[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <20180502.130024.362366635740254052.davem@davemloft.net>
Date: Wed, 02 May 2018 13:00:24 -0400 (EDT)
From: David Miller <davem@...emloft.net>
To: dsahern@...il.com
Cc: brouer@...hat.com, netdev@...r.kernel.org, borkmann@...earbox.net,
ast@...nel.org, shm@...ulusnetworks.com, roopa@...ulusnetworks.com,
toke@...e.dk, john.fastabend@...il.com, bernat@...fy.cx
Subject: Re: [RFC v2 bpf-next 8/9] bpf: Provide helper to do lookups in
kernel FIB table
From: David Ahern <dsahern@...il.com>
Date: Wed, 2 May 2018 09:37:21 -0600
> To share numbers from recent testing I did using Vincent's modules,
> lookup times in nsec (using local_clock) with MULTIPLE_TABLES config
> disabled for IPv4 and IPv6
>
> IPv4 IPv6-dst IPv6-fib6
> baseline 49 126 52
>
> I have other cases with combinations of configs and rules, but this
> shows the best possible case.
>
> IPv6 needs some more work to improve speeds with MULTIPLE_TABLES enabled
> (separate local and main tables unlike IPv4) and IPV6_SUBTREES enabled.
Yes, like for ipv4 sharing local and main tables will help a lot.
Powered by blists - more mailing lists