Re: [PATCH net-next] bridge: del external_learned fdbs from device on flush or ageout

On Sun, Jun 14, 2015 at 3:14 PM, Rami Rosen [off-list ref] wrote:

Hi,

You mention bridge doing ageing vs. device doing
ageing. AFAIK, there is no way to prevent a bridge from doing aging by
a sysfs entry. Do you think that preventing bridge aging via sysfs is
needed?

The topic has been discussed before, but not at the patch level, yet,
AFIAK.  I had this XXX comment in
Documentation/networking/switchdev.txt:

     XXX: how to turn off ageing in kernel on a per-port basis or
otherwise prevent the kernel from ageing out the FDB entry?

I think the preference would be to use netlink over sysfs.  If we made
this a per-port flag, we could add bool IFLA_BRPORT_AGEING flag (and
BR_AGEING) for IFLA_PROTINFO.

There is also the need to push the bridge's ageing timeout value down
to the device.  This could be done with switchdev attr, e.g.
SWITCHDEV_ATTR_AGEING_TIMEOUT.  (Maybe there is already netlink
plumbing to do this...need to check).

Are you going to work on it?  ;)

-scott