[an error occurred while processing this directive]

HP OpenVMS Systems Documentation

Content starts here HP TCP/IP Services for OpenVMS

HP TCP/IP Services for OpenVMS
Guide to IPv6


Previous Contents Index

7.4.3 Address Translation and Conversion Functions

The following functions are available for node name to address translation:

Function Description
gethostbyname() Returns IPv4 addresses.
getaddrinfo() Protocol-independent function for mapping names to addresses.
freeaddrinfo() Returns addrinfo() structures and dynamic storage to the system.

The following functions are available for address to node name translation:

Function Description
gethostbyaddr() Returns a node name for an IPv4 address.
getnameinfo() Protocol-independent function for mapping addresses to names.
freeaddrinfo() Returns addrinfo() structures and dynamic storage to the system.

The following address conversion functions convert both IPv4 and IPv6 addresses.

Function Description
inet_pton() Converts an address in its standard text presentation form to its numeric binary form, in network byte order.
inet_ntop() Converts a numeric address to a text string suitable for presentation.

7.4.4 Address-Testing Macros

Table 7-2 lists the currently defined address-testing macros and the return value for a valid test. To use these macros, include the following file in your application:



#include <in.h>

Table 7-2 Summary of Address-Testing Macros
Macro Return
IN6_IS_ADDR_UNSPECIFIED True, if specified type.
IN6_IS_ADDR_LOOPBACK True, if specified type.
IN6_IS_ADDR_MULTICAST True, if specified type.
IN6_IS_ADDR_LINKLOCAL True, if specified type.
IN6_IS_ADDR_SITELOCAL True, if specified type.
IN6_IS_ADDR_V4MAPPED True, if specified type.
IN6_IS_ADDR_V4COMPAT True, if specified type.
IN6_IS_ADDR_MC_NODELOCAL True, if specified scope.
IN6_IS_ADDR_MC_LINKLOCAL True, if specified scope.
IN6_IS_ADDR_MC_SITELOCAL True, if specified scope.
IN6_IS_ADDR_MC_ORGLOCAL True, if specified scope.
IN6_IS_ADDR_MC_GLOBAL True, if specified scope.
IN6_ARE_ADDR_EQUAL True, if addresses are equal.

7.5 Advanced API

The advanced API provides support for advanced applications that may need knowledge of IPv6 headers. These applications commonly use raw sockets to access IPv6 or ICMPv6 header fields. The advanced interface provides the following:

  • Support for portable interfaces for applications that use raw sockets under IPv6
  • Functions to access router headers
  • Functions to access option headers

7.5.1 Using IPv6 Raw Sockets

Raw sockets are used in both IPv4 and IPv6 to bypass the TCP and UDP transport layers.

Table 7-3 describes the principal differences between IPv4 and IPv6 raw sockets.

Table 7-3 Differences Between IPv4 and IPv6 Raw Sockets
  IPv4 IPv6
Use Access ICMPv4, IGMPv4, and to read and write IPv4 datagrams that contain a protocol field the kernel does not recognize. Access ICMPv6 and to read and write IPv6 datagrams that contain a Next Header field the kernel does not recognize.
Byte order Not specified. Network byte order for all data sent and received.
Send and receive complete packets Yes No. Uses ancillary data objects to transfer extension headers and hop limit information.

For output, applications can modify all fields, except for the flow label field, by using ancillary data or socket options, or both.

For input, applications can access all fields, except for the flow label, version number, and Next Header fields, and all extension headers by using ancillary data.

For IPv6 raw sockets other than ICMPv6 raw sockets, the application must set the IPV6_CHECKSUM socket option. For example:


int offset = 2;
setsockopt (fd, IPPROTO_IPV6, IPV6_CHECKSUM, &offset, sizeof(offset));

This enables the kernel to compute and store a checksum for output and to verify the checksum on input. This relieves the application from having to perform source address selection on all outgoing packets. This socket option is disabled by default. You can explicitly disable this option by setting the offset variable to -1.

Using IPv6 raw sockets, an application can access the following information:

  • ICMPv6 messages
  • IPv6 header
  • Routing header
  • IPv6 options headers: hop-by-hop options header and destination options header

The following sections describe how to access this information.

7.5.1.1 Accessing ICMPv6 Messages

An ICMPv6 raw socket is a socket that is created by calling the socket function with the PF_INET6 , SOCK_RAW , and IPPROTO_ICMPV6 arguments.

The kernel calculates and inserts the ICMPv6 checksum for all outbound ICMPv6 packets and verifies the checksum for all received packets. If the received checksum is incorrect, the packet is discarded.

Because ICMPv6 is a superset of ICMPv4, an ICMPv6 raw socket can receive many more messages than an ICMPv4 raw socket. By default, when you create an ICMPv6 raw socket, it passes all ICMPv6 message types to an application. An application, however, does not need access to all messages. An application can specify the ICMPv6 message types it wants passed by creating an ICMPv6 filter.

The ICMPv6 filter has a datatype of struct icmp6_filter . Use getsockopt() to retrieve the current filter and setsockopt() to store the filter. For example, to enable filtering of ICMPv6 messages, use the ICMP6_FILTER option, as follows:


struct icmp6_filter myfilter;

setsockopt (fd, IPPROTO_ICMPV6, IPV6_FILTER, &(myfilter), (sizeof)(myfilter));

The value of myfilter is an ICMPv6 message type between 0 and 255.

Table 7-4 describes the ICMPv6 filter macros.

Table 7-4 ICMPv6 Filtering Macros
Macro Description
ICMP6_FILTER_SETPASSALL Passes all ICMPv6 messages to an application.
ICMP6_FILTER_SETBLOCKALL Blocks all ICMPv6 messages from being passed to an application.
ICMP6_FILTER_SETPASS Passes ICMPv6 messages of a given type to an application.
ICMP6_FILTER_SETBLOCK Blocks ICMPv6 messages of a given type from being passed to an application.
ICMP6_FILTER_WILLPASS Returns true, if specified message type is passed to application.
ICMP6_FILTER_WILLBLOCK True, if the specified message type is blocked from being passed to an application.

To clear an installed filter, call setsockopt() for the ICMP_FILTER option with a zero-length filter.

The kernel does not perform any validity checks on message type, message content, or packet structure. The application is responsible for checking them.

7.5.1.2 Accessing the IPv6 Header

When using IPv6 raw sockets, applications must be able to receive the IPv6 header content. To receive this optional information, use the setsockopt() function with the appropriate socket option.

Table 7-5 describes the socket options for receiving optional information.

Table 7-5 Optional Information and Socket Options
Optional Information Socket Option cmsg_type
Source and destination IPv6 address, and sending and receiving interface IPV6_RECVPKTINFO IPV6_PKTINFO
Hop limit IPV6_RECVHOPLIMIT IPV6_HOPLIMIT
Routing header IPV6_RECVRTHDR IPV6_RTHDR
Hop-by-Hop options IPV6_RECVHOPOPTS IPV6_HOPOPTS
Destination options IPV6_RECVDSTOPTS IPV6_DSTOPTS

The recvmsg() function returns the received data as one or more ancillary data objects in a cmsghdr data structure.

To determine the value of a socket option, use the getsockopt() function with the corresponding option. If the IPV6_RECVPKTINFO option is not set, the function returns an in6_pktinfo data structure with ipi6_addr set to in6addr_any and ipi6_addr set to zero. For other options, the function returns an option_len value of zero if there is no option value.

An application can receive the following IPv6 header information as ancillary data from incoming packets:

  • Destination IPv6 address
  • Interface index
  • Hop limit

The IPv6 address and interface index are contained in a in6_pktinfo data structure that is received as ancillary data with the recvmsg() function. the in6_pktinfo data structure is defined in in.h . The tasks associated with the IPv6 header are:

  • Receiving an IPv6 address
    If the IPV6_RECVPKTINFO option is enabled, the recvmsg() function returns a in6_pktinfo data structure as ancillary data. The ipi6_addr member contains the destination IPv6 address from the received packet. For TCP sockets, the destination address is the local address of the connection.
  • Receiving an interface
    If the IPV6_RECVPKTINFO option is enabled, the recvmsg() function returns a in6_pktinfo data structure as ancillary data. The ipi6_ifindex member contains the interface index of the interface that received the packet.
  • Receiving a hop limit
    If the IPV6_RECVHOPLIMIT option is enabled, the recvmsg() function returns a cmsghdr data structure as ancillary data. The cmsg_type member is IPV6_HOPLIMIT and the cmsg_data[] member contains the first byte of the integer hop limit.

7.5.1.3 Accessing the IPv6 Routing Header

The advanced sockets API enables you to access the IPv6 routing header. The routing header is an IPv6 extension header that enables an application to perform source routing. RFC 2460 defines the type 0 routing header, which supports up to 127 intermediate nodes, or 128 hops.

Table 7-6 describes the sockets calls that an application uses to build and examine routing headers.

Table 7-6 Socket Calls for Routing Header Name Description
Function Description
inet6_rth_space() Returns the number of bytes required for a routing header.
inet6_rth_init() Initializes buffer data for a routing header.
inet6_rth_add() Adds one address to a routing header.
inet6_rth_reverse() Reverses the order of fields in a routing header.
inet6_rth_segments() Returns the number of segments, or addresses, in a routing header.
inet6_rth_getaddr() Fetches one address from a routing header.

The tasks associated with the routing header are:

  • Receiving a routing header
    To receive a routing header, an application calls setsockopt() with the IPV6_RECVRTHDR option enabled.
    For each received routing header, the kernel passes one ancillary data object in a cmsghdr structure with the cmsg_type member set to IPV6_RTHDR . An application processes a routing header by calling inet6_rth_reverse() , inet6_rth_segments() , and inet6_rth_getaddr() .
  • Sending a routing header
    To send a routing header, an application specifies the header either as ancillary data in a call to sendmsg() or by calling setsockopt() . An application can remove a sticky routing header by calling setsockopt() for the IPV6_RTHDR option and specifying a option length of zero.
    When using ancillary data,the application sets cmsg_level member to IPPROTO_IPV6 and the cmsg_type member to IPV6_RTHDR . Use the inet6_rth_space() , inet6_rth_init() , and inet6_rth_add() calls to build the routing header.
    When an application specifies a routing header, the destination address specified in a call to the connect() , sendto() , or sendmsg() function is the final destination of the datagram. Therefore, the routing header contains the addresses of all intermediate nodes.
    Because of the order of extension headers specified in RFC 2460, an application can specify only one outgoing routing header.

7.5.1.4 Accessing the IPv6 Options Headers

The advanced sockets API enables applications to access the following IPv6 options headers:

  • Hop-by-hop header
    A single hop-by-hop options header can contain a variable number of hop-by-hop options. Each option is encoded with a type, length, and value (TLV). The application uses sticky options or ancillary data to communicate this information with the kernel.
  • Destination header
    One or more destination options headers can contain a variable number of destination options. A destination options header appearing before a routing header is processed by the first and subsequent destinations specified in the routing header. A destination option appearing after the routing header is processed only by the final destination. Each option is encoded with a type, length, and value (TLV). The application uses sticky options or ancillary data to communicate this information with the kernel.

See RFC 2460 for additional information about the alignment requirements of the headers and ordering of the extensions headers.

Table 7-7 lists the sockets calls that an application uses to build and examine hop-by-hop and destination headers.

Table 7-7 Socket Calls for Options Headers
Function Description
inet6_opt_init() Initializes buffer data for options.
inet6_opt_append() Adds an option to the options header.
inet6_opt_finish() Finishes adding options to the options header.
inet6_opt_set_val() Adds one component of the option content to the options header.
inet6_opt_next() Extracts the next option from the options header.
inet6_opt_find() Extracts an option of a specified type from the options header.
inet6_opt_get_val() Retrieves one component of the option content from the options header.

The tasks associate with options headers are:

  • Receiving hop-by-hop options
    To receive a hop-by-hop options header, an application calls setsockopt() with the IPV6_RECVHOPOPTS option enabled.
    When using ancillary data, the kernel passes a hop-by-hop options header to the application and sets the cmsg_level member to IPPROTO_IPV6 and the cmsg_type member to IPV6_HOPOPTS .
    An application retrieves these options by calling inet6_opt_next() , inet6_opt_find() , and inet6_opt_get_val() .
  • Sending hop-by-hop options
    To send a hop-by-hop options header, an application specifies the header either as ancillary data in a call to sendmsg() or by calling setsockopt() An application can remove a sticky hop-by-hop options header by calling setsockopt() for the IPV6_HOPOPTS option and specifying a option length of zero (0).
    When using ancillary data, all hop-by-hop options are specified by a single ancillary data object. The application sets cmsg_level member to IPPROTO_IPV6 and the cmsg_type member to IPV6_HOPOPTS . Use the inet6_opt_init() , inet6_opt_append() , inet6_opt_finish() , and inet6_opt_set_val() calls to build the option header.
  • Receiving destination options
    To receive a destination options header, an application calls setsockopt() with the IPV6_RECVDSTOPTS option enabled. The kernel passes each destination option to the application as one ancillary data object and sets the cmsg_level member to IPPROTO_IPV6 and the cmsg_type member to IPV6_DSTOPTS .
    An application processes these options by calling inet6_opt_next() , inet6_opt_find() , and inet6_opt_get_val() .
  • Sending destination options
    To send a destination options header, an application specifies the header either as ancillary data in a call to sendmsg() or by calling setsockopt() .
    An application can remove a sticky hop-by-hop options header by calling setsockopt() for either the IPV6_RTHDRDSTOPTS or the IPV6_DSTOPTS option and specifying a option length of zero (0).
    In accordance with RFC 2460, the API assumes that the extension headers are in order. Only one set of destination options can precede a routing header and only one set of destination options can follow a routing header.
    Each set can contain one or more options, but each set is considered a single extension header.
    When using ancillary data, the application passes a destination options header to the kernel in one of the following ways:
    • For destination options that precede a routing header, the application sets the cmsg_level member to IPPROTO_IPV6 and the cmsg_type member to IPV6_RTHDRDSTOPTS . Any setsockopt() or ancillary data is ignored unless the application explicitly specifies its own routing header.
    • For destination options that follow a routing header or when no routing header is specified, the application sets the cmsg_level member to IPPROTO_IPV6 and the cmsg_type member to IPV6_DSTOPTS .

    An application builds these options by calling inet6_opt_init() , inet6_opt_append() , inet6_opt_finish() , and inet6_opt_set_val() .


Previous Next Contents Index