tcp_in.c

lwip-1.4.0

/**
 * @file
 * Transmission Control Protocol, incoming traffic
 *
 * The input processing functions of the TCP layer.
 *
 * These functions are generally called in the order (ip_input() ->)
 * tcp_input() -> * tcp_process() -> tcp_receive() (-> application).
 * 
 */

/*
 * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
 * 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. The name of the author may not be used to endorse or promote products
 *    derived from this software without specific prior written permission.
 *
 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``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 AUTHOR 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.
 *
 * This file is part of the lwIP TCP/IP stack.
 *
 * Author: Adam Dunkels <[e-mail]>
 *
 */

#include "lwip/opt.h"

#if LWIP_TCP /* don't build if not configured for use in lwipopts.h */

#include "lwip/tcp_impl.h"
#include "lwip/def.h"
#include "lwip/ip_addr.h"
#include "lwip/netif.h"
#include "lwip/mem.h"
#include "lwip/memp.h"
#include "lwip/inet_chksum.h"
#include "lwip/stats.h"
#include "lwip/snmp.h"
#include "arch/perf.h"

/* These variables are global to all functions involved in the input
   processing of TCP segments. They are set by the tcp_input()
   function. */
static struct tcp_seg inseg;
static struct tcp_hdr *tcphdr;
static struct ip_hdr *iphdr;
static u32_t seqno, ackno;
static u8_t flags;
static u16_t tcplen;

static u8_t recv_flags;
static struct pbuf *recv_data;

struct tcp_pcb *tcp_input_pcb;

/* Forward declarations. */
static err_t tcp_process(struct tcp_pcb *pcb);
static void tcp_receive(struct tcp_pcb *pcb);
static void tcp_parseopt(struct tcp_pcb *pcb);

static err_t tcp_listen_input(struct tcp_pcb_listen *pcb);
static err_t tcp_timewait_input(struct tcp_pcb *pcb);

/**
 * The initial input processing of TCP. It verifies the TCP header, demultiplexes
 * the segment between the PCBs and passes it on to tcp_process(), which implements
 * the TCP finite state machine. This function is called by the IP layer (in
 * ip_input()).
 *
 * @param p received TCP segment to process (p->payload pointing to the IP header)
 * @param inp network interface on which this segment was received
 */
void
tcp_input(struct pbuf *p, struct netif *inp)
{
  struct tcp_pcb *pcb, *prev;
  struct tcp_pcb_listen *lpcb;
#if SO_REUSE
  struct tcp_pcb *lpcb_prev = NULL;
  struct tcp_pcb_listen *lpcb_any = NULL;
#endif /* SO_REUSE */
  u8_t hdrlen;
  err_t err;

  PERF_START;

  TCP_STATS_INC(tcp.recv);
  snmp_inc_tcpinsegs();

  iphdr = (struct ip_hdr *)p->payload;
  tcphdr = (struct tcp_hdr *)((u8_t *)p->payload + IPH_HL(iphdr) * 4);

#if TCP_INPUT_DEBUG
  tcp_debug_print(tcphdr);
#endif

  /* remove header from payload */
  if (pbuf_header(p, -((s16_t)(IPH_HL(iphdr) * 4))) || (p->tot_len < sizeof(struct tcp_hdr))) {
    /* drop short packets */
    LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_input: short packet (%"U16_F" bytes) discarded\n", p->tot_len));
    TCP_STATS_INC(tcp.lenerr);
    TCP_STATS_INC(tcp.drop);
    snmp_inc_tcpinerrs();
    pbuf_free(p);
    return;
  }

  /* Don't even process incoming broadcasts/multicasts. */
  if (ip_addr_isbroadcast(&current_iphdr_dest, inp) ||
      ip_addr_ismulticast(&current_iphdr_dest)) {
    TCP_STATS_INC(tcp.proterr);
    TCP_STATS_INC(tcp.drop);
    snmp_inc_tcpinerrs();
    pbuf_free(p);
    return;
  }

#if CHECKSUM_CHECK_TCP
  /* Verify TCP checksum. */
  if (inet_chksum_pseudo(p, ip_current_src_addr(), ip_current_dest_addr(),
      IP_PROTO_TCP, p->tot_len) != 0) {
      LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_input: packet discarded due to failing checksum 0x%04"X16_F"\n",
        inet_chksum_pseudo(p, ip_current_src_addr(), ip_current_dest_addr(),
      IP_PROTO_TCP, p->tot_len)));
#if TCP_DEBUG
    tcp_debug_print(tcphdr);
#endif /* TCP_DEBUG */
    TCP_STATS_INC(tcp.chkerr);
    TCP_STATS_INC(tcp.drop);
    snmp_inc_tcpinerrs();
    pbuf_free(p);
    return;
  }
#endif

  /* Move the payload pointer in the pbuf so that it points to the
     TCP data instead of the TCP header. */
  hdrlen = TCPH_HDRLEN(tcphdr);
  if(pbuf_header(p, -(hdrlen * 4))){
    /* drop short packets */
    LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_input: short packet\n"));
    TCP_STATS_INC(tcp.lenerr);
    TCP_STATS_INC(tcp.drop);
    snmp_inc_tcpinerrs();
    pbuf_free(p);
    return;
  }

  /* Convert fields in TCP header to host byte order. */
  tcphdr->src = ntohs(tcphdr->src);
  tcphdr->dest = ntohs(tcphdr->dest);
  seqno = tcphdr->seqno = ntohl(tcphdr->seqno);
  ackno = tcphdr->ackno = ntohl(tcphdr->ackno);
  tcphdr->wnd = ntohs(tcphdr->wnd);

  flags = TCPH_FLAGS(tcphdr);
  tcplen = p->tot_len + ((flags & (TCP_FIN | TCP_SYN)) ? 1 : 0);

  /* Demultiplex an incoming segment. First, we check if it is destined
     for an active connection. */
  prev = NULL;

  
  for(pcb = tcp_active_pcbs; pcb != NULL; pcb = pcb->next) {
    LWIP_ASSERT("tcp_input: active pcb->state != CLOSED", pcb->state != CLOSED);
    LWIP_ASSERT("tcp_input: active pcb->state != TIME-WAIT", pcb->state != TIME_WAIT);
    LWIP_ASSERT("tcp_input: active pcb->state != LISTEN", pcb->state != LISTEN);
    if (pcb->remote_port == tcphdr->src &&
       pcb->local_port == tcphdr->dest &&
       ip_addr_cmp(&(pcb->remote_ip), &current_iphdr_src) &&
       ip_addr_cmp(&(pcb->local_ip), &current_iphdr_dest)) {

      /* Move this PCB to the front of the list so that subsequent
         lookups will be faster (we exploit locality in TCP segment
         arrivals). */
      LWIP_ASSERT("tcp_input: pcb->next != pcb (before cache)", pcb->next != pcb);
      if (prev != NULL) {
        prev->next = pcb->next;
        pcb->next = tcp_active_pcbs;
        tcp_active_pcbs = pcb;
      }
      LWIP_ASSERT("tcp_input: pcb->next != pcb (after cache)", pcb->next != pcb);
      break;
    }
    prev = pcb;
  }

  if (pcb == NULL) {
    /* If it did not go to an active connection, we check the connections
       in the TIME-WAIT state. */
    for(pcb = tcp_tw_pcbs; pcb != NULL; pcb = pcb->next) {
      LWIP_ASSERT("tcp_input: TIME-WAIT pcb->state == TIME-WAIT", pcb->state == TIME_WAIT);
      if (pcb->remote_port == tcphdr->src &&
         pcb->local_port == tcphdr->dest &&
         ip_addr_cmp(&(pcb->remote_ip), &current_iphdr_src) &&
         ip_addr_cmp(&(pcb->local_ip), &current_iphdr_dest)) {
        /* We don't really care enough to move this PCB to the front
           of the list since we are not very likely to receive that
           many segments for connections in TIME-WAIT. */
        LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_input: packed for TIME_WAITing connection.\n"));
        tcp_timewait_input(pcb);
        pbuf_free(p);
        return;
      }
    }

    /* Finally, if we still did not get a match, we check all PCBs that
       are LISTENing for incoming connections. */
    prev = NULL;
    for(lpcb = tcp_listen_pcbs.listen_pcbs; lpcb != NULL; lpcb = lpcb->next) {
      if (lpcb->local_port == tcphdr->dest) {
#if SO_REUSE
        if (ip_addr_cmp(&(lpcb->local_ip), &current_iphdr_dest)) {
          /* found an exact match */
          break;
        } else if(ip_addr_isany(&(lpcb->local_ip))) {
          /* found an ANY-match */
          lpcb_any = lpcb;
          lpcb_prev = prev;
        }
#else /* SO_REUSE */
        if (ip_addr_cmp(&(lpcb->local_ip), &current_iphdr_dest) ||
            ip_addr_isany(&(lpcb->local_ip))) {
          /* found a match */
          break;
        }
#endif /* SO_REUSE */
      }
      prev = (struct tcp_pcb *)lpcb;
    }
#if SO_REUSE
    /* first try specific local IP */
    if (lpcb == NULL) {
      /* only pass to ANY if no specific local IP has been found */
      lpcb = lpcb_any;
      prev = lpcb_prev;
    }
#endif /* SO_REUSE */
    if (lpcb != NULL) {
      /* Move this PCB to the front of the list so that subsequent
         lookups will be faster (we exploit locality in TCP segment
         arrivals). */
      if (prev != NULL) {
        ((struct tcp_pcb_listen *)prev)->next = lpcb->next;
              /* our successor is the remainder of the listening list */
        lpcb->next = tcp_listen_pcbs.listen_pcbs;
              /* put this listening pcb at the head of the listening list */
        tcp_listen_pcbs.listen_pcbs = lpcb;
      }
    
      LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_input: packed for LISTENing connection.\n"));
      tcp_listen_input(lpcb);
      pbuf_free(p);
      return;
    }
  }

#if TCP_INPUT_DEBUG
  LWIP_DEBUGF(TCP_INPUT_DEBUG, ("+-+-+-+-+-+-+-+-+-+-+-+-+-+- tcp_input: flags "));
  tcp_debug_print_flags(TCPH_FLAGS(tcphdr));
  LWIP_DEBUGF(TCP_INPUT_DEBUG, ("-+-+-+-+-+-+-+-+-+-+-+-+-+-+\n"));
#endif /* TCP_INPUT_DEBUG */


  if (pcb != NULL) {
    /* The incoming segment belongs to a connection. */
#if TCP_INPUT_DEBUG
#if TCP_DEBUG
    tcp_debug_print_state(pcb->state);
#endif /* TCP_DEBUG */
#endif /* TCP_INPUT_DEBUG */

    /* Set up a tcp_seg structure. */
    inseg.next = NULL;
    inseg.len = p->tot_len;
    inseg.p = p;
    inseg.tcphdr = tcphdr;

    recv_data = NULL;
    recv_flags = 0;

    /* If there is data which was previously "refused" by upper layer */
    if (pcb->refused_data != NULL) {
      /* Notify again application with data previously received. */
      LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_input: notify kept packet\n"));
      TCP_EVENT_RECV(pcb, pcb->refused_data, ERR_OK, err);
      if (err == ERR_OK) {
        pcb->refused_data = NULL;
      } else if ((err == ERR_ABRT) || (tcplen > 0)) {
        /* if err == ERR_ABRT, 'pcb' is already deallocated */
        /* Drop incoming packets because pcb is "full" (only if the incoming
           segment contains data). */
        LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_input: drop incoming packets, because pcb is \"full\"\n"));
        TCP_STATS_INC(tcp.drop);
        snmp_inc_tcpinerrs();
        pbuf_free(p);
        return;
      }
    }
    tcp_input_pcb = pcb;
    err = tcp_process(pcb);
    /* A return value of ERR_ABRT means that tcp_abort() was called
       and that the pcb has been freed. If so, we don't do anything. */
    if (err != ERR_ABRT) {
      if (recv_flags & TF_RESET) {
        /* TF_RESET means that the connection was reset by the other
           end. We then call the error callback to inform the
           application that the connection is dead before we
           deallocate the PCB. */
        TCP_EVENT_ERR(pcb->errf, pcb->callback_arg, ERR_RST);
        tcp_pcb_remove(&tcp_active_pcbs, pcb);
        memp_free(MEMP_TCP_PCB, pcb);
      } else if (recv_flags & TF_CLOSED) {
        /* The connection has been closed and we will deallocate the
           PCB. */
        tcp_pcb_remove(&tcp_active_pcbs, pcb);
        memp_free(MEMP_TCP_PCB, pcb);
      } else {
        err = ERR_OK;
        /* If the application has registered a "sent" function to be
           called when new send buffer space is available, we call it
           now. */
        if (pcb->acked > 0) {
          TCP_EVENT_SENT(pcb, pcb->acked, err);
          if (err == ERR_ABRT) {
            goto aborted;
          }
        }

        if (recv_data != NULL) {
          LWIP_ASSERT("pcb->refused_data == NULL", pcb->refused_data == NULL);
          if (pcb->flags & TF_RXCLOSED) {
            /* received data although already closed -> abort (send RST) to
               notify the remote host that not all data has been processed */
            pbuf_free(recv_data);
            tcp_abort(pcb);
            goto aborted;
          }
          if (flags & TCP_PSH) {
            recv_data->flags |= PBUF_FLAG_PUSH;
          }

          /* Notify application that data has been received. */
          TCP_EVENT_RECV(pcb, recv_data, ERR_OK, err);
          if (err == ERR_ABRT) {
            goto aborted;
          }

          /* If the upper layer can't receive this data, store it */
          if (err != ERR_OK) {
            pcb->refused_data = recv_data;
            LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_input: keep incoming packet, because pcb is \"full\"\n"));
          }
        }

        /* If a FIN segment was received, we call the callback
           function with a NULL buffer to indicate EOF. */
        if (recv_flags & TF_GOT_FIN) {
          /* correct rcv_wnd as the application won't call tcp_recved()
             for the FIN's seqno */
          if (pcb->rcv_wnd != TCP_WND) {
            pcb->rcv_wnd++;
          }
          TCP_EVENT_CLOSED(pcb, err);
          if (err == ERR_ABRT) {
            goto aborted;
          }
        }

        tcp_input_pcb = NULL;
        /* Try to send something out. */
        tcp_output(pcb);
#if TCP_INPUT_DEBUG
#if TCP_DEBUG