diff options
Diffstat (limited to 'net/tipc/socket.c')
| -rw-r--r-- | net/tipc/socket.c | 1598 |
1 files changed, 876 insertions, 722 deletions
diff --git a/net/tipc/socket.c b/net/tipc/socket.c index cd0bb77f267..ef0475568f9 100644 --- a/net/tipc/socket.c +++ b/net/tipc/socket.c @@ -1,8 +1,8 @@ /* * net/tipc/socket.c: TIPC socket API * - * Copyright (c) 2001-2007, Ericsson AB - * Copyright (c) 2004-2008, Wind River Systems + * Copyright (c) 2001-2007, 2012-2014, Ericsson AB + * Copyright (c) 2004-2008, 2010-2013, Wind River Systems * All rights reserved. * * Redistribution and use in source and binary forms, with or without @@ -34,54 +34,29 @@ * 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/poll.h> -#include <linux/fcntl.h> -#include <linux/gfp.h> -#include <asm/string.h> -#include <asm/atomic.h> -#include <net/sock.h> - -#include <linux/tipc.h> -#include <linux/tipc_config.h> - #include "core.h" #include "port.h" +#include "node.h" + +#include <linux/export.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; - long conn_timeout; -}; - -#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 int tipc_backlog_rcv(struct sock *sk, struct sk_buff *skb); +static void tipc_data_ready(struct sock *sk); +static void tipc_write_space(struct sock *sk); +static int tipc_release(struct socket *sock); +static int tipc_accept(struct socket *sock, struct socket *new_sock, int flags); 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); +static struct proto tipc_proto_kern; /* * Revised TIPC socket locking policy: @@ -128,32 +103,16 @@ static atomic_t tipc_queue_size = ATOMIC_INIT(0); * - port reference */ +#include "socket.h" + /** * 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); - } + kfree_skb(__skb_dequeue(&sk->sk_receive_queue)); } /** @@ -161,19 +120,16 @@ static void discard_rx_queue(struct sock *sk) * * 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))) { + 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 + * tipc_sk_create - create a TIPC socket * @net: network namespace (must be default network) * @sock: pre-allocated socket structure * @protocol: protocol indicator (must be 0) @@ -184,20 +140,17 @@ static void reject_rx_queue(struct sock *sk) * * Returns 0 on success, errno otherwise */ - -static int tipc_create(struct net *net, struct socket *sock, int protocol, - int kern) +static int tipc_sk_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; + struct tipc_sock *tsk; + struct tipc_port *port; + u32 ref; /* Validate arguments */ - - if (!net_eq(net, &init_net)) - return -EAFNOSUPPORT; - if (unlikely(protocol != 0)) return -EPROTONOSUPPORT; @@ -220,44 +173,116 @@ static int tipc_create(struct net *net, struct socket *sock, int protocol, } /* Allocate socket's protocol area */ + if (!kern) + sk = sk_alloc(net, AF_TIPC, GFP_KERNEL, &tipc_proto); + else + sk = sk_alloc(net, AF_TIPC, GFP_KERNEL, &tipc_proto_kern); - sk = sk_alloc(net, AF_TIPC, GFP_KERNEL, &tipc_proto); if (sk == NULL) return -ENOMEM; - /* Allocate TIPC port for socket to use */ + tsk = tipc_sk(sk); + port = &tsk->port; - tp_ptr = tipc_createport_raw(sk, &dispatch, &wakeupdispatch, - TIPC_LOW_IMPORTANCE); - if (unlikely(!tp_ptr)) { + ref = tipc_port_init(port, TIPC_LOW_IMPORTANCE); + if (!ref) { + pr_warn("Socket registration failed, ref. table exhausted\n"); 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 = msecs_to_jiffies(CONN_TIMEOUT_DEFAULT); - - spin_unlock_bh(tp_ptr->lock); + sk->sk_backlog_rcv = tipc_backlog_rcv; + sk->sk_rcvbuf = sysctl_tipc_rmem[1]; + sk->sk_data_ready = tipc_data_ready; + sk->sk_write_space = tipc_write_space; + tsk->conn_timeout = CONN_TIMEOUT_DEFAULT; + atomic_set(&tsk->dupl_rcvcnt, 0); + tipc_port_unlock(port); if (sock->state == SS_READY) { - tipc_set_portunreturnable(tp_ptr->ref, 1); + tipc_port_set_unreturnable(port, true); if (sock->type == SOCK_DGRAM) - tipc_set_portunreliable(tp_ptr->ref, 1); + tipc_port_set_unreliable(port, true); + } + return 0; +} + +/** + * tipc_sock_create_local - create TIPC socket from inside TIPC module + * @type: socket type - SOCK_RDM or SOCK_SEQPACKET + * + * We cannot use sock_creat_kern here because it bumps module user count. + * Since socket owner and creator is the same module we must make sure + * that module count remains zero for module local sockets, otherwise + * we cannot do rmmod. + * + * Returns 0 on success, errno otherwise + */ +int tipc_sock_create_local(int type, struct socket **res) +{ + int rc; + + rc = sock_create_lite(AF_TIPC, type, 0, res); + if (rc < 0) { + pr_err("Failed to create kernel socket\n"); + return rc; } + tipc_sk_create(&init_net, *res, 0, 1); - atomic_inc(&tipc_user_count); return 0; } /** - * release - destroy a TIPC socket + * tipc_sock_release_local - release socket created by tipc_sock_create_local + * @sock: the socket to be released. + * + * Module reference count is not incremented when such sockets are created, + * so we must keep it from being decremented when they are released. + */ +void tipc_sock_release_local(struct socket *sock) +{ + tipc_release(sock); + sock->ops = NULL; + sock_release(sock); +} + +/** + * tipc_sock_accept_local - accept a connection on a socket created + * with tipc_sock_create_local. Use this function to avoid that + * module reference count is inadvertently incremented. + * + * @sock: the accepting socket + * @newsock: reference to the new socket to be created + * @flags: socket flags + */ + +int tipc_sock_accept_local(struct socket *sock, struct socket **newsock, + int flags) +{ + struct sock *sk = sock->sk; + int ret; + + ret = sock_create_lite(sk->sk_family, sk->sk_type, + sk->sk_protocol, newsock); + if (ret < 0) + return ret; + + ret = tipc_accept(sock, *newsock, flags); + if (ret < 0) { + sock_release(*newsock); + return ret; + } + (*newsock)->ops = sock->ops; + return ret; +} + +/** + * tipc_release - destroy a TIPC socket * @sock: socket to destroy * * This routine cleans up any messages that are still queued on the socket. @@ -272,72 +297,64 @@ static int tipc_create(struct net *net, struct socket *sock, int protocol, * * Returns 0 on success, errno otherwise */ - -static int release(struct socket *sock) +static int tipc_release(struct socket *sock) { struct sock *sk = sock->sk; - struct tipc_port *tport; + struct tipc_sock *tsk; + struct tipc_port *port; 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); + tsk = tipc_sk(sk); + port = &tsk->port; 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); + if (TIPC_SKB_CB(buf)->handle != NULL) + kfree_skb(buf); else { if ((sock->state == SS_CONNECTING) || (sock->state == SS_CONNECTED)) { sock->state = SS_DISCONNECTING; - tipc_disconnect(tport->ref); + tipc_port_disconnect(port->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) + /* Destroy TIPC port; also disconnects an active connection and + * sends a 'FIN-' to peer. */ - - res = tipc_deleteport(tport->ref); + tipc_port_destroy(port); /* Discard any remaining (connection-based) messages in receive queue */ - - discard_rx_queue(sk); + __skb_queue_purge(&sk->sk_receive_queue); /* 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; + return 0; } /** - * bind - associate or disassocate TIPC name(s) with a socket + * tipc_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 @@ -351,32 +368,53 @@ static int release(struct socket *sock) * 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) +static int tipc_bind(struct socket *sock, struct sockaddr *uaddr, + int uaddr_len) { + struct sock *sk = sock->sk; struct sockaddr_tipc *addr = (struct sockaddr_tipc *)uaddr; - u32 portref = tipc_sk_port(sock->sk)->ref; + struct tipc_sock *tsk = tipc_sk(sk); + int res = -EINVAL; - if (unlikely(!uaddr_len)) - return tipc_withdraw(portref, 0, NULL); + lock_sock(sk); + if (unlikely(!uaddr_len)) { + res = tipc_withdraw(&tsk->port, 0, NULL); + goto exit; + } - if (uaddr_len < sizeof(struct sockaddr_tipc)) - return -EINVAL; - if (addr->family != AF_TIPC) - return -EAFNOSUPPORT; + if (uaddr_len < sizeof(struct sockaddr_tipc)) { + res = -EINVAL; + goto exit; + } + if (addr->family != AF_TIPC) { + res = -EAFNOSUPPORT; + goto exit; + } if (addr->addrtype == TIPC_ADDR_NAME) addr->addr.nameseq.upper = addr->addr.nameseq.lower; - else if (addr->addrtype != TIPC_ADDR_NAMESEQ) - return -EAFNOSUPPORT; + else if (addr->addrtype != TIPC_ADDR_NAMESEQ) { + res = -EAFNOSUPPORT; + goto exit; + } + + if ((addr->addr.nameseq.type < TIPC_RESERVED_TYPES) && + (addr->addr.nameseq.type != TIPC_TOP_SRV) && + (addr->addr.nameseq.type != TIPC_CFG_SRV)) { + res = -EACCES; + goto exit; + } - return (addr->scope > 0) ? - tipc_publish(portref, addr->scope, &addr->addr.nameseq) : - tipc_withdraw(portref, -addr->scope, &addr->addr.nameseq); + res = (addr->scope > 0) ? + tipc_publish(&tsk->port, addr->scope, &addr->addr.nameseq) : + tipc_withdraw(&tsk->port, -addr->scope, &addr->addr.nameseq); +exit: + release_sock(sk); + return res; } /** - * get_name - get port ID of socket or peer socket + * tipc_getname - 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 @@ -386,24 +424,23 @@ static int bind(struct socket *sock, struct sockaddr *uaddr, int uaddr_len) * * 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). + * a completely predictable manner). */ - -static int get_name(struct socket *sock, struct sockaddr *uaddr, - int *uaddr_len, int peer) +static int tipc_getname(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); + struct tipc_sock *tsk = 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; + addr->addr.id.ref = tipc_port_peerport(&tsk->port); + addr->addr.id.node = tipc_port_peernode(&tsk->port); } else { - addr->addr.id.ref = tsock->p->ref; + addr->addr.id.ref = tsk->port.ref; addr->addr.id.node = tipc_own_addr; } @@ -417,7 +454,7 @@ static int get_name(struct socket *sock, struct sockaddr *uaddr, } /** - * poll - read and possibly block on pollmask + * tipc_poll - read and possibly block on pollmask * @file: file structure associated with the socket * @sock: socket for which to calculate the poll bits * @wait: ??? @@ -435,7 +472,7 @@ static int get_name(struct socket *sock, struct sockaddr *uaddr, * socket state flags set * ------------ --------- * unconnected no read flags - * no write flags + * POLLOUT if port is not congested * * connecting POLLIN/POLLRDNORM if ACK/NACK in rx queue * no write flags @@ -456,19 +493,23 @@ static int get_name(struct socket *sock, struct sockaddr *uaddr, * 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) +static unsigned int tipc_poll(struct file *file, struct socket *sock, + poll_table *wait) { struct sock *sk = sock->sk; + struct tipc_sock *tsk = tipc_sk(sk); u32 mask = 0; - poll_wait(file, sk_sleep(sk), wait); + sock_poll_wait(file, sk_sleep(sk), wait); switch ((int)sock->state) { + case SS_UNCONNECTED: + if (!tsk->port.congested) + mask |= POLLOUT; + break; case SS_READY: case SS_CONNECTED: - if (!tipc_sk_port(sk)->congested) + if (!tsk->port.congested) mask |= POLLOUT; /* fall thru' */ case SS_CONNECTING: @@ -494,7 +535,6 @@ static unsigned int poll(struct file *file, struct socket *sock, * * 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; @@ -506,6 +546,8 @@ static int dest_name_check(struct sockaddr_tipc *dest, struct msghdr *m) 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))) @@ -514,8 +556,34 @@ 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_sock *tsk = tipc_sk(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, !tsk->port.congested); + finish_wait(sk_sleep(sk), &wait); + } while (!done); + return 0; +} + + /** - * send_msg - send message in connectionless manner + * tipc_sendmsg - send message in connectionless manner * @iocb: if NULL, indicates that socket lock is already held * @sock: socket structure * @m: message to send @@ -528,14 +596,15 @@ static int dest_name_check(struct sockaddr_tipc *dest, struct msghdr *m) * * 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) +static int tipc_sendmsg(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; + struct tipc_sock *tsk = tipc_sk(sk); + struct tipc_port *port = &tsk->port; + DECLARE_SOCKADDR(struct sockaddr_tipc *, dest, m->msg_name); int needs_conn; + long timeo; int res = -EINVAL; if (unlikely(!dest)) @@ -543,6 +612,8 @@ static int send_msg(struct kiocb *iocb, struct socket *sock, if (unlikely((m->msg_namelen < sizeof(*dest)) || (dest->family != AF_TIPC))) return -EINVAL; + if (total_len > TIPC_MAX_USER_MSG_SIZE) + return -EMSGSIZE; if (iocb) lock_sock(sk); @@ -557,63 +628,54 @@ static int send_msg(struct kiocb *iocb, struct socket *sock, res = -EISCONN; goto exit; } - if ((tport->published) || - ((sock->type == SOCK_STREAM) && (total_len != 0))) { + if (tsk->port.published) { 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; + tsk->port.conn_type = dest->addr.name.name.type; + tsk->port.conn_instance = dest->addr.name.name.instance; } /* Abort any pending connection attempts (very unlikely) */ - reject_rx_queue(sk); } + timeo = sock_sndtimeo(sk, m->msg_flags & MSG_DONTWAIT); do { if (dest->addrtype == TIPC_ADDR_NAME) { - if ((res = dest_name_check(dest, m))) + res = dest_name_check(dest, m); + if (res) break; - res = tipc_send2name(tport->ref, + res = tipc_send2name(port, &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, + m->msg_iov, + total_len); + } else if (dest->addrtype == TIPC_ADDR_ID) { + res = tipc_send2port(port, &dest->addr.id, - m->msg_iovlen, - m->msg_iov); - } - else if (dest->addrtype == TIPC_ADDR_MCAST) { + m->msg_iov, + total_len); + } else if (dest->addrtype == TIPC_ADDR_MCAST) { if (needs_conn) { res = -EOPNOTSUPP; break; } - if ((res = dest_name_check(dest, m))) + res = dest_name_check(dest, m); + if (res) break; - res = tipc_multicast(tport->ref, - &dest->addr.nameseq, - m->msg_iovlen, - m->msg_iov); + res = tipc_port_mcast_xmit(port, + &dest->addr.nameseq, + m->msg_iov, + total_len); } if (likely(res != -ELINKCONG)) { - if (needs_conn && (res >= 0)) { + 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_sleep(sk), - !tport->congested); - lock_sock(sk); + res = tipc_wait_for_sndmsg(sock, &timeo); if (res) break; } while (1); @@ -624,8 +686,37 @@ exit: return res; } +static int tipc_wait_for_sndpkt(struct socket *sock, long *timeo_p) +{ + struct sock *sk = sock->sk; + struct tipc_sock *tsk = tipc_sk(sk); + struct tipc_port *port = &tsk->port; + 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, + (!port->congested || !port->connected)); + finish_wait(sk_sleep(sk), &wait); + } while (!done); + return 0; +} + /** - * send_packet - send a connection-oriented message + * tipc_send_packet - send a connection-oriented message * @iocb: if NULL, indicates that socket lock is already held * @sock: socket structure * @m: message to send @@ -635,55 +726,50 @@ exit: * * 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) +static int tipc_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; + struct tipc_sock *tsk = tipc_sk(sk); + DECLARE_SOCKADDR(struct sockaddr_tipc *, dest, m->msg_name); + int res = -EINVAL; + long timeo; /* Handle implied connection establishment */ - if (unlikely(dest)) - return send_msg(iocb, sock, m, total_len); + return tipc_sendmsg(iocb, sock, m, total_len); + + if (total_len > TIPC_MAX_USER_MSG_SIZE) + return -EMSGSIZE; if (iocb) lock_sock(sk); - do { - if (unlikely(sock->state != SS_CONNECTED)) { - if (sock->state == SS_DISCONNECTING) - res = -EPIPE; - else - res = -ENOTCONN; - break; - } + if (unlikely(sock->state != SS_CONNECTED)) { + if (sock->state == SS_DISCONNECTING) + res = -EPIPE; + else + res = -ENOTCONN; + goto exit; + } - res = tipc_send(tport->ref, m->msg_iovlen, m->msg_iov); - if (likely(res != -ELINKCONG)) { - break; - } - if (m->msg_flags & MSG_DONTWAIT) { - res = -EWOULDBLOCK; + timeo = sock_sndtimeo(sk, m->msg_flags & MSG_DONTWAIT); + do { + res = tipc_send(&tsk->port, m->msg_iov, total_len); + if (likely(res != -ELINKCONG)) break; - } - release_sock(sk); - res = wait_event_interruptible(*sk_sleep(sk), - (!tport->congested || !tport->connected)); - lock_sock(sk); + res = tipc_wait_for_sndpkt(sock, &timeo); if (res) break; } while (1); - +exit: if (iocb) release_sock(sk); return res; } /** - * send_stream - send stream-oriented data + * tipc_send_stream - send stream-oriented data * @iocb: (unused) * @sock: socket structure * @m: data to send @@ -694,12 +780,11 @@ static int send_packet(struct kiocb *iocb, struct socket *sock, * 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) +static int tipc_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 tipc_sock *tsk = tipc_sk(sk); struct msghdr my_msg; struct iovec my_iov; struct iovec *curr_iov; @@ -714,18 +799,12 @@ static int send_stream(struct kiocb *iocb, struct socket *sock, 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 (sock->state == SS_UNCONNECTED) + res = tipc_send_packet(NULL, sock, m, total_len); + else + res = sock->state == SS_DISCONNECTING ? -EPIPE : -ENOTCONN; + goto exit; } if (unlikely(m->msg_name)) { @@ -733,6 +812,11 @@ static int send_stream(struct kiocb *iocb, struct socket *sock, goto exit; } + if (total_len > (unsigned int)INT_MAX) { + res = -EMSGSIZE; + goto exit; + } + /* * Send each iovec entry using one or more messages * @@ -740,7 +824,6 @@ static int send_stream(struct kiocb *iocb, struct socket *sock, * (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; @@ -749,21 +832,23 @@ static int send_stream(struct kiocb *iocb, struct socket *sock, my_msg.msg_name = NULL; bytes_sent = 0; - hdr_size = msg_hdr_sz(&tport->phdr); + hdr_size = msg_hdr_sz(&tsk->port.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; + bytes_to_send = tsk->port.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) { + res = tipc_send_packet(NULL, sock, &my_msg, + bytes_to_send); + if (res < 0) { if (bytes_sent) res = bytes_sent; goto exit; @@ -783,25 +868,25 @@ exit: /** * auto_connect - complete connection setup to a remote port - * @sock: socket structure + * @tsk: tipc socket structure * @msg: peer's response message * * Returns 0 on success, errno otherwise */ - -static int auto_connect(struct socket *sock, struct tipc_msg *msg) +static int auto_connect(struct tipc_sock *tsk, struct tipc_msg *msg) { - struct tipc_sock *tsock = tipc_sk(sock->sk); + struct tipc_port *port = &tsk->port; + struct socket *sock = tsk->sk.sk_socket; + 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_port_connect(port->ref, port, &peer); - 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)); + if (msg_importance(msg) > TIPC_CRITICAL_IMPORTANCE) + return -EINVAL; + msg_set_importance(&port->phdr, (u32)msg_importance(msg)); sock->state = SS_CONNECTED; return 0; } @@ -813,18 +898,18 @@ static int auto_connect(struct socket *sock, struct tipc_msg *msg) * * 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; + DECLARE_SOCKADDR(struct sockaddr_tipc *, addr, m->msg_name); if (addr) { addr->family = AF_TIPC; addr->addrtype = TIPC_ADDR_ID; + memset(&addr->addr, 0, sizeof(addr->addr)); 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 */ + addr->addr.name.domain = 0; /* could leave uninitialized */ + addr->scope = 0; /* could leave uninitialized */ m->msg_namelen = sizeof(struct sockaddr_tipc); } } @@ -839,9 +924,8 @@ static void set_orig_addr(struct msghdr *m, struct tipc_msg *msg) * * Returns 0 if successful, otherwise errno */ - static int anc_data_recv(struct msghdr *m, struct tipc_msg *msg, - struct tipc_port *tport) + struct tipc_port *tport) { u32 anc_data[3]; u32 err; @@ -853,21 +937,22 @@ static int anc_data_recv(struct msghdr *m, struct tipc_msg *msg, 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)))) + 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: @@ -891,15 +976,50 @@ static int anc_data_recv(struct msghdr *m, struct tipc_msg *msg, default: has_name = 0; } - if (has_name && - (res = put_cmsg(m, SOL_TIPC, TIPC_DESTNAME, 12, anc_data))) - return res; + if (has_name) { + res = put_cmsg(m, SOL_TIPC, TIPC_DESTNAME, 12, anc_data); + if (res) + return res; + } return 0; } +static int tipc_wait_for_rcvmsg(struct socket *sock, long *timeop) +{ + struct sock *sk = sock->sk; + DEFINE_WAIT(wait); + long timeo = *timeop; + int err; + + for (;;) { + prepare_to_wait(sk_sleep(sk), &wait, TASK_INTERRUPTIBLE); + if (timeo && 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); + *timeop = timeo; + return err; +} + /** - * recv_msg - receive packet-oriented message + * tipc_recvmsg - receive packet-oriented message * @iocb: (unused) * @m: descriptor for message info * @buf_len: total size of user buffer area @@ -910,23 +1030,20 @@ static int anc_data_recv(struct msghdr *m, struct tipc_msg *msg, * * 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) +static int tipc_recvmsg(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 tipc_sock *tsk = tipc_sk(sk); + struct tipc_port *port = &tsk->port; struct sk_buff *buf; struct tipc_msg *msg; + long timeo; 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; @@ -937,72 +1054,44 @@ static int recv_msg(struct kiocb *iocb, struct socket *sock, goto exit; } + 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 (flags & MSG_DONTWAIT) { - res = -EWOULDBLOCK; - goto exit; - } - release_sock(sk); - res = wait_event_interruptible(*sk_sleep(sk), - (!skb_queue_empty(&sk->sk_receive_queue) || - (sock->state == SS_DISCONNECTING))); - lock_sock(sk); - if (res) - goto exit; - } + 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); 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); + res = anc_data_recv(m, msg, port); 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; + res = skb_copy_datagram_iovec(buf, msg_hdr_sz(msg), + m->msg_iov, sz); + if (res) goto exit; - } res = sz; } else { if ((sock->state == SS_READY) || @@ -1013,11 +1102,10 @@ restart: } /* 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); + (++port->conn_unacked >= TIPC_CONNACK_INTV)) + tipc_acknowledge(port->ref, port->conn_unacked); advance_rx_queue(sk); } exit: @@ -1026,7 +1114,7 @@ exit: } /** - * recv_stream - receive stream-oriented data + * tipc_recv_stream - receive stream-oriented data * @iocb: (unused) * @m: descriptor for message info * @buf_len: total size of user buffer area @@ -1037,106 +1125,82 @@ exit: * * 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) +static int tipc_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 tipc_sock *tsk = tipc_sk(sk); + struct tipc_port *port = &tsk->port; struct sk_buff *buf; struct tipc_msg *msg; + long timeo; unsigned int sz; int sz_to_copy, target, needed; int sz_copied = 0; - 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))) { + if (unlikely(sock->state == SS_UNCONNECTED)) { res = -ENOTCONN; goto exit; } target = sock_rcvlowat(sk, flags & MSG_WAITALL, buf_len); + 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 (flags & MSG_DONTWAIT) { - res = -EWOULDBLOCK; - goto exit; - } - release_sock(sk); - res = wait_event_interruptible(*sk_sleep(sk), - (!skb_queue_empty(&sk->sk_receive_queue) || - (sock->state == SS_DISCONNECTING))); - lock_sock(sk); - if (res) - goto exit; - } + 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); 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); + res = anc_data_recv(m, msg, port); 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; + u32 offset = (u32)(unsigned long)(TIPC_SKB_CB(buf)->handle); + sz -= offset; 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; + + res = skb_copy_datagram_iovec(buf, msg_hdr_sz(msg) + offset, + m->msg_iov, sz_to_copy); + if (res) 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; + TIPC_SKB_CB(buf)->handle = + (void *)(unsigned long)(offset + sz_to_copy); goto exit; } - - crs += sz_to_copy; } else { if (sz_copied != 0) goto exit; /* can't add error msg to valid data */ @@ -1148,15 +1212,13 @@ restart: } /* 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); + if (unlikely(++port->conn_unacked >= TIPC_CONNACK_INTV)) + tipc_acknowledge(port->ref, port->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(&sk->sk_receive_queue) || (sz_copied < target)) && /* and more is ready or required */ @@ -1170,32 +1232,144 @@ exit: } /** - * 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 + * tipc_write_space - wake up thread if port congestion is released + * @sk: socket + */ +static void tipc_write_space(struct sock *sk) +{ + struct socket_wq *wq; + + rcu_read_lock(); + wq = rcu_dereference(sk->sk_wq); + if (wq_has_sleeper(wq)) + wake_up_interruptible_sync_poll(&wq->wait, POLLOUT | + POLLWRNORM | POLLWRBAND); + rcu_read_unlock(); +} + +/** + * tipc_data_ready - wake up threads to indicate messages have been received + * @sk: socket + * @len: the length of messages + */ +static void tipc_data_ready(struct sock *sk) +{ + struct socket_wq *wq; + + rcu_read_lock(); + wq = rcu_dereference(sk->sk_wq); + if (wq_has_sleeper(wq)) + wake_up_interruptible_sync_poll(&wq->wait, POLLIN | + POLLRDNORM | POLLRDBAND); + rcu_read_unlock(); +} + +/** + * filter_connect - Handle all incoming messages for a connection-based socket + * @tsk: TIPC socket + * @msg: message * - * Returns 1 if queue is unable to accept message, 0 otherwise + * Returns TIPC error status code and socket error status code + * once it encounters some errors */ +static u32 filter_connect(struct tipc_sock *tsk, struct sk_buff **buf) +{ + struct sock *sk = &tsk->sk; + struct tipc_port *port = &tsk->port; + struct socket *sock = sk->sk_socket; + struct tipc_msg *msg = buf_msg(*buf); + + u32 retval = TIPC_ERR_NO_PORT; + int res; + + if (msg_mcast(msg)) + return retval; + + switch ((int)sock->state) { + case SS_CONNECTED: + /* Accept only connection-based messages sent by peer */ + if (msg_connected(msg) && tipc_port_peer_msg(port, msg)) { + if (unlikely(msg_errcode(msg))) { + sock->state = SS_DISCONNECTING; + __tipc_port_disconnect(port); + } + retval = TIPC_OK; + } + break; + case SS_CONNECTING: + /* Accept only ACK or NACK message */ + if (unlikely(msg_errcode(msg))) { + sock->state = SS_DISCONNECTING; + sk->sk_err = ECONNREFUSED; + retval = TIPC_OK; + break; + } + + if (unlikely(!msg_connected(msg))) + break; + + res = auto_connect(tsk, msg); + if (res) { + sock->state = SS_DISCONNECTING; + sk->sk_err = -res; + retval = TIPC_OK; + break; + } + + /* If an incoming message is an 'ACK-', it should be + * discarded here because it doesn't contain useful + * data. In addition, we should try to wake up + * connect() routine if sleeping. + */ + if (msg_data_sz(msg) == 0) { + kfree_skb(*buf); + *buf = NULL; + if (waitqueue_active(sk_sleep(sk))) + wake_up_interruptible(sk_sleep(sk)); + } + retval = TIPC_OK; + break; + case SS_LISTENING: + case SS_UNCONNECTED: + /* Accept only SYN message */ + if (!msg_connected(msg) && !(msg_errcode(msg))) + retval = TIPC_OK; + break; + case SS_DISCONNECTING: + break; + default: + pr_err("Unknown socket state %u\n", sock->state); + } + return retval; +} -static int rx_queue_full(struct tipc_msg *msg, u32 queue_size, u32 base) +/** + * rcvbuf_limit - get proper overload limit of socket receive queue + * @sk: socket + * @buf: message + * + * For all connection oriented messages, irrespective of importance, + * the default overload value (i.e. 67MB) is set as limit. + * + * For all connectionless messages, by default new queue limits are + * as belows: + * + * TIPC_LOW_IMPORTANCE (4 MB) + * TIPC_MEDIUM_IMPORTANCE (8 MB) + * TIPC_HIGH_IMPORTANCE (16 MB) + * TIPC_CRITICAL_IMPORTANCE (32 MB) + * + * Returns overload limit according to corresponding message importance + */ +static unsigned int rcvbuf_limit(struct sock *sk, struct sk_buff *buf) { - 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; + struct tipc_msg *msg = buf_msg(buf); if (msg_connected(msg)) - threshold *= 4; + return sysctl_tipc_rmem[2]; - return queue_size >= threshold; + return sk->sk_rcvbuf >> TIPC_CRITICAL_IMPORTANCE << + msg_importance(msg); } /** @@ -1210,95 +1384,42 @@ static int rx_queue_full(struct tipc_msg *msg, u32 queue_size, u32 base) * * 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_sock *tsk = tipc_sk(sk); struct tipc_msg *msg = buf_msg(buf); - u32 recv_q_len; + unsigned int limit = rcvbuf_limit(sk, buf); + u32 res = TIPC_OK; /* 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 (msg_type(msg) > TIPC_DIRECT_MSG) + return TIPC_ERR_NO_PORT; if (sock->state == SS_READY) { - if (msg_connected(msg)) { - msg_dbg(msg, "dispatch filter 1\n"); + if (msg_connected(msg)) 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; - } - } + res = filter_connect(tsk, &buf); + if (res != TIPC_OK || buf == NULL) + return res; } /* Reject message if there isn't room to queue it */ + if (sk_rmem_alloc_get(sk) + buf->truesize >= limit) + return TIPC_ERR_OVERLOAD; - 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); + /* Enqueue message */ + TIPC_SKB_CB(buf)->handle = NULL; __skb_queue_tail(&sk->sk_receive_queue, buf); + skb_set_owner_r(buf, sk); - /* 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_sleep(sk))) - wake_up_interruptible(sk_sleep(sk)); + sk->sk_data_ready(sk); return TIPC_OK; } /** - * backlog_rcv - handle incoming message from backlog queue + * tipc_backlog_rcv - handle incoming message from backlog queue * @sk: socket * @buf: message * @@ -1306,70 +1427,100 @@ static u32 filter_rcv(struct sock *sk, struct sk_buff *buf) * * Returns 0 */ - -static int backlog_rcv(struct sock *sk, struct sk_buff *buf) +static int tipc_backlog_rcv(struct sock *sk, struct sk_buff *buf) { u32 res; + struct tipc_sock *tsk = tipc_sk(sk); + uint truesize = buf->truesize; res = filter_rcv(sk, buf); - if (res) + if (unlikely(res)) tipc_reject_msg(buf, res); + + if (atomic_read(&tsk->dupl_rcvcnt) < TIPC_CONN_OVERLOAD_LIMIT) + atomic_add(truesize, &tsk->dupl_rcvcnt); + 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) + * tipc_sk_rcv - handle incoming message + * @buf: buffer containing arriving message + * Consumes buffer + * Returns 0 if success, or errno: -EHOSTUNREACH */ - -static u32 dispatch(struct tipc_port *tport, struct sk_buff *buf) +int tipc_sk_rcv(struct sk_buff *buf) { - struct sock *sk = (struct sock *)tport->usr_handle; - u32 res; + struct tipc_sock *tsk; + struct tipc_port *port; + struct sock *sk; + u32 dport = msg_destport(buf_msg(buf)); + int err = TIPC_OK; + uint limit; - /* - * 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 - */ + /* Forward unresolved named message */ + if (unlikely(!dport)) { + tipc_net_route_msg(buf); + return 0; + } + /* Validate destination */ + port = tipc_port_lock(dport); + if (unlikely(!port)) { + err = TIPC_ERR_NO_PORT; + goto exit; + } + + tsk = tipc_port_to_sock(port); + sk = &tsk->sk; + + /* Queue message */ bh_lock_sock(sk); + if (!sock_owned_by_user(sk)) { - res = filter_rcv(sk, buf); + err = filter_rcv(sk, buf); } else { - if (sk_add_backlog(sk, buf)) - res = TIPC_ERR_OVERLOAD; - else - res = TIPC_OK; + if (sk->sk_backlog.len == 0) + atomic_set(&tsk->dupl_rcvcnt, 0); + limit = rcvbuf_limit(sk, buf) + atomic_read(&tsk->dupl_rcvcnt); + if (sk_add_backlog(sk, buf, limit)) + err = TIPC_ERR_OVERLOAD; } + bh_unlock_sock(sk); + tipc_port_unlock(port); - return res; + if (likely(!err)) + return 0; +exit: + tipc_reject_msg(buf, err); + return -EHOSTUNREACH; } -/** - * wakeupdispatch - wake up port after congestion - * @tport: port to wakeup - * - * Called with port lock already taken. - */ - -static void wakeupdispatch(struct tipc_port *tport) +static int tipc_wait_for_connect(struct socket *sock, long *timeo_p) { - struct sock *sk = (struct sock *)tport->usr_handle; + struct sock *sk = sock->sk; + DEFINE_WAIT(wait); + int done; - if (waitqueue_active(sk_sleep(sk))) - wake_up_interruptible(sk_sleep(sk)); + 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 + * tipc_connect - establish a connection to another TIPC port * @sock: socket structure * @dest: socket address for destination port * @destlen: size of socket address data structure @@ -1377,131 +1528,93 @@ static void wakeupdispatch(struct tipc_port *tport) * * Returns 0 on success, errno otherwise */ - -static int connect(struct socket *sock, struct sockaddr *dest, int destlen, - int flags) +static int tipc_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; - long timeout; + long timeout = (flags & O_NONBLOCK) ? 0 : tipc_sk(sk)->conn_timeout; + socket_state previous; 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 = -EOPNOTSUPP; - 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; - } + previous = sock->state; + switch (sock->state) { + case SS_UNCONNECTED: + /* Send a 'SYN-' to destination */ + m.msg_name = dest; + m.msg_namelen = destlen; - /* Wait until an 'ACK' or 'RST' arrives, or a timeout occurs */ + /* If connect is in non-blocking case, set MSG_DONTWAIT to + * indicate send_msg() is never blocked. + */ + if (!timeout) + m.msg_flags = MSG_DONTWAIT; - timeout = tipc_sk(sk)->conn_timeout; - release_sock(sk); - res = wait_event_interruptible_timeout(*sk_sleep(sk), - (!skb_queue_empty(&sk->sk_receive_queue) || - (sock->state != SS_CONNECTING)), - timeout ? timeout : MAX_SCHEDULE_TIMEOUT); - lock_sock(sk); + res = tipc_sendmsg(NULL, sock, &m, 0); + if ((res < 0) && (res != -EWOULDBLOCK)) + goto exit; - 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; + /* Just entered SS_CONNECTING state; the only + * difference is that return value in non-blocking + * case is EINPROGRESS, rather than EALREADY. + */ + res = -EINPROGRESS; + case SS_CONNECTING: + 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; + break; } - exit: release_sock(sk); return res; } /** - * listen - allow socket to listen for incoming connections + * tipc_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) +static int tipc_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) + if (sock->state != SS_UNCONNECTED) res = -EINVAL; else { sock->state = SS_LISTENING; @@ -1512,101 +1625,126 @@ 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 (timeo && 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 + * tipc_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) +static int tipc_accept(struct socket *sock, struct socket *new_sock, int flags) { - struct sock *sk = sock->sk; + struct sock *new_sk, *sk = sock->sk; struct sk_buff *buf; + struct tipc_port *new_port; + struct tipc_msg *msg; + struct tipc_portid peer; + u32 new_ref; + long timeo; 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_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); - res = tipc_create(sock_net(sock->sk), new_sock, 0, 0); - if (!res) { - struct sock *new_sk = new_sock->sk; - struct tipc_sock *new_tsock = tipc_sk(new_sk); - struct tipc_port *new_tport = new_tsock->p; - u32 new_ref = new_tport->ref; - 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 */ + res = tipc_sk_create(sock_net(sock->sk), new_sock, 0, 1); + if (res) + goto exit; - new_tsock->peer_name.ref = msg_origport(msg); - new_tsock->peer_name.node = msg_orignode(msg); - tipc_connect2port(new_ref, &new_tsock->peer_name); - new_sock->state = SS_CONNECTED; + new_sk = new_sock->sk; + new_port = &tipc_sk(new_sk)->port; + new_ref = new_port->ref; + msg = buf_msg(buf); - 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); - } + /* we lock on new_sk; but lockdep sees the lock on sk */ + lock_sock_nested(new_sk, SINGLE_DEPTH_NESTING); - /* - * Respond to 'SYN-' by discarding it & returning 'ACK'-. - * Respond to 'SYN+' by queuing it on new socket. - */ + /* + * 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 */ + peer.ref = msg_origport(msg); + peer.node = msg_orignode(msg); + tipc_port_connect(new_ref, &peer); + new_sock->state = SS_CONNECTED; + + tipc_port_set_importance(new_port, msg_importance(msg)); + if (msg_named(msg)) { + new_port->conn_type = msg_nametype(msg); + new_port->conn_instance = msg_nameinst(msg); + } - msg_dbg(msg,"<ACC<: "); - if (!msg_data_sz(msg)) { - struct msghdr m = {NULL,}; + /* + * Respond to 'SYN-' by discarding it & returning 'ACK'-. + * Respond to 'SYN+' by queuing it on new socket. + */ + 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); + advance_rx_queue(sk); + tipc_send_packet(NULL, new_sock, &m, 0); + } else { + __skb_dequeue(&sk->sk_receive_queue); + __skb_queue_head(&new_sk->sk_receive_queue, buf); + skb_set_owner_r(buf, new_sk); } + release_sock(new_sk); exit: release_sock(sk); return res; } /** - * shutdown - shutdown socket connection + * tipc_shutdown - shutdown socket connection * @sock: socket structure * @how: direction to close (must be SHUT_RDWR) * @@ -1614,11 +1752,11 @@ exit: * * Returns 0 on success, errno otherwise */ - -static int shutdown(struct socket *sock, int how) +static int tipc_shutdown(struct socket *sock, int how) { struct sock *sk = sock->sk; - struct tipc_port *tport = tipc_sk_port(sk); + struct tipc_sock *tsk = tipc_sk(sk); + struct tipc_port *port = &tsk->port; struct sk_buff *buf; int res; @@ -1631,19 +1769,18 @@ static int shutdown(struct socket *sock, int how) case SS_CONNECTING: case SS_CONNECTED: - /* Disconnect and send a 'FIN+' or 'FIN-' message to peer */ restart: + /* Disconnect and send a 'FIN+' or 'FIN-' message to peer */ 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); + if (TIPC_SKB_CB(buf)->handle != NULL) { + kfree_skb(buf); goto restart; } - tipc_disconnect(tport->ref); + tipc_port_disconnect(port->ref); tipc_reject_msg(buf, TIPC_CONN_SHUTDOWN); } else { - tipc_shutdown(tport->ref); + tipc_port_shutdown(port->ref); } sock->state = SS_DISCONNECTING; @@ -1652,11 +1789,11 @@ restart: case SS_DISCONNECTING: - /* Discard any unreceived messages; wake up sleeping tasks */ + /* Discard any unreceived messages */ + __skb_queue_purge(&sk->sk_receive_queue); - discard_rx_queue(sk); - if (waitqueue_active(sk_sleep(sk))) - wake_up_interruptible(sk_sleep(sk)); + /* Wake up anyone sleeping in poll */ + sk->sk_state_change(sk); res = 0; break; @@ -1669,7 +1806,7 @@ restart: } /** - * setsockopt - set socket option + * tipc_setsockopt - set socket option * @sock: socket structure * @lvl: option level * @opt: option identifier @@ -1681,12 +1818,12 @@ restart: * * Returns 0 on success, errno otherwise */ - -static int setsockopt(struct socket *sock, - int lvl, int opt, char __user *ov, unsigned int ol) +static int tipc_setsockopt(struct socket *sock, int lvl, int opt, + char __user *ov, unsigned int ol) { struct sock *sk = sock->sk; - struct tipc_port *tport = tipc_sk_port(sk); + struct tipc_sock *tsk = tipc_sk(sk); + struct tipc_port *port = &tsk->port; u32 value; int res; @@ -1696,26 +1833,27 @@ static int setsockopt(struct socket *sock, return -ENOPROTOOPT; if (ol < sizeof(value)) return -EINVAL; - if ((res = get_user(value, (u32 __user *)ov))) + res = get_user(value, (u32 __user *)ov); + if (res) return res; lock_sock(sk); switch (opt) { case TIPC_IMPORTANCE: - res = tipc_set_portimportance(tport->ref, value); + tipc_port_set_importance(port, value); break; case TIPC_SRC_DROPPABLE: if (sock->type != SOCK_STREAM) - res = tipc_set_portunreliable(tport->ref, value); + tipc_port_set_unreliable(port, value); else res = -ENOPROTOOPT; break; case TIPC_DEST_DROPPABLE: - res = tipc_set_portunreturnable(tport->ref, value); + tipc_port_set_unreturnable(port, value); break; case TIPC_CONN_TIMEOUT: - tipc_sk(sk)->conn_timeout = msecs_to_jiffies(value); + tipc_sk(sk)->conn_timeout = value; /* no need to set "res", since already 0 at this point */ break; default: @@ -1728,7 +1866,7 @@ static int setsockopt(struct socket *sock, } /** - * getsockopt - get socket option + * tipc_getsockopt - get socket option * @sock: socket structure * @lvl: option level * @opt: option identifier @@ -1740,12 +1878,12 @@ static int setsockopt(struct socket *sock, * * Returns 0 on success, errno otherwise */ - -static int getsockopt(struct socket *sock, - int lvl, int opt, char __user *ov, int __user *ol) +static int tipc_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); + struct tipc_sock *tsk = tipc_sk(sk); + struct tipc_port *port = &tsk->port; int len; u32 value; int res; @@ -1754,29 +1892,30 @@ static int getsockopt(struct socket *sock, return put_user(0, ol); if (lvl != SOL_TIPC) return -ENOPROTOOPT; - if ((res = get_user(len, ol))) + res = get_user(len, ol); + if (res) return res; lock_sock(sk); switch (opt) { case TIPC_IMPORTANCE: - res = tipc_portimportance(tport->ref, &value); + value = tipc_port_importance(port); break; case TIPC_SRC_DROPPABLE: - res = tipc_portunreliable(tport->ref, &value); + value = tipc_port_unreliable(port); break; case TIPC_DEST_DROPPABLE: - res = tipc_portunreturnable(tport->ref, &value); + value = tipc_port_unreturnable(port); break; case TIPC_CONN_TIMEOUT: - value = jiffies_to_msecs(tipc_sk(sk)->conn_timeout); + value = tipc_sk(sk)->conn_timeout; /* no need to set "res", since already 0 at this point */ break; - case TIPC_NODE_RECVQ_DEPTH: - value = (u32)atomic_read(&tipc_queue_size); + case TIPC_NODE_RECVQ_DEPTH: + value = 0; /* was tipc_queue_size, now obsolete */ break; - case TIPC_SOCK_RECVQ_DEPTH: + case TIPC_SOCK_RECVQ_DEPTH: value = skb_queue_len(&sk->sk_receive_queue); break; default: @@ -1785,99 +1924,122 @@ static int getsockopt(struct socket *sock, 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); - } + if (res) + return res; /* "get" failed */ - return res; + if (len < sizeof(value)) + return -EINVAL; + + if (copy_to_user(ov, &value, sizeof(value))) + return -EFAULT; + + return put_user(sizeof(value), ol); } -/** - * Protocol switches for the various types of TIPC sockets - */ +int tipc_ioctl(struct socket *sk, unsigned int cmd, unsigned long arg) +{ + struct tipc_sioc_ln_req lnr; + void __user *argp = (void __user *)arg; + + switch (cmd) { + case SIOCGETLINKNAME: + if (copy_from_user(&lnr, argp, sizeof(lnr))) + return -EFAULT; + if (!tipc_node_get_linkname(lnr.bearer_id, lnr.peer, + lnr.linkname, TIPC_MAX_LINK_NAME)) { + if (copy_to_user(argp, &lnr, sizeof(lnr))) + return -EFAULT; + return 0; + } + return -EADDRNOTAVAIL; + break; + default: + return -ENOIOCTLCMD; + } +} + +/* Protocol switches for the various types of TIPC sockets */ static const struct proto_ops msg_ops = { - .owner = THIS_MODULE, + .owner = THIS_MODULE, .family = AF_TIPC, - .release = release, - .bind = bind, - .connect = connect, + .release = tipc_release, + .bind = tipc_bind, + .connect = tipc_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, + .accept = sock_no_accept, + .getname = tipc_getname, + .poll = tipc_poll, + .ioctl = tipc_ioctl, + .listen = sock_no_listen, + .shutdown = tipc_shutdown, + .setsockopt = tipc_setsockopt, + .getsockopt = tipc_getsockopt, + .sendmsg = tipc_sendmsg, + .recvmsg = tipc_recvmsg, .mmap = sock_no_mmap, .sendpage = sock_no_sendpage }; static const struct proto_ops packet_ops = { - .owner = THIS_MODULE, + .owner = THIS_MODULE, .family = AF_TIPC, - .release = release, - .bind = bind, - .connect = connect, + .release = tipc_release, + .bind = tipc_bind, + .connect = tipc_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, + .accept = tipc_accept, + .getname = tipc_getname, + .poll = tipc_poll, + .ioctl = tipc_ioctl, + .listen = tipc_listen, + .shutdown = tipc_shutdown, + .setsockopt = tipc_setsockopt, + .getsockopt = tipc_getsockopt, + .sendmsg = tipc_send_packet, + .recvmsg = tipc_recvmsg, .mmap = sock_no_mmap, .sendpage = sock_no_sendpage }; static const struct proto_ops stream_ops = { - .owner = THIS_MODULE, + .owner = THIS_MODULE, .family = AF_TIPC, - .release = release, - .bind = bind, - .connect = connect, + .release = tipc_release, + .bind = tipc_bind, + .connect = tipc_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, + .accept = tipc_accept, + .getname = tipc_getname, + .poll = tipc_poll, + .ioctl = tipc_ioctl, + .listen = tipc_listen, + .shutdown = tipc_shutdown, + .setsockopt = tipc_setsockopt, + .getsockopt = tipc_getsockopt, + .sendmsg = tipc_send_stream, + .recvmsg = tipc_recv_stream, .mmap = sock_no_mmap, .sendpage = sock_no_sendpage }; static const struct net_proto_family tipc_family_ops = { - .owner = THIS_MODULE, + .owner = THIS_MODULE, .family = AF_TIPC, - .create = tipc_create + .create = tipc_sk_create }; static struct proto tipc_proto = { .name = "TIPC", .owner = THIS_MODULE, - .obj_size = sizeof(struct tipc_sock) + .obj_size = sizeof(struct tipc_sock), + .sysctl_rmem = sysctl_tipc_rmem +}; + +static struct proto tipc_proto_kern = { + .name = "TIPC", + .obj_size = sizeof(struct tipc_sock), + .sysctl_rmem = sysctl_tipc_rmem }; /** @@ -1891,18 +2053,16 @@ int tipc_socket_init(void) res = proto_register(&tipc_proto, 1); if (res) { - err("Failed to register TIPC protocol type\n"); + pr_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"); + pr_err("Failed to register TIPC socket type\n"); proto_unregister(&tipc_proto); goto out; } - - sockets_enabled = 1; out: return res; } @@ -1910,14 +2070,8 @@ int tipc_socket_init(void) /** * 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); } - |