raw.c

linux 内核源代码

/*
 *	RAW sockets for IPv6
 *	Linux INET6 implementation
 *
 *	Authors:
 *	Pedro Roque		<[e-mail]>
 *
 *	Adapted from linux/net/ipv4/raw.c
 *
 *	$Id: raw.c,v 1.51 2002/02/01 22:01:04 davem Exp $
 *
 *	Fixes:
 *	Hideaki YOSHIFUJI	:	sin6_scope_id support
 *	YOSHIFUJI,H.@USAGI	:	raw checksum (RFC2292(bis) compliance)
 *	Kazunori MIYAZAWA @USAGI:	change process style to use ip6_append_data
 *
 *	This program is free software; you can redistribute it and/or
 *      modify it under the terms of the GNU General Public License
 *      as published by the Free Software Foundation; either version
 *      2 of the License, or (at your option) any later version.
 */

#include <linux/errno.h>
#include <linux/types.h>
#include <linux/socket.h>
#include <linux/sockios.h>
#include <linux/net.h>
#include <linux/in6.h>
#include <linux/netdevice.h>
#include <linux/if_arp.h>
#include <linux/icmpv6.h>
#include <linux/netfilter.h>
#include <linux/netfilter_ipv6.h>
#include <linux/skbuff.h>
#include <asm/uaccess.h>
#include <asm/ioctls.h>

#include <net/net_namespace.h>
#include <net/ip.h>
#include <net/sock.h>
#include <net/snmp.h>

#include <net/ipv6.h>
#include <net/ndisc.h>
#include <net/protocol.h>
#include <net/ip6_route.h>
#include <net/ip6_checksum.h>
#include <net/addrconf.h>
#include <net/transp_v6.h>
#include <net/udp.h>
#include <net/inet_common.h>
#include <net/tcp_states.h>
#if defined(CONFIG_IPV6_MIP6) || defined(CONFIG_IPV6_MIP6_MODULE)
#include <net/mip6.h>
#endif

#include <net/rawv6.h>
#include <net/xfrm.h>

#include <linux/proc_fs.h>
#include <linux/seq_file.h>

struct hlist_head raw_v6_htable[RAWV6_HTABLE_SIZE];
DEFINE_RWLOCK(raw_v6_lock);

static void raw_v6_hash(struct sock *sk)
{
	struct hlist_head *list = &raw_v6_htable[inet_sk(sk)->num &
						 (RAWV6_HTABLE_SIZE - 1)];

	write_lock_bh(&raw_v6_lock);
	sk_add_node(sk, list);
	sock_prot_inc_use(sk->sk_prot);
	write_unlock_bh(&raw_v6_lock);
}

static void raw_v6_unhash(struct sock *sk)
{
	write_lock_bh(&raw_v6_lock);
	if (sk_del_node_init(sk))
		sock_prot_dec_use(sk->sk_prot);
	write_unlock_bh(&raw_v6_lock);
}


/* Grumble... icmp and ip_input want to get at this... */
struct sock *__raw_v6_lookup(struct sock *sk, unsigned short num,
			     struct in6_addr *loc_addr, struct in6_addr *rmt_addr,
			     int dif)
{
	struct hlist_node *node;
	int is_multicast = ipv6_addr_is_multicast(loc_addr);

	sk_for_each_from(sk, node)
		if (inet_sk(sk)->num == num) {
			struct ipv6_pinfo *np = inet6_sk(sk);

			if (!ipv6_addr_any(&np->daddr) &&
			    !ipv6_addr_equal(&np->daddr, rmt_addr))
				continue;

			if (sk->sk_bound_dev_if && sk->sk_bound_dev_if != dif)
				continue;

			if (!ipv6_addr_any(&np->rcv_saddr)) {
				if (ipv6_addr_equal(&np->rcv_saddr, loc_addr))
					goto found;
				if (is_multicast &&
				    inet6_mc_check(sk, loc_addr, rmt_addr))
					goto found;
				continue;
			}
			goto found;
		}
	sk = NULL;
found:
	return sk;
}

/*
 *	0 - deliver
 *	1 - block
 */
static __inline__ int icmpv6_filter(struct sock *sk, struct sk_buff *skb)
{
	struct icmp6hdr *icmph;
	struct raw6_sock *rp = raw6_sk(sk);

	if (pskb_may_pull(skb, sizeof(struct icmp6hdr))) {
		__u32 *data = &rp->filter.data[0];
		int bit_nr;

		icmph = (struct icmp6hdr *) skb->data;
		bit_nr = icmph->icmp6_type;

		return (data[bit_nr >> 5] & (1 << (bit_nr & 31))) != 0;
	}
	return 0;
}

#if defined(CONFIG_IPV6_MIP6) || defined(CONFIG_IPV6_MIP6_MODULE)
static int (*mh_filter)(struct sock *sock, struct sk_buff *skb);

int rawv6_mh_filter_register(int (*filter)(struct sock *sock,
					   struct sk_buff *skb))
{
	rcu_assign_pointer(mh_filter, filter);
	return 0;
}
EXPORT_SYMBOL(rawv6_mh_filter_register);

int rawv6_mh_filter_unregister(int (*filter)(struct sock *sock,
					     struct sk_buff *skb))
{
	rcu_assign_pointer(mh_filter, NULL);
	synchronize_rcu();
	return 0;
}
EXPORT_SYMBOL(rawv6_mh_filter_unregister);

#endif

/*
 *	demultiplex raw sockets.
 *	(should consider queueing the skb in the sock receive_queue
 *	without calling rawv6.c)
 *
 *	Caller owns SKB so we must make clones.
 */
int ipv6_raw_deliver(struct sk_buff *skb, int nexthdr)
{
	struct in6_addr *saddr;
	struct in6_addr *daddr;
	struct sock *sk;
	int delivered = 0;
	__u8 hash;

	saddr = &ipv6_hdr(skb)->saddr;
	daddr = saddr + 1;

	hash = nexthdr & (MAX_INET_PROTOS - 1);

	read_lock(&raw_v6_lock);
	sk = sk_head(&raw_v6_htable[hash]);

	/*
	 *	The first socket found will be delivered after
	 *	delivery to transport protocols.
	 */

	if (sk == NULL)
		goto out;

	sk = __raw_v6_lookup(sk, nexthdr, daddr, saddr, IP6CB(skb)->iif);

	while (sk) {
		int filtered;

		delivered = 1;
		switch (nexthdr) {
		case IPPROTO_ICMPV6:
			filtered = icmpv6_filter(sk, skb);
			break;

#if defined(CONFIG_IPV6_MIP6) || defined(CONFIG_IPV6_MIP6_MODULE)
		case IPPROTO_MH:
		{
			/* XXX: To validate MH only once for each packet,
			 * this is placed here. It should be after checking
			 * xfrm policy, however it doesn't. The checking xfrm
			 * policy is placed in rawv6_rcv() because it is
			 * required for each socket.
			 */
			int (*filter)(struct sock *sock, struct sk_buff *skb);

			filter = rcu_dereference(mh_filter);
			filtered = filter ? filter(sk, skb) : 0;
			break;
		}
#endif
		default:
			filtered = 0;
			break;
		}

		if (filtered < 0)
			break;
		if (filtered == 0) {
			struct sk_buff *clone = skb_clone(skb, GFP_ATOMIC);

			/* Not releasing hash table! */
			if (clone) {
				nf_reset(clone);
				rawv6_rcv(sk, clone);
			}
		}
		sk = __raw_v6_lookup(sk_next(sk), nexthdr, daddr, saddr,
				     IP6CB(skb)->iif);
	}
out:
	read_unlock(&raw_v6_lock);
	return delivered;
}

/* This cleans up af_inet6 a bit. -DaveM */
static int rawv6_bind(struct sock *sk, struct sockaddr *uaddr, int addr_len)
{
	struct inet_sock *inet = inet_sk(sk);
	struct ipv6_pinfo *np = inet6_sk(sk);
	struct sockaddr_in6 *addr = (struct sockaddr_in6 *) uaddr;
	__be32 v4addr = 0;
	int addr_type;
	int err;

	if (addr_len < SIN6_LEN_RFC2133)
		return -EINVAL;
	addr_type = ipv6_addr_type(&addr->sin6_addr);

	/* Raw sockets are IPv6 only */
	if (addr_type == IPV6_ADDR_MAPPED)
		return(-EADDRNOTAVAIL);

	lock_sock(sk);

	err = -EINVAL;
	if (sk->sk_state != TCP_CLOSE)
		goto out;

	/* Check if the address belongs to the host. */
	if (addr_type != IPV6_ADDR_ANY) {
		struct net_device *dev = NULL;

		if (addr_type & IPV6_ADDR_LINKLOCAL) {
			if (addr_len >= sizeof(struct sockaddr_in6) &&
			    addr->sin6_scope_id) {
				/* Override any existing binding, if another
				 * one is supplied by user.
				 */
				sk->sk_bound_dev_if = addr->sin6_scope_id;
			}

			/* Binding to link-local address requires an interface */
			if (!sk->sk_bound_dev_if)
				goto out;

			dev = dev_get_by_index(&init_net, sk->sk_bound_dev_if);
			if (!dev) {
				err = -ENODEV;
				goto out;
			}
		}

		/* ipv4 addr of the socket is invalid.  Only the
		 * unspecified and mapped address have a v4 equivalent.
		 */
		v4addr = LOOPBACK4_IPV6;
		if (!(addr_type & IPV6_ADDR_MULTICAST))	{
			err = -EADDRNOTAVAIL;
			if (!ipv6_chk_addr(&addr->sin6_addr, dev, 0)) {
				if (dev)
					dev_put(dev);
				goto out;
			}
		}
		if (dev)
			dev_put(dev);
	}

	inet->rcv_saddr = inet->saddr = v4addr;
	ipv6_addr_copy(&np->rcv_saddr, &addr->sin6_addr);
	if (!(addr_type & IPV6_ADDR_MULTICAST))
		ipv6_addr_copy(&np->saddr, &addr->sin6_addr);
	err = 0;
out:
	release_sock(sk);
	return err;
}

void rawv6_err(struct sock *sk, struct sk_buff *skb,
	       struct inet6_skb_parm *opt,
	       int type, int code, int offset, __be32 info)
{
	struct inet_sock *inet = inet_sk(sk);
	struct ipv6_pinfo *np = inet6_sk(sk);
	int err;
	int harderr;

	/* Report error on raw socket, if:
	   1. User requested recverr.
	   2. Socket is connected (otherwise the error indication
	      is useless without recverr and error is hard.
	 */
	if (!np->recverr && sk->sk_state != TCP_ESTABLISHED)
		return;

	harderr = icmpv6_err_convert(type, code, &err);
	if (type == ICMPV6_PKT_TOOBIG)
		harderr = (np->pmtudisc == IPV6_PMTUDISC_DO);

	if (np->recverr) {
		u8 *payload = skb->data;
		if (!inet->hdrincl)
			payload += offset;
		ipv6_icmp_error(sk, skb, err, 0, ntohl(info), payload);
	}

	if (np->recverr || harderr) {
		sk->sk_err = err;
		sk->sk_error_report(sk);
	}
}

static inline int rawv6_rcv_skb(struct sock * sk, struct sk_buff * skb)
{
	if ((raw6_sk(sk)->checksum || sk->sk_filter) &&
	    skb_checksum_complete(skb)) {
		/* FIXME: increment a raw6 drops counter here */
		kfree_skb(skb);
		return 0;
	}

	/* Charge it to the socket. */
	if (sock_queue_rcv_skb(sk,skb)<0) {
		/* FIXME: increment a raw6 drops counter here */
		kfree_skb(skb);
		return 0;
	}

	return 0;
}

/*
 *	This is next to useless...
 *	if we demultiplex in network layer we don't need the extra call
 *	just to queue the skb...
 *	maybe we could have the network decide upon a hint if it
 *	should call raw_rcv for demultiplexing
 */
int rawv6_rcv(struct sock *sk, struct sk_buff *skb)
{
	struct inet_sock *inet = inet_sk(sk);
	struct raw6_sock *rp = raw6_sk(sk);

	if (!xfrm6_policy_check(sk, XFRM_POLICY_IN, skb)) {
		kfree_skb(skb);
		return NET_RX_DROP;
	}

	if (!rp->checksum)
		skb->ip_summed = CHECKSUM_UNNECESSARY;

	if (skb->ip_summed == CHECKSUM_COMPLETE) {
		skb_postpull_rcsum(skb, skb_network_header(skb),
				   skb_network_header_len(skb));
		if (!csum_ipv6_magic(&ipv6_hdr(skb)->saddr,
				     &ipv6_hdr(skb)->daddr,
				     skb->len, inet->num, skb->csum))
			skb->ip_summed = CHECKSUM_UNNECESSARY;
	}
	if (!skb_csum_unnecessary(skb))
		skb->csum = ~csum_unfold(csum_ipv6_magic(&ipv6_hdr(skb)->saddr,
							 &ipv6_hdr(skb)->daddr,
							 skb->len,
							 inet->num, 0));

	if (inet->hdrincl) {
		if (skb_checksum_complete(skb)) {
			/* FIXME: increment a raw6 drops counter here */
			kfree_skb(skb);
			return 0;
		}
	}

	rawv6_rcv_skb(sk, skb);
	return 0;
}


/*
 *	This should be easy, if there is something there
 *	we return it, otherwise we block.
 */

static int rawv6_recvmsg(struct kiocb *iocb, struct sock *sk,
		  struct msghdr *msg, size_t len,
		  int noblock, int flags, int *addr_len)
{
	struct ipv6_pinfo *np = inet6_sk(sk);
	struct sockaddr_in6 *sin6 = (struct sockaddr_in6 *)msg->msg_name;
	struct sk_buff *skb;
	size_t copied;
	int err;

	if (flags & MSG_OOB)
		return -EOPNOTSUPP;

	if (addr_len)
		*addr_len=sizeof(*sin6);

	if (flags & MSG_ERRQUEUE)
		return ipv6_recv_error(sk, msg, len);

	skb = skb_recv_datagram(sk, flags, noblock, &err);
	if (!skb)
		goto out;

	copied = skb->len;
	if (copied > len) {
		copied = len;
		msg->msg_flags |= MSG_TRUNC;
	}

	if (skb_csum_unnecessary(skb)) {
		err = skb_copy_datagram_iovec(skb, 0, msg->msg_iov, copied);
	} else if (msg->msg_flags&MSG_TRUNC) {
		if (__skb_checksum_complete(skb))
			goto csum_copy_err;
		err = skb_copy_datagram_iovec(skb, 0, msg->msg_iov, copied);
	} else {
		err = skb_copy_and_csum_datagram_iovec(skb, 0, msg->msg_iov);
		if (err == -EINVAL)
			goto csum_copy_err;
	}
	if (err)
		goto out_free;

	/* Copy the address. */
	if (sin6) {
		sin6->sin6_family = AF_INET6;
		sin6->sin6_port = 0;
		ipv6_addr_copy(&sin6->sin6_addr, &ipv6_hdr(skb)->saddr);
		sin6->sin6_flowinfo = 0;
		sin6->sin6_scope_id = 0;
		if (ipv6_addr_type(&sin6->sin6_addr) & IPV6_ADDR_LINKLOCAL)
			sin6->sin6_scope_id = IP6CB(skb)->iif;
	}

	sock_recv_timestamp(msg, sk, skb);

	if (np->rxopt.all)
		datagram_recv_ctl(sk, msg, skb);

	err = copied;
	if (flags & MSG_TRUNC)
		err = skb->len;

out_free:
	skb_free_datagram(sk, skb);
out:
	return err;

csum_copy_err:
	skb_kill_datagram(sk, skb, flags);

	/* Error for blocking case is chosen to masquerade
	   as some normal condition.
	 */
	err = (flags&MSG_DONTWAIT) ? -EAGAIN : -EHOSTUNREACH;
	/* FIXME: increment a raw6 drops counter here */
	goto out;
}

static int rawv6_push_pending_frames(struct sock *sk, struct flowi *fl,
				     struct raw6_sock *rp)
{
	struct sk_buff *skb;
	int err = 0;
	int offset;
	int len;
	int total_len;
	__wsum tmp_csum;
	__sum16 csum;

	if (!rp->checksum)
		goto send;

	if ((skb = skb_peek(&sk->sk_write_queue)) == NULL)
		goto out;

	offset = rp->offset;
	total_len = inet_sk(sk)->cork.length - (skb_network_header(skb) -
						skb->data);
	if (offset >= total_len - 1) {
		err = -EINVAL;
		ip6_flush_pending_frames(sk);
		goto out;
	}

	/* should be check HW csum miyazawa */
	if (skb_queue_len(&sk->sk_write_queue) == 1) {
		/*
		 * Only one fragment on the socket.
		 */
		tmp_csum = skb->csum;
	} else {
		struct sk_buff *csum_skb = NULL;
		tmp_csum = 0;

		skb_queue_walk(&sk->sk_write_queue, skb) {
			tmp_csum = csum_add(tmp_csum, skb->csum);

			if (csum_skb)
				continue;

			len = skb->len - skb_transport_offset(skb);
			if (offset >= len) {
				offset -= len;
				continue;
			}

			csum_skb = skb;
		}

		skb = csum_skb;
	}

	offset += skb_transport_offset(skb);
	if (skb_copy_bits(skb, offset, &csum, 2))
		BUG();

	/* in case cksum was not initialized */
	if (unlikely(csum))
		tmp_csum = csum_sub(tmp_csum, csum_unfold(csum));

	csum = csum_ipv6_magic(&fl->fl6_src,
				   &fl->fl6_dst,
				   total_len, fl->proto, tmp_csum);

	if (csum == 0 && fl->proto == IPPROTO_UDP)
		csum = CSUM_MANGLED_0;

	if (skb_store_bits(skb, offset, &csum, 2))
		BUG();

send:
	err = ip6_push_pending_frames(sk);
out:
	return err;
}

static int rawv6_send_hdrinc(struct sock *sk, void *from, int length,
			struct flowi *fl, struct rt6_info *rt,
			unsigned int flags)
{
	struct ipv6_pinfo *np = inet6_sk(sk);
	struct ipv6hdr *iph;
	struct sk_buff *skb;
	unsigned int hh_len;
	int err;

	if (length > rt->u.dst.dev->mtu) {
		ipv6_local_error(sk, EMSGSIZE, fl, rt->u.dst.dev->mtu);
		return -EMSGSIZE;
	}
	if (flags&MSG_PROBE)
		goto out;

	hh_len = LL_RESERVED_SPACE(rt->u.dst.dev);

	skb = sock_alloc_send_skb(sk, length+hh_len+15,
				  flags&MSG_DONTWAIT, &err);
	if (skb == NULL)
		goto error;
	skb_reserve(skb, hh_len);

	skb->priority = sk->sk_priority;
	skb->dst = dst_clone(&rt->u.dst);

	skb_put(skb, length);
	skb_reset_network_header(skb);
	iph = ipv6_hdr(skb);

	skb->ip_summed = CHECKSUM_NONE;

	skb->transport_header = skb->network_header;
	err = memcpy_fromiovecend((void *)iph, from, 0, length);
	if (err)
		goto error_fault;

	IP6_INC_STATS(rt->rt6i_idev, IPSTATS_MIB_OUTREQUESTS);
	err = NF_HOOK(PF_INET6, NF_IP6_LOCAL_OUT, skb, NULL, rt->u.dst.dev,
		      dst_output);
	if (err > 0)
		err = np->recverr ? net_xmit_errno(err) : 0;
	if (err)
		goto error;
out:
	return 0;

error_fault:
	err = -EFAULT;
	kfree_skb(skb);
error:
	IP6_INC_STATS(rt->rt6i_idev, IPSTATS_MIB_OUTDISCARDS);
	return err;
}

static int rawv6_probe_proto_opt(struct flowi *fl, struct msghdr *msg)
{
	struct iovec *iov;
	u8 __user *type = NULL;
	u8 __user *code = NULL;
	u8 len = 0;
	int probed = 0;
	int i;

	if (!msg->msg_iov)
		return 0;

	for (i = 0; i < msg->msg_iovlen; i++) {
		iov = &msg->msg_iov[i];
		if (!iov)
			continue;

		switch (fl->proto) {
		case IPPROTO_ICMPV6:
			/* check if one-byte field is readable or not. */
			if (iov->iov_base && iov->iov_len < 1)
				break;