linux_dsm_epyc7002/net/tipc/socket.c

1885 lines
47 KiB
C
Raw Normal View History

/*
* net/tipc/socket.c: TIPC socket API
*
* Copyright (c) 2001-2007, 2012 Ericsson AB
* Copyright (c) 2004-2008, 2010-2012, Wind River Systems
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions are met:
*
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* 3. Neither the names of the copyright holders nor the names of its
* contributors may be used to endorse or promote products derived from
* this software without specific prior written permission.
*
* Alternatively, this software may be distributed under the terms of the
* GNU General Public License ("GPL") version 2 as published by the Free
* Software Foundation.
*
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
* AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
* LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
* CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
* SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
* INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
* CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
* ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
* POSSIBILITY OF SUCH DAMAGE.
*/
#include "core.h"
#include "port.h"
#include <linux/export.h>
#include <net/sock.h>
#define SS_LISTENING -1 /* socket is listening */
#define SS_READY -2 /* socket is connectionless */
#define CONN_OVERLOAD_LIMIT ((TIPC_FLOW_CONTROL_WIN * 2 + 1) * \
SKB_TRUESIZE(TIPC_MAX_USER_MSG_SIZE))
#define CONN_TIMEOUT_DEFAULT 8000 /* default connect timeout = 8s */
struct tipc_sock {
struct sock sk;
struct tipc_port *p;
struct tipc_portid peer_name;
unsigned int conn_timeout;
};
#define tipc_sk(sk) ((struct tipc_sock *)(sk))
#define tipc_sk_port(sk) (tipc_sk(sk)->p)
#define tipc_rx_ready(sock) (!skb_queue_empty(&sock->sk->sk_receive_queue) || \
(sock->state == SS_DISCONNECTING))
static int backlog_rcv(struct sock *sk, struct sk_buff *skb);
static u32 dispatch(struct tipc_port *tport, struct sk_buff *buf);
static void wakeupdispatch(struct tipc_port *tport);
static void tipc_data_ready(struct sock *sk, int len);
static void tipc_write_space(struct sock *sk);
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;
/*
* Revised TIPC socket locking policy:
*
* Most socket operations take the standard socket lock when they start
* and hold it until they finish (or until they need to sleep). Acquiring
* this lock grants the owner exclusive access to the fields of the socket
* data structures, with the exception of the backlog queue. A few socket
* operations can be done without taking the socket lock because they only
* read socket information that never changes during the life of the socket.
*
* Socket operations may acquire the lock for the associated TIPC port if they
* need to perform an operation on the port. If any routine needs to acquire
* both the socket lock and the port lock it must take the socket lock first
* to avoid the risk of deadlock.
*
* The dispatcher handling incoming messages cannot grab the socket lock in
* the standard fashion, since invoked it runs at the BH level and cannot block.
* Instead, it checks to see if the socket lock is currently owned by someone,
* and either handles the message itself or adds it to the socket's backlog
* queue; in the latter case the queued message is processed once the process
* owning the socket lock releases it.
*
* NOTE: Releasing the socket lock while an operation is sleeping overcomes
* the problem of a blocked socket operation preventing any other operations
* from occurring. However, applications must be careful if they have
* multiple threads trying to send (or receive) on the same socket, as these
* operations might interfere with each other. For example, doing a connect
* and a receive at the same time might allow the receive to consume the
* ACK message meant for the connect. While additional work could be done
* to try and overcome this, it doesn't seem to be worthwhile at the present.
*
* NOTE: Releasing the socket lock while an operation is sleeping also ensures
* that another operation that must be performed in a non-blocking manner is
* not delayed for very long because the lock has already been taken.
*
* NOTE: This code assumes that certain fields of a port/socket pair are
* constant over its lifetime; such fields can be examined without taking
* the socket lock and/or port lock, and do not need to be re-read even
* after resuming processing after waiting. These fields include:
* - socket type
* - pointer to socket sk structure (aka tipc_sock structure)
* - pointer to port structure
* - port reference
*/
/**
* advance_rx_queue - discard first buffer in socket receive queue
*
* Caller must hold socket lock
*/
static void advance_rx_queue(struct sock *sk)
{
kfree_skb(__skb_dequeue(&sk->sk_receive_queue));
}
/**
* reject_rx_queue - reject all buffers in socket receive queue
*
* Caller must hold socket lock
*/
static void reject_rx_queue(struct sock *sk)
{
struct sk_buff *buf;
while ((buf = __skb_dequeue(&sk->sk_receive_queue)))
tipc_reject_msg(buf, TIPC_ERR_NO_PORT);
}
/**
* tipc_create - create a TIPC socket
* @net: network namespace (must be default network)
* @sock: pre-allocated socket structure
* @protocol: protocol indicator (must be 0)
* @kern: caused by kernel or by userspace?
*
* This routine creates additional data structures used by the TIPC socket,
* initializes them, and links them together.
*
* Returns 0 on success, errno otherwise
*/
static int tipc_create(struct net *net, struct socket *sock, int protocol,
int kern)
{
const struct proto_ops *ops;
socket_state state;
struct sock *sk;
struct tipc_port *tp_ptr;
/* Validate arguments */
if (unlikely(protocol != 0))
return -EPROTONOSUPPORT;
switch (sock->type) {
case SOCK_STREAM:
ops = &stream_ops;
state = SS_UNCONNECTED;
break;
case SOCK_SEQPACKET:
ops = &packet_ops;
state = SS_UNCONNECTED;
break;
case SOCK_DGRAM:
case SOCK_RDM:
ops = &msg_ops;
state = SS_READY;
break;
default:
return -EPROTOTYPE;
}
/* Allocate socket's protocol area */
sk = sk_alloc(net, AF_TIPC, GFP_KERNEL, &tipc_proto);
if (sk == NULL)
return -ENOMEM;
/* Allocate TIPC port for socket to use */
tp_ptr = tipc_createport_raw(sk, &dispatch, &wakeupdispatch,
TIPC_LOW_IMPORTANCE);
if (unlikely(!tp_ptr)) {
sk_free(sk);
return -ENOMEM;
}
/* Finish initializing socket data structures */
sock->ops = ops;
sock->state = state;
sock_init_data(sock, sk);
sk->sk_backlog_rcv = backlog_rcv;
sk->sk_data_ready = tipc_data_ready;
sk->sk_write_space = tipc_write_space;
tipc_sk(sk)->p = tp_ptr;
tipc_sk(sk)->conn_timeout = CONN_TIMEOUT_DEFAULT;
spin_unlock_bh(tp_ptr->lock);
if (sock->state == SS_READY) {
tipc_set_portunreturnable(tp_ptr->ref, 1);
if (sock->type == SOCK_DGRAM)
tipc_set_portunreliable(tp_ptr->ref, 1);
}
return 0;
}
/**
* release - destroy a TIPC socket
* @sock: socket to destroy
*
* This routine cleans up any messages that are still queued on the socket.
* For DGRAM and RDM socket types, all queued messages are rejected.
* For SEQPACKET and STREAM socket types, the first message is rejected
* and any others are discarded. (If the first message on a STREAM socket
* is partially-read, it is discarded and the next one is rejected instead.)
*
* NOTE: Rejected messages are not necessarily returned to the sender! They
* are returned or discarded according to the "destination droppable" setting
* specified for the message by the sender.
*
* Returns 0 on success, errno otherwise
*/
static int release(struct socket *sock)
{
struct sock *sk = sock->sk;
struct tipc_port *tport;
struct sk_buff *buf;
int res;
/*
* Exit if socket isn't fully initialized (occurs when a failed accept()
* releases a pre-allocated child socket that was never used)
*/
if (sk == NULL)
return 0;
tport = tipc_sk_port(sk);
lock_sock(sk);
/*
* Reject all unreceived messages, except on an active connection
* (which disconnects locally & sends a 'FIN+' to peer)
*/
while (sock->state != SS_DISCONNECTING) {
buf = __skb_dequeue(&sk->sk_receive_queue);
if (buf == NULL)
break;
if (TIPC_SKB_CB(buf)->handle != 0)
kfree_skb(buf);
else {
if ((sock->state == SS_CONNECTING) ||
(sock->state == SS_CONNECTED)) {
sock->state = SS_DISCONNECTING;
tipc_disconnect(tport->ref);
}
tipc_reject_msg(buf, TIPC_ERR_NO_PORT);
}
}
/*
* Delete TIPC port; this ensures no more messages are queued
* (also disconnects an active connection & sends a 'FIN-' to peer)
*/
res = tipc_deleteport(tport->ref);
/* Discard any remaining (connection-based) messages in receive queue */
__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;
return res;
}
/**
* bind - associate or disassocate TIPC name(s) with a socket
* @sock: socket structure
* @uaddr: socket address describing name(s) and desired operation
* @uaddr_len: size of socket address data structure
*
* Name and name sequence binding is indicated using a positive scope value;
* a negative scope value unbinds the specified name. Specifying no name
* (i.e. a socket address length of 0) unbinds all names from the socket.
*
* Returns 0 on success, errno otherwise
*
* NOTE: This routine doesn't need to take the socket lock since it doesn't
* access any non-constant socket information.
*/
static int bind(struct socket *sock, struct sockaddr *uaddr, int uaddr_len)
{
struct sockaddr_tipc *addr = (struct sockaddr_tipc *)uaddr;
u32 portref = tipc_sk_port(sock->sk)->ref;
if (unlikely(!uaddr_len))
return tipc_withdraw(portref, 0, NULL);
if (uaddr_len < sizeof(struct sockaddr_tipc))
return -EINVAL;
if (addr->family != AF_TIPC)
return -EAFNOSUPPORT;
if (addr->addrtype == TIPC_ADDR_NAME)
addr->addr.nameseq.upper = addr->addr.nameseq.lower;
else if (addr->addrtype != TIPC_ADDR_NAMESEQ)
return -EAFNOSUPPORT;
if (addr->addr.nameseq.type < TIPC_RESERVED_TYPES)
return -EACCES;
return (addr->scope > 0) ?
tipc_publish(portref, addr->scope, &addr->addr.nameseq) :
tipc_withdraw(portref, -addr->scope, &addr->addr.nameseq);
}
/**
* get_name - get port ID of socket or peer socket
* @sock: socket structure
* @uaddr: area for returned socket address
* @uaddr_len: area for returned length of socket address
* @peer: 0 = own ID, 1 = current peer ID, 2 = current/former peer ID
*
* Returns 0 on success, errno otherwise
*
* NOTE: This routine doesn't need to take the socket lock since it only
* accesses socket information that is unchanging (or which changes in
* a completely predictable manner).
*/
static int get_name(struct socket *sock, struct sockaddr *uaddr,
int *uaddr_len, int peer)
{
struct sockaddr_tipc *addr = (struct sockaddr_tipc *)uaddr;
struct tipc_sock *tsock = tipc_sk(sock->sk);
memset(addr, 0, sizeof(*addr));
if (peer) {
if ((sock->state != SS_CONNECTED) &&
((peer != 2) || (sock->state != SS_DISCONNECTING)))
return -ENOTCONN;
addr->addr.id.ref = tsock->peer_name.ref;
addr->addr.id.node = tsock->peer_name.node;
} else {
addr->addr.id.ref = tsock->p->ref;
addr->addr.id.node = tipc_own_addr;
}
*uaddr_len = sizeof(*addr);
addr->addrtype = TIPC_ADDR_ID;
addr->family = AF_TIPC;
addr->scope = 0;
addr->addr.name.domain = 0;
return 0;
}
/**
* poll - read and possibly block on pollmask
* @file: file structure associated with the socket
* @sock: socket for which to calculate the poll bits
* @wait: ???
*
* Returns pollmask value
*
* COMMENTARY:
* It appears that the usual socket locking mechanisms are not useful here
* since the pollmask info is potentially out-of-date the moment this routine
* exits. TCP and other protocols seem to rely on higher level poll routines
* to handle any preventable race conditions, so TIPC will do the same ...
*
* TIPC sets the returned events as follows:
*
* socket state flags set
* ------------ ---------
* unconnected no read flags
* POLLOUT if port is not congested
*
* connecting POLLIN/POLLRDNORM if ACK/NACK in rx queue
* no write flags
*
* connected POLLIN/POLLRDNORM if data in rx queue
* POLLOUT if port is not congested
*
* disconnecting POLLIN/POLLRDNORM/POLLHUP
* no write flags
*
* listening POLLIN if SYN in rx queue
* no write flags
*
* ready POLLIN/POLLRDNORM if data in rx queue
* [connectionless] POLLOUT (since port cannot be congested)
*
* IMPORTANT: The fact that a read or write operation is indicated does NOT
* imply that the operation will succeed, merely that it should be performed
* and will not block.
*/
static unsigned int poll(struct file *file, struct socket *sock,
poll_table *wait)
{
struct sock *sk = sock->sk;
u32 mask = 0;
sock_poll_wait(file, sk_sleep(sk), wait);
switch ((int)sock->state) {
case SS_UNCONNECTED:
if (!tipc_sk_port(sk)->congested)
mask |= POLLOUT;
break;
case SS_READY:
case SS_CONNECTED:
if (!tipc_sk_port(sk)->congested)
mask |= POLLOUT;
/* fall thru' */
case SS_CONNECTING:
case SS_LISTENING:
if (!skb_queue_empty(&sk->sk_receive_queue))
mask |= (POLLIN | POLLRDNORM);
break;
case SS_DISCONNECTING:
mask = (POLLIN | POLLRDNORM | POLLHUP);
break;
}
return mask;
}
/**
* dest_name_check - verify user is permitted to send to specified port name
* @dest: destination address
* @m: descriptor for message to be sent
*
* Prevents restricted configuration commands from being issued by
* unauthorized users.
*
* Returns 0 if permission is granted, otherwise errno
*/
static int dest_name_check(struct sockaddr_tipc *dest, struct msghdr *m)
{
struct tipc_cfg_msg_hdr hdr;
if (likely(dest->addr.name.name.type >= TIPC_RESERVED_TYPES))
return 0;
if (likely(dest->addr.name.name.type == TIPC_TOP_SRV))
return 0;
if (likely(dest->addr.name.name.type != TIPC_CFG_SRV))
return -EACCES;
if (!m->msg_iovlen || (m->msg_iov[0].iov_len < sizeof(hdr)))
return -EMSGSIZE;
if (copy_from_user(&hdr, m->msg_iov[0].iov_base, sizeof(hdr)))
return -EFAULT;
if ((ntohs(hdr.tcm_type) & 0xC000) && (!capable(CAP_NET_ADMIN)))
return -EACCES;
return 0;
}
/**
* send_msg - send message in connectionless manner
* @iocb: if NULL, indicates that socket lock is already held
* @sock: socket structure
* @m: message to send
* @total_len: length of message
*
* Message must have an destination specified explicitly.
* Used for SOCK_RDM and SOCK_DGRAM messages,
* and for 'SYN' messages on SOCK_SEQPACKET and SOCK_STREAM connections.
* (Note: 'SYN+' is prohibited on SOCK_STREAM.)
*
* Returns the number of bytes sent on success, or errno otherwise
*/
static int send_msg(struct kiocb *iocb, struct socket *sock,
struct msghdr *m, size_t total_len)
{
struct sock *sk = sock->sk;
struct tipc_port *tport = tipc_sk_port(sk);
struct sockaddr_tipc *dest = (struct sockaddr_tipc *)m->msg_name;
int needs_conn;
long timeout_val;
int res = -EINVAL;
if (unlikely(!dest))
return -EDESTADDRREQ;
if (unlikely((m->msg_namelen < sizeof(*dest)) ||
(dest->family != AF_TIPC)))
return -EINVAL;
if (total_len > TIPC_MAX_USER_MSG_SIZE)
return -EMSGSIZE;
if (iocb)
lock_sock(sk);
needs_conn = (sock->state != SS_READY);
if (unlikely(needs_conn)) {
if (sock->state == SS_LISTENING) {
res = -EPIPE;
goto exit;
}
if (sock->state != SS_UNCONNECTED) {
res = -EISCONN;
goto exit;
}
if ((tport->published) ||
((sock->type == SOCK_STREAM) && (total_len != 0))) {
res = -EOPNOTSUPP;
goto exit;
}
if (dest->addrtype == TIPC_ADDR_NAME) {
tport->conn_type = dest->addr.name.name.type;
tport->conn_instance = dest->addr.name.name.instance;
}
/* Abort any pending connection attempts (very unlikely) */
reject_rx_queue(sk);
}
timeout_val = sock_sndtimeo(sk, m->msg_flags & MSG_DONTWAIT);
do {
if (dest->addrtype == TIPC_ADDR_NAME) {
res = dest_name_check(dest, m);
if (res)
break;
res = tipc_send2name(tport->ref,
&dest->addr.name.name,
dest->addr.name.domain,
m->msg_iovlen,
m->msg_iov,
total_len);
} else if (dest->addrtype == TIPC_ADDR_ID) {
res = tipc_send2port(tport->ref,
&dest->addr.id,
m->msg_iovlen,
m->msg_iov,
total_len);
} else if (dest->addrtype == TIPC_ADDR_MCAST) {
if (needs_conn) {
res = -EOPNOTSUPP;
break;
}
res = dest_name_check(dest, m);
if (res)
break;
res = tipc_multicast(tport->ref,
&dest->addr.nameseq,
m->msg_iovlen,
m->msg_iov,
total_len);
}
if (likely(res != -ELINKCONG)) {
if (needs_conn && (res >= 0))
sock->state = SS_CONNECTING;
break;
}
if (timeout_val <= 0L) {
res = timeout_val ? timeout_val : -EWOULDBLOCK;
break;
}
release_sock(sk);
timeout_val = wait_event_interruptible_timeout(*sk_sleep(sk),
!tport->congested, timeout_val);
lock_sock(sk);
} while (1);
exit:
if (iocb)
release_sock(sk);
return res;
}
/**
* send_packet - send a connection-oriented message
* @iocb: if NULL, indicates that socket lock is already held
* @sock: socket structure
* @m: message to send
* @total_len: length of message
*
* Used for SOCK_SEQPACKET messages and SOCK_STREAM data.
*
* Returns the number of bytes sent on success, or errno otherwise
*/
static int send_packet(struct kiocb *iocb, struct socket *sock,
struct msghdr *m, size_t total_len)
{
struct sock *sk = sock->sk;
struct tipc_port *tport = tipc_sk_port(sk);
struct sockaddr_tipc *dest = (struct sockaddr_tipc *)m->msg_name;
long timeout_val;
int res;
/* Handle implied connection establishment */
if (unlikely(dest))
return send_msg(iocb, sock, m, total_len);
if (total_len > TIPC_MAX_USER_MSG_SIZE)
return -EMSGSIZE;
if (iocb)
lock_sock(sk);
timeout_val = sock_sndtimeo(sk, m->msg_flags & MSG_DONTWAIT);
do {
if (unlikely(sock->state != SS_CONNECTED)) {
if (sock->state == SS_DISCONNECTING)
res = -EPIPE;
else
res = -ENOTCONN;
break;
}
res = tipc_send(tport->ref, m->msg_iovlen, m->msg_iov,
total_len);
if (likely(res != -ELINKCONG))
break;
if (timeout_val <= 0L) {
res = timeout_val ? timeout_val : -EWOULDBLOCK;
break;
}
release_sock(sk);
timeout_val = wait_event_interruptible_timeout(*sk_sleep(sk),
(!tport->congested || !tport->connected), timeout_val);
lock_sock(sk);
} while (1);
if (iocb)
release_sock(sk);
return res;
}
/**
* send_stream - send stream-oriented data
* @iocb: (unused)
* @sock: socket structure
* @m: data to send
* @total_len: total length of data to be sent
*
* Used for SOCK_STREAM data.
*
* Returns the number of bytes sent on success (or partial success),
* or errno if no data sent
*/
static int send_stream(struct kiocb *iocb, struct socket *sock,
struct msghdr *m, size_t total_len)
{
struct sock *sk = sock->sk;
struct tipc_port *tport = tipc_sk_port(sk);
struct msghdr my_msg;
struct iovec my_iov;
struct iovec *curr_iov;
int curr_iovlen;
char __user *curr_start;
u32 hdr_size;
int curr_left;
int bytes_to_send;
int bytes_sent;
int res;
lock_sock(sk);
/* Handle special cases where there is no connection */
if (unlikely(sock->state != SS_CONNECTED)) {
if (sock->state == SS_UNCONNECTED) {
res = send_packet(NULL, sock, m, total_len);
goto exit;
} else if (sock->state == SS_DISCONNECTING) {
res = -EPIPE;
goto exit;
} else {
res = -ENOTCONN;
goto exit;
}
}
if (unlikely(m->msg_name)) {
res = -EISCONN;
goto exit;
}
if (total_len > (unsigned int)INT_MAX) {
res = -EMSGSIZE;
goto exit;
}
/*
* Send each iovec entry using one or more messages
*
* Note: This algorithm is good for the most likely case
* (i.e. one large iovec entry), but could be improved to pass sets
* of small iovec entries into send_packet().
*/
curr_iov = m->msg_iov;
curr_iovlen = m->msg_iovlen;
my_msg.msg_iov = &my_iov;
my_msg.msg_iovlen = 1;
my_msg.msg_flags = m->msg_flags;
my_msg.msg_name = NULL;
bytes_sent = 0;
hdr_size = msg_hdr_sz(&tport->phdr);
while (curr_iovlen--) {
curr_start = curr_iov->iov_base;
curr_left = curr_iov->iov_len;
while (curr_left) {
bytes_to_send = tport->max_pkt - hdr_size;
if (bytes_to_send > TIPC_MAX_USER_MSG_SIZE)
bytes_to_send = TIPC_MAX_USER_MSG_SIZE;
if (curr_left < bytes_to_send)
bytes_to_send = curr_left;
my_iov.iov_base = curr_start;
my_iov.iov_len = bytes_to_send;
res = send_packet(NULL, sock, &my_msg, bytes_to_send);
if (res < 0) {
if (bytes_sent)
res = bytes_sent;
goto exit;
}
curr_left -= bytes_to_send;
curr_start += bytes_to_send;
bytes_sent += bytes_to_send;
}
curr_iov++;
}
res = bytes_sent;
exit:
release_sock(sk);
return res;
}
/**
* auto_connect - complete connection setup to a remote port
* @sock: socket structure
* @msg: peer's response message
*
* Returns 0 on success, errno otherwise
*/
static int auto_connect(struct socket *sock, struct tipc_msg *msg)
{
struct tipc_sock *tsock = tipc_sk(sock->sk);
tipc: introduce non-blocking socket connect TIPC has so far only supported blocking connect(), meaning that a call to connect() doesn't return until either the connection is fully established, or an error occurs. This has proved insufficient for many users, so we now introduce non-blocking connect(), analogous to how this is done in TCP and other protocols. With this feature, if a connection cannot be established instantly, connect() will return the error code "-EINPROGRESS". If the user later calls connect() again, he will either have the return code "-EALREADY" or "-EISCONN", depending on whether the connection has been established or not. The user must have explicitly set the socket to be non-blocking (SOCK_NONBLOCK or O_NONBLOCK, depending on method used), so unless for some reason they had set this already (the socket would anyway remain blocking in current TIPC) this change should be completely backwards compatible. It is also now possible to call select() or poll() to wait for the completion of a connection. An effect of the above is that the actual completion of a connection may now be performed asynchronously, independent of the calls from user space. Therefore, we now execute this code in BH context, in the function filter_rcv(), which is executed upon reception of messages in the socket. Signed-off-by: Ying Xue <ying.xue@windriver.com> Signed-off-by: Jon Maloy <jon.maloy@ericsson.com> [PG: minor refactoring for improved connect/disconnect function names] Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
2012-11-30 06:51:19 +07:00
struct tipc_port *p_ptr;
tsock->peer_name.ref = msg_origport(msg);
tsock->peer_name.node = msg_orignode(msg);
tipc: introduce non-blocking socket connect TIPC has so far only supported blocking connect(), meaning that a call to connect() doesn't return until either the connection is fully established, or an error occurs. This has proved insufficient for many users, so we now introduce non-blocking connect(), analogous to how this is done in TCP and other protocols. With this feature, if a connection cannot be established instantly, connect() will return the error code "-EINPROGRESS". If the user later calls connect() again, he will either have the return code "-EALREADY" or "-EISCONN", depending on whether the connection has been established or not. The user must have explicitly set the socket to be non-blocking (SOCK_NONBLOCK or O_NONBLOCK, depending on method used), so unless for some reason they had set this already (the socket would anyway remain blocking in current TIPC) this change should be completely backwards compatible. It is also now possible to call select() or poll() to wait for the completion of a connection. An effect of the above is that the actual completion of a connection may now be performed asynchronously, independent of the calls from user space. Therefore, we now execute this code in BH context, in the function filter_rcv(), which is executed upon reception of messages in the socket. Signed-off-by: Ying Xue <ying.xue@windriver.com> Signed-off-by: Jon Maloy <jon.maloy@ericsson.com> [PG: minor refactoring for improved connect/disconnect function names] Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
2012-11-30 06:51:19 +07:00
p_ptr = tipc_port_deref(tsock->p->ref);
if (!p_ptr)
return -EINVAL;
__tipc_connect(tsock->p->ref, p_ptr, &tsock->peer_name);
if (msg_importance(msg) > TIPC_CRITICAL_IMPORTANCE)
return -EINVAL;
msg_set_importance(&p_ptr->phdr, (u32)msg_importance(msg));
sock->state = SS_CONNECTED;
return 0;
}
/**
* set_orig_addr - capture sender's address for received message
* @m: descriptor for message info
* @msg: received message header
*
* Note: Address is not captured if not requested by receiver.
*/
static void set_orig_addr(struct msghdr *m, struct tipc_msg *msg)
{
struct sockaddr_tipc *addr = (struct sockaddr_tipc *)m->msg_name;
if (addr) {
addr->family = AF_TIPC;
addr->addrtype = TIPC_ADDR_ID;
addr->addr.id.ref = msg_origport(msg);
addr->addr.id.node = msg_orignode(msg);
addr->addr.name.domain = 0; /* could leave uninitialized */
addr->scope = 0; /* could leave uninitialized */
m->msg_namelen = sizeof(struct sockaddr_tipc);
}
}
/**
* anc_data_recv - optionally capture ancillary data for received message
* @m: descriptor for message info
* @msg: received message header
* @tport: TIPC port associated with message
*
* Note: Ancillary data is not captured if not requested by receiver.
*
* Returns 0 if successful, otherwise errno
*/
static int anc_data_recv(struct msghdr *m, struct tipc_msg *msg,
struct tipc_port *tport)
{
u32 anc_data[3];
u32 err;
u32 dest_type;
int has_name;
int res;
if (likely(m->msg_controllen == 0))
return 0;
/* Optionally capture errored message object(s) */
err = msg ? msg_errcode(msg) : 0;
if (unlikely(err)) {
anc_data[0] = err;
anc_data[1] = msg_data_sz(msg);
res = put_cmsg(m, SOL_TIPC, TIPC_ERRINFO, 8, anc_data);
if (res)
return res;
if (anc_data[1]) {
res = put_cmsg(m, SOL_TIPC, TIPC_RETDATA, anc_data[1],
msg_data(msg));
if (res)
return res;
}
}
/* Optionally capture message destination object */
dest_type = msg ? msg_type(msg) : TIPC_DIRECT_MSG;
switch (dest_type) {
case TIPC_NAMED_MSG:
has_name = 1;
anc_data[0] = msg_nametype(msg);
anc_data[1] = msg_namelower(msg);
anc_data[2] = msg_namelower(msg);
break;
case TIPC_MCAST_MSG:
has_name = 1;
anc_data[0] = msg_nametype(msg);
anc_data[1] = msg_namelower(msg);
anc_data[2] = msg_nameupper(msg);
break;
case TIPC_CONN_MSG:
has_name = (tport->conn_type != 0);
anc_data[0] = tport->conn_type;
anc_data[1] = tport->conn_instance;
anc_data[2] = tport->conn_instance;
break;
default:
has_name = 0;
}
if (has_name) {
res = put_cmsg(m, SOL_TIPC, TIPC_DESTNAME, 12, anc_data);
if (res)
return res;
}
return 0;
}
/**
* recv_msg - receive packet-oriented message
* @iocb: (unused)
* @m: descriptor for message info
* @buf_len: total size of user buffer area
* @flags: receive flags
*
* Used for SOCK_DGRAM, SOCK_RDM, and SOCK_SEQPACKET messages.
* If the complete message doesn't fit in user area, truncate it.
*
* Returns size of returned message data, errno otherwise
*/
static int recv_msg(struct kiocb *iocb, struct socket *sock,
struct msghdr *m, size_t buf_len, int flags)
{
struct sock *sk = sock->sk;
struct tipc_port *tport = tipc_sk_port(sk);
struct sk_buff *buf;
struct tipc_msg *msg;
long timeout;
unsigned int sz;
u32 err;
int res;
/* Catch invalid receive requests */
if (unlikely(!buf_len))
return -EINVAL;
lock_sock(sk);
if (unlikely(sock->state == SS_UNCONNECTED)) {
res = -ENOTCONN;
goto exit;
}
timeout = sock_rcvtimeo(sk, flags & MSG_DONTWAIT);
restart:
/* Look for a message in receive queue; wait if necessary */
while (skb_queue_empty(&sk->sk_receive_queue)) {
if (sock->state == SS_DISCONNECTING) {
res = -ENOTCONN;
goto exit;
}
if (timeout <= 0L) {
res = timeout ? timeout : -EWOULDBLOCK;
goto exit;
}
release_sock(sk);
timeout = wait_event_interruptible_timeout(*sk_sleep(sk),
tipc_rx_ready(sock),
timeout);
lock_sock(sk);
}
/* Look at first message in receive queue */
buf = skb_peek(&sk->sk_receive_queue);
msg = buf_msg(buf);
sz = msg_data_sz(msg);
err = msg_errcode(msg);
/* Discard an empty non-errored message & try again */
if ((!sz) && (!err)) {
advance_rx_queue(sk);
goto restart;
}
/* Capture sender's address (optional) */
set_orig_addr(m, msg);
/* Capture ancillary data (optional) */
res = anc_data_recv(m, msg, tport);
if (res)
goto exit;
/* Capture message data (if valid) & compute return value (always) */
if (!err) {
if (unlikely(buf_len < sz)) {
sz = buf_len;
m->msg_flags |= MSG_TRUNC;
}
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) ||
((err == TIPC_CONN_SHUTDOWN) || m->msg_control))
res = 0;
else
res = -ECONNRESET;
}
/* Consume received message (optional) */
if (likely(!(flags & MSG_PEEK))) {
if ((sock->state != SS_READY) &&
(++tport->conn_unacked >= TIPC_FLOW_CONTROL_WIN))
tipc_acknowledge(tport->ref, tport->conn_unacked);
advance_rx_queue(sk);
}
exit:
release_sock(sk);
return res;
}
/**
* recv_stream - receive stream-oriented data
* @iocb: (unused)
* @m: descriptor for message info
* @buf_len: total size of user buffer area
* @flags: receive flags
*
* Used for SOCK_STREAM messages only. If not enough data is available
* will optionally wait for more; never truncates data.
*
* Returns size of returned message data, errno otherwise
*/
static int recv_stream(struct kiocb *iocb, struct socket *sock,
struct msghdr *m, size_t buf_len, int flags)
{
struct sock *sk = sock->sk;
struct tipc_port *tport = tipc_sk_port(sk);
struct sk_buff *buf;
struct tipc_msg *msg;
long timeout;
unsigned int sz;
int sz_to_copy, target, needed;
int sz_copied = 0;
u32 err;
int res = 0;
/* Catch invalid receive attempts */
if (unlikely(!buf_len))
return -EINVAL;
lock_sock(sk);
if (unlikely((sock->state == SS_UNCONNECTED) ||
(sock->state == SS_CONNECTING))) {
res = -ENOTCONN;
goto exit;
}
target = sock_rcvlowat(sk, flags & MSG_WAITALL, buf_len);
timeout = sock_rcvtimeo(sk, flags & MSG_DONTWAIT);
restart:
/* Look for a message in receive queue; wait if necessary */
while (skb_queue_empty(&sk->sk_receive_queue)) {
if (sock->state == SS_DISCONNECTING) {
res = -ENOTCONN;
goto exit;
}
if (timeout <= 0L) {
res = timeout ? timeout : -EWOULDBLOCK;
goto exit;
}
release_sock(sk);
timeout = wait_event_interruptible_timeout(*sk_sleep(sk),
tipc_rx_ready(sock),
timeout);
lock_sock(sk);
}
/* Look at first message in receive queue */
buf = skb_peek(&sk->sk_receive_queue);
msg = buf_msg(buf);
sz = msg_data_sz(msg);
err = msg_errcode(msg);
/* Discard an empty non-errored message & try again */
if ((!sz) && (!err)) {
advance_rx_queue(sk);
goto restart;
}
/* Optionally capture sender's address & ancillary data of first msg */
if (sz_copied == 0) {
set_orig_addr(m, msg);
res = anc_data_recv(m, msg, tport);
if (res)
goto exit;
}
/* Capture message data (if valid) & compute return value (always) */
if (!err) {
u32 offset = (u32)(unsigned long)(TIPC_SKB_CB(buf)->handle);
sz -= offset;
needed = (buf_len - sz_copied);
sz_to_copy = (sz <= needed) ? sz : needed;
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 =
(void *)(unsigned long)(offset + sz_to_copy);
goto exit;
}
} else {
if (sz_copied != 0)
goto exit; /* can't add error msg to valid data */
if ((err == TIPC_CONN_SHUTDOWN) || m->msg_control)
res = 0;
else
res = -ECONNRESET;
}
/* Consume received message (optional) */
if (likely(!(flags & MSG_PEEK))) {
if (unlikely(++tport->conn_unacked >= TIPC_FLOW_CONTROL_WIN))
tipc_acknowledge(tport->ref, tport->conn_unacked);
advance_rx_queue(sk);
}
/* Loop around if more data is required */
if ((sz_copied < buf_len) && /* didn't get all requested data */
(!skb_queue_empty(&sk->sk_receive_queue) ||
(sz_copied < target)) && /* and more is ready or required */
(!(flags & MSG_PEEK)) && /* and aren't just peeking at data */
(!err)) /* and haven't reached a FIN */
goto restart;
exit:
release_sock(sk);
return sz_copied ? sz_copied : res;
}
/**
* 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, int len)
{
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
* @tsock: TIPC socket
* @msg: message
*
* Returns TIPC error status code and socket error status code
* once it encounters some errors
*/
static u32 filter_connect(struct tipc_sock *tsock, struct sk_buff **buf)
{
struct socket *sock = tsock->sk.sk_socket;
struct tipc_msg *msg = buf_msg(*buf);
tipc: introduce non-blocking socket connect TIPC has so far only supported blocking connect(), meaning that a call to connect() doesn't return until either the connection is fully established, or an error occurs. This has proved insufficient for many users, so we now introduce non-blocking connect(), analogous to how this is done in TCP and other protocols. With this feature, if a connection cannot be established instantly, connect() will return the error code "-EINPROGRESS". If the user later calls connect() again, he will either have the return code "-EALREADY" or "-EISCONN", depending on whether the connection has been established or not. The user must have explicitly set the socket to be non-blocking (SOCK_NONBLOCK or O_NONBLOCK, depending on method used), so unless for some reason they had set this already (the socket would anyway remain blocking in current TIPC) this change should be completely backwards compatible. It is also now possible to call select() or poll() to wait for the completion of a connection. An effect of the above is that the actual completion of a connection may now be performed asynchronously, independent of the calls from user space. Therefore, we now execute this code in BH context, in the function filter_rcv(), which is executed upon reception of messages in the socket. Signed-off-by: Ying Xue <ying.xue@windriver.com> Signed-off-by: Jon Maloy <jon.maloy@ericsson.com> [PG: minor refactoring for improved connect/disconnect function names] Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
2012-11-30 06:51:19 +07:00
struct sock *sk = &tsock->sk;
u32 retval = TIPC_ERR_NO_PORT;
tipc: introduce non-blocking socket connect TIPC has so far only supported blocking connect(), meaning that a call to connect() doesn't return until either the connection is fully established, or an error occurs. This has proved insufficient for many users, so we now introduce non-blocking connect(), analogous to how this is done in TCP and other protocols. With this feature, if a connection cannot be established instantly, connect() will return the error code "-EINPROGRESS". If the user later calls connect() again, he will either have the return code "-EALREADY" or "-EISCONN", depending on whether the connection has been established or not. The user must have explicitly set the socket to be non-blocking (SOCK_NONBLOCK or O_NONBLOCK, depending on method used), so unless for some reason they had set this already (the socket would anyway remain blocking in current TIPC) this change should be completely backwards compatible. It is also now possible to call select() or poll() to wait for the completion of a connection. An effect of the above is that the actual completion of a connection may now be performed asynchronously, independent of the calls from user space. Therefore, we now execute this code in BH context, in the function filter_rcv(), which is executed upon reception of messages in the socket. Signed-off-by: Ying Xue <ying.xue@windriver.com> Signed-off-by: Jon Maloy <jon.maloy@ericsson.com> [PG: minor refactoring for improved connect/disconnect function names] Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
2012-11-30 06:51:19 +07:00
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(tsock->p, msg)) {
if (unlikely(msg_errcode(msg))) {
sock->state = SS_DISCONNECTING;
__tipc_disconnect(tsock->p);
}
retval = TIPC_OK;
}
break;
case SS_CONNECTING:
/* Accept only ACK or NACK message */
tipc: introduce non-blocking socket connect TIPC has so far only supported blocking connect(), meaning that a call to connect() doesn't return until either the connection is fully established, or an error occurs. This has proved insufficient for many users, so we now introduce non-blocking connect(), analogous to how this is done in TCP and other protocols. With this feature, if a connection cannot be established instantly, connect() will return the error code "-EINPROGRESS". If the user later calls connect() again, he will either have the return code "-EALREADY" or "-EISCONN", depending on whether the connection has been established or not. The user must have explicitly set the socket to be non-blocking (SOCK_NONBLOCK or O_NONBLOCK, depending on method used), so unless for some reason they had set this already (the socket would anyway remain blocking in current TIPC) this change should be completely backwards compatible. It is also now possible to call select() or poll() to wait for the completion of a connection. An effect of the above is that the actual completion of a connection may now be performed asynchronously, independent of the calls from user space. Therefore, we now execute this code in BH context, in the function filter_rcv(), which is executed upon reception of messages in the socket. Signed-off-by: Ying Xue <ying.xue@windriver.com> Signed-off-by: Jon Maloy <jon.maloy@ericsson.com> [PG: minor refactoring for improved connect/disconnect function names] Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
2012-11-30 06:51:19 +07:00
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(sock, msg);
if (res) {
sock->state = SS_DISCONNECTING;
sk->sk_err = res;
retval = TIPC_OK;
tipc: introduce non-blocking socket connect TIPC has so far only supported blocking connect(), meaning that a call to connect() doesn't return until either the connection is fully established, or an error occurs. This has proved insufficient for many users, so we now introduce non-blocking connect(), analogous to how this is done in TCP and other protocols. With this feature, if a connection cannot be established instantly, connect() will return the error code "-EINPROGRESS". If the user later calls connect() again, he will either have the return code "-EALREADY" or "-EISCONN", depending on whether the connection has been established or not. The user must have explicitly set the socket to be non-blocking (SOCK_NONBLOCK or O_NONBLOCK, depending on method used), so unless for some reason they had set this already (the socket would anyway remain blocking in current TIPC) this change should be completely backwards compatible. It is also now possible to call select() or poll() to wait for the completion of a connection. An effect of the above is that the actual completion of a connection may now be performed asynchronously, independent of the calls from user space. Therefore, we now execute this code in BH context, in the function filter_rcv(), which is executed upon reception of messages in the socket. Signed-off-by: Ying Xue <ying.xue@windriver.com> Signed-off-by: Jon Maloy <jon.maloy@ericsson.com> [PG: minor refactoring for improved connect/disconnect function names] Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
2012-11-30 06:51:19 +07:00
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;
}
/**
* 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 (5MB)
* TIPC_MEDIUM_IMPORTANCE (10MB)
* TIPC_HIGH_IMPORTANCE (20MB)
* TIPC_CRITICAL_IMPORTANCE (40MB)
*
* Returns overload limit according to corresponding message importance
*/
static unsigned int rcvbuf_limit(struct sock *sk, struct sk_buff *buf)
{
struct tipc_msg *msg = buf_msg(buf);
unsigned int limit;
if (msg_connected(msg))
limit = CONN_OVERLOAD_LIMIT;
else
limit = sk->sk_rcvbuf << (msg_importance(msg) + 5);
return limit;
}
/**
* filter_rcv - validate incoming message
* @sk: socket
* @buf: message
*
* Enqueues message on receive queue if acceptable; optionally handles
* disconnect indication for a connected socket.
*
* Called with socket lock already taken; port lock may also be taken.
*
* Returns TIPC error status code (TIPC_OK if message is not to be rejected)
*/
static u32 filter_rcv(struct sock *sk, struct sk_buff *buf)
{
struct socket *sock = sk->sk_socket;
struct tipc_msg *msg = buf_msg(buf);
unsigned int limit = rcvbuf_limit(sk, buf);
u32 res = TIPC_OK;
/* Reject message if it is wrong sort of message for socket */
if (msg_type(msg) > TIPC_DIRECT_MSG)
return TIPC_ERR_NO_PORT;
if (sock->state == SS_READY) {
if (msg_connected(msg))
return TIPC_ERR_NO_PORT;
} else {
res = filter_connect(tipc_sk(sk), &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;
/* Enqueue message */
TIPC_SKB_CB(buf)->handle = 0;
__skb_queue_tail(&sk->sk_receive_queue, buf);
skb_set_owner_r(buf, sk);
sk->sk_data_ready(sk, 0);
return TIPC_OK;
}
/**
* backlog_rcv - handle incoming message from backlog queue
* @sk: socket
* @buf: message
*
* Caller must hold socket lock, but not port lock.
*
* Returns 0
*/
static int backlog_rcv(struct sock *sk, struct sk_buff *buf)
{
u32 res;
res = filter_rcv(sk, buf);
if (res)
tipc_reject_msg(buf, res);
return 0;
}
/**
* dispatch - handle incoming message
* @tport: TIPC port that received message
* @buf: message
*
* Called with port lock already taken.
*
* Returns TIPC error status code (TIPC_OK if message is not to be rejected)
*/
static u32 dispatch(struct tipc_port *tport, struct sk_buff *buf)
{
struct sock *sk = (struct sock *)tport->usr_handle;
u32 res;
/*
* Process message if socket is unlocked; otherwise add to backlog queue
*
* This code is based on sk_receive_skb(), but must be distinct from it
* since a TIPC-specific filter/reject mechanism is utilized
*/
bh_lock_sock(sk);
if (!sock_owned_by_user(sk)) {
res = filter_rcv(sk, buf);
} else {
if (sk_add_backlog(sk, buf, rcvbuf_limit(sk, buf)))
res = TIPC_ERR_OVERLOAD;
else
res = TIPC_OK;
}
bh_unlock_sock(sk);
return res;
}
/**
* wakeupdispatch - wake up port after congestion
* @tport: port to wakeup
*
* Called with port lock already taken.
*/
static void wakeupdispatch(struct tipc_port *tport)
{
struct sock *sk = (struct sock *)tport->usr_handle;
sk->sk_write_space(sk);
}
/**
* connect - establish a connection to another TIPC port
* @sock: socket structure
* @dest: socket address for destination port
* @destlen: size of socket address data structure
* @flags: file-related flags associated with socket
*
* Returns 0 on success, errno otherwise
*/
static int connect(struct socket *sock, struct sockaddr *dest, int destlen,
int flags)
{
struct sock *sk = sock->sk;
struct sockaddr_tipc *dst = (struct sockaddr_tipc *)dest;
struct msghdr m = {NULL,};
unsigned int timeout;
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;
}
/*
* 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;
}
tipc: introduce non-blocking socket connect TIPC has so far only supported blocking connect(), meaning that a call to connect() doesn't return until either the connection is fully established, or an error occurs. This has proved insufficient for many users, so we now introduce non-blocking connect(), analogous to how this is done in TCP and other protocols. With this feature, if a connection cannot be established instantly, connect() will return the error code "-EINPROGRESS". If the user later calls connect() again, he will either have the return code "-EALREADY" or "-EISCONN", depending on whether the connection has been established or not. The user must have explicitly set the socket to be non-blocking (SOCK_NONBLOCK or O_NONBLOCK, depending on method used), so unless for some reason they had set this already (the socket would anyway remain blocking in current TIPC) this change should be completely backwards compatible. It is also now possible to call select() or poll() to wait for the completion of a connection. An effect of the above is that the actual completion of a connection may now be performed asynchronously, independent of the calls from user space. Therefore, we now execute this code in BH context, in the function filter_rcv(), which is executed upon reception of messages in the socket. Signed-off-by: Ying Xue <ying.xue@windriver.com> Signed-off-by: Jon Maloy <jon.maloy@ericsson.com> [PG: minor refactoring for improved connect/disconnect function names] Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
2012-11-30 06:51:19 +07:00
timeout = (flags & O_NONBLOCK) ? 0 : tipc_sk(sk)->conn_timeout;
tipc: introduce non-blocking socket connect TIPC has so far only supported blocking connect(), meaning that a call to connect() doesn't return until either the connection is fully established, or an error occurs. This has proved insufficient for many users, so we now introduce non-blocking connect(), analogous to how this is done in TCP and other protocols. With this feature, if a connection cannot be established instantly, connect() will return the error code "-EINPROGRESS". If the user later calls connect() again, he will either have the return code "-EALREADY" or "-EISCONN", depending on whether the connection has been established or not. The user must have explicitly set the socket to be non-blocking (SOCK_NONBLOCK or O_NONBLOCK, depending on method used), so unless for some reason they had set this already (the socket would anyway remain blocking in current TIPC) this change should be completely backwards compatible. It is also now possible to call select() or poll() to wait for the completion of a connection. An effect of the above is that the actual completion of a connection may now be performed asynchronously, independent of the calls from user space. Therefore, we now execute this code in BH context, in the function filter_rcv(), which is executed upon reception of messages in the socket. Signed-off-by: Ying Xue <ying.xue@windriver.com> Signed-off-by: Jon Maloy <jon.maloy@ericsson.com> [PG: minor refactoring for improved connect/disconnect function names] Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
2012-11-30 06:51:19 +07:00
switch (sock->state) {
case SS_UNCONNECTED:
/* Send a 'SYN-' to destination */
m.msg_name = dest;
m.msg_namelen = destlen;
/* If connect is in non-blocking case, set MSG_DONTWAIT to
* indicate send_msg() is never blocked.
*/
if (!timeout)
m.msg_flags = MSG_DONTWAIT;
res = send_msg(NULL, sock, &m, 0);
if ((res < 0) && (res != -EWOULDBLOCK))
goto exit;
/* Just entered SS_CONNECTING state; the only
* difference is that return value in non-blocking
* case is EINPROGRESS, rather than EALREADY.
*/
res = -EINPROGRESS;
break;
case SS_CONNECTING:
res = -EALREADY;
break;
case SS_CONNECTED:
res = -EISCONN;
break;
default:
res = -EINVAL;
goto exit;
tipc: introduce non-blocking socket connect TIPC has so far only supported blocking connect(), meaning that a call to connect() doesn't return until either the connection is fully established, or an error occurs. This has proved insufficient for many users, so we now introduce non-blocking connect(), analogous to how this is done in TCP and other protocols. With this feature, if a connection cannot be established instantly, connect() will return the error code "-EINPROGRESS". If the user later calls connect() again, he will either have the return code "-EALREADY" or "-EISCONN", depending on whether the connection has been established or not. The user must have explicitly set the socket to be non-blocking (SOCK_NONBLOCK or O_NONBLOCK, depending on method used), so unless for some reason they had set this already (the socket would anyway remain blocking in current TIPC) this change should be completely backwards compatible. It is also now possible to call select() or poll() to wait for the completion of a connection. An effect of the above is that the actual completion of a connection may now be performed asynchronously, independent of the calls from user space. Therefore, we now execute this code in BH context, in the function filter_rcv(), which is executed upon reception of messages in the socket. Signed-off-by: Ying Xue <ying.xue@windriver.com> Signed-off-by: Jon Maloy <jon.maloy@ericsson.com> [PG: minor refactoring for improved connect/disconnect function names] Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
2012-11-30 06:51:19 +07:00
}
tipc: introduce non-blocking socket connect TIPC has so far only supported blocking connect(), meaning that a call to connect() doesn't return until either the connection is fully established, or an error occurs. This has proved insufficient for many users, so we now introduce non-blocking connect(), analogous to how this is done in TCP and other protocols. With this feature, if a connection cannot be established instantly, connect() will return the error code "-EINPROGRESS". If the user later calls connect() again, he will either have the return code "-EALREADY" or "-EISCONN", depending on whether the connection has been established or not. The user must have explicitly set the socket to be non-blocking (SOCK_NONBLOCK or O_NONBLOCK, depending on method used), so unless for some reason they had set this already (the socket would anyway remain blocking in current TIPC) this change should be completely backwards compatible. It is also now possible to call select() or poll() to wait for the completion of a connection. An effect of the above is that the actual completion of a connection may now be performed asynchronously, independent of the calls from user space. Therefore, we now execute this code in BH context, in the function filter_rcv(), which is executed upon reception of messages in the socket. Signed-off-by: Ying Xue <ying.xue@windriver.com> Signed-off-by: Jon Maloy <jon.maloy@ericsson.com> [PG: minor refactoring for improved connect/disconnect function names] Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
2012-11-30 06:51:19 +07:00
if (sock->state == SS_CONNECTING) {
if (!timeout)
goto exit;
tipc: introduce non-blocking socket connect TIPC has so far only supported blocking connect(), meaning that a call to connect() doesn't return until either the connection is fully established, or an error occurs. This has proved insufficient for many users, so we now introduce non-blocking connect(), analogous to how this is done in TCP and other protocols. With this feature, if a connection cannot be established instantly, connect() will return the error code "-EINPROGRESS". If the user later calls connect() again, he will either have the return code "-EALREADY" or "-EISCONN", depending on whether the connection has been established or not. The user must have explicitly set the socket to be non-blocking (SOCK_NONBLOCK or O_NONBLOCK, depending on method used), so unless for some reason they had set this already (the socket would anyway remain blocking in current TIPC) this change should be completely backwards compatible. It is also now possible to call select() or poll() to wait for the completion of a connection. An effect of the above is that the actual completion of a connection may now be performed asynchronously, independent of the calls from user space. Therefore, we now execute this code in BH context, in the function filter_rcv(), which is executed upon reception of messages in the socket. Signed-off-by: Ying Xue <ying.xue@windriver.com> Signed-off-by: Jon Maloy <jon.maloy@ericsson.com> [PG: minor refactoring for improved connect/disconnect function names] Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
2012-11-30 06:51:19 +07:00
/* Wait until an 'ACK' or 'RST' arrives, or a timeout occurs */
release_sock(sk);
res = wait_event_interruptible_timeout(*sk_sleep(sk),
sock->state != SS_CONNECTING,
timeout ? (long)msecs_to_jiffies(timeout)
: MAX_SCHEDULE_TIMEOUT);
lock_sock(sk);
if (res <= 0) {
if (res == 0)
res = -ETIMEDOUT;
else
tipc: introduce non-blocking socket connect TIPC has so far only supported blocking connect(), meaning that a call to connect() doesn't return until either the connection is fully established, or an error occurs. This has proved insufficient for many users, so we now introduce non-blocking connect(), analogous to how this is done in TCP and other protocols. With this feature, if a connection cannot be established instantly, connect() will return the error code "-EINPROGRESS". If the user later calls connect() again, he will either have the return code "-EALREADY" or "-EISCONN", depending on whether the connection has been established or not. The user must have explicitly set the socket to be non-blocking (SOCK_NONBLOCK or O_NONBLOCK, depending on method used), so unless for some reason they had set this already (the socket would anyway remain blocking in current TIPC) this change should be completely backwards compatible. It is also now possible to call select() or poll() to wait for the completion of a connection. An effect of the above is that the actual completion of a connection may now be performed asynchronously, independent of the calls from user space. Therefore, we now execute this code in BH context, in the function filter_rcv(), which is executed upon reception of messages in the socket. Signed-off-by: Ying Xue <ying.xue@windriver.com> Signed-off-by: Jon Maloy <jon.maloy@ericsson.com> [PG: minor refactoring for improved connect/disconnect function names] Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
2012-11-30 06:51:19 +07:00
; /* leave "res" unchanged */
goto exit;
}
}
tipc: introduce non-blocking socket connect TIPC has so far only supported blocking connect(), meaning that a call to connect() doesn't return until either the connection is fully established, or an error occurs. This has proved insufficient for many users, so we now introduce non-blocking connect(), analogous to how this is done in TCP and other protocols. With this feature, if a connection cannot be established instantly, connect() will return the error code "-EINPROGRESS". If the user later calls connect() again, he will either have the return code "-EALREADY" or "-EISCONN", depending on whether the connection has been established or not. The user must have explicitly set the socket to be non-blocking (SOCK_NONBLOCK or O_NONBLOCK, depending on method used), so unless for some reason they had set this already (the socket would anyway remain blocking in current TIPC) this change should be completely backwards compatible. It is also now possible to call select() or poll() to wait for the completion of a connection. An effect of the above is that the actual completion of a connection may now be performed asynchronously, independent of the calls from user space. Therefore, we now execute this code in BH context, in the function filter_rcv(), which is executed upon reception of messages in the socket. Signed-off-by: Ying Xue <ying.xue@windriver.com> Signed-off-by: Jon Maloy <jon.maloy@ericsson.com> [PG: minor refactoring for improved connect/disconnect function names] Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
2012-11-30 06:51:19 +07:00
if (unlikely(sock->state == SS_DISCONNECTING))
res = sock_error(sk);
else
res = 0;
exit:
release_sock(sk);
return res;
}
/**
* listen - allow socket to listen for incoming connections
* @sock: socket structure
* @len: (unused)
*
* Returns 0 on success, errno otherwise
*/
static int listen(struct socket *sock, int len)
{
struct sock *sk = sock->sk;
int res;
lock_sock(sk);
if (sock->state != SS_UNCONNECTED)
res = -EINVAL;
else {
sock->state = SS_LISTENING;
res = 0;
}
release_sock(sk);
return res;
}
/**
* accept - wait for connection request
* @sock: listening socket
* @newsock: new socket that is to be connected
* @flags: file-related flags associated with socket
*
* Returns 0 on success, errno otherwise
*/
static int accept(struct socket *sock, struct socket *new_sock, int flags)
{
struct sock *new_sk, *sk = sock->sk;
struct sk_buff *buf;
struct tipc_sock *new_tsock;
struct tipc_port *new_tport;
struct tipc_msg *msg;
u32 new_ref;
int res;
lock_sock(sk);
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;
}
buf = skb_peek(&sk->sk_receive_queue);
res = tipc_create(sock_net(sock->sk), new_sock, 0, 0);
if (res)
goto exit;
new_sk = new_sock->sk;
new_tsock = tipc_sk(new_sk);
new_tport = new_tsock->p;
new_ref = new_tport->ref;
msg = buf_msg(buf);
/* we lock on new_sk; but lockdep sees the lock on sk */
lock_sock_nested(new_sk, SINGLE_DEPTH_NESTING);
/*
* 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 */
new_tsock->peer_name.ref = msg_origport(msg);
new_tsock->peer_name.node = msg_orignode(msg);
tipc_connect(new_ref, &new_tsock->peer_name);
new_sock->state = SS_CONNECTED;
tipc_set_portimportance(new_ref, msg_importance(msg));
if (msg_named(msg)) {
new_tport->conn_type = msg_nametype(msg);
new_tport->conn_instance = msg_nameinst(msg);
}
/*
* Respond to 'SYN-' by discarding it & returning 'ACK'-.
* Respond to 'SYN+' by queuing it on new socket.
*/
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);
skb_set_owner_r(buf, new_sk);
}
release_sock(new_sk);
exit:
release_sock(sk);
return res;
}
/**
* shutdown - shutdown socket connection
* @sock: socket structure
* @how: direction to close (must be SHUT_RDWR)
*
* Terminates connection (if necessary), then purges socket's receive queue.
*
* Returns 0 on success, errno otherwise
*/
static int shutdown(struct socket *sock, int how)
{
struct sock *sk = sock->sk;
struct tipc_port *tport = tipc_sk_port(sk);
struct sk_buff *buf;
int res;
if (how != SHUT_RDWR)
return -EINVAL;
lock_sock(sk);
switch (sock->state) {
case SS_CONNECTING:
case SS_CONNECTED:
restart:
/* Disconnect and send a 'FIN+' or 'FIN-' message to peer */
buf = __skb_dequeue(&sk->sk_receive_queue);
if (buf) {
if (TIPC_SKB_CB(buf)->handle != 0) {
kfree_skb(buf);
goto restart;
}
tipc_disconnect(tport->ref);
tipc_reject_msg(buf, TIPC_CONN_SHUTDOWN);
} else {
tipc_shutdown(tport->ref);
}
sock->state = SS_DISCONNECTING;
/* fall through */
case SS_DISCONNECTING:
/* Discard any unreceived messages */
__skb_queue_purge(&sk->sk_receive_queue);
/* Wake up anyone sleeping in poll */
sk->sk_state_change(sk);
res = 0;
break;
default:
res = -ENOTCONN;
}
release_sock(sk);
return res;
}
/**
* setsockopt - set socket option
* @sock: socket structure
* @lvl: option level
* @opt: option identifier
* @ov: pointer to new option value
* @ol: length of option value
*
* For stream sockets only, accepts and ignores all IPPROTO_TCP options
* (to ease compatibility).
*
* Returns 0 on success, errno otherwise
*/
static int setsockopt(struct socket *sock,
int lvl, int opt, char __user *ov, unsigned int ol)
{
struct sock *sk = sock->sk;
struct tipc_port *tport = tipc_sk_port(sk);
u32 value;
int res;
if ((lvl == IPPROTO_TCP) && (sock->type == SOCK_STREAM))
return 0;
if (lvl != SOL_TIPC)
return -ENOPROTOOPT;
if (ol < sizeof(value))
return -EINVAL;
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);
break;
case TIPC_SRC_DROPPABLE:
if (sock->type != SOCK_STREAM)
res = tipc_set_portunreliable(tport->ref, value);
else
res = -ENOPROTOOPT;
break;
case TIPC_DEST_DROPPABLE:
res = tipc_set_portunreturnable(tport->ref, value);
break;
case TIPC_CONN_TIMEOUT:
tipc_sk(sk)->conn_timeout = value;
/* no need to set "res", since already 0 at this point */
break;
default:
res = -EINVAL;
}
release_sock(sk);
return res;
}
/**
* getsockopt - get socket option
* @sock: socket structure
* @lvl: option level
* @opt: option identifier
* @ov: receptacle for option value
* @ol: receptacle for length of option value
*
* For stream sockets only, returns 0 length result for all IPPROTO_TCP options
* (to ease compatibility).
*
* Returns 0 on success, errno otherwise
*/
static int getsockopt(struct socket *sock,
int lvl, int opt, char __user *ov, int __user *ol)
{
struct sock *sk = sock->sk;
struct tipc_port *tport = tipc_sk_port(sk);
int len;
u32 value;
int res;
if ((lvl == IPPROTO_TCP) && (sock->type == SOCK_STREAM))
return put_user(0, ol);
if (lvl != SOL_TIPC)
return -ENOPROTOOPT;
res = get_user(len, ol);
if (res)
return res;
lock_sock(sk);
switch (opt) {
case TIPC_IMPORTANCE:
res = tipc_portimportance(tport->ref, &value);
break;
case TIPC_SRC_DROPPABLE:
res = tipc_portunreliable(tport->ref, &value);
break;
case TIPC_DEST_DROPPABLE:
res = tipc_portunreturnable(tport->ref, &value);
break;
case TIPC_CONN_TIMEOUT:
value = tipc_sk(sk)->conn_timeout;
/* no need to set "res", since already 0 at this point */
break;
case TIPC_NODE_RECVQ_DEPTH:
value = 0; /* was tipc_queue_size, now obsolete */
break;
case TIPC_SOCK_RECVQ_DEPTH:
value = skb_queue_len(&sk->sk_receive_queue);
break;
default:
res = -EINVAL;
}
release_sock(sk);
if (res)
return res; /* "get" failed */
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 */
static const struct proto_ops msg_ops = {
.owner = THIS_MODULE,
.family = AF_TIPC,
.release = release,
.bind = bind,
.connect = connect,
.socketpair = sock_no_socketpair,
.accept = sock_no_accept,
.getname = get_name,
.poll = poll,
.ioctl = sock_no_ioctl,
.listen = sock_no_listen,
.shutdown = shutdown,
.setsockopt = setsockopt,
.getsockopt = getsockopt,
.sendmsg = send_msg,
.recvmsg = recv_msg,
.mmap = sock_no_mmap,
.sendpage = sock_no_sendpage
};
static const struct proto_ops packet_ops = {
.owner = THIS_MODULE,
.family = AF_TIPC,
.release = release,
.bind = bind,
.connect = connect,
.socketpair = sock_no_socketpair,
.accept = accept,
.getname = get_name,
.poll = poll,
.ioctl = sock_no_ioctl,
.listen = listen,
.shutdown = shutdown,
.setsockopt = setsockopt,
.getsockopt = getsockopt,
.sendmsg = send_packet,
.recvmsg = recv_msg,
.mmap = sock_no_mmap,
.sendpage = sock_no_sendpage
};
static const struct proto_ops stream_ops = {
.owner = THIS_MODULE,
.family = AF_TIPC,
.release = release,
.bind = bind,
.connect = connect,
.socketpair = sock_no_socketpair,
.accept = accept,
.getname = get_name,
.poll = poll,
.ioctl = sock_no_ioctl,
.listen = listen,
.shutdown = shutdown,
.setsockopt = setsockopt,
.getsockopt = getsockopt,
.sendmsg = send_stream,
.recvmsg = recv_stream,
.mmap = sock_no_mmap,
.sendpage = sock_no_sendpage
};
static const struct net_proto_family tipc_family_ops = {
.owner = THIS_MODULE,
.family = AF_TIPC,
.create = tipc_create
};
static struct proto tipc_proto = {
.name = "TIPC",
.owner = THIS_MODULE,
.obj_size = sizeof(struct tipc_sock)
};
/**
* tipc_socket_init - initialize TIPC socket interface
*
* Returns 0 on success, errno otherwise
*/
int tipc_socket_init(void)
{
int res;
res = proto_register(&tipc_proto, 1);
if (res) {
pr_err("Failed to register TIPC protocol type\n");
goto out;
}
res = sock_register(&tipc_family_ops);
if (res) {
pr_err("Failed to register TIPC socket type\n");
proto_unregister(&tipc_proto);
goto out;
}
sockets_enabled = 1;
out:
return res;
}
/**
* tipc_socket_stop - stop TIPC socket interface
*/
void tipc_socket_stop(void)
{
if (!sockets_enabled)
return;
sockets_enabled = 0;
sock_unregister(tipc_family_ops.family);
proto_unregister(&tipc_proto);
}