Re: [PATCH net 1/2] vxlan: unregister on namespace exit

[Pravin Shelar <pshelar@nicira.com>]

On Thu, Jul 18, 2013 at 9:58 AM, Stephen Hemminger
<stephen@networkplumber.org> wrote:
> On Thu, 18 Jul 2013 09:40:07 -0700
> Pravin Shelar <pshelar@nicira.com> wrote:
>
>> On Thu, Jul 18, 2013 at 8:38 AM, Stephen Hemminger
>> <stephen@networkplumber.org> wrote:
>> > Fix memory leaks and other badness from VXLAN network namespace
>> > teardown. When network namespace is removed, all the vxlan devices should
>> > be unregistered (not closed).
>> >
>> > Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
>> >
>> >
>> > --- a/drivers/net/vxlan.c       2013-07-17 16:32:10.826955800 -0700
>> > +++ b/drivers/net/vxlan.c       2013-07-17 16:47:35.275690741 -0700
>> > @@ -1878,10 +1878,12 @@ static __net_exit void vxlan_exit_net(st
>> >  {
>> >         struct vxlan_net *vn = net_generic(net, vxlan_net_id);
>> >         struct vxlan_dev *vxlan;
>> > +       LIST_HEAD(list);
>> >
>> >         rtnl_lock();
>> > -       list_for_each_entry(vxlan, &vn->vxlan_list, next)
>> > -               dev_close(vxlan->dev);
>> > +       list_for_each_entry(vxlan, &vn->vxlan_list, next)
>> > +               unregister_netdevice_queue(vxlan->dev, &list);
>> > +       unregister_netdevice_many(&list);
>> >         rtnl_unlock();
>> >  }
>> >
>> Looks good.
>> Reviewed-by: Pravin B Shelar <pshelar@nicira.com>
>
> vxlan might be a candidate for integration with the generic ip_tunnel
> code, but haven't looked at all the details.

ip_tunnel already does same on net-exit.