On 3/2/2017 4:03 PM, Andrew Rybchenko wrote:
> From: Roman Zhukov <roman.zhu...@oktetlabs.ru>
>
> Exact match of IP protocol, source and destination
> addresses is supported by parser.
> EtherType match is enforced to IPv4 EtherType.
>
> Signed-off-by: Roman Zhukov <roman.zhu...@oktetlabs.ru>
> Signed-off-by: Andrew Rybchenko <arybche...@solarflare.com>
> Reviewed-by: Andrew Lee <a...@solarflare.com>
<...>
>
> +static int
> +sfc_flow_parse_ipv4(const struct rte_flow_item *item,
Adding a function comment to all parser types can be useful. What do you
think?
It can help to verify implementation against intention. Also gives quick
information about supported parser without digging from function each time.
A list of valid/expected spec, mask and last values for that parser.
Also details like "if the mask is NULL, default mask will be used" can
be documented here.
> + efx_filter_spec_t *efx_spec,
> + struct rte_flow_error *error)
> +{
<...>
