39. OCTEON TX2 Poll Mode driver

The OCTEON TX2 ETHDEV PMD (librte_pmd_octeontx2) provides poll mode ethdev driver support for the inbuilt network device found in Marvell OCTEON TX2 SoC family as well as for their virtual functions (VF) in SR-IOV context.

More information can be found at Marvell Official Website.

39.1. Features

Features of the OCTEON TX2 Ethdev PMD are:

  • Packet type information
  • Promiscuous mode
  • Jumbo frames
  • SR-IOV VF
  • Lock-free Tx queue
  • Multiple queues for TX and RX
  • Receiver Side Scaling (RSS)
  • MAC/VLAN filtering
  • Generic flow API
  • Inner and Outer Checksum offload
  • VLAN/QinQ stripping and insertion
  • Port hardware statistics
  • Link state information
  • Link flow control
  • MTU update
  • Scatter-Gather IO support
  • Vector Poll mode driver
  • Debug utilities - Context dump and error interrupt support
  • IEEE1588 timestamping
  • HW offloaded ethdev Rx queue to eventdev event queue packet injection
  • Support Rx interrupt

39.2. Prerequisites

See Marvell OCTEON TX2 Platform Guide for setup information.

39.3. Compile time Config Options

The following options may be modified in the config file.

  • CONFIG_RTE_LIBRTE_OCTEONTX2_PMD (default y)

    Toggle compilation of the librte_pmd_octeontx2 driver.

39.4. Driver compilation and testing

Refer to the document compiling and testing a PMD for a NIC for details.

To compile the OCTEON TX2 PMD for Linux arm64 gcc, use arm64-octeontx2-linux-gcc as target.

  1. Running testpmd:

    Follow instructions available in the document compiling and testing a PMD for a NIC to run testpmd.

    Example output:

    ./build/app/testpmd -c 0x300 -w 0002:02:00.0 -- --portmask=0x1 --nb-cores=1 --port-topology=loop --rxq=1 --txq=1
    EAL: Detected 24 lcore(s)
    EAL: Detected 1 NUMA nodes
    EAL: Multi-process socket /var/run/dpdk/rte/mp_socket
    EAL: No available hugepages reported in hugepages-2048kB
    EAL: Probing VFIO support...
    EAL: VFIO support initialized
    EAL: PCI device 0002:02:00.0 on NUMA socket 0
    EAL:   probe driver: 177d:a063 net_octeontx2
    EAL:   using IOMMU type 1 (Type 1)
    testpmd: create a new mbuf pool <mbuf_pool_socket_0>: n=267456, size=2176, socket=0
    testpmd: preferred mempool ops selected: octeontx2_npa
    Configuring Port 0 (socket 0)
    PMD: Port 0: Link Up - speed 40000 Mbps - full-duplex
    
    Port 0: link state change event
    Port 0: 36:10:66:88:7A:57
    Checking link statuses...
    Done
    No commandline core given, start packet forwarding
    io packet forwarding - ports=1 - cores=1 - streams=1 - NUMA support enabled, MP allocation mode: native
    Logical Core 9 (socket 0) forwards packets on 1 streams:
      RX P=0/Q=0 (socket 0) -> TX P=0/Q=0 (socket 0) peer=02:00:00:00:00:00
    
      io packet forwarding packets/burst=32
      nb forwarding cores=1 - nb forwarding ports=1
      port 0: RX queue number: 1 Tx queue number: 1
        Rx offloads=0x0 Tx offloads=0x10000
        RX queue: 0
          RX desc=512 - RX free threshold=0
          RX threshold registers: pthresh=0 hthresh=0  wthresh=0
          RX Offloads=0x0
        TX queue: 0
          TX desc=512 - TX free threshold=0
          TX threshold registers: pthresh=0 hthresh=0  wthresh=0
          TX offloads=0x10000 - TX RS bit threshold=0
    Press enter to exit
    

39.5. Runtime Config Options

  • HW offload ptype parsing disable (default 0)

    Packet type parsing is HW offloaded by default and this feature may be toggled using ptype_disable devargs parameter.

  • Rx&Tx scalar mode enable (default 0)

    Ethdev supports both scalar and vector mode, it may be selected at runtime using scalar_enable devargs parameter.

  • RSS reta size (default 64)

    RSS redirection table size may be configured during runtime using reta_size devargs parameter.

    For example:

    -w 0002:02:00.0,reta_size=256
    

    With the above configuration, reta table of size 256 is populated.

  • Flow priority levels (default 3)

    RTE Flow priority levels can be configured during runtime using flow_max_priority devargs parameter.

    For example:

    -w 0002:02:00.0,flow_max_priority=10
    

    With the above configuration, priority level was set to 10 (0-9). Max priority level supported is 32.

  • Reserve Flow entries (default 8)

    RTE flow entries can be pre allocated and the size of pre allocation can be selected runtime using flow_prealloc_size devargs parameter.

    For example:

    -w 0002:02:00.0,flow_prealloc_size=4
    

    With the above configuration, pre alloc size was set to 4. Max pre alloc size supported is 32.

  • Max SQB buffer count (default 512)

    Send queue descriptor buffer count may be limited during runtime using max_sqb_count devargs parameter.

    For example:

    -w 0002:02:00.0,max_sqb_count=64
    

    With the above configuration, each send queue’s decscriptor buffer count is limited to a maximum of 64 buffers.

Note

Above devarg parameters are configurable per device, user needs to pass the parameters to all the PCIe devices if application requires to configure on all the ethdev ports.

39.6. Limitations

39.6.1. mempool_octeontx2 external mempool handler dependency

The OCTEON TX2 SoC family NIC has inbuilt HW assisted external mempool manager. net_octeontx2 pmd only works with mempool_octeontx2 mempool handler as it is performance wise most effective way for packet allocation and Tx buffer recycling on OCTEON TX2 SoC platform.

39.6.2. CRC striping

The OCTEON TX2 SoC family NICs strip the CRC for every packet being received by the host interface irrespective of the offload configuration.

39.7. Debugging Options

Table 39.1 OCTEON TX2 ethdev debug options
# Component EAL log command
1 NIX –log-level=’pmd.net.octeontx2,8’
2 NPC –log-level=’pmd.net.octeontx2.flow,8’

39.8. RTE Flow Support

The OCTEON TX2 SoC family NIC has support for the following patterns and actions.

Patterns:

Table 39.2 Item types
# Pattern Type
1 RTE_FLOW_ITEM_TYPE_ETH
2 RTE_FLOW_ITEM_TYPE_VLAN
3 RTE_FLOW_ITEM_TYPE_E_TAG
4 RTE_FLOW_ITEM_TYPE_IPV4
5 RTE_FLOW_ITEM_TYPE_IPV6
6 RTE_FLOW_ITEM_TYPE_ARP_ETH_IPV4
7 RTE_FLOW_ITEM_TYPE_MPLS
8 RTE_FLOW_ITEM_TYPE_ICMP
9 RTE_FLOW_ITEM_TYPE_UDP
10 RTE_FLOW_ITEM_TYPE_TCP
11 RTE_FLOW_ITEM_TYPE_SCTP
12 RTE_FLOW_ITEM_TYPE_ESP
13 RTE_FLOW_ITEM_TYPE_GRE
14 RTE_FLOW_ITEM_TYPE_NVGRE
15 RTE_FLOW_ITEM_TYPE_VXLAN
16 RTE_FLOW_ITEM_TYPE_GTPC
17 RTE_FLOW_ITEM_TYPE_GTPU
18 RTE_FLOW_ITEM_TYPE_GENEVE
19 RTE_FLOW_ITEM_TYPE_VXLAN_GPE
20 RTE_FLOW_ITEM_TYPE_IPV6_EXT
21 RTE_FLOW_ITEM_TYPE_VOID
22 RTE_FLOW_ITEM_TYPE_ANY
23 RTE_FLOW_ITEM_TYPE_GRE_KEY

Note

RTE_FLOW_ITEM_TYPE_GRE_KEY works only when checksum and routing bits in the GRE header are equal to 0.

Actions:

Table 39.3 Ingress action types
# Action Type
1 RTE_FLOW_ACTION_TYPE_VOID
2 RTE_FLOW_ACTION_TYPE_MARK
3 RTE_FLOW_ACTION_TYPE_FLAG
4 RTE_FLOW_ACTION_TYPE_COUNT
5 RTE_FLOW_ACTION_TYPE_DROP
6 RTE_FLOW_ACTION_TYPE_QUEUE
7 RTE_FLOW_ACTION_TYPE_RSS
8 RTE_FLOW_ACTION_TYPE_SECURITY
9 RTE_FLOW_ACTION_TYPE_PF
10 RTE_FLOW_ACTION_TYPE_VF
Table 39.4 Egress action types
# Action Type
1 RTE_FLOW_ACTION_TYPE_COUNT
2 RTE_FLOW_ACTION_TYPE_DROP