On Sun, Aug 2, 2020 at 8:17 PM <pbhagavat...@marvell.com> wrote:
>
> From: Pavan Nikhilesh <pbhagavat...@marvell.com>
>
> The l3fwd example should use the reserved IPv4/v6 reserved address
> ranges defined in RFC5735 and RFC5180 and RFC863 discard protocol for
> the port number in the exact match mode of L3 forwarding.
>
> Signed-off-by: Pavan Nikhilesh <pbhagavat...@marvell.com>
> ---
> examples/l3fwd/l3fwd_em.c | 87 ++++++++++++++++++++++++++-------------
> 1 file changed, 59 insertions(+), 28 deletions(-)
Would such a change affect current users?
If so, please add a release note update.
I see DTS uses l3fwd.
If the change has an impact, probably worth reporting it on the dts ml.
>
> diff --git a/examples/l3fwd/l3fwd_em.c b/examples/l3fwd/l3fwd_em.c
> index fdbee70b4..481b6dcce 100644
> --- a/examples/l3fwd/l3fwd_em.c
> +++ b/examples/l3fwd/l3fwd_em.c
[snip]
> @@ -429,7 +460,7 @@ populate_ipv4_many_flow_into_table(const struct rte_hash
> *h,
> switch (i & (NUMBER_PORT_USED - 1)) {
> case 0:
> entry = ipv4_l3fwd_em_route_array[0];
> - entry.key.ip_dst = RTE_IPV4(101, c, b, a);
> + entry.key.ip_dst = RTE_IPV4(9, c, b, a);
> break;
> case 1:
> entry = ipv4_l3fwd_em_route_array[1];
> @@ -437,11 +468,11 @@ populate_ipv4_many_flow_into_table(const struct
> rte_hash *h,
> break;
> case 2:
> entry = ipv4_l3fwd_em_route_array[2];
> - entry.key.ip_dst = RTE_IPV4(111, c, b, a);
> + entry.key.ip_dst = RTE_IPV4(91, c, b, a);
> break;
> case 3:
> entry = ipv4_l3fwd_em_route_array[3];
> - entry.key.ip_dst = RTE_IPV4(211, c, b, a);
> + entry.key.ip_dst = RTE_IPV4(29, c, b, a);
> break;
> };
> convert_ipv4_5tuple(&entry.key, &newkey);
Why such 9, 91, 29 values? Do they have special properties?
--
David Marchand
