/* * CDDL HEADER START * * The contents of this file are subject to the terms of the * Common Development and Distribution License (the "License"). * You may not use this file except in compliance with the License. * * You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE * or http://www.opensolaris.org/os/licensing. * See the License for the specific language governing permissions * and limitations under the License. * * When distributing Covered Code, include this CDDL HEADER in each * file and include the License file at usr/src/OPENSOLARIS.LICENSE. * If applicable, add the following below this CDDL HEADER, with the * fields enclosed by brackets "[]" replaced with your own identifying * information: Portions Copyright [yyyy] [name of copyright owner] * * CDDL HEADER END */ /* * Copyright 2006 Sun Microsystems, Inc. All rights reserved. * Use is subject to license terms. */ /* Copyright (c) 1990 Mentat Inc. */ #ifndef _INET_TCP_H #define _INET_TCP_H #pragma ident "%Z%%M% %I% %E% SMI" #ifdef __cplusplus extern "C" { #endif #include #include #include #include #include #include #include /* * Private (and possibly temporary) ioctl used by configuration code * to lock in the "default" stream for detached closes. */ #define TCP_IOC_DEFAULT_Q (('T' << 8) + 51) /* TCP states */ #define TCPS_CLOSED -6 #define TCPS_IDLE -5 /* idle (opened, but not bound) */ #define TCPS_BOUND -4 /* bound, ready to connect or accept */ #define TCPS_LISTEN -3 /* listening for connection */ #define TCPS_SYN_SENT -2 /* active, have sent syn */ #define TCPS_SYN_RCVD -1 /* have received syn (and sent ours) */ /* states < TCPS_ESTABLISHED are those where connections not established */ #define TCPS_ESTABLISHED 0 /* established */ #define TCPS_CLOSE_WAIT 1 /* rcvd fin, waiting for close */ /* states > TCPS_CLOSE_WAIT are those where user has closed */ #define TCPS_FIN_WAIT_1 2 /* have closed and sent fin */ #define TCPS_CLOSING 3 /* closed, xchd FIN, await FIN ACK */ #define TCPS_LAST_ACK 4 /* had fin and close; await FIN ACK */ /* states > TCPS_CLOSE_WAIT && < TCPS_FIN_WAIT_2 await ACK of FIN */ #define TCPS_FIN_WAIT_2 5 /* have closed, fin is acked */ #define TCPS_TIME_WAIT 6 /* in 2*msl quiet wait after close */ /* * Internal flags used in conjunction with the packet header flags. * Used in tcp_rput_data to keep track of what needs to be done. */ #define TH_LIMIT_XMIT 0x0400 /* Limited xmit is needed */ #define TH_XMIT_NEEDED 0x0800 /* Window opened - send queued data */ #define TH_REXMIT_NEEDED 0x1000 /* Time expired for unacked data */ #define TH_ACK_NEEDED 0x2000 /* Send an ack now. */ #define TH_NEED_SACK_REXMIT 0x4000 /* Use SACK info to retransmission */ #define TH_ACK_TIMER_NEEDED 0x8000 /* Start the delayed ACK timer */ #define TH_ORDREL_NEEDED 0x10000 /* Generate an ordrel indication */ #define TH_MARKNEXT_NEEDED 0x20000 /* Data should have MSGMARKNEXT */ #define TH_SEND_URP_MARK 0x40000 /* Send up tcp_urp_mark_mp */ /* * TCP sequence numbers are 32 bit integers operated * on with modular arithmetic. These macros can be * used to compare such integers. */ #define SEQ_LT(a, b) ((int32_t)((a)-(b)) < 0) #define SEQ_LEQ(a, b) ((int32_t)((a)-(b)) <= 0) #define SEQ_GT(a, b) ((int32_t)((a)-(b)) > 0) #define SEQ_GEQ(a, b) ((int32_t)((a)-(b)) >= 0) /* TCP Protocol header */ typedef struct tcphdr_s { uint8_t th_lport[2]; /* Source port */ uint8_t th_fport[2]; /* Destination port */ uint8_t th_seq[4]; /* Sequence number */ uint8_t th_ack[4]; /* Acknowledgement number */ uint8_t th_offset_and_rsrvd[1]; /* Offset to the packet data */ uint8_t th_flags[1]; uint8_t th_win[2]; /* Allocation number */ uint8_t th_sum[2]; /* TCP checksum */ uint8_t th_urp[2]; /* Urgent pointer */ } tcph_t; #define TCP_HDR_LENGTH(tcph) (((tcph)->th_offset_and_rsrvd[0] >>2) &(0xF << 2)) #define TCP_MAX_COMBINED_HEADER_LENGTH (60 + 60) /* Maxed out ip + tcp */ #define TCP_MAX_IP_OPTIONS_LENGTH (60 - IP_SIMPLE_HDR_LENGTH) #define TCP_MAX_HDR_LENGTH 60 #define TCP_MAX_TCP_OPTIONS_LENGTH (60 - sizeof (tcph_t)) #define TCP_MIN_HEADER_LENGTH 20 #define TCP_MAXWIN 65535 #define TCP_PORT_LEN sizeof (in_port_t) #define TCP_MAX_WINSHIFT 14 #define TCP_MAX_LARGEWIN (TCP_MAXWIN << TCP_MAX_WINSHIFT) #define TCPIP_HDR_LENGTH(mp, n) \ (n) = IPH_HDR_LENGTH((mp)->b_rptr), \ (n) += TCP_HDR_LENGTH((tcph_t *)&(mp)->b_rptr[(n)]) /* TCP Protocol header (used if the header is known to be 32-bit aligned) */ typedef struct tcphdra_s { in_port_t tha_lport; /* Source port */ in_port_t tha_fport; /* Destination port */ uint32_t tha_seq; /* Sequence number */ uint32_t tha_ack; /* Acknowledgement number */ uint8_t tha_offset_and_reserved; /* Offset to the packet data */ uint8_t tha_flags; uint16_t tha_win; /* Allocation number */ uint16_t tha_sum; /* TCP checksum */ uint16_t tha_urp; /* Urgent pointer */ } tcpha_t; struct conn_s; /* * Control structure for each open TCP stream, * defined only within the kernel or for a kmem user. * NOTE: tcp_reinit_values MUST have a line for each field in this structure! */ #if (defined(_KERNEL) || defined(_KMEMUSER)) typedef struct tcp_s { /* Pointer to previous bind hash next. */ struct tcp_s *tcp_time_wait_next; /* Pointer to next T/W block */ struct tcp_s *tcp_time_wait_prev; /* Pointer to previous T/W next */ clock_t tcp_time_wait_expire; struct conn_s *tcp_connp; int32_t tcp_state; int32_t tcp_rcv_ws; /* My window scale power */ int32_t tcp_snd_ws; /* Sender's window scale power */ uint32_t tcp_ts_recent; /* Timestamp of earliest unacked */ /* data segment */ clock_t tcp_rto; /* Round trip timeout */ clock_t tcp_last_rcv_lbolt; /* lbolt on last packet, used for PAWS */ uint32_t tcp_snxt; /* Senders next seq num */ uint32_t tcp_swnd; /* Senders window (relative to suna) */ uint32_t tcp_mss; /* Max segment size */ uint32_t tcp_iss; /* Initial send seq num */ uint32_t tcp_rnxt; /* Seq we expect to recv next */ uint32_t tcp_rwnd; queue_t *tcp_rq; /* Our upstream neighbor (client) */ queue_t *tcp_wq; /* Our downstream neighbor */ /* Fields arranged in approximate access order along main paths */ mblk_t *tcp_xmit_head; /* Head of rexmit list */ mblk_t *tcp_xmit_last; /* last valid data seen by tcp_wput */ mblk_t *tcp_xmit_tail; /* Last rexmit data sent */ uint32_t tcp_unsent; /* # of bytes in hand that are unsent */ uint32_t tcp_xmit_tail_unsent; /* # of unsent bytes in xmit_tail */ uint32_t tcp_suna; /* Sender unacknowledged */ uint32_t tcp_rexmit_nxt; /* Next rexmit seq num */ uint32_t tcp_rexmit_max; /* Max retran seq num */ int32_t tcp_snd_burst; /* Send burst factor */ uint32_t tcp_cwnd; /* Congestion window */ int32_t tcp_cwnd_cnt; /* cwnd cnt in congestion avoidance */ uint32_t tcp_ibsegs; /* Inbound segments on this stream */ uint32_t tcp_obsegs; /* Outbound segments on this stream */ uint32_t tcp_naglim; /* Tunable nagle limit */ uint32_t tcp_valid_bits; #define TCP_ISS_VALID 0x1 /* Is the tcp_iss seq num active? */ #define TCP_FSS_VALID 0x2 /* Is the tcp_fss seq num active? */ #define TCP_URG_VALID 0x4 /* Is the tcp_urg seq num active? */ #define TCP_OFO_FIN_VALID 0x8 /* Has TCP received an out of order FIN? */ int32_t tcp_xmit_hiwater; /* Send buffer high water mark. */ timeout_id_t tcp_timer_tid; /* Control block for timer service */ uchar_t tcp_timer_backoff; /* Backoff shift count. */ int64_t tcp_last_recv_time; /* Last time we receive a segment. */ uint32_t tcp_init_cwnd; /* Initial cwnd (start/restart) */ /* * Following socket options are set by sockfs outside the squeue * and we want to separate these bit fields from the other bit fields * set by TCP to avoid grabbing locks. sockfs ensures that only one * thread in sockfs can set a socket option at a time on a conn_t. * However TCP may read these options concurrently. The linger option * needs atomicity since tcp_lingertime also needs to be in sync. * However TCP uses it only during close, and by then no socket option * can come down. So we don't need any locks, instead just separating * the sockfs settable bit fields from the other bit fields is * sufficient. */ uint32_t tcp_debug : 1, /* SO_DEBUG "socket" option. */ tcp_dontroute : 1, /* SO_DONTROUTE "socket" option. */ tcp_broadcast : 1, /* SO_BROADCAST "socket" option. */ tcp_useloopback : 1, /* SO_USELOOPBACK "socket" option. */ tcp_oobinline : 1, /* SO_OOBINLINE "socket" option. */ tcp_dgram_errind : 1, /* SO_DGRAM_ERRIND option */ tcp_linger : 1, /* SO_LINGER turned on */ tcp_reuseaddr : 1, /* SO_REUSEADDR "socket" option. */ tcp_junk_to_bit_31 : 24; /* Following manipulated by TCP under squeue protection */ uint32_t tcp_urp_last_valid : 1, /* Is tcp_urp_last valid? */ tcp_hard_binding : 1, /* If we've started a full bind */ tcp_hard_bound : 1, /* If we've done a full bind with IP */ tcp_fin_acked : 1, /* Has our FIN been acked? */ tcp_fin_rcvd : 1, /* Have we seen a FIN? */ tcp_fin_sent : 1, /* Have we sent our FIN yet? */ tcp_ordrel_done : 1, /* Have we sent the ord_rel upstream? */ tcp_flow_stopped : 1, /* Have we flow controlled xmitter? */ tcp_detached : 1, /* If we're detached from a stream */ tcp_bind_pending : 1, /* Client is waiting for bind ack */ tcp_unbind_pending : 1, /* Client sent T_UNBIND_REQ */ tcp_deferred_clean_death : 1, /* defer tcp endpoint cleanup etc. */ tcp_conn_def_q0: 1, /* move from q0 to q deferred */ tcp_ka_enabled: 1, /* Connection KeepAlive Timer needed */ tcp_zero_win_probe: 1, /* Zero win probing is in progress */ tcp_loopback: 1, /* src and dst are the same machine */ tcp_localnet: 1, /* src and dst are on the same subnet */ tcp_syn_defense: 1, /* For defense against SYN attack */ #define tcp_dontdrop tcp_syn_defense tcp_set_timer : 1, tcp_active_open: 1, /* This is a active open */ tcp_timeout : 1, /* qbufcall failed, qtimeout pending */ tcp_rexmit : 1, /* TCP is retransmitting */ tcp_snd_sack_ok : 1, /* Can use SACK for this connection */ tcp_empty_flag : 1, /* Empty flag for future use */ tcp_recvdstaddr : 1, /* return T_EXTCONN_IND with dst addr */ tcp_hwcksum : 1, /* The NIC is capable of hwcksum */ tcp_ip_forward_progress : 1, tcp_anon_priv_bind : 1, tcp_ecn_ok : 1, /* Can use ECN for this connection */ tcp_ecn_echo_on : 1, /* Need to do ECN echo */ tcp_ecn_cwr_sent : 1, /* ECN_CWR has been sent */ tcp_cwr : 1; /* Cwnd has reduced recently */ /* Following manipulated by TCP under squeue protection */ uint32_t tcp_mdt : 1, /* Lower layer is capable of MDT */ tcp_snd_ts_ok : 1, tcp_snd_ws_ok : 1, tcp_exclbind : 1, /* ``exclusive'' binding */ tcp_reserved_port : 1, tcp_hdr_grown : 1, tcp_in_free_list : 1, tcp_snd_zcopy_on : 1, /* xmit zero-copy enabled */ tcp_snd_zcopy_aware : 1, /* client is zero-copy aware */ tcp_xmit_zc_clean : 1, /* the xmit list is free of zc-mblk */ tcp_wait_for_eagers : 1, /* Wait for eagers to disappear */ tcp_accept_error : 1, /* Error during TLI accept */ tcp_send_discon_ind : 1, /* TLI accept err, send discon ind */ tcp_cork : 1, /* tcp_cork option */ tcp_pad_to_bit_31 : 18; uint32_t tcp_if_mtu; /* Outgoing interface MTU. */ mblk_t *tcp_reass_head; /* Out of order reassembly list head */ mblk_t *tcp_reass_tail; /* Out of order reassembly list tail */ tcp_sack_info_t *tcp_sack_info; #define tcp_pipe tcp_sack_info->tcp_pipe #define tcp_fack tcp_sack_info->tcp_fack #define tcp_sack_snxt tcp_sack_info->tcp_sack_snxt #define tcp_max_sack_blk tcp_sack_info->tcp_max_sack_blk #define tcp_num_sack_blk tcp_sack_info->tcp_num_sack_blk #define tcp_sack_list tcp_sack_info->tcp_sack_list #define tcp_num_notsack_blk tcp_sack_info->tcp_num_notsack_blk #define tcp_cnt_notsack_list tcp_sack_info->tcp_cnt_notsack_list #define tcp_notsack_list tcp_sack_info->tcp_notsack_list mblk_t *tcp_rcv_list; /* Queued until push, urgent data, */ mblk_t *tcp_rcv_last_head; /* optdata, or the count exceeds */ mblk_t *tcp_rcv_last_tail; /* tcp_rcv_push_wait. */ uint32_t tcp_rcv_cnt; /* tcp_rcv_list is b_next chain. */ uint32_t tcp_cwnd_ssthresh; /* Congestion window */ uint32_t tcp_cwnd_max; uint32_t tcp_csuna; /* Clear (no rexmits in window) suna */ clock_t tcp_rtt_sa; /* Round trip smoothed average */ clock_t tcp_rtt_sd; /* Round trip smoothed deviation */ clock_t tcp_rtt_update; /* Round trip update(s) */ clock_t tcp_ms_we_have_waited; /* Total retrans time */ uint32_t tcp_swl1; /* These help us avoid using stale */ uint32_t tcp_swl2; /* packets to update state */ uint32_t tcp_rack; /* Seq # we have acked */ uint32_t tcp_rack_cnt; /* # of segs we have deferred ack */ uint32_t tcp_rack_cur_max; /* # of segs we may defer ack for now */ uint32_t tcp_rack_abs_max; /* # of segs we may defer ack ever */ timeout_id_t tcp_ack_tid; /* Delayed ACK timer ID */ timeout_id_t tcp_push_tid; /* Push timer ID */ uint32_t tcp_max_swnd; /* Maximum swnd we have seen */ struct tcp_s *tcp_listener; /* Our listener */ int32_t tcp_xmit_lowater; /* Send buffer low water mark. */ uint32_t tcp_irs; /* Initial recv seq num */ uint32_t tcp_fss; /* Final/fin send seq num */ uint32_t tcp_urg; /* Urgent data seq num */ clock_t tcp_first_timer_threshold; /* When to prod IP */ clock_t tcp_second_timer_threshold; /* When to give up completely */ clock_t tcp_first_ctimer_threshold; /* 1st threshold while connecting */ clock_t tcp_second_ctimer_threshold; /* 2nd ... while connecting */ int tcp_lingertime; /* Close linger time (in seconds) */ uint32_t tcp_urp_last; /* Last urp for which signal sent */ mblk_t *tcp_urp_mp; /* T_EXDATA_IND for urgent byte */ mblk_t *tcp_urp_mark_mp; /* zero-length marked/unmarked msg */ int tcp_conn_req_cnt_q0; /* # of conn reqs in SYN_RCVD */ int tcp_conn_req_cnt_q; /* # of conn reqs in ESTABLISHED */ int tcp_conn_req_max; /* # of ESTABLISHED conn reqs allowed */ t_scalar_t tcp_conn_req_seqnum; /* Incrementing pending conn req ID */ #define tcp_ip_addr_cache tcp_reass_tail /* Cache ip addresses that */ /* complete the 3-way handshake */ kmutex_t tcp_eager_lock; struct tcp_s *tcp_eager_next_q; /* next eager in ESTABLISHED state */ struct tcp_s *tcp_eager_last_q; /* last eager in ESTABLISHED state */ struct tcp_s *tcp_eager_next_q0; /* next eager in SYN_RCVD state */ struct tcp_s *tcp_eager_prev_q0; /* prev eager in SYN_RCVD state */ /* all eagers form a circular list */ union { mblk_t *tcp_eager_conn_ind; /* T_CONN_IND waiting for 3rd ack. */ mblk_t *tcp_opts_conn_req; /* T_CONN_REQ w/ options processed */ } tcp_conn; uint32_t tcp_syn_rcvd_timeout; /* How many SYN_RCVD timeout in q0 */ /* TCP Keepalive Timer members */ int32_t tcp_ka_last_intrvl; /* Last probe interval */ timeout_id_t tcp_ka_tid; /* Keepalive timer ID */ uint32_t tcp_ka_interval; /* Keepalive interval */ uint32_t tcp_ka_abort_thres; /* Keepalive abort threshold */ int32_t tcp_client_errno; /* How the client screwed up */ char *tcp_iphc; /* Buffer holding tcp/ip hdr template */ int tcp_iphc_len; /* actual allocated buffer size */ int32_t tcp_hdr_len; /* Byte len of combined TCP/IP hdr */ ipha_t *tcp_ipha; /* IPv4 header in the buffer */ ip6_t *tcp_ip6h; /* IPv6 header in the buffer */ int tcp_ip_hdr_len; /* Byte len of our current IPvx hdr */ tcph_t *tcp_tcph; /* tcp header within combined hdr */ int32_t tcp_tcp_hdr_len; /* tcp header len within combined */ uint32_t tcp_sum; /* checksum to compensate for source */ /* routed packets. Host byte order */ uint16_t tcp_last_sent_len; /* Record length for nagle */ uint16_t tcp_dupack_cnt; /* # of consequtive duplicate acks */ kmutex_t *tcp_acceptor_lockp; /* Ptr to tf_lock */ timeout_id_t tcp_ordrelid; /* qbufcall/qtimeout id */ t_uscalar_t tcp_acceptor_id; /* ACCEPTOR_id */ int tcp_ipsec_overhead; /* * Address family that app wishes returned addrsses to be in. * Currently taken from address family used in T_BIND_REQ, but * should really come from family used in original socket() call. * Value can be AF_INET or AF_INET6. */ uint_t tcp_family; /* * used for a quick test to determine if any ancillary bits are * set */ uint_t tcp_ipv6_recvancillary; /* Flags */ #define TCP_IPV6_RECVPKTINFO 0x01 /* IPV6_RECVPKTINFO option */ #define TCP_IPV6_RECVHOPLIMIT 0x02 /* IPV6_RECVHOPLIMIT option */ #define TCP_IPV6_RECVHOPOPTS 0x04 /* IPV6_RECVHOPOPTS option */ #define TCP_IPV6_RECVDSTOPTS 0x08 /* IPV6_RECVDSTOPTS option */ #define TCP_IPV6_RECVRTHDR 0x10 /* IPV6_RECVRTHDR option */ #define TCP_IPV6_RECVRTDSTOPTS 0x20 /* IPV6_RECVRTHDRDSTOPTS option */ #define TCP_IPV6_RECVTCLASS 0x40 /* IPV6_RECVTCLASS option */ #define TCP_OLD_IPV6_RECVDSTOPTS 0x80 /* old IPV6_RECVDSTOPTS option */ uint_t tcp_recvifindex; /* Last received IPV6_RCVPKTINFO */ uint_t tcp_recvhops; /* Last received IPV6_RECVHOPLIMIT */ uint_t tcp_recvtclass; /* Last received IPV6_RECVTCLASS */ ip6_hbh_t *tcp_hopopts; /* Last received IPV6_RECVHOPOPTS */ ip6_dest_t *tcp_dstopts; /* Last received IPV6_RECVDSTOPTS */ ip6_dest_t *tcp_rtdstopts; /* Last recvd IPV6_RECVRTHDRDSTOPTS */ ip6_rthdr_t *tcp_rthdr; /* Last received IPV6_RECVRTHDR */ uint_t tcp_hopoptslen; uint_t tcp_dstoptslen; uint_t tcp_rtdstoptslen; uint_t tcp_rthdrlen; mblk_t *tcp_timercache; cred_t *tcp_cred; /* Credentials when this was opened */ pid_t tcp_cpid; /* Process id when this was opened */ union { struct { uchar_t v4_ttl; /* Dup of tcp_ipha.iph_type_of_service */ uchar_t v4_tos; /* Dup of tcp_ipha.iph_ttl */ } v4_hdr_info; struct { uint_t v6_vcf; /* Dup of tcp_ip6h.ip6h_vcf */ uchar_t v6_hops; /* Dup of tcp_ip6h.ip6h_hops */ } v6_hdr_info; } tcp_hdr_info; #define tcp_ttl tcp_hdr_info.v4_hdr_info.v4_ttl #define tcp_tos tcp_hdr_info.v4_hdr_info.v4_tos #define tcp_ip6_vcf tcp_hdr_info.v6_hdr_info.v6_vcf #define tcp_ip6_hops tcp_hdr_info.v6_hdr_info.v6_hops ushort_t tcp_ipversion; uint_t tcp_bound_if; /* IPV6_BOUND_IF */ #define tcp_lport tcp_connp->conn_lport #define tcp_fport tcp_connp->conn_fport #define tcp_ports tcp_connp->conn_ports #define tcp_remote tcp_connp->conn_rem #define tcp_ip_src tcp_connp->conn_src #define tcp_remote_v6 tcp_connp->conn_remv6 #define tcp_ip_src_v6 tcp_connp->conn_srcv6 #define tcp_bound_source_v6 tcp_connp->conn_bound_source_v6 #define tcp_bound_source tcp_connp->conn_bound_source kmutex_t tcp_closelock; kcondvar_t tcp_closecv; uint8_t tcp_closed; uint8_t tcp_closeflags; uint8_t tcp_cleandeathtag; mblk_t tcp_closemp; timeout_id_t tcp_linger_tid; /* Linger timer ID */ void *tcp_tracebuf; struct tcp_s *tcp_acceptor_hash; /* Acceptor hash chain */ struct tcp_s **tcp_ptpahn; /* Pointer to previous accept hash next. */ struct tcp_s *tcp_bind_hash; /* Bind hash chain */ struct tcp_s **tcp_ptpbhn; boolean_t tcp_ire_ill_check_done; uint_t tcp_maxpsz; /* * used for Multidata Transmit */ uint_t tcp_mdt_hdr_head; /* leading header fragment extra space */ uint_t tcp_mdt_hdr_tail; /* trailing header fragment extra space */ int tcp_mdt_max_pld; /* maximum payload buffers per Multidata */ uint32_t tcp_ofo_fin_seq; /* Recv out of order FIN seq num */ uint32_t tcp_cwr_snd_max; uint_t tcp_drop_opt_ack_cnt; /* # tcp generated optmgmt */ ip6_pkt_t tcp_sticky_ipp; /* Sticky options */ #define tcp_ipp_fields tcp_sticky_ipp.ipp_fields /* valid fields */ #define tcp_ipp_ifindex tcp_sticky_ipp.ipp_ifindex /* pktinfo ifindex */ #define tcp_ipp_addr tcp_sticky_ipp.ipp_addr /* pktinfo src/dst addr */ #define tcp_ipp_hoplimit tcp_sticky_ipp.ipp_hoplimit #define tcp_ipp_hopoptslen tcp_sticky_ipp.ipp_hopoptslen #define tcp_ipp_rtdstoptslen tcp_sticky_ipp.ipp_rtdstoptslen #define tcp_ipp_rthdrlen tcp_sticky_ipp.ipp_rthdrlen #define tcp_ipp_dstoptslen tcp_sticky_ipp.ipp_dstoptslen #define tcp_ipp_hopopts tcp_sticky_ipp.ipp_hopopts #define tcp_ipp_rtdstopts tcp_sticky_ipp.ipp_rtdstopts #define tcp_ipp_rthdr tcp_sticky_ipp.ipp_rthdr #define tcp_ipp_dstopts tcp_sticky_ipp.ipp_dstopts #define tcp_ipp_nexthop tcp_sticky_ipp.ipp_nexthop #define tcp_ipp_use_min_mtu tcp_sticky_ipp.ipp_use_min_mtu struct tcp_s *tcp_saved_listener; /* saved value of listener */ uint32_t tcp_in_ack_unsent; /* ACK for unsent data cnt. */ /* * The following fusion-related fields are protected by squeue. */ struct tcp_s *tcp_loopback_peer; /* peer tcp for loopback */ mblk_t *tcp_fused_sigurg_mp; /* M_PCSIG mblk for SIGURG */ size_t tcp_fuse_rcv_hiwater; /* fusion receive queue size */ uint_t tcp_fuse_rcv_unread_hiwater; /* max # of outstanding pkts */ /* * The following fusion-related fields and bit fields are to be * manipulated with squeue protection or with tcp_fuse_lock held. */ kmutex_t tcp_fuse_lock; uint_t tcp_fuse_rcv_unread_cnt; /* # of outstanding pkts */ uint32_t tcp_fused : 1, /* loopback tcp in fusion mode */ tcp_unfusable : 1, /* fusion not allowed on endpoint */ tcp_fused_sigurg : 1, /* send SIGURG upon draining */ tcp_direct_sockfs : 1, /* direct calls to sockfs */ tcp_fuse_syncstr_stopped : 1, /* synchronous streams stopped */ tcp_fuse_to_bit_31 : 27; /* * This variable is accessed without any lock protection * and therefore must not be declared as a bit field along * with the rest which require such condition. */ boolean_t tcp_issocket; /* this is a socket tcp */ uint32_t tcp_squeue_bytes; /* * Kernel SSL session information */ boolean_t tcp_kssl_pending; /* waiting for 1st SSL rec. */ boolean_t tcp_kssl_inhandshake; /* during SSL handshake */ kssl_ent_t tcp_kssl_ent; /* SSL table entry */ kssl_ctx_t tcp_kssl_ctx; /* SSL session */ uint_t tcp_label_len; /* length of cached label */ } tcp_t; extern void tcp_free(tcp_t *tcp); extern void tcp_ddi_init(void); extern void tcp_ddi_destroy(void); extern void tcp_xmit_listeners_reset(mblk_t *mp, uint_t ip_hdr_len); extern void tcp_conn_request(void *arg, mblk_t *mp, void *arg2); extern void tcp_conn_request_unbound(void *arg, mblk_t *mp, void *arg2); extern void tcp_input(void *arg, mblk_t *mp, void *arg2); extern void tcp_rput_data(void *arg, mblk_t *mp, void *arg2); extern void *tcp_get_conn(void *arg); extern void tcp_time_wait_collector(void *arg); extern int tcp_snmp_get(queue_t *, mblk_t *); extern int tcp_snmp_set(queue_t *, int, int, uchar_t *, int len); /* * The TCP Fanout structure. * The hash tables and their linkage (tcp_*_hash_next, tcp_ptp*hn) are * protected by the per-bucket tf_lock. Each tcp_t * inserted in the list points back at this lock using tcp_*_lockp. * * The listener and acceptor hash queues are lists of tcp_t. */ /* listener hash and acceptor hash queue head */ typedef struct tf_s { tcp_t *tf_tcp; kmutex_t tf_lock; } tf_t; extern mib2_tcp_t tcp_mib; #endif /* (defined(_KERNEL) || defined(_KMEMUSER)) */ /* Contract private interface between TCP and Clustering. */ #define CL_TCPI_V1 1 /* cl_tcpi_version number */ typedef struct cl_tcp_info_s { ushort_t cl_tcpi_version; /* cl_tcp_info_t's version no */ ushort_t cl_tcpi_ipversion; /* IP version */ int32_t cl_tcpi_state; /* TCP state */ in_port_t cl_tcpi_lport; /* Local port */ in_port_t cl_tcpi_fport; /* Remote port */ in6_addr_t cl_tcpi_laddr_v6; /* Local IP address */ in6_addr_t cl_tcpi_faddr_v6; /* Remote IP address */ #ifdef _KERNEL /* Note: V4_PART_OF_V6 is meant to be used only for _KERNEL defined stuff */ #define cl_tcpi_laddr V4_PART_OF_V6(cl_tcpi_laddr_v6) #define cl_tcpi_faddr V4_PART_OF_V6(cl_tcpi_faddr_v6) #endif /* _KERNEL */ } cl_tcp_info_t; /* * Hook functions to enable cluster networking * On non-clustered systems these vectors must always be NULL. */ extern void (*cl_inet_listen)(uint8_t, sa_family_t, uint8_t *, in_port_t); extern void (*cl_inet_unlisten)(uint8_t, sa_family_t, uint8_t *, in_port_t); /* * Contracted Consolidation Private ioctl for aborting TCP connections. * In order to keep the offsets and size of the structure the same between * a 32-bit application and a 64-bit amd64 kernel, we use a #pragma * pack(4). */ #define TCP_IOC_ABORT_CONN (('T' << 8) + 91) #if _LONG_LONG_ALIGNMENT == 8 && _LONG_LONG_ALIGNMENT_32 == 4 #pragma pack(4) #endif typedef struct tcp_ioc_abort_conn_s { struct sockaddr_storage ac_local; /* local addr and port */ struct sockaddr_storage ac_remote; /* remote addr and port */ int32_t ac_start; /* start state */ int32_t ac_end; /* end state */ int32_t ac_zoneid; /* zoneid */ } tcp_ioc_abort_conn_t; #if _LONG_LONG_ALIGNMENT == 8 && _LONG_LONG_ALIGNMENT_32 == 4 #pragma pack() #endif #if (defined(_KERNEL) || defined(_KMEMUSER)) extern void tcp_rput_other(tcp_t *tcp, mblk_t *mp); #endif #ifdef __cplusplus } #endif #endif /* _INET_TCP_H */