On Fri Oct 18 2024, Joe Damato wrote:
> On Tue, Oct 15, 2024 at 12:27:01PM +0200, Kurt Kanzenbach wrote:
>> On Mon Oct 14 2024, Joe Damato wrote:
>
> [...]
>
>> > diff --git a/drivers/net/ethernet/intel/igc/igc_main.c 
>> > b/drivers/net/ethernet/intel/igc/igc_main.c
>> > index 7964bbedb16c..59c00acfa0ed 100644
>> > --- a/drivers/net/ethernet/intel/igc/igc_main.c
>> > +++ b/drivers/net/ethernet/intel/igc/igc_main.c
>> > @@ -4948,6 +4948,47 @@ static int igc_sw_init(struct igc_adapter *adapter)
>> >    return 0;
>> >  }
>> >  
>> > +void igc_set_queue_napi(struct igc_adapter *adapter, int q_idx,
>> > +                  struct napi_struct *napi)
>> > +{
>> > +  if (adapter->flags & IGC_FLAG_QUEUE_PAIRS) {
>> > +          netif_queue_set_napi(adapter->netdev, q_idx,
>> > +                               NETDEV_QUEUE_TYPE_RX, napi);
>> > +          netif_queue_set_napi(adapter->netdev, q_idx,
>> > +                               NETDEV_QUEUE_TYPE_TX, napi);
>> > +  } else {
>> > +          if (q_idx < adapter->num_rx_queues) {
>> > +                  netif_queue_set_napi(adapter->netdev, q_idx,
>> > +                                       NETDEV_QUEUE_TYPE_RX, napi);
>> > +          } else {
>> > +                  q_idx -= adapter->num_rx_queues;
>> > +                  netif_queue_set_napi(adapter->netdev, q_idx,
>> > +                                       NETDEV_QUEUE_TYPE_TX, napi);
>> > +          }
>> > +  }
>> > +}
>> 
>> In addition, to what Vinicius said. I think this can be done
>> simpler. Something like this?
>> 
>> void igc_set_queue_napi(struct igc_adapter *adapter, int vector,
>>                      struct napi_struct *napi)
>> {
>>      struct igc_q_vector *q_vector = adapter->q_vector[vector];
>> 
>>      if (q_vector->rx.ring)
>>              netif_queue_set_napi(adapter->netdev, vector, 
>> NETDEV_QUEUE_TYPE_RX, napi);
>> 
>>      if (q_vector->tx.ring)
>>              netif_queue_set_napi(adapter->netdev, vector, 
>> NETDEV_QUEUE_TYPE_TX, napi);
>> }
>
> I tried this suggestion but this does not result in correct output
> in the case where IGC_FLAG_QUEUE_PAIRS is disabled.
>
> The output from netlink:
>
> $ ./tools/net/ynl/cli.py --spec Documentation/netlink/specs/netdev.yaml \
>                              --dump queue-get --json='{"ifindex": 2}'
>
> [{'id': 0, 'ifindex': 2, 'napi-id': 8193, 'type': 'rx'},
>  {'id': 1, 'ifindex': 2, 'napi-id': 8194, 'type': 'rx'},
>  {'id': 0, 'ifindex': 2, 'type': 'tx'},
>  {'id': 1, 'ifindex': 2, 'type': 'tx'}]
>
> Note the lack of a napi-id for the TX queues. This typically happens
> when the linking is not done correctly; netif_queue_set_napi should
> take a queue id as the second parameter.
>
> I believe the suggested code above should be modified to be as
> follows to use ring->queue_index:
>
>   if (q_vector->rx.ring)
>     netif_queue_set_napi(adapter->netdev,
>                          q_vector->rx.ring->queue_index,
>                          NETDEV_QUEUE_TYPE_RX, napi);
>   
>   if (q_vector->tx.ring)
>     netif_queue_set_napi(adapter->netdev,
>                          q_vector->tx.ring->queue_index,
>                          NETDEV_QUEUE_TYPE_TX, napi);

LGTM. Thanks.

Attachment: signature.asc
Description: PGP signature

Reply via email to