All of lore.kernel.org
 help / color / mirror / Atom feed
* [RFCv2 net-next 000/167] net: extend the netdev_features_t
@ 2021-09-29 15:50 Jian Shen
  2021-09-29 15:50 ` [RFCv2 net-next 001/167] net: convert the prototype of netdev_intersect_features Jian Shen
                   ` (170 more replies)
  0 siblings, 171 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:50 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

For the prototype of netdev_features_t is u64, and the number
of netdevice feature bits is 64 now. So there is no space to
introduce new feature bit.

This patchset try to solve it by change the prototype of
netdev_features_t from u64 to bitmap. With this change,
it's necessary to introduce a set of bitmap operation helpers
for netdev features. Meanwhile, the functions which use
netdev_features_t as return value are also need to be changed,
return the result as an output parameter.

With above changes, it will affect hundreds of files, and all the
nic drivers. To make it easy to be reviewed, split the changes
to 167 patches to 5 parts.

patch 1~22: convert the prototype which use netdev_features_t
as return value
patch 24: introduce fake helpers for bitmap operation
patch 25~165: use netdev_feature_xxx helpers
patch 166: use macro __DECLARE_NETDEV_FEATURE_MASK to replace
netdev_feature_t declaration.
patch 167: change the type of netdev_features_t to bitmap,
and rewrite the bitmap helpers.

Sorry to send a so huge patchset, I wanna to get more suggestions
to finish this work, to make it much more reviewable and feasible.

The former discussing for the changes, see [1]
[1]. https://www.spinics.net/lists/netdev/msg753528.html

ChangeLog:
V1->V2:
use bitmap by comment from Andrew Lunn

Jian Shen (167):
  net: convert the prototype of netdev_intersect_features
  net: convert the prototype of netdev_get_wanted_features
  net: convert the prototype of net_mpls_features
  net: convert the prototype of harmonize_features
  net: convert the prototype of gso_features_check
  net: convert the prototype of vlan_features_check
  net: convert the prototype of vxlan_features_check
  net: convert the prototype of dflt_features_check
  net: convert the prototype of ndo_features_check
  net: convert the prototype of netif_skb_features
  net: convert the prototype of ndo_fix_features
  net: convert the prototype of netdev_fix_features
  net: convert the prototype of netdev_sync_upper_features
  net: convert the prototype of br_features_recompute
  net: convert the prototype of netdev_add_tso_features
  net: convert the prototype of netdev_increment_features
  net: convert the prototype of hsr_features_recompute
  net: mlx5: convert prototype of mlx5e_ipsec_feature_check,
    mlx5e_tunnel_features_check and mlx5e_fix_uplink_rep_features
  net: sfc: convert the prototype of xxx_supported_features
  net: qlogic: convert the prototype of qlcnic_process_flags
  net: realtek: convert the prototype of rtl8168evl_fix_tso
  ethtool: convert the prototype of ethtool_get_feature_mask
  net: add netdev feature helpers
  net: core: use netdev feature helpers
  skbuff: use netdev feature helpers
  net: vlan: use netdev feature helpers
  bridge: use netdev feature helpers
  ipvlan: use netdev feature helpers
  veth: use netdev feature helpers
  bonding: use netdev feature helpers
  net: tun: use netdev feature helpers
  net: tap: use netdev feature helpers
  net: geneve: use netdev feature helpers
  hv_netvsc: use netdev feature helpers
  macvlan: use netdev feature helpers
  macsec: use netdev feature helpers
  net: tls: use netdev feature helpers
  s390: qeth: use netdev feature helpers
  dsa: use netdev feature helpers
  macvtap: use netdev feature helpers
  team: use netdev feature helpers
  vmxnet3: use netdev feature helpers
  net_failover: use netdev feature helpers
  pktgen: use netdev feature helpers
  net: sched: use netdev feature helpers
  netdevsim: use netdev feature helpers
  virtio_net: use netdev feature helpers
  net: ipv4: use netdev feature helpers
  net: ipv6: use netdev feature helpers
  net: hsr: use netdev feature helpers
  net: mpls: use netdev feature helpers
  net: nsh: use netdev feature helpers
  net: decnet: use netdev feature helpers
  net: dccp: use netdev feature helpers
  net: l2tp: use netdev feature helpers
  net: ntb_netdev: use netdev feature helpers
  net: thunderbolt: use netdev feature helpers
  net: phonet: use netdev feature helpers
  net: vrf: use netdev feature helpers
  net: sctp: use netdev feature helpers
  vxlan: use netdev feature helpers
  xen-netback: use netdev feature helpers
  xen-netfront: use netdev feature helpers
  sock: use netdev feature helpers
  sunrpc: use netdev feature helpers
  net: caif: use netdev feature helpers
  net: loopback: use netdev feature helpers
  net: dummy: use netdev_feature helpers
  net: mac80211: use netdev feature helpers
  net: ifb: use netdev feature helpers
  net: bareudp: use netdev feature helpers
  net: rionet: use netdev feature helpers
  net: gtp: use netdev feature helpers
  net: vsockmon: use netdev feature helpers
  net: nlmon: use netdev feature helpers
  net: wireguard: use netdev feature helpers
  net: can: use netdev feature helpers
  net: ppp: use netdev feature helpers
  net: ipa: use netdev feature helpers
  net: fjes: use netdev feature helpers
  net: usb: use netdev feature helpers
  net: wireless: use netdev feature helpers
  net: realtek: use netdev feature helpers
  net: broadcom: use netdev feature helpers
  net: intel: use netdev feature helpers
  net: hisilicon: use netdev feature helpers
  net: mellanox: use netdev feature helpers
  net: atlantic: use netdev feature helpers
  net: atheros: use netdev feature helpers
  net: chelsio: use netdev feature helpers
  net: davicom: use netdev feature helpers
  net: freescale: use netdev feature helpers
  net: synopsys: use netdev feature helpers
  net: sfc: use netdev feature helpers
  net: qualcomm: use netdev feature helpers
  net: nvidia: use netdev feature helpers
  net: faraday: use netdev feature helpers
  net: google: use netdev feature helpers
  net: hinic: use netdev feature helpers
  net: ibm: use netdev feature helpers
  net: ionic: use netdev feature helpers
  net: jme: use netdev feature helpers
  net: micrel: use netdev feature helpers
  net: cavium: use netdev feature helpers
  net: cadence: use netdev feature helpers
  net: mediatek: use netdev feature helpers
  net: marvell: use netdev feature helpers
  net: socionext: use netdev feature helpers
  net: qlogic: use netdev feature helpers
  net: nfp: use netdev feature helpers
  net: mscc: use netdev feature helpers
  net: oki-semi: use netdev feature helpers
  net: renesas: use netdev feature helpers
  net: neterion: use netdev feature helpers
  net: cortina: use netdev feature helpers
  net: stmmac: use netdev feature helpers
  net: sxgbe: use netdev feature helpers
  net: xgmac: use netdev feature helpers
  net: altera: use netdev feature helpers
  net: ti: use netdev feature helpers
  net: benet: use netdev feature helpers
  net: amd: use netdev feature helpers
  net: bna: use netdev feature helpers
  net: enic: use netdev feature helpers
  net: 3com: use netdev feature helpers
  net: aeroflex: use netdev feature helpers
  net: sun: use netdev feature helpers
  net: mana: use netdev feature helpers
  net: myricom: use netdev feature helpers
  net: alacritech: use netdev feature helpers
  net: toshiba: use netdev feature helpers
  net: tehuti: use netdev feature helpers
  net: alteon: use netdev feature helpers
  net: ena: use netdev feature helpers
  net: sgi: use netdev feature helpers
  net: microchip: use netdev feature helpers
  net: ni: use netdev feature helpers
  net: apm: use netdev feature helpers
  net: natsemi: use netdev feature helpers
  net: xilinx: use netdev feature helpers
  net: pasemi: use netdev feature helpers
  net: rocker: use netdev feature helpers
  net: silan: use netdev feature helpers
  net: adaptec: use netdev feature helpers
  net: tundra: use netdev feature helpers
  net: via: use netdev feature helpers
  net: wiznet: use netdev feature helpers
  net: dnet: use netdev feature helpers
  net: ethoc: use netdev feature helpers
  RDMA: ipoib: use netdev feature helpers
  um: use netdev feature helpers
  scsi: fcoe: use netdev feature helpers
  net: ipvs: use netdev feature helpers
  net: xfrm: use netdev feature helpers
  net: cirrus: use netdev feature helpers
  net: ec_bhf: use netdev feature helpers
  net: hamradio: use netdev feature helpers
  net: batman: use netdev feature helpers
  net: ieee802154: use netdev feature helpers
  test_bpf: change the prototype of features
  net: openvswitch: use netdev feature helpers
  firewire: use netdev feature helpers
  staging: qlge: use netdev feature helpers
  staging: octeon: use netdev feature helpers
  net: sock: add helper sk_nocaps_add_gso()
  treewide: introduce macro __DECLARE_NETDEV_FEATURE_MASK
  net: extend the type of netdev_features_t to bitmap

 arch/um/drivers/vector_kern.c                 |  15 +-
 drivers/firewire/net.c                        |   3 +-
 drivers/infiniband/hw/hfi1/vnic_main.c        |   7 +-
 drivers/infiniband/ulp/ipoib/ipoib_cm.c       |   2 +-
 drivers/infiniband/ulp/ipoib/ipoib_ib.c       |   2 +-
 drivers/infiniband/ulp/ipoib/ipoib_main.c     |  21 +-
 drivers/infiniband/ulp/ipoib/ipoib_verbs.c    |   2 +-
 drivers/net/bareudp.c                         |  16 +-
 drivers/net/bonding/bond_main.c               | 136 +++---
 drivers/net/bonding/bond_options.c            |  12 +-
 drivers/net/caif/caif_serial.c                |   2 +-
 drivers/net/can/dev/dev.c                     |   3 +-
 drivers/net/can/slcan.c                       |   3 +-
 drivers/net/dsa/xrs700x/xrs700x.c             |   6 +-
 drivers/net/dummy.c                           |  14 +-
 drivers/net/ethernet/3com/3c59x.c             |   9 +-
 drivers/net/ethernet/3com/typhoon.c           |  10 +-
 drivers/net/ethernet/adaptec/starfire.c       |   9 +-
 drivers/net/ethernet/aeroflex/greth.c         |  12 +-
 drivers/net/ethernet/alacritech/slicoss.c     |   5 +-
 drivers/net/ethernet/alteon/acenic.c          |   7 +-
 drivers/net/ethernet/altera/altera_tse_main.c |  11 +-
 .../net/ethernet/amazon/ena/ena_admin_defs.h  |   2 +-
 drivers/net/ethernet/amazon/ena/ena_netdev.c  |  50 ++-
 drivers/net/ethernet/amd/amd8111e.c           |   8 +-
 drivers/net/ethernet/amd/xgbe/xgbe-dev.c      |  20 +-
 drivers/net/ethernet/amd/xgbe/xgbe-drv.c      |  93 ++--
 drivers/net/ethernet/amd/xgbe/xgbe-main.c     |  68 +--
 drivers/net/ethernet/amd/xgbe/xgbe.h          |   2 +-
 drivers/net/ethernet/apm/xgene-v2/main.c      |   5 +-
 .../net/ethernet/apm/xgene/xgene_enet_main.c  |  21 +-
 .../ethernet/aquantia/atlantic/aq_filters.c   |  12 +-
 .../ethernet/aquantia/atlantic/aq_macsec.c    |   2 +-
 .../net/ethernet/aquantia/atlantic/aq_main.c  |  27 +-
 .../net/ethernet/aquantia/atlantic/aq_nic.c   |  28 +-
 .../net/ethernet/aquantia/atlantic/aq_nic.h   |   2 +-
 .../net/ethernet/aquantia/atlantic/aq_ring.c  |   3 +-
 .../aquantia/atlantic/hw_atl/hw_atl_b0.c      |   3 +-
 drivers/net/ethernet/atheros/alx/main.c       |  23 +-
 .../net/ethernet/atheros/atl1c/atl1c_main.c   |  33 +-
 .../net/ethernet/atheros/atl1e/atl1e_main.c   |  44 +-
 drivers/net/ethernet/atheros/atlx/atl1.c      |  15 +-
 drivers/net/ethernet/atheros/atlx/atl2.c      |  27 +-
 drivers/net/ethernet/atheros/atlx/atlx.c      |  20 +-
 drivers/net/ethernet/broadcom/b44.c           |   2 +-
 drivers/net/ethernet/broadcom/bcmsysport.c    |  24 +-
 drivers/net/ethernet/broadcom/bgmac.c         |   8 +-
 drivers/net/ethernet/broadcom/bnx2.c          |  48 +-
 .../net/ethernet/broadcom/bnx2x/bnx2x_cmn.c   |  67 +--
 .../net/ethernet/broadcom/bnx2x/bnx2x_cmn.h   |   3 +-
 .../net/ethernet/broadcom/bnx2x/bnx2x_main.c  |  93 ++--
 drivers/net/ethernet/broadcom/bnxt/bnxt.c     | 171 ++++---
 drivers/net/ethernet/broadcom/bnxt/bnxt_tc.c  |   4 +-
 drivers/net/ethernet/broadcom/bnxt/bnxt_vfr.c |  11 +-
 drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.c |   3 +-
 .../net/ethernet/broadcom/genet/bcmgenet.c    |  12 +-
 drivers/net/ethernet/broadcom/tg3.c           |  62 +--
 drivers/net/ethernet/brocade/bna/bnad.c       |  48 +-
 drivers/net/ethernet/cadence/macb_main.c      |  81 ++--
 drivers/net/ethernet/calxeda/xgmac.c          |  20 +-
 .../net/ethernet/cavium/liquidio/lio_core.c   |   6 +-
 .../net/ethernet/cavium/liquidio/lio_main.c   |  95 ++--
 .../ethernet/cavium/liquidio/lio_vf_main.c    |  79 ++--
 .../net/ethernet/cavium/thunder/nicvf_main.c  |  50 ++-
 .../ethernet/cavium/thunder/nicvf_queues.c    |   2 +-
 drivers/net/ethernet/chelsio/cxgb/cxgb2.c     |  50 ++-
 drivers/net/ethernet/chelsio/cxgb/sge.c       |  12 +-
 .../net/ethernet/chelsio/cxgb3/cxgb3_main.c   |  60 ++-
 drivers/net/ethernet/chelsio/cxgb3/sge.c      |  10 +-
 .../net/ethernet/chelsio/cxgb4/cxgb4_fcoe.c   |  16 +-
 .../net/ethernet/chelsio/cxgb4/cxgb4_main.c   |  88 ++--
 .../net/ethernet/chelsio/cxgb4/cxgb4_tc_u32.h |   3 +-
 drivers/net/ethernet/chelsio/cxgb4/sge.c      |  14 +-
 .../ethernet/chelsio/cxgb4vf/cxgb4vf_main.c   |  39 +-
 drivers/net/ethernet/chelsio/cxgb4vf/sge.c    |   8 +-
 .../chelsio/inline_crypto/chtls/chtls_main.c  |   3 +-
 drivers/net/ethernet/cirrus/ep93xx_eth.c      |   2 +-
 drivers/net/ethernet/cisco/enic/enic_main.c   |  74 +--
 drivers/net/ethernet/cortina/gemini.c         |  18 +-
 drivers/net/ethernet/davicom/dm9000.c         |  24 +-
 drivers/net/ethernet/dnet.c                   |   2 +-
 drivers/net/ethernet/ec_bhf.c                 |   2 +-
 drivers/net/ethernet/emulex/benet/be_main.c   |  68 +--
 drivers/net/ethernet/ethoc.c                  |   2 +-
 drivers/net/ethernet/faraday/ftgmac100.c      |  38 +-
 .../net/ethernet/freescale/dpaa/dpaa_eth.c    |  23 +-
 .../net/ethernet/freescale/dpaa2/dpaa2-eth.c  |  44 +-
 .../ethernet/freescale/dpaa2/dpaa2-switch.c   |   5 +-
 drivers/net/ethernet/freescale/enetc/enetc.c  |  34 +-
 .../net/ethernet/freescale/enetc/enetc_pf.c   |  38 +-
 .../net/ethernet/freescale/enetc/enetc_qos.c  |   2 +-
 .../net/ethernet/freescale/enetc/enetc_vf.c   |  16 +-
 drivers/net/ethernet/freescale/fec_main.c     |  29 +-
 .../ethernet/freescale/fs_enet/fs_enet-main.c |   2 +-
 drivers/net/ethernet/freescale/gianfar.c      |  39 +-
 .../net/ethernet/freescale/gianfar_ethtool.c  |  11 +-
 drivers/net/ethernet/google/gve/gve_adminq.c  |   6 +-
 drivers/net/ethernet/google/gve/gve_main.c    |  31 +-
 drivers/net/ethernet/google/gve/gve_rx.c      |   8 +-
 drivers/net/ethernet/google/gve/gve_rx_dqo.c  |   8 +-
 drivers/net/ethernet/hisilicon/hix5hd2_gmac.c |   8 +-
 drivers/net/ethernet/hisilicon/hns/hns_enet.c |  52 ++-
 .../net/ethernet/hisilicon/hns3/hns3_enet.c   | 109 +++--
 .../net/ethernet/huawei/hinic/hinic_main.c    |  80 ++--
 drivers/net/ethernet/huawei/hinic/hinic_rx.c  |   5 +-
 drivers/net/ethernet/ibm/ehea/ehea_main.c     |  22 +-
 drivers/net/ethernet/ibm/emac/core.c          |   7 +-
 drivers/net/ethernet/ibm/ibmveth.c            |  76 ++--
 drivers/net/ethernet/ibm/ibmvnic.c            |  69 +--
 drivers/net/ethernet/intel/e100.c             |  26 +-
 drivers/net/ethernet/intel/e1000/e1000_main.c |  82 ++--
 drivers/net/ethernet/intel/e1000e/netdev.c    | 129 +++---
 drivers/net/ethernet/intel/fm10k/fm10k_main.c |   9 +-
 .../net/ethernet/intel/fm10k/fm10k_netdev.c   |  65 +--
 drivers/net/ethernet/intel/fm10k/fm10k_pci.c  |  11 +-
 drivers/net/ethernet/intel/i40e/i40e_main.c   | 122 ++---
 drivers/net/ethernet/intel/i40e/i40e_txrx.c   |   9 +-
 drivers/net/ethernet/intel/iavf/iavf_main.c   | 153 ++++---
 drivers/net/ethernet/intel/iavf/iavf_txrx.c   |  12 +-
 drivers/net/ethernet/intel/ice/ice_main.c     | 177 +++++---
 drivers/net/ethernet/intel/ice/ice_txrx_lib.c |   9 +-
 drivers/net/ethernet/intel/igb/igb_ethtool.c  |   2 +-
 drivers/net/ethernet/intel/igb/igb_main.c     | 150 ++++---
 drivers/net/ethernet/intel/igbvf/netdev.c     |  86 ++--
 drivers/net/ethernet/intel/igc/igc_ethtool.c  |   3 +-
 drivers/net/ethernet/intel/igc/igc_main.c     | 130 +++---
 drivers/net/ethernet/intel/ixgb/ixgb_main.c   |  49 +-
 .../net/ethernet/intel/ixgbe/ixgbe_dcb_nl.c   |   3 +-
 .../net/ethernet/intel/ixgbe/ixgbe_ethtool.c  |   2 +-
 drivers/net/ethernet/intel/ixgbe/ixgbe_fcoe.c |   7 +-
 drivers/net/ethernet/intel/ixgbe/ixgbe_lib.c  |   3 +-
 drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 215 +++++----
 .../net/ethernet/intel/ixgbe/ixgbe_sriov.c    |   6 +-
 drivers/net/ethernet/intel/ixgbevf/ipsec.c    |   9 +-
 .../net/ethernet/intel/ixgbevf/ixgbevf_main.c | 105 +++--
 drivers/net/ethernet/jme.c                    |  49 +-
 drivers/net/ethernet/marvell/mv643xx_eth.c    |  14 +-
 drivers/net/ethernet/marvell/mvneta.c         |  22 +-
 .../net/ethernet/marvell/mvpp2/mvpp2_main.c   |  73 +--
 .../marvell/octeontx2/nic/otx2_common.c       |   3 +-
 .../marvell/octeontx2/nic/otx2_ethtool.c      |   4 +-
 .../ethernet/marvell/octeontx2/nic/otx2_pf.c  |  85 ++--
 .../marvell/octeontx2/nic/otx2_txrx.c         |   8 +-
 .../ethernet/marvell/octeontx2/nic/otx2_vf.c  |  35 +-
 .../ethernet/marvell/prestera/prestera_main.c |   3 +-
 drivers/net/ethernet/marvell/skge.c           |  11 +-
 drivers/net/ethernet/marvell/sky2.c           |  75 ++--
 drivers/net/ethernet/mediatek/mtk_eth_soc.c   |  68 +--
 drivers/net/ethernet/mediatek/mtk_eth_soc.h   |   2 +-
 .../net/ethernet/mellanox/mlx4/en_ethtool.c   |   4 +-
 drivers/net/ethernet/mellanox/mlx4/en_main.c  |   8 +-
 .../net/ethernet/mellanox/mlx4/en_netdev.c    | 198 +++++----
 .../net/ethernet/mellanox/mlx4/en_resources.c |   4 +-
 drivers/net/ethernet/mellanox/mlx4/en_rx.c    |  19 +-
 drivers/net/ethernet/mellanox/mlx4/mlx4_en.h  |   5 +-
 drivers/net/ethernet/mellanox/mlx5/core/en.h  |   5 +-
 .../mellanox/mlx5/core/en_accel/ipsec.c       |  18 +-
 .../mellanox/mlx5/core/en_accel/ipsec_rxtx.h  |  15 +-
 .../mellanox/mlx5/core/en_accel/ktls.c        |  14 +-
 .../mellanox/mlx5/core/en_accel/tls.c         |  16 +-
 .../net/ethernet/mellanox/mlx5/core/en_arfs.c |   6 +-
 .../ethernet/mellanox/mlx5/core/en_ethtool.c  |   3 +-
 .../net/ethernet/mellanox/mlx5/core/en_fs.c   |   3 +-
 .../net/ethernet/mellanox/mlx5/core/en_main.c | 250 ++++++-----
 .../net/ethernet/mellanox/mlx5/core/en_rep.c  |  25 +-
 .../net/ethernet/mellanox/mlx5/core/en_rx.c   |   9 +-
 .../net/ethernet/mellanox/mlx5/core/en_tc.c   |   2 +-
 .../ethernet/mellanox/mlx5/core/ipoib/ipoib.c |  19 +-
 .../net/ethernet/mellanox/mlxsw/spectrum.c    |  36 +-
 drivers/net/ethernet/micrel/ksz884x.c         |  10 +-
 drivers/net/ethernet/microchip/lan743x_main.c |   7 +-
 .../ethernet/microchip/sparx5/sparx5_fdma.c   |   2 +-
 drivers/net/ethernet/microsoft/mana/mana_en.c |  20 +-
 drivers/net/ethernet/mscc/ocelot.c            |   3 +-
 drivers/net/ethernet/mscc/ocelot_net.c        |  18 +-
 .../net/ethernet/myricom/myri10ge/myri10ge.c  |  31 +-
 drivers/net/ethernet/natsemi/ns83820.c        |  13 +-
 drivers/net/ethernet/neterion/s2io.c          |  32 +-
 .../net/ethernet/neterion/vxge/vxge-main.c    |  49 +-
 .../net/ethernet/netronome/nfp/crypto/tls.c   |  12 +-
 .../ethernet/netronome/nfp/nfp_net_common.c   | 118 +++--
 .../net/ethernet/netronome/nfp/nfp_net_repr.c |  76 ++--
 drivers/net/ethernet/netronome/nfp/nfp_port.c |   3 +-
 drivers/net/ethernet/ni/nixge.c               |   3 +-
 drivers/net/ethernet/nvidia/forcedeth.c       |  66 +--
 .../ethernet/oki-semi/pch_gbe/pch_gbe_main.c  |  13 +-
 .../ethernet/oki-semi/pch_gbe/pch_gbe_param.c |   6 +-
 drivers/net/ethernet/pasemi/pasemi_mac.c      |   6 +-
 .../net/ethernet/pensando/ionic/ionic_lif.c   | 125 +++---
 .../net/ethernet/pensando/ionic/ionic_txrx.c  |   9 +-
 .../ethernet/qlogic/netxen/netxen_nic_init.c  |   3 +-
 .../ethernet/qlogic/netxen/netxen_nic_main.c  |  50 ++-
 drivers/net/ethernet/qlogic/qede/qede.h       |  10 +-
 .../net/ethernet/qlogic/qede/qede_ethtool.c   |   2 +-
 .../net/ethernet/qlogic/qede/qede_filter.c    |  19 +-
 drivers/net/ethernet/qlogic/qede/qede_fp.c    |  23 +-
 drivers/net/ethernet/qlogic/qede/qede_main.c  |  67 +--
 drivers/net/ethernet/qlogic/qla3xxx.c         |   5 +-
 drivers/net/ethernet/qlogic/qlcnic/qlcnic.h   |   3 +-
 .../net/ethernet/qlogic/qlcnic/qlcnic_hw.c    |  72 +--
 .../net/ethernet/qlogic/qlcnic/qlcnic_io.c    |   3 +-
 .../net/ethernet/qlogic/qlcnic/qlcnic_main.c  |  57 +--
 drivers/net/ethernet/qualcomm/emac/emac-mac.c |   6 +-
 drivers/net/ethernet/qualcomm/emac/emac.c     |  25 +-
 .../ethernet/qualcomm/rmnet/rmnet_map_data.c  |  11 +-
 .../net/ethernet/qualcomm/rmnet/rmnet_vnd.c   |  10 +-
 drivers/net/ethernet/realtek/8139cp.c         |  39 +-
 drivers/net/ethernet/realtek/8139too.c        |  28 +-
 drivers/net/ethernet/realtek/r8169_main.c     |  85 ++--
 drivers/net/ethernet/renesas/ravb.h           |   4 +-
 drivers/net/ethernet/renesas/ravb_main.c      |  24 +-
 drivers/net/ethernet/renesas/sh_eth.c         |  28 +-
 drivers/net/ethernet/rocker/rocker_main.c     |   3 +-
 .../net/ethernet/samsung/sxgbe/sxgbe_main.c   |  19 +-
 drivers/net/ethernet/sfc/ef10.c               |  41 +-
 drivers/net/ethernet/sfc/ef100_nic.c          |  43 +-
 drivers/net/ethernet/sfc/ef100_rx.c           |   6 +-
 drivers/net/ethernet/sfc/ef100_tx.c           |  11 +-
 drivers/net/ethernet/sfc/ef10_sriov.c         |   6 +-
 drivers/net/ethernet/sfc/efx.c                |  32 +-
 drivers/net/ethernet/sfc/efx_common.c         |  41 +-
 drivers/net/ethernet/sfc/efx_common.h         |   4 +-
 drivers/net/ethernet/sfc/falcon/efx.c         |  44 +-
 drivers/net/ethernet/sfc/falcon/falcon.c      |   4 +-
 drivers/net/ethernet/sfc/falcon/net_driver.h  |   9 +-
 drivers/net/ethernet/sfc/falcon/rx.c          |   6 +-
 drivers/net/ethernet/sfc/farch.c              |   2 +-
 drivers/net/ethernet/sfc/mcdi_filters.c       |  21 +-
 drivers/net/ethernet/sfc/mcdi_port_common.c   |   3 +-
 drivers/net/ethernet/sfc/net_driver.h         |   9 +-
 drivers/net/ethernet/sfc/rx.c                 |   3 +-
 drivers/net/ethernet/sfc/rx_common.c          |   5 +-
 drivers/net/ethernet/sfc/siena.c              |   4 +-
 drivers/net/ethernet/sgi/ioc3-eth.c           |  11 +-
 drivers/net/ethernet/silan/sc92031.c          |   6 +-
 drivers/net/ethernet/socionext/netsec.c       |  10 +-
 drivers/net/ethernet/socionext/sni_ave.c      |   6 +-
 .../net/ethernet/stmicro/stmmac/dwmac4_core.c |   3 +-
 .../net/ethernet/stmicro/stmmac/stmmac_main.c |  62 +--
 .../stmicro/stmmac/stmmac_selftests.c         |   6 +-
 drivers/net/ethernet/sun/cassini.c            |   5 +-
 drivers/net/ethernet/sun/ldmvsw.c             |   5 +-
 drivers/net/ethernet/sun/niu.c                |  11 +-
 drivers/net/ethernet/sun/sungem.c             |  11 +-
 drivers/net/ethernet/sun/sunhme.c             |  14 +-
 drivers/net/ethernet/sun/sunvnet.c            |   8 +-
 drivers/net/ethernet/sun/sunvnet_common.c     |   5 +-
 .../net/ethernet/synopsys/dwc-xlgmac-common.c |  45 +-
 drivers/net/ethernet/synopsys/dwc-xlgmac-hw.c |  20 +-
 .../net/ethernet/synopsys/dwc-xlgmac-net.c    |  41 +-
 drivers/net/ethernet/synopsys/dwc-xlgmac.h    |   2 +-
 drivers/net/ethernet/tehuti/tehuti.c          |  22 +-
 drivers/net/ethernet/ti/am65-cpsw-nuss.c      |  23 +-
 drivers/net/ethernet/ti/cpsw.c                |   6 +-
 drivers/net/ethernet/ti/cpsw_new.c            |   7 +-
 drivers/net/ethernet/ti/netcp_core.c          |   9 +-
 drivers/net/ethernet/toshiba/ps3_gelic_net.c  |  14 +-
 drivers/net/ethernet/toshiba/spider_net.c     |  12 +-
 drivers/net/ethernet/tundra/tsi108_eth.c      |   3 +-
 drivers/net/ethernet/via/via-rhine.c          |  10 +-
 drivers/net/ethernet/via/via-velocity.c       |  14 +-
 drivers/net/ethernet/wiznet/w5100.c           |   2 +-
 drivers/net/ethernet/wiznet/w5300.c           |   2 +-
 drivers/net/ethernet/xilinx/ll_temac_main.c   |   5 +-
 .../net/ethernet/xilinx/xilinx_axienet_main.c |   9 +-
 drivers/net/fjes/fjes_main.c                  |   3 +-
 drivers/net/geneve.c                          |  18 +-
 drivers/net/gtp.c                             |   2 +-
 drivers/net/hamradio/bpqether.c               |   3 +-
 drivers/net/hyperv/netvsc_bpf.c               |   2 +-
 drivers/net/hyperv/netvsc_drv.c               |  46 +-
 drivers/net/hyperv/rndis_filter.c             |  30 +-
 drivers/net/ifb.c                             |  12 +-
 drivers/net/ipa/ipa_modem.c                   |   3 +-
 drivers/net/ipvlan/ipvlan.h                   |   2 +-
 drivers/net/ipvlan/ipvlan_main.c              |  40 +-
 drivers/net/ipvlan/ipvtap.c                   |   5 +-
 drivers/net/loopback.c                        |  20 +-
 drivers/net/macsec.c                          |  49 +-
 drivers/net/macvlan.c                         |  58 ++-
 drivers/net/macvtap.c                         |   5 +-
 drivers/net/net_failover.c                    |  77 ++--
 drivers/net/netdevsim/ipsec.c                 |   4 +-
 drivers/net/netdevsim/netdev.c                |  15 +-
 drivers/net/nlmon.c                           |   5 +-
 drivers/net/ntb_netdev.c                      |   5 +-
 drivers/net/ppp/ppp_generic.c                 |   2 +-
 drivers/net/rionet.c                          |   3 +-
 drivers/net/tap.c                             |  38 +-
 drivers/net/team/team.c                       |  90 ++--
 drivers/net/thunderbolt.c                     |  10 +-
 drivers/net/tun.c                             |  53 ++-
 drivers/net/usb/aqc111.c                      |  42 +-
 drivers/net/usb/ax88179_178a.c                |  26 +-
 drivers/net/usb/cdc-phonet.c                  |   2 +-
 drivers/net/usb/cdc_mbim.c                    |   4 +-
 drivers/net/usb/lan78xx.c                     |  46 +-
 drivers/net/usb/r8152.c                       |  71 +--
 drivers/net/usb/smsc75xx.c                    |  14 +-
 drivers/net/usb/smsc95xx.c                    |  19 +-
 drivers/net/veth.c                            |  67 +--
 drivers/net/virtio_net.c                      |  50 ++-
 drivers/net/vmxnet3/vmxnet3_drv.c             |  75 ++--
 drivers/net/vmxnet3/vmxnet3_ethtool.c         |  88 ++--
 drivers/net/vmxnet3/vmxnet3_int.h             |   4 +-
 drivers/net/vrf.c                             |  20 +-
 drivers/net/vsockmon.c                        |   6 +-
 drivers/net/vxlan.c                           |  21 +-
 drivers/net/wireguard/device.c                |   8 +-
 drivers/net/wireless/ath/ath10k/mac.c         |   7 +-
 drivers/net/wireless/ath/ath11k/mac.c         |   4 +-
 drivers/net/wireless/ath/ath6kl/main.c        |  15 +-
 drivers/net/wireless/ath/ath6kl/txrx.c        |   6 +-
 drivers/net/wireless/ath/wil6210/netdev.c     |  11 +-
 .../broadcom/brcm80211/brcmfmac/core.c        |   4 +-
 .../net/wireless/intel/iwlwifi/cfg/22000.c    |   4 +-
 drivers/net/wireless/intel/iwlwifi/cfg/8000.c |   2 +-
 drivers/net/wireless/intel/iwlwifi/cfg/9000.c |   2 +-
 .../net/wireless/intel/iwlwifi/dvm/mac80211.c |   7 +-
 .../net/wireless/intel/iwlwifi/iwl-config.h   |   2 +-
 .../net/wireless/intel/iwlwifi/mvm/mac80211.c |  17 +-
 drivers/net/wireless/intel/iwlwifi/mvm/rx.c   |   2 +-
 drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c |   2 +-
 drivers/net/wireless/intel/iwlwifi/mvm/tx.c   |  10 +-
 .../net/wireless/mediatek/mt76/mt7615/init.c  |   3 +-
 .../net/wireless/mediatek/mt76/mt7915/init.c  |   3 +-
 .../net/wireless/mediatek/mt76/mt7921/init.c  |   3 +-
 drivers/net/xen-netback/interface.c           |  27 +-
 drivers/net/xen-netfront.c                    |  53 ++-
 drivers/s390/net/qeth_core.h                  |   7 +-
 drivers/s390/net/qeth_core_main.c             | 139 +++---
 drivers/s390/net/qeth_l2_main.c               |  44 +-
 drivers/s390/net/qeth_l3_main.c               |  49 +-
 drivers/scsi/fcoe/fcoe.c                      |  14 +-
 drivers/staging/octeon/ethernet.c             |   5 +-
 drivers/staging/qlge/qlge_main.c              |  59 +--
 drivers/usb/gadget/function/f_phonet.c        |   2 +-
 include/linux/if_macvlan.h                    |   2 +-
 include/linux/if_tap.h                        |   2 +-
 include/linux/if_vlan.h                       |  22 +-
 include/linux/netdev_features.h               | 181 +++++++-
 include/linux/netdevice.h                     | 122 ++---
 include/linux/skbuff.h                        |   6 +-
 include/net/mac80211.h                        |   2 +-
 include/net/pkt_cls.h                         |   2 +-
 include/net/sock.h                            |  22 +-
 include/net/udp.h                             |   8 +-
 include/net/udp_tunnel.h                      |  12 +-
 include/net/vxlan.h                           |  13 +-
 lib/test_bpf.c                                |  21 +-
 lib/vsprintf.c                                |   4 +-
 net/8021q/vlan.c                              |  11 +-
 net/8021q/vlan.h                              |  24 +-
 net/8021q/vlan_core.c                         |   6 +-
 net/8021q/vlan_dev.c                          |  57 +--
 net/batman-adv/soft-interface.c               |   5 +-
 net/bridge/br_device.c                        |  21 +-
 net/bridge/br_if.c                            |  19 +-
 net/bridge/br_private.h                       |   3 +-
 net/core/dev.c                                | 420 ++++++++++--------
 net/core/netpoll.c                            |   4 +-
 net/core/pktgen.c                             |   6 +-
 net/core/skbuff.c                             |  10 +-
 net/core/skmsg.c                              |   3 +-
 net/core/sock.c                               |  17 +-
 net/dccp/ipv4.c                               |   2 +-
 net/dccp/ipv6.c                               |   7 +-
 net/decnet/af_decnet.c                        |   2 +-
 net/decnet/dn_nsp_out.c                       |   2 +-
 net/dsa/slave.c                               |  24 +-
 net/ethtool/features.c                        |  94 ++--
 net/ethtool/ioctl.c                           | 158 ++++---
 net/hsr/hsr_device.c                          |  40 +-
 net/hsr/hsr_forward.c                         |  11 +-
 net/hsr/hsr_framereg.c                        |   3 +-
 net/hsr/hsr_slave.c                           |   3 +-
 net/ieee802154/6lowpan/core.c                 |   2 +-
 net/ieee802154/core.c                         |  14 +-
 net/ipv4/af_inet.c                            |   5 +-
 net/ipv4/esp4_offload.c                       |  28 +-
 net/ipv4/gre_offload.c                        |   8 +-
 net/ipv4/ip_gre.c                             |  36 +-
 net/ipv4/ip_output.c                          |  27 +-
 net/ipv4/ip_tunnel.c                          |   4 +-
 net/ipv4/ip_vti.c                             |   2 +-
 net/ipv4/ipip.c                               |   6 +-
 net/ipv4/ipmr.c                               |   2 +-
 net/ipv4/tcp.c                                |  10 +-
 net/ipv4/tcp_ipv4.c                           |   6 +-
 net/ipv4/tcp_offload.c                        |   5 +-
 net/ipv4/tcp_output.c                         |   2 +-
 net/ipv4/udp_offload.c                        |  28 +-
 net/ipv6/af_inet6.c                           |   2 +-
 net/ipv6/esp6_offload.c                       |  19 +-
 net/ipv6/inet6_connection_sock.c              |   2 +-
 net/ipv6/ip6_gre.c                            |  17 +-
 net/ipv6/ip6_offload.c                        |   3 +-
 net/ipv6/ip6_output.c                         |  24 +-
 net/ipv6/ip6_tunnel.c                         |   9 +-
 net/ipv6/ip6mr.c                              |   2 +-
 net/ipv6/sit.c                                |   9 +-
 net/ipv6/tcp_ipv6.c                           |   2 +-
 net/ipv6/udp_offload.c                        |   2 +-
 net/l2tp/l2tp_eth.c                           |   2 +-
 net/mac80211/iface.c                          |  10 +-
 net/mac80211/main.c                           |   5 +-
 net/mpls/mpls_gso.c                           |   4 +-
 net/netfilter/ipvs/ip_vs_proto_sctp.c         |   3 +-
 net/nsh/nsh.c                                 |   2 +-
 net/openvswitch/datapath.c                    |   5 +-
 net/openvswitch/vport-internal_dev.c          |  24 +-
 net/phonet/pep-gprs.c                         |   3 +-
 net/sched/sch_cake.c                          |   6 +-
 net/sched/sch_netem.c                         |   6 +-
 net/sched/sch_taprio.c                        |   6 +-
 net/sched/sch_tbf.c                           |   6 +-
 net/sctp/offload.c                            |  12 +-
 net/sctp/output.c                             |   2 +-
 net/sunrpc/sunrpc.h                           |   2 +-
 net/tls/tls_device.c                          |  10 +-
 net/wireless/core.c                           |  14 +-
 net/xfrm/xfrm_device.c                        |  31 +-
 net/xfrm/xfrm_interface.c                     |   6 +-
 net/xfrm/xfrm_output.c                        |   4 +-
 424 files changed, 6584 insertions(+), 4459 deletions(-)

-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 001/167] net: convert the prototype of netdev_intersect_features
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
@ 2021-09-29 15:50 ` Jian Shen
  2021-09-29 15:50 ` [RFCv2 net-next 002/167] net: convert the prototype of netdev_get_wanted_features Jian Shen
                   ` (169 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:50 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

For the origin type for netdev_features_t would be changed to
be unsigned long * from u64, so changes the prototype of
netdev_intersect_features for adaption.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 drivers/net/ethernet/netronome/nfp/nfp_net_repr.c | 2 +-
 include/linux/netdevice.h                         | 7 ++++---
 net/8021q/vlan_dev.c                              | 8 ++++----
 net/core/dev.c                                    | 8 ++++----
 4 files changed, 13 insertions(+), 12 deletions(-)

diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_repr.c b/drivers/net/ethernet/netronome/nfp/nfp_net_repr.c
index 369f6ae700c7..f04b79f04a9d 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_net_repr.c
+++ b/drivers/net/ethernet/netronome/nfp/nfp_net_repr.c
@@ -246,7 +246,7 @@ nfp_repr_fix_features(struct net_device *netdev, netdev_features_t features)
 	if (lower_features & (NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM))
 		lower_features |= NETIF_F_HW_CSUM;
 
-	features = netdev_intersect_features(features, lower_features);
+	netdev_intersect_features(&features, features, lower_features);
 	features |= old_features & (NETIF_F_SOFT_FEATURES | NETIF_F_HW_TC);
 	features |= NETIF_F_LLTX;
 
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index d79163208dfd..5d5129d4791f 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -5017,8 +5017,9 @@ const char *netdev_drivername(const struct net_device *dev);
 
 void linkwatch_run_queue(void);
 
-static inline netdev_features_t netdev_intersect_features(netdev_features_t f1,
-							  netdev_features_t f2)
+static inline void netdev_intersect_features(netdev_features_t *ret,
+					     netdev_features_t f1,
+					     netdev_features_t f2)
 {
 	if ((f1 ^ f2) & NETIF_F_HW_CSUM) {
 		if (f1 & NETIF_F_HW_CSUM)
@@ -5027,7 +5028,7 @@ static inline netdev_features_t netdev_intersect_features(netdev_features_t f1,
 			f2 |= (NETIF_F_IP_CSUM|NETIF_F_IPV6_CSUM);
 	}
 
-	return f1 & f2;
+	*ret = f1 & f2;
 }
 
 static inline netdev_features_t netdev_get_wanted_features(
diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c
index 0c21d1fec852..2987201ec93d 100644
--- a/net/8021q/vlan_dev.c
+++ b/net/8021q/vlan_dev.c
@@ -640,16 +640,16 @@ static netdev_features_t vlan_dev_fix_features(struct net_device *dev,
 	netdev_features_t old_features = features;
 	netdev_features_t lower_features;
 
-	lower_features = netdev_intersect_features((real_dev->vlan_features |
-						    NETIF_F_RXCSUM),
-						   real_dev->features);
+	netdev_intersect_features(&lower_features,
+				  (real_dev->vlan_features | NETIF_F_RXCSUM),
+				  real_dev->features);
 
 	/* Add HW_CSUM setting to preserve user ability to control
 	 * checksum offload on the vlan device.
 	 */
 	if (lower_features & (NETIF_F_IP_CSUM|NETIF_F_IPV6_CSUM))
 		lower_features |= NETIF_F_HW_CSUM;
-	features = netdev_intersect_features(features, lower_features);
+	netdev_intersect_features(&features, features, lower_features);
 	features |= old_features & (NETIF_F_SOFT_FEATURES | NETIF_F_GSO_SOFTWARE);
 	features |= NETIF_F_LLTX;
 
diff --git a/net/core/dev.c b/net/core/dev.c
index fa989ab63f29..6ee56cd01d53 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -3548,10 +3548,10 @@ netdev_features_t netif_skb_features(struct sk_buff *skb)
 		features &= dev->hw_enc_features;
 
 	if (skb_vlan_tagged(skb))
-		features = netdev_intersect_features(features,
-						     dev->vlan_features |
-						     NETIF_F_HW_VLAN_CTAG_TX |
-						     NETIF_F_HW_VLAN_STAG_TX);
+		netdev_intersect_features(&features, features,
+					  dev->vlan_features |
+					  NETIF_F_HW_VLAN_CTAG_TX |
+					  NETIF_F_HW_VLAN_STAG_TX);
 
 	if (dev->netdev_ops->ndo_features_check)
 		features &= dev->netdev_ops->ndo_features_check(skb, dev,
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 002/167] net: convert the prototype of netdev_get_wanted_features
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
  2021-09-29 15:50 ` [RFCv2 net-next 001/167] net: convert the prototype of netdev_intersect_features Jian Shen
@ 2021-09-29 15:50 ` Jian Shen
  2021-09-29 15:50 ` [RFCv2 net-next 003/167] net: convert the prototype of net_mpls_features Jian Shen
                   ` (168 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:50 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

For the origin type for netdev_features_t would be changed to
be unsigned long * from u64, so changes the prototype of
netdev_get_wanted_features for adaption.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 include/linux/netdevice.h | 6 +++---
 net/core/dev.c            | 2 +-
 2 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 5d5129d4791f..a5598d617789 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -5031,10 +5031,10 @@ static inline void netdev_intersect_features(netdev_features_t *ret,
 	*ret = f1 & f2;
 }
 
-static inline netdev_features_t netdev_get_wanted_features(
-	struct net_device *dev)
+static inline void netdev_get_wanted_features(struct net_device *dev,
+					      netdev_features_t *wanted)
 {
-	return (dev->features & ~dev->hw_features) | dev->wanted_features;
+	*wanted = (dev->features & ~dev->hw_features) | dev->wanted_features;
 }
 netdev_features_t netdev_increment_features(netdev_features_t all,
 	netdev_features_t one, netdev_features_t mask);
diff --git a/net/core/dev.c b/net/core/dev.c
index 6ee56cd01d53..e2b4ae74c999 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -9939,7 +9939,7 @@ int __netdev_update_features(struct net_device *dev)
 
 	ASSERT_RTNL();
 
-	features = netdev_get_wanted_features(dev);
+	netdev_get_wanted_features(dev, &features);
 
 	if (dev->netdev_ops->ndo_fix_features)
 		features = dev->netdev_ops->ndo_fix_features(dev, features);
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 003/167] net: convert the prototype of net_mpls_features
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
  2021-09-29 15:50 ` [RFCv2 net-next 001/167] net: convert the prototype of netdev_intersect_features Jian Shen
  2021-09-29 15:50 ` [RFCv2 net-next 002/167] net: convert the prototype of netdev_get_wanted_features Jian Shen
@ 2021-09-29 15:50 ` Jian Shen
  2021-09-29 15:50 ` [RFCv2 net-next 004/167] net: convert the prototype of harmonize_features Jian Shen
                   ` (167 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:50 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

For the origin type for netdev_features_t would be changed to
be unsigned long * from u64, so changes the prototype of
net_mpls_features for adaption.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 net/core/dev.c | 17 ++++++-----------
 1 file changed, 6 insertions(+), 11 deletions(-)

diff --git a/net/core/dev.c b/net/core/dev.c
index e2b4ae74c999..81e93be4cedb 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -3444,21 +3444,16 @@ static int illegal_highdma(struct net_device *dev, struct sk_buff *skb)
  * instead of standard features for the netdev.
  */
 #if IS_ENABLED(CONFIG_NET_MPLS_GSO)
-static netdev_features_t net_mpls_features(struct sk_buff *skb,
-					   netdev_features_t features,
-					   __be16 type)
+static void net_mpls_features(struct sk_buff *skb, netdev_features_t *features,
+			      __be16 type)
 {
 	if (eth_p_mpls(type))
-		features &= skb->dev->mpls_features;
-
-	return features;
+		*features &= skb->dev->mpls_features;
 }
 #else
-static netdev_features_t net_mpls_features(struct sk_buff *skb,
-					   netdev_features_t features,
-					   __be16 type)
+static void net_mpls_features(struct sk_buff *skb, netdev_features_t *features,
+			      __be16 type)
 {
-	return features;
 }
 #endif
 
@@ -3468,7 +3463,7 @@ static netdev_features_t harmonize_features(struct sk_buff *skb,
 	__be16 type;
 
 	type = skb_network_protocol(skb, NULL);
-	features = net_mpls_features(skb, features, type);
+	net_mpls_features(skb, &features, type);
 
 	if (skb->ip_summed != CHECKSUM_NONE &&
 	    !can_checksum_protocol(features, type)) {
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 004/167] net: convert the prototype of harmonize_features
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (2 preceding siblings ...)
  2021-09-29 15:50 ` [RFCv2 net-next 003/167] net: convert the prototype of net_mpls_features Jian Shen
@ 2021-09-29 15:50 ` Jian Shen
  2021-09-29 15:50 ` [RFCv2 net-next 005/167] net: convert the prototype of gso_features_check Jian Shen
                   ` (166 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:50 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

For the origin type for netdev_features_t would be changed to
be unsigned long * from u64, so changes the prototype of
harmonize_features for adaption.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 net/core/dev.c | 16 +++++++---------
 1 file changed, 7 insertions(+), 9 deletions(-)

diff --git a/net/core/dev.c b/net/core/dev.c
index 81e93be4cedb..ef77e080504b 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -3457,22 +3457,19 @@ static void net_mpls_features(struct sk_buff *skb, netdev_features_t *features,
 }
 #endif
 
-static netdev_features_t harmonize_features(struct sk_buff *skb,
-	netdev_features_t features)
+static void harmonize_features(struct sk_buff *skb, netdev_features_t *features)
 {
 	__be16 type;
 
 	type = skb_network_protocol(skb, NULL);
-	net_mpls_features(skb, &features, type);
+	net_mpls_features(skb, features, type);
 
 	if (skb->ip_summed != CHECKSUM_NONE &&
-	    !can_checksum_protocol(features, type)) {
-		features &= ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
+	    !can_checksum_protocol(*features, type)) {
+		*features &= ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
 	}
 	if (illegal_highdma(skb->dev, skb))
-		features &= ~NETIF_F_SG;
-
-	return features;
+		*features &= ~NETIF_F_SG;
 }
 
 netdev_features_t passthru_features_check(struct sk_buff *skb,
@@ -3554,7 +3551,8 @@ netdev_features_t netif_skb_features(struct sk_buff *skb)
 	else
 		features &= dflt_features_check(skb, dev, features);
 
-	return harmonize_features(skb, features);
+	harmonize_features(skb, &features);
+	return features;
 }
 EXPORT_SYMBOL(netif_skb_features);
 
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 005/167] net: convert the prototype of gso_features_check
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (3 preceding siblings ...)
  2021-09-29 15:50 ` [RFCv2 net-next 004/167] net: convert the prototype of harmonize_features Jian Shen
@ 2021-09-29 15:50 ` Jian Shen
  2021-09-29 15:50 ` [RFCv2 net-next 006/167] net: convert the prototype of vlan_features_check Jian Shen
                   ` (165 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:50 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

For the origin type for netdev_features_t would be changed to
be unsigned long * from u64, so changes the prototype of
gso_features_check for adaption.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 net/core/dev.c | 23 ++++++++++++-----------
 1 file changed, 12 insertions(+), 11 deletions(-)

diff --git a/net/core/dev.c b/net/core/dev.c
index ef77e080504b..b3426559bac7 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -3487,18 +3487,21 @@ static netdev_features_t dflt_features_check(struct sk_buff *skb,
 	return vlan_features_check(skb, features);
 }
 
-static netdev_features_t gso_features_check(const struct sk_buff *skb,
-					    struct net_device *dev,
-					    netdev_features_t features)
+static void gso_features_check(const struct sk_buff *skb,
+			       struct net_device *dev,
+			       netdev_features_t *features)
 {
 	u16 gso_segs = skb_shinfo(skb)->gso_segs;
 
-	if (gso_segs > dev->gso_max_segs)
-		return features & ~NETIF_F_GSO_MASK;
+	if (gso_segs > dev->gso_max_segs) {
+		*features &= ~NETIF_F_GSO_MASK;
+		return;
+	}
 
 	if (!skb_shinfo(skb)->gso_type) {
 		skb_warn_bad_offload(skb);
-		return features & ~NETIF_F_GSO_MASK;
+		*features &= ~NETIF_F_GSO_MASK;
+		return;
 	}
 
 	/* Support for GSO partial features requires software
@@ -3508,7 +3511,7 @@ static netdev_features_t gso_features_check(const struct sk_buff *skb,
 	 * segmented the frame.
 	 */
 	if (!(skb_shinfo(skb)->gso_type & SKB_GSO_PARTIAL))
-		features &= ~dev->gso_partial_features;
+		*features &= ~dev->gso_partial_features;
 
 	/* Make sure to clear the IPv4 ID mangling feature if the
 	 * IPv4 header has the potential to be fragmented.
@@ -3518,10 +3521,8 @@ static netdev_features_t gso_features_check(const struct sk_buff *skb,
 				    inner_ip_hdr(skb) : ip_hdr(skb);
 
 		if (!(iph->frag_off & htons(IP_DF)))
-			features &= ~NETIF_F_TSO_MANGLEID;
+			*features &= ~NETIF_F_TSO_MANGLEID;
 	}
-
-	return features;
 }
 
 netdev_features_t netif_skb_features(struct sk_buff *skb)
@@ -3530,7 +3531,7 @@ netdev_features_t netif_skb_features(struct sk_buff *skb)
 	netdev_features_t features = dev->features;
 
 	if (skb_is_gso(skb))
-		features = gso_features_check(skb, dev, features);
+		gso_features_check(skb, dev, &features);
 
 	/* If encapsulation offload request, verify we are testing
 	 * hardware encapsulation features instead of standard
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 006/167] net: convert the prototype of vlan_features_check
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (4 preceding siblings ...)
  2021-09-29 15:50 ` [RFCv2 net-next 005/167] net: convert the prototype of gso_features_check Jian Shen
@ 2021-09-29 15:50 ` Jian Shen
  2021-09-29 15:50 ` [RFCv2 net-next 007/167] net: convert the prototype of vxlan_features_check Jian Shen
                   ` (164 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:50 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

For the origin type for netdev_features_t would be changed to
be unsigned long * from u64, so changes the prototype of
vlan_features_check for adaption.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 drivers/net/ethernet/amd/xgbe/xgbe-drv.c            |  2 +-
 drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c    |  2 +-
 drivers/net/ethernet/broadcom/bnxt/bnxt.c           |  2 +-
 drivers/net/ethernet/emulex/benet/be_main.c         |  2 +-
 drivers/net/ethernet/mellanox/mlx4/en_netdev.c      |  2 +-
 drivers/net/ethernet/mellanox/mlx5/core/en_main.c   |  2 +-
 drivers/net/ethernet/netronome/nfp/nfp_net_common.c |  2 +-
 drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c    |  2 +-
 drivers/net/ethernet/realtek/8139cp.c               |  3 ++-
 drivers/net/ethernet/realtek/r8169_main.c           |  3 ++-
 drivers/net/usb/lan78xx.c                           |  2 +-
 drivers/s390/net/qeth_core_main.c                   |  3 ++-
 include/linux/if_vlan.h                             | 12 +++++-------
 net/core/dev.c                                      |  3 ++-
 14 files changed, 22 insertions(+), 20 deletions(-)

diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-drv.c b/drivers/net/ethernet/amd/xgbe/xgbe-drv.c
index 17a585adfb49..3fb15f675ddf 100644
--- a/drivers/net/ethernet/amd/xgbe/xgbe-drv.c
+++ b/drivers/net/ethernet/amd/xgbe/xgbe-drv.c
@@ -2271,7 +2271,7 @@ static netdev_features_t xgbe_features_check(struct sk_buff *skb,
 					     struct net_device *netdev,
 					     netdev_features_t features)
 {
-	features = vlan_features_check(skb, features);
+	vlan_features_check(skb, &features);
 	features = vxlan_features_check(skb, features);
 
 	return features;
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
index ae87296ae1ff..105aae3e21bf 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
@@ -12856,7 +12856,7 @@ static netdev_features_t bnx2x_features_check(struct sk_buff *skb,
 		     !skb_gso_validate_mac_len(skb, 9700)))
 		features &= ~NETIF_F_GSO_MASK;
 
-	features = vlan_features_check(skb, features);
+	vlan_features_check(skb, &features);
 	return vxlan_features_check(skb, features);
 }
 
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index 0fba01db336c..775dfcdd35f2 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -11097,7 +11097,7 @@ static netdev_features_t bnxt_features_check(struct sk_buff *skb,
 	struct bnxt *bp = netdev_priv(dev);
 	u8 *l4_proto;
 
-	features = vlan_features_check(skb, features);
+	vlan_features_check(skb, &features);
 	switch (vlan_get_protocol(skb)) {
 	case htons(ETH_P_IP):
 		if (!skb->encapsulation)
diff --git a/drivers/net/ethernet/emulex/benet/be_main.c b/drivers/net/ethernet/emulex/benet/be_main.c
index 361c1c87c183..556242d32d93 100644
--- a/drivers/net/ethernet/emulex/benet/be_main.c
+++ b/drivers/net/ethernet/emulex/benet/be_main.c
@@ -5092,7 +5092,7 @@ static netdev_features_t be_features_check(struct sk_buff *skb,
 	 * Q-in-Q packets.
 	 * Offload features for normal (non tunnel) packets are unchanged.
 	 */
-	features = vlan_features_check(skb, features);
+	vlan_features_check(skb, &features);
 	if (!skb->encapsulation ||
 	    !(adapter->flags & BE_FLAGS_VXLAN_OFFLOADS))
 		return features;
diff --git a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
index 8af7f2827322..a3c2a3e2d392 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
@@ -2678,7 +2678,7 @@ static netdev_features_t mlx4_en_features_check(struct sk_buff *skb,
 						struct net_device *dev,
 						netdev_features_t features)
 {
-	features = vlan_features_check(skb, features);
+	vlan_features_check(skb, &features);
 	features = vxlan_features_check(skb, features);
 
 	/* The ConnectX-3 doesn't support outer IPv6 checksums but it does
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index 3fd515e7bf30..b6aa5da06776 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -3910,7 +3910,7 @@ netdev_features_t mlx5e_features_check(struct sk_buff *skb,
 {
 	struct mlx5e_priv *priv = netdev_priv(netdev);
 
-	features = vlan_features_check(skb, features);
+	vlan_features_check(skb, &features);
 	features = vxlan_features_check(skb, features);
 
 	/* Validate if the tunneled packet is being offloaded by HW */
diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
index 5bfa22accf2c..58e7d98d0dd6 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
+++ b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
@@ -3605,7 +3605,7 @@ nfp_net_features_check(struct sk_buff *skb, struct net_device *dev,
 	u8 l4_hdr;
 
 	/* We can't do TSO over double tagged packets (802.1AD) */
-	features &= vlan_features_check(skb, features);
+	vlan_features_check(skb, &features);
 
 	if (!skb->encapsulation)
 		return features;
diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c
index 75960a29f80e..8f3a2a021082 100644
--- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c
+++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c
@@ -496,7 +496,7 @@ static netdev_features_t qlcnic_features_check(struct sk_buff *skb,
 					       struct net_device *dev,
 					       netdev_features_t features)
 {
-	features = vlan_features_check(skb, features);
+	vlan_features_check(skb, &features);
 	return vxlan_features_check(skb, features);
 }
 
diff --git a/drivers/net/ethernet/realtek/8139cp.c b/drivers/net/ethernet/realtek/8139cp.c
index 2b84b4565e64..d2135fe99cd7 100644
--- a/drivers/net/ethernet/realtek/8139cp.c
+++ b/drivers/net/ethernet/realtek/8139cp.c
@@ -1860,7 +1860,8 @@ static netdev_features_t cp_features_check(struct sk_buff *skb,
 	if (skb_shinfo(skb)->gso_size > MSSMask)
 		features &= ~NETIF_F_TSO;
 
-	return vlan_features_check(skb, features);
+	vlan_features_check(skb, &features);
+	return features;
 }
 static const struct net_device_ops cp_netdev_ops = {
 	.ndo_open		= cp_open,
diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c
index 0199914440ab..98e7ffa6aa0a 100644
--- a/drivers/net/ethernet/realtek/r8169_main.c
+++ b/drivers/net/ethernet/realtek/r8169_main.c
@@ -4374,7 +4374,8 @@ static netdev_features_t rtl8169_features_check(struct sk_buff *skb,
 			features &= ~NETIF_F_CSUM_MASK;
 	}
 
-	return vlan_features_check(skb, features);
+	vlan_features_check(skb, &features);
+	return features;
 }
 
 static void rtl8169_pcierr_interrupt(struct net_device *dev)
diff --git a/drivers/net/usb/lan78xx.c b/drivers/net/usb/lan78xx.c
index 793f8fbe0069..bf477ea4ac26 100644
--- a/drivers/net/usb/lan78xx.c
+++ b/drivers/net/usb/lan78xx.c
@@ -3983,7 +3983,7 @@ static netdev_features_t lan78xx_features_check(struct sk_buff *skb,
 	if (skb->len + TX_OVERHEAD > MAX_SINGLE_PACKET_SIZE)
 		features &= ~NETIF_F_GSO_MASK;
 
-	features = vlan_features_check(skb, features);
+	vlan_features_check(skb, &features);
 	features = vxlan_features_check(skb, features);
 
 	return features;
diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c
index e9807d2996a9..3b1903f8790a 100644
--- a/drivers/s390/net/qeth_core_main.c
+++ b/drivers/s390/net/qeth_core_main.c
@@ -6998,7 +6998,8 @@ netdev_features_t qeth_features_check(struct sk_buff *skb,
 			features &= ~NETIF_F_SG;
 	}
 
-	return vlan_features_check(skb, features);
+	vlan_features_check(skb, &features);
+	return features;
 }
 EXPORT_SYMBOL_GPL(qeth_features_check);
 
diff --git a/include/linux/if_vlan.h b/include/linux/if_vlan.h
index 41a518336673..2337538ef015 100644
--- a/include/linux/if_vlan.h
+++ b/include/linux/if_vlan.h
@@ -727,8 +727,8 @@ static inline bool skb_vlan_tagged_multi(struct sk_buff *skb)
  *
  * Returns features without unsafe ones if the skb has multiple tags.
  */
-static inline netdev_features_t vlan_features_check(struct sk_buff *skb,
-						    netdev_features_t features)
+static inline void vlan_features_check(struct sk_buff *skb,
+				       netdev_features_t *features)
 {
 	if (skb_vlan_tagged_multi(skb)) {
 		/* In the case of multi-tagged packets, use a direct mask
@@ -736,12 +736,10 @@ static inline netdev_features_t vlan_features_check(struct sk_buff *skb,
 		 * sure that only devices supporting NETIF_F_HW_CSUM will
 		 * have checksum offloading support.
 		 */
-		features &= NETIF_F_SG | NETIF_F_HIGHDMA | NETIF_F_HW_CSUM |
-			    NETIF_F_FRAGLIST | NETIF_F_HW_VLAN_CTAG_TX |
-			    NETIF_F_HW_VLAN_STAG_TX;
+		*features &= NETIF_F_SG | NETIF_F_HIGHDMA | NETIF_F_HW_CSUM |
+			     NETIF_F_FRAGLIST | NETIF_F_HW_VLAN_CTAG_TX |
+			     NETIF_F_HW_VLAN_STAG_TX;
 	}
-
-	return features;
 }
 
 /**
diff --git a/net/core/dev.c b/net/core/dev.c
index b3426559bac7..85d894e06f4e 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -3484,7 +3484,8 @@ static netdev_features_t dflt_features_check(struct sk_buff *skb,
 					     struct net_device *dev,
 					     netdev_features_t features)
 {
-	return vlan_features_check(skb, features);
+	vlan_features_check(skb, &features);
+	return features;
 }
 
 static void gso_features_check(const struct sk_buff *skb,
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 007/167] net: convert the prototype of vxlan_features_check
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (5 preceding siblings ...)
  2021-09-29 15:50 ` [RFCv2 net-next 006/167] net: convert the prototype of vlan_features_check Jian Shen
@ 2021-09-29 15:50 ` Jian Shen
  2021-09-29 15:50 ` [RFCv2 net-next 008/167] net: convert the prototype of dflt_features_check Jian Shen
                   ` (163 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:50 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

For the origin type for netdev_features_t would be changed to
be unsigned long * from u64, so changes the prototype of
vxlan_features_check for adaption.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 drivers/net/ethernet/amd/xgbe/xgbe-drv.c          |  2 +-
 drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c  |  3 ++-
 drivers/net/ethernet/cisco/enic/enic_main.c       |  2 +-
 drivers/net/ethernet/mellanox/mlx4/en_netdev.c    |  2 +-
 drivers/net/ethernet/mellanox/mlx5/core/en_main.c |  2 +-
 drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c  |  3 ++-
 drivers/net/usb/lan78xx.c                         |  2 +-
 include/net/vxlan.h                               | 13 ++++++-------
 8 files changed, 15 insertions(+), 14 deletions(-)

diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-drv.c b/drivers/net/ethernet/amd/xgbe/xgbe-drv.c
index 3fb15f675ddf..29bca9fabcde 100644
--- a/drivers/net/ethernet/amd/xgbe/xgbe-drv.c
+++ b/drivers/net/ethernet/amd/xgbe/xgbe-drv.c
@@ -2272,7 +2272,7 @@ static netdev_features_t xgbe_features_check(struct sk_buff *skb,
 					     netdev_features_t features)
 {
 	vlan_features_check(skb, &features);
-	features = vxlan_features_check(skb, features);
+	vxlan_features_check(skb, &features);
 
 	return features;
 }
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
index 105aae3e21bf..f444515452a5 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
@@ -12857,7 +12857,8 @@ static netdev_features_t bnx2x_features_check(struct sk_buff *skb,
 		features &= ~NETIF_F_GSO_MASK;
 
 	vlan_features_check(skb, &features);
-	return vxlan_features_check(skb, features);
+	vxlan_features_check(skb, &features);
+	return features;
 }
 
 static int __bnx2x_vlan_configure_vid(struct bnx2x *bp, u16 vid, bool add)
diff --git a/drivers/net/ethernet/cisco/enic/enic_main.c b/drivers/net/ethernet/cisco/enic/enic_main.c
index 13dbdd5d07b4..558a07c5b4bc 100644
--- a/drivers/net/ethernet/cisco/enic/enic_main.c
+++ b/drivers/net/ethernet/cisco/enic/enic_main.c
@@ -253,7 +253,7 @@ static netdev_features_t enic_features_check(struct sk_buff *skb,
 	if (!skb->encapsulation)
 		return features;
 
-	features = vxlan_features_check(skb, features);
+	vxlan_features_check(skb, &features);
 
 	switch (vlan_get_protocol(skb)) {
 	case htons(ETH_P_IPV6):
diff --git a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
index a3c2a3e2d392..dd6a41182446 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
@@ -2679,7 +2679,7 @@ static netdev_features_t mlx4_en_features_check(struct sk_buff *skb,
 						netdev_features_t features)
 {
 	vlan_features_check(skb, &features);
-	features = vxlan_features_check(skb, features);
+	vxlan_features_check(skb, &features);
 
 	/* The ConnectX-3 doesn't support outer IPv6 checksums but it does
 	 * support inner IPv6 checksums and segmentation so  we need to
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index b6aa5da06776..4518a490cdcd 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -3911,7 +3911,7 @@ netdev_features_t mlx5e_features_check(struct sk_buff *skb,
 	struct mlx5e_priv *priv = netdev_priv(netdev);
 
 	vlan_features_check(skb, &features);
-	features = vxlan_features_check(skb, features);
+	vxlan_features_check(skb, &features);
 
 	/* Validate if the tunneled packet is being offloaded by HW */
 	if (skb->encapsulation &&
diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c
index 8f3a2a021082..d823f2a22472 100644
--- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c
+++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c
@@ -497,7 +497,8 @@ static netdev_features_t qlcnic_features_check(struct sk_buff *skb,
 					       netdev_features_t features)
 {
 	vlan_features_check(skb, &features);
-	return vxlan_features_check(skb, features);
+	vxlan_features_check(skb, &features);
+	return features;
 }
 
 static const struct net_device_ops qlcnic_netdev_ops = {
diff --git a/drivers/net/usb/lan78xx.c b/drivers/net/usb/lan78xx.c
index bf477ea4ac26..9bfee75cbaf1 100644
--- a/drivers/net/usb/lan78xx.c
+++ b/drivers/net/usb/lan78xx.c
@@ -3984,7 +3984,7 @@ static netdev_features_t lan78xx_features_check(struct sk_buff *skb,
 		features &= ~NETIF_F_GSO_MASK;
 
 	vlan_features_check(skb, &features);
-	features = vxlan_features_check(skb, features);
+	vxlan_features_check(skb, &features);
 
 	return features;
 }
diff --git a/include/net/vxlan.h b/include/net/vxlan.h
index 08537aa14f7c..a801e59e2313 100644
--- a/include/net/vxlan.h
+++ b/include/net/vxlan.h
@@ -296,13 +296,13 @@ struct vxlan_dev {
 struct net_device *vxlan_dev_create(struct net *net, const char *name,
 				    u8 name_assign_type, struct vxlan_config *conf);
 
-static inline netdev_features_t vxlan_features_check(struct sk_buff *skb,
-						     netdev_features_t features)
+static inline void vxlan_features_check(struct sk_buff *skb,
+					netdev_features_t *features)
 {
 	u8 l4_hdr = 0;
 
 	if (!skb->encapsulation)
-		return features;
+		return;
 
 	switch (vlan_get_protocol(skb)) {
 	case htons(ETH_P_IP):
@@ -312,7 +312,7 @@ static inline netdev_features_t vxlan_features_check(struct sk_buff *skb,
 		l4_hdr = ipv6_hdr(skb)->nexthdr;
 		break;
 	default:
-		return features;
+		return;
 	}
 
 	if ((l4_hdr == IPPROTO_UDP) &&
@@ -321,10 +321,9 @@ static inline netdev_features_t vxlan_features_check(struct sk_buff *skb,
 	     (skb_inner_mac_header(skb) - skb_transport_header(skb) !=
 	      sizeof(struct udphdr) + sizeof(struct vxlanhdr)) ||
 	     (skb->ip_summed != CHECKSUM_NONE &&
-	      !can_checksum_protocol(features, inner_eth_hdr(skb)->h_proto))))
-		return features & ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
+	      !can_checksum_protocol(*features, inner_eth_hdr(skb)->h_proto))))
+		*features &= ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
 
-	return features;
 }
 
 /* IP header + UDP + VXLAN + Ethernet header */
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 008/167] net: convert the prototype of dflt_features_check
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (6 preceding siblings ...)
  2021-09-29 15:50 ` [RFCv2 net-next 007/167] net: convert the prototype of vxlan_features_check Jian Shen
@ 2021-09-29 15:50 ` Jian Shen
  2021-09-29 15:50 ` [RFCv2 net-next 009/167] net: convert the prototype of ndo_features_check Jian Shen
                   ` (162 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:50 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

For the origin type for netdev_features_t would be changed to
be unsigned long * from u64, so changes the prototype of
dflt_features_check for adaption.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 net/core/dev.c | 11 +++++------
 1 file changed, 5 insertions(+), 6 deletions(-)

diff --git a/net/core/dev.c b/net/core/dev.c
index 85d894e06f4e..814e6e7ee579 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -3480,12 +3480,11 @@ netdev_features_t passthru_features_check(struct sk_buff *skb,
 }
 EXPORT_SYMBOL(passthru_features_check);
 
-static netdev_features_t dflt_features_check(struct sk_buff *skb,
-					     struct net_device *dev,
-					     netdev_features_t features)
+static void dflt_features_check(struct sk_buff *skb,
+				struct net_device *dev,
+				netdev_features_t *features)
 {
-	vlan_features_check(skb, &features);
-	return features;
+	vlan_features_check(skb, features);
 }
 
 static void gso_features_check(const struct sk_buff *skb,
@@ -3551,7 +3550,7 @@ netdev_features_t netif_skb_features(struct sk_buff *skb)
 		features &= dev->netdev_ops->ndo_features_check(skb, dev,
 								features);
 	else
-		features &= dflt_features_check(skb, dev, features);
+		dflt_features_check(skb, dev, &features);
 
 	harmonize_features(skb, &features);
 	return features;
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 009/167] net: convert the prototype of ndo_features_check
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (7 preceding siblings ...)
  2021-09-29 15:50 ` [RFCv2 net-next 008/167] net: convert the prototype of dflt_features_check Jian Shen
@ 2021-09-29 15:50 ` Jian Shen
  2021-09-29 15:50 ` [RFCv2 net-next 010/167] net: convert the prototype of netif_skb_features Jian Shen
                   ` (161 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:50 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

For the origin type for netdev_features_t would be changed to
be unsigned long * from u64, so changes the prototype of
ndo_features_check for adaption.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 drivers/net/ethernet/amd/xgbe/xgbe-drv.c      | 11 ++---
 .../net/ethernet/broadcom/bnx2x/bnx2x_main.c  | 12 +++---
 drivers/net/ethernet/broadcom/bnxt/bnxt.c     | 15 ++++---
 drivers/net/ethernet/cadence/macb_main.c      | 20 +++++----
 .../net/ethernet/chelsio/cxgb4/cxgb4_main.c   | 11 +++--
 drivers/net/ethernet/cisco/enic/enic_main.c   | 13 +++---
 drivers/net/ethernet/emulex/benet/be_main.c   | 23 +++++------
 .../net/ethernet/hisilicon/hns3/hns3_enet.c   | 11 ++---
 drivers/net/ethernet/ibm/ibmvnic.c            |  9 ++--
 .../net/ethernet/intel/fm10k/fm10k_netdev.c   |  9 ++--
 drivers/net/ethernet/intel/i40e/i40e_main.c   | 13 +++---
 drivers/net/ethernet/intel/iavf/iavf_main.c   | 13 +++---
 drivers/net/ethernet/intel/ice/ice_main.c     | 15 ++++---
 drivers/net/ethernet/intel/igb/igb_main.c     | 41 ++++++++++---------
 drivers/net/ethernet/intel/igbvf/netdev.c     | 37 +++++++++--------
 drivers/net/ethernet/intel/igc/igc_main.c     | 37 +++++++++--------
 drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 41 ++++++++++---------
 .../net/ethernet/intel/ixgbevf/ixgbevf_main.c | 37 +++++++++--------
 .../net/ethernet/mellanox/mlx4/en_netdev.c    | 13 +++---
 drivers/net/ethernet/mellanox/mlx5/core/en.h  |  5 +--
 .../net/ethernet/mellanox/mlx5/core/en_main.c | 15 +++----
 .../ethernet/netronome/nfp/nfp_net_common.c   | 22 +++++-----
 drivers/net/ethernet/qlogic/qede/qede.h       |  5 +--
 drivers/net/ethernet/qlogic/qede/qede_fp.c    | 20 ++++-----
 .../net/ethernet/qlogic/qlcnic/qlcnic_main.c  | 10 ++---
 drivers/net/ethernet/realtek/8139cp.c         | 10 ++---
 drivers/net/ethernet/realtek/r8169_main.c     | 18 ++++----
 drivers/net/ethernet/sfc/efx_common.c         | 13 +++---
 drivers/net/ethernet/sfc/efx_common.h         |  4 +-
 drivers/net/usb/lan78xx.c                     | 14 +++----
 drivers/net/usb/r8152.c                       | 11 ++---
 drivers/net/vmxnet3/vmxnet3_ethtool.c         | 15 +++----
 drivers/net/vmxnet3/vmxnet3_int.h             |  4 +-
 drivers/s390/net/qeth_core.h                  |  5 +--
 drivers/s390/net/qeth_core_main.c             | 18 ++++----
 drivers/s390/net/qeth_l3_main.c               | 10 ++---
 include/linux/netdevice.h                     |  9 ++--
 net/core/dev.c                                |  9 ++--
 38 files changed, 280 insertions(+), 318 deletions(-)

diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-drv.c b/drivers/net/ethernet/amd/xgbe/xgbe-drv.c
index 29bca9fabcde..cc1f1a7a46ae 100644
--- a/drivers/net/ethernet/amd/xgbe/xgbe-drv.c
+++ b/drivers/net/ethernet/amd/xgbe/xgbe-drv.c
@@ -2267,14 +2267,11 @@ static int xgbe_set_features(struct net_device *netdev,
 	return 0;
 }
 
-static netdev_features_t xgbe_features_check(struct sk_buff *skb,
-					     struct net_device *netdev,
-					     netdev_features_t features)
+static void xgbe_features_check(struct sk_buff *skb, struct net_device *netdev,
+				netdev_features_t *features)
 {
-	vlan_features_check(skb, &features);
-	vxlan_features_check(skb, &features);
-
-	return features;
+	vlan_features_check(skb, features);
+	vxlan_features_check(skb, features);
 }
 
 static const struct net_device_ops xgbe_netdev_ops = {
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
index f444515452a5..1f817dfa5b64 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
@@ -12834,9 +12834,8 @@ static int bnx2x_get_phys_port_id(struct net_device *netdev,
 	return 0;
 }
 
-static netdev_features_t bnx2x_features_check(struct sk_buff *skb,
-					      struct net_device *dev,
-					      netdev_features_t features)
+static void bnx2x_features_check(struct sk_buff *skb, struct net_device *dev,
+				 netdev_features_t *features)
 {
 	/*
 	 * A skb with gso_size + header length > 9700 will cause a
@@ -12854,11 +12853,10 @@ static netdev_features_t bnx2x_features_check(struct sk_buff *skb,
 	if (unlikely(skb_is_gso(skb) &&
 		     (skb_shinfo(skb)->gso_size > 9000) &&
 		     !skb_gso_validate_mac_len(skb, 9700)))
-		features &= ~NETIF_F_GSO_MASK;
+		*features &= ~NETIF_F_GSO_MASK;
 
-	vlan_features_check(skb, &features);
-	vxlan_features_check(skb, &features);
-	return features;
+	vlan_features_check(skb, features);
+	vxlan_features_check(skb, features);
 }
 
 static int __bnx2x_vlan_configure_vid(struct bnx2x *bp, u16 vid, bool add)
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index 775dfcdd35f2..4689d053aff8 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -11090,31 +11090,30 @@ static bool bnxt_tunl_check(struct bnxt *bp, struct sk_buff *skb, u8 l4_proto)
 	return false;
 }
 
-static netdev_features_t bnxt_features_check(struct sk_buff *skb,
-					     struct net_device *dev,
-					     netdev_features_t features)
+static void bnxt_features_check(struct sk_buff *skb, struct net_device *dev,
+				netdev_features_t *features)
 {
 	struct bnxt *bp = netdev_priv(dev);
 	u8 *l4_proto;
 
-	vlan_features_check(skb, &features);
+	vlan_features_check(skb, features);
 	switch (vlan_get_protocol(skb)) {
 	case htons(ETH_P_IP):
 		if (!skb->encapsulation)
-			return features;
+			return;
 		l4_proto = &ip_hdr(skb)->protocol;
 		if (bnxt_tunl_check(bp, skb, *l4_proto))
-			return features;
+			return;
 		break;
 	case htons(ETH_P_IPV6):
 		if (!bnxt_exthdr_check(bp, skb, skb_network_offset(skb),
 				       &l4_proto))
 			break;
 		if (!l4_proto || bnxt_tunl_check(bp, skb, *l4_proto))
-			return features;
+			return;
 		break;
 	}
-	return features & ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
+	*features &= ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
 }
 
 int bnxt_dbg_hwrm_rd_reg(struct bnxt *bp, u32 reg_off, u16 num_words,
diff --git a/drivers/net/ethernet/cadence/macb_main.c b/drivers/net/ethernet/cadence/macb_main.c
index e2730b3e1a57..e30ee19d9ba2 100644
--- a/drivers/net/ethernet/cadence/macb_main.c
+++ b/drivers/net/ethernet/cadence/macb_main.c
@@ -2035,9 +2035,8 @@ static unsigned int macb_tx_map(struct macb *bp,
 	return 0;
 }
 
-static netdev_features_t macb_features_check(struct sk_buff *skb,
-					     struct net_device *dev,
-					     netdev_features_t features)
+static void macb_features_check(struct sk_buff *skb, struct net_device *dev,
+				netdev_features_t *features)
 {
 	unsigned int nr_frags, f;
 	unsigned int hdrlen;
@@ -2046,7 +2045,7 @@ static netdev_features_t macb_features_check(struct sk_buff *skb,
 
 	/* there is only one buffer or protocol is not UDP */
 	if (!skb_is_nonlinear(skb) || (ip_hdr(skb)->protocol != IPPROTO_UDP))
-		return features;
+		return;
 
 	/* length of header */
 	hdrlen = skb_transport_offset(skb);
@@ -2055,8 +2054,10 @@ static netdev_features_t macb_features_check(struct sk_buff *skb,
 	 * When software supplies two or more payload buffers all payload buffers
 	 * apart from the last must be a multiple of 8 bytes in size.
 	 */
-	if (!IS_ALIGNED(skb_headlen(skb) - hdrlen, MACB_TX_LEN_ALIGN))
-		return features & ~MACB_NETIF_LSO;
+	if (!IS_ALIGNED(skb_headlen(skb) - hdrlen, MACB_TX_LEN_ALIGN)) {
+		*features &= ~MACB_NETIF_LSO;
+		return;
+	}
 
 	nr_frags = skb_shinfo(skb)->nr_frags;
 	/* No need to check last fragment */
@@ -2064,10 +2065,11 @@ static netdev_features_t macb_features_check(struct sk_buff *skb,
 	for (f = 0; f < nr_frags; f++) {
 		const skb_frag_t *frag = &skb_shinfo(skb)->frags[f];
 
-		if (!IS_ALIGNED(skb_frag_size(frag), MACB_TX_LEN_ALIGN))
-			return features & ~MACB_NETIF_LSO;
+		if (!IS_ALIGNED(skb_frag_size(frag), MACB_TX_LEN_ALIGN)) {
+			*features &= ~MACB_NETIF_LSO;
+			return;
+		}
 	}
-	return features;
 }
 
 static inline int macb_clear_csum(struct sk_buff *skb)
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c
index 0d9cda4ab303..a654169b9dfc 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c
@@ -3834,22 +3834,21 @@ static const struct udp_tunnel_nic_info cxgb_udp_tunnels = {
 	},
 };
 
-static netdev_features_t cxgb_features_check(struct sk_buff *skb,
-					     struct net_device *dev,
-					     netdev_features_t features)
+static void cxgb_features_check(struct sk_buff *skb, struct net_device *dev,
+				netdev_features_t *features)
 {
 	struct port_info *pi = netdev_priv(dev);
 	struct adapter *adapter = pi->adapter;
 
 	if (CHELSIO_CHIP_VERSION(adapter->params.chip) < CHELSIO_T6)
-		return features;
+		return;
 
 	/* Check if hw supports offload for this packet */
 	if (!skb->encapsulation || cxgb_encap_offload_supported(skb))
-		return features;
+		return;
 
 	/* Offload is not supported for this encapsulated packet */
-	return features & ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
+	*features &= ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
 }
 
 static netdev_features_t cxgb_fix_features(struct net_device *dev,
diff --git a/drivers/net/ethernet/cisco/enic/enic_main.c b/drivers/net/ethernet/cisco/enic/enic_main.c
index 558a07c5b4bc..5bc075331343 100644
--- a/drivers/net/ethernet/cisco/enic/enic_main.c
+++ b/drivers/net/ethernet/cisco/enic/enic_main.c
@@ -240,9 +240,8 @@ static const struct udp_tunnel_nic_info enic_udp_tunnels = {
 	},
 };
 
-static netdev_features_t enic_features_check(struct sk_buff *skb,
-					     struct net_device *dev,
-					     netdev_features_t features)
+static void enic_features_check(struct sk_buff *skb, struct net_device *dev,
+				netdev_features_t *features)
 {
 	const struct ethhdr *eth = (struct ethhdr *)skb_inner_mac_header(skb);
 	struct enic *enic = netdev_priv(dev);
@@ -251,9 +250,9 @@ static netdev_features_t enic_features_check(struct sk_buff *skb,
 	u8 proto;
 
 	if (!skb->encapsulation)
-		return features;
+		return;
 
-	vxlan_features_check(skb, &features);
+	vxlan_features_check(skb, features);
 
 	switch (vlan_get_protocol(skb)) {
 	case htons(ETH_P_IPV6):
@@ -291,10 +290,10 @@ static netdev_features_t enic_features_check(struct sk_buff *skb,
 	if (port  != enic->vxlan.vxlan_udp_port_number)
 		goto out;
 
-	return features;
+	return;
 
 out:
-	return features & ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
+	*features &= ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
 }
 
 int enic_is_dynamic(struct enic *enic)
diff --git a/drivers/net/ethernet/emulex/benet/be_main.c b/drivers/net/ethernet/emulex/benet/be_main.c
index 556242d32d93..27e8a1d95a9a 100644
--- a/drivers/net/ethernet/emulex/benet/be_main.c
+++ b/drivers/net/ethernet/emulex/benet/be_main.c
@@ -5064,9 +5064,8 @@ static struct be_cmd_work *be_alloc_work(struct be_adapter *adapter,
 	return work;
 }
 
-static netdev_features_t be_features_check(struct sk_buff *skb,
-					   struct net_device *dev,
-					   netdev_features_t features)
+static void be_features_check(struct sk_buff *skb, struct net_device *dev,
+			      netdev_features_t *features)
 {
 	struct be_adapter *adapter = netdev_priv(dev);
 	u8 l4_hdr = 0;
@@ -5076,7 +5075,7 @@ static netdev_features_t be_features_check(struct sk_buff *skb,
 		 * to Lancer and BE3 HW. Disable TSO6 feature.
 		 */
 		if (!skyhawk_chip(adapter) && is_ipv6_ext_hdr(skb))
-			features &= ~NETIF_F_TSO6;
+			*features &= ~NETIF_F_TSO6;
 
 		/* Lancer cannot handle the packet with MSS less than 256.
 		 * Also it can't handle a TSO packet with a single segment
@@ -5085,17 +5084,17 @@ static netdev_features_t be_features_check(struct sk_buff *skb,
 		if (lancer_chip(adapter) &&
 		    (skb_shinfo(skb)->gso_size < 256 ||
 		     skb_shinfo(skb)->gso_segs == 1))
-			features &= ~NETIF_F_GSO_MASK;
+			*features &= ~NETIF_F_GSO_MASK;
 	}
 
 	/* The code below restricts offload features for some tunneled and
 	 * Q-in-Q packets.
 	 * Offload features for normal (non tunnel) packets are unchanged.
 	 */
-	vlan_features_check(skb, &features);
+	vlan_features_check(skb, features);
 	if (!skb->encapsulation ||
 	    !(adapter->flags & BE_FLAGS_VXLAN_OFFLOADS))
-		return features;
+		return;
 
 	/* It's an encapsulated packet and VxLAN offloads are enabled. We
 	 * should disable tunnel offload features if it's not a VxLAN packet,
@@ -5111,7 +5110,7 @@ static netdev_features_t be_features_check(struct sk_buff *skb,
 		l4_hdr = ipv6_hdr(skb)->nexthdr;
 		break;
 	default:
-		return features;
+		return;
 	}
 
 	if (l4_hdr != IPPROTO_UDP ||
@@ -5120,10 +5119,10 @@ static netdev_features_t be_features_check(struct sk_buff *skb,
 	    skb_inner_mac_header(skb) - skb_transport_header(skb) !=
 		sizeof(struct udphdr) + sizeof(struct vxlanhdr) ||
 	    !adapter->vxlan_port ||
-	    udp_hdr(skb)->dest != adapter->vxlan_port)
-		return features & ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
-
-	return features;
+	    udp_hdr(skb)->dest != adapter->vxlan_port) {
+		*features &= ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
+		return;
+	}
 }
 
 static int be_get_phys_port_id(struct net_device *dev,
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c
index adc54a726661..e8179ac6d4e3 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c
@@ -2353,9 +2353,8 @@ static int hns3_nic_set_features(struct net_device *netdev,
 	return 0;
 }
 
-static netdev_features_t hns3_features_check(struct sk_buff *skb,
-					     struct net_device *dev,
-					     netdev_features_t features)
+static void hns3_features_check(struct sk_buff *skb, struct net_device *dev,
+				netdev_features_t *features)
 {
 #define HNS3_MAX_HDR_LEN	480U
 #define HNS3_MAX_L4_HDR_LEN	60U
@@ -2363,7 +2362,7 @@ static netdev_features_t hns3_features_check(struct sk_buff *skb,
 	size_t len;
 
 	if (skb->ip_summed != CHECKSUM_PARTIAL)
-		return features;
+		return;
 
 	if (skb->encapsulation)
 		len = skb_inner_transport_header(skb) - skb->data;
@@ -2379,9 +2378,7 @@ static netdev_features_t hns3_features_check(struct sk_buff *skb,
 	 * len of 480 bytes.
 	 */
 	if (len > HNS3_MAX_HDR_LEN)
-		features &= ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
-
-	return features;
+		*features &= ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
 }
 
 static void hns3_nic_get_stats64(struct net_device *netdev,
diff --git a/drivers/net/ethernet/ibm/ibmvnic.c b/drivers/net/ethernet/ibm/ibmvnic.c
index 8f17096e614d..b8ee2e3977b5 100644
--- a/drivers/net/ethernet/ibm/ibmvnic.c
+++ b/drivers/net/ethernet/ibm/ibmvnic.c
@@ -3003,9 +3003,8 @@ static int ibmvnic_change_mtu(struct net_device *netdev, int new_mtu)
 	return wait_for_reset(adapter);
 }
 
-static netdev_features_t ibmvnic_features_check(struct sk_buff *skb,
-						struct net_device *dev,
-						netdev_features_t features)
+static void ibmvnic_features_check(struct sk_buff *skb, struct net_device *dev,
+				   netdev_features_t *features)
 {
 	/* Some backing hardware adapters can not
 	 * handle packets with a MSS less than 224
@@ -3014,10 +3013,8 @@ static netdev_features_t ibmvnic_features_check(struct sk_buff *skb,
 	if (skb_is_gso(skb)) {
 		if (skb_shinfo(skb)->gso_size < 224 ||
 		    skb_shinfo(skb)->gso_segs == 1)
-			features &= ~NETIF_F_GSO_MASK;
+			*features &= ~NETIF_F_GSO_MASK;
 	}
-
-	return features;
 }
 
 static const struct net_device_ops ibmvnic_netdev_ops = {
diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_netdev.c b/drivers/net/ethernet/intel/fm10k/fm10k_netdev.c
index 2fb52bd6fc0e..a311bcdfbff2 100644
--- a/drivers/net/ethernet/intel/fm10k/fm10k_netdev.c
+++ b/drivers/net/ethernet/intel/fm10k/fm10k_netdev.c
@@ -1504,14 +1504,13 @@ static void fm10k_dfwd_del_station(struct net_device *dev, void *priv)
 	}
 }
 
-static netdev_features_t fm10k_features_check(struct sk_buff *skb,
-					      struct net_device *dev,
-					      netdev_features_t features)
+static void fm10k_features_check(struct sk_buff *skb, struct net_device *dev,
+				 netdev_features_t *features)
 {
 	if (!skb->encapsulation || fm10k_tx_encap_offload(skb))
-		return features;
+		return;
 
-	return features & ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
+	*features &= ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
 }
 
 static const struct net_device_ops fm10k_netdev_ops = {
diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index 2f20980dd9a5..5246c6abbd7d 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -12873,9 +12873,8 @@ static int i40e_ndo_bridge_getlink(struct sk_buff *skb, u32 pid, u32 seq,
  * @dev: This physical port's netdev
  * @features: Offload features that the stack believes apply
  **/
-static netdev_features_t i40e_features_check(struct sk_buff *skb,
-					     struct net_device *dev,
-					     netdev_features_t features)
+static void i40e_features_check(struct sk_buff *skb, struct net_device *dev,
+				netdev_features_t *features)
 {
 	size_t len;
 
@@ -12884,13 +12883,13 @@ static netdev_features_t i40e_features_check(struct sk_buff *skb,
 	 * checking for CHECKSUM_PARTIAL
 	 */
 	if (skb->ip_summed != CHECKSUM_PARTIAL)
-		return features;
+		return;
 
 	/* We cannot support GSO if the MSS is going to be less than
 	 * 64 bytes.  If it is then we need to drop support for GSO.
 	 */
 	if (skb_is_gso(skb) && (skb_shinfo(skb)->gso_size < 64))
-		features &= ~NETIF_F_GSO_MASK;
+		*features &= ~NETIF_F_GSO_MASK;
 
 	/* MACLEN can support at most 63 words */
 	len = skb_network_header(skb) - skb->data;
@@ -12920,9 +12919,9 @@ static netdev_features_t i40e_features_check(struct sk_buff *skb,
 	 * by TCP, which is at most 15 dwords
 	 */
 
-	return features;
+	return;
 out_err:
-	return features & ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
+	*features &= ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
 }
 
 /**
diff --git a/drivers/net/ethernet/intel/iavf/iavf_main.c b/drivers/net/ethernet/intel/iavf/iavf_main.c
index 23762a7ef740..b066f1864b3f 100644
--- a/drivers/net/ethernet/intel/iavf/iavf_main.c
+++ b/drivers/net/ethernet/intel/iavf/iavf_main.c
@@ -3379,9 +3379,8 @@ static int iavf_set_features(struct net_device *netdev,
  * @dev: This physical port's netdev
  * @features: Offload features that the stack believes apply
  **/
-static netdev_features_t iavf_features_check(struct sk_buff *skb,
-					     struct net_device *dev,
-					     netdev_features_t features)
+static void iavf_features_check(struct sk_buff *skb, struct net_device *dev,
+				netdev_features_t *features)
 {
 	size_t len;
 
@@ -3390,13 +3389,13 @@ static netdev_features_t iavf_features_check(struct sk_buff *skb,
 	 * checking for CHECKSUM_PARTIAL
 	 */
 	if (skb->ip_summed != CHECKSUM_PARTIAL)
-		return features;
+		return;
 
 	/* We cannot support GSO if the MSS is going to be less than
 	 * 64 bytes.  If it is then we need to drop support for GSO.
 	 */
 	if (skb_is_gso(skb) && (skb_shinfo(skb)->gso_size < 64))
-		features &= ~NETIF_F_GSO_MASK;
+		*features &= ~NETIF_F_GSO_MASK;
 
 	/* MACLEN can support at most 63 words */
 	len = skb_network_header(skb) - skb->data;
@@ -3426,9 +3425,9 @@ static netdev_features_t iavf_features_check(struct sk_buff *skb,
 	 * by TCP, which is at most 15 dwords
 	 */
 
-	return features;
+	return;
 out_err:
-	return features & ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
+	*features &= ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
 }
 
 /**
diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c
index 909e5cd98054..1ad630b8d103 100644
--- a/drivers/net/ethernet/intel/ice/ice_main.c
+++ b/drivers/net/ethernet/intel/ice/ice_main.c
@@ -7176,10 +7176,9 @@ int ice_stop(struct net_device *netdev)
  * @netdev: This port's netdev
  * @features: Offload features that the stack believes apply
  */
-static netdev_features_t
-ice_features_check(struct sk_buff *skb,
-		   struct net_device __always_unused *netdev,
-		   netdev_features_t features)
+static void ice_features_check(struct sk_buff *skb,
+			       struct net_device __always_unused *netdev,
+			       netdev_features_t *features)
 {
 	size_t len;
 
@@ -7188,13 +7187,13 @@ ice_features_check(struct sk_buff *skb,
 	 * checking for CHECKSUM_PARTIAL
 	 */
 	if (skb->ip_summed != CHECKSUM_PARTIAL)
-		return features;
+		return;
 
 	/* We cannot support GSO if the MSS is going to be less than
 	 * 64 bytes. If it is then we need to drop support for GSO.
 	 */
 	if (skb_is_gso(skb) && (skb_shinfo(skb)->gso_size < 64))
-		features &= ~NETIF_F_GSO_MASK;
+		*features &= ~NETIF_F_GSO_MASK;
 
 	len = skb_network_header(skb) - skb->data;
 	if (len > ICE_TXD_MACLEN_MAX || len & 0x1)
@@ -7215,9 +7214,9 @@ ice_features_check(struct sk_buff *skb,
 			goto out_rm_features;
 	}
 
-	return features;
+	return;
 out_rm_features:
-	return features & ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
+	*features &= ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
 }
 
 static const struct net_device_ops ice_netdev_safe_mode_ops = {
diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c
index 751de06019a0..82b59adf8034 100644
--- a/drivers/net/ethernet/intel/igb/igb_main.c
+++ b/drivers/net/ethernet/intel/igb/igb_main.c
@@ -2505,37 +2505,38 @@ static int igb_ndo_fdb_add(struct ndmsg *ndm, struct nlattr *tb[],
 #define IGB_MAX_MAC_HDR_LEN	127
 #define IGB_MAX_NETWORK_HDR_LEN	511
 
-static netdev_features_t
-igb_features_check(struct sk_buff *skb, struct net_device *dev,
-		   netdev_features_t features)
+static void igb_features_check(struct sk_buff *skb, struct net_device *dev,
+			       netdev_features_t *features)
 {
 	unsigned int network_hdr_len, mac_hdr_len;
 
 	/* Make certain the headers can be described by a context descriptor */
 	mac_hdr_len = skb_network_header(skb) - skb->data;
-	if (unlikely(mac_hdr_len > IGB_MAX_MAC_HDR_LEN))
-		return features & ~(NETIF_F_HW_CSUM |
-				    NETIF_F_SCTP_CRC |
-				    NETIF_F_GSO_UDP_L4 |
-				    NETIF_F_HW_VLAN_CTAG_TX |
-				    NETIF_F_TSO |
-				    NETIF_F_TSO6);
+	if (unlikely(mac_hdr_len > IGB_MAX_MAC_HDR_LEN)) {
+		*features &= ~(NETIF_F_HW_CSUM |
+			       NETIF_F_SCTP_CRC |
+			       NETIF_F_GSO_UDP_L4 |
+			       NETIF_F_HW_VLAN_CTAG_TX |
+			       NETIF_F_TSO |
+			       NETIF_F_TSO6);
+		return;
+	}
 
 	network_hdr_len = skb_checksum_start(skb) - skb_network_header(skb);
-	if (unlikely(network_hdr_len >  IGB_MAX_NETWORK_HDR_LEN))
-		return features & ~(NETIF_F_HW_CSUM |
-				    NETIF_F_SCTP_CRC |
-				    NETIF_F_GSO_UDP_L4 |
-				    NETIF_F_TSO |
-				    NETIF_F_TSO6);
+	if (unlikely(network_hdr_len >  IGB_MAX_NETWORK_HDR_LEN)) {
+		*features &= ~(NETIF_F_HW_CSUM |
+			       NETIF_F_SCTP_CRC |
+			       NETIF_F_GSO_UDP_L4 |
+			       NETIF_F_TSO |
+			       NETIF_F_TSO6);
+		return;
+	}
 
 	/* We can only support IPV4 TSO in tunnels if we can mangle the
 	 * inner IP ID field, so strip TSO if MANGLEID is not supported.
 	 */
-	if (skb->encapsulation && !(features & NETIF_F_TSO_MANGLEID))
-		features &= ~NETIF_F_TSO;
-
-	return features;
+	if (skb->encapsulation && !(*features & NETIF_F_TSO_MANGLEID))
+		*features &= ~NETIF_F_TSO;
 }
 
 static void igb_offload_apply(struct igb_adapter *adapter, s32 queue)
diff --git a/drivers/net/ethernet/intel/igbvf/netdev.c b/drivers/net/ethernet/intel/igbvf/netdev.c
index d32e72d953c8..92a46ac4ee1f 100644
--- a/drivers/net/ethernet/intel/igbvf/netdev.c
+++ b/drivers/net/ethernet/intel/igbvf/netdev.c
@@ -2619,35 +2619,36 @@ static int igbvf_set_features(struct net_device *netdev,
 #define IGBVF_MAX_MAC_HDR_LEN		127
 #define IGBVF_MAX_NETWORK_HDR_LEN	511
 
-static netdev_features_t
-igbvf_features_check(struct sk_buff *skb, struct net_device *dev,
-		     netdev_features_t features)
+static void igbvf_features_check(struct sk_buff *skb, struct net_device *dev,
+				 netdev_features_t *features)
 {
 	unsigned int network_hdr_len, mac_hdr_len;
 
 	/* Make certain the headers can be described by a context descriptor */
 	mac_hdr_len = skb_network_header(skb) - skb->data;
-	if (unlikely(mac_hdr_len > IGBVF_MAX_MAC_HDR_LEN))
-		return features & ~(NETIF_F_HW_CSUM |
-				    NETIF_F_SCTP_CRC |
-				    NETIF_F_HW_VLAN_CTAG_TX |
-				    NETIF_F_TSO |
-				    NETIF_F_TSO6);
+	if (unlikely(mac_hdr_len > IGBVF_MAX_MAC_HDR_LEN)) {
+		*features &= ~(NETIF_F_HW_CSUM |
+			       NETIF_F_SCTP_CRC |
+			       NETIF_F_HW_VLAN_CTAG_TX |
+			       NETIF_F_TSO |
+			       NETIF_F_TSO6);
+		return;
+	}
 
 	network_hdr_len = skb_checksum_start(skb) - skb_network_header(skb);
-	if (unlikely(network_hdr_len >  IGBVF_MAX_NETWORK_HDR_LEN))
-		return features & ~(NETIF_F_HW_CSUM |
-				    NETIF_F_SCTP_CRC |
-				    NETIF_F_TSO |
-				    NETIF_F_TSO6);
+	if (unlikely(network_hdr_len >  IGBVF_MAX_NETWORK_HDR_LEN)) {
+		*features &= ~(NETIF_F_HW_CSUM |
+			       NETIF_F_SCTP_CRC |
+			       NETIF_F_TSO |
+			       NETIF_F_TSO6);
+		return;
+	}
 
 	/* We can only support IPV4 TSO in tunnels if we can mangle the
 	 * inner IP ID field, so strip TSO if MANGLEID is not supported.
 	 */
-	if (skb->encapsulation && !(features & NETIF_F_TSO_MANGLEID))
-		features &= ~NETIF_F_TSO;
-
-	return features;
+	if (skb->encapsulation && !(*features & NETIF_F_TSO_MANGLEID))
+		*features &= ~NETIF_F_TSO;
 }
 
 static const struct net_device_ops igbvf_netdev_ops = {
diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c
index 0e19b4d02e62..fdb3ed051456 100644
--- a/drivers/net/ethernet/intel/igc/igc_main.c
+++ b/drivers/net/ethernet/intel/igc/igc_main.c
@@ -4955,35 +4955,36 @@ static int igc_set_features(struct net_device *netdev,
 	return 1;
 }
 
-static netdev_features_t
-igc_features_check(struct sk_buff *skb, struct net_device *dev,
-		   netdev_features_t features)
+static void igc_features_check(struct sk_buff *skb, struct net_device *dev,
+			       netdev_features_t *features)
 {
 	unsigned int network_hdr_len, mac_hdr_len;
 
 	/* Make certain the headers can be described by a context descriptor */
 	mac_hdr_len = skb_network_header(skb) - skb->data;
-	if (unlikely(mac_hdr_len > IGC_MAX_MAC_HDR_LEN))
-		return features & ~(NETIF_F_HW_CSUM |
-				    NETIF_F_SCTP_CRC |
-				    NETIF_F_HW_VLAN_CTAG_TX |
-				    NETIF_F_TSO |
-				    NETIF_F_TSO6);
+	if (unlikely(mac_hdr_len > IGC_MAX_MAC_HDR_LEN)) {
+		*features &= ~(NETIF_F_HW_CSUM |
+			       NETIF_F_SCTP_CRC |
+			       NETIF_F_HW_VLAN_CTAG_TX |
+			       NETIF_F_TSO |
+			       NETIF_F_TSO6);
+		return;
+	}
 
 	network_hdr_len = skb_checksum_start(skb) - skb_network_header(skb);
-	if (unlikely(network_hdr_len >  IGC_MAX_NETWORK_HDR_LEN))
-		return features & ~(NETIF_F_HW_CSUM |
-				    NETIF_F_SCTP_CRC |
-				    NETIF_F_TSO |
-				    NETIF_F_TSO6);
+	if (unlikely(network_hdr_len >  IGC_MAX_NETWORK_HDR_LEN)) {
+		*features &= ~(NETIF_F_HW_CSUM |
+			       NETIF_F_SCTP_CRC |
+			       NETIF_F_TSO |
+			       NETIF_F_TSO6);
+		return;
+	}
 
 	/* We can only support IPv4 TSO in tunnels if we can mangle the
 	 * inner IP ID field, so strip TSO if MANGLEID is not supported.
 	 */
-	if (skb->encapsulation && !(features & NETIF_F_TSO_MANGLEID))
-		features &= ~NETIF_F_TSO;
-
-	return features;
+	if (skb->encapsulation && !(*features & NETIF_F_TSO_MANGLEID))
+		*features &= ~NETIF_F_TSO;
 }
 
 static void igc_tsync_interrupt(struct igc_adapter *adapter)
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
index 24e06ba6f5e9..e39d3983a455 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
@@ -10067,43 +10067,44 @@ static void ixgbe_fwd_del(struct net_device *pdev, void *priv)
 #define IXGBE_MAX_MAC_HDR_LEN		127
 #define IXGBE_MAX_NETWORK_HDR_LEN	511
 
-static netdev_features_t
-ixgbe_features_check(struct sk_buff *skb, struct net_device *dev,
-		     netdev_features_t features)
+static void ixgbe_features_check(struct sk_buff *skb, struct net_device *dev,
+				 netdev_features_t *features)
 {
 	unsigned int network_hdr_len, mac_hdr_len;
 
 	/* Make certain the headers can be described by a context descriptor */
 	mac_hdr_len = skb_network_header(skb) - skb->data;
-	if (unlikely(mac_hdr_len > IXGBE_MAX_MAC_HDR_LEN))
-		return features & ~(NETIF_F_HW_CSUM |
-				    NETIF_F_SCTP_CRC |
-				    NETIF_F_GSO_UDP_L4 |
-				    NETIF_F_HW_VLAN_CTAG_TX |
-				    NETIF_F_TSO |
-				    NETIF_F_TSO6);
+	if (unlikely(mac_hdr_len > IXGBE_MAX_MAC_HDR_LEN)) {
+		*features &= ~(NETIF_F_HW_CSUM |
+			       NETIF_F_SCTP_CRC |
+			       NETIF_F_GSO_UDP_L4 |
+			       NETIF_F_HW_VLAN_CTAG_TX |
+			       NETIF_F_TSO |
+			       NETIF_F_TSO6);
+		return;
+	}
 
 	network_hdr_len = skb_checksum_start(skb) - skb_network_header(skb);
-	if (unlikely(network_hdr_len >  IXGBE_MAX_NETWORK_HDR_LEN))
-		return features & ~(NETIF_F_HW_CSUM |
-				    NETIF_F_SCTP_CRC |
-				    NETIF_F_GSO_UDP_L4 |
-				    NETIF_F_TSO |
-				    NETIF_F_TSO6);
+	if (unlikely(network_hdr_len >  IXGBE_MAX_NETWORK_HDR_LEN)) {
+		*features &= ~(NETIF_F_HW_CSUM |
+			       NETIF_F_SCTP_CRC |
+			       NETIF_F_GSO_UDP_L4 |
+			       NETIF_F_TSO |
+			       NETIF_F_TSO6);
+		return;
+	}
 
 	/* We can only support IPV4 TSO in tunnels if we can mangle the
 	 * inner IP ID field, so strip TSO if MANGLEID is not supported.
 	 * IPsec offoad sets skb->encapsulation but still can handle
 	 * the TSO, so it's the exception.
 	 */
-	if (skb->encapsulation && !(features & NETIF_F_TSO_MANGLEID)) {
+	if (skb->encapsulation && !(*features & NETIF_F_TSO_MANGLEID)) {
 #ifdef CONFIG_IXGBE_IPSEC
 		if (!secpath_exists(skb))
 #endif
-			features &= ~NETIF_F_TSO;
+			*features &= ~NETIF_F_TSO;
 	}
-
-	return features;
 }
 
 static int ixgbe_xdp_setup(struct net_device *dev, struct bpf_prog *prog)
diff --git a/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c b/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c
index c714e1ecd308..f2f6d00960c5 100644
--- a/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c
+++ b/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c
@@ -4381,35 +4381,36 @@ static void ixgbevf_get_stats(struct net_device *netdev,
 #define IXGBEVF_MAX_MAC_HDR_LEN		127
 #define IXGBEVF_MAX_NETWORK_HDR_LEN	511
 
-static netdev_features_t
-ixgbevf_features_check(struct sk_buff *skb, struct net_device *dev,
-		       netdev_features_t features)
+static void ixgbevf_features_check(struct sk_buff *skb, struct net_device *dev,
+				   netdev_features_t *features)
 {
 	unsigned int network_hdr_len, mac_hdr_len;
 
 	/* Make certain the headers can be described by a context descriptor */
 	mac_hdr_len = skb_network_header(skb) - skb->data;
-	if (unlikely(mac_hdr_len > IXGBEVF_MAX_MAC_HDR_LEN))
-		return features & ~(NETIF_F_HW_CSUM |
-				    NETIF_F_SCTP_CRC |
-				    NETIF_F_HW_VLAN_CTAG_TX |
-				    NETIF_F_TSO |
-				    NETIF_F_TSO6);
+	if (unlikely(mac_hdr_len > IXGBEVF_MAX_MAC_HDR_LEN)) {
+		*features &= ~(NETIF_F_HW_CSUM |
+			       NETIF_F_SCTP_CRC |
+			       NETIF_F_HW_VLAN_CTAG_TX |
+			       NETIF_F_TSO |
+			       NETIF_F_TSO6);
+		return;
+	}
 
 	network_hdr_len = skb_checksum_start(skb) - skb_network_header(skb);
-	if (unlikely(network_hdr_len >  IXGBEVF_MAX_NETWORK_HDR_LEN))
-		return features & ~(NETIF_F_HW_CSUM |
-				    NETIF_F_SCTP_CRC |
-				    NETIF_F_TSO |
-				    NETIF_F_TSO6);
+	if (unlikely(network_hdr_len >  IXGBEVF_MAX_NETWORK_HDR_LEN)) {
+		*features &= ~(NETIF_F_HW_CSUM |
+			       NETIF_F_SCTP_CRC |
+			       NETIF_F_TSO |
+			       NETIF_F_TSO6);
+		return;
+	}
 
 	/* We can only support IPV4 TSO in tunnels if we can mangle the
 	 * inner IP ID field, so strip TSO if MANGLEID is not supported.
 	 */
-	if (skb->encapsulation && !(features & NETIF_F_TSO_MANGLEID))
-		features &= ~NETIF_F_TSO;
-
-	return features;
+	if (skb->encapsulation && !(*features & NETIF_F_TSO_MANGLEID))
+		*features &= ~NETIF_F_TSO;
 }
 
 static int ixgbevf_xdp_setup(struct net_device *dev, struct bpf_prog *prog)
diff --git a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
index dd6a41182446..354c63aa726d 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
@@ -2674,12 +2674,11 @@ static const struct udp_tunnel_nic_info mlx4_udp_tunnels = {
 	},
 };
 
-static netdev_features_t mlx4_en_features_check(struct sk_buff *skb,
-						struct net_device *dev,
-						netdev_features_t features)
+static void mlx4_en_features_check(struct sk_buff *skb, struct net_device *dev,
+				   netdev_features_t *features)
 {
-	vlan_features_check(skb, &features);
-	vxlan_features_check(skb, &features);
+	vlan_features_check(skb, features);
+	vxlan_features_check(skb, features);
 
 	/* The ConnectX-3 doesn't support outer IPv6 checksums but it does
 	 * support inner IPv6 checksums and segmentation so  we need to
@@ -2692,10 +2691,8 @@ static netdev_features_t mlx4_en_features_check(struct sk_buff *skb,
 		if (!priv->vxlan_port ||
 		    (ip_hdr(skb)->version != 4) ||
 		    (udp_hdr(skb)->dest != priv->vxlan_port))
-			features &= ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
+			*features &= ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
 	}
-
-	return features;
 }
 
 static int mlx4_en_set_tx_maxrate(struct net_device *dev, int queue_index, u32 maxrate)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h
index f5d147c3141a..daf276c43d8a 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h
@@ -1129,9 +1129,8 @@ void mlx5e_build_nic_params(struct mlx5e_priv *priv, struct mlx5e_xsk *xsk, u16
 void mlx5e_rx_dim_work(struct work_struct *work);
 void mlx5e_tx_dim_work(struct work_struct *work);
 
-netdev_features_t mlx5e_features_check(struct sk_buff *skb,
-				       struct net_device *netdev,
-				       netdev_features_t features);
+void mlx5e_features_check(struct sk_buff *skb, struct net_device *netdev,
+			  netdev_features_t *features);
 int mlx5e_set_features(struct net_device *netdev, netdev_features_t features);
 #ifdef CONFIG_MLX5_ESWITCH
 int mlx5e_set_vf_mac(struct net_device *dev, int vf, u8 *mac);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index 4518a490cdcd..6fdc2a793c1f 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -3904,21 +3904,18 @@ static netdev_features_t mlx5e_tunnel_features_check(struct mlx5e_priv *priv,
 	return features & ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
 }
 
-netdev_features_t mlx5e_features_check(struct sk_buff *skb,
-				       struct net_device *netdev,
-				       netdev_features_t features)
+void mlx5e_features_check(struct sk_buff *skb, struct net_device *netdev,
+			  netdev_features_t *features)
 {
 	struct mlx5e_priv *priv = netdev_priv(netdev);
 
-	vlan_features_check(skb, &features);
-	vxlan_features_check(skb, &features);
+	vlan_features_check(skb, features);
+	vxlan_features_check(skb, features);
 
 	/* Validate if the tunneled packet is being offloaded by HW */
 	if (skb->encapsulation &&
-	    (features & NETIF_F_CSUM_MASK || features & NETIF_F_GSO_MASK))
-		return mlx5e_tunnel_features_check(priv, skb, features);
-
-	return features;
+	    (*features & NETIF_F_CSUM_MASK || *features & NETIF_F_GSO_MASK))
+		*features = mlx5e_tunnel_features_check(priv, skb, *features);
 }
 
 static void mlx5e_tx_timeout_work(struct work_struct *work)
diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
index 58e7d98d0dd6..6261596cfda6 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
+++ b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
@@ -3598,17 +3598,16 @@ static int nfp_net_set_features(struct net_device *netdev,
 	return 0;
 }
 
-static netdev_features_t
-nfp_net_features_check(struct sk_buff *skb, struct net_device *dev,
-		       netdev_features_t features)
+static void nfp_net_features_check(struct sk_buff *skb, struct net_device *dev,
+				   netdev_features_t *features)
 {
 	u8 l4_hdr;
 
 	/* We can't do TSO over double tagged packets (802.1AD) */
-	vlan_features_check(skb, &features);
+	vlan_features_check(skb, features);
 
 	if (!skb->encapsulation)
-		return features;
+		return;
 
 	/* Ensure that inner L4 header offset fits into TX descriptor field */
 	if (skb_is_gso(skb)) {
@@ -3621,7 +3620,7 @@ nfp_net_features_check(struct sk_buff *skb, struct net_device *dev,
 		 * metadata prepend - 8B
 		 */
 		if (unlikely(hdrlen > NFP_NET_LSO_MAX_HDR_SZ - 8))
-			features &= ~NETIF_F_GSO_MASK;
+			*features &= ~NETIF_F_GSO_MASK;
 	}
 
 	/* VXLAN/GRE check */
@@ -3633,7 +3632,8 @@ nfp_net_features_check(struct sk_buff *skb, struct net_device *dev,
 		l4_hdr = ipv6_hdr(skb)->nexthdr;
 		break;
 	default:
-		return features & ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
+		*features &= ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
+		return;
 	}
 
 	if (skb->inner_protocol_type != ENCAP_TYPE_ETHER ||
@@ -3641,10 +3641,10 @@ nfp_net_features_check(struct sk_buff *skb, struct net_device *dev,
 	    (l4_hdr != IPPROTO_UDP && l4_hdr != IPPROTO_GRE) ||
 	    (l4_hdr == IPPROTO_UDP &&
 	     (skb_inner_mac_header(skb) - skb_transport_header(skb) !=
-	      sizeof(struct udphdr) + sizeof(struct vxlanhdr))))
-		return features & ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
-
-	return features;
+	      sizeof(struct udphdr) + sizeof(struct vxlanhdr)))) {
+		*features &= ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
+		return;
+	}
 }
 
 static int
diff --git a/drivers/net/ethernet/qlogic/qede/qede.h b/drivers/net/ethernet/qlogic/qede/qede.h
index f90dcfe9ee68..c1f26a2e374d 100644
--- a/drivers/net/ethernet/qlogic/qede/qede.h
+++ b/drivers/net/ethernet/qlogic/qede/qede.h
@@ -527,9 +527,8 @@ int qede_xdp_transmit(struct net_device *dev, int n_frames,
 		      struct xdp_frame **frames, u32 flags);
 u16 qede_select_queue(struct net_device *dev, struct sk_buff *skb,
 		      struct net_device *sb_dev);
-netdev_features_t qede_features_check(struct sk_buff *skb,
-				      struct net_device *dev,
-				      netdev_features_t features);
+void qede_features_check(struct sk_buff *skb, struct net_device *dev,
+			 netdev_features_t *features);
 int qede_alloc_rx_buffer(struct qede_rx_queue *rxq, bool allow_lazy);
 int qede_free_tx_pkt(struct qede_dev *edev,
 		     struct qede_tx_queue *txq, int *len);
diff --git a/drivers/net/ethernet/qlogic/qede/qede_fp.c b/drivers/net/ethernet/qlogic/qede/qede_fp.c
index 065e9004598e..8a459d2ca983 100644
--- a/drivers/net/ethernet/qlogic/qede/qede_fp.c
+++ b/drivers/net/ethernet/qlogic/qede/qede_fp.c
@@ -1747,9 +1747,8 @@ u16 qede_select_queue(struct net_device *dev, struct sk_buff *skb,
 /* 8B udp header + 8B base tunnel header + 32B option length */
 #define QEDE_MAX_TUN_HDR_LEN 48
 
-netdev_features_t qede_features_check(struct sk_buff *skb,
-				      struct net_device *dev,
-				      netdev_features_t features)
+void qede_features_check(struct sk_buff *skb, struct net_device *dev,
+			 netdev_features_t *features)
 {
 	if (skb->encapsulation) {
 		u8 l4_proto = 0;
@@ -1762,7 +1761,7 @@ netdev_features_t qede_features_check(struct sk_buff *skb,
 			l4_proto = ipv6_hdr(skb)->nexthdr;
 			break;
 		default:
-			return features;
+			return;
 		}
 
 		/* Disable offloads for geneve tunnels, as HW can't parse
@@ -1780,16 +1779,17 @@ netdev_features_t qede_features_check(struct sk_buff *skb,
 			if ((skb_inner_mac_header(skb) -
 			     skb_transport_header(skb)) > hdrlen ||
 			     (ntohs(udp_hdr(skb)->dest) != vxln_port &&
-			      ntohs(udp_hdr(skb)->dest) != gnv_port))
-				return features & ~(NETIF_F_CSUM_MASK |
-						    NETIF_F_GSO_MASK);
+			      ntohs(udp_hdr(skb)->dest) != gnv_port)) {
+				*features &= ~(NETIF_F_CSUM_MASK |
+					       NETIF_F_GSO_MASK);
+				return;
+			}
 		} else if (l4_proto == IPPROTO_IPIP) {
 			/* IPIP tunnels are unknown to the device or at least unsupported natively,
 			 * offloads for them can't be done trivially, so disable them for such skb.
 			 */
-			return features & ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
+			*features &= ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
+			return;
 		}
 	}
-
-	return features;
 }
diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c
index d823f2a22472..192d74665c20 100644
--- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c
+++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c
@@ -492,13 +492,11 @@ static const struct udp_tunnel_nic_info qlcnic_udp_tunnels = {
 	},
 };
 
-static netdev_features_t qlcnic_features_check(struct sk_buff *skb,
-					       struct net_device *dev,
-					       netdev_features_t features)
+static void qlcnic_features_check(struct sk_buff *skb, struct net_device *dev,
+				  netdev_features_t *features)
 {
-	vlan_features_check(skb, &features);
-	vxlan_features_check(skb, &features);
-	return features;
+	vlan_features_check(skb, features);
+	vxlan_features_check(skb, features);
 }
 
 static const struct net_device_ops qlcnic_netdev_ops = {
diff --git a/drivers/net/ethernet/realtek/8139cp.c b/drivers/net/ethernet/realtek/8139cp.c
index d2135fe99cd7..663a12598cad 100644
--- a/drivers/net/ethernet/realtek/8139cp.c
+++ b/drivers/net/ethernet/realtek/8139cp.c
@@ -1853,15 +1853,13 @@ static void cp_set_d3_state (struct cp_private *cp)
 	pci_set_power_state (cp->pdev, PCI_D3hot);
 }
 
-static netdev_features_t cp_features_check(struct sk_buff *skb,
-					   struct net_device *dev,
-					   netdev_features_t features)
+static void cp_features_check(struct sk_buff *skb, struct net_device *dev,
+			      netdev_features_t *features)
 {
 	if (skb_shinfo(skb)->gso_size > MSSMask)
-		features &= ~NETIF_F_TSO;
+		*features &= ~NETIF_F_TSO;
 
-	vlan_features_check(skb, &features);
-	return features;
+	vlan_features_check(skb, features);
 }
 static const struct net_device_ops cp_netdev_ops = {
 	.ndo_open		= cp_open,
diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c
index 98e7ffa6aa0a..3d753ddd1a89 100644
--- a/drivers/net/ethernet/realtek/r8169_main.c
+++ b/drivers/net/ethernet/realtek/r8169_main.c
@@ -4347,35 +4347,33 @@ static netdev_features_t rtl8168evl_fix_tso(struct sk_buff *skb,
 	return features;
 }
 
-static netdev_features_t rtl8169_features_check(struct sk_buff *skb,
-						struct net_device *dev,
-						netdev_features_t features)
+static void rtl8169_features_check(struct sk_buff *skb, struct net_device *dev,
+				   netdev_features_t *features)
 {
 	int transport_offset = skb_transport_offset(skb);
 	struct rtl8169_private *tp = netdev_priv(dev);
 
 	if (skb_is_gso(skb)) {
 		if (tp->mac_version == RTL_GIGA_MAC_VER_34)
-			features = rtl8168evl_fix_tso(skb, features);
+			*features = rtl8168evl_fix_tso(skb, *features);
 
 		if (transport_offset > GTTCPHO_MAX &&
 		    rtl_chip_supports_csum_v2(tp))
-			features &= ~NETIF_F_ALL_TSO;
+			*features &= ~NETIF_F_ALL_TSO;
 	} else if (skb->ip_summed == CHECKSUM_PARTIAL) {
 		/* work around hw bug on some chip versions */
 		if (skb->len < ETH_ZLEN)
-			features &= ~NETIF_F_CSUM_MASK;
+			*features &= ~NETIF_F_CSUM_MASK;
 
 		if (rtl_quirk_packet_padto(tp, skb))
-			features &= ~NETIF_F_CSUM_MASK;
+			*features &= ~NETIF_F_CSUM_MASK;
 
 		if (transport_offset > TCPHO_MAX &&
 		    rtl_chip_supports_csum_v2(tp))
-			features &= ~NETIF_F_CSUM_MASK;
+			*features &= ~NETIF_F_CSUM_MASK;
 	}
 
-	vlan_features_check(skb, &features);
-	return features;
+	vlan_features_check(skb, features);
 }
 
 static void rtl8169_pcierr_interrupt(struct net_device *dev)
diff --git a/drivers/net/ethernet/sfc/efx_common.c b/drivers/net/ethernet/sfc/efx_common.c
index 896b59253197..a101fa50a5cc 100644
--- a/drivers/net/ethernet/sfc/efx_common.c
+++ b/drivers/net/ethernet/sfc/efx_common.c
@@ -1353,25 +1353,24 @@ static bool efx_can_encap_offloads(struct efx_nic *efx, struct sk_buff *skb)
 	}
 }
 
-netdev_features_t efx_features_check(struct sk_buff *skb, struct net_device *dev,
-				     netdev_features_t features)
+void efx_features_check(struct sk_buff *skb, struct net_device *dev,
+			netdev_features_t *features)
 {
 	struct efx_nic *efx = netdev_priv(dev);
 
 	if (skb->encapsulation) {
-		if (features & NETIF_F_GSO_MASK)
+		if (*features & NETIF_F_GSO_MASK)
 			/* Hardware can only do TSO with at most 208 bytes
 			 * of headers.
 			 */
 			if (skb_inner_transport_offset(skb) >
 			    EFX_TSO2_MAX_HDRLEN)
-				features &= ~(NETIF_F_GSO_MASK);
-		if (features & (NETIF_F_GSO_MASK | NETIF_F_CSUM_MASK))
+				*features &= ~(NETIF_F_GSO_MASK);
+		if (*features & (NETIF_F_GSO_MASK | NETIF_F_CSUM_MASK))
 			if (!efx_can_encap_offloads(efx, skb))
-				features &= ~(NETIF_F_GSO_MASK |
+				*features &= ~(NETIF_F_GSO_MASK |
 					      NETIF_F_CSUM_MASK);
 	}
-	return features;
 }
 
 int efx_get_phys_port_id(struct net_device *net_dev,
diff --git a/drivers/net/ethernet/sfc/efx_common.h b/drivers/net/ethernet/sfc/efx_common.h
index 65513fd0cf6c..c4fea3997704 100644
--- a/drivers/net/ethernet/sfc/efx_common.h
+++ b/drivers/net/ethernet/sfc/efx_common.h
@@ -105,8 +105,8 @@ int efx_change_mtu(struct net_device *net_dev, int new_mtu);
 
 extern const struct pci_error_handlers efx_err_handlers;
 
-netdev_features_t efx_features_check(struct sk_buff *skb, struct net_device *dev,
-				     netdev_features_t features);
+void efx_features_check(struct sk_buff *skb, struct net_device *dev,
+			netdev_features_t *features);
 
 int efx_get_phys_port_id(struct net_device *net_dev,
 			 struct netdev_phys_item_id *ppid);
diff --git a/drivers/net/usb/lan78xx.c b/drivers/net/usb/lan78xx.c
index 9bfee75cbaf1..7a60cc0f6aad 100644
--- a/drivers/net/usb/lan78xx.c
+++ b/drivers/net/usb/lan78xx.c
@@ -3976,17 +3976,15 @@ static void lan78xx_tx_timeout(struct net_device *net, unsigned int txqueue)
 	tasklet_schedule(&dev->bh);
 }
 
-static netdev_features_t lan78xx_features_check(struct sk_buff *skb,
-						struct net_device *netdev,
-						netdev_features_t features)
+static void lan78xx_features_check(struct sk_buff *skb,
+				   struct net_device *netdev,
+				   netdev_features_t *features)
 {
 	if (skb->len + TX_OVERHEAD > MAX_SINGLE_PACKET_SIZE)
-		features &= ~NETIF_F_GSO_MASK;
+		*features &= ~NETIF_F_GSO_MASK;
 
-	vlan_features_check(skb, &features);
-	vxlan_features_check(skb, &features);
-
-	return features;
+	vlan_features_check(skb, features);
+	vxlan_features_check(skb, features);
 }
 
 static const struct net_device_ops lan78xx_netdev_ops = {
diff --git a/drivers/net/usb/r8152.c b/drivers/net/usb/r8152.c
index 60ba9b734055..5bb327b8d8e0 100644
--- a/drivers/net/usb/r8152.c
+++ b/drivers/net/usb/r8152.c
@@ -2741,20 +2741,17 @@ static void _rtl8152_set_rx_mode(struct net_device *netdev)
 	netif_wake_queue(netdev);
 }
 
-static netdev_features_t
-rtl8152_features_check(struct sk_buff *skb, struct net_device *dev,
-		       netdev_features_t features)
+static void rtl8152_features_check(struct sk_buff *skb, struct net_device *dev,
+				   netdev_features_t *features)
 {
 	u32 mss = skb_shinfo(skb)->gso_size;
 	int max_offset = mss ? GTTCPHO_MAX : TCPHO_MAX;
 	int offset = skb_transport_offset(skb);
 
 	if ((mss || skb->ip_summed == CHECKSUM_PARTIAL) && offset > max_offset)
-		features &= ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
+		*features &= ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
 	else if ((skb->len + sizeof(struct tx_desc)) > agg_buf_sz)
-		features &= ~NETIF_F_GSO_MASK;
-
-	return features;
+		*features &= ~NETIF_F_GSO_MASK;
 }
 
 static netdev_tx_t rtl8152_start_xmit(struct sk_buff *skb,
diff --git a/drivers/net/vmxnet3/vmxnet3_ethtool.c b/drivers/net/vmxnet3/vmxnet3_ethtool.c
index 5dd8360b21a0..e840547cd19e 100644
--- a/drivers/net/vmxnet3/vmxnet3_ethtool.c
+++ b/drivers/net/vmxnet3/vmxnet3_ethtool.c
@@ -256,9 +256,8 @@ netdev_features_t vmxnet3_fix_features(struct net_device *netdev,
 	return features;
 }
 
-netdev_features_t vmxnet3_features_check(struct sk_buff *skb,
-					 struct net_device *netdev,
-					 netdev_features_t features)
+void vmxnet3_features_check(struct sk_buff *skb, struct net_device *netdev,
+			    netdev_features_t *features)
 {
 	struct vmxnet3_adapter *adapter = netdev_priv(netdev);
 
@@ -277,7 +276,8 @@ netdev_features_t vmxnet3_features_check(struct sk_buff *skb,
 			l4_proto = ipv6_hdr(skb)->nexthdr;
 			break;
 		default:
-			return features & ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
+			*features &= ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
+			return;
 		}
 
 		switch (l4_proto) {
@@ -288,14 +288,15 @@ netdev_features_t vmxnet3_features_check(struct sk_buff *skb,
 			if (port != GENEVE_UDP_PORT &&
 			    port != IANA_VXLAN_UDP_PORT &&
 			    port != VXLAN_UDP_PORT) {
-				return features & ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
+				*features &= ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
+				return;
 			}
 			break;
 		default:
-			return features & ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
+			*features &= ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
+			return;
 		}
 	}
-	return features;
 }
 
 static void vmxnet3_enable_encap_offloads(struct net_device *netdev)
diff --git a/drivers/net/vmxnet3/vmxnet3_int.h b/drivers/net/vmxnet3/vmxnet3_int.h
index 7027ff483fa5..a711759ebb00 100644
--- a/drivers/net/vmxnet3/vmxnet3_int.h
+++ b/drivers/net/vmxnet3/vmxnet3_int.h
@@ -478,9 +478,9 @@ vmxnet3_rq_destroy_all(struct vmxnet3_adapter *adapter);
 netdev_features_t
 vmxnet3_fix_features(struct net_device *netdev, netdev_features_t features);
 
-netdev_features_t
+void
 vmxnet3_features_check(struct sk_buff *skb,
-		       struct net_device *netdev, netdev_features_t features);
+		       struct net_device *netdev, netdev_features_t *features);
 
 int
 vmxnet3_set_features(struct net_device *netdev, netdev_features_t features);
diff --git a/drivers/s390/net/qeth_core.h b/drivers/s390/net/qeth_core.h
index a5aa0bdc61d6..58bd0dc43695 100644
--- a/drivers/s390/net/qeth_core.h
+++ b/drivers/s390/net/qeth_core.h
@@ -1094,9 +1094,8 @@ int qeth_setassparms_cb(struct qeth_card *, struct qeth_reply *, unsigned long);
 int qeth_set_features(struct net_device *, netdev_features_t);
 void qeth_enable_hw_features(struct net_device *dev);
 netdev_features_t qeth_fix_features(struct net_device *, netdev_features_t);
-netdev_features_t qeth_features_check(struct sk_buff *skb,
-				      struct net_device *dev,
-				      netdev_features_t features);
+void qeth_features_check(struct sk_buff *skb, struct net_device *dev,
+			 netdev_features_t *features);
 void qeth_get_stats64(struct net_device *dev, struct rtnl_link_stats64 *stats);
 int qeth_set_real_num_tx_queues(struct qeth_card *card, unsigned int count);
 u16 qeth_iqd_select_queue(struct net_device *dev, struct sk_buff *skb,
diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c
index 3b1903f8790a..46b5ad0171fa 100644
--- a/drivers/s390/net/qeth_core_main.c
+++ b/drivers/s390/net/qeth_core_main.c
@@ -6946,9 +6946,8 @@ netdev_features_t qeth_fix_features(struct net_device *dev,
 }
 EXPORT_SYMBOL_GPL(qeth_fix_features);
 
-netdev_features_t qeth_features_check(struct sk_buff *skb,
-				      struct net_device *dev,
-				      netdev_features_t features)
+void qeth_features_check(struct sk_buff *skb, struct net_device *dev,
+			 netdev_features_t *features)
 {
 	struct qeth_card *card = dev->ml_priv;
 
@@ -6957,7 +6956,7 @@ netdev_features_t qeth_features_check(struct sk_buff *skb,
 	    READ_ONCE(card->options.isolation) != ISOLATION_MODE_FWD) {
 		netdev_features_t restricted = 0;
 
-		if (skb_is_gso(skb) && !netif_needs_gso(skb, features))
+		if (skb_is_gso(skb) && !netif_needs_gso(skb, *features))
 			restricted |= NETIF_F_ALL_TSO;
 
 		switch (vlan_get_protocol(skb)) {
@@ -6966,14 +6965,14 @@ netdev_features_t qeth_features_check(struct sk_buff *skb,
 				restricted |= NETIF_F_IP_CSUM;
 
 			if (restricted && qeth_next_hop_is_local_v4(card, skb))
-				features &= ~restricted;
+				*features &= ~restricted;
 			break;
 		case htons(ETH_P_IPV6):
 			if (!card->info.has_lp2lp_cso_v6)
 				restricted |= NETIF_F_IPV6_CSUM;
 
 			if (restricted && qeth_next_hop_is_local_v6(card, skb))
-				features &= ~restricted;
+				*features &= ~restricted;
 			break;
 		default:
 			break;
@@ -6987,7 +6986,7 @@ netdev_features_t qeth_features_check(struct sk_buff *skb,
 	 * additional buffer element. This reduces buffer utilization, and
 	 * hurts throughput. So compress small segments into one element.
 	 */
-	if (netif_needs_gso(skb, features)) {
+	if (netif_needs_gso(skb, *features)) {
 		/* match skb_segment(): */
 		unsigned int doffset = skb->data - skb_mac_header(skb);
 		unsigned int hsize = skb_shinfo(skb)->gso_size;
@@ -6995,11 +6994,10 @@ netdev_features_t qeth_features_check(struct sk_buff *skb,
 
 		/* linearize only if resulting skb allocations are order-0: */
 		if (SKB_DATA_ALIGN(hroom + doffset + hsize) <= SKB_MAX_HEAD(0))
-			features &= ~NETIF_F_SG;
+			*features &= ~NETIF_F_SG;
 	}
 
-	vlan_features_check(skb, &features);
-	return features;
+	vlan_features_check(skb, features);
 }
 EXPORT_SYMBOL_GPL(qeth_features_check);
 
diff --git a/drivers/s390/net/qeth_l3_main.c b/drivers/s390/net/qeth_l3_main.c
index 6fd3e288f059..1d3d0377fe4b 100644
--- a/drivers/s390/net/qeth_l3_main.c
+++ b/drivers/s390/net/qeth_l3_main.c
@@ -1804,13 +1804,13 @@ qeth_l3_neigh_setup(struct net_device *dev, struct neigh_parms *np)
 	return 0;
 }
 
-static netdev_features_t qeth_l3_osa_features_check(struct sk_buff *skb,
-						    struct net_device *dev,
-						    netdev_features_t features)
+static void qeth_l3_osa_features_check(struct sk_buff *skb,
+				       struct net_device *dev,
+				       netdev_features_t *features)
 {
 	if (vlan_get_protocol(skb) != htons(ETH_P_IP))
-		features &= ~NETIF_F_HW_VLAN_CTAG_TX;
-	return qeth_features_check(skb, dev, features);
+		*features &= ~NETIF_F_HW_VLAN_CTAG_TX;
+	qeth_features_check(skb, dev, features);
 }
 
 static u16 qeth_l3_iqd_select_queue(struct net_device *dev, struct sk_buff *skb,
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index a5598d617789..09233e7df8f1 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -1352,9 +1352,9 @@ struct net_device_ops {
 	int			(*ndo_stop)(struct net_device *dev);
 	netdev_tx_t		(*ndo_start_xmit)(struct sk_buff *skb,
 						  struct net_device *dev);
-	netdev_features_t	(*ndo_features_check)(struct sk_buff *skb,
+	void			(*ndo_features_check)(struct sk_buff *skb,
 						      struct net_device *dev,
-						      netdev_features_t features);
+						      netdev_features_t *features);
 	u16			(*ndo_select_queue)(struct net_device *dev,
 						    struct sk_buff *skb,
 						    struct net_device *sb_dev);
@@ -5056,9 +5056,8 @@ void netdev_change_features(struct net_device *dev);
 void netif_stacked_transfer_operstate(const struct net_device *rootdev,
 					struct net_device *dev);
 
-netdev_features_t passthru_features_check(struct sk_buff *skb,
-					  struct net_device *dev,
-					  netdev_features_t features);
+void passthru_features_check(struct sk_buff *skb, struct net_device *dev,
+			     netdev_features_t *features);
 netdev_features_t netif_skb_features(struct sk_buff *skb);
 
 static inline bool net_gso_ok(netdev_features_t features, int gso_type)
diff --git a/net/core/dev.c b/net/core/dev.c
index 814e6e7ee579..43b81dc6b815 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -3472,11 +3472,9 @@ static void harmonize_features(struct sk_buff *skb, netdev_features_t *features)
 		*features &= ~NETIF_F_SG;
 }
 
-netdev_features_t passthru_features_check(struct sk_buff *skb,
-					  struct net_device *dev,
-					  netdev_features_t features)
+void passthru_features_check(struct sk_buff *skb, struct net_device *dev,
+			     netdev_features_t *features)
 {
-	return features;
 }
 EXPORT_SYMBOL(passthru_features_check);
 
@@ -3547,8 +3545,7 @@ netdev_features_t netif_skb_features(struct sk_buff *skb)
 					  NETIF_F_HW_VLAN_STAG_TX);
 
 	if (dev->netdev_ops->ndo_features_check)
-		features &= dev->netdev_ops->ndo_features_check(skb, dev,
-								features);
+		dev->netdev_ops->ndo_features_check(skb, dev, &features);
 	else
 		dflt_features_check(skb, dev, &features);
 
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 010/167] net: convert the prototype of netif_skb_features
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (8 preceding siblings ...)
  2021-09-29 15:50 ` [RFCv2 net-next 009/167] net: convert the prototype of ndo_features_check Jian Shen
@ 2021-09-29 15:50 ` Jian Shen
  2021-09-29 15:50 ` [RFCv2 net-next 011/167] net: convert the prototype of ndo_fix_features Jian Shen
                   ` (160 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:50 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

For the origin type for netdev_features_t would be changed to
be unsigned long * from u64, so changes the prototype of
netif_skb_features for adaption.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 drivers/net/xen-netfront.c |  9 +++++++--
 include/linux/netdevice.h  |  2 +-
 net/core/dev.c             | 20 ++++++++++----------
 net/core/netpoll.c         |  2 +-
 net/ipv4/ip_output.c       |  2 +-
 net/ipv6/ip6_output.c      |  2 +-
 net/sched/sch_cake.c       |  3 ++-
 net/sched/sch_netem.c      |  3 ++-
 net/sched/sch_taprio.c     |  3 ++-
 net/sched/sch_tbf.c        |  3 ++-
 10 files changed, 29 insertions(+), 20 deletions(-)

diff --git a/drivers/net/xen-netfront.c b/drivers/net/xen-netfront.c
index e31b98403f31..65c134ac2be5 100644
--- a/drivers/net/xen-netfront.c
+++ b/drivers/net/xen-netfront.c
@@ -679,6 +679,7 @@ static netdev_tx_t xennet_start_xmit(struct sk_buff *skb, struct net_device *dev
 	struct netfront_queue *queue = NULL;
 	struct xennet_gnttab_make_txreq info = { };
 	unsigned int num_queues = dev->real_num_tx_queues;
+	netdev_features_t features;
 	u16 queue_index;
 	struct sk_buff *nskb;
 
@@ -730,8 +731,12 @@ static netdev_tx_t xennet_start_xmit(struct sk_buff *skb, struct net_device *dev
 	spin_lock_irqsave(&queue->tx_lock, flags);
 
 	if (unlikely(!netif_carrier_ok(dev) ||
-		     (slots > 1 && !xennet_can_sg(dev)) ||
-		     netif_needs_gso(skb, netif_skb_features(skb)))) {
+		     (slots > 1 && !xennet_can_sg(dev)))) {
+		spin_unlock_irqrestore(&queue->tx_lock, flags);
+		goto drop;
+	}
+	netif_skb_features(skb, &features);
+	if (unlikely(netif_needs_gso(skb, features))) {
 		spin_unlock_irqrestore(&queue->tx_lock, flags);
 		goto drop;
 	}
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 09233e7df8f1..d62edd4c99a9 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -5058,7 +5058,7 @@ void netif_stacked_transfer_operstate(const struct net_device *rootdev,
 
 void passthru_features_check(struct sk_buff *skb, struct net_device *dev,
 			     netdev_features_t *features);
-netdev_features_t netif_skb_features(struct sk_buff *skb);
+void netif_skb_features(struct sk_buff *skb, netdev_features_t *features);
 
 static inline bool net_gso_ok(netdev_features_t features, int gso_type)
 {
diff --git a/net/core/dev.c b/net/core/dev.c
index 43b81dc6b815..8f6316bee565 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -3523,34 +3523,34 @@ static void gso_features_check(const struct sk_buff *skb,
 	}
 }
 
-netdev_features_t netif_skb_features(struct sk_buff *skb)
+void netif_skb_features(struct sk_buff *skb, netdev_features_t *features)
 {
 	struct net_device *dev = skb->dev;
-	netdev_features_t features = dev->features;
+
+	*features = dev->features;
 
 	if (skb_is_gso(skb))
-		gso_features_check(skb, dev, &features);
+		gso_features_check(skb, dev, features);
 
 	/* If encapsulation offload request, verify we are testing
 	 * hardware encapsulation features instead of standard
 	 * features for the netdev
 	 */
 	if (skb->encapsulation)
-		features &= dev->hw_enc_features;
+		*features &= dev->hw_enc_features;
 
 	if (skb_vlan_tagged(skb))
-		netdev_intersect_features(&features, features,
+		netdev_intersect_features(features, *features,
 					  dev->vlan_features |
 					  NETIF_F_HW_VLAN_CTAG_TX |
 					  NETIF_F_HW_VLAN_STAG_TX);
 
 	if (dev->netdev_ops->ndo_features_check)
-		dev->netdev_ops->ndo_features_check(skb, dev, &features);
+		dev->netdev_ops->ndo_features_check(skb, dev, features);
 	else
-		dflt_features_check(skb, dev, &features);
+		dflt_features_check(skb, dev, features);
 
-	harmonize_features(skb, &features);
-	return features;
+	harmonize_features(skb, features);
 }
 EXPORT_SYMBOL(netif_skb_features);
 
@@ -3635,7 +3635,7 @@ static struct sk_buff *validate_xmit_skb(struct sk_buff *skb, struct net_device
 {
 	netdev_features_t features;
 
-	features = netif_skb_features(skb);
+	netif_skb_features(skb, &features);
 	skb = validate_xmit_vlan(skb, features);
 	if (unlikely(!skb))
 		goto out_null;
diff --git a/net/core/netpoll.c b/net/core/netpoll.c
index edfc0f8011f8..d2fc92b89a5f 100644
--- a/net/core/netpoll.c
+++ b/net/core/netpoll.c
@@ -77,7 +77,7 @@ static netdev_tx_t netpoll_start_xmit(struct sk_buff *skb,
 	netdev_tx_t status = NETDEV_TX_OK;
 	netdev_features_t features;
 
-	features = netif_skb_features(skb);
+	netif_skb_features(skb, &features);
 
 	if (skb_vlan_tag_present(skb) &&
 	    !vlan_hw_offload_capable(features, skb->vlan_proto)) {
diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c
index 9bca57ef8b83..8d552f1b7f62 100644
--- a/net/ipv4/ip_output.c
+++ b/net/ipv4/ip_output.c
@@ -255,7 +255,7 @@ static int ip_finish_output_gso(struct net *net, struct sock *sk,
 	 *    bridged to a NETIF_F_TSO tunnel stacked over an interface with an
 	 *    insufficient MTU.
 	 */
-	features = netif_skb_features(skb);
+	netif_skb_features(skb, &features);
 	BUILD_BUG_ON(sizeof(*IPCB(skb)) > SKB_GSO_CB_OFFSET);
 	segs = skb_gso_segment(skb, features & ~NETIF_F_GSO_MASK);
 	if (IS_ERR_OR_NULL(segs)) {
diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c
index 12f985f43bcc..46a0867bcedf 100644
--- a/net/ipv6/ip6_output.c
+++ b/net/ipv6/ip6_output.c
@@ -146,7 +146,7 @@ ip6_finish_output_gso_slowpath_drop(struct net *net, struct sock *sk,
 	 * describing the cases where GSO segment length exceeds the
 	 * egress MTU.
 	 */
-	features = netif_skb_features(skb);
+	netif_skb_features(skb, &features);
 	segs = skb_gso_segment(skb, features & ~NETIF_F_GSO_MASK);
 	if (IS_ERR_OR_NULL(segs)) {
 		kfree_skb(skb);
diff --git a/net/sched/sch_cake.c b/net/sched/sch_cake.c
index 3c2300d14468..e650ec5dc791 100644
--- a/net/sched/sch_cake.c
+++ b/net/sched/sch_cake.c
@@ -1740,9 +1740,10 @@ static s32 cake_enqueue(struct sk_buff *skb, struct Qdisc *sch,
 
 	if (skb_is_gso(skb) && q->rate_flags & CAKE_FLAG_SPLIT_GSO) {
 		struct sk_buff *segs, *nskb;
-		netdev_features_t features = netif_skb_features(skb);
+		netdev_features_t features;
 		unsigned int slen = 0, numsegs = 0;
 
+		netif_skb_features(skb, &features);
 		segs = skb_gso_segment(skb, features & ~NETIF_F_GSO_MASK);
 		if (IS_ERR_OR_NULL(segs))
 			return qdisc_drop(skb, sch, to_free);
diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c
index 0c345e43a09a..414d57e017b9 100644
--- a/net/sched/sch_netem.c
+++ b/net/sched/sch_netem.c
@@ -413,8 +413,9 @@ static struct sk_buff *netem_segment(struct sk_buff *skb, struct Qdisc *sch,
 				     struct sk_buff **to_free)
 {
 	struct sk_buff *segs;
-	netdev_features_t features = netif_skb_features(skb);
+	netdev_features_t features;
 
+	netif_skb_features(skb, &features);
 	segs = skb_gso_segment(skb, features & ~NETIF_F_GSO_MASK);
 
 	if (IS_ERR_OR_NULL(segs)) {
diff --git a/net/sched/sch_taprio.c b/net/sched/sch_taprio.c
index 1ab2fc933a21..d7fe4a2cc14f 100644
--- a/net/sched/sch_taprio.c
+++ b/net/sched/sch_taprio.c
@@ -453,10 +453,11 @@ static int taprio_enqueue(struct sk_buff *skb, struct Qdisc *sch,
 	 */
 	if (skb_is_gso(skb) && !FULL_OFFLOAD_IS_ENABLED(q->flags)) {
 		unsigned int slen = 0, numsegs = 0, len = qdisc_pkt_len(skb);
-		netdev_features_t features = netif_skb_features(skb);
+		netdev_features_t features;
 		struct sk_buff *segs, *nskb;
 		int ret;
 
+		netif_skb_features(skb, &features);
 		segs = skb_gso_segment(skb, features & ~NETIF_F_GSO_MASK);
 		if (IS_ERR_OR_NULL(segs))
 			return qdisc_drop(skb, sch, to_free);
diff --git a/net/sched/sch_tbf.c b/net/sched/sch_tbf.c
index 78e79029dc63..99e6d7265e7f 100644
--- a/net/sched/sch_tbf.c
+++ b/net/sched/sch_tbf.c
@@ -192,10 +192,11 @@ static int tbf_segment(struct sk_buff *skb, struct Qdisc *sch,
 {
 	struct tbf_sched_data *q = qdisc_priv(sch);
 	struct sk_buff *segs, *nskb;
-	netdev_features_t features = netif_skb_features(skb);
+	netdev_features_t features;
 	unsigned int len = 0, prev_len = qdisc_pkt_len(skb);
 	int ret, nb;
 
+	netif_skb_features(skb, &features);
 	segs = skb_gso_segment(skb, features & ~NETIF_F_GSO_MASK);
 
 	if (IS_ERR_OR_NULL(segs))
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 011/167] net: convert the prototype of ndo_fix_features
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (9 preceding siblings ...)
  2021-09-29 15:50 ` [RFCv2 net-next 010/167] net: convert the prototype of netif_skb_features Jian Shen
@ 2021-09-29 15:50 ` Jian Shen
  2021-09-29 15:50 ` [RFCv2 net-next 012/167] net: convert the prototype of netdev_fix_features Jian Shen
                   ` (159 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:50 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

For the origin type for netdev_features_t would be changed to
be unsigned long * from u64, so changes the prototype of
ndo_fix_features for adaption.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 arch/um/drivers/vector_kern.c                 |  7 ++--
 drivers/infiniband/ulp/ipoib/ipoib_main.c     |  7 ++--
 drivers/net/bonding/bond_main.c               | 24 ++++++-------
 drivers/net/ethernet/amd/xgbe/xgbe-drv.c      | 28 +++++++--------
 drivers/net/ethernet/atheros/alx/main.c       |  8 ++---
 .../net/ethernet/atheros/atl1c/atl1c_main.c   | 14 ++++----
 .../net/ethernet/atheros/atl1e/atl1e_main.c   | 12 +++----
 drivers/net/ethernet/atheros/atlx/atl2.c      | 12 +++----
 drivers/net/ethernet/atheros/atlx/atlx.c      | 12 +++----
 .../net/ethernet/broadcom/bnx2x/bnx2x_cmn.c   | 29 +++++++---------
 .../net/ethernet/broadcom/bnx2x/bnx2x_cmn.h   |  3 +-
 drivers/net/ethernet/broadcom/bnxt/bnxt.c     | 27 +++++++--------
 drivers/net/ethernet/broadcom/tg3.c           |  8 ++---
 .../net/ethernet/cavium/liquidio/lio_main.c   | 34 +++++++++----------
 .../ethernet/cavium/liquidio/lio_vf_main.c    | 30 ++++++++--------
 .../net/ethernet/cavium/thunder/nicvf_main.c  | 10 +++---
 drivers/net/ethernet/chelsio/cxgb/cxgb2.c     | 11 +++---
 .../net/ethernet/chelsio/cxgb3/cxgb3_main.c   | 12 +++----
 .../net/ethernet/chelsio/cxgb4/cxgb4_main.c   | 10 +++---
 .../ethernet/chelsio/cxgb4vf/cxgb4vf_main.c   | 12 +++----
 drivers/net/ethernet/cortina/gemini.c         |  8 ++---
 drivers/net/ethernet/hisilicon/hns/hns_enet.c |  7 ++--
 .../net/ethernet/huawei/hinic/hinic_main.c    | 10 +++---
 drivers/net/ethernet/ibm/ibmveth.c            | 10 +++---
 drivers/net/ethernet/intel/e1000/e1000_main.c | 12 +++----
 drivers/net/ethernet/intel/e1000e/netdev.c    | 14 ++++----
 drivers/net/ethernet/intel/iavf/iavf_main.c   |  8 ++---
 drivers/net/ethernet/intel/igb/igb_main.c     | 12 +++----
 drivers/net/ethernet/intel/igc/igc_main.c     | 12 +++----
 drivers/net/ethernet/intel/ixgb/ixgb_main.c   | 10 +++---
 drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 16 ++++-----
 drivers/net/ethernet/jme.c                    |  7 ++--
 drivers/net/ethernet/marvell/mvneta.c         |  8 ++---
 .../ethernet/marvell/octeontx2/nic/otx2_pf.c  | 12 +++----
 drivers/net/ethernet/marvell/sky2.c           | 16 ++++-----
 drivers/net/ethernet/mediatek/mtk_eth_soc.c   | 10 +++---
 .../net/ethernet/mellanox/mlx4/en_netdev.c    | 12 +++----
 .../net/ethernet/mellanox/mlx5/core/en_main.c | 14 ++++----
 .../net/ethernet/neterion/vxge/vxge-main.c    | 10 +++---
 .../net/ethernet/netronome/nfp/nfp_net_repr.c | 14 ++++----
 drivers/net/ethernet/nvidia/forcedeth.c       | 10 +++---
 .../ethernet/qlogic/netxen/netxen_nic_main.c  | 10 +++---
 drivers/net/ethernet/qlogic/qede/qede.h       |  3 +-
 .../net/ethernet/qlogic/qede/qede_filter.c    |  9 ++---
 drivers/net/ethernet/qlogic/qlcnic/qlcnic.h   |  3 +-
 .../net/ethernet/qlogic/qlcnic/qlcnic_hw.c    | 15 ++++----
 drivers/net/ethernet/realtek/r8169_main.c     | 10 +++---
 .../net/ethernet/stmicro/stmmac/stmmac_main.c | 14 ++++----
 drivers/net/hyperv/netvsc_drv.c               | 12 +++----
 drivers/net/ipvlan/ipvlan_main.c              | 18 +++++-----
 drivers/net/macsec.c                          | 16 ++++-----
 drivers/net/macvlan.c                         | 20 +++++------
 drivers/net/team/team.c                       | 20 +++++------
 drivers/net/tun.c                             |  7 ++--
 drivers/net/veth.c                            | 10 +++---
 drivers/net/vmxnet3/vmxnet3_ethtool.c         | 10 +++---
 drivers/net/vmxnet3/vmxnet3_int.h             |  4 +--
 drivers/net/xen-netback/interface.c           | 16 ++++-----
 drivers/net/xen-netfront.c                    | 22 ++++++------
 drivers/s390/net/qeth_core.h                  |  2 +-
 drivers/s390/net/qeth_core_main.c             | 16 ++++-----
 include/linux/netdevice.h                     |  4 +--
 net/8021q/vlan_dev.c                          | 14 ++++----
 net/bridge/br_device.c                        |  5 ++-
 net/core/dev.c                                |  2 +-
 net/hsr/hsr_device.c                          |  6 ++--
 66 files changed, 349 insertions(+), 461 deletions(-)

diff --git a/arch/um/drivers/vector_kern.c b/arch/um/drivers/vector_kern.c
index cde6db184c26..88cc24a58742 100644
--- a/arch/um/drivers/vector_kern.c
+++ b/arch/um/drivers/vector_kern.c
@@ -1342,11 +1342,10 @@ static void vector_net_tx_timeout(struct net_device *dev, unsigned int txqueue)
 	schedule_work(&vp->reset_tx);
 }
 
-static netdev_features_t vector_fix_features(struct net_device *dev,
-	netdev_features_t features)
+static void vector_fix_features(struct net_device *dev,
+				netdev_features_t *features)
 {
-	features &= ~(NETIF_F_IP_CSUM|NETIF_F_IPV6_CSUM);
-	return features;
+	*features &= ~(NETIF_F_IP_CSUM|NETIF_F_IPV6_CSUM);
 }
 
 static int vector_set_features(struct net_device *dev,
diff --git a/drivers/infiniband/ulp/ipoib/ipoib_main.c b/drivers/infiniband/ulp/ipoib/ipoib_main.c
index 0aa8629fdf62..488d50a82a87 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib_main.c
+++ b/drivers/infiniband/ulp/ipoib/ipoib_main.c
@@ -214,14 +214,13 @@ static int ipoib_stop(struct net_device *dev)
 	return 0;
 }
 
-static netdev_features_t ipoib_fix_features(struct net_device *dev, netdev_features_t features)
+static void ipoib_fix_features(struct net_device *dev,
+			       netdev_features_t *features)
 {
 	struct ipoib_dev_priv *priv = ipoib_priv(dev);
 
 	if (test_bit(IPOIB_FLAG_ADMIN_CM, &priv->flags))
-		features &= ~(NETIF_F_IP_CSUM | NETIF_F_TSO);
-
-	return features;
+		*features &= ~(NETIF_F_IP_CSUM | NETIF_F_TSO);
 }
 
 static int ipoib_change_mtu(struct net_device *dev, int new_mtu)
diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index 77dc79a7f574..469509260a51 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -1361,8 +1361,8 @@ static void bond_netpoll_cleanup(struct net_device *bond_dev)
 
 /*---------------------------------- IOCTL ----------------------------------*/
 
-static netdev_features_t bond_fix_features(struct net_device *dev,
-					   netdev_features_t features)
+static void bond_fix_features(struct net_device *dev,
+			      netdev_features_t *features)
 {
 	struct bonding *bond = netdev_priv(dev);
 	struct list_head *iter;
@@ -1371,24 +1371,22 @@ static netdev_features_t bond_fix_features(struct net_device *dev,
 
 #if IS_ENABLED(CONFIG_TLS_DEVICE)
 	if (bond_sk_check(bond))
-		features |= BOND_TLS_FEATURES;
+		*features |= BOND_TLS_FEATURES;
 	else
-		features &= ~BOND_TLS_FEATURES;
+		*features &= ~BOND_TLS_FEATURES;
 #endif
 
-	mask = features;
+	mask = *features;
 
-	features &= ~NETIF_F_ONE_FOR_ALL;
-	features |= NETIF_F_ALL_FOR_ALL;
+	*features &= ~NETIF_F_ONE_FOR_ALL;
+	*features |= NETIF_F_ALL_FOR_ALL;
 
 	bond_for_each_slave(bond, slave, iter) {
-		features = netdev_increment_features(features,
-						     slave->dev->features,
-						     mask);
+		*features = netdev_increment_features(*features,
+						      slave->dev->features,
+						      mask);
 	}
-	features = netdev_add_tso_features(features, mask);
-
-	return features;
+	*features = netdev_add_tso_features(*features, mask);
 }
 
 #define BOND_VLAN_FEATURES	(NETIF_F_HW_CSUM | NETIF_F_SG | \
diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-drv.c b/drivers/net/ethernet/amd/xgbe/xgbe-drv.c
index cc1f1a7a46ae..dff9eecac6e9 100644
--- a/drivers/net/ethernet/amd/xgbe/xgbe-drv.c
+++ b/drivers/net/ethernet/amd/xgbe/xgbe-drv.c
@@ -2182,8 +2182,8 @@ static int xgbe_setup_tc(struct net_device *netdev, enum tc_setup_type type,
 	return 0;
 }
 
-static netdev_features_t xgbe_fix_features(struct net_device *netdev,
-					   netdev_features_t features)
+static void xgbe_fix_features(struct net_device *netdev,
+			      netdev_features_t *features)
 {
 	struct xgbe_prv_data *pdata = netdev_priv(netdev);
 	netdev_features_t vxlan_base;
@@ -2191,38 +2191,36 @@ static netdev_features_t xgbe_fix_features(struct net_device *netdev,
 	vxlan_base = NETIF_F_GSO_UDP_TUNNEL | NETIF_F_RX_UDP_TUNNEL_PORT;
 
 	if (!pdata->hw_feat.vxn)
-		return features;
+		return;
 
 	/* VXLAN CSUM requires VXLAN base */
-	if ((features & NETIF_F_GSO_UDP_TUNNEL_CSUM) &&
-	    !(features & NETIF_F_GSO_UDP_TUNNEL)) {
+	if ((*features & NETIF_F_GSO_UDP_TUNNEL_CSUM) &&
+	    !(*features & NETIF_F_GSO_UDP_TUNNEL)) {
 		netdev_notice(netdev,
 			      "forcing tx udp tunnel support\n");
-		features |= NETIF_F_GSO_UDP_TUNNEL;
+		*features |= NETIF_F_GSO_UDP_TUNNEL;
 	}
 
 	/* Can't do one without doing the other */
-	if ((features & vxlan_base) != vxlan_base) {
+	if ((*features & vxlan_base) != vxlan_base) {
 		netdev_notice(netdev,
 			      "forcing both tx and rx udp tunnel support\n");
-		features |= vxlan_base;
+		*features |= vxlan_base;
 	}
 
-	if (features & (NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM)) {
-		if (!(features & NETIF_F_GSO_UDP_TUNNEL_CSUM)) {
+	if (*features & (NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM)) {
+		if (!(*features & NETIF_F_GSO_UDP_TUNNEL_CSUM)) {
 			netdev_notice(netdev,
 				      "forcing tx udp tunnel checksumming on\n");
-			features |= NETIF_F_GSO_UDP_TUNNEL_CSUM;
+			*features |= NETIF_F_GSO_UDP_TUNNEL_CSUM;
 		}
 	} else {
-		if (features & NETIF_F_GSO_UDP_TUNNEL_CSUM) {
+		if (*features & NETIF_F_GSO_UDP_TUNNEL_CSUM) {
 			netdev_notice(netdev,
 				      "forcing tx udp tunnel checksumming off\n");
-			features &= ~NETIF_F_GSO_UDP_TUNNEL_CSUM;
+			*features &= ~NETIF_F_GSO_UDP_TUNNEL_CSUM;
 		}
 	}
-
-	return features;
 }
 
 static int xgbe_set_features(struct net_device *netdev,
diff --git a/drivers/net/ethernet/atheros/alx/main.c b/drivers/net/ethernet/atheros/alx/main.c
index 4ea157efca86..922c600fd292 100644
--- a/drivers/net/ethernet/atheros/alx/main.c
+++ b/drivers/net/ethernet/atheros/alx/main.c
@@ -1097,13 +1097,11 @@ static int alx_init_sw(struct alx_priv *alx)
 }
 
 
-static netdev_features_t alx_fix_features(struct net_device *netdev,
-					  netdev_features_t features)
+static void alx_fix_features(struct net_device *netdev,
+			     netdev_features_t *features)
 {
 	if (netdev->mtu > ALX_MAX_TSO_PKT_SIZE)
-		features &= ~(NETIF_F_TSO | NETIF_F_TSO6);
-
-	return features;
+		*features &= ~(NETIF_F_TSO | NETIF_F_TSO6);
 }
 
 static void alx_netif_stop(struct alx_priv *alx)
diff --git a/drivers/net/ethernet/atheros/atl1c/atl1c_main.c b/drivers/net/ethernet/atheros/atl1c/atl1c_main.c
index 1c258e4ddc96..66c0985adb43 100644
--- a/drivers/net/ethernet/atheros/atl1c/atl1c_main.c
+++ b/drivers/net/ethernet/atheros/atl1c/atl1c_main.c
@@ -504,8 +504,8 @@ static void atl1c_set_rxbufsize(struct atl1c_adapter *adapter,
 	adapter->rx_frag_size = roundup_pow_of_two(head_size);
 }
 
-static netdev_features_t atl1c_fix_features(struct net_device *netdev,
-	netdev_features_t features)
+static void atl1c_fix_features(struct net_device *netdev,
+			       netdev_features_t *features)
 {
 	struct atl1c_adapter *adapter = netdev_priv(netdev);
 	struct atl1c_hw *hw = &adapter->hw;
@@ -514,17 +514,15 @@ static netdev_features_t atl1c_fix_features(struct net_device *netdev,
 	 * Since there is no support for separate rx/tx vlan accel
 	 * enable/disable make sure tx flag is always in same state as rx.
 	 */
-	if (features & NETIF_F_HW_VLAN_CTAG_RX)
-		features |= NETIF_F_HW_VLAN_CTAG_TX;
+	if (*features & NETIF_F_HW_VLAN_CTAG_RX)
+		*features |= NETIF_F_HW_VLAN_CTAG_TX;
 	else
-		features &= ~NETIF_F_HW_VLAN_CTAG_TX;
+		*features &= ~NETIF_F_HW_VLAN_CTAG_TX;
 
 	if (hw->nic_type != athr_mt) {
 		if (netdev->mtu > MAX_TSO_FRAME_SIZE)
-			features &= ~(NETIF_F_TSO | NETIF_F_TSO6);
+			*features &= ~(NETIF_F_TSO | NETIF_F_TSO6);
 	}
-
-	return features;
 }
 
 static int atl1c_set_features(struct net_device *netdev,
diff --git a/drivers/net/ethernet/atheros/atl1e/atl1e_main.c b/drivers/net/ethernet/atheros/atl1e/atl1e_main.c
index 2e22483a9040..ea99949c91eb 100644
--- a/drivers/net/ethernet/atheros/atl1e/atl1e_main.c
+++ b/drivers/net/ethernet/atheros/atl1e/atl1e_main.c
@@ -382,19 +382,17 @@ static int atl1e_set_mac_addr(struct net_device *netdev, void *p)
 	return 0;
 }
 
-static netdev_features_t atl1e_fix_features(struct net_device *netdev,
-	netdev_features_t features)
+static void atl1e_fix_features(struct net_device *netdev,
+			       netdev_features_t *features)
 {
 	/*
 	 * Since there is no support for separate rx/tx vlan accel
 	 * enable/disable make sure tx flag is always in same state as rx.
 	 */
-	if (features & NETIF_F_HW_VLAN_CTAG_RX)
-		features |= NETIF_F_HW_VLAN_CTAG_TX;
+	if (*features & NETIF_F_HW_VLAN_CTAG_RX)
+		*features |= NETIF_F_HW_VLAN_CTAG_TX;
 	else
-		features &= ~NETIF_F_HW_VLAN_CTAG_TX;
-
-	return features;
+		*features &= ~NETIF_F_HW_VLAN_CTAG_TX;
 }
 
 static int atl1e_set_features(struct net_device *netdev,
diff --git a/drivers/net/ethernet/atheros/atlx/atl2.c b/drivers/net/ethernet/atheros/atlx/atl2.c
index b69298ddb647..c4d303ce284c 100644
--- a/drivers/net/ethernet/atheros/atlx/atl2.c
+++ b/drivers/net/ethernet/atheros/atlx/atl2.c
@@ -371,19 +371,17 @@ static void atl2_restore_vlan(struct atl2_adapter *adapter)
 	atl2_vlan_mode(adapter->netdev, adapter->netdev->features);
 }
 
-static netdev_features_t atl2_fix_features(struct net_device *netdev,
-	netdev_features_t features)
+static void atl2_fix_features(struct net_device *netdev,
+			      netdev_features_t *features)
 {
 	/*
 	 * Since there is no support for separate rx/tx vlan accel
 	 * enable/disable make sure tx flag is always in same state as rx.
 	 */
-	if (features & NETIF_F_HW_VLAN_CTAG_RX)
-		features |= NETIF_F_HW_VLAN_CTAG_TX;
+	if (*features & NETIF_F_HW_VLAN_CTAG_RX)
+		*features |= NETIF_F_HW_VLAN_CTAG_TX;
 	else
-		features &= ~NETIF_F_HW_VLAN_CTAG_TX;
-
-	return features;
+		*features &= ~NETIF_F_HW_VLAN_CTAG_TX;
 }
 
 static int atl2_set_features(struct net_device *netdev,
diff --git a/drivers/net/ethernet/atheros/atlx/atlx.c b/drivers/net/ethernet/atheros/atlx/atlx.c
index 0941d07d0833..3b0dbc3a5896 100644
--- a/drivers/net/ethernet/atheros/atlx/atlx.c
+++ b/drivers/net/ethernet/atheros/atlx/atlx.c
@@ -237,19 +237,17 @@ static void atlx_restore_vlan(struct atlx_adapter *adapter)
 	atlx_vlan_mode(adapter->netdev, adapter->netdev->features);
 }
 
-static netdev_features_t atlx_fix_features(struct net_device *netdev,
-	netdev_features_t features)
+static void atlx_fix_features(struct net_device *netdev,
+			      netdev_features_t *features)
 {
 	/*
 	 * Since there is no support for separate rx/tx vlan accel
 	 * enable/disable make sure tx flag is always in same state as rx.
 	 */
-	if (features & NETIF_F_HW_VLAN_CTAG_RX)
-		features |= NETIF_F_HW_VLAN_CTAG_TX;
+	if (*features & NETIF_F_HW_VLAN_CTAG_RX)
+		*features |= NETIF_F_HW_VLAN_CTAG_TX;
 	else
-		features &= ~NETIF_F_HW_VLAN_CTAG_TX;
-
-	return features;
+		*features &= ~NETIF_F_HW_VLAN_CTAG_TX;
 }
 
 static int atlx_set_features(struct net_device *netdev,
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c
index b5d954cb409a..d74510306068 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c
@@ -4899,38 +4899,35 @@ int bnx2x_change_mtu(struct net_device *dev, int new_mtu)
 	return bnx2x_reload_if_running(dev);
 }
 
-netdev_features_t bnx2x_fix_features(struct net_device *dev,
-				     netdev_features_t features)
+void bnx2x_fix_features(struct net_device *dev, netdev_features_t *features)
 {
 	struct bnx2x *bp = netdev_priv(dev);
 
 	if (pci_num_vf(bp->pdev)) {
-		netdev_features_t changed = dev->features ^ features;
+		netdev_features_t changed = dev->features ^ *features;
 
 		/* Revert the requested changes in features if they
 		 * would require internal reload of PF in bnx2x_set_features().
 		 */
-		if (!(features & NETIF_F_RXCSUM) && !bp->disable_tpa) {
-			features &= ~NETIF_F_RXCSUM;
-			features |= dev->features & NETIF_F_RXCSUM;
+		if (!(*features & NETIF_F_RXCSUM) && !bp->disable_tpa) {
+			*features &= ~NETIF_F_RXCSUM;
+			*features |= dev->features & NETIF_F_RXCSUM;
 		}
 
 		if (changed & NETIF_F_LOOPBACK) {
-			features &= ~NETIF_F_LOOPBACK;
-			features |= dev->features & NETIF_F_LOOPBACK;
+			*features &= ~NETIF_F_LOOPBACK;
+			*features |= dev->features & NETIF_F_LOOPBACK;
 		}
 	}
 
 	/* TPA requires Rx CSUM offloading */
-	if (!(features & NETIF_F_RXCSUM))
-		features &= ~NETIF_F_LRO;
+	if (!(*features & NETIF_F_RXCSUM))
+		*features &= ~NETIF_F_LRO;
 
-	if (!(features & NETIF_F_GRO) || !bnx2x_mtu_allows_gro(dev->mtu))
-		features &= ~NETIF_F_GRO_HW;
-	if (features & NETIF_F_GRO_HW)
-		features &= ~NETIF_F_LRO;
-
-	return features;
+	if (!(*features & NETIF_F_GRO) || !bnx2x_mtu_allows_gro(dev->mtu))
+		*features &= ~NETIF_F_GRO_HW;
+	if (*features & NETIF_F_GRO_HW)
+		*features &= ~NETIF_F_LRO;
 }
 
 int bnx2x_set_features(struct net_device *dev, netdev_features_t features)
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h
index d8b1824c334d..4c66ef3e04bf 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h
@@ -606,8 +606,7 @@ int bnx2x_change_mtu(struct net_device *dev, int new_mtu);
 int bnx2x_fcoe_get_wwn(struct net_device *dev, u64 *wwn, int type);
 #endif
 
-netdev_features_t bnx2x_fix_features(struct net_device *dev,
-				     netdev_features_t features);
+void bnx2x_fix_features(struct net_device *dev, netdev_features_t *features);
 int bnx2x_set_features(struct net_device *dev, netdev_features_t features);
 
 /**
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index 4689d053aff8..7d9166876e95 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -10894,39 +10894,38 @@ static bool bnxt_rfs_capable(struct bnxt *bp)
 #endif
 }
 
-static netdev_features_t bnxt_fix_features(struct net_device *dev,
-					   netdev_features_t features)
+static void bnxt_fix_features(struct net_device *dev,
+			      netdev_features_t *features)
 {
 	struct bnxt *bp = netdev_priv(dev);
 	netdev_features_t vlan_features;
 
-	if ((features & NETIF_F_NTUPLE) && !bnxt_rfs_capable(bp))
-		features &= ~NETIF_F_NTUPLE;
+	if ((*features & NETIF_F_NTUPLE) && !bnxt_rfs_capable(bp))
+		*features &= ~NETIF_F_NTUPLE;
 
 	if (bp->flags & BNXT_FLAG_NO_AGG_RINGS)
-		features &= ~(NETIF_F_LRO | NETIF_F_GRO_HW);
+		*features &= ~(NETIF_F_LRO | NETIF_F_GRO_HW);
 
-	if (!(features & NETIF_F_GRO))
-		features &= ~NETIF_F_GRO_HW;
+	if (!(*features & NETIF_F_GRO))
+		*features &= ~NETIF_F_GRO_HW;
 
-	if (features & NETIF_F_GRO_HW)
-		features &= ~NETIF_F_LRO;
+	if (*features & NETIF_F_GRO_HW)
+		*features &= ~NETIF_F_LRO;
 
 	/* Both CTAG and STAG VLAN accelaration on the RX side have to be
 	 * turned on or off together.
 	 */
-	vlan_features = features & BNXT_HW_FEATURE_VLAN_ALL_RX;
+	vlan_features = *features & BNXT_HW_FEATURE_VLAN_ALL_RX;
 	if (vlan_features != BNXT_HW_FEATURE_VLAN_ALL_RX) {
 		if (dev->features & BNXT_HW_FEATURE_VLAN_ALL_RX)
-			features &= ~BNXT_HW_FEATURE_VLAN_ALL_RX;
+			*features &= ~BNXT_HW_FEATURE_VLAN_ALL_RX;
 		else if (vlan_features)
-			features |= BNXT_HW_FEATURE_VLAN_ALL_RX;
+			*features |= BNXT_HW_FEATURE_VLAN_ALL_RX;
 	}
 #ifdef CONFIG_BNXT_SRIOV
 	if (BNXT_VF(bp) && bp->vf.vlan)
-		features &= ~BNXT_HW_FEATURE_VLAN_ALL_RX;
+		*features &= ~BNXT_HW_FEATURE_VLAN_ALL_RX;
 #endif
-	return features;
 }
 
 static int bnxt_set_features(struct net_device *dev, netdev_features_t features)
diff --git a/drivers/net/ethernet/broadcom/tg3.c b/drivers/net/ethernet/broadcom/tg3.c
index 5e0e0e70d801..df0d6a35f093 100644
--- a/drivers/net/ethernet/broadcom/tg3.c
+++ b/drivers/net/ethernet/broadcom/tg3.c
@@ -8302,15 +8302,13 @@ static void tg3_set_loopback(struct net_device *dev, netdev_features_t features)
 	}
 }
 
-static netdev_features_t tg3_fix_features(struct net_device *dev,
-	netdev_features_t features)
+static void tg3_fix_features(struct net_device *dev,
+			     netdev_features_t *features)
 {
 	struct tg3 *tp = netdev_priv(dev);
 
 	if (dev->mtu > ETH_DATA_LEN && tg3_flag(tp, 5780_CLASS))
-		features &= ~NETIF_F_ALL_TSO;
-
-	return features;
+		*features &= ~NETIF_F_ALL_TSO;
 }
 
 static int tg3_set_features(struct net_device *dev, netdev_features_t features)
diff --git a/drivers/net/ethernet/cavium/liquidio/lio_main.c b/drivers/net/ethernet/cavium/liquidio/lio_main.c
index dafc79bd34f4..43c256ad2790 100644
--- a/drivers/net/ethernet/cavium/liquidio/lio_main.c
+++ b/drivers/net/ethernet/cavium/liquidio/lio_main.c
@@ -2718,38 +2718,36 @@ static const struct udp_tunnel_nic_info liquidio_udp_tunnels = {
  * @request: features requested
  * Return: updated features list
  */
-static netdev_features_t liquidio_fix_features(struct net_device *netdev,
-					       netdev_features_t request)
+static void liquidio_fix_features(struct net_device *netdev,
+				  netdev_features_t *request)
 {
 	struct lio *lio = netdev_priv(netdev);
 
-	if ((request & NETIF_F_RXCSUM) &&
+	if ((*request & NETIF_F_RXCSUM) &&
 	    !(lio->dev_capability & NETIF_F_RXCSUM))
-		request &= ~NETIF_F_RXCSUM;
+		*request &= ~NETIF_F_RXCSUM;
 
-	if ((request & NETIF_F_HW_CSUM) &&
+	if ((*request & NETIF_F_HW_CSUM) &&
 	    !(lio->dev_capability & NETIF_F_HW_CSUM))
-		request &= ~NETIF_F_HW_CSUM;
+		*request &= ~NETIF_F_HW_CSUM;
 
-	if ((request & NETIF_F_TSO) && !(lio->dev_capability & NETIF_F_TSO))
-		request &= ~NETIF_F_TSO;
+	if ((*request & NETIF_F_TSO) && !(lio->dev_capability & NETIF_F_TSO))
+		*request &= ~NETIF_F_TSO;
 
-	if ((request & NETIF_F_TSO6) && !(lio->dev_capability & NETIF_F_TSO6))
-		request &= ~NETIF_F_TSO6;
+	if ((*request & NETIF_F_TSO6) && !(lio->dev_capability & NETIF_F_TSO6))
+		*request &= ~NETIF_F_TSO6;
 
-	if ((request & NETIF_F_LRO) && !(lio->dev_capability & NETIF_F_LRO))
-		request &= ~NETIF_F_LRO;
+	if ((*request & NETIF_F_LRO) && !(lio->dev_capability & NETIF_F_LRO))
+		*request &= ~NETIF_F_LRO;
 
 	/*Disable LRO if RXCSUM is off */
-	if (!(request & NETIF_F_RXCSUM) && (netdev->features & NETIF_F_LRO) &&
+	if (!(*request & NETIF_F_RXCSUM) && (netdev->features & NETIF_F_LRO) &&
 	    (lio->dev_capability & NETIF_F_LRO))
-		request &= ~NETIF_F_LRO;
+		*request &= ~NETIF_F_LRO;
 
-	if ((request & NETIF_F_HW_VLAN_CTAG_FILTER) &&
+	if ((*request & NETIF_F_HW_VLAN_CTAG_FILTER) &&
 	    !(lio->dev_capability & NETIF_F_HW_VLAN_CTAG_FILTER))
-		request &= ~NETIF_F_HW_VLAN_CTAG_FILTER;
-
-	return request;
+		*request &= ~NETIF_F_HW_VLAN_CTAG_FILTER;
 }
 
 /**
diff --git a/drivers/net/ethernet/cavium/liquidio/lio_vf_main.c b/drivers/net/ethernet/cavium/liquidio/lio_vf_main.c
index f6396ac64006..1c4c039dff9b 100644
--- a/drivers/net/ethernet/cavium/liquidio/lio_vf_main.c
+++ b/drivers/net/ethernet/cavium/liquidio/lio_vf_main.c
@@ -1815,34 +1815,32 @@ static const struct udp_tunnel_nic_info liquidio_udp_tunnels = {
  * @param request features requested
  * @returns updated features list
  */
-static netdev_features_t liquidio_fix_features(struct net_device *netdev,
-					       netdev_features_t request)
+static void liquidio_fix_features(struct net_device *netdev,
+				  netdev_features_t *request)
 {
 	struct lio *lio = netdev_priv(netdev);
 
-	if ((request & NETIF_F_RXCSUM) &&
+	if ((*request & NETIF_F_RXCSUM) &&
 	    !(lio->dev_capability & NETIF_F_RXCSUM))
-		request &= ~NETIF_F_RXCSUM;
+		*request &= ~NETIF_F_RXCSUM;
 
-	if ((request & NETIF_F_HW_CSUM) &&
+	if ((*request & NETIF_F_HW_CSUM) &&
 	    !(lio->dev_capability & NETIF_F_HW_CSUM))
-		request &= ~NETIF_F_HW_CSUM;
+		*request &= ~NETIF_F_HW_CSUM;
 
-	if ((request & NETIF_F_TSO) && !(lio->dev_capability & NETIF_F_TSO))
-		request &= ~NETIF_F_TSO;
+	if ((*request & NETIF_F_TSO) && !(lio->dev_capability & NETIF_F_TSO))
+		*request &= ~NETIF_F_TSO;
 
-	if ((request & NETIF_F_TSO6) && !(lio->dev_capability & NETIF_F_TSO6))
-		request &= ~NETIF_F_TSO6;
+	if ((*request & NETIF_F_TSO6) && !(lio->dev_capability & NETIF_F_TSO6))
+		*request &= ~NETIF_F_TSO6;
 
-	if ((request & NETIF_F_LRO) && !(lio->dev_capability & NETIF_F_LRO))
-		request &= ~NETIF_F_LRO;
+	if ((*request & NETIF_F_LRO) && !(lio->dev_capability & NETIF_F_LRO))
+		*request &= ~NETIF_F_LRO;
 
 	/* Disable LRO if RXCSUM is off */
-	if (!(request & NETIF_F_RXCSUM) && (netdev->features & NETIF_F_LRO) &&
+	if (!(*request & NETIF_F_RXCSUM) && (netdev->features & NETIF_F_LRO) &&
 	    (lio->dev_capability & NETIF_F_LRO))
-		request &= ~NETIF_F_LRO;
-
-	return request;
+		*request &= ~NETIF_F_LRO;
 }
 
 /** \brief Net device set features
diff --git a/drivers/net/ethernet/cavium/thunder/nicvf_main.c b/drivers/net/ethernet/cavium/thunder/nicvf_main.c
index 2b87565781a0..781138a71458 100644
--- a/drivers/net/ethernet/cavium/thunder/nicvf_main.c
+++ b/drivers/net/ethernet/cavium/thunder/nicvf_main.c
@@ -1773,16 +1773,14 @@ static int nicvf_config_loopback(struct nicvf *nic,
 	return nicvf_send_msg_to_pf(nic, &mbx);
 }
 
-static netdev_features_t nicvf_fix_features(struct net_device *netdev,
-					    netdev_features_t features)
+static void nicvf_fix_features(struct net_device *netdev,
+			       netdev_features_t *features)
 {
 	struct nicvf *nic = netdev_priv(netdev);
 
-	if ((features & NETIF_F_LOOPBACK) &&
+	if ((*features & NETIF_F_LOOPBACK) &&
 	    netif_running(netdev) && !nic->loopback_supported)
-		features &= ~NETIF_F_LOOPBACK;
-
-	return features;
+		*features &= ~NETIF_F_LOOPBACK;
 }
 
 static int nicvf_set_features(struct net_device *netdev,
diff --git a/drivers/net/ethernet/chelsio/cxgb/cxgb2.c b/drivers/net/ethernet/chelsio/cxgb/cxgb2.c
index d246eee4b6d5..3fcd628fa449 100644
--- a/drivers/net/ethernet/chelsio/cxgb/cxgb2.c
+++ b/drivers/net/ethernet/chelsio/cxgb/cxgb2.c
@@ -858,19 +858,16 @@ static int t1_set_mac_addr(struct net_device *dev, void *p)
 	return 0;
 }
 
-static netdev_features_t t1_fix_features(struct net_device *dev,
-	netdev_features_t features)
+static void t1_fix_features(struct net_device *dev, netdev_features_t *features)
 {
 	/*
 	 * Since there is no support for separate rx/tx vlan accel
 	 * enable/disable make sure tx flag is always in same state as rx.
 	 */
-	if (features & NETIF_F_HW_VLAN_CTAG_RX)
-		features |= NETIF_F_HW_VLAN_CTAG_TX;
+	if (*features & NETIF_F_HW_VLAN_CTAG_RX)
+		*features |= NETIF_F_HW_VLAN_CTAG_TX;
 	else
-		features &= ~NETIF_F_HW_VLAN_CTAG_TX;
-
-	return features;
+		*features &= ~NETIF_F_HW_VLAN_CTAG_TX;
 }
 
 static int t1_set_features(struct net_device *dev, netdev_features_t features)
diff --git a/drivers/net/ethernet/chelsio/cxgb3/cxgb3_main.c b/drivers/net/ethernet/chelsio/cxgb3/cxgb3_main.c
index 38e47703f9ab..140b40e5c54c 100644
--- a/drivers/net/ethernet/chelsio/cxgb3/cxgb3_main.c
+++ b/drivers/net/ethernet/chelsio/cxgb3/cxgb3_main.c
@@ -2593,19 +2593,17 @@ static int cxgb_set_mac_addr(struct net_device *dev, void *p)
 	return 0;
 }
 
-static netdev_features_t cxgb_fix_features(struct net_device *dev,
-	netdev_features_t features)
+static void cxgb_fix_features(struct net_device *dev,
+			      netdev_features_t *features)
 {
 	/*
 	 * Since there is no support for separate rx/tx vlan accel
 	 * enable/disable make sure tx flag is always in same state as rx.
 	 */
-	if (features & NETIF_F_HW_VLAN_CTAG_RX)
-		features |= NETIF_F_HW_VLAN_CTAG_TX;
+	if (*features & NETIF_F_HW_VLAN_CTAG_RX)
+		*features |= NETIF_F_HW_VLAN_CTAG_TX;
 	else
-		features &= ~NETIF_F_HW_VLAN_CTAG_TX;
-
-	return features;
+		*features &= ~NETIF_F_HW_VLAN_CTAG_TX;
 }
 
 static int cxgb_set_features(struct net_device *dev, netdev_features_t features)
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c
index a654169b9dfc..238416724a7c 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c
@@ -3851,14 +3851,12 @@ static void cxgb_features_check(struct sk_buff *skb, struct net_device *dev,
 	*features &= ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
 }
 
-static netdev_features_t cxgb_fix_features(struct net_device *dev,
-					   netdev_features_t features)
+static void cxgb_fix_features(struct net_device *dev,
+			      netdev_features_t *features)
 {
 	/* Disable GRO, if RX_CSUM is disabled */
-	if (!(features & NETIF_F_RXCSUM))
-		features &= ~NETIF_F_GRO;
-
-	return features;
+	if (!(*features & NETIF_F_RXCSUM))
+		*features &= ~NETIF_F_GRO;
 }
 
 static const struct net_device_ops cxgb4_netdev_ops = {
diff --git a/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c b/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c
index 4920a80a0460..6d46d460a0a1 100644
--- a/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c
+++ b/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c
@@ -1173,19 +1173,17 @@ static int cxgb4vf_change_mtu(struct net_device *dev, int new_mtu)
 	return ret;
 }
 
-static netdev_features_t cxgb4vf_fix_features(struct net_device *dev,
-	netdev_features_t features)
+static void cxgb4vf_fix_features(struct net_device *dev,
+				 netdev_features_t *features)
 {
 	/*
 	 * Since there is no support for separate rx/tx vlan accel
 	 * enable/disable make sure tx flag is always in same state as rx.
 	 */
-	if (features & NETIF_F_HW_VLAN_CTAG_RX)
-		features |= NETIF_F_HW_VLAN_CTAG_TX;
+	if (*features & NETIF_F_HW_VLAN_CTAG_RX)
+		*features |= NETIF_F_HW_VLAN_CTAG_TX;
 	else
-		features &= ~NETIF_F_HW_VLAN_CTAG_TX;
-
-	return features;
+		*features &= ~NETIF_F_HW_VLAN_CTAG_TX;
 }
 
 static int cxgb4vf_set_features(struct net_device *dev,
diff --git a/drivers/net/ethernet/cortina/gemini.c b/drivers/net/ethernet/cortina/gemini.c
index 6e745ca4c433..44397d1f44e4 100644
--- a/drivers/net/ethernet/cortina/gemini.c
+++ b/drivers/net/ethernet/cortina/gemini.c
@@ -1976,13 +1976,11 @@ static int gmac_change_mtu(struct net_device *netdev, int new_mtu)
 	return 0;
 }
 
-static netdev_features_t gmac_fix_features(struct net_device *netdev,
-					   netdev_features_t features)
+static void gmac_fix_features(struct net_device *netdev,
+			      netdev_features_t *features)
 {
 	if (netdev->mtu + ETH_HLEN + VLAN_HLEN > MTU_SIZE_BIT_MASK)
-		features &= ~GMAC_OFFLOAD_FEATURES;
-
-	return features;
+		*features &= ~GMAC_OFFLOAD_FEATURES;
 }
 
 static int gmac_set_features(struct net_device *netdev,
diff --git a/drivers/net/ethernet/hisilicon/hns/hns_enet.c b/drivers/net/ethernet/hisilicon/hns/hns_enet.c
index 343c605c4be8..1e08c18c813f 100644
--- a/drivers/net/ethernet/hisilicon/hns/hns_enet.c
+++ b/drivers/net/ethernet/hisilicon/hns/hns_enet.c
@@ -1793,20 +1793,19 @@ static int hns_nic_set_features(struct net_device *netdev,
 	return 0;
 }
 
-static netdev_features_t hns_nic_fix_features(
-		struct net_device *netdev, netdev_features_t features)
+static void hns_nic_fix_features(struct net_device *netdev,
+				 netdev_features_t *features)
 {
 	struct hns_nic_priv *priv = netdev_priv(netdev);
 
 	switch (priv->enet_ver) {
 	case AE_VERSION_1:
-		features &= ~(NETIF_F_TSO | NETIF_F_TSO6 |
+		*features &= ~(NETIF_F_TSO | NETIF_F_TSO6 |
 				NETIF_F_HW_VLAN_CTAG_FILTER);
 		break;
 	default:
 		break;
 	}
-	return features;
 }
 
 static int hns_nic_uc_sync(struct net_device *netdev, const unsigned char *addr)
diff --git a/drivers/net/ethernet/huawei/hinic/hinic_main.c b/drivers/net/ethernet/huawei/hinic/hinic_main.c
index 9965e8d5d0a9..cce66faa477c 100644
--- a/drivers/net/ethernet/huawei/hinic/hinic_main.c
+++ b/drivers/net/ethernet/huawei/hinic/hinic_main.c
@@ -888,18 +888,16 @@ static int hinic_set_features(struct net_device *netdev,
 			    features, false);
 }
 
-static netdev_features_t hinic_fix_features(struct net_device *netdev,
-					    netdev_features_t features)
+static void hinic_fix_features(struct net_device *netdev,
+			       netdev_features_t *features)
 {
 	struct hinic_dev *nic_dev = netdev_priv(netdev);
 
 	/* If Rx checksum is disabled, then LRO should also be disabled */
-	if (!(features & NETIF_F_RXCSUM)) {
+	if (!(*features & NETIF_F_RXCSUM)) {
 		netif_info(nic_dev, drv, netdev, "disabling LRO as RXCSUM is off\n");
-		features &= ~NETIF_F_LRO;
+		*features &= ~NETIF_F_LRO;
 	}
-
-	return features;
 }
 
 static const struct net_device_ops hinic_netdev_ops = {
diff --git a/drivers/net/ethernet/ibm/ibmveth.c b/drivers/net/ethernet/ibm/ibmveth.c
index 3aedb680adb8..7884d17d666f 100644
--- a/drivers/net/ethernet/ibm/ibmveth.c
+++ b/drivers/net/ethernet/ibm/ibmveth.c
@@ -742,8 +742,8 @@ static void netdev_get_drvinfo(struct net_device *dev,
 	strlcpy(info->version, ibmveth_driver_version, sizeof(info->version));
 }
 
-static netdev_features_t ibmveth_fix_features(struct net_device *dev,
-	netdev_features_t features)
+static void ibmveth_fix_features(struct net_device *dev,
+				 netdev_features_t *features)
 {
 	/*
 	 * Since the ibmveth firmware interface does not have the
@@ -754,10 +754,8 @@ static netdev_features_t ibmveth_fix_features(struct net_device *dev,
 	 * checksummed.
 	 */
 
-	if (!(features & NETIF_F_RXCSUM))
-		features &= ~NETIF_F_CSUM_MASK;
-
-	return features;
+	if (!(*features & NETIF_F_RXCSUM))
+		*features &= ~NETIF_F_CSUM_MASK;
 }
 
 static int ibmveth_set_csum_offload(struct net_device *dev, u32 data)
diff --git a/drivers/net/ethernet/intel/e1000/e1000_main.c b/drivers/net/ethernet/intel/e1000/e1000_main.c
index bed4f040face..e333ca1e7395 100644
--- a/drivers/net/ethernet/intel/e1000/e1000_main.c
+++ b/drivers/net/ethernet/intel/e1000/e1000_main.c
@@ -787,18 +787,16 @@ static int e1000_is_need_ioport(struct pci_dev *pdev)
 	}
 }
 
-static netdev_features_t e1000_fix_features(struct net_device *netdev,
-	netdev_features_t features)
+static void e1000_fix_features(struct net_device *netdev,
+			       netdev_features_t *features)
 {
 	/* Since there is no support for separate Rx/Tx vlan accel
 	 * enable/disable make sure Tx flag is always in same state as Rx.
 	 */
-	if (features & NETIF_F_HW_VLAN_CTAG_RX)
-		features |= NETIF_F_HW_VLAN_CTAG_TX;
+	if (*features & NETIF_F_HW_VLAN_CTAG_RX)
+		*features |= NETIF_F_HW_VLAN_CTAG_TX;
 	else
-		features &= ~NETIF_F_HW_VLAN_CTAG_TX;
-
-	return features;
+		*features &= ~NETIF_F_HW_VLAN_CTAG_TX;
 }
 
 static int e1000_set_features(struct net_device *netdev,
diff --git a/drivers/net/ethernet/intel/e1000e/netdev.c b/drivers/net/ethernet/intel/e1000e/netdev.c
index 900b3ab998bd..5dd183e0cb0f 100644
--- a/drivers/net/ethernet/intel/e1000e/netdev.c
+++ b/drivers/net/ethernet/intel/e1000e/netdev.c
@@ -7287,25 +7287,23 @@ static void e1000_eeprom_checks(struct e1000_adapter *adapter)
 	}
 }
 
-static netdev_features_t e1000_fix_features(struct net_device *netdev,
-					    netdev_features_t features)
+static void e1000_fix_features(struct net_device *netdev,
+			       netdev_features_t *features)
 {
 	struct e1000_adapter *adapter = netdev_priv(netdev);
 	struct e1000_hw *hw = &adapter->hw;
 
 	/* Jumbo frame workaround on 82579 and newer requires CRC be stripped */
 	if ((hw->mac.type >= e1000_pch2lan) && (netdev->mtu > ETH_DATA_LEN))
-		features &= ~NETIF_F_RXFCS;
+		*features &= ~NETIF_F_RXFCS;
 
 	/* Since there is no support for separate Rx/Tx vlan accel
 	 * enable/disable make sure Tx flag is always in same state as Rx.
 	 */
-	if (features & NETIF_F_HW_VLAN_CTAG_RX)
-		features |= NETIF_F_HW_VLAN_CTAG_TX;
+	if (*features & NETIF_F_HW_VLAN_CTAG_RX)
+		*features |= NETIF_F_HW_VLAN_CTAG_TX;
 	else
-		features &= ~NETIF_F_HW_VLAN_CTAG_TX;
-
-	return features;
+		*features &= ~NETIF_F_HW_VLAN_CTAG_TX;
 }
 
 static int e1000_set_features(struct net_device *netdev,
diff --git a/drivers/net/ethernet/intel/iavf/iavf_main.c b/drivers/net/ethernet/intel/iavf/iavf_main.c
index b066f1864b3f..9a086211af4a 100644
--- a/drivers/net/ethernet/intel/iavf/iavf_main.c
+++ b/drivers/net/ethernet/intel/iavf/iavf_main.c
@@ -3437,17 +3437,15 @@ static void iavf_features_check(struct sk_buff *skb, struct net_device *dev,
  *
  * Returns fixed-up features bits
  **/
-static netdev_features_t iavf_fix_features(struct net_device *netdev,
-					   netdev_features_t features)
+static void iavf_fix_features(struct net_device *netdev,
+			      netdev_features_t *features)
 {
 	struct iavf_adapter *adapter = netdev_priv(netdev);
 
 	if (!(adapter->vf_res->vf_cap_flags & VIRTCHNL_VF_OFFLOAD_VLAN))
-		features &= ~(NETIF_F_HW_VLAN_CTAG_TX |
+		*features &= ~(NETIF_F_HW_VLAN_CTAG_TX |
 			      NETIF_F_HW_VLAN_CTAG_RX |
 			      NETIF_F_HW_VLAN_CTAG_FILTER);
-
-	return features;
 }
 
 static const struct net_device_ops iavf_netdev_ops = {
diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c
index 82b59adf8034..2e4a53c76c60 100644
--- a/drivers/net/ethernet/intel/igb/igb_main.c
+++ b/drivers/net/ethernet/intel/igb/igb_main.c
@@ -2433,18 +2433,16 @@ void igb_reset(struct igb_adapter *adapter)
 	igb_get_phy_info(hw);
 }
 
-static netdev_features_t igb_fix_features(struct net_device *netdev,
-	netdev_features_t features)
+static void igb_fix_features(struct net_device *netdev,
+			     netdev_features_t *features)
 {
 	/* Since there is no support for separate Rx/Tx vlan accel
 	 * enable/disable make sure Tx flag is always in same state as Rx.
 	 */
-	if (features & NETIF_F_HW_VLAN_CTAG_RX)
-		features |= NETIF_F_HW_VLAN_CTAG_TX;
+	if (*features & NETIF_F_HW_VLAN_CTAG_RX)
+		*features |= NETIF_F_HW_VLAN_CTAG_TX;
 	else
-		features &= ~NETIF_F_HW_VLAN_CTAG_TX;
-
-	return features;
+		*features &= ~NETIF_F_HW_VLAN_CTAG_TX;
 }
 
 static int igb_set_features(struct net_device *netdev,
diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c
index fdb3ed051456..13b89a742ebc 100644
--- a/drivers/net/ethernet/intel/igc/igc_main.c
+++ b/drivers/net/ethernet/intel/igc/igc_main.c
@@ -4915,18 +4915,16 @@ static void igc_get_stats64(struct net_device *netdev,
 	spin_unlock(&adapter->stats64_lock);
 }
 
-static netdev_features_t igc_fix_features(struct net_device *netdev,
-					  netdev_features_t features)
+static void igc_fix_features(struct net_device *netdev,
+			     netdev_features_t *features)
 {
 	/* Since there is no support for separate Rx/Tx vlan accel
 	 * enable/disable make sure Tx flag is always in same state as Rx.
 	 */
-	if (features & NETIF_F_HW_VLAN_CTAG_RX)
-		features |= NETIF_F_HW_VLAN_CTAG_TX;
+	if (*features & NETIF_F_HW_VLAN_CTAG_RX)
+		*features |= NETIF_F_HW_VLAN_CTAG_TX;
 	else
-		features &= ~NETIF_F_HW_VLAN_CTAG_TX;
-
-	return features;
+		*features &= ~NETIF_F_HW_VLAN_CTAG_TX;
 }
 
 static int igc_set_features(struct net_device *netdev,
diff --git a/drivers/net/ethernet/intel/ixgb/ixgb_main.c b/drivers/net/ethernet/intel/ixgb/ixgb_main.c
index 1588376d4c67..34c1bfc75b7b 100644
--- a/drivers/net/ethernet/intel/ixgb/ixgb_main.c
+++ b/drivers/net/ethernet/intel/ixgb/ixgb_main.c
@@ -293,17 +293,15 @@ ixgb_reset(struct ixgb_adapter *adapter)
 	}
 }
 
-static netdev_features_t
-ixgb_fix_features(struct net_device *netdev, netdev_features_t features)
+static void ixgb_fix_features(struct net_device *netdev,
+			      netdev_features_t *features)
 {
 	/*
 	 * Tx VLAN insertion does not work per HW design when Rx stripping is
 	 * disabled.
 	 */
-	if (!(features & NETIF_F_HW_VLAN_CTAG_RX))
-		features &= ~NETIF_F_HW_VLAN_CTAG_TX;
-
-	return features;
+	if (!(*features & NETIF_F_HW_VLAN_CTAG_RX))
+		*features &= ~NETIF_F_HW_VLAN_CTAG_TX;
 }
 
 static int
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
index e39d3983a455..c43c99a44914 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
@@ -9682,25 +9682,23 @@ void ixgbe_do_reset(struct net_device *netdev)
 		ixgbe_reset(adapter);
 }
 
-static netdev_features_t ixgbe_fix_features(struct net_device *netdev,
-					    netdev_features_t features)
+static void ixgbe_fix_features(struct net_device *netdev,
+			       netdev_features_t *features)
 {
 	struct ixgbe_adapter *adapter = netdev_priv(netdev);
 
 	/* If Rx checksum is disabled, then RSC/LRO should also be disabled */
-	if (!(features & NETIF_F_RXCSUM))
-		features &= ~NETIF_F_LRO;
+	if (!(*features & NETIF_F_RXCSUM))
+		*features &= ~NETIF_F_LRO;
 
 	/* Turn off LRO if not RSC capable */
 	if (!(adapter->flags2 & IXGBE_FLAG2_RSC_CAPABLE))
-		features &= ~NETIF_F_LRO;
+		*features &= ~NETIF_F_LRO;
 
-	if (adapter->xdp_prog && (features & NETIF_F_LRO)) {
+	if (adapter->xdp_prog && (*features & NETIF_F_LRO)) {
 		e_dev_err("LRO is not supported with XDP\n");
-		features &= ~NETIF_F_LRO;
+		*features &= ~NETIF_F_LRO;
 	}
-
-	return features;
 }
 
 static void ixgbe_reset_l2fw_offload(struct ixgbe_adapter *adapter)
diff --git a/drivers/net/ethernet/jme.c b/drivers/net/ethernet/jme.c
index 1bdc4f23e1e5..11749bd7276d 100644
--- a/drivers/net/ethernet/jme.c
+++ b/drivers/net/ethernet/jme.c
@@ -2659,12 +2659,11 @@ jme_set_msglevel(struct net_device *netdev, u32 value)
 	jme->msg_enable = value;
 }
 
-static netdev_features_t
-jme_fix_features(struct net_device *netdev, netdev_features_t features)
+static void jme_fix_features(struct net_device *netdev,
+			     netdev_features_t *features)
 {
 	if (netdev->mtu > 1900)
-		features &= ~(NETIF_F_ALL_TSO | NETIF_F_CSUM_MASK);
-	return features;
+		*features &= ~(NETIF_F_ALL_TSO | NETIF_F_CSUM_MASK);
 }
 
 static int
diff --git a/drivers/net/ethernet/marvell/mvneta.c b/drivers/net/ethernet/marvell/mvneta.c
index 9d460a270601..a4093977cd2b 100644
--- a/drivers/net/ethernet/marvell/mvneta.c
+++ b/drivers/net/ethernet/marvell/mvneta.c
@@ -3770,19 +3770,17 @@ static int mvneta_change_mtu(struct net_device *dev, int mtu)
 	return 0;
 }
 
-static netdev_features_t mvneta_fix_features(struct net_device *dev,
-					     netdev_features_t features)
+static void mvneta_fix_features(struct net_device *dev,
+				netdev_features_t *features)
 {
 	struct mvneta_port *pp = netdev_priv(dev);
 
 	if (pp->tx_csum_limit && dev->mtu > pp->tx_csum_limit) {
-		features &= ~(NETIF_F_IP_CSUM | NETIF_F_TSO);
+		*features &= ~(NETIF_F_IP_CSUM | NETIF_F_TSO);
 		netdev_info(dev,
 			    "Disable IP checksum for MTU greater than %dB\n",
 			    pp->tx_csum_limit);
 	}
-
-	return features;
 }
 
 /* Get mac address */
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
index ab07964dbf0a..cf4513ec02e6 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
@@ -1813,15 +1813,13 @@ static netdev_tx_t otx2_xmit(struct sk_buff *skb, struct net_device *netdev)
 	return NETDEV_TX_OK;
 }
 
-static netdev_features_t otx2_fix_features(struct net_device *dev,
-					   netdev_features_t features)
+static void otx2_fix_features(struct net_device *dev,
+			      netdev_features_t *features)
 {
-	if (features & NETIF_F_HW_VLAN_CTAG_RX)
-		features |= NETIF_F_HW_VLAN_STAG_RX;
+	if (*features & NETIF_F_HW_VLAN_CTAG_RX)
+		*features |= NETIF_F_HW_VLAN_STAG_RX;
 	else
-		features &= ~NETIF_F_HW_VLAN_STAG_RX;
-
-	return features;
+		*features &= ~NETIF_F_HW_VLAN_STAG_RX;
 }
 
 static void otx2_set_rx_mode(struct net_device *netdev)
diff --git a/drivers/net/ethernet/marvell/sky2.c b/drivers/net/ethernet/marvell/sky2.c
index 3cb9c1271328..8b1af582c250 100644
--- a/drivers/net/ethernet/marvell/sky2.c
+++ b/drivers/net/ethernet/marvell/sky2.c
@@ -4358,8 +4358,8 @@ static int sky2_set_eeprom(struct net_device *dev, struct ethtool_eeprom *eeprom
 	return sky2_vpd_write(sky2->hw, cap, data, eeprom->offset, eeprom->len);
 }
 
-static netdev_features_t sky2_fix_features(struct net_device *dev,
-	netdev_features_t features)
+static void sky2_fix_features(struct net_device *dev,
+			      netdev_features_t *features)
 {
 	const struct sky2_port *sky2 = netdev_priv(dev);
 	const struct sky2_hw *hw = sky2->hw;
@@ -4369,18 +4369,16 @@ static netdev_features_t sky2_fix_features(struct net_device *dev,
 	 */
 	if (dev->mtu > ETH_DATA_LEN && hw->chip_id == CHIP_ID_YUKON_EC_U) {
 		netdev_info(dev, "checksum offload not possible with jumbo frames\n");
-		features &= ~(NETIF_F_TSO | NETIF_F_SG | NETIF_F_CSUM_MASK);
+		*features &= ~(NETIF_F_TSO | NETIF_F_SG | NETIF_F_CSUM_MASK);
 	}
 
 	/* Some hardware requires receive checksum for RSS to work. */
-	if ( (features & NETIF_F_RXHASH) &&
-	     !(features & NETIF_F_RXCSUM) &&
-	     (sky2->hw->flags & SKY2_HW_RSS_CHKSUM)) {
+	if ((*features & NETIF_F_RXHASH) &&
+	    !(*features & NETIF_F_RXCSUM) &&
+	    (sky2->hw->flags & SKY2_HW_RSS_CHKSUM)) {
 		netdev_info(dev, "receive hashing forces receive checksum\n");
-		features |= NETIF_F_RXCSUM;
+		*features |= NETIF_F_RXCSUM;
 	}
-
-	return features;
 }
 
 static int sky2_set_features(struct net_device *dev, netdev_features_t features)
diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c
index 398c23cec815..6a6ef1c29657 100644
--- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c
+++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c
@@ -2021,21 +2021,19 @@ static int mtk_hwlro_get_fdir_all(struct net_device *dev,
 	return 0;
 }
 
-static netdev_features_t mtk_fix_features(struct net_device *dev,
-					  netdev_features_t features)
+static void mtk_fix_features(struct net_device *dev,
+			     netdev_features_t *features)
 {
-	if (!(features & NETIF_F_LRO)) {
+	if (!(*features & NETIF_F_LRO)) {
 		struct mtk_mac *mac = netdev_priv(dev);
 		int ip_cnt = mtk_hwlro_get_ip_cnt(mac);
 
 		if (ip_cnt) {
 			netdev_info(dev, "RX flow is programmed, LRO should keep on\n");
 
-			features |= NETIF_F_LRO;
+			*features |= NETIF_F_LRO;
 		}
 	}
-
-	return features;
 }
 
 static int mtk_set_features(struct net_device *dev, netdev_features_t features)
diff --git a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
index 354c63aa726d..5d95e878bad2 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
@@ -2493,8 +2493,8 @@ static int mlx4_en_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 	}
 }
 
-static netdev_features_t mlx4_en_fix_features(struct net_device *netdev,
-					      netdev_features_t features)
+static void mlx4_en_fix_features(struct net_device *netdev,
+				 netdev_features_t *features)
 {
 	struct mlx4_en_priv *en_priv = netdev_priv(netdev);
 	struct mlx4_en_dev *mdev = en_priv->mdev;
@@ -2503,13 +2503,11 @@ static netdev_features_t mlx4_en_fix_features(struct net_device *netdev,
 	 * enable/disable make sure S-TAG flag is always in same state as
 	 * C-TAG.
 	 */
-	if (features & NETIF_F_HW_VLAN_CTAG_RX &&
+	if (*features & NETIF_F_HW_VLAN_CTAG_RX &&
 	    !(mdev->dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_SKIP_OUTER_VLAN))
-		features |= NETIF_F_HW_VLAN_STAG_RX;
+		*features |= NETIF_F_HW_VLAN_STAG_RX;
 	else
-		features &= ~NETIF_F_HW_VLAN_STAG_RX;
-
-	return features;
+		*features &= ~NETIF_F_HW_VLAN_STAG_RX;
 }
 
 static int mlx4_en_set_features(struct net_device *netdev,
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index 6fdc2a793c1f..4092210fb079 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -3397,8 +3397,8 @@ static netdev_features_t mlx5e_fix_uplink_rep_features(struct net_device *netdev
 	return features;
 }
 
-static netdev_features_t mlx5e_fix_features(struct net_device *netdev,
-					    netdev_features_t features)
+static void mlx5e_fix_features(struct net_device *netdev,
+			       netdev_features_t *features)
 {
 	struct mlx5e_priv *priv = netdev_priv(netdev);
 	struct mlx5e_params *params;
@@ -3410,20 +3410,20 @@ static netdev_features_t mlx5e_fix_features(struct net_device *netdev,
 		/* HW strips the outer C-tag header, this is a problem
 		 * for S-tag traffic.
 		 */
-		features &= ~NETIF_F_HW_VLAN_CTAG_RX;
+		*features &= ~NETIF_F_HW_VLAN_CTAG_RX;
 		if (!params->vlan_strip_disable)
 			netdev_warn(netdev, "Dropping C-tag vlan stripping offload due to S-tag vlan\n");
 	}
 
 	if (!MLX5E_GET_PFLAG(params, MLX5E_PFLAG_RX_STRIDING_RQ)) {
-		if (features & NETIF_F_LRO) {
+		if (*features & NETIF_F_LRO) {
 			netdev_warn(netdev, "Disabling LRO, not supported in legacy RQ\n");
-			features &= ~NETIF_F_LRO;
+			*features &= ~NETIF_F_LRO;
 		}
 	}
 
 	if (MLX5E_GET_PFLAG(params, MLX5E_PFLAG_RX_CQE_COMPRESS)) {
-		features &= ~NETIF_F_RXHASH;
+		*features &= ~NETIF_F_RXHASH;
 		if (netdev->features & NETIF_F_RXHASH)
 			netdev_warn(netdev, "Disabling rxhash, not supported when CQE compress is active\n");
 	}
@@ -3432,8 +3432,6 @@ static netdev_features_t mlx5e_fix_features(struct net_device *netdev,
 		features = mlx5e_fix_uplink_rep_features(netdev, features);
 
 	mutex_unlock(&priv->state_lock);
-
-	return features;
 }
 
 static bool mlx5e_xsk_validate_mtu(struct net_device *netdev,
diff --git a/drivers/net/ethernet/neterion/vxge/vxge-main.c b/drivers/net/ethernet/neterion/vxge/vxge-main.c
index df4a3f3da83a..5f720e97a558 100644
--- a/drivers/net/ethernet/neterion/vxge/vxge-main.c
+++ b/drivers/net/ethernet/neterion/vxge/vxge-main.c
@@ -2639,19 +2639,17 @@ static void vxge_poll_vp_lockup(struct timer_list *t)
 	mod_timer(&vdev->vp_lockup_timer, jiffies + HZ / 1000);
 }
 
-static netdev_features_t vxge_fix_features(struct net_device *dev,
-	netdev_features_t features)
+static void vxge_fix_features(struct net_device *dev,
+			      netdev_features_t *features)
 {
-	netdev_features_t changed = dev->features ^ features;
+	netdev_features_t changed = dev->features ^ *features;
 
 	/* Enabling RTH requires some of the logic in vxge_device_register and a
 	 * vpath reset.  Due to these restrictions, only allow modification
 	 * while the interface is down.
 	 */
 	if ((changed & NETIF_F_RXHASH) && netif_running(dev))
-		features ^= NETIF_F_RXHASH;
-
-	return features;
+		*features ^= NETIF_F_RXHASH;
 }
 
 static int vxge_set_features(struct net_device *dev, netdev_features_t features)
diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_repr.c b/drivers/net/ethernet/netronome/nfp/nfp_net_repr.c
index f04b79f04a9d..fcb2e30e8ac7 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_net_repr.c
+++ b/drivers/net/ethernet/netronome/nfp/nfp_net_repr.c
@@ -232,11 +232,11 @@ static int nfp_repr_open(struct net_device *netdev)
 	return err;
 }
 
-static netdev_features_t
-nfp_repr_fix_features(struct net_device *netdev, netdev_features_t features)
+static void nfp_repr_fix_features(struct net_device *netdev,
+				  netdev_features_t *features)
 {
 	struct nfp_repr *repr = netdev_priv(netdev);
-	netdev_features_t old_features = features;
+	netdev_features_t old_features = *features;
 	netdev_features_t lower_features;
 	struct net_device *lower_dev;
 
@@ -246,11 +246,9 @@ nfp_repr_fix_features(struct net_device *netdev, netdev_features_t features)
 	if (lower_features & (NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM))
 		lower_features |= NETIF_F_HW_CSUM;
 
-	netdev_intersect_features(&features, features, lower_features);
-	features |= old_features & (NETIF_F_SOFT_FEATURES | NETIF_F_HW_TC);
-	features |= NETIF_F_LLTX;
-
-	return features;
+	netdev_intersect_features(features, *features, lower_features);
+	*features |= old_features & (NETIF_F_SOFT_FEATURES | NETIF_F_HW_TC);
+	*features |= NETIF_F_LLTX;
 }
 
 const struct net_device_ops nfp_repr_netdev_ops = {
diff --git a/drivers/net/ethernet/nvidia/forcedeth.c b/drivers/net/ethernet/nvidia/forcedeth.c
index ef3fb4cc90af..e1f16988cb75 100644
--- a/drivers/net/ethernet/nvidia/forcedeth.c
+++ b/drivers/net/ethernet/nvidia/forcedeth.c
@@ -4920,14 +4920,12 @@ static int nv_set_loopback(struct net_device *dev, netdev_features_t features)
 	return retval;
 }
 
-static netdev_features_t nv_fix_features(struct net_device *dev,
-	netdev_features_t features)
+static void nv_fix_features(struct net_device *dev,
+			    netdev_features_t *features)
 {
 	/* vlan is dependent on rx checksum offload */
-	if (features & (NETIF_F_HW_VLAN_CTAG_TX|NETIF_F_HW_VLAN_CTAG_RX))
-		features |= NETIF_F_RXCSUM;
-
-	return features;
+	if (*features & (NETIF_F_HW_VLAN_CTAG_TX | NETIF_F_HW_VLAN_CTAG_RX))
+		*features |= NETIF_F_RXCSUM;
 }
 
 static void nv_vlan_mode(struct net_device *dev, netdev_features_t features)
diff --git a/drivers/net/ethernet/qlogic/netxen/netxen_nic_main.c b/drivers/net/ethernet/qlogic/netxen/netxen_nic_main.c
index 344ea1143454..251668839926 100644
--- a/drivers/net/ethernet/qlogic/netxen/netxen_nic_main.c
+++ b/drivers/net/ethernet/qlogic/netxen/netxen_nic_main.c
@@ -517,16 +517,14 @@ static void netxen_set_multicast_list(struct net_device *dev)
 	adapter->set_multi(dev);
 }
 
-static netdev_features_t netxen_fix_features(struct net_device *dev,
-	netdev_features_t features)
+static void netxen_fix_features(struct net_device *dev,
+				netdev_features_t *features)
 {
-	if (!(features & NETIF_F_RXCSUM)) {
+	if (!(*features & NETIF_F_RXCSUM)) {
 		netdev_info(dev, "disabling LRO as RXCSUM is off\n");
 
-		features &= ~NETIF_F_LRO;
+		*features &= ~NETIF_F_LRO;
 	}
-
-	return features;
 }
 
 static int netxen_set_features(struct net_device *dev,
diff --git a/drivers/net/ethernet/qlogic/qede/qede.h b/drivers/net/ethernet/qlogic/qede/qede.h
index c1f26a2e374d..ada71452d454 100644
--- a/drivers/net/ethernet/qlogic/qede/qede.h
+++ b/drivers/net/ethernet/qlogic/qede/qede.h
@@ -545,8 +545,7 @@ int qede_vlan_rx_kill_vid(struct net_device *dev, __be16 proto, u16 vid);
 void qede_vlan_mark_nonconfigured(struct qede_dev *edev);
 int qede_configure_vlan_filters(struct qede_dev *edev);
 
-netdev_features_t qede_fix_features(struct net_device *dev,
-				    netdev_features_t features);
+void qede_fix_features(struct net_device *dev, netdev_features_t *features);
 int qede_set_features(struct net_device *dev, netdev_features_t features);
 void qede_set_rx_mode(struct net_device *ndev);
 void qede_config_rx_mode(struct net_device *ndev);
diff --git a/drivers/net/ethernet/qlogic/qede/qede_filter.c b/drivers/net/ethernet/qlogic/qede/qede_filter.c
index f99b085b56a5..ea89a3afa206 100644
--- a/drivers/net/ethernet/qlogic/qede/qede_filter.c
+++ b/drivers/net/ethernet/qlogic/qede/qede_filter.c
@@ -910,16 +910,13 @@ static void qede_set_features_reload(struct qede_dev *edev,
 	edev->ndev->features = args->u.features;
 }
 
-netdev_features_t qede_fix_features(struct net_device *dev,
-				    netdev_features_t features)
+void qede_fix_features(struct net_device *dev, netdev_features_t *features)
 {
 	struct qede_dev *edev = netdev_priv(dev);
 
 	if (edev->xdp_prog || edev->ndev->mtu > PAGE_SIZE ||
-	    !(features & NETIF_F_GRO))
-		features &= ~NETIF_F_GRO_HW;
-
-	return features;
+	    !(*features & NETIF_F_GRO))
+		*features &= ~NETIF_F_GRO_HW;
 }
 
 int qede_set_features(struct net_device *dev, netdev_features_t features)
diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic.h b/drivers/net/ethernet/qlogic/qlcnic/qlcnic.h
index be7abee160e7..7fbf895becdd 100644
--- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic.h
+++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic.h
@@ -1622,8 +1622,7 @@ int qlcnic_82xx_read_phys_port_id(struct qlcnic_adapter *);
 int qlcnic_fw_cmd_set_mtu(struct qlcnic_adapter *adapter, int mtu);
 int qlcnic_fw_cmd_set_drv_version(struct qlcnic_adapter *, u32);
 int qlcnic_change_mtu(struct net_device *netdev, int new_mtu);
-netdev_features_t qlcnic_fix_features(struct net_device *netdev,
-	netdev_features_t features);
+void qlcnic_fix_features(struct net_device *netdev, netdev_features_t *features);
 int qlcnic_set_features(struct net_device *netdev, netdev_features_t features);
 int qlcnic_config_bridged_mode(struct qlcnic_adapter *adapter, u32 enable);
 void qlcnic_update_cmd_producer(struct qlcnic_host_tx_ring *);
diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_hw.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_hw.c
index 4b8bc46f55c2..e6ed7f8413b4 100644
--- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_hw.c
+++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_hw.c
@@ -1053,8 +1053,7 @@ static netdev_features_t qlcnic_process_flags(struct qlcnic_adapter *adapter,
 	return features;
 }
 
-netdev_features_t qlcnic_fix_features(struct net_device *netdev,
-	netdev_features_t features)
+void qlcnic_fix_features(struct net_device *netdev, netdev_features_t *features)
 {
 	struct qlcnic_adapter *adapter = netdev_priv(netdev);
 	netdev_features_t changed;
@@ -1062,10 +1061,10 @@ netdev_features_t qlcnic_fix_features(struct net_device *netdev,
 	if (qlcnic_82xx_check(adapter) &&
 	    (adapter->flags & QLCNIC_ESWITCH_ENABLED)) {
 		if (adapter->flags & QLCNIC_APP_CHANGED_FLAGS) {
-			features = qlcnic_process_flags(adapter, features);
+			*features = qlcnic_process_flags(adapter, *features);
 		} else {
-			changed = features ^ netdev->features;
-			features ^= changed & (NETIF_F_RXCSUM |
+			changed = *features ^ netdev->features;
+			*features ^= changed & (NETIF_F_RXCSUM |
 					       NETIF_F_IP_CSUM |
 					       NETIF_F_IPV6_CSUM |
 					       NETIF_F_TSO |
@@ -1073,10 +1072,8 @@ netdev_features_t qlcnic_fix_features(struct net_device *netdev,
 		}
 	}
 
-	if (!(features & NETIF_F_RXCSUM))
-		features &= ~NETIF_F_LRO;
-
-	return features;
+	if (!(*features & NETIF_F_RXCSUM))
+		*features &= ~NETIF_F_LRO;
 }
 
 
diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c
index 3d753ddd1a89..8affae2b55a6 100644
--- a/drivers/net/ethernet/realtek/r8169_main.c
+++ b/drivers/net/ethernet/realtek/r8169_main.c
@@ -1431,19 +1431,17 @@ static int rtl8169_get_regs_len(struct net_device *dev)
 	return R8169_REGS_SIZE;
 }
 
-static netdev_features_t rtl8169_fix_features(struct net_device *dev,
-	netdev_features_t features)
+static void rtl8169_fix_features(struct net_device *dev,
+				 netdev_features_t *features)
 {
 	struct rtl8169_private *tp = netdev_priv(dev);
 
 	if (dev->mtu > TD_MSS_MAX)
-		features &= ~NETIF_F_ALL_TSO;
+		*features &= ~NETIF_F_ALL_TSO;
 
 	if (dev->mtu > ETH_DATA_LEN &&
 	    tp->mac_version > RTL_GIGA_MAC_VER_06)
-		features &= ~(NETIF_F_CSUM_MASK | NETIF_F_ALL_TSO);
-
-	return features;
+		*features &= ~(NETIF_F_CSUM_MASK | NETIF_F_ALL_TSO);
 }
 
 static void rtl_set_rx_config_features(struct rtl8169_private *tp,
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index fd4c6517125e..aa977cac3c10 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -5461,16 +5461,16 @@ static int stmmac_change_mtu(struct net_device *dev, int new_mtu)
 	return 0;
 }
 
-static netdev_features_t stmmac_fix_features(struct net_device *dev,
-					     netdev_features_t features)
+static void stmmac_fix_features(struct net_device *dev,
+				netdev_features_t *features)
 {
 	struct stmmac_priv *priv = netdev_priv(dev);
 
 	if (priv->plat->rx_coe == STMMAC_RX_COE_NONE)
-		features &= ~NETIF_F_RXCSUM;
+		*features &= ~NETIF_F_RXCSUM;
 
 	if (!priv->plat->tx_coe)
-		features &= ~NETIF_F_CSUM_MASK;
+		*features &= ~NETIF_F_CSUM_MASK;
 
 	/* Some GMAC devices have a bugged Jumbo frame support that
 	 * needs to have the Tx COE disabled for oversized frames
@@ -5478,17 +5478,15 @@ static netdev_features_t stmmac_fix_features(struct net_device *dev,
 	 * the TX csum insertion in the TDES and not use SF.
 	 */
 	if (priv->plat->bugged_jumbo && (dev->mtu > ETH_DATA_LEN))
-		features &= ~NETIF_F_CSUM_MASK;
+		*features &= ~NETIF_F_CSUM_MASK;
 
 	/* Disable tso if asked by ethtool */
 	if ((priv->plat->tso_en) && (priv->dma_cap.tsoen)) {
-		if (features & NETIF_F_TSO)
+		if (*features & NETIF_F_TSO)
 			priv->tso = true;
 		else
 			priv->tso = false;
 	}
-
-	return features;
 }
 
 static int stmmac_set_features(struct net_device *netdev,
diff --git a/drivers/net/hyperv/netvsc_drv.c b/drivers/net/hyperv/netvsc_drv.c
index 382bebc2420d..5371328422ec 100644
--- a/drivers/net/hyperv/netvsc_drv.c
+++ b/drivers/net/hyperv/netvsc_drv.c
@@ -1919,21 +1919,19 @@ static int netvsc_set_ringparam(struct net_device *ndev,
 	return ret;
 }
 
-static netdev_features_t netvsc_fix_features(struct net_device *ndev,
-					     netdev_features_t features)
+static void netvsc_fix_features(struct net_device *ndev,
+				netdev_features_t *features)
 {
 	struct net_device_context *ndevctx = netdev_priv(ndev);
 	struct netvsc_device *nvdev = rtnl_dereference(ndevctx->nvdev);
 
 	if (!nvdev || nvdev->destroy)
-		return features;
+		return;
 
-	if ((features & NETIF_F_LRO) && netvsc_xdp_get(nvdev)) {
-		features ^= NETIF_F_LRO;
+	if ((*features & NETIF_F_LRO) && netvsc_xdp_get(nvdev)) {
+		*features ^= NETIF_F_LRO;
 		netdev_info(ndev, "Skip LRO - unsupported with XDP\n");
 	}
-
-	return features;
 }
 
 static int netvsc_set_features(struct net_device *ndev,
diff --git a/drivers/net/ipvlan/ipvlan_main.c b/drivers/net/ipvlan/ipvlan_main.c
index c0b21a5580d5..8fcc91c0b0f4 100644
--- a/drivers/net/ipvlan/ipvlan_main.c
+++ b/drivers/net/ipvlan/ipvlan_main.c
@@ -235,19 +235,17 @@ static netdev_tx_t ipvlan_start_xmit(struct sk_buff *skb,
 	return ret;
 }
 
-static netdev_features_t ipvlan_fix_features(struct net_device *dev,
-					     netdev_features_t features)
+static void ipvlan_fix_features(struct net_device *dev,
+				netdev_features_t *features)
 {
 	struct ipvl_dev *ipvlan = netdev_priv(dev);
 
-	features |= NETIF_F_ALL_FOR_ALL;
-	features &= (ipvlan->sfeatures | ~IPVLAN_FEATURES);
-	features = netdev_increment_features(ipvlan->phy_dev->features,
-					     features, features);
-	features |= IPVLAN_ALWAYS_ON;
-	features &= (IPVLAN_FEATURES | IPVLAN_ALWAYS_ON);
-
-	return features;
+	*features |= NETIF_F_ALL_FOR_ALL;
+	*features &= (ipvlan->sfeatures | ~IPVLAN_FEATURES);
+	*features = netdev_increment_features(ipvlan->phy_dev->features,
+					      *features, *features);
+	*features |= IPVLAN_ALWAYS_ON;
+	*features &= (IPVLAN_FEATURES | IPVLAN_ALWAYS_ON);
 }
 
 static void ipvlan_change_rx_flags(struct net_device *dev, int change)
diff --git a/drivers/net/macsec.c b/drivers/net/macsec.c
index 93dc48b9b4f2..601b833a40bd 100644
--- a/drivers/net/macsec.c
+++ b/drivers/net/macsec.c
@@ -3470,20 +3470,20 @@ static void macsec_dev_uninit(struct net_device *dev)
 	free_percpu(dev->tstats);
 }
 
-static netdev_features_t macsec_fix_features(struct net_device *dev,
-					     netdev_features_t features)
+static void macsec_fix_features(struct net_device *dev,
+				netdev_features_t *features)
 {
 	struct macsec_dev *macsec = macsec_priv(dev);
 	struct net_device *real_dev = macsec->real_dev;
 
-	if (macsec_is_offloaded(macsec))
-		return REAL_DEV_FEATURES(real_dev);
+	if (macsec_is_offloaded(macsec)) {
+		*features = REAL_DEV_FEATURES(real_dev);
+		return;
+	}
 
-	features &= (real_dev->features & SW_MACSEC_FEATURES) |
+	*features &= (real_dev->features & SW_MACSEC_FEATURES) |
 		    NETIF_F_GSO_SOFTWARE | NETIF_F_SOFT_FEATURES;
-	features |= NETIF_F_LLTX;
-
-	return features;
+	*features |= NETIF_F_LLTX;
 }
 
 static int macsec_dev_open(struct net_device *dev)
diff --git a/drivers/net/macvlan.c b/drivers/net/macvlan.c
index 35f46ad040b0..27cd9c08bb1e 100644
--- a/drivers/net/macvlan.c
+++ b/drivers/net/macvlan.c
@@ -1067,23 +1067,21 @@ static int macvlan_ethtool_get_ts_info(struct net_device *dev,
 	return 0;
 }
 
-static netdev_features_t macvlan_fix_features(struct net_device *dev,
-					      netdev_features_t features)
+static void macvlan_fix_features(struct net_device *dev,
+				 netdev_features_t *features)
 {
 	struct macvlan_dev *vlan = netdev_priv(dev);
 	netdev_features_t lowerdev_features = vlan->lowerdev->features;
 	netdev_features_t mask;
 
-	features |= NETIF_F_ALL_FOR_ALL;
-	features &= (vlan->set_features | ~MACVLAN_FEATURES);
-	mask = features;
+	*features |= NETIF_F_ALL_FOR_ALL;
+	*features &= (vlan->set_features | ~MACVLAN_FEATURES);
+	mask = *features;
 
-	lowerdev_features &= (features | ~NETIF_F_LRO);
-	features = netdev_increment_features(lowerdev_features, features, mask);
-	features |= ALWAYS_ON_FEATURES;
-	features &= (ALWAYS_ON_FEATURES | MACVLAN_FEATURES);
-
-	return features;
+	lowerdev_features &= (*features | ~NETIF_F_LRO);
+	*features = netdev_increment_features(lowerdev_features, *features, mask);
+	*features |= ALWAYS_ON_FEATURES;
+	*features &= (ALWAYS_ON_FEATURES | MACVLAN_FEATURES);
 }
 
 #ifdef CONFIG_NET_POLL_CONTROLLER
diff --git a/drivers/net/team/team.c b/drivers/net/team/team.c
index dd7917cab2b1..340be925d4eb 100644
--- a/drivers/net/team/team.c
+++ b/drivers/net/team/team.c
@@ -1995,28 +1995,26 @@ static int team_del_slave(struct net_device *dev, struct net_device *port_dev)
 	return err;
 }
 
-static netdev_features_t team_fix_features(struct net_device *dev,
-					   netdev_features_t features)
+static void team_fix_features(struct net_device *dev,
+			      netdev_features_t *features)
 {
 	struct team_port *port;
 	struct team *team = netdev_priv(dev);
 	netdev_features_t mask;
 
-	mask = features;
-	features &= ~NETIF_F_ONE_FOR_ALL;
-	features |= NETIF_F_ALL_FOR_ALL;
+	mask = *features;
+	*features &= ~NETIF_F_ONE_FOR_ALL;
+	*features |= NETIF_F_ALL_FOR_ALL;
 
 	rcu_read_lock();
 	list_for_each_entry_rcu(port, &team->port_list, list) {
-		features = netdev_increment_features(features,
-						     port->dev->features,
-						     mask);
+		*features = netdev_increment_features(*features,
+						      port->dev->features,
+						      mask);
 	}
 	rcu_read_unlock();
 
-	features = netdev_add_tso_features(features, mask);
-
-	return features;
+	*features = netdev_add_tso_features(*features, mask);
 }
 
 static int team_change_carrier(struct net_device *dev, bool new_carrier)
diff --git a/drivers/net/tun.c b/drivers/net/tun.c
index fecc9a1d293a..d89a9874eb37 100644
--- a/drivers/net/tun.c
+++ b/drivers/net/tun.c
@@ -1079,12 +1079,13 @@ static void tun_net_mclist(struct net_device *dev)
 	 */
 }
 
-static netdev_features_t tun_net_fix_features(struct net_device *dev,
-	netdev_features_t features)
+static void tun_net_fix_features(struct net_device *dev,
+				 netdev_features_t *features)
 {
 	struct tun_struct *tun = netdev_priv(dev);
 
-	return (features & tun->set_features) | (features & ~TUN_USER_FEATURES);
+	*features = (*features & tun->set_features) |
+			(*features & ~TUN_USER_FEATURES);
 }
 
 static void tun_set_headroom(struct net_device *dev, int new_hr)
diff --git a/drivers/net/veth.c b/drivers/net/veth.c
index 50eb43e5bf45..82c900d7ba7b 100644
--- a/drivers/net/veth.c
+++ b/drivers/net/veth.c
@@ -1382,8 +1382,8 @@ static int veth_get_iflink(const struct net_device *dev)
 	return iflink;
 }
 
-static netdev_features_t veth_fix_features(struct net_device *dev,
-					   netdev_features_t features)
+static void veth_fix_features(struct net_device *dev,
+			      netdev_features_t *features)
 {
 	struct veth_priv *priv = netdev_priv(dev);
 	struct net_device *peer;
@@ -1393,12 +1393,10 @@ static netdev_features_t veth_fix_features(struct net_device *dev,
 		struct veth_priv *peer_priv = netdev_priv(peer);
 
 		if (peer_priv->_xdp_prog)
-			features &= ~NETIF_F_GSO_SOFTWARE;
+			*features &= ~NETIF_F_GSO_SOFTWARE;
 	}
 	if (priv->_xdp_prog)
-		features |= NETIF_F_GRO;
-
-	return features;
+		*features |= NETIF_F_GRO;
 }
 
 static int veth_set_features(struct net_device *dev,
diff --git a/drivers/net/vmxnet3/vmxnet3_ethtool.c b/drivers/net/vmxnet3/vmxnet3_ethtool.c
index e840547cd19e..6a5827c21c63 100644
--- a/drivers/net/vmxnet3/vmxnet3_ethtool.c
+++ b/drivers/net/vmxnet3/vmxnet3_ethtool.c
@@ -246,14 +246,12 @@ vmxnet3_get_strings(struct net_device *netdev, u32 stringset, u8 *buf)
 		ethtool_sprintf(&buf, vmxnet3_global_stats[i].desc);
 }
 
-netdev_features_t vmxnet3_fix_features(struct net_device *netdev,
-				       netdev_features_t features)
+void vmxnet3_fix_features(struct net_device *netdev,
+			  netdev_features_t *features)
 {
 	/* If Rx checksum is disabled, then LRO should also be disabled */
-	if (!(features & NETIF_F_RXCSUM))
-		features &= ~NETIF_F_LRO;
-
-	return features;
+	if (!(*features & NETIF_F_RXCSUM))
+		*features &= ~NETIF_F_LRO;
 }
 
 void vmxnet3_features_check(struct sk_buff *skb, struct net_device *netdev,
diff --git a/drivers/net/vmxnet3/vmxnet3_int.h b/drivers/net/vmxnet3/vmxnet3_int.h
index a711759ebb00..aed4e1bf9298 100644
--- a/drivers/net/vmxnet3/vmxnet3_int.h
+++ b/drivers/net/vmxnet3/vmxnet3_int.h
@@ -475,8 +475,8 @@ vmxnet3_tq_destroy_all(struct vmxnet3_adapter *adapter);
 void
 vmxnet3_rq_destroy_all(struct vmxnet3_adapter *adapter);
 
-netdev_features_t
-vmxnet3_fix_features(struct net_device *netdev, netdev_features_t features);
+void
+vmxnet3_fix_features(struct net_device *netdev, netdev_features_t *features);
 
 void
 vmxnet3_features_check(struct sk_buff *skb,
diff --git a/drivers/net/xen-netback/interface.c b/drivers/net/xen-netback/interface.c
index c58996c1e230..f964f0a402a6 100644
--- a/drivers/net/xen-netback/interface.c
+++ b/drivers/net/xen-netback/interface.c
@@ -374,23 +374,21 @@ static int xenvif_change_mtu(struct net_device *dev, int mtu)
 	return 0;
 }
 
-static netdev_features_t xenvif_fix_features(struct net_device *dev,
-	netdev_features_t features)
+static void xenvif_fix_features(struct net_device *dev,
+				netdev_features_t *features)
 {
 	struct xenvif *vif = netdev_priv(dev);
 
 	if (!vif->can_sg)
-		features &= ~NETIF_F_SG;
+		*features &= ~NETIF_F_SG;
 	if (~(vif->gso_mask) & GSO_BIT(TCPV4))
-		features &= ~NETIF_F_TSO;
+		*features &= ~NETIF_F_TSO;
 	if (~(vif->gso_mask) & GSO_BIT(TCPV6))
-		features &= ~NETIF_F_TSO6;
+		*features &= ~NETIF_F_TSO6;
 	if (!vif->ip_csum)
-		features &= ~NETIF_F_IP_CSUM;
+		*features &= ~NETIF_F_IP_CSUM;
 	if (!vif->ipv6_csum)
-		features &= ~NETIF_F_IPV6_CSUM;
-
-	return features;
+		*features &= ~NETIF_F_IPV6_CSUM;
 }
 
 static const struct xenvif_stat {
diff --git a/drivers/net/xen-netfront.c b/drivers/net/xen-netfront.c
index 65c134ac2be5..79752f16277a 100644
--- a/drivers/net/xen-netfront.c
+++ b/drivers/net/xen-netfront.c
@@ -1386,29 +1386,27 @@ static void xennet_release_rx_bufs(struct netfront_queue *queue)
 	spin_unlock_bh(&queue->rx_lock);
 }
 
-static netdev_features_t xennet_fix_features(struct net_device *dev,
-	netdev_features_t features)
+static void xennet_fix_features(struct net_device *dev,
+				netdev_features_t *features)
 {
 	struct netfront_info *np = netdev_priv(dev);
 
-	if (features & NETIF_F_SG &&
+	if (*features & NETIF_F_SG &&
 	    !xenbus_read_unsigned(np->xbdev->otherend, "feature-sg", 0))
-		features &= ~NETIF_F_SG;
+		*features &= ~NETIF_F_SG;
 
-	if (features & NETIF_F_IPV6_CSUM &&
+	if (*features & NETIF_F_IPV6_CSUM &&
 	    !xenbus_read_unsigned(np->xbdev->otherend,
 				  "feature-ipv6-csum-offload", 0))
-		features &= ~NETIF_F_IPV6_CSUM;
+		*features &= ~NETIF_F_IPV6_CSUM;
 
-	if (features & NETIF_F_TSO &&
+	if (*features & NETIF_F_TSO &&
 	    !xenbus_read_unsigned(np->xbdev->otherend, "feature-gso-tcpv4", 0))
-		features &= ~NETIF_F_TSO;
+		*features &= ~NETIF_F_TSO;
 
-	if (features & NETIF_F_TSO6 &&
+	if (*features & NETIF_F_TSO6 &&
 	    !xenbus_read_unsigned(np->xbdev->otherend, "feature-gso-tcpv6", 0))
-		features &= ~NETIF_F_TSO6;
-
-	return features;
+		*features &= ~NETIF_F_TSO6;
 }
 
 static int xennet_set_features(struct net_device *dev,
diff --git a/drivers/s390/net/qeth_core.h b/drivers/s390/net/qeth_core.h
index 58bd0dc43695..5fd8151b355b 100644
--- a/drivers/s390/net/qeth_core.h
+++ b/drivers/s390/net/qeth_core.h
@@ -1093,7 +1093,7 @@ int qeth_hw_trap(struct qeth_card *, enum qeth_diags_trap_action);
 int qeth_setassparms_cb(struct qeth_card *, struct qeth_reply *, unsigned long);
 int qeth_set_features(struct net_device *, netdev_features_t);
 void qeth_enable_hw_features(struct net_device *dev);
-netdev_features_t qeth_fix_features(struct net_device *, netdev_features_t);
+void qeth_fix_features(struct net_device *dev, netdev_features_t *features);
 void qeth_features_check(struct sk_buff *skb, struct net_device *dev,
 			 netdev_features_t *features);
 void qeth_get_stats64(struct net_device *dev, struct rtnl_link_stats64 *stats);
diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c
index 46b5ad0171fa..6a1941e5fd51 100644
--- a/drivers/s390/net/qeth_core_main.c
+++ b/drivers/s390/net/qeth_core_main.c
@@ -6923,26 +6923,24 @@ int qeth_set_features(struct net_device *dev, netdev_features_t features)
 }
 EXPORT_SYMBOL_GPL(qeth_set_features);
 
-netdev_features_t qeth_fix_features(struct net_device *dev,
-				    netdev_features_t features)
+void qeth_fix_features(struct net_device *dev, netdev_features_t *features)
 {
 	struct qeth_card *card = dev->ml_priv;
 
 	QETH_CARD_TEXT(card, 2, "fixfeat");
 	if (!qeth_is_supported(card, IPA_OUTBOUND_CHECKSUM))
-		features &= ~NETIF_F_IP_CSUM;
+		*features &= ~NETIF_F_IP_CSUM;
 	if (!qeth_is_supported6(card, IPA_OUTBOUND_CHECKSUM_V6))
-		features &= ~NETIF_F_IPV6_CSUM;
+		*features &= ~NETIF_F_IPV6_CSUM;
 	if (!qeth_is_supported(card, IPA_INBOUND_CHECKSUM) &&
 	    !qeth_is_supported6(card, IPA_INBOUND_CHECKSUM_V6))
-		features &= ~NETIF_F_RXCSUM;
+		*features &= ~NETIF_F_RXCSUM;
 	if (!qeth_is_supported(card, IPA_OUTBOUND_TSO))
-		features &= ~NETIF_F_TSO;
+		*features &= ~NETIF_F_TSO;
 	if (!qeth_is_supported6(card, IPA_OUTBOUND_TSO))
-		features &= ~NETIF_F_TSO6;
+		*features &= ~NETIF_F_TSO6;
 
-	QETH_CARD_HEX(card, 2, &features, sizeof(features));
-	return features;
+	QETH_CARD_HEX(card, 2, features, sizeof(*features));
 }
 EXPORT_SYMBOL_GPL(qeth_fix_features);
 
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index d62edd4c99a9..43eb57fa9434 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -1481,8 +1481,8 @@ struct net_device_ops {
 						      bool all_slaves);
 	struct net_device*	(*ndo_sk_get_lower_dev)(struct net_device *dev,
 							struct sock *sk);
-	netdev_features_t	(*ndo_fix_features)(struct net_device *dev,
-						    netdev_features_t features);
+	void			(*ndo_fix_features)(struct net_device *dev,
+						    netdev_features_t *features);
 	int			(*ndo_set_features)(struct net_device *dev,
 						    netdev_features_t features);
 	int			(*ndo_neigh_construct)(struct net_device *dev,
diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c
index 2987201ec93d..9f90a587d4e5 100644
--- a/net/8021q/vlan_dev.c
+++ b/net/8021q/vlan_dev.c
@@ -633,11 +633,11 @@ void vlan_dev_uninit(struct net_device *dev)
 	}
 }
 
-static netdev_features_t vlan_dev_fix_features(struct net_device *dev,
-	netdev_features_t features)
+static void vlan_dev_fix_features(struct net_device *dev,
+				  netdev_features_t *features)
 {
 	struct net_device *real_dev = vlan_dev_priv(dev)->real_dev;
-	netdev_features_t old_features = features;
+	netdev_features_t old_features = *features;
 	netdev_features_t lower_features;
 
 	netdev_intersect_features(&lower_features,
@@ -649,11 +649,9 @@ static netdev_features_t vlan_dev_fix_features(struct net_device *dev,
 	 */
 	if (lower_features & (NETIF_F_IP_CSUM|NETIF_F_IPV6_CSUM))
 		lower_features |= NETIF_F_HW_CSUM;
-	netdev_intersect_features(&features, features, lower_features);
-	features |= old_features & (NETIF_F_SOFT_FEATURES | NETIF_F_GSO_SOFTWARE);
-	features |= NETIF_F_LLTX;
-
-	return features;
+	netdev_intersect_features(features, *features, lower_features);
+	*features |= old_features & (NETIF_F_SOFT_FEATURES | NETIF_F_GSO_SOFTWARE);
+	*features |= NETIF_F_LLTX;
 }
 
 static int vlan_ethtool_get_link_ksettings(struct net_device *dev,
diff --git a/net/bridge/br_device.c b/net/bridge/br_device.c
index 8d6bab244c4a..622559aff2dd 100644
--- a/net/bridge/br_device.c
+++ b/net/bridge/br_device.c
@@ -288,12 +288,11 @@ static int br_get_link_ksettings(struct net_device *dev,
 	return 0;
 }
 
-static netdev_features_t br_fix_features(struct net_device *dev,
-	netdev_features_t features)
+static void br_fix_features(struct net_device *dev, netdev_features_t *features)
 {
 	struct net_bridge *br = netdev_priv(dev);
 
-	return br_features_recompute(br, features);
+	*features = br_features_recompute(br, *features);
 }
 
 #ifdef CONFIG_NET_POLL_CONTROLLER
diff --git a/net/core/dev.c b/net/core/dev.c
index 8f6316bee565..8a2de66e709b 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -9933,7 +9933,7 @@ int __netdev_update_features(struct net_device *dev)
 	netdev_get_wanted_features(dev, &features);
 
 	if (dev->netdev_ops->ndo_fix_features)
-		features = dev->netdev_ops->ndo_fix_features(dev, features);
+		dev->netdev_ops->ndo_fix_features(dev, &features);
 
 	/* driver might be less strict about feature dependencies */
 	features = netdev_fix_features(dev, features);
diff --git a/net/hsr/hsr_device.c b/net/hsr/hsr_device.c
index 26c32407f029..acaf48a1e136 100644
--- a/net/hsr/hsr_device.c
+++ b/net/hsr/hsr_device.c
@@ -201,12 +201,12 @@ static netdev_features_t hsr_features_recompute(struct hsr_priv *hsr,
 	return features;
 }
 
-static netdev_features_t hsr_fix_features(struct net_device *dev,
-					  netdev_features_t features)
+static void hsr_fix_features(struct net_device *dev,
+			     netdev_features_t *features)
 {
 	struct hsr_priv *hsr = netdev_priv(dev);
 
-	return hsr_features_recompute(hsr, features);
+	*features = hsr_features_recompute(hsr, *features);
 }
 
 static netdev_tx_t hsr_dev_xmit(struct sk_buff *skb, struct net_device *dev)
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 012/167] net: convert the prototype of netdev_fix_features
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (10 preceding siblings ...)
  2021-09-29 15:50 ` [RFCv2 net-next 011/167] net: convert the prototype of ndo_fix_features Jian Shen
@ 2021-09-29 15:50 ` Jian Shen
  2021-09-29 15:51 ` [RFCv2 net-next 013/167] net: convert the prototype of netdev_sync_upper_features Jian Shen
                   ` (158 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:50 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

For the origin type for netdev_features_t would be changed to
be unsigned long * from u64, so changes the prototype of
netdev_fix_features for adaption.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 net/core/dev.c | 78 ++++++++++++++++++++++++--------------------------
 1 file changed, 38 insertions(+), 40 deletions(-)

diff --git a/net/core/dev.c b/net/core/dev.c
index 8a2de66e709b..a03a01e5339e 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -9826,99 +9826,97 @@ static void netdev_sync_lower_features(struct net_device *upper,
 	}
 }
 
-static netdev_features_t netdev_fix_features(struct net_device *dev,
-	netdev_features_t features)
+static void netdev_fix_features(struct net_device *dev,
+				netdev_features_t *features)
 {
 	/* Fix illegal checksum combinations */
-	if ((features & NETIF_F_HW_CSUM) &&
-	    (features & (NETIF_F_IP_CSUM|NETIF_F_IPV6_CSUM))) {
+	if ((*features & NETIF_F_HW_CSUM) &&
+	    (*features & (NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM))) {
 		netdev_warn(dev, "mixed HW and IP checksum settings.\n");
-		features &= ~(NETIF_F_IP_CSUM|NETIF_F_IPV6_CSUM);
+		*features &= ~(NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM);
 	}
 
 	/* TSO requires that SG is present as well. */
-	if ((features & NETIF_F_ALL_TSO) && !(features & NETIF_F_SG)) {
+	if ((*features & NETIF_F_ALL_TSO) && !(*features & NETIF_F_SG)) {
 		netdev_dbg(dev, "Dropping TSO features since no SG feature.\n");
-		features &= ~NETIF_F_ALL_TSO;
+		*features &= ~NETIF_F_ALL_TSO;
 	}
 
-	if ((features & NETIF_F_TSO) && !(features & NETIF_F_HW_CSUM) &&
-					!(features & NETIF_F_IP_CSUM)) {
+	if ((*features & NETIF_F_TSO) && !(*features & NETIF_F_HW_CSUM) &&
+	    !(*features & NETIF_F_IP_CSUM)) {
 		netdev_dbg(dev, "Dropping TSO features since no CSUM feature.\n");
-		features &= ~NETIF_F_TSO;
-		features &= ~NETIF_F_TSO_ECN;
+		*features &= ~NETIF_F_TSO;
+		*features &= ~NETIF_F_TSO_ECN;
 	}
 
-	if ((features & NETIF_F_TSO6) && !(features & NETIF_F_HW_CSUM) &&
-					 !(features & NETIF_F_IPV6_CSUM)) {
+	if ((*features & NETIF_F_TSO6) && !(*features & NETIF_F_HW_CSUM) &&
+	    !(*features & NETIF_F_IPV6_CSUM)) {
 		netdev_dbg(dev, "Dropping TSO6 features since no CSUM feature.\n");
-		features &= ~NETIF_F_TSO6;
+		*features &= ~NETIF_F_TSO6;
 	}
 
 	/* TSO with IPv4 ID mangling requires IPv4 TSO be enabled */
-	if ((features & NETIF_F_TSO_MANGLEID) && !(features & NETIF_F_TSO))
-		features &= ~NETIF_F_TSO_MANGLEID;
+	if ((*features & NETIF_F_TSO_MANGLEID) && !(*features & NETIF_F_TSO))
+		*features &= ~NETIF_F_TSO_MANGLEID;
 
 	/* TSO ECN requires that TSO is present as well. */
-	if ((features & NETIF_F_ALL_TSO) == NETIF_F_TSO_ECN)
-		features &= ~NETIF_F_TSO_ECN;
+	if ((*features & NETIF_F_ALL_TSO) == NETIF_F_TSO_ECN)
+		*features &= ~NETIF_F_TSO_ECN;
 
 	/* Software GSO depends on SG. */
-	if ((features & NETIF_F_GSO) && !(features & NETIF_F_SG)) {
+	if ((*features & NETIF_F_GSO) && !(*features & NETIF_F_SG)) {
 		netdev_dbg(dev, "Dropping NETIF_F_GSO since no SG feature.\n");
-		features &= ~NETIF_F_GSO;
+		*features &= ~NETIF_F_GSO;
 	}
 
 	/* GSO partial features require GSO partial be set */
-	if ((features & dev->gso_partial_features) &&
-	    !(features & NETIF_F_GSO_PARTIAL)) {
+	if ((*features & dev->gso_partial_features) &&
+	    !(*features & NETIF_F_GSO_PARTIAL)) {
 		netdev_dbg(dev,
 			   "Dropping partially supported GSO features since no GSO partial.\n");
-		features &= ~dev->gso_partial_features;
+		*features &= ~dev->gso_partial_features;
 	}
 
-	if (!(features & NETIF_F_RXCSUM)) {
+	if (!(*features & NETIF_F_RXCSUM)) {
 		/* NETIF_F_GRO_HW implies doing RXCSUM since every packet
 		 * successfully merged by hardware must also have the
 		 * checksum verified by hardware.  If the user does not
 		 * want to enable RXCSUM, logically, we should disable GRO_HW.
 		 */
-		if (features & NETIF_F_GRO_HW) {
+		if (*features & NETIF_F_GRO_HW) {
 			netdev_dbg(dev, "Dropping NETIF_F_GRO_HW since no RXCSUM feature.\n");
-			features &= ~NETIF_F_GRO_HW;
+			*features &= ~NETIF_F_GRO_HW;
 		}
 	}
 
 	/* LRO/HW-GRO features cannot be combined with RX-FCS */
-	if (features & NETIF_F_RXFCS) {
-		if (features & NETIF_F_LRO) {
+	if (*features & NETIF_F_RXFCS) {
+		if (*features & NETIF_F_LRO) {
 			netdev_dbg(dev, "Dropping LRO feature since RX-FCS is requested.\n");
-			features &= ~NETIF_F_LRO;
+			*features &= ~NETIF_F_LRO;
 		}
 
-		if (features & NETIF_F_GRO_HW) {
+		if (*features & NETIF_F_GRO_HW) {
 			netdev_dbg(dev, "Dropping HW-GRO feature since RX-FCS is requested.\n");
-			features &= ~NETIF_F_GRO_HW;
+			*features &= ~NETIF_F_GRO_HW;
 		}
 	}
 
-	if (features & NETIF_F_HW_TLS_TX) {
-		bool ip_csum = (features & (NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM)) ==
+	if (*features & NETIF_F_HW_TLS_TX) {
+		bool ip_csum = (*features & (NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM)) ==
 			(NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM);
-		bool hw_csum = features & NETIF_F_HW_CSUM;
+		bool hw_csum = *features & NETIF_F_HW_CSUM;
 
 		if (!ip_csum && !hw_csum) {
 			netdev_dbg(dev, "Dropping TLS TX HW offload feature since no CSUM feature.\n");
-			features &= ~NETIF_F_HW_TLS_TX;
+			*features &= ~NETIF_F_HW_TLS_TX;
 		}
 	}
 
-	if ((features & NETIF_F_HW_TLS_RX) && !(features & NETIF_F_RXCSUM)) {
+	if ((*features & NETIF_F_HW_TLS_RX) && !(*features & NETIF_F_RXCSUM)) {
 		netdev_dbg(dev, "Dropping TLS RX HW offload feature since no RXCSUM feature.\n");
-		features &= ~NETIF_F_HW_TLS_RX;
+		*features &= ~NETIF_F_HW_TLS_RX;
 	}
-
-	return features;
 }
 
 int __netdev_update_features(struct net_device *dev)
@@ -9936,7 +9934,7 @@ int __netdev_update_features(struct net_device *dev)
 		dev->netdev_ops->ndo_fix_features(dev, &features);
 
 	/* driver might be less strict about feature dependencies */
-	features = netdev_fix_features(dev, features);
+	netdev_fix_features(dev, &features);
 
 	/* some features can't be enabled if they're off on an upper device */
 	netdev_for_each_upper_dev_rcu(dev, upper, iter)
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 013/167] net: convert the prototype of netdev_sync_upper_features
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (11 preceding siblings ...)
  2021-09-29 15:50 ` [RFCv2 net-next 012/167] net: convert the prototype of netdev_fix_features Jian Shen
@ 2021-09-29 15:51 ` Jian Shen
  2021-09-29 15:51 ` [RFCv2 net-next 014/167] net: convert the prototype of br_features_recompute Jian Shen
                   ` (157 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:51 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

For the origin type for netdev_features_t would be changed to
be unsigned long * from u64, so changes the prototype of
netdev_sync_upper_features for adaption.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 net/core/dev.c | 15 +++++++--------
 1 file changed, 7 insertions(+), 8 deletions(-)

diff --git a/net/core/dev.c b/net/core/dev.c
index a03a01e5339e..58c46131126b 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -9782,8 +9782,9 @@ static void net_set_todo(struct net_device *dev)
 	dev_net(dev)->dev_unreg_count++;
 }
 
-static netdev_features_t netdev_sync_upper_features(struct net_device *lower,
-	struct net_device *upper, netdev_features_t features)
+static void netdev_sync_upper_features(struct net_device *lower,
+				       struct net_device *upper,
+				       netdev_features_t *features)
 {
 	netdev_features_t upper_disables = NETIF_F_UPPER_DISABLES;
 	netdev_features_t feature;
@@ -9791,15 +9792,13 @@ static netdev_features_t netdev_sync_upper_features(struct net_device *lower,
 
 	for_each_netdev_feature(upper_disables, feature_bit) {
 		feature = __NETIF_F_BIT(feature_bit);
-		if (!(upper->wanted_features & feature)
-		    && (features & feature)) {
+		if (!(upper->wanted_features & feature) &&
+		    (*features & feature)) {
 			netdev_dbg(lower, "Dropping feature %pNF, upper dev %s has it off.\n",
 				   &feature, upper->name);
-			features &= ~feature;
+			*features &= ~feature;
 		}
 	}
-
-	return features;
 }
 
 static void netdev_sync_lower_features(struct net_device *upper,
@@ -9938,7 +9937,7 @@ int __netdev_update_features(struct net_device *dev)
 
 	/* some features can't be enabled if they're off on an upper device */
 	netdev_for_each_upper_dev_rcu(dev, upper, iter)
-		features = netdev_sync_upper_features(dev, upper, features);
+		netdev_sync_upper_features(dev, upper, &features);
 
 	if (dev->features == features)
 		goto sync_lower;
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 014/167] net: convert the prototype of br_features_recompute
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (12 preceding siblings ...)
  2021-09-29 15:51 ` [RFCv2 net-next 013/167] net: convert the prototype of netdev_sync_upper_features Jian Shen
@ 2021-09-29 15:51 ` Jian Shen
  2021-09-29 15:51 ` [RFCv2 net-next 015/167] net: convert the prototype of netdev_add_tso_features Jian Shen
                   ` (156 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:51 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

For the origin type for netdev_features_t would be changed to
be unsigned long * from u64, so changes the prototype of
br_features_recompute for adaption.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 net/bridge/br_device.c  |  2 +-
 net/bridge/br_if.c      | 17 +++++++----------
 net/bridge/br_private.h |  3 +--
 3 files changed, 9 insertions(+), 13 deletions(-)

diff --git a/net/bridge/br_device.c b/net/bridge/br_device.c
index 622559aff2dd..fc508b9cbaa9 100644
--- a/net/bridge/br_device.c
+++ b/net/bridge/br_device.c
@@ -292,7 +292,7 @@ static void br_fix_features(struct net_device *dev, netdev_features_t *features)
 {
 	struct net_bridge *br = netdev_priv(dev);
 
-	*features = br_features_recompute(br, *features);
+	br_features_recompute(br, features);
 }
 
 #ifdef CONFIG_NET_POLL_CONTROLLER
diff --git a/net/bridge/br_if.c b/net/bridge/br_if.c
index 4a02f8bb278a..b2afbea2133e 100644
--- a/net/bridge/br_if.c
+++ b/net/bridge/br_if.c
@@ -532,25 +532,22 @@ static void br_set_gso_limits(struct net_bridge *br)
 /*
  * Recomputes features using slave's features
  */
-netdev_features_t br_features_recompute(struct net_bridge *br,
-	netdev_features_t features)
+void br_features_recompute(struct net_bridge *br, netdev_features_t *features)
 {
 	struct net_bridge_port *p;
 	netdev_features_t mask;
 
 	if (list_empty(&br->port_list))
-		return features;
+		return;
 
-	mask = features;
-	features &= ~NETIF_F_ONE_FOR_ALL;
+	mask = *features;
+	*features &= ~NETIF_F_ONE_FOR_ALL;
 
 	list_for_each_entry(p, &br->port_list, list) {
-		features = netdev_increment_features(features,
-						     p->dev->features, mask);
+		*features = netdev_increment_features(*features,
+						      p->dev->features, mask);
 	}
-	features = netdev_add_tso_features(features, mask);
-
-	return features;
+	*features = netdev_add_tso_features(*features, mask);
 }
 
 /* called with RTNL */
diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h
index b4cef3a97f12..1f5cff791827 100644
--- a/net/bridge/br_private.h
+++ b/net/bridge/br_private.h
@@ -824,8 +824,7 @@ int br_add_if(struct net_bridge *br, struct net_device *dev,
 	      struct netlink_ext_ack *extack);
 int br_del_if(struct net_bridge *br, struct net_device *dev);
 void br_mtu_auto_adjust(struct net_bridge *br);
-netdev_features_t br_features_recompute(struct net_bridge *br,
-					netdev_features_t features);
+void br_features_recompute(struct net_bridge *br, netdev_features_t *features);
 void br_port_flags_change(struct net_bridge_port *port, unsigned long mask);
 void br_manage_promisc(struct net_bridge *br);
 int nbp_backup_change(struct net_bridge_port *p, struct net_device *backup_dev);
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 015/167] net: convert the prototype of netdev_add_tso_features
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (13 preceding siblings ...)
  2021-09-29 15:51 ` [RFCv2 net-next 014/167] net: convert the prototype of br_features_recompute Jian Shen
@ 2021-09-29 15:51 ` Jian Shen
  2021-09-29 15:51 ` [RFCv2 net-next 016/167] net: convert the prototype of netdev_increment_features Jian Shen
                   ` (155 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:51 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

For the origin type for netdev_features_t would be changed to
be unsigned long * from u64, so changes the prototype of
netdev_add_tso_features for adaption.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 drivers/net/bonding/bond_main.c | 2 +-
 drivers/net/team/team.c         | 2 +-
 include/linux/netdevice.h       | 6 +++---
 net/bridge/br_if.c              | 2 +-
 4 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index 469509260a51..878c92746ada 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -1386,7 +1386,7 @@ static void bond_fix_features(struct net_device *dev,
 						      slave->dev->features,
 						      mask);
 	}
-	*features = netdev_add_tso_features(*features, mask);
+	netdev_add_tso_features(features, mask);
 }
 
 #define BOND_VLAN_FEATURES	(NETIF_F_HW_CSUM | NETIF_F_SG | \
diff --git a/drivers/net/team/team.c b/drivers/net/team/team.c
index 340be925d4eb..706572b7a313 100644
--- a/drivers/net/team/team.c
+++ b/drivers/net/team/team.c
@@ -2014,7 +2014,7 @@ static void team_fix_features(struct net_device *dev,
 	}
 	rcu_read_unlock();
 
-	*features = netdev_add_tso_features(*features, mask);
+	netdev_add_tso_features(features, mask);
 }
 
 static int team_change_carrier(struct net_device *dev, bool new_carrier)
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 43eb57fa9434..e826435ab847 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -5043,10 +5043,10 @@ netdev_features_t netdev_increment_features(netdev_features_t all,
  * Performing the GSO segmentation before last device
  * is a performance improvement.
  */
-static inline netdev_features_t netdev_add_tso_features(netdev_features_t features,
-							netdev_features_t mask)
+static inline void netdev_add_tso_features(netdev_features_t *features,
+					   netdev_features_t mask)
 {
-	return netdev_increment_features(features, NETIF_F_ALL_TSO, mask);
+	*features = netdev_increment_features(*features, NETIF_F_ALL_TSO, mask);
 }
 
 int __netdev_update_features(struct net_device *dev);
diff --git a/net/bridge/br_if.c b/net/bridge/br_if.c
index b2afbea2133e..749971ab1088 100644
--- a/net/bridge/br_if.c
+++ b/net/bridge/br_if.c
@@ -547,7 +547,7 @@ void br_features_recompute(struct net_bridge *br, netdev_features_t *features)
 		*features = netdev_increment_features(*features,
 						      p->dev->features, mask);
 	}
-	*features = netdev_add_tso_features(*features, mask);
+	netdev_add_tso_features(features, mask);
 }
 
 /* called with RTNL */
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 016/167] net: convert the prototype of netdev_increment_features
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (14 preceding siblings ...)
  2021-09-29 15:51 ` [RFCv2 net-next 015/167] net: convert the prototype of netdev_add_tso_features Jian Shen
@ 2021-09-29 15:51 ` Jian Shen
  2021-09-29 15:51 ` [RFCv2 net-next 017/167] net: convert the prototype of hsr_features_recompute Jian Shen
                   ` (154 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:51 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

For the origin type for netdev_features_t would be changed to
be unsigned long * from u64, so changes the prototype of
netdev_increment_features for adaption.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 drivers/net/bonding/bond_main.c  | 28 ++++++++++++++--------------
 drivers/net/ipvlan/ipvlan_main.c |  4 ++--
 drivers/net/macvlan.c            |  2 +-
 drivers/net/net_failover.c       | 28 ++++++++++++----------------
 drivers/net/team/team.c          | 19 ++++++++-----------
 include/linux/netdevice.h        |  6 +++---
 net/bridge/br_if.c               |  4 ++--
 net/core/dev.c                   |  7 ++++---
 net/hsr/hsr_device.c             |  5 ++---
 9 files changed, 48 insertions(+), 55 deletions(-)

diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index 878c92746ada..b1c13fa3c677 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -1382,9 +1382,8 @@ static void bond_fix_features(struct net_device *dev,
 	*features |= NETIF_F_ALL_FOR_ALL;
 
 	bond_for_each_slave(bond, slave, iter) {
-		*features = netdev_increment_features(*features,
-						      slave->dev->features,
-						      mask);
+		netdev_increment_features(features, *features,
+					  slave->dev->features, mask);
 	}
 	netdev_add_tso_features(features, mask);
 }
@@ -1423,22 +1422,23 @@ static void bond_compute_features(struct bonding *bond)
 	mpls_features &= NETIF_F_ALL_FOR_ALL;
 
 	bond_for_each_slave(bond, slave, iter) {
-		vlan_features = netdev_increment_features(vlan_features,
-			slave->dev->vlan_features, BOND_VLAN_FEATURES);
+		netdev_increment_features(&vlan_features, vlan_features,
+					  slave->dev->vlan_features,
+					  BOND_VLAN_FEATURES);
 
-		enc_features = netdev_increment_features(enc_features,
-							 slave->dev->hw_enc_features,
-							 BOND_ENC_FEATURES);
+		netdev_increment_features(&enc_features, enc_features,
+					  slave->dev->hw_enc_features,
+					  BOND_ENC_FEATURES);
 
 #ifdef CONFIG_XFRM_OFFLOAD
-		xfrm_features = netdev_increment_features(xfrm_features,
-							  slave->dev->hw_enc_features,
-							  BOND_XFRM_FEATURES);
+		netdev_increment_features(&xfrm_features, xfrm_features,
+					  slave->dev->hw_enc_features,
+					  BOND_XFRM_FEATURES);
 #endif /* CONFIG_XFRM_OFFLOAD */
 
-		mpls_features = netdev_increment_features(mpls_features,
-							  slave->dev->mpls_features,
-							  BOND_MPLS_FEATURES);
+		netdev_increment_features(&mpls_features, mpls_features,
+					  slave->dev->mpls_features,
+					  BOND_MPLS_FEATURES);
 
 		dst_release_flag &= slave->dev->priv_flags;
 		if (slave->dev->hard_header_len > max_hard_header_len)
diff --git a/drivers/net/ipvlan/ipvlan_main.c b/drivers/net/ipvlan/ipvlan_main.c
index 8fcc91c0b0f4..6a0b7bd2d3ae 100644
--- a/drivers/net/ipvlan/ipvlan_main.c
+++ b/drivers/net/ipvlan/ipvlan_main.c
@@ -242,8 +242,8 @@ static void ipvlan_fix_features(struct net_device *dev,
 
 	*features |= NETIF_F_ALL_FOR_ALL;
 	*features &= (ipvlan->sfeatures | ~IPVLAN_FEATURES);
-	*features = netdev_increment_features(ipvlan->phy_dev->features,
-					      *features, *features);
+	netdev_increment_features(features, ipvlan->phy_dev->features,
+				  *features, *features);
 	*features |= IPVLAN_ALWAYS_ON;
 	*features &= (IPVLAN_FEATURES | IPVLAN_ALWAYS_ON);
 }
diff --git a/drivers/net/macvlan.c b/drivers/net/macvlan.c
index 27cd9c08bb1e..3c408653e864 100644
--- a/drivers/net/macvlan.c
+++ b/drivers/net/macvlan.c
@@ -1079,7 +1079,7 @@ static void macvlan_fix_features(struct net_device *dev,
 	mask = *features;
 
 	lowerdev_features &= (*features | ~NETIF_F_LRO);
-	*features = netdev_increment_features(lowerdev_features, *features, mask);
+	netdev_increment_features(features, lowerdev_features, *features, mask);
 	*features |= ALWAYS_ON_FEATURES;
 	*features &= (ALWAYS_ON_FEATURES | MACVLAN_FEATURES);
 }
diff --git a/drivers/net/net_failover.c b/drivers/net/net_failover.c
index 2a4892402ed8..878cad216aaf 100644
--- a/drivers/net/net_failover.c
+++ b/drivers/net/net_failover.c
@@ -390,14 +390,12 @@ static void net_failover_compute_features(struct net_device *dev)
 
 	primary_dev = rcu_dereference(nfo_info->primary_dev);
 	if (primary_dev) {
-		vlan_features =
-			netdev_increment_features(vlan_features,
-						  primary_dev->vlan_features,
-						  FAILOVER_VLAN_FEATURES);
-		enc_features =
-			netdev_increment_features(enc_features,
-						  primary_dev->hw_enc_features,
-						  FAILOVER_ENC_FEATURES);
+		netdev_increment_features(&vlan_features, vlan_features,
+					  primary_dev->vlan_features,
+					  FAILOVER_VLAN_FEATURES);
+		netdev_increment_features(&enc_features, enc_features,
+					  primary_dev->hw_enc_features,
+					  FAILOVER_ENC_FEATURES);
 
 		dst_release_flag &= primary_dev->priv_flags;
 		if (primary_dev->hard_header_len > max_hard_header_len)
@@ -406,14 +404,12 @@ static void net_failover_compute_features(struct net_device *dev)
 
 	standby_dev = rcu_dereference(nfo_info->standby_dev);
 	if (standby_dev) {
-		vlan_features =
-			netdev_increment_features(vlan_features,
-						  standby_dev->vlan_features,
-						  FAILOVER_VLAN_FEATURES);
-		enc_features =
-			netdev_increment_features(enc_features,
-						  standby_dev->hw_enc_features,
-						  FAILOVER_ENC_FEATURES);
+		netdev_increment_features(&vlan_features, vlan_features,
+					  standby_dev->vlan_features,
+					  FAILOVER_VLAN_FEATURES);
+		netdev_increment_features(&enc_features, enc_features,
+					  standby_dev->hw_enc_features,
+					  FAILOVER_ENC_FEATURES);
 
 		dst_release_flag &= standby_dev->priv_flags;
 		if (standby_dev->hard_header_len > max_hard_header_len)
diff --git a/drivers/net/team/team.c b/drivers/net/team/team.c
index 706572b7a313..fa96ee62c91a 100644
--- a/drivers/net/team/team.c
+++ b/drivers/net/team/team.c
@@ -994,14 +994,12 @@ static void __team_compute_features(struct team *team)
 
 	rcu_read_lock();
 	list_for_each_entry_rcu(port, &team->port_list, list) {
-		vlan_features = netdev_increment_features(vlan_features,
-					port->dev->vlan_features,
-					TEAM_VLAN_FEATURES);
-		enc_features =
-			netdev_increment_features(enc_features,
-						  port->dev->hw_enc_features,
-						  TEAM_ENC_FEATURES);
-
+		netdev_increment_features(&vlan_features, vlan_features,
+					  port->dev->vlan_features,
+					  TEAM_VLAN_FEATURES);
+		netdev_increment_features(&enc_features, enc_features,
+					  port->dev->hw_enc_features,
+					  TEAM_ENC_FEATURES);
 
 		dst_release_flag &= port->dev->priv_flags;
 		if (port->dev->hard_header_len > max_hard_header_len)
@@ -2008,9 +2006,8 @@ static void team_fix_features(struct net_device *dev,
 
 	rcu_read_lock();
 	list_for_each_entry_rcu(port, &team->port_list, list) {
-		*features = netdev_increment_features(*features,
-						      port->dev->features,
-						      mask);
+		netdev_increment_features(features, *features,
+					  port->dev->features, mask);
 	}
 	rcu_read_unlock();
 
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index e826435ab847..7e2678a9d769 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -5036,8 +5036,8 @@ static inline void netdev_get_wanted_features(struct net_device *dev,
 {
 	*wanted = (dev->features & ~dev->hw_features) | dev->wanted_features;
 }
-netdev_features_t netdev_increment_features(netdev_features_t all,
-	netdev_features_t one, netdev_features_t mask);
+void netdev_increment_features(netdev_features_t *ret, netdev_features_t all,
+			       netdev_features_t one, netdev_features_t mask);
 
 /* Allow TSO being used on stacked device :
  * Performing the GSO segmentation before last device
@@ -5046,7 +5046,7 @@ netdev_features_t netdev_increment_features(netdev_features_t all,
 static inline void netdev_add_tso_features(netdev_features_t *features,
 					   netdev_features_t mask)
 {
-	*features = netdev_increment_features(*features, NETIF_F_ALL_TSO, mask);
+	netdev_increment_features(features, *features, NETIF_F_ALL_TSO, mask);
 }
 
 int __netdev_update_features(struct net_device *dev);
diff --git a/net/bridge/br_if.c b/net/bridge/br_if.c
index 749971ab1088..914ca4b2d07c 100644
--- a/net/bridge/br_if.c
+++ b/net/bridge/br_if.c
@@ -544,8 +544,8 @@ void br_features_recompute(struct net_bridge *br, netdev_features_t *features)
 	*features &= ~NETIF_F_ONE_FOR_ALL;
 
 	list_for_each_entry(p, &br->port_list, list) {
-		*features = netdev_increment_features(*features,
-						      p->dev->features, mask);
+		netdev_increment_features(features, *features, p->dev->features,
+					  mask);
 	}
 	netdev_add_tso_features(features, mask);
 }
diff --git a/net/core/dev.c b/net/core/dev.c
index 58c46131126b..6663dd4ed7ff 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -11311,6 +11311,7 @@ static int dev_cpu_dead(unsigned int oldcpu)
 
 /**
  *	netdev_increment_features - increment feature set by one
+ *	@ret: result feature set
  *	@all: current feature set
  *	@one: new feature set
  *	@mask: mask feature set
@@ -11319,8 +11320,8 @@ static int dev_cpu_dead(unsigned int oldcpu)
  *	@one to the master device with current feature set @all.  Will not
  *	enable anything that is off in @mask. Returns the new feature set.
  */
-netdev_features_t netdev_increment_features(netdev_features_t all,
-	netdev_features_t one, netdev_features_t mask)
+void netdev_increment_features(netdev_features_t *ret, netdev_features_t all,
+			       netdev_features_t one, netdev_features_t mask)
 {
 	if (mask & NETIF_F_HW_CSUM)
 		mask |= NETIF_F_CSUM_MASK;
@@ -11333,7 +11334,7 @@ netdev_features_t netdev_increment_features(netdev_features_t all,
 	if (all & NETIF_F_HW_CSUM)
 		all &= ~(NETIF_F_CSUM_MASK & ~NETIF_F_HW_CSUM);
 
-	return all;
+	*ret = all;
 }
 EXPORT_SYMBOL(netdev_increment_features);
 
diff --git a/net/hsr/hsr_device.c b/net/hsr/hsr_device.c
index acaf48a1e136..16e0efd8b528 100644
--- a/net/hsr/hsr_device.c
+++ b/net/hsr/hsr_device.c
@@ -194,9 +194,8 @@ static netdev_features_t hsr_features_recompute(struct hsr_priv *hsr,
 	 */
 	features &= ~NETIF_F_ONE_FOR_ALL;
 	hsr_for_each_port(hsr, port)
-		features = netdev_increment_features(features,
-						     port->dev->features,
-						     mask);
+		netdev_increment_features(&features, features,
+					  port->dev->features, mask);
 
 	return features;
 }
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 017/167] net: convert the prototype of hsr_features_recompute
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (15 preceding siblings ...)
  2021-09-29 15:51 ` [RFCv2 net-next 016/167] net: convert the prototype of netdev_increment_features Jian Shen
@ 2021-09-29 15:51 ` Jian Shen
  2021-09-29 15:51 ` [RFCv2 net-next 018/167] net: mlx5: convert prototype of mlx5e_ipsec_feature_check, mlx5e_tunnel_features_check and mlx5e_fix_uplink_rep_features Jian Shen
                   ` (153 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:51 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

For the origin type for netdev_features_t would be changed to
be unsigned long * from u64, so changes the prototype of
hsr_features_recompute for adaption.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 net/hsr/hsr_device.c | 14 ++++++--------
 1 file changed, 6 insertions(+), 8 deletions(-)

diff --git a/net/hsr/hsr_device.c b/net/hsr/hsr_device.c
index 16e0efd8b528..7e6dd372711f 100644
--- a/net/hsr/hsr_device.c
+++ b/net/hsr/hsr_device.c
@@ -177,13 +177,13 @@ static int hsr_dev_close(struct net_device *dev)
 	return 0;
 }
 
-static netdev_features_t hsr_features_recompute(struct hsr_priv *hsr,
-						netdev_features_t features)
+static void hsr_features_recompute(struct hsr_priv *hsr,
+				   netdev_features_t *features)
 {
 	netdev_features_t mask;
 	struct hsr_port *port;
 
-	mask = features;
+	mask = *features;
 
 	/* Mask out all features that, if supported by one device, should be
 	 * enabled for all devices (see NETIF_F_ONE_FOR_ALL).
@@ -192,12 +192,10 @@ static netdev_features_t hsr_features_recompute(struct hsr_priv *hsr,
 	 * that were in features originally, and also is in NETIF_F_ONE_FOR_ALL,
 	 * may become enabled.
 	 */
-	features &= ~NETIF_F_ONE_FOR_ALL;
+	*features &= ~NETIF_F_ONE_FOR_ALL;
 	hsr_for_each_port(hsr, port)
-		netdev_increment_features(&features, features,
+		netdev_increment_features(features, *features,
 					  port->dev->features, mask);
-
-	return features;
 }
 
 static void hsr_fix_features(struct net_device *dev,
@@ -205,7 +203,7 @@ static void hsr_fix_features(struct net_device *dev,
 {
 	struct hsr_priv *hsr = netdev_priv(dev);
 
-	*features = hsr_features_recompute(hsr, *features);
+	hsr_features_recompute(hsr, features);
 }
 
 static netdev_tx_t hsr_dev_xmit(struct sk_buff *skb, struct net_device *dev)
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 018/167] net: mlx5: convert prototype of mlx5e_ipsec_feature_check, mlx5e_tunnel_features_check and mlx5e_fix_uplink_rep_features
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (16 preceding siblings ...)
  2021-09-29 15:51 ` [RFCv2 net-next 017/167] net: convert the prototype of hsr_features_recompute Jian Shen
@ 2021-09-29 15:51 ` Jian Shen
  2021-09-29 15:51 ` [RFCv2 net-next 019/167] net: sfc: convert the prototype of xxx_supported_features Jian Shen
                   ` (152 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:51 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

For the origin type for netdev_features_t would be changed to
be unsigned long * from u64, so changes the prototype of
mlx5e_tunnel_features_check, mlx5e_ipsec_feature_check
and mlx5e_fix_uplink_rep_features for adaption.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 .../mellanox/mlx5/core/en_accel/ipsec_rxtx.h  | 15 ++++----
 .../net/ethernet/mellanox/mlx5/core/en_main.c | 34 +++++++++----------
 2 files changed, 23 insertions(+), 26 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_rxtx.h b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_rxtx.h
index 5120a59361e6..62ab97cca915 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_rxtx.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_rxtx.h
@@ -93,8 +93,8 @@ static inline bool mlx5e_ipsec_eseg_meta(struct mlx5_wqe_eth_seg *eseg)
 void mlx5e_ipsec_tx_build_eseg(struct mlx5e_priv *priv, struct sk_buff *skb,
 			       struct mlx5_wqe_eth_seg *eseg);
 
-static inline netdev_features_t
-mlx5e_ipsec_feature_check(struct sk_buff *skb, netdev_features_t features)
+static inline void
+mlx5e_ipsec_feature_check(struct sk_buff *skb, netdev_features_t *features)
 {
 	struct xfrm_offload *xo = xfrm_offload(skb);
 	struct sec_path *sp = skb_sec_path(skb);
@@ -118,13 +118,12 @@ mlx5e_ipsec_feature_check(struct sk_buff *skb, netdev_features_t features)
 				goto out_disable;
 		}
 
-		return features;
-
+		return;
 	}
 
 	/* Disable CSUM and GSO for software IPsec */
 out_disable:
-	return features & ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
+	*features &= ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
 }
 
 #else
@@ -140,9 +139,9 @@ static inline bool mlx5e_ipsec_eseg_meta(struct mlx5_wqe_eth_seg *eseg)
 }
 
 static inline bool mlx5_ipsec_is_rx_flow(struct mlx5_cqe64 *cqe) { return false; }
-static inline netdev_features_t
-mlx5e_ipsec_feature_check(struct sk_buff *skb, netdev_features_t features)
-{ return features & ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK); }
+static inline void
+mlx5e_ipsec_feature_check(struct sk_buff *skb, netdev_features_t *features)
+{ *features &= ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK); }
 #endif /* CONFIG_MLX5_EN_IPSEC */
 
 #endif /* __MLX5E_IPSEC_RXTX_H__ */
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index 4092210fb079..07f88bc4be1c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -3379,22 +3379,20 @@ int mlx5e_set_features(struct net_device *netdev, netdev_features_t features)
 	return 0;
 }
 
-static netdev_features_t mlx5e_fix_uplink_rep_features(struct net_device *netdev,
-						       netdev_features_t features)
+static void mlx5e_fix_uplink_rep_features(struct net_device *netdev,
+					  netdev_features_t *features)
 {
-	features &= ~NETIF_F_HW_TLS_RX;
+	*features &= ~NETIF_F_HW_TLS_RX;
 	if (netdev->features & NETIF_F_HW_TLS_RX)
 		netdev_warn(netdev, "Disabling hw_tls_rx, not supported in switchdev mode\n");
 
-	features &= ~NETIF_F_HW_TLS_TX;
+	*features &= ~NETIF_F_HW_TLS_TX;
 	if (netdev->features & NETIF_F_HW_TLS_TX)
 		netdev_warn(netdev, "Disabling hw_tls_tx, not supported in switchdev mode\n");
 
-	features &= ~NETIF_F_NTUPLE;
+	*features &= ~NETIF_F_NTUPLE;
 	if (netdev->features & NETIF_F_NTUPLE)
 		netdev_warn(netdev, "Disabling ntuple, not supported in switchdev mode\n");
-
-	return features;
 }
 
 static void mlx5e_fix_features(struct net_device *netdev,
@@ -3429,7 +3427,7 @@ static void mlx5e_fix_features(struct net_device *netdev,
 	}
 
 	if (mlx5e_is_uplink_rep(priv))
-		features = mlx5e_fix_uplink_rep_features(netdev, features);
+		mlx5e_fix_uplink_rep_features(netdev, &features);
 
 	mutex_unlock(&priv->state_lock);
 }
@@ -3847,9 +3845,9 @@ static bool mlx5e_gre_tunnel_inner_proto_offload_supported(struct mlx5_core_dev
 	return false;
 }
 
-static netdev_features_t mlx5e_tunnel_features_check(struct mlx5e_priv *priv,
-						     struct sk_buff *skb,
-						     netdev_features_t features)
+static void mlx5e_tunnel_features_check(struct mlx5e_priv *priv,
+					struct sk_buff *skb,
+					netdev_features_t *features)
 {
 	unsigned int offset = 0;
 	struct udphdr *udph;
@@ -3870,12 +3868,12 @@ static netdev_features_t mlx5e_tunnel_features_check(struct mlx5e_priv *priv,
 	switch (proto) {
 	case IPPROTO_GRE:
 		if (mlx5e_gre_tunnel_inner_proto_offload_supported(priv->mdev, skb))
-			return features;
+			return;
 		break;
 	case IPPROTO_IPIP:
 	case IPPROTO_IPV6:
 		if (mlx5e_tunnel_proto_supported_tx(priv->mdev, IPPROTO_IPIP))
-			return features;
+			return;
 		break;
 	case IPPROTO_UDP:
 		udph = udp_hdr(skb);
@@ -3883,23 +3881,23 @@ static netdev_features_t mlx5e_tunnel_features_check(struct mlx5e_priv *priv,
 
 		/* Verify if UDP port is being offloaded by HW */
 		if (mlx5_vxlan_lookup_port(priv->mdev->vxlan, port))
-			return features;
+			return;
 
 #if IS_ENABLED(CONFIG_GENEVE)
 		/* Support Geneve offload for default UDP port */
 		if (port == GENEVE_UDP_PORT && mlx5_geneve_tx_allowed(priv->mdev))
-			return features;
+			return;
 #endif
 		break;
 #ifdef CONFIG_MLX5_EN_IPSEC
 	case IPPROTO_ESP:
-		return mlx5e_ipsec_feature_check(skb, features);
+		mlx5e_ipsec_feature_check(skb, *features);
 #endif
 	}
 
 out:
 	/* Disable CSUM and GSO if the udp dport is not offloaded by HW */
-	return features & ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
+	*features &= ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
 }
 
 void mlx5e_features_check(struct sk_buff *skb, struct net_device *netdev,
@@ -3913,7 +3911,7 @@ void mlx5e_features_check(struct sk_buff *skb, struct net_device *netdev,
 	/* Validate if the tunneled packet is being offloaded by HW */
 	if (skb->encapsulation &&
 	    (*features & NETIF_F_CSUM_MASK || *features & NETIF_F_GSO_MASK))
-		*features = mlx5e_tunnel_features_check(priv, skb, *features);
+		mlx5e_tunnel_features_check(priv, skb, features);
 }
 
 static void mlx5e_tx_timeout_work(struct work_struct *work)
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 019/167] net: sfc: convert the prototype of xxx_supported_features
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (17 preceding siblings ...)
  2021-09-29 15:51 ` [RFCv2 net-next 018/167] net: mlx5: convert prototype of mlx5e_ipsec_feature_check, mlx5e_tunnel_features_check and mlx5e_fix_uplink_rep_features Jian Shen
@ 2021-09-29 15:51 ` Jian Shen
  2021-09-30  0:28   ` Edward Cree
  2021-09-29 15:51 ` [RFCv2 net-next 020/167] net: qlogic: convert the prototype of qlcnic_process_flags Jian Shen
                   ` (151 subsequent siblings)
  170 siblings, 1 reply; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:51 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

For the origin type for netdev_features_t would be changed to
be unsigned long * from u64, so changes the prototype of
xxx_supported_features for adaption.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 drivers/net/ethernet/sfc/falcon/net_driver.h | 5 +++--
 drivers/net/ethernet/sfc/mcdi_filters.c      | 5 ++++-
 drivers/net/ethernet/sfc/net_driver.h        | 5 +++--
 3 files changed, 10 insertions(+), 5 deletions(-)

diff --git a/drivers/net/ethernet/sfc/falcon/net_driver.h b/drivers/net/ethernet/sfc/falcon/net_driver.h
index a381cf9ec4f3..6fabfe7f02f5 100644
--- a/drivers/net/ethernet/sfc/falcon/net_driver.h
+++ b/drivers/net/ethernet/sfc/falcon/net_driver.h
@@ -1298,11 +1298,12 @@ static inline struct ef4_rx_buffer *ef4_rx_buffer(struct ef4_rx_queue *rx_queue,
  * If a feature is fixed, it does not present in hw_features, but
  * always in features.
  */
-static inline netdev_features_t ef4_supported_features(const struct ef4_nic *efx)
+static inline void ef4_supported_features(const struct ef4_nic *efx,
+					  netdev_features_t *supported)
 {
 	const struct net_device *net_dev = efx->net_dev;
 
-	return net_dev->features | net_dev->hw_features;
+	*supported = net_dev->features | net_dev->hw_features;
 }
 
 /* Get the current TX queue insert index. */
diff --git a/drivers/net/ethernet/sfc/mcdi_filters.c b/drivers/net/ethernet/sfc/mcdi_filters.c
index 1523be77b9db..8e788c3a6f2a 100644
--- a/drivers/net/ethernet/sfc/mcdi_filters.c
+++ b/drivers/net/ethernet/sfc/mcdi_filters.c
@@ -1298,6 +1298,7 @@ int efx_mcdi_filter_table_probe(struct efx_nic *efx, bool multicast_chaining)
 {
 	struct net_device *net_dev = efx->net_dev;
 	struct efx_mcdi_filter_table *table;
+	netdev_features_t supported;
 	int rc;
 
 	if (!efx_rwsem_assert_write_locked(&efx->filter_sem))
@@ -1319,7 +1320,9 @@ int efx_mcdi_filter_table_probe(struct efx_nic *efx, bool multicast_chaining)
 		rc = efx_mcdi_filter_table_probe_matches(efx, table, true);
 	if (rc)
 		goto fail;
-	if ((efx_supported_features(efx) & NETIF_F_HW_VLAN_CTAG_FILTER) &&
+
+	efx_supported_features(efx, &supported);
+	if ((supported & NETIF_F_HW_VLAN_CTAG_FILTER) &&
 	    !(efx_mcdi_filter_match_supported(table, false,
 		(EFX_FILTER_MATCH_OUTER_VID | EFX_FILTER_MATCH_LOC_MAC)) &&
 	      efx_mcdi_filter_match_supported(table, false,
diff --git a/drivers/net/ethernet/sfc/net_driver.h b/drivers/net/ethernet/sfc/net_driver.h
index f6981810039d..bf962ca160e7 100644
--- a/drivers/net/ethernet/sfc/net_driver.h
+++ b/drivers/net/ethernet/sfc/net_driver.h
@@ -1681,11 +1681,12 @@ efx_channel_tx_old_fill_level(struct efx_channel *channel)
  * If a feature is fixed, it does not present in hw_features, but
  * always in features.
  */
-static inline netdev_features_t efx_supported_features(const struct efx_nic *efx)
+static inline void efx_supported_features(const struct efx_nic *efx,
+					  netdev_features_t *supported)
 {
 	const struct net_device *net_dev = efx->net_dev;
 
-	return net_dev->features | net_dev->hw_features;
+	*supported = net_dev->features | net_dev->hw_features;
 }
 
 /* Get the current TX queue insert index. */
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 020/167] net: qlogic: convert the prototype of qlcnic_process_flags
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (18 preceding siblings ...)
  2021-09-29 15:51 ` [RFCv2 net-next 019/167] net: sfc: convert the prototype of xxx_supported_features Jian Shen
@ 2021-09-29 15:51 ` Jian Shen
  2021-09-29 15:51 ` [RFCv2 net-next 021/167] net: realtek: convert the prototype of rtl8168evl_fix_tso Jian Shen
                   ` (150 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:51 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

For the origin type for netdev_features_t would be changed to
be unsigned long * from u64, so changes the prototype of
qlcnic_process_flags for adaption.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 .../net/ethernet/qlogic/qlcnic/qlcnic_hw.c    | 24 +++++++++----------
 1 file changed, 11 insertions(+), 13 deletions(-)

diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_hw.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_hw.c
index e6ed7f8413b4..2367923e3e3f 100644
--- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_hw.c
+++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_hw.c
@@ -1020,37 +1020,35 @@ int qlcnic_change_mtu(struct net_device *netdev, int mtu)
 	return rc;
 }
 
-static netdev_features_t qlcnic_process_flags(struct qlcnic_adapter *adapter,
-					      netdev_features_t features)
+static void qlcnic_process_flags(struct qlcnic_adapter *adapter,
+				 netdev_features_t *features)
 {
 	u32 offload_flags = adapter->offload_flags;
 
 	if (offload_flags & BIT_0) {
-		features |= NETIF_F_RXCSUM | NETIF_F_IP_CSUM |
-			    NETIF_F_IPV6_CSUM;
+		*features |= NETIF_F_RXCSUM | NETIF_F_IP_CSUM |
+			     NETIF_F_IPV6_CSUM;
 		adapter->rx_csum = 1;
 		if (QLCNIC_IS_TSO_CAPABLE(adapter)) {
 			if (!(offload_flags & BIT_1))
-				features &= ~NETIF_F_TSO;
+				*features &= ~NETIF_F_TSO;
 			else
-				features |= NETIF_F_TSO;
+				*features |= NETIF_F_TSO;
 
 			if (!(offload_flags & BIT_2))
-				features &= ~NETIF_F_TSO6;
+				*features &= ~NETIF_F_TSO6;
 			else
-				features |= NETIF_F_TSO6;
+				*features |= NETIF_F_TSO6;
 		}
 	} else {
-		features &= ~(NETIF_F_RXCSUM |
+		*features &= ~(NETIF_F_RXCSUM |
 			      NETIF_F_IP_CSUM |
 			      NETIF_F_IPV6_CSUM);
 
 		if (QLCNIC_IS_TSO_CAPABLE(adapter))
-			features &= ~(NETIF_F_TSO | NETIF_F_TSO6);
+			*features &= ~(NETIF_F_TSO | NETIF_F_TSO6);
 		adapter->rx_csum = 0;
 	}
-
-	return features;
 }
 
 void qlcnic_fix_features(struct net_device *netdev, netdev_features_t *features)
@@ -1061,7 +1059,7 @@ void qlcnic_fix_features(struct net_device *netdev, netdev_features_t *features)
 	if (qlcnic_82xx_check(adapter) &&
 	    (adapter->flags & QLCNIC_ESWITCH_ENABLED)) {
 		if (adapter->flags & QLCNIC_APP_CHANGED_FLAGS) {
-			*features = qlcnic_process_flags(adapter, *features);
+			qlcnic_process_flags(adapter, features);
 		} else {
 			changed = *features ^ netdev->features;
 			*features ^= changed & (NETIF_F_RXCSUM |
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 021/167] net: realtek: convert the prototype of rtl8168evl_fix_tso
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (19 preceding siblings ...)
  2021-09-29 15:51 ` [RFCv2 net-next 020/167] net: qlogic: convert the prototype of qlcnic_process_flags Jian Shen
@ 2021-09-29 15:51 ` Jian Shen
  2021-09-29 15:51 ` [RFCv2 net-next 022/167] ethtool: convert the prototype of ethtool_get_feature_mask Jian Shen
                   ` (149 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:51 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

For the origin type for netdev_features_t would be changed to
be unsigned long * from u64, so changes the prototype of
rtl8168evl_fix_tso for adaption.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 drivers/net/ethernet/realtek/r8169_main.c | 13 +++++--------
 1 file changed, 5 insertions(+), 8 deletions(-)

diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c
index 8affae2b55a6..c876246cf8c1 100644
--- a/drivers/net/ethernet/realtek/r8169_main.c
+++ b/drivers/net/ethernet/realtek/r8169_main.c
@@ -4326,23 +4326,20 @@ static unsigned int rtl_last_frag_len(struct sk_buff *skb)
 }
 
 /* Workaround for hw issues with TSO on RTL8168evl */
-static netdev_features_t rtl8168evl_fix_tso(struct sk_buff *skb,
-					    netdev_features_t features)
+static void rtl8168evl_fix_tso(struct sk_buff *skb, netdev_features_t *features)
 {
 	/* IPv4 header has options field */
 	if (vlan_get_protocol(skb) == htons(ETH_P_IP) &&
 	    ip_hdrlen(skb) > sizeof(struct iphdr))
-		features &= ~NETIF_F_ALL_TSO;
+		*features &= ~NETIF_F_ALL_TSO;
 
 	/* IPv4 TCP header has options field */
 	else if (skb_shinfo(skb)->gso_type & SKB_GSO_TCPV4 &&
 		 tcp_hdrlen(skb) > sizeof(struct tcphdr))
-		features &= ~NETIF_F_ALL_TSO;
+		*features &= ~NETIF_F_ALL_TSO;
 
 	else if (rtl_last_frag_len(skb) <= 6)
-		features &= ~NETIF_F_ALL_TSO;
-
-	return features;
+		*features &= ~NETIF_F_ALL_TSO;
 }
 
 static void rtl8169_features_check(struct sk_buff *skb, struct net_device *dev,
@@ -4353,7 +4350,7 @@ static void rtl8169_features_check(struct sk_buff *skb, struct net_device *dev,
 
 	if (skb_is_gso(skb)) {
 		if (tp->mac_version == RTL_GIGA_MAC_VER_34)
-			*features = rtl8168evl_fix_tso(skb, *features);
+			rtl8168evl_fix_tso(skb, features);
 
 		if (transport_offset > GTTCPHO_MAX &&
 		    rtl_chip_supports_csum_v2(tp))
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 022/167] ethtool: convert the prototype of ethtool_get_feature_mask
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (20 preceding siblings ...)
  2021-09-29 15:51 ` [RFCv2 net-next 021/167] net: realtek: convert the prototype of rtl8168evl_fix_tso Jian Shen
@ 2021-09-29 15:51 ` Jian Shen
  2021-09-29 15:51 ` [RFCv2 net-next 023/167] net: add netdev feature helpers Jian Shen
                   ` (148 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:51 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

For the origin type for netdev_features_t would be changed to
be unsigned long * from u64, so changes the prototype of
ethtool_get_feature_mask for adaption.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 net/ethtool/ioctl.c | 34 ++++++++++++++++++++--------------
 1 file changed, 20 insertions(+), 14 deletions(-)

diff --git a/net/ethtool/ioctl.c b/net/ethtool/ioctl.c
index bf6e8c2f9bf7..661b75dee9fd 100644
--- a/net/ethtool/ioctl.c
+++ b/net/ethtool/ioctl.c
@@ -195,30 +195,35 @@ static void __ethtool_get_strings(struct net_device *dev,
 		ops->get_strings(dev, stringset, data);
 }
 
-static netdev_features_t ethtool_get_feature_mask(u32 eth_cmd)
+static void ethtool_get_feature_mask(u32 eth_cmd, netdev_features_t *mask)
 {
 	/* feature masks of legacy discrete ethtool ops */
 
 	switch (eth_cmd) {
 	case ETHTOOL_GTXCSUM:
 	case ETHTOOL_STXCSUM:
-		return NETIF_F_CSUM_MASK | NETIF_F_FCOE_CRC |
-		       NETIF_F_SCTP_CRC;
+		*mask = NETIF_F_CSUM_MASK | NETIF_F_FCOE_CRC | NETIF_F_SCTP_CRC;
+		break;
 	case ETHTOOL_GRXCSUM:
 	case ETHTOOL_SRXCSUM:
-		return NETIF_F_RXCSUM;
+		*mask = NETIF_F_RXCSUM;
+		break;
 	case ETHTOOL_GSG:
 	case ETHTOOL_SSG:
-		return NETIF_F_SG | NETIF_F_FRAGLIST;
+		*mask = NETIF_F_SG | NETIF_F_FRAGLIST;
+		break;
 	case ETHTOOL_GTSO:
 	case ETHTOOL_STSO:
-		return NETIF_F_ALL_TSO;
+		*mask = NETIF_F_ALL_TSO;
+		break;
 	case ETHTOOL_GGSO:
 	case ETHTOOL_SGSO:
-		return NETIF_F_GSO;
+		*mask = NETIF_F_GSO;
+		break;
 	case ETHTOOL_GGRO:
 	case ETHTOOL_SGRO:
-		return NETIF_F_GRO;
+		*mask = NETIF_F_GRO;
+		break;
 	default:
 		BUG();
 	}
@@ -227,11 +232,12 @@ static netdev_features_t ethtool_get_feature_mask(u32 eth_cmd)
 static int ethtool_get_one_feature(struct net_device *dev,
 	char __user *useraddr, u32 ethcmd)
 {
-	netdev_features_t mask = ethtool_get_feature_mask(ethcmd);
-	struct ethtool_value edata = {
-		.cmd = ethcmd,
-		.data = !!(dev->features & mask),
-	};
+	struct ethtool_value edata;
+	netdev_features_t mask;
+
+	ethtool_get_feature_mask(ethcmd, &mask);
+	edata.cmd = ethcmd;
+	edata.data = !!(dev->features & mask);
 
 	if (copy_to_user(useraddr, &edata, sizeof(edata)))
 		return -EFAULT;
@@ -247,7 +253,7 @@ static int ethtool_set_one_feature(struct net_device *dev,
 	if (copy_from_user(&edata, useraddr, sizeof(edata)))
 		return -EFAULT;
 
-	mask = ethtool_get_feature_mask(ethcmd);
+	ethtool_get_feature_mask(ethcmd, &mask);
 	mask &= dev->hw_features;
 	if (!mask)
 		return -EOPNOTSUPP;
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 023/167] net: add netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (21 preceding siblings ...)
  2021-09-29 15:51 ` [RFCv2 net-next 022/167] ethtool: convert the prototype of ethtool_get_feature_mask Jian Shen
@ 2021-09-29 15:51 ` Jian Shen
  2021-09-29 15:51 ` [RFCv2 net-next 024/167] net: core: use " Jian Shen
                   ` (147 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:51 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

For the prototype of netdev_features_t will be changed from
u64 to unsigned long *, so it's necessary to add a set of
helpers to do the logic operation.

This is a temporary patch, used to compatible with the
current prototype of netdev_features_t. The parameters styles
are not consistent for among these helpers, in order to minimize
the modification when change the netdev_features_t to bitmap.

It also introduce a new macro NETDEV_FEATURE_DWORDS, used for
scenario define the features as an array, so it can be
initialized when define structure.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 include/linux/netdev_features.h | 135 ++++++++++++++++++++++++++++++++
 1 file changed, 135 insertions(+)

diff --git a/include/linux/netdev_features.h b/include/linux/netdev_features.h
index 2c6b9e416225..b3fa05c88eef 100644
--- a/include/linux/netdev_features.h
+++ b/include/linux/netdev_features.h
@@ -101,6 +101,8 @@ enum {
 	/**/NETDEV_FEATURE_COUNT
 };
 
+#define NETDEV_FEATURE_DWORDS	DIV_ROUND_UP(NETDEV_FEATURE_COUNT, 64)
+
 /* copy'n'paste compression ;) */
 #define __NETIF_F_BIT(bit)	((netdev_features_t)1 << (bit))
 #define __NETIF_F(name)		__NETIF_F_BIT(NETIF_F_##name##_BIT)
@@ -261,4 +263,137 @@ static inline int find_next_netdev_feature(u64 feature, unsigned long start)
 				 NETIF_F_GSO_UDP_TUNNEL |		\
 				 NETIF_F_GSO_UDP_TUNNEL_CSUM)
 
+static inline void netdev_feature_zero(netdev_features_t *dst)
+{
+	*dst = 0;
+}
+
+static inline void netdev_feature_fill(netdev_features_t *dst)
+{
+	*dst = ~0;
+}
+
+static inline void netdev_feature_copy(netdev_features_t *dst,
+				       const netdev_features_t src)
+{
+	*dst = src;
+}
+
+static inline void netdev_feature_and(netdev_features_t *dst,
+				      const netdev_features_t a,
+				      const netdev_features_t b)
+{
+	*dst = a & b;
+}
+
+static inline void netdev_feature_or(netdev_features_t *dst,
+				     const netdev_features_t a,
+				     const netdev_features_t b)
+{
+	*dst = a | b;
+}
+
+static inline void netdev_feature_xor(netdev_features_t *dst,
+				      const netdev_features_t a,
+				      const netdev_features_t b)
+{
+	*dst = a ^ b;
+}
+
+static inline bool netdev_feature_empty(netdev_features_t src)
+{
+	return src == 0;
+}
+
+static inline bool netdev_feature_equal(const netdev_features_t src1,
+					const netdev_features_t src2)
+{
+	return src1 == src2;
+}
+
+static inline int netdev_feature_andnot(netdev_features_t *dst,
+					const netdev_features_t src1,
+					const netdev_features_t src2)
+{
+	*dst = src1 & ~src2;
+	return 0;
+}
+
+static inline void netdev_feature_set_bit(int nr, netdev_features_t *addr)
+{
+	*addr |= __NETIF_F_BIT(nr);
+}
+
+static inline void netdev_feature_clear_bit(int nr, netdev_features_t *addr)
+{
+	*addr &= ~(__NETIF_F_BIT(nr));
+}
+
+static inline void netdev_feature_mod_bit(int nr, netdev_features_t *addr,
+					  int set)
+{
+	if (set)
+		netdev_feature_set_bit(nr, addr);
+	else
+		netdev_feature_clear_bit(nr, addr);
+}
+
+static inline void netdev_feature_change_bit(int nr, netdev_features_t *addr)
+{
+	*addr ^= __NETIF_F_BIT(nr);
+}
+
+static inline int netdev_feature_test_bit(int nr, const netdev_features_t addr)
+{
+	return (addr & __NETIF_F_BIT(nr)) > 0;
+}
+
+static inline void netdev_feature_set_bit_array(const int *array,
+						int array_size,
+						netdev_features_t *addr)
+{
+	int i;
+
+	for (i = 0; i < array_size; i++)
+		netdev_feature_set_bit(array[i], addr);
+}
+
+/* only be used for the first 64 bits features */
+static inline void netdev_feature_set_bits(u64 bits, netdev_features_t *addr)
+{
+	*addr |= bits;
+}
+
+/* only be used for the first 64 bits features */
+static inline void netdev_feature_clear_bits(u64 bits, netdev_features_t *addr)
+{
+	*addr &= ~bits;
+}
+
+/* only be used for the first 64 bits features */
+static inline bool netdev_feature_test_bits(u64 bits,
+					    const netdev_features_t addr)
+{
+	return (addr & bits) > 0;
+}
+
+/* only be used for the first 64 bits features */
+static inline void netdev_feature_and_bits(u64 bits,
+					   netdev_features_t *addr)
+{
+	*addr &= bits;
+}
+
+static inline int netdev_feature_intersects(const netdev_features_t src1,
+					    const netdev_features_t src2)
+{
+	return (src1 & src2) > 0;
+}
+
+static inline int netdev_feature_subset(const netdev_features_t src1,
+					const netdev_features_t src2)
+{
+	return (src1 & src2) == src2;
+}
+
 #endif	/* _LINUX_NETDEV_FEATURES_H */
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 024/167] net: core: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (22 preceding siblings ...)
  2021-09-29 15:51 ` [RFCv2 net-next 023/167] net: add netdev feature helpers Jian Shen
@ 2021-09-29 15:51 ` Jian Shen
  2021-09-29 15:51 ` [RFCv2 net-next 025/167] skbuff: " Jian Shen
                   ` (146 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:51 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 include/linux/netdevice.h |  78 ++++++----
 net/core/dev.c            | 315 +++++++++++++++++++++++---------------
 2 files changed, 236 insertions(+), 157 deletions(-)

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 7e2678a9d769..56e642a72997 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -2271,7 +2271,8 @@ struct net_device {
 
 static inline bool netif_elide_gro(const struct net_device *dev)
 {
-	if (!(dev->features & NETIF_F_GRO) || dev->xdp_prog)
+	if (!netdev_feature_test_bit(NETIF_F_GRO_BIT, dev->features) ||
+	    dev->xdp_prog)
 		return true;
 	return false;
 }
@@ -4515,25 +4516,28 @@ static inline void netif_tx_unlock_bh(struct net_device *dev)
 	local_bh_enable();
 }
 
-#define HARD_TX_LOCK(dev, txq, cpu) {			\
-	if ((dev->features & NETIF_F_LLTX) == 0) {	\
-		__netif_tx_lock(txq, cpu);		\
-	} else {					\
-		__netif_tx_acquire(txq);		\
-	}						\
+#define HARD_TX_LOCK(dev, txq, cpu) {				\
+	if (!netdev_feature_test_bit(NETIF_F_LLTX_BIT,		\
+				     dev->features)) {		\
+		__netif_tx_lock(txq, cpu);			\
+	} else {						\
+		__netif_tx_acquire(txq);			\
+	}							\
 }
 
-#define HARD_TX_TRYLOCK(dev, txq)			\
-	(((dev->features & NETIF_F_LLTX) == 0) ?	\
-		__netif_tx_trylock(txq) :		\
+#define HARD_TX_TRYLOCK(dev, txq)				\
+	(!netdev_feature_test_bit(NETIF_F_LLTX_BIT,		\
+				  dev->features) ?		\
+		__netif_tx_trylock(txq) :			\
 		__netif_tx_acquire(txq))
 
-#define HARD_TX_UNLOCK(dev, txq) {			\
-	if ((dev->features & NETIF_F_LLTX) == 0) {	\
-		__netif_tx_unlock(txq);			\
-	} else {					\
-		__netif_tx_release(txq);		\
-	}						\
+#define HARD_TX_UNLOCK(dev, txq) {				\
+	if (!netdev_feature_test_bit(NETIF_F_LLTX_BIT,		\
+				     dev->features)) {		\
+		__netif_tx_unlock(txq);				\
+	} else {						\
+		__netif_tx_release(txq);			\
+	}							\
 }
 
 static inline void netif_tx_disable(struct net_device *dev)
@@ -4943,20 +4947,20 @@ static inline bool can_checksum_protocol(netdev_features_t features,
 					 __be16 protocol)
 {
 	if (protocol == htons(ETH_P_FCOE))
-		return !!(features & NETIF_F_FCOE_CRC);
+		return netdev_feature_test_bit(NETIF_F_FCOE_CRC_BIT, features);
 
 	/* Assume this is an IP checksum (not SCTP CRC) */
 
-	if (features & NETIF_F_HW_CSUM) {
+	if (netdev_feature_test_bit(NETIF_F_HW_CSUM_BIT, features)) {
 		/* Can checksum everything */
 		return true;
 	}
 
 	switch (protocol) {
 	case htons(ETH_P_IP):
-		return !!(features & NETIF_F_IP_CSUM);
+		return netdev_feature_test_bit(NETIF_F_IP_CSUM_BIT, features);
 	case htons(ETH_P_IPV6):
-		return !!(features & NETIF_F_IPV6_CSUM);
+		return netdev_feature_test_bit(NETIF_F_IPV6_CSUM_BIT, features);
 	default:
 		return false;
 	}
@@ -5021,21 +5025,26 @@ static inline void netdev_intersect_features(netdev_features_t *ret,
 					     netdev_features_t f1,
 					     netdev_features_t f2)
 {
-	if ((f1 ^ f2) & NETIF_F_HW_CSUM) {
-		if (f1 & NETIF_F_HW_CSUM)
-			f1 |= (NETIF_F_IP_CSUM|NETIF_F_IPV6_CSUM);
+	if (netdev_feature_test_bit(NETIF_F_HW_CSUM_BIT, f1) !=
+	    netdev_feature_test_bit(NETIF_F_HW_CSUM_BIT, f2)) {
+		if (netdev_feature_test_bit(NETIF_F_HW_CSUM_BIT, f1))
+			netdev_feature_set_bits(NETIF_F_IP_CSUM |
+						NETIF_F_IPV6_CSUM, &f1);
 		else
-			f2 |= (NETIF_F_IP_CSUM|NETIF_F_IPV6_CSUM);
+			netdev_feature_set_bits(NETIF_F_IP_CSUM |
+						NETIF_F_IPV6_CSUM, &f2);
 	}
 
-	*ret = f1 & f2;
+	netdev_feature_and(ret, f1, f2);
 }
 
 static inline void netdev_get_wanted_features(struct net_device *dev,
 					      netdev_features_t *wanted)
 {
-	*wanted = (dev->features & ~dev->hw_features) | dev->wanted_features;
+	netdev_feature_andnot(wanted, dev->features, dev->hw_features);
+	netdev_feature_or(wanted, *wanted, dev->wanted_features);
 }
+
 void netdev_increment_features(netdev_features_t *ret, netdev_features_t all,
 			       netdev_features_t one, netdev_features_t mask);
 
@@ -5046,7 +5055,11 @@ void netdev_increment_features(netdev_features_t *ret, netdev_features_t all,
 static inline void netdev_add_tso_features(netdev_features_t *features,
 					   netdev_features_t mask)
 {
-	netdev_increment_features(features, *features, NETIF_F_ALL_TSO, mask);
+	netdev_features_t one;
+
+	netdev_feature_zero(&one);
+	netdev_feature_set_bits(NETIF_F_ALL_TSO, &one);
+	netdev_increment_features(features, *features, one, mask);
 }
 
 int __netdev_update_features(struct net_device *dev);
@@ -5062,7 +5075,7 @@ void netif_skb_features(struct sk_buff *skb, netdev_features_t *features);
 
 static inline bool net_gso_ok(netdev_features_t features, int gso_type)
 {
-	netdev_features_t feature = (netdev_features_t)gso_type << NETIF_F_GSO_SHIFT;
+	netdev_features_t feature;
 
 	/* check flags correspondence */
 	BUILD_BUG_ON(SKB_GSO_TCPV4   != (NETIF_F_TSO >> NETIF_F_GSO_SHIFT));
@@ -5085,13 +5098,18 @@ static inline bool net_gso_ok(netdev_features_t features, int gso_type)
 	BUILD_BUG_ON(SKB_GSO_UDP_L4 != (NETIF_F_GSO_UDP_L4 >> NETIF_F_GSO_SHIFT));
 	BUILD_BUG_ON(SKB_GSO_FRAGLIST != (NETIF_F_GSO_FRAGLIST >> NETIF_F_GSO_SHIFT));
 
-	return (features & feature) == feature;
+	netdev_feature_zero(&feature);
+	netdev_feature_set_bits((u64)gso_type << NETIF_F_GSO_SHIFT, &feature);
+
+	return netdev_feature_subset(features, feature);
+
 }
 
 static inline bool skb_gso_ok(struct sk_buff *skb, netdev_features_t features)
 {
 	return net_gso_ok(features, skb_shinfo(skb)->gso_type) &&
-	       (!skb_has_frag_list(skb) || (features & NETIF_F_FRAGLIST));
+	       (!skb_has_frag_list(skb) ||
+	       netdev_feature_test_bit(NETIF_F_FRAGLIST_BIT, features));
 }
 
 static inline bool netif_needs_gso(struct sk_buff *skb,
diff --git a/net/core/dev.c b/net/core/dev.c
index 6663dd4ed7ff..17a93ec301b6 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -1636,10 +1636,10 @@ void dev_disable_lro(struct net_device *dev)
 	struct net_device *lower_dev;
 	struct list_head *iter;
 
-	dev->wanted_features &= ~NETIF_F_LRO;
+	netdev_feature_clear_bit(NETIF_F_LRO_BIT, &dev->wanted_features);
 	netdev_update_features(dev);
 
-	if (unlikely(dev->features & NETIF_F_LRO))
+	if (unlikely(netdev_feature_test_bit(NETIF_F_LRO_BIT, dev->features)))
 		netdev_WARN(dev, "failed to disable LRO!\n");
 
 	netdev_for_each_lower_dev(dev, lower_dev, iter)
@@ -1657,10 +1657,12 @@ EXPORT_SYMBOL(dev_disable_lro);
  */
 static void dev_disable_gro_hw(struct net_device *dev)
 {
-	dev->wanted_features &= ~NETIF_F_GRO_HW;
+	netdev_feature_clear_bit(NETIF_F_GRO_HW_BIT, &dev->wanted_features);
 	netdev_update_features(dev);
 
-	if (unlikely(dev->features & NETIF_F_GRO_HW))
+	if (unlikely(netdev_feature_test_bit(NETIF_F_GRO_HW_BIT,
+					     dev->features)))
+
 		netdev_WARN(dev, "failed to disable GRO_HW!\n");
 }
 
@@ -3379,13 +3381,19 @@ struct sk_buff *__skb_gso_segment(struct sk_buff *skb,
 	 * support segmentation on this frame without needing additional
 	 * work.
 	 */
-	if (features & NETIF_F_GSO_PARTIAL) {
-		netdev_features_t partial_features = NETIF_F_GSO_ROBUST;
+	if (netdev_feature_test_bit(NETIF_F_GSO_PARTIAL_BIT, features)) {
+		netdev_features_t partial_features;
 		struct net_device *dev = skb->dev;
 
-		partial_features |= dev->features & dev->gso_partial_features;
-		if (!skb_gso_ok(skb, features | partial_features))
-			features &= ~NETIF_F_GSO_PARTIAL;
+		netdev_feature_and(&partial_features, dev->features,
+				   dev->gso_partial_features);
+		netdev_feature_set_bit(NETIF_F_GSO_ROBUST_BIT,
+				       &partial_features);
+		netdev_feature_or(&partial_features, partial_features,
+				  features);
+		if (!skb_gso_ok(skb, partial_features))
+			netdev_feature_clear_bit(NETIF_F_GSO_PARTIAL_BIT,
+						 &features);
 	}
 
 	BUILD_BUG_ON(SKB_GSO_CB_OFFSET +
@@ -3428,7 +3436,7 @@ static int illegal_highdma(struct net_device *dev, struct sk_buff *skb)
 #ifdef CONFIG_HIGHMEM
 	int i;
 
-	if (!(dev->features & NETIF_F_HIGHDMA)) {
+	if (!netdev_feature_test_bit(NETIF_F_HIGHDMA_BIT, dev->features)) {
 		for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
 			skb_frag_t *frag = &skb_shinfo(skb)->frags[i];
 
@@ -3448,7 +3456,8 @@ static void net_mpls_features(struct sk_buff *skb, netdev_features_t *features,
 			      __be16 type)
 {
 	if (eth_p_mpls(type))
-		*features &= skb->dev->mpls_features;
+		netdev_feature_and(features, *features,
+				   skb->dev->mpls_features);
 }
 #else
 static void net_mpls_features(struct sk_buff *skb, netdev_features_t *features,
@@ -3466,10 +3475,11 @@ static void harmonize_features(struct sk_buff *skb, netdev_features_t *features)
 
 	if (skb->ip_summed != CHECKSUM_NONE &&
 	    !can_checksum_protocol(*features, type)) {
-		*features &= ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
+		netdev_feature_clear_bits(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK,
+					  features);
 	}
 	if (illegal_highdma(skb->dev, skb))
-		*features &= ~NETIF_F_SG;
+		netdev_feature_clear_bit(NETIF_F_SG_BIT, features);
 }
 
 void passthru_features_check(struct sk_buff *skb, struct net_device *dev,
@@ -3492,13 +3502,13 @@ static void gso_features_check(const struct sk_buff *skb,
 	u16 gso_segs = skb_shinfo(skb)->gso_segs;
 
 	if (gso_segs > dev->gso_max_segs) {
-		*features &= ~NETIF_F_GSO_MASK;
+		netdev_feature_clear_bits(NETIF_F_GSO_MASK, features);
 		return;
 	}
 
 	if (!skb_shinfo(skb)->gso_type) {
 		skb_warn_bad_offload(skb);
-		*features &= ~NETIF_F_GSO_MASK;
+		netdev_feature_clear_bits(NETIF_F_GSO_MASK, features);
 		return;
 	}
 
@@ -3509,7 +3519,8 @@ static void gso_features_check(const struct sk_buff *skb,
 	 * segmented the frame.
 	 */
 	if (!(skb_shinfo(skb)->gso_type & SKB_GSO_PARTIAL))
-		*features &= ~dev->gso_partial_features;
+		netdev_feature_andnot(features, *features,
+				      dev->gso_partial_features);
 
 	/* Make sure to clear the IPv4 ID mangling feature if the
 	 * IPv4 header has the potential to be fragmented.
@@ -3519,7 +3530,8 @@ static void gso_features_check(const struct sk_buff *skb,
 				    inner_ip_hdr(skb) : ip_hdr(skb);
 
 		if (!(iph->frag_off & htons(IP_DF)))
-			*features &= ~NETIF_F_TSO_MANGLEID;
+			netdev_feature_clear_bit(NETIF_F_TSO_MANGLEID_BIT,
+						 features);
 	}
 }
 
@@ -3527,7 +3539,7 @@ void netif_skb_features(struct sk_buff *skb, netdev_features_t *features)
 {
 	struct net_device *dev = skb->dev;
 
-	*features = dev->features;
+	netdev_feature_copy(features, dev->features);
 
 	if (skb_is_gso(skb))
 		gso_features_check(skb, dev, features);
@@ -3537,13 +3549,16 @@ void netif_skb_features(struct sk_buff *skb, netdev_features_t *features)
 	 * features for the netdev
 	 */
 	if (skb->encapsulation)
-		*features &= dev->hw_enc_features;
+		netdev_feature_and(features, *features, dev->hw_enc_features);
+
+	if (skb_vlan_tagged(skb)) {
+		netdev_features_t tmp;
 
-	if (skb_vlan_tagged(skb))
-		netdev_intersect_features(features, *features,
-					  dev->vlan_features |
-					  NETIF_F_HW_VLAN_CTAG_TX |
-					  NETIF_F_HW_VLAN_STAG_TX);
+		netdev_feature_copy(&tmp, dev->vlan_features);
+		netdev_feature_set_bits(NETIF_F_HW_VLAN_CTAG_TX_BIT |
+					NETIF_F_HW_VLAN_STAG_TX_BIT, &tmp);
+		netdev_intersect_features(features, *features, tmp);
+	}
 
 	if (dev->netdev_ops->ndo_features_check)
 		dev->netdev_ops->ndo_features_check(skb, dev, features);
@@ -3613,13 +3628,14 @@ int skb_csum_hwoffload_help(struct sk_buff *skb,
 			    const netdev_features_t features)
 {
 	if (unlikely(skb_csum_is_sctp(skb)))
-		return !!(features & NETIF_F_SCTP_CRC) ? 0 :
-			skb_crc32c_csum_help(skb);
+		return netdev_feature_test_bit(NETIF_F_SCTP_CRC_BIT, features) ?
+			0 : skb_crc32c_csum_help(skb);
 
-	if (features & NETIF_F_HW_CSUM)
+	if (netdev_feature_test_bit(NETIF_F_HW_CSUM_BIT, features))
 		return 0;
 
-	if (features & (NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM)) {
+	if (netdev_feature_test_bits(NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM,
+				     features)) {
 		switch (skb->csum_offset) {
 		case offsetof(struct tcphdr, check):
 		case offsetof(struct udphdr, check):
@@ -4351,7 +4367,8 @@ set_rps_cpu(struct net_device *dev, struct sk_buff *skb,
 
 		/* Should we steer this flow to a different hardware queue? */
 		if (!skb_rx_queue_recorded(skb) || !dev->rx_cpu_rmap ||
-		    !(dev->features & NETIF_F_NTUPLE))
+		    !netdev_feature_test_bit(NETIF_F_NTUPLE_BIT,
+					     dev->features))
 			goto out;
 		rxq_index = cpu_rmap_lookup_index(dev->rx_cpu_rmap, next_cpu);
 		if (rxq_index == skb_get_rx_queue(skb))
@@ -9786,17 +9803,19 @@ static void netdev_sync_upper_features(struct net_device *lower,
 				       struct net_device *upper,
 				       netdev_features_t *features)
 {
-	netdev_features_t upper_disables = NETIF_F_UPPER_DISABLES;
-	netdev_features_t feature;
+	netdev_features_t upper_disables;
 	int feature_bit;
 
+	netdev_feature_zero(&upper_disables);
+	netdev_feature_set_bits(NETIF_F_UPPER_DISABLES, &upper_disables);
+
 	for_each_netdev_feature(upper_disables, feature_bit) {
-		feature = __NETIF_F_BIT(feature_bit);
-		if (!(upper->wanted_features & feature) &&
-		    (*features & feature)) {
-			netdev_dbg(lower, "Dropping feature %pNF, upper dev %s has it off.\n",
-				   &feature, upper->name);
-			*features &= ~feature;
+		if (!netdev_feature_test_bit(feature_bit,
+					     upper->wanted_features) &&
+		    netdev_feature_test_bit(feature_bit, *features)) {
+			netdev_dbg(lower, "Dropping feature bit %d, upper dev %s has it off.\n",
+				   feature_bit, upper->name);
+			netdev_feature_clear_bit(feature_bit, features);
 		}
 	}
 }
@@ -9804,21 +9823,25 @@ static void netdev_sync_upper_features(struct net_device *lower,
 static void netdev_sync_lower_features(struct net_device *upper,
 	struct net_device *lower, netdev_features_t features)
 {
-	netdev_features_t upper_disables = NETIF_F_UPPER_DISABLES;
-	netdev_features_t feature;
+	netdev_features_t upper_disables;
 	int feature_bit;
 
+	netdev_feature_zero(&upper_disables);
+	netdev_feature_set_bits(NETIF_F_UPPER_DISABLES, &upper_disables);
+
 	for_each_netdev_feature(upper_disables, feature_bit) {
-		feature = __NETIF_F_BIT(feature_bit);
-		if (!(features & feature) && (lower->features & feature)) {
-			netdev_dbg(upper, "Disabling feature %pNF on lower dev %s.\n",
-				   &feature, lower->name);
-			lower->wanted_features &= ~feature;
+		if (!netdev_feature_test_bit(feature_bit, features) &&
+		    netdev_feature_test_bit(feature_bit, lower->features)) {
+			netdev_dbg(upper, "Disabling feature bit %d on lower dev %s.\n",
+				   feature_bit, lower->name);
+			netdev_feature_clear_bit(feature_bit,
+						 &lower->wanted_features);
 			__netdev_update_features(lower);
 
-			if (unlikely(lower->features & feature))
-				netdev_WARN(upper, "failed to disable %pNF on %s!\n",
-					    &feature, lower->name);
+			if (unlikely(netdev_feature_test_bit(feature_bit,
+							     lower->features)))
+				netdev_WARN(upper, "failed to disable feature bit %d on %s!\n",
+					    feature_bit, lower->name);
 			else
 				netdev_features_change(lower);
 		}
@@ -9829,92 +9852,102 @@ static void netdev_fix_features(struct net_device *dev,
 				netdev_features_t *features)
 {
 	/* Fix illegal checksum combinations */
-	if ((*features & NETIF_F_HW_CSUM) &&
-	    (*features & (NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM))) {
+	if (netdev_feature_test_bit(NETIF_F_HW_CSUM_BIT, *features) &&
+	    netdev_feature_test_bits(NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM,
+				     *features)) {
 		netdev_warn(dev, "mixed HW and IP checksum settings.\n");
-		*features &= ~(NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM);
+		netdev_feature_clear_bits(NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM,
+					  features);
 	}
 
 	/* TSO requires that SG is present as well. */
-	if ((*features & NETIF_F_ALL_TSO) && !(*features & NETIF_F_SG)) {
+	if (netdev_feature_test_bits(NETIF_F_ALL_TSO, *features) &&
+	    !netdev_feature_test_bit(NETIF_F_SG_BIT, *features)) {
 		netdev_dbg(dev, "Dropping TSO features since no SG feature.\n");
-		*features &= ~NETIF_F_ALL_TSO;
+		netdev_feature_clear_bits(NETIF_F_ALL_TSO, features);
 	}
 
-	if ((*features & NETIF_F_TSO) && !(*features & NETIF_F_HW_CSUM) &&
-	    !(*features & NETIF_F_IP_CSUM)) {
+	if (netdev_feature_test_bit(NETIF_F_TSO_BIT, *features) &&
+	    !netdev_feature_test_bit(NETIF_F_HW_CSUM_BIT, *features) &&
+	    !netdev_feature_test_bit(NETIF_F_IP_CSUM_BIT, *features)) {
 		netdev_dbg(dev, "Dropping TSO features since no CSUM feature.\n");
-		*features &= ~NETIF_F_TSO;
-		*features &= ~NETIF_F_TSO_ECN;
+		netdev_feature_clear_bit(NETIF_F_TSO_BIT, features);
+		netdev_feature_clear_bit(NETIF_F_TSO_ECN_BIT, features);
 	}
 
-	if ((*features & NETIF_F_TSO6) && !(*features & NETIF_F_HW_CSUM) &&
-	    !(*features & NETIF_F_IPV6_CSUM)) {
+	if (netdev_feature_test_bit(NETIF_F_TSO6_BIT, *features) &&
+	    !netdev_feature_test_bit(NETIF_F_HW_CSUM_BIT, *features) &&
+	    !netdev_feature_test_bit(NETIF_F_IPV6_CSUM_BIT, *features)) {
 		netdev_dbg(dev, "Dropping TSO6 features since no CSUM feature.\n");
-		*features &= ~NETIF_F_TSO6;
+		netdev_feature_clear_bit(NETIF_F_TSO6_BIT, features);
 	}
 
 	/* TSO with IPv4 ID mangling requires IPv4 TSO be enabled */
-	if ((*features & NETIF_F_TSO_MANGLEID) && !(*features & NETIF_F_TSO))
-		*features &= ~NETIF_F_TSO_MANGLEID;
+	if (netdev_feature_test_bit(NETIF_F_TSO_MANGLEID_BIT, *features) &&
+	    !netdev_feature_test_bit(NETIF_F_TSO_BIT, *features))
+		netdev_feature_clear_bit(NETIF_F_TSO_MANGLEID_BIT, features);
 
 	/* TSO ECN requires that TSO is present as well. */
-	if ((*features & NETIF_F_ALL_TSO) == NETIF_F_TSO_ECN)
-		*features &= ~NETIF_F_TSO_ECN;
+	if (netdev_feature_test_bit(NETIF_F_TSO_ECN_BIT, *features) &&
+	    !netdev_feature_test_bits(NETIF_F_ALL_TSO & ~NETIF_F_TSO_ECN,
+				      *features))
+		netdev_feature_clear_bit(NETIF_F_TSO_ECN_BIT, features);
 
 	/* Software GSO depends on SG. */
-	if ((*features & NETIF_F_GSO) && !(*features & NETIF_F_SG)) {
+	if (netdev_feature_test_bit(NETIF_F_GSO_BIT, *features) &&
+	    !netdev_feature_test_bit(NETIF_F_SG_BIT, *features)) {
 		netdev_dbg(dev, "Dropping NETIF_F_GSO since no SG feature.\n");
-		*features &= ~NETIF_F_GSO;
+		netdev_feature_clear_bit(NETIF_F_GSO_BIT, features);
 	}
 
 	/* GSO partial features require GSO partial be set */
-	if ((*features & dev->gso_partial_features) &&
-	    !(*features & NETIF_F_GSO_PARTIAL)) {
+	if (netdev_feature_intersects(*features, dev->gso_partial_features) &&
+	    !netdev_feature_test_bit(NETIF_F_GSO_PARTIAL_BIT, *features)) {
 		netdev_dbg(dev,
 			   "Dropping partially supported GSO features since no GSO partial.\n");
-		*features &= ~dev->gso_partial_features;
+		netdev_feature_andnot(features, *features,
+				      dev->gso_partial_features);
 	}
 
-	if (!(*features & NETIF_F_RXCSUM)) {
+	if (!netdev_feature_test_bit(NETIF_F_RXCSUM_BIT, *features)) {
 		/* NETIF_F_GRO_HW implies doing RXCSUM since every packet
 		 * successfully merged by hardware must also have the
 		 * checksum verified by hardware.  If the user does not
 		 * want to enable RXCSUM, logically, we should disable GRO_HW.
 		 */
-		if (*features & NETIF_F_GRO_HW) {
+		if (netdev_feature_test_bit(NETIF_F_GRO_HW_BIT, *features)) {
 			netdev_dbg(dev, "Dropping NETIF_F_GRO_HW since no RXCSUM feature.\n");
-			*features &= ~NETIF_F_GRO_HW;
+			netdev_feature_clear_bit(NETIF_F_GRO_HW_BIT, features);
 		}
 	}
 
 	/* LRO/HW-GRO features cannot be combined with RX-FCS */
-	if (*features & NETIF_F_RXFCS) {
-		if (*features & NETIF_F_LRO) {
+	if (netdev_feature_test_bit(NETIF_F_RXFCS_BIT, *features)) {
+		if (netdev_feature_test_bit(NETIF_F_LRO_BIT, *features)) {
 			netdev_dbg(dev, "Dropping LRO feature since RX-FCS is requested.\n");
-			*features &= ~NETIF_F_LRO;
+			netdev_feature_clear_bit(NETIF_F_LRO_BIT, features);
 		}
 
-		if (*features & NETIF_F_GRO_HW) {
+		if (netdev_feature_test_bit(NETIF_F_GRO_HW_BIT, *features)) {
 			netdev_dbg(dev, "Dropping HW-GRO feature since RX-FCS is requested.\n");
-			*features &= ~NETIF_F_GRO_HW;
+			netdev_feature_clear_bit(NETIF_F_GRO_HW_BIT, features);
 		}
 	}
 
-	if (*features & NETIF_F_HW_TLS_TX) {
-		bool ip_csum = (*features & (NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM)) ==
-			(NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM);
-		bool hw_csum = *features & NETIF_F_HW_CSUM;
-
-		if (!ip_csum && !hw_csum) {
+	if (netdev_feature_test_bit(NETIF_F_HW_TLS_TX_BIT, *features)) {
+		if ((!netdev_feature_test_bit(NETIF_F_IP_CSUM_BIT, *features) ||
+		     !netdev_feature_test_bit(NETIF_F_IPV6_CSUM_BIT, *features)) &&
+		    !netdev_feature_test_bit(NETIF_F_HW_CSUM_BIT, *features)) {
 			netdev_dbg(dev, "Dropping TLS TX HW offload feature since no CSUM feature.\n");
-			*features &= ~NETIF_F_HW_TLS_TX;
+			netdev_feature_clear_bit(NETIF_F_HW_TLS_TX_BIT,
+						 features);
 		}
 	}
 
-	if ((*features & NETIF_F_HW_TLS_RX) && !(*features & NETIF_F_RXCSUM)) {
+	if (netdev_feature_test_bit(NETIF_F_HW_TLS_RX_BIT, *features) &&
+	    !netdev_feature_test_bit(NETIF_F_RXCSUM_BIT, *features)) {
 		netdev_dbg(dev, "Dropping TLS RX HW offload feature since no RXCSUM feature.\n");
-		*features &= ~NETIF_F_HW_TLS_RX;
+		netdev_feature_clear_bit(NETIF_F_HW_TLS_RX_BIT, features);
 	}
 }
 
@@ -9939,7 +9972,7 @@ int __netdev_update_features(struct net_device *dev)
 	netdev_for_each_upper_dev_rcu(dev, upper, iter)
 		netdev_sync_upper_features(dev, upper, &features);
 
-	if (dev->features == features)
+	if (netdev_feature_equal(dev->features, features))
 		goto sync_lower;
 
 	netdev_dbg(dev, "Features changed: %pNF -> %pNF\n",
@@ -9968,9 +10001,11 @@ int __netdev_update_features(struct net_device *dev)
 		netdev_sync_lower_features(dev, lower, features);
 
 	if (!err) {
-		netdev_features_t diff = features ^ dev->features;
+		netdev_features_t diff;
+
+		netdev_feature_xor(&diff, features, dev->features);
 
-		if (diff & NETIF_F_RX_UDP_TUNNEL_PORT) {
+		if (netdev_feature_test_bit(NETIF_F_RX_UDP_TUNNEL_PORT_BIT, diff)) {
 			/* udp_tunnel_{get,drop}_rx_info both need
 			 * NETIF_F_RX_UDP_TUNNEL_PORT enabled on the
 			 * device, or they won't do anything.
@@ -9978,33 +10013,36 @@ int __netdev_update_features(struct net_device *dev)
 			 * *before* calling udp_tunnel_get_rx_info,
 			 * but *after* calling udp_tunnel_drop_rx_info.
 			 */
-			if (features & NETIF_F_RX_UDP_TUNNEL_PORT) {
-				dev->features = features;
+			if (netdev_feature_test_bit(NETIF_F_RX_UDP_TUNNEL_PORT_BIT,
+						    features)) {
+				netdev_feature_copy(&dev->features, features);
 				udp_tunnel_get_rx_info(dev);
 			} else {
 				udp_tunnel_drop_rx_info(dev);
 			}
 		}
 
-		if (diff & NETIF_F_HW_VLAN_CTAG_FILTER) {
-			if (features & NETIF_F_HW_VLAN_CTAG_FILTER) {
-				dev->features = features;
+		if (netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_FILTER_BIT, diff)) {
+			if (netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_FILTER_BIT,
+						    features)) {
+				netdev_feature_copy(&dev->features, features);
 				err |= vlan_get_rx_ctag_filter_info(dev);
 			} else {
 				vlan_drop_rx_ctag_filter_info(dev);
 			}
 		}
 
-		if (diff & NETIF_F_HW_VLAN_STAG_FILTER) {
-			if (features & NETIF_F_HW_VLAN_STAG_FILTER) {
-				dev->features = features;
+		if (netdev_feature_test_bit(NETIF_F_HW_VLAN_STAG_FILTER_BIT, diff)) {
+			if (netdev_feature_test_bit(NETIF_F_HW_VLAN_STAG_FILTER_BIT,
+						    features)) {
+				netdev_feature_copy(&dev->features, features);
 				err |= vlan_get_rx_stag_filter_info(dev);
 			} else {
 				vlan_drop_rx_stag_filter_info(dev);
 			}
 		}
 
-		dev->features = features;
+		netdev_feature_copy(&dev->features, features);
 	}
 
 	return err < 0 ? 0 : 1;
@@ -10230,8 +10268,10 @@ int register_netdevice(struct net_device *dev)
 		}
 	}
 
-	if (((dev->hw_features | dev->features) &
-	     NETIF_F_HW_VLAN_CTAG_FILTER) &&
+	if ((netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_FILTER_BIT,
+				     dev->hw_features) ||
+	     netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_FILTER_BIT,
+				     dev->features)) &&
 	    (!dev->netdev_ops->ndo_vlan_rx_add_vid ||
 	     !dev->netdev_ops->ndo_vlan_rx_kill_vid)) {
 		netdev_WARN(dev, "Buggy VLAN acceleration in driver!\n");
@@ -10248,44 +10288,54 @@ int register_netdevice(struct net_device *dev)
 	/* Transfer changeable features to wanted_features and enable
 	 * software offloads (GSO and GRO).
 	 */
-	dev->hw_features |= (NETIF_F_SOFT_FEATURES | NETIF_F_SOFT_FEATURES_OFF);
-	dev->features |= NETIF_F_SOFT_FEATURES;
+	netdev_feature_set_bits(NETIF_F_SOFT_FEATURES | NETIF_F_SOFT_FEATURES_OFF,
+				&dev->hw_features);
+	netdev_feature_set_bits(NETIF_F_SOFT_FEATURES, &dev->features);
 
 	if (dev->udp_tunnel_nic_info) {
-		dev->features |= NETIF_F_RX_UDP_TUNNEL_PORT;
-		dev->hw_features |= NETIF_F_RX_UDP_TUNNEL_PORT;
+		netdev_feature_set_bit(NETIF_F_RX_UDP_TUNNEL_PORT_BIT,
+				       &dev->features);
+		netdev_feature_set_bit(NETIF_F_RX_UDP_TUNNEL_PORT_BIT,
+				       &dev->hw_features);
 	}
 
-	dev->wanted_features = dev->features & dev->hw_features;
+	netdev_feature_and(&dev->wanted_features, dev->features,
+			   dev->hw_features);
 
 	if (!(dev->flags & IFF_LOOPBACK))
-		dev->hw_features |= NETIF_F_NOCACHE_COPY;
+		netdev_feature_set_bit(NETIF_F_NOCACHE_COPY_BIT,
+				       &dev->hw_features);
 
 	/* If IPv4 TCP segmentation offload is supported we should also
 	 * allow the device to enable segmenting the frame with the option
 	 * of ignoring a static IP ID value.  This doesn't enable the
 	 * feature itself but allows the user to enable it later.
 	 */
-	if (dev->hw_features & NETIF_F_TSO)
-		dev->hw_features |= NETIF_F_TSO_MANGLEID;
-	if (dev->vlan_features & NETIF_F_TSO)
-		dev->vlan_features |= NETIF_F_TSO_MANGLEID;
-	if (dev->mpls_features & NETIF_F_TSO)
-		dev->mpls_features |= NETIF_F_TSO_MANGLEID;
-	if (dev->hw_enc_features & NETIF_F_TSO)
-		dev->hw_enc_features |= NETIF_F_TSO_MANGLEID;
+	if (netdev_feature_test_bit(NETIF_F_TSO_BIT, dev->hw_features))
+		netdev_feature_set_bit(NETIF_F_TSO_MANGLEID_BIT,
+				       &dev->hw_features);
+	if (netdev_feature_test_bit(NETIF_F_TSO_BIT, dev->vlan_features))
+		netdev_feature_set_bit(NETIF_F_TSO_MANGLEID_BIT,
+				       &dev->vlan_features);
+	if (netdev_feature_test_bit(NETIF_F_TSO_BIT, dev->mpls_features))
+		netdev_feature_set_bit(NETIF_F_TSO_MANGLEID_BIT,
+				       &dev->mpls_features);
+	if (netdev_feature_test_bit(NETIF_F_TSO_BIT, dev->hw_enc_features))
+		netdev_feature_set_bit(NETIF_F_TSO_MANGLEID_BIT,
+				       &dev->hw_enc_features);
 
 	/* Make NETIF_F_HIGHDMA inheritable to VLAN devices.
 	 */
-	dev->vlan_features |= NETIF_F_HIGHDMA;
+	netdev_feature_set_bit(NETIF_F_HIGHDMA_BIT, &dev->vlan_features);
 
 	/* Make NETIF_F_SG inheritable to tunnel devices.
 	 */
-	dev->hw_enc_features |= NETIF_F_SG | NETIF_F_GSO_PARTIAL;
+	netdev_feature_set_bits(NETIF_F_SG | NETIF_F_GSO_PARTIAL,
+				&dev->hw_enc_features);
 
 	/* Make NETIF_F_SG inheritable to MPLS.
 	 */
-	dev->mpls_features |= NETIF_F_SG;
+	netdev_feature_set_bit(NETIF_F_SG_BIT, &dev->mpls_features);
 
 	ret = call_netdevice_notifiers(NETDEV_POST_INIT, dev);
 	ret = notifier_to_errno(ret);
@@ -11126,7 +11176,7 @@ int __dev_change_net_namespace(struct net_device *dev, struct net *net,
 
 	/* Don't allow namespace local devices to be moved. */
 	err = -EINVAL;
-	if (dev->features & NETIF_F_NETNS_LOCAL)
+	if (netdev_feature_test_bit(NETIF_F_NETNS_LOCAL_BIT, dev->features))
 		goto out;
 
 	/* Ensure the device has been registrered */
@@ -11323,18 +11373,28 @@ static int dev_cpu_dead(unsigned int oldcpu)
 void netdev_increment_features(netdev_features_t *ret, netdev_features_t all,
 			       netdev_features_t one, netdev_features_t mask)
 {
-	if (mask & NETIF_F_HW_CSUM)
-		mask |= NETIF_F_CSUM_MASK;
-	mask |= NETIF_F_VLAN_CHALLENGED;
+	netdev_features_t tmp;
 
-	all |= one & (NETIF_F_ONE_FOR_ALL | NETIF_F_CSUM_MASK) & mask;
-	all &= one | ~NETIF_F_ALL_FOR_ALL;
+	if (netdev_feature_test_bit(NETIF_F_HW_CSUM_BIT, mask))
+		netdev_feature_set_bits(NETIF_F_CSUM_MASK, &mask);
+	netdev_feature_set_bit(NETIF_F_VLAN_CHALLENGED_BIT, &mask);
 
-	/* If one device supports hw checksumming, set for all. */
-	if (all & NETIF_F_HW_CSUM)
-		all &= ~(NETIF_F_CSUM_MASK & ~NETIF_F_HW_CSUM);
+	netdev_feature_copy(ret, all);
+	netdev_feature_zero(&tmp);
+	netdev_feature_set_bits(NETIF_F_ONE_FOR_ALL | NETIF_F_CSUM_MASK, &tmp);
+	netdev_feature_and(&tmp, tmp, one);
+	netdev_feature_and(&tmp, tmp, mask);
+	netdev_feature_or(ret, *ret, tmp);
 
-	*ret = all;
+	netdev_feature_fill(&tmp);
+	netdev_feature_clear_bits(NETIF_F_ALL_FOR_ALL, &tmp);
+	netdev_feature_or(&tmp, tmp, one);
+	netdev_feature_and(ret, *ret, tmp);
+
+	/* If one device supports hw checksumming, set for all. */
+	if (netdev_feature_test_bit(NETIF_F_HW_CSUM_BIT, *ret))
+		netdev_feature_clear_bits(NETIF_F_CSUM_MASK & ~NETIF_F_HW_CSUM,
+					  ret);
 }
 EXPORT_SYMBOL(netdev_increment_features);
 
@@ -11487,7 +11547,8 @@ static void __net_exit default_device_exit(struct net *net)
 		char fb_name[IFNAMSIZ];
 
 		/* Ignore unmoveable devices (i.e. loopback) */
-		if (dev->features & NETIF_F_NETNS_LOCAL)
+		if (netdev_feature_test_bit(NETIF_F_NETNS_LOCAL_BIT,
+					    dev->features))
 			continue;
 
 		/* Leave virtual devices for the generic cleanup */
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 025/167] skbuff: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (23 preceding siblings ...)
  2021-09-29 15:51 ` [RFCv2 net-next 024/167] net: core: use " Jian Shen
@ 2021-09-29 15:51 ` Jian Shen
  2021-09-29 15:51 ` [RFCv2 net-next 026/167] net: vlan: " Jian Shen
                   ` (145 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:51 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 include/linux/skbuff.h |  6 ++++--
 net/core/skbuff.c      | 10 ++++++----
 2 files changed, 10 insertions(+), 6 deletions(-)

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 841e2f0f5240..5dbb04a5f29f 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -3730,8 +3730,10 @@ static inline bool skb_needs_linearize(struct sk_buff *skb,
 				       netdev_features_t features)
 {
 	return skb_is_nonlinear(skb) &&
-	       ((skb_has_frag_list(skb) && !(features & NETIF_F_FRAGLIST)) ||
-		(skb_shinfo(skb)->nr_frags && !(features & NETIF_F_SG)));
+	       ((skb_has_frag_list(skb) &&
+		 !netdev_feature_test_bit(NETIF_F_FRAGLIST_BIT, features)) ||
+		(skb_shinfo(skb)->nr_frags &&
+		!netdev_feature_test_bit(NETIF_F_SG_BIT, features)));
 }
 
 static inline void skb_copy_from_linear_data(const struct sk_buff *skb,
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 74601bbc56ac..e229c1c0ce41 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -3992,7 +3992,7 @@ struct sk_buff *skb_segment(struct sk_buff *head_skb,
 		 * skbs; we do so by disabling SG.
 		 */
 		if (mss != GSO_BY_FRAGS && mss != skb_headlen(head_skb))
-			features &= ~NETIF_F_SG;
+			netdev_feature_clear_bit(NETIF_F_SG_BIT, &features);
 	}
 
 	__skb_push(head_skb, doffset);
@@ -4000,11 +4000,12 @@ struct sk_buff *skb_segment(struct sk_buff *head_skb,
 	if (unlikely(!proto))
 		return ERR_PTR(-EINVAL);
 
-	sg = !!(features & NETIF_F_SG);
+	sg = !!netdev_feature_test_bit(NETIF_F_SG_BIT, features);
 	csum = !!can_checksum_protocol(features, proto);
 
 	if (sg && csum && (mss != GSO_BY_FRAGS))  {
-		if (!(features & NETIF_F_GSO_PARTIAL)) {
+		if (!netdev_feature_test_bit(NETIF_F_GSO_PARTIAL_BIT,
+					     features)) {
 			struct sk_buff *iter;
 			unsigned int frag_len;
 
@@ -4261,7 +4262,8 @@ struct sk_buff *skb_segment(struct sk_buff *head_skb,
 		unsigned short gso_size = skb_shinfo(head_skb)->gso_size;
 
 		/* Update type to add partial and then remove dodgy if set */
-		type |= (features & NETIF_F_GSO_PARTIAL) / NETIF_F_GSO_PARTIAL * SKB_GSO_PARTIAL;
+		type |= netdev_feature_test_bit(NETIF_F_GSO_PARTIAL_BIT,
+						features) * SKB_GSO_PARTIAL;
 		type &= ~SKB_GSO_DODGY;
 
 		/* Update GSO info and prepare to start updating headers on
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 026/167] net: vlan: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (24 preceding siblings ...)
  2021-09-29 15:51 ` [RFCv2 net-next 025/167] skbuff: " Jian Shen
@ 2021-09-29 15:51 ` Jian Shen
  2021-09-29 15:51 ` [RFCv2 net-next 027/167] bridge: " Jian Shen
                   ` (144 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:51 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 include/linux/if_vlan.h | 16 ++++++++------
 net/8021q/vlan.c        | 11 ++++++----
 net/8021q/vlan.h        | 24 ++++++++++++---------
 net/8021q/vlan_core.c   |  6 ++++--
 net/8021q/vlan_dev.c    | 47 ++++++++++++++++++++++++-----------------
 5 files changed, 63 insertions(+), 41 deletions(-)

diff --git a/include/linux/if_vlan.h b/include/linux/if_vlan.h
index 2337538ef015..6565a62db842 100644
--- a/include/linux/if_vlan.h
+++ b/include/linux/if_vlan.h
@@ -313,9 +313,11 @@ static inline bool eth_type_vlan(__be16 ethertype)
 static inline bool vlan_hw_offload_capable(netdev_features_t features,
 					   __be16 proto)
 {
-	if (proto == htons(ETH_P_8021Q) && features & NETIF_F_HW_VLAN_CTAG_TX)
+	if (proto == htons(ETH_P_8021Q) &&
+	    netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_TX_BIT, features))
 		return true;
-	if (proto == htons(ETH_P_8021AD) && features & NETIF_F_HW_VLAN_STAG_TX)
+	if (proto == htons(ETH_P_8021AD) &&
+	    netdev_feature_test_bit(NETIF_F_HW_VLAN_STAG_TX_BIT, features))
 		return true;
 	return false;
 }
@@ -563,7 +565,8 @@ static inline int __vlan_hwaccel_get_tag(const struct sk_buff *skb,
  */
 static inline int vlan_get_tag(const struct sk_buff *skb, u16 *vlan_tci)
 {
-	if (skb->dev->features & NETIF_F_HW_VLAN_CTAG_TX) {
+	if (netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_TX_BIT,
+				    skb->dev->features)) {
 		return __vlan_hwaccel_get_tag(skb, vlan_tci);
 	} else {
 		return __vlan_get_tag(skb, vlan_tci);
@@ -736,9 +739,10 @@ static inline void vlan_features_check(struct sk_buff *skb,
 		 * sure that only devices supporting NETIF_F_HW_CSUM will
 		 * have checksum offloading support.
 		 */
-		*features &= NETIF_F_SG | NETIF_F_HIGHDMA | NETIF_F_HW_CSUM |
-			     NETIF_F_FRAGLIST | NETIF_F_HW_VLAN_CTAG_TX |
-			     NETIF_F_HW_VLAN_STAG_TX;
+		netdev_feature_and_bits(NETIF_F_SG | NETIF_F_HIGHDMA |
+					NETIF_F_HW_CSUM | NETIF_F_FRAGLIST |
+					NETIF_F_HW_VLAN_CTAG_TX |
+					NETIF_F_HW_VLAN_STAG_TX, features);
 	}
 }
 
diff --git a/net/8021q/vlan.c b/net/8021q/vlan.c
index 55275ef9a31a..8a50deff286a 100644
--- a/net/8021q/vlan.c
+++ b/net/8021q/vlan.c
@@ -134,7 +134,8 @@ int vlan_check_real_dev(struct net_device *real_dev,
 {
 	const char *name = real_dev->name;
 
-	if (real_dev->features & NETIF_F_VLAN_CHALLENGED) {
+	if (netdev_feature_test_bit(NETIF_F_VLAN_CHALLENGED_BIT,
+				    real_dev->features)) {
 		pr_info("VLANs not supported on %s\n", name);
 		NL_SET_ERR_MSG_MOD(extack, "VLANs not supported on device");
 		return -EOPNOTSUPP;
@@ -339,7 +340,7 @@ static void vlan_transfer_features(struct net_device *dev,
 
 	vlandev->priv_flags &= ~IFF_XMIT_DST_RELEASE;
 	vlandev->priv_flags |= (vlan->real_dev->priv_flags & IFF_XMIT_DST_RELEASE);
-	vlandev->hw_enc_features = vlan_tnl_features(vlan->real_dev);
+	vlan_tnl_features(vlan->real_dev, &vlandev->hw_enc_features);
 
 	netdev_update_features(vlandev);
 }
@@ -386,13 +387,15 @@ static int vlan_device_event(struct notifier_block *unused, unsigned long event,
 	}
 
 	if ((event == NETDEV_UP) &&
-	    (dev->features & NETIF_F_HW_VLAN_CTAG_FILTER)) {
+	    netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_FILTER_BIT,
+				    dev->features)) {
 		pr_info("adding VLAN 0 to HW filter on device %s\n",
 			dev->name);
 		vlan_vid_add(dev, htons(ETH_P_8021Q), 0);
 	}
 	if (event == NETDEV_DOWN &&
-	    (dev->features & NETIF_F_HW_VLAN_CTAG_FILTER))
+	    netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_FILTER_BIT,
+				    dev->features))
 		vlan_vid_del(dev, htons(ETH_P_8021Q), 0);
 
 	vlan_info = rtnl_dereference(dev->vlan_info);
diff --git a/net/8021q/vlan.h b/net/8021q/vlan.h
index 1a705a4ef7fa..9ed1d1c8f547 100644
--- a/net/8021q/vlan.h
+++ b/net/8021q/vlan.h
@@ -103,17 +103,21 @@ static inline struct net_device *vlan_find_dev(struct net_device *real_dev,
 	return NULL;
 }
 
-static inline netdev_features_t vlan_tnl_features(struct net_device *real_dev)
+static inline void vlan_tnl_features(struct net_device *real_dev,
+				     netdev_features_t *tnl)
 {
-	netdev_features_t ret;
-
-	ret = real_dev->hw_enc_features &
-	      (NETIF_F_CSUM_MASK | NETIF_F_GSO_SOFTWARE |
-	       NETIF_F_GSO_ENCAP_ALL);
-
-	if ((ret & NETIF_F_GSO_ENCAP_ALL) && (ret & NETIF_F_CSUM_MASK))
-		return (ret & ~NETIF_F_CSUM_MASK) | NETIF_F_HW_CSUM;
-	return 0;
+	netdev_feature_zero(tnl);
+	netdev_feature_set_bits(NETIF_F_CSUM_MASK | NETIF_F_GSO_SOFTWARE |
+				NETIF_F_GSO_ENCAP_ALL, tnl);
+	netdev_feature_and(tnl, *tnl, real_dev->hw_enc_features);
+
+	if (netdev_feature_test_bits(NETIF_F_GSO_ENCAP_ALL, *tnl) &&
+	    netdev_feature_test_bits(NETIF_F_CSUM_MASK, *tnl)) {
+		netdev_feature_clear_bits(NETIF_F_CSUM_MASK, tnl);
+		netdev_feature_set_bit(NETIF_F_HW_CSUM_BIT, tnl);
+		return;
+	}
+	netdev_feature_zero(tnl);
 }
 
 #define vlan_group_for_each_dev(grp, i, dev) \
diff --git a/net/8021q/vlan_core.c b/net/8021q/vlan_core.c
index 59bc13b5f14f..757b2f18978f 100644
--- a/net/8021q/vlan_core.c
+++ b/net/8021q/vlan_core.c
@@ -169,10 +169,12 @@ struct vlan_vid_info {
 static bool vlan_hw_filter_capable(const struct net_device *dev, __be16 proto)
 {
 	if (proto == htons(ETH_P_8021Q) &&
-	    dev->features & NETIF_F_HW_VLAN_CTAG_FILTER)
+	    netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_FILTER_BIT,
+				    dev->features))
 		return true;
 	if (proto == htons(ETH_P_8021AD) &&
-	    dev->features & NETIF_F_HW_VLAN_STAG_FILTER)
+	    netdev_feature_test_bit(NETIF_F_HW_VLAN_STAG_FILTER_BIT,
+				    dev->features))
 		return true;
 	return false;
 }
diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c
index 9f90a587d4e5..a010612cf27b 100644
--- a/net/8021q/vlan_dev.c
+++ b/net/8021q/vlan_dev.c
@@ -566,21 +566,24 @@ static int vlan_dev_init(struct net_device *dev)
 	if (vlan->flags & VLAN_FLAG_BRIDGE_BINDING)
 		dev->state |= (1 << __LINK_STATE_NOCARRIER);
 
-	dev->hw_features = NETIF_F_HW_CSUM | NETIF_F_SG |
-			   NETIF_F_FRAGLIST | NETIF_F_GSO_SOFTWARE |
-			   NETIF_F_GSO_ENCAP_ALL |
-			   NETIF_F_HIGHDMA | NETIF_F_SCTP_CRC |
-			   NETIF_F_ALL_FCOE;
-
-	dev->features |= dev->hw_features | NETIF_F_LLTX;
+	netdev_feature_zero(&dev->hw_features);
+	netdev_feature_set_bits(NETIF_F_HW_CSUM | NETIF_F_SG |
+				NETIF_F_FRAGLIST | NETIF_F_GSO_SOFTWARE |
+				NETIF_F_GSO_ENCAP_ALL |
+				NETIF_F_HIGHDMA | NETIF_F_SCTP_CRC |
+				NETIF_F_ALL_FCOE, &dev->hw_features);
+
+	netdev_feature_or(&dev->features, dev->features, dev->hw_features);
+	netdev_feature_set_bit(NETIF_F_LLTX_BIT, &dev->features);
 	dev->gso_max_size = real_dev->gso_max_size;
 	dev->gso_max_segs = real_dev->gso_max_segs;
-	if (dev->features & NETIF_F_VLAN_FEATURES)
+	if (netdev_feature_test_bits(NETIF_F_VLAN_FEATURES, dev->features))
 		netdev_warn(real_dev, "VLAN features are set incorrectly.  Q-in-Q configurations may not work correctly.\n");
 
-	dev->vlan_features = real_dev->vlan_features & ~NETIF_F_ALL_FCOE;
-	dev->hw_enc_features = vlan_tnl_features(real_dev);
-	dev->mpls_features = real_dev->mpls_features;
+	netdev_feature_copy(&dev->vlan_features, real_dev->vlan_features);
+	netdev_feature_clear_bits(NETIF_F_ALL_FCOE, &dev->vlan_features);
+	vlan_tnl_features(real_dev, &dev->hw_enc_features);
+	netdev_feature_copy(&dev->mpls_features, real_dev->mpls_features);
 
 	/* ipv6 shared card related stuff */
 	dev->dev_id = real_dev->dev_id;
@@ -637,21 +640,27 @@ static void vlan_dev_fix_features(struct net_device *dev,
 				  netdev_features_t *features)
 {
 	struct net_device *real_dev = vlan_dev_priv(dev)->real_dev;
-	netdev_features_t old_features = *features;
 	netdev_features_t lower_features;
+	netdev_features_t old_features;
+	netdev_features_t tmp;
 
-	netdev_intersect_features(&lower_features,
-				  (real_dev->vlan_features | NETIF_F_RXCSUM),
-				  real_dev->features);
+	netdev_feature_copy(&tmp, real_dev->vlan_features);
+	netdev_feature_set_bit(NETIF_F_RXCSUM_BIT, &tmp);
+	netdev_feature_copy(&old_features, *features);
+	netdev_intersect_features(&lower_features, tmp, real_dev->features);
 
 	/* Add HW_CSUM setting to preserve user ability to control
 	 * checksum offload on the vlan device.
 	 */
-	if (lower_features & (NETIF_F_IP_CSUM|NETIF_F_IPV6_CSUM))
-		lower_features |= NETIF_F_HW_CSUM;
+	if (netdev_feature_test_bits(NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM,
+				     lower_features))
+		netdev_feature_set_bit(NETIF_F_HW_CSUM_BIT, &lower_features);
 	netdev_intersect_features(features, *features, lower_features);
-	*features |= old_features & (NETIF_F_SOFT_FEATURES | NETIF_F_GSO_SOFTWARE);
-	*features |= NETIF_F_LLTX;
+
+	netdev_feature_and_bits(NETIF_F_SOFT_FEATURES | NETIF_F_GSO_SOFTWARE,
+				&old_features);
+	netdev_feature_or(features, *features, old_features);
+	netdev_feature_set_bit(NETIF_F_LLTX_BIT, features);
 }
 
 static int vlan_ethtool_get_link_ksettings(struct net_device *dev,
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 027/167] bridge: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (25 preceding siblings ...)
  2021-09-29 15:51 ` [RFCv2 net-next 026/167] net: vlan: " Jian Shen
@ 2021-09-29 15:51 ` Jian Shen
  2021-09-29 15:51 ` [RFCv2 net-next 028/167] ipvlan: " Jian Shen
                   ` (143 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:51 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 net/bridge/br_device.c | 16 +++++++++++-----
 net/bridge/br_if.c     |  4 ++--
 2 files changed, 13 insertions(+), 7 deletions(-)

diff --git a/net/bridge/br_device.c b/net/bridge/br_device.c
index fc508b9cbaa9..c9894b9944ab 100644
--- a/net/bridge/br_device.c
+++ b/net/bridge/br_device.c
@@ -490,11 +490,17 @@ void br_dev_setup(struct net_device *dev)
 	SET_NETDEV_DEVTYPE(dev, &br_type);
 	dev->priv_flags = IFF_EBRIDGE | IFF_NO_QUEUE;
 
-	dev->features = COMMON_FEATURES | NETIF_F_LLTX | NETIF_F_NETNS_LOCAL |
-			NETIF_F_HW_VLAN_CTAG_TX | NETIF_F_HW_VLAN_STAG_TX;
-	dev->hw_features = COMMON_FEATURES | NETIF_F_HW_VLAN_CTAG_TX |
-			   NETIF_F_HW_VLAN_STAG_TX;
-	dev->vlan_features = COMMON_FEATURES;
+	netdev_feature_zero(&dev->features);
+	netdev_feature_set_bits(COMMON_FEATURES | NETIF_F_LLTX |
+				NETIF_F_NETNS_LOCAL |
+				NETIF_F_HW_VLAN_CTAG_TX |
+				NETIF_F_HW_VLAN_STAG_TX,
+				&dev->features);
+	netdev_feature_zero(&dev->hw_features);
+	netdev_feature_set_bits(COMMON_FEATURES | NETIF_F_HW_VLAN_CTAG_TX |
+				NETIF_F_HW_VLAN_STAG_TX, &dev->hw_features);
+	netdev_feature_zero(&dev->vlan_features);
+	netdev_feature_set_bits(COMMON_FEATURES, &dev->vlan_features);
 
 	br->dev = dev;
 	spin_lock_init(&br->lock);
diff --git a/net/bridge/br_if.c b/net/bridge/br_if.c
index 914ca4b2d07c..c0a71fd6f772 100644
--- a/net/bridge/br_if.c
+++ b/net/bridge/br_if.c
@@ -540,8 +540,8 @@ void br_features_recompute(struct net_bridge *br, netdev_features_t *features)
 	if (list_empty(&br->port_list))
 		return;
 
-	mask = *features;
-	*features &= ~NETIF_F_ONE_FOR_ALL;
+	netdev_feature_copy(&mask, *features);
+	netdev_feature_clear_bits(NETIF_F_ONE_FOR_ALL, features);
 
 	list_for_each_entry(p, &br->port_list, list) {
 		netdev_increment_features(features, *features, p->dev->features,
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 028/167] ipvlan: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (26 preceding siblings ...)
  2021-09-29 15:51 ` [RFCv2 net-next 027/167] bridge: " Jian Shen
@ 2021-09-29 15:51 ` Jian Shen
  2021-09-29 15:51 ` [RFCv2 net-next 029/167] veth: " Jian Shen
                   ` (142 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:51 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 drivers/net/ipvlan/ipvlan_main.c | 28 ++++++++++++++++++----------
 drivers/net/ipvlan/ipvtap.c      |  5 +++--
 2 files changed, 21 insertions(+), 12 deletions(-)

diff --git a/drivers/net/ipvlan/ipvlan_main.c b/drivers/net/ipvlan/ipvlan_main.c
index 6a0b7bd2d3ae..84ad786ae18d 100644
--- a/drivers/net/ipvlan/ipvlan_main.c
+++ b/drivers/net/ipvlan/ipvlan_main.c
@@ -135,11 +135,14 @@ static int ipvlan_init(struct net_device *dev)
 
 	dev->state = (dev->state & ~IPVLAN_STATE_MASK) |
 		     (phy_dev->state & IPVLAN_STATE_MASK);
-	dev->features = phy_dev->features & IPVLAN_FEATURES;
-	dev->features |= IPVLAN_ALWAYS_ON;
-	dev->vlan_features = phy_dev->vlan_features & IPVLAN_FEATURES;
-	dev->vlan_features |= IPVLAN_ALWAYS_ON_OFLOADS;
-	dev->hw_enc_features |= dev->features;
+	netdev_feature_copy(&dev->features, phy_dev->features);
+	netdev_feature_and_bits(IPVLAN_FEATURES, &dev->features);
+	netdev_feature_set_bits(IPVLAN_ALWAYS_ON, &dev->features);
+	netdev_feature_copy(&dev->vlan_features, phy_dev->vlan_features);
+	netdev_feature_and_bits(IPVLAN_FEATURES, &dev->vlan_features);
+	netdev_feature_set_bits(IPVLAN_ALWAYS_ON_OFLOADS, &dev->vlan_features);
+	netdev_feature_or(&dev->hw_enc_features, dev->hw_enc_features,
+			  dev->features);
 	dev->gso_max_size = phy_dev->gso_max_size;
 	dev->gso_max_segs = phy_dev->gso_max_segs;
 	dev->hard_header_len = phy_dev->hard_header_len;
@@ -239,13 +242,17 @@ static void ipvlan_fix_features(struct net_device *dev,
 				netdev_features_t *features)
 {
 	struct ipvl_dev *ipvlan = netdev_priv(dev);
+	netdev_features_t tmp;
 
-	*features |= NETIF_F_ALL_FOR_ALL;
-	*features &= (ipvlan->sfeatures | ~IPVLAN_FEATURES);
+	netdev_feature_set_bits(NETIF_F_ALL_FOR_ALL, features);
+	netdev_feature_fill(&tmp);
+	netdev_feature_clear_bits(IPVLAN_FEATURES, &tmp);
+	netdev_feature_or(&tmp, tmp, ipvlan->sfeatures);
+	netdev_feature_and(features, *features, tmp);
 	netdev_increment_features(features, ipvlan->phy_dev->features,
 				  *features, *features);
-	*features |= IPVLAN_ALWAYS_ON;
-	*features &= (IPVLAN_FEATURES | IPVLAN_ALWAYS_ON);
+	netdev_feature_set_bits(IPVLAN_ALWAYS_ON, features);
+	netdev_feature_and_bits(IPVLAN_FEATURES | IPVLAN_ALWAYS_ON, features);
 }
 
 static void ipvlan_change_rx_flags(struct net_device *dev, int change)
@@ -567,7 +574,8 @@ int ipvlan_link_new(struct net *src_net, struct net_device *dev,
 
 	ipvlan->phy_dev = phy_dev;
 	ipvlan->dev = dev;
-	ipvlan->sfeatures = IPVLAN_FEATURES;
+	netdev_feature_zero(&ipvlan->sfeatures);
+	netdev_feature_set_bits(IPVLAN_FEATURES, &ipvlan->sfeatures);
 	if (!tb[IFLA_MTU])
 		ipvlan_adjust_mtu(ipvlan, phy_dev);
 	INIT_LIST_HEAD(&ipvlan->addrs);
diff --git a/drivers/net/ipvlan/ipvtap.c b/drivers/net/ipvlan/ipvtap.c
index 1cedb634f4f7..fd11ce806cc7 100644
--- a/drivers/net/ipvlan/ipvtap.c
+++ b/drivers/net/ipvlan/ipvtap.c
@@ -70,7 +70,7 @@ static void ipvtap_update_features(struct tap_dev *tap,
 	struct ipvtap_dev *vlantap = container_of(tap, struct ipvtap_dev, tap);
 	struct ipvl_dev *vlan = &vlantap->vlan;
 
-	vlan->sfeatures = features;
+	netdev_feature_copy(&vlan->sfeatures, features);
 	netdev_update_features(vlan->dev);
 }
 
@@ -86,7 +86,8 @@ static int ipvtap_newlink(struct net *src_net, struct net_device *dev,
 	/* Since macvlan supports all offloads by default, make
 	 * tap support all offloads also.
 	 */
-	vlantap->tap.tap_features = TUN_OFFLOADS;
+	netdev_feature_zero(&vlantap->tap.tap_features);
+	netdev_feature_set_bits(TUN_OFFLOADS, &vlantap->tap.tap_features);
 	vlantap->tap.count_tx_dropped = ipvtap_count_tx_dropped;
 	vlantap->tap.update_features =	ipvtap_update_features;
 	vlantap->tap.count_rx_dropped = ipvtap_count_rx_dropped;
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 029/167] veth: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (27 preceding siblings ...)
  2021-09-29 15:51 ` [RFCv2 net-next 028/167] ipvlan: " Jian Shen
@ 2021-09-29 15:51 ` Jian Shen
  2021-09-29 15:51 ` [RFCv2 net-next 030/167] bonding: " Jian Shen
                   ` (141 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:51 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 drivers/net/veth.c | 61 ++++++++++++++++++++++++++++------------------
 1 file changed, 37 insertions(+), 24 deletions(-)

diff --git a/drivers/net/veth.c b/drivers/net/veth.c
index 82c900d7ba7b..39bfa024861c 100644
--- a/drivers/net/veth.c
+++ b/drivers/net/veth.c
@@ -310,9 +310,10 @@ static bool veth_skb_is_eligible_for_gro(const struct net_device *dev,
 					 const struct net_device *rcv,
 					 const struct sk_buff *skb)
 {
-	return !(dev->features & NETIF_F_ALL_TSO) ||
+	return !netdev_feature_test_bits(NETIF_F_ALL_TSO, dev->features) ||
 		(skb->destructor == sock_wfree &&
-		 rcv->features & (NETIF_F_GRO_FRAGLIST | NETIF_F_GRO_UDP_FWD));
+		 netdev_feature_test_bits(NETIF_F_GRO_FRAGLIST |
+					  NETIF_F_GRO_UDP_FWD, rcv->features));
 }
 
 static netdev_tx_t veth_xmit(struct sk_buff *skb, struct net_device *dev)
@@ -993,7 +994,7 @@ static void veth_napi_del(struct net_device *dev)
 
 static bool veth_gro_requested(const struct net_device *dev)
 {
-	return !!(dev->wanted_features & NETIF_F_GRO);
+	return netdev_feature_test_bit(NETIF_F_GRO_BIT, dev->wanted_features);
 }
 
 static int veth_enable_xdp_range(struct net_device *dev, int start, int end,
@@ -1075,7 +1076,8 @@ static int veth_enable_xdp(struct net_device *dev)
 				/* user-space did not require GRO, but adding XDP
 				 * is supposed to get GRO working
 				 */
-				dev->features |= NETIF_F_GRO;
+				netdev_feature_set_bit(NETIF_F_GRO_BIT,
+						       &dev->features);
 				netdev_features_change(dev);
 			}
 		}
@@ -1104,7 +1106,8 @@ static void veth_disable_xdp(struct net_device *dev)
 		 * enabled it, clear it now
 		 */
 		if (!veth_gro_requested(dev) && netif_running(dev)) {
-			dev->features &= ~NETIF_F_GRO;
+			netdev_feature_clear_bit(NETIF_F_GRO_BIT,
+						 &dev->features);
 			netdev_features_change(dev);
 		}
 	}
@@ -1393,23 +1396,27 @@ static void veth_fix_features(struct net_device *dev,
 		struct veth_priv *peer_priv = netdev_priv(peer);
 
 		if (peer_priv->_xdp_prog)
-			*features &= ~NETIF_F_GSO_SOFTWARE;
+			netdev_feature_clear_bits(NETIF_F_GSO_SOFTWARE,
+						  features);
 	}
 	if (priv->_xdp_prog)
-		*features |= NETIF_F_GRO;
+		netdev_feature_set_bit(NETIF_F_GRO_BIT, features);
 }
 
 static int veth_set_features(struct net_device *dev,
 			     netdev_features_t features)
 {
-	netdev_features_t changed = features ^ dev->features;
 	struct veth_priv *priv = netdev_priv(dev);
+	netdev_features_t changed;
 	int err;
 
-	if (!(changed & NETIF_F_GRO) || !(dev->flags & IFF_UP) || priv->_xdp_prog)
+	netdev_feature_xor(&changed, features, dev->features);
+
+	if (!netdev_feature_test_bit(NETIF_F_GRO_BIT, changed) ||
+	    !(dev->flags & IFF_UP) || priv->_xdp_prog)
 		return 0;
 
-	if (features & NETIF_F_GRO) {
+	if (netdev_feature_test_bit(NETIF_F_GRO_BIT, features)) {
 		err = veth_napi_enable(dev);
 		if (err)
 			return err;
@@ -1486,7 +1493,8 @@ static int veth_xdp_set(struct net_device *dev, struct bpf_prog *prog,
 		}
 
 		if (!old_prog) {
-			peer->hw_features &= ~NETIF_F_GSO_SOFTWARE;
+			netdev_feature_clear_bits(NETIF_F_GSO_SOFTWARE,
+						  &peer->hw_features);
 			peer->max_mtu = max_mtu;
 		}
 	}
@@ -1497,7 +1505,8 @@ static int veth_xdp_set(struct net_device *dev, struct bpf_prog *prog,
 				veth_disable_xdp(dev);
 
 			if (peer) {
-				peer->hw_features |= NETIF_F_GSO_SOFTWARE;
+				netdev_feature_set_bits(NETIF_F_GSO_SOFTWARE,
+							&peer->hw_features);
 				peer->max_mtu = ETH_MAX_MTU;
 			}
 		}
@@ -1562,20 +1571,24 @@ static void veth_setup(struct net_device *dev)
 
 	dev->netdev_ops = &veth_netdev_ops;
 	dev->ethtool_ops = &veth_ethtool_ops;
-	dev->features |= NETIF_F_LLTX;
-	dev->features |= VETH_FEATURES;
-	dev->vlan_features = dev->features &
-			     ~(NETIF_F_HW_VLAN_CTAG_TX |
-			       NETIF_F_HW_VLAN_STAG_TX |
-			       NETIF_F_HW_VLAN_CTAG_RX |
-			       NETIF_F_HW_VLAN_STAG_RX);
+	netdev_feature_set_bit(NETIF_F_LLTX_BIT, &dev->features);
+	netdev_feature_set_bits(VETH_FEATURES, &dev->features);
+	netdev_feature_copy(&dev->vlan_features, dev->features);
+	netdev_feature_clear_bits(NETIF_F_HW_VLAN_CTAG_TX |
+				  NETIF_F_HW_VLAN_STAG_TX |
+				  NETIF_F_HW_VLAN_CTAG_RX |
+				  NETIF_F_HW_VLAN_STAG_RX,
+				  &dev->vlan_features);
 	dev->needs_free_netdev = true;
 	dev->priv_destructor = veth_dev_free;
 	dev->max_mtu = ETH_MAX_MTU;
 
-	dev->hw_features = VETH_FEATURES;
-	dev->hw_enc_features = VETH_FEATURES;
-	dev->mpls_features = NETIF_F_HW_CSUM | NETIF_F_GSO_SOFTWARE;
+	netdev_feature_zero(&dev->hw_features);
+	netdev_feature_set_bits(VETH_FEATURES, &dev->hw_features);
+	netdev_feature_copy(&dev->hw_enc_features, dev->hw_features);
+	netdev_feature_zero(&dev->mpls_features);
+	netdev_feature_set_bits(NETIF_F_HW_CSUM | NETIF_F_GSO_SOFTWARE,
+				&dev->mpls_features);
 }
 
 /*
@@ -1602,8 +1615,8 @@ static struct rtnl_link_ops veth_link_ops;
 
 static void veth_disable_gro(struct net_device *dev)
 {
-	dev->features &= ~NETIF_F_GRO;
-	dev->wanted_features &= ~NETIF_F_GRO;
+	netdev_feature_clear_bit(NETIF_F_GRO_BIT, &dev->features);
+	netdev_feature_clear_bit(NETIF_F_GRO_BIT, &dev->wanted_features);
 	netdev_update_features(dev);
 }
 
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 030/167] bonding: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (28 preceding siblings ...)
  2021-09-29 15:51 ` [RFCv2 net-next 029/167] veth: " Jian Shen
@ 2021-09-29 15:51 ` Jian Shen
  2021-09-29 15:51 ` [RFCv2 net-next 031/167] net: tun: " Jian Shen
                   ` (140 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:51 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 drivers/net/bonding/bond_main.c    | 106 ++++++++++++++++++-----------
 drivers/net/bonding/bond_options.c |  12 ++--
 2 files changed, 76 insertions(+), 42 deletions(-)

diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index b1c13fa3c677..9c6f37589733 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -1371,15 +1371,15 @@ static void bond_fix_features(struct net_device *dev,
 
 #if IS_ENABLED(CONFIG_TLS_DEVICE)
 	if (bond_sk_check(bond))
-		*features |= BOND_TLS_FEATURES;
+		netdev_feature_set_bits(BOND_TLS_FEATURES, features);
 	else
-		*features &= ~BOND_TLS_FEATURES;
+		netdev_feature_clear_bits(BOND_TLS_FEATURES, features);
 #endif
 
-	mask = *features;
+	netdev_feature_copy(&mask, *features);
 
-	*features &= ~NETIF_F_ONE_FOR_ALL;
-	*features |= NETIF_F_ALL_FOR_ALL;
+	netdev_feature_clear_bits(NETIF_F_ONE_FOR_ALL, features);
+	netdev_feature_set_bits(NETIF_F_ALL_FOR_ALL, features);
 
 	bond_for_each_slave(bond, slave, iter) {
 		netdev_increment_features(features, *features,
@@ -1403,12 +1403,16 @@ static void bond_compute_features(struct bonding *bond)
 {
 	unsigned int dst_release_flag = IFF_XMIT_DST_RELEASE |
 					IFF_XMIT_DST_RELEASE_PERM;
-	netdev_features_t vlan_features = BOND_VLAN_FEATURES;
-	netdev_features_t enc_features  = BOND_ENC_FEATURES;
+	netdev_features_t vlan_features;
+	netdev_features_t enc_features;
+	netdev_features_t vlan_mask;
+	netdev_features_t enc_mask;
 #ifdef CONFIG_XFRM_OFFLOAD
-	netdev_features_t xfrm_features  = BOND_XFRM_FEATURES;
+	netdev_features_t xfrm_features;
+	netdev_features_t xfrm_mask;
 #endif /* CONFIG_XFRM_OFFLOAD */
-	netdev_features_t mpls_features  = BOND_MPLS_FEATURES;
+	netdev_features_t mpls_features;
+	netdev_features_t mpls_mask;
 	struct net_device *bond_dev = bond->dev;
 	struct list_head *iter;
 	struct slave *slave;
@@ -1416,29 +1420,44 @@ static void bond_compute_features(struct bonding *bond)
 	unsigned int gso_max_size = GSO_MAX_SIZE;
 	u16 gso_max_segs = GSO_MAX_SEGS;
 
+	netdev_feature_zero(&vlan_features);
+	netdev_feature_set_bits(BOND_VLAN_FEATURES, &vlan_features);
+	netdev_feature_copy(&vlan_mask, vlan_features);
+	netdev_feature_zero(&enc_features);
+	netdev_feature_set_bits(BOND_ENC_FEATURES, &enc_features);
+	netdev_feature_copy(&enc_mask, enc_features);
+#ifdef CONFIG_XFRM_OFFLOAD
+	netdev_feature_zero(&xfrm_features);
+	netdev_feature_set_bits(BOND_XFRM_FEATURES, &xfrm_features);
+	netdev_feature_copy(&xfrm_mask, xfrm_features);
+#endif /* CONFIG_XFRM_OFFLOAD */
+	netdev_feature_zero(&mpls_features);
+	netdev_feature_set_bits(BOND_MPLS_FEATURES, &mpls_features);
+	netdev_feature_copy(&mpls_mask, mpls_features);
+
 	if (!bond_has_slaves(bond))
 		goto done;
-	vlan_features &= NETIF_F_ALL_FOR_ALL;
-	mpls_features &= NETIF_F_ALL_FOR_ALL;
+	netdev_feature_and_bits(NETIF_F_ALL_FOR_ALL, &vlan_features);
+	netdev_feature_and_bits(NETIF_F_ALL_FOR_ALL, &mpls_features);
 
 	bond_for_each_slave(bond, slave, iter) {
 		netdev_increment_features(&vlan_features, vlan_features,
 					  slave->dev->vlan_features,
-					  BOND_VLAN_FEATURES);
+					  vlan_mask);
 
 		netdev_increment_features(&enc_features, enc_features,
 					  slave->dev->hw_enc_features,
-					  BOND_ENC_FEATURES);
+					  enc_mask);
 
 #ifdef CONFIG_XFRM_OFFLOAD
 		netdev_increment_features(&xfrm_features, xfrm_features,
 					  slave->dev->hw_enc_features,
-					  BOND_XFRM_FEATURES);
+					  xfrm_mask);
 #endif /* CONFIG_XFRM_OFFLOAD */
 
 		netdev_increment_features(&mpls_features, mpls_features,
 					  slave->dev->mpls_features,
-					  BOND_MPLS_FEATURES);
+					  mpls_mask);
 
 		dst_release_flag &= slave->dev->priv_flags;
 		if (slave->dev->hard_header_len > max_hard_header_len)
@@ -1450,14 +1469,16 @@ static void bond_compute_features(struct bonding *bond)
 	bond_dev->hard_header_len = max_hard_header_len;
 
 done:
-	bond_dev->vlan_features = vlan_features;
-	bond_dev->hw_enc_features = enc_features | NETIF_F_GSO_ENCAP_ALL |
-				    NETIF_F_HW_VLAN_CTAG_TX |
-				    NETIF_F_HW_VLAN_STAG_TX;
+	netdev_feature_copy(&bond_dev->vlan_features, vlan_features);
+	netdev_feature_copy(&bond_dev->hw_enc_features, enc_features);
+	netdev_feature_set_bits(NETIF_F_GSO_ENCAP_ALL |
+				NETIF_F_HW_VLAN_CTAG_TX |
+				NETIF_F_HW_VLAN_STAG_TX, &mpls_features);
 #ifdef CONFIG_XFRM_OFFLOAD
-	bond_dev->hw_enc_features |= xfrm_features;
+	netdev_feature_or(&bond_dev->hw_enc_features, bond_dev->hw_enc_features,
+			  xfrm_features);
 #endif /* CONFIG_XFRM_OFFLOAD */
-	bond_dev->mpls_features = mpls_features;
+	netdev_feature_copy(&bond_dev->mpls_features, mpls_features);
 	bond_dev->gso_max_segs = gso_max_segs;
 	netif_set_gso_max_size(bond_dev, gso_max_size);
 
@@ -1781,7 +1802,8 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev,
 
 	/* vlan challenged mutual exclusion */
 	/* no need to lock since we're protected by rtnl_lock */
-	if (slave_dev->features & NETIF_F_VLAN_CHALLENGED) {
+	if (netdev_feature_test_bit(NETIF_F_VLAN_CHALLENGED_BIT,
+				    slave_dev->features)) {
 		slave_dbg(bond_dev, slave_dev, "is NETIF_F_VLAN_CHALLENGED\n");
 		if (vlan_uses_dev(bond_dev)) {
 			SLAVE_NL_ERR(bond_dev, slave_dev, extack,
@@ -1794,7 +1816,7 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev,
 		slave_dbg(bond_dev, slave_dev, "is !NETIF_F_VLAN_CHALLENGED\n");
 	}
 
-	if (slave_dev->features & NETIF_F_HW_ESP)
+	if (netdev_feature_test_bit(NETIF_F_HW_ESP_BIT, slave_dev->features))
 		slave_dbg(bond_dev, slave_dev, "is esp-hw-offload capable\n");
 
 	/* Old ifenslave binaries are no longer supported.  These can
@@ -2087,7 +2109,7 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev,
 	}
 #endif
 
-	if (!(bond_dev->features & NETIF_F_LRO))
+	if (!netdev_feature_test_bit(NETIF_F_LRO_BIT, bond_dev->features))
 		dev_disable_lro(slave_dev);
 
 	res = netdev_rx_handler_register(slave_dev, bond_handle_frame,
@@ -2285,8 +2307,9 @@ static int __bond_release_one(struct net_device *bond_dev,
 	struct slave *slave, *oldcurrent;
 	struct sockaddr_storage ss;
 	int old_flags = bond_dev->flags;
-	netdev_features_t old_features = bond_dev->features;
+	netdev_features_t old_features;
 
+	netdev_feature_copy(&old_features, bond_dev->features);
 	/* slave is not a slave or master is not master of this slave */
 	if (!(slave_dev->flags & IFF_SLAVE) ||
 	    !netdev_has_upper_dev(slave_dev, bond_dev)) {
@@ -2390,8 +2413,9 @@ static int __bond_release_one(struct net_device *bond_dev,
 	}
 
 	bond_compute_features(bond);
-	if (!(bond_dev->features & NETIF_F_VLAN_CHALLENGED) &&
-	    (old_features & NETIF_F_VLAN_CHALLENGED))
+	if (!netdev_feature_test_bit(NETIF_F_VLAN_CHALLENGED_BIT,
+				     bond_dev->features) &&
+	    netdev_feature_test_bit(NETIF_F_VLAN_CHALLENGED_BIT, old_features))
 		slave_info(bond_dev, slave_dev, "last VLAN challenged slave left bond - VLAN blocking is removed\n");
 
 	vlan_vids_del_by_dev(slave_dev, bond_dev);
@@ -5406,7 +5430,7 @@ void bond_setup(struct net_device *bond_dev)
 #endif /* CONFIG_XFRM_OFFLOAD */
 
 	/* don't acquire bond device's netif_tx_lock when transmitting */
-	bond_dev->features |= NETIF_F_LLTX;
+	netdev_feature_set_bit(NETIF_F_LLTX_BIT, &bond_dev->features);
 
 	/* By default, we declare the bond to be fully
 	 * VLAN hardware accelerated capable. Special
@@ -5416,24 +5440,30 @@ void bond_setup(struct net_device *bond_dev)
 	 */
 
 	/* Don't allow bond devices to change network namespaces. */
-	bond_dev->features |= NETIF_F_NETNS_LOCAL;
+	netdev_feature_set_bit(NETIF_F_NETNS_LOCAL_BIT, &bond_dev->features);
 
-	bond_dev->hw_features = BOND_VLAN_FEATURES |
+	netdev_feature_zero(&bond_dev->hw_features);
+	netdev_feature_set_bits(BOND_VLAN_FEATURES |
 				NETIF_F_HW_VLAN_CTAG_RX |
-				NETIF_F_HW_VLAN_CTAG_FILTER;
-
-	bond_dev->hw_features |= NETIF_F_GSO_ENCAP_ALL;
-	bond_dev->features |= bond_dev->hw_features;
-	bond_dev->features |= NETIF_F_HW_VLAN_CTAG_TX | NETIF_F_HW_VLAN_STAG_TX;
+				NETIF_F_HW_VLAN_CTAG_FILTER,
+				&bond_dev->hw_features);
+
+	netdev_feature_set_bits(NETIF_F_GSO_ENCAP_ALL, &bond_dev->hw_features);
+	netdev_feature_or(&bond_dev->features, bond_dev->features,
+			  bond_dev->hw_features);
+	netdev_feature_set_bits(NETIF_F_HW_VLAN_CTAG_TX |
+				NETIF_F_HW_VLAN_STAG_TX,
+				&bond_dev->features);
 #ifdef CONFIG_XFRM_OFFLOAD
-	bond_dev->hw_features |= BOND_XFRM_FEATURES;
+	netdev_feature_set_bits(BOND_XFRM_FEATURES, &bond_dev->hw_features);
 	/* Only enable XFRM features if this is an active-backup config */
 	if (BOND_MODE(bond) == BOND_MODE_ACTIVEBACKUP)
-		bond_dev->features |= BOND_XFRM_FEATURES;
+		netdev_feature_set_bits(BOND_XFRM_FEATURES,
+					&bond_dev->features);
 #endif /* CONFIG_XFRM_OFFLOAD */
 #if IS_ENABLED(CONFIG_TLS_DEVICE)
 	if (bond_sk_check(bond))
-		bond_dev->features |= BOND_TLS_FEATURES;
+		netdev_feature_set_bits(BOND_TLS_FEATURES, &bond_dev->features);
 #endif
 }
 
diff --git a/drivers/net/bonding/bond_options.c b/drivers/net/bonding/bond_options.c
index a8fde3bc458f..b0c3a28c2448 100644
--- a/drivers/net/bonding/bond_options.c
+++ b/drivers/net/bonding/bond_options.c
@@ -778,9 +778,11 @@ static bool bond_set_xfrm_features(struct bonding *bond)
 		return false;
 
 	if (BOND_MODE(bond) == BOND_MODE_ACTIVEBACKUP)
-		bond->dev->wanted_features |= BOND_XFRM_FEATURES;
+		netdev_feature_set_bits(BOND_XFRM_FEATURES,
+					&bond->dev->wanted_features);
 	else
-		bond->dev->wanted_features &= ~BOND_XFRM_FEATURES;
+		netdev_feature_clear_bits(BOND_XFRM_FEATURES,
+					  &bond->dev->wanted_features);
 
 	return true;
 }
@@ -791,9 +793,11 @@ static bool bond_set_tls_features(struct bonding *bond)
 		return false;
 
 	if (bond_sk_check(bond))
-		bond->dev->wanted_features |= BOND_TLS_FEATURES;
+		netdev_feature_set_bits(BOND_TLS_FEATURES,
+					&bond->dev->wanted_features);
 	else
-		bond->dev->wanted_features &= ~BOND_TLS_FEATURES;
+		netdev_feature_clear_bits(BOND_TLS_FEATURES,
+					  &bond->dev->wanted_features);
 
 	return true;
 }
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 031/167] net: tun: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (29 preceding siblings ...)
  2021-09-29 15:51 ` [RFCv2 net-next 030/167] bonding: " Jian Shen
@ 2021-09-29 15:51 ` Jian Shen
  2021-09-29 15:51 ` [RFCv2 net-next 032/167] net: tap: " Jian Shen
                   ` (139 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:51 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 drivers/net/tun.c | 48 ++++++++++++++++++++++++++++++-----------------
 1 file changed, 31 insertions(+), 17 deletions(-)

diff --git a/drivers/net/tun.c b/drivers/net/tun.c
index d89a9874eb37..9e9cbafc870b 100644
--- a/drivers/net/tun.c
+++ b/drivers/net/tun.c
@@ -1083,9 +1083,11 @@ static void tun_net_fix_features(struct net_device *dev,
 				 netdev_features_t *features)
 {
 	struct tun_struct *tun = netdev_priv(dev);
+	netdev_features_t tmp;
 
-	*features = (*features & tun->set_features) |
-			(*features & ~TUN_USER_FEATURES);
+	netdev_feature_and(&tmp, *features, tun->set_features);
+	netdev_feature_clear_bits(TUN_USER_FEATURES, features);
+	netdev_feature_or(features, *features, tmp);
 }
 
 static void tun_set_headroom(struct net_device *dev, int new_hr)
@@ -2727,13 +2729,18 @@ static int tun_set_iff(struct net *net, struct file *file, struct ifreq *ifr)
 		tun_net_init(dev);
 		tun_flow_init(tun);
 
-		dev->hw_features = NETIF_F_SG | NETIF_F_FRAGLIST |
-				   TUN_USER_FEATURES | NETIF_F_HW_VLAN_CTAG_TX |
-				   NETIF_F_HW_VLAN_STAG_TX;
-		dev->features = dev->hw_features | NETIF_F_LLTX;
-		dev->vlan_features = dev->features &
-				     ~(NETIF_F_HW_VLAN_CTAG_TX |
-				       NETIF_F_HW_VLAN_STAG_TX);
+		netdev_feature_zero(&dev->hw_features);
+		netdev_feature_set_bits(NETIF_F_SG | NETIF_F_FRAGLIST |
+					TUN_USER_FEATURES |
+					NETIF_F_HW_VLAN_CTAG_TX |
+					NETIF_F_HW_VLAN_STAG_TX,
+					&dev->hw_features);
+		netdev_feature_copy(&dev->features, dev->hw_features);
+		netdev_feature_set_bit(NETIF_F_LLTX_BIT, &dev->features);
+		netdev_feature_copy(&dev->vlan_features, dev->features);
+		netdev_feature_clear_bits(NETIF_F_HW_VLAN_CTAG_TX |
+					  NETIF_F_HW_VLAN_STAG_TX,
+					  &dev->vlan_features);
 
 		tun->flags = (tun->flags & ~TUN_FEATURES) |
 			      (ifr->ifr_flags & TUN_FEATURES);
@@ -2795,21 +2802,26 @@ static void tun_get_iff(struct tun_struct *tun, struct ifreq *ifr)
  * privs required. */
 static int set_offload(struct tun_struct *tun, unsigned long arg)
 {
-	netdev_features_t features = 0;
+	netdev_features_t features;
+
+	netdev_feature_zero(&features);
 
 	if (arg & TUN_F_CSUM) {
-		features |= NETIF_F_HW_CSUM;
+		netdev_feature_set_bit(NETIF_F_HW_CSUM_BIT, &features);
 		arg &= ~TUN_F_CSUM;
 
 		if (arg & (TUN_F_TSO4|TUN_F_TSO6)) {
 			if (arg & TUN_F_TSO_ECN) {
-				features |= NETIF_F_TSO_ECN;
+				netdev_feature_set_bit(NETIF_F_TSO_ECN_BIT,
+						       &features);
 				arg &= ~TUN_F_TSO_ECN;
 			}
 			if (arg & TUN_F_TSO4)
-				features |= NETIF_F_TSO;
+				netdev_feature_set_bit(NETIF_F_TSO_BIT,
+						       &features);
 			if (arg & TUN_F_TSO6)
-				features |= NETIF_F_TSO6;
+				netdev_feature_set_bit(NETIF_F_TSO6_BIT,
+						       &features);
 			arg &= ~(TUN_F_TSO4|TUN_F_TSO6);
 		}
 
@@ -2821,9 +2833,11 @@ static int set_offload(struct tun_struct *tun, unsigned long arg)
 	if (arg)
 		return -EINVAL;
 
-	tun->set_features = features;
-	tun->dev->wanted_features &= ~TUN_USER_FEATURES;
-	tun->dev->wanted_features |= features;
+	netdev_feature_copy(&tun->set_features, features);
+	netdev_feature_clear_bits(TUN_USER_FEATURES,
+				  &tun->dev->wanted_features);
+	netdev_feature_or(&tun->dev->wanted_features,
+			  tun->dev->wanted_features, features);
 	netdev_update_features(tun->dev);
 
 	return 0;
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 032/167] net: tap: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (30 preceding siblings ...)
  2021-09-29 15:51 ` [RFCv2 net-next 031/167] net: tun: " Jian Shen
@ 2021-09-29 15:51 ` Jian Shen
  2021-09-29 15:51 ` [RFCv2 net-next 033/167] net: geneve: " Jian Shen
                   ` (138 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:51 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 drivers/net/tap.c | 36 ++++++++++++++++++++++--------------
 1 file changed, 22 insertions(+), 14 deletions(-)

diff --git a/drivers/net/tap.c b/drivers/net/tap.c
index 8e3a28ba6b28..03158fa9476c 100644
--- a/drivers/net/tap.c
+++ b/drivers/net/tap.c
@@ -321,7 +321,10 @@ rx_handler_result_t tap_handle_frame(struct sk_buff **pskb)
 	struct net_device *dev = skb->dev;
 	struct tap_dev *tap;
 	struct tap_queue *q;
-	netdev_features_t features = TAP_FEATURES;
+	netdev_features_t features;
+
+	netdev_feature_zero(&features);
+	netdev_feature_set_bits(TAP_FEATURES, &features);
 
 	tap = tap_dev_get_rcu(dev);
 	if (!tap)
@@ -338,7 +341,7 @@ rx_handler_result_t tap_handle_frame(struct sk_buff **pskb)
 	 * enabled.
 	 */
 	if (q->flags & IFF_VNET_HDR)
-		features |= tap->tap_features;
+		netdev_feature_or(&features, features, tap->tap_features);
 	if (netif_needs_gso(skb, features)) {
 		struct sk_buff *segs = __skb_gso_segment(skb, features, false);
 		struct sk_buff *next;
@@ -368,7 +371,7 @@ rx_handler_result_t tap_handle_frame(struct sk_buff **pskb)
 		 *	  check, we either support them all or none.
 		 */
 		if (skb->ip_summed == CHECKSUM_PARTIAL &&
-		    !(features & NETIF_F_CSUM_MASK) &&
+		    !netdev_feature_test_bits(NETIF_F_CSUM_MASK, features) &&
 		    skb_checksum_help(skb))
 			goto drop;
 		if (ptr_ring_produce(&q->ring, skb))
@@ -536,7 +539,8 @@ static int tap_open(struct inode *inode, struct file *file)
 	 * The macvlan supports zerocopy iff the lower device supports zero
 	 * copy so we don't have to look at the lower device directly.
 	 */
-	if ((tap->dev->features & NETIF_F_HIGHDMA) && (tap->dev->features & NETIF_F_SG))
+	if (netdev_feature_test_bit(NETIF_F_HIGHDMA_BIT, tap->dev->features) &&
+	    netdev_feature_test_bit(NETIF_F_SG_BIT, tap->dev->features))
 		sock_set_flag(&q->sk, SOCK_ZEROCOPY);
 
 	err = tap_set_queue(tap, file, q);
@@ -921,24 +925,28 @@ static int set_offload(struct tap_queue *q, unsigned long arg)
 {
 	struct tap_dev *tap;
 	netdev_features_t features;
-	netdev_features_t feature_mask = 0;
+	netdev_features_t feature_mask;
 
 	tap = rtnl_dereference(q->tap);
 	if (!tap)
 		return -ENOLINK;
 
-	features = tap->dev->features;
+	netdev_feature_zero(&feature_mask);
+	netdev_feature_copy(&features, tap->dev->features);
 
 	if (arg & TUN_F_CSUM) {
-		feature_mask = NETIF_F_HW_CSUM;
+		netdev_feature_set_bit(NETIF_F_HW_CSUM_BIT, &feature_mask);
 
 		if (arg & (TUN_F_TSO4 | TUN_F_TSO6)) {
 			if (arg & TUN_F_TSO_ECN)
-				feature_mask |= NETIF_F_TSO_ECN;
+				netdev_feature_set_bit(NETIF_F_TSO_ECN_BIT,
+						       &feature_mask);
 			if (arg & TUN_F_TSO4)
-				feature_mask |= NETIF_F_TSO;
+				netdev_feature_set_bit(NETIF_F_TSO_BIT,
+						       &feature_mask);
 			if (arg & TUN_F_TSO6)
-				feature_mask |= NETIF_F_TSO6;
+				netdev_feature_set_bit(NETIF_F_TSO6_BIT,
+						       &feature_mask);
 		}
 	}
 
@@ -950,15 +958,15 @@ static int set_offload(struct tap_queue *q, unsigned long arg)
 	 * When user space turns off TSO, we turn off GSO/LRO so that
 	 * user-space will not receive TSO frames.
 	 */
-	if (feature_mask & (NETIF_F_TSO | NETIF_F_TSO6))
-		features |= RX_OFFLOADS;
+	if (netdev_feature_test_bits(NETIF_F_TSO | NETIF_F_TSO6, feature_mask))
+		netdev_feature_set_bits(RX_OFFLOADS, &features);
 	else
-		features &= ~RX_OFFLOADS;
+		netdev_feature_clear_bits(RX_OFFLOADS, &features);
 
 	/* tap_features are the same as features on tun/tap and
 	 * reflect user expectations.
 	 */
-	tap->tap_features = feature_mask;
+	netdev_feature_copy(&tap->tap_features, feature_mask);
 	if (tap->update_features)
 		tap->update_features(tap, features);
 
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 033/167] net: geneve: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (31 preceding siblings ...)
  2021-09-29 15:51 ` [RFCv2 net-next 032/167] net: tap: " Jian Shen
@ 2021-09-29 15:51 ` Jian Shen
  2021-09-29 15:51 ` [RFCv2 net-next 034/167] hv_netvsc: " Jian Shen
                   ` (137 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:51 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 drivers/net/geneve.c | 18 ++++++++++--------
 1 file changed, 10 insertions(+), 8 deletions(-)

diff --git a/drivers/net/geneve.c b/drivers/net/geneve.c
index 1ab94b5f9bbf..d2179c103300 100644
--- a/drivers/net/geneve.c
+++ b/drivers/net/geneve.c
@@ -1219,14 +1219,16 @@ static void geneve_setup(struct net_device *dev)
 
 	SET_NETDEV_DEVTYPE(dev, &geneve_type);
 
-	dev->features    |= NETIF_F_LLTX;
-	dev->features    |= NETIF_F_SG | NETIF_F_HW_CSUM | NETIF_F_FRAGLIST;
-	dev->features    |= NETIF_F_RXCSUM;
-	dev->features    |= NETIF_F_GSO_SOFTWARE;
-
-	dev->hw_features |= NETIF_F_SG | NETIF_F_HW_CSUM | NETIF_F_FRAGLIST;
-	dev->hw_features |= NETIF_F_RXCSUM;
-	dev->hw_features |= NETIF_F_GSO_SOFTWARE;
+	netdev_feature_set_bit(NETIF_F_LLTX_BIT, &dev->features);
+	netdev_feature_set_bits(NETIF_F_SG | NETIF_F_HW_CSUM | NETIF_F_FRAGLIST,
+				&dev->features);
+	netdev_feature_set_bit(NETIF_F_RXCSUM_BIT, &dev->features);
+	netdev_feature_set_bits(NETIF_F_GSO_SOFTWARE, &dev->features);
+
+	netdev_feature_set_bits(NETIF_F_SG | NETIF_F_HW_CSUM | NETIF_F_FRAGLIST,
+				&dev->hw_features);
+	netdev_feature_set_bit(NETIF_F_RXCSUM_BIT, &dev->hw_features);
+	netdev_feature_set_bits(NETIF_F_GSO_SOFTWARE, &dev->hw_features);
 
 	/* MTU range: 68 - (something less than 65535) */
 	dev->min_mtu = ETH_MIN_MTU;
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 034/167] hv_netvsc: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (32 preceding siblings ...)
  2021-09-29 15:51 ` [RFCv2 net-next 033/167] net: geneve: " Jian Shen
@ 2021-09-29 15:51 ` Jian Shen
  2021-09-29 15:51 ` [RFCv2 net-next 035/167] macvlan: " Jian Shen
                   ` (136 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:51 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 drivers/net/hyperv/netvsc_bpf.c   |  2 +-
 drivers/net/hyperv/netvsc_drv.c   | 38 ++++++++++++++++++-------------
 drivers/net/hyperv/rndis_filter.c | 30 +++++++++++++++---------
 3 files changed, 42 insertions(+), 28 deletions(-)

diff --git a/drivers/net/hyperv/netvsc_bpf.c b/drivers/net/hyperv/netvsc_bpf.c
index aa877da113f8..058a92696cbc 100644
--- a/drivers/net/hyperv/netvsc_bpf.c
+++ b/drivers/net/hyperv/netvsc_bpf.c
@@ -114,7 +114,7 @@ int netvsc_xdp_set(struct net_device *dev, struct bpf_prog *prog,
 		return -EOPNOTSUPP;
 	}
 
-	if (prog && (dev->features & NETIF_F_LRO)) {
+	if (prog && netdev_feature_test_bit(NETIF_F_LRO_BIT, dev->features)) {
 		netdev_err(dev, "XDP: not support LRO\n");
 		NL_SET_ERR_MSG_MOD(extack, "XDP: not support LRO");
 
diff --git a/drivers/net/hyperv/netvsc_drv.c b/drivers/net/hyperv/netvsc_drv.c
index 5371328422ec..22875a7ef1ff 100644
--- a/drivers/net/hyperv/netvsc_drv.c
+++ b/drivers/net/hyperv/netvsc_drv.c
@@ -892,13 +892,15 @@ static struct sk_buff *netvsc_alloc_recv_skb(struct net_device *net,
 	}
 
 	/* Do L4 checksum offload if enabled and present. */
-	if ((ppi_flags & NVSC_RSC_CSUM_INFO) && (net->features & NETIF_F_RXCSUM)) {
+	if ((ppi_flags & NVSC_RSC_CSUM_INFO) &&
+	    netdev_feature_test_bit(NETIF_F_RXCSUM_BIT, net->features)) {
 		if (csum_info->receive.tcp_checksum_succeeded ||
 		    csum_info->receive.udp_checksum_succeeded)
 			skb->ip_summed = CHECKSUM_UNNECESSARY;
 	}
 
-	if ((ppi_flags & NVSC_RSC_HASH_INFO) && (net->features & NETIF_F_RXHASH))
+	if ((ppi_flags & NVSC_RSC_HASH_INFO) &&
+	    netdev_feature_test_bit(NETIF_F_RXHASH_BIT, net->features))
 		skb_set_hash(skb, *hash_info, PKT_HASH_TYPE_L4);
 
 	if (ppi_flags & NVSC_RSC_VLAN) {
@@ -1205,7 +1207,8 @@ static void netvsc_init_settings(struct net_device *dev)
 	ndc->speed = SPEED_UNKNOWN;
 	ndc->duplex = DUPLEX_FULL;
 
-	dev->features = NETIF_F_LRO;
+	netdev_feature_zero(&dev->features);
+	netdev_feature_set_bit(NETIF_F_LRO_BIT, &dev->features);
 }
 
 static int netvsc_get_link_ksettings(struct net_device *dev,
@@ -1928,8 +1931,9 @@ static void netvsc_fix_features(struct net_device *ndev,
 	if (!nvdev || nvdev->destroy)
 		return;
 
-	if ((*features & NETIF_F_LRO) && netvsc_xdp_get(nvdev)) {
-		*features ^= NETIF_F_LRO;
+	if (netdev_feature_test_bit(NETIF_F_LRO_BIT, *features) &&
+	    netvsc_xdp_get(nvdev)) {
+		netdev_feature_change_bit(NETIF_F_LRO_BIT, features);
 		netdev_info(ndev, "Skip LRO - unsupported with XDP\n");
 	}
 }
@@ -1937,22 +1941,24 @@ static void netvsc_fix_features(struct net_device *ndev,
 static int netvsc_set_features(struct net_device *ndev,
 			       netdev_features_t features)
 {
-	netdev_features_t change = features ^ ndev->features;
 	struct net_device_context *ndevctx = netdev_priv(ndev);
 	struct netvsc_device *nvdev = rtnl_dereference(ndevctx->nvdev);
 	struct net_device *vf_netdev = rtnl_dereference(ndevctx->vf_netdev);
 	struct ndis_offload_params offloads;
+	netdev_features_t change;
 	int ret = 0;
 
 	if (!nvdev || nvdev->destroy)
 		return -ENODEV;
 
-	if (!(change & NETIF_F_LRO))
+	netdev_feature_xor(&change, features, ndev->features);
+
+	if (!netdev_feature_test_bit(NETIF_F_LRO_BIT, change))
 		goto syncvf;
 
 	memset(&offloads, 0, sizeof(struct ndis_offload_params));
 
-	if (features & NETIF_F_LRO) {
+	if (netdev_feature_test_bit(NETIF_F_LRO_BIT, features)) {
 		offloads.rsc_ip_v4 = NDIS_OFFLOAD_PARAMETERS_RSC_ENABLED;
 		offloads.rsc_ip_v6 = NDIS_OFFLOAD_PARAMETERS_RSC_ENABLED;
 	} else {
@@ -1963,15 +1969,15 @@ static int netvsc_set_features(struct net_device *ndev,
 	ret = rndis_filter_set_offload_params(ndev, nvdev, &offloads);
 
 	if (ret) {
-		features ^= NETIF_F_LRO;
-		ndev->features = features;
+		netdev_feature_change_bit(NETIF_F_LRO_BIT, &features);
+		netdev_feature_copy(&ndev->features, features);
 	}
 
 syncvf:
 	if (!vf_netdev)
 		return ret;
 
-	vf_netdev->wanted_features = features;
+	netdev_feature_copy(&vf_netdev->wanted_features, features);
 	netdev_update_features(vf_netdev);
 
 	return ret;
@@ -2385,7 +2391,7 @@ static int netvsc_register_vf(struct net_device *vf_netdev)
 	if (ndev->needed_headroom < vf_netdev->needed_headroom)
 		ndev->needed_headroom = vf_netdev->needed_headroom;
 
-	vf_netdev->wanted_features = ndev->features;
+	netdev_feature_copy(&vf_netdev->wanted_features, ndev->features);
 	netdev_update_features(vf_netdev);
 
 	prog = netvsc_xdp_get(netvsc_dev);
@@ -2550,10 +2556,10 @@ static int netvsc_probe(struct hv_device *dev,
 		schedule_work(&nvdev->subchan_work);
 
 	/* hw_features computed in rndis_netdev_set_hwcaps() */
-	net->features = net->hw_features |
-		NETIF_F_HIGHDMA | NETIF_F_HW_VLAN_CTAG_TX |
-		NETIF_F_HW_VLAN_CTAG_RX;
-	net->vlan_features = net->features;
+	netdev_feature_copy(&net->features, net->hw_features);
+	netdev_feature_set_bits(NETIF_F_HIGHDMA | NETIF_F_HW_VLAN_CTAG_TX |
+				NETIF_F_HW_VLAN_CTAG_RX, &net->features);
+	netdev_feature_copy(&net->vlan_features, net->features);
 
 	netdev_lockdep_set_classes(net);
 
diff --git a/drivers/net/hyperv/rndis_filter.c b/drivers/net/hyperv/rndis_filter.c
index f6c9c2a670f9..115ebe0b970a 100644
--- a/drivers/net/hyperv/rndis_filter.c
+++ b/drivers/net/hyperv/rndis_filter.c
@@ -1348,6 +1348,7 @@ static int rndis_netdev_set_hwcaps(struct rndis_device *rndis_device,
 	struct ndis_offload hwcaps;
 	struct ndis_offload_params offloads;
 	unsigned int gso_max_size = GSO_MAX_SIZE;
+	netdev_features_t tmp;
 	int ret;
 
 	/* Find HW offload capabilities */
@@ -1362,24 +1363,26 @@ static int rndis_netdev_set_hwcaps(struct rndis_device *rndis_device,
 	offloads.ip_v4_csum = NDIS_OFFLOAD_PARAMETERS_TX_RX_DISABLED;
 
 	/* Reset previously set hw_features flags */
-	net->hw_features &= ~NETVSC_SUPPORTED_HW_FEATURES;
+	netdev_feature_clear_bits(NETVSC_SUPPORTED_HW_FEATURES,
+				  &net->hw_features);
 	net_device_ctx->tx_checksum_mask = 0;
 
 	/* Compute tx offload settings based on hw capabilities */
-	net->hw_features |= NETIF_F_RXCSUM;
-	net->hw_features |= NETIF_F_SG;
-	net->hw_features |= NETIF_F_RXHASH;
+	netdev_feature_set_bit(NETIF_F_RXCSUM_BIT, &net->hw_features);
+	netdev_feature_set_bit(NETIF_F_SG_BIT, &net->hw_features);
+	netdev_feature_set_bit(NETIF_F_RXHASH_BIT, &net->hw_features);
 
 	if ((hwcaps.csum.ip4_txcsum & NDIS_TXCSUM_ALL_TCP4) == NDIS_TXCSUM_ALL_TCP4) {
 		/* Can checksum TCP */
-		net->hw_features |= NETIF_F_IP_CSUM;
+		netdev_feature_set_bit(NETIF_F_IP_CSUM_BIT, &net->hw_features);
 		net_device_ctx->tx_checksum_mask |= TRANSPORT_INFO_IPV4_TCP;
 
 		offloads.tcp_ip_v4_csum = NDIS_OFFLOAD_PARAMETERS_TX_RX_ENABLED;
 
 		if (hwcaps.lsov2.ip4_encap & NDIS_OFFLOAD_ENCAP_8023) {
 			offloads.lso_v2_ipv4 = NDIS_OFFLOAD_PARAMETERS_LSOV2_ENABLED;
-			net->hw_features |= NETIF_F_TSO;
+			netdev_feature_set_bit(NETIF_F_TSO_BIT,
+					       &net->hw_features);
 
 			if (hwcaps.lsov2.ip4_maxsz < gso_max_size)
 				gso_max_size = hwcaps.lsov2.ip4_maxsz;
@@ -1392,7 +1395,8 @@ static int rndis_netdev_set_hwcaps(struct rndis_device *rndis_device,
 	}
 
 	if ((hwcaps.csum.ip6_txcsum & NDIS_TXCSUM_ALL_TCP6) == NDIS_TXCSUM_ALL_TCP6) {
-		net->hw_features |= NETIF_F_IPV6_CSUM;
+		netdev_feature_set_bit(NETIF_F_IPV6_CSUM_BIT,
+				       &net->hw_features);
 
 		offloads.tcp_ip_v6_csum = NDIS_OFFLOAD_PARAMETERS_TX_RX_ENABLED;
 		net_device_ctx->tx_checksum_mask |= TRANSPORT_INFO_IPV6_TCP;
@@ -1400,7 +1404,8 @@ static int rndis_netdev_set_hwcaps(struct rndis_device *rndis_device,
 		if ((hwcaps.lsov2.ip6_encap & NDIS_OFFLOAD_ENCAP_8023) &&
 		    (hwcaps.lsov2.ip6_opts & NDIS_LSOV2_CAP_IP6) == NDIS_LSOV2_CAP_IP6) {
 			offloads.lso_v2_ipv6 = NDIS_OFFLOAD_PARAMETERS_LSOV2_ENABLED;
-			net->hw_features |= NETIF_F_TSO6;
+			netdev_feature_set_bit(NETIF_F_TSO6_BIT,
+					       &net->hw_features);
 
 			if (hwcaps.lsov2.ip6_maxsz < gso_max_size)
 				gso_max_size = hwcaps.lsov2.ip6_maxsz;
@@ -1413,9 +1418,9 @@ static int rndis_netdev_set_hwcaps(struct rndis_device *rndis_device,
 	}
 
 	if (hwcaps.rsc.ip4 && hwcaps.rsc.ip6) {
-		net->hw_features |= NETIF_F_LRO;
+		netdev_feature_set_bit(NETIF_F_LRO_BIT, &net->hw_features);
 
-		if (net->features & NETIF_F_LRO) {
+		if (netdev_feature_test_bit(NETIF_F_LRO_BIT, net->features)) {
 			offloads.rsc_ip_v4 = NDIS_OFFLOAD_PARAMETERS_RSC_ENABLED;
 			offloads.rsc_ip_v6 = NDIS_OFFLOAD_PARAMETERS_RSC_ENABLED;
 		} else {
@@ -1427,7 +1432,10 @@ static int rndis_netdev_set_hwcaps(struct rndis_device *rndis_device,
 	/* In case some hw_features disappeared we need to remove them from
 	 * net->features list as they're no longer supported.
 	 */
-	net->features &= ~NETVSC_SUPPORTED_HW_FEATURES | net->hw_features;
+	netdev_feature_fill(&tmp);
+	netdev_feature_clear_bits(NETVSC_SUPPORTED_HW_FEATURES, &tmp);
+	netdev_feature_or(&tmp, tmp, net->hw_features);
+	netdev_feature_and(&net->features, net->features, tmp);
 
 	netif_set_gso_max_size(net, gso_max_size);
 
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 035/167] macvlan: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (33 preceding siblings ...)
  2021-09-29 15:51 ` [RFCv2 net-next 034/167] hv_netvsc: " Jian Shen
@ 2021-09-29 15:51 ` Jian Shen
  2021-09-29 15:51 ` [RFCv2 net-next 036/167] macsec: " Jian Shen
                   ` (135 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:51 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 drivers/net/macvlan.c | 49 +++++++++++++++++++++++++++----------------
 1 file changed, 31 insertions(+), 18 deletions(-)

diff --git a/drivers/net/macvlan.c b/drivers/net/macvlan.c
index 3c408653e864..9ac38e22ccfc 100644
--- a/drivers/net/macvlan.c
+++ b/drivers/net/macvlan.c
@@ -620,7 +620,8 @@ static int macvlan_open(struct net_device *dev)
 	/* Attempt to populate accel_priv which is used to offload the L2
 	 * forwarding requests for unicast packets.
 	 */
-	if (lowerdev->features & NETIF_F_HW_L2FW_DOFFLOAD)
+	if (netdev_feature_test_bit(NETIF_F_HW_L2FW_DOFFLOAD_BIT,
+				    lowerdev->features))
 		vlan->accel_priv =
 		      lowerdev->netdev_ops->ndo_dfwd_add_station(lowerdev, dev);
 
@@ -893,12 +894,15 @@ static int macvlan_init(struct net_device *dev)
 
 	dev->state		= (dev->state & ~MACVLAN_STATE_MASK) |
 				  (lowerdev->state & MACVLAN_STATE_MASK);
-	dev->features 		= lowerdev->features & MACVLAN_FEATURES;
-	dev->features		|= ALWAYS_ON_FEATURES;
-	dev->hw_features	|= NETIF_F_LRO;
-	dev->vlan_features	= lowerdev->vlan_features & MACVLAN_FEATURES;
-	dev->vlan_features	|= ALWAYS_ON_OFFLOADS;
-	dev->hw_enc_features    |= dev->features;
+	netdev_feature_copy(&dev->features, lowerdev->features);
+	netdev_feature_and_bits(MACVLAN_FEATURES, &dev->features);
+	netdev_feature_set_bits(ALWAYS_ON_FEATURES, &dev->features);
+	netdev_feature_set_bit(NETIF_F_LRO_BIT, &dev->hw_features);
+	netdev_feature_copy(&dev->vlan_features, lowerdev->vlan_features);
+	netdev_feature_and_bits(MACVLAN_FEATURES, &dev->vlan_features);
+	netdev_feature_set_bits(ALWAYS_ON_OFFLOADS, &dev->vlan_features);
+	netdev_feature_or(&dev->hw_enc_features, dev->hw_enc_features,
+			  dev->features);
 	dev->gso_max_size	= lowerdev->gso_max_size;
 	dev->gso_max_segs	= lowerdev->gso_max_segs;
 	dev->hard_header_len	= lowerdev->hard_header_len;
@@ -1071,17 +1075,25 @@ static void macvlan_fix_features(struct net_device *dev,
 				 netdev_features_t *features)
 {
 	struct macvlan_dev *vlan = netdev_priv(dev);
-	netdev_features_t lowerdev_features = vlan->lowerdev->features;
-	netdev_features_t mask;
-
-	*features |= NETIF_F_ALL_FOR_ALL;
-	*features &= (vlan->set_features | ~MACVLAN_FEATURES);
-	mask = *features;
-
-	lowerdev_features &= (*features | ~NETIF_F_LRO);
+	netdev_features_t lowerdev_features;
+	netdev_features_t mask, tmp;
+
+	netdev_feature_copy(&lowerdev_features, vlan->lowerdev->features);
+	netdev_feature_set_bits(NETIF_F_ALL_FOR_ALL, features);
+	netdev_feature_fill(&tmp);
+	netdev_feature_clear_bits(MACVLAN_FEATURES, &tmp);
+	netdev_feature_or(&tmp, tmp, vlan->set_features);
+	netdev_feature_and(features, *features, tmp);
+	netdev_feature_copy(&mask, *features);
+
+	netdev_feature_fill(&tmp);
+	netdev_feature_clear_bit(NETIF_F_LRO_BIT, &tmp);
+	netdev_feature_or(&tmp, tmp, *features);
+	netdev_feature_and(&lowerdev_features, lowerdev_features, tmp);
 	netdev_increment_features(features, lowerdev_features, *features, mask);
-	*features |= ALWAYS_ON_FEATURES;
-	*features &= (ALWAYS_ON_FEATURES | MACVLAN_FEATURES);
+	netdev_feature_set_bits(ALWAYS_ON_FEATURES, features);
+	netdev_feature_and_bits(ALWAYS_ON_FEATURES | MACVLAN_FEATURES,
+				features);
 }
 
 #ifdef CONFIG_NET_POLL_CONTROLLER
@@ -1458,7 +1470,8 @@ int macvlan_common_newlink(struct net *src_net, struct net_device *dev,
 	vlan->lowerdev = lowerdev;
 	vlan->dev      = dev;
 	vlan->port     = port;
-	vlan->set_features = MACVLAN_FEATURES;
+	netdev_feature_zero(&vlan->set_features);
+	netdev_feature_set_bits(MACVLAN_FEATURES, &vlan->set_features);
 
 	vlan->mode     = MACVLAN_MODE_VEPA;
 	if (data && data[IFLA_MACVLAN_MODE])
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 036/167] macsec: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (34 preceding siblings ...)
  2021-09-29 15:51 ` [RFCv2 net-next 035/167] macvlan: " Jian Shen
@ 2021-09-29 15:51 ` Jian Shen
  2021-09-29 15:51 ` [RFCv2 net-next 037/167] net: tls: " Jian Shen
                   ` (134 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:51 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 drivers/net/macsec.c | 39 ++++++++++++++++++++++++---------------
 1 file changed, 24 insertions(+), 15 deletions(-)

diff --git a/drivers/net/macsec.c b/drivers/net/macsec.c
index 601b833a40bd..5526d54c0a14 100644
--- a/drivers/net/macsec.c
+++ b/drivers/net/macsec.c
@@ -347,7 +347,8 @@ static bool macsec_check_offload(enum macsec_offload offload,
 		return macsec->real_dev->phydev &&
 		       macsec->real_dev->phydev->macsec_ops;
 	else if (offload == MACSEC_OFFLOAD_MAC)
-		return macsec->real_dev->features & NETIF_F_HW_MACSEC &&
+		return netdev_feature_test_bit(NETIF_F_HW_MACSEC_BIT,
+					       macsec->real_dev->features) &&
 		       macsec->real_dev->macsec_ops;
 
 	return false;
@@ -3419,13 +3420,6 @@ static netdev_tx_t macsec_start_xmit(struct sk_buff *skb,
 #define SW_MACSEC_FEATURES \
 	(NETIF_F_SG | NETIF_F_HIGHDMA | NETIF_F_FRAGLIST)
 
-/* If h/w offloading is enabled, use real device features save for
- *   VLAN_FEATURES - they require additional ops
- *   HW_MACSEC - no reason to report it
- */
-#define REAL_DEV_FEATURES(dev) \
-	((dev)->features & ~(NETIF_F_VLAN_FEATURES | NETIF_F_HW_MACSEC))
-
 static int macsec_dev_init(struct net_device *dev)
 {
 	struct macsec_dev *macsec = macsec_priv(dev);
@@ -3443,10 +3437,19 @@ static int macsec_dev_init(struct net_device *dev)
 	}
 
 	if (macsec_is_offloaded(macsec)) {
-		dev->features = REAL_DEV_FEATURES(real_dev);
+		/* If h/w offloading is enabled, use real device features save for
+		 *   VLAN_FEATURES - they require additional ops
+		 *   HW_MACSEC - no reason to report it
+		 */
+		netdev_feature_copy(&dev->features, real_dev->features);
+		netdev_feature_clear_bits(NETIF_F_VLAN_FEATURES |
+					  NETIF_F_HW_MACSEC,
+					  &dev->features);
 	} else {
-		dev->features = real_dev->features & SW_MACSEC_FEATURES;
-		dev->features |= NETIF_F_LLTX | NETIF_F_GSO_SOFTWARE;
+		netdev_feature_copy(&dev->features, real_dev->features);
+		netdev_feature_and_bits(SW_MACSEC_FEATURES, &dev->features);
+		netdev_feature_set_bits(NETIF_F_LLTX | NETIF_F_GSO_SOFTWARE,
+					&dev->features);
 	}
 
 	dev->needed_headroom = real_dev->needed_headroom +
@@ -3475,15 +3478,21 @@ static void macsec_fix_features(struct net_device *dev,
 {
 	struct macsec_dev *macsec = macsec_priv(dev);
 	struct net_device *real_dev = macsec->real_dev;
+	netdev_features_t tmp;
 
 	if (macsec_is_offloaded(macsec)) {
-		*features = REAL_DEV_FEATURES(real_dev);
+		netdev_feature_copy(features, real_dev->features);
+		netdev_feature_clear_bits(NETIF_F_VLAN_FEATURES |
+					  NETIF_F_HW_MACSEC, features);
 		return;
 	}
 
-	*features &= (real_dev->features & SW_MACSEC_FEATURES) |
-		    NETIF_F_GSO_SOFTWARE | NETIF_F_SOFT_FEATURES;
-	*features |= NETIF_F_LLTX;
+	netdev_feature_copy(&tmp, real_dev->features);
+	netdev_feature_and_bits(SW_MACSEC_FEATURES, &tmp);
+	netdev_feature_set_bits(NETIF_F_GSO_SOFTWARE | NETIF_F_SOFT_FEATURES,
+				&tmp);
+	netdev_feature_and(features, *features, tmp);
+	netdev_feature_set_bit(NETIF_F_LLTX_BIT, features);
 }
 
 static int macsec_dev_open(struct net_device *dev)
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 037/167] net: tls: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (35 preceding siblings ...)
  2021-09-29 15:51 ` [RFCv2 net-next 036/167] macsec: " Jian Shen
@ 2021-09-29 15:51 ` Jian Shen
  2021-09-29 15:51 ` [RFCv2 net-next 038/167] s390: qeth: " Jian Shen
                   ` (133 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:51 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 net/tls/tls_device.c | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)

diff --git a/net/tls/tls_device.c b/net/tls/tls_device.c
index b932469ee69c..94189107995f 100644
--- a/net/tls/tls_device.c
+++ b/net/tls/tls_device.c
@@ -1125,7 +1125,7 @@ int tls_set_device_offload(struct sock *sk, struct tls_context *ctx)
 		goto disable_cad;
 	}
 
-	if (!(netdev->features & NETIF_F_HW_TLS_TX)) {
+	if (!netdev_feature_test_bit(NETIF_F_HW_TLS_TX_BIT, netdev->features)) {
 		rc = -EOPNOTSUPP;
 		goto release_netdev;
 	}
@@ -1200,7 +1200,7 @@ int tls_set_device_offload_rx(struct sock *sk, struct tls_context *ctx)
 		return -EINVAL;
 	}
 
-	if (!(netdev->features & NETIF_F_HW_TLS_RX)) {
+	if (!netdev_feature_test_bit(NETIF_F_HW_TLS_RX_BIT, netdev->features)) {
 		rc = -EOPNOTSUPP;
 		goto release_netdev;
 	}
@@ -1361,7 +1361,8 @@ static int tls_dev_event(struct notifier_block *this, unsigned long event,
 	struct net_device *dev = netdev_notifier_info_to_dev(ptr);
 
 	if (!dev->tlsdev_ops &&
-	    !(dev->features & (NETIF_F_HW_TLS_RX | NETIF_F_HW_TLS_TX)))
+	    !netdev_feature_test_bits(NETIF_F_HW_TLS_RX | NETIF_F_HW_TLS_TX,
+				      dev->features))
 		return NOTIFY_DONE;
 
 	switch (event) {
@@ -1369,7 +1370,8 @@ static int tls_dev_event(struct notifier_block *this, unsigned long event,
 	case NETDEV_FEAT_CHANGE:
 		if (netif_is_bond_master(dev))
 			return NOTIFY_DONE;
-		if ((dev->features & NETIF_F_HW_TLS_RX) &&
+		if (netdev_feature_test_bit(NETIF_F_HW_TLS_RX_BIT,
+					    dev->features) &&
 		    !dev->tlsdev_ops->tls_dev_resync)
 			return NOTIFY_BAD;
 
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 038/167] s390: qeth: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (36 preceding siblings ...)
  2021-09-29 15:51 ` [RFCv2 net-next 037/167] net: tls: " Jian Shen
@ 2021-09-29 15:51 ` Jian Shen
  2021-09-29 15:51 ` [RFCv2 net-next 039/167] dsa: " Jian Shen
                   ` (132 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:51 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 drivers/s390/net/qeth_core_main.c | 122 ++++++++++++++++++------------
 drivers/s390/net/qeth_l2_main.c   |  44 +++++++----
 drivers/s390/net/qeth_l3_main.c   |  41 ++++++----
 3 files changed, 130 insertions(+), 77 deletions(-)

diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c
index 6a1941e5fd51..21e4a356c717 100644
--- a/drivers/s390/net/qeth_core_main.c
+++ b/drivers/s390/net/qeth_core_main.c
@@ -5608,7 +5608,8 @@ static void qeth_receive_skb(struct qeth_card *card, struct sk_buff *skb,
 		return;
 	}
 
-	if (is_cso && (card->dev->features & NETIF_F_RXCSUM)) {
+	if (is_cso && netdev_feature_test_bit(NETIF_F_RXCSUM_BIT,
+					      card->dev->features)) {
 		skb->ip_summed = CHECKSUM_UNNECESSARY;
 		QETH_CARD_STAT_INC(card, rx_skb_csum);
 	} else {
@@ -6385,10 +6386,10 @@ static struct net_device *qeth_alloc_netdev(struct qeth_card *card)
 
 	dev->ethtool_ops = &qeth_ethtool_ops;
 	dev->priv_flags &= ~IFF_TX_SKB_SHARING;
-	dev->hw_features |= NETIF_F_SG;
-	dev->vlan_features |= NETIF_F_SG;
+	netdev_feature_set_bit(NETIF_F_SG_BIT, &dev->hw_features);
+	netdev_feature_set_bit(NETIF_F_SG_BIT, &dev->vlan_features);
 	if (IS_IQD(card))
-		dev->features |= NETIF_F_SG;
+		netdev_feature_set_bit(NETIF_F_SG_BIT, &dev->features);
 
 	return dev;
 }
@@ -6835,18 +6836,21 @@ void qeth_enable_hw_features(struct net_device *dev)
 	struct qeth_card *card = dev->ml_priv;
 	netdev_features_t features;
 
-	features = dev->features;
+	netdev_feature_copy(&features, dev->features);
 	/* force-off any feature that might need an IPA sequence.
 	 * netdev_update_features() will restart them.
 	 */
-	dev->features &= ~dev->hw_features;
+	netdev_feature_andnot(&dev->features, dev->features,
+			      dev->hw_features);
 	/* toggle VLAN filter, so that VIDs are re-programmed: */
 	if (IS_LAYER2(card) && IS_VM_NIC(card)) {
-		dev->features &= ~NETIF_F_HW_VLAN_CTAG_FILTER;
-		dev->wanted_features |= NETIF_F_HW_VLAN_CTAG_FILTER;
+		netdev_feature_clear_bit(NETIF_F_HW_VLAN_CTAG_FILTER_BIT,
+					 &dev->features);
+		netdev_feature_set_bit(NETIF_F_HW_VLAN_CTAG_FILTER_BIT,
+				       &dev->wanted_features);
 	}
 	netdev_update_features(dev);
-	if (features != dev->features)
+	if (!netdev_feature_equal(features, dev->features))
 		dev_warn(&card->gdev->dev,
 			 "Device recovery failed to restore all offload features\n");
 }
@@ -6856,69 +6860,88 @@ static void qeth_check_restricted_features(struct qeth_card *card,
 					   netdev_features_t changed,
 					   netdev_features_t actual)
 {
-	netdev_features_t ipv6_features = NETIF_F_TSO6;
-	netdev_features_t ipv4_features = NETIF_F_TSO;
+	netdev_features_t ipv6_features;
+	netdev_features_t ipv4_features;
 
+	netdev_features_zero(&ipv6_features);
+	netdev_features_zero(&ipv4_features);
+	netdev_features_set_bit(NETIF_F_TSO6_BIT, &ipv6_features);
+	netdev_features_set_bit(NETIF_F_TSO_BIT, &ipv4_features);
 	if (!card->info.has_lp2lp_cso_v6)
-		ipv6_features |= NETIF_F_IPV6_CSUM;
+		netdev_features_set_bit(NETIF_F_IPV6_CSUM_BIT, &ipv6_features);
 	if (!card->info.has_lp2lp_cso_v4)
-		ipv4_features |= NETIF_F_IP_CSUM;
+		netdev_features_set_bit(NETIF_F_IP_CSUM_BIT, &ipv4_features);
 
-	if ((changed & ipv6_features) && !(actual & ipv6_features))
+	if (netdev_feature_intersects(ipv6_features, changed) &&
+	    !netdev_feature_intersects(ipv6_features, actual))
 		qeth_flush_local_addrs6(card);
-	if ((changed & ipv4_features) && !(actual & ipv4_features))
+	if (netdev_feature_intersects(ipv4_features, changed) &&
+	    !netdev_feature_intersects(ipv4_features, actual))
 		qeth_flush_local_addrs4(card);
 }
 
 int qeth_set_features(struct net_device *dev, netdev_features_t features)
 {
 	struct qeth_card *card = dev->ml_priv;
-	netdev_features_t changed = dev->features ^ features;
+	netdev_features_t changed, tmp1, tmp2;
 	int rc = 0;
 
 	QETH_CARD_TEXT(card, 2, "setfeat");
 	QETH_CARD_HEX(card, 2, &features, sizeof(features));
 
-	if ((changed & NETIF_F_IP_CSUM)) {
-		rc = qeth_set_ipa_csum(card, features & NETIF_F_IP_CSUM,
+	netdev_feature_xor(&changed, dev->features, features);
+	if (netdev_feature_test_bit(NETIF_F_IP_CSUM_BIT, changed)) {
+		rc = qeth_set_ipa_csum(card,
+				       netdev_feature_test_bit(NETIF_F_IP_CSUM_BIT,
+							       features),
 				       IPA_OUTBOUND_CHECKSUM, QETH_PROT_IPV4,
 				       &card->info.has_lp2lp_cso_v4);
 		if (rc)
-			changed ^= NETIF_F_IP_CSUM;
+			netdev_feature_change_bit(NETIF_F_IP_CSUM_BIT, &changed);
 	}
-	if (changed & NETIF_F_IPV6_CSUM) {
-		rc = qeth_set_ipa_csum(card, features & NETIF_F_IPV6_CSUM,
+	if (netdev_feature_test_bit(NETIF_F_IPV6_CSUM_BIT, changed)) {
+		rc = qeth_set_ipa_csum(card,
+				       netdev_feature_test_bit(NETIF_F_IPV6_CSUM_BIT,
+							       features),
 				       IPA_OUTBOUND_CHECKSUM, QETH_PROT_IPV6,
 				       &card->info.has_lp2lp_cso_v6);
 		if (rc)
-			changed ^= NETIF_F_IPV6_CSUM;
+			netdev_feature_change_bit(NETIF_F_IPV6_CSUM_BIT, &changed);
 	}
-	if (changed & NETIF_F_RXCSUM) {
-		rc = qeth_set_ipa_rx_csum(card, features & NETIF_F_RXCSUM);
+	if (netdev_feature_test_bit(NETIF_F_RXCSUM_BIT, changed)) {
+		rc = qeth_set_ipa_rx_csum(card,
+					  netdev_feature_test_bit(NETIF_F_RXCSUM_BIT,
+								  features));
 		if (rc)
-			changed ^= NETIF_F_RXCSUM;
+			netdev_feature_change_bit(NETIF_F_RXCSUM_BIT, &changed);
 	}
-	if (changed & NETIF_F_TSO) {
-		rc = qeth_set_ipa_tso(card, features & NETIF_F_TSO,
+	if (netdev_feature_test_bit(NETIF_F_TSO_BIT, changed)) {
+		rc = qeth_set_ipa_tso(card,
+				      netdev_feature_test_bit(NETIF_F_TSO_BIT,
+							      features),
 				      QETH_PROT_IPV4);
 		if (rc)
-			changed ^= NETIF_F_TSO;
+			netdev_feature_change_bit(NETIF_F_TSO_BIT, &changed);
 	}
-	if (changed & NETIF_F_TSO6) {
-		rc = qeth_set_ipa_tso(card, features & NETIF_F_TSO6,
+	if (netdev_feature_test_bit(NETIF_F_TSO6_BIT, changed)) {
+		rc = qeth_set_ipa_tso(card,
+				      netdev_feature_test_bit(NETIF_F_TSO6_BIT,
+							      features),
 				      QETH_PROT_IPV6);
 		if (rc)
-			changed ^= NETIF_F_TSO6;
+			netdev_feature_change_bit(NETIF_F_TSO6_BIT, &changed);
 	}
 
-	qeth_check_restricted_features(card, dev->features ^ features,
-				       dev->features ^ changed);
+	netdev_feature_xor(&tmp1, dev->features, features);
+	netdev_feature_xor(&tmp2, dev->features, changed);
+	qeth_check_restricted_features(card, tmp1, tmp2);
 
 	/* everything changed successfully? */
-	if ((dev->features ^ features) == changed)
+	netdev_feature_xor(&tmp1, dev->features, features);
+	if (netdev_feature_equal(tmp1, changed))
 		return 0;
 	/* something went wrong. save changed features and return error */
-	dev->features ^= changed;
+	netdev_feature_xor(&dev->features, dev->features, changed);
 	return -EIO;
 }
 EXPORT_SYMBOL_GPL(qeth_set_features);
@@ -6929,16 +6952,16 @@ void qeth_fix_features(struct net_device *dev, netdev_features_t *features)
 
 	QETH_CARD_TEXT(card, 2, "fixfeat");
 	if (!qeth_is_supported(card, IPA_OUTBOUND_CHECKSUM))
-		*features &= ~NETIF_F_IP_CSUM;
+		netdev_feature_clear_bit(NETIF_F_IP_CSUM_BIT, features);
 	if (!qeth_is_supported6(card, IPA_OUTBOUND_CHECKSUM_V6))
-		*features &= ~NETIF_F_IPV6_CSUM;
+		netdev_feature_clear_bit(NETIF_F_IPV6_CSUM_BIT, features);
 	if (!qeth_is_supported(card, IPA_INBOUND_CHECKSUM) &&
 	    !qeth_is_supported6(card, IPA_INBOUND_CHECKSUM_V6))
-		*features &= ~NETIF_F_RXCSUM;
+		netdev_feature_clear_bit(NETIF_F_RXCSUM_BIT, features);
 	if (!qeth_is_supported(card, IPA_OUTBOUND_TSO))
-		*features &= ~NETIF_F_TSO;
+		netdev_feature_clear_bit(NETIF_F_TSO_BIT, features);
 	if (!qeth_is_supported6(card, IPA_OUTBOUND_TSO))
-		*features &= ~NETIF_F_TSO6;
+		netdev_feature_clear_bit(NETIF_F_TSO6_BIT, features);
 
 	QETH_CARD_HEX(card, 2, features, sizeof(*features));
 }
@@ -6952,25 +6975,30 @@ void qeth_features_check(struct sk_buff *skb, struct net_device *dev,
 	/* Traffic with local next-hop is not eligible for some offloads: */
 	if (skb->ip_summed == CHECKSUM_PARTIAL &&
 	    READ_ONCE(card->options.isolation) != ISOLATION_MODE_FWD) {
-		netdev_features_t restricted = 0;
+		netdev_features_t restricted;
 
+		netdev_feature_zero(&restricted);
 		if (skb_is_gso(skb) && !netif_needs_gso(skb, *features))
-			restricted |= NETIF_F_ALL_TSO;
+			netdev_feature_set_bits(NETIF_F_ALL_TSO, &restricted);
 
 		switch (vlan_get_protocol(skb)) {
 		case htons(ETH_P_IP):
 			if (!card->info.has_lp2lp_cso_v4)
-				restricted |= NETIF_F_IP_CSUM;
+				netdev_feature_set_bit(NETIF_F_IP_CSUM_BIT,
+						       &restricted);
 
 			if (restricted && qeth_next_hop_is_local_v4(card, skb))
-				*features &= ~restricted;
+				netdev_feature_andnot(features, features,
+						      &restricted);
 			break;
 		case htons(ETH_P_IPV6):
 			if (!card->info.has_lp2lp_cso_v6)
-				restricted |= NETIF_F_IPV6_CSUM;
+				netdev_feature_set_bit(NETIF_F_IPV6_CSUM_BIT,
+						       &restricted);
 
 			if (restricted && qeth_next_hop_is_local_v6(card, skb))
-				*features &= ~restricted;
+				netdev_feature_andnot(features, *features,
+						      restricted);
 			break;
 		default:
 			break;
@@ -6992,7 +7020,7 @@ void qeth_features_check(struct sk_buff *skb, struct net_device *dev,
 
 		/* linearize only if resulting skb allocations are order-0: */
 		if (SKB_DATA_ALIGN(hroom + doffset + hsize) <= SKB_MAX_HEAD(0))
-			*features &= ~NETIF_F_SG;
+			netdev_feature_clear_bit(NETIF_F_SG_BIT, features);
 	}
 
 	vlan_features_check(skb, features);
diff --git a/drivers/s390/net/qeth_l2_main.c b/drivers/s390/net/qeth_l2_main.c
index dc6c00768d91..c9750d3e842c 100644
--- a/drivers/s390/net/qeth_l2_main.c
+++ b/drivers/s390/net/qeth_l2_main.c
@@ -1079,40 +1079,54 @@ static int qeth_l2_setup_netdev(struct qeth_card *card)
 	card->dev->priv_flags |= IFF_UNICAST_FLT;
 
 	if (IS_OSM(card)) {
-		card->dev->features |= NETIF_F_VLAN_CHALLENGED;
+		netdev_feature_set_bit(NETIF_F_VLAN_CHALLENGED_BIT,
+				       &card->dev->features);
 	} else {
 		if (!IS_VM_NIC(card))
-			card->dev->hw_features |= NETIF_F_HW_VLAN_CTAG_FILTER;
-		card->dev->features |= NETIF_F_HW_VLAN_CTAG_FILTER;
+			netdev_feature_set_bit(NETIF_F_HW_VLAN_CTAG_FILTER_BIT,
+					       &card->dev->hw_features);
+		netdev_feature_set_bit(NETIF_F_HW_VLAN_CTAG_FILTER_BIT,
+				       &card->dev->features);
 	}
 
 	if (IS_OSD(card) && !IS_VM_NIC(card)) {
-		card->dev->features |= NETIF_F_SG;
+		netdev_feature_set_bit(NETIF_F_SG_BIT, &card->dev->features);
 		/* OSA 3S and earlier has no RX/TX support */
 		if (qeth_is_supported(card, IPA_OUTBOUND_CHECKSUM)) {
-			card->dev->hw_features |= NETIF_F_IP_CSUM;
-			card->dev->vlan_features |= NETIF_F_IP_CSUM;
+			netdev_feature_set_bit(NETIF_F_IP_CSUM_BIT,
+					       &card->dev->hw_features);
+			netdev_feature_set_bit(NETIF_F_IP_CSUM_BIT,
+					       &card->dev->vlan_features);
 		}
 	}
 	if (qeth_is_supported6(card, IPA_OUTBOUND_CHECKSUM_V6)) {
-		card->dev->hw_features |= NETIF_F_IPV6_CSUM;
-		card->dev->vlan_features |= NETIF_F_IPV6_CSUM;
+		netdev_feature_set_bit(NETIF_F_IPV6_CSUM_BIT,
+				       &card->dev->hw_features);
+		netdev_feature_set_bit(NETIF_F_IPV6_CSUM_BIT,
+				       &card->dev->vlan_features);
 	}
 	if (qeth_is_supported(card, IPA_INBOUND_CHECKSUM) ||
 	    qeth_is_supported6(card, IPA_INBOUND_CHECKSUM_V6)) {
-		card->dev->hw_features |= NETIF_F_RXCSUM;
-		card->dev->vlan_features |= NETIF_F_RXCSUM;
+		netdev_feature_set_bit(NETIF_F_RXCSUM_BIT,
+				       &card->dev->hw_features);
+		netdev_feature_set_bit(NETIF_F_RXCSUM_BIT,
+				       &card->dev->vlan_features);
 	}
 	if (qeth_is_supported(card, IPA_OUTBOUND_TSO)) {
-		card->dev->hw_features |= NETIF_F_TSO;
-		card->dev->vlan_features |= NETIF_F_TSO;
+		netdev_feature_set_bit(NETIF_F_TSO_BIT,
+				       &card->dev->hw_features);
+		netdev_feature_set_bit(NETIF_F_TSO_BIT,
+				       &card->dev->vlan_features);
 	}
 	if (qeth_is_supported6(card, IPA_OUTBOUND_TSO)) {
-		card->dev->hw_features |= NETIF_F_TSO6;
-		card->dev->vlan_features |= NETIF_F_TSO6;
+		netdev_feature_set_bit(NETIF_F_TSO6_BIT,
+				       &card->dev->hw_features);
+		netdev_feature_set_bit(NETIF_F_TSO6_BIT,
+				       &card->dev->vlan_features);
 	}
 
-	if (card->dev->hw_features & (NETIF_F_TSO | NETIF_F_TSO6)) {
+	if (netdev_feature_test_bits(NETIF_F_TSO | NETIF_F_TSO6,
+				     card->dev->hw_features)) {
 		card->dev->needed_headroom = sizeof(struct qeth_hdr_tso);
 		netif_keep_dst(card->dev);
 		netif_set_gso_max_size(card->dev,
diff --git a/drivers/s390/net/qeth_l3_main.c b/drivers/s390/net/qeth_l3_main.c
index 1d3d0377fe4b..54416c37f52f 100644
--- a/drivers/s390/net/qeth_l3_main.c
+++ b/drivers/s390/net/qeth_l3_main.c
@@ -1809,7 +1809,8 @@ static void qeth_l3_osa_features_check(struct sk_buff *skb,
 				       netdev_features_t *features)
 {
 	if (vlan_get_protocol(skb) != htons(ETH_P_IP))
-		*features &= ~NETIF_F_HW_VLAN_CTAG_TX;
+		netdev_feature_clear_bit(NETIF_F_HW_VLAN_CTAG_TX_BIT,
+					 features);
 	qeth_features_check(skb, dev, features);
 }
 
@@ -1878,26 +1879,35 @@ static int qeth_l3_setup_netdev(struct qeth_card *card)
 		dev->dev_id = qeth_l3_get_unique_id(card, dev->dev_id);
 
 		if (!IS_VM_NIC(card)) {
-			card->dev->features |= NETIF_F_SG;
-			card->dev->hw_features |= NETIF_F_TSO |
-				NETIF_F_RXCSUM | NETIF_F_IP_CSUM;
-			card->dev->vlan_features |= NETIF_F_TSO |
-				NETIF_F_RXCSUM | NETIF_F_IP_CSUM;
+			netdev_feature_set_bit(NETIF_F_SG_BIT,
+					       &card->dev->features);
+			netdev_feature_set_bits(NETIF_F_TSO | NETIF_F_RXCSUM |
+						NETIF_F_IP_CSUM,
+						&card->dev->hw_features);
+			netdev_feature_set_bits(NETIF_F_TSO | NETIF_F_RXCSUM |
+						NETIF_F_IP_CSUM,
+						&card->dev->vlan_features);
 		}
 
 		if (qeth_is_supported6(card, IPA_OUTBOUND_CHECKSUM_V6)) {
-			card->dev->hw_features |= NETIF_F_IPV6_CSUM;
-			card->dev->vlan_features |= NETIF_F_IPV6_CSUM;
+			netdev_feature_set_bit(NETIF_F_IPV6_CSUM_BIT,
+					       &card->dev->hw_features);
+			netdev_feature_set_bit(NETIF_F_IPV6_CSUM_BIT,
+					       &card->dev->vlan_features);
 		}
 		if (qeth_is_supported6(card, IPA_OUTBOUND_TSO)) {
-			card->dev->hw_features |= NETIF_F_TSO6;
-			card->dev->vlan_features |= NETIF_F_TSO6;
+			netdev_feature_set_bit(NETIF_F_TSO6_BIT,
+					       &card->dev->hw_features);
+			netdev_feature_set_bit(NETIF_F_TSO6_BIT,
+					       &card->dev->vlan_features);
 		}
 
 		/* allow for de-acceleration of NETIF_F_HW_VLAN_CTAG_TX: */
-		if (card->dev->hw_features & NETIF_F_TSO6)
+		if (netdev_feature_test_bit(NETIF_F_TSO6_BIT,
+					    card->dev->hw_features))
 			headroom = sizeof(struct qeth_hdr_tso) + VLAN_HLEN;
-		else if (card->dev->hw_features & NETIF_F_TSO)
+		else if (netdev_feature_test_bit(NETIF_F_TSO_BIT,
+						 card->dev->hw_features))
 			headroom = sizeof(struct qeth_hdr_tso);
 		else
 			headroom = sizeof(struct qeth_hdr) + VLAN_HLEN;
@@ -1913,11 +1923,12 @@ static int qeth_l3_setup_netdev(struct qeth_card *card)
 		return -ENODEV;
 
 	card->dev->needed_headroom = headroom;
-	card->dev->features |=	NETIF_F_HW_VLAN_CTAG_TX |
-				NETIF_F_HW_VLAN_CTAG_RX;
+	netdev_feature_set_bits(NETIF_F_HW_VLAN_CTAG_TX |
+				NETIF_F_HW_VLAN_CTAG_RX, &card->dev->features);
 
 	netif_keep_dst(card->dev);
-	if (card->dev->hw_features & (NETIF_F_TSO | NETIF_F_TSO6))
+	if (netdev_feature_test_bits(NETIF_F_TSO | NETIF_F_TSO6,
+				     card->dev->hw_features))
 		netif_set_gso_max_size(card->dev,
 				       PAGE_SIZE * (QETH_MAX_BUFFER_ELEMENTS(card) - 1));
 
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 039/167] dsa: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (37 preceding siblings ...)
  2021-09-29 15:51 ` [RFCv2 net-next 038/167] s390: qeth: " Jian Shen
@ 2021-09-29 15:51 ` Jian Shen
  2021-09-29 15:51 ` [RFCv2 net-next 040/167] macvtap: " Jian Shen
                   ` (131 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:51 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 drivers/net/dsa/xrs700x/xrs700x.c |  6 ++++--
 net/dsa/slave.c                   | 24 +++++++++++++++---------
 2 files changed, 19 insertions(+), 11 deletions(-)

diff --git a/drivers/net/dsa/xrs700x/xrs700x.c b/drivers/net/dsa/xrs700x/xrs700x.c
index 469420941054..f5b5807b2313 100644
--- a/drivers/net/dsa/xrs700x/xrs700x.c
+++ b/drivers/net/dsa/xrs700x/xrs700x.c
@@ -639,7 +639,8 @@ static int xrs700x_hsr_join(struct dsa_switch *ds, int port,
 	hsr_pair[1] = partner->index;
 	for (i = 0; i < ARRAY_SIZE(hsr_pair); i++) {
 		slave = dsa_to_port(ds, hsr_pair[i])->slave;
-		slave->features |= XRS7000X_SUPPORTED_HSR_FEATURES;
+		netdev_feature_set_bits(XRS7000X_SUPPORTED_HSR_FEATURES,
+					&slave->features);
 	}
 
 	return 0;
@@ -693,7 +694,8 @@ static int xrs700x_hsr_leave(struct dsa_switch *ds, int port,
 	hsr_pair[1] = partner->index;
 	for (i = 0; i < ARRAY_SIZE(hsr_pair); i++) {
 		slave = dsa_to_port(ds, hsr_pair[i])->slave;
-		slave->features &= ~XRS7000X_SUPPORTED_HSR_FEATURES;
+		netdev_feature_clear_bits(XRS7000X_SUPPORTED_HSR_FEATURES,
+					  &slave->features);
 	}
 
 	return 0;
diff --git a/net/dsa/slave.c b/net/dsa/slave.c
index a2bf2d8ac65b..dbd53653b5f0 100644
--- a/net/dsa/slave.c
+++ b/net/dsa/slave.c
@@ -1460,12 +1460,14 @@ int dsa_slave_manage_vlan_filtering(struct net_device *slave,
 	int err;
 
 	if (vlan_filtering) {
-		slave->features |= NETIF_F_HW_VLAN_CTAG_FILTER;
+		netdev_feature_set_bit(NETIF_F_HW_VLAN_CTAG_FILTER_BIT,
+				       &slave->features);
 
 		err = vlan_for_each(slave, dsa_slave_restore_vlan, slave);
 		if (err) {
 			vlan_for_each(slave, dsa_slave_clear_vlan, slave);
-			slave->features &= ~NETIF_F_HW_VLAN_CTAG_FILTER;
+			netdev_feature_clear_bit(NETIF_F_HW_VLAN_CTAG_FILTER_BIT,
+						 &slave->features);
 			return err;
 		}
 	} else {
@@ -1473,7 +1475,8 @@ int dsa_slave_manage_vlan_filtering(struct net_device *slave,
 		if (err)
 			return err;
 
-		slave->features &= ~NETIF_F_HW_VLAN_CTAG_FILTER;
+		netdev_feature_clear_bit(NETIF_F_HW_VLAN_CTAG_FILTER_BIT,
+					 &slave->features);
 	}
 
 	return 0;
@@ -1883,13 +1886,16 @@ void dsa_slave_setup_tagger(struct net_device *slave)
 
 	p->xmit = cpu_dp->tag_ops->xmit;
 
-	slave->features = master->vlan_features | NETIF_F_HW_TC;
-	slave->hw_features |= NETIF_F_HW_TC;
-	slave->features |= NETIF_F_LLTX;
+	netdev_feature_copy(&slave->features, master->vlan_features);
+	netdev_feature_set_bit(NETIF_F_HW_TC_BIT, &slave->features);
+	netdev_feature_set_bit(NETIF_F_HW_TC_BIT, &slave->hw_features);
+	netdev_feature_set_bit(NETIF_F_LLTX_BIT, &slave->features);
 	if (slave->needed_tailroom)
-		slave->features &= ~(NETIF_F_SG | NETIF_F_FRAGLIST);
+		netdev_feature_clear_bits(NETIF_F_SG | NETIF_F_FRAGLIST,
+					  &slave->features);
 	if (ds->needs_standalone_vlan_filtering)
-		slave->features |= NETIF_F_HW_VLAN_CTAG_FILTER;
+		netdev_feature_set_bit(NETIF_F_HW_VLAN_CTAG_FILTER_BIT,
+				       &slave->features);
 }
 
 static struct lock_class_key dsa_slave_netdev_xmit_lock_key;
@@ -1968,7 +1974,7 @@ int dsa_slave_create(struct dsa_port *port)
 
 	SET_NETDEV_DEV(slave_dev, port->ds->dev);
 	slave_dev->dev.of_node = port->dn;
-	slave_dev->vlan_features = master->vlan_features;
+	netdev_feature_copy(&slave_dev->vlan_features, master->vlan_features);
 
 	p = netdev_priv(slave_dev);
 	slave_dev->tstats = netdev_alloc_pcpu_stats(struct pcpu_sw_netstats);
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 040/167] macvtap: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (38 preceding siblings ...)
  2021-09-29 15:51 ` [RFCv2 net-next 039/167] dsa: " Jian Shen
@ 2021-09-29 15:51 ` Jian Shen
  2021-09-29 15:51 ` [RFCv2 net-next 041/167] team: " Jian Shen
                   ` (130 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:51 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 drivers/net/macvtap.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/net/macvtap.c b/drivers/net/macvtap.c
index 694e2f5dbbe5..2162080133bb 100644
--- a/drivers/net/macvtap.c
+++ b/drivers/net/macvtap.c
@@ -74,7 +74,7 @@ static void macvtap_update_features(struct tap_dev *tap,
 	struct macvtap_dev *vlantap = container_of(tap, struct macvtap_dev, tap);
 	struct macvlan_dev *vlan = &vlantap->vlan;
 
-	vlan->set_features = features;
+	netdev_feature_copy(&vlan->set_features, features);
 	netdev_update_features(vlan->dev);
 }
 
@@ -90,7 +90,8 @@ static int macvtap_newlink(struct net *src_net, struct net_device *dev,
 	/* Since macvlan supports all offloads by default, make
 	 * tap support all offloads also.
 	 */
-	vlantap->tap.tap_features = TUN_OFFLOADS;
+	netdev_feature_zero(&vlantap->tap.tap_features);
+	netdev_feature_set_bits(TUN_OFFLOADS, &vlantap->tap.tap_features);
 
 	/* Register callbacks for rx/tx drops accounting and updating
 	 * net_device features
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 041/167] team: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (39 preceding siblings ...)
  2021-09-29 15:51 ` [RFCv2 net-next 040/167] macvtap: " Jian Shen
@ 2021-09-29 15:51 ` Jian Shen
  2021-09-29 15:51 ` [RFCv2 net-next 042/167] vmxnet3: " Jian Shen
                   ` (129 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:51 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 drivers/net/team/team.c | 69 +++++++++++++++++++++++++----------------
 1 file changed, 42 insertions(+), 27 deletions(-)

diff --git a/drivers/net/team/team.c b/drivers/net/team/team.c
index fa96ee62c91a..207771e0d1c8 100644
--- a/drivers/net/team/team.c
+++ b/drivers/net/team/team.c
@@ -985,21 +985,30 @@ static void team_port_disable(struct team *team,
 static void __team_compute_features(struct team *team)
 {
 	struct team_port *port;
-	netdev_features_t vlan_features = TEAM_VLAN_FEATURES &
-					  NETIF_F_ALL_FOR_ALL;
-	netdev_features_t enc_features  = TEAM_ENC_FEATURES;
 	unsigned short max_hard_header_len = ETH_HLEN;
 	unsigned int dst_release_flag = IFF_XMIT_DST_RELEASE |
 					IFF_XMIT_DST_RELEASE_PERM;
+	netdev_features_t vlan_features;
+	netdev_features_t enc_features;
+	netdev_features_t mask_vlan;
+	netdev_features_t mask_enc;
+
+	netdev_feature_zero(&vlan_features);
+	netdev_feature_set_bits(TEAM_VLAN_FEATURES & NETIF_F_ALL_FOR_ALL,
+				&vlan_features);
+	netdev_feature_zero(&enc_features);
+	netdev_feature_set_bits(TEAM_ENC_FEATURES, &enc_features);
+	netdev_feature_zero(&mask_vlan);
+	netdev_feature_set_bits(TEAM_VLAN_FEATURES, &mask_vlan);
+	netdev_feature_zero(&mask_enc);
+	netdev_feature_set_bits(TEAM_ENC_FEATURES, &mask_enc);
 
 	rcu_read_lock();
 	list_for_each_entry_rcu(port, &team->port_list, list) {
 		netdev_increment_features(&vlan_features, vlan_features,
-					  port->dev->vlan_features,
-					  TEAM_VLAN_FEATURES);
+					  port->dev->vlan_features, mask_vlan);
 		netdev_increment_features(&enc_features, enc_features,
-					  port->dev->hw_enc_features,
-					  TEAM_ENC_FEATURES);
+					  port->dev->hw_enc_features, mask_enc);
 
 		dst_release_flag &= port->dev->priv_flags;
 		if (port->dev->hard_header_len > max_hard_header_len)
@@ -1007,10 +1016,12 @@ static void __team_compute_features(struct team *team)
 	}
 	rcu_read_unlock();
 
-	team->dev->vlan_features = vlan_features;
-	team->dev->hw_enc_features = enc_features | NETIF_F_GSO_ENCAP_ALL |
-				     NETIF_F_HW_VLAN_CTAG_TX |
-				     NETIF_F_HW_VLAN_STAG_TX;
+	netdev_feature_copy(&team->dev->vlan_features, vlan_features);
+	netdev_feature_copy(&team->dev->hw_enc_features, enc_features);
+	netdev_feature_set_bits(NETIF_F_GSO_ENCAP_ALL |
+				NETIF_F_HW_VLAN_CTAG_TX |
+				NETIF_F_HW_VLAN_STAG_TX,
+				&team->dev->hw_enc_features);
 	team->dev->hard_header_len = max_hard_header_len;
 
 	team->dev->priv_flags &= ~IFF_XMIT_DST_RELEASE;
@@ -1161,7 +1172,8 @@ static int team_port_add(struct team *team, struct net_device *port_dev,
 		return -EBUSY;
 	}
 
-	if (port_dev->features & NETIF_F_VLAN_CHALLENGED &&
+	if (netdev_feature_test_bit(NETIF_F_VLAN_CHALLENGED_BIT,
+				    port_dev->features) &&
 	    vlan_uses_dev(dev)) {
 		NL_SET_ERR_MSG(extack, "Device is VLAN challenged and team device has VLAN set up");
 		netdev_err(dev, "Device %s is VLAN challenged and team device has VLAN set up\n",
@@ -1226,7 +1238,7 @@ static int team_port_add(struct team *team, struct net_device *port_dev,
 		goto err_enable_netpoll;
 	}
 
-	if (!(dev->features & NETIF_F_LRO))
+	if (!netdev_feature_test_bit(NETIF_F_LRO_BIT, dev->features))
 		dev_disable_lro(port_dev);
 
 	err = netdev_rx_handler_register(port_dev, team_handle_frame,
@@ -2000,9 +2012,9 @@ static void team_fix_features(struct net_device *dev,
 	struct team *team = netdev_priv(dev);
 	netdev_features_t mask;
 
-	mask = *features;
-	*features &= ~NETIF_F_ONE_FOR_ALL;
-	*features |= NETIF_F_ALL_FOR_ALL;
+	netdev_feature_copy(&mask, *features);
+	netdev_feature_clear_bits(NETIF_F_ONE_FOR_ALL, features);
+	netdev_feature_set_bits(NETIF_F_ALL_FOR_ALL, features);
 
 	rcu_read_lock();
 	list_for_each_entry_rcu(port, &team->port_list, list) {
@@ -2160,19 +2172,22 @@ static void team_setup(struct net_device *dev)
 	 */
 	dev->priv_flags |= IFF_UNICAST_FLT | IFF_LIVE_ADDR_CHANGE;
 
-	dev->features |= NETIF_F_LLTX;
-	dev->features |= NETIF_F_GRO;
+	netdev_feature_set_bit(NETIF_F_LLTX_BIT, &dev->features);
+	netdev_feature_set_bit(NETIF_F_GRO_BIT, &dev->features);
 
 	/* Don't allow team devices to change network namespaces. */
-	dev->features |= NETIF_F_NETNS_LOCAL;
-
-	dev->hw_features = TEAM_VLAN_FEATURES |
-			   NETIF_F_HW_VLAN_CTAG_RX |
-			   NETIF_F_HW_VLAN_CTAG_FILTER;
-
-	dev->hw_features |= NETIF_F_GSO_ENCAP_ALL;
-	dev->features |= dev->hw_features;
-	dev->features |= NETIF_F_HW_VLAN_CTAG_TX | NETIF_F_HW_VLAN_STAG_TX;
+	netdev_feature_set_bit(NETIF_F_NETNS_LOCAL_BIT, &dev->features);
+
+	netdev_feature_set_bits(TEAM_VLAN_FEATURES |
+				NETIF_F_HW_VLAN_CTAG_RX |
+				NETIF_F_HW_VLAN_CTAG_FILTER,
+				&dev->hw_features);
+
+	netdev_feature_set_bits(NETIF_F_GSO_ENCAP_ALL, &dev->hw_features);
+	netdev_feature_or(&dev->features, dev->features, dev->hw_features);
+	netdev_feature_set_bits(NETIF_F_HW_VLAN_CTAG_TX |
+				NETIF_F_HW_VLAN_STAG_TX,
+				&dev->features);
 }
 
 static int team_newlink(struct net *src_net, struct net_device *dev,
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 042/167] vmxnet3: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (40 preceding siblings ...)
  2021-09-29 15:51 ` [RFCv2 net-next 041/167] team: " Jian Shen
@ 2021-09-29 15:51 ` Jian Shen
  2021-09-29 15:51 ` [RFCv2 net-next 043/167] net_failover: " Jian Shen
                   ` (128 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:51 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 drivers/net/vmxnet3/vmxnet3_drv.c     | 75 +++++++++++++++++----------
 drivers/net/vmxnet3/vmxnet3_ethtool.c | 73 ++++++++++++++++----------
 2 files changed, 94 insertions(+), 54 deletions(-)

diff --git a/drivers/net/vmxnet3/vmxnet3_drv.c b/drivers/net/vmxnet3/vmxnet3_drv.c
index 142f70670f5c..06acbad37a6d 100644
--- a/drivers/net/vmxnet3/vmxnet3_drv.c
+++ b/drivers/net/vmxnet3/vmxnet3_drv.c
@@ -1223,7 +1223,9 @@ vmxnet3_rx_csum(struct vmxnet3_adapter *adapter,
 		struct sk_buff *skb,
 		union Vmxnet3_GenericDesc *gdesc)
 {
-	if (!gdesc->rcd.cnc && adapter->netdev->features & NETIF_F_RXCSUM) {
+	if (!gdesc->rcd.cnc &&
+	    netdev_feature_test_bit(NETIF_F_RXCSUM_BIT,
+				    adapter->netdev->features)) {
 		if (gdesc->rcd.v4 &&
 		    (le32_to_cpu(gdesc->dword[3]) &
 		     VMXNET3_RCD_CSUM_OK) == VMXNET3_RCD_CSUM_OK) {
@@ -1478,7 +1480,8 @@ vmxnet3_rq_rx_complete(struct vmxnet3_rx_queue *rq,
 
 #ifdef VMXNET3_RSS
 			if (rcd->rssType != VMXNET3_RCD_RSS_TYPE_NONE &&
-			    (adapter->netdev->features & NETIF_F_RXHASH)) {
+			    netdev_feature_test_bit(NETIF_F_RXHASH_BIT,
+						    adapter->netdev->features)) {
 				enum pkt_hash_types hash_type;
 
 				switch (rcd->rssType) {
@@ -1581,7 +1584,8 @@ vmxnet3_rq_rx_complete(struct vmxnet3_rx_queue *rq,
 					(union Vmxnet3_GenericDesc *)rcd);
 			skb->protocol = eth_type_trans(skb, adapter->netdev);
 			if (!rcd->tcp ||
-			    !(adapter->netdev->features & NETIF_F_LRO))
+			    !netdev_feature_test_bit(NETIF_F_LRO_BIT,
+						     adapter->netdev->features))
 				goto not_lro;
 
 			if (segCnt != 0 && mss != 0) {
@@ -1612,7 +1616,8 @@ vmxnet3_rq_rx_complete(struct vmxnet3_rx_queue *rq,
 			if (unlikely(rcd->ts))
 				__vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q), rcd->tci);
 
-			if (adapter->netdev->features & NETIF_F_LRO)
+			if (netdev_feature_test_bit(NETIF_F_LRO_BIT,
+						    adapter->netdev->features))
 				netif_receive_skb(skb);
 			else
 				napi_gro_receive(&rq->napi, skb);
@@ -2501,18 +2506,22 @@ vmxnet3_setup_driver_shared(struct vmxnet3_adapter *adapter)
 	devRead->misc.ddLen = cpu_to_le32(sizeof(struct vmxnet3_adapter));
 
 	/* set up feature flags */
-	if (adapter->netdev->features & NETIF_F_RXCSUM)
+	if (netdev_feature_test_bit(NETIF_F_RXCSUM_BIT,
+				    adapter->netdev->features))
 		devRead->misc.uptFeatures |= UPT1_F_RXCSUM;
 
-	if (adapter->netdev->features & NETIF_F_LRO) {
+	if (netdev_feature_test_bit(NETIF_F_LRO_BIT,
+				    adapter->netdev->features)) {
 		devRead->misc.uptFeatures |= UPT1_F_LRO;
 		devRead->misc.maxNumRxSG = cpu_to_le16(1 + MAX_SKB_FRAGS);
 	}
-	if (adapter->netdev->features & NETIF_F_HW_VLAN_CTAG_RX)
+	if (netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_RX_BIT,
+				    adapter->netdev->features))
 		devRead->misc.uptFeatures |= UPT1_F_RXVLAN;
 
-	if (adapter->netdev->features & (NETIF_F_GSO_UDP_TUNNEL |
-					 NETIF_F_GSO_UDP_TUNNEL_CSUM))
+	if (netdev_feature_test_bits(NETIF_F_GSO_UDP_TUNNEL |
+				     NETIF_F_GSO_UDP_TUNNEL_CSUM,
+				     adapter->netdev->features))
 		devRead->misc.uptFeatures |= UPT1_F_RXINNEROFLD;
 
 	devRead->misc.mtu = cpu_to_le32(adapter->netdev->mtu);
@@ -3163,28 +3172,39 @@ vmxnet3_declare_features(struct vmxnet3_adapter *adapter, bool dma64)
 {
 	struct net_device *netdev = adapter->netdev;
 
-	netdev->hw_features = NETIF_F_SG | NETIF_F_RXCSUM |
-		NETIF_F_HW_CSUM | NETIF_F_HW_VLAN_CTAG_TX |
-		NETIF_F_HW_VLAN_CTAG_RX | NETIF_F_TSO | NETIF_F_TSO6 |
-		NETIF_F_LRO;
+	netdev_feature_zero(&netdev->hw_features);
+	netdev_feature_set_bits(NETIF_F_SG | NETIF_F_RXCSUM |
+				NETIF_F_HW_CSUM | NETIF_F_HW_VLAN_CTAG_TX |
+				NETIF_F_HW_VLAN_CTAG_RX | NETIF_F_TSO |
+				NETIF_F_TSO6 | NETIF_F_LRO,
+				&netdev->hw_features);
 
 	if (VMXNET3_VERSION_GE_4(adapter)) {
-		netdev->hw_features |= NETIF_F_GSO_UDP_TUNNEL |
-				NETIF_F_GSO_UDP_TUNNEL_CSUM;
+		netdev_feature_set_bits(NETIF_F_GSO_UDP_TUNNEL |
+					NETIF_F_GSO_UDP_TUNNEL_CSUM,
+					&netdev->hw_features);
 
-		netdev->hw_enc_features = NETIF_F_SG | NETIF_F_RXCSUM |
-			NETIF_F_HW_CSUM | NETIF_F_HW_VLAN_CTAG_TX |
-			NETIF_F_HW_VLAN_CTAG_RX | NETIF_F_TSO | NETIF_F_TSO6 |
-			NETIF_F_LRO | NETIF_F_GSO_UDP_TUNNEL |
-			NETIF_F_GSO_UDP_TUNNEL_CSUM;
+		netdev_feature_zero(&netdev->hw_enc_features);
+		netdev_feature_set_bits(NETIF_F_SG | NETIF_F_RXCSUM |
+					NETIF_F_HW_CSUM |
+					NETIF_F_HW_VLAN_CTAG_TX |
+					NETIF_F_HW_VLAN_CTAG_RX |
+					NETIF_F_TSO | NETIF_F_TSO6 |
+					NETIF_F_LRO | NETIF_F_GSO_UDP_TUNNEL |
+					NETIF_F_GSO_UDP_TUNNEL_CSUM,
+					&netdev->hw_enc_features);
 	}
 
 	if (dma64)
-		netdev->hw_features |= NETIF_F_HIGHDMA;
-	netdev->vlan_features = netdev->hw_features &
-				~(NETIF_F_HW_VLAN_CTAG_TX |
-				  NETIF_F_HW_VLAN_CTAG_RX);
-	netdev->features = netdev->hw_features | NETIF_F_HW_VLAN_CTAG_FILTER;
+		netdev_feature_set_bit(NETIF_F_HIGHDMA_BIT,
+				       &netdev->hw_features);
+	netdev_feature_copy(&netdev->vlan_features, netdev->hw_features);
+	netdev_feature_clear_bits(NETIF_F_HW_VLAN_CTAG_TX |
+				  NETIF_F_HW_VLAN_CTAG_RX,
+				  &netdev->vlan_features);
+	netdev_feature_copy(&netdev->features, netdev->hw_features);
+	netdev_feature_set_bit(NETIF_F_HW_VLAN_CTAG_FILTER_BIT,
+			       &netdev->features);
 }
 
 
@@ -3629,8 +3649,9 @@ vmxnet3_probe_device(struct pci_dev *pdev,
 	if (adapter->num_rx_queues > 1 &&
 	    adapter->intr.type == VMXNET3_IT_MSIX) {
 		adapter->rss = true;
-		netdev->hw_features |= NETIF_F_RXHASH;
-		netdev->features |= NETIF_F_RXHASH;
+		netdev_feature_set_bit(NETIF_F_RXHASH_BIT,
+				       &netdev->hw_features);
+		netdev_feature_set_bit(NETIF_F_RXHASH_BIT, &netdev->features);
 		dev_dbg(&pdev->dev, "RSS is enabled.\n");
 	} else {
 		adapter->rss = false;
diff --git a/drivers/net/vmxnet3/vmxnet3_ethtool.c b/drivers/net/vmxnet3/vmxnet3_ethtool.c
index 6a5827c21c63..05e422b48774 100644
--- a/drivers/net/vmxnet3/vmxnet3_ethtool.c
+++ b/drivers/net/vmxnet3/vmxnet3_ethtool.c
@@ -250,8 +250,8 @@ void vmxnet3_fix_features(struct net_device *netdev,
 			  netdev_features_t *features)
 {
 	/* If Rx checksum is disabled, then LRO should also be disabled */
-	if (!(*features & NETIF_F_RXCSUM))
-		*features &= ~NETIF_F_LRO;
+	if (!netdev_feature_test_bit(NETIF_F_RXCSUM_BIT, *features))
+		netdev_feature_clear_bit(NETIF_F_LRO_BIT, features);
 }
 
 void vmxnet3_features_check(struct sk_buff *skb, struct net_device *netdev,
@@ -274,7 +274,9 @@ void vmxnet3_features_check(struct sk_buff *skb, struct net_device *netdev,
 			l4_proto = ipv6_hdr(skb)->nexthdr;
 			break;
 		default:
-			*features &= ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
+			netdev_feature_clear_bits(NETIF_F_CSUM_MASK |
+						  NETIF_F_GSO_MASK,
+						  features);
 			return;
 		}
 
@@ -286,12 +288,16 @@ void vmxnet3_features_check(struct sk_buff *skb, struct net_device *netdev,
 			if (port != GENEVE_UDP_PORT &&
 			    port != IANA_VXLAN_UDP_PORT &&
 			    port != VXLAN_UDP_PORT) {
-				*features &= ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
+				netdev_feature_clear_bits(NETIF_F_CSUM_MASK |
+							  NETIF_F_GSO_MASK,
+							  features);
 				return;
 			}
 			break;
 		default:
-			*features &= ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
+			netdev_feature_clear_bits(NETIF_F_CSUM_MASK |
+						  NETIF_F_GSO_MASK,
+						  features);
 			return;
 		}
 	}
@@ -302,11 +308,14 @@ static void vmxnet3_enable_encap_offloads(struct net_device *netdev)
 	struct vmxnet3_adapter *adapter = netdev_priv(netdev);
 
 	if (VMXNET3_VERSION_GE_4(adapter)) {
-		netdev->hw_enc_features |= NETIF_F_SG | NETIF_F_RXCSUM |
-			NETIF_F_HW_CSUM | NETIF_F_HW_VLAN_CTAG_TX |
-			NETIF_F_HW_VLAN_CTAG_RX | NETIF_F_TSO | NETIF_F_TSO6 |
-			NETIF_F_LRO | NETIF_F_GSO_UDP_TUNNEL |
-			NETIF_F_GSO_UDP_TUNNEL_CSUM;
+		netdev_feature_set_bits(NETIF_F_SG | NETIF_F_RXCSUM |
+					NETIF_F_HW_CSUM |
+					NETIF_F_HW_VLAN_CTAG_TX |
+					NETIF_F_HW_VLAN_CTAG_RX |
+					NETIF_F_TSO | NETIF_F_TSO6 |
+					NETIF_F_LRO | NETIF_F_GSO_UDP_TUNNEL |
+					NETIF_F_GSO_UDP_TUNNEL_CSUM,
+					&netdev->hw_enc_features);
 	}
 }
 
@@ -315,26 +324,33 @@ static void vmxnet3_disable_encap_offloads(struct net_device *netdev)
 	struct vmxnet3_adapter *adapter = netdev_priv(netdev);
 
 	if (VMXNET3_VERSION_GE_4(adapter)) {
-		netdev->hw_enc_features &= ~(NETIF_F_SG | NETIF_F_RXCSUM |
-			NETIF_F_HW_CSUM | NETIF_F_HW_VLAN_CTAG_TX |
-			NETIF_F_HW_VLAN_CTAG_RX | NETIF_F_TSO | NETIF_F_TSO6 |
-			NETIF_F_LRO | NETIF_F_GSO_UDP_TUNNEL |
-			NETIF_F_GSO_UDP_TUNNEL_CSUM);
+		netdev_feature_clear_bits(NETIF_F_SG | NETIF_F_RXCSUM |
+					  NETIF_F_HW_CSUM |
+					  NETIF_F_HW_VLAN_CTAG_TX |
+					  NETIF_F_HW_VLAN_CTAG_RX |
+					  NETIF_F_TSO | NETIF_F_TSO6 |
+					  NETIF_F_LRO | NETIF_F_GSO_UDP_TUNNEL |
+					  NETIF_F_GSO_UDP_TUNNEL_CSUM,
+					  &netdev->hw_enc_features);
 	}
 }
 
 int vmxnet3_set_features(struct net_device *netdev, netdev_features_t features)
 {
 	struct vmxnet3_adapter *adapter = netdev_priv(netdev);
+	u64 tun_offload_mask = NETIF_F_GSO_UDP_TUNNEL |
+				NETIF_F_GSO_UDP_TUNNEL_CSUM;
+	netdev_features_t changed;
 	unsigned long flags;
-	netdev_features_t changed = features ^ netdev->features;
-	netdev_features_t tun_offload_mask = NETIF_F_GSO_UDP_TUNNEL |
-					     NETIF_F_GSO_UDP_TUNNEL_CSUM;
-	u8 udp_tun_enabled = (netdev->features & tun_offload_mask) != 0;
-
-	if (changed & (NETIF_F_RXCSUM | NETIF_F_LRO |
-		       NETIF_F_HW_VLAN_CTAG_RX | tun_offload_mask)) {
-		if (features & NETIF_F_RXCSUM)
+	u8 udp_tun_enabled;
+
+	netdev_feature_xor(&changed, netdev->features, features);
+	udp_tun_enabled = netdev_feature_test_bits(tun_offload_mask,
+						   netdev->features);
+	if (netdev_feature_test_bits(NETIF_F_RXCSUM | NETIF_F_LRO |
+				     NETIF_F_HW_VLAN_CTAG_RX |
+				     tun_offload_mask, changed)) {
+		if (netdev_feature_test_bit(NETIF_F_RXCSUM_BIT, features))
 			adapter->shared->devRead.misc.uptFeatures |=
 			UPT1_F_RXCSUM;
 		else
@@ -342,25 +358,28 @@ int vmxnet3_set_features(struct net_device *netdev, netdev_features_t features)
 			~UPT1_F_RXCSUM;
 
 		/* update hardware LRO capability accordingly */
-		if (features & NETIF_F_LRO)
+		if (netdev_feature_test_bit(NETIF_F_LRO_BIT, features))
 			adapter->shared->devRead.misc.uptFeatures |=
 							UPT1_F_LRO;
 		else
 			adapter->shared->devRead.misc.uptFeatures &=
 							~UPT1_F_LRO;
 
-		if (features & NETIF_F_HW_VLAN_CTAG_RX)
+		if (netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_RX_BIT,
+					    features))
 			adapter->shared->devRead.misc.uptFeatures |=
 			UPT1_F_RXVLAN;
 		else
 			adapter->shared->devRead.misc.uptFeatures &=
 			~UPT1_F_RXVLAN;
 
-		if ((features & tun_offload_mask) != 0 && !udp_tun_enabled) {
+		if (netdev_feature_test_bits(tun_offload_mask, features) &&
+		    !udp_tun_enabled) {
 			vmxnet3_enable_encap_offloads(netdev);
 			adapter->shared->devRead.misc.uptFeatures |=
 			UPT1_F_RXINNEROFLD;
-		} else if ((features & tun_offload_mask) == 0 &&
+		} else if (!netdev_feature_test_bits(tun_offload_mask,
+						     features) &&
 			   udp_tun_enabled) {
 			vmxnet3_disable_encap_offloads(netdev);
 			adapter->shared->devRead.misc.uptFeatures &=
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 043/167] net_failover: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (41 preceding siblings ...)
  2021-09-29 15:51 ` [RFCv2 net-next 042/167] vmxnet3: " Jian Shen
@ 2021-09-29 15:51 ` Jian Shen
  2021-09-29 15:51 ` [RFCv2 net-next 044/167] pktgen: " Jian Shen
                   ` (127 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:51 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 drivers/net/net_failover.c | 57 +++++++++++++++++++++++++-------------
 1 file changed, 37 insertions(+), 20 deletions(-)

diff --git a/drivers/net/net_failover.c b/drivers/net/net_failover.c
index 878cad216aaf..704e691530f7 100644
--- a/drivers/net/net_failover.c
+++ b/drivers/net/net_failover.c
@@ -379,23 +379,33 @@ static rx_handler_result_t net_failover_handle_frame(struct sk_buff **pskb)
 
 static void net_failover_compute_features(struct net_device *dev)
 {
-	netdev_features_t vlan_features = FAILOVER_VLAN_FEATURES &
-					  NETIF_F_ALL_FOR_ALL;
-	netdev_features_t enc_features  = FAILOVER_ENC_FEATURES;
 	unsigned short max_hard_header_len = ETH_HLEN;
 	unsigned int dst_release_flag = IFF_XMIT_DST_RELEASE |
 					IFF_XMIT_DST_RELEASE_PERM;
 	struct net_failover_info *nfo_info = netdev_priv(dev);
 	struct net_device *primary_dev, *standby_dev;
+	netdev_features_t vlan_features;
+	netdev_features_t enc_features;
+	netdev_features_t vlan_mask;
+	netdev_features_t enc_mask;
+
+	netdev_feature_zero(&vlan_features);
+	netdev_feature_set_bits(FAILOVER_VLAN_FEATURES & NETIF_F_ALL_FOR_ALL,
+				&vlan_features);
+	netdev_feature_zero(&enc_features);
+	netdev_feature_set_bits(FAILOVER_ENC_FEATURES, &enc_features);
+	netdev_feature_zero(&vlan_mask);
+	netdev_feature_set_bits(FAILOVER_VLAN_FEATURES, &vlan_mask);
+	netdev_feature_copy(&enc_mask, enc_features);
 
 	primary_dev = rcu_dereference(nfo_info->primary_dev);
 	if (primary_dev) {
 		netdev_increment_features(&vlan_features, vlan_features,
 					  primary_dev->vlan_features,
-					  FAILOVER_VLAN_FEATURES);
+					  vlan_mask);
 		netdev_increment_features(&enc_features, enc_features,
 					  primary_dev->hw_enc_features,
-					  FAILOVER_ENC_FEATURES);
+					  enc_features);
 
 		dst_release_flag &= primary_dev->priv_flags;
 		if (primary_dev->hard_header_len > max_hard_header_len)
@@ -406,18 +416,19 @@ static void net_failover_compute_features(struct net_device *dev)
 	if (standby_dev) {
 		netdev_increment_features(&vlan_features, vlan_features,
 					  standby_dev->vlan_features,
-					  FAILOVER_VLAN_FEATURES);
+					  vlan_mask);
 		netdev_increment_features(&enc_features, enc_features,
 					  standby_dev->hw_enc_features,
-					  FAILOVER_ENC_FEATURES);
+					  enc_features);
 
 		dst_release_flag &= standby_dev->priv_flags;
 		if (standby_dev->hard_header_len > max_hard_header_len)
 			max_hard_header_len = standby_dev->hard_header_len;
 	}
 
-	dev->vlan_features = vlan_features;
-	dev->hw_enc_features = enc_features | NETIF_F_GSO_ENCAP_ALL;
+	netdev_feature_copy(&dev->vlan_features, vlan_features);
+	netdev_feature_copy(&dev->hw_enc_features, enc_features);
+	netdev_feature_set_bits(NETIF_F_GSO_ENCAP_ALL, &dev->hw_enc_features);
 	dev->hard_header_len = max_hard_header_len;
 
 	dev->priv_flags &= ~IFF_XMIT_DST_RELEASE;
@@ -481,7 +492,8 @@ static int net_failover_slave_pre_register(struct net_device *slave_dev,
 				  !dev_is_pci(slave_dev->dev.parent)))
 		return -EINVAL;
 
-	if (failover_dev->features & NETIF_F_VLAN_CHALLENGED &&
+	if (netdev_feature_test_bit(NETIF_F_VLAN_CHALLENGED_BIT,
+				    failover_dev->features) &&
 	    vlan_uses_dev(failover_dev)) {
 		netdev_err(failover_dev, "Device %s is VLAN challenged and failover device has VLAN set up\n",
 			   failover_dev->name);
@@ -731,18 +743,23 @@ struct failover *net_failover_create(struct net_device *standby_dev)
 				       IFF_TX_SKB_SHARING);
 
 	/* don't acquire failover netdev's netif_tx_lock when transmitting */
-	failover_dev->features |= NETIF_F_LLTX;
+	netdev_feature_set_bit(NETIF_F_LLTX_BIT, &failover_dev->features);
 
 	/* Don't allow failover devices to change network namespaces. */
-	failover_dev->features |= NETIF_F_NETNS_LOCAL;
-
-	failover_dev->hw_features = FAILOVER_VLAN_FEATURES |
-				    NETIF_F_HW_VLAN_CTAG_TX |
-				    NETIF_F_HW_VLAN_CTAG_RX |
-				    NETIF_F_HW_VLAN_CTAG_FILTER;
-
-	failover_dev->hw_features |= NETIF_F_GSO_ENCAP_ALL;
-	failover_dev->features |= failover_dev->hw_features;
+	netdev_feature_set_bit(NETIF_F_NETNS_LOCAL_BIT,
+			       &failover_dev->features);
+
+	netdev_feature_zero(&failover_dev->hw_features);
+	netdev_feature_set_bits(FAILOVER_VLAN_FEATURES |
+				NETIF_F_HW_VLAN_CTAG_TX |
+				NETIF_F_HW_VLAN_CTAG_RX |
+				NETIF_F_HW_VLAN_CTAG_FILTER,
+				&failover_dev->hw_features);
+
+	netdev_feature_set_bits(NETIF_F_GSO_ENCAP_ALL,
+				&failover_dev->hw_features);
+	netdev_feature_or(&failover_dev->features, failover_dev->features,
+			  failover_dev->hw_features);
 
 	memcpy(failover_dev->dev_addr, standby_dev->dev_addr,
 	       failover_dev->addr_len);
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 044/167] pktgen: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (42 preceding siblings ...)
  2021-09-29 15:51 ` [RFCv2 net-next 043/167] net_failover: " Jian Shen
@ 2021-09-29 15:51 ` Jian Shen
  2021-09-29 15:51 ` [RFCv2 net-next 045/167] net: sched: " Jian Shen
                   ` (126 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:51 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 net/core/pktgen.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/net/core/pktgen.c b/net/core/pktgen.c
index a3d74e2704c4..4a4a866ff97d 100644
--- a/net/core/pktgen.c
+++ b/net/core/pktgen.c
@@ -2961,7 +2961,8 @@ static struct sk_buff *fill_packet_ipv4(struct net_device *odev,
 
 	if (!(pkt_dev->flags & F_UDPCSUM)) {
 		skb->ip_summed = CHECKSUM_NONE;
-	} else if (odev->features & (NETIF_F_HW_CSUM | NETIF_F_IP_CSUM)) {
+	} else if (netdev_feature_test_bits(NETIF_F_HW_CSUM | NETIF_F_IP_CSUM,
+					    odev->features)) {
 		skb->ip_summed = CHECKSUM_PARTIAL;
 		skb->csum = 0;
 		udp4_hwcsum(skb, iph->saddr, iph->daddr);
@@ -3096,7 +3097,8 @@ static struct sk_buff *fill_packet_ipv6(struct net_device *odev,
 
 	if (!(pkt_dev->flags & F_UDPCSUM)) {
 		skb->ip_summed = CHECKSUM_NONE;
-	} else if (odev->features & (NETIF_F_HW_CSUM | NETIF_F_IPV6_CSUM)) {
+	} else if (netdev_feature_test_bits(NETIF_F_HW_CSUM | NETIF_F_IPV6_CSUM,
+					    odev->features)) {
 		skb->ip_summed = CHECKSUM_PARTIAL;
 		skb->csum_start = skb_transport_header(skb) - skb->head;
 		skb->csum_offset = offsetof(struct udphdr, check);
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 045/167] net: sched: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (43 preceding siblings ...)
  2021-09-29 15:51 ` [RFCv2 net-next 044/167] pktgen: " Jian Shen
@ 2021-09-29 15:51 ` Jian Shen
  2021-09-29 15:51 ` [RFCv2 net-next 046/167] netdevsim: " Jian Shen
                   ` (125 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:51 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 include/net/pkt_cls.h  | 2 +-
 net/sched/sch_cake.c   | 3 ++-
 net/sched/sch_netem.c  | 3 ++-
 net/sched/sch_taprio.c | 3 ++-
 net/sched/sch_tbf.c    | 3 ++-
 5 files changed, 9 insertions(+), 5 deletions(-)

diff --git a/include/net/pkt_cls.h b/include/net/pkt_cls.h
index 83a6d0792180..228881d40dd8 100644
--- a/include/net/pkt_cls.h
+++ b/include/net/pkt_cls.h
@@ -636,7 +636,7 @@ struct tc_cls_u32_offload {
 
 static inline bool tc_can_offload(const struct net_device *dev)
 {
-	return dev->features & NETIF_F_HW_TC;
+	return netdev_feature_test_bit(NETIF_F_HW_TC_BIT, dev->features);
 }
 
 static inline bool tc_can_offload_extack(const struct net_device *dev,
diff --git a/net/sched/sch_cake.c b/net/sched/sch_cake.c
index e650ec5dc791..5d783cd72290 100644
--- a/net/sched/sch_cake.c
+++ b/net/sched/sch_cake.c
@@ -1744,7 +1744,8 @@ static s32 cake_enqueue(struct sk_buff *skb, struct Qdisc *sch,
 		unsigned int slen = 0, numsegs = 0;
 
 		netif_skb_features(skb, &features);
-		segs = skb_gso_segment(skb, features & ~NETIF_F_GSO_MASK);
+		netdev_feature_clear_bits(NETIF_F_GSO_MASK, &features);
+		segs = skb_gso_segment(skb, features);
 		if (IS_ERR_OR_NULL(segs))
 			return qdisc_drop(skb, sch, to_free);
 
diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c
index 414d57e017b9..4aabeb206777 100644
--- a/net/sched/sch_netem.c
+++ b/net/sched/sch_netem.c
@@ -416,7 +416,8 @@ static struct sk_buff *netem_segment(struct sk_buff *skb, struct Qdisc *sch,
 	netdev_features_t features;
 
 	netif_skb_features(skb, &features);
-	segs = skb_gso_segment(skb, features & ~NETIF_F_GSO_MASK);
+	netdev_feature_clear_bits(NETIF_F_GSO_MASK, &features);
+	segs = skb_gso_segment(skb, features);
 
 	if (IS_ERR_OR_NULL(segs)) {
 		qdisc_drop(skb, sch, to_free);
diff --git a/net/sched/sch_taprio.c b/net/sched/sch_taprio.c
index d7fe4a2cc14f..e42deaf723b1 100644
--- a/net/sched/sch_taprio.c
+++ b/net/sched/sch_taprio.c
@@ -458,7 +458,8 @@ static int taprio_enqueue(struct sk_buff *skb, struct Qdisc *sch,
 		int ret;
 
 		netif_skb_features(skb, &features);
-		segs = skb_gso_segment(skb, features & ~NETIF_F_GSO_MASK);
+		netdev_feature_clear_bits(NETIF_F_GSO_MASK, &features);
+		segs = skb_gso_segment(skb, features);
 		if (IS_ERR_OR_NULL(segs))
 			return qdisc_drop(skb, sch, to_free);
 
diff --git a/net/sched/sch_tbf.c b/net/sched/sch_tbf.c
index 99e6d7265e7f..8ac786f467a0 100644
--- a/net/sched/sch_tbf.c
+++ b/net/sched/sch_tbf.c
@@ -197,7 +197,8 @@ static int tbf_segment(struct sk_buff *skb, struct Qdisc *sch,
 	int ret, nb;
 
 	netif_skb_features(skb, &features);
-	segs = skb_gso_segment(skb, features & ~NETIF_F_GSO_MASK);
+	netdev_feature_clear_bits(NETIF_F_GSO_MASK, &features);
+	segs = skb_gso_segment(skb, features);
 
 	if (IS_ERR_OR_NULL(segs))
 		return qdisc_drop(skb, sch, to_free);
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 046/167] netdevsim: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (44 preceding siblings ...)
  2021-09-29 15:51 ` [RFCv2 net-next 045/167] net: sched: " Jian Shen
@ 2021-09-29 15:51 ` Jian Shen
  2021-09-29 15:51 ` [RFCv2 net-next 047/167] virtio_net: " Jian Shen
                   ` (124 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:51 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 drivers/net/netdevsim/ipsec.c  |  4 ++--
 drivers/net/netdevsim/netdev.c | 15 ++++++++-------
 2 files changed, 10 insertions(+), 9 deletions(-)

diff --git a/drivers/net/netdevsim/ipsec.c b/drivers/net/netdevsim/ipsec.c
index b80ed2ffd45e..3b77c376e34f 100644
--- a/drivers/net/netdevsim/ipsec.c
+++ b/drivers/net/netdevsim/ipsec.c
@@ -280,8 +280,8 @@ void nsim_ipsec_init(struct netdevsim *ns)
 				 NETIF_F_HW_ESP_TX_CSUM | \
 				 NETIF_F_GSO_ESP)
 
-	ns->netdev->features |= NSIM_ESP_FEATURES;
-	ns->netdev->hw_enc_features |= NSIM_ESP_FEATURES;
+	netdev_feature_set_bits(NSIM_ESP_FEATURES, &ns->netdev->features);
+	netdev_feature_set_bits(NSIM_ESP_FEATURES, &ns->netdev->hw_enc_features);
 
 	ns->ipsec.pfile = debugfs_create_file("ipsec", 0400,
 					      ns->nsim_dev_port->ddir, ns,
diff --git a/drivers/net/netdevsim/netdev.c b/drivers/net/netdevsim/netdev.c
index 50572e0f1f52..895b5487bbee 100644
--- a/drivers/net/netdevsim/netdev.c
+++ b/drivers/net/netdevsim/netdev.c
@@ -232,7 +232,8 @@ nsim_set_features(struct net_device *dev, netdev_features_t features)
 {
 	struct netdevsim *ns = netdev_priv(dev);
 
-	if ((dev->features & NETIF_F_HW_TC) > (features & NETIF_F_HW_TC))
+	if (netdev_feature_test_bit(NETIF_F_HW_TC_BIT, dev->features) &&
+	    !netdev_feature_test_bit(NETIF_F_HW_TC_BIT, features))
 		return nsim_bpf_disable_tc(ns);
 
 	return 0;
@@ -288,12 +289,12 @@ static void nsim_setup(struct net_device *dev)
 	dev->flags &= ~IFF_MULTICAST;
 	dev->priv_flags |= IFF_LIVE_ADDR_CHANGE |
 			   IFF_NO_QUEUE;
-	dev->features |= NETIF_F_HIGHDMA |
-			 NETIF_F_SG |
-			 NETIF_F_FRAGLIST |
-			 NETIF_F_HW_CSUM |
-			 NETIF_F_TSO;
-	dev->hw_features |= NETIF_F_HW_TC;
+	netdev_feature_set_bits(NETIF_F_HIGHDMA |
+				NETIF_F_SG |
+				NETIF_F_FRAGLIST |
+				NETIF_F_HW_CSUM |
+				NETIF_F_TSO, &dev->features);
+	netdev_feature_set_bit(NETIF_F_HW_TC_BIT, &dev->hw_features);
 	dev->max_mtu = ETH_MAX_MTU;
 }
 
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 047/167] virtio_net: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (45 preceding siblings ...)
  2021-09-29 15:51 ` [RFCv2 net-next 046/167] netdevsim: " Jian Shen
@ 2021-09-29 15:51 ` Jian Shen
  2021-09-29 15:51 ` [RFCv2 net-next 048/167] net: ipv4: " Jian Shen
                   ` (123 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:51 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 drivers/net/virtio_net.c | 50 +++++++++++++++++++++++++---------------
 1 file changed, 32 insertions(+), 18 deletions(-)

diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index 2ed49884565f..5307bc4f3c99 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -2648,11 +2648,12 @@ static int virtnet_set_features(struct net_device *dev,
 	u64 offloads;
 	int err;
 
-	if ((dev->features ^ features) & NETIF_F_GRO_HW) {
+	if (netdev_feature_test_bit(NETIF_F_GRO_HW_BIT, dev->features) !=
+	    netdev_feature_test_bit(NETIF_F_GRO_HW_BIT, features)) {
 		if (vi->xdp_enabled)
 			return -EBUSY;
 
-		if (features & NETIF_F_GRO_HW)
+		if (netdev_feature_test_bit(NETIF_F_GRO_HW_BIT, features))
 			offloads = vi->guest_offloads_capable;
 		else
 			offloads = vi->guest_offloads_capable &
@@ -2907,7 +2908,8 @@ static int virtnet_find_vqs(struct virtnet_info *vi)
 	if (vi->has_cvq) {
 		vi->cvq = vqs[total_vqs - 1];
 		if (virtio_has_feature(vi->vdev, VIRTIO_NET_F_CTRL_VLAN))
-			vi->dev->features |= NETIF_F_HW_VLAN_CTAG_FILTER;
+			netdev_feature_set_bit(NETIF_F_HW_VLAN_CTAG_FILTER_BIT,
+					       &vi->dev->features);
 	}
 
 	for (i = 0; i < vi->max_queue_pairs; i++) {
@@ -3118,7 +3120,8 @@ static int virtnet_probe(struct virtio_device *vdev)
 	dev->priv_flags |= IFF_UNICAST_FLT | IFF_LIVE_ADDR_CHANGE |
 			   IFF_TX_SKB_NO_LINEAR;
 	dev->netdev_ops = &virtnet_netdev;
-	dev->features = NETIF_F_HIGHDMA;
+	netdev_feature_zero(&dev->features);
+	netdev_feature_set_bit(NETIF_F_HIGHDMA_BIT, &dev->features);
 
 	dev->ethtool_ops = &virtnet_ethtool_ops;
 	SET_NETDEV_DEV(dev, &vdev->dev);
@@ -3126,37 +3129,48 @@ static int virtnet_probe(struct virtio_device *vdev)
 	/* Do we support "hardware" checksums? */
 	if (virtio_has_feature(vdev, VIRTIO_NET_F_CSUM)) {
 		/* This opens up the world of extra features. */
-		dev->hw_features |= NETIF_F_HW_CSUM | NETIF_F_SG;
+		netdev_feature_set_bits(NETIF_F_HW_CSUM | NETIF_F_SG,
+					&dev->hw_features);
 		if (csum)
-			dev->features |= NETIF_F_HW_CSUM | NETIF_F_SG;
+			netdev_feature_set_bits(NETIF_F_HW_CSUM | NETIF_F_SG,
+						&dev->features);
 
 		if (virtio_has_feature(vdev, VIRTIO_NET_F_GSO)) {
-			dev->hw_features |= NETIF_F_TSO
-				| NETIF_F_TSO_ECN | NETIF_F_TSO6;
+			netdev_feature_set_bits(NETIF_F_TSO | NETIF_F_TSO_ECN |
+						NETIF_F_TSO6,
+						&dev->hw_features);
 		}
 		/* Individual feature bits: what can host handle? */
 		if (virtio_has_feature(vdev, VIRTIO_NET_F_HOST_TSO4))
-			dev->hw_features |= NETIF_F_TSO;
+			netdev_feature_set_bit(NETIF_F_TSO_BIT,
+					       &dev->hw_features);
 		if (virtio_has_feature(vdev, VIRTIO_NET_F_HOST_TSO6))
-			dev->hw_features |= NETIF_F_TSO6;
+			netdev_feature_set_bit(NETIF_F_TSO6_BIT,
+					       &dev->hw_features);
 		if (virtio_has_feature(vdev, VIRTIO_NET_F_HOST_ECN))
-			dev->hw_features |= NETIF_F_TSO_ECN;
+			netdev_feature_set_bit(NETIF_F_TSO_ECN_BIT,
+					       &dev->hw_features);
 
-		dev->features |= NETIF_F_GSO_ROBUST;
+		netdev_feature_set_bit(NETIF_F_GSO_ROBUST_BIT, &dev->features);
 
-		if (gso)
-			dev->features |= dev->hw_features & NETIF_F_ALL_TSO;
+		if (gso) {
+			netdev_features_t tmp;
+
+			netdev_feature_copy(&tmp, dev->hw_features);
+			netdev_feature_and_bits(NETIF_F_ALL_TSO, &tmp);
+			netdev_feature_or(&dev->features, dev->features, tmp);
+		}
 		/* (!csum && gso) case will be fixed by register_netdev() */
 	}
 	if (virtio_has_feature(vdev, VIRTIO_NET_F_GUEST_CSUM))
-		dev->features |= NETIF_F_RXCSUM;
+		netdev_feature_set_bit(NETIF_F_RXCSUM_BIT, &dev->features);
 	if (virtio_has_feature(vdev, VIRTIO_NET_F_GUEST_TSO4) ||
 	    virtio_has_feature(vdev, VIRTIO_NET_F_GUEST_TSO6))
-		dev->features |= NETIF_F_GRO_HW;
+		netdev_feature_set_bit(NETIF_F_GRO_HW_BIT, &dev->features);
 	if (virtio_has_feature(vdev, VIRTIO_NET_F_CTRL_GUEST_OFFLOADS))
-		dev->hw_features |= NETIF_F_GRO_HW;
+		netdev_feature_set_bit(NETIF_F_GRO_HW_BIT, &dev->hw_features);
 
-	dev->vlan_features = dev->features;
+	netdev_feature_copy(&dev->vlan_features, dev->features);
 
 	/* MTU range: 68 - 65535 */
 	dev->min_mtu = MIN_MTU;
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 048/167] net: ipv4: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (46 preceding siblings ...)
  2021-09-29 15:51 ` [RFCv2 net-next 047/167] virtio_net: " Jian Shen
@ 2021-09-29 15:51 ` Jian Shen
  2021-09-29 15:51 ` [RFCv2 net-next 049/167] net: ipv6: " Jian Shen
                   ` (122 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:51 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 include/net/udp.h        |  8 ++++++--
 include/net/udp_tunnel.h | 12 ++++++++----
 net/ipv4/af_inet.c       |  3 ++-
 net/ipv4/esp4_offload.c  | 28 +++++++++++++++++-----------
 net/ipv4/gre_offload.c   |  8 +++++---
 net/ipv4/ip_gre.c        | 36 ++++++++++++++++++++++--------------
 net/ipv4/ip_output.c     | 23 +++++++++++++++--------
 net/ipv4/ip_tunnel.c     |  4 +++-
 net/ipv4/ip_vti.c        |  2 +-
 net/ipv4/ipip.c          |  6 +++---
 net/ipv4/ipmr.c          |  2 +-
 net/ipv4/tcp_offload.c   |  5 ++++-
 net/ipv4/udp_offload.c   | 28 +++++++++++++++++-----------
 13 files changed, 104 insertions(+), 61 deletions(-)

diff --git a/include/net/udp.h b/include/net/udp.h
index 360df454356c..3855b9775cbf 100644
--- a/include/net/udp.h
+++ b/include/net/udp.h
@@ -481,14 +481,18 @@ void udpv6_encap_enable(void);
 static inline struct sk_buff *udp_rcv_segment(struct sock *sk,
 					      struct sk_buff *skb, bool ipv4)
 {
-	netdev_features_t features = NETIF_F_SG;
+	netdev_features_t features;
 	struct sk_buff *segs;
 
+	netdev_feature_zero(&features);
+	netdev_feature_set_bit(NETIF_F_SG_BIT, &features);
+
 	/* Avoid csum recalculation by skb_segment unless userspace explicitly
 	 * asks for the final checksum values
 	 */
 	if (!inet_get_convert_csum(sk))
-		features |= NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM;
+		netdev_feature_set_bits(NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM,
+					&features);
 
 	/* UDP segmentation expects packets of type CHECKSUM_PARTIAL or
 	 * CHECKSUM_NONE in __udp_gso_segment. UDP GRO indeed builds partial
diff --git a/include/net/udp_tunnel.h b/include/net/udp_tunnel.h
index afc7ce713657..68c1a540a733 100644
--- a/include/net/udp_tunnel.h
+++ b/include/net/udp_tunnel.h
@@ -129,7 +129,8 @@ void udp_tunnel_notify_del_rx_port(struct socket *sock, unsigned short type);
 static inline void udp_tunnel_get_rx_info(struct net_device *dev)
 {
 	ASSERT_RTNL();
-	if (!(dev->features & NETIF_F_RX_UDP_TUNNEL_PORT))
+	if (!netdev_feature_test_bit(NETIF_F_RX_UDP_TUNNEL_PORT_BIT,
+				     dev->features))
 		return;
 	call_netdevice_notifiers(NETDEV_UDP_TUNNEL_PUSH_INFO, dev);
 }
@@ -137,7 +138,8 @@ static inline void udp_tunnel_get_rx_info(struct net_device *dev)
 static inline void udp_tunnel_drop_rx_info(struct net_device *dev)
 {
 	ASSERT_RTNL();
-	if (!(dev->features & NETIF_F_RX_UDP_TUNNEL_PORT))
+	if (!netdev_feature_test_bit(NETIF_F_RX_UDP_TUNNEL_PORT_BIT,
+				     dev->features))
 		return;
 	call_netdevice_notifiers(NETDEV_UDP_TUNNEL_DROP_INFO, dev);
 }
@@ -326,7 +328,8 @@ udp_tunnel_nic_set_port_priv(struct net_device *dev, unsigned int table,
 static inline void
 udp_tunnel_nic_add_port(struct net_device *dev, struct udp_tunnel_info *ti)
 {
-	if (!(dev->features & NETIF_F_RX_UDP_TUNNEL_PORT))
+	if (!netdev_feature_test_bit(NETIF_F_RX_UDP_TUNNEL_PORT_BIT,
+				     dev->features))
 		return;
 	if (udp_tunnel_nic_ops)
 		udp_tunnel_nic_ops->add_port(dev, ti);
@@ -335,7 +338,8 @@ udp_tunnel_nic_add_port(struct net_device *dev, struct udp_tunnel_info *ti)
 static inline void
 udp_tunnel_nic_del_port(struct net_device *dev, struct udp_tunnel_info *ti)
 {
-	if (!(dev->features & NETIF_F_RX_UDP_TUNNEL_PORT))
+	if (!netdev_feature_test_bit(NETIF_F_RX_UDP_TUNNEL_PORT_BIT,
+				     dev->features))
 		return;
 	if (udp_tunnel_nic_ops)
 		udp_tunnel_nic_ops->del_port(dev, ti);
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index 40558033f857..aeff7736f260 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -1357,7 +1357,8 @@ struct sk_buff *inet_gso_segment(struct sk_buff *skb,
 
 	encap = SKB_GSO_CB(skb)->encap_level > 0;
 	if (encap)
-		features &= skb->dev->hw_enc_features;
+		netdev_feature_and(&features, features,
+				   skb->dev->hw_enc_features);
 	SKB_GSO_CB(skb)->encap_level += ihl;
 
 	skb_reset_transport_header(skb);
diff --git a/net/ipv4/esp4_offload.c b/net/ipv4/esp4_offload.c
index 8e4e9aa12130..f7e77a9176d6 100644
--- a/net/ipv4/esp4_offload.c
+++ b/net/ipv4/esp4_offload.c
@@ -189,8 +189,8 @@ static struct sk_buff *esp4_gso_segment(struct sk_buff *skb,
 	struct xfrm_state *x;
 	struct ip_esp_hdr *esph;
 	struct crypto_aead *aead;
-	netdev_features_t esp_features = features;
 	struct xfrm_offload *xo = xfrm_offload(skb);
+	netdev_features_t esp_features;
 	struct sec_path *sp;
 
 	if (!xo)
@@ -214,14 +214,19 @@ static struct sk_buff *esp4_gso_segment(struct sk_buff *skb,
 
 	skb->encap_hdr_csum = 1;
 
-	if ((!(skb->dev->gso_partial_features & NETIF_F_HW_ESP) &&
-	     !(features & NETIF_F_HW_ESP)) || x->xso.dev != skb->dev)
-		esp_features = features & ~(NETIF_F_SG | NETIF_F_CSUM_MASK |
-					    NETIF_F_SCTP_CRC);
-	else if (!(features & NETIF_F_HW_ESP_TX_CSUM) &&
-		 !(skb->dev->gso_partial_features & NETIF_F_HW_ESP_TX_CSUM))
-		esp_features = features & ~(NETIF_F_CSUM_MASK |
-					    NETIF_F_SCTP_CRC);
+	netdev_feature_copy(&esp_features, features);
+	if ((!netdev_feature_test_bit(NETIF_F_HW_ESP_BIT,
+				      skb->dev->gso_partial_features) &&
+	     !netdev_feature_test_bit(NETIF_F_HW_ESP_BIT, features)) ||
+	    x->xso.dev != skb->dev)
+		netdev_feature_clear_bits(NETIF_F_SG | NETIF_F_CSUM_MASK |
+					  NETIF_F_SCTP_CRC, &esp_features);
+	else if (!netdev_feature_test_bit(NETIF_F_HW_ESP_TX_CSUM_BIT,
+					  features) &&
+		 !netdev_feature_test_bit(NETIF_F_HW_ESP_TX_CSUM_BIT,
+					  skb->dev->gso_partial_features))
+		netdev_feature_clear_bits(NETIF_F_CSUM_MASK | NETIF_F_SCTP_CRC,
+					  &esp_features);
 
 	xo->flags |= XFRM_GSO_SEGMENT;
 
@@ -261,8 +266,9 @@ static int esp_xmit(struct xfrm_state *x, struct sk_buff *skb,  netdev_features_
 	if (!xo)
 		return -EINVAL;
 
-	if ((!(features & NETIF_F_HW_ESP) &&
-	     !(skb->dev->gso_partial_features & NETIF_F_HW_ESP)) ||
+	if ((!netdev_feature_test_bit(NETIF_F_HW_ESP_BIT, features) &&
+	     !netdev_feature_test_bit(NETIF_F_HW_ESP_BIT,
+				      skb->dev->gso_partial_features)) ||
 	    x->xso.dev != skb->dev) {
 		xo->flags |= CRYPTO_FALLBACK;
 		hw_offload = false;
diff --git a/net/ipv4/gre_offload.c b/net/ipv4/gre_offload.c
index 1121a9d5fed9..65f22fa749be 100644
--- a/net/ipv4/gre_offload.c
+++ b/net/ipv4/gre_offload.c
@@ -43,14 +43,16 @@ static struct sk_buff *gre_gso_segment(struct sk_buff *skb,
 	need_csum = !!(skb_shinfo(skb)->gso_type & SKB_GSO_GRE_CSUM);
 	skb->encap_hdr_csum = need_csum;
 
-	features &= skb->dev->hw_enc_features;
+	netdev_feature_and(&features, features, skb->dev->hw_enc_features);
+
 	if (need_csum)
-		features &= ~NETIF_F_SCTP_CRC;
+		netdev_feature_clear_bit(NETIF_F_SCTP_CRC_BIT, &features);
 
 	need_ipsec = skb_dst(skb) && dst_xfrm(skb_dst(skb));
 	/* Try to offload checksum if possible */
 	offload_csum = !!(need_csum && !need_ipsec &&
-			  (skb->dev->features & NETIF_F_HW_CSUM));
+			  netdev_feature_test_bit(NETIF_F_HW_CSUM_BIT,
+						  skb->dev->features));
 
 	/* segment inner packet. */
 	segs = skb_mac_gso_segment(skb, features);
diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c
index 0fe6c936dc54..e083ef6a8454 100644
--- a/net/ipv4/ip_gre.c
+++ b/net/ipv4/ip_gre.c
@@ -767,16 +767,22 @@ static void ipgre_link_update(struct net_device *dev, bool set_mtu)
 	if (!(tunnel->parms.o_flags & TUNNEL_SEQ)) {
 		if (!(tunnel->parms.o_flags & TUNNEL_CSUM) ||
 		    tunnel->encap.type == TUNNEL_ENCAP_NONE) {
-			dev->features |= NETIF_F_GSO_SOFTWARE;
-			dev->hw_features |= NETIF_F_GSO_SOFTWARE;
+			netdev_feature_set_bits(NETIF_F_GSO_SOFTWARE,
+						&dev->features);
+			netdev_feature_set_bits(NETIF_F_GSO_SOFTWARE,
+						&dev->hw_features);
 		} else {
-			dev->features &= ~NETIF_F_GSO_SOFTWARE;
-			dev->hw_features &= ~NETIF_F_GSO_SOFTWARE;
+			netdev_feature_clear_bits(NETIF_F_GSO_SOFTWARE,
+						  &dev->features);
+			netdev_feature_clear_bits(NETIF_F_GSO_SOFTWARE,
+						  &dev->hw_features);
 		}
-		dev->features |= NETIF_F_LLTX;
+		netdev_feature_set_bit(NETIF_F_LLTX_BIT, &dev->features);
 	} else {
-		dev->hw_features &= ~NETIF_F_GSO_SOFTWARE;
-		dev->features &= ~(NETIF_F_LLTX | NETIF_F_GSO_SOFTWARE);
+		netdev_feature_clear_bits(NETIF_F_GSO_SOFTWARE,
+					  &dev->hw_features);
+		netdev_feature_clear_bits(NETIF_F_LLTX | NETIF_F_GSO_SOFTWARE,
+					  &dev->features);
 	}
 }
 
@@ -958,8 +964,8 @@ static void __gre_tunnel_init(struct net_device *dev)
 	tunnel->hlen = tunnel->tun_hlen + tunnel->encap_hlen;
 	dev->needed_headroom = tunnel->hlen + sizeof(tunnel->parms.iph);
 
-	dev->features		|= GRE_FEATURES;
-	dev->hw_features	|= GRE_FEATURES;
+	netdev_feature_set_bits(GRE_FEATURES, &dev->features);
+	netdev_feature_set_bits(GRE_FEATURES, &dev->hw_features);
 
 	if (!(tunnel->parms.o_flags & TUNNEL_SEQ)) {
 		/* TCP offload with GRE SEQ is not supported, nor
@@ -968,14 +974,16 @@ static void __gre_tunnel_init(struct net_device *dev)
 		 */
 		if (!(tunnel->parms.o_flags & TUNNEL_CSUM) ||
 		    (tunnel->encap.type == TUNNEL_ENCAP_NONE)) {
-			dev->features    |= NETIF_F_GSO_SOFTWARE;
-			dev->hw_features |= NETIF_F_GSO_SOFTWARE;
+			netdev_feature_set_bits(NETIF_F_GSO_SOFTWARE,
+						&dev->features);
+			netdev_feature_set_bits(NETIF_F_GSO_SOFTWARE,
+						&dev->hw_features);
 		}
 
 		/* Can use a lockless transmit, unless we generate
 		 * output sequences
 		 */
-		dev->features |= NETIF_F_LLTX;
+		netdev_feature_set_bit(NETIF_F_LLTX_BIT, &dev->features);
 	}
 }
 
@@ -1303,8 +1311,8 @@ static int erspan_tunnel_init(struct net_device *dev)
 	tunnel->hlen = tunnel->tun_hlen + tunnel->encap_hlen +
 		       erspan_hdr_len(tunnel->erspan_ver);
 
-	dev->features		|= GRE_FEATURES;
-	dev->hw_features	|= GRE_FEATURES;
+	netdev_feature_set_bits(GRE_FEATURES, &dev->features);
+	netdev_feature_set_bits(GRE_FEATURES, &dev->hw_features);
 	dev->priv_flags		|= IFF_LIVE_ADDR_CHANGE;
 	netif_keep_dst(dev);
 
diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c
index 8d552f1b7f62..60a693c7ae79 100644
--- a/net/ipv4/ip_output.c
+++ b/net/ipv4/ip_output.c
@@ -257,7 +257,8 @@ static int ip_finish_output_gso(struct net *net, struct sock *sk,
 	 */
 	netif_skb_features(skb, &features);
 	BUILD_BUG_ON(sizeof(*IPCB(skb)) > SKB_GSO_CB_OFFSET);
-	segs = skb_gso_segment(skb, features & ~NETIF_F_GSO_MASK);
+	netdev_feature_clear_bits(NETIF_F_GSO_MASK, &features);
+	segs = skb_gso_segment(skb, features);
 	if (IS_ERR_OR_NULL(segs)) {
 		kfree_skb(skb);
 		return -ENOMEM;
@@ -996,9 +997,11 @@ static int __ip_append_data(struct sock *sk,
 	 */
 	if (transhdrlen &&
 	    length + fragheaderlen <= mtu &&
-	    rt->dst.dev->features & (NETIF_F_HW_CSUM | NETIF_F_IP_CSUM) &&
+	    netdev_feature_test_bits(NETIF_F_HW_CSUM | NETIF_F_IP_CSUM,
+				     rt->dst.dev->features) &&
 	    (!(flags & MSG_MORE) || cork->gso_size) &&
-	    (!exthdrlen || (rt->dst.dev->features & NETIF_F_HW_ESP_TX_CSUM)))
+	    (!exthdrlen || netdev_feature_test_bit(NETIF_F_HW_ESP_TX_CSUM_BIT,
+						   rt->dst.dev->features)))
 		csummode = CHECKSUM_PARTIAL;
 
 	if (flags & MSG_ZEROCOPY && length && sock_flag(sk, SOCK_ZEROCOPY)) {
@@ -1006,7 +1009,8 @@ static int __ip_append_data(struct sock *sk,
 		if (!uarg)
 			return -ENOBUFS;
 		extra_uref = !skb_zcopy(skb);	/* only ref on new uarg */
-		if (rt->dst.dev->features & NETIF_F_SG &&
+		if (netdev_feature_test_bit(NETIF_F_SG_BIT,
+					    rt->dst.dev->features) &&
 		    csummode == CHECKSUM_PARTIAL) {
 			paged = true;
 		} else {
@@ -1069,11 +1073,13 @@ static int __ip_append_data(struct sock *sk,
 				alloc_extra += rt->dst.trailer_len;
 
 			if ((flags & MSG_MORE) &&
-			    !(rt->dst.dev->features&NETIF_F_SG))
+			    !netdev_feature_test_bit(NETIF_F_SG_BIT,
+						     rt->dst.dev->features))
 				alloclen = mtu;
 			else if (!paged &&
 				 (fraglen + alloc_extra < SKB_MAX_ALLOC ||
-				  !(rt->dst.dev->features & NETIF_F_SG)))
+				  !netdev_feature_test_bit(NETIF_F_SG_BIT,
+							   rt->dst.dev->features)))
 				alloclen = fraglen;
 			else {
 				alloclen = min_t(int, fraglen, MAX_HEADER);
@@ -1161,7 +1167,8 @@ static int __ip_append_data(struct sock *sk,
 		if (copy > length)
 			copy = length;
 
-		if (!(rt->dst.dev->features&NETIF_F_SG) &&
+		if (!netdev_feature_test_bit(NETIF_F_SG_BIT,
+					     rt->dst.dev->features) &&
 		    skb_tailroom(skb) >= copy) {
 			unsigned int off;
 
@@ -1340,7 +1347,7 @@ ssize_t	ip_append_page(struct sock *sk, struct flowi4 *fl4, struct page *page,
 	if (cork->flags & IPCORK_OPT)
 		opt = cork->opt;
 
-	if (!(rt->dst.dev->features & NETIF_F_SG))
+	if (!netdev_feature_test_bit(NETIF_F_SG_BIT, rt->dst.dev->features))
 		return -EOPNOTSUPP;
 
 	hh_len = LL_RESERVED_SPACE(rt->dst.dev);
diff --git a/net/ipv4/ip_tunnel.c b/net/ipv4/ip_tunnel.c
index fe9101d3d69e..ac5671e5ea28 100644
--- a/net/ipv4/ip_tunnel.c
+++ b/net/ipv4/ip_tunnel.c
@@ -1072,7 +1072,9 @@ int ip_tunnel_init_net(struct net *net, unsigned int ip_tnl_net_id,
 	 * Allowing to move it to another netns is clearly unsafe.
 	 */
 	if (!IS_ERR(itn->fb_tunnel_dev)) {
-		itn->fb_tunnel_dev->features |= NETIF_F_NETNS_LOCAL;
+		netdev_feature_set_bit(NETIF_F_NETNS_LOCAL_BIT,
+				       &itn->fb_tunnel_dev->features);
+
 		itn->fb_tunnel_dev->mtu = ip_tunnel_bind_dev(itn->fb_tunnel_dev);
 		ip_tunnel_add(itn, netdev_priv(itn->fb_tunnel_dev));
 		itn->type = itn->fb_tunnel_dev->type;
diff --git a/net/ipv4/ip_vti.c b/net/ipv4/ip_vti.c
index efe25a0172e6..bf94ef6a1f2c 100644
--- a/net/ipv4/ip_vti.c
+++ b/net/ipv4/ip_vti.c
@@ -430,7 +430,7 @@ static int vti_tunnel_init(struct net_device *dev)
 
 	dev->flags		= IFF_NOARP;
 	dev->addr_len		= 4;
-	dev->features		|= NETIF_F_LLTX;
+	netdev_feature_set_bit(NETIF_F_LLTX_BIT, &dev->features);
 	netif_keep_dst(dev);
 
 	return ip_tunnel_init(dev);
diff --git a/net/ipv4/ipip.c b/net/ipv4/ipip.c
index 3aa78ccbec3e..b587d472fad5 100644
--- a/net/ipv4/ipip.c
+++ b/net/ipv4/ipip.c
@@ -368,11 +368,11 @@ static void ipip_tunnel_setup(struct net_device *dev)
 	dev->type		= ARPHRD_TUNNEL;
 	dev->flags		= IFF_NOARP;
 	dev->addr_len		= 4;
-	dev->features		|= NETIF_F_LLTX;
+	netdev_feature_set_bit(NETIF_F_LLTX_BIT, &dev->features);
 	netif_keep_dst(dev);
 
-	dev->features		|= IPIP_FEATURES;
-	dev->hw_features	|= IPIP_FEATURES;
+	netdev_feature_set_bits(IPIP_FEATURES, &dev->features);
+	netdev_feature_set_bits(IPIP_FEATURES, &dev->hw_features);
 	ip_tunnel_setup(dev, ipip_net_id);
 }
 
diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c
index 2dda856ca260..750a4a102014 100644
--- a/net/ipv4/ipmr.c
+++ b/net/ipv4/ipmr.c
@@ -532,7 +532,7 @@ static void reg_vif_setup(struct net_device *dev)
 	dev->flags		= IFF_NOARP;
 	dev->netdev_ops		= &reg_vif_netdev_ops;
 	dev->needs_free_netdev	= true;
-	dev->features		|= NETIF_F_NETNS_LOCAL;
+	netdev_feature_set_bit(NETIF_F_NETNS_LOCAL_BIT, &dev->features);
 }
 
 static struct net_device *ipmr_reg_vif(struct net *net, struct mr_table *mrt)
diff --git a/net/ipv4/tcp_offload.c b/net/ipv4/tcp_offload.c
index fc61cd3fea65..4bf0a0d382f6 100644
--- a/net/ipv4/tcp_offload.c
+++ b/net/ipv4/tcp_offload.c
@@ -65,6 +65,7 @@ struct sk_buff *tcp_gso_segment(struct sk_buff *skb,
 	struct sk_buff *gso_skb = skb;
 	__sum16 newcheck;
 	bool ooo_okay, copy_destructor;
+	netdev_features_t tmp;
 
 	th = tcp_hdr(skb);
 	thlen = th->doff * 4;
@@ -81,7 +82,9 @@ struct sk_buff *tcp_gso_segment(struct sk_buff *skb,
 	if (unlikely(skb->len <= mss))
 		goto out;
 
-	if (skb_gso_ok(skb, features | NETIF_F_GSO_ROBUST)) {
+	netdev_feature_copy(&tmp, features);
+	netdev_feature_set_bit(NETIF_F_GSO_ROBUST_BIT, &tmp);
+	if (skb_gso_ok(skb, tmp)) {
 		/* Packet is from an untrusted source, reset gso_segs. */
 
 		skb_shinfo(skb)->gso_segs = DIV_ROUND_UP(skb->len, mss);
diff --git a/net/ipv4/udp_offload.c b/net/ipv4/udp_offload.c
index 86d32a1e62ac..e4c14e142bde 100644
--- a/net/ipv4/udp_offload.c
+++ b/net/ipv4/udp_offload.c
@@ -63,22 +63,26 @@ static struct sk_buff *__skb_udp_tunnel_segment(struct sk_buff *skb,
 	/* Try to offload checksum if possible */
 	offload_csum = !!(need_csum &&
 			  !need_ipsec &&
-			  (skb->dev->features &
-			   (is_ipv6 ? (NETIF_F_HW_CSUM | NETIF_F_IPV6_CSUM) :
-				      (NETIF_F_HW_CSUM | NETIF_F_IP_CSUM))));
-
-	features &= skb->dev->hw_enc_features;
+			  is_ipv6 ?
+			  netdev_feature_test_bits(NETIF_F_HW_CSUM |
+						   NETIF_F_IPV6_CSUM,
+						   skb->dev->features) :
+			  netdev_feature_test_bits(NETIF_F_HW_CSUM |
+						   NETIF_F_IP_CSUM,
+						   skb->dev->features));
+
+	netdev_feature_and(&features, features, skb->dev->hw_enc_features);
 	if (need_csum)
-		features &= ~NETIF_F_SCTP_CRC;
+		netdev_feature_clear_bit(NETIF_F_SCTP_CRC_BIT, &features);
 
 	/* The only checksum offload we care about from here on out is the
 	 * outer one so strip the existing checksum feature flags and
 	 * instead set the flag based on our outer checksum offload value.
 	 */
 	if (remcsum) {
-		features &= ~NETIF_F_CSUM_MASK;
+		netdev_feature_clear_bits(NETIF_F_CSUM_MASK, &features);
 		if (!need_csum || offload_csum)
-			features |= NETIF_F_HW_CSUM;
+			netdev_feature_set_bit(NETIF_F_HW_CSUM_BIT, &features);
 	}
 
 	/* segment inner packet. */
@@ -414,7 +418,7 @@ static struct sk_buff *udp4_ufo_fragment(struct sk_buff *skb,
 	 * software prior to segmenting the frame.
 	 */
 	if (!skb->encap_hdr_csum)
-		features |= NETIF_F_HW_CSUM;
+		netdev_feature_set_bit(NETIF_F_HW_CSUM_BIT, &features);
 
 	/* Fragment the skb. IP headers of the fragments are updated in
 	 * inet_gso_segment()
@@ -520,10 +524,12 @@ struct sk_buff *udp_gro_receive(struct list_head *head, struct sk_buff *skb,
 	 */
 	NAPI_GRO_CB(skb)->is_flist = 0;
 	if (!sk || !udp_sk(sk)->gro_receive) {
-		if (skb->dev->features & NETIF_F_GRO_FRAGLIST)
+		if (netdev_feature_test_bit(NETIF_F_GRO_FRAGLIST_BIT,
+					    skb->dev->features))
 			NAPI_GRO_CB(skb)->is_flist = sk ? !udp_sk(sk)->gro_enabled : 1;
 
-		if ((!sk && (skb->dev->features & NETIF_F_GRO_UDP_FWD)) ||
+		if ((!sk && netdev_feature_test_bit(NETIF_F_GRO_UDP_FWD_BIT,
+						    skb->dev->features)) ||
 		    (sk && udp_sk(sk)->gro_enabled) || NAPI_GRO_CB(skb)->is_flist)
 			return call_gro_receive(udp_gro_receive_segment, head, skb);
 
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 049/167] net: ipv6: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (47 preceding siblings ...)
  2021-09-29 15:51 ` [RFCv2 net-next 048/167] net: ipv4: " Jian Shen
@ 2021-09-29 15:51 ` Jian Shen
  2021-09-29 15:51 ` [RFCv2 net-next 050/167] net: hsr: " Jian Shen
                   ` (121 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:51 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 net/ipv6/esp6_offload.c | 19 +++++++++++--------
 net/ipv6/ip6_gre.c      | 17 ++++++++++-------
 net/ipv6/ip6_offload.c  |  3 ++-
 net/ipv6/ip6_output.c   | 18 ++++++++++++------
 net/ipv6/ip6_tunnel.c   |  9 +++++----
 net/ipv6/ip6mr.c        |  2 +-
 net/ipv6/sit.c          |  9 +++++----
 net/ipv6/udp_offload.c  |  2 +-
 8 files changed, 47 insertions(+), 32 deletions(-)

diff --git a/net/ipv6/esp6_offload.c b/net/ipv6/esp6_offload.c
index a349d4798077..5c2e0da49804 100644
--- a/net/ipv6/esp6_offload.c
+++ b/net/ipv6/esp6_offload.c
@@ -228,8 +228,8 @@ static struct sk_buff *esp6_gso_segment(struct sk_buff *skb,
 	struct xfrm_state *x;
 	struct ip_esp_hdr *esph;
 	struct crypto_aead *aead;
-	netdev_features_t esp_features = features;
 	struct xfrm_offload *xo = xfrm_offload(skb);
+	netdev_features_t esp_features;
 	struct sec_path *sp;
 
 	if (!xo)
@@ -253,12 +253,14 @@ static struct sk_buff *esp6_gso_segment(struct sk_buff *skb,
 
 	skb->encap_hdr_csum = 1;
 
-	if (!(features & NETIF_F_HW_ESP) || x->xso.dev != skb->dev)
-		esp_features = features & ~(NETIF_F_SG | NETIF_F_CSUM_MASK |
-					    NETIF_F_SCTP_CRC);
-	else if (!(features & NETIF_F_HW_ESP_TX_CSUM))
-		esp_features = features & ~(NETIF_F_CSUM_MASK |
-					    NETIF_F_SCTP_CRC);
+	netdev_feature_copy(&esp_features, features);
+	if (!netdev_feature_test_bit(NETIF_F_HW_ESP_BIT, features) ||
+	    x->xso.dev != skb->dev)
+		netdev_feature_clear_bits(NETIF_F_SG | NETIF_F_CSUM_MASK |
+					  NETIF_F_SCTP_CRC, &esp_features);
+	else if (!netdev_feature_test_bit(NETIF_F_HW_ESP_TX_CSUM_BIT, features))
+		netdev_feature_clear_bits(NETIF_F_CSUM_MASK | NETIF_F_SCTP_CRC,
+					  &esp_features);
 
 	xo->flags |= XFRM_GSO_SEGMENT;
 
@@ -298,7 +300,8 @@ static int esp6_xmit(struct xfrm_state *x, struct sk_buff *skb,  netdev_features
 	if (!xo)
 		return -EINVAL;
 
-	if (!(features & NETIF_F_HW_ESP) || x->xso.dev != skb->dev) {
+	if (!netdev_feature_test_bit(NETIF_F_HW_ESP_BIT, features) ||
+	    x->xso.dev != skb->dev) {
 		xo->flags |= CRYPTO_FALLBACK;
 		hw_offload = false;
 	}
diff --git a/net/ipv6/ip6_gre.c b/net/ipv6/ip6_gre.c
index 3ad201d372d8..336ddf0b3586 100644
--- a/net/ipv6/ip6_gre.c
+++ b/net/ipv6/ip6_gre.c
@@ -385,7 +385,7 @@ static struct ip6_tnl *ip6gre_tunnel_locate(struct net *net,
 
 	/* Can use a lockless transmit, unless we generate output sequences */
 	if (!(nt->parms.o_flags & TUNNEL_SEQ))
-		dev->features |= NETIF_F_LLTX;
+		netdev_feature_set_bit(NETIF_F_LLTX_BIT, &dev->features);
 
 	ip6gre_tunnel_link(ign, nt);
 	return nt;
@@ -1439,8 +1439,8 @@ static void ip6gre_tnl_init_features(struct net_device *dev)
 {
 	struct ip6_tnl *nt = netdev_priv(dev);
 
-	dev->features		|= GRE6_FEATURES;
-	dev->hw_features	|= GRE6_FEATURES;
+	netdev_feature_set_bits(GRE6_FEATURES, &dev->features);
+	netdev_feature_set_bits(GRE6_FEATURES, &dev->hw_features);
 
 	if (!(nt->parms.o_flags & TUNNEL_SEQ)) {
 		/* TCP offload with GRE SEQ is not supported, nor
@@ -1449,14 +1449,16 @@ static void ip6gre_tnl_init_features(struct net_device *dev)
 		 */
 		if (!(nt->parms.o_flags & TUNNEL_CSUM) ||
 		    nt->encap.type == TUNNEL_ENCAP_NONE) {
-			dev->features    |= NETIF_F_GSO_SOFTWARE;
-			dev->hw_features |= NETIF_F_GSO_SOFTWARE;
+			netdev_feature_set_bits(NETIF_F_GSO_SOFTWARE,
+						&dev->features);
+			netdev_feature_set_bits(NETIF_F_GSO_SOFTWARE,
+						&dev->hw_features);
 		}
 
 		/* Can use a lockless transmit, unless we generate
 		 * output sequences
 		 */
-		dev->features |= NETIF_F_LLTX;
+		netdev_feature_set_bit(NETIF_F_LLTX_BIT, &dev->features);
 	}
 }
 
@@ -1598,7 +1600,8 @@ static int __net_init ip6gre_init_net(struct net *net)
 	/* FB netdevice is special: we have one, and only one per netns.
 	 * Allowing to move it to another netns is clearly unsafe.
 	 */
-	ign->fb_tunnel_dev->features |= NETIF_F_NETNS_LOCAL;
+	netdev_feature_set_bit(NETIF_F_NETNS_LOCAL_BIT,
+			       &ign->fb_tunnel_dev->features);
 
 
 	ip6gre_fb_tunnel_init(ign->fb_tunnel_dev);
diff --git a/net/ipv6/ip6_offload.c b/net/ipv6/ip6_offload.c
index 1b9827ff8ccf..4c3ba4b4ba51 100644
--- a/net/ipv6/ip6_offload.c
+++ b/net/ipv6/ip6_offload.c
@@ -93,7 +93,8 @@ static struct sk_buff *ipv6_gso_segment(struct sk_buff *skb,
 
 	encap = SKB_GSO_CB(skb)->encap_level > 0;
 	if (encap)
-		features &= skb->dev->hw_enc_features;
+		netdev_feature_and(&features, features,
+				   skb->dev->hw_enc_features);
 	SKB_GSO_CB(skb)->encap_level += sizeof(*ipv6h);
 
 	ipv6h = ipv6_hdr(skb);
diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c
index 46a0867bcedf..78e01d4faf09 100644
--- a/net/ipv6/ip6_output.c
+++ b/net/ipv6/ip6_output.c
@@ -147,7 +147,8 @@ ip6_finish_output_gso_slowpath_drop(struct net *net, struct sock *sk,
 	 * egress MTU.
 	 */
 	netif_skb_features(skb, &features);
-	segs = skb_gso_segment(skb, features & ~NETIF_F_GSO_MASK);
+	netdev_feature_clear_bits(NETIF_F_GSO_MASK, &features);
+	segs = skb_gso_segment(skb, features);
 	if (IS_ERR_OR_NULL(segs)) {
 		kfree_skb(skb);
 		return -ENOMEM;
@@ -1512,7 +1513,8 @@ static int __ip6_append_data(struct sock *sk,
 	    headersize == sizeof(struct ipv6hdr) &&
 	    length <= mtu - headersize &&
 	    (!(flags & MSG_MORE) || cork->gso_size) &&
-	    rt->dst.dev->features & (NETIF_F_IPV6_CSUM | NETIF_F_HW_CSUM))
+	    netdev_feature_test_bits(NETIF_F_IPV6_CSUM | NETIF_F_HW_CSUM,
+				     rt->dst.dev->features))
 		csummode = CHECKSUM_PARTIAL;
 
 	if (flags & MSG_ZEROCOPY && length && sock_flag(sk, SOCK_ZEROCOPY)) {
@@ -1520,7 +1522,8 @@ static int __ip6_append_data(struct sock *sk,
 		if (!uarg)
 			return -ENOBUFS;
 		extra_uref = !skb_zcopy(skb);	/* only ref on new uarg */
-		if (rt->dst.dev->features & NETIF_F_SG &&
+		if (netdev_feature_test_bit(NETIF_F_SG_BIT,
+					    rt->dst.dev->features) &&
 		    csummode == CHECKSUM_PARTIAL) {
 			paged = true;
 		} else {
@@ -1598,11 +1601,13 @@ static int __ip6_append_data(struct sock *sk,
 			alloc_extra += sizeof(struct frag_hdr);
 
 			if ((flags & MSG_MORE) &&
-			    !(rt->dst.dev->features&NETIF_F_SG))
+			    !netdev_feature_test_bit(NETIF_F_SG_BIT,
+						     rt->dst.dev->features))
 				alloclen = mtu;
 			else if (!paged &&
 				 (fraglen + alloc_extra < SKB_MAX_ALLOC ||
-				  !(rt->dst.dev->features & NETIF_F_SG)))
+				  !netdev_feature_test_bit(NETIF_F_SG_BIT,
+							   rt->dst.dev->features)))
 				alloclen = fraglen;
 			else {
 				alloclen = min_t(int, fraglen, MAX_HEADER);
@@ -1705,7 +1710,8 @@ static int __ip6_append_data(struct sock *sk,
 		if (copy > length)
 			copy = length;
 
-		if (!(rt->dst.dev->features&NETIF_F_SG) &&
+		if (!netdev_feature_test_bit(NETIF_F_SG_BIT,
+					     rt->dst.dev->features) &&
 		    skb_tailroom(skb) >= copy) {
 			unsigned int off;
 
diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c
index 20a67efda47f..012f3ad0e7c0 100644
--- a/net/ipv6/ip6_tunnel.c
+++ b/net/ipv6/ip6_tunnel.c
@@ -1833,11 +1833,11 @@ static void ip6_tnl_dev_setup(struct net_device *dev)
 	dev->type = ARPHRD_TUNNEL6;
 	dev->flags |= IFF_NOARP;
 	dev->addr_len = sizeof(struct in6_addr);
-	dev->features |= NETIF_F_LLTX;
+	netdev_feature_set_bit(NETIF_F_LLTX_BIT, &dev->features);
 	netif_keep_dst(dev);
 
-	dev->features		|= IPXIPX_FEATURES;
-	dev->hw_features	|= IPXIPX_FEATURES;
+	netdev_feature_set_bits(IPXIPX_FEATURES, &dev->features);
+	netdev_feature_set_bits(IPXIPX_FEATURES, &dev->hw_features);
 
 	/* This perm addr will be used as interface identifier by IPv6 */
 	dev->addr_assign_type = NET_ADDR_RANDOM;
@@ -2279,7 +2279,8 @@ static int __net_init ip6_tnl_init_net(struct net *net)
 	/* FB netdevice is special: we have one, and only one per netns.
 	 * Allowing to move it to another netns is clearly unsafe.
 	 */
-	ip6n->fb_tnl_dev->features |= NETIF_F_NETNS_LOCAL;
+	netdev_feature_set_bit(NETIF_F_NETNS_LOCAL_BIT,
+			       &ip6n->fb_tnl_dev->features);
 
 	err = ip6_fb_tnl_dev_init(ip6n->fb_tnl_dev);
 	if (err < 0)
diff --git a/net/ipv6/ip6mr.c b/net/ipv6/ip6mr.c
index 36ed9efb8825..daf2a4b13112 100644
--- a/net/ipv6/ip6mr.c
+++ b/net/ipv6/ip6mr.c
@@ -636,7 +636,7 @@ static void reg_vif_setup(struct net_device *dev)
 	dev->flags		= IFF_NOARP;
 	dev->netdev_ops		= &reg_vif_netdev_ops;
 	dev->needs_free_netdev	= true;
-	dev->features		|= NETIF_F_NETNS_LOCAL;
+	netdev_feature_set_bit(NETIF_F_NETNS_LOCAL_BIT, &dev->features);
 }
 
 static struct net_device *ip6mr_reg_vif(struct net *net, struct mr_table *mrt)
diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c
index ef0c7a7c18e2..58aa23267d41 100644
--- a/net/ipv6/sit.c
+++ b/net/ipv6/sit.c
@@ -1438,9 +1438,9 @@ static void ipip6_tunnel_setup(struct net_device *dev)
 	dev->flags		= IFF_NOARP;
 	netif_keep_dst(dev);
 	dev->addr_len		= 4;
-	dev->features		|= NETIF_F_LLTX;
-	dev->features		|= SIT_FEATURES;
-	dev->hw_features	|= SIT_FEATURES;
+	netdev_feature_set_bit(NETIF_F_LLTX_BIT, &dev->features);
+	netdev_feature_set_bits(SIT_FEATURES, &dev->features);
+	netdev_feature_set_bits(SIT_FEATURES, &dev->hw_features);
 }
 
 static int ipip6_tunnel_init(struct net_device *dev)
@@ -1918,7 +1918,8 @@ static int __net_init sit_init_net(struct net *net)
 	/* FB netdevice is special: we have one, and only one per netns.
 	 * Allowing to move it to another netns is clearly unsafe.
 	 */
-	sitn->fb_tunnel_dev->features |= NETIF_F_NETNS_LOCAL;
+	netdev_feature_set_bit(NETIF_F_NETNS_LOCAL_BIT,
+			       &sitn->fb_tunnel_dev->features);
 
 	err = register_netdev(sitn->fb_tunnel_dev);
 	if (err)
diff --git a/net/ipv6/udp_offload.c b/net/ipv6/udp_offload.c
index b3d9ed96e5ea..5911c603bd0b 100644
--- a/net/ipv6/udp_offload.c
+++ b/net/ipv6/udp_offload.c
@@ -69,7 +69,7 @@ static struct sk_buff *udp6_ufo_fragment(struct sk_buff *skb,
 		 * software prior to segmenting the frame.
 		 */
 		if (!skb->encap_hdr_csum)
-			features |= NETIF_F_HW_CSUM;
+			netdev_feature_set_bit(NETIF_F_HW_CSUM_BIT, &features);
 
 		/* Check if there is enough headroom to insert fragment header. */
 		tnl_hlen = skb_tnl_header_len(skb);
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 050/167] net: hsr: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (48 preceding siblings ...)
  2021-09-29 15:51 ` [RFCv2 net-next 049/167] net: ipv6: " Jian Shen
@ 2021-09-29 15:51 ` Jian Shen
  2021-09-29 15:51 ` [RFCv2 net-next 051/167] net: mpls: " Jian Shen
                   ` (120 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:51 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 net/hsr/hsr_device.c   | 21 ++++++++++++---------
 net/hsr/hsr_forward.c  | 11 +++++++----
 net/hsr/hsr_framereg.c |  3 ++-
 net/hsr/hsr_slave.c    |  3 ++-
 4 files changed, 23 insertions(+), 15 deletions(-)

diff --git a/net/hsr/hsr_device.c b/net/hsr/hsr_device.c
index 7e6dd372711f..b7ba16bdb8e0 100644
--- a/net/hsr/hsr_device.c
+++ b/net/hsr/hsr_device.c
@@ -183,7 +183,7 @@ static void hsr_features_recompute(struct hsr_priv *hsr,
 	netdev_features_t mask;
 	struct hsr_port *port;
 
-	mask = *features;
+	netdev_feature_copy(&mask, *features);
 
 	/* Mask out all features that, if supported by one device, should be
 	 * enabled for all devices (see NETIF_F_ONE_FOR_ALL).
@@ -192,7 +192,7 @@ static void hsr_features_recompute(struct hsr_priv *hsr,
 	 * that were in features originally, and also is in NETIF_F_ONE_FOR_ALL,
 	 * may become enabled.
 	 */
-	*features &= ~NETIF_F_ONE_FOR_ALL;
+	netdev_feature_clear_bits(NETIF_F_ONE_FOR_ALL, features);
 	hsr_for_each_port(hsr, port)
 		netdev_increment_features(features, *features,
 					  port->dev->features, mask);
@@ -445,22 +445,25 @@ void hsr_dev_setup(struct net_device *dev)
 
 	dev->needs_free_netdev = true;
 
-	dev->hw_features = NETIF_F_SG | NETIF_F_FRAGLIST | NETIF_F_HIGHDMA |
-			   NETIF_F_GSO_MASK | NETIF_F_HW_CSUM |
-			   NETIF_F_HW_VLAN_CTAG_TX;
+	netdev_feature_zero(&dev->hw_features);
+	netdev_feature_set_bits(NETIF_F_SG | NETIF_F_FRAGLIST |
+				NETIF_F_HIGHDMA |
+				NETIF_F_GSO_MASK | NETIF_F_HW_CSUM |
+				NETIF_F_HW_VLAN_CTAG_TX, &dev->hw_features);
 
-	dev->features = dev->hw_features;
+	netdev_feature_copy(&dev->features, dev->hw_features);
 
 	/* Prevent recursive tx locking */
-	dev->features |= NETIF_F_LLTX;
+	netdev_feature_set_bit(NETIF_F_LLTX_BIT, &dev->features);
+
 	/* VLAN on top of HSR needs testing and probably some work on
 	 * hsr_header_create() etc.
 	 */
-	dev->features |= NETIF_F_VLAN_CHALLENGED;
+	netdev_feature_set_bit(NETIF_F_VLAN_CHALLENGED_BIT, &dev->features);
 	/* Not sure about this. Taken from bridge code. netdev_features.h says
 	 * it means "Does not change network namespaces".
 	 */
-	dev->features |= NETIF_F_NETNS_LOCAL;
+	netdev_feature_set_bit(NETIF_F_NETNS_LOCAL_BIT, &dev->features);
 }
 
 /* Return true if dev is a HSR master; return false otherwise.
diff --git a/net/hsr/hsr_forward.c b/net/hsr/hsr_forward.c
index ceb8afb2a62f..6b3aa5f93be0 100644
--- a/net/hsr/hsr_forward.c
+++ b/net/hsr/hsr_forward.c
@@ -249,7 +249,8 @@ struct sk_buff *hsr_create_tagged_frame(struct hsr_frame_info *frame,
 		/* set the lane id properly */
 		hsr_set_path_id(hsr_ethhdr, port);
 		return skb_clone(frame->skb_hsr, GFP_ATOMIC);
-	} else if (port->dev->features & NETIF_F_HW_HSR_TAG_INS) {
+	} else if (netdev_feature_test_bit(NETIF_F_HW_HSR_TAG_INS_BIT,
+					   port->dev->features)) {
 		return skb_clone(frame->skb_std, GFP_ATOMIC);
 	}
 
@@ -293,7 +294,8 @@ struct sk_buff *prp_create_tagged_frame(struct hsr_frame_info *frame,
 			return NULL;
 		}
 		return skb_clone(frame->skb_prp, GFP_ATOMIC);
-	} else if (port->dev->features & NETIF_F_HW_HSR_TAG_INS) {
+	} else if (netdev_feature_test_bit(NETIF_F_HW_HSR_TAG_INS_BIT,
+					   port->dev->features)) {
 		return skb_clone(frame->skb_std, GFP_ATOMIC);
 	}
 
@@ -349,7 +351,7 @@ bool prp_drop_frame(struct hsr_frame_info *frame, struct hsr_port *port)
 
 bool hsr_drop_frame(struct hsr_frame_info *frame, struct hsr_port *port)
 {
-	if (port->dev->features & NETIF_F_HW_HSR_FWD)
+	if (netdev_feature_test_bit(NETIF_F_HW_HSR_FWD_BIT, port->dev->features))
 		return prp_drop_frame(frame, port);
 
 	return false;
@@ -390,7 +392,8 @@ static void hsr_forward_do(struct hsr_frame_info *frame)
 		/* If hardware duplicate generation is enabled, only send out
 		 * one port.
 		 */
-		if ((port->dev->features & NETIF_F_HW_HSR_DUP) && sent)
+		if (netdev_feature_test_bit(NETIF_F_HW_HSR_DUP_BIT,
+					    port->dev->features) && sent)
 			continue;
 
 		/* Don't send frame over port where it has been sent before.
diff --git a/net/hsr/hsr_framereg.c b/net/hsr/hsr_framereg.c
index e31949479305..d0b1d9fe4f4f 100644
--- a/net/hsr/hsr_framereg.c
+++ b/net/hsr/hsr_framereg.c
@@ -397,7 +397,8 @@ void hsr_register_frame_in(struct hsr_node *node, struct hsr_port *port,
 	 * ensures entries of restarted nodes gets pruned so that they can
 	 * re-register and resume communications.
 	 */
-	if (!(port->dev->features & NETIF_F_HW_HSR_TAG_RM) &&
+	if (!netdev_feature_test_bit(NETIF_F_HW_HSR_TAG_RM_BIT,
+				     port->dev->features) &&
 	    seq_nr_before(sequence_nr, node->seq_out[port->type]))
 		return;
 
diff --git a/net/hsr/hsr_slave.c b/net/hsr/hsr_slave.c
index b70e6bbf6021..a21a3035f6d2 100644
--- a/net/hsr/hsr_slave.c
+++ b/net/hsr/hsr_slave.c
@@ -54,7 +54,8 @@ static rx_handler_result_t hsr_handle_frame(struct sk_buff **pskb)
 	 */
 	protocol = eth_hdr(skb)->h_proto;
 
-	if (!(port->dev->features & NETIF_F_HW_HSR_TAG_RM) &&
+	if (!netdev_feature_test_bit(NETIF_F_HW_HSR_TAG_RM_BIT,
+				     port->dev->features) &&
 	    hsr->proto_ops->invalid_dan_ingress_frame &&
 	    hsr->proto_ops->invalid_dan_ingress_frame(protocol))
 		goto finish_pass;
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 051/167] net: mpls: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (49 preceding siblings ...)
  2021-09-29 15:51 ` [RFCv2 net-next 050/167] net: hsr: " Jian Shen
@ 2021-09-29 15:51 ` Jian Shen
  2021-09-29 15:51 ` [RFCv2 net-next 052/167] net: nsh: " Jian Shen
                   ` (119 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:51 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 net/mpls/mpls_gso.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/mpls/mpls_gso.c b/net/mpls/mpls_gso.c
index 1482259de9b5..65994c68e126 100644
--- a/net/mpls/mpls_gso.c
+++ b/net/mpls/mpls_gso.c
@@ -43,7 +43,7 @@ static struct sk_buff *mpls_gso_segment(struct sk_buff *skb,
 	skb_reset_mac_header(skb);
 
 	/* Segment inner packet. */
-	mpls_features = skb->dev->mpls_features & features;
+	netdev_feature_and(&mpls_features, skb->dev->mpls_features, features);
 	segs = skb_mac_gso_segment(skb, mpls_features);
 	if (IS_ERR_OR_NULL(segs)) {
 		skb_gso_error_unwind(skb, mpls_protocol, mpls_hlen, mac_offset,
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 052/167] net: nsh: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (50 preceding siblings ...)
  2021-09-29 15:51 ` [RFCv2 net-next 051/167] net: mpls: " Jian Shen
@ 2021-09-29 15:51 ` Jian Shen
  2021-09-29 15:51 ` [RFCv2 net-next 053/167] net: decnet: " Jian Shen
                   ` (118 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:51 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 net/nsh/nsh.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/nsh/nsh.c b/net/nsh/nsh.c
index e9ca007718b7..9cc1b15891ef 100644
--- a/net/nsh/nsh.c
+++ b/net/nsh/nsh.c
@@ -104,7 +104,7 @@ static struct sk_buff *nsh_gso_segment(struct sk_buff *skb,
 	skb->mac_len = proto == htons(ETH_P_TEB) ? ETH_HLEN : 0;
 	skb->protocol = proto;
 
-	features &= NETIF_F_SG;
+	netdev_feature_and_bits(NETIF_F_SG, &features);
 	segs = skb_mac_gso_segment(skb, features);
 	if (IS_ERR_OR_NULL(segs)) {
 		skb_gso_error_unwind(skb, htons(ETH_P_NSH), nsh_len,
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 053/167] net: decnet: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (51 preceding siblings ...)
  2021-09-29 15:51 ` [RFCv2 net-next 052/167] net: nsh: " Jian Shen
@ 2021-09-29 15:51 ` Jian Shen
  2021-09-29 15:51 ` [RFCv2 net-next 054/167] net: dccp: " Jian Shen
                   ` (117 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:51 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 net/decnet/af_decnet.c  | 2 +-
 net/decnet/dn_nsp_out.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/net/decnet/af_decnet.c b/net/decnet/af_decnet.c
index dc92a67baea3..83138e013c46 100644
--- a/net/decnet/af_decnet.c
+++ b/net/decnet/af_decnet.c
@@ -949,7 +949,7 @@ static int __dn_connect(struct sock *sk, struct sockaddr_dn *addr, int addrlen,
 	if (dn_route_output_sock(&sk->sk_dst_cache, &fld, sk, flags) < 0)
 		goto out;
 	dst = __sk_dst_get(sk);
-	sk->sk_route_caps = dst->dev->features;
+	netdev_feature_copy(&sk->sk_route_caps, dst->dev->features);
 	sock->state = SS_CONNECTING;
 	scp->state = DN_CI;
 	scp->segsize_loc = dst_metric_advmss(dst);
diff --git a/net/decnet/dn_nsp_out.c b/net/decnet/dn_nsp_out.c
index eadc89583168..613e81bb6ed8 100644
--- a/net/decnet/dn_nsp_out.c
+++ b/net/decnet/dn_nsp_out.c
@@ -89,7 +89,7 @@ static void dn_nsp_send(struct sk_buff *skb)
 	fld.flowidn_proto = DNPROTO_NSP;
 	if (dn_route_output_sock(&sk->sk_dst_cache, &fld, sk, 0) == 0) {
 		dst = sk_dst_get(sk);
-		sk->sk_route_caps = dst->dev->features;
+		netdev_feature_copy(&sk->sk_route_caps, dst->dev->features);
 		goto try_again;
 	}
 
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 054/167] net: dccp: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (52 preceding siblings ...)
  2021-09-29 15:51 ` [RFCv2 net-next 053/167] net: decnet: " Jian Shen
@ 2021-09-29 15:51 ` Jian Shen
  2021-09-29 15:51 ` [RFCv2 net-next 055/167] net: l2tp: " Jian Shen
                   ` (116 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:51 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 net/dccp/ipv4.c | 2 +-
 net/dccp/ipv6.c | 7 ++++---
 2 files changed, 5 insertions(+), 4 deletions(-)

diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c
index 0ea29270d7e5..f6161a100dbe 100644
--- a/net/dccp/ipv4.c
+++ b/net/dccp/ipv4.c
@@ -138,7 +138,7 @@ int dccp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
 	 */
 	dccp_set_state(sk, DCCP_CLOSED);
 	ip_rt_put(rt);
-	sk->sk_route_caps = 0;
+	netdev_feature_zero(&sk->sk_route_caps);
 	inet->inet_dport = 0;
 	goto out;
 }
diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c
index fa663518fa0e..0f966e02ddc4 100644
--- a/net/dccp/ipv6.c
+++ b/net/dccp/ipv6.c
@@ -489,8 +489,9 @@ static struct sock *dccp_v6_request_recv_sock(const struct sock *sk,
 	 */
 
 	ip6_dst_store(newsk, dst, NULL, NULL);
-	newsk->sk_route_caps = dst->dev->features & ~(NETIF_F_IP_CSUM |
-						      NETIF_F_TSO);
+	netdev_feature_copy(&newsk->sk_route_caps, dst->dev->features);
+	netdev_feature_clear_bits(NETIF_F_IP_CSUM | NETIF_F_TSO,
+				  &newsk->sk_route_caps);
 	newdp6 = (struct dccp6_sock *)newsk;
 	newinet = inet_sk(newsk);
 	newinet->pinet6 = &newdp6->inet6;
@@ -970,7 +971,7 @@ static int dccp_v6_connect(struct sock *sk, struct sockaddr *uaddr,
 	__sk_dst_reset(sk);
 failure:
 	inet->inet_dport = 0;
-	sk->sk_route_caps = 0;
+	netdev_feature_zero(&sk->sk_route_caps);
 	return err;
 }
 
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 055/167] net: l2tp: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (53 preceding siblings ...)
  2021-09-29 15:51 ` [RFCv2 net-next 054/167] net: dccp: " Jian Shen
@ 2021-09-29 15:51 ` Jian Shen
  2021-09-29 15:51 ` [RFCv2 net-next 056/167] net: ntb_netdev: " Jian Shen
                   ` (115 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:51 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 net/l2tp/l2tp_eth.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/l2tp/l2tp_eth.c b/net/l2tp/l2tp_eth.c
index 6cd97c75445c..9fb916cec24e 100644
--- a/net/l2tp/l2tp_eth.c
+++ b/net/l2tp/l2tp_eth.c
@@ -117,7 +117,7 @@ static void l2tp_eth_dev_setup(struct net_device *dev)
 	SET_NETDEV_DEVTYPE(dev, &l2tpeth_type);
 	ether_setup(dev);
 	dev->priv_flags		&= ~IFF_TX_SKB_SHARING;
-	dev->features		|= NETIF_F_LLTX;
+	netdev_feature_set_bit(NETIF_F_LLTX_BIT, &dev->features);
 	dev->netdev_ops		= &l2tp_eth_netdev_ops;
 	dev->needs_free_netdev	= true;
 }
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 056/167] net: ntb_netdev: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (54 preceding siblings ...)
  2021-09-29 15:51 ` [RFCv2 net-next 055/167] net: l2tp: " Jian Shen
@ 2021-09-29 15:51 ` Jian Shen
  2021-09-29 15:51 ` [RFCv2 net-next 057/167] net: thunderbolt: " Jian Shen
                   ` (114 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:51 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 drivers/net/ntb_netdev.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ntb_netdev.c b/drivers/net/ntb_netdev.c
index a5bab614ff84..aaaf118e778f 100644
--- a/drivers/net/ntb_netdev.c
+++ b/drivers/net/ntb_netdev.c
@@ -420,11 +420,12 @@ static int ntb_netdev_probe(struct device *client_dev)
 	dev = netdev_priv(ndev);
 	dev->ndev = ndev;
 	dev->pdev = pdev;
-	ndev->features = NETIF_F_HIGHDMA;
+	netdev_feature_zero(&ndev->features);
+	netdev_feature_set_bit(NETIF_F_HIGHDMA_BIT, &ndev->features);
 
 	ndev->priv_flags |= IFF_LIVE_ADDR_CHANGE;
 
-	ndev->hw_features = ndev->features;
+	netdev_feature_copy(&ndev->hw_features, ndev->features);
 	ndev->watchdog_timeo = msecs_to_jiffies(NTB_TX_TIMEOUT_MS);
 
 	eth_random_addr(ndev->perm_addr);
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 057/167] net: thunderbolt: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (55 preceding siblings ...)
  2021-09-29 15:51 ` [RFCv2 net-next 056/167] net: ntb_netdev: " Jian Shen
@ 2021-09-29 15:51 ` Jian Shen
  2021-09-29 15:51 ` [RFCv2 net-next 058/167] net: phonet: " Jian Shen
                   ` (113 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:51 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 drivers/net/thunderbolt.c | 10 +++++++---
 1 file changed, 7 insertions(+), 3 deletions(-)

diff --git a/drivers/net/thunderbolt.c b/drivers/net/thunderbolt.c
index 9a6a8353e192..0fd4cd06fa6b 100644
--- a/drivers/net/thunderbolt.c
+++ b/drivers/net/thunderbolt.c
@@ -1257,9 +1257,13 @@ static int tbnet_probe(struct tb_service *svc, const struct tb_service_id *id)
 	 * we need to announce support for most of the offloading
 	 * features here.
 	 */
-	dev->hw_features = NETIF_F_SG | NETIF_F_ALL_TSO | NETIF_F_GRO |
-			   NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM;
-	dev->features = dev->hw_features | NETIF_F_HIGHDMA;
+	netdev_feature_zero(&dev->hw_features);
+	netdev_feature_set_bits(NETIF_F_SG | NETIF_F_ALL_TSO | NETIF_F_GRO |
+				NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM,
+				&dev->hw_features);
+	netdev_feature_copy(&dev->features, dev->hw_features);
+	netdev_feature_set_bit(NETIF_F_HIGHDMA_BIT, &dev->features);
+
 	dev->hard_header_len += sizeof(struct thunderbolt_ip_frame_header);
 
 	netif_napi_add(dev, &net->napi, tbnet_poll, NAPI_POLL_WEIGHT);
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 058/167] net: phonet: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (56 preceding siblings ...)
  2021-09-29 15:51 ` [RFCv2 net-next 057/167] net: thunderbolt: " Jian Shen
@ 2021-09-29 15:51 ` Jian Shen
  2021-09-29 15:51 ` [RFCv2 net-next 059/167] net: vrf: " Jian Shen
                   ` (112 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:51 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 net/phonet/pep-gprs.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/net/phonet/pep-gprs.c b/net/phonet/pep-gprs.c
index 1f5df0432d37..0f6e6c1f3ec1 100644
--- a/net/phonet/pep-gprs.c
+++ b/net/phonet/pep-gprs.c
@@ -212,7 +212,8 @@ static const struct net_device_ops gprs_netdev_ops = {
 
 static void gprs_setup(struct net_device *dev)
 {
-	dev->features		= NETIF_F_FRAGLIST;
+	netdev_feature_zero(&dev->features);
+	netdev_feature_set_bit(NETIF_F_FRAGLIST_BIT, &dev->features);
 	dev->type		= ARPHRD_PHONET_PIPE;
 	dev->flags		= IFF_POINTOPOINT | IFF_NOARP;
 	dev->mtu		= GPRS_DEFAULT_MTU;
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 059/167] net: vrf: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (57 preceding siblings ...)
  2021-09-29 15:51 ` [RFCv2 net-next 058/167] net: phonet: " Jian Shen
@ 2021-09-29 15:51 ` Jian Shen
  2021-09-29 15:51 ` [RFCv2 net-next 060/167] net: sctp: " Jian Shen
                   ` (111 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:51 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 drivers/net/vrf.c | 20 +++++++++++---------
 1 file changed, 11 insertions(+), 9 deletions(-)

diff --git a/drivers/net/vrf.c b/drivers/net/vrf.c
index bf2fac913942..aee9e0fa0842 100644
--- a/drivers/net/vrf.c
+++ b/drivers/net/vrf.c
@@ -1653,21 +1653,23 @@ static void vrf_setup(struct net_device *dev)
 	eth_hw_addr_random(dev);
 
 	/* don't acquire vrf device's netif_tx_lock when transmitting */
-	dev->features |= NETIF_F_LLTX;
+	netdev_feature_set_bit(NETIF_F_LLTX_BIT, &dev->features);
 
 	/* don't allow vrf devices to change network namespaces. */
-	dev->features |= NETIF_F_NETNS_LOCAL;
+	netdev_feature_set_bit(NETIF_F_NETNS_LOCAL_BIT, &dev->features);
 
 	/* does not make sense for a VLAN to be added to a vrf device */
-	dev->features   |= NETIF_F_VLAN_CHALLENGED;
+	netdev_feature_set_bit(NETIF_F_VLAN_CHALLENGED_BIT, &dev->features);
 
 	/* enable offload features */
-	dev->features   |= NETIF_F_GSO_SOFTWARE;
-	dev->features   |= NETIF_F_RXCSUM | NETIF_F_HW_CSUM | NETIF_F_SCTP_CRC;
-	dev->features   |= NETIF_F_SG | NETIF_F_FRAGLIST | NETIF_F_HIGHDMA;
-
-	dev->hw_features = dev->features;
-	dev->hw_enc_features = dev->features;
+	netdev_feature_set_bits(NETIF_F_GSO_SOFTWARE, &dev->features);
+	netdev_feature_set_bits(NETIF_F_RXCSUM | NETIF_F_HW_CSUM | NETIF_F_SCTP_CRC,
+				&dev->features);
+	netdev_feature_set_bits(NETIF_F_SG | NETIF_F_FRAGLIST | NETIF_F_HIGHDMA,
+				&dev->features);
+
+	netdev_feature_copy(&dev->hw_features, dev->features);
+	netdev_feature_copy(&dev->hw_enc_features, dev->features);
 
 	/* default to no qdisc; user can add if desired */
 	dev->priv_flags |= IFF_NO_QUEUE;
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 060/167] net: sctp: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (58 preceding siblings ...)
  2021-09-29 15:51 ` [RFCv2 net-next 059/167] net: vrf: " Jian Shen
@ 2021-09-29 15:51 ` Jian Shen
  2021-09-29 15:51 ` [RFCv2 net-next 061/167] vxlan: " Jian Shen
                   ` (110 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:51 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 net/sctp/offload.c | 12 +++++++++---
 net/sctp/output.c  |  2 +-
 2 files changed, 10 insertions(+), 4 deletions(-)

diff --git a/net/sctp/offload.c b/net/sctp/offload.c
index eb874e3c399a..c49464811614 100644
--- a/net/sctp/offload.c
+++ b/net/sctp/offload.c
@@ -39,6 +39,7 @@ static struct sk_buff *sctp_gso_segment(struct sk_buff *skb,
 					netdev_features_t features)
 {
 	struct sk_buff *segs = ERR_PTR(-EINVAL);
+	netdev_features_t tmp;
 	struct sctphdr *sh;
 
 	if (!skb_is_gso_sctp(skb))
@@ -50,7 +51,9 @@ static struct sk_buff *sctp_gso_segment(struct sk_buff *skb,
 
 	__skb_pull(skb, sizeof(*sh));
 
-	if (skb_gso_ok(skb, features | NETIF_F_GSO_ROBUST)) {
+	netdev_feature_copy(&tmp, features);
+	netdev_feature_set_bit(NETIF_F_GSO_ROBUST_BIT, &tmp);
+	if (skb_gso_ok(skb, tmp)) {
 		/* Packet is from an untrusted source, reset gso_segs. */
 		struct skb_shared_info *pinfo = skb_shinfo(skb);
 		struct sk_buff *frag_iter;
@@ -68,12 +71,15 @@ static struct sk_buff *sctp_gso_segment(struct sk_buff *skb,
 		goto out;
 	}
 
-	segs = skb_segment(skb, (features | NETIF_F_HW_CSUM) & ~NETIF_F_SG);
+	netdev_feature_copy(&tmp, features);
+	netdev_feature_set_bit(NETIF_F_HW_CSUM_BIT, &tmp);
+	netdev_feature_clear_bit(NETIF_F_SG_BIT, &tmp);
+	segs = skb_segment(skb, tmp);
 	if (IS_ERR(segs))
 		goto out;
 
 	/* All that is left is update SCTP CRC if necessary */
-	if (!(features & NETIF_F_SCTP_CRC)) {
+	if (!netdev_feature_test_bit(NETIF_F_SCTP_CRC_BIT, features)) {
 		for (skb = segs; skb; skb = skb->next) {
 			if (skb->ip_summed == CHECKSUM_PARTIAL) {
 				sh = sctp_hdr(skb);
diff --git a/net/sctp/output.c b/net/sctp/output.c
index 4dfb5ea82b05..bbaadde69358 100644
--- a/net/sctp/output.c
+++ b/net/sctp/output.c
@@ -543,7 +543,7 @@ static int sctp_packet_pack(struct sctp_packet *packet,
 	if (sctp_checksum_disable)
 		return 1;
 
-	if (!(tp->dst->dev->features & NETIF_F_SCTP_CRC) ||
+	if (!netdev_feature_test_bit(NETIF_F_SCTP_CRC_BIT, tp->dst->dev->features) ||
 	    dst_xfrm(tp->dst) || packet->ipfragok || tp->encap_port) {
 		struct sctphdr *sh =
 			(struct sctphdr *)skb_transport_header(head);
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 061/167] vxlan: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (59 preceding siblings ...)
  2021-09-29 15:51 ` [RFCv2 net-next 060/167] net: sctp: " Jian Shen
@ 2021-09-29 15:51 ` Jian Shen
  2021-09-29 15:51 ` [RFCv2 net-next 062/167] xen-netback: " Jian Shen
                   ` (109 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:51 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 drivers/net/vxlan.c | 21 ++++++++++++---------
 include/net/vxlan.h |  4 ++--
 2 files changed, 14 insertions(+), 11 deletions(-)

diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c
index 141635a35c28..1978dfc71056 100644
--- a/drivers/net/vxlan.c
+++ b/drivers/net/vxlan.c
@@ -3294,15 +3294,18 @@ static void vxlan_setup(struct net_device *dev)
 	dev->needs_free_netdev = true;
 	SET_NETDEV_DEVTYPE(dev, &vxlan_type);
 
-	dev->features	|= NETIF_F_LLTX;
-	dev->features	|= NETIF_F_SG | NETIF_F_HW_CSUM | NETIF_F_FRAGLIST;
-	dev->features   |= NETIF_F_RXCSUM;
-	dev->features   |= NETIF_F_GSO_SOFTWARE;
-
-	dev->vlan_features = dev->features;
-	dev->hw_features |= NETIF_F_SG | NETIF_F_HW_CSUM | NETIF_F_FRAGLIST;
-	dev->hw_features |= NETIF_F_RXCSUM;
-	dev->hw_features |= NETIF_F_GSO_SOFTWARE;
+	netdev_feature_set_bit(NETIF_F_LLTX_BIT, &dev->features);
+	netdev_feature_set_bits(NETIF_F_SG | NETIF_F_HW_CSUM | NETIF_F_FRAGLIST,
+				&dev->features);
+	netdev_feature_set_bit(NETIF_F_RXCSUM_BIT, &dev->features);
+	netdev_feature_set_bits(NETIF_F_GSO_SOFTWARE, &dev->features);
+
+	netdev_feature_copy(&dev->vlan_features, dev->features);
+	netdev_feature_set_bits(NETIF_F_SG | NETIF_F_HW_CSUM | NETIF_F_FRAGLIST,
+				&dev->hw_features);
+	netdev_feature_set_bit(NETIF_F_RXCSUM_BIT, &dev->hw_features);
+	netdev_feature_set_bits(NETIF_F_GSO_SOFTWARE, &dev->hw_features);
+
 	netif_keep_dst(dev);
 	dev->priv_flags |= IFF_NO_QUEUE;
 
diff --git a/include/net/vxlan.h b/include/net/vxlan.h
index a801e59e2313..dba14ed12c5d 100644
--- a/include/net/vxlan.h
+++ b/include/net/vxlan.h
@@ -322,8 +322,8 @@ static inline void vxlan_features_check(struct sk_buff *skb,
 	      sizeof(struct udphdr) + sizeof(struct vxlanhdr)) ||
 	     (skb->ip_summed != CHECKSUM_NONE &&
 	      !can_checksum_protocol(*features, inner_eth_hdr(skb)->h_proto))))
-		*features &= ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
-
+		netdev_feature_clear_bits(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK,
+					  features);
 }
 
 /* IP header + UDP + VXLAN + Ethernet header */
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 062/167] xen-netback: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (60 preceding siblings ...)
  2021-09-29 15:51 ` [RFCv2 net-next 061/167] vxlan: " Jian Shen
@ 2021-09-29 15:51 ` Jian Shen
  2021-09-29 15:51 ` [RFCv2 net-next 063/167] xen-netfront: " Jian Shen
                   ` (108 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:51 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 drivers/net/xen-netback/interface.c | 21 ++++++++++++---------
 1 file changed, 12 insertions(+), 9 deletions(-)

diff --git a/drivers/net/xen-netback/interface.c b/drivers/net/xen-netback/interface.c
index f964f0a402a6..8acfc1f1e716 100644
--- a/drivers/net/xen-netback/interface.c
+++ b/drivers/net/xen-netback/interface.c
@@ -380,15 +380,15 @@ static void xenvif_fix_features(struct net_device *dev,
 	struct xenvif *vif = netdev_priv(dev);
 
 	if (!vif->can_sg)
-		*features &= ~NETIF_F_SG;
+		netdev_feature_clear_bit(NETIF_F_SG_BIT, features);
 	if (~(vif->gso_mask) & GSO_BIT(TCPV4))
-		*features &= ~NETIF_F_TSO;
+		netdev_feature_clear_bit(NETIF_F_TSO_BIT, features);
 	if (~(vif->gso_mask) & GSO_BIT(TCPV6))
-		*features &= ~NETIF_F_TSO6;
+		netdev_feature_clear_bit(NETIF_F_TSO6_BIT, features);
 	if (!vif->ip_csum)
-		*features &= ~NETIF_F_IP_CSUM;
+		netdev_feature_clear_bit(NETIF_F_IP_CSUM_BIT, features);
 	if (!vif->ipv6_csum)
-		*features &= ~NETIF_F_IPV6_CSUM;
+		netdev_feature_clear_bit(NETIF_F_IPV6_CSUM_BIT, features);
 }
 
 static const struct xenvif_stat {
@@ -532,10 +532,13 @@ struct xenvif *xenvif_alloc(struct device *parent, domid_t domid,
 	INIT_LIST_HEAD(&vif->fe_mcast_addr);
 
 	dev->netdev_ops	= &xenvif_netdev_ops;
-	dev->hw_features = NETIF_F_SG |
-		NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM |
-		NETIF_F_TSO | NETIF_F_TSO6 | NETIF_F_FRAGLIST;
-	dev->features = dev->hw_features | NETIF_F_RXCSUM;
+	netdev_feature_zero(&dev->hw_features);
+	netdev_feature_set_bits(NETIF_F_SG |
+				NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM |
+				NETIF_F_TSO | NETIF_F_TSO6 | NETIF_F_FRAGLIST,
+				&dev->hw_features);
+	netdev_feature_copy(&dev->features, dev->hw_features);
+	netdev_feature_set_bit(NETIF_F_RXCSUM_BIT, &dev->features);
 	dev->ethtool_ops = &xenvif_ethtool_ops;
 
 	dev->tx_queue_len = XENVIF_QUEUE_LENGTH;
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 063/167] xen-netfront: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (61 preceding siblings ...)
  2021-09-29 15:51 ` [RFCv2 net-next 062/167] xen-netback: " Jian Shen
@ 2021-09-29 15:51 ` Jian Shen
  2021-09-29 15:51 ` [RFCv2 net-next 064/167] sock: " Jian Shen
                   ` (107 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:51 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 drivers/net/xen-netfront.c | 38 +++++++++++++++++++++-----------------
 1 file changed, 21 insertions(+), 17 deletions(-)

diff --git a/drivers/net/xen-netfront.c b/drivers/net/xen-netfront.c
index 79752f16277a..00c30e6dbe7c 100644
--- a/drivers/net/xen-netfront.c
+++ b/drivers/net/xen-netfront.c
@@ -231,7 +231,7 @@ static const struct attribute_group xennet_dev_group;
 
 static bool xennet_can_sg(struct net_device *dev)
 {
-	return dev->features & NETIF_F_SG;
+	return netdev_feature_test_bit(NETIF_F_SG_BIT, dev->features);
 }
 
 
@@ -1391,28 +1391,29 @@ static void xennet_fix_features(struct net_device *dev,
 {
 	struct netfront_info *np = netdev_priv(dev);
 
-	if (*features & NETIF_F_SG &&
+	if (netdev_feature_test_bit(NETIF_F_SG_BIT, *features) &&
 	    !xenbus_read_unsigned(np->xbdev->otherend, "feature-sg", 0))
-		*features &= ~NETIF_F_SG;
+		netdev_feature_clear_bit(NETIF_F_SG_BIT, features);
 
-	if (*features & NETIF_F_IPV6_CSUM &&
+	if (netdev_feature_test_bit(NETIF_F_IPV6_CSUM_BIT, *features) &&
 	    !xenbus_read_unsigned(np->xbdev->otherend,
 				  "feature-ipv6-csum-offload", 0))
-		*features &= ~NETIF_F_IPV6_CSUM;
+		netdev_feature_clear_bit(NETIF_F_IPV6_CSUM_BIT, features);
 
-	if (*features & NETIF_F_TSO &&
+	if (netdev_feature_test_bit(NETIF_F_TSO_BIT, *features) &&
 	    !xenbus_read_unsigned(np->xbdev->otherend, "feature-gso-tcpv4", 0))
-		*features &= ~NETIF_F_TSO;
+		netdev_feature_clear_bit(NETIF_F_TSO_BIT, features);
 
-	if (*features & NETIF_F_TSO6 &&
+	if (netdev_feature_test_bit(NETIF_F_TSO6_BIT, *features) &&
 	    !xenbus_read_unsigned(np->xbdev->otherend, "feature-gso-tcpv6", 0))
-		*features &= ~NETIF_F_TSO6;
+		netdev_feature_clear_bit(NETIF_F_TSO6_BIT, features);
 }
 
 static int xennet_set_features(struct net_device *dev,
 	netdev_features_t features)
 {
-	if (!(features & NETIF_F_SG) && dev->mtu > ETH_DATA_LEN) {
+	if (!netdev_feature_test_bit(NETIF_F_SG_BIT, features) &&
+	    dev->mtu > ETH_DATA_LEN) {
 		netdev_info(dev, "Reducing MTU because no SG offload");
 		dev->mtu = ETH_DATA_LEN;
 	}
@@ -1604,19 +1605,22 @@ static struct net_device *xennet_create_dev(struct xenbus_device *dev)
 
 	netdev->netdev_ops	= &xennet_netdev_ops;
 
-	netdev->features        = NETIF_F_IP_CSUM | NETIF_F_RXCSUM |
-				  NETIF_F_GSO_ROBUST;
-	netdev->hw_features	= NETIF_F_SG |
-				  NETIF_F_IPV6_CSUM |
-				  NETIF_F_TSO | NETIF_F_TSO6;
-
+	netdev_feature_zero(&netdev->features);
+	netdev_feature_set_bits(NETIF_F_IP_CSUM | NETIF_F_RXCSUM |
+				NETIF_F_GSO_ROBUST, &netdev->features);
+	netdev_feature_zero(&netdev->hw_features);
+	netdev_feature_set_bits(NETIF_F_SG |
+				NETIF_F_IPV6_CSUM |
+				NETIF_F_TSO | NETIF_F_TSO6,
+				&netdev->hw_features);
 	/*
          * Assume that all hw features are available for now. This set
          * will be adjusted by the call to netdev_update_features() in
          * xennet_connect() which is the earliest point where we can
          * negotiate with the backend regarding supported features.
          */
-	netdev->features |= netdev->hw_features;
+	netdev_feature_or(&netdev->features, netdev->features,
+			  netdev->hw_features);
 
 	netdev->ethtool_ops = &xennet_ethtool_ops;
 	netdev->min_mtu = ETH_MIN_MTU;
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 064/167] sock: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (62 preceding siblings ...)
  2021-09-29 15:51 ` [RFCv2 net-next 063/167] xen-netfront: " Jian Shen
@ 2021-09-29 15:51 ` Jian Shen
  2021-09-29 15:51 ` [RFCv2 net-next 065/167] sunrpc: " Jian Shen
                   ` (106 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:51 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 include/net/sock.h               |  7 ++++---
 net/core/skmsg.c                 |  3 ++-
 net/core/sock.c                  | 17 +++++++++++------
 net/ipv4/af_inet.c               |  2 +-
 net/ipv4/tcp.c                   | 10 ++++++----
 net/ipv4/tcp_ipv4.c              |  2 +-
 net/ipv6/af_inet6.c              |  2 +-
 net/ipv6/inet6_connection_sock.c |  2 +-
 net/ipv6/tcp_ipv6.c              |  2 +-
 9 files changed, 28 insertions(+), 19 deletions(-)

diff --git a/include/net/sock.h b/include/net/sock.h
index 879980de3dcd..5c76b5c7ee34 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -2050,8 +2050,8 @@ void sk_setup_caps(struct sock *sk, struct dst_entry *dst);
 
 static inline void sk_nocaps_add(struct sock *sk, netdev_features_t flags)
 {
-	sk->sk_route_nocaps |= flags;
-	sk->sk_route_caps &= ~flags;
+	netdev_feature_or(&sk->sk_route_nocaps, sk->sk_route_nocaps, flags);
+	netdev_feature_andnot(&sk->sk_route_caps, sk->sk_route_caps, flags);
 }
 
 static inline int skb_do_copy_data_nocache(struct sock *sk, struct sk_buff *skb,
@@ -2063,7 +2063,8 @@ static inline int skb_do_copy_data_nocache(struct sock *sk, struct sk_buff *skb,
 		if (!csum_and_copy_from_iter_full(to, copy, &csum, from))
 			return -EFAULT;
 		skb->csum = csum_block_add(skb->csum, csum, offset);
-	} else if (sk->sk_route_caps & NETIF_F_NOCACHE_COPY) {
+	} else if (netdev_feature_test_bit(NETIF_F_NOCACHE_COPY_BIT,
+					   sk->sk_route_caps)) {
 		if (!copy_from_iter_full_nocache(to, copy, from))
 			return -EFAULT;
 	} else if (!copy_from_iter_full(to, copy, from))
diff --git a/net/core/skmsg.c b/net/core/skmsg.c
index 2d6249b28928..2ff4a64eaa62 100644
--- a/net/core/skmsg.c
+++ b/net/core/skmsg.c
@@ -379,7 +379,8 @@ int sk_msg_memcopy_from_iter(struct sock *sk, struct iov_iter *from,
 		copy = (buf_size > bytes) ? bytes : buf_size;
 		to = sg_virt(sge) + msg->sg.copybreak;
 		msg->sg.copybreak += copy;
-		if (sk->sk_route_caps & NETIF_F_NOCACHE_COPY)
+		if (netdev_feature_test_bit(NETIF_F_NOCACHE_COPY_BIT,
+					    sk->sk_route_caps))
 			ret = copy_from_iter_nocache(to, copy, from);
 		else
 			ret = copy_from_iter(to, copy, from);
diff --git a/net/core/sock.c b/net/core/sock.c
index 512e629f9780..84a414606eb7 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -2146,15 +2146,20 @@ void sk_setup_caps(struct sock *sk, struct dst_entry *dst)
 	u32 max_segs = 1;
 
 	sk_dst_set(sk, dst);
-	sk->sk_route_caps = dst->dev->features | sk->sk_route_forced_caps;
-	if (sk->sk_route_caps & NETIF_F_GSO)
-		sk->sk_route_caps |= NETIF_F_GSO_SOFTWARE;
-	sk->sk_route_caps &= ~sk->sk_route_nocaps;
+	netdev_feature_or(&sk->sk_route_caps, dst->dev->features,
+			  sk->sk_route_forced_caps);
+	if (netdev_feature_test_bit(NETIF_F_GSO_BIT, sk->sk_route_caps))
+		netdev_feature_set_bits(NETIF_F_GSO_SOFTWARE,
+					&sk->sk_route_caps);
+	netdev_feature_andnot(&sk->sk_route_caps, sk->sk_route_caps,
+			      sk->sk_route_nocaps);
 	if (sk_can_gso(sk)) {
 		if (dst->header_len && !xfrm_dst_offload_ok(dst)) {
-			sk->sk_route_caps &= ~NETIF_F_GSO_MASK;
+			netdev_feature_clear_bits(NETIF_F_GSO_MASK,
+						  &sk->sk_route_caps);
 		} else {
-			sk->sk_route_caps |= NETIF_F_SG | NETIF_F_HW_CSUM;
+			netdev_feature_set_bits(NETIF_F_SG | NETIF_F_HW_CSUM,
+						&sk->sk_route_caps);
 			sk->sk_gso_max_size = dst->dev->gso_max_size;
 			max_segs = max_t(u32, dst->dev->gso_max_segs, 1);
 		}
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index aeff7736f260..e3384a815537 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -1295,7 +1295,7 @@ int inet_sk_rebuild_header(struct sock *sk)
 		err = PTR_ERR(rt);
 
 		/* Routing failed... */
-		sk->sk_route_caps = 0;
+		netdev_feature_zero(&sk->sk_route_caps);
 		/*
 		 * Other protocols have to map its equivalent state to TCP_SYN_SENT.
 		 * DCCP maps its DCCP_REQUESTING state to TCP_SYN_SENT. -acme
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 414c179c28e0..d573815a3186 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -457,7 +457,8 @@ void tcp_init_sock(struct sock *sk)
 	WRITE_ONCE(sk->sk_rcvbuf, sock_net(sk)->ipv4.sysctl_tcp_rmem[1]);
 
 	sk_sockets_allocated_inc(sk);
-	sk->sk_route_forced_caps = NETIF_F_GSO;
+	netdev_feature_zero(&sk->sk_route_forced_caps);
+	netdev_feature_set_bit(NETIF_F_GSO_BIT, &sk->sk_route_forced_caps);
 }
 EXPORT_SYMBOL(tcp_init_sock);
 
@@ -1106,7 +1107,7 @@ EXPORT_SYMBOL_GPL(do_tcp_sendpages);
 int tcp_sendpage_locked(struct sock *sk, struct page *page, int offset,
 			size_t size, int flags)
 {
-	if (!(sk->sk_route_caps & NETIF_F_SG))
+	if (!netdev_feature_test_bit(NETIF_F_SG_BIT, sk->sk_route_caps))
 		return sock_no_sendpage_locked(sk, page, offset, size, flags);
 
 	tcp_rate_check_app_limited(sk);  /* is sending application-limited? */
@@ -1166,7 +1167,7 @@ static int tcp_sendmsg_fastopen(struct sock *sk, struct msghdr *msg,
 		if (err) {
 			tcp_set_state(sk, TCP_CLOSE);
 			inet->inet_dport = 0;
-			sk->sk_route_caps = 0;
+			netdev_feature_zero(&sk->sk_route_caps);
 		}
 	}
 	flags = (msg->msg_flags & MSG_DONTWAIT) ? O_NONBLOCK : 0;
@@ -1205,7 +1206,8 @@ int tcp_sendmsg_locked(struct sock *sk, struct msghdr *msg, size_t size)
 			goto out_err;
 		}
 
-		zc = sk->sk_route_caps & NETIF_F_SG;
+		zc = netdev_feature_test_bit(NETIF_F_SG_BIT,
+					     sk->sk_route_caps);
 		if (!zc)
 			uarg->zerocopy = 0;
 	}
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 29a57bd159f0..a3720304fa81 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -323,7 +323,7 @@ int tcp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
 	 */
 	tcp_set_state(sk, TCP_CLOSE);
 	ip_rt_put(rt);
-	sk->sk_route_caps = 0;
+	netdev_feature_zero(&sk->sk_route_caps);
 	inet->inet_dport = 0;
 	return err;
 }
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index b5878bb8e419..94a4e1adce7b 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -833,7 +833,7 @@ int inet6_sk_rebuild_header(struct sock *sk)
 
 		dst = ip6_dst_lookup_flow(sock_net(sk), sk, &fl6, final_p);
 		if (IS_ERR(dst)) {
-			sk->sk_route_caps = 0;
+			netdev_feature_zero(&sk->sk_route_caps);
 			sk->sk_err_soft = -PTR_ERR(dst);
 			return PTR_ERR(dst);
 		}
diff --git a/net/ipv6/inet6_connection_sock.c b/net/ipv6/inet6_connection_sock.c
index 5a9f4d722f35..6d8314846355 100644
--- a/net/ipv6/inet6_connection_sock.c
+++ b/net/ipv6/inet6_connection_sock.c
@@ -121,7 +121,7 @@ int inet6_csk_xmit(struct sock *sk, struct sk_buff *skb, struct flowi *fl_unused
 	dst = inet6_csk_route_socket(sk, &fl6);
 	if (IS_ERR(dst)) {
 		sk->sk_err_soft = -PTR_ERR(dst);
-		sk->sk_route_caps = 0;
+		netdev_feature_zero(&sk->sk_route_caps);
 		kfree_skb(skb);
 		return PTR_ERR(dst);
 	}
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 8cf5ff2e9504..9152b6013a1d 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -341,7 +341,7 @@ static int tcp_v6_connect(struct sock *sk, struct sockaddr *uaddr,
 	tcp_set_state(sk, TCP_CLOSE);
 failure:
 	inet->inet_dport = 0;
-	sk->sk_route_caps = 0;
+	netdev_feature_zero(&sk->sk_route_caps);
 	return err;
 }
 
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 065/167] sunrpc: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (63 preceding siblings ...)
  2021-09-29 15:51 ` [RFCv2 net-next 064/167] sock: " Jian Shen
@ 2021-09-29 15:51 ` Jian Shen
  2021-09-29 15:51 ` [RFCv2 net-next 066/167] net: caif: " Jian Shen
                   ` (105 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:51 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 net/sunrpc/sunrpc.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/sunrpc/sunrpc.h b/net/sunrpc/sunrpc.h
index 2f59464e6524..30dd2c8efadb 100644
--- a/net/sunrpc/sunrpc.h
+++ b/net/sunrpc/sunrpc.h
@@ -44,7 +44,7 @@ static inline int sock_is_loopback(struct sock *sk)
 	rcu_read_lock();
 	dst = rcu_dereference(sk->sk_dst_cache);
 	if (dst && dst->dev &&
-	    (dst->dev->features & NETIF_F_LOOPBACK))
+	    (netdev_feature_test_bit(NETIF_F_LOOPBACK_BIT, dst->dev->features)))
 		loopback = 1;
 	rcu_read_unlock();
 	return loopback;
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 066/167] net: caif: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (64 preceding siblings ...)
  2021-09-29 15:51 ` [RFCv2 net-next 065/167] sunrpc: " Jian Shen
@ 2021-09-29 15:51 ` Jian Shen
  2021-09-29 15:51 ` [RFCv2 net-next 067/167] net: loopback: " Jian Shen
                   ` (104 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:51 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 drivers/net/caif/caif_serial.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/caif/caif_serial.c b/drivers/net/caif/caif_serial.c
index 2a7af611d43a..96255e886ce0 100644
--- a/drivers/net/caif/caif_serial.c
+++ b/drivers/net/caif/caif_serial.c
@@ -398,7 +398,7 @@ static void caifdev_setup(struct net_device *dev)
 {
 	struct ser_device *serdev = netdev_priv(dev);
 
-	dev->features = 0;
+	netdev_feature_zero(&dev->features);
 	dev->netdev_ops = &netdev_ops;
 	dev->type = ARPHRD_CAIF;
 	dev->flags = IFF_POINTOPOINT | IFF_NOARP;
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 067/167] net: loopback: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (65 preceding siblings ...)
  2021-09-29 15:51 ` [RFCv2 net-next 066/167] net: caif: " Jian Shen
@ 2021-09-29 15:51 ` Jian Shen
  2021-09-29 15:51 ` [RFCv2 net-next 068/167] net: dummy: use netdev_feature helpers Jian Shen
                   ` (103 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:51 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 drivers/net/loopback.c | 20 +++++++++-----------
 1 file changed, 9 insertions(+), 11 deletions(-)

diff --git a/drivers/net/loopback.c b/drivers/net/loopback.c
index a1c77cc00416..23cad6442959 100644
--- a/drivers/net/loopback.c
+++ b/drivers/net/loopback.c
@@ -174,17 +174,15 @@ static void gen_lo_setup(struct net_device *dev,
 	dev->flags		= IFF_LOOPBACK;
 	dev->priv_flags		|= IFF_LIVE_ADDR_CHANGE | IFF_NO_QUEUE;
 	netif_keep_dst(dev);
-	dev->hw_features	= NETIF_F_GSO_SOFTWARE;
-	dev->features		= NETIF_F_SG | NETIF_F_FRAGLIST
-		| NETIF_F_GSO_SOFTWARE
-		| NETIF_F_HW_CSUM
-		| NETIF_F_RXCSUM
-		| NETIF_F_SCTP_CRC
-		| NETIF_F_HIGHDMA
-		| NETIF_F_LLTX
-		| NETIF_F_NETNS_LOCAL
-		| NETIF_F_VLAN_CHALLENGED
-		| NETIF_F_LOOPBACK;
+	netdev_feature_zero(&dev->hw_features);
+	netdev_feature_set_bits(NETIF_F_GSO_SOFTWARE, &dev->hw_features);
+	netdev_feature_zero(&dev->features);
+	netdev_feature_set_bits(NETIF_F_SG | NETIF_F_FRAGLIST |
+				NETIF_F_GSO_SOFTWARE | NETIF_F_HW_CSUM |
+				NETIF_F_RXCSUM | NETIF_F_SCTP_CRC |
+				NETIF_F_HIGHDMA | NETIF_F_LLTX |
+				NETIF_F_NETNS_LOCAL | NETIF_F_VLAN_CHALLENGED |
+				NETIF_F_LOOPBACK, &dev->hw_features);
 	dev->ethtool_ops	= eth_ops;
 	dev->header_ops		= hdr_ops;
 	dev->netdev_ops		= dev_ops;
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 068/167] net: dummy: use netdev_feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (66 preceding siblings ...)
  2021-09-29 15:51 ` [RFCv2 net-next 067/167] net: loopback: " Jian Shen
@ 2021-09-29 15:51 ` Jian Shen
  2021-09-29 15:51 ` [RFCv2 net-next 069/167] net: mac80211: use netdev feature helpers Jian Shen
                   ` (102 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:51 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 drivers/net/dummy.c | 14 ++++++++------
 1 file changed, 8 insertions(+), 6 deletions(-)

diff --git a/drivers/net/dummy.c b/drivers/net/dummy.c
index f82ad7419508..54ebb638a57b 100644
--- a/drivers/net/dummy.c
+++ b/drivers/net/dummy.c
@@ -123,12 +123,14 @@ static void dummy_setup(struct net_device *dev)
 	dev->flags |= IFF_NOARP;
 	dev->flags &= ~IFF_MULTICAST;
 	dev->priv_flags |= IFF_LIVE_ADDR_CHANGE | IFF_NO_QUEUE;
-	dev->features	|= NETIF_F_SG | NETIF_F_FRAGLIST;
-	dev->features	|= NETIF_F_GSO_SOFTWARE;
-	dev->features	|= NETIF_F_HW_CSUM | NETIF_F_HIGHDMA | NETIF_F_LLTX;
-	dev->features	|= NETIF_F_GSO_ENCAP_ALL;
-	dev->hw_features |= dev->features;
-	dev->hw_enc_features |= dev->features;
+	netdev_feature_set_bits(NETIF_F_SG | NETIF_F_FRAGLIST, &dev->features);
+	netdev_feature_set_bits(NETIF_F_GSO_SOFTWARE, &dev->features);
+	netdev_feature_set_bits(NETIF_F_HW_CSUM | NETIF_F_HIGHDMA |
+				NETIF_F_LLTX, &dev->features);
+	netdev_feature_set_bits(NETIF_F_GSO_ENCAP_ALL, &dev->features);
+	netdev_feature_or(&dev->hw_features, dev->hw_features, dev->features);
+	netdev_feature_or(&dev->hw_enc_features, dev->hw_enc_features,
+			  dev->features);
 	eth_hw_addr_random(dev);
 
 	dev->min_mtu = 0;
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 069/167] net: mac80211: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (67 preceding siblings ...)
  2021-09-29 15:51 ` [RFCv2 net-next 068/167] net: dummy: use netdev_feature helpers Jian Shen
@ 2021-09-29 15:51 ` Jian Shen
  2021-09-29 15:51 ` [RFCv2 net-next 070/167] net: ifb: " Jian Shen
                   ` (101 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:51 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 net/mac80211/iface.c | 10 +++++++---
 net/mac80211/main.c  |  5 ++++-
 2 files changed, 11 insertions(+), 4 deletions(-)

diff --git a/net/mac80211/iface.c b/net/mac80211/iface.c
index 62c95597704b..7e7431144727 100644
--- a/net/mac80211/iface.c
+++ b/net/mac80211/iface.c
@@ -2034,13 +2034,17 @@ int ieee80211_if_add(struct ieee80211_local *local, const char *name,
 	ieee80211_setup_sdata(sdata, type);
 
 	if (ndev) {
+		netdev_features_t tmp;
+
 		ndev->ieee80211_ptr->use_4addr = params->use_4addr;
 		if (type == NL80211_IFTYPE_STATION)
 			sdata->u.mgd.use_4addr = params->use_4addr;
 
-		ndev->features |= local->hw.netdev_features;
-		ndev->hw_features |= ndev->features &
-					MAC80211_SUPPORTED_FEATURES_TX;
+		netdev_feature_or(&ndev->features, ndev->features,
+				  local->hw.netdev_features);
+		netdev_feature_copy(&tmp, ndev->features);
+		netdev_feature_and_bits(MAC80211_SUPPORTED_FEATURES_TX, &tmp);
+		netdev_feature_or(&ndev->hw_features, ndev->hw_features, tmp);
 
 		netdev_set_default_ethtool_ops(ndev, &ieee80211_ethtool_ops);
 
diff --git a/net/mac80211/main.c b/net/mac80211/main.c
index 45fb517591ee..06142de8f8db 100644
--- a/net/mac80211/main.c
+++ b/net/mac80211/main.c
@@ -911,6 +911,7 @@ int ieee80211_register_hw(struct ieee80211_hw *hw)
 	int channels, max_bitrates;
 	bool supp_ht, supp_vht, supp_he;
 	struct cfg80211_chan_def dflt_chandef = {};
+	netdev_features_t tmp;
 
 	if (ieee80211_hw_check(hw, QUEUE_CONTROL) &&
 	    (local->hw.offchannel_tx_hw_queue == IEEE80211_INVAL_HW_QUEUE ||
@@ -960,7 +961,9 @@ int ieee80211_register_hw(struct ieee80211_hw *hw)
 	}
 
 	/* Only HW csum features are currently compatible with mac80211 */
-	if (WARN_ON(hw->netdev_features & ~MAC80211_SUPPORTED_FEATURES))
+	netdev_feature_copy(&tmp, hw->netdev_features);
+	netdev_feature_clear_bits(MAC80211_SUPPORTED_FEATURES, &tmp);
+	if (WARN_ON(!netdev_feature_empty(tmp)))
 		return -EINVAL;
 
 	if (hw->max_report_rates == 0)
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 070/167] net: ifb: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (68 preceding siblings ...)
  2021-09-29 15:51 ` [RFCv2 net-next 069/167] net: mac80211: use netdev feature helpers Jian Shen
@ 2021-09-29 15:51 ` Jian Shen
  2021-09-29 15:51 ` [RFCv2 net-next 071/167] net: bareudp: " Jian Shen
                   ` (100 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:51 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 drivers/net/ifb.c | 12 +++++++-----
 1 file changed, 7 insertions(+), 5 deletions(-)

diff --git a/drivers/net/ifb.c b/drivers/net/ifb.c
index e9258a9f3702..c7cf224f8bed 100644
--- a/drivers/net/ifb.c
+++ b/drivers/net/ifb.c
@@ -213,11 +213,13 @@ static void ifb_setup(struct net_device *dev)
 	ether_setup(dev);
 	dev->tx_queue_len = TX_Q_LIMIT;
 
-	dev->features |= IFB_FEATURES;
-	dev->hw_features |= dev->features;
-	dev->hw_enc_features |= dev->features;
-	dev->vlan_features |= IFB_FEATURES & ~(NETIF_F_HW_VLAN_CTAG_TX |
-					       NETIF_F_HW_VLAN_STAG_TX);
+	netdev_feature_set_bits(IFB_FEATURES, &dev->features);
+	netdev_feature_or(&dev->hw_features, dev->hw_features, dev->features);
+	netdev_feature_or(&dev->hw_enc_features, dev->hw_enc_features,
+			  dev->features);
+	netdev_feature_set_bits(IFB_FEATURES & ~(NETIF_F_HW_VLAN_CTAG_TX |
+						 NETIF_F_HW_VLAN_STAG_TX),
+				&dev->vlan_features);
 
 	dev->flags |= IFF_NOARP;
 	dev->flags &= ~IFF_MULTICAST;
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 071/167] net: bareudp: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (69 preceding siblings ...)
  2021-09-29 15:51 ` [RFCv2 net-next 070/167] net: ifb: " Jian Shen
@ 2021-09-29 15:51 ` Jian Shen
  2021-09-29 15:51 ` [RFCv2 net-next 072/167] net: rionet: " Jian Shen
                   ` (99 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:51 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 drivers/net/bareudp.c | 16 +++++++++-------
 1 file changed, 9 insertions(+), 7 deletions(-)

diff --git a/drivers/net/bareudp.c b/drivers/net/bareudp.c
index 54e321a695ce..88be2288a78c 100644
--- a/drivers/net/bareudp.c
+++ b/drivers/net/bareudp.c
@@ -534,13 +534,15 @@ static void bareudp_setup(struct net_device *dev)
 	dev->netdev_ops = &bareudp_netdev_ops;
 	dev->needs_free_netdev = true;
 	SET_NETDEV_DEVTYPE(dev, &bareudp_type);
-	dev->features    |= NETIF_F_SG | NETIF_F_HW_CSUM | NETIF_F_FRAGLIST;
-	dev->features    |= NETIF_F_RXCSUM;
-	dev->features    |= NETIF_F_LLTX;
-	dev->features    |= NETIF_F_GSO_SOFTWARE;
-	dev->hw_features |= NETIF_F_SG | NETIF_F_HW_CSUM | NETIF_F_FRAGLIST;
-	dev->hw_features |= NETIF_F_RXCSUM;
-	dev->hw_features |= NETIF_F_GSO_SOFTWARE;
+	netdev_feature_set_bits(NETIF_F_SG | NETIF_F_HW_CSUM | NETIF_F_FRAGLIST,
+				&dev->features);
+	netdev_feature_set_bit(NETIF_F_RXCSUM_BIT, &dev->features);
+	netdev_feature_set_bit(NETIF_F_LLTX_BIT, &dev->features);
+	netdev_feature_set_bits(NETIF_F_GSO_SOFTWARE, &dev->features);
+	netdev_feature_set_bits(NETIF_F_SG | NETIF_F_HW_CSUM | NETIF_F_FRAGLIST,
+				&dev->hw_features);
+	netdev_feature_set_bit(NETIF_F_RXCSUM_BIT, &dev->hw_features);
+	netdev_feature_set_bits(NETIF_F_GSO_SOFTWARE, &dev->hw_features);
 	dev->hard_header_len = 0;
 	dev->addr_len = 0;
 	dev->mtu = ETH_DATA_LEN;
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 072/167] net: rionet: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (70 preceding siblings ...)
  2021-09-29 15:51 ` [RFCv2 net-next 071/167] net: bareudp: " Jian Shen
@ 2021-09-29 15:51 ` Jian Shen
  2021-09-29 15:52 ` [RFCv2 net-next 073/167] net: gtp: " Jian Shen
                   ` (98 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:51 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 drivers/net/rionet.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/rionet.c b/drivers/net/rionet.c
index 2056d6ad04b5..eac3e882cf5a 100644
--- a/drivers/net/rionet.c
+++ b/drivers/net/rionet.c
@@ -513,7 +513,8 @@ static int rionet_setup_netdev(struct rio_mport *mport, struct net_device *ndev)
 	/* MTU range: 68 - 4082 */
 	ndev->min_mtu = ETH_MIN_MTU;
 	ndev->max_mtu = RIONET_MAX_MTU;
-	ndev->features = NETIF_F_LLTX;
+	netdev_feature_zero(&ndev->features);
+	netdev_feature_set_bit(NETIF_F_LLTX_BIT, &ndev->features);
 	SET_NETDEV_DEV(ndev, &mport->dev);
 	ndev->ethtool_ops = &rionet_ethtool_ops;
 
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 073/167] net: gtp: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (71 preceding siblings ...)
  2021-09-29 15:51 ` [RFCv2 net-next 072/167] net: rionet: " Jian Shen
@ 2021-09-29 15:52 ` Jian Shen
  2021-09-29 15:52 ` [RFCv2 net-next 074/167] net: vsockmon: " Jian Shen
                   ` (97 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:52 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 drivers/net/gtp.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/gtp.c b/drivers/net/gtp.c
index 30e0a10595a1..ddc55d3432f7 100644
--- a/drivers/net/gtp.c
+++ b/drivers/net/gtp.c
@@ -639,7 +639,7 @@ static void gtp_link_setup(struct net_device *dev)
 	dev->flags = IFF_POINTOPOINT | IFF_NOARP | IFF_MULTICAST;
 
 	dev->priv_flags	|= IFF_NO_QUEUE;
-	dev->features	|= NETIF_F_LLTX;
+	netdev_feature_set_bit(NETIF_F_LLTX_BIT, &dev->features);
 	netif_keep_dst(dev);
 
 	dev->needed_headroom	= LL_MAX_HEADER + max_gtp_header_len;
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 074/167] net: vsockmon: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (72 preceding siblings ...)
  2021-09-29 15:52 ` [RFCv2 net-next 073/167] net: gtp: " Jian Shen
@ 2021-09-29 15:52 ` Jian Shen
  2021-09-29 15:52 ` [RFCv2 net-next 075/167] net: nlmon: " Jian Shen
                   ` (96 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:52 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 drivers/net/vsockmon.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/net/vsockmon.c b/drivers/net/vsockmon.c
index b1bb1b04b664..047d04115fd9 100644
--- a/drivers/net/vsockmon.c
+++ b/drivers/net/vsockmon.c
@@ -106,8 +106,10 @@ static void vsockmon_setup(struct net_device *dev)
 	dev->ethtool_ops = &vsockmon_ethtool_ops;
 	dev->needs_free_netdev = true;
 
-	dev->features = NETIF_F_SG | NETIF_F_FRAGLIST |
-			NETIF_F_HIGHDMA | NETIF_F_LLTX;
+	netdev_feature_zero(&dev->features);
+	netdev_feature_set_bits(NETIF_F_SG | NETIF_F_FRAGLIST |
+				NETIF_F_HIGHDMA | NETIF_F_LLTX,
+				&dev->features);
 
 	dev->flags = IFF_NOARP;
 
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 075/167] net: nlmon: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (73 preceding siblings ...)
  2021-09-29 15:52 ` [RFCv2 net-next 074/167] net: vsockmon: " Jian Shen
@ 2021-09-29 15:52 ` Jian Shen
  2021-09-29 15:52 ` [RFCv2 net-next 076/167] net: wireguard: " Jian Shen
                   ` (95 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:52 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 drivers/net/nlmon.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/net/nlmon.c b/drivers/net/nlmon.c
index 5e19a6839dea..66af97852b76 100644
--- a/drivers/net/nlmon.c
+++ b/drivers/net/nlmon.c
@@ -89,8 +89,9 @@ static void nlmon_setup(struct net_device *dev)
 	dev->ethtool_ops = &nlmon_ethtool_ops;
 	dev->needs_free_netdev = true;
 
-	dev->features = NETIF_F_SG | NETIF_F_FRAGLIST |
-			NETIF_F_HIGHDMA | NETIF_F_LLTX;
+	netdev_feature_zero(&dev->features);
+	netdev_feature_set_bits(NETIF_F_SG | NETIF_F_FRAGLIST |
+				NETIF_F_HIGHDMA | NETIF_F_LLTX, &dev->features);
 	dev->flags = IFF_NOARP;
 
 	/* That's rather a softlimit here, which, of course,
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 076/167] net: wireguard: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (74 preceding siblings ...)
  2021-09-29 15:52 ` [RFCv2 net-next 075/167] net: nlmon: " Jian Shen
@ 2021-09-29 15:52 ` Jian Shen
  2021-09-29 15:52 ` [RFCv2 net-next 077/167] net: can: " Jian Shen
                   ` (94 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:52 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 drivers/net/wireguard/device.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/net/wireguard/device.c b/drivers/net/wireguard/device.c
index 551ddaaaf540..9c58f5208909 100644
--- a/drivers/net/wireguard/device.c
+++ b/drivers/net/wireguard/device.c
@@ -271,10 +271,10 @@ static void wg_setup(struct net_device *dev)
 	dev->type = ARPHRD_NONE;
 	dev->flags = IFF_POINTOPOINT | IFF_NOARP;
 	dev->priv_flags |= IFF_NO_QUEUE;
-	dev->features |= NETIF_F_LLTX;
-	dev->features |= WG_NETDEV_FEATURES;
-	dev->hw_features |= WG_NETDEV_FEATURES;
-	dev->hw_enc_features |= WG_NETDEV_FEATURES;
+	netdev_feature_set_bit(NETIF_F_LLTX_BIT, &dev->features);
+	netdev_feature_set_bits(WG_NETDEV_FEATURES, &dev->features);
+	netdev_feature_set_bits(WG_NETDEV_FEATURES, &dev->hw_features);
+	netdev_feature_set_bits(WG_NETDEV_FEATURES, &dev->hw_enc_features);
 	dev->mtu = ETH_DATA_LEN - overhead;
 	dev->max_mtu = round_down(INT_MAX, MESSAGE_PADDING_MULTIPLE) - overhead;
 
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 077/167] net: can: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (75 preceding siblings ...)
  2021-09-29 15:52 ` [RFCv2 net-next 076/167] net: wireguard: " Jian Shen
@ 2021-09-29 15:52 ` Jian Shen
  2021-09-29 15:52 ` [RFCv2 net-next 078/167] net: ppp: " Jian Shen
                   ` (93 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:52 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 drivers/net/can/dev/dev.c | 3 ++-
 drivers/net/can/slcan.c   | 3 ++-
 2 files changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/net/can/dev/dev.c b/drivers/net/can/dev/dev.c
index e3d840b81357..97c61943125e 100644
--- a/drivers/net/can/dev/dev.c
+++ b/drivers/net/can/dev/dev.c
@@ -233,7 +233,8 @@ void can_setup(struct net_device *dev)
 
 	/* New-style flags. */
 	dev->flags = IFF_NOARP;
-	dev->features = NETIF_F_HW_CSUM;
+	netdev_feature_zero(&dev->features);
+	netdev_feature_set_bit(NETIF_F_HW_CSUM_BIT, &dev->features);
 }
 
 /* Allocate and setup space for the CAN network device */
diff --git a/drivers/net/can/slcan.c b/drivers/net/can/slcan.c
index d42ec7d1bc14..8af2cf00a021 100644
--- a/drivers/net/can/slcan.c
+++ b/drivers/net/can/slcan.c
@@ -450,7 +450,8 @@ static void slc_setup(struct net_device *dev)
 
 	/* New-style flags. */
 	dev->flags		= IFF_NOARP;
-	dev->features           = NETIF_F_HW_CSUM;
+	netdev_feature_zero(&dev->features);
+	netdev_feature_set_bit(NETIF_F_HW_CSUM_BIT, &dev->features);
 }
 
 /******************************************
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 078/167] net: ppp: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (76 preceding siblings ...)
  2021-09-29 15:52 ` [RFCv2 net-next 077/167] net: can: " Jian Shen
@ 2021-09-29 15:52 ` Jian Shen
  2021-09-29 15:52 ` [RFCv2 net-next 079/167] net: ipa: " Jian Shen
                   ` (92 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:52 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 drivers/net/ppp/ppp_generic.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ppp/ppp_generic.c b/drivers/net/ppp/ppp_generic.c
index fb52cd175b45..d1bc1f10ddb7 100644
--- a/drivers/net/ppp/ppp_generic.c
+++ b/drivers/net/ppp/ppp_generic.c
@@ -1610,7 +1610,7 @@ static void ppp_setup(struct net_device *dev)
 	dev->netdev_ops = &ppp_netdev_ops;
 	SET_NETDEV_DEVTYPE(dev, &ppp_type);
 
-	dev->features |= NETIF_F_LLTX;
+	netdev_feature_set_bit(NETIF_F_LLTX_BIT, &dev->features);
 
 	dev->hard_header_len = PPP_HDRLEN;
 	dev->mtu = PPP_MRU;
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 079/167] net: ipa: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (77 preceding siblings ...)
  2021-09-29 15:52 ` [RFCv2 net-next 078/167] net: ppp: " Jian Shen
@ 2021-09-29 15:52 ` Jian Shen
  2021-09-29 15:52 ` [RFCv2 net-next 080/167] net: fjes: " Jian Shen
                   ` (91 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:52 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 drivers/net/ipa/ipa_modem.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ipa/ipa_modem.c b/drivers/net/ipa/ipa_modem.c
index ad116bcc0580..a7657e16a72a 100644
--- a/drivers/net/ipa/ipa_modem.c
+++ b/drivers/net/ipa/ipa_modem.c
@@ -216,7 +216,8 @@ static void ipa_modem_netdev_setup(struct net_device *netdev)
 	netdev->needed_headroom = sizeof(struct rmnet_map_header);
 	netdev->needed_tailroom = IPA_NETDEV_TAILROOM;
 	netdev->watchdog_timeo = IPA_NETDEV_TIMEOUT * HZ;
-	netdev->hw_features = NETIF_F_SG;
+	netdev_feature_zero(&netdev->hw_features);
+	netdev_feature_set_bit(NETIF_F_SG_BIT, &netdev->hw_features);
 }
 
 /** ipa_modem_suspend() - suspend callback
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 080/167] net: fjes: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (78 preceding siblings ...)
  2021-09-29 15:52 ` [RFCv2 net-next 079/167] net: ipa: " Jian Shen
@ 2021-09-29 15:52 ` Jian Shen
  2021-09-29 15:52 ` [RFCv2 net-next 081/167] net: usb: " Jian Shen
                   ` (90 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:52 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 drivers/net/fjes/fjes_main.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/fjes/fjes_main.c b/drivers/net/fjes/fjes_main.c
index 185c8a398681..583d21ad25c8 100644
--- a/drivers/net/fjes/fjes_main.c
+++ b/drivers/net/fjes/fjes_main.c
@@ -1344,7 +1344,8 @@ static void fjes_netdev_setup(struct net_device *netdev)
 	netdev->mtu = fjes_support_mtu[3];
 	netdev->min_mtu = fjes_support_mtu[0];
 	netdev->max_mtu = fjes_support_mtu[3];
-	netdev->features |= NETIF_F_HW_VLAN_CTAG_FILTER;
+	netdev_feature_set_bit(NETIF_F_HW_VLAN_CTAG_FILTER_BIT,
+			       &netdev->features);
 }
 
 static void fjes_irq_watch_task(struct work_struct *work)
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 081/167] net: usb: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (79 preceding siblings ...)
  2021-09-29 15:52 ` [RFCv2 net-next 080/167] net: fjes: " Jian Shen
@ 2021-09-29 15:52 ` Jian Shen
  2021-09-29 15:52 ` [RFCv2 net-next 082/167] net: wireless: " Jian Shen
                   ` (89 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:52 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 drivers/net/usb/aqc111.c               | 42 +++++++++--------
 drivers/net/usb/ax88179_178a.c         | 26 ++++++-----
 drivers/net/usb/cdc-phonet.c           |  2 +-
 drivers/net/usb/cdc_mbim.c             |  4 +-
 drivers/net/usb/lan78xx.c              | 34 ++++++++------
 drivers/net/usb/r8152.c                | 64 ++++++++++++++++----------
 drivers/net/usb/smsc75xx.c             | 14 +++---
 drivers/net/usb/smsc95xx.c             | 19 +++++---
 drivers/usb/gadget/function/f_phonet.c |  2 +-
 9 files changed, 123 insertions(+), 84 deletions(-)

diff --git a/drivers/net/usb/aqc111.c b/drivers/net/usb/aqc111.c
index 73b97f4cc1ec..1363f64f4c86 100644
--- a/drivers/net/usb/aqc111.c
+++ b/drivers/net/usb/aqc111.c
@@ -567,27 +567,28 @@ static int aqc111_set_features(struct net_device *net,
 {
 	struct usbnet *dev = netdev_priv(net);
 	struct aqc111_data *aqc111_data = dev->driver_priv;
-	netdev_features_t changed = net->features ^ features;
+	netdev_features_t changed;
 	u16 reg16 = 0;
 	u8 reg8 = 0;
 
-	if (changed & NETIF_F_IP_CSUM) {
+	netdev_feature_xor(&changed, net->features, features);
+	if (netdev_feature_test_bit(NETIF_F_IP_CSUM_BIT, changed)) {
 		aqc111_read_cmd(dev, AQ_ACCESS_MAC, SFR_TXCOE_CTL, 1, 1, &reg8);
 		reg8 ^= SFR_TXCOE_TCP | SFR_TXCOE_UDP;
 		aqc111_write_cmd(dev, AQ_ACCESS_MAC, SFR_TXCOE_CTL,
 				 1, 1, &reg8);
 	}
 
-	if (changed & NETIF_F_IPV6_CSUM) {
+	if (netdev_feature_test_bit(NETIF_F_IPV6_CSUM_BIT, changed)) {
 		aqc111_read_cmd(dev, AQ_ACCESS_MAC, SFR_TXCOE_CTL, 1, 1, &reg8);
 		reg8 ^= SFR_TXCOE_TCPV6 | SFR_TXCOE_UDPV6;
 		aqc111_write_cmd(dev, AQ_ACCESS_MAC, SFR_TXCOE_CTL,
 				 1, 1, &reg8);
 	}
 
-	if (changed & NETIF_F_RXCSUM) {
+	if (netdev_feature_test_bit(NETIF_F_RXCSUM_BIT, changed)) {
 		aqc111_read_cmd(dev, AQ_ACCESS_MAC, SFR_RXCOE_CTL, 1, 1, &reg8);
-		if (features & NETIF_F_RXCSUM) {
+		if (netdev_feature_test_bit(NETIF_F_RXCSUM_BIT, features)) {
 			aqc111_data->rx_checksum = 1;
 			reg8 &= ~(SFR_RXCOE_IP | SFR_RXCOE_TCP | SFR_RXCOE_UDP |
 				  SFR_RXCOE_TCPV6 | SFR_RXCOE_UDPV6);
@@ -600,8 +601,9 @@ static int aqc111_set_features(struct net_device *net,
 		aqc111_write_cmd(dev, AQ_ACCESS_MAC, SFR_RXCOE_CTL,
 				 1, 1, &reg8);
 	}
-	if (changed & NETIF_F_HW_VLAN_CTAG_FILTER) {
-		if (features & NETIF_F_HW_VLAN_CTAG_FILTER) {
+	if (netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_FILTER_BIT, changed)) {
+		if (netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_FILTER_BIT,
+					    features)) {
 			u16 i = 0;
 
 			for (i = 0; i < 256; i++) {
@@ -731,9 +733,10 @@ static int aqc111_bind(struct usbnet *dev, struct usb_interface *intf)
 	if (usb_device_no_sg_constraint(dev->udev))
 		dev->can_dma_sg = 1;
 
-	dev->net->hw_features |= AQ_SUPPORT_HW_FEATURE;
-	dev->net->features |= AQ_SUPPORT_FEATURE;
-	dev->net->vlan_features |= AQ_SUPPORT_VLAN_FEATURE;
+	netdev_feature_set_bits(AQ_SUPPORT_HW_FEATURE, &dev->net->hw_features);
+	netdev_feature_set_bits(AQ_SUPPORT_FEATURE, &dev->net->features);
+	netdev_feature_set_bits(AQ_SUPPORT_VLAN_FEATURE,
+				&dev->net->vlan_features);
 
 	netif_set_gso_max_size(dev->net, 65535);
 
@@ -881,17 +884,17 @@ static void aqc111_configure_csum_offload(struct usbnet *dev)
 {
 	u8 reg8 = 0;
 
-	if (dev->net->features & NETIF_F_RXCSUM) {
+	if (netdev_feature_test_bit(NETIF_F_RXCSUM_BIT, dev->net->features)) {
 		reg8 |= SFR_RXCOE_IP | SFR_RXCOE_TCP | SFR_RXCOE_UDP |
 			SFR_RXCOE_TCPV6 | SFR_RXCOE_UDPV6;
 	}
 	aqc111_write_cmd(dev, AQ_ACCESS_MAC, SFR_RXCOE_CTL, 1, 1, &reg8);
 
 	reg8 = 0;
-	if (dev->net->features & NETIF_F_IP_CSUM)
+	if (netdev_feature_test_bit(NETIF_F_IP_CSUM_BIT, dev->net->features))
 		reg8 |= SFR_TXCOE_IP | SFR_TXCOE_TCP | SFR_TXCOE_UDP;
 
-	if (dev->net->features & NETIF_F_IPV6_CSUM)
+	if (netdev_feature_test_bit(NETIF_F_IPV6_CSUM_BIT, dev->net->features))
 		reg8 |= SFR_TXCOE_TCPV6 | SFR_TXCOE_UDPV6;
 
 	aqc111_write_cmd(dev, AQ_ACCESS_MAC, SFR_TXCOE_CTL, 1, 1, &reg8);
@@ -908,7 +911,8 @@ static int aqc111_link_reset(struct usbnet *dev)
 
 		/* Vlan Tag Filter */
 		reg8 = SFR_VLAN_CONTROL_VSO;
-		if (dev->net->features & NETIF_F_HW_VLAN_CTAG_FILTER)
+		if (netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_FILTER_BIT,
+					    dev->net->features))
 			reg8 |= SFR_VLAN_CONTROL_VFE;
 
 		aqc111_write_cmd(dev, AQ_ACCESS_MAC, SFR_VLAN_ID_CONTROL,
@@ -996,9 +1000,10 @@ static int aqc111_reset(struct usbnet *dev)
 	if (usb_device_no_sg_constraint(dev->udev))
 		dev->can_dma_sg = 1;
 
-	dev->net->hw_features |= AQ_SUPPORT_HW_FEATURE;
-	dev->net->features |= AQ_SUPPORT_FEATURE;
-	dev->net->vlan_features |= AQ_SUPPORT_VLAN_FEATURE;
+	netdev_feature_set_bits(AQ_SUPPORT_HW_FEATURE, &dev->net->hw_features);
+	netdev_feature_set_bits(AQ_SUPPORT_FEATURE, &dev->net->features);
+	netdev_feature_set_bits(AQ_SUPPORT_VLAN_FEATURE,
+				&dev->net->vlan_features);
 
 	/* Power up ethernet PHY */
 	aqc111_data->phy_cfg = AQ_PHY_POWER_EN;
@@ -1209,7 +1214,8 @@ static struct sk_buff *aqc111_tx_fixup(struct usbnet *dev, struct sk_buff *skb,
 			   AQ_TX_DESC_VLAN_SHIFT;
 	}
 
-	if (!dev->can_dma_sg && (dev->net->features & NETIF_F_SG) &&
+	if (!dev->can_dma_sg &&
+	    netdev_feature_test_bit(NETIF_F_SG_BIT, dev->net->features) &&
 	    skb_linearize(skb))
 		return NULL;
 
diff --git a/drivers/net/usb/ax88179_178a.c b/drivers/net/usb/ax88179_178a.c
index f25448a08870..0099d55f9c90 100644
--- a/drivers/net/usb/ax88179_178a.c
+++ b/drivers/net/usb/ax88179_178a.c
@@ -952,21 +952,23 @@ ax88179_set_features(struct net_device *net, netdev_features_t features)
 {
 	u8 tmp;
 	struct usbnet *dev = netdev_priv(net);
-	netdev_features_t changed = net->features ^ features;
+	netdev_features_t changed;
 
-	if (changed & NETIF_F_IP_CSUM) {
+	netdev_feature_xor(&changed, net->features, features);
+
+	if (netdev_feature_test_bit(NETIF_F_IP_CSUM_BIT, changed)) {
 		ax88179_read_cmd(dev, AX_ACCESS_MAC, AX_TXCOE_CTL, 1, 1, &tmp);
 		tmp ^= AX_TXCOE_TCP | AX_TXCOE_UDP;
 		ax88179_write_cmd(dev, AX_ACCESS_MAC, AX_TXCOE_CTL, 1, 1, &tmp);
 	}
 
-	if (changed & NETIF_F_IPV6_CSUM) {
+	if (netdev_feature_test_bit(NETIF_F_IPV6_CSUM_BIT, changed)) {
 		ax88179_read_cmd(dev, AX_ACCESS_MAC, AX_TXCOE_CTL, 1, 1, &tmp);
 		tmp ^= AX_TXCOE_TCPV6 | AX_TXCOE_UDPV6;
 		ax88179_write_cmd(dev, AX_ACCESS_MAC, AX_TXCOE_CTL, 1, 1, &tmp);
 	}
 
-	if (changed & NETIF_F_RXCSUM) {
+	if (netdev_feature_test_bit(NETIF_F_RXCSUM_BIT, changed)) {
 		ax88179_read_cmd(dev, AX_ACCESS_MAC, AX_RXCOE_CTL, 1, 1, &tmp);
 		tmp ^= AX_RXCOE_IP | AX_RXCOE_TCP | AX_RXCOE_UDP |
 		       AX_RXCOE_TCPV6 | AX_RXCOE_UDPV6;
@@ -1377,11 +1379,11 @@ static int ax88179_bind(struct usbnet *dev, struct usb_interface *intf)
 	dev->mii.phy_id = 0x03;
 	dev->mii.supports_gmii = 1;
 
-	dev->net->features |= NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM |
-			      NETIF_F_RXCSUM;
+	netdev_feature_set_bits(NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM |
+				NETIF_F_RXCSUM, &dev->net->features);
 
-	dev->net->hw_features |= NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM |
-				 NETIF_F_RXCSUM;
+	netdev_feature_set_bits(NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM |
+				 NETIF_F_RXCSUM, &dev->net->hw_features);
 
 	/* Enable checksum offload */
 	*tmp = AX_RXCOE_IP | AX_RXCOE_TCP | AX_RXCOE_UDP |
@@ -1664,11 +1666,11 @@ static int ax88179_reset(struct usbnet *dev)
 	ax88179_write_cmd(dev, AX_ACCESS_MAC, AX_PAUSE_WATERLVL_HIGH,
 			  1, 1, tmp);
 
-	dev->net->features |= NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM |
-			      NETIF_F_RXCSUM;
+	netdev_feature_set_bits(NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM |
+				NETIF_F_RXCSUM, &dev->net->features);
 
-	dev->net->hw_features |= NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM |
-				 NETIF_F_RXCSUM;
+	netdev_feature_set_bits(NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM |
+				 NETIF_F_RXCSUM, &dev->net->hw_features);
 
 	/* Enable checksum offload */
 	*tmp = AX_RXCOE_IP | AX_RXCOE_TCP | AX_RXCOE_UDP |
diff --git a/drivers/net/usb/cdc-phonet.c b/drivers/net/usb/cdc-phonet.c
index e1da9102a540..a65be627dc03 100644
--- a/drivers/net/usb/cdc-phonet.c
+++ b/drivers/net/usb/cdc-phonet.c
@@ -275,7 +275,7 @@ static const struct net_device_ops usbpn_ops = {
 
 static void usbpn_setup(struct net_device *dev)
 {
-	dev->features		= 0;
+	netdev_feature_zero(&dev->features);
 	dev->netdev_ops		= &usbpn_ops;
 	dev->header_ops		= &phonet_header_ops;
 	dev->type		= ARPHRD_PHONET;
diff --git a/drivers/net/usb/cdc_mbim.c b/drivers/net/usb/cdc_mbim.c
index 82bb5ed94c48..964e5260364a 100644
--- a/drivers/net/usb/cdc_mbim.c
+++ b/drivers/net/usb/cdc_mbim.c
@@ -184,7 +184,9 @@ static int cdc_mbim_bind(struct usbnet *dev, struct usb_interface *intf)
 	dev->net->flags |= IFF_NOARP;
 
 	/* no need to put the VLAN tci in the packet headers */
-	dev->net->features |= NETIF_F_HW_VLAN_CTAG_TX | NETIF_F_HW_VLAN_CTAG_FILTER;
+	netdev_feature_set_bits(NETIF_F_HW_VLAN_CTAG_TX |
+				NETIF_F_HW_VLAN_CTAG_FILTER,
+				&dev->net->features);
 
 	/* monitor VLAN additions and removals */
 	dev->net->netdev_ops = &cdc_mbim_netdev_ops;
diff --git a/drivers/net/usb/lan78xx.c b/drivers/net/usb/lan78xx.c
index 7a60cc0f6aad..aad719ebf318 100644
--- a/drivers/net/usb/lan78xx.c
+++ b/drivers/net/usb/lan78xx.c
@@ -2445,7 +2445,7 @@ static int lan78xx_set_features(struct net_device *netdev,
 
 	spin_lock_irqsave(&pdata->rfe_ctl_lock, flags);
 
-	if (features & NETIF_F_RXCSUM) {
+	if (netdev_feature_test_bit(NETIF_F_RXCSUM_BIT, features)) {
 		pdata->rfe_ctl |= RFE_CTL_TCPUDP_COE_ | RFE_CTL_IP_COE_;
 		pdata->rfe_ctl |= RFE_CTL_ICMP_COE_ | RFE_CTL_IGMP_COE_;
 	} else {
@@ -2453,12 +2453,12 @@ static int lan78xx_set_features(struct net_device *netdev,
 		pdata->rfe_ctl &= ~(RFE_CTL_ICMP_COE_ | RFE_CTL_IGMP_COE_);
 	}
 
-	if (features & NETIF_F_HW_VLAN_CTAG_RX)
+	if (netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_RX_BIT, features))
 		pdata->rfe_ctl |= RFE_CTL_VLAN_STRIP_;
 	else
 		pdata->rfe_ctl &= ~RFE_CTL_VLAN_STRIP_;
 
-	if (features & NETIF_F_HW_VLAN_CTAG_FILTER)
+	if (netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_FILTER_BIT, features))
 		pdata->rfe_ctl |= RFE_CTL_VLAN_FILTER_;
 	else
 		pdata->rfe_ctl &= ~RFE_CTL_VLAN_FILTER_;
@@ -3250,24 +3250,28 @@ static int lan78xx_bind(struct lan78xx_net *dev, struct usb_interface *intf)
 
 	INIT_WORK(&pdata->set_vlan, lan78xx_deferred_vlan_write);
 
-	dev->net->features = 0;
+	netdev_feature_zero(&dev->net->features);
 
 	if (DEFAULT_TX_CSUM_ENABLE)
-		dev->net->features |= NETIF_F_HW_CSUM;
+		netdev_feature_set_bit(NETIF_F_HW_CSUM_BIT,
+				       &dev->net->features);
 
 	if (DEFAULT_RX_CSUM_ENABLE)
-		dev->net->features |= NETIF_F_RXCSUM;
+		netdev_feature_set_bit(NETIF_F_RXCSUM_BIT, &dev->net->features);
 
 	if (DEFAULT_TSO_CSUM_ENABLE)
-		dev->net->features |= NETIF_F_TSO | NETIF_F_TSO6 | NETIF_F_SG;
+		netdev_feature_set_bits(NETIF_F_TSO | NETIF_F_TSO6 | NETIF_F_SG,
+					&dev->net->features);
 
 	if (DEFAULT_VLAN_RX_OFFLOAD)
-		dev->net->features |= NETIF_F_HW_VLAN_CTAG_RX;
+		netdev_feature_set_bit(NETIF_F_HW_VLAN_CTAG_RX_BIT,
+				       &dev->net->features);
 
 	if (DEFAULT_VLAN_FILTER_ENABLE)
-		dev->net->features |= NETIF_F_HW_VLAN_CTAG_FILTER;
+		netdev_feature_set_bit(NETIF_F_HW_VLAN_CTAG_FILTER_BIT,
+				       &dev->net->features);
 
-	dev->net->hw_features = dev->net->features;
+	netdev_feature_copy(&dev->net->hw_features, dev->net->features);
 
 	ret = lan78xx_setup_irq_domain(dev);
 	if (ret < 0) {
@@ -3334,10 +3338,11 @@ static void lan78xx_rx_csum_offload(struct lan78xx_net *dev,
 	/* HW Checksum offload appears to be flawed if used when not stripping
 	 * VLAN headers. Drop back to S/W checksums under these conditions.
 	 */
-	if (!(dev->net->features & NETIF_F_RXCSUM) ||
+	if (!netdev_feature_test_bit(NETIF_F_RXCSUM_BIT, dev->net->features) ||
 	    unlikely(rx_cmd_a & RX_CMD_A_ICSM_) ||
 	    ((rx_cmd_a & RX_CMD_A_FVTG_) &&
-	     !(dev->net->features & NETIF_F_HW_VLAN_CTAG_RX))) {
+	     !netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_RX_BIT,
+				      dev->net->features))) {
 		skb->ip_summed = CHECKSUM_NONE;
 	} else {
 		skb->csum = ntohs((u16)(rx_cmd_b >> RX_CMD_B_CSUM_SHIFT_));
@@ -3349,7 +3354,8 @@ static void lan78xx_rx_vlan_offload(struct lan78xx_net *dev,
 				    struct sk_buff *skb,
 				    u32 rx_cmd_a, u32 rx_cmd_b)
 {
-	if ((dev->net->features & NETIF_F_HW_VLAN_CTAG_RX) &&
+	if (netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_RX_BIT,
+				    dev->net->features) &&
 	    (rx_cmd_a & RX_CMD_A_FVTG_))
 		__vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q),
 				       (rx_cmd_b & 0xffff));
@@ -3981,7 +3987,7 @@ static void lan78xx_features_check(struct sk_buff *skb,
 				   netdev_features_t *features)
 {
 	if (skb->len + TX_OVERHEAD > MAX_SINGLE_PACKET_SIZE)
-		*features &= ~NETIF_F_GSO_MASK;
+		netdev_feature_clear_bits(NETIF_F_GSO_MASK, features);
 
 	vlan_features_check(skb, features);
 	vxlan_features_check(skb, features);
diff --git a/drivers/net/usb/r8152.c b/drivers/net/usb/r8152.c
index 5bb327b8d8e0..31fb92ae23c2 100644
--- a/drivers/net/usb/r8152.c
+++ b/drivers/net/usb/r8152.c
@@ -2092,11 +2092,13 @@ static void r8152_csum_workaround(struct r8152 *tp, struct sk_buff *skb,
 				  struct sk_buff_head *list)
 {
 	if (skb_shinfo(skb)->gso_size) {
-		netdev_features_t features = tp->netdev->features;
 		struct sk_buff *segs, *seg, *next;
 		struct sk_buff_head seg_list;
+		netdev_features_t features;
 
-		features &= ~(NETIF_F_SG | NETIF_F_IPV6_CSUM | NETIF_F_TSO6);
+		netdev_feature_copy(&features, tp->netdev->features);
+		netdev_feature_clear_bits(NETIF_F_SG | NETIF_F_IPV6_CSUM |
+					  NETIF_F_TSO6, &features);
 		segs = skb_gso_segment(skb, features);
 		if (IS_ERR(segs) || !segs)
 			goto drop;
@@ -2333,7 +2335,7 @@ static u8 r8152_rx_csum(struct r8152 *tp, struct rx_desc *rx_desc)
 	u8 checksum = CHECKSUM_NONE;
 	u32 opts2, opts3;
 
-	if (!(tp->netdev->features & NETIF_F_RXCSUM))
+	if (!netdev_feature_test_bit(NETIF_F_RXCSUM_BIT, tp->netdev->features))
 		goto return_result;
 
 	opts2 = le32_to_cpu(rx_desc->opts2);
@@ -2749,9 +2751,10 @@ static void rtl8152_features_check(struct sk_buff *skb, struct net_device *dev,
 	int offset = skb_transport_offset(skb);
 
 	if ((mss || skb->ip_summed == CHECKSUM_PARTIAL) && offset > max_offset)
-		*features &= ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
+		netdev_feature_clear_bits(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK,
+					  features);
 	else if ((skb->len + sizeof(struct tx_desc)) > agg_buf_sz)
-		*features &= ~NETIF_F_GSO_MASK;
+		netdev_feature_clear_bits(NETIF_F_GSO_MASK, features);
 }
 
 static netdev_tx_t rtl8152_start_xmit(struct sk_buff *skb,
@@ -3230,18 +3233,21 @@ static void rtl_rx_vlan_en(struct r8152 *tp, bool enable)
 static int rtl8152_set_features(struct net_device *dev,
 				netdev_features_t features)
 {
-	netdev_features_t changed = features ^ dev->features;
 	struct r8152 *tp = netdev_priv(dev);
+	netdev_features_t changed;
 	int ret;
 
 	ret = usb_autopm_get_interface(tp->intf);
 	if (ret < 0)
 		goto out;
 
+	netdev_feature_xor(&changed, dev->features, features);
+
 	mutex_lock(&tp->control);
 
-	if (changed & NETIF_F_HW_VLAN_CTAG_RX) {
-		if (features & NETIF_F_HW_VLAN_CTAG_RX)
+	if (netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_RX_BIT, changed)) {
+		if (netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_RX_BIT,
+					    features))
 			rtl_rx_vlan_en(tp, true);
 		else
 			rtl_rx_vlan_en(tp, false);
@@ -5405,7 +5411,8 @@ static void r8152b_exit_oob(struct r8152 *tp)
 	ocp_write_dword(tp, MCU_TYPE_USB, USB_TX_DMA,
 			TEST_MODE_DISABLE | TX_SIZE_ADJUST1);
 
-	rtl_rx_vlan_en(tp, tp->netdev->features & NETIF_F_HW_VLAN_CTAG_RX);
+	rtl_rx_vlan_en(tp, netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_RX_BIT,
+						   tp->netdev->features));
 
 	ocp_write_word(tp, MCU_TYPE_PLA, PLA_RMS, RTL8152_RMS);
 
@@ -5850,7 +5857,8 @@ static void r8153_first_init(struct r8152 *tp)
 
 	wait_oob_link_list_ready(tp);
 
-	rtl_rx_vlan_en(tp, tp->netdev->features & NETIF_F_HW_VLAN_CTAG_RX);
+	rtl_rx_vlan_en(tp, netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_RX_BIT,
+						   tp->netdev->features));
 
 	rtl8153_change_mtu(tp);
 
@@ -6351,7 +6359,8 @@ static void rtl8153c_up(struct r8152 *tp)
 
 	wait_oob_link_list_ready(tp);
 
-	rtl_rx_vlan_en(tp, tp->netdev->features & NETIF_F_HW_VLAN_CTAG_RX);
+	rtl_rx_vlan_en(tp, netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_RX_BIT,
+						   tp->netdev->features));
 
 	rtl8153c_change_mtu(tp);
 
@@ -6456,7 +6465,8 @@ static void rtl8156_up(struct r8152 *tp)
 	ocp_data &= ~MCU_BORW_EN;
 	ocp_write_word(tp, MCU_TYPE_PLA, PLA_SFF_STS_7, ocp_data);
 
-	rtl_rx_vlan_en(tp, tp->netdev->features & NETIF_F_HW_VLAN_CTAG_RX);
+	rtl_rx_vlan_en(tp, netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_RX_BIT,
+						   tp->netdev->features));
 
 	rtl8156_change_mtu(tp);
 
@@ -9569,21 +9579,27 @@ static int rtl8152_probe(struct usb_interface *intf,
 	netdev->netdev_ops = &rtl8152_netdev_ops;
 	netdev->watchdog_timeo = RTL8152_TX_TIMEOUT;
 
-	netdev->features |= NETIF_F_RXCSUM | NETIF_F_IP_CSUM | NETIF_F_SG |
-			    NETIF_F_TSO | NETIF_F_FRAGLIST | NETIF_F_IPV6_CSUM |
-			    NETIF_F_TSO6 | NETIF_F_HW_VLAN_CTAG_RX |
-			    NETIF_F_HW_VLAN_CTAG_TX;
-	netdev->hw_features = NETIF_F_RXCSUM | NETIF_F_IP_CSUM | NETIF_F_SG |
-			      NETIF_F_TSO | NETIF_F_FRAGLIST |
-			      NETIF_F_IPV6_CSUM | NETIF_F_TSO6 |
-			      NETIF_F_HW_VLAN_CTAG_RX | NETIF_F_HW_VLAN_CTAG_TX;
-	netdev->vlan_features = NETIF_F_SG | NETIF_F_IP_CSUM | NETIF_F_TSO |
+	netdev_feature_set_bits(NETIF_F_RXCSUM | NETIF_F_IP_CSUM | NETIF_F_SG |
+				NETIF_F_TSO | NETIF_F_FRAGLIST |
+				NETIF_F_IPV6_CSUM | NETIF_F_TSO6 |
+				NETIF_F_HW_VLAN_CTAG_RX |
+				NETIF_F_HW_VLAN_CTAG_TX, &netdev->features);
+	netdev_feature_zero(&netdev->hw_features);
+	netdev_feature_set_bits(NETIF_F_RXCSUM | NETIF_F_IP_CSUM | NETIF_F_SG |
+				NETIF_F_TSO | NETIF_F_FRAGLIST |
+				NETIF_F_IPV6_CSUM | NETIF_F_TSO6 |
+				NETIF_F_HW_VLAN_CTAG_RX |
+				NETIF_F_HW_VLAN_CTAG_TX, &netdev->hw_features);
+	netdev_feature_zero(&netdev->vlan_features);
+	netdev_feature_set_bits(NETIF_F_SG | NETIF_F_IP_CSUM | NETIF_F_TSO |
 				NETIF_F_HIGHDMA | NETIF_F_FRAGLIST |
-				NETIF_F_IPV6_CSUM | NETIF_F_TSO6;
+				NETIF_F_IPV6_CSUM | NETIF_F_TSO6,
+				&netdev->vlan_features);
 
 	if (tp->version == RTL_VER_01) {
-		netdev->features &= ~NETIF_F_RXCSUM;
-		netdev->hw_features &= ~NETIF_F_RXCSUM;
+		netdev_feature_clear_bit(NETIF_F_RXCSUM_BIT, &netdev->features);
+		netdev_feature_clear_bit(NETIF_F_RXCSUM_BIT,
+					 &netdev->hw_features);
 	}
 
 	if (le16_to_cpu(udev->descriptor.idVendor) == VENDOR_ID_LENOVO) {
diff --git a/drivers/net/usb/smsc75xx.c b/drivers/net/usb/smsc75xx.c
index 76f7af161313..87ad30084d2b 100644
--- a/drivers/net/usb/smsc75xx.c
+++ b/drivers/net/usb/smsc75xx.c
@@ -942,7 +942,7 @@ static int smsc75xx_set_features(struct net_device *netdev,
 
 	spin_lock_irqsave(&pdata->rfe_ctl_lock, flags);
 
-	if (features & NETIF_F_RXCSUM)
+	if (netdev_feature_test_bit(NETIF_F_RXCSUM_BIT, features))
 		pdata->rfe_ctl |= RFE_CTL_TCPUDP_CKM | RFE_CTL_IP_CKM;
 	else
 		pdata->rfe_ctl &= ~(RFE_CTL_TCPUDP_CKM | RFE_CTL_IP_CKM);
@@ -1472,13 +1472,15 @@ static int smsc75xx_bind(struct usbnet *dev, struct usb_interface *intf)
 	INIT_WORK(&pdata->set_multicast, smsc75xx_deferred_multicast_write);
 
 	if (DEFAULT_TX_CSUM_ENABLE)
-		dev->net->features |= NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM;
+		netdev_feature_set_bits(NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM,
+					&dev->net->features);
 
 	if (DEFAULT_RX_CSUM_ENABLE)
-		dev->net->features |= NETIF_F_RXCSUM;
+		netdev_feature_set_bit(NETIF_F_RXCSUM_BIT, &dev->net->features);
 
-	dev->net->hw_features = NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM |
-				NETIF_F_RXCSUM;
+	netdev_feature_set_bits(NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM |
+				NETIF_F_RXCSUM,
+				&dev->net->hw_features);
 
 	ret = smsc75xx_wait_ready(dev, 0);
 	if (ret < 0) {
@@ -2167,7 +2169,7 @@ static int smsc75xx_resume(struct usb_interface *intf)
 static void smsc75xx_rx_csum_offload(struct usbnet *dev, struct sk_buff *skb,
 				     u32 rx_cmd_a, u32 rx_cmd_b)
 {
-	if (!(dev->net->features & NETIF_F_RXCSUM) ||
+	if (!netdev_feature_test_bit(NETIF_F_RXCSUM_BIT, dev->net->features) ||
 	    unlikely(rx_cmd_a & RX_CMD_A_LCSM)) {
 		skb->ip_summed = CHECKSUM_NONE;
 	} else {
diff --git a/drivers/net/usb/smsc95xx.c b/drivers/net/usb/smsc95xx.c
index 7d953974eb9b..1bedb67af06f 100644
--- a/drivers/net/usb/smsc95xx.c
+++ b/drivers/net/usb/smsc95xx.c
@@ -615,12 +615,12 @@ static int smsc95xx_set_features(struct net_device *netdev,
 	if (ret < 0)
 		return ret;
 
-	if (features & NETIF_F_IP_CSUM)
+	if (netdev_feature_test_bit(NETIF_F_IP_CSUM_BIT, features))
 		read_buf |= Tx_COE_EN_;
 	else
 		read_buf &= ~Tx_COE_EN_;
 
-	if (features & NETIF_F_RXCSUM)
+	if (netdev_feature_test_bit(NETIF_F_RXCSUM_BIT, features))
 		read_buf |= Rx_COE_EN_;
 	else
 		read_buf &= ~Rx_COE_EN_;
@@ -1080,11 +1080,14 @@ static int smsc95xx_bind(struct usbnet *dev, struct usb_interface *intf)
 	 * for ipv4 packets.
 	 */
 	if (DEFAULT_TX_CSUM_ENABLE)
-		dev->net->features |= NETIF_F_IP_CSUM;
+		netdev_feature_set_bit(NETIF_F_IP_CSUM_BIT,
+				       &dev->net->features);
 	if (DEFAULT_RX_CSUM_ENABLE)
-		dev->net->features |= NETIF_F_RXCSUM;
+		netdev_feature_set_bit(NETIF_F_RXCSUM_BIT, &dev->net->features);
 
-	dev->net->hw_features = NETIF_F_IP_CSUM | NETIF_F_RXCSUM;
+	netdev_feature_zero(&dev->net->hw_features);
+	netdev_feature_set_bits(NETIF_F_IP_CSUM | NETIF_F_RXCSUM,
+				&dev->net->hw_features);
 	set_bit(EVENT_NO_IP_ALIGN, &dev->flags);
 
 	smsc95xx_init_mac_address(dev);
@@ -1824,7 +1827,8 @@ static int smsc95xx_rx_fixup(struct usbnet *dev, struct sk_buff *skb)
 
 			/* last frame in this batch */
 			if (skb->len == size) {
-				if (dev->net->features & NETIF_F_RXCSUM)
+				if (netdev_feature_test_bit(NETIF_F_RXCSUM_BIT,
+							    dev->net->features))
 					smsc95xx_rx_csum_offload(skb);
 				skb_trim(skb, skb->len - 4); /* remove fcs */
 				skb->truesize = size + sizeof(struct sk_buff);
@@ -1842,7 +1846,8 @@ static int smsc95xx_rx_fixup(struct usbnet *dev, struct sk_buff *skb)
 			ax_skb->data = packet;
 			skb_set_tail_pointer(ax_skb, size);
 
-			if (dev->net->features & NETIF_F_RXCSUM)
+			if (netdev_feature_test_bit(NETIF_F_RXCSUM_BIT,
+						    dev->net->features))
 				smsc95xx_rx_csum_offload(ax_skb);
 			skb_trim(ax_skb, ax_skb->len - 4); /* remove fcs */
 			ax_skb->truesize = size + sizeof(struct sk_buff);
diff --git a/drivers/usb/gadget/function/f_phonet.c b/drivers/usb/gadget/function/f_phonet.c
index 0b468f5d55bc..1d63d28425e6 100644
--- a/drivers/usb/gadget/function/f_phonet.c
+++ b/drivers/usb/gadget/function/f_phonet.c
@@ -267,7 +267,7 @@ static const struct net_device_ops pn_netdev_ops = {
 
 static void pn_net_setup(struct net_device *dev)
 {
-	dev->features		= 0;
+	netdev_feature_zero(&dev->features);
 	dev->type		= ARPHRD_PHONET;
 	dev->flags		= IFF_POINTOPOINT | IFF_NOARP;
 	dev->mtu		= PHONET_DEV_MTU;
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 082/167] net: wireless: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (80 preceding siblings ...)
  2021-09-29 15:52 ` [RFCv2 net-next 081/167] net: usb: " Jian Shen
@ 2021-09-29 15:52 ` Jian Shen
  2021-09-29 15:52 ` [RFCv2 net-next 083/167] net: realtek: " Jian Shen
                   ` (88 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:52 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

For iwiwifi, it initializes the features in struct. So it's
unable to change the prototype from u64 to bitmap. Define it
as u64 array instead.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 drivers/net/wireless/ath/ath10k/mac.c           |  7 +++++--
 drivers/net/wireless/ath/ath11k/mac.c           |  4 +++-
 drivers/net/wireless/ath/ath6kl/main.c          | 15 ++++++++++-----
 drivers/net/wireless/ath/ath6kl/txrx.c          |  6 ++++--
 drivers/net/wireless/ath/wil6210/netdev.c       | 11 +++++++----
 .../wireless/broadcom/brcm80211/brcmfmac/core.c |  4 ++--
 drivers/net/wireless/intel/iwlwifi/cfg/22000.c  |  4 ++--
 drivers/net/wireless/intel/iwlwifi/cfg/8000.c   |  2 +-
 drivers/net/wireless/intel/iwlwifi/cfg/9000.c   |  2 +-
 .../net/wireless/intel/iwlwifi/dvm/mac80211.c   |  7 +++++--
 drivers/net/wireless/intel/iwlwifi/iwl-config.h |  2 +-
 .../net/wireless/intel/iwlwifi/mvm/mac80211.c   | 17 +++++++++++------
 drivers/net/wireless/intel/iwlwifi/mvm/rx.c     |  2 +-
 drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c   |  2 +-
 drivers/net/wireless/intel/iwlwifi/mvm/tx.c     | 10 +++++++---
 .../net/wireless/mediatek/mt76/mt7615/init.c    |  3 ++-
 .../net/wireless/mediatek/mt76/mt7915/init.c    |  3 ++-
 .../net/wireless/mediatek/mt76/mt7921/init.c    |  3 ++-
 net/wireless/core.c                             | 15 ++++++++++-----
 19 files changed, 77 insertions(+), 42 deletions(-)

diff --git a/drivers/net/wireless/ath/ath10k/mac.c b/drivers/net/wireless/ath/ath10k/mac.c
index c272b290fa73..493d15f4d4e9 100644
--- a/drivers/net/wireless/ath/ath10k/mac.c
+++ b/drivers/net/wireless/ath/ath10k/mac.c
@@ -10119,8 +10119,11 @@ int ath10k_mac_register(struct ath10k *ar)
 	if (ar->hw_params.dynamic_sar_support)
 		ar->hw->wiphy->sar_capa = &ath10k_sar_capa;
 
-	if (!test_bit(ATH10K_FLAG_RAW_MODE, &ar->dev_flags))
-		ar->hw->netdev_features = NETIF_F_HW_CSUM;
+	if (!test_bit(ATH10K_FLAG_RAW_MODE, &ar->dev_flags)) {
+		netdev_feature_zero(&ar->hw->netdev_features);
+		netdev_feature_set_bit(NETIF_F_HW_CSUM_BIT,
+				       &ar->hw->netdev_features);
+	}
 
 	if (IS_ENABLED(CONFIG_ATH10K_DFS_CERTIFIED)) {
 		/* Init ath dfs pattern detector */
diff --git a/drivers/net/wireless/ath/ath11k/mac.c b/drivers/net/wireless/ath/ath11k/mac.c
index e9b3689331ec..ba892c4ab3eb 100644
--- a/drivers/net/wireless/ath/ath11k/mac.c
+++ b/drivers/net/wireless/ath/ath11k/mac.c
@@ -6570,7 +6570,9 @@ static int __ath11k_mac_register(struct ath11k *ar)
 	ath11k_reg_init(ar);
 
 	if (!test_bit(ATH11K_FLAG_RAW_MODE, &ab->dev_flags)) {
-		ar->hw->netdev_features = NETIF_F_HW_CSUM;
+		netdev_feature_zero(&ar->hw->netdev_features);
+		netdev_feature_set_bit(NETIF_F_HW_CSUM_BIT,
+				       &ar->hw->netdev_features);
 		ieee80211_hw_set(ar->hw, SW_CRYPTO_CONTROL);
 		ieee80211_hw_set(ar->hw, SUPPORT_FAST_XMIT);
 	}
diff --git a/drivers/net/wireless/ath/ath6kl/main.c b/drivers/net/wireless/ath/ath6kl/main.c
index d3aa9e7a37c2..cbf0e5dcac02 100644
--- a/drivers/net/wireless/ath/ath6kl/main.c
+++ b/drivers/net/wireless/ath/ath6kl/main.c
@@ -1126,24 +1126,28 @@ static int ath6kl_set_features(struct net_device *dev,
 	struct ath6kl *ar = vif->ar;
 	int err = 0;
 
-	if ((features & NETIF_F_RXCSUM) &&
+	if (netdev_feature_test_bit(NETIF_F_RXCSUM_BIT, features) &&
 	    (ar->rx_meta_ver != WMI_META_VERSION_2)) {
 		ar->rx_meta_ver = WMI_META_VERSION_2;
 		err = ath6kl_wmi_set_rx_frame_format_cmd(ar->wmi,
 							 vif->fw_vif_idx,
 							 ar->rx_meta_ver, 0, 0);
 		if (err) {
-			dev->features = features & ~NETIF_F_RXCSUM;
+			netdev_feature_copy(&dev->features, features);
+			netdev_feature_clear_bit(NETIF_F_RXCSUM_BIT,
+						 &dev->features);
 			return err;
 		}
-	} else if (!(features & NETIF_F_RXCSUM) &&
+	} else if (!netdev_feature_test_bit(NETIF_F_RXCSUM_BIT, features) &&
 		   (ar->rx_meta_ver == WMI_META_VERSION_2)) {
 		ar->rx_meta_ver = 0;
 		err = ath6kl_wmi_set_rx_frame_format_cmd(ar->wmi,
 							 vif->fw_vif_idx,
 							 ar->rx_meta_ver, 0, 0);
 		if (err) {
-			dev->features = features | NETIF_F_RXCSUM;
+			netdev_feature_copy(&dev->features, features);
+			netdev_feature_set_bit(NETIF_F_RXCSUM_BIT,
+					       &dev->features);
 			return err;
 		}
 	}
@@ -1305,7 +1309,8 @@ void init_netdev(struct net_device *dev)
 
 	if (!test_bit(ATH6KL_FW_CAPABILITY_NO_IP_CHECKSUM,
 		      ar->fw_capabilities))
-		dev->hw_features |= NETIF_F_IP_CSUM | NETIF_F_RXCSUM;
+		netdev_feature_set_bits(NETIF_F_IP_CSUM | NETIF_F_RXCSUM,
+					&dev->hw_features);
 
 	return;
 }
diff --git a/drivers/net/wireless/ath/ath6kl/txrx.c b/drivers/net/wireless/ath/ath6kl/txrx.c
index b22ed499f7ba..6c7569c20d00 100644
--- a/drivers/net/wireless/ath/ath6kl/txrx.c
+++ b/drivers/net/wireless/ath/ath6kl/txrx.c
@@ -391,7 +391,8 @@ netdev_tx_t ath6kl_data_tx(struct sk_buff *skb, struct net_device *dev)
 	}
 
 	if (test_bit(WMI_ENABLED, &ar->flag)) {
-		if ((dev->features & NETIF_F_IP_CSUM) &&
+		if (netdev_feature_test_bit(NETIF_F_IP_CSUM_BIT,
+					    dev->features) &&
 		    (csum == CHECKSUM_PARTIAL)) {
 			csum_start = skb->csum_start -
 					(skb_network_header(skb) - skb->head) +
@@ -410,7 +411,8 @@ netdev_tx_t ath6kl_data_tx(struct sk_buff *skb, struct net_device *dev)
 			goto fail_tx;
 		}
 
-		if ((dev->features & NETIF_F_IP_CSUM) &&
+		if (netdev_feature_test_bit(NETIF_F_IP_CSUM_BIT,
+					    dev->features) &&
 		    (csum == CHECKSUM_PARTIAL)) {
 			meta_v2.csum_start = csum_start;
 			meta_v2.csum_dest = csum_dest;
diff --git a/drivers/net/wireless/ath/wil6210/netdev.c b/drivers/net/wireless/ath/wil6210/netdev.c
index 0913f0bf60e7..b29a4d6c3c0f 100644
--- a/drivers/net/wireless/ath/wil6210/netdev.c
+++ b/drivers/net/wireless/ath/wil6210/netdev.c
@@ -335,11 +335,14 @@ wil_vif_alloc(struct wil6210_priv *wil, const char *name,
 	ndev->netdev_ops = &wil_netdev_ops;
 	wil_set_ethtoolops(ndev);
 	ndev->ieee80211_ptr = wdev;
-	ndev->hw_features = NETIF_F_HW_CSUM | NETIF_F_RXCSUM |
-			    NETIF_F_SG | NETIF_F_GRO |
-			    NETIF_F_TSO | NETIF_F_TSO6;
+	netdev_feature_zero(&ndev->hw_features);
+	netdev_feature_set_bits(NETIF_F_HW_CSUM | NETIF_F_RXCSUM |
+				NETIF_F_SG | NETIF_F_GRO |
+				NETIF_F_TSO | NETIF_F_TSO6,
+				&ndev->hw_features);
+
+	netdev_feature_or(&ndev->features, ndev->features, ndev->hw_features);
 
-	ndev->features |= ndev->hw_features;
 	SET_NETDEV_DEV(ndev, wiphy_dev(wdev->wiphy));
 	wdev->netdev = ndev;
 	return vif;
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.c
index db5f8535fdb5..d04bff1a60f4 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.c
@@ -616,9 +616,9 @@ static int brcmf_netdev_open(struct net_device *ndev)
 	/* Get current TOE mode from dongle */
 	if (brcmf_fil_iovar_int_get(ifp, "toe_ol", &toe_ol) >= 0
 	    && (toe_ol & TOE_TX_CSUM_OL) != 0)
-		ndev->features |= NETIF_F_IP_CSUM;
+		netdev_feature_set_bit(NETIF_F_IP_CSUM_BIT, &ndev->features);
 	else
-		ndev->features &= ~NETIF_F_IP_CSUM;
+		netdev_feature_clear_bit(NETIF_F_IP_CSUM_BIT, &ndev->features);
 
 	if (brcmf_cfg80211_up(ndev)) {
 		bphy_err(drvr, "failed to bring up cfg80211\n");
diff --git a/drivers/net/wireless/intel/iwlwifi/cfg/22000.c b/drivers/net/wireless/intel/iwlwifi/cfg/22000.c
index d8231cc821ae..fd2408f80676 100644
--- a/drivers/net/wireless/intel/iwlwifi/cfg/22000.c
+++ b/drivers/net/wireless/intel/iwlwifi/cfg/22000.c
@@ -150,7 +150,7 @@ static const struct iwl_ht_params iwl_22000_ht_params = {
 	.dccm2_len = IWL_22000_DCCM2_LEN,				\
 	.smem_offset = IWL_22000_SMEM_OFFSET,				\
 	.smem_len = IWL_22000_SMEM_LEN,					\
-	.features = IWL_TX_CSUM_NETIF_FLAGS | NETIF_F_RXCSUM,		\
+	.features = { IWL_TX_CSUM_NETIF_FLAGS | NETIF_F_RXCSUM },		\
 	.apmg_not_supported = true,					\
 	.trans.mq_rx_supported = true,					\
 	.vht_mu_mimo_supported = true,					\
@@ -227,7 +227,7 @@ static const struct iwl_ht_params iwl_22000_ht_params = {
 	.dccm2_len = IWL_22000_DCCM2_LEN,				\
 	.smem_offset = IWL_22000_SMEM_OFFSET,				\
 	.smem_len = IWL_22000_SMEM_LEN,					\
-	.features = IWL_TX_CSUM_NETIF_FLAGS | NETIF_F_RXCSUM,		\
+	.features = { IWL_TX_CSUM_NETIF_FLAGS | NETIF_F_RXCSUM },	\
 	.apmg_not_supported = true,					\
 	.trans.mq_rx_supported = true,					\
 	.vht_mu_mimo_supported = true,					\
diff --git a/drivers/net/wireless/intel/iwlwifi/cfg/8000.c b/drivers/net/wireless/intel/iwlwifi/cfg/8000.c
index a6454287d506..d4e33cd76702 100644
--- a/drivers/net/wireless/intel/iwlwifi/cfg/8000.c
+++ b/drivers/net/wireless/intel/iwlwifi/cfg/8000.c
@@ -81,7 +81,7 @@ static const struct iwl_tt_params iwl8000_tt_params = {
 	.trans.base_params = &iwl8000_base_params,			\
 	.led_mode = IWL_LED_RF_STATE,					\
 	.nvm_hw_section_num = 10,					\
-	.features = NETIF_F_RXCSUM,					\
+	.features = { NETIF_F_RXCSUM },					\
 	.non_shared_ant = ANT_A,					\
 	.dccm_offset = IWL8260_DCCM_OFFSET,				\
 	.dccm_len = IWL8260_DCCM_LEN,					\
diff --git a/drivers/net/wireless/intel/iwlwifi/cfg/9000.c b/drivers/net/wireless/intel/iwlwifi/cfg/9000.c
index 7a7ca06d46c1..36d0049cb900 100644
--- a/drivers/net/wireless/intel/iwlwifi/cfg/9000.c
+++ b/drivers/net/wireless/intel/iwlwifi/cfg/9000.c
@@ -84,7 +84,7 @@ static const struct iwl_tt_params iwl9000_tt_params = {
 	.dccm2_len = IWL9000_DCCM2_LEN,					\
 	.smem_offset = IWL9000_SMEM_OFFSET,				\
 	.smem_len = IWL9000_SMEM_LEN,					\
-	.features = IWL_TX_CSUM_NETIF_FLAGS | NETIF_F_RXCSUM,		\
+	.features = { IWL_TX_CSUM_NETIF_FLAGS | NETIF_F_RXCSUM },	\
 	.thermal_params = &iwl9000_tt_params,				\
 	.apmg_not_supported = true,					\
 	.num_rbds = 512,						\
diff --git a/drivers/net/wireless/intel/iwlwifi/dvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/dvm/mac80211.c
index 75e7665773c5..81df8565a22f 100644
--- a/drivers/net/wireless/intel/iwlwifi/dvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/dvm/mac80211.c
@@ -101,8 +101,11 @@ int iwlagn_mac_setup_register(struct iwl_priv *priv,
 	ieee80211_hw_set(hw, SUPPORT_FAST_XMIT);
 	ieee80211_hw_set(hw, WANT_MONITOR_VIF);
 
-	if (priv->trans->max_skb_frags)
-		hw->netdev_features = NETIF_F_HIGHDMA | NETIF_F_SG;
+	if (priv->trans->max_skb_frags) {
+		netdev_feature_zero(&hw->netdev_features);
+		netdev_feature_set_bits(NETIF_F_HIGHDMA | NETIF_F_SG,
+					&hw->netdev_features);
+	}
 
 	hw->offchannel_tx_hw_queue = IWL_AUX_QUEUE;
 	hw->radiotap_mcs_details |= IEEE80211_RADIOTAP_MCS_HAVE_FMT;
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-config.h b/drivers/net/wireless/intel/iwlwifi/iwl-config.h
index 7eb534df5331..01acacf0219e 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-config.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-config.h
@@ -366,7 +366,7 @@ struct iwl_cfg {
 	enum iwl_nvm_type nvm_type;
 	u32 max_data_size;
 	u32 max_inst_size;
-	netdev_features_t features;
+	u64 features[NETDEV_FEATURE_DWORDS];
 	u32 dccm_offset;
 	u32 dccm_len;
 	u32 dccm2_offset;
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index 3a4585222d6d..fdfda9cc55f3 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -387,8 +387,11 @@ int iwl_mvm_mac_setup_register(struct iwl_mvm *mvm)
 	if (mvm->trans->num_rx_queues > 1)
 		ieee80211_hw_set(hw, USES_RSS);
 
-	if (mvm->trans->max_skb_frags)
-		hw->netdev_features = NETIF_F_HIGHDMA | NETIF_F_SG;
+	if (mvm->trans->max_skb_frags) {
+		netdev_feature_zero(&hw->netdev_features);
+		netdev_feature_set_bits(NETIF_F_HIGHDMA | NETIF_F_SG,
+					&hw->netdev_features);
+	}
 
 	hw->queues = IEEE80211_NUM_ACS;
 	hw->offchannel_tx_hw_queue = IWL_MVM_OFFCHANNEL_QUEUE;
@@ -703,10 +706,11 @@ int iwl_mvm_mac_setup_register(struct iwl_mvm *mvm)
 		hw->wiphy->features |= NL80211_FEATURE_TDLS_CHANNEL_SWITCH;
 	}
 
-	hw->netdev_features |= mvm->cfg->features;
+	netdev_feature_set_bits(mvm->cfg->features[0], &hw->netdev_features);
 	if (!iwl_mvm_is_csum_supported(mvm))
-		hw->netdev_features &= ~(IWL_TX_CSUM_NETIF_FLAGS |
-					 NETIF_F_RXCSUM);
+		netdev_feature_clear_bits(IWL_TX_CSUM_NETIF_FLAGS |
+					  NETIF_F_RXCSUM,
+					  &hw->netdev_features);
 
 	if (mvm->cfg->vht_mu_mimo_supported)
 		wiphy_ext_feature_set(hw->wiphy,
@@ -1451,7 +1455,8 @@ static int iwl_mvm_mac_add_interface(struct ieee80211_hw *hw,
 		goto out_unlock;
 	}
 
-	mvmvif->features |= hw->netdev_features;
+	netdev_feature_or(&mvmvif->features, mvmvif->features,
+			  hw->netdev_features);
 
 	ret = iwl_mvm_mac_ctxt_add(mvm, vif);
 	if (ret)
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rx.c b/drivers/net/wireless/intel/iwlwifi/mvm/rx.c
index 8ef5399ad9be..21d97eb6c207 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rx.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rx.c
@@ -276,7 +276,7 @@ static void iwl_mvm_rx_csum(struct ieee80211_sta *sta,
 	struct iwl_mvm_sta *mvmsta = iwl_mvm_sta_from_mac80211(sta);
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(mvmsta->vif);
 
-	if (mvmvif->features & NETIF_F_RXCSUM &&
+	if (netdev_feature_test_bit(NETIF_F_RXCSUM_BIT, mvmvif->features) &&
 	    status & RX_MPDU_RES_STATUS_CSUM_DONE &&
 	    status & RX_MPDU_RES_STATUS_CSUM_OK)
 		skb->ip_summed = CHECKSUM_UNNECESSARY;
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
index c12f303cf652..a0daf93c36c1 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
@@ -472,7 +472,7 @@ static void iwl_mvm_rx_csum(struct iwl_mvm *mvm,
 
 		mvmvif = iwl_mvm_vif_from_mac80211(mvmsta->vif);
 
-		if (mvmvif->features & NETIF_F_RXCSUM &&
+		if (netdev_feature_test_bit(NETIF_F_RXCSUM_BIT, mvmvif->features) &&
 		    flags & IWL_RX_L3L4_TCP_UDP_CSUM_OK &&
 		    (flags & IWL_RX_L3L4_IP_HDR_CSUM_OK ||
 		     l3_prot == IWL_RX_L3_TYPE_IPV6 ||
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
index 0a13c2bda2ee..cfe0fcb1a0ba 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
@@ -53,7 +53,8 @@ static u16 iwl_mvm_tx_csum(struct iwl_mvm *mvm, struct sk_buff *skb,
 		goto out;
 
 	/* We do not expect to be requested to csum stuff we do not support */
-	if (WARN_ONCE(!(mvm->hw->netdev_features & IWL_TX_CSUM_NETIF_FLAGS) ||
+	if (WARN_ONCE(!netdev_feature_test_bits(IWL_TX_CSUM_NETIF_FLAGS,
+						mvm->hw->netdev_features) ||
 		      (skb->protocol != htons(ETH_P_IP) &&
 		       skb->protocol != htons(ETH_P_IPV6)),
 		      "No support for requested checksum\n")) {
@@ -836,9 +837,12 @@ static int iwl_mvm_tx_tso(struct iwl_mvm *mvm, struct sk_buff *skb,
 	unsigned int mss = skb_shinfo(skb)->gso_size;
 	unsigned int num_subframes, tcp_payload_len, subf_len, max_amsdu_len;
 	u16 snap_ip_tcp, pad;
-	netdev_features_t netdev_flags = NETIF_F_CSUM_MASK | NETIF_F_SG;
+	netdev_features_t netdev_flags;
 	u8 tid;
 
+	netdev_feature_zero(&netdev_flags);
+	netdev_feature_set_bits(NETIF_F_CSUM_MASK | NETIF_F_SG, &netdev_flags);
+
 	snap_ip_tcp = 8 + skb_transport_header(skb) - skb_network_header(skb) +
 		tcp_hdrlen(skb);
 
@@ -854,7 +858,7 @@ static int iwl_mvm_tx_tso(struct iwl_mvm *mvm, struct sk_buff *skb,
 	if (skb->protocol == htons(ETH_P_IPV6) &&
 	    ((struct ipv6hdr *)skb_network_header(skb))->nexthdr !=
 	    IPPROTO_TCP) {
-		netdev_flags &= ~NETIF_F_CSUM_MASK;
+		netdev_feature_clear_bits(NETIF_F_CSUM_MASK, &netdev_flags);
 		return iwl_mvm_tx_tso_segment(skb, 1, netdev_flags, mpdus_skb);
 	}
 
diff --git a/drivers/net/wireless/mediatek/mt76/mt7615/init.c b/drivers/net/wireless/mediatek/mt76/mt7615/init.c
index 2f1ac644e018..e690e095ba60 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7615/init.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7615/init.c
@@ -363,7 +363,8 @@ mt7615_init_wiphy(struct ieee80211_hw *hw)
 	hw->max_rates = 3;
 	hw->max_report_rates = 7;
 	hw->max_rate_tries = 11;
-	hw->netdev_features = NETIF_F_RXCSUM;
+	netdev_feature_zero(&hw->netdev_features);
+	netdev_feature_set_bit(NETIF_F_RXCSUM_BIT, &hw->netdev_features);
 
 	hw->radiotap_timestamp.units_pos =
 		IEEE80211_RADIOTAP_TIMESTAMP_UNIT_US;
diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/init.c b/drivers/net/wireless/mediatek/mt76/mt7915/init.c
index 4798d6344305..a3ab1a5f3f94 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7915/init.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7915/init.c
@@ -217,7 +217,8 @@ mt7915_init_wiphy(struct ieee80211_hw *hw)
 	hw->queues = 4;
 	hw->max_rx_aggregation_subframes = IEEE80211_MAX_AMPDU_BUF;
 	hw->max_tx_aggregation_subframes = IEEE80211_MAX_AMPDU_BUF;
-	hw->netdev_features = NETIF_F_RXCSUM;
+	netdev_feature_zero(&hw->netdev_features);
+	netdev_feature_set_bit(NETIF_F_RXCSUM_BIT, &hw->netdev_features);
 
 	hw->radiotap_timestamp.units_pos =
 		IEEE80211_RADIOTAP_TIMESTAMP_UNIT_US;
diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/init.c b/drivers/net/wireless/mediatek/mt76/mt7921/init.c
index a9ce10b98827..368d6cfb2881 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/init.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/init.c
@@ -51,7 +51,8 @@ mt7921_init_wiphy(struct ieee80211_hw *hw)
 	hw->queues = 4;
 	hw->max_rx_aggregation_subframes = 64;
 	hw->max_tx_aggregation_subframes = 128;
-	hw->netdev_features = NETIF_F_RXCSUM;
+	netdev_feature_zero(&hw->netdev_features);
+	netdev_feature_set_bit(NETIF_F_RXCSUM_BIT, &hw->netdev_features);
 
 	hw->radiotap_timestamp.units_pos =
 		IEEE80211_RADIOTAP_TIMESTAMP_UNIT_US;
diff --git a/net/wireless/core.c b/net/wireless/core.c
index 03323121ca50..5849ebf7ca5f 100644
--- a/net/wireless/core.c
+++ b/net/wireless/core.c
@@ -164,11 +164,13 @@ int cfg80211_switch_netns(struct cfg80211_registered_device *rdev,
 	list_for_each_entry(wdev, &rdev->wiphy.wdev_list, list) {
 		if (!wdev->netdev)
 			continue;
-		wdev->netdev->features &= ~NETIF_F_NETNS_LOCAL;
+		netdev_feature_clear_bit(NETIF_F_NETNS_LOCAL_BIT,
+					 &wdev->netdev->features);
 		err = dev_change_net_namespace(wdev->netdev, net, "wlan%d");
 		if (err)
 			break;
-		wdev->netdev->features |= NETIF_F_NETNS_LOCAL;
+		netdev_feature_set_bit(NETIF_F_NETNS_LOCAL_BIT,
+				       &wdev->netdev->features);
 	}
 
 	if (err) {
@@ -180,11 +182,13 @@ int cfg80211_switch_netns(struct cfg80211_registered_device *rdev,
 						     list) {
 			if (!wdev->netdev)
 				continue;
-			wdev->netdev->features &= ~NETIF_F_NETNS_LOCAL;
+			netdev_feature_clear_bit(NETIF_F_NETNS_LOCAL_BIT,
+						 &wdev->netdev->features);
 			err = dev_change_net_namespace(wdev->netdev, net,
 							"wlan%d");
 			WARN_ON(err);
-			wdev->netdev->features |= NETIF_F_NETNS_LOCAL;
+			netdev_feature_set_bit(NETIF_F_NETNS_LOCAL_BIT,
+					       &wdev->netdev->features);
 		}
 
 		return err;
@@ -1384,7 +1388,8 @@ static int cfg80211_netdev_notifier_call(struct notifier_block *nb,
 		SET_NETDEV_DEVTYPE(dev, &wiphy_type);
 		wdev->netdev = dev;
 		/* can only change netns with wiphy */
-		dev->features |= NETIF_F_NETNS_LOCAL;
+		netdev_feature_set_bit(NETIF_F_NETNS_LOCAL_BIT,
+				       &dev->features);
 
 		cfg80211_init_wdev(wdev);
 		break;
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 083/167] net: realtek: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (81 preceding siblings ...)
  2021-09-29 15:52 ` [RFCv2 net-next 082/167] net: wireless: " Jian Shen
@ 2021-09-29 15:52 ` Jian Shen
  2021-09-29 15:52 ` [RFCv2 net-next 084/167] net: broadcom: " Jian Shen
                   ` (87 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:52 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 drivers/net/ethernet/realtek/8139cp.c     | 32 ++++++-----
 drivers/net/ethernet/realtek/8139too.c    | 28 ++++++----
 drivers/net/ethernet/realtek/r8169_main.c | 65 ++++++++++++++---------
 3 files changed, 74 insertions(+), 51 deletions(-)

diff --git a/drivers/net/ethernet/realtek/8139cp.c b/drivers/net/ethernet/realtek/8139cp.c
index 663a12598cad..5ef5b41c7aab 100644
--- a/drivers/net/ethernet/realtek/8139cp.c
+++ b/drivers/net/ethernet/realtek/8139cp.c
@@ -1461,17 +1461,18 @@ static int cp_set_features(struct net_device *dev, netdev_features_t features)
 	struct cp_private *cp = netdev_priv(dev);
 	unsigned long flags;
 
-	if (!((dev->features ^ features) & NETIF_F_RXCSUM))
+	if (netdev_feature_test_bit(NETIF_F_RXCSUM_BIT, dev->features) ==
+	    netdev_feature_test_bit(NETIF_F_RXCSUM_BIT, features))
 		return 0;
 
 	spin_lock_irqsave(&cp->lock, flags);
 
-	if (features & NETIF_F_RXCSUM)
+	if (netdev_feature_test_bit(NETIF_F_RXCSUM_BIT, features))
 		cp->cpcmd |= RxChkSum;
 	else
 		cp->cpcmd &= ~RxChkSum;
 
-	if (features & NETIF_F_HW_VLAN_CTAG_RX)
+	if (netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_RX_BIT, features))
 		cp->cpcmd |= RxVlanOn;
 	else
 		cp->cpcmd &= ~RxVlanOn;
@@ -1857,7 +1858,7 @@ static void cp_features_check(struct sk_buff *skb, struct net_device *dev,
 			      netdev_features_t *features)
 {
 	if (skb_shinfo(skb)->gso_size > MSSMask)
-		*features &= ~NETIF_F_TSO;
+		netdev_feature_clear_bit(NETIF_F_TSO_BIT, features);
 
 	vlan_features_check(skb, features);
 }
@@ -1960,8 +1961,8 @@ static int cp_init_one (struct pci_dev *pdev, const struct pci_device_id *ent)
 	cp->cpcmd = (pci_using_dac ? PCIDAC : 0) |
 		    PCIMulRW | RxChkSum | CpRxOn | CpTxOn;
 
-	dev->features |= NETIF_F_RXCSUM;
-	dev->hw_features |= NETIF_F_RXCSUM;
+	netdev_feature_set_bit(NETIF_F_RXCSUM_BIT, &dev->features);
+	netdev_feature_set_bit(NETIF_F_RXCSUM_BIT, &dev->hw_features);
 
 	regs = ioremap(pciaddr, CP_REGS_SIZE);
 	if (!regs) {
@@ -1986,16 +1987,19 @@ static int cp_init_one (struct pci_dev *pdev, const struct pci_device_id *ent)
 	dev->ethtool_ops = &cp_ethtool_ops;
 	dev->watchdog_timeo = TX_TIMEOUT;
 
-	dev->features |= NETIF_F_SG | NETIF_F_IP_CSUM | NETIF_F_TSO |
-		NETIF_F_HW_VLAN_CTAG_TX | NETIF_F_HW_VLAN_CTAG_RX;
+	netdev_feature_set_bits(NETIF_F_SG | NETIF_F_IP_CSUM | NETIF_F_TSO |
+				NETIF_F_HW_VLAN_CTAG_TX |
+				NETIF_F_HW_VLAN_CTAG_RX, &dev->features);
 
 	if (pci_using_dac)
-		dev->features |= NETIF_F_HIGHDMA;
-
-	dev->hw_features |= NETIF_F_SG | NETIF_F_IP_CSUM | NETIF_F_TSO |
-		NETIF_F_HW_VLAN_CTAG_TX | NETIF_F_HW_VLAN_CTAG_RX;
-	dev->vlan_features = NETIF_F_SG | NETIF_F_IP_CSUM | NETIF_F_TSO |
-		NETIF_F_HIGHDMA;
+		netdev_feature_set_bit(NETIF_F_HIGHDMA_BIT, &dev->features);
+
+	netdev_feature_set_bits(NETIF_F_SG | NETIF_F_IP_CSUM | NETIF_F_TSO |
+				NETIF_F_HW_VLAN_CTAG_TX |
+				NETIF_F_HW_VLAN_CTAG_RX, &dev->hw_features);
+	netdev_feature_zero(&dev->vlan_features);
+	netdev_feature_set_bits(NETIF_F_SG | NETIF_F_IP_CSUM | NETIF_F_TSO |
+				NETIF_F_HIGHDMA, &dev->vlan_features);
 
 	/* MTU range: 60 - 4096 */
 	dev->min_mtu = CP_MIN_MTU;
diff --git a/drivers/net/ethernet/realtek/8139too.c b/drivers/net/ethernet/realtek/8139too.c
index 2e6923cc653e..59b5a95bcaa4 100644
--- a/drivers/net/ethernet/realtek/8139too.c
+++ b/drivers/net/ethernet/realtek/8139too.c
@@ -900,18 +900,21 @@ static struct net_device *rtl8139_init_board(struct pci_dev *pdev)
 static int rtl8139_set_features(struct net_device *dev, netdev_features_t features)
 {
 	struct rtl8139_private *tp = netdev_priv(dev);
+	netdev_features_t changed;
 	unsigned long flags;
-	netdev_features_t changed = features ^ dev->features;
 	void __iomem *ioaddr = tp->mmio_addr;
 
-	if (!(changed & (NETIF_F_RXALL)))
+	netdev_feature_xor(&changed, features, dev->features);
+
+	if (!netdev_feature_test_bit(NETIF_F_RXALL_BIT, changed))
 		return 0;
 
 	spin_lock_irqsave(&tp->lock, flags);
 
-	if (changed & NETIF_F_RXALL) {
+	if (netdev_feature_test_bit(NETIF_F_RXALL_BIT, changed)) {
 		int rx_mode = tp->rx_config;
-		if (features & NETIF_F_RXALL)
+
+		if (netdev_feature_test_bit(NETIF_F_RXALL_BIT, features))
 			rx_mode |= (AcceptErr | AcceptRunt);
 		else
 			rx_mode &= ~(AcceptErr | AcceptRunt);
@@ -1007,11 +1010,12 @@ static int rtl8139_init_one(struct pci_dev *pdev,
 	 * through the use of skb_copy_and_csum_dev we enable these
 	 * features
 	 */
-	dev->features |= NETIF_F_SG | NETIF_F_HW_CSUM | NETIF_F_HIGHDMA;
-	dev->vlan_features = dev->features;
+	netdev_feature_set_bits(NETIF_F_SG | NETIF_F_HW_CSUM | NETIF_F_HIGHDMA,
+				&dev->features);
+	netdev_feature_copy(&dev->vlan_features, dev->features);
 
-	dev->hw_features |= NETIF_F_RXALL;
-	dev->hw_features |= NETIF_F_RXFCS;
+	netdev_feature_set_bit(NETIF_F_RXALL_BIT, &dev->hw_features);
+	netdev_feature_set_bit(NETIF_F_RXFCS_BIT, &dev->hw_features);
 
 	/* MTU range: 68 - 1770 */
 	dev->min_mtu = ETH_MIN_MTU;
@@ -1968,7 +1972,8 @@ static int rtl8139_rx(struct net_device *dev, struct rtl8139_private *tp,
 		/* read size+status of next frame from DMA ring buffer */
 		rx_status = le32_to_cpu (*(__le32 *) (rx_ring + ring_offset));
 		rx_size = rx_status >> 16;
-		if (likely(!(dev->features & NETIF_F_RXFCS)))
+		if (likely(!netdev_feature_test_bit(NETIF_F_RXFCS_BIT,
+						    dev->features)))
 			pkt_size = rx_size - 4;
 		else
 			pkt_size = rx_size;
@@ -2009,7 +2014,8 @@ static int rtl8139_rx(struct net_device *dev, struct rtl8139_private *tp,
 		if (unlikely((rx_size > (MAX_ETH_FRAME_SIZE+4)) ||
 			     (rx_size < 8) ||
 			     (!(rx_status & RxStatusOK)))) {
-			if ((dev->features & NETIF_F_RXALL) &&
+			if (netdev_feature_test_bit(NETIF_F_RXALL_BIT,
+						    dev->features) &&
 			    (rx_size <= (MAX_ETH_FRAME_SIZE + 4)) &&
 			    (rx_size >= 8) &&
 			    (!(rx_status & RxStatusOK))) {
@@ -2580,7 +2586,7 @@ static void __set_rx_mode (struct net_device *dev)
 		}
 	}
 
-	if (dev->features & NETIF_F_RXALL)
+	if (netdev_feature_test_bit(NETIF_F_RXALL_BIT, dev->features))
 		rx_mode |= (AcceptErr | AcceptRunt);
 
 	/* We can safely update without stopping the chip. */
diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c
index c876246cf8c1..3fdc1da3eb52 100644
--- a/drivers/net/ethernet/realtek/r8169_main.c
+++ b/drivers/net/ethernet/realtek/r8169_main.c
@@ -1437,11 +1437,12 @@ static void rtl8169_fix_features(struct net_device *dev,
 	struct rtl8169_private *tp = netdev_priv(dev);
 
 	if (dev->mtu > TD_MSS_MAX)
-		*features &= ~NETIF_F_ALL_TSO;
+		netdev_feature_clear_bits(NETIF_F_ALL_TSO, features);
 
 	if (dev->mtu > ETH_DATA_LEN &&
 	    tp->mac_version > RTL_GIGA_MAC_VER_06)
-		*features &= ~(NETIF_F_CSUM_MASK | NETIF_F_ALL_TSO);
+		netdev_feature_clear_bits(NETIF_F_CSUM_MASK | NETIF_F_ALL_TSO,
+					  features);
 }
 
 static void rtl_set_rx_config_features(struct rtl8169_private *tp,
@@ -1449,13 +1450,14 @@ static void rtl_set_rx_config_features(struct rtl8169_private *tp,
 {
 	u32 rx_config = RTL_R32(tp, RxConfig);
 
-	if (features & NETIF_F_RXALL)
+	if (netdev_feature_test_bit(NETIF_F_RXALL_BIT, features))
 		rx_config |= RX_CONFIG_ACCEPT_ERR_MASK;
 	else
 		rx_config &= ~RX_CONFIG_ACCEPT_ERR_MASK;
 
 	if (rtl_is_8125(tp)) {
-		if (features & NETIF_F_HW_VLAN_CTAG_RX)
+		if (netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_RX_BIT,
+					    features))
 			rx_config |= RX_VLAN_8125;
 		else
 			rx_config &= ~RX_VLAN_8125;
@@ -1471,13 +1473,14 @@ static int rtl8169_set_features(struct net_device *dev,
 
 	rtl_set_rx_config_features(tp, features);
 
-	if (features & NETIF_F_RXCSUM)
+	if (netdev_feature_test_bit(NETIF_F_RXCSUM_BIT, features))
 		tp->cp_cmd |= RxChkSum;
 	else
 		tp->cp_cmd &= ~RxChkSum;
 
 	if (!rtl_is_8125(tp)) {
-		if (features & NETIF_F_HW_VLAN_CTAG_RX)
+		if (netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_RX_BIT,
+					    features))
 			tp->cp_cmd |= RxVlan;
 		else
 			tp->cp_cmd &= ~RxVlan;
@@ -4331,15 +4334,15 @@ static void rtl8168evl_fix_tso(struct sk_buff *skb, netdev_features_t *features)
 	/* IPv4 header has options field */
 	if (vlan_get_protocol(skb) == htons(ETH_P_IP) &&
 	    ip_hdrlen(skb) > sizeof(struct iphdr))
-		*features &= ~NETIF_F_ALL_TSO;
+		netdev_feature_clear_bits(NETIF_F_ALL_TSO, features);
 
 	/* IPv4 TCP header has options field */
 	else if (skb_shinfo(skb)->gso_type & SKB_GSO_TCPV4 &&
 		 tcp_hdrlen(skb) > sizeof(struct tcphdr))
-		*features &= ~NETIF_F_ALL_TSO;
+		netdev_feature_clear_bits(NETIF_F_ALL_TSO, features);
 
 	else if (rtl_last_frag_len(skb) <= 6)
-		*features &= ~NETIF_F_ALL_TSO;
+		netdev_feature_clear_bits(NETIF_F_ALL_TSO, features);
 }
 
 static void rtl8169_features_check(struct sk_buff *skb, struct net_device *dev,
@@ -4354,18 +4357,18 @@ static void rtl8169_features_check(struct sk_buff *skb, struct net_device *dev,
 
 		if (transport_offset > GTTCPHO_MAX &&
 		    rtl_chip_supports_csum_v2(tp))
-			*features &= ~NETIF_F_ALL_TSO;
+			netdev_feature_clear_bits(NETIF_F_ALL_TSO, features);
 	} else if (skb->ip_summed == CHECKSUM_PARTIAL) {
 		/* work around hw bug on some chip versions */
 		if (skb->len < ETH_ZLEN)
-			*features &= ~NETIF_F_CSUM_MASK;
+			netdev_feature_clear_bits(NETIF_F_CSUM_MASK, features);
 
 		if (rtl_quirk_packet_padto(tp, skb))
-			*features &= ~NETIF_F_CSUM_MASK;
+			netdev_feature_clear_bits(NETIF_F_CSUM_MASK, features);
 
 		if (transport_offset > TCPHO_MAX &&
 		    rtl_chip_supports_csum_v2(tp))
-			*features &= ~NETIF_F_CSUM_MASK;
+			netdev_feature_clear_bits(NETIF_F_CSUM_MASK, features);
 	}
 
 	vlan_features_check(skb, features);
@@ -4491,14 +4494,16 @@ static int rtl_rx(struct net_device *dev, struct rtl8169_private *tp, int budget
 			if (status & RxCRC)
 				dev->stats.rx_crc_errors++;
 
-			if (!(dev->features & NETIF_F_RXALL))
+			if (!netdev_feature_test_bit(NETIF_F_RXALL_BIT,
+						     dev->features))
 				goto release_descriptor;
 			else if (status & RxRWT || !(status & (RxRUNT | RxCRC)))
 				goto release_descriptor;
 		}
 
 		pkt_size = status & GENMASK(13, 0);
-		if (likely(!(dev->features & NETIF_F_RXFCS)))
+		if (likely(!netdev_feature_test_bit(NETIF_F_RXFCS_BIT,
+						    dev->features)))
 			pkt_size -= ETH_FCS_LEN;
 
 		/* The driver does not support incoming fragmented frames.
@@ -5317,7 +5322,7 @@ static int rtl_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	if (sizeof(dma_addr_t) > 4 && tp->mac_version >= RTL_GIGA_MAC_VER_18 &&
 	    !dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(64)))
-		dev->features |= NETIF_F_HIGHDMA;
+		netdev_feature_set_bit(NETIF_F_HIGHDMA_BIT, &dev->features);
 
 	rtl_init_rxcfg(tp);
 
@@ -5341,9 +5346,13 @@ static int rtl_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	netif_napi_add(dev, &tp->napi, rtl8169_poll, NAPI_POLL_WEIGHT);
 
-	dev->hw_features = NETIF_F_IP_CSUM | NETIF_F_RXCSUM |
-			   NETIF_F_HW_VLAN_CTAG_TX | NETIF_F_HW_VLAN_CTAG_RX;
-	dev->vlan_features = NETIF_F_SG | NETIF_F_IP_CSUM | NETIF_F_TSO;
+	netdev_feature_zero(&dev->hw_features);
+	netdev_feature_set_bits(NETIF_F_IP_CSUM | NETIF_F_RXCSUM |
+				NETIF_F_HW_VLAN_CTAG_TX |
+				NETIF_F_HW_VLAN_CTAG_RX, &dev->hw_features);
+	netdev_feature_zero(&dev->vlan_features);
+	netdev_feature_set_bits(NETIF_F_SG | NETIF_F_IP_CSUM | NETIF_F_TSO,
+				&dev->vlan_features);
 	dev->priv_flags |= IFF_LIVE_ADDR_CHANGE;
 
 	/*
@@ -5352,12 +5361,14 @@ static int rtl_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	 */
 	if (tp->mac_version == RTL_GIGA_MAC_VER_05)
 		/* Disallow toggling */
-		dev->hw_features &= ~NETIF_F_HW_VLAN_CTAG_RX;
+		netdev_feature_clear_bit(NETIF_F_HW_VLAN_CTAG_RX_BIT,
+					 &dev->hw_features);
 
 	if (rtl_chip_supports_csum_v2(tp))
-		dev->hw_features |= NETIF_F_IPV6_CSUM;
+		netdev_feature_set_bit(NETIF_F_IPV6_CSUM_BIT,
+				       &dev->hw_features);
 
-	dev->features |= dev->hw_features;
+	netdev_feature_or(&dev->features, dev->features, dev->hw_features);
 
 	/* There has been a number of reports that using SG/TSO results in
 	 * tx timeouts. However for a lot of people SG/TSO works fine.
@@ -5365,17 +5376,19 @@ static int rtl_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	 * enable them. Use at own risk!
 	 */
 	if (rtl_chip_supports_csum_v2(tp)) {
-		dev->hw_features |= NETIF_F_SG | NETIF_F_TSO | NETIF_F_TSO6;
+		netdev_feature_set_bits(NETIF_F_SG | NETIF_F_TSO | NETIF_F_TSO6,
+					&dev->hw_features);
 		dev->gso_max_size = RTL_GSO_MAX_SIZE_V2;
 		dev->gso_max_segs = RTL_GSO_MAX_SEGS_V2;
 	} else {
-		dev->hw_features |= NETIF_F_SG | NETIF_F_TSO;
+		netdev_feature_set_bits(NETIF_F_SG | NETIF_F_TSO,
+					&dev->hw_features);
 		dev->gso_max_size = RTL_GSO_MAX_SIZE_V1;
 		dev->gso_max_segs = RTL_GSO_MAX_SEGS_V1;
 	}
 
-	dev->hw_features |= NETIF_F_RXALL;
-	dev->hw_features |= NETIF_F_RXFCS;
+	netdev_feature_set_bit(NETIF_F_RXALL_BIT, &dev->hw_features);
+	netdev_feature_set_bit(NETIF_F_RXFCS_BIT, &dev->hw_features);
 
 	/* configure chip for default features */
 	rtl8169_set_features(dev, dev->features);
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 084/167] net: broadcom: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (82 preceding siblings ...)
  2021-09-29 15:52 ` [RFCv2 net-next 083/167] net: realtek: " Jian Shen
@ 2021-09-29 15:52 ` Jian Shen
  2021-09-29 15:52 ` [RFCv2 net-next 085/167] net: intel: " Jian Shen
                   ` (86 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:52 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 drivers/net/ethernet/broadcom/b44.c           |   2 +-
 drivers/net/ethernet/broadcom/bcmsysport.c    |  24 +--
 drivers/net/ethernet/broadcom/bgmac.c         |   8 +-
 drivers/net/ethernet/broadcom/bnx2.c          |  49 +++---
 .../net/ethernet/broadcom/bnx2x/bnx2x_cmn.c   |  62 ++++---
 .../net/ethernet/broadcom/bnx2x/bnx2x_main.c  |  84 ++++++----
 drivers/net/ethernet/broadcom/bnxt/bnxt.c     | 152 +++++++++++-------
 drivers/net/ethernet/broadcom/bnxt/bnxt_tc.c  |   4 +-
 drivers/net/ethernet/broadcom/bnxt/bnxt_vfr.c |  11 +-
 drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.c |   3 +-
 .../net/ethernet/broadcom/genet/bcmgenet.c    |  12 +-
 drivers/net/ethernet/broadcom/tg3.c           |  56 ++++---
 12 files changed, 279 insertions(+), 188 deletions(-)

diff --git a/drivers/net/ethernet/broadcom/b44.c b/drivers/net/ethernet/broadcom/b44.c
index 38b465452902..d9e9742dec91 100644
--- a/drivers/net/ethernet/broadcom/b44.c
+++ b/drivers/net/ethernet/broadcom/b44.c
@@ -2353,7 +2353,7 @@ static int b44_init_one(struct ssb_device *sdev,
 	SET_NETDEV_DEV(dev, sdev->dev);
 
 	/* No interesting netdevice features in this card... */
-	dev->features |= 0;
+	netdev_feature_set_bits(0, &dev->features);
 
 	bp = netdev_priv(dev);
 	bp->sdev = sdev;
diff --git a/drivers/net/ethernet/broadcom/bcmsysport.c b/drivers/net/ethernet/broadcom/bcmsysport.c
index 7fa1b695400d..0d686804fa6e 100644
--- a/drivers/net/ethernet/broadcom/bcmsysport.c
+++ b/drivers/net/ethernet/broadcom/bcmsysport.c
@@ -122,7 +122,7 @@ static void bcm_sysport_set_rx_csum(struct net_device *dev,
 	struct bcm_sysport_priv *priv = netdev_priv(dev);
 	u32 reg;
 
-	priv->rx_chk_en = !!(wanted & NETIF_F_RXCSUM);
+	priv->rx_chk_en = netdev_feature_test_bit(NETIF_F_RXCSUM_BIT, wanted);
 	reg = rxchk_readl(priv, RXCHK_CONTROL);
 	/* Clear L2 header checks, which would prevent BPDUs
 	 * from being received.
@@ -162,8 +162,10 @@ static void bcm_sysport_set_tx_csum(struct net_device *dev,
 	/* Hardware transmit checksum requires us to enable the Transmit status
 	 * block prepended to the packet contents
 	 */
-	priv->tsb_en = !!(wanted & (NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM |
-				    NETIF_F_HW_VLAN_CTAG_TX));
+	priv->tsb_en = netdev_feature_test_bits(NETIF_F_IP_CSUM |
+						NETIF_F_IPV6_CSUM |
+						NETIF_F_HW_VLAN_CTAG_TX,
+						wanted);
 	reg = tdma_readl(priv, TDMA_CONTROL);
 	if (priv->tsb_en)
 		reg |= tdma_control_bit(priv, TSB_EN);
@@ -180,7 +182,7 @@ static void bcm_sysport_set_tx_csum(struct net_device *dev,
 	tdma_writel(priv, reg, TDMA_CONTROL);
 
 	/* Default TPID is ETH_P_8021AD, change to ETH_P_8021Q */
-	if (wanted & NETIF_F_HW_VLAN_CTAG_TX)
+	if (netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_TX_BIT, wanted))
 		tdma_writel(priv, ETH_P_8021Q, TDMA_TPID);
 }
 
@@ -1543,7 +1545,8 @@ static int bcm_sysport_init_tx_ring(struct bcm_sysport_priv *priv,
 	/* Adjust the packet size calculations if SYSTEMPORT is responsible
 	 * for HW insertion of VLAN tags
 	 */
-	if (priv->netdev->features & NETIF_F_HW_VLAN_CTAG_TX)
+	if (netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_TX_BIT,
+				    priv->netdev->features))
 		reg = VLAN_HLEN << RING_PKT_SIZE_ADJ_SHIFT;
 	tdma_writel(priv, reg, TDMA_DESC_RING_PCP_DEI_VID(index));
 
@@ -2567,11 +2570,12 @@ static int bcm_sysport_probe(struct platform_device *pdev)
 	dev->netdev_ops = &bcm_sysport_netdev_ops;
 	netif_napi_add(dev, &priv->napi, bcm_sysport_poll, 64);
 
-	dev->features |= NETIF_F_RXCSUM | NETIF_F_HIGHDMA |
-			 NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM |
-			 NETIF_F_HW_VLAN_CTAG_TX;
-	dev->hw_features |= dev->features;
-	dev->vlan_features |= dev->features;
+	netdev_feature_set_bits(NETIF_F_RXCSUM | NETIF_F_HIGHDMA |
+				NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM |
+				NETIF_F_HW_VLAN_CTAG_TX, &dev->features);
+	netdev_feature_or(&dev->hw_features, dev->hw_features, dev->features);
+	netdev_feature_or(&dev->vlan_features, dev->vlan_features,
+			  dev->features);
 	dev->max_mtu = UMAC_MAX_MTU_SIZE;
 
 	/* Request the WOL interrupt and advertise suspend if available */
diff --git a/drivers/net/ethernet/broadcom/bgmac.c b/drivers/net/ethernet/broadcom/bgmac.c
index fe4d99abd548..c99800b8d812 100644
--- a/drivers/net/ethernet/broadcom/bgmac.c
+++ b/drivers/net/ethernet/broadcom/bgmac.c
@@ -1535,9 +1535,11 @@ int bgmac_enet_probe(struct bgmac *bgmac)
 		goto err_dma_free;
 	}
 
-	net_dev->features = NETIF_F_SG | NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM;
-	net_dev->hw_features = net_dev->features;
-	net_dev->vlan_features = net_dev->features;
+	netdev_feature_zero(&net_dev->features);
+	netdev_feature_set_bits(NETIF_F_SG | NETIF_F_IP_CSUM |
+				NETIF_F_IPV6_CSUM, &net_dev->features);
+	netdev_feature_copy(&net_dev->hw_features, net_dev->features);
+	netdev_feature_copy(&net_dev->vlan_features, net_dev->features);
 
 	/* Omit FCS from max MTU size */
 	net_dev->max_mtu = BGMAC_RX_MAX_FRAME_SIZE - ETH_FCS_LEN;
diff --git a/drivers/net/ethernet/broadcom/bnx2.c b/drivers/net/ethernet/broadcom/bnx2.c
index 8c83973adca5..29f7b0b03d64 100644
--- a/drivers/net/ethernet/broadcom/bnx2.c
+++ b/drivers/net/ethernet/broadcom/bnx2.c
@@ -3258,7 +3258,8 @@ bnx2_rx_int(struct bnx2 *bp, struct bnx2_napi *bnapi, int budget)
 		}
 
 		skb_checksum_none_assert(skb);
-		if ((bp->dev->features & NETIF_F_RXCSUM) &&
+		if (netdev_feature_test_bit(NETIF_F_RXCSUM_BIT,
+					    bp->dev->features) &&
 			(status & (L2_FHDR_STATUS_TCP_SEGMENT |
 			L2_FHDR_STATUS_UDP_DATAGRAM))) {
 
@@ -3266,7 +3267,8 @@ bnx2_rx_int(struct bnx2 *bp, struct bnx2_napi *bnapi, int budget)
 					      L2_FHDR_ERRORS_UDP_XSUM)) == 0))
 				skb->ip_summed = CHECKSUM_UNNECESSARY;
 		}
-		if ((bp->dev->features & NETIF_F_RXHASH) &&
+		if (netdev_feature_test_bit(NETIF_F_RXHASH_BIT,
+					    bp->dev->features) &&
 		    ((status & L2_FHDR_STATUS_USE_RXHASH) ==
 		     L2_FHDR_STATUS_USE_RXHASH))
 			skb_set_hash(skb, rx_hdr->l2_fhdr_hash,
@@ -3586,7 +3588,8 @@ bnx2_set_rx_mode(struct net_device *dev)
 	rx_mode = bp->rx_mode & ~(BNX2_EMAC_RX_MODE_PROMISCUOUS |
 				  BNX2_EMAC_RX_MODE_KEEP_VLAN_TAG);
 	sort_mode = 1 | BNX2_RPM_SORT_USER0_BC_EN;
-	if (!(dev->features & NETIF_F_HW_VLAN_CTAG_RX) &&
+	if (!netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_RX_BIT,
+				     dev->features) &&
 	     (bp->flags & BNX2_FLAG_CAN_KEEP_VLAN))
 		rx_mode |= BNX2_EMAC_RX_MODE_KEEP_VLAN_TAG;
 	if (dev->flags & IFF_PROMISC) {
@@ -7747,18 +7750,23 @@ static int
 bnx2_set_features(struct net_device *dev, netdev_features_t features)
 {
 	struct bnx2 *bp = netdev_priv(dev);
+	netdev_features_t tmp;
 
 	/* TSO with VLAN tag won't work with current firmware */
-	if (features & NETIF_F_HW_VLAN_CTAG_TX)
-		dev->vlan_features |= (dev->hw_features & NETIF_F_ALL_TSO);
-	else
-		dev->vlan_features &= ~NETIF_F_ALL_TSO;
+	if (netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_TX_BIT, features)) {
+		netdev_feature_copy(&tmp, dev->hw_features);
+		netdev_feature_and_bits(NETIF_F_ALL_TSO, &tmp);
+		netdev_feature_or(&dev->vlan_features, dev->vlan_features,
+				  tmp);
+	} else {
+		netdev_feature_clear_bits(NETIF_F_ALL_TSO, &dev->vlan_features);
+	}
 
-	if ((!!(features & NETIF_F_HW_VLAN_CTAG_RX) !=
+	if ((netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_RX_BIT, features) !=
 	    !!(bp->rx_mode & BNX2_EMAC_RX_MODE_KEEP_VLAN_TAG)) &&
 	    netif_running(dev)) {
 		bnx2_netif_stop(bp, false);
-		dev->features = features;
+		netdev_feature_copy(&dev->features, features);
 		bnx2_set_rx_mode(dev);
 		bnx2_fw_sync(bp, BNX2_DRV_MSG_CODE_KEEP_VLAN_UPDATE, 0, 1);
 		bnx2_netif_start(bp, false);
@@ -8208,7 +8216,7 @@ bnx2_init_board(struct pci_dev *pdev, struct net_device *dev)
 
 	/* Configure DMA attributes. */
 	if (dma_set_mask(&pdev->dev, dma_mask) == 0) {
-		dev->features |= NETIF_F_HIGHDMA;
+		netdev_feature_set_bit(NETIF_F_HIGHDMA_BIT, &dev->features);
 		rc = dma_set_coherent_mask(&pdev->dev, persist_dma_mask);
 		if (rc) {
 			dev_err(&pdev->dev,
@@ -8576,22 +8584,27 @@ bnx2_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	memcpy(dev->dev_addr, bp->mac_addr, ETH_ALEN);
 
-	dev->hw_features = NETIF_F_IP_CSUM | NETIF_F_SG |
-		NETIF_F_TSO | NETIF_F_TSO_ECN |
-		NETIF_F_RXHASH | NETIF_F_RXCSUM;
+	netdev_feature_zero(&dev->hw_features);
+	netdev_feature_set_bits(NETIF_F_IP_CSUM | NETIF_F_SG |
+				NETIF_F_TSO | NETIF_F_TSO_ECN |
+				NETIF_F_RXHASH | NETIF_F_RXCSUM,
+				&dev->hw_features);
 
 	if (BNX2_CHIP(bp) == BNX2_CHIP_5709)
-		dev->hw_features |= NETIF_F_IPV6_CSUM | NETIF_F_TSO6;
+		netdev_feature_set_bits(NETIF_F_IPV6_CSUM | NETIF_F_TSO6,
+					&dev->hw_features);
 
-	dev->vlan_features = dev->hw_features;
-	dev->hw_features |= NETIF_F_HW_VLAN_CTAG_TX | NETIF_F_HW_VLAN_CTAG_RX;
-	dev->features |= dev->hw_features;
+	netdev_feature_copy(&dev->vlan_features, dev->hw_features);
+	netdev_feature_set_bits(NETIF_F_HW_VLAN_CTAG_TX |
+				NETIF_F_HW_VLAN_CTAG_RX, &dev->hw_features);
+	netdev_feature_or(&dev->features, dev->features, dev->hw_features);
 	dev->priv_flags |= IFF_UNICAST_FLT;
 	dev->min_mtu = MIN_ETHERNET_PACKET_SIZE;
 	dev->max_mtu = MAX_ETHERNET_JUMBO_PACKET_SIZE;
 
 	if (!(bp->flags & BNX2_FLAG_CAN_KEEP_VLAN))
-		dev->hw_features &= ~NETIF_F_HW_VLAN_CTAG_RX;
+		netdev_feature_clear_bit(NETIF_F_HW_VLAN_CTAG_RX_BIT,
+					 &dev->hw_features);
 
 	if ((rc = register_netdev(dev))) {
 		dev_err(&pdev->dev, "Cannot register net device\n");
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c
index d74510306068..b238312fd0c7 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c
@@ -412,7 +412,7 @@ static u32 bnx2x_get_rxhash(const struct bnx2x *bp,
 			    enum pkt_hash_types *rxhash_type)
 {
 	/* Get Toeplitz hash from CQE */
-	if ((bp->dev->features & NETIF_F_RXHASH) &&
+	if (netdev_feature_test_bit(NETIF_F_RXHASH_BIT, bp->dev->features) &&
 	    (cqe->status_flags & ETH_FAST_PATH_RX_CQE_RSS_HASH_FLG)) {
 		enum eth_rss_hash_type htype;
 
@@ -1073,7 +1073,8 @@ static int bnx2x_rx_int(struct bnx2x_fastpath *fp, int budget)
 
 		skb_checksum_none_assert(skb);
 
-		if (bp->dev->features & NETIF_F_RXCSUM)
+		if (netdev_feature_test_bit(NETIF_F_RXCSUM_BIT,
+					    bp->dev->features))
 			bnx2x_csum_validate(skb, cqe, fp,
 					    bnx2x_fp_qstats(bp, fp));
 
@@ -2491,9 +2492,10 @@ static void bnx2x_bz_fp(struct bnx2x *bp, int index)
 	/* set the tpa flag for each queue. The tpa flag determines the queue
 	 * minimal size so it must be set prior to queue memory allocation
 	 */
-	if (bp->dev->features & NETIF_F_LRO)
+	if (netdev_feature_test_bit(NETIF_F_LRO_BIT, bp->dev->features))
 		fp->mode = TPA_MODE_LRO;
-	else if (bp->dev->features & NETIF_F_GRO_HW)
+	else if (netdev_feature_test_bit(NETIF_F_GRO_HW_BIT,
+					 bp->dev->features))
 		fp->mode = TPA_MODE_GRO;
 	else
 		fp->mode = TPA_MODE_DISABLED;
@@ -4891,7 +4893,7 @@ int bnx2x_change_mtu(struct net_device *dev, int new_mtu)
 	dev->mtu = new_mtu;
 
 	if (!bnx2x_mtu_allows_gro(new_mtu))
-		dev->features &= ~NETIF_F_GRO_HW;
+		netdev_feature_clear_bit(NETIF_F_GRO_HW_BIT, &dev->features);
 
 	if (IS_PF(bp) && SHMEM2_HAS(bp, curr_cfg))
 		SHMEM2_WR(bp, curr_cfg, CURR_CFG_MET_OS);
@@ -4904,42 +4906,54 @@ void bnx2x_fix_features(struct net_device *dev, netdev_features_t *features)
 	struct bnx2x *bp = netdev_priv(dev);
 
 	if (pci_num_vf(bp->pdev)) {
-		netdev_features_t changed = dev->features ^ *features;
+		netdev_features_t changed;
+
+		netdev_feature_xor(&changed, dev->features, *features);
 
 		/* Revert the requested changes in features if they
 		 * would require internal reload of PF in bnx2x_set_features().
 		 */
-		if (!(*features & NETIF_F_RXCSUM) && !bp->disable_tpa) {
-			*features &= ~NETIF_F_RXCSUM;
-			*features |= dev->features & NETIF_F_RXCSUM;
+		if (!netdev_feature_test_bit(NETIF_F_RXCSUM_BIT, *features) &&
+		    !bp->disable_tpa) {
+			if (netdev_feature_test_bit(NETIF_F_RXCSUM_BIT,
+						    dev->features))
+				netdev_feature_set_bit(NETIF_F_RXCSUM_BIT,
+						       features);
 		}
 
-		if (changed & NETIF_F_LOOPBACK) {
-			*features &= ~NETIF_F_LOOPBACK;
-			*features |= dev->features & NETIF_F_LOOPBACK;
+		if (netdev_feature_test_bit(NETIF_F_LOOPBACK_BIT, changed)) {
+			netdev_feature_clear_bit(NETIF_F_LOOPBACK_BIT,
+						 features);
+			if (netdev_feature_test_bit(NETIF_F_LOOPBACK_BIT,
+						    dev->features))
+				netdev_feature_set_bit(NETIF_F_LOOPBACK_BIT,
+						       features);
 		}
 	}
 
 	/* TPA requires Rx CSUM offloading */
-	if (!(*features & NETIF_F_RXCSUM))
-		*features &= ~NETIF_F_LRO;
-
-	if (!(*features & NETIF_F_GRO) || !bnx2x_mtu_allows_gro(dev->mtu))
-		*features &= ~NETIF_F_GRO_HW;
-	if (*features & NETIF_F_GRO_HW)
-		*features &= ~NETIF_F_LRO;
+	if (!netdev_feature_test_bit(NETIF_F_RXCSUM_BIT, *features))
+		netdev_feature_clear_bit(NETIF_F_LRO_BIT, features);
+
+	if (!netdev_feature_test_bit(NETIF_F_GRO_BIT, *features) ||
+	    !bnx2x_mtu_allows_gro(dev->mtu))
+		netdev_feature_clear_bit(NETIF_F_GRO_HW_BIT, features);
+	if (netdev_feature_test_bit(NETIF_F_GRO_HW_BIT, *features))
+		netdev_feature_clear_bit(NETIF_F_LRO_BIT, features);
 }
 
 int bnx2x_set_features(struct net_device *dev, netdev_features_t features)
 {
 	struct bnx2x *bp = netdev_priv(dev);
-	netdev_features_t changes = features ^ dev->features;
+	netdev_features_t changes;
 	bool bnx2x_reload = false;
 	int rc;
 
+	netdev_feature_xor(&changes, dev->features, features);
+
 	/* VFs or non SRIOV PFs should be able to change loopback feature */
 	if (!pci_num_vf(bp->pdev)) {
-		if (features & NETIF_F_LOOPBACK) {
+		if (netdev_feature_test_bit(NETIF_F_LOOPBACK_BIT, features)) {
 			if (bp->link_params.loopback_mode != LOOPBACK_BMAC) {
 				bp->link_params.loopback_mode = LOOPBACK_BMAC;
 				bnx2x_reload = true;
@@ -4953,14 +4967,14 @@ int bnx2x_set_features(struct net_device *dev, netdev_features_t features)
 	}
 
 	/* Don't care about GRO changes */
-	changes &= ~NETIF_F_GRO;
+	netdev_feature_clear_bit(NETIF_F_GRO_BIT, &changes);
 
-	if (changes)
+	if (!netdev_feature_empty(changes))
 		bnx2x_reload = true;
 
 	if (bnx2x_reload) {
 		if (bp->recovery_state == BNX2X_RECOVERY_DONE) {
-			dev->features = features;
+			netdev_feature_copy(&dev->features, features);
 			rc = bnx2x_reload_if_running(dev);
 			return rc ? rc : 1;
 		}
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
index 1f817dfa5b64..67fbe0ebf3b6 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
@@ -3393,9 +3393,9 @@ static void bnx2x_drv_info_ether_stat(struct bnx2x *bp)
 				ether_stat->mac_local + MAC_PAD, MAC_PAD,
 				ETH_ALEN);
 	ether_stat->mtu_size = bp->dev->mtu;
-	if (bp->dev->features & NETIF_F_RXCSUM)
+	if (netdev_feature_test_bit(NETIF_F_RXCSUM_BIT, bp->dev->features))
 		ether_stat->feature_flags |= FEATURE_ETH_CHKSUM_OFFLOAD_MASK;
-	if (bp->dev->features & NETIF_F_TSO)
+	if (netdev_feature_test_bit(NETIF_F_TSO_BIT, bp->dev->features))
 		ether_stat->feature_flags |= FEATURE_ETH_LSO_MASK;
 	ether_stat->feature_flags |= bp->common.boot_mode;
 
@@ -12332,8 +12332,10 @@ static int bnx2x_init_bp(struct bnx2x *bp)
 
 	/* Set TPA flags */
 	if (bp->disable_tpa) {
-		bp->dev->hw_features &= ~(NETIF_F_LRO | NETIF_F_GRO_HW);
-		bp->dev->features &= ~(NETIF_F_LRO | NETIF_F_GRO_HW);
+		netdev_feature_clear_bits(NETIF_F_LRO | NETIF_F_GRO_HW,
+					  &bp->dev->hw_features);
+		netdev_feature_clear_bits(NETIF_F_LRO | NETIF_F_GRO_HW,
+					  &bp->dev->features);
 	}
 
 	if (CHIP_IS_E1(bp))
@@ -12853,7 +12855,7 @@ static void bnx2x_features_check(struct sk_buff *skb, struct net_device *dev,
 	if (unlikely(skb_is_gso(skb) &&
 		     (skb_shinfo(skb)->gso_size > 9000) &&
 		     !skb_gso_validate_mac_len(skb, 9700)))
-		*features &= ~NETIF_F_GSO_MASK;
+		netdev_feature_clear_bits(NETIF_F_GSO_MASK, features);
 
 	vlan_features_check(skb, features);
 	vxlan_features_check(skb, features);
@@ -13190,52 +13192,62 @@ static int bnx2x_init_dev(struct bnx2x *bp, struct pci_dev *pdev,
 
 	dev->priv_flags |= IFF_UNICAST_FLT;
 
-	dev->hw_features = NETIF_F_SG | NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM |
-		NETIF_F_TSO | NETIF_F_TSO_ECN | NETIF_F_TSO6 |
-		NETIF_F_RXCSUM | NETIF_F_LRO | NETIF_F_GRO | NETIF_F_GRO_HW |
-		NETIF_F_RXHASH | NETIF_F_HW_VLAN_CTAG_TX;
+	netdev_feature_zero(&dev->hw_features);
+	netdev_feature_set_bits(NETIF_F_SG | NETIF_F_IP_CSUM |
+				NETIF_F_IPV6_CSUM | NETIF_F_TSO |
+				NETIF_F_TSO_ECN | NETIF_F_TSO6 |
+				NETIF_F_RXCSUM | NETIF_F_LRO | NETIF_F_GRO |
+				NETIF_F_GRO_HW | NETIF_F_RXHASH |
+				NETIF_F_HW_VLAN_CTAG_TX, &dev->hw_features);
 	if (!chip_is_e1x) {
-		dev->hw_features |= NETIF_F_GSO_GRE | NETIF_F_GSO_GRE_CSUM |
-				    NETIF_F_GSO_IPXIP4 |
-				    NETIF_F_GSO_UDP_TUNNEL |
-				    NETIF_F_GSO_UDP_TUNNEL_CSUM |
-				    NETIF_F_GSO_PARTIAL;
-
-		dev->hw_enc_features =
-			NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM | NETIF_F_SG |
-			NETIF_F_TSO | NETIF_F_TSO_ECN | NETIF_F_TSO6 |
-			NETIF_F_GSO_IPXIP4 |
-			NETIF_F_GSO_GRE | NETIF_F_GSO_GRE_CSUM |
-			NETIF_F_GSO_UDP_TUNNEL | NETIF_F_GSO_UDP_TUNNEL_CSUM |
-			NETIF_F_GSO_PARTIAL;
-
-		dev->gso_partial_features = NETIF_F_GSO_GRE_CSUM |
-					    NETIF_F_GSO_UDP_TUNNEL_CSUM;
+		netdev_feature_set_bits(NETIF_F_GSO_GRE | NETIF_F_GSO_GRE_CSUM |
+				NETIF_F_GSO_IPXIP4 | NETIF_F_GSO_UDP_TUNNEL |
+				NETIF_F_GSO_UDP_TUNNEL_CSUM |
+				NETIF_F_GSO_PARTIAL, &dev->hw_features);
+
+		netdev_feature_zero(&dev->hw_enc_features);
+		netdev_feature_set_bits(NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM |
+				NETIF_F_SG | NETIF_F_TSO |
+				NETIF_F_TSO_ECN | NETIF_F_TSO6 |
+				NETIF_F_GSO_IPXIP4 | NETIF_F_GSO_GRE |
+				NETIF_F_GSO_GRE_CSUM | NETIF_F_GSO_UDP_TUNNEL |
+				NETIF_F_GSO_UDP_TUNNEL_CSUM |
+				NETIF_F_GSO_PARTIAL, &dev->hw_enc_features);
+
+		netdev_feature_zero(&dev->gso_partial_features);
+		netdev_feature_set_bits(NETIF_F_GSO_GRE_CSUM |
+					NETIF_F_GSO_UDP_TUNNEL_CSUM,
+					&dev->gso_partial_features);
 
 		if (IS_PF(bp))
 			dev->udp_tunnel_nic_info = &bnx2x_udp_tunnels;
 	}
 
-	dev->vlan_features = NETIF_F_SG | NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM |
-		NETIF_F_TSO | NETIF_F_TSO_ECN | NETIF_F_TSO6 | NETIF_F_HIGHDMA;
+	netdev_feature_zero(&dev->vlan_features);
+	netdev_feature_set_bits(NETIF_F_SG | NETIF_F_IP_CSUM |
+				NETIF_F_IPV6_CSUM | NETIF_F_TSO |
+				NETIF_F_TSO_ECN | NETIF_F_TSO6 |
+				NETIF_F_HIGHDMA, &dev->vlan_features);
 
 	if (IS_PF(bp)) {
 		if (chip_is_e1x)
 			bp->accept_any_vlan = true;
 		else
-			dev->hw_features |= NETIF_F_HW_VLAN_CTAG_FILTER;
+			netdev_feature_set_bit(NETIF_F_HW_VLAN_CTAG_FILTER_BIT,
+					       &dev->hw_features);
 	}
 	/* For VF we'll know whether to enable VLAN filtering after
 	 * getting a response to CHANNEL_TLV_ACQUIRE from PF.
 	 */
 
-	dev->features |= dev->hw_features | NETIF_F_HW_VLAN_CTAG_RX;
-	dev->features |= NETIF_F_HIGHDMA;
-	if (dev->features & NETIF_F_LRO)
-		dev->features &= ~NETIF_F_GRO_HW;
+	netdev_feature_or(&dev->features, dev->features, dev->hw_features);
+	netdev_feature_set_bit(NETIF_F_HW_VLAN_CTAG_RX_BIT, &dev->features);
+	netdev_feature_set_bit(NETIF_F_HIGHDMA_BIT, &dev->features);
+	if (netdev_feature_test_bit(NETIF_F_LRO_BIT, dev->features))
+		netdev_feature_clear_bit(NETIF_F_GRO_HW_BIT, &dev->features);
 
 	/* Add Loopback capability to the device */
-	dev->hw_features |= NETIF_F_LOOPBACK;
+	netdev_feature_set_bit(NETIF_F_LOOPBACK_BIT, &dev->features);
 
 #ifdef BCM_DCBNL
 	dev->dcbnl_ops = &bnx2x_dcbnl_ops;
@@ -13938,8 +13950,10 @@ static int bnx2x_init_one(struct pci_dev *pdev,
 #ifdef CONFIG_BNX2X_SRIOV
 		/* VF with OLD Hypervisor or old PF do not support filtering */
 		if (bp->acquire_resp.pfdev_info.pf_cap & PFVF_CAP_VLAN_FILTER) {
-			dev->hw_features |= NETIF_F_HW_VLAN_CTAG_FILTER;
-			dev->features |= NETIF_F_HW_VLAN_CTAG_FILTER;
+			netdev_feature_set_bit(NETIF_F_HW_VLAN_CTAG_FILTER_BIT,
+					       &dev->hw_features);
+			netdev_feature_set_bit(NETIF_F_HW_VLAN_CTAG_FILTER_BIT,
+					       &dev->features);
 		}
 #endif
 	}
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index 7d9166876e95..24d6f883c243 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -1704,7 +1704,8 @@ static inline struct sk_buff *bnxt_tpa_end(struct bnxt *bp,
 		skb_set_hash(skb, tpa_info->rss_hash, tpa_info->hash_type);
 
 	if ((tpa_info->flags2 & RX_CMP_FLAGS2_META_FORMAT_VLAN) &&
-	    (skb->dev->features & BNXT_HW_FEATURE_VLAN_ALL_RX)) {
+	    (netdev_feature_test_bits(BNXT_HW_FEATURE_VLAN_ALL_RX,
+				      skb->dev->features))) {
 		__be16 vlan_proto = htons(tpa_info->metadata >>
 					  RX_CMP_FLAGS2_METADATA_TPID_SFT);
 		u16 vtag = tpa_info->metadata & RX_CMP_FLAGS2_METADATA_TCI_MASK;
@@ -1940,7 +1941,8 @@ static int bnxt_rx_pkt(struct bnxt *bp, struct bnxt_cp_ring_info *cpr,
 
 	if ((rxcmp1->rx_cmp_flags2 &
 	     cpu_to_le32(RX_CMP_FLAGS2_META_FORMAT_VLAN)) &&
-	    (skb->dev->features & BNXT_HW_FEATURE_VLAN_ALL_RX)) {
+	     netdev_feature_test_bits(BNXT_HW_FEATURE_VLAN_ALL_RX,
+				      skb->dev->features)) {
 		u32 meta_data = le32_to_cpu(rxcmp1->rx_cmp_meta_data);
 		u16 vtag = meta_data & RX_CMP_FLAGS2_METADATA_TCI_MASK;
 		__be16 vlan_proto = htons(meta_data >>
@@ -1956,13 +1958,15 @@ static int bnxt_rx_pkt(struct bnxt *bp, struct bnxt_cp_ring_info *cpr,
 
 	skb_checksum_none_assert(skb);
 	if (RX_CMP_L4_CS_OK(rxcmp1)) {
-		if (dev->features & NETIF_F_RXCSUM) {
+		if (netdev_feature_test_bit(NETIF_F_RXCSUM_BIT,
+					    dev->features)) {
 			skb->ip_summed = CHECKSUM_UNNECESSARY;
 			skb->csum_level = RX_CMP_ENCAP(rxcmp1);
 		}
 	} else {
 		if (rxcmp1->rx_cmp_cfa_code_errors_v2 & RX_CMP_L4_CS_ERR_BITS) {
-			if (dev->features & NETIF_F_RXCSUM)
+			if (netdev_feature_test_bit(NETIF_F_RXCSUM_BIT,
+						    dev->features))
 				bnapi->cp_ring.sw_stats.rx.rx_l4_csum_errors++;
 		}
 	}
@@ -3756,9 +3760,10 @@ void bnxt_set_tpa_flags(struct bnxt *bp)
 	bp->flags &= ~BNXT_FLAG_TPA;
 	if (bp->flags & BNXT_FLAG_NO_AGG_RINGS)
 		return;
-	if (bp->dev->features & NETIF_F_LRO)
+	if (netdev_feature_test_bit(NETIF_F_LRO_BIT, bp->dev->features))
 		bp->flags |= BNXT_FLAG_LRO;
-	else if (bp->dev->features & NETIF_F_GRO_HW)
+	else if (netdev_feature_test_bit(NETIF_F_GRO_HW_BIT,
+					 bp->dev->features))
 		bp->flags |= BNXT_FLAG_GRO;
 }
 
@@ -6344,8 +6349,10 @@ static int __bnxt_reserve_rings(struct bnxt *bp)
 
 			bp->flags &= ~BNXT_FLAG_AGG_RINGS;
 			bp->flags |= BNXT_FLAG_NO_AGG_RINGS;
-			bp->dev->hw_features &= ~NETIF_F_LRO;
-			bp->dev->features &= ~NETIF_F_LRO;
+			netdev_feature_clear_bit(NETIF_F_LRO_BIT,
+						 &bp->dev->hw_features);
+			netdev_feature_clear_bit(NETIF_F_LRO_BIT,
+						 &bp->dev->features);
 			bnxt_set_ring_params(bp);
 		}
 	}
@@ -10189,7 +10196,8 @@ static int __bnxt_open_nic(struct bnxt *bp, bool irq_re_init, bool link_re_init)
 	if ((bp->flags & BNXT_FLAG_RFS) &&
 	    !(bp->flags & BNXT_FLAG_USING_MSIX)) {
 		/* disable RFS if falling back to INTA */
-		bp->dev->hw_features &= ~NETIF_F_NTUPLE;
+		netdev_feature_clear_bit(NETIF_F_NTUPLE_BIT,
+					 &bp->dev->hw_features);
 		bp->flags &= ~BNXT_FLAG_RFS;
 	}
 
@@ -10897,34 +10905,42 @@ static bool bnxt_rfs_capable(struct bnxt *bp)
 static void bnxt_fix_features(struct net_device *dev,
 			      netdev_features_t *features)
 {
+	netdev_features_t vlan_features, tmp;
 	struct bnxt *bp = netdev_priv(dev);
-	netdev_features_t vlan_features;
 
-	if ((*features & NETIF_F_NTUPLE) && !bnxt_rfs_capable(bp))
-		*features &= ~NETIF_F_NTUPLE;
+	if (netdev_feature_test_bit(NETIF_F_NTUPLE_BIT, *features) &&
+	    !bnxt_rfs_capable(bp))
+		netdev_feature_clear_bit(NETIF_F_NTUPLE_BIT, features);
 
 	if (bp->flags & BNXT_FLAG_NO_AGG_RINGS)
-		*features &= ~(NETIF_F_LRO | NETIF_F_GRO_HW);
+		netdev_feature_clear_bits(NETIF_F_LRO | NETIF_F_GRO_HW,
+					  features);
 
-	if (!(*features & NETIF_F_GRO))
-		*features &= ~NETIF_F_GRO_HW;
+	if (!netdev_feature_test_bit(NETIF_F_GRO_BIT, *features))
+		netdev_feature_clear_bit(NETIF_F_GRO_HW_BIT, features);
 
-	if (*features & NETIF_F_GRO_HW)
-		*features &= ~NETIF_F_LRO;
+	if (netdev_feature_test_bit(NETIF_F_GRO_HW_BIT, *features))
+		netdev_feature_clear_bit(NETIF_F_LRO_BIT, features);
 
 	/* Both CTAG and STAG VLAN accelaration on the RX side have to be
 	 * turned on or off together.
 	 */
-	vlan_features = *features & BNXT_HW_FEATURE_VLAN_ALL_RX;
-	if (vlan_features != BNXT_HW_FEATURE_VLAN_ALL_RX) {
-		if (dev->features & BNXT_HW_FEATURE_VLAN_ALL_RX)
-			*features &= ~BNXT_HW_FEATURE_VLAN_ALL_RX;
-		else if (vlan_features)
-			*features |= BNXT_HW_FEATURE_VLAN_ALL_RX;
+	netdev_feature_zero(&tmp);
+	netdev_feature_set_bits(BNXT_HW_FEATURE_VLAN_ALL_RX, &tmp);
+	netdev_feature_and(&vlan_features, *features, tmp);
+	if (!netdev_feature_equal(vlan_features, tmp)) {
+		if (netdev_feature_test_bits(BNXT_HW_FEATURE_VLAN_ALL_RX,
+					     dev->features))
+			netdev_feature_clear_bits(BNXT_HW_FEATURE_VLAN_ALL_RX,
+						  features);
+		else if (!netdev_feature_empty(vlan_features))
+			netdev_feature_set_bits(BNXT_HW_FEATURE_VLAN_ALL_RX,
+						features);
 	}
 #ifdef CONFIG_BNXT_SRIOV
 	if (BNXT_VF(bp) && bp->vf.vlan)
-		*features &= ~BNXT_HW_FEATURE_VLAN_ALL_RX;
+		netdev_feature_clear_bits(BNXT_HW_FEATURE_VLAN_ALL_RX,
+					  features);
 #endif
 }
 
@@ -10938,18 +10954,18 @@ static int bnxt_set_features(struct net_device *dev, netdev_features_t features)
 	bool update_tpa = false;
 
 	flags &= ~BNXT_FLAG_ALL_CONFIG_FEATS;
-	if (features & NETIF_F_GRO_HW)
+	if (netdev_feature_test_bit(NETIF_F_GRO_HW_BIT, features))
 		flags |= BNXT_FLAG_GRO;
-	else if (features & NETIF_F_LRO)
+	else if (netdev_feature_test_bit(NETIF_F_LRO_BIT, features))
 		flags |= BNXT_FLAG_LRO;
 
 	if (bp->flags & BNXT_FLAG_NO_AGG_RINGS)
 		flags &= ~BNXT_FLAG_TPA;
 
-	if (features & BNXT_HW_FEATURE_VLAN_ALL_RX)
+	if (netdev_feature_test_bits(BNXT_HW_FEATURE_VLAN_ALL_RX, features))
 		flags |= BNXT_FLAG_STRIP_VLAN;
 
-	if (features & NETIF_F_NTUPLE)
+	if (netdev_feature_test_bit(NETIF_F_NTUPLE_BIT, features))
 		flags |= BNXT_FLAG_RFS;
 
 	changes = flags ^ bp->flags;
@@ -11112,7 +11128,8 @@ static void bnxt_features_check(struct sk_buff *skb, struct net_device *dev,
 			return;
 		break;
 	}
-	*features &= ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
+	netdev_feature_clear_bits(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK,
+				  features);
 }
 
 int bnxt_dbg_hwrm_rd_reg(struct bnxt *bp, u32 reg_off, u16 num_words,
@@ -11956,14 +11973,15 @@ static void bnxt_set_dflt_rfs(struct bnxt *bp)
 {
 	struct net_device *dev = bp->dev;
 
-	dev->hw_features &= ~NETIF_F_NTUPLE;
-	dev->features &= ~NETIF_F_NTUPLE;
+	netdev_feature_clear_bit(NETIF_F_NTUPLE_BIT, &dev->hw_features);
+	netdev_feature_clear_bit(NETIF_F_NTUPLE_BIT, &dev->features);
 	bp->flags &= ~BNXT_FLAG_RFS;
 	if (bnxt_rfs_supported(bp)) {
-		dev->hw_features |= NETIF_F_NTUPLE;
+		netdev_feature_set_bit(NETIF_F_NTUPLE_BIT, &dev->hw_features);
 		if (bnxt_rfs_capable(bp)) {
 			bp->flags |= BNXT_FLAG_RFS;
-			dev->features |= NETIF_F_NTUPLE;
+			netdev_feature_set_bit(NETIF_F_NTUPLE_BIT,
+					       &dev->features);
 		}
 	}
 }
@@ -12948,8 +12966,10 @@ static int bnxt_get_dflt_rings(struct bnxt *bp, int *max_rx, int *max_tx,
 			return rc;
 		}
 		bp->flags |= BNXT_FLAG_NO_AGG_RINGS;
-		bp->dev->hw_features &= ~(NETIF_F_LRO | NETIF_F_GRO_HW);
-		bp->dev->features &= ~(NETIF_F_LRO | NETIF_F_GRO_HW);
+		netdev_feature_clear_bits(NETIF_F_LRO | NETIF_F_GRO_HW,
+					  &bp->dev->hw_features);
+		netdev_feature_clear_bits(NETIF_F_LRO | NETIF_F_GRO_HW,
+					  &bp->dev->features);
 		bnxt_set_ring_params(bp);
 	}
 
@@ -13064,7 +13084,7 @@ static int bnxt_init_dflt_ring_mode(struct bnxt *bp)
 	bp->tx_nr_rings_per_tc = bp->tx_nr_rings;
 	if (bnxt_rfs_supported(bp) && bnxt_rfs_capable(bp)) {
 		bp->flags |= BNXT_FLAG_RFS;
-		bp->dev->features |= NETIF_F_NTUPLE;
+		netdev_feature_set_bit(NETIF_F_NTUPLE_BIT, &bp->dev->features);
 	}
 init_dflt_ring_err:
 	bnxt_ulp_irq_restart(bp, rc);
@@ -13260,37 +13280,46 @@ static int bnxt_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 		goto init_err_pci_clean;
 	}
 
-	dev->hw_features = NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM | NETIF_F_SG |
-			   NETIF_F_TSO | NETIF_F_TSO6 |
-			   NETIF_F_GSO_UDP_TUNNEL | NETIF_F_GSO_GRE |
-			   NETIF_F_GSO_IPXIP4 |
-			   NETIF_F_GSO_UDP_TUNNEL_CSUM | NETIF_F_GSO_GRE_CSUM |
-			   NETIF_F_GSO_PARTIAL | NETIF_F_RXHASH |
-			   NETIF_F_RXCSUM | NETIF_F_GRO;
+	netdev_feature_zero(&dev->hw_features);
+	netdev_feature_set_bits(NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM |
+				NETIF_F_SG | NETIF_F_TSO | NETIF_F_TSO6 |
+				NETIF_F_GSO_UDP_TUNNEL | NETIF_F_GSO_GRE |
+				NETIF_F_GSO_IPXIP4 |
+				NETIF_F_GSO_UDP_TUNNEL_CSUM |
+				NETIF_F_GSO_GRE_CSUM | NETIF_F_GSO_PARTIAL |
+				NETIF_F_RXHASH | NETIF_F_RXCSUM | NETIF_F_GRO,
+				&dev->hw_features);
 
 	if (BNXT_SUPPORTS_TPA(bp))
-		dev->hw_features |= NETIF_F_LRO;
-
-	dev->hw_enc_features =
-			NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM | NETIF_F_SG |
-			NETIF_F_TSO | NETIF_F_TSO6 |
-			NETIF_F_GSO_UDP_TUNNEL | NETIF_F_GSO_GRE |
-			NETIF_F_GSO_UDP_TUNNEL_CSUM | NETIF_F_GSO_GRE_CSUM |
-			NETIF_F_GSO_IPXIP4 | NETIF_F_GSO_PARTIAL;
+		netdev_feature_set_bit(NETIF_F_LRO_BIT, &dev->hw_features);
+
+	netdev_feature_zero(&dev->hw_enc_features);
+	netdev_feature_set_bits(NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM |
+				NETIF_F_SG | NETIF_F_TSO | NETIF_F_TSO6 |
+				NETIF_F_GSO_UDP_TUNNEL | NETIF_F_GSO_GRE |
+				NETIF_F_GSO_UDP_TUNNEL_CSUM |
+				NETIF_F_GSO_GRE_CSUM | NETIF_F_GSO_IPXIP4 |
+				NETIF_F_GSO_PARTIAL, &dev->hw_enc_features);
 	dev->udp_tunnel_nic_info = &bnxt_udp_tunnels;
 
-	dev->gso_partial_features = NETIF_F_GSO_UDP_TUNNEL_CSUM |
-				    NETIF_F_GSO_GRE_CSUM;
-	dev->vlan_features = dev->hw_features | NETIF_F_HIGHDMA;
+	netdev_feature_zero(&dev->gso_partial_features);
+	netdev_feature_set_bits(NETIF_F_GSO_UDP_TUNNEL_CSUM |
+				NETIF_F_GSO_GRE_CSUM,
+				&dev->gso_partial_features);
+	netdev_feature_copy(&dev->vlan_features, dev->hw_features);
+	netdev_feature_set_bit(NETIF_F_HIGHDMA_BIT, &dev->vlan_features);
 	if (bp->fw_cap & BNXT_FW_CAP_VLAN_RX_STRIP)
-		dev->hw_features |= BNXT_HW_FEATURE_VLAN_ALL_RX;
+		netdev_feature_set_bits(BNXT_HW_FEATURE_VLAN_ALL_RX,
+					&dev->hw_features);
 	if (bp->fw_cap & BNXT_FW_CAP_VLAN_TX_INSERT)
-		dev->hw_features |= BNXT_HW_FEATURE_VLAN_ALL_TX;
+		netdev_feature_set_bits(BNXT_HW_FEATURE_VLAN_ALL_TX,
+					&dev->hw_features);
 	if (BNXT_SUPPORTS_TPA(bp))
-		dev->hw_features |= NETIF_F_GRO_HW;
-	dev->features |= dev->hw_features | NETIF_F_HIGHDMA;
-	if (dev->features & NETIF_F_GRO_HW)
-		dev->features &= ~NETIF_F_LRO;
+		netdev_feature_set_bit(NETIF_F_GRO_HW_BIT, &dev->hw_features);
+	netdev_feature_or(&dev->features, dev->features, dev->hw_features);
+	netdev_feature_set_bit(NETIF_F_HIGHDMA_BIT, &dev->features);
+	if (netdev_feature_test_bit(NETIF_F_GRO_HW_BIT, dev->features))
+		netdev_feature_clear_bit(NETIF_F_LRO_BIT, &dev->features);
 	dev->priv_flags |= IFF_UNICAST_FLT;
 
 #ifdef CONFIG_BNXT_SRIOV
@@ -13339,7 +13368,8 @@ static int bnxt_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	bnxt_fw_init_one_p3(bp);
 
-	if (dev->hw_features & BNXT_HW_FEATURE_VLAN_ALL_RX)
+	if (netdev_feature_test_bits(BNXT_HW_FEATURE_VLAN_ALL_RX,
+				     dev->hw_features))
 		bp->flags |= BNXT_FLAG_STRIP_VLAN;
 
 	rc = bnxt_init_int_mode(bp);
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_tc.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_tc.c
index e6a4a768b10b..ff88f5ccb782 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_tc.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_tc.c
@@ -2052,8 +2052,8 @@ int bnxt_init_tc(struct bnxt *bp)
 		goto destroy_decap_table;
 
 	tc_info->enabled = true;
-	bp->dev->hw_features |= NETIF_F_HW_TC;
-	bp->dev->features |= NETIF_F_HW_TC;
+	netdev_feature_set_bit(NETIF_F_HW_TC_BIT, &bp->dev->hw_features);
+	netdev_feature_set_bit(NETIF_F_HW_TC_BIT, &bp->dev->features);
 	bp->tc_info = tc_info;
 
 	/* init indirect block notifications */
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_vfr.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_vfr.c
index 9401936b74fa..d77beadf5921 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_vfr.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_vfr.c
@@ -468,11 +468,12 @@ static void bnxt_vf_rep_netdev_init(struct bnxt *bp, struct bnxt_vf_rep *vf_rep,
 	/* Just inherit all the featues of the parent PF as the VF-R
 	 * uses the RX/TX rings of the parent PF
 	 */
-	dev->hw_features = pf_dev->hw_features;
-	dev->gso_partial_features = pf_dev->gso_partial_features;
-	dev->vlan_features = pf_dev->vlan_features;
-	dev->hw_enc_features = pf_dev->hw_enc_features;
-	dev->features |= pf_dev->features;
+	netdev_feature_copy(&dev->hw_features, pf_dev->hw_features);
+	netdev_feature_copy(&dev->gso_partial_features,
+			    pf_dev->gso_partial_features);
+	netdev_feature_copy(&dev->vlan_features, pf_dev->vlan_features);
+	netdev_feature_copy(&dev->hw_enc_features, pf_dev->hw_enc_features);
+	netdev_feature_or(&dev->features, dev->features, pf_dev->features);
 	bnxt_vf_rep_eth_addr_gen(bp->pf.mac_addr, vf_rep->vf_idx,
 				 dev->perm_addr);
 	ether_addr_copy(dev->dev_addr, dev->perm_addr);
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.c
index c8083df5e0ab..a86f3c3db767 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.c
@@ -297,7 +297,8 @@ static int bnxt_xdp_set(struct bnxt *bp, struct bpf_prog *prog)
 		bnxt_get_max_rings(bp, &rx, &tx, true);
 		if (rx > 1) {
 			bp->flags &= ~BNXT_FLAG_NO_AGG_RINGS;
-			bp->dev->hw_features |= NETIF_F_LRO;
+			netdev_feature_set_bit(NETIF_F_LRO_BIT,
+					       &bp->dev->hw_features);
 		}
 	}
 	bp->tx_nr_rings_xdp = tx_xdp;
diff --git a/drivers/net/ethernet/broadcom/genet/bcmgenet.c b/drivers/net/ethernet/broadcom/genet/bcmgenet.c
index 6a8234bc9428..1b12d0ae8f1f 100644
--- a/drivers/net/ethernet/broadcom/genet/bcmgenet.c
+++ b/drivers/net/ethernet/broadcom/genet/bcmgenet.c
@@ -2285,7 +2285,8 @@ static unsigned int bcmgenet_desc_rx(struct bcmgenet_rx_ring *ring,
 
 		status = (struct status_64 *)skb->data;
 		dma_length_status = status->length_status;
-		if (dev->features & NETIF_F_RXCSUM) {
+		if (netdev_feature_test_bit(NETIF_F_RXCSUM_BIT,
+					    dev->features)) {
 			rx_csum = (__force __be16)(status->rx_csum & 0xffff);
 			skb->csum = (__force __wsum)ntohs(rx_csum);
 			skb->ip_summed = CHECKSUM_COMPLETE;
@@ -4005,10 +4006,11 @@ static int bcmgenet_probe(struct platform_device *pdev)
 	priv->msg_enable = netif_msg_init(-1, GENET_MSG_DEFAULT);
 
 	/* Set default features */
-	dev->features |= NETIF_F_SG | NETIF_F_HIGHDMA | NETIF_F_HW_CSUM |
-			 NETIF_F_RXCSUM;
-	dev->hw_features |= dev->features;
-	dev->vlan_features |= dev->features;
+	netdev_feature_set_bits(NETIF_F_SG | NETIF_F_HIGHDMA | NETIF_F_HW_CSUM |
+				NETIF_F_RXCSUM, &dev->features);
+	netdev_feature_or(&dev->hw_features, dev->hw_features, dev->features);
+	netdev_feature_or(&dev->vlan_features, dev->vlan_features,
+			  dev->features);
 
 	/* Request the WOL interrupt and advertise suspend if available */
 	priv->wol_irq_disabled = true;
diff --git a/drivers/net/ethernet/broadcom/tg3.c b/drivers/net/ethernet/broadcom/tg3.c
index df0d6a35f093..0406951a3a3e 100644
--- a/drivers/net/ethernet/broadcom/tg3.c
+++ b/drivers/net/ethernet/broadcom/tg3.c
@@ -6918,7 +6918,8 @@ static int tg3_rx(struct tg3_napi *tnapi, int budget)
 			tg3_hwclock_to_timestamp(tp, tstamp,
 						 skb_hwtstamps(skb));
 
-		if ((tp->dev->features & NETIF_F_RXCSUM) &&
+		if (netdev_feature_test_bit(NETIF_F_RXCSUM_BIT,
+					    tp->dev->features) &&
 		    (desc->type_flags & RXD_FLAG_TCPUDP_CSUM) &&
 		    (((desc->ip_tcp_csum & RXD_TCPCSUM_MASK)
 		      >> RXD_TCPCSUM_SHIFT) == 0xffff))
@@ -7860,6 +7861,7 @@ static int tg3_tso_bug(struct tg3 *tp, struct tg3_napi *tnapi,
 {
 	u32 frag_cnt_est = skb_shinfo(skb)->gso_segs * 3;
 	struct sk_buff *segs, *seg, *next;
+	netdev_features_t tmp;
 
 	/* Estimate the number of fragments in the worst case */
 	if (unlikely(tg3_tx_avail(tnapi) <= frag_cnt_est)) {
@@ -7877,8 +7879,9 @@ static int tg3_tso_bug(struct tg3 *tp, struct tg3_napi *tnapi,
 		netif_tx_wake_queue(txq);
 	}
 
-	segs = skb_gso_segment(skb, tp->dev->features &
-				    ~(NETIF_F_TSO | NETIF_F_TSO6));
+	netdev_feature_copy(&tmp, tp->dev->features);
+	netdev_feature_clear_bits(NETIF_F_TSO | NETIF_F_TSO6, &tmp);
+	segs = skb_gso_segment(skb, tmp);
 	if (IS_ERR(segs) || !segs)
 		goto tg3_tso_bug_end;
 
@@ -8280,7 +8283,7 @@ static void tg3_set_loopback(struct net_device *dev, netdev_features_t features)
 {
 	struct tg3 *tp = netdev_priv(dev);
 
-	if (features & NETIF_F_LOOPBACK) {
+	if (netdev_feature_test_bit(NETIF_F_LOOPBACK_BIT, features)) {
 		if (tp->mac_mode & MAC_MODE_PORT_INT_LPBACK)
 			return;
 
@@ -8308,14 +8311,16 @@ static void tg3_fix_features(struct net_device *dev,
 	struct tg3 *tp = netdev_priv(dev);
 
 	if (dev->mtu > ETH_DATA_LEN && tg3_flag(tp, 5780_CLASS))
-		*features &= ~NETIF_F_ALL_TSO;
+		netdev_feature_clear_bits(NETIF_F_ALL_TSO, features);
 }
 
 static int tg3_set_features(struct net_device *dev, netdev_features_t features)
 {
-	netdev_features_t changed = dev->features ^ features;
+	netdev_features_t changed;
 
-	if ((changed & NETIF_F_LOOPBACK) && netif_running(dev))
+	netdev_feature_xor(&changed, dev->features, features);
+	if (netdev_feature_test_bit(NETIF_F_LOOPBACK_BIT, changed) &&
+	    netif_running(dev))
 		tg3_set_loopback(dev, features);
 
 	return 0;
@@ -11648,7 +11653,7 @@ static int tg3_start(struct tg3 *tp, bool reset_phy, bool test_irq,
 	 * Reset loopback feature if it was turned on while the device was down
 	 * make sure that it's installed properly now.
 	 */
-	if (dev->features & NETIF_F_LOOPBACK)
+	if (netdev_feature_test_bit(NETIF_F_LOOPBACK_BIT, dev->features))
 		tg3_set_loopback(dev, dev->features);
 
 	return 0;
@@ -17558,7 +17563,9 @@ static int tg3_init_one(struct pci_dev *pdev,
 	u32 sndmbx, rcvmbx, intmbx;
 	char str[40];
 	u64 dma_mask, persist_dma_mask;
-	netdev_features_t features = 0;
+	netdev_features_t features;
+
+	netdev_feature_zero(&features);
 
 	err = pci_enable_device(pdev);
 	if (err) {
@@ -17701,7 +17708,7 @@ static int tg3_init_one(struct pci_dev *pdev,
 	if (dma_mask > DMA_BIT_MASK(32)) {
 		err = dma_set_mask(&pdev->dev, dma_mask);
 		if (!err) {
-			features |= NETIF_F_HIGHDMA;
+			netdev_feature_set_bit(NETIF_F_HIGHDMA_BIT, &features);
 			err = dma_set_coherent_mask(&pdev->dev,
 						    persist_dma_mask);
 			if (err < 0) {
@@ -17726,10 +17733,12 @@ static int tg3_init_one(struct pci_dev *pdev,
 	 * to hardware bugs.
 	 */
 	if (tg3_chip_rev_id(tp) != CHIPREV_ID_5700_B0) {
-		features |= NETIF_F_SG | NETIF_F_IP_CSUM | NETIF_F_RXCSUM;
+		netdev_feature_set_bits(NETIF_F_SG | NETIF_F_IP_CSUM |
+					NETIF_F_RXCSUM, &features);
 
 		if (tg3_flag(tp, 5755_PLUS))
-			features |= NETIF_F_IPV6_CSUM;
+			netdev_feature_set_bit(NETIF_F_IPV6_CSUM_BIT,
+					       &features);
 	}
 
 	/* TSO is on by default on chips that support hardware TSO.
@@ -17739,23 +17748,24 @@ static int tg3_init_one(struct pci_dev *pdev,
 	if ((tg3_flag(tp, HW_TSO_1) ||
 	     tg3_flag(tp, HW_TSO_2) ||
 	     tg3_flag(tp, HW_TSO_3)) &&
-	    (features & NETIF_F_IP_CSUM))
-		features |= NETIF_F_TSO;
+	    netdev_feature_test_bit(NETIF_F_IP_CSUM_BIT, features))
+		netdev_feature_set_bit(NETIF_F_TSO_BIT, &features);
 	if (tg3_flag(tp, HW_TSO_2) || tg3_flag(tp, HW_TSO_3)) {
-		if (features & NETIF_F_IPV6_CSUM)
-			features |= NETIF_F_TSO6;
+		if (netdev_feature_test_bit(NETIF_F_IPV6_CSUM_BIT, features))
+			netdev_feature_set_bit(NETIF_F_TSO6_BIT, &features);
 		if (tg3_flag(tp, HW_TSO_3) ||
 		    tg3_asic_rev(tp) == ASIC_REV_5761 ||
 		    (tg3_asic_rev(tp) == ASIC_REV_5784 &&
 		     tg3_chip_rev(tp) != CHIPREV_5784_AX) ||
 		    tg3_asic_rev(tp) == ASIC_REV_5785 ||
 		    tg3_asic_rev(tp) == ASIC_REV_57780)
-			features |= NETIF_F_TSO_ECN;
+			netdev_feature_set_bit(NETIF_F_TSO_ECN_BIT, &features);
 	}
 
-	dev->features |= features | NETIF_F_HW_VLAN_CTAG_TX |
-			 NETIF_F_HW_VLAN_CTAG_RX;
-	dev->vlan_features |= features;
+	netdev_feature_or(&dev->features, dev->features, features);
+	netdev_feature_set_bits(NETIF_F_HW_VLAN_CTAG_TX |
+			NETIF_F_HW_VLAN_CTAG_RX, &dev->features);
+	netdev_feature_or(&dev->vlan_features, dev->vlan_features, features);
 
 	/*
 	 * Add loopback capability only for a subset of devices that support
@@ -17765,9 +17775,9 @@ static int tg3_init_one(struct pci_dev *pdev,
 	if (tg3_asic_rev(tp) != ASIC_REV_5780 &&
 	    !tg3_flag(tp, CPMU_PRESENT))
 		/* Add the loopback capability */
-		features |= NETIF_F_LOOPBACK;
+		netdev_feature_set_bit(NETIF_F_LOOPBACK_BIT, &features);
 
-	dev->hw_features |= features;
+	netdev_feature_or(&dev->hw_features, dev->hw_features, features);
 	dev->priv_flags |= IFF_UNICAST_FLT;
 
 	/* MTU range: 60 - 9000 or 1500, depending on hardware */
@@ -17902,7 +17912,7 @@ static int tg3_init_one(struct pci_dev *pdev,
 	}
 
 	netdev_info(dev, "RXcsums[%d] LinkChgREG[%d] MIirq[%d] ASF[%d] TSOcap[%d]\n",
-		    (dev->features & NETIF_F_RXCSUM) != 0,
+		    netdev_feature_test_bit(NETIF_F_RXCSUM_BIT, dev->features),
 		    tg3_flag(tp, USE_LINKCHG_REG) != 0,
 		    (tp->phy_flags & TG3_PHYFLG_USE_MI_INTERRUPT) != 0,
 		    tg3_flag(tp, ENABLE_ASF) != 0,
-- 
2.33.0


^ permalink raw reply	[flat|nested] 181+ messages in thread

* [RFCv2 net-next 085/167] net: intel: use netdev feature helpers
  2021-09-29 15:50 [RFCv2 net-next 000/167] net: extend the netdev_features_t Jian Shen
                   ` (83 preceding siblings ...)
  2021-09-29 15:52 ` [RFCv2 net-next 084/167] net: broadcom: " Jian Shen
@ 2021-09-29 15:52 ` Jian Shen
  2021-09-29 15:52 ` [RFCv2 net-next 086/167] net: hisilicon: " Jian Shen
                   ` (85 subsequent siblings)
  170 siblings, 0 replies; 181+ messages in thread
From: Jian Shen @ 2021-09-29 15:52 UTC (permalink / raw)
  To: davem, kuba, andrew, hkallweit1; +Cc: netdev, linuxarm

Use netdev_feature_xxx helpers to replace the logical operation
for netdev features.

Signed-off-by: Jian Shen <shenjian15@huawei.com>
---
 drivers/net/ethernet/intel/e100.c             |  26 ++-
 drivers/net/ethernet/intel/e1000/e1000_main.c |  76 ++++---
 drivers/net/ethernet/intel/e1000e/netdev.c    | 123 ++++++-----
 drivers/net/ethernet/intel/fm10k/fm10k_main.c |   9 +-
 .../net/ethernet/intel/fm10k/fm10k_netdev.c   |  57 ++---
 drivers/net/ethernet/intel/fm10k/fm10k_pci.c  |  11 +-
 drivers/net/ethernet/intel/i40e/i40e_main.c   | 109 ++++++----
 drivers/net/ethernet/intel/i40e/i40e_txrx.c   |   9 +-
 drivers/net/ethernet/intel/iavf/iavf_main.c   | 134 +++++++-----
 drivers/net/ethernet/intel/iavf/iavf_txrx.c   |  12 +-
 drivers/net/ethernet/intel/ice/ice_main.c     | 158 ++++++++------
 drivers/net/ethernet/intel/ice/ice_txrx_lib.c |   9 +-
 drivers/net/ethernet/intel/igb/igb_ethtool.c  |   2 +-
 drivers/net/ethernet/intel/igb/igb_main.c     | 129 +++++++-----
 drivers/net/ethernet/intel/igbvf/netdev.c     |  71 ++++---
 drivers/net/ethernet/intel/igc/igc_ethtool.c  |   3 +-
 drivers/net/ethernet/intel/igc/igc_main.c     | 109 ++++++----
 drivers/net/ethernet/intel/ixgb/ixgb_main.c   |  43 ++--
 .../net/ethernet/intel/ixgbe/ixgbe_dcb_nl.c   |   3 +-
 .../net/ethernet/intel/ixgbe/ixgbe_ethtool.c  |   2 +-
 drivers/net/ethernet/intel/ixgbe/ixgbe_fcoe.c |   7 +-
 drivers/net/ethernet/intel/ixgbe/ixgbe_lib.c  |   3 +-
 drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 194 +++++++++++-------
 .../net/ethernet/intel/ixgbe/ixgbe_sriov.c    |   6 +-
 drivers/net/ethernet/intel/ixgbevf/ipsec.c    |   9 +-
 .../net/ethernet/intel/ixgbevf/ixgbevf_main.c |  90 ++++----
 26 files changed, 850 insertions(+), 554 deletions(-)

diff --git a/drivers/net/ethernet/intel/e100.c b/drivers/net/ethernet/intel/e100.c
index 373eb027b925..abec49f6b541 100644
--- a/drivers/net/ethernet/intel/e100.c
+++ b/drivers/net/ethernet/intel/e100.c
@@ -1114,7 +1114,8 @@ static int e100_configure(struct nic *nic, struct cb *cb, struct sk_buff *skb)
 		config->promiscuous_mode = 0x1;		/* 1=on, 0=off */
 	}
 
-	if (unlikely(netdev->features & NETIF_F_RXFCS))
+	if (unlikely(netdev_feature_test_bit(NETIF_F_RXFCS_BIT,
+					     netdev->features)))
 		config->rx_crc_transfer = 0x1;	/* 1=save, 0=discard */
 
 	if (nic->flags & multicast_all)
@@ -1141,7 +1142,7 @@ static int e100_configure(struct nic *nic, struct cb *cb, struct sk_buff *skb)
 		}
 	}
 
-	if (netdev->features & NETIF_F_RXALL) {
+	if (netdev_feature_test_bit(NETIF_F_RXALL_BIT, netdev->features)) {
 		config->rx_save_overruns = 0x1; /* 1=save, 0=discard */
 		config->rx_save_bad_frames = 0x1;       /* 1=save, 0=discard */
 		config->rx_discard_short_frames = 0x0;  /* 1=discard, 0=save */
@@ -1990,7 +1991,8 @@ static int e100_rx_indicate(struct nic *nic, struct rx *rx,
 	}
 
 	/* Get actual data size */
-	if (unlikely(dev->features & NETIF_F_RXFCS))
+	if (unlikely(netdev_feature_test_bit(NETIF_F_RXFCS_BIT,
+					     dev->features)))
 		fcs_pad = 4;
 	actual_size = le16_to_cpu(rfd->actual_size) & 0x3FFF;
 	if (unlikely(actual_size > RFD_BUF_LEN - sizeof(struct rfd)))
@@ -2021,7 +2023,8 @@ static int e100_rx_indicate(struct nic *nic, struct rx *rx,
 	/* If we are receiving all frames, then don't bother
 	 * checking for errors.
 	 */
-	if (unlikely(dev->features & NETIF_F_RXALL)) {
+	if (unlikely(netdev_feature_test_bit(NETIF_F_RXALL_BIT,
+					     dev->features))) {
 		if (actual_size > ETH_DATA_LEN + VLAN_ETH_HLEN + fcs_pad)
 			/* Received oversized frame, but keep it. */
 			nic->rx_over_length_errors++;
@@ -2792,12 +2795,14 @@ static int e100_set_features(struct net_device *netdev,
 			     netdev_features_t features)
 {
 	struct nic *nic = netdev_priv(netdev);
-	netdev_features_t changed = features ^ netdev->features;
+	netdev_features_t changed;
 
-	if (!(changed & (NETIF_F_RXFCS | NETIF_F_RXALL)))
+	netdev_feature_xor(&changed, features, netdev->features);
+
+	if (!netdev_feature_test_bits(NETIF_F_RXFCS | NETIF_F_RXALL, changed))
 		return 0;
 
-	netdev->features = features;
+	netdev_feature_copy(&netdev->features, features);
 	e100_exec_cb(nic, NULL, e100_configure);
 	return 1;
 }
@@ -2826,9 +2831,9 @@ static int e100_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	if (!(netdev = alloc_etherdev(sizeof(struct nic))))
 		return -ENOMEM;
 
-	netdev->hw_features |= NETIF_F_RXFCS;
+	netdev_feature_set_bit(NETIF_F_RXFCS_BIT, &netdev->hw_features);
 	netdev->priv_flags |= IFF_SUPP_NOFCS;
-	netdev->hw_features |= NETIF_F_RXALL;
+	netdev_feature_set_bit(NETIF_F_RXALL_BIT, &netdev->hw_features);
 
 	netdev->netdev_ops = &e100_netdev_ops;
 	netdev->ethtool_ops = &e100_ethtool_ops;
@@ -2885,7 +2890,8 @@ static int e100_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	/* D100 MAC doesn't allow rx of vlan packets with normal MTU */
 	if (nic->mac < mac_82558_D101_A4)
-		netdev->features |= NETIF_F_VLAN_CHALLENGED;
+		netdev_feature_set_bit(NETIF_F_VLAN_CHALLENGED_BIT,
+				       &netdev->features);
 
 	/* locks must be initialized before calling hw_reset */
 	spin_lock_init(&nic->cb_lock);
diff --git a/drivers/net/ethernet/intel/e1000/e1000_main.c b/drivers/net/ethernet/intel/e1000/e1000_main.c
index e333ca1e7395..34536befc4ea 100644
--- a/drivers/net/ethernet/intel/e1000/e1000_main.c
+++ b/drivers/net/ethernet/intel/e1000/e1000_main.c
@@ -793,26 +793,30 @@ static void e1000_fix_features(struct net_device *netdev,
 	/* Since there is no support for separate Rx/Tx vlan accel
 	 * enable/disable make sure Tx flag is always in same state as Rx.
 	 */
-	if (*features & NETIF_F_HW_VLAN_CTAG_RX)
-		*features |= NETIF_F_HW_VLAN_CTAG_TX;
+	if (netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_RX_BIT, *features))
+		netdev_feature_set_bit(NETIF_F_HW_VLAN_CTAG_TX_BIT, features);
 	else
-		*features &= ~NETIF_F_HW_VLAN_CTAG_TX;
+		netdev_feature_clear_bit(NETIF_F_HW_VLAN_CTAG_TX_BIT,
+					 features);
 }
 
 static int e1000_set_features(struct net_device *netdev,
 	netdev_features_t features)
 {
 	struct e1000_adapter *adapter = netdev_priv(netdev);
-	netdev_features_t changed = features ^ netdev->features;
+	netdev_features_t changed;
 
-	if (changed & NETIF_F_HW_VLAN_CTAG_RX)
+	netdev_feature_xor(&changed, features, netdev->features);
+
+	if (netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_RX_BIT, changed))
 		e1000_vlan_mode(netdev, features);
 
-	if (!(changed & (NETIF_F_RXCSUM | NETIF_F_RXALL)))
+	if (!netdev_feature_test_bits(NETIF_F_RXCSUM | NETIF_F_RXALL, changed))
 		return 0;
 
-	netdev->features = features;
-	adapter->rx_csum = !!(features & NETIF_F_RXCSUM);
+	netdev_feature_copy(&netdev->features, features);
+	adapter->rx_csum = netdev_feature_test_bit(NETIF_F_RXCSUM_BIT,
+						   features);
 
 	if (netif_running(netdev))
 		e1000_reinit_locked(adapter);
@@ -1033,32 +1037,40 @@ static int e1000_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	}
 
 	if (hw->mac_type >= e1000_82543) {
-		netdev->hw_features = NETIF_F_SG |
-				   NETIF_F_HW_CSUM |
-				   NETIF_F_HW_VLAN_CTAG_RX;
-		netdev->features = NETIF_F_HW_VLAN_CTAG_TX |
-				   NETIF_F_HW_VLAN_CTAG_FILTER;
+		netdev_feature_zero(&netdev->hw_features);
+		netdev_feature_set_bits(NETIF_F_SG |
+					NETIF_F_HW_CSUM |
+					NETIF_F_HW_VLAN_CTAG_RX,
+					&netdev->hw_features);
+		netdev_feature_zero(&netdev->features);
+		netdev_feature_set_bits(NETIF_F_HW_VLAN_CTAG_TX |
+					NETIF_F_HW_VLAN_CTAG_FILTER,
+					&netdev->features);
 	}
 
 	if ((hw->mac_type >= e1000_82544) &&
 	   (hw->mac_type != e1000_82547))
-		netdev->hw_features |= NETIF_F_TSO;
+		netdev_feature_set_bit(NETIF_F_TSO_BIT, &netdev->hw_features);
 
 	netdev->priv_flags |= IFF_SUPP_NOFCS;
 
-	netdev->features |= netdev->hw_features;
-	netdev->hw_features |= (NETIF_F_RXCSUM |
+	netdev_feature_or(&netdev->features, netdev->features,
+			  netdev->hw_features);
+	netdev_feature_set_bits(NETIF_F_RXCSUM |
 				NETIF_F_RXALL |
-				NETIF_F_RXFCS);
+				NETIF_F_RXFCS,
+				&netdev->hw_features);
 
 	if (pci_using_dac) {
-		netdev->features |= NETIF_F_HIGHDMA;
-		netdev->vlan_features |= NETIF_F_HIGHDMA;
+		netdev_feature_set_bit(NETIF_F_HIGHDMA_BIT, &netdev->features);
+		netdev_feature_set_bit(NETIF_F_HIGHDMA_BIT,
+				       &netdev->vlan_features);
 	}
 
-	netdev->vlan_features |= (NETIF_F_TSO |
-				  NETIF_F_HW_CSUM |
-				  NETIF_F_SG);
+	netdev_feature_set_bits(NETIF_F_TSO |
+				NETIF_F_HW_CSUM |
+				NETIF_F_SG,
+				&netdev->vlan_features);
 
 	/* Do not set IFF_UNICAST_FLT for VMWare's 82545EM */
 	if (hw->device_id != E1000_DEV_ID_82545EM_COPPER ||
@@ -1822,7 +1834,8 @@ static void e1000_setup_rctl(struct e1000_adapter *adapter)
 	}
 
 	/* This is useful for sniffing bad packets. */
-	if (adapter->netdev->features & NETIF_F_RXALL) {
+	if (netdev_feature_test_bit(NETIF_F_RXALL_BIT,
+				    adapter->netdev->features)) {
 		/* UPE and MPE will be handled by normal PROMISC logic
 		 * in e1000e_set_rx_mode
 		 */
@@ -4176,7 +4189,8 @@ static bool e1000_clean_jumbo_rx_irq(struct e1000_adapter *adapter,
 						    rx_desc->errors,
 						    length, mapped)) {
 				length--;
-			} else if (netdev->features & NETIF_F_RXALL) {
+			} else if (netdev_feature_test_bit(NETIF_F_RXALL_BIT,
+							   netdev->features)) {
 				goto process_skb;
 			} else {
 				/* an error means any chain goes out the window
@@ -4227,7 +4241,8 @@ static bool e1000_clean_jumbo_rx_irq(struct e1000_adapter *adapter,
 				if (length <= copybreak) {
 					u8 *vaddr;
 
-					if (likely(!(netdev->features & NETIF_F_RXFCS)))
+					if (likely(!netdev_feature_test_bit(NETIF_F_RXFCS_BIT,
+									    netdev->features)))
 						length -= 4;
 					skb = e1000_alloc_rx_skb(adapter,
 								 length);
@@ -4273,7 +4288,8 @@ static bool e1000_clean_jumbo_rx_irq(struct e1000_adapter *adapter,
 				  le16_to_cpu(rx_desc->csum), skb);
 
 		total_rx_bytes += (skb->len - 4); /* don't count FCS */
-		if (likely(!(netdev->features & NETIF_F_RXFCS)))
+		if (likely(!netdev_feature_test_bit(NETIF_F_RXFCS_BIT,
+						    netdev->features)))
 			pskb_trim(skb, skb->len - 4);
 		total_rx_packets++;
 
@@ -4428,7 +4444,8 @@ static bool e1000_clean_rx_irq(struct e1000_adapter *adapter,
 						    rx_desc->errors,
 						    length, data)) {
 				length--;
-			} else if (netdev->features & NETIF_F_RXALL) {
+			} else if (netdev_feature_test_bit(NETIF_F_RXALL_BIT,
+							   netdev->features)) {
 				goto process_skb;
 			} else {
 				dev_kfree_skb(skb);
@@ -4440,7 +4457,8 @@ static bool e1000_clean_rx_irq(struct e1000_adapter *adapter,
 		total_rx_bytes += (length - 4); /* don't count FCS */
 		total_rx_packets++;
 
-		if (likely(!(netdev->features & NETIF_F_RXFCS)))
+		if (likely(!netdev_feature_test_bit(NETIF_F_RXFCS_BIT,
+						    netdev->features)))
 			/* adjust length to remove Ethernet CRC, this must be
 			 * done after the TBI_ACCEPT workaround above
 			 */
@@ -4888,7 +4906,7 @@ static void __e1000_vlan_mode(struct e1000_adapter *adapter,
 	u32 ctrl;
 
 	ctrl = er32(CTRL);
-	if (features & NETIF_F_HW_VLAN_CTAG_RX) {
+	if (netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_RX_BIT, features)) {
 		/* enable VLAN tag insert/strip */
 		ctrl |= E1000_CTRL_VME;
 	} else {
diff --git a/drivers/net/ethernet/intel/e1000e/netdev.c b/drivers/net/ethernet/intel/e1000e/netdev.c
index 5dd183e0cb0f..a299f2789164 100644
--- a/drivers/net/ethernet/intel/e1000e/netdev.c
+++ b/drivers/net/ethernet/intel/e1000e/netdev.c
@@ -576,7 +576,8 @@ static void e1000_rx_checksum(struct e1000_adapter *adapter, u32 status_err,
 	skb_checksum_none_assert(skb);
 
 	/* Rx checksum disabled */
-	if (!(adapter->netdev->features & NETIF_F_RXCSUM))
+	if (!netdev_feature_test_bit(NETIF_F_RXCSUM_BIT,
+				     adapter->netdev->features))
 		return;
 
 	/* Ignore Checksum bit is set */
@@ -894,7 +895,7 @@ static void e1000_alloc_jumbo_rx_buffers(struct e1000_ring *rx_ring,
 static inline void e1000_rx_hash(struct net_device *netdev, __le32 rss,
 				 struct sk_buff *skb)
 {
-	if (netdev->features & NETIF_F_RXHASH)
+	if (netdev_feature_test_bit(NETIF_F_RXHASH_BIT, netdev->features))
 		skb_set_hash(skb, le32_to_cpu(rss), PKT_HASH_TYPE_L3);
 }
 
@@ -976,7 +977,8 @@ static bool e1000_clean_rx_irq(struct e1000_ring *rx_ring, int *work_done,
 		}
 
 		if (unlikely((staterr & E1000_RXDEXT_ERR_FRAME_ERR_MASK) &&
-			     !(netdev->features & NETIF_F_RXALL))) {
+			     !netdev_feature_test_bit(NETIF_F_RXALL_BIT,
+						      netdev->features))) {
 			/* recycle */
 			buffer_info->skb = skb;
 			goto next_desc;
@@ -988,7 +990,8 @@ static bool e1000_clean_rx_irq(struct e1000_ring *rx_ring, int *work_done,
 			 * but keep the FCS bytes out of the total_rx_bytes
 			 * counter
 			 */
-			if (netdev->features & NETIF_F_RXFCS)
+			if (netdev_feature_test_bit(NETIF_F_RXFCS_BIT,
+						    netdev->features))
 				total_rx_bytes -= 4;
 			else
 				length -= 4;
@@ -1362,7 +1365,8 @@ static bool e1000_clean_rx_irq_ps(struct e1000_ring *rx_ring, int *work_done,
 		}
 
 		if (unlikely((staterr & E1000_RXDEXT_ERR_FRAME_ERR_MASK) &&
-			     !(netdev->features & NETIF_F_RXALL))) {
+			     !netdev_feature_test_bit(NETIF_F_RXALL_BIT,
+						      netdev->features))) {
 			dev_kfree_skb_irq(skb);
 			goto next_desc;
 		}
@@ -1413,7 +1417,8 @@ static bool e1000_clean_rx_irq_ps(struct e1000_ring *rx_ring, int *work_done,
 
 				/* remove the CRC */
 				if (!(adapter->flags2 & FLAG2_CRC_STRIPPING)) {
-					if (!(netdev->features & NETIF_F_RXFCS))
+					if (!netdev_feature_test_bit(NETIF_F_RXFCS_BIT,
+								     netdev->features))
 						l1 -= 4;
 				}
 
@@ -1442,7 +1447,8 @@ static bool e1000_clean_rx_irq_ps(struct e1000_ring *rx_ring, int *work_done,
 		 * this whole operation can get a little cpu intensive
 		 */
 		if (!(adapter->flags2 & FLAG2_CRC_STRIPPING)) {
-			if (!(netdev->features & NETIF_F_RXFCS))
+			if (!netdev_feature_test_bit(NETIF_F_RXFCS_BIT,
+						     netdev->features))
 				pskb_trim(skb, skb->len - 4);
 		}
 
@@ -1557,7 +1563,8 @@ static bool e1000_clean_jumbo_rx_irq(struct e1000_ring *rx_ring, int *work_done,
 		/* errors is only valid for DD + EOP descriptors */
 		if (unlikely((staterr & E1000_RXD_STAT_EOP) &&
 			     ((staterr & E1000_RXDEXT_ERR_FRAME_ERR_MASK) &&
-			      !(netdev->features & NETIF_F_RXALL)))) {
+			      !netdev_feature_test_bit(NETIF_F_RXALL_BIT,
+						       netdev->features)))) {
 			/* recycle both page and skb */
 			buffer_info->skb = skb;
 			/* an error means any chain goes out the window too */
@@ -3165,7 +3172,8 @@ static void e1000_setup_rctl(struct e1000_adapter *adapter)
 	}
 
 	/* This is useful for sniffing bad packets. */
-	if (adapter->netdev->features & NETIF_F_RXALL) {
+	if (netdev_feature_test_bit(NETIF_F_RXALL_BIT,
+				    adapter->netdev->features)) {
 		/* UPE and MPE will be handled by normal PROMISC logic
 		 * in e1000e_set_rx_mode
 		 */
@@ -3270,7 +3278,8 @@ static void e1000_configure_rx(struct e1000_adapter *adapter)
 
 	/* Enable Receive Checksum Offload for TCP and UDP */
 	rxcsum = er32(RXCSUM);
-	if (adapter->netdev->features & NETIF_F_RXCSUM)
+	if (netdev_feature_test_bit(NETIF_F_RXCSUM_BIT,
+				    adapter->netdev->features))
 		rxcsum |= E1000_RXCSUM_TUOFL;
 	else
 		rxcsum &= ~E1000_RXCSUM_TUOFL;
@@ -3451,7 +3460,8 @@ static void e1000e_set_rx_mode(struct net_device *netdev)
 
 	ew32(RCTL, rctl);
 
-	if (netdev->features & NETIF_F_HW_VLAN_CTAG_RX)
+	if (netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_RX_BIT,
+				    netdev->features))
 		e1000e_vlan_strip_enable(adapter);
 	else
 		e1000e_vlan_strip_disable(adapter);
@@ -3765,7 +3775,8 @@ static void e1000_configure(struct e1000_adapter *adapter)
 
 	e1000_configure_tx(adapter);
 
-	if (adapter->netdev->features & NETIF_F_RXHASH)
+	if (netdev_feature_test_bit(NETIF_F_RXHASH_BIT,
+				    adapter->netdev->features))
 		e1000e_setup_rss_hash(adapter);
 	e1000_setup_rctl(adapter);
 	e1000_configure_rx(adapter);
@@ -5304,20 +5315,26 @@ static void e1000_watchdog_task(struct work_struct *work)
 				case SPEED_10:
 				case SPEED_100:
 					e_info("10/100 speed: disabling TSO\n");
-					netdev->features &= ~NETIF_F_TSO;
-					netdev->features &= ~NETIF_F_TSO6;
+					netdev_feature_clear_bit(NETIF_F_TSO_BIT,
+								 &netdev->features);
+					netdev_feature_clear_bit(NETIF_F_TSO6_BIT,
+								 &netdev->features);
 					break;
 				case SPEED_1000:
-					netdev->features |= NETIF_F_TSO;
-					netdev->features |= NETIF_F_TSO6;
+					netdev_feature_clear_bit(NETIF_F_TSO_BIT,
+								 &netdev->features);
+					netdev_feature_clear_bit(NETIF_F_TSO6_BIT,
+								 &netdev->features);
 					break;
 				default:
 					/* oops */
 					break;
 				}
 				if (hw->mac.type == e1000_pch_spt) {
-					netdev->features &= ~NETIF_F_TSO;
-					netdev->features &= ~NETIF_F_TSO6;
+					netdev_feature_clear_bit(NETIF_F_TSO_BIT,
+								 &netdev->features);
+					netdev_feature_clear_bit(NETIF_F_TSO6_BIT,
+								 &netdev->features);
 				}
 			}
 
@@ -7295,33 +7312,38 @@ static void e1000_fix_features(struct net_device *netdev,
 
 	/* Jumbo frame workaround on 82579 and newer requires CRC be stripped */
 	if ((hw->mac.type >= e1000_pch2lan) && (netdev->mtu > ETH_DATA_LEN))
-		*features &= ~NETIF_F_RXFCS;
+		netdev_feature_clear_bit(NETIF_F_RXFCS_BIT, features);
 
 	/* Since there is no support for separate Rx/Tx vlan accel
 	 * enable/disable make sure Tx flag is always in same state as Rx.
 	 */
-	if (*features & NETIF_F_HW_VLAN_CTAG_RX)
-		*features |= NETIF_F_HW_VLAN_CTAG_TX;
+	if (netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_RX_BIT, *features))
+		netdev_feature_set_bit(NETIF_F_HW_VLAN_CTAG_TX_BIT, features);
 	else
-		*features &= ~NETIF_F_HW_VLAN_CTAG_TX;
+		netdev_feature_clear_bit(NETIF_F_HW_VLAN_CTAG_TX_BIT,
+					 features);
 }
 
 static int e1000_set_features(struct net_device *netdev,
 			      netdev_features_t features)
 {
 	struct e1000_adapter *adapter = netdev_priv(netdev);
-	netdev_features_t changed = features ^ netdev->features;
+	netdev_features_t changed;
 
-	if (changed & (NETIF_F_TSO | NETIF_F_TSO6))
+	netdev_feature_xor(&changed, features, netdev->features);
+
+	if (netdev_feature_test_bits(NETIF_F_TSO | NETIF_F_TSO6, changed))
 		adapter->flags |= FLAG_TSO_FORCE;
 
-	if (!(changed & (NETIF_F_HW_VLAN_CTAG_RX | NETIF_F_HW_VLAN_CTAG_TX |
-			 NETIF_F_RXCSUM | NETIF_F_RXHASH | NETIF_F_RXFCS |
-			 NETIF_F_RXALL)))
+	if (!netdev_feature_test_bits(NETIF_F_HW_VLAN_CTAG_RX |
+				      NETIF_F_HW_VLAN_CTAG_TX | NETIF_F_RXCSUM |
+				      NETIF_F_RXHASH | NETIF_F_RXFCS |
+				      NETIF_F_RXALL | NETIF_F_HW_VLAN_CTAG_RX,
+				      changed))
 		return 0;
 
-	if (changed & NETIF_F_RXFCS) {
-		if (features & NETIF_F_RXFCS) {
+	if (netdev_feature_test_bit(NETIF_F_RXFCS_BIT, changed)) {
+		if (netdev_feature_test_bit(NETIF_F_RXFCS_BIT, features)) {
 			adapter->flags2 &= ~FLAG2_CRC_STRIPPING;
 		} else {
 			/* We need to take it back to defaults, which might mean
@@ -7334,7 +7356,7 @@ static int e1000_set_features(struct net_device *netdev,
 		}
 	}
 
-	netdev->features = features;
+	netdev_feature_copy(&netdev->features, features);
 
 	if (netif_running(netdev))
 		e1000e_reinit_locked(adapter);
@@ -7524,34 +7546,39 @@ static int e1000_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 			 "PHY reset is blocked due to SOL/IDER session.\n");
 
 	/* Set initial default active device features */
-	netdev->features = (NETIF_F_SG |
-			    NETIF_F_HW_VLAN_CTAG_RX |
-			    NETIF_F_HW_VLAN_CTAG_TX |
-			    NETIF_F_TSO |
-			    NETIF_F_TSO6 |
-			    NETIF_F_RXHASH |
-			    NETIF_F_RXCSUM |
-			    NETIF_F_HW_CSUM);
+	netdev_feature_zero(&netdev->features);
+	netdev_feature_set_bits(NETIF_F_SG |
+				NETIF_F_HW_VLAN_CTAG_RX |
+				NETIF_F_HW_VLAN_CTAG_TX |
+				NETIF_F_TSO |
+				NETIF_F_TSO6 |
+				NETIF_F_RXHASH |
+				NETIF_F_RXCSUM |
+				NETIF_F_HW_CSUM,
+				&netdev->features);
 
 	/* Set user-changeable features (subset of all device features) */
-	netdev->hw_features = netdev->features;
-	netdev->hw_features |= NETIF_F_RXFCS;
+	netdev_feature_copy(&netdev->hw_features, netdev->features);
+	netdev_feature_set_bit(NETIF_F_RXFCS_BIT, &netdev->hw_features);
 	netdev->priv_flags |= IFF_SUPP_NOFCS;
-	netdev->hw_features |= NETIF_F_RXALL;
+	netdev_feature_set_bit(NETIF_F_RXALL_BIT, &netdev->hw_features);
 
 	if (adapter->flags & FLAG_HAS_HW_VLAN_FILTER)
-		netdev->features |= NETIF_F_HW_VLAN_CTAG_FILTER;
+		netdev_feature_set_bit(NETIF_F_HW_VLAN_CTAG_FILTER_BIT,
+				       &netdev->features);
 
-	netdev->vlan_features |= (NETIF_F_SG |
-				  NETIF_F_TSO |
-				  NETIF_F_TSO6 |
-				  NETIF_F_HW_CSUM);
+	netdev_feature_set_bits(NETIF_F_SG |
+				NETIF_F_TSO |
+				NETIF_F_TSO6 |
+				NETIF_F_HW_CSUM,
+				&netdev->vlan_features);
 
 	netdev->priv_flags |= IFF_UNICAST_FLT;
 
 	if (pci_using_dac) {
-		netdev->features |= NETIF_F_HIGHDMA;
-		netdev->vlan_features |= NETIF_F_HIGHDMA;
+		netdev_feature_set_bit(NETIF_F_HIGHDMA_BIT, &netdev->features);
+		netdev_feature_set_bit(NETIF_F_HIGHDMA_BIT,
+				       &netdev->vlan_features);
 	}
 
 	/* MTU range: 68 - max_hw_frame_size */
diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_main.c b/drivers/net/ethernet/intel/fm10k/fm10k_main.c
index 3362f26d7f99..ed2871d62804 100644
--- a/drivers/net/ethernet/intel/fm10k/fm10k_main.c
+++ b/drivers/net/ethernet/intel/fm10k/fm10k_main.c
@@ -352,7 +352,8 @@ static inline void fm10k_rx_checksum(struct fm10k_ring *ring,
 	skb_checksum_none_assert(skb);
 
 	/* Rx checksum disabled via ethtool */
-	if (!(ring->netdev->features & NETIF_F_RXCSUM))
+	if (!netdev_feature_test_bit(NETIF_F_RXCSUM_BIT,
+				     ring->netdev->features))
 		return;
 
 	/* TCP/UDP checksum error bit is set */
@@ -388,7 +389,8 @@ static inline void fm10k_rx_hash(struct fm10k_ring *ring,
 {
 	u16 rss_type;
 
-	if (!(ring->netdev->features & NETIF_F_RXHASH))
+	if (!netdev_feature_test_bit(NETIF_F_RXHASH_BIT,
+				     ring->netdev->features))
 		return;
 
 	rss_type = le16_to_cpu(rx_desc->w.pkt_info) & FM10K_RXD_RSSTYPE_MASK;
@@ -772,7 +774,8 @@ static int fm10k_tso(struct fm10k_ring *tx_ring,
 	return 1;
 
 err_vxlan:
-	tx_ring->netdev->features &= ~NETIF_F_GSO_UDP_TUNNEL;
+	netdev_feature_clear_bit(NETIF_F_GSO_UDP_TUNNEL_BIT,
+				 &tx_ring->netdev->features);
 	if (net_ratelimit())
 		netdev_err(tx_ring->netdev,
 			   "TSO requested for unsupported tunnel, disabling offload\n");
diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_netdev.c b/drivers/net/ethernet/intel/fm10k/fm10k_netdev.c
index a311bcdfbff2..e3747f310c2e 100644
--- a/drivers/net/ethernet/intel/fm10k/fm10k_netdev.c
+++ b/drivers/net/ethernet/intel/fm10k/fm10k_netdev.c
@@ -1510,7 +1510,8 @@ static void fm10k_features_check(struct sk_buff *skb, struct net_device *dev,
 	if (!skb->encapsulation || fm10k_tx_encap_offload(skb))
 		return;
 
-	*features &= ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
+	netdev_feature_clear_bits(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK,
+				  features);
 }
 
 static const struct net_device_ops fm10k_netdev_ops = {
@@ -1556,50 +1557,56 @@ struct net_device *fm10k_alloc_netdev(const struct fm10k_info *info)
 	interface->msg_enable = BIT(DEFAULT_DEBUG_LEVEL_SHIFT) - 1;
 
 	/* configure default features */
-	dev->features |= NETIF_F_IP_CSUM |
-			 NETIF_F_IPV6_CSUM |
-			 NETIF_F_SG |
-			 NETIF_F_TSO |
-			 NETIF_F_TSO6 |
-			 NETIF_F_TSO_ECN |
-			 NETIF_F_RXHASH |
-			 NETIF_F_RXCSUM;
+	netdev_feature_set_bits(NETIF_F_IP_CSUM |
+				NETIF_F_IPV6_CSUM |
+				NETIF_F_SG |
+				NETIF_F_TSO |
+				NETIF_F_TSO6 |
+				NETIF_F_TSO_ECN |
+				NETIF_F_RXHASH |
+				NETIF_F_RXCSUM,
+				&dev->features);
 
 	/* Only the PF can support VXLAN and NVGRE tunnel offloads */
 	if (info->mac == fm10k_mac_pf) {
-		dev->hw_enc_features = NETIF_F_IP_CSUM |
-				       NETIF_F_TSO |
-				       NETIF_F_TSO6 |
-				       NETIF_F_TSO_ECN |
-				       NETIF_F_GSO_UDP_TUNNEL |
-				       NETIF_F_IPV6_CSUM |
-				       NETIF_F_SG;
-
-		dev->features |= NETIF_F_GSO_UDP_TUNNEL;
+		netdev_feature_set_bits(NETIF_F_IP_CSUM |
+					NETIF_F_TSO |
+					NETIF_F_TSO6 |
+					NETIF_F_TSO_ECN |
+					NETIF_F_GSO_UDP_TUNNEL |
+					NETIF_F_IPV6_CSUM |
+					NETIF_F_SG,
+					&dev->hw_enc_features);
+
+		netdev_feature_set_bit(NETIF_F_GSO_UDP_TUNNEL_BIT,
+				       &dev->features);
 
 		dev->udp_tunnel_nic_info = &fm10k_udp_tunnels;
 	}
 
 	/* all features defined to this point should be changeable */
-	hw_features = dev->features;
+	netdev_feature_copy(&hw_features, dev->features);
 
 	/* allow user to enable L2 forwarding acceleration */
-	hw_features |= NETIF_F_HW_L2FW_DOFFLOAD;
+	netdev_feature_set_bit(NETIF_F_HW_L2FW_DOFFLOAD_BIT, &hw_features);
 
 	/* configure VLAN features */
-	dev->vlan_features |= dev->features;
+	netdev_feature_or(&dev->vlan_features, dev->vlan_features,
+			  dev->features);
 
 	/* we want to leave these both on as we cannot disable VLAN tag
 	 * insertion or stripping on the hardware since it is contained
 	 * in the FTAG and not in the frame itself.
 	 */
-	dev->features |= NETIF_F_HW_VLAN_CTAG_TX |
-			 NETIF_F_HW_VLAN_CTAG_RX |
-			 NETIF_F_HW_VLAN_CTAG_FILTER;
+	netdev_feature_set_bits(NETIF_F_HW_VLAN_CTAG_TX |
+				NETIF_F_HW_VLAN_CTAG_RX |
+				NETIF_F_HW_VLAN_CTAG_FILTER,
+				&dev->features);
 
 	dev->priv_flags |= IFF_UNICAST_FLT;
 
-	dev->hw_features |= hw_features;
+	netdev_feature_or(&dev->hw_features, dev->hw_features,
+			  hw_features);
 
 	/* MTU range: 68 - 15342 */
 	dev->min_mtu = ETH_MIN_MTU;
diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_pci.c b/drivers/net/ethernet/intel/fm10k/fm10k_pci.c
index adfa2768f024..48c1206246d0 100644
--- a/drivers/net/ethernet/intel/fm10k/fm10k_pci.c
+++ b/drivers/net/ethernet/intel/fm10k/fm10k_pci.c
@@ -305,9 +305,11 @@ static int fm10k_handle_reset(struct fm10k_intfc *interface)
 		}
 
 		if (hw->mac.vlan_override)
-			netdev->features &= ~NETIF_F_HW_VLAN_CTAG_RX;
+			netdev_feature_clear_bit(NETIF_F_HW_VLAN_CTAG_RX_BIT,
+						 &netdev->features);
 		else
-			netdev->features |= NETIF_F_HW_VLAN_CTAG_RX;
+			netdev_feature_set_bit(NETIF_F_HW_VLAN_CTAG_RX_BIT,
+					       &netdev->features);
 	}
 
 	err = netif_running(netdev) ? fm10k_open(netdev) : 0;
@@ -2010,8 +2012,9 @@ static int fm10k_sw_init(struct fm10k_intfc *interface,
 
 	/* update netdev with DMA restrictions */
 	if (dma_get_mask(&pdev->dev) > DMA_BIT_MASK(32)) {
-		netdev->features |= NETIF_F_HIGHDMA;
-		netdev->vlan_features |= NETIF_F_HIGHDMA;
+		netdev_feature_set_bit(NETIF_F_HIGHDMA_BIT, &netdev->features);
+		netdev_feature_set_bit(NETIF_F_HIGHDMA_BIT,
+				       &netdev->vlan_features);
 	}
 
 	/* reset and initialize the hardware so it is in a known state */
diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index 5246c6abbd7d..5dddd63bde74 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -2962,7 +2962,8 @@ static void i40e_restore_vlan(struct i40e_vsi *vsi)
 	if (!vsi->netdev)
 		return;
 
-	if (vsi->netdev->features & NETIF_F_HW_VLAN_CTAG_RX)
+	if (netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_RX_BIT,
+				    vsi->netdev->features))
 		i40e_vlan_stripping_enable(vsi);
 	else
 		i40e_vlan_stripping_disable(vsi);
@@ -12541,7 +12542,7 @@ bool i40e_set_ntuple(struct i40e_pf *pf, netdev_features_t features)
 	/* Check if Flow Director n-tuple support was enabled or disabled.  If
 	 * the state changed, we need to reset.
 	 */
-	if (features & NETIF_F_NTUPLE) {
+	if (netdev_feature_test_bit(NETIF_F_NTUPLE_BIT, features)) {
 		/* Enable filters and mark for reset */
 		if (!(pf->flags & I40E_FLAG_FD_SB_ENABLED))
 			need_reset = true;
@@ -12610,24 +12611,27 @@ static int i40e_set_features(struct net_device *netdev,
 	struct i40e_pf *pf = vsi->back;
 	bool need_reset;
 
-	if (features & NETIF_F_RXHASH && !(netdev->features & NETIF_F_RXHASH))
+	if (netdev_feature_test_bit(NETIF_F_RXHASH_BIT, features) &&
+	    !netdev_feature_test_bit(NETIF_F_RXHASH_BIT, netdev->features))
 		i40e_pf_config_rss(pf);
-	else if (!(features & NETIF_F_RXHASH) &&
-		 netdev->features & NETIF_F_RXHASH)
+	else if (!netdev_feature_test_bit(NETIF_F_RXHASH_BIT, features) &&
+		 netdev_feature_test_bit(NETIF_F_RXHASH_BIT, netdev->features))
 		i40e_clear_rss_lut(vsi);
 
-	if (features & NETIF_F_HW_VLAN_CTAG_RX)
+	if (netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_RX_BIT, features))
 		i40e_vlan_stripping_enable(vsi);
 	else
 		i40e_vlan_stripping_disable(vsi);
 
-	if (!(features & NETIF_F_HW_TC) && pf->num_cloud_filters) {
+	if (!netdev_feature_test_bit(NETIF_F_HW_TC_BIT, features) &&
+	    pf->num_cloud_filters) {
 		dev_err(&pf->pdev->dev,
 			"Offloaded tc filters active, can't turn hw_tc_offload off");
 		return -EINVAL;
 	}
 
-	if (!(features & NETIF_F_HW_L2FW_DOFFLOAD) && vsi->macvlan_cnt)
+	if (!netdev_feature_test_bit(NETIF_F_HW_L2FW_DOFFLOAD_BIT, features) &&
+	    vsi->macvlan_cnt)
 		i40e_del_all_macvlans(vsi);
 
 	need_reset = i40e_set_ntuple(pf, features);
@@ -12889,7 +12893,7 @@ static void i40e_features_check(struct sk_buff *skb, struct net_device *dev,
 	 * 64 bytes.  If it is then we need to drop support for GSO.
 	 */
 	if (skb_is_gso(skb) && (skb_shinfo(skb)->gso_size < 64))
-		*features &= ~NETIF_F_GSO_MASK;
+		netdev_feature_clear_bits(NETIF_F_GSO_MASK, features);
 
 	/* MACLEN can support at most 63 words */
 	len = skb_network_header(skb) - skb->data;
@@ -12921,7 +12925,8 @@ static void i40e_features_check(struct sk_buff *skb, struct net_device *dev,
 
 	return;
 out_err:
-	*features &= ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
+	netdev_feature_clear_bits(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK,
+				  features);
 }
 
 /**
@@ -13324,53 +13329,71 @@ static int i40e_config_netdev(struct i40e_vsi *vsi)
 	np = netdev_priv(netdev);
 	np->vsi = vsi;
 
-	hw_enc_features = NETIF_F_SG			|
-			  NETIF_F_IP_CSUM		|
-			  NETIF_F_IPV6_CSUM		|
-			  NETIF_F_HIGHDMA		|
-			  NETIF_F_SOFT_FEATURES		|
-			  NETIF_F_TSO			|
-			  NETIF_F_TSO_ECN		|
-			  NETIF_F_TSO6			|
-			  NETIF_F_GSO_GRE		|
-			  NETIF_F_GSO_GRE_CSUM		|
-			  NETIF_F_GSO_PARTIAL		|
-			  NETIF_F_GSO_IPXIP4		|
-			  NETIF_F_GSO_IPXIP6		|
-			  NETIF_F_GSO_UDP_TUNNEL	|
-			  NETIF_F_GSO_UDP_TUNNEL_CSUM	|
-			  NETIF_F_GSO_UDP_L4		|
-			  NETIF_F_SCTP_CRC		|
-			  NETIF_F_RXHASH		|
-			  NETIF_F_RXCSUM		|
-			  0;
+	netdev_feature_zero(&hw_enc_features);
+	netdev_feature_set_bits(NETIF_F_SG_BIT			|
+				NETIF_F_IP_CSUM_BIT		|
+				NETIF_F_IPV6_CSUM_BIT		|
+				NETIF_F_HIGHDMA_BIT		|
+				NETIF_F_GSO_BIT			|
+				NETIF_F_GRO_BIT			|
+				NETIF_F_TSO_BIT			|
+				NETIF_F_TSO_ECN_BIT		|
+				NETIF_F_TSO6_BIT		|
+				NETIF_F_GSO_GRE_BIT		|
+				NETIF_F_GSO_GRE_CSUM_BIT	|
+				NETIF_F_GSO_PARTIAL_BIT		|
+				NETIF_F_GSO_IPXIP4_BIT		|
+				NETIF_F_GSO_IPXIP6_BIT		|
+				NETIF_F_GSO_UDP_TUNNEL_BIT	|
+				NETIF_F_GSO_UDP_TUNNEL_CSUM_BIT	|
+				NETIF_F_GSO_UDP_L4_BIT		|
+				NETIF_F_SCTP_CRC_BIT		|
+				NETIF_F_RXHASH_BIT		|
+				NETIF_F_RXCSUM_BIT,
+				&hw_enc_features);
 
 	if (!(pf->hw_features & I40E_HW_OUTER_UDP_CSUM_CAPABLE))
-		netdev->gso_partial_features |= NETIF_F_GSO_UDP_TUNNEL_CSUM;
+		netdev_feature_set_bit(NETIF_F_GSO_UDP_TUNNEL_CSUM_BIT,
+				       &netdev->gso_partial_features);
 
 	netdev->udp_tunnel_nic_info = &pf->udp_tunnel_nic;
 
-	netdev->gso_partial_features |= NETIF_F_GSO_GRE_CSUM;
+	netdev_feature_set_bit(NETIF_F_GSO_GRE_CSUM_BIT,
+			       &netdev->gso_partial_features);
 
-	netdev->hw_enc_features |= hw_enc_features;
+	netdev_feature_or(&netdev->hw_enc_features, netdev->hw_enc_features,
+			  hw_enc_features);
 
 	/* record features VLANs can make use of */
-	netdev->vlan_features |= hw_enc_features | NETIF_F_TSO_MANGLEID;
+	netdev_feature_or(&netdev->vlan_features, netdev->vlan_features,
+			  hw_enc_features);
+	netdev_feature_set_bit(NETIF_F_TSO_MANGLEID_BIT,
+			       &netdev->vlan_features);
 
 	/* enable macvlan offloads */
-	netdev->hw_features |= NETIF_F_HW_L2FW_DOFFLOAD;
+	netdev_feature_set_bit(NETIF_F_HW_L2FW_DOFFLOAD_BIT,
+			       &netdev->hw_features);
 
-	hw_features = hw_enc_features		|
-		      NETIF_F_HW_VLAN_CTAG_TX	|
-		      NETIF_F_HW_VLAN_CTAG_RX;
+	netdev_feature_copy(&hw_features, hw_enc_features);
+	netdev_feature_set_bit(NETIF_F_HW_VLAN_CTAG_TX_BIT,
+			       &hw_features);
+	netdev_feature_set_bit(NETIF_F_HW_VLAN_CTAG_RX_BIT,
+			       &hw_features);
 
-	if (!(pf->flags & I40E_FLAG_MFP_ENABLED))
-		hw_features |= NETIF_F_NTUPLE | NETIF_F_HW_TC;
+	if (!(pf->flags & I40E_FLAG_MFP_ENABLED)) {
+		netdev_feature_set_bit(NETIF_F_NTUPLE_BIT, &hw_features);
+		netdev_feature_set_bit(NETIF_F_HW_TC_BIT, &hw_features);
+	}
 
-	netdev->hw_features |= hw_features;
+	netdev_feature_or(&netdev->hw_features, netdev->hw_features,
+			  hw_features);
 
-	netdev->features |= hw_features | NETIF_F_HW_VLAN_CTAG_FILTER;
-	netdev->hw_enc_features |= NETIF_F_TSO_MANGLEID;
+	netdev_feature_or(&netdev->features, netdev->features,
+			  hw_features);
+	netdev_feature_set_bit(NETIF_F_HW_VLAN_CTAG_FILTER_BIT,
+			       &netdev->features);
+	netdev_feature_set_bit(NETIF_F_TSO_MANGLEID_BIT,
+			       &netdev->hw_enc_features);
 
 	if (vsi->type == I40E_VSI_MAIN) {
 		SET_NETDEV_DEV(netdev, &pf->pdev->dev);
diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
index 10a83e5385c7..0e94bd8bcd59 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
@@ -1793,7 +1793,8 @@ static inline void i40e_rx_checksum(struct i40e_vsi *vsi,
 	skb_checksum_none_assert(skb);
 
 	/* Rx csum enabled and ip headers found? */
-	if (!(vsi->netdev->features & NETIF_F_RXCSUM))
+	if (!netdev_feature_test_bit(NETIF_F_RXCSUM_BIT,
+				     vsi->netdev->features))
 		return;
 
 	/* did the hardware decode the packet and checksum? */
@@ -1895,7 +1896,8 @@ static inline void i40e_rx_hash(struct i40e_ring *ring,
 		cpu_to_le64((u64)I40E_RX_DESC_FLTSTAT_RSS_HASH <<
 			    I40E_RX_DESC_STATUS_FLTSTAT_SHIFT);
 
-	if (!(ring->netdev->features & NETIF_F_RXHASH))
+	if (!netdev_feature_test_bit(NETIF_F_RXHASH_BIT,
+				     ring->netdev->features))
 		return;
 
 	if ((rx_desc->wb.qword1.status_error_len & rss_mask) == rss_mask) {
@@ -2944,7 +2946,8 @@ static inline int i40e_tx_prepare_vlan_flags(struct sk_buff *skb,
 	u32  tx_flags = 0;
 
 	if (protocol == htons(ETH_P_8021Q) &&
-	    !(tx_ring->netdev->features & NETIF_F_HW_VLAN_CTAG_TX)) {
+	    !netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_TX_BIT,
+				     tx_ring->netdev->features)) {
 		/* When HW VLAN acceleration is turned off by the user the
 		 * stack sets the protocol to 8021q so that the driver
 		 * can take any steps required to support the SW only
diff --git a/drivers/net/ethernet/intel/iavf/iavf_main.c b/drivers/net/ethernet/intel/iavf/iavf_main.c
index 9a086211af4a..9f6b6cabdca1 100644
--- a/drivers/net/ethernet/intel/iavf/iavf_main.c
+++ b/drivers/net/ethernet/intel/iavf/iavf_main.c
@@ -1890,7 +1890,7 @@ static int iavf_init_get_resources(struct iavf_adapter *adapter)
 				 err);
 	}
 	dev_info(&pdev->dev, "MAC address: %pM\n", adapter->hw.mac.addr);
-	if (netdev->features & NETIF_F_GRO)
+	if (netdev_feature_test_bit(NETIF_F_GRO_BIT, netdev->features))
 		dev_info(&pdev->dev, "GRO is enabled\n");
 
 	adapter->state = __IAVF_DOWN;
@@ -3354,15 +3354,20 @@ static int iavf_set_features(struct net_device *netdev,
 			     netdev_features_t features)
 {
 	struct iavf_adapter *adapter = netdev_priv(netdev);
+	netdev_features_t changed;
 
+	netdev_feature_xor(&changed, netdev->features, features);
 	/* Don't allow changing VLAN_RX flag when adapter is not capable
 	 * of VLAN offload
 	 */
 	if (!VLAN_ALLOWED(adapter)) {
-		if ((netdev->features ^ features) & NETIF_F_HW_VLAN_CTAG_RX)
+		if (netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_RX_BIT,
+					    changed))
 			return -EINVAL;
-	} else if ((netdev->features ^ features) & NETIF_F_HW_VLAN_CTAG_RX) {
-		if (features & NETIF_F_HW_VLAN_CTAG_RX)
+	} else if (netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_RX_BIT,
+					   changed)) {
+		if (netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_RX_BIT,
+					    features))
 			adapter->aq_required |=
 				IAVF_FLAG_AQ_ENABLE_VLAN_STRIPPING;
 		else
@@ -3395,7 +3400,7 @@ static void iavf_features_check(struct sk_buff *skb, struct net_device *dev,
 	 * 64 bytes.  If it is then we need to drop support for GSO.
 	 */
 	if (skb_is_gso(skb) && (skb_shinfo(skb)->gso_size < 64))
-		*features &= ~NETIF_F_GSO_MASK;
+		netdev_feature_clear_bits(NETIF_F_GSO_MASK, features);
 
 	/* MACLEN can support at most 63 words */
 	len = skb_network_header(skb) - skb->data;
@@ -3427,7 +3432,8 @@ static void iavf_features_check(struct sk_buff *skb, struct net_device *dev,
 
 	return;
 out_err:
-	*features &= ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
+	netdev_feature_clear_bits(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK,
+				  features);
 }
 
 /**
@@ -3443,9 +3449,10 @@ static void iavf_fix_features(struct net_device *netdev,
 	struct iavf_adapter *adapter = netdev_priv(netdev);
 
 	if (!(adapter->vf_res->vf_cap_flags & VIRTCHNL_VF_OFFLOAD_VLAN))
-		*features &= ~(NETIF_F_HW_VLAN_CTAG_TX |
-			      NETIF_F_HW_VLAN_CTAG_RX |
-			      NETIF_F_HW_VLAN_CTAG_FILTER);
+		netdev_feature_clear_bits(NETIF_F_HW_VLAN_CTAG_TX |
+					  NETIF_F_HW_VLAN_CTAG_RX |
+					  NETIF_F_HW_VLAN_CTAG_FILTER,
+					  features);
 }
 
 static const struct net_device_ops iavf_netdev_ops = {
@@ -3530,84 +3537,103 @@ int iavf_process_config(struct iavf_adapter *adapter)
 	}
 	adapter->num_req_queues = 0;
 
-	hw_enc_features = NETIF_F_SG			|
-			  NETIF_F_IP_CSUM		|
-			  NETIF_F_IPV6_CSUM		|
-			  NETIF_F_HIGHDMA		|
-			  NETIF_F_SOFT_FEATURES	|
-			  NETIF_F_TSO			|
-			  NETIF_F_TSO_ECN		|
-			  NETIF_F_TSO6			|
-			  NETIF_F_SCTP_CRC		|
-			  NETIF_F_RXHASH		|
-			  NETIF_F_RXCSUM		|
-			  0;
+	netdev_feature_zero(&hw_enc_features);
+	netdev_feature_set_bits(NETIF_F_SG		|
+				NETIF_F_IP_CSUM		|
+				NETIF_F_IPV6_CSUM	|
+				NETIF_F_HIGHDMA		|
+				NETIF_F_SOFT_FEATURES	|
+				NETIF_F_TSO		|
+				NETIF_F_TSO_ECN		|
+				NETIF_F_TSO6		|
+				NETIF_F_SCTP_CRC	|
+				NETIF_F_RXHASH		|
+				NETIF_F_RXCSUM,
+				&hw_enc_features);
 
 	/* advertise to stack only if offloads for encapsulated packets is
 	 * supported
 	 */
 	if (vfres->vf_cap_flags & VIRTCHNL_VF_OFFLOAD_ENCAP) {
-		hw_enc_features |= NETIF_F_GSO_UDP_TUNNEL	|
-				   NETIF_F_GSO_GRE		|
-				   NETIF_F_GSO_GRE_CSUM		|
-				   NETIF_F_GSO_IPXIP4		|
-				   NETIF_F_GSO_IPXIP6		|
-				   NETIF_F_GSO_UDP_TUNNEL_CSUM	|
-				   NETIF_F_GSO_PARTIAL		|
-				   0;
+		netdev_feature_set_bits(NETIF_F_GSO_UDP_TUNNEL		|
+					NETIF_F_GSO_GRE			|
+					NETIF_F_GSO_GRE_CSUM		|
+					NETIF_F_GSO_IPXIP4		|
+					NETIF_F_GSO_IPXIP6		|
+					NETIF_F_GSO_UDP_TUNNEL_CSUM	|
+					NETIF_F_GSO_PARTIAL,
+					&hw_enc_features);
 
 		if (!(vfres->vf_cap_flags &
 		      VIRTCHNL_VF_OFFLOAD_ENCAP_CSUM))
-			netdev->gso_partial_features |=
-				NETIF_F_GSO_UDP_TUNNEL_CSUM;
+			netdev_feature_set_bit(NETIF_F_GSO_UDP_TUNNEL_CSUM_BIT,
+					       &netdev->gso_partial_features);
 
-		netdev->gso_partial_features |= NETIF_F_GSO_GRE_CSUM;
-		netdev->hw_enc_features |= NETIF_F_TSO_MANGLEID;
-		netdev->hw_enc_features |= hw_enc_features;
+		netdev_feature_set_bit(NETIF_F_GSO_GRE_CSUM_BIT,
+				       &netdev->gso_partial_features);
+		netdev_feature_set_bit(NETIF_F_TSO_MANGLEID_BIT,
+				       &netdev->hw_enc_features);
+		netdev_feature_or(&netdev->hw_enc_features,
+				  netdev->hw_enc_features, hw_enc_features);
 	}
 	/* record features VLANs can make use of */
-	netdev->vlan_features |= hw_enc_features | NETIF_F_TSO_MANGLEID;
+	netdev_feature_or(&netdev->vlan_features,
+			  netdev->vlan_features, hw_enc_features);
+	netdev_feature_set_bit(NETIF_F_TSO_MANGLEID_BIT,
+			       &netdev->vlan_features);
 
 	/* Write features and hw_features separately to avoid polluting
 	 * with, or dropping, features that are set when we registered.
 	 */
-	hw_features = hw_enc_features;
+	netdev_feature_copy(&hw_features, hw_enc_features);
 
 	/* Enable VLAN features if supported */
 	if (vfres->vf_cap_flags & VIRTCHNL_VF_OFFLOAD_VLAN)
-		hw_features |= (NETIF_F_HW_VLAN_CTAG_TX |
-				NETIF_F_HW_VLAN_CTAG_RX);
+		netdev_feature_set_bits(NETIF_F_HW_VLAN_CTAG_TX |
+					NETIF_F_HW_VLAN_CTAG_RX, &hw_features);
 	/* Enable cloud filter if ADQ is supported */
 	if (vfres->vf_cap_flags & VIRTCHNL_VF_OFFLOAD_ADQ)
-		hw_features |= NETIF_F_HW_TC;
+		netdev_feature_set_bit(NETIF_F_HW_TC_BIT, &hw_features);
 	if (vfres->vf_cap_flags & VIRTCHNL_VF_OFFLOAD_USO)
-		hw_features |= NETIF_F_GSO_UDP_L4;
+		netdev_feature_set_bit(NETIF_F_GSO_UDP_L4_BIT, &hw_features);
 
-	netdev->hw_features |= hw_features;
+	netdev_feature_or(&netdev->hw_features, netdev->hw_features,
+			  hw_features);
 
-	netdev->features |= hw_features;
+	netdev_feature_or(&netdev->features, netdev->features, hw_features);
 
 	if (vfres->vf_cap_flags & VIRTCHNL_VF_OFFLOAD_VLAN)
-		netdev->features |= NETIF_F_HW_VLAN_CTAG_FILTER;
+		netdev_feature_set_bit(NETIF_F_HW_VLAN_CTAG_FILTER_BIT,
+				       &netdev->features);
 
 	netdev->priv_flags |= IFF_UNICAST_FLT;
 
 	/* Do not turn on offloads when they are requested to be turned off.
 	 * TSO needs minimum 576 bytes to work correctly.
 	 */
-	if (netdev->wanted_features) {
-		if (!(netdev->wanted_features & NETIF_F_TSO) ||
+	if (!netdev_feature_empty(netdev->wanted_features)) {
+		if (!netdev_feature_test_bit(NETIF_F_TSO_BIT,
+					     netdev->wanted_features) ||
 		    netdev->mtu < 576)
-			netdev->features &= ~NETIF_F_TSO;
-		if (!(netdev->wanted_features & NETIF_F_TSO6) ||
+			netdev_feature_clear_bit(NETIF_F_TSO_BIT,
+						 &netdev->features);
+		if (!netdev_feature_test_bit(NETIF_F_TSO6_BIT,
+					     netdev->wanted_features) ||
 		    netdev->mtu < 576)
-			netdev->features &= ~NETIF_F_TSO6;
-		if (!(netdev->wanted_features & NETIF_F_TSO_ECN))
-			netdev->features &= ~NETIF_F_TSO_ECN;
-		if (!(netdev->wanted_features & NETIF_F_GRO))
-			netdev->features &= ~NETIF_F_GRO;
-		if (!(netdev->wanted_features & NETIF_F_GSO))
-			netdev->features &= ~NETIF_F_GSO;
+			netdev_feature_clear_bit(NETIF_F_TSO6_BIT,
+						 &netdev->features);
+		if (!netdev_feature_test_bit(NETIF_F_TSO_ECN_BIT,
+					     netdev->wanted_features))
+			netdev_feature_clear_bit(NETIF_F_TSO_ECN_BIT,
+						 &netdev->features);
+		if (!netdev_feature_test_bit(NETIF_F_GRO_BIT,
+					     netdev->wanted_features))
+			netdev_feature_clear_bit(NETIF_F_GRO_BIT,
+						 &netdev->features);
+		if (!netdev_feature_test_bit(NETIF_F_GSO_BIT,
+					     netdev->wanted_features))
+			netdev_feature_clear_bit(NETIF_F_GSO_BIT,
+						 &netdev->features);
 	}
 
 	adapter->vsi.id = adapter->vsi_res->vsi_id;
diff --git a/drivers/net/ethernet/intel/iavf/iavf_txrx.c b/drivers/net/ethernet/intel/iavf/iavf_txrx.c
index 3525eab8e9f9..2293ad268f9a 100644
--- a/drivers/net/ethernet/intel/iavf/iavf_txrx.c
+++ b/drivers/net/ethernet/intel/iavf/iavf_txrx.c
@@ -862,7 +862,8 @@ static void iavf_receive_skb(struct iavf_ring *rx_ring,
 {
 	struct iavf_q_vector *q_vector = rx_ring->q_vector;
 
-	if ((rx_ring->netdev->features & NETIF_F_HW_VLAN_CTAG_RX) &&
+	if (netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_RX_BIT,
+				    rx_ring->netdev->features) &&
 	    (vlan_tag & VLAN_VID_MASK))
 		__vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q), vlan_tag);
 
@@ -963,7 +964,8 @@ static inline void iavf_rx_checksum(struct iavf_vsi *vsi,
 	skb_checksum_none_assert(skb);
 
 	/* Rx csum enabled and ip headers found? */
-	if (!(vsi->netdev->features & NETIF_F_RXCSUM))
+	if (!netdev_feature_test_bit(NETIF_F_RXCSUM_BIT,
+				     vsi->netdev->features))
 		return;
 
 	/* did the hardware decode the packet and checksum? */
@@ -1058,7 +1060,8 @@ static inline void iavf_rx_hash(struct iavf_ring *ring,
 		cpu_to_le64((u64)IAVF_RX_DESC_FLTSTAT_RSS_HASH <<
 			    IAVF_RX_DESC_STATUS_FLTSTAT_SHIFT);
 
-	if (ring->netdev->features & NETIF_F_RXHASH)
+	if (netdev_feature_test_bit(NETIF_F_RXHASH_BIT,
+				    ring->netdev->features))
 		return;
 
 	if ((rx_desc->wb.qword1.status_error_len & rss_mask) == rss_mask) {
@@ -1789,7 +1792,8 @@ static inline int iavf_tx_prepare_vlan_flags(struct sk_buff *skb,
 	u32  tx_flags = 0;
 
 	if (protocol == htons(ETH_P_8021Q) &&
-	    !(tx_ring->netdev->features & NETIF_F_HW_VLAN_CTAG_TX)) {
+	    !netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_TX_BIT,
+				     tx_ring->netdev->features)) {
 		/* When HW VLAN acceleration is turned off by the user the
 		 * stack sets the protocol to 8021q so that the driver
 		 * can take any steps required to support the SW only
diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c
index 1ad630b8d103..682b14bebdee 100644
--- a/drivers/net/ethernet/intel/ice/ice_main.c
+++ b/drivers/net/ethernet/intel/ice/ice_main.c
@@ -3065,53 +3065,74 @@ static void ice_set_netdev_features(struct net_device *netdev)
 
 	if (ice_is_safe_mode(pf)) {
 		/* safe mode */
-		netdev->features = NETIF_F_SG | NETIF_F_HIGHDMA;
-		netdev->hw_features = netdev->features;
+		netdev_feature_zero(&netdev->features);
+		netdev_feature_set_bits(NETIF_F_SG | NETIF_F_HIGHDMA,
+					&netdev->features);
+		netdev_feature_copy(&netdev->hw_features, netdev->features);
 		return;
 	}
 
-	dflt_features = NETIF_F_SG	|
-			NETIF_F_HIGHDMA	|
-			NETIF_F_NTUPLE	|
-			NETIF_F_RXHASH;
-
-	csumo_features = NETIF_F_RXCSUM	  |
-			 NETIF_F_IP_CSUM  |
-			 NETIF_F_SCTP_CRC |
-			 NETIF_F_IPV6_CSUM;
-
-	vlano_features = NETIF_F_HW_VLAN_CTAG_FILTER |
-			 NETIF_F_HW_VLAN_CTAG_TX     |
-			 NETIF_F_HW_VLAN_CTAG_RX;
-
-	tso_features = NETIF_F_TSO			|
-		       NETIF_F_TSO_ECN			|
-		       NETIF_F_TSO6			|
-		       NETIF_F_GSO_GRE			|
-		       NETIF_F_GSO_UDP_TUNNEL		|
-		       NETIF_F_GSO_GRE_CSUM		|
-		       NETIF_F_GSO_UDP_TUNNEL_CSUM	|
-		       NETIF_F_GSO_PARTIAL		|
-		       NETIF_F_GSO_IPXIP4		|
-		       NETIF_F_GSO_IPXIP6		|
-		       NETIF_F_GSO_UDP_L4;
-
-	netdev->gso_partial_features |= NETIF_F_GSO_UDP_TUNNEL_CSUM |
-					NETIF_F_GSO_GRE_CSUM;
+	netdev_feature_zero(&dflt_features);
+	netdev_feature_set_bits(NETIF_F_SG		|
+				NETIF_F_HIGHDMA		|
+				NETIF_F_NTUPLE		|
+				NETIF_F_RXHASH,
+				&dflt_features);
+
+	netdev_feature_zero(&csumo_features);
+	netdev_feature_set_bits(NETIF_F_RXCSUM		|
+				NETIF_F_IP_CSUM		|
+				NETIF_F_SCTP_CRC	|
+				NETIF_F_IPV6_CSUM,
+				&csumo_features);
+
+	netdev_feature_zero(&vlano_features);
+	netdev_feature_set_bits(NETIF_F_HW_VLAN_CTAG_FILTER	|
+				NETIF_F_HW_VLAN_CTAG_TX		|
+				NETIF_F_HW_VLAN_CTAG_RX,
+				&vlano_features);
+
+	netdev_feature_zero(&tso_features);
+	netdev_feature_set_bits(NETIF_F_TSO			|
+				NETIF_F_TSO_ECN			|
+				NETIF_F_TSO6			|
+				NETIF_F_GSO_GRE			|
+				NETIF_F_GSO_UDP_TUNNEL		|
+				NETIF_F_GSO_GRE_CSUM		|
+				NETIF_F_GSO_UDP_TUNNEL_CSUM	|
+				NETIF_F_GSO_PARTIAL		|
+				NETIF_F_GSO_IPXIP4		|
+				NETIF_F_GSO_IPXIP6		|
+				NETIF_F_GSO_UDP_L4,
+				&tso_features);
+
+	netdev_feature_set_bits(NETIF_F_GSO_UDP_TUNNEL_CSUM |
+				NETIF_F_GSO_GRE_CSUM,
+				&netdev->gso_partial_features);
 	/* set features that user can change */
-	netdev->hw_features = dflt_features | csumo_features |
-			      vlano_features | tso_features;
+	netdev_feature_or(&netdev->hw_features, dflt_features,
+			  csumo_features);
+	netdev_feature_or(&netdev->hw_features, netdev->hw_features,
+			  vlano_features);
+	netdev_feature_or(&netdev->hw_features, netdev->hw_features,
+			  tso_features);
 
 	/* add support for HW_CSUM on packets with MPLS header */
-	netdev->mpls_features =  NETIF_F_HW_CSUM;
+	netdev_feature_zero(&netdev->mpls_features);
+	netdev_feature_set_bit(NETIF_F_HW_CSUM_BIT, &netdev->mpls_features);
 
 	/* enable features */
-	netdev->features |= netdev->hw_features;
+	netdev_feature_or(&netdev->features, netdev->features,
+			  netdev->hw_features);
 	/* encap and VLAN devices inherit default, csumo and tso features */
-	netdev->hw_enc_features |= dflt_features | csumo_features |
-				   tso_features;
-	netdev->vlan_features |= dflt_features | csumo_features |
-				 tso_features;
+	netdev_feature_or(&netdev->hw_enc_features, dflt_features,
+			  csumo_features);
+	netdev_feature_or(&netdev->hw_enc_features, netdev->hw_enc_features,
+			  tso_features);
+	netdev_feature_or(&netdev->vlan_features, dflt_features,
+			  csumo_features);
+	netdev_feature_or(&netdev->vlan_features, netdev->vlan_features,
+			  tso_features);
 }
 
 /**
@@ -5355,39 +5376,51 @@ ice_set_features(struct net_device *netdev, netdev_features_t features)
 	/* Multiple features can be changed in one call so keep features in
 	 * separate if/else statements to guarantee each feature is checked
 	 */
-	if (features & NETIF_F_RXHASH && !(netdev->features & NETIF_F_RXHASH))
+	if (netdev_feature_test_bit(NETIF_F_RXHASH_BIT, features) &&
+	    !netdev_feature_test_bit(NETIF_F_RXHASH_BIT, netdev->features))
 		ice_vsi_manage_rss_lut(vsi, true);
-	else if (!(features & NETIF_F_RXHASH) &&
-		 netdev->features & NETIF_F_RXHASH)
+	else if (!netdev_feature_test_bit(NETIF_F_RXHASH_BIT, features) &&
+		 netdev_feature_test_bit(NETIF_F_RXHASH_BIT, netdev->features))
 		ice_vsi_manage_rss_lut(vsi, false);
 
-	if ((features & NETIF_F_HW_VLAN_CTAG_RX) &&
-	    !(netdev->features & NETIF_F_HW_VLAN_CTAG_RX))
+	if (netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_RX_BIT, features) &&
+	    !netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_RX_BIT,
+				     netdev->features))
 		ret = ice_vsi_manage_vlan_stripping(vsi, true);
-	else if (!(features & NETIF_F_HW_VLAN_CTAG_RX) &&
-		 (netdev->features & NETIF_F_HW_VLAN_CTAG_RX))
+	else if (!netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_RX_BIT,
+					  features) &&
+		 netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_RX_BIT,
+					 netdev->features))
 		ret = ice_vsi_manage_vlan_stripping(vsi, false);
 
-	if ((features & NETIF_F_HW_VLAN_CTAG_TX) &&
-	    !(netdev->features & NETIF_F_HW_VLAN_CTAG_TX))
+	if (netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_TX_BIT, features) &&
+	    !netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_TX_BIT,
+				     netdev->features))
 		ret = ice_vsi_manage_vlan_insertion(vsi);
-	else if (!(features & NETIF_F_HW_VLAN_CTAG_TX) &&
-		 (netdev->features & NETIF_F_HW_VLAN_CTAG_TX))
+	else if (!netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_TX_BIT,
+					  features) &&
+		 netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_TX_BIT,
+					 netdev->features))
 		ret = ice_vsi_manage_vlan_insertion(vsi);
 
-	if ((features & NETIF_F_HW_VLAN_CTAG_FILTER) &&
-	    !(netdev->features & NETIF_F_HW_VLAN_CTAG_FILTER))
+	if (netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_FILTER_BIT,
+				    features) &&
+	    !netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_FILTER_BIT,
+				     netdev->features))
 		ret = ice_cfg_vlan_pruning(vsi, true, false);
-	else if (!(features & NETIF_F_HW_VLAN_CTAG_FILTER) &&
-		 (netdev->features & NETIF_F_HW_VLAN_CTAG_FILTER))
+	else if (!netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_FILTER_BIT,
+					  features) &&
+		 netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_FILTER_BIT,
+					 netdev->features))
 		ret = ice_cfg_vlan_pruning(vsi, false, false);
 
-	if ((features & NETIF_F_NTUPLE) &&
-	    !(netdev->features & NETIF_F_NTUPLE)) {
+	if (netdev_feature_test_bit(NETIF_F_NTUPLE_BIT, features) &&
+	    !netdev_feature_test_bit(NETIF_F_NTUPLE_BIT, netdev->features)) {
 		ice_vsi_manage_fdir(vsi, true);
 		ice_init_arfs(vsi);
-	} else if (!(features & NETIF_F_NTUPLE) &&
-		 (netdev->features & NETIF_F_NTUPLE)) {
+	} else if (!netdev_feature_test_bit(NETIF_F_NTUPLE_BIT, features) &&
+		   netdev_feature_test_bit(NETIF_F_NTUPLE_BIT,
+					   netdev->features)) {
 		ice_vsi_manage_fdir(vsi, false);
 		ice_clear_arfs(vsi);
 	}
@@ -5403,9 +5436,11 @@ static int ice_vsi_vlan_setup(struct ice_vsi *vsi)
 {
 	int ret = 0;
 
-	if (vsi->netdev->features & NETIF_F_HW_VLAN_CTAG_RX)
+	if (netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_RX_BIT,
+				    vsi->netdev->features))
 		ret = ice_vsi_manage_vlan_stripping(vsi, true);
-	if (vsi->netdev->features & NETIF_F_HW_VLAN_CTAG_TX)
+	if (netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_TX_BIT,
+				    vsi->netdev->features))
 		ret = ice_vsi_manage_vlan_insertion(vsi);
 
 	return ret;
@@ -7193,7 +7228,7 @@ static void ice_features_check(struct sk_buff *skb,
 	 * 64 bytes. If it is then we need to drop support for GSO.
 	 */
 	if (skb_is_gso(skb) && (skb_shinfo(skb)->gso_size < 64))
-		*features &= ~NETIF_F_GSO_MASK;
+		netdev_feature_clear_bits(NETIF_F_GSO_MASK, features);
 
 	len = skb_network_header(skb) - skb->data;
 	if (len > ICE_TXD_MACLEN_MAX || len & 0x1)
@@ -7216,7 +7251,8 @@ static void ice_features_check(struct sk_buff *skb,
 
 	return;
 out_rm_features:
-	*features &= ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
+	netdev_feature_clear_bits(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK,
+				  features);
 }
 
 static const struct net_device_ops ice_netdev_safe_mode_ops = {
diff --git a/drivers/net/ethernet/intel/ice/ice_txrx_lib.c b/drivers/net/ethernet/intel/ice/ice_txrx_lib.c
index 171397dcf00a..12211d2d0ed3 100644
--- a/drivers/net/ethernet/intel/ice/ice_txrx_lib.c
+++ b/drivers/net/ethernet/intel/ice/ice_txrx_lib.c
@@ -72,7 +72,8 @@ ice_rx_hash(struct ice_ring *rx_ring, union ice_32b_rx_flex_desc *rx_desc,
 	struct ice_32b_rx_flex_desc_nic *nic_mdid;
 	u32 hash;
 
-	if (!(rx_ring->netdev->features & NETIF_F_RXHASH))
+	if (!netdev_feature_test_bit(NETIF_F_RXHASH_BIT,
+				     rx_ring->netdev->features))
 		return;
 
 	if (rx_desc->wb.rxdid != ICE_RXDID_FLEX_NIC)
@@ -110,7 +111,8 @@ ice_rx_csum(struct ice_ring *ring, struct sk_buff *skb,
 	skb_checksum_none_assert(skb);
 
 	/* check if Rx checksum is enabled */
-	if (!(ring->netdev->features & NETIF_F_RXCSUM))
+	if (!netdev_feature_test_bit(NETIF_F_RXCSUM_BIT,
+				     ring->netdev->features))
 		return;
 
 	/* check if HW has decoded the packet and checksum */
@@ -205,7 +207,8 @@ ice_process_skb_fields(struct ice_ring *rx_ring,
 void
 ice_receive_skb(struct ice_ring *rx_ring, struct sk_buff *skb, u16 vlan_tag)
 {
-	if ((rx_ring->netdev->features & NETIF_F_HW_VLAN_CTAG_RX) &&
+	if (netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_RX_BIT,
+				    rx_ring->netdev->features)  &&
 	    (vlan_tag & VLAN_VID_MASK))
 		__vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q), vlan_tag);
 	napi_gro_receive(&rx_ring->q_vector->napi, skb);
diff --git a/drivers/net/ethernet/intel/igb/igb_ethtool.c b/drivers/net/ethernet/intel/igb/igb_ethtool.c
index fb1029352c3e..89b4f28ad71c 100644
--- a/drivers/net/ethernet/intel/igb/igb_ethtool.c
+++ b/drivers/net/ethernet/intel/igb/igb_ethtool.c
@@ -2901,7 +2901,7 @@ static int igb_add_ethtool_nfc_entry(struct igb_adapter *adapter,
 	struct igb_nfc_filter *input, *rule;
 	int err = 0;
 
-	if (!(netdev->hw_features & NETIF_F_NTUPLE))
+	if (!netdev_feature_test_bit(NETIF_F_NTUPLE_BIT, netdev->hw_features))
 		return -EOPNOTSUPP;
 
 	/* Don't allow programming if the action is a queue greater than
diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c
index 2e4a53c76c60..1f764d4fbd14 100644
--- a/drivers/net/ethernet/intel/igb/igb_main.c
+++ b/drivers/net/ethernet/intel/igb/igb_main.c
@@ -2439,25 +2439,28 @@ static void igb_fix_features(struct net_device *netdev,
 	/* Since there is no support for separate Rx/Tx vlan accel
 	 * enable/disable make sure Tx flag is always in same state as Rx.
 	 */
-	if (*features & NETIF_F_HW_VLAN_CTAG_RX)
-		*features |= NETIF_F_HW_VLAN_CTAG_TX;
+	if (netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_RX_BIT, *features))
+		netdev_feature_set_bit(NETIF_F_HW_VLAN_CTAG_TX_BIT, features);
 	else
-		*features &= ~NETIF_F_HW_VLAN_CTAG_TX;
+		netdev_feature_clear_bit(NETIF_F_HW_VLAN_CTAG_TX_BIT,
+					 features);
 }
 
 static int igb_set_features(struct net_device *netdev,
 	netdev_features_t features)
 {
-	netdev_features_t changed = netdev->features ^ features;
 	struct igb_adapter *adapter = netdev_priv(netdev);
+	netdev_features_t changed;
 
-	if (changed & NETIF_F_HW_VLAN_CTAG_RX)
+	netdev_feature_xor(&changed, netdev->features, features);
+
+	if (netdev_feature_test_bit(NETIF_F_HW_VLAN_CTAG_RX_BIT, changed))
 		igb_vlan_mode(netdev, features);
 
-	if (!(changed & (NETIF_F_RXALL | NETIF_F_NTUPLE)))
+	if (!netdev_feature_test_bits(NETIF_F_RXALL | NETIF_F_NTUPLE, changed))
 		return 0;
 
-	if (!(features & NETIF_F_NTUPLE)) {
+	if (!netdev_feature_test_bit(NETIF_F_NTUPLE_BIT, features)) {
 		struct hlist_node *node2;
 		struct igb_nfc_filter *rule;
 
@@ -2472,7 +2475,7 @@ static int igb_set_features(struct net_device *netdev,
 		adapter->nfc_filter_count = 0;
 	}
 
-	netdev->features = features;
+	netdev_feature_copy(&netdev->features, features);
 
 	if (netif_running(netdev))
 		igb_reinit_locked(adapter);
@@ -2511,30 +2514,33 @@ static void igb_features_check(struct sk_buff *skb, struct net_device *dev,
 	/* Make certain the headers can be described by a context descriptor */
 	mac_hdr_len = skb_network_header(skb) - skb->data;
 	if (unlikely(mac_hdr_len > IGB_MAX_MAC_HDR_LEN)) {
-		*features &= ~(NETIF_F_HW_CSUM |
-			       NETIF_F_SCTP_CRC |
-			       NETIF_F_GSO_UDP_L4 |
-			       NETIF_F_HW_VLAN_CTAG_TX |
-			       NETIF_F_TSO |
-			       NETIF_F_TSO6);
+		netdev_feature_clear_bits(NETIF_F_HW_CSUM |
+					NETIF_F_SCTP_CRC |
+					NETIF_F_GSO_UDP_L4 |
+					NETIF_F_HW_VLAN_CTAG_TX |
+					NETIF_F_TSO |
+					NETIF_F_TSO6,
+					features);
 		return;
 	}
 
 	network_hdr_len = skb_checksum_start(skb) - skb_network_header(skb);
 	if (unlikely(network_hdr_len >  IGB_MAX_NETWORK_HDR_LEN)) {
-		*features &= ~(NETIF_F_HW_CSUM |
-			       NETIF_F_SCTP_CRC |
-			       NETIF_F_GSO_UDP_L4 |
-			       NETIF_F_TSO |
-			       NETIF_F_TSO6);
+		netdev_feature_clear_bits(NETIF_F_HW_CSUM |
+					NETIF_F_SCTP_CRC |
+					NETIF_F_GSO_UDP_L4 |
+					NETIF_F_TSO |
+					NETIF_F_TSO6,
+					features);
 		return;
 	}
 
 	/* We can only support IPV4 TSO in tunnels if we can mangle the
 	 * inner IP ID field, so strip TSO if MANGLEID is not supported.
 	 */
-	if (skb->encapsulation && !(*features & NETIF_F_TSO_MANGLEID))
-		*features &= ~NETIF_F_TSO;
+	if (skb->encapsulation &&
+	    !netdev_feature_test_bit(NETIF_F_TSO_MANGLEID_BIT, *features))
+		netdev_feature_clear_bit(NETIF_F_TSO_BIT, features);
 }
 
 static void igb_offload_apply(struct igb_adapter *adapter, s32 queue)
@@ -3266,18 +3272,20 @@ static int igb_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	 * set by igb_sw_init so we should use an or instead of an
 	 * assignment.
 	 */
-	netdev->features |= NETIF_F_SG |
-			    NETIF_F_TSO |
-			    NETIF_F_TSO6 |
-			    NETIF_F_RXHASH |
-			    NETIF_F_RXCSUM |
-			    NETIF_F_HW_CSUM;
+	netdev_feature_set_bits(NETIF_F_SG |
+				NETIF_F_TSO |
+				NETIF_F_TSO6 |
+				NETIF_F_RXHASH |
+				NETIF_F_RXCSUM |
+				NETIF_F_HW_CSUM,
+				&netdev->features);
 
 	if (hw->mac.type >= e1000_82576)
-		netdev->features |= NETIF_F_SCTP_CRC | NETIF_F_GSO_UDP_L4;
+		netdev_feature_set_bits(NETIF_F_SCTP_CRC | NETIF_F_GSO_UDP_L4,
+					&netdev->features);
 
 	if (hw->mac.type >= e1000_i350)
-		netdev->features |= NETIF_F_HW_TC;
+		netdev_feature_set_bit(NETIF_F_HW_TC_BIT, &netdev->features);
 
 #define IGB_GSO_PARTIAL_FEATURES (NETIF_F_GSO_GRE | \
 				  NETIF_F_GSO_GRE_CSUM | \
@@ -3286,29 +3294,41 @@ static int igb_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 				  NETIF_F_GSO_UDP_TUNNEL | \
 				  NETIF_F_GSO_UDP_TUNNEL_CSUM)
 
-	netdev->gso_partial_features = IGB_GSO_PARTIAL_FEATURES;
-	netdev->features |= NETIF_F_GSO_PARTIAL | IGB_GSO_PARTIAL_FEATURES;
+	netdev_feature_zero(&netdev->gso_partial_features);
+	netdev_feature_set_bits(IGB_GSO_PARTIAL_FEATURES,
+				&netdev->gso_partial_features);
+	netdev_feature_set_bits(NETIF_F_GSO_PARTIAL |
+				IGB_GSO_PARTIAL_FEATURES,
+				&netdev->features);
 
 	/* copy netdev features into list of user selectable features */
-	netdev->hw_features |= netdev->features |
-			       NETIF_F_HW_VLAN_CTAG_RX |
-			       NETIF_F_HW_VLAN_CTAG_TX |
-			       NETIF_F_RXALL;
+	netdev_feature_or(&netdev->hw_features, netdev->hw_features,
+			  netdev->features);
+	netdev_feature_set_bits(NETIF_F_HW_VLAN_CTAG_RX |
+				NETIF_F_HW_VLAN_CTAG_TX |
+				NETIF_F_RXALL,
+				&netdev->hw_features);
 
 	if (hw->mac.type >= e1000_i350)
-		netdev->hw_features |= NETIF_F_NTUPLE;
+		netdev_feature_set_bit(NETIF_F_NTUPLE_BIT,
+				       &netdev->hw_features);
 
 	if (pci_using_dac)
-		netdev->features |= NETIF_F_HIGHDMA;
+		netdev_feature_set_bit(NETIF_F_HIGHDMA_BIT, &netdev->features);
 
-	netdev->vlan_features |= netdev->features | NETIF_F_TSO_MANGLEID;
-	netdev->mpls_features |= NETIF_F_HW_CSUM;
-	netdev->hw_enc_features |= netdev->vlan_features;
+	netdev_feature_or(&netdev->vlan_features, netdev->vlan_features,
+			  netdev->features);
+	netdev_feature_set_bit(NETIF_F_TSO_MANGLEID_BIT,
+			       &netdev->vlan_features);
+	netdev_feature_set_bit(NETIF_F_HW_CSUM_BIT, &netdev->mpls_features);
+	netdev_feature_or(&netdev->hw_enc_features, netdev->hw_enc_features,
+			  netdev->vlan_features);
 
 	/* set this bit last since it cannot be part of vlan_features */
-	netdev->features |= NETIF_F_HW_VLAN_CTAG_FILTER |
-			    NETIF_F_HW_VLAN_CTAG_RX |
-			    NETIF_F_HW_VLAN_CTAG_TX;
+	netdev_feature_set_bits(NETIF_F_HW_VLAN_CTAG_FILTER |
+				NETIF_F_HW_VLAN_CTAG_RX |
+				NETIF_F_HW_VLAN_CTAG_TX,
+				&netdev->features);
 
 	netdev->priv_flags |= IFF_SUPP_NOFCS;
 
@@ -4539,7 +4559,8 @@ void igb_setup_rctl(struct igb_adapter *adapter)
 	}
 
 	/* This is useful for sniffing bad packets. */
-	if (adapter->netdev->features & NETIF_F_RXALL) {
+	if (netdev_feature_test_bit(NETIF_F_RXALL_BIT,
+				    adapter->netdev->features)) {
 		/* UPE and MPE will be handled by normal PROMISC logic
 		 * in e1000e_set_rx_mode
 		 */
@@ -5045,7 +5066,8 @@ static int igb_vlan_promisc_enable(struct igb_adapter *adapter)
 	case e1000_i211:
 	case e1000_i350:
 		/* VLAN filtering needed for VLAN prio filter */
-		if (adapter->netdev->features & NETIF_F_NTUPLE)
+		if (netdev_feature_test_bit(NETIF_F_NTUPLE_BIT,
+					    adapter->netdev->features))
 			break;
 		fallthrough;
 	case e1000_82576:
@@ -5217,7 +5239,7 @@ static void igb_set_rx_mode(struct net_device *netdev)
 
 	/* disable VLAN filtering for modes that require it */
 	if ((netdev->flags & IFF_PROMISC) ||
-	    (netdev->features & NETIF_F_RXALL)) {
+	    netdev_feature_test_bit(NETIF_F_RXALL_BIT, netdev->features)) {
 		/* if we fail to set all rules then just clear VFE */
 		if (igb_vlan_promisc_enable(adapter))
 			rctl &= ~E1000_RCTL_VFE;
@@ -8467,7 +8489,8 @@ static inline void igb_rx_checksum(struct igb_ring *ring,
 		return;
 
 	/* Rx checksum disabled via ethtool */
-	if (!(ring->netdev->features & NETIF_F_RXCSUM))
+	if (!netdev_feature_test_bit(NETIF_F_RXCSUM_BIT,
+				     ring->netdev->features))
 		return;
 
 	/* TCP/UDP checksum error bit is set */
@@ -8500,7 +8523,8 @@ static inline void igb_rx_hash(struct igb_ring *ring,
 			       union e1000_adv_rx_desc *rx_desc,
 			       struct sk_buff *skb)
 {
-	if (ring->netdev->features & NETIF_F_RXHASH)
+	if (netdev_feature_test_bit(NETIF_F_RXHASH_BIT,
+				    ring->netdev->features))
 		skb_set_hash(skb,
 			     le32_to_cpu(rx_desc->wb.lower.hi_dword.rss),
 			     PKT_HASH_TYPE_L3);
@@ -8558,7 +8582,8 @@ static bool igb_cleanup_headers(struct igb_ring *rx_ring,
 	if (unlikely((igb_test_staterr(rx_desc,
 				       E1000_RXDEXT_ERR_FRAME_ERR_MASK)))) {
 		struct net_device *netdev = rx_ring->netdev;
-		if (!(netdev->features & NETIF_F_RXALL)) {
+		if (!netdev_feature_test_bit(NETIF_F_RXALL_BIT,
+					     netdev->features)) {
 			dev_kfree_skb_any(s