advance_rx_queue(sk);
                        }
                } else {
                        if (sock->state == SS_CONNECTED) {
                                res = -EISCONN;
                        } else {
                                res = -ECONNREFUSED;
                        }
                }
        } else {
                if (res == 0)
                        res = -ETIMEDOUT;
                else
                        ; /* leave "res" unchanged */
                sock->state = SS_DISCONNECTING;
        }

exit:
        release_sock(sk);
        return res;
}

/**
 * listen - allow socket to listen for incoming connections
 * @sock: socket structure
 * @len: (unused)
 *
 * Returns 0 on success, errno otherwise
 */

static int listen(struct socket *sock, int len)
{
        struct sock *sk = sock->sk;
        int res;

        lock_sock(sk);

        if (sock->state == SS_READY)
                res = -EOPNOTSUPP;
        else if (sock->state != SS_UNCONNECTED)
                res = -EINVAL;
        else {
                sock->state = SS_LISTENING;
                res = 0;
        }

        release_sock(sk);
        return res;
}

/**
 * accept - wait for connection request
 * @sock: listening socket
 * @newsock: new socket that is to be connected
 * @flags: file-related flags associated with socket
 *
 * Returns 0 on success, errno otherwise
 */

static int accept(struct socket *sock, struct socket *new_sock, int flags)
{
        struct sock *sk = sock->sk;
        struct sk_buff *buf;
        int res;

        lock_sock(sk);

        if (sock->state == SS_READY) {
                res = -EOPNOTSUPP;
                goto exit;
        }
        if (sock->state != SS_LISTENING) {
                res = -EINVAL;
                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->sk_sleep,
                                (!skb_queue_empty(&sk->sk_receive_queue)));
                lock_sock(sk);
                if (res)
                        goto exit;
        }

        buf = skb_peek(&sk->sk_receive_queue);

        res = tipc_create(sock_net(sock->sk), new_sock, 0);
        if (!res) {
                struct sock *new_sk = new_sock->sk;
                struct tipc_port *new_tport = tipc_sk_port(new_sk);
                u32 new_ref = new_tport->ref;
                struct tipc_portid id;
                struct tipc_msg *msg = buf_msg(buf);

                lock_sock(new_sk);

                /*
                 * Reject any stray messages received by new socket
                 * before the socket lock was taken (very, very unlikely)
                 */

                reject_rx_queue(new_sk);

                /* Connect new socket to it's peer */

                id.ref = msg_origport(msg);
                id.node = msg_orignode(msg);
                tipc_connect2port(new_ref, &id);
                new_sock->state = SS_CONNECTED;

                tipc_set_portimportance(new_ref, msg_importance(msg));
                if (msg_named(msg)) {
                        new_tport->conn_type = msg_nametype(msg);
                        new_tport->conn_instance = msg_nameinst(msg);
                }

                /*
                 * Respond to 'SYN-' by discarding it & returning 'ACK'-.
                 * Respond to 'SYN+' by queuing it on new socket.
                 */

                msg_dbg(msg,"<ACC<: ");
                if (!msg_data_sz(msg)) {
                        struct msghdr m = {NULL,};

                        advance_rx_queue(sk);
                        send_packet(NULL, new_sock, &m, 0);
                } else {
                        __skb_dequeue(&sk->sk_receive_queue);
                        __skb_queue_head(&new_sk->sk_receive_queue, buf);
                }
                release_sock(new_sk);
        }
exit:
        release_sock(sk);
        return res;
}

/**
 * shutdown - shutdown socket connection
 * @sock: socket structure
 * @how: direction to close (must be SHUT_RDWR)
 *
 * Terminates connection (if necessary), then purges socket's receive queue.
 *
 * Returns 0 on success, errno otherwise
 */

static int shutdown(struct socket *sock, int how)
{
        struct sock *sk = sock->sk;
        struct tipc_port *tport = tipc_sk_port(sk);
        struct sk_buff *buf;
        int res;

        if (how != SHUT_RDWR)
                return -EINVAL;

        lock_sock(sk);

        switch (sock->state) {
        case SS_CONNECTING:
        case SS_CONNECTED:

                /* Disconnect and send a 'FIN+' or 'FIN-' message to peer */
restart:
                buf = __skb_dequeue(&sk->sk_receive_queue);
                if (buf) {
                        atomic_dec(&tipc_queue_size);
                        if (TIPC_SKB_CB(buf)->handle != msg_data(buf_msg(buf))) {
                                buf_discard(buf);
                                goto restart;
                        }
                        tipc_disconnect(tport->ref);
                        tipc_reject_msg(buf, TIPC_CONN_SHUTDOWN);
                } else {
                        tipc_shutdown(tport->ref);
                }

                sock->state = SS_DISCONNECTING;

                /* fall through */

        case SS_DISCONNECTING:

                /* Discard any unreceived messages; wake up sleeping tasks */

                discard_rx_queue(sk);
                if (waitqueue_active(sk->sk_sleep))
                        wake_up_interruptible(sk->sk_sleep);
                res = 0;
                break;

        default:
                res = -ENOTCONN;
        }

        release_sock(sk);
        return res;
}

/**
 * setsockopt - set socket option
 * @sock: socket structure
 * @lvl: option level
 * @opt: option identifier
 * @ov: pointer to new option value
 * @ol: length of option value
 *
 * For stream sockets only, accepts and ignores all IPPROTO_TCP options
 * (to ease compatibility).
 *
 * Returns 0 on success, errno otherwise
 */

static int setsockopt(struct socket *sock,
                      int lvl, int opt, char __user *ov, int ol)
{
        struct sock *sk = sock->sk;
        struct tipc_port *tport = tipc_sk_port(sk);
        u32 value;
        int res;

        if ((lvl == IPPROTO_TCP) && (sock->type == SOCK_STREAM))
                return 0;
        if (lvl != SOL_TIPC)
                return -ENOPROTOOPT;
        if (ol < sizeof(value))
                return -EINVAL;
        if ((res = get_user(value, (u32 __user *)ov)))
                return res;

        lock_sock(sk);

        switch (opt) {
        case TIPC_IMPORTANCE:
                res = tipc_set_portimportance(tport->ref, value);
                break;
        case TIPC_SRC_DROPPABLE:
                if (sock->type != SOCK_STREAM)
                        res = tipc_set_portunreliable(tport->ref, value);
                else
                        res = -ENOPROTOOPT;
                break;
        case TIPC_DEST_DROPPABLE:
                res = tipc_set_portunreturnable(tport->ref, value);
                break;
        case TIPC_CONN_TIMEOUT:
                sk->sk_rcvtimeo = msecs_to_jiffies(value);
                /* no need to set "res", since already 0 at this point */
                break;
        default:
                res = -EINVAL;
        }

        release_sock(sk);

        return res;
}

/**
 * getsockopt - get socket option
 * @sock: socket structure
 * @lvl: option level
 * @opt: option identifier
 * @ov: receptacle for option value
 * @ol: receptacle for length of option value
 *
 * For stream sockets only, returns 0 length result for all IPPROTO_TCP options
 * (to ease compatibility).
 *
 * Returns 0 on success, errno otherwise
 */

static int getsockopt(struct socket *sock,
                      int lvl, int opt, char __user *ov, int __user *ol)
{
        struct sock *sk = sock->sk;
        struct tipc_port *tport = tipc_sk_port(sk);
        int len;
        u32 value;
        int res;

        if ((lvl == IPPROTO_TCP) && (sock->type == SOCK_STREAM))
                return put_user(0, ol);
        if (lvl != SOL_TIPC)
                return -ENOPROTOOPT;
        if ((res = get_user(len, ol)))
                return res;

        lock_sock(sk);

        switch (opt) {
        case TIPC_IMPORTANCE:
                res = tipc_portimportance(tport->ref, &value);
                break;
        case TIPC_SRC_DROPPABLE:
                res = tipc_portunreliable(tport->ref, &value);
                break;
        case TIPC_DEST_DROPPABLE:
                res = tipc_portunreturnable(tport->ref, &value);
                break;
        case TIPC_CONN_TIMEOUT:
                value = jiffies_to_msecs(sk->sk_rcvtimeo);
                /* no need to set "res", since already 0 at this point */
                break;
        default:
                res = -EINVAL;
        }

        release_sock(sk);

        if (res) {
                /* "get" failed */
        }
        else if (len < sizeof(value)) {
                res = -EINVAL;
        }
        else if (copy_to_user(ov, &value, sizeof(value))) {
                res = -EFAULT;
        }
        else {
                res = put_user(sizeof(value), ol);
        }

        return res;
}

/**
 * Protocol switches for the various types of TIPC sockets
 */

static const struct proto_ops msg_ops = {
        .owner          = THIS_MODULE,
        .family         = AF_TIPC,
        .release        = release,
        .bind           = bind,
        .connect        = connect,
        .socketpair     = sock_no_socketpair,
        .accept         = accept,
        .getname        = get_name,
        .poll           = poll,
        .ioctl          = sock_no_ioctl,
        .listen         = listen,
        .shutdown       = shutdown,
        .setsockopt     = setsockopt,
        .getsockopt     = getsockopt,
        .sendmsg        = send_msg,
        .recvmsg        = recv_msg,
        .mmap           = sock_no_mmap,
        .sendpage       = sock_no_sendpage
};

static const struct proto_ops packet_ops = {
        .owner          = THIS_MODULE,
        .family         = AF_TIPC,
        .release        = release,
        .bind           = bind,
        .connect        = connect,
        .socketpair     = sock_no_socketpair,
        .accept         = accept,
        .getname        = get_name,
        .poll           = poll,
        .ioctl          = sock_no_ioctl,
        .listen         = listen,
        .shutdown       = shutdown,
        .setsockopt     = setsockopt,
        .getsockopt     = getsockopt,
        .sendmsg        = send_packet,
        .recvmsg        = recv_msg,
        .mmap           = sock_no_mmap,
        .sendpage       = sock_no_sendpage
};

static const struct proto_ops stream_ops = {
        .owner          = THIS_MODULE,
        .family         = AF_TIPC,
        .release        = release,
        .bind           = bind,
        .connect        = connect,
        .socketpair     = sock_no_socketpair,
        .accept         = accept,
        .getname        = get_name,
        .poll           = poll,
        .ioctl          = sock_no_ioctl,
        .listen         = listen,
        .shutdown       = shutdown,
        .setsockopt     = setsockopt,
        .getsockopt     = getsockopt,
        .sendmsg        = send_stream,
        .recvmsg        = recv_stream,
        .mmap           = sock_no_mmap,
        .sendpage       = sock_no_sendpage
};

static const struct net_proto_family tipc_family_ops = {
        .owner          = THIS_MODULE,
        .family         = AF_TIPC,
        .create         = tipc_create
};

static struct proto tipc_proto = {
        .name           = "TIPC",
        .owner          = THIS_MODULE,
        .obj_size       = sizeof(struct tipc_sock)
};

/**
 * tipc_socket_init - initialize TIPC socket interface
 *
 * Returns 0 on success, errno otherwise
 */
int tipc_socket_init(void)
{
        int res;

        res = proto_register(&tipc_proto, 1);
        if (res) {
                err("Failed to register TIPC protocol type\n");
                goto out;
        }

        res = sock_register(&tipc_family_ops);
        if (res) {
                err("Failed to register TIPC socket type\n");
                proto_unregister(&tipc_proto);
                goto out;
        }

        sockets_enabled = 1;
 out:
        return res;
}

/**
 * tipc_socket_stop - stop TIPC socket interface
 */

void tipc_socket_stop(void)
{
        if (!sockets_enabled)
                return;

        sockets_enabled = 0;
        sock_unregister(tipc_family_ops.family);
        proto_unregister(&tipc_proto);
}