/*      $NetBSD: tcp_timer.c,v 1.99 2022/11/04 09:00:58 ozaki-r Exp $   */

/*
* Copyright (C) 1995, 1996, 1997, and 1998 WIDE Project.
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
*    notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
*    notice, this list of conditions and the following disclaimer in the
*    documentation and/or other materials provided with the distribution.
* 3. Neither the name of the project nor the names of its contributors
*    may be used to endorse or promote products derived from this software
*    without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE PROJECT AND CONTRIBUTORS ``AS IS'' AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED.  IN NO EVENT SHALL THE PROJECT OR CONTRIBUTORS BE LIABLE
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*/

/*-
* Copyright (c) 1997, 1998, 2001, 2005 The NetBSD Foundation, Inc.
* All rights reserved.
*
* This code is derived from software contributed to The NetBSD Foundation
* by Jason R. Thorpe and Kevin M. Lahey of the Numerical Aerospace Simulation
* Facility, NASA Ames Research Center.
* This code is derived from software contributed to The NetBSD Foundation
* by Charles M. Hannum.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
*    notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
*    notice, this list of conditions and the following disclaimer in the
*    documentation and/or other materials provided with the distribution.
*
* THIS SOFTWARE IS PROVIDED BY THE NETBSD FOUNDATION, INC. AND CONTRIBUTORS
* ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
* TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
* PURPOSE ARE DISCLAIMED.  IN NO EVENT SHALL THE FOUNDATION OR CONTRIBUTORS
* BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
* CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
* SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
* INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
* CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
* ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
* POSSIBILITY OF SUCH DAMAGE.
*/

/*
* Copyright (c) 1982, 1986, 1988, 1990, 1993, 1995
*      The Regents of the University of California.  All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
*    notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
*    notice, this list of conditions and the following disclaimer in the
*    documentation and/or other materials provided with the distribution.
* 3. Neither the name of the University nor the names of its contributors
*    may be used to endorse or promote products derived from this software
*    without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*
*      @(#)tcp_timer.c 8.2 (Berkeley) 5/24/95
*/

#include <sys/cdefs.h>
__KERNEL_RCSID(0, "$NetBSD: tcp_timer.c,v 1.99 2022/11/04 09:00:58 ozaki-r Exp $");

#ifdef _KERNEL_OPT
#include "opt_inet.h"
#include "opt_tcp_debug.h"
#include "opt_net_mpsafe.h"
#endif

#include <sys/param.h>
#include <sys/systm.h>
#include <sys/mbuf.h>
#include <sys/socket.h>
#include <sys/socketvar.h>
#include <sys/protosw.h>
#include <sys/errno.h>
#include <sys/kernel.h>
#include <sys/callout.h>
#include <sys/workqueue.h>
#include <sys/cprng.h>

#include <net/if.h>

#include <netinet/in.h>
#include <netinet/in_systm.h>
#include <netinet/ip.h>
#include <netinet/in_pcb.h>
#include <netinet/ip_var.h>
#include <netinet/ip_icmp.h>

#ifdef INET6
#include <netinet/ip6.h>
#include <netinet6/in6_pcb.h>
#endif

#include <netinet/tcp.h>
#include <netinet/tcp_fsm.h>
#include <netinet/tcp_seq.h>
#include <netinet/tcp_timer.h>
#include <netinet/tcp_var.h>
#include <netinet/tcp_private.h>
#include <netinet/tcp_congctl.h>
#ifdef TCP_DEBUG
#include <netinet/tcp_debug.h>
#endif

/*
* Various tunable timer parameters.  These are initialized in tcp_init(),
* unless they are patched.
*/
u_int   tcp_keepinit = 0;
u_int   tcp_keepidle = 0;
u_int   tcp_keepintvl = 0;
u_int   tcp_keepcnt = 0;                /* max idle probes */

int     tcp_maxpersistidle = 0;         /* max idle time in persist */

static callout_t        tcp_slowtimo_ch;
#ifdef NET_MPSAFE
static struct workqueue *tcp_slowtimo_wq;
static struct work      tcp_slowtimo_wk;
#endif

static void tcp_slowtimo_work(struct work *, void *);
static void tcp_slowtimo(void *);

/*
* Time to delay the ACK.  This is initialized in tcp_init(), unless
* its patched.
*/
int     tcp_delack_ticks = 0;

void    tcp_timer_rexmt(void *);
void    tcp_timer_persist(void *);
void    tcp_timer_keep(void *);
void    tcp_timer_2msl(void *);

const tcp_timer_func_t tcp_timer_funcs[TCPT_NTIMERS] = {
       tcp_timer_rexmt,
       tcp_timer_persist,
       tcp_timer_keep,
       tcp_timer_2msl,
};

/*
* Timer state initialization, called from tcp_init().
*/
void
tcp_timer_init(void)
{

       if (tcp_keepinit == 0)
               tcp_keepinit = TCPTV_KEEP_INIT;

       if (tcp_keepidle == 0)
               tcp_keepidle = TCPTV_KEEP_IDLE;

       if (tcp_keepintvl == 0)
               tcp_keepintvl = TCPTV_KEEPINTVL;

       if (tcp_keepcnt == 0)
               tcp_keepcnt = TCPTV_KEEPCNT;

       if (tcp_maxpersistidle == 0)
               tcp_maxpersistidle = TCPTV_KEEP_IDLE;

       if (tcp_delack_ticks == 0)
               tcp_delack_ticks = TCP_DELACK_TICKS;
}

void
tcp_slowtimo_init(void)
{
#ifdef NET_MPSAFE
       int error;

       error = workqueue_create(&tcp_slowtimo_wq, "tcp_slowtimo",
           tcp_slowtimo_work, NULL, PRI_SOFTNET, IPL_SOFTNET, WQ_MPSAFE);
       if (error != 0)
               panic("%s: workqueue_create failed (%d)\n", __func__, error);
#endif
       callout_init(&tcp_slowtimo_ch, CALLOUT_MPSAFE);
       callout_reset(&tcp_slowtimo_ch, 1, tcp_slowtimo, NULL);
}

/*
* Callout to process delayed ACKs for a TCPCB.
*/
void
tcp_delack(void *arg)
{
       struct tcpcb *tp = arg;

       /*
        * If tcp_output() wasn't able to transmit the ACK
        * for whatever reason, it will restart the delayed
        * ACK callout.
        */

       mutex_enter(softnet_lock);
       if ((tp->t_flags & (TF_DEAD | TF_DELACK)) != TF_DELACK) {
               mutex_exit(softnet_lock);
               return;
       }
       if (!callout_expired(&tp->t_delack_ch)) {
               mutex_exit(softnet_lock);
               return;
       }

       tp->t_flags |= TF_ACKNOW;
       KERNEL_LOCK(1, NULL);
       (void) tcp_output(tp);
       KERNEL_UNLOCK_ONE(NULL);
       mutex_exit(softnet_lock);
}

/*
* Tcp protocol timeout routine called every 500 ms.
* Updates the timers in all active tcb's and
* causes finite state machine actions if timers expire.
*/
static void
tcp_slowtimo_work(struct work *wk, void *arg)
{

       mutex_enter(softnet_lock);
       tcp_iss_seq += TCP_ISSINCR + (TCP_ISS_RANDOM_MASK & cprng_fast32());
       tcp_now++;                                      /* for timestamps */
       mutex_exit(softnet_lock);

       callout_schedule(&tcp_slowtimo_ch, hz / PR_SLOWHZ);
}

static void
tcp_slowtimo(void *arg)
{

#ifdef NET_MPSAFE
       workqueue_enqueue(tcp_slowtimo_wq, &tcp_slowtimo_wk, NULL);
#else
       tcp_slowtimo_work(NULL, NULL);
#endif
}

/*
* Cancel all timers for TCP tp.
*/
void
tcp_canceltimers(struct tcpcb *tp)
{
       int i;

       for (i = 0; i < TCPT_NTIMERS; i++)
               TCP_TIMER_DISARM(tp, i);
}

const int       tcp_backoff[TCP_MAXRXTSHIFT + 1] =
   { 1, 2, 4, 8, 16, 32, 64, 64, 64, 64, 64, 64, 64 };

const int       tcp_totbackoff = 511;   /* sum of tcp_backoff[] */

/*
* TCP timer processing.
*/

void
tcp_timer_rexmt(void *arg)
{
       struct tcpcb *tp = arg;
       uint32_t rto;
#ifdef TCP_DEBUG
       struct socket *so = NULL;
       short ostate;
#endif

       mutex_enter(softnet_lock);
       if ((tp->t_flags & TF_DEAD) != 0) {
               mutex_exit(softnet_lock);
               return;
       }
       if (!callout_expired(&tp->t_timer[TCPT_REXMT])) {
               mutex_exit(softnet_lock);
               return;
       }

       KERNEL_LOCK(1, NULL);
       if ((tp->t_flags & TF_PMTUD_PEND) && tp->t_inpcb &&
           SEQ_GEQ(tp->t_pmtud_th_seq, tp->snd_una) &&
           SEQ_LT(tp->t_pmtud_th_seq, (int)(tp->snd_una + tp->t_ourmss))) {
               extern struct sockaddr_in icmpsrc;
               struct icmp icmp;

               tp->t_flags &= ~TF_PMTUD_PEND;

               /* XXX create fake icmp message with relevant entries */
               icmp.icmp_nextmtu = tp->t_pmtud_nextmtu;
               icmp.icmp_ip.ip_len = tp->t_pmtud_ip_len;
               icmp.icmp_ip.ip_hl = tp->t_pmtud_ip_hl;
               icmpsrc.sin_addr = in4p_faddr(tp->t_inpcb);
               icmp_mtudisc(&icmp, icmpsrc.sin_addr);

               /*
                * Notify all connections to the same peer about
                * new mss and trigger retransmit.
                */
               inpcb_notifyall(&tcbtable, icmpsrc.sin_addr, EMSGSIZE,
                   tcp_mtudisc);
               KERNEL_UNLOCK_ONE(NULL);
               mutex_exit(softnet_lock);
               return;
       }
#ifdef TCP_DEBUG
       so = tp->t_inpcb->inp_socket;
       ostate = tp->t_state;
#endif /* TCP_DEBUG */

       /*
        * Clear the SACK scoreboard, reset FACK estimate.
        */
       tcp_free_sackholes(tp);
       tp->snd_fack = tp->snd_una;

       /*
        * Retransmission timer went off.  Message has not
        * been acked within retransmit interval.  Back off
        * to a longer retransmit interval and retransmit one segment.
        */

       if (++tp->t_rxtshift > TCP_MAXRXTSHIFT) {
               tp->t_rxtshift = TCP_MAXRXTSHIFT;
               TCP_STATINC(TCP_STAT_TIMEOUTDROP);
               tp = tcp_drop(tp, tp->t_softerror ?
                   tp->t_softerror : ETIMEDOUT);
               goto out;
       }
       TCP_STATINC(TCP_STAT_REXMTTIMEO);
       rto = TCP_REXMTVAL(tp);
       if (rto < tp->t_rttmin)
               rto = tp->t_rttmin;
       TCPT_RANGESET(tp->t_rxtcur, rto * tcp_backoff[tp->t_rxtshift],
           tp->t_rttmin, TCPTV_REXMTMAX);
       TCP_TIMER_ARM(tp, TCPT_REXMT, tp->t_rxtcur);

       /*
        * If we are losing and we are trying path MTU discovery,
        * try turning it off.  This will avoid black holes in
        * the network which suppress or fail to send "packet
        * too big" ICMP messages.  We should ideally do
        * lots more sophisticated searching to find the right
        * value here...
        */
       if (tp->t_mtudisc && tp->t_rxtshift > TCP_MAXRXTSHIFT / 6) {
               TCP_STATINC(TCP_STAT_PMTUBLACKHOLE);

               /* try turning PMTUD off or try using IPv6 minimum MTU */
               tp->t_mtudisc = 0;

               /* XXX: more sophisticated Black hole recovery code? */
       }

       /*
        * If losing, let the lower level know and try for
        * a better route.  Also, if we backed off this far,
        * our srtt estimate is probably bogus.  Clobber it
        * so we'll take the next rtt measurement as our srtt;
        * move the current srtt into rttvar to keep the current
        * retransmit times until then.
        */
       if (tp->t_rxtshift > TCP_MAXRXTSHIFT / 4) {
               inpcb_losing(tp->t_inpcb);
               /*
                * This operation is not described in RFC2988.  The
                * point is to keep srtt+4*rttvar constant, so we
                * should shift right 2 bits to divide by 4, and then
                * shift right one bit because the storage
                * representation of rttvar is 1/16s vs 1/32s for
                * srtt.
                */
               tp->t_rttvar += (tp->t_srtt >> TCP_RTT_SHIFT);
               tp->t_srtt = 0;
       }
       tp->snd_nxt = tp->snd_una;
       tp->snd_high = tp->snd_max;
       /*
        * If timing a segment in this window, stop the timer.
        */
       tp->t_rtttime = 0;
       /*
        * Remember if we are retransmitting a SYN, because if
        * we do, set the initial congestion window must be set
        * to 1 segment.
        */
       if (tp->t_state == TCPS_SYN_SENT)
               tp->t_flags |= TF_SYN_REXMT;

       /*
        * Adjust congestion control parameters.
        */
       tp->t_congctl->slow_retransmit(tp);

       (void) tcp_output(tp);

out:
#ifdef TCP_DEBUG
       if (tp && so->so_options & SO_DEBUG)
               tcp_trace(TA_USER, ostate, tp, NULL,
                   PRU_SLOWTIMO | (TCPT_REXMT << 8));
#endif
       KERNEL_UNLOCK_ONE(NULL);
       mutex_exit(softnet_lock);
}

void
tcp_timer_persist(void *arg)
{
       struct tcpcb *tp = arg;
       uint32_t rto;
#ifdef TCP_DEBUG
       struct socket *so = NULL;
       short ostate;
#endif

       mutex_enter(softnet_lock);
       if ((tp->t_flags & TF_DEAD) != 0) {
               mutex_exit(softnet_lock);
               return;
       }
       if (!callout_expired(&tp->t_timer[TCPT_PERSIST])) {
               mutex_exit(softnet_lock);
               return;
       }

       KERNEL_LOCK(1, NULL);
#ifdef TCP_DEBUG
       so = tp->t_inpcb->inp_socket;
       ostate = tp->t_state;
#endif /* TCP_DEBUG */

       /*
        * Persistance timer into zero window.
        * Force a byte to be output, if possible.
        */

       /*
        * Hack: if the peer is dead/unreachable, we do not
        * time out if the window is closed.  After a full
        * backoff, drop the connection if the idle time
        * (no responses to probes) reaches the maximum
        * backoff that we would use if retransmitting.
        */
       rto = TCP_REXMTVAL(tp);
       if (rto < tp->t_rttmin)
               rto = tp->t_rttmin;
       if (tp->t_rxtshift == TCP_MAXRXTSHIFT &&
           ((tcp_now - tp->t_rcvtime) >= tcp_maxpersistidle ||
           (tcp_now - tp->t_rcvtime) >= rto * tcp_totbackoff)) {
               TCP_STATINC(TCP_STAT_PERSISTDROPS);
               tp = tcp_drop(tp, ETIMEDOUT);
               goto out;
       }
       TCP_STATINC(TCP_STAT_PERSISTTIMEO);
       tcp_setpersist(tp);
       tp->t_force = 1;
       (void) tcp_output(tp);
       tp->t_force = 0;

out:
#ifdef TCP_DEBUG
       if (tp && so->so_options & SO_DEBUG)
               tcp_trace(TA_USER, ostate, tp, NULL,
                   PRU_SLOWTIMO | (TCPT_PERSIST << 8));
#endif
       KERNEL_UNLOCK_ONE(NULL);
       mutex_exit(softnet_lock);
}

void
tcp_timer_keep(void *arg)
{
       struct tcpcb *tp = arg;
       struct socket *so = NULL;       /* Quell compiler warning */
#ifdef TCP_DEBUG
       short ostate;
#endif

       mutex_enter(softnet_lock);
       if ((tp->t_flags & TF_DEAD) != 0) {
               mutex_exit(softnet_lock);
               return;
       }
       if (!callout_expired(&tp->t_timer[TCPT_KEEP])) {
               mutex_exit(softnet_lock);
               return;
       }

       KERNEL_LOCK(1, NULL);

#ifdef TCP_DEBUG
       ostate = tp->t_state;
#endif /* TCP_DEBUG */

       /*
        * Keep-alive timer went off; send something
        * or drop connection if idle for too long.
        */

       TCP_STATINC(TCP_STAT_KEEPTIMEO);
       if (TCPS_HAVEESTABLISHED(tp->t_state) == 0)
               goto dropit;
       so = tp->t_inpcb->inp_socket;
       KASSERT(so != NULL);
       if (so->so_options & SO_KEEPALIVE &&
           tp->t_state <= TCPS_CLOSE_WAIT) {
               if ((tp->t_maxidle > 0) &&
                   ((tcp_now - tp->t_rcvtime) >=
                    tp->t_keepidle + tp->t_maxidle))
                       goto dropit;
               /*
                * Send a packet designed to force a response
                * if the peer is up and reachable:
                * either an ACK if the connection is still alive,
                * or an RST if the peer has closed the connection
                * due to timeout or reboot.
                * Using sequence number tp->snd_una-1
                * causes the transmitted zero-length segment
                * to lie outside the receive window;
                * by the protocol spec, this requires the
                * correspondent TCP to respond.
                */
               TCP_STATINC(TCP_STAT_KEEPPROBE);

               (void)tcp_respond(tp, tp->t_template,
                   NULL, NULL, tp->rcv_nxt,
                   tp->snd_una - 1, 0);

               TCP_TIMER_ARM(tp, TCPT_KEEP, tp->t_keepintvl);
       } else
               TCP_TIMER_ARM(tp, TCPT_KEEP, tp->t_keepidle);

#ifdef TCP_DEBUG
       if (tp && so->so_options & SO_DEBUG)
               tcp_trace(TA_USER, ostate, tp, NULL,
                   PRU_SLOWTIMO | (TCPT_KEEP << 8));
#endif
       KERNEL_UNLOCK_ONE(NULL);
       mutex_exit(softnet_lock);
       return;

dropit:
       TCP_STATINC(TCP_STAT_KEEPDROPS);
       (void) tcp_drop(tp, ETIMEDOUT);
       KERNEL_UNLOCK_ONE(NULL);
       mutex_exit(softnet_lock);
}

void
tcp_timer_2msl(void *arg)
{
       struct tcpcb *tp = arg;
#ifdef TCP_DEBUG
       struct socket *so = NULL;
       short ostate;
#endif

       mutex_enter(softnet_lock);
       if ((tp->t_flags & TF_DEAD) != 0) {
               mutex_exit(softnet_lock);
               return;
       }
       if (!callout_expired(&tp->t_timer[TCPT_2MSL])) {
               mutex_exit(softnet_lock);
               return;
       }

       /*
        * 2 MSL timeout went off, clear the SACK scoreboard, reset
        * the FACK estimate.
        */
       KERNEL_LOCK(1, NULL);
       tcp_free_sackholes(tp);
       tp->snd_fack = tp->snd_una;

#ifdef TCP_DEBUG
       so = tp->t_inpcb->inp_socket;
       ostate = tp->t_state;
#endif /* TCP_DEBUG */

       /*
        * 2 MSL timeout in shutdown went off.  If we're closed but
        * still waiting for peer to close and connection has been idle
        * too long, or if 2MSL time is up from TIME_WAIT, delete connection
        * control block.  Otherwise, check again in a bit.
        */
       if (tp->t_state != TCPS_TIME_WAIT &&
           ((tp->t_maxidle == 0) ||
           ((tcp_now - tp->t_rcvtime) <= tp->t_maxidle)))
           TCP_TIMER_ARM(tp, TCPT_2MSL, tp->t_keepintvl);
       else
               tp = tcp_close(tp);

#ifdef TCP_DEBUG
       if (tp && so->so_options & SO_DEBUG)
               tcp_trace(TA_USER, ostate, tp, NULL,
                   PRU_SLOWTIMO | (TCPT_2MSL << 8));
#endif
       KERNEL_UNLOCK_ONE(NULL);
       mutex_exit(softnet_lock);
}