| Message ID | 20210219130001.49600-1-haiyue.wang@intel.com |
|---|---|
| Headers | show |
| Series | Enable Intel VF flow director with DDP | expand |
Haiyue Wang wrote: > The Intel E810 Series supports a programmable pipeline for a domain > specific protocols classification, for example GTP by Dynamic Device > Personalization (DDP) profile. > > The E810 PF has introduced flex-bytes support by ethtool user-def option > allowing for packet deeper matching based on an offset and value for DDP > usage. > > For making VF also benefit from this flexible protocol classification, some > new virtchnl messages are defined and handled by PF, so VF can query this > new flow director capability, and use ethtool with extending the user-def > option to configure Rx flow classification. > > The new user-def 0xAAAABBBBCCCCDDDD: BBBB is the 2 byte pattern while AAAA > corresponds to its offset in the packet. Similarly DDDD is the 2 byte > pattern with CCCC being the corresponding offset. The offset ranges from > 0x0 to 0x1F7 (up to 504 bytes into the packet). The offset starts from > the beginning of the packet. Why must you use the ethtool user-def interface like this? I suspect that you should be using tc-u32. At the very least you should cover why you're *not* using u32 in the cover letter, otherwise I'm sure the question will come up. > This feature can be used to allow customers to set flow director rules > for protocols headers that are beyond standard ones supported by ethtool > (e.g. PFCP or GTP-U). > > Like for matching GTP-U's TEID value 0x10203040: > ethtool -N ens787f0v0 flow-type udp4 dst-port 2152 \ >
The Intel E810 Series supports a programmable pipeline for a domain specific protocols classification, for example GTP by Dynamic Device Personalization (DDP) profile. The E810 PF has introduced flex-bytes support by ethtool user-def option allowing for packet deeper matching based on an offset and value for DDP usage. For making VF also benefit from this flexible protocol classification, some new virtchnl messages are defined and handled by PF, so VF can query this new flow director capability, and use ethtool with extending the user-def option to configure Rx flow classification. The new user-def 0xAAAABBBBCCCCDDDD: BBBB is the 2 byte pattern while AAAA corresponds to its offset in the packet. Similarly DDDD is the 2 byte pattern with CCCC being the corresponding offset. The offset ranges from 0x0 to 0x1F7 (up to 504 bytes into the packet). The offset starts from the beginning of the packet. This feature can be used to allow customers to set flow director rules for protocols headers that are beyond standard ones supported by ethtool (e.g. PFCP or GTP-U). Like for matching GTP-U's TEID value 0x10203040: ethtool -N ens787f0v0 flow-type udp4 dst-port 2152 \ user-def 0x002e102000303040 action 13 v4: Refactor the patch set. v3: Fix the flags bit conflict v2: Fix the building and checkpatch warnings. Haiyue Wang (5): iavf: Add framework to enable ethtool ntuple filters iavf: Support IPv4 Flow Director filters iavf: Support IPv6 Flow Director filters iavf: Support Ethernet Type Flow Director filters iavf: Enable flex-bytes support Qi Zhang (13): ice: Add more basic protocol support for flow filter ice: Support non word aligned input set field ice: Add more advanced protocol support in flow filter ice: Support to separate GTP-U uplink and downlink ice: Enhanced IPv4 and IPv6 flow filter ice: Add support for per VF ctrl VSI enabling ice: Enable FDIR Configure for AVF ice: Add FDIR pattern action parser for VF ice: Add new actions support for VF FDIR ice: Add non-IP Layer2 protocol FDIR filter for AVF ice: Add GTPU FDIR filter for AVF ice: Add more FDIR filter type for AVF ice: Check FDIR program status for AVF drivers/net/ethernet/intel/iavf/Makefile | 2 +- drivers/net/ethernet/intel/iavf/iavf.h | 12 + .../net/ethernet/intel/iavf/iavf_ethtool.c | 631 +++++ drivers/net/ethernet/intel/iavf/iavf_fdir.c | 773 ++++++ drivers/net/ethernet/intel/iavf/iavf_fdir.h | 113 + drivers/net/ethernet/intel/iavf/iavf_main.c | 27 + .../net/ethernet/intel/iavf/iavf_virtchnl.c | 195 ++ drivers/net/ethernet/intel/ice/Makefile | 2 +- drivers/net/ethernet/intel/ice/ice.h | 5 +- .../net/ethernet/intel/ice/ice_ethtool_fdir.c | 4 + drivers/net/ethernet/intel/ice/ice_fdir.c | 486 +++- drivers/net/ethernet/intel/ice/ice_fdir.h | 58 + .../net/ethernet/intel/ice/ice_flex_pipe.c | 528 +++- .../net/ethernet/intel/ice/ice_flex_pipe.h | 3 +- .../net/ethernet/intel/ice/ice_flex_type.h | 78 + drivers/net/ethernet/intel/ice/ice_flow.c | 727 +++++- drivers/net/ethernet/intel/ice/ice_flow.h | 160 +- .../net/ethernet/intel/ice/ice_hw_autogen.h | 19 + .../net/ethernet/intel/ice/ice_lan_tx_rx.h | 22 + drivers/net/ethernet/intel/ice/ice_lib.c | 64 +- drivers/net/ethernet/intel/ice/ice_main.c | 11 +- .../ethernet/intel/ice/ice_protocol_type.h | 10 + drivers/net/ethernet/intel/ice/ice_txrx.c | 5 + drivers/net/ethernet/intel/ice/ice_type.h | 18 + .../ethernet/intel/ice/ice_virtchnl_fdir.c | 2206 +++++++++++++++++ .../ethernet/intel/ice/ice_virtchnl_fdir.h | 49 + .../net/ethernet/intel/ice/ice_virtchnl_pf.c | 79 +- .../net/ethernet/intel/ice/ice_virtchnl_pf.h | 8 + include/linux/avf/virtchnl.h | 278 +++ 29 files changed, 6489 insertions(+), 84 deletions(-) create mode 100644 drivers/net/ethernet/intel/iavf/iavf_fdir.c create mode 100644 drivers/net/ethernet/intel/iavf/iavf_fdir.h create mode 100644 drivers/net/ethernet/intel/ice/ice_virtchnl_fdir.c create mode 100644 drivers/net/ethernet/intel/ice/ice_virtchnl_fdir.h