diff options
Diffstat (limited to 'net/tipc/socket.c')
-rw-r--r-- | net/tipc/socket.c | 1722 |
1 files changed, 1722 insertions, 0 deletions
diff --git a/net/tipc/socket.c b/net/tipc/socket.c new file mode 100644 index 000000000000..22dcb724e760 --- /dev/null +++ b/net/tipc/socket.c @@ -0,0 +1,1722 @@ +/* + * net/tipc/socket.c: TIPC socket API + * + * Copyright (c) 2003-2005, Ericsson Research Canada + * Copyright (c) 2004-2005, Wind River Systems + * Copyright (c) 2005-2006, Ericsson AB + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * Redistributions of source code must retain the above copyright notice, this + * list of conditions and the following disclaimer. + * 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. + * 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. + * + * 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/version.h> +#include <linux/fcntl.h> +#include <linux/version.h> +#include <asm/semaphore.h> +#include <asm/string.h> +#include <asm/atomic.h> +#include <net/sock.h> + +#include <linux/tipc.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 + +struct tipc_sock { + struct sock sk; + struct tipc_port *p; + struct semaphore sem; +}; + +#define tipc_sk(sk) ((struct tipc_sock*)sk) + +static u32 dispatch(struct tipc_port *tport, struct sk_buff *buf); +static void wakeupdispatch(struct tipc_port *tport); + +static struct proto_ops packet_ops; +static struct proto_ops stream_ops; +static struct proto_ops msg_ops; + +static struct proto tipc_proto; + +static int sockets_enabled = 0; + +static atomic_t tipc_queue_size = ATOMIC_INIT(0); + + +/* + * sock_lock(): Lock a port/socket pair. lock_sock() can + * not be used here, since the same lock must protect ports + * with non-socket interfaces. + * See net.c for description of locking policy. + */ +static inline void sock_lock(struct tipc_sock* tsock) +{ + spin_lock_bh(tsock->p->lock); +} + +/* + * sock_unlock(): Unlock a port/socket pair + */ +static inline void sock_unlock(struct tipc_sock* tsock) +{ + spin_unlock_bh(tsock->p->lock); +} + +/** + * pollmask - determine the current set of poll() events for a socket + * @sock: socket structure + * + * 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! + * + * Returns pollmask value + */ + +static inline u32 pollmask(struct socket *sock) +{ + u32 mask; + + if ((skb_queue_len(&sock->sk->sk_receive_queue) != 0) || + (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; +} + + +/** + * advance_queue - discard first buffer in queue + * @tsock: TIPC socket + */ + +static inline void advance_queue(struct tipc_sock *tsock) +{ + sock_lock(tsock); + buf_discard(skb_dequeue(&tsock->sk.sk_receive_queue)); + sock_unlock(tsock); + atomic_dec(&tipc_queue_size); +} + +/** + * tipc_create - create a TIPC socket + * @sock: pre-allocated socket structure + * @protocol: protocol indicator (must be 0) + * + * This routine creates and attaches a 'struct sock' to the 'struct socket', + * then create and attaches a TIPC port to the 'struct sock' part. + * + * Returns 0 on success, errno otherwise + */ +static int tipc_create(struct socket *sock, int protocol) +{ + struct tipc_sock *tsock; + struct tipc_port *port; + struct sock *sk; + u32 ref; + + if ((sock->type != SOCK_STREAM) && + (sock->type != SOCK_SEQPACKET) && + (sock->type != SOCK_DGRAM) && + (sock->type != SOCK_RDM)) + return -EPROTOTYPE; + + if (unlikely(protocol != 0)) + return -EPROTONOSUPPORT; + + ref = tipc_createport_raw(0, &dispatch, &wakeupdispatch, TIPC_LOW_IMPORTANCE); + if (unlikely(!ref)) + return -ENOMEM; + + sock->state = SS_UNCONNECTED; + + switch (sock->type) { + case SOCK_STREAM: + sock->ops = &stream_ops; + break; + case SOCK_SEQPACKET: + sock->ops = &packet_ops; + break; + case SOCK_DGRAM: + tipc_set_portunreliable(ref, 1); + /* fall through */ + case SOCK_RDM: + tipc_set_portunreturnable(ref, 1); + sock->ops = &msg_ops; + sock->state = SS_READY; + break; + } + + sk = sk_alloc(AF_TIPC, GFP_KERNEL, &tipc_proto, 1); + if (!sk) { + tipc_deleteport(ref); + return -ENOMEM; + } + + sock_init_data(sock, sk); + init_waitqueue_head(sk->sk_sleep); + sk->sk_rcvtimeo = 8 * HZ; /* default connect timeout = 8s */ + + tsock = tipc_sk(sk); + port = tipc_get_port(ref); + + tsock->p = port; + port->usr_handle = tsock; + + init_MUTEX(&tsock->sem); + + dbg("sock_create: %x\n",tsock); + + 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 tipc_sock *tsock = tipc_sk(sock->sk); + struct sock *sk = sock->sk; + int res = TIPC_OK; + struct sk_buff *buf; + + dbg("sock_delete: %x\n",tsock); + if (!tsock) + return 0; + down_interruptible(&tsock->sem); + if (!sock->sk) { + up(&tsock->sem); + return 0; + } + + /* Reject unreceived messages, unless no longer connected */ + + while (sock->state != SS_DISCONNECTING) { + sock_lock(tsock); + buf = skb_dequeue(&sk->sk_receive_queue); + if (!buf) + tsock->p->usr_handle = 0; + sock_unlock(tsock); + if (!buf) + break; + if (TIPC_SKB_CB(buf)->handle != msg_data(buf_msg(buf))) + buf_discard(buf); + else + tipc_reject_msg(buf, TIPC_ERR_NO_PORT); + atomic_dec(&tipc_queue_size); + } + + /* Delete TIPC port */ + + res = tipc_deleteport(tsock->p->ref); + sock->sk = NULL; + + /* Discard any remaining messages */ + + while ((buf = skb_dequeue(&sk->sk_receive_queue))) { + buf_discard(buf); + atomic_dec(&tipc_queue_size); + } + + up(&tsock->sem); + + sock_put(sk); + + 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 + */ + +static int bind(struct socket *sock, struct sockaddr *uaddr, int uaddr_len) +{ + struct tipc_sock *tsock = tipc_sk(sock->sk); + struct sockaddr_tipc *addr = (struct sockaddr_tipc *)uaddr; + int res; + + if (down_interruptible(&tsock->sem)) + return -ERESTARTSYS; + + if (unlikely(!uaddr_len)) { + res = tipc_withdraw(tsock->p->ref, 0, 0); + goto exit; + } + + 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) { + res = -EAFNOSUPPORT; + goto exit; + } + + if (addr->scope > 0) + res = tipc_publish(tsock->p->ref, addr->scope, + &addr->addr.nameseq); + else + res = tipc_withdraw(tsock->p->ref, -addr->scope, + &addr->addr.nameseq); +exit: + up(&tsock->sem); + return res; +} + +/** + * 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 + */ + +static int get_name(struct socket *sock, struct sockaddr *uaddr, + int *uaddr_len, int peer) +{ + struct tipc_sock *tsock = tipc_sk(sock->sk); + struct sockaddr_tipc *addr = (struct sockaddr_tipc *)uaddr; + u32 res; + + if (down_interruptible(&tsock->sem)) + return -ERESTARTSYS; + + *uaddr_len = sizeof(*addr); + addr->addrtype = TIPC_ADDR_ID; + addr->family = AF_TIPC; + addr->scope = 0; + if (peer) + res = tipc_peer(tsock->p->ref, &addr->addr.id); + else + res = tipc_ownidentity(tsock->p->ref, &addr->addr.id); + addr->addr.name.domain = 0; + + up(&tsock->sem); + return res; +} + +/** + * 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 the pollmask + */ + +static unsigned int poll(struct file *file, struct socket *sock, + poll_table *wait) +{ + poll_wait(file, sock->sk->sk_sleep, wait); + /* NEED LOCK HERE? */ + return pollmask(sock); +} + +/** + * 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 inline 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: (unused) + * @sock: socket structure + * @m: message to send + * @total_len: (unused) + * + * 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 tipc_sock *tsock = tipc_sk(sock->sk); + struct sockaddr_tipc *dest = (struct sockaddr_tipc *)m->msg_name; + struct sk_buff *buf; + int needs_conn; + int res = -EINVAL; + + if (unlikely(!dest)) + return -EDESTADDRREQ; + if (unlikely(dest->family != AF_TIPC)) + return -EINVAL; + + needs_conn = (sock->state != SS_READY); + if (unlikely(needs_conn)) { + if (sock->state == SS_LISTENING) + return -EPIPE; + if (sock->state != SS_UNCONNECTED) + return -EISCONN; + if ((tsock->p->published) || + ((sock->type == SOCK_STREAM) && (total_len != 0))) + return -EOPNOTSUPP; + } + + if (down_interruptible(&tsock->sem)) + return -ERESTARTSYS; + + if (needs_conn) { + + /* Abort any pending connection attempts (very unlikely) */ + + while ((buf = skb_dequeue(&sock->sk->sk_receive_queue))) { + tipc_reject_msg(buf, TIPC_ERR_NO_PORT); + atomic_dec(&tipc_queue_size); + } + + sock->state = SS_CONNECTING; + } + + do { + if (dest->addrtype == TIPC_ADDR_NAME) { + if ((res = dest_name_check(dest, m))) + goto exit; + res = tipc_send2name(tsock->p->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(tsock->p->ref, + &dest->addr.id, + m->msg_iovlen, + m->msg_iov); + } + else if (dest->addrtype == TIPC_ADDR_MCAST) { + if (needs_conn) { + res = -EOPNOTSUPP; + goto exit; + } + if ((res = dest_name_check(dest, m))) + goto exit; + res = tipc_multicast(tsock->p->ref, + &dest->addr.nameseq, + 0, + m->msg_iovlen, + m->msg_iov); + } + if (likely(res != -ELINKCONG)) { +exit: + up(&tsock->sem); + return res; + } + if (m->msg_flags & MSG_DONTWAIT) { + res = -EWOULDBLOCK; + goto exit; + } + if (wait_event_interruptible(*sock->sk->sk_sleep, + !tsock->p->congested)) { + res = -ERESTARTSYS; + goto exit; + } + } while (1); +} + +/** + * send_packet - send a connection-oriented message + * @iocb: (unused) + * @sock: socket structure + * @m: message to send + * @total_len: (unused) + * + * 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 tipc_sock *tsock = tipc_sk(sock->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 (down_interruptible(&tsock->sem)) { + return -ERESTARTSYS; + } + + if (unlikely(sock->state != SS_CONNECTED)) { + if (sock->state == SS_DISCONNECTING) + res = -EPIPE; + else + res = -ENOTCONN; + goto exit; + } + + do { + res = tipc_send(tsock->p->ref, m->msg_iovlen, m->msg_iov); + if (likely(res != -ELINKCONG)) { +exit: + up(&tsock->sem); + return res; + } + if (m->msg_flags & MSG_DONTWAIT) { + res = -EWOULDBLOCK; + goto exit; + } + if (wait_event_interruptible(*sock->sk->sk_sleep, + !tsock->p->congested)) { + res = -ERESTARTSYS; + goto exit; + } + } while (1); +} + +/** + * 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 errno otherwise + */ + + +static int send_stream(struct kiocb *iocb, struct socket *sock, + struct msghdr *m, size_t total_len) +{ + struct msghdr my_msg; + struct iovec my_iov; + struct iovec *curr_iov; + int curr_iovlen; + char __user *curr_start; + int curr_left; + int bytes_to_send; + int res; + + if (likely(total_len <= TIPC_MAX_USER_MSG_SIZE)) + return send_packet(iocb, sock, m, total_len); + + /* Can only send large data streams if already connected */ + + if (unlikely(sock->state != SS_CONNECTED)) { + if (sock->state == SS_DISCONNECTING) + return -EPIPE; + else + return -ENOTCONN; + } + + /* + * 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(). + */ + + my_msg = *m; + curr_iov = my_msg.msg_iov; + curr_iovlen = my_msg.msg_iovlen; + my_msg.msg_iov = &my_iov; + my_msg.msg_iovlen = 1; + + while (curr_iovlen--) { + curr_start = curr_iov->iov_base; + curr_left = curr_iov->iov_len; + + while (curr_left) { + bytes_to_send = (curr_left < TIPC_MAX_USER_MSG_SIZE) + ? curr_left : TIPC_MAX_USER_MSG_SIZE; + my_iov.iov_base = curr_start; + my_iov.iov_len = bytes_to_send; + if ((res = send_packet(iocb, sock, &my_msg, 0)) < 0) + return res; + curr_left -= bytes_to_send; + curr_start += bytes_to_send; + } + + curr_iov++; + } + + return total_len; +} + +/** + * auto_connect - complete connection setup to a remote port + * @sock: socket structure + * @tsock: TIPC-specific socket structure + * @msg: peer's response message + * + * Returns 0 on success, errno otherwise + */ + +static int auto_connect(struct socket *sock, struct tipc_sock *tsock, + struct tipc_msg *msg) +{ + 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(tsock->p->ref, &peer); + tipc_set_portimportance(tsock->p->ref, msg_importance(msg)); + sock->state = SS_CONNECTED; + return 0; +} + +/** + * set_orig_addr - capture sender's address for received message + * @m: descriptor for message info + * @msg: received message header + * + * Note: Address is not captured if not requested by receiver. + */ + +static inline 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 inline 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 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_SOCKET, TIPC_ERRINFO, 8, anc_data))) + return res; + if (anc_data[1] && + (res = put_cmsg(m, SOL_SOCKET, 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: + anc_data[0] = msg_nametype(msg); + anc_data[1] = msg_namelower(msg); + anc_data[2] = msg_namelower(msg); + break; + case TIPC_MCAST_MSG: + anc_data[0] = msg_nametype(msg); + anc_data[1] = msg_namelower(msg); + anc_data[2] = msg_nameupper(msg); + break; + case TIPC_CONN_MSG: + anc_data[0] = tport->conn_type; + anc_data[1] = tport->conn_instance; + anc_data[2] = tport->conn_instance; + break; + default: + anc_data[0] = 0; + } + if (anc_data[0] && + (res = put_cmsg(m, SOL_SOCKET, 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 tipc_sock *tsock = tipc_sk(sock->sk); + struct sk_buff *buf; + struct tipc_msg *msg; + unsigned int q_len; + unsigned int sz; + u32 err; + int res; + + /* Currently doesn't support receiving into multiple iovec entries */ + + if (m->msg_iovlen != 1) + return -EOPNOTSUPP; + + /* Catch invalid receive attempts */ + + if (unlikely(!buf_len)) + return -EINVAL; + + if (sock->type == SOCK_SEQPACKET) { + if (unlikely(sock->state == SS_UNCONNECTED)) + return -ENOTCONN; + if (unlikely((sock->state == SS_DISCONNECTING) && + (skb_queue_len(&sock->sk->sk_receive_queue) == 0))) + return -ENOTCONN; + } + + /* Look for a message in receive queue; wait if necessary */ + + if (unlikely(down_interruptible(&tsock->sem))) + return -ERESTARTSYS; + +restart: + if (unlikely((skb_queue_len(&sock->sk->sk_receive_queue) == 0) && + (flags & MSG_DONTWAIT))) { + res = -EWOULDBLOCK; + goto exit; + } + + if ((res = wait_event_interruptible( + *sock->sk->sk_sleep, + ((q_len = skb_queue_len(&sock->sk->sk_receive_queue)) || + (sock->state == SS_DISCONNECTING))) )) { + goto exit; + } + + /* Catch attempt to receive on an already terminated connection */ + /* [THIS CHECK MAY OVERLAP WITH AN EARLIER CHECK] */ + + if (!q_len) { + res = -ENOTCONN; + goto exit; + } + + /* Get access to first message in receive queue */ + + buf = skb_peek(&sock->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)) { + if ((res = auto_connect(sock, tsock, msg))) + goto exit; + } + + /* Discard an empty non-errored message & try again */ + + if ((!sz) && (!err)) { + advance_queue(tsock); + goto restart; + } + + /* Capture sender's address (optional) */ + + set_orig_addr(m, msg); + + /* Capture ancillary data (optional) */ + + if ((res = anc_data_recv(m, msg, tsock->p))) + 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 (unlikely(++tsock->p->conn_unacked >= TIPC_FLOW_CONTROL_WIN)) + tipc_acknowledge(tsock->p->ref, tsock->p->conn_unacked); + advance_queue(tsock); + } +exit: + up(&tsock->sem); + 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 tipc_sock *tsock = tipc_sk(sock->sk); + struct sk_buff *buf; + struct tipc_msg *msg; + unsigned int q_len; + unsigned int sz; + int sz_to_copy; + int sz_copied = 0; + int needed; + char *crs = m->msg_iov->iov_base; + unsigned char *buf_crs; + u32 err; + int res; + + /* Currently doesn't support receiving into multiple iovec entries */ + + if (m->msg_iovlen != 1) + return -EOPNOTSUPP; + + /* Catch invalid receive attempts */ + + if (unlikely(!buf_len)) + return -EINVAL; + + if (unlikely(sock->state == SS_DISCONNECTING)) { + if (skb_queue_len(&sock->sk->sk_receive_queue) == 0) + return -ENOTCONN; + } else if (unlikely(sock->state != SS_CONNECTED)) + return -ENOTCONN; + + /* Look for a message in receive queue; wait if necessary */ + + if (unlikely(down_interruptible(&tsock->sem))) + return -ERESTARTSYS; + +restart: + if (unlikely((skb_queue_len(&sock->sk->sk_receive_queue) == 0) && + (flags & MSG_DONTWAIT))) { + res = (sz_copied == 0) ? -EWOULDBLOCK : 0; + goto exit; + } + + if ((res = wait_event_interruptible( + *sock->sk->sk_sleep, + ((q_len = skb_queue_len(&sock->sk->sk_receive_queue)) || + (sock->state == SS_DISCONNECTING))) )) { + goto exit; + } + + /* Catch attempt to receive on an already terminated connection */ + /* [THIS CHECK MAY OVERLAP WITH AN EARLIER CHECK] */ + + if (!q_len) { + res = -ENOTCONN; + goto exit; + } + + /* Get access to first message in receive queue */ + + buf = skb_peek(&sock->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_queue(tsock); + goto restart; + } + + /* Optionally capture sender's address & ancillary data of first msg */ + + if (sz_copied == 0) { + set_orig_addr(m, msg); + if ((res = anc_data_recv(m, msg, tsock->p))) + goto exit; + } + + /* Capture message data (if valid) & compute return value (always) */ + + if (!err) { + buf_crs = (unsigned char *)(TIPC_SKB_CB(buf)->handle); + sz = buf->tail - 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(++tsock->p->conn_unacked >= TIPC_FLOW_CONTROL_WIN)) + tipc_acknowledge(tsock->p->ref, tsock->p->conn_unacked); + advance_queue(tsock); + } + + /* Loop around if more data is required */ + + if ((sz_copied < buf_len) /* didn't get all requested data */ + && (flags & MSG_WAITALL) /* ... and need to wait for more */ + && (!(flags & MSG_PEEK)) /* ... and aren't just peeking at data */ + && (!err) /* ... and haven't reached a FIN */ + ) + goto restart; + +exit: + up(&tsock->sem); + return res ? res : sz_copied; +} + +/** + * queue_overloaded - test if queue overload condition exists + * @queue_size: current size of queue + * @base: nominal maximum size of queue + * @msg: message to be added to queue + * + * Returns 1 if queue is currently overloaded, 0 otherwise + */ + +static int queue_overloaded(u32 queue_size, u32 base, struct tipc_msg *msg) +{ + 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); +} + +/** + * async_disconnect - wrapper function used to disconnect port + * @portref: TIPC port reference (passed as pointer-sized value) + */ + +static void async_disconnect(unsigned long portref) +{ + tipc_disconnect((u32)portref); +} + +/** + * dispatch - handle arriving message + * @tport: TIPC port that received message + * @buf: message + * + * Called with port locked. Must not take socket lock to avoid deadlock risk. + * + * 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 tipc_msg *msg = buf_msg(buf); + struct tipc_sock *tsock = (struct tipc_sock *)tport->usr_handle; + struct socket *sock; + u32 recv_q_len; + + /* Reject message if socket is closing */ + + if (!tsock) + return TIPC_ERR_NO_PORT; + + /* 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 + */ + sock = tsock->sk.sk_socket; + 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 */ + + if (unlikely((u32)atomic_read(&tipc_queue_size) > + OVERLOAD_LIMIT_BASE)) { + if (queue_overloaded(atomic_read(&tipc_queue_size), + OVERLOAD_LIMIT_BASE, msg)) + return TIPC_ERR_OVERLOAD; + } + recv_q_len = skb_queue_len(&tsock->sk.sk_receive_queue); + if (unlikely(recv_q_len > (OVERLOAD_LIMIT_BASE / 2))) { + if (queue_overloaded(recv_q_len, + OVERLOAD_LIMIT_BASE / 2, msg)) + return TIPC_ERR_OVERLOAD; + } + + /* Initiate connection termination for an incoming 'FIN' */ + + if (unlikely(msg_errcode(msg) && (sock->state == SS_CONNECTED))) { + sock->state = SS_DISCONNECTING; + /* Note: Use signal since port lock is already taken! */ + k_signal((Handler)async_disconnect, tport->ref); + } + + /* Enqueue message (finally!) */ |