ACK/Cmnt: [Pull Request][jammy] Add OVS Internal Port HW Offload to mlx5 driver

Tim Gardner tim.gardner at canonical.com
Fri Aug 5 14:13:03 UTC 2022


On 8/4/22 07:05, Zachary Tahenakos wrote:
> BugLink: https://bugs.launchpad.net/bugs/1983498 
> <https://bugs.launchpad.net/bugs/1983498>
> 
> SRU Justification
> 
> [Impact]
> 
> A request has been made, originating from 
> https://bugs.launchpad.net/bugs/1980730 
> <https://bugs.launchpad.net/bugs/1980730>, to add support for OVS 
> internal port offload for the mlx5 driver. This feature takes care of 
> certain traffic workloads that are not currently being offloaded to the 
> hardware.
> 
> [Test Case]
> 
> Tested by the requester.
> 
> [Where things could go wrong]
> 
> These changes could potentially introduce regressions in behavior of the 
> mlx5 driver.
> 
> The following changes since commit 4cc5e60c9399a10400737bda19312574b5a5c7a2:
> 
>    UBUNTU: Ubuntu-5.15.0-45.48 (2022-07-19 18:57:17 -0300)
> 
> are available in the Git repository at:
> 
> https://git.launchpad.net/~ztahenakos/ubuntu/+source/linux/+git/jammy 
> <https://git.launchpad.net/~ztahenakos/ubuntu/+source/linux/+git/jammy> 
> 5195a9ae669b266eb35cd8ee604923d4c5e5058a
> 
> for you to fetch changes up to 5195a9ae669b266eb35cd8ee604923d4c5e5058a:
> 
>    net/mlx5e: TC, fix decap fallback to uplink when int port not 
> supported (2022-08-03 15:10:10 -0400)
> 
> ----------------------------------------------------------------
> Ariel Levkovich (11):
>        net/mlx5e: Refactor rx handler of represetor device
>        net/mlx5e: Use generic name for the forwarding dev pointer
>        net/mlx5: E-Switch, Add ovs internal port mapping to metadata support
>        net/mlx5e: Accept action skbedit in the tc actions list
>        net/mlx5e: Offload tc rules that redirect to ovs internal port
>        net/mlx5e: Offload internal port as encap route device
>        net/mlx5e: Add indirect tc offload of ovs internal port
>        net/mlx5e: Term table handling of internal port rules
>        net/mlx5: Support internal port as decap route device
>        net/mlx5e: Fix wrong source vport matching on tunnel rule
>        net/mlx5e: TC, fix decap fallback to uplink when int port not 
> supported
> 
> Christophe JAILLET (1):
>        net/mlx5: Fix some error handling paths in 'mlx5e_tc_add_fdb_flow()'
> 
> Dima Chumak (2):
>        net/mlx5e: Enable TC offload for ingress MACVLAN
>        net/mlx5e: Fix nullptr on deleting mirroring rule
> 
> Gal Pressman (1):
>        net/mlx5e: Fix skb memory leak when TC classifier action offloads 
> are disabled
> 
> Roi Dayan (2):
>        net/mlx5e: TC, Fix memory leak with rules with internal port
>        net/mlx5e: Avoid implicit modify hdr for decap drop rule
> 
> Vlad Buslov (1):
>        net/mlx5e: Support accept action
> 
>   drivers/net/ethernet/mellanox/mlx5/core/Makefile   |   2 +-
>   .../net/ethernet/mellanox/mlx5/core/en/rep/tc.c    | 132 ++++--
>   .../net/ethernet/mellanox/mlx5/core/en/rep/tc.h    |  14 +-
>   .../ethernet/mellanox/mlx5/core/en/tc/int_port.c   | 457 
> +++++++++++++++++++++
>   .../ethernet/mellanox/mlx5/core/en/tc/int_port.h   |  65 +++
>   .../net/ethernet/mellanox/mlx5/core/en/tc_priv.h   |   2 +
>   .../net/ethernet/mellanox/mlx5/core/en/tc_tun.c    |  33 +-
>   .../ethernet/mellanox/mlx5/core/en/tc_tun_encap.c  |  35 ++
>   drivers/net/ethernet/mellanox/mlx5/core/en_rep.c   |  13 +-
>   drivers/net/ethernet/mellanox/mlx5/core/en_rep.h   |   4 +
>   drivers/net/ethernet/mellanox/mlx5/core/en_rx.c    |  22 +-
>   drivers/net/ethernet/mellanox/mlx5/core/en_tc.c    | 205 ++++++++-
>   drivers/net/ethernet/mellanox/mlx5/core/en_tc.h    |  11 +-
>   drivers/net/ethernet/mellanox/mlx5/core/eswitch.h  |  16 +
>   .../ethernet/mellanox/mlx5/core/eswitch_offloads.c |  84 ++--
>   .../mellanox/mlx5/core/eswitch_offloads_termtbl.c  |   7 +-
>   16 files changed, 998 insertions(+), 104 deletions(-)
>   create mode 100644 
> drivers/net/ethernet/mellanox/mlx5/core/en/tc/int_port.c
>   create mode 100644 
> drivers/net/ethernet/mellanox/mlx5/core/en/tc/int_port.h
> 
> 
Acked-by: Tim Gardner <tim.gardner at canonical.com>

Clean cherry picks, but I'm a little confused by "net/mlx5: Fix some 
error handling paths in 'mlx5e_tc_add_fdb_flow()'". How can you have 2 
cherry pick commit IDs for one patch ? I did note that both commit IDs 
produced identical patches, which I also find confusing.

-- 
-----------
Tim Gardner
Canonical, Inc



More information about the kernel-team mailing list