tcp_ipv6.c

嵌入式系统设计与实验教材二源码linux内核移植与编译

	kfree_skb(skb);
	return 0;

discard_and_relse:
	sock_put(sk);
	goto discard_it;

do_time_wait:
	if (skb->len < (th->doff<<2) || tcp_checksum_complete(skb)) {
		TCP_INC_STATS_BH(TcpInErrs);
		sock_put(sk);
		goto discard_it;
	}

	switch(tcp_timewait_state_process((struct tcp_tw_bucket *)sk,
					  skb, th, skb->len)) {
	case TCP_TW_SYN:
	{
		struct sock *sk2;

		sk2 = tcp_v6_lookup_listener(&skb->nh.ipv6h->daddr, ntohs(th->dest), tcp_v6_iif(skb));
		if (sk2 != NULL) {
			tcp_tw_deschedule((struct tcp_tw_bucket *)sk);
			tcp_timewait_kill((struct tcp_tw_bucket *)sk);
			tcp_tw_put((struct tcp_tw_bucket *)sk);
			sk = sk2;
			goto process;
		}
		/* Fall through to ACK */
	}
	case TCP_TW_ACK:
		tcp_v6_timewait_ack(sk, skb);
		break;
	case TCP_TW_RST:
		goto no_tcp_socket;
	case TCP_TW_SUCCESS:;
	}
	goto discard_it;
}

static int tcp_v6_rebuild_header(struct sock *sk)
{
	int err;
	struct dst_entry *dst;
	struct ipv6_pinfo *np = &sk->net_pinfo.af_inet6;

	dst = __sk_dst_check(sk, np->dst_cookie);

	if (dst == NULL) {
		struct flowi fl;

		fl.proto = IPPROTO_TCP;
		fl.nl_u.ip6_u.daddr = &np->daddr;
		fl.nl_u.ip6_u.saddr = &np->saddr;
		fl.fl6_flowlabel = np->flow_label;
		fl.oif = sk->bound_dev_if;
		fl.uli_u.ports.dport = sk->dport;
		fl.uli_u.ports.sport = sk->sport;

		if (np->opt && np->opt->srcrt) {
			struct rt0_hdr *rt0 = (struct rt0_hdr *) np->opt->srcrt;
			fl.nl_u.ip6_u.daddr = rt0->addr;
		}

		dst = ip6_route_output(sk, &fl);

		if (dst->error) {
			err = dst->error;
			dst_release(dst);
			sk->route_caps = 0;
			return err;
		}

		ip6_dst_store(sk, dst, NULL);
		sk->route_caps = dst->dev->features&~NETIF_F_IP_CSUM;
	}

	return 0;
}

static int tcp_v6_xmit(struct sk_buff *skb)
{
	struct sock *sk = skb->sk;
	struct ipv6_pinfo * np = &sk->net_pinfo.af_inet6;
	struct flowi fl;
	struct dst_entry *dst;

	fl.proto = IPPROTO_TCP;
	fl.fl6_dst = &np->daddr;
	fl.fl6_src = &np->saddr;
	fl.fl6_flowlabel = np->flow_label;
	IP6_ECN_flow_xmit(sk, fl.fl6_flowlabel);
	fl.oif = sk->bound_dev_if;
	fl.uli_u.ports.sport = sk->sport;
	fl.uli_u.ports.dport = sk->dport;

	if (np->opt && np->opt->srcrt) {
		struct rt0_hdr *rt0 = (struct rt0_hdr *) np->opt->srcrt;
		fl.nl_u.ip6_u.daddr = rt0->addr;
	}

	dst = __sk_dst_check(sk, np->dst_cookie);

	if (dst == NULL) {
		dst = ip6_route_output(sk, &fl);

		if (dst->error) {
			sk->err_soft = -dst->error;
			dst_release(dst);
			return -sk->err_soft;
		}

		ip6_dst_store(sk, dst, NULL);
	}

	skb->dst = dst_clone(dst);

	/* Restore final destination back after routing done */
	fl.nl_u.ip6_u.daddr = &np->daddr;

	return ip6_xmit(sk, skb, &fl, np->opt);
}

static void v6_addr2sockaddr(struct sock *sk, struct sockaddr * uaddr)
{
	struct ipv6_pinfo * np = &sk->net_pinfo.af_inet6;
	struct sockaddr_in6 *sin6 = (struct sockaddr_in6 *) uaddr;

	sin6->sin6_family = AF_INET6;
	memcpy(&sin6->sin6_addr, &np->daddr, sizeof(struct in6_addr));
	sin6->sin6_port	= sk->dport;
	/* We do not store received flowlabel for TCP */
	sin6->sin6_flowinfo = 0;
	sin6->sin6_scope_id = 0;
	if (sk->bound_dev_if && ipv6_addr_type(&sin6->sin6_addr)&IPV6_ADDR_LINKLOCAL)
		sin6->sin6_scope_id = sk->bound_dev_if;
}

static int tcp_v6_remember_stamp(struct sock *sk)
{
	/* Alas, not yet... */
	return 0;
}

static struct tcp_func ipv6_specific = {
	tcp_v6_xmit,
	tcp_v6_send_check,
	tcp_v6_rebuild_header,
	tcp_v6_conn_request,
	tcp_v6_syn_recv_sock,
	tcp_v6_hash_connecting,
	tcp_v6_remember_stamp,
	sizeof(struct ipv6hdr),

	ipv6_setsockopt,
	ipv6_getsockopt,
	v6_addr2sockaddr,
	sizeof(struct sockaddr_in6)
};

/*
 *	TCP over IPv4 via INET6 API
 */

static struct tcp_func ipv6_mapped = {
	ip_queue_xmit,
	tcp_v4_send_check,
	tcp_v4_rebuild_header,
	tcp_v6_conn_request,
	tcp_v6_syn_recv_sock,
	tcp_v4_hash_connecting,
	tcp_v4_remember_stamp,
	sizeof(struct iphdr),

	ipv6_setsockopt,
	ipv6_getsockopt,
	v6_addr2sockaddr,
	sizeof(struct sockaddr_in6)
};



/* NOTE: A lot of things set to zero explicitly by call to
 *       sk_alloc() so need not be done here.
 */
static int tcp_v6_init_sock(struct sock *sk)
{
	struct tcp_opt *tp = &(sk->tp_pinfo.af_tcp);

	skb_queue_head_init(&tp->out_of_order_queue);
	tcp_init_xmit_timers(sk);
	tcp_prequeue_init(tp);

	tp->rto  = TCP_TIMEOUT_INIT;
	tp->mdev = TCP_TIMEOUT_INIT;

	/* So many TCP implementations out there (incorrectly) count the
	 * initial SYN frame in their delayed-ACK and congestion control
	 * algorithms that we must have the following bandaid to talk
	 * efficiently to them.  -DaveM
	 */
	tp->snd_cwnd = 2;

	/* See draft-stevens-tcpca-spec-01 for discussion of the
	 * initialization of these values.
	 */
	tp->snd_ssthresh = 0x7fffffff;
	tp->snd_cwnd_clamp = ~0;
	tp->mss_cache = 536;

	tp->reordering = sysctl_tcp_reordering;

	sk->state = TCP_CLOSE;

	sk->tp_pinfo.af_tcp.af_specific = &ipv6_specific;

	sk->write_space = tcp_write_space;
	sk->use_write_queue = 1;

	sk->sndbuf = sysctl_tcp_wmem[1];
	sk->rcvbuf = sysctl_tcp_rmem[1];

	atomic_inc(&tcp_sockets_allocated);

	return 0;
}

static int tcp_v6_destroy_sock(struct sock *sk)
{
	struct tcp_opt *tp = &(sk->tp_pinfo.af_tcp);

	tcp_clear_xmit_timers(sk);

	/* Cleanup up the write buffer. */
  	tcp_writequeue_purge(sk);

	/* Cleans up our, hopefully empty, out_of_order_queue. */
  	__skb_queue_purge(&tp->out_of_order_queue);

	/* Clean prequeue, it must be empty really */
	__skb_queue_purge(&tp->ucopy.prequeue);

	/* Clean up a referenced TCP bind bucket. */
	if(sk->prev != NULL)
		tcp_put_port(sk);

	/* If sendmsg cached page exists, toss it. */
	if (tp->sndmsg_page != NULL)
		__free_page(tp->sndmsg_page);

	atomic_dec(&tcp_sockets_allocated);

	return inet6_destroy_sock(sk);
}

/* Proc filesystem TCPv6 sock list dumping. */
static void get_openreq6(struct sock *sk, struct open_request *req, char *tmpbuf, int i, int uid)
{
	struct in6_addr *dest, *src;
	int ttd = req->expires - jiffies;

	if (ttd < 0)
		ttd = 0;

	src = &req->af.v6_req.loc_addr;
	dest = &req->af.v6_req.rmt_addr;
	sprintf(tmpbuf,
		"%4d: %08X%08X%08X%08X:%04X %08X%08X%08X%08X:%04X "
		"%02X %08X:%08X %02X:%08X %08X %5d %8d %d %d %p",
		i,
		src->s6_addr32[0], src->s6_addr32[1],
		src->s6_addr32[2], src->s6_addr32[3],
		ntohs(sk->sport),
		dest->s6_addr32[0], dest->s6_addr32[1],
		dest->s6_addr32[2], dest->s6_addr32[3],
		ntohs(req->rmt_port),
		TCP_SYN_RECV,
		0,0, /* could print option size, but that is af dependent. */
		1,   /* timers active (only the expire timer) */  
		ttd, 
		req->retrans,
		uid,
		0,  /* non standard timer */  
		0, /* open_requests have no inode */
		0, req);
}

static void get_tcp6_sock(struct sock *sp, char *tmpbuf, int i)
{
	struct in6_addr *dest, *src;
	__u16 destp, srcp;
	int timer_active;
	unsigned long timer_expires;
	struct tcp_opt *tp = &sp->tp_pinfo.af_tcp;

	dest  = &sp->net_pinfo.af_inet6.daddr;
	src   = &sp->net_pinfo.af_inet6.rcv_saddr;
	destp = ntohs(sp->dport);
	srcp  = ntohs(sp->sport);
	if (tp->pending == TCP_TIME_RETRANS) {
		timer_active	= 1;
		timer_expires	= tp->timeout;
	} else if (tp->pending == TCP_TIME_PROBE0) {
		timer_active	= 4;
		timer_expires	= tp->timeout;
	} else if (timer_pending(&sp->timer)) {
		timer_active	= 2;
		timer_expires	= sp->timer.expires;
	} else {
		timer_active	= 0;
		timer_expires = jiffies;
	}

	sprintf(tmpbuf,
		"%4d: %08X%08X%08X%08X:%04X %08X%08X%08X%08X:%04X "
		"%02X %08X:%08X %02X:%08lX %08X %5d %8d %lu %d %p %u %u %u %u %d",
		i,
		src->s6_addr32[0], src->s6_addr32[1],
		src->s6_addr32[2], src->s6_addr32[3], srcp,
		dest->s6_addr32[0], dest->s6_addr32[1],
		dest->s6_addr32[2], dest->s6_addr32[3], destp,
		sp->state, 
		tp->write_seq-tp->snd_una, tp->rcv_nxt-tp->copied_seq,
		timer_active, timer_expires-jiffies,
		tp->retransmits,
		sock_i_uid(sp),
		tp->probes_out,
		sock_i_ino(sp),
		atomic_read(&sp->refcnt), sp,
		tp->rto, tp->ack.ato, (tp->ack.quick<<1)|tp->ack.pingpong,
		tp->snd_cwnd, tp->snd_ssthresh>=0xFFFF?-1:tp->snd_ssthresh
		);
}

static void get_timewait6_sock(struct tcp_tw_bucket *tw, char *tmpbuf, int i)
{
	struct in6_addr *dest, *src;
	__u16 destp, srcp;
	int ttd = tw->ttd - jiffies;

	if (ttd < 0)
		ttd = 0;

	dest  = &tw->v6_daddr;
	src   = &tw->v6_rcv_saddr;
	destp = ntohs(tw->dport);
	srcp  = ntohs(tw->sport);

	sprintf(tmpbuf,
		"%4d: %08X%08X%08X%08X:%04X %08X%08X%08X%08X:%04X "
		"%02X %08X:%08X %02X:%08X %08X %5d %8d %d %d %p",
		i,
		src->s6_addr32[0], src->s6_addr32[1],
		src->s6_addr32[2], src->s6_addr32[3], srcp,
		dest->s6_addr32[0], dest->s6_addr32[1],
		dest->s6_addr32[2], dest->s6_addr32[3], destp,
		tw->substate, 0, 0,
		3, ttd, 0, 0, 0, 0,
		atomic_read(&tw->refcnt), tw);
}

#define LINE_LEN 190
#define LINE_FMT "%-190s\n"

int tcp6_get_info(char *buffer, char **start, off_t offset, int length)
{
	int len = 0, num = 0, i;
	off_t begin, pos = 0;
	char tmpbuf[LINE_LEN+2];

	if (offset < LINE_LEN+1)
		len += sprintf(buffer, LINE_FMT,
			       "  sl  "						/* 6 */
			       "local_address                         "		/* 38 */
			       "remote_address                        "		/* 38 */
			       "st tx_queue rx_queue tr tm->when retrnsmt"	/* 41 */
			       "   uid  timeout inode");			/* 21 */
										/*----*/
										/*144 */

	pos = LINE_LEN+1;

	/* First, walk listening socket table. */
	tcp_listen_lock();
	for(i = 0; i < TCP_LHTABLE_SIZE; i++) {
		struct sock *sk = tcp_listening_hash[i];
		struct tcp_listen_opt *lopt;
		int k;

		for (sk = tcp_listening_hash[i]; sk; sk = sk->next, num++) {
			struct open_request *req;
			int uid;
			struct tcp_opt *tp = &(sk->tp_pinfo.af_tcp);

			if (sk->family != PF_INET6)
				continue;
			pos += LINE_LEN+1;
			if (pos >= offset) {
				get_tcp6_sock(sk, tmpbuf, num);
				len += sprintf(buffer+len, LINE_FMT, tmpbuf);
				if (pos >= offset + length) {
					tcp_listen_unlock();
					goto out_no_bh;
				}
			}

			uid = sock_i_uid(sk);
			read_lock_bh(&tp->syn_wait_lock);
			lopt = tp->listen_opt;
			if (lopt && lopt->qlen != 0) {
				for (k=0; k<TCP_SYNQ_HSIZE; k++) {
					for (req = lopt->syn_table[k]; req; req = req->dl_next, num++) {
						if (req->class->family != PF_INET6)
							continue;
						pos += LINE_LEN+1;
						if (pos <= offset)
							continue;
						get_openreq6(sk, req, tmpbuf, num, uid);
						len += sprintf(buffer+len, LINE_FMT, tmpbuf);
						if (pos >= offset + length) { 
							read_unlock_bh(&tp->syn_wait_lock);
							tcp_listen_unlock();
							goto out_no_bh;
						}
					}
				}
			}
			read_unlock_bh(&tp->syn_wait_lock);

			/* Completed requests are in normal socket hash table */
		}
	}
	tcp_listen_unlock();

	local_bh_disable();

	/* Next, walk established hash chain. */
	for (i = 0; i < tcp_ehash_size; i++) {
		struct tcp_ehash_bucket *head = &tcp_ehash[i];
		struct sock *sk;
		struct tcp_tw_bucket *tw;

		read_lock(&head->lock);
		for(sk = head->chain; sk; sk = sk->next, num++) {
			if (sk->family != PF_INET6)
				continue;
			pos += LINE_LEN+1;
			if (pos <= offset)
				continue;
			get_tcp6_sock(sk, tmpbuf, num);
			len += sprintf(buffer+len, LINE_FMT, tmpbuf);
			if (pos >= offset + length) {
				read_unlock(&head->lock);
				goto out;
			}
		}
		for (tw = (struct tcp_tw_bucket *)tcp_ehash[i+tcp_ehash_size].chain;
		     tw != NULL;
		     tw = (struct tcp_tw_bucket *)tw->next, num++) {
			if (tw->family != PF_INET6)
				continue;
			pos += LINE_LEN+1;
			if (pos <= offset)
				continue;
			get_timewait6_sock(tw, tmpbuf, num);
			len += sprintf(buffer+len, LINE_FMT, tmpbuf);
			if (pos >= offset + length) {
				read_unlock(&head->lock);
				goto out;
			}
		}
		read_unlock(&head->lock);
	}

out:
	local_bh_enable();
out_no_bh:

	begin = len - (pos - offset);
	*start = buffer + begin;
	len -= begin;
	if (len > length)
		len = length;
	if (len < 0)
		len = 0; 
	return len;
}

struct proto tcpv6_prot = {
	name:		"TCPv6",
	close:		tcp_close,
	connect:	tcp_v6_connect,
	disconnect:	tcp_disconnect,
	accept:		tcp_accept,
	ioctl:		tcp_ioctl,
	init:		tcp_v6_init_sock,
	destroy:	tcp_v6_destroy_sock,
	shutdown:	tcp_shutdown,
	setsockopt:	tcp_setsockopt,
	getsockopt:	tcp_getsockopt,
	sendmsg:	tcp_sendmsg,
	recvmsg:	tcp_recvmsg,
	backlog_rcv:	tcp_v6_do_rcv,
	hash:		tcp_v6_hash,
	unhash:		tcp_unhash,
	get_port:	tcp_v6_get_port,
};

static struct inet6_protocol tcpv6_protocol =
{
	tcp_v6_rcv,		/* TCP handler		*/
	tcp_v6_err,		/* TCP error control	*/
	NULL,			/* next			*/
	IPPROTO_TCP,		/* protocol ID		*/
	0,			/* copy			*/
	NULL,			/* data			*/
	"TCPv6"			/* name			*/
};

extern struct proto_ops inet6_stream_ops;

static struct inet_protosw tcpv6_protosw = {
	type:        SOCK_STREAM,
	protocol:    IPPROTO_TCP,
	prot:        &tcpv6_prot,
	ops:         &inet6_stream_ops,
	capability:  -1,
	no_check:    0,
	flags:       INET_PROTOSW_PERMANENT,
};

void __init tcpv6_init(void)
{
	/* register inet6 protocol */
	inet6_add_protocol(&tcpv6_protocol);
	inet6_register_protosw(&tcpv6_protosw);
}