diff options
Diffstat (limited to 'include/net/tcp.h')
| -rw-r--r-- | include/net/tcp.h | 2402 |
1 files changed, 1048 insertions, 1354 deletions
diff --git a/include/net/tcp.h b/include/net/tcp.h index f730935b824..7286db80e8b 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -18,365 +18,43 @@ #ifndef _TCP_H #define _TCP_H -#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/bug.h> #include <linux/slab.h> #include <linux/cache.h> #include <linux/percpu.h> +#include <linux/skbuff.h> +#include <linux/dmaengine.h> +#include <linux/crypto.h> +#include <linux/cryptohash.h> +#include <linux/kref.h> +#include <linux/ktime.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> #include <net/sock.h> #include <net/snmp.h> #include <net/ip.h> -#if defined(CONFIG_IPV6) || defined (CONFIG_IPV6_MODULE) -#include <linux/ipv6.h> -#endif -#include <linux/seq_file.h> - -/* This is for all connections with a full identity, no wildcards. - * New scheme, half the table is for TIME_WAIT, the other half is - * for the rest. I'll experiment with dynamic table growth later. - */ -struct tcp_ehash_bucket { - rwlock_t lock; - struct hlist_head chain; -} __attribute__((__aligned__(8))); - -/* This is for listening sockets, thus all sockets which possess wildcards. */ -#define TCP_LHTABLE_SIZE 32 /* Yes, really, this is all you need. */ - -/* There are a few simple rules, which allow for local port reuse by - * an application. In essence: - * - * 1) Sockets bound to different interfaces may share a local port. - * Failing that, goto test 2. - * 2) If all sockets have sk->sk_reuse set, and none of them are in - * TCP_LISTEN state, the port may be shared. - * Failing that, goto test 3. - * 3) If all sockets are bound to a specific inet_sk(sk)->rcv_saddr local - * address, and none of them are the same, the port may be - * shared. - * Failing this, the port cannot be shared. - * - * The interesting point, is test #2. This is what an FTP server does - * all day. To optimize this case we use a specific flag bit defined - * below. As we add sockets to a bind bucket list, we perform a - * check of: (newsk->sk_reuse && (newsk->sk_state != TCP_LISTEN)) - * As long as all sockets added to a bind bucket pass this test, - * the flag bit will be set. - * The resulting situation is that tcp_v[46]_verify_bind() can just check - * for this flag bit, if it is set and the socket trying to bind has - * sk->sk_reuse set, we don't even have to walk the owners list at all, - * we return that it is ok to bind this socket to the requested local port. - * - * Sounds like a lot of work, but it is worth it. In a more naive - * implementation (ie. current FreeBSD etc.) the entire list of ports - * must be walked for each data port opened by an ftp server. Needless - * to say, this does not scale at all. With a couple thousand FTP - * users logged onto your box, isn't it nice to know that new data - * ports are created in O(1) time? I thought so. ;-) -DaveM - */ -struct tcp_bind_bucket { - unsigned short port; - signed short fastreuse; - struct hlist_node node; - struct hlist_head owners; -}; - -#define tb_for_each(tb, node, head) hlist_for_each_entry(tb, node, head, node) - -struct tcp_bind_hashbucket { - spinlock_t lock; - struct hlist_head chain; -}; - -static inline struct tcp_bind_bucket *__tb_head(struct tcp_bind_hashbucket *head) -{ - return hlist_entry(head->chain.first, struct tcp_bind_bucket, node); -} - -static inline struct tcp_bind_bucket *tb_head(struct tcp_bind_hashbucket *head) -{ - return hlist_empty(&head->chain) ? NULL : __tb_head(head); -} - -extern struct tcp_hashinfo { - /* This is for sockets with full identity only. Sockets here will - * always be without wildcards and will have the following invariant: - * - * TCP_ESTABLISHED <= sk->sk_state < TCP_CLOSE - * - * First half of the table is for sockets not in TIME_WAIT, second half - * is for TIME_WAIT sockets only. - */ - struct tcp_ehash_bucket *__tcp_ehash; - - /* Ok, let's try this, I give up, we do need a local binding - * TCP hash as well as the others for fast bind/connect. - */ - struct tcp_bind_hashbucket *__tcp_bhash; - - int __tcp_bhash_size; - int __tcp_ehash_size; - - /* All sockets in TCP_LISTEN state will be in here. This is the only - * table where wildcard'd TCP sockets can exist. Hash function here - * is just local port number. - */ - struct hlist_head __tcp_listening_hash[TCP_LHTABLE_SIZE]; - - /* All the above members are written once at bootup and - * never written again _or_ are predominantly read-access. - * - * Now align to a new cache line as all the following members - * are often dirty. - */ - rwlock_t __tcp_lhash_lock ____cacheline_aligned; - atomic_t __tcp_lhash_users; - wait_queue_head_t __tcp_lhash_wait; - spinlock_t __tcp_portalloc_lock; -} tcp_hashinfo; - -#define tcp_ehash (tcp_hashinfo.__tcp_ehash) -#define tcp_bhash (tcp_hashinfo.__tcp_bhash) -#define tcp_ehash_size (tcp_hashinfo.__tcp_ehash_size) -#define tcp_bhash_size (tcp_hashinfo.__tcp_bhash_size) -#define tcp_listening_hash (tcp_hashinfo.__tcp_listening_hash) -#define tcp_lhash_lock (tcp_hashinfo.__tcp_lhash_lock) -#define tcp_lhash_users (tcp_hashinfo.__tcp_lhash_users) -#define tcp_lhash_wait (tcp_hashinfo.__tcp_lhash_wait) -#define tcp_portalloc_lock (tcp_hashinfo.__tcp_portalloc_lock) - -extern kmem_cache_t *tcp_bucket_cachep; -extern struct tcp_bind_bucket *tcp_bucket_create(struct tcp_bind_hashbucket *head, - unsigned short snum); -extern void tcp_bucket_destroy(struct tcp_bind_bucket *tb); -extern void tcp_bucket_unlock(struct sock *sk); -extern int tcp_port_rover; - -/* These are AF independent. */ -static __inline__ int tcp_bhashfn(__u16 lport) -{ - return (lport & (tcp_bhash_size - 1)); -} - -extern void tcp_bind_hash(struct sock *sk, struct tcp_bind_bucket *tb, - unsigned short snum); - -#if (BITS_PER_LONG == 64) -#define TCP_ADDRCMP_ALIGN_BYTES 8 -#else -#define TCP_ADDRCMP_ALIGN_BYTES 4 -#endif - -/* This is a TIME_WAIT bucket. It works around the memory consumption - * problems of sockets in such a state on heavily loaded servers, but - * without violating the protocol specification. - */ -struct tcp_tw_bucket { - /* - * Now struct sock also uses sock_common, so please just - * don't add nothing before this first member (__tw_common) --acme - */ - struct sock_common __tw_common; -#define tw_family __tw_common.skc_family -#define tw_state __tw_common.skc_state -#define tw_reuse __tw_common.skc_reuse -#define tw_bound_dev_if __tw_common.skc_bound_dev_if -#define tw_node __tw_common.skc_node -#define tw_bind_node __tw_common.skc_bind_node -#define tw_refcnt __tw_common.skc_refcnt - volatile unsigned char tw_substate; - unsigned char tw_rcv_wscale; - __u16 tw_sport; - /* Socket demultiplex comparisons on incoming packets. */ - /* these five are in inet_sock */ - __u32 tw_daddr - __attribute__((aligned(TCP_ADDRCMP_ALIGN_BYTES))); - __u32 tw_rcv_saddr; - __u16 tw_dport; - __u16 tw_num; - /* And these are ours. */ - int tw_hashent; - int tw_timeout; - __u32 tw_rcv_nxt; - __u32 tw_snd_nxt; - __u32 tw_rcv_wnd; - __u32 tw_ts_recent; - long tw_ts_recent_stamp; - unsigned long tw_ttd; - struct tcp_bind_bucket *tw_tb; - struct hlist_node tw_death_node; -#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) - struct in6_addr tw_v6_daddr; - struct in6_addr tw_v6_rcv_saddr; - int tw_v6_ipv6only; -#endif -}; - -static __inline__ void tw_add_node(struct tcp_tw_bucket *tw, - struct hlist_head *list) -{ - hlist_add_head(&tw->tw_node, list); -} - -static __inline__ void tw_add_bind_node(struct tcp_tw_bucket *tw, - struct hlist_head *list) -{ - hlist_add_head(&tw->tw_bind_node, list); -} - -static inline int tw_dead_hashed(struct tcp_tw_bucket *tw) -{ - return tw->tw_death_node.pprev != NULL; -} - -static __inline__ void tw_dead_node_init(struct tcp_tw_bucket *tw) -{ - tw->tw_death_node.pprev = NULL; -} - -static __inline__ void __tw_del_dead_node(struct tcp_tw_bucket *tw) -{ - __hlist_del(&tw->tw_death_node); - tw_dead_node_init(tw); -} - -static __inline__ int tw_del_dead_node(struct tcp_tw_bucket *tw) -{ - if (tw_dead_hashed(tw)) { - __tw_del_dead_node(tw); - return 1; - } - return 0; -} - -#define tw_for_each(tw, node, head) \ - hlist_for_each_entry(tw, node, head, tw_node) - -#define tw_for_each_inmate(tw, node, jail) \ - hlist_for_each_entry(tw, node, jail, tw_death_node) - -#define tw_for_each_inmate_safe(tw, node, safe, jail) \ - hlist_for_each_entry_safe(tw, node, safe, jail, tw_death_node) - -#define tcptw_sk(__sk) ((struct tcp_tw_bucket *)(__sk)) - -static inline u32 tcp_v4_rcv_saddr(const struct sock *sk) -{ - return likely(sk->sk_state != TCP_TIME_WAIT) ? - inet_sk(sk)->rcv_saddr : tcptw_sk(sk)->tw_rcv_saddr; -} +#include <net/tcp_states.h> +#include <net/inet_ecn.h> +#include <net/dst.h> -#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) -static inline struct in6_addr *__tcp_v6_rcv_saddr(const struct sock *sk) -{ - return likely(sk->sk_state != TCP_TIME_WAIT) ? - &inet6_sk(sk)->rcv_saddr : &tcptw_sk(sk)->tw_v6_rcv_saddr; -} - -static inline struct in6_addr *tcp_v6_rcv_saddr(const struct sock *sk) -{ - return sk->sk_family == AF_INET6 ? __tcp_v6_rcv_saddr(sk) : NULL; -} - -#define tcptw_sk_ipv6only(__sk) (tcptw_sk(__sk)->tw_v6_ipv6only) - -static inline int tcp_v6_ipv6only(const struct sock *sk) -{ - return likely(sk->sk_state != TCP_TIME_WAIT) ? - ipv6_only_sock(sk) : tcptw_sk_ipv6only(sk); -} -#else -# define __tcp_v6_rcv_saddr(__sk) NULL -# define tcp_v6_rcv_saddr(__sk) NULL -# define tcptw_sk_ipv6only(__sk) 0 -# define tcp_v6_ipv6only(__sk) 0 -#endif - -extern kmem_cache_t *tcp_timewait_cachep; - -static inline void tcp_tw_put(struct tcp_tw_bucket *tw) -{ - if (atomic_dec_and_test(&tw->tw_refcnt)) { -#ifdef INET_REFCNT_DEBUG - printk(KERN_DEBUG "tw_bucket %p released\n", tw); -#endif - kmem_cache_free(tcp_timewait_cachep, tw); - } -} - -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 tcp_tw_bucket *tw); +#include <linux/seq_file.h> +#include <linux/memcontrol.h> +extern struct inet_hashinfo tcp_hashinfo; -/* Socket demux engine toys. */ -#ifdef __BIG_ENDIAN -#define TCP_COMBINED_PORTS(__sport, __dport) \ - (((__u32)(__sport)<<16) | (__u32)(__dport)) -#else /* __LITTLE_ENDIAN */ -#define TCP_COMBINED_PORTS(__sport, __dport) \ - (((__u32)(__dport)<<16) | (__u32)(__sport)) -#endif - -#if (BITS_PER_LONG == 64) -#ifdef __BIG_ENDIAN -#define TCP_V4_ADDR_COOKIE(__name, __saddr, __daddr) \ - __u64 __name = (((__u64)(__saddr))<<32)|((__u64)(__daddr)); -#else /* __LITTLE_ENDIAN */ -#define TCP_V4_ADDR_COOKIE(__name, __saddr, __daddr) \ - __u64 __name = (((__u64)(__daddr))<<32)|((__u64)(__saddr)); -#endif /* __BIG_ENDIAN */ -#define TCP_IPV4_MATCH(__sk, __cookie, __saddr, __daddr, __ports, __dif)\ - (((*((__u64 *)&(inet_sk(__sk)->daddr)))== (__cookie)) && \ - ((*((__u32 *)&(inet_sk(__sk)->dport)))== (__ports)) && \ - (!((__sk)->sk_bound_dev_if) || ((__sk)->sk_bound_dev_if == (__dif)))) -#define TCP_IPV4_TW_MATCH(__sk, __cookie, __saddr, __daddr, __ports, __dif)\ - (((*((__u64 *)&(tcptw_sk(__sk)->tw_daddr))) == (__cookie)) && \ - ((*((__u32 *)&(tcptw_sk(__sk)->tw_dport))) == (__ports)) && \ - (!((__sk)->sk_bound_dev_if) || ((__sk)->sk_bound_dev_if == (__dif)))) -#else /* 32-bit arch */ -#define TCP_V4_ADDR_COOKIE(__name, __saddr, __daddr) -#define TCP_IPV4_MATCH(__sk, __cookie, __saddr, __daddr, __ports, __dif)\ - ((inet_sk(__sk)->daddr == (__saddr)) && \ - (inet_sk(__sk)->rcv_saddr == (__daddr)) && \ - ((*((__u32 *)&(inet_sk(__sk)->dport)))== (__ports)) && \ - (!((__sk)->sk_bound_dev_if) || ((__sk)->sk_bound_dev_if == (__dif)))) -#define TCP_IPV4_TW_MATCH(__sk, __cookie, __saddr, __daddr, __ports, __dif)\ - ((tcptw_sk(__sk)->tw_daddr == (__saddr)) && \ - (tcptw_sk(__sk)->tw_rcv_saddr == (__daddr)) && \ - ((*((__u32 *)&(tcptw_sk(__sk)->tw_dport))) == (__ports)) && \ - (!((__sk)->sk_bound_dev_if) || ((__sk)->sk_bound_dev_if == (__dif)))) -#endif /* 64-bit arch */ - -#define TCP_IPV6_MATCH(__sk, __saddr, __daddr, __ports, __dif) \ - (((*((__u32 *)&(inet_sk(__sk)->dport)))== (__ports)) && \ - ((__sk)->sk_family == AF_INET6) && \ - ipv6_addr_equal(&inet6_sk(__sk)->daddr, (__saddr)) && \ - ipv6_addr_equal(&inet6_sk(__sk)->rcv_saddr, (__daddr)) && \ - (!((__sk)->sk_bound_dev_if) || ((__sk)->sk_bound_dev_if == (__dif)))) - -/* These can have wildcards, don't try too hard. */ -static __inline__ int tcp_lhashfn(unsigned short num) -{ - return num & (TCP_LHTABLE_SIZE - 1); -} - -static __inline__ int tcp_sk_listen_hashfn(struct sock *sk) -{ - return tcp_lhashfn(inet_sk(sk)->num); -} +extern struct percpu_counter tcp_orphan_count; +void tcp_time_wait(struct sock *sk, int state, int timeo); #define MAX_TCP_HEADER (128 + MAX_HEADER) +#define MAX_TCP_OPTION_SPACE 40 /* * Never offer a window over 32767 without using window scaling. Some @@ -387,8 +65,8 @@ static __inline__ int tcp_sk_listen_hashfn(struct sock *sk) /* Minimal accepted MSS. It is (60+60+8) - (20+20). */ #define TCP_MIN_MSS 88U -/* Minimal RCV_MSS. */ -#define TCP_MIN_RCVMSS 536U +/* The least MTU to use for probing */ +#define TCP_BASE_MSS 512 /* After receiving this amount of duplicate ACKs fast retransmit starts. */ #define TCP_FASTRETRANS_THRESH 3 @@ -418,17 +96,21 @@ static __inline__ int tcp_sk_listen_hashfn(struct sock *sk) * 15 is ~13-30min depending on RTO. */ -#define TCP_SYN_RETRIES 5 /* number of times to retry active opening a - * connection: ~180sec is RFC minumum */ - -#define TCP_SYNACK_RETRIES 5 /* number of times to retry passive opening a - * connection: ~180sec is RFC minumum */ - - -#define TCP_ORPHAN_RETRIES 7 /* number of times to retry on an orphaned - * socket. 7 is ~50sec-16min. +#define TCP_SYN_RETRIES 6 /* This is how many retries are done + * when active opening a connection. + * RFC1122 says the minimum retry MUST + * be at least 180secs. Nevertheless + * this value is corresponding to + * 63secs of retransmission with the + * current initial RTO. */ +#define TCP_SYNACK_RETRIES 5 /* This is how may retries are done + * when passive opening a connection. + * This is corresponding to 31secs of + * retransmission with the current + * initial RTO. + */ #define TCP_TIMEWAIT_LEN (60*HZ) /* how long to wait to destroy TIME-WAIT * state, about 60 seconds */ @@ -449,7 +131,13 @@ static __inline__ int tcp_sk_listen_hashfn(struct sock *sk) #endif #define TCP_RTO_MAX ((unsigned)(120*HZ)) #define TCP_RTO_MIN ((unsigned)(HZ/5)) -#define TCP_TIMEOUT_INIT ((unsigned)(3*HZ)) /* RFC 1122 initial RTO value */ +#define TCP_TIMEOUT_INIT ((unsigned)(1*HZ)) /* RFC6298 2.1 initial RTO value */ +#define TCP_TIMEOUT_FALLBACK ((unsigned)(3*HZ)) /* RFC 1122 initial RTO value, now + * used as a fallback RTO for the + * initial data transmission if no + * valid RTT sample has been acquired, + * most likely due to retrans in 3WHS. + */ #define TCP_RESOURCE_PROBE_INTERVAL ((unsigned)(HZ/2U)) /* Maximal interval between probes * for local resources. @@ -465,7 +153,6 @@ static __inline__ int tcp_sk_listen_hashfn(struct sock *sk) #define MAX_TCP_SYNCNT 127 #define TCP_SYNQ_INTERVAL (HZ/5) /* Period of SYNACK timer */ -#define TCP_SYNQ_HSIZE 512 /* Size of SYNACK hash table */ #define TCP_PAWS_24DAYS (60 * 60 * 24 * 24) #define TCP_PAWS_MSL 60 /* Per-host timestamps are invalidated @@ -478,52 +165,6 @@ static __inline__ int tcp_sk_listen_hashfn(struct sock *sk) * 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 - -#define BICTCP_BETA_SCALE 1024 /* Scale factor beta calculation - * max_cwnd = snd_cwnd * beta - */ -#define BICTCP_MAX_INCREMENT 32 /* - * Limit on the amount of - * increment allowed during - * binary search. - */ -#define BICTCP_FUNC_OF_MIN_INCR 11 /* - * log(B/Smin)/log(B/(B-1))+1, - * Smin:min increment - * B:log factor - */ -#define BICTCP_B 4 /* - * In binary search, - * go to point (max+min)/N - */ - /* * TCP option */ @@ -535,6 +176,12 @@ static __inline__ int tcp_sk_listen_hashfn(struct sock *sk) #define TCPOPT_SACK_PERM 4 /* SACK Permitted */ #define TCPOPT_SACK 5 /* SACK Block */ #define TCPOPT_TIMESTAMP 8 /* Better RTT estimations/PAWS */ +#define TCPOPT_MD5SIG 19 /* MD5 Signature (RFC2385) */ +#define TCPOPT_EXP 254 /* Experimental */ +/* Magic number to be after the option value for sharing TCP + * experimental options. See draft-ietf-tcpm-experimental-options-00.txt + */ +#define TCPOPT_FASTOPEN_MAGIC 0xF989 /* * TCP option lengths @@ -544,6 +191,8 @@ static __inline__ int tcp_sk_listen_hashfn(struct sock *sk) #define TCPOLEN_WINDOW 3 #define TCPOLEN_SACK_PERM 2 #define TCPOLEN_TIMESTAMP 10 +#define TCPOLEN_MD5SIG 18 +#define TCPOLEN_EXP_FASTOPEN_BASE 4 /* But this is what stacks really send out. */ #define TCPOLEN_TSTAMP_ALIGNED 12 @@ -552,23 +201,41 @@ static __inline__ int tcp_sk_listen_hashfn(struct sock *sk) #define TCPOLEN_SACK_BASE 2 #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 */ +#define TCPOLEN_MD5SIG_ALIGNED 20 +#define TCPOLEN_MSS_ALIGNED 4 /* 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 */ +#define TCP_NAGLE_PUSH 4 /* Cork is overridden for already queued data */ + +/* TCP thin-stream limits */ +#define TCP_THIN_LINEAR_RETRIES 6 /* After 6 linear retries, do exp. backoff */ + +/* TCP initial congestion window as per draft-hkchu-tcpm-initcwnd-01 */ +#define TCP_INIT_CWND 10 + +/* Bit Flags for sysctl_tcp_fastopen */ +#define TFO_CLIENT_ENABLE 1 +#define TFO_SERVER_ENABLE 2 +#define TFO_CLIENT_NO_COOKIE 4 /* Data in SYN w/o cookie option */ + +/* Accept SYN data w/o any cookie option */ +#define TFO_SERVER_COOKIE_NOT_REQD 0x200 + +/* Force enable TFO on all listeners, i.e., not requiring the + * TCP_FASTOPEN socket option. SOCKOPT1/2 determine how to set max_qlen. + */ +#define TFO_SERVER_WO_SOCKOPT1 0x400 +#define TFO_SERVER_WO_SOCKOPT2 0x800 + +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; @@ -578,17 +245,16 @@ extern int sysctl_tcp_retries1; extern int sysctl_tcp_retries2; extern int sysctl_tcp_orphan_retries; extern int sysctl_tcp_syncookies; +extern int sysctl_tcp_fastopen; extern int sysctl_tcp_retrans_collapse; extern int sysctl_tcp_stdurg; 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 int sysctl_tcp_dsack; -extern int sysctl_tcp_mem[3]; +extern long sysctl_tcp_mem[3]; extern int sysctl_tcp_wmem[3]; extern int sysctl_tcp_rmem[3]; extern int sysctl_tcp_app_win; @@ -596,181 +262,140 @@ extern int sysctl_tcp_adv_win_scale; extern int sysctl_tcp_tw_reuse; extern int sysctl_tcp_frto; extern int sysctl_tcp_low_latency; -extern int sysctl_tcp_westwood; -extern int sysctl_tcp_vegas_cong_avoid; -extern int sysctl_tcp_vegas_alpha; -extern int sysctl_tcp_vegas_beta; -extern int sysctl_tcp_vegas_gamma; +extern int sysctl_tcp_dma_copybreak; extern int sysctl_tcp_nometrics_save; -extern int sysctl_tcp_bic; -extern int sysctl_tcp_bic_fast_convergence; -extern int sysctl_tcp_bic_low_window; -extern int sysctl_tcp_bic_beta; extern int sysctl_tcp_moderate_rcvbuf; extern int sysctl_tcp_tso_win_divisor; - -extern atomic_t tcp_memory_allocated; -extern atomic_t tcp_sockets_allocated; +extern int sysctl_tcp_mtu_probing; +extern int sysctl_tcp_base_mss; +extern int sysctl_tcp_workaround_signed_windows; +extern int sysctl_tcp_slow_start_after_idle; +extern int sysctl_tcp_thin_linear_timeouts; +extern int sysctl_tcp_thin_dupack; +extern int sysctl_tcp_early_retrans; +extern int sysctl_tcp_limit_output_bytes; +extern int sysctl_tcp_challenge_ack_limit; +extern unsigned int sysctl_tcp_notsent_lowat; +extern int sysctl_tcp_min_tso_segs; +extern int sysctl_tcp_autocorking; + +extern atomic_long_t tcp_memory_allocated; +extern struct percpu_counter 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) - */ - -struct tcp_func { - int (*queue_xmit) (struct sk_buff *skb, - int ipfragok); - - void (*send_check) (struct sock *sk, - struct tcphdr *th, - int len, - struct sk_buff *skb); - - int (*rebuild_header) (struct sock *sk); - - int (*conn_request) (struct sock *sk, - struct sk_buff *skb); - - struct sock * (*syn_recv_sock) (struct sock *sk, - struct sk_buff *skb, - struct request_sock *req, - struct dst_entry *dst); - - int (*remember_stamp) (struct sock *sk); - - __u16 net_header_len; - - int (*setsockopt) (struct sock *sk, - int level, - int optname, - char __user *optval, - int optlen); - - int (*getsockopt) (struct sock *sk, - int level, - int optname, - char __user *optval, - int __user *optlen); - - - void (*addr2sockaddr) (struct sock *sk, - struct sockaddr *); - - int sockaddr_len; -}; - /* * The next routines deal with comparing 32 bit unsigned ints * and worry about wraparound (automatic with unsigned arithmetic). */ -static inline int before(__u32 seq1, __u32 seq2) +static inline bool before(__u32 seq1, __u32 seq2) { return (__s32)(seq1-seq2) < 0; } - -static inline int after(__u32 seq1, __u32 seq2) -{ - return (__s32)(seq2-seq1) < 0; -} - +#define after(seq2, seq1) before(seq1, seq2) /* is s2<=s1<=s3 ? */ -static inline int between(__u32 seq1, __u32 seq2, __u32 seq3) +static inline bool between(__u32 seq1, __u32 seq2, __u32 seq3) { return seq3 - seq2 >= seq1 - seq2; } - -extern struct proto tcp_prot; - -DECLARE_SNMP_STAT(struct tcp_mib, tcp_statistics); -#define TCP_INC_STATS(field) SNMP_INC_STATS(tcp_statistics, field) -#define TCP_INC_STATS_BH(field) SNMP_INC_STATS_BH(tcp_statistics, field) -#define TCP_INC_STATS_USER(field) SNMP_INC_STATS_USER(tcp_statistics, field) -#define TCP_DEC_STATS(field) SNMP_DEC_STATS(tcp_statistics, field) -#define TCP_ADD_STATS_BH(field, val) SNMP_ADD_STATS_BH(tcp_statistics, field, val) -#define TCP_ADD_STATS_USER(field, val) SNMP_ADD_STATS_USER(tcp_statistics, field, val) - -extern void tcp_put_port(struct sock *sk); -extern void tcp_inherit_port(struct sock *sk, struct sock *child); - -extern void tcp_v4_err(struct sk_buff *skb, u32); - -extern void tcp_shutdown (struct sock *sk, int how); - -extern int tcp_v4_rcv(struct sk_buff *skb); - -extern int tcp_v4_remember_stamp(struct sock *sk); - -extern int tcp_v4_tw_remember_stamp(struct tcp_tw_bucket *tw); - -extern int tcp_sendmsg(struct kiocb *iocb, struct sock *sk, - struct msghdr *msg, size_t size); -extern ssize_t tcp_sendpage(struct socket *sock, struct page *page, int offset, size_t size, int flags); - -extern int tcp_ioctl(struct sock *sk, - int cmd, - unsigned long arg); - -extern int tcp_rcv_state_process(struct sock *sk, - struct sk_buff *skb, - struct tcphdr *th, - unsigned len); - -extern int tcp_rcv_established(struct sock *sk, - struct sk_buff *skb, - struct tcphdr *th, - unsigned len); - -extern void tcp_rcv_space_adjust(struct sock *sk); - -enum tcp_ack_state_t -{ - TCP_ACK_SCHED = 1, - TCP_ACK_TIMER = 2, - TCP_ACK_PUSHED= 4 -}; - -static inline void tcp_schedule_ack(struct tcp_sock *tp) +static inline bool tcp_out_of_memory(struct sock *sk) { - tp->ack.pending |= TCP_ACK_SCHED; + if (sk->sk_wmem_queued > SOCK_MIN_SNDBUF && + sk_memory_allocated(sk) > sk_prot_mem_limits(sk, 2)) + return true; + return false; } -static inline int tcp_ack_scheduled(struct tcp_sock *tp) +static inline bool tcp_too_many_orphans(struct sock *sk, int shift) { - return tp->ack.pending&TCP_ACK_SCHED; -} + struct percpu_counter *ocp = sk->sk_prot->orphan_count; + int orphans = percpu_counter_read_positive(ocp); -static __inline__ void tcp_dec_quickack_mode(struct tcp_sock *tp) -{ - if (tp->ack.quick && --tp->ack.quick == 0) { - /* Leaving quickack mode we deflate ATO. */ - tp->ack.ato = TCP_ATO_MIN; + if (orphans << shift > sysctl_tcp_max_orphans) { + orphans = percpu_counter_sum_positive(ocp); + if (orphans << shift > sysctl_tcp_max_orphans) + return true; } + return false; } -extern void tcp_enter_quickack_mode(struct tcp_sock *tp); +bool tcp_check_oom(struct sock *sk, int shift); -static __inline__ void tcp_delack_init(struct tcp_sock *tp) +/* syncookies: remember time of last synqueue overflow */ +static inline void tcp_synq_overflow(struct sock *sk) { - memset(&tp->ack, 0, sizeof(tp->ack)); + tcp_sk(sk)->rx_opt.ts_recent_stamp = jiffies; } -static inline void tcp_clear_options(struct tcp_options_received *rx_opt) +/* syncookies: no recent synqueue overflow on this listening socket? */ +static inline bool tcp_synq_no_recent_overflow(const struct sock *sk) { - rx_opt->tstamp_ok = rx_opt->sack_ok = rx_opt->wscale_ok = rx_opt->snd_wscale = 0; + unsigned long last_overflow = tcp_sk(sk)->rx_opt.ts_recent_stamp; + return time_after(jiffies, last_overflow + TCP_TIMEOUT_FALLBACK); } -enum tcp_tw_status -{ +extern struct proto tcp_prot; + +#define TCP_INC_STATS(net, field) SNMP_INC_STATS((net)->mib.tcp_statistics, field) +#define TCP_INC_STATS_BH(net, field) SNMP_INC_STATS_BH((net)->mib.tcp_statistics, field) +#define TCP_DEC_STATS(net, field) SNMP_DEC_STATS((net)->mib.tcp_statistics, field) +#define TCP_ADD_STATS_USER(net, field, val) SNMP_ADD_STATS_USER((net)->mib.tcp_statistics, field, val) +#define TCP_ADD_STATS(net, field, val) SNMP_ADD_STATS((net)->mib.tcp_statistics, field, val) + +void tcp_tasklet_init(void); + +void tcp_v4_err(struct sk_buff *skb, u32); + +void tcp_shutdown(struct sock *sk, int how); + +void tcp_v4_early_demux(struct sk_buff *skb); +int tcp_v4_rcv(struct sk_buff *skb); + +int tcp_v4_tw_remember_stamp(struct inet_timewait_sock *tw); +int tcp_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, + size_t size); +int tcp_sendpage(struct sock *sk, struct page *page, int offset, size_t size, + int flags); +void tcp_release_cb(struct sock *sk); +void tcp_wfree(struct sk_buff *skb); +void tcp_write_timer_handler(struct sock *sk); +void tcp_delack_timer_handler(struct sock *sk); +int tcp_ioctl(struct sock *sk, int cmd, unsigned long arg); +int tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb, + const struct tcphdr *th, unsigned int len); +void tcp_rcv_established(struct sock *sk, struct sk_buff *skb, + const struct tcphdr *th, unsigned int len); +void tcp_rcv_space_adjust(struct sock *sk); +void tcp_cleanup_rbuf(struct sock *sk, int copied); +int tcp_twsk_unique(struct sock *sk, struct sock *sktw, void *twp); +void tcp_twsk_destructor(struct sock *sk); +ssize_t tcp_splice_read(struct socket *sk, loff_t *ppos, + struct pipe_inode_info *pipe, size_t len, + unsigned int flags); + +static inline void tcp_dec_quickack_mode(struct sock *sk, + const unsigned int pkts) +{ + 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. */ + icsk->icsk_ack.ato = TCP_ATO_MIN; + } else + icsk->icsk_ack.quick -= pkts; + } +} + +#define TCP_ECN_OK 1 +#define TCP_ECN_QUEUE_CWR 2 +#define TCP_ECN_DEMAND_CWR 4 +#define TCP_ECN_SEEN 8 + +enum tcp_tw_status { TCP_TW_SUCCESS = 0, TCP_TW_RST = 1, TCP_TW_ACK = 2, @@ -778,249 +403,263 @@ enum tcp_tw_status }; -extern enum tcp_tw_status tcp_timewait_state_process(struct tcp_tw_bucket *tw, - struct sk_buff *skb, - struct tcphdr *th, - unsigned len); - -extern struct sock * tcp_check_req(struct sock *sk,struct sk_buff *skb, - struct request_sock *req, - struct request_sock **prev); -extern int tcp_child_process(struct sock *parent, - struct sock *child, - struct sk_buff *skb); -extern void tcp_enter_frto(struct sock *sk); -extern void tcp_enter_loss(struct sock *sk, int how); -extern void tcp_clear_retrans(struct tcp_sock *tp); -extern void tcp_update_metrics(struct sock *sk); - -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, - int optname, - char __user *optval, - int __user *optlen); -extern int tcp_setsockopt(struct sock *sk, int level, - int optname, char __user *optval, - int optlen); -extern void tcp_set_keepalive(struct sock *sk, int val); -extern int tcp_recvmsg(struct kiocb *iocb, struct sock *sk, - struct msghdr *msg, - 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); +enum tcp_tw_status tcp_timewait_state_process(struct inet_timewait_sock *tw, + struct sk_buff *skb, + const struct tcphdr *th); +struct sock *tcp_check_req(struct sock *sk, struct sk_buff *skb, + struct request_sock *req, struct request_sock **prev, + bool fastopen); +int tcp_child_process(struct sock *parent, struct sock *child, + struct sk_buff *skb); +void tcp_enter_loss(struct sock *sk, int how); +void tcp_clear_retrans(struct tcp_sock *tp); +void tcp_update_metrics(struct sock *sk); +void tcp_init_metrics(struct sock *sk); +void tcp_metrics_init(void); +bool tcp_peer_is_proven(struct request_sock *req, struct dst_entry *dst, + bool paws_check); +bool tcp_remember_stamp(struct sock *sk); +bool tcp_tw_remember_stamp(struct inet_timewait_sock *tw); +void tcp_fetch_timewait_stamp(struct sock *sk, struct dst_entry *dst); +void tcp_disable_fack(struct tcp_sock *tp); +void tcp_close(struct sock *sk, long timeout); +void tcp_init_sock(struct sock *sk); +unsigned int tcp_poll(struct file *file, struct socket *sock, + struct poll_table_struct *wait); +int tcp_getsockopt(struct sock *sk, int level, int optname, + char __user *optval, int __user *optlen); +int tcp_setsockopt(struct sock *sk, int level, int optname, + char __user *optval, unsigned int optlen); +int compat_tcp_getsockopt(struct sock *sk, int level, int optname, + char __user *optval, int __user *optlen); +int compat_tcp_setsockopt(struct sock *sk, int level, int optname, + char __user *optval, unsigned int optlen); +void tcp_set_keepalive(struct sock *sk, int val); +void tcp_syn_ack_timeout(struct sock *sk, struct request_sock *req); +int tcp_recvmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, + size_t len, int nonblock, int flags, int *addr_len); +void tcp_parse_options(const struct sk_buff *skb, + struct tcp_options_received *opt_rx, + int estab, struct tcp_fastopen_cookie *foc); +const u8 *tcp_parse_md5sig_option(const struct tcphdr *th); /* * TCP v4 functions exported for the inet6 API */ -extern int tcp_v4_rebuild_header(struct sock *sk); - -extern int tcp_v4_build_header(struct sock *sk, - struct sk_buff *skb); - -extern void tcp_v4_send_check(struct sock *sk, - struct tcphdr *th, int len, - struct sk_buff *skb); - -extern int tcp_v4_conn_request(struct sock *sk, - struct sk_buff *skb); - -extern struct sock * tcp_create_openreq_child(struct sock *sk, - struct request_sock *req, - struct sk_buff *skb); - -extern struct sock * tcp_v4_syn_recv_sock(struct sock *sk, - struct sk_buff *skb, - struct request_sock *req, - struct dst_entry *dst); - -extern int tcp_v4_do_rcv(struct sock *sk, - struct sk_buff *skb); - -extern int tcp_v4_connect(struct sock *sk, - struct sockaddr *uaddr, - int addr_len); - -extern int tcp_connect(struct sock *sk); +void tcp_v4_send_check(struct sock *sk, struct sk_buff *skb); +int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb); +struct sock *tcp_create_openreq_child(struct sock *sk, + struct request_sock *req, + struct sk_buff *skb); +struct sock *tcp_v4_syn_recv_sock(struct sock *sk, struct sk_buff *skb, + struct request_sock *req, + struct dst_entry *dst); +int tcp_v4_do_rcv(struct sock *sk, struct sk_buff *skb); +int tcp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len); +int tcp_connect(struct sock *sk); +struct sk_buff *tcp_make_synack(struct sock *sk, struct dst_entry *dst, + struct request_sock *req, + struct tcp_fastopen_cookie *foc); +int tcp_disconnect(struct sock *sk, int flags); + +void tcp_finish_connect(struct sock *sk, struct sk_buff *skb); +int tcp_send_rcvq(struct sock *sk, struct msghdr *msg, size_t size); +void inet_sk_rx_dst_set(struct sock *sk, const struct sk_buff *skb); -extern struct sk_buff * tcp_make_synack(struct sock *sk, - struct dst_entry *dst, - struct request_sock *req); - -extern int tcp_disconnect(struct sock *sk, int flags); - -extern void tcp_unhash(struct sock *sk); +/* From syncookies.c */ +int __cookie_v4_check(const struct iphdr *iph, const struct tcphdr *th, + u32 cookie); +struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb, + struct ip_options *opt); +#ifdef CONFIG_SYN_COOKIES + +/* Syncookies use a monotonic timer which increments every 60 seconds. + * This counter is used both as a hash input and partially encoded into + * the cookie value. A cookie is only validated further if the delta + * between the current counter value and the encoded one is less than this, + * i.e. a sent cookie is valid only at most for 2*60 seconds (or less if + * the counter advances immediately after a cookie is generated). + */ +#define MAX_SYNCOOKIE_AGE 2 -extern int tcp_v4_hash_connecting(struct sock *sk); +static inline u32 tcp_cookie_time(void) +{ + u64 val = get_jiffies_64(); + do_div(val, 60 * HZ); + return val; +} -/* From syncookies.c */ -extern struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb, - struct ip_options *opt); -extern __u32 cookie_v4_init_sequence(struct sock *sk, struct sk_buff *skb, - __u16 *mss); +u32 __cookie_v4_init_sequence(const struct iphdr *iph, const struct tcphdr *th, + u16 *mssp); +__u32 cookie_v4_init_sequence(struct sock *sk, struct sk_buff *skb, __u16 *mss); +#else +static inline __u32 cookie_v4_init_sequence(struct sock *sk, + struct sk_buff *skb, + __u16 *mss) +{ + return 0; +} +#endif +__u32 cookie_init_timestamp(struct request_sock *req); +bool cookie_check_timestamp(struct tcp_options_received *opt, struct net *net, + bool *ecn_ok); + +/* From net/ipv6/syncookies.c */ +int __cookie_v6_check(const struct ipv6hdr *iph, const struct tcphdr *th, + u32 cookie); +struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb); +#ifdef CONFIG_SYN_COOKIES +u32 __cookie_v6_init_sequence(const struct ipv6hdr *iph, + const struct tcphdr *th, u16 *mssp); +__u32 cookie_v6_init_sequence(struct sock *sk, const struct sk_buff *skb, + __u16 *mss); +#else +static inline __u32 cookie_v6_init_sequence(struct sock *sk, + struct sk_buff *skb, + __u16 *mss) +{ + return 0; +} +#endif /* tcp_output.c */ -extern int tcp_write_xmit(struct sock *, int nonagle); -extern int tcp_retransmit_skb(struct sock *, struct sk_buff *); -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 void tcp_send_probe0(struct sock *); -extern void tcp_send_partial(struct sock *); -extern int tcp_write_wakeup(struct sock *); -extern void tcp_send_fin(struct sock *sk); -extern void tcp_send_active_reset(struct sock *sk, int priority); -extern int tcp_send_synack(struct sock *); -extern void tcp_push_one(struct sock *, unsigned mss_now); -extern void tcp_send_ack(struct sock *sk); -extern void tcp_send_delayed_ack(struct sock *sk); +void __tcp_push_pending_frames(struct sock *sk, unsigned int cur_mss, + int nonagle); +bool tcp_may_send_now(struct sock *sk); +int __tcp_retransmit_skb(struct sock *, struct sk_buff *); +int tcp_retransmit_skb(struct sock *, struct sk_buff *); +void tcp_retransmit_timer(struct sock *sk); +void tcp_xmit_retransmit_queue(struct sock *); +void tcp_simple_retransmit(struct sock *); +int tcp_trim_head(struct sock *, struct sk_buff *, u32); +int tcp_fragment(struct sock *, struct sk_buff *, u32, unsigned int, gfp_t); + +void tcp_send_probe0(struct sock *); +void tcp_send_partial(struct sock *); +int tcp_write_wakeup(struct sock *); +void tcp_send_fin(struct sock *sk); +void tcp_send_active_reset(struct sock *sk, gfp_t priority); +int tcp_send_synack(struct sock *); +bool tcp_syn_flood_action(struct sock *sk, const struct sk_buff *skb, + const char *proto); +void tcp_push_one(struct sock *, unsigned int mss_now); +void tcp_send_ack(struct sock *sk); +void tcp_send_delayed_ack(struct sock *sk); +void tcp_send_loss_probe(struct sock *sk); +bool tcp_schedule_loss_probe(struct sock *sk); + +/* tcp_input.c */ +void tcp_resume_early_retransmit(struct sock *sk); +void tcp_rearm_rto(struct sock *sk); +void tcp_reset(struct sock *sk); /* tcp_timer.c */ -extern void tcp_init_xmit_timers(struct sock *); -extern void tcp_clear_xmit_timers(struct sock *); +void tcp_init_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); +unsigned int tcp_sync_mss(struct sock *sk, u32 pmtu); +unsigned int tcp_current_mss(struct sock *sk); -#ifdef TCP_DEBUG -extern const char tcp_timer_bug_msg[]; -#endif +/* Bound MSS / TSO packet size with the half of the window */ +static inline int tcp_bound_to_half_wnd(struct tcp_sock *tp, int pktsize) +{ + int cutoff; -/* tcp_diag.c */ -extern void tcp_get_info(struct sock *, struct tcp_info *); + /* When peer uses tiny windows, there is no use in packetizing + * to sub-MSS pieces for the sake of SWS or making sure there + * are enough packets in the pipe for fast recovery. + * + * On the other hand, for extremely large MSS devices, handling + * smaller than MSS windows in this way does make sense. + */ + if (tp->max_window >= 512) + cutoff = (tp->max_window >> 1); + else + cutoff = tp->max_window; + + if (cutoff && pktsize > cutoff) + return max_t(int, cutoff, 68U - tp->tcp_header_len); + else + return pktsize; +} + +/* tcp.c */ +void tcp_get_info(const struct sock *, struct tcp_info *); /* Read 'sendfile()'-style from a TCP socket */ typedef int (*sk_read_actor_t)(read_descriptor_t *, struct sk_buff *, unsigned int, size_t); -extern int tcp_read_sock(struct sock *sk, read_descriptor_t *desc, - sk_read_actor_t recv_actor); +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; +void tcp_initialize_rcv_mss(struct sock *sk); -#ifdef TCP_CLEAR_TIMERS - sk_stop_timer(sk, &tp->delack_timer); -#endif - break; - default: -#ifdef TCP_DEBUG - printk(tcp_timer_bug_msg); -#endif - return; - }; +int tcp_mtu_to_mss(struct sock *sk, int pmtu); +int tcp_mss_to_mtu(struct sock *sk, int mss); +void tcp_mtup_init(struct sock *sk); +void tcp_init_buffer_space(struct sock *sk); -} - -/* - * Reset the retransmission timer - */ -static inline void tcp_reset_xmit_timer(struct sock *sk, int what, unsigned long when) +static inline void tcp_bound_rto(const struct sock *sk) { - 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; - }; + if (inet_csk(sk)->icsk_rto > TCP_RTO_MAX) + inet_csk(sk)->icsk_rto = TCP_RTO_MAX; } -/* 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. - * It's better to underestimate the RCV_MSS rather than overestimate. - * Overestimations make us ACKing less frequently than needed. - * Underestimations are more easy to detect and fix by tcp_measure_rcv_mss(). - */ - -static inline void tcp_initialize_rcv_mss(struct sock *sk) +static inline u32 __tcp_set_rto(const struct tcp_sock *tp) { - struct tcp_sock *tp = tcp_sk(sk); - unsigned int hint = min(tp->advmss, tp->mss_cache_std); - - hint = min(hint, tp->rcv_wnd/2); - hint = min(hint, TCP_MIN_RCVMSS); - hint = max(hint, TCP_MIN_MSS); - - tp->ack.rcv_mss = hint; + return usecs_to_jiffies((tp->srtt_us >> 3) + tp->rttvar_us); } -static __inline__ void __tcp_fast_path_on(struct tcp_sock *tp, u32 snd_wnd) +static inline void __tcp_fast_path_on(struct tcp_sock *tp, u32 snd_wnd) { tp->pred_flags = htonl((tp->tcp_header_len << 26) | ntohl(TCP_FLAG_ACK) | snd_wnd); } -static __inline__ void tcp_fast_path_on(struct tcp_sock *tp) +static inline void tcp_fast_path_on(struct tcp_sock *tp) { __tcp_fast_path_on(tp, tp->snd_wnd >> tp->rx_opt.snd_wscale); } -static inline void tcp_fast_path_check(struct sock *sk, struct tcp_sock *tp) +static inline void tcp_fast_path_check(struct sock *sk) { - if (skb_queue_len(&tp->out_of_order_queue) == 0 && + struct tcp_sock *tp = tcp_sk(sk); + + if (skb_queue_empty(&tp->out_of_order_queue) && tp->rcv_wnd && atomic_read(&sk->sk_rmem_alloc) < sk->sk_rcvbuf && !tp->urg_data) tcp_fast_path_on(tp); } +/* Compute the actual rto_min value */ +static inline u32 tcp_rto_min(struct sock *sk) +{ + const struct dst_entry *dst = __sk_dst_get(sk); + u32 rto_min = TCP_RTO_MIN; + + if (dst && dst_metric_locked(dst, RTAX_RTO_MIN)) + rto_min = dst_metric_rtt(dst, RTAX_RTO_MIN); + return rto_min; +} + +static inline u32 tcp_rto_min_us(struct sock *sk) +{ + return jiffies_to_usecs(tcp_rto_min(sk)); +} + /* Compute the actual receive window we are currently advertising. * Rcv_nxt can be after the window if our peer push more data * than the offered window. */ -static __inline__ u32 tcp_receive_window(const struct tcp_sock *tp) +static inline u32 tcp_receive_window(const struct tcp_sock *tp) { s32 win = tp->rcv_wup + tp->rcv_wnd - tp->rcv_nxt; @@ -1033,107 +672,222 @@ static __inline__ u32 tcp_receive_window(const struct tcp_sock *tp) * scaling applied to the result. The caller does these things * if necessary. This is a "raw" window selection. */ -extern u32 __tcp_select_window(struct sock *sk); +u32 __tcp_select_window(struct sock *sk); + +void tcp_send_window_probe(struct sock *sk); /* TCP timestamps are only 32-bits, this causes a slight * complication on 64-bit systems since we store a snapshot - * of jiffies in the buffer control blocks below. We decidely - * only use of the low 32-bits of jiffies and hide the ugly + * of jiffies in the buffer control blocks below. We decided + * to use only the low 32-bits of jiffies and hide the ugly * casts with the following macro. */ #define tcp_time_stamp ((__u32)(jiffies)) -/* This is what the send packet queueing engine uses to pass - * TCP per-packet control information to the transmission - * code. We also store the host-order sequence numbers in - * here too. This is 36 bytes on 32-bit architectures, - * 40 bytes on 64-bit machines, if this grows please adjust - * skbuff.h:skbuff->cb[xxx] size appropriately. +#define tcp_flag_byte(th) (((u_int8_t *)th)[13]) + +#define TCPHDR_FIN 0x01 +#define TCPHDR_SYN 0x02 +#define TCPHDR_RST 0x04 +#define TCPHDR_PSH 0x08 +#define TCPHDR_ACK 0x10 +#define TCPHDR_URG 0x20 +#define TCPHDR_ECE 0x40 +#define TCPHDR_CWR 0x80 + +/* This is what the send packet queuing engine uses to pass + * TCP per-packet control information to the transmission code. + * We also store the host-order sequence numbers in here too. + * This is 44 bytes if IPV6 is enabled. + * If this grows please adjust skbuff.h:skbuff->cb[xxx] size appropriately. */ struct tcp_skb_cb { union { struct inet_skb_parm h4; -#if defined(CONFIG_IPV6) || defined (CONFIG_IPV6_MODULE) +#if IS_ENABLED(CONFIG_IPV6) struct inet6_skb_parm h6; #endif } header; /* For incoming frames */ __u32 seq; /* Starting sequence number */ __u32 end_seq; /* SEQ + FIN + SYN + datalen */ __u32 when; /* used to compute rtt's */ - __u8 flags; /* TCP header flags. */ - - /* NOTE: These must match up to the flags byte in a - * real TCP header. - */ -#define TCPCB_FLAG_FIN 0x01 -#define TCPCB_FLAG_SYN 0x02 -#define TCPCB_FLAG_RST 0x04 -#define TCPCB_FLAG_PSH 0x08 -#define TCPCB_FLAG_ACK 0x10 -#define TCPCB_FLAG_URG 0x20 -#define TCPCB_FLAG_ECE 0x40 -#define TCPCB_FLAG_CWR 0x80 + __u8 tcp_flags; /* TCP header flags. (tcp[13]) */ __u8 sacked; /* State flags for SACK/FACK. */ #define TCPCB_SACKED_ACKED 0x01 /* SKB ACK'd by a SACK block */ #define TCPCB_SACKED_RETRANS 0x02 /* SKB retransmitted */ #define TCPCB_LOST 0x04 /* SKB is lost */ #define TCPCB_TAGBITS 0x07 /* All tag bits */ - #define TCPCB_EVER_RETRANS 0x80 /* Ever retransmitted frame */ #define TCPCB_RETRANS (TCPCB_SACKED_RETRANS|TCPCB_EVER_RETRANS) -#define TCPCB_URG 0x20 /* Urgent pointer advenced here */ - -#define TCPCB_AT_TAIL (TCPCB_URG) - - __u16 urg_ptr; /* Valid w/URG flags is set. */ + __u8 ip_dsfield; /* IPv4 tos or IPv6 dsfield */ + /* 1 byte hole */ __u32 ack_seq; /* Sequence number ACK'd */ }; #define TCP_SKB_CB(__skb) ((struct tcp_skb_cb *)&((__skb)->cb[0])) -#include <net/tcp_ecn.h> +/* RFC3168 : 6.1.1 SYN packets must not have ECT/ECN bits set + * + * If we receive a SYN packet with these bits set, it means a network is + * playing bad games with TOS bits. In order to avoid possible false congestion + * notifications, we disable TCP ECN negociation. + */ +static inline void +TCP_ECN_create_request(struct request_sock *req, const struct sk_buff *skb, + struct net *net) +{ + const struct tcphdr *th = tcp_hdr(skb); + + if (net->ipv4.sysctl_tcp_ecn && th->ece && th->cwr && + INET_ECN_is_not_ect(TCP_SKB_CB(skb)->ip_dsfield)) + inet_rsk(req)->ecn_ok = 1; +} /* Due to TSO, an SKB can be composed of multiple actual * packets. To keep these tracked properly, we use this. */ static inline int tcp_skb_pcount(const struct sk_buff *skb) { - return skb_shinfo(skb)->tso_segs; + return skb_shinfo(skb)->gso_segs; } /* This is valid iff tcp_skb_pcount() > 1. */ static inline int tcp_skb_mss(const struct sk_buff *skb) { - return skb_shinfo(skb)->tso_size; + return skb_shinfo(skb)->gso_size; } -static inline void tcp_dec_pcount_approx(__u32 *count, - const struct sk_buff *skb) +/* Events passed to congestion control interface */ +enum tcp_ca_event { + CA_EVENT_TX_START, /* first transmit when no packets in flight */ + CA_EVENT_CWND_RESTART, /* congestion window restart */ + CA_EVENT_COMPLETE_CWR, /* end of congestion recovery */ + CA_EVENT_LOSS, /* loss timeout */ + CA_EVENT_FAST_ACK, /* in sequence ack */ + CA_EVENT_SLOW_ACK, /* other ack */ +}; + +/* + * Interface for adding new TCP congestion control handlers + */ +#define TCP_CA_NAME_MAX 16 +#define TCP_CA_MAX 128 +#define TCP_CA_BUF_MAX (TCP_CA_NAME_MAX*TCP_CA_MAX) + +#define TCP_CONG_NON_RESTRICTED 0x1 + +struct tcp_congestion_ops { + struct list_head list; + unsigned long flags; + + /* initialize private data (optional) */ + void (*init)(struct sock *sk); + /* cleanup private data (optional) */ + void (*release)(struct sock *sk); + + /* return slow start threshold (required) */ + u32 (*ssthresh)(struct sock *sk); + /* do new cwnd calculation (required) */ + void (*cong_avoid)(struct sock *sk, u32 ack, u32 acked); + /* call before changing ca_state (optional) */ + void (*set_state)(struct sock *sk, u8 new_state); + /* call when cwnd event occurs (optional) */ + void (*cwnd_event)(struct sock *sk, enum tcp_ca_event ev); + /* new value of cwnd after loss (optional) */ + u32 (*undo_cwnd)(struct sock *sk); + /* hook for packet ack accounting (optional) */ + void (*pkts_acked)(struct sock *sk, u32 num_acked, s32 rtt_us); + /* 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; +}; + +int tcp_register_congestion_control(struct tcp_congestion_ops *type); +void tcp_unregister_congestion_control(struct tcp_congestion_ops *type); + +void tcp_init_congestion_control(struct sock *sk); +void tcp_cleanup_congestion_control(struct sock *sk); +int tcp_set_default_congestion_control(const char *name); +void tcp_get_default_congestion_control(char *name); +void tcp_get_available_congestion_control(char *buf, size_t len); +void tcp_get_allowed_congestion_control(char *buf, size_t len); +int tcp_set_allowed_congestion_control(char *allowed); +int tcp_set_congestion_control(struct sock *sk, const char *name); +int tcp_slow_start(struct tcp_sock *tp, u32 acked); +void tcp_cong_avoid_ai(struct tcp_sock *tp, u32 w); + +extern struct tcp_congestion_ops tcp_init_congestion_ops; +u32 tcp_reno_ssthresh(struct sock *sk); +void tcp_reno_cong_avoid(struct sock *sk, u32 ack, u32 acked); +extern struct tcp_congestion_ops tcp_reno; + +static inline void tcp_set_ca_state(struct sock *sk, const u8 ca_state) { - if (*count) { - *count -= tcp_skb_pcount(skb); - if ((int)*count < 0) - *count = 0; - } + 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 sock *sk, const enum tcp_ca_event 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); } -static inline void tcp_packets_out_inc(struct sock *sk, - struct tcp_sock *tp, - const struct sk_buff *skb) +/* These functions determine how the current flow behaves in respect of SACK + * handling. SACK is negotiated with the peer, and therefore it can vary + * between different flows. + * + * tcp_is_sack - SACK enabled + * tcp_is_reno - No SACK + * tcp_is_fack - FACK enabled, implies SACK enabled + */ +static inline int tcp_is_sack(const struct tcp_sock *tp) +{ + return tp->rx_opt.sack_ok; +} + +static inline bool tcp_is_reno(const struct tcp_sock *tp) { - int orig = tp->packets_out; + return !tcp_is_sack(tp); +} + +static inline bool tcp_is_fack(const struct tcp_sock *tp) +{ + return tp->rx_opt.sack_ok & TCP_FACK_ENABLED; +} - tp->packets_out += tcp_skb_pcount(skb); - if (!orig) - tcp_reset_xmit_timer(sk, TCP_TIME_RETRANS, tp->rto); +static inline void tcp_enable_fack(struct tcp_sock *tp) +{ + tp->rx_opt.sack_ok |= TCP_FACK_ENABLED; +} + +/* TCP early-retransmit (ER) is similar to but more conservative than + * the thin-dupack feature. Enable ER only if thin-dupack is disabled. + */ +static inline void tcp_enable_early_retrans(struct tcp_sock *tp) +{ + tp->do_early_retrans = sysctl_tcp_early_retrans && + sysctl_tcp_early_retrans < 4 && !sysctl_tcp_thin_dupack && + sysctl_tcp_reordering == 3; +} + +static inline void tcp_disable_early_retrans(struct tcp_sock *tp) +{ + tp->do_early_retrans = 0; } -static inline void tcp_packets_out_dec(struct tcp_sock *tp, - const struct sk_buff *skb) +static inline unsigned int tcp_left_out(const struct tcp_sock *tp) { - tp->packets_out -= tcp_skb_pcount(skb); + return tp->sacked_out + tp->lost_out; } /* This determines how many packets are "in the network" to the best @@ -1150,103 +904,33 @@ static inline void tcp_packets_out_dec(struct tcp_sock *tp, * "Packets left network, but not honestly ACKed yet" PLUS * "Packets fast retransmitted" */ -static __inline__ unsigned int tcp_packets_in_flight(const struct tcp_sock *tp) +static inline unsigned int tcp_packets_in_flight(const struct tcp_sock *tp) { - return (tp->packets_out - tp->left_out + tp->retrans_out); + return tp->packets_out - tcp_left_out(tp) + tp->retrans_out; } -/* - * Which congestion algorithim is in use on the connection. - */ -#define tcp_is_vegas(__tp) ((__tp)->adv_cong == TCP_VEGAS) -#define tcp_is_westwood(__tp) ((__tp)->adv_cong == TCP_WESTWOOD) -#define tcp_is_bic(__tp) ((__tp)->adv_cong == TCP_BIC) - -/* Recalculate snd_ssthresh, we want to set it to: - * - * Reno: - * one half the current congestion window, but no - * less than two segments - * - * BIC: - * behave like Reno until low_window is reached, - * then increase congestion window slowly - */ -static inline __u32 tcp_recalc_ssthresh(struct tcp_sock *tp) -{ - if (tcp_is_bic(tp)) { - if (sysctl_tcp_bic_fast_convergence && - tp->snd_cwnd < tp->bictcp.last_max_cwnd) - tp->bictcp.last_max_cwnd = (tp->snd_cwnd * - (BICTCP_BETA_SCALE - + sysctl_tcp_bic_beta)) - / (2 * BICTCP_BETA_SCALE); - else - tp->bictcp.last_max_cwnd = tp->snd_cwnd; - - if (tp->snd_cwnd > sysctl_tcp_bic_low_window) - return max((tp->snd_cwnd * sysctl_tcp_bic_beta) - / BICTCP_BETA_SCALE, 2U); - } - - return max(tp->snd_cwnd >> 1U, 2U); -} +#define TCP_INFINITE_SSTHRESH 0x7fffffff -/* Stop taking Vegas samples for now. */ -#define tcp_vegas_disable(__tp) ((__tp)->vegas.doing_vegas_now = 0) - -static inline void tcp_vegas_enable(struct tcp_sock *tp) +static inline bool tcp_in_initial_slowstart(const struct tcp_sock *tp) { - /* There are several situations when we must "re-start" Vegas: - * - * o when a connection is established - * o after an RTO - * o after fast recovery - * o when we send a packet and there is no outstanding - * unacknowledged data (restarting an idle connection) - * - * In these circumstances we cannot do a Vegas calculation at the - * end of the first RTT, because any calculation we do is using - * stale info -- both the saved cwnd and congestion feedback are - * stale. - * - * Instead we must wait until the completion of an RTT during - * which we actually receive ACKs. - */ - - /* Begin taking Vegas samples next time we send something. */ - tp->vegas.doing_vegas_now = 1; - - /* Set the beginning of the next send window. */ - tp->vegas.beg_snd_nxt = tp->snd_nxt; - - tp->vegas.cntRTT = 0; - tp->vegas.minRTT = 0x7fffffff; + return tp->snd_ssthresh >= TCP_INFINITE_SSTHRESH; } -/* Should we be taking Vegas samples right now? */ -#define tcp_vegas_enabled(__tp) ((__tp)->vegas.doing_vegas_now) - -extern void tcp_ca_init(struct tcp_sock *tp); - -static inline void tcp_set_ca_state(struct tcp_sock *tp, u8 ca_state) +static inline bool tcp_in_cwnd_reduction(const struct sock *sk) { - if (tcp_is_vegas(tp)) { - if (ca_state == TCP_CA_Open) - tcp_vegas_enable(tp); - else - tcp_vegas_disable(tp); - } - tp->ca_state = ca_state; + return (TCPF_CA_CWR | TCPF_CA_Recovery) & + (1 << inet_csk(sk)->icsk_ca_state); } /* If cwnd > ssthresh, we may raise ssthresh to be half-way to cwnd. - * The exception is rate halving phase, when cwnd is decreasing towards + * The exception is cwnd reduction 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 (tcp_in_cwnd_reduction(sk)) return tp->snd_ssthresh; else return max(tp->snd_ssthresh, @@ -1254,599 +938,592 @@ static inline __u32 tcp_current_ssthresh(struct tcp_sock *tp) (tp->snd_cwnd >> 2))); } -static inline void tcp_sync_left_out(struct tcp_sock *tp) +/* Use define here intentionally to get WARN_ON location shown at the caller */ +#define tcp_verify_left_out(tp) WARN_ON(tcp_left_out(tp) > tp->packets_out) + +void tcp_enter_cwr(struct sock *sk, const int set_ssthresh); +__u32 tcp_init_cwnd(const struct tcp_sock *tp, const struct dst_entry *dst); + +/* The maximum number of MSS of available cwnd for which TSO defers + * sending if not using sysctl_tcp_tso_win_divisor. + */ +static inline __u32 tcp_max_tso_deferred_mss(const struct tcp_sock *tp) { - if (tp->rx_opt.sack_ok && - (tp->sacked_out >= tp->packets_out - tp->lost_out)) - tp->sacked_out = tp->packets_out - tp->lost_out; - tp->left_out = tp->sacked_out + tp->lost_out; + return 3; } -extern void tcp_cwnd_application_limited(struct sock *sk); +/* Slow start with delack produces 3 packets of burst, so that + * it is safe "de facto". This will be the default - same as + * the default reordering threshold - but if reordering increases, + * we must be able to allow cwnd to burst at least this much in order + * to not pull it back when holes are filled. + */ +static __inline__ __u32 tcp_max_burst(const struct tcp_sock *tp) +{ + return tp->reordering; +} -/* Congestion window validation. (RFC2861) */ +/* Returns end sequence number of the receiver's advertised window */ +static inline u32 tcp_wnd_end(const struct tcp_sock *tp) +{ + return tp->snd_una + tp->snd_wnd; +} -static inline void tcp_cwnd_validate(struct sock *sk, struct tcp_sock *tp) +/* We follow the spirit of RFC2861 to validate cwnd but implement a more + * flexible approach. The RFC suggests cwnd should not be raised unless + * it was fully used previously. And that's exactly what we do in + * congestion avoidance mode. But in slow start we allow cwnd to grow + * as long as the application has used half the cwnd. + * Example : + * cwnd is 10 (IW10), but application sends 9 frames. + * We allow cwnd to reach 18 when all frames are ACKed. + * This check is safe because it's as aggressive as slow start which already + * risks 100% overshoot. The advantage is that we discourage application to + * either send more filler packets or data to artificially blow up the cwnd + * usage, and allow application-limited process to probe bw more aggressively. + */ +static inline bool tcp_is_cwnd_limited(const struct sock *sk) { - __u32 packets_out = tp->packets_out; + const struct tcp_sock *tp = tcp_sk(sk); - if (packets_out >= tp->snd_cwnd) { - /* Network is feed fully. */ - tp->snd_cwnd_used = 0; - tp->snd_cwnd_stamp = tcp_time_stamp; - } else { - /* Network starves. */ - if (tp->packets_out > tp->snd_cwnd_used) - tp->snd_cwnd_used = tp->packets_out; + /* If in slow start, ensure cwnd grows to twice what was ACKed. */ + if (tp->snd_cwnd <= tp->snd_ssthresh) + return tp->snd_cwnd < 2 * tp->max_packets_out; - if ((s32)(tcp_time_stamp - tp->snd_cwnd_stamp) >= tp->rto) - tcp_cwnd_application_limited(sk); - } + return tp->is_cwnd_limited; } -/* Set slow start threshould and cwnd not falling to slow start */ -static inline void __tcp_enter_cwr(struct tcp_sock *tp) +static inline void tcp_check_probe_timer(struct sock *sk) { - tp->undo_marker = 0; - tp->snd_ssthresh = tcp_recalc_ssthresh(tp); - tp->snd_cwnd = min(tp->snd_cwnd, - tcp_packets_in_flight(tp) + 1U); - tp->snd_cwnd_cnt = 0; - tp->high_seq = tp->snd_nxt; - tp->snd_cwnd_stamp = tcp_time_stamp; - TCP_ECN_queue_cwr(tp); + const struct tcp_sock *tp = tcp_sk(sk); + 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_enter_cwr(struct tcp_sock *tp) +static inline void tcp_init_wl(struct tcp_sock *tp, u32 seq) { - tp->prior_ssthresh = 0; - if (tp->ca_state < TCP_CA_CWR) { - __tcp_enter_cwr(tp); - tcp_set_ca_state(tp, TCP_CA_CWR); - } + tp->snd_wl1 = seq; } -extern __u32 tcp_init_cwnd(struct tcp_sock *tp, struct dst_entry *dst); +static inline void tcp_update_wl(struct tcp_sock *tp, u32 seq) +{ + tp->snd_wl1 = seq; +} -/* Slow start with delack produces 3 packets of burst, so that - * it is safe "de facto". +/* + * Calculate(/check) TCP checksum */ -static __inline__ __u32 tcp_max_burst(const struct tcp_sock *tp) +static inline __sum16 tcp_v4_check(int len, __be32 saddr, + __be32 daddr, __wsum base) { - return 3; + return csum_tcpudp_magic(saddr,daddr,len,IPPROTO_TCP,base); } -static __inline__ int tcp_minshall_check(const struct tcp_sock *tp) +static inline __sum16 __tcp_checksum_complete(struct sk_buff *skb) { - return after(tp->snd_sml,tp->snd_una) && - !after(tp->snd_sml, tp->snd_nxt); + return __skb_checksum_complete(skb); } -static __inline__ void tcp_minshall_update(struct tcp_sock *tp, int mss, - const struct sk_buff *skb) +static inline bool tcp_checksum_complete(struct sk_buff *skb) { - if (skb->len < mss) - tp->snd_sml = TCP_SKB_CB(skb)->end_seq; + return !skb_csum_unnecessary(skb) && + __tcp_checksum_complete(skb); } -/* Return 0, if packet can be sent now without violation Nagle's rules: - 1. It is full sized. - 2. Or it contains FIN. - 3. Or TCP_NODELAY was set. - 4. Or TCP_CORK is not set, and all sent packets are ACKed. - With Minshall's modification: all sent small packets are ACKed. - */ +/* Prequeue for VJ style copy to user, combined with checksumming. */ -static __inline__ int -tcp_nagle_check(const struct tcp_sock *tp, const struct sk_buff *skb, - unsigned mss_now, int nonagle) +static inline void tcp_prequeue_init(struct tcp_sock *tp) { - return (skb->len < mss_now && - !(TCP_SKB_CB(skb)->flags & TCPCB_FLAG_FIN) && - ((nonagle&TCP_NAGLE_CORK) || - (!nonagle && - tp->packets_out && - tcp_minshall_check(tp)))); + tp->ucopy.task = NULL; + tp->ucopy.len = 0; + tp->ucopy.memory = 0; + skb_queue_head_init(&tp->ucopy.prequeue); +#ifdef CONFIG_NET_DMA + tp->ucopy.dma_chan = NULL; + tp->ucopy.wakeup = 0; + tp->ucopy.pinned_list = NULL; + tp->ucopy.dma_cookie = 0; +#endif } -extern void tcp_set_skb_tso_segs(struct sock *, struct sk_buff *); +bool tcp_prequeue(struct sock *sk, struct sk_buff *skb); -/* This checks if the data bearing packet SKB (usually sk->sk_send_head) - * should be put on the wire right now. - */ -static __inline__ int tcp_snd_test(struct sock *sk, - struct sk_buff *skb, - unsigned cur_mss, int nonagle) -{ - struct tcp_sock *tp = tcp_sk(sk); - int pkts = tcp_skb_pcount(skb); +#undef STATE_TRACE - if (!pkts) { - tcp_set_skb_tso_segs(sk, skb); - pkts = tcp_skb_pcount(skb); - } +#ifdef STATE_TRACE +static const char *statename[]={ + "Unused","Established","Syn Sent","Syn Recv", + "Fin Wait 1","Fin Wait 2","Time Wait", "Close", + "Close Wait","Last ACK","Listen","Closing" +}; +#endif +void tcp_set_state(struct sock *sk, int state); - /* RFC 1122 - section 4.2.3.4 - * - * We must queue if - * - * a) The right edge of this frame exceeds the window - * b) There are packets in flight and we have a small segment - * [SWS avoidance and Nagle algorithm] - * (part of SWS is done on packetization) - * Minshall version sounds: there are no _small_ - * segments in flight. (tcp_nagle_check) - * c) We have too many packets 'in flight' - * - * Don't use the nagle rule for urgent data (or - * for the final FIN -DaveM). - * - * Also, Nagle rule does not apply to frames, which - * sit in the middle of queue (they have no chances - * to get new data) and if room at tail of skb is - * not enough to save something seriously (<32 for now). - */ +void tcp_done(struct sock *sk); - /* Don't be strict about the congestion window for the - * final FIN frame. -DaveM - */ - return (((nonagle&TCP_NAGLE_PUSH) || tp->urg_mode - || !tcp_nagle_check(tp, skb, cur_mss, nonagle)) && - (((tcp_packets_in_flight(tp) + (pkts-1)) < tp->snd_cwnd) || - (TCP_SKB_CB(skb)->flags & TCPCB_FLAG_FIN)) && - !after(TCP_SKB_CB(skb)->end_seq, tp->snd_una + tp->snd_wnd)); +static inline void tcp_sack_reset(struct tcp_options_received *rx_opt) +{ + rx_opt->dsack = 0; + rx_opt->num_sacks = 0; } -static __inline__ void tcp_check_probe_timer(struct sock *sk, struct tcp_sock *tp) +u32 tcp_default_init_rwnd(u32 mss); + +/* Determine a window scaling and initial window to offer. */ +void tcp_select_initial_window(int __space, __u32 mss, __u32 *rcv_wnd, + __u32 *window_clamp, int wscale_ok, + __u8 *rcv_wscale, __u32 init_rcv_wnd); + +static inline int tcp_win_from_space(int space) { - if (!tp->packets_out && !tp->pending) - tcp_reset_xmit_timer(sk, TCP_TIME_PROBE0, tp->rto); + return sysctl_tcp_adv_win_scale<=0 ? + (space>>(-sysctl_tcp_adv_win_scale)) : + space - (space>>sysctl_tcp_adv_win_scale); } -static __inline__ int tcp_skb_is_last(const struct sock *sk, - const struct sk_buff *skb) +/* Note: caller must be prepared to deal with negative returns */ +static inline int tcp_space(const struct sock *sk) { - return skb->next == (struct sk_buff *)&sk->sk_write_queue; -} + return tcp_win_from_space(sk->sk_rcvbuf - + atomic_read(&sk->sk_rmem_alloc)); +} -/* Push out any pending frames which were held back due to - * TCP_CORK or attempt at coalescing tiny packets. - * The socket must be locked by the caller. - */ -static __inline__ void __tcp_push_pending_frames(struct sock *sk, - struct tcp_sock *tp, - unsigned cur_mss, - int nonagle) -{ - struct sk_buff *skb = sk->sk_send_head; - - if (skb) { - if (!tcp_skb_is_last(sk, skb)) - nonagle = TCP_NAGLE_PUSH; - if (!tcp_snd_test(sk, skb, cur_mss, nonagle) || - tcp_write_xmit(sk, nonagle)) - tcp_check_probe_timer(sk, tp); - } - tcp_cwnd_validate(sk, tp); +static inline int tcp_full_space(const struct sock *sk) +{ + return tcp_win_from_space(sk->sk_rcvbuf); } -static __inline__ void tcp_push_pending_frames(struct sock *sk, - struct tcp_sock *tp) +static inline void tcp_openreq_init(struct request_sock *req, + struct tcp_options_received *rx_opt, + struct sk_buff *skb) { - __tcp_push_pending_frames(sk, tp, tcp_current_mss(sk, 1), tp->nonagle); + struct inet_request_sock *ireq = inet_rsk(req); + + req->rcv_wnd = 0; /* So that tcp_send_synack() knows! */ + req->cookie_ts = 0; + tcp_rsk(req)->rcv_isn = TCP_SKB_CB(skb)->seq; + tcp_rsk(req)->rcv_nxt = TCP_SKB_CB(skb)->seq + 1; + tcp_rsk(req)->snt_synack = 0; + req->mss = rx_opt->mss_clamp; + req->ts_recent = rx_opt->saw_tstamp ? rx_opt->rcv_tsval : 0; + ireq->tstamp_ok = rx_opt->tstamp_ok; + ireq->sack_ok = rx_opt->sack_ok; + ireq->snd_wscale = rx_opt->snd_wscale; + ireq->wscale_ok = rx_opt->wscale_ok; + ireq->acked = 0; + ireq->ecn_ok = 0; + ireq->ir_rmt_port = tcp_hdr(skb)->source; + ireq->ir_num = ntohs(tcp_hdr(skb)->dest); } -static __inline__ int tcp_may_send_now(struct sock *sk, struct tcp_sock *tp) -{ - struct sk_buff *skb = sk->sk_send_head; +extern void tcp_openreq_init_rwin(struct request_sock *req, + struct sock *sk, struct dst_entry *dst); + +void tcp_enter_memory_pressure(struct sock *sk); - return (skb && - tcp_snd_test(sk, skb, tcp_current_mss(sk, 1), - tcp_skb_is_last(sk, skb) ? TCP_NAGLE_PUSH : tp->nonagle)); +static inline int keepalive_intvl_when(const struct tcp_sock *tp) +{ + return tp->keepalive_intvl ? : sysctl_tcp_keepalive_intvl; } -static __inline__ void tcp_init_wl(struct tcp_sock *tp, u32 ack, u32 seq) +static inline int keepalive_time_when(const struct tcp_sock *tp) { - tp->snd_wl1 = seq; + return tp->keepalive_time ? : sysctl_tcp_keepalive_time; } -static __inline__ void tcp_update_wl(struct tcp_sock *tp, u32 ack, u32 seq) +static inline int keepalive_probes(const struct tcp_sock *tp) { - tp->snd_wl1 = seq; + return tp->keepalive_probes ? : sysctl_tcp_keepalive_probes; } -extern void tcp_destroy_sock(struct sock *sk); +static inline u32 keepalive_time_elapsed(const struct tcp_sock *tp) +{ + const struct inet_connection_sock *icsk = &tp->inet_conn; + return min_t(u32, tcp_time_stamp - icsk->icsk_ack.lrcvtime, + tcp_time_stamp - tp->rcv_tstamp); +} -/* - * Calculate(/check) TCP checksum - */ -static __inline__ u16 tcp_v4_check(struct tcphdr *th, int len, - unsigned long saddr, unsigned long daddr, - unsigned long base) +static inline int tcp_fin_time(const struct sock *sk) { - return csum_tcpudp_magic(saddr,daddr,len,IPPROTO_TCP,base); + int fin_timeout = tcp_sk(sk)->linger2 ? : sysctl_tcp_fin_timeout; + const int rto = inet_csk(sk)->icsk_rto; + + if (fin_timeout < (rto << 2) - (rto >> 1)) + fin_timeout = (rto << 2) - (rto >> 1); + + return fin_timeout; } -static __inline__ int __tcp_checksum_complete(struct sk_buff *skb) +static inline bool tcp_paws_check(const struct tcp_options_received *rx_opt, + int paws_win) { - return (unsigned short)csum_fold(skb_checksum(skb, 0, skb->len, skb->csum)); + if ((s32)(rx_opt->ts_recent - rx_opt->rcv_tsval) <= paws_win) + return true; + if (unlikely(get_seconds() >= rx_opt->ts_recent_stamp + TCP_PAWS_24DAYS)) + return true; + /* + * Some OSes send SYN and SYNACK messages with tsval=0 tsecr=0, + * then following tcp messages have valid values. Ignore 0 value, + * or else 'negative' tsval might forbid us to accept their packets. + */ + if (!rx_opt->ts_recent) + return true; + return false; } -static __inline__ int tcp_checksum_complete(struct sk_buff *skb) +static inline bool tcp_paws_reject(const struct tcp_options_received *rx_opt, + int rst) { - return skb->ip_summed != CHECKSUM_UNNECESSARY && - __tcp_checksum_complete(skb); -} + if (tcp_paws_check(rx_opt, 0)) + return false; -/* Prequeue for VJ style copy to user, combined with checksumming. */ + /* RST segments are not recommended to carry timestamp, + and, if they do, it is recommended to ignore PAWS because + "their cleanup function should take precedence over timestamps." + Certainly, it is mistake. It is necessary to understand the reasons + of this constraint to relax it: if peer reboots, clock may go + out-of-sync and half-open connections will not be reset. + Actually, the problem would be not existing if all + the implementations followed draft about maintaining clock + via reboots. Linux-2.2 DOES NOT! -static __inline__ void tcp_prequeue_init(struct tcp_sock *tp) + However, we can relax time bounds for RST segments to MSL. + */ + if (rst && get_seconds() >= rx_opt->ts_recent_stamp + TCP_PAWS_MSL) + return false; + return true; +} + +static inline void tcp_mib_init(struct net *net) { - tp->ucopy.task = NULL; - tp->ucopy.len = 0; - tp->ucopy.memory = 0; - skb_queue_head_init(&tp->ucopy.prequeue); + /* See RFC 2012 */ + TCP_ADD_STATS_USER(net, TCP_MIB_RTOALGORITHM, 1); + TCP_ADD_STATS_USER(net, TCP_MIB_RTOMIN, TCP_RTO_MIN*1000/HZ); + TCP_ADD_STATS_USER(net, TCP_MIB_RTOMAX, TCP_RTO_MAX*1000/HZ); + TCP_ADD_STATS_USER(net, TCP_MIB_MAXCONN, -1); } -/* Packet is added to VJ-style prequeue for processing in process - * context, if a reader task is waiting. Apparently, this exciting - * idea (VJ's mail "Re: query about TCP header on tcp-ip" of 07 Sep 93) - * failed somewhere. Latency? Burstiness? Well, at least now we will - * see, why it failed. 8)8) --ANK - * - * NOTE: is this not too big to inline? - */ -static __inline__ int tcp_prequeue(struct sock *sk, struct sk_buff *skb) +/* from STCP */ +static inline void tcp_clear_retrans_hints_partial(struct tcp_sock *tp) { - struct tcp_sock *tp = tcp_sk(sk); + tp->lost_skb_hint = NULL; +} - if (!sysctl_tcp_low_latency && tp->ucopy.task) { - __skb_queue_tail(&tp->ucopy.prequeue, skb); - tp->ucopy.memory += skb->truesize; - if (tp->ucopy.memory > sk->sk_rcvbuf) { - struct sk_buff *skb1; - - BUG_ON(sock_owned_by_user(sk)); - - while ((skb1 = __skb_dequeue(&tp->ucopy.prequeue)) != NULL) { - sk->sk_backlog_rcv(sk, skb1); - NET_INC_STATS_BH(LINUX_MIB_TCPPREQUEUEDROPPED); - } - - 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); - } - return 1; - } - return 0; +static inline void tcp_clear_all_retrans_hints(struct tcp_sock *tp) +{ + tcp_clear_retrans_hints_partial(tp); + tp->retransmit_skb_hint = NULL; } +/* MD5 Signature */ +struct crypto_hash; -#undef STATE_TRACE +union tcp_md5_addr { + struct in_addr a4; +#if IS_ENABLED(CONFIG_IPV6) + struct in6_addr a6; +#endif +}; -#ifdef STATE_TRACE -static const char *statename[]={ - "Unused","Established","Syn Sent","Syn Recv", - "Fin Wait 1","Fin Wait 2","Time Wait", "Close", - "Close Wait","Last ACK","Listen","Closing" +/* - key database */ +struct tcp_md5sig_key { + struct hlist_node node; + u8 keylen; + u8 family; /* AF_INET or AF_INET6 */ + union tcp_md5_addr addr; + u8 key[TCP_MD5SIG_MAXKEYLEN]; + struct rcu_head rcu; }; -#endif -static __inline__ void tcp_set_state(struct sock *sk, int state) -{ - int oldstate = sk->sk_state; +/* - sock block */ +struct tcp_md5sig_info { + struct hlist_head head; + struct rcu_head rcu; +}; - switch (state) { - case TCP_ESTABLISHED: - if (oldstate != TCP_ESTABLISHED) - TCP_INC_STATS(TCP_MIB_CURRESTAB); - break; +/* - pseudo header */ +struct tcp4_pseudohdr { + __be32 saddr; + __be32 daddr; + __u8 pad; + __u8 protocol; + __be16 len; +}; - case TCP_CLOSE: - if (oldstate == TCP_CLOSE_WAIT || oldstate == TCP_ESTABLISHED) - TCP_INC_STATS(TCP_MIB_ESTABRESETS); +struct tcp6_pseudohdr { + struct in6_addr saddr; + struct in6_addr daddr; + __be32 len; + __be32 protocol; /* including padding */ +}; - sk->sk_prot->unhash(sk); - if (tcp_sk(sk)->bind_hash && - !(sk->sk_userlocks & SOCK_BINDPORT_LOCK)) - tcp_put_port(sk); - /* fall through */ - default: - if (oldstate==TCP_ESTABLISHED) - TCP_DEC_STATS(TCP_MIB_CURRESTAB); - } +union tcp_md5sum_block { + struct tcp4_pseudohdr ip4; +#if IS_ENABLED(CONFIG_IPV6) + struct tcp6_pseudohdr ip6; +#endif +}; - /* Change state AFTER socket is unhashed to avoid closed - * socket sitting in hash tables. - */ - sk->sk_state = state; +/* - pool: digest algorithm, hash description and scratch buffer */ +struct tcp_md5sig_pool { + struct hash_desc md5_desc; + union tcp_md5sum_block md5_blk; +}; -#ifdef STATE_TRACE - SOCK_DEBUG(sk, "TCP sk=%p, State %s -> %s\n",sk, statename[oldstate],statename[state]); -#endif +/* - functions */ +int tcp_v4_md5_hash_skb(char *md5_hash, struct tcp_md5sig_key *key, + const struct sock *sk, const struct request_sock *req, + const struct sk_buff *skb); +int tcp_md5_do_add(struct sock *sk, const union tcp_md5_addr *addr, + int family, const u8 *newkey, u8 newkeylen, gfp_t gfp); +int tcp_md5_do_del(struct sock *sk, const union tcp_md5_addr *addr, + int family); +struct tcp_md5sig_key *tcp_v4_md5_lookup(struct sock *sk, + struct sock *addr_sk); + +#ifdef CONFIG_TCP_MD5SIG +struct tcp_md5sig_key *tcp_md5_do_lookup(struct sock *sk, + const union tcp_md5_addr *addr, + int family); +#define tcp_twsk_md5_key(twsk) ((twsk)->tw_md5_key) +#else +static inline struct tcp_md5sig_key *tcp_md5_do_lookup(struct sock *sk, + const union tcp_md5_addr *addr, + int family) +{ + return NULL; } +#define tcp_twsk_md5_key(twsk) NULL +#endif -static __inline__ void tcp_done(struct sock *sk) +bool tcp_alloc_md5sig_pool(void); + +struct tcp_md5sig_pool *tcp_get_md5sig_pool(void); +static inline void tcp_put_md5sig_pool(void) { - tcp_set_state(sk, TCP_CLOSE); - tcp_clear_xmit_timers(sk); + local_bh_enable(); +} - sk->sk_shutdown = SHUTDOWN_MASK; +int tcp_md5_hash_header(struct tcp_md5sig_pool *, const struct tcphdr *); +int tcp_md5_hash_skb_data(struct tcp_md5sig_pool *, const struct sk_buff *, + unsigned int header_len); +int tcp_md5_hash_key(struct tcp_md5sig_pool *hp, + const struct tcp_md5sig_key *key); - if (!sock_flag(sk, SOCK_DEAD)) - sk->sk_state_change(sk); - else - tcp_destroy_sock(sk); -} +/* From tcp_fastopen.c */ +void tcp_fastopen_cache_get(struct sock *sk, u16 *mss, + struct tcp_fastopen_cookie *cookie, int *syn_loss, + unsigned long *last_syn_loss); +void tcp_fastopen_cache_set(struct sock *sk, u16 mss, + struct tcp_fastopen_cookie *cookie, bool syn_lost); +struct tcp_fastopen_request { + /* Fast Open cookie. Size 0 means a cookie request */ + struct tcp_fastopen_cookie cookie; + struct msghdr *data; /* data in MSG_FASTOPEN */ + size_t size; + int copied; /* queued in tcp_connect() */ +}; +void tcp_free_fastopen_req(struct tcp_sock *tp); + +extern struct tcp_fastopen_context __rcu *tcp_fastopen_ctx; +int tcp_fastopen_reset_cipher(void *key, unsigned int len); +bool tcp_try_fastopen(struct sock *sk, struct sk_buff *skb, + struct request_sock *req, + struct tcp_fastopen_cookie *foc, + struct dst_entry *dst); +void tcp_fastopen_init_key_once(bool publish); +#define TCP_FASTOPEN_KEY_LENGTH 16 + +/* Fastopen key context */ +struct tcp_fastopen_context { + struct crypto_cipher *tfm; + __u8 key[TCP_FASTOPEN_KEY_LENGTH]; + struct rcu_head rcu; +}; -static __inline__ void tcp_sack_reset(struct tcp_options_received *rx_opt) +/* write queue abstraction */ +static inline void tcp_write_queue_purge(struct sock *sk) { - rx_opt->dsack = 0; - rx_opt->eff_sacks = 0; - rx_opt->num_sacks = 0; + struct sk_buff *skb; + + while ((skb = __skb_dequeue(&sk->sk_write_queue)) != NULL) + sk_wmem_free_skb(sk, skb); + sk_mem_reclaim(sk); + tcp_clear_all_retrans_hints(tcp_sk(sk)); } -static __inline__ void tcp_build_and_update_options(__u32 *ptr, struct tcp_sock *tp, __u32 tstamp) +static inline struct sk_buff *tcp_write_queue_head(const struct sock *sk) { - if (tp->rx_opt.tstamp_ok) { - *ptr++ = __constant_htonl((TCPOPT_NOP << 24) | - (TCPOPT_NOP << 16) | - (TCPOPT_TIMESTAMP << 8) | - TCPOLEN_TIMESTAMP); - *ptr++ = htonl(tstamp); - *ptr++ = htonl(tp->rx_opt.ts_recent); - } - if (tp->rx_opt.eff_sacks) { - struct tcp_sack_block *sp = tp->rx_opt.dsack ? tp->duplicate_sack : tp->selective_acks; - int this_sack; - - *ptr++ = __constant_htonl((TCPOPT_NOP << 24) | - (TCPOPT_NOP << 16) | - (TCPOPT_SACK << 8) | - (TCPOLEN_SACK_BASE + - (tp->rx_opt.eff_sacks * TCPOLEN_SACK_PERBLOCK))); - for(this_sack = 0; this_sack < tp->rx_opt.eff_sacks; this_sack++) { - *ptr++ = htonl(sp[this_sack].start_seq); - *ptr++ = htonl(sp[this_sack].end_seq); - } - if (tp->rx_opt.dsack) { - tp->rx_opt.dsack = 0; - tp->rx_opt.eff_sacks--; - } - } + return skb_peek(&sk->sk_write_queue); } -/* Construct a tcp options header for a SYN or SYN_ACK packet. - * If this is every changed make sure to change the definition of - * MAX_SYN_SIZE to match the new maximum number of options that you - * can generate. - */ -static inline void tcp_syn_build_options(__u32 *ptr, int mss, int ts, int sack, - int offer_wscale, int wscale, __u32 tstamp, __u32 ts_recent) -{ - /* We always get an MSS option. - * The option bytes which will be seen in normal data - * packets should timestamps be used, must be in the MSS - * advertised. But we subtract them from tp->mss_cache so - * that calculations in tcp_sendmsg are simpler etc. - * So account for this fact here if necessary. If we - * don't do this correctly, as a receiver we won't - * recognize data packets as being full sized when we - * should, and thus we won't abide by the delayed ACK - * rules correctly. - * SACKs don't matter, we never delay an ACK when we - * have any of those going out. - */ - *ptr++ = htonl((TCPOPT_MSS << 24) | (TCPOLEN_MSS << 16) | mss); - if (ts) { - if(sack) - *ptr++ = __constant_htonl((TCPOPT_SACK_PERM << 24) | (TCPOLEN_SACK_PERM << 16) | - (TCPOPT_TIMESTAMP << 8) | TCPOLEN_TIMESTAMP); - else - *ptr++ = __constant_htonl((TCPOPT_NOP << 24) | (TCPOPT_NOP << 16) | - (TCPOPT_TIMESTAMP << 8) | TCPOLEN_TIMESTAMP); - *ptr++ = htonl(tstamp); /* TSVAL */ - *ptr++ = htonl(ts_recent); /* TSECR */ - } else if(sack) - *ptr++ = __constant_htonl((TCPOPT_NOP << 24) | (TCPOPT_NOP << 16) | - (TCPOPT_SACK_PERM << 8) | TCPOLEN_SACK_PERM); - if (offer_wscale) - *ptr++ = htonl((TCPOPT_NOP << 24) | (TCPOPT_WINDOW << 16) | (TCPOLEN_WINDOW << 8) | (wscale)); +static inline struct sk_buff *tcp_write_queue_tail(const struct sock *sk) +{ + return skb_peek_tail(&sk->sk_write_queue); } -/* Determine a window scaling and initial window to offer. */ -extern void tcp_select_initial_window(int __space, __u32 mss, - __u32 *rcv_wnd, __u32 *window_clamp, - int wscale_ok, __u8 *rcv_wscale); - -static inline int tcp_win_from_space(int space) +static inline struct sk_buff *tcp_write_queue_next(const struct sock *sk, + const struct sk_buff *skb) { - return sysctl_tcp_adv_win_scale<=0 ? - (space>>(-sysctl_tcp_adv_win_scale)) : - space - (space>>sysctl_tcp_adv_win_scale); + return skb_queue_next(&sk->sk_write_queue, skb); } -/* Note: caller must be prepared to deal with negative returns */ -static inline int tcp_space(const struct sock *sk) +static inline struct sk_buff *tcp_write_queue_prev(const struct sock *sk, + const struct sk_buff *skb) { - return tcp_win_from_space(sk->sk_rcvbuf - - atomic_read(&sk->sk_rmem_alloc)); -} + return skb_queue_prev(&sk->sk_write_queue, skb); +} -static inline int tcp_full_space(const struct sock *sk) +#define tcp_for_write_queue(skb, sk) \ + skb_queue_walk(&(sk)->sk_write_queue, skb) + +#define tcp_for_write_queue_from(skb, sk) \ + skb_queue_walk_from(&(sk)->sk_write_queue, skb) + +#define tcp_for_write_queue_from_safe(skb, tmp, sk) \ + skb_queue_walk_from_safe(&(sk)->sk_write_queue, skb, tmp) + +static inline struct sk_buff *tcp_send_head(const struct sock *sk) { - return tcp_win_from_space(sk->sk_rcvbuf); + return sk->sk_send_head; } -static inline void tcp_acceptq_queue(struct sock *sk, struct request_sock *req, - struct sock *child) +static inline bool tcp_skb_is_last(const struct sock *sk, + const struct sk_buff *skb) { - reqsk_queue_add(&tcp_sk(sk)->accept_queue, req, sk, child); + return skb_queue_is_last(&sk->sk_write_queue, skb); } -static inline void -tcp_synq_removed(struct sock *sk, struct request_sock *req) +static inline void tcp_advance_send_head(struct sock *sk, const struct sk_buff *skb) { - if (reqsk_queue_removed(&tcp_sk(sk)->accept_queue, req) == 0) - tcp_delete_keepalive_timer(sk); + if (tcp_skb_is_last(sk, skb)) + sk->sk_send_head = NULL; + else + sk->sk_send_head = tcp_write_queue_next(sk, skb); } -static inline void tcp_synq_added(struct sock *sk) +static inline void tcp_check_send_head(struct sock *sk, struct sk_buff *skb_unlinked) { - if (reqsk_queue_added(&tcp_sk(sk)->accept_queue) == 0) - tcp_reset_keepalive_timer(sk, TCP_TIMEOUT_INIT); + if (sk->sk_send_head == skb_unlinked) + sk->sk_send_head = NULL; } -static inline int tcp_synq_len(struct sock *sk) +static inline void tcp_init_send_head(struct sock *sk) { - return reqsk_queue_len(&tcp_sk(sk)->accept_queue); + sk->sk_send_head = NULL; } -static inline int tcp_synq_young(struct sock *sk) +static inline void __tcp_add_write_queue_tail(struct sock *sk, struct sk_buff *skb) { - return reqsk_queue_len_young(&tcp_sk(sk)->accept_queue); + __skb_queue_tail(&sk->sk_write_queue, skb); } -static inline int tcp_synq_is_full(struct sock *sk) +static inline void tcp_add_write_queue_tail(struct sock *sk, struct sk_buff *skb) { - return reqsk_queue_is_full(&tcp_sk(sk)->accept_queue); + __tcp_add_write_queue_tail(sk, skb); + + /* Queue it, remembering where we must start sending. */ + if (sk->sk_send_head == NULL) { + sk->sk_send_head = skb; + + if (tcp_sk(sk)->highest_sack == NULL) + tcp_sk(sk)->highest_sack = skb; + } } -static inline void tcp_synq_unlink(struct tcp_sock *tp, struct request_sock *req, - struct request_sock **prev) +static inline void __tcp_add_write_queue_head(struct sock *sk, struct sk_buff *skb) { - reqsk_queue_unlink(&tp->accept_queue, req, prev); + __skb_queue_head(&sk->sk_write_queue, skb); } -static inline void tcp_synq_drop(struct sock *sk, struct request_sock *req, - struct request_sock **prev) +/* Insert buff after skb on the write queue of sk. */ +static inline void tcp_insert_write_queue_after(struct sk_buff *skb, + struct sk_buff *buff, + struct sock *sk) { - tcp_synq_unlink(tcp_sk(sk), req, prev); - tcp_synq_removed(sk, req); - reqsk_free(req); + __skb_queue_after(&sk->sk_write_queue, skb, buff); } -static __inline__ void tcp_openreq_init(struct request_sock *req, - struct tcp_options_received *rx_opt, - struct sk_buff *skb) +/* Insert new before skb on the write queue of sk. */ +static inline void tcp_insert_write_queue_before(struct sk_buff *new, + struct sk_buff *skb, + struct sock *sk) { - struct inet_request_sock *ireq = inet_rsk(req); + __skb_queue_before(&sk->sk_write_queue, skb, new); - req->rcv_wnd = 0; /* So that tcp_send_synack() knows! */ - tcp_rsk(req)->rcv_isn = TCP_SKB_CB(skb)->seq; - req->mss = rx_opt->mss_clamp; - req->ts_recent = rx_opt->saw_tstamp ? rx_opt->rcv_tsval : 0; - ireq->tstamp_ok = rx_opt->tstamp_ok; - ireq->sack_ok = rx_opt->sack_ok; - ireq->snd_wscale = rx_opt->snd_wscale; - ireq->wscale_ok = rx_opt->wscale_ok; - ireq->acked = 0; - ireq->ecn_ok = 0; - ireq->rmt_port = skb->h.th->source; + if (sk->sk_send_head == skb) + sk->sk_send_head = new; } -extern void tcp_enter_memory_pressure(void); - -extern void tcp_listen_wlock(void); - -/* - We may sleep inside this lock. - * - If sleeping is not required (or called from BH), - * use plain read_(un)lock(&tcp_lhash_lock). - */ - -static inline void tcp_listen_lock(void) +static inline void tcp_unlink_write_queue(struct sk_buff *skb, struct sock *sk) { - /* read_lock synchronizes to candidates to writers */ - read_lock(&tcp_lhash_lock); - atomic_inc(&tcp_lhash_users); - read_unlock(&tcp_lhash_lock); + __skb_unlink(skb, &sk->sk_write_queue); } -static inline void tcp_listen_unlock(void) +static inline bool tcp_write_queue_empty(struct sock *sk) { - if (atomic_dec_and_test(&tcp_lhash_users)) - wake_up(&tcp_lhash_wait); + return skb_queue_empty(&sk->sk_write_queue); } -static inline int keepalive_intvl_when(const struct tcp_sock *tp) +static inline void tcp_push_pending_frames(struct sock *sk) { - return tp->keepalive_intvl ? : sysctl_tcp_keepalive_intvl; -} + if (tcp_send_head(sk)) { + struct tcp_sock *tp = tcp_sk(sk); -static inline int keepalive_time_when(const struct tcp_sock *tp) -{ - return tp->keepalive_time ? : sysctl_tcp_keepalive_time; + __tcp_push_pending_frames(sk, tcp_current_mss(sk), tp->nonagle); + } } -static inline int tcp_fin_time(const struct tcp_sock *tp) +/* Start sequence of the skb just after the highest skb with SACKed + * bit, valid only if sacked_out > 0 or when the caller has ensured + * validity by itself. + */ +static inline u32 tcp_highest_sack_seq(struct tcp_sock *tp) { - int fin_timeout = tp->linger2 ? : sysctl_tcp_fin_timeout; + if (!tp->sacked_out) + return tp->snd_una; - if (fin_timeout < (tp->rto<<2) - (tp->rto>>1)) - fin_timeout = (tp->rto<<2) - (tp->rto>>1); + if (tp->highest_sack == NULL) + return tp->snd_nxt; - return fin_timeout; + return TCP_SKB_CB(tp->highest_sack)->seq; } -static inline int tcp_paws_check(const struct tcp_options_received *rx_opt, int rst) +static inline void tcp_advance_highest_sack(struct sock *sk, struct sk_buff *skb) { - if ((s32)(rx_opt->rcv_tsval - rx_opt->ts_recent) >= 0) - return 0; - if (xtime.tv_sec >= rx_opt->ts_recent_stamp + TCP_PAWS_24DAYS) - return 0; - - /* RST segments are not recommended to carry timestamp, - and, if they do, it is recommended to ignore PAWS because - "their cleanup function should take precedence over timestamps." - Certainly, it is mistake. It is necessary to understand the reasons - of this constraint to relax it: if peer reboots, clock may go - out-of-sync and half-open connections will not be reset. - Actually, the problem would be not existing if all - the implementations followed draft about maintaining clock - via reboots. Linux-2.2 DOES NOT! - - However, we can relax time bounds for RST segments to MSL. - */ - if (rst && xtime.tv_sec >= rx_opt->ts_recent_stamp + TCP_PAWS_MSL) - return 0; - return 1; + tcp_sk(sk)->highest_sack = tcp_skb_is_last(sk, skb) ? NULL : + tcp_write_queue_next(sk, skb); } -static inline void tcp_v4_setup_caps(struct sock *sk, struct dst_entry *dst) +static inline struct sk_buff *tcp_highest_sack(struct sock *sk) { - sk->sk_route_caps = dst->dev->features; - if (sk->sk_route_caps & NETIF_F_TSO) { - if (sock_flag(sk, SOCK_NO_LARGESEND) || dst->header_len) - sk->sk_route_caps &= ~NETIF_F_TSO; - } + return tcp_sk(sk)->highest_sack; } -#define TCP_CHECK_TIMER(sk) do { } while (0) +static inline void tcp_highest_sack_reset(struct sock *sk) +{ + tcp_sk(sk)->highest_sack = tcp_write_queue_head(sk); +} -static inline int tcp_use_frto(const struct sock *sk) +/* Called when old skb is about to be deleted (to be combined with new skb) */ +static inline void tcp_highest_sack_combine(struct sock *sk, + struct sk_buff *old, + struct sk_buff *new) { - const struct tcp_sock *tp = tcp_sk(sk); - - /* F-RTO must be activated in sysctl and there must be some - * unsent new data, and the advertised window should allow - * sending it. - */ - return (sysctl_tcp_frto && sk->sk_send_head && - !after(TCP_SKB_CB(sk->sk_send_head)->end_seq, - tp->snd_una + tp->snd_wnd)); + if (tcp_sk(sk)->sacked_out && (old == tcp_sk(sk)->highest_sack)) + tcp_sk(sk)->highest_sack = new; } -static inline void tcp_mib_init(void) +/* Determines whether this is a thin stream (which may suffer from + * increased latency). Used to trigger latency-reducing mechanisms. + */ +static inline bool tcp_stream_is_thin(struct tcp_sock *tp) { - /* See RFC 2012 */ - TCP_ADD_STATS_USER(TCP_MIB_RTOALGORITHM, 1); - TCP_ADD_STATS_USER(TCP_MIB_RTOMIN, TCP_RTO_MIN*1000/HZ); - TCP_ADD_STATS_USER(TCP_MIB_RTOMAX, TCP_RTO_MAX*1000/HZ); - TCP_ADD_STATS_USER(TCP_MIB_MAXCONN, -1); + return tp->packets_out < 4 && !tcp_in_initial_slowstart(tp); } /* /proc */ @@ -1854,74 +1531,91 @@ enum tcp_seq_states { TCP_SEQ_STATE_LISTENING, TCP_SEQ_STATE_OPENREQ, TCP_SEQ_STATE_ESTABLISHED, - TCP_SEQ_STATE_TIME_WAIT, }; +int tcp_seq_open(struct inode *inode, struct file *file); + struct tcp_seq_afinfo { - struct module *owner; - char *name; - sa_family_t family; - int (*seq_show) (struct seq_file *m, void *v); - struct file_operations *seq_fops; + char *name; + sa_family_t family; + const struct file_operations *seq_fops; + struct seq_operations seq_ops; }; struct tcp_iter_state { + struct seq_net_private p; sa_family_t family; enum tcp_seq_states state; struct sock *syn_wait_sk; - int bucket, sbucket, num, uid; - struct seq_operations seq_ops; + int bucket, offset, sbucket, num; + kuid_t uid; + loff_t last_pos; }; -extern int tcp_proc_register(struct tcp_seq_afinfo *afinfo); -extern void tcp_proc_unregister(struct tcp_seq_afinfo *afinfo); +int tcp_proc_register(struct net *net, struct tcp_seq_afinfo *afinfo); +void tcp_proc_unregister(struct net *net, struct tcp_seq_afinfo *afinfo); -/* TCP Westwood functions and constants */ +extern struct request_sock_ops tcp_request_sock_ops; +extern struct request_sock_ops tcp6_request_sock_ops; -#define TCP_WESTWOOD_INIT_RTT (20*HZ) /* maybe too conservative?! */ -#define TCP_WESTWOOD_RTT_MIN (HZ/20) /* 50ms */ +void tcp_v4_destroy_sock(struct sock *sk); -static inline void tcp_westwood_update_rtt(struct tcp_sock *tp, __u32 rtt_seq) -{ - if (tcp_is_westwood(tp)) - tp->westwood.rtt = rtt_seq; -} +struct sk_buff *tcp_gso_segment(struct sk_buff *skb, + netdev_features_t features); +struct sk_buff **tcp_gro_receive(struct sk_buff **head, struct sk_buff *skb); +int tcp_gro_complete(struct sk_buff *skb); + +void __tcp_v4_send_check(struct sk_buff *skb, __be32 saddr, __be32 daddr); -static inline __u32 __tcp_westwood_bw_rttmin(const struct tcp_sock *tp) +static inline u32 tcp_notsent_lowat(const struct tcp_sock *tp) { - return max((tp->westwood.bw_est) * (tp->westwood.rtt_min) / - (__u32) (tp->mss_cache_std), - 2U); + return tp->notsent_lowat ?: sysctl_tcp_notsent_lowat; } -static inline __u32 tcp_westwood_bw_rttmin(const struct tcp_sock *tp) +static inline bool tcp_stream_memory_free(const struct sock *sk) { - return tcp_is_westwood(tp) ? __tcp_westwood_bw_rttmin(tp) : 0; + const struct tcp_sock *tp = tcp_sk(sk); + u32 notsent_bytes = tp->write_seq - tp->snd_nxt; + + return notsent_bytes < tcp_notsent_lowat(tp); } -static inline int tcp_westwood_ssthresh(struct tcp_sock *tp) -{ - __u32 ssthresh = 0; +#ifdef CONFIG_PROC_FS +int tcp4_proc_init(void); +void tcp4_proc_exit(void); +#endif - if (tcp_is_westwood(tp)) { - ssthresh = __tcp_westwood_bw_rttmin(tp); - if (ssthresh) - tp->snd_ssthresh = ssthresh; - } +/* TCP af-specific functions */ +struct tcp_sock_af_ops { +#ifdef CONFIG_TCP_MD5SIG + struct tcp_md5sig_key *(*md5_lookup) (struct sock *sk, + struct sock *addr_sk); + int (*calc_md5_hash) (char *location, + struct tcp_md5sig_key *md5, + const struct sock *sk, + const struct request_sock *req, + const struct sk_buff *skb); + int (*md5_parse) (struct sock *sk, + char __user *optval, + int optlen); +#endif +}; - return (ssthresh != 0); -} +struct tcp_request_sock_ops { +#ifdef CONFIG_TCP_MD5SIG + struct tcp_md5sig_key *(*md5_lookup) (struct sock *sk, + struct request_sock *req); + int (*calc_md5_hash) (char *location, + struct tcp_md5sig_key *md5, + const struct sock *sk, + const struct request_sock *req, + const struct sk_buff *skb); +#endif +}; -static inline int tcp_westwood_cwnd(struct tcp_sock *tp) -{ - __u32 cwnd = 0; +int tcpv4_offload_init(void); - if (tcp_is_westwood(tp)) { - cwnd = __tcp_westwood_bw_rttmin(tp); - if (cwnd) - tp->snd_cwnd = cwnd; - } +void tcp_v4_init(void); +void tcp_init(void); - return (cwnd != 0); -} #endif /* _TCP_H */ |