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

@@ -1,13 +1,89 @@
+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 (/DSP) architectures showing these problems.
+    We still have to fix some of these issues neatly.
+
 HISTORY
 
-(current)
+(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:
+
+  * Fixed TCP bug for SYN_SENT to ESTABLISHED state transition.
+  * Fixed TCP bug in dequeueing of FIN from out of order segment queue.
+  * Fixed two possible NULL references in rare cases.
+
+(STABLE-0_6_6)
+
+  ++ Bug fixes:
+
+  * Fixed DHCP which did not include the IP address in DECLINE messages.
+
+  ++ Changes:
+
+  * etharp.c has been hauled over a bit.
+
+(STABLE-0_6_5)
+
+  ++ Bug fixes:
+
+  * Fixed TCP bug induced by bad window resizing with unidirectional TCP traffic.
+  * Packets sent from ARP queue had invalid source hardware address.
+
+  ++ Changes:
+
+  * Pass-by ARP requests do now update the cache.
+
+  ++ New features:
+
+  * No longer dependent on ctype.h.
+  * New socket options.
+  * Raw IP pcb support.
+
+(STABLE-0_6_4)
+
+  ++ Bug fixes:
+
+  * Some debug formatters and casts fixed.
+  * Numereous fixes in PPP.
+
+  ++ Changes:
+
+  * DEBUGF now is LWIP_DEBUGF
+  * pbuf_dechain() has been re-enabled.
+  * Mentioned the changed use of CVS branches in README.
 
 (STABLE-0_6_3)
 
   ++ Bug fixes:
-  
+
   * Fixed pool pbuf memory leak in pbuf_alloc().
     Occured if not enough PBUF_POOL pbufs for a packet pbuf chain.
     Reported by Savin Zlobec.
@@ -20,11 +96,11 @@ HISTORY
   * Added PPP stack contributed by Marc Boucher
 
   ++ Changes:
-  
+
   * Now drops short packets for ICMP/UDP/TCP protocols. More robust.
 
   * ARP queueuing now queues the latest packet instead of the first.
-    This is the recommended behaviour, but can be overridden in
+    This is the RFC recommended behaviour, but can be overridden in
     lwipopts.h.
 
 (0.6.2)
@@ -80,10 +156,10 @@ HISTORY
 
   * netif's have a dhcp field that must be initialized to NULL by the driver.
     See the contrib/ports/c16x cs8900 driver as a driver example.
-  
+
 (0.5.x) This file has been unmaintained up to 0.6.1. All changes are
   logged in CVS but have not been explained here.
-  
+
 (0.5.3) Changes since version 0.5.2
 
   ++ Bugfixes:
@@ -115,7 +191,7 @@ HISTORY
   * pbuf_dechain() did not update the ->tot_len field of the tail.
 
   * Aborted TCP connections were not handled correctly in all
-    situations. 
+    situations.
 
   ++ Other changes:
 
@@ -123,7 +199,7 @@ HISTORY
 
   * The ->len field in the tcp_seg structure now counts the actual
     amount of data, and does not add one for SYN and FIN segments.
-    
+
 (0.5.1) Changes since version 0.5.0
 
   ++ New features:
@@ -133,16 +209,16 @@ HISTORY
   * Preliminary support for cross platform packed structs.
 
   * ARP timer now implemented.
-  
+
   ++ Bugfixes:
 
   * TCP output queue length was badly initialized when opening
     connections.
 
   * TCP delayed ACKs were not sent correctly.
-  
+
   * Explicit initialization of BSS segment variables.
-  
+
   * read() in BSD socket library could drop data.
 
   * Problems with memory alignment.
@@ -156,7 +232,7 @@ HISTORY
 
   * IP multicast address tests had endianess problems.
 
-  * ARP requests had wrong destination hardware address. 
+  * ARP requests had wrong destination hardware address.
 
   ++ Other changes:
 
@@ -166,8 +242,8 @@ HISTORY
 
   * TCP and UDP ->dest_* struct members where changed to ->remote_*.
 
-  * ntoh* macros are now null definitions for big endian CPUs. 
-  
+  * ntoh* macros are now null definitions for big endian CPUs.
+
 (0.5.0) Changes since version 0.4.2
 
   ++ New features:
@@ -175,9 +251,9 @@ HISTORY
   * Redesigned operating system emulation layer to make porting easier.
 
   * Better control over TCP output buffers.
-    
+
   * Documenation added.
-  
+
   ++ Bugfixes:
 
   * Locking issues in buffer management.
@@ -190,7 +266,7 @@ HISTORY
 
   * Directory structure somewhat changed; the core/ tree has been
     collapsed.
-  
+
 (0.4.2) Changes since version 0.4.1
 
   ++ New features:
@@ -213,7 +289,7 @@ HISTORY
 
   * Variable++ have in appliciable cases been translated to ++variable
     since some compilers generate better code in the latter case.
-    
+
 (0.4.1) Changes since version 0.4
 
   ++ New features:
@@ -225,43 +301,43 @@ HISTORY
   * UDP: experimental support for UDP-Lite extensions.
 
   ++ Bugfixes:
-    
+
   * TCP: out of order segments were in some cases handled incorrectly,
     and this has now been fixed. Delayed acknowledgements was broken
     in 0.4, has now been fixed. Binding to an address that is in use
     now results in an error. Reset connections sometimes hung an
-    application; this has been fixed. 
+    application; this has been fixed.
 
   * Checksum calculation sometimes failed for chained pbufs with odd
     lengths. This has been fixed.
-    
+
   * API: a lot of bug fixes in the API. The UDP API has been improved
     and tested. Error reporting and handling has been
     improved. Logical flaws and race conditions for incoming TCP
-    connections has been found and removed. 
- 
+    connections has been found and removed.
+
   * Memory manager: alignment issues. Reallocating memory sometimes
     failed, this has been fixed.
 
-  * Generic library: bcopy was flawed and has been fixed. 
+  * Generic library: bcopy was flawed and has been fixed.
 
   ++ Other changes:
-  
+
   * API: all datatypes has been changed from generic ones such as
     ints, to specified ones such as u16_t. Functions that return
     errors now have the correct type (err_t).
-  
+
   * General: A lot of code cleaned up and debugging code removed. Many
     portability issues have been fixed.
 
-  * The license was changed; the advertising clause was removed. 
+  * The license was changed; the advertising clause was removed.
 
   * C64 port added.
 
   * Thanks: Huge thanks go to Dagan Galarneau, Horst Garnetzke, Petri
     Kosunen, Mikael Caleres, and Frits Wilmink for reporting and
     fixing bugs!
-  
+
 (0.4) Changes since version 0.3.1
 
   * Memory management has been radically changed; instead of
@@ -269,14 +345,14 @@ HISTORY
     rapidly allocated and deallocated is now kept in pools. Allocation
     and deallocation from those memory pools is very fast. The shared
     heap is still present but is used less frequently.
-    
+
   * The memory, memory pool, and packet buffer subsystems now support
     4-, 2-, or 1-byte alignment.
 
   * "Out of memory" situations are handled in a more robust way.
- 
+
   * Stack usage has been reduced.
-    
+
   * Easier configuration of lwIP parameters such as memory usage,
     TTLs, statistics gathering, etc. All configuration parameters are
     now kept in a single header file "lwipopts.h".
@@ -290,7 +366,7 @@ HISTORY
 
   * The code for the RTXC architecture has been implemented, tested
     and put to use.
-    
+
   * Bugs have been found and corrected in the TCP, UDP, IP, API, and
     the Internet checksum modules.
 
@@ -299,7 +375,7 @@ HISTORY
 
   * The license has been changed slightly to conform more with the
     original BSD license, including the advertisement clause.
-  
+
 (0.3.1) Changes since version 0.3
 
   * Fix of a fatal bug in the buffer management. Pbufs with allocated
@@ -325,10 +401,10 @@ HISTORY
     to free some memory and retry the allocation.
 
   * Much testing has been done with limited memory
-    configurations. lwIP now does a better job when overloaded. 
+    configurations. lwIP now does a better job when overloaded.
 
   * Some bugfixes and improvements to the buffer (pbuf) subsystem.
-    
+
   * Many bugfixes in the TCP code:
 
     - Fixed a bug in tcp_close().
@@ -355,15 +431,15 @@ HISTORY
     - TCP retransmission timeout backoffs are not correctly computed
       (ala BSD). After a number of retransmissions, TCP now gives up
       the connection.
-      
+
   * TCP connections now are kept on three lists, one for active
     connections, one for listening connections, and one for
     connections that are in TIME-WAIT. This greatly speeds up the fast
     timeout processing for sending delayed ACKs.
-  
+
   * TCP now provides proper feedback to the application when a
     connection has been successfully set up.
-  
+
   * More comments have been added to the code. The code has also been
     somewhat cleaned up.
 

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

@@ -2,51 +2,99 @@ 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
 
 
-1 Anonymous CVS checkouts and updates
--------------------------------------
 
-Obtain lwIP from the STABLE tree (on the CVS main trunk)
+1 Obtaining lwIP from the CVS repository
+----------------------------------------
 
-cvs -d:pserver:anoncvs@subversions.gnu.org:/cvsroot/lwip login
-cvs -d:pserver:anoncvs@subversions.gnu.org:/cvsroot/lwip checkout lwip
+To perform an anonymous CVS checkout of the main trunk (this is where
+bug fixes and incremental enhancements occur), do this:
 
-Or, obtain a specific release as follows:
+export CVS_RSH=ssh
+cvs -d:ext:anoncvs@subversions.gnu.org:/cvsroot/lwip checkout lwip
 
-cvs -d:pserver:anoncvs@subversions.gnu.org:/cvsroot/lwip login
-cvs -d:pserver:anoncvs@subversions.gnu.org:/cvsroot/lwip checkout -r STABLE-0_6_3 -d lwip-0.6.3 lwip
+(If SSH asks about authenticity of the host, you can check the key
+ fingerprint against http://savannah.nongnu.org/cvs/?group=lwip)
 
-Or, obtain a development branch as follows:
+Or, obtain a stable branch (updated with bug fixes only) as follows:
+cvs -d:ext:anoncvs@subversions.gnu.org:/cvsroot/lwip checkout -r STABLE-0_7 -d lwip-0.7 lwip
 
-cvs -d:pserver:anoncvs@subversions.gnu.org:/cvsroot/lwip login
-cvs -d:pserver:anoncvs@subversions.gnu.org:/cvsroot/lwip checkout -r DEVEL -d lwip-devel lwip
+Or, obtain a specific (fixed) release as follows:
+cvs -d:ext:anoncvs@subversions.gnu.org:/cvsroot/lwip checkout -r STABLE-0_7_0 -d lwip-0.7.0 lwip
 
+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 Merging from DEVEL branch to main trunk (stable branch)
----------------------------------------------------------
+3 Committers/developers CVS access using SSH
+--------------------------------------------
 
-Merging from DEVEL to main requires that the DEVEL branch
-has a tag indicating the previous merger. This tag is
-called 'merged_from_DEVEL_to_main'.
+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)
+--------------------------------------------------
+
+Merging is a delicate process in CVS and requires the
+following disciplined steps in order to prevent conflicts
+in the future. Conflicts can be hard to solve!
+
+Merging from branch A to branch B requires that the A branch
+has a tag indicating the previous merger. This tag is called
+'merged_from_A_to_B'. After merging, the tag is moved in the
+A branch to remember this merger for future merge actions.
 
 IMPORTANT: AFTER COMMITTING A SUCCESFUL MERGE IN THE
-REPOSITORY, THE TAG MUST BE SET ON THE SOURCE OF THE
-MERGE BRANCH (REPLACING EXISTING TAGS WITH THE SAME NAME).
+REPOSITORY, THE TAG MUST BE SET ON THE SOURCE BRANCH OF THE
+MERGE ACTION (REPLACING EXISTING TAGS WITH THE SAME NAME).
 
 Merge all changes in DEVEL since our last merge to main:
 
 In the working copy of the main trunk:
 cvs update -P -jmerged_from_DEVEL_to_main -jDEVEL 
 
-Now move the tag in the DEVEL branch to this merge point,
-so we can use this for future merges.
+(This will apply the changes between 'merged_from_DEVEL_to_main'
+and 'DEVEL' to your work set of files)
 
-cvs -z4 rtag -F -r DEVEL merged_from_DEVEL_to_main lwip 
+We can now commit the merge result.
+cvs commit -R -m "Merged from DEVEL to main." 
+
+If this worked out OK, we now move the tag in the DEVEL branch
+to this merge point, so we can use this point for future merges:
+
+cvs rtag -F -r DEVEL merged_from_DEVEL_to_main lwip 
 
 4 How to release lwIP
 ---------------------
@@ -58,12 +106,14 @@ Login CVS using pserver authentication, then export a clean copy of the
 tagged tree. Export is similar to a checkout, except that the CVS metadata
 is not created locally. 
 
-cvs -d:pserver:anoncvs@subversions.gnu.org:/cvsroot/lwip login
-cvs -d:pserver:anoncvs@subversions.gnu.org:/cvsroot/lwip export -r STABLE-0_6_3 -d lwip-0.6.3 lwip
+export CVS_RSH=ssh
+cvs -d:ext:anoncvs@subversions.gnu.org:/cvsroot/lwip export -r STABLE-0_6_3 -d lwip-0.6.3 lwip
 
-Archive this directory using tar.
+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":
 
@@ -74,17 +124,15 @@ Now, make a new release by creating a new directory for it (these are
 Savannah conventions so that it shows up in the Files list real nice):
 
 mkdir stable.pkg
-cd stable.pkg
-mkdir 0.6.3
-cd 0.6.3
+mkdir stable.pkg 0.6.3
 
 We can now copy the tar archive we made earlier into the release directory:
 
-cp /../../../lwip-0.6.3.tar.gz .
+cp ../../../lwip-0.6.3.tar.gz .
 
 Finally, synchronize this directory upwards to Savannah:
 
-rsync -n -e "ssh -1" -t -u -v -r . likewise@savannah.nongnu.org:/upload/lwip
+rsync -n -e "ssh -1" -t -u -v -r *.pkg likewise@savannah.nongnu.org:/upload/lwip
 
 This does a "dry run": no files are modified! After you have confirmed that
 this is what you intended to do, remove "-n" and actually synchronize for

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, 
@@ -38,13 +38,13 @@
 #include "lwip/api_msg.h"
 #include "lwip/memp.h"
 
-/*-----------------------------------------------------------------------------------*/
+
 struct
 netbuf *netbuf_new(void)
 {
   struct netbuf *buf;
 
-  buf = memp_mallocp(MEMP_NETBUF);
+  buf = memp_malloc(MEMP_NETBUF);
   if (buf != NULL) {
     buf->p = NULL;
     buf->ptr = NULL;
@@ -53,7 +53,7 @@ netbuf *netbuf_new(void)
     return NULL;
   }
 }
-/*-----------------------------------------------------------------------------------*/
+
 void
 netbuf_delete(struct netbuf *buf)
 {
@@ -62,10 +62,10 @@ netbuf_delete(struct netbuf *buf)
       pbuf_free(buf->p);
       buf->p = buf->ptr = NULL;
     }
-    memp_freep(MEMP_NETBUF, buf);
+    memp_free(MEMP_NETBUF, buf);
   }
 }
-/*-----------------------------------------------------------------------------------*/
+
 void *
 netbuf_alloc(struct netbuf *buf, u16_t size)
 {
@@ -80,7 +80,7 @@ netbuf_alloc(struct netbuf *buf, u16_t size)
   buf->ptr = buf->p;
   return buf->p->payload;
 }
-/*-----------------------------------------------------------------------------------*/
+
 void
 netbuf_free(struct netbuf *buf)
 {
@@ -89,7 +89,7 @@ netbuf_free(struct netbuf *buf)
   }
   buf->p = buf->ptr = NULL;
 }
-/*-----------------------------------------------------------------------------------*/
+
 void
 netbuf_ref(struct netbuf *buf, void *dataptr, u16_t size)
 {
@@ -101,21 +101,21 @@ netbuf_ref(struct netbuf *buf, void *dataptr, u16_t size)
   buf->p->len = buf->p->tot_len = size;
   buf->ptr = buf->p;
 }
-/*-----------------------------------------------------------------------------------*/
+
 void
 netbuf_chain(struct netbuf *head, struct netbuf *tail)
 {
   pbuf_chain(head->p, tail->p);
   head->ptr = head->p;
-  memp_freep(MEMP_NETBUF, tail);
+  memp_free(MEMP_NETBUF, tail);
 }
-/*-----------------------------------------------------------------------------------*/
+
 u16_t
 netbuf_len(struct netbuf *buf)
 {
   return buf->p->tot_len;
 }
-/*-----------------------------------------------------------------------------------*/
+
 err_t
 netbuf_data(struct netbuf *buf, void **dataptr, u16_t *len)
 {
@@ -126,7 +126,7 @@ netbuf_data(struct netbuf *buf, void **dataptr, u16_t *len)
   *len = buf->ptr->len;
   return ERR_OK;
 }
-/*-----------------------------------------------------------------------------------*/
+
 s8_t
 netbuf_next(struct netbuf *buf)
 {
@@ -139,13 +139,13 @@ netbuf_next(struct netbuf *buf)
   }
   return 0;
 }
-/*-----------------------------------------------------------------------------------*/
+
 void
 netbuf_first(struct netbuf *buf)
 {
   buf->ptr = buf->p;
 }
-/*-----------------------------------------------------------------------------------*/
+
 void
 netbuf_copy_partial(struct netbuf *buf, void *dataptr, u16_t len, u16_t offset)
 {
@@ -154,7 +154,7 @@ netbuf_copy_partial(struct netbuf *buf, void *dataptr, u16_t len, u16_t offset)
 
   left = 0;
 
-  if (buf == NULL) {
+  if(buf == NULL || dataptr == NULL) {
     return;
   }
   
@@ -174,39 +174,43 @@ netbuf_copy_partial(struct netbuf *buf, void *dataptr, u16_t len, u16_t offset)
     }
   }
 }
-/*-----------------------------------------------------------------------------------*/
+
 void
 netbuf_copy(struct netbuf *buf, void *dataptr, u16_t len)
 {
   netbuf_copy_partial(buf, dataptr, len, 0);
 }
-/*-----------------------------------------------------------------------------------*/
+
 struct ip_addr *
 netbuf_fromaddr(struct netbuf *buf)
 {
   return buf->fromaddr;
 }
-/*-----------------------------------------------------------------------------------*/
+
 u16_t
 netbuf_fromport(struct netbuf *buf)
 {
   return buf->fromport;
 }
-/*-----------------------------------------------------------------------------------*/
+
 struct
-netconn *netconn_new(enum netconn_type t)
+netconn *netconn_new_with_proto_and_callback(enum netconn_type t, u16_t proto,
+                                   void (*callback)(struct netconn *, enum netconn_evt, u16_t len))
 {
   struct netconn *conn;
+  struct api_msg *msg;
 
-  conn = memp_mallocp(MEMP_NETCONN);
+  conn = memp_malloc(MEMP_NETCONN);
   if (conn == NULL) {
     return NULL;
   }
+  
+  conn->err = ERR_OK;
   conn->type = t;
   conn->pcb.tcp = NULL;
 
   if ((conn->mbox = sys_mbox_new()) == SYS_MBOX_NULL) {
-    memp_freep(MEMP_NETCONN, conn);
+    memp_free(MEMP_NETCONN, conn);
     return NULL;
   }
   conn->recvmbox = SYS_MBOX_NULL;
@@ -214,25 +218,44 @@ netconn *netconn_new(enum netconn_type t)
   conn->sem = SYS_SEM_NULL;
   conn->state = NETCONN_NONE;
   conn->socket = 0;
-  conn->callback = 0;
+  conn->callback = callback;
   conn->recv_avail = 0;
+
+  if((msg = memp_malloc(MEMP_API_MSG)) == NULL) {
+    memp_free(MEMP_NETCONN, conn);
+    return NULL;
+  }
+  
+  msg->type = API_MSG_NEWCONN;
+  msg->msg.msg.bc.port = proto; /* misusing the port field */
+  msg->msg.conn = conn;
+  api_msg_post(msg);  
+  sys_mbox_fetch(conn->mbox, NULL);
+  memp_free(MEMP_API_MSG, msg);
+
+  if ( conn->err != ERR_OK ) {
+    memp_free(MEMP_NETCONN, conn);
+    return NULL;
+  }
+
   return conn;
 }
-/*-----------------------------------------------------------------------------------*/
+
+
+struct
+netconn *netconn_new(enum netconn_type t)
+{
+  return netconn_new_with_proto_and_callback(t,0,NULL);
+}
+
 struct
 netconn *netconn_new_with_callback(enum netconn_type t,
                                    void (*callback)(struct netconn *, enum netconn_evt, u16_t len))
 {
-    struct netconn *conn;
-    
-    /* get a netconn and then initialize callback pointer and socket */
-    conn = netconn_new(t);
-    if (conn)
-        conn->callback = callback;
-    return conn;
+  return netconn_new_with_proto_and_callback(t,0,callback);
 }
 
-/*-----------------------------------------------------------------------------------*/
+
 err_t
 netconn_delete(struct netconn *conn)
 {
@@ -243,7 +266,7 @@ netconn_delete(struct netconn *conn)
     return ERR_OK;
   }
   
-  if ((msg = memp_mallocp(MEMP_API_MSG)) == NULL) {
+  if ((msg = memp_malloc(MEMP_API_MSG)) == NULL) {
     return ERR_MEM;
   }
   
@@ -251,7 +274,7 @@ netconn_delete(struct netconn *conn)
   msg->msg.conn = conn;
   api_msg_post(msg);  
   sys_mbox_fetch(conn->mbox, NULL);
-  memp_freep(MEMP_API_MSG, msg);
+  memp_free(MEMP_API_MSG, msg);
 
   /* Drain the recvmbox. */
   if (conn->recvmbox != SYS_MBOX_NULL) {
@@ -286,18 +309,21 @@ netconn_delete(struct netconn *conn)
   memp_free(MEMP_NETCONN, conn);
   return ERR_OK;
 }
-/*-----------------------------------------------------------------------------------*/
+
 enum netconn_type
 netconn_type(struct netconn *conn)
 {
   return conn->type;
 }
-/*-----------------------------------------------------------------------------------*/
+
 err_t
 netconn_peer(struct netconn *conn, struct ip_addr *addr,
        u16_t *port)
 {
   switch (conn->type) {
+  case NETCONN_RAW:
+    /* return an error as connecting is only a helper for upper layers */
+    return ERR_CONN;
   case NETCONN_UDPLITE:
   case NETCONN_UDPNOCHKSUM:
   case NETCONN_UDP:
@@ -316,12 +342,16 @@ netconn_peer(struct netconn *conn, struct ip_addr *addr,
   }
   return (conn->err = ERR_OK);
 }
-/*-----------------------------------------------------------------------------------*/
+
 err_t
 netconn_addr(struct netconn *conn, struct ip_addr **addr,
        u16_t *port)
 {
   switch (conn->type) {
+  case NETCONN_RAW:
+    *addr = &(conn->pcb.raw->local_ip);
+    *port = conn->pcb.raw->protocol;
+    break;
   case NETCONN_UDPLITE:
   case NETCONN_UDPNOCHKSUM:
   case NETCONN_UDP:
@@ -335,7 +365,7 @@ netconn_addr(struct netconn *conn, struct ip_addr **addr,
   }
   return (conn->err = ERR_OK);
 }
-/*-----------------------------------------------------------------------------------*/
+
 err_t
 netconn_bind(struct netconn *conn, struct ip_addr *addr,
       u16_t port)
@@ -353,7 +383,7 @@ netconn_bind(struct netconn *conn, struct ip_addr *addr,
     }
   }
   
-  if ((msg = memp_mallocp(MEMP_API_MSG)) == NULL) {
+  if ((msg = memp_malloc(MEMP_API_MSG)) == NULL) {
     return (conn->err = ERR_MEM);
   }
   msg->type = API_MSG_BIND;
@@ -362,11 +392,11 @@ netconn_bind(struct netconn *conn, struct ip_addr *addr,
   msg->msg.msg.bc.port = port;
   api_msg_post(msg);
   sys_mbox_fetch(conn->mbox, NULL);
-  memp_freep(MEMP_API_MSG, msg);
+  memp_free(MEMP_API_MSG, msg);
   return conn->err;
 }
 
-/*-----------------------------------------------------------------------------------*/
+
 err_t
 netconn_connect(struct netconn *conn, struct ip_addr *addr,
        u16_t port)
@@ -384,7 +414,7 @@ netconn_connect(struct netconn *conn, struct ip_addr *addr,
     }
   }
   
-  if ((msg = memp_mallocp(MEMP_API_MSG)) == NULL) {
+  if ((msg = memp_malloc(MEMP_API_MSG)) == NULL) {
     return ERR_MEM;
   }
   msg->type = API_MSG_CONNECT;
@@ -393,7 +423,7 @@ netconn_connect(struct netconn *conn, struct ip_addr *addr,
   msg->msg.msg.bc.port = port;
   api_msg_post(msg);
   sys_mbox_fetch(conn->mbox, NULL);
-  memp_freep(MEMP_API_MSG, msg);
+  memp_free(MEMP_API_MSG, msg);
   return conn->err;
 }
 
@@ -406,18 +436,18 @@ netconn_disconnect(struct netconn *conn)
     return ERR_VAL;
   }
 
-  if ((msg = memp_mallocp(MEMP_API_MSG)) == NULL) {
+  if ((msg = memp_malloc(MEMP_API_MSG)) == NULL) {
     return ERR_MEM;
   }
   msg->type = API_MSG_DISCONNECT;
   msg->msg.conn = conn;  
   api_msg_post(msg);
   sys_mbox_fetch(conn->mbox, NULL);
-  memp_freep(MEMP_API_MSG, msg);
+  memp_free(MEMP_API_MSG, msg);
   return conn->err;
 
 }
-/*-----------------------------------------------------------------------------------*/
+
 err_t
 netconn_listen(struct netconn *conn)
 {
@@ -434,17 +464,17 @@ netconn_listen(struct netconn *conn)
     }
   }
   
-  if ((msg = memp_mallocp(MEMP_API_MSG)) == NULL) {
+  if ((msg = memp_malloc(MEMP_API_MSG)) == NULL) {
     return (conn->err = ERR_MEM);
   }
   msg->type = API_MSG_LISTEN;
   msg->msg.conn = conn;
   api_msg_post(msg);
   sys_mbox_fetch(conn->mbox, NULL);
-  memp_freep(MEMP_API_MSG, msg);
+  memp_free(MEMP_API_MSG, msg);
   return conn->err;
 }
-/*-----------------------------------------------------------------------------------*/
+
 struct netconn *
 netconn_accept(struct netconn *conn)
 {
@@ -461,7 +491,7 @@ netconn_accept(struct netconn *conn)
   
   return newconn;
 }
-/*-----------------------------------------------------------------------------------*/
+
 struct netbuf *
 netconn_recv(struct netconn *conn)
 {
@@ -490,7 +520,7 @@ netconn_recv(struct netconn *conn)
     }
 
 
-    buf = memp_mallocp(MEMP_NETBUF);
+    buf = memp_malloc(MEMP_NETBUF);
 
     if (buf == NULL) {
       conn->err = ERR_MEM;
@@ -514,7 +544,7 @@ netconn_recv(struct netconn *conn)
     /* If we are closed, we indicate that we no longer wish to receive
        data by setting conn->recvmbox to SYS_MBOX_NULL. */
     if (p == NULL) {
-      memp_freep(MEMP_NETBUF, buf);
+      memp_free(MEMP_NETBUF, buf);
       sys_mbox_free(conn->recvmbox);
       conn->recvmbox = SYS_MBOX_NULL;
       return NULL;
@@ -526,7 +556,7 @@ netconn_recv(struct netconn *conn)
     buf->fromaddr = NULL;
 
     /* Let the stack know that we have taken the data. */
-    if ((msg = memp_mallocp(MEMP_API_MSG)) == NULL) {
+    if ((msg = memp_malloc(MEMP_API_MSG)) == NULL) {
       conn->err = ERR_MEM;
       return buf;
     }
@@ -540,7 +570,7 @@ netconn_recv(struct netconn *conn)
     api_msg_post(msg);
 
     sys_mbox_fetch(conn->mbox, NULL);
-    memp_freep(MEMP_API_MSG, msg);
+    memp_free(MEMP_API_MSG, msg);
   } else {
     sys_mbox_fetch(conn->recvmbox, (void **)&buf);
   conn->recv_avail -= buf->p->tot_len;
@@ -557,7 +587,7 @@ netconn_recv(struct netconn *conn)
 
   return buf;
 }
-/*-----------------------------------------------------------------------------------*/
+
 err_t
 netconn_send(struct netconn *conn, struct netbuf *buf)
 {
@@ -571,7 +601,7 @@ netconn_send(struct netconn *conn, struct netbuf *buf)
     return conn->err;
   }
 
-  if ((msg = memp_mallocp(MEMP_API_MSG)) == NULL) {
+  if ((msg = memp_malloc(MEMP_API_MSG)) == NULL) {
     return (conn->err = ERR_MEM);
   }
 
@@ -582,10 +612,10 @@ netconn_send(struct netconn *conn, struct netbuf *buf)
   api_msg_post(msg);
 
   sys_mbox_fetch(conn->mbox, NULL);
-  memp_freep(MEMP_API_MSG, msg);
+  memp_free(MEMP_API_MSG, msg);
   return conn->err;
 }
-/*-----------------------------------------------------------------------------------*/
+
 err_t
 netconn_write(struct netconn *conn, void *dataptr, u16_t size, u8_t copy)
 {
@@ -607,7 +637,7 @@ netconn_write(struct netconn *conn, void *dataptr, u16_t size, u8_t copy)
     }
   }
 
-  if ((msg = memp_mallocp(MEMP_API_MSG)) == NULL) {
+  if ((msg = memp_malloc(MEMP_API_MSG)) == NULL) {
     return (conn->err = ERR_MEM);
   }
   msg->type = API_MSG_WRITE;
@@ -652,7 +682,7 @@ netconn_write(struct netconn *conn, void *dataptr, u16_t size, u8_t copy)
     }
   }
  ret:
-  memp_freep(MEMP_API_MSG, msg);
+  memp_free(MEMP_API_MSG, msg);
   conn->state = NETCONN_NONE;
   if (conn->sem != SYS_SEM_NULL) {
     sys_sem_free(conn->sem);
@@ -661,7 +691,7 @@ netconn_write(struct netconn *conn, void *dataptr, u16_t size, u8_t copy)
   
   return conn->err;
 }
-/*-----------------------------------------------------------------------------------*/
+
 err_t
 netconn_close(struct netconn *conn)
 {
@@ -670,7 +700,7 @@ netconn_close(struct netconn *conn)
   if (conn == NULL) {
     return ERR_VAL;
   }
-  if ((msg = memp_mallocp(MEMP_API_MSG)) == NULL) {
+  if ((msg = memp_malloc(MEMP_API_MSG)) == NULL) {
     return (conn->err = ERR_MEM);
   }
 
@@ -686,13 +716,13 @@ netconn_close(struct netconn *conn)
     goto again;
   }
   conn->state = NETCONN_NONE;
-  memp_freep(MEMP_API_MSG, msg);
+  memp_free(MEMP_API_MSG, msg);
   return conn->err;
 }
-/*-----------------------------------------------------------------------------------*/
+
 err_t
 netconn_err(struct netconn *conn)
 {
   return conn->err;
 }
-/*-----------------------------------------------------------------------------------*/
+

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, 
@@ -37,6 +37,37 @@
 #include "lwip/sys.h"
 #include "lwip/tcpip.h"
 
+#if LWIP_RAW
+static int
+recv_raw(void *arg, struct raw_pcb *pcb, struct pbuf *p,
+    struct ip_addr *addr)
+{
+  struct netbuf *buf;
+  struct netconn *conn;
+
+  conn = arg;
+  if (!conn) return 0;
+
+  if (conn->recvmbox != SYS_MBOX_NULL) {
+    if (!(buf = memp_malloc(MEMP_NETBUF))) {
+      return 0;
+    }
+    pbuf_ref(p);
+    buf->p = p;
+    buf->ptr = p;
+    buf->fromaddr = addr;
+    buf->fromport = pcb->protocol;
+
+    conn->recv_avail += p->tot_len;
+    /* Register event with callback */
+    if (conn->callback)
+        (*conn->callback)(conn, NETCONN_EVT_RCVPLUS, p->tot_len);
+    sys_mbox_post(conn->recvmbox, buf);
+  }
+
+  return 0; /* do not eat the packet */
+}
+#endif
 #if LWIP_UDP
 static void
 recv_udp(void *arg, struct udp_pcb *pcb, struct pbuf *p,
@@ -52,7 +83,7 @@ recv_udp(void *arg, struct udp_pcb *pcb, struct pbuf *p,
     return;
   }
   if (conn->recvmbox != SYS_MBOX_NULL) {
-    buf = memp_mallocp(MEMP_NETBUF);
+    buf = memp_malloc(MEMP_NETBUF);
     if (buf == NULL) {
       pbuf_free(p);
       return;
@@ -72,7 +103,7 @@ recv_udp(void *arg, struct udp_pcb *pcb, struct pbuf *p,
 }
 #endif /* LWIP_UDP */
 #if LWIP_TCP
-/*-----------------------------------------------------------------------------------*/
+
 static err_t
 recv_tcp(void *arg, struct tcp_pcb *pcb, struct pbuf *p, err_t err)
 {
@@ -103,7 +134,7 @@ recv_tcp(void *arg, struct tcp_pcb *pcb, struct pbuf *p, err_t err)
   return ERR_OK;
 }
 
-/*-----------------------------------------------------------------------------------*/
+
 static err_t
 poll_tcp(void *arg, struct tcp_pcb *pcb)
 {
@@ -117,7 +148,7 @@ poll_tcp(void *arg, struct tcp_pcb *pcb)
   }
   return ERR_OK;
 }
-/*-----------------------------------------------------------------------------------*/
+
 static err_t
 sent_tcp(void *arg, struct tcp_pcb *pcb, u16_t len)
 {
@@ -134,7 +165,7 @@ sent_tcp(void *arg, struct tcp_pcb *pcb, u16_t len)
   
   return ERR_OK;
 }
-/*-----------------------------------------------------------------------------------*/
+
 static void
 err_tcp(void *arg, err_t err)
 {
@@ -165,7 +196,7 @@ err_tcp(void *arg, err_t err)
     sys_sem_signal(conn->sem);
   }
 }
-/*-----------------------------------------------------------------------------------*/
+
 static void
 setup_tcp(struct netconn *conn)
 {
@@ -178,7 +209,7 @@ setup_tcp(struct netconn *conn)
   tcp_poll(pcb, poll_tcp, 4);
   tcp_err(pcb, err_tcp);
 }
-/*-----------------------------------------------------------------------------------*/
+
 static err_t
 accept_function(void *arg, struct tcp_pcb *newpcb, err_t err)
 {
@@ -193,7 +224,7 @@ accept_function(void *arg, struct tcp_pcb *newpcb, err_t err)
 #endif /* API_MSG_DEBUG */
   conn = (struct netconn *)arg;
   mbox = conn->acceptmbox;
-  newconn = memp_mallocp(MEMP_NETCONN);
+  newconn = memp_malloc(MEMP_NETCONN);
   if (newconn == NULL) {
     return ERR_MEM;
   }
@@ -234,17 +265,83 @@ accept_function(void *arg, struct tcp_pcb *newpcb, err_t err)
   return ERR_OK;
 }
 #endif /* LWIP_TCP */
-/*-----------------------------------------------------------------------------------*/
+
 static void
 do_newconn(struct api_msg_msg *msg)
 {
+   if(msg->conn->pcb.tcp != NULL) {
+   /* This "new" connection already has a PCB allocated. */
+   /* Is this an error condition? Should it be deleted? 
+      We currently just are happy and return. */
+     sys_mbox_post(msg->conn->mbox, NULL);
+     return;
+   }
+
+   msg->conn->err = ERR_OK;
+
+   /* Allocate a PCB for this connection */
+   switch(msg->conn->type) {
+#if LWIP_RAW
+   case NETCONN_RAW:
+      msg->conn->pcb.raw = raw_new(msg->msg.bc.port); /* misusing the port field */
+      raw_recv(msg->conn->pcb.raw, recv_raw, msg->conn);
+     break;
+#endif
+#if LWIP_UDP
+   case NETCONN_UDPLITE:
+      msg->conn->pcb.udp = udp_new();
+      if(msg->conn->pcb.udp == NULL) {
+         msg->conn->err = ERR_MEM;
+         break;
+      }
+      udp_setflags(msg->conn->pcb.udp, UDP_FLAGS_UDPLITE);
+      udp_recv(msg->conn->pcb.udp, recv_udp, msg->conn);
+      break;
+   case NETCONN_UDPNOCHKSUM:
+      msg->conn->pcb.udp = udp_new();
+      if(msg->conn->pcb.udp == NULL) {
+         msg->conn->err = ERR_MEM;
+         break;
+      }
+      udp_setflags(msg->conn->pcb.udp, UDP_FLAGS_NOCHKSUM);
+      udp_recv(msg->conn->pcb.udp, recv_udp, msg->conn);
+      break;
+   case NETCONN_UDP:
+      msg->conn->pcb.udp = udp_new();
+      if(msg->conn->pcb.udp == NULL) {
+         msg->conn->err = ERR_MEM;
+         break;
+      }
+      udp_recv(msg->conn->pcb.udp, recv_udp, msg->conn);
+      break;
+#endif /* LWIP_UDP */
+#if LWIP_TCP
+   case NETCONN_TCP:
+      msg->conn->pcb.tcp = tcp_new();
+      if(msg->conn->pcb.tcp == NULL) {
+         msg->conn->err = ERR_MEM;
+         break;
+      }
+      setup_tcp(msg->conn);
+      break;
+#endif
+   }
+   
+  
+  sys_mbox_post(msg->conn->mbox, NULL);
 }
-/*-----------------------------------------------------------------------------------*/
+
+
 static void
 do_delconn(struct api_msg_msg *msg)
 {
   if (msg->conn->pcb.tcp != NULL) {
     switch (msg->conn->type) {
+#if LWIP_RAW
+    case NETCONN_RAW:
+      raw_remove(msg->conn->pcb.raw);
+      break;
+#endif
 #if LWIP_UDP
     case NETCONN_UDPLITE:
       /* FALLTHROUGH */
@@ -287,12 +384,18 @@ do_delconn(struct api_msg_msg *msg)
     sys_mbox_post(msg->conn->mbox, NULL);
   }
 }
-/*-----------------------------------------------------------------------------------*/
+
 static void
 do_bind(struct api_msg_msg *msg)
 {
   if (msg->conn->pcb.tcp == NULL) {
     switch (msg->conn->type) {
+#if LWIP_RAW
+    case NETCONN_RAW:
+      msg->conn->pcb.raw = raw_new(msg->msg.bc.port); /* misusing the port field as protocol */
+      raw_recv(msg->conn->pcb.raw, recv_raw, msg->conn);
+      break;
+#endif
 #if LWIP_UDP
     case NETCONN_UDPLITE:
       msg->conn->pcb.udp = udp_new();
@@ -319,6 +422,11 @@ do_bind(struct api_msg_msg *msg)
     }
   }
   switch (msg->conn->type) {
+#if LWIP_RAW
+  case NETCONN_RAW:
+    msg->conn->err = raw_bind(msg->conn->pcb.raw,msg->msg.bc.ipaddr);
+    break;
+#endif
 #if LWIP_UDP
   case NETCONN_UDPLITE:
     /* FALLTHROUGH */
@@ -339,7 +447,7 @@ do_bind(struct api_msg_msg *msg)
   sys_mbox_post(msg->conn->mbox, NULL);
 }
 #if LWIP_TCP
-/*-----------------------------------------------------------------------------------*/
+
 static err_t
 do_connected(void *arg, struct tcp_pcb *pcb, err_t err)
 {
@@ -359,12 +467,18 @@ do_connected(void *arg, struct tcp_pcb *pcb, err_t err)
   return ERR_OK;
 }
 #endif  
-/*-----------------------------------------------------------------------------------*/
+
 static void
 do_connect(struct api_msg_msg *msg)
 {
   if (msg->conn->pcb.tcp == NULL) {
     switch (msg->conn->type) {
+#if LWIP_RAW
+    case NETCONN_RAW:
+      msg->conn->pcb.raw = raw_new(msg->msg.bc.port); /* misusing the port field as protocol */
+      raw_recv(msg->conn->pcb.raw, recv_raw, msg->conn);
+      break;
+#endif
 #if LWIP_UDP
     case NETCONN_UDPLITE:
       msg->conn->pcb.udp = udp_new();
@@ -410,6 +524,12 @@ do_connect(struct api_msg_msg *msg)
     }
   }
   switch (msg->conn->type) {
+#if LWIP_RAW
+  case NETCONN_RAW:
+    raw_connect(msg->conn->pcb.raw, msg->msg.bc.ipaddr);
+    sys_mbox_post(msg->conn->mbox, NULL);
+    break;
+#endif
 #if LWIP_UDP
   case NETCONN_UDPLITE:
     /* FALLTHROUGH */
@@ -428,6 +548,7 @@ do_connect(struct api_msg_msg *msg)
     do_connected);
     /*tcp_output(msg->conn->pcb.tcp);*/
 #endif
+
   default:
     break;
   }
@@ -438,6 +559,11 @@ do_disconnect(struct api_msg_msg *msg)
 {
 
   switch (msg->conn->type) {
+#if LWIP_RAW
+  case NETCONN_RAW:
+    /* Do nothing as connecting is only a helper for upper lwip layers */
+    break;
+#endif
 #if LWIP_UDP
   case NETCONN_UDPLITE:
     /* FALLTHROUGH */
@@ -453,12 +579,17 @@ do_disconnect(struct api_msg_msg *msg)
   sys_mbox_post(msg->conn->mbox, NULL);
 }
 
-/*-----------------------------------------------------------------------------------*/
+
 static void
 do_listen(struct api_msg_msg *msg)
 {
   if (msg->conn->pcb.tcp != NULL) {
     switch (msg->conn->type) {
+#if LWIP_RAW
+    case NETCONN_RAW:
+      LWIP_DEBUGF(API_MSG_DEBUG, ("api_msg: listen RAW: cannot listen for RAW.\n"));
+      break;
+#endif
 #if LWIP_UDP
     case NETCONN_UDPLITE:
       /* FALLTHROUGH */
@@ -491,12 +622,17 @@ do_listen(struct api_msg_msg *msg)
   }
   sys_mbox_post(msg->conn->mbox, NULL);
 }
-/*-----------------------------------------------------------------------------------*/
+
 static void
 do_accept(struct api_msg_msg *msg)
 {
   if (msg->conn->pcb.tcp != NULL) {
     switch (msg->conn->type) {
+#if LWIP_RAW
+    case NETCONN_RAW:
+      LWIP_DEBUGF(API_MSG_DEBUG, ("api_msg: accept RAW: cannot accept for RAW.\n"));
+      break;
+#endif
 #if LWIP_UDP
     case NETCONN_UDPLITE:
       /* FALLTHROUGH */
@@ -511,12 +647,17 @@ do_accept(struct api_msg_msg *msg)
     }
   }
 }
-/*-----------------------------------------------------------------------------------*/
+
 static void
 do_send(struct api_msg_msg *msg)
 {
   if (msg->conn->pcb.tcp != NULL) {
     switch (msg->conn->type) {
+#if LWIP_RAW
+    case NETCONN_RAW:
+      raw_send(msg->conn->pcb.raw, msg->msg.p);
+      break;
+#endif
 #if LWIP_UDP
     case NETCONN_UDPLITE:
       /* FALLTHROUGH */
@@ -532,7 +673,7 @@ do_send(struct api_msg_msg *msg)
   }
   sys_mbox_post(msg->conn->mbox, NULL);
 }
-/*-----------------------------------------------------------------------------------*/
+
 static void
 do_recv(struct api_msg_msg *msg)
 {
@@ -545,7 +686,7 @@ do_recv(struct api_msg_msg *msg)
 #endif  
   sys_mbox_post(msg->conn->mbox, NULL);
 }
-/*-----------------------------------------------------------------------------------*/
+
 static void
 do_write(struct api_msg_msg *msg)
 {
@@ -554,6 +695,11 @@ do_write(struct api_msg_msg *msg)
 #endif  
   if (msg->conn->pcb.tcp != NULL) {
     switch (msg->conn->type) {
+#if LWIP_RAW
+    case NETCONN_RAW:
+      msg->conn->err = ERR_VAL;
+      break;
+#endif
 #if LWIP_UDP 
     case NETCONN_UDPLITE:
       /* FALLTHROUGH */
@@ -571,7 +717,7 @@ do_write(struct api_msg_msg *msg)
    segments when new outgoing data arrives from the user if any
    previously transmitted data on the connection remains
    unacknowledged. */
-      if (err == ERR_OK && msg->conn->pcb.tcp->unacked == NULL) {
+      if(err == ERR_OK && (msg->conn->pcb.tcp->unacked == NULL || (msg->conn->pcb.tcp->flags & TF_NODELAY)) ) {
   tcp_output(msg->conn->pcb.tcp);
       }
       msg->conn->err = err;
@@ -588,7 +734,7 @@ do_write(struct api_msg_msg *msg)
   }
   sys_mbox_post(msg->conn->mbox, NULL);
 }
-/*-----------------------------------------------------------------------------------*/
+
 static void
 do_close(struct api_msg_msg *msg)
 {
@@ -598,6 +744,10 @@ do_close(struct api_msg_msg *msg)
 
   if (msg->conn->pcb.tcp != NULL) {
     switch (msg->conn->type) {
+#if LWIP_RAW
+    case NETCONN_RAW:
+      break;
+#endif
 #if LWIP_UDP
     case NETCONN_UDPLITE:
       /* FALLTHROUGH */
@@ -619,7 +769,7 @@ do_close(struct api_msg_msg *msg)
   }
   sys_mbox_post(msg->conn->mbox, NULL);
 }
-/*-----------------------------------------------------------------------------------*/
+
 typedef void (* api_msg_decode)(struct api_msg_msg *msg);
 static api_msg_decode decode[API_MSG_MAX] = {
   do_newconn,
@@ -639,12 +789,12 @@ api_msg_input(struct api_msg *msg)
 {  
   decode[msg->type](&(msg->msg));
 }
-/*-----------------------------------------------------------------------------------*/
+
 void
 api_msg_post(struct api_msg *msg)
 {
   tcpip_apimsg(msg);
 }
-/*-----------------------------------------------------------------------------------*/
+
 
 

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, 
@@ -47,13 +47,13 @@ static char *err_strerr[] = {"Ok.",
            "Address in use."
 };
 
-/*-----------------------------------------------------------------------------------*/
+
 char *
 lwip_strerr(err_t err)
 {
   return err_strerr[-err];
 
 }
-/*-----------------------------------------------------------------------------------*/
+
 
 #endif /* LWIP_DEBUG */

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
@@ -99,7 +100,7 @@ static int err_to_errno_table[11] = {
       set_errno(sk->err); \
 } while (0)
 
-/*-----------------------------------------------------------------------------------*/
+
 static struct lwip_socket *
 get_socket(int s)
 {
@@ -121,7 +122,7 @@ get_socket(int s)
 
   return sock;
 }
-/*-----------------------------------------------------------------------------------*/
+
 static int
 alloc_socket(struct netconn *newconn)
 {
@@ -150,7 +151,7 @@ alloc_socket(struct netconn *newconn)
   sys_sem_signal(socksem);
   return -1;
 }
-/*-----------------------------------------------------------------------------------*/
+
 int
 lwip_accept(int s, struct sockaddr *addr, socklen_t *addrlen)
 {
@@ -164,6 +165,7 @@ lwip_accept(int s, struct sockaddr *addr, socklen_t *addrlen)
   LWIP_DEBUGF(SOCKETS_DEBUG, ("lwip_accept(%d)...\n", s));
   sock = get_socket(s);
   if (!sock) {
+    set_errno(EBADF);
     return -1;
   }
 
@@ -197,16 +199,14 @@ lwip_accept(int s, struct sockaddr *addr, socklen_t *addrlen)
   newconn->socket = newsock;
   sys_sem_signal(socksem);
 
-#if SOCKETS_DEBUG
   LWIP_DEBUGF(SOCKETS_DEBUG, ("lwip_accept(%d) returning new sock=%d addr=", s, newsock));
   ip_addr_debug_print(SOCKETS_DEBUG, &naddr);
   LWIP_DEBUGF(SOCKETS_DEBUG, (" port=%u\n", port));
-#endif
 
   sock_set_errno(sock, 0);
   return newsock;
 }
-/*-----------------------------------------------------------------------------------*/
+
 int
 lwip_bind(int s, struct sockaddr *name, socklen_t namelen)
 {
@@ -217,17 +217,16 @@ lwip_bind(int s, struct sockaddr *name, socklen_t namelen)
 
   sock = get_socket(s);
   if (!sock) {
+    set_errno(EBADF);
     return -1;
   }
 
   local_addr.addr = ((struct sockaddr_in *)name)->sin_addr.s_addr;
   local_port = ((struct sockaddr_in *)name)->sin_port;
 
-#if SOCKETS_DEBUG
   LWIP_DEBUGF(SOCKETS_DEBUG, ("lwip_bind(%d, addr=", s));
   ip_addr_debug_print(SOCKETS_DEBUG, &local_addr);
   LWIP_DEBUGF(SOCKETS_DEBUG, (" port=%u)\n", ntohs(local_port)));
-#endif
 
   err = netconn_bind(sock->conn, &local_addr, ntohs(local_port));
 
@@ -241,7 +240,7 @@ lwip_bind(int s, struct sockaddr *name, socklen_t namelen)
   sock_set_errno(sock, 0);
   return 0;
 }
-/*-----------------------------------------------------------------------------------*/
+
 int
 lwip_close(int s)
 {
@@ -257,6 +256,7 @@ lwip_close(int s)
   sock = get_socket(s);
   if (!sock) {
       sys_sem_signal(socksem);
+      set_errno(EBADF);
       return -1;
   }
 
@@ -271,7 +271,7 @@ lwip_close(int s)
   sock_set_errno(sock, 0);
   return 0;
 }
-/*-----------------------------------------------------------------------------------*/
+
 int
 lwip_connect(int s, struct sockaddr *name, socklen_t namelen)
 {
@@ -280,6 +280,7 @@ lwip_connect(int s, struct sockaddr *name, socklen_t namelen)
 
   sock = get_socket(s);
   if (!sock) {
+    set_errno(EBADF);
     return -1;
   }
 
@@ -293,11 +294,9 @@ lwip_connect(int s, struct sockaddr *name, socklen_t namelen)
     remote_addr.addr = ((struct sockaddr_in *)name)->sin_addr.s_addr;
     remote_port = ((struct sockaddr_in *)name)->sin_port;
 
-#if SOCKETS_DEBUG
     LWIP_DEBUGF(SOCKETS_DEBUG, ("lwip_connect(%d, addr=", s));
     ip_addr_debug_print(SOCKETS_DEBUG, &remote_addr);
     LWIP_DEBUGF(SOCKETS_DEBUG, (" port=%u)\n", ntohs(remote_port)));
-#endif
 
     err = netconn_connect(sock->conn, &remote_addr, ntohs(remote_port));
    }
@@ -312,7 +311,7 @@ lwip_connect(int s, struct sockaddr *name, socklen_t namelen)
   sock_set_errno(sock, 0);
   return 0;
 }
-/*-----------------------------------------------------------------------------------*/
+
 int
 lwip_listen(int s, int backlog)
 {
@@ -322,6 +321,7 @@ lwip_listen(int s, int backlog)
   LWIP_DEBUGF(SOCKETS_DEBUG, ("lwip_listen(%d, backlog=%d)\n", s, backlog));
   sock = get_socket(s);
   if (!sock) {
+    set_errno(EBADF);
     return -1;
   }
 
@@ -336,7 +336,7 @@ lwip_listen(int s, int backlog)
   sock_set_errno(sock, 0);
   return 0;
 }
-/*-----------------------------------------------------------------------------------*/
+
 int
 lwip_recvfrom(int s, void *mem, int len, unsigned int flags,
         struct sockaddr *from, socklen_t *fromlen)
@@ -351,6 +351,7 @@ lwip_recvfrom(int s, void *mem, int len, unsigned int flags,
   LWIP_DEBUGF(SOCKETS_DEBUG, ("lwip_recvfrom(%d, %p, %d, 0x%x, ..)\n", s, mem, len, flags));
   sock = get_socket(s);
   if (!sock) {
+    set_errno(EBADF);
     return -1;
   }
 
@@ -411,11 +412,9 @@ lwip_recvfrom(int s, void *mem, int len, unsigned int flags,
 
     memcpy(from, &sin, *fromlen);
 
-#if SOCKETS_DEBUG
     LWIP_DEBUGF(SOCKETS_DEBUG, ("lwip_recvfrom(%d): addr=", s));
     ip_addr_debug_print(SOCKETS_DEBUG, addr);
     LWIP_DEBUGF(SOCKETS_DEBUG, (" port=%u len=%u\n", port, copylen));
-#endif
   } else {
 #if SOCKETS_DEBUG > 0
     addr = netbuf_fromaddr(buf);
@@ -444,19 +443,19 @@ lwip_recvfrom(int s, void *mem, int len, unsigned int flags,
   sock_set_errno(sock, 0);
   return copylen;
 }
-/*-----------------------------------------------------------------------------------*/
+
 int
 lwip_read(int s, void *mem, int len)
 {
   return lwip_recvfrom(s, mem, len, 0, NULL, NULL);
 }
-/*-----------------------------------------------------------------------------------*/
+
 int
 lwip_recv(int s, void *mem, int len, unsigned int flags)
 {
   return lwip_recvfrom(s, mem, len, flags, NULL, NULL);
 }
-/*-----------------------------------------------------------------------------------*/
+
 int
 lwip_send(int s, void *data, int size, unsigned int flags)
 {
@@ -468,11 +467,15 @@ lwip_send(int s, void *data, int size, unsigned int flags)
 
   sock = get_socket(s);
   if (!sock) {
+    set_errno(EBADF);
     return -1;
   }
 
   switch (netconn_type(sock->conn)) {
+  case NETCONN_RAW:
   case NETCONN_UDP:
+  case NETCONN_UDPLITE:
+  case NETCONN_UDPNOCHKSUM:
     /* create a buffer */
     buf = netbuf_new();
 
@@ -509,7 +512,7 @@ lwip_send(int s, void *data, int size, unsigned int flags)
   sock_set_errno(sock, 0);
   return size;
 }
-/*-----------------------------------------------------------------------------------*/
+
 int
 lwip_sendto(int s, void *data, int size, unsigned int flags,
        struct sockaddr *to, socklen_t tolen)
@@ -521,6 +524,7 @@ lwip_sendto(int s, void *data, int size, unsigned int flags,
 
   sock = get_socket(s);
   if (!sock) {
+    set_errno(EBADF);
     return -1;
   }
 
@@ -530,11 +534,9 @@ lwip_sendto(int s, void *data, int size, unsigned int flags,
   remote_addr.addr = ((struct sockaddr_in *)to)->sin_addr.s_addr;
   remote_port = ((struct sockaddr_in *)to)->sin_port;
 
-#if SOCKETS_DEBUG
   LWIP_DEBUGF(SOCKETS_DEBUG, ("lwip_sendto(%d, data=%p, size=%d, flags=0x%x to=", s, data, size, flags));
   ip_addr_debug_print(SOCKETS_DEBUG, &remote_addr);
   LWIP_DEBUGF(SOCKETS_DEBUG, (" port=%u\n", ntohs(remote_port)));
-#endif
 
   netconn_connect(sock->conn, &remote_addr, ntohs(remote_port));
 
@@ -548,7 +550,7 @@ lwip_sendto(int s, void *data, int size, unsigned int flags,
   netconn_disconnect(sock->conn);
   return ret;
 }
-/*-----------------------------------------------------------------------------------*/
+
 int
 lwip_socket(int domain, int type, int protocol)
 {
@@ -557,6 +559,10 @@ lwip_socket(int domain, int type, int protocol)
 
   /* create a netconn */
   switch (type) {
+  case SOCK_RAW:
+    conn = netconn_new_with_proto_and_callback(NETCONN_RAW, protocol, event_callback);
+    LWIP_DEBUGF(SOCKETS_DEBUG, ("lwip_socket(%s, SOCK_RAW, %d) = ", domain == PF_INET ? "PF_INET" : "UNKNOWN", protocol));
+    break;
   case SOCK_DGRAM:
     conn = netconn_new_with_callback(NETCONN_UDP, event_callback);
     LWIP_DEBUGF(SOCKETS_DEBUG, ("lwip_socket(%s, SOCK_DGRAM, %d) = ", domain == PF_INET ? "PF_INET" : "UNKNOWN", protocol));
@@ -589,14 +595,14 @@ lwip_socket(int domain, int type, int protocol)
   set_errno(0);
   return i;
 }
-/*-----------------------------------------------------------------------------------*/
+
 int
 lwip_write(int s, void *data, int size)
 {
    return lwip_send(s, data, size, 0);
 }
 
-/*-----------------------------------------------------------------------------------*/
+
 static int
 lwip_selscan(int maxfdp1, fd_set *readset, fd_set *writeset, fd_set *exceptset)
 {
@@ -643,7 +649,7 @@ lwip_selscan(int maxfdp1, fd_set *readset, fd_set *writeset, fd_set *exceptset)
 }
 
 
-/*-----------------------------------------------------------------------------------*/
+
 int
 lwip_select(int maxfdp1, fd_set *readset, fd_set *writeset, fd_set *exceptset,
                struct timeval *timeout)
@@ -723,7 +729,7 @@ lwip_select(int maxfdp1, fd_set *readset, fd_set *writeset, fd_set *exceptset,
             /* Wait forever */
             msectimeout = 0;
         else
-            msectimeout =  ((timeout->tv_sec * 1000) + (timeout->tv_usec /1000));
+            msectimeout =  ((timeout->tv_sec * 1000) + ((timeout->tv_usec + 500)/1000));
 
         i = sys_sem_wait_timeout(select_cb.sem, msectimeout);
 
@@ -789,7 +795,7 @@ lwip_select(int maxfdp1, fd_set *readset, fd_set *writeset, fd_set *exceptset,
     return nready;
 }
 
-/*-----------------------------------------------------------------------------------*/
+
 static void
 event_callback(struct netconn *conn, enum netconn_evt evt, u16_t len)
 {
@@ -878,7 +884,7 @@ event_callback(struct netconn *conn, enum netconn_evt evt, u16_t len)
 
 }
 
-/*-----------------------------------------------------------------------------------*/
+
 
 
 int lwip_shutdown(int s, int how)
@@ -895,6 +901,7 @@ int lwip_getpeername (int s, struct sockaddr *name, socklen_t *namelen)
 
   sock = get_socket(s);
   if (!sock) {
+    set_errno(EBADF);
     return -1;
   }
 
@@ -905,11 +912,9 @@ int lwip_getpeername (int s, struct sockaddr *name, socklen_t *namelen)
   /* get the IP address and port of the remote host */
   netconn_peer(sock->conn, &naddr, &sin.sin_port);
 
-#if SOCKETS_DEBUG
   LWIP_DEBUGF(SOCKETS_DEBUG, ("lwip_getpeername(%d, addr=", s));
   ip_addr_debug_print(SOCKETS_DEBUG, &naddr);
   LWIP_DEBUGF(SOCKETS_DEBUG, (" port=%d)\n", sin.sin_port));
-#endif
 
   sin.sin_port = htons(sin.sin_port);
   sin.sin_addr.s_addr = naddr.addr;
@@ -930,6 +935,7 @@ int lwip_getsockname (int s, struct sockaddr *name, socklen_t *namelen)
 
   sock = get_socket(s);
   if (!sock) {
+    set_errno(EBADF);
     return -1;
   }
 
@@ -940,11 +946,9 @@ int lwip_getsockname (int s, struct sockaddr *name, socklen_t *namelen)
   /* get the IP address and port of the remote host */
   netconn_addr(sock->conn, &naddr, &sin.sin_port);
 
-#if SOCKETS_DEBUG
   LWIP_DEBUGF(SOCKETS_DEBUG, ("lwip_getsockname(%d, addr=", s));
   ip_addr_debug_print(SOCKETS_DEBUG, naddr);
   LWIP_DEBUGF(SOCKETS_DEBUG, (" port=%d)\n", sin.sin_port));
-#endif
 
   sin.sin_port = htons(sin.sin_port);
   sin.sin_addr.s_addr = naddr->addr;
@@ -959,32 +963,190 @@ int lwip_getsockname (int s, struct sockaddr *name, socklen_t *namelen)
 
 int lwip_getsockopt (int s, int level, int optname, void *optval, socklen_t *optlen)
 {
-  int err = ENOSYS;
+  int err = 0;
   struct lwip_socket *sock = get_socket(s);
 
-  if (!sock) {
+  if(!sock) {
+   	set_errno(EBADF);
     return -1;
   }
 
-  if (level == SOL_SOCKET) {
-    switch (optname) {
+  if( NULL == optval || NULL == optlen ) {
+    sock_set_errno( sock, EFAULT );
+    return -1;
+  }
+
+  /* Do length and type checks for the various options first, to keep it readable. */
+  switch( level ) {
+   
+/* Level: SOL_SOCKET */
+  case SOL_SOCKET:
+      switch(optname) {
+         
+      case SO_ACCEPTCONN:
+      case SO_BROADCAST:
+      /* UNIMPL case SO_DEBUG: */
+      /* UNIMPL case SO_DONTROUTE: */
       case SO_ERROR:
-        if (!optval || !optlen || (*optlen != sizeof(int))) {
+      case SO_KEEPALIVE:
+      /* UNIMPL case SO_OOBINLINE: */
+      /* UNIMPL case SO_RCVBUF: */
+      /* UNIMPL case SO_SNDBUF: */
+      /* UNIMPL case SO_RCVLOWAT: */
+      /* UNIMPL case SO_SNDLOWAT: */
+#if SO_REUSE
+      case SO_REUSEADDR:
+      case SO_REUSEPORT:
+#endif /* SO_REUSE */
+      case SO_TYPE:
+      /* UNIMPL case SO_USELOOPBACK: */
+        if( *optlen < sizeof(int) ) {
           err = EINVAL;
-          break;
         }
-        *(int *)optval = sock->err;
-        sock->err = 0;
-        LWIP_DEBUGF(SOCKETS_DEBUG, ("lwip_getsockopt(%d, SOL_SOCKET, SO_ERROR) = %d\n", s, *(int *)optval));
-        err = 0;
-        break;
+          break;
+
       default:
         LWIP_DEBUGF(SOCKETS_DEBUG, ("lwip_getsockopt(%d, SOL_SOCKET, UNIMPL: optname=0x%x, ..)\n", s, optname));
+        err = ENOPROTOOPT;
+      }  /* switch */
+      break;
+                     
+/* Level: IPPROTO_IP */
+  case IPPROTO_IP:
+      switch(optname) {
+      /* UNIMPL case IP_HDRINCL: */
+      /* UNIMPL case IP_RCVDSTADDR: */
+      /* UNIMPL case IP_RCVIF: */
+      case IP_TTL:
+      case IP_TOS:
+        if( *optlen < sizeof(int) ) {
+          err = EINVAL;
+        }
+        break;
+
+      default:
+        LWIP_DEBUGF(SOCKETS_DEBUG, ("lwip_getsockopt(%d, IPPROTO_IP, UNIMPL: optname=0x%x, ..)\n", s, optname));
+        err = ENOPROTOOPT;
+      }  /* switch */
+      break;
+         
+/* Level: IPPROTO_TCP */
+  case IPPROTO_TCP:
+      if( *optlen < sizeof(int) ) {
+        err = EINVAL;
         break;
     }
-  } else {
-    LWIP_DEBUGF(SOCKETS_DEBUG, ("lwip_getsockopt(%d, level=0x%x, UNIMPL: optname=0x%x, ..)\n", s, level, optname));
+      
+      /* If this is no TCP socket, ignore any options. */
+      if ( sock->conn->type != NETCONN_TCP ) return 0;
+
+      switch( optname ) {
+      case TCP_NODELAY:
+      case TCP_KEEPALIVE:
+        break;
+         
+      default:
+        LWIP_DEBUGF(SOCKETS_DEBUG, ("lwip_getsockopt(%d, IPPROTO_TCP, UNIMPL: optname=0x%x, ..)\n", s, optname));
+        err = ENOPROTOOPT;
+      }  /* switch */
+      break;
+
+/* UNDEFINED LEVEL */
+  default:
+      LWIP_DEBUGF(SOCKETS_DEBUG, ("lwip_getsockopt(%d, level=0x%x, UNIMPL: optname=0x%x, ..)\n", s, level, optname));
+      err = ENOPROTOOPT;
+  }  /* switch */
+
+   
+  if( 0 != err ) {
+    sock_set_errno(sock, err);
+    return -1;
   }
+   
+
+
+  /* Now do the actual option processing */
+
+  switch(level) {
+   
+/* Level: SOL_SOCKET */
+  case SOL_SOCKET:
+    switch( optname ) {
+
+    /* The option flags */
+    case SO_ACCEPTCONN:
+    case SO_BROADCAST:
+    /* UNIMPL case SO_DEBUG: */
+    /* UNIMPL case SO_DONTROUTE: */
+    case SO_KEEPALIVE:
+    /* UNIMPL case SO_OOBINCLUDE: */
+#if SO_REUSE
+    case SO_REUSEADDR:
+    case SO_REUSEPORT:
+#endif /* SO_REUSE */
+    /*case SO_USELOOPBACK: UNIMPL */
+      *(int*)optval = sock->conn->pcb.tcp->so_options & optname;
+      LWIP_DEBUGF(SOCKETS_DEBUG, ("lwip_getsockopt(%d, SOL_SOCKET, optname=0x%x, ..) = %s\n", s, optname, (*(int*)optval?"on":"off")));
+      break;
+
+    case SO_TYPE:
+      switch (sock->conn->type) {
+      case NETCONN_RAW:
+        *(int*)optval = SOCK_RAW;
+        break;
+      case NETCONN_TCP:
+        *(int*)optval = SOCK_STREAM;
+        break;
+      case NETCONN_UDP:
+      case NETCONN_UDPLITE:
+      case NETCONN_UDPNOCHKSUM:
+        *(int*)optval = SOCK_DGRAM;
+        break;
+      default: /* unrecognized socket type */
+        *(int*)optval = sock->conn->type;
+        LWIP_DEBUGF(SOCKETS_DEBUG, ("lwip_getsockopt(%d, SOL_SOCKET, SO_TYPE): unrecognized socket type %d\n", s, *(int *)optval));
+      }  /* switch */
+      LWIP_DEBUGF(SOCKETS_DEBUG, ("lwip_getsockopt(%d, SOL_SOCKET, SO_TYPE) = %d\n", s, *(int *)optval));
+      break;
+
+    case SO_ERROR:
+      *(int *)optval = sock->err;
+      sock->err = 0;
+      LWIP_DEBUGF(SOCKETS_DEBUG, ("lwip_getsockopt(%d, SOL_SOCKET, SO_ERROR) = %d\n", s, *(int *)optval));
+      break;
+    }  /* switch */
+    break;
+
+/* Level: IPPROTO_IP */
+  case IPPROTO_IP:
+    switch( optname ) {
+    case IP_TTL:
+      *(int*)optval = sock->conn->pcb.tcp->ttl;
+      LWIP_DEBUGF(SOCKETS_DEBUG, ("lwip_getsockopt(%d, IPPROTO_IP, IP_TTL) = %d\n", s, *(int *)optval));
+      break;
+    case IP_TOS:
+      *(int*)optval = sock->conn->pcb.tcp->tos;
+      LWIP_DEBUGF(SOCKETS_DEBUG, ("lwip_getsockopt(%d, IPPROTO_IP, IP_TOS) = %d\n", s, *(int *)optval));
+      break;
+    }  /* switch */
+    break;
+
+/* Level: IPPROTO_TCP */
+  case IPPROTO_TCP:
+    switch( optname ) {
+    case TCP_NODELAY:
+      *(int*)optval = (sock->conn->pcb.tcp->flags & TF_NODELAY);
+      LWIP_DEBUGF(SOCKETS_DEBUG, ("lwip_getsockopt(%d, IPPROTO_TCP, TCP_NODELAY) = %s\n", s, (*(int*)optval)?"on":"off") );
+      break;
+    case TCP_KEEPALIVE:
+      *(int*)optval = sock->conn->pcb.tcp->keepalive;
+      LWIP_DEBUGF(SOCKETS_DEBUG, ("lwip_getsockopt(%d, IPPROTO_IP, TCP_KEEPALIVE) = %d\n", s, *(int *)optval));
+      break;
+    }  /* switch */
+    break;
+  }
+
+
   sock_set_errno(sock, err);
   return err ? -1 : 0;
 }
@@ -992,27 +1154,165 @@ int lwip_getsockopt (int s, int level, int optname, void *optval, socklen_t *opt
 int lwip_setsockopt (int s, int level, int optname, const void *optval, socklen_t optlen)
 {
   struct lwip_socket *sock = get_socket(s);
-  int err = ENOSYS;
+  int err = 0;
 
-  if (!sock) {
+  if(!sock) {
+   	set_errno(EBADF);
     return -1;
   }
 
-  if (level == SOL_SOCKET) {
-    switch (optname) {
-      case SO_REUSEADDR:
-        LWIP_DEBUGF(SOCKETS_DEBUG, ("lwip_setsockopt(%d, SOL_SOCKET, SO_REUSEADDR, ..)\n", s));
-           /* XXX just pretend we support this for now */
-          err = 0;
+  if( NULL == optval ) {
+    sock_set_errno( sock, EFAULT );
+    return -1;
+  }
+
+
+  /* Do length and type checks for the various options first, to keep it readable. */
+  switch( level ) {
+
+/* Level: SOL_SOCKET */
+  case SOL_SOCKET:
+    switch(optname) {
+
+    case SO_BROADCAST:
+    /* UNIMPL case SO_DEBUG: */
+    /* UNIMPL case SO_DONTROUTE: */
+    case SO_KEEPALIVE:
+    /* UNIMPL case SO_OOBINLINE: */
+    /* UNIMPL case SO_RCVBUF: */
+    /* UNIMPL case SO_SNDBUF: */
+    /* UNIMPL case SO_RCVLOWAT: */
+    /* UNIMPL case SO_SNDLOWAT: */
+#if SO_REUSE
+    case SO_REUSEADDR:
+    case SO_REUSEPORT:
+#endif /* SO_REUSE */
+    /* UNIMPL case SO_USELOOPBACK: */
+      if( optlen < sizeof(int) ) {
+        err = EINVAL;
+      }
+      break;
+    default:
+      LWIP_DEBUGF(SOCKETS_DEBUG, ("lwip_setsockopt(%d, SOL_SOCKET, UNIMPL: optname=0x%x, ..)\n", s, optname));
+      err = ENOPROTOOPT;
+    }  /* switch */
+    break;
+
+/* Level: IPPROTO_IP */
+  case IPPROTO_IP:
+    switch(optname) {
+    /* UNIMPL case IP_HDRINCL: */
+    /* UNIMPL case IP_RCVDSTADDR: */
+    /* UNIMPL case IP_RCVIF: */
+    case IP_TTL:
+    case IP_TOS:
+      if( optlen < sizeof(int) ) {
+        err = EINVAL;
+      }
         break;
       default:
-        LWIP_DEBUGF(SOCKETS_DEBUG, ("lwip_setsockopt(%d, SOL_SOCKET, UNIMPL: optname=0x%x, ..)\n", s, optname));
+      LWIP_DEBUGF(SOCKETS_DEBUG, ("lwip_setsockopt(%d, IPPROTO_IP, UNIMPL: optname=0x%x, ..)\n", s, optname));
+      err = ENOPROTOOPT;
+    }  /* switch */
+    break;
+
+/* Level: IPPROTO_TCP */
+  case IPPROTO_TCP:
+    if( optlen < sizeof(int) ) {
+      err = EINVAL;
         break;
     }
-  } else {
+
+    /* If this is no TCP socket, ignore any options. */
+    if ( sock->conn->type != NETCONN_TCP ) return 0;
+
+    switch( optname ) {
+    case TCP_NODELAY:
+    case TCP_KEEPALIVE:
+      break;
+
+    default:
+      LWIP_DEBUGF(SOCKETS_DEBUG, ("lwip_setsockopt(%d, IPPROTO_TCP, UNIMPL: optname=0x%x, ..)\n", s, optname));
+      err = ENOPROTOOPT;
+    }  /* switch */
+    break;
+
+/* UNDEFINED LEVEL */      
+  default:
     LWIP_DEBUGF(SOCKETS_DEBUG, ("lwip_setsockopt(%d, level=0x%x, UNIMPL: optname=0x%x, ..)\n", s, level, optname));
+    err = ENOPROTOOPT;
+  }  /* switch */
+
+
+  if( 0 != err ) {
+    sock_set_errno(sock, err);
+    return -1;
   }
 
+
+
+  /* Now do the actual option processing */
+
+  switch(level) {
+
+/* Level: SOL_SOCKET */
+  case SOL_SOCKET:
+    switch(optname) {
+
+    /* The option flags */
+    case SO_BROADCAST:
+    /* UNIMPL case SO_DEBUG: */
+    /* UNIMPL case SO_DONTROUTE: */
+    case SO_KEEPALIVE:
+    /* UNIMPL case SO_OOBINCLUDE: */
+#if SO_REUSE
+    case SO_REUSEADDR:
+    case SO_REUSEPORT:
+#endif /* SO_REUSE */
+    /* UNIMPL case SO_USELOOPBACK: */
+      if ( *(int*)optval ) {
+        sock->conn->pcb.tcp->so_options |= optname;
+      } else {
+        sock->conn->pcb.tcp->so_options &= ~optname;
+      }
+      LWIP_DEBUGF(SOCKETS_DEBUG, ("lwip_setsockopt(%d, SOL_SOCKET, optname=0x%x, ..) -> %s\n", s, optname, (*(int*)optval?"on":"off")));
+      break;
+    }  /* switch */
+    break;
+
+/* Level: IPPROTO_IP */
+  case IPPROTO_IP:
+    switch( optname ) {
+    case IP_TTL:
+      sock->conn->pcb.tcp->ttl = (u8_t)(*(int*)optval);
+      LWIP_DEBUGF(SOCKETS_DEBUG, ("lwip_setsockopt(%d, IPPROTO_IP, IP_TTL, ..) -> %u\n", s, sock->conn->pcb.tcp->ttl));
+      break;
+    case IP_TOS:
+      sock->conn->pcb.tcp->tos = (u8_t)(*(int*)optval);
+      LWIP_DEBUGF(SOCKETS_DEBUG, ("lwip_setsockopt(%d, IPPROTO_IP, IP_TOS, ..)-> %u\n", s, sock->conn->pcb.tcp->tos));
+      break;
+    }  /* switch */
+    break;
+
+/* Level: IPPROTO_TCP */
+  case IPPROTO_TCP:
+    switch( optname ) {
+    case TCP_NODELAY:
+      if ( *(int*)optval ) {
+        sock->conn->pcb.tcp->flags |= TF_NODELAY;
+      } else {
+        sock->conn->pcb.tcp->flags &= ~TF_NODELAY;
+      }
+      LWIP_DEBUGF(SOCKETS_DEBUG, ("lwip_setsockopt(%d, IPPROTO_TCP, TCP_NODELAY) -> %s\n", s, (*(int *)optval)?"on":"off") );
+      break;
+    case TCP_KEEPALIVE:
+      sock->conn->pcb.tcp->keepalive = (u32_t)(*(int*)optval);
+      LWIP_DEBUGF(SOCKETS_DEBUG, ("lwip_setsockopt(%d, IPPROTO_TCP, TCP_KEEPALIVE) -> %u\n", s, sock->conn->pcb.tcp->keepalive));
+      break;
+    }  /* switch */
+    break;
+  }  /* switch */
+
   sock_set_errno(sock, err);
   return err ? -1 : 0;
 }
@@ -1021,7 +1321,8 @@ int lwip_ioctl(int s, long cmd, void *argp)
 {
   struct lwip_socket *sock = get_socket(s);
 
-  if (!sock) {
+  if(!sock) {
+   	set_errno(EBADF);
     return -1;
   }
 
@@ -1035,7 +1336,7 @@ int lwip_ioctl(int s, long cmd, void *argp)
     *((u16_t*)argp) = sock->conn->recv_avail;
 
     LWIP_DEBUGF(SOCKETS_DEBUG, ("lwip_ioctl(%d, FIONREAD, %p) = %u\n", s, argp, *((u16_t*)argp)));
-      sock_set_errno(sock, 0);
+    sock_set_errno(sock, 0);
     return 0;
 
   case FIONBIO:
@@ -1044,7 +1345,7 @@ int lwip_ioctl(int s, long cmd, void *argp)
     else
       sock->flags &= ~O_NONBLOCK;
     LWIP_DEBUGF(SOCKETS_DEBUG, ("lwip_ioctl(%d, FIONBIO, %d)\n", s, !!(sock->flags & O_NONBLOCK)));
-      sock_set_errno(sock, 0);
+    sock_set_errno(sock, 0);
     return 0;
 
   default:

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, 
@@ -50,7 +50,7 @@ static sys_mbox_t mbox;
 static int tcpip_tcp_timer_active = 0;
 
 
-/*-----------------------------------------------------------------------------------*/
+
 static void
 tcpip_tcp_timer(void *arg)
 {
@@ -73,7 +73,7 @@ tcp_timer_needed(void)
   }
 }
 #endif /* LWIP_TCP */
-/*-----------------------------------------------------------------------------------*/
+
 static void
 tcpip_thread(void *arg)
 {
@@ -110,16 +110,16 @@ tcpip_thread(void *arg)
     default:
       break;
     }
-    memp_freep(MEMP_TCPIP_MSG, msg);
+    memp_free(MEMP_TCPIP_MSG, msg);
   }
 }
-/*-----------------------------------------------------------------------------------*/
+
 err_t
 tcpip_input(struct pbuf *p, struct netif *inp)
 {
   struct tcpip_msg *msg;
   
-  msg = memp_mallocp(MEMP_TCPIP_MSG);
+  msg = memp_malloc(MEMP_TCPIP_MSG);
   if (msg == NULL) {
     pbuf_free(p);    
     return ERR_MEM;  
@@ -131,13 +131,13 @@ tcpip_input(struct pbuf *p, struct netif *inp)
   sys_mbox_post(mbox, msg);
   return ERR_OK;
 }
-/*-----------------------------------------------------------------------------------*/
+
 err_t
 tcpip_callback(void (*f)(void *ctx), void *ctx)
 {
   struct tcpip_msg *msg;
   
-  msg = memp_mallocp(MEMP_TCPIP_MSG);
+  msg = memp_malloc(MEMP_TCPIP_MSG);
   if (msg == NULL) {
     return ERR_MEM;  
   }
@@ -148,12 +148,12 @@ tcpip_callback(void (*f)(void *ctx), void *ctx)
   sys_mbox_post(mbox, msg);
   return ERR_OK;
 }
-/*-----------------------------------------------------------------------------------*/
+
 void
 tcpip_apimsg(struct api_msg *apimsg)
 {
   struct tcpip_msg *msg;
-  msg = memp_mallocp(MEMP_TCPIP_MSG);
+  msg = memp_malloc(MEMP_TCPIP_MSG);
   if (msg == NULL) {
     memp_free(MEMP_API_MSG, apimsg);
     return;
@@ -162,7 +162,7 @@ tcpip_apimsg(struct api_msg *apimsg)
   msg->msg.apimsg = apimsg;
   sys_mbox_post(mbox, msg);
 }
-/*-----------------------------------------------------------------------------------*/
+
 void
 tcpip_init(void (* initfunc)(void *), void *arg)
 {
@@ -171,7 +171,7 @@ tcpip_init(void (* initfunc)(void *), void *arg)
   mbox = sys_mbox_new();
   sys_thread_new(tcpip_thread, NULL, TCPIP_THREAD_PRIO);
 }
-/*-----------------------------------------------------------------------------------*/
+
 
 
 

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;
@@ -192,6 +194,7 @@ static void dhcp_handle_offer(struct netif *netif)
     /* remember offered address */
     ip_addr_set(&dhcp->offered_ip_addr, (struct ip_addr *)&dhcp->msg_in->yiaddr);
     LWIP_DEBUGF(DHCP_DEBUG | DBG_STATE, ("dhcp_handle_offer(): offer for 0x%08lx\n", dhcp->offered_ip_addr.addr));
+
     dhcp_select(netif);
   }
 }
@@ -403,7 +406,6 @@ static void dhcp_t2_timeout(struct netif *netif)
 }
 
 /**
- * Extract options from the server ACK message.
  *
  * @param netif the netif under DHCP control
  */
@@ -646,6 +648,9 @@ static err_t dhcp_decline(struct netif *netif)
     dhcp_option(dhcp, DHCP_OPTION_MAX_MSG_SIZE, DHCP_OPTION_MAX_MSG_SIZE_LEN);
     dhcp_option_short(dhcp, 576);
 
+    dhcp_option(dhcp, DHCP_OPTION_REQUESTED_IP, 4);
+    dhcp_option_long(dhcp, ntohl(dhcp->offered_ip_addr.addr));
+
     dhcp_option_trailer(dhcp);
     /* resize pbuf to reflect true size of options */
     pbuf_realloc(dhcp->p_out, sizeof(struct dhcp_msg) - DHCP_OPTIONS_LEN + dhcp->options_out_len);
@@ -702,15 +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"));
@@ -877,10 +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 {
@@ -1239,10 +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*/;
@@ -1414,3 +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,
@@ -30,14 +30,14 @@
  *
  */
 
-/*-----------------------------------------------------------------------------------*/
+
 /* inet.c
  *
  * Functions common to all TCP/IP modules, such as the Internet checksum and the
  * byte order functions.
  *
  */
-/*-----------------------------------------------------------------------------------*/
+
 
 #include "lwip/opt.h"
 
@@ -47,7 +47,7 @@
 #include "lwip/inet.h"
 
 
-/*-----------------------------------------------------------------------------------*/
+
 static u16_t
 lwip_chksum(void *dataptr, int len)
 {
@@ -75,12 +75,12 @@ lwip_chksum(void *dataptr, int len)
 
   return (u16_t)acc;
 }
-/*-----------------------------------------------------------------------------------*/
+
 /* inet_chksum_pseudo:
  *
  * Calculates the pseudo Internet checksum used by TCP and UDP for a pbuf chain.
  */
-/*-----------------------------------------------------------------------------------*/
+
 u16_t
 inet_chksum_pseudo(struct pbuf *p,
        struct ip_addr *src, struct ip_addr *dest,
@@ -124,13 +124,13 @@ inet_chksum_pseudo(struct pbuf *p,
   LWIP_DEBUGF(INET_DEBUG, ("inet_chksum_pseudo(): pbuf chain lwip_chksum()=%lx\n", acc));
   return ~(acc & 0xffffUL);
 }
-/*-----------------------------------------------------------------------------------*/
+
 /* inet_chksum:
  *
  * Calculates the Internet checksum over a portion of memory. Used primarely for IP
  * and ICMP.
  */
-/*-----------------------------------------------------------------------------------*/
+
 u16_t
 inet_chksum(void *dataptr, u16_t len)
 {
@@ -142,7 +142,7 @@ inet_chksum(void *dataptr, u16_t len)
   }
   return ~(acc & 0xffff);
 }
-/*-----------------------------------------------------------------------------------*/
+
 u16_t
 inet_chksum_pbuf(struct pbuf *p)
 {
@@ -169,7 +169,17 @@ inet_chksum_pbuf(struct pbuf *p)
   return ~(acc & 0xffffUL);
 }
 
-/*-----------------------------------------------------------------------------------*/
+/* Here for now until needed in other places in lwIP */
+#ifndef isascii
+#define in_range(c, lo, up)  ((u8_t)c >= lo && (u8_t)c <= up)
+#define isascii(c)           in_range(c, 0x20, 0x7f)
+#define isdigit(c)           in_range(c, '0', '9')
+#define isxdigit(c)          (isdigit(c) || in_range(c, 'a', 'f') || in_range(c, 'A', 'F'))
+#define islower(c)           in_range(c, 'a', 'z')
+#define isspace(c)           (c == ' ' || c == '\f' || c == '\n' || c == '\r' || c == '\t' || c == '\v')
+#endif		
+		
+
  /*
   * Ascii internet address interpretation routine.
   * The value returned is in network order.
@@ -290,9 +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;
@@ -301,7 +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 {
@@ -318,24 +329,24 @@ u8_t *inet_ntoa(u32_t addr)
   return str;
 }
 
-/*-----------------------------------------------------------------------------------*/
+
 #ifndef BYTE_ORDER
 #error BYTE_ORDER is not defined
 #endif
 #if BYTE_ORDER == LITTLE_ENDIAN
-/*-----------------------------------------------------------------------------------*/
+
 u16_t
 htons(u16_t n)
 {
   return ((n & 0xff) << 8) | ((n & 0xff00) >> 8);
 }
-/*-----------------------------------------------------------------------------------*/
+
 u16_t
 ntohs(u16_t n)
 {
   return htons(n);
 }
-/*-----------------------------------------------------------------------------------*/
+
 u32_t
 htonl(u32_t n)
 {
@@ -344,11 +355,11 @@ htonl(u32_t n)
     ((n & 0xff0000) >> 8) |
     ((n & 0xff000000) >> 24);
 }
-/*-----------------------------------------------------------------------------------*/
+
 u32_t
 ntohl(u32_t n)
 {
   return htonl(n);
 }
-/*-----------------------------------------------------------------------------------*/
+
 #endif /* BYTE_ORDER == LITTLE_ENDIAN */

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, 
@@ -30,14 +30,14 @@
  *
  */
 
-/*-----------------------------------------------------------------------------------*/
+
 /* inet6.c
  *
  * Functions common to all TCP/IP modules, such as the Internet checksum and the
  * byte order functions.
  *
  */
-/*-----------------------------------------------------------------------------------*/
+
 
 #include "lwip/opt.h"
 
@@ -45,7 +45,7 @@
 #include "lwip/inet.h"
 
 
-/*-----------------------------------------------------------------------------------*/
+
 /* chksum:
  *
  * Sums up all 16 bit words in a memory portion. Also includes any odd byte.
@@ -54,7 +54,7 @@
  * For now, this is not optimized. Must be optimized for the particular processor
  * arcitecture on which it is to run. Preferebly coded in assembler.
  */
-/*-----------------------------------------------------------------------------------*/
+
 static u32_t
 chksum(void *dataptr, u16_t len)
 {
@@ -74,12 +74,12 @@ chksum(void *dataptr, u16_t len)
   return acc;
 
 }
-/*-----------------------------------------------------------------------------------*/
+
 /* inet_chksum_pseudo:
  *
  * Calculates the pseudo Internet checksum used by TCP and UDP for a pbuf chain.
  */
-/*-----------------------------------------------------------------------------------*/
+
 u16_t
 inet_chksum_pseudo(struct pbuf *p,
        struct ip_addr *src, struct ip_addr *dest,
@@ -122,13 +122,13 @@ inet_chksum_pseudo(struct pbuf *p,
   }
   return ~(acc & 0xffff);
 }
-/*-----------------------------------------------------------------------------------*/
+
 /* inet_chksum:
  *
  * Calculates the Internet checksum over a portion of memory. Used primarely for IP
  * and ICMP.
  */
-/*-----------------------------------------------------------------------------------*/
+
 u16_t
 inet_chksum(void *dataptr, u16_t len)
 {
@@ -139,7 +139,7 @@ inet_chksum(void *dataptr, u16_t len)
   sum += (sum >> 16);
   return ~(sum & 0xffff);
 }
-/*-----------------------------------------------------------------------------------*/
+
 u16_t
 inet_chksum_pbuf(struct pbuf *p)
 {
@@ -165,4 +165,4 @@ inet_chksum_pbuf(struct pbuf *p)
   }
   return ~(acc & 0xffff);
 }
-/*-----------------------------------------------------------------------------------*/
+

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,
@@ -43,7 +43,7 @@
 #include "lwip/stats.h"
 
 #include "lwip/snmp.h"
-/*-----------------------------------------------------------------------------------*/
+
 void
 icmp_input(struct pbuf *p, struct netif *inp)
 {
@@ -54,9 +54,7 @@ icmp_input(struct pbuf *p, struct netif *inp)
   struct ip_addr tmpaddr;
   u16_t hlen;
 
-#ifdef ICMP_STATS
-  ++lwip_stats.icmp.recv;
-#endif /* ICMP_STATS */
+  ICMP_STATS_INC(icmp.recv);
   snmp_inc_icmpinmsgs();
 
 
@@ -65,9 +63,7 @@ icmp_input(struct pbuf *p, struct netif *inp)
   if (pbuf_header(p, -((s16_t)hlen)) || (p->tot_len < sizeof(u16_t)*2)) {
     LWIP_DEBUGF(ICMP_DEBUG, ("icmp_input: short ICMP (%u bytes) received\n", p->tot_len));
     pbuf_free(p);
-#ifdef ICMP_STATS
-    ++lwip_stats.icmp.lenerr;
-#endif /* ICMP_STATS */
+    ICMP_STATS_INC(icmp.lenerr);
     snmp_inc_icmpinerrors();
     return;
   }
@@ -76,23 +72,18 @@ 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"));
-#ifdef ICMP_STATS
-      ++lwip_stats.icmp.err;
-#endif /* ICMP_STATS */
+      ICMP_STATS_INC(icmp.err);
       pbuf_free(p);
       return;
     }
     LWIP_DEBUGF(ICMP_DEBUG, ("icmp_input: ping\n"));
-    LWIP_DEBUGF(DEMO_DEBUG, ("Pong!\n"));
     if (p->tot_len < sizeof(struct icmp_echo_hdr)) {
       LWIP_DEBUGF(ICMP_DEBUG, ("icmp_input: bad ICMP echo received\n"));
       pbuf_free(p);
-#ifdef ICMP_STATS
-      ++lwip_stats.icmp.lenerr;
-#endif /* ICMP_STATS */
+      ICMP_STATS_INC(icmp.lenerr);
       snmp_inc_icmpinerrors();
 
       return;
@@ -101,9 +92,7 @@ icmp_input(struct pbuf *p, struct netif *inp)
     if (inet_chksum_pbuf(p) != 0) {
       LWIP_DEBUGF(ICMP_DEBUG, ("icmp_input: checksum failed for received ICMP echo\n"));
       pbuf_free(p);
-#ifdef ICMP_STATS
-      ++lwip_stats.icmp.chkerr;
-#endif /* ICMP_STATS */
+      ICMP_STATS_INC(icmp.chkerr);
       snmp_inc_icmpinerrors();
       return;
     }
@@ -117,28 +106,24 @@ icmp_input(struct pbuf *p, struct netif *inp)
     } else {
       iecho->chksum += htons(ICMP_ECHO << 8);
     }
-#ifdef ICMP_STATS
-    ++lwip_stats.icmp.xmit;
-#endif /* ICMP_STATS */
+    ICMP_STATS_INC(icmp.xmit);
     /* increase number of messages attempted to send */
     snmp_inc_icmpoutmsgs();
     /* increase number of echo replies attempted to send */
     snmp_inc_icmpoutechoreps();
 
     pbuf_header(p, hlen);
-    ip_output_if (p, &(iphdr->src), IP_HDRINCL,
-     IPH_TTL(iphdr), IP_PROTO_ICMP, inp);
+    ip_output_if(p, &(iphdr->src), IP_HDRINCL,
+		 IPH_TTL(iphdr), 0, IP_PROTO_ICMP, inp);
     break;
   default:
   LWIP_DEBUGF(ICMP_DEBUG, ("icmp_input: ICMP type %d code %d not supported.\n", (int)type, (int)code));
-#ifdef ICMP_STATS
-    ++lwip_stats.icmp.proterr;
-    ++lwip_stats.icmp.drop;
-#endif /* ICMP_STATS */
+    ICMP_STATS_INC(icmp.proterr);
+    ICMP_STATS_INC(icmp.drop);
   }
   pbuf_free(p);
 }
-/*-----------------------------------------------------------------------------------*/
+
 void
 icmp_dest_unreach(struct pbuf *p, enum icmp_dur_type t)
 {
@@ -160,19 +145,17 @@ icmp_dest_unreach(struct pbuf *p, enum icmp_dur_type t)
   /* calculate checksum */
   idur->chksum = 0;
   idur->chksum = inet_chksum(idur, q->len);
-#ifdef ICMP_STATS
-  ++lwip_stats.icmp.xmit;
-#endif /* ICMP_STATS */
+  ICMP_STATS_INC(icmp.xmit);
   /* increase number of messages attempted to send */
   snmp_inc_icmpoutmsgs();
   /* increase number of destination unreachable messages attempted to send */
   snmp_inc_icmpoutdestunreachs();
 
   ip_output(q, NULL, &(iphdr->src),
-      ICMP_TTL, IP_PROTO_ICMP);
+	    ICMP_TTL, 0, IP_PROTO_ICMP);
   pbuf_free(q);
 }
-/*-----------------------------------------------------------------------------------*/
+
 #if IP_FORWARD
 void
 icmp_time_exceeded(struct pbuf *p, enum icmp_te_type t)
@@ -184,13 +167,11 @@ icmp_time_exceeded(struct pbuf *p, enum icmp_te_type t)
   q = pbuf_alloc(PBUF_IP, 8 + IP_HLEN + 8, PBUF_RAM);
 
   iphdr = p->payload;
-#if ICMP_DEBUG
   LWIP_DEBUGF(ICMP_DEBUG, ("icmp_time_exceeded from "));
   ip_addr_debug_print(ICMP_DEBUG, &(iphdr->src));
   LWIP_DEBUGF(ICMP_DEBUG, (" to "));
   ip_addr_debug_print(ICMP_DEBUG, &(iphdr->dest));
   LWIP_DEBUGF(ICMP_DEBUG, ("\n"));
-#endif /* ICMP_DEBNUG */
 
   tehdr = q->payload;
   ICMPH_TYPE_SET(tehdr, ICMP_TE);
@@ -202,19 +183,17 @@ icmp_time_exceeded(struct pbuf *p, enum icmp_te_type t)
   /* calculate checksum */
   tehdr->chksum = 0;
   tehdr->chksum = inet_chksum(tehdr, q->len);
-#ifdef ICMP_STATS
-  ++lwip_stats.icmp.xmit;
-#endif /* ICMP_STATS */
+  ICMP_STATS_INC(icmp.xmit);
   /* increase number of messages attempted to send */
   snmp_inc_icmpoutmsgs();
   /* increase number of destination unreachable messages attempted to send */
   snmp_inc_icmpouttimeexcds();
   ip_output(q, NULL, &(iphdr->src),
-      ICMP_TTL, IP_PROTO_ICMP);
+	    ICMP_TTL, 0, IP_PROTO_ICMP);
   pbuf_free(q);
 }
 
-#endif /* IP_FORWARDING > 0 */
+#endif /* IP_FORWARD */
 
 
 

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"
@@ -49,6 +46,7 @@
 #include "lwip/inet.h"
 #include "lwip/netif.h"
 #include "lwip/icmp.h"
+#include "lwip/raw.h"
 #include "lwip/udp.h"
 #include "lwip/tcp.h"
 
@@ -61,75 +59,24 @@
 #  include "lwip/dhcp.h"
 #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 > 0
-  case IP_PROTO_UDP:
-    return udp_lookup(iphdr, inp);
-#endif /* LWIP_UDP */
-#if LWIP_TCP > 0
-  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
  * IP address given to the function.
  */
-/*-----------------------------------------------------------------------------------*/
+
 struct netif *
 ip_route(struct ip_addr *dest)
 {
@@ -147,14 +94,13 @@ ip_route(struct ip_addr *dest)
   return netif_default;
 }
 #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.
  */
-/*-----------------------------------------------------------------------------------*/
+
 static void
 ip_forward(struct pbuf *p, struct ip_hdr *iphdr, struct netif *inp)
 {
@@ -170,7 +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();
@@ -199,10 +145,8 @@ ip_forward(struct pbuf *p, struct ip_hdr *iphdr, struct netif *inp)
   LWIP_DEBUGF(IP_DEBUG, ("ip_forward: forwarding packet to 0x%lx\n",
                     iphdr->dest.addr));
 
-#ifdef IP_STATS
-  ++lwip_stats.ip.fw;
-  ++lwip_stats.ip.xmit;
-#endif /* IP_STATS */
+  IP_STATS_INC(ip.fw);
+  IP_STATS_INC(ip.xmit);
     snmp_inc_ipforwdatagrams();
 
   PERF_STOP("ip_forward");
@@ -210,9 +154,8 @@ ip_forward(struct pbuf *p, struct ip_hdr *iphdr, struct netif *inp)
   netif->output(netif, p, (struct ip_addr *)&(iphdr->dest));
 }
 #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
@@ -221,30 +164,24 @@ ip_forward(struct pbuf *p, struct ip_hdr *iphdr, struct netif *inp)
  *
  * Finally, the packet is sent to the upper layer protocol input function.
  */
-/*-----------------------------------------------------------------------------------*/
+
 err_t
 ip_input(struct pbuf *p, struct netif *inp) {
   static struct ip_hdr *iphdr;
   static struct netif *netif;
   static u16_t iphdrlen;
 
-#ifdef IP_STATS
-  ++lwip_stats.ip.recv;
-#endif /* IP_STATS */
+  IP_STATS_INC(ip.recv);
   snmp_inc_ipinreceives();
 
   /* identify the IP header */
   iphdr = p->payload;
   if (IPH_V(iphdr) != 4) {
     LWIP_DEBUGF(IP_DEBUG | 1, ("IP packet dropped due to bad version number %u\n", IPH_V(iphdr)));
-#if IP_DEBUG
     ip_debug_print(p);
-#endif /* IP_DEBUG */
     pbuf_free(p);
-#ifdef IP_STATS
-    ++lwip_stats.ip.err;
-    ++lwip_stats.ip.drop;
-#endif /* IP_STATS */
+    IP_STATS_INC(ip.err);
+    IP_STATS_INC(ip.drop);
     snmp_inc_ipunknownprotos();
     return ERR_OK;
   }
@@ -259,32 +196,28 @@ ip_input(struct pbuf *p, struct netif *inp) {
       iphdrlen, p->len));
     /* free (drop) packet pbufs */
     pbuf_free(p);
-#ifdef IP_STATS
-    ++lwip_stats.ip.lenerr;
-    ++lwip_stats.ip.drop;
-#endif /* IP_STATS */
+    IP_STATS_INC(ip.lenerr);
+    IP_STATS_INC(ip.drop);
     snmp_inc_ipindiscards();
     return ERR_OK;
   }
 
   /* 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)));
-#if IP_DEBUG
     ip_debug_print(p);
-#endif /* IP_DEBUG */
     pbuf_free(p);
-#ifdef IP_STATS
-    ++lwip_stats.ip.chkerr;
-    ++lwip_stats.ip.drop;
-#endif /* IP_STATS */
+    IP_STATS_INC(ip.chkerr);
+    IP_STATS_INC(ip.drop);
     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? */
@@ -302,7 +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)) {
@@ -315,8 +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) {
@@ -335,7 +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);
     }
@@ -347,59 +280,59 @@ 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))));
-#ifdef IP_STATS
-    ++lwip_stats.ip.opterr;
-    ++lwip_stats.ip.drop;
-#endif /* IP_STATS */
+    IP_STATS_INC(ip.opterr);
+    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);
-#ifdef IP_STATS
-    ++lwip_stats.ip.opterr;
-    ++lwip_stats.ip.drop;
-#endif /* IP_STATS */
+    IP_STATS_INC(ip.opterr);
+    IP_STATS_INC(ip.drop);
     snmp_inc_ipunknownprotos();
     return ERR_OK;
   }
 #endif /* IP_OPTIONS == 0 */
 
   /* send to upper layers */
-#if IP_DEBUG
   LWIP_DEBUGF(IP_DEBUG, ("ip_input: \n"));
   ip_debug_print(p);
   LWIP_DEBUGF(IP_DEBUG, ("ip_input: p->len %d p->tot_len %d\n", p->len, p->tot_len));
-#endif /* IP_DEBUG */
+
+#if LWIP_RAW
+  /* raw input did not eat the packet? */
+  if (raw_input(p, inp) == 0) {
+#endif /* LWIP_RAW */
 
   switch (IPH_PROTO(iphdr)) {
-#if LWIP_UDP > 0
+#if LWIP_UDP
   case IP_PROTO_UDP:
+  case IP_PROTO_UDPLITE:
     snmp_inc_ipindelivers();
     udp_input(p, inp);
     break;
 #endif /* LWIP_UDP */
-#if LWIP_TCP > 0
+#if LWIP_TCP
   case IP_PROTO_TCP:
     snmp_inc_ipindelivers();
     tcp_input(p, inp);
@@ -411,8 +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);
     }
@@ -420,28 +353,26 @@ ip_input(struct pbuf *p, struct netif *inp) {
 
     LWIP_DEBUGF(IP_DEBUG | 2, ("Unsupported transport protocol %d\n", IPH_PROTO(iphdr)));
 
-#ifdef IP_STATS
-    ++lwip_stats.ip.proterr;
-    ++lwip_stats.ip.drop;
-#endif /* IP_STATS */
+    IP_STATS_INC(ip.proterr);
+    IP_STATS_INC(ip.drop);
     snmp_inc_ipunknownprotos();
-
   }
+#if LWIP_RAW
+  } /* LWIP_RAW */
+#endif
   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
  * interface is filled in as source address.
  */
-/*-----------------------------------------------------------------------------------*/
+
 err_t
-ip_output_if (struct pbuf *p, struct ip_addr *src, struct ip_addr *dest,
-             u8_t ttl,
+ip_output_if(struct pbuf *p, struct ip_addr *src, struct ip_addr *dest,
+             u8_t ttl, u8_t tos,
              u8_t proto, struct netif *netif)
 {
   static struct ip_hdr *iphdr;
@@ -453,9 +384,7 @@ ip_output_if (struct pbuf *p, struct ip_addr *src, struct ip_addr *dest,
     if (pbuf_header(p, IP_HLEN)) {
       LWIP_DEBUGF(IP_DEBUG | 2, ("ip_output: not enough room for IP header in pbuf\n"));
 
-#ifdef IP_STATS
-      ++lwip_stats.ip.err;
-#endif /* IP_STATS */
+      IP_STATS_INC(ip.err);
       snmp_inc_ipoutdiscards();
       return ERR_BUF;
     }
@@ -467,7 +396,7 @@ ip_output_if (struct pbuf *p, struct ip_addr *src, struct ip_addr *dest,
 
     ip_addr_set(&(iphdr->dest), dest);
 
-    IPH_VHLTOS_SET(iphdr, 4, IP_HLEN / 4, 0);
+    IPH_VHLTOS_SET(iphdr, 4, IP_HLEN / 4, tos);
     IPH_LEN_SET(iphdr, htons(p->tot_len));
     IPH_OFFSET_SET(iphdr, htons(IP_DF));
     IPH_ID_SET(iphdr, htons(ip_id));
@@ -480,7 +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);
@@ -492,44 +423,38 @@ ip_output_if (struct pbuf *p, struct ip_addr *src, struct ip_addr *dest,
     return ip_frag(p,netif,dest);
 #endif
 
-#ifdef IP_STATS
-  lwip_stats.ip.xmit++;
-#endif /* IP_STATS */
+  IP_STATS_INC(ip.xmit);
+
   LWIP_DEBUGF(IP_DEBUG, ("ip_output_if: %c%c%u\n", netif->name[0], netif->name[1], netif->num));
-#if IP_DEBUG
   ip_debug_print(p);
-#endif /* IP_DEBUG */
 
   LWIP_DEBUGF(IP_DEBUG, ("netif->output()"));
 
   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.
  */
-/*-----------------------------------------------------------------------------------*/
+
 err_t
 ip_output(struct pbuf *p, struct ip_addr *src, struct ip_addr *dest,
-          u8_t ttl, u8_t proto)
+          u8_t ttl, u8_t tos, u8_t proto)
 {
   struct netif *netif;
 
   if ((netif = ip_route(dest)) == NULL) {
     LWIP_DEBUGF(IP_DEBUG | 2, ("ip_output: No route to 0x%lx\n", dest->addr));
 
-#ifdef IP_STATS
-    ++lwip_stats.ip.rterr;
-#endif /* IP_STATS */
+    IP_STATS_INC(ip.rterr);
     snmp_inc_ipoutdiscards();
     return ERR_RTE;
   }
 
-  return ip_output_if (p, src, dest, ttl, proto, netif);
+  return ip_output_if(p, src, dest, ttl, tos, proto, netif);
 }
-/*-----------------------------------------------------------------------------------*/
+
 #if IP_DEBUG
 void
 ip_debug_print(struct pbuf *p)
@@ -573,7 +498,7 @@ ip_debug_print(struct pbuf *p)
   LWIP_DEBUGF(IP_DEBUG, ("+-------------------------------+\n"));
 }
 #endif /* IP_DEBUG */
-/*-----------------------------------------------------------------------------------*/
+
 
 
 

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"
@@ -109,9 +107,7 @@ ip_reass(struct pbuf *p)
   u16_t offset, len;
   u16_t i;
 
-#ifdef IP_STATS
-  ++lwip_stats.ip_frag.recv;
-#endif /* IP_STATS */
+  IPFRAG_STATS_INC(ip_frag.recv);
 
   iphdr = (struct ip_hdr *) ip_reassbuf;
   fraghdr = (struct ip_hdr *) p->payload;
@@ -135,9 +131,7 @@ ip_reass(struct pbuf *p)
       ip_addr_cmp(&iphdr->dest, &fraghdr->dest) &&
       IPH_ID(iphdr) == IPH_ID(fraghdr)) {
     LWIP_DEBUGF(IP_REASS_DEBUG, ("ip_reass: matching old packet\n"));
-#ifdef IP_STATS
-    ++lwip_stats.ip_frag.cachehit;
-#endif /* IP_STATS */
+    IPFRAG_STATS_INC(ip_frag.cachehit);
     /* Find out the offset in the reassembly buffer where we should
        copy the fragment. */
     len = ntohs(IPH_LEN(fraghdr)) - IPH_HL(fraghdr) * 4;
@@ -255,13 +249,9 @@ ip_reass(struct pbuf *p)
     q->len > ip_reasslen - i ? ip_reasslen - i : q->len);
     i += q->len;
   }
-#ifdef IP_STATS
-  ++lwip_stats.ip_frag.fw;
-#endif /* IP_STATS */
+  IPFRAG_STATS_INC(ip_frag.fw);
       } else {
-#ifdef IP_STATS
-  ++lwip_stats.ip_frag.memerr;
-#endif /* IP_STATS */
+  IPFRAG_STATS_INC(ip_frag.memerr);
       }
       LWIP_DEBUGF(IP_REASS_DEBUG, ("ip_reass: p %p\n", (void*)p));
       return p;
@@ -269,9 +259,7 @@ ip_reass(struct pbuf *p)
   }
 
 nullreturn:
-#ifdef IP_STATS
-  ++lwip_stats.ip_frag.drop;
-#endif /* IP_STATS */
+  IPFRAG_STATS_INC(ip_frag.drop);
   pbuf_free(p);
   return NULL;
 }
@@ -304,7 +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);
@@ -347,9 +334,7 @@ ip_frag(struct pbuf *p, struct netif *netif, struct ip_addr *dest)
     header = pbuf_alloc(PBUF_LINK, 0, PBUF_RAM);
     pbuf_chain(header, rambuf);
     netif->output(netif, header, dest);
-#ifdef IP_STATS
-    ++lwip_stats.ip_frag.xmit;
-#endif /* IP_STATS */
+    IPFRAG_STATS_INC(ip_frag.xmit);
     pbuf_free(header);
 
     left -= cop;

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,
@@ -42,7 +42,7 @@
 
 #include "lwip/stats.h"
 
-/*-----------------------------------------------------------------------------------*/
+
 void
 icmp_input(struct pbuf *p, struct netif *inp)
 {
@@ -113,7 +113,7 @@ icmp_input(struct pbuf *p, struct netif *inp)
 
   pbuf_free(p);
 }
-/*-----------------------------------------------------------------------------------*/
+
 void
 icmp_dest_unreach(struct pbuf *p, enum icmp_dur_type t)
 {
@@ -143,7 +143,7 @@ icmp_dest_unreach(struct pbuf *p, enum icmp_dur_type t)
       (struct ip_addr *)&(iphdr->src), ICMP_TTL, IP_PROTO_ICMP);
   pbuf_free(q);
 }
-/*-----------------------------------------------------------------------------------*/
+
 void
 icmp_time_exceeded(struct pbuf *p, enum icmp_te_type t)
 {

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,
@@ -31,13 +31,13 @@
  */
 
 
-/*-----------------------------------------------------------------------------------*/
+
 /* ip.c
  *
  * This is the code for the IP layer for IPv6.
  *
  */
-/*-----------------------------------------------------------------------------------*/
+
 
 #include "lwip/opt.h"
 
@@ -53,24 +53,24 @@
 #include "lwip/stats.h"
 
 #include "arch/perf.h"
-/*-----------------------------------------------------------------------------------*/
+
 /* ip_init:
  *
  * Initializes the IP layer.
  */
-/*-----------------------------------------------------------------------------------*/
+
 void
 ip_init(void)
 {
 }
-/*-----------------------------------------------------------------------------------*/
+
 /* 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 IP address given to the function.
  */
-/*-----------------------------------------------------------------------------------*/
+
 struct netif *
 ip_route(struct ip_addr *dest)
 {
@@ -84,14 +84,14 @@ ip_route(struct ip_addr *dest)
 
   return netif_default;
 }
-/*-----------------------------------------------------------------------------------*/
+
 /* 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.
  */
-/*-----------------------------------------------------------------------------------*/
+
 static void
 ip_forward(struct pbuf *p, struct ip_hdr *iphdr)
 {
@@ -142,7 +142,7 @@ ip_forward(struct pbuf *p, struct ip_hdr *iphdr)
 
   netif->output(netif, p, (struct ip_addr *)&(iphdr->dest));
 }
-/*-----------------------------------------------------------------------------------*/
+
 /* ip_input:
  *
  * This function is called by the network interface device driver when an IP packet is
@@ -152,7 +152,7 @@ ip_forward(struct pbuf *p, struct ip_hdr *iphdr)
  *
  * Finally, the packet is sent to the upper layer protocol input function.
  */
-/*-----------------------------------------------------------------------------------*/
+
 void
 ip_input(struct pbuf *p, struct netif *inp) {
   struct ip_hdr *iphdr;
@@ -249,14 +249,14 @@ ip_input(struct pbuf *p, struct netif *inp) {
   PERF_STOP("ip_input");
 }
 
-/*-----------------------------------------------------------------------------------*/
+
 /* 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 interface is filled in as source address.
  */
-/*-----------------------------------------------------------------------------------*/
+
 err_t
 ip_output_if (struct pbuf *p, struct ip_addr *src, struct ip_addr *dest,
        u8_t ttl,
@@ -311,13 +311,13 @@ ip_output_if (struct pbuf *p, struct ip_addr *src, struct ip_addr *dest,
   PERF_STOP("ip_output_if");
   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.
  */
-/*-----------------------------------------------------------------------------------*/
+
 err_t
 ip_output(struct pbuf *p, struct ip_addr *src, struct ip_addr *dest,
     u8_t ttl, u8_t proto)
@@ -333,7 +333,7 @@ ip_output(struct pbuf *p, struct ip_addr *src, struct ip_addr *dest,
 
   return ip_output_if (p, src, dest, ttl, proto, netif);
 }
-/*-----------------------------------------------------------------------------------*/
+
 #if IP_DEBUG
 void
 ip_debug_print(struct pbuf *p)
@@ -383,4 +383,4 @@ ip_debug_print(struct pbuf *p)
   LWIP_DEBUGF(IP_DEBUG, ("+-------------------------------+\n"));
 }
 #endif /* IP_DEBUG */
-/*-----------------------------------------------------------------------------------*/
+

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, 
@@ -33,7 +33,7 @@
 #include "lwip/ip_addr.h"
 #include "lwip/inet.h"
 
-/*-----------------------------------------------------------------------------------*/
+
 int
 ip_addr_maskcmp(struct ip_addr *addr1, struct ip_addr *addr2,
                 struct ip_addr *mask)
@@ -44,7 +44,7 @@ ip_addr_maskcmp(struct ip_addr *addr1, struct ip_addr *addr2,
          (addr1->addr[3] & mask->addr[3]) == (addr2->addr[3] & mask->addr[3]));
         
 }
-/*-----------------------------------------------------------------------------------*/
+
 int
 ip_addr_cmp(struct ip_addr *addr1, struct ip_addr *addr2)
 {
@@ -53,7 +53,7 @@ ip_addr_cmp(struct ip_addr *addr1, struct ip_addr *addr2)
          addr1->addr[2] == addr2->addr[2] &&
          addr1->addr[3] == addr2->addr[3]);
 }
-/*-----------------------------------------------------------------------------------*/
+
 void
 ip_addr_set(struct ip_addr *dest, struct ip_addr *src)
 {
@@ -63,7 +63,7 @@ ip_addr_set(struct ip_addr *dest, struct ip_addr *src)
   dest->addr[2] = src->addr[2];
   dest->addr[3] = src->addr[3];*/
 }
-/*-----------------------------------------------------------------------------------*/
+
 int
 ip_addr_isany(struct ip_addr *addr)
 {
@@ -71,7 +71,7 @@ ip_addr_isany(struct ip_addr *addr)
   return((addr->addr[0] | addr->addr[1] | addr->addr[2] | addr->addr[3]) == 0);
 }
 
-/*-----------------------------------------------------------------------------------*/
+
 /*#if IP_DEBUG*/
 void
 ip_addr_debug_print(struct ip_addr *addr)
@@ -87,4 +87,4 @@ ip_addr_debug_print(struct ip_addr *addr)
          ntohl(addr->addr[3]) & 0xffff);
 }
 /*#endif*/ /* IP_DEBUG */
-/*-----------------------------------------------------------------------------------*/
+

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 */
@@ -127,7 +129,7 @@ mem_init(void)
 
   lfree = (struct mem *)ram;
 
-#ifdef MEM_STATS
+#if MEM_STATS
   lwip_stats.mem.avail = MEM_SIZE;
 #endif /* MEM_STATS */
 }
@@ -148,7 +150,7 @@ mem_free(void *rmem)
   
   if ((u8_t *)rmem < (u8_t *)ram || (u8_t *)rmem >= (u8_t *)ram_end) {
     LWIP_DEBUGF(MEM_DEBUG | 3, ("mem_free: illegal memory\n"));
-#ifdef MEM_STATS
+#if MEM_STATS
     ++lwip_stats.mem.err;
 #endif /* MEM_STATS */
     sys_sem_signal(mem_sem);
@@ -164,7 +166,7 @@ mem_free(void *rmem)
     lfree = mem;
   }
   
-#ifdef MEM_STATS
+#if MEM_STATS
   lwip_stats.mem.used -= mem->next - ((u8_t *)mem - ram);
   
 #endif /* MEM_STATS */
@@ -183,7 +185,7 @@ mem_reallocm(void *rmem, mem_size_t newsize)
   mem_free(rmem);
   return nmem;
 }
-/*-----------------------------------------------------------------------------------*/
+
 void *
 mem_realloc(void *rmem, mem_size_t newsize)
 {
@@ -215,7 +217,7 @@ mem_realloc(void *rmem, mem_size_t newsize)
   ptr = (u8_t *)mem - ram;
 
   size = mem->next - ptr - SIZEOF_STRUCT_MEM;
-#ifdef MEM_STATS
+#if MEM_STATS
   lwip_stats.mem.used -= (size - newsize);
 #endif /* MEM_STATS */
   
@@ -273,7 +275,7 @@ mem_malloc(mem_size_t size)
       
       mem2->used = 0;      
       mem->used = 1;
-#ifdef MEM_STATS
+#if MEM_STATS
       lwip_stats.mem.used += (size + SIZEOF_STRUCT_MEM);
       /*      if (lwip_stats.mem.max < lwip_stats.mem.used) {
         lwip_stats.mem.max = lwip_stats.mem.used;
@@ -292,14 +294,14 @@ 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;
     }    
   }
   LWIP_DEBUGF(MEM_DEBUG | 2, ("mem_malloc: could not allocate %d bytes\n", (int)size));
-#ifdef MEM_STATS
+#if MEM_STATS
   ++lwip_stats.mem.err;
 #endif /* MEM_STATS */  
   sys_sem_signal(mem_sem);

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, 
@@ -36,6 +36,7 @@
 
 #include "lwip/pbuf.h"
 #include "lwip/udp.h"
+#include "lwip/raw.h"
 #include "lwip/tcp.h"
 #include "lwip/api.h"
 #include "lwip/api_msg.h"
@@ -54,6 +55,7 @@ static struct memp *memp_tab[MEMP_MAX];
 
 static const u16_t memp_sizes[MEMP_MAX] = {
   sizeof(struct pbuf),
+  sizeof(struct raw_pcb),
   sizeof(struct udp_pcb),
   sizeof(struct tcp_pcb),
   sizeof(struct tcp_pcb_listen),
@@ -67,6 +69,7 @@ static const u16_t memp_sizes[MEMP_MAX] = {
 
 static const u16_t memp_num[MEMP_MAX] = {
   MEMP_NUM_PBUF,
+  MEMP_NUM_RAW_PCB,
   MEMP_NUM_UDP_PCB,
   MEMP_NUM_TCP_PCB,
   MEMP_NUM_TCP_PCB_LISTEN,
@@ -81,6 +84,9 @@ static const u16_t memp_num[MEMP_MAX] = {
 static u8_t memp_memory[(MEMP_NUM_PBUF *
        MEM_ALIGN_SIZE(sizeof(struct pbuf) +
           sizeof(struct memp)) +
+      MEMP_NUM_RAW_PCB *
+       MEM_ALIGN_SIZE(sizeof(struct raw_pcb) +
+          sizeof(struct memp)) +
       MEMP_NUM_UDP_PCB *
        MEM_ALIGN_SIZE(sizeof(struct udp_pcb) +
           sizeof(struct memp)) +
@@ -109,12 +115,12 @@ static u8_t memp_memory[(MEMP_NUM_PBUF *
        MEM_ALIGN_SIZE(sizeof(struct sys_timeout) +
           sizeof(struct memp)))];
 
-/*-----------------------------------------------------------------------------------*/
+
 #if !SYS_LIGHTWEIGHT_PROT
 static sys_sem_t mutex;
 #endif
-/*-----------------------------------------------------------------------------------*/
-#ifdef LWIP_DEBUG
+
+#if MEMP_SANITY_CHECK
 static int
 memp_sanity(void)
 {
@@ -134,8 +140,8 @@ memp_sanity(void)
   }
   return 1;
 }
-#endif /* LWIP_DEBUG */
-/*-----------------------------------------------------------------------------------*/
+#endif /* MEMP_SANITY_CHECK*/
+
 void
 memp_init(void)
 {
@@ -143,7 +149,7 @@ memp_init(void)
   u16_t i, j;
   u16_t size;
       
-#ifdef MEMP_STATS
+#if MEMP_STATS
   for(i = 0; i < MEMP_MAX; ++i) {
     lwip_stats.memp[i].used = lwip_stats.memp[i].max =
       lwip_stats.memp[i].err = 0;
@@ -176,102 +182,93 @@ memp_init(void)
 
   
 }
-/*-----------------------------------------------------------------------------------*/
+
 void *
 memp_malloc(memp_t type)
 {
   struct memp *memp;
   void *mem;
+#if SYS_LIGHTWEIGHT_PROT
+  SYS_ARCH_DECL_PROTECT(old_level);
+#endif
  
   LWIP_ASSERT("memp_malloc: type < MEMP_MAX", type < MEMP_MAX);
 
+#if SYS_LIGHTWEIGHT_PROT
+  SYS_ARCH_PROTECT(old_level);
+#else /* SYS_LIGHTWEIGHT_PROT */  
+  sys_sem_wait(mutex);
+#endif /* SYS_LIGHTWEIGHT_PROT */  
+
   memp = memp_tab[type];
   
   if (memp != NULL) {    
     memp_tab[type] = memp->next;    
     memp->next = NULL;
-#ifdef MEMP_STATS
+#if MEMP_STATS
     ++lwip_stats.memp[type].used;
     if (lwip_stats.memp[type].used > lwip_stats.memp[type].max) {
       lwip_stats.memp[type].max = lwip_stats.memp[type].used;
     }
 #endif /* MEMP_STATS */
+#if SYS_LIGHTWEIGHT_PROT
+    SYS_ARCH_UNPROTECT(old_level);
+#else /* SYS_LIGHTWEIGHT_PROT */
+    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));
-    /* initialize memp memory with zeroes */
-    memset(mem, 0, memp_sizes[type]);  
     return mem;
   } else {
     LWIP_DEBUGF(MEMP_DEBUG | 2, ("memp_malloc: out of memory in pool %d\n", type));
-#ifdef MEMP_STATS
+#if MEMP_STATS
     ++lwip_stats.memp[type].err;
 #endif /* MEMP_STATS */
-    return NULL;
-  }
-}
-/*-----------------------------------------------------------------------------------*/
-void *
-memp_mallocp(memp_t type)
-{
-  void *mem;
-#if SYS_LIGHTWEIGHT_PROT
-  SYS_ARCH_DECL_PROTECT(old_level);
-  SYS_ARCH_PROTECT(old_level);
-#else /* SYS_LIGHTWEIGHT_PROT */  
-  sys_sem_wait(mutex);
-#endif /* SYS_LIGHTWEIGHT_PROT */  
-
-  mem = memp_malloc(type);
-
 #if SYS_LIGHTWEIGHT_PROT
   SYS_ARCH_UNPROTECT(old_level);
 #else /* SYS_LIGHTWEIGHT_PROT */
   sys_sem_signal(mutex);
 #endif /* SYS_LIGHTWEIGHT_PROT */  
-  return mem;
+    return NULL;
+  }
 }
-/*-----------------------------------------------------------------------------------*/
+
 void
 memp_free(memp_t type, void *mem)
 {
   struct memp *memp;
+#if SYS_LIGHTWEIGHT_PROT
+  SYS_ARCH_DECL_PROTECT(old_level);
+#endif /* SYS_LIGHTWEIGHT_PROT */  
 
   if (mem == NULL) {
     return;
   }
   memp = (struct memp *)((u8_t *)mem - sizeof(struct memp));
 
-#ifdef MEMP_STATS
+#if SYS_LIGHTWEIGHT_PROT
+    SYS_ARCH_PROTECT(old_level);
+#else /* SYS_LIGHTWEIGHT_PROT */  
+  sys_sem_wait(mutex);
+#endif /* SYS_LIGHTWEIGHT_PROT */  
+
+#if MEMP_STATS
   lwip_stats.memp[type].used--; 
 #endif /* MEMP_STATS */
   
   memp->next = memp_tab[type]; 
   memp_tab[type] = memp;
 
+#if MEMP_SANITY_CHECK
   LWIP_ASSERT("memp sanity", memp_sanity());
-
-  return;
-}
-/*-----------------------------------------------------------------------------------*/
-void 
-memp_freep(memp_t type, void *mem)
-{
-#if SYS_LIGHTWEIGHT_PROT
-    SYS_ARCH_DECL_PROTECT(old_level);
-    SYS_ARCH_PROTECT(old_level);
-#else /* SYS_LIGHTWEIGHT_PROT */  
-  sys_sem_wait(mutex);
-#endif /* SYS_LIGHTWEIGHT_PROT */  
-
-  memp_free(type, mem);
+#endif  
 
 #if SYS_LIGHTWEIGHT_PROT
   SYS_ARCH_UNPROTECT(old_level);
 #else /* SYS_LIGHTWEIGHT_PROT */
   sys_sem_signal(mutex);
 #endif /* SYS_LIGHTWEIGHT_PROT */  
-
 }
-/*-----------------------------------------------------------------------------------*/
+

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,
@@ -39,44 +39,37 @@
 #include "lwip/opt.h"
 
 #include "lwip/def.h"
-#include "lwip/mem.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.
  */
 struct netif *
-netif_add(struct ip_addr *ipaddr, struct ip_addr *netmask,
+netif_add(struct netif *netif, struct ip_addr *ipaddr, struct ip_addr *netmask,
   struct ip_addr *gw,
   void *state,
   err_t (* init)(struct netif *netif),
   err_t (* input)(struct pbuf *p, struct netif *netif))
 {
-  struct netif *netif;
   static int netifnum = 0;
 
-  /* allocate netif structure */
-  netif = mem_malloc(sizeof(struct netif));
-
-  if (netif == NULL) {
-    LWIP_DEBUGF(NETIF_DEBUG, ("netif_add(): out of memory for netif\n"));
-    return NULL;
-  }
+  
 #if LWIP_DHCP
   /* netif not under DHCP control by default */
   netif->dhcp = NULL;
@@ -90,14 +83,12 @@ netif_add(struct ip_addr *ipaddr, struct ip_addr *netmask,
 
   /* call user specified initialization function for netif */
   if (init(netif) != ERR_OK) {
-    mem_free(netif);
     return NULL;
   }
 
   /* add this netif to the list */
   netif->next = netif_list;
   netif_list = netif;
-#if NETIF_DEBUG
   LWIP_DEBUGF(NETIF_DEBUG, ("netif: added interface %c%c IP addr ",
     netif->name[0], netif->name[1]));
   ip_addr_debug_print(NETIF_DEBUG, ipaddr);
@@ -106,7 +97,6 @@ netif_add(struct ip_addr *ipaddr, struct ip_addr *netmask,
   LWIP_DEBUGF(NETIF_DEBUG, (" gw "));
   ip_addr_debug_print(NETIF_DEBUG, gw);
   LWIP_DEBUGF(NETIF_DEBUG, ("\n"));
-#endif /* NETIF_DEBUG */
   return netif;
 }
 
@@ -144,7 +134,6 @@ void netif_remove(struct netif * netif)
     /* reset default netif */
     netif_default = NULL;
   LWIP_DEBUGF( NETIF_DEBUG, ("netif_remove: removed netif\n") );
-  mem_free( netif );
 }
 
 struct netif *
@@ -170,7 +159,7 @@ netif_find(char *name)
   LWIP_DEBUGF(NETIF_DEBUG, ("netif_find: didn't find %c%c\n", name[0], name[1]));
   return NULL;
 }
-/*-----------------------------------------------------------------------------------*/
+
 void
 netif_set_ipaddr(struct netif *netif, struct ip_addr *ipaddr)
 {
@@ -183,7 +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) {
@@ -191,25 +180,32 @@ netif_set_ipaddr(struct netif *netif, struct ip_addr *ipaddr)
       if (ip_addr_cmp(&(pcb->local_ip), &(netif->ip_addr))) {
         /* this connection must be aborted */
         struct tcp_pcb *next = pcb->next;
-        LWIP_DEBUGF(NETIF_DEBUG | 1, ("netif_set_ipaddr: aborting pcb %p\n", (void *)pcb));
+        LWIP_DEBUGF(NETIF_DEBUG | 1, ("netif_set_ipaddr: aborting TCP pcb %p\n", (void *)pcb));
         tcp_abort(pcb);
         pcb = next;
       } else {
         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
-         * is set to listen to the new one instead
-         * TODO: how do we know it is _listening_? */
+         * is set to listen to the new one instead */
         ip_addr_set(&(lpcb->local_ip), 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),
@@ -217,7 +213,7 @@ netif_set_ipaddr(struct netif *netif, struct ip_addr *ipaddr)
     (unsigned int)(ntohl(netif->ip_addr.addr) >> 8 & 0xff),
     (unsigned int)(ntohl(netif->ip_addr.addr) & 0xff)));
 }
-/*-----------------------------------------------------------------------------------*/
+
 void
 netif_set_gw(struct netif *netif, struct ip_addr *gw)
 {
@@ -229,7 +225,7 @@ netif_set_gw(struct netif *netif, struct ip_addr *gw)
            (unsigned int)(ntohl(netif->gw.addr) >> 8 & 0xff),
            (unsigned int)(ntohl(netif->gw.addr) & 0xff)));
 }
-/*-----------------------------------------------------------------------------------*/
+
 void
 netif_set_netmask(struct netif *netif, struct ip_addr *netmask)
 {
@@ -241,7 +237,7 @@ netif_set_netmask(struct netif *netif, struct ip_addr *netmask)
            (unsigned int)(ntohl(netif->netmask.addr) >> 8 & 0xff),
            (unsigned int)(ntohl(netif->netmask.addr) & 0xff)));
 }
-/*-----------------------------------------------------------------------------------*/
+
 void
 netif_set_default(struct netif *netif)
 {
@@ -249,10 +245,10 @@ netif_set_default(struct netif *netif)
   LWIP_DEBUGF(NETIF_DEBUG, ("netif: setting default interface %c%c\n",
            netif ? netif->name[0] : '\'', netif ? netif->name[1] : '\''));
 }
-/*-----------------------------------------------------------------------------------*/
+
 void
 netif_init(void)
 {
   netif_list = netif_default = NULL;
 }
-/*-----------------------------------------------------------------------------------*/
+

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,
@@ -100,9 +103,9 @@ 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);
 
-#ifdef PBUF_STATS
+#if PBUF_STATS
   lwip_stats.pbuf.avail = PBUF_POOL_SIZE;
 #endif /* PBUF_STATS */
 
@@ -113,6 +116,7 @@ pbuf_init(void)
     p->next = (struct pbuf *)((u8_t *)p + PBUF_POOL_BUFSIZE + sizeof(struct pbuf));
     p->len = p->tot_len = PBUF_POOL_BUFSIZE;
     p->payload = MEM_ALIGN((void *)((u8_t *)p + sizeof(struct pbuf)));
+    p->flags = PBUF_FLAG_POOL;
     q = p;
     p = p->next;
   }
@@ -143,7 +147,7 @@ pbuf_pool_alloc(void)
   /* Next, check the actual pbuf pool, but if the pool is locked, we
      pretend to be out of buffers and return NULL. */
   if (pbuf_pool_free_lock) {
-#ifdef PBUF_STATS
+#if PBUF_STATS
     ++lwip_stats.pbuf.alloc_locked;
 #endif /* PBUF_STATS */
     return NULL;
@@ -156,7 +160,7 @@ pbuf_pool_alloc(void)
       pbuf_pool = p->next;
     }
 #if !SYS_LIGHTWEIGHT_PROT
-#ifdef PBUF_STATS
+#if PBUF_STATS
   } else {
     ++lwip_stats.pbuf.alloc_locked;
 #endif /* PBUF_STATS */
@@ -164,7 +168,7 @@ pbuf_pool_alloc(void)
   pbuf_pool_alloc_lock = 0;
 #endif /* SYS_LIGHTWEIGHT_PROT */
 
-#ifdef PBUF_STATS
+#if PBUF_STATS
   if (p != NULL) {
     ++lwip_stats.pbuf.used;
     if (lwip_stats.pbuf.used > lwip_stats.pbuf.max) {
@@ -243,7 +247,7 @@ pbuf_alloc(pbuf_layer l, u16_t length, pbuf_flag flag)
     p = pbuf_pool_alloc();
     LWIP_DEBUGF(PBUF_DEBUG | DBG_TRACE | 3, ("pbuf_alloc: allocated pbuf %p\n", (void *)p));
     if (p == NULL) {
-#ifdef PBUF_STATS
+#if PBUF_STATS
       ++lwip_stats.pbuf.err;
 #endif /* PBUF_STATS */
       return NULL;
@@ -253,13 +257,11 @@ 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 */
     p->len = length > PBUF_POOL_BUFSIZE - offset? PBUF_POOL_BUFSIZE - offset: length;
-    /* set pbuf type */
-    p->flags = PBUF_FLAG_POOL;
     /* set reference count (needed here in case we fail) */
     p->ref = 1;
 
@@ -274,7 +276,7 @@ pbuf_alloc(pbuf_layer l, u16_t length, pbuf_flag flag)
       q = pbuf_pool_alloc();
       if (q == NULL) {
        LWIP_DEBUGF(PBUF_DEBUG | 2, ("pbuf_alloc: Out of pbufs in pool.\n"));
-#ifdef PBUF_STATS
+#if PBUF_STATS
         ++lwip_stats.pbuf.err;
 #endif /* PBUF_STATS */
         /* free chain so far allocated */
@@ -289,10 +291,9 @@ pbuf_alloc(pbuf_layer l, u16_t length, pbuf_flag flag)
       q->tot_len = rem_len;
       /* this pbuf length is pool size, unless smaller sized tail */
       q->len = rem_len > PBUF_POOL_BUFSIZE? PBUF_POOL_BUFSIZE: rem_len;
-      q->flags = PBUF_FLAG_POOL;
       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;
@@ -300,7 +301,7 @@ pbuf_alloc(pbuf_layer l, u16_t length, pbuf_flag flag)
       r = q;
     }
     /* end of chain */
-    //r->next = NULL;
+    /*r->next = NULL;*/
 
     break;
   case PBUF_RAM:
@@ -316,14 +317,14 @@ 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:
   /* pbuf references existing (externally allocated) RAM payload? */
   case PBUF_REF:
     /* only allocate memory for the pbuf structure */
-    p = memp_mallocp(MEMP_PBUF);
+    p = memp_malloc(MEMP_PBUF);
     if (p == NULL) {
       LWIP_DEBUGF(PBUF_DEBUG | DBG_TRACE | 2, ("pbuf_alloc: Could not allocate MEMP_PBUF for PBUF_%s.\n", flag == PBUF_ROM?"ROM":"REF"));
       return NULL;
@@ -345,7 +346,7 @@ pbuf_alloc(pbuf_layer l, u16_t length, pbuf_flag flag)
 }
 
 
-#ifdef PBUF_STATS
+#if PBUF_STATS
 #define DEC_PBUF_STATS do { --lwip_stats.pbuf.used; } while (0)
 #else /* PBUF_STATS */
 #define DEC_PBUF_STATS
@@ -412,7 +413,7 @@ pbuf_realloc(struct pbuf *p, u16_t new_len)
   /* first, step over any pbufs that should remain in the chain */
   rem_len = new_len;
   q = p;
-  /* this pbuf should be kept? */
+  /* should this pbuf be kept? */
   while (rem_len > q->len) {
     /* decrease remaining length by pbuf length */
     rem_len -= q->len;
@@ -452,32 +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 */
@@ -485,50 +489,56 @@ 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;
 }
 
 /**
- * Free a pbuf (chain) from usage, de-allocate non-used head of 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, until
- * a non-zero reference count is encountered, or the end of the chain is
- * reached.
+ * For a pbuf chain, this is repeated for each pbuf in the chain,
+ * up to the first pbuf which has a non-zero reference count after
+ * decrementing. (This might de-allocate the whole chain.)
  *
- * @param pbuf pbuf (chain) to be freed from one user.
+ * @param pbuf The pbuf (chain) to be dereferenced.
  *
- * @return the number of unreferenced pbufs that were de-allocated
+ * @return the number of pbufs that were de-allocated
  * from the head of the chain.
  *
- * @note May 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).
  *
  * @internal examples:
  *
+ * Assuming existing chains a->b->c with the following reference
+ * counts, calling pbuf_free(a) results in:
+ * 
  * 1->2->3 becomes ...1->3
  * 3->3->3 becomes 2->3->3
- * 1->1->2 becomes ....->1
+ * 1->1->2 becomes ......1
  * 2->1->1 becomes 1->1->1
  * 1->1->1 becomes .......
  *
@@ -540,6 +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;
@@ -576,7 +588,7 @@ pbuf_free(struct pbuf *p)
         PBUF_POOL_FREE(p);
       /* a ROM or RAM referencing pbuf */
       } else if (p->flags == PBUF_FLAG_ROM || p->flags == PBUF_FLAG_REF) {
-        memp_freep(MEMP_PBUF, p);
+        memp_free(MEMP_PBUF, p);
       /* p->flags == PBUF_FLAG_RAM */
       } else {
         mem_free(p);
@@ -619,7 +631,6 @@ pbuf_clen(struct pbuf *p)
 }
 
 /**
- *
  * Increment the reference count of the pbuf.
  *
  * @param p pbuf to increase reference counter of
@@ -638,12 +649,47 @@ pbuf_ref(struct pbuf *p)
 }
 
 /**
- *
- * Chain two pbufs (or pbuf chains) together. They must belong to the same packet.
- *
+ * Concatenate two pbufs (each may be a pbuf chain) and take over
+ * the caller's reference of the tail pbuf.
+ * 
+ * @note The caller MAY NOT reference the tail pbuf afterwards.
+ * Use pbuf_chain() for that purpose.
+ * 
+ * @see pbuf_chain()
+ */
+
+void
+pbuf_cat(struct pbuf *h, struct pbuf *t)
+{
+  struct pbuf *p;
+
+  LWIP_ASSERT("h != NULL", h != NULL);
+  LWIP_ASSERT("t != NULL", t != NULL);
+  if ((h == NULL) || (t == NULL)) return;
+
+  /* proceed to last pbuf of chain */
+  for (p = h; p->next != NULL; p = p->next) {
+    /* add total length of second chain to all totals of first chain */
+    p->tot_len += t->tot_len;
+  }
+  /* { p is last pbuf of first h chain, p->next == NULL } */
+  LWIP_ASSERT("p->tot_len == p->len (of last pbuf in chain)", p->tot_len == p->len);
+  /* add total length of second chain to last pbuf total of first chain */
+  p->tot_len += t->tot_len;
+  /* chain last pbuf of head (p) with first of tail (t) */
+  p->next = t;
+}
+
+/**
+ * Chain two pbufs (or pbuf chains) together.
+ * 
+ * The caller MUST call pbuf_free(t) once it has stopped
+ * using it. Use pbuf_cat() instead if you no longer use t.
+ * 
  * @param h head pbuf (chain)
  * @param t tail pbuf (chain)
- * @note May not be called on a packet queue.
+ * @note The pbufs MUST belong to the same packet.
+ * @note MAY NOT be called on a packet queue.
  *
  * The ->tot_len fields of all pbufs of the head chain are adjusted.
  * The ->next field of the last pbuf of the head chain is adjusted.
@@ -653,28 +699,10 @@ pbuf_ref(struct pbuf *p)
 void
 pbuf_chain(struct pbuf *h, struct pbuf *t)
 {
-  struct pbuf *p;
-
-  LWIP_ASSERT("h != NULL", h != NULL);
-  LWIP_ASSERT("t != NULL", t != NULL);
-
-  if (t == NULL)
-    return;
-
-  /* proceed to last pbuf of chain */
-  for (p = h; p->next != NULL; p = p->next) {
-    /* add total length of second chain to all totals of first chain */
-    p->tot_len += t->tot_len;
-  }
-  /* p is last pbuf of first h chain */
-  LWIP_ASSERT("p->tot_len == p->len (of last pbuf in chain)", p->tot_len == p->len);
-  /* add total length of second chain to last pbuf total of first chain */
-  p->tot_len += t->tot_len;
-  /* chain last pbuf of h chain (p) with first of tail (t) */
-  p->next = t;
-  /* t is now referenced to one more time */
+  pbuf_cat(h, t);
+  /* t is now referenced by h */
   pbuf_ref(t);
-  LWIP_DEBUGF(PBUF_DEBUG | DBG_FRESH | 2, ("pbuf_chain: %p references %p\n", (void *)p, (void *)t));
+  LWIP_DEBUGF(PBUF_DEBUG | DBG_FRESH | 2, ("pbuf_chain: %p references %p\n", (void *)h, (void *)t));
 }
 
 /* For packet queueing. Note that queued packets must be dequeued first
@@ -690,11 +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) {
@@ -702,25 +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;
   }
-  /* chain last pbuf of h chain (p) with first of tail (t) */
+  /* { 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;
-  /* t 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).
  *
@@ -733,17 +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
@@ -764,7 +811,7 @@ pbuf_dequeue(struct pbuf *p)
  * by pbuf_take()!
  *
  * @note Any replaced pbufs will be freed through pbuf_free().
- * This may allocate them if they become no longer referenced.
+ * This may deallocate them if they become no longer referenced.
  *
  * @param p Head of pbuf chain to process
  *
@@ -889,27 +936,3 @@ pbuf_dechain(struct pbuf *p)
   LWIP_ASSERT("p->tot_len == p->len", p->tot_len == p->len);
   return (tail_gone > 0? NULL: q);
 }
-
-/* TODO: This function is unused in the lwIP stack and will be deprecated. This is due
- * to the new way chains are built. */
-#if 0
-/**
- *
- * Increment the reference count of all pbufs in a chain.
- *
- * @param p first pbuf of chain
- *
- */
-void
-pbuf_ref_chain(struct pbuf *p)
-{
-  SYS_ARCH_DECL_PROTECT(old_level);
-  SYS_ARCH_PROTECT(old_level);
-
-  while (p != NULL) {
-    ++p->ref;
-    p = p->next;
-  }
-  SYS_ARCH_UNPROTECT(old_level);
-}
-#endif

src/core/raw.c

@@ -0,0 +1,324 @@
+/**
+ * @file
+ * 
+ * Implementation of raw protocol PCBs for low-level handling of
+ * different types of protocols besides (or overriding) those
+ * already available in lwIP.
+ *
+ */
+/*
+ * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without modification,
+ * are permitted provided that the following conditions are met:
+ *
+ * 1. Redistributions of source code must retain the above copyright notice,
+ *    this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright notice,
+ *    this list of conditions and the following disclaimer in the documentation
+ *    and/or other materials provided with the distribution.
+ * 3. The name of the author may not be used to endorse or promote products
+ *    derived from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR IMPLIED
+ * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
+ * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT
+ * SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
+ * EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT
+ * OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING
+ * IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY
+ * OF SUCH DAMAGE.
+ *
+ * This file is part of the lwIP TCP/IP stack.
+ *
+ * Author: Adam Dunkels <adam@sics.se>
+ *
+ */
+
+#include "lwip/opt.h"
+
+#include "lwip/def.h"
+#include "lwip/memp.h"
+#include "lwip/inet.h"
+#include "lwip/ip_addr.h"
+#include "lwip/netif.h"
+#include "lwip/raw.h"
+
+#include "lwip/stats.h"
+
+#include "arch/perf.h"
+#include "lwip/snmp.h"
+
+#if LWIP_RAW
+
+/** The list of RAW PCBs */
+static struct raw_pcb *raw_pcbs = NULL;
+
+void
+raw_init(void)
+{
+  raw_pcbs = NULL;
+}
+
+/**
+ * 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 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 - 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).
+ *
+ */
+u8_t
+raw_input(struct pbuf *p, struct netif *inp)
+{
+  struct raw_pcb *pcb;
+  struct ip_hdr *iphdr;
+  int proto;
+  u8_t eaten = 0;
+
+  iphdr = p->payload;
+  proto = IPH_PROTO(iphdr);
+
+  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) {
+      /* 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;
+        }
+      }
+      /* no receive callback function was set for this raw PCB */
+      /* drop the packet */
+    }
+    pcb = pcb->next;
+  }
+  return eaten;
+}
+
+/**
+ * Bind a RAW PCB.
+ *
+ * @param pcb RAW PCB to be bound with a local address ipaddr.
+ * @param ipaddr local IP address to bind with. Use IP_ADDR_ANY to
+ * bind to all local interfaces.
+ *
+ * @return lwIP error code.
+ * - ERR_OK. Successful. No error occured.
+ * - ERR_USE. The specified IP address is already bound to by
+ * another RAW PCB.
+ *
+ * @see raw_disconnect()
+ */
+err_t
+raw_bind(struct raw_pcb *pcb, struct ip_addr *ipaddr)
+{
+  ip_addr_set(&pcb->local_ip, ipaddr);
+  return ERR_OK;
+}
+
+/**
+ * Connect an RAW PCB. This function is required by upper layers
+ * of lwip. Using the raw api you could use raw_sendto() instead
+ *
+ * This will associate the RAW PCB with the remote address.
+ *
+ * @param pcb RAW PCB to be connected with remote address ipaddr and port.
+ * @param ipaddr remote IP address to connect with.
+ *
+ * @return lwIP error code
+ *
+ * @see raw_disconnect() and raw_sendto()
+ */
+err_t
+raw_connect(struct raw_pcb *pcb, struct ip_addr *ipaddr)
+{
+  ip_addr_set(&pcb->remote_ip, ipaddr);
+  return ERR_OK;
+}
+
+
+/**
+ * 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,
+         u8_t (* recv)(void *arg, struct raw_pcb *upcb, struct pbuf *p,
+                      struct ip_addr *addr),
+         void *recv_arg)
+{
+  /* remember recv() callback and user data */
+  pcb->recv = recv;
+  pcb->recv_arg = recv_arg;
+}
+
+/**
+ * Send the raw IP packet to the given address. Note that actually you cannot
+ * 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 IP packet
+ *
+ */
+err_t
+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_sendto\n"));
+  
+  /* not enough space to add an IP header to first pbuf in given p chain? */
+  if (pbuf_header(p, IP_HLEN)) {
+    /* allocate header in new pbuf */
+    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_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_sendto: added header pbuf %p before given pbuf %p\n", (void *)q, (void *)p));
+  }  else {
+    /* first pbuf q equals given pbuf */
+    q = p;
+    pbuf_header(q, -IP_HLEN);
+  }
+  
+  if ((netif = ip_route(ipaddr)) == NULL) {
+    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);
+    }
+    return ERR_RTE;
+  }
+
+  if (ip_addr_isany(&pcb->local_ip)) {
+    /* use outgoing network interface IP address as source address */
+    src_ip = &(netif->ip_addr);
+  } else {
+    /* use RAW PCB local IP address as source address */
+    src_ip = &(pcb->local_ip);
+  }
+
+  err = ip_output_if (q, src_ip, ipaddr, pcb->ttl, pcb->tos, pcb->protocol, netif);
+
+  /* did we chain a header earlier? */
+  if (q != p) {
+    /* free the header */
+    pbuf_free(q);
+  }
+  return err;
+}
+
+/**
+ * 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 IP packet
+ *
+ */
+err_t
+raw_send(struct raw_pcb *pcb, struct pbuf *p)
+{
+  return raw_sendto(pcb, p, &pcb->remote_ip);
+}
+
+/**
+ * Remove an RAW PCB.
+ *
+ * @param pcb RAW PCB to be removed. The PCB is removed from the list of
+ * RAW PCB's and the data structure is freed from memory.
+ *
+ * @see raw_new()
+ */
+void
+raw_remove(struct raw_pcb *pcb)
+{
+  struct raw_pcb *pcb2;
+  /* pcb to be removed is first in list? */
+  if (raw_pcbs == pcb) {
+    /* make list start at 2nd pcb */
+    raw_pcbs = raw_pcbs->next;
+    /* 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) {
+      /* remove pcb from list */
+      pcb2->next = pcb->next;
+    }
+  }
+  memp_free(MEMP_RAW_PCB, pcb);
+}
+
+/**
+ * Create a RAW PCB.
+ *
+ * @return The RAW PCB which was created. NULL if the PCB data structure
+ * could not be allocated.
+ *
+ * @param proto the protocol number of the IPs payload (e.g. IP_PROTO_ICMP)
+ *
+ * @see raw_remove()
+ */
+struct raw_pcb *
+raw_new(u16_t proto) {
+  struct raw_pcb *pcb;
+
+  LWIP_DEBUGF(RAW_DEBUG | DBG_TRACE | 3, ("raw_new\n"));
+
+  pcb = memp_malloc(MEMP_RAW_PCB);
+  /* could allocate RAW PCB? */
+  if (pcb != NULL) {
+    /* initialize PCB to all zeroes */
+    memset(pcb, 0, sizeof(struct raw_pcb));
+    pcb->protocol = proto;
+    pcb->ttl = RAW_TTL;
+    pcb->next = raw_pcbs;
+    raw_pcbs = pcb;
+  }
+  return pcb;
+}
+
+#endif /* LWIP_RAW */

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, 
@@ -41,11 +41,74 @@
 
 #if LWIP_STATS
 struct stats_ lwip_stats;
-/*-----------------------------------------------------------------------------------*/
+
 void
 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,
@@ -44,7 +44,7 @@ struct sswt_cb
 };
 
 
-/*-----------------------------------------------------------------------------------*/
+
 void
 sys_mbox_fetch(sys_mbox_t mbox, void **msg)
 {
@@ -96,7 +96,7 @@ sys_mbox_fetch(sys_mbox_t mbox, void **msg)
 
   }
 }
-/*-----------------------------------------------------------------------------------*/
+
 void
 sys_sem_wait(sys_sem_t sem)
 {
@@ -152,7 +152,7 @@ sys_sem_wait(sys_sem_t sem)
 
   }
 }
-/*-----------------------------------------------------------------------------------*/
+
 void
 sys_timeout(u32_t msecs, sys_timeout_handler h, void *arg)
 {
@@ -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;
       }
     }
   }
@@ -203,7 +203,7 @@ sys_timeout(u32_t msecs, sys_timeout_handler h, void *arg)
 /* Go through timeout list (for this task only) and remove the first matching entry,
    even though the timeout has not triggered yet.
 */
-/*-----------------------------------------------------------------------------------*/
+
 void
 sys_untimeout(sys_timeout_handler h, void *arg)
 {
@@ -238,7 +238,7 @@ sys_untimeout(sys_timeout_handler h, void *arg)
 
 
 
-/*-----------------------------------------------------------------------------------*/
+
 static void
 sswt_handler(void *arg)
 {
@@ -252,7 +252,7 @@ sswt_handler(void *arg)
 /* Wait for a semaphore with timeout (specified in ms) */
 /* timeout = 0: wait forever */
 /* Returns 0 on timeout. 1 otherwise */
-/*-----------------------------------------------------------------------------------*/
+
 int
 sys_sem_wait_timeout(sys_sem_t sem, u32_t timeout)
 {
@@ -279,7 +279,7 @@ sys_sem_wait_timeout(sys_sem_t sem, u32_t timeout)
 
 }
 
-/*-----------------------------------------------------------------------------------*/
+
 void
 sys_msleep(u32_t ms)
 {
@@ -289,6 +289,6 @@ sys_msleep(u32_t ms)
 
   sys_sem_free(delaysem);
 }
-/*-----------------------------------------------------------------------------------*/
+
 
 #endif /* NO_SYS */

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;
@@ -95,41 +87,28 @@ tcp_init(void)
   tcp_timer = 0;
   
 }
-/*-----------------------------------------------------------------------------------*/
-/*
- * tcp_tmr():
- *
+
+/**
  * Called periodically to dispatch TCP timers.
  *
  */
-/*-----------------------------------------------------------------------------------*/
 void
 tcp_tmr(void)
 {
-  ++tcp_timer;
-  if (tcp_timer == 10) {
-    tcp_timer = 0;
-  }
-  
-  if (tcp_timer & 1) {
-    /* Call tcp_fasttmr() every 200 ms, i.e., every other timer
-       tcp_tmr() is called. */
-    tcp_fasttmr();
-  }
-  if (tcp_timer == 0 || tcp_timer == 5) {
-    /* Call tcp_slowtmr() every 500 ms, i.e., every fifth timer
+  /* Call tcp_fasttmr() every 250 ms */
+  tcp_fasttmr();
+
+  if (++tcp_timer & 1) {
+    /* Call tcp_tmr() every 500 ms, i.e., every other timer
        tcp_tmr() is called. */
     tcp_slowtmr();
   }
 }
-/*-----------------------------------------------------------------------------------*/
-/*
- * tcp_close():
- *
+
+/**
  * Closes the connection held by the PCB.
  *
  */
-/*-----------------------------------------------------------------------------------*/
 err_t
 tcp_close(struct tcp_pcb *pcb)
 {
@@ -143,7 +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;
@@ -154,11 +133,6 @@ tcp_close(struct tcp_pcb *pcb)
     pcb = NULL;
     break;
   case SYN_RCVD:
-    err = tcp_send_ctrl(pcb, TCP_FIN);
-    if (err == ERR_OK) {
-      pcb->state = FIN_WAIT_1;
-    }
-    break;
   case ESTABLISHED:
     err = tcp_send_ctrl(pcb, TCP_FIN);
     if (err == ERR_OK) {
@@ -183,16 +157,13 @@ 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)
 {
@@ -240,27 +211,28 @@ tcp_abort(struct tcp_pcb *pcb)
     tcp_rst(seqno, ackno, &local_ip, &remote_ip, local_port, remote_port);
   }
 }
-/*-----------------------------------------------------------------------------------*/
-/*
- * 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.
  *
  */
-/*-----------------------------------------------------------------------------------*/
+
 err_t
 tcp_bind(struct tcp_pcb *pcb, struct ip_addr *ipaddr, u16_t port)
 {
   struct tcp_pcb *cpcb;
+#if SO_REUSE
+  int reuse_port_all_set = 1;
+#endif /* SO_REUSE */
 
   if (port == 0) {
     port = tcp_new_port();
   }
-
+#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)) ||
@@ -280,6 +252,102 @@ tcp_bind(struct tcp_pcb *pcb, struct ip_addr *ipaddr, u16_t port)
       }
     }
   }
+#else /* SO_REUSE */
+  /* Search through list of PCB's in LISTEN state. 
+     
+  If there is a PCB bound to specified port and IP_ADDR_ANY another PCB can be bound to the interface IP
+  or to the loopback address on the same port if SOF_REUSEADDR is set. Any combination of PCB's bound to 
+  the same local port, but to one address out of {IP_ADDR_ANY, 127.0.0.1, interface IP} at a time is valid.
+  But no two PCB's bound to same local port and same local address is valid.
+  
+  If SOF_REUSEPORT is set several PCB's can be bound to same local port and same local address also. But then 
+  all PCB's must have the SOF_REUSEPORT option set.
+  
+  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.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) {
+          LWIP_DEBUGF(TCP_DEBUG, ("tcp_bind: in listening PCB's: SO_REUSEPORT set and same address.\n"));
+          reuse_port_all_set = (reuse_port_all_set && (cpcb->so_options & SOF_REUSEPORT));
+        }
+        else {
+          LWIP_DEBUGF(TCP_DEBUG, ("tcp_bind: in listening PCB's: SO_REUSEPORT not set and same address.\n"));
+          return ERR_USE;
+        }
+      }
+      else if((ip_addr_isany(ipaddr) && !ip_addr_isany(&(cpcb->local_ip))) ||
+              (!ip_addr_isany(ipaddr) && ip_addr_isany(&(cpcb->local_ip)))) {
+        if(!(pcb->so_options & SOF_REUSEADDR) && !(pcb->so_options & SOF_REUSEPORT)) {
+          LWIP_DEBUGF(TCP_DEBUG, ("tcp_bind: in listening PCB's SO_REUSEPORT or SO_REUSEADDR not set and not the same address.\n"));
+          return ERR_USE;
+        }      
+        else {
+          LWIP_DEBUGF(TCP_DEBUG, ("tcp_bind: in listening PCB's SO_REUSEPORT or SO_REUSEADDR set and not the same address.\n"));
+        }     
+      }
+    }
+  }
+
+  /* Search through list of PCB's in a state in which they can accept or send data. Same decription as for 
+     PCB's in state LISTEN applies to this PCB's regarding the options SOF_REUSEADDR and SOF_REUSEPORT. */
+  for(cpcb = tcp_active_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) {
+          LWIP_DEBUGF(TCP_DEBUG, ("tcp_bind: in active PCB's SO_REUSEPORT set and same address.\n"));
+          reuse_port_all_set = (reuse_port_all_set && (cpcb->so_options & SOF_REUSEPORT));
+        }
+        else {
+          LWIP_DEBUGF(TCP_DEBUG, ("tcp_bind: in active PCB's SO_REUSEPORT not set and same address.\n"));
+          return ERR_USE;
+        }
+      }
+      else if((ip_addr_isany(ipaddr) && !ip_addr_isany(&(cpcb->local_ip))) ||
+              (!ip_addr_isany(ipaddr) && ip_addr_isany(&(cpcb->local_ip)))) {
+        if(!(pcb->so_options & SOF_REUSEADDR) && !(pcb->so_options & SOF_REUSEPORT)) {
+          LWIP_DEBUGF(TCP_DEBUG, ("tcp_bind: in active PCB's SO_REUSEPORT or SO_REUSEADDR not set and not the same address.\n"));
+          return ERR_USE;
+        }   
+        else {
+          LWIP_DEBUGF(TCP_DEBUG, ("tcp_bind: in active PCB's SO_REUSEPORT or SO_REUSEADDR set and not the same address.\n"));
+        }        
+      }
+    }
+  }
+
+  /* Search through list of PCB's in TIME_WAIT state. If SO_REUSEADDR is set a bound combination [IP, port} 
+     can be rebound. The same applies when SOF_REUSEPORT is set. 
+     
+     If SOF_REUSEPORT is set several PCB's can be bound to same local port and same local address also. But then 
+     all PCB's must have the SOF_REUSEPORT option set.
+     
+     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 = tcp_tw_pcbs; cpcb != NULL; cpcb = cpcb->next) {
+    if(cpcb->local_port == port) {
+      if(ip_addr_cmp(&(cpcb->local_ip), ipaddr)) {
+        if(!(pcb->so_options & SOF_REUSEADDR) && !(pcb->so_options & SOF_REUSEPORT)) {
+          LWIP_DEBUGF(TCP_DEBUG, ("tcp_bind: in TIME_WAIT PCB's SO_REUSEPORT or SO_REUSEADDR not set and same address.\n"));
+          return ERR_USE;
+        }
+        else if(pcb->so_options & SOF_REUSEPORT) {
+          LWIP_DEBUGF(TCP_DEBUG, ("tcp_bind: in TIME_WAIT PCB's SO_REUSEPORT set and same address.\n"));
+          reuse_port_all_set = (reuse_port_all_set && (cpcb->so_options & SOF_REUSEPORT));
+        }
+      }
+    }
+  }
+
+  /* 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) {
+    LWIP_DEBUGF(TCP_DEBUG, ("tcp_bind: not all sockets have SO_REUSEPORT set.\n"));
+    return ERR_USE;
+  }
+#endif /* SO_REUSE */
+
   if (!ip_addr_isany(ipaddr)) {
     pcb->local_ip = *ipaddr;
   }
@@ -298,17 +366,14 @@ tcp_accept_null(void *arg, struct tcp_pcb *pcb, err_t err)
   return ERR_ABRT;
 }
 #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)
 {
@@ -325,47 +390,56 @@ tcp_listen(struct tcp_pcb *pcb)
   lpcb->callback_arg = pcb->callback_arg;
   lpcb->local_port = pcb->local_port;
   lpcb->state = LISTEN;
+  lpcb->so_options = pcb->so_options;
+  lpcb->so_options |= SOF_ACCEPTCONN;
+  lpcb->ttl = pcb->ttl;
+  lpcb->tos = pcb->tos;
   ip_addr_set(&lpcb->local_ip, &pcb->local_ip);
   memp_free(MEMP_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)
 {
-  pcb->rcv_wnd += len;
-  if (pcb->rcv_wnd > TCP_WND) {
+  if ((u32_t)pcb->rcv_wnd + len > TCP_WND) {
     pcb->rcv_wnd = TCP_WND;
+  } else {
+    pcb->rcv_wnd += 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);
   }
 
   LWIP_DEBUGF(TCP_DEBUG, ("tcp_recved: recveived %u bytes, wnd %u (%u).\n",
          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)
 {
@@ -391,22 +465,19 @@ 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;
     }
   }
   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))
@@ -453,15 +524,12 @@ 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)
 {
@@ -527,6 +595,21 @@ tcp_slowtmr(void)
       }
     }
 
+   /* Check if KEEPALIVE should be sent */
+   if((pcb->so_options & SOF_KEEPALIVE) && ((pcb->state == ESTABLISHED) || (pcb->state == CLOSE_WAIT))) {
+      if((u32_t)(tcp_ticks - pcb->tmr) > (pcb->keepalive + TCP_MAXIDLE) / TCP_SLOW_INTERVAL)  {
+         LWIP_DEBUGF(TCP_DEBUG, ("tcp_slowtmr: KEEPALIVE timeout. Aborting connection to %u.%u.%u.%u.\n",
+                                 ip4_addr1(&pcb->remote_ip), ip4_addr2(&pcb->remote_ip),
+                                 ip4_addr3(&pcb->remote_ip), ip4_addr4(&pcb->remote_ip)));
+
+         tcp_abort(pcb);
+      }
+      else if((u32_t)(tcp_ticks - pcb->tmr) > (pcb->keepalive + pcb->keep_cnt * TCP_KEEPINTVL) / TCP_SLOW_INTERVAL) {
+         tcp_keepalive(pcb);
+         pcb->keep_cnt++;
+      }
+   }
+
     /* If this PCB has queued out of sequence data, but has been
        inactive for too long, will drop the data (it will eventually
        be retransmitted). */
@@ -622,13 +705,10 @@ tcp_slowtmr(void)
     }
   }
 }
-/*-----------------------------------------------------------------------------------*/
-/*
- * tcp_fasttmr():
- *
- * Is called every TCP_FINE_TIMEOUT (100 ms) and sends delayed ACKs.
+
+/**
+ * Is called every TCP_FAST_INTERVAL (250 ms) and sends delayed ACKs.
  */
-/*-----------------------------------------------------------------------------------*/
 void
 tcp_fasttmr(void)
 {
@@ -643,75 +723,60 @@ tcp_fasttmr(void)
     }
   }
 }
-/*-----------------------------------------------------------------------------------*/
-/*
- * tcp_segs_free():
- *
+
+/**
  * Deallocates a list of TCP segments (tcp_seg structures).
  *
  */
-/*-----------------------------------------------------------------------------------*/
 u8_t
 tcp_segs_free(struct tcp_seg *seg)
 {
   u8_t count = 0;
   struct tcp_seg *next;
- again:  
-  if (seg != NULL) {
+  while (seg != NULL) {
     next = seg->next;
     count += tcp_seg_free(seg);
     seg = next;
-    goto again;
   }
   return count;
 }
-/*-----------------------------------------------------------------------------------*/
-/*
- * tcp_seg_free():
- *
+
+/**
  * Frees a TCP segment.
  *
  */
-/*-----------------------------------------------------------------------------------*/
 u8_t
 tcp_seg_free(struct tcp_seg *seg)
 {
   u8_t count = 0;
   
   if (seg != NULL) {
-    if (seg->p == NULL) {
-      memp_free(MEMP_TCP_SEG, seg);
-    } else {
+    if (seg->p != NULL) {
       count = pbuf_free(seg->p);
 #if TCP_DEBUG
       seg->p = NULL;
 #endif /* TCP_DEBUG */
-      memp_free(MEMP_TCP_SEG, seg);
     }
+    memp_free(MEMP_TCP_SEG, seg);
   }
   return count;
 }
-/*-----------------------------------------------------------------------------------*/
-/*
- * tcp_setprio():
- *
+
+/**
  * Sets the priority of a connection.
  *
  */
-/*-----------------------------------------------------------------------------------*/
 void
 tcp_setprio(struct tcp_pcb *pcb, u8_t prio)
 {
   pcb->prio = prio;
 }
-/*-----------------------------------------------------------------------------------*/
-/*
- * tcp_seg_copy():
- *
+#if TCP_QUEUE_OOSEQ
+
+/**
  * Returns a copy of the given TCP segment.
  *
  */ 
-/*-----------------------------------------------------------------------------------*/
 struct tcp_seg *
 tcp_seg_copy(struct tcp_seg *seg)
 {
@@ -725,7 +790,8 @@ tcp_seg_copy(struct tcp_seg *seg)
   pbuf_ref(cseg->p);
   return cseg;
 }
-/*-----------------------------------------------------------------------------------*/
+#endif
+
 #if LWIP_CALLBACK_API
 static err_t
 tcp_recv_null(void *arg, struct tcp_pcb *pcb, struct pbuf *p, err_t err)
@@ -739,7 +805,7 @@ tcp_recv_null(void *arg, struct tcp_pcb *pcb, struct pbuf *p, err_t err)
   return ERR_OK;
 }
 #endif /* LWIP_CALLBACK_API */
-/*-----------------------------------------------------------------------------------*/
+
 static void
 tcp_kill_prio(u8_t prio)
 {
@@ -770,7 +836,7 @@ tcp_kill_prio(u8_t prio)
   }      
 }
 
-/*-----------------------------------------------------------------------------------*/
+
 static void
 tcp_kill_timewait(void)
 {
@@ -792,8 +858,8 @@ tcp_kill_timewait(void)
   }      
 }
 
-/*-----------------------------------------------------------------------------------*/
-/*-----------------------------------------------------------------------------------*/
+
+
 struct tcp_pcb *
 tcp_alloc(u8_t prio)
 {
@@ -817,6 +883,8 @@ tcp_alloc(u8_t prio)
     pcb->snd_buf = TCP_SND_BUF;
     pcb->snd_queuelen = 0;
     pcb->rcv_wnd = TCP_WND;
+    pcb->tos = 0;
+    pcb->ttl = TCP_TTL;
     pcb->mss = TCP_MSS;
     pcb->rto = 3000 / TCP_SLOW_INTERVAL;
     pcb->sa = 0;
@@ -836,24 +904,29 @@ tcp_alloc(u8_t prio)
 #if LWIP_CALLBACK_API
     pcb->recv = tcp_recv_null;
 #endif /* LWIP_CALLBACK_API */  
+    
+    /* Init KEEPALIVE timer */
+    pcb->keepalive = TCP_KEEPDEFAULT;
+    pcb->keep_cnt = 0;
   }
   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 *
 tcp_new(void)
 {
   return tcp_alloc(TCP_PRIO_NORMAL);
 }
-/*-----------------------------------------------------------------------------------*/
+
 /*
  * tcp_arg():
  *
@@ -861,73 +934,71 @@ tcp_new(void)
  * functions.
  *
  */ 
-/*-----------------------------------------------------------------------------------*/
+
 void
 tcp_arg(struct tcp_pcb *pcb, void *arg)
 {  
   pcb->callback_arg = arg;
 }
-/*-----------------------------------------------------------------------------------*/
-/*
- * tcp_recv():
- *
+#if LWIP_CALLBACK_API
+
+/**
  * Used to specify the function that should be called when a TCP
  * connection receives data.
  *
  */ 
-/*-----------------------------------------------------------------------------------*/
-#if LWIP_CALLBACK_API
 void
 tcp_recv(struct tcp_pcb *pcb,
    err_t (* recv)(void *arg, struct tcp_pcb *tpcb, struct pbuf *p, err_t err))
 {
   pcb->recv = recv;
 }
-#endif /* LWIP_CALLBACK_API */
-/*-----------------------------------------------------------------------------------*/
-/*
- * tcp_sent():
- *
+
+/**
  * Used to specify the function that should be called when TCP data
  * has been successfully delivered to the remote host.
  *
  */ 
-/*-----------------------------------------------------------------------------------*/
-#if LWIP_CALLBACK_API
+
 void
 tcp_sent(struct tcp_pcb *pcb,
    err_t (* sent)(void *arg, struct tcp_pcb *tpcb, u16_t len))
 {
   pcb->sent = sent;
 }
-#endif /* LWIP_CALLBACK_API */
-/*-----------------------------------------------------------------------------------*/
-/*
- * tcp_err():
- *
+
+/**
  * Used to specify the function that should be called when a fatal error
  * has occured on the connection.
  *
  */ 
-/*-----------------------------------------------------------------------------------*/
-#if LWIP_CALLBACK_API
 void
 tcp_err(struct tcp_pcb *pcb,
    void (* errf)(void *arg, err_t err))
 {
   pcb->errf = errf;
 }
-#endif /* LWIP_CALLBACK_API */
-/*-----------------------------------------------------------------------------------*/
-/*
- * tcp_poll():
+
+/**
+ * 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))
+{
+  ((struct tcp_pcb_listen *)pcb)->accept = accept;
+}
+#endif /* LWIP_CALLBACK_API */
+
+
+/**
  * 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)
@@ -937,31 +1008,11 @@ tcp_poll(struct tcp_pcb *pcb,
 #endif /* LWIP_CALLBACK_API */  
   pcb->pollinterval = interval;
 }
-/*-----------------------------------------------------------------------------------*/
-/*
- * tcp_accept():
- *
- * Used for specifying the function that should be called when a
- * LISTENing connection has been connected to another host.
- *
- */ 
-/*-----------------------------------------------------------------------------------*/
-#if LWIP_CALLBACK_API
-void
-tcp_accept(struct tcp_pcb *pcb,
-     err_t (* accept)(void *arg, struct tcp_pcb *newpcb, err_t err))
-{
-  ((struct tcp_pcb_listen *)pcb)->accept = accept;
-}
-#endif /* LWIP_CALLBACK_API */
-/*-----------------------------------------------------------------------------------*/
-/*
- * tcp_pcb_purge():
- *
+
+/**
  * Purges a TCP PCB. Removes any buffered data and frees the buffer memory.
  *
  */
-/*-----------------------------------------------------------------------------------*/
 void
 tcp_pcb_purge(struct tcp_pcb *pcb)
 {
@@ -971,7 +1022,6 @@ tcp_pcb_purge(struct tcp_pcb *pcb)
 
     LWIP_DEBUGF(TCP_DEBUG, ("tcp_pcb_purge\n"));
     
-#if TCP_DEBUG
     if (pcb->unsent != NULL) {    
       LWIP_DEBUGF(TCP_DEBUG, ("tcp_pcb_purge: not all data sent\n"));
     }
@@ -982,28 +1032,20 @@ tcp_pcb_purge(struct tcp_pcb *pcb)
     if (pcb->ooseq != NULL) {    
       LWIP_DEBUGF(TCP_DEBUG, ("tcp_pcb_purge: data left on ->ooseq\n"));
     }
-#endif
-#endif /* TCP_DEBUG */
-    tcp_segs_free(pcb->unsent);
-#if TCP_QUEUE_OOSEQ
+    
     tcp_segs_free(pcb->ooseq);
+    pcb->ooseq = NULL;
 #endif /* TCP_QUEUE_OOSEQ */
+    tcp_segs_free(pcb->unsent);
     tcp_segs_free(pcb->unacked);
-    pcb->unacked = pcb->unsent =
-#if TCP_QUEUE_OOSEQ
-      pcb->ooseq =
-#endif /* TCP_QUEUE_OOSEQ */
-      NULL;
+    pcb->unacked = pcb->unsent = NULL;
   }
 }
-/*-----------------------------------------------------------------------------------*/
-/*
- * 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)
 {
@@ -1022,14 +1064,11 @@ 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)
 {
@@ -1038,31 +1077,31 @@ tcp_next_iss(void)
   iss += tcp_ticks;       /* XXX */
   return iss;
 }
-/*-----------------------------------------------------------------------------------*/
+
 #if TCP_DEBUG || TCP_INPUT_DEBUG || TCP_OUTPUT_DEBUG
 void
 tcp_debug_print(struct tcp_hdr *tcphdr)
 {
   LWIP_DEBUGF(TCP_DEBUG, ("TCP header:\n"));
   LWIP_DEBUGF(TCP_DEBUG, ("+-------------------------------+\n"));
-  LWIP_DEBUGF(TCP_DEBUG, ("|      %04x     |      %04x     | (src port, dest port)\n",
-         tcphdr->src, tcphdr->dest));
+  LWIP_DEBUGF(TCP_DEBUG, ("|    %5u      |    %5u      | (src port, dest port)\n",
+         ntohs(tcphdr->src), ntohs(tcphdr->dest)));
   LWIP_DEBUGF(TCP_DEBUG, ("+-------------------------------+\n"));
-  LWIP_DEBUGF(TCP_DEBUG, ("|            %08lu           | (seq no)\n",
-          tcphdr->seqno));
+  LWIP_DEBUGF(TCP_DEBUG, ("|           %010lu          | (seq no)\n",
+          ntohl(tcphdr->seqno)));
   LWIP_DEBUGF(TCP_DEBUG, ("+-------------------------------+\n"));
-  LWIP_DEBUGF(TCP_DEBUG, ("|            %08lu           | (ack no)\n",
-         tcphdr->ackno));
+  LWIP_DEBUGF(TCP_DEBUG, ("|           %010lu          | (ack no)\n",
+         ntohl(tcphdr->ackno)));
   LWIP_DEBUGF(TCP_DEBUG, ("+-------------------------------+\n"));
-  LWIP_DEBUGF(TCP_DEBUG, ("| %2u |    |%u%u%u%u%u|    %5u      | (offset, flags (",
-       TCPH_OFFSET(tcphdr),
-         TCPH_FLAGS(tcphdr) >> 4 & 1,
+  LWIP_DEBUGF(TCP_DEBUG, ("| %2u |   |%u%u%u%u%u%u|     %5u     | (hdrlen, flags (",
+       TCPH_HDRLEN(tcphdr),
+         TCPH_FLAGS(tcphdr) >> 5 & 1,
          TCPH_FLAGS(tcphdr) >> 4 & 1,
          TCPH_FLAGS(tcphdr) >> 3 & 1,
          TCPH_FLAGS(tcphdr) >> 2 & 1,
          TCPH_FLAGS(tcphdr) >> 1 & 1,
          TCPH_FLAGS(tcphdr) & 1,
-         tcphdr->wnd));
+         ntohs(tcphdr->wnd)));
   tcp_debug_print_flags(TCPH_FLAGS(tcphdr));
   LWIP_DEBUGF(TCP_DEBUG, ("), win)\n"));
   LWIP_DEBUGF(TCP_DEBUG, ("+-------------------------------+\n"));
@@ -1070,7 +1109,7 @@ tcp_debug_print(struct tcp_hdr *tcphdr)
          ntohs(tcphdr->chksum), ntohs(tcphdr->urgp)));
   LWIP_DEBUGF(TCP_DEBUG, ("+-------------------------------+\n"));
 }
-/*-----------------------------------------------------------------------------------*/
+
 void
 tcp_debug_print_state(enum tcp_state s)
 {
@@ -1111,7 +1150,7 @@ tcp_debug_print_state(enum tcp_state s)
    break;
   }
 }
-/*-----------------------------------------------------------------------------------*/
+
 void
 tcp_debug_print_flags(u8_t flags)
 {
@@ -1133,8 +1172,14 @@ tcp_debug_print_flags(u8_t flags)
   if (flags & TCP_URG) {
     LWIP_DEBUGF(TCP_DEBUG, ("URG "));
   }
+  if (flags & TCP_ECE) {
+    LWIP_DEBUGF(TCP_DEBUG, ("ECE "));
+  }
+  if (flags & TCP_CWR) {
+    LWIP_DEBUGF(TCP_DEBUG, ("CWR "));
+  }
 }
-/*-----------------------------------------------------------------------------------*/
+
 void
 tcp_debug_print_pcbs(void)
 {
@@ -1147,7 +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));
@@ -1161,7 +1206,7 @@ tcp_debug_print_pcbs(void)
     tcp_debug_print_state(pcb->state);
   }    
 }
-/*-----------------------------------------------------------------------------------*/
+
 int
 tcp_pcbs_sane(void)
 {
@@ -1178,7 +1223,7 @@ tcp_pcbs_sane(void)
 }
 #endif /* TCP_DEBUG */
 #endif /* LWIP_TCP */
-/*-----------------------------------------------------------------------------------*/
+
 
 
 

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"
@@ -85,7 +80,7 @@ static void tcp_parseopt(struct tcp_pcb *pcb);
 static err_t tcp_listen_input(struct tcp_pcb_listen *pcb);
 static err_t tcp_timewait_input(struct tcp_pcb *pcb);
 
-/*-----------------------------------------------------------------------------------*/
+
 /* tcp_input:
  *
  * The initial input processing of TCP. It verifies the TCP header, demultiplexes
@@ -93,45 +88,50 @@ static err_t tcp_timewait_input(struct tcp_pcb *pcb);
  * the TCP finite state machine. This function is called by the IP layer (in
  * ip_input()).
  */
-/*-----------------------------------------------------------------------------------*/
+
 void
 tcp_input(struct pbuf *p, struct netif *inp)
 {
   struct tcp_pcb *pcb, *prev;
   struct tcp_pcb_listen *lpcb;
-  u8_t offset;
+  u8_t hdrlen;
   err_t err;
 
+#if SO_REUSE
+  struct tcp_pcb *pcb_temp;
+  int reuse = 0;
+  int reuse_port = 0;
+#endif /* SO_REUSE */
 
   PERF_START;
 
-
-#ifdef TCP_STATS
-  ++lwip_stats.tcp.recv;
-#endif /* TCP_STATS */
+  TCP_STATS_INC(tcp.recv);
 
   iphdr = p->payload;
   tcphdr = (struct tcp_hdr *)((u8_t *)p->payload + IPH_HL(iphdr) * 4);
 
+#if TCP_INPUT_DEBUG
+  tcp_debug_print(tcphdr);
+#endif
+
   /* remove header from payload */
   if (pbuf_header(p, -((s16_t)(IPH_HL(iphdr) * 4))) || (p->tot_len < sizeof(struct tcp_hdr))) {
     /* drop short packets */
     LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_input: short packet (%u bytes) discarded\n", p->tot_len));
-#ifdef TCP_STATS
-    ++lwip_stats.tcp.lenerr;
-    ++lwip_stats.tcp.drop;
-#endif /* TCP_STATS */
+    TCP_STATS_INC(tcp.lenerr);
+    TCP_STATS_INC(tcp.drop);
     pbuf_free(p);
     return;
   }
 
   /* 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),
@@ -142,20 +142,18 @@ tcp_input(struct pbuf *p, struct netif *inp)
 #if TCP_DEBUG
     tcp_debug_print(tcphdr);
 #endif /* TCP_DEBUG */
-#ifdef TCP_STATS
-    ++lwip_stats.tcp.chkerr;
-    ++lwip_stats.tcp.drop;
-#endif /* TCP_STATS */
+    TCP_STATS_INC(tcp.chkerr);
+    TCP_STATS_INC(tcp.drop);
 
     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. */
-  offset = TCPH_OFFSET(tcphdr) >> 4;
-  pbuf_header(p, -(offset * 4));
+  hdrlen = TCPH_HDRLEN(tcphdr);
+  pbuf_header(p, -(hdrlen * 4));
 
   /* Convert fields in TCP header to host byte order. */
   tcphdr->src = ntohs(tcphdr->src);
@@ -170,7 +168,17 @@ tcp_input(struct pbuf *p, struct netif *inp)
   /* Demultiplex an incoming segment. First, we check if it is destined
      for an active connection. */
   prev = NULL;
+
+#if SO_REUSE
+  pcb_temp = tcp_active_pcbs;
+  
+ again_1:
+  
+  /* Iterate through the TCP pcb list for a fully matching pcb */
+  for(pcb = pcb_temp; pcb != NULL; pcb = pcb->next) {
+#else  /* SO_REUSE */
   for(pcb = tcp_active_pcbs; pcb != NULL; pcb = pcb->next) {
+#endif  /* SO_REUSE */
     LWIP_ASSERT("tcp_input: active pcb->state != CLOSED", pcb->state != CLOSED);
     LWIP_ASSERT("tcp_input: active pcb->state != TIME-WAIT", pcb->state != TIME_WAIT);
     LWIP_ASSERT("tcp_input: active pcb->state != LISTEN", pcb->state != LISTEN);
@@ -179,6 +187,32 @@ tcp_input(struct pbuf *p, struct netif *inp)
        ip_addr_cmp(&(pcb->remote_ip), &(iphdr->src)) &&
        ip_addr_cmp(&(pcb->local_ip), &(iphdr->dest))) {
 
+#if SO_REUSE
+      if(pcb->so_options & SOF_REUSEPORT) {
+        if(reuse) {
+          /* We processed one PCB already */
+          LWIP_DEBUGF(TCP_INPUT_DEBUG,("tcp_input: second or later PCB and SOF_REUSEPORT set.\n"));
+        } else {
+          /* First PCB with this address */
+          LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_input: first PCB and SOF_REUSEPORT set.\n"));
+          reuse = 1;
+        }
+        
+        reuse_port = 1; 
+        p->ref++;
+        
+        /* We want to search on next socket after receiving */
+        pcb_temp = pcb->next;
+        
+        LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_input: reference counter on PBUF set to %i\n", p->ref));
+      } else  {
+        if(reuse) {
+          /* We processed one PCB already */
+          LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_input: second or later PCB but SOF_REUSEPORT not set !\n"));
+        }
+      }
+#endif /* SO_REUSE */
+
       /* Move this PCB to the front of the list so that subsequent
    lookups will be faster (we exploit locality in TCP segment
    arrivals). */
@@ -217,7 +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) {
@@ -227,9 +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"));
@@ -323,33 +357,50 @@ tcp_input(struct pbuf *p, struct netif *inp)
     tcp_debug_print_state(pcb->state);
 #endif /* TCP_DEBUG */
 #endif /* TCP_INPUT_DEBUG */
+#if SO_REUSE
+    /* First socket should receive now */
+    if(reuse_port) {
+      LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_input: searching next PCB.\n"));
+      reuse_port = 0;
+      
+      /* We are searching connected sockets */
+      goto again_1;
+    }
+#endif /* SO_REUSE */
 
   } else {
+#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);
+      goto end;
+    }
+#endif /* SO_REUSE */
     /* If no matching PCB was found, send a TCP RST (reset) to the
        sender. */
     LWIP_DEBUGF(TCP_RST_DEBUG, ("tcp_input: no PCB match found, resetting.\n"));
     if (!(TCPH_FLAGS(tcphdr) & TCP_RST)) {
-#ifdef TCP_STATS
-      ++lwip_stats.tcp.proterr;
-      ++lwip_stats.tcp.drop;
-#endif /* TCP_STATS */
+      TCP_STATS_INC(tcp.proterr);
+      TCP_STATS_INC(tcp.drop);
       tcp_rst(ackno, seqno + tcplen,
         &(iphdr->dest), &(iphdr->src),
         tcphdr->dest, tcphdr->src);
     }
     pbuf_free(p);
   }
-
+#if SO_REUSE
+ end:
+#endif /* SO_REUSE */
   LWIP_ASSERT("tcp_input: tcp_pcbs_sane()", tcp_pcbs_sane());
   PERF_STOP("tcp_input");
 }
-/*-----------------------------------------------------------------------------------*/
+
 /* tcp_listen_input():
  *
  * Called by tcp_input() when a segment arrives for a listening
  * connection.
  */
-/*-----------------------------------------------------------------------------------*/
+
 static err_t
 tcp_listen_input(struct tcp_pcb_listen *pcb)
 {
@@ -366,16 +417,14 @@ tcp_listen_input(struct tcp_pcb_listen *pcb)
       &(iphdr->dest), &(iphdr->src),
       tcphdr->dest, tcphdr->src);
   } else if (flags & TCP_SYN) {
-    LWIP_DEBUGF(DEMO_DEBUG, ("TCP connection request %u -> %u.\n", tcphdr->src, tcphdr->dest));
+    LWIP_DEBUGF(TCP_DEBUG, ("TCP connection request %u -> %u.\n", tcphdr->src, tcphdr->dest));
     npcb = tcp_alloc(pcb->prio);
     /* If a new PCB could not be created (probably due to lack of memory),
        we don't do anything, but rely on the sender will retransmit the
        SYN at a time when we have more memory available. */
     if (npcb == NULL) {
       LWIP_DEBUGF(TCP_DEBUG, ("tcp_listen_input: could not allocate PCB\n"));
-#ifdef TCP_STATS
-      ++lwip_stats.tcp.memerr;
-#endif /* TCP_STATS */
+      TCP_STATS_INC(tcp.memerr);
       return ERR_MEM;
     }
     /* Set up the new PCB. */
@@ -387,12 +436,13 @@ tcp_listen_input(struct tcp_pcb_listen *pcb)
     npcb->rcv_nxt = seqno + 1;
     npcb->snd_wnd = tcphdr->wnd;
     npcb->ssthresh = npcb->snd_wnd;
-    npcb->snd_wl1 = seqno;
+    npcb->snd_wl1 = seqno - 1;/* initialise to seqno-1 to force window update */
     npcb->callback_arg = pcb->callback_arg;
 #if LWIP_CALLBACK_API
     npcb->accept = pcb->accept;
 #endif /* LWIP_CALLBACK_API */
-
+    /* inherit socket options */
+    npcb->so_options = pcb->so_options & (SOF_DEBUG|SOF_DONTROUTE|SOF_KEEPALIVE|SOF_OOBINLINE|SOF_LINGER);
     /* Register the new PCB so that we can begin receiving segments
        for it. */
     TCP_REG(&tcp_active_pcbs, npcb);
@@ -411,13 +461,13 @@ tcp_listen_input(struct tcp_pcb_listen *pcb)
   }
   return ERR_OK;
 }
-/*-----------------------------------------------------------------------------------*/
+
 /* tcp_timewait_input():
  *
  * Called by tcp_input() when a segment arrives for a connection in
  * TIME_WAIT.
  */
-/*-----------------------------------------------------------------------------------*/
+
 static err_t
 tcp_timewait_input(struct tcp_pcb *pcb)
 {
@@ -429,7 +479,7 @@ tcp_timewait_input(struct tcp_pcb *pcb)
   }
   return tcp_output(pcb);
 }
-/*-----------------------------------------------------------------------------------*/
+
 /* tcp_process
  *
  * Implements the TCP state machine. Called by tcp_input. In some
@@ -437,7 +487,7 @@ tcp_timewait_input(struct tcp_pcb *pcb)
  * argument will be freed by the caller (tcp_input()) unless the
  * recv_data pointer in the pcb is set.
  */
-/*-----------------------------------------------------------------------------------*/
+
 static err_t
 tcp_process(struct tcp_pcb *pcb)
 {
@@ -479,17 +529,19 @@ tcp_process(struct tcp_pcb *pcb)
 
   /* Update the PCB (in)activity timer. */
   pcb->tmr = tcp_ticks;
+  pcb->keep_cnt = 0;
 
   /* Do different things depending on the TCP state. */
   switch (pcb->state) {
   case SYN_SENT:
     LWIP_DEBUGF(TCP_INPUT_DEBUG, ("SYN-SENT: ackno %lu pcb->snd_nxt %lu unacked %lu\n", ackno,
      pcb->snd_nxt, ntohl(pcb->unacked->tcphdr->seqno)));
-    if (flags & (TCP_ACK | TCP_SYN) &&
-       ackno == ntohl(pcb->unacked->tcphdr->seqno) + 1) {
+    if ((flags & TCP_ACK) && (flags & TCP_SYN)
+        && ackno == ntohl(pcb->unacked->tcphdr->seqno) + 1) {
       pcb->rcv_nxt = seqno + 1;
       pcb->lastack = ackno;
-      pcb->snd_wnd = pcb->snd_wl1 = tcphdr->wnd;
+      pcb->snd_wnd = tcphdr->wnd;
+      pcb->snd_wl1 = seqno - 1; /* initialise to seqno - 1 to force window update */
       pcb->state = ESTABLISHED;
       pcb->cwnd = pcb->mss;
       --pcb->snd_queuelen;
@@ -502,7 +554,7 @@ tcp_process(struct tcp_pcb *pcb)
       tcp_parseopt(pcb);
 
       /* Call the user specified function to call when sucessfully
-   connected. */
+       * connected. */
       TCP_EVENT_CONNECTED(pcb, ERR_OK, err);
       tcp_ack(pcb);
     }
@@ -511,22 +563,24 @@ tcp_process(struct tcp_pcb *pcb)
     if (flags & TCP_ACK &&
        !(flags & TCP_RST)) {
       if (TCP_SEQ_LT(pcb->lastack, ackno) &&
-   TCP_SEQ_LEQ(ackno, pcb->snd_nxt)) {
+          TCP_SEQ_LEQ(ackno, pcb->snd_nxt)) {
         pcb->state = ESTABLISHED;
-        LWIP_DEBUGF(DEMO_DEBUG, ("TCP connection established %u -> %u.\n", inseg.tcphdr->src, inseg.tcphdr->dest));
-  LWIP_ASSERT("pcb->accept != NULL", pcb->accept != NULL);
-  /* Call the accept function. */
-  TCP_EVENT_ACCEPT(pcb, ERR_OK, err);
-  if (err != ERR_OK) {
-    /* If the accept function returns with an error, we abort
-       the connection. */
-    tcp_abort(pcb);
-    return ERR_ABRT;
-  }
-  /* If there was any data contained within this ACK,
-     we'd better pass it on to the application as well. */
-  tcp_receive(pcb);
-  pcb->cwnd = pcb->mss;
+        LWIP_DEBUGF(TCP_DEBUG, ("TCP connection established %u -> %u.\n", inseg.tcphdr->src, inseg.tcphdr->dest));
+#if LWIP_CALLBACK_API
+        LWIP_ASSERT("pcb->accept != NULL", pcb->accept != NULL);
+#endif
+        /* Call the accept function. */
+        TCP_EVENT_ACCEPT(pcb, ERR_OK, err);
+        if (err != ERR_OK) {
+          /* If the accept function returns with an error, we abort
+           * the connection. */
+          tcp_abort(pcb);
+          return ERR_ABRT;
+        }
+        /* If there was any data contained within this ACK,
+         * we'd better pass it on to the application as well. */
+        tcp_receive(pcb);
+        pcb->cwnd = pcb->mss;
       }
     }
     break;
@@ -543,7 +597,7 @@ tcp_process(struct tcp_pcb *pcb)
     tcp_receive(pcb);
     if (flags & TCP_FIN) {
       if (flags & TCP_ACK && ackno == pcb->snd_nxt) {
-        LWIP_DEBUGF(DEMO_DEBUG,
+        LWIP_DEBUGF(TCP_DEBUG,
          ("TCP connection closed %d -> %d.\n", inseg.tcphdr->src, inseg.tcphdr->dest));
   tcp_ack_now(pcb);
   tcp_pcb_purge(pcb);
@@ -561,7 +615,7 @@ tcp_process(struct tcp_pcb *pcb)
   case FIN_WAIT_2:
     tcp_receive(pcb);
     if (flags & TCP_FIN) {
-      LWIP_DEBUGF(DEMO_DEBUG, ("TCP connection closed %u -> %u.\n", inseg.tcphdr->src, inseg.tcphdr->dest));
+      LWIP_DEBUGF(TCP_DEBUG, ("TCP connection closed %u -> %u.\n", inseg.tcphdr->src, inseg.tcphdr->dest));
       tcp_ack_now(pcb);
       tcp_pcb_purge(pcb);
       TCP_RMV(&tcp_active_pcbs, pcb);
@@ -572,7 +626,7 @@ tcp_process(struct tcp_pcb *pcb)
   case CLOSING:
     tcp_receive(pcb);
     if (flags & TCP_ACK && ackno == pcb->snd_nxt) {
-      LWIP_DEBUGF(DEMO_DEBUG, ("TCP connection closed %u -> %u.\n", inseg.tcphdr->src, inseg.tcphdr->dest));
+      LWIP_DEBUGF(TCP_DEBUG, ("TCP connection closed %u -> %u.\n", inseg.tcphdr->src, inseg.tcphdr->dest));
       tcp_ack_now(pcb);
       tcp_pcb_purge(pcb);
       TCP_RMV(&tcp_active_pcbs, pcb);
@@ -583,7 +637,7 @@ tcp_process(struct tcp_pcb *pcb)
   case LAST_ACK:
     tcp_receive(pcb);
     if (flags & TCP_ACK && ackno == pcb->snd_nxt) {
-      LWIP_DEBUGF(DEMO_DEBUG, ("TCP connection closed %u -> %u.\n", inseg.tcphdr->src, inseg.tcphdr->dest));
+      LWIP_DEBUGF(TCP_DEBUG, ("TCP connection closed %u -> %u.\n", inseg.tcphdr->src, inseg.tcphdr->dest));
       pcb->state = CLOSED;
       recv_flags = TF_CLOSED;
     }
@@ -594,7 +648,7 @@ tcp_process(struct tcp_pcb *pcb)
 
   return ERR_OK;
 }
-/*-----------------------------------------------------------------------------------*/
+
 /* tcp_receive:
  *
  * Called by tcp_process. Checks if the given segment is an ACK for outstanding
@@ -606,7 +660,7 @@ tcp_process(struct tcp_pcb *pcb)
  * If the incoming segment constitutes an ACK for a segment that was used for RTT
  * estimation, the RTT is estimated here as well.
  */
-/*-----------------------------------------------------------------------------------*/
+
 static void
 tcp_receive(struct tcp_pcb *pcb)
 {
@@ -934,7 +988,7 @@ tcp_receive(struct tcp_pcb *pcb)
     inseg.p = NULL;
   }
   if (TCPH_FLAGS(inseg.tcphdr) & TCP_FIN) {
-    LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_receive: received FIN."));
+    LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_receive: received FIN.\n"));
     recv_flags = TF_GOT_FIN;
   }
 
@@ -957,15 +1011,14 @@ tcp_receive(struct tcp_pcb *pcb)
       /* Chain this pbuf onto the pbuf that we will pass to
          the application. */
       if (recv_data) {
-              pbuf_chain(recv_data, cseg->p);
-              pbuf_free(cseg->p);
+              pbuf_cat(recv_data, cseg->p);
             } else {
         recv_data = cseg->p;
       }
       cseg->p = NULL;
     }
-    if (flags & TCP_FIN) {
-      LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_receive: dequeued FIN."));
+    if (TCPH_FLAGS(cseg->tcphdr) & TCP_FIN) {
+      LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_receive: dequeued FIN.\n"));
       recv_flags = TF_GOT_FIN;
     }
 
@@ -1103,7 +1156,7 @@ tcp_receive(struct tcp_pcb *pcb)
     }
   }
 }
-/*-----------------------------------------------------------------------------------*/
+
 /*
  * tcp_parseopt:
  *
@@ -1111,7 +1164,7 @@ tcp_receive(struct tcp_pcb *pcb)
  * from uIP with only small changes.)
  *
  */
-/*-----------------------------------------------------------------------------------*/
+
 static void
 tcp_parseopt(struct tcp_pcb *pcb)
 {
@@ -1122,8 +1175,8 @@ tcp_parseopt(struct tcp_pcb *pcb)
   opts = (u8_t *)tcphdr + TCP_HLEN;
 
   /* Parse the TCP MSS option, if present. */
-  if ((TCPH_OFFSET(tcphdr) & 0xf0) > 0x50) {
-    for(c = 0; c < ((TCPH_OFFSET(tcphdr) >> 4) - 5) << 2 ;) {
+  if(TCPH_HDRLEN(tcphdr) > 0x5) {
+    for(c = 0; c < (TCPH_HDRLEN(tcphdr) - 5) << 2 ;) {
       opt = opts[c];
       if (opt == 0x00) {
         /* End of options. */
@@ -1153,5 +1206,5 @@ tcp_parseopt(struct tcp_pcb *pcb)
   }
 }
 #endif /* LWIP_TCP */
-/*-----------------------------------------------------------------------------------*/
+
 

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"
@@ -63,15 +59,21 @@
 /* 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)
 {
   return tcp_enqueue(pcb, NULL, 0, flags, 1, NULL, 0);
 
 }
-/*-----------------------------------------------------------------------------------*/
+
+/*
+ * 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)
 {
@@ -90,7 +92,7 @@ tcp_write(struct tcp_pcb *pcb, const void *arg, u16_t len, u8_t copy)
     return ERR_CONN;
   }
 }
-/*-----------------------------------------------------------------------------------*/
+
 err_t
 tcp_enqueue(struct tcp_pcb *pcb, void *arg, u16_t len,
       u8_t flags, u8_t copy,
@@ -133,7 +135,7 @@ tcp_enqueue(struct tcp_pcb *pcb, void *arg, u16_t len,
     pcb->unsent != NULL);
   }
 
-  seg = NULL;
+  seg = useg = NULL;
   seglen = 0;
 
   /* First, break up the data into segments and tuck them together in
@@ -154,12 +156,13 @@ tcp_enqueue(struct tcp_pcb *pcb, void *arg, u16_t len,
     seg->p = NULL;
 
     if (queue == NULL) {
-      queue = seg;
+      useg = queue = seg;
     }
     else {
       /* Attach the segment to the end of the queued segments. */
-      for (useg = queue; useg->next != NULL; useg = useg->next);
+      LWIP_ASSERT("useg != NULL", useg != NULL);
       useg->next = seg;
+      useg = seg;
     }
 
     /* If copy is set, memory should be allocated
@@ -210,9 +213,8 @@ tcp_enqueue(struct tcp_pcb *pcb, void *arg, u16_t len,
       }
       ++queuelen;
 
-      /* Chain the headers and data pbufs together. */
-      pbuf_chain(seg->p, p);
-      pbuf_free(p);
+      /* Concatenate the headers and data pbufs together. */
+      pbuf_cat(seg->p, p);
       p = NULL;
     }
 
@@ -234,9 +236,7 @@ tcp_enqueue(struct tcp_pcb *pcb, void *arg, u16_t len,
 
       LWIP_DEBUGF(TCP_OUTPUT_DEBUG | 2, ("tcp_enqueue: no room for TCP header in pbuf.\n"));
 
-#ifdef TCP_STATS
-      ++lwip_stats.tcp.err;
-#endif /* TCP_STATS */
+      TCP_STATS_INC(tcp.err);
       goto memerr;
     }
     seg->tcphdr = seg->p->payload;
@@ -249,10 +249,10 @@ tcp_enqueue(struct tcp_pcb *pcb, void *arg, u16_t len,
 
     /* Copy the options into the header, if they are present. */
     if (optdata == NULL) {
-      TCPH_OFFSET_SET(seg->tcphdr, 5 << 4);
+      TCPH_HDRLEN_SET(seg->tcphdr, 5);
     }
     else {
-      TCPH_OFFSET_SET(seg->tcphdr, (5 + optlen / 4) << 4);
+      TCPH_HDRLEN_SET(seg->tcphdr, (5 + optlen / 4));
       /* Copy options into data portion of segment.
        Options can thus only be sent in non data carrying
        segments such as SYN|ACK. */
@@ -288,11 +288,7 @@ tcp_enqueue(struct tcp_pcb *pcb, void *arg, u16_t len,
     useg->len + queue->len <= pcb->mss) {
     /* Remove TCP header from first segment. */
     pbuf_header(queue->p, -TCP_HLEN);
-    pbuf_chain(useg->p, queue->p);
-    /* Free buffer which was merged. Note that the previous pbuf_chain call
-     * will have incremented the ref count, so here the ref count will still
-     * be 1 for the 1 pointer still being used on this buffer. */
-    pbuf_free(queue->p);
+    pbuf_cat(useg->p, queue->p);
     useg->len += queue->len;
     useg->next = queue->next;
 
@@ -327,14 +323,12 @@ tcp_enqueue(struct tcp_pcb *pcb, void *arg, u16_t len,
   /* Set the PSH flag in the last segment that we enqueued, but only
   if the segment has data (indicated by seglen > 0). */
   if (seg != NULL && seglen > 0 && seg->tcphdr != NULL) {
-    TCPH_FLAGS_SET(seg->tcphdr, TCPH_FLAGS(seg->tcphdr) | TCP_PSH);
+    TCPH_SET_FLAG(seg->tcphdr, TCP_PSH);
   }
 
   return ERR_OK;
   memerr:
-#ifdef TCP_STATS
-  ++lwip_stats.tcp.memerr;
-#endif /* TCP_STATS */
+  TCP_STATS_INC(tcp.memerr);
 
   if (queue != NULL) {
     tcp_segs_free(queue);
@@ -347,7 +341,7 @@ tcp_enqueue(struct tcp_pcb *pcb, void *arg, u16_t len,
   LWIP_DEBUGF(TCP_QLEN_DEBUG | DBG_STATE, ("tcp_enqueue: %d (with mem err)\n", pcb->snd_queuelen));
   return ERR_MEM;
 }
-/*-----------------------------------------------------------------------------------*/
+
 /* find out what we can send and send it */
 err_t
 tcp_output(struct tcp_pcb *pcb)
@@ -373,6 +367,13 @@ tcp_output(struct tcp_pcb *pcb)
 
   seg = pcb->unsent;
 
+  /* useg should point to last segment on unacked queue */
+  useg = pcb->unacked;
+  if (useg != NULL) {
+    for (; useg->next != NULL; useg = useg->next);
+  }                                                                             
+
+   
   /* If the TF_ACK_NOW flag is set, we check if there is data that is
      to be sent. If data is to be sent out, we'll just piggyback our
      acknowledgement with the outgoing segment. If no data will be
@@ -398,13 +399,15 @@ tcp_output(struct tcp_pcb *pcb)
     TCPH_FLAGS_SET(tcphdr, TCP_ACK);
     tcphdr->wnd = htons(pcb->rcv_wnd);
     tcphdr->urgp = 0;
-    TCPH_OFFSET_SET(tcphdr, 5 << 4);
+    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), TCP_TTL,
+    ip_output(p, &(pcb->local_ip), &(pcb->remote_ip), pcb->ttl, pcb->tos,
         IP_PROTO_TCP);
     pbuf_free(p);
 
@@ -443,7 +446,7 @@ tcp_output(struct tcp_pcb *pcb)
     pcb->unsent = seg->next;
 
     if (pcb->state != SYN_SENT) {
-      TCPH_FLAGS_SET(seg->tcphdr, TCPH_FLAGS(seg->tcphdr) | TCP_ACK);
+      TCPH_SET_FLAG(seg->tcphdr, TCP_ACK);
       pcb->flags &= ~(TF_ACK_DELAY | TF_ACK_NOW);
     }
 
@@ -457,11 +460,10 @@ tcp_output(struct tcp_pcb *pcb)
       seg->next = NULL;
       if (pcb->unacked == NULL) {
         pcb->unacked = seg;
-
-
+        useg = seg;
       } else {
-        for (useg = pcb->unacked; useg->next != NULL; useg = useg->next);
         useg->next = seg;
+        useg = useg->next;
       }
     } else {
       tcp_seg_free(seg);
@@ -470,7 +472,7 @@ tcp_output(struct tcp_pcb *pcb)
   }
   return ERR_OK;
 }
-/*-----------------------------------------------------------------------------------*/
+
 static void
 tcp_output_segment(struct tcp_seg *seg, struct tcp_pcb *pcb)
 {
@@ -519,18 +521,18 @@ 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);
-#ifdef TCP_STATS
-  ++lwip_stats.tcp.xmit;
-#endif /* TCP_STATS */
+#endif
+  TCP_STATS_INC(tcp.xmit);
 
-  ip_output(seg->p, &(pcb->local_ip), &(pcb->remote_ip), TCP_TTL,
+  ip_output(seg->p, &(pcb->local_ip), &(pcb->remote_ip), pcb->ttl, pcb->tos,
       IP_PROTO_TCP);
 }
-/*-----------------------------------------------------------------------------------*/
+
 void
 tcp_rst(u32_t seqno, u32_t ackno,
   struct ip_addr *local_ip, struct ip_addr *remote_ip,
@@ -552,20 +554,20 @@ tcp_rst(u32_t seqno, u32_t ackno,
   TCPH_FLAGS_SET(tcphdr, TCP_RST | TCP_ACK);
   tcphdr->wnd = htons(TCP_WND);
   tcphdr->urgp = 0;
-  TCPH_OFFSET_SET(tcphdr, 5 << 4);
+  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);
-
-#ifdef TCP_STATS
-  ++lwip_stats.tcp.xmit;
-#endif /* TCP_STATS */
-  ip_output(p, local_ip, remote_ip, TCP_TTL, IP_PROTO_TCP);
+#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);
   pbuf_free(p);
   LWIP_DEBUGF(TCP_RST_DEBUG, ("tcp_rst: seqno %lu ackno %lu.\n", seqno, ackno));
 }
-/*-----------------------------------------------------------------------------------*/
+
 void
 tcp_rexmit(struct tcp_pcb *pcb)
 {
@@ -595,6 +597,49 @@ tcp_rexmit(struct tcp_pcb *pcb)
   tcp_output(pcb);
 
 }
+
+void
+tcp_keepalive(struct tcp_pcb *pcb)
+{
+   struct pbuf *p;
+   struct tcp_hdr *tcphdr;
+
+   LWIP_DEBUGF(TCP_DEBUG, ("tcp_keepalive: sending KEEPALIVE probe to %u.%u.%u.%u\n",
+                           ip4_addr1(&pcb->remote_ip), ip4_addr2(&pcb->remote_ip),
+                           ip4_addr3(&pcb->remote_ip), ip4_addr4(&pcb->remote_ip)));
+
+   LWIP_DEBUGF(TCP_DEBUG, ("tcp_keepalive: tcp_ticks %ld   pcb->tmr %ld  pcb->keep_cnt %ld\n", tcp_ticks, pcb->tmr, pcb->keep_cnt));
+   
+   p = pbuf_alloc(PBUF_IP, TCP_HLEN, PBUF_RAM);
+
+   if(p == NULL) {
+      LWIP_DEBUGF(TCP_DEBUG, ("tcp_keepalive: could not allocate memory for pbuf\n"));
+      return;
+   }
+
+   tcphdr = p->payload;
+   tcphdr->src = htons(pcb->local_port);
+   tcphdr->dest = htons(pcb->remote_port);
+   tcphdr->seqno = htonl(pcb->snd_nxt - 1);
+   tcphdr->ackno = htonl(pcb->rcv_nxt);
+   tcphdr->wnd = htons(pcb->rcv_wnd);
+   tcphdr->urgp = 0;
+   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 */
+  ip_output(p, &pcb->local_ip, &pcb->remote_ip, pcb->ttl, 0, IP_PROTO_TCP);
+
+  pbuf_free(p);
+
+  LWIP_DEBUGF(TCP_RST_DEBUG, ("tcp_keepalive: seqno %lu ackno %lu.\n", pcb->snd_nxt - 1, pcb->rcv_nxt));
+}
+
 #endif /* LWIP_TCP */
 
 

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,
@@ -35,22 +35,22 @@
  *
  */
 
-/*-----------------------------------------------------------------------------------*/
+
 /* udp.c
  *
  * The code for the User Datagram Protocol UDP.
  *
  */
-/*-----------------------------------------------------------------------------------*/
+
 #include "lwip/opt.h"
 
 #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"
 
@@ -64,87 +64,13 @@ struct udp_pcb *udp_pcbs = NULL;
 
 static struct udp_pcb *pcb_cache = NULL;
 
-#if UDP_DEBUG
-int udp_debug_print(struct udp_hdr *udphdr);
-#endif /* UDP_DEBUG */
 
-/*-----------------------------------------------------------------------------------*/
 void
 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.
  *
@@ -163,21 +89,24 @@ udp_input(struct pbuf *p, struct netif *inp)
   struct ip_hdr *iphdr;
   u16_t src, dest;
 
+#if SO_REUSE
+  struct udp_pcb *pcb_temp;
+  int reuse = 0;
+  int reuse_port_1 = 0;
+  int reuse_port_2 = 0;
+#endif /* SO_REUSE */
+  
   PERF_START;
 
-#ifdef UDP_STATS
-  ++lwip_stats.udp.recv;
-#endif /* UDP_STATS */
+  UDP_STATS_INC(udp.recv);
 
   iphdr = p->payload;
 
   if (pbuf_header(p, -((s16_t)(UDP_HLEN + IPH_HL(iphdr) * 4)))) {
     /* drop short packets */
     LWIP_DEBUGF(UDP_DEBUG, ("udp_input: short UDP datagram (%u bytes) discarded\n", p->tot_len));
-#ifdef UDP_STATS
-    ++lwip_stats.udp.lenerr;
-    ++lwip_stats.udp.drop;
-#endif /* UDP_STATS */
+    UDP_STATS_INC(udp.lenerr);
+    UDP_STATS_INC(udp.drop);
     snmp_inc_udpinerrors();
     pbuf_free(p);
     goto end;
@@ -190,9 +119,7 @@ udp_input(struct pbuf *p, struct netif *inp)
   src = ntohs(udphdr->src);
   dest = ntohs(udphdr->dest);
 
-#if UDP_DEBUG
   udp_debug_print(udphdr);
-#endif /* UDP_DEBUG */
 
   /* print the UDP source and destination */
   LWIP_DEBUGF(UDP_DEBUG, ("udp (%u.%u.%u.%u, %u) <-- (%u.%u.%u.%u, %u)\n",
@@ -200,8 +127,18 @@ udp_input(struct pbuf *p, struct netif *inp)
     ip4_addr3(&iphdr->dest), ip4_addr4(&iphdr->dest), ntohs(udphdr->dest),
     ip4_addr1(&iphdr->src), ip4_addr2(&iphdr->src),
     ip4_addr3(&iphdr->src), ip4_addr4(&iphdr->src), ntohs(udphdr->src)));
+
+#if SO_REUSE
+  pcb_temp = udp_pcbs;
+  
+ again_1:
+  
+  /* Iterate through the UDP pcb list for a fully matching pcb */
+  for(pcb = pcb_temp; pcb != NULL; pcb = pcb->next) {
+#else  /* SO_REUSE */ 
   /* Iterate through the UDP pcb list for a fully matching pcb */
   for(pcb = udp_pcbs; pcb != NULL; pcb = pcb->next) {
+#endif  /* SO_REUSE */ 
     /* print the PCB local and remote address */
     LWIP_DEBUGF(UDP_DEBUG, ("pcb (%u.%u.%u.%u, %u) --- (%u.%u.%u.%u, %u)\n",
       ip4_addr1(&pcb->local_ip), ip4_addr2(&pcb->local_ip),
@@ -221,6 +158,27 @@ 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)))) {
+#if SO_REUSE
+      if(pcb->so_options & SOF_REUSEPORT) {
+        if(reuse) {
+          /* We processed one PCB already */
+          LWIP_DEBUGF(UDP_DEBUG, ("udp_input: second or later PCB and SOF_REUSEPORT set.\n"));
+        } else {
+          /* First PCB with this address */
+          LWIP_DEBUGF(UDP_DEBUG, ("udp_input: first PCB and SOF_REUSEPORT set.\n"));
+          reuse = 1;
+        }
+        
+        reuse_port_1 = 1; 
+        p->ref++;
+        LWIP_DEBUGF(UDP_DEBUG, ("udp_input: reference counter on PBUF set to %i\n", p->ref));
+      } else {
+        if(reuse) {
+          /* We processed one PCB already */
+          LWIP_DEBUGF(UDP_DEBUG, ("udp_input: second or later PCB but SOF_REUSEPORT not set !\n"));
+        }
+      }
+#endif /* SO_REUSE */
       break;
     }
   }
@@ -228,7 +186,16 @@ udp_input(struct pbuf *p, struct netif *inp)
   if (pcb == NULL) {
     /* Iterate through the UDP PCB list for a pcb that matches
        the local address. */
+
+#if SO_REUSE
+    pcb_temp = udp_pcbs;
+    
+  again_2:
+
+    for(pcb = pcb_temp; pcb != NULL; pcb = pcb->next) {
+#else  /* SO_REUSE */ 
     for(pcb = udp_pcbs; pcb != NULL; pcb = pcb->next) {
+#endif  /* SO_REUSE */ 
       LWIP_DEBUGF(UDP_DEBUG, ("pcb (%u.%u.%u.%u, %u) --- (%u.%u.%u.%u, %u)\n",
         ip4_addr1(&pcb->local_ip), ip4_addr2(&pcb->local_ip),
         ip4_addr3(&pcb->local_ip), ip4_addr4(&pcb->local_ip), pcb->local_port,
@@ -242,7 +209,28 @@ 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)))) {
-         break;
+#if SO_REUSE
+        if(pcb->so_options & SOF_REUSEPORT) {
+          if(reuse) {
+            /* We processed one PCB already */
+            LWIP_DEBUGF(UDP_DEBUG, ("udp_input: second or later PCB and SOF_REUSEPORT set.\n"));
+          } else {
+            /* First PCB with this address */
+            LWIP_DEBUGF(UDP_DEBUG, ("udp_input: first PCB and SOF_REUSEPORT set.\n"));
+            reuse = 1;
+          }
+          
+          reuse_port_2 = 1; 
+          p->ref++;
+          LWIP_DEBUGF(UDP_DEBUG, ("udp_input: reference counter on PBUF set to %i\n", p->ref));
+        } else {
+          if(reuse) {
+            /* We processed one PCB already */
+            LWIP_DEBUGF(UDP_DEBUG, ("udp_input: second or later PCB but SOF_REUSEPORT not set !\n"));
+          }
+        }
+#endif /* SO_REUSE */
+        break;
       }
     }
   }
@@ -258,56 +246,80 @@ 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) {
   LWIP_DEBUGF(UDP_DEBUG | 2, ("udp_input: UDP Lite datagram discarded due to failing checksum\n"));
-#ifdef UDP_STATS
-  ++lwip_stats.udp.chkerr;
-  ++lwip_stats.udp.drop;
-#endif /* UDP_STATS */
+  UDP_STATS_INC(udp.chkerr);
+  UDP_STATS_INC(udp.drop);
   snmp_inc_udpinerrors();
   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),
         IP_PROTO_UDP, p->tot_len) != 0) {
     LWIP_DEBUGF(UDP_DEBUG | 2, ("udp_input: UDP datagram discarded due to failing checksum\n"));
 
-#ifdef UDP_STATS
-    ++lwip_stats.udp.chkerr;
-    ++lwip_stats.udp.drop;
-#endif /* UDP_STATS */
+    UDP_STATS_INC(udp.chkerr);
+    UDP_STATS_INC(udp.drop);
     snmp_inc_udpinerrors();
     pbuf_free(p);
     goto end;
   }
       }
+#endif
     }
     pbuf_header(p, -UDP_HLEN);
     if (pcb != NULL) {
       snmp_inc_udpindatagrams();
       pcb->recv(pcb->recv_arg, pcb, p, &(iphdr->src), src);
+#if SO_REUSE
+      /* First socket should receive now */
+      if(reuse_port_1 || reuse_port_2) {
+        /* We want to search on next socket after receiving */
+        pcb_temp = pcb->next;
+        
+        if(reuse_port_1) {
+          /* We are searching connected sockets */
+          reuse_port_1 = 0;
+          reuse_port_2 = 0;
+          goto again_1;
+        } else {
+          /* We are searching unconnected sockets */
+          reuse_port_1 = 0;
+          reuse_port_2 = 0;
+          goto again_2;
+        }
+      }
+#endif /* SO_REUSE */ 
     } else {
+#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);
+        goto end;
+      }
+#endif /* SO_REUSE */
       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;
   icmp_dest_unreach(p, ICMP_DUR_PORT);
       }
-#ifdef UDP_STATS
-      ++lwip_stats.udp.proterr;
-      ++lwip_stats.udp.drop;
-#endif /* UDP_STATS */
+      UDP_STATS_INC(udp.proterr);
+      UDP_STATS_INC(udp.drop);
     snmp_inc_udpnoports();
       pbuf_free(p);
     }
@@ -318,6 +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.
  *
@@ -327,10 +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)
@@ -355,7 +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) {
@@ -367,25 +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 } */
   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));
-#ifdef UDP_STATS
-    ++lwip_stats.udp.rterr;
-#endif /* UDP_STATS */
+    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);
@@ -402,38 +453,47 @@ 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 */
-    err = ip_output_if (q, src_ip, &pcb->remote_ip, UDP_TTL, IP_PROTO_UDPLITE, netif);
-    snmp_inc_udpoutdatagrams();
+    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);    
+  /* 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, UDP_TTL, IP_PROTO_UDP, netif);
+    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 */
-    pbuf_free(q);
+    /* free the header pbuf */
+    pbuf_free(q); q = NULL;
+    /* { p is still referenced by the caller, and will live on } */
   }
 
-#ifdef UDP_STATS
-  ++lwip_stats.udp.xmit;
-#endif /* UDP_STATS */
+  UDP_STATS_INC(udp.xmit);
   return err;
 }
 
@@ -457,9 +517,13 @@ udp_bind(struct udp_pcb *pcb, struct ip_addr *ipaddr, u16_t port)
 {
   struct udp_pcb *ipcb;
   u8_t rebind;
+#if SO_REUSE
+  int reuse_port_all_set = 1;
+#endif /* SO_REUSE */
   LWIP_DEBUGF(UDP_DEBUG | DBG_TRACE | 3, ("udp_bind(ipaddr = "));
   ip_addr_debug_print(UDP_DEBUG, ipaddr);
   LWIP_DEBUGF(UDP_DEBUG | DBG_TRACE | 3, (", port = %u)\n", port));
+
   rebind = 0;
   /* Check for double bind and rebind of the same pcb */
   for (ipcb = udp_pcbs; ipcb != NULL; ipcb = ipcb->next) {
@@ -470,11 +534,13 @@ udp_bind(struct udp_pcb *pcb, struct ip_addr *ipaddr, u16_t port)
       /* pcb already in list, just rebind */
       rebind = 1;
     }
+
+#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
    combine with implementation of UDP PCB flags. Leon Woestenberg. */
-#if 0
+#ifdef LWIP_UDP_TODO
     /* port matches that of PCB in list? */
     else if ((ipcb->local_port == port) &&
        /* IP address matches, or one is IP_ADDR_ANY? */
@@ -486,8 +552,52 @@ udp_bind(struct udp_pcb *pcb, struct ip_addr *ipaddr, u16_t port)
       return ERR_USE;
     }
 #endif
+
+#else /* SO_REUSE */
+      /* Search through list of PCB's. 
+         
+      If there is a PCB bound to specified port and IP_ADDR_ANY another PCB can be bound to the interface IP
+      or to the loopback address on the same port if SOF_REUSEADDR is set. Any combination of PCB's bound to 
+      the same local port, but to one address out of {IP_ADDR_ANY, 127.0.0.1, interface IP} at a time is valid.
+      But no two PCB's bound to same local port and same local address is valid.
+      
+      If SOF_REUSEPORT is set several PCB's can be bound to same local port and same local address also. But then 
+      all PCB's must have the SOF_REUSEPORT option set.
+      
+      When the two options aren't set and specified port is already bound, ERR_USE is returned saying that 
+      address is already in use. */
+    else if (ipcb->local_port == port) {
+      if(ip_addr_cmp(&(ipcb->local_ip), ipaddr)) {
+        if(pcb->so_options & SOF_REUSEPORT) {
+          LWIP_DEBUGF(UDP_DEBUG, ("udp_bind: in UDP PCB's SO_REUSEPORT set and same address.\n"));
+          reuse_port_all_set = (reuse_port_all_set && (ipcb->so_options & SOF_REUSEPORT));
+        }
+        else {
+          LWIP_DEBUGF(UDP_DEBUG, ("udp_bind: in UDP PCB's SO_REUSEPORT not set and same address.\n"));
+          return ERR_USE;
+        }
+      }
+      else if((ip_addr_isany(ipaddr) && !ip_addr_isany(&(ipcb->local_ip))) ||
+              (!ip_addr_isany(ipaddr) && ip_addr_isany(&(ipcb->local_ip)))) {
+        if(!(pcb->so_options & SOF_REUSEADDR) && !(pcb->so_options & SOF_REUSEPORT)) {
+          LWIP_DEBUGF(UDP_DEBUG, ("udp_bind: in UDP PCB's SO_REUSEPORT or SO_REUSEADDR not set and not the same address.\n"));
+          return ERR_USE;
+        }           
+      }
+    }
+#endif /* SO_REUSE */
+
   }
-  /* bind local address */
+
+#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) {
+    LWIP_DEBUGF(UDP_DEBUG, ("udp_bind: not all sockets have SO_REUSEPORT set.\n"));
+    return ERR_USE;
+  }
+#endif /* SO_REUSE */
+
   ip_addr_set(&pcb->local_ip, ipaddr);
   /* no port specified? */
   if (port == 0) {
@@ -552,16 +662,14 @@ udp_connect(struct udp_pcb *pcb, struct ip_addr *ipaddr, u16_t port)
   pcb->remote_port = port;
   pcb->flags |= UDP_FLAGS_CONNECTED;
 /** TODO: this functionality belongs in upper layers */
-#if 0
+#ifdef LWIP_UDP_TODO
   /* Nail down local IP for netconn_addr()/getsockname() */
   if (ip_addr_isany(&pcb->local_ip) && !ip_addr_isany(&pcb->remote_ip)) {
     struct netif *netif;
 
     if ((netif = ip_route(&(pcb->remote_ip))) == NULL) {
       LWIP_DEBUGF(UDP_DEBUG, ("udp_connect: No route to 0x%lx\n", pcb->remote_ip.addr));
-#ifdef UDP_STATS
-        ++lwip_stats.udp.rterr;
-#endif /* UDP_STATS */
+        UDP_STATS_INC(udp.rterr);
       return ERR_RTE;
     }
     /** TODO: this will bind the udp pcb locally, to the interface which
@@ -594,9 +702,13 @@ 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;
 }
-/*-----------------------------------------------------------------------------------*/
+
 void
 udp_recv(struct udp_pcb *pcb,
    void (* recv)(void *arg, struct udp_pcb *upcb, struct pbuf *p,
@@ -649,10 +761,13 @@ udp_new(void) {
   if (pcb != NULL) {
     /* initialize PCB to all zeroes */
     memset(pcb, 0, sizeof(struct udp_pcb));
+    pcb->ttl = UDP_TTL;
   }
+  
+  
   return pcb;
 }
-/*-----------------------------------------------------------------------------------*/
+
 #if UDP_DEBUG
 int
 udp_debug_print(struct udp_hdr *udphdr)
@@ -668,7 +783,7 @@ udp_debug_print(struct udp_hdr *udphdr)
   return 0;
 }
 #endif /* UDP_DEBUG */
-/*-----------------------------------------------------------------------------------*/
+
 #endif /* LWIP_UDP */
 
 

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
@@ -67,6 +67,13 @@ u8_t *inet_ntoa(u32_t addr); /* returns ptr to static buffer; not reentrant! */
 #define htonl(x) (x)
 #define ntohl(x) (x)
 #else
+#ifdef LWIP_PREFIX_BYTEORDER_FUNCS
+/* workaround for naming collisions on some platforms */
+#define htons lwip_htons
+#define ntohs lwip_ntohs
+#define htonl lwip_htonl
+#define ntohl lwip_ntohl
+#endif
 u16_t htons(u16_t x);
 u16_t ntohs(u16_t x);
 u32_t htonl(u32_t x);

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, 
@@ -43,13 +43,12 @@
 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,
-    u8_t ttl, u8_t proto);
+		u8_t ttl, u8_t tos, u8_t proto);
 err_t ip_output_if(struct pbuf *p, struct ip_addr *src, struct ip_addr *dest,
-       u8_t ttl, u8_t proto,
+		   u8_t ttl, u8_t tos, u8_t proto,
        struct netif *netif);
 
 #define IP_HLEN 20
@@ -67,6 +66,36 @@ err_t ip_output_if(struct pbuf *p, struct ip_addr *src, struct ip_addr *dest,
 #endif /* IP_HDRINCL */
 #define IP_HDRINCL  NULL
 
+
+/* This is the common part of all PCB types. It needs to be at the
+   beginning of a PCB type definition. It is located here so that
+   changes to this common part are made in one location instead of
+   having to change all PCB structs. */
+#define IP_PCB struct ip_addr local_ip; \
+  struct ip_addr remote_ip; \
+   /* Socket options */  \
+  u16_t so_options;      \
+   /* Type Of Service */ \
+  u8_t tos;              \
+  /* Time To Live */     \
+  u8_t ttl
+
+/*
+ * Option flags per-socket. These are the same like SO_XXX.
+ */
+#define	SOF_DEBUG	    (u16_t)0x0001U		/* turn on debugging info recording */
+#define	SOF_ACCEPTCONN	(u16_t)0x0002U		/* socket has had listen() */
+#define	SOF_REUSEADDR	(u16_t)0x0004U		/* allow local address reuse */
+#define	SOF_KEEPALIVE	(u16_t)0x0008U		/* keep connections alive */
+#define	SOF_DONTROUTE	(u16_t)0x0010U		/* just use interface addresses */
+#define	SOF_BROADCAST	(u16_t)0x0020U		/* permit sending of broadcast msgs */
+#define	SOF_USELOOPBACK	(u16_t)0x0040U		/* bypass hardware when possible */
+#define	SOF_LINGER	    (u16_t)0x0080U		/* linger on close if data present */
+#define	SOF_OOBINLINE	(u16_t)0x0100U		/* leave received OOB data in line */
+#define	SOF_REUSEPORT	(u16_t)0x0200U		/* allow local address & port reuse */
+
+
+
 #ifdef PACK_STRUCT_USE_INCLUDES
 #  include "arch/bpstruct.h"
 #endif
@@ -117,6 +146,8 @@ PACK_STRUCT_END
 
 #if IP_DEBUG
 void ip_debug_print(struct pbuf *p);
+#else
+#define ip_debug_print(p)
 #endif /* IP_DEBUG */
 
 #endif /* __LWIP_IP_H__ */

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, 
@@ -38,6 +38,7 @@
 
 #include "lwip/ip.h"
 
+#include "lwip/raw.h"
 #include "lwip/udp.h"
 #include "lwip/tcp.h"
 
@@ -50,7 +51,8 @@ enum netconn_type {
   NETCONN_TCP,
   NETCONN_UDP,
   NETCONN_UDPLITE,
-  NETCONN_UDPNOCHKSUM
+  NETCONN_UDPNOCHKSUM,
+  NETCONN_RAW
 };
 
 enum netconn_state {
@@ -82,6 +84,7 @@ struct netconn {
   union {
     struct tcp_pcb *tcp;
     struct udp_pcb *udp;
+    struct raw_pcb *raw;
   } pcb;
   err_t err;
   sys_mbox_t mbox;
@@ -121,6 +124,9 @@ struct netconn *  netconn_new     (enum netconn_type type);
 struct
 netconn *netconn_new_with_callback(enum netconn_type t,
                                    void (*callback)(struct netconn *, enum netconn_evt, u16_t len));
+struct
+netconn *netconn_new_with_proto_and_callback(enum netconn_type t, u16_t proto,
+                                   void (*callback)(struct netconn *, enum netconn_evt, u16_t len));
 err_t             netconn_delete  (struct netconn *conn);
 enum netconn_type netconn_type    (struct netconn *conn);
 err_t             netconn_peer    (struct netconn *conn,

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, 
@@ -61,19 +61,19 @@
 /** flag for LWIP_DEBUGF to halt after printing this debug message */
 #define DBG_HALT    0x08U
 
-#ifdef LWIP_DEBUG
-# ifndef LWIP_NOASSERT
+#ifndef LWIP_NOASSERT
 #  define LWIP_ASSERT(x,y) do { if(!(y)) LWIP_PLATFORM_ASSERT(x); } while(0)
-# else
+#else
 #  define LWIP_ASSERT(x,y) 
-# endif
+#endif
+
+#ifdef LWIP_DEBUG
 /** print debug message only if debug message type is enabled...
  *  AND is of correct type AND is at least DBG_LEVEL
  */
 #  define LWIP_DEBUGF(debug,x) do { if (((debug) & DBG_ON) && ((debug) & DBG_TYPES_ON) && (((debug) & DBG_MASK_LEVEL) >= DBG_MIN_LEVEL)) { LWIP_PLATFORM_DIAG(x); if ((debug) & DBG_HALT) while(1); } } while(0)
 #  define LWIP_ERROR(x)   do { LWIP_PLATFORM_DIAG(x); } while(0)  
 #else /* LWIP_DEBUG */
-#  define LWIP_ASSERT(x,y) 
 #  define LWIP_DEBUGF(debug,x) 
 #  define LWIP_ERROR(x)  
 #endif /* LWIP_DEBUG */

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
@@ -191,10 +191,10 @@ void dhcp_fine_tmr(void);
 #define DHCP_OPTION_MESSAGE_TYPE_LEN 1
 
 
-#define DHCP_OPTION_SERVER_ID 54 /* RFC 2131 9.7, server IP address */
-#define DHCP_OPTION_PARAMETER_REQUEST_LIST 55 /* RFC 2131 9.8, requested option types */
+#define DHCP_OPTION_SERVER_ID 54 /* RFC 2132 9.7, server IP address */
+#define DHCP_OPTION_PARAMETER_REQUEST_LIST 55 /* RFC 2132 9.8, requested option types */
 
-#define DHCP_OPTION_MAX_MSG_SIZE 57 /* RFC 2131 9.10, message size accepted >= 576 */
+#define DHCP_OPTION_MAX_MSG_SIZE 57 /* RFC 2132 9.10, message size accepted >= 576 */
 #define DHCP_OPTION_MAX_MSG_SIZE_LEN 2
 
 #define DHCP_OPTION_T1 58 /* T1 renewal time */

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, 
@@ -49,9 +49,13 @@ void mem_free(void *mem);
 void *mem_realloc(void *mem, mem_size_t size);
 void *mem_reallocm(void *mem, mem_size_t size);
 
+#ifndef MEM_ALIGN_SIZE
 #define MEM_ALIGN_SIZE(size) (((size) + MEM_ALIGNMENT - 1) & ~(MEM_ALIGNMENT-1))
+#endif
 
+#ifndef MEM_ALIGN
 #define MEM_ALIGN(addr) ((void *)(((mem_ptr_t)(addr) + MEM_ALIGNMENT - 1) & ~(mem_ptr_t)(MEM_ALIGNMENT-1)))
+#endif
 
 #endif /* __LWIP_MEM_H__ */
 

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, 
@@ -37,6 +37,7 @@
 
 typedef enum {
   MEMP_PBUF,
+  MEMP_RAW_PCB,
   MEMP_UDP_PCB,
   MEMP_TCP_PCB,
   MEMP_TCP_PCB_LISTEN,
@@ -58,8 +59,5 @@ void *memp_malloc(memp_t type);
 void *memp_realloc(memp_t fromtype, memp_t totype, void *mem);
 void memp_free(memp_t type, void *mem);
 
-void *memp_mallocp(memp_t type);
-void memp_freep(memp_t type, void *mem);
-
 #endif /* __LWIP_MEMP_H__  */
     

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. */
@@ -115,7 +119,7 @@ extern struct netif *netif_default;
 /* netif_init() must be called first. */
 void netif_init(void);
 
-struct netif *netif_add(struct ip_addr *ipaddr, struct ip_addr *netmask,
+struct netif *netif_add(struct netif *netif, struct ip_addr *ipaddr, struct ip_addr *netmask,
       struct ip_addr *gw,
       void *state,
       err_t (* init)(struct netif *netif),

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, 
@@ -32,9 +32,9 @@
 #ifndef __LWIP_OPT_H__
 #define __LWIP_OPT_H__
 
-#include "lwip/debug.h"
 /* Include user defined options first */
 #include "lwipopts.h"
+#include "lwip/debug.h"
 
 /* Define default values for unconfigured parameters. */
 
@@ -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. */
@@ -74,6 +78,11 @@ a lot of data that needs to be copied, this should be set high. */
 #define MEMP_NUM_PBUF                   16
 #endif
 
+/* Number of raw connection PCBs */
+#ifndef MEMP_NUM_RAW_PCB
+#define MEMP_NUM_RAW_PCB                4
+#endif
+
 /* MEMP_NUM_UDP_PCB: the number of UDP protocol control blocks. One
    per active UDP "connection". */
 #ifndef MEMP_NUM_UDP_PCB
@@ -166,14 +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 ---------- */
@@ -211,6 +216,16 @@ a lot of data that needs to be copied, this should be set high. */
 #define ICMP_TTL                        255
 #endif
 
+/* ---------- RAW options ---------- */
+
+#ifndef LWIP_RAW
+#define LWIP_RAW                        1
+#endif
+
+#ifndef RAW_TTL
+#define RAW_TTL                        255
+#endif
+
 /* ---------- DHCP options ---------- */
 
 #ifndef LWIP_DHCP
@@ -288,8 +303,10 @@ a lot of data that needs to be copied, this should be set high. */
 #define TCP_SNDLOWAT                    TCP_SND_BUF/2
 #endif
 
-
-
+/* Support loop interface (127.0.0.1) */
+#ifndef LWIP_HAVE_LOOPIF
+#define LWIP_HAVE_LOOPIF		1
+#endif
 
 #ifndef LWIP_EVENT_API
 #define LWIP_EVENT_API                  0
@@ -320,6 +337,14 @@ a lot of data that needs to be copied, this should be set high. */
 #define DEFAULT_THREAD_PRIO             1
 #endif
 
+
+/* ---------- Socket Options ---------- */
+/* Enable SO_REUSEADDR and SO_REUSEPORT options */ 
+#ifndef SO_REUSE
+# define SO_REUSE 1
+#endif                                                                        
+
+
 /* ---------- Statistics options ---------- */
 #ifndef LWIP_STATS
 #define LWIP_STATS                      1
@@ -327,15 +352,68 @@ a lot of data that needs to be copied, this should be set high. */
 
 #if LWIP_STATS
 
-#define LINK_STATS
-#define IP_STATS
-#define ICMP_STATS
-#define UDP_STATS
-#define TCP_STATS
-#define MEM_STATS
-#define MEMP_STATS
-#define PBUF_STATS
-#define SYS_STATS
+#ifndef LWIP_STATS_DISPLAY
+#define LWIP_STATS_DISPLAY 0
+#endif
+
+#ifndef LINK_STATS
+#define LINK_STATS	1
+#endif
+
+#ifndef IP_STATS
+#define IP_STATS	1
+#endif
+
+#ifndef IPFRAG_STATS
+#define IPFRAG_STATS	1
+#endif
+
+#ifndef ICMP_STATS
+#define ICMP_STATS	1
+#endif
+
+#ifndef UDP_STATS
+#define UDP_STATS	1
+#endif
+
+#ifndef TCP_STATS
+#define TCP_STATS	1
+#endif
+
+#ifndef MEM_STATS
+#define MEM_STATS	1
+#endif
+
+#ifndef MEMP_STATS
+#define MEMP_STATS	1
+#endif
+
+#ifndef PBUF_STATS
+#define PBUF_STATS	1
+#endif
+
+#ifndef SYS_STATS
+#define SYS_STATS	1
+#endif
+
+#ifndef RAW_STATS
+#define RAW_STATS	0
+#endif
+
+#else
+
+#define LINK_STATS	0
+#define IP_STATS	0
+#define IPFRAG_STATS	0
+#define ICMP_STATS	0
+#define UDP_STATS	0
+#define TCP_STATS	0
+#define MEM_STATS	0
+#define MEMP_STATS	0
+#define PBUF_STATS	0
+#define SYS_STATS	0
+#define RAW_STATS	0
+#define LWIP_STATS_DISPLAY	0
 
 #endif /* LWIP_STATS */
 
@@ -426,6 +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 */
 
@@ -433,10 +536,6 @@ a lot of data that needs to be copied, this should be set high. */
 #define DBG_TYPES_ON                    0
 #endif
 
-#ifndef DEMO_DEBUG
-#define DEMO_DEBUG                      DBG_OFF
-#endif
-
 #ifndef ETHARP_DEBUG
 #define ETHARP_DEBUG                    DBG_OFF
 #endif
@@ -477,6 +576,10 @@ a lot of data that needs to be copied, this should be set high. */
 #define IP_REASS_DEBUG                  DBG_OFF
 #endif
 
+#ifndef RAW_DEBUG
+#define RAW_DEBUG                       DBG_OFF
+#endif
+
 #ifndef MEM_DEBUG
 #define MEM_DEBUG                       DBG_OFF
 #endif

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, 
@@ -29,7 +29,7 @@
  * Author: Adam Dunkels <adam@sics.se>
  *
  */
-/*-----------------------------------------------------------------------------------*/
+
 #ifndef __LWIP_PBUF_H__
 #define __LWIP_PBUF_H__
 
@@ -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);
@@ -108,8 +103,11 @@ void pbuf_ref(struct pbuf *p);
 void pbuf_ref_chain(struct pbuf *p);
 u8_t pbuf_free(struct pbuf *p);
 u8_t pbuf_clen(struct pbuf *p);  
+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

@@ -0,0 +1,74 @@
+/*
+ * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without modification,
+ * are permitted provided that the following conditions are met:
+ *
+ * 1. Redistributions of source code must retain the above copyright notice,
+ *    this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright notice,
+ *    this list of conditions and the following disclaimer in the documentation
+ *    and/or other materials provided with the distribution.
+ * 3. The name of the author may not be used to endorse or promote products
+ *    derived from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR IMPLIED
+ * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
+ * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT
+ * SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
+ * EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT
+ * OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING
+ * IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY
+ * OF SUCH DAMAGE.
+ *
+ * This file is part of the lwIP TCP/IP stack.
+ *
+ * Author: Adam Dunkels <adam@sics.se>
+ *
+ */
+#ifndef __LWIP_RAW_H__
+#define __LWIP_RAW_H__
+
+#include "lwip/arch.h"
+
+#include "lwip/pbuf.h"
+#include "lwip/inet.h"
+#include "lwip/ip.h"
+
+struct raw_pcb {
+/* Common members of all PCB types */
+  IP_PCB;
+
+  struct raw_pcb *next;
+
+  u16_t protocol;
+
+  u8_t (* recv)(void *arg, struct raw_pcb *pcb, struct pbuf *p,
+    struct ip_addr *addr);
+  void *recv_arg;
+};
+
+/* The following functions is the application layer interface to the
+   RAW code. */
+struct raw_pcb * raw_new        (u16_t proto);
+void             raw_remove     (struct raw_pcb *pcb);
+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,
+                                 u8_t (* recv)(void *arg, struct raw_pcb *pcb,
+                                              struct pbuf *p,
+                                              struct ip_addr *addr),
+                                 void *recv_arg);
+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. */
+u8_t              raw_input      (struct pbuf *p, struct netif *inp);
+void             raw_init       (void);
+
+
+#endif /* __LWIP_RAW_H__ */

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, 
@@ -70,6 +70,7 @@ struct sockaddr {
 #define  SO_USELOOPBACK  0x0040    /* bypass hardware when possible */
 #define  SO_LINGER  0x0080    /* linger on close if data present */
 #define  SO_OOBINLINE  0x0100    /* leave received OOB data in line */
+#define	 SO_REUSEPORT	0x0200		/* allow local address & port reuse */
 
 #define SO_DONTLINGER   (int)(~SO_LINGER)
 
@@ -117,6 +118,36 @@ struct linger {
 #define MSG_DONTWAIT    0x40            /* Nonblocking i/o for this operation only */
 
 
+/*
+ * Options for level IPPROTO_IP
+ */
+#define IP_TOS       1
+#define IP_TTL       2
+
+
+#define IPTOS_TOS_MASK          0x1E
+#define IPTOS_TOS(tos)          ((tos) & IPTOS_TOS_MASK)
+#define IPTOS_LOWDELAY          0x10
+#define IPTOS_THROUGHPUT        0x08
+#define IPTOS_RELIABILITY       0x04
+#define IPTOS_LOWCOST           0x02
+#define IPTOS_MINCOST           IPTOS_LOWCOST
+
+/*
+ * Definitions for IP precedence (also in ip_tos) (hopefully unused)
+ */
+#define IPTOS_PREC_MASK                 0xe0
+#define IPTOS_PREC(tos)                ((tos) & IPTOS_PREC_MASK)
+#define IPTOS_PREC_NETCONTROL           0xe0
+#define IPTOS_PREC_INTERNETCONTROL      0xc0
+#define IPTOS_PREC_CRITIC_ECP           0xa0
+#define IPTOS_PREC_FLASHOVERRIDE        0x80
+#define IPTOS_PREC_FLASH                0x60
+#define IPTOS_PREC_IMMEDIATE            0x40
+#define IPTOS_PREC_PRIORITY             0x20
+#define IPTOS_PREC_ROUTINE              0x00
+
+
 /*
  * Commands for ioctlsocket(),  taken from the BSD file fcntl.h.
  *
@@ -174,10 +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, 
@@ -101,9 +101,56 @@ extern struct stats_ lwip_stats;
 
 
 void stats_init(void);
+
+#define STATS_INC(x) ++lwip_stats.x
 #else
 #define stats_init()
+#define STATS_INC(x)
 #endif /* LWIP_STATS */
+
+#if TCP_STATS
+#define TCP_STATS_INC(x) STATS_INC(x)
+#else
+#define TCP_STATS_INC(x)
+#endif
+
+#if UDP_STATS
+#define UDP_STATS_INC(x) STATS_INC(x)
+#else
+#define UDP_STATS_INC(x)
+#endif
+
+#if ICMP_STATS
+#define ICMP_STATS_INC(x) STATS_INC(x)
+#else
+#define ICMP_STATS_INC(x)
+#endif
+
+#if IP_STATS
+#define IP_STATS_INC(x) STATS_INC(x)
+#else
+#define IP_STATS_INC(x)
+#endif
+
+#if IPFRAG_STATS
+#define IPFRAG_STATS_INC(x) STATS_INC(x)
+#else
+#define IPFRAG_STATS_INC(x)
+#endif
+
+#if LINK_STATS
+#define LINK_STATS_INC(x) STATS_INC(x)
+#else
+#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;
@@ -53,7 +51,7 @@ void             tcp_init    (void);  /* Must be called first to
            initialize TCP. */
 void             tcp_tmr     (void);  /* Must be called every
            TCP_TMR_INTERVAL
-           ms. (Typically 100 ms). */
+           ms. (Typically 250 ms). */
 /* Application program's interface: */
 struct tcp_pcb * tcp_new     (void);
 struct tcp_pcb * tcp_alloc   (u8_t prio);
@@ -115,30 +113,32 @@ void             tcp_rexmit  (struct tcp_pcb *pcb);
 #define TCP_SEQ_GT(a,b)     ((s32_t)((a)-(b)) > 0)
 #define TCP_SEQ_GEQ(a,b)    ((s32_t)((a)-(b)) >= 0)
 
-#define TCP_FIN 0x01
-#define TCP_SYN 0x02
-#define TCP_RST 0x04
-#define TCP_PSH 0x08
-#define TCP_ACK 0x10
-#define TCP_URG 0x20
+#define TCP_FIN 0x01U
+#define TCP_SYN 0x02U
+#define TCP_RST 0x04U
+#define TCP_PSH 0x08U
+#define TCP_ACK 0x10U
+#define TCP_URG 0x20U
+#define TCP_ECE 0x40U
+#define TCP_CWR 0x80U
 
-#define TCP_FLAGS 0x3f
+#define TCP_FLAGS 0x3fU
 
 /* Length of the TCP header, excluding options. */
 #define TCP_HLEN 20
 
 #ifndef TCP_TMR_INTERVAL
-#define TCP_TMR_INTERVAL       100  /* The TCP timer interval in
+#define TCP_TMR_INTERVAL       250  /* The TCP timer interval in
                                        milliseconds. */
 #endif /* TCP_TMR_INTERVAL */
 
 #ifndef TCP_FAST_INTERVAL
-#define TCP_FAST_INTERVAL      200  /* the fine grained timeout in
+#define TCP_FAST_INTERVAL      TCP_TMR_INTERVAL /* the fine grained timeout in
                                        milliseconds */
 #endif /* TCP_FAST_INTERVAL */
 
 #ifndef TCP_SLOW_INTERVAL
-#define TCP_SLOW_INTERVAL      500  /* the coarse grained timeout in
+#define TCP_SLOW_INTERVAL      (2*TCP_TMR_INTERVAL)  /* the coarse grained timeout in
                                        milliseconds */
 #endif /* TCP_SLOW_INTERVAL */
 
@@ -149,6 +149,19 @@ void             tcp_rexmit  (struct tcp_pcb *pcb);
 
 #define TCP_MSL 60000  /* The maximum segment lifetime in microseconds */
 
+/*
+ * User-settable options (used with setsockopt).
+ */
+#define	TCP_NODELAY	   0x01	   /* don't delay send to coalesce packets */
+#define TCP_KEEPALIVE  0x02    /* send KEEPALIVE probes when idle for pcb->keepalive miliseconds */
+
+/* Keepalive values */
+#define  TCP_KEEPDEFAULT   7200000                       /* KEEPALIVE timer in miliseconds */
+#define  TCP_KEEPINTVL     75000                         /* Time between KEEPALIVE probes in miliseconds */
+#define  TCP_KEEPCNT       9                             /* Counter for KEEPALIVE probes */
+#define  TCP_MAXIDLE       TCP_KEEPCNT * TCP_KEEPINTVL   /* Maximum KEEPALIVE probe time */
+
+
 #ifdef PACK_STRUCT_USE_INCLUDES
 #  include "arch/bpstruct.h"
 #endif
@@ -158,7 +171,7 @@ struct tcp_hdr {
   PACK_STRUCT_FIELD(u16_t dest);
   PACK_STRUCT_FIELD(u32_t seqno);
   PACK_STRUCT_FIELD(u32_t ackno);
-  PACK_STRUCT_FIELD(u16_t _offset_flags);
+  PACK_STRUCT_FIELD(u16_t _hdrlen_rsvd_flags);
   PACK_STRUCT_FIELD(u16_t wnd);
   PACK_STRUCT_FIELD(u16_t chksum);
   PACK_STRUCT_FIELD(u16_t urgp);
@@ -168,11 +181,15 @@ PACK_STRUCT_END
 #  include "arch/epstruct.h"
 #endif
 
-#define TCPH_OFFSET(hdr) (ntohs((hdr)->_offset_flags) >> 8)
-#define TCPH_FLAGS(hdr) (ntohs((hdr)->_offset_flags) & 0xff)
+#define TCPH_OFFSET(phdr) (ntohs((phdr)->_hdrlen_rsvd_flags) >> 8)
+#define TCPH_HDRLEN(phdr) (ntohs((phdr)->_hdrlen_rsvd_flags) >> 12)
+#define TCPH_FLAGS(phdr)  (ntohs((phdr)->_hdrlen_rsvd_flags) & TCP_FLAGS)
 
-#define TCPH_OFFSET_SET(hdr, offset) (hdr)->_offset_flags = htons(((offset) << 8) | TCPH_FLAGS(hdr))
-#define TCPH_FLAGS_SET(hdr, flags) (hdr)->_offset_flags = htons((TCPH_OFFSET(hdr) << 8) | (flags))
+#define TCPH_OFFSET_SET(phdr, offset) (phdr)->_hdrlen_rsvd_flags = htons(((offset) << 8) | TCPH_FLAGS(phdr))
+#define TCPH_HDRLEN_SET(phdr, len) (phdr)->_hdrlen_rsvd_flags = htons(((len) << 12) | TCPH_FLAGS(phdr))
+#define TCPH_FLAGS_SET(phdr, flags) (phdr)->_hdrlen_rsvd_flags = htons((ntohs((phdr)->_hdrlen_rsvd_flags) & ~TCP_FLAGS) | (flags))
+#define TCPH_SET_FLAG(phdr, flags ) (phdr)->_hdrlen_rsvd_flags = htons(ntohs((phdr)->_hdrlen_rsvd_flags) | (flags))
+#define TCPH_UNSET_FLAG(phdr, flags) (phdr)->_hdrlen_rsvd_flags = htons(ntohs((phdr)->_hdrlen_rsvd_flags) | (TCPH_FLAGS(phdr) & ~(flags)) )
 
 #define TCP_TCPLEN(seg) ((seg)->len + ((TCPH_FLAGS((seg)->tcphdr) & TCP_FIN || \
           TCPH_FLAGS((seg)->tcphdr) & TCP_SYN)? 1: 0))
@@ -194,17 +211,30 @@ enum tcp_state {
 
 /* the TCP protocol control block */
 struct tcp_pcb {
+/* Common members of all PCB types */
+  IP_PCB;
+
+/* Protocol specific PCB members */
+
   struct tcp_pcb *next;   /* for the linked list */
+
+  enum tcp_state state;   /* TCP state */
+
   u8_t prio;
   void *callback_arg;
 
-  struct ip_addr local_ip;
   u16_t local_port;
-  enum tcp_state state;   /* TCP state */
-  
-  struct ip_addr remote_ip;
   u16_t remote_port;
   
+  u8_t flags;
+#define TF_ACK_DELAY (u8_t)0x01U   /* Delayed ACK. */
+#define TF_ACK_NOW   (u8_t)0x02U   /* Immediate ACK. */
+#define TF_INFR      (u8_t)0x04U   /* In fast recovery. */
+#define TF_RESET     (u8_t)0x08U   /* Connection was reset. */
+#define TF_CLOSED    (u8_t)0x10U   /* Connection was sucessfully closed. */
+#define TF_GOT_FIN   (u8_t)0x20U   /* Connection was closed by the remote end. */
+#define TF_NODELAY   (u8_t)0x40U   /* Disable Nagle algorithm */
+
   /* receiver varables */
   u32_t rcv_nxt;   /* next seqno expected */
   u16_t rcv_wnd;   /* receiver window */
@@ -217,14 +247,6 @@ struct tcp_pcb {
   u16_t rtime;
   
   u16_t mss;   /* maximum segment size */
-
-  u8_t flags;
-#define TF_ACK_DELAY 0x01U   /* Delayed ACK. */
-#define TF_ACK_NOW   0x02U   /* Immediate ACK. */
-#define TF_INFR      0x04U   /* In fast recovery. */
-#define TF_RESET     0x08U   /* Connection was reset. */
-#define TF_CLOSED    0x10U   /* Connection was sucessfully closed. */
-#define TF_GOT_FIN   0x20U   /* Connection was closed by the remote end. */
   
   /* RTT estimation variables. */
   u16_t rttest; /* RTT estimate in 500ms ticks */
@@ -282,21 +304,32 @@ struct tcp_pcb {
   /* Function to be called whenever a fatal error occurs. */
   void (* errf)(void *arg, err_t err);
 #endif /* LWIP_CALLBACK_API */
+
+  /* idle time before KEEPALIVE is sent */
+  u32_t keepalive;
+  
+  /* KEEPALIVE counter */
+  u8_t keep_cnt;
 };
 
 struct tcp_pcb_listen {  
+/* Common members of all PCB types */
+  IP_PCB;
+
+/* Protocol specific PCB members */
   struct tcp_pcb_listen *next;   /* for the linked list */
-  u8_t prio;
-  void *callback_arg;
   
-  struct ip_addr local_ip;
-  u16_t local_port; 
   /* Even if state is obviously LISTEN this is here for
    * field compatibility with tpc_pcb to which it is cast sometimes
    * Until a cleaner solution emerges this is here.FIXME
    */ 
   enum tcp_state state;   /* TCP state */
 
+  u8_t prio;
+  void *callback_arg;
+  
+  u16_t local_port; 
+
 #if LWIP_CALLBACK_API
   /* Function to call when a listener has been connected. */
   err_t (* accept)(void *arg, struct tcp_pcb *newpcb, err_t err);
@@ -396,6 +429,8 @@ void tcp_rst(u32_t seqno, u32_t ackno,
 
 u32_t tcp_next_iss(void);
 
+void tcp_keepalive(struct tcp_pcb *pcb);
+
 extern struct tcp_pcb *tcp_input_pcb;
 extern u32_t tcp_ticks;
 
@@ -406,7 +441,11 @@ void tcp_debug_print_state(enum tcp_state s);
 void tcp_debug_print_pcbs(void);
 int tcp_pcbs_sane(void);
 #else
-#define tcp_pcbs_sane() 1
+#  define tcp_debug_print(tcphdr)
+#  define tcp_debug_print_flags(flags)
+#  define tcp_debug_print_state(s)
+#  define tcp_debug_print_pcbs()
+#  define tcp_pcbs_sane() 1
 #endif /* TCP_DEBUG */
 
 #if NO_SYS
@@ -416,7 +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, 
@@ -38,8 +38,6 @@
 #include "lwip/inet.h"
 #include "lwip/ip.h"
 
-#include "lwip/err.h"
-
 #define UDP_HLEN 8
 
 struct udp_hdr {
@@ -54,12 +52,16 @@ struct udp_hdr {
 #define UDP_FLAGS_CONNECTED  0x04U
 
 struct udp_pcb {
+/* Common members of all PCB types */
+  IP_PCB;
+
+/* Protocol specific PCB members */
+
   struct udp_pcb *next;
 
-  struct ip_addr local_ip, remote_ip;
+  u8_t flags;
   u16_t local_port, remote_port;
   
-  u8_t flags;
   u16_t chksum_len;
   
   void (* recv)(void *arg, struct udp_pcb *pcb, struct pbuf *p,
@@ -82,18 +84,21 @@ 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);
 
-
+#if UDP_DEBUG
+int udp_debug_print(struct udp_hdr *udphdr);
+#else
+#define udp_debug_print(udphdr)
+#endif
 #endif /* __LWIP_UDP_H__ */
 
 

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

@@ -4,12 +4,18 @@
  *
  * Functionally, ARP is divided into two parts. The first maps an IP address
  * to a physical address when sending a packet, and the second part answers
- * requests from other machines.
+ * requests from other machines for our physical address.
  *
+ * 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,
@@ -36,42 +42,8 @@
  *
  * This file is part of the lwIP TCP/IP stack.
  *
- * Author: Adam Dunkels <adam@sics.se>
- *
  */
 
-/*
- * 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"
@@ -83,10 +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
 
@@ -103,7 +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 {
@@ -111,6 +88,9 @@ struct etharp_entry {
   struct eth_addr ethaddr;
   enum etharp_state state;
 #if ARP_QUEUEING
+  /** 
+   * Pointer to queue of pending outgoing packets on this ARP entry.
+   * Must be at most a single packet for now. */
   struct pbuf *p;
 #endif
   u8_t ctime;
@@ -119,22 +99,24 @@ struct etharp_entry {
 static const struct eth_addr ethbroadcast = {{0xff,0xff,0xff,0xff,0xff,0xff}};
 static struct etharp_entry arp_table[ARP_TABLE_SIZE];
 
-static struct pbuf *update_arp_entry(struct netif *netif, struct ip_addr *ipaddr, struct eth_addr *ethaddr, u8_t flags);
+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);
 /**
  * Initializes ARP module.
  */
 void
 etharp_init(void)
 {
-  u8_t i;
+  s8_t i;
   /* clear ARP entries */
   for(i = 0; i < ARP_TABLE_SIZE; ++i) {
     arp_table[i].state = ETHARP_STATE_EMPTY;
 #if ARP_QUEUEING
     arp_table[i].p = NULL;
 #endif
+    arp_table[i].ctime = 0;
   }
 }
 
@@ -147,90 +129,104 @@ etharp_init(void)
 void
 etharp_tmr(void)
 {
-  u8_t i;
+  s8_t i;
 
   LWIP_DEBUGF(ETHARP_DEBUG, ("etharp_timer\n"));
   /* remove expired entries from the ARP table */
   for (i = 0; i < ARP_TABLE_SIZE; ++i) {
     arp_table[i].ctime++;
+    /* a resolved/stable entry? */
     if ((arp_table[i].state == ETHARP_STATE_STABLE) &&
+         /* entry has become old? */
         (arp_table[i].ctime >= ARP_MAXAGE)) {
       LWIP_DEBUGF(ETHARP_DEBUG, ("etharp_timer: expired stable entry %u.\n", i));
-      arp_table[i].state = ETHARP_STATE_EMPTY;
-#if ARP_QUEUEING
-      if (arp_table[i].p != NULL) {
-        /* remove any queued packet */
-        LWIP_DEBUGF(ETHARP_DEBUG, ("etharp_timer: freeing packet queue %p.\n", i, (void *)(arp_table[i].p)));
-        pbuf_free(arp_table[i].p);
-        arp_table[i].p = NULL;
-      }
-#endif
+      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)) {
-      arp_table[i].state = ETHARP_STATE_EMPTY;
       LWIP_DEBUGF(ETHARP_DEBUG, ("etharp_timer: expired pending entry %u.\n", i));
+      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 */
-        LWIP_DEBUGF(ETHARP_DEBUG, ("etharp_timer: freeing packet queue %p.\n", i, (void *)(arp_table[i].p)));
+        /* 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;
     }
   }
 }
 
 /**
- * Return an empty ARP entry or, if the table is full, ARP_TABLE_SIZE if all
- * entries are pending, otherwise the oldest entry.
+ * Return an empty ARP entry (possibly recycling the oldest stable entry).
  *
- * @return The ARP entry index that is available, ARP_TABLE_SIZE if no usable
+ * @return The ARP entry index that is available, ERR_MEM if no usable
  * entry is found.
  */
-static u8_t
+static s8_t
 find_arp_entry(void)
 {
-  u8_t i, j, maxtime;
+  s8_t i, j;
+  u8_t maxtime = 0;
 
-  /* Try to find an unused entry in the ARP table. */
+  j = ARP_TABLE_SIZE;
+  /* search ARP table for an unused or old entry */
   for (i = 0; i < ARP_TABLE_SIZE; ++i) {
+  	/* empty entry? */
     if (arp_table[i].state == ETHARP_STATE_EMPTY) {
-      LWIP_DEBUGF(ETHARP_DEBUG, ("find_arp_entry: found empty entry %u\n", i));
-      break;
+      LWIP_DEBUGF(ETHARP_DEBUG, ("find_arp_entry: returning empty entry %u\n", i));
+      return i;
+  	/* stable entry? */
+    } else if (arp_table[i].state == ETHARP_STATE_STABLE) {
+      /* remember entry with oldest stable entry in j */
+      if (arp_table[i].ctime >= maxtime) maxtime = arp_table[j = i].ctime;
     }
   }
+  /* no empty entry found? */
+  if (i == ARP_TABLE_SIZE) {
+  	LWIP_DEBUGF(ETHARP_DEBUG, ("find_arp_entry: found oldest stable entry %u\n", j));
+    /* fall-back to oldest stable */
+  	i = j;
+  }
+  /* no available entry found? */
+  if (i == ARP_TABLE_SIZE) {
+    LWIP_DEBUGF(ETHARP_DEBUG, ("find_arp_entry: no replacable entry could be found\n"));
+    /* return failure */
+    return ERR_MEM;
+  }
 
-  /* If no unused entry is found, we try to find the oldest entry and
-     throw it away. If all entries are new and have 0 ctime drop one  */
-  if (i == ARP_TABLE_SIZE) {
-    maxtime = 0;
-    j = ARP_TABLE_SIZE;
-    for (i = 0; i < ARP_TABLE_SIZE; ++i) {
-      /* remember entry with oldest stable entry in j*/
-      if ((arp_table[i].state == ETHARP_STATE_STABLE) &&
-#if ARP_QUEUEING /* do not want to re-use an entry with queued packets */
-      (arp_table[i].p == NULL) &&
+  /* clean up the oldest stable entry (to be recycled) */
+  if (arp_table[i].state == ETHARP_STATE_STABLE) {
+#if ARP_QUEUEING
+    /* 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
-      (arp_table[i].ctime >= maxtime)) {
-        maxtime = arp_table[i].ctime;
-        j = i;
-      }
-    }
-    if (j != ARP_TABLE_SIZE) {
-      LWIP_DEBUGF(ETHARP_DEBUG, ("find_arp_entry: found oldest stable entry %u\n", j));
-    } else {
-      LWIP_DEBUGF(ETHARP_DEBUG, ("find_arp_entry: no replacable entry could be found\n"));
-    }
-    i = j;
+    LWIP_DEBUGF(ETHARP_DEBUG | DBG_TRACE, ("find_arp_entry: recycling oldest stable entry %u\n", i));
+    arp_table[i].state = ETHARP_STATE_EMPTY;
   }
-  LWIP_DEBUGF(ETHARP_DEBUG, ("find_arp_entry: returning %u, state %u\n", i, arp_table[i].state));
+  LWIP_DEBUGF(ETHARP_DEBUG, ("find_arp_entry: returning %u\n", i));
   return i;
 }
 
 /**
  * Update (or insert) a IP/MAC address pair in the ARP cache.
  *
+ * If a pending entry is resolved, any queued packets will be sent
+ * at this point.
+ * 
  * @param ipaddr IP address of the inserted ARP entry.
  * @param ethaddr Ethernet address of the inserted ARP entry.
  * @param flags Defines behaviour:
@@ -245,23 +241,25 @@ find_arp_entry(void)
 static struct pbuf *
 update_arp_entry(struct netif *netif, struct ip_addr *ipaddr, struct eth_addr *ethaddr, u8_t flags)
 {
-  u8_t i, k;
+  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));
@@ -269,12 +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? (possible 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) {
@@ -282,55 +276,49 @@ update_arp_entry(struct netif *netif, struct ip_addr *ipaddr, struct eth_addr *e
         }
         /* reset time stamp */
         arp_table[i].ctime = 0;
+/* this is where we will send out queued packets! */
 #if ARP_QUEUEING
-        p = arp_table[i].p;
-        /* queued packet present? */
-        if (p != NULL) {
-          /* NULL attached buffer immediately */
-          arp_table[i].p = 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.\n"));
+          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);
         }
 #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. */
     i = find_arp_entry();
-    if (i == ARP_TABLE_SIZE) {
+    if (i == ERR_MEM) {
       LWIP_DEBUGF(ETHARP_DEBUG | DBG_TRACE, ("update_arp_entry: no available entry found\n"));
       return NULL;
     }
-    /* see if find_arp_entry() gave us an old stable, or empty entry to re-use */
-    if (arp_table[i].state == ETHARP_STATE_STABLE) {
-      LWIP_DEBUGF(ETHARP_DEBUG | DBG_TRACE, ("update_arp_entry: overwriting old stable entry %u\n", i));
-      /* stable entries should have no queued packets (TODO: allow later) */
-#if ARP_QUEUEING
-      LWIP_ASSERT("update_arp_entry: arp_table[i].p == NULL", arp_table[i].p == NULL);
-#endif
-    } else {
-      LWIP_DEBUGF(ETHARP_DEBUG | DBG_TRACE | DBG_STATE, ("update_arp_entry: filling empty entry %u with state %u\n", i, arp_table[i].state));
-      LWIP_ASSERT("update_arp_entry: arp_table[i].state == ETHARP_STATE_EMPTY", arp_table[i].state == ETHARP_STATE_EMPTY);
-    }
     /* set IP address */
     ip_addr_set(&arp_table[i].ipaddr, ipaddr);
     /* set Ethernet hardware address */
@@ -354,7 +342,7 @@ update_arp_entry(struct netif *netif, struct ip_addr *ipaddr, struct eth_addr *e
 }
 
 /**
- * Updates the ARP table using the given packet.
+ * Updates the ARP table using the given IP packet.
  *
  * Uses the incoming IP packet's source address to update the
  * ARP cache for the local network. The function does not alter
@@ -368,7 +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;
@@ -379,18 +367,18 @@ 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;
 }
 
 
 /**
- * Responds to ARP requests, updates ARP entries and sends queued IP packets.
+ * Responds to ARP requests to us. Upon ARP replies to us, add entry to cache  
+ * send out queued IP packets. Updates cache with snooped address pairs.
  *
  * Should be called for incoming ARP packets. The pbuf in the argument
  * is freed by this function.
@@ -403,46 +391,65 @@ 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;
 
   /* drop short ARP packets */
   if (p->tot_len < sizeof(struct etharp_hdr)) {
-    LWIP_DEBUGF(ETHARP_DEBUG | DBG_TRACE | 1, ("etharp_arp_input: packet too short (%d/%d)\n", 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(&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;
-    }
-    /* update the ARP cache */
-    update_arp_entry(netif, &(hdr->sipaddr), &(hdr->shwaddr), 0);
     /* ARP request for our address? */
-    if (ip_addr_cmp(&(hdr->dipaddr), &(netif->ip_addr))) {
+    if (for_us) {
 
       LWIP_DEBUGF(ETHARP_DEBUG | DBG_TRACE, ("etharp_arp_input: replying to ARP request for our IP address\n"));
       /* 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];
@@ -454,36 +461,29 @@ etharp_arp_input(struct netif *netif, struct eth_addr *ethaddr, struct pbuf *p)
       hdr->hwtype = htons(HWTYPE_ETHERNET);
       ARPH_HWLEN_SET(hdr, netif->hwaddr_len);
 
-        hdr->proto = htons(ETHTYPE_IP);
+      hdr->proto = htons(ETHTYPE_IP);
       ARPH_PROTOLEN_SET(hdr, sizeof(struct ip_addr));
 
-        hdr->ethhdr.type = htons(ETHTYPE_ARP);
+      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. */
+    /* 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 */
-    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
-    /* ARP reply directed to us? */
-    if (ip_addr_cmp(&(hdr->dipaddr), &(netif->ip_addr))) {
-      LWIP_DEBUGF(ETHARP_DEBUG | DBG_TRACE, ("etharp_arp_input: incoming ARP reply is for us\n"));
-      /* update_the ARP cache, ask to insert */
-      update_arp_entry(netif, &(hdr->sipaddr), &(hdr->shwaddr), ARP_INSERT_FLAG);
-    /* ARP reply not directed to us */
-    } else {
-      LWIP_DEBUGF(ETHARP_DEBUG | DBG_TRACE, ("etharp_arp_input: incoming ARP reply is not for us\n"));
-      /* update the destination address pair */
-      update_arp_entry(netif, &(hdr->sipaddr), &(hdr->shwaddr), 0);
-      /* update the destination address pair */
-      update_arp_entry(netif, &(hdr->dipaddr), &(hdr->dhwaddr), 0);
-    }
     break;
   default:
     LWIP_DEBUGF(ETHARP_DEBUG | DBG_TRACE, ("etharp_arp_input: ARP unknown opcode type %d\n", htons(hdr->opcode)));
@@ -491,9 +491,6 @@ etharp_arp_input(struct netif *netif, struct eth_addr *ethaddr, struct pbuf *p)
   }
   /* free ARP packet */
   pbuf_free(p);
-  p = NULL;
-  /* nothing to send, we did it! */
-  return NULL;
 }
 
 /**
@@ -503,51 +500,50 @@ etharp_arp_input(struct netif *netif, struct eth_addr *ethaddr, struct pbuf *p)
  * returned, ready to be sent.
  *
  * If ARP does not have the Ethernet address in cache the packet is
- * queued and a ARP request is sent (on a best-effort basis). This
- * ARP request is returned as a pbuf, which should be sent by the
- * caller.
- *
- * If ARP failed to allocate resources, NULL is returned.
+ * queued (if enabled and space available) and a ARP request is sent.
+ * This ARP request is returned as a pbuf, which should be sent by
+ * the caller.
  *
  * A returned non-NULL packet should be sent by the caller.
  *
+ * If ARP failed to allocate resources, NULL is returned.
+ *
  * @param netif The lwIP network interface which the IP packet will be sent on.
  * @param ipaddr The IP address of the packet destination.
  * @param pbuf The pbuf(s) containing the IP packet to be sent.
  *
- * @return If non-NULL, a packet ready to be sent.
+ * @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;
-  u8_t i;
+  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"));
-#ifdef LINK_STATS
-    ++lwip_stats.link.lenerr;
-#endif /* LINK_STATS */
-    return NULL;
+    LINK_STATS_INC(link.lenerr);
+    pbuf_free(q);
+    return ERR_BUF;
   }
 
-  /* obtain source Ethernet address of the given interface */
-  srcaddr = (struct eth_addr *)netif->hwaddr;
-
   /* 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;
   }
@@ -555,7 +551,7 @@ etharp_output(struct netif *netif, struct ip_addr *ipaddr, struct pbuf *q)
   else if (ip_addr_ismulticast(ipaddr)) {
     /* Hash IP multicast address to MAC address. */
     mcastaddr.addr[0] = 0x01;
-    mcastaddr.addr[1] = 0x0;
+    mcastaddr.addr[1] = 0x00;
     mcastaddr.addr[2] = 0x5e;
     mcastaddr.addr[3] = ip4_addr2(ipaddr) & 0x7f;
     mcastaddr.addr[4] = ip4_addr3(ipaddr);
@@ -565,80 +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? */
-    /* this occurs if the packet is routed to the default gateway on this interface */
+    /* outside local network? */
     if (!ip_addr_maskcmp(ipaddr, &(netif->ip_addr), &(netif->netmask))) {
-      /* gateway available? */
-      if (netif->gw.addr != 0)
-      {
-        /* use the gateway IP address */
+      /* interface has default gateway? */
+      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 */
-        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 */
-      etharp_query(netif, ipaddr, q);
-      /* 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) {
-    /* A valid IP->MAC address mapping was found, so we construct the
-    Ethernet header for the outgoing packet. */
+    /* 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.
  *
@@ -647,30 +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;
-  u8_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 stable 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;
       }
     }
@@ -678,91 +704,52 @@ err_t etharp_query(struct netif *netif, struct ip_addr *ipaddr, struct pbuf *q)
   /* queried address not yet in ARP table? */
   if (i == ARP_TABLE_SIZE) {
     LWIP_DEBUGF(ETHARP_DEBUG | DBG_TRACE, ("etharp_query: IP address not found in ARP table\n"));
-    /* find an available entry */
+    /* find an available (unused or old) entry */
     i = find_arp_entry();
     /* bail out if no ARP entries are available */
-    if (i == ARP_TABLE_SIZE) {
-      LWIP_DEBUGF(ETHARP_DEBUG | 2, ("etharp_query: no more ARP entries available.\n"));
+    if (i == ERR_MEM) {
+      LWIP_DEBUGF(ETHARP_DEBUG | 2, ("etharp_query: no more ARP entries available. Should seldom occur.\n"));
       return ERR_MEM;
     }
-    /* we will now recycle entry i */
-    LWIP_DEBUGF(ETHARP_DEBUG | DBG_TRACE, ("etharp_query: created ARP table entry %u.\n", i));
     /* i is available, create ARP entry */
-    ip_addr_set(&arp_table[i].ipaddr, ipaddr);
-    arp_table[i].ctime = 0;
     arp_table[i].state = ETHARP_STATE_PENDING;
-#if ARP_QUEUEING
-    /* free queued packet, as entry is now invalidated */
-    if (arp_table[i].p != NULL) {
-      pbuf_free(arp_table[i].p);
-      arp_table[i].p = NULL;
-      LWIP_DEBUGF(ETHARP_DEBUG | DBG_TRACE | 3, ("etharp_query: dropped packet on ARP queue. Should not occur.\n"));
-    }
-#endif
+    ip_addr_set(&arp_table[i].ipaddr, ipaddr);
+    arp_table[i].p = NULL;
+    LWIP_DEBUGF(ETHARP_DEBUG | DBG_TRACE, ("etharp_query: added pending entry %u for IP address\n", i));
   }
-#if ARP_QUEUEING
-  /* any pbuf to queue and queue is empty? */
+
+  /* { i is either a (new or existing) PENDING or STABLE entry } */
+  
+  /* packet given? */
   if (q != NULL) {
-/* yield later packets over older packets? */
-#if ARP_QUEUE_FIRST == 0
-    /* earlier queued packet on this entry? */
-    if (arp_table[i].p != NULL) {
-      pbuf_free(arp_table[i].p);
-      arp_table[i].p = NULL;
-      LWIP_DEBUGF(ETHARP_DEBUG | DBG_TRACE | 3, ("etharp_query: dropped packet on ARP queue. Should not occur.\n"));
-      /* fall-through into next if */
-    }
-#endif
-    /* packet can be queued? */
-    if (arp_table[i].p == NULL) {
-      /* copy PBUF_REF referenced payloads into PBUF_RAM */
-      q = pbuf_take(q);
-      /* remember pbuf to queue, if any */
-      arp_table[i].p = q;
-      /* pbufs are 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));
-    }
-  }
-#endif
-  /* ARP request? */
-  if (perform_arp_request)
-  {
-    /* 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->dhwaddr.addr[j] = 0x00;
-        hdr->shwaddr.addr[j] = srcaddr->addr[j];
+    /* 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];
       }
-      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];
+      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; 
       }
-      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"));
+#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;
@@ -87,7 +84,7 @@ low_level_init(struct netif *netif)
  
   /* Do whatever else is needed to initialize interface. */  
 }
-/*-----------------------------------------------------------------------------------*/
+
 /*
  * low_level_output():
  *
@@ -96,7 +93,6 @@ low_level_init(struct netif *netif)
  * might be chained.
  *
  */
-/*-----------------------------------------------------------------------------------*/
 
 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++;
@@ -120,7 +124,7 @@ low_level_output(struct ethernetif *ethernetif, struct pbuf *p)
 
   return ERR_OK;
 }
-/*-----------------------------------------------------------------------------------*/
+
 /*
  * low_level_input():
  *
@@ -128,7 +132,7 @@ low_level_output(struct ethernetif *ethernetif, struct pbuf *p)
  * packet from the interface into the pbuf.
  *
  */
-/*-----------------------------------------------------------------------------------*/
+
 static struct pbuf *
 low_level_input(struct ethernetif *ethernetif)
 {
@@ -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 */      
@@ -165,106 +183,26 @@ low_level_input(struct ethernetif *ethernetif)
 
   return p;  
 }
-/*-----------------------------------------------------------------------------------*/
+
 /*
  * ethernetif_output():
  *
  * 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.
  *
  */
-/*-----------------------------------------------------------------------------------*/
+
 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);
+ 
 }
-/*-----------------------------------------------------------------------------------*/
+
 /*
  * ethernetif_input():
  *
@@ -274,7 +212,7 @@ ethernetif_output(struct netif *netif, struct pbuf *p,
  * interface.
  *
  */
-/*-----------------------------------------------------------------------------------*/
+
 static void
 ethernetif_input(struct netif *netif)
 {
@@ -282,46 +220,50 @@ 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);
 }
-/*-----------------------------------------------------------------------------------*/
+
 /*
  * ethernetif_init():
  *
@@ -330,13 +272,20 @@ arp_timer(void *arg)
  * actual setup of the hardware.
  *
  */
-/*-----------------------------------------------------------------------------------*/
+
 void
 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;
@@ -346,8 +295,9 @@ 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, 
@@ -29,9 +29,12 @@
  * Author: Adam Dunkels <adam@sics.se>
  *
  */
-#include "lwip/mem.h"
 #include "lwip/opt.h"
+
+#if LWIP_HAVE_LOOPIF
+
 #include "netif/loopif.h"
+#include "lwip/mem.h"
 
 #if defined(LWIP_DEBUG) && defined(LWIP_TCPDUMP)
 #include "netif/tcpdump.h"
@@ -40,13 +43,23 @@
 #include "lwip/tcp.h"
 #include "lwip/ip.h"
 
-/*-----------------------------------------------------------------------------------*/
+static void
+loopif_input( void * arg )
+{
+	struct netif *netif = (struct netif *)( ((void **)arg)[ 0 ] );
+	struct pbuf *r = (struct pbuf *)( ((void **)arg)[ 1 ] );
+
+	mem_free( arg );
+	netif -> input( r, netif );
+}
+
 static err_t
 loopif_output(struct netif *netif, struct pbuf *p,
        struct ip_addr *ipaddr)
 {
   struct pbuf *q, *r;
   char *ptr;
+  void **arg;
 
 #if defined(LWIP_DEBUG) && defined(LWIP_TCPDUMP)
   tcpdump(p);
@@ -60,21 +73,43 @@ loopif_output(struct netif *netif, struct pbuf *p,
       memcpy(ptr, q->payload, q->len);
       ptr += q->len;
     }
-    netif->input(r, netif);
+
+    arg = mem_malloc( sizeof( void *[2]));
+	if( NULL == arg ) {
+		return ERR_MEM;
+	}
+	
+	arg[0] = netif;
+	arg[1] = r;
+	/**
+	 * workaround (patch #1779) to try to prevent bug #2595:
+	 * When connecting to "localhost" with the loopif interface,
+	 * tcp_output doesn't get the opportunity to finnish sending the
+	 * segment before tcp_process gets it, resulting in tcp_process
+	 * referencing pcb->unacked-> which still is NULL.
+	 * 
+	 * TODO: Is there still a race condition here? Leon
+	 */
+	sys_timeout( 1, loopif_input, arg );
+	
     return ERR_OK;    
   }
   return ERR_MEM;
 }
-/*-----------------------------------------------------------------------------------*/
+
 err_t
 loopif_init(struct netif *netif)
 {
   netif->name[0] = 'l';
   netif->name[1] = 'o';
+#if 0 /** TODO: I think this should be enabled, or not? Leon */
+  netif->input = loopif_input;
+#endif
   netif->output = loopif_output;
   return ERR_OK;
 }
-/*-----------------------------------------------------------------------------------*/
+
+#endif /* LWIP_HAVE_LOOPIF */
 
 
 

src/netif/ppp/auth.c

@@ -122,7 +122,7 @@ static int  get_pap_passwd (int, char *, char *);
 static int  have_pap_secret (void);
 static int  have_chap_secret (char *, char *, u32_t);
 static int  ip_addr_check (u32_t, struct wordlist *);
-#if PAP_SUPPORT > 0 || CHAP_SUPPORT > 0
+#if 0 /* PAP_SUPPORT > 0 || CHAP_SUPPORT > 0 */
 static void set_allowed_addrs(int unit, struct wordlist *addrs);
 static void free_wordlist (struct wordlist *);
 #endif
@@ -194,7 +194,7 @@ void link_terminated(int unit)
     if (logged_in)
         logout();
     lcp_phase[unit] = PHASE_DEAD;
-    ppp_trace(LOG_NOTICE, "Connection terminated.\n");
+    AUTHDEBUG((LOG_NOTICE, "Connection terminated.\n"));
 	pppMainWakeup(unit);
 }
 
@@ -257,7 +257,7 @@ void link_established(int unit)
          * of "" and a password of "".  If that's not OK, boot it out.
          */
         if (!wo->neg_upap || !null_login(unit)) {
-            ppp_trace(LOG_WARNING, "peer refused to authenticate\n");
+            AUTHDEBUG((LOG_WARNING, "peer refused to authenticate\n"));
             lcp_close(unit, "peer refused to authenticate");
             return;
         }
@@ -294,7 +294,7 @@ void link_established(int unit)
         if (ppp_settings.passwd[0] == 0) {
             passwd_from_file = 1;
             if (!get_pap_passwd(unit, ppp_settings.user, ppp_settings.passwd))
-                ppp_trace(LOG_ERR, "No secret found for PAP login\n");
+                AUTHDEBUG((LOG_ERR, "No secret found for PAP login\n"));
         }
         upap_authwithpeer(unit, ppp_settings.user, ppp_settings.passwd);
         auth |= PAP_WITHPEER;
@@ -337,8 +337,8 @@ void auth_peer_success(int unit, u16_t protocol, char *name, int namelen)
         pbit = PAP_PEER;
         break;
     default:
-        ppp_trace(LOG_WARNING, "auth_peer_success: unknown protocol %x\n",
-               protocol);
+        AUTHDEBUG((LOG_WARNING, "auth_peer_success: unknown protocol %x\n",
+               protocol));
         return;
     }
     
@@ -400,8 +400,8 @@ void auth_withpeer_success(int unit, u16_t protocol)
         pbit = PAP_WITHPEER;
         break;
     default:
-        ppp_trace(LOG_WARNING, "auth_peer_success: unknown protocol %x\n",
-               protocol);
+        AUTHDEBUG((LOG_WARNING, "auth_peer_success: unknown protocol %x\n",
+               protocol));
         pbit = 0;
     }
     
@@ -476,7 +476,7 @@ void auth_reset(int unit)
     
     AUTHDEBUG((LOG_INFO, "auth_reset: %d\n", unit));
     ao->neg_upap = !ppp_settings.refuse_pap && (ppp_settings.passwd[0] != 0 || get_pap_passwd(unit, NULL, NULL));
-    ao->neg_chap = !ppp_settings.refuse_chap && have_chap_secret(ppp_settings.user, ppp_settings.remote_name, (u32_t)0);
+    ao->neg_chap = !ppp_settings.refuse_chap && ppp_settings.passwd[0] != 0 /*have_chap_secret(ppp_settings.user, ppp_settings.remote_name, (u32_t)0)*/;
     
     if (go->neg_upap && !have_pap_secret())
         go->neg_upap = 0;
@@ -541,7 +541,7 @@ int check_passwd(
          * On 10'th, drop the connection.
          */
         if (attempts++ >= 10) {
-            ppp_trace(LOG_WARNING, "%d LOGIN FAILURES BY %s\n", attempts, user);
+            AUTHDEBUG((LOG_WARNING, "%d LOGIN FAILURES BY %s\n", attempts, user));
             /*ppp_panic("Excess Bad Logins");*/
         }
         if (attempts > 3) {
@@ -605,7 +605,24 @@ int get_secret(
 )
 {
 #if 1
+    int len;
+    struct wordlist *addrs;
+    
+    addrs = NULL;
+
+    if(!client || !client[0] || strcmp(client, ppp_settings.user)) {
 	return 0;
+    }
+
+    len = strlen(ppp_settings.passwd);
+    if (len > MAXSECRETLEN) {
+        AUTHDEBUG((LOG_ERR, "Secret for %s on %s is too long\n", client, server));
+        len = MAXSECRETLEN;
+    }
+    BCOPY(ppp_settings.passwd, secret, len);
+    *secret_len = len;
+    
+    return 1;
 #else
     int ret = 0, len;
     struct wordlist *addrs;
@@ -623,7 +640,7 @@ int get_secret(
     
     len = strlen(secbuf);
     if (len > MAXSECRETLEN) {
-        ppp_trace(LOG_ERR, "Secret for %s on %s is too long\n", client, server);
+        AUTHDEBUG((LOG_ERR, "Secret for %s on %s is too long\n", client, server));
         len = MAXSECRETLEN;
     }
     BCOPY(secbuf, secret, len);
@@ -736,7 +753,7 @@ static void check_idle(void *arg)
     itime = LWIP_MIN(idle.xmit_idle, idle.recv_idle);
     if (itime >= ppp_settings.idle_time_limit) {
         /* link is idle: shut it down. */
-        ppp_trace(LOG_INFO, "Terminating connection due to lack of activity.\n");
+        AUTHDEBUG((LOG_INFO, "Terminating connection due to lack of activity.\n"));
         lcp_close(0, "Link inactive");
     } else {
         TIMEOUT(check_idle, NULL, ppp_settings.idle_time_limit - itime);
@@ -750,7 +767,7 @@ static void connect_time_expired(void *arg)
 {
 	(void)arg;
 
-    ppp_trace(LOG_INFO, "Connect time expired\n");
+    AUTHDEBUG((LOG_INFO, "Connect time expired\n"));
     lcp_close(0, "Connect time expired");   /* Close connection */
 }
 
@@ -841,7 +858,7 @@ static int have_chap_secret(char *client, char *server, u32_t remote)
 }
 
 
-#if PAP_SUPPORT > 0 || CHAP_SUPPORT > 0
+#if 0 /* PAP_SUPPORT > 0 || CHAP_SUPPORT > 0 */
 /*
  * set_allowed_addrs() - set the list of allowed addresses.
  */
@@ -891,7 +908,7 @@ static int ip_addr_check(u32_t addr, struct wordlist *addrs)
     return 1;
 }
 
-#if PAP_SUPPORT > 0 || CHAP_SUPPORT
+#if 0 /* PAP_SUPPORT > 0 || CHAP_SUPPORT */
 /*
  * free_wordlist - release memory allocated for a wordlist.
  */

src/netif/ppp/chap.c

@@ -245,7 +245,7 @@ static void ChapChallengeTimeout(void *arg)
 	
 	if (cstate->chal_transmits >= cstate->max_transmits) {
 		/* give up on peer */
-		ppp_trace(LOG_ERR, "Peer failed to respond to CHAP challenge\n");
+		CHAPDEBUG((LOG_ERR, "Peer failed to respond to CHAP challenge\n"));
 		cstate->serverstate = CHAPSS_BADAUTH;
 		auth_peer_fail(cstate->unit, PPP_CHAP);
 		return;
@@ -405,7 +405,7 @@ static void ChapInput(int unit, u_char *inpacket, int packet_len)
 		break;
 	
 	default:				/* Need code reject? */
-		ppp_trace(LOG_WARNING, "Unknown CHAP code (%d) received.\n", code);
+		CHAPDEBUG((LOG_WARNING, "Unknown CHAP code (%d) received.\n", code));
 		break;
 	}
 }
@@ -466,7 +466,7 @@ static void ChapReceiveChallenge(chap_state *cstate, u_char *inp, int id, int le
 	if (!get_secret(cstate->unit, cstate->resp_name, rhostname,
 			    secret, &secret_len, 0)) {
 		secret_len = 0;		/* assume null secret if can't find one */
-		ppp_trace(LOG_WARNING, "No CHAP secret found for authenticating us to %s\n", rhostname);
+		CHAPDEBUG((LOG_WARNING, "No CHAP secret found for authenticating us to %s\n", rhostname));
 	}
 	
 	/* cancel response send timeout if necessary */
@@ -576,8 +576,8 @@ static void ChapReceiveResponse(chap_state *cstate, u_char *inp, int id, int len
 	if (!get_secret(cstate->unit, rhostname, cstate->chal_name,
 	secret, &secret_len, 1)) {
 /*        CHAPDEBUG((LOG_WARNING, TL_CHAP, "No CHAP secret found for authenticating %s\n", rhostname)); */
-		ppp_trace(LOG_WARNING, "No CHAP secret found for authenticating %s\n",
-		rhostname);
+		CHAPDEBUG((LOG_WARNING, "No CHAP secret found for authenticating %s\n",
+		rhostname));
 	} else {
 	
 		/*  generate MD based on negotiated type */
@@ -614,7 +614,7 @@ static void ChapReceiveResponse(chap_state *cstate, u_char *inp, int id, int len
 		if (cstate->chal_interval != 0)
 			TIMEOUT(ChapRechallenge, cstate, cstate->chal_interval);
 	} else {
-		ppp_trace(LOG_ERR, "CHAP peer authentication failed\n");
+		CHAPDEBUG((LOG_ERR, "CHAP peer authentication failed\n"));
 		cstate->serverstate = CHAPSS_BADAUTH;
 		auth_peer_fail(cstate->unit, PPP_CHAP);
 	}
@@ -675,7 +675,7 @@ static void ChapReceiveFailure(chap_state *cstate, u_char *inp, u_char id, int l
 	if (len > 0)
 		PRINTMSG(inp, len);
 	
-	ppp_trace(LOG_ERR, "CHAP authentication failed\n");
+	CHAPDEBUG((LOG_ERR, "CHAP authentication failed\n"));
 	auth_withpeer_fail(cstate->unit, PPP_CHAP);
 }
 

src/netif/ppp/chpms.h

@@ -29,7 +29,7 @@
 *   Ported to lwIP.
 * 98-01-30 Guy Lancaster <lancasterg@acm.org>, Global Election Systems Inc.
 *	Original built from BSD network code.
-******************************************************************************
+******************************************************************************/
 /*
  * chap.h - Challenge Handshake Authentication Protocol definitions.
  *
@@ -51,7 +51,7 @@
  * IMPLIED WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED
  * WARRANTIES OF MERCHANTIBILITY AND FITNESS FOR A PARTICULAR PURPOSE.
  *
- * $Id: chpms.h,v 1.1 2003/05/27 14:37:56 jani 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.
  */
@@ -1192,12 +1206,12 @@ static void ipcp_up(fsm *f)
 		ho->hisaddr = wo->hisaddr;
 	
 	if (ho->hisaddr == 0) {
-		ppp_trace(LOG_ERR, "Could not determine remote IP address\n");
+		IPCPDEBUG((LOG_ERR, "Could not determine remote IP address\n"));
 		ipcp_close(f->unit, "Could not determine remote IP address");
 		return;
 	}
 	if (go->ouraddr == 0) {
-		ppp_trace(LOG_ERR, "Could not determine local IP address\n");
+		IPCPDEBUG((LOG_ERR, "Could not determine local IP address\n"));
 		ipcp_close(f->unit, "Could not determine local IP address");
 		return;
 	}
@@ -1210,8 +1224,8 @@ static void ipcp_up(fsm *f)
 	 * Check that the peer is allowed to use the IP address it wants.
 	 */
 	if (!auth_ip_addr(f->unit, ho->hisaddr)) {
-		ppp_trace(LOG_ERR, "Peer is not authorized to use remote address %s\n",
-				inet_ntoa(ho->hisaddr));
+		IPCPDEBUG((LOG_ERR, "Peer is not authorized to use remote address %s\n",
+				inet_ntoa(ho->hisaddr)));
 		ipcp_close(f->unit, "Unauthorized remote IP address");
 		return;
 	}

src/netif/ppp/pap.c

@@ -216,7 +216,7 @@ static void upap_timeout(void *arg)
 	
 	if (u->us_transmits >= u->us_maxtransmits) {
 		/* give up in disgust */
-		ppp_trace(LOG_ERR, "No response to PAP authenticate-requests\n");
+		UPAPDEBUG((LOG_ERR, "No response to PAP authenticate-requests\n"));
 		u->us_clientstate = UPAPCS_BADAUTH;
 		auth_withpeer_fail(u->us_unit, PPP_PAP);
 		return;
@@ -299,11 +299,11 @@ static void upap_protrej(int unit)
 	upap_state *u = &upap[unit];
 	
 	if (u->us_clientstate == UPAPCS_AUTHREQ) {
-		ppp_trace(LOG_ERR, "PAP authentication failed due to protocol-reject\n");
+		UPAPDEBUG((LOG_ERR, "PAP authentication failed due to protocol-reject\n"));
 		auth_withpeer_fail(unit, PPP_PAP);
 	}
 	if (u->us_serverstate == UPAPSS_LISTEN) {
-		ppp_trace(LOG_ERR, "PAP authentication of peer failed (protocol-reject)\n");
+		UPAPDEBUG((LOG_ERR, "PAP authentication of peer failed (protocol-reject)\n"));
 		auth_peer_fail(unit, PPP_PAP);
 	}
 	upap_lowerdown(unit);
@@ -518,7 +518,7 @@ static void upap_rauthnak(
 	
 	u->us_clientstate = UPAPCS_BADAUTH;
 	
-	ppp_trace(LOG_ERR, "PAP authentication failed\n");
+	UPAPDEBUG((LOG_ERR, "PAP authentication failed\n"));
 	auth_withpeer_fail(u->us_unit, PPP_PAP);
 }
 

src/netif/ppp/ppp.c

@@ -153,7 +153,7 @@ typedef struct PPPControl_s {
     struct vjcompress vjComp;           /* Van Jabobsen compression header. */
 #endif
 
-    struct netif *netif;
+    struct netif netif;
 
     struct ppp_addrs addrs;
 
@@ -290,13 +290,12 @@ void pppInit(void)
     
 	memset(&ppp_settings, 0, sizeof(ppp_settings));
 	ppp_settings.usepeerdns = 1;
-	ppp_settings.refuse_chap = (CHAP_SUPPORT == 0);
+	pppSetAuth(PPPAUTHTYPE_NONE, NULL, NULL);
 
 	magicInit();
 
     for (i = 0; i < NUM_PPP; i++) {
         pppControl[i].openFlag = 0;
-        pppControl[i].netif = NULL;
 
 		subnetMask = htonl(0xffffff00);
     
@@ -307,25 +306,69 @@ void pppInit(void)
             (*protp->init)(i);
     }
 
-#ifdef LINK_STATS
+#if LINK_STATS
     /* Clear the statistics. */
     memset(&lwip_stats.link, 0, sizeof(lwip_stats.link));
 #endif
 }
 
-void pppSetAuth(const char *user, const char *passwd)
+void pppSetAuth(enum pppAuthType authType, const char *user, const char *passwd)
 {
-	if(user) {
-		strncpy(ppp_settings.user, user, sizeof(ppp_settings.user)-1);
-		ppp_settings.user[sizeof(ppp_settings.user)-1] = '\0';
-	} else
-		ppp_settings.user[0] = '\0';
+    switch(authType) {
+	case PPPAUTHTYPE_NONE:
+	default:
+#ifdef LWIP_PPP_STRICT_PAP_REJECT
+	    ppp_settings.refuse_pap = 1;
+#else
+	    /* some providers request pap and accept an empty login/pw */
+	    ppp_settings.refuse_pap = 0;
+#endif
+	    ppp_settings.refuse_chap = 1;
+	    break;
+	case PPPAUTHTYPE_ANY:
+/* Warning: Using PPPAUTHTYPE_ANY might have security consequences.
+ * RFC 1994 says:
+ *
+ * In practice, within or associated with each PPP server, there is a
+ * database which associates "user" names with authentication
+ * information ("secrets").  It is not anticipated that a particular
+ * named user would be authenticated by multiple methods.  This would
+ * make the user vulnerable to attacks which negotiate the least secure
+ * method from among a set (such as PAP rather than CHAP).  If the same
+ * secret was used, PAP would reveal the secret to be used later with
+ * CHAP.
+ *
+ * Instead, for each user name there should be an indication of exactly
+ * one method used to authenticate that user name.  If a user needs to
+ * make use of different authentication methods under different
+ * circumstances, then distinct user names SHOULD be employed, each of
+ * which identifies exactly one authentication method.
+ *
+ */
+	    ppp_settings.refuse_pap = 0;
+	    ppp_settings.refuse_chap = 0;
+	    break;
+	case PPPAUTHTYPE_PAP:
+	    ppp_settings.refuse_pap = 0;
+	    ppp_settings.refuse_chap = 1;
+	    break;
+	case PPPAUTHTYPE_CHAP:
+	    ppp_settings.refuse_pap = 1;
+	    ppp_settings.refuse_chap = 0;
+	    break;
+    }
 
-	if(passwd) {
-		strncpy(ppp_settings.passwd, passwd, sizeof(ppp_settings.passwd)-1);
-		ppp_settings.passwd[sizeof(ppp_settings.passwd)-1] = '\0';
-	} else
-		ppp_settings.passwd[0] = '\0';
+    if(user) {
+	strncpy(ppp_settings.user, user, sizeof(ppp_settings.user)-1);
+	ppp_settings.user[sizeof(ppp_settings.user)-1] = '\0';
+    } else
+	ppp_settings.user[0] = '\0';
+
+    if(passwd) {
+	strncpy(ppp_settings.passwd, passwd, sizeof(ppp_settings.passwd)-1);
+	ppp_settings.passwd[sizeof(ppp_settings.passwd)-1] = '\0';
+    } else
+	ppp_settings.passwd[0] = '\0';
 }
 
 /* Open a new PPP connection using the given I/O device.
@@ -438,7 +481,7 @@ static void nPut(PPPControl *pc, struct pbuf *nb)
 	    if((c = sio_write(pc->fd, b->payload, b->len)) != b->len) {
 		PPPDEBUG((LOG_WARNING,
 			    "PPP nPut: incomplete sio_write(%d,, %u) = %d\n", pc->fd, b->len, c));
-#ifdef LINK_STATS
+#if LINK_STATS
 		lwip_stats.link.err++;
 #endif /* LINK_STATS */
 		pc->lastXMit = 0; /* prepend PPP_FLAG to next packet */
@@ -447,7 +490,7 @@ static void nPut(PPPControl *pc, struct pbuf *nb)
 	}
 	pbuf_free(nb);
 
-#ifdef LINK_STATS
+#if LINK_STATS
 	lwip_stats.link.xmit++;
 #endif /* LINK_STATS */
 }
@@ -471,7 +514,7 @@ static struct pbuf *pppAppend(u_char c, struct pbuf *nb, ext_accm *outACCM)
 	if (tb) {
 	    nb->next = tb;
         }
-#ifdef LINK_STATS
+#if LINK_STATS
 	else {
 	    lwip_stats.link.memerr++;
 	}
@@ -493,14 +536,14 @@ static struct pbuf *pppAppend(u_char c, struct pbuf *nb, ext_accm *outACCM)
 /* Send a packet on the given connection. */
 static err_t pppifOutput(struct netif *netif, struct pbuf *pb, struct ip_addr *ipaddr)
 {
-	int pd = (int)netif->state;
-	u_short protocol = PPP_IP;
+    int pd = (int)netif->state;
+    u_short protocol = PPP_IP;
     PPPControl *pc = &pppControl[pd];
     u_int fcsOut = PPP_INITFCS;
     struct pbuf *headMB = NULL, *tailMB = NULL, *p;
     u_char c;
 
-	(void)ipaddr;
+    (void)ipaddr;
 
     /* Validate parameters. */
     /* We let any protocol value go through - it can't hurt us
@@ -508,7 +551,7 @@ static err_t pppifOutput(struct netif *netif, struct pbuf *pb, struct ip_addr *i
 	if (pd < 0 || pd >= NUM_PPP || !pc->openFlag || !pb) {
         PPPDEBUG((LOG_WARNING, "pppifOutput[%d]: bad parms prot=%d pb=%p\n",
                     pd, protocol, pb));
-#ifdef LINK_STATS
+#if LINK_STATS
 		lwip_stats.link.opterr++;
 		lwip_stats.link.drop++;
 #endif
@@ -518,7 +561,7 @@ static err_t pppifOutput(struct netif *netif, struct pbuf *pb, struct ip_addr *i
     /* Check that the link is up. */
 	if (lcp_phase[pd] == PHASE_DEAD) {
         PPPDEBUG((LOG_ERR, "pppifOutput[%d]: link not up\n", pd));
-#ifdef LINK_STATS
+#if LINK_STATS
 		lwip_stats.link.rterr++;
 		lwip_stats.link.drop++;
 #endif
@@ -529,7 +572,7 @@ static err_t pppifOutput(struct netif *netif, struct pbuf *pb, struct ip_addr *i
 	headMB = pbuf_alloc(PBUF_RAW, 0, PBUF_POOL);
     if (headMB == NULL) {
         PPPDEBUG((LOG_WARNING, "pppifOutput[%d]: first alloc fail\n", pd));
-#ifdef LINK_STATS
+#if LINK_STATS
 		lwip_stats.link.memerr++;
 		lwip_stats.link.drop++;
 #endif /* LINK_STATS */
@@ -556,7 +599,7 @@ static err_t pppifOutput(struct netif *netif, struct pbuf *pb, struct ip_addr *i
             break;
         default:
             PPPDEBUG((LOG_WARNING, "pppifOutput[%d]: bad IP packet\n", pd));
-#ifdef LINK_STATS
+#if LINK_STATS
 			lwip_stats.link.proterr++;
 			lwip_stats.link.drop++;
 #endif
@@ -618,7 +661,7 @@ static err_t pppifOutput(struct netif *netif, struct pbuf *pb, struct ip_addr *i
                     "pppifOutput[%d]: Alloc err - dropping proto=%d\n", 
                     pd, protocol));
         pbuf_free(headMB);
-#ifdef LINK_STATS
+#if LINK_STATS
 		lwip_stats.link.memerr++;
 		lwip_stats.link.drop++;
 #endif
@@ -707,7 +750,7 @@ int pppWrite(int pd, const u_char *s, int n)
     struct pbuf *headMB = NULL, *tailMB;
 	headMB = pbuf_alloc(PBUF_RAW, 0, PBUF_POOL);
     if (headMB == NULL) {
-#ifdef LINK_STATS
+#if LINK_STATS
 		lwip_stats.link.memerr++;
 		lwip_stats.link.proterr++;
 #endif /* LINK_STATS */
@@ -747,7 +790,7 @@ int pppWrite(int pd, const u_char *s, int n)
                 "pppWrite[%d]: Alloc err - dropping pbuf len=%d\n", pd, headMB->len));
 /*                "pppWrite[%d]: Alloc err - dropping %d:%.*H", pd, headMB->len, LWIP_MIN(headMB->len * 2, 40), headMB->payload)); */
 		pbuf_free(headMB);
-#ifdef LINK_STATS
+#if LINK_STATS
 		lwip_stats.link.memerr++;
 		lwip_stats.link.proterr++;
 #endif /* LINK_STATS */
@@ -959,12 +1002,10 @@ int sifup(int pd)
         st = 0;
         PPPDEBUG((LOG_WARNING, "sifup[%d]: bad parms\n", pd));
     } else {
-		if(pc->netif)
-			netif_remove(pc->netif);
-		pc->netif = netif_add(&pc->addrs.our_ipaddr, &pc->addrs.netmask, &pc->addrs.his_ipaddr, (void *)pd, pppifNetifInit, ip_input);
-		if(pc->netif) {
-        	pc->if_up = 1;
-        	pc->errCode = PPPERR_NONE;
+		netif_remove(&pc->netif);
+		if (netif_add(&pc->netif, &pc->addrs.our_ipaddr, &pc->addrs.netmask, &pc->addrs.his_ipaddr, (void *)pd, pppifNetifInit, ip_input)) {
+        		pc->if_up = 1;
+        		pc->errCode = PPPERR_NONE;
 
 			PPPDEBUG((LOG_DEBUG, "sifup: unit %d: linkStatusCB=%lx errCode=%d\n", pd, pc->linkStatusCB, pc->errCode));
 			if(pc->linkStatusCB)
@@ -1002,12 +1043,10 @@ int sifdown(int pd)
         PPPDEBUG((LOG_WARNING, "sifdown[%d]: bad parms\n", pd));
     } else {
         pc->if_up = 0;
-		if(pc->netif)
-			netif_remove(pc->netif);
-		pc->netif = NULL;
-		PPPDEBUG((LOG_DEBUG, "sifdown: unit %d: linkStatusCB=%lx errCode=%d\n", pd, pc->linkStatusCB, pc->errCode));
-		if(pc->linkStatusCB)
-			pc->linkStatusCB(pc->linkStatusCtx, PPPERR_CONNECT, NULL);
+	netif_remove(&pc->netif);
+	PPPDEBUG((LOG_DEBUG, "sifdown: unit %d: linkStatusCB=%lx errCode=%d\n", pd, pc->linkStatusCB, pc->errCode));
+	if(pc->linkStatusCB)
+		pc->linkStatusCB(pc->linkStatusCtx, PPPERR_CONNECT, NULL);
 	}
     return st;
 }
@@ -1082,7 +1121,7 @@ int sifdefaultroute(int pd, u32_t l, u32_t g)
         st = 0;
         PPPDEBUG((LOG_WARNING, "sifup[%d]: bad parms\n", pd));
     } else {
-		netif_set_default(pc->netif);
+		netif_set_default(&pc->netif);
     }
 
     /* TODO: check how PPP handled the netMask, previously not set by ipSetDefault */
@@ -1254,7 +1293,7 @@ static void pppInput(void *arg)
 
     pbuf_header(nb, -(int)sizeof(struct pppInputHeader));
 
-#ifdef LINK_STATS
+#if LINK_STATS
     lwip_stats.link.recv++;
 #endif /* LINK_STATS */
 
@@ -1280,7 +1319,7 @@ static void pppInput(void *arg)
          * pass the result to IP.
          */
         if (vj_uncompress_tcp(&nb, &pppControl[pd].vjComp) >= 0) {
-            pppControl[pd].netif->input(nb, pppControl[pd].netif);
+            pppControl[pd].netif.input(nb, &pppControl[pd].netif);
 			return;
         }
 	/* Something's wrong so drop it. */
@@ -1298,7 +1337,7 @@ static void pppInput(void *arg)
          * the packet to IP.
          */
         if (vj_uncompress_uncomp(nb, &pppControl[pd].vjComp) >= 0) {
-            pppControl[pd].netif->input(nb, pppControl[pd].netif);
+            pppControl[pd].netif.input(nb, &pppControl[pd].netif);
 			return;
         }
 	/* Something's wrong so drop it. */
@@ -1312,7 +1351,7 @@ static void pppInput(void *arg)
 	break;
     case PPP_IP:            /* Internet Protocol */
         PPPDEBUG((LOG_INFO, "pppInput[%d]: ip in pbuf len=%d\n", pd, nb->len));
-        pppControl[pd].netif->input(nb, pppControl[pd].netif);
+        pppControl[pd].netif.input(nb, &pppControl[pd].netif);
 		return;
     default:
 	{
@@ -1344,7 +1383,7 @@ static void pppInput(void *arg)
     }
 
 drop:
-#ifdef LINK_STATS
+#if LINK_STATS
     lwip_stats.link.drop++;
 #endif
 
@@ -1374,7 +1413,7 @@ static void pppDrop(PPPControl *pc)
     vj_uncompress_err(&pc->vjComp);
 #endif
 
-#ifdef LINK_STATS
+#if LINK_STATS
     lwip_stats.link.drop++;
 #endif /* LINK_STATS */
 }
@@ -1412,7 +1451,7 @@ static void pppInProc(int pd, u_char *s, int l)
                     PPPDEBUG((LOG_WARNING,
                                 "pppInProc[%d]: Dropping incomplete packet %d\n", 
                                 pd, pc->inState));
-#ifdef LINK_STATS
+#if LINK_STATS
 					lwip_stats.link.lenerr++;
 #endif
                     pppDrop(pc);
@@ -1422,7 +1461,7 @@ static void pppInProc(int pd, u_char *s, int l)
                     PPPDEBUG((LOG_INFO,
                                 "pppInProc[%d]: Dropping bad fcs 0x%04X proto=0x%04X\n", 
                                 pd, pc->inFCS, pc->inProtocol));
-#ifdef LINK_STATS
+#if LINK_STATS
 					lwip_stats.link.chkerr++;
 #endif
                     pppDrop(pc);
@@ -1436,14 +1475,12 @@ static void pppInProc(int pd, u_char *s, int l)
 
 			pc->inTail->tot_len = pc->inTail->len;
 			if (pc->inTail != pc->inHead) {
-			    pbuf_chain(pc->inHead, pc->inTail);
-			    pbuf_free(pc->inTail);
+			    pbuf_cat(pc->inHead, pc->inTail);
 			}
 		    } else {
 			pc->inTail->tot_len = pc->inTail->len;
 			if (pc->inTail != pc->inHead) {
-			    pbuf_chain(pc->inHead, pc->inTail);
-			    pbuf_free(pc->inTail);
+			    pbuf_cat(pc->inHead, pc->inTail);
 			}
 
 			pbuf_realloc(pc->inHead, pc->inHead->tot_len - 2);
@@ -1454,7 +1491,7 @@ static void pppInProc(int pd, u_char *s, int l)
                     	PPPDEBUG((LOG_ERR,
 				    "pppInProc[%d]: tcpip_callback() failed, dropping packet\n", pd));
 			pbuf_free(pc->inHead);
-#ifdef LINK_STATS
+#if LINK_STATS
 			lwip_stats.link.drop++;
 #endif
 		    }
@@ -1539,8 +1576,7 @@ static void pppInProc(int pd, u_char *s, int l)
 		    if(pc->inTail) {
 			pc->inTail->tot_len = pc->inTail->len;
 			if (pc->inTail != pc->inHead) {
-			    pbuf_chain(pc->inHead, pc->inTail);
-			    pbuf_free(pc->inTail);
+			    pbuf_cat(pc->inHead, pc->inTail);
 			}
 		    }
                     /* If we haven't started a packet, we need a packet header. */
@@ -1550,7 +1586,7 @@ static void pppInProc(int pd, u_char *s, int l)
                          * higher layers deal with it.  Continue processing
                          * the received pbuf chain in case a new packet starts. */
                         PPPDEBUG((LOG_ERR, "pppInProc[%d]: NO FREE MBUFS!\n", pd));
-#ifdef LINK_STATS
+#if LINK_STATS
 						lwip_stats.link.memerr++;
 #endif /* LINK_STATS */
                         pppDrop(pc);

src/netif/ppp/ppp.h

@@ -208,8 +208,11 @@ enum NPmode {
 #define BCMP(s0, s1, l)     memcmp((u_char *)(s0), (u_char *)(s1), (l))
 #define BCOPY(s, d, l)      memcpy((d), (s), (l))
 #define BZERO(s, n)         memset(s, 0, n)
-
+#if PPP_DEBUG
 #define PRINTMSG(m, l)  { m[l] = '\0'; ppp_trace(LOG_INFO, "Remote message: %s\n", m); }
+#else
+#define PRINTMSG(m, l)
+#endif
 
 /*
  * MAKEHEADER - Add PPP Header fields to a packet.
@@ -313,7 +316,7 @@ struct ppp_settings {
 	int  maxconnect;         /* Maximum connect time (seconds) */
 
 	char user[MAXNAMELEN + 1];/* Username for PAP */
-	char passwd[MAXNAMELEN + 1];           /* Password for PAP */
+	char passwd[MAXSECRETLEN + 1];           /* Password for PAP, secret for CHAP */
 	char our_name[MAXNAMELEN + 1];         /* Our name for authentication purposes */
 	char remote_name[MAXNAMELEN + 1];      /* Peer's name for authentication */
 };
@@ -340,7 +343,33 @@ extern struct protent *ppp_protocols[];/* Table of pointers to supported protoco
 /* Initialize the PPP subsystem. */
 void pppInit(void);
 
-void pppSetAuth(const char *user, const char *passwd);
+/* Warning: Using PPPAUTHTYPE_ANY might have security consequences.
+ * RFC 1994 says:
+ *
+ * In practice, within or associated with each PPP server, there is a
+ * database which associates "user" names with authentication
+ * information ("secrets").  It is not anticipated that a particular
+ * named user would be authenticated by multiple methods.  This would
+ * make the user vulnerable to attacks which negotiate the least secure
+ * method from among a set (such as PAP rather than CHAP).  If the same
+ * secret was used, PAP would reveal the secret to be used later with
+ * CHAP.
+ *
+ * Instead, for each user name there should be an indication of exactly
+ * one method used to authenticate that user name.  If a user needs to
+ * make use of different authentication methods under different
+ * circumstances, then distinct user names SHOULD be employed, each of
+ * which identifies exactly one authentication method.
+ *
+ */
+enum pppAuthType {
+    PPPAUTHTYPE_NONE,
+    PPPAUTHTYPE_ANY,
+    PPPAUTHTYPE_PAP,
+    PPPAUTHTYPE_CHAP
+};
+
+void pppSetAuth(enum pppAuthType authType, const char *user, const char *passwd);
 
 /*
  * Open a new PPP connection using the given I/O device.

src/netif/ppp/pppdebug.h

@@ -61,14 +61,13 @@ void ppp_trace(int level, const char *format,...);
 
 #if PPP_DEBUG > 0
 
-#define AUTHDEBUG(a) ppp_trace##a
-#define IPCPDEBUG(a) ppp_trace##a
-#define UPAPDEBUG(a) ppp_trace##a
-#define LCPDEBUG(a) ppp_trace##a
-#define FSMDEBUG(a) ppp_trace##a
-#define CHAPDEBUG(a) ppp_trace##a
-
-#define PPPDEBUG(a) ppp_trace##a
+#define AUTHDEBUG(a) ppp_trace a
+#define IPCPDEBUG(a) ppp_trace a
+#define UPAPDEBUG(a) ppp_trace a
+#define LCPDEBUG(a) ppp_trace a
+#define FSMDEBUG(a) ppp_trace a
+#define CHAPDEBUG(a) ppp_trace a
+#define PPPDEBUG(a) ppp_trace a
 
 #define TRACELCP 1
 

src/netif/ppp/randm.c

@@ -32,6 +32,7 @@
 *****************************************************************************/
 
 #include "ppp.h"
+#if PPP_SUPPORT > 0
 #include "md5.h"
 #include "randm.h"
 
@@ -237,4 +238,5 @@ u32_t avRandom()
 
 
 #endif /* MD5_SUPPORT */
+#endif /* PPP_SUPPORT */
 

src/netif/ppp/vj.c

@@ -35,7 +35,7 @@
 
 #if VJ_SUPPORT > 0
 
-#ifdef LINK_STATS
+#if LINK_STATS
 #define INCR(counter) ++comp->stats.counter
 #else
 #define INCR(counter)
@@ -611,8 +611,7 @@ int vj_uncompress_tcp(
 			*nb = NULL;
 			goto bad;
 		}
-		pbuf_chain(np, n0);
-		pbuf_free(n0);
+		pbuf_cat(np, n0);
 		n0 = np;
 	}
 	LWIP_ASSERT("n0->len >= cs->cs_hlen", n0->len >= cs->cs_hlen);

src/netif/ppp/vj.h

@@ -1,7 +1,7 @@
 /*
  * Definitions for tcp compression routines.
  *
- * $Id: vj.h,v 1.2 2003/06/02 11:12:56 jani 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.
@@ -136,7 +136,7 @@ struct vjcompress {
     u_short flags;
     u_char maxSlotIndex;
     u_char compressSlot;	/* Flag indicating OK to compress slot ID. */
-#ifdef LINK_STATS
+#if LINK_STATS
     struct vjstat stats;
 #endif
     struct cstate tstate[MAX_SLOTS];	/* xmit connection states */

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 
@@ -115,9 +115,7 @@ slipif_input( struct netif * netif )
   /* Received whole packet. */
   pbuf_realloc(q, recved);
   
-#ifdef LINK_STATS
-  ++lwip_stats.link.recv;
-#endif /* LINK_STATS */         
+  LINK_STATS_INC(link.recv);
   
   LWIP_DEBUGF(SLIP_DEBUG, ("slipif: Got packet\n"));
   return q;
@@ -141,16 +139,13 @@ slipif_input( struct netif * netif )
   LWIP_DEBUGF(SLIP_DEBUG, ("slipif_input: alloc\n"));
   p = pbuf_alloc(PBUF_LINK, PBUF_POOL_BUFSIZE, PBUF_POOL);
 
-#ifdef LINK_STATS           
   if (p == NULL) {
-    ++lwip_stats.link.drop;
+    LINK_STATS_INC(link.drop);
     LWIP_DEBUGF(SLIP_DEBUG, ("slipif_input: no new pbuf! (DROP)\n"));
   }
-#endif /* LINK_STATS */                  
   
   if (q != NULL) {
-    pbuf_chain(q, p);
-    pbuf_free(p);
+    pbuf_cat(q, p);
   } else {
     q = p;
   }
@@ -204,6 +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)