Re: [PATCH net-next] net: hsr: Add support for MC filtering at the slave device

From: Ravi Gunasekaran
Date: Mon Nov 20 2023 - 23:18:27 EST




On 11/20/2023 5:10 PM, Wojciech Drewek wrote:
>
> On 20.11.2023 12:01, Ravi Gunasekaran wrote:
>> From: Murali Karicheri <m-karicheri2@xxxxxx>
>>
>> When MC (multicast) list is updated by the networking layer due to a
>> user command and as well as when allmulti flag is set, it needs to be
>> passed to the enslaved Ethernet devices. This patch allows this
>> to happen by implementing ndo_change_rx_flags() and ndo_set_rx_mode()
>> API calls that in turns pass it to the slave devices using
>> existing API calls.
>>
>> Signed-off-by: Murali Karicheri <m-karicheri2@xxxxxx>
>> Signed-off-by: Ravi Gunasekaran <r-gunasekaran@xxxxxx>
>> ---
> One small nit, besides that:
> Reviewed-by: Wojciech Drewek <wojciech.drewek@xxxxxxxxx>
>
>> net/hsr/hsr_device.c | 67 +++++++++++++++++++++++++++++++++++++++++++-
>> 1 file changed, 66 insertions(+), 1 deletion(-)
>>
>> diff --git a/net/hsr/hsr_device.c b/net/hsr/hsr_device.c
>> index 306f942c3b28..4e8f4a3cefbf 100644
>> --- a/net/hsr/hsr_device.c
>> +++ b/net/hsr/hsr_device.c
>> @@ -173,7 +173,24 @@ static int hsr_dev_open(struct net_device *dev)
>>
>> static int hsr_dev_close(struct net_device *dev)
>> {
>> - /* Nothing to do here. */
>> + struct hsr_port *port;
>> + struct hsr_priv *hsr;
>> +
>> + hsr = netdev_priv(dev);
>> + hsr_for_each_port(hsr, port) {
>> + if (port->type == HSR_PT_MASTER)
>> + continue;
>> + switch (port->type) {
>> + case HSR_PT_SLAVE_A:
>> + case HSR_PT_SLAVE_B:
>> + dev_uc_unsync(port->dev, dev);
>> + dev_mc_unsync(port->dev, dev);
>> + break;
>> + default:
>> + break;
>> + }
>> + }
>> +
>> return 0;
>> }
>>
>> @@ -404,12 +421,60 @@ void hsr_del_ports(struct hsr_priv *hsr)
>> hsr_del_port(port);
>> }
>>
>> +static void hsr_ndo_set_rx_mode(struct net_device *dev)
> I wouldn't add "ndo" to the callback name, let's not stand out from other ndo callbacks here.

Thank you for reviewing the patch. I will address the above comment and send out
a v2.

>> +{
>> + struct hsr_port *port;
>> + struct hsr_priv *hsr;
>> +
>> + hsr = netdev_priv(dev);
>> +
>> + hsr_for_each_port(hsr, port) {
>> + if (port->type == HSR_PT_MASTER)
>> + continue;
>> + switch (port->type) {
>> + case HSR_PT_SLAVE_A:
>> + case HSR_PT_SLAVE_B:
>> + dev_mc_sync_multiple(port->dev, dev);
>> + dev_uc_sync_multiple(port->dev, dev);
>> + break;
>> + default:
>> + break;
>> + }
>> + }
>> +}
>> +
>> +static void hsr_change_rx_flags(struct net_device *dev, int change)
>> +{
>> + struct hsr_port *port;
>> + struct hsr_priv *hsr;
>> +
>> + hsr = netdev_priv(dev);
>> +
>> + hsr_for_each_port(hsr, port) {
>> + if (port->type == HSR_PT_MASTER)
>> + continue;
>> + switch (port->type) {
>> + case HSR_PT_SLAVE_A:
>> + case HSR_PT_SLAVE_B:
>> + if (change & IFF_ALLMULTI)
>> + dev_set_allmulti(port->dev,
>> + dev->flags &
>> + IFF_ALLMULTI ? 1 : -1);
>> + break;
>> + default:
>> + break;
>> + }
>> + }
>> +}
>> +
>> static const struct net_device_ops hsr_device_ops = {
>> .ndo_change_mtu = hsr_dev_change_mtu,
>> .ndo_open = hsr_dev_open,
>> .ndo_stop = hsr_dev_close,
>> .ndo_start_xmit = hsr_dev_xmit,
>> + .ndo_change_rx_flags = hsr_change_rx_flags,
>> .ndo_fix_features = hsr_fix_features,
>> + .ndo_set_rx_mode = hsr_ndo_set_rx_mode,
>> };
>>
>> static struct device_type hsr_type = {
>>
>> base-commit: 5a82d69d48c82e89aef44483d2a129f869f3506a