Fix three bugs during tests.

etharp_output() now returns err_t.
etharp_query() now sends a packet directly if the IP address is stable.

CHANGELOG

@@ -1,12 +1,37 @@
-TODO
+FUTURE
 
   * TODO: The lwIP source code makes some invalid assumptions on processor
     word-length, storage sizes and alignment. See the mailing lists for
-    problems with exoteric architectures showing these problems.
-    We still have to fix this neatly.
+    problems with exoteric (/DSP) architectures showing these problems.
+    We still have to fix some of these issues neatly.
 
 HISTORY
 
+(HEAD)
+
+  ++ Changes:
+
+  2004-04-29 Leon Woestenberg <leon.woestenberg@gmx.net>
+  * tcp*.c: Cleaned up source comment documentation for Doxygen processing.
+  * opt.h: ETHHARP_ALWAYS_INSERT option removed to comply with ARP RFC.
+  * etharp.c: update_arp_entry() only adds new ARP entries when adviced to by
+    the caller. This deprecates the ETHHARP_ALWAYS_INSERT overrule option.
+
+  ++ Bug fixes:
+
+  2004-04-27 Leon Woestenberg <leon.woestenberg@gmx.net>
+  * etharp.c: Applied patch of bug #8708 by Toni Mountifield with a solution
+    suggested by Timmy Brolin. Fix for 32-bit processors that cannot access
+    non-aligned 32-bit words, such as soms 32-bit TCP/IP header fields. Fix
+    is to prefix the 14-bit Ethernet headers with two padding bytes.
+
+  2004-04-23 Leon Woestenberg <leon.woestenberg@gmx.net>
+  * ip_addr.c: Fix in the ip_addr_isbroadcast() check.
+  * etharp.c: Fixed the case where the packet that initiates the ARP request
+    is not queued, and gets lost. Fixed the case where the packets destination
+    address is already known; we now always queue the packet and perform an ARP
+    request.
+  
 (STABLE-0_7_0)
 
   ++ Bug fixes:

doc/rawapi.txt

@@ -272,6 +272,11 @@ level of complexity of UDP, the interface is significantly simpler.
   Sets the remote end of the pcb. This function does not generate any
   network traffic, but only set the remote address of the pcb.
 
+- err_t udp_disconnect(struct udp_pcb *pcb)
+
+  Remove the remote end of the pcb. This function does not generate
+  any network traffic, but only removes the remote address of the pcb.
+
 - err_t udp_send(struct udp_pcb *pcb, struct pbuf *p)
 
   Sends the pbuf p. The pbuf is not deallocated.

doc/savannah.txt

@@ -2,7 +2,7 @@ Daily Use Guide for using Savannah for lwIP
 
 Table of Contents:
 
-1 - Anonymous CVS checkouts and updates (to be elaborated)
+1 - Obtaining lwIP from the CVS repository
 2 - Committers/developers CVS access using SSH (to be written)
 3 - Merging from DEVEL branch to main trunk (stable branch)
 4 - How to release lwIP
@@ -30,6 +30,39 @@ cvs -d:ext:anoncvs@subversions.gnu.org:/cvsroot/lwip checkout -r STABLE-0_7_0 -d
 Or, obtain a development branch (considered unstable!) as follows:
 cvs -d:ext:anoncvs@subversions.gnu.org:/cvsroot/lwip checkout -r DEVEL -d lwip-DEVEL lwip
 
+3 Committers/developers CVS access using SSH
+--------------------------------------------
+
+The Savannah server uses SSH (Secure Shell) protocol 2 authentication and encryption.
+As such, CVS commits to the server occur through a SSH tunnel for project members.
+To create a SSH2 key pair in UNIX-like environments, do this:
+
+ssh-keygen -t dsa
+
+Under Windows, a recommended SSH client is "PuTTY", freely available with good
+documentation and a graphic user interface. Use its key generator.
+
+Now paste the id_dsa.pub contents into your Savannah account public key list. Wait
+a while so that Savannah can update its configuration (This can take minutes).
+
+Try to login using SSH:
+
+ssh -v your_login@subversions.gnu.org
+
+If it tells you:
+
+Authenticating with public key "your_key_name"...
+Server refused to allocate pty
+
+then you could login; Savannah refuses to give you a shell - which is OK, as we
+are allowed to use SSH for CVS only. Now, you should be able to do this:
+
+export CVS_RSH=ssh
+cvs -d:ext:your_login@subversions.gnu.org:/cvsroot/lwip checkout lwip
+
+after which you can edit your local files with bug fixes or new features and
+commit them. Make sure you know what you are doing when using CVS to make
+changes on the repository. If in doubt, ask on the lwip-members mailing list.
 
 3 Merging from DEVEL branch to main trunk (stable)
 --------------------------------------------------
@@ -79,6 +112,8 @@ cvs -d:ext:anoncvs@subversions.gnu.org:/cvsroot/lwip export -r STABLE-0_6_3 -d l
 Archive this directory using tar, gzip'd, bzip2'd and zip'd.
 
 tar czvf lwip-0.6.3.tar.gz lwip-0.6.3
+tar cjvf lwip-0.6.3.tar.bz2 lwip-0.6.3
+zip -r lwip-0.6.3.zip lwip-0.6.3
 
 First, make a local release directory to work in, I use "lwip-releases":
 

src/api/api_lib.c

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2001-2003 Swedish Institute of Computer Science.
+ * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
  * All rights reserved. 
  * 
  * Redistribution and use in source and binary forms, with or without modification, 

src/api/api_msg.c

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2001-2003 Swedish Institute of Computer Science.
+ * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
  * All rights reserved. 
  * 
  * Redistribution and use in source and binary forms, with or without modification, 

src/api/err.c

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2001-2003 Swedish Institute of Computer Science.
+ * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
  * All rights reserved. 
  * 
  * Redistribution and use in source and binary forms, with or without modification, 

src/api/sockets.c

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2001-2003 Swedish Institute of Computer Science.
+ * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without modification,
@@ -37,6 +37,7 @@
 #include "lwip/arch.h"
 #include "lwip/sys.h"
 
+#define LWIP_TIMEVAL_PRIVATE
 #include "lwip/sockets.h"
 
 #define NUM_SOCKETS MEMP_NUM_NETCONN
@@ -993,7 +994,7 @@ int lwip_getsockopt (int s, int level, int optname, void *optval, socklen_t *opt
       /* UNIMPL case SO_SNDBUF: */
       /* UNIMPL case SO_RCVLOWAT: */
       /* UNIMPL case SO_SNDLOWAT: */
-#ifdef SO_REUSE
+#if SO_REUSE
       case SO_REUSEADDR:
       case SO_REUSEPORT:
 #endif /* SO_REUSE */
@@ -1079,7 +1080,7 @@ int lwip_getsockopt (int s, int level, int optname, void *optval, socklen_t *opt
     /* UNIMPL case SO_DONTROUTE: */
     case SO_KEEPALIVE:
     /* UNIMPL case SO_OOBINCLUDE: */
-#ifdef SO_REUSE
+#if SO_REUSE
     case SO_REUSEADDR:
     case SO_REUSEPORT:
 #endif /* SO_REUSE */
@@ -1182,7 +1183,7 @@ int lwip_setsockopt (int s, int level, int optname, const void *optval, socklen_
     /* UNIMPL case SO_SNDBUF: */
     /* UNIMPL case SO_RCVLOWAT: */
     /* UNIMPL case SO_SNDLOWAT: */
-#ifdef SO_REUSE
+#if SO_REUSE
     case SO_REUSEADDR:
     case SO_REUSEPORT:
 #endif /* SO_REUSE */
@@ -1264,7 +1265,7 @@ int lwip_setsockopt (int s, int level, int optname, const void *optval, socklen_
     /* UNIMPL case SO_DONTROUTE: */
     case SO_KEEPALIVE:
     /* UNIMPL case SO_OOBINCLUDE: */
-#ifdef SO_REUSE
+#if SO_REUSE
     case SO_REUSEADDR:
     case SO_REUSEPORT:
 #endif /* SO_REUSE */

src/api/tcpip.c

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2001-2003 Swedish Institute of Computer Science.
+ * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
  * All rights reserved. 
  * 
  * Redistribution and use in source and binary forms, with or without modification, 

src/core/dhcp.c

@@ -6,8 +6,8 @@
 
 /*
  *
- * Copyright (c) 2001-2003 Leon Woestenberg <leon.woestenberg@gmx.net>
- * Copyright (c) 2001-2003 Axon Digital Design B.V., The Netherlands.
+ * Copyright (c) 2001-2004 Leon Woestenberg <leon.woestenberg@gmx.net>
+ * Copyright (c) 2001-2004 Axon Digital Design B.V., The Netherlands.
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without modification,
@@ -70,15 +70,17 @@
 #include "lwip/stats.h"
 #include "lwip/mem.h"
 #include "lwip/udp.h"
+#include "lwip/ip_addr.h"
 #include "lwip/netif.h"
 #include "lwip/inet.h"
-#include "lwip/ip_addr.h"
 #include "netif/etharp.h"
 
 #include "lwip/sys.h"
 #include "lwip/opt.h"
 #include "lwip/dhcp.h"
 
+#if LWIP_DHCP /* don't build if not configured for use in lwipopt.h */
+
 /** global transaction identifier, must be
  *  unique for each DHCP request. */
 static u32_t xid = 0xABCD0000;
@@ -404,7 +406,6 @@ static void dhcp_t2_timeout(struct netif *netif)
 }
 
 /**
- * Extract options from the server ACK message.
  *
  * @param netif the netif under DHCP control
  */
@@ -706,15 +707,9 @@ static err_t dhcp_discover(struct netif *netif)
     /* set receive callback function with netif as user data */
     udp_recv(dhcp->pcb, dhcp_recv, netif);
     udp_bind(dhcp->pcb, IP_ADDR_ANY, DHCP_CLIENT_PORT);
-    udp_connect(dhcp->pcb, IP_ADDR_BROADCAST, DHCP_SERVER_PORT);
-
-    LWIP_DEBUGF(DHCP_DEBUG | DBG_TRACE, ("dhcp_discover: send()ing\n"));
-
-    udp_send(dhcp->pcb, dhcp->p_out);
-    LWIP_DEBUGF(DHCP_DEBUG | DBG_TRACE, ("dhcp_discover: bind()ing\n"));
-    udp_bind(dhcp->pcb, IP_ADDR_ANY, DHCP_CLIENT_PORT);
-    LWIP_DEBUGF(DHCP_DEBUG | DBG_TRACE, ("dhcp_discover: connect()ing\n"));
     udp_connect(dhcp->pcb, IP_ADDR_ANY, DHCP_SERVER_PORT);
+    LWIP_DEBUGF(DHCP_DEBUG | DBG_TRACE, ("dhcp_discover: sendto(DISCOVER, IP_ADDR_BROADCAST, DHCP_SERVER_PORT)\n"));
+    udp_sendto(dhcp->pcb, dhcp->p_out, IP_ADDR_BROADCAST, DHCP_SERVER_PORT);
     LWIP_DEBUGF(DHCP_DEBUG | DBG_TRACE, ("dhcp_discover: deleting()ing\n"));
     dhcp_delete_request(netif);
     LWIP_DEBUGF(DHCP_DEBUG | DBG_TRACE | DBG_STATE, ("dhcp_discover: SELECTING\n"));
@@ -881,10 +876,11 @@ static err_t dhcp_rebind(struct netif *netif)
 
     pbuf_realloc(dhcp->p_out, sizeof(struct dhcp_msg) - DHCP_OPTIONS_LEN + dhcp->options_out_len);
 
+    /* set remote IP association to any DHCP server */
     udp_bind(dhcp->pcb, IP_ADDR_ANY, DHCP_CLIENT_PORT);
-    udp_connect(dhcp->pcb, IP_ADDR_BROADCAST, DHCP_SERVER_PORT);
-    udp_send(dhcp->pcb, dhcp->p_out);
     udp_connect(dhcp->pcb, IP_ADDR_ANY, DHCP_SERVER_PORT);
+    /* broadcast to server */
+    udp_sendto(dhcp->pcb, dhcp->p_out, IP_ADDR_BROADCAST, DHCP_SERVER_PORT);
     dhcp_delete_request(netif);
     LWIP_DEBUGF(DHCP_DEBUG | DBG_TRACE | DBG_STATE, ("dhcp_rebind: REBINDING\n"));
   } else {
@@ -1243,10 +1239,10 @@ static err_t dhcp_create_request(struct netif *netif)
   dhcp->msg_out->xid = htonl(dhcp->xid);
   dhcp->msg_out->secs = 0;
   dhcp->msg_out->flags = 0;
-  dhcp->msg_out->ciaddr = netif->ip_addr.addr;
-  dhcp->msg_out->yiaddr = 0;
-  dhcp->msg_out->siaddr = 0;
-  dhcp->msg_out->giaddr = 0;
+  dhcp->msg_out->ciaddr.addr = netif->ip_addr.addr;
+  dhcp->msg_out->yiaddr.addr = 0;
+  dhcp->msg_out->siaddr.addr = 0;
+  dhcp->msg_out->giaddr.addr = 0;
   for (i = 0; i < DHCP_CHADDR_LEN; i++) {
     /* copy netif hardware address, pad with zeroes */
     dhcp->msg_out->chaddr[i] = (i < netif->hwaddr_len) ? netif->hwaddr[i] : 0/* pad byte*/;
@@ -1418,3 +1414,5 @@ static u32_t dhcp_get_option_long(u8_t *ptr)
   LWIP_DEBUGF(DHCP_DEBUG, ("option long value=%lu\n", value));
   return value;
 }
+
+#endif /* LWIP_DHCP */

src/core/inet.c

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2001-2003 Swedish Institute of Computer Science.
+ * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without modification,
@@ -300,9 +300,10 @@ inet_chksum_pbuf(struct pbuf *p)
 /* Convert numeric IP address into decimal dotted ASCII representation.
  * returns ptr to static buffer; not reentrant!
  */
-u8_t *inet_ntoa(u32_t addr)
+char *inet_ntoa(struct in_addr addr)
 {
   static u8_t str[16];
+  u32_t s_addr = addr.s_addr;
   u8_t inv[3];
   u8_t *rp;
   u8_t *ap;
@@ -311,7 +312,7 @@ u8_t *inet_ntoa(u32_t addr)
   u8_t i;
 
   rp = str;
-  ap = (u8_t *)&addr;
+  ap = (char *)&s_addr;
   for(n = 0; n < 4; n++) {
     i = 0;
     do {

src/core/inet6.c

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2001-2003 Swedish Institute of Computer Science.
+ * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
  * All rights reserved. 
  * 
  * Redistribution and use in source and binary forms, with or without modification, 

src/core/ipv4/icmp.c

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2001-2003 Swedish Institute of Computer Science.
+ * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without modification,
@@ -72,8 +72,8 @@ icmp_input(struct pbuf *p, struct netif *inp)
   code = *(((u8_t *)p->payload)+1);
   switch (type) {
   case ICMP_ECHO:
-    if (ip_addr_isbroadcast(&iphdr->dest, &inp->netmask) ||
-       ip_addr_ismulticast(&iphdr->dest)) {
+    /* broadcast or multicast destination address? */
+    if (ip_addr_isbroadcast(&iphdr->dest, inp) || ip_addr_ismulticast(&iphdr->dest)) {
       LWIP_DEBUGF(ICMP_DEBUG, ("Smurf.\n"));
       ICMP_STATS_INC(icmp.err);
       pbuf_free(p);

src/core/ipv4/ip.c

@@ -1,5 +1,12 @@
+/* @file
+ *
+ * This is the IP layer implementation for incoming and outgoing IP traffic.
+ * 
+ * @see ip_frag.c
+ *
+ */
 /*
- * Copyright (c) 2001-2003 Swedish Institute of Computer Science.
+ * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without modification,
@@ -30,18 +37,8 @@
  *
  */
 
-
-
-/* ip.c
- *
- * This is the code for the IP layer.
- *
- */
-
-
 #include "lwip/opt.h"
 
-
 #include "lwip/def.h"
 #include "lwip/mem.h"
 #include "lwip/ip.h"
@@ -63,69 +60,17 @@
 #endif /* LWIP_DHCP */
 
 
-/* ip_init:
- *
+/**
  * Initializes the IP layer.
  */
 
 void
 ip_init(void)
 {
+  /* no initializations as of yet */
 }
 
-/* ip_lookup:
- *
- * An experimental feature that will be changed in future versions. Do
- * not depend on it yet...
- */
-
-#ifdef LWIP_DEBUG
-u8_t
-ip_lookup(void *header, struct netif *inp)
-{
-  struct ip_hdr *iphdr;
-
-  iphdr = header;
-
-  /* not IP v4? */
-  if (IPH_V(iphdr) != 4) {
-    return 0;
-  }
-
-  /* Immediately accept/decline packets that are fragments or has
-     options. */
-#if IP_REASSEMBLY == 0
-  /*  if ((IPH_OFFSET(iphdr) & htons(IP_OFFMASK | IP_MF)) != 0) {
-    return 0;
-    }*/
-#endif /* IP_REASSEMBLY == 0 */
-
-#if IP_OPTIONS == 0
-  if (IPH_HL(iphdr) != 5) {
-    return 0;
-  }
-#endif /* IP_OPTIONS == 0 */
-
-  switch (IPH_PROTO(iphdr)) {
-#if LWIP_UDP
-  case IP_PROTO_UDP:
-  case IP_PROTO_UDPLITE:
-    return udp_lookup(iphdr, inp);
-#endif /* LWIP_UDP */
-#if LWIP_TCP
-  case IP_PROTO_TCP:
-    return 1;
-#endif /* LWIP_TCP */
-  case IP_PROTO_ICMP:
-    return 1;
-  default:
-    return 0;
-  }
-}
-#endif /* LWIP_DEBUG */
-
-/* ip_route:
- *
+/**
  * Finds the appropriate network interface for a given IP address. It
  * searches the list of network interfaces linearly. A match is found
  * if the masked IP address of the network interface equals the masked
@@ -150,8 +95,7 @@ ip_route(struct ip_addr *dest)
 }
 #if IP_FORWARD
 
-/* ip_forward:
- *
+/**
  * Forwards an IP packet. It finds an appropriate route for the
  * packet, decrements the TTL value of the packet, adjusts the
  * checksum and outputs the packet on the appropriate interface.
@@ -172,7 +116,7 @@ ip_forward(struct pbuf *p, struct ip_hdr *iphdr, struct netif *inp)
     return;
   }
   /* Do not forward packets onto the same network interface on which
-     they arrived. */
+   * they arrived. */
   if (netif == inp) {
     LWIP_DEBUGF(IP_DEBUG, ("ip_forward: not bouncing packets back on incoming interface.\n"));
     snmp_inc_ipnoroutes();
@@ -211,8 +155,7 @@ ip_forward(struct pbuf *p, struct ip_hdr *iphdr, struct netif *inp)
 }
 #endif /* IP_FORWARD */
 
-/* ip_input:
- *
+/**
  * This function is called by the network interface device driver when
  * an IP packet is received. The function does the basic checks of the
  * IP header such as packet size being at least larger than the header
@@ -260,6 +203,7 @@ ip_input(struct pbuf *p, struct netif *inp) {
   }
 
   /* verify checksum */
+#if CHECKSUM_CHECK_IP
   if (inet_chksum(iphdr, iphdrlen) != 0) {
 
     LWIP_DEBUGF(IP_DEBUG | 2, ("Checksum (0x%x) failed, IP packet dropped.\n", inet_chksum(iphdr, iphdrlen)));
@@ -270,9 +214,10 @@ ip_input(struct pbuf *p, struct netif *inp) {
     snmp_inc_ipindiscards();
     return ERR_OK;
   }
+#endif
 
   /* Trim pbuf. This should have been done at the netif layer,
-     but we'll do it anyway just to be sure that its done. */
+   * but we'll do it anyway just to be sure that its done. */
   pbuf_realloc(p, ntohs(IPH_LEN(iphdr)));
 
   /* is this packet for us? */
@@ -290,7 +235,7 @@ ip_input(struct pbuf *p, struct netif *inp) {
       /* unicast to this interface address? */
       if (ip_addr_cmp(&(iphdr->dest), &(netif->ip_addr)) ||
         /* or broadcast matching this interface network address? */
-        (ip_addr_isbroadcast(&(iphdr->dest), &(netif->netmask)) &&
+        (ip_addr_isbroadcast(&(iphdr->dest), netif) &&
          ip_addr_maskcmp(&(iphdr->dest), &(netif->ip_addr), &(netif->netmask))) ||
          /* or restricted broadcast? */
          ip_addr_cmp(&(iphdr->dest), IP_ADDR_BROADCAST)) {
@@ -303,8 +248,8 @@ ip_input(struct pbuf *p, struct netif *inp) {
   }
 #if LWIP_DHCP
   /* Pass DHCP messages regardless of destination address. DHCP traffic is addressed
-     using link layer addressing (such as Ethernet MAC) so we must not filter on IP.
-     According to RFC 1542 section 3.1.1, referred by RFC 2131). */
+   * using link layer addressing (such as Ethernet MAC) so we must not filter on IP.
+   * According to RFC 1542 section 3.1.1, referred by RFC 2131). */
   if (netif == NULL) {
     /* remote port is DHCP server? */
     if (IPH_PROTO(iphdr) == IP_PROTO_UDP) {
@@ -323,7 +268,7 @@ ip_input(struct pbuf *p, struct netif *inp) {
     LWIP_DEBUGF(IP_DEBUG | DBG_TRACE | 1, ("ip_input: packet not for us.\n"));
 #if IP_FORWARD
     /* non-broadcast packet? */
-    if (!ip_addr_isbroadcast(&(iphdr->dest), &(inp->netmask))) {
+    if (!ip_addr_isbroadcast(&(iphdr->dest), inp)) {
       /* try to forward IP packet on (other) interfaces */
       ip_forward(p, iphdr, inp);
     }
@@ -335,19 +280,19 @@ ip_input(struct pbuf *p, struct netif *inp) {
     pbuf_free(p);
     return ERR_OK;
   }
-
-#if IP_REASSEMBLY
+  /* packet consists of multiple fragments? */
   if ((IPH_OFFSET(iphdr) & htons(IP_OFFMASK | IP_MF)) != 0) {
+#if IP_REASSEMBLY /* packet fragment reassembly code present? */
     LWIP_DEBUGF(IP_DEBUG, ("IP packet is a fragment (id=0x%04x tot_len=%u len=%u MF=%u offset=%u), calling ip_reass()\n",
       ntohs(IPH_ID(iphdr)), p->tot_len, ntohs(IPH_LEN(iphdr)), !!(IPH_OFFSET(iphdr) & htons(IP_MF)), (ntohs(IPH_OFFSET(iphdr)) & IP_OFFMASK)*8));
+    /* reassemble the packet*/
     p = ip_reass(p);
+    /* packet not fully reassembled yet? */
     if (p == NULL) {
       return ERR_OK;
     }
     iphdr = p->payload;
-  }
-#else /* IP_REASSEMBLY */
-  if ((IPH_OFFSET(iphdr) & htons(IP_OFFMASK | IP_MF)) != 0) {
+#else /* IP_REASSEMBLY == 0, no packet fragment reassembly code present */
     pbuf_free(p);
     LWIP_DEBUGF(IP_DEBUG | 2, ("IP packet dropped since it was fragmented (0x%x) (while IP_REASSEMBLY == 0).\n",
                   ntohs(IPH_OFFSET(iphdr))));
@@ -355,10 +300,10 @@ ip_input(struct pbuf *p, struct netif *inp) {
     IP_STATS_INC(ip.drop);
     snmp_inc_ipunknownprotos();
     return ERR_OK;
-  }
 #endif /* IP_REASSEMBLY */
+  }
 
-#if IP_OPTIONS == 0
+#if IP_OPTIONS == 0 /* no support for IP options in the IP header? */
   if (iphdrlen > IP_HLEN) {
     LWIP_DEBUGF(IP_DEBUG | 2, ("IP packet dropped since there were IP options (while IP_OPTIONS == 0).\n"));
     pbuf_free(p);
@@ -375,7 +320,8 @@ ip_input(struct pbuf *p, struct netif *inp) {
   LWIP_DEBUGF(IP_DEBUG, ("ip_input: p->len %d p->tot_len %d\n", p->len, p->tot_len));
 
 #if LWIP_RAW
-  if (!raw_input(p, inp)) {
+  /* raw input did not eat the packet? */
+  if (raw_input(p, inp) == 0) {
 #endif /* LWIP_RAW */
 
   switch (IPH_PROTO(iphdr)) {
@@ -398,8 +344,8 @@ ip_input(struct pbuf *p, struct netif *inp) {
     break;
   default:
     /* send ICMP destination protocol unreachable unless is was a broadcast */
-    if (!ip_addr_isbroadcast(&(iphdr->dest), &(inp->netmask)) &&
-       !ip_addr_ismulticast(&(iphdr->dest))) {
+    if (!ip_addr_isbroadcast(&(iphdr->dest), inp) &&
+        !ip_addr_ismulticast(&(iphdr->dest))) {
       p->payload = iphdr;
       icmp_dest_unreach(p, ICMP_DUR_PROTO);
     }
@@ -410,7 +356,6 @@ ip_input(struct pbuf *p, struct netif *inp) {
     IP_STATS_INC(ip.proterr);
     IP_STATS_INC(ip.drop);
     snmp_inc_ipunknownprotos();
-
   }
 #if LWIP_RAW
   } /* LWIP_RAW */
@@ -418,9 +363,7 @@ ip_input(struct pbuf *p, struct netif *inp) {
   return ERR_OK;
 }
 
-
-/* ip_output_if:
- *
+/**
  * Sends an IP packet on a network interface. This function constructs
  * the IP header and calculates the IP header checksum. If the source
  * IP address is NULL, the IP address of the outgoing network
@@ -466,7 +409,9 @@ ip_output_if(struct pbuf *p, struct ip_addr *src, struct ip_addr *dest,
     }
 
     IPH_CHKSUM_SET(iphdr, 0);
+#if CHECKSUM_GEN_IP
     IPH_CHKSUM_SET(iphdr, inet_chksum(iphdr, IP_HLEN));
+#endif
   } else {
     iphdr = p->payload;
     dest = &(iphdr->dest);
@@ -488,8 +433,7 @@ ip_output_if(struct pbuf *p, struct ip_addr *src, struct ip_addr *dest,
   return netif->output(netif, p, dest);
 }
 
-/* ip_output:
- *
+/**
  * Simple interface to ip_output_if. It finds the outgoing network
  * interface and calls upon ip_output_if to do the actual work.
  */

src/core/ipv4/ip_addr.c

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2001-2003 Swedish Institute of Computer Science.
+ * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
  * All rights reserved. 
  * 
  * Redistribution and use in source and binary forms, with or without modification, 
@@ -32,7 +32,38 @@
 
 #include "lwip/ip_addr.h"
 #include "lwip/inet.h"
+#include "lwip/netif.h"
 
 /* used by IP_ADDR_ANY and IP_ADDR_BROADCAST in ip_addr.h */
 const struct ip_addr ip_addr_any = { 0x00000000UL };
 const struct ip_addr ip_addr_broadcast = { 0xffffffffUL };
+
+/* Determine if an address is a broadcast address on a network interface 
+ * 
+ * @param addr address to be checked
+ * @param netif the network interface against which the address is checked
+ * @return returns non-zero if the address is a broadcast address
+ *
+ */
+
+u8_t ip_addr_isbroadcast(struct ip_addr *addr, struct netif *netif)
+{
+  /* all ones (broadcast) or all zeroes (old skool broadcast) */
+  if ((addr->addr == ip_addr_broadcast.addr) ||
+      (addr->addr == ip_addr_any.addr))
+    return 1;
+  /* no broadcast support on this network interface? */
+  else if ((netif->flags & NETIF_FLAG_BROADCAST) == 0)
+    /* the given address cannot be a broadcast address
+     * nor can we check against any broadcast addresses */
+    return 0;
+  /* address matches network interface address exactly? => no broadcast */
+  else if (addr->addr == netif->ip_addr.addr)
+    return 0;
+  /* host identifier bits are all ones? => network broadcast address */
+  else if ((addr->addr & ~netif->netmask.addr) ==
+           (ip_addr_broadcast.addr & ~netif->netmask.addr))
+    return 1;
+  else
+    return 0;
+}

src/core/ipv4/ip_frag.c

@@ -1,5 +1,11 @@
+/* @file
+ * 
+ * This is the IP packet segmentation and reassembly implementation.
+ *
+ */
+
 /*
- * Copyright (c) 2001-2003 Swedish Institute of Computer Science.
+ * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
  * All rights reserved. 
  * 
  * Redistribution and use in source and binary forms, with or without modification, 
@@ -31,14 +37,6 @@
  * 
  */
 
-
-/* ip_frag.c
- *
- * This is the code for IP segmentation and reassembly
- *
- */
-
-
 #include "lwip/opt.h"
 #include "lwip/sys.h"
 #include "lwip/ip.h"
@@ -294,7 +292,6 @@ ip_frag(struct pbuf *p, struct netif *netif, struct ip_addr *dest)
   rambuf->tot_len = rambuf->len = mtu;
   rambuf->payload = MEM_ALIGN((void *)buf);
 
-
   /* Copy the IP header in it */
   iphdr = rambuf->payload;
   memcpy(iphdr, p->payload, IP_HLEN);

src/core/ipv6/icmp6.c

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2001-2003 Swedish Institute of Computer Science.
+ * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without modification,

src/core/ipv6/ip6.c

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2001-2003 Swedish Institute of Computer Science.
+ * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without modification,

src/core/ipv6/ip6_addr.c

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2001-2003 Swedish Institute of Computer Science.
+ * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
  * All rights reserved. 
  * 
  * Redistribution and use in source and binary forms, with or without modification, 

src/core/mem.c

@@ -5,7 +5,7 @@
  */
 
 /* 
- * Copyright (c) 2001-2003 Swedish Institute of Computer Science.
+ * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
  * All rights reserved. 
  * 
  * Redistribution and use in source and binary forms, with or without modification, 
@@ -54,6 +54,8 @@ struct mem {
   u16_t used;
 #elif MEM_ALIGNMENT == 4
   u32_t used;
+#elif MEM_ALIGNMENT == 8
+  u64_t used;
 #else
 #error "unhandled MEM_ALIGNMENT size"
 #endif /* MEM_ALIGNMENT */
@@ -292,7 +294,7 @@ mem_malloc(mem_size_t size)
       }
       sys_sem_signal(mem_sem);
       LWIP_ASSERT("mem_malloc: allocated memory not above ram_end.",
-       (u32_t)mem + SIZEOF_STRUCT_MEM + size <= (u32_t)ram_end);
+       (mem_ptr_t)mem + SIZEOF_STRUCT_MEM + size <= (mem_ptr_t)ram_end);
       LWIP_ASSERT("mem_malloc: allocated memory properly aligned.",
        (unsigned long)((u8_t *)mem + SIZEOF_STRUCT_MEM) % MEM_ALIGNMENT == 0);
       return (u8_t *)mem + SIZEOF_STRUCT_MEM;

src/core/memp.c

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2001-2003 Swedish Institute of Computer Science.
+ * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
  * All rights reserved. 
  * 
  * Redistribution and use in source and binary forms, with or without modification, 
@@ -120,7 +120,7 @@ static u8_t memp_memory[(MEMP_NUM_PBUF *
 static sys_sem_t mutex;
 #endif
 
-#ifndef LWIP_NOASSERT
+#if MEMP_SANITY_CHECK
 static int
 memp_sanity(void)
 {
@@ -140,7 +140,7 @@ memp_sanity(void)
   }
   return 1;
 }
-#endif /* LWIP_DEBUG */
+#endif /* MEMP_SANITY_CHECK*/
 
 void
 memp_init(void)
@@ -217,7 +217,7 @@ memp_malloc(memp_t type)
     sys_sem_signal(mutex);
 #endif /* SYS_LIGHTWEIGHT_PROT */  
     LWIP_ASSERT("memp_malloc: memp properly aligned",
-     ((u32_t)MEM_ALIGN((u8_t *)memp + sizeof(struct memp)) % MEM_ALIGNMENT) == 0);
+     ((mem_ptr_t)MEM_ALIGN((u8_t *)memp + sizeof(struct memp)) % MEM_ALIGNMENT) == 0);
 
     mem = MEM_ALIGN((u8_t *)memp + sizeof(struct memp));
     return mem;
@@ -261,7 +261,9 @@ memp_free(memp_t type, void *mem)
   memp->next = memp_tab[type]; 
   memp_tab[type] = memp;
 
+#if MEMP_SANITY_CHECK
   LWIP_ASSERT("memp sanity", memp_sanity());
+#endif  
 
 #if SYS_LIGHTWEIGHT_PROT
   SYS_ARCH_UNPROTECT(old_level);

src/core/netif.c

@@ -5,7 +5,7 @@
  */
 
 /*
- * Copyright (c) 2001-2003 Swedish Institute of Computer Science.
+ * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without modification,
@@ -39,23 +39,24 @@
 #include "lwip/opt.h"
 
 #include "lwip/def.h"
-#include "lwip/netif.h"
 #include "lwip/ip_addr.h"
+#include "lwip/netif.h"
 #include "lwip/tcp.h"
 
-
 struct netif *netif_list = NULL;
 struct netif *netif_default = NULL;
 
 /**
  * Add a network interface to the list of lwIP netifs.
  *
+ * @param netif a pre-allocated netif structure
  * @param ipaddr IP address for the new netif
  * @param netmask network mask for the new netif
  * @param gw default gateway IP address for the new netif
  * @param state opaque data passed to the new netif
  * @param init callback function that initializes the interface
- * @param input callback function that...
+ * @param input callback function that is called to pass
+ * ingress packets up in the protocol layer stack.
  *
  * @return netif, or NULL if failed.
  */
@@ -171,7 +172,7 @@ netif_set_ipaddr(struct netif *netif, struct ip_addr *ipaddr)
   /* address is actually being changed? */
   if ((ip_addr_cmp(ipaddr, &(netif->ip_addr))) == 0)
   {
-    extern struct tcp_pcb *tcp_active_pcbs;
+    /* extern struct tcp_pcb *tcp_active_pcbs; defined by tcp.h */
     LWIP_DEBUGF(NETIF_DEBUG | 1, ("netif_set_ipaddr: netif address being changed\n"));
     pcb = tcp_active_pcbs;
     while (pcb != NULL) {
@@ -186,7 +187,7 @@ netif_set_ipaddr(struct netif *netif, struct ip_addr *ipaddr)
         pcb = pcb->next;
       }
     }
-    for (lpcb = tcp_listen_pcbs; lpcb != NULL; lpcb = lpcb->next) {
+    for (lpcb = tcp_listen_pcbs.listen_pcbs; lpcb != NULL; lpcb = lpcb->next) {
       /* PCB bound to current local interface address? */
       if (ip_addr_cmp(&(lpcb->local_ip), &(netif->ip_addr))) {
         /* The PCB is listening to the old ipaddr and
@@ -197,6 +198,14 @@ netif_set_ipaddr(struct netif *netif, struct ip_addr *ipaddr)
   }
 #endif
   ip_addr_set(&(netif->ip_addr), ipaddr);
+#if 0 /* only allowed for Ethernet interfaces TODO: how can we check? */
+  /** For Ethernet network interfaces, we would like to send a
+   *  "gratuitous ARP"; this is an ARP packet sent by a node in order
+   *  to spontaneously cause other nodes to update an entry in their
+   *  ARP cache. From RFC 3220 "IP Mobility Support for IPv4" section 4.6.
+   */ 
+  etharp_query(netif, ipaddr, NULL);
+#endif
   LWIP_DEBUGF(NETIF_DEBUG | DBG_TRACE | DBG_STATE | 3, ("netif: IP address of interface %c%c set to %u.%u.%u.%u\n",
     netif->name[0], netif->name[1],
     (unsigned int)(ntohl(netif->ip_addr.addr) >> 24 & 0xff),

src/core/pbuf.c

@@ -11,12 +11,15 @@
  * list. This is called a "pbuf chain".
  *
  * Multiple packets may be queued, also using this singly linked list.
- * This is called a "packet queue". So, a packet queue consists of one
- * or more pbuf chains, each of which consist of one or more pbufs.
+ * This is called a "packet queue".
+ * 
+ * So, a packet queue consists of one or more pbuf chains, each of
+ * which consist of one or more pbufs. Currently, queues are only
+ * supported in a limited section of lwIP, this is the etharp queueing
+ * code. Outside of this section no packet queues are supported yet.
+ * 
  * The differences between a pbuf chain and a packet queue are very
- * subtle. Currently, queues are only supported in a limited section
- * of lwIP, this is the etharp queueing code. Outside of this section
- * no packet queues are supported as of yet.
+ * precise but subtle. 
  *
  * The last pbuf of a packet has a ->tot_len field that equals the
  * ->len field. It can be found by traversing the list. If the last
@@ -28,7 +31,7 @@
  */
 
 /*
- * Copyright (c) 2001-2003 Swedish Institute of Computer Science.
+ * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without modification,
@@ -100,7 +103,7 @@ pbuf_init(void)
   u16_t i;
 
   pbuf_pool = (struct pbuf *)&pbuf_pool_memory[0];
-  LWIP_ASSERT("pbuf_init: pool aligned", (long)pbuf_pool % MEM_ALIGNMENT == 0);
+  LWIP_ASSERT("pbuf_init: pool aligned", (mem_ptr_t)pbuf_pool % MEM_ALIGNMENT == 0);
 
 #if PBUF_STATS
   lwip_stats.pbuf.avail = PBUF_POOL_SIZE;
@@ -254,7 +257,7 @@ pbuf_alloc(pbuf_layer l, u16_t length, pbuf_flag flag)
     /* make the payload pointer point 'offset' bytes into pbuf data memory */
     p->payload = MEM_ALIGN((void *)((u8_t *)p + (sizeof(struct pbuf) + offset)));
     LWIP_ASSERT("pbuf_alloc: pbuf p->payload properly aligned",
-            ((u32_t)p->payload % MEM_ALIGNMENT) == 0);
+            ((mem_ptr_t)p->payload % MEM_ALIGNMENT) == 0);
     /* the total length of the pbuf chain is the requested size */
     p->tot_len = length;
     /* set the length of the first pbuf in the chain */
@@ -290,7 +293,7 @@ pbuf_alloc(pbuf_layer l, u16_t length, pbuf_flag flag)
       q->len = rem_len > PBUF_POOL_BUFSIZE? PBUF_POOL_BUFSIZE: rem_len;
       q->payload = (void *)((u8_t *)q + sizeof(struct pbuf));
       LWIP_ASSERT("pbuf_alloc: pbuf q->payload properly aligned",
-              ((u32_t)q->payload % MEM_ALIGNMENT) == 0);
+              ((mem_ptr_t)q->payload % MEM_ALIGNMENT) == 0);
       q->ref = 1;
       /* calculate remaining length to be allocated */
       rem_len -= q->len;
@@ -314,7 +317,7 @@ pbuf_alloc(pbuf_layer l, u16_t length, pbuf_flag flag)
     p->flags = PBUF_FLAG_RAM;
 
     LWIP_ASSERT("pbuf_alloc: pbuf->payload properly aligned",
-           ((u32_t)p->payload % MEM_ALIGNMENT) == 0);
+           ((mem_ptr_t)p->payload % MEM_ALIGNMENT) == 0);
     break;
   /* pbuf references existing (static constant) ROM payload? */
   case PBUF_ROM:
@@ -450,32 +453,35 @@ pbuf_realloc(struct pbuf *p, u16_t new_len)
  *
  * The ->payload, ->tot_len and ->len fields are adjusted.
  *
- * @param hdr_size Number of bytes to increment header size which
+ * @param hdr_size_inc Number of bytes to increment header size which
  * increases the size of the pbuf. New space is on the front.
  * (Using a negative value decreases the header size.)
+ * If hdr_size_inc is 0, this function does nothing and returns succesful.
  *
  * PBUF_ROM and PBUF_REF type buffers cannot have their sizes increased, so
  * the call will fail. A check is made that the increase in header size does
  * not move the payload pointer in front of the start of the buffer.
- * @return 1 on failure, 0 on success.
+ * @return non-zero on failure, zero on success.
  *
- * @note May not be called on a packet queue.
  */
 u8_t
-pbuf_header(struct pbuf *p, s16_t header_size)
+pbuf_header(struct pbuf *p, s16_t header_size_increment)
 {
   void *payload;
 
+  LWIP_ASSERT("p != NULL", p != NULL);
+  if ((header_size_increment == 0) || (p == NULL)) return 0;
+ 
   /* remember current payload pointer */
   payload = p->payload;
 
   /* pbuf types containing payloads? */
   if (p->flags == PBUF_FLAG_RAM || p->flags == PBUF_FLAG_POOL) {
     /* set new payload pointer */
-    p->payload = (u8_t *)p->payload - header_size;
+    p->payload = (u8_t *)p->payload - header_size_increment;
     /* boundary check fails? */
     if ((u8_t *)p->payload < (u8_t *)p + sizeof(struct pbuf)) {
-      LWIP_DEBUGF( PBUF_DEBUG | 2, ("pbuf_header: failed as %p < %p\n",
+      LWIP_DEBUGF( PBUF_DEBUG | 2, ("pbuf_header: failed as %p < %p (not enough space for new header size)\n",
         (u8_t *)p->payload,
         (u8_t *)p + sizeof(struct pbuf)) );\
       /* restore old payload pointer */
@@ -483,35 +489,37 @@ pbuf_header(struct pbuf *p, s16_t header_size)
       /* bail out unsuccesfully */
       return 1;
     }
-  /* pbuf types refering to payloads? */
+  /* pbuf types refering to external payloads? */
   } else if (p->flags == PBUF_FLAG_REF || p->flags == PBUF_FLAG_ROM) {
     /* hide a header in the payload? */
-    if ((header_size < 0) && (header_size - p->len <= 0)) {
+    if ((header_size_increment < 0) && (header_size_increment - p->len <= 0)) {
       /* increase payload pointer */
-      p->payload = (u8_t *)p->payload - header_size;
+      p->payload = (u8_t *)p->payload - header_size_increment;
     } else {
       /* cannot expand payload to front (yet!)
        * bail out unsuccesfully */
       return 1;
     }
   }
-  LWIP_DEBUGF( PBUF_DEBUG, ("pbuf_header: old %p new %p (%d)\n", (void *)payload, (void *)p->payload, header_size) );
   /* modify pbuf length fields */
-  p->len += header_size;
-  p->tot_len += header_size;
+  p->len += header_size_increment;
+  p->tot_len += header_size_increment;
+
+  LWIP_DEBUGF( PBUF_DEBUG, ("pbuf_header: old %p new %p (%d)\n",
+    (void *)payload, (void *)p->payload, header_size_increment));
 
   return 0;
 }
 
 /**
- * Dereference a pbuf (chain) and deallocate any no-longer-used
- * pbufs at the head of this chain.
+ * Dereference a pbuf chain or queue and deallocate any no-longer-used
+ * pbufs at the head of this chain or queue.
  *
- * Decrements the pbuf reference count. If it reaches
- * zero, the pbuf is deallocated.
+ * Decrements the pbuf reference count. If it reaches zero, the pbuf is
+ * deallocated.
  *
  * For a pbuf chain, this is repeated for each pbuf in the chain,
- * up to a pbuf which has a non-zero reference count after
+ * up to the first pbuf which has a non-zero reference count after
  * decrementing. (This might de-allocate the whole chain.)
  *
  * @param pbuf The pbuf (chain) to be dereferenced.
@@ -519,7 +527,7 @@ pbuf_header(struct pbuf *p, s16_t header_size)
  * @return the number of pbufs that were de-allocated
  * from the head of the chain.
  *
- * @note MUST NOT be called on a packet queue.
+ * @note MUST NOT be called on a packet queue (Not verified to work yet).
  * @note the reference counter of a pbuf equals the number of pointers
  * that refer to the pbuf (or into the pbuf).
  *
@@ -542,6 +550,8 @@ pbuf_free(struct pbuf *p)
   u8_t count;
   SYS_ARCH_DECL_PROTECT(old_level);
 
+  LWIP_ASSERT("p != NULL", p != NULL);
+  /* if assertions are disabled, proceed with debug output */
   if (p == NULL) {
     LWIP_DEBUGF(PBUF_DEBUG | DBG_TRACE | 2, ("pbuf_free(p == NULL) was called.\n"));
     return 0;
@@ -708,11 +718,13 @@ pbuf_chain(struct pbuf *h, struct pbuf *t)
 void
 pbuf_queue(struct pbuf *p, struct pbuf *n)
 {
+#if PBUF_DEBUG /* remember head of queue */
+  struct pbuf *q = p;
+#endif
+  /* programmer stupidity checks */
   LWIP_ASSERT("p != NULL", p != NULL);
   LWIP_ASSERT("n != NULL", n != NULL);
-
-  if ((p == NULL) || (n == NULL))
-    return;
+  if ((p == NULL) || (n == NULL)) return;
 
   /* iterate through all packets on queue */
   while (p->next != NULL) {
@@ -720,25 +732,34 @@ pbuf_queue(struct pbuf *p, struct pbuf *n)
 #if PBUF_DEBUG
     /* iterate through all pbufs in packet */
     while (p->tot_len != p->len) {
+      /* make sure invariant condition holds */
+      LWIP_ASSERT("p->len < p->tot_len", p->len < p->tot_len);
       /* make sure each packet is complete */
       LWIP_ASSERT("p->next != NULL", p->next != NULL);
       p = p->next;
+      /* { p->tot_len == p->len => p is last pbuf of a packet } */
     }
 #endif
-    /* now p->tot_len == p->len */
+    /* { p->tot_len == p->len and p is last pbuf of a packet } */
     /* proceed to next packet on queue */
-    p = p->next;
+    if (p->next != NULL) p = p->next;
   }
+  /* { p->tot_len == p->len and p->next == NULL } ==>
+   * { p is last pbuf of last packet on queue } */
   /* chain last pbuf of queue with n */
   p->next = n;
-  /* n is now referenced to one more time */
+  /* n is now referenced to by the (packet p in the) queue */
   pbuf_ref(n);
-  LWIP_DEBUGF(PBUF_DEBUG | DBG_FRESH | 2, ("pbuf_queue: referencing queued packet %p\n", (void *)n));
+  LWIP_DEBUGF(PBUF_DEBUG | DBG_FRESH | 2,
+    ("pbuf_queue: newly queued packet %p sits after packet %p in queue %p\n",
+    (void *)n, (void *)p, (void *)q));
 }
 
 /**
  * Remove a packet from the head of a queue.
  *
+ * The caller MUST reference the remainder of the queue (as returned).
+ * 
  * @param p pointer to first packet on the queue which will be dequeued.
  * @return first packet on the remaining queue (NULL if no further packets).
  *
@@ -751,17 +772,25 @@ pbuf_dequeue(struct pbuf *p)
 
   /* iterate through all pbufs in packet */
   while (p->tot_len != p->len) {
+    /* make sure invariant condition holds */
+    LWIP_ASSERT("p->len < p->tot_len", p->len < p->tot_len);
     /* make sure each packet is complete */
     LWIP_ASSERT("p->next != NULL", p->next != NULL);
     p = p->next;
   }
+  /* { p->tot_len == p->len } => p is the last pbuf of the first packet */
   /* remember next packet on queue */
   q = p->next;
   /* dequeue p from queue */
   p->next = NULL;
-  /* q is now referenced to one less time */
-  pbuf_free(q);
-  LWIP_DEBUGF(PBUF_DEBUG | DBG_FRESH | 2, ("pbuf_dequeue: dereferencing remaining queue %p\n", (void *)q));
+  /* any next packet on queue? */
+  if (q != NULL) {
+    /* although q is no longer referenced by p, it MUST be referenced by
+     * the caller, who is maintaining this packet queue */
+    LWIP_DEBUGF(PBUF_DEBUG | DBG_FRESH | 2, ("pbuf_dequeue: at least one packet on queue, first %p\n", (void *)q));
+  } else {
+    LWIP_DEBUGF(PBUF_DEBUG | DBG_FRESH | 2, ("pbuf_dequeue: no further packets on queue\n"));
+  }
   return q;
 }
 #endif

src/core/raw.c

@@ -1,10 +1,13 @@
 /**
  * @file
- * Raw Access module
+ * 
+ * Implementation of raw protocol PCBs for low-level handling of
+ * different types of protocols besides (or overriding) those
+ * already available in lwIP.
  *
  */
 /*
- * Copyright (c) 2001-2003 Swedish Institute of Computer Science.
+ * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without modification,
@@ -35,20 +38,13 @@
  *
  */
 
-
-/* raw.c
- *
- * The code for the Raw Access to the IP
- *
- */
-
 #include "lwip/opt.h"
 
 #include "lwip/def.h"
 #include "lwip/memp.h"
 #include "lwip/inet.h"
-#include "lwip/netif.h"
 #include "lwip/ip_addr.h"
+#include "lwip/netif.h"
 #include "lwip/raw.h"
 
 #include "lwip/stats.h"
@@ -57,11 +53,10 @@
 #include "lwip/snmp.h"
 
 #if LWIP_RAW
-/* The list of RAW PCBs */
 
+/** The list of RAW PCBs */
 static struct raw_pcb *raw_pcbs = NULL;
 
-
 void
 raw_init(void)
 {
@@ -69,42 +64,54 @@ raw_init(void)
 }
 
 /**
- * Determine if in incoming IP packet is covered by a RAW pcb and
- * and process it if possible
+ * Determine if in incoming IP packet is covered by a RAW PCB
+ * and if so, pass it to a user-provided receive callback function.
  *
  * Given an incoming IP datagram (as a chain of pbufs) this function
- * finds a corresponding RAW PCB and
+ * finds a corresponding RAW PCB and calls the corresponding receive
+ * callback function.
  *
  * @param pbuf pbuf to be demultiplexed to a RAW PCB.
  * @param netif network interface on which the datagram was received.
- * @return 0 if packet is not eated (pbuf needs to be freed then)
- *         or 1 if the packet has been eaten (pbuf needs not to be freed
- *         then)
+ * @Return - 1 if the packet has been eaten by a RAW PCB receive
+ *           callback function. The caller MAY NOT not reference the
+ *           packet any longer, and MAY NOT call pbuf_free().
+ * @return - 0 if packet is not eaten (pbuf is still referenced by the
+ *           caller).
  *
  */
-int
+u8_t
 raw_input(struct pbuf *p, struct netif *inp)
 {
   struct raw_pcb *pcb;
   struct ip_hdr *iphdr;
   int proto;
-  int rc = 0;
+  u8_t eaten = 0;
 
   iphdr = p->payload;
   proto = IPH_PROTO(iphdr);
 
-  for(pcb = raw_pcbs; pcb != NULL; pcb = pcb->next) {
+  pcb = raw_pcbs;
+  /* loop through all raw pcbs until the packet is eaten by one */
+  /* this allows multiple pcbs to match against the packet by design */
+  while ((eaten == 0) && (pcb != NULL)) {
     if (pcb->protocol == proto) {
-      if (pcb->recv) {
-        if (!pcb->recv(pcb->recv_arg, pcb, p, &(iphdr->src)))
-          return 0;
+      /* receive callback function available? */
+      if (pcb->recv != NULL) {
+        /* the receive callback function did not eat the packet? */
+        if (pcb->recv(pcb->recv_arg, pcb, p, &(iphdr->src)) != 0)
+        {
+          /* receive function ate the packet */
+          p = NULL;
+          eaten = 1;
+        }
       }
-      pbuf_free(p);
-      rc = 1;
-      break;
+      /* no receive callback function was set for this raw PCB */
+      /* drop the packet */
     }
+    pcb = pcb->next;
   }
-  return rc;
+  return eaten;
 }
 
 /**
@@ -116,7 +123,7 @@ raw_input(struct pbuf *p, struct netif *inp)
  *
  * @return lwIP error code.
  * - ERR_OK. Successful. No error occured.
- * - ERR_USE. The specified ipaddr is already bound to by
+ * - ERR_USE. The specified IP address is already bound to by
  * another RAW PCB.
  *
  * @see raw_disconnect()
@@ -130,7 +137,7 @@ raw_bind(struct raw_pcb *pcb, struct ip_addr *ipaddr)
 
 /**
  * Connect an RAW PCB. This function is required by upper layers
- * of lwip. Using the raw api you could use raw_send_to() instead
+ * of lwip. Using the raw api you could use raw_sendto() instead
  *
  * This will associate the RAW PCB with the remote address.
  *
@@ -139,7 +146,7 @@ raw_bind(struct raw_pcb *pcb, struct ip_addr *ipaddr)
  *
  * @return lwIP error code
  *
- * @see raw_disconnect() and raw_send_to()
+ * @see raw_disconnect() and raw_sendto()
  */
 err_t
 raw_connect(struct raw_pcb *pcb, struct ip_addr *ipaddr)
@@ -150,14 +157,21 @@ raw_connect(struct raw_pcb *pcb, struct ip_addr *ipaddr)
 
 
 /**
- * Set the callback function if a RAW packet with the pcb's protocol
- * is received. If the callback function returns a value unequal 0
- * the raw packet is "eaten" and not forwarded to any other raw pcb
- * including lwip itself
+ * Set the callback function for received packets that match the
+ * raw PCB's protocol and binding. 
+ * 
+ * The callback function MUST either
+ * - eat the packet by calling pbuf_free() and returning non-zero. The
+ *   packet will not be passed to other raw PCBs or other protocol layers.
+ * - not free the packet, and return zero. The packet will be matched
+ *   against further PCBs and/or forwarded to another protocol layers.
+ * 
+ * @return non-zero if the packet was free()d, zero if the packet remains
+ * available for others.
  */
 void
 raw_recv(struct raw_pcb *pcb,
-         int (* recv)(void *arg, struct raw_pcb *upcb, struct pbuf *p,
+         u8_t (* recv)(void *arg, struct raw_pcb *upcb, struct pbuf *p,
                       struct ip_addr *addr),
          void *recv_arg)
 {
@@ -168,25 +182,25 @@ raw_recv(struct raw_pcb *pcb,
 
 /**
  * Send the raw IP packet to the given address. Note that actually you cannot
- * modify the IP headers (this is inconsitent with the receive callback where
- * you actually get the IP headers), you can only specifiy the ip payload here.
- * It requires some more changes in LWIP. (there will be a raw_send() function
- * then)
+ * modify the IP headers (this is inconsistent with the receive callback where
+ * you actually get the IP headers), you can only specify the IP payload here.
+ * It requires some more changes in lwIP. (there will be a raw_send() function
+ * then.)
  *
  * @param pcb the raw pcb which to send
- * @param p the ip payload to send
- * @param ipaddr the destination address of the whole IP packet
+ * @param p the IP payload to send
+ * @param ipaddr the destination address of the IP packet
  *
  */
 err_t
-raw_send_to(struct raw_pcb *pcb, struct pbuf *p, struct ip_addr *ipaddr)
+raw_sendto(struct raw_pcb *pcb, struct pbuf *p, struct ip_addr *ipaddr)
 {
   err_t err;
   struct netif *netif;
   struct ip_addr *src_ip;
   struct pbuf *q; /* q will be sent down the stack */
   
-  LWIP_DEBUGF(RAW_DEBUG | DBG_TRACE | 3, ("raw_send_to\n"));
+  LWIP_DEBUGF(RAW_DEBUG | DBG_TRACE | 3, ("raw_sendto\n"));
   
   /* not enough space to add an IP header to first pbuf in given p chain? */
   if (pbuf_header(p, IP_HLEN)) {
@@ -194,13 +208,13 @@ raw_send_to(struct raw_pcb *pcb, struct pbuf *p, struct ip_addr *ipaddr)
     q = pbuf_alloc(PBUF_IP, 0, PBUF_RAM);
     /* new header pbuf could not be allocated? */
     if (q == NULL) {
-      LWIP_DEBUGF(RAW_DEBUG | DBG_TRACE | 2, ("raw_send_to: could not allocate header\n"));
+      LWIP_DEBUGF(RAW_DEBUG | DBG_TRACE | 2, ("raw_sendto: could not allocate header\n"));
       return ERR_MEM;
     }
     /* chain header q in front of given pbuf p */
     pbuf_chain(q, p);
     /* { first pbuf q points to header pbuf } */
-    LWIP_DEBUGF(RAW_DEBUG, ("raw_send_to: added header pbuf %p before given pbuf %p\n", (void *)q, (void *)p));
+    LWIP_DEBUGF(RAW_DEBUG, ("raw_sendto: added header pbuf %p before given pbuf %p\n", (void *)q, (void *)p));
   }  else {
     /* first pbuf q equals given pbuf */
     q = p;
@@ -208,10 +222,11 @@ raw_send_to(struct raw_pcb *pcb, struct pbuf *p, struct ip_addr *ipaddr)
   }
   
   if ((netif = ip_route(ipaddr)) == NULL) {
-    LWIP_DEBUGF(RAW_DEBUG | 1, ("raw_send_to: No route to 0x%lx\n", ipaddr->addr));
+    LWIP_DEBUGF(RAW_DEBUG | 1, ("raw_sendto: No route to 0x%lx\n", ipaddr->addr));
 #if RAW_STATS
     /*    ++lwip_stats.raw.rterr;*/
 #endif /* RAW_STATS */
+    /* free any temporary header pbuf allocated by pbuf_header() */
     if (q != p) {
       pbuf_free(q);
     }
@@ -240,14 +255,14 @@ raw_send_to(struct raw_pcb *pcb, struct pbuf *p, struct ip_addr *ipaddr)
  * Send the raw IP packet to the address given by raw_connect()
  *
  * @param pcb the raw pcb which to send
- * @param p the ip payload to send
- * @param ipaddr the destination address of the whole IP packet
+ * @param p the IP payload to send
+ * @param ipaddr the destination address of the IP packet
  *
  */
 err_t
 raw_send(struct raw_pcb *pcb, struct pbuf *p)
 {
-  return raw_send_to(pcb,p,&pcb->remote_ip);
+  return raw_sendto(pcb, p, &pcb->remote_ip);
 }
 
 /**
@@ -266,7 +281,7 @@ raw_remove(struct raw_pcb *pcb)
   if (raw_pcbs == pcb) {
     /* make list start at 2nd pcb */
     raw_pcbs = raw_pcbs->next;
-  /* pcb not 1st in list */
+    /* pcb not 1st in list */
   } else for(pcb2 = raw_pcbs; pcb2 != NULL; pcb2 = pcb2->next) {
     /* find pcb in raw_pcbs list */
     if (pcb2->next != NULL && pcb2->next == pcb) {
@@ -303,7 +318,6 @@ raw_new(u16_t proto) {
     pcb->next = raw_pcbs;
     raw_pcbs = pcb;
   }
-
   return pcb;
 }
 

src/core/stats.c

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2001-2003 Swedish Institute of Computer Science.
+ * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
  * All rights reserved. 
  * 
  * Redistribution and use in source and binary forms, with or without modification, 
@@ -47,5 +47,68 @@ stats_init(void)
 {
   memset(&lwip_stats, 0, sizeof(struct stats_));
 }
+#if LWIP_STATS_DISPLAY
+void
+stats_display_proto(struct stats_proto *proto, char *name)
+{
+  LWIP_PLATFORM_DIAG(("\n%s\n\t", name));
+  LWIP_PLATFORM_DIAG(("xmit: %d\n\t", proto->xmit)); 
+  LWIP_PLATFORM_DIAG(("rexmit: %d\n\t", proto->rexmit)); 
+  LWIP_PLATFORM_DIAG(("recv: %d\n\t", proto->recv)); 
+  LWIP_PLATFORM_DIAG(("fw: %d\n\t", proto->fw)); 
+  LWIP_PLATFORM_DIAG(("drop: %d\n\t", proto->drop)); 
+  LWIP_PLATFORM_DIAG(("chkerr: %d\n\t", proto->chkerr)); 
+  LWIP_PLATFORM_DIAG(("lenerr: %d\n\t", proto->lenerr)); 
+  LWIP_PLATFORM_DIAG(("memerr: %d\n\t", proto->memerr)); 
+  LWIP_PLATFORM_DIAG(("rterr: %d\n\t", proto->rterr)); 
+  LWIP_PLATFORM_DIAG(("proterr: %d\n\t", proto->proterr)); 
+  LWIP_PLATFORM_DIAG(("opterr: %d\n\t", proto->opterr)); 
+  LWIP_PLATFORM_DIAG(("err: %d\n\t", proto->err)); 
+  LWIP_PLATFORM_DIAG(("cachehit: %d\n", proto->cachehit)); 
+}
+
+void
+stats_display_pbuf(struct stats_pbuf *pbuf)
+{
+  LWIP_PLATFORM_DIAG(("\nPBUF\n\t"));
+  LWIP_PLATFORM_DIAG(("avail: %d\n\t", pbuf->avail)); 
+  LWIP_PLATFORM_DIAG(("used: %d\n\t", pbuf->used)); 
+  LWIP_PLATFORM_DIAG(("max: %d\n\t", pbuf->max)); 
+  LWIP_PLATFORM_DIAG(("err: %d\n\t", pbuf->err)); 
+  LWIP_PLATFORM_DIAG(("alloc_locked: %d\n\t", pbuf->alloc_locked)); 
+  LWIP_PLATFORM_DIAG(("refresh_locked: %d\n", pbuf->refresh_locked)); 
+}
+
+void
+stats_display_mem(struct stats_mem *mem, char *name)
+{
+  LWIP_PLATFORM_DIAG(("\n MEM %s\n\t", name));
+  LWIP_PLATFORM_DIAG(("avail: %d\n\t", mem->avail)); 
+  LWIP_PLATFORM_DIAG(("used: %d\n\t", mem->used)); 
+  LWIP_PLATFORM_DIAG(("max: %d\n\t", mem->max)); 
+  LWIP_PLATFORM_DIAG(("err: %d\n", mem->err));
+  
+}
+
+void
+stats_display(void)
+{
+  int i;
+  char * memp_names[] = {"PBUF", "RAW_PCB", "UDP_PCB", "TCP_PCB", "TCP_PCB_LISTEN",
+	  		"TCP_SEG", "NETBUF", "NETCONN", "API_MSG", "TCP_MSG", "TIMEOUT"};
+  stats_display_proto(&lwip_stats.link, "LINK");
+  stats_display_proto(&lwip_stats.ip_frag, "IP_FRAG");
+  stats_display_proto(&lwip_stats.ip, "IP");
+  stats_display_proto(&lwip_stats.icmp, "ICMP");
+  stats_display_proto(&lwip_stats.udp, "UDP");
+  stats_display_proto(&lwip_stats.tcp, "TCP");
+  stats_display_pbuf(&lwip_stats.pbuf);
+  stats_display_mem(&lwip_stats.mem, "HEAP");
+  for (i = 0; i < MEMP_MAX; i++) {
+    stats_display_mem(&lwip_stats.memp[i], memp_names[i]);
+  }
+	
+}
+#endif /* LWIP_STATS_DISPLAY */
 #endif /* LWIP_STATS */
 

src/core/sys.c

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2001-2003 Swedish Institute of Computer Science.
+ * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without modification,
@@ -174,6 +174,7 @@ sys_timeout(u32_t msecs, sys_timeout_handler h, void *arg)
     (void *)timeout, msecs, (void *)h, (void *)arg));
 
   LWIP_ASSERT("sys_timeout: timeouts != NULL", timeouts != NULL);
+
   if (timeouts->next == NULL) {
     timeouts->next = timeout;
     return;
@@ -186,14 +187,13 @@ sys_timeout(u32_t msecs, sys_timeout_handler h, void *arg)
   } else {
     for(t = timeouts->next; t != NULL; t = t->next) {
       timeout->time -= t->time;
-      if (t->next == NULL ||
-   t->next->time > timeout->time) {
-  if (t->next != NULL) {
-    t->next->time -= timeout->time;
-  }
-  timeout->next = t->next;
-  t->next = timeout;
-  break;
+      if (t->next == NULL || t->next->time > timeout->time) {
+        if (t->next != NULL) {
+          t->next->time -= timeout->time;
+        }
+        timeout->next = t->next;
+        t->next = timeout;
+        break;
       }
     }
   }

src/core/tcp.c

@@ -2,10 +2,15 @@
  * @file
  *
  * Transmission Control Protocol for IP
+ *
+ * This file contains common functions for the TCP implementation, such as functinos
+ * for manipulating the data structures and the TCP timer functions. TCP functions
+ * related to input and output is found in tcp_in.c and tcp_out.c respectively.
+ *
  */
 
 /*
- * Copyright (c) 2001-2003 Swedish Institute of Computer Science.
+ * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
  * All rights reserved. 
  * 
  * Redistribution and use in source and binary forms, with or without modification, 
@@ -36,17 +41,6 @@
  *
  */
 
-
-/* tcp.c
- *
- * This file contains common functions for the TCP implementation, such as functinos
- * for manipulating the data structures and the TCP timer functions. TCP functions
- * related to input and output is found in tcp_input.c and tcp_output.c respectively.
- *
- */
-
-
-
 #include "lwip/opt.h"
 #include "lwip/def.h"
 #include "lwip/mem.h"
@@ -62,30 +56,28 @@ const u8_t tcp_backoff[13] =
     { 1, 2, 3, 4, 5, 6, 7, 7, 7, 7, 7, 7, 7};
 
 /* The TCP PCB lists. */
-struct tcp_pcb_listen *tcp_listen_pcbs;  /* List of all TCP PCBs in LISTEN state. */
-struct tcp_pcb *tcp_active_pcbs;  /* List of all TCP PCBs that are in a
-         state in which they accept or send
-         data. */
-struct tcp_pcb *tcp_tw_pcbs;      /* List of all TCP PCBs in TIME-WAIT. */
+
+/** List of all TCP PCBs in LISTEN state */
+union tcp_listen_pcbs_t tcp_listen_pcbs;
+/** List of all TCP PCBs that are in a state in which
+ * they accept or send data. */
+struct tcp_pcb *tcp_active_pcbs;  
+/** List of all TCP PCBs in TIME-WAIT state */
+struct tcp_pcb *tcp_tw_pcbs;
 
 struct tcp_pcb *tcp_tmp_pcb;
 
 static u8_t tcp_timer;
-
 static u16_t tcp_new_port(void);
 
-
-/*
- * tcp_init():
- *
+/**
  * Initializes the TCP layer.
  */
-
 void
 tcp_init(void)
 {
   /* Clear globals. */
-  tcp_listen_pcbs = NULL;
+  tcp_listen_pcbs.listen_pcbs = NULL;
   tcp_active_pcbs = NULL;
   tcp_tw_pcbs = NULL;
   tcp_tmp_pcb = NULL;
@@ -96,13 +88,10 @@ tcp_init(void)
   
 }
 
-/*
- * tcp_tmr():
- *
+/**
  * Called periodically to dispatch TCP timers.
  *
  */
-
 void
 tcp_tmr(void)
 {
@@ -116,13 +105,10 @@ tcp_tmr(void)
   }
 }
 
-/*
- * tcp_close():
- *
+/**
  * Closes the connection held by the PCB.
  *
  */
-
 err_t
 tcp_close(struct tcp_pcb *pcb)
 {
@@ -136,7 +122,7 @@ tcp_close(struct tcp_pcb *pcb)
   switch (pcb->state) {
   case LISTEN:
     err = ERR_OK;
-    tcp_pcb_remove((struct tcp_pcb **)&tcp_listen_pcbs, pcb);
+    tcp_pcb_remove((struct tcp_pcb **)&tcp_listen_pcbs.pcbs, pcb);
     memp_free(MEMP_TCP_PCB_LISTEN, pcb);
     pcb = NULL;
     break;
@@ -172,15 +158,12 @@ tcp_close(struct tcp_pcb *pcb)
   return err;
 }
 
-/*
- * tcp_abort()
- *
+/**
  * Aborts a connection by sending a RST to the remote host and deletes
  * the local protocol control block. This is done when a connection is
  * killed because of shortage of memory.
  *
  */
-
 void
 tcp_abort(struct tcp_pcb *pcb)
 {
@@ -229,9 +212,7 @@ tcp_abort(struct tcp_pcb *pcb)
   }
 }
 
-/*
- * tcp_bind():
- *
+/**
  * Binds the connection to a local portnumber and IP address. If the
  * IP address is not given (i.e., ipaddr == NULL), the IP address of
  * the outgoing network interface is used instead.
@@ -242,16 +223,16 @@ err_t
 tcp_bind(struct tcp_pcb *pcb, struct ip_addr *ipaddr, u16_t port)
 {
   struct tcp_pcb *cpcb;
-#ifdef SO_REUSE
+#if SO_REUSE
   int reuse_port_all_set = 1;
 #endif /* SO_REUSE */
 
   if (port == 0) {
     port = tcp_new_port();
   }
-#ifndef SO_REUSE
+#if SO_REUSE == 0
   /* Check if the address already is in use. */
-  for(cpcb = (struct tcp_pcb *)tcp_listen_pcbs;
+  for(cpcb = (struct tcp_pcb *)tcp_listen_pcbs.pcbs;
       cpcb != NULL; cpcb = cpcb->next) {
     if (cpcb->local_port == port) {
       if (ip_addr_isany(&(cpcb->local_ip)) ||
@@ -284,7 +265,7 @@ tcp_bind(struct tcp_pcb *pcb, struct ip_addr *ipaddr, u16_t port)
   
   When the two options aren't set and specified port is already bound, ERR_USE is returned saying that 
   address is already in use. */
-  for(cpcb = (struct tcp_pcb *)tcp_listen_pcbs; cpcb != NULL; cpcb = cpcb->next) {
+  for(cpcb = (struct tcp_pcb *)tcp_listen_pcbs.pcbs; cpcb != NULL; cpcb = cpcb->next) {
     if(cpcb->local_port == port) {
       if(ip_addr_cmp(&(cpcb->local_ip), ipaddr)) {
         if(pcb->so_options & SOF_REUSEPORT) {
@@ -386,16 +367,13 @@ tcp_accept_null(void *arg, struct tcp_pcb *pcb, err_t err)
 }
 #endif /* LWIP_CALLBACK_API */
 
-/*
- * tcp_listen():
- *
+/**
  * Set the state of the connection to be LISTEN, which means that it
  * is able to accept incoming connections. The protocol control block
  * is reallocated in order to consume less memory. Setting the
  * connection to LISTEN is an irreversible process.
  *
  */
-
 struct tcp_pcb *
 tcp_listen(struct tcp_pcb *pcb)
 {
@@ -421,19 +399,16 @@ tcp_listen(struct tcp_pcb *pcb)
 #if LWIP_CALLBACK_API
   lpcb->accept = tcp_accept_null;
 #endif /* LWIP_CALLBACK_API */
-  TCP_REG(&tcp_listen_pcbs, lpcb);
+  TCP_REG(&tcp_listen_pcbs.listen_pcbs, lpcb);
   return (struct tcp_pcb *)lpcb;
 }
 
-/*
- * tcp_recved():
- *
+/**
  * This function should be called by the application when it has
  * processed the data. The purpose is to advertise a larger window
  * when the data has been processed.
  *
  */
-
 void
 tcp_recved(struct tcp_pcb *pcb, u16_t len)
 {
@@ -444,6 +419,16 @@ tcp_recved(struct tcp_pcb *pcb, u16_t len)
   }
   if (!(pcb->flags & TF_ACK_DELAY) &&
      !(pcb->flags & TF_ACK_NOW)) {
+    /*
+     * We send an ACK here (if one is not already pending, hence
+     * the above tests) as tcp_recved() implies that the application
+     * has processed some data, and so we can open the receiver's
+     * window to allow more to be transmitted.  This could result in
+     * two ACKs being sent for each received packet in some limited cases
+     * (where the application is only receiving data, and is slow to
+     * process it) but it is necessary to guarantee that the sender can
+     * continue to transmit.
+     */
     tcp_ack(pcb);
   }
 
@@ -451,13 +436,10 @@ tcp_recved(struct tcp_pcb *pcb, u16_t len)
          len, pcb->rcv_wnd, TCP_WND - pcb->rcv_wnd));
 }
 
-/*
- * tcp_new_port():
- *
+/**
  * A nastly hack featuring 'goto' statements that allocates a
  * new TCP local port.
  */
-
 static u16_t
 tcp_new_port(void)
 {
@@ -483,7 +465,7 @@ tcp_new_port(void)
       goto again;
     }
   }
-  for(pcb = (struct tcp_pcb *)tcp_listen_pcbs; pcb != NULL; pcb = pcb->next) {
+  for(pcb = (struct tcp_pcb *)tcp_listen_pcbs.pcbs; pcb != NULL; pcb = pcb->next) {
     if (pcb->local_port == port) {
       goto again;
     }
@@ -491,14 +473,11 @@ tcp_new_port(void)
   return port;
 }
 
-/*
- * tcp_connect():
- *
+/**
  * Connects to another host. The function given as the "connected"
  * argument will be called when the connection has been established.
  *
  */
-
 err_t
 tcp_connect(struct tcp_pcb *pcb, struct ip_addr *ipaddr, u16_t port,
       err_t (* connected)(void *arg, struct tcp_pcb *tpcb, err_t err))
@@ -546,14 +525,11 @@ tcp_connect(struct tcp_pcb *pcb, struct ip_addr *ipaddr, u16_t port,
   return ret;
 } 
 
-/*
- * tcp_slowtmr():
- *
+/**
  * Called every 500 ms and implements the retransmission timer and the timer that
  * removes PCBs that have been in TIME-WAIT for enough time. It also increments
  * various timers such as the inactivity timer in each PCB.
  */
-
 void
 tcp_slowtmr(void)
 {
@@ -730,12 +706,9 @@ tcp_slowtmr(void)
   }
 }
 
-/*
- * tcp_fasttmr():
- *
+/**
  * Is called every TCP_FAST_INTERVAL (250 ms) and sends delayed ACKs.
  */
-
 void
 tcp_fasttmr(void)
 {
@@ -751,13 +724,10 @@ tcp_fasttmr(void)
   }
 }
 
-/*
- * tcp_segs_free():
- *
+/**
  * Deallocates a list of TCP segments (tcp_seg structures).
  *
  */
-
 u8_t
 tcp_segs_free(struct tcp_seg *seg)
 {
@@ -771,13 +741,10 @@ tcp_segs_free(struct tcp_seg *seg)
   return count;
 }
 
-/*
- * tcp_seg_free():
- *
+/**
  * Frees a TCP segment.
  *
  */
-
 u8_t
 tcp_seg_free(struct tcp_seg *seg)
 {
@@ -795,13 +762,10 @@ tcp_seg_free(struct tcp_seg *seg)
   return count;
 }
 
-/*
- * tcp_setprio():
- *
+/**
  * Sets the priority of a connection.
  *
  */
-
 void
 tcp_setprio(struct tcp_pcb *pcb, u8_t prio)
 {
@@ -809,13 +773,10 @@ tcp_setprio(struct tcp_pcb *pcb, u8_t prio)
 }
 #if TCP_QUEUE_OOSEQ
 
-/*
- * tcp_seg_copy():
- *
+/**
  * Returns a copy of the given TCP segment.
  *
  */ 
-
 struct tcp_seg *
 tcp_seg_copy(struct tcp_seg *seg)
 {
@@ -951,12 +912,13 @@ tcp_alloc(u8_t prio)
   return pcb;
 }
 
-/*
- * tcp_new():
- *
+/**
  * Creates a new TCP protocol control block but doesn't place it on
  * any of the TCP PCB lists.
  *
+ * @internal: Maybe there should be a idle TCP PCB list where these
+ * PCBs are put on. We can then implement port reservation using
+ * tcp_bind(). Currently, we lack this (BSD socket type of) feature.
  */
 
 struct tcp_pcb *
@@ -980,14 +942,11 @@ tcp_arg(struct tcp_pcb *pcb, void *arg)
 }
 #if LWIP_CALLBACK_API
 
-/*
- * tcp_recv():
- *
+/**
  * Used to specify the function that should be called when a TCP
  * connection receives data.
  *
  */ 
-
 void
 tcp_recv(struct tcp_pcb *pcb,
    err_t (* recv)(void *arg, struct tcp_pcb *tpcb, struct pbuf *p, err_t err))
@@ -995,9 +954,7 @@ tcp_recv(struct tcp_pcb *pcb,
   pcb->recv = recv;
 }
 
-/*
- * tcp_sent():
- *
+/**
  * Used to specify the function that should be called when TCP data
  * has been successfully delivered to the remote host.
  *
@@ -1010,14 +967,11 @@ tcp_sent(struct tcp_pcb *pcb,
   pcb->sent = sent;
 }
 
-/*
- * tcp_err():
- *
+/**
  * Used to specify the function that should be called when a fatal error
  * has occured on the connection.
  *
  */ 
-
 void
 tcp_err(struct tcp_pcb *pcb,
    void (* errf)(void *arg, err_t err))
@@ -1025,14 +979,11 @@ tcp_err(struct tcp_pcb *pcb,
   pcb->errf = errf;
 }
 
-/*
- * tcp_accept():
- *
+/**
  * Used for specifying the function that should be called when a
  * LISTENing connection has been connected to another host.
  *
  */ 
-
 void
 tcp_accept(struct tcp_pcb *pcb,
      err_t (* accept)(void *arg, struct tcp_pcb *newpcb, err_t err))
@@ -1042,15 +993,12 @@ tcp_accept(struct tcp_pcb *pcb,
 #endif /* LWIP_CALLBACK_API */
 
 
-/*
- * tcp_poll():
- *
+/**
  * Used to specify the function that should be called periodically
  * from TCP. The interval is specified in terms of the TCP coarse
  * timer interval, which is called twice a second.
  *
  */ 
-
 void
 tcp_poll(struct tcp_pcb *pcb,
    err_t (* poll)(void *arg, struct tcp_pcb *tpcb), u8_t interval)
@@ -1061,13 +1009,10 @@ tcp_poll(struct tcp_pcb *pcb,
   pcb->pollinterval = interval;
 }
 
-/*
- * tcp_pcb_purge():
- *
+/**
  * Purges a TCP PCB. Removes any buffered data and frees the buffer memory.
  *
  */
-
 void
 tcp_pcb_purge(struct tcp_pcb *pcb)
 {
@@ -1097,13 +1042,10 @@ tcp_pcb_purge(struct tcp_pcb *pcb)
   }
 }
 
-/*
- * tcp_pcb_remove():
- *
+/**
  * Purges the PCB and removes it from a PCB list. Any delayed ACKs are sent first.
  *
  */
-
 void
 tcp_pcb_remove(struct tcp_pcb **pcblist, struct tcp_pcb *pcb)
 {
@@ -1123,13 +1065,10 @@ tcp_pcb_remove(struct tcp_pcb **pcblist, struct tcp_pcb *pcb)
   LWIP_ASSERT("tcp_pcb_remove: tcp_pcbs_sane()", tcp_pcbs_sane());
 }
 
-/*
- * tcp_next_iss():
- *
+/**
  * Calculates a new initial sequence number for new connections.
  *
  */
-
 u32_t
 tcp_next_iss(void)
 {
@@ -1253,7 +1192,7 @@ tcp_debug_print_pcbs(void)
     tcp_debug_print_state(pcb->state);
   }    
   LWIP_DEBUGF(TCP_DEBUG, ("Listen PCB states:\n"));
-  for(pcb = (struct tcp_pcb *)tcp_listen_pcbs; pcb != NULL; pcb = pcb->next) {
+  for(pcb = (struct tcp_pcb *)tcp_listen_pcbs.pcbs; pcb != NULL; pcb = pcb->next) {
     LWIP_DEBUGF(TCP_DEBUG, ("Local port %u, foreign port %u snd_nxt %lu rcv_nxt %lu ",
                        pcb->local_port, pcb->remote_port,
                        pcb->snd_nxt, pcb->rcv_nxt));

src/core/tcp_in.c

@@ -2,10 +2,16 @@
  * @file
  *
  * Transmission Control Protocol, incoming traffic
+ *
+ * The input processing functions of TCP.
+ *
+ * These functions are generally called in the order (ip_input() ->) tcp_input() ->
+ * tcp_process() -> tcp_receive() (-> application).
+ * 
  */
 
 /*
- * Copyright (c) 2001-2003 Swedish Institute of Computer Science.
+ * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without modification,
@@ -36,21 +42,10 @@
  *
  */
 
-
-/* tcp_input.c
- *
- * The input processing functions of TCP.
- *
- * These functions are generally called in the order (ip_input() ->) tcp_input() ->
- * tcp_process() -> tcp_receive() (-> application).
- *
- */
-
-
-
 #include "lwip/def.h"
 #include "lwip/opt.h"
 
+#include "lwip/ip_addr.h"
 #include "lwip/netif.h"
 #include "lwip/mem.h"
 #include "lwip/memp.h"
@@ -102,7 +97,7 @@ tcp_input(struct pbuf *p, struct netif *inp)
   u8_t hdrlen;
   err_t err;
 
-#ifdef SO_REUSE
+#if SO_REUSE
   struct tcp_pcb *pcb_temp;
   int reuse = 0;
   int reuse_port = 0;
@@ -130,12 +125,13 @@ tcp_input(struct pbuf *p, struct netif *inp)
   }
 
   /* Don't even process incoming broadcasts/multicasts. */
-  if (ip_addr_isbroadcast(&(iphdr->dest), &(inp->netmask)) ||
-     ip_addr_ismulticast(&(iphdr->dest))) {
+  if (ip_addr_isbroadcast(&(iphdr->dest), inp) ||
+      ip_addr_ismulticast(&(iphdr->dest))) {
     pbuf_free(p);
     return;
   }
 
+#if CHECKSUM_CHECK_TCP
   /* Verify TCP checksum. */
   if (inet_chksum_pseudo(p, (struct ip_addr *)&(iphdr->src),
       (struct ip_addr *)&(iphdr->dest),
@@ -152,7 +148,7 @@ tcp_input(struct pbuf *p, struct netif *inp)
     pbuf_free(p);
     return;
   }
-
+#endif
 
   /* Move the payload pointer in the pbuf so that it points to the
      TCP data instead of the TCP header. */
@@ -173,7 +169,7 @@ tcp_input(struct pbuf *p, struct netif *inp)
      for an active connection. */
   prev = NULL;
 
-#ifdef SO_REUSE
+#if SO_REUSE
   pcb_temp = tcp_active_pcbs;
   
  again_1:
@@ -191,7 +187,7 @@ tcp_input(struct pbuf *p, struct netif *inp)
        ip_addr_cmp(&(pcb->remote_ip), &(iphdr->src)) &&
        ip_addr_cmp(&(pcb->local_ip), &(iphdr->dest))) {
 
-#ifdef SO_REUSE
+#if SO_REUSE
       if(pcb->so_options & SOF_REUSEPORT) {
         if(reuse) {
           /* We processed one PCB already */
@@ -255,7 +251,7 @@ tcp_input(struct pbuf *p, struct netif *inp)
   /* Finally, if we still did not get a match, we check all PCBs that
      are LISTENing for incoming connections. */
     prev = NULL;
-    for(lpcb = tcp_listen_pcbs; lpcb != NULL; lpcb = lpcb->next) {
+    for(lpcb = tcp_listen_pcbs.listen_pcbs; lpcb != NULL; lpcb = lpcb->next) {
       if ((ip_addr_isany(&(lpcb->local_ip)) ||
     ip_addr_cmp(&(lpcb->local_ip), &(iphdr->dest))) &&
    lpcb->local_port == tcphdr->dest) {
@@ -265,9 +261,9 @@ tcp_input(struct pbuf *p, struct netif *inp)
   if (prev != NULL) {
     ((struct tcp_pcb_listen *)prev)->next = lpcb->next;
           /* our successor is the remainder of the listening list */
-    lpcb->next = tcp_listen_pcbs;
+    lpcb->next = tcp_listen_pcbs.listen_pcbs;
           /* put this listening pcb at the head of the listening list */
-    tcp_listen_pcbs = lpcb;
+    tcp_listen_pcbs.listen_pcbs = lpcb;
   }
 
   LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_input: packed for LISTENing connection.\n"));
@@ -361,7 +357,7 @@ tcp_input(struct pbuf *p, struct netif *inp)
     tcp_debug_print_state(pcb->state);
 #endif /* TCP_DEBUG */
 #endif /* TCP_INPUT_DEBUG */
-#ifdef SO_REUSE
+#if SO_REUSE
     /* First socket should receive now */
     if(reuse_port) {
       LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_input: searching next PCB.\n"));
@@ -373,7 +369,7 @@ tcp_input(struct pbuf *p, struct netif *inp)
 #endif /* SO_REUSE */
 
   } else {
-#ifdef SO_REUSE
+#if SO_REUSE
     if(reuse) {
       LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_input: freeing PBUF with reference counter set to %i\n", p->ref));
       pbuf_free(p);
@@ -392,7 +388,7 @@ tcp_input(struct pbuf *p, struct netif *inp)
     }
     pbuf_free(p);
   }
-#ifdef SO_REUSE
+#if SO_REUSE
  end:
 #endif /* SO_REUSE */
   LWIP_ASSERT("tcp_input: tcp_pcbs_sane()", tcp_pcbs_sane());

src/core/tcp_out.c

@@ -2,9 +2,13 @@
  * @file
  *
  * Transmission Control Protocol, outgoing traffic
+ *
+ * The output functions of TCP.
+ *
  */
+
 /*
- * Copyright (c) 2001-2003 Swedish Institute of Computer Science.
+ * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without modification,
@@ -35,15 +39,6 @@
  *
  */
 
-
-/* tcp_output.c
- *
- * The output functions of TCP.
- *
- */
-
-
-
 #include "lwip/def.h"
 #include "lwip/opt.h"
 
@@ -51,6 +46,7 @@
 #include "lwip/memp.h"
 #include "lwip/sys.h"
 
+#include "lwip/ip_addr.h"
 #include "lwip/netif.h"
 
 #include "lwip/inet.h"
@@ -63,8 +59,6 @@
 /* Forward declarations.*/
 static void tcp_output_segment(struct tcp_seg *seg, struct tcp_pcb *pcb);
 
-
-
 err_t
 tcp_send_ctrl(struct tcp_pcb *pcb, u8_t flags)
 {
@@ -72,6 +66,14 @@ tcp_send_ctrl(struct tcp_pcb *pcb, u8_t flags)
 
 }
 
+/*
+ * NB. tcp_write() enqueues data for sending, but does not send it
+ * straight away.  It waits in the expectation of more data being sent
+ * soon (as it can send them more efficiently by combining them
+ * together).  To prompt the system to send data now, call
+ * tcp_output() after calling tcp_write().
+ */
+
 err_t
 tcp_write(struct tcp_pcb *pcb, const void *arg, u16_t len, u8_t copy)
 {
@@ -400,9 +402,10 @@ tcp_output(struct tcp_pcb *pcb)
     TCPH_HDRLEN_SET(tcphdr, 5);
 
     tcphdr->chksum = 0;
+#if CHECKSUM_GEN_TCP
     tcphdr->chksum = inet_chksum_pseudo(p, &(pcb->local_ip), &(pcb->remote_ip),
           IP_PROTO_TCP, p->tot_len);
-
+#endif
 
     ip_output(p, &(pcb->local_ip), &(pcb->remote_ip), pcb->ttl, pcb->tos,
         IP_PROTO_TCP);
@@ -518,10 +521,12 @@ tcp_output_segment(struct tcp_seg *seg, struct tcp_pcb *pcb)
   seg->p->payload = seg->tcphdr;
 
   seg->tcphdr->chksum = 0;
+#if CHECKSUM_GEN_TCP
   seg->tcphdr->chksum = inet_chksum_pseudo(seg->p,
              &(pcb->local_ip),
              &(pcb->remote_ip),
              IP_PROTO_TCP, seg->p->tot_len);
+#endif
   TCP_STATS_INC(tcp.xmit);
 
   ip_output(seg->p, &(pcb->local_ip), &(pcb->remote_ip), pcb->ttl, pcb->tos,
@@ -552,9 +557,10 @@ tcp_rst(u32_t seqno, u32_t ackno,
   TCPH_HDRLEN_SET(tcphdr, 5);
 
   tcphdr->chksum = 0;
+#if CHECKSUM_GEN_TCP
   tcphdr->chksum = inet_chksum_pseudo(p, local_ip, remote_ip,
               IP_PROTO_TCP, p->tot_len);
-
+#endif
   TCP_STATS_INC(tcp.xmit);
    /* Send output with hardcoded TTL since we have no access to the pcb */
   ip_output(p, local_ip, remote_ip, TCP_TTL, 0, IP_PROTO_TCP);
@@ -621,8 +627,9 @@ tcp_keepalive(struct tcp_pcb *pcb)
    TCPH_HDRLEN_SET(tcphdr, 5);
    
    tcphdr->chksum = 0;
+#if CHECKSUM_GEN_TCP
    tcphdr->chksum = inet_chksum_pseudo(p, &pcb->local_ip, &pcb->remote_ip, IP_PROTO_TCP, p->tot_len);
-
+#endif
   TCP_STATS_INC(tcp.xmit);
 
    /* Send output to IP */

src/core/udp.c

@@ -4,7 +4,7 @@
  *
  */
 /*
- * Copyright (c) 2001-2003 Swedish Institute of Computer Science.
+ * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without modification,
@@ -47,10 +47,10 @@
 #include "lwip/def.h"
 #include "lwip/memp.h"
 #include "lwip/inet.h"
+#include "lwip/ip_addr.h"
 #include "lwip/netif.h"
 #include "lwip/udp.h"
 #include "lwip/icmp.h"
-#include "lwip/ip_addr.h"
 
 #include "lwip/stats.h"
 
@@ -71,76 +71,6 @@ udp_init(void)
   udp_pcbs = pcb_cache = NULL;
 }
 
-
-/* udp_lookup:
- *
- * An experimental feature that will be changed in future versions. Do
- * not depend on it yet...
- */
-
-#ifdef LWIP_DEBUG
-u8_t
-udp_lookup(struct ip_hdr *iphdr, struct netif *inp)
-{
-  struct udp_pcb *pcb;
-  struct udp_hdr *udphdr;
-  u16_t src, dest;
-
-    PERF_START;
-  (void)inp;
-
-    udphdr = (struct udp_hdr *)(u8_t *)iphdr + IPH_HL(iphdr) * 4;
-
-  src = ntohs(udphdr->src);
-  dest = ntohs(udphdr->dest);
-
-    pcb = pcb_cache;
-  if (pcb != NULL &&
-    pcb->remote_port == src &&
-    pcb->local_port == dest &&
-    (ip_addr_isany(&pcb->remote_ip) ||
-    ip_addr_cmp(&(pcb->remote_ip), &(iphdr->src))) &&
-    (ip_addr_isany(&pcb->local_ip) ||
-    ip_addr_cmp(&(pcb->local_ip), &(iphdr->dest)))) {
-    return 1;
-  }
-  else {
-    for(pcb = udp_pcbs; pcb != NULL; pcb = pcb->next) {
-      if (pcb->remote_port == src &&
-   pcb->local_port == dest &&
-   (ip_addr_isany(&pcb->remote_ip) ||
-    ip_addr_cmp(&(pcb->remote_ip), &(iphdr->src))) &&
-   (ip_addr_isany(&pcb->local_ip) ||
-    ip_addr_cmp(&(pcb->local_ip), &(iphdr->dest)))) {
-  pcb_cache = pcb;
-        break;
-        }
-    }
-
-    if (pcb == NULL) {
-      for(pcb = udp_pcbs; pcb != NULL; pcb = pcb->next) {
-  if (pcb->remote_port == 0 &&
-     pcb->local_port == dest &&
-     (ip_addr_isany(&pcb->remote_ip) ||
-      ip_addr_cmp(&(pcb->remote_ip), &(iphdr->src))) &&
-     (ip_addr_isany(&pcb->local_ip) ||
-      ip_addr_cmp(&(pcb->local_ip), &(iphdr->dest)))) {
-        break;
-        }
-      }
-    }
-  }
-
-  PERF_STOP("udp_lookup");
-
-  if (pcb != NULL) {
-    return 1;
-  }
-  else {
-    return 1;
-  }
-}
-#endif /* LWIP_DEBUG */
 /**
  * Process an incoming UDP datagram.
  *
@@ -159,7 +89,7 @@ udp_input(struct pbuf *p, struct netif *inp)
   struct ip_hdr *iphdr;
   u16_t src, dest;
 
-#ifdef SO_REUSE
+#if SO_REUSE
   struct udp_pcb *pcb_temp;
   int reuse = 0;
   int reuse_port_1 = 0;
@@ -198,7 +128,7 @@ udp_input(struct pbuf *p, struct netif *inp)
     ip4_addr1(&iphdr->src), ip4_addr2(&iphdr->src),
     ip4_addr3(&iphdr->src), ip4_addr4(&iphdr->src), ntohs(udphdr->src)));
 
-#ifdef SO_REUSE
+#if SO_REUSE
   pcb_temp = udp_pcbs;
   
  again_1:
@@ -228,7 +158,7 @@ udp_input(struct pbuf *p, struct netif *inp)
        (ip_addr_isany(&pcb->local_ip) ||
        /* PCB local IP address matches UDP destination IP address? */
         ip_addr_cmp(&(pcb->local_ip), &(iphdr->dest)))) {
-#ifdef SO_REUSE
+#if SO_REUSE
       if(pcb->so_options & SOF_REUSEPORT) {
         if(reuse) {
           /* We processed one PCB already */
@@ -257,7 +187,7 @@ udp_input(struct pbuf *p, struct netif *inp)
     /* Iterate through the UDP PCB list for a pcb that matches
        the local address. */
 
-#ifdef SO_REUSE
+#if SO_REUSE
     pcb_temp = udp_pcbs;
     
   again_2:
@@ -279,7 +209,7 @@ udp_input(struct pbuf *p, struct netif *inp)
         (ip_addr_isany(&pcb->local_ip) ||
         /* ...matching interface address? */
         ip_addr_cmp(&(pcb->local_ip), &(iphdr->dest)))) {
-#ifdef SO_REUSE
+#if SO_REUSE
         if(pcb->so_options & SOF_REUSEPORT) {
           if(reuse) {
             /* We processed one PCB already */
@@ -316,6 +246,7 @@ udp_input(struct pbuf *p, struct netif *inp)
     if (IPH_PROTO(iphdr) == IP_PROTO_UDPLITE) {
 #endif /* IPv4 */
       /* Do the UDP Lite checksum */
+#if CHECKSUM_CHECK_UDP
       if (inet_chksum_pseudo(p, (struct ip_addr *)&(iphdr->src),
          (struct ip_addr *)&(iphdr->dest),
          IP_PROTO_UDPLITE, ntohs(udphdr->len)) != 0) {
@@ -326,7 +257,9 @@ udp_input(struct pbuf *p, struct netif *inp)
   pbuf_free(p);
   goto end;
       }
+#endif
     } else {
+#if CHECKSUM_CHECK_UDP
       if (udphdr->chksum != 0) {
   if (inet_chksum_pseudo(p, (struct ip_addr *)&(iphdr->src),
        (struct ip_addr *)&(iphdr->dest),
@@ -340,12 +273,13 @@ udp_input(struct pbuf *p, struct netif *inp)
     goto end;
   }
       }
+#endif
     }
     pbuf_header(p, -UDP_HLEN);
     if (pcb != NULL) {
       snmp_inc_udpindatagrams();
       pcb->recv(pcb->recv_arg, pcb, p, &(iphdr->src), src);
-#ifdef SO_REUSE
+#if SO_REUSE
       /* First socket should receive now */
       if(reuse_port_1 || reuse_port_2) {
         /* We want to search on next socket after receiving */
@@ -365,7 +299,7 @@ udp_input(struct pbuf *p, struct netif *inp)
       }
 #endif /* SO_REUSE */ 
     } else {
-#ifdef SO_REUSE
+#if SO_REUSE
       if(reuse) {
         LWIP_DEBUGF(UDP_DEBUG, ("udp_input: freeing PBUF with reference counter set to %i\n", p->ref));
         pbuf_free(p);
@@ -375,10 +309,10 @@ udp_input(struct pbuf *p, struct netif *inp)
       LWIP_DEBUGF(UDP_DEBUG | DBG_TRACE, ("udp_input: not for us.\n"));
 
       /* No match was found, send ICMP destination port unreachable unless
-   destination address was broadcast/multicast. */
+      destination address was broadcast/multicast. */
 
-      if (!ip_addr_isbroadcast(&iphdr->dest, &inp->netmask) &&
-   !ip_addr_ismulticast(&iphdr->dest)) {
+      if (!ip_addr_isbroadcast(&iphdr->dest, inp) &&
+          !ip_addr_ismulticast(&iphdr->dest)) {
 
   /* adjust pbuf pointer */
   p->payload = iphdr;
@@ -396,6 +330,46 @@ udp_input(struct pbuf *p, struct netif *inp)
 
   PERF_STOP("udp_input");
 }
+
+/**
+ * Send data to a specified address using UDP.
+ *
+ * @param pcb UDP PCB used to send the data.
+ * @param pbuf chain of pbuf's to be sent.
+ * @param dst_ip Destination IP address.
+ * @param dst_port Destination UDP port.
+ *
+ * If the PCB already has a remote address association, it will
+ * be restored after the data is sent.
+ * 
+ * @return lwIP error code.
+ * - ERR_OK. Successful. No error occured.
+ * - ERR_MEM. Out of memory.
+ * - ERR_RTE. Could not find route to destination address.
+ *
+ * @see udp_disconnect() udp_send()
+ */
+err_t
+udp_sendto(struct udp_pcb *pcb, struct pbuf *p,
+  struct ip_addr *dst_ip, u16_t dst_port)
+{
+  err_t err;
+  struct ip_addr pcb_remote_ip;
+  u16_t pcb_remote_port;
+  /* remember remote peer address of PCB */
+  pcb_remote_ip.addr = pcb->remote_ip.addr;
+  pcb_remote_port = pcb->remote_port;
+  /* copy packet destination address to PCB remote peer address */
+  pcb->remote_ip.addr = dst_ip->addr;
+  pcb->remote_port = dst_port;
+  /* send to the packet destination address */
+  err = udp_send(pcb, p);
+  /* reset PCB remote peer address */
+  pcb->remote_ip.addr = pcb_remote_ip.addr;
+  pcb->remote_port = pcb_remote_port;
+  return err;
+}
+
 /**
  * Send data using UDP.
  *
@@ -405,10 +379,9 @@ udp_input(struct pbuf *p, struct netif *inp)
  * @return lwIP error code.
  * - ERR_OK. Successful. No error occured.
  * - ERR_MEM. Out of memory.
- * - ERR_USE. The specified ipaddr and port are already bound to by
- * another UDP PCB.
+ * - ERR_RTE. Could not find route to destination address.
  *
- * @see udp_disconnect()
+ * @see udp_disconnect() udp_sendto()
  */
 err_t
 udp_send(struct udp_pcb *pcb, struct pbuf *p)
@@ -433,7 +406,7 @@ udp_send(struct udp_pcb *pcb, struct pbuf *p)
 
   /* not enough space to add an UDP header to first pbuf in given p chain? */
   if (pbuf_header(p, UDP_HLEN)) {
-    /* allocate header in new pbuf */
+    /* allocate header in a seperate new pbuf */
     q = pbuf_alloc(PBUF_IP, UDP_HLEN, PBUF_RAM);
     /* new header pbuf could not be allocated? */
     if (q == NULL) {
@@ -445,23 +418,25 @@ udp_send(struct udp_pcb *pcb, struct pbuf *p)
     /* { first pbuf q points to header pbuf } */
     LWIP_DEBUGF(UDP_DEBUG, ("udp_send: added header pbuf %p before given pbuf %p\n", (void *)q, (void *)p));
   /* adding a header within p succeeded */
-  }  else {
+  } else {
     /* first pbuf q equals given pbuf */
     q = p;
     LWIP_DEBUGF(UDP_DEBUG, ("udp_send: added header in given pbuf %p\n", (void *)p));
   }
-  /* { q now represents the packet to be sent */
+  /* { q now represents the packet to be sent } */
   udphdr = q->payload;
   udphdr->src = htons(pcb->local_port);
   udphdr->dest = htons(pcb->remote_port);
-  udphdr->chksum = 0x0000;
+  /* in UDP, 0 checksum means 'no checksum' */
+  udphdr->chksum = 0x0000; 
 
+  /* find the outgoing network interface for this packet */
   if ((netif = ip_route(&(pcb->remote_ip))) == NULL) {
     LWIP_DEBUGF(UDP_DEBUG | 1, ("udp_send: No route to 0x%lx\n", pcb->remote_ip.addr));
     UDP_STATS_INC(udp.rterr);
     return ERR_RTE;
   }
-  /* using IP_ANY_ADDR? */
+  /* PCB local address is IP_ANY_ADDR? */
   if (ip_addr_isany(&pcb->local_ip)) {
     /* use outgoing network interface IP address as source address */
     src_ip = &(netif->ip_addr);
@@ -478,34 +453,44 @@ udp_send(struct udp_pcb *pcb, struct pbuf *p)
     /* set UDP message length in UDP header */
     udphdr->len = htons(pcb->chksum_len);
     /* calculate checksum */
+#if CHECKSUM_GEN_UDP
     udphdr->chksum = inet_chksum_pseudo(q, src_ip, &(pcb->remote_ip),
           IP_PROTO_UDP, pcb->chksum_len);
     /* chksum zero must become 0xffff, as zero means 'no checksum' */
     if (udphdr->chksum == 0x0000) udphdr->chksum = 0xffff;
+#else
+    udphdr->chksum = 0x0000;
+#endif
     /* output to IP */
+    LWIP_DEBUGF(UDP_DEBUG, ("udp_send: ip_output_if (,,,,IP_PROTO_UDPLITE,)\n"));
     err = ip_output_if (q, src_ip, &pcb->remote_ip, pcb->ttl, pcb->tos, IP_PROTO_UDPLITE, netif);    
-    snmp_inc_udpoutdatagrams();
+  /* UDP */
   } else {
     LWIP_DEBUGF(UDP_DEBUG, ("udp_send: UDP packet length %u\n", q->tot_len));
     udphdr->len = htons(q->tot_len);
     /* calculate checksum */
+#if CHECKSUM_GEN_UDP
     if ((pcb->flags & UDP_FLAGS_NOCHKSUM) == 0) {
       udphdr->chksum = inet_chksum_pseudo(q, src_ip, &pcb->remote_ip, IP_PROTO_UDP, q->tot_len);
       /* chksum zero must become 0xffff, as zero means 'no checksum' */
       if (udphdr->chksum == 0x0000) udphdr->chksum = 0xffff;
     }
+#else
+    udphdr->chksum = 0x0000;
+#endif
     LWIP_DEBUGF(UDP_DEBUG, ("udp_send: UDP checksum 0x%04x\n", udphdr->chksum));
-    snmp_inc_udpoutdatagrams();
     LWIP_DEBUGF(UDP_DEBUG, ("udp_send: ip_output_if (,,,,IP_PROTO_UDP,)\n"));
     /* output to IP */
     err = ip_output_if(q, src_ip, &pcb->remote_ip, pcb->ttl, pcb->tos, IP_PROTO_UDP, netif);    
   }
+  /* TODO: must this be increased even if error occured? */
+  snmp_inc_udpoutdatagrams();
 
-  /* did we chain a header earlier? */
+  /* did we chain a seperate header pbuf earlier? */
   if (q != p) {
-    /* free the header */
-    /* p is also still referenced by the caller, and will live on */
-    pbuf_free(q);
+    /* free the header pbuf */
+    pbuf_free(q); q = NULL;
+    /* { p is still referenced by the caller, and will live on } */
   }
 
   UDP_STATS_INC(udp.xmit);
@@ -532,7 +517,7 @@ udp_bind(struct udp_pcb *pcb, struct ip_addr *ipaddr, u16_t port)
 {
   struct udp_pcb *ipcb;
   u8_t rebind;
-#ifdef SO_REUSE
+#if SO_REUSE
   int reuse_port_all_set = 1;
 #endif /* SO_REUSE */
   LWIP_DEBUGF(UDP_DEBUG | DBG_TRACE | 3, ("udp_bind(ipaddr = "));
@@ -550,7 +535,7 @@ udp_bind(struct udp_pcb *pcb, struct ip_addr *ipaddr, u16_t port)
       rebind = 1;
     }
 
-#ifndef SO_REUSE
+#if SO_REUSE == 0
 /* this code does not allow upper layer to share a UDP port for
    listening to broadcast or multicast traffic (See SO_REUSE_ADDR and
    SO_REUSE_PORT under *BSD). TODO: See where it fits instead, OR
@@ -604,7 +589,7 @@ udp_bind(struct udp_pcb *pcb, struct ip_addr *ipaddr, u16_t port)
 
   }
 
-#ifdef SO_REUSE
+#if SO_REUSE
   /* If SOF_REUSEPORT isn't set in all PCB's bound to specified port and local address specified then 
      {IP, port} can't be reused. */
   if(!reuse_port_all_set) {
@@ -717,6 +702,10 @@ udp_connect(struct udp_pcb *pcb, struct ip_addr *ipaddr, u16_t port)
 void
 udp_disconnect(struct udp_pcb *pcb)
 {
+  /* reset remote address association */
+  ip_addr_set(&pcb->remote_ip, IP_ADDR_ANY);
+  pcb->remote_port = 0;
+  /* mark PCB as unconnected */
   pcb->flags &= ~UDP_FLAGS_CONNECTED;
 }
 

src/include/ipv4/lwip/icmp.h

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2001-2003 Swedish Institute of Computer Science.
+ * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
  * All rights reserved. 
  * 
  * Redistribution and use in source and binary forms, with or without modification, 
@@ -37,6 +37,7 @@
 #include "lwip/opt.h"
 #include "lwip/pbuf.h"
 
+#include "lwip/ip_addr.h"
 #include "lwip/netif.h"
 
 #define ICMP_ER 0      /* echo reply */

src/include/ipv4/lwip/inet.h

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2001-2003 Swedish Institute of Computer Science.
+ * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
  * All rights reserved. 
  * 
  * Redistribution and use in source and binary forms, with or without modification, 
@@ -46,7 +46,7 @@ u16_t inet_chksum_pseudo(struct pbuf *p,
 
 u32_t inet_addr(const char *cp);
 int inet_aton(const char *cp, struct in_addr *addr);
-u8_t *inet_ntoa(u32_t addr); /* returns ptr to static buffer; not reentrant! */
+char *inet_ntoa(struct in_addr addr); /* returns ptr to static buffer; not reentrant! */
 
 #ifdef htons
 #undef htons

src/include/ipv4/lwip/ip.h

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2001-2003 Swedish Institute of Computer Science.
+ * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
  * All rights reserved. 
  * 
  * Redistribution and use in source and binary forms, with or without modification, 
@@ -43,7 +43,6 @@
 struct netif;
 
 void ip_init(void);
-u8_t ip_lookup(void *header, struct netif *inp);
 struct netif *ip_route(struct ip_addr *dest);
 err_t ip_input(struct pbuf *p, struct netif *inp);
 err_t ip_output(struct pbuf *p, struct ip_addr *src, struct ip_addr *dest,

src/include/ipv4/lwip/ip_addr.h

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2001-2003 Swedish Institute of Computer Science.
+ * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without modification,
@@ -34,6 +34,40 @@
 
 #include "lwip/arch.h"
 
+#ifdef PACK_STRUCT_USE_INCLUDES
+#  include "arch/bpstruct.h"
+#endif
+PACK_STRUCT_BEGIN
+struct ip_addr {
+  PACK_STRUCT_FIELD(u32_t addr);
+} PACK_STRUCT_STRUCT;
+PACK_STRUCT_END
+#ifdef PACK_STRUCT_USE_INCLUDES
+#  include "arch/epstruct.h"
+#endif
+
+#ifdef PACK_STRUCT_USE_INCLUDES
+#  include "arch/bpstruct.h"
+#endif
+PACK_STRUCT_BEGIN
+struct ip_addr2 {
+  PACK_STRUCT_FIELD(u16_t addrw[2]);
+} PACK_STRUCT_STRUCT;
+PACK_STRUCT_END
+#ifdef PACK_STRUCT_USE_INCLUDES
+#  include "arch/epstruct.h"
+#endif
+
+/* For compatibility with BSD code */
+struct in_addr {
+  u32_t s_addr;
+};
+
+struct netif;
+
+extern const struct ip_addr ip_addr_any;
+extern const struct ip_addr ip_addr_broadcast;
+
 /** IP_ADDR_ can be used as a fixed IP address
  *  for the wildcard and the broadcast address
  */
@@ -76,32 +110,13 @@
 
 #define IN_LOOPBACKNET      127         /* official! */
 
-#ifdef PACK_STRUCT_USE_INCLUDES
-#  include "arch/bpstruct.h"
-#endif
-PACK_STRUCT_BEGIN
-struct ip_addr {
-  PACK_STRUCT_FIELD(u32_t addr);
-} PACK_STRUCT_STRUCT;
-PACK_STRUCT_END
-#ifdef PACK_STRUCT_USE_INCLUDES
-#  include "arch/epstruct.h"
-#endif
-
-/* For compatibility with BSD code */
-struct in_addr {
-  u32_t s_addr;
-};
-
-extern const struct ip_addr ip_addr_any;
-extern const struct ip_addr ip_addr_broadcast;
 
 #define IP4_ADDR(ipaddr, a,b,c,d) (ipaddr)->addr = htonl(((u32_t)(a & 0xff) << 24) | ((u32_t)(b & 0xff) << 16) | \
                                                          ((u32_t)(c & 0xff) << 8) | (u32_t)(d & 0xff))
 
 #define ip_addr_set(dest, src) (dest)->addr = \
                                ((src) == NULL? 0:\
-                               ((struct ip_addr *)src)->addr)
+                               (src)->addr)
 #define ip_addr_maskcmp(addr1, addr2, mask) (((addr1)->addr & \
                                               (mask)->addr) == \
                                              ((addr2)->addr & \
@@ -110,10 +125,7 @@ extern const struct ip_addr ip_addr_broadcast;
 
 #define ip_addr_isany(addr1) ((addr1) == NULL || (addr1)->addr == 0)
 
-#define ip_addr_isbroadcast(addr1, mask) (((((addr1)->addr) & ~((mask)->addr)) == \
-           (0xffffffff & ~((mask)->addr))) || \
-                                         ((addr1)->addr == 0xffffffff) || \
-                                         ((addr1)->addr == 0x00000000))
+u8_t ip_addr_isbroadcast(struct ip_addr *, struct netif *);
 
 #define ip_addr_ismulticast(addr1) (((addr1)->addr & ntohl(0xf0000000)) == ntohl(0xe0000000))
 

src/include/ipv4/lwip/ip_frag.h

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2001-2003 Swedish Institute of Computer Science.
+ * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
  * All rights reserved. 
  * 
  * Redistribution and use in source and binary forms, with or without modification, 

src/include/ipv6/lwip/icmp.h

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2001-2003 Swedish Institute of Computer Science.
+ * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
  * All rights reserved. 
  * 
  * Redistribution and use in source and binary forms, with or without modification, 

src/include/ipv6/lwip/inet.h

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2001-2003 Swedish Institute of Computer Science.
+ * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
  * All rights reserved. 
  * 
  * Redistribution and use in source and binary forms, with or without modification, 

src/include/ipv6/lwip/ip.h

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2001-2003 Swedish Institute of Computer Science.
+ * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
  * All rights reserved. 
  * 
  * Redistribution and use in source and binary forms, with or without modification, 

src/include/ipv6/lwip/ip_addr.h

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2001-2003 Swedish Institute of Computer Science.
+ * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
  * All rights reserved. 
  * 
  * Redistribution and use in source and binary forms, with or without modification, 

src/include/lwip/api.h

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2001-2003 Swedish Institute of Computer Science.
+ * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
  * All rights reserved. 
  * 
  * Redistribution and use in source and binary forms, with or without modification, 

src/include/lwip/api_msg.h

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2001-2003 Swedish Institute of Computer Science.
+ * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
  * All rights reserved. 
  * 
  * Redistribution and use in source and binary forms, with or without modification, 

src/include/lwip/arch.h

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2001-2003 Swedish Institute of Computer Science.
+ * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
  * All rights reserved. 
  * 
  * Redistribution and use in source and binary forms, with or without modification, 

src/include/lwip/debug.h

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2001-2003 Swedish Institute of Computer Science.
+ * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
  * All rights reserved. 
  * 
  * Redistribution and use in source and binary forms, with or without modification, 

src/include/lwip/def.h

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2001-2003 Swedish Institute of Computer Science.
+ * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
  * All rights reserved. 
  * 
  * Redistribution and use in source and binary forms, with or without modification, 

src/include/lwip/dhcp.h

@@ -70,10 +70,10 @@ struct dhcp_msg
   PACK_STRUCT_FIELD(u32_t xid);
   PACK_STRUCT_FIELD(u16_t secs);
   PACK_STRUCT_FIELD(u16_t flags);
-  PACK_STRUCT_FIELD(u32_t ciaddr);
-  PACK_STRUCT_FIELD(u32_t yiaddr);
-  PACK_STRUCT_FIELD(u32_t siaddr);
-  PACK_STRUCT_FIELD(u32_t giaddr);
+  PACK_STRUCT_FIELD(struct ip_addr ciaddr);
+  PACK_STRUCT_FIELD(struct ip_addr yiaddr);
+  PACK_STRUCT_FIELD(struct ip_addr siaddr);
+  PACK_STRUCT_FIELD(struct ip_addr giaddr);
 #define DHCP_CHADDR_LEN 16U
   PACK_STRUCT_FIELD(u8_t chaddr[DHCP_CHADDR_LEN]);
 #define DHCP_SNAME_LEN 64U

src/include/lwip/err.h

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2001-2003 Swedish Institute of Computer Science.
+ * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
  * All rights reserved. 
  * 
  * Redistribution and use in source and binary forms, with or without modification, 
@@ -59,6 +59,7 @@ typedef s8_t err_t;
 #define ERR_USE  -10     /* Address in use.          */
 
 #define ERR_IF   -11     /* Low-level netif error    */
+#define ERR_ISCONN -12   /* Already connected.       */
 
 
 #ifdef LWIP_DEBUG

src/include/lwip/mem.h

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2001-2003 Swedish Institute of Computer Science.
+ * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
  * All rights reserved. 
  * 
  * Redistribution and use in source and binary forms, with or without modification, 

src/include/lwip/memp.h

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2001-2003 Swedish Institute of Computer Science.
+ * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
  * All rights reserved. 
  * 
  * Redistribution and use in source and binary forms, with or without modification, 

src/include/lwip/netif.h

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2001-2003 Swedish Institute of Computer Science.
+ * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
  * All rights reserved. 
  * 
  * Redistribution and use in source and binary forms, with or without modification, 
@@ -40,7 +40,9 @@
 
 #include "lwip/inet.h"
 #include "lwip/pbuf.h"
-#include "lwip/dhcp.h"
+#if LWIP_DHCP
+#  include "lwip/dhcp.h"
+#endif
 
 /** must be the maximum of all used hardware address lengths
     across all types of interfaces in use */
@@ -50,7 +52,8 @@
 
 /** whether the network interface is 'up'. this is
  * a software flag used to control whether this network
- * interface is enabled and processes traffic */
+ * interface is enabled and processes traffic.
+ * TODO: who should act on this flag, lwIP stack or driver?? */
 #define NETIF_FLAG_UP 0x1U
 /** if set, the netif has broadcast capability */
 #define NETIF_FLAG_BROADCAST 0x2U
@@ -59,35 +62,36 @@
 /** if set, the interface is configured using DHCP */
 #define NETIF_FLAG_DHCP 0x08U
 /** if set, the interface has an active link
- *  (set by the interface) */
+ *  (set by the network interface driver) */
 #define NETIF_FLAG_LINK_UP 0x10U
 
-/** generic data structure used for all lwIP network interfaces */
+/** Generic data structure used for all lwIP network interfaces.
+ *  The following fields should be filled in by the initialization
+ *  function for the device driver: hwaddr_len, hwaddr[], mtu, flags */
+
 struct netif {
   /** pointer to next in linked list */
   struct netif *next;
-  /** The following fields should be filled in by the
-      initialization function for the device driver. */
-  
+
   /** IP address configuration in network byte order */
   struct ip_addr ip_addr;
   struct ip_addr netmask;
   struct ip_addr gw;
 
   /** This function is called by the network device driver
-      to pass a packet up the TCP/IP stack. */
+   *  to pass a packet up the TCP/IP stack. */
   err_t (* input)(struct pbuf *p, struct netif *inp);
   /** This function is called by the IP module when it wants
-      to send a packet on the interface. This function typically
-      first resolves the hardware address, then sends the packet. */
+   *  to send a packet on the interface. This function typically
+   *  first resolves the hardware address, then sends the packet. */
   err_t (* output)(struct netif *netif, struct pbuf *p,
        struct ip_addr *ipaddr);
   /** This function is called by the ARP module when it wants
-      to send a packet on the interface. This function outputs
-      the pbuf as-is on the link medium. */
+   *  to send a packet on the interface. This function outputs
+   *  the pbuf as-is on the link medium. */
   err_t (* linkoutput)(struct netif *netif, struct pbuf *p);
   /** This field can be set by the device driver and could point
-      to state information for the device. */
+   *  to state information for the device. */
   void *state;
 #if LWIP_DHCP
   /** the DHCP client state information for this netif */
@@ -99,12 +103,12 @@ struct netif {
   unsigned char hwaddr[NETIF_MAX_HWADDR_LEN];
   /** maximum transfer unit (in bytes) */
   u16_t mtu;
+  /** flags (see NETIF_FLAG_ above) */
+  u8_t flags;
   /** descriptive abbreviation */
   char name[2];
   /** number of this interface */
   u8_t num;
-  /** NETIF_FLAG_* */
-  u8_t flags;
 };
 
 /** The list of network interfaces. */

src/include/lwip/opt.h

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2001-2003 Swedish Institute of Computer Science.
+ * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
  * All rights reserved. 
  * 
  * Redistribution and use in source and binary forms, with or without modification, 
@@ -67,6 +67,10 @@ a lot of data that needs to be copied, this should be set high. */
 #define MEM_SIZE                        1600
 #endif
 
+#ifndef MEMP_SANITY_CHECK
+#define MEMP_SANITY_CHECK		0
+#endif
+
 /* MEMP_NUM_PBUF: the number of memp struct pbufs. If the application
    sends a lot of data out of ROM (or other static memory), this
    should be set high. */
@@ -171,14 +175,10 @@ a lot of data that needs to be copied, this should be set high. */
 #ifndef ARP_QUEUE_FIRST
 #define ARP_QUEUE_FIRST                 0
 #endif
-/**
- * If defined to 1, cache entries are updated or added for every kind of ARP traffic
- * or broadcast IP traffic. Recommended for routers.
- * If defined to 0, only existing cache entries are updated. Entries are added when
- * lwIP is sending to them. Recommended for embedded devices.
- */
-#ifndef ETHARP_ALWAYS_INSERT
-#define ETHARP_ALWAYS_INSERT            1
+
+/* This option is removed to comply with the ARP standard */
+#ifdef ETHARP_ALWAYS_INSERT
+#error ETHARP_ALWAYS_INSERT option is deprecated. Remove it from your lwipopts.h.
 #endif
 
 /* ---------- IP options ---------- */
@@ -218,6 +218,10 @@ a lot of data that needs to be copied, this should be set high. */
 
 /* ---------- RAW options ---------- */
 
+#ifndef LWIP_RAW
+#define LWIP_RAW                        1
+#endif
+
 #ifndef RAW_TTL
 #define RAW_TTL                        255
 #endif
@@ -348,6 +352,10 @@ a lot of data that needs to be copied, this should be set high. */
 
 #if LWIP_STATS
 
+#ifndef LWIP_STATS_DISPLAY
+#define LWIP_STATS_DISPLAY 0
+#endif
+
 #ifndef LINK_STATS
 #define LINK_STATS	1
 #endif
@@ -405,6 +413,7 @@ a lot of data that needs to be copied, this should be set high. */
 #define PBUF_STATS	0
 #define SYS_STATS	0
 #define RAW_STATS	0
+#define LWIP_STATS_DISPLAY	0
 
 #endif /* LWIP_STATS */
 
@@ -495,6 +504,31 @@ a lot of data that needs to be copied, this should be set high. */
 
 #endif /* PPP_SUPPORT */
 
+/* checksum options - set to zero for hardware checksum support */
+
+#ifndef CHECKSUM_GEN_IP
+#define CHECKSUM_GEN_IP                 1
+#endif
+ 
+#ifndef CHECKSUM_GEN_UDP
+#define CHECKSUM_GEN_UDP                1
+#endif
+ 
+#ifndef CHECKSUM_GEN_TCP
+#define CHECKSUM_GEN_TCP                1
+#endif
+ 
+#ifndef CHECKSUM_CHECK_IP
+#define CHECKSUM_CHECK_IP               1
+#endif
+ 
+#ifndef CHECKSUM_CHECK_UDP
+#define CHECKSUM_CHECK_UDP              1
+#endif
+
+#ifndef CHECKSUM_CHECK_TCP
+#define CHECKSUM_CHECK_TCP              1
+#endif
 
 /* Debugging options all default to off */
 

src/include/lwip/pbuf.h

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2001-2003 Swedish Institute of Computer Science.
+ * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
  * All rights reserved. 
  * 
  * Redistribution and use in source and binary forms, with or without modification, 
@@ -53,8 +53,8 @@ typedef enum {
   PBUF_POOL
 } pbuf_flag;
 
-/* Definitions for the pbuf flag field (these are not the flags that
-   are passed to pbuf_alloc()). */
+/* Definitions for the pbuf flag field. These are NOT the flags that
+ * are passed to pbuf_alloc(). */
 #define PBUF_FLAG_RAM   0x00U    /* Flags that pbuf data is stored in RAM */
 #define PBUF_FLAG_ROM   0x01U    /* Flags that pbuf data is stored in ROM */
 #define PBUF_FLAG_POOL  0x02U    /* Flags that the pbuf comes from the pbuf pool */
@@ -79,10 +79,10 @@ struct pbuf {
    */
   u16_t tot_len;
   
-  /* length of this buffer */
+  /** length of this buffer */
   u16_t len;  
 
-  /* flags telling the type of pbuf */
+  /** flags telling the type of pbuf, see PBUF_FLAG_ */
   u16_t flags;
   
   /**
@@ -94,11 +94,6 @@ struct pbuf {
   
 };
 
-/* pbuf_init():
-
-   Initializes the pbuf module. The num parameter determines how many
-   pbufs that should be allocated to the pbuf pool, and the size
-   parameter specifies the size of the data allocated to those.  */
 void pbuf_init(void);
 
 struct pbuf *pbuf_alloc(pbuf_layer l, u16_t size, pbuf_flag flag);
@@ -112,5 +107,7 @@ void pbuf_cat(struct pbuf *h, struct pbuf *t);
 void pbuf_chain(struct pbuf *h, struct pbuf *t);
 struct pbuf *pbuf_take(struct pbuf *f);
 struct pbuf *pbuf_dechain(struct pbuf *p);
+void pbuf_queue(struct pbuf *p, struct pbuf *n);
+struct pbuf * pbuf_dequeue(struct pbuf *p);
 
 #endif /* __LWIP_PBUF_H__ */

src/include/lwip/raw.h

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2001-2003 Swedish Institute of Computer Science.
+ * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without modification,
@@ -46,7 +46,7 @@ struct raw_pcb {
 
   u16_t protocol;
 
-  int (* recv)(void *arg, struct raw_pcb *pcb, struct pbuf *p,
+  u8_t (* recv)(void *arg, struct raw_pcb *pcb, struct pbuf *p,
     struct ip_addr *addr);
   void *recv_arg;
 };
@@ -59,15 +59,15 @@ err_t            raw_bind       (struct raw_pcb *pcb, struct ip_addr *ipaddr);
 err_t            raw_connect    (struct raw_pcb *pcb, struct ip_addr *ipaddr);
 
 void             raw_recv       (struct raw_pcb *pcb,
-                                 int (* recv)(void *arg, struct raw_pcb *pcb,
+                                 u8_t (* recv)(void *arg, struct raw_pcb *pcb,
                                               struct pbuf *p,
                                               struct ip_addr *addr),
                                  void *recv_arg);
-err_t            raw_send_to    (struct raw_pcb *pcb, struct pbuf *p, struct ip_addr *ipaddr);
+err_t            raw_sendto    (struct raw_pcb *pcb, struct pbuf *p, struct ip_addr *ipaddr);
 err_t            raw_send       (struct raw_pcb *pcb, struct pbuf *p);
 
 /* The following functions are the lower layer interface to RAW. */
-int              raw_input      (struct pbuf *p, struct netif *inp);
+u8_t              raw_input      (struct pbuf *p, struct netif *inp);
 void             raw_init       (void);
 
 

src/include/lwip/sio.h

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2001-2003 Swedish Institute of Computer Science.
+ * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
  * All rights reserved. 
  * 
  * Redistribution and use in source and binary forms, with or without modification, 

src/include/lwip/sockets.h

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2001-2003 Swedish Institute of Computer Science.
+ * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
  * All rights reserved. 
  * 
  * Redistribution and use in source and binary forms, with or without modification, 
@@ -205,10 +205,16 @@ struct linger {
           unsigned char fd_bits [(FD_SETSIZE+7)/8];
         } fd_set;
 
+/* 
+ * only define this in sockets.c so it does not interfere
+ * with other projects namespaces where timeval is present
+ */ 
+#ifdef LWIP_TIMEVAL_PRIVATE
   struct timeval {
     long    tv_sec;         /* seconds */
     long    tv_usec;        /* and microseconds */
   };
+#endif
 
 #endif
 

src/include/lwip/stats.h

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2001-2003 Swedish Institute of Computer Science.
+ * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
  * All rights reserved. 
  * 
  * Redistribution and use in source and binary forms, with or without modification, 
@@ -144,6 +144,13 @@ void stats_init(void);
 #define LINK_STATS_INC(x)
 #endif
 
+/* Display of statistics */
+#if LWIP_STATS_DISPLAY
+void stats_display(void);
+#else
+#define stats_display()
+#endif
+
 #endif /* __LWIP_STATS_H__ */
 
 

src/include/lwip/sys.h

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2001-2003 Swedish Institute of Computer Science.
+ * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
  * All rights reserved. 
  * 
  * Redistribution and use in source and binary forms, with or without modification, 

src/include/lwip/tcp.h

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2001-2003 Swedish Institute of Computer Science.
+ * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
  * All rights reserved. 
  * 
  * Redistribution and use in source and binary forms, with or without modification, 
@@ -40,8 +40,6 @@
 #include "lwip/ip.h"
 #include "lwip/icmp.h"
 
-#include "lwip/sys.h"
-
 #include "lwip/err.h"
 
 struct tcp_pcb;
@@ -457,7 +455,11 @@ void tcp_timer_needed(void);
 #endif
 
 /* The TCP PCB lists. */
-extern struct tcp_pcb_listen *tcp_listen_pcbs;  /* List of all TCP PCBs in LISTEN state. */
+union tcp_listen_pcbs_t { /* List of all TCP PCBs in LISTEN state. */
+	struct tcp_pcb_listen *listen_pcbs; 
+	struct tcp_pcb *pcbs;
+};
+extern union tcp_listen_pcbs_t tcp_listen_pcbs;
 extern struct tcp_pcb *tcp_active_pcbs;  /* List of all TCP PCBs that are in a
               state in which they accept or send
               data. */

src/include/lwip/tcpip.h

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2001-2003 Swedish Institute of Computer Science.
+ * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
  * All rights reserved. 
  * 
  * Redistribution and use in source and binary forms, with or without modification, 

src/include/lwip/udp.h

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2001-2003 Swedish Institute of Computer Science.
+ * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
  * All rights reserved. 
  * 
  * Redistribution and use in source and binary forms, with or without modification, 
@@ -84,14 +84,13 @@ void             udp_recv       (struct udp_pcb *pcb,
                  struct ip_addr *addr,
                  u16_t port),
          void *recv_arg);
+err_t            udp_sendto     (struct udp_pcb *pcb, struct pbuf *p, struct ip_addr *dst_ip, u16_t dst_port);
 err_t            udp_send       (struct udp_pcb *pcb, struct pbuf *p);
 
 #define          udp_flags(pcb)  ((pcb)->flags)
 #define          udp_setflags(pcb, f)  ((pcb)->flags = (f))
 
-
 /* The following functions are the lower layer interface to UDP. */
-u8_t             udp_lookup     (struct ip_hdr *iphdr, struct netif *inp);
 void             udp_input      (struct pbuf *p, struct netif *inp);
 void             udp_init       (void);
 

src/include/netif/etharp.h

@@ -1,5 +1,7 @@
 /*
  * Copyright (c) 2001-2003 Swedish Institute of Computer Science.
+ * Copyright (c) 2003-2004 Leon Woestenberg <leon.woestenberg@axon.tv>
+ * Copyright (c) 2003-2004 Axon Digital Design B.V., The Netherlands.
  * All rights reserved. 
  * 
  * Redistribution and use in source and binary forms, with or without modification, 
@@ -33,6 +35,10 @@
 #ifndef __NETIF_ETHARP_H__
 #define __NETIF_ETHARP_H__
 
+#ifndef ETH_PAD_SIZE
+#define ETH_PAD_SIZE 0
+#endif
+
 #include "lwip/pbuf.h"
 #include "lwip/ip_addr.h"
 #include "lwip/netif.h"
@@ -49,6 +55,9 @@ PACK_STRUCT_END
 
 PACK_STRUCT_BEGIN
 struct eth_hdr {
+#if ETH_PAD_SIZE
+  PACK_STRUCT_FIELD(u8_t padding[ETH_PAD_SIZE]);
+#endif
   PACK_STRUCT_FIELD(struct eth_addr dest);
   PACK_STRUCT_FIELD(struct eth_addr src);
   PACK_STRUCT_FIELD(u16_t type);
@@ -64,9 +73,9 @@ struct etharp_hdr {
   PACK_STRUCT_FIELD(u16_t _hwlen_protolen);
   PACK_STRUCT_FIELD(u16_t opcode);
   PACK_STRUCT_FIELD(struct eth_addr shwaddr);
-  PACK_STRUCT_FIELD(struct ip_addr sipaddr);
+  PACK_STRUCT_FIELD(struct ip_addr2 sipaddr);
   PACK_STRUCT_FIELD(struct eth_addr dhwaddr);
-  PACK_STRUCT_FIELD(struct ip_addr dipaddr);
+  PACK_STRUCT_FIELD(struct ip_addr2 dipaddr);
 } PACK_STRUCT_STRUCT;
 PACK_STRUCT_END
 
@@ -86,10 +95,10 @@ struct ethip_hdr {
 
 void etharp_init(void);
 void etharp_tmr(void);
-struct pbuf *etharp_ip_input(struct netif *netif, struct pbuf *p);
-struct pbuf *etharp_arp_input(struct netif *netif, struct eth_addr *ethaddr,
+void etharp_ip_input(struct netif *netif, struct pbuf *p);
+void etharp_arp_input(struct netif *netif, struct eth_addr *ethaddr,
          struct pbuf *p);
-struct pbuf *etharp_output(struct netif *netif, struct ip_addr *ipaddr,
+err_t etharp_output(struct netif *netif, struct ip_addr *ipaddr,
          struct pbuf *q);
 err_t etharp_query(struct netif *netif, struct ip_addr *ipaddr, struct pbuf *q);
 

src/include/netif/loopif.h

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2001-2003 Swedish Institute of Computer Science.
+ * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
  * All rights reserved. 
  * 
  * Redistribution and use in source and binary forms, with or without modification, 

src/netif/etharp.c

@@ -6,12 +6,16 @@
  * to a physical address when sending a packet, and the second part answers
  * requests from other machines for our physical address.
  *
- * This implementation complies with RFC 826 (Ethernet ARP) and supports
- * Gratuitious ARP from RFC3220 (IP Mobility Support for IPv4) section 4.6.
+ * This implementation complies with RFC 826 (Ethernet ARP). It supports
+ * Gratuitious ARP from RFC3220 (IP Mobility Support for IPv4) section 4.6
+ * if an interface calls etharp_query(our_netif, its_ip_addr, NULL) upon
+ * address change.
  */
 
 /*
  * Copyright (c) 2001-2003 Swedish Institute of Computer Science.
+ * Copyright (c) 2003-2004 Leon Woestenberg <leon.woestenberg@axon.tv>
+ * Copyright (c) 2003-2004 Axon Digital Design B.V., The Netherlands.
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without modification,
@@ -38,49 +42,8 @@
  *
  * This file is part of the lwIP TCP/IP stack.
  *
- * Author: Adam Dunkels <adam@sics.se>
- *
  */
 
-/**
- * TODO:
- * - pbufs should be sent from the queue once an ARP entry state
- *   goes from PENDING to STABLE.
- * - Non-PENDING entries MUST NOT have queued packets.
- */
-
-/*
- * TODO:
- *
-RFC 3220 4.6          IP Mobility Support for IPv4          January 2002
-
-      -  A Gratuitous ARP [45] is an ARP packet sent by a node in order
-         to spontaneously cause other nodes to update an entry in their
-         ARP cache.  A gratuitous ARP MAY use either an ARP Request or
-         an ARP Reply packet.  In either case, the ARP Sender Protocol
-         Address and ARP Target Protocol Address are both set to the IP
-         address of the cache entry to be updated, and the ARP Sender
-         Hardware Address is set to the link-layer address to which this
-         cache entry should be updated.  When using an ARP Reply packet,
-         the Target Hardware Address is also set to the link-layer
-         address to which this cache entry should be updated (this field
-         is not used in an ARP Request packet).
-
-         In either case, for a gratuitous ARP, the ARP packet MUST be
-         transmitted as a local broadcast packet on the local link.  As
-         specified in [36], any node receiving any ARP packet (Request
-         or Reply) MUST update its local ARP cache with the Sender
-         Protocol and Hardware Addresses in the ARP packet, if the
-         receiving node has an entry for that IP address already in its
-         ARP cache.  This requirement in the ARP protocol applies even
-         for ARP Request packets, and for ARP Reply packets that do not
-         match any ARP Request transmitted by the receiving node [36].
-*
-  My suggestion would be to send a ARP request for our newly obtained
-  address upon configuration of an Ethernet interface.
-
-*/
-
 #include "lwip/opt.h"
 #include "lwip/inet.h"
 #include "netif/etharp.h"
@@ -92,10 +55,13 @@ RFC 3220 4.6          IP Mobility Support for IPv4          January 2002
 #  include "lwip/dhcp.h"
 #endif
 
+/* allows new queueing code to be disabled (0) for regression testing */
+#define ARP_NEW_QUEUE 1
+
 /** the time an ARP entry stays valid after its last update, (120 * 10) seconds = 20 minutes. */
 #define ARP_MAXAGE 120
-/** the time an ARP entry stays pending after first request, (2 * 10) seconds = 20 seconds. */
-#define ARP_MAXPENDING 2
+/** the time an ARP entry stays pending after first request, (1 * 10) seconds = 10 seconds. */
+#define ARP_MAXPENDING 1
 
 #define HWTYPE_ETHERNET 1
 
@@ -112,7 +78,9 @@ RFC 3220 4.6          IP Mobility Support for IPv4          January 2002
 enum etharp_state {
   ETHARP_STATE_EMPTY,
   ETHARP_STATE_PENDING,
-  ETHARP_STATE_STABLE
+  ETHARP_STATE_STABLE,
+  /** @internal convenience transitional state used in etharp_tmr() */
+  ETHARP_STATE_EXPIRED
 };
 
 struct etharp_entry {
@@ -132,12 +100,9 @@ static const struct eth_addr ethbroadcast = {{0xff,0xff,0xff,0xff,0xff,0xff}};
 static struct etharp_entry arp_table[ARP_TABLE_SIZE];
 
 static s8_t find_arp_entry(void);
+/** ask update_arp_entry() to add instead of merely update an ARP entry */
 #define ARP_INSERT_FLAG 1
 static struct pbuf *update_arp_entry(struct netif *netif, struct ip_addr *ipaddr, struct eth_addr *ethaddr, u8_t flags);
-#if ARP_QUEUEING
-static struct pbuf *etharp_enqueue(s8_t i, struct pbuf *q);
-static u8_t etharp_dequeue(s8_t i);
-#endif
 /**
  * Initializes ARP module.
  */
@@ -175,24 +140,27 @@ etharp_tmr(void)
          /* entry has become old? */
         (arp_table[i].ctime >= ARP_MAXAGE)) {
       LWIP_DEBUGF(ETHARP_DEBUG, ("etharp_timer: expired stable entry %u.\n", i));
-      goto empty;
+      arp_table[i].state = ETHARP_STATE_EXPIRED;
     /* an unresolved/pending entry? */
     } else if ((arp_table[i].state == ETHARP_STATE_PENDING) &&
          /* entry unresolved/pending for too long? */
         (arp_table[i].ctime >= ARP_MAXPENDING)) {
       LWIP_DEBUGF(ETHARP_DEBUG, ("etharp_timer: expired pending entry %u.\n", i));
-  empty:
-      /* empty old entry */      
-      arp_table[i].state = ETHARP_STATE_EMPTY;
+      arp_table[i].state = ETHARP_STATE_EXPIRED;
+    }
+    /* clean up entries that have just been expired */
+    if (arp_table[i].state == ETHARP_STATE_EXPIRED) {
 #if ARP_QUEUEING
       /* and empty packet queue */
       if (arp_table[i].p != NULL) {
-        /* remove any queued packet */
+        /* remove all queued packets */
         LWIP_DEBUGF(ETHARP_DEBUG, ("etharp_timer: freeing entry %u, packet queue %p.\n", i, (void *)(arp_table[i].p)));
         pbuf_free(arp_table[i].p);
         arp_table[i].p = NULL;
       }
 #endif
+      /* recycle entry for re-use */      
+      arp_table[i].state = ETHARP_STATE_EMPTY;
     }
   }
 }
@@ -235,82 +203,24 @@ find_arp_entry(void)
     return ERR_MEM;
   }
 
-  /* clean up the recycled stable entry */
+  /* clean up the oldest stable entry (to be recycled) */
   if (arp_table[i].state == ETHARP_STATE_STABLE) {
 #if ARP_QUEUEING
-    /* free packets on queue */
-    etharp_dequeue(i);
+    /* and empty the packet queue */
+    if (arp_table[i].p != NULL) {
+      LWIP_DEBUGF(ETHARP_DEBUG, ("find_arp_entry: freeing entry %u, packet queue %p.\n", i, (void *)(arp_table[i].p)));
+      /* remove all queued packets */
+      pbuf_free(arp_table[i].p);
+      arp_table[i].p = NULL;
+    }
 #endif
     LWIP_DEBUGF(ETHARP_DEBUG | DBG_TRACE, ("find_arp_entry: recycling oldest stable entry %u\n", i));
     arp_table[i].state = ETHARP_STATE_EMPTY;
-    arp_table[i].ctime = 0;
   }
   LWIP_DEBUGF(ETHARP_DEBUG, ("find_arp_entry: returning %u\n", i));
   return i;
 }
 
-#if ARP_QUEUEING
-/*
- * Enqueues a pbuf (chain) on an ARP entry.
- * 
- * Places the pbuf (chain) on the queue (if space allows). The
- * caller may safely free the pbuf (chain) afterwards, as the
- * pbufs will be referenced by the queue and copies are made of
- * pbufs referencing external payloads.
- * 
- * @ i the ARP entry index
- * @arg q the pbuf (chain) to be queued on the ARP entry
- * 
- * @return Returns the new head of queue of the ARP entry.
- * 
- */
-static struct pbuf *
-etharp_enqueue(s8_t i, struct pbuf *q)
-{
-  /* any pbuf to queue? */
-  if (q != NULL) {
-/* queue later packet over earliers? TODO: Implement multiple pbuf queue */
-#if ARP_QUEUE_FIRST == 0
-    /* remove any pbufs on queue */
-    u8_t deq = etharp_dequeue(i);
-    if (deq > 0) LWIP_DEBUGF(ETHARP_DEBUG | DBG_TRACE | 3, ("etharp_query: dequeued %u pbufs from ARP entry %u. Should not occur.\n", deq, i));
-#endif
-    /* packet can be queued? TODO: Implement multiple pbuf queue */
-    if (arp_table[i].p == NULL) {
-      /* copy any PBUF_REF referenced payloads into PBUF_RAM */
-      q = pbuf_take(q);
-      /* add pbuf to queue */
-      arp_table[i].p = q;
-      /* pbuf (chain) now queued, increase the reference count */
-      pbuf_ref(q);
-      LWIP_DEBUGF(ETHARP_DEBUG | DBG_TRACE | DBG_STATE, ("etharp_query: queued packet %p on ARP entry %u.\n", (void *)q, i));
-    }
-  }
-  return arp_table[i].p;
-}
-
-/**
- * Dequeues any pbufs queued on an ARP entry
- * 
- * @return number of pbufs removed from the queue
- * 
- * TODO: decide what is a sensible return value?
- */
-static u8_t
-etharp_dequeue(s8_t i)
-{
-  /* queued packets on a stable entry (work in progress) */
-  if (arp_table[i].p != NULL) {
-  	/* queue no longer references pbuf */
-  	pbuf_free(arp_table[i].p);
-    arp_table[i].p = NULL;
-    return 1;
-  } else {
-    return 0;
-  }
-}
-#endif
-
 /**
  * Update (or insert) a IP/MAC address pair in the ARP cache.
  *
@@ -334,20 +244,22 @@ update_arp_entry(struct netif *netif, struct ip_addr *ipaddr, struct eth_addr *e
   s8_t i, k;
   LWIP_DEBUGF(ETHARP_DEBUG | DBG_TRACE | 3, ("update_arp_entry()\n"));
   LWIP_ASSERT("netif->hwaddr_len != 0", netif->hwaddr_len != 0);
-  LWIP_DEBUGF(ETHARP_DEBUG | DBG_TRACE, ("update_arp_entry: %u.%u.%u.%u - %02x:%02x:%02x:%02x:%02x:%02x\n", ip4_addr1(ipaddr), ip4_addr2(ipaddr), ip4_addr3(ipaddr), ip4_addr4(ipaddr),
-  ethaddr->addr[0], ethaddr->addr[1], ethaddr->addr[2], ethaddr->addr[3], ethaddr->addr[4], ethaddr->addr[5]));
+  LWIP_DEBUGF(ETHARP_DEBUG | DBG_TRACE, ("update_arp_entry: %u.%u.%u.%u - %02x:%02x:%02x:%02x:%02x:%02x\n",
+                                        ip4_addr1(ipaddr), ip4_addr2(ipaddr), ip4_addr3(ipaddr), ip4_addr4(ipaddr), 
+                                        ethaddr->addr[0], ethaddr->addr[1], ethaddr->addr[2],
+                                        ethaddr->addr[3], ethaddr->addr[4], ethaddr->addr[5]));
   /* do not update for 0.0.0.0 addresses */
   if (ipaddr->addr == 0) {
     LWIP_DEBUGF(ETHARP_DEBUG | DBG_TRACE, ("update_arp_entry: will not add 0.0.0.0 to ARP cache\n"));
     return NULL;
   }
-  /* Walk through the ARP mapping table and try to find an entry to
-  update. If none is found, the IP -> MAC address mapping is
-  inserted in the ARP table. */
+  /* Walk through the ARP mapping table and try to find an entry to update.
+   * If none is found, a new IP -> MAC address mapping is inserted. */
   for (i = 0; i < ARP_TABLE_SIZE; ++i) {
     /* Check if the source IP address of the incoming packet matches
-    the IP address in this ARP table entry. */
-    if (ip_addr_cmp(ipaddr, &arp_table[i].ipaddr)) {
+     * the IP address in this ARP table entry. */
+    if (arp_table[i].state != ETHARP_STATE_EMPTY &&
+    	ip_addr_cmp(ipaddr, &arp_table[i].ipaddr)) {
       /* pending entry? */
       if (arp_table[i].state == ETHARP_STATE_PENDING) {
         LWIP_DEBUGF(ETHARP_DEBUG | DBG_TRACE, ("update_arp_entry: pending entry %u goes stable\n", i));
@@ -355,12 +267,8 @@ update_arp_entry(struct netif *netif, struct ip_addr *ipaddr, struct eth_addr *e
         arp_table[i].state = ETHARP_STATE_STABLE;
         /* fall-through to next if */
       }
-      /* stable entry? (possibly just marked to become stable) */
+      /* stable entry? (possibly just marked stable) */
       if (arp_table[i].state == ETHARP_STATE_STABLE) {
-#if ARP_QUEUEING
-        struct pbuf *p;
-        struct eth_hdr *ethhdr;
-#endif
         LWIP_DEBUGF(ETHARP_DEBUG | DBG_TRACE, ("update_arp_entry: updating stable entry %u\n", i));
         /* An old entry found, update this and return. */
         for (k = 0; k < netif->hwaddr_len; ++k) {
@@ -370,52 +278,39 @@ update_arp_entry(struct netif *netif, struct ip_addr *ipaddr, struct eth_addr *e
         arp_table[i].ctime = 0;
 /* this is where we will send out queued packets! */
 #if ARP_QUEUEING
-        /* get the first packet on the queue (if any) */
-        p = arp_table[i].p;
-        /* queued packet present? */
-        while (p != NULL) {
-          struct pbuf *q, *n;
-          /* search for second packet on queue (n) */
-          q = p;
-          while (q->tot_len > q->len) {
-          	/* proceed to next pbuf of this packet */
-          	LWIP_ASSERT("q->next ! NULL", q->next != NULL);
-          	q = q->next;
-          }
-          /* { q = last pbuf of first packet, q->tot_len = q->len } */
-          n = q->next;
-          /* { n = first pbuf of 2nd packet, or NULL if no 2nd packet } */
-          /* terminate the first packet pbuf chain */
-          q->next = NULL;
+        while (arp_table[i].p != NULL) {
+          /* get the first packet on the queue (if any) */
+          struct pbuf *p = arp_table[i].p;
+          /* Ethernet header */
+          struct eth_hdr *ethhdr = p->payload;;
+          /* remember (and reference) remainder of queue */
+          /* note: this will also terminate the p pbuf chain */
+          arp_table[i].p = pbuf_dequeue(p);
           /* fill-in Ethernet header */
-          ethhdr = p->payload;
           for (k = 0; k < netif->hwaddr_len; ++k) {
             ethhdr->dest.addr[k] = ethaddr->addr[k];
             ethhdr->src.addr[k] = netif->hwaddr[k];
           }
           ethhdr->type = htons(ETHTYPE_IP);
-          LWIP_DEBUGF(ETHARP_DEBUG | DBG_TRACE, ("update_arp_entry: sending queued IP packet %p.\n",(void *)p));
+          LWIP_DEBUGF(ETHARP_DEBUG | DBG_TRACE, ("update_arp_entry: sending queued IP packet %p.\n", (void *)p));
           /* send the queued IP packet */
           netif->linkoutput(netif, p);
           /* free the queued IP packet */
           pbuf_free(p);
-          /* proceed to next packet on queue */
-          p = n;
         }
-        /* NULL attached buffer*/
-        arp_table[i].p = NULL;
 #endif
+        /* IP addresses should only occur once in the ARP entry, we are done */
         return NULL;
       }
-    } /* if */
-  } /* for */
+    } /* if STABLE */
+  } /* for all ARP entries */
 
   /* no matching ARP entry was found */
   LWIP_ASSERT("update_arp_entry: i == ARP_TABLE_SIZE", i == ARP_TABLE_SIZE);
 
   LWIP_DEBUGF(ETHARP_DEBUG | DBG_TRACE, ("update_arp_entry: IP address not yet in table\n"));
-  /* allowed to insert an entry? */
-  if ((ETHARP_ALWAYS_INSERT) || (flags & ARP_INSERT_FLAG))
+  /* allowed to insert a new entry? */
+  if (flags & ARP_INSERT_FLAG)
   {
     LWIP_DEBUGF(ETHARP_DEBUG | DBG_TRACE, ("update_arp_entry: adding entry to table\n"));
     /* find an empty or old entry. */
@@ -461,7 +356,7 @@ update_arp_entry(struct netif *netif, struct ip_addr *ipaddr, struct eth_addr *e
  *
  * @see pbuf_free()
  */
-struct pbuf *
+void
 etharp_ip_input(struct netif *netif, struct pbuf *p)
 {
   struct ethip_hdr *hdr;
@@ -472,13 +367,12 @@ etharp_ip_input(struct netif *netif, struct pbuf *p)
   /* source is on local network? */
   if (!ip_addr_maskcmp(&(hdr->ip.src), &(netif->ip_addr), &(netif->netmask))) {
     /* do nothing */
-    return NULL;
+    return;
   }
 
   LWIP_DEBUGF(ETHARP_DEBUG | DBG_TRACE, ("etharp_ip_input: updating ETHARP table.\n"));
   /* update ARP table, ask to insert entry */
   update_arp_entry(netif, &(hdr->ip.src), &(hdr->eth.src), ARP_INSERT_FLAG);
-  return NULL;
 }
 
 
@@ -497,10 +391,12 @@ etharp_ip_input(struct netif *netif, struct pbuf *p)
  *
  * @see pbuf_free()
  */
-struct pbuf *
+void
 etharp_arp_input(struct netif *netif, struct eth_addr *ethaddr, struct pbuf *p)
 {
   struct etharp_hdr *hdr;
+  /* these are aligned properly, whereas the ARP header fields might not be */
+  struct ip_addr sipaddr, dipaddr;
   u8_t i;
   u8_t for_us;
 
@@ -508,33 +404,43 @@ etharp_arp_input(struct netif *netif, struct eth_addr *ethaddr, struct pbuf *p)
   if (p->tot_len < sizeof(struct etharp_hdr)) {
     LWIP_DEBUGF(ETHARP_DEBUG | DBG_TRACE | 1, ("etharp_arp_input: packet dropped, too short (%d/%d)\n", p->tot_len, sizeof(struct etharp_hdr)));
     pbuf_free(p);
-    return NULL;
+    return;
   }
 
   hdr = p->payload;
  
+  /* get aligned copies of addresses */
+  *(struct ip_addr2 *)&sipaddr = hdr->sipaddr;
+  *(struct ip_addr2 *)&dipaddr = hdr->dipaddr;
+
   /* this interface is not configured? */
   if (netif->ip_addr.addr == 0) {
     for_us = 0;
   } else {
     /* ARP packet directed to us? */
-    for_us = ip_addr_cmp(&(hdr->dipaddr), &(netif->ip_addr));
+    for_us = ip_addr_cmp(&dipaddr, &(netif->ip_addr));
   }
 
+  /* ARP message directed to us? */
+  if (for_us) {
+    /* add IP address in ARP cache; assume requester wants to talk to us.
+     * can result in directly sending the queued packets for this host. */
+    update_arp_entry(netif, &sipaddr, &(hdr->shwaddr), ARP_INSERT_FLAG);
+  /* ARP message not directed to us? */
+  } else {
+    /* update the source IP address in the cache, if present */
+    update_arp_entry(netif, &sipaddr, &(hdr->shwaddr), 0);
+  }
+
+  /* now act on the message itself */
   switch (htons(hdr->opcode)) {
   /* ARP request? */
   case ARP_REQUEST:
     /* ARP request. If it asked for our address, we send out a
-    reply. In any case, we time-stamp any existing ARP entry,
-    and possiby send out an IP packet that was queued on it. */
+     * reply. In any case, we time-stamp any existing ARP entry,
+     * and possiby send out an IP packet that was queued on it. */
 
     LWIP_DEBUGF (ETHARP_DEBUG | DBG_TRACE, ("etharp_arp_input: incoming ARP request\n"));
-    /* we are not configured? */
-    if (netif->ip_addr.addr == 0) {
-      LWIP_DEBUGF(ETHARP_DEBUG | DBG_TRACE, ("etharp_arp_input: we are unconfigured, ARP request ignored.\n"));
-      pbuf_free(p);
-      return NULL;
-    }
     /* ARP request for our address? */
     if (for_us) {
 
@@ -542,8 +448,8 @@ etharp_arp_input(struct netif *netif, struct eth_addr *ethaddr, struct pbuf *p)
       /* re-use pbuf to send ARP reply */
       hdr->opcode = htons(ARP_REPLY);
 
-      ip_addr_set(&(hdr->dipaddr), &(hdr->sipaddr));
-      ip_addr_set(&(hdr->sipaddr), &(netif->ip_addr));
+      hdr->dipaddr = hdr->sipaddr;
+      hdr->sipaddr = *(struct ip_addr2 *)&netif->ip_addr;
 
       for(i = 0; i < netif->hwaddr_len; ++i) {
         hdr->dhwaddr.addr[i] = hdr->shwaddr.addr[i];
@@ -561,41 +467,30 @@ etharp_arp_input(struct netif *netif, struct eth_addr *ethaddr, struct pbuf *p)
       hdr->ethhdr.type = htons(ETHTYPE_ARP);
       /* return ARP reply */
       netif->linkoutput(netif, p);
-
+    /* we are not configured? */
+    } else if (netif->ip_addr.addr == 0) {
+      /* { for_us == 0 and netif->ip_addr.addr == 0 } */
+      LWIP_DEBUGF(ETHARP_DEBUG | DBG_TRACE, ("etharp_arp_input: we are unconfigured, ARP request ignored.\n"));
     /* request was not directed to us */
     } else {
-      LWIP_DEBUGF(ETHARP_DEBUG | DBG_TRACE, ("etharp_arp_input: incoming ARP request was not for us.\n"));
+      /* { for_us == 0 and netif->ip_addr.addr != 0 } */
+      LWIP_DEBUGF(ETHARP_DEBUG | DBG_TRACE, ("etharp_arp_input: ARP request was not for us.\n"));
     }
     break;
   case ARP_REPLY:
-    /* ARP reply. We insert or update the ARP table later. */
+    /* ARP reply. We already updated the ARP cache earlier. */
     LWIP_DEBUGF(ETHARP_DEBUG | DBG_TRACE, ("etharp_arp_input: incoming ARP reply\n"));
 #if (LWIP_DHCP && DHCP_DOES_ARP_CHECK)
-    /* DHCP needs to know about ARP replies to our address */
-    if (for_us) dhcp_arp_reply(netif, &hdr->sipaddr);
+    /* DHCP wants to know about ARP replies to our wanna-have-address */
+    if (for_us) dhcp_arp_reply(netif, &sipaddr);
 #endif
     break;
   default:
     LWIP_DEBUGF(ETHARP_DEBUG | DBG_TRACE, ("etharp_arp_input: ARP unknown opcode type %d\n", htons(hdr->opcode)));
     break;
   }
-  /* add or update entries in the ARP cache */
-  if (for_us) {
-    /* insert IP address in ARP cache (assume requester wants to talk to us)
-     * we might even send out a queued packet to this host */
-    update_arp_entry(netif, &(hdr->sipaddr), &(hdr->shwaddr), ARP_INSERT_FLAG);
-  /* request was not directed to us, but snoop anyway */
-  } else {
-    /* update or insert the source IP address in the cache */
-    update_arp_entry(netif, &(hdr->sipaddr), &(hdr->shwaddr), 0);
-    /* update or insert the destination IP address pair in the cache */
-    update_arp_entry(netif, &(hdr->dipaddr), &(hdr->dhwaddr), 0);
-  }
   /* free ARP packet */
   pbuf_free(p);
-  p = NULL;
-  /* nothing to send, we did it! */
-  return NULL;
 }
 
 /**
@@ -619,33 +514,36 @@ etharp_arp_input(struct netif *netif, struct eth_addr *ethaddr, struct pbuf *p)
  *
  * @return If non-NULL, a packet ready to be sent by caller.
  *
+ * @return
+ * - ERR_BUF Could not make room for Ethernet header.
+ * - ERR_MEM Hardware address unknown, and no more ARP entries available
+ *   to query for address or queue the packet.
+ * - ERR_RTE No route to destination (no gateway to external networks).
  */
-struct pbuf *
+err_t
 etharp_output(struct netif *netif, struct ip_addr *ipaddr, struct pbuf *q)
 {
   struct eth_addr *dest, *srcaddr, mcastaddr;
   struct eth_hdr *ethhdr;
   s8_t i;
+  err_t result = ERR_OK;
 
-  /* make room for Ethernet header */
+  /* make room for Ethernet header - should not fail*/
   if (pbuf_header(q, sizeof(struct eth_hdr)) != 0) {
-    /* The pbuf_header() call shouldn't fail, and we'll just bail
-    out if it does.. */
+    /* bail out */
     LWIP_DEBUGF(ETHARP_DEBUG | DBG_TRACE | 2, ("etharp_output: could not allocate room for header.\n"));
     LINK_STATS_INC(link.lenerr);
-    return NULL;
+    pbuf_free(q);
+    return ERR_BUF;
   }
 
   /* assume unresolved Ethernet address */
   dest = NULL;
-  /* Construct Ethernet header. Start with looking up deciding which
-     MAC address to use as a destination address. Broadcasts and
-     multicasts are special, all other addresses are looked up in the
-     ARP table. */
+  /* Determine on destination hardware address. Broadcasts and multicasts
+   * are special, other IP addresses are looked up in the ARP table. */
 
   /* destination IP address is an IP broadcast address? */
-  if (ip_addr_isany(ipaddr) ||
-    ip_addr_isbroadcast(ipaddr, &(netif->netmask))) {
+  if (ip_addr_isany(ipaddr) || ip_addr_isbroadcast(ipaddr, netif)) {
     /* broadcast on Ethernet also */
     dest = (struct eth_addr *)&ethbroadcast;
   }
@@ -663,86 +561,67 @@ etharp_output(struct netif *netif, struct ip_addr *ipaddr, struct pbuf *q)
   }
   /* destination IP address is an IP unicast address */
   else {
-    /* destination IP network address not on local network?
-     * IP layer wants us to forward to the default gateway */
+    /* outside local network? */
     if (!ip_addr_maskcmp(ipaddr, &(netif->ip_addr), &(netif->netmask))) {
       /* interface has default gateway? */
-      if (netif->gw.addr != 0)
-      {
-        /* route to default gateway IP address */
+      if (netif->gw.addr != 0) {
+        /* send to hardware address of default gateway IP address */
         ipaddr = &(netif->gw);
-      }
-      /* no gateway available? */
-      else
-      {
-        /* IP destination address outside local network, but no gateway available */
-        /* { packet is discarded } */
-        return NULL;
+      /* no default gateway available? */
+      } else {
+        /* destination unreachable, discard packet */
+        pbuf_free(q);
+        return ERR_RTE;
       }
     }
-
-    /* Ethernet address for IP destination address is in ARP cache? */
-    for (i = 0; i < ARP_TABLE_SIZE; ++i) {
-      /* match found? */
-      if (arp_table[i].state == ETHARP_STATE_STABLE &&
-        ip_addr_cmp(ipaddr, &arp_table[i].ipaddr)) {
-        dest = &arp_table[i].ethaddr;
-        break;
-      }
-    }
-    /* could not find the destination Ethernet address in ARP cache? */
-    if (dest == NULL) {
-      /* ARP query for the IP address, submit this IP packet for queueing */
-      /* TODO: How do we handle netif->ipaddr == ipaddr? */
-      etharp_query(netif, ipaddr, q);
-      /* { packet was queued (ERR_OK), or discarded } */
-      /* return nothing */
-      return NULL;
-    }
-    /* destination Ethernet address resolved from ARP cache */
-    else
-    {
-      /* fallthrough */
-    }
-  }
+    result = etharp_query(netif, ipaddr, q);
+  } /* else unicast */
 
   /* destination Ethernet address known */
   if (dest != NULL) {
     /* obtain source Ethernet address of the given interface */
     srcaddr = (struct eth_addr *)netif->hwaddr;
-
     /* A valid IP->MAC address mapping was found, fill in the
      * Ethernet header for the outgoing packet */
     ethhdr = q->payload;
-
     for(i = 0; i < netif->hwaddr_len; i++) {
       ethhdr->dest.addr[i] = dest->addr[i];
       ethhdr->src.addr[i] = srcaddr->addr[i];
     }
-
     ethhdr->type = htons(ETHTYPE_IP);
-    /* return the outgoing packet */
-    return q;
+    /* send packet */
+    result = netif->linkoutput(netif, q);
   }
   /* never reached; here for safety */
-  return NULL;
+  pbuf_free(q);
+  return result;
 }
 
 /**
  * Send an ARP request for the given IP address.
  *
- * Sends an ARP request for the given IP address, unless
- * a request for this address is already pending. Optionally
- * queues an outgoing packet on the resulting ARP entry.
+ * If the IP address was not yet in the cache, a pending ARP cache entry
+ * is added and an ARP request is sent for the given address. The packet
+ * is queued on this entry.
  *
+ * If the IP address was already pending in the cache, a new ARP request
+ * is sent for the given address. The packet is queued on this entry.
+ *
+ * If the IP address was already stable in the cache, the packet is
+ * directly sent. An ARP request is sent out.
+ * 
  * @param netif The lwIP network interface where ipaddr
  * must be queried for.
  * @param ipaddr The IP address to be resolved.
- * @param q If non-NULL, a pbuf that must be queued on the
- * ARP entry for the ipaddr IP address.
- *
- * @return NULL.
+ * @param q If non-NULL, a pbuf that must be delivered to the IP address.
  *
+ * @return
+ * - ERR_BUF Could not make room for Ethernet header.
+ * - ERR_MEM Hardware address unknown, and no more ARP entries available
+ *   to query for address or queue the packet.
+ * - ERR_MEM Could not queue packet due to memory shortage.
+ * - ERR_RTE No route to destination (no gateway to external networks).
+  *
  * @note Might be used in the future by manual IP configuration
  * as well.
  *
@@ -751,29 +630,73 @@ etharp_output(struct netif *netif, struct ip_addr *ipaddr, struct pbuf *q)
  */
 err_t etharp_query(struct netif *netif, struct ip_addr *ipaddr, struct pbuf *q)
 {
-  struct eth_addr *srcaddr;
-  struct etharp_hdr *hdr;
+  struct pbuf *p;
+  struct eth_addr * srcaddr = (struct eth_addr *)netif->hwaddr;
   err_t result = ERR_OK;
-  s8_t i;
-  u8_t perform_arp_request = 1;
-  /* prevent 'unused argument' warning if ARP_QUEUEING == 0 */
-  (void)q;
-  srcaddr = (struct eth_addr *)netif->hwaddr;
-  /* bail out if this IP address is pending */
+  s8_t i; /* ARP entry index */
+  u8_t k; /* Ethernet address octet index */
+
+  /* Do three things in this order (by design):
+   * 
+   * 1) send out ARP request
+   * 2) find entry in ARP cache
+   * 3) handle the packet
+   */
+
+  /* allocate a pbuf for the outgoing ARP request packet */
+  p = pbuf_alloc(PBUF_LINK, sizeof(struct etharp_hdr), PBUF_RAM);
+  /* could allocate a pbuf for an ARP request? */
+  if (p != NULL) {
+    struct etharp_hdr *hdr = p->payload;
+    LWIP_DEBUGF(ETHARP_DEBUG | DBG_TRACE, ("etharp_query: sending ARP request.\n"));
+    hdr->opcode = htons(ARP_REQUEST);
+    for (k = 0; k < netif->hwaddr_len; k++)
+    {
+      hdr->shwaddr.addr[k] = srcaddr->addr[k];
+      /* the hardware address is what we ask for, in
+       * a request it is a don't-care value, we use zeroes */
+      hdr->dhwaddr.addr[k] = 0x00;
+    }
+    hdr->dipaddr = *(struct ip_addr2 *)ipaddr;
+    hdr->sipaddr = *(struct ip_addr2 *)&netif->ip_addr;
+
+    hdr->hwtype = htons(HWTYPE_ETHERNET);
+    ARPH_HWLEN_SET(hdr, netif->hwaddr_len);
+
+    hdr->proto = htons(ETHTYPE_IP);
+    ARPH_PROTOLEN_SET(hdr, sizeof(struct ip_addr));
+    for (k = 0; k < netif->hwaddr_len; ++k)
+    {
+      /* broadcast to all network interfaces on the local network */
+      hdr->ethhdr.dest.addr[k] = 0xff;
+      hdr->ethhdr.src.addr[k] = srcaddr->addr[k];
+    }
+    hdr->ethhdr.type = htons(ETHTYPE_ARP);
+    /* send ARP query */
+    result = netif->linkoutput(netif, p);
+    /* free ARP query packet */
+    pbuf_free(p);
+    p = NULL;
+  /* could not allocate pbuf for ARP request */
+  } else {
+    result = ERR_MEM;
+    LWIP_DEBUGF(ETHARP_DEBUG | DBG_TRACE | 2, ("etharp_query: could not allocate pbuf for ARP request.\n"));
+  }
+
+  /* search entry of queried IP address in the ARP cache */
   for (i = 0; i < ARP_TABLE_SIZE; ++i) {
-    if (ip_addr_cmp(ipaddr, &arp_table[i].ipaddr)) {
+    /* valid ARP cache entry with matching IP address? */
+    if (arp_table[i].state != ETHARP_STATE_EMPTY &&
+    	ip_addr_cmp(ipaddr, &arp_table[i].ipaddr)) {
+      /* pending entry? */
       if (arp_table[i].state == ETHARP_STATE_PENDING) {
-        LWIP_DEBUGF(ETHARP_DEBUG | DBG_TRACE | DBG_STATE, ("etharp_query: requested IP already pending as entry %u\n", i));
-        /* break out of for-loop, user may wish to queue a packet on a pending entry */
-        /* TODO: we will issue a new ARP request, which should not occur too often */
-        /* we might want to run a faster timer on ARP to limit this */
+        LWIP_DEBUGF(ETHARP_DEBUG | DBG_TRACE | DBG_STATE, ("etharp_query: requested IP already pending in entry %u\n", i));
+        /* { i != ARP_TABLE_SIZE } */
         break;
       }
       else if (arp_table[i].state == ETHARP_STATE_STABLE) {
-        LWIP_DEBUGF(ETHARP_DEBUG | DBG_TRACE | DBG_STATE, ("etharp_query: requested IP already stable as entry %u\n", i));
-        /* user may wish to queue a packet on a stable entry, so we proceed without ARP requesting */
-        /* TODO: even if the ARP entry is stable, we might do an ARP request anyway */
-        perform_arp_request = 0;
+        LWIP_DEBUGF(ETHARP_DEBUG | DBG_TRACE | DBG_STATE, ("etharp_query: requested IP already stable in entry %u\n", i));
+        /* { i != ARP_TABLE_SIZE } */
         break;
       }
     }
@@ -791,53 +714,42 @@ err_t etharp_query(struct netif *netif, struct ip_addr *ipaddr, struct pbuf *q)
     /* i is available, create ARP entry */
     arp_table[i].state = ETHARP_STATE_PENDING;
     ip_addr_set(&arp_table[i].ipaddr, ipaddr);
-  /* queried address was already in ARP table */
-  } else {
-#if ARP_QUEUEING
-    etharp_enqueue(i, q);
-#endif
+    arp_table[i].p = NULL;
+    LWIP_DEBUGF(ETHARP_DEBUG | DBG_TRACE, ("etharp_query: added pending entry %u for IP address\n", i));
   }
-  /* ARP request? */
-  if (perform_arp_request)
-  {
-    struct pbuf *p;
-    /* allocate a pbuf for the outgoing ARP request packet */
-    p = pbuf_alloc(PBUF_LINK, sizeof(struct etharp_hdr), PBUF_RAM);
-    /* could allocate pbuf? */
-    if (p != NULL) {
-      u8_t j;
-      LWIP_DEBUGF(ETHARP_DEBUG | DBG_TRACE, ("etharp_query: sending ARP request.\n"));
-      hdr = p->payload;
-      hdr->opcode = htons(ARP_REQUEST);
-      for (j = 0; j < netif->hwaddr_len; ++j)
-      {
-        hdr->shwaddr.addr[j] = srcaddr->addr[j];
-        /* the hardware address is what we ask for, in
-         * a request it is a don't-care, we use 0's */
-        hdr->dhwaddr.addr[j] = 0x00;
-      }
-      ip_addr_set(&(hdr->dipaddr), ipaddr);
-      ip_addr_set(&(hdr->sipaddr), &(netif->ip_addr));
 
-      hdr->hwtype = htons(HWTYPE_ETHERNET);
-      ARPH_HWLEN_SET(hdr, netif->hwaddr_len);
-
-      hdr->proto = htons(ETHTYPE_IP);
-      ARPH_PROTOLEN_SET(hdr, sizeof(struct ip_addr));
-      for (j = 0; j < netif->hwaddr_len; ++j)
-      {
-        hdr->ethhdr.dest.addr[j] = 0xff;
-        hdr->ethhdr.src.addr[j] = srcaddr->addr[j];
+  /* { i is either a (new or existing) PENDING or STABLE entry } */
+  
+  /* packet given? */
+  if (q != NULL) {
+    /* stable entry? */
+    if (arp_table[i].state == ETHARP_STATE_STABLE) {
+      /* we have a valid IP->Ethernet address mapping,
+       * fill in the Ethernet header for the outgoing packet */
+      struct eth_hdr *ethhdr = q->payload;
+      for(k = 0; k < netif->hwaddr_len; k++) {
+        ethhdr->dest.addr[k] = arp_table[i].ethaddr.addr[k];
+        ethhdr->src.addr[k]  = srcaddr->addr[k];
       }
-      hdr->ethhdr.type = htons(ETHTYPE_ARP);
-      /* send ARP query */
-      result = netif->linkoutput(netif, p);
-      /* free ARP query packet */
-      pbuf_free(p);
-      p = NULL;
-    } else {
-      result = ERR_MEM;
-      LWIP_DEBUGF(ETHARP_DEBUG | DBG_TRACE | 2, ("etharp_query: could not allocate pbuf for ARP request.\n"));
+      ethhdr->type = htons(ETHTYPE_IP);
+      LWIP_DEBUGF(ETHARP_DEBUG | DBG_TRACE, ("etharp_query: sending packet %p\n", (void *)q));
+      /* send the packet */
+      result = netif->linkoutput(netif, q);
+#if ARP_QUEUEING /* queue the given q packet */
+    /* pending entry? (either just created or already pending */
+    } else if (arp_table[i].state == ETHARP_STATE_PENDING) {
+      /* copy any PBUF_REF referenced payloads into PBUF_RAM */
+      /* (the caller assumes the referenced payload can be freed) */
+      p = pbuf_take(q);
+      /* queue packet */
+      if (p != NULL) {
+        pbuf_queue(arp_table[i].p, p);
+        LWIP_DEBUGF(ETHARP_DEBUG | DBG_TRACE, ("etharp_query: queued packet %p on ARP entry %d\n", (void *)q, i));
+      } else {
+        LWIP_DEBUGF(ETHARP_DEBUG | DBG_TRACE, ("etharp_query: could not queue a copy of PBUF_REF packet %p (out of memory)\n", (void *)q));
+        result = ERR_MEM; 
+      }
+#endif
     }
   }
   return result;

src/netif/ethernetif.c

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2001-2003 Swedish Institute of Computer Science.
+ * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
  * All rights reserved. 
  * 
  * Redistribution and use in source and binary forms, with or without modification, 
@@ -35,8 +35,6 @@
  * drivers for lwIP. Add code to the low_level functions and do a
  * search-and-replace for the word "ethernetif" to replace it with
  * something that better describes your network interface.
- *
- * THIS CODE NEEDS TO BE FIXED - IT IS NOT In SYNC WITH CURRENT ETHARP API
  */
 
 #include "lwip/opt.h"
@@ -45,11 +43,11 @@
 #include "lwip/pbuf.h"
 #include "lwip/sys.h"
 
-#include "netif/arp.h"
+#include "netif/etharp.h"
 
 /* Define those to better describe your network interface. */
 #define IFNAME0 'e'
-#define IFNAME1 't'
+#define IFNAME1 'n'
 
 struct ethernetif {
   struct eth_addr *ethaddr;
@@ -63,7 +61,6 @@ static void  ethernetif_input(struct netif *netif);
 static err_t ethernetif_output(struct netif *netif, struct pbuf *p,
              struct ip_addr *ipaddr);
 
-
 static void
 low_level_init(struct netif *netif)
 {
@@ -77,7 +74,7 @@ low_level_init(struct netif *netif)
   /* set MAC hardware address */
   netif->hwaddr[0] = ;
   ...
-  netif->hwaddr[6] = ;
+  netif->hwaddr[5] = ;
 
   /* maximum transfer unit */
   netif->mtu = 1500;
@@ -97,7 +94,6 @@ low_level_init(struct netif *netif)
  *
  */
 
-
 static err_t
 low_level_output(struct ethernetif *ethernetif, struct pbuf *p)
 {
@@ -105,6 +101,10 @@ low_level_output(struct ethernetif *ethernetif, struct pbuf *p)
 
   initiate transfer();
   
+#if ETH_PAD_SIZE
+  pbuf_header(p, -ETH_PAD_SIZE);			/* drop the padding word */
+#endif
+
   for(q = p; q != NULL; q = q->next) {
     /* Send the data from the pbuf to the interface, one pbuf at a
        time. The size of the data in each pbuf is kept in the ->len
@@ -113,6 +113,10 @@ low_level_output(struct ethernetif *ethernetif, struct pbuf *p)
   }
 
   signal that packet should be sent();
+
+#if ETH_PAD_SIZE
+  pbuf_header(p, ETH_PAD_SIZE);			/* reclaim the padding word */
+#endif
   
 #ifdef LINK_STATS
   lwip_stats.link.xmit++;
@@ -139,19 +143,33 @@ low_level_input(struct ethernetif *ethernetif)
      variable. */
   len = ;
 
+#if ETH_PAD_SIZE
+  len += ETH_PAD_SIZE;						/* allow room for Ethernet padding */
+#endif
+
   /* We allocate a pbuf chain of pbufs from the pool. */
   p = pbuf_alloc(PBUF_RAW, len, PBUF_POOL);
   
   if (p != NULL) {
+
+#if ETH_PAD_SIZE
+    pbuf_header(p, -ETH_PAD_SIZE);			/* drop the padding word */
+#endif
+
     /* We iterate over the pbuf chain until we have read the entire
-       packet into the pbuf. */
+     * packet into the pbuf. */
     for(q = p; q != NULL; q = q->next) {
       /* Read enough bytes to fill this pbuf in the chain. The
-         available data in the pbuf is given by the q->len
-         variable. */
+       * available data in the pbuf is given by the q->len
+       * variable. */
       read data into(q->payload, q->len);
     }
     acknowledge that packet has been read();
+
+#if ETH_PAD_SIZE
+    pbuf_header(p, ETH_PAD_SIZE);			/* reclaim the padding word */
+#endif
+
 #ifdef LINK_STATS
     lwip_stats.link.recv++;
 #endif /* LINK_STATS */      
@@ -171,7 +189,7 @@ low_level_input(struct ethernetif *ethernetif)
  *
  * This function is called by the TCP/IP stack when an IP packet
  * should be sent. It calls the function called low_level_output() to
- * do the actuall transmission of the packet.
+ * do the actual transmission of the packet.
  *
  */
 
@@ -179,90 +197,10 @@ static err_t
 ethernetif_output(struct netif *netif, struct pbuf *p,
       struct ip_addr *ipaddr)
 {
-  struct ethernetif *ethernetif;
-  struct pbuf *q;
-  struct eth_hdr *ethhdr;
-  struct eth_addr *dest, mcastaddr;
-  struct ip_addr *queryaddr;
-  err_t err;
-  u8_t i;
   
-  ethernetif = netif->state;
-
-  /* Make room for Ethernet header. */
-  if (pbuf_header(p, 14) != 0) {
-    /* The pbuf_header() call shouldn't fail, but we allocate an extra
-       pbuf just in case. */
-    q = pbuf_alloc(PBUF_LINK, 14, PBUF_RAM);
-    if (q == NULL) {
-#ifdef LINK_STATS
-      lwip_stats.link.drop++;
-      lwip_stats.link.memerr++;
-#endif /* LINK_STATS */      
-      return ERR_MEM;
-    }
-    pbuf_chain(q, p);
-    p = q;
-  }
-
-  /* Construct Ethernet header. Start with looking up deciding which
-     MAC address to use as a destination address. Broadcasts and
-     multicasts are special, all other addresses are looked up in the
-     ARP table. */
-  queryaddr = ipaddr;
-  if (ip_addr_isany(ipaddr) ||
-     ip_addr_isbroadcast(ipaddr, &(netif->netmask))) {
-    dest = (struct eth_addr *)&ethbroadcast;
-  } else if (ip_addr_ismulticast(ipaddr)) {
-    /* Hash IP multicast address to MAC address. */
-    mcastaddr.addr[0] = 0x01;
-    mcastaddr.addr[1] = 0x0;
-    mcastaddr.addr[2] = 0x5e;
-    mcastaddr.addr[3] = ip4_addr2(ipaddr) & 0x7f;
-    mcastaddr.addr[4] = ip4_addr3(ipaddr);
-    mcastaddr.addr[5] = ip4_addr4(ipaddr);
-    dest = &mcastaddr;
-  } else {
-
-    if (ip_addr_maskcmp(ipaddr, &(netif->ip_addr), &(netif->netmask))) {
-      /* Use destination IP address if the destination is on the same
-         subnet as we are. */
-      queryaddr = ipaddr;
-    } else {
-      /* Otherwise we use the default router as the address to send
-         the Ethernet frame to. */
-      queryaddr = &(netif->gw);
-    }
-    dest = arp_lookup(queryaddr);
-  }
-
-
-  /* If the arp_lookup() didn't find an address, we send out an ARP
-     query for the IP address. */
-  if (dest == NULL) {
-    q = arp_query(netif, ethernetif->ethaddr, queryaddr);
-    if (q != NULL) {
-      err = low_level_output(ethernetif, q);
-      pbuf_free(q);
-      return err;
-    }
-#ifdef LINK_STATS
-    lwip_stats.link.drop++;
-    lwip_stats.link.memerr++;
-#endif /* LINK_STATS */          
-    return ERR_MEM;
-  }
-  ethhdr = p->payload;
-
-  for(i = 0; i < 6; i++) {
-    ethhdr->dest.addr[i] = dest->addr[i];
-    ethhdr->src.addr[i] = ethernetif->ethaddr->addr[i];
-  }
-  
-  ethhdr->type = htons(ETHTYPE_IP);
-  
-  return low_level_output(ethernetif, p);
-
+ /* resolve hardware address, then send (or queue) packet */
+  return etharp_output(netif, ipaddr, p);
+ 
 }
 
 /*
@@ -282,43 +220,47 @@ ethernetif_input(struct netif *netif)
   struct eth_hdr *ethhdr;
   struct pbuf *p;
 
-
   ethernetif = netif->state;
   
+  /* move received packet into a new pbuf */
   p = low_level_input(ethernetif);
-
-  if (p != NULL) {
+  /* no packet could be read, silently ignore this */
+  if (p == NULL) return;
+  /* points to packet payload, which starts with an Ethernet header */
+  ethhdr = p->payload;
 
 #ifdef LINK_STATS
-    lwip_stats.link.recv++;
+  lwip_stats.link.recv++;
 #endif /* LINK_STATS */
 
-    ethhdr = p->payload;
+  ethhdr = p->payload;
     
-    switch (htons(ethhdr->type)) {
-    case ETHTYPE_IP:
-      arp_ip_input(netif, p);
-      pbuf_header(p, -14);
-      netif->input(p, netif);
-      break;
+  switch (htons(ethhdr->type)) {
+  /* IP packet? */
+  case ETHTYPE_IP:
+    /* update ARP table */
+    etharp_ip_input(netif, p);
+    /* skip Ethernet header */
+    pbuf_header(p, -sizeof(struct eth_hdr));
+    /* pass to network layer */
+    netif->input(p, netif);
+    break;
+      
     case ETHTYPE_ARP:
-      p = arp_arp_input(netif, ethernetif->ethaddr, p);
-      if (p != NULL) {
-  low_level_output(ethernetif, p);
-  pbuf_free(p);
-      }
+      /* pass p to ARP module  */
+      etharp_arp_input(netif, ethernetif->ethaddr, p);
       break;
     default:
       pbuf_free(p);
+      p = NULL;
       break;
-    }
   }
 }
 
 static void
 arp_timer(void *arg)
 {
-  arp_tmr();
+  etharp_tmr();
   sys_timeout(ARP_TMR_INTERVAL, arp_timer, NULL);
 }
 
@@ -337,6 +279,13 @@ ethernetif_init(struct netif *netif)
   struct ethernetif *ethernetif;
     
   ethernetif = mem_malloc(sizeof(struct ethernetif));
+  
+  if (ethernetif == NULL)
+  {
+  	LWIP_DEBUGF(NETIF_DEBUG, ("ethernetif_init: out of memory\n"));
+  	return ERR_MEM;
+  }
+  
   netif->state = ethernetif;
   netif->name[0] = IFNAME0;
   netif->name[1] = IFNAME1;
@@ -346,7 +295,8 @@ ethernetif_init(struct netif *netif)
   ethernetif->ethaddr = (struct eth_addr *)&(netif->hwaddr[0]);
   
   low_level_init(netif);
-  arp_init();
+
+  etharp_init();
 
   sys_timeout(ARP_TMR_INTERVAL, arp_timer, NULL);
 }

src/netif/loopif.c

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2001-2003 Swedish Institute of Computer Science.
+ * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
  * All rights reserved. 
  * 
  * Redistribution and use in source and binary forms, with or without modification, 

src/netif/ppp/auth.c

@@ -610,7 +610,7 @@ int get_secret(
     
     addrs = NULL;
 
-    if(!client || !client[0] && strcmp(client, ppp_settings.user)) {
+    if(!client || !client[0] || strcmp(client, ppp_settings.user)) {
 	return 0;
     }
 

src/netif/ppp/chpms.h

@@ -51,7 +51,7 @@
  * IMPLIED WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED
  * WARRANTIES OF MERCHANTIBILITY AND FITNESS FOR A PARTICULAR PURPOSE.
  *
- * $Id: chpms.h,v 1.2 2003/11/14 14:56:31 likewise Exp $
+ * $Id: chpms.h,v 1.3 2004/02/07 00:30:03 likewise Exp $
  */
 
 #ifndef CHPMS_H

src/netif/ppp/ipcp.c

@@ -178,6 +178,20 @@ static fsm_callbacks ipcp_callbacks = { /* IPCP callback routines */
 /*** LOCAL FUNCTION DEFINITIONS ***/
 /**********************************/
 
+/*
+ * Non-standard inet_ntoa left here for compat with original ppp
+ * sources. Assumes u32_t instead of struct in_addr.
+ */ 
+
+char * _inet_ntoa(u32_t n)
+{
+	struct in_addr ia;
+	ia.s_addr = n;
+	return inet_ntoa(ia);
+}
+
+#define inet_ntoa _inet_ntoa
+
 /*
  * ipcp_init - Initialize IPCP.
  */

src/netif/ppp/vj.h

@@ -1,7 +1,7 @@
 /*
  * Definitions for tcp compression routines.
  *
- * $Id: vj.h,v 1.3 2003/11/14 14:56:31 likewise Exp $
+ * $Id: vj.h,v 1.4 2004/02/07 00:30:03 likewise Exp $
  *
  * Copyright (c) 1989 Regents of the University of California.
  * All rights reserved.

src/netif/slipif.c

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2001, Swedish Institute of Computer Science.
+ * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
  * All rights reserved. 
  *
  * Redistribution and use in source and binary forms, with or without 
@@ -199,6 +199,7 @@ slipif_init(struct netif *netif)
   netif->name[1] = 'l';
   netif->output = slipif_output;
   netif->mtu = 1500;  
+  netif->flags = NETIF_FLAG_POINTTOPOINT;
 
   netif->state = sio_open(netif->num);
   if (!netif->state)