/*
* 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);
}
}
}
/*
* Partial ack handling within a sack recovery episode. Keeping this very
* simple for now. When a partial ack is received, force snd_cwnd to a value
* that will allow the sender to transmit no more than 2 segments. If
* necessary, a better scheme can be adopted at a later point, but for now,
* the goal is to prevent the sender from bursting a large amount of data in
* the midst of sack recovery.
*/
void
tcp_sack_partialack(struct tcpcb *tp, struct tcphdr *th)
{
int num_segs = 1;
//ScenSim-Port// INP_WLOCK_ASSERT(tp->t_inpcb);
tcp_timer_activate(tp, TT_REXMT, 0);
tp->t_rtttime = 0;
/* Send one or 2 segments based on how much new data was acked. */
if ((BYTES_THIS_ACK(tp, th) / tp->t_maxseg) >= 2)
num_segs = 2;
tp->snd_cwnd = (tp->sackhint.sack_bytes_rexmit +
(tp->snd_nxt - tp->sack_newdata) + num_segs * tp->t_maxseg);
if (tp->snd_cwnd > tp->snd_ssthresh)
tp->snd_cwnd = tp->snd_ssthresh;
assert(curvnet->ctx != 0); //ScenSim-Port//
UpdateCwndForStatistics(curvnet->ctx, tp->snd_cwnd); //ScenSim-Port//
tp->t_flags |= TF_ACKNOW;
(void) tcp_output(tp);
}
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;
}
