Re: [PATCH net] net: drop bad gso csum_start and offset in virtio_net_hdr

[Paolo Abeni <pabeni@xxxxxxxxxx>]

On 7/26/24 04:32, Willem de Bruijn wrot> @@ -182,6 +171,11 @@ static 
inline int virtio_net_hdr_to_skb(struct sk_buff *skb,
>   			if (gso_type != SKB_GSO_UDP_L4)
>   				return -EINVAL;
>   			break;
> +		case SKB_GSO_TCPV4:
> +		case SKB_GSO_TCPV6:

I think we need to add here an additional check:

			if (!(hdr->flags & VIRTIO_NET_HDR_F_NEEDS_CSUM))
				return -EINVAL;

> +			if (skb->csum_offset != offsetof(struct tcphdr, check))
> +				return -EINVAL;
> +			break;
>   		}
>   
>   		/* Kernel has a special handling for GSO_BY_FRAGS. */
> diff --git a/net/ipv4/tcp_offload.c b/net/ipv4/tcp_offload.c
> index 4b791e74529e1..9e49ffcc77071 100644
> --- a/net/ipv4/tcp_offload.c
> +++ b/net/ipv4/tcp_offload.c
> @@ -140,6 +140,9 @@ struct sk_buff *tcp_gso_segment(struct sk_buff *skb,
>   	if (thlen < sizeof(*th))
>   		goto out;
>   
> +	if (unlikely(skb->csum_start != skb->transport_header))
> +		goto out;

Given that for packet injected from user-space, the transport offset is 
set to csum_start by skb_partial_csum_set(), do we need the above check? 
If so, why don't we need another similar one for csum_offset even here?

Thanks,

Paolo