[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Mon, 4 May 2020 21:36:08 -0600
From: David Ahern <dsahern@...il.com>
To: Roopa Prabhu <roopa@...ulusnetworks.com>, davem@...emloft.net
Cc: netdev@...r.kernel.org, nikolay@...ulusnetworks.com,
idosch@...lanox.com, jiri@...lanox.com, petrm@...lanox.com
Subject: Re: [RFC PATCH net-next 2/5] vxlan: ecmp support for mac fdb entries
On 5/4/20 4:28 PM, Roopa Prabhu wrote:
> diff --git a/include/uapi/linux/neighbour.h b/include/uapi/linux/neighbour.h
> index cd144e3..eefcda8 100644
> --- a/include/uapi/linux/neighbour.h
> +++ b/include/uapi/linux/neighbour.h
> @@ -29,6 +29,7 @@ enum {
> NDA_LINK_NETNSID,
> NDA_SRC_VNI,
> NDA_PROTOCOL, /* Originator of entry */
> + NDA_NH_ID,
> __NDA_MAX
> };
those attributes are shared by neighbor and bridge code for example.
nda_policy should reject it, and a new attribute provides a means for
starting strict checking (add NDA_UNSPEC with .strict_start_type =
NDA_NH_ID).
Similar for the fdb code in rtnetlink.c. Shame AF_BRIDGE parsing does
not have a policy attribute.
Powered by blists - more mailing lists