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 };
static __inline__ unsigned int tcp_listen_poll(struct sock *sk,
poll_table *wait)
{
- return !reqsk_queue_empty(&tcp_sk(sk)->accept_queue) ? (POLLIN | POLLRDNORM) : 0;
+ return !reqsk_queue_empty(&inet_csk(sk)->icsk_accept_queue) ? (POLLIN | POLLRDNORM) : 0;
}
/*
return put_user(answ, (int __user *)arg);
}
-
-int tcp_listen_start(struct sock *sk)
+int inet_csk_listen_start(struct sock *sk, const int nr_table_entries)
{
struct inet_sock *inet = inet_sk(sk);
- struct tcp_sock *tp = tcp_sk(sk);
- int rc = reqsk_queue_alloc(&tp->accept_queue, TCP_SYNQ_HSIZE);
+ struct inet_connection_sock *icsk = inet_csk(sk);
+ int rc = reqsk_queue_alloc(&icsk->icsk_accept_queue, nr_table_entries);
if (rc != 0)
return rc;
sk->sk_max_ack_backlog = 0;
sk->sk_ack_backlog = 0;
- tcp_delack_init(tp);
+ inet_csk_delack_init(sk);
/* There is race window here: we announce ourselves listening,
* but this transition is still not validated by get_port().
}
sk->sk_state = TCP_CLOSE;
- __reqsk_queue_destroy(&tp->accept_queue);
+ __reqsk_queue_destroy(&icsk->icsk_accept_queue);
return -EADDRINUSE;
}
+EXPORT_SYMBOL_GPL(inet_csk_listen_start);
+
/*
* This routine closes sockets which have been at least partially
* opened, but not yet accepted.
*/
-
-static void tcp_listen_stop (struct sock *sk)
+void inet_csk_listen_stop(struct sock *sk)
{
- struct tcp_sock *tp = tcp_sk(sk);
+ struct inet_connection_sock *icsk = inet_csk(sk);
struct request_sock *acc_req;
struct request_sock *req;
- tcp_delete_keepalive_timer(sk);
+ inet_csk_delete_keepalive_timer(sk);
/* make all the listen_opt local to us */
- acc_req = reqsk_queue_yank_acceptq(&tp->accept_queue);
+ 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)
* To be honest, we are not able to make either
* of the variants now. --ANK
*/
- reqsk_queue_destroy(&tp->accept_queue);
+ reqsk_queue_destroy(&icsk->icsk_accept_queue);
while ((req = acc_req) != NULL) {
struct sock *child = req->sk;
BUG_TRAP(!sock_owned_by_user(child));
sock_hold(child);
- tcp_disconnect(child, O_NONBLOCK);
+ sk->sk_prot->disconnect(child, O_NONBLOCK);
sock_orphan(child);
- atomic_inc(&tcp_orphan_count);
+ atomic_inc(sk->sk_prot->orphan_count);
- tcp_destroy_sock(child);
+ inet_csk_destroy_sock(child);
bh_unlock_sock(child);
local_bh_enable();
BUG_TRAP(!sk->sk_ack_backlog);
}
+EXPORT_SYMBOL_GPL(inet_csk_listen_stop);
+
static inline void tcp_mark_push(struct tcp_sock *tp, struct sk_buff *skb)
{
TCP_SKB_CB(skb)->flags |= TCPCB_FLAG_PSH;
BUG_TRAP(!skb || before(tp->copied_seq, TCP_SKB_CB(skb)->end_seq));
#endif
- if (tcp_ack_scheduled(tp)) {
+ if (inet_csk_ack_scheduled(sk)) {
+ const struct inet_connection_sock *icsk = inet_csk(sk);
/* Delayed ACKs frequently hit locked sockets during bulk
* receive. */
- if (tp->ack.blocked ||
+ if (icsk->icsk_ack.blocked ||
/* Once-per-two-segments ACK was not sent by tcp_input.c */
- tp->rcv_nxt - tp->rcv_wup > tp->ack.rcv_mss ||
+ tp->rcv_nxt - tp->rcv_wup > icsk->icsk_ack.rcv_mss ||
/*
* If this read emptied read buffer, we send ACK, if
* connection is not bidirectional, user drained
* receive buffer and there was a small segment
* in queue.
*/
- (copied > 0 && (tp->ack.pending & TCP_ACK_PUSHED) &&
- !tp->ack.pingpong && !atomic_read(&sk->sk_rmem_alloc)))
+ (copied > 0 && (icsk->icsk_ack.pending & ICSK_ACK_PUSHED) &&
+ !icsk->icsk_ack.pingpong && !atomic_read(&sk->sk_rmem_alloc)))
time_to_ack = 1;
}
* can assume the socket waitqueue is inactive and nobody will
* try to jump onto it.
*/
-void tcp_destroy_sock(struct sock *sk)
+void inet_csk_destroy_sock(struct sock *sk)
{
BUG_TRAP(sk->sk_state == TCP_CLOSE);
BUG_TRAP(sock_flag(sk, SOCK_DEAD));
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_sk(sk)->bind_hash);
+ BUG_TRAP(!inet_sk(sk)->num || inet_csk(sk)->icsk_bind_hash);
sk->sk_prot->destroy(sk);
sk_refcnt_debug_release(sk);
- atomic_dec(&tcp_orphan_count);
+ atomic_dec(sk->sk_prot->orphan_count);
sock_put(sk);
}
tcp_set_state(sk, TCP_CLOSE);
/* Special case. */
- tcp_listen_stop(sk);
+ inet_csk_listen_stop(sk);
goto adjudge_to_death;
}
tcp_send_active_reset(sk, GFP_ATOMIC);
NET_INC_STATS_BH(LINUX_MIB_TCPABORTONLINGER);
} else {
- int tmo = tcp_fin_time(tp);
+ const int tmo = tcp_fin_time(sk);
if (tmo > TCP_TIMEWAIT_LEN) {
- tcp_reset_keepalive_timer(sk, tcp_fin_time(tp));
+ 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:
int tcp_disconnect(struct sock *sk, int flags)
{
struct inet_sock *inet = inet_sk(sk);
+ struct inet_connection_sock *icsk = inet_csk(sk);
struct tcp_sock *tp = tcp_sk(sk);
int err = 0;
int old_state = sk->sk_state;
/* 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->srtt = 0;
if ((tp->write_seq += tp->max_window + 2) == 0)
tp->write_seq = 1;
- tp->backoff = 0;
+ icsk->icsk_backoff = 0;
tp->snd_cwnd = 2;
tp->probes_out = 0;
tp->packets_out = 0;
tp->snd_cwnd_cnt = 0;
tcp_set_ca_state(tp, TCP_CA_Open);
tcp_clear_retrans(tp);
- tcp_delack_init(tp);
+ inet_csk_delack_init(sk);
sk->sk_send_head = NULL;
tp->rx_opt.saw_tstamp = 0;
tcp_sack_reset(&tp->rx_opt);
__sk_dst_reset(sk);
- BUG_TRAP(!inet->num || inet->bind_hash);
+ BUG_TRAP(!inet->num || icsk->icsk_bind_hash);
sk->sk_error_report(sk);
return err;
}
-/*
- * Wait for an incoming connection, avoid race
- * conditions. This must be called with the socket locked.
- */
-static int wait_for_connect(struct sock *sk, long timeo)
-{
- struct tcp_sock *tp = tcp_sk(sk);
- DEFINE_WAIT(wait);
- int err;
-
- /*
- * True wake-one mechanism for incoming connections: only
- * one process gets woken up, not the 'whole herd'.
- * Since we do not 'race & poll' for established sockets
- * anymore, the common case will execute the loop only once.
- *
- * Subtle issue: "add_wait_queue_exclusive()" will be added
- * after any current non-exclusive waiters, and we know that
- * it will always _stay_ after any new non-exclusive waiters
- * because all non-exclusive waiters are added at the
- * beginning of the wait-queue. As such, it's ok to "drop"
- * our exclusiveness temporarily when we get woken up without
- * having to remove and re-insert us on the wait queue.
- */
- for (;;) {
- prepare_to_wait_exclusive(sk->sk_sleep, &wait,
- TASK_INTERRUPTIBLE);
- release_sock(sk);
- if (reqsk_queue_empty(&tp->accept_queue))
- timeo = schedule_timeout(timeo);
- lock_sock(sk);
- err = 0;
- if (!reqsk_queue_empty(&tp->accept_queue))
- break;
- err = -EINVAL;
- if (sk->sk_state != TCP_LISTEN)
- break;
- err = sock_intr_errno(timeo);
- if (signal_pending(current))
- break;
- err = -EAGAIN;
- if (!timeo)
- break;
- }
- finish_wait(sk->sk_sleep, &wait);
- return err;
-}
-
-/*
- * This will accept the next outstanding connection.
- */
-
-struct sock *tcp_accept(struct sock *sk, int flags, int *err)
-{
- struct tcp_sock *tp = tcp_sk(sk);
- struct sock *newsk;
- int error;
-
- lock_sock(sk);
-
- /* We need to make sure that this socket is listening,
- * and that it has something pending.
- */
- error = -EINVAL;
- if (sk->sk_state != TCP_LISTEN)
- goto out_err;
-
- /* Find already established connection */
- if (reqsk_queue_empty(&tp->accept_queue)) {
- long timeo = sock_rcvtimeo(sk, flags & O_NONBLOCK);
-
- /* If this is a non blocking socket don't sleep */
- error = -EAGAIN;
- if (!timeo)
- goto out_err;
-
- error = wait_for_connect(sk, timeo);
- if (error)
- goto out_err;
- }
-
- newsk = reqsk_queue_get_child(&tp->accept_queue, sk);
- BUG_TRAP(newsk->sk_state != TCP_SYN_RECV);
-out:
- release_sock(sk);
- return newsk;
-out_err:
- newsk = NULL;
- *err = error;
- goto out;
-}
-
/*
* Socket option code for TCP.
*/
int optlen)
{
struct tcp_sock *tp = tcp_sk(sk);
+ struct inet_connection_sock *icsk = inet_csk(sk);
int val;
int err = 0;
elapsed = tp->keepalive_time - elapsed;
else
elapsed = 0;
- tcp_reset_keepalive_timer(sk, elapsed);
+ inet_csk_reset_keepalive_timer(sk, elapsed);
}
}
break;
if (val < 1 || val > MAX_TCP_SYNCNT)
err = -EINVAL;
else
- tp->syn_retries = val;
+ icsk->icsk_syn_retries = val;
break;
case TCP_LINGER2:
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;
case TCP_QUICKACK:
if (!val) {
- tp->ack.pingpong = 1;
+ icsk->icsk_ack.pingpong = 1;
} else {
- tp->ack.pingpong = 0;
+ icsk->icsk_ack.pingpong = 0;
if ((1 << sk->sk_state) &
(TCPF_ESTABLISHED | TCPF_CLOSE_WAIT) &&
- tcp_ack_scheduled(tp)) {
- tp->ack.pending |= TCP_ACK_PUSHED;
+ inet_csk_ack_scheduled(sk)) {
+ icsk->icsk_ack.pending |= ICSK_ACK_PUSHED;
cleanup_rbuf(sk, 1);
if (!(val & 1))
- tp->ack.pingpong = 1;
+ icsk->icsk_ack.pingpong = 1;
}
}
break;
void tcp_get_info(struct sock *sk, struct tcp_info *info)
{
struct tcp_sock *tp = tcp_sk(sk);
+ const struct inet_connection_sock *icsk = inet_csk(sk);
u32 now = tcp_time_stamp;
memset(info, 0, sizeof(*info));
info->tcpi_state = sk->sk_state;
info->tcpi_ca_state = tp->ca_state;
- info->tcpi_retransmits = tp->retransmits;
+ info->tcpi_retransmits = icsk->icsk_retransmits;
info->tcpi_probes = tp->probes_out;
- info->tcpi_backoff = tp->backoff;
+ info->tcpi_backoff = icsk->icsk_backoff;
if (tp->rx_opt.tstamp_ok)
info->tcpi_options |= TCPI_OPT_TIMESTAMPS;
if (tp->ecn_flags&TCP_ECN_OK)
info->tcpi_options |= TCPI_OPT_ECN;
- info->tcpi_rto = jiffies_to_usecs(tp->rto);
- info->tcpi_ato = jiffies_to_usecs(tp->ack.ato);
+ info->tcpi_rto = jiffies_to_usecs(icsk->icsk_rto);
+ info->tcpi_ato = jiffies_to_usecs(icsk->icsk_ack.ato);
info->tcpi_snd_mss = tp->mss_cache;
- info->tcpi_rcv_mss = tp->ack.rcv_mss;
+ info->tcpi_rcv_mss = icsk->icsk_ack.rcv_mss;
info->tcpi_unacked = tp->packets_out;
info->tcpi_sacked = tp->sacked_out;
info->tcpi_fackets = tp->fackets_out;
info->tcpi_last_data_sent = jiffies_to_msecs(now - tp->lsndtime);
- info->tcpi_last_data_recv = jiffies_to_msecs(now - tp->ack.lrcvtime);
+ info->tcpi_last_data_recv = jiffies_to_msecs(now - icsk->icsk_ack.lrcvtime);
info->tcpi_last_ack_recv = jiffies_to_msecs(now - tp->rcv_tstamp);
info->tcpi_pmtu = tp->pmtu_cookie;
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 = tp->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 = !tp->ack.pingpong;
+ val = !icsk->icsk_ack.pingpong;
break;
case TCP_CONGESTION:
tcp_register_congestion_control(&tcp_reno);
}
-EXPORT_SYMBOL(tcp_accept);
EXPORT_SYMBOL(tcp_close);
-EXPORT_SYMBOL(tcp_destroy_sock);
+EXPORT_SYMBOL(inet_csk_destroy_sock);
EXPORT_SYMBOL(tcp_disconnect);
EXPORT_SYMBOL(tcp_getsockopt);
EXPORT_SYMBOL(tcp_ioctl);
git://bbs.cooldavid.org / net-next-2.6.git / blobdiff