On 23/10/15 15:44, Vivien Didelot wrote:
> On Oct. Friday 23 (43) 01:20 PM, Florian Fainelli wrote:
>> On 23/10/15 12:28, Vivien Didelot wrote:
>>> On Oct. Friday 23 (43) 11:38 AM, Florian Fainelli wrote:
>>>> +static int bcm_sf2_sw_fdb_del(struct dsa_switch *ds, int port,
>>>> + const struct switchdev_obj_port_fdb *fdb)
>>>> +{
>>>> + struct bcm_sf2_priv *priv = ds_to_priv(ds);
>>>> +
>>>> + return bcm_sf2_arl_op(priv, 0, port, fdb->addr, fdb->vid, false);
>>>> +}
>>>
>>> I'm wondering if you are populating the FDB of the invalid VLAN 0 here.
>>>
>>> Does your ARL consider that fdb->vid == 0 means "this port's FDB" and
>>> not "FDB of VLAN 0"?
>>
>> (please trim your replies)
>
> Noted, thanks.
>
>> I do not think this matters right now, since 802.1q is not currently
>> enabled/supported in the driver, but maybe I am trivializing this?
>
> I meant that, when you issue the following command:
>
> bridge fdb add de:ea:be:ef:12:34 dev swp3
>
> you will get vid == 0 in bcm_sf2_sw_fdb_add(). So I was wondering if
> your code was populating the FDB of the VLAN 0, instead of the choosen
> FDB associated to the given port.
It is populating the FDB for the given port right now, based on the
current switch configuration done by the driver. But thanks for noting
that, as this may need changing in the future.
--
Florian
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
