On Fri, Jan 09, 2026 at 02:25:52PM +0000, Bruce Richardson wrote:
> On Fri, Jan 09, 2026 at 03:21:30PM +0100, Robin Jarry wrote:
> > With a regular iavf device, rte_eth_dev_info_get reports 256 maximum TX
> > and RX queues. Trying to configure a port with 20 queues returns an
> > error:
> > 
> >   ERR: IAVF_DRIVER: iavf_dev_configure(): large VF is not supported
> > 
> > When the large VF feature isn't supported by the PF kernel driver,
> > ethdev info must not report 256 supported queues but 16.
> > 
> > Fixes: e436cd43835b ("net/iavf: negotiate large VF and request more queues")
> > Cc: [email protected]
> > 
> > Signed-off-by: Robin Jarry <[email protected]>
> > ---
> >  drivers/net/intel/iavf/iavf_ethdev.c | 10 ++++++++--
> >  1 file changed, 8 insertions(+), 2 deletions(-)
> > 
> > diff --git a/drivers/net/intel/iavf/iavf_ethdev.c 
> > b/drivers/net/intel/iavf/iavf_ethdev.c
> > index 15e49fe24814..59328ebc5ad2 100644
> > --- a/drivers/net/intel/iavf/iavf_ethdev.c
> > +++ b/drivers/net/intel/iavf/iavf_ethdev.c
> > @@ -1126,12 +1126,18 @@ iavf_dev_info_get(struct rte_eth_dev *dev, struct 
> > rte_eth_dev_info *dev_info)
> >     struct iavf_adapter *adapter =
> >             IAVF_DEV_PRIVATE_TO_ADAPTER(dev->data->dev_private);
> >     struct iavf_info *vf = &adapter->vf;
> > +   uint16_t max_queue_pairs;
> >  
> >     if (adapter->closed)
> >             return -EIO;
> >  
> > -   dev_info->max_rx_queues = IAVF_MAX_NUM_QUEUES_LV;
> > -   dev_info->max_tx_queues = IAVF_MAX_NUM_QUEUES_LV;
> > +   if (vf->vf_res->vf_cap_flags & VIRTCHNL_VF_LARGE_NUM_QPAIRS)
> > +           max_queue_pairs = IAVF_MAX_NUM_QUEUES_LV;
> > +   else
> > +           max_queue_pairs = IAVF_MAX_NUM_QUEUES_DFLT;
> > +
> > +   dev_info->max_rx_queues = max_queue_pairs;
> > +   dev_info->max_tx_queues = max_queue_pairs;
> >     dev_info->min_rx_bufsize = IAVF_BUF_SIZE_MIN;
> >     dev_info->max_rx_pktlen = IAVF_FRAME_SIZE_MAX;
> >     dev_info->max_mtu = dev_info->max_rx_pktlen - IAVF_ETH_OVERHEAD;
> > -- 
> Makes sense.
> 
> Acked-by: Bruce Richardson <[email protected]>
> 
Applied to dpdk-next-net-intel.

/Bruce

Reply via email to