[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <70077e59-1614-b059-a85e-42863f63144f@gmail.com>
Date: Fri, 21 Feb 2020 10:31:40 -0700
From: David Ahern <dsahern@...il.com>
To: Carmine Scarpitta <carmine.scarpitta@...roma2.it>
Cc: davem@...emloft.net, kuznet@....inr.ac.ru, yoshfuji@...ux-ipv6.org,
kuba@...nel.org, netdev@...r.kernel.org,
linux-kernel@...r.kernel.org, ahmed.abdelsalam@...i.it,
dav.lebrun@...il.com, andrea.mayer@...roma2.it,
paolo.lungaroni@...t.it, hiroki.shirokura@...ecorp.com
Subject: Re: [net-next 1/2] Perform IPv4 FIB lookup in a predefined FIB table
On 2/20/20 3:33 PM, Carmine Scarpitta wrote:
> As you can see, what is missing is having SRv6 End.DT4 supported to do
> decapsulation and VRF lookup.
Thanks for the detailed explanation and reference. Both this comment and
the slides reference VRF and per-tenant table lookup which is the same
as VRF.
If you allocated a VRF device for the tenant, the SRv6 header pop and
route lookup can be done very similar to how MPLS works when the goal is
to do a route lookup vs forwarding to a specific nexthop, be it main table:
ip -f mpls route add 200 dev lo
or VRF table
ip -f mpls route add 200 dev red
Take a look at the mpls code. Unless I am missing something this use
case is very similar.
Powered by blists - more mailing lists