#include <net/addrconf.h>
#include <net/snmp.h>
#include <net/dsfield.h>
+#include <net/timewait_sock.h>
#include <asm/uaccess.h>
#include <linux/proc_fs.h>
#include <linux/seq_file.h>
+/* Socket used for sending RSTs and ACKs */
+static struct socket *tcp6_socket;
+
static void tcp_v6_send_reset(struct sk_buff *skb);
static void tcp_v6_reqsk_send_ack(struct sk_buff *skb, struct request_sock *req);
static void tcp_v6_send_check(struct sock *sk, int len,
}
}
-static int __tcp_v6_check_established(struct sock *sk, const __u16 lport,
- struct inet_timewait_sock **twp)
-{
- struct inet_sock *inet = inet_sk(sk);
- const struct ipv6_pinfo *np = inet6_sk(sk);
- const struct in6_addr *daddr = &np->rcv_saddr;
- const struct in6_addr *saddr = &np->daddr;
- const int dif = sk->sk_bound_dev_if;
- const u32 ports = INET_COMBINED_PORTS(inet->dport, lport);
- unsigned int hash = inet6_ehashfn(daddr, inet->num, saddr, inet->dport);
- struct inet_ehash_bucket *head = inet_ehash_bucket(&tcp_hashinfo, hash);
- struct sock *sk2;
- const struct hlist_node *node;
- struct inet_timewait_sock *tw;
-
- prefetch(head->chain.first);
- write_lock(&head->lock);
-
- /* Check TIME-WAIT sockets first. */
- sk_for_each(sk2, node, &(head + tcp_hashinfo.ehash_size)->chain) {
- const struct tcp6_timewait_sock *tcp6tw = tcp6_twsk(sk2);
-
- tw = inet_twsk(sk2);
-
- if(*((__u32 *)&(tw->tw_dport)) == ports &&
- sk2->sk_family == PF_INET6 &&
- ipv6_addr_equal(&tcp6tw->tw_v6_daddr, saddr) &&
- ipv6_addr_equal(&tcp6tw->tw_v6_rcv_saddr, daddr) &&
- sk2->sk_bound_dev_if == sk->sk_bound_dev_if) {
- const struct tcp_timewait_sock *tcptw = tcp_twsk(sk2);
- struct tcp_sock *tp = tcp_sk(sk);
-
- if (tcptw->tw_ts_recent_stamp &&
- (!twp ||
- (sysctl_tcp_tw_reuse &&
- xtime.tv_sec - tcptw->tw_ts_recent_stamp > 1))) {
- /* See comment in tcp_ipv4.c */
- tp->write_seq = tcptw->tw_snd_nxt + 65535 + 2;
- if (!tp->write_seq)
- tp->write_seq = 1;
- tp->rx_opt.ts_recent = tcptw->tw_ts_recent;
- tp->rx_opt.ts_recent_stamp = tcptw->tw_ts_recent_stamp;
- sock_hold(sk2);
- goto unique;
- } else
- goto not_unique;
- }
- }
- tw = NULL;
-
- /* And established part... */
- sk_for_each(sk2, node, &head->chain) {
- if (INET6_MATCH(sk2, hash, saddr, daddr, ports, dif))
- goto not_unique;
- }
-
-unique:
- BUG_TRAP(sk_unhashed(sk));
- __sk_add_node(sk, &head->chain);
- sk->sk_hash = hash;
- sock_prot_inc_use(sk->sk_prot);
- write_unlock(&head->lock);
-
- if (twp) {
- *twp = tw;
- NET_INC_STATS_BH(LINUX_MIB_TIMEWAITRECYCLED);
- } else if (tw) {
- /* Silly. Should hash-dance instead... */
- inet_twsk_deschedule(tw, &tcp_death_row);
- NET_INC_STATS_BH(LINUX_MIB_TIMEWAITRECYCLED);
-
- inet_twsk_put(tw);
- }
- return 0;
-
-not_unique:
- write_unlock(&head->lock);
- return -EADDRNOTAVAIL;
-}
-
-static inline u32 tcpv6_port_offset(const struct sock *sk)
-{
- const struct inet_sock *inet = inet_sk(sk);
- const struct ipv6_pinfo *np = inet6_sk(sk);
-
- return secure_tcpv6_port_ephemeral(np->rcv_saddr.s6_addr32,
- np->daddr.s6_addr32,
- inet->dport);
-}
-
-static int tcp_v6_hash_connect(struct sock *sk)
-{
- unsigned short snum = inet_sk(sk)->num;
- struct inet_bind_hashbucket *head;
- struct inet_bind_bucket *tb;
- int ret;
-
- if (!snum) {
- int low = sysctl_local_port_range[0];
- int high = sysctl_local_port_range[1];
- int range = high - low;
- int i;
- int port;
- static u32 hint;
- u32 offset = hint + tcpv6_port_offset(sk);
- struct hlist_node *node;
- struct inet_timewait_sock *tw = NULL;
-
- local_bh_disable();
- for (i = 1; i <= range; i++) {
- port = low + (i + offset) % range;
- head = &tcp_hashinfo.bhash[inet_bhashfn(port, tcp_hashinfo.bhash_size)];
- spin_lock(&head->lock);
-
- /* Does not bother with rcv_saddr checks,
- * because the established check is already
- * unique enough.
- */
- inet_bind_bucket_for_each(tb, node, &head->chain) {
- if (tb->port == port) {
- BUG_TRAP(!hlist_empty(&tb->owners));
- if (tb->fastreuse >= 0)
- goto next_port;
- if (!__tcp_v6_check_established(sk,
- port,
- &tw))
- goto ok;
- goto next_port;
- }
- }
-
- tb = inet_bind_bucket_create(tcp_hashinfo.bind_bucket_cachep, head, port);
- if (!tb) {
- spin_unlock(&head->lock);
- break;
- }
- tb->fastreuse = -1;
- goto ok;
-
- next_port:
- spin_unlock(&head->lock);
- }
- local_bh_enable();
-
- return -EADDRNOTAVAIL;
-
-ok:
- hint += i;
-
- /* Head lock still held and bh's disabled */
- inet_bind_hash(sk, tb, port);
- if (sk_unhashed(sk)) {
- inet_sk(sk)->sport = htons(port);
- __inet6_hash(&tcp_hashinfo, sk);
- }
- spin_unlock(&head->lock);
-
- if (tw) {
- inet_twsk_deschedule(tw, &tcp_death_row);
- inet_twsk_put(tw);
- }
-
- ret = 0;
- goto out;
- }
-
- head = &tcp_hashinfo.bhash[inet_bhashfn(snum, tcp_hashinfo.bhash_size)];
- tb = inet_csk(sk)->icsk_bind_hash;
- spin_lock_bh(&head->lock);
-
- if (sk_head(&tb->owners) == sk && !sk->sk_bind_node.next) {
- __inet6_hash(&tcp_hashinfo, sk);
- spin_unlock_bh(&head->lock);
- return 0;
- } else {
- spin_unlock(&head->lock);
- /* No definite answer... Walk to established hash table */
- ret = __tcp_v6_check_established(sk, snum, NULL);
-out:
- local_bh_enable();
- return ret;
- }
-}
-
static int tcp_v6_connect(struct sock *sk, struct sockaddr *uaddr,
int addr_len)
{
struct sockaddr_in6 *usin = (struct sockaddr_in6 *) uaddr;
- struct inet_sock *inet = inet_sk(sk);
+ struct inet_sock *inet = inet_sk(sk);
+ struct inet_connection_sock *icsk = inet_csk(sk);
struct ipv6_pinfo *np = inet6_sk(sk);
struct tcp_sock *tp = tcp_sk(sk);
struct in6_addr *saddr = NULL, *final_p = NULL, final;
*/
if (addr_type == IPV6_ADDR_MAPPED) {
- u32 exthdrlen = tp->ext_header_len;
+ u32 exthdrlen = icsk->icsk_ext_hdr_len;
struct sockaddr_in sin;
SOCK_DEBUG(sk, "connect: ipv4 mapped\n");
sin.sin_port = usin->sin6_port;
sin.sin_addr.s_addr = usin->sin6_addr.s6_addr32[3];
- inet_csk(sk)->icsk_af_ops = &ipv6_mapped;
+ icsk->icsk_af_ops = &ipv6_mapped;
sk->sk_backlog_rcv = tcp_v4_do_rcv;
err = tcp_v4_connect(sk, (struct sockaddr *)&sin, sizeof(sin));
if (err) {
- tp->ext_header_len = exthdrlen;
- inet_csk(sk)->icsk_af_ops = &ipv6_specific;
+ icsk->icsk_ext_hdr_len = exthdrlen;
+ icsk->icsk_af_ops = &ipv6_specific;
sk->sk_backlog_rcv = tcp_v6_do_rcv;
goto failure;
} else {
sk->sk_route_caps = dst->dev->features &
~(NETIF_F_IP_CSUM | NETIF_F_TSO);
- tp->ext_header_len = 0;
+ icsk->icsk_ext_hdr_len = 0;
if (np->opt)
- tp->ext_header_len = np->opt->opt_flen + np->opt->opt_nflen;
+ icsk->icsk_ext_hdr_len = (np->opt->opt_flen +
+ np->opt->opt_nflen);
tp->rx_opt.mss_clamp = IPV6_MIN_MTU - sizeof(struct tcphdr) - sizeof(struct ipv6hdr);
inet->dport = usin->sin6_port;
tcp_set_state(sk, TCP_SYN_SENT);
- err = tcp_v6_hash_connect(sk);
+ err = inet6_hash_connect(&tcp_death_row, sk);
if (err)
goto late_failure;
} else
dst_hold(dst);
- if (tp->pmtu_cookie > dst_mtu(dst)) {
+ if (inet_csk(sk)->icsk_pmtu_cookie > dst_mtu(dst)) {
tcp_sync_mss(sk, dst_mtu(dst));
tcp_simple_retransmit(sk);
} /* else let the usual retransmit timer handle it */
done:
if (opt && opt != np->opt)
sock_kfree_s(sk, opt, opt->tot_len);
+ dst_release(dst);
return err;
}
.send_reset = tcp_v6_send_reset
};
-static int ipv6_opt_accepted(struct sock *sk, struct sk_buff *skb)
-{
- struct ipv6_pinfo *np = inet6_sk(sk);
- struct inet6_skb_parm *opt = IP6CB(skb);
-
- if (np->rxopt.all) {
- if ((opt->hop && (np->rxopt.bits.hopopts || np->rxopt.bits.ohopopts)) ||
- ((IPV6_FLOWINFO_MASK & *(u32*)skb->nh.raw) && np->rxopt.bits.rxflow) ||
- (opt->srcrt && (np->rxopt.bits.srcrt || np->rxopt.bits.osrcrt)) ||
- ((opt->dst1 || opt->dst0) && (np->rxopt.bits.dstopts || np->rxopt.bits.odstopts)))
- return 1;
- }
- return 0;
-}
-
+static struct timewait_sock_ops tcp6_timewait_sock_ops = {
+ .twsk_obj_size = sizeof(struct tcp6_timewait_sock),
+ .twsk_unique = tcp_twsk_unique,
+};
static void tcp_v6_send_check(struct sock *sk, int len, struct sk_buff *skb)
{
if (!ip6_dst_lookup(NULL, &buff->dst, &fl)) {
if (xfrm_lookup(&buff->dst, &fl, NULL, 0) >= 0) {
- ip6_xmit(NULL, buff, &fl, NULL, 0);
+ ip6_xmit(tcp6_socket->sk, buff, &fl, NULL, 0);
TCP_INC_STATS_BH(TCP_MIB_OUTSEGS);
TCP_INC_STATS_BH(TCP_MIB_OUTRSTS);
return;
if (!ip6_dst_lookup(NULL, &buff->dst, &fl)) {
if (xfrm_lookup(&buff->dst, &fl, NULL, 0) >= 0) {
- ip6_xmit(NULL, buff, &fl, NULL, 0);
+ ip6_xmit(tcp6_socket->sk, buff, &fl, NULL, 0);
TCP_INC_STATS_BH(TCP_MIB_OUTSEGS);
return;
}
worked with IPv6 icsk.icsk_af_ops.
Sync it now.
*/
- tcp_sync_mss(newsk, newtp->pmtu_cookie);
+ tcp_sync_mss(newsk, inet_csk(newsk)->icsk_pmtu_cookie);
return newsk;
}
sock_kfree_s(sk, opt, opt->tot_len);
}
- newtp->ext_header_len = 0;
+ inet_csk(newsk)->icsk_ext_hdr_len = 0;
if (newnp->opt)
- newtp->ext_header_len = newnp->opt->opt_nflen +
- newnp->opt->opt_flen;
+ inet_csk(newsk)->icsk_ext_hdr_len = (newnp->opt->opt_nflen +
+ newnp->opt->opt_flen);
tcp_sync_mss(newsk, dst_mtu(dst));
newtp->advmss = dst_metric(dst, RTAX_ADVMSS);
return 0;
}
-static int tcp_v6_rcv(struct sk_buff **pskb, unsigned int *nhoffp)
+static int tcp_v6_rcv(struct sk_buff **pskb)
{
struct sk_buff *skb = *pskb;
struct tcphdr *th;
icsk->icsk_af_ops = &ipv6_specific;
icsk->icsk_ca_ops = &tcp_init_congestion_ops;
+ icsk->icsk_sync_mss = tcp_sync_mss;
sk->sk_write_space = sk_stream_write_space;
sock_set_flag(sk, SOCK_USE_WRITE_QUEUE);
{
struct in6_addr *dest, *src;
__u16 destp, srcp;
- struct tcp6_timewait_sock *tcp6tw = tcp6_twsk((struct sock *)tw);
+ struct inet6_timewait_sock *tw6 = inet6_twsk((struct sock *)tw);
int ttd = tw->tw_ttd - jiffies;
if (ttd < 0)
ttd = 0;
- dest = &tcp6tw->tw_v6_daddr;
- src = &tcp6tw->tw_v6_rcv_saddr;
+ dest = &tw6->tw_v6_daddr;
+ src = &tw6->tw_v6_rcv_saddr;
destp = ntohs(tw->tw_dport);
srcp = ntohs(tw->tw_sport);
.sysctl_rmem = sysctl_tcp_rmem,
.max_header = MAX_TCP_HEADER,
.obj_size = sizeof(struct tcp6_sock),
- .twsk_obj_size = sizeof(struct tcp6_timewait_sock),
+ .twsk_prot = &tcp6_timewait_sock_ops,
.rsk_prot = &tcp6_request_sock_ops,
};
.ops = &inet6_stream_ops,
.capability = -1,
.no_check = 0,
- .flags = INET_PROTOSW_PERMANENT,
+ .flags = INET_PROTOSW_PERMANENT |
+ INET_PROTOSW_ICSK,
};
void __init tcpv6_init(void)
{
+ int err;
+
/* register inet6 protocol */
if (inet6_add_protocol(&tcpv6_protocol, IPPROTO_TCP) < 0)
printk(KERN_ERR "tcpv6_init: Could not register protocol\n");
inet6_register_protosw(&tcpv6_protosw);
+
+ err = sock_create_kern(PF_INET6, SOCK_RAW, IPPROTO_TCP, &tcp6_socket);
+ if (err < 0)
+ panic("Failed to create the TCPv6 control socket.\n");
+ tcp6_socket->sk->sk_allocation = GFP_ATOMIC;
+
+ /* Unhash it so that IP input processing does not even
+ * see it, we do not wish this socket to see incoming
+ * packets.
+ */
+ tcp6_socket->sk->sk_prot->unhash(tcp6_socket->sk);
}