ipv4.c

linux 内核源代码

/*
 *  net/dccp/ipv4.c
 *
 *  An implementation of the DCCP protocol
 *  Arnaldo Carvalho de Melo <[e-mail]>
 *
 *	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/dccp.h>
#include <linux/icmp.h>
#include <linux/module.h>
#include <linux/skbuff.h>
#include <linux/random.h>

#include <net/icmp.h>
#include <net/inet_common.h>
#include <net/inet_hashtables.h>
#include <net/inet_sock.h>
#include <net/protocol.h>
#include <net/sock.h>
#include <net/timewait_sock.h>
#include <net/tcp_states.h>
#include <net/xfrm.h>

#include "ackvec.h"
#include "ccid.h"
#include "dccp.h"
#include "feat.h"

/*
 * This is the global socket data structure used for responding to
 * the Out-of-the-blue (OOTB) packets. A control sock will be created
 * for this socket at the initialization time.
 */
static struct socket *dccp_v4_ctl_socket;

static int dccp_v4_get_port(struct sock *sk, const unsigned short snum)
{
	return inet_csk_get_port(&dccp_hashinfo, sk, snum,
				 inet_csk_bind_conflict);
}

int dccp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
{
	struct inet_sock *inet = inet_sk(sk);
	struct dccp_sock *dp = dccp_sk(sk);
	const struct sockaddr_in *usin = (struct sockaddr_in *)uaddr;
	struct rtable *rt;
	__be32 daddr, nexthop;
	int tmp;
	int err;

	dp->dccps_role = DCCP_ROLE_CLIENT;

	if (addr_len < sizeof(struct sockaddr_in))
		return -EINVAL;

	if (usin->sin_family != AF_INET)
		return -EAFNOSUPPORT;

	nexthop = daddr = usin->sin_addr.s_addr;
	if (inet->opt != NULL && inet->opt->srr) {
		if (daddr == 0)
			return -EINVAL;
		nexthop = inet->opt->faddr;
	}

	tmp = ip_route_connect(&rt, nexthop, inet->saddr,
			       RT_CONN_FLAGS(sk), sk->sk_bound_dev_if,
			       IPPROTO_DCCP,
			       inet->sport, usin->sin_port, sk, 1);
	if (tmp < 0)
		return tmp;

	if (rt->rt_flags & (RTCF_MULTICAST | RTCF_BROADCAST)) {
		ip_rt_put(rt);
		return -ENETUNREACH;
	}

	if (inet->opt == NULL || !inet->opt->srr)
		daddr = rt->rt_dst;

	if (inet->saddr == 0)
		inet->saddr = rt->rt_src;
	inet->rcv_saddr = inet->saddr;

	inet->dport = usin->sin_port;
	inet->daddr = daddr;

	inet_csk(sk)->icsk_ext_hdr_len = 0;
	if (inet->opt != NULL)
		inet_csk(sk)->icsk_ext_hdr_len = inet->opt->optlen;
	/*
	 * Socket identity is still unknown (sport may be zero).
	 * However we set state to DCCP_REQUESTING and not releasing socket
	 * lock select source port, enter ourselves into the hash tables and
	 * complete initialization after this.
	 */
	dccp_set_state(sk, DCCP_REQUESTING);
	err = inet_hash_connect(&dccp_death_row, sk);
	if (err != 0)
		goto failure;

	err = ip_route_newports(&rt, IPPROTO_DCCP, inet->sport, inet->dport,
				sk);
	if (err != 0)
		goto failure;

	/* OK, now commit destination to socket.  */
	sk_setup_caps(sk, &rt->u.dst);

	dp->dccps_iss = secure_dccp_sequence_number(inet->saddr, inet->daddr,
						    inet->sport, inet->dport);
	inet->id = dp->dccps_iss ^ jiffies;

	err = dccp_connect(sk);
	rt = NULL;
	if (err != 0)
		goto failure;
out:
	return err;
failure:
	/*
	 * This unhashes the socket and releases the local port, if necessary.
	 */
	dccp_set_state(sk, DCCP_CLOSED);
	ip_rt_put(rt);
	sk->sk_route_caps = 0;
	inet->dport = 0;
	goto out;
}

EXPORT_SYMBOL_GPL(dccp_v4_connect);

/*
 * This routine does path mtu discovery as defined in RFC1191.
 */
static inline void dccp_do_pmtu_discovery(struct sock *sk,
					  const struct iphdr *iph,
					  u32 mtu)
{
	struct dst_entry *dst;
	const struct inet_sock *inet = inet_sk(sk);
	const struct dccp_sock *dp = dccp_sk(sk);

	/* We are not interested in DCCP_LISTEN and request_socks (RESPONSEs
	 * send out by Linux are always < 576bytes so they should go through
	 * unfragmented).
	 */
	if (sk->sk_state == DCCP_LISTEN)
		return;

	/* We don't check in the destentry if pmtu discovery is forbidden
	 * on this route. We just assume that no packet_to_big packets
	 * are send back when pmtu discovery is not active.
	 * There is a small race when the user changes this flag in the
	 * route, but I think that's acceptable.
	 */
	if ((dst = __sk_dst_check(sk, 0)) == NULL)
		return;

	dst->ops->update_pmtu(dst, mtu);

	/* Something is about to be wrong... Remember soft error
	 * for the case, if this connection will not able to recover.
	 */
	if (mtu < dst_mtu(dst) && ip_dont_fragment(sk, dst))
		sk->sk_err_soft = EMSGSIZE;

	mtu = dst_mtu(dst);

	if (inet->pmtudisc != IP_PMTUDISC_DONT &&
	    inet_csk(sk)->icsk_pmtu_cookie > mtu) {
		dccp_sync_mss(sk, mtu);

		/*
		 * From RFC 4340, sec. 14.1:
		 *
		 *	DCCP-Sync packets are the best choice for upward
		 *	probing, since DCCP-Sync probes do not risk application
		 *	data loss.
		 */
		dccp_send_sync(sk, dp->dccps_gsr, DCCP_PKT_SYNC);
	} /* else let the usual retransmit timer handle it */
}

/*
 * This routine is called by the ICMP module when it gets some sort of error
 * condition. If err < 0 then the socket should be closed and the error
 * returned to the user. If err > 0 it's just the icmp type << 8 | icmp code.
 * After adjustment header points to the first 8 bytes of the tcp header. We
 * need to find the appropriate port.
 *
 * The locking strategy used here is very "optimistic". When someone else
 * accesses the socket the ICMP is just dropped and for some paths there is no
 * check at all. A more general error queue to queue errors for later handling
 * is probably better.
 */
static void dccp_v4_err(struct sk_buff *skb, u32 info)
{
	const struct iphdr *iph = (struct iphdr *)skb->data;
	const struct dccp_hdr *dh = (struct dccp_hdr *)(skb->data +
							(iph->ihl << 2));
	struct dccp_sock *dp;
	struct inet_sock *inet;
	const int type = icmp_hdr(skb)->type;
	const int code = icmp_hdr(skb)->code;
	struct sock *sk;
	__u64 seq;
	int err;

	if (skb->len < (iph->ihl << 2) + 8) {
		ICMP_INC_STATS_BH(ICMP_MIB_INERRORS);
		return;
	}

	sk = inet_lookup(&dccp_hashinfo, iph->daddr, dh->dccph_dport,
			 iph->saddr, dh->dccph_sport, inet_iif(skb));
	if (sk == NULL) {
		ICMP_INC_STATS_BH(ICMP_MIB_INERRORS);
		return;
	}

	if (sk->sk_state == DCCP_TIME_WAIT) {
		inet_twsk_put(inet_twsk(sk));
		return;
	}

	bh_lock_sock(sk);
	/* If too many ICMPs get dropped on busy
	 * servers this needs to be solved differently.
	 */
	if (sock_owned_by_user(sk))
		NET_INC_STATS_BH(LINUX_MIB_LOCKDROPPEDICMPS);

	if (sk->sk_state == DCCP_CLOSED)
		goto out;

	dp = dccp_sk(sk);
	seq = dccp_hdr_seq(dh);
	if ((1 << sk->sk_state) & ~(DCCPF_REQUESTING | DCCPF_LISTEN) &&
	    !between48(seq, dp->dccps_swl, dp->dccps_swh)) {
		NET_INC_STATS_BH(LINUX_MIB_OUTOFWINDOWICMPS);
		goto out;
	}

	switch (type) {
	case ICMP_SOURCE_QUENCH:
		/* Just silently ignore these. */
		goto out;
	case ICMP_PARAMETERPROB:
		err = EPROTO;
		break;
	case ICMP_DEST_UNREACH:
		if (code > NR_ICMP_UNREACH)
			goto out;

		if (code == ICMP_FRAG_NEEDED) { /* PMTU discovery (RFC1191) */
			if (!sock_owned_by_user(sk))
				dccp_do_pmtu_discovery(sk, iph, info);
			goto out;
		}

		err = icmp_err_convert[code].errno;
		break;
	case ICMP_TIME_EXCEEDED:
		err = EHOSTUNREACH;
		break;
	default:
		goto out;
	}

	switch (sk->sk_state) {
		struct request_sock *req , **prev;
	case DCCP_LISTEN:
		if (sock_owned_by_user(sk))
			goto out;
		req = inet_csk_search_req(sk, &prev, dh->dccph_dport,
					  iph->daddr, iph->saddr);
		if (!req)
			goto out;

		/*
		 * ICMPs are not backlogged, hence we cannot get an established
		 * socket here.
		 */
		BUG_TRAP(!req->sk);

		if (seq != dccp_rsk(req)->dreq_iss) {
			NET_INC_STATS_BH(LINUX_MIB_OUTOFWINDOWICMPS);
			goto out;
		}
		/*
		 * Still in RESPOND, just remove it silently.
		 * There is no good way to pass the error to the newly
		 * created socket, and POSIX does not want network
		 * errors returned from accept().
		 */
		inet_csk_reqsk_queue_drop(sk, req, prev);
		goto out;

	case DCCP_REQUESTING:
	case DCCP_RESPOND:
		if (!sock_owned_by_user(sk)) {
			DCCP_INC_STATS_BH(DCCP_MIB_ATTEMPTFAILS);
			sk->sk_err = err;

			sk->sk_error_report(sk);

			dccp_done(sk);
		} else
			sk->sk_err_soft = err;
		goto out;
	}

	/* If we've already connected we will keep trying
	 * until we time out, or the user gives up.
	 *
	 * rfc1122 4.2.3.9 allows to consider as hard errors
	 * only PROTO_UNREACH and PORT_UNREACH (well, FRAG_FAILED too,
	 * but it is obsoleted by pmtu discovery).
	 *
	 * Note, that in modern internet, where routing is unreliable
	 * and in each dark corner broken firewalls sit, sending random
	 * errors ordered by their masters even this two messages finally lose
	 * their original sense (even Linux sends invalid PORT_UNREACHs)
	 *
	 * Now we are in compliance with RFCs.
	 *							--ANK (980905)
	 */

	inet = inet_sk(sk);
	if (!sock_owned_by_user(sk) && inet->recverr) {
		sk->sk_err = err;
		sk->sk_error_report(sk);
	} else /* Only an error on timeout */
		sk->sk_err_soft = err;
out:
	bh_unlock_sock(sk);
	sock_put(sk);
}

static inline __sum16 dccp_v4_csum_finish(struct sk_buff *skb,
				      __be32 src, __be32 dst)
{
	return csum_tcpudp_magic(src, dst, skb->len, IPPROTO_DCCP, skb->csum);
}

void dccp_v4_send_check(struct sock *sk, int unused, struct sk_buff *skb)
{
	const struct inet_sock *inet = inet_sk(sk);
	struct dccp_hdr *dh = dccp_hdr(skb);

	dccp_csum_outgoing(skb);
	dh->dccph_checksum = dccp_v4_csum_finish(skb, inet->saddr, inet->daddr);
}

EXPORT_SYMBOL_GPL(dccp_v4_send_check);

static inline u64 dccp_v4_init_sequence(const struct sk_buff *skb)
{
	return secure_dccp_sequence_number(ip_hdr(skb)->daddr,
					   ip_hdr(skb)->saddr,
					   dccp_hdr(skb)->dccph_dport,
					   dccp_hdr(skb)->dccph_sport);
}

/*
 * The three way handshake has completed - we got a valid ACK or DATAACK -
 * now create the new socket.
 *
 * This is the equivalent of TCP's tcp_v4_syn_recv_sock
 */
struct sock *dccp_v4_request_recv_sock(struct sock *sk, struct sk_buff *skb,
				       struct request_sock *req,
				       struct dst_entry *dst)
{
	struct inet_request_sock *ireq;
	struct inet_sock *newinet;
	struct sock *newsk;

	if (sk_acceptq_is_full(sk))
		goto exit_overflow;

	if (dst == NULL && (dst = inet_csk_route_req(sk, req)) == NULL)
		goto exit;

	newsk = dccp_create_openreq_child(sk, req, skb);
	if (newsk == NULL)
		goto exit;

	sk_setup_caps(newsk, dst);

	newinet		   = inet_sk(newsk);
	ireq		   = inet_rsk(req);
	newinet->daddr	   = ireq->rmt_addr;
	newinet->rcv_saddr = ireq->loc_addr;
	newinet->saddr	   = ireq->loc_addr;
	newinet->opt	   = ireq->opt;
	ireq->opt	   = NULL;
	newinet->mc_index  = inet_iif(skb);
	newinet->mc_ttl	   = ip_hdr(skb)->ttl;
	newinet->id	   = jiffies;

	dccp_sync_mss(newsk, dst_mtu(dst));

	__inet_hash(&dccp_hashinfo, newsk, 0);
	__inet_inherit_port(&dccp_hashinfo, sk, newsk);

	return newsk;

exit_overflow:
	NET_INC_STATS_BH(LINUX_MIB_LISTENOVERFLOWS);
exit:
	NET_INC_STATS_BH(LINUX_MIB_LISTENDROPS);
	dst_release(dst);
	return NULL;
}

EXPORT_SYMBOL_GPL(dccp_v4_request_recv_sock);

static struct sock *dccp_v4_hnd_req(struct sock *sk, struct sk_buff *skb)
{
	const struct dccp_hdr *dh = dccp_hdr(skb);
	const struct iphdr *iph = ip_hdr(skb);
	struct sock *nsk;
	struct request_sock **prev;
	/* Find possible connection requests. */
	struct request_sock *req = inet_csk_search_req(sk, &prev,
						       dh->dccph_sport,
						       iph->saddr, iph->daddr);
	if (req != NULL)
		return dccp_check_req(sk, skb, req, prev);

	nsk = inet_lookup_established(&dccp_hashinfo,
				      iph->saddr, dh->dccph_sport,
				      iph->daddr, dh->dccph_dport,
				      inet_iif(skb));
	if (nsk != NULL) {
		if (nsk->sk_state != DCCP_TIME_WAIT) {
			bh_lock_sock(nsk);
			return nsk;
		}
		inet_twsk_put(inet_twsk(nsk));
		return NULL;
	}

	return sk;
}

static struct dst_entry* dccp_v4_route_skb(struct sock *sk,
					   struct sk_buff *skb)
{
	struct rtable *rt;
	struct flowi fl = { .oif = ((struct rtable *)skb->dst)->rt_iif,
			    .nl_u = { .ip4_u =
				      { .daddr = ip_hdr(skb)->saddr,
					.saddr = ip_hdr(skb)->daddr,
					.tos = RT_CONN_FLAGS(sk) } },
			    .proto = sk->sk_protocol,
			    .uli_u = { .ports =
				       { .sport = dccp_hdr(skb)->dccph_dport,
					 .dport = dccp_hdr(skb)->dccph_sport }
				     }
			  };

	security_skb_classify_flow(skb, &fl);
	if (ip_route_output_flow(&rt, &fl, sk, 0)) {
		IP_INC_STATS_BH(IPSTATS_MIB_OUTNOROUTES);
		return NULL;
	}

	return &rt->u.dst;
}

static int dccp_v4_send_response(struct sock *sk, struct request_sock *req,
				 struct dst_entry *dst)
{
	int err = -1;
	struct sk_buff *skb;

	/* First, grab a route. */

	if (dst == NULL && (dst = inet_csk_route_req(sk, req)) == NULL)
		goto out;

	skb = dccp_make_response(sk, dst, req);
	if (skb != NULL) {
		const struct inet_request_sock *ireq = inet_rsk(req);
		struct dccp_hdr *dh = dccp_hdr(skb);

		dh->dccph_checksum = dccp_v4_csum_finish(skb, ireq->loc_addr,
							      ireq->rmt_addr);
		memset(&(IPCB(skb)->opt), 0, sizeof(IPCB(skb)->opt));
		err = ip_build_and_send_pkt(skb, sk, ireq->loc_addr,
					    ireq->rmt_addr,
					    ireq->opt);
		err = net_xmit_eval(err);
	}

out:
	dst_release(dst);
	return err;
}

static void dccp_v4_ctl_send_reset(struct sock *sk, struct sk_buff *rxskb)
{
	int err;
	const struct iphdr *rxiph;
	struct sk_buff *skb;
	struct dst_entry *dst;

	/* Never send a reset in response to a reset. */
	if (dccp_hdr(rxskb)->dccph_type == DCCP_PKT_RESET)
		return;

	if (((struct rtable *)rxskb->dst)->rt_type != RTN_LOCAL)
		return;

	dst = dccp_v4_route_skb(dccp_v4_ctl_socket->sk, rxskb);