NAK: [SRU][J:linux-bluefield][PATCH v1 00/32] support IPsec full offload
Tim Gardner
tim.gardner at canonical.com
Thu Sep 7 14:39:17 UTC 2023
On 9/6/23 10:30 AM, Tony Duan wrote:
> BugLink: https://bugs.launchpad.net/bugs/2034578
>
> Align Kernel IPsec Full offload implementation in the DPU to the upstream Full
> offload in all components: OFED, Strongswan, etc. This commit is in order for
> DPU Kernel IPsec to include policy offload and be fully aligned to what
> ConnectX Kernel customers will use.
>
> Most of the changes are already in uptream. In this commit, it contains 32
> patches include 4 reverts, 27 cherry-picks and 1 conflict resolver.
> Most of the patches are clean cherry-pick. The conflict are mainly
> focus on driver part and netdev_***() related functions. I ignored
> driver change and keep dev_***() unchanged.
>
> This commit also depends on
> "[SRU][J:linux-bluefield][PATCH v1 0/1] UBUNTU: [Config] bluefield: disable MLX5, and vendor drivers."
> (https://lists.ubuntu.com/archives/kernel-team/2023-September/142502.html)
> which is still under review.
>
> Merege request: https://code.launchpad.net/~yifeid/ubuntu/+source/linux-bluefield/+git/linux-bluefield/+merge/450800
>
> Edward Cree (1):
> netlink: add support for formatted extack messages
>
> Leon Romanovsky (21):
> xfrm: delete not used number of external headers
> xfrm: rename xfrm_state_offload struct to allow reuse
> xfrm: store and rely on direction to construct offload flags
> xfrm: drop not needed flags variable in XFRM offload struct
> xfrm: Remove not-used total variable
> xfrm: add new packet offload flag
> xfrm: allow state packet offload mode
> xfrm: add an interface to offload policy
> xfrm: add TX datapath support for IPsec packet offload mode
> xfrm: speed-up lookup of HW policies
> xfrm: add support to HW update soft and hard limits
> xfrm: document IPsec packet offload mode
> xfrm: extend add policy callback to set failure reason
> xfrm: extend add state callback to set failure reason
> netlink: provide an ability to set default extack message
> xfrm: don't require advance ESN callback for packet offload
> xfrm: release all offloaded policy memory
> xfrm: add missed call to delete offloaded policies
> xfrm: Support UDP encapsulation in packet offload mode
> xfrm: delete offloaded policy
> xfrm: don't skip free of empty state in acquire policy
>
> Raed Salem (2):
> xfrm: add new device offload acquire flag
> xfrm: copy_to_user_state fetch offloaded SA packets/bytes statistics
>
> Sabrina Dubroca (3):
> xfrm: add extack support to xfrm_dev_state_add
> xfrm: add extack to verify_policy_type
> xfrm: propagate extack to all netlink doit handlers
>
> Tony Duan (5):
> Revert "UBUNTU: SAUCE: net/xfrm: Fix XFRM flags validity check"
> Revert "UBUNTU: SAUCE: xfrm: Check if_id in xfrm full offload"
> Revert "UBUNTU: SAUCE: net/xfrm: Add support for xfrm full offload"
> Revert "UBUNTU: SAUCE: net/xfrm: IPsec full offload support for
> lifetime limit"
> UBUNTU: SAUCE: xfrm: fix conflict for netdev and tx stats
>
> Documentation/networking/xfrm_device.rst | 64 +++++-
> drivers/net/bonding/bond_main.c | 8 +-
> drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c | 5 +-
> .../chelsio/inline_crypto/ch_ipsec/chcr_ipsec.c | 10 +-
> drivers/net/ethernet/intel/ixgbe/ixgbe_ipsec.c | 11 +-
> drivers/net/ethernet/intel/ixgbevf/ipsec.c | 9 +-
> .../ethernet/mellanox/mlx5/core/en_accel/ipsec.c | 7 +-
> drivers/net/netdevsim/ipsec.c | 8 +-
> include/linux/netdevice.h | 6 +-
> include/linux/netlink.h | 39 +++-
> include/net/xfrm.h | 99 +++++++++-
> include/uapi/linux/xfrm.h | 7 +-
> net/bridge/br_switchdev.c | 10 +-
> net/xfrm/xfrm_device.c | 146 ++++++++++++--
> net/xfrm/xfrm_output.c | 22 +--
> net/xfrm/xfrm_policy.c | 127 +++++++-----
> net/xfrm/xfrm_state.c | 215 ++++++++++++++++-----
> net/xfrm/xfrm_user.c | 121 ++++++++----
> 18 files changed, 708 insertions(+), 206 deletions(-)
>
How about a normal pull request ? We don't do merges. Plus, the
Launchpad instructions don't work and I'm too lazy to figure it out.
--
-----------
Tim Gardner
Canonical, Inc
More information about the kernel-team
mailing list