tcp.c

lwip-1.4.0

     we have to check the pcbs in TIME-WAIT state, also.
     We do not dump TIME_WAIT pcb's; they can still be matched by incoming
     packets using both local and remote IP addresses and ports to distinguish.
   */
  if ((pcb->so_options & SOF_REUSEADDR) != 0) {
    max_pcb_list = NUM_TCP_PCB_LISTS_NO_TIME_WAIT;
  }
#endif /* SO_REUSE */

  if (port == 0) {
    port = tcp_new_port();
  }

  /* Check if the address already is in use (on all lists) */
  for (i = 0; i < max_pcb_list; i++) {
    for(cpcb = *tcp_pcb_lists[i]; cpcb != NULL; cpcb = cpcb->next) {
      if (cpcb->local_port == port) {
#if SO_REUSE
        /* Omit checking for the same port if both pcbs have REUSEADDR set.
           For SO_REUSEADDR, the duplicate-check for a 5-tuple is done in
           tcp_connect. */
        if (((pcb->so_options & SOF_REUSEADDR) == 0) ||
          ((cpcb->so_options & SOF_REUSEADDR) == 0))
#endif /* SO_REUSE */
        {
          if (ip_addr_isany(&(cpcb->local_ip)) ||
              ip_addr_isany(ipaddr) ||
              ip_addr_cmp(&(cpcb->local_ip), ipaddr)) {
            return ERR_USE;
          }
        }
      }
    }
  }

  if (!ip_addr_isany(ipaddr)) {
    pcb->local_ip = *ipaddr;
  }
  pcb->local_port = port;
  TCP_REG(&tcp_bound_pcbs, pcb);
  LWIP_DEBUGF(TCP_DEBUG, ("tcp_bind: bind to port %"U16_F"\n", port));
  return ERR_OK;
}
#if LWIP_CALLBACK_API
/**
 * Default accept callback if no accept callback is specified by the user.
 */
static err_t
tcp_accept_null(void *arg, struct tcp_pcb *pcb, err_t err)
{
  LWIP_UNUSED_ARG(arg);
  LWIP_UNUSED_ARG(pcb);
  LWIP_UNUSED_ARG(err);

  return ERR_ABRT;
}
#endif /* LWIP_CALLBACK_API */

/**
 * Set the state of the connection to be LISTEN, which means that it
 * is able to accept incoming connections. The protocol control block
 * is reallocated in order to consume less memory. Setting the
 * connection to LISTEN is an irreversible process.
 *
 * @param pcb the original tcp_pcb
 * @param backlog the incoming connections queue limit
 * @return tcp_pcb used for listening, consumes less memory.
 *
 * @note The original tcp_pcb is freed. This function therefore has to be
 *       called like this:
 *             tpcb = tcp_listen(tpcb);
 */
struct tcp_pcb *
tcp_listen_with_backlog(struct tcp_pcb *pcb, u8_t backlog)
{
  struct tcp_pcb_listen *lpcb;

  LWIP_UNUSED_ARG(backlog);
  LWIP_ERROR("tcp_listen: pcb already connected", pcb->state == CLOSED, return NULL);

  /* already listening? */
  if (pcb->state == LISTEN) {
    return pcb;
  }
#if SO_REUSE
  if ((pcb->so_options & SOF_REUSEADDR) != 0) {
    /* Since SOF_REUSEADDR allows reusing a local address before the pcb's usage
       is declared (listen-/connection-pcb), we have to make sure now that
       this port is only used once for every local IP. */
    for(lpcb = tcp_listen_pcbs.listen_pcbs; lpcb != NULL; lpcb = lpcb->next) {
      if (lpcb->local_port == pcb->local_port) {
        if (ip_addr_cmp(&lpcb->local_ip, &pcb->local_ip)) {
          /* this address/port is already used */
          return NULL;
        }
      }
    }
  }
#endif /* SO_REUSE */
  lpcb = (struct tcp_pcb_listen *)memp_malloc(MEMP_TCP_PCB_LISTEN);
  if (lpcb == NULL) {
    return NULL;
  }
  lpcb->callback_arg = pcb->callback_arg;
  lpcb->local_port = pcb->local_port;
  lpcb->state = LISTEN;
  lpcb->prio = pcb->prio;
  lpcb->so_options = pcb->so_options;
  lpcb->so_options |= SOF_ACCEPTCONN;
  lpcb->ttl = pcb->ttl;
  lpcb->tos = pcb->tos;
  ip_addr_copy(lpcb->local_ip, pcb->local_ip);
  if (pcb->local_port != 0) {
    TCP_RMV(&tcp_bound_pcbs, pcb);
  }
  memp_free(MEMP_TCP_PCB, pcb);
#if LWIP_CALLBACK_API
  lpcb->accept = tcp_accept_null;
#endif /* LWIP_CALLBACK_API */
#if TCP_LISTEN_BACKLOG
  lpcb->accepts_pending = 0;
  lpcb->backlog = (backlog ? backlog : 1);
#endif /* TCP_LISTEN_BACKLOG */
  TCP_REG(&tcp_listen_pcbs.pcbs, (struct tcp_pcb *)lpcb);
  return (struct tcp_pcb *)lpcb;
}

/** 
 * Update the state that tracks the available window space to advertise.
 *
 * Returns how much extra window would be advertised if we sent an
 * update now.
 */
u32_t tcp_update_rcv_ann_wnd(struct tcp_pcb *pcb)
{
  u32_t new_right_edge = pcb->rcv_nxt + pcb->rcv_wnd;

  if (TCP_SEQ_GEQ(new_right_edge, pcb->rcv_ann_right_edge + LWIP_MIN((TCP_WND / 2), pcb->mss))) {
    /* we can advertise more window */
    pcb->rcv_ann_wnd = pcb->rcv_wnd;
    return new_right_edge - pcb->rcv_ann_right_edge;
  } else {
    if (TCP_SEQ_GT(pcb->rcv_nxt, pcb->rcv_ann_right_edge)) {
      /* Can happen due to other end sending out of advertised window,
       * but within actual available (but not yet advertised) window */
      pcb->rcv_ann_wnd = 0;
    } else {
      /* keep the right edge of window constant */
      u32_t new_rcv_ann_wnd = pcb->rcv_ann_right_edge - pcb->rcv_nxt;
      LWIP_ASSERT("new_rcv_ann_wnd <= 0xffff", new_rcv_ann_wnd <= 0xffff);
      pcb->rcv_ann_wnd = (u16_t)new_rcv_ann_wnd;
    }
    return 0;
  }
}

/**
 * This function should be called by the application when it has
 * processed the data. The purpose is to advertise a larger window
 * when the data has been processed.
 *
 * @param pcb the tcp_pcb for which data is read
 * @param len the amount of bytes that have been read by the application
 */
void
tcp_recved(struct tcp_pcb *pcb, u16_t len)
{
  int wnd_inflation;

  LWIP_ASSERT("tcp_recved: len would wrap rcv_wnd\n",
              len <= 0xffff - pcb->rcv_wnd );

  pcb->rcv_wnd += len;
  if (pcb->rcv_wnd > TCP_WND) {
    pcb->rcv_wnd = TCP_WND;
  }

  wnd_inflation = tcp_update_rcv_ann_wnd(pcb);

  /* If the change in the right edge of window is significant (default
   * watermark is TCP_WND/4), then send an explicit update now.
   * Otherwise wait for a packet to be sent in the normal course of
   * events (or more window to be available later) */
  if (wnd_inflation >= TCP_WND_UPDATE_THRESHOLD) {
    tcp_ack_now(pcb);
    tcp_output(pcb);
  }

  LWIP_DEBUGF(TCP_DEBUG, ("tcp_recved: recveived %"U16_F" bytes, wnd %"U16_F" (%"U16_F").\n",
         len, pcb->rcv_wnd, TCP_WND - pcb->rcv_wnd));
}

/**
 * A nastly hack featuring 'goto' statements that allocates a
 * new TCP local port.
 *
 * @return a new (free) local TCP port number
 */
static u16_t
tcp_new_port(void)
{
  int i;
  struct tcp_pcb *pcb;
#ifndef TCP_LOCAL_PORT_RANGE_START
/* From http://www.iana.org/assignments/port-numbers:
   "The Dynamic and/or Private Ports are those from 49152 through 65535" */
#define TCP_LOCAL_PORT_RANGE_START  0xc000
#define TCP_LOCAL_PORT_RANGE_END    0xffff
#endif
  static u16_t port = TCP_LOCAL_PORT_RANGE_START;
  
 again:
  if (port++ >= TCP_LOCAL_PORT_RANGE_END) {
    port = TCP_LOCAL_PORT_RANGE_START;
  }
  /* Check all PCB lists. */
  for (i = 0; i < NUM_TCP_PCB_LISTS; i++) {
    for(pcb = *tcp_pcb_lists[i]; pcb != NULL; pcb = pcb->next) {
      if (pcb->local_port == port) {
        goto again;
      }
    }
  }
  return port;
}

/**
 * Connects to another host. The function given as the "connected"
 * argument will be called when the connection has been established.
 *
 * @param pcb the tcp_pcb used to establish the connection
 * @param ipaddr the remote ip address to connect to
 * @param port the remote tcp port to connect to
 * @param connected callback function to call when connected (or on error)
 * @return ERR_VAL if invalid arguments are given
 *         ERR_OK if connect request has been sent
 *         other err_t values if connect request couldn't be sent
 */
err_t
tcp_connect(struct tcp_pcb *pcb, ip_addr_t *ipaddr, u16_t port,
      tcp_connected_fn connected)
{
  err_t ret;
  u32_t iss;
  u16_t old_local_port;

  LWIP_ERROR("tcp_connect: can only connect from state CLOSED", pcb->state == CLOSED, return ERR_ISCONN);

  LWIP_DEBUGF(TCP_DEBUG, ("tcp_connect to port %"U16_F"\n", port));
  if (ipaddr != NULL) {
    pcb->remote_ip = *ipaddr;
  } else {
    return ERR_VAL;
  }
  pcb->remote_port = port;

  /* check if we have a route to the remote host */
  if (ip_addr_isany(&(pcb->local_ip))) {
    /* no local IP address set, yet. */
    struct netif *netif = ip_route(&(pcb->remote_ip));
    if (netif == NULL) {
      /* Don't even try to send a SYN packet if we have no route
         since that will fail. */
      return ERR_RTE;
    }
    /* Use the netif's IP address as local address. */
    ip_addr_copy(pcb->local_ip, netif->ip_addr);
  }

  old_local_port = pcb->local_port;
  if (pcb->local_port == 0) {
    pcb->local_port = tcp_new_port();
  }
#if SO_REUSE
  if ((pcb->so_options & SOF_REUSEADDR) != 0) {
    /* Since SOF_REUSEADDR allows reusing a local address, we have to make sure
       now that the 5-tuple is unique. */
    struct tcp_pcb *cpcb;
    int i;
    /* Don't check listen- and bound-PCBs, check active- and TIME-WAIT PCBs. */
    for (i = 2; i < NUM_TCP_PCB_LISTS; i++) {
      for(cpcb = *tcp_pcb_lists[i]; cpcb != NULL; cpcb = cpcb->next) {
        if ((cpcb->local_port == pcb->local_port) &&
            (cpcb->remote_port == port) &&
            ip_addr_cmp(&cpcb->local_ip, &pcb->local_ip) &&
            ip_addr_cmp(&cpcb->remote_ip, ipaddr)) {
          /* linux returns EISCONN here, but ERR_USE should be OK for us */
          return ERR_USE;
        }
      }
    }
  }
#endif /* SO_REUSE */
  iss = tcp_next_iss();
  pcb->rcv_nxt = 0;
  pcb->snd_nxt = iss;
  pcb->lastack = iss - 1;
  pcb->snd_lbb = iss - 1;
  pcb->rcv_wnd = TCP_WND;
  pcb->rcv_ann_wnd = TCP_WND;
  pcb->rcv_ann_right_edge = pcb->rcv_nxt;
  pcb->snd_wnd = TCP_WND;
  /* As initial send MSS, we use TCP_MSS but limit it to 536.
     The send MSS is updated when an MSS option is received. */
  pcb->mss = (TCP_MSS > 536) ? 536 : TCP_MSS;
#if TCP_CALCULATE_EFF_SEND_MSS
  pcb->mss = tcp_eff_send_mss(pcb->mss, ipaddr);
#endif /* TCP_CALCULATE_EFF_SEND_MSS */
  pcb->cwnd = 1;
  pcb->ssthresh = pcb->mss * 10;
#if LWIP_CALLBACK_API
  pcb->connected = connected;
#else /* LWIP_CALLBACK_API */  
  LWIP_UNUSED_ARG(connected);
#endif /* LWIP_CALLBACK_API */

  /* Send a SYN together with the MSS option. */
  ret = tcp_enqueue_flags(pcb, TCP_SYN);
  if (ret == ERR_OK) {
    /* SYN segment was enqueued, changed the pcbs state now */
    pcb->state = SYN_SENT;
    if (old_local_port != 0) {
      TCP_RMV(&tcp_bound_pcbs, pcb);
    }
    TCP_REG(&tcp_active_pcbs, pcb);
    snmp_inc_tcpactiveopens();

    tcp_output(pcb);
  }
  return ret;
}

/**
 * Called every 500 ms and implements the retransmission timer and the timer that
 * removes PCBs that have been in TIME-WAIT for enough time. It also increments
 * various timers such as the inactivity timer in each PCB.
 *
 * Automatically called from tcp_tmr().
 */
void
tcp_slowtmr(void)
{
  struct tcp_pcb *pcb, *prev;
  u16_t eff_wnd;
  u8_t pcb_remove;      /* flag if a PCB should be removed */
  u8_t pcb_reset;       /* flag if a RST should be sent when removing */
  err_t err;

  err = ERR_OK;

  ++tcp_ticks;

  /* Steps through all of the active PCBs. */
  prev = NULL;
  pcb = tcp_active_pcbs;
  if (pcb == NULL) {
    LWIP_DEBUGF(TCP_DEBUG, ("tcp_slowtmr: no active pcbs\n"));
  }
  while (pcb != NULL) {
    LWIP_DEBUGF(TCP_DEBUG, ("tcp_slowtmr: processing active pcb\n"));
    LWIP_ASSERT("tcp_slowtmr: active pcb->state != CLOSED\n", pcb->state != CLOSED);
    LWIP_ASSERT("tcp_slowtmr: active pcb->state != LISTEN\n", pcb->state != LISTEN);
    LWIP_ASSERT("tcp_slowtmr: active pcb->state != TIME-WAIT\n", pcb->state != TIME_WAIT);

    pcb_remove = 0;
    pcb_reset = 0;

    if (pcb->state == SYN_SENT && pcb->nrtx == TCP_SYNMAXRTX) {
      ++pcb_remove;
      LWIP_DEBUGF(TCP_DEBUG, ("tcp_slowtmr: max SYN retries reached\n"));
    }
    else if (pcb->nrtx == TCP_MAXRTX) {
      ++pcb_remove;
      LWIP_DEBUGF(TCP_DEBUG, ("tcp_slowtmr: max DATA retries reached\n"));
    } else {
      if (pcb->persist_backoff > 0) {
        /* If snd_wnd is zero, use persist timer to send 1 byte probes
         * instead of using the standard retransmission mechanism. */
        pcb->persist_cnt++;
        if (pcb->persist_cnt >= tcp_persist_backoff[pcb->persist_backoff-1]) {
          pcb->persist_cnt = 0;
          if (pcb->persist_backoff < sizeof(tcp_persist_backoff)) {
            pcb->persist_backoff++;
          }
          tcp_zero_window_probe(pcb);
        }
      } else {
        /* Increase the retransmission timer if it is running */
        if(pcb->rtime >= 0)
          ++pcb->rtime;

        if (pcb->unacked != NULL && pcb->rtime >= pcb->rto) {
          /* Time for a retransmission. */
          LWIP_DEBUGF(TCP_RTO_DEBUG, ("tcp_slowtmr: rtime %"S16_F
                                      " pcb->rto %"S16_F"\n",
                                      pcb->rtime, pcb->rto));

          /* Double retransmission time-out unless we are trying to
           * connect to somebody (i.e., we are in SYN_SENT). */
          if (pcb->state != SYN_SENT) {
            pcb->rto = ((pcb->sa >> 3) + pcb->sv) << tcp_backoff[pcb->nrtx];
          }

          /* Reset the retransmission timer. */
          pcb->rtime = 0;

          /* Reduce congestion window and ssthresh. */
          eff_wnd = LWIP_MIN(pcb->cwnd, pcb->snd_wnd);
          pcb->ssthresh = eff_wnd >> 1;