* ACK bit.
* Andi Kleen : Implemented fast path mtu discovery.
* Fixed many serious bugs in the
- * open_request handling and moved
+ * request_sock handling and moved
* most of it into the af independent code.
* Added tail drop and some other bugfixes.
* Added new listen sematics.
return (jhash_2words(raddr, (u32) rport, rnd) & (TCP_SYNQ_HSIZE - 1));
}
-static struct open_request *tcp_v4_search_req(struct tcp_sock *tp,
- struct open_request ***prevp,
+static struct request_sock *tcp_v4_search_req(struct tcp_sock *tp,
+ struct request_sock ***prevp,
__u16 rport,
__u32 raddr, __u32 laddr)
{
- struct tcp_listen_opt *lopt = tp->listen_opt;
- struct open_request *req, **prev;
+ struct listen_sock *lopt = tp->accept_queue.listen_opt;
+ struct request_sock *req, **prev;
for (prev = &lopt->syn_table[tcp_v4_synq_hash(raddr, rport, lopt->hash_rnd)];
(req = *prev) != NULL;
if (ireq->rmt_port == rport &&
ireq->rmt_addr == raddr &&
ireq->loc_addr == laddr &&
- TCP_INET_FAMILY(req->class->family)) {
+ TCP_INET_FAMILY(req->rsk_ops->family)) {
BUG_TRAP(!req->sk);
*prevp = prev;
break;
return req;
}
-static void tcp_v4_synq_add(struct sock *sk, struct open_request *req)
+static void tcp_v4_synq_add(struct sock *sk, struct request_sock *req)
{
struct tcp_sock *tp = tcp_sk(sk);
- struct tcp_listen_opt *lopt = tp->listen_opt;
+ struct listen_sock *lopt = tp->accept_queue.listen_opt;
u32 h = tcp_v4_synq_hash(inet_rsk(req)->rmt_addr, inet_rsk(req)->rmt_port, lopt->hash_rnd);
- req->expires = jiffies + TCP_TIMEOUT_INIT;
- req->retrans = 0;
- req->sk = NULL;
- req->dl_next = lopt->syn_table[h];
-
- write_lock(&tp->syn_wait_lock);
- lopt->syn_table[h] = req;
- write_unlock(&tp->syn_wait_lock);
-
+ reqsk_queue_hash_req(&tp->accept_queue, h, req, TCP_TIMEOUT_INIT);
tcp_synq_added(sk);
}
}
switch (sk->sk_state) {
- struct open_request *req, **prev;
+ struct request_sock *req, **prev;
case TCP_LISTEN:
if (sock_owned_by_user(sk))
goto out;
tcp_tw_put(tw);
}
-static void tcp_v4_or_send_ack(struct sk_buff *skb, struct open_request *req)
+static void tcp_v4_reqsk_send_ack(struct sk_buff *skb, struct request_sock *req)
{
tcp_v4_send_ack(skb, tcp_rsk(req)->snt_isn + 1, tcp_rsk(req)->rcv_isn + 1, req->rcv_wnd,
req->ts_recent);
}
static struct dst_entry* tcp_v4_route_req(struct sock *sk,
- struct open_request *req)
+ struct request_sock *req)
{
struct rtable *rt;
const struct inet_request_sock *ireq = inet_rsk(req);
/*
* Send a SYN-ACK after having received an ACK.
- * This still operates on a open_request only, not on a big
+ * This still operates on a request_sock only, not on a big
* socket.
*/
-static int tcp_v4_send_synack(struct sock *sk, struct open_request *req,
+static int tcp_v4_send_synack(struct sock *sk, struct request_sock *req,
struct dst_entry *dst)
{
const struct inet_request_sock *ireq = inet_rsk(req);
}
/*
- * IPv4 open_request destructor.
+ * IPv4 request_sock destructor.
*/
-static void tcp_v4_or_free(struct open_request *req)
+static void tcp_v4_reqsk_destructor(struct request_sock *req)
{
if (inet_rsk(req)->opt)
kfree(inet_rsk(req)->opt);
}
/*
- * Save and compile IPv4 options into the open_request if needed.
+ * Save and compile IPv4 options into the request_sock if needed.
*/
static inline struct ip_options *tcp_v4_save_options(struct sock *sk,
struct sk_buff *skb)
return dopt;
}
-/*
- * Maximum number of SYN_RECV sockets in queue per LISTEN socket.
- * One SYN_RECV socket costs about 80bytes on a 32bit machine.
- * It would be better to replace it with a global counter for all sockets
- * but then some measure against one socket starving all other sockets
- * would be needed.
- *
- * It was 128 by default. Experiments with real servers show, that
- * it is absolutely not enough even at 100conn/sec. 256 cures most
- * of problems. This value is adjusted to 128 for very small machines
- * (<=32Mb of memory) and to 1024 on normal or better ones (>=256Mb).
- * Further increasing requires to change hash table size.
- */
-int sysctl_max_syn_backlog = 256;
-
-struct or_calltable or_ipv4 = {
+struct request_sock_ops tcp_request_sock_ops = {
.family = PF_INET,
.obj_size = sizeof(struct tcp_request_sock),
.rtx_syn_ack = tcp_v4_send_synack,
- .send_ack = tcp_v4_or_send_ack,
- .destructor = tcp_v4_or_free,
+ .send_ack = tcp_v4_reqsk_send_ack,
+ .destructor = tcp_v4_reqsk_destructor,
.send_reset = tcp_v4_send_reset,
};
{
struct inet_request_sock *ireq;
struct tcp_options_received tmp_opt;
- struct open_request *req;
+ struct request_sock *req;
__u32 saddr = skb->nh.iph->saddr;
__u32 daddr = skb->nh.iph->daddr;
__u32 isn = TCP_SKB_CB(skb)->when;
if (sk_acceptq_is_full(sk) && tcp_synq_young(sk) > 1)
goto drop;
- req = tcp_openreq_alloc(&or_ipv4);
+ req = reqsk_alloc(&tcp_request_sock_ops);
if (!req)
goto drop;
goto drop_and_free;
if (want_cookie) {
- tcp_openreq_free(req);
+ reqsk_free(req);
} else {
tcp_v4_synq_add(sk, req);
}
return 0;
drop_and_free:
- tcp_openreq_free(req);
+ reqsk_free(req);
drop:
TCP_INC_STATS_BH(TCP_MIB_ATTEMPTFAILS);
return 0;
* now create the new socket.
*/
struct sock *tcp_v4_syn_recv_sock(struct sock *sk, struct sk_buff *skb,
- struct open_request *req,
+ struct request_sock *req,
struct dst_entry *dst)
{
struct inet_request_sock *ireq;
struct iphdr *iph = skb->nh.iph;
struct tcp_sock *tp = tcp_sk(sk);
struct sock *nsk;
- struct open_request **prev;
+ struct request_sock **prev;
/* Find possible connection requests. */
- struct open_request *req = tcp_v4_search_req(tp, &prev, th->source,
+ struct request_sock *req = tcp_v4_search_req(tp, &prev, th->source,
iph->saddr, iph->daddr);
if (req)
return tcp_check_req(sk, skb, req, prev);
*/
tp->snd_ssthresh = 0x7fffffff; /* Infinity */
tp->snd_cwnd_clamp = ~0;
- tp->mss_cache_std = tp->mss_cache = 536;
+ tp->mss_cache = 536;
tp->reordering = sysctl_tcp_reordering;
+ tp->ca_ops = &tcp_init_congestion_ops;
sk->sk_state = TCP_CLOSE;
tcp_clear_xmit_timers(sk);
+ tcp_cleanup_congestion_control(tp);
+
/* Cleanup up the write buffer. */
sk_stream_writequeue_purge(sk);
++st->num;
if (st->state == TCP_SEQ_STATE_OPENREQ) {
- struct open_request *req = cur;
+ struct request_sock *req = cur;
tp = tcp_sk(st->syn_wait_sk);
req = req->dl_next;
while (1) {
while (req) {
- if (req->class->family == st->family) {
+ if (req->rsk_ops->family == st->family) {
cur = req;
goto out;
}
if (++st->sbucket >= TCP_SYNQ_HSIZE)
break;
get_req:
- req = tp->listen_opt->syn_table[st->sbucket];
+ req = tp->accept_queue.listen_opt->syn_table[st->sbucket];
}
sk = sk_next(st->syn_wait_sk);
st->state = TCP_SEQ_STATE_LISTENING;
- read_unlock_bh(&tp->syn_wait_lock);
+ read_unlock_bh(&tp->accept_queue.syn_wait_lock);
} else {
tp = tcp_sk(sk);
- read_lock_bh(&tp->syn_wait_lock);
- if (tp->listen_opt && tp->listen_opt->qlen)
+ read_lock_bh(&tp->accept_queue.syn_wait_lock);
+ if (reqsk_queue_len(&tp->accept_queue))
goto start_req;
- read_unlock_bh(&tp->syn_wait_lock);
+ read_unlock_bh(&tp->accept_queue.syn_wait_lock);
sk = sk_next(sk);
}
get_sk:
goto out;
}
tp = tcp_sk(sk);
- read_lock_bh(&tp->syn_wait_lock);
- if (tp->listen_opt && tp->listen_opt->qlen) {
+ read_lock_bh(&tp->accept_queue.syn_wait_lock);
+ if (reqsk_queue_len(&tp->accept_queue)) {
start_req:
st->uid = sock_i_uid(sk);
st->syn_wait_sk = sk;
st->sbucket = 0;
goto get_req;
}
- read_unlock_bh(&tp->syn_wait_lock);
+ read_unlock_bh(&tp->accept_queue.syn_wait_lock);
}
if (++st->bucket < TCP_LHTABLE_SIZE) {
sk = sk_head(&tcp_listening_hash[st->bucket]);
case TCP_SEQ_STATE_OPENREQ:
if (v) {
struct tcp_sock *tp = tcp_sk(st->syn_wait_sk);
- read_unlock_bh(&tp->syn_wait_lock);
+ read_unlock_bh(&tp->accept_queue.syn_wait_lock);
}
case TCP_SEQ_STATE_LISTENING:
if (v != SEQ_START_TOKEN)
memset(afinfo->seq_fops, 0, sizeof(*afinfo->seq_fops));
}
-static void get_openreq4(struct sock *sk, struct open_request *req,
+static void get_openreq4(struct sock *sk, struct request_sock *req,
char *tmpbuf, int i, int uid)
{
const struct inet_request_sock *ireq = inet_rsk(req);
.sysctl_rmem = sysctl_tcp_rmem,
.max_header = MAX_TCP_HEADER,
.obj_size = sizeof(struct tcp_sock),
- .rsk_prot = &or_ipv4,
+ .rsk_prot = &tcp_request_sock_ops,
};
EXPORT_SYMBOL(tcp_proc_unregister);
#endif
EXPORT_SYMBOL(sysctl_local_port_range);
-EXPORT_SYMBOL(sysctl_max_syn_backlog);
EXPORT_SYMBOL(sysctl_tcp_low_latency);
EXPORT_SYMBOL(sysctl_tcp_tw_reuse);