Re: [bpf-next V1 PATCH 2/8] i40e: implement flush flag for ndo_xdp_xmit

[Song Liu]

On Wed, May 30, 2018 at 11:00 AM, Jesper Dangaard Brouer
<bro...@redhat.com> wrote:
> Signed-off-by: Jesper Dangaard Brouer <bro...@redhat.com>

I guess we still need to say something in the commit message?

> ---
>  drivers/net/ethernet/intel/i40e/i40e_txrx.c |    5 ++++-
>  1 file changed, 4 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c 
> b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
> index c0451d6e0790..03c1446f0465 100644
> --- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
> +++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
> @@ -3685,7 +3685,7 @@ int i40e_xdp_xmit(struct net_device *dev, int n, struct 
> xdp_frame **frames,
>         if (!i40e_enabled_xdp_vsi(vsi) || queue_index >= vsi->num_queue_pairs)
>                 return -ENXIO;
>
> -       if (unlikely(flags & ~XDP_XMIT_FLAGS_NONE))
> +       if (unlikely(flags & ~XDP_XMIT_FLAGS_MASK))
>                 return -EINVAL;
>
>         for (i = 0; i < n; i++) {
> @@ -3699,6 +3699,9 @@ int i40e_xdp_xmit(struct net_device *dev, int n, struct 
> xdp_frame **frames,
>                 }
>         }
>
> +       if (unlikely(flags & XDP_XMIT_FLUSH))
> +               i40e_xdp_ring_update_tail(vsi->xdp_rings[queue_index]);
> +
>         return n - drops;

Do we still flush when drops > 0?

Thanks,
Song

>  }
>
>