Skip to content
Snippets Groups Projects
socket.c 161 KiB
Newer Older
	int addrlen;
	rwlock_t *addr_lock;
	int err = 0;
	size_t space_left;
	int bytes_copied;

	if (len <= sizeof(struct sctp_getaddrs))
		return -EINVAL;

	if (copy_from_user(&getaddrs, optval, sizeof(struct sctp_getaddrs)))
		return -EFAULT;

	/*
	 *  For UDP-style sockets, id specifies the association to query.
	 *  If the id field is set to the value '0' then the locally bound
	 *  addresses are returned without regard to any particular
	 *  association.
	 */
	if (0 == getaddrs.assoc_id) {
		bp = &sctp_sk(sk)->ep->base.bind_addr;
		addr_lock = &sctp_sk(sk)->ep->base.addr_lock;
	} else {
		asoc = sctp_id2assoc(sk, getaddrs.assoc_id);
		if (!asoc)
			return -EINVAL;
		bp = &asoc->base.bind_addr;
		addr_lock = &asoc->base.addr_lock;
	}

	to = optval + offsetof(struct sctp_getaddrs,addrs);
	space_left = len - sizeof(struct sctp_getaddrs) -
			 offsetof(struct sctp_getaddrs,addrs);

	sctp_read_lock(addr_lock);

	/* If the endpoint is bound to 0.0.0.0 or ::0, get the valid
	 * addresses from the global local address list.
	 */
	if (sctp_list_single_entry(&bp->address_list)) {
		addr = list_entry(bp->address_list.next,
				  struct sctp_sockaddr_entry, list);
		if (sctp_is_any(&addr->a)) {
			cnt = sctp_copy_laddrs_to_user(sk, bp->port,
						       &to, space_left);
			if (cnt < 0) {
				err = cnt;
				goto unlock;
			}
			goto copy_getaddrs;		
		}
	}

	list_for_each(pos, &bp->address_list) {
		addr = list_entry(pos, struct sctp_sockaddr_entry, list);
		memcpy(&temp, &addr->a, sizeof(temp));
		sctp_get_pf_specific(sk->sk_family)->addr_v4map(sp, &temp);
		addrlen = sctp_get_af_specific(temp.sa.sa_family)->sockaddr_len;
		if(space_left < addrlen)
			return -ENOMEM; /*fixme: right error?*/
		temp.v4.sin_port = htons(temp.v4.sin_port);
		if (copy_to_user(to, &temp, addrlen)) {
			err = -EFAULT;
			goto unlock;
		}
		to += addrlen;
		cnt ++;
		space_left -= addrlen;
	}

copy_getaddrs:
	if (put_user(cnt, &((struct sctp_getaddrs __user *)optval)->addr_num))
		return -EFAULT;
	bytes_copied = ((char __user *)to) - optval;
	if (put_user(bytes_copied, optlen))
		return -EFAULT;

unlock:
	sctp_read_unlock(addr_lock);
	return err;
}

Linus Torvalds's avatar
Linus Torvalds committed
/* 7.1.10 Set Primary Address (SCTP_PRIMARY_ADDR)
 *
 * Requests that the local SCTP stack use the enclosed peer address as
 * the association primary.  The enclosed address must be one of the
 * association peer's addresses.
 */
static int sctp_getsockopt_primary_addr(struct sock *sk, int len,
					char __user *optval, int __user *optlen)
{
	struct sctp_prim prim;
	struct sctp_association *asoc;
	struct sctp_sock *sp = sctp_sk(sk);

	if (len != sizeof(struct sctp_prim))
		return -EINVAL;

	if (copy_from_user(&prim, optval, sizeof(struct sctp_prim)))
		return -EFAULT;

	asoc = sctp_id2assoc(sk, prim.ssp_assoc_id);
	if (!asoc)
		return -EINVAL;

	if (!asoc->peer.primary_path)
		return -ENOTCONN;
	
	asoc->peer.primary_path->ipaddr.v4.sin_port =
		htons(asoc->peer.primary_path->ipaddr.v4.sin_port);
	memcpy(&prim.ssp_addr, &asoc->peer.primary_path->ipaddr,
	       sizeof(union sctp_addr));
	asoc->peer.primary_path->ipaddr.v4.sin_port =
		ntohs(asoc->peer.primary_path->ipaddr.v4.sin_port);

	sctp_get_pf_specific(sk->sk_family)->addr_v4map(sp,
			(union sctp_addr *)&prim.ssp_addr);

	if (copy_to_user(optval, &prim, sizeof(struct sctp_prim)))
		return -EFAULT;

	return 0;
}

/*
 * 7.1.11  Set Adaption Layer Indicator (SCTP_ADAPTION_LAYER)
 *
 * Requests that the local endpoint set the specified Adaption Layer
 * Indication parameter for all future INIT and INIT-ACK exchanges.
 */
static int sctp_getsockopt_adaption_layer(struct sock *sk, int len,
				  char __user *optval, int __user *optlen)
{
Linus Torvalds's avatar
Linus Torvalds committed

	if (len != sizeof(struct sctp_setadaption))
Linus Torvalds's avatar
Linus Torvalds committed
		return -EINVAL;

	adaption.ssb_adaption_ind = sctp_sk(sk)->adaption_ind;
	if (copy_to_user(optval, &adaption, len))
Linus Torvalds's avatar
Linus Torvalds committed
		return -EFAULT;
Linus Torvalds's avatar
Linus Torvalds committed
	return 0;
}

/*
 *
 * 7.1.14 Set default send parameters (SCTP_DEFAULT_SEND_PARAM)
 *
 *   Applications that wish to use the sendto() system call may wish to
 *   specify a default set of parameters that would normally be supplied
 *   through the inclusion of ancillary data.  This socket option allows
 *   such an application to set the default sctp_sndrcvinfo structure.


 *   The application that wishes to use this socket option simply passes
 *   in to this call the sctp_sndrcvinfo structure defined in Section
 *   5.2.2) The input parameters accepted by this call include
 *   sinfo_stream, sinfo_flags, sinfo_ppid, sinfo_context,
 *   sinfo_timetolive.  The user must provide the sinfo_assoc_id field in
 *   to this call if the caller is using the UDP model.
 *
 *   For getsockopt, it get the default sctp_sndrcvinfo structure.
 */
static int sctp_getsockopt_default_send_param(struct sock *sk,
					int len, char __user *optval,
					int __user *optlen)
{
	struct sctp_sndrcvinfo info;
	struct sctp_association *asoc;
	struct sctp_sock *sp = sctp_sk(sk);

	if (len != sizeof(struct sctp_sndrcvinfo))
		return -EINVAL;
	if (copy_from_user(&info, optval, sizeof(struct sctp_sndrcvinfo)))
		return -EFAULT;

	asoc = sctp_id2assoc(sk, info.sinfo_assoc_id);
	if (!asoc && info.sinfo_assoc_id && sctp_style(sk, UDP))
		return -EINVAL;

	if (asoc) {
		info.sinfo_stream = asoc->default_stream;
		info.sinfo_flags = asoc->default_flags;
		info.sinfo_ppid = asoc->default_ppid;
		info.sinfo_context = asoc->default_context;
		info.sinfo_timetolive = asoc->default_timetolive;
	} else {
		info.sinfo_stream = sp->default_stream;
		info.sinfo_flags = sp->default_flags;
		info.sinfo_ppid = sp->default_ppid;
		info.sinfo_context = sp->default_context;
		info.sinfo_timetolive = sp->default_timetolive;
	}

	if (copy_to_user(optval, &info, sizeof(struct sctp_sndrcvinfo)))
		return -EFAULT;

	return 0;
}

/*
 *
 * 7.1.5 SCTP_NODELAY
 *
 * Turn on/off any Nagle-like algorithm.  This means that packets are
 * generally sent as soon as possible and no unnecessary delays are
 * introduced, at the cost of more packets in the network.  Expects an
 * integer boolean flag.
 */

static int sctp_getsockopt_nodelay(struct sock *sk, int len,
				   char __user *optval, int __user *optlen)
{
	int val;

	if (len < sizeof(int))
		return -EINVAL;

	len = sizeof(int);
	val = (sctp_sk(sk)->nodelay == 1);
	if (put_user(len, optlen))
		return -EFAULT;
	if (copy_to_user(optval, &val, len))
		return -EFAULT;
	return 0;
}

/*
 *
 * 7.1.1 SCTP_RTOINFO
 *
 * The protocol parameters used to initialize and bound retransmission
 * timeout (RTO) are tunable. sctp_rtoinfo structure is used to access
 * and modify these parameters.
 * All parameters are time values, in milliseconds.  A value of 0, when
 * modifying the parameters, indicates that the current value should not
 * be changed.
 *
 */
static int sctp_getsockopt_rtoinfo(struct sock *sk, int len,
				char __user *optval,
				int __user *optlen) {
	struct sctp_rtoinfo rtoinfo;
	struct sctp_association *asoc;

	if (len != sizeof (struct sctp_rtoinfo))
		return -EINVAL;

	if (copy_from_user(&rtoinfo, optval, sizeof (struct sctp_rtoinfo)))
		return -EFAULT;

	asoc = sctp_id2assoc(sk, rtoinfo.srto_assoc_id);

	if (!asoc && rtoinfo.srto_assoc_id && sctp_style(sk, UDP))
		return -EINVAL;

	/* Values corresponding to the specific association. */
	if (asoc) {
		rtoinfo.srto_initial = jiffies_to_msecs(asoc->rto_initial);
		rtoinfo.srto_max = jiffies_to_msecs(asoc->rto_max);
		rtoinfo.srto_min = jiffies_to_msecs(asoc->rto_min);
	} else {
		/* Values corresponding to the endpoint. */
		struct sctp_sock *sp = sctp_sk(sk);

		rtoinfo.srto_initial = sp->rtoinfo.srto_initial;
		rtoinfo.srto_max = sp->rtoinfo.srto_max;
		rtoinfo.srto_min = sp->rtoinfo.srto_min;
	}

	if (put_user(len, optlen))
		return -EFAULT;

	if (copy_to_user(optval, &rtoinfo, len))
		return -EFAULT;

	return 0;
}

/*
 *
 * 7.1.2 SCTP_ASSOCINFO
 *
 * This option is used to tune the the maximum retransmission attempts
 * of the association.
 * Returns an error if the new association retransmission value is
 * greater than the sum of the retransmission value  of the peer.
 * See [SCTP] for more information.
 *
 */
static int sctp_getsockopt_associnfo(struct sock *sk, int len,
				     char __user *optval,
				     int __user *optlen)
{

	struct sctp_assocparams assocparams;
	struct sctp_association *asoc;
	struct list_head *pos;
	int cnt = 0;

	if (len != sizeof (struct sctp_assocparams))
		return -EINVAL;

	if (copy_from_user(&assocparams, optval,
			sizeof (struct sctp_assocparams)))
		return -EFAULT;

	asoc = sctp_id2assoc(sk, assocparams.sasoc_assoc_id);

	if (!asoc && assocparams.sasoc_assoc_id && sctp_style(sk, UDP))
		return -EINVAL;

	/* Values correspoinding to the specific association */
Linus Torvalds's avatar
Linus Torvalds committed
		assocparams.sasoc_asocmaxrxt = asoc->max_retrans;
		assocparams.sasoc_peer_rwnd = asoc->peer.rwnd;
		assocparams.sasoc_local_rwnd = asoc->a_rwnd;
		assocparams.sasoc_cookie_life = (asoc->cookie_life.tv_sec
						* 1000) +
						(asoc->cookie_life.tv_usec
						/ 1000);

		list_for_each(pos, &asoc->peer.transport_addr_list) {
			cnt ++;
		}

		assocparams.sasoc_number_peer_destinations = cnt;
	} else {
		/* Values corresponding to the endpoint */
		struct sctp_sock *sp = sctp_sk(sk);

		assocparams.sasoc_asocmaxrxt = sp->assocparams.sasoc_asocmaxrxt;
		assocparams.sasoc_peer_rwnd = sp->assocparams.sasoc_peer_rwnd;
		assocparams.sasoc_local_rwnd = sp->assocparams.sasoc_local_rwnd;
		assocparams.sasoc_cookie_life =
					sp->assocparams.sasoc_cookie_life;
		assocparams.sasoc_number_peer_destinations =
					sp->assocparams.
					sasoc_number_peer_destinations;
	}

	if (put_user(len, optlen))
		return -EFAULT;

	if (copy_to_user(optval, &assocparams, len))
		return -EFAULT;

	return 0;
}

/*
 * 7.1.16 Set/clear IPv4 mapped addresses (SCTP_I_WANT_MAPPED_V4_ADDR)
 *
 * This socket option is a boolean flag which turns on or off mapped V4
 * addresses.  If this option is turned on and the socket is type
 * PF_INET6, then IPv4 addresses will be mapped to V6 representation.
 * If this option is turned off, then no mapping will be done of V4
 * addresses and a user will receive both PF_INET6 and PF_INET type
 * addresses on the socket.
 */
static int sctp_getsockopt_mappedv4(struct sock *sk, int len,
				    char __user *optval, int __user *optlen)
{
	int val;
	struct sctp_sock *sp = sctp_sk(sk);

	if (len < sizeof(int))
		return -EINVAL;

	len = sizeof(int);
	val = sp->v4mapped;
	if (put_user(len, optlen))
		return -EFAULT;
	if (copy_to_user(optval, &val, len))
		return -EFAULT;

	return 0;
}

/*
 * 7.1.17 Set the maximum fragrmentation size (SCTP_MAXSEG)
 *
 * This socket option specifies the maximum size to put in any outgoing
 * SCTP chunk.  If a message is larger than this size it will be
 * fragmented by SCTP into the specified size.  Note that the underlying
 * SCTP implementation may fragment into smaller sized chunks when the
 * PMTU of the underlying association is smaller than the value set by
 * the user.
 */
static int sctp_getsockopt_maxseg(struct sock *sk, int len,
				  char __user *optval, int __user *optlen)
{
	int val;

	if (len < sizeof(int))
		return -EINVAL;

	len = sizeof(int);

	val = sctp_sk(sk)->user_frag;
	if (put_user(len, optlen))
		return -EFAULT;
	if (copy_to_user(optval, &val, len))
		return -EFAULT;

	return 0;
}

SCTP_STATIC int sctp_getsockopt(struct sock *sk, int level, int optname,
				char __user *optval, int __user *optlen)
{
	int retval = 0;
	int len;

	SCTP_DEBUG_PRINTK("sctp_getsockopt(sk: %p... optname: %d)\n",
			  sk, optname);
Linus Torvalds's avatar
Linus Torvalds committed

	/* I can hardly begin to describe how wrong this is.  This is
	 * so broken as to be worse than useless.  The API draft
	 * REALLY is NOT helpful here...  I am not convinced that the
	 * semantics of getsockopt() with a level OTHER THAN SOL_SCTP
	 * are at all well-founded.
	 */
	if (level != SOL_SCTP) {
		struct sctp_af *af = sctp_sk(sk)->pf->af;

		retval = af->getsockopt(sk, level, optname, optval, optlen);
		return retval;
	}

	if (get_user(len, optlen))
		return -EFAULT;

	sctp_lock_sock(sk);

	switch (optname) {
	case SCTP_STATUS:
		retval = sctp_getsockopt_sctp_status(sk, len, optval, optlen);
		break;
	case SCTP_DISABLE_FRAGMENTS:
		retval = sctp_getsockopt_disable_fragments(sk, len, optval,
							   optlen);
		break;
	case SCTP_EVENTS:
		retval = sctp_getsockopt_events(sk, len, optval, optlen);
		break;
	case SCTP_AUTOCLOSE:
		retval = sctp_getsockopt_autoclose(sk, len, optval, optlen);
		break;
	case SCTP_SOCKOPT_PEELOFF:
		retval = sctp_getsockopt_peeloff(sk, len, optval, optlen);
		break;
	case SCTP_PEER_ADDR_PARAMS:
		retval = sctp_getsockopt_peer_addr_params(sk, len, optval,
							  optlen);
		break;
	case SCTP_DELAYED_ACK_TIME:
		retval = sctp_getsockopt_delayed_ack_time(sk, len, optval,
							  optlen);
		break;
Linus Torvalds's avatar
Linus Torvalds committed
	case SCTP_INITMSG:
		retval = sctp_getsockopt_initmsg(sk, len, optval, optlen);
		break;
	case SCTP_GET_PEER_ADDRS_NUM_OLD:
		retval = sctp_getsockopt_peer_addrs_num_old(sk, len, optval,
							    optlen);
		break;
	case SCTP_GET_LOCAL_ADDRS_NUM_OLD:
		retval = sctp_getsockopt_local_addrs_num_old(sk, len, optval,
							     optlen);
		break;
	case SCTP_GET_PEER_ADDRS_OLD:
		retval = sctp_getsockopt_peer_addrs_old(sk, len, optval,
Linus Torvalds's avatar
Linus Torvalds committed
							optlen);
		break;
	case SCTP_GET_LOCAL_ADDRS_OLD:
		retval = sctp_getsockopt_local_addrs_old(sk, len, optval,
Linus Torvalds's avatar
Linus Torvalds committed
4478 4479 4480 4481 4482 4483 4484 4485 4486 4487 4488 4489 4490 4491 4492 4493 4494 4495 4496 4497 4498 4499 4500 4501 4502 4503 4504 4505 4506 4507 4508 4509 4510 4511 4512 4513 4514 4515 4516 4517 4518 4519 4520 4521 4522 4523 4524 4525 4526 4527 4528 4529 4530 4531 4532 4533 4534 4535 4536 4537 4538 4539 4540 4541 4542 4543 4544 4545 4546 4547 4548 4549 4550 4551 4552 4553 4554 4555 4556 4557 4558 4559 4560 4561 4562 4563 4564 4565 4566 4567 4568 4569 4570 4571 4572 4573 4574 4575 4576 4577 4578 4579 4580 4581 4582 4583 4584 4585 4586 4587 4588 4589 4590 4591 4592 4593 4594 4595 4596 4597 4598 4599 4600 4601 4602 4603 4604 4605 4606 4607 4608 4609 4610 4611 4612 4613 4614 4615 4616 4617 4618 4619 4620 4621 4622 4623 4624 4625 4626 4627 4628 4629 4630 4631 4632 4633 4634 4635 4636 4637 4638 4639 4640 4641 4642 4643 4644 4645 4646 4647 4648 4649 4650 4651 4652 4653 4654 4655 4656 4657 4658 4659 4660 4661 4662 4663 4664 4665 4666 4667 4668 4669 4670 4671 4672 4673 4674 4675 4676 4677 4678 4679 4680 4681 4682 4683 4684 4685 4686 4687 4688 4689 4690 4691 4692 4693 4694 4695 4696 4697 4698 4699 4700 4701 4702 4703 4704 4705 4706 4707 4708 4709 4710 4711 4712 4713 4714 4715 4716 4717 4718 4719 4720 4721 4722 4723 4724 4725 4726 4727 4728 4729 4730 4731 4732 4733 4734 4735 4736 4737 4738 4739 4740 4741 4742 4743 4744 4745 4746 4747 4748 4749 4750 4751 4752 4753 4754 4755 4756 4757 4758 4759 4760 4761 4762 4763 4764 4765 4766 4767 4768 4769 4770 4771 4772 4773 4774 4775 4776 4777 4778 4779 4780 4781 4782 4783 4784 4785 4786 4787 4788 4789 4790 4791 4792 4793 4794 4795 4796 4797 4798 4799 4800 4801 4802 4803 4804 4805 4806 4807 4808 4809 4810 4811 4812 4813 4814 4815 4816 4817 4818 4819 4820 4821 4822 4823 4824 4825 4826 4827 4828 4829 4830 4831 4832 4833 4834 4835 4836 4837 4838 4839 4840 4841 4842 4843 4844 4845 4846 4847 4848 4849 4850 4851 4852 4853 4854 4855 4856 4857 4858
							 optlen);
		break;
	case SCTP_GET_PEER_ADDRS:
		retval = sctp_getsockopt_peer_addrs(sk, len, optval,
						    optlen);
		break;
	case SCTP_GET_LOCAL_ADDRS:
		retval = sctp_getsockopt_local_addrs(sk, len, optval,
						     optlen);
		break;
	case SCTP_DEFAULT_SEND_PARAM:
		retval = sctp_getsockopt_default_send_param(sk, len,
							    optval, optlen);
		break;
	case SCTP_PRIMARY_ADDR:
		retval = sctp_getsockopt_primary_addr(sk, len, optval, optlen);
		break;
	case SCTP_NODELAY:
		retval = sctp_getsockopt_nodelay(sk, len, optval, optlen);
		break;
	case SCTP_RTOINFO:
		retval = sctp_getsockopt_rtoinfo(sk, len, optval, optlen);
		break;
	case SCTP_ASSOCINFO:
		retval = sctp_getsockopt_associnfo(sk, len, optval, optlen);
		break;
	case SCTP_I_WANT_MAPPED_V4_ADDR:
		retval = sctp_getsockopt_mappedv4(sk, len, optval, optlen);
		break;
	case SCTP_MAXSEG:
		retval = sctp_getsockopt_maxseg(sk, len, optval, optlen);
		break;
	case SCTP_GET_PEER_ADDR_INFO:
		retval = sctp_getsockopt_peer_addr_info(sk, len, optval,
							optlen);
		break;
	case SCTP_ADAPTION_LAYER:
		retval = sctp_getsockopt_adaption_layer(sk, len, optval,
							optlen);
		break;
	default:
		retval = -ENOPROTOOPT;
		break;
	};

	sctp_release_sock(sk);
	return retval;
}

static void sctp_hash(struct sock *sk)
{
	/* STUB */
}

static void sctp_unhash(struct sock *sk)
{
	/* STUB */
}

/* Check if port is acceptable.  Possibly find first available port.
 *
 * The port hash table (contained in the 'global' SCTP protocol storage
 * returned by struct sctp_protocol *sctp_get_protocol()). The hash
 * table is an array of 4096 lists (sctp_bind_hashbucket). Each
 * list (the list number is the port number hashed out, so as you
 * would expect from a hash function, all the ports in a given list have
 * such a number that hashes out to the same list number; you were
 * expecting that, right?); so each list has a set of ports, with a
 * link to the socket (struct sock) that uses it, the port number and
 * a fastreuse flag (FIXME: NPI ipg).
 */
static struct sctp_bind_bucket *sctp_bucket_create(
	struct sctp_bind_hashbucket *head, unsigned short snum);

static long sctp_get_port_local(struct sock *sk, union sctp_addr *addr)
{
	struct sctp_bind_hashbucket *head; /* hash list */
	struct sctp_bind_bucket *pp; /* hash list port iterator */
	unsigned short snum;
	int ret;

	/* NOTE:  Remember to put this back to net order. */
	addr->v4.sin_port = ntohs(addr->v4.sin_port);
	snum = addr->v4.sin_port;

	SCTP_DEBUG_PRINTK("sctp_get_port() begins, snum=%d\n", snum);
	sctp_local_bh_disable();

	if (snum == 0) {
		/* Search for an available port.
		 *
		 * 'sctp_port_rover' was the last port assigned, so
		 * we start to search from 'sctp_port_rover +
		 * 1'. What we do is first check if port 'rover' is
		 * already in the hash table; if not, we use that; if
		 * it is, we try next.
		 */
		int low = sysctl_local_port_range[0];
		int high = sysctl_local_port_range[1];
		int remaining = (high - low) + 1;
		int rover;
		int index;

		sctp_spin_lock(&sctp_port_alloc_lock);
		rover = sctp_port_rover;
		do {
			rover++;
			if ((rover < low) || (rover > high))
				rover = low;
			index = sctp_phashfn(rover);
			head = &sctp_port_hashtable[index];
			sctp_spin_lock(&head->lock);
			for (pp = head->chain; pp; pp = pp->next)
				if (pp->port == rover)
					goto next;
			break;
		next:
			sctp_spin_unlock(&head->lock);
		} while (--remaining > 0);
		sctp_port_rover = rover;
		sctp_spin_unlock(&sctp_port_alloc_lock);

		/* Exhausted local port range during search? */
		ret = 1;
		if (remaining <= 0)
			goto fail;

		/* OK, here is the one we will use.  HEAD (the port
		 * hash table list entry) is non-NULL and we hold it's
		 * mutex.
		 */
		snum = rover;
	} else {
		/* We are given an specific port number; we verify
		 * that it is not being used. If it is used, we will
		 * exahust the search in the hash list corresponding
		 * to the port number (snum) - we detect that with the
		 * port iterator, pp being NULL.
		 */
		head = &sctp_port_hashtable[sctp_phashfn(snum)];
		sctp_spin_lock(&head->lock);
		for (pp = head->chain; pp; pp = pp->next) {
			if (pp->port == snum)
				goto pp_found;
		}
	}
	pp = NULL;
	goto pp_not_found;
pp_found:
	if (!hlist_empty(&pp->owner)) {
		/* We had a port hash table hit - there is an
		 * available port (pp != NULL) and it is being
		 * used by other socket (pp->owner not empty); that other
		 * socket is going to be sk2.
		 */
		int reuse = sk->sk_reuse;
		struct sock *sk2;
		struct hlist_node *node;

		SCTP_DEBUG_PRINTK("sctp_get_port() found a possible match\n");
		if (pp->fastreuse && sk->sk_reuse)
			goto success;

		/* Run through the list of sockets bound to the port
		 * (pp->port) [via the pointers bind_next and
		 * bind_pprev in the struct sock *sk2 (pp->sk)]. On each one,
		 * we get the endpoint they describe and run through
		 * the endpoint's list of IP (v4 or v6) addresses,
		 * comparing each of the addresses with the address of
		 * the socket sk. If we find a match, then that means
		 * that this port/socket (sk) combination are already
		 * in an endpoint.
		 */
		sk_for_each_bound(sk2, node, &pp->owner) {
			struct sctp_endpoint *ep2;
			ep2 = sctp_sk(sk2)->ep;

			if (reuse && sk2->sk_reuse)
				continue;

			if (sctp_bind_addr_match(&ep2->base.bind_addr, addr,
						 sctp_sk(sk))) {
				ret = (long)sk2;
				goto fail_unlock;
			}
		}
		SCTP_DEBUG_PRINTK("sctp_get_port(): Found a match\n");
	}
pp_not_found:
	/* If there was a hash table miss, create a new port.  */
	ret = 1;
	if (!pp && !(pp = sctp_bucket_create(head, snum)))
		goto fail_unlock;

	/* In either case (hit or miss), make sure fastreuse is 1 only
	 * if sk->sk_reuse is too (that is, if the caller requested
	 * SO_REUSEADDR on this socket -sk-).
	 */
	if (hlist_empty(&pp->owner))
		pp->fastreuse = sk->sk_reuse ? 1 : 0;
	else if (pp->fastreuse && !sk->sk_reuse)
		pp->fastreuse = 0;

	/* We are set, so fill up all the data in the hash table
	 * entry, tie the socket list information with the rest of the
	 * sockets FIXME: Blurry, NPI (ipg).
	 */
success:
	inet_sk(sk)->num = snum;
	if (!sctp_sk(sk)->bind_hash) {
		sk_add_bind_node(sk, &pp->owner);
		sctp_sk(sk)->bind_hash = pp;
	}
	ret = 0;

fail_unlock:
	sctp_spin_unlock(&head->lock);

fail:
	sctp_local_bh_enable();
	addr->v4.sin_port = htons(addr->v4.sin_port);
	return ret;
}

/* Assign a 'snum' port to the socket.  If snum == 0, an ephemeral
 * port is requested.
 */
static int sctp_get_port(struct sock *sk, unsigned short snum)
{
	long ret;
	union sctp_addr addr;
	struct sctp_af *af = sctp_sk(sk)->pf->af;

	/* Set up a dummy address struct from the sk. */
	af->from_sk(&addr, sk);
	addr.v4.sin_port = htons(snum);

	/* Note: sk->sk_num gets filled in if ephemeral port request. */
	ret = sctp_get_port_local(sk, &addr);

	return (ret ? 1 : 0);
}

/*
 * 3.1.3 listen() - UDP Style Syntax
 *
 *   By default, new associations are not accepted for UDP style sockets.
 *   An application uses listen() to mark a socket as being able to
 *   accept new associations.
 */
SCTP_STATIC int sctp_seqpacket_listen(struct sock *sk, int backlog)
{
	struct sctp_sock *sp = sctp_sk(sk);
	struct sctp_endpoint *ep = sp->ep;

	/* Only UDP style sockets that are not peeled off are allowed to
	 * listen().
	 */
	if (!sctp_style(sk, UDP))
		return -EINVAL;

	/* If backlog is zero, disable listening. */
	if (!backlog) {
		if (sctp_sstate(sk, CLOSED))
			return 0;
		
		sctp_unhash_endpoint(ep);
		sk->sk_state = SCTP_SS_CLOSED;
	}

	/* Return if we are already listening. */
	if (sctp_sstate(sk, LISTENING))
		return 0;
		
	/*
	 * If a bind() or sctp_bindx() is not called prior to a listen()
	 * call that allows new associations to be accepted, the system
	 * picks an ephemeral port and will choose an address set equivalent
	 * to binding with a wildcard address.
	 *
	 * This is not currently spelled out in the SCTP sockets
	 * extensions draft, but follows the practice as seen in TCP
	 * sockets.
	 */
	if (!ep->base.bind_addr.port) {
		if (sctp_autobind(sk))
			return -EAGAIN;
	}
	sk->sk_state = SCTP_SS_LISTENING;
	sctp_hash_endpoint(ep);
	return 0;
}

/*
 * 4.1.3 listen() - TCP Style Syntax
 *
 *   Applications uses listen() to ready the SCTP endpoint for accepting
 *   inbound associations.
 */
SCTP_STATIC int sctp_stream_listen(struct sock *sk, int backlog)
{
	struct sctp_sock *sp = sctp_sk(sk);
	struct sctp_endpoint *ep = sp->ep;

	/* If backlog is zero, disable listening. */
	if (!backlog) {
		if (sctp_sstate(sk, CLOSED))
			return 0;
		
		sctp_unhash_endpoint(ep);
		sk->sk_state = SCTP_SS_CLOSED;
	}

	if (sctp_sstate(sk, LISTENING))
		return 0;

	/*
	 * If a bind() or sctp_bindx() is not called prior to a listen()
	 * call that allows new associations to be accepted, the system
	 * picks an ephemeral port and will choose an address set equivalent
	 * to binding with a wildcard address.
	 *
	 * This is not currently spelled out in the SCTP sockets
	 * extensions draft, but follows the practice as seen in TCP
	 * sockets.
	 */
	if (!ep->base.bind_addr.port) {
		if (sctp_autobind(sk))
			return -EAGAIN;
	}
	sk->sk_state = SCTP_SS_LISTENING;
	sk->sk_max_ack_backlog = backlog;
	sctp_hash_endpoint(ep);
	return 0;
}

/*
 *  Move a socket to LISTENING state.
 */
int sctp_inet_listen(struct socket *sock, int backlog)
{
	struct sock *sk = sock->sk;
	struct crypto_tfm *tfm=NULL;
	int err = -EINVAL;

	if (unlikely(backlog < 0))
		goto out;

	sctp_lock_sock(sk);

	if (sock->state != SS_UNCONNECTED)
		goto out;

	/* Allocate HMAC for generating cookie. */
	if (sctp_hmac_alg) {
		tfm = sctp_crypto_alloc_tfm(sctp_hmac_alg, 0);
		if (!tfm) {
			err = -ENOSYS;
			goto out;
		}
	}

	switch (sock->type) {
	case SOCK_SEQPACKET:
		err = sctp_seqpacket_listen(sk, backlog);
		break;
	case SOCK_STREAM:
		err = sctp_stream_listen(sk, backlog);
		break;
	default:
		break;
	};
	if (err)
		goto cleanup;

	/* Store away the transform reference. */
	sctp_sk(sk)->hmac = tfm;
out:
	sctp_release_sock(sk);
	return err;
cleanup:
Linus Torvalds's avatar
Linus Torvalds committed
	goto out;
}

/*
 * This function is done by modeling the current datagram_poll() and the
 * tcp_poll().  Note that, based on these implementations, we don't
 * lock the socket in this function, even though it seems that,
 * ideally, locking or some other mechanisms can be used to ensure
 * the integrity of the counters (sndbuf and wmem_alloc) used
Linus Torvalds's avatar
Linus Torvalds committed
 * in this place.  We assume that we don't need locks either until proven
 * otherwise.
 *
 * Another thing to note is that we include the Async I/O support
 * here, again, by modeling the current TCP/UDP code.  We don't have
 * a good way to test with it yet.
 */
unsigned int sctp_poll(struct file *file, struct socket *sock, poll_table *wait)
{
	struct sock *sk = sock->sk;
	struct sctp_sock *sp = sctp_sk(sk);
	unsigned int mask;

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

	/* A TCP-style listening socket becomes readable when the accept queue
	 * is not empty.
	 */
	if (sctp_style(sk, TCP) && sctp_sstate(sk, LISTENING))
		return (!list_empty(&sp->ep->asocs)) ?
		       	(POLLIN | POLLRDNORM) : 0;

	mask = 0;

	/* Is there any exceptional events?  */
	if (sk->sk_err || !skb_queue_empty(&sk->sk_error_queue))
		mask |= POLLERR;
	if (sk->sk_shutdown == SHUTDOWN_MASK)
		mask |= POLLHUP;

	/* Is it readable?  Reconsider this code with TCP-style support.  */
	if (!skb_queue_empty(&sk->sk_receive_queue) ||
	    (sk->sk_shutdown & RCV_SHUTDOWN))
		mask |= POLLIN | POLLRDNORM;

	/* The association is either gone or not ready.  */
	if (!sctp_style(sk, UDP) && sctp_sstate(sk, CLOSED))
		return mask;

	/* Is it writable?  */
	if (sctp_writeable(sk)) {
		mask |= POLLOUT | POLLWRNORM;
	} else {
		set_bit(SOCK_ASYNC_NOSPACE, &sk->sk_socket->flags);
		/*
		 * Since the socket is not locked, the buffer
		 * might be made available after the writeable check and
		 * before the bit is set.  This could cause a lost I/O
		 * signal.  tcp_poll() has a race breaker for this race
		 * condition.  Based on their implementation, we put
		 * in the following code to cover it as well.
		 */
		if (sctp_writeable(sk))
			mask |= POLLOUT | POLLWRNORM;
	}
	return mask;
}

/********************************************************************
 * 2nd Level Abstractions
 ********************************************************************/

static struct sctp_bind_bucket *sctp_bucket_create(
	struct sctp_bind_hashbucket *head, unsigned short snum)
{
	struct sctp_bind_bucket *pp;

	pp = kmem_cache_alloc(sctp_bucket_cachep, SLAB_ATOMIC);
	SCTP_DBG_OBJCNT_INC(bind_bucket);
	if (pp) {
		pp->port = snum;
		pp->fastreuse = 0;
		INIT_HLIST_HEAD(&pp->owner);
		if ((pp->next = head->chain) != NULL)
			pp->next->pprev = &pp->next;
		head->chain = pp;
		pp->pprev = &head->chain;
	}
	return pp;
}

/* Caller must hold hashbucket lock for this tb with local BH disabled */
static void sctp_bucket_destroy(struct sctp_bind_bucket *pp)
{
	if (hlist_empty(&pp->owner)) {
		if (pp->next)
			pp->next->pprev = pp->pprev;
		*(pp->pprev) = pp->next;
		kmem_cache_free(sctp_bucket_cachep, pp);
		SCTP_DBG_OBJCNT_DEC(bind_bucket);
	}
}

/* Release this socket's reference to a local port.  */
static inline void __sctp_put_port(struct sock *sk)
{
	struct sctp_bind_hashbucket *head =
		&sctp_port_hashtable[sctp_phashfn(inet_sk(sk)->num)];
	struct sctp_bind_bucket *pp;

	sctp_spin_lock(&head->lock);
	pp = sctp_sk(sk)->bind_hash;
	__sk_del_bind_node(sk);
	sctp_sk(sk)->bind_hash = NULL;
	inet_sk(sk)->num = 0;
	sctp_bucket_destroy(pp);
	sctp_spin_unlock(&head->lock);
}

void sctp_put_port(struct sock *sk)
{
	sctp_local_bh_disable();
	__sctp_put_port(sk);
	sctp_local_bh_enable();
}

/*
 * The system picks an ephemeral port and choose an address set equivalent
 * to binding with a wildcard address.
 * One of those addresses will be the primary address for the association.
 * This automatically enables the multihoming capability of SCTP.
 */
static int sctp_autobind(struct sock *sk)
{
	union sctp_addr autoaddr;
	struct sctp_af *af;
	unsigned short port;

	/* Initialize a local sockaddr structure to INADDR_ANY. */
	af = sctp_sk(sk)->pf->af;

	port = htons(inet_sk(sk)->num);