Re: [PATCH v21] virtio-net: support inner header hash

[Cornelia Huck <cohuck@redhat.com>]

On Mon, Jul 03 2023, Heng Qi <hengqi@linux.alibaba.com> wrote:

(...)

> +\paragraph{Inner Header Hash}
> +\label{sec:Device Types / Network Device / Device Operation / Processing of Incoming Packets / Inner Header Hash}
> +
> +If VIRTIO_NET_F_HASH_TUNNEL has been negotiated, the driver can send the command
> +VIRTIO_NET_CTRL_HASH_TUNNEL_SET to configure the calculation of the inner header hash.
> +
> +struct virtnet_hash_tunnel {
> +    le32 enabled_tunnel_types;
> +};
> +
> +#define VIRTIO_NET_CTRL_HASH_TUNNEL 7
> + #define VIRTIO_NET_CTRL_HASH_TUNNEL_SET 0

This needs to be wrapped in \begin{lstlisting}..\end{lstlisting}, can do
so when applying.

(...)

> +Encapsulation types applicable for inner header hash:
> +\begin{lstlisting}
> +#define VIRTIO_NET_HASH_TUNNEL_TYPE_GRE_2784    (1 << 0) /* \hyperref[intro:gre_rfc2784]{[GRE_rfc2784]} */
> +#define VIRTIO_NET_HASH_TUNNEL_TYPE_GRE_2890    (1 << 1) /* \hyperref[intro:gre_rfc2890]{[GRE_rfc2890]} */
> +#define VIRTIO_NET_HASH_TUNNEL_TYPE_GRE_7676    (1 << 2) /* \hyperref[intro:gre_rfc7676]{[GRE_rfc7676]} */
> +#define VIRTIO_NET_HASH_TUNNEL_TYPE_GRE_UDP     (1 << 3) /* \hyperref[intro:gre_in_udp_rfc8086]{[GRE-in-UDP]} */
> +#define VIRTIO_NET_HASH_TUNNEL_TYPE_VXLAN       (1 << 4) /* \hyperref[intro:vxlan]{[VXLAN]} */
> +#define VIRTIO_NET_HASH_TUNNEL_TYPE_VXLAN_GPE   (1 << 5) /* \hyperref[intro:vxlan_gpe]{[VXLAN-GPE]} */
> +#define VIRTIO_NET_HASH_TUNNEL_TYPE_GENEVE      (1 << 6) /* \hyperref[intro:geneve]{[GENEVE]} */
> +#define VIRTIO_NET_HASH_TUNNEL_TYPE_IPIP        (1 << 7) /* \hyperref[intro:ipip]{[IPIP]} */
> +#define VIRTIO_NET_HASH_TUNNEL_TYPE_NVGRE       (1 << 8) /* \hyperref[intro:nvgre]{[NVGRE]} */
> +\end{lstlisting}

I'm afraid this one doesn't come out quite as intended, we'll end up
with verbatim "\hyperref" text instead of a link. Anyone have a good
idea on how to fix that?

I'd prefer to push this now with the first issue addressed and to do an
(editorial) patch on top to deal with the second issue (unless someone
can come up with a really trivial fix for it, then I can apply that
straightaway.)