jamal wrote:
On Fri, 2006-07-04 at 11:31 -0700, David Daney wrote:
jamal wrote:
It exists. Just use it. For testing just use the ip utility.
But why does it exist? There must be one or more reasons that it exists.
There is only one reason it exists: to define an IP address as being
link local.
This influences what address is used as the src when going on different
paths which also have link scope.
It must not exist to cause IPv4 ARP to do broadcast as specified in RFC
3927, or we would not be having this conversation.
RFC 3927 defines what addresses are legit link local scope. The kernel
says what to do with addresses that are link local scopes. It knows this
if you tell it is a link local address.
The point that I think you are missing is that RFC 3927 defines what
happens only on the 169.254.0.0/16 subnet and more specifically how ARP
packets should be formed and sent on that particular subnet.
For link local addresses outside of the 169.254.0.0/16 network RFC 3927
does not apply.
Patching the ARP so that it does RFC 3927 ARP broadcasting for link
local addresses outside of 169.254.0.0/16 would be incorrect.
Overloading its current semantics, will cause in unnecessary ARP
broadcasts in non RFC 3927 cases.
Refer to what i said above.
cheers,
jamal
-
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at http://vger.kernel.org/majordomo-info.html
