bsd_socket.c

Go to the documentation of this file.

/**
 * @file bsd_socket.c
 * @brief BSD socket API
 *
 * @section License
 *
 * SPDX-License-Identifier: GPL-2.0-or-later
 *
 * Copyright (C) 2010-2025 Oryx Embedded SARL. All rights reserved.
 *
 * This file is part of CycloneTCP Open.
 *
 * 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.
 *
 * @author Oryx Embedded SARL (www.oryx-embedded.com)
 * @version 2.5.0
 **/
 
//Switch to the appropriate trace level
#define TRACE_LEVEL BSD_SOCKET_TRACE_LEVEL
 
//Dependencies
#include "core/net.h"
#include "core/bsd_socket.h"
#include "core/socket.h"
#include "core/socket_misc.h"
#include "debug.h"
 
//Check TCP/IP stack configuration
#if (BSD_SOCKET_SUPPORT == ENABLED)
 
//Dependencies
#include "core/bsd_socket_options.h"
#include "core/bsd_socket_misc.h"
 
//Common IPv6 addresses
const struct in6_addr in6addr_any =
   {{0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00}};
 
const struct in6_addr in6addr_loopback =
   {{0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x01}};
 
 
/**
 * @brief Create a socket that is bound to a specific transport service provider
 * @param[in] family Address family
 * @param[in] type Type specification for the new socket
 * @param[in] protocol Protocol to be used
 * @return On success, a file descriptor for the new socket is returned.
 *   On failure, SOCKET_ERROR is returned
 **/
 
int_t socket(int_t family, int_t type, int_t protocol)
{
   Socket *sock;
 
   //Check address family
   if(family == AF_INET || family == AF_INET6)
   {
      //Create a socket
      sock = socketOpen(type, protocol);
   }
   else if(family == AF_PACKET)
   {
      //Create a socket
      sock = socketOpen(SOCKET_TYPE_RAW_ETH, ntohs(protocol));
   }
   else
   {
      //The address family is not valid
      return SOCKET_ERROR;
   }
 
   //Failed to create a new socket?
   if(sock == NULL)
   {
      //Report an error
      return SOCKET_ERROR;
   }
 
   //Return the socket descriptor
   return sock->descriptor;
}
 
 
/**
 * @brief Associate a local address with a socket
 * @param[in] s Descriptor identifying an unbound socket
 * @param[in] addr Local address to assign to the bound socket
 * @param[in] addrlen Length in bytes of the address
 * @return If no error occurs, bind returns SOCKET_SUCCESS.
 *   Otherwise, it returns SOCKET_ERROR
 **/
 
int_t bind(int_t s, const struct sockaddr *addr, socklen_t addrlen)
{
   error_t error;
   uint16_t port;
   IpAddr ipAddr;
   Socket *sock;
 
   //Make sure the socket descriptor is valid
   if(s < 0 || s >= SOCKET_MAX_COUNT)
   {
      return SOCKET_ERROR;
   }
 
   //Point to the socket structure
   sock = &socketTable[s];
 
   //Check the length of the address
   if(addrlen < (socklen_t) sizeof(SOCKADDR))
   {
      //Report an error
      socketSetErrnoCode(sock, EINVAL);
      return SOCKET_ERROR;
   }
 
#if (IPV4_SUPPORT == ENABLED)
   //IPv4 address?
   if(addr->sa_family == AF_INET &&
      addrlen >= (socklen_t) sizeof(SOCKADDR_IN))
   {
      //Point to the IPv4 address information
      SOCKADDR_IN *sa = (SOCKADDR_IN *) addr;
 
      //Get port number
      port = ntohs(sa->sin_port);
 
      //Copy IPv4 address
      ipAddr.length = sizeof(Ipv4Addr);
      ipAddr.ipv4Addr = sa->sin_addr.s_addr;
   }
   else
#endif
#if (IPV6_SUPPORT == ENABLED)
   //IPv6 address?
   if(addr->sa_family == AF_INET6 &&
      addrlen >= (socklen_t) sizeof(SOCKADDR_IN6))
   {
      //Point to the IPv6 address information
      SOCKADDR_IN6 *sa = (SOCKADDR_IN6 *) addr;
 
      //Get port number
      port = ntohs(sa->sin6_port);
 
      //Copy IPv6 address
      if(ipv6CompAddr(sa->sin6_addr.s6_addr, &in6addr_any))
      {
         ipAddr.length = 0;
         ipAddr.ipv6Addr = IPV6_UNSPECIFIED_ADDR;
      }
      else
      {
         ipAddr.length = sizeof(Ipv6Addr);
         ipv6CopyAddr(&ipAddr.ipv6Addr, sa->sin6_addr.s6_addr);
      }
   }
   else
#endif
   //Invalid address?
   {
      //Report an error
      socketSetErrnoCode(sock, EINVAL);
      return SOCKET_ERROR;
   }
 
   //Associate the local address with the socket
   error = socketBind(sock, &ipAddr, port);
 
   //Any error to report?
   if(error)
   {
      socketTranslateErrorCode(sock, error);
      return SOCKET_ERROR;
   }
 
   //Successful processing
   return SOCKET_SUCCESS;
}
 
 
/**
 * @brief Establish a connection to a specified socket
 * @param[in] s Descriptor identifying an unconnected socket
 * @param[in] addr Address to which the connection should be established
 * @param[in] addrlen Length in bytes of the address
 * @return If no error occurs, connect returns SOCKET_SUCCESS.
 *   Otherwise, it returns SOCKET_ERROR
 **/
 
int_t connect(int_t s, const struct sockaddr *addr, socklen_t addrlen)
{
   error_t error;
   uint16_t port;
   IpAddr ipAddr;
   Socket *sock;
 
   //Make sure the socket descriptor is valid
   if(s < 0 || s >= SOCKET_MAX_COUNT)
   {
      return SOCKET_ERROR;
   }
 
   //Point to the socket structure
   sock = &socketTable[s];
 
   //Check the length of the address
   if(addrlen < (socklen_t) sizeof(SOCKADDR))
   {
      socketSetErrnoCode(sock, EINVAL);
      return SOCKET_ERROR;
   }
 
#if (IPV4_SUPPORT == ENABLED)
   //IPv4 address?
   if(addr->sa_family == AF_INET &&
      addrlen >= (socklen_t) sizeof(SOCKADDR_IN))
   {
      //Point to the IPv4 address information
      SOCKADDR_IN *sa = (SOCKADDR_IN *) addr;
 
      //Get port number
      port = ntohs(sa->sin_port);
 
      //Copy IPv4 address
      ipAddr.length = sizeof(Ipv4Addr);
      ipAddr.ipv4Addr = sa->sin_addr.s_addr;
   }
   else
#endif
#if (IPV6_SUPPORT == ENABLED)
   //IPv6 address?
   if(addr->sa_family == AF_INET6 &&
      addrlen >= (socklen_t) sizeof(SOCKADDR_IN6))
   {
      //Point to the IPv6 address information
      SOCKADDR_IN6 *sa = (SOCKADDR_IN6 *) addr;
 
      //Get port number
      port = ntohs(sa->sin6_port);
 
      //Copy IPv6 address
      if(ipv6CompAddr(sa->sin6_addr.s6_addr, &in6addr_any))
      {
         ipAddr.length = 0;
         ipAddr.ipv6Addr = IPV6_UNSPECIFIED_ADDR;
      }
      else
      {
         ipAddr.length = sizeof(Ipv6Addr);
         ipv6CopyAddr(&ipAddr.ipv6Addr, sa->sin6_addr.s6_addr);
      }
   }
   else
#endif
   //Invalid address?
   {
      //Report an error
      socketSetErrnoCode(sock, EINVAL);
      return SOCKET_ERROR;
   }
 
   //Establish connection
   error = socketConnect(sock, &ipAddr, port);
 
   //Check status code
   if(error == NO_ERROR)
   {
      //Successful processing
      return SOCKET_SUCCESS;
   }
   else if(error == ERROR_TIMEOUT)
   {
      //Non-blocking socket?
      if(sock->timeout == 0)
      {
         //The connection cannot be completed immediately
         socketSetErrnoCode(sock, EINPROGRESS);
      }
      else
      {
         //Timeout while attempting connection
         socketSetErrnoCode(sock, ETIMEDOUT);
      }
 
      //Report an error
      return SOCKET_ERROR;
   }
   else
   {
      //Report an error
      socketTranslateErrorCode(sock, error);
      return SOCKET_ERROR;
   }
}
 
 
/**
 * @brief Place a socket in the listening state
 *
 * Place a socket in a state in which it is listening for an incoming connection
 *
 * @param[in] s Descriptor identifying a bound, unconnected socket
 * @param[in] backlog Maximum length of the queue of pending connections
 * @return If no error occurs, listen returns SOCKET_SUCCESS.
 *   Otherwise, it returns SOCKET_ERROR
 **/
 
int_t listen(int_t s, int_t backlog)
{
   error_t error;
   Socket *sock;
 
   //Make sure the socket descriptor is valid
   if(s < 0 || s >= SOCKET_MAX_COUNT)
   {
      return SOCKET_ERROR;
   }
 
   //Point to the socket structure
   sock = &socketTable[s];
 
   //Place the socket in the listening state
   error = socketListen(sock, backlog);
 
   //Any error to report?
   if(error)
   {
      socketTranslateErrorCode(sock, error);
      return SOCKET_ERROR;
   }
 
   //Successful processing
   return SOCKET_SUCCESS;
}
 
 
/**
 * @brief Permit an incoming connection attempt on a socket
 * @param[in] s Descriptor that identifies a socket in the listening state
 * @param[out] addr Address of the connecting entity (optional)
 * @param[in,out] addrlen Length in bytes of the address (optional)
 * @return If no error occurs, accept returns a descriptor for the new socket.
 *   Otherwise, it returns SOCKET_ERROR
 **/
 
int_t accept(int_t s, struct sockaddr *addr, socklen_t *addrlen)
{
   uint16_t port;
   IpAddr ipAddr;
   Socket *sock;
   Socket *newSock;
 
   //Make sure the socket descriptor is valid
   if(s < 0 || s >= SOCKET_MAX_COUNT)
   {
      return SOCKET_ERROR;
   }
 
   //Point to the socket structure
   sock = &socketTable[s];
 
   //Permit an incoming connection attempt on a socket
   newSock = socketAccept(sock, &ipAddr, &port);
 
   //No connection request is pending in the SYN queue?
   if(newSock == NULL)
   {
      //Report an error
      socketSetErrnoCode(sock, EWOULDBLOCK);
      return SOCKET_ERROR;
   }
 
   //The address parameter is optional
   if(addr != NULL && addrlen != NULL)
   {
#if (IPV4_SUPPORT == ENABLED)
      //IPv4 address?
      if(ipAddr.length == sizeof(Ipv4Addr) &&
         *addrlen >= (socklen_t) sizeof(SOCKADDR_IN))
      {
         //Point to the IPv4 address information
         SOCKADDR_IN *sa = (SOCKADDR_IN *) addr;
 
         //Set address family and port number
         sa->sin_family = AF_INET;
         sa->sin_port = htons(port);
 
         //Copy IPv4 address
         sa->sin_addr.s_addr = ipAddr.ipv4Addr;
 
         //Return the actual length of the address
         *addrlen = sizeof(SOCKADDR_IN);
      }
      else
#endif
#if (IPV6_SUPPORT == ENABLED)
      //IPv6 address?
      if(ipAddr.length == sizeof(Ipv6Addr) &&
         *addrlen >= (socklen_t) sizeof(SOCKADDR_IN6))
      {
         //Point to the IPv6 address information
         SOCKADDR_IN6 *sa = (SOCKADDR_IN6 *) addr;
 
         //Set address family and port number
         sa->sin6_family = AF_INET6;
         sa->sin6_port = htons(port);
         sa->sin6_flowinfo = 0;
         sa->sin6_scope_id = 0;
 
         //Copy IPv6 address
         ipv6CopyAddr(sa->sin6_addr.s6_addr, &ipAddr.ipv6Addr);
 
         //Return the actual length of the address
         *addrlen = sizeof(SOCKADDR_IN6);
      }
      else
#endif
      //Invalid address?
      {
         //Close socket
         socketClose(newSock);
 
         //Report an error
         socketSetErrnoCode(sock, EINVAL);
         return SOCKET_ERROR;
      }
   }
 
   //Return the descriptor to the new socket
   return newSock->descriptor;
}
 
 
/**
 * @brief Send data to a connected socket
 * @param[in] s Descriptor that identifies a connected socket
 * @param[in] data Pointer to a buffer containing the data to be transmitted
 * @param[in] length Number of bytes to be transmitted
 * @param[in] flags Set of flags that influences the behavior of this function
 * @return If no error occurs, send returns the total number of bytes sent,
 *   which can be less than the number requested to be sent in the
 *   length parameter. Otherwise, a value of SOCKET_ERROR is returned
 **/
 
int_t send(int_t s, const void *data, size_t length, int_t flags)
{
   error_t error;
   size_t written;
   uint_t socketFlags;
   Socket *sock;
 
   //Make sure the socket descriptor is valid
   if(s < 0 || s >= SOCKET_MAX_COUNT)
   {
      return SOCKET_ERROR;
   }
 
   //Point to the socket structure
   sock = &socketTable[s];
 
   //The flags parameter can be used to influence the behavior of the function
   socketFlags = 0;
 
   //The MSG_DONTROUTE flag specifies that the data should not be subject
   //to routing
   if((flags & MSG_DONTROUTE) != 0)
   {
      socketFlags |= SOCKET_FLAG_DONT_ROUTE;
   }
 
   //The TCP_NODELAY option disables the Nagle algorithm for TCP sockets
   if((sock->options & SOCKET_OPTION_TCP_NO_DELAY) != 0)
   {
      socketFlags |= SOCKET_FLAG_NO_DELAY;
   }
 
   //Send data
   error = socketSend(sock, data, length, &written, socketFlags);
 
   //Any error to report?
   if(error == ERROR_TIMEOUT)
   {
      //Check whether some data has been written
      if(written > 0)
      {
         //If a timeout error occurs and some data has been written, the
         //count of bytes transferred so far is returned...
      }
      else
      {
         //If no data has been written, a value of SOCKET_ERROR is returned
         socketTranslateErrorCode(sock, error);
         return SOCKET_ERROR;
      }
   }
   else if(error != NO_ERROR)
   {
      //Otherwise, a value of SOCKET_ERROR is returned
      socketTranslateErrorCode(sock, error);
      return SOCKET_ERROR;
   }
 
   //Return the number of bytes transferred so far
   return written;
}
 
 
/**
 * @brief Send a datagram to a specific destination
 * @param[in] s Descriptor that identifies a socket
 * @param[in] data Pointer to a buffer containing the data to be transmitted
 * @param[in] length Number of bytes to be transmitted
 * @param[in] flags Set of flags that influences the behavior of this function
 * @param[in] addr Destination address
 * @param[in] addrlen Length in bytes of the destination address
 * @return If no error occurs, sendto returns the total number of bytes sent,
 *   which can be less than the number requested to be sent in the
 *   length parameter. Otherwise, a value of SOCKET_ERROR is returned
 **/
 
int_t sendto(int_t s, const void *data, size_t length, int_t flags,
   const struct sockaddr *addr, socklen_t addrlen)
{
   error_t error;
   size_t written;
   uint_t socketFlags;
   uint16_t port;
   IpAddr ipAddr;
   Socket *sock;
 
   //Make sure the socket descriptor is valid
   if(s < 0 || s >= SOCKET_MAX_COUNT)
   {
      return SOCKET_ERROR;
   }
 
   //Point to the socket structure
   sock = &socketTable[s];
 
   //The flags parameter can be used to influence the behavior of the function
   socketFlags = 0;
 
   //The MSG_DONTROUTE flag specifies that the data should not be subject
   //to routing
   if((flags & MSG_DONTROUTE) != 0)
   {
      socketFlags |= SOCKET_FLAG_DONT_ROUTE;
   }
 
   //The TCP_NODELAY option disables the Nagle algorithm for TCP sockets
   if((sock->options & SOCKET_OPTION_TCP_NO_DELAY) != 0)
   {
      socketFlags |= SOCKET_FLAG_NO_DELAY;
   }
 
   //Check the length of the address
   if(addrlen < (socklen_t) sizeof(SOCKADDR))
   {
      //Report an error
      socketSetErrnoCode(sock, EINVAL);
      return SOCKET_ERROR;
   }
 
#if (IPV4_SUPPORT == ENABLED)
   //IPv4 address?
   if(addr->sa_family == AF_INET &&
      addrlen >= (socklen_t) sizeof(SOCKADDR_IN))
   {
      //Point to the IPv4 address information
      SOCKADDR_IN *sa = (SOCKADDR_IN *) addr;
 
      //Get port number
      port = ntohs(sa->sin_port);
 
      //Copy IPv4 address
      ipAddr.length = sizeof(Ipv4Addr);
      ipAddr.ipv4Addr = sa->sin_addr.s_addr;
   }
   else
#endif
#if (IPV6_SUPPORT == ENABLED)
   //IPv6 address?
   if(addr->sa_family == AF_INET6 &&
      addrlen >= (socklen_t) sizeof(SOCKADDR_IN6))
   {
      //Point to the IPv6 address information
      SOCKADDR_IN6 *sa = (SOCKADDR_IN6 *) addr;
 
      //Get port number
      port = ntohs(sa->sin6_port);
 
      //Copy IPv6 address
      ipAddr.length = sizeof(Ipv6Addr);
      ipv6CopyAddr(&ipAddr.ipv6Addr, sa->sin6_addr.s6_addr);
   }
   else
#endif
   //Invalid address?
   {
      //Report an error
      socketSetErrnoCode(sock, EINVAL);
      return SOCKET_ERROR;
   }
 
   //Send data
   error = socketSendTo(sock, &ipAddr, port, data, length, &written,
      socketFlags);
 
   //Any error to report?
   if(error == ERROR_TIMEOUT)
   {
      //Check whether some data has been written
      if(written > 0)
      {
         //If a timeout error occurs and some data has been written, the
         //count of bytes transferred so far is returned...
      }
      else
      {
         //If no data has been written, a value of SOCKET_ERROR is returned
         socketTranslateErrorCode(sock, error);
         return SOCKET_ERROR;
      }
   }
   else if(error != NO_ERROR)
   {
      //Otherwise, a value of SOCKET_ERROR is returned
      socketTranslateErrorCode(sock, error);
      return SOCKET_ERROR;
   }
 
   //Return the number of bytes transferred so far
   return written;
}
 
 
/**
 * @brief Send a message
 * @param[in] s Descriptor that identifies a socket
 * @param[in] msg Pointer to the structure describing the message
 * @param[in] flags Set of flags that influences the behavior of this function
 * @return If no error occurs, sendmsg returns the total number of bytes sent,
 *   which can be less than the number requested to be sent in the
 *   length parameter. Otherwise, a value of SOCKET_ERROR is returned
 **/
 
int_t sendmsg(int_t s, struct msghdr *msg, int_t flags)
{
   error_t error;
   uint_t socketFlags;
   Socket *sock;
   SocketMsg message;
   SOCKADDR *addr;
 
   //Make sure the socket descriptor is valid
   if(s < 0 || s >= SOCKET_MAX_COUNT)
   {
      return SOCKET_ERROR;
   }
 
   //Point to the socket structure
   sock = &socketTable[s];
 
   //Check parameters
   if(msg == NULL || msg->msg_iov == NULL || msg->msg_iovlen != 1)
   {
      socketSetErrnoCode(sock, EINVAL);
      return SOCKET_ERROR;
   }
 
   //Point to the message to be transmitted
   message = SOCKET_DEFAULT_MSG;
   message.data = msg->msg_iov[0].iov_base;
   message.length = msg->msg_iov[0].iov_len;
 
   //Check the length of the address
   if(msg->msg_namelen < (socklen_t) sizeof(SOCKADDR))
   {
      //Report an error
      socketSetErrnoCode(sock, EINVAL);
      return SOCKET_ERROR;
   }
 
   //Point to the destination address
   addr = (SOCKADDR *) msg->msg_name;
 
#if (IPV4_SUPPORT == ENABLED)
   //IPv4 address?
   if(addr->sa_family == AF_INET &&
      msg->msg_namelen >= (socklen_t) sizeof(SOCKADDR_IN))
   {
      //Point to the IPv4 address information
      SOCKADDR_IN *sa = (SOCKADDR_IN *) addr;
 
      //Get port number
      message.destPort = ntohs(sa->sin_port);
 
      //Copy IPv4 address
      message.destIpAddr.length = sizeof(Ipv4Addr);
      message.destIpAddr.ipv4Addr = sa->sin_addr.s_addr;
   }
   else
#endif
#if (IPV6_SUPPORT == ENABLED)
   //IPv6 address?
   if(addr->sa_family == AF_INET6 &&
      msg->msg_namelen >= (socklen_t) sizeof(SOCKADDR_IN6))
   {
      //Point to the IPv6 address information
      SOCKADDR_IN6 *sa = (SOCKADDR_IN6 *) addr;
 
      //Get port number
      message.destPort = ntohs(sa->sin6_port);
 
      //Copy IPv6 address
      message.destIpAddr.length = sizeof(Ipv6Addr);
      ipv6CopyAddr(&message.destIpAddr.ipv6Addr, sa->sin6_addr.s6_addr);
   }
   else
#endif
   //Invalid address?
   {
      //Report an error
      socketSetErrnoCode(sock, EINVAL);
      return SOCKET_ERROR;
   }
 
   //The ancillary data buffer parameter is optional
   if(msg->msg_control != NULL)
   {
      uint_t n;
      int_t *val;
      CMSGHDR *cmsg;
 
      //Point to the first control message
      n = 0;
 
      //Loop through control messages
      while((n + sizeof(CMSGHDR)) <= msg->msg_controllen)
      {
         //Point to the ancillary data header
         cmsg = (CMSGHDR *) ((uint8_t *) msg->msg_control + n);
 
         //Check the length of the control message
         if(cmsg->cmsg_len >= sizeof(CMSGHDR) &&
            cmsg->cmsg_len <= (msg->msg_controllen - n))
         {
#if (IPV4_SUPPORT == ENABLED)
            //IPv4 protocol?
            if(addr->sa_family == AF_INET && cmsg->cmsg_level == IPPROTO_IP)
            {
               //Check control message type
               if(cmsg->cmsg_type == IP_PKTINFO &&
                  cmsg->cmsg_len >= CMSG_LEN(sizeof(IN_PKTINFO)))
               {
                  //Point to the ancillary data value
                  IN_PKTINFO *pktInfo = (IN_PKTINFO *) CMSG_DATA(cmsg);
 
                  //Specify source IPv4 address
                  message.srcIpAddr.length = sizeof(Ipv4Addr);
                  message.srcIpAddr.ipv4Addr = pktInfo->ipi_addr.s_addr;
               }
               else if(cmsg->cmsg_type == IP_TOS &&
                  cmsg->cmsg_len >= CMSG_LEN(sizeof(int_t)))
               {
                  //Point to the ancillary data value
                  val = (int_t *) CMSG_DATA(cmsg);
                  //Specify ToS value
                  message.tos = (uint8_t) *val;
               }
               else if(cmsg->cmsg_type == IP_TTL &&
                  cmsg->cmsg_len >= CMSG_LEN(sizeof(int_t)))
               {
                  //Point to the ancillary data value
                  val = (int_t *) CMSG_DATA(cmsg);
                  //Specify TTL value
                  message.ttl = (uint8_t) *val;
               }
               else if(cmsg->cmsg_type == IP_DONTFRAG &&
                  cmsg->cmsg_len >= CMSG_LEN(sizeof(int_t)))
               {
                  //Point to the ancillary data value
                  val = (int_t *) CMSG_DATA(cmsg);
 
                  //This option can be used to set the "don't fragment" flag
                  //on IP packets
                  message.dontFrag = (*val != 0) ? TRUE : FALSE;
               }
               else
               {
                  //Unknown control message type
               }
            }
            else
#endif
#if (IPV6_SUPPORT == ENABLED)
            //IPv6 protocol?
            if(addr->sa_family == AF_INET6 && cmsg->cmsg_level == IPPROTO_IPV6)
            {
               //Check control message type
               if(cmsg->cmsg_type == IPV6_PKTINFO &&
                  cmsg->cmsg_len >= CMSG_LEN(sizeof(IN_PKTINFO)))
               {
                  //Point to the ancillary data value
                  IN6_PKTINFO *pktInfo = (IN6_PKTINFO *) CMSG_DATA(cmsg);
 
                  //Specify source IPv6 address
                  message.srcIpAddr.length = sizeof(Ipv6Addr);
                  ipv6CopyAddr(&message.srcIpAddr.ipv6Addr, pktInfo->ipi6_addr.s6_addr);
               }
               else if(cmsg->cmsg_type == IPV6_TCLASS &&
                  cmsg->cmsg_len >= CMSG_LEN(sizeof(int_t)))
               {
                  //Point to the ancillary data value
                  val = (int_t *) CMSG_DATA(cmsg);
                  //Specify Traffic Class value
                  message.tos = (uint8_t) *val;
               }
               else if(cmsg->cmsg_type == IPV6_HOPLIMIT &&
                  cmsg->cmsg_len >= CMSG_LEN(sizeof(int_t)))
               {
                  //Point to the ancillary data value
                  val = (int_t *) CMSG_DATA(cmsg);
                  //Specify Hop Limit value
                  message.ttl = (uint8_t) *val;
               }
               else if(cmsg->cmsg_type == IPV6_DONTFRAG &&
                  cmsg->cmsg_len >= CMSG_LEN(sizeof(int_t)))
               {
                  //Point to the ancillary data value
                  val = (int_t *) CMSG_DATA(cmsg);
 
                  //This option be used to turn off the automatic inserting
                  //of a fragment header for UDP and raw sockets
                  message.dontFrag = (*val != 0) ? TRUE : FALSE;
               }
               else
               {
                  //Unknown control message type
               }
            }
            //Unknown protocol?
            else
#endif
            {
               //Discard control message
            }
 
            //Next control message
            n += cmsg->cmsg_len;
         }
         else
         {
            //Malformed control message
            break;
         }
      }
   }
 
   //The flags parameter can be used to influence the behavior of the function
   socketFlags = 0;
 
   //The MSG_DONTROUTE flag specifies that the data should not be subject
   //to routing
   if((flags & MSG_DONTROUTE) != 0)
   {
      socketFlags |= SOCKET_FLAG_DONT_ROUTE;
   }
 
   //The TCP_NODELAY option disables the Nagle algorithm for TCP sockets
   if((sock->options & SOCKET_OPTION_TCP_NO_DELAY) != 0)
   {
      socketFlags |= SOCKET_FLAG_NO_DELAY;
   }
 
   //Send message
   error = socketSendMsg(sock, &message, socketFlags);
 
   //Any error to report?
   if(error != NO_ERROR)
   {
      //Otherwise, a value of SOCKET_ERROR is returned
      socketTranslateErrorCode(sock, error);
      return SOCKET_ERROR;
   }
 
   //Return the number of bytes transferred so far
   return message.length;
}
 
 
/**
 * @brief Receive data from a connected socket
 * @param[in] s Descriptor that identifies a connected socket
 * @param[out] data Buffer where to store the incoming data
 * @param[in] size Maximum number of bytes that can be received
 * @param[in] flags Set of flags that influences the behavior of this function
 * @return If no error occurs, recv returns the number of bytes received. If the
 *   connection has been gracefully closed, the return value is zero.
 *   Otherwise, a value of SOCKET_ERROR is returned
 **/
 
int_t recv(int_t s, void *data, size_t size, int_t flags)
{
   error_t error;
   size_t received;
   uint_t socketFlags;
   Socket *sock;
 
   //Make sure the socket descriptor is valid
   if(s < 0 || s >= SOCKET_MAX_COUNT)
   {
      return SOCKET_ERROR;
   }
 
   //Point to the socket structure
   sock = &socketTable[s];
 
   //The flags parameter can be used to influence the behavior of the function
   socketFlags = 0;
 
   //When the MSG_PEEK flag is specified, the data is copied into the buffer,
   //but is not removed from the input queue
   if((flags & MSG_PEEK) != 0)
   {
      socketFlags |= SOCKET_FLAG_PEEK;
   }
 
   //When the MSG_WAITALL flag is specified, the receive request will complete
   //when the buffer supplied by the caller is completely full
   if((flags & MSG_WAITALL) != 0)
   {
      socketFlags |= SOCKET_FLAG_WAIT_ALL;
   }
 
   //The MSG_DONTWAIT flag enables non-blocking operation
   if((flags & MSG_DONTWAIT) != 0)
   {
      socketFlags |= SOCKET_FLAG_DONT_WAIT;
   }
 
   //Receive data
   error = socketReceive(sock, data, size, &received, socketFlags);
 
   //Any error to report?
   if(error == ERROR_END_OF_STREAM)
   {
      //If the connection has been gracefully closed, the return value is zero
      return 0;
   }
   else if(error != NO_ERROR)
   {
      //Otherwise, a value of SOCKET_ERROR is returned
      socketTranslateErrorCode(sock, error);
      return SOCKET_ERROR;
   }
 
   //Return the number of bytes received
   return received;
}
 
 
/**
 * @brief Receive a datagram
 * @param[in] s Descriptor that identifies a socket
 * @param[out] data Buffer where to store the incoming data
 * @param[in] size Maximum number of bytes that can be received
 * @param[in] flags Set of flags that influences the behavior of this function
 * @param[out] addr Source address upon return (optional)
 * @param[in,out] addrlen Length in bytes of the address (optional)
 * @return If no error occurs, recvfrom returns the number of bytes received.
 *   Otherwise, a value of SOCKET_ERROR is returned
 **/
 
int_t recvfrom(int_t s, void *data, size_t size, int_t flags,
   struct sockaddr *addr, socklen_t *addrlen)
{
   error_t error;
   size_t received;
   uint_t socketFlags;
   uint16_t port;
   IpAddr ipAddr;
   Socket *sock;
 
   //Make sure the socket descriptor is valid
   if(s < 0 || s >= SOCKET_MAX_COUNT)
   {
      return SOCKET_ERROR;
   }
 
   //Point to the socket structure
   sock = &socketTable[s];
 
   //The flags parameter can be used to influence the behavior of the function
   socketFlags = 0;
 
   //When the MSG_PEEK flag is specified, the data is copied into the buffer,
   //but is not removed from the input queue
   if((flags & MSG_PEEK) != 0)
   {
      socketFlags |= SOCKET_FLAG_PEEK;
   }
 
   //When the MSG_WAITALL flag is specified, the receive request will complete
   //when the buffer supplied by the caller is completely full
   if((flags & MSG_WAITALL) != 0)
   {
      socketFlags |= SOCKET_FLAG_WAIT_ALL;
   }
 
   //The MSG_DONTWAIT flag enables non-blocking operation
   if((flags & MSG_DONTWAIT) != 0)
   {
      socketFlags |= SOCKET_FLAG_DONT_WAIT;
   }
 
   //Receive data
   error = socketReceiveFrom(sock, &ipAddr, &port, data, size, &received,
      socketFlags);
 
   //Any error to report?
   if(error == ERROR_END_OF_STREAM)
   {
      //If the connection has been gracefully closed, the return value is zero
      return 0;
   }
   else if(error != NO_ERROR)
   {
      //Otherwise, a value of SOCKET_ERROR is returned
      socketTranslateErrorCode(sock, error);
      return SOCKET_ERROR;
   }
 
   //The address parameter is optional
   if(addr != NULL && addrlen != NULL)
   {
#if (IPV4_SUPPORT == ENABLED)
      //IPv4 address?
      if(ipAddr.length == sizeof(Ipv4Addr) &&
         *addrlen >= (socklen_t) sizeof(SOCKADDR_IN))
      {
         //Point to the IPv4 address information
         SOCKADDR_IN *sa = (SOCKADDR_IN *) addr;
 
         //Set address family and port number
         sa->sin_family = AF_INET;
         sa->sin_port = htons(port);
 
         //Copy IPv4 address
         sa->sin_addr.s_addr = ipAddr.ipv4Addr;
 
         //Return the actual length of the address
         *addrlen = sizeof(SOCKADDR_IN);
      }
      else
#endif
#if (IPV6_SUPPORT == ENABLED)
      //IPv6 address?
      if(ipAddr.length == sizeof(Ipv6Addr) &&
         *addrlen >= (socklen_t) sizeof(SOCKADDR_IN6))
      {
         //Point to the IPv6 address information
         SOCKADDR_IN6 *sa = (SOCKADDR_IN6 *) addr;
 
         //Set address family and port number
         sa->sin6_family = AF_INET6;
         sa->sin6_port = htons(port);
         sa->sin6_flowinfo = 0;
         sa->sin6_scope_id = 0;
 
         //Copy IPv6 address
         ipv6CopyAddr(sa->sin6_addr.s6_addr, &ipAddr.ipv6Addr);
 
         //Return the actual length of the address
         *addrlen = sizeof(SOCKADDR_IN6);
      }
      else
#endif
      //Invalid address?
      {
         //Report an error
         socketSetErrnoCode(sock, EINVAL);
         return SOCKET_ERROR;
      }
   }
 
   //Return the number of bytes received
   return received;
}
 
 
/**
 * @brief Receive a message
 * @param[in] s Descriptor that identifies a socket
 * @param[in,out] msg Pointer to the structure describing the message
 * @param[in] flags Set of flags that influences the behavior of this function
 * @return If no error occurs, recvmsg returns the number of bytes received.
 *   Otherwise, a value of SOCKET_ERROR is returned
 **/
 
int_t recvmsg(int_t s, struct msghdr *msg, int_t flags)
{
   error_t error;
   size_t n;
   uint_t socketFlags;
   Socket *sock;
   SocketMsg message;
 
   //Make sure the socket descriptor is valid
   if(s < 0 || s >= SOCKET_MAX_COUNT)
   {
      return SOCKET_ERROR;
   }
 
   //Point to the socket structure
   sock = &socketTable[s];
 
   //Check parameters
   if(msg == NULL || msg->msg_iov == NULL || msg->msg_iovlen != 1)
   {
      socketSetErrnoCode(sock, EINVAL);
      return SOCKET_ERROR;
   }
 
   //Point to the receive buffer
   message = SOCKET_DEFAULT_MSG;
   message.data = msg->msg_iov[0].iov_base;
   message.size = msg->msg_iov[0].iov_len;
 
   //The flags parameter can be used to influence the behavior of the function
   socketFlags = 0;
 
   //When the MSG_PEEK flag is specified, the data is copied into the buffer,
   //but is not removed from the input queue
   if((flags & MSG_PEEK) != 0)
   {
      socketFlags |= SOCKET_FLAG_PEEK;
   }
 
   //When the MSG_WAITALL flag is specified, the receive request will complete
   //when the buffer supplied by the caller is completely full
   if((flags & MSG_WAITALL) != 0)
   {
      socketFlags |= SOCKET_FLAG_WAIT_ALL;
   }
 
   //The MSG_DONTWAIT flag enables non-blocking operation
   if((flags & MSG_DONTWAIT) != 0)
   {
      socketFlags |= SOCKET_FLAG_DONT_WAIT;
   }
 
   //Receive message
   error = socketReceiveMsg(sock, &message, socketFlags);
 
   //Any error to report?
   if(error)
   {
      socketTranslateErrorCode(sock, error);
      return SOCKET_ERROR;
   }
 
   //The source address parameter is optional
   if(msg->msg_name != NULL)
   {
#if (IPV4_SUPPORT == ENABLED)
      //IPv4 address?
      if(message.srcIpAddr.length == sizeof(Ipv4Addr) &&
         msg->msg_namelen >= (socklen_t) sizeof(SOCKADDR_IN))
      {
         //Point to the IPv4 address information
         SOCKADDR_IN *sa = (SOCKADDR_IN *) msg->msg_name;
 
         //Set address family and port number
         sa->sin_family = AF_INET;
         sa->sin_port = htons(message.srcPort);
 
         //Copy IPv4 address
         sa->sin_addr.s_addr = message.srcIpAddr.ipv4Addr;
 
         //Return the actual length of the address
         msg->msg_namelen = sizeof(SOCKADDR_IN);
      }
      else
#endif
#if (IPV6_SUPPORT == ENABLED)
      //IPv6 address?
      if(message.srcIpAddr.length == sizeof(Ipv6Addr) &&
         msg->msg_namelen >= (socklen_t) sizeof(SOCKADDR_IN6))
      {
         //Point to the IPv6 address information
         SOCKADDR_IN6 *sa = (SOCKADDR_IN6 *) msg->msg_name;
 
         //Set address family and port number
         sa->sin6_family = AF_INET6;
         sa->sin6_port = htons(message.srcPort);
         sa->sin6_flowinfo = 0;
         sa->sin6_scope_id = 0;
 
         //Copy IPv6 address
         ipv6CopyAddr(sa->sin6_addr.s6_addr, &message.srcIpAddr.ipv6Addr);
 
         //Return the actual length of the address
         msg->msg_namelen = sizeof(SOCKADDR_IN6);
      }
      else
#endif
      //Invalid address?
      {
         //Report an error
         socketSetErrnoCode(sock, EINVAL);
         return SOCKET_ERROR;
      }
   }
   else
   {
      msg->msg_namelen = 0;
   }
 
   //Clear flags
   msg->msg_flags = 0;
 
   //Length of the ancillary data buffer
   n = 0;
 
   //The ancillary data buffer parameter is optional
   if(msg->msg_control != NULL)
   {
#if (IPV4_SUPPORT == ENABLED)
      //IPv4 address?
      if(message.destIpAddr.length == sizeof(Ipv4Addr))
      {
         int_t *val;
         CMSGHDR *cmsg;
         IN_PKTINFO *pktInfo;
 
         //The IP_PKTINFO option allows an application to enable or disable
         //the return of IPv4 packet information
         if((sock->options & SOCKET_OPTION_IPV4_PKT_INFO) != 0)
         {
            //Make sure there is enough room to add the control message
            if((n + CMSG_SPACE(sizeof(IN_PKTINFO))) <= msg->msg_controllen)
            {
               //Point to the ancillary data header
               cmsg = (CMSGHDR *) ((uint8_t *) msg->msg_control + n);
 
               //Format ancillary data header
               cmsg->cmsg_len = CMSG_LEN(sizeof(IN_PKTINFO));
               cmsg->cmsg_level = IPPROTO_IP;
               cmsg->cmsg_type = IP_PKTINFO;
 
               //Point to the ancillary data value
               pktInfo = (IN_PKTINFO *) CMSG_DATA(cmsg);
 
               //Format packet information
               pktInfo->ipi_ifindex = message.interface->index + 1;
               pktInfo->ipi_addr.s_addr = message.destIpAddr.ipv4Addr;
 
               //Adjust the actual length of the ancillary data buffer
               n += CMSG_SPACE(sizeof(IN_PKTINFO));
            }
            else
            {
               //When the control message buffer is too short to store all
               //messages, the MSG_CTRUNC flag must be set
               msg->msg_flags |= MSG_CTRUNC;
            }
         }
 
         //The IP_RECVTOS option allows an application to enable or disable
         //the return of ToS header field on received datagrams
         if((sock->options & SOCKET_OPTION_IPV4_RECV_TOS) != 0)
         {
            //Make sure there is enough room to add the control message
            if((n + CMSG_SPACE(sizeof(int_t))) <= msg->msg_controllen)
            {
               //Point to the ancillary data header
               cmsg = (CMSGHDR *) ((uint8_t *) msg->msg_control + n);
 
               //Format ancillary data header
               cmsg->cmsg_len = CMSG_LEN(sizeof(int_t));
               cmsg->cmsg_level = IPPROTO_IP;
               cmsg->cmsg_type = IP_TOS;
 
               //Point to the ancillary data value
               val = (int_t *) CMSG_DATA(cmsg);
               //Set ancillary data value
               *val = message.tos;
 
               //Adjust the actual length of the ancillary data buffer
               n += CMSG_SPACE(sizeof(int_t));
            }
            else
            {
               //When the control message buffer is too short to store all
               //messages, the MSG_CTRUNC flag must be set
               msg->msg_flags |= MSG_CTRUNC;
            }
         }
 
         //The IP_RECVTTL option allows an application to enable or disable
         //the return of TTL header field on received datagrams
         if((sock->options & SOCKET_OPTION_IPV4_RECV_TTL) != 0)
         {
            //Make sure there is enough room to add the control message
            if((n + CMSG_SPACE(sizeof(int_t))) <= msg->msg_controllen)
            {
               //Point to the ancillary data header
               cmsg = (CMSGHDR *) ((uint8_t *) msg->msg_control + n);
 
               //Format ancillary data header
               cmsg->cmsg_len = CMSG_LEN(sizeof(int_t));
               cmsg->cmsg_level = IPPROTO_IP;
               cmsg->cmsg_type = IP_TTL;
 
               //Point to the ancillary data value
               val = (int_t *) CMSG_DATA(cmsg);
               //Set ancillary data value
               *val = message.ttl;
 
               //Adjust the actual length of the ancillary data buffer
               n += CMSG_SPACE(sizeof(int_t));
            }
            else
            {
               //When the control message buffer is too short to store all
               //messages, the MSG_CTRUNC flag must be set
               msg->msg_flags |= MSG_CTRUNC;
            }
         }
      }
      else
#endif
#if (IPV6_SUPPORT == ENABLED)
      //IPv6 address?
      if(message.destIpAddr.length == sizeof(Ipv6Addr))
      {
         int_t *val;
         CMSGHDR *cmsg;
         IN6_PKTINFO *pktInfo;
 
         //The IPV6_PKTINFO option allows an application to enable or disable
         //the return of IPv6 packet information
         if((sock->options & SOCKET_OPTION_IPV6_PKT_INFO) != 0)
         {
            //Make sure there is enough room to add the control message
            if((n + CMSG_SPACE(sizeof(IN6_PKTINFO))) <= msg->msg_controllen)
            {
               //Point to the ancillary data header
               cmsg = (CMSGHDR *) ((uint8_t *) msg->msg_control + n);
 
               //Format ancillary data header
               cmsg->cmsg_len = CMSG_LEN(sizeof(IN6_PKTINFO));
               cmsg->cmsg_level = IPPROTO_IPV6;
               cmsg->cmsg_type = IPV6_PKTINFO;
 
               //Point to the ancillary data value
               pktInfo = (IN6_PKTINFO *) CMSG_DATA(cmsg);
 
               //Format packet information
               pktInfo->ipi6_ifindex = message.interface->index + 1;
               ipv6CopyAddr(pktInfo->ipi6_addr.s6_addr, &message.destIpAddr.ipv6Addr);
 
               //Adjust the actual length of the ancillary data buffer
               n += CMSG_SPACE(sizeof(IN6_PKTINFO));
            }
            else
            {
               //When the control message buffer is too short to store all
               //messages, the MSG_CTRUNC flag must be set
               msg->msg_flags |= MSG_CTRUNC;
            }
         }
 
         //The IPV6_RECVTCLASS option allows an application to enable or disable
         //the return of Traffic Class header field on received datagrams
         if((sock->options & SOCKET_OPTION_IPV6_RECV_TRAFFIC_CLASS) != 0)
         {
            //Make sure there is enough room to add the control message
            if((n + CMSG_SPACE(sizeof(int_t))) <= msg->msg_controllen)
            {
               //Point to the ancillary data header
               cmsg = (CMSGHDR *) ((uint8_t *) msg->msg_control + n);
 
               //Format ancillary data header
               cmsg->cmsg_len = CMSG_LEN(sizeof(int_t));
               cmsg->cmsg_level = IPPROTO_IPV6;
               cmsg->cmsg_type = IPV6_TCLASS;
 
               //Point to the ancillary data value
               val = (int_t *) CMSG_DATA(cmsg);
               //Set ancillary data value
               *val = message.tos;
 
               //Adjust the actual length of the ancillary data buffer
               n += CMSG_SPACE(sizeof(int_t));
            }
            else
            {
               //When the control message buffer is too short to store all
               //messages, the MSG_CTRUNC flag must be set
               msg->msg_flags |= MSG_CTRUNC;
            }
         }
 
         //The IPV6_RECVHOPLIMIT option allows an application to enable or
         //disable the return of Hop Limit header field on received datagrams
         if((sock->options & SOCKET_OPTION_IPV6_RECV_HOP_LIMIT) != 0)
         {
            //Make sure there is enough room to add the control message
            if((n + CMSG_SPACE(sizeof(int_t))) <= msg->msg_controllen)
            {
               //Point to the ancillary data header
               cmsg = (CMSGHDR *) ((uint8_t *) msg->msg_control + n);
 
               //Format ancillary data header
               cmsg->cmsg_len = CMSG_LEN(sizeof(int_t));
               cmsg->cmsg_level = IPPROTO_IPV6;
               cmsg->cmsg_type = IPV6_HOPLIMIT;
 
               //Point to the ancillary data value
               val = (int_t *) CMSG_DATA(cmsg);
               //Set ancillary data value
               *val = message.ttl;
 
               //Adjust the actual length of the ancillary data buffer
               n += CMSG_SPACE(sizeof(int_t));
            }
            else
            {
               //When the control message buffer is too short to store all
               //messages, the MSG_CTRUNC flag must be set
               msg->msg_flags |= MSG_CTRUNC;
            }
         }
      }
      else
#endif
      //Invalid address?
      {
         //Just for sanity
      }
   }
 
   //Length of the actual length of the ancillary data buffer
   msg->msg_controllen = n;
 
   //Return the number of bytes received
   return message.length;
}
 
 
/**
 * @brief Retrieves the local name for a socket
 * @param[in] s Descriptor identifying a socket
 * @param[out] addr Address of the socket
 * @param[in,out] addrlen Length in bytes of the address
 * @return If no error occurs, getsockname returns SOCKET_SUCCESS
 *   Otherwise, it returns SOCKET_ERROR
 **/
 
int_t getsockname(int_t s, struct sockaddr *addr, socklen_t *addrlen)
{
   int_t ret;
   Socket *sock;
 
   //Make sure the socket descriptor is valid
   if(s < 0 || s >= SOCKET_MAX_COUNT)
   {
      return SOCKET_ERROR;
   }
 
   //Point to the socket structure
   sock = &socketTable[s];
 
   //Get exclusive access
   osAcquireMutex(&netMutex);
 
   //Check whether the socket has been bound to an address
   if(sock->localIpAddr.length != 0)
   {
#if (IPV4_SUPPORT == ENABLED)
      //IPv4 address?
      if(sock->localIpAddr.length == sizeof(Ipv4Addr) &&
         *addrlen >= (socklen_t) sizeof(SOCKADDR_IN))
      {
         //Point to the IPv4 address information
         SOCKADDR_IN *sa = (SOCKADDR_IN *) addr;
 
         //Set address family and port number
         sa->sin_family = AF_INET;
         sa->sin_port = htons(sock->localPort);
 
         //Copy IPv4 address
         sa->sin_addr.s_addr = sock->localIpAddr.ipv4Addr;
 
         //Return the actual length of the address
         *addrlen = sizeof(SOCKADDR_IN);
         //Successful processing
         ret = SOCKET_SUCCESS;
      }
      else
#endif
#if (IPV6_SUPPORT == ENABLED)
      //IPv6 address?
      if(sock->localIpAddr.length == sizeof(Ipv6Addr) &&
         *addrlen >= (socklen_t) sizeof(SOCKADDR_IN6))
      {
         //Point to the IPv6 address information
         SOCKADDR_IN6 *sa = (SOCKADDR_IN6 *) addr;
 
         //Set address family and port number
         sa->sin6_family = AF_INET6;
         sa->sin6_port = htons(sock->localPort);
         sa->sin6_flowinfo = 0;
         sa->sin6_scope_id = 0;
 
         //Copy IPv6 address
         ipv6CopyAddr(sa->sin6_addr.s6_addr, &sock->localIpAddr.ipv6Addr);
 
         //Return the actual length of the address
         *addrlen = sizeof(SOCKADDR_IN6);
         //Successful processing
         ret = SOCKET_SUCCESS;
      }
      else
#endif
      {
         //The specified length is not valid
         socketSetErrnoCode(sock, EINVAL);
         ret = SOCKET_ERROR;
      }
   }
   else
   {
      //The socket is not bound to any address
      socketSetErrnoCode(sock, ENOTCONN);
      ret = SOCKET_ERROR;
   }
 
   //Release exclusive access
   osReleaseMutex(&netMutex);
 
   //return status code
   return ret;
}
 
 
/**
 * @brief Retrieves the address of the peer to which a socket is connected
 * @param[in] s Descriptor identifying a socket
 * @param[out] addr Address of the peer
 * @param[in,out] addrlen Length in bytes of the address
 * @return If no error occurs, getpeername returns SOCKET_SUCCESS
 *   Otherwise, it returns SOCKET_ERROR
 **/
 
int_t getpeername(int_t s, struct sockaddr *addr, socklen_t *addrlen)
{
   int_t ret;
   Socket *sock;
 
   //Make sure the socket descriptor is valid
   if(s < 0 || s >= SOCKET_MAX_COUNT)
   {
      return SOCKET_ERROR;
   }
 
   //Point to the socket structure
   sock = &socketTable[s];
 
   //Get exclusive access
   osAcquireMutex(&netMutex);
 
   //Check whether the socket is connected to a peer
   if(sock->remoteIpAddr.length != 0)
   {
#if (IPV4_SUPPORT == ENABLED)
      //IPv4 address?
      if(sock->remoteIpAddr.length == sizeof(Ipv4Addr) &&
         *addrlen >= (socklen_t) sizeof(SOCKADDR_IN))
      {
         //Point to the IPv4 address information
         SOCKADDR_IN *sa = (SOCKADDR_IN *) addr;
 
         //Set address family and port number
         sa->sin_family = AF_INET;
         sa->sin_port = htons(sock->remotePort);
 
         //Copy IPv4 address
         sa->sin_addr.s_addr = sock->remoteIpAddr.ipv4Addr;
 
         //Return the actual length of the address
         *addrlen = sizeof(SOCKADDR_IN);
         //Successful processing
         ret = SOCKET_SUCCESS;
      }
      else
#endif
#if (IPV6_SUPPORT == ENABLED)
      //IPv6 address?
      if(sock->remoteIpAddr.length == sizeof(Ipv6Addr) &&
         *addrlen >= (socklen_t) sizeof(SOCKADDR_IN6))
      {
         //Point to the IPv6 address information
         SOCKADDR_IN6 *sa = (SOCKADDR_IN6 *) addr;
 
         //Set address family and port number
         sa->sin6_family = AF_INET6;
         sa->sin6_port = htons(sock->remotePort);
         sa->sin6_flowinfo = 0;
         sa->sin6_scope_id = 0;
 
         //Copy IPv6 address
         ipv6CopyAddr(sa->sin6_addr.s6_addr, &sock->remoteIpAddr.ipv6Addr);
 
         //Return the actual length of the address
         *addrlen = sizeof(SOCKADDR_IN6);
         //Successful processing
         ret = SOCKET_SUCCESS;
      }
      else
#endif
      {
         //The specified length is not valid
         socketSetErrnoCode(sock, EINVAL);
         ret = SOCKET_ERROR;
      }
   }
   else
   {
      //The socket is not connected to any peer
      socketSetErrnoCode(sock, ENOTCONN);
      ret = SOCKET_ERROR;
   }
 
   //Release exclusive access
   osReleaseMutex(&netMutex);
 
   //return status code
   return ret;
}
 
 
/**
 * @brief The setsockopt function sets a socket option
 * @param[in] s Descriptor that identifies a socket
 * @param[in] level The level at which the option is defined
 * @param[in] optname The socket option for which the value is to be set
 * @param[in] optval A pointer to the buffer in which the value for the
 *   requested option is specified
 * @param[in] optlen The size, in bytes, of the buffer pointed to by the optval
 *   parameter
 * @return If no error occurs, setsockopt returns SOCKET_SUCCESS
 *   Otherwise, it returns SOCKET_ERROR
 **/
 
int_t setsockopt(int_t s, int_t level, int_t optname, const void *optval,
   socklen_t optlen)
{
   int_t ret;
   Socket *sock;
 
   //Make sure the socket descriptor is valid
   if(s < 0 || s >= SOCKET_MAX_COUNT)
   {
      return SOCKET_ERROR;
   }
 
   //Point to the socket structure
   sock = &socketTable[s];
 
   //Make sure the option is valid
   if(optval != NULL)
   {
      //Check option level
      if(level == SOL_SOCKET)
      {
         //Check option type
         if(optname == SO_REUSEADDR)
         {
            //Set SO_REUSEADDR option
            ret = socketSetSoReuseAddrOption(sock, optval, optlen);
         }
         else if(optname == SO_BROADCAST)
         {
            //Set SO_BROADCAST option
            ret = socketSetSoBroadcastOption(sock, optval, optlen);
         }
         else if(optname == SO_SNDTIMEO)
         {
            //Set SO_SNDTIMEO option
            ret = socketSetSoSndTimeoOption(sock, optval, optlen);
         }
         else if(optname == SO_RCVTIMEO)
         {
            //Set SO_RCVTIMEO option
            ret = socketSetSoRcvTimeoOption(sock, optval, optlen);
         }
         else if(optname == SO_SNDBUF)
         {
            //Set SO_SNDBUF option
            ret = socketSetSoSndBufOption(sock, optval, optlen);
         }
         else if(optname == SO_RCVBUF)
         {
            //Set SO_RCVBUF option
            ret = socketSetSoRcvBufOption(sock, optval, optlen);
         }
         else if(optname == SO_KEEPALIVE)
         {
            //Set SO_KEEPALIVE option
            ret = socketSetSoKeepAliveOption(sock, optval, optlen);
         }
         else if(optname == SO_NO_CHECK)
         {
            //Set SO_NO_CHECK option
            ret = socketSetSoNoCheckOption(sock, optval, optlen);
         }
         else
         {
            //Unknown option
            socketSetErrnoCode(sock, ENOPROTOOPT);
            //Report an error
            ret = SOCKET_ERROR;
         }
      }
      else if(level == IPPROTO_IP)
      {
         //Check option type
         if(optname == IP_TOS)
         {
            //Set IP_TOS option
            ret = socketSetIpTosOption(sock, optval, optlen);
         }
         else if(optname == IP_TTL)
         {
            //Set IP_TTL option
            ret = socketSetIpTtlOption(sock, optval, optlen);
         }
         else if(optname == IP_MULTICAST_IF)
         {
            //Set IP_MULTICAST_IF option
            ret = socketSetIpMulticastIfOption(sock, optval, optlen);
         }
         else if(optname == IP_MULTICAST_TTL)
         {
            //Set IP_MULTICAST_TTL option
            ret = socketSetIpMulticastTtlOption(sock, optval, optlen);
         }
         else if(optname == IP_MULTICAST_LOOP)
         {
            //Set IP_MULTICAST_LOOP option
            ret = socketSetIpMulticastLoopOption(sock, optval, optlen);
         }
         else if(optname == IP_ADD_MEMBERSHIP)
         {
            //Set IP_ADD_MEMBERSHIP option
            ret = socketSetIpAddMembershipOption(sock, optval, optlen);
         }
         else if(optname == IP_DROP_MEMBERSHIP)
         {
            //Set IP_DROP_MEMBERSHIP option
            ret = socketSetIpDropMembershipOption(sock, optval, optlen);
         }
         else if(optname == IP_BLOCK_SOURCE)
         {
            //Set IP_BLOCK_SOURCE option
            ret = socketSetIpBlockSourceOption(sock, optval, optlen);
         }
         else if(optname == IP_UNBLOCK_SOURCE)
         {
            //Set IP_UNBLOCK_SOURCE option
            ret = socketSetIpUnblockSourceOption(sock, optval, optlen);
         }
         else if(optname == IP_ADD_SOURCE_MEMBERSHIP)
         {
            //Set IP_ADD_SOURCE_MEMBERSHIP option
            ret = socketSetIpAddSourceMembershipOption(sock, optval, optlen);
         }
         else if(optname == IP_DROP_SOURCE_MEMBERSHIP)
         {
            //Set IP_DROP_SOURCE_MEMBERSHIP option
            ret = socketSetIpDropSourceMembershipOption(sock, optval, optlen);
         }
         else if(optname == MCAST_JOIN_GROUP)
         {
            //Set MCAST_JOIN_GROUP option
            ret = socketSetMcastJoinGroupOption(sock, optval, optlen);
         }
         else if(optname == MCAST_LEAVE_GROUP)
         {
            //Set MCAST_LEAVE_GROUP option
            ret = socketSetMcastLeaveGroupOption(sock, optval, optlen);
         }
         else if(optname == MCAST_BLOCK_SOURCE)
         {
            //Set MCAST_BLOCK_SOURCE option
            ret = socketSetMcastBlockSourceOption(sock, optval, optlen);
         }
         else if(optname == MCAST_UNBLOCK_SOURCE)
         {
            //Set MCAST_UNBLOCK_SOURCE option
            ret = socketSetMcastUnblockSourceOption(sock, optval, optlen);
         }
         else if(optname == MCAST_JOIN_SOURCE_GROUP)
         {
            //Set MCAST_JOIN_SOURCE_GROUP option
            ret = socketSetMcastJoinSourceGroupOption(sock, optval, optlen);
         }
         else if(optname == MCAST_LEAVE_SOURCE_GROUP)
         {
            //Set MCAST_LEAVE_SOURCE_GROUP option
            ret = socketSetMcastLeaveSourceGroupOption(sock, optval, optlen);
         }
         else if(optname == IP_DONTFRAG)
         {
            //Set IP_DONTFRAG option
            ret = socketSetIpDontFragOption(sock, optval, optlen);
         }
         else if(optname == IP_PKTINFO)
         {
            //Set IP_PKTINFO option
            ret = socketSetIpPktInfoOption(sock, optval, optlen);
         }
         else if(optname == IP_RECVTOS)
         {
            //Set IP_RECVTOS option
            ret = socketSetIpRecvTosOption(sock, optval, optlen);
         }
         else if(optname == IP_RECVTTL)
         {
            //Set IP_RECVTTL option
            ret = socketSetIpRecvTtlOption(sock, optval, optlen);
         }
         else
         {
            //Unknown option
            socketSetErrnoCode(sock, ENOPROTOOPT);
            ret = SOCKET_ERROR;
         }
      }
      else if(level == IPPROTO_IPV6)
      {
         //Check option type
         if(optname == IPV6_TCLASS)
         {
            //Set IPV6_TCLASS option
            ret = socketSetIpv6TrafficClassOption(sock, optval, optlen);
         }
         else if(optname == IPV6_UNICAST_HOPS)
         {
            //Set IPV6_UNICAST_HOPS option
            ret = socketSetIpv6UnicastHopsOption(sock, optval, optlen);
         }
         else if(optname == IPV6_MULTICAST_IF)
         {
            //Set IPV6_MULTICAST_IF option
            ret = socketSetIpv6MulticastIfOption(sock, optval, optlen);
         }
         else if(optname == IPV6_MULTICAST_HOPS)
         {
            //Set IPV6_MULTICAST_HOPS option
            ret = socketSetIpv6MulticastHopsOption(sock, optval, optlen);
         }
         else if(optname == IPV6_MULTICAST_LOOP)
         {
            //Set IPV6_MULTICAST_LOOP option
            ret = socketSetIpv6MulticastLoopOption(sock, optval, optlen);
         }
         else if(optname == IPV6_ADD_MEMBERSHIP)
         {
            //Set IPV6_ADD_MEMBERSHIP option
            ret = socketSetIpv6AddMembershipOption(sock, optval, optlen);
         }
         else if(optname == IPV6_DROP_MEMBERSHIP)
         {
            //Set IPV6_DROP_MEMBERSHIP option
            ret = socketSetIpv6DropMembershipOption(sock, optval, optlen);
         }
         else if(optname == MCAST_JOIN_GROUP)
         {
            //Set MCAST_JOIN_GROUP option
            ret = socketSetMcastJoinGroupOption(sock, optval, optlen);
         }
         else if(optname == MCAST_LEAVE_GROUP)
         {
            //Set MCAST_LEAVE_GROUP option
            ret = socketSetMcastLeaveGroupOption(sock, optval, optlen);
         }
         else if(optname == MCAST_BLOCK_SOURCE)
         {
            //Set MCAST_BLOCK_SOURCE option
            ret = socketSetMcastBlockSourceOption(sock, optval, optlen);
         }
         else if(optname == MCAST_UNBLOCK_SOURCE)
         {
            //Set MCAST_UNBLOCK_SOURCE option
            ret = socketSetMcastUnblockSourceOption(sock, optval, optlen);
         }
         else if(optname == MCAST_JOIN_SOURCE_GROUP)
         {
            //Set MCAST_JOIN_SOURCE_GROUP option
            ret = socketSetMcastJoinSourceGroupOption(sock, optval, optlen);
         }
         else if(optname == MCAST_LEAVE_SOURCE_GROUP)
         {
            //Set MCAST_LEAVE_SOURCE_GROUP option
            ret = socketSetMcastLeaveSourceGroupOption(sock, optval, optlen);
         }
         else if(optname == IPV6_V6ONLY)
         {
            //Set IPV6_V6ONLY option
            ret = socketSetIpv6OnlyOption(sock, optval, optlen);
         }
         else if(optname == IPV6_DONTFRAG)
         {
            //Set IPV6_DONTFRAG option
            ret = socketSetIpv6DontFragOption(sock, optval, optlen);
         }
         else if(optname == IPV6_PKTINFO)
         {
            //Set IPV6_PKTINFO option
            ret = socketSetIpv6PktInfoOption(sock, optval, optlen);
         }
         else if(optname == IPV6_RECVTCLASS)
         {
            //Set IPV6_RECVTCLASS option
            ret = socketSetIpv6RecvTrafficClassOption(sock, optval, optlen);
         }
         else if(optname == IPV6_RECVHOPLIMIT)
         {
            //Set IPV6_RECVHOPLIMIT option
            ret = socketSetIpv6RecvHopLimitOption(sock, optval, optlen);
         }
         else
         {
            //Unknown option
            socketSetErrnoCode(sock, ENOPROTOOPT);
            ret = SOCKET_ERROR;
         }
      }
      else if(level == IPPROTO_TCP)
      {
         //Check option type
         if(optname == TCP_NODELAY)
         {
            //Set TCP_NODELAY option
            ret = socketSetTcpNoDelayOption(sock, optval, optlen);
         }
         else if(optname == TCP_MAXSEG)
         {
            //Set TCP_MAXSEG option
            ret = socketSetTcpMaxSegOption(sock, optval, optlen);
         }
         else if(optname == TCP_KEEPIDLE)
         {
            //Set TCP_KEEPIDLE option
            ret = socketSetTcpKeepIdleOption(sock, optval, optlen);
         }
         else if(optname == TCP_KEEPINTVL)
         {
            //Set TCP_KEEPINTVL option
            ret = socketSetTcpKeepIntvlOption(sock, optval, optlen);
         }
         else if(optname == TCP_KEEPCNT)
         {
            //Set TCP_KEEPCNT option
            ret = socketSetTcpKeepCntOption(sock, optval, optlen);
         }
         else
         {
            //Unknown option
            socketSetErrnoCode(sock, ENOPROTOOPT);
            ret = SOCKET_ERROR;
         }
      }
      else
      {
         //The specified level is not valid
         socketSetErrnoCode(sock, EINVAL);
         ret = SOCKET_ERROR;
      }
   }
   else
   {
      //The option is not valid
      socketSetErrnoCode(sock, EFAULT);
      ret = SOCKET_ERROR;
   }
 
   //return status code
   return ret;
}
 
 
/**
 * @brief The getsockopt function retrieves a socket option
 * @param[in] s Descriptor that identifies a socket
 * @param[in] level The level at which the option is defined
 * @param[in] optname The socket option for which the value is to be retrieved
 * @param[out] optval A pointer to the buffer in which the value for the
 *   requested option is to be returned
 * @param[in,out] optlen The size, in bytes, of the buffer pointed to by the
 *   optval parameter
 * @return If no error occurs, getsockopt returns SOCKET_SUCCESS
 *   Otherwise, it returns SOCKET_ERROR
 **/
 
int_t getsockopt(int_t s, int_t level, int_t optname, void *optval,
   socklen_t *optlen)
{
   int_t ret;
   Socket *sock;
 
   //Make sure the socket descriptor is valid
   if(s < 0 || s >= SOCKET_MAX_COUNT)
   {
      return SOCKET_ERROR;
   }
 
   //Point to the socket structure
   sock = &socketTable[s];
 
   //Get exclusive access
   osAcquireMutex(&netMutex);
 
   //Make sure the option is valid
   if(optval != NULL)
   {
      //Check option level
      if(level == SOL_SOCKET)
      {
         //Check option type
         if(optname == SO_REUSEADDR)
         {
            //Get SO_REUSEADDR option
            ret = socketGetSoReuseAddrOption(sock, optval, optlen);
         }
         else if(optname == SO_TYPE)
         {
            //Get SO_TYPE option
            ret = socketGetSoTypeOption(sock, optval, optlen);
         }
         else if(optname == SO_ERROR)
         {
            //Get SO_ERROR option
            ret = socketGetSoErrorOption(sock, optval, optlen);
         }
         else if(optname == SO_BROADCAST)
         {
            //Get SO_BROADCAST option
            ret = socketGetSoBroadcastOption(sock, optval, optlen);
         }
         else if(optname == SO_SNDTIMEO)
         {
            //Get SO_SNDTIMEO option
            ret = socketGetSoSndTimeoOption(sock, optval, optlen);
         }
         else if(optname == SO_RCVTIMEO)
         {
            //Get SO_RCVTIMEO option
            ret = socketGetSoRcvTimeoOption(sock, optval, optlen);
         }
         else if(optname == SO_SNDBUF)
         {
            //Get SO_SNDBUF option
            ret = socketGetSoSndBufOption(sock, optval, optlen);
         }
         else if(optname == SO_RCVBUF)
         {
            //Get SO_RCVBUF option
            ret = socketGetSoRcvBufOption(sock, optval, optlen);
         }
         else if(optname == SO_KEEPALIVE)
         {
            //Get SO_KEEPALIVE option
            ret = socketGetSoKeepAliveOption(sock, optval, optlen);
         }
         else if(optname == SO_NO_CHECK)
         {
            //Get SO_NO_CHECK option
            ret = socketGetSoNoCheckOption(sock, optval, optlen);
         }
         else
         {
            //Unknown option
            socketSetErrnoCode(sock, ENOPROTOOPT);
            ret = SOCKET_ERROR;
         }
      }
      else if(level == IPPROTO_IP)
      {
         //Check option type
         if(optname == IP_TOS)
         {
            //Get IP_TOS option
            ret = socketGetIpTosOption(sock, optval, optlen);
         }
         else if(optname == IP_TTL)
         {
            //Get IP_TTL option
            ret = socketGetIpTtlOption(sock, optval, optlen);
         }
         else if(optname == IP_MULTICAST_TTL)
         {
            //Get IP_MULTICAST_TTL option
            ret = socketGetIpMulticastTtlOption(sock, optval, optlen);
         }
         else if(optname == IP_MULTICAST_LOOP)
         {
            //Get IP_MULTICAST_LOOP option
            ret = socketGetIpMulticastLoopOption(sock, optval, optlen);
         }
         else if(optname == IP_ADD_MEMBERSHIP ||
            optname == IP_DROP_MEMBERSHIP ||
            optname == IP_BLOCK_SOURCE ||
            optname == IP_UNBLOCK_SOURCE ||
            optname == IP_ADD_SOURCE_MEMBERSHIP ||
            optname == IP_DROP_SOURCE_MEMBERSHIP ||
            optname == MCAST_JOIN_GROUP ||
            optname == MCAST_LEAVE_GROUP ||
            optname == MCAST_BLOCK_SOURCE ||
            optname == MCAST_UNBLOCK_SOURCE ||
            optname == MCAST_JOIN_SOURCE_GROUP ||
            optname == MCAST_LEAVE_SOURCE_GROUP)
         {
            //When any of these options are used with getsockopt, the error
            //generated is EOPNOTSUPP (refer to RFC 3678, section 4.1.3)
            socketSetErrnoCode(sock, EOPNOTSUPP);
            ret = SOCKET_ERROR;
         }
         else if(optname == IP_DONTFRAG)
         {
            //Get IP_DONTFRAG option
            ret = socketGetIpDontFragOption(sock, optval, optlen);
         }
         else if(optname == IP_PKTINFO)
         {
            //Get IP_PKTINFO option
            ret = socketGetIpPktInfoOption(sock, optval, optlen);
         }
         else if(optname == IP_RECVTOS)
         {
            //Get IP_RECVTOS option
            ret = socketGetIpRecvTosOption(sock, optval, optlen);
         }
         else if(optname == IP_RECVTTL)
         {
            //Get IP_RECVTTL option
            ret = socketGetIpRecvTtlOption(sock, optval, optlen);
         }
         else
         {
            //Unknown option
            socketSetErrnoCode(sock, ENOPROTOOPT);
            ret = SOCKET_ERROR;
         }
      }
      else if(level == IPPROTO_IPV6)
      {
         //Check option type
         if(optname == IPV6_TCLASS)
         {
            //Get IPV6_TCLASS option
            ret = socketGetIpv6TrafficClassOption(sock, optval, optlen);
         }
         else if(optname == IPV6_UNICAST_HOPS)
         {
            //Get IPV6_UNICAST_HOPS option
            ret = socketGetIpv6UnicastHopsOption(sock, optval, optlen);
         }
         else if(optname == IPV6_MULTICAST_HOPS)
         {
            //Get IPV6_MULTICAST_HOPS option
            ret = socketGetIpv6MulticastHopsOption(sock, optval, optlen);
         }
         else if(optname == IPV6_MULTICAST_LOOP)
         {
            //Get IPV6_MULTICAST_LOOP option
            ret = socketGetIpv6MulticastLoopOption(sock, optval, optlen);
         }
         else if(optname == IPV6_ADD_MEMBERSHIP ||
            optname == IPV6_DROP_MEMBERSHIP ||
            optname == MCAST_JOIN_GROUP ||
            optname == MCAST_LEAVE_GROUP ||
            optname == MCAST_BLOCK_SOURCE ||
            optname == MCAST_UNBLOCK_SOURCE ||
            optname == MCAST_JOIN_SOURCE_GROUP ||
            optname == MCAST_LEAVE_SOURCE_GROUP)
         {
            //When any of these options are used with getsockopt, the error
            //generated is EOPNOTSUPP
            socketSetErrnoCode(sock, EOPNOTSUPP);
            ret = SOCKET_ERROR;
         }
         else if(optname == IPV6_V6ONLY)
         {
            //Get IPV6_V6ONLY option
            ret = socketGetIpv6OnlyOption(sock, optval, optlen);
         }
         else if(optname == IPV6_DONTFRAG)
         {
            //Get IPV6_DONTFRAG option
            ret = socketGetIpv6DontFragOption(sock, optval, optlen);
         }
         else if(optname == IPV6_PKTINFO)
         {
            //Get IPV6_PKTINFO option
            ret = socketGetIpv6PktInfoOption(sock, optval, optlen);
         }
         else if(optname == IPV6_RECVTCLASS)
         {
            //Get IPV6_RECVTCLASS option
            ret = socketGetIpv6RecvTrafficClassOption(sock, optval, optlen);
         }
         else if(optname == IPV6_RECVHOPLIMIT)
         {
            //Get IPV6_RECVHOPLIMIT option
            ret = socketGetIpv6RecvHopLimitOption(sock, optval, optlen);
         }
         else
         {
            //Unknown option
            socketSetErrnoCode(sock, ENOPROTOOPT);
            ret = SOCKET_ERROR;
         }
      }
      else if(level == IPPROTO_TCP)
      {
         //Check option type
         if(optname == TCP_NODELAY)
         {
            //Get TCP_NODELAY option
            ret = socketGetTcpNoDelayOption(sock, optval, optlen);
         }
         else if(optname == TCP_MAXSEG)
         {
            //Get TCP_MAXSEG option
            ret = socketGetTcpMaxSegOption(sock, optval, optlen);
         }
         else if(optname == TCP_KEEPIDLE)
         {
            //Get TCP_KEEPIDLE option
            ret = socketGetTcpKeepIdleOption(sock, optval, optlen);
         }
         else if(optname == TCP_KEEPINTVL)
         {
            //Get TCP_KEEPINTVL option
            ret = socketGetTcpKeepIntvlOption(sock, optval, optlen);
         }
         else if(optname == TCP_KEEPCNT)
         {
            //Get TCP_KEEPCNT option
            ret = socketGetTcpKeepCntOption(sock, optval, optlen);
         }
         else
         {
            //Unknown option
            socketSetErrnoCode(sock, ENOPROTOOPT);
            ret = SOCKET_ERROR;
         }
      }
      else
      {
         //The specified level is not valid
         socketSetErrnoCode(sock, EINVAL);
         ret = SOCKET_ERROR;
      }
   }
   else
   {
      //The option is not valid
      socketSetErrnoCode(sock, EFAULT);
      ret = SOCKET_ERROR;
   }
 
   //Release exclusive access
   osReleaseMutex(&netMutex);
 
   //return status code
   return ret;
}
 
 
/**
 * @brief Set multicast source filter (IPv4 only)
 * @param[in] s Descriptor that identifies a socket
 * @param[in] interface Local IP address of the interface
 * @param[in] group IP multicast address of the group
 * @param[in] fmode Multicast filter mode (MCAST_INCLUDE or MCAST_EXCLUDE)
 * @param[in] numsrc Number of source addresses in the slist array
 * @param[in] slist Array of IP addresses of sources to include or exclude
 *   depending on the filter mode
 * @return If no error occurs, setipv4sourcefilter returns SOCKET_SUCCESS
 *   Otherwise, it returns SOCKET_ERROR
 **/
 
int_t setipv4sourcefilter(int_t s, struct in_addr interface, struct in_addr group,
   uint32_t fmode, uint_t numsrc, struct in_addr *slist)
{
#if (SOCKET_MAX_MULTICAST_SOURCES > 0)
   error_t error;
   uint_t i;
   Socket *sock;
   IpFilterMode filterMode;
   IpAddr groupAddr;
   IpAddr sources[SOCKET_MAX_MULTICAST_SOURCES];
 
   //Make sure the socket descriptor is valid
   if(s < 0 || s >= SOCKET_MAX_COUNT)
   {
      return SOCKET_ERROR;
   }
 
   //Point to the socket structure
   sock = &socketTable[s];
 
   //Copy group address
   groupAddr.length = sizeof(Ipv4Addr);
   groupAddr.ipv4Addr = group.s_addr;
 
   //Check filter mode
   if(fmode == MCAST_INCLUDE)
   {
      //Select source-specific filter mode
      filterMode = IP_FILTER_MODE_INCLUDE;
   }
   else if(fmode == MCAST_INCLUDE)
   {
      //Select any-source filter mode
      filterMode = IP_FILTER_MODE_EXCLUDE;
   }
   else
   {
      //Report an error
      socketSetErrnoCode(sock, EINVAL);
      return SOCKET_ERROR;
   }
 
   //If the implementation imposes a limit on the maximum number of sources
   //in a source filter, ENOBUFS is generated when the operation would exceed
   //the maximum (refer to RFC3678, section 4.2.1)
   if(numsrc > SOCKET_MAX_MULTICAST_SOURCES)
   {
      socketSetErrnoCode(sock, ENOBUFS);
      return SOCKET_ERROR;
   }
 
   //If numsrc is 0, a NULL pointer may be supplied
   if(numsrc > 0 && slist == NULL)
   {
      socketSetErrnoCode(sock, EINVAL);
      return SOCKET_ERROR;
   }
 
   //Copy the list of source addresses to include or exclude
   for(i = 0; i < numsrc; i++)
   {
      sources[i].length = sizeof(Ipv4Addr);
      sources[i].ipv4Addr = slist[i].s_addr;
   }
 
   //Set multicast source filter
   error = socketSetMulticastSourceFilter(sock, &groupAddr, filterMode,
      sources, numsrc);
 
   //Any error to report?
   if(error)
   {
      socketTranslateErrorCode(sock, error);
      return SOCKET_ERROR;
   }
 
   //Successful processing
   return SOCKET_SUCCESS;
#else
   //Not implemented
   return SOCKET_ERROR;
#endif
}
 
 
/**
 * @brief Get multicast source filter (IPv4 only)
 * @param[in] s Descriptor that identifies a socket
 * @param[in] interface Local IP address of the interface
 * @param[in] group IP multicast address of the group
 * @param[out] fmode Multicast filter mode (MCAST_INCLUDE or MCAST_EXCLUDE)
 * @param[out] numsrc On input, the numsrc argument holds the number of source
 *   addresses that will fit in the slist array. On output, the numsrc argument
 *   will hold the total number of sources in the filter
 * @param[out] slist Buffer into which an array of IP addresses of included or
 *   excluded (depending on the filter mode) sources will be written. If numsrc
 *   was 0 on input, a NULL pointer may be supplied
 * @return If no error occurs, setipv4sourcefilter returns SOCKET_SUCCESS
 *   Otherwise, it returns SOCKET_ERROR
 **/
 
int_t getipv4sourcefilter(int_t s, struct in_addr interface, struct in_addr group,
   uint32_t *fmode, uint_t *numsrc, struct in_addr *slist)
{
#if (SOCKET_MAX_MULTICAST_SOURCES > 0)
   error_t error;
   uint_t i;
   uint_t numSources;
   Socket *sock;
   IpFilterMode filterMode;
   IpAddr groupAddr;
   IpAddr sources[SOCKET_MAX_MULTICAST_SOURCES];
 
   //Make sure the socket descriptor is valid
   if(s < 0 || s >= SOCKET_MAX_COUNT)
   {
      return SOCKET_ERROR;
   }
 
   //Point to the socket structure
   sock = &socketTable[s];
 
   //Check parameters
   if(fmode == NULL || numsrc == NULL)
   {
      socketSetErrnoCode(sock, EINVAL);
      return SOCKET_ERROR;
   }
 
   //If numsrc was 0 on input, a NULL pointer may be supplied
   if(*numsrc > 0 && slist == NULL)
   {
      socketSetErrnoCode(sock, EINVAL);
      return SOCKET_ERROR;
   }
 
   //Copy group address
   groupAddr.length = sizeof(Ipv4Addr);
   groupAddr.ipv4Addr = group.s_addr;
 
   //Set multicast source filter
   error = socketGetMulticastSourceFilter(sock, &groupAddr, &filterMode,
      sources, &numSources);
 
   //Any error to report?
   if(error)
   {
      socketTranslateErrorCode(sock, error);
      return SOCKET_ERROR;
   }
 
   //Filter mode
   if(filterMode == IP_FILTER_MODE_INCLUDE)
   {
      *fmode = MCAST_INCLUDE;
   }
   else
   {
      *fmode = MCAST_EXCLUDE;
   }
 
   //The slist parameter will hold as many source addresses as fit, up to the
   //minimum of the array size passed in as the original numsrc value and the
   //total number of sources in the filter (refer to RFC3678, section 4.2.2)
   for(i = 0; i < *numsrc && i < numSources; i++)
   {
      slist[i].s_addr = sources[i].ipv4Addr;
   }
 
   //On return, numsrc is always updated to be the total number of sources in
   //the filter
   *numsrc = numSources;
 
   //Successful processing
   return SOCKET_SUCCESS;
#else
   //Not implemented
   return SOCKET_ERROR;
#endif
}
 
 
/**
 * @brief Set multicast source filter
 * @param[in] s Descriptor that identifies a socket
 * @param[in] interface Local IP address of the interface
 * @param[in] group IP multicast address of the group
 * @param[in] grouplen Length in bytes of the group
 * @param[in] fmode Multicast filter mode (MCAST_INCLUDE or MCAST_EXCLUDE)
 * @param[in] numsrc Number of source addresses in the slist array
 * @param[in] slist Array of IP addresses of sources to include or exclude
 *   depending on the filter mode
 * @return If no error occurs, setipv4sourcefilter returns SOCKET_SUCCESS
 *   Otherwise, it returns SOCKET_ERROR
 **/
 
int_t setsourcefilter(int_t s, uint32_t interface, struct sockaddr *group,
   socklen_t grouplen, uint32_t fmode, uint_t numsrc,
   struct sockaddr_storage *slist)
{
#if (SOCKET_MAX_MULTICAST_SOURCES > 0)
   error_t error;
   uint_t i;
   Socket *sock;
   IpFilterMode filterMode;
   IpAddr groupAddr;
   IpAddr sources[SOCKET_MAX_MULTICAST_SOURCES];
 
   //Make sure the socket descriptor is valid
   if(s < 0 || s >= SOCKET_MAX_COUNT)
   {
      return SOCKET_ERROR;
   }
 
   //Point to the socket structure
   sock = &socketTable[s];
 
#if (IPV4_SUPPORT == ENABLED)
   //IPv4 group address?
   if(group->sa_family == AF_INET &&
      grouplen >= (socklen_t) sizeof(SOCKADDR_IN))
   {
      //Point to the IPv4 address information
      SOCKADDR_IN *sa = (SOCKADDR_IN *) group;
 
      //Copy IPv4 address
      groupAddr.length = sizeof(Ipv4Addr);
      groupAddr.ipv4Addr = sa->sin_addr.s_addr;
   }
   else
#endif
#if (IPV6_SUPPORT == ENABLED)
   //IPv6 group address?
   if(group->sa_family == AF_INET6 &&
      grouplen >= (socklen_t) sizeof(SOCKADDR_IN6))
   {
      //Point to the IPv6 address information
      SOCKADDR_IN6 *sa = (SOCKADDR_IN6 *) group;
 
      //Copy IPv6 address
      groupAddr.length = sizeof(Ipv6Addr);
      ipv6CopyAddr(&groupAddr.ipv6Addr, sa->sin6_addr.s6_addr);
   }
   else
#endif
   //Invalid group address?
   {
      //Report an error
      socketSetErrnoCode(sock, EINVAL);
      return SOCKET_ERROR;
   }
 
   //Check filter mode
   if(fmode == MCAST_INCLUDE)
   {
      //Select source-specific filter mode
      filterMode = IP_FILTER_MODE_INCLUDE;
   }
   else if(fmode == MCAST_INCLUDE)
   {
      //Select any-source filter mode
      filterMode = IP_FILTER_MODE_EXCLUDE;
   }
   else
   {
      //Report an error
      socketSetErrnoCode(sock, EINVAL);
      return SOCKET_ERROR;
   }
 
   //If the implementation imposes a limit on the maximum number of sources
   //in a source filter, ENOBUFS is generated when the operation would exceed
   //the maximum (refer to RFC3678, section 5.2.1)
   if(numsrc > SOCKET_MAX_MULTICAST_SOURCES)
   {
      socketSetErrnoCode(sock, ENOBUFS);
      return SOCKET_ERROR;
   }
 
   //If numsrc is 0, a NULL pointer may be supplied
   if(numsrc > 0 && slist == NULL)
   {
      socketSetErrnoCode(sock, EINVAL);
      return SOCKET_ERROR;
   }
 
   //Copy the list of source addresses to include or exclude
   for(i = 0; i < numsrc; i++)
   {
#if (IPV4_SUPPORT == ENABLED)
      //IPv4 source address?
      if(slist[i].ss_family == AF_INET &&
         slist[i].ss_family == group->sa_family)
      {
         //Point to the IPv4 address information
         SOCKADDR_IN *sa = (SOCKADDR_IN *) &slist[i];
 
         //Copy IPv4 address
         sources[i].length = sizeof(Ipv4Addr);
         sources[i].ipv4Addr = sa->sin_addr.s_addr;
      }
      else
#endif
#if (IPV6_SUPPORT == ENABLED)
      //IPv6 source address?
      if(slist[i].ss_family == AF_INET6 &&
         slist[i].ss_family == group->sa_family)
      {
         //Point to the IPv6 address information
         SOCKADDR_IN6 *sa = (SOCKADDR_IN6 *) &slist[i];
 
         //Copy IPv6 address
         sources[i].length = sizeof(Ipv6Addr);
         ipv6CopyAddr(&sources[i].ipv6Addr, sa->sin6_addr.s6_addr);
      }
      else
#endif
      //Invalid source address?
      {
         //Report an error
         socketSetErrnoCode(sock, EINVAL);
         return SOCKET_ERROR;
      }
   }
 
   //Set multicast source filter
   error = socketSetMulticastSourceFilter(sock, &groupAddr, filterMode,
      sources, numsrc);
 
   //Any error to report?
   if(error)
   {
      socketTranslateErrorCode(sock, error);
      return SOCKET_ERROR;
   }
 
   //Successful processing
   return SOCKET_SUCCESS;
#else
   //Not implemented
   return SOCKET_ERROR;
#endif
}
 
 
/**
 * @brief Get multicast source filter
 * @param[in] s Descriptor that identifies a socket
 * @param[in] interface Local IP address of the interface
 * @param[in] group IP multicast address of the group
 * @param[in] grouplen Length in bytes of the group
 * @param[out] fmode Multicast filter mode (MCAST_INCLUDE or MCAST_EXCLUDE)
 * @param[out] numsrc On input, the numsrc argument holds the number of source
 *   addresses that will fit in the slist array. On output, the numsrc argument
 *   will hold the total number of sources in the filter
 * @param[out] slist Buffer into which an array of IP addresses of included or
 *   excluded (depending on the filter mode) sources will be written. If numsrc
 *   was 0 on input, a NULL pointer may be supplied
 * @return If no error occurs, setipv4sourcefilter returns SOCKET_SUCCESS
 *   Otherwise, it returns SOCKET_ERROR
 **/
 
int_t getsourcefilter(int_t s, uint32_t interface, struct sockaddr *group,
   socklen_t grouplen, uint32_t *fmode, uint_t *numsrc,
   struct sockaddr_storage *slist)
{
#if (SOCKET_MAX_MULTICAST_SOURCES > 0)
   error_t error;
   uint_t i;
   uint_t numSources;
   Socket *sock;
   IpFilterMode filterMode;
   IpAddr groupAddr;
   IpAddr sources[SOCKET_MAX_MULTICAST_SOURCES];
 
   //Make sure the socket descriptor is valid
   if(s < 0 || s >= SOCKET_MAX_COUNT)
   {
      return SOCKET_ERROR;
   }
 
   //Point to the socket structure
   sock = &socketTable[s];
 
   //Check parameters
   if(fmode == NULL || numsrc == NULL)
   {
      socketSetErrnoCode(sock, EINVAL);
      return SOCKET_ERROR;
   }
 
   //If numsrc was 0 on input, a NULL pointer may be supplied
   if(*numsrc > 0 && slist == NULL)
   {
      socketSetErrnoCode(sock, EINVAL);
      return SOCKET_ERROR;
   }
 
#if (IPV4_SUPPORT == ENABLED)
   //IPv4 group address?
   if(group->sa_family == AF_INET &&
      grouplen >= (socklen_t) sizeof(SOCKADDR_IN))
   {
      //Point to the IPv4 address information
      SOCKADDR_IN *sa = (SOCKADDR_IN *) group;
 
      //Copy IPv4 address
      groupAddr.length = sizeof(Ipv4Addr);
      groupAddr.ipv4Addr = sa->sin_addr.s_addr;
   }
   else
#endif
#if (IPV6_SUPPORT == ENABLED)
   //IPv6 group address?
   if(group->sa_family == AF_INET6 &&
      grouplen >= (socklen_t) sizeof(SOCKADDR_IN6))
   {
      //Point to the IPv6 address information
      SOCKADDR_IN6 *sa = (SOCKADDR_IN6 *) group;
 
      //Copy IPv6 address
      groupAddr.length = sizeof(Ipv6Addr);
      ipv6CopyAddr(&groupAddr.ipv6Addr, sa->sin6_addr.s6_addr);
   }
   else
#endif
   //Invalid group address?
   {
      //Report an error
      socketSetErrnoCode(sock, EINVAL);
      return SOCKET_ERROR;
   }
 
   //Set multicast source filter
   error = socketGetMulticastSourceFilter(sock, &groupAddr, &filterMode,
      sources, &numSources);
 
   //Any error to report?
   if(error)
   {
      socketTranslateErrorCode(sock, error);
      return SOCKET_ERROR;
   }
 
   //Filter mode
   if(filterMode == IP_FILTER_MODE_INCLUDE)
   {
      *fmode = MCAST_INCLUDE;
   }
   else
   {
      *fmode = MCAST_EXCLUDE;
   }
 
   //The slist parameter will hold as many source addresses as fit, up to the
   //minimum of the array size passed in as the original numsrc value and the
   //total number of sources in the filter (refer to RFC3678, section 5.2.2)
   for(i = 0; i < *numsrc && i < numSources; i++)
   {
#if (IPV4_SUPPORT == ENABLED)
      //IPv4 source address?
      if(sock->localIpAddr.length == sizeof(Ipv4Addr))
      {
         //Point to the IPv4 address information
         SOCKADDR_IN *sa = (SOCKADDR_IN *) &slist[i];
 
         //Set address family and port number
         sa->sin_family = AF_INET;
         sa->sin_port = 0;
 
         //Copy IPv4 address
         sa->sin_addr.s_addr = sources[i].ipv4Addr;
      }
      else
#endif
#if (IPV6_SUPPORT == ENABLED)
      //IPv6 source address?
      if(sock->localIpAddr.length == sizeof(Ipv6Addr))
      {
         //Point to the IPv4 address information
         SOCKADDR_IN6 *sa = (SOCKADDR_IN6 *) &slist[i];
 
         //Set address family and port number
         sa->sin6_family = AF_INET6;
         sa->sin6_port = 0;
         sa->sin6_flowinfo = 0;
         sa->sin6_scope_id = 0;
 
         //Copy IPv6 address
         ipv6CopyAddr(sa->sin6_addr.s6_addr, &sources[i].ipv6Addr);
      }
      else
#endif
      //Invalid source address?
      {
         //Just for sanity
         osMemset(&slist[i], 0, sizeof(SOCKADDR_STORAGE));
      }
   }
 
   //On return, numsrc is always updated to be the total number of sources in
   //the filter
   *numsrc = numSources;
 
   //Successful processing
   return SOCKET_SUCCESS;
#else
   //Not implemented
   return SOCKET_ERROR;
#endif
}
 
 
/**
 * @brief Control the I/O mode of a socket
 * @param[in] s Descriptor that identifies a socket
 * @param[in] cmd A command to perform on the socket
 * @param[in,out] arg A pointer to a parameter
 * @return If no error occurs, setsockopt returns SOCKET_SUCCESS
 *   Otherwise, it returns SOCKET_ERROR
 **/
 
int_t ioctlsocket(int_t s, uint32_t cmd, void *arg)
{
   int_t ret;
   uint_t *val;
   Socket *sock;
 
   //Make sure the socket descriptor is valid
   if(s < 0 || s >= SOCKET_MAX_COUNT)
   {
      return SOCKET_ERROR;
   }
 
   //Point to the socket structure
   sock = &socketTable[s];
 
   //Get exclusive access
   osAcquireMutex(&netMutex);
 
   //Make sure the parameter is valid
   if(arg != NULL)
   {
      //Check command type
      switch(cmd)
      {
      //Enable or disable non-blocking mode
      case FIONBIO:
         //Cast the parameter to the relevant type
         val = (uint_t *) arg;
         //Enable blocking or non-blocking operation
         sock->timeout = (*val != 0) ? 0 : INFINITE_DELAY;
         //Successful processing
         ret = SOCKET_SUCCESS;
         break;
 
#if (TCP_SUPPORT == ENABLED)
      //Get the number of bytes that are immediately available for reading
      case FIONREAD:
         //Cast the parameter to the relevant type
         val = (uint_t *) arg;
         //Return the actual value
         *val = sock->rcvUser;
         //Successful processing
         ret = SOCKET_SUCCESS;
         break;
 
      //Get the number of bytes written to the send queue but not yet
      //acknowledged by the other side of the connection
      case FIONWRITE:
         //Cast the parameter to the relevant type
         val = (uint_t *) arg;
         //Return the actual value
         *val = sock->sndUser + sock->sndNxt - sock->sndUna;
         //Successful processing
         ret = SOCKET_SUCCESS;
         break;
 
      //Get the free space  in the send queue
      case FIONSPACE:
         //Cast the parameter to the relevant type
         val = (uint_t *) arg;
         //Return the actual value
         *val = sock->txBufferSize - (sock->sndUser + sock->sndNxt -
            sock->sndUna);
         //Successful processing
         ret = SOCKET_SUCCESS;
         break;
#endif
 
      //Unknown command?
      default:
         //Report an error
         socketSetErrnoCode(sock, EINVAL);
         ret = SOCKET_ERROR;
         break;
      }
   }
   else
   {
      //The parameter is not valid
      socketSetErrnoCode(sock, EFAULT);
      ret = SOCKET_ERROR;
   }
 
   //Release exclusive access
   osReleaseMutex(&netMutex);
 
   //return status code
   return ret;
}
 
 
/**
 * @brief Perform specific operation
 * @param[in] s Descriptor that identifies a socket
 * @param[in] cmd A command to perform on the socket
 * @param[in] arg Argument value
 * @return If no error occurs, setsockopt returns SOCKET_SUCCESS
 *   Otherwise, it returns SOCKET_ERROR
 **/
 
int_t fcntl(int_t s, int_t cmd, int_t arg)
{
   int_t ret;
   Socket *sock;
 
   //Make sure the socket descriptor is valid
   if(s < 0 || s >= SOCKET_MAX_COUNT)
   {
      return SOCKET_ERROR;
   }
 
   //Point to the socket structure
   sock = &socketTable[s];
 
   //Get exclusive access
   osAcquireMutex(&netMutex);
 
   //Check command type
   switch(cmd)
   {
   //Get the file status flags?
   case F_GETFL:
      //Return (as the function result) the file descriptor flags
      ret = (sock->timeout == 0) ? O_NONBLOCK : 0;
      break;
 
   //Set the file status flags?
   case F_SETFL:
      //Enable blocking or non-blocking operation
      sock->timeout = ((arg & O_NONBLOCK) != 0) ? 0 : INFINITE_DELAY;
      //Successful processing
      ret = SOCKET_SUCCESS;
      break;
 
   //Unknown command?
   default:
      //Report an error
      socketSetErrnoCode(sock, EINVAL);
      ret = SOCKET_ERROR;
      break;
   }
 
   //Release exclusive access
   osReleaseMutex(&netMutex);
 
   //return status code
   return ret;
}
 
 
/**
 * @brief The shutdown function disables sends or receives on a socket
 * @param[in] s Descriptor that identifies a socket
 * @param[in] how A flag that describes what types of operation will no longer be allowed
 * @return If no error occurs, shutdown returns SOCKET_SUCCESS
 *   Otherwise, it returns SOCKET_ERROR
 **/
 
int_t shutdown(int_t s, int_t how)
{
   error_t error;
   Socket *sock;
 
   //Make sure the socket descriptor is valid
   if(s < 0 || s >= SOCKET_MAX_COUNT)
   {
      return SOCKET_ERROR;
   }
 
   //Point to the socket structure
   sock = &socketTable[s];
 
   //Shut down socket
   error = socketShutdown(sock, how);
 
   //Any error to report?
   if(error)
   {
      socketTranslateErrorCode(sock, error);
      return SOCKET_ERROR;
   }
 
   //Successful processing
   return SOCKET_SUCCESS;
}
 
 
/**
 * @brief The closesocket function closes an existing socket
 * @param[in] s Descriptor that identifies a socket
 * @return If no error occurs, closesocket returns SOCKET_SUCCESS
 *   Otherwise, it returns SOCKET_ERROR
 **/
 
int_t closesocket(int_t s)
{
   Socket *sock;
 
   //Make sure the socket descriptor is valid
   if(s < 0 || s >= SOCKET_MAX_COUNT)
   {
      return SOCKET_ERROR;
   }
 
   //Point to the socket structure
   sock = &socketTable[s];
 
   //Close socket
   socketClose(sock);
 
   //Successful processing
   return SOCKET_SUCCESS;
}
 
 
/**
 * @brief Determine the status of one or more sockets
 *
 * The select function determines the status of one or more sockets,
 * waiting if necessary, to perform synchronous I/O
 *
 * @param[in] nfds Unused parameter included only for compatibility with
 *   Berkeley socket
 * @param[in,out] readfds An optional pointer to a set of sockets to be
 *   checked for readability
 * @param[in,out] writefds An optional pointer to a set of sockets to be
 *   checked for writability
 * @param[in,out] exceptfds An optional pointer to a set of sockets to be
 *   checked for errors
 * @param[in] timeout The maximum time for select to wait. Set the timeout
 *   parameter to null for blocking operations
 * @return The select function returns the total number of socket handles that
 *   are ready and contained in the fd_set structures, zero if the time limit
 *   expired, or SOCKET_ERROR if an error occurred
 **/
 
int_t select(int_t nfds, fd_set *readfds, fd_set *writefds, fd_set *exceptfds,
   const struct timeval *timeout)
{
   int_t i;
   int_t j;
   int_t n;
   int_t s;
   systime_t time;
   uint_t eventMask;
   uint_t eventFlags;
   OsEvent event;
   fd_set *fds;
 
   //Parse all the descriptor sets
   for(i = 0; i < 3; i++)
   {
      //Select the suitable descriptor set
      switch(i)
      {
      case 0:
         //Set of sockets to be checked for readability
         fds = readfds;
         break;
 
      case 1:
         //Set of sockets to be checked for writability
         fds = writefds;
         break;
 
      default:
         //Set of sockets to be checked for errors
         fds = exceptfds;
         break;
      }
 
      //Each descriptor is optional and may be omitted
      if(fds != NULL)
      {
         //Parse the current set of sockets
         for(j = 0; j < fds->fd_count; j++)
         {
            //Invalid socket descriptor?
            if(fds->fd_array[j] < 0 || fds->fd_array[j] >= SOCKET_MAX_COUNT)
            {
               //Report an error
               return SOCKET_ERROR;
            }
         }
      }
   }
 
   //Create an event object to get notified of socket events
   if(!osCreateEvent(&event))
   {
      //Failed to create event
      return SOCKET_ERROR;
   }
 
   //Parse all the descriptor sets
   for(i = 0; i < 3; i++)
   {
      //Select the suitable descriptor set
      switch(i)
      {
      case 0:
         //Set of sockets to be checked for readability
         fds = readfds;
         eventMask = SOCKET_EVENT_RX_READY;
         break;
 
      case 1:
         //Set of sockets to be checked for writability
         fds = writefds;
         eventMask = SOCKET_EVENT_TX_READY;
         break;
 
      default:
         //Set of sockets to be checked for errors
         fds = exceptfds;
         eventMask = SOCKET_EVENT_CLOSED;
         break;
      }
 
      //Each descriptor is optional and may be omitted
      if(fds != NULL)
      {
         //Parse the current set of sockets
         for(j = 0; j < fds->fd_count; j++)
         {
            //Get the descriptor associated with the current entry
            s = fds->fd_array[j];
            //Subscribe to the requested events
            socketRegisterEvents(&socketTable[s], &event, eventMask);
         }
      }
   }
 
   //Retrieve timeout value
   if(timeout != NULL)
   {
      time = timeout->tv_sec * 1000 + timeout->tv_usec / 1000;
   }
   else
   {
      time = INFINITE_DELAY;
   }
 
   //Block the current task until an event occurs
   osWaitForEvent(&event, time);
 
   //Count the number of events in the signaled state
   n = 0;
 
   //Parse all the descriptor sets
   for(i = 0; i < 3; i++)
   {
      //Select the suitable descriptor set
      switch(i)
      {
      case 0:
         //Set of sockets to be checked for readability
         fds = readfds;
         eventMask = SOCKET_EVENT_RX_READY;
         break;
 
      case 1:
         //Set of sockets to be checked for writability
         fds = writefds;
         eventMask = SOCKET_EVENT_TX_READY;
         break;
 
      default:
         //Set of sockets to be checked for errors
         fds = exceptfds;
         eventMask = SOCKET_EVENT_CLOSED;
         break;
      }
 
      //Each descriptor is optional and may be omitted
      if(fds != NULL)
      {
         //Parse the current set of sockets
         for(j = 0; j < fds->fd_count; )
         {
            //Get the descriptor associated with the current entry
            s = fds->fd_array[j];
            //Retrieve event flags for the current socket
            eventFlags = socketGetEvents(&socketTable[s]);
            //Unsubscribe previously registered events
            socketUnregisterEvents(&socketTable[s]);
 
            //Event flag is set?
            if(eventFlags & eventMask)
            {
               //Track the number of events in the signaled state
               n++;
               //Jump to the next socket descriptor
               j++;
            }
            else
            {
               //Remove descriptor from the current set
               socketFdClr(fds, s);
            }
         }
      }
   }
 
   //Delete event object
   osDeleteEvent(&event);
   //Return the number of events in the signaled state
   return n;
}
 
 
/**
 * @brief Get system host name
 * @param[out] name Output buffer where to store the system host name
 * @param[in] len Length of the buffer, in bytes
 * @return On success, zero is returned. On error, -1 is returned
 **/
 
int_t gethostname(char_t *name, size_t len)
{
   size_t n;
   NetInterface *interface;
 
   //Check parameter
   if(name == NULL)
   {
      //Report an error
      BSD_SOCKET_SET_ERRNO(EINVAL);
      return -1;
   }
 
   //Select the default network interface
   interface = netGetDefaultInterface();
 
   //Retrieve the length of the host name
   n = osStrlen(interface->hostname);
 
   //Make sure the buffer is large enough to hold the string
   if(len <= n)
   {
      //Report an error
      BSD_SOCKET_SET_ERRNO(ENAMETOOLONG);
      return -1;
   }
 
   //Copy the host name
   osStrcpy(name, interface->hostname);
 
   //Successful processing
   return 0;
}
 
 
/**
 * @brief Host name resolution
 * @param[in] name Name of the host to resolve
 * @return Pointer to the hostent structure or a NULL if an error occurs
 **/
 
struct hostent *gethostbyname(const char_t *name)
{
   int_t herrno;
   static HOSTENT result;
 
   //The hostent structure returned by the function resides in static
   //memory area
   return gethostbyname_r(name, &result, NULL, 0, &herrno);
}
 
 
/**
 * @brief Host name resolution (reentrant version)
 * @param[in] name Name of the host to resolve
 * @param[in] result Pointer to a hostent structure where the function can
 *   store the host entry
 * @param[out] buf Pointer to a temporary work buffer
 * @param[in] buflen Length of the temporary work buffer
 * @param[out] h_errnop Pointer to a location where the function can store an
 *   h_errno value if an error occurs
 * @return Pointer to the hostent structure or a NULL if an error occurs
 **/
 
struct hostent *gethostbyname_r(const char_t *name, struct hostent *result,
   char_t *buf, size_t buflen, int_t *h_errnop)
{
   error_t error;
   IpAddr ipAddr;
 
   //Check input parameters
   if(name == NULL || result == NULL)
   {
      //Report an error
      *h_errnop = NO_RECOVERY;
      return NULL;
   }
 
   //Resolve host address
   error = getHostByName(NULL, name, &ipAddr, 0);
   //Address resolution failed?
   if(error)
   {
      //Report an error
      *h_errnop = HOST_NOT_FOUND;
      return NULL;
   }
 
#if (IPV4_SUPPORT == ENABLED)
   //IPv4 address?
   if(ipAddr.length == sizeof(Ipv4Addr))
   {
      //Set address family
      result->h_addrtype = AF_INET;
      result->h_length = sizeof(Ipv4Addr);
 
      //Copy IPv4 address
      ipv4CopyAddr(result->h_addr, &ipAddr.ipv4Addr);
   }
   else
#endif
#if (IPV6_SUPPORT == ENABLED)
   //IPv6 address?
   if(ipAddr.length == sizeof(Ipv6Addr))
   {
      //Set address family
      result->h_addrtype = AF_INET6;
      result->h_length = sizeof(Ipv6Addr);
 
      //Copy IPv6 address
      ipv6CopyAddr(result->h_addr, &ipAddr.ipv6Addr);
   }
   else
#endif
   //Invalid address?
   {
      //Report an error
      *h_errnop = NO_ADDRESS;
      return NULL;
   }
 
   //Successful host name resolution
   *h_errnop = NETDB_SUCCESS;
 
   //Return a pointer to the hostent structure
   return result;
}
 
 
/**
 * @brief Convert host and service names to socket address
 * @param[in] node Host name or numerical network address
 * @param[in] service Service name or decimal number
 * @param[in] hints Criteria for selecting the socket address structures
 * @param[out] res Dynamically allocated list of socket address structures
 * @return On success, zero is returned. On error, a non-zero value is returned
 **/
 
int_t getaddrinfo(const char_t *node, const char_t *service,
   const struct addrinfo *hints, struct addrinfo **res)
{
   error_t error;
   size_t n;
   char_t *p;
   uint_t flags;
   uint32_t port;
   IpAddr ipAddr;
   ADDRINFO h;
   ADDRINFO *ai;
 
   //Check whether both node and service name are NULL
   if(node == NULL && service == NULL)
      return EAI_NONAME;
 
   //The hints argument is optional
   if(hints != NULL)
   {
      //If hints is not NULL, it points to an addrinfo structure that specifies
      //criteria that limit the set of socket addresses that will be returned
      h = *hints;
   }
   else
   {
      //If hints is NULL, then default criteria are used
      osMemset(&h, 0, sizeof(ADDRINFO));
      h.ai_family = AF_UNSPEC;
      h.ai_socktype = 0;
      h.ai_protocol = 0;
      h.ai_flags = 0;
   }
 
   //The user may provide a hint to choose between IPv4 and IPv6
   if(h.ai_family == AF_UNSPEC)
   {
      //The value AF_UNSPEC indicates that function should return socket
      //addresses for any address family (either IPv4 or IPv6)
      flags = 0;
   }
#if (IPV4_SUPPORT == ENABLED)
   else if(h.ai_family == AF_INET)
   {
      //The value AF_INET indicates that the function should return an IPv4
      //address
      flags = HOST_TYPE_IPV4;
   }
#endif
#if (IPV6_SUPPORT == ENABLED)
   else if(h.ai_family == AF_INET6)
   {
      //The value AF_INET6 indicates that the function should return an IPv6
      //address
      flags = HOST_TYPE_IPV6;
   }
#endif
   else
   {
      //The requested address family is not supported
      return EAI_FAMILY;
   }
 
   //Check whether a host name or numerical network address is specified
   if(node != NULL)
   {
      //If the AI_NUMERICHOST flag, then node must be a numerical network
      //address
      if((h.ai_flags & AI_NUMERICHOST) != 0)
      {
         //Convert the string representation to a binary IP address
         error = ipStringToAddr(node, &ipAddr);
      }
      else
      {
         //Resolve host address
         error = getHostByName(NULL, node, &ipAddr, flags);
      }
 
      //Check status code
      if(error == NO_ERROR)
      {
         //Successful host name resolution
      }
      else if(error == ERROR_IN_PROGRESS)
      {
         //Host name resolution is in progress
         return EAI_AGAIN;
      }
      else
      {
         //Permanent failure indication
         return EAI_FAIL;
      }
   }
   else
   {
      //Check flags
      if((h.ai_flags & AI_PASSIVE) != 0)
      {
#if (IPV4_SUPPORT == ENABLED)
         //IPv4 address family?
         if(h.ai_family == AF_INET || h.ai_family == AF_UNSPEC)
         {
            ipAddr.length = sizeof(Ipv4Addr);
            ipAddr.ipv4Addr = IPV4_UNSPECIFIED_ADDR;
         }
         else
#endif
#if (IPV6_SUPPORT == ENABLED)
         //IPv6 address family?
         if(h.ai_family == AF_INET6 || h.ai_family == AF_UNSPEC)
         {
            ipAddr.length = sizeof(Ipv6Addr);
            ipAddr.ipv6Addr = IPV6_UNSPECIFIED_ADDR;
         }
         else
#endif
         //Unknown address family?
         {
            //Report an error
            return EAI_ADDRFAMILY;
         }
      }
      else
      {
         //Invalid flags
         return EAI_BADFLAGS;
      }
   }
 
   //Only service names containing a numeric port number are supported
   port = osStrtoul(service, &p, 10);
   //Invalid service name?
   if(*p != '\0')
   {
      //The requested service is not available
      return EAI_SERVICE;
   }
 
#if (IPV4_SUPPORT == ENABLED)
   //IPv4 address?
   if(ipAddr.length == sizeof(Ipv4Addr))
   {
      //Select the relevant socket family
      h.ai_family = AF_INET;
      //Get the length of the corresponding socket address
      n = sizeof(SOCKADDR_IN);
   }
   else
#endif
#if (IPV6_SUPPORT == ENABLED)
   //IPv6 address?
   if(ipAddr.length == sizeof(Ipv6Addr))
   {
      //Select the relevant socket family
      h.ai_family = AF_INET6;
      //Get the length of the corresponding socket address
      n = sizeof(SOCKADDR_IN6);
   }
   else
#endif
   //Unknown address?
   {
      //Report an error
      return EAI_ADDRFAMILY;
   }
 
   //Allocate a memory buffer to hold the address information structure
   ai = osAllocMem(sizeof(ADDRINFO) + n);
   //Failed to allocate memory?
   if(ai == NULL)
   {
      //Out of memory
      return EAI_MEMORY;
   }
 
   //Initialize address information structure
   osMemset(ai, 0, sizeof(ADDRINFO) + n);
   ai->ai_family = h.ai_family;
   ai->ai_socktype = h.ai_socktype;
   ai->ai_protocol = h.ai_protocol;
   ai->ai_addr = (SOCKADDR *) ((uint8_t *) ai + sizeof(ADDRINFO));
   ai->ai_addrlen = n;
   ai->ai_next = NULL;
 
#if (IPV4_SUPPORT == ENABLED)
   //IPv4 address?
   if(ipAddr.length == sizeof(Ipv4Addr))
   {
      //Point to the IPv4 address information
      SOCKADDR_IN *sa = (SOCKADDR_IN *) ai->ai_addr;
 
      //Set address family and port number
      sa->sin_family = AF_INET;
      sa->sin_port = htons(port);
 
      //Copy IPv4 address
      sa->sin_addr.s_addr = ipAddr.ipv4Addr;
   }
   else
#endif
#if (IPV6_SUPPORT == ENABLED)
   //IPv6 address?
   if(ipAddr.length == sizeof(Ipv6Addr))
   {
      //Point to the IPv6 address information
      SOCKADDR_IN6 *sa = (SOCKADDR_IN6 *) ai->ai_addr;
 
      //Set address family and port number
      sa->sin6_family = AF_INET6;
      sa->sin6_port = htons(port);
      sa->sin6_flowinfo = 0;
      sa->sin6_scope_id = 0;
 
      //Copy IPv6 address
      ipv6CopyAddr(sa->sin6_addr.s6_addr, &ipAddr.ipv6Addr);
   }
   else
#endif
   //Unknown address?
   {
      //Clean up side effects
      osFreeMem(ai);
      //Report an error
      return EAI_ADDRFAMILY;
   }
 
   //Return a pointer to the allocated address information
   *res = ai;
 
   //Successful processing
   return 0;
}
 
 
/**
 * @brief Free socket address structures
 * @param[in] res Dynamically allocated list of socket address structures
 **/
 
void freeaddrinfo(struct addrinfo *res)
{
   ADDRINFO *next;
 
   //Free the list of socket address structures
   while(res != NULL)
   {
      //Get next entry
      next = res->ai_next;
      //Free current socket address structure
      osFreeMem(res);
      //Point to the next entry
      res = next;
   }
}
 
 
/**
 * @brief Convert a socket address to a corresponding host and service
 * @param[in] addr Generic socket address structure
 * @param[in] addrlen Length in bytes of the address
 * @param[out] host Output buffer where to store the host name
 * @param[in] hostlen Length of the host name buffer, in bytes
 * @param[out] serv Output buffer where to store the service name
 * @param[in] servlen Length of the service name buffer, in bytes
 * @param[in] flags Set of flags that influences the behavior of this function
 * @return On success, zero is returned. On error, a non-zero value is returned
 **/
 
int_t getnameinfo(const struct sockaddr *addr, socklen_t addrlen,
   char_t *host, size_t hostlen, char_t *serv, size_t servlen, int flags)
{
   uint16_t port;
 
   //At least one of hostname or service name must be requested
   if(host == NULL && serv == NULL)
      return EAI_NONAME;
 
#if (IPV4_SUPPORT == ENABLED)
   //IPv4 address?
   if(addr->sa_family == AF_INET &&
      addrlen >= (socklen_t) sizeof(SOCKADDR_IN))
   {
      SOCKADDR_IN *sa;
      Ipv4Addr ipv4Addr;
 
      //Point to the IPv4 address information
      sa = (SOCKADDR_IN *) addr;
 
      //The caller can specify that no host name is required
      if(host != NULL)
      {
         //Make sure the buffer is large enough to hold the host name
         if(hostlen < 16)
            return EAI_OVERFLOW;
 
         //Copy the binary representation of the IPv4 address
         ipv4CopyAddr(&ipv4Addr, &sa->sin_addr.s_addr);
         //Convert the IPv4 address to dot-decimal notation
         ipv4AddrToString(ipv4Addr, host);
      }
 
      //Retrieve port number
      port = ntohs(sa->sin_port);
   }
   else
#endif
#if (IPV6_SUPPORT == ENABLED)
   //IPv6 address?
   if(addr->sa_family == AF_INET6 &&
      addrlen >= (socklen_t) sizeof(SOCKADDR_IN6))
   {
      SOCKADDR_IN6 *sa;
      Ipv6Addr ipv6Addr;
 
      //Point to the IPv6 address information
      sa = (SOCKADDR_IN6 *) addr;
 
      //The caller can specify that no host name is required
      if(host != NULL)
      {
         //Make sure the buffer is large enough to hold the host name
         if(hostlen < 40)
            return EAI_OVERFLOW;
 
         //Copy the binary representation of the IPv6 address
         ipv6CopyAddr(&ipv6Addr, sa->sin6_addr.s6_addr);
         //Convert the IPv6 address to string representation
         ipv6AddrToString(&ipv6Addr, host);
      }
 
      //Retrieve port number
      port = ntohs(sa->sin6_port);
   }
   else
#endif
   //Invalid address?
   {
      //The address family was not recognized, or the address length was
      //invalid for the specified family
      return EAI_FAMILY;
   }
 
   //The caller can specify that no service name is required
   if(serv != NULL)
   {
      //Make sure the buffer is large enough to hold the service name
      if(servlen < 6)
         return EAI_OVERFLOW;
 
      //Convert the port number to string representation
      osSprintf(serv, "%" PRIu16, ntohs(port));
   }
 
   //Successful processing
   return 0;
}
 
 
/**
 * @brief Map an interface name into its corresponding index
 * @param[in] ifname Name of the interface
 * @return If ifname is the name of an interface, then the function returns the
 *   interface index corresponding to name ifname. Otherwise, the function
 *   returns zero
 **/
 
uint_t if_nametoindex(const char_t *ifname)
{
   uint_t i;
   uint_t index;
 
   //Initialize interface index
   index = 0;
 
   //Valid parameter?
   if(ifname != NULL)
   {
      //Loop through network interfaces
      for(i = 0; i < NET_INTERFACE_COUNT; i++)
      {
         //Matching interface name?
         if(osStrcmp(netInterface[i].name, ifname) == 0)
         {
            //Save the interface index
            index = netInterface[i].index + 1;
            //We are done
            break;
         }
      }
   }
 
   //Return the index corresponding to interface name
   return index;
}
 
 
/**
 * @brief Map an interface index into its corresponding name
 * @param[in] ifindex Interface index
 * @param[in] ifname Buffer of at least IF_NAMESIZE bytes
 * @return If ifindex is an interface index, then the function returns the value
 *   supplied in ifname, which points to a buffer now containing the interface
 *   name. Otherwise, the function returns a NULL pointer
 **/
 
char_t *if_indextoname(uint_t ifindex, char_t *ifname)
{
   uint_t i;
   char_t *name;
 
   //Initialize interface name
   name = NULL;
 
   //Valid parameters?
   if(ifindex != 0 && ifname != NULL)
   {
      //Loop through network interfaces
      for(i = 0; i < NET_INTERFACE_COUNT; i++)
      {
         //Matching interface index?
         if((netInterface[i].index + 1) == ifindex)
         {
            //Copy the name of the interface
            osStrcpy(ifname, netInterface[i].name);
            name = ifname;
 
            //We are done
            break;
         }
      }
   }
 
   //Return the name of the interface
   return name;
}
 
 
/**
 * @brief Convert a dot-decimal string into binary data in network byte order
 * @param[in] cp NULL-terminated string representing the IPv4 address
 * @return Binary data in network byte order
 **/
 
in_addr_t inet_addr(const char_t *cp)
{
#if (IPV4_SUPPORT == ENABLED)
   error_t error;
   Ipv4Addr ipv4Addr;
 
   //Convert a dot-decimal string to a binary IPv4 address
   error = ipv4StringToAddr(cp, &ipv4Addr);
 
   //Check status code
   if(error)
   {
      //The input is invalid
      return INADDR_NONE;
   }
   else
   {
      //Return the binary representation
      return ipv4Addr;
   }
#else
   //IPv4 is not implemented
   return INADDR_NONE;
#endif
}
 
 
/**
 * @brief Convert a dot-decimal string into binary form
 * @param[in] cp NULL-terminated string representing the IPv4 address
 * @param[out] inp Binary data in network byte order
 * @return The function returns non-zero if the address is valid, zero if not
 **/
 
int_t inet_aton(const char_t *cp, struct in_addr *inp)
{
#if (IPV4_SUPPORT == ENABLED)
   error_t error;
   Ipv4Addr ipv4Addr;
 
   //Convert a dot-decimal string to a binary IPv4 address
   error = ipv4StringToAddr(cp, &ipv4Addr);
 
   //Check status code
   if(error)
   {
      //The input is invalid
      return 0;
   }
   else
   {
      //Copy the binary representation of the IPv4 address
      inp->s_addr = ipv4Addr;
      //The conversion succeeded
      return 1;
   }
#else
   //IPv4 is not implemented
   return 0;
#endif
}
 
 
/**
 * @brief Convert a binary IPv4 address to dot-decimal notation
 * @param[in] in Binary representation of the IPv4 address
 * @return Pointer to the formatted string
 **/
 
const char_t *inet_ntoa(struct in_addr in)
{
   static char_t buf[16];
 
   //The string returned by the function resides in static memory area
   return inet_ntoa_r(in, buf, sizeof(buf));
}
 
 
/**
 * @brief Convert a binary IPv4 address to dot-decimal notation (reentrant version)
 * @param[in] in Binary representation of the IPv4 address
 * @param[out] buf Pointer to the buffer where to format the string
 * @param[in] buflen Number of bytes available in the buffer
 * @return Pointer to the formatted string
 **/
 
const char_t *inet_ntoa_r(struct in_addr in, char_t *buf, socklen_t buflen)
{
   //Properly terminate the string
   buf[0] = '\0';
 
#if (IPV4_SUPPORT == ENABLED)
   //Check the length of the buffer
   if(buflen >= 16)
   {
      //Convert the binary IPv4 address to dot-decimal notation
      ipv4AddrToString(in.s_addr, buf);
   }
#endif
 
   //Return a pointer to the formatted string
   return buf;
}
 
 
/**
 * @brief Convert an IPv4 or IPv6 address from text to binary form
 * @param[in] af Address family
 * @param[in] src NULL-terminated string representing the IP address
 * @param[out] dst Binary representation of the IP address
 * @return The function returns 1 on success. 0 is returned if the address
 *   is not valid. If the address family is not valid, -1 is returned
 **/
 
int_t inet_pton(int_t af, const char_t *src, void *dst)
{
   error_t error;
 
#if (IPV4_SUPPORT == ENABLED)
   //IPv4 address?
   if(af == AF_INET)
   {
      Ipv4Addr ipv4Addr;
 
      //Convert the IPv4 address from text to binary form
      error = ipv4StringToAddr(src, &ipv4Addr);
 
      //Check status code
      if(error)
      {
         //The input is invalid
         return 0;
      }
      else
      {
         //Copy the binary representation of the IPv4 address
         ipv4CopyAddr(dst, &ipv4Addr);
         //The conversion succeeded
         return 1;
      }
   }
   else
#endif
#if (IPV6_SUPPORT == ENABLED)
   //IPv6 address?
   if(af == AF_INET6)
   {
      Ipv6Addr ipv6Addr;
 
      //Convert the IPv6 address from text to binary form
      error = ipv6StringToAddr(src, &ipv6Addr);
 
      //Check status code
      if(error)
      {
         //The input is invalid
         return 0;
      }
      else
      {
         //Copy the binary representation of the IPv6 address
         ipv6CopyAddr(dst, &ipv6Addr);
         //The conversion succeeded
         return 1;
      }
   }
   else
#endif
   //Invalid address family?
   {
      //Report an error
      return -1;
   }
}
 
 
/**
 * @brief Convert an IPv4 or IPv6 address from binary to text
 * @param[in] af Address family
 * @param[in] src Binary representation of the IP address
 * @param[out] dst NULL-terminated string representing the IP address
 * @param[in] size Number of bytes available in the buffer
 * @return On success, the function returns a pointer to the formatted string.
 *   NULL is returned if there was an error
 **/
 
const char_t *inet_ntop(int_t af, const void *src, char_t *dst, socklen_t size)
{
#if (IPV4_SUPPORT == ENABLED)
   //IPv4 address?
   if(af == AF_INET && size >= INET_ADDRSTRLEN)
   {
      Ipv4Addr ipv4Addr;
 
      //Copy the binary representation of the IPv4 address
      ipv4CopyAddr(&ipv4Addr, src);
 
      //Convert the IPv4 address to dot-decimal notation
      return ipv4AddrToString(ipv4Addr, dst);
   }
   else
#endif
#if (IPV6_SUPPORT == ENABLED)
   //IPv6 address?
   if(af == AF_INET6 && size >= INET6_ADDRSTRLEN)
   {
      Ipv6Addr ipv6Addr;
 
      //Copy the binary representation of the IPv6 address
      ipv6CopyAddr(&ipv6Addr, src);
 
      //Convert the IPv6 address to string representation
      return ipv6AddrToString(&ipv6Addr, dst);
   }
   else
#endif
   //Invalid address family?
   {
      //Report an error
      return NULL;
   }
}
 
#endif