af_unix.c

linux 内核源代码

};

static const struct proto_ops unix_dgram_ops = {
	.family =	PF_UNIX,
	.owner =	THIS_MODULE,
	.release =	unix_release,
	.bind =		unix_bind,
	.connect =	unix_dgram_connect,
	.socketpair =	unix_socketpair,
	.accept =	sock_no_accept,
	.getname =	unix_getname,
	.poll =		datagram_poll,
	.ioctl =	unix_ioctl,
	.listen =	sock_no_listen,
	.shutdown =	unix_shutdown,
	.setsockopt =	sock_no_setsockopt,
	.getsockopt =	sock_no_getsockopt,
	.sendmsg =	unix_dgram_sendmsg,
	.recvmsg =	unix_dgram_recvmsg,
	.mmap =		sock_no_mmap,
	.sendpage =	sock_no_sendpage,
};

static const struct proto_ops unix_seqpacket_ops = {
	.family =	PF_UNIX,
	.owner =	THIS_MODULE,
	.release =	unix_release,
	.bind =		unix_bind,
	.connect =	unix_stream_connect,
	.socketpair =	unix_socketpair,
	.accept =	unix_accept,
	.getname =	unix_getname,
	.poll =		datagram_poll,
	.ioctl =	unix_ioctl,
	.listen =	unix_listen,
	.shutdown =	unix_shutdown,
	.setsockopt =	sock_no_setsockopt,
	.getsockopt =	sock_no_getsockopt,
	.sendmsg =	unix_seqpacket_sendmsg,
	.recvmsg =	unix_dgram_recvmsg,
	.mmap =		sock_no_mmap,
	.sendpage =	sock_no_sendpage,
};

static struct proto unix_proto = {
	.name	  = "UNIX",
	.owner	  = THIS_MODULE,
	.obj_size = sizeof(struct unix_sock),
};

/*
 * AF_UNIX sockets do not interact with hardware, hence they
 * dont trigger interrupts - so it's safe for them to have
 * bh-unsafe locking for their sk_receive_queue.lock. Split off
 * this special lock-class by reinitializing the spinlock key:
 */
static struct lock_class_key af_unix_sk_receive_queue_lock_key;

static struct sock * unix_create1(struct net *net, struct socket *sock)
{
	struct sock *sk = NULL;
	struct unix_sock *u;

	atomic_inc(&unix_nr_socks);
	if (atomic_read(&unix_nr_socks) > 2 * get_max_files())
		goto out;

	sk = sk_alloc(net, PF_UNIX, GFP_KERNEL, &unix_proto);
	if (!sk)
		goto out;

	sock_init_data(sock,sk);
	lockdep_set_class(&sk->sk_receive_queue.lock,
				&af_unix_sk_receive_queue_lock_key);

	sk->sk_write_space	= unix_write_space;
	sk->sk_max_ack_backlog	= sysctl_unix_max_dgram_qlen;
	sk->sk_destruct		= unix_sock_destructor;
	u	  = unix_sk(sk);
	u->dentry = NULL;
	u->mnt	  = NULL;
	spin_lock_init(&u->lock);
	atomic_set(&u->inflight, 0);
	INIT_LIST_HEAD(&u->link);
	mutex_init(&u->readlock); /* single task reading lock */
	init_waitqueue_head(&u->peer_wait);
	unix_insert_socket(unix_sockets_unbound, sk);
out:
	if (sk == NULL)
		atomic_dec(&unix_nr_socks);
	return sk;
}

static int unix_create(struct net *net, struct socket *sock, int protocol)
{
	if (net != &init_net)
		return -EAFNOSUPPORT;

	if (protocol && protocol != PF_UNIX)
		return -EPROTONOSUPPORT;

	sock->state = SS_UNCONNECTED;

	switch (sock->type) {
	case SOCK_STREAM:
		sock->ops = &unix_stream_ops;
		break;
		/*
		 *	Believe it or not BSD has AF_UNIX, SOCK_RAW though
		 *	nothing uses it.
		 */
	case SOCK_RAW:
		sock->type=SOCK_DGRAM;
	case SOCK_DGRAM:
		sock->ops = &unix_dgram_ops;
		break;
	case SOCK_SEQPACKET:
		sock->ops = &unix_seqpacket_ops;
		break;
	default:
		return -ESOCKTNOSUPPORT;
	}

	return unix_create1(net, sock) ? 0 : -ENOMEM;
}

static int unix_release(struct socket *sock)
{
	struct sock *sk = sock->sk;

	if (!sk)
		return 0;

	sock->sk = NULL;

	return unix_release_sock (sk, 0);
}

static int unix_autobind(struct socket *sock)
{
	struct sock *sk = sock->sk;
	struct unix_sock *u = unix_sk(sk);
	static u32 ordernum = 1;
	struct unix_address * addr;
	int err;

	mutex_lock(&u->readlock);

	err = 0;
	if (u->addr)
		goto out;

	err = -ENOMEM;
	addr = kzalloc(sizeof(*addr) + sizeof(short) + 16, GFP_KERNEL);
	if (!addr)
		goto out;

	addr->name->sun_family = AF_UNIX;
	atomic_set(&addr->refcnt, 1);

retry:
	addr->len = sprintf(addr->name->sun_path+1, "%05x", ordernum) + 1 + sizeof(short);
	addr->hash = unix_hash_fold(csum_partial((void*)addr->name, addr->len, 0));

	spin_lock(&unix_table_lock);
	ordernum = (ordernum+1)&0xFFFFF;

	if (__unix_find_socket_byname(addr->name, addr->len, sock->type,
				      addr->hash)) {
		spin_unlock(&unix_table_lock);
		/* Sanity yield. It is unusual case, but yet... */
		if (!(ordernum&0xFF))
			yield();
		goto retry;
	}
	addr->hash ^= sk->sk_type;

	__unix_remove_socket(sk);
	u->addr = addr;
	__unix_insert_socket(&unix_socket_table[addr->hash], sk);
	spin_unlock(&unix_table_lock);
	err = 0;

out:	mutex_unlock(&u->readlock);
	return err;
}

static struct sock *unix_find_other(struct sockaddr_un *sunname, int len,
				    int type, unsigned hash, int *error)
{
	struct sock *u;
	struct nameidata nd;
	int err = 0;

	if (sunname->sun_path[0]) {
		err = path_lookup(sunname->sun_path, LOOKUP_FOLLOW, &nd);
		if (err)
			goto fail;
		err = vfs_permission(&nd, MAY_WRITE);
		if (err)
			goto put_fail;

		err = -ECONNREFUSED;
		if (!S_ISSOCK(nd.dentry->d_inode->i_mode))
			goto put_fail;
		u=unix_find_socket_byinode(nd.dentry->d_inode);
		if (!u)
			goto put_fail;

		if (u->sk_type == type)
			touch_atime(nd.mnt, nd.dentry);

		path_release(&nd);

		err=-EPROTOTYPE;
		if (u->sk_type != type) {
			sock_put(u);
			goto fail;
		}
	} else {
		err = -ECONNREFUSED;
		u=unix_find_socket_byname(sunname, len, type, hash);
		if (u) {
			struct dentry *dentry;
			dentry = unix_sk(u)->dentry;
			if (dentry)
				touch_atime(unix_sk(u)->mnt, dentry);
		} else
			goto fail;
	}
	return u;

put_fail:
	path_release(&nd);
fail:
	*error=err;
	return NULL;
}


static int unix_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len)
{
	struct sock *sk = sock->sk;
	struct unix_sock *u = unix_sk(sk);
	struct sockaddr_un *sunaddr=(struct sockaddr_un *)uaddr;
	struct dentry * dentry = NULL;
	struct nameidata nd;
	int err;
	unsigned hash;
	struct unix_address *addr;
	struct hlist_head *list;

	err = -EINVAL;
	if (sunaddr->sun_family != AF_UNIX)
		goto out;

	if (addr_len==sizeof(short)) {
		err = unix_autobind(sock);
		goto out;
	}

	err = unix_mkname(sunaddr, addr_len, &hash);
	if (err < 0)
		goto out;
	addr_len = err;

	mutex_lock(&u->readlock);

	err = -EINVAL;
	if (u->addr)
		goto out_up;

	err = -ENOMEM;
	addr = kmalloc(sizeof(*addr)+addr_len, GFP_KERNEL);
	if (!addr)
		goto out_up;

	memcpy(addr->name, sunaddr, addr_len);
	addr->len = addr_len;
	addr->hash = hash ^ sk->sk_type;
	atomic_set(&addr->refcnt, 1);

	if (sunaddr->sun_path[0]) {
		unsigned int mode;
		err = 0;
		/*
		 * Get the parent directory, calculate the hash for last
		 * component.
		 */
		err = path_lookup(sunaddr->sun_path, LOOKUP_PARENT, &nd);
		if (err)
			goto out_mknod_parent;

		dentry = lookup_create(&nd, 0);
		err = PTR_ERR(dentry);
		if (IS_ERR(dentry))
			goto out_mknod_unlock;

		/*
		 * All right, let's create it.
		 */
		mode = S_IFSOCK |
		       (SOCK_INODE(sock)->i_mode & ~current->fs->umask);
		err = vfs_mknod(nd.dentry->d_inode, dentry, mode, 0);
		if (err)
			goto out_mknod_dput;
		mutex_unlock(&nd.dentry->d_inode->i_mutex);
		dput(nd.dentry);
		nd.dentry = dentry;

		addr->hash = UNIX_HASH_SIZE;
	}

	spin_lock(&unix_table_lock);

	if (!sunaddr->sun_path[0]) {
		err = -EADDRINUSE;
		if (__unix_find_socket_byname(sunaddr, addr_len,
					      sk->sk_type, hash)) {
			unix_release_addr(addr);
			goto out_unlock;
		}

		list = &unix_socket_table[addr->hash];
	} else {
		list = &unix_socket_table[dentry->d_inode->i_ino & (UNIX_HASH_SIZE-1)];
		u->dentry = nd.dentry;
		u->mnt    = nd.mnt;
	}

	err = 0;
	__unix_remove_socket(sk);
	u->addr = addr;
	__unix_insert_socket(list, sk);

out_unlock:
	spin_unlock(&unix_table_lock);
out_up:
	mutex_unlock(&u->readlock);
out:
	return err;

out_mknod_dput:
	dput(dentry);
out_mknod_unlock:
	mutex_unlock(&nd.dentry->d_inode->i_mutex);
	path_release(&nd);
out_mknod_parent:
	if (err==-EEXIST)
		err=-EADDRINUSE;
	unix_release_addr(addr);
	goto out_up;
}

static void unix_state_double_lock(struct sock *sk1, struct sock *sk2)
{
	if (unlikely(sk1 == sk2) || !sk2) {
		unix_state_lock(sk1);
		return;
	}
	if (sk1 < sk2) {
		unix_state_lock(sk1);
		unix_state_lock_nested(sk2);
	} else {
		unix_state_lock(sk2);
		unix_state_lock_nested(sk1);
	}
}

static void unix_state_double_unlock(struct sock *sk1, struct sock *sk2)
{
	if (unlikely(sk1 == sk2) || !sk2) {
		unix_state_unlock(sk1);
		return;
	}
	unix_state_unlock(sk1);
	unix_state_unlock(sk2);
}

static int unix_dgram_connect(struct socket *sock, struct sockaddr *addr,
			      int alen, int flags)
{
	struct sock *sk = sock->sk;
	struct sockaddr_un *sunaddr=(struct sockaddr_un*)addr;
	struct sock *other;
	unsigned hash;
	int err;

	if (addr->sa_family != AF_UNSPEC) {
		err = unix_mkname(sunaddr, alen, &hash);
		if (err < 0)
			goto out;
		alen = err;

		if (test_bit(SOCK_PASSCRED, &sock->flags) &&
		    !unix_sk(sk)->addr && (err = unix_autobind(sock)) != 0)
			goto out;

restart:
		other=unix_find_other(sunaddr, alen, sock->type, hash, &err);
		if (!other)
			goto out;

		unix_state_double_lock(sk, other);

		/* Apparently VFS overslept socket death. Retry. */
		if (sock_flag(other, SOCK_DEAD)) {
			unix_state_double_unlock(sk, other);
			sock_put(other);
			goto restart;
		}

		err = -EPERM;
		if (!unix_may_send(sk, other))
			goto out_unlock;

		err = security_unix_may_send(sk->sk_socket, other->sk_socket);
		if (err)
			goto out_unlock;

	} else {
		/*
		 *	1003.1g breaking connected state with AF_UNSPEC
		 */
		other = NULL;
		unix_state_double_lock(sk, other);
	}

	/*
	 * If it was connected, reconnect.
	 */
	if (unix_peer(sk)) {
		struct sock *old_peer = unix_peer(sk);
		unix_peer(sk)=other;
		unix_state_double_unlock(sk, other);

		if (other != old_peer)
			unix_dgram_disconnected(sk, old_peer);
		sock_put(old_peer);
	} else {
		unix_peer(sk)=other;
		unix_state_double_unlock(sk, other);
	}
	return 0;

out_unlock:
	unix_state_double_unlock(sk, other);
	sock_put(other);
out:
	return err;
}

static long unix_wait_for_peer(struct sock *other, long timeo)
{
	struct unix_sock *u = unix_sk(other);
	int sched;
	DEFINE_WAIT(wait);

	prepare_to_wait_exclusive(&u->peer_wait, &wait, TASK_INTERRUPTIBLE);

	sched = !sock_flag(other, SOCK_DEAD) &&
		!(other->sk_shutdown & RCV_SHUTDOWN) &&
		(skb_queue_len(&other->sk_receive_queue) >
		 other->sk_max_ack_backlog);

	unix_state_unlock(other);

	if (sched)
		timeo = schedule_timeout(timeo);

	finish_wait(&u->peer_wait, &wait);
	return timeo;
}

static int unix_stream_connect(struct socket *sock, struct sockaddr *uaddr,
			       int addr_len, int flags)
{
	struct sockaddr_un *sunaddr=(struct sockaddr_un *)uaddr;
	struct sock *sk = sock->sk;
	struct unix_sock *u = unix_sk(sk), *newu, *otheru;
	struct sock *newsk = NULL;
	struct sock *other = NULL;
	struct sk_buff *skb = NULL;
	unsigned hash;
	int st;
	int err;
	long timeo;

	err = unix_mkname(sunaddr, addr_len, &hash);
	if (err < 0)
		goto out;
	addr_len = err;

	if (test_bit(SOCK_PASSCRED, &sock->flags)
		&& !u->addr && (err = unix_autobind(sock)) != 0)
		goto out;

	timeo = sock_sndtimeo(sk, flags & O_NONBLOCK);

	/* First of all allocate resources.
	   If we will make it after state is locked,
	   we will have to recheck all again in any case.
	 */

	err = -ENOMEM;

	/* create new sock for complete connection */
	newsk = unix_create1(sk->sk_net, NULL);
	if (newsk == NULL)
		goto out;

	/* Allocate skb for sending to listening sock */
	skb = sock_wmalloc(newsk, 1, 0, GFP_KERNEL);
	if (skb == NULL)
		goto out;

restart:
	/*  Find listening sock. */
	other = unix_find_other(sunaddr, addr_len, sk->sk_type, hash, &err);
	if (!other)
		goto out;

	/* Latch state of peer */
	unix_state_lock(other);

	/* Apparently VFS overslept socket death. Retry. */
	if (sock_flag(other, SOCK_DEAD)) {
		unix_state_unlock(other);
		sock_put(other);
		goto restart;
	}

	err = -ECONNREFUSED;
	if (other->sk_state != TCP_LISTEN)
		goto out_unlock;

	if (skb_queue_len(&other->sk_receive_queue) >
	    other->sk_max_ack_backlog) {