Internet Engineering Task Force                   R. E. Gilligan (Sun)
INTERNET-DRAFT                                   S. Thomson (Bellcore)
                                                    J. Bound (Digital)

                                                        July 7,

                                                    November 21, 1995

                IPv6 Program Interfaces for BSD Systems
                   <draft-ietf-ipngwg-bsd-api-02.txt>
                   <draft-ietf-ipngwg-bsd-api-03.txt>

Abstract

In order to implement the version 6 Internet Protocol (IPv6) [1] in an
operating system based on Berkeley Unix (4.x BSD), changes must be made
to the application program interface (API).  TCP/IP applications written
for BSD-based operating systems have in the past enjoyed a high degree
of portability because most of the systems derived from BSD provide the
same API, known informally as "the socket interface".  We would like the
same portability with IPv6.  This memo presents a set of extensions to
the BSD socket API to support IPv6.  The changes include a new data
structure to carry IPv6 addresses, new name to address translation
library functions, new address conversion functions, and some new
setsockopt() options.  The extensions are designed to provide access to
IPv6 features, while introducing a minimum of change into the system and
providing complete compatibility for existing IPv4 applications.

Status of this Memo

This document is an Internet Draft.  Internet Drafts are working
documents of the Internet Engineering Task Force (IETF), its Areas,
and its Working Groups.  Note that other groups may also distribute
working documents as Internet Drafts.

Internet Drafts are draft documents valid for a maximum of six months.
This Internet Draft expires on January 6, May 21, 1996.  Internet Drafts may be
updated, replaced, or obsoleted by other documents at any time.  It is
not appropriate to use Internet Drafts as reference material or to cite
them other than as a "working draft" or "work in progress."

To learn the current status of any Internet-Draft, please check the
1id-abstracts.txt listing contained in the Internet-Drafts Shadow
Directories on ds.internic.net, nic.nordu.net, ftp.isi.edu, or
munnari.oz.au.

Distribution of this memo is unlimited.

1.  Introduction.

While IPv4 addresses are 32-bits long, IPv6 nodes are identified by
128-bit addresses.  The socket interface API make the size of an IP address
quite visible to an application; virtually all TCP/IP applications for
BSD-based systems have knowledge of the size of an IP address.  Those
parts of the API that expose the addresses need to be extended to
accommodate the larger IPv6 address size.  IPv6 also introduces new
features, some of which must be made visible to applications via the
API.  This paper defines a set of extensions to the socket interface API to
support the larger address size and new features of IPv6.

This specification is preliminary.  The  These API extensions are expected to
evolve as we gain more implementation experience.

2.  Design Considerations

There are a number of important considerations in designing changes to
this well-worn API:

   -    The extended API should provide both source and binary
        compatibility for programs written to the original API.  That
        is, existing program binaries should continue to operate when
        run on a system supporting the new API.  In addition, existing
        applications that are re-compiled and run on a system supporting
        the new API should continue to operate.  Simply put, the API
        changes for IPv6 should not break existing programs.

   -    The changes to the API should be as small as possible in order
        to simplify the task of converting existing IPv4 applications to
        IPv6.

   -    Where possible, applications should be able to use the extended
        API to interoperate with both IPv6 and IPv4 hosts.  Applications
        should not need to know which type of host they are
        communicating with.

   -    IPv6 addresses carried in data structures should be 64-bit
        aligned.  This is necessary in order to obtain optimum
        performance on 64-bit machine architectures.

Because of the importance of providing IPv4 compatibility in the API,
our
these extensions are explicitly designed to operate on machines that
provide complete support for both IPv4 and IPv6.  A subset of this API
could probably be designed for operation on systems that support only
IPv6.  However, this is not addressed in this document.

2.1.  Overview of Changes  What Needs to be Changed

The socket interface API consists of a few distinct components:

   -    Core socket functions.

   -    Address data structures.

   -    Name-to-address translation functions.

   -    Address conversion functions.

The core socket functions -- those functions that deal with such things
as setting up and tearing down TCP connections, and sending and
receiving UDP packets -- were designed to be transport independent.
Where protocol addresses are passed as function arguments, they are
carried via opaque pointers.  A protocol specific address data structure
is defined for each protocol that the socket functions support.
Applications must cast these protocol specific address structures into
the generic "sockaddr" data type when using the socket functions.  These
functions need not change for IPv6, but a new IPv6 specific address data
structure is needed.

The "sockaddr_in" structure is the protocol specific data structure for
IPv4.  This data structure actually includes 8-octets of unused space,
and it is tempting to try to use this space to adapt the sockaddr_in
structure to IPv6.  Unfortunately, the sockaddr_in structure is not
large enough to hold the 16-octet IPv6 address as well as the other
information (2-octet address family and 2-octet port number) that is
needed.  So a new address data structure must be defined for IPv6.

The name-to-address translation functions in the socket interface are
gethostbyname() and gethostbyaddr().  Gethostbyname() does not provide
enough flexibility to accommodate more than one protocol family.  To
solve this problem, we introduced a new name-to-address translation
function which is analogous to gethostbyname(), but supports addresses
in both the IPv4 and IPv6 address families.  Gethostbyaddr() does not,
strictly speaking, need to be replaced since it carries an address
family argument and can be extended to support both address families
without introducing compatibility problems.  However, we have chosen to
introduce a new function to maintain symmetry with the replacement to
gethostbyname().  The new functions both carry an address family
parameter, so they can be extended to operate with other protocol
families in addition to IPv4 and IPv6.

The address conversion functions -- inet_ntoa() and inet_addr() --
convert IPv4 addresses between binary and printable form.  These
functions are quite specific to 32-bit IPv4 addresses.  We have designed

two analogous functions which convert both IPv4 and IPv6 addresses, and
carry an address type parameter so that they can be extended to other
protocol families as well.

Finally, a few miscellaneous features are needed to support IPv6.  A new
interface is needed in order to support the IPv6 flow label and priority
header fields.  New interfaces are needed in order to receive IPv6
multicast packets and control the sending of multicast packets.  And an
interface is necessary in order to pass IPv6 source route information
between the application and the system.

3.  Implementation Experience

A few issues exposed in experimenting with prototype implementations of
IPv6 helped to guide the design of this API:

First, we discovered that, by providing a way to represent the addresses
of IPv4 nodes as IPv6 addresses, we could greatly simplify  Socket Interface

This section specifies the
applications' task socket interface changes for IPv6.

The data types of providing IPv4 compatibility.  New applications
could interoperate with IPv4 nodes by using the new API and expressing structure elements given in the addresses of IPv4 nodes following section
are intended to be examples, not absolute requirements.  System
implementations may use other types if they interoperate with are appropriate.  In some
cases, such as IPv6 addresses.
For example, when a client application could open field of a TCP connection to an IPv4
server by giving data structure holds a protocol value,
the IPv6 representation structure field must be of the server's IPv4 address some minimum size.  These size
requirements are noted in the connect() call.  Most applications do not even need to know whether
the peer is an IPv4 or IPv6 node.  Such applications can simply treat
IPv6 addresses as opaque values; They need not understand text.  For example, since the
"structure" by which IPv4 addresses UDP and TCP
port values are encoded within IPv6 addresses.
Yet 16-bit quantities, the structure can sin6_port field must be decoded by those applications that do need to
know whether the peer at least
a 16-bit data types.  The sin6_port field is IPv6 or IPv4.  This should prove to be specified as a
significant simplification since most applications will need to
interoperate with both IPv4 and IPv6 nodes for some time to come.

Second, we learned that existing applications written to the IPv4 API
could be made to interoperate with IPv6 nodes to a limited degree.  This
technique does not work for all applications, u_int16_t
type, but does for certain
applications, such as those an implementation may use any data type that do not "look at" the peer is at least
16-bits long.

3.1.  New Address Family

A new address that family macro, named AF_INET6, is provided by the API.  (e.g. defined in
<sys/socket.h>.  The AF_INET6 definition is used to distinguish between
the source original sockaddr_in address provided by data structure, and the
recvfrom() function when a UDP packet new
sockaddr_in6 data structure.

A new protocol family macro, named PF_INET6, is received, or defined in
<sys/socket.h>.  Like most of the client address
returned by other protocol family macros, this
will usually be defined to have the accept() function.)

Third, we learned that same value as the common application practice of passing open
socket descriptors between processes across an exec() call can cause
problems.  It corresponding
address family macro:

        #define PF_INET6        AF_INET6

The PF_INET6 is possible, for example, for an application using used in the
extended API first argument to pass an open socket the socket() function to
indicate that an older application using the
original API.  The old application could be confused if the IPv6 socket
functions return is being created.

3.2. IPv6 address structures to it.  The solution designed
was Address Data Structure

A new data structure to provide hold a mechanism by single IPv6 address is defined as
follows:

        struct in6_addr {
                u_char  s6_addr[16];    /* IPv6 address */
        }

This data structure contains an array of sixteen 8-bit elements, which applications could have explicit
control over what form of addresses are returned.

4.  Interface Specification

This section specifies the interface changes for IPv6.
make up one 128-bit IPv6 address.  The data types of IPv6 address is stored in network
byte order.

Applications obtain the declaration for this structure elements given in by including
the following section
are intended to be examples, not absolute requirements.  System
implementations may use other types if they are appropriate. system header file <netinet/in.h>.

3.3.  Socket Address Structure for 4.3 BSD-Based Systems

In some
cases, such as when a field of the socket interface, a different protocol-specific data structure holds a protocol value, is
defined to carry the structure field must be addresses for each of some minimum size.  These size
requirements are noted in the text.  For example, since the UDP and TCP
port values are 16-bit quantities, the sin6_port field must protocol suite.  Each
protocol-specific data structure is designed so it can be at least cast into a 16-bit
protocol-independent data types.  We specify structure -- the sin6_port field as "sockaddr" structure.  Each
has a u_short type,
but an implementation may use any "family" field which overlays the "sa_family" of the sockaddr data
structure.  This field can be used to identify the type that of the data
structure.

The sockaddr_in structure is at least 16-bits
long.

4.1.  New Address Family

A new the protocol-specific address family macro, named AF_INET6, is defined in
<sys/socket.h>.  The AF_INET6 definition data
structure for IPv4.  It is used to distinguish pass addresses between
the original sockaddr_in address data structure, applications
and the new
sockaddr_in6 data structure.

A new protocol family macro, named PF_INET6, is defined system in
<sys/socket.h>.  Like most of the other protocol family macros, this
will usually be defined to have the same value as the corresponding
address family macro:

        #define PF_INET6        AF_INET6

The PF_INET6 is used in the first argument to the socket() function to
indicate that an IPv6 socket is being created.

4.2. IPv6 Address Data Structure

A new data structure to hold a single IPv6 address is defined in
<netinet/in.h>:

        struct in_addr6 {
                u_char  s6_addr[16];    /* IPv6 address */
        }

This data structure contains an array of sixteen 8-bit elements, which
make up one 128-bit IPv6 address.

The IPv6 address is stored in network byte order.

4.3.  Socket Address Structure for 4.3 BSD-Based Systems

In the socket interface, a different protocol-specific data structure
is defined to carry the addresses for each of the protocol suite.
Each protocol-specific data structure is designed so it can be cast
into a protocol-independent data structure -- the "sockaddr"
structure.  Each has a "family" field which overlays the "sa_family"
of the sockaddr data structure.  This field can be used to identify
the type of the data structure. functions. The sockaddr_in structure is the protocol-specific address data following structure for IPv4.  It is used to pass addresses between applications
and the system in the socket functions.  We have
defined the following
structure in <netinet/in.h> to carry IPv6 addresses:

        struct sockaddr_in6 {
                u_short
                u_int16_t       sin6_family;    /* AF_INET6 */
                u_short
                u_int16_t       sin6_port;      /* Transport layer port # */
                u_long
                u_int32_t       sin6_flowinfo;  /* IPv6 flow information */
                struct in_addr6 in6_addr sin6_addr;      /* IPv6 address */
        };

This structure is designed to be compatible with the sockaddr data
structure used in the 4.3 BSD release.

The sin6_family field is used to identify this as a sockaddr_in6
structure.  This field is designed to overlay the sa_family field when
the buffer is cast to a sockaddr data structure.  The value of this
field must be AF_INET6.

The sin6_port field is used to store the 16-bit UDP or TCP port
number.  This field is used in the same way as the sin_port field of
the sockaddr_in structure.  The port number is stored in network byte
order.

The sin6_flowinfo field is a 32-bit field that is used to store three
pieces of information: the 24-bit IPv6 flow label, the 4-bit priority

field, and a 1-bit loose/strict source routing flag.  The IPv6 flow
label is represented as the low-order 24-bits of the 32-bit field.  The
priority is represented in the next 4-bits above this, and the
loose/strict flag is the 1 bit above this.  The high-order 3 bits of
this field are reserved.  The sin6_flowinfo field is stored in network
byte order.  The use of the flow label and priority fields are explained
in sec 4.9.  The use of the loose/strict flag is discussed in section
4.10.

The sin6_addr field is a single in_addr6 in6_addr structure (defined in the
previous section).  This field holds one 128-bit IPv6 address.  The
address is stored in network byte order.

The ordering of elements in this structure is specifically designed so
that the sin6_addr field will be aligned on a 64-bit boundary.  This is
done for optimum performance on 64-bit architectures.

4.4.

Applications obtain the declaration of the sockaddr_in6 structure by
including the system header file <netinet/in.h>.

3.4. Socket Address Structure for 4.4 BSD-Based Systems

The 4.4 BSD release includes a small, but incompatible change to the
socket interface.  The "sa_family" field of the sockaddr data structure
was changed from a 16-bit value to an 8-bit value, and the space saved
used to hold a length field, named "sa_len". The sockaddr_in6 data
structure given in the previous section can not be correctly cast into
the newer sockaddr data structure.  For this reason, we have defined the following
alternative IPv6 address data structure is provided to be used on
systems based on 4.4 BSD:

        #define SIN6_LEN

        struct sockaddr_in6 {
                u_char          sin6_len;       /* length of this struct */
                u_char          sin6_family;    /* AF_INET6 */
                u_short
                u_int16_t       sin6_port;      /* Transport layer port # */
                u_long
                u_int32_t       sin6_flowinfo;  /* IPv6 flow information */
                struct in_addr6 in6_addr sin6_addr;      /* IPv6 address */
        };

This structure is defined in the <netinet/in.h> header file.

The only differences between this data structure and the 4.3 BSD variant
are the inclusion of the length field, and the change of the family
field to a 8-bit data type.  The definitions of all the other fields are
identical to the 4.3 BSD variant defined in the previous section.

Systems that provide this version of the sockaddr_in6 data structure
must include also declare the SIN6_LEN macro definition in <netinet/in.h>. as a result of including the

<netinet/in.h> header file.  This macro allows applications to determine
whether they are being built on a system that supports the 4.3 BSD or
4.4 BSD variants of the data structure.  Applications can be written to
run on both systems by simply making their assignments and use of the
sin6_len field conditional on the SIN6_LEN field.  For example, to fill
in an IPv6 address structure in an application, one might write:

        struct sockaddr_in6 sin6;

        bzero((char *) &sin6, sizeof(struct sockaddr_in6));
        #ifdef SIN6_LEN
        sin6.sin6_len = sizeof(struct sockaddr_in6);
        #endif
        sin6.sin6_family = AF_INET6;
        sin6.sin6_port = htons(23);

Note that the size of the sockaddr_in6 structure is larger than the size
of the sockaddr structure.  Applications that use the sockaddr_in6
structure need to be aware that they can not use sizeof(sockaddr) to
allocate a buffer to hold a sockaddr_in6 structure.  They should use
sizeof(sockaddr_in6) instead.

4.5.

3.5.  The Socket Functions

Applications use the socket() function to create a socket descriptor
that represents a communication endpoint.  The arguments to the socket()
function tell the system which protocol to use, and what format address
structure will be used in subsequent functions.  For example, to create
an IPv4/TCP socket, applications make the call:

        s = socket (PF_INET, SOCK_STREAM, 0);

To create an IPv4/UDP socket, applications make the call:

        s = socket (PF_INET, SOCK_DGRAM, 0);

Applications may create IPv6/TCP and IPv6/UDP sockets by simply using
the constant PF_INET6 instead of PF_INET in the first argument.  For
example, to create an IPv6/TCP socket, applications make the call:

        s = socket (PF_INET6, SOCK_STREAM, 0);

To create an IPv6/UDP socket, applications make the call:

        s = socket (PF_INET6, SOCK_DGRAM, 0);

Once the application has created a PF_INET6 socket, it must use the

sockaddr_in6 address structure when passing addresses in to the system.
The functions which the application uses to pass addresses into the
system are:

           bind()
           connect()
           sendmsg()
           sendto()

The system will use the sockaddr_in6 address structure to return
addresses to applications that are using PF_INET6 sockets.  The
functions that return an address from the system to an application
are:

           accept()
           recvfrom()
           recvmsg()
           getpeername()
           getsockname()

No changes to the syntax of the socket functions are needed to support
IPv6, since the all of the "address carrying" functions use an opaque
address pointer, and carry an address length as a function argument.

4.6.

3.6.  Compatibility with IPv4 Applications

In order to support the large base of applications using the original
API, system implementations must provide complete source and binary
compatibility with the original API.  This means that systems must
continue to support PF_INET sockets and the sockaddr_in addresses
structure.  Applications must be able to create IPv4/TCP and IPv4/UDP
sockets using the PF_INET constant in the socket() function, as
described in the previous section.  Applications should be able to hold
a combination of IPv4/TCP, IPv4/UDP, IPv6/TCP and IPv6/UDP sockets
simultaneously within the same process.

Applications using the original API should continue to operate as they
did on systems supporting only IPv4.  That is, they should continue to
interoperate with IPv4 nodes.  It is not clear, though, how, or even if,
those IPv4 applications should interoperate with IPv6 nodes.  The open
issues section (section 7) 9) discusses some of the alternatives.

4.7.

3.7.  Compatibility with IPv4 Nodes

The API also provides a different type of compatibility: the ability for
applications using the extended API to interoperate with IPv4 nodes.

This feature uses the IPv4-mapped IPv6 address format defined in the
IPv6 addressing architecture specification [3].  This address format
allows the IPv4 address of an IPv4 node to be represented as an IPv6
address.  The IPv4 address is encoded into the low-order 32-bits of the
IPv6 address, and the high-order 96-bits hold the fixed prefix
0:0:0:0:0:FFFF.  IPv4-mapped addresses are written as follows:

        ::FFFF:<IPv4-address>

Applications may use PF_INET6 sockets to open TCP connections to IPv4
nodes, or send UDP packets to IPv4 nodes, by simply encoding the
destination's IPv4 address as an IPv4-mapped IPv6 address, and passing
that address, within a sockaddr_in6 structure, in the connect() or
sendto() call.  When applications use PF_INET6 sockets to accept TCP
connections from IPv4 nodes, or receive UDP packets from IPv4 nodes, the
system returns the peer's address to the application in the accept(),
recvfrom(), or getpeername() call using a sockaddr_in6 structure encoded
this way.

We expect that few

Few applications will likely need to know which type of node they are
interoperating with.  However, for those applications that do need to
know, the following function inet6_isipv4addr() function, defined in section 6.3, is provided:

        int is_ipv4_addr (const struct in_addr6 *ap);
provided.

3.8.  Flow Information

The "ap" argument IPv6 header has a 24-bit field to this function points hold a "flow label", and a 4-bit
field to hold a buffer holding an IPv6
address "priority" value.  Applications have control over what
values for these fields are used in network byte order.  The function returns true (non-zero)
if packets that address is an IPv4-mapped address, they originate, and returns 0 otherwise.
When an application using
have access to the extended API accepts a TCP connection,
or receives a UDP packet, it may determine whether field values of packets that they receive.

The sin6_flowinfo field of the peer sockaddr_in6 structure encodes three
pieces of information: IPv6 flow label, IPv6 priority, and a
strict/loose source routing flag which is discussed in section 4.2.
Applications use this field to set the flow label and priority in IPv6
headers of packets they generate, and to retrieve the flow label and
priority from the packets they receive.  The header fields of an IPv4
node
actively opened TCP connection are set by applying assigning in the is_ipv4_addr() function to sin6_flowinfo
field of the destination address returned
by accept() or recvfrom().

4.8.  Sockets Passed Across exec()

Unix allows open sockets to sockaddr_in6 structure passed in the
connect() function.  The same technique can be used with the
sockaddr_in6 structure passed across an exec() call.  It is a
relatively common application practice in to pass open sockets across
exec() calls.  Because of this, it is possible for an application
using the original API to pass an open PF_INET socket to an
application that is expecting sendto() or sendmsg() function
to receive a PF_INET6 socket.
Similarly, it is possible for an application using set the extended API to
pass an open PF_INET6 socket to an application using flow label and priority fields of UDP packets.  Similarly,
the original API,
which would be equipped only to deal with PF_INET sockets.  Either flow label and priority values of
these cases could cause problems, because received UDP packets and accepted
TCP connections are reflected in the application which is
passed sin6_flowinfo field of the open socket might not know how
sockaddr_in6 structure returned to decode the address
structures returned in subsequent socket functions.

To remedy this problem, we have defined a new setsockopt() option that
allows an application to "transform" a PF_INET6 socket into a PF_INET
socket by the recvfrom(),
recvmsg(), and vice-versa.

An IPv6 application that is passed an open socket from accept() functions.  And an unknown
process application may use specify the IP_ADDRFORM setsockopt() option
flow label and priority to "convert" use in transmitted packets of a passively
accepted TCP connection, by setting the
socket to PF_INET6.  Once that has been done, sin6_flowinfo field of the system will return
sockaddr_in6

address structures passed in subsequent socket functions.
Similarly, an IPv6 application that is about to pass an open PF_INET6
socket the bind() function.

Implementations provide two bitmask constant declarations to a program that may not be IPv6 capable may "downgrade" help
applications select out the
socket flow label and priority fields.  These
constants are:

        IPV6_FLOWINFO_FLOWLABEL
        IPV6_FLOWINFO_PRIORITY

These constants can be applied to PF_INET before calling exec().  After that, the system will
return sockaddr_in address structures sin6_flowinfo field of addresses
returned to the application that was
exec()'ed.

The macro definition application, for IP_ADDRFORM is in <netinet/in.h>.

The IP_ADDRFORM option is at example:

        struct sockaddr_in6 sin6;
        . . .
        recvfrom(s, buf, buflen, flags, (struct sockaddr *) &sin6, &fromlen);
        . . .
        received_flowlabel = sin6.sin6_flowinfo & IPV6_FLOWINFO_FLOWLABEL;
        received_priority = sin6.sin6_flowinfo & IPV6_FLOWINFO_PRIORITY;

On the IPPROTO_IP level.  The only valid
option sending side, applications are responsible for selecting the flow
label value.  The system provides constant declarations for the IPv6
priority values defined in the IPv6 specification [1].  These constants
are:

        IPV6_PRIORITY_UNCHARACTERIZED
        IPV6_PRIORITY_FILLER
        IPV6_PRIORITY_UNATTENDED
        IPV6_PRIORITY_RESERVED1
        IPV6_PRIORITY_BULK
        IPV6_PRIORITY_RESERVED2
        IPV6_PRIORITY_INTERACTIVE
        IPV6_PRIORITY_CONTROL
        IPV6_PRIORITY_8
        IPV6_PRIORITY_9
        IPV6_PRIORITY_10
        IPV6_PRIORITY_11
        IPV6_PRIORITY_12
        IPV6_PRIORITY_13
        IPV6_PRIORITY_14
        IPV6_PRIORITY_15

Applications can use these constants along with the flow label they
selected to assign the sin6_flowinfo field, for example:

        struct sockaddr_in6 sin6;
        . . .
        send_flowlabel = . . . ;
        . . .

        sin6.sin6_flowinfo = IPV6_PRIORITY_UNATTENDED |
                             (IPV6_FLOWINFO_FLOWLABEL & send_flowlabel);

The macro declarations for these constants are obtained by including
the header file <netinet/in.h>.

3.9. Binding to System-Selected Address

While the bind() function allows applications to select the source IP
address of UDP packets and TCP connections, applications often wish to
let the system to select the source address for them.  In IPv4, this
is done by specifying the IPv4 address represented by the symbolic
constant INADDR_ANY in the bind() call, or by simply by skipping the
bind() entirely.

A symbolic constant can not be used for IPv6 because the address is
not a scalar type.  Instead, the system provides a global variable
holding the distinguished IPv6 address that can be used in the bind()
call to instruct the system to select the source IPv6 address.  The
global variable is an in6_addr type structure named "ipv6addr_any."
The extern declaration for this is:

        extern struct in6_addr ipv6addr_any;

Applications use ipv6addr_any similarly to the way they use INADDR_ANY
in IPv4.  For example, to bind a socket to port number 23, but let the
system select the source address, an application could use the
following code:

        struct sockaddr_in6 sin6;
        . . .
        sin6.sin6_family = AF_INET6;
        sin6.sin6_flowinfo = 0;
        sin6.sin6_port = htons(23);
        sin6.sin6_addr = ipv6addr_any;
        . . .
        if (bind(s, (struct sockaddr *) &sin6, sizeof(sin6)) == -1)
                . . .

Note that the value of ipv6addr_any can not be the all-zeros IPv6
address, since that can be used as a valid IPv6 source address in some
cases.

3.10. Communicating with Local Services

Applications may need to send UDP packets to, or originate TCP

connections to, services residing on the local node.  In IPv4, they can
do this by using the constant IPv4 address INADDR_LOOPBACK in their
connect(), sendto(), or sendmsg() call.

For IPv6, the system provides a global variable holding a distinguished
IPv6 address that can be used to contact local TCP and UDP services.
This variable is an in6_addr type structure named "ipv6addr_loopback."
The extern declaration for this variable is:

        extern struct in6_addr ipv6addr_loopback;

Applications use ipv6addr_loopback as they would use INADDR_LOOPBACK
in IPv4 applications.  For example, to open a TCP connection to the
local telnet server, an application could use the following code:

        struct sockaddr_in6 sin6;
        . . .
        sin6.sin6_family = AF_INET6;
        sin6.sin6_flowinfo = 0;
        sin6.sin6_port = htons(23);
        sin6.sin6_addr = ipv6addr_loopback;
        . . .
        if (connect(s, (struct sockaddr *) &sin6, sizeof(sin6)) == -1)
                . . .

4. Socket Options

A number of new socket options are defined for IPv6.  All of these new
options are at the IPPROTO_IPV6 level.  That is, the "level" parameter
in the getsockopt() and setsockopt() call is IPPROTO_IPV6 when using
these options.  The constant name prefix IPV6_ is used in all of the new
socket options.  This serves to clearly identify these options as
applying to IPv6.

The macro declaration for IPPROTO_IPV6, the new IPv6 socket options, and
related constants defined in this section are obtained by including the
header file <netinet/in.h>

4.1  Changing Socket Type

Unix allows open sockets to be passed between processes via the exec()
call and other means.  It is a relatively common application practice to
pass open sockets across exec() calls.  Thus it is possible for an
application using the original API to pass an open PF_INET socket to an
application that is expecting to receive a PF_INET6 socket.  Similarly,
it is possible for an application using the extended API to pass an open
PF_INET6 socket to an application using the original API, which would be

equipped only to deal with PF_INET sockets.  Either of these cases could
cause problems, because the application which is passed the open socket
might not know how to decode the address structures returned in
subsequent socket functions.

To remedy this problem, a new setsockopt() option is defined that allows
an application to "transform" a PF_INET6 socket into a PF_INET socket
and vice-versa.

An IPv6 application that is passed an open socket from an unknown
process may use the IPV6_ADDRFORM setsockopt() option to "convert" the
socket to PF_INET6.  Once that has been done, the system will return
sockaddr_in6 address structures in subsequent socket functions.
Similarly, an IPv6 application that is about to pass an open PF_INET6
socket to a program that may not be IPv6 capable may "downgrade" the
socket to PF_INET before calling exec().  After that, the system will
return sockaddr_in address structures to the application that was
exec()'ed.

The IPV6_ADDRFORM option is at the IPPROTO_IP level.  The only valid
option values are PF_INET6 and PF_INET.  For example, to convert a

PF_INET6 socket
PF_INET6 socket to PF_INET, a program would call:

        int addrform = PF_INET;

        if (setsockopt(s, IPPROTO_IPV6, IPV6_ADDRFORM, (char *) &addrform,
                sizeof(addrform)) == -1)
                perror("setsockopt IPV6_ADDRFORM");

An application may use IPV6_ADDRFORM in the getsockopt() function to
learn whether an open socket is a PF_INET of PF_INET6 socket.  For
example:

        int addrform;
        int len = sizeof(int);

        if (getsockopt(s, IPPROTO_IPV6, IPV6_ADDRFORM, (char *) &addrform,
                &len) == -1)
                perror("getsockopt IPV6_ADDRFORM");
        if (addrform == PF_INET)
                printf("This is an IPv4 socket.\n");
        else if (addrform == PF_INET6)
                printf("This is an IPv6 socket.\n");
        else
                printf("This system is broken.\n");

4.2.  Handling IPv6 Source Routes

IPv6 makes more use of the source routing mechanism than IPv4.  In order
for source routing to operate properly, the node receiving a request
packet that bears a source route must reverse that source route when
sending the reply.  In the case of TCP, the reversal can be done in the
transport protocol implementation transparently to the application.  But
in the case of UDP, the application must perform the reversal itself.
The transport protocol code can not perform the reversal for UDP packets
because a UDP application may receive a number of requests and generate
replies asynchronously.  A "reply" sent by an application may not match
the "request" most recently passed up to the application.

The API for source routing has two components: providing a source route
to be used with originated traffic -- actively opened TCP connections
and UDP packets being sent; and retrieving the source route of received
traffic -- passively accepted TCP connections and received UDP packets.
An application may always provide a source route with TCP connections
being originated and UDP packets being sent.  But to PF_INET, receive source
routes, the application must enable an option.

To provide a program would call:

        int addrform = PF_INET;

        if (setsockopt(s, IPPROTO_IP, IP_ADDRFORM, (char *) &addrform,
                sizeof(addrform)) == -1)
                perror("setsockopt IP_ADDRFORM");

An source route, an application may use IP_ADDRFORM simply provides an array of
sockaddr_in6 data structures in the getsckopt() msg_name field of the msghdr
structure of a sendmsg() function, or the address argument of the
sendto() function to learn
whether an open socket is (when sending a PF_INET UDP packet), or the address argument
of PF_INET6 socket. the connect() function (when actively opening a TCP connection).  For example:

        int addrform;
        int len = sizeof(int);

        if (getsockopt(s, IPPROTO_IP, IP_ADDRFORM, (char *) &addrform,
                &len) == -1)
                perror("getsockopt IP_ADDRFORM");
        if (addrform == PF_INET)
                printf("This is an IPv4 socket.\n");
        else if (addrform == PF_INET6)
                printf("This is an IPv6 socket.\n");
        else
                printf("This system
sendto() and connect(), the length argument of the function is broken.\n");

4.9.  Flow Information

The IPv6 header has a 24-bit the total
length, in octets, of the array.  For sendmsg(), the msg_namelen field
of the msghdr structure specifies the total length of the array.  The
elements of the array represent the full source route, including both
source and destination endpoint address.  The elements of the array are
ordered from destination to hold a "flow label", source.  That is, the first element of the
array represents the destination endpoint address, and the last element
of the array represents the source endpoint address.  If the application
provides a 4-bit source route, the source endpoint address can not be omitted.
The sin6_addr field of the source endpoint address may be set to hold a "priority".  Applications have control over what values
for these fields are used zero,
however, in packets that they originate, and have
access to which case the system will select an appropriate source
address.  The sin6_port field values of packets that they receive. the destination endpoint address must
be assigned.  The sin6_flowinfo sin6_port field of the sockaddr_in6 structure is used source endpoint address may be
set to carry zero, in which case the flow information between system will select an appropriate source
port number.  The sin6_port fields of the intermediate addresses must be
set to zero.

The application and also has control over the system.  An
application may specify a flow label and priority to use loose/strict source routing
flag that is defined in the
transmitted packets of an actively opened TCP connection IPv6 specification [1].  It does this by
setting or clearing the loose/strict flag contained in the sin6_flowinfo
field of the destination address sockaddr_in6 structure
passed and intermediate addresses.  On the receive
side, the implementation uses the loose/strict flag in the connect() function.  An address array
returned to the application may specify to indicate the flow
label and priority loose/strict status of each
hop.

The implementation provides a set of constant definitions to use in transmitted UDP packets by simplify
getting and setting the
sin6_flowinfo field loose/strict flag for each of the destination address sockaddr_in6 structure
passed in hops of a
source route.  The following constant is used to select the sendto() function.  If an application does not care what
values loose/strict
flag from the sin6_flowinfo field:

        IPV6_FLOWINFO_SRFLAG

In addition, two constants are used, it should set provided which represent the flowinfo value two states
of this flag:

        IPV6_SRFLAG_STRICT
        IPV6_SRFLAG_LOOSE

These constants can be used to inspect the source route flags of
received addresses, for example:

        struct sockaddr_in6 sin6[3];
        . . .
        if ((sin6[0].sin6_flowinfo & IPV6_FLOWINFO_SRFLAG) ==
            IPV6_SRFLAG_STRICT)
            . . .

And they can also be used to zero.

An application may specify set the source route flags:

        struct sockaddr_in6 sin6[3];
        . . .
        sin6[0].sin6_flowinfo =
                (sin6[0].sin6_flowinfo & ~IPV6_FLOWINFO_SRFLAG) |
                IPV6_SRFLAG_STRICT;

The flow label and priority to use in
transmitted packets sub-fields of a passively accepted TCP connection, by setting the sin6_flowinfo field of the
destination endpoint address passed in may be set, but the bind() function.

The flow label and priority that appeared in received UDP packets are
passed up these fields must be
set to the application zero in the sin6_flowinfo field of the intermediate and source

address sockaddr_in6 structure that is returned in the recvfrom() call. endpoint address.

The flow information that appeared in the received SYN segment arrangement of a
passively accepted TCP connection is returned to the application address structures in the
source address sin6_flowinfo field of the sockaddr_in6 structure that is buffer passed
to sendmsg(), connect() or sendto() is shown in the accept() call.

4.10.  Handling IPv6 figure below:

        +--------------------+
        |                    |
        |  sockaddr_in6[0]   |  Destination Endpoint Address
        |                    |
        +--------------------+
        |                    |
        |  sockaddr_in6[1]   |  Last Source-Route Hop Address
        |                    |
        +--------------------+
        .                    .
        .                    .
        .                    .
        +--------------------+
        |                    |
        | sockaddr_in6[N-1]  |  First Source-Route Hop Address
        |                    |
        +--------------------+
        |                    |
        |  sockaddr_in6[N]   |  Source Routes

IPv6 makes more use of the source routing mechanism than IPv4.  In order
for source routing to operate properly, the node receiving a request
packet that bears Endpoint Address
        |                    |
        +--------------------+

               Address buffer when sending a source route must reverse that source route when
sending the reply.  In the case of TCP, the reversal can be done in the
transport protocol implementation transparently to the application.  But
in

The IPV6_RECVSRCRT setsockopt() option controls the case reception of UDP, the application source
routes.  The option is disabled by default.  Applications must perform
explicitly enable the reversal itself.
The transport protocol code can not perform option using the reversal for UDP packets
because a UDP application may setsockopt() function in order to
receive a number source routes.

The IPV6_RECVSRCRT option is at the IPPROTO_IPV6 level.  An example of requests and generate
replies asynchronously.  A "reply" sent by
how an application may not match might use this option is:

        int on = 1;             /* value == 1 means enable the "request" most recently passed up to option */

        if (setsockopt(s, IPPROTO_IPV6, IPV6_RECVSRCRT, (char *) &on,
                sizeof(on)) == -1)
                perror("setsockopt IPV6_RECVSRCRT");

When the application.

The API for source routing has two components: providing IPV6_RECVSRCRT option is disabled, only a source route single sockaddr_in6
address structure is returned to be used with originated traffic -- actively opened TCP connections
and UDP packets being sent -- applications in the address argument of
the recvfrom() and retrieving accept() functions.  This address represents the
source route endpoint address of
received traffic -- passively accepted TCP connections and received the UDP
packets.  An application may always provide a source route with packet received or the TCP
connections being originated and UDP packets being sent.  But to receive
source routes, connection
accepted.

When the application must enable an option.

To provide a source route, an application simply provides an array IPV6_RECVSRCRT option is enabled, the msg_name field of
sockaddr_in6 data structures in the
msghdr of the recvmsg() function, or the address argument of the sendto()
recvfrom() function (when sending a receiving UDP packet), or packets) and the connect() function accept()
functions (when
actively opening a passively accepting TCP connection).  The length argument of the function
is the total length, in octets, of the array.  The elements connections) points to an array

of sockaddr_in6 structures.  When the array
represent function returns, the full source route, including both array will
hold two elements -- source and destination
endpoint address. address -- when the received
UDP packet or TCP SYN packet does not carry a source route.  The array
will hold more than two elements of when the received packet carries a
source route.

The addresses in the array are ordered from
destination source to source. destination.  That
is, the first element of the array
represents the destination holds source endpoint address, and address of the
received packet.  Following this in the array are the intermediate hops
in the order in which they were visited.  The last element of the array represents
holds the source destination endpoint address.  If  Note that this is the application
provides a source route, opposite
of the order specified for sending.  This ordering was chosen so that
the source endpoint address array returned in a recvmsg() or recvfrom() call can not be omitted.
The sin6_addr field of used
in a subsequent sendmsg() or sendto() call without requiring the source endpoint address may be set
application to zero,
however, re-order the addresses in which case the system will select array.  Similarly, the
address array received in an appropriate source
address. accept() call can be used unchanged in a
subsequent connect() call.

The sin6_port address length argument of the recvfrom() and accept() functions,
and the msg_namelen field of the destination endpoint msghdr field in the recvmsg() function,
indicate the length, in octets, of the full address must
be assigned. array.  This
argument is a value-result parameter.  The sin_port field application sets the maximum
size of the source endpoint address may be
set buffer when it makes the call, and the system
modifies the value to zero, in which case return the actual size of the buffer to the
application.

The sin6_port field of the system first and last array elements (source and
destination endpoint address) will select an appropriate hold the source and destination UDP
or TCP port number. number of the received packet.  The sin6_port fields field of the
intermediate
addresses must elements of the array will be set to zero.

The flow label and priority sub-fields of the sin6_flowinfo field of the
destination
source endpoint address may be set, but will hold the these fields must be

set to zero in flow label and priority values of
the received packet.  The flow label and priority sub-fields of the
intermediate addresses and source the destination endpoint address. address will be
zero.  The loose/strict flag of the sin6_flowid sin6_flowinfo field of the destination source
endpoint address and the intermediate addresses may will be set according to 0 or 1.  If the
flag is set to one,
the leg of flags in the end-to-end path TO that address
should received packet.  The macros defined above can be treated as a strict source route.  If used
to inspect the flag is 0, that leg
should be treated as a loose source route.  The loose/strict flag of the
source endpoint each hop.

The address must be set buffer returned to 0.

The arrangement of the address structures application in the address buffer passed
to connect() recvfrom() or sendto()
accept() functions when the IPV6_RECVSRCRT option is enabled is shown in the figure
below:

        +--------------------+
        |                    |
        |  sockaddr_in6[0]   |  Destination  Source Endpoint Address
        |                    |
        +--------------------+
        |                    |
        |  sockaddr_in6[1]   |  Last  First Source-Route Hop Address
        |                    |
        +--------------------+
        .                    .
        .                    .
        .                    .
        +--------------------+
        |                    |
        | sockaddr_in6[N-1]  |  First  Last Source-Route Hop Address
        |                    |
        +--------------------+
        |                    |
        |  sockaddr_in6[N]   |  Source  Destination Endpoint Address
        |                    |
        +--------------------+

              Address buffer when sending receiving a source route

The IP_RCVSRCRT setsockopt() option controls the reception of

IPv6 allows a source
routes.  The option is disabled by default.  Applications route with up to 23 intermediate hops.  Since the
it must
explicitly enable also receive the option using source and destination endpoint addresses, the setsockopt() function in order
application must provide a buffer capable of holding 25 addresses to
receive such a source routes.

The macro definition for IP_RCVSRCRT is route.  Implementations provide the following
constant declaration in <netinet/in.h>.

The IP_RCVSRCRT option is at order to allow applications to simply declare
storage for the IPPROTO_IP level.  An example of how an
application might largest possible source route:

        IPV6_SR_MAXADDR

Applications can use this option is:

        int on = 1;             /* value == 1 means enable the option */

        if (setsockopt(s, IPPROTO_IP, IP_RCVSRCRT, (char *) &on,
                sizeof(on)) == -1)
                perror("setsockopt IP_RCVSRCRT");

When constant like this:

        struct sockaddr_in6 sin6[IPV6_SR_MAXADDR];

It may be impractical for some applications to allocate space to hold
the IP_RCVSRCRT option is disabled, only largest possible source route.  Thus a single sockaddr_in6
address structure is returned received source route may be
too large to applications in fit into the address argument
of buffer provided by the recvfrom() and accept() functions.  This application.  In this
circumstance, the system should return only a single address represents element --
the source endpoint address of the UDP packet received or the TCP
connection accepted.

When -- to the IP_RCVSRCRT option application.  This case is enabled, the address argument of the
recvfrom() function (when receiving UDP packets) and the accept()
functions (when passively accepting TCP connections) points clearly
distinguishable to an array
of sockaddr_in6 structures.  When the function returns, application because in all other cases, the array will
hold
system returns at least two address elements -- the source and
destination address -- when the received endpoint addresses.

4.3.  Receiving Interface Determination

Some applications run on multi-homed hosts need to determine which
interface UDP packet packets were received on or TCP SYN packet connections are bound to.
While the source routing interface described in the previous section
returns the destination address of the packet, this does not carry a source route.  The array
will hold more than two elements when necessarily
identify the receiving interface.  Some cases where it does not are:

   -    When the received packet carries a
source route. is multicast.  The addresses destination address
        in this case is an IPv6 multicast address, not the array are ordered from source to destination.  That
is, the first element of the array holds source endpoint address of an
        interface.

   -    When the
received packet.  Following this in node is operating as an IPv6 router.  The node may
        receive packets on interfaces other than the array one they are
        addressed to.

   -    When the intermediate hops
in received packet is sent to the order in node's link-local
        address which they were visited. is being used on multiple interfaces.

The last element address of the array
holds receiving interface is returned to the destination endpoint address.  Note application
similarly to the way that source routes are returned.  A new
setsockopt() option named IPV6_RECVIF is provided at the IPPROTO_IPV6
level.  If this option is enabled, the opposite
of system returns an additional
sockaddr_in6 structure to the order specified for sending.  This ordering was chosen so that application, holding the IPv6 address array received of
the receiving interface, in a recvfrom() call the recvfrom(), recvmsg(), or accept()
functions.

The option is enabled like this:

        int on = 1;             /* value == 1 means enable the option */

        if (setsockopt(s, IPPROTO_IPV6, IPV6_RECVIF, (char *) &on,
                sizeof(on)) == -1)
                perror("setsockopt IPV6_RECVIF");

This option can be used in a
subsequent sendto() call without requiring conjunction with the application to re-order IP_RECVSRCRT option.
When the addresses in IPV6_RECVIF option is enabled, the array.  Similarly, buffer returned to the
application is structured like this:

        +--------------------+  - - - - - - - - - - - - - - -
        |                    |
        |  sockaddr_in6[0]   |
        |                    |
        +--------------------+
        .                    .  Source Address, or
        .                    .  Full Source Route
        .                    .
        +--------------------+
        |                    |
        | sockaddr_in6[N-1]  |
        |                    |
        +--------------------+  - - - - - - - - - - - - - - -
        |                    |
        |  sockaddr_in6[N]   |  Receiving Interface Address
        |                    |
        +--------------------+  - - - - - - - - - - - - - - -

              Address buffer with receiving interface address array received in an
accept() call can be used unchanged in a subsequent connect() call.

The last address length argument of the recvfrom() and accept() functions
indicate the length, in octets, of the full address array.  This
argument array is a value-result parameter.  The application sets the maximum
size of the an IPv6 address buffer when it makes of the call, receiving
interface.  Since interfaces in IPv6 may have more than one address, and
some addresses (e.g. link-local addresses) may be used on more than one
interface, the system
modifies the value to return the actual size of should select an address that uniquely identifies
the buffer interface.

4.4.  Sending Interface Specification

Applications may also need to specify the
application.

The sin6_port field of the first and last array elements (source and
destination endpoint address) will hold the source and destination outgoing interface that
originated UDP or TCP port number of the received packet.  The sin6_port field of the
intermediate elements of the array will be zero.

The flow label and priority sub-fields of the sin6_flowinfo field of the packets should use.  This is accomplished like
source endpoint address will hold the flow label and priority values of
the received packet. route selection.  The flow label and priority sub-fields of the
intermediate addresses and application may provide an additional
sockaddr_in6 structure in its sendto(), sendmsg() or connect() call
specifying the destination endpoint address will be
zero.  The loose/strict flag of the sin6_flowinfo field of the outgoing interface.  Unlike source
endpoint address and route
selection, the intermediate addresses will outgoing interface address can only be 1 included if a new
option is enabled.  The new option is needed so that the leg of system can
differentiate between the end-to-end path originating FROM that application's specification of an outgoing
interface address was strict. and a source route.

The

loose/strict flag of new option option is named IPV6_SENDIF and is at the destination endpoint address will IPPROTO_IPV6
level.  It can be 0.

The address buffer returned to enabled like this:

        int on = 1;             /* value == 1 means enable the application option */

        if (setsockopt(s, IPPROTO_IPV6, IPV6_SENDIF, (char *) &on,
                sizeof(on)) == -1)
                perror("setsockopt IPV6_SENDIF");

This option can be used in the recvfrom() or
accept() functions when the IP_RCVSRCRT conjunction with source route specification.
If this option is enabled is shown
below: enabled, the application passes in an address array
structured as follows:

        +--------------------+  - - - - - - - - - - - - - - -
        |                    |
        |  sockaddr_in6[0]   |  Source Endpoint Address
        |                    |
        +--------------------+
        |                    |
        |  sockaddr_in6[1]   |  First Source-Route Hop Address
        |                    |
        +--------------------+
        .                    .  Destination Address, or
        .                    .  Full Source Route
        .                    .
        +--------------------+
        |                    |
        | sockaddr_in6[N-1]  |  Last Source-Route Hop Address
        |                    |
        +--------------------+  - - - - - - - - - - - - - - -
        |                    |
        |  sockaddr_in6[N]   |  Destination Endpoint  Sending Interface Address
        |                    |
        +--------------------+

              Address buffer when receiving a source route

Since IPv6 allows the number of elements in a source route to be very
large, it is impractical for all applications that have enabled the
reception of source routes to provide buffer space to hold the maximum
number of elements.  Some applications may choose a buffer size that is
appropriate for their own use.  This means that it is possible that a
received source route may be too large to fit into the buffer provided
by the application.  In this circumstance, the system should return only
a single  - - - - - - - - - - - - - - -

              Address buffer with sending interface address element -- the source endpoint

The last address -- to in the
application.  This case array is clearly distinguishable to the application
because in all other cases, an IPv6 address of the system returns at least two sending
interface.  Applications should use an address
elements -- that uniquely identifies
the source and destination endpoint addresses.

4.11. interface to use.

4.5.  Unicast Hop Limit

A new setsockopt() option is used to control the hop limit used in
outgoing unicast IPv6 packets.  The name of this option is
IP_UNICAST_HOPS,
IPV6_UNICAST_HOPS, and it is used at the IPPROTO_IP IPPROTO_IPV6 layer.  The macro
definition for IP_UNICAST_HOPS resides in the <netinet/in.h> header

file.  The
following example illustrates how it is used:

        int hoplimit = 10;

        if (setsockopt(s, IPPROTO_IP, IP_UNICAST_HOPS, IPPROTO_IPV6, IPV6_UNICAST_HOPS, (char *) &hoplimit,
                sizeof(hoplimit)) == -1)
                perror("setsockopt IP_UNICAST_HOPS); IPV6_UNICAST_HOPS");

When the IP_UNICAST_HOPS IPV6_UNICAST_HOPS option is set with setsockopt(), the option
value given is used as the hop limit for all subsequent unicast packets
sent via that socket.  If the option is not set, the system selects a
default value.

The IP_UNICAST_HOPS IPV6_UNICAST_HOPS option may be used in the getsockopt() function to

determine the hop limit value that the system will use for subsequent
unicast packets sent via that socket.  For example:

        int hoplimit;
        int len = sizeof(hoplimit);

        if (getsockopt(s, IPPROTO_IP, IP_UNICAST_HOPS, IPPROTO_IPV6, IPV6_UNICAST_HOPS, (char *) &hoplimit,
                &len) == -1)
                perror("getsockopt IP_UNICAST_HOPS); IPV6_UNICAST_HOPS");
        else
                printf("Using %d for hop limit.\n", hoplimit);

4.12.

4.6.  Sending and Receiving Multicast Packets

IPv6 applications may send UDP multicast packets by simply specifying an
IPv6 multicast address in the address argument of the sendto() function.

A few setsockopt options at the IPPROTO_IP IPPROTO_IPV6 layer are used to control
some of the parameters of sending multicast packets.  These options are
optional: applications may send multicast packets without using these
options.  The setsockopt() options for controlling the sending of
multicast packets are summarized below:

        IP_MULTICAST_IF

        IPV6_MULTICAST_IF

                Set the interface to use for outgoing multicast packets.

        IP_MULTICAST_HOPS
                The argument is an IPv6 address of the interface to use.

                Argument type: struct in6_addr

        IPV6_MULTICAST_HOPS

                Set the hop limit to use for outgoing multicast packets.
                (Note a separate option - IP_UNICAST_HOPS IPV6_UNICAST_HOPS - is
                provided to set the hop limit to use for outgoing
                unicast packets.)

        IP_MULTICAST_LOOP

                Argument type: unsigned int

        IPV6_MULTICAST_LOOP

                Controls whether outgoing multicast packets sent should
                be delivered back to the local application.  A toggle.
                If the option is set to 1, multicast packets are looped
                back.  If it is set to 0, they are not.

                Argument type: unsigned int

The reception of multicast packets is controlled by the two setsockopt()
options summarized below:

        IP_ADD_MEMBERSHIP

        IPV6_ADD_MEMBERSHIP

                Join a multicast group.  Requests that multicast packets
                sent to a particular multicast address be delivered to
                this socket.

        IP_DROP_MEMBERSHIP  The argument is the IPv6 multicast address
                of the group to join.

                Argument type: struct in6_addr

        IPV6_DROP_MEMBERSHIP

                Leave a multicast group.  Requests that multicast
                packets sent to a particular multicast address no longer
                be delivered to this socket.

4.13.  The argument is the IPv6
                multicast address of the group to join.

                Argument type: struct in6_addr

5. Library Functions

New library functions are needed to lookup IPv6 addresses in the name
service, and to manipulate IPv6 addresses.

5.1.  Name-to-Address Translation Functions

We have defined two

Two new functions analogous to gethostbyname() and gethostbyaddr() have
been defined which support addresses in both IPv4 and IPv6 addresses.  The names of
the new functions are hostname2addr() and addr2hostname().  These
functions were designed to have semantics similar to gethostbyname() and
gethostbyaddr(), so that existing IPv4 applications can be easily ported
to IPv6.

The new functions differ from the old in one important way that is not
related to IPv6: The old functions could not safely be used by
multi-threaded applications, while the new ones can.  There are two
multi-threading problems with the old functions.  First, the return
value of the old functions is a pointer to a single static buffer
belonging to the library.  The new functions return a dynamically
allocated buffer, and a third new function, named freehostent(), is
provided to free that storage.  Second, the IPv4 and IPv6
address families. old functions returned their
error code in a global variable (h_errno).  The names of the new functions are hostname2addr()
and addr2hostname().  These functions were designed to have semantics
similar to gethostbyname() and gethostbyaddr(), so carry a
pointer that existing IPv4
applications can be easily ported allows the library to IPv6.

Hostname2addr() return the error code into storage
provided by the caller.

The hostname2addr() function is defined similarly similar to gethostbyname(), but enables
applications to specify the type of address to be looked up:

        struct hostent *hostname2addr(const *hostname2addr(
                const char *name,
                int af); af,
                int *error);

This new function looks up the given hostname argument name in the name service and
returns
and, if the lookup succeeds, returns a completed hostent structure if structure.  If
the lookup succeeds, and fails, the function returns NULL
otherwise.  The name argument and an error code is
returned in the domain name of the host buffer pointed to look up. by the argument error. The af argument
specifies the type of the address -- IPv4 (AF_INET) or IPv6 (AF_INET6)
-- to return to the caller in the h_addr_list field of the hostent
structure.

If the af argument is AF_INET, hostname2addr() behaves much like
gethostbyname.  It queries the name service for IPv4 addresses and, if
any are found, returns a hostent structure that includes an array of
IPv4 addresses.  Each IPv4 address is encoded in network byte order.

If the af argument is AF_INET6, the processing is as follows: the hostname2addr() function first queries the name service
for IPv6 addresses. If IPv6 addresses are found, they are returned in an array in
the hostent structure.  If no IPv6 addresses are found, the  The function
queries may also query the name service for
IPv4 addresses. records.  If this is done, any IPv4 addresses are
found, they found are returned to

the application encoded as IPv4-mapped IPv4-compatible IPv6 addresses.  As in IPv4,
each  The
determination of whether to query for IPv4 addresses is system specific.
Systems that support querying for IPv4 addresses should provide a
system-wide configuration switch allowing the system administrator to
enable or disable that feature.

IPv6 address addresses returned in by the hostent structure is hostname2addr() function are encoded in
network byte order.

The second new function, called addr2hostname(), is defined in exactly
the same way as like the
gethostbyaddr() function, except that it now but supports both the IPv4 and IPv6 address
families:

        struct hostent *addr2hostname(const *addr2hostname(
                const void *addr,
                int len, addrlen,
                int af,
                int af); *error);

The addr2hostname() function performs an address-to-name lookup on the
address
specified, specified by the addr argument, returning a completed hostent
structure if the lookup
succeeds, or NULL, if succeeds.  If the lookup fails. fails, the function
returns NULL and an error code is returned in the buffer pointed to by
the argument error.

The addrlen argument specifies the length of the address (in octets)
pointed to by the addr argument.

The af argument specifies the address family of the addr argument.  This
function supports both the AF_INET and AF_INET6 address families.  If
the af argument is AF_INET, then len must be specified to be 4-octets and addr must refer refers to an IPv4
address. address and
addrlen must have the value 4.  If af is AF_INET6, then len must be specified as 16-octets and addr must refer to represents an
IPv6 address.  If address and addrlen must have the addr argument is value 16.  In the latter case,
the caller may present an IPv4-mapped IPv6 address, address in the addr argument.
If this is done, an IPv4 address-to-name lookup is performed on the
embedded IPv4 address.

A

The hostent structure returned by both of these functions is allocated
by the library.  Applications use the freehostent() function to return
the hostent structure to the library after they are done using it:

        void freehostent(
                struct hostent *hp);

Applications may not access the hostent structure after they have
returned it to the library.

Another new name-to-address translation library function is now under

development at Berkeley.  This new function, named getconninfo(), will
subsume the functionality of gethostbyname(), hostname2addr(), as well
as the getservbyname() and getservbyport() functions.  The new function
is specifically designed to be "transport independent", so it should be
directly usable by IPv6 applications.

System implementations should provide the addr2hostname() and
hostname2addr() functions in order to simplify the porting of existing
IPv4 applications to IPv6.  System implementations may also provide the
getconninfo() function, once it is defined, so that newly written
applications can be transport independent.

Specification

The specification of the getconninfo() function is published as a
separate
specification document [2], not included in this spec.

Implementations must retain the BSD gethostbyname() and gethostbyaddr()
functions in order to provide source and binary compatibility for
existing applications.

4.14.

Applications obtain the function prototype declarations for
hostname2addr() and addr2hostname() by including the header file
<netdb.h>.

5.3.  Address Conversion Functions

BSD Unix provides two functions, inet_addr() and inet_ntoa(), to convert
an IPv4 address address between binary and printable form.  IPv6 applications
need similar functions.  The following two functions convert both IPv6
and IPv4 addresses:

        int ascii2addr(
                int af,
                const char *cp,
                void *ap);

and

        char *addr2ascii(
                int af,
                const void *ap,
                int len,
                char *cp);

The first function converts an ascii string to an address in the address
family specified by the af argument.  Currently AF_INET and AF_INET6
address families are supported.  The cp argument points to the ascii
string being passed in.  The ap argument points to a buffer into which

the function stores the address.  Ascii2addr() returns the length of the
address in octets if the conversion succeeds, and -1 otherwise. The
function does not modify the storage pointed to by ap if the conversion
fails. The application must ensure that the buffer referred to by ap is
large enough to hold the converted address.

If the af argument is AF_INET, the function accepts a string in the
standard IPv4 dotted decimal form:

        ddd.ddd.ddd.ddd

where ddd is a one to three digit decimal number between binary 0 and printable form. 255.

If the af argument is AF_INET6, then the function accepts a string in
one of the standard IPv6 applications
need similar functions.  We have printing forms defined in the following two functions to
convert both IPv6 and IPv4 addresses:

        int ascii2addr(int af, const char *cp, void *ap);

and

        char *addr2ascii(int af, const void *ap, int len, char *cp); addressing
architecture specification [3].

The first second function converts an ascii string address into a printable string.  The af
argument specifies the form of the address.  This can be AF_INET or
AF_INET6.  The ap argument points to a buffer holding an IPv4 address in if
the af argument is AF_INET, and an IPv6 address
family specified by if the af argument.  Currently AF_INET and AF_INET6 argument is
AF_INET6.  The len field specifies the length in octets of the address families are supported.
pointed to by ap, and must be 4 if af is AF_INET, or 16 if af is
AF_INET6.  The cp argument points to a buffer that the function can use
to store the ascii
string being passed in.  The ap string.  If the cp argument points is NULL, the function
uses its own private static buffer.  If the application specifies a cp
argument, it must be large enough to hold the ascii conversion of the
address specified as an argument, including the terminating null octet.
For IPv6 addresses, the buffer must be at least 46-octets.  For IPv4
addresses, the buffer must be at least 16-octets.  In order to a buffer into which allow
applications to easily declare buffers of the function stores proper size to store IPv4
and IPv6 addresses in string form, implementations should provide the address.  Ascii2addr()
following constants, made available to applications that include
<netinet/in.h>:

        #define INET_ADDRSTRLEN         16
        #define INET6_ADDRSTRLEN        46

The addr2ascii() function returns a pointer to the length of buffer containing the
address in octets
ascii string if the conversion succeeds, and -1 NULL otherwise.  The
function does not modify the storage pointed to by ap cp if the conversion
fails. The application must ensure that

Applications obtain the buffer referred to prototype declarations for addr2ascii() and
ascii2addr() by ap is
large enough to hold the converted address.

If the af argument is AF_INET, the function accepts a string in including the
standard header file <arpa/inet.h>.

5.3. Embedded IPv4 dotted decimal form:

        ddd.ddd.ddd.ddd

where ddd Addresses

The IPv4-mapped IPv6 address format is a one used to three digit decimal number between 0 and 255.

If the af argument is AF_INET6, then the function accepts represent IPv4 addresses
as IPv6 addresses.  Most applications should be able to to manipulate
IPv6 addresses as opaque 16-bit quantities, without needing to know
whether they represent IPv4 addresses.  However, a string in
one of the standard few applications may
need to determine whether an IPv6 printing forms defined in the addressing
architecture specification [3].

The second function converts address is an IPv4-mapped address into a printable string.  The af
argument specifies the form of the address.  This can be AF_INET or
AF_INET6.
not.  The ap following function is provided for those applications:

        int inet6_isipv4addr (const struct in6_addr *addr);

The "addr" argument to this function points to a buffer holding an IPv4 address if
the af argument is AF_INET, and an IPv6
address if the af argument is
AF_INET6.  The len field specifies the length in octets of the network byte order.  The function returns true (non-zero) if
that address
pointed to by ap, is an IPv4-mapped address, and must returns 0 otherwise.

This function could be 4 if af used by server applications to determine whether
the peer is AF_INET, an IPv4 node or 16 if af is
AF_INET6.  The cp argument points to an IPv6 node.  After accepting a buffer that TCP
connection via accept(), or receiving a UDP packet via recvfrom(), the function
application can use
to store apply the ascii string.  If inet6_isipv4addr() function to the cp argument is NULL, returned
address.

Applications obtain the prototype for this function
uses its own private static buffer.  If by including the application specifies
header file <arpa/inet.h>.

6.  Security Considerations

IPv6 provides a cp
argument, it must number of new security mechanisms, many of which need to
be large enough accessible to hold the ascii conversion of applications.  A companion document detailing the
address specified as an argument, including
extensions to the terminating null octet.
For socket interfaces to support IPv6 addresses, the buffer must be at least 46-octets.  For IPv4
addresses, security is being
written [4].  At some point in the buffer must future, that document and this one
may be at least 16-octets.

The addr2ascii() function returns merged into a pointer to single API specification.

7. Change History

Changes from the July 1995 Edition

   -    Changed u_long and u_short types in structures to u_int32_t and
        u_int16_t for consistency and clarity.

   -    Added implementation-provided constants for IPv4 and IPv6 text
        address buffer containing the
ascii string if the conversion succeeds, length.

   -    Defined a set of constants for subfields of sin6_flowid and NULL otherwise.  The
function does not modify for
        priority values.

   -    Defined constants for getting and setting the storage pointed to by cp if source route flag.

   -    Define where ansi prototypes for hostname2addr(),
        addr2hostname(), addr2ascii(), ascii2addr(), and
        ipv6_isipv4addr() reside.

   -    Clarified the conversion
fails.

5.  Security Considerations

IPv6 provides include file requirements.  Say that the structure
        definitions are defined as a number result of new security mechanisms, many of which need including the header file
        <netinet/in.h>, not that the structures are necessarily defined
        there.

   -    Removed underscore chars from is_ipv4_addr() function name for
        BSD compatibility.

   -    Added inet6_ prefix to
be accessible is_ipv4_addr() function name to applications.  A companion avoid
        name space conflicts.

   -    Changes setsockopt option naming convention to use IPV6_ prefix
        instead of IP_ so that there is clearly no ambiguity with IPv4
        options.  Also, use level IPPROTO_IPV6 for these options.

   -    Made hostname2addr() and addr2hostname() functions thread-safe.

   -    Added support for sendmsg() and recvmsg() in source routing
        section.

   -    Changed in_addr6 to in6_addr for consistency.

   -    Re-structured document detailing into sub-sections.

   -    Deleted the
extensions implementation experience section.  It was too
        wordy.

   -    Added argument types to the multicast socket interfaces to support IPv6 security is being
written [4].  At some point in options.

   -    Added constant for largest source route array buffer.

   -    Added the future, that document freehostent() function.

   -    Added receving interface determination and this one
may be merged into a single API specification.

6. Change History sending interface
        selection options.

   -    Added definitions of ipv6addr_any and ipv6addr_loopback.

   -    Added text making the lookup of IPv4 addresses by
        hostname2addr() optional.

Changes from the June 1995 Edition

   -    Added capability for application to select loose or strict
        source routing.

Changes from the March 1995 Edition

   -    Changed the definition of the ipv6_addr structure to be an array
        of sixteen chars instead of four longs.  This change is
        necessary to support machines which implement the socket
        interface, but do not have a 32-bit addressable word.  Virtually
        all machines which provide the socket interface do support an
        8-bit addressable data type.

   -    Added a more detailed explanation that the data types defined in
        this documented are not intended to be hard and fast
        requirements.  Systems may use other data types if they wish.

   -    Added a note flagging the fact that the sockaddr_in6 structure
        is not the same size as the sockaddr structure.

   -    Changed the sin6_flowlabel field to sin6_flowinfo to accommodate
        the addition of the priority field to the IPv6 header.

Changes from the October 1994 Edition

   -    Added variant of sockaddr_in6 for 4.4 BSD-based systems (sa_len
        compatibility).

   -    Removed references to SIT transition specification, and added
        reference to addressing architecture document, for definition of
        IPv4-mapped addresses.

   -    Added a solution to the problem of the application not providing
        enough buffer space to hold a received source route.

   -    Moved discussion of IPv4 applications interoperating with IPv6
        nodes to open issues section.

   -    Added length parameter to addr2ascii() function to be consistent
        with addr2hostname().

   -    Changed IP_MULTICAST_TTL to IP_MULTICAST_HOPS to match IPv6
        terminology, and added IP_UNICAST_HOPS option to match
        IP_MULTICAST_HOPS.

   -    Removed specification of numeric values for AF_INET6,
        IP_ADDRFORM, and IP_RCVSRCRT, since they need not be the same on
        different implementations.

   -    Added a definition for the in_addr6 IPv6 address data
        structure.  Added this so that applications could use
        sizeof(struct in_addr6) to get the size of an IPv6 address,
        and so that a structured type could be used in the
        is_ipv4_addr().

7.

8. Open Issues

A few open issues for IPv6 socket interface API specification remain,
including:

   -    The multicast API needs to be documented in more detail.

   -    Should we add a timeout parameter to hostname2addr() and
        addr2hostname()?  DNS lookups need to be given some finite
        timeout interval, so it might be nice to let the application
        specify that interval.

   -    Can the IPV6_ADDRFORM option really be implemented?

   -    An interface is needed to allocate flow labels.  Should one be
        defined in this spec, or left for another document?

   -    Can existing IPv4 applications interoperate with IPv6 nodes?

7.1.
        This issue is discussed in more detail in the following section.

8.1. IPv4 Applications Interoperating with IPv6 Nodes

This problem primarily has to do with the how IPv4 applications
represent addresses of IPv6 nodes.  What address should be returned to
the application when an IPv6/UDP packet is received, or an IPv6/TCP
connection is accepted?  The peer's address could be any arbitrary
128-bit IPv6 address.  But the application is only equipped to deal with
32-bit IPv4 addresses encoded in sockaddr_in data structures.

We have not discovered any solution that provides complete transparent
interoperability with IPv6 nodes for applications using the original
IPv4 API.  However, two techniques that partially solve the problem are:

   1)   Prohibit communication between IPv4 applications and IPv6 nodes.
        Only UDP packets received from IPv4 nodes would be passed up to
        the application, and only TCP connections received from IPv4
        nodes would be accepted.  UDP packets from IPv6 nodes would be
        dropped, and TCP connections from IPv6 nodes would be refused.

   2)   The system could generate a local 32-bit cookie to represent the
        full 128-bit IPv6 address, and pass this value to the
        application.  The system would maintain a mapping from cookie
        value into the 128-bit IPv6 address that it represents.  When
        the application passed a cookie back into the system (for
        example, in a sendto() or connect() call) the system would use
        the 128-bit IPv6 address that the cookie represents.

        The cookie would have to be chosen so as to be an invalid IPv4
        address (e.g. an address on net 127.0.0.0), and the system would
        have to make sure that these cookie values did not escape into
        the Internet as the source or destination addresses of IPv4
        packets.

Both of these techniques have drawbacks.  This is an area for further
study.  System implementors may use one of these techniques or implement
another solution.

Acknowledgments

Thanks to the many people who made suggestions and provided feedback to
to the numerous revisions of this document, including: Marc Hasson, Ran Atkinson,
Fred Baker, Dave Borman, Andrew Cherenson, Alex Conta, Alan Cox, Steve
Deering, Francis Dupont, Robert Elz, Marc Hasson, Tom Herbert, Christian
Huitema, Wan-Yen Hsu, Alex Conta, Richard Stevens, Dan
McDonald, Alan Lloyd, Christian Huitema, Steve Deering, Andrew
Cherenson, Charles Lynn, Ran Atkinson, Dan McDonald, Erik
Nordmark, Josh Osborne,
Glenn Trewitt, Fred Baker, Robert Elz, Richard Stevens, Dean D.  Throop, Glenn Trewitt,
and Francis
Dupont. Carl Williams.  Craig Partridge suggested the addr2ascii() and
ascii2addr() functions.

Ramesh Govindan made a number of contributions and co-authored an
earlier version of this paper.

References

  [1]   R. Hinden. "Internet Protocol, Version 6 (IPv6) Specification".
        Internet Draft.  June 1995.

  [2]   Keith Sklower. "Getconninfo(): An alternative to Gethostbyname()"
        Internet Draft.  June 1995.

  [3]   R. Hinden., S. Deering. "IP Version 6 Addressing Architecture".
        Internet Draft. June 1995.

  [4]   D. McDonald. "IPv6 Security API for BSD Sockets".  Internet
        Draft. January 1995.

Authors' Address

        Jim Bound
        Digital Equipment Corporation
        110 Spitbrook Road ZK3-3/U14
        Nashua, NH 03062-2698
        Phone: +1 603 881 0400
        Email: bound@zk3.dec.com
        Susan Thomson
        Bell Communications Research
        MRE 2P-343, 445 South Street
        Morristown, NJ 07960
        Telephone: +1 201 829 4514
        Email: set@thumper.bellcore.com

        Robert E. Gilligan
        Mailstop MPK 17-202
        Sun Microsystems, Inc.
        2550 Garcia Avenue
        Mailstop UMTV05-44
        Mountain View, CA 94043-1100
        Phone: +1 415 336 1012 786 5151
        Email: bob.gilligan@eng.sun.com gilligan@eng.sun.com