Merge pull request #422 from BernardXiong/master

Add lwIP/NAT, DHCP server and dlib(LIBC of IAR)

9 files changed, 1768 insertions, 5 deletions

diff --git a/components/net/lwip-1.4.1/src/core/ipv4/ip.c b/components/net/lwip-1.4.1/src/core/ipv4/ip.c
index 95d2db404c..17bcd3929a 100644
--- a/components/net/lwip-1.4.1/src/core/ipv4/ip.c
+++ b/components/net/lwip-1.4.1/src/core/ipv4/ip.c
@@ -312,6 +312,11 @@ ip_input(struct pbuf *p, struct netif *inp)
   int check_ip_src=1;
 #endif /* IP_ACCEPT_LINK_LAYER_ADDRESSING */
 
+#if IP_NAT
+  extern u8_t ip_nat_input(struct pbuf *p);
+  extern u8_t ip_nat_out(struct pbuf *p);
+#endif
+
   IP_STATS_INC(ip.recv);
   snmp_inc_ipinreceives();
 
@@ -487,15 +492,30 @@ ip_input(struct pbuf *p, struct netif *inp)
 
   /* packet not for us? */
   if (netif == NULL) {
+#if IP_FORWARD || IP_NAT
+    u8_t taken = 0;
+#endif /* IP_FORWARD || IP_NAT */
     /* packet not for us, route or discard */
     LWIP_DEBUGF(IP_DEBUG | LWIP_DBG_TRACE, ("ip_input: packet not for us.\n"));
-#if IP_FORWARD
+#if IP_FORWARD || IP_NAT
     /* non-broadcast packet? */
-    if (!ip_addr_isbroadcast(&current_iphdr_dest, inp)) {
-      /* try to forward IP packet on (other) interfaces */
-      ip_forward(p, iphdr, inp);
-    } else
+    if (!ip_addr_isbroadcast(&(iphdr->dest), inp)) {
+#if IP_NAT
+      /* check if we want to perform NAT with this packet. */
+      taken = ip_nat_out(p);
+      if (!taken)
+#endif /* IP_NAT */
+      {
+#if IP_FORWARD
+        /* try to forward IP packet on (other) interfaces */
+        if (ip_forward(p, iphdr, inp) != NULL) {
+          taken = 1;
+        }
 #endif /* IP_FORWARD */
+      }
+    }
+    if (!taken)
+#endif /* IP_FORWARD || IP_NAT */
     {
       snmp_inc_ipinaddrerrors();
       snmp_inc_ipindiscards();
@@ -553,6 +573,13 @@ ip_input(struct pbuf *p, struct netif *inp)
   current_netif = inp;
   current_header = iphdr;
 
+#if IP_NAT
+  if (!ip_addr_isbroadcast(&(iphdr->dest), inp) &&
+      (ip_nat_input(p) != 0)) {
+     LWIP_DEBUGF(IP_DEBUG, ("ip_input: packet consumed by nat layer\n"));
+  } else
+#endif /* IP_NAT */
+
 #if LWIP_RAW
   /* raw input did not eat the packet? */
   if (raw_input(p, inp) == 0)
diff --git a/components/net/lwip-1.4.1/src/lwipopts.h b/components/net/lwip-1.4.1/src/lwipopts.h
index bff172e280..cb7def8ae6 100644
--- a/components/net/lwip-1.4.1/src/lwipopts.h
+++ b/components/net/lwip-1.4.1/src/lwipopts.h
@@ -161,6 +161,12 @@
  */
 #define SYS_LIGHTWEIGHT_PROT        (NO_SYS==0)
 
+#ifdef LWIP_USING_NAT
+#define IP_NAT                      1
+#else
+#define IP_NAT                      0
+#endif
+
 /* ---------- TCP options ---------- */
 #ifdef RT_LWIP_TCP
 #define LWIP_TCP                    1
diff --git a/components/net/lwip_dhcpd/SConscript b/components/net/lwip_dhcpd/SConscript
new file mode 100644
index 0000000000..d5704c8c8b
--- /dev/null
+++ b/components/net/lwip_dhcpd/SConscript
@@ -0,0 +1,10 @@
+from building import *
+
+cwd = GetCurrentDir()
+src = Glob('*.c')
+
+CPPPATH = [cwd]
+
+group = DefineGroup('LwIP', src, depend = ['RT_USING_LWIP', 'LWIP_USING_DHCPD'], CPPPATH = CPPPATH)
+
+Return('group')
diff --git a/components/net/lwip_dhcpd/dhcp_server.c b/components/net/lwip_dhcpd/dhcp_server.c
new file mode 100644
index 0000000000..47547a598b
--- /dev/null
+++ b/components/net/lwip_dhcpd/dhcp_server.c
@@ -0,0 +1,398 @@
+/*
+ * File      : dhcp_server.c
+ *             A simple DHCP server implementation
+ *
+ * This file is part of RT-Thread RTOS
+ * COPYRIGHT (C) 2013-2015, Shanghai Real-Thread Technology Co., Ltd
+ * http://www.rt-thread.com
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License as published by
+ *  the Free Software Foundation; either version 2 of the License, or
+ *  (at your option) any later version.
+ *
+ *  This program is distributed in the hope that it will be useful,
+ *  but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *  GNU General Public License for more details.
+ *
+ *  You should have received a copy of the GNU General Public License along
+ *  with this program; if not, write to the Free Software Foundation, Inc.,
+ *  51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
+ *
+ * Change Logs:
+ * Date           Author       Notes
+ * 2013-01-30     aozima       the first version
+ * 2013-08-08     aozima       support different network segments.
+ * 2015-01-30     bernard      release to RT-Thread RTOS.
+ */
+
+#include <stdio.h>
+#include <stdint.h>
+#include <rtthread.h>
+
+#include <lwip/opt.h>
+#include <lwip/sockets.h>
+#include <lwip/inet_chksum.h>
+#include <netif/etharp.h>
+#include <netif/ethernetif.h>
+#include <lwip/ip.h>
+
+/* DHCP server option */
+
+/* allocated client ip range */
+#ifndef DHCPD_CLIENT_IP_MIN
+#define DHCPD_CLIENT_IP_MIN     2
+#endif
+#ifndef DHCPD_CLIENT_IP_MAX
+#define DHCPD_CLIENT_IP_MAX     254
+#endif
+
+/* the DHCP server address */
+#ifndef DHCPD_SERVER_IPADDR0
+#define DHCPD_SERVER_IPADDR0      192UL
+#define DHCPD_SERVER_IPADDR1      168UL
+#define DHCPD_SERVER_IPADDR2      169UL
+#define DHCPD_SERVER_IPADDR3      1UL
+#endif
+
+//#define DHCP_DEBUG_PRINTF
+
+#ifdef  DHCP_DEBUG_PRINTF
+#define DEBUG_PRINTF        rt_kprintf("[DHCP] "); rt_kprintf
+#else
+#define DEBUG_PRINTF(...)
+#endif /* DHCP_DEBUG_PRINTF */
+
+/* we need some routines in the DHCP of lwIP */
+#undef  LWIP_DHCP
+#define LWIP_DHCP   1
+#include <lwip/dhcp.h>
+
+/* buffer size for receive DHCP packet */
+#define BUFSZ               1024
+
+static uint8_t next_client_ip = DHCPD_CLIENT_IP_MIN;
+static rt_err_t _low_level_dhcp_send(struct netif *netif,
+                                     const void *buffer,
+                                     rt_size_t size)
+{
+    struct pbuf *p;
+    struct eth_hdr *ethhdr;
+    struct ip_hdr *iphdr;
+    struct udp_hdr *udphdr;
+
+    p = pbuf_alloc(PBUF_LINK,
+                   SIZEOF_ETH_HDR + sizeof(struct ip_hdr)
+                   + sizeof(struct udp_hdr) + size,
+                   PBUF_RAM);
+    if (p == RT_NULL) return -RT_ENOMEM;
+
+    ethhdr = (struct eth_hdr *)p->payload;
+    iphdr  = (struct ip_hdr *)((char *)ethhdr + SIZEOF_ETH_HDR);
+    udphdr = (struct udp_hdr *)((char *)iphdr + sizeof(struct ip_hdr));
+
+    ETHADDR32_COPY(&ethhdr->dest, (struct eth_addr *)&ethbroadcast);
+    ETHADDR16_COPY(&ethhdr->src, netif->hwaddr);
+    ethhdr->type = PP_HTONS(ETHTYPE_IP);
+
+    iphdr->src.addr  = 0x00000000; /* src: 0.0.0.0 */
+    iphdr->dest.addr = 0xFFFFFFFF; /* src: 255.255.255.255 */
+
+    IPH_VHL_SET(iphdr, 4, IP_HLEN / 4);
+    IPH_TOS_SET(iphdr, 0x00);
+    IPH_LEN_SET(iphdr, htons(IP_HLEN + sizeof(struct udp_hdr) + size));
+    IPH_ID_SET(iphdr, htons(2));
+    IPH_OFFSET_SET(iphdr, 0);
+    IPH_TTL_SET(iphdr, 255);
+    IPH_PROTO_SET(iphdr, IP_PROTO_UDP);
+    IPH_CHKSUM_SET(iphdr, 0);
+    IPH_CHKSUM_SET(iphdr, inet_chksum(iphdr, IP_HLEN));
+
+    udphdr->src = htons(DHCP_SERVER_PORT);
+    udphdr->dest = htons(DHCP_CLIENT_PORT);
+    udphdr->len = htons(sizeof(struct udp_hdr) + size);
+    udphdr->chksum = 0;
+
+    memcpy((char *)udphdr + sizeof(struct udp_hdr),
+           buffer, size);
+
+    return netif->linkoutput(netif, p);
+}
+
+static void dhcpd_thread_entry(void *parameter)
+{
+    struct netif *netif = RT_NULL;
+    int sock;
+    int bytes_read;
+    char *recv_data;
+    rt_uint32_t addr_len;
+    struct sockaddr_in server_addr, client_addr;
+    struct dhcp_msg *msg;
+    int optval = 1;
+
+    /* get ethernet interface. */
+    netif = (struct netif*) parameter;
+    RT_ASSERT(netif != RT_NULL);
+
+    /* our DHCP server information */
+    DEBUG_PRINTF("DHCP server IP: %d.%d.%d.%d  client IP: %d.%d.%d.%d-%d\n",
+                 DHCPD_SERVER_IPADDR0, DHCPD_SERVER_IPADDR1,
+                 DHCPD_SERVER_IPADDR2, DHCPD_SERVER_IPADDR3,
+                 DHCPD_SERVER_IPADDR0, DHCPD_SERVER_IPADDR1,
+                 DHCPD_SERVER_IPADDR2, DHCPD_CLIENT_IP_MIN, DHCPD_CLIENT_IP_MAX);
+
+    /* allocate buffer for receive */
+    recv_data = rt_malloc(BUFSZ);
+    if (recv_data == RT_NULL)
+    {
+        /* No memory */
+        DEBUG_PRINTF("Out of memory\n");
+        return;
+    }
+
+    /* create a socket with UDP */
+    if ((sock = socket(AF_INET, SOCK_DGRAM, 0)) == -1)
+    {
+        DEBUG_PRINTF("create socket failed, errno = %d\n", errno);
+        rt_free(recv_data);
+        return;
+    }
+
+    /* set to receive broadcast packet */
+    setsockopt(sock, SOL_SOCKET, SO_BROADCAST, &optval, sizeof(optval));
+
+    /* initialize server address */
+    server_addr.sin_family = AF_INET;
+    server_addr.sin_port = htons(DHCP_SERVER_PORT);
+    server_addr.sin_addr.s_addr = INADDR_ANY;
+    rt_memset(&(server_addr.sin_zero), 0, sizeof(server_addr.sin_zero));
+
+    /* bind socket to the server address */
+    if (bind(sock, (struct sockaddr *)&server_addr,
+             sizeof(struct sockaddr)) == -1)
+    {
+        /* bind failed. */
+        DEBUG_PRINTF("bind server address failed, errno=%d\n", errno);
+        rt_free(recv_data);
+        return;
+    }
+
+    addr_len = sizeof(struct sockaddr);
+    DEBUG_PRINTF("DHCP server listen on port %d...\n", DHCP_SERVER_PORT);
+
+    while (1)
+    {
+        bytes_read = recvfrom(sock, recv_data, BUFSZ - 1, 0,
+                              (struct sockaddr *)&client_addr, &addr_len);
+        if (bytes_read < DHCP_MSG_LEN)
+        {
+            DEBUG_PRINTF("packet too short, wait for next!\n");
+            continue;
+        }
+
+        msg = (struct dhcp_msg *)recv_data;
+        /* check message type to make sure we can handle it */
+        if ((msg->op != DHCP_BOOTREQUEST) || (msg->cookie != PP_HTONL(DHCP_MAGIC_COOKIE)))
+        {
+            continue;
+        }
+
+        /* handler. */
+        {
+            uint8_t *dhcp_opt;
+            uint8_t option;
+            uint8_t length;
+
+            uint8_t message_type = 0;
+            uint8_t finished = 0;
+            uint32_t request_ip  = 0;
+
+            dhcp_opt = (uint8_t *)msg + DHCP_OPTIONS_OFS;
+            while (finished == 0)
+            {
+                option = *dhcp_opt;
+                length = *(dhcp_opt + 1);
+
+                switch (option)
+                {
+                case DHCP_OPTION_REQUESTED_IP:
+                    request_ip = *(dhcp_opt + 2) << 24 | *(dhcp_opt + 3) << 16
+                                 | *(dhcp_opt + 4) << 8 | *(dhcp_opt + 5);
+                    break;
+
+                case DHCP_OPTION_END:
+                    finished = 1;
+                    break;
+
+                case DHCP_OPTION_MESSAGE_TYPE:
+                    message_type = *(dhcp_opt + 2);
+                    break;
+
+                default:
+                    break;
+                } /* switch(option) */
+
+                dhcp_opt += (2 + length);
+            }
+
+            /* reply. */
+            dhcp_opt = (uint8_t *)msg + DHCP_OPTIONS_OFS;
+
+            /* check. */
+            if (request_ip)
+            {
+                uint32_t client_ip = DHCPD_SERVER_IPADDR0 << 24 | DHCPD_SERVER_IPADDR1 << 16
+                                     | DHCPD_SERVER_IPADDR2 << 8 | (next_client_ip);
+
+                if (request_ip != client_ip)
+                {
+                    *dhcp_opt++ = DHCP_OPTION_MESSAGE_TYPE;
+                    *dhcp_opt++ = DHCP_OPTION_MESSAGE_TYPE_LEN;
+                    *dhcp_opt++ = DHCP_NAK;
+                    *dhcp_opt++ = DHCP_OPTION_END;
+
+                    DEBUG_PRINTF("requested IP invalid, reply DHCP_NAK\n");
+                    if (netif != RT_NULL)
+                    {
+                        int send_byte = (dhcp_opt - (uint8_t *)msg);
+                        _low_level_dhcp_send(netif, msg, send_byte);
+                        DEBUG_PRINTF("DHCP server send %d byte\n", send_byte);
+                    }
+                    next_client_ip++;
+                    if (next_client_ip > DHCPD_CLIENT_IP_MAX)
+                        next_client_ip = DHCPD_CLIENT_IP_MIN;
+                    continue;
+                }
+            }
+
+            if (message_type == DHCP_DISCOVER)
+            {
+                DEBUG_PRINTF("request DHCP_DISCOVER\n");
+                DEBUG_PRINTF("reply   DHCP_OFFER\n");
+
+                // DHCP_OPTION_MESSAGE_TYPE
+                *dhcp_opt++ = DHCP_OPTION_MESSAGE_TYPE;
+                *dhcp_opt++ = DHCP_OPTION_MESSAGE_TYPE_LEN;
+                *dhcp_opt++ = DHCP_OFFER;
+
+                // DHCP_OPTION_SERVER_ID
+                *dhcp_opt++ = DHCP_OPTION_SERVER_ID;
+                *dhcp_opt++ = 4;
+                *dhcp_opt++ = DHCPD_SERVER_IPADDR0;
+                *dhcp_opt++ = DHCPD_SERVER_IPADDR1;
+                *dhcp_opt++ = DHCPD_SERVER_IPADDR2;
+                *dhcp_opt++ = DHCPD_SERVER_IPADDR3;
+
+                // DHCP_OPTION_LEASE_TIME
+                *dhcp_opt++ = DHCP_OPTION_LEASE_TIME;
+                *dhcp_opt++ = 4;
+                *dhcp_opt++ = 0x00;
+                *dhcp_opt++ = 0x01;
+                *dhcp_opt++ = 0x51;
+                *dhcp_opt++ = 0x80;
+            }
+            else if (message_type == DHCP_REQUEST)
+            {
+                DEBUG_PRINTF("request DHCP_REQUEST\n");
+                DEBUG_PRINTF("reply   DHCP_ACK\n");
+
+                // DHCP_OPTION_MESSAGE_TYPE
+                *dhcp_opt++ = DHCP_OPTION_MESSAGE_TYPE;
+                *dhcp_opt++ = DHCP_OPTION_MESSAGE_TYPE_LEN;
+                *dhcp_opt++ = DHCP_ACK;
+
+                // DHCP_OPTION_SERVER_ID
+                *dhcp_opt++ = DHCP_OPTION_SERVER_ID;
+                *dhcp_opt++ = 4;
+                *dhcp_opt++ = DHCPD_SERVER_IPADDR0;
+                *dhcp_opt++ = DHCPD_SERVER_IPADDR1;
+                *dhcp_opt++ = DHCPD_SERVER_IPADDR2;
+                *dhcp_opt++ = DHCPD_SERVER_IPADDR3;
+
+                // DHCP_OPTION_SUBNET_MASK
+                *dhcp_opt++ = DHCP_OPTION_SUBNET_MASK;
+                *dhcp_opt++ = 4;
+                *dhcp_opt++ = 0xFF;
+                *dhcp_opt++ = 0xFF;
+                *dhcp_opt++ = 0xFF;
+                *dhcp_opt++ = 0x00;
+
+#ifdef DHCPD_USING_ROUTER
+                // DHCP_OPTION_ROUTER
+                *dhcp_opt++ = DHCP_OPTION_ROUTER;
+                *dhcp_opt++ = 4;
+                *dhcp_opt++ = DHCPD_SERVER_IPADDR0;
+                *dhcp_opt++ = DHCPD_SERVER_IPADDR1;
+                *dhcp_opt++ = DHCPD_SERVER_IPADDR2;
+                *dhcp_opt++ = 1;
+#endif
+
+                // DHCP_OPTION_DNS_SERVER, use the default DNS server address in lwIP
+                *dhcp_opt++ = DHCP_OPTION_DNS_SERVER;
+                *dhcp_opt++ = 4;
+                *dhcp_opt++ = 208;
+                *dhcp_opt++ = 67;
+                *dhcp_opt++ = 222;
+                *dhcp_opt++ = 222;
+
+                // DHCP_OPTION_LEASE_TIME
+                *dhcp_opt++ = DHCP_OPTION_LEASE_TIME;
+                *dhcp_opt++ = 4;
+                *dhcp_opt++ = 0x00;
+                *dhcp_opt++ = 0x01;
+                *dhcp_opt++ = 0x51;
+                *dhcp_opt++ = 0x80;
+            }
+            else
+            {
+                DEBUG_PRINTF("un handle message:%d\n", message_type);
+            }
+
+            // append DHCP_OPTION_END
+            *dhcp_opt++ = DHCP_OPTION_END;
+
+            /* send reply. */
+            if ((message_type == DHCP_DISCOVER) || (message_type == DHCP_REQUEST))
+            {
+                msg->op = DHCP_BOOTREPLY;
+                IP4_ADDR(&msg->yiaddr,
+                         DHCPD_SERVER_IPADDR0, DHCPD_SERVER_IPADDR1,
+                         DHCPD_SERVER_IPADDR2, next_client_ip);
+
+                client_addr.sin_addr.s_addr = INADDR_BROADCAST;
+
+                if (netif != RT_NULL)
+                {
+                    int send_byte = (dhcp_opt - (uint8_t *)msg);
+                    _low_level_dhcp_send(netif, msg, send_byte);
+                    DEBUG_PRINTF("DHCP server send %d byte\n", send_byte);
+                }
+            }
+        } /* handler. */
+    }
+}
+
+void dhcpd_start(char* netif_name)
+{
+    rt_thread_t thread;
+    struct netif *netif = RT_NULL;
+
+    /* find ethernet interface. */
+    netif = netif_find(netif_name);
+    if (netif == RT_NULL)
+    {
+        DEBUG_PRINTF("Not found network interface:%s\n", netif_name);
+    }
+
+    thread = rt_thread_create("dhcpd",
+                              dhcpd_thread_entry, netif,
+                              1024,
+                              RT_THREAD_PRIORITY_MAX - 3,
+                              2);
+    if (thread != RT_NULL)
+    {
+        rt_thread_startup(thread);
+    }
+}
+
diff --git a/components/net/lwip_dhcpd/dhcp_server.h b/components/net/lwip_dhcpd/dhcp_server.h
new file mode 100644
index 0000000000..4bd1163706
--- /dev/null
+++ b/components/net/lwip_dhcpd/dhcp_server.h
@@ -0,0 +1,36 @@
+/*
+ * File      : dhcp_server.h
+ *             A simple DHCP server implementation
+ *
+ * This file is part of RT-Thread RTOS
+ * COPYRIGHT (C) 2013-2015, Shanghai Real-Thread Technology Co., Ltd
+ * http://www.rt-thread.com
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License as published by
+ *  the Free Software Foundation; either version 2 of the License, or
+ *  (at your option) any later version.
+ *
+ *  This program is distributed in the hope that it will be useful,
+ *  but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *  GNU General Public License for more details.
+ *
+ *  You should have received a copy of the GNU General Public License along
+ *  with this program; if not, write to the Free Software Foundation, Inc.,
+ *  51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
+ *
+ * Change Logs:
+ * Date           Author       Notes
+ * 2013-01-30     aozima       the first version
+ * 2013-08-08     aozima       support different network segments.
+ * 2015-01-30     bernard      release to RT-Thread RTOS.
+ */
+
+#ifndef DHCPV4_SERVER_H__
+#define DHCPV4_SERVER_H__
+
+void dhcpd_start(char* netif_name);
+
+#endif
+
diff --git a/components/net/lwip_nat/README.md b/components/net/lwip_nat/README.md
new file mode 100644
index 0000000000..9672a11b54
--- /dev/null
+++ b/components/net/lwip_nat/README.md
@@ -0,0 +1,19 @@
+lwIP NAT componenent
+
+If you want to use lwIP NAT componenent, please define LWIP_USING_NAT in rtconfig.h. 
+
+In this case the network 213.129.231.168/29 is nat'ed when packets are sent to the 
+destination network 10.0.0.0/24 (untypical example - most users will have the other 
+way around).
+
+Use following code to add a NAT entry: 
+
+  ip_nat_entry_t nat_entry;
+ 
+  nat_entry.out_if = (struct netif *)&emac_if1;
+  nat_entry.in_if = (struct netif *)&emac_if2;
+  IP4_ADDR(&nat_entry.source_net, 213, 129, 231, 168);
+  IP4_ADDR(&nat_entry.source_netmask, 255, 255, 255, 248);
+  IP4_ADDR(&nat_entry.dest_net, 10, 0, 0, 0);
+  IP4_ADDR(&nat_entry.source_netmask, 255, 0, 0, 0);
+  ip_nat_add(&_nat_entry);
diff --git a/components/net/lwip_nat/SConscript b/components/net/lwip_nat/SConscript
new file mode 100644
index 0000000000..ce684f677a
--- /dev/null
+++ b/components/net/lwip_nat/SConscript
@@ -0,0 +1,10 @@
+from building import *
+
+cwd = GetCurrentDir()
+src = Glob('*.c')
+
+CPPPATH = [cwd]
+
+group = DefineGroup('LwIP', src, depend = ['RT_USING_LWIP', 'LWIP_USING_NAT'], CPPPATH = CPPPATH)
+
+Return('group')
diff --git a/components/net/lwip_nat/ipv4_nat.c b/components/net/lwip_nat/ipv4_nat.c
new file mode 100644
index 0000000000..13c1877f2c
--- /dev/null
+++ b/components/net/lwip_nat/ipv4_nat.c
@@ -0,0 +1,1154 @@
+/**
+ * NAT - NAT implementation for lwIP supporting TCP/UDP and ICMP.
+ * Copyright (c) 2009 Christian Walter, ?Embedded Solutions, Vienna 2009.
+ *
+ * Copyright (c) 2010 lwIP project ;-)
+ * 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.
+ */
+
+/*
+ * File      : ipv4_nat.c
+ * This file is part of RT-Thread RTOS
+ * COPYRIGHT (C) 2015, RT-Thread Development Team
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License as published by
+ *  the Free Software Foundation; either version 2 of the License, or
+ *  (at your option) any later version.
+ *
+ *  This program is distributed in the hope that it will be useful,
+ *  but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *  GNU General Public License for more details.
+ *
+ *  You should have received a copy of the GNU General Public License along
+ *  with this program; if not, write to the Free Software Foundation, Inc.,
+ *  51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
+ *
+ * Change Logs:
+ * Date           Author       Notes
+ * 2015-01-26     Hichard      porting to RT-Thread
+ * 2015-01-27     Bernard      code cleanup for lwIP in RT-Thread
+ */
+
+/*
+ * TODOS:
+ *  - we should decide if we want to use static tables for NAT or a linked
+ *    list.
+ *  - we should allocate icmp ping id if multiple clients are sending
+ *    ping requests.
+ *  - maybe we could hash the identifiers for TCP, ICMP and UDP and use
+ *    a single table structure. This would reduce the code amount although
+ *    it will cost performance.
+ *  - NAT code must check for broadcast addresses and NOT forward
+ *    them.
+ *
+ *  - netif_remove must notify NAT code when a NAT'ed interface is removed
+ *  - allocate NAT entries from a new memp pool instead of the heap
+ *  - let ttl be ticks, not seconds
+ *
+ * HOWTO USE:
+ *
+ * Shows how to create NAT between a PPP interface and an internal NIC.
+ * In this case the network 213.129.231.168/29 is nat'ed when packets
+ * are sent to the destination network 10.0.0.0/24 (untypical example -
+ * most users will have the other way around).
+ *
+ * Step 1) Execute when network interfaces are ready.
+ *
+ * new_nat_entry.out_if = (struct netif *)&PPP_IF;
+ * new_nat_entry.in_if = (struct netif *)&EMAC_if;
+ * IP4_ADDR(&new_nat_entry.source_net, 213, 129, 231, 168);
+ * IP4_ADDR(&new_nat_entry.source_netmask, 255, 255, 255, 248);
+ * IP4_ADDR(&new_nat_entry.dest_net, 10, 0, 0, 0);
+ * IP4_ADDR(&new_nat_entry.source_netmask, 255, 0, 0, 0);
+ * ip_nat_add(&new_nat_entry);
+ */
+
+#include "ipv4_nat.h"
+#include "lwip/opt.h"
+
+#ifdef LWIP_USING_NAT
+
+#include "lwip/ip.h"
+#include "lwip/inet.h"
+#include "lwip/netif.h"
+#include "lwip/ip_addr.h"
+#include "lwip/icmp.h"
+#include "lwip/tcp_impl.h"
+#include "lwip/udp.h"
+#include "lwip/mem.h"
+#include "lwip/sys.h"
+#include "lwip/timers.h"
+#include "netif/etharp.h"
+
+#include <limits.h>
+#include <string.h>
+
+/** Define this to enable debug output of this module */
+#ifndef LWIP_NAT_DEBUG
+#define LWIP_NAT_DEBUG      LWIP_DBG_OFF
+#endif
+
+#define LWIP_NAT_TTL_INFINITE                    (INT_MAX)
+#define LWIP_NAT_DEFAULT_TTL_SECONDS             (128)
+#define LWIP_NAT_FORWARD_HEADER_SIZE_MIN         (sizeof(struct eth_hdr))
+
+#define LWIP_NAT_DEFAULT_STATE_TABLES_ICMP       (4)
+#define LWIP_NAT_DEFAULT_STATE_TABLES_TCP        (32)
+#define LWIP_NAT_DEFAULT_STATE_TABLES_UDP        (32)
+
+#define LWIP_NAT_DEFAULT_TCP_SOURCE_PORT         (40000)
+#define LWIP_NAT_DEFAULT_UDP_SOURCE_PORT         (40000)
+
+#define IPNAT_ENTRY_RESET(x) do { \
+  (x)->ttl = 0; \
+} while(0)
+
+typedef struct ip_nat_conf
+{
+  struct ip_nat_conf *next;
+  ip_nat_entry_t      entry;
+} ip_nat_conf_t;
+
+typedef struct ip_nat_entry_common
+{
+  s32_t           ttl; /* @todo: do we really need this to be signed?? */
+  ip_addr_t       source;
+  ip_addr_t       dest;
+  ip_nat_conf_t   *cfg;
+} ip_nat_entry_common_t;
+
+typedef struct ip_nat_entries_icmp
+{
+  ip_nat_entry_common_t common;
+  u16_t                 id;
+  u16_t                 seqno;
+} ip_nat_entries_icmp_t;
+
+typedef struct ip_nat_entries_tcp
+{
+  ip_nat_entry_common_t common;
+  u16_t                 nport;
+  u16_t                 sport;
+  u16_t                 dport;
+} ip_nat_entries_tcp_t;
+
+typedef struct ip_nat_entries_udp
+{
+  ip_nat_entry_common_t common;
+  u16_t                 nport;
+  u16_t                 sport;
+  u16_t                 dport;
+} ip_nat_entries_udp_t;
+
+typedef union u_nat_entry
+{
+  ip_nat_entry_common_t *cmn;
+  ip_nat_entries_tcp_t  *tcp;
+  ip_nat_entries_icmp_t *icmp;
+  ip_nat_entries_udp_t  *udp;
+} nat_entry_t;
+
+static ip_nat_conf_t *ip_nat_cfg = NULL;
+static ip_nat_entries_icmp_t ip_nat_icmp_table[LWIP_NAT_DEFAULT_STATE_TABLES_ICMP];
+static ip_nat_entries_tcp_t ip_nat_tcp_table[LWIP_NAT_DEFAULT_STATE_TABLES_TCP];
+static ip_nat_entries_udp_t ip_nat_udp_table[LWIP_NAT_DEFAULT_STATE_TABLES_UDP];
+
+/* ----------------------- Static functions (COMMON) --------------------*/
+static void     ip_nat_chksum_adjust(u8_t *chksum, const u8_t *optr, s16_t olen, const u8_t *nptr, s16_t nlen);
+static void     ip_nat_cmn_init(ip_nat_conf_t *nat_config, const struct ip_hdr *iphdr,
+                                 ip_nat_entry_common_t *nat_entry);
+static ip_nat_conf_t *ip_nat_shallnat(const struct ip_hdr *iphdr);
+static void     ip_nat_reset_state(ip_nat_conf_t *cfg);
+
+/* ----------------------- Static functions (DEBUG) ---------------------*/
+#if defined(LWIP_DEBUG) && (LWIP_NAT_DEBUG & LWIP_DBG_ON)
+static void     ip_nat_dbg_dump(const char *msg, const struct ip_hdr *iphdr);
+static void     ip_nat_dbg_dump_ip(const ip_addr_t *addr);
+static void     ip_nat_dbg_dump_icmp_nat_entry(const char *msg, const ip_nat_entries_icmp_t *nat_entry);
+static void     ip_nat_dbg_dump_tcp_nat_entry(const char *msg, const ip_nat_entries_tcp_t *nat_entry);
+static void     ip_nat_dbg_dump_udp_nat_entry(const char *msg, const ip_nat_entries_udp_t *nat_entry);
+static void     ip_nat_dbg_dump_init(ip_nat_conf_t *ip_nat_cfg_new);
+static void     ip_nat_dbg_dump_remove(ip_nat_conf_t *cur);
+#else /* defined(LWIP_DEBUG) && (LWIP_NAT_DEBUG & LWIP_DBG_ON) */
+#define ip_nat_dbg_dump(msg, iphdr)
+#define ip_nat_dbg_dump_ip(addr)
+#define ip_nat_dbg_dump_icmp_nat_entry(msg, nat_entry)
+#define ip_nat_dbg_dump_tcp_nat_entry(msg, nat_entry)
+#define ip_nat_dbg_dump_udp_nat_entry(msg, nat_entry)
+#define ip_nat_dbg_dump_init(ip_nat_cfg_new)
+#define ip_nat_dbg_dump_remove(cur)
+#endif /* defined(LWIP_DEBUG) && (LWIP_NAT_DEBUG & LWIP_DBG_ON) */
+
+/* ----------------------- Static functions (TCP) -----------------------*/
+static ip_nat_entries_tcp_t *ip_nat_tcp_lookup_incoming(const struct ip_hdr *iphdr, const struct tcp_hdr *tcphdr);
+static ip_nat_entries_tcp_t *ip_nat_tcp_lookup_outgoing(ip_nat_conf_t *nat_config,
+                                                         const struct ip_hdr *iphdr, const struct tcp_hdr *tcphdr,
+                                                         u8_t allocate);
+
+/* ----------------------- Static functions (UDP) -----------------------*/
+static ip_nat_entries_udp_t *ip_nat_udp_lookup_incoming(const struct ip_hdr *iphdr, const struct udp_hdr *udphdr);
+static ip_nat_entries_udp_t *ip_nat_udp_lookup_outgoing(ip_nat_conf_t *nat_config,
+                                                         const struct ip_hdr *iphdr, const struct udp_hdr *udphdr,
+                                                         u8_t allocate);
+
+/**
+ * Timer callback function that calls ip_nat_tmr() and reschedules itself.
+ *
+ * @param arg unused argument
+ */
+static void
+nat_timer(void *arg)
+{
+  LWIP_UNUSED_ARG(arg);
+  LWIP_DEBUGF(TIMERS_DEBUG, ("tcpip: nat_timer()\n"));
+
+  ip_nat_tmr();
+  sys_timeout(LWIP_NAT_TMR_INTERVAL_SEC, nat_timer, NULL);
+}
+
+/** Initialize this module */
+void
+ip_nat_init(void)
+{
+  int i;
+  extern void lwip_ip_input_set_hook(int (*hook)(struct pbuf *p, struct netif *inp));
+
+  /* @todo: this can be omitted since we trust static variables
+            to be initialized to zero */
+  for (i = 0; i < LWIP_NAT_DEFAULT_STATE_TABLES_ICMP; i++) {
+    IPNAT_ENTRY_RESET(&ip_nat_icmp_table[i].common);
+  }
+  for (i = 0; i < LWIP_NAT_DEFAULT_STATE_TABLES_TCP; i++) {
+    IPNAT_ENTRY_RESET(&ip_nat_tcp_table[i].common);
+  }
+  for (i = 0; i < LWIP_NAT_DEFAULT_STATE_TABLES_UDP; i++) {
+    IPNAT_ENTRY_RESET(&ip_nat_udp_table[i].common);
+  }
+
+  /* we must lock scheduler to protect following code */
+  rt_enter_critical();
+
+  /* add a lwip timer for NAT */
+  sys_timeout(LWIP_NAT_TMR_INTERVAL_SEC, nat_timer, NULL);
+
+  /* un-protect */
+  rt_exit_critical();
+}
+
+/** Allocate a new ip_nat_conf_t item */
+static ip_nat_conf_t*
+ip_nat_alloc(void)
+{
+  ip_nat_conf_t *ret = (ip_nat_conf_t*)mem_malloc(sizeof(ip_nat_conf_t));
+  return ret;
+}
+
+/** Free a removed ip_nat_conf_t item */
+static void
+ip_nat_free(ip_nat_conf_t *item)
+{
+  LWIP_ASSERT("item != NULL", item != NULL);
+  mem_free(item);
+}
+
+/** Add a new NAT entry
+ *
+ * @param new_entry pointer to a structure used to initialize the entry
+ * @return ERR_OK if succeeded
+ */
+err_t
+ip_nat_add(const ip_nat_entry_t *new_entry)
+{
+  err_t err = ERR_VAL;
+  ip_nat_conf_t *cur = ip_nat_cfg;
+  ip_nat_conf_t *ip_nat_cfg_new = ip_nat_alloc();
+  LWIP_ASSERT("new_entry != NULL", new_entry != NULL);
+
+  if (ip_nat_cfg_new != NULL) {
+    SMEMCPY(&ip_nat_cfg_new->entry, new_entry, sizeof(ip_nat_entry_t));
+    ip_nat_cfg_new->next = NULL;
+
+    ip_nat_dbg_dump_init(ip_nat_cfg_new);
+
+    if (ip_nat_cfg == NULL) {
+      ip_nat_cfg = ip_nat_cfg_new;
+    } else {
+      /* @todo: do we really need to enqueue the new entry at the end?? */
+      while (cur->next != NULL) {
+        cur = cur->next;
+      }
+      cur->next = ip_nat_cfg_new;
+    }
+    err = ERR_OK;
+  } else {
+    err = ERR_MEM;
+  }
+  return err;
+}
+
+/** Remove a NAT entry previously added by 'ip_nat_add()'.
+ *
+ * @param remove_entry describes the entry to remove
+ */
+void
+ip_nat_remove(const ip_nat_entry_t *remove_entry)
+{
+  ip_nat_conf_t *cur = ip_nat_cfg;
+  ip_nat_conf_t *next;
+  ip_nat_conf_t *previous = NULL;
+
+  while (cur != NULL) {
+    /* Remove the NAT interfaces */
+    if ((cur->entry.source_net.addr     == remove_entry->source_net.addr) &&
+        (cur->entry.source_netmask.addr == remove_entry->source_netmask.addr) &&
+        (cur->entry.dest_net.addr       == remove_entry->dest_net.addr) &&
+        (cur->entry.dest_netmask.addr   == remove_entry->dest_netmask.addr) &&
+        (cur->entry.out_if              == remove_entry->out_if) &&
+        (cur->entry.in_if               == remove_entry->in_if))
+    {
+      ip_nat_dbg_dump_remove(cur);
+
+      ip_nat_reset_state(cur);
+      next = cur->next;
+      if (cur == ip_nat_cfg) {
+        ip_nat_cfg = next;
+      } else {
+        LWIP_ASSERT("NULL != previous", NULL != previous);
+        previous->next = next;
+      }
+      /* free 'cur' or there will be a memory leak */
+      ip_nat_free(cur);
+      return;
+    } else {
+      previous = cur;
+      cur = cur->next;
+    }
+  }
+}
+
+/** Reset a NAT configured entry to be reused.
+ * Effectively calls IPNAT_ENTRY_RESET() on 'cfg'.
+ *
+ * @param cfg NAT entry to reset
+ */
+static void
+ip_nat_reset_state(ip_nat_conf_t *cfg)
+{
+  int i;
+
+  /* @todo: optimize this!!!
+     why do we search for it anyway, if we have the pointer??? */
+  for (i = 0; i < LWIP_NAT_DEFAULT_STATE_TABLES_ICMP; i++) {
+    if(ip_nat_icmp_table[i].common.cfg == cfg) {
+      IPNAT_ENTRY_RESET(&ip_nat_icmp_table[i].common);
+    }
+  }
+  for (i = 0; i < LWIP_NAT_DEFAULT_STATE_TABLES_TCP; i++) {
+    if(ip_nat_icmp_table[i].common.cfg == cfg) {
+      IPNAT_ENTRY_RESET(&ip_nat_tcp_table[i].common);
+    }
+  }
+  for (i = 0; i < LWIP_NAT_DEFAULT_STATE_TABLES_UDP; i++) {
+    if(ip_nat_icmp_table[i].common.cfg == cfg) {
+      IPNAT_ENTRY_RESET(&ip_nat_udp_table[i].common);
+    }
+  }
+}
+
+/** Check if this packet should be routed or should be translated
+ *
+ * @param iphdr the IP header to check
+ * @return - a NAT entry if the packet shall be translated,
+ *         - NULL if the packet shall be routed normally
+ */
+static ip_nat_conf_t *
+ip_nat_shallnat(const struct ip_hdr *iphdr)
+{
+  ip_nat_conf_t *nat_config = ip_nat_cfg;
+
+  for (nat_config = ip_nat_cfg; nat_config != NULL; nat_config = nat_config->next) {
+      if (ip_addr_netcmp(&(iphdr->dest), &(nat_config->entry.dest_net),
+                       &(nat_config->entry.dest_netmask)) ||
+      ip_addr_netcmp(&(iphdr->src), &(nat_config->entry.source_net),
+                     &(nat_config->entry.source_netmask))) {
+        break;
+    }
+  }
+
+  return nat_config;
+}
+
+/** Check if the IP header can be hidden and if the remaining packet
+ * is long enough. p->payload is reset to the IP header on return.
+ *
+ * @param p received packet, p->payload pointing to IP header
+ * @param min_size minimum p->tot_len after hiding IP header
+ * @return a pointer to the next header (after IP header),
+ *         NULL if hiding IP header fails or the packet is too short
+ */
+static void*
+ip_nat_check_header(struct pbuf *p, u16_t min_size)
+{
+  void *ret = NULL;
+  struct ip_hdr  *iphdr = (struct ip_hdr*)p->payload;
+  s16_t iphdr_len = IPH_HL(iphdr) * 4;
+
+  if(!pbuf_header(p, -iphdr_len)) {
+    if(p->tot_len >= min_size) {
+      ret = p->payload;
+    }
+    /* Restore pbuf payload pointer from previous header check. */
+    pbuf_header(p, iphdr_len);
+  }
+  return ret;
+}
+
+/** Input processing: check if a received packet belongs to a NAT entry
+ * and if so, translated it and send it on.
+ *
+ * @param p received packet
+ * @return 1 if the packet has been consumed (it was a NAT packet),
+ *         0 if the packet has not been consumed (no NAT packet)
+ */
+u8_t
+ip_nat_input(struct pbuf *p)
+{
+  struct ip_hdr        *iphdr = (struct ip_hdr*)p->payload;
+  struct tcp_hdr       *tcphdr;
+  struct udp_hdr       *udphdr;
+  struct icmp_echo_hdr *icmphdr;
+  nat_entry_t           nat_entry;
+  err_t                 err;
+  u8_t                  consumed = 0;
+  int                   i;
+  struct pbuf          *q = NULL;
+
+  nat_entry.cmn = NULL;
+  ip_nat_dbg_dump("ip_nat_in: checking nat for", iphdr);
+
+  switch (IPH_PROTO(iphdr)) {
+    case IP_PROTO_TCP:
+      tcphdr = (struct tcp_hdr*)ip_nat_check_header(p, sizeof(struct tcp_hdr));
+      if (tcphdr == NULL) {
+        LWIP_DEBUGF(LWIP_NAT_DEBUG, ("ip_nat_input: short tcp packet (%" U16_F " bytes) discarded\n", p->tot_len));
+      } else {
+        nat_entry.tcp = ip_nat_tcp_lookup_incoming(iphdr, tcphdr);
+        if (nat_entry.tcp != NULL) {
+          /* Refresh TCP entry */
+          nat_entry.tcp->common.ttl = LWIP_NAT_DEFAULT_TTL_SECONDS;
+          tcphdr->dest = nat_entry.tcp->sport;
+          /* Adjust TCP checksum for changed destination port */
+          ip_nat_chksum_adjust((u8_t *)&(tcphdr->chksum),
+            (u8_t *)&(nat_entry.tcp->nport), 2, (u8_t *)&(tcphdr->dest), 2);
+          /* Adjust TCP checksum for changing dest IP address */
+          ip_nat_chksum_adjust((u8_t *)&(tcphdr->chksum),
+            (u8_t *)&(nat_entry.cmn->cfg->entry.out_if->ip_addr.addr), 4,
+            (u8_t *)&(nat_entry.cmn->source.addr), 4);
+
+          consumed = 1;
+        }
+      }
+      break;
+
+    case IP_PROTO_UDP:
+      udphdr = (struct udp_hdr *)ip_nat_check_header(p, sizeof(struct udp_hdr));
+      if (udphdr == NULL) {
+        LWIP_DEBUGF(LWIP_NAT_DEBUG,
+          ("ip_nat_input: short udp packet (%" U16_F " bytes) discarded\n",
+          p->tot_len));
+      } else {
+        nat_entry.udp = ip_nat_udp_lookup_incoming(iphdr, udphdr);
+        if (nat_entry.udp != NULL) {
+          /* Refresh UDP entry */
+          nat_entry.udp->common.ttl = LWIP_NAT_DEFAULT_TTL_SECONDS;
+          udphdr->dest = nat_entry.udp->sport;
+          /* Adjust UDP checksum for changed destination port */
+          ip_nat_chksum_adjust((u8_t *)&(udphdr->chksum),
+            (u8_t *)&(nat_entry.udp->nport), 2, (u8_t *)&(udphdr->dest), 2);
+          /* Adjust UDP checksum for changing dest IP address */
+          ip_nat_chksum_adjust((u8_t *)&(udphdr->chksum),
+            (u8_t *)&(nat_entry.cmn->cfg->entry.out_if->ip_addr.addr), 4,
+            (u8_t *)&(nat_entry.cmn->source.addr), 4);
+
+          consumed = 1;
+        }
+      }
+      break;
+
+    case IP_PROTO_ICMP:
+      icmphdr = (struct icmp_echo_hdr *)ip_nat_check_header(p, sizeof(struct icmp_echo_hdr));
+      if (icmphdr == NULL) {
+        LWIP_DEBUGF(LWIP_NAT_DEBUG,
+          ("ip_nat_out: short icmp echo reply packet (%" U16_F " bytes) discarded\n",
+          p->tot_len));
+      } else {
+        if (ICMP_ER == ICMPH_TYPE(icmphdr)) {
+          for (i = 0; i < LWIP_NAT_DEFAULT_STATE_TABLES_ICMP; i++) {
+            nat_entry.icmp = &ip_nat_icmp_table[i];
+            if ((nat_entry.icmp->common.ttl) &&
+                (iphdr->src.addr == nat_entry.icmp->common.dest.addr) &&
+                (nat_entry.icmp->id == icmphdr->id) &&
+                (nat_entry.icmp->seqno == icmphdr->seqno)) {
+              ip_nat_dbg_dump_icmp_nat_entry("found existing nat entry: ", nat_entry.icmp);
+              consumed = 1;
+              IPNAT_ENTRY_RESET(nat_entry.cmn);
+              break;
+            }
+          }
+        }
+      }
+      break;
+
+    default:
+      break;
+  }
+
+  if(consumed) {
+    /* packet consumed, send it out on in_if */
+    struct netif *in_if;
+
+    /* check if the pbuf has room for link headers */
+    if (pbuf_header(p, PBUF_LINK_HLEN)) {
+      /* pbuf has no room for link headers, allocate an extra pbuf */
+      q = pbuf_alloc(PBUF_LINK, 0, PBUF_RAM);
+      if (q == NULL) {
+        LWIP_DEBUGF(LWIP_NAT_DEBUG, ("ip_nat_input: no pbuf for outgoing header\n"));
+        // rt_kprintf("ip_nat_input: no pbuf for outgoing header\n");
+        /* @todo: stats? */
+        pbuf_free(p);
+        p = NULL;
+        return 1;
+      } else {
+        pbuf_cat(q, p);
+      }
+    } else {
+      /* restore p->payload to IP header */
+      if (pbuf_header(p, -PBUF_LINK_HLEN)) {
+        LWIP_DEBUGF(LWIP_NAT_DEBUG, ("ip_nat_input: restoring header failed\n"));
+        // rt_kprintf("ip_nat_input: restoring header failed\n");
+        /* @todo: stats? */
+        pbuf_free(p);
+        p = NULL;
+        return 1;
+      }
+      else q = p;
+    }
+    /* if we come here, q is the pbuf to send (either points to p or to a chain) */
+    in_if = nat_entry.cmn->cfg->entry.in_if;
+    iphdr->dest.addr = nat_entry.cmn->source.addr;
+    ip_nat_chksum_adjust((u8_t *) & IPH_CHKSUM(iphdr),
+      (u8_t *) & (nat_entry.cmn->cfg->entry.out_if->ip_addr.addr), 4,
+      (u8_t *) & (iphdr->dest.addr), 4);
+
+    ip_nat_dbg_dump("ip_nat_input: packet back to source after nat: ", iphdr);
+    LWIP_DEBUGF(LWIP_NAT_DEBUG, ("ip_nat_input: sending packet on interface ("));
+    ip_nat_dbg_dump_ip(&(in_if->ip_addr));
+    LWIP_DEBUGF(LWIP_NAT_DEBUG, (")\n"));
+
+    err = in_if->output(in_if, q, (ip_addr_t *)&(iphdr->dest));
+    if(err != ERR_OK) {
+      LWIP_DEBUGF(LWIP_NAT_DEBUG,
+        ("ip_nat_input: failed to send rewritten packet. link layer returned %d\n",
+        err));
+      // rt_kprintf("ip_nat_input: failed to send rewritten packet. link layer returned %d\n", err);
+    }
+    /* now that q (and/or p) is sent (or not), give up the reference to it
+       this frees the input pbuf (p) as we have consumed it. */
+    pbuf_free(q);
+  }
+  return consumed;
+}
+
+/** Check if one NAT entry timed out */
+static void
+ip_nat_check_timeout(ip_nat_entry_common_t *nat_entry)
+{
+  if(nat_entry->ttl > 0) {
+    if(nat_entry->ttl != LWIP_NAT_TTL_INFINITE) {
+      /* this is not a 'no-timeout' entry */
+      if(nat_entry->ttl > LWIP_NAT_TMR_INTERVAL_SEC) {
+        nat_entry->ttl -= LWIP_NAT_TMR_INTERVAL_SEC;
+      } else {
+        nat_entry->ttl = 0;
+      }
+    }
+  }
+}
+
+/** The NAT timer function, to be called at an interval of
+ * LWIP_NAT_TMR_INTERVAL_SEC seconds.
+ */
+void
+ip_nat_tmr(void)
+{
+  int i;
+
+  LWIP_DEBUGF(LWIP_NAT_DEBUG, ("ip_nat_tmr: removing old entries\n"));
+
+  for(i = 0; i < LWIP_NAT_DEFAULT_STATE_TABLES_ICMP; i++) {
+    ip_nat_check_timeout((ip_nat_entry_common_t *) & ip_nat_icmp_table[i]);
+  }
+  for(i = 0; i < LWIP_NAT_DEFAULT_STATE_TABLES_TCP; i++) {
+    ip_nat_check_timeout((ip_nat_entry_common_t *) & ip_nat_tcp_table[i]);
+  }
+  for(i = 0; i < LWIP_NAT_DEFAULT_STATE_TABLES_UDP; i++) {
+    ip_nat_check_timeout((ip_nat_entry_common_t *) & ip_nat_udp_table[i]);
+  }
+}
+
+/** Vheck if we want to perform NAT with this packet. If so, send it out on
+ * the correct interface.
+ *
+ * @param p the packet to test/send
+ * @return 1: the packet has been sent using NAT,
+ *         0: the packet did not belong to a NAT entry
+ */
+u8_t
+ip_nat_out(struct pbuf *p)
+{
+  u8_t                  sent = 0;
+  err_t                 err;
+  struct ip_hdr        *iphdr = p->payload;
+  struct icmp_echo_hdr *icmphdr;
+  struct tcp_hdr       *tcphdr;
+  struct udp_hdr       *udphdr;
+  ip_nat_conf_t        *nat_config;
+  nat_entry_t           nat_entry;
+  int             i;
+
+  nat_entry.cmn = NULL;
+
+  ip_nat_dbg_dump("ip_nat_out: checking nat for", iphdr);
+
+  /* Check if this packet should be routed or should be translated */
+  nat_config = ip_nat_shallnat(iphdr);
+  if (nat_config != NULL ) {
+    if (nat_config->entry.out_if == NULL) {
+      LWIP_DEBUGF(LWIP_NAT_DEBUG, ("ip_nat_out: no external interface for nat table entry\n"));
+    } else {
+      switch (IPH_PROTO(iphdr))
+      {
+      case IP_PROTO_TCP:
+        tcphdr = (struct tcp_hdr *)ip_nat_check_header(p, sizeof(struct tcp_hdr));
+        if (tcphdr == NULL) {
+          LWIP_DEBUGF(LWIP_NAT_DEBUG,
+            ("ip_nat_out: short tcp packet (%" U16_F " bytes) discarded\n", p->tot_len));
+        } else {
+          nat_entry.tcp = ip_nat_tcp_lookup_outgoing(nat_config, iphdr, tcphdr, 1);
+          if (nat_entry.tcp != NULL) {
+            /* Adjust TCP checksum for changing source port */
+            tcphdr->src = nat_entry.tcp->nport;
+            ip_nat_chksum_adjust((u8_t *)&(tcphdr->chksum),
+              (u8_t *)&(nat_entry.tcp->sport), 2, (u8_t *)&(tcphdr->src), 2);
+            /* Adjust TCP checksum for changing source IP address */
+            ip_nat_chksum_adjust((u8_t *)&(tcphdr->chksum),
+              (u8_t *)&(nat_entry.cmn->source.addr), 4,
+              (u8_t *)&(nat_entry.cmn->cfg->entry.out_if->ip_addr.addr), 4);
+          }
+        }
+        break;
+
+      case IP_PROTO_UDP:
+        udphdr = (struct udp_hdr *)ip_nat_check_header(p, sizeof(struct udp_hdr));
+        if (udphdr == NULL) {
+          LWIP_DEBUGF(LWIP_NAT_DEBUG,
+            ("ip_nat_out: short udp packet (%" U16_F " bytes) discarded\n", p->tot_len));
+        } else {
+          nat_entry.udp = ip_nat_udp_lookup_outgoing(nat_config, iphdr, udphdr, 1);
+          if (nat_entry.udp != NULL) {
+            /* Adjust UDP checksum for changing source port */
+            udphdr->src = nat_entry.udp->nport;
+            ip_nat_chksum_adjust((u8_t *)&(udphdr->chksum),
+              (u8_t *)&(nat_entry.udp->sport), 2, (u8_t *) & (udphdr->src), 2);
+            /* Adjust UDP checksum for changing source IP address */
+            ip_nat_chksum_adjust((u8_t *)&(udphdr->chksum),
+              (u8_t *)&(nat_entry.cmn->source.addr), 4,
+              (u8_t *)&(nat_entry.cmn->cfg->entry.out_if->ip_addr.addr), 4);
+          }
+        }
+        break;
+
+      case IP_PROTO_ICMP:
+        icmphdr = (struct icmp_echo_hdr *)ip_nat_check_header(p, sizeof(struct icmp_echo_hdr));
+        if(icmphdr == NULL) {
+          LWIP_DEBUGF(LWIP_NAT_DEBUG,
+            ("ip_nat_out: short icmp echo packet (%" U16_F " bytes) discarded\n", p->tot_len));
+        } else {
+          if (ICMPH_TYPE(icmphdr) == ICMP_ECHO) {
+            for (i = 0; i < LWIP_NAT_DEFAULT_STATE_TABLES_ICMP; i++) {
+              if (!ip_nat_icmp_table[i].common.ttl) {
+                nat_entry.icmp = &ip_nat_icmp_table[i];
+                ip_nat_cmn_init(nat_config, iphdr, nat_entry.cmn);
+                nat_entry.icmp->id = icmphdr->id;
+                nat_entry.icmp->seqno = icmphdr->seqno;
+                ip_nat_dbg_dump_icmp_nat_entry(" ip_nat_out: created new NAT entry ", nat_entry.icmp);
+                break;
+              }
+            }
+            if (NULL == nat_entry.icmp)
+            {
+              LWIP_DEBUGF(LWIP_NAT_DEBUG, ("ip_nat_out: no more NAT entries for ICMP available\n"));
+            }
+          }
+        }
+        break;
+      default:
+        break;
+      }
+
+      if (nat_entry.cmn != NULL) {
+        struct netif *out_if = nat_entry.cmn->cfg->entry.out_if;
+        /* Exchange the IP source address with the address of the interface
+        * where the packet will be sent.
+        */
+        /* @todo: check nat_config->entry.out_if agains nat_entry.cmn->cfg->entry.out_if */
+        iphdr->src.addr = nat_config->entry.out_if->ip_addr.addr;
+        ip_nat_chksum_adjust((u8_t *) & IPH_CHKSUM(iphdr),
+          (u8_t *) & (nat_entry.cmn->source.addr), 4, (u8_t *) & iphdr->src.addr, 4);
+
+        ip_nat_dbg_dump("ip_nat_out: rewritten packet", iphdr);
+        LWIP_DEBUGF(LWIP_NAT_DEBUG, ("ip_nat_out: sending packet on interface ("));
+        ip_nat_dbg_dump_ip(&(out_if->ip_addr));
+        LWIP_DEBUGF(LWIP_NAT_DEBUG, (")\n"));
+
+        err = out_if->output(out_if, p, (ip_addr_t *)&(iphdr->dest));
+        if (err != ERR_OK) {
+          LWIP_DEBUGF(LWIP_NAT_DEBUG,
+            ("ip_nat_out: failed to send rewritten packet. link layer returned %d\n", err));
+          // rt_kprintf("ip_nat_out: failed to send rewritten packet. link layer returned %d\n", err);
+        } else {
+          sent = 1;
+        }
+      }
+    }
+  }
+
+  return sent;
+}
+
+/** Initialize common parts of a NAT entry
+ *
+ * @param nat_config NAT config entry
+ * @param iphdr IP header from which to initialize the entry
+ * @param nat_entry entry to initialize
+ */
+static void
+ip_nat_cmn_init(ip_nat_conf_t *nat_config, const struct ip_hdr *iphdr, ip_nat_entry_common_t *nat_entry)
+{
+  LWIP_ASSERT("NULL != nat_entry", NULL != nat_entry);
+  LWIP_ASSERT("NULL != nat_config", NULL != nat_config);
+  LWIP_ASSERT("NULL != iphdr", NULL != iphdr);
+  nat_entry->cfg = nat_config;
+  nat_entry->dest = *((ip_addr_t *)&iphdr->dest);
+  nat_entry->source = *((ip_addr_t *)&iphdr->src);
+  nat_entry->ttl = LWIP_NAT_DEFAULT_TTL_SECONDS;
+}
+
+/**
+ * This function checks for incoming packets if we already have a NAT entry.
+ * If yes a pointer to the NAT entry is returned. Otherwise NULL.
+ *
+ * @param nat_config NAT configuration.
+ * @param iphdr The IP header.
+ * @param udphdr The UDP header.
+ * @return A pointer to an existing NAT entry or
+ *         NULL if none is found.
+ */
+static ip_nat_entries_udp_t *
+ip_nat_udp_lookup_incoming(const struct ip_hdr *iphdr, const struct udp_hdr *udphdr)
+{
+  int i;
+  ip_nat_entries_udp_t *nat_entry = NULL;
+
+  for (i = 0; i < LWIP_NAT_DEFAULT_STATE_TABLES_UDP; i++) {
+    if (ip_nat_udp_table[i].common.ttl) {
+      if ((iphdr->src.addr == ip_nat_udp_table[i].common.dest.addr) &&
+          (udphdr->src == ip_nat_udp_table[i].dport) &&
+          (udphdr->dest == ip_nat_udp_table[i].nport)) {
+        nat_entry = &ip_nat_udp_table[i];
+        ip_nat_dbg_dump_udp_nat_entry("ip_nat_udp_lookup_incoming: found existing nat entry: ",
+                                      nat_entry);
+        break;
+      }
+    }
+  }
+  return nat_entry;
+}
+
+/**
+ * This function checks if we already have a NAT entry for this UDP connection.
+ * If yes the a pointer to this NAT entry is returned.
+ *
+ * @param iphdr The IP header.
+ * @param udphdr The UDP header.
+ * @param allocate If no existing NAT entry is found and this flag is true
+ *        a NAT entry is allocated.
+ */
+static ip_nat_entries_udp_t *
+ip_nat_udp_lookup_outgoing(ip_nat_conf_t *nat_config, const struct ip_hdr *iphdr,
+                           const struct udp_hdr *udphdr, u8_t allocate)
+{
+  int i;
+  nat_entry_t nat_entry;
+  int last_free = -1;
+
+  nat_entry.cmn = NULL;
+  for (i = 0; i < LWIP_NAT_DEFAULT_STATE_TABLES_UDP; i++) {
+    if (ip_nat_udp_table[i].common.ttl) {
+      if ((iphdr->src.addr == ip_nat_udp_table[i].common.source.addr) &&
+          (iphdr->dest.addr == ip_nat_udp_table[i].common.dest.addr) &&
+          (udphdr->src == ip_nat_udp_table[i].sport) &&
+          (udphdr->dest == ip_nat_udp_table[i].dport)) {
+        nat_entry.udp = &ip_nat_udp_table[i];
+
+        ip_nat_dbg_dump_udp_nat_entry("ip_nat_udp_lookup_outgoing: found existing nat entry: ",
+                                      nat_entry.udp);
+        break;
+      }
+    } else {
+        last_free = i;
+    }
+  }
+  if (nat_entry.cmn == NULL) {
+    if (allocate) {
+      if (last_free != -1) {
+        nat_entry.udp = &ip_nat_udp_table[last_free];
+        nat_entry.udp->nport = htons((u16_t) (LWIP_NAT_DEFAULT_UDP_SOURCE_PORT + i));
+        nat_entry.udp->sport = udphdr->src;
+        nat_entry.udp->dport = udphdr->dest;
+        ip_nat_cmn_init(nat_config, iphdr, nat_entry.cmn);
+
+        ip_nat_dbg_dump_udp_nat_entry("ip_nat_udp_lookup_outgoing: created new nat entry: ",
+                                      nat_entry.udp);
+      } else {
+        LWIP_DEBUGF(LWIP_NAT_DEBUG, ("ip_nat_udp_lookup_outgoing: no more NAT entries available\n"));
+        // rt_kprintf("ip_nat_udp_lookup_outgoing: no more NAT entries available\n");
+      }
+    }
+  }
+  return nat_entry.udp;
+}
+
+/**
+ * This function checks for incoming packets if we already have a NAT entry.
+ * If yes a pointer to the NAT entry is returned. Otherwise NULL.
+ *
+ * @param nat_config NAT configuration.
+ * @param iphdr The IP header.
+ * @param tcphdr The TCP header.
+ * @return A pointer to an existing NAT entry or NULL if none is found.
+ */
+static ip_nat_entries_tcp_t *
+ip_nat_tcp_lookup_incoming(const struct ip_hdr *iphdr, const struct tcp_hdr *tcphdr)
+{
+  int i;
+  ip_nat_entries_tcp_t *nat_entry = NULL;
+
+  for (i = 0; i < LWIP_NAT_DEFAULT_STATE_TABLES_TCP; i++) {
+    if (ip_nat_tcp_table[i].common.ttl) {
+      if ((iphdr->src.addr == ip_nat_tcp_table[i].common.dest.addr) &&
+          (tcphdr->src == ip_nat_tcp_table[i].dport) &&
+          (tcphdr->dest == ip_nat_tcp_table[i].nport)) {
+        nat_entry = &ip_nat_tcp_table[i];
+
+        ip_nat_dbg_dump_tcp_nat_entry("ip_nat_tcp_lookup_incoming: found existing nat entry: ",
+                                      nat_entry);
+        break;
+      }
+    }
+  }
+  return nat_entry;
+}
+
+/**
+ * This function checks if we already have a NAT entry for this TCP connection.
+ * If yes the a pointer to this NAT entry is returned.
+ *
+ * @param iphdr The IP header.
+ * @param tcphdr The TCP header.
+ * @param allocate If no existing NAT entry is found and this flag is true
+ *   a NAT entry is allocated.
+ */
+static ip_nat_entries_tcp_t *
+ip_nat_tcp_lookup_outgoing(ip_nat_conf_t *nat_config, const struct ip_hdr *iphdr,
+                           const struct tcp_hdr *tcphdr, u8_t allocate)
+{
+  int i;
+  nat_entry_t nat_entry;
+  int last_free = -1;
+
+  nat_entry.cmn = NULL;
+  for (i = 0; i < LWIP_NAT_DEFAULT_STATE_TABLES_TCP; i++) {
+    if (ip_nat_tcp_table[i].common.ttl) {
+      if ((iphdr->src.addr == ip_nat_tcp_table[i].common.source.addr) &&
+          (iphdr->dest.addr == ip_nat_tcp_table[i].common.dest.addr) &&
+          (tcphdr->src == ip_nat_tcp_table[i].sport) &&
+          (tcphdr->dest == ip_nat_tcp_table[i].dport)) {
+        nat_entry.tcp = &ip_nat_tcp_table[i];
+
+        ip_nat_dbg_dump_tcp_nat_entry("ip_nat_tcp_lookup_outgoing: found existing nat entry: ",
+                                      nat_entry.tcp);
+        break;
+      }
+    } else {
+        last_free = i;
+    }
+  }
+  if (nat_entry.cmn == NULL) {
+    if (allocate) {
+      if (last_free != -1) {
+        nat_entry.tcp = &ip_nat_tcp_table[last_free];
+        nat_entry.tcp->nport = htons((u16_t) (LWIP_NAT_DEFAULT_TCP_SOURCE_PORT + i));
+        nat_entry.tcp->sport = tcphdr->src;
+        nat_entry.tcp->dport = tcphdr->dest;
+        ip_nat_cmn_init(nat_config, iphdr, nat_entry.cmn);
+
+        ip_nat_dbg_dump_tcp_nat_entry("ip_nat_tcp_lookup_outgoing: created new nat entry: ",
+                                      nat_entry.tcp);
+      } else {
+        LWIP_DEBUGF(LWIP_NAT_DEBUG, ("ip_nat_udp_lookup_outgoing: no more NAT entries available\n"));
+        // rt_kprintf("ip_nat_udp_lookup_outgoing: no more NAT entries available\n");
+      }
+    }
+  }
+  return nat_entry.tcp;
+}
+
+/** Adjusts the checksum of a NAT'ed packet without having to completely recalculate it
+ * @todo: verify this works for little- and big-endian
+ *
+ * @param chksum points to the chksum in the packet
+ * @param optr points to the old data in the packet
+ * @param olen length of old data
+ * @param nptr points to the new data in the packet
+ * @param nlen length of new data
+ */
+static void
+ip_nat_chksum_adjust(u8_t *chksum, const u8_t *optr, s16_t olen, const u8_t *nptr, s16_t nlen)
+{
+  s32_t x, oldval, newval;
+
+  LWIP_ASSERT("NULL != chksum", NULL != chksum);
+  LWIP_ASSERT("NULL != optr", NULL != optr);
+  LWIP_ASSERT("NULL != nptr", NULL != nptr);
+  LWIP_DEBUGF(LWIP_NAT_DEBUG, ("ip_nat_chksum_adjust: chksum=%p, optr=%p, olen=%" U16_F ", nptr=%p, nlen=%" U16_F "\n",
+    chksum, optr, olen, nptr, nlen));
+  x = chksum[0] * 256 + chksum[1];
+  x = ~x & 0xFFFF;
+  while (olen) {
+    oldval = optr[0] * 256 + optr[1];
+    optr += 2;
+    x -= oldval & 0xffff;
+    if (x <= 0) {
+      x--;
+      x &= 0xffff;
+    }
+    olen -= 2;
+  }
+  while (nlen) {
+    newval = nptr[0] * 256 + nptr[1];
+    nptr += 2;
+    x += newval & 0xffff;
+    if (x & 0x10000) {
+      x++;
+      x &= 0xffff;
+    }
+    nlen -= 2;
+  }
+  x = ~x & 0xFFFF;
+  chksum[0] = x / 256;
+  chksum[1] = x & 0xff;
+  LWIP_DEBUGF(LWIP_NAT_DEBUG, ("ip_nat_chksum_adjust: chksum = 0x%x\n", *((u16_t *) chksum)));
+}
+
+#if defined(LWIP_DEBUG) && (LWIP_NAT_DEBUG & LWIP_DBG_ON)
+/**
+ * This function dumps an IP address
+ *
+ * @param addr IP address
+ */
+static void
+ip_nat_dbg_dump_ip(const ip_addr_t *addr)
+{
+  LWIP_ASSERT("NULL != addr", NULL != addr);
+  LWIP_DEBUGF(LWIP_NAT_DEBUG, ("%" U16_F ".%" U16_F ".%" U16_F ".%" U16_F,
+    ip4_addr1(addr), ip4_addr2(addr), ip4_addr3(addr), ip4_addr4(addr)));
+}
+
+/**
+ * This function dumps an IP header
+ *
+ * @param msg a message to print
+ * @param iphdr IP header
+ */
+static void
+ip_nat_dbg_dump(const char *msg, const struct ip_hdr *iphdr)
+{
+  LWIP_ASSERT("NULL != msg", NULL != msg);
+  LWIP_ASSERT("NULL != iphdr", NULL != iphdr);
+  LWIP_DEBUGF(LWIP_NAT_DEBUG, ("%s: IP: (", msg));
+  ip_nat_dbg_dump_ip((ip_addr_t *)&(iphdr->src));
+  LWIP_DEBUGF(LWIP_NAT_DEBUG, (" --> "));
+  ip_nat_dbg_dump_ip((ip_addr_t *)&(iphdr->dest));
+  LWIP_DEBUGF(LWIP_NAT_DEBUG, (" id=%" U16_F ", chksum=%" U16_F ")\n",
+    ntohs(IPH_ID(iphdr)), ntohs(IPH_CHKSUM(iphdr))));
+}
+
+/**
+ * This function dumps an ICMP echo reply/recho request nat entry.
+ *
+ * @param msg a message to print
+ * @param nat_entry the ICMP NAT entry to print
+ */
+static void
+ip_nat_dbg_dump_icmp_nat_entry(const char *msg, const ip_nat_entries_icmp_t *nat_entry)
+{
+  LWIP_ASSERT("NULL != msg", NULL != msg);
+  LWIP_ASSERT("NULL != nat_entry", NULL != nat_entry);
+  LWIP_ASSERT("NULL != nat_entry->common.cfg", NULL != nat_entry->common.cfg);
+  LWIP_ASSERT("NULL != nat_entry->common.cfg->entry.out_if",
+    NULL != nat_entry->common.cfg->entry.out_if);
+  LWIP_DEBUGF(LWIP_NAT_DEBUG, ("%s", msg));
+  LWIP_DEBUGF(LWIP_NAT_DEBUG, ("ICMP : ("));
+  ip_nat_dbg_dump_ip(&(nat_entry->common.source));
+  LWIP_DEBUGF(LWIP_NAT_DEBUG, (" --> "));
+  ip_nat_dbg_dump_ip(&(nat_entry->common.dest));
+  LWIP_DEBUGF(LWIP_NAT_DEBUG, (" id=%" U16_F, ntohs(nat_entry->id)));
+  LWIP_DEBUGF(LWIP_NAT_DEBUG, (", seq=%" U16_F, ntohs(nat_entry->seqno)));
+  LWIP_DEBUGF(LWIP_NAT_DEBUG, (") mapped at ("));
+  ip_nat_dbg_dump_ip(&(nat_entry->common.cfg->entry.out_if->ip_addr));
+  LWIP_DEBUGF(LWIP_NAT_DEBUG, (" --> "));
+  ip_nat_dbg_dump_ip(&(nat_entry->common.dest));
+  LWIP_DEBUGF(LWIP_NAT_DEBUG, (" id=%" U16_F, ntohs(nat_entry->id)));
+  LWIP_DEBUGF(LWIP_NAT_DEBUG, (", seq=%" U16_F, ntohs(nat_entry->seqno)));
+  LWIP_DEBUGF(LWIP_NAT_DEBUG, (")\n"));
+}
+
+/**
+ * This function dumps an TCP nat entry.
+ *
+ * @param msg a message to print
+ * @param nat_entry the TCP NAT entry to print
+ */
+static void
+ip_nat_dbg_dump_tcp_nat_entry(const char *msg, const ip_nat_entries_tcp_t *nat_entry)
+{
+  LWIP_ASSERT("NULL != msg", NULL != msg);
+  LWIP_ASSERT("NULL != nat_entry", NULL != nat_entry);
+  LWIP_ASSERT("NULL != nat_entry->common.cfg", NULL != nat_entry->common.cfg);
+  LWIP_ASSERT("NULL != nat_entry->common.cfg->entry.out_if",
+    NULL != nat_entry->common.cfg->entry.out_if);
+  LWIP_DEBUGF(LWIP_NAT_DEBUG, ("%s", msg));
+  LWIP_DEBUGF(LWIP_NAT_DEBUG, ("TCP : ("));
+  ip_nat_dbg_dump_ip(&(nat_entry->common.source));
+  LWIP_DEBUGF(LWIP_NAT_DEBUG, (":%" U16_F, ntohs(nat_entry->sport)));
+  LWIP_DEBUGF(LWIP_NAT_DEBUG, (" --> "));
+  ip_nat_dbg_dump_ip(&(nat_entry->common.dest));
+  LWIP_DEBUGF(LWIP_NAT_DEBUG, (":%" U16_F, ntohs(nat_entry->dport)));
+  LWIP_DEBUGF(LWIP_NAT_DEBUG, (") mapped at ("));
+  ip_nat_dbg_dump_ip(&(nat_entry->common.cfg->entry.out_if->ip_addr));
+  LWIP_DEBUGF(LWIP_NAT_DEBUG, (":%" U16_F, ntohs(nat_entry->nport)));
+  LWIP_DEBUGF(LWIP_NAT_DEBUG, (" --> "));
+  ip_nat_dbg_dump_ip(&(nat_entry->common.dest));
+  LWIP_DEBUGF(LWIP_NAT_DEBUG, (":%" U16_F, ntohs(nat_entry->dport)));
+  LWIP_DEBUGF(LWIP_NAT_DEBUG, (")\n"));
+}
+
+/**
+ * This function dumps a UDP NAT entry.
+ *
+ * @param msg a message to print
+ * @param nat_entry the UDP NAT entry to print
+ */
+static void
+ip_nat_dbg_dump_udp_nat_entry(const char *msg, const ip_nat_entries_udp_t *nat_entry)
+{
+  LWIP_ASSERT("NULL != msg", NULL != msg);
+  LWIP_ASSERT("NULL != nat_entry", NULL != nat_entry);
+  LWIP_ASSERT("NULL != nat_entry->common.cfg", NULL != nat_entry->common.cfg);
+  LWIP_ASSERT("NULL != nat_entry->common.cfg->entry.out_if",
+    NULL != nat_entry->common.cfg->entry.out_if);
+  LWIP_DEBUGF(LWIP_NAT_DEBUG, ("%s", msg));
+  LWIP_DEBUGF(LWIP_NAT_DEBUG, ("UDP : ("));
+  ip_nat_dbg_dump_ip(&(nat_entry->common.source));
+  LWIP_DEBUGF(LWIP_NAT_DEBUG, (":%" U16_F, ntohs(nat_entry->sport)));
+  LWIP_DEBUGF(LWIP_NAT_DEBUG, (" --> "));
+  ip_nat_dbg_dump_ip(&(nat_entry->common.dest));
+  LWIP_DEBUGF(LWIP_NAT_DEBUG, (":%" U16_F, ntohs(nat_entry->dport)));
+  LWIP_DEBUGF(LWIP_NAT_DEBUG, (") mapped at ("));
+  ip_nat_dbg_dump_ip(&(nat_entry->common.cfg->entry.out_if->ip_addr));
+  LWIP_DEBUGF(LWIP_NAT_DEBUG, (":%" U16_F, ntohs(nat_entry->nport)));
+  LWIP_DEBUGF(LWIP_NAT_DEBUG, (" --> "));
+  ip_nat_dbg_dump_ip(&(nat_entry->common.dest));
+  LWIP_DEBUGF(LWIP_NAT_DEBUG, (":%" U16_F, ntohs(nat_entry->dport)));
+  LWIP_DEBUGF(LWIP_NAT_DEBUG, (")\n"));
+}
+
+/** Prints some info when creating a new NAT entry */
+static void
+ip_nat_dbg_dump_init(ip_nat_conf_t *ip_nat_cfg_new)
+{
+  LWIP_DEBUGF(LWIP_NAT_DEBUG, ("ip_nat_init: added new NAT interface\n"));
+  LWIP_DEBUGF(LWIP_NAT_DEBUG, ("ip_nat_init:   "));
+  ip_nat_dbg_dump_ip(&(ip_nat_cfg_new->entry.source_net));
+  LWIP_DEBUGF(LWIP_NAT_DEBUG, ("/"));
+  ip_nat_dbg_dump_ip(&(ip_nat_cfg_new->entry.source_netmask));
+  LWIP_DEBUGF(LWIP_NAT_DEBUG, ("@"));
+  ip_nat_dbg_dump_ip(&(ip_nat_cfg_new->entry.in_if->ip_addr));
+  LWIP_DEBUGF(LWIP_NAT_DEBUG, (" --> "));
+  ip_nat_dbg_dump_ip(&(ip_nat_cfg_new->entry.dest_net));
+  LWIP_DEBUGF(LWIP_NAT_DEBUG, ("/"));
+  ip_nat_dbg_dump_ip(&(ip_nat_cfg_new->entry.dest_netmask));
+  LWIP_DEBUGF(LWIP_NAT_DEBUG, ("@"));
+  ip_nat_dbg_dump_ip(&(ip_nat_cfg_new->entry.out_if->ip_addr));
+  LWIP_DEBUGF(LWIP_NAT_DEBUG, ("\n"));
+}
+
+/** Prints some info when removing a NAT entry */
+static void
+ip_nat_dbg_dump_remove(ip_nat_conf_t *cur)
+{
+  LWIP_DEBUGF(LWIP_NAT_DEBUG, ("ip_nat_remove: removing existing NAT interface\n"));
+  LWIP_DEBUGF(LWIP_NAT_DEBUG, ("ip_nat_remove:   "));
+  ip_nat_dbg_dump_ip(&(cur->entry.source_net));
+  LWIP_DEBUGF(LWIP_NAT_DEBUG, ("/"));
+  ip_nat_dbg_dump_ip(&(cur->entry.source_netmask));
+  LWIP_DEBUGF(LWIP_NAT_DEBUG, ("@"));
+  ip_nat_dbg_dump_ip(&(cur->entry.in_if->ip_addr));
+  LWIP_DEBUGF(LWIP_NAT_DEBUG, (" --> "));
+  ip_nat_dbg_dump_ip(&(cur->entry.dest_net));
+  LWIP_DEBUGF(LWIP_NAT_DEBUG, ("/"));
+  ip_nat_dbg_dump_ip(&(cur->entry.dest_netmask));
+  LWIP_DEBUGF(LWIP_NAT_DEBUG, ("@"));
+  ip_nat_dbg_dump_ip(&(cur->entry.out_if->ip_addr));
+  LWIP_DEBUGF(LWIP_NAT_DEBUG, ("\n"));
+}
+#endif /* defined(LWIP_DEBUG) && (LWIP_NAT_DEBUG & LWIP_DBG_ON) */
+
+#endif /* IP_NAT */
diff --git a/components/net/lwip_nat/ipv4_nat.h b/components/net/lwip_nat/ipv4_nat.h
new file mode 100644
index 0000000000..e369c781da
--- /dev/null
+++ b/components/net/lwip_nat/ipv4_nat.h
@@ -0,0 +1,103 @@
+/**
+ * NAT - NAT implementation for lwIP supporting TCP/UDP and ICMP.
+ * Copyright (c) 2009 Christian Walter, ?Embedded Solutions, Vienna 2009.
+ *
+ * Copyright (c) 2010 lwIP project ;-)
+ * 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.
+ */
+
+/*
+ * File      : ipv4_nat.h
+ * This file is part of RT-Thread RTOS
+ * COPYRIGHT (C) 2015, RT-Thread Development Team
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License as published by
+ *  the Free Software Foundation; either version 2 of the License, or
+ *  (at your option) any later version.
+ *
+ *  This program is distributed in the hope that it will be useful,
+ *  but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *  GNU General Public License for more details.
+ *
+ *  You should have received a copy of the GNU General Public License along
+ *  with this program; if not, write to the Free Software Foundation, Inc.,
+ *  51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
+ *
+ * Change Logs:
+ * Date           Author       Notes
+ * 2015-01-26     Hichard      porting to RT-Thread
+ * 2015-01-27     Bernard      code cleanup for lwIP in RT-Thread
+ */
+
+#ifndef __LWIP_NAT_H__
+#define __LWIP_NAT_H__
+
+#include <rtthread.h>
+
+#ifdef LWIP_USING_NAT
+
+#include "lwip/err.h"
+#include "lwip/ip_addr.h"
+#include "lwip/opt.h"
+
+/** Timer interval at which to call ip_nat_tmr() */
+#define LWIP_NAT_TMR_INTERVAL_SEC        (30*1000)
+
+#ifdef __cplusplus
+extern "C" {
+#endif /* __cplusplus */
+
+struct netif;
+struct pbuf;
+
+typedef struct ip_nat_entry
+{
+  ip_addr_t    source_net;
+  ip_addr_t    source_netmask;
+  ip_addr_t    dest_net;
+  ip_addr_t    dest_netmask;
+  struct netif *out_if;
+  struct netif *in_if;
+} ip_nat_entry_t;
+
+void  ip_nat_init(void);
+void  ip_nat_tmr(void);
+u8_t  ip_nat_input(struct pbuf *p);
+u8_t  ip_nat_out(struct pbuf *p);
+
+err_t ip_nat_add(const ip_nat_entry_t *new_entry);
+void  ip_nat_remove(const ip_nat_entry_t *remove_entry);
+
+#ifdef __cplusplus
+}
+#endif /* __cplusplus */
+
+#endif /* IP_NAT */
+
+#endif /* __LWIP_NAT_H__ */