12. CXGBE Poll Mode Driver

The CXGBE PMD (librte_pmd_cxgbe) provides poll mode driver support for Chelsio Terminator 10/25/40/100 Gbps family of adapters. CXGBE PMD has support for the latest Linux and FreeBSD operating systems.

CXGBEVF PMD provides poll mode driver support for SR-IOV Virtual functions and has support for the latest Linux operating systems.

More information can be found at Chelsio Communications Official Website.

12.1. Features

CXGBE and CXGBEVF PMD has support for:

  • Multiple queues for TX and RX
  • Receiver Side Steering (RSS) Receiver Side Steering (RSS) on IPv4, IPv6, IPv4-TCP/UDP, IPv6-TCP/UDP. For 4-tuple, enabling ‘RSS on TCP’ and ‘RSS on TCP + UDP’ is supported.
  • VLAN filtering
  • Checksum offload
  • Promiscuous mode
  • All multicast mode
  • Port hardware statistics
  • Jumbo frames
  • Flow API - Support for both Wildcard (LE-TCAM) and Exact (HASH) match filters.

12.2. Limitations

The Chelsio Terminator series of devices provide two/four ports but expose a single PCI bus address, thus, librte_pmd_cxgbe registers itself as a PCI driver that allocates one Ethernet device per detected port.

For this reason, one cannot whitelist/blacklist a single port without whitelisting/blacklisting the other ports on the same device.

12.3. Supported Chelsio T5 NICs

  • 1G NICs: T502-BT
  • 10G NICs: T520-BT, T520-CR, T520-LL-CR, T520-SO-CR, T540-CR
  • 40G NICs: T580-CR, T580-LP-CR, T580-SO-CR
  • Other T5 NICs: T522-CR

12.4. Supported Chelsio T6 NICs

  • 25G NICs: T6425-CR, T6225-CR, T6225-LL-CR, T6225-SO-CR
  • 100G NICs: T62100-CR, T62100-LP-CR, T62100-SO-CR

12.5. Supported SR-IOV Chelsio NICs

SR-IOV virtual functions are supported on all the Chelsio NICs listed in Supported Chelsio T5 NICs and Supported Chelsio T6 NICs.

12.6. Prerequisites

  • Requires firmware version 1.24.11.0 and higher. Visit Chelsio Download Center to get latest firmware bundled with the latest Chelsio Unified Wire package.

    For Linux, installing and loading the latest cxgb4 kernel driver from the Chelsio Unified Wire package should get you the latest firmware. More information can be obtained from the User Guide that is bundled with the Chelsio Unified Wire package.

    For FreeBSD, the latest firmware obtained from the Chelsio Unified Wire package must be manually flashed via cxgbetool available in FreeBSD source repository.

    Instructions on how to manually flash the firmware are given in section Linux Installation for Linux and section FreeBSD Installation for FreeBSD.

12.7. Pre-Installation Configuration

12.7.1. Config File Options

The following options can be modified in the .config file. Please note that enabling debugging options may affect system performance.

  • CONFIG_RTE_LIBRTE_CXGBE_PMD (default y)

    Toggle compilation of librte_pmd_cxgbe driver.

    Note

    This controls compilation of both CXGBE and CXGBEVF PMD.

12.7.2. Runtime Options

The following devargs options can be enabled at runtime. They must be passed as part of EAL arguments. For example,

testpmd -w 02:00.4,keep_ovlan=1 -- -i

12.7.2.1. Common Runtime Options

  • keep_ovlan (default 0)

    Toggle behavior to keep/strip outer VLAN in Q-in-Q packets. If enabled, the outer VLAN tag is preserved in Q-in-Q packets. Otherwise, the outer VLAN tag is stripped in Q-in-Q packets.

  • tx_mode_latency (default 0)

    When set to 1, Tx doesn’t wait for max number of packets to get coalesced and sends the packets immediately at the end of the current Tx burst. When set to 0, Tx waits across multiple Tx bursts until the max number of packets have been coalesced. In this case, Tx only sends the coalesced packets to hardware once the max coalesce limit has been reached.

12.7.2.2. CXGBE VF Only Runtime Options

  • force_link_up (default 0)

    When set to 1, CXGBEVF PMD always forces link as up for all VFs on underlying Chelsio NICs. This enables multiple VFs on the same NIC to send traffic to each other even when the physical link is down.

12.7.2.3. CXGBE PF Only Runtime Options

  • filtermode (default 0)

    Apart from the 4-tuple (IP src/dst addresses and TCP/UDP src/dst port addresses), there are only 40-bits available to match other fields in packet headers. So, filtermode devarg allows user to dynamically select a 40-bit supported match field combination for LETCAM (wildcard) filters.

    Default value of 0 makes driver pick the combination configured in the firmware configuration file on the adapter.

    The supported flags and their corresponding values are shown in table below. These flags can be OR’d to create 1 of the multiple supported combinations for LETCAM filters.

    FLAG VALUE
    Physical Port 0x1
    PFVF 0x2
    Destination MAC 0x4
    Ethertype 0x8
    Inner VLAN 0x10
    Outer VLAN 0x20
    IP TOS 0x40
    IP Protocol 0x80

    The supported filtermode combinations and their corresponding OR’d values are shown in table below.

    FILTERMODE COMBINATIONS VALUE
    Protocol, TOS, Outer VLAN, Port 0xE1
    Protocol, TOS, Outer VLAN 0xE0
    Protocol, TOS, Inner VLAN, Port 0xD1
    Protocol, TOS, Inner VLAN 0xD0
    Protocol, TOS, PFVF, Port 0xC3
    Protocol, TOS, PFVF 0xC2
    Protocol, TOS, Port 0xC1
    Protocol, TOS 0xC0
    Protocol, Outer VLAN, Port 0xA1
    Protocol, Outer VLAN 0xA0
    Protocol, Inner VLAN, Port 0x91
    Protocol, Inner VLAN 0x90
    Protocol, Ethertype, DstMAC, Port 0x8D
    Protocol, Ethertype, DstMAC 0x8C
    Protocol, Ethertype, Port 0x89
    Protocol, Ethertype 0x88
    Protocol, DstMAC, PFVF, Port 0x87
    Protocol, DstMAC, PFVF 0x86
    Protocol, DstMAC, Port 0x85
    Protocol, DstMAC 0x84
    Protocol, PFVF, Port 0x83
    Protocol, PFVF 0x82
    Protocol, Port 0x81
    Protocol 0x80
    TOS, Outer VLAN, Port 0x61
    TOS, Outer VLAN 0x60
    TOS, Inner VLAN, Port 0x51
    TOS, Inner VLAN 0x50
    TOS, Ethertype, DstMAC, Port 0x4D
    TOS, Ethertype, DstMAC 0x4C
    TOS, Ethertype, Port 0x49
    TOS, Ethertype 0x48
    TOS, DstMAC, PFVF, Port 0x47
    TOS, DstMAC, PFVF 0x46
    TOS, DstMAC, Port 0x45
    TOS, DstMAC 0x44
    TOS, PFVF, Port 0x43
    TOS, PFVF 0x42
    TOS, Port 0x41
    TOS 0x40
    Outer VLAN, Inner VLAN, Port 0x31
    Outer VLAN, Ethertype, Port 0x29
    Outer VLAN, Ethertype 0x28
    Outer VLAN, DstMAC, Port 0x25
    Outer VLAN, DstMAC 0x24
    Outer VLAN, Port 0x21
    Outer VLAN 0x20
    Inner VLAN, Ethertype, Port 0x19
    Inner VLAN, Ethertype 0x18
    Inner VLAN, DstMAC, Port 0x15
    Inner VLAN, DstMAC 0x14
    Inner VLAN, Port 0x11
    Inner VLAN 0x10
    Ethertype, DstMAC, Port 0xD
    Ethertype, DstMAC 0xC
    Ethertype, PFVF, Port 0xB
    Ethertype, PFVF 0xA
    Ethertype, Port 0x9
    Ethertype 0x8
    DstMAC, PFVF, Port 0x7
    DstMAC, PFVF 0x6
    DstMAC, Port 0x5
    Destination MAC 0x4
    PFVF, Port 0x3
    PFVF 0x2
    Physical Port 0x1

    For example, to enable matching ethertype field in Ethernet header, and protocol field in IPv4 header, the filtermode combination must be given as:

    testpmd -w 02:00.4,filtermode=0x88 -- -i
    
  • filtermask (default 0)

    filtermask devarg works similar to filtermode, but is used to configure a filter mode combination for HASH (exact-match) filters.

    Note

    The combination chosen for filtermask devarg must be a subset of the combination chosen for filtermode devarg.

    Default value of 0 makes driver pick the combination configured in the firmware configuration file on the adapter.

    Note that the filter rule will only be inserted in HASH region, if the rule contains all the fields specified in the filtermask combination. Otherwise, the filter rule will get inserted in LETCAM region.

    The same combination list explained in the tables in filtermode devarg section earlier applies for filtermask devarg, as well.

    For example, to enable matching only protocol field in IPv4 header, the filtermask combination must be given as:

    testpmd -w 02:00.4,filtermode=0x88,filtermask=0x80 -- -i
    

12.8. Driver compilation and testing

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

12.9. Linux

12.9.1. Linux Installation

Steps to manually install the latest firmware from the downloaded Chelsio Unified Wire package for Linux operating system are as follows:

  1. Load the kernel module:

    modprobe cxgb4
    
  2. Use ifconfig to get the interface name assigned to Chelsio card:

    ifconfig -a | grep "00:07:43"
    

    Example output:

    p1p1      Link encap:Ethernet  HWaddr 00:07:43:2D:EA:C0
    p1p2      Link encap:Ethernet  HWaddr 00:07:43:2D:EA:C8
    
  3. Install cxgbtool:

    cd <path_to_uwire>/tools/cxgbtool
    make install
    
  4. Use cxgbtool to load the firmware config file onto the card:

    cxgbtool p1p1 loadcfg <path_to_uwire>/src/network/firmware/t5-config.txt
    
  5. Use cxgbtool to load the firmware image onto the card:

    cxgbtool p1p1 loadfw <path_to_uwire>/src/network/firmware/t5fw-*.bin
    
  6. Unload and reload the kernel module:

    modprobe -r cxgb4
    modprobe cxgb4
    
  7. Verify with ethtool:

    ethtool -i p1p1 | grep "firmware"
    

    Example output:

    firmware-version: 1.24.11.0, TP 0.1.23.2
    

12.9.2. Running testpmd

This section demonstrates how to launch testpmd with Chelsio devices managed by librte_pmd_cxgbe in Linux operating system.

  1. Load the kernel module:

    modprobe cxgb4
    
  2. Get the PCI bus addresses of the interfaces bound to cxgb4 driver:

    dmesg | tail -2
    

    Example output:

    cxgb4 0000:02:00.4 p1p1: renamed from eth0
    cxgb4 0000:02:00.4 p1p2: renamed from eth1
    

    Note

    Both the interfaces of a Chelsio 2-port adapter are bound to the same PCI bus address.

  3. Unload the kernel module:

    modprobe -ar cxgb4 csiostor
    
  4. Running testpmd

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

    Note

    Currently, CXGBE PMD only supports the binding of PF4 for Chelsio NICs.

    Example output:

    [...]
    EAL: PCI device 0000:02:00.4 on NUMA socket -1
    EAL:   probe driver: 1425:5401 rte_cxgbe_pmd
    EAL:   PCI memory mapped at 0x7fd7c0200000
    EAL:   PCI memory mapped at 0x7fd77cdfd000
    EAL:   PCI memory mapped at 0x7fd7c10b7000
    PMD: rte_cxgbe_pmd: fw: 1.24.11.0, TP: 0.1.23.2
    PMD: rte_cxgbe_pmd: Coming up as MASTER: Initializing adapter
    Interactive-mode selected
    Configuring Port 0 (socket 0)
    Port 0: 00:07:43:2D:EA:C0
    Configuring Port 1 (socket 0)
    Port 1: 00:07:43:2D:EA:C8
    Checking link statuses...
    PMD: rte_cxgbe_pmd: Port0: passive DA port module inserted
    PMD: rte_cxgbe_pmd: Port1: passive DA port module inserted
    Port 0 Link Up - speed 10000 Mbps - full-duplex
    Port 1 Link Up - speed 10000 Mbps - full-duplex
    Done
    testpmd>
    

    Note

    Flow control pause TX/RX is disabled by default and can be enabled via testpmd. Refer section Enable/Disable Flow Control for more details.

12.9.3. Configuring SR-IOV Virtual Functions

This section demonstrates how to enable SR-IOV virtual functions on Chelsio NICs and demonstrates how to run testpmd with SR-IOV virtual functions.

  1. Load the kernel module:

    modprobe cxgb4
    
  2. Get the PCI bus addresses of the interfaces bound to cxgb4 driver:

    dmesg | tail -2
    

    Example output:

    cxgb4 0000:02:00.4 p1p1: renamed from eth0
    cxgb4 0000:02:00.4 p1p2: renamed from eth1
    

    Note

    Both the interfaces of a Chelsio 2-port adapter are bound to the same PCI bus address.

  3. Use ifconfig to get the interface name assigned to Chelsio card:

    ifconfig -a | grep "00:07:43"
    

    Example output:

    p1p1      Link encap:Ethernet  HWaddr 00:07:43:2D:EA:C0
    p1p2      Link encap:Ethernet  HWaddr 00:07:43:2D:EA:C8
    
  4. Bring up the interfaces:

    ifconfig p1p1 up
    ifconfig p1p2 up
    
  5. Instantiate SR-IOV Virtual Functions. PF0..3 can be used for SR-IOV VFs. Multiple VFs can be instantiated on each of PF0..3. To instantiate one SR-IOV VF on each PF0 and PF1:

    echo 1 > /sys/bus/pci/devices/0000\:02\:00.0/sriov_numvfs
    echo 1 > /sys/bus/pci/devices/0000\:02\:00.1/sriov_numvfs
    
  6. Get the PCI bus addresses of the virtual functions:

    lspci | grep -i "Chelsio" | grep -i "VF"
    

    Example output:

    02:01.0 Ethernet controller: Chelsio Communications Inc T540-CR Unified Wire Ethernet Controller [VF]
    02:01.1 Ethernet controller: Chelsio Communications Inc T540-CR Unified Wire Ethernet Controller [VF]
    
  7. Running testpmd

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

    Example output:

    [...]
    EAL: PCI device 0000:02:01.0 on NUMA socket 0
    EAL:   probe driver: 1425:5803 net_cxgbevf
    PMD: rte_cxgbe_pmd: Firmware version: 1.24.11.0
    PMD: rte_cxgbe_pmd: TP Microcode version: 0.1.23.2
    PMD: rte_cxgbe_pmd: Chelsio rev 0
    PMD: rte_cxgbe_pmd: No bootstrap loaded
    PMD: rte_cxgbe_pmd: No Expansion ROM loaded
    PMD: rte_cxgbe_pmd:  0000:02:01.0 Chelsio rev 0 1G/10GBASE-SFP
    EAL: PCI device 0000:02:01.1 on NUMA socket 0
    EAL:   probe driver: 1425:5803 net_cxgbevf
    PMD: rte_cxgbe_pmd: Firmware version: 1.24.11.0
    PMD: rte_cxgbe_pmd: TP Microcode version: 0.1.23.2
    PMD: rte_cxgbe_pmd: Chelsio rev 0
    PMD: rte_cxgbe_pmd: No bootstrap loaded
    PMD: rte_cxgbe_pmd: No Expansion ROM loaded
    PMD: rte_cxgbe_pmd:  0000:02:01.1 Chelsio rev 0 1G/10GBASE-SFP
    Configuring Port 0 (socket 0)
    Port 0: 06:44:29:44:40:00
    Configuring Port 1 (socket 0)
    Port 1: 06:44:29:44:40:10
    Checking link statuses...
    Done
    testpmd>
    

12.10. FreeBSD

12.10.1. FreeBSD Installation

Steps to manually install the latest firmware from the downloaded Chelsio Unified Wire package for FreeBSD operating system are as follows:

  1. Load the kernel module:

    kldload if_cxgbe
    
  2. Use dmesg to get the t5nex instance assigned to the Chelsio card:

    dmesg | grep "t5nex"
    

    Example output:

    t5nex0: <Chelsio T520-CR> irq 16 at device 0.4 on pci2
    cxl0: <port 0> on t5nex0
    cxl1: <port 1> on t5nex0
    t5nex0: PCIe x8, 2 ports, 14 MSI-X interrupts, 31 eq, 13 iq
    

    In the example above, a Chelsio T520-CR card is bound to a t5nex0 instance.

  3. Install cxgbetool from FreeBSD source repository:

    cd <path_to_FreeBSD_source>/tools/tools/cxgbetool/
    make && make install
    
  4. Use cxgbetool to load the firmware image onto the card:

    cxgbetool t5nex0 loadfw <path_to_uwire>/src/network/firmware/t5fw-*.bin
    
  5. Unload and reload the kernel module:

    kldunload if_cxgbe
    kldload if_cxgbe
    
  6. Verify with sysctl:

    sysctl -a | grep "t5nex" | grep "firmware"
    

    Example output:

    dev.t5nex.0.firmware_version: 1.24.11.0
    

12.10.2. Running testpmd

This section demonstrates how to launch testpmd with Chelsio devices managed by librte_pmd_cxgbe in FreeBSD operating system.

  1. Change to DPDK source directory where the target has been compiled in section Driver compilation and testing:

    cd <DPDK-source-directory>
    
  2. Copy the contigmem kernel module to /boot/kernel directory:

    cp x86_64-native-freebsd-clang/kmod/contigmem.ko /boot/kernel/
    
  3. Add the following lines to /boot/loader.conf:

    # reserve 2 x 1G blocks of contiguous memory using contigmem driver
    hw.contigmem.num_buffers=2
    hw.contigmem.buffer_size=1073741824
    # load contigmem module during boot process
    contigmem_load="YES"
    

    The above lines load the contigmem kernel module during boot process and allocate 2 x 1G blocks of contiguous memory to be used for DPDK later on. This is to avoid issues with potential memory fragmentation during later system up time, which may result in failure of allocating the contiguous memory required for the contigmem kernel module.

  4. Restart the system and ensure the contigmem module is loaded successfully:

    reboot
    kldstat | grep "contigmem"
    

    Example output:

    2    1 0xffffffff817f1000 3118     contigmem.ko
    
  5. Repeat step 1 to ensure that you are in the DPDK source directory.

  6. Load the cxgbe kernel module:

    kldload if_cxgbe
    
  7. Get the PCI bus addresses of the interfaces bound to t5nex driver:

    pciconf -l | grep "t5nex"
    

    Example output:

    t5nex0@pci0:2:0:4: class=0x020000 card=0x00001425 chip=0x54011425 rev=0x00
    

    In the above example, the t5nex0 is bound to 2:0:4 bus address.

    Note

    Both the interfaces of a Chelsio 2-port adapter are bound to the same PCI bus address.

  8. Unload the kernel module:

    kldunload if_cxgbe
    
  9. Set the PCI bus addresses to hw.nic_uio.bdfs kernel environment parameter:

    kenv hw.nic_uio.bdfs="2:0:4"
    

    This automatically binds 2:0:4 to nic_uio kernel driver when it is loaded in the next step.

    Note

    Currently, CXGBE PMD only supports the binding of PF4 for Chelsio NICs.

  10. Load nic_uio kernel driver:

    kldload ./x86_64-native-freebsd-clang/kmod/nic_uio.ko
    
  11. Start testpmd with basic parameters:

    ./x86_64-native-freebsd-clang/app/testpmd -l 0-3 -n 4 -w 0000:02:00.4 -- -i
    

    Example output:

    [...]
    EAL: PCI device 0000:02:00.4 on NUMA socket 0
    EAL:   probe driver: 1425:5401 rte_cxgbe_pmd
    EAL:   PCI memory mapped at 0x8007ec000
    EAL:   PCI memory mapped at 0x842800000
    EAL:   PCI memory mapped at 0x80086c000
    PMD: rte_cxgbe_pmd: fw: 1.24.11.0, TP: 0.1.23.2
    PMD: rte_cxgbe_pmd: Coming up as MASTER: Initializing adapter
    Interactive-mode selected
    Configuring Port 0 (socket 0)
    Port 0: 00:07:43:2D:EA:C0
    Configuring Port 1 (socket 0)
    Port 1: 00:07:43:2D:EA:C8
    Checking link statuses...
    PMD: rte_cxgbe_pmd: Port0: passive DA port module inserted
    PMD: rte_cxgbe_pmd: Port1: passive DA port module inserted
    Port 0 Link Up - speed 10000 Mbps - full-duplex
    Port 1 Link Up - speed 10000 Mbps - full-duplex
    Done
    testpmd>
    

Note

Flow control pause TX/RX is disabled by default and can be enabled via testpmd. Refer section Enable/Disable Flow Control for more details.

12.11. Sample Application Notes

12.11.1. Enable/Disable Flow Control

Flow control pause TX/RX is disabled by default and can be enabled via testpmd as follows:

testpmd> set flow_ctrl rx on tx on 0 0 0 0 mac_ctrl_frame_fwd off autoneg on 0
testpmd> set flow_ctrl rx on tx on 0 0 0 0 mac_ctrl_frame_fwd off autoneg on 1

To disable again, run:

testpmd> set flow_ctrl rx off tx off 0 0 0 0 mac_ctrl_frame_fwd off autoneg off 0
testpmd> set flow_ctrl rx off tx off 0 0 0 0 mac_ctrl_frame_fwd off autoneg off 1

12.11.2. Jumbo Mode

There are two ways to enable sending and receiving of jumbo frames via testpmd. One method involves using the mtu command, which changes the mtu of an individual port without having to stop the selected port. Another method involves stopping all the ports first and then running max-pkt-len command to configure the mtu of all the ports with a single command.

  • To configure each port individually, run the mtu command as follows:

    testpmd> port config mtu 0 9000
    testpmd> port config mtu 1 9000
    
  • To configure all the ports at once, stop all the ports first and run the max-pkt-len command as follows:

    testpmd> port stop all
    testpmd> port config all max-pkt-len 9000