Programming with sockets

Moving to IPv6

The version 6 Internet Protocol (IPv6) provides all the functionality of its predecessor (IPv4), but includes a much larger network address space. This was necessary for a number of reasons:

to increase the number of available IP addresses
to streamline and optimize routing the functionality of the Internet
to simplify the packet header processing
to provide improved option support

Changing the address structure has a major impact on all aspects of IP inter-networking; the transition from IPv4 to IPv6 has to be done over period of time.

In UnixWare, the first stage of this transition is to provide a set of functions that enable you to develop ``IPv6-ready'' applications, while still using an IPv4-based network. These applications can function in both IPv4 and IPv6 environments. When you migrate to an IPv6 network, the new ``IPv6-ready'' applications will be able to continue working without modification. The impact of the UnixWare IPv6 implementation is described in the following topics:

``Internet Protocol version 6 (IPv6)'' describes the IPv6 protocol including its various address formats, and explains how IPv4-mapped IPv6 addresses allow ``IPv6-ready'' applications to work in an IPv4 environment.
``The IPv6 sockaddr structure'' describes the new data structures required to hold the larger IPv6 address.
``Address and protocol families'' lists the new address and protocol family constants for use with IPv6.
``Related functions'' introduces the new and modified functions you need to use to create new applications.
``Passing sockets between applications'' describes a mechanism you can use to freely pass sockets between IPv6 and IPv4 applications.
``Porting IPv4 applications to IPv6'' outlines the modifications you must make to existing IPv4 applications so that they become ``IPv6 ready''.
``Porting IPv6 -- examples'' shows some examples of what modifications are required to ``IPv4'' code to make it ``IPv6 ready''.