> -Original Message-
> From: David Miller
> Sent: Wednesday, July 24, 2019 12:03 AM
> To: Ioana Ciocoi Radulescu
> Cc: netdev@vger.kernel.org; Ioana Ciornei ; Vladimir
> Oltean
> Subject: Re: [PATCH net-next] dpaa2-eth: Don't use netif_receive_skb_list for
>
On 7/23/19 7:28 PM, Ioana Radulescu wrote:
> Using Rx skb bulking for all frames may negatively impact the
> performance in some TCP termination scenarios, as it effectively
> bypasses GRO.
>
> - list_add_tail(&skb->list, ch->rx_list);
> + if (frame_is_tcp(fd, fas))
> + na
From: Ioana Radulescu
Date: Tue, 23 Jul 2019 20:28:43 +0300
> Using Rx skb bulking for all frames may negatively impact the
> performance in some TCP termination scenarios, as it effectively
> bypasses GRO.
"may"?
Please provide numbers so that we know exactly whether it actually
hurts performa
On Tue, 2019-07-23 at 20:28 +0300, Ioana Radulescu wrote:
> Using Rx skb bulking for all frames may negatively impact the
> performance in some TCP termination scenarios, as it effectively
> bypasses GRO.
>
> Look at the hardware parse results of each ingress frame to see
> if a TCP header is pres
Using Rx skb bulking for all frames may negatively impact the
performance in some TCP termination scenarios, as it effectively
bypasses GRO.
Look at the hardware parse results of each ingress frame to see
if a TCP header is present or not; for TCP frames fall back to
the old implementation.
Signe
