tcp_in.c

lwip-1.4.0

    break;
  }
  return ERR_OK;
}

#if TCP_QUEUE_OOSEQ
/**
 * Insert segment into the list (segments covered with new one will be deleted)
 *
 * Called from tcp_receive()
 */
static void
tcp_oos_insert_segment(struct tcp_seg *cseg, struct tcp_seg *next)
{
  struct tcp_seg *old_seg;

  if (TCPH_FLAGS(cseg->tcphdr) & TCP_FIN) {
    /* received segment overlaps all following segments */
    tcp_segs_free(next);
    next = NULL;
  }
  else {
    /* delete some following segments
       oos queue may have segments with FIN flag */
    while (next &&
           TCP_SEQ_GEQ((seqno + cseg->len),
                      (next->tcphdr->seqno + next->len))) {
      /* cseg with FIN already processed */
      if (TCPH_FLAGS(next->tcphdr) & TCP_FIN) {
        TCPH_SET_FLAG(cseg->tcphdr, TCP_FIN);
      }
      old_seg = next;
      next = next->next;
      tcp_seg_free(old_seg);
    }
    if (next &&
        TCP_SEQ_GT(seqno + cseg->len, next->tcphdr->seqno)) {
      /* We need to trim the incoming segment. */
      cseg->len = (u16_t)(next->tcphdr->seqno - seqno);
      pbuf_realloc(cseg->p, cseg->len);
    }
  }
  cseg->next = next;
}
#endif /* TCP_QUEUE_OOSEQ */

/**
 * Called by tcp_process. Checks if the given segment is an ACK for outstanding
 * data, and if so frees the memory of the buffered data. Next, is places the
 * segment on any of the receive queues (pcb->recved or pcb->ooseq). If the segment
 * is buffered, the pbuf is referenced by pbuf_ref so that it will not be freed until
 * i it has been removed from the buffer.
 *
 * If the incoming segment constitutes an ACK for a segment that was used for RTT
 * estimation, the RTT is estimated here as well.
 *
 * Called from tcp_process().
 */
static void
tcp_receive(struct tcp_pcb *pcb)
{
  struct tcp_seg *next;
#if TCP_QUEUE_OOSEQ
  struct tcp_seg *prev, *cseg;
#endif /* TCP_QUEUE_OOSEQ */
  struct pbuf *p;
  s32_t off;
  s16_t m;
  u32_t right_wnd_edge;
  u16_t new_tot_len;
  int found_dupack = 0;

  if (flags & TCP_ACK) {
    right_wnd_edge = pcb->snd_wnd + pcb->snd_wl2;

    /* Update window. */
    if (TCP_SEQ_LT(pcb->snd_wl1, seqno) ||
       (pcb->snd_wl1 == seqno && TCP_SEQ_LT(pcb->snd_wl2, ackno)) ||
       (pcb->snd_wl2 == ackno && tcphdr->wnd > pcb->snd_wnd)) {
      pcb->snd_wnd = tcphdr->wnd;
      pcb->snd_wl1 = seqno;
      pcb->snd_wl2 = ackno;
      if (pcb->snd_wnd > 0 && pcb->persist_backoff > 0) {
          pcb->persist_backoff = 0;
      }
      LWIP_DEBUGF(TCP_WND_DEBUG, ("tcp_receive: window update %"U16_F"\n", pcb->snd_wnd));
#if TCP_WND_DEBUG
    } else {
      if (pcb->snd_wnd != tcphdr->wnd) {
        LWIP_DEBUGF(TCP_WND_DEBUG, 
                    ("tcp_receive: no window update lastack %"U32_F" ackno %"
                     U32_F" wl1 %"U32_F" seqno %"U32_F" wl2 %"U32_F"\n",
                     pcb->lastack, ackno, pcb->snd_wl1, seqno, pcb->snd_wl2));
      }
#endif /* TCP_WND_DEBUG */
    }

    /* (From Stevens TCP/IP Illustrated Vol II, p970.) Its only a
     * duplicate ack if:
     * 1) It doesn't ACK new data 
     * 2) length of received packet is zero (i.e. no payload) 
     * 3) the advertised window hasn't changed 
     * 4) There is outstanding unacknowledged data (retransmission timer running)
     * 5) The ACK is == biggest ACK sequence number so far seen (snd_una)
     * 
     * If it passes all five, should process as a dupack: 
     * a) dupacks < 3: do nothing 
     * b) dupacks == 3: fast retransmit 
     * c) dupacks > 3: increase cwnd 
     * 
     * If it only passes 1-3, should reset dupack counter (and add to
     * stats, which we don't do in lwIP)
     *
     * If it only passes 1, should reset dupack counter
     *
     */

    /* Clause 1 */
    if (TCP_SEQ_LEQ(ackno, pcb->lastack)) {
      pcb->acked = 0;
      /* Clause 2 */
      if (tcplen == 0) {
        /* Clause 3 */
        if (pcb->snd_wl2 + pcb->snd_wnd == right_wnd_edge){
          /* Clause 4 */
          if (pcb->rtime >= 0) {
            /* Clause 5 */
            if (pcb->lastack == ackno) {
              found_dupack = 1;
              if (pcb->dupacks + 1 > pcb->dupacks)
                ++pcb->dupacks;
              if (pcb->dupacks > 3) {
                /* Inflate the congestion window, but not if it means that
                   the value overflows. */
                if ((u16_t)(pcb->cwnd + pcb->mss) > pcb->cwnd) {
                  pcb->cwnd += pcb->mss;
                }
              } else if (pcb->dupacks == 3) {
                /* Do fast retransmit */
                tcp_rexmit_fast(pcb);
              }
            }
          }
        }
      }
      /* If Clause (1) or more is true, but not a duplicate ack, reset
       * count of consecutive duplicate acks */
      if (!found_dupack) {
        pcb->dupacks = 0;
      }
    } else if (TCP_SEQ_BETWEEN(ackno, pcb->lastack+1, pcb->snd_nxt)){
      /* We come here when the ACK acknowledges new data. */

      /* Reset the "IN Fast Retransmit" flag, since we are no longer
         in fast retransmit. Also reset the congestion window to the
         slow start threshold. */
      if (pcb->flags & TF_INFR) {
        pcb->flags &= ~TF_INFR;
        pcb->cwnd = pcb->ssthresh;
      }

      /* Reset the number of retransmissions. */
      pcb->nrtx = 0;

      /* Reset the retransmission time-out. */
      pcb->rto = (pcb->sa >> 3) + pcb->sv;

      /* Update the send buffer space. Diff between the two can never exceed 64K? */
      pcb->acked = (u16_t)(ackno - pcb->lastack);

      pcb->snd_buf += pcb->acked;

      /* Reset the fast retransmit variables. */
      pcb->dupacks = 0;
      pcb->lastack = ackno;

      /* Update the congestion control variables (cwnd and
         ssthresh). */
      if (pcb->state >= ESTABLISHED) {
        if (pcb->cwnd < pcb->ssthresh) {
          if ((u16_t)(pcb->cwnd + pcb->mss) > pcb->cwnd) {
            pcb->cwnd += pcb->mss;
          }
          LWIP_DEBUGF(TCP_CWND_DEBUG, ("tcp_receive: slow start cwnd %"U16_F"\n", pcb->cwnd));
        } else {
          u16_t new_cwnd = (pcb->cwnd + pcb->mss * pcb->mss / pcb->cwnd);
          if (new_cwnd > pcb->cwnd) {
            pcb->cwnd = new_cwnd;
          }
          LWIP_DEBUGF(TCP_CWND_DEBUG, ("tcp_receive: congestion avoidance cwnd %"U16_F"\n", pcb->cwnd));
        }
      }
      LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_receive: ACK for %"U32_F", unacked->seqno %"U32_F":%"U32_F"\n",
                                    ackno,
                                    pcb->unacked != NULL?
                                    ntohl(pcb->unacked->tcphdr->seqno): 0,
                                    pcb->unacked != NULL?
                                    ntohl(pcb->unacked->tcphdr->seqno) + TCP_TCPLEN(pcb->unacked): 0));

      /* Remove segment from the unacknowledged list if the incoming
         ACK acknowlegdes them. */
      while (pcb->unacked != NULL &&
             TCP_SEQ_LEQ(ntohl(pcb->unacked->tcphdr->seqno) +
                         TCP_TCPLEN(pcb->unacked), ackno)) {
        LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_receive: removing %"U32_F":%"U32_F" from pcb->unacked\n",
                                      ntohl(pcb->unacked->tcphdr->seqno),
                                      ntohl(pcb->unacked->tcphdr->seqno) +
                                      TCP_TCPLEN(pcb->unacked)));

        next = pcb->unacked;
        pcb->unacked = pcb->unacked->next;

        LWIP_DEBUGF(TCP_QLEN_DEBUG, ("tcp_receive: queuelen %"U16_F" ... ", (u16_t)pcb->snd_queuelen));
        LWIP_ASSERT("pcb->snd_queuelen >= pbuf_clen(next->p)", (pcb->snd_queuelen >= pbuf_clen(next->p)));
        /* Prevent ACK for FIN to generate a sent event */
        if ((pcb->acked != 0) && ((TCPH_FLAGS(next->tcphdr) & TCP_FIN) != 0)) {
          pcb->acked--;
        }

        pcb->snd_queuelen -= pbuf_clen(next->p);
        tcp_seg_free(next);

        LWIP_DEBUGF(TCP_QLEN_DEBUG, ("%"U16_F" (after freeing unacked)\n", (u16_t)pcb->snd_queuelen));
        if (pcb->snd_queuelen != 0) {
          LWIP_ASSERT("tcp_receive: valid queue length", pcb->unacked != NULL ||
                      pcb->unsent != NULL);
        }
      }

      /* If there's nothing left to acknowledge, stop the retransmit
         timer, otherwise reset it to start again */
      if(pcb->unacked == NULL)
        pcb->rtime = -1;
      else
        pcb->rtime = 0;

      pcb->polltmr = 0;
    } else {
      /* Fix bug bug #21582: out of sequence ACK, didn't really ack anything */
      pcb->acked = 0;
    }

    /* We go through the ->unsent list to see if any of the segments
       on the list are acknowledged by the ACK. This may seem
       strange since an "unsent" segment shouldn't be acked. The
       rationale is that lwIP puts all outstanding segments on the
       ->unsent list after a retransmission, so these segments may
       in fact have been sent once. */
    while (pcb->unsent != NULL &&
           TCP_SEQ_BETWEEN(ackno, ntohl(pcb->unsent->tcphdr->seqno) + 
                           TCP_TCPLEN(pcb->unsent), pcb->snd_nxt)) {
      LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_receive: removing %"U32_F":%"U32_F" from pcb->unsent\n",
                                    ntohl(pcb->unsent->tcphdr->seqno), ntohl(pcb->unsent->tcphdr->seqno) +
                                    TCP_TCPLEN(pcb->unsent)));

      next = pcb->unsent;
      pcb->unsent = pcb->unsent->next;
      LWIP_DEBUGF(TCP_QLEN_DEBUG, ("tcp_receive: queuelen %"U16_F" ... ", (u16_t)pcb->snd_queuelen));
      LWIP_ASSERT("pcb->snd_queuelen >= pbuf_clen(next->p)", (pcb->snd_queuelen >= pbuf_clen(next->p)));
      /* Prevent ACK for FIN to generate a sent event */
      if ((pcb->acked != 0) && ((TCPH_FLAGS(next->tcphdr) & TCP_FIN) != 0)) {
        pcb->acked--;
      }
      pcb->snd_queuelen -= pbuf_clen(next->p);
      tcp_seg_free(next);
      LWIP_DEBUGF(TCP_QLEN_DEBUG, ("%"U16_F" (after freeing unsent)\n", (u16_t)pcb->snd_queuelen));
      if (pcb->snd_queuelen != 0) {
        LWIP_ASSERT("tcp_receive: valid queue length",
          pcb->unacked != NULL || pcb->unsent != NULL);
      }
    }
    /* End of ACK for new data processing. */

    LWIP_DEBUGF(TCP_RTO_DEBUG, ("tcp_receive: pcb->rttest %"U32_F" rtseq %"U32_F" ackno %"U32_F"\n",
                                pcb->rttest, pcb->rtseq, ackno));

    /* RTT estimation calculations. This is done by checking if the
       incoming segment acknowledges the segment we use to take a
       round-trip time measurement. */
    if (pcb->rttest && TCP_SEQ_LT(pcb->rtseq, ackno)) {
      /* diff between this shouldn't exceed 32K since this are tcp timer ticks
         and a round-trip shouldn't be that long... */
      m = (s16_t)(tcp_ticks - pcb->rttest);

      LWIP_DEBUGF(TCP_RTO_DEBUG, ("tcp_receive: experienced rtt %"U16_F" ticks (%"U16_F" msec).\n",
                                  m, m * TCP_SLOW_INTERVAL));

      /* This is taken directly from VJs original code in his paper */
      m = m - (pcb->sa >> 3);
      pcb->sa += m;
      if (m < 0) {
        m = -m;
      }
      m = m - (pcb->sv >> 2);
      pcb->sv += m;
      pcb->rto = (pcb->sa >> 3) + pcb->sv;

      LWIP_DEBUGF(TCP_RTO_DEBUG, ("tcp_receive: RTO %"U16_F" (%"U16_F" milliseconds)\n",
                                  pcb->rto, pcb->rto * TCP_SLOW_INTERVAL));

      pcb->rttest = 0;
    }
  }

  /* If the incoming segment contains data, we must process it
     further. */
  if (tcplen > 0) {
    /* This code basically does three things:

    +) If the incoming segment contains data that is the next
    in-sequence data, this data is passed to the application. This
    might involve trimming the first edge of the data. The rcv_nxt
    variable and the advertised window are adjusted.

    +) If the incoming segment has data that is above the next
    sequence number expected (->rcv_nxt), the segment is placed on
    the ->ooseq queue. This is done by finding the appropriate
    place in the ->ooseq queue (which is ordered by sequence
    number) and trim the segment in both ends if needed. An
    immediate ACK is sent to indicate that we received an
    out-of-sequence segment.

    +) Finally, we check if the first segment on the ->ooseq queue
    now is in sequence (i.e., if rcv_nxt >= ooseq->seqno). If
    rcv_nxt > ooseq->seqno, we must trim the first edge of the
    segment on ->ooseq before we adjust rcv_nxt. The data in the
    segments that are now on sequence are chained onto the
    incoming segment so that we only need to call the application
    once.
    */

    /* First, we check if we must trim the first edge. We have to do
       this if the sequence number of the incoming segment is less
       than rcv_nxt, and the sequence number plus the length of the
       segment is larger than rcv_nxt. */
    /*    if (TCP_SEQ_LT(seqno, pcb->rcv_nxt)){
          if (TCP_SEQ_LT(pcb->rcv_nxt, seqno + tcplen)) {*/
    if (TCP_SEQ_BETWEEN(pcb->rcv_nxt, seqno + 1, seqno + tcplen - 1)){
      /* Trimming the first edge is done by pushing the payload
         pointer in the pbuf downwards. This is somewhat tricky since
         we do not want to discard the full contents of the pbuf up to
         the new starting point of the data since we have to keep the
         TCP header which is present in the first pbuf in the chain.

         What is done is really quite a nasty hack: the first pbuf in
         the pbuf chain is pointed to by inseg.p. Since we need to be
         able to deallocate the whole pbuf, we cannot change this
         inseg.p pointer to point to any of the later pbufs in the
         chain. Instead, we point the ->payload pointer in the first
         pbuf to data in one of the later pbufs. We also set the
         inseg.data pointer to point to the right place. This way, the
         ->p pointer will still point to the first pbuf, but the
         ->p->payload pointer will point to data in another pbuf.

         After we are done with adjusting the pbuf pointers we must
         adjust the ->data pointer in the seg and the segment
         length.*/

      off = pcb->rcv_nxt - seqno;
      p = inseg.p;
      LWIP_ASSERT("inseg.p != NULL", inseg.p);
      LWIP_ASSERT("insane offset!", (off < 0x7fff));
      if (inseg.p->len < off) {
        LWIP_ASSERT("pbuf too short!", (((s32_t)inseg.p->tot_len) >= off));
        new_tot_len = (u16_t)(inseg.p->tot_len - off);
        while (p->len < off) {
          off -= p->len;
          /* KJM following line changed (with addition of new_tot_len var)
             to fix bug #9076
             inseg.p->tot_len -= p->len; */
          p->tot_len = new_tot_len;
          p->len = 0;
          p = p->next;
        }
        if(pbuf_header(p, (s16_t)-off)) {
          /* Do we need to cope with this failing?  Assert for now */
          LWIP_ASSERT("pbuf_header failed", 0);
        }
      } else {
        if(pbuf_header(inseg.p, (s16_t)-off)) {
          /* Do we need to cope with this failing?  Assert for now */
          LWIP_ASSERT("pbuf_header failed", 0);
        }
      }
      inseg.len -= (u16_t)(pcb->rcv_nxt - seqno);
      inseg.tcphdr->seqno = seqno = pcb->rcv_nxt;
    }
    else {
      if (TCP_SEQ_LT(seqno, pcb->rcv_nxt)){
        /* the whole segment is < rcv_nxt */
        /* must be a duplicate of a packet that has already been correctly handled */