[RFC net-next 0/8] use tc_can_offload_extack() throughout the drivers

[Jakub Kicinski <jakub.kicinski@netronome.com>]

Hi!

I have a number of patches for drivers to use tc_can_offload_extack()
as requested, but I wanted to ask for early comments and one specific
question - should we assume that type_data in block_cbs is always going
to contain struct tc_cls_common_offload as first argument?  In that
case we could just make it that struct instead of void?

It wasn't clear to me this assumption was safe therefore I pushed the
checks after enum tc_setup_type is checked against known values.  And
there is only one driver now (cxgb4) which supports two classifiers in
the same code base, so checking tc_can_offload before type_data is
cast doesn't buy much LOC.

Jiri, WDYT?

(i40e patch is queued up for later to avoid conflicts.)

Jakub Kicinski (8):
  pkt_cls: make tc_can_offload_extack() check chain index
  mlx5: use tc_can_offload_cls()
  mlxsw: use tc_can_offload_cls()
  cls_bpf: drivers: don't try to use extack before callback type is
    validated

 drivers/net/ethernet/broadcom/bnxt/bnxt.c           |  2 +-
 drivers/net/ethernet/broadcom/bnxt/bnxt_tc.c        |  2 +-
 drivers/net/ethernet/broadcom/bnxt/bnxt_vfr.c       |  2 +-
 drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c     |  7 ++-----
 drivers/net/ethernet/intel/ixgbe/ixgbe_main.c       |  5 +----
 drivers/net/ethernet/mellanox/mlx5/core/en_main.c   |  5 +----
 drivers/net/ethernet/mellanox/mlx5/core/en_rep.c    |  5 +----
 drivers/net/ethernet/mellanox/mlxsw/spectrum.c      |  5 +----
 drivers/net/ethernet/netronome/nfp/bpf/main.c       |  9 ++-------
 drivers/net/ethernet/netronome/nfp/flower/offload.c | 11 +++--------
 drivers/net/netdevsim/bpf.c                         | 10 ++--------
 include/net/pkt_cls.h                               | 20 +++++++++++++-------
 12 files changed, 29 insertions(+), 54 deletions(-)

-- 
2.15.1