/*
* Initiate (or continue) disconnect.
* If embryonic state, just send reset (once).
* If in ``let data drain'' option and linger null, just drop.
* Otherwise (hard), mark socket disconnecting and drop
* current input data; switch states based on user close, and
* send segment to peer (with FIN).
*/
static void
tcp_disconnect(struct tcpcb *tp)
{
struct inpcb *inp = tp->t_inpcb;
struct socket *so = inp->inp_socket;
INP_INFO_WLOCK_ASSERT(&V_tcbinfo);
INP_WLOCK_ASSERT(inp);
/*
* Neither tcp_close() nor tcp_drop() should return NULL, as the
* socket is still open.
*/
if (tp->t_state < TCPS_ESTABLISHED) {
tp = tcp_close(tp);
KASSERT(tp != NULL,
("tcp_disconnect: tcp_close() returned NULL"));
} else if ((so->so_options & SO_LINGER) && so->so_linger == 0) {
tp = tcp_drop(tp, 0);
KASSERT(tp != NULL,
("tcp_disconnect: tcp_drop() returned NULL"));
} else {
soisdisconnecting(so);
sbflush(&so->so_rcv);
tcp_usrclosed(tp);
if (!(inp->inp_flags & INP_DROPPED))
tcp_output_disconnect(tp);
}
}
static void
rip_delhash(struct inpcb *inp)
{
INP_INFO_WLOCK_ASSERT(inp->inp_pcbinfo);
INP_WLOCK_ASSERT(inp);
LIST_REMOVE(inp, inp_hash);
}
/*
* tcp_detach is called when the socket layer loses its final reference
* to the socket, be it a file descriptor reference, a reference from TCP,
* etc. At this point, there is only one case in which we will keep around
* inpcb state: time wait.
*
* This function can probably be re-absorbed back into tcp_usr_detach() now
* that there is a single detach path.
*/
static void
tcp_detach(struct socket *so, struct inpcb *inp)
{
struct tcpcb *tp;
INP_INFO_WLOCK_ASSERT(&V_tcbinfo);
INP_WLOCK_ASSERT(inp);
KASSERT(so->so_pcb == inp, ("tcp_detach: so_pcb != inp"));
KASSERT(inp->inp_socket == so, ("tcp_detach: inp_socket != so"));
tp = intotcpcb(inp);
if (inp->inp_flags & INP_TIMEWAIT) {
/*
* There are two cases to handle: one in which the time wait
* state is being discarded (INP_DROPPED), and one in which
* this connection will remain in timewait. In the former,
* it is time to discard all state (except tcptw, which has
* already been discarded by the timewait close code, which
* should be further up the call stack somewhere). In the
* latter case, we detach from the socket, but leave the pcb
* present until timewait ends.
*
* XXXRW: Would it be cleaner to free the tcptw here?
*/
if (inp->inp_flags & INP_DROPPED) {
KASSERT(tp == NULL, ("tcp_detach: INP_TIMEWAIT && "
"INP_DROPPED && tp != NULL"));
in_pcbdetach(inp);
in_pcbfree(inp);
} else {
in_pcbdetach(inp);
INP_WUNLOCK(inp);
}
} else {
/*
* If the connection is not in timewait, we consider two
* two conditions: one in which no further processing is
* necessary (dropped || embryonic), and one in which TCP is
* not yet done, but no longer requires the socket, so the
* pcb will persist for the time being.
*
* XXXRW: Does the second case still occur?
*/
if (inp->inp_flags & INP_DROPPED ||
tp->t_state < TCPS_SYN_SENT) {
tcp_discardcb(tp);
in_pcbdetach(inp);
in_pcbfree(inp);
} else
in_pcbdetach(inp);
}
}
/*
* Common subroutine to open a TCP connection to remote host specified
* by struct sockaddr_in in mbuf *nam. Call in_pcbbind to assign a local
* port number if needed. Call in_pcbconnect_setup to do the routing and
* to choose a local host address (interface). If there is an existing
* incarnation of the same connection in TIME-WAIT state and if the remote
* host was sending CC options and if the connection duration was < MSL, then
* truncate the previous TIME-WAIT state and proceed.
* Initialize connection parameters and enter SYN-SENT state.
*/
static int
tcp_connect(struct tcpcb *tp, struct sockaddr *nam, struct thread *td)
{
struct inpcb *inp = tp->t_inpcb, *oinp;
struct socket *so = inp->inp_socket;
struct in_addr laddr;
u_short lport;
int error;
INP_INFO_WLOCK_ASSERT(&V_tcbinfo);
INP_WLOCK_ASSERT(inp);
if (inp->inp_lport == 0) {
error = in_pcbbind(inp, (struct sockaddr *)0, td->td_ucred);
if (error)
return error;
}
/*
* Cannot simply call in_pcbconnect, because there might be an
* earlier incarnation of this same connection still in
* TIME_WAIT state, creating an ADDRINUSE error.
*/
laddr = inp->inp_laddr;
lport = inp->inp_lport;
error = in_pcbconnect_setup(inp, nam, &laddr.s_addr, &lport,
&inp->inp_faddr.s_addr, &inp->inp_fport, &oinp, td->td_ucred);
if (error && oinp == NULL)
return error;
if (oinp)
return EADDRINUSE;
inp->inp_laddr = laddr;
in_pcbrehash(inp);
/*
* Compute window scaling to request:
* Scale to fit into sweet spot. See tcp_syncache.c.
* XXX: This should move to tcp_output().
*/
while (tp->request_r_scale < TCP_MAX_WINSHIFT &&
(TCP_MAXWIN << tp->request_r_scale) < sb_max)
tp->request_r_scale++;
soisconnecting(so);
TCPSTAT_INC(tcps_connattempt);
tp->t_state = TCPS_SYN_SENT;
tcp_timer_activate(tp, TT_KEEP, tcp_keepinit);
tp->iss = tcp_new_isn(tp);
tp->t_bw_rtseq = tp->iss;
tcp_sendseqinit(tp);
return 0;
}
/*
* User issued close, and wish to trail through shutdown states:
* if never received SYN, just forget it. If got a SYN from peer,
* but haven't sent FIN, then go to FIN_WAIT_1 state to send peer a FIN.
* If already got a FIN from peer, then almost done; go to LAST_ACK
* state. In all other cases, have already sent FIN to peer (e.g.
* after PRU_SHUTDOWN), and just have to play tedious game waiting
* for peer to send FIN or not respond to keep-alives, etc.
* We can let the user exit from the close as soon as the FIN is acked.
*/
static void
tcp_usrclosed(struct tcpcb *tp)
{
INP_INFO_WLOCK_ASSERT(&V_tcbinfo);
INP_WLOCK_ASSERT(tp->t_inpcb);
switch (tp->t_state) {
case TCPS_LISTEN:
tcp_offload_listen_close(tp);
/* FALLTHROUGH */
case TCPS_CLOSED:
tp->t_state = TCPS_CLOSED;
tp = tcp_close(tp);
/*
* tcp_close() should never return NULL here as the socket is
* still open.
*/
KASSERT(tp != NULL,
("tcp_usrclosed: tcp_close() returned NULL"));
break;
case TCPS_SYN_SENT:
case TCPS_SYN_RECEIVED:
tp->t_flags |= TF_NEEDFIN;
break;
case TCPS_ESTABLISHED:
tp->t_state = TCPS_FIN_WAIT_1;
break;
case TCPS_CLOSE_WAIT:
tp->t_state = TCPS_LAST_ACK;
break;
}
if (tp->t_state >= TCPS_FIN_WAIT_2) {
soisdisconnected(tp->t_inpcb->inp_socket);
/* Prevent the connection hanging in FIN_WAIT_2 forever. */
if (tp->t_state == TCPS_FIN_WAIT_2) {
int timeout;
timeout = (tcp_fast_finwait2_recycle) ?
tcp_finwait2_timeout : tcp_maxidle;
tcp_timer_activate(tp, TT_2MSL, timeout);
}
}
}
static void
rip_inshash(struct inpcb *inp)
{
struct inpcbinfo *pcbinfo = inp->inp_pcbinfo;
struct inpcbhead *pcbhash;
int hash;
INP_INFO_WLOCK_ASSERT(pcbinfo);
INP_WLOCK_ASSERT(inp);
if (inp->inp_ip_p != 0 &&
inp->inp_laddr.s_addr != INADDR_ANY &&
inp->inp_faddr.s_addr != INADDR_ANY) {
hash = INP_PCBHASH_RAW(inp->inp_ip_p, inp->inp_laddr.s_addr,
inp->inp_faddr.s_addr, pcbinfo->ipi_hashmask);
} else
hash = 0;
pcbhash = &pcbinfo->ipi_hashbase[hash];
LIST_INSERT_HEAD(pcbhash, inp, inp_hash);
}
/*
* Do a send by putting data in output queue and updating urgent
* marker if URG set. Possibly send more data. Unlike the other
* pru_*() routines, the mbuf chains are our responsibility. We
* must either enqueue them or free them. The other pru_* routines
* generally are caller-frees.
*/
static int
tcp_usr_send(struct socket *so, int flags, struct mbuf *m,
struct sockaddr *nam, struct mbuf *control, struct thread *td)
{
int error = 0;
struct inpcb *inp;
struct tcpcb *tp = NULL;
int headlocked = 0;
#ifdef INET6
int isipv6;
#endif
TCPDEBUG0;
/*
* We require the pcbinfo lock in two cases:
*
* (1) An implied connect is taking place, which can result in
* binding IPs and ports and hence modification of the pcb hash
* chains.
*
* (2) PRUS_EOF is set, resulting in explicit close on the send.
*/
if ((nam != NULL) || (flags & PRUS_EOF)) {
INP_INFO_WLOCK(&V_tcbinfo);
headlocked = 1;
}
inp = sotoinpcb(so);
KASSERT(inp != NULL, ("tcp_usr_send: inp == NULL"));
INP_WLOCK(inp);
if (inp->inp_flags & (INP_TIMEWAIT | INP_DROPPED)) {
if (control)
m_freem(control);
if (m)
m_freem(m);
error = ECONNRESET;
goto out;
}
#ifdef INET6
isipv6 = nam && nam->sa_family == AF_INET6;
#endif /* INET6 */
tp = intotcpcb(inp);
TCPDEBUG1();
if (control) {
/* TCP doesn't do control messages (rights, creds, etc) */
if (control->m_len) {
m_freem(control);
if (m)
m_freem(m);
error = EINVAL;
goto out;
}
m_freem(control); /* empty control, just free it */
}
if (!(flags & PRUS_OOB)) {
sbappendstream(&so->so_snd, m);
if (nam && tp->t_state < TCPS_SYN_SENT) {
/*
* Do implied connect if not yet connected,
* initialize window to default value, and
* initialize maxseg/maxopd using peer's cached
* MSS.
*/
INP_INFO_WLOCK_ASSERT(&V_tcbinfo);
#ifdef INET6
if (isipv6)
error = tcp6_connect(tp, nam, td);
else
#endif /* INET6 */
error = tcp_connect(tp, nam, td);
if (error)
goto out;
tp->snd_wnd = TTCP_CLIENT_SND_WND;
tcp_mss(tp, -1);
}
if (flags & PRUS_EOF) {
/*
* Close the send side of the connection after
* the data is sent.
*/
INP_INFO_WLOCK_ASSERT(&V_tcbinfo);
socantsendmore(so);
tcp_usrclosed(tp);
}
if (headlocked) {
INP_INFO_WUNLOCK(&V_tcbinfo);
headlocked = 0;
}
if (!(inp->inp_flags & INP_DROPPED)) {
if (flags & PRUS_MORETOCOME)
tp->t_flags |= TF_MORETOCOME;
error = tcp_output_send(tp);
if (flags & PRUS_MORETOCOME)
tp->t_flags &= ~TF_MORETOCOME;
}
} else {
/*
* XXXRW: PRUS_EOF not implemented with PRUS_OOB?
*/
SOCKBUF_LOCK(&so->so_snd);
if (sbspace(&so->so_snd) < -512) {
SOCKBUF_UNLOCK(&so->so_snd);
m_freem(m);
error = ENOBUFS;
goto out;
}
/*
* According to RFC961 (Assigned Protocols),
* the urgent pointer points to the last octet
* of urgent data. We continue, however,
* to consider it to indicate the first octet
* of data past the urgent section.
* Otherwise, snd_up should be one lower.
*/
sbappendstream_locked(&so->so_snd, m);
SOCKBUF_UNLOCK(&so->so_snd);
if (nam && tp->t_state < TCPS_SYN_SENT) {
/*
* Do implied connect if not yet connected,
* initialize window to default value, and
* initialize maxseg/maxopd using peer's cached
* MSS.
*/
INP_INFO_WLOCK_ASSERT(&V_tcbinfo);
#ifdef INET6
if (isipv6)
error = tcp6_connect(tp, nam, td);
else
#endif /* INET6 */
error = tcp_connect(tp, nam, td);
if (error)
goto out;
tp->snd_wnd = TTCP_CLIENT_SND_WND;
tcp_mss(tp, -1);
INP_INFO_WUNLOCK(&V_tcbinfo);
headlocked = 0;
} else if (nam) {
INP_INFO_WUNLOCK(&V_tcbinfo);
headlocked = 0;
}
tp->snd_up = tp->snd_una + so->so_snd.sb_cc;
tp->t_flags |= TF_FORCEDATA;
error = tcp_output_send(tp);
tp->t_flags &= ~TF_FORCEDATA;
}
out:
TCPDEBUG2((flags & PRUS_OOB) ? PRU_SENDOOB :
((flags & PRUS_EOF) ? PRU_SEND_EOF : PRU_SEND));
INP_WUNLOCK(inp);
if (headlocked)
INP_INFO_WUNLOCK(&V_tcbinfo);
return (error);
}
static int
tcp6_connect(struct tcpcb *tp, struct sockaddr *nam, struct thread *td)
{
struct inpcb *inp = tp->t_inpcb, *oinp;
struct socket *so = inp->inp_socket;
struct sockaddr_in6 *sin6 = (struct sockaddr_in6 *)nam;
struct in6_addr addr6;
int error;
INP_INFO_WLOCK_ASSERT(&V_tcbinfo);
INP_WLOCK_ASSERT(inp);
if (inp->inp_lport == 0) {
error = in6_pcbbind(inp, (struct sockaddr *)0, td->td_ucred);
if (error)
return error;
}
/*
* Cannot simply call in_pcbconnect, because there might be an
* earlier incarnation of this same connection still in
* TIME_WAIT state, creating an ADDRINUSE error.
* in6_pcbladdr() also handles scope zone IDs.
*/
error = in6_pcbladdr(inp, nam, &addr6);
if (error)
return error;
oinp = in6_pcblookup_hash(inp->inp_pcbinfo,
&sin6->sin6_addr, sin6->sin6_port,
IN6_IS_ADDR_UNSPECIFIED(&inp->in6p_laddr)
? &addr6
: &inp->in6p_laddr,
inp->inp_lport, 0, NULL);
if (oinp)
return EADDRINUSE;
if (IN6_IS_ADDR_UNSPECIFIED(&inp->in6p_laddr))
inp->in6p_laddr = addr6;
inp->in6p_faddr = sin6->sin6_addr;
inp->inp_fport = sin6->sin6_port;
/* update flowinfo - draft-itojun-ipv6-flowlabel-api-00 */
inp->inp_flow &= ~IPV6_FLOWLABEL_MASK;
if (inp->inp_flags & IN6P_AUTOFLOWLABEL)
inp->inp_flow |=
(htonl(ip6_randomflowlabel()) & IPV6_FLOWLABEL_MASK);
in_pcbrehash(inp);
/* Compute window scaling to request. */
while (tp->request_r_scale < TCP_MAX_WINSHIFT &&
(TCP_MAXWIN << tp->request_r_scale) < sb_max)
tp->request_r_scale++;
soisconnecting(so);
TCPSTAT_INC(tcps_connattempt);
tp->t_state = TCPS_SYN_SENT;
tcp_timer_activate(tp, TT_KEEP, tcp_keepinit);
tp->iss = tcp_new_isn(tp);
tp->t_bw_rtseq = tp->iss;
tcp_sendseqinit(tp);
return 0;
}