int sysctl_tcp_fin_timeout = TCP_FIN_TIMEOUT;
-DEFINE_SNMP_STAT(struct tcp_mib, tcp_statistics);
+DEFINE_SNMP_STAT(struct tcp_mib, tcp_statistics) __read_mostly;
atomic_t tcp_orphan_count = ATOMIC_INIT(0);
+EXPORT_SYMBOL_GPL(tcp_orphan_count);
+
int sysctl_tcp_mem[3];
int sysctl_tcp_wmem[3] = { 4 * 1024, 16 * 1024, 128 * 1024 };
int sysctl_tcp_rmem[3] = { 4 * 1024, 87380, 87380 * 2 };
EXPORT_SYMBOL(tcp_enter_memory_pressure);
-/*
- * LISTEN is a special case for poll..
- */
-static __inline__ unsigned int tcp_listen_poll(struct sock *sk,
- poll_table *wait)
-{
- return !reqsk_queue_empty(&inet_csk(sk)->icsk_accept_queue) ? (POLLIN | POLLRDNORM) : 0;
-}
-
/*
* Wait for a TCP event.
*
poll_wait(file, sk->sk_sleep, wait);
if (sk->sk_state == TCP_LISTEN)
- return tcp_listen_poll(sk, wait);
+ return inet_csk_listen_poll(sk);
/* Socket is not locked. We are protected from async events
by poll logic and correct handling of state changes
return put_user(answ, (int __user *)arg);
}
-
-int tcp_listen_start(struct sock *sk)
-{
- struct inet_sock *inet = inet_sk(sk);
- struct inet_connection_sock *icsk = inet_csk(sk);
- int rc = reqsk_queue_alloc(&icsk->icsk_accept_queue, TCP_SYNQ_HSIZE);
-
- if (rc != 0)
- return rc;
-
- sk->sk_max_ack_backlog = 0;
- sk->sk_ack_backlog = 0;
- inet_csk_delack_init(sk);
-
- /* There is race window here: we announce ourselves listening,
- * but this transition is still not validated by get_port().
- * It is OK, because this socket enters to hash table only
- * after validation is complete.
- */
- sk->sk_state = TCP_LISTEN;
- if (!sk->sk_prot->get_port(sk, inet->num)) {
- inet->sport = htons(inet->num);
-
- sk_dst_reset(sk);
- sk->sk_prot->hash(sk);
-
- return 0;
- }
-
- sk->sk_state = TCP_CLOSE;
- __reqsk_queue_destroy(&icsk->icsk_accept_queue);
- return -EADDRINUSE;
-}
-
-/*
- * This routine closes sockets which have been at least partially
- * opened, but not yet accepted.
- */
-
-static void tcp_listen_stop (struct sock *sk)
-{
- struct inet_connection_sock *icsk = inet_csk(sk);
- struct request_sock *acc_req;
- struct request_sock *req;
-
- inet_csk_delete_keepalive_timer(sk);
-
- /* make all the listen_opt local to us */
- acc_req = reqsk_queue_yank_acceptq(&icsk->icsk_accept_queue);
-
- /* Following specs, it would be better either to send FIN
- * (and enter FIN-WAIT-1, it is normal close)
- * or to send active reset (abort).
- * Certainly, it is pretty dangerous while synflood, but it is
- * bad justification for our negligence 8)
- * To be honest, we are not able to make either
- * of the variants now. --ANK
- */
- reqsk_queue_destroy(&icsk->icsk_accept_queue);
-
- while ((req = acc_req) != NULL) {
- struct sock *child = req->sk;
-
- acc_req = req->dl_next;
-
- local_bh_disable();
- bh_lock_sock(child);
- BUG_TRAP(!sock_owned_by_user(child));
- sock_hold(child);
-
- tcp_disconnect(child, O_NONBLOCK);
-
- sock_orphan(child);
-
- atomic_inc(&tcp_orphan_count);
-
- tcp_destroy_sock(child);
-
- bh_unlock_sock(child);
- local_bh_enable();
- sock_put(child);
-
- sk_acceptq_removed(sk);
- __reqsk_free(req);
- }
- BUG_TRAP(!sk->sk_ack_backlog);
-}
-
static inline void tcp_mark_push(struct tcp_sock *tp, struct sk_buff *skb)
{
TCP_SKB_CB(skb)->flags |= TCPCB_FLAG_PSH;
tcp_mark_push(tp, skb);
goto new_segment;
}
- if (sk->sk_forward_alloc < copy &&
- !sk_stream_mem_schedule(sk, copy, 0))
+ if (!sk_stream_wmem_schedule(sk, copy))
goto wait_for_memory;
if (can_coalesce) {
if (off == PAGE_SIZE) {
put_page(page);
TCP_PAGE(sk) = page = NULL;
+ off = 0;
}
- }
+ } else
+ off = 0;
+
+ if (copy > PAGE_SIZE - off)
+ copy = PAGE_SIZE - off;
+
+ if (!sk_stream_wmem_schedule(sk, copy))
+ goto wait_for_memory;
if (!page) {
/* Allocate new cache page. */
if (!(page = sk_stream_alloc_page(sk)))
goto wait_for_memory;
- off = 0;
}
- if (copy > PAGE_SIZE - off)
- copy = PAGE_SIZE - off;
-
/* Time to copy data. We are close to
* the end! */
err = skb_copy_to_page(sk, from, skb, page,
}
}
-/*
- * At this point, there should be no process reference to this
- * socket, and thus no user references at all. Therefore we
- * can assume the socket waitqueue is inactive and nobody will
- * try to jump onto it.
- */
-void tcp_destroy_sock(struct sock *sk)
-{
- BUG_TRAP(sk->sk_state == TCP_CLOSE);
- BUG_TRAP(sock_flag(sk, SOCK_DEAD));
-
- /* It cannot be in hash table! */
- BUG_TRAP(sk_unhashed(sk));
-
- /* If it has not 0 inet_sk(sk)->num, it must be bound */
- BUG_TRAP(!inet_sk(sk)->num || inet_csk(sk)->icsk_bind_hash);
-
- sk->sk_prot->destroy(sk);
-
- sk_stream_kill_queues(sk);
-
- xfrm_sk_free_policy(sk);
-
- sk_refcnt_debug_release(sk);
-
- atomic_dec(&tcp_orphan_count);
- sock_put(sk);
-}
-
void tcp_close(struct sock *sk, long timeout)
{
struct sk_buff *skb;
tcp_set_state(sk, TCP_CLOSE);
/* Special case. */
- tcp_listen_stop(sk);
+ inet_csk_listen_stop(sk);
goto adjudge_to_death;
}
if (tmo > TCP_TIMEWAIT_LEN) {
inet_csk_reset_keepalive_timer(sk, tcp_fin_time(sk));
} else {
- atomic_inc(&tcp_orphan_count);
+ atomic_inc(sk->sk_prot->orphan_count);
tcp_time_wait(sk, TCP_FIN_WAIT2, tmo);
goto out;
}
}
if (sk->sk_state != TCP_CLOSE) {
sk_stream_mem_reclaim(sk);
- if (atomic_read(&tcp_orphan_count) > sysctl_tcp_max_orphans ||
+ if (atomic_read(sk->sk_prot->orphan_count) > sysctl_tcp_max_orphans ||
(sk->sk_wmem_queued > SOCK_MIN_SNDBUF &&
atomic_read(&tcp_memory_allocated) > sysctl_tcp_mem[2])) {
if (net_ratelimit())
NET_INC_STATS_BH(LINUX_MIB_TCPABORTONMEMORY);
}
}
- atomic_inc(&tcp_orphan_count);
+ atomic_inc(sk->sk_prot->orphan_count);
if (sk->sk_state == TCP_CLOSE)
- tcp_destroy_sock(sk);
+ inet_csk_destroy_sock(sk);
/* Otherwise, socket is reprieved until protocol close. */
out:
/* ABORT function of RFC793 */
if (old_state == TCP_LISTEN) {
- tcp_listen_stop(sk);
+ inet_csk_listen_stop(sk);
} else if (tcp_need_reset(old_state) ||
(tp->snd_nxt != tp->write_seq &&
(1 << old_state) & (TCPF_CLOSING | TCPF_LAST_ACK))) {
tp->write_seq = 1;
icsk->icsk_backoff = 0;
tp->snd_cwnd = 2;
- tp->probes_out = 0;
+ icsk->icsk_probes_out = 0;
tp->packets_out = 0;
tp->snd_ssthresh = 0x7fffffff;
tp->snd_cwnd_cnt = 0;
- tcp_set_ca_state(tp, TCP_CA_Open);
+ tcp_set_ca_state(sk, TCP_CA_Open);
tcp_clear_retrans(tp);
inet_csk_delack_init(sk);
sk->sk_send_head = NULL;
name[val] = 0;
lock_sock(sk);
- err = tcp_set_congestion_control(tp, name);
+ err = tcp_set_congestion_control(sk, name);
release_sock(sk);
return err;
}
break;
case TCP_DEFER_ACCEPT:
- tp->defer_accept = 0;
+ icsk->icsk_accept_queue.rskq_defer_accept = 0;
if (val > 0) {
/* Translate value in seconds to number of
* retransmits */
- while (tp->defer_accept < 32 &&
+ while (icsk->icsk_accept_queue.rskq_defer_accept < 32 &&
val > ((TCP_TIMEOUT_INIT / HZ) <<
- tp->defer_accept))
- tp->defer_accept++;
- tp->defer_accept++;
+ icsk->icsk_accept_queue.rskq_defer_accept))
+ icsk->icsk_accept_queue.rskq_defer_accept++;
+ icsk->icsk_accept_queue.rskq_defer_accept++;
}
break;
memset(info, 0, sizeof(*info));
info->tcpi_state = sk->sk_state;
- info->tcpi_ca_state = tp->ca_state;
+ info->tcpi_ca_state = icsk->icsk_ca_state;
info->tcpi_retransmits = icsk->icsk_retransmits;
- info->tcpi_probes = tp->probes_out;
+ info->tcpi_probes = icsk->icsk_probes_out;
info->tcpi_backoff = icsk->icsk_backoff;
if (tp->rx_opt.tstamp_ok)
int tcp_getsockopt(struct sock *sk, int level, int optname, char __user *optval,
int __user *optlen)
{
+ struct inet_connection_sock *icsk = inet_csk(sk);
struct tcp_sock *tp = tcp_sk(sk);
int val, len;
val = tp->keepalive_probes ? : sysctl_tcp_keepalive_probes;
break;
case TCP_SYNCNT:
- val = inet_csk(sk)->icsk_syn_retries ? : sysctl_tcp_syn_retries;
+ val = icsk->icsk_syn_retries ? : sysctl_tcp_syn_retries;
break;
case TCP_LINGER2:
val = tp->linger2;
val = (val ? : sysctl_tcp_fin_timeout) / HZ;
break;
case TCP_DEFER_ACCEPT:
- val = !tp->defer_accept ? 0 : ((TCP_TIMEOUT_INIT / HZ) <<
- (tp->defer_accept - 1));
+ val = !icsk->icsk_accept_queue.rskq_defer_accept ? 0 :
+ ((TCP_TIMEOUT_INIT / HZ) << (icsk->icsk_accept_queue.rskq_defer_accept - 1));
break;
case TCP_WINDOW_CLAMP:
val = tp->window_clamp;
return 0;
}
case TCP_QUICKACK:
- val = !inet_csk(sk)->icsk_ack.pingpong;
+ val = !icsk->icsk_ack.pingpong;
break;
case TCP_CONGESTION:
len = min_t(unsigned int, len, TCP_CA_NAME_MAX);
if (put_user(len, optlen))
return -EFAULT;
- if (copy_to_user(optval, tp->ca_ops->name, len))
+ if (copy_to_user(optval, icsk->icsk_ca_ops->name, len))
return -EFAULT;
return 0;
default:
if (order >= 4) {
sysctl_local_port_range[0] = 32768;
sysctl_local_port_range[1] = 61000;
- sysctl_tcp_max_tw_buckets = 180000;
+ tcp_death_row.sysctl_max_tw_buckets = 180000;
sysctl_tcp_max_orphans = 4096 << (order - 4);
sysctl_max_syn_backlog = 1024;
} else if (order < 3) {
sysctl_local_port_range[0] = 1024 * (3 - order);
- sysctl_tcp_max_tw_buckets >>= (3 - order);
+ tcp_death_row.sysctl_max_tw_buckets >>= (3 - order);
sysctl_tcp_max_orphans >>= (3 - order);
sysctl_max_syn_backlog = 128;
}
}
EXPORT_SYMBOL(tcp_close);
-EXPORT_SYMBOL(tcp_destroy_sock);
EXPORT_SYMBOL(tcp_disconnect);
EXPORT_SYMBOL(tcp_getsockopt);
EXPORT_SYMBOL(tcp_ioctl);