Support on the switch includes expanded address capability, header format simplification, improved
support of extensions and options, and hardware parsing of the extension header. The switch supports
hop-by-hop extension header packets, which are routed or bridged in software.

The switch provides IPv6 routing capability over native Ethernet Inter-Switch Link (ISL) or 802.1Q
trunk ports for static routes, Routing Information Protocol (RIP) for IPv6, and Open Shortest Path First
(OSPF) Version 3 Protocol. It supports up to 16 equal-cost routes and can simultaneously forward IPv4
and IPv6 frames at line rate.

128-Bit Wide Unicast Addresses

The switch supports aggregatable global unicast addresses and link-local unicast addresses. It does not
support site-local unicast addresses.

•

Aggregatable global unicast addresses are IPv6 addresses from the aggregatable global unicast
prefix. The address structure enables strict aggregation of routing prefixes and limits the number of
routing table entries in the global routing table. These addresses are used on links that are aggregated
through organizations and eventually to the Internet service provider.

These addresses are defined by a global routing prefix, a subnet ID, and an interface ID. Current
global unicast address allocation uses the range of addresses that start with binary value 001
(2000::/3). Addresses with a prefix of 2000::/3(001) through E000::/3(111) must have 64-bit
interface identifiers in the extended unique identifier (EUI)-64 format.

•

Link local unicast addresses can be automatically configured on any interface by using the link-local
prefix FE80::/10(1111 1110 10) and the interface identifier in the modified EUI format. Link-local
addresses are used in the neighbor discovery protocol (NDP) and the stateless autoconfiguration