#define TCP_DEBUG 1
#define FASTRETRANS_DEBUG 1
-/* Cancel timers, when they are not required. */
-#undef TCP_CLEAR_TIMERS
-
#include <linux/config.h>
#include <linux/list.h>
#include <linux/tcp.h>
#include <linux/slab.h>
#include <linux/cache.h>
#include <linux/percpu.h>
+
+#include <net/inet_connection_sock.h>
+#include <net/inet_timewait_sock.h>
#include <net/inet_hashtables.h>
#include <net/checksum.h>
#include <net/request_sock.h>
extern struct inet_hashinfo tcp_hashinfo;
extern atomic_t tcp_orphan_count;
-extern int tcp_tw_count;
extern void tcp_time_wait(struct sock *sk, int state, int timeo);
-extern void tcp_tw_deschedule(struct inet_timewait_sock *tw);
#define MAX_TCP_HEADER (128 + MAX_HEADER)
* timestamps. It must be less than
* minimal timewait lifetime.
*/
-
-#define TCP_TW_RECYCLE_SLOTS_LOG 5
-#define TCP_TW_RECYCLE_SLOTS (1<<TCP_TW_RECYCLE_SLOTS_LOG)
-
-/* If time > 4sec, it is "slow" path, no recycling is required,
- so that we select tick to get range about 4 seconds.
- */
-
-#if HZ <= 16 || HZ > 4096
-# error Unsupported: HZ <= 16 or HZ > 4096
-#elif HZ <= 32
-# define TCP_TW_RECYCLE_TICK (5+2-TCP_TW_RECYCLE_SLOTS_LOG)
-#elif HZ <= 64
-# define TCP_TW_RECYCLE_TICK (6+2-TCP_TW_RECYCLE_SLOTS_LOG)
-#elif HZ <= 128
-# define TCP_TW_RECYCLE_TICK (7+2-TCP_TW_RECYCLE_SLOTS_LOG)
-#elif HZ <= 256
-# define TCP_TW_RECYCLE_TICK (8+2-TCP_TW_RECYCLE_SLOTS_LOG)
-#elif HZ <= 512
-# define TCP_TW_RECYCLE_TICK (9+2-TCP_TW_RECYCLE_SLOTS_LOG)
-#elif HZ <= 1024
-# define TCP_TW_RECYCLE_TICK (10+2-TCP_TW_RECYCLE_SLOTS_LOG)
-#elif HZ <= 2048
-# define TCP_TW_RECYCLE_TICK (11+2-TCP_TW_RECYCLE_SLOTS_LOG)
-#else
-# define TCP_TW_RECYCLE_TICK (12+2-TCP_TW_RECYCLE_SLOTS_LOG)
-#endif
/*
* TCP option
*/
#define TCPOLEN_SACK_BASE_ALIGNED 4
#define TCPOLEN_SACK_PERBLOCK 8
-#define TCP_TIME_RETRANS 1 /* Retransmit timer */
-#define TCP_TIME_DACK 2 /* Delayed ack timer */
-#define TCP_TIME_PROBE0 3 /* Zero window probe timer */
-#define TCP_TIME_KEEPOPEN 4 /* Keepalive timer */
-
/* Flags in tp->nonagle */
#define TCP_NAGLE_OFF 1 /* Nagle's algo is disabled */
#define TCP_NAGLE_CORK 2 /* Socket is corked */
#define TCP_NAGLE_PUSH 4 /* Cork is overriden for already queued data */
+extern struct inet_timewait_death_row tcp_death_row;
+
/* sysctl variables for tcp */
extern int sysctl_tcp_timestamps;
extern int sysctl_tcp_window_scaling;
extern int sysctl_tcp_sack;
extern int sysctl_tcp_fin_timeout;
-extern int sysctl_tcp_tw_recycle;
extern int sysctl_tcp_keepalive_time;
extern int sysctl_tcp_keepalive_probes;
extern int sysctl_tcp_keepalive_intvl;
extern int sysctl_tcp_rfc1337;
extern int sysctl_tcp_abort_on_overflow;
extern int sysctl_tcp_max_orphans;
-extern int sysctl_tcp_max_tw_buckets;
extern int sysctl_tcp_fack;
extern int sysctl_tcp_reordering;
extern int sysctl_tcp_ecn;
extern atomic_t tcp_sockets_allocated;
extern int tcp_memory_pressure;
-#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
-#define TCP_INET_FAMILY(fam) ((fam) == AF_INET)
-#else
-#define TCP_INET_FAMILY(fam) 1
-#endif
-
/*
* Pointers to address related TCP functions
* (i.e. things that depend on the address family)
extern void tcp_rcv_space_adjust(struct sock *sk);
-enum tcp_ack_state_t
+static inline void tcp_dec_quickack_mode(struct sock *sk,
+ const unsigned int pkts)
{
- TCP_ACK_SCHED = 1,
- TCP_ACK_TIMER = 2,
- TCP_ACK_PUSHED= 4
-};
-
-static inline void tcp_schedule_ack(struct tcp_sock *tp)
-{
- tp->ack.pending |= TCP_ACK_SCHED;
-}
-
-static inline int tcp_ack_scheduled(struct tcp_sock *tp)
-{
- return tp->ack.pending&TCP_ACK_SCHED;
-}
-
-static __inline__ void tcp_dec_quickack_mode(struct tcp_sock *tp, unsigned int pkts)
-{
- if (tp->ack.quick) {
- if (pkts >= tp->ack.quick) {
- tp->ack.quick = 0;
+ struct inet_connection_sock *icsk = inet_csk(sk);
+ if (icsk->icsk_ack.quick) {
+ if (pkts >= icsk->icsk_ack.quick) {
+ icsk->icsk_ack.quick = 0;
/* Leaving quickack mode we deflate ATO. */
- tp->ack.ato = TCP_ATO_MIN;
+ icsk->icsk_ack.ato = TCP_ATO_MIN;
} else
- tp->ack.quick -= pkts;
+ icsk->icsk_ack.quick -= pkts;
}
}
-extern void tcp_enter_quickack_mode(struct tcp_sock *tp);
-
-static __inline__ void tcp_delack_init(struct tcp_sock *tp)
-{
- memset(&tp->ack, 0, sizeof(tp->ack));
-}
+extern void tcp_enter_quickack_mode(struct sock *sk);
static inline void tcp_clear_options(struct tcp_options_received *rx_opt)
{
extern void tcp_close(struct sock *sk,
long timeout);
-extern struct sock * tcp_accept(struct sock *sk, int flags, int *err);
extern unsigned int tcp_poll(struct file * file, struct socket *sock, struct poll_table_struct *wait);
extern int tcp_getsockopt(struct sock *sk, int level,
size_t len, int nonblock,
int flags, int *addr_len);
-extern int tcp_listen_start(struct sock *sk);
-
extern void tcp_parse_options(struct sk_buff *skb,
struct tcp_options_received *opt_rx,
int estab);
extern void tcp_xmit_retransmit_queue(struct sock *);
extern void tcp_simple_retransmit(struct sock *);
extern int tcp_trim_head(struct sock *, struct sk_buff *, u32);
+extern int tcp_fragment(struct sock *, struct sk_buff *, u32, unsigned int);
extern void tcp_send_probe0(struct sock *);
extern void tcp_send_partial(struct sock *);
/* tcp_timer.c */
extern void tcp_init_xmit_timers(struct sock *);
-extern void tcp_clear_xmit_timers(struct sock *);
+static inline void tcp_clear_xmit_timers(struct sock *sk)
+{
+ inet_csk_clear_xmit_timers(sk);
+}
-extern void tcp_delete_keepalive_timer(struct sock *);
-extern void tcp_reset_keepalive_timer(struct sock *, unsigned long);
extern unsigned int tcp_sync_mss(struct sock *sk, u32 pmtu);
extern unsigned int tcp_current_mss(struct sock *sk, int large);
-#ifdef TCP_DEBUG
-extern const char tcp_timer_bug_msg[];
-#endif
-
-/* tcp_diag.c */
+/* tcp.c */
extern void tcp_get_info(struct sock *, struct tcp_info *);
/* Read 'sendfile()'-style from a TCP socket */
extern int tcp_read_sock(struct sock *sk, read_descriptor_t *desc,
sk_read_actor_t recv_actor);
-static inline void tcp_clear_xmit_timer(struct sock *sk, int what)
-{
- struct tcp_sock *tp = tcp_sk(sk);
-
- switch (what) {
- case TCP_TIME_RETRANS:
- case TCP_TIME_PROBE0:
- tp->pending = 0;
-
-#ifdef TCP_CLEAR_TIMERS
- sk_stop_timer(sk, &tp->retransmit_timer);
-#endif
- break;
- case TCP_TIME_DACK:
- tp->ack.blocked = 0;
- tp->ack.pending = 0;
-
-#ifdef TCP_CLEAR_TIMERS
- sk_stop_timer(sk, &tp->delack_timer);
-#endif
- break;
- default:
-#ifdef TCP_DEBUG
- printk(tcp_timer_bug_msg);
-#endif
- return;
- };
-
-}
-
-/*
- * Reset the retransmission timer
- */
-static inline void tcp_reset_xmit_timer(struct sock *sk, int what, unsigned long when)
-{
- struct tcp_sock *tp = tcp_sk(sk);
-
- if (when > TCP_RTO_MAX) {
-#ifdef TCP_DEBUG
- printk(KERN_DEBUG "reset_xmit_timer sk=%p %d when=0x%lx, caller=%p\n", sk, what, when, current_text_addr());
-#endif
- when = TCP_RTO_MAX;
- }
-
- switch (what) {
- case TCP_TIME_RETRANS:
- case TCP_TIME_PROBE0:
- tp->pending = what;
- tp->timeout = jiffies+when;
- sk_reset_timer(sk, &tp->retransmit_timer, tp->timeout);
- break;
-
- case TCP_TIME_DACK:
- tp->ack.pending |= TCP_ACK_TIMER;
- tp->ack.timeout = jiffies+when;
- sk_reset_timer(sk, &tp->delack_timer, tp->ack.timeout);
- break;
-
- default:
-#ifdef TCP_DEBUG
- printk(tcp_timer_bug_msg);
-#endif
- return;
- };
-}
-
/* Initialize RCV_MSS value.
* RCV_MSS is an our guess about MSS used by the peer.
* We haven't any direct information about the MSS.
hint = min(hint, TCP_MIN_RCVMSS);
hint = max(hint, TCP_MIN_MSS);
- tp->ack.rcv_mss = hint;
+ inet_csk(sk)->icsk_ack.rcv_mss = hint;
}
static __inline__ void __tcp_fast_path_on(struct tcp_sock *tp, u32 snd_wnd)
tp->packets_out += tcp_skb_pcount(skb);
if (!orig)
- tcp_reset_xmit_timer(sk, TCP_TIME_RETRANS, tp->rto);
+ inet_csk_reset_xmit_timer(sk, ICSK_TIME_RETRANS,
+ inet_csk(sk)->icsk_rto, TCP_RTO_MAX);
}
static inline void tcp_packets_out_dec(struct tcp_sock *tp,
struct list_head list;
/* initialize private data (optional) */
- void (*init)(struct tcp_sock *tp);
+ void (*init)(struct sock *sk);
/* cleanup private data (optional) */
- void (*release)(struct tcp_sock *tp);
+ void (*release)(struct sock *sk);
/* return slow start threshold (required) */
- u32 (*ssthresh)(struct tcp_sock *tp);
+ u32 (*ssthresh)(struct sock *sk);
/* lower bound for congestion window (optional) */
- u32 (*min_cwnd)(struct tcp_sock *tp);
+ u32 (*min_cwnd)(struct sock *sk);
/* do new cwnd calculation (required) */
- void (*cong_avoid)(struct tcp_sock *tp, u32 ack,
+ void (*cong_avoid)(struct sock *sk, u32 ack,
u32 rtt, u32 in_flight, int good_ack);
/* round trip time sample per acked packet (optional) */
- void (*rtt_sample)(struct tcp_sock *tp, u32 usrtt);
+ void (*rtt_sample)(struct sock *sk, u32 usrtt);
/* call before changing ca_state (optional) */
- void (*set_state)(struct tcp_sock *tp, u8 new_state);
+ void (*set_state)(struct sock *sk, u8 new_state);
/* call when cwnd event occurs (optional) */
- void (*cwnd_event)(struct tcp_sock *tp, enum tcp_ca_event ev);
+ void (*cwnd_event)(struct sock *sk, enum tcp_ca_event ev);
/* new value of cwnd after loss (optional) */
- u32 (*undo_cwnd)(struct tcp_sock *tp);
+ u32 (*undo_cwnd)(struct sock *sk);
/* hook for packet ack accounting (optional) */
- void (*pkts_acked)(struct tcp_sock *tp, u32 num_acked);
- /* get info for tcp_diag (optional) */
- void (*get_info)(struct tcp_sock *tp, u32 ext, struct sk_buff *skb);
+ void (*pkts_acked)(struct sock *sk, u32 num_acked);
+ /* get info for inet_diag (optional) */
+ void (*get_info)(struct sock *sk, u32 ext, struct sk_buff *skb);
char name[TCP_CA_NAME_MAX];
struct module *owner;
extern int tcp_register_congestion_control(struct tcp_congestion_ops *type);
extern void tcp_unregister_congestion_control(struct tcp_congestion_ops *type);
-extern void tcp_init_congestion_control(struct tcp_sock *tp);
-extern void tcp_cleanup_congestion_control(struct tcp_sock *tp);
+extern void tcp_init_congestion_control(struct sock *sk);
+extern void tcp_cleanup_congestion_control(struct sock *sk);
extern int tcp_set_default_congestion_control(const char *name);
extern void tcp_get_default_congestion_control(char *name);
-extern int tcp_set_congestion_control(struct tcp_sock *tp, const char *name);
+extern int tcp_set_congestion_control(struct sock *sk, const char *name);
extern struct tcp_congestion_ops tcp_init_congestion_ops;
-extern u32 tcp_reno_ssthresh(struct tcp_sock *tp);
-extern void tcp_reno_cong_avoid(struct tcp_sock *tp, u32 ack,
+extern u32 tcp_reno_ssthresh(struct sock *sk);
+extern void tcp_reno_cong_avoid(struct sock *sk, u32 ack,
u32 rtt, u32 in_flight, int flag);
-extern u32 tcp_reno_min_cwnd(struct tcp_sock *tp);
+extern u32 tcp_reno_min_cwnd(struct sock *sk);
extern struct tcp_congestion_ops tcp_reno;
-static inline void tcp_set_ca_state(struct tcp_sock *tp, u8 ca_state)
+static inline void tcp_set_ca_state(struct sock *sk, const u8 ca_state)
{
- if (tp->ca_ops->set_state)
- tp->ca_ops->set_state(tp, ca_state);
- tp->ca_state = ca_state;
+ struct inet_connection_sock *icsk = inet_csk(sk);
+
+ if (icsk->icsk_ca_ops->set_state)
+ icsk->icsk_ca_ops->set_state(sk, ca_state);
+ icsk->icsk_ca_state = ca_state;
}
-static inline void tcp_ca_event(struct tcp_sock *tp, enum tcp_ca_event event)
+static inline void tcp_ca_event(struct sock *sk, const enum tcp_ca_event event)
{
- if (tp->ca_ops->cwnd_event)
- tp->ca_ops->cwnd_event(tp, event);
+ const struct inet_connection_sock *icsk = inet_csk(sk);
+
+ if (icsk->icsk_ca_ops->cwnd_event)
+ icsk->icsk_ca_ops->cwnd_event(sk, event);
}
/* This determines how many packets are "in the network" to the best
* The exception is rate halving phase, when cwnd is decreasing towards
* ssthresh.
*/
-static inline __u32 tcp_current_ssthresh(struct tcp_sock *tp)
+static inline __u32 tcp_current_ssthresh(const struct sock *sk)
{
- if ((1<<tp->ca_state)&(TCPF_CA_CWR|TCPF_CA_Recovery))
+ const struct tcp_sock *tp = tcp_sk(sk);
+ if ((1 << inet_csk(sk)->icsk_ca_state) & (TCPF_CA_CWR | TCPF_CA_Recovery))
return tp->snd_ssthresh;
else
return max(tp->snd_ssthresh,
}
/* Set slow start threshold and cwnd not falling to slow start */
-static inline void __tcp_enter_cwr(struct tcp_sock *tp)
+static inline void __tcp_enter_cwr(struct sock *sk)
{
+ const struct inet_connection_sock *icsk = inet_csk(sk);
+ struct tcp_sock *tp = tcp_sk(sk);
+
tp->undo_marker = 0;
- tp->snd_ssthresh = tp->ca_ops->ssthresh(tp);
+ tp->snd_ssthresh = icsk->icsk_ca_ops->ssthresh(sk);
tp->snd_cwnd = min(tp->snd_cwnd,
tcp_packets_in_flight(tp) + 1U);
tp->snd_cwnd_cnt = 0;
TCP_ECN_queue_cwr(tp);
}
-static inline void tcp_enter_cwr(struct tcp_sock *tp)
+static inline void tcp_enter_cwr(struct sock *sk)
{
+ struct tcp_sock *tp = tcp_sk(sk);
+
tp->prior_ssthresh = 0;
- if (tp->ca_state < TCP_CA_CWR) {
- __tcp_enter_cwr(tp);
- tcp_set_ca_state(tp, TCP_CA_CWR);
+ if (inet_csk(sk)->icsk_ca_state < TCP_CA_CWR) {
+ __tcp_enter_cwr(sk);
+ tcp_set_ca_state(sk, TCP_CA_CWR);
}
}
static __inline__ void tcp_check_probe_timer(struct sock *sk, struct tcp_sock *tp)
{
- if (!tp->packets_out && !tp->pending)
- tcp_reset_xmit_timer(sk, TCP_TIME_PROBE0, tp->rto);
+ const struct inet_connection_sock *icsk = inet_csk(sk);
+ if (!tp->packets_out && !icsk->icsk_pending)
+ inet_csk_reset_xmit_timer(sk, ICSK_TIME_PROBE0,
+ icsk->icsk_rto, TCP_RTO_MAX);
}
static __inline__ void tcp_push_pending_frames(struct sock *sk,
tp->snd_wl1 = seq;
}
-extern void tcp_destroy_sock(struct sock *sk);
-
-
/*
* Calculate(/check) TCP checksum
*/
tp->ucopy.memory = 0;
} else if (skb_queue_len(&tp->ucopy.prequeue) == 1) {
wake_up_interruptible(sk->sk_sleep);
- if (!tcp_ack_scheduled(tp))
- tcp_reset_xmit_timer(sk, TCP_TIME_DACK, (3*TCP_RTO_MIN)/4);
+ if (!inet_csk_ack_scheduled(sk))
+ inet_csk_reset_xmit_timer(sk, ICSK_TIME_DACK,
+ (3 * TCP_RTO_MIN) / 4,
+ TCP_RTO_MAX);
}
return 1;
}
TCP_INC_STATS(TCP_MIB_ESTABRESETS);
sk->sk_prot->unhash(sk);
- if (inet_sk(sk)->bind_hash &&
+ if (inet_csk(sk)->icsk_bind_hash &&
!(sk->sk_userlocks & SOCK_BINDPORT_LOCK))
inet_put_port(&tcp_hashinfo, sk);
/* fall through */
if (!sock_flag(sk, SOCK_DEAD))
sk->sk_state_change(sk);
else
- tcp_destroy_sock(sk);
+ inet_csk_destroy_sock(sk);
}
static __inline__ void tcp_sack_reset(struct tcp_options_received *rx_opt)
return tcp_win_from_space(sk->sk_rcvbuf);
}
-static inline void tcp_acceptq_queue(struct sock *sk, struct request_sock *req,
- struct sock *child)
-{
- reqsk_queue_add(&tcp_sk(sk)->accept_queue, req, sk, child);
-}
-
-static inline void
-tcp_synq_removed(struct sock *sk, struct request_sock *req)
-{
- if (reqsk_queue_removed(&tcp_sk(sk)->accept_queue, req) == 0)
- tcp_delete_keepalive_timer(sk);
-}
-
-static inline void tcp_synq_added(struct sock *sk)
-{
- if (reqsk_queue_added(&tcp_sk(sk)->accept_queue) == 0)
- tcp_reset_keepalive_timer(sk, TCP_TIMEOUT_INIT);
-}
-
-static inline int tcp_synq_len(struct sock *sk)
-{
- return reqsk_queue_len(&tcp_sk(sk)->accept_queue);
-}
-
-static inline int tcp_synq_young(struct sock *sk)
-{
- return reqsk_queue_len_young(&tcp_sk(sk)->accept_queue);
-}
-
-static inline int tcp_synq_is_full(struct sock *sk)
-{
- return reqsk_queue_is_full(&tcp_sk(sk)->accept_queue);
-}
-
-static inline void tcp_synq_unlink(struct tcp_sock *tp, struct request_sock *req,
- struct request_sock **prev)
-{
- reqsk_queue_unlink(&tp->accept_queue, req, prev);
-}
-
-static inline void tcp_synq_drop(struct sock *sk, struct request_sock *req,
- struct request_sock **prev)
-{
- tcp_synq_unlink(tcp_sk(sk), req, prev);
- tcp_synq_removed(sk, req);
- reqsk_free(req);
-}
-
static __inline__ void tcp_openreq_init(struct request_sock *req,
struct tcp_options_received *rx_opt,
struct sk_buff *skb)
return tp->keepalive_time ? : sysctl_tcp_keepalive_time;
}
-static inline int tcp_fin_time(const struct tcp_sock *tp)
+static inline int tcp_fin_time(const struct sock *sk)
{
- int fin_timeout = tp->linger2 ? : sysctl_tcp_fin_timeout;
+ int fin_timeout = tcp_sk(sk)->linger2 ? : sysctl_tcp_fin_timeout;
+ const int rto = inet_csk(sk)->icsk_rto;
- if (fin_timeout < (tp->rto<<2) - (tp->rto>>1))
- fin_timeout = (tp->rto<<2) - (tp->rto>>1);
+ if (fin_timeout < (rto << 2) - (rto >> 1))
+ fin_timeout = (rto << 2) - (rto >> 1);
return fin_timeout;
}
extern int tcp_proc_register(struct tcp_seq_afinfo *afinfo);
extern void tcp_proc_unregister(struct tcp_seq_afinfo *afinfo);
+extern struct request_sock_ops tcp_request_sock_ops;
+
+extern int tcp_v4_destroy_sock(struct sock *sk);
+
+#ifdef CONFIG_PROC_FS
+extern int tcp4_proc_init(void);
+extern void tcp4_proc_exit(void);
+#endif
+
+extern void tcp_v4_init(struct net_proto_family *ops);
+extern void tcp_init(void);
+
#endif /* _TCP_H */
projects / linux-2.6 / blobdiff