[Intel-wired-lan] [RFC PATCH net-next 0/5] ice: L2TPv3 offload support

[Jiri Pirko]

Fri, Aug 26, 2022 at 01:36:55PM CEST, wojciech.drewek at intel.com wrote:
>
>
>> -----Original Message-----
>> From: Jiri Pirko <jiri at resnulli.us>
>> Sent: piątek, 26 sierpnia 2022 13:16
>> To: Drewek, Wojciech <wojciech.drewek at intel.com>
>> Cc: netdev at vger.kernel.org; Lobakin, Alexandr <alexandr.lobakin at intel.com>; Brandeburg, Jesse <jesse.brandeburg at intel.com>;
>> Nguyen, Anthony L <anthony.l.nguyen at intel.com>; davem at davemloft.net; edumazet at google.com; kuba at kernel.org;
>> pabeni at redhat.com; jhs at mojatatu.com; xiyou.wangcong at gmail.com; marcin.szycik at linux.intel.com;
>> michal.swiatkowski at linux.intel.com; kurt at linutronix.de; boris.sukholitko at broadcom.com; vladbu at nvidia.com;
>> komachi.yoshiki at gmail.com; paulb at nvidia.com; baowen.zheng at corigine.com; louis.peens at corigine.com;
>> simon.horman at corigine.com; pablo at netfilter.org; maksym.glubokiy at plvision.eu; intel-wired-lan at lists.osuosl.org;
>> jchapman at katalix.com; gnault at redhat.com
>> Subject: Re: [RFC PATCH net-next 0/5] ice: L2TPv3 offload support
>> 
>> Fri, Aug 26, 2022 at 01:00:54PM CEST, wojciech.drewek at intel.com wrote:
>> >Add support for dissecting L2TPv3 session id in flow dissector. Add support
>> >for this field in tc-flower and support offloading L2TPv3. Finally, add
>> >support for hardware offload of L2TPv3 packets based on session id in
>> >switchdev mode in ice driver.
>> >
>> >Example filter:
>> >  # tc filter add dev $PF1 ingress prio 1 protocol ip \
>> >      flower \
>> >        ip_proto l2tp \
>> >        l2tpv3_sid 1234 \
>> >        skip_sw \
>> >      action mirred egress redirect dev $VF1_PR
>> >
>> >Changes in iproute2 are required to use the new fields.
>> >
>> >ICE COMMS DDP package is required to create a filter in ice.
>> 
>> I don't understand what do you mean by this. Could you please explain
>> what this mysterious "ICE COMMS DDP package" is? Do I understand it
>> correctly that without it, the solution would not work?
>
>Sorry, I'll include more precise description in the next version.
>DDP (Dynamic Device Personalization) is a firmware package that contains definitions
>protocol's headers and packets. It allows you  to add support for the new protocol to the
>NIC card without rebooting.  If the DDP package does not support L2TPv3 then hw offload 
>will not work, however sw offload will still work.

Hmm, so it is some FW part? Why do we care about it here in patchset
description?


>
>More info on DDP:
>https://www.intel.com/content/www/us/en/architecture-and-technology/ethernet/dynamic-device-personalization-brief.html
>
>> 
>> >
>> >Marcin Szycik (1):
>> >  ice: Add L2TPv3 hardware offload support
>> >
>> >Wojciech Drewek (4):
>> >  uapi: move IPPROTO_L2TP to in.h
>> >  flow_dissector: Add L2TPv3 dissectors
>> >  net/sched: flower: Add L2TPv3 filter
>> >  flow_offload: Introduce flow_match_l2tpv3
>> >
>> > .../ethernet/intel/ice/ice_protocol_type.h    |  8 +++
>> > drivers/net/ethernet/intel/ice/ice_switch.c   | 70 ++++++++++++++++++-
>> > drivers/net/ethernet/intel/ice/ice_tc_lib.c   | 27 ++++++-
>> > drivers/net/ethernet/intel/ice/ice_tc_lib.h   |  6 ++
>> > include/net/flow_dissector.h                  |  9 +++
>> > include/net/flow_offload.h                    |  6 ++
>> > include/uapi/linux/in.h                       |  2 +
>> > include/uapi/linux/l2tp.h                     |  2 -
>> > include/uapi/linux/pkt_cls.h                  |  2 +
>> > net/core/flow_dissector.c                     | 28 ++++++++
>> > net/core/flow_offload.c                       |  7 ++
>> > net/sched/cls_flower.c                        | 16 +++++
>> > 12 files changed, 179 insertions(+), 4 deletions(-)
>> >
>> >--
>> >2.31.1
>> >