On 2/6/23 14:02, Edward Cree wrote:
> On 02/02/2023 11:14, alejandro.lucero-palau@xxxxxxx wrote:
>> From: Alejandro Lucero <alejandro.lucero-palau@xxxxxxx>
>>
>> Using the data when enumerating mports, create devlink ports just before
>> netdevs are registered and remove those devlink ports after netdev has
>> been unregistered.
>>
>> Signed-off-by: Alejandro Lucero <alejandro.lucero-palau@xxxxxxx>
> ...
>> @@ -297,6 +298,7 @@ int efx_ef100_vfrep_create(struct efx_nic *efx, unsigned int i)
>> i, rc);
>> goto fail1;
>> }
>> + ef100_rep_set_devlink_port(efv);
>> rc = register_netdev(efv->net_dev);
>> if (rc) {
>> pci_err(efx->pci_dev,
>> @@ -308,6 +310,7 @@ int efx_ef100_vfrep_create(struct efx_nic *efx, unsigned int i)
>> efv->net_dev->name);
>> return 0;
>> fail2:
>> + ef100_rep_unset_devlink_port(efv);
>> efx_ef100_deconfigure_rep(efv);
>> fail1:
>> efx_ef100_rep_destroy_netdev(efv);
>> @@ -323,6 +326,7 @@ void efx_ef100_vfrep_destroy(struct efx_nic *efx, struct efx_rep *efv)
>> return;
>> netif_dbg(efx, drv, rep_dev, "Removing VF representor\n");
>> unregister_netdev(rep_dev);
>> + ef100_rep_unset_devlink_port(efv);
>> efx_ef100_deconfigure_rep(efv);
>> efx_ef100_rep_destroy_netdev(efv);
>> }
> Would it make sense to move these calls into
> efx_ef100_[de]configure_rep()? It's responsible for other
> MAE/m-port related stuff (and is also common with remote reps
> when they arrive).
Uhmm, not sure about this.
I would say configure/deconfigure reps is more driver's internal and
those devlink related calls are about the driver using an external API.
Indeed due to this relationship with register/unregister_netdev, I think
it is more visible where they are at the moment.
