C:\tcp_in.c C:\tcp_in.new.c
/** /**
 * @file  * @file
 * Transmission Control Protocol, incoming traffic  * Transmission Control Protocol, incoming traffic
 *  *
 * The input processing functions of the TCP layer.  * The input processing functions of the TCP layer.
 *  *
 * These functions are generally called in the order (ip_input() ->)  * These functions are generally called in the order (ip_input() ->)
 * tcp_input() -> * tcp_process() -> tcp_receive() (-> application).  * tcp_input() -> * tcp_process() -> tcp_receive() (-> application).
 *  *
 */   */ 
   
/* /*
 * Copyright (c) 2001-2004 Swedish Institute of Computer Science.  * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
 * All rights reserved.  * All rights reserved.
 *  *
 * Redistribution and use in source and binary forms, with or without modification,  * Redistribution and use in source and binary forms, with or without modification,
 * are permitted provided that the following conditions are met:  * are permitted provided that the following conditions are met:
 *  *
 * 1. Redistributions of source code must retain the above copyright notice,  * 1. Redistributions of source code must retain the above copyright notice,
 *    this list of conditions and the following disclaimer.  *    this list of conditions and the following disclaimer.
 * 2. Redistributions in binary form must reproduce the above copyright notice,  * 2. Redistributions in binary form must reproduce the above copyright notice,
 *    this list of conditions and the following disclaimer in the documentation  *    this list of conditions and the following disclaimer in the documentation
 *    and/or other materials provided with the distribution.  *    and/or other materials provided with the distribution.
 * 3. The name of the author may not be used to endorse or promote products  * 3. The name of the author may not be used to endorse or promote products
 *    derived from this software without specific prior written permission.  *    derived from this software without specific prior written permission.
 *  *
 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR IMPLIED  * THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR IMPLIED
 * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF  * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
 * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT  * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT
 * SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,  * SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
 * EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT  * EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT
 * OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS  * OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
 * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN  * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
 * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING  * 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  * IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY
 * OF SUCH DAMAGE.  * OF SUCH DAMAGE.
 *  *
 * This file is part of the lwIP TCP/IP stack.  * This file is part of the lwIP TCP/IP stack.
 *  *
 * Author: Adam Dunkels <adam@sics.se>  * Author: Adam Dunkels <adam@sics.se>
 *  *
 */   */ 
   
#include "lwip/opt.h" #include "lwip/opt.h"
   
#if LWIP_TCP /* don't build if not configured for use in lwipopts.h */  #if LWIP_TCP /* don't build if not configured for use in lwipopts.h */ 
   
#include "lwip/priv/tcp_priv.h" #include "lwip/priv/tcp_priv.h"
#include "lwip/def.h" #include "lwip/def.h"
#include "lwip/ip_addr.h" #include "lwip/ip_addr.h"
#include "lwip/netif.h" #include "lwip/netif.h"
#include "lwip/mem.h" #include "lwip/mem.h"
#include "lwip/memp.h" #include "lwip/memp.h"
#include "lwip/inet_chksum.h" #include "lwip/inet_chksum.h"
#include "lwip/stats.h" #include "lwip/stats.h"
#include "lwip/ip6.h" #include "lwip/ip6.h"
#include "lwip/ip6_addr.h" #include "lwip/ip6_addr.h"
#if LWIP_ND6_TCP_REACHABILITY_HINTS #if LWIP_ND6_TCP_REACHABILITY_HINTS
#include "lwip/nd6.h" #include "lwip/nd6.h"
#endif /* LWIP_ND6_TCP_REACHABILITY_HINTS */  #endif /* LWIP_ND6_TCP_REACHABILITY_HINTS */ 
   
#include <string.h> #include <string.h>
   
#ifdef LWIP_HOOK_FILENAME #ifdef LWIP_HOOK_FILENAME
#include LWIP_HOOK_FILENAME #include LWIP_HOOK_FILENAME
#endif #endif
   
/** Initial CWND calculation as defined RFC 2581 */  /** Initial CWND calculation as defined RFC 2581 */ 
#define LWIP_TCP_CALC_INITIAL_CWND(mss) ((tcpwnd_size_t)LWIP_MIN((4U * (mss)), LWIP_MAX((2U * (mss)), 4380U))) #define LWIP_TCP_CALC_INITIAL_CWND(mss) ((tcpwnd_size_t)LWIP_MIN((4U * (mss)), LWIP_MAX((2U * (mss)), 4380U)))
   
/* These variables are global to all functions involved in the input /* These variables are global to all functions involved in the input
   processing of TCP segments. They are set by the tcp_input()    processing of TCP segments. They are set by the tcp_input()
   function. */     function. */ 
static struct tcp_seg inseg; static struct tcp_seg inseg;
static struct tcp_hdr *tcphdr; static struct tcp_hdr *tcphdr;
static u16_t tcphdr_optlen; static u16_t tcphdr_optlen;
static u16_t tcphdr_opt1len; static u16_t tcphdr_opt1len;
static u8_t *tcphdr_opt2; static u8_t *tcphdr_opt2;
static u16_t tcp_optidx; static u16_t tcp_optidx;
static u32_t seqno, ackno; static u32_t seqno, ackno;
static tcpwnd_size_t recv_acked; static tcpwnd_size_t recv_acked;
static u16_t tcplen; static u16_t tcplen;
static u8_t flags; static u8_t flags;
   
static u8_t recv_flags; static u8_t recv_flags;
static struct pbuf *recv_data; static struct pbuf *recv_data;
   
struct tcp_pcb *tcp_input_pcb; struct tcp_pcb *tcp_input_pcb;
   
/* Forward declarations. */  /* Forward declarations. */ 
static err_t tcp_process(struct tcp_pcb *pcb); static err_t tcp_process(struct tcp_pcb *pcb);
static void tcp_receive(struct tcp_pcb *pcb); static void tcp_receive(struct tcp_pcb *pcb);
static void tcp_parseopt(struct tcp_pcb *pcb); static void tcp_parseopt(struct tcp_pcb *pcb);
   
static void tcp_listen_input(struct tcp_pcb_listen *pcb); static void tcp_listen_input(struct tcp_pcb_listen *pcb);
static void tcp_timewait_input(struct tcp_pcb *pcb); static void tcp_timewait_input(struct tcp_pcb *pcb);
   
static int tcp_input_delayed_close(struct tcp_pcb *pcb); static int tcp_input_delayed_close(struct tcp_pcb *pcb);
   
#if LWIP_TCP_SACK_OUT #if LWIP_TCP_SACK_OUT
static void tcp_add_sack(struct tcp_pcb *pcb, u32_t left, u32_t right); static void tcp_add_sack(struct tcp_pcb *pcb, u32_t left, u32_t right);
static void tcp_remove_sacks_lt(struct tcp_pcb *pcb, u32_t seq); static void tcp_remove_sacks_lt(struct tcp_pcb *pcb, u32_t seq);
#if defined(TCP_OOSEQ_BYTES_LIMIT) || defined(TCP_OOSEQ_PBUFS_LIMIT) #if defined(TCP_OOSEQ_BYTES_LIMIT) || defined(TCP_OOSEQ_PBUFS_LIMIT)
static void tcp_remove_sacks_gt(struct tcp_pcb *pcb, u32_t seq); static void tcp_remove_sacks_gt(struct tcp_pcb *pcb, u32_t seq);
#endif /* TCP_OOSEQ_BYTES_LIMIT || TCP_OOSEQ_PBUFS_LIMIT */  #endif /* TCP_OOSEQ_BYTES_LIMIT || TCP_OOSEQ_PBUFS_LIMIT */ 
#endif /* LWIP_TCP_SACK_OUT */  #endif /* LWIP_TCP_SACK_OUT */ 
   
/** /**
 * The initial input processing of TCP. It verifies the TCP header, demultiplexes  * 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 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  * the TCP finite state machine. This function is called by the IP layer (in
 * ip_input()).  * ip_input()).
 *  *
 * @param p received TCP segment to process (p->payload pointing to the TCP header)  * @param p received TCP segment to process (p->payload pointing to the TCP header)
 * @param inp network interface on which this segment was received  * @param inp network interface on which this segment was received
 */   */ 
void void
tcp_input(struct pbuf *p, struct netif *inp) tcp_input(struct pbuf *p, struct netif *inp)
{ {
  struct tcp_pcb *pcb, *prev;   struct tcp_pcb *pcb, *prev;
  struct tcp_pcb_listen *lpcb;   struct tcp_pcb_listen *lpcb;
#if SO_REUSE #if SO_REUSE
  struct tcp_pcb *lpcb_prev = NULL;   struct tcp_pcb *lpcb_prev = NULL;
  struct tcp_pcb_listen *lpcb_any = NULL;   struct tcp_pcb_listen *lpcb_any = NULL;
#endif /* SO_REUSE */  #endif /* SO_REUSE */ 
  u8_t hdrlen_bytes;   u8_t hdrlen_bytes;
  err_t err;   err_t err;
   
  LWIP_UNUSED_ARG(inp);   LWIP_UNUSED_ARG(inp);
  LWIP_ASSERT_CORE_LOCKED();   LWIP_ASSERT_CORE_LOCKED();
  LWIP_ASSERT("tcp_input: invalid pbuf", p != NULL);   LWIP_ASSERT("tcp_input: invalid pbuf", p != NULL);
   
  PERF_START;   PERF_START;
   
  TCP_STATS_INC(tcp.recv);   TCP_STATS_INC(tcp.recv);
  MIB2_STATS_INC(mib2.tcpinsegs);   MIB2_STATS_INC(mib2.tcpinsegs);
   
  tcphdr = (struct tcp_hdr *)p->payload;   tcphdr = (struct tcp_hdr *)p->payload;
   
#if TCP_INPUT_DEBUG #if TCP_INPUT_DEBUG
  tcp_debug_print(tcphdr);   tcp_debug_print(tcphdr);
#endif #endif
   
  /* Check that TCP header fits in payload */    /* Check that TCP header fits in payload */ 
  if (p->len < TCP_HLEN) {   if (p->len < TCP_HLEN) {
    /* drop short packets */      /* drop short packets */ 
    LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_input: short packet (%"U16_F" bytes) discarded\n", p->tot_len));     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.lenerr);
    goto dropped;     goto dropped;
  }   }
   
  /* Don't even process incoming broadcasts/multicasts. */    /* Don't even process incoming broadcasts/multicasts. */ 
  if (ip_addr_isbroadcast(ip_current_dest_addr(), ip_current_netif()) ||   if (ip_addr_isbroadcast(ip_current_dest_addr(), ip_current_netif()) ||
      ip_addr_ismulticast(ip_current_dest_addr())) {       ip_addr_ismulticast(ip_current_dest_addr())) {
    TCP_STATS_INC(tcp.proterr);     TCP_STATS_INC(tcp.proterr);
    goto dropped;     goto dropped;
  }   }
   
#if CHECKSUM_CHECK_TCP #if CHECKSUM_CHECK_TCP
  IF__NETIF_CHECKSUM_ENABLED(inp, NETIF_CHECKSUM_CHECK_TCP) {   IF__NETIF_CHECKSUM_ENABLED(inp, NETIF_CHECKSUM_CHECK_TCP) {
    /* Verify TCP checksum. */      /* Verify TCP checksum. */ 
    u16_t chksum = ip_chksum_pseudo(p, IP_PROTO_TCP, p->tot_len,     u16_t chksum = ip_chksum_pseudo(p, IP_PROTO_TCP, p->tot_len,
                                    ip_current_src_addr(), ip_current_dest_addr());                                     ip_current_src_addr(), ip_current_dest_addr());
    if (chksum != 0) {     if (chksum != 0) {
      LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_input: packet discarded due to failing checksum 0x%04"X16_F"\n",       LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_input: packet discarded due to failing checksum 0x%04"X16_F"\n",
                                    chksum));                                     chksum));
      tcp_debug_print(tcphdr);       tcp_debug_print(tcphdr);
      TCP_STATS_INC(tcp.chkerr);       TCP_STATS_INC(tcp.chkerr);
      goto dropped;       goto dropped;
    }     }
  }   }
#endif /* CHECKSUM_CHECK_TCP */  #endif /* CHECKSUM_CHECK_TCP */ 
   
  /* sanity-check header length */    /* sanity-check header length */ 
  hdrlen_bytes = TCPH_HDRLEN_BYTES(tcphdr);   hdrlen_bytes = TCPH_HDRLEN_BYTES(tcphdr);
  if ((hdrlen_bytes < TCP_HLEN) || (hdrlen_bytes > p->tot_len)) {   if ((hdrlen_bytes < TCP_HLEN) || (hdrlen_bytes > p->tot_len)) {
    LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_input: invalid header length (%"U16_F")\n", (u16_t)hdrlen_bytes));     LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_input: invalid header length (%"U16_F")\n", (u16_t)hdrlen_bytes));
    TCP_STATS_INC(tcp.lenerr);     TCP_STATS_INC(tcp.lenerr);
    goto dropped;     goto dropped;
  }   }
   
  /* Move the payload pointer in the pbuf so that it points to the   /* Move the payload pointer in the pbuf so that it points to the
     TCP data instead of the TCP header. */       TCP data instead of the TCP header. */ 
  tcphdr_optlen = (u16_t)(hdrlen_bytes - TCP_HLEN);   tcphdr_optlen = (u16_t)(hdrlen_bytes - TCP_HLEN);
  tcphdr_opt2 = NULL;   tcphdr_opt2 = NULL;
  if (p->len >= hdrlen_bytes) {   if (p->len >= hdrlen_bytes) {
    /* all options are in the first pbuf */      /* all options are in the first pbuf */ 
    tcphdr_opt1len = tcphdr_optlen;     tcphdr_opt1len = tcphdr_optlen;
    pbuf_remove_header(p, hdrlen_bytes); /* cannot fail */      pbuf_remove_header(p, hdrlen_bytes); /* cannot fail */ 
  } else {   } else {
    u16_t opt2len;     u16_t opt2len;
    /* TCP header fits into first pbuf, options don't - data is in the next pbuf */      /* TCP header fits into first pbuf, options don't - data is in the next pbuf */ 
    /* there must be a next pbuf, due to hdrlen_bytes sanity check above */      /* there must be a next pbuf, due to hdrlen_bytes sanity check above */ 
    LWIP_ASSERT("p->next != NULL", p->next != NULL);     LWIP_ASSERT("p->next != NULL", p->next != NULL);
   
    /* advance over the TCP header (cannot fail) */      /* advance over the TCP header (cannot fail) */ 
    pbuf_remove_header(p, TCP_HLEN);     pbuf_remove_header(p, TCP_HLEN);
   
    /* determine how long the first and second parts of the options are */      /* determine how long the first and second parts of the options are */ 
    tcphdr_opt1len = p->len;     tcphdr_opt1len = p->len;
    opt2len = (u16_t)(tcphdr_optlen - tcphdr_opt1len);     opt2len = (u16_t)(tcphdr_optlen - tcphdr_opt1len);
   
    /* options continue in the next pbuf: set p to zero length and hide the     /* options continue in the next pbuf: set p to zero length and hide the
        options in the next pbuf (adjusting p->tot_len) */          options in the next pbuf (adjusting p->tot_len) */ 
    pbuf_remove_header(p, tcphdr_opt1len);     pbuf_remove_header(p, tcphdr_opt1len);
   
    /* check that the options fit in the second pbuf */      /* check that the options fit in the second pbuf */ 
    if (opt2len > p->next->len) {     if (opt2len > p->next->len) {
      /* drop short packets */        /* drop short packets */ 
      LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_input: options overflow second pbuf (%"U16_F" bytes)\n", p->next->len));       LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_input: options overflow second pbuf (%"U16_F" bytes)\n", p->next->len));
      TCP_STATS_INC(tcp.lenerr);       TCP_STATS_INC(tcp.lenerr);
      goto dropped;       goto dropped;
    }     }
   
    /* remember the pointer to the second part of the options */      /* remember the pointer to the second part of the options */ 
    tcphdr_opt2 = (u8_t *)p->next->payload;     tcphdr_opt2 = (u8_t *)p->next->payload;
   
    /* advance p->next to point after the options, and manually     /* advance p->next to point after the options, and manually
        adjust p->tot_len to keep it consistent with the changed p->next */          adjust p->tot_len to keep it consistent with the changed p->next */ 
    pbuf_remove_header(p->next, opt2len);     pbuf_remove_header(p->next, opt2len);
    p->tot_len = (u16_t)(p->tot_len - opt2len);     p->tot_len = (u16_t)(p->tot_len - opt2len);
   
    LWIP_ASSERT("p->len == 0", p->len == 0);     LWIP_ASSERT("p->len == 0", p->len == 0);
    LWIP_ASSERT("p->tot_len == p->next->tot_len", p->tot_len == p->next->tot_len);     LWIP_ASSERT("p->tot_len == p->next->tot_len", p->tot_len == p->next->tot_len);
  }   }
   
  /* Convert fields in TCP header to host byte order. */    /* Convert fields in TCP header to host byte order. */ 
  tcphdr->src = lwip_ntohs(tcphdr->src);   tcphdr->src = lwip_ntohs(tcphdr->src);
  tcphdr->dest = lwip_ntohs(tcphdr->dest);   tcphdr->dest = lwip_ntohs(tcphdr->dest);
  seqno = tcphdr->seqno = lwip_ntohl(tcphdr->seqno);   seqno = tcphdr->seqno = lwip_ntohl(tcphdr->seqno);
  ackno = tcphdr->ackno = lwip_ntohl(tcphdr->ackno);   ackno = tcphdr->ackno = lwip_ntohl(tcphdr->ackno);
  tcphdr->wnd = lwip_ntohs(tcphdr->wnd);   tcphdr->wnd = lwip_ntohs(tcphdr->wnd);
   
  flags = TCPH_FLAGS(tcphdr);   flags = TCPH_FLAGS(tcphdr);
  tcplen = p->tot_len;   tcplen = p->tot_len;
  if (flags & (TCP_FIN | TCP_SYN)) {   if (flags & (TCP_FIN | TCP_SYN)) {
    tcplen++;     tcplen++;
    if (tcplen < p->tot_len) {     if (tcplen < p->tot_len) {
      /* u16_t overflow, cannot handle this */        /* u16_t overflow, cannot handle this */ 
      LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_input: length u16_t overflow, cannot handle this\n"));       LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_input: length u16_t overflow, cannot handle this\n"));
      TCP_STATS_INC(tcp.lenerr);       TCP_STATS_INC(tcp.lenerr);
      goto dropped;       goto dropped;
    }     }
  }   }
   
  /* Demultiplex an incoming segment. First, we check if it is destined   /* Demultiplex an incoming segment. First, we check if it is destined
     for an active connection. */       for an active connection. */ 
  prev = NULL;   prev = NULL;
   
  for (pcb = tcp_active_pcbs; pcb != NULL; pcb = pcb->next) {   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 != CLOSED", pcb->state != CLOSED);
    LWIP_ASSERT("tcp_input: active pcb->state != TIME-WAIT", pcb->state != TIME_WAIT);     LWIP_ASSERT("tcp_input: active pcb->state != TIME-WAIT", pcb->state != TIME_WAIT);
    LWIP_ASSERT("tcp_input: active pcb->state != LISTEN", pcb->state != LISTEN);     LWIP_ASSERT("tcp_input: active pcb->state != LISTEN", pcb->state != LISTEN);
   
    /* check if PCB is bound to specific netif */      /* check if PCB is bound to specific netif */ 
    if ((pcb->netif_idx != NETIF_NO_INDEX) &&     if ((pcb->netif_idx != NETIF_NO_INDEX) &&
        (pcb->netif_idx != netif_get_index(ip_data.current_input_netif))) {         (pcb->netif_idx != netif_get_index(ip_data.current_input_netif))) {
      prev = pcb;       prev = pcb;
      continue;       continue;
    }     }
   
    if (pcb->remote_port == tcphdr->src &&     if (pcb->remote_port == tcphdr->src &&
        pcb->local_port == tcphdr->dest &&         pcb->local_port == tcphdr->dest &&
        ip_addr_cmp(&pcb->remote_ip, ip_current_src_addr()) &&         ip_addr_cmp(&pcb->remote_ip, ip_current_src_addr()) &&
        ip_addr_cmp(&pcb->local_ip, ip_current_dest_addr())) {         ip_addr_cmp(&pcb->local_ip, ip_current_dest_addr())) {
      /* Move this PCB to the front of the list so that subsequent       /* Move this PCB to the front of the list so that subsequent
         lookups will be faster (we exploit locality in TCP segment          lookups will be faster (we exploit locality in TCP segment
         arrivals). */           arrivals). */ 
      LWIP_ASSERT("tcp_input: pcb->next != pcb (before cache)", pcb->next != pcb);       LWIP_ASSERT("tcp_input: pcb->next != pcb (before cache)", pcb->next != pcb);
      if (prev != NULL) {       if (prev != NULL) {
        prev->next = pcb->next;         prev->next = pcb->next;
        pcb->next = tcp_active_pcbs;         pcb->next = tcp_active_pcbs;
        tcp_active_pcbs = pcb;         tcp_active_pcbs = pcb;
      } else {       } else {
        TCP_STATS_INC(tcp.cachehit);         TCP_STATS_INC(tcp.cachehit);
      }       }
      LWIP_ASSERT("tcp_input: pcb->next != pcb (after cache)", pcb->next != pcb);       LWIP_ASSERT("tcp_input: pcb->next != pcb (after cache)", pcb->next != pcb);
      break;       break;
    }     }
    prev = pcb;     prev = pcb;
  }   }
   
  if (pcb == NULL) {   if (pcb == NULL) {
    /* If it did not go to an active connection, we check the connections     /* If it did not go to an active connection, we check the connections
       in the TIME-WAIT state. */         in the TIME-WAIT state. */ 
    for (pcb = tcp_tw_pcbs; pcb != NULL; pcb = pcb->next) {     for (pcb = tcp_tw_pcbs; pcb != NULL; pcb = pcb->next) {
      LWIP_ASSERT("tcp_input: TIME-WAIT pcb->state == TIME-WAIT", pcb->state == TIME_WAIT);       LWIP_ASSERT("tcp_input: TIME-WAIT pcb->state == TIME-WAIT", pcb->state == TIME_WAIT);
   
      /* check if PCB is bound to specific netif */        /* check if PCB is bound to specific netif */ 
      if ((pcb->netif_idx != NETIF_NO_INDEX) &&       if ((pcb->netif_idx != NETIF_NO_INDEX) &&
          (pcb->netif_idx != netif_get_index(ip_data.current_input_netif))) {           (pcb->netif_idx != netif_get_index(ip_data.current_input_netif))) {
        continue;         continue;
      }       }
   
      if (pcb->remote_port == tcphdr->src &&       if (pcb->remote_port == tcphdr->src &&
          pcb->local_port == tcphdr->dest &&           pcb->local_port == tcphdr->dest &&
          ip_addr_cmp(&pcb->remote_ip, ip_current_src_addr()) &&           ip_addr_cmp(&pcb->remote_ip, ip_current_src_addr()) &&
          ip_addr_cmp(&pcb->local_ip, ip_current_dest_addr())) {           ip_addr_cmp(&pcb->local_ip, ip_current_dest_addr())) {
        /* We don't really care enough to move this PCB to the front         /* 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            of the list since we are not very likely to receive that
           many segments for connections in TIME-WAIT. */             many segments for connections in TIME-WAIT. */ 
        LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_input: packed for TIME_WAITing connection.\n"));         LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_input: packed for TIME_WAITing connection.\n"));
#ifdef LWIP_HOOK_TCP_INPACKET_PCB #ifdef LWIP_HOOK_TCP_INPACKET_PCB
        if (LWIP_HOOK_TCP_INPACKET_PCB(pcb, tcphdr, tcphdr_optlen, tcphdr_opt1len,         if (LWIP_HOOK_TCP_INPACKET_PCB(pcb, tcphdr, tcphdr_optlen, tcphdr_opt1len,
                                       tcphdr_opt2, p) == ERR_OK)                                        tcphdr_opt2, p) == ERR_OK)
#endif #endif
        {         {
          tcp_timewait_input(pcb);           tcp_timewait_input(pcb);
        }         }
        pbuf_free(p);         pbuf_free(p);
        return;         return;
      }       }
    }     }
   
    /* Finally, if we still did not get a match, we check all PCBs that     /* Finally, if we still did not get a match, we check all PCBs that
       are LISTENing for incoming connections. */         are LISTENing for incoming connections. */ 
    prev = NULL;     prev = NULL;
    for (lpcb = tcp_listen_pcbs.listen_pcbs; lpcb != NULL; lpcb = lpcb->next) {     for (lpcb = tcp_listen_pcbs.listen_pcbs; lpcb != NULL; lpcb = lpcb->next) {
      /* check if PCB is bound to specific netif */        /* check if PCB is bound to specific netif */ 
      if ((lpcb->netif_idx != NETIF_NO_INDEX) &&       if ((lpcb->netif_idx != NETIF_NO_INDEX) &&
          (lpcb->netif_idx != netif_get_index(ip_data.current_input_netif))) {           (lpcb->netif_idx != netif_get_index(ip_data.current_input_netif))) {
        prev = (struct tcp_pcb *)lpcb;         prev = (struct tcp_pcb *)lpcb;
        continue;         continue;
      }       }
   
      if (lpcb->local_port == tcphdr->dest) {       if (lpcb->local_port == tcphdr->dest) {
        if (IP_IS_ANY_TYPE_VAL(lpcb->local_ip)) {         if (IP_IS_ANY_TYPE_VAL(lpcb->local_ip)) {
          /* found an ANY TYPE (IPv4/IPv6) match */            /* found an ANY TYPE (IPv4/IPv6) match */ 
#if SO_REUSE #if SO_REUSE
          lpcb_any = lpcb;           lpcb_any = lpcb;
          lpcb_prev = prev;           lpcb_prev = prev;
#else /* SO_REUSE */  #else /* SO_REUSE */ 
          break;           break;
#endif /* SO_REUSE */  #endif /* SO_REUSE */ 
        } else if (IP_ADDR_PCB_VERSION_MATCH_EXACT(lpcb, ip_current_dest_addr())) {         } else if (IP_ADDR_PCB_VERSION_MATCH_EXACT(lpcb, ip_current_dest_addr())) {
          if (ip_addr_cmp(&lpcb->local_ip, ip_current_dest_addr())) {           if (ip_addr_cmp(&lpcb->local_ip, ip_current_dest_addr())) {
            /* found an exact match */              /* found an exact match */ 
            break;             break;
          } else if (ip_addr_isany(&lpcb->local_ip)) {           } else if (ip_addr_isany(&lpcb->local_ip)) {
            /* found an ANY-match */              /* found an ANY-match */ 
#if SO_REUSE #if SO_REUSE
            lpcb_any = lpcb;             lpcb_any = lpcb;
            lpcb_prev = prev;             lpcb_prev = prev;
#else /* SO_REUSE */  #else /* SO_REUSE */ 
            break;             break;
#endif /* SO_REUSE */  #endif /* SO_REUSE */ 
          }           }
        }         }
      }       }
      prev = (struct tcp_pcb *)lpcb;       prev = (struct tcp_pcb *)lpcb;
    }     }
#if SO_REUSE #if SO_REUSE
    /* first try specific local IP */      /* first try specific local IP */ 
    if (lpcb == NULL) {     if (lpcb == NULL) {
      /* only pass to ANY if no specific local IP has been found */        /* only pass to ANY if no specific local IP has been found */ 
      lpcb = lpcb_any;       lpcb = lpcb_any;
      prev = lpcb_prev;       prev = lpcb_prev;
    }     }
#endif /* SO_REUSE */  #endif /* SO_REUSE */ 
    if (lpcb != NULL) {     if (lpcb != NULL) {
      /* Move this PCB to the front of the list so that subsequent       /* Move this PCB to the front of the list so that subsequent
         lookups will be faster (we exploit locality in TCP segment          lookups will be faster (we exploit locality in TCP segment
         arrivals). */           arrivals). */ 
      if (prev != NULL) {       if (prev != NULL) {
        ((struct tcp_pcb_listen *)prev)->next = lpcb->next;         ((struct tcp_pcb_listen *)prev)->next = lpcb->next;
        /* our successor is the remainder of the listening list */          /* our successor is the remainder of the listening list */ 
        lpcb->next = tcp_listen_pcbs.listen_pcbs;         lpcb->next = tcp_listen_pcbs.listen_pcbs;
        /* put this listening pcb at the head of the listening list */          /* put this listening pcb at the head of the listening list */ 
        tcp_listen_pcbs.listen_pcbs = lpcb;         tcp_listen_pcbs.listen_pcbs = lpcb;
      } else {       } else {
        TCP_STATS_INC(tcp.cachehit);         TCP_STATS_INC(tcp.cachehit);
      }       }
   
      LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_input: packed for LISTENing connection.\n"));       LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_input: packed for LISTENing connection.\n"));
#ifdef LWIP_HOOK_TCP_INPACKET_PCB #ifdef LWIP_HOOK_TCP_INPACKET_PCB
      if (LWIP_HOOK_TCP_INPACKET_PCB((struct tcp_pcb *)lpcb, tcphdr, tcphdr_optlen,       if (LWIP_HOOK_TCP_INPACKET_PCB((struct tcp_pcb *)lpcb, tcphdr, tcphdr_optlen,
                                     tcphdr_opt1len, tcphdr_opt2, p) == ERR_OK)                                      tcphdr_opt1len, tcphdr_opt2, p) == ERR_OK)
#endif #endif
      {       {
        tcp_listen_input(lpcb);         tcp_listen_input(lpcb);
      }       }
      pbuf_free(p);       pbuf_free(p);
      return;       return;
    }     }
  }   }
   
#if TCP_INPUT_DEBUG #if TCP_INPUT_DEBUG
  LWIP_DEBUGF(TCP_INPUT_DEBUG, ("+-+-+-+-+-+-+-+-+-+-+-+-+-+- tcp_input: flags "));   LWIP_DEBUGF(TCP_INPUT_DEBUG, ("+-+-+-+-+-+-+-+-+-+-+-+-+-+- tcp_input: flags "));
  tcp_debug_print_flags(TCPH_FLAGS(tcphdr));   tcp_debug_print_flags(TCPH_FLAGS(tcphdr));
  LWIP_DEBUGF(TCP_INPUT_DEBUG, ("-+-+-+-+-+-+-+-+-+-+-+-+-+-+\n"));   LWIP_DEBUGF(TCP_INPUT_DEBUG, ("-+-+-+-+-+-+-+-+-+-+-+-+-+-+\n"));
#endif /* TCP_INPUT_DEBUG */  #endif /* TCP_INPUT_DEBUG */ 
   
   
#ifdef LWIP_HOOK_TCP_INPACKET_PCB #ifdef LWIP_HOOK_TCP_INPACKET_PCB
  if ((pcb != NULL) && LWIP_HOOK_TCP_INPACKET_PCB(pcb, tcphdr, tcphdr_optlen,   if ((pcb != NULL) && LWIP_HOOK_TCP_INPACKET_PCB(pcb, tcphdr, tcphdr_optlen,
      tcphdr_opt1len, tcphdr_opt2, p) != ERR_OK) {       tcphdr_opt1len, tcphdr_opt2, p) != ERR_OK) {
    pbuf_free(p);     pbuf_free(p);
    return;     return;
  }   }
#endif #endif
  if (pcb != NULL) {   if (pcb != NULL) {
    /* The incoming segment belongs to a connection. */      /* The incoming segment belongs to a connection. */ 
#if TCP_INPUT_DEBUG #if TCP_INPUT_DEBUG
    tcp_debug_print_state(pcb->state);     tcp_debug_print_state(pcb->state);
#endif /* TCP_INPUT_DEBUG */  #endif /* TCP_INPUT_DEBUG */ 
   
    /* Set up a tcp_seg structure. */      /* Set up a tcp_seg structure. */ 
    inseg.next = NULL;     inseg.next = NULL;
    inseg.len = p->tot_len;     inseg.len = p->tot_len;
    inseg.p = p;     inseg.p = p;
    inseg.tcphdr = tcphdr;     inseg.tcphdr = tcphdr;
   
    recv_data = NULL;     recv_data = NULL;
    recv_flags = 0;     recv_flags = 0;
    recv_acked = 0;     recv_acked = 0;
   
    if (flags & TCP_PSH) {     if (flags & TCP_PSH) {
      p->flags |= PBUF_FLAG_PUSH;       p->flags |= PBUF_FLAG_PUSH;
    }     }
   
    /* If there is data which was previously "refused" by upper layer */      /* If there is data which was previously "refused" by upper layer */ 
    if (pcb->refused_data != NULL) {     if (pcb->refused_data != NULL) {
      if ((tcp_process_refused_data(pcb) == ERR_ABRT) ||       if ((tcp_process_refused_data(pcb) == ERR_ABRT) ||
          ((pcb->refused_data != NULL) && (tcplen > 0))) {           ((pcb->refused_data != NULL) && (tcplen > 0))) {
        /* pcb has been aborted or refused data is still refused and the new         /* pcb has been aborted or refused data is still refused and the new
           segment contains data */             segment contains data */ 
        if (pcb->rcv_ann_wnd == 0) {         if (pcb->rcv_ann_wnd == 0) {
          /* this is a zero-window probe, we respond to it with current RCV.NXT           /* this is a zero-window probe, we respond to it with current RCV.NXT
          and drop the data segment */            and drop the data segment */ 
          tcp_send_empty_ack(pcb);           tcp_send_empty_ack(pcb);
        }         }
        TCP_STATS_INC(tcp.drop);         TCP_STATS_INC(tcp.drop);
        MIB2_STATS_INC(mib2.tcpinerrs);         MIB2_STATS_INC(mib2.tcpinerrs);
        goto aborted;         goto aborted;
      }       }
    }     }
    tcp_input_pcb = pcb;     tcp_input_pcb = pcb;
    err = tcp_process(pcb);     err = tcp_process(pcb);
    /* A return value of ERR_ABRT means that tcp_abort() was called     /* 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. */         and that the pcb has been freed. If so, we don't do anything. */ 
    if (err != ERR_ABRT) {     if (err != ERR_ABRT) {
      if (recv_flags & TF_RESET) {       if (recv_flags & TF_RESET) {
        /* TF_RESET means that the connection was reset by the other         /* TF_RESET means that the connection was reset by the other
           end. We then call the error callback to inform the            end. We then call the error callback to inform the
           application that the connection is dead before we            application that the connection is dead before we
           deallocate the PCB. */             deallocate the PCB. */ 
        TCP_EVENT_ERR(pcb->state, pcb->errf, pcb->callback_arg, ERR_RST);         TCP_EVENT_ERR(pcb->state, pcb->errf, pcb->callback_arg, ERR_RST);
        tcp_pcb_remove(&tcp_active_pcbs, pcb);         tcp_pcb_remove(&tcp_active_pcbs, pcb);
        tcp_free(pcb);         tcp_free(pcb);
      } else {       } else {
        err = ERR_OK;         err = ERR_OK;
        /* If the application has registered a "sent" function to be         /* If the application has registered a "sent" function to be
           called when new send buffer space is available, we call it            called when new send buffer space is available, we call it
           now. */             now. */ 
        if (recv_acked > 0) {         if (recv_acked > 0) {
          u16_t acked16;           u16_t acked16;
#if LWIP_WND_SCALE #if LWIP_WND_SCALE
          /* recv_acked is u32_t but the sent callback only takes a u16_t,           /* recv_acked is u32_t but the sent callback only takes a u16_t,
             so we might have to call it multiple times. */               so we might have to call it multiple times. */ 
          u32_t acked = recv_acked;           u32_t acked = recv_acked;
          while (acked > 0) {           while (acked > 0) {
            acked16 = (u16_t)LWIP_MIN(acked, 0xffffu);             acked16 = (u16_t)LWIP_MIN(acked, 0xffffu);
            acked -= acked16;             acked -= acked16;
#else #else
          {           {
            acked16 = recv_acked;             acked16 = recv_acked;
#endif #endif
            TCP_EVENT_SENT(pcb, (u16_t)acked16, err);             TCP_EVENT_SENT(pcb, (u16_t)acked16, err);
            if (err == ERR_ABRT) {             if (err == ERR_ABRT) {
.                /* 若recv_data不为NULL则需要释放pbuf */ 
                if (recv_data != NULL) {
                  pbuf_free(recv_data);
                }
              goto aborted;               goto aborted;
            }             }
          }           }
          recv_acked = 0;           recv_acked = 0;
        }         }
        if (tcp_input_delayed_close(pcb)) {         if (tcp_input_delayed_close(pcb)) {
.            /* 若recv_data不为NULL则需要释放pbuf */ 
            if (recv_data != NULL) {
              pbuf_free(recv_data);
            }
          goto aborted;           goto aborted;
        }         }
#if TCP_QUEUE_OOSEQ && LWIP_WND_SCALE #if TCP_QUEUE_OOSEQ && LWIP_WND_SCALE
        while (recv_data != NULL) {         while (recv_data != NULL) {
          struct pbuf *rest = NULL;           struct pbuf *rest = NULL;
          pbuf_split_64k(recv_data, &rest);           pbuf_split_64k(recv_data, &rest);
#else /* TCP_QUEUE_OOSEQ && LWIP_WND_SCALE */  #else /* TCP_QUEUE_OOSEQ && LWIP_WND_SCALE */ 
        if (recv_data != NULL) {         if (recv_data != NULL) {
#endif /* TCP_QUEUE_OOSEQ && LWIP_WND_SCALE */  #endif /* TCP_QUEUE_OOSEQ && LWIP_WND_SCALE */ 
   
          LWIP_ASSERT("pcb->refused_data == NULL", pcb->refused_data == NULL);           LWIP_ASSERT("pcb->refused_data == NULL", pcb->refused_data == NULL);
          if (pcb->flags & TF_RXCLOSED) {           if (pcb->flags & TF_RXCLOSED) {
            /* received data although already closed -> abort (send RST) to             /* received data although already closed -> abort (send RST) to
               notify the remote host that not all data has been processed */                 notify the remote host that not all data has been processed */ 
            pbuf_free(recv_data);             pbuf_free(recv_data);
#if TCP_QUEUE_OOSEQ && LWIP_WND_SCALE #if TCP_QUEUE_OOSEQ && LWIP_WND_SCALE
            if (rest != NULL) {             if (rest != NULL) {
              pbuf_free(rest);               pbuf_free(rest);
            }             }
#endif /* TCP_QUEUE_OOSEQ && LWIP_WND_SCALE */  #endif /* TCP_QUEUE_OOSEQ && LWIP_WND_SCALE */ 
            tcp_abort(pcb);             tcp_abort(pcb);
            goto aborted;             goto aborted;
          }           }
   
          /* Notify application that data has been received. */            /* Notify application that data has been received. */ 
          TCP_EVENT_RECV(pcb, recv_data, ERR_OK, err);           TCP_EVENT_RECV(pcb, recv_data, ERR_OK, err);
          if (err == ERR_ABRT) {           if (err == ERR_ABRT) {
#if TCP_QUEUE_OOSEQ && LWIP_WND_SCALE #if TCP_QUEUE_OOSEQ && LWIP_WND_SCALE
            if (rest != NULL) {             if (rest != NULL) {
              pbuf_free(rest);               pbuf_free(rest);
            }             }
#endif /* TCP_QUEUE_OOSEQ && LWIP_WND_SCALE */  #endif /* TCP_QUEUE_OOSEQ && LWIP_WND_SCALE */ 
            goto aborted;             goto aborted;
          }           }
   
          /* If the upper layer can't receive this data, store it */            /* If the upper layer can't receive this data, store it */ 
          if (err != ERR_OK) {           if (err != ERR_OK) {
#if TCP_QUEUE_OOSEQ && LWIP_WND_SCALE #if TCP_QUEUE_OOSEQ && LWIP_WND_SCALE
            if (rest != NULL) {             if (rest != NULL) {
              pbuf_cat(recv_data, rest);               pbuf_cat(recv_data, rest);
            }             }
#endif /* TCP_QUEUE_OOSEQ && LWIP_WND_SCALE */  #endif /* TCP_QUEUE_OOSEQ && LWIP_WND_SCALE */ 
            pcb->refused_data = recv_data;             pcb->refused_data = recv_data;
            LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_input: keep incoming packet, because pcb is \"full\"\n"));             LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_input: keep incoming packet, because pcb is \"full\"\n"));
#if TCP_QUEUE_OOSEQ && LWIP_WND_SCALE #if TCP_QUEUE_OOSEQ && LWIP_WND_SCALE
            break;             break;
          } else {           } else {
            /* Upper layer received the data, go on with the rest if > 64K */              /* Upper layer received the data, go on with the rest if > 64K */ 
            recv_data = rest;             recv_data = rest;
#endif /* TCP_QUEUE_OOSEQ && LWIP_WND_SCALE */  #endif /* TCP_QUEUE_OOSEQ && LWIP_WND_SCALE */ 
          }           }
        }         }
   
        /* If a FIN segment was received, we call the callback         /* If a FIN segment was received, we call the callback
           function with a NULL buffer to indicate EOF. */             function with a NULL buffer to indicate EOF. */ 
        if (recv_flags & TF_GOT_FIN) {         if (recv_flags & TF_GOT_FIN) {
          if (pcb->refused_data != NULL) {           if (pcb->refused_data != NULL) {
            /* Delay this if we have refused data. */              /* Delay this if we have refused data. */ 
            pcb->refused_data->flags |= PBUF_FLAG_TCP_FIN;             pcb->refused_data->flags |= PBUF_FLAG_TCP_FIN;
          } else {           } else {
            /* correct rcv_wnd as the application won't call tcp_recved()             /* correct rcv_wnd as the application won't call tcp_recved()
               for the FIN's seqno */                 for the FIN's seqno */ 
            if (pcb->rcv_wnd != TCP_WND_MAX(pcb)) {             if (pcb->rcv_wnd != TCP_WND_MAX(pcb)) {
              pcb->rcv_wnd++;               pcb->rcv_wnd++;
            }             }
            TCP_EVENT_CLOSED(pcb, err);             TCP_EVENT_CLOSED(pcb, err);
            if (err == ERR_ABRT) {             if (err == ERR_ABRT) {
              goto aborted;               goto aborted;
            }             }
          }           }
        }         }
   
        tcp_input_pcb = NULL;         tcp_input_pcb = NULL;
        if (tcp_input_delayed_close(pcb)) {         if (tcp_input_delayed_close(pcb)) {
          goto aborted;           goto aborted;
        }         }
        /* Try to send something out. */          /* Try to send something out. */ 
        tcp_output(pcb);         tcp_output(pcb);
#if TCP_INPUT_DEBUG #if TCP_INPUT_DEBUG
#if TCP_DEBUG #if TCP_DEBUG
        tcp_debug_print_state(pcb->state);         tcp_debug_print_state(pcb->state);
#endif /* TCP_DEBUG */  #endif /* TCP_DEBUG */ 
#endif /* TCP_INPUT_DEBUG */  #endif /* TCP_INPUT_DEBUG */ 
      }       }
    }     }
    /* Jump target if pcb has been aborted in a callback (by calling tcp_abort()).     /* Jump target if pcb has been aborted in a callback (by calling tcp_abort()).
       Below this line, 'pcb' may not be dereferenced! */         Below this line, 'pcb' may not be dereferenced! */ 
aborted: aborted:
    tcp_input_pcb = NULL;     tcp_input_pcb = NULL;
    recv_data = NULL;     recv_data = NULL;
   
    /* give up our reference to inseg.p */      /* give up our reference to inseg.p */ 
    if (inseg.p != NULL) {     if (inseg.p != NULL) {
      pbuf_free(inseg.p);       pbuf_free(inseg.p);
      inseg.p = NULL;       inseg.p = NULL;
    }     }
  } else {   } else {
    /* If no matching PCB was found, send a TCP RST (reset) to the     /* If no matching PCB was found, send a TCP RST (reset) to the
       sender. */         sender. */ 
    LWIP_DEBUGF(TCP_RST_DEBUG, ("tcp_input: no PCB match found, resetting.\n"));     LWIP_DEBUGF(TCP_RST_DEBUG, ("tcp_input: no PCB match found, resetting.\n"));
    if (!(TCPH_FLAGS(tcphdr) & TCP_RST)) {     if (!(TCPH_FLAGS(tcphdr) & TCP_RST)) {
      TCP_STATS_INC(tcp.proterr);       TCP_STATS_INC(tcp.proterr);
      TCP_STATS_INC(tcp.drop);       TCP_STATS_INC(tcp.drop);
      tcp_rst(NULL, ackno, seqno + tcplen, ip_current_dest_addr(),       tcp_rst(NULL, ackno, seqno + tcplen, ip_current_dest_addr(),
              ip_current_src_addr(), tcphdr->dest, tcphdr->src);               ip_current_src_addr(), tcphdr->dest, tcphdr->src);
    }     }
    pbuf_free(p);     pbuf_free(p);
  }   }
   
  LWIP_ASSERT("tcp_input: tcp_pcbs_sane()", tcp_pcbs_sane());   LWIP_ASSERT("tcp_input: tcp_pcbs_sane()", tcp_pcbs_sane());
  PERF_STOP("tcp_input");   PERF_STOP("tcp_input");
  return;   return;
dropped: dropped:
  TCP_STATS_INC(tcp.drop);   TCP_STATS_INC(tcp.drop);
  MIB2_STATS_INC(mib2.tcpinerrs);   MIB2_STATS_INC(mib2.tcpinerrs);
  pbuf_free(p);   pbuf_free(p);
} }
   
/** Called from tcp_input to check for TF_CLOSED flag. This results in closing /** Called from tcp_input to check for TF_CLOSED flag. This results in closing
 * and deallocating a pcb at the correct place to ensure noone references it  * and deallocating a pcb at the correct place to ensure noone references it
 * any more.  * any more.
 * @returns 1 if the pcb has been closed and deallocated, 0 otherwise  * @returns 1 if the pcb has been closed and deallocated, 0 otherwise
 */   */ 
static int static int
tcp_input_delayed_close(struct tcp_pcb *pcb) tcp_input_delayed_close(struct tcp_pcb *pcb)
{ {
  LWIP_ASSERT("tcp_input_delayed_close: invalid pcb", pcb != NULL);   LWIP_ASSERT("tcp_input_delayed_close: invalid pcb", pcb != NULL);
   
  if (recv_flags & TF_CLOSED) {   if (recv_flags & TF_CLOSED) {
    /* The connection has been closed and we will deallocate the     /* The connection has been closed and we will deallocate the
        PCB. */          PCB. */ 
    if (!(pcb->flags & TF_RXCLOSED)) {     if (!(pcb->flags & TF_RXCLOSED)) {
      /* Connection closed although the application has only shut down the       /* Connection closed although the application has only shut down the
          tx side: call the PCB's err callback and indicate the closure to           tx side: call the PCB's err callback and indicate the closure to
          ensure the application doesn't continue using the PCB. */            ensure the application doesn't continue using the PCB. */ 
      TCP_EVENT_ERR(pcb->state, pcb->errf, pcb->callback_arg, ERR_CLSD);       TCP_EVENT_ERR(pcb->state, pcb->errf, pcb->callback_arg, ERR_CLSD);
    }     }
    tcp_pcb_remove(&tcp_active_pcbs, pcb);     tcp_pcb_remove(&tcp_active_pcbs, pcb);
    tcp_free(pcb);     tcp_free(pcb);
    return 1;     return 1;
  }   }
  return 0;   return 0;
} }
   
/** /**
 * Called by tcp_input() when a segment arrives for a listening  * Called by tcp_input() when a segment arrives for a listening
 * connection (from tcp_input()).  * connection (from tcp_input()).
 *  *
 * @param pcb the tcp_pcb_listen for which a segment arrived  * @param pcb the tcp_pcb_listen for which a segment arrived
 *  *
 * @note the segment which arrived is saved in global variables, therefore only the pcb  * @note the segment which arrived is saved in global variables, therefore only the pcb
 *       involved is passed as a parameter to this function  *       involved is passed as a parameter to this function
 */   */ 
static void static void
tcp_listen_input(struct tcp_pcb_listen *pcb) tcp_listen_input(struct tcp_pcb_listen *pcb)
{ {
  struct tcp_pcb *npcb;   struct tcp_pcb *npcb;
  u32_t iss;   u32_t iss;
  err_t rc;   err_t rc;
   
  if (flags & TCP_RST) {   if (flags & TCP_RST) {
    /* An incoming RST should be ignored. Return. */      /* An incoming RST should be ignored. Return. */ 
    return;     return;
  }   }
   
  LWIP_ASSERT("tcp_listen_input: invalid pcb", pcb != NULL);   LWIP_ASSERT("tcp_listen_input: invalid pcb", pcb != NULL);
   
  /* In the LISTEN state, we check for incoming SYN segments,   /* In the LISTEN state, we check for incoming SYN segments,
     creates a new PCB, and responds with a SYN|ACK. */       creates a new PCB, and responds with a SYN|ACK. */ 
  if (flags & TCP_ACK) {   if (flags & TCP_ACK) {
    /* For incoming segments with the ACK flag set, respond with a     /* For incoming segments with the ACK flag set, respond with a
       RST. */         RST. */ 
    LWIP_DEBUGF(TCP_RST_DEBUG, ("tcp_listen_input: ACK in LISTEN, sending reset\n"));     LWIP_DEBUGF(TCP_RST_DEBUG, ("tcp_listen_input: ACK in LISTEN, sending reset\n"));
    tcp_rst((const struct tcp_pcb *)pcb, ackno, seqno + tcplen, ip_current_dest_addr(),     tcp_rst((const struct tcp_pcb *)pcb, ackno, seqno + tcplen, ip_current_dest_addr(),
            ip_current_src_addr(), tcphdr->dest, tcphdr->src);             ip_current_src_addr(), tcphdr->dest, tcphdr->src);
  } else if (flags & TCP_SYN) {   } else if (flags & TCP_SYN) {
    LWIP_DEBUGF(TCP_DEBUG, ("TCP connection request %"U16_F" -> %"U16_F".\n", tcphdr->src, tcphdr->dest));     LWIP_DEBUGF(TCP_DEBUG, ("TCP connection request %"U16_F" -> %"U16_F".\n", tcphdr->src, tcphdr->dest));
#if TCP_LISTEN_BACKLOG #if TCP_LISTEN_BACKLOG
    if (pcb->accepts_pending >= pcb->backlog) {     if (pcb->accepts_pending >= pcb->backlog) {
      LWIP_DEBUGF(TCP_DEBUG, ("tcp_listen_input: listen backlog exceeded for port %"U16_F"\n", tcphdr->dest));       LWIP_DEBUGF(TCP_DEBUG, ("tcp_listen_input: listen backlog exceeded for port %"U16_F"\n", tcphdr->dest));
      return;       return;
    }     }
#endif /* TCP_LISTEN_BACKLOG */  #endif /* TCP_LISTEN_BACKLOG */ 
    npcb = tcp_alloc(pcb->prio);     npcb = tcp_alloc(pcb->prio);
    /* If a new PCB could not be created (probably due to lack of memory),     /* If a new PCB could not be created (probably due to lack of memory),
       we don't do anything, but rely on the sender will retransmit the        we don't do anything, but rely on the sender will retransmit the
       SYN at a time when we have more memory available. */         SYN at a time when we have more memory available. */ 
    if (npcb == NULL) {     if (npcb == NULL) {
      err_t err;       err_t err;
      LWIP_DEBUGF(TCP_DEBUG, ("tcp_listen_input: could not allocate PCB\n"));       LWIP_DEBUGF(TCP_DEBUG, ("tcp_listen_input: could not allocate PCB\n"));
      TCP_STATS_INC(tcp.memerr);       TCP_STATS_INC(tcp.memerr);
      TCP_EVENT_ACCEPT(pcb, NULL, pcb->callback_arg, ERR_MEM, err);       TCP_EVENT_ACCEPT(pcb, NULL, pcb->callback_arg, ERR_MEM, err);
      LWIP_UNUSED_ARG(err); /* err not useful here */        LWIP_UNUSED_ARG(err); /* err not useful here */ 
      return;       return;
    }     }
#if TCP_LISTEN_BACKLOG #if TCP_LISTEN_BACKLOG
    pcb->accepts_pending++;     pcb->accepts_pending++;
    tcp_set_flags(npcb, TF_BACKLOGPEND);     tcp_set_flags(npcb, TF_BACKLOGPEND);
#endif /* TCP_LISTEN_BACKLOG */  #endif /* TCP_LISTEN_BACKLOG */ 
    /* Set up the new PCB. */      /* Set up the new PCB. */ 
    ip_addr_copy(npcb->local_ip, *ip_current_dest_addr());     ip_addr_copy(npcb->local_ip, *ip_current_dest_addr());
    ip_addr_copy(npcb->remote_ip, *ip_current_src_addr());     ip_addr_copy(npcb->remote_ip, *ip_current_src_addr());
    npcb->local_port = pcb->local_port;     npcb->local_port = pcb->local_port;
    npcb->remote_port = tcphdr->src;     npcb->remote_port = tcphdr->src;
    npcb->state = SYN_RCVD;     npcb->state = SYN_RCVD;
    npcb->rcv_nxt = seqno + 1;     npcb->rcv_nxt = seqno + 1;
    npcb->rcv_ann_right_edge = npcb->rcv_nxt;     npcb->rcv_ann_right_edge = npcb->rcv_nxt;
    iss = tcp_next_iss(npcb);     iss = tcp_next_iss(npcb);
    npcb->snd_wl2 = iss;     npcb->snd_wl2 = iss;
    npcb->snd_nxt = iss;     npcb->snd_nxt = iss;
    npcb->lastack = iss;     npcb->lastack = iss;
    npcb->snd_lbb = iss;     npcb->snd_lbb = iss;
    npcb->snd_wl1 = seqno - 1;/* initialise to seqno-1 to force window update */      npcb->snd_wl1 = seqno - 1;/* initialise to seqno-1 to force window update */ 
    npcb->callback_arg = pcb->callback_arg;     npcb->callback_arg = pcb->callback_arg;
#if LWIP_CALLBACK_API || TCP_LISTEN_BACKLOG #if LWIP_CALLBACK_API || TCP_LISTEN_BACKLOG
    npcb->listener = pcb;     npcb->listener = pcb;
#endif /* LWIP_CALLBACK_API || TCP_LISTEN_BACKLOG */  #endif /* LWIP_CALLBACK_API || TCP_LISTEN_BACKLOG */ 
    /* inherit socket options */      /* inherit socket options */ 
    npcb->so_options = pcb->so_options & SOF_INHERITED;     npcb->so_options = pcb->so_options & SOF_INHERITED;
    npcb->netif_idx = pcb->netif_idx;     npcb->netif_idx = pcb->netif_idx;
    /* Register the new PCB so that we can begin receiving segments     /* Register the new PCB so that we can begin receiving segments
       for it. */         for it. */ 
    TCP_REG_ACTIVE(npcb);     TCP_REG_ACTIVE(npcb);
   
    /* Parse any options in the SYN. */      /* Parse any options in the SYN. */ 
    tcp_parseopt(npcb);     tcp_parseopt(npcb);
    npcb->snd_wnd = tcphdr->wnd;     npcb->snd_wnd = tcphdr->wnd;
    npcb->snd_wnd_max = npcb->snd_wnd;     npcb->snd_wnd_max = npcb->snd_wnd;
   
#if TCP_CALCULATE_EFF_SEND_MSS #if TCP_CALCULATE_EFF_SEND_MSS
    npcb->mss = tcp_eff_send_mss(npcb->mss, &npcb->local_ip, &npcb->remote_ip);     npcb->mss = tcp_eff_send_mss(npcb->mss, &npcb->local_ip, &npcb->remote_ip);
#endif /* TCP_CALCULATE_EFF_SEND_MSS */  #endif /* TCP_CALCULATE_EFF_SEND_MSS */ 
   
    MIB2_STATS_INC(mib2.tcppassiveopens);     MIB2_STATS_INC(mib2.tcppassiveopens);
   
#if LWIP_TCP_PCB_NUM_EXT_ARGS #if LWIP_TCP_PCB_NUM_EXT_ARGS
    if (tcp_ext_arg_invoke_callbacks_passive_open(pcb, npcb) != ERR_OK) {     if (tcp_ext_arg_invoke_callbacks_passive_open(pcb, npcb) != ERR_OK) {
      tcp_abandon(npcb, 0);       tcp_abandon(npcb, 0);
      return;       return;
    }     }
#endif #endif
   
    /* Send a SYN|ACK together with the MSS option. */      /* Send a SYN|ACK together with the MSS option. */ 
    rc = tcp_enqueue_flags(npcb, TCP_SYN | TCP_ACK);     rc = tcp_enqueue_flags(npcb, TCP_SYN | TCP_ACK);
    if (rc != ERR_OK) {     if (rc != ERR_OK) {
      tcp_abandon(npcb, 0);       tcp_abandon(npcb, 0);
      return;       return;
    }     }
    tcp_output(npcb);     tcp_output(npcb);
  }   }
  return;   return;
} }
   
/** /**
 * Called by tcp_input() when a segment arrives for a connection in  * Called by tcp_input() when a segment arrives for a connection in
 * TIME_WAIT.  * TIME_WAIT.
 *  *
 * @param pcb the tcp_pcb for which a segment arrived  * @param pcb the tcp_pcb for which a segment arrived
 *  *
 * @note the segment which arrived is saved in global variables, therefore only the pcb  * @note the segment which arrived is saved in global variables, therefore only the pcb
 *       involved is passed as a parameter to this function  *       involved is passed as a parameter to this function
 */   */ 
static void static void
tcp_timewait_input(struct tcp_pcb *pcb) tcp_timewait_input(struct tcp_pcb *pcb)
{ {
  /* RFC 1337: in TIME_WAIT, ignore RST and ACK FINs + any 'acceptable' segments */    /* RFC 1337: in TIME_WAIT, ignore RST and ACK FINs + any 'acceptable' segments */ 
  /* RFC 793 3.9 Event Processing - Segment Arrives:   /* RFC 793 3.9 Event Processing - Segment Arrives:
   * - first check sequence number - we skip that one in TIME_WAIT (always    * - first check sequence number - we skip that one in TIME_WAIT (always
   *   acceptable since we only send ACKs)    *   acceptable since we only send ACKs)
   * - second check the RST bit (... return) */     * - second check the RST bit (... return) */ 
  if (flags & TCP_RST) {   if (flags & TCP_RST) {
    return;     return;
  }   }
   
  LWIP_ASSERT("tcp_timewait_input: invalid pcb", pcb != NULL);   LWIP_ASSERT("tcp_timewait_input: invalid pcb", pcb != NULL);
   
  /* - fourth, check the SYN bit, */    /* - fourth, check the SYN bit, */ 
  if (flags & TCP_SYN) {   if (flags & TCP_SYN) {
    /* If an incoming segment is not acceptable, an acknowledgment     /* If an incoming segment is not acceptable, an acknowledgment
       should be sent in reply */         should be sent in reply */ 
    if (TCP_SEQ_BETWEEN(seqno, pcb->rcv_nxt, pcb->rcv_nxt + pcb->rcv_wnd)) {     if (TCP_SEQ_BETWEEN(seqno, pcb->rcv_nxt, pcb->rcv_nxt + pcb->rcv_wnd)) {
      /* If the SYN is in the window it is an error, send a reset */        /* If the SYN is in the window it is an error, send a reset */ 
      tcp_rst(pcb, ackno, seqno + tcplen, ip_current_dest_addr(),       tcp_rst(pcb, ackno, seqno + tcplen, ip_current_dest_addr(),
              ip_current_src_addr(), tcphdr->dest, tcphdr->src);               ip_current_src_addr(), tcphdr->dest, tcphdr->src);
      return;       return;
    }     }
  } else if (flags & TCP_FIN) {   } else if (flags & TCP_FIN) {
    /* - eighth, check the FIN bit: Remain in the TIME-WAIT state.     /* - eighth, check the FIN bit: Remain in the TIME-WAIT state.
         Restart the 2 MSL time-wait timeout.*/           Restart the 2 MSL time-wait timeout.*/ 
    pcb->tmr = tcp_ticks;     pcb->tmr = tcp_ticks;
  }   }
   
  if ((tcplen > 0)) {   if ((tcplen > 0)) {
    /* Acknowledge data, FIN or out-of-window SYN */      /* Acknowledge data, FIN or out-of-window SYN */ 
    tcp_ack_now(pcb);     tcp_ack_now(pcb);
    tcp_output(pcb);     tcp_output(pcb);
  }   }
  return;   return;
} }
   
/** /**
 * Implements the TCP state machine. Called by tcp_input. In some  * Implements the TCP state machine. Called by tcp_input. In some
 * states tcp_receive() is called to receive data. The tcp_seg  * states tcp_receive() is called to receive data. The tcp_seg
 * argument will be freed by the caller (tcp_input()) unless the  * argument will be freed by the caller (tcp_input()) unless the
 * recv_data pointer in the pcb is set.  * recv_data pointer in the pcb is set.
 *  *
 * @param pcb the tcp_pcb for which a segment arrived  * @param pcb the tcp_pcb for which a segment arrived
 *  *
 * @note the segment which arrived is saved in global variables, therefore only the pcb  * @note the segment which arrived is saved in global variables, therefore only the pcb
 *       involved is passed as a parameter to this function  *       involved is passed as a parameter to this function
 */   */ 
static err_t static err_t
tcp_process(struct tcp_pcb *pcb) tcp_process(struct tcp_pcb *pcb)
{ {
  struct tcp_seg *rseg;   struct tcp_seg *rseg;
  u8_t acceptable = 0;   u8_t acceptable = 0;
  err_t err;   err_t err;
   
  err = ERR_OK;   err = ERR_OK;
   
  LWIP_ASSERT("tcp_process: invalid pcb", pcb != NULL);   LWIP_ASSERT("tcp_process: invalid pcb", pcb != NULL);
   
  /* Process incoming RST segments. */    /* Process incoming RST segments. */ 
  if (flags & TCP_RST) {   if (flags & TCP_RST) {
    /* First, determine if the reset is acceptable. */      /* First, determine if the reset is acceptable. */ 
    if (pcb->state == SYN_SENT) {     if (pcb->state == SYN_SENT) {
      /* "In the SYN-SENT state (a RST received in response to an initial SYN),       /* "In the SYN-SENT state (a RST received in response to an initial SYN),
          the RST is acceptable if the ACK field acknowledges the SYN." */            the RST is acceptable if the ACK field acknowledges the SYN." */ 
      if (ackno == pcb->snd_nxt) {       if (ackno == pcb->snd_nxt) {
        acceptable = 1;         acceptable = 1;
      }       }
    } else {     } else {
      /* "In all states except SYN-SENT, all reset (RST) segments are validated       /* "In all states except SYN-SENT, all reset (RST) segments are validated
          by checking their SEQ-fields." */            by checking their SEQ-fields." */ 
      if (seqno == pcb->rcv_nxt) {       if (seqno == pcb->rcv_nxt) {
        acceptable = 1;         acceptable = 1;
      } else  if (TCP_SEQ_BETWEEN(seqno, pcb->rcv_nxt,       } else  if (TCP_SEQ_BETWEEN(seqno, pcb->rcv_nxt,
                                  pcb->rcv_nxt + pcb->rcv_wnd)) {                                   pcb->rcv_nxt + pcb->rcv_wnd)) {
        /* If the sequence number is inside the window, we send a challenge ACK         /* If the sequence number is inside the window, we send a challenge ACK
           and wait for a re-send with matching sequence number.            and wait for a re-send with matching sequence number.
           This follows RFC 5961 section 3.2 and addresses CVE-2004-0230            This follows RFC 5961 section 3.2 and addresses CVE-2004-0230
           (RST spoofing attack), which is present in RFC 793 RST handling. */             (RST spoofing attack), which is present in RFC 793 RST handling. */ 
        tcp_ack_now(pcb);         tcp_ack_now(pcb);
      }       }
    }     }
   
    if (acceptable) {     if (acceptable) {
      LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_process: Connection RESET\n"));       LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_process: Connection RESET\n"));
      LWIP_ASSERT("tcp_input: pcb->state != CLOSED", pcb->state != CLOSED);       LWIP_ASSERT("tcp_input: pcb->state != CLOSED", pcb->state != CLOSED);
      recv_flags |= TF_RESET;       recv_flags |= TF_RESET;
      tcp_clear_flags(pcb, TF_ACK_DELAY);       tcp_clear_flags(pcb, TF_ACK_DELAY);
      return ERR_RST;       return ERR_RST;
    } else {     } else {
      LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_process: unacceptable reset seqno %"U32_F" rcv_nxt %"U32_F"\n",       LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_process: unacceptable reset seqno %"U32_F" rcv_nxt %"U32_F"\n",
                                    seqno, pcb->rcv_nxt));                                     seqno, pcb->rcv_nxt));
      LWIP_DEBUGF(TCP_DEBUG, ("tcp_process: unacceptable reset seqno %"U32_F" rcv_nxt %"U32_F"\n",       LWIP_DEBUGF(TCP_DEBUG, ("tcp_process: unacceptable reset seqno %"U32_F" rcv_nxt %"U32_F"\n",
                              seqno, pcb->rcv_nxt));                               seqno, pcb->rcv_nxt));
      return ERR_OK;       return ERR_OK;
    }     }
  }   }
   
  if ((flags & TCP_SYN) && (pcb->state != SYN_SENT && pcb->state != SYN_RCVD)) {   if ((flags & TCP_SYN) && (pcb->state != SYN_SENT && pcb->state != SYN_RCVD)) {
    /* Cope with new connection attempt after remote end crashed */      /* Cope with new connection attempt after remote end crashed */ 
    tcp_ack_now(pcb);     tcp_ack_now(pcb);
    return ERR_OK;     return ERR_OK;
  }   }
   
  if ((pcb->flags & TF_RXCLOSED) == 0) {   if ((pcb->flags & TF_RXCLOSED) == 0) {
    /* Update the PCB (in)activity timer unless rx is closed (see tcp_shutdown) */      /* Update the PCB (in)activity timer unless rx is closed (see tcp_shutdown) */ 
    pcb->tmr = tcp_ticks;     pcb->tmr = tcp_ticks;
  }   }
  pcb->keep_cnt_sent = 0;   pcb->keep_cnt_sent = 0;
  pcb->persist_probe = 0;   pcb->persist_probe = 0;
   
  tcp_parseopt(pcb);   tcp_parseopt(pcb);
   
  /* Do different things depending on the TCP state. */    /* Do different things depending on the TCP state. */ 
  switch (pcb->state) {   switch (pcb->state) {
    case SYN_SENT:     case SYN_SENT:
      LWIP_DEBUGF(TCP_INPUT_DEBUG, ("SYN-SENT: ackno %"U32_F" pcb->snd_nxt %"U32_F" unacked %s %"U32_F"\n",       LWIP_DEBUGF(TCP_INPUT_DEBUG, ("SYN-SENT: ackno %"U32_F" pcb->snd_nxt %"U32_F" unacked %s %"U32_F"\n",
                                    ackno, pcb->snd_nxt, pcb->unacked ? "" : " empty:",                                     ackno, pcb->snd_nxt, pcb->unacked ? "" : " empty:",
                                    pcb->unacked ? lwip_ntohl(pcb->unacked->tcphdr->seqno) : 0));                                     pcb->unacked ? lwip_ntohl(pcb->unacked->tcphdr->seqno) : 0));
      /* received SYN ACK with expected sequence number? */        /* received SYN ACK with expected sequence number? */ 
      if ((flags & TCP_ACK) && (flags & TCP_SYN)       if ((flags & TCP_ACK) && (flags & TCP_SYN)
          && (ackno == pcb->lastack + 1)) {           && (ackno == pcb->lastack + 1)) {
        pcb->rcv_nxt = seqno + 1;         pcb->rcv_nxt = seqno + 1;
        pcb->rcv_ann_right_edge = pcb->rcv_nxt;         pcb->rcv_ann_right_edge = pcb->rcv_nxt;
        pcb->lastack = ackno;         pcb->lastack = ackno;
        pcb->snd_wnd = tcphdr->wnd;         pcb->snd_wnd = tcphdr->wnd;
        pcb->snd_wnd_max = pcb->snd_wnd;         pcb->snd_wnd_max = pcb->snd_wnd;
        pcb->snd_wl1 = seqno - 1; /* initialise to seqno - 1 to force window update */          pcb->snd_wl1 = seqno - 1; /* initialise to seqno - 1 to force window update */ 
        pcb->state = ESTABLISHED;         pcb->state = ESTABLISHED;
   
#if TCP_CALCULATE_EFF_SEND_MSS #if TCP_CALCULATE_EFF_SEND_MSS
        pcb->mss = tcp_eff_send_mss(pcb->mss, &pcb->local_ip, &pcb->remote_ip);         pcb->mss = tcp_eff_send_mss(pcb->mss, &pcb->local_ip, &pcb->remote_ip);
#endif /* TCP_CALCULATE_EFF_SEND_MSS */  #endif /* TCP_CALCULATE_EFF_SEND_MSS */ 
   
        pcb->cwnd = LWIP_TCP_CALC_INITIAL_CWND(pcb->mss);         pcb->cwnd = LWIP_TCP_CALC_INITIAL_CWND(pcb->mss);
        LWIP_DEBUGF(TCP_CWND_DEBUG, ("tcp_process (SENT): cwnd %"TCPWNDSIZE_F         LWIP_DEBUGF(TCP_CWND_DEBUG, ("tcp_process (SENT): cwnd %"TCPWNDSIZE_F
                                     " ssthresh %"TCPWNDSIZE_F"\n",                                      " ssthresh %"TCPWNDSIZE_F"\n",
                                     pcb->cwnd, pcb->ssthresh));                                      pcb->cwnd, pcb->ssthresh));
        LWIP_ASSERT("pcb->snd_queuelen > 0", (pcb->snd_queuelen > 0));         LWIP_ASSERT("pcb->snd_queuelen > 0", (pcb->snd_queuelen > 0));
        --pcb->snd_queuelen;         --pcb->snd_queuelen;
        LWIP_DEBUGF(TCP_QLEN_DEBUG, ("tcp_process: SYN-SENT --queuelen %"TCPWNDSIZE_F"\n", (tcpwnd_size_t)pcb->snd_queuelen));         LWIP_DEBUGF(TCP_QLEN_DEBUG, ("tcp_process: SYN-SENT --queuelen %"TCPWNDSIZE_F"\n", (tcpwnd_size_t)pcb->snd_queuelen));
        rseg = pcb->unacked;         rseg = pcb->unacked;
        if (rseg == NULL) {         if (rseg == NULL) {
          /* might happen if tcp_output fails in tcp_rexmit_rto()           /* might happen if tcp_output fails in tcp_rexmit_rto()
             in which case the segment is on the unsent list */               in which case the segment is on the unsent list */ 
          rseg = pcb->unsent;           rseg = pcb->unsent;
          LWIP_ASSERT("no segment to free", rseg != NULL);           LWIP_ASSERT("no segment to free", rseg != NULL);
          pcb->unsent = rseg->next;           pcb->unsent = rseg->next;
        } else {         } else {
          pcb->unacked = rseg->next;           pcb->unacked = rseg->next;
        }         }
        tcp_seg_free(rseg);         tcp_seg_free(rseg);
   
        /* If there's nothing left to acknowledge, stop the retransmit         /* If there's nothing left to acknowledge, stop the retransmit
           timer, otherwise reset it to start again */             timer, otherwise reset it to start again */ 
        if (pcb->unacked == NULL) {         if (pcb->unacked == NULL) {
          pcb->rtime = -1;           pcb->rtime = -1;
        } else {         } else {
          pcb->rtime = 0;           pcb->rtime = 0;
          pcb->nrtx = 0;           pcb->nrtx = 0;
        }         }
   
        /* Call the user specified function to call when successfully         /* Call the user specified function to call when successfully
         * connected. */           * connected. */ 
        TCP_EVENT_CONNECTED(pcb, ERR_OK, err);         TCP_EVENT_CONNECTED(pcb, ERR_OK, err);
        if (err == ERR_ABRT) {         if (err == ERR_ABRT) {
          return ERR_ABRT;           return ERR_ABRT;
        }         }
        tcp_ack_now(pcb);         tcp_ack_now(pcb);
      }       }
      /* received ACK? possibly a half-open connection */        /* received ACK? possibly a half-open connection */ 
      else if (flags & TCP_ACK) {       else if (flags & TCP_ACK) {
        /* send a RST to bring the other side in a non-synchronized state. */          /* send a RST to bring the other side in a non-synchronized state. */ 
        tcp_rst(pcb, ackno, seqno + tcplen, ip_current_dest_addr(),         tcp_rst(pcb, ackno, seqno + tcplen, ip_current_dest_addr(),
                ip_current_src_addr(), tcphdr->dest, tcphdr->src);                 ip_current_src_addr(), tcphdr->dest, tcphdr->src);
        /* Resend SYN immediately (don't wait for rto timeout) to establish         /* Resend SYN immediately (don't wait for rto timeout) to establish
          connection faster, but do not send more SYNs than we otherwise would           connection faster, but do not send more SYNs than we otherwise would
          have, or we might get caught in a loop on loopback interfaces. */            have, or we might get caught in a loop on loopback interfaces. */ 
        if (pcb->nrtx < TCP_SYNMAXRTX) {         if (pcb->nrtx < TCP_SYNMAXRTX) {
          pcb->rtime = 0;           pcb->rtime = 0;
          tcp_rexmit_rto(pcb);           tcp_rexmit_rto(pcb);
        }         }
      }       }
      break;       break;
    case SYN_RCVD:     case SYN_RCVD:
      if (flags & TCP_ACK) {       if (flags & TCP_ACK) {
        /* expected ACK number? */          /* expected ACK number? */ 
        if (TCP_SEQ_BETWEEN(ackno, pcb->lastack + 1, pcb->snd_nxt)) {         if (TCP_SEQ_BETWEEN(ackno, pcb->lastack + 1, pcb->snd_nxt)) {
          pcb->state = ESTABLISHED;           pcb->state = ESTABLISHED;
          LWIP_DEBUGF(TCP_DEBUG, ("TCP connection established %"U16_F" -> %"U16_F".\n", inseg.tcphdr->src, inseg.tcphdr->dest));           LWIP_DEBUGF(TCP_DEBUG, ("TCP connection established %"U16_F" -> %"U16_F".\n", inseg.tcphdr->src, inseg.tcphdr->dest));
#if LWIP_CALLBACK_API || TCP_LISTEN_BACKLOG #if LWIP_CALLBACK_API || TCP_LISTEN_BACKLOG
          if (pcb->listener == NULL) {           if (pcb->listener == NULL) {
            /* listen pcb might be closed by now */              /* listen pcb might be closed by now */ 
            err = ERR_VAL;             err = ERR_VAL;
          } else           } else
#endif /* LWIP_CALLBACK_API || TCP_LISTEN_BACKLOG */  #endif /* LWIP_CALLBACK_API || TCP_LISTEN_BACKLOG */ 
          {           {
#if LWIP_CALLBACK_API #if LWIP_CALLBACK_API
            LWIP_ASSERT("pcb->listener->accept != NULL", pcb->listener->accept != NULL);             LWIP_ASSERT("pcb->listener->accept != NULL", pcb->listener->accept != NULL);
#endif #endif
            tcp_backlog_accepted(pcb);             tcp_backlog_accepted(pcb);
            /* Call the accept function. */              /* Call the accept function. */ 
            TCP_EVENT_ACCEPT(pcb->listener, pcb, pcb->callback_arg, ERR_OK, err);             TCP_EVENT_ACCEPT(pcb->listener, pcb, pcb->callback_arg, ERR_OK, err);
          }           }
          if (err != ERR_OK) {           if (err != ERR_OK) {
            /* If the accept function returns with an error, we abort             /* If the accept function returns with an error, we abort
             * the connection. */               * the connection. */ 
            /* Already aborted? */              /* Already aborted? */ 
            if (err != ERR_ABRT) {             if (err != ERR_ABRT) {
              tcp_abort(pcb);               tcp_abort(pcb);
            }             }
            return ERR_ABRT;             return ERR_ABRT;
          }           }
          /* If there was any data contained within this ACK,           /* If there was any data contained within this ACK,
           * we'd better pass it on to the application as well. */             * we'd better pass it on to the application as well. */ 
          tcp_receive(pcb);           tcp_receive(pcb);
   
          /* Prevent ACK for SYN to generate a sent event */            /* Prevent ACK for SYN to generate a sent event */ 
          if (recv_acked != 0) {           if (recv_acked != 0) {
            recv_acked--;             recv_acked--;
          }           }
   
          pcb->cwnd = LWIP_TCP_CALC_INITIAL_CWND(pcb->mss);           pcb->cwnd = LWIP_TCP_CALC_INITIAL_CWND(pcb->mss);
          LWIP_DEBUGF(TCP_CWND_DEBUG, ("tcp_process (SYN_RCVD): cwnd %"TCPWNDSIZE_F           LWIP_DEBUGF(TCP_CWND_DEBUG, ("tcp_process (SYN_RCVD): cwnd %"TCPWNDSIZE_F
                                       " ssthresh %"TCPWNDSIZE_F"\n",                                        " ssthresh %"TCPWNDSIZE_F"\n",
                                       pcb->cwnd, pcb->ssthresh));                                        pcb->cwnd, pcb->ssthresh));
   
          if (recv_flags & TF_GOT_FIN) {           if (recv_flags & TF_GOT_FIN) {
            tcp_ack_now(pcb);             tcp_ack_now(pcb);
            pcb->state = CLOSE_WAIT;             pcb->state = CLOSE_WAIT;
          }           }
        } else {         } else {
          /* incorrect ACK number, send RST */            /* incorrect ACK number, send RST */ 
          tcp_rst(pcb, ackno, seqno + tcplen, ip_current_dest_addr(),           tcp_rst(pcb, ackno, seqno + tcplen, ip_current_dest_addr(),
                  ip_current_src_addr(), tcphdr->dest, tcphdr->src);                   ip_current_src_addr(), tcphdr->dest, tcphdr->src);
        }         }
      } else if ((flags & TCP_SYN) && (seqno == pcb->rcv_nxt - 1)) {       } else if ((flags & TCP_SYN) && (seqno == pcb->rcv_nxt - 1)) {
        /* Looks like another copy of the SYN - retransmit our SYN-ACK */          /* Looks like another copy of the SYN - retransmit our SYN-ACK */ 
        tcp_rexmit(pcb);         tcp_rexmit(pcb);
      }       }
      break;       break;
    case CLOSE_WAIT:     case CLOSE_WAIT:
    /* FALLTHROUGH */      /* FALLTHROUGH */ 
    case ESTABLISHED:     case ESTABLISHED:
      tcp_receive(pcb);       tcp_receive(pcb);
      if (recv_flags & TF_GOT_FIN) { /* passive close */        if (recv_flags & TF_GOT_FIN) { /* passive close */ 
        tcp_ack_now(pcb);         tcp_ack_now(pcb);
        pcb->state = CLOSE_WAIT;         pcb->state = CLOSE_WAIT;
      }       }
      break;       break;
    case FIN_WAIT_1:     case FIN_WAIT_1:
      tcp_receive(pcb);       tcp_receive(pcb);
      if (recv_flags & TF_GOT_FIN) {       if (recv_flags & TF_GOT_FIN) {
        if ((flags & TCP_ACK) && (ackno == pcb->snd_nxt) &&         if ((flags & TCP_ACK) && (ackno == pcb->snd_nxt) &&
            pcb->unsent == NULL) {             pcb->unsent == NULL) {
          LWIP_DEBUGF(TCP_DEBUG,           LWIP_DEBUGF(TCP_DEBUG,
                      ("TCP connection closed: FIN_WAIT_1 %"U16_F" -> %"U16_F".\n", inseg.tcphdr->src, inseg.tcphdr->dest));                       ("TCP connection closed: FIN_WAIT_1 %"U16_F" -> %"U16_F".\n", inseg.tcphdr->src, inseg.tcphdr->dest));
          tcp_ack_now(pcb);           tcp_ack_now(pcb);
          tcp_pcb_purge(pcb);           tcp_pcb_purge(pcb);
          TCP_RMV_ACTIVE(pcb);           TCP_RMV_ACTIVE(pcb);
          pcb->state = TIME_WAIT;           pcb->state = TIME_WAIT;
          TCP_REG(&tcp_tw_pcbs, pcb);           TCP_REG(&tcp_tw_pcbs, pcb);
        } else {         } else {
          tcp_ack_now(pcb);           tcp_ack_now(pcb);
          pcb->state = CLOSING;           pcb->state = CLOSING;
        }         }
      } else if ((flags & TCP_ACK) && (ackno == pcb->snd_nxt) &&       } else if ((flags & TCP_ACK) && (ackno == pcb->snd_nxt) &&
                 pcb->unsent == NULL) {                  pcb->unsent == NULL) {
        pcb->state = FIN_WAIT_2;         pcb->state = FIN_WAIT_2;
      }       }
      break;       break;
    case FIN_WAIT_2:     case FIN_WAIT_2:
      tcp_receive(pcb);       tcp_receive(pcb);
      if (recv_flags & TF_GOT_FIN) {       if (recv_flags & TF_GOT_FIN) {
        LWIP_DEBUGF(TCP_DEBUG, ("TCP connection closed: FIN_WAIT_2 %"U16_F" -> %"U16_F".\n", inseg.tcphdr->src, inseg.tcphdr->dest));         LWIP_DEBUGF(TCP_DEBUG, ("TCP connection closed: FIN_WAIT_2 %"U16_F" -> %"U16_F".\n", inseg.tcphdr->src, inseg.tcphdr->dest));
        tcp_ack_now(pcb);         tcp_ack_now(pcb);
        tcp_pcb_purge(pcb);         tcp_pcb_purge(pcb);
        TCP_RMV_ACTIVE(pcb);         TCP_RMV_ACTIVE(pcb);
        pcb->state = TIME_WAIT;         pcb->state = TIME_WAIT;
        TCP_REG(&tcp_tw_pcbs, pcb);         TCP_REG(&tcp_tw_pcbs, pcb);
      }       }
      break;       break;
    case CLOSING:     case CLOSING:
      tcp_receive(pcb);       tcp_receive(pcb);
      if ((flags & TCP_ACK) && ackno == pcb->snd_nxt && pcb->unsent == NULL) {       if ((flags & TCP_ACK) && ackno == pcb->snd_nxt && pcb->unsent == NULL) {
        LWIP_DEBUGF(TCP_DEBUG, ("TCP connection closed: CLOSING %"U16_F" -> %"U16_F".\n", inseg.tcphdr->src, inseg.tcphdr->dest));         LWIP_DEBUGF(TCP_DEBUG, ("TCP connection closed: CLOSING %"U16_F" -> %"U16_F".\n", inseg.tcphdr->src, inseg.tcphdr->dest));
        tcp_pcb_purge(pcb);         tcp_pcb_purge(pcb);
        TCP_RMV_ACTIVE(pcb);         TCP_RMV_ACTIVE(pcb);
        pcb->state = TIME_WAIT;         pcb->state = TIME_WAIT;
        TCP_REG(&tcp_tw_pcbs, pcb);         TCP_REG(&tcp_tw_pcbs, pcb);
      }       }
      break;       break;
    case LAST_ACK:     case LAST_ACK:
      tcp_receive(pcb);       tcp_receive(pcb);
      if ((flags & TCP_ACK) && ackno == pcb->snd_nxt && pcb->unsent == NULL) {       if ((flags & TCP_ACK) && ackno == pcb->snd_nxt && pcb->unsent == NULL) {
        LWIP_DEBUGF(TCP_DEBUG, ("TCP connection closed: LAST_ACK %"U16_F" -> %"U16_F".\n", inseg.tcphdr->src, inseg.tcphdr->dest));         LWIP_DEBUGF(TCP_DEBUG, ("TCP connection closed: LAST_ACK %"U16_F" -> %"U16_F".\n", inseg.tcphdr->src, inseg.tcphdr->dest));
        /* bugfix #21699: don't set pcb->state to CLOSED here or we risk leaking segments */          /* bugfix #21699: don't set pcb->state to CLOSED here or we risk leaking segments */ 
        recv_flags |= TF_CLOSED;         recv_flags |= TF_CLOSED;
      }       }
      break;       break;
    default:     default:
      break;       break;
  }   }
  return ERR_OK;   return ERR_OK;
} }
   
#if TCP_QUEUE_OOSEQ #if TCP_QUEUE_OOSEQ
/** /**
 * Insert segment into the list (segments covered with new one will be deleted)  * Insert segment into the list (segments covered with new one will be deleted)
 *  *
 * Called from tcp_receive()  * Called from tcp_receive()
 */   */ 
static void static void
tcp_oos_insert_segment(struct tcp_seg *cseg, struct tcp_seg *next) tcp_oos_insert_segment(struct tcp_seg *cseg, struct tcp_seg *next)
{ {
  struct tcp_seg *old_seg;   struct tcp_seg *old_seg;
   
  LWIP_ASSERT("tcp_oos_insert_segment: invalid cseg", cseg != NULL);   LWIP_ASSERT("tcp_oos_insert_segment: invalid cseg", cseg != NULL);
   
  if (TCPH_FLAGS(cseg->tcphdr) & TCP_FIN) {   if (TCPH_FLAGS(cseg->tcphdr) & TCP_FIN) {
    /* received segment overlaps all following segments */      /* received segment overlaps all following segments */ 
    tcp_segs_free(next);     tcp_segs_free(next);
    next = NULL;     next = NULL;
  } else {   } else {
    /* delete some following segments     /* delete some following segments
       oos queue may have segments with FIN flag */         oos queue may have segments with FIN flag */ 
    while (next &&     while (next &&
           TCP_SEQ_GEQ((seqno + cseg->len),            TCP_SEQ_GEQ((seqno + cseg->len),
                       (next->tcphdr->seqno + next->len))) {                        (next->tcphdr->seqno + next->len))) {
      /* cseg with FIN already processed */        /* cseg with FIN already processed */ 
      if (TCPH_FLAGS(next->tcphdr) & TCP_FIN) {       if (TCPH_FLAGS(next->tcphdr) & TCP_FIN) {
        TCPH_SET_FLAG(cseg->tcphdr, TCP_FIN);         TCPH_SET_FLAG(cseg->tcphdr, TCP_FIN);
      }       }
      old_seg = next;       old_seg = next;
      next = next->next;       next = next->next;
      tcp_seg_free(old_seg);       tcp_seg_free(old_seg);
    }     }
    if (next &&     if (next &&
        TCP_SEQ_GT(seqno + cseg->len, next->tcphdr->seqno)) {         TCP_SEQ_GT(seqno + cseg->len, next->tcphdr->seqno)) {
      /* We need to trim the incoming segment. */        /* We need to trim the incoming segment. */ 
      cseg->len = (u16_t)(next->tcphdr->seqno - seqno);       cseg->len = (u16_t)(next->tcphdr->seqno - seqno);
      pbuf_realloc(cseg->p, cseg->len);       pbuf_realloc(cseg->p, cseg->len);
    }     }
  }   }
  cseg->next = next;   cseg->next = next;
} }
#endif /* TCP_QUEUE_OOSEQ */  #endif /* TCP_QUEUE_OOSEQ */ 
   
/** Remove segments from a list if the incoming ACK acknowledges them */  /** Remove segments from a list if the incoming ACK acknowledges them */ 
static struct tcp_seg * static struct tcp_seg *
tcp_free_acked_segments(struct tcp_pcb *pcb, struct tcp_seg *seg_list, const char *dbg_list_name, tcp_free_acked_segments(struct tcp_pcb *pcb, struct tcp_seg *seg_list, const char *dbg_list_name,
                        struct tcp_seg *dbg_other_seg_list)                         struct tcp_seg *dbg_other_seg_list)
{ {
  struct tcp_seg *next;   struct tcp_seg *next;
  u16_t clen;   u16_t clen;
   
  LWIP_UNUSED_ARG(dbg_list_name);   LWIP_UNUSED_ARG(dbg_list_name);
  LWIP_UNUSED_ARG(dbg_other_seg_list);   LWIP_UNUSED_ARG(dbg_other_seg_list);
   
  while (seg_list != NULL &&   while (seg_list != NULL &&
         TCP_SEQ_LEQ(lwip_ntohl(seg_list->tcphdr->seqno) +          TCP_SEQ_LEQ(lwip_ntohl(seg_list->tcphdr->seqno) +
                     TCP_TCPLEN(seg_list), ackno)) {                      TCP_TCPLEN(seg_list), ackno)) {
    LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_receive: removing %"U32_F":%"U32_F" from pcb->%s\n",     LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_receive: removing %"U32_F":%"U32_F" from pcb->%s\n",
                                  lwip_ntohl(seg_list->tcphdr->seqno),                                   lwip_ntohl(seg_list->tcphdr->seqno),
                                  lwip_ntohl(seg_list->tcphdr->seqno) + TCP_TCPLEN(seg_list),                                   lwip_ntohl(seg_list->tcphdr->seqno) + TCP_TCPLEN(seg_list),
                                  dbg_list_name));                                   dbg_list_name));
   
    next = seg_list;     next = seg_list;
    seg_list = seg_list->next;     seg_list = seg_list->next;
   
    clen = pbuf_clen(next->p);     clen = pbuf_clen(next->p);
    LWIP_DEBUGF(TCP_QLEN_DEBUG, ("tcp_receive: queuelen %"TCPWNDSIZE_F" ... ",     LWIP_DEBUGF(TCP_QLEN_DEBUG, ("tcp_receive: queuelen %"TCPWNDSIZE_F" ... ",
                                 (tcpwnd_size_t)pcb->snd_queuelen));                                  (tcpwnd_size_t)pcb->snd_queuelen));
    LWIP_ASSERT("pcb->snd_queuelen >= pbuf_clen(next->p)", (pcb->snd_queuelen >= clen));     LWIP_ASSERT("pcb->snd_queuelen >= pbuf_clen(next->p)", (pcb->snd_queuelen >= clen));
   
    pcb->snd_queuelen = (u16_t)(pcb->snd_queuelen - clen);     pcb->snd_queuelen = (u16_t)(pcb->snd_queuelen - clen);
    recv_acked = (tcpwnd_size_t)(recv_acked + next->len);     recv_acked = (tcpwnd_size_t)(recv_acked + next->len);
    tcp_seg_free(next);     tcp_seg_free(next);
   
    LWIP_DEBUGF(TCP_QLEN_DEBUG, ("%"TCPWNDSIZE_F" (after freeing %s)\n",     LWIP_DEBUGF(TCP_QLEN_DEBUG, ("%"TCPWNDSIZE_F" (after freeing %s)\n",
                                 (tcpwnd_size_t)pcb->snd_queuelen,                                  (tcpwnd_size_t)pcb->snd_queuelen,
                                 dbg_list_name));                                  dbg_list_name));
    if (pcb->snd_queuelen != 0) {     if (pcb->snd_queuelen != 0) {
      LWIP_ASSERT("tcp_receive: valid queue length",       LWIP_ASSERT("tcp_receive: valid queue length",
                  seg_list != NULL || dbg_other_seg_list != NULL);                   seg_list != NULL || dbg_other_seg_list != NULL);
    }     }
  }   }
  return seg_list;   return seg_list;
} }
   
/** /**
 * Called by tcp_process. Checks if the given segment is an ACK for outstanding  * 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, it places the  * data, and if so frees the memory of the buffered data. Next, it places the
 * segment on any of the receive queues (pcb->recved or pcb->ooseq). If the segment  * 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  * is buffered, the pbuf is referenced by pbuf_ref so that it will not be freed until
 * it has been removed from the buffer.  * it has been removed from the buffer.
 *  *
 * If the incoming segment constitutes an ACK for a segment that was used for RTT  * If the incoming segment constitutes an ACK for a segment that was used for RTT
 * estimation, the RTT is estimated here as well.  * estimation, the RTT is estimated here as well.
 *  *
 * Called from tcp_process().  * Called from tcp_process().
 */   */ 
static void static void
tcp_receive(struct tcp_pcb *pcb) tcp_receive(struct tcp_pcb *pcb)
{ {
  s16_t m;   s16_t m;
  u32_t right_wnd_edge;   u32_t right_wnd_edge;
  int found_dupack = 0;   int found_dupack = 0;
   
  LWIP_ASSERT("tcp_receive: invalid pcb", pcb != NULL);   LWIP_ASSERT("tcp_receive: invalid pcb", pcb != NULL);
  LWIP_ASSERT("tcp_receive: wrong state", pcb->state >= ESTABLISHED);   LWIP_ASSERT("tcp_receive: wrong state", pcb->state >= ESTABLISHED);
   
  if (flags & TCP_ACK) {   if (flags & TCP_ACK) {
    right_wnd_edge = pcb->snd_wnd + pcb->snd_wl2;     right_wnd_edge = pcb->snd_wnd + pcb->snd_wl2;
   
    /* Update window. */      /* Update window. */ 
    if (TCP_SEQ_LT(pcb->snd_wl1, seqno) ||     if (TCP_SEQ_LT(pcb->snd_wl1, seqno) ||
        (pcb->snd_wl1 == seqno && TCP_SEQ_LT(pcb->snd_wl2, ackno)) ||         (pcb->snd_wl1 == seqno && TCP_SEQ_LT(pcb->snd_wl2, ackno)) ||
        (pcb->snd_wl2 == ackno && (u32_t)SND_WND_SCALE(pcb, tcphdr->wnd) > pcb->snd_wnd)) {         (pcb->snd_wl2 == ackno && (u32_t)SND_WND_SCALE(pcb, tcphdr->wnd) > pcb->snd_wnd)) {
      pcb->snd_wnd = SND_WND_SCALE(pcb, tcphdr->wnd);       pcb->snd_wnd = SND_WND_SCALE(pcb, tcphdr->wnd);
      /* keep track of the biggest window announced by the remote host to calculate       /* keep track of the biggest window announced by the remote host to calculate
         the maximum segment size */           the maximum segment size */ 
      if (pcb->snd_wnd_max < pcb->snd_wnd) {       if (pcb->snd_wnd_max < pcb->snd_wnd) {
        pcb->snd_wnd_max = pcb->snd_wnd;         pcb->snd_wnd_max = pcb->snd_wnd;
      }       }
      pcb->snd_wl1 = seqno;       pcb->snd_wl1 = seqno;
      pcb->snd_wl2 = ackno;       pcb->snd_wl2 = ackno;
      LWIP_DEBUGF(TCP_WND_DEBUG, ("tcp_receive: window update %"TCPWNDSIZE_F"\n", pcb->snd_wnd));       LWIP_DEBUGF(TCP_WND_DEBUG, ("tcp_receive: window update %"TCPWNDSIZE_F"\n", pcb->snd_wnd));
#if TCP_WND_DEBUG #if TCP_WND_DEBUG
    } else {     } else {
      if (pcb->snd_wnd != (tcpwnd_size_t)SND_WND_SCALE(pcb, tcphdr->wnd)) {       if (pcb->snd_wnd != (tcpwnd_size_t)SND_WND_SCALE(pcb, tcphdr->wnd)) {
        LWIP_DEBUGF(TCP_WND_DEBUG,         LWIP_DEBUGF(TCP_WND_DEBUG,
                    ("tcp_receive: no window update lastack %"U32_F" ackno %"                      ("tcp_receive: no window update lastack %"U32_F" ackno %" 
                     U32_F" wl1 %"U32_F" seqno %"U32_F" wl2 %"U32_F"\n",                      U32_F" wl1 %"U32_F" seqno %"U32_F" wl2 %"U32_F"\n",
                     pcb->lastack, ackno, pcb->snd_wl1, seqno, pcb->snd_wl2));                      pcb->lastack, ackno, pcb->snd_wl1, seqno, pcb->snd_wl2));
      }       }
#endif /* TCP_WND_DEBUG */  #endif /* TCP_WND_DEBUG */ 
    }     }
   
    /* (From Stevens TCP/IP Illustrated Vol II, p970.) Its only a     /* (From Stevens TCP/IP Illustrated Vol II, p970.) Its only a
     * duplicate ack if:      * duplicate ack if:
     * 1) It doesn't ACK new data      * 1) It doesn't ACK new data
     * 2) length of received packet is zero (i.e. no payload)      * 2) length of received packet is zero (i.e. no payload)
     * 3) the advertised window hasn't changed      * 3) the advertised window hasn't changed
     * 4) There is outstanding unacknowledged data (retransmission timer running)      * 4) There is outstanding unacknowledged data (retransmission timer running)
     * 5) The ACK is == biggest ACK sequence number so far seen (snd_una)      * 5) The ACK is == biggest ACK sequence number so far seen (snd_una)
     *      *
     * If it passes all five, should process as a dupack:      * If it passes all five, should process as a dupack:
     * a) dupacks < 3: do nothing      * a) dupacks < 3: do nothing
     * b) dupacks == 3: fast retransmit      * b) dupacks == 3: fast retransmit
     * c) dupacks > 3: increase cwnd      * c) dupacks > 3: increase cwnd
     *      *
     * If it only passes 1-3, should reset dupack counter (and add to      * If it only passes 1-3, should reset dupack counter (and add to
     * stats, which we don't do in lwIP)      * stats, which we don't do in lwIP)
     *      *
     * If it only passes 1, should reset dupack counter      * If it only passes 1, should reset dupack counter
     *      *
     */       */ 
   
    /* Clause 1 */      /* Clause 1 */ 
    if (TCP_SEQ_LEQ(ackno, pcb->lastack)) {     if (TCP_SEQ_LEQ(ackno, pcb->lastack)) {
      /* Clause 2 */        /* Clause 2 */ 
      if (tcplen == 0) {       if (tcplen == 0) {
        /* Clause 3 */          /* Clause 3 */ 
        if (pcb->snd_wl2 + pcb->snd_wnd == right_wnd_edge) {         if (pcb->snd_wl2 + pcb->snd_wnd == right_wnd_edge) {
          /* Clause 4 */            /* Clause 4 */ 
          if (pcb->rtime >= 0) {           if (pcb->rtime >= 0) {
            /* Clause 5 */              /* Clause 5 */ 
            if (pcb->lastack == ackno) {             if (pcb->lastack == ackno) {
              found_dupack = 1;               found_dupack = 1;
              if ((u8_t)(pcb->dupacks + 1) > pcb->dupacks) {               if ((u8_t)(pcb->dupacks + 1) > pcb->dupacks) {
                ++pcb->dupacks;                 ++pcb->dupacks;
              }               }
              if (pcb->dupacks > 3) {               if (pcb->dupacks > 3) {
                /* Inflate the congestion window */                  /* Inflate the congestion window */ 
                TCP_WND_INC(pcb->cwnd, pcb->mss);                 TCP_WND_INC(pcb->cwnd, pcb->mss);
              }               }
              if (pcb->dupacks >= 3) {               if (pcb->dupacks >= 3) {
                /* Do fast retransmit (checked via TF_INFR, not via dupacks count) */                  /* Do fast retransmit (checked via TF_INFR, not via dupacks count) */ 
                tcp_rexmit_fast(pcb);                 tcp_rexmit_fast(pcb);
              }               }
            }             }
          }           }
        }         }
      }       }
      /* If Clause (1) or more is true, but not a duplicate ack, reset       /* If Clause (1) or more is true, but not a duplicate ack, reset
       * count of consecutive duplicate acks */         * count of consecutive duplicate acks */ 
      if (!found_dupack) {       if (!found_dupack) {
        pcb->dupacks = 0;         pcb->dupacks = 0;
      }       }
    } else if (TCP_SEQ_BETWEEN(ackno, pcb->lastack + 1, pcb->snd_nxt)) {     } else if (TCP_SEQ_BETWEEN(ackno, pcb->lastack + 1, pcb->snd_nxt)) {
      /* We come here when the ACK acknowledges new data. */        /* We come here when the ACK acknowledges new data. */ 
      tcpwnd_size_t acked;       tcpwnd_size_t acked;
   
      /* Reset the "IN Fast Retransmit" flag, since we are no longer       /* Reset the "IN Fast Retransmit" flag, since we are no longer
         in fast retransmit. Also reset the congestion window to the          in fast retransmit. Also reset the congestion window to the
         slow start threshold. */           slow start threshold. */ 
      if (pcb->flags & TF_INFR) {       if (pcb->flags & TF_INFR) {
        tcp_clear_flags(pcb, TF_INFR);         tcp_clear_flags(pcb, TF_INFR);
        pcb->cwnd = pcb->ssthresh;         pcb->cwnd = pcb->ssthresh;
        pcb->bytes_acked = 0;         pcb->bytes_acked = 0;
      }       }
   
      /* Reset the number of retransmissions. */        /* Reset the number of retransmissions. */ 
      pcb->nrtx = 0;       pcb->nrtx = 0;
   
      /* Reset the retransmission time-out. */        /* Reset the retransmission time-out. */ 
      pcb->rto = (s16_t)((pcb->sa >> 3) + pcb->sv);       pcb->rto = (s16_t)((pcb->sa >> 3) + pcb->sv);
   
      /* Record how much data this ACK acks */        /* Record how much data this ACK acks */ 
      acked = (tcpwnd_size_t)(ackno - pcb->lastack);       acked = (tcpwnd_size_t)(ackno - pcb->lastack);
   
      /* Reset the fast retransmit variables. */        /* Reset the fast retransmit variables. */ 
      pcb->dupacks = 0;       pcb->dupacks = 0;
      pcb->lastack = ackno;       pcb->lastack = ackno;
   
      /* Update the congestion control variables (cwnd and       /* Update the congestion control variables (cwnd and
         ssthresh). */           ssthresh). */ 
      if (pcb->state >= ESTABLISHED) {       if (pcb->state >= ESTABLISHED) {
        if (pcb->cwnd < pcb->ssthresh) {         if (pcb->cwnd < pcb->ssthresh) {
          tcpwnd_size_t increase;           tcpwnd_size_t increase;
          /* limit to 1 SMSS segment during period following RTO */            /* limit to 1 SMSS segment during period following RTO */ 
          u8_t num_seg = (pcb->flags & TF_RTO) ? 1 : 2;           u8_t num_seg = (pcb->flags & TF_RTO) ? 1 : 2;
          /* RFC 3465, section 2.2 Slow Start */            /* RFC 3465, section 2.2 Slow Start */ 
          increase = LWIP_MIN(acked, (tcpwnd_size_t)(num_seg * pcb->mss));           increase = LWIP_MIN(acked, (tcpwnd_size_t)(num_seg * pcb->mss));
          TCP_WND_INC(pcb->cwnd, increase);           TCP_WND_INC(pcb->cwnd, increase);
          LWIP_DEBUGF(TCP_CWND_DEBUG, ("tcp_receive: slow start cwnd %"TCPWNDSIZE_F"\n", pcb->cwnd));           LWIP_DEBUGF(TCP_CWND_DEBUG, ("tcp_receive: slow start cwnd %"TCPWNDSIZE_F"\n", pcb->cwnd));
        } else {         } else {
          /* RFC 3465, section 2.1 Congestion Avoidance */            /* RFC 3465, section 2.1 Congestion Avoidance */ 
          TCP_WND_INC(pcb->bytes_acked, acked);           TCP_WND_INC(pcb->bytes_acked, acked);
          if (pcb->bytes_acked >= pcb->cwnd) {           if (pcb->bytes_acked >= pcb->cwnd) {
            pcb->bytes_acked = (tcpwnd_size_t)(pcb->bytes_acked - pcb->cwnd);             pcb->bytes_acked = (tcpwnd_size_t)(pcb->bytes_acked - pcb->cwnd);
            TCP_WND_INC(pcb->cwnd, pcb->mss);             TCP_WND_INC(pcb->cwnd, pcb->mss);
          }           }
          LWIP_DEBUGF(TCP_CWND_DEBUG, ("tcp_receive: congestion avoidance cwnd %"TCPWNDSIZE_F"\n", pcb->cwnd));           LWIP_DEBUGF(TCP_CWND_DEBUG, ("tcp_receive: congestion avoidance cwnd %"TCPWNDSIZE_F"\n", pcb->cwnd));
        }         }
      }       }
      LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_receive: ACK for %"U32_F", unacked->seqno %"U32_F":%"U32_F"\n",       LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_receive: ACK for %"U32_F", unacked->seqno %"U32_F":%"U32_F"\n",
                                    ackno,                                     ackno,
                                    pcb->unacked != NULL ?                                     pcb->unacked != NULL ?
                                    lwip_ntohl(pcb->unacked->tcphdr->seqno) : 0,                                     lwip_ntohl(pcb->unacked->tcphdr->seqno) : 0,
                                    pcb->unacked != NULL ?                                     pcb->unacked != NULL ?
                                    lwip_ntohl(pcb->unacked->tcphdr->seqno) + TCP_TCPLEN(pcb->unacked) : 0));                                     lwip_ntohl(pcb->unacked->tcphdr->seqno) + TCP_TCPLEN(pcb->unacked) : 0));
   
      /* Remove segment from the unacknowledged list if the incoming       /* Remove segment from the unacknowledged list if the incoming
         ACK acknowledges them. */           ACK acknowledges them. */ 
      pcb->unacked = tcp_free_acked_segments(pcb, pcb->unacked, "unacked", pcb->unsent);       pcb->unacked = tcp_free_acked_segments(pcb, pcb->unacked, "unacked", pcb->unsent);
      /* We go through the ->unsent list to see if any of the segments       /* We go through the ->unsent list to see if any of the segments
         on the list are acknowledged by the ACK. This may seem          on the list are acknowledged by the ACK. This may seem
         strange since an "unsent" segment shouldn't be acked. The          strange since an "unsent" segment shouldn't be acked. The
         rationale is that lwIP puts all outstanding segments on the          rationale is that lwIP puts all outstanding segments on the
         ->unsent list after a retransmission, so these segments may          ->unsent list after a retransmission, so these segments may
         in fact have been sent once. */           in fact have been sent once. */ 
      pcb->unsent = tcp_free_acked_segments(pcb, pcb->unsent, "unsent", pcb->unacked);       pcb->unsent = tcp_free_acked_segments(pcb, pcb->unsent, "unsent", pcb->unacked);
   
      /* If there's nothing left to acknowledge, stop the retransmit       /* If there's nothing left to acknowledge, stop the retransmit
         timer, otherwise reset it to start again */           timer, otherwise reset it to start again */ 
      if (pcb->unacked == NULL) {       if (pcb->unacked == NULL) {
        pcb->rtime = -1;         pcb->rtime = -1;
      } else {       } else {
        pcb->rtime = 0;         pcb->rtime = 0;
      }       }
   
      pcb->polltmr = 0;       pcb->polltmr = 0;
   
#if TCP_OVERSIZE #if TCP_OVERSIZE
      if (pcb->unsent == NULL) {       if (pcb->unsent == NULL) {
        pcb->unsent_oversize = 0;         pcb->unsent_oversize = 0;
      }       }
#endif /* TCP_OVERSIZE */  #endif /* TCP_OVERSIZE */ 
   
#if LWIP_IPV6 && LWIP_ND6_TCP_REACHABILITY_HINTS #if LWIP_IPV6 && LWIP_ND6_TCP_REACHABILITY_HINTS
      if (ip_current_is_v6()) {       if (ip_current_is_v6()) {
        /* Inform neighbor reachability of forward progress. */          /* Inform neighbor reachability of forward progress. */ 
        nd6_reachability_hint(ip6_current_src_addr());         nd6_reachability_hint(ip6_current_src_addr());
      }       }
#endif /* LWIP_IPV6 && LWIP_ND6_TCP_REACHABILITY_HINTS*/  #endif /* LWIP_IPV6 && LWIP_ND6_TCP_REACHABILITY_HINTS*/ 
   
      pcb->snd_buf = (tcpwnd_size_t)(pcb->snd_buf + recv_acked);       pcb->snd_buf = (tcpwnd_size_t)(pcb->snd_buf + recv_acked);
      /* check if this ACK ends our retransmission of in-flight data */        /* check if this ACK ends our retransmission of in-flight data */ 
      if (pcb->flags & TF_RTO) {       if (pcb->flags & TF_RTO) {
        /* RTO is done if         /* RTO is done if
            1) both queues are empty or             1) both queues are empty or
            2) unacked is empty and unsent head contains data not part of RTO or             2) unacked is empty and unsent head contains data not part of RTO or
            3) unacked head contains data not part of RTO */              3) unacked head contains data not part of RTO */ 
        if (pcb->unacked == NULL) {         if (pcb->unacked == NULL) {
          if ((pcb->unsent == NULL) ||           if ((pcb->unsent == NULL) ||
              (TCP_SEQ_LEQ(pcb->rto_end, lwip_ntohl(pcb->unsent->tcphdr->seqno)))) {               (TCP_SEQ_LEQ(pcb->rto_end, lwip_ntohl(pcb->unsent->tcphdr->seqno)))) {
            tcp_clear_flags(pcb, TF_RTO);             tcp_clear_flags(pcb, TF_RTO);
          }           }
        } else if (TCP_SEQ_LEQ(pcb->rto_end, lwip_ntohl(pcb->unacked->tcphdr->seqno))) {         } else if (TCP_SEQ_LEQ(pcb->rto_end, lwip_ntohl(pcb->unacked->tcphdr->seqno))) {
          tcp_clear_flags(pcb, TF_RTO);           tcp_clear_flags(pcb, TF_RTO);
        }         }
      }       }
      /* End of ACK for new data processing. */        /* End of ACK for new data processing. */ 
    } else {     } else {
      /* Out of sequence ACK, didn't really ack anything */        /* Out of sequence ACK, didn't really ack anything */ 
      tcp_send_empty_ack(pcb);       tcp_send_empty_ack(pcb);
    }     }
   
    LWIP_DEBUGF(TCP_RTO_DEBUG, ("tcp_receive: pcb->rttest %"U32_F" rtseq %"U32_F" ackno %"U32_F"\n",     LWIP_DEBUGF(TCP_RTO_DEBUG, ("tcp_receive: pcb->rttest %"U32_F" rtseq %"U32_F" ackno %"U32_F"\n",
                                pcb->rttest, pcb->rtseq, ackno));                                 pcb->rttest, pcb->rtseq, ackno));
   
    /* RTT estimation calculations. This is done by checking if the     /* RTT estimation calculations. This is done by checking if the
       incoming segment acknowledges the segment we use to take a        incoming segment acknowledges the segment we use to take a
       round-trip time measurement. */         round-trip time measurement. */ 
    if (pcb->rttest && TCP_SEQ_LT(pcb->rtseq, ackno)) {     if (pcb->rttest && TCP_SEQ_LT(pcb->rtseq, ackno)) {
      /* diff between this shouldn't exceed 32K since this are tcp timer ticks       /* diff between this shouldn't exceed 32K since this are tcp timer ticks
         and a round-trip shouldn't be that long... */           and a round-trip shouldn't be that long... */ 
      m = (s16_t)(tcp_ticks - pcb->rttest);       m = (s16_t)(tcp_ticks - pcb->rttest);
   
      LWIP_DEBUGF(TCP_RTO_DEBUG, ("tcp_receive: experienced rtt %"U16_F" ticks (%"U16_F" msec).\n",       LWIP_DEBUGF(TCP_RTO_DEBUG, ("tcp_receive: experienced rtt %"U16_F" ticks (%"U16_F" msec).\n",
                                  m, (u16_t)(m * TCP_SLOW_INTERVAL)));                                   m, (u16_t)(m * TCP_SLOW_INTERVAL)));
   
      /* This is taken directly from VJs original code in his paper */        /* This is taken directly from VJs original code in his paper */ 
      m = (s16_t)(m - (pcb->sa >> 3));       m = (s16_t)(m - (pcb->sa >> 3));
      pcb->sa = (s16_t)(pcb->sa + m);       pcb->sa = (s16_t)(pcb->sa + m);
      if (m < 0) {       if (m < 0) {
        m = (s16_t) - m;         m = (s16_t) - m;
      }       }
      m = (s16_t)(m - (pcb->sv >> 2));       m = (s16_t)(m - (pcb->sv >> 2));
      pcb->sv = (s16_t)(pcb->sv + m);       pcb->sv = (s16_t)(pcb->sv + m);
      pcb->rto = (s16_t)((pcb->sa >> 3) + pcb->sv);       pcb->rto = (s16_t)((pcb->sa >> 3) + pcb->sv);
   
      LWIP_DEBUGF(TCP_RTO_DEBUG, ("tcp_receive: RTO %"U16_F" (%"U16_F" milliseconds)\n",       LWIP_DEBUGF(TCP_RTO_DEBUG, ("tcp_receive: RTO %"U16_F" (%"U16_F" milliseconds)\n",
                                  pcb->rto, (u16_t)(pcb->rto * TCP_SLOW_INTERVAL)));                                   pcb->rto, (u16_t)(pcb->rto * TCP_SLOW_INTERVAL)));
   
      pcb->rttest = 0;       pcb->rttest = 0;
    }     }
  }   }
   
  /* If the incoming segment contains data, we must process it   /* If the incoming segment contains data, we must process it
     further unless the pcb already received a FIN.      further unless the pcb already received a FIN.
     (RFC 793, chapter 3.9, "SEGMENT ARRIVES" in states CLOSE-WAIT, CLOSING,      (RFC 793, chapter 3.9, "SEGMENT ARRIVES" in states CLOSE-WAIT, CLOSING,
     LAST-ACK and TIME-WAIT: "Ignore the segment text.") */       LAST-ACK and TIME-WAIT: "Ignore the segment text.") */ 
  if ((tcplen > 0) && (pcb->state < CLOSE_WAIT)) {   if ((tcplen > 0) && (pcb->state < CLOSE_WAIT)) {
    /* This code basically does three things:     /* This code basically does three things:
   
    +) If the incoming segment contains data that is the next     +) If the incoming segment contains data that is the next
    in-sequence data, this data is passed to the application. This     in-sequence data, this data is passed to the application. This
    might involve trimming the first edge of the data. The rcv_nxt     might involve trimming the first edge of the data. The rcv_nxt
    variable and the advertised window are adjusted.     variable and the advertised window are adjusted.
   
    +) If the incoming segment has data that is above the next     +) If the incoming segment has data that is above the next
    sequence number expected (->rcv_nxt), the segment is placed on     sequence number expected (->rcv_nxt), the segment is placed on
    the ->ooseq queue. This is done by finding the appropriate     the ->ooseq queue. This is done by finding the appropriate
    place in the ->ooseq queue (which is ordered by sequence     place in the ->ooseq queue (which is ordered by sequence
    number) and trim the segment in both ends if needed. An     number) and trim the segment in both ends if needed. An
    immediate ACK is sent to indicate that we received an     immediate ACK is sent to indicate that we received an
    out-of-sequence segment.     out-of-sequence segment.
   
    +) Finally, we check if the first segment on the ->ooseq queue     +) Finally, we check if the first segment on the ->ooseq queue
    now is in sequence (i.e., if rcv_nxt >= ooseq->seqno). If     now is in sequence (i.e., if rcv_nxt >= ooseq->seqno). If
    rcv_nxt > ooseq->seqno, we must trim the first edge of the     rcv_nxt > ooseq->seqno, we must trim the first edge of the
    segment on ->ooseq before we adjust rcv_nxt. The data in the     segment on ->ooseq before we adjust rcv_nxt. The data in the
    segments that are now on sequence are chained onto the     segments that are now on sequence are chained onto the
    incoming segment so that we only need to call the application     incoming segment so that we only need to call the application
    once.     once.
    */      */ 
   
    /* First, we check if we must trim the first edge. We have to do     /* 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        this if the sequence number of the incoming segment is less
       than rcv_nxt, and the sequence number plus the length of the        than rcv_nxt, and the sequence number plus the length of the
       segment is larger than rcv_nxt. */         segment is larger than rcv_nxt. */ 
    /*    if (TCP_SEQ_LT(seqno, pcb->rcv_nxt)) {     /*    if (TCP_SEQ_LT(seqno, pcb->rcv_nxt)) {
          if (TCP_SEQ_LT(pcb->rcv_nxt, seqno + tcplen)) {*/            if (TCP_SEQ_LT(pcb->rcv_nxt, seqno + tcplen)) {*/ 
    if (TCP_SEQ_BETWEEN(pcb->rcv_nxt, seqno + 1, seqno + tcplen - 1)) {     if (TCP_SEQ_BETWEEN(pcb->rcv_nxt, seqno + 1, seqno + tcplen - 1)) {
      /* Trimming the first edge is done by pushing the payload       /* Trimming the first edge is done by pushing the payload
         pointer in the pbuf downwards. This is somewhat tricky since          pointer in the pbuf downwards. This is somewhat tricky since
         we do not want to discard the full contents of the pbuf up to          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          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.          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          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          the pbuf chain is pointed to by inseg.p. Since we need to be
         able to deallocate the whole pbuf, we cannot change this          able to deallocate the whole pbuf, we cannot change this
         inseg.p pointer to point to any of the later pbufs in the          inseg.p pointer to point to any of the later pbufs in the
         chain. Instead, we point the ->payload pointer in the first          chain. Instead, we point the ->payload pointer in the first
         pbuf to data in one of the later pbufs. We also set the          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          inseg.data pointer to point to the right place. This way, the
         ->p pointer will still point to the first pbuf, but the          ->p pointer will still point to the first pbuf, but the
         ->p->payload pointer will point to data in another pbuf.          ->p->payload pointer will point to data in another pbuf.
   
         After we are done with adjusting the pbuf pointers we must          After we are done with adjusting the pbuf pointers we must
         adjust the ->data pointer in the seg and the segment          adjust the ->data pointer in the seg and the segment
         length.*/           length.*/ 
   
      struct pbuf *p = inseg.p;       struct pbuf *p = inseg.p;
      u32_t off32 = pcb->rcv_nxt - seqno;       u32_t off32 = pcb->rcv_nxt - seqno;
      u16_t new_tot_len, off;       u16_t new_tot_len, off;
      LWIP_ASSERT("inseg.p != NULL", inseg.p);       LWIP_ASSERT("inseg.p != NULL", inseg.p);
      LWIP_ASSERT("insane offset!", (off32 < 0xffff));       LWIP_ASSERT("insane offset!", (off32 < 0xffff));
      off = (u16_t)off32;       off = (u16_t)off32;
      LWIP_ASSERT("pbuf too short!", (((s32_t)inseg.p->tot_len) >= off));       LWIP_ASSERT("pbuf too short!", (((s32_t)inseg.p->tot_len) >= off));
      inseg.len -= off;       inseg.len -= off;
      new_tot_len = (u16_t)(inseg.p->tot_len - off);       new_tot_len = (u16_t)(inseg.p->tot_len - off);
      while (p->len < off) {       while (p->len < off) {
        off -= p->len;         off -= p->len;
        /* all pbufs up to and including this one have len==0, so tot_len is equal */          /* all pbufs up to and including this one have len==0, so tot_len is equal */ 
        p->tot_len = new_tot_len;         p->tot_len = new_tot_len;
        p->len = 0;         p->len = 0;
        p = p->next;         p = p->next;
      }       }
      /* cannot fail... */        /* cannot fail... */ 
      pbuf_remove_header(p, off);       pbuf_remove_header(p, off);
      inseg.tcphdr->seqno = seqno = pcb->rcv_nxt;       inseg.tcphdr->seqno = seqno = pcb->rcv_nxt;
    } else {     } else {
      if (TCP_SEQ_LT(seqno, pcb->rcv_nxt)) {       if (TCP_SEQ_LT(seqno, pcb->rcv_nxt)) {
        /* the whole segment is < rcv_nxt */          /* the whole segment is < rcv_nxt */ 
        /* must be a duplicate of a packet that has already been correctly handled */          /* must be a duplicate of a packet that has already been correctly handled */ 
   
        LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_receive: duplicate seqno %"U32_F"\n", seqno));         LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_receive: duplicate seqno %"U32_F"\n", seqno));
        tcp_ack_now(pcb);         tcp_ack_now(pcb);
      }       }
    }     }
   
    /* The sequence number must be within the window (above rcv_nxt     /* The sequence number must be within the window (above rcv_nxt
       and below rcv_nxt + rcv_wnd) in order to be further        and below rcv_nxt + rcv_wnd) in order to be further
       processed. */         processed. */ 
    if (TCP_SEQ_BETWEEN(seqno, pcb->rcv_nxt,     if (TCP_SEQ_BETWEEN(seqno, pcb->rcv_nxt,
                        pcb->rcv_nxt + pcb->rcv_wnd - 1)) {                         pcb->rcv_nxt + pcb->rcv_wnd - 1)) {
      if (pcb->rcv_nxt == seqno) {       if (pcb->rcv_nxt == seqno) {
        /* The incoming segment is the next in sequence. We check if         /* The incoming segment is the next in sequence. We check if
           we have to trim the end of the segment and update rcv_nxt            we have to trim the end of the segment and update rcv_nxt
           and pass the data to the application. */             and pass the data to the application. */ 
        tcplen = TCP_TCPLEN(&inseg);         tcplen = TCP_TCPLEN(&inseg);
   
        if (tcplen > pcb->rcv_wnd) {         if (tcplen > pcb->rcv_wnd) {
          LWIP_DEBUGF(TCP_INPUT_DEBUG,           LWIP_DEBUGF(TCP_INPUT_DEBUG,
                      ("tcp_receive: other end overran receive window"                        ("tcp_receive: other end overran receive window" 
                       "seqno %"U32_F" len %"U16_F" right edge %"U32_F"\n",                        "seqno %"U32_F" len %"U16_F" right edge %"U32_F"\n",
                       seqno, tcplen, pcb->rcv_nxt + pcb->rcv_wnd));                        seqno, tcplen, pcb->rcv_nxt + pcb->rcv_wnd));
          if (TCPH_FLAGS(inseg.tcphdr) & TCP_FIN) {           if (TCPH_FLAGS(inseg.tcphdr) & TCP_FIN) {
            /* Must remove the FIN from the header as we're trimming             /* Must remove the FIN from the header as we're trimming
             * that byte of sequence-space from the packet */               * that byte of sequence-space from the packet */ 
            TCPH_FLAGS_SET(inseg.tcphdr, TCPH_FLAGS(inseg.tcphdr) & ~(unsigned int)TCP_FIN);             TCPH_FLAGS_SET(inseg.tcphdr, TCPH_FLAGS(inseg.tcphdr) & ~(unsigned int)TCP_FIN);
          }           }
          /* Adjust length of segment to fit in the window. */            /* Adjust length of segment to fit in the window. */ 
          TCPWND_CHECK16(pcb->rcv_wnd);           TCPWND_CHECK16(pcb->rcv_wnd);
          inseg.len = (u16_t)pcb->rcv_wnd;           inseg.len = (u16_t)pcb->rcv_wnd;
          if (TCPH_FLAGS(inseg.tcphdr) & TCP_SYN) {           if (TCPH_FLAGS(inseg.tcphdr) & TCP_SYN) {
            inseg.len -= 1;             inseg.len -= 1;
          }           }
          pbuf_realloc(inseg.p, inseg.len);           pbuf_realloc(inseg.p, inseg.len);
          tcplen = TCP_TCPLEN(&inseg);           tcplen = TCP_TCPLEN(&inseg);
          LWIP_ASSERT("tcp_receive: segment not trimmed correctly to rcv_wnd\n",           LWIP_ASSERT("tcp_receive: segment not trimmed correctly to rcv_wnd\n",
                      (seqno + tcplen) == (pcb->rcv_nxt + pcb->rcv_wnd));                       (seqno + tcplen) == (pcb->rcv_nxt + pcb->rcv_wnd));
        }         }
#if TCP_QUEUE_OOSEQ #if TCP_QUEUE_OOSEQ
        /* Received in-sequence data, adjust ooseq data if:         /* Received in-sequence data, adjust ooseq data if:
           - FIN has been received or            - FIN has been received or
           - inseq overlaps with ooseq */             - inseq overlaps with ooseq */ 
        if (pcb->ooseq != NULL) {         if (pcb->ooseq != NULL) {
          if (TCPH_FLAGS(inseg.tcphdr) & TCP_FIN) {           if (TCPH_FLAGS(inseg.tcphdr) & TCP_FIN) {
            LWIP_DEBUGF(TCP_INPUT_DEBUG,             LWIP_DEBUGF(TCP_INPUT_DEBUG,
                        ("tcp_receive: received in-order FIN, binning ooseq queue\n"));                         ("tcp_receive: received in-order FIN, binning ooseq queue\n"));
            /* Received in-order FIN means anything that was received             /* Received in-order FIN means anything that was received
             * out of order must now have been received in-order, so              * out of order must now have been received in-order, so
             * bin the ooseq queue */               * bin the ooseq queue */ 
            while (pcb->ooseq != NULL) {             while (pcb->ooseq != NULL) {
              struct tcp_seg *old_ooseq = pcb->ooseq;               struct tcp_seg *old_ooseq = pcb->ooseq;
              pcb->ooseq = pcb->ooseq->next;               pcb->ooseq = pcb->ooseq->next;
              tcp_seg_free(old_ooseq);               tcp_seg_free(old_ooseq);
            }             }
          } else {           } else {
            struct tcp_seg *next = pcb->ooseq;             struct tcp_seg *next = pcb->ooseq;
            /* Remove all segments on ooseq that are covered by inseg already.             /* Remove all segments on ooseq that are covered by inseg already.
             * FIN is copied from ooseq to inseg if present. */               * FIN is copied from ooseq to inseg if present. */ 
            while (next &&             while (next &&
                   TCP_SEQ_GEQ(seqno + tcplen,                    TCP_SEQ_GEQ(seqno + tcplen,
                               next->tcphdr->seqno + next->len)) {                                next->tcphdr->seqno + next->len)) {
              struct tcp_seg *tmp;               struct tcp_seg *tmp;
              /* inseg cannot have FIN here (already processed above) */                /* inseg cannot have FIN here (already processed above) */ 
              if ((TCPH_FLAGS(next->tcphdr) & TCP_FIN) != 0 &&               if ((TCPH_FLAGS(next->tcphdr) & TCP_FIN) != 0 &&
                  (TCPH_FLAGS(inseg.tcphdr) & TCP_SYN) == 0) {                   (TCPH_FLAGS(inseg.tcphdr) & TCP_SYN) == 0) {
                TCPH_SET_FLAG(inseg.tcphdr, TCP_FIN);                 TCPH_SET_FLAG(inseg.tcphdr, TCP_FIN);
                tcplen = TCP_TCPLEN(&inseg);                 tcplen = TCP_TCPLEN(&inseg);
              }               }
              tmp = next;               tmp = next;
              next = next->next;               next = next->next;
              tcp_seg_free(tmp);               tcp_seg_free(tmp);
            }             }
            /* Now trim right side of inseg if it overlaps with the first             /* Now trim right side of inseg if it overlaps with the first
             * segment on ooseq */               * segment on ooseq */ 
            if (next &&             if (next &&
                TCP_SEQ_GT(seqno + tcplen,                 TCP_SEQ_GT(seqno + tcplen,
                           next->tcphdr->seqno)) {                            next->tcphdr->seqno)) {
              /* inseg cannot have FIN here (already processed above) */                /* inseg cannot have FIN here (already processed above) */ 
              inseg.len = (u16_t)(next->tcphdr->seqno - seqno);               inseg.len = (u16_t)(next->tcphdr->seqno - seqno);
              if (TCPH_FLAGS(inseg.tcphdr) & TCP_SYN) {               if (TCPH_FLAGS(inseg.tcphdr) & TCP_SYN) {
                inseg.len -= 1;                 inseg.len -= 1;
              }               }
              pbuf_realloc(inseg.p, inseg.len);               pbuf_realloc(inseg.p, inseg.len);
              tcplen = TCP_TCPLEN(&inseg);               tcplen = TCP_TCPLEN(&inseg);
              LWIP_ASSERT("tcp_receive: segment not trimmed correctly to ooseq queue\n",               LWIP_ASSERT("tcp_receive: segment not trimmed correctly to ooseq queue\n",
                          (seqno + tcplen) == next->tcphdr->seqno);                           (seqno + tcplen) == next->tcphdr->seqno);
            }             }
            pcb->ooseq = next;             pcb->ooseq = next;
          }           }
        }         }
#endif /* TCP_QUEUE_OOSEQ */  #endif /* TCP_QUEUE_OOSEQ */ 
   
        pcb->rcv_nxt = seqno + tcplen;         pcb->rcv_nxt = seqno + tcplen;
   
        /* Update the receiver's (our) window. */          /* Update the receiver's (our) window. */ 
        LWIP_ASSERT("tcp_receive: tcplen > rcv_wnd\n", pcb->rcv_wnd >= tcplen);         LWIP_ASSERT("tcp_receive: tcplen > rcv_wnd\n", pcb->rcv_wnd >= tcplen);
        pcb->rcv_wnd -= tcplen;         pcb->rcv_wnd -= tcplen;
   
        tcp_update_rcv_ann_wnd(pcb);         tcp_update_rcv_ann_wnd(pcb);
   
        /* If there is data in the segment, we make preparations to         /* If there is data in the segment, we make preparations to
           pass this up to the application. The ->recv_data variable            pass this up to the application. The ->recv_data variable
           is used for holding the pbuf that goes to the            is used for holding the pbuf that goes to the
           application. The code for reassembling out-of-sequence data            application. The code for reassembling out-of-sequence data
           chains its data on this pbuf as well.            chains its data on this pbuf as well.
   
           If the segment was a FIN, we set the TF_GOT_FIN flag that will            If the segment was a FIN, we set the TF_GOT_FIN flag that will
           be used to indicate to the application that the remote side has            be used to indicate to the application that the remote side has
           closed its end of the connection. */             closed its end of the connection. */ 
        if (inseg.p->tot_len > 0) {         if (inseg.p->tot_len > 0) {
          recv_data = inseg.p;           recv_data = inseg.p;
          /* Since this pbuf now is the responsibility of the           /* Since this pbuf now is the responsibility of the
             application, we delete our reference to it so that we won't              application, we delete our reference to it so that we won't
             (mistakingly) deallocate it. */               (mistakingly) deallocate it. */ 
          inseg.p = NULL;           inseg.p = NULL;
        }         }
        if (TCPH_FLAGS(inseg.tcphdr) & TCP_FIN) {         if (TCPH_FLAGS(inseg.tcphdr) & TCP_FIN) {
          LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_receive: received FIN.\n"));           LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_receive: received FIN.\n"));
          recv_flags |= TF_GOT_FIN;           recv_flags |= TF_GOT_FIN;
        }         }
   
#if TCP_QUEUE_OOSEQ #if TCP_QUEUE_OOSEQ
        /* We now check if we have segments on the ->ooseq queue that         /* We now check if we have segments on the ->ooseq queue that
           are now in sequence. */             are now in sequence. */ 
        while (pcb->ooseq != NULL &&         while (pcb->ooseq != NULL &&
               pcb->ooseq->tcphdr->seqno == pcb->rcv_nxt) {                pcb->ooseq->tcphdr->seqno == pcb->rcv_nxt) {
   
          struct tcp_seg *cseg = pcb->ooseq;           struct tcp_seg *cseg = pcb->ooseq;
          seqno = pcb->ooseq->tcphdr->seqno;           seqno = pcb->ooseq->tcphdr->seqno;
   
          pcb->rcv_nxt += TCP_TCPLEN(cseg);           pcb->rcv_nxt += TCP_TCPLEN(cseg);
          LWIP_ASSERT("tcp_receive: ooseq tcplen > rcv_wnd\n",           LWIP_ASSERT("tcp_receive: ooseq tcplen > rcv_wnd\n",
                      pcb->rcv_wnd >= TCP_TCPLEN(cseg));                       pcb->rcv_wnd >= TCP_TCPLEN(cseg));
          pcb->rcv_wnd -= TCP_TCPLEN(cseg);           pcb->rcv_wnd -= TCP_TCPLEN(cseg);
   
          tcp_update_rcv_ann_wnd(pcb);           tcp_update_rcv_ann_wnd(pcb);
   
          if (cseg->p->tot_len > 0) {           if (cseg->p->tot_len > 0) {
            /* Chain this pbuf onto the pbuf that we will pass to             /* Chain this pbuf onto the pbuf that we will pass to
               the application. */                 the application. */ 
            /* With window scaling, this can overflow recv_data->tot_len, but             /* With window scaling, this can overflow recv_data->tot_len, but
               that's not a problem since we explicitly fix that before passing                that's not a problem since we explicitly fix that before passing
               recv_data to the application. */                 recv_data to the application. */ 
            if (recv_data) {             if (recv_data) {
              pbuf_cat(recv_data, cseg->p);               pbuf_cat(recv_data, cseg->p);
            } else {             } else {
              recv_data = cseg->p;               recv_data = cseg->p;
            }             }
            cseg->p = NULL;             cseg->p = NULL;
          }           }
          if (TCPH_FLAGS(cseg->tcphdr) & TCP_FIN) {           if (TCPH_FLAGS(cseg->tcphdr) & TCP_FIN) {
            LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_receive: dequeued FIN.\n"));             LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_receive: dequeued FIN.\n"));
            recv_flags |= TF_GOT_FIN;             recv_flags |= TF_GOT_FIN;
            if (pcb->state == ESTABLISHED) { /* force passive close or we can move to active close */              if (pcb->state == ESTABLISHED) { /* force passive close or we can move to active close */ 
              pcb->state = CLOSE_WAIT;               pcb->state = CLOSE_WAIT;
            }             }
          }           }
   
          pcb->ooseq = cseg->next;           pcb->ooseq = cseg->next;
          tcp_seg_free(cseg);           tcp_seg_free(cseg);
        }         }
#if LWIP_TCP_SACK_OUT #if LWIP_TCP_SACK_OUT
        if (pcb->flags & TF_SACK) {         if (pcb->flags & TF_SACK) {
          if (pcb->ooseq != NULL) {           if (pcb->ooseq != NULL) {
            /* Some segments may have been removed from ooseq, let's remove all SACKs that             /* Some segments may have been removed from ooseq, let's remove all SACKs that
               describe anything before the new beginning of that list. */                 describe anything before the new beginning of that list. */ 
            tcp_remove_sacks_lt(pcb, pcb->ooseq->tcphdr->seqno);             tcp_remove_sacks_lt(pcb, pcb->ooseq->tcphdr->seqno);
          } else if (LWIP_TCP_SACK_VALID(pcb, 0)) {           } else if (LWIP_TCP_SACK_VALID(pcb, 0)) {
            /* ooseq has been cleared. Nothing to SACK */              /* ooseq has been cleared. Nothing to SACK */ 
            memset(pcb->rcv_sacks, 0, sizeof(pcb->rcv_sacks));             memset(pcb->rcv_sacks, 0, sizeof(pcb->rcv_sacks));
          }           }
        }         }
#endif /* LWIP_TCP_SACK_OUT */  #endif /* LWIP_TCP_SACK_OUT */ 
#endif /* TCP_QUEUE_OOSEQ */  #endif /* TCP_QUEUE_OOSEQ */ 
   
   
        /* Acknowledge the segment(s). */          /* Acknowledge the segment(s). */ 
        tcp_ack(pcb);         tcp_ack(pcb);
   
#if LWIP_TCP_SACK_OUT #if LWIP_TCP_SACK_OUT
        if (LWIP_TCP_SACK_VALID(pcb, 0)) {         if (LWIP_TCP_SACK_VALID(pcb, 0)) {
          /* Normally the ACK for the data received could be piggy-backed on a data packet,           /* Normally the ACK for the data received could be piggy-backed on a data packet,
             but lwIP currently does not support including SACKs in data packets. So we force              but lwIP currently does not support including SACKs in data packets. So we force
             it to respond with an empty ACK packet (only if there is at least one SACK to be sent).              it to respond with an empty ACK packet (only if there is at least one SACK to be sent).
             NOTE: tcp_send_empty_ack() on success clears the ACK flags (set by tcp_ack()) */               NOTE: tcp_send_empty_ack() on success clears the ACK flags (set by tcp_ack()) */ 
          tcp_send_empty_ack(pcb);           tcp_send_empty_ack(pcb);
        }         }
#endif /* LWIP_TCP_SACK_OUT */  #endif /* LWIP_TCP_SACK_OUT */ 
   
#if LWIP_IPV6 && LWIP_ND6_TCP_REACHABILITY_HINTS #if LWIP_IPV6 && LWIP_ND6_TCP_REACHABILITY_HINTS
        if (ip_current_is_v6()) {         if (ip_current_is_v6()) {
          /* Inform neighbor reachability of forward progress. */            /* Inform neighbor reachability of forward progress. */ 
          nd6_reachability_hint(ip6_current_src_addr());           nd6_reachability_hint(ip6_current_src_addr());
        }         }
#endif /* LWIP_IPV6 && LWIP_ND6_TCP_REACHABILITY_HINTS*/  #endif /* LWIP_IPV6 && LWIP_ND6_TCP_REACHABILITY_HINTS*/ 
   
      } else {       } else {
        /* We get here if the incoming segment is out-of-sequence. */          /* We get here if the incoming segment is out-of-sequence. */ 
   
#if TCP_QUEUE_OOSEQ #if TCP_QUEUE_OOSEQ
        /* We queue the segment on the ->ooseq queue. */          /* We queue the segment on the ->ooseq queue. */ 
        if (pcb->ooseq == NULL) {         if (pcb->ooseq == NULL) {
          pcb->ooseq = tcp_seg_copy(&inseg);           pcb->ooseq = tcp_seg_copy(&inseg);
#if LWIP_TCP_SACK_OUT #if LWIP_TCP_SACK_OUT
          if (pcb->flags & TF_SACK) {           if (pcb->flags & TF_SACK) {
            /* All the SACKs should be invalid, so we can simply store the most recent one: */              /* All the SACKs should be invalid, so we can simply store the most recent one: */ 
            pcb->rcv_sacks[0].left = seqno;             pcb->rcv_sacks[0].left = seqno;
            pcb->rcv_sacks[0].right = seqno + inseg.len;             pcb->rcv_sacks[0].right = seqno + inseg.len;
          }           }
#endif /* LWIP_TCP_SACK_OUT */  #endif /* LWIP_TCP_SACK_OUT */ 
        } else {         } else {
          /* If the queue is not empty, we walk through the queue and           /* If the queue is not empty, we walk through the queue and
             try to find a place where the sequence number of the              try to find a place where the sequence number of the
             incoming segment is between the sequence numbers of the              incoming segment is between the sequence numbers of the
             previous and the next segment on the ->ooseq queue. That is              previous and the next segment on the ->ooseq queue. That is
             the place where we put the incoming segment. If needed, we              the place where we put the incoming segment. If needed, we
             trim the second edges of the previous and the incoming              trim the second edges of the previous and the incoming
             segment so that it will fit into the sequence.              segment so that it will fit into the sequence.
   
             If the incoming segment has the same sequence number as a              If the incoming segment has the same sequence number as a
             segment on the ->ooseq queue, we discard the segment that              segment on the ->ooseq queue, we discard the segment that
             contains less data. */               contains less data. */ 
   
#if LWIP_TCP_SACK_OUT #if LWIP_TCP_SACK_OUT
          /* This is the left edge of the lowest possible SACK range.           /* This is the left edge of the lowest possible SACK range.
             It may start before the newly received segment (possibly adjusted below). */               It may start before the newly received segment (possibly adjusted below). */ 
          u32_t sackbeg = TCP_SEQ_LT(seqno, pcb->ooseq->tcphdr->seqno) ? seqno : pcb->ooseq->tcphdr->seqno;           u32_t sackbeg = TCP_SEQ_LT(seqno, pcb->ooseq->tcphdr->seqno) ? seqno : pcb->ooseq->tcphdr->seqno;
#endif /* LWIP_TCP_SACK_OUT */  #endif /* LWIP_TCP_SACK_OUT */ 
          struct tcp_seg *next, *prev = NULL;           struct tcp_seg *next, *prev = NULL;
          for (next = pcb->ooseq; next != NULL; next = next->next) {           for (next = pcb->ooseq; next != NULL; next = next->next) {
            if (seqno == next->tcphdr->seqno) {             if (seqno == next->tcphdr->seqno) {
              /* The sequence number of the incoming segment is the               /* The sequence number of the incoming segment is the
                 same as the sequence number of the segment on                  same as the sequence number of the segment on
                 ->ooseq. We check the lengths to see which one to                  ->ooseq. We check the lengths to see which one to
                 discard. */                   discard. */ 
              if (inseg.len > next->len) {               if (inseg.len > next->len) {
                /* The incoming segment is larger than the old                 /* The incoming segment is larger than the old
                   segment. We replace some segments with the new                    segment. We replace some segments with the new
                   one. */                     one. */ 
                struct tcp_seg *cseg = tcp_seg_copy(&inseg);                 struct tcp_seg *cseg = tcp_seg_copy(&inseg);
                if (cseg != NULL) {                 if (cseg != NULL) {
                  if (prev != NULL) {                   if (prev != NULL) {
                    prev->next = cseg;                     prev->next = cseg;
                  } else {                   } else {
                    pcb->ooseq = cseg;                     pcb->ooseq = cseg;
                  }                   }
                  tcp_oos_insert_segment(cseg, next);                   tcp_oos_insert_segment(cseg, next);
                }                 }
                break;                 break;
              } else {               } else {
                /* Either the lengths are the same or the incoming                 /* Either the lengths are the same or the incoming
                   segment was smaller than the old one; in either                    segment was smaller than the old one; in either
                   case, we ditch the incoming segment. */                     case, we ditch the incoming segment. */ 
                break;                 break;
              }               }
            } else {             } else {
              if (prev == NULL) {               if (prev == NULL) {
                if (TCP_SEQ_LT(seqno, next->tcphdr->seqno)) {                 if (TCP_SEQ_LT(seqno, next->tcphdr->seqno)) {
                  /* The sequence number of the incoming segment is lower                   /* The sequence number of the incoming segment is lower
                     than the sequence number of the first segment on the                      than the sequence number of the first segment on the
                     queue. We put the incoming segment first on the                      queue. We put the incoming segment first on the
                     queue. */                       queue. */ 
                  struct tcp_seg *cseg = tcp_seg_copy(&inseg);                   struct tcp_seg *cseg = tcp_seg_copy(&inseg);
                  if (cseg != NULL) {                   if (cseg != NULL) {
                    pcb->ooseq = cseg;                     pcb->ooseq = cseg;
                    tcp_oos_insert_segment(cseg, next);                     tcp_oos_insert_segment(cseg, next);
                  }                   }
                  break;                   break;
                }                 }
              } else {               } else {
                /*if (TCP_SEQ_LT(prev->tcphdr->seqno, seqno) &&                 /*if (TCP_SEQ_LT(prev->tcphdr->seqno, seqno) &&
                  TCP_SEQ_LT(seqno, next->tcphdr->seqno)) {*/                    TCP_SEQ_LT(seqno, next->tcphdr->seqno)) {*/ 
                if (TCP_SEQ_BETWEEN(seqno, prev->tcphdr->seqno + 1, next->tcphdr->seqno - 1)) {                 if (TCP_SEQ_BETWEEN(seqno, prev->tcphdr->seqno + 1, next->tcphdr->seqno - 1)) {
                  /* The sequence number of the incoming segment is in                   /* The sequence number of the incoming segment is in
                     between the sequence numbers of the previous and                      between the sequence numbers of the previous and
                     the next segment on ->ooseq. We trim trim the previous                      the next segment on ->ooseq. We trim trim the previous
                     segment, delete next segments that included in received segment                      segment, delete next segments that included in received segment
                     and trim received, if needed. */                       and trim received, if needed. */ 
                  struct tcp_seg *cseg = tcp_seg_copy(&inseg);                   struct tcp_seg *cseg = tcp_seg_copy(&inseg);
                  if (cseg != NULL) {                   if (cseg != NULL) {
                    if (TCP_SEQ_GT(prev->tcphdr->seqno + prev->len, seqno)) {                     if (TCP_SEQ_GT(prev->tcphdr->seqno + prev->len, seqno)) {
                      /* We need to trim the prev segment. */                        /* We need to trim the prev segment. */ 
                      prev->len = (u16_t)(seqno - prev->tcphdr->seqno);                       prev->len = (u16_t)(seqno - prev->tcphdr->seqno);
                      pbuf_realloc(prev->p, prev->len);                       pbuf_realloc(prev->p, prev->len);
                    }                     }
                    prev->next = cseg;                     prev->next = cseg;
                    tcp_oos_insert_segment(cseg, next);                     tcp_oos_insert_segment(cseg, next);
                  }                   }
                  break;                   break;
                }                 }
              }               }
   
#if LWIP_TCP_SACK_OUT #if LWIP_TCP_SACK_OUT
              /* The new segment goes after the 'next' one. If there is a "hole" in sequence numbers               /* The new segment goes after the 'next' one. If there is a "hole" in sequence numbers
                 between 'prev' and the beginning of 'next', we want to move sackbeg. */                   between 'prev' and the beginning of 'next', we want to move sackbeg. */ 
              if (prev != NULL && prev->tcphdr->seqno + prev->len != next->tcphdr->seqno) {               if (prev != NULL && prev->tcphdr->seqno + prev->len != next->tcphdr->seqno) {
                sackbeg = next->tcphdr->seqno;                 sackbeg = next->tcphdr->seqno;
              }               }
#endif /* LWIP_TCP_SACK_OUT */  #endif /* LWIP_TCP_SACK_OUT */ 
   
              /* We don't use 'prev' below, so let's set it to current 'next'.               /* We don't use 'prev' below, so let's set it to current 'next'.
                 This way even if we break the loop below, 'prev' will be pointing                  This way even if we break the loop below, 'prev' will be pointing
                 at the segment right in front of the newly added one. */                   at the segment right in front of the newly added one. */ 
              prev = next;               prev = next;
   
              /* If the "next" segment is the last segment on the               /* If the "next" segment is the last segment on the
                 ooseq queue, we add the incoming segment to the end                  ooseq queue, we add the incoming segment to the end
                 of the list. */                   of the list. */ 
              if (next->next == NULL &&               if (next->next == NULL &&
                  TCP_SEQ_GT(seqno, next->tcphdr->seqno)) {                   TCP_SEQ_GT(seqno, next->tcphdr->seqno)) {
                if (TCPH_FLAGS(next->tcphdr) & TCP_FIN) {                 if (TCPH_FLAGS(next->tcphdr) & TCP_FIN) {
                  /* segment "next" already contains all data */                    /* segment "next" already contains all data */ 
                  break;                   break;
                }                 }
                next->next = tcp_seg_copy(&inseg);                 next->next = tcp_seg_copy(&inseg);
                if (next->next != NULL) {                 if (next->next != NULL) {
                  if (TCP_SEQ_GT(next->tcphdr->seqno + next->len, seqno)) {                   if (TCP_SEQ_GT(next->tcphdr->seqno + next->len, seqno)) {
                    /* We need to trim the last segment. */                      /* We need to trim the last segment. */ 
                    next->len = (u16_t)(seqno - next->tcphdr->seqno);                     next->len = (u16_t)(seqno - next->tcphdr->seqno);
                    pbuf_realloc(next->p, next->len);                     pbuf_realloc(next->p, next->len);
                  }                   }
                  /* check if the remote side overruns our receive window */                    /* check if the remote side overruns our receive window */ 
                  if (TCP_SEQ_GT((u32_t)tcplen + seqno, pcb->rcv_nxt + (u32_t)pcb->rcv_wnd)) {                   if (TCP_SEQ_GT((u32_t)tcplen + seqno, pcb->rcv_nxt + (u32_t)pcb->rcv_wnd)) {
                    LWIP_DEBUGF(TCP_INPUT_DEBUG,                     LWIP_DEBUGF(TCP_INPUT_DEBUG,
                                ("tcp_receive: other end overran receive window"                                  ("tcp_receive: other end overran receive window" 
                                 "seqno %"U32_F" len %"U16_F" right edge %"U32_F"\n",                                  "seqno %"U32_F" len %"U16_F" right edge %"U32_F"\n",
                                 seqno, tcplen, pcb->rcv_nxt + pcb->rcv_wnd));                                  seqno, tcplen, pcb->rcv_nxt + pcb->rcv_wnd));
                    if (TCPH_FLAGS(next->next->tcphdr) & TCP_FIN) {                     if (TCPH_FLAGS(next->next->tcphdr) & TCP_FIN) {
                      /* Must remove the FIN from the header as we're trimming                       /* Must remove the FIN from the header as we're trimming
                       * that byte of sequence-space from the packet */                         * that byte of sequence-space from the packet */ 
                      TCPH_FLAGS_SET(next->next->tcphdr, TCPH_FLAGS(next->next->tcphdr) & ~TCP_FIN);                       TCPH_FLAGS_SET(next->next->tcphdr, TCPH_FLAGS(next->next->tcphdr) & ~TCP_FIN);
                    }                     }
                    /* Adjust length of segment to fit in the window. */                      /* Adjust length of segment to fit in the window. */ 
                    next->next->len = (u16_t)(pcb->rcv_nxt + pcb->rcv_wnd - seqno);                     next->next->len = (u16_t)(pcb->rcv_nxt + pcb->rcv_wnd - seqno);
                    pbuf_realloc(next->next->p, next->next->len);                     pbuf_realloc(next->next->p, next->next->len);
                    tcplen = TCP_TCPLEN(next->next);                     tcplen = TCP_TCPLEN(next->next);
                    LWIP_ASSERT("tcp_receive: segment not trimmed correctly to rcv_wnd\n",                     LWIP_ASSERT("tcp_receive: segment not trimmed correctly to rcv_wnd\n",
                                (seqno + tcplen) == (pcb->rcv_nxt + pcb->rcv_wnd));                                 (seqno + tcplen) == (pcb->rcv_nxt + pcb->rcv_wnd));
                  }                   }
                }                 }
                break;                 break;
              }               }
            }             }
          }           }
   
#if LWIP_TCP_SACK_OUT #if LWIP_TCP_SACK_OUT
          if (pcb->flags & TF_SACK) {           if (pcb->flags & TF_SACK) {
            if (prev == NULL) {             if (prev == NULL) {
              /* The new segment is at the beginning. sackbeg should already be set properly.               /* The new segment is at the beginning. sackbeg should already be set properly.
                 We need to find the right edge. */                   We need to find the right edge. */ 
              next = pcb->ooseq;               next = pcb->ooseq;
            } else if (prev->next != NULL) {             } else if (prev->next != NULL) {
              /* The new segment was added after 'prev'. If there is a "hole" between 'prev' and 'prev->next',               /* The new segment was added after 'prev'. If there is a "hole" between 'prev' and 'prev->next',
                 we need to move sackbeg. After that we should find the right edge. */                   we need to move sackbeg. After that we should find the right edge. */ 
              next = prev->next;               next = prev->next;
              if (prev->tcphdr->seqno + prev->len != next->tcphdr->seqno) {               if (prev->tcphdr->seqno + prev->len != next->tcphdr->seqno) {
                sackbeg = next->tcphdr->seqno;                 sackbeg = next->tcphdr->seqno;
              }               }
            } else {             } else {
              next = NULL;               next = NULL;
            }             }
            if (next != NULL) {             if (next != NULL) {
              u32_t sackend = next->tcphdr->seqno;               u32_t sackend = next->tcphdr->seqno;
              for ( ; (next != NULL) && (sackend == next->tcphdr->seqno); next = next->next) {               for ( ; (next != NULL) && (sackend == next->tcphdr->seqno); next = next->next) {
                sackend += next->len;                 sackend += next->len;
              }               }
              tcp_add_sack(pcb, sackbeg, sackend);               tcp_add_sack(pcb, sackbeg, sackend);
            }             }
          }           }
#endif /* LWIP_TCP_SACK_OUT */  #endif /* LWIP_TCP_SACK_OUT */ 
        }         }
#if defined(TCP_OOSEQ_BYTES_LIMIT) || defined(TCP_OOSEQ_PBUFS_LIMIT) #if defined(TCP_OOSEQ_BYTES_LIMIT) || defined(TCP_OOSEQ_PBUFS_LIMIT)
        {         {
          /* Check that the data on ooseq doesn't exceed one of the limits           /* Check that the data on ooseq doesn't exceed one of the limits
             and throw away everything above that limit. */               and throw away everything above that limit. */ 
#ifdef TCP_OOSEQ_BYTES_LIMIT #ifdef TCP_OOSEQ_BYTES_LIMIT
          const u32_t ooseq_max_blen = TCP_OOSEQ_BYTES_LIMIT(pcb);           const u32_t ooseq_max_blen = TCP_OOSEQ_BYTES_LIMIT(pcb);
          u32_t ooseq_blen = 0;           u32_t ooseq_blen = 0;
#endif #endif
#ifdef TCP_OOSEQ_PBUFS_LIMIT #ifdef TCP_OOSEQ_PBUFS_LIMIT
          const u16_t ooseq_max_qlen = TCP_OOSEQ_PBUFS_LIMIT(pcb);           const u16_t ooseq_max_qlen = TCP_OOSEQ_PBUFS_LIMIT(pcb);
          u16_t ooseq_qlen = 0;           u16_t ooseq_qlen = 0;
#endif #endif
          struct tcp_seg *next, *prev = NULL;           struct tcp_seg *next, *prev = NULL;
          for (next = pcb->ooseq; next != NULL; prev = next, next = next->next) {           for (next = pcb->ooseq; next != NULL; prev = next, next = next->next) {
            struct pbuf *p = next->p;             struct pbuf *p = next->p;
            int stop_here = 0;             int stop_here = 0;
#ifdef TCP_OOSEQ_BYTES_LIMIT #ifdef TCP_OOSEQ_BYTES_LIMIT
            ooseq_blen += p->tot_len;             ooseq_blen += p->tot_len;
            if (ooseq_blen > ooseq_max_blen) {             if (ooseq_blen > ooseq_max_blen) {
              stop_here = 1;               stop_here = 1;
            }             }
#endif #endif
#ifdef TCP_OOSEQ_PBUFS_LIMIT #ifdef TCP_OOSEQ_PBUFS_LIMIT
            ooseq_qlen += pbuf_clen(p);             ooseq_qlen += pbuf_clen(p);
            if (ooseq_qlen > ooseq_max_qlen) {             if (ooseq_qlen > ooseq_max_qlen) {
              stop_here = 1;               stop_here = 1;
            }             }
#endif #endif
            if (stop_here) {             if (stop_here) {
#if LWIP_TCP_SACK_OUT #if LWIP_TCP_SACK_OUT
              if (pcb->flags & TF_SACK) {               if (pcb->flags & TF_SACK) {
                /* Let's remove all SACKs from next's seqno up. */                  /* Let's remove all SACKs from next's seqno up. */ 
                tcp_remove_sacks_gt(pcb, next->tcphdr->seqno);                 tcp_remove_sacks_gt(pcb, next->tcphdr->seqno);
              }               }
#endif /* LWIP_TCP_SACK_OUT */  #endif /* LWIP_TCP_SACK_OUT */ 
              /* too much ooseq data, dump this and everything after it */                /* too much ooseq data, dump this and everything after it */ 
              tcp_segs_free(next);               tcp_segs_free(next);
              if (prev == NULL) {               if (prev == NULL) {
                /* first ooseq segment is too much, dump the whole queue */                  /* first ooseq segment is too much, dump the whole queue */ 
                pcb->ooseq = NULL;                 pcb->ooseq = NULL;
              } else {               } else {
                /* just dump 'next' and everything after it */                  /* just dump 'next' and everything after it */ 
                prev->next = NULL;                 prev->next = NULL;
              }               }
              break;               break;
            }             }
          }           }
        }         }
#endif /* TCP_OOSEQ_BYTES_LIMIT || TCP_OOSEQ_PBUFS_LIMIT */  #endif /* TCP_OOSEQ_BYTES_LIMIT || TCP_OOSEQ_PBUFS_LIMIT */ 
#endif /* TCP_QUEUE_OOSEQ */  #endif /* TCP_QUEUE_OOSEQ */ 
   
        /* We send the ACK packet after we've (potentially) dealt with SACKs,         /* We send the ACK packet after we've (potentially) dealt with SACKs,
           so they can be included in the acknowledgment. */             so they can be included in the acknowledgment. */ 
        tcp_send_empty_ack(pcb);         tcp_send_empty_ack(pcb);
      }       }
    } else {     } else {
      /* The incoming segment is not within the window. */        /* The incoming segment is not within the window. */ 
      tcp_send_empty_ack(pcb);       tcp_send_empty_ack(pcb);
    }     }
  } else {   } else {
    /* Segments with length 0 is taken care of here. Segments that     /* Segments with length 0 is taken care of here. Segments that
       fall out of the window are ACKed. */         fall out of the window are ACKed. */ 
    if (!TCP_SEQ_BETWEEN(seqno, pcb->rcv_nxt, pcb->rcv_nxt + pcb->rcv_wnd - 1)) {     if (!TCP_SEQ_BETWEEN(seqno, pcb->rcv_nxt, pcb->rcv_nxt + pcb->rcv_wnd - 1)) {
      tcp_ack_now(pcb);       tcp_ack_now(pcb);
    }     }
  }   }
} }
   
static u8_t static u8_t
tcp_get_next_optbyte(void) tcp_get_next_optbyte(void)
{ {
  u16_t optidx = tcp_optidx++;   u16_t optidx = tcp_optidx++;
  if ((tcphdr_opt2 == NULL) || (optidx < tcphdr_opt1len)) {   if ((tcphdr_opt2 == NULL) || (optidx < tcphdr_opt1len)) {
    u8_t *opts = (u8_t *)tcphdr + TCP_HLEN;     u8_t *opts = (u8_t *)tcphdr + TCP_HLEN;
    return opts[optidx];     return opts[optidx];
  } else {   } else {
    u8_t idx = (u8_t)(optidx - tcphdr_opt1len);     u8_t idx = (u8_t)(optidx - tcphdr_opt1len);
    return tcphdr_opt2[idx];     return tcphdr_opt2[idx];
  }   }
} }
   
/** /**
 * Parses the options contained in the incoming segment.  * Parses the options contained in the incoming segment.
 *  *
 * Called from tcp_listen_input() and tcp_process().  * Called from tcp_listen_input() and tcp_process().
 * Currently, only the MSS option is supported!  * Currently, only the MSS option is supported!
 *  *
 * @param pcb the tcp_pcb for which a segment arrived  * @param pcb the tcp_pcb for which a segment arrived
 */   */ 
static void static void
tcp_parseopt(struct tcp_pcb *pcb) tcp_parseopt(struct tcp_pcb *pcb)
{ {
  u8_t data;   u8_t data;
  u16_t mss;   u16_t mss;
#if LWIP_TCP_TIMESTAMPS #if LWIP_TCP_TIMESTAMPS
  u32_t tsval;   u32_t tsval;
#endif #endif
   
  LWIP_ASSERT("tcp_parseopt: invalid pcb", pcb != NULL);   LWIP_ASSERT("tcp_parseopt: invalid pcb", pcb != NULL);
   
  /* Parse the TCP MSS option, if present. */    /* Parse the TCP MSS option, if present. */ 
  if (tcphdr_optlen != 0) {   if (tcphdr_optlen != 0) {
    for (tcp_optidx = 0; tcp_optidx < tcphdr_optlen; ) {     for (tcp_optidx = 0; tcp_optidx < tcphdr_optlen; ) {
      u8_t opt = tcp_get_next_optbyte();       u8_t opt = tcp_get_next_optbyte();
      switch (opt) {       switch (opt) {
        case LWIP_TCP_OPT_EOL:         case LWIP_TCP_OPT_EOL:
          /* End of options. */            /* End of options. */ 
          LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_parseopt: EOL\n"));           LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_parseopt: EOL\n"));
          return;           return;
        case LWIP_TCP_OPT_NOP:         case LWIP_TCP_OPT_NOP:
          /* NOP option. */            /* NOP option. */ 
          LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_parseopt: NOP\n"));           LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_parseopt: NOP\n"));
          break;           break;
        case LWIP_TCP_OPT_MSS:         case LWIP_TCP_OPT_MSS:
          LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_parseopt: MSS\n"));           LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_parseopt: MSS\n"));
          if (tcp_get_next_optbyte() != LWIP_TCP_OPT_LEN_MSS || (tcp_optidx - 2 + LWIP_TCP_OPT_LEN_MSS) > tcphdr_optlen) {           if (tcp_get_next_optbyte() != LWIP_TCP_OPT_LEN_MSS || (tcp_optidx - 2 + LWIP_TCP_OPT_LEN_MSS) > tcphdr_optlen) {
            /* Bad length */              /* Bad length */ 
            LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_parseopt: bad length\n"));             LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_parseopt: bad length\n"));
            return;             return;
          }           }
          /* An MSS option with the right option length. */            /* An MSS option with the right option length. */ 
          mss = (u16_t)(tcp_get_next_optbyte() << 8);           mss = (u16_t)(tcp_get_next_optbyte() << 8);
          mss |= tcp_get_next_optbyte();           mss |= tcp_get_next_optbyte();
          /* Limit the mss to the configured TCP_MSS and prevent division by zero */            /* Limit the mss to the configured TCP_MSS and prevent division by zero */ 
          pcb->mss = ((mss > TCP_MSS) || (mss == 0)) ? TCP_MSS : mss;           pcb->mss = ((mss > TCP_MSS) || (mss == 0)) ? TCP_MSS : mss;
          break;           break;
#if LWIP_WND_SCALE #if LWIP_WND_SCALE
        case LWIP_TCP_OPT_WS:         case LWIP_TCP_OPT_WS:
          LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_parseopt: WND_SCALE\n"));           LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_parseopt: WND_SCALE\n"));
          if (tcp_get_next_optbyte() != LWIP_TCP_OPT_LEN_WS || (tcp_optidx - 2 + LWIP_TCP_OPT_LEN_WS) > tcphdr_optlen) {           if (tcp_get_next_optbyte() != LWIP_TCP_OPT_LEN_WS || (tcp_optidx - 2 + LWIP_TCP_OPT_LEN_WS) > tcphdr_optlen) {
            /* Bad length */              /* Bad length */ 
            LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_parseopt: bad length\n"));             LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_parseopt: bad length\n"));
            return;             return;
          }           }
          /* An WND_SCALE option with the right option length. */            /* An WND_SCALE option with the right option length. */ 
          data = tcp_get_next_optbyte();           data = tcp_get_next_optbyte();
          /* If syn was received with wnd scale option,           /* If syn was received with wnd scale option,
             activate wnd scale opt, but only if this is not a retransmission */               activate wnd scale opt, but only if this is not a retransmission */ 
          if ((flags & TCP_SYN) && !(pcb->flags & TF_WND_SCALE)) {           if ((flags & TCP_SYN) && !(pcb->flags & TF_WND_SCALE)) {
            pcb->snd_scale = data;             pcb->snd_scale = data;
            if (pcb->snd_scale > 14U) {             if (pcb->snd_scale > 14U) {
              pcb->snd_scale = 14U;               pcb->snd_scale = 14U;
            }             }
            pcb->rcv_scale = TCP_RCV_SCALE;             pcb->rcv_scale = TCP_RCV_SCALE;
            tcp_set_flags(pcb, TF_WND_SCALE);             tcp_set_flags(pcb, TF_WND_SCALE);
            /* window scaling is enabled, we can use the full receive window */              /* window scaling is enabled, we can use the full receive window */ 
            LWIP_ASSERT("window not at default value", pcb->rcv_wnd == TCPWND_MIN16(TCP_WND));             LWIP_ASSERT("window not at default value", pcb->rcv_wnd == TCPWND_MIN16(TCP_WND));
            LWIP_ASSERT("window not at default value", pcb->rcv_ann_wnd == TCPWND_MIN16(TCP_WND));             LWIP_ASSERT("window not at default value", pcb->rcv_ann_wnd == TCPWND_MIN16(TCP_WND));
            pcb->rcv_wnd = pcb->rcv_ann_wnd = TCP_WND;             pcb->rcv_wnd = pcb->rcv_ann_wnd = TCP_WND;
          }           }
          break;           break;
#endif /* LWIP_WND_SCALE */  #endif /* LWIP_WND_SCALE */ 
#if LWIP_TCP_TIMESTAMPS #if LWIP_TCP_TIMESTAMPS
        case LWIP_TCP_OPT_TS:         case LWIP_TCP_OPT_TS:
          LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_parseopt: TS\n"));           LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_parseopt: TS\n"));
          if (tcp_get_next_optbyte() != LWIP_TCP_OPT_LEN_TS || (tcp_optidx - 2 + LWIP_TCP_OPT_LEN_TS) > tcphdr_optlen) {           if (tcp_get_next_optbyte() != LWIP_TCP_OPT_LEN_TS || (tcp_optidx - 2 + LWIP_TCP_OPT_LEN_TS) > tcphdr_optlen) {
            /* Bad length */              /* Bad length */ 
            LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_parseopt: bad length\n"));             LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_parseopt: bad length\n"));
            return;             return;
          }           }
          /* TCP timestamp option with valid length */            /* TCP timestamp option with valid length */ 
          tsval = tcp_get_next_optbyte();           tsval = tcp_get_next_optbyte();
          tsval |= (tcp_get_next_optbyte() << 8);           tsval |= (tcp_get_next_optbyte() << 8);
          tsval |= (tcp_get_next_optbyte() << 16);           tsval |= (tcp_get_next_optbyte() << 16);
          tsval |= (tcp_get_next_optbyte() << 24);           tsval |= (tcp_get_next_optbyte() << 24);
          if (flags & TCP_SYN) {           if (flags & TCP_SYN) {
            pcb->ts_recent = lwip_ntohl(tsval);             pcb->ts_recent = lwip_ntohl(tsval);
            /* Enable sending timestamps in every segment now that we know             /* Enable sending timestamps in every segment now that we know
               the remote host supports it. */                 the remote host supports it. */ 
            tcp_set_flags(pcb, TF_TIMESTAMP);             tcp_set_flags(pcb, TF_TIMESTAMP);
          } else if (TCP_SEQ_BETWEEN(pcb->ts_lastacksent, seqno, seqno + tcplen)) {           } else if (TCP_SEQ_BETWEEN(pcb->ts_lastacksent, seqno, seqno + tcplen)) {
            pcb->ts_recent = lwip_ntohl(tsval);             pcb->ts_recent = lwip_ntohl(tsval);
          }           }
          /* Advance to next option (6 bytes already read) */            /* Advance to next option (6 bytes already read) */ 
          tcp_optidx += LWIP_TCP_OPT_LEN_TS - 6;           tcp_optidx += LWIP_TCP_OPT_LEN_TS - 6;
          break;           break;
#endif /* LWIP_TCP_TIMESTAMPS */  #endif /* LWIP_TCP_TIMESTAMPS */ 
#if LWIP_TCP_SACK_OUT #if LWIP_TCP_SACK_OUT
        case LWIP_TCP_OPT_SACK_PERM:         case LWIP_TCP_OPT_SACK_PERM:
          LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_parseopt: SACK_PERM\n"));           LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_parseopt: SACK_PERM\n"));
          if (tcp_get_next_optbyte() != LWIP_TCP_OPT_LEN_SACK_PERM || (tcp_optidx - 2 + LWIP_TCP_OPT_LEN_SACK_PERM) > tcphdr_optlen) {           if (tcp_get_next_optbyte() != LWIP_TCP_OPT_LEN_SACK_PERM || (tcp_optidx - 2 + LWIP_TCP_OPT_LEN_SACK_PERM) > tcphdr_optlen) {
            /* Bad length */              /* Bad length */ 
            LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_parseopt: bad length\n"));             LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_parseopt: bad length\n"));
            return;             return;
          }           }
          /* TCP SACK_PERM option with valid length */            /* TCP SACK_PERM option with valid length */ 
          if (flags & TCP_SYN) {           if (flags & TCP_SYN) {
            /* We only set it if we receive it in a SYN (or SYN+ACK) packet */              /* We only set it if we receive it in a SYN (or SYN+ACK) packet */ 
            tcp_set_flags(pcb, TF_SACK);             tcp_set_flags(pcb, TF_SACK);
          }           }
          break;           break;
#endif /* LWIP_TCP_SACK_OUT */  #endif /* LWIP_TCP_SACK_OUT */ 
        default:         default:
          LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_parseopt: other\n"));           LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_parseopt: other\n"));
          data = tcp_get_next_optbyte();           data = tcp_get_next_optbyte();
          if (data < 2) {           if (data < 2) {
            LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_parseopt: bad length\n"));             LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_parseopt: bad length\n"));
            /* If the length field is zero, the options are malformed             /* If the length field is zero, the options are malformed
               and we don't process them further. */                 and we don't process them further. */ 
            return;             return;
          }           }
          /* All other options have a length field, so that we easily           /* All other options have a length field, so that we easily
             can skip past them. */               can skip past them. */ 
          tcp_optidx += data - 2;           tcp_optidx += data - 2;
      }       }
    }     }
  }   }
} }
   
void void
tcp_trigger_input_pcb_close(void) tcp_trigger_input_pcb_close(void)
{ {
  recv_flags |= TF_CLOSED;   recv_flags |= TF_CLOSED;
} }
   
#if LWIP_TCP_SACK_OUT #if LWIP_TCP_SACK_OUT
/** /**
 * Called by tcp_receive() to add new SACK entry.  * Called by tcp_receive() to add new SACK entry.
 *  *
 * The new SACK entry will be placed at the beginning of rcv_sacks[], as the newest one.  * The new SACK entry will be placed at the beginning of rcv_sacks[], as the newest one.
 * Existing SACK entries will be "pushed back", to preserve their order.  * Existing SACK entries will be "pushed back", to preserve their order.
 * This is the behavior described in RFC 2018, section 4.  * This is the behavior described in RFC 2018, section 4.
 *  *
 * @param pcb the tcp_pcb for which a segment arrived  * @param pcb the tcp_pcb for which a segment arrived
 * @param left the left side of the SACK (the first sequence number)  * @param left the left side of the SACK (the first sequence number)
 * @param right the right side of the SACK (the first sequence number past this SACK)  * @param right the right side of the SACK (the first sequence number past this SACK)
 */   */ 
static void static void
tcp_add_sack(struct tcp_pcb *pcb, u32_t left, u32_t right) tcp_add_sack(struct tcp_pcb *pcb, u32_t left, u32_t right)
{ {
  u8_t i;   u8_t i;
  u8_t unused_idx;   u8_t unused_idx;
   
  if ((pcb->flags & TF_SACK) == 0 || !TCP_SEQ_LT(left, right)) {   if ((pcb->flags & TF_SACK) == 0 || !TCP_SEQ_LT(left, right)) {
    return;     return;
  }   }
   
  /* First, let's remove all SACKs that are no longer needed (because they overlap with the newest one),   /* First, let's remove all SACKs that are no longer needed (because they overlap with the newest one),
     while moving all other SACKs forward.      while moving all other SACKs forward.
     We run this loop for all entries, until we find the first invalid one.      We run this loop for all entries, until we find the first invalid one.
     There is no point checking after that. */       There is no point checking after that. */ 
  for (i = unused_idx = 0; (i < LWIP_TCP_MAX_SACK_NUM) && LWIP_TCP_SACK_VALID(pcb, i); ++i) {   for (i = unused_idx = 0; (i < LWIP_TCP_MAX_SACK_NUM) && LWIP_TCP_SACK_VALID(pcb, i); ++i) {
    /* We only want to use SACK at [i] if it doesn't overlap with left:right range.     /* We only want to use SACK at [i] if it doesn't overlap with left:right range.
       It does not overlap if its right side is before the newly added SACK,        It does not overlap if its right side is before the newly added SACK,
       or if its left side is after the newly added SACK.        or if its left side is after the newly added SACK.
       NOTE: The equality should not really happen, but it doesn't hurt. */         NOTE: The equality should not really happen, but it doesn't hurt. */ 
    if (TCP_SEQ_LEQ(pcb->rcv_sacks[i].right, left) || TCP_SEQ_LEQ(right, pcb->rcv_sacks[i].left)) {     if (TCP_SEQ_LEQ(pcb->rcv_sacks[i].right, left) || TCP_SEQ_LEQ(right, pcb->rcv_sacks[i].left)) {
      if (unused_idx != i) {       if (unused_idx != i) {
        /* We don't need to copy if it's already in the right spot */          /* We don't need to copy if it's already in the right spot */ 
        pcb->rcv_sacks[unused_idx] = pcb->rcv_sacks[i];         pcb->rcv_sacks[unused_idx] = pcb->rcv_sacks[i];
      }       }
      ++unused_idx;       ++unused_idx;
    }     }
  }   }
   
  /* Now 'unused_idx' is the index of the first invalid SACK entry,   /* Now 'unused_idx' is the index of the first invalid SACK entry,
     anywhere between 0 (no valid entries) and LWIP_TCP_MAX_SACK_NUM (all entries are valid).      anywhere between 0 (no valid entries) and LWIP_TCP_MAX_SACK_NUM (all entries are valid).
     We want to clear this and all following SACKs.      We want to clear this and all following SACKs.
     However, we will be adding another one in the front (and shifting everything else back).      However, we will be adding another one in the front (and shifting everything else back).
     So let's just iterate from the back, and set each entry to the one to the left if it's valid,      So let's just iterate from the back, and set each entry to the one to the left if it's valid,
     or to 0 if it is not. */       or to 0 if it is not. */ 
  for (i = LWIP_TCP_MAX_SACK_NUM - 1; i > 0; --i) {   for (i = LWIP_TCP_MAX_SACK_NUM - 1; i > 0; --i) {
    /* [i] is the index we are setting, and the value should be at index [i-1],     /* [i] is the index we are setting, and the value should be at index [i-1],
       or 0 if that index is unused (>= unused_idx). */         or 0 if that index is unused (>= unused_idx). */ 
    if (i - 1 >= unused_idx) {     if (i - 1 >= unused_idx) {
      /* [i-1] is unused. Let's clear [i]. */        /* [i-1] is unused. Let's clear [i]. */ 
      pcb->rcv_sacks[i].left = pcb->rcv_sacks[i].right = 0;       pcb->rcv_sacks[i].left = pcb->rcv_sacks[i].right = 0;
    } else {     } else {
      pcb->rcv_sacks[i] = pcb->rcv_sacks[i - 1];       pcb->rcv_sacks[i] = pcb->rcv_sacks[i - 1];
    }     }
  }   }
   
  /* And now we can store the newest SACK */    /* And now we can store the newest SACK */ 
  pcb->rcv_sacks[0].left = left;   pcb->rcv_sacks[0].left = left;
  pcb->rcv_sacks[0].right = right;   pcb->rcv_sacks[0].right = right;
} }
   
/** /**
 * Called to remove a range of SACKs.  * Called to remove a range of SACKs.
 *  *
 * SACK entries will be removed or adjusted to not acknowledge any sequence  * SACK entries will be removed or adjusted to not acknowledge any sequence
 * numbers that are less than 'seq' passed. It not only invalidates entries,  * numbers that are less than 'seq' passed. It not only invalidates entries,
 * but also moves all entries that are still valid to the beginning.  * but also moves all entries that are still valid to the beginning.
 *  *
 * @param pcb the tcp_pcb to modify  * @param pcb the tcp_pcb to modify
 * @param seq the lowest sequence number to keep in SACK entries  * @param seq the lowest sequence number to keep in SACK entries
 */   */ 
static void static void
tcp_remove_sacks_lt(struct tcp_pcb *pcb, u32_t seq) tcp_remove_sacks_lt(struct tcp_pcb *pcb, u32_t seq)
{ {
  u8_t i;   u8_t i;
  u8_t unused_idx;   u8_t unused_idx;
   
  /* We run this loop for all entries, until we find the first invalid one.   /* We run this loop for all entries, until we find the first invalid one.
     There is no point checking after that. */       There is no point checking after that. */ 
  for (i = unused_idx = 0; (i < LWIP_TCP_MAX_SACK_NUM) && LWIP_TCP_SACK_VALID(pcb, i); ++i) {   for (i = unused_idx = 0; (i < LWIP_TCP_MAX_SACK_NUM) && LWIP_TCP_SACK_VALID(pcb, i); ++i) {
    /* We only want to use SACK at index [i] if its right side is > 'seq'. */      /* We only want to use SACK at index [i] if its right side is > 'seq'. */ 
    if (TCP_SEQ_GT(pcb->rcv_sacks[i].right, seq)) {     if (TCP_SEQ_GT(pcb->rcv_sacks[i].right, seq)) {
      if (unused_idx != i) {       if (unused_idx != i) {
        /* We only copy it if it's not in the right spot already. */          /* We only copy it if it's not in the right spot already. */ 
        pcb->rcv_sacks[unused_idx] = pcb->rcv_sacks[i];         pcb->rcv_sacks[unused_idx] = pcb->rcv_sacks[i];
      }       }
      /* NOTE: It is possible that its left side is < 'seq', in which case we should adjust it. */        /* NOTE: It is possible that its left side is < 'seq', in which case we should adjust it. */ 
      if (TCP_SEQ_LT(pcb->rcv_sacks[unused_idx].left, seq)) {       if (TCP_SEQ_LT(pcb->rcv_sacks[unused_idx].left, seq)) {
        pcb->rcv_sacks[unused_idx].left = seq;         pcb->rcv_sacks[unused_idx].left = seq;
      }       }
      ++unused_idx;       ++unused_idx;
    }     }
  }   }
   
  /* We also need to invalidate everything from 'unused_idx' till the end */    /* We also need to invalidate everything from 'unused_idx' till the end */ 
  for (i = unused_idx; i < LWIP_TCP_MAX_SACK_NUM; ++i) {   for (i = unused_idx; i < LWIP_TCP_MAX_SACK_NUM; ++i) {
    pcb->rcv_sacks[i].left = pcb->rcv_sacks[i].right = 0;     pcb->rcv_sacks[i].left = pcb->rcv_sacks[i].right = 0;
  }   }
} }
   
#if defined(TCP_OOSEQ_BYTES_LIMIT) || defined(TCP_OOSEQ_PBUFS_LIMIT) #if defined(TCP_OOSEQ_BYTES_LIMIT) || defined(TCP_OOSEQ_PBUFS_LIMIT)
/** /**
 * Called to remove a range of SACKs.  * Called to remove a range of SACKs.
 *  *
 * SACK entries will be removed or adjusted to not acknowledge any sequence  * SACK entries will be removed or adjusted to not acknowledge any sequence
 * numbers that are greater than (or equal to) 'seq' passed. It not only invalidates entries,  * numbers that are greater than (or equal to) 'seq' passed. It not only invalidates entries,
 * but also moves all entries that are still valid to the beginning.  * but also moves all entries that are still valid to the beginning.
 *  *
 * @param pcb the tcp_pcb to modify  * @param pcb the tcp_pcb to modify
 * @param seq the highest sequence number to keep in SACK entries  * @param seq the highest sequence number to keep in SACK entries
 */   */ 
static void static void
tcp_remove_sacks_gt(struct tcp_pcb *pcb, u32_t seq) tcp_remove_sacks_gt(struct tcp_pcb *pcb, u32_t seq)
{ {
  u8_t i;   u8_t i;
  u8_t unused_idx;   u8_t unused_idx;
   
  /* We run this loop for all entries, until we find the first invalid one.   /* We run this loop for all entries, until we find the first invalid one.
     There is no point checking after that. */       There is no point checking after that. */ 
  for (i = unused_idx = 0; (i < LWIP_TCP_MAX_SACK_NUM) && LWIP_TCP_SACK_VALID(pcb, i); ++i) {   for (i = unused_idx = 0; (i < LWIP_TCP_MAX_SACK_NUM) && LWIP_TCP_SACK_VALID(pcb, i); ++i) {
    /* We only want to use SACK at index [i] if its left side is < 'seq'. */      /* We only want to use SACK at index [i] if its left side is < 'seq'. */ 
    if (TCP_SEQ_LT(pcb->rcv_sacks[i].left, seq)) {     if (TCP_SEQ_LT(pcb->rcv_sacks[i].left, seq)) {
      if (unused_idx != i) {       if (unused_idx != i) {
        /* We only copy it if it's not in the right spot already. */          /* We only copy it if it's not in the right spot already. */ 
        pcb->rcv_sacks[unused_idx] = pcb->rcv_sacks[i];         pcb->rcv_sacks[unused_idx] = pcb->rcv_sacks[i];
      }       }
      /* NOTE: It is possible that its right side is > 'seq', in which case we should adjust it. */        /* NOTE: It is possible that its right side is > 'seq', in which case we should adjust it. */ 
      if (TCP_SEQ_GT(pcb->rcv_sacks[unused_idx].right, seq)) {       if (TCP_SEQ_GT(pcb->rcv_sacks[unused_idx].right, seq)) {
        pcb->rcv_sacks[unused_idx].right = seq;         pcb->rcv_sacks[unused_idx].right = seq;
      }       }
      ++unused_idx;       ++unused_idx;
    }     }
  }   }
   
  /* We also need to invalidate everything from 'unused_idx' till the end */    /* We also need to invalidate everything from 'unused_idx' till the end */ 
  for (i = unused_idx; i < LWIP_TCP_MAX_SACK_NUM; ++i) {   for (i = unused_idx; i < LWIP_TCP_MAX_SACK_NUM; ++i) {
    pcb->rcv_sacks[i].left = pcb->rcv_sacks[i].right = 0;     pcb->rcv_sacks[i].left = pcb->rcv_sacks[i].right = 0;
  }   }
} }
#endif /* TCP_OOSEQ_BYTES_LIMIT || TCP_OOSEQ_PBUFS_LIMIT */  #endif /* TCP_OOSEQ_BYTES_LIMIT || TCP_OOSEQ_PBUFS_LIMIT */ 
   
#endif /* LWIP_TCP_SACK_OUT */  #endif /* LWIP_TCP_SACK_OUT */ 
   
#endif /* LWIP_TCP */  #endif /* LWIP_TCP */