Hi,
On Mon, 2019-01-28 at 09:50 +0100, Steffen Klassert wrote:
> diff --git a/net/ipv4/udp_offload.c b/net/ipv4/udp_offload.c
> index 584635db9231..c0be33216750 100644
> --- a/net/ipv4/udp_offload.c
> +++ b/net/ipv4/udp_offload.c
[...]
> @@ -379,8 +397,17 @@ static struct sk_buff *udp_gro_receive_segment(struct
> list_head *head,
> * Under small packet flood GRO count could elsewhere grow a lot
> * leading to execessive truesize values
> */
> - if (!skb_gro_receive(p, skb) &&
> - NAPI_GRO_CB(p)->count >= UDP_GRO_CNT_MAX)
> + if (NAPI_GRO_CB(skb)->is_flist) {
> + if (!pskb_may_pull(skb, skb_gro_offset(skb)))
> + return NULL;
> + ret = skb_gro_receive_list(p, skb);
> + } else {
> + skb_gro_postpull_rcsum(skb, uh, sizeof(struct udphdr));
> +
> + ret = skb_gro_receive(p, skb);
> + }
> +
> + if (!ret && NAPI_GRO_CB(p)->count > UDP_GRO_CNT_MAX)
^^
Minor nitpick: here we may want to preserve the '>=' operator.
Note: I've not finished looking at the patches yet, I'll try to provide
some benck-marking and it will take some time.
Cheers,
Paolo
