On Tue, Mar 21, 2017 at 12:04:47PM +0800, Jason Wang wrote:
> For the socket that exports its skb array, we can use lockless polling
> to avoid touching spinlock during busy polling.
> 
> Signed-off-by: Jason Wang <jasow...@redhat.com>
> ---
>  drivers/vhost/net.c | 7 +++++--
>  1 file changed, 5 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/vhost/net.c b/drivers/vhost/net.c
> index 53f09f2..41153a3 100644
> --- a/drivers/vhost/net.c
> +++ b/drivers/vhost/net.c
> @@ -551,10 +551,13 @@ static int peek_head_len(struct vhost_net_virtqueue 
> *rvq, struct sock *sk)
>       return len;
>  }
>  
> -static int sk_has_rx_data(struct sock *sk)
> +static int sk_has_rx_data(struct vhost_net_virtqueue *rvq, struct sock *sk)
>  {
>       struct socket *sock = sk->sk_socket;
>  
> +     if (rvq->rx_array)
> +             return !__skb_array_empty(rvq->rx_array);
> +
>       if (sock->ops->peek_len)
>               return sock->ops->peek_len(sock);
>  

I don't see which patch adds __skb_array_empty.

> @@ -579,7 +582,7 @@ static int vhost_net_rx_peek_head_len(struct vhost_net 
> *net,
>               endtime = busy_clock() + vq->busyloop_timeout;
>  
>               while (vhost_can_busy_poll(&net->dev, endtime) &&
> -                    !sk_has_rx_data(sk) &&
> +                    !sk_has_rx_data(rvq, sk) &&
>                      vhost_vq_avail_empty(&net->dev, vq))
>                       cpu_relax();
>  
> -- 
> 2.7.4

Reply via email to