[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <54A2355B.6060903@cumulusnetworks.com>
Date: Mon, 29 Dec 2014 21:17:15 -0800
From: roopa <roopa@...ulusnetworks.com>
To: Scott Feldman <sfeldma@...il.com>
CC: Netdev <netdev@...r.kernel.org>, shemminger@...tta.com,
"vyasevic@...hat.com" <vyasevic@...hat.com>,
Wilson Kok <wkok@...ulusnetworks.com>
Subject: Re: [PATCH 6/6] bridge: add vlan info to bridge setlink and dellink
notification messages
On 12/29/14, 3:42 PM, Scott Feldman wrote:
> On Mon, Dec 29, 2014 at 1:05 PM, <roopa@...ulusnetworks.com> wrote:
>> From: Roopa Prabhu <roopa@...ulusnetworks.com>
>>
>> vlan add/deletes are not notified to userspace today. This patch fixes it.
>> Notifications will contain vlans compressed into ranges whereever applicable.
> Why is this notification needed? On VLAN add/del, the port driver
> will already get called if it implements ndo_vlan_rx_add_vid and
> ndo_vlan_rx_kill_vid.
This is strictly for userspace.
> The port driver already gets the notification
> it needs to filter vlans. User space can query for vlan port
> membership if it needs to know.
You mean request a dump ?. yes it can if it needs all the configured vlans.
But for notifications, we must at the least notify what changed with
respect to vlans with RTM_SETLINK/DELLINK, no ?
>
>> Signed-off-by: Roopa Prabhu <roopa@...ulusnetworks.com>
>> ---
>> net/bridge/br_netlink.c | 3 ++-
>> net/core/rtnetlink.c | 3 ++-
>> 2 files changed, 4 insertions(+), 2 deletions(-)
>>
>> diff --git a/net/bridge/br_netlink.c b/net/bridge/br_netlink.c
>> index 16bdd5a..cebfb03 100644
>> --- a/net/bridge/br_netlink.c
>> +++ b/net/bridge/br_netlink.c
>> @@ -303,7 +303,8 @@ void br_ifinfo_notify(int event, struct net_bridge_port *port)
>> if (skb == NULL)
>> goto errout;
>>
>> - err = br_fill_ifinfo(skb, port, 0, 0, event, 0, 0, port->dev);
>> + err = br_fill_ifinfo(skb, port, 0, 0, event, 0,
>> + RTEXT_FILTER_BRVLAN_COMPRESSED, port->dev);
> This doesn't look right. The skb wasn't allocated large enough to
> hold vlan info. If this is working in your tests, you're getting
> lucky due to some skb padding.
>
> skb was allocated with br_nlmsg_size(), which doesn't include vlan info.
>
>> if (err < 0) {
>> /* -EMSGSIZE implies BUG in br_nlmsg_size() */
>> WARN_ON(err == -EMSGSIZE);
>> diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c
>> index d06107d..dad5fb6 100644
>> --- a/net/core/rtnetlink.c
>> +++ b/net/core/rtnetlink.c
>> @@ -2878,7 +2878,8 @@ static int rtnl_bridge_notify(struct net_device *dev, u16 flags)
>>
>> if ((!flags || (flags & BRIDGE_FLAGS_MASTER)) &&
>> br_dev && br_dev->netdev_ops->ndo_bridge_getlink) {
>> - err = br_dev->netdev_ops->ndo_bridge_getlink(skb, 0, 0, dev, 0);
>> + err = br_dev->netdev_ops->ndo_bridge_getlink(skb, 0, 0, dev,
>> + RTEXT_FILTER_BRVLAN_COMPRESSED);
>> if (err < 0)
>> goto errout;
>> }
>> --
>> 1.7.10.4
>>
>> --
>> To unsubscribe from this list: send the line "unsubscribe netdev" in
>> the body of a message to majordomo@...r.kernel.org
>> More majordomo info at http://vger.kernel.org/majordomo-info.html
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Powered by blists - more mailing lists