[Intel-wired-lan] [PATCH 2/3 net-next v5] igb: refactor XDP registration

Alexander Lobakin alexandr.lobakin at intel.com
Tue Jan 18 15:05:12 UTC 2022


From: Corinna Vinschen <vinschen at redhat.com>
Date: Mon, 17 Jan 2022 19:29:14 +0100

> On changing the RX ring parameters igb uses a hack to avoid a warning
> when calling xdp_rxq_info_reg via igb_setup_rx_resources.  It just
> clears the struct xdp_rxq_info content.
> 
> Change this to unregister if we're already registered instead.  Align
> code to the igc code.
> 
> Fixes: 9cbc948b5a20c ("igb: add XDP support")
> Signed-off-by: Corinna Vinschen <vinschen at redhat.com>
> ---
>  drivers/net/ethernet/intel/igb/igb_ethtool.c |  4 ----
>  drivers/net/ethernet/intel/igb/igb_main.c    | 12 +++++++++---
>  2 files changed, 9 insertions(+), 7 deletions(-)
> 
> diff --git a/drivers/net/ethernet/intel/igb/igb_ethtool.c b/drivers/net/ethernet/intel/igb/igb_ethtool.c
> index 51a2dcaf553d..2a5782063f4c 100644
> --- a/drivers/net/ethernet/intel/igb/igb_ethtool.c
> +++ b/drivers/net/ethernet/intel/igb/igb_ethtool.c
> @@ -965,10 +965,6 @@ static int igb_set_ringparam(struct net_device *netdev,
>  			memcpy(&temp_ring[i], adapter->rx_ring[i],
>  			       sizeof(struct igb_ring));
>  
> -			/* Clear copied XDP RX-queue info */
> -			memset(&temp_ring[i].xdp_rxq, 0,
> -			       sizeof(temp_ring[i].xdp_rxq));
> -
>  			temp_ring[i].count = new_rx_count;
>  			err = igb_setup_rx_resources(&temp_ring[i]);
>  			if (err) {
> diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c
> index 38ba92022cd4..cea89d301bfd 100644
> --- a/drivers/net/ethernet/intel/igb/igb_main.c
> +++ b/drivers/net/ethernet/intel/igb/igb_main.c
> @@ -4352,7 +4352,7 @@ int igb_setup_rx_resources(struct igb_ring *rx_ring)
>  {
>  	struct igb_adapter *adapter = netdev_priv(rx_ring->netdev);
>  	struct device *dev = rx_ring->dev;
> -	int size;
> +	int size, res;
>  
>  	size = sizeof(struct igb_rx_buffer) * rx_ring->count;
>  
> @@ -4376,9 +4376,15 @@ int igb_setup_rx_resources(struct igb_ring *rx_ring)
>  	rx_ring->xdp_prog = adapter->xdp_prog;
>  
>  	/* XDP RX-queue info */
> -	if (xdp_rxq_info_reg(&rx_ring->xdp_rxq, rx_ring->netdev,
> -			     rx_ring->queue_index, 0) < 0)
> +	if (xdp_rxq_info_is_reg(&rx_ring->xdp_rxq))
> +		xdp_rxq_info_unreg(&rx_ring->xdp_rxq);
> +	res = xdp_rxq_info_reg(&rx_ring->xdp_rxq, rx_ring->netdev,
> +			       rx_ring->queue_index, 0);
> +	if (res < 0) {
> +		dev_err(dev, "Failed to register xdp_rxq index %u\n",
> +			rx_ring->queue_index);
>  		goto err;

Error path always returns -ENOMEM, even in this case, and reports
that it failed to allocate memory for rings. Handle this correctly
and return `res` instead and without one more error message?

> +	}

As I mentioned a bit above, `res` is unused here as an error code,
only to test the value on < 0. Does it make sense to add a new
variable?

>  
>  	return 0;
>  
> -- 
> 2.27.0

Thanks,
Al


More information about the Intel-wired-lan mailing list