Darren Reed wrote:
Is there a reason why ip_wput_ire() shouldn't be heavily rototilled? (apart
from funding, that is)
ip_wput/ip_wput_ire needs to be simplified.
I'm looking at what this would take and it involved making ip_newroute*
return an IRE_CACHE (even if the L2 information isn't known), and
radically simplifying all the multirt code.
I'm trying to find time to write up things in a form where people can
comment on the ideas.
Erik
_______________________________________________
networking-discuss mailing list
[email protected]
