[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <55843522.7060804@cumulusnetworks.com>
Date: Fri, 19 Jun 2015 08:28:34 -0700
From: roopa <roopa@...ulusnetworks.com>
To: Robert Shearman <rshearma@...cade.com>
CC: ebiederm@...ssion.com, tgraf@...g.ch, davem@...emloft.net,
netdev@...r.kernel.org
Subject: Re: [PATCH net-next RFC v2 2/3] ipv4: add support for light weight
tunnel encap attributes
On 6/19/15, 8:19 AM, Robert Shearman wrote:
> On 19/06/15 05:49, Roopa Prabhu wrote:
>> From: Roopa Prabhu <roopa@...ulusnetworks.com>
>>
>> Introduces two netlink attributes RTA_ENCAP_TYPE and
>> RTA_ENCAP to support attaching encap information to ipv4 routes.
>
> Surely RTA_ENCAP_TYPE should be part of RTA_ENCAP, since the type
> doesn't make sense without the data and vice versa?
I went back and forth on this. And started with what you are saying
above. But then I wanted RTA_ENCAP netlink policy to be declared by
individual lwtunnel drivers.
And to determine which RTA_ENCAP netlink policy to pick, you need to
know the RTA_ENCAP policy type (or lwtunnel type)
which is encoded in RTA_ENCAP_TYPE. And I did not want to introduce
another level of nest in RTA_ENCAP (because for nexthops we are already
2 levels deep when parsing RTA_ENCAP).
Hence, fib code first looks for RTA_ENCAP and if RTA_ENCAP is specified,
RTA_ENCAP_TYPE is a required attribute. My iproute2 patches handles this
and makes sure
there is an RTA_ENCAP_TYPE specified with RTA_ENCAP.
thanks,
Roopa
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
Powered by blists - more mailing lists