1 .. SPDX-License-Identifier: BSD-3-Clause
2 Copyright(c) 2010-2016 Intel Corporation.
6 Testpmd Runtime Functions
7 =========================
9 Where the testpmd application is started in interactive mode, (``-i|--interactive``),
10 it displays a prompt that can be used to start and stop forwarding,
11 configure the application, display statistics (including the extended NIC
12 statistics aka xstats) , set the Flow Director and other tasks::
16 The testpmd prompt has some, limited, readline support.
17 Common bash command-line functions such as ``Ctrl+a`` and ``Ctrl+e`` to go to the start and end of the prompt line are supported
18 as well as access to the command history via the up-arrow.
20 There is also support for tab completion.
21 If you type a partial command and hit ``<TAB>`` you get a list of the available completions:
23 .. code-block:: console
25 testpmd> show port <TAB>
27 info [Mul-choice STRING]: show|clear port info|stats|xstats|fdir|stat_qmap|dcb_tc|cap X
28 info [Mul-choice STRING]: show|clear port info|stats|xstats|fdir|stat_qmap|dcb_tc|cap all
29 stats [Mul-choice STRING]: show|clear port info|stats|xstats|fdir|stat_qmap|dcb_tc|cap X
30 stats [Mul-choice STRING]: show|clear port info|stats|xstats|fdir|stat_qmap|dcb_tc|cap all
36 Some examples in this document are too long to fit on one line are shown wrapped at `"\\"` for display purposes::
38 testpmd> set flow_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \
39 (pause_time) (send_xon) (port_id)
41 In the real ``testpmd>`` prompt these commands should be on a single line.
46 The testpmd has on-line help for the functions that are available at runtime.
47 These are divided into sections and can be accessed using help, help section or help all:
49 .. code-block:: console
53 help control : Start and stop forwarding.
54 help display : Displaying port, stats and config information.
55 help config : Configuration information.
56 help ports : Configuring ports.
57 help registers : Reading and setting port registers.
58 help filters : Filters configuration help.
59 help all : All of the above sections.
62 Command File Functions
63 ----------------------
65 To facilitate loading large number of commands or to avoid cutting and pasting where not
66 practical or possible testpmd supports alternative methods for executing commands.
68 * If started with the ``--cmdline-file=FILENAME`` command line argument testpmd
69 will execute all CLI commands contained within the file immediately before
70 starting packet forwarding or entering interactive mode.
72 .. code-block:: console
74 ./testpmd -n4 -r2 ... -- -i --cmdline-file=/home/ubuntu/flow-create-commands.txt
75 Interactive-mode selected
76 CLI commands to be read from /home/ubuntu/flow-create-commands.txt
77 Configuring Port 0 (socket 0)
78 Port 0: 7C:FE:90:CB:74:CE
79 Configuring Port 1 (socket 0)
80 Port 1: 7C:FE:90:CB:74:CA
81 Checking link statuses...
82 Port 0 Link Up - speed 10000 Mbps - full-duplex
83 Port 1 Link Up - speed 10000 Mbps - full-duplex
89 Flow rule #498 created
90 Flow rule #499 created
91 Read all CLI commands from /home/ubuntu/flow-create-commands.txt
95 * At run-time additional commands can be loaded in bulk by invoking the ``load FILENAME``
98 .. code-block:: console
100 testpmd> load /home/ubuntu/flow-create-commands.txt
105 Flow rule #498 created
106 Flow rule #499 created
107 Read all CLI commands from /home/ubuntu/flow-create-commands.txt
111 In all cases output from any included command will be displayed as standard output.
112 Execution will continue until the end of the file is reached regardless of
113 whether any errors occur. The end user must examine the output to determine if
114 any failures occurred.
123 Start packet forwarding with current configuration::
130 Start packet forwarding with current configuration after sending specified number of bursts of packets::
132 testpmd> start tx_first (""|burst_num)
134 The default burst number is 1 when ``burst_num`` not presented.
139 Stop packet forwarding, and display accumulated statistics::
154 The functions in the following sections are used to display information about the
155 testpmd configuration or the NIC status.
160 Display information for a given port or all ports::
162 testpmd> show port (info|summary|stats|xstats|fdir|stat_qmap|dcb_tc|cap) (port_id|all)
164 The available information categories are:
166 * ``info``: General port information such as MAC address.
168 * ``summary``: Brief port summary such as Device Name, Driver Name etc.
170 * ``stats``: RX/TX statistics.
172 * ``xstats``: RX/TX extended NIC statistics.
174 * ``fdir``: Flow Director information and statistics.
176 * ``stat_qmap``: Queue statistics mapping.
178 * ``dcb_tc``: DCB information such as TC mapping.
180 * ``cap``: Supported offload capabilities.
184 .. code-block:: console
186 testpmd> show port info 0
188 ********************* Infos for port 0 *********************
190 MAC address: XX:XX:XX:XX:XX:XX
192 memory allocation on the socket: 0
194 Link speed: 40000 Mbps
195 Link duplex: full-duplex
196 Promiscuous mode: enabled
197 Allmulticast mode: disabled
198 Maximum number of MAC addresses: 64
199 Maximum number of MAC addresses of hash filtering: 0
201 strip on, filter on, extend off, qinq strip off
202 Redirection table size: 512
203 Supported flow types:
221 show port (module_eeprom|eeprom)
222 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
224 Display the EEPROM information of a port::
226 testpmd> show port (port_id) (module_eeprom|eeprom)
231 Display the rss redirection table entry indicated by masks on port X::
233 testpmd> show port (port_id) rss reta (size) (mask0, mask1...)
235 size is used to indicate the hardware supported reta size
240 Display the RSS hash functions and RSS hash key of a port::
242 testpmd> show port (port_id) rss-hash [key]
247 Clear the port statistics and forward engine statistics for a given port or for all ports::
249 testpmd> clear port (info|stats|xstats|fdir|stat_qmap) (port_id|all)
253 testpmd> clear port stats all
258 Display information for a given port's RX/TX queue::
260 testpmd> show (rxq|txq) info (port_id) (queue_id)
262 show desc status(rxq|txq)
263 ~~~~~~~~~~~~~~~~~~~~~~~~~
265 Display information for a given port's RX/TX descriptor status::
267 testpmd> show port (port_id) (rxq|txq) (queue_id) desc (desc_id) status
273 Displays the configuration of the application.
274 The configuration comes from the command-line, the runtime or the application defaults::
276 testpmd> show config (rxtx|cores|fwd|txpkts|txtimes)
278 The available information categories are:
280 * ``rxtx``: RX/TX configuration items.
282 * ``cores``: List of forwarding cores.
284 * ``fwd``: Packet forwarding configuration.
286 * ``txpkts``: Packets to TX configuration.
288 * ``txtimes``: Burst time pattern for Tx only mode.
292 .. code-block:: console
294 testpmd> show config rxtx
296 io packet forwarding - CRC stripping disabled - packets/burst=16
297 nb forwarding cores=2 - nb forwarding ports=1
298 RX queues=1 - RX desc=128 - RX free threshold=0
299 RX threshold registers: pthresh=8 hthresh=8 wthresh=4
300 TX queues=1 - TX desc=512 - TX free threshold=0
301 TX threshold registers: pthresh=36 hthresh=0 wthresh=0
302 TX RS bit threshold=0 - TXQ flags=0x0
307 Set the packet forwarding mode::
309 testpmd> set fwd (io|mac|macswap|flowgen| \
310 rxonly|txonly|csum|icmpecho|noisy|5tswap) (""|retry)
312 ``retry`` can be specified for forwarding engines except ``rx_only``.
314 The available information categories are:
316 * ``io``: Forwards packets "as-is" in I/O mode.
317 This is the fastest possible forwarding operation as it does not access packets data.
318 This is the default mode.
320 * ``mac``: Changes the source and the destination Ethernet addresses of packets before forwarding them.
321 Default application behavior is to set source Ethernet address to that of the transmitting interface, and destination
322 address to a dummy value (set during init). The user may specify a target destination Ethernet address via the 'eth-peer' or
323 'eth-peers-configfile' command-line options. It is not currently possible to specify a specific source Ethernet address.
325 * ``macswap``: MAC swap forwarding mode.
326 Swaps the source and the destination Ethernet addresses of packets before forwarding them.
328 * ``flowgen``: Multi-flow generation mode.
329 Originates a number of flows (with varying destination IP addresses), and terminate receive traffic.
331 * ``rxonly``: Receives packets but doesn't transmit them.
333 * ``txonly``: Generates and transmits packets without receiving any.
335 * ``csum``: Changes the checksum field with hardware or software methods depending on the offload flags on the packet.
337 * ``icmpecho``: Receives a burst of packets, lookup for ICMP echo requests and, if any, send back ICMP echo replies.
339 * ``ieee1588``: Demonstrate L2 IEEE1588 V2 PTP timestamping for RX and TX. Requires ``CONFIG_RTE_LIBRTE_IEEE1588=y``.
341 * ``noisy``: Noisy neighbor simulation.
342 Simulate more realistic behavior of a guest machine engaged in receiving
343 and sending packets performing Virtual Network Function (VNF).
345 * ``5tswap``: Swap the source and destination of L2,L3,L4 if they exist.
347 L2 swaps the source address and destination address of Ethernet, as same as ``macswap``.
349 L3 swaps the source address and destination address of IP (v4 and v6).
351 L4 swaps the source port and destination port of transport layer (TCP and UDP).
355 testpmd> set fwd rxonly
357 Set rxonly packet forwarding mode
363 When running, forwarding engines maintain statistics from the time they have been started.
364 Example for the io forwarding engine, with some packet drops on the tx side::
366 testpmd> show fwd stats all
368 ------- Forward Stats for RX Port= 0/Queue= 0 -> TX Port= 1/Queue= 0 -------
369 RX-packets: 274293770 TX-packets: 274293642 TX-dropped: 128
371 ------- Forward Stats for RX Port= 1/Queue= 0 -> TX Port= 0/Queue= 0 -------
372 RX-packets: 274301850 TX-packets: 274301850 TX-dropped: 0
374 ---------------------- Forward statistics for port 0 ----------------------
375 RX-packets: 274293802 RX-dropped: 0 RX-total: 274293802
376 TX-packets: 274301862 TX-dropped: 0 TX-total: 274301862
377 ----------------------------------------------------------------------------
379 ---------------------- Forward statistics for port 1 ----------------------
380 RX-packets: 274301894 RX-dropped: 0 RX-total: 274301894
381 TX-packets: 274293706 TX-dropped: 128 TX-total: 274293834
382 ----------------------------------------------------------------------------
384 +++++++++++++++ Accumulated forward statistics for all ports+++++++++++++++
385 RX-packets: 548595696 RX-dropped: 0 RX-total: 548595696
386 TX-packets: 548595568 TX-dropped: 128 TX-total: 548595696
387 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
391 Enabling CONFIG_RTE_TEST_PMD_RECORD_CORE_CYCLES appends "CPU cycles/packet" stats, like:
393 CPU cycles/packet=xx.dd (total cycles=xxxx / total RX packets=xxxx) at xxx MHz clock
398 Clear the forwarding engines statistics::
400 testpmd> clear fwd stats all
405 Display an RX descriptor for a port RX queue::
407 testpmd> read rxd (port_id) (queue_id) (rxd_id)
411 testpmd> read rxd 0 0 4
412 0x0000000B - 0x001D0180 / 0x0000000B - 0x001D0180
417 Display a TX descriptor for a port TX queue::
419 testpmd> read txd (port_id) (queue_id) (txd_id)
423 testpmd> read txd 0 0 4
424 0x00000001 - 0x24C3C440 / 0x000F0000 - 0x2330003C
429 Get loaded dynamic device personalization (DDP) package info list::
431 testpmd> ddp get list (port_id)
436 Display information about dynamic device personalization (DDP) profile::
438 testpmd> ddp get info (profile_path)
443 Display VF statistics::
445 testpmd> show vf stats (port_id) (vf_id)
450 Reset VF statistics::
452 testpmd> clear vf stats (port_id) (vf_id)
454 show port pctype mapping
455 ~~~~~~~~~~~~~~~~~~~~~~~~
457 List all items from the pctype mapping table::
459 testpmd> show port (port_id) pctype mapping
461 show rx offloading capabilities
462 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
464 List all per queue and per port Rx offloading capabilities of a port::
466 testpmd> show port (port_id) rx_offload capabilities
468 show rx offloading configuration
469 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
471 List port level and all queue level Rx offloading configuration::
473 testpmd> show port (port_id) rx_offload configuration
475 show tx offloading capabilities
476 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
478 List all per queue and per port Tx offloading capabilities of a port::
480 testpmd> show port (port_id) tx_offload capabilities
482 show tx offloading configuration
483 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
485 List port level and all queue level Tx offloading configuration::
487 testpmd> show port (port_id) tx_offload configuration
489 show tx metadata setting
490 ~~~~~~~~~~~~~~~~~~~~~~~~
492 Show Tx metadata value set for a specific port::
494 testpmd> show port (port_id) tx_metadata
496 show port supported ptypes
497 ~~~~~~~~~~~~~~~~~~~~~~~~~~
499 Show ptypes supported for a specific port::
501 testpmd> show port (port_id) ptypes
503 set port supported ptypes
504 ~~~~~~~~~~~~~~~~~~~~~~~~~
506 set packet types classification for a specific port::
508 testpmd> set port (port_id) ptypes_mask (mask)
510 show port mac addresses info
511 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
513 Show mac addresses added for a specific port::
515 testpmd> show port (port_id) macs
518 show port multicast mac addresses info
519 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
521 Show multicast mac addresses added for a specific port::
523 testpmd> show port (port_id) mcast_macs
528 Show general information about devices probed::
530 testpmd> show device info (<identifier>|all)
534 .. code-block:: console
536 testpmd> show device info net_pcap0
538 ********************* Infos for device net_pcap0 *********************
540 Driver name: net_pcap
541 Devargs: iface=enP2p6s0,phy_mac=1
542 Connect to socket: -1
545 MAC address: 1E:37:93:28:04:B8
546 Device name: net_pcap0
551 Dumps all physical memory segment layouts::
553 testpmd> dump_physmem
558 Dumps the layout of all memory zones::
560 testpmd> dump_memzone
565 Dumps the memory usage of all sockets::
567 testpmd> dump_socket_mem
572 Dumps the size of all memory structures::
574 testpmd> dump_struct_sizes
579 Dumps the status of all or specific element in DPDK rings::
581 testpmd> dump_ring [ring_name]
586 Dumps the statistics of all or specific memory pool::
588 testpmd> dump_mempool [mempool_name]
593 Dumps the user device list::
595 testpmd> dump_devargs
600 Dumps the log level for all the dpdk modules::
602 testpmd> dump_log_types
604 show (raw_encap|raw_decap)
605 ~~~~~~~~~~~~~~~~~~~~~~~~~~
607 Display content of raw_encap/raw_decap buffers in hex::
609 testpmd> show <raw_encap|raw_decap> <index>
610 testpmd> show <raw_encap|raw_decap> all
614 testpmd> show raw_encap 6
616 index: 6 at [0x1c565b0], len=50
617 00000000: 00 00 00 00 00 00 16 26 36 46 56 66 08 00 45 00 | .......&6FVf..E.
618 00000010: 00 00 00 00 00 00 00 11 00 00 C0 A8 01 06 C0 A8 | ................
619 00000020: 03 06 00 00 00 FA 00 00 00 00 08 00 00 00 00 00 | ................
623 Configuration Functions
624 -----------------------
626 The testpmd application can be configured from the runtime as well as from the command-line.
628 This section details the available configuration functions that are available.
632 Configuration changes only become active when forwarding is started/restarted.
637 Reset forwarding to the default configuration::
644 Set the debug verbosity level::
646 testpmd> set verbose (level)
648 Available levels are as following:
650 * ``0`` silent except for error.
651 * ``1`` fully verbose except for Tx packets.
652 * ``2`` fully verbose except for Rx packets.
653 * ``> 2`` fully verbose.
658 Set the log level for a log type::
660 testpmd> set log global|(type) (level)
664 * ``type`` is the log name.
666 * ``level`` is the log level.
668 For example, to change the global log level::
670 testpmd> set log global (level)
672 Regexes can also be used for type. To change log level of user1, user2 and user3::
674 testpmd> set log user[1-3] (level)
679 Set the number of ports used by the application:
683 This is equivalent to the ``--nb-ports`` command-line option.
688 Set the number of cores used by the application::
690 testpmd> set nbcore (num)
692 This is equivalent to the ``--nb-cores`` command-line option.
696 The number of cores used must not be greater than number of ports used multiplied by the number of queues per port.
701 Set the forwarding cores hexadecimal mask::
703 testpmd> set coremask (mask)
705 This is equivalent to the ``--coremask`` command-line option.
709 The master lcore is reserved for command line parsing only and cannot be masked on for packet forwarding.
714 Set the forwarding ports hexadecimal mask::
716 testpmd> set portmask (mask)
718 This is equivalent to the ``--portmask`` command-line option.
720 set record-core-cycles
721 ~~~~~~~~~~~~~~~~~~~~~~
723 Set the recording of CPU cycles::
725 testpmd> set record-core-cycles (on|off)
729 * ``on`` enables measurement of CPU cycles per packet.
731 * ``off`` disables measurement of CPU cycles per packet.
733 This is equivalent to the ``--record-core-cycles command-line`` option.
735 set record-burst-stats
736 ~~~~~~~~~~~~~~~~~~~~~~
738 Set the displaying of RX and TX bursts::
740 testpmd> set record-burst-stats (on|off)
744 * ``on`` enables display of RX and TX bursts.
746 * ``off`` disables display of RX and TX bursts.
748 This is equivalent to the ``--record-burst-stats command-line`` option.
753 Set number of packets per burst::
755 testpmd> set burst (num)
757 This is equivalent to the ``--burst command-line`` option.
759 When retry is enabled, the transmit delay time and number of retries can also be set::
761 testpmd> set burst tx delay (microseconds) retry (num)
766 Set the length of each segment of the TX-ONLY packets or length of packet for FLOWGEN mode::
768 testpmd> set txpkts (x[,y]*)
770 Where x[,y]* represents a CSV list of values, without white space.
775 Configure the timing burst pattern for Tx only mode. This command enables
776 the packet send scheduling on dynamic timestamp mbuf field and configures
777 timing pattern in Tx only mode. In this mode, if scheduling is enabled
778 application provides timestamps in the packets being sent. It is possible
779 to configure delay (in unspecified device clock units) between bursts
780 and between the packets within the burst::
782 testpmd> set txtimes (inter),(intra)
786 * ``inter`` is the delay between the bursts in the device clock units.
787 If ``intra`` is zero, this is the time between the beginnings of the
788 first packets in the neighbour bursts, if ``intra`` is not zero,
789 ``inter`` specifies the time between the beginning of the first packet
790 of the current burst and the beginning of the last packet of the
791 previous burst. If ``inter`` parameter is zero the send scheduling
792 on timestamps is disabled (default).
794 * ``intra`` is the delay between the packets within the burst specified
795 in the device clock units. The number of packets in the burst is defined
796 by regular burst setting. If ``intra`` parameter is zero no timestamps
797 provided in the packets excepting the first one in the burst.
799 As the result the bursts of packet will be transmitted with specific
800 delays between the packets within the burst and specific delay between
801 the bursts. The rte_eth_read_clock() must be supported by the device(s)
802 and is supposed to be engaged to get the current device clock value
803 and provide the reference for the timestamps. If there is no supported
804 rte_eth_read_clock() there will be no send scheduling provided on the port.
809 Set the split policy for the TX packets, applicable for TX-ONLY and CSUM forwarding modes::
811 testpmd> set txsplit (off|on|rand)
815 * ``off`` disable packet copy & split for CSUM mode.
817 * ``on`` split outgoing packet into multiple segments. Size of each segment
818 and number of segments per packet is determined by ``set txpkts`` command
821 * ``rand`` same as 'on', but number of segments per each packet is a random value between 1 and total number of segments.
826 Set the list of forwarding cores::
828 testpmd> set corelist (x[,y]*)
830 For example, to change the forwarding cores:
832 .. code-block:: console
834 testpmd> set corelist 3,1
835 testpmd> show config fwd
837 io packet forwarding - ports=2 - cores=2 - streams=2 - NUMA support disabled
838 Logical Core 3 (socket 0) forwards packets on 1 streams:
839 RX P=0/Q=0 (socket 0) -> TX P=1/Q=0 (socket 0) peer=02:00:00:00:00:01
840 Logical Core 1 (socket 0) forwards packets on 1 streams:
841 RX P=1/Q=0 (socket 0) -> TX P=0/Q=0 (socket 0) peer=02:00:00:00:00:00
845 The cores are used in the same order as specified on the command line.
850 Set the list of forwarding ports::
852 testpmd> set portlist (x[,y]*)
854 For example, to change the port forwarding:
856 .. code-block:: console
858 testpmd> set portlist 0,2,1,3
859 testpmd> show config fwd
861 io packet forwarding - ports=4 - cores=1 - streams=4
862 Logical Core 3 (socket 0) forwards packets on 4 streams:
863 RX P=0/Q=0 (socket 0) -> TX P=2/Q=0 (socket 0) peer=02:00:00:00:00:01
864 RX P=2/Q=0 (socket 0) -> TX P=0/Q=0 (socket 0) peer=02:00:00:00:00:00
865 RX P=1/Q=0 (socket 0) -> TX P=3/Q=0 (socket 0) peer=02:00:00:00:00:03
866 RX P=3/Q=0 (socket 0) -> TX P=1/Q=0 (socket 0) peer=02:00:00:00:00:02
871 Select how to retrieve new ports created after "port attach" command::
873 testpmd> set port setup on (iterator|event)
875 For each new port, a setup is done.
876 It will find the probed ports via RTE_ETH_FOREACH_MATCHING_DEV loop
877 in iterator mode, or via RTE_ETH_EVENT_NEW in event mode.
882 Enable/disable tx loopback::
884 testpmd> set tx loopback (port_id) (on|off)
889 set drop enable bit for all queues::
891 testpmd> set all queues drop (port_id) (on|off)
893 set split drop enable (for VF)
894 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
896 set split drop enable bit for VF from PF::
898 testpmd> set vf split drop (port_id) (vf_id) (on|off)
900 set mac antispoof (for VF)
901 ~~~~~~~~~~~~~~~~~~~~~~~~~~
903 Set mac antispoof for a VF from the PF::
905 testpmd> set vf mac antispoof (port_id) (vf_id) (on|off)
910 Enable/disable MACsec offload::
912 testpmd> set macsec offload (port_id) on encrypt (on|off) replay-protect (on|off)
913 testpmd> set macsec offload (port_id) off
918 Configure MACsec secure connection (SC)::
920 testpmd> set macsec sc (tx|rx) (port_id) (mac) (pi)
924 The pi argument is ignored for tx.
925 Check the NIC Datasheet for hardware limits.
930 Configure MACsec secure association (SA)::
932 testpmd> set macsec sa (tx|rx) (port_id) (idx) (an) (pn) (key)
936 The IDX value must be 0 or 1.
937 Check the NIC Datasheet for hardware limits.
939 set broadcast mode (for VF)
940 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
942 Set broadcast mode for a VF from the PF::
944 testpmd> set vf broadcast (port_id) (vf_id) (on|off)
949 Set the VLAN strip for a queue on a port::
951 testpmd> vlan set stripq (on|off) (port_id,queue_id)
953 vlan set stripq (for VF)
954 ~~~~~~~~~~~~~~~~~~~~~~~~
956 Set VLAN strip for all queues in a pool for a VF from the PF::
958 testpmd> set vf vlan stripq (port_id) (vf_id) (on|off)
960 vlan set insert (for VF)
961 ~~~~~~~~~~~~~~~~~~~~~~~~
963 Set VLAN insert for a VF from the PF::
965 testpmd> set vf vlan insert (port_id) (vf_id) (vlan_id)
967 vlan set tag (for VF)
968 ~~~~~~~~~~~~~~~~~~~~~
970 Set VLAN tag for a VF from the PF::
972 testpmd> set vf vlan tag (port_id) (vf_id) (on|off)
974 vlan set antispoof (for VF)
975 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
977 Set VLAN antispoof for a VF from the PF::
979 testpmd> set vf vlan antispoof (port_id) (vf_id) (on|off)
981 vlan set (strip|filter|qinq_strip|extend)
982 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
983 Set the VLAN strip/filter/QinQ strip/extend on for a port::
985 testpmd> vlan set (strip|filter|qinq_strip|extend) (on|off) (port_id)
990 Set the inner or outer VLAN TPID for packet filtering on a port::
992 testpmd> vlan set (inner|outer) tpid (value) (port_id)
996 TPID value must be a 16-bit number (value <= 65536).
1001 Add a VLAN ID, or all identifiers, to the set of VLAN identifiers filtered by port ID::
1003 testpmd> rx_vlan add (vlan_id|all) (port_id)
1007 VLAN filter must be set on that port. VLAN ID < 4096.
1008 Depending on the NIC used, number of vlan_ids may be limited to the maximum entries
1009 in VFTA table. This is important if enabling all vlan_ids.
1014 Remove a VLAN ID, or all identifiers, from the set of VLAN identifiers filtered by port ID::
1016 testpmd> rx_vlan rm (vlan_id|all) (port_id)
1018 rx_vlan add (for VF)
1019 ~~~~~~~~~~~~~~~~~~~~
1021 Add a VLAN ID, to the set of VLAN identifiers filtered for VF(s) for port ID::
1023 testpmd> rx_vlan add (vlan_id) port (port_id) vf (vf_mask)
1028 Remove a VLAN ID, from the set of VLAN identifiers filtered for VF(s) for port ID::
1030 testpmd> rx_vlan rm (vlan_id) port (port_id) vf (vf_mask)
1035 Add a tunnel filter on a port::
1037 testpmd> tunnel_filter add (port_id) (outer_mac) (inner_mac) (ip_addr) \
1038 (inner_vlan) (vxlan|nvgre|ipingre|vxlan-gpe) (imac-ivlan|imac-ivlan-tenid|\
1039 imac-tenid|imac|omac-imac-tenid|oip|iip) (tenant_id) (queue_id)
1041 The available information categories are:
1043 * ``vxlan``: Set tunnel type as VXLAN.
1045 * ``nvgre``: Set tunnel type as NVGRE.
1047 * ``ipingre``: Set tunnel type as IP-in-GRE.
1049 * ``vxlan-gpe``: Set tunnel type as VXLAN-GPE
1051 * ``imac-ivlan``: Set filter type as Inner MAC and VLAN.
1053 * ``imac-ivlan-tenid``: Set filter type as Inner MAC, VLAN and tenant ID.
1055 * ``imac-tenid``: Set filter type as Inner MAC and tenant ID.
1057 * ``imac``: Set filter type as Inner MAC.
1059 * ``omac-imac-tenid``: Set filter type as Outer MAC, Inner MAC and tenant ID.
1061 * ``oip``: Set filter type as Outer IP.
1063 * ``iip``: Set filter type as Inner IP.
1067 testpmd> tunnel_filter add 0 68:05:CA:28:09:82 00:00:00:00:00:00 \
1068 192.168.2.2 0 ipingre oip 1 1
1070 Set an IP-in-GRE tunnel on port 0, and the filter type is Outer IP.
1072 tunnel_filter remove
1073 ~~~~~~~~~~~~~~~~~~~~
1075 Remove a tunnel filter on a port::
1077 testpmd> tunnel_filter rm (port_id) (outer_mac) (inner_mac) (ip_addr) \
1078 (inner_vlan) (vxlan|nvgre|ipingre|vxlan-gpe) (imac-ivlan|imac-ivlan-tenid|\
1079 imac-tenid|imac|omac-imac-tenid|oip|iip) (tenant_id) (queue_id)
1084 Add an UDP port for VXLAN packet filter on a port::
1086 testpmd> rx_vxlan_port add (udp_port) (port_id)
1088 rx_vxlan_port remove
1089 ~~~~~~~~~~~~~~~~~~~~
1091 Remove an UDP port for VXLAN packet filter on a port::
1093 testpmd> rx_vxlan_port rm (udp_port) (port_id)
1098 Set hardware insertion of VLAN IDs in packets sent on a port::
1100 testpmd> tx_vlan set (port_id) vlan_id[, vlan_id_outer]
1102 For example, set a single VLAN ID (5) insertion on port 0::
1106 Or, set double VLAN ID (inner: 2, outer: 3) insertion on port 1::
1114 Set port based hardware insertion of VLAN ID in packets sent on a port::
1116 testpmd> tx_vlan set pvid (port_id) (vlan_id) (on|off)
1121 Disable hardware insertion of a VLAN header in packets sent on a port::
1123 testpmd> tx_vlan reset (port_id)
1128 Select hardware or software calculation of the checksum when
1129 transmitting a packet using the ``csum`` forwarding engine::
1131 testpmd> csum set (ip|udp|tcp|sctp|outer-ip|outer-udp) (hw|sw) (port_id)
1135 * ``ip|udp|tcp|sctp`` always relate to the inner layer.
1137 * ``outer-ip`` relates to the outer IP layer (only for IPv4) in the case where the packet is recognized
1138 as a tunnel packet by the forwarding engine (vxlan, gre and ipip are
1139 supported). See also the ``csum parse-tunnel`` command.
1141 * ``outer-udp`` relates to the outer UDP layer in the case where the packet is recognized
1142 as a tunnel packet by the forwarding engine (vxlan, vxlan-gpe are
1143 supported). See also the ``csum parse-tunnel`` command.
1147 Check the NIC Datasheet for hardware limits.
1152 Set RSS queue region span on a port::
1154 testpmd> set port (port_id) queue-region region_id (value) \
1155 queue_start_index (value) queue_num (value)
1157 Set flowtype mapping on a RSS queue region on a port::
1159 testpmd> set port (port_id) queue-region region_id (value) flowtype (value)
1163 * For the flowtype(pctype) of packet,the specific index for each type has
1164 been defined in file i40e_type.h as enum i40e_filter_pctype.
1166 Set user priority mapping on a RSS queue region on a port::
1168 testpmd> set port (port_id) queue-region UP (value) region_id (value)
1170 Flush all queue region related configuration on a port::
1172 testpmd> set port (port_id) queue-region flush (on|off)
1176 * ``on``: is just an enable function which server for other configuration,
1177 it is for all configuration about queue region from up layer,
1178 at first will only keep in DPDK software stored in driver,
1179 only after "flush on", it commit all configuration to HW.
1181 * ``"off``: is just clean all configuration about queue region just now,
1182 and restore all to DPDK i40e driver default config when start up.
1184 Show all queue region related configuration info on a port::
1186 testpmd> show port (port_id) queue-region
1190 Queue region only support on PF by now, so these command is
1191 only for configuration of queue region on PF port.
1196 Define how tunneled packets should be handled by the csum forward
1199 testpmd> csum parse-tunnel (on|off) (tx_port_id)
1201 If enabled, the csum forward engine will try to recognize supported
1202 tunnel headers (vxlan, gre, ipip).
1204 If disabled, treat tunnel packets as non-tunneled packets (a inner
1205 header is handled as a packet payload).
1209 The port argument is the TX port like in the ``csum set`` command.
1213 Consider a packet in packet like the following::
1215 eth_out/ipv4_out/udp_out/vxlan/eth_in/ipv4_in/tcp_in
1217 * If parse-tunnel is enabled, the ``ip|udp|tcp|sctp`` parameters of ``csum set``
1218 command relate to the inner headers (here ``ipv4_in`` and ``tcp_in``), and the
1219 ``outer-ip|outer-udp`` parameter relates to the outer headers (here ``ipv4_out`` and ``udp_out``).
1221 * If parse-tunnel is disabled, the ``ip|udp|tcp|sctp`` parameters of ``csum set``
1222 command relate to the outer headers, here ``ipv4_out`` and ``udp_out``.
1227 Display tx checksum offload configuration::
1229 testpmd> csum show (port_id)
1234 Enable TCP Segmentation Offload (TSO) in the ``csum`` forwarding engine::
1236 testpmd> tso set (segsize) (port_id)
1240 Check the NIC datasheet for hardware limits.
1245 Display the status of TCP Segmentation Offload::
1247 testpmd> tso show (port_id)
1252 Set tso segment size of tunneled packets for a port in csum engine::
1254 testpmd> tunnel_tso set (tso_segsz) (port_id)
1259 Display the status of tunneled TCP Segmentation Offload for a port::
1261 testpmd> tunnel_tso show (port_id)
1266 Enable or disable GRO in ``csum`` forwarding engine::
1268 testpmd> set port <port_id> gro on|off
1270 If enabled, the csum forwarding engine will perform GRO on the TCP/IPv4
1271 packets received from the given port.
1273 If disabled, packets received from the given port won't be performed
1274 GRO. By default, GRO is disabled for all ports.
1278 When enable GRO for a port, TCP/IPv4 packets received from the port
1279 will be performed GRO. After GRO, all merged packets have bad
1280 checksums, since the GRO library doesn't re-calculate checksums for
1281 the merged packets. Therefore, if users want the merged packets to
1282 have correct checksums, please select HW IP checksum calculation and
1283 HW TCP checksum calculation for the port which the merged packets are
1289 Display GRO configuration for a given port::
1291 testpmd> show port <port_id> gro
1296 Set the cycle to flush the GROed packets from reassembly tables::
1298 testpmd> set gro flush <cycles>
1300 When enable GRO, the csum forwarding engine performs GRO on received
1301 packets, and the GROed packets are stored in reassembly tables. Users
1302 can use this command to determine when the GROed packets are flushed
1303 from the reassembly tables.
1305 The ``cycles`` is measured in GRO operation times. The csum forwarding
1306 engine flushes the GROed packets from the tables every ``cycles`` GRO
1309 By default, the value of ``cycles`` is 1, which means flush GROed packets
1310 from the reassembly tables as soon as one GRO operation finishes. The value
1311 of ``cycles`` should be in the range of 1 to ``GRO_MAX_FLUSH_CYCLES``.
1313 Please note that the large value of ``cycles`` may cause the poor TCP/IP
1314 stack performance. Because the GROed packets are delayed to arrive the
1315 stack, thus causing more duplicated ACKs and TCP retransmissions.
1320 Toggle per-port GSO support in ``csum`` forwarding engine::
1322 testpmd> set port <port_id> gso on|off
1324 If enabled, the csum forwarding engine will perform GSO on supported IPv4
1325 packets, transmitted on the given port.
1327 If disabled, packets transmitted on the given port will not undergo GSO.
1328 By default, GSO is disabled for all ports.
1332 When GSO is enabled on a port, supported IPv4 packets transmitted on that
1333 port undergo GSO. Afterwards, the segmented packets are represented by
1334 multi-segment mbufs; however, the csum forwarding engine doesn't calculation
1335 of checksums for GSO'd segments in SW. As a result, if users want correct
1336 checksums in GSO segments, they should enable HW checksum calculation for
1339 For example, HW checksum calculation for VxLAN GSO'd packets may be enabled
1340 by setting the following options in the csum forwarding engine:
1342 testpmd> csum set outer_ip hw <port_id>
1344 testpmd> csum set ip hw <port_id>
1346 testpmd> csum set tcp hw <port_id>
1348 UDP GSO is the same as IP fragmentation, which treats the UDP header
1349 as the payload and does not modify it during segmentation. That is,
1350 after UDP GSO, only the first output fragment has the original UDP
1351 header. Therefore, users need to enable HW IP checksum calculation
1352 and SW UDP checksum calculation for GSO-enabled ports, if they want
1353 correct checksums for UDP/IPv4 packets.
1358 Set the maximum GSO segment size (measured in bytes), which includes the
1359 packet header and the packet payload for GSO-enabled ports (global)::
1361 testpmd> set gso segsz <length>
1366 Display the status of Generic Segmentation Offload for a given port::
1368 testpmd> show port <port_id> gso
1373 Add an alternative MAC address to a port::
1375 testpmd> mac_addr add (port_id) (XX:XX:XX:XX:XX:XX)
1380 Remove a MAC address from a port::
1382 testpmd> mac_addr remove (port_id) (XX:XX:XX:XX:XX:XX)
1387 To add the multicast MAC address to/from the set of multicast addresses
1390 testpmd> mcast_addr add (port_id) (mcast_addr)
1395 To remove the multicast MAC address to/from the set of multicast addresses
1398 testpmd> mcast_addr remove (port_id) (mcast_addr)
1400 mac_addr add (for VF)
1401 ~~~~~~~~~~~~~~~~~~~~~
1403 Add an alternative MAC address for a VF to a port::
1405 testpmd> mac_add add port (port_id) vf (vf_id) (XX:XX:XX:XX:XX:XX)
1410 Set the default MAC address for a port::
1412 testpmd> mac_addr set (port_id) (XX:XX:XX:XX:XX:XX)
1414 mac_addr set (for VF)
1415 ~~~~~~~~~~~~~~~~~~~~~
1417 Set the MAC address for a VF from the PF::
1419 testpmd> set vf mac addr (port_id) (vf_id) (XX:XX:XX:XX:XX:XX)
1424 Set the forwarding peer address for certain port::
1426 testpmd> set eth-peer (port_id) (peer_addr)
1428 This is equivalent to the ``--eth-peer`` command-line option.
1433 Set the unicast hash filter(s) on/off for a port::
1435 testpmd> set port (port_id) uta (XX:XX:XX:XX:XX:XX|all) (on|off)
1440 Set the promiscuous mode on for a port or for all ports.
1441 In promiscuous mode packets are not dropped if they aren't for the specified MAC address::
1443 testpmd> set promisc (port_id|all) (on|off)
1448 Set the allmulti mode for a port or for all ports::
1450 testpmd> set allmulti (port_id|all) (on|off)
1452 Same as the ifconfig (8) option. Controls how multicast packets are handled.
1454 set promisc (for VF)
1455 ~~~~~~~~~~~~~~~~~~~~
1457 Set the unicast promiscuous mode for a VF from PF.
1458 It's supported by Intel i40e NICs now.
1459 In promiscuous mode packets are not dropped if they aren't for the specified MAC address::
1461 testpmd> set vf promisc (port_id) (vf_id) (on|off)
1463 set allmulticast (for VF)
1464 ~~~~~~~~~~~~~~~~~~~~~~~~~
1466 Set the multicast promiscuous mode for a VF from PF.
1467 It's supported by Intel i40e NICs now.
1468 In promiscuous mode packets are not dropped if they aren't for the specified MAC address::
1470 testpmd> set vf allmulti (port_id) (vf_id) (on|off)
1472 set tx max bandwidth (for VF)
1473 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1475 Set TX max absolute bandwidth (Mbps) for a VF from PF::
1477 testpmd> set vf tx max-bandwidth (port_id) (vf_id) (max_bandwidth)
1479 set tc tx min bandwidth (for VF)
1480 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1482 Set all TCs' TX min relative bandwidth (%) for a VF from PF::
1484 testpmd> set vf tc tx min-bandwidth (port_id) (vf_id) (bw1, bw2, ...)
1486 set tc tx max bandwidth (for VF)
1487 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1489 Set a TC's TX max absolute bandwidth (Mbps) for a VF from PF::
1491 testpmd> set vf tc tx max-bandwidth (port_id) (vf_id) (tc_no) (max_bandwidth)
1493 set tc strict link priority mode
1494 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1496 Set some TCs' strict link priority mode on a physical port::
1498 testpmd> set tx strict-link-priority (port_id) (tc_bitmap)
1500 set tc tx min bandwidth
1501 ~~~~~~~~~~~~~~~~~~~~~~~
1503 Set all TCs' TX min relative bandwidth (%) globally for all PF and VFs::
1505 testpmd> set tc tx min-bandwidth (port_id) (bw1, bw2, ...)
1510 Set the link flow control parameter on a port::
1512 testpmd> set flow_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \
1513 (pause_time) (send_xon) mac_ctrl_frame_fwd (on|off) \
1514 autoneg (on|off) (port_id)
1518 * ``high_water`` (integer): High threshold value to trigger XOFF.
1520 * ``low_water`` (integer): Low threshold value to trigger XON.
1522 * ``pause_time`` (integer): Pause quota in the Pause frame.
1524 * ``send_xon`` (0/1): Send XON frame.
1526 * ``mac_ctrl_frame_fwd``: Enable receiving MAC control frames.
1528 * ``autoneg``: Change the auto-negotiation parameter.
1533 Set the priority flow control parameter on a port::
1535 testpmd> set pfc_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \
1536 (pause_time) (priority) (port_id)
1540 * ``high_water`` (integer): High threshold value.
1542 * ``low_water`` (integer): Low threshold value.
1544 * ``pause_time`` (integer): Pause quota in the Pause frame.
1546 * ``priority`` (0-7): VLAN User Priority.
1551 Set statistics mapping (qmapping 0..15) for RX/TX queue on port::
1553 testpmd> set stat_qmap (tx|rx) (port_id) (queue_id) (qmapping)
1555 For example, to set rx queue 2 on port 0 to mapping 5::
1557 testpmd>set stat_qmap rx 0 2 5
1559 set xstats-hide-zero
1560 ~~~~~~~~~~~~~~~~~~~~
1562 Set the option to hide zero values for xstats display::
1564 testpmd> set xstats-hide-zero on|off
1568 By default, the zero values are displayed for xstats.
1570 set port - rx/tx (for VF)
1571 ~~~~~~~~~~~~~~~~~~~~~~~~~
1573 Set VF receive/transmit from a port::
1575 testpmd> set port (port_id) vf (vf_id) (rx|tx) (on|off)
1577 set port - mac address filter (for VF)
1578 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1580 Add/Remove unicast or multicast MAC addr filter for a VF::
1582 testpmd> set port (port_id) vf (vf_id) (mac_addr) \
1583 (exact-mac|exact-mac-vlan|hashmac|hashmac-vlan) (on|off)
1585 set port - rx mode(for VF)
1586 ~~~~~~~~~~~~~~~~~~~~~~~~~~
1588 Set the VF receive mode of a port::
1590 testpmd> set port (port_id) vf (vf_id) \
1591 rxmode (AUPE|ROPE|BAM|MPE) (on|off)
1593 The available receive modes are:
1595 * ``AUPE``: Accepts untagged VLAN.
1597 * ``ROPE``: Accepts unicast hash.
1599 * ``BAM``: Accepts broadcast packets.
1601 * ``MPE``: Accepts all multicast packets.
1603 set port - tx_rate (for Queue)
1604 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1606 Set TX rate limitation for a queue on a port::
1608 testpmd> set port (port_id) queue (queue_id) rate (rate_value)
1610 set port - tx_rate (for VF)
1611 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
1613 Set TX rate limitation for queues in VF on a port::
1615 testpmd> set port (port_id) vf (vf_id) rate (rate_value) queue_mask (queue_mask)
1617 set port - mirror rule
1618 ~~~~~~~~~~~~~~~~~~~~~~
1620 Set pool or vlan type mirror rule for a port::
1622 testpmd> set port (port_id) mirror-rule (rule_id) \
1623 (pool-mirror-up|pool-mirror-down|vlan-mirror) \
1624 (poolmask|vlanid[,vlanid]*) dst-pool (pool_id) (on|off)
1626 Set link mirror rule for a port::
1628 testpmd> set port (port_id) mirror-rule (rule_id) \
1629 (uplink-mirror|downlink-mirror) dst-pool (pool_id) (on|off)
1631 For example to enable mirror traffic with vlan 0,1 to pool 0::
1633 set port 0 mirror-rule 0 vlan-mirror 0,1 dst-pool 0 on
1635 reset port - mirror rule
1636 ~~~~~~~~~~~~~~~~~~~~~~~~
1638 Reset a mirror rule for a port::
1640 testpmd> reset port (port_id) mirror-rule (rule_id)
1645 Set the flush on RX streams before forwarding.
1646 The default is flush ``on``.
1647 Mainly used with PCAP drivers to turn off the default behavior of flushing the first 512 packets on RX streams::
1649 testpmd> set flush_rx off
1654 Set the bypass mode for the lowest port on bypass enabled NIC::
1656 testpmd> set bypass mode (normal|bypass|isolate) (port_id)
1661 Set the event required to initiate specified bypass mode for the lowest port on a bypass enabled::
1663 testpmd> set bypass event (timeout|os_on|os_off|power_on|power_off) \
1664 mode (normal|bypass|isolate) (port_id)
1668 * ``timeout``: Enable bypass after watchdog timeout.
1670 * ``os_on``: Enable bypass when OS/board is powered on.
1672 * ``os_off``: Enable bypass when OS/board is powered off.
1674 * ``power_on``: Enable bypass when power supply is turned on.
1676 * ``power_off``: Enable bypass when power supply is turned off.
1682 Set the bypass watchdog timeout to ``n`` seconds where 0 = instant::
1684 testpmd> set bypass timeout (0|1.5|2|3|4|8|16|32)
1689 Show the bypass configuration for a bypass enabled NIC using the lowest port on the NIC::
1691 testpmd> show bypass config (port_id)
1696 Set link up for a port::
1698 testpmd> set link-up port (port id)
1703 Set link down for a port::
1705 testpmd> set link-down port (port id)
1710 Enable E-tag insertion for a VF on a port::
1712 testpmd> E-tag set insertion on port-tag-id (value) port (port_id) vf (vf_id)
1714 Disable E-tag insertion for a VF on a port::
1716 testpmd> E-tag set insertion off port (port_id) vf (vf_id)
1718 Enable/disable E-tag stripping on a port::
1720 testpmd> E-tag set stripping (on|off) port (port_id)
1722 Enable/disable E-tag based forwarding on a port::
1724 testpmd> E-tag set forwarding (on|off) port (port_id)
1726 Add an E-tag forwarding filter on a port::
1728 testpmd> E-tag set filter add e-tag-id (value) dst-pool (pool_id) port (port_id)
1730 Delete an E-tag forwarding filter on a port::
1731 testpmd> E-tag set filter del e-tag-id (value) port (port_id)
1736 Load a dynamic device personalization (DDP) profile and store backup profile::
1738 testpmd> ddp add (port_id) (profile_path[,backup_profile_path])
1743 Delete a dynamic device personalization profile and restore backup profile::
1745 testpmd> ddp del (port_id) (backup_profile_path)
1750 List all items from the ptype mapping table::
1752 testpmd> ptype mapping get (port_id) (valid_only)
1756 * ``valid_only``: A flag indicates if only list valid items(=1) or all itemss(=0).
1758 Replace a specific or a group of software defined ptype with a new one::
1760 testpmd> ptype mapping replace (port_id) (target) (mask) (pkt_type)
1764 * ``target``: A specific software ptype or a mask to represent a group of software ptypes.
1766 * ``mask``: A flag indicate if "target" is a specific software ptype(=0) or a ptype mask(=1).
1768 * ``pkt_type``: The new software ptype to replace the old ones.
1770 Update hardware defined ptype to software defined packet type mapping table::
1772 testpmd> ptype mapping update (port_id) (hw_ptype) (sw_ptype)
1776 * ``hw_ptype``: hardware ptype as the index of the ptype mapping table.
1778 * ``sw_ptype``: software ptype as the value of the ptype mapping table.
1780 Reset ptype mapping table::
1782 testpmd> ptype mapping reset (port_id)
1784 config per port Rx offloading
1785 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1787 Enable or disable a per port Rx offloading on all Rx queues of a port::
1789 testpmd> port config (port_id) rx_offload (offloading) on|off
1791 * ``offloading``: can be any of these offloading capability:
1792 vlan_strip, ipv4_cksum, udp_cksum, tcp_cksum, tcp_lro,
1793 qinq_strip, outer_ipv4_cksum, macsec_strip,
1794 header_split, vlan_filter, vlan_extend, jumbo_frame,
1795 scatter, timestamp, security, keep_crc, rss_hash
1797 This command should be run when the port is stopped, or else it will fail.
1799 config per queue Rx offloading
1800 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1802 Enable or disable a per queue Rx offloading only on a specific Rx queue::
1804 testpmd> port (port_id) rxq (queue_id) rx_offload (offloading) on|off
1806 * ``offloading``: can be any of these offloading capability:
1807 vlan_strip, ipv4_cksum, udp_cksum, tcp_cksum, tcp_lro,
1808 qinq_strip, outer_ipv4_cksum, macsec_strip,
1809 header_split, vlan_filter, vlan_extend, jumbo_frame,
1810 scatter, timestamp, security, keep_crc
1812 This command should be run when the port is stopped, or else it will fail.
1814 config per port Tx offloading
1815 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1817 Enable or disable a per port Tx offloading on all Tx queues of a port::
1819 testpmd> port config (port_id) tx_offload (offloading) on|off
1821 * ``offloading``: can be any of these offloading capability:
1822 vlan_insert, ipv4_cksum, udp_cksum, tcp_cksum,
1823 sctp_cksum, tcp_tso, udp_tso, outer_ipv4_cksum,
1824 qinq_insert, vxlan_tnl_tso, gre_tnl_tso,
1825 ipip_tnl_tso, geneve_tnl_tso, macsec_insert,
1826 mt_lockfree, multi_segs, mbuf_fast_free, security
1828 This command should be run when the port is stopped, or else it will fail.
1830 config per queue Tx offloading
1831 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1833 Enable or disable a per queue Tx offloading only on a specific Tx queue::
1835 testpmd> port (port_id) txq (queue_id) tx_offload (offloading) on|off
1837 * ``offloading``: can be any of these offloading capability:
1838 vlan_insert, ipv4_cksum, udp_cksum, tcp_cksum,
1839 sctp_cksum, tcp_tso, udp_tso, outer_ipv4_cksum,
1840 qinq_insert, vxlan_tnl_tso, gre_tnl_tso,
1841 ipip_tnl_tso, geneve_tnl_tso, macsec_insert,
1842 mt_lockfree, multi_segs, mbuf_fast_free, security
1844 This command should be run when the port is stopped, or else it will fail.
1846 Config VXLAN Encap outer layers
1847 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1849 Configure the outer layer to encapsulate a packet inside a VXLAN tunnel::
1851 set vxlan ip-version (ipv4|ipv6) vni (vni) udp-src (udp-src) \
1852 udp-dst (udp-dst) ip-src (ip-src) ip-dst (ip-dst) eth-src (eth-src) \
1855 set vxlan-with-vlan ip-version (ipv4|ipv6) vni (vni) udp-src (udp-src) \
1856 udp-dst (udp-dst) ip-src (ip-src) ip-dst (ip-dst) vlan-tci (vlan-tci) \
1857 eth-src (eth-src) eth-dst (eth-dst)
1859 set vxlan-tos-ttl ip-version (ipv4|ipv6) vni (vni) udp-src (udp-src) \
1860 udp-dst (udp-dst) ip-tos (ip-tos) ip-ttl (ip-ttl) ip-src (ip-src) \
1861 ip-dst (ip-dst) eth-src (eth-src) eth-dst (eth-dst)
1863 These commands will set an internal configuration inside testpmd, any following
1864 flow rule using the action vxlan_encap will use the last configuration set.
1865 To have a different encapsulation header, one of those commands must be called
1866 before the flow rule creation.
1868 Config NVGRE Encap outer layers
1869 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1871 Configure the outer layer to encapsulate a packet inside a NVGRE tunnel::
1873 set nvgre ip-version (ipv4|ipv6) tni (tni) ip-src (ip-src) ip-dst (ip-dst) \
1874 eth-src (eth-src) eth-dst (eth-dst)
1875 set nvgre-with-vlan ip-version (ipv4|ipv6) tni (tni) ip-src (ip-src) \
1876 ip-dst (ip-dst) vlan-tci (vlan-tci) eth-src (eth-src) eth-dst (eth-dst)
1878 These commands will set an internal configuration inside testpmd, any following
1879 flow rule using the action nvgre_encap will use the last configuration set.
1880 To have a different encapsulation header, one of those commands must be called
1881 before the flow rule creation.
1886 Configure the l2 to be used when encapsulating a packet with L2::
1888 set l2_encap ip-version (ipv4|ipv6) eth-src (eth-src) eth-dst (eth-dst)
1889 set l2_encap-with-vlan ip-version (ipv4|ipv6) vlan-tci (vlan-tci) \
1890 eth-src (eth-src) eth-dst (eth-dst)
1892 Those commands will set an internal configuration inside testpmd, any following
1893 flow rule using the action l2_encap will use the last configuration set.
1894 To have a different encapsulation header, one of those commands must be called
1895 before the flow rule creation.
1900 Configure the l2 to be removed when decapsulating a packet with L2::
1902 set l2_decap ip-version (ipv4|ipv6)
1903 set l2_decap-with-vlan ip-version (ipv4|ipv6)
1905 Those commands will set an internal configuration inside testpmd, any following
1906 flow rule using the action l2_decap will use the last configuration set.
1907 To have a different encapsulation header, one of those commands must be called
1908 before the flow rule creation.
1910 Config MPLSoGRE Encap outer layers
1911 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1913 Configure the outer layer to encapsulate a packet inside a MPLSoGRE tunnel::
1915 set mplsogre_encap ip-version (ipv4|ipv6) label (label) \
1916 ip-src (ip-src) ip-dst (ip-dst) eth-src (eth-src) eth-dst (eth-dst)
1917 set mplsogre_encap-with-vlan ip-version (ipv4|ipv6) label (label) \
1918 ip-src (ip-src) ip-dst (ip-dst) vlan-tci (vlan-tci) \
1919 eth-src (eth-src) eth-dst (eth-dst)
1921 These commands will set an internal configuration inside testpmd, any following
1922 flow rule using the action mplsogre_encap will use the last configuration set.
1923 To have a different encapsulation header, one of those commands must be called
1924 before the flow rule creation.
1926 Config MPLSoGRE Decap outer layers
1927 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1929 Configure the outer layer to decapsulate MPLSoGRE packet::
1931 set mplsogre_decap ip-version (ipv4|ipv6)
1932 set mplsogre_decap-with-vlan ip-version (ipv4|ipv6)
1934 These commands will set an internal configuration inside testpmd, any following
1935 flow rule using the action mplsogre_decap will use the last configuration set.
1936 To have a different decapsulation header, one of those commands must be called
1937 before the flow rule creation.
1939 Config MPLSoUDP Encap outer layers
1940 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1942 Configure the outer layer to encapsulate a packet inside a MPLSoUDP tunnel::
1944 set mplsoudp_encap ip-version (ipv4|ipv6) label (label) udp-src (udp-src) \
1945 udp-dst (udp-dst) ip-src (ip-src) ip-dst (ip-dst) \
1946 eth-src (eth-src) eth-dst (eth-dst)
1947 set mplsoudp_encap-with-vlan ip-version (ipv4|ipv6) label (label) \
1948 udp-src (udp-src) udp-dst (udp-dst) ip-src (ip-src) ip-dst (ip-dst) \
1949 vlan-tci (vlan-tci) eth-src (eth-src) eth-dst (eth-dst)
1951 These commands will set an internal configuration inside testpmd, any following
1952 flow rule using the action mplsoudp_encap will use the last configuration set.
1953 To have a different encapsulation header, one of those commands must be called
1954 before the flow rule creation.
1956 Config MPLSoUDP Decap outer layers
1957 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1959 Configure the outer layer to decapsulate MPLSoUDP packet::
1961 set mplsoudp_decap ip-version (ipv4|ipv6)
1962 set mplsoudp_decap-with-vlan ip-version (ipv4|ipv6)
1964 These commands will set an internal configuration inside testpmd, any following
1965 flow rule using the action mplsoudp_decap will use the last configuration set.
1966 To have a different decapsulation header, one of those commands must be called
1967 before the flow rule creation.
1969 Config Raw Encapsulation
1970 ~~~~~~~~~~~~~~~~~~~~~~~~~
1972 Configure the raw data to be used when encapsulating a packet by
1973 rte_flow_action_raw_encap::
1975 set raw_encap {index} {item} [/ {item} [...]] / end_set
1977 There are multiple global buffers for ``raw_encap``, this command will set one
1978 internal buffer index by ``{index}``.
1979 If there is no ``{index}`` specified::
1981 set raw_encap {item} [/ {item} [...]] / end_set
1983 the default index ``0`` is used.
1984 In order to use different encapsulating header, ``index`` must be specified
1985 during the flow rule creation::
1987 testpmd> flow create 0 egress pattern eth / ipv4 / end actions
1988 raw_encap index 2 / end
1990 Otherwise the default index ``0`` is used.
1992 Config Raw Decapsulation
1993 ~~~~~~~~~~~~~~~~~~~~~~~~
1995 Configure the raw data to be used when decapsulating a packet by
1996 rte_flow_action_raw_decap::
1998 set raw_decap {index} {item} [/ {item} [...]] / end_set
2000 There are multiple global buffers for ``raw_decap``, this command will set
2001 one internal buffer index by ``{index}``.
2002 If there is no ``{index}`` specified::
2004 set raw_decap {item} [/ {item} [...]] / end_set
2006 the default index ``0`` is used.
2007 In order to use different decapsulating header, ``index`` must be specified
2008 during the flow rule creation::
2010 testpmd> flow create 0 egress pattern eth / ipv4 / end actions
2011 raw_encap index 3 / end
2013 Otherwise the default index ``0`` is used.
2018 The following sections show functions for configuring ports.
2022 Port configuration changes only become active when forwarding is started/restarted.
2027 Attach a port specified by pci address or virtual device args::
2029 testpmd> port attach (identifier)
2031 To attach a new pci device, the device should be recognized by kernel first.
2032 Then it should be moved under DPDK management.
2033 Finally the port can be attached to testpmd.
2035 For example, to move a pci device using ixgbe under DPDK management:
2037 .. code-block:: console
2039 # Check the status of the available devices.
2040 ./usertools/dpdk-devbind.py --status
2042 Network devices using DPDK-compatible driver
2043 ============================================
2046 Network devices using kernel driver
2047 ===================================
2048 0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused=
2051 # Bind the device to igb_uio.
2052 sudo ./usertools/dpdk-devbind.py -b igb_uio 0000:0a:00.0
2055 # Recheck the status of the devices.
2056 ./usertools/dpdk-devbind.py --status
2057 Network devices using DPDK-compatible driver
2058 ============================================
2059 0000:0a:00.0 '82599ES 10-Gigabit' drv=igb_uio unused=
2061 To attach a port created by virtual device, above steps are not needed.
2063 For example, to attach a port whose pci address is 0000:0a:00.0.
2065 .. code-block:: console
2067 testpmd> port attach 0000:0a:00.0
2068 Attaching a new port...
2069 EAL: PCI device 0000:0a:00.0 on NUMA socket -1
2070 EAL: probe driver: 8086:10fb rte_ixgbe_pmd
2071 EAL: PCI memory mapped at 0x7f83bfa00000
2072 EAL: PCI memory mapped at 0x7f83bfa80000
2073 PMD: eth_ixgbe_dev_init(): MAC: 2, PHY: 18, SFP+: 5
2074 PMD: eth_ixgbe_dev_init(): port 0 vendorID=0x8086 deviceID=0x10fb
2075 Port 0 is attached. Now total ports is 1
2078 For example, to attach a port created by pcap PMD.
2080 .. code-block:: console
2082 testpmd> port attach net_pcap0
2083 Attaching a new port...
2084 PMD: Initializing pmd_pcap for net_pcap0
2085 PMD: Creating pcap-backed ethdev on numa socket 0
2086 Port 0 is attached. Now total ports is 1
2089 In this case, identifier is ``net_pcap0``.
2090 This identifier format is the same as ``--vdev`` format of DPDK applications.
2092 For example, to re-attach a bonded port which has been previously detached,
2093 the mode and slave parameters must be given.
2095 .. code-block:: console
2097 testpmd> port attach net_bond_0,mode=0,slave=1
2098 Attaching a new port...
2099 EAL: Initializing pmd_bond for net_bond_0
2100 EAL: Create bonded device net_bond_0 on port 0 in mode 0 on socket 0.
2101 Port 0 is attached. Now total ports is 1
2108 Detach a specific port::
2110 testpmd> port detach (port_id)
2112 Before detaching a port, the port should be stopped and closed.
2114 For example, to detach a pci device port 0.
2116 .. code-block:: console
2118 testpmd> port stop 0
2121 testpmd> port close 0
2125 testpmd> port detach 0
2127 EAL: PCI device 0000:0a:00.0 on NUMA socket -1
2128 EAL: remove driver: 8086:10fb rte_ixgbe_pmd
2129 EAL: PCI memory unmapped at 0x7f83bfa00000
2130 EAL: PCI memory unmapped at 0x7f83bfa80000
2134 For example, to detach a virtual device port 0.
2136 .. code-block:: console
2138 testpmd> port stop 0
2141 testpmd> port close 0
2145 testpmd> port detach 0
2147 PMD: Closing pcap ethdev on numa socket 0
2148 Port 'net_pcap0' is detached. Now total ports is 0
2151 To remove a pci device completely from the system, first detach the port from testpmd.
2152 Then the device should be moved under kernel management.
2153 Finally the device can be removed using kernel pci hotplug functionality.
2155 For example, to move a pci device under kernel management:
2157 .. code-block:: console
2159 sudo ./usertools/dpdk-devbind.py -b ixgbe 0000:0a:00.0
2161 ./usertools/dpdk-devbind.py --status
2163 Network devices using DPDK-compatible driver
2164 ============================================
2167 Network devices using kernel driver
2168 ===================================
2169 0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused=igb_uio
2171 To remove a port created by a virtual device, above steps are not needed.
2176 Start all ports or a specific port::
2178 testpmd> port start (port_id|all)
2183 Stop all ports or a specific port::
2185 testpmd> port stop (port_id|all)
2190 Close all ports or a specific port::
2192 testpmd> port close (port_id|all)
2197 Reset all ports or a specific port::
2199 testpmd> port reset (port_id|all)
2201 User should stop port(s) before resetting and (re-)start after reset.
2203 port config - queue ring size
2204 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2206 Configure a rx/tx queue ring size::
2208 testpmd> port (port_id) (rxq|txq) (queue_id) ring_size (value)
2210 Only take effect after command that (re-)start the port or command that setup specific queue.
2212 port start/stop queue
2213 ~~~~~~~~~~~~~~~~~~~~~
2215 Start/stop a rx/tx queue on a specific port::
2217 testpmd> port (port_id) (rxq|txq) (queue_id) (start|stop)
2219 port config - queue deferred start
2220 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2222 Switch on/off deferred start of a specific port queue::
2224 testpmd> port (port_id) (rxq|txq) (queue_id) deferred_start (on|off)
2227 ~~~~~~~~~~~~~~~~~~~~~
2229 Setup a rx/tx queue on a specific port::
2231 testpmd> port (port_id) (rxq|txq) (queue_id) setup
2233 Only take effect when port is started.
2238 Set the speed and duplex mode for all ports or a specific port::
2240 testpmd> port config (port_id|all) speed (10|100|1000|10000|25000|40000|50000|100000|200000|auto) \
2241 duplex (half|full|auto)
2243 port config - queues/descriptors
2244 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2246 Set number of queues/descriptors for rxq, txq, rxd and txd::
2248 testpmd> port config all (rxq|txq|rxd|txd) (value)
2250 This is equivalent to the ``--rxq``, ``--txq``, ``--rxd`` and ``--txd`` command-line options.
2252 port config - max-pkt-len
2253 ~~~~~~~~~~~~~~~~~~~~~~~~~
2255 Set the maximum packet length::
2257 testpmd> port config all max-pkt-len (value)
2259 This is equivalent to the ``--max-pkt-len`` command-line option.
2261 port config - max-lro-pkt-size
2262 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2264 Set the maximum LRO aggregated packet size::
2266 testpmd> port config all max-lro-pkt-size (value)
2268 This is equivalent to the ``--max-lro-pkt-size`` command-line option.
2270 port config - Drop Packets
2271 ~~~~~~~~~~~~~~~~~~~~~~~~~~
2273 Enable or disable packet drop on all RX queues of all ports when no receive buffers available::
2275 testpmd> port config all drop-en (on|off)
2277 Packet dropping when no receive buffers available is off by default.
2279 The ``on`` option is equivalent to the ``--enable-drop-en`` command-line option.
2284 Set the RSS (Receive Side Scaling) mode on or off::
2286 testpmd> port config all rss (all|default|eth|vlan|ip|tcp|udp|sctp|ether|port|vxlan|geneve|nvgre|vxlan-gpe|l2tpv3|esp|ah|pfcp|none)
2288 RSS is on by default.
2290 The ``all`` option is equivalent to eth|vlan|ip|tcp|udp|sctp|ether|l2tpv3|esp|ah|pfcp.
2292 The ``default`` option enables all supported RSS types reported by device info.
2294 The ``none`` option is equivalent to the ``--disable-rss`` command-line option.
2296 port config - RSS Reta
2297 ~~~~~~~~~~~~~~~~~~~~~~
2299 Set the RSS (Receive Side Scaling) redirection table::
2301 testpmd> port config all rss reta (hash,queue)[,(hash,queue)]
2306 Set the DCB mode for an individual port::
2308 testpmd> port config (port_id) dcb vt (on|off) (traffic_class) pfc (on|off)
2310 The traffic class should be 4 or 8.
2315 Set the number of packets per burst::
2317 testpmd> port config all burst (value)
2319 This is equivalent to the ``--burst`` command-line option.
2321 port config - Threshold
2322 ~~~~~~~~~~~~~~~~~~~~~~~
2324 Set thresholds for TX/RX queues::
2326 testpmd> port config all (threshold) (value)
2328 Where the threshold type can be:
2330 * ``txpt:`` Set the prefetch threshold register of the TX rings, 0 <= value <= 255.
2332 * ``txht:`` Set the host threshold register of the TX rings, 0 <= value <= 255.
2334 * ``txwt:`` Set the write-back threshold register of the TX rings, 0 <= value <= 255.
2336 * ``rxpt:`` Set the prefetch threshold register of the RX rings, 0 <= value <= 255.
2338 * ``rxht:`` Set the host threshold register of the RX rings, 0 <= value <= 255.
2340 * ``rxwt:`` Set the write-back threshold register of the RX rings, 0 <= value <= 255.
2342 * ``txfreet:`` Set the transmit free threshold of the TX rings, 0 <= value <= txd.
2344 * ``rxfreet:`` Set the transmit free threshold of the RX rings, 0 <= value <= rxd.
2346 * ``txrst:`` Set the transmit RS bit threshold of TX rings, 0 <= value <= txd.
2348 These threshold options are also available from the command-line.
2353 Set the value of ether-type for E-tag::
2355 testpmd> port config (port_id|all) l2-tunnel E-tag ether-type (value)
2357 Enable/disable the E-tag support::
2359 testpmd> port config (port_id|all) l2-tunnel E-tag (enable|disable)
2361 port config pctype mapping
2362 ~~~~~~~~~~~~~~~~~~~~~~~~~~
2364 Reset pctype mapping table::
2366 testpmd> port config (port_id) pctype mapping reset
2368 Update hardware defined pctype to software defined flow type mapping table::
2370 testpmd> port config (port_id) pctype mapping update (pctype_id_0[,pctype_id_1]*) (flow_type_id)
2374 * ``pctype_id_x``: hardware pctype id as index of bit in bitmask value of the pctype mapping table.
2376 * ``flow_type_id``: software flow type id as the index of the pctype mapping table.
2378 port config input set
2379 ~~~~~~~~~~~~~~~~~~~~~
2381 Config RSS/FDIR/FDIR flexible payload input set for some pctype::
2383 testpmd> port config (port_id) pctype (pctype_id) \
2384 (hash_inset|fdir_inset|fdir_flx_inset) \
2385 (get|set|clear) field (field_idx)
2387 Clear RSS/FDIR/FDIR flexible payload input set for some pctype::
2389 testpmd> port config (port_id) pctype (pctype_id) \
2390 (hash_inset|fdir_inset|fdir_flx_inset) clear all
2394 * ``pctype_id``: hardware packet classification types.
2395 * ``field_idx``: hardware field index.
2397 port config udp_tunnel_port
2398 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
2400 Add/remove UDP tunnel port for VXLAN/GENEVE tunneling protocols::
2402 testpmd> port config (port_id) udp_tunnel_port add|rm vxlan|geneve|vxlan-gpe (udp_port)
2404 port config tx_metadata
2405 ~~~~~~~~~~~~~~~~~~~~~~~
2407 Set Tx metadata value per port.
2408 testpmd will add this value to any Tx packet sent from this port::
2410 testpmd> port config (port_id) tx_metadata (value)
2415 Set/clear dynamic flag per port.
2416 testpmd will register this flag in the mbuf (same registration
2417 for both Tx and Rx). Then set/clear this flag for each Tx
2418 packet sent from this port. The set bit only works for Tx packet::
2420 testpmd> port config (port_id) dynf (name) (set|clear)
2425 To configure MTU(Maximum Transmission Unit) on devices using testpmd::
2427 testpmd> port config mtu (port_id) (value)
2429 port config rss hash key
2430 ~~~~~~~~~~~~~~~~~~~~~~~~
2432 To configure the RSS hash key used to compute the RSS
2433 hash of input [IP] packets received on port::
2435 testpmd> port config <port_id> rss-hash-key (ipv4|ipv4-frag|\
2436 ipv4-tcp|ipv4-udp|ipv4-sctp|ipv4-other|\
2437 ipv6|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|\
2438 ipv6-other|l2-payload|ipv6-ex|ipv6-tcp-ex|\
2439 ipv6-udp-ex <string of hex digits \
2440 (variable length, NIC dependent)>)
2445 The following sections show functions for device operations.
2450 Detach a device specified by pci address or virtual device args::
2452 testpmd> device detach (identifier)
2454 Before detaching a device associated with ports, the ports should be stopped and closed.
2456 For example, to detach a pci device whose address is 0002:03:00.0.
2458 .. code-block:: console
2460 testpmd> device detach 0002:03:00.0
2461 Removing a device...
2462 Port 1 is now closed
2463 EAL: Releasing pci mapped resource for 0002:03:00.0
2464 EAL: Calling pci_unmap_resource for 0002:03:00.0 at 0x218a050000
2465 EAL: Calling pci_unmap_resource for 0002:03:00.0 at 0x218c050000
2466 Device 0002:03:00.0 is detached
2467 Now total ports is 1
2469 For example, to detach a port created by pcap PMD.
2471 .. code-block:: console
2473 testpmd> device detach net_pcap0
2474 Removing a device...
2475 Port 0 is now closed
2476 Device net_pcap0 is detached
2477 Now total ports is 0
2480 In this case, identifier is ``net_pcap0``.
2481 This identifier format is the same as ``--vdev`` format of DPDK applications.
2483 Link Bonding Functions
2484 ----------------------
2486 The Link Bonding functions make it possible to dynamically create and
2487 manage link bonding devices from within testpmd interactive prompt.
2489 create bonded device
2490 ~~~~~~~~~~~~~~~~~~~~
2492 Create a new bonding device::
2494 testpmd> create bonded device (mode) (socket)
2496 For example, to create a bonded device in mode 1 on socket 0::
2498 testpmd> create bonded device 1 0
2499 created new bonded device (port X)
2504 Adds Ethernet device to a Link Bonding device::
2506 testpmd> add bonding slave (slave id) (port id)
2508 For example, to add Ethernet device (port 6) to a Link Bonding device (port 10)::
2510 testpmd> add bonding slave 6 10
2513 remove bonding slave
2514 ~~~~~~~~~~~~~~~~~~~~
2516 Removes an Ethernet slave device from a Link Bonding device::
2518 testpmd> remove bonding slave (slave id) (port id)
2520 For example, to remove Ethernet slave device (port 6) to a Link Bonding device (port 10)::
2522 testpmd> remove bonding slave 6 10
2527 Set the Link Bonding mode of a Link Bonding device::
2529 testpmd> set bonding mode (value) (port id)
2531 For example, to set the bonding mode of a Link Bonding device (port 10) to broadcast (mode 3)::
2533 testpmd> set bonding mode 3 10
2538 Set an Ethernet slave device as the primary device on a Link Bonding device::
2540 testpmd> set bonding primary (slave id) (port id)
2542 For example, to set the Ethernet slave device (port 6) as the primary port of a Link Bonding device (port 10)::
2544 testpmd> set bonding primary 6 10
2549 Set the MAC address of a Link Bonding device::
2551 testpmd> set bonding mac (port id) (mac)
2553 For example, to set the MAC address of a Link Bonding device (port 10) to 00:00:00:00:00:01::
2555 testpmd> set bonding mac 10 00:00:00:00:00:01
2557 set bonding xmit_balance_policy
2558 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2560 Set the transmission policy for a Link Bonding device when it is in Balance XOR mode::
2562 testpmd> set bonding xmit_balance_policy (port_id) (l2|l23|l34)
2564 For example, set a Link Bonding device (port 10) to use a balance policy of layer 3+4 (IP addresses & UDP ports)::
2566 testpmd> set bonding xmit_balance_policy 10 l34
2569 set bonding mon_period
2570 ~~~~~~~~~~~~~~~~~~~~~~
2572 Set the link status monitoring polling period in milliseconds for a bonding device.
2574 This adds support for PMD slave devices which do not support link status interrupts.
2575 When the mon_period is set to a value greater than 0 then all PMD's which do not support
2576 link status ISR will be queried every polling interval to check if their link status has changed::
2578 testpmd> set bonding mon_period (port_id) (value)
2580 For example, to set the link status monitoring polling period of bonded device (port 5) to 150ms::
2582 testpmd> set bonding mon_period 5 150
2585 set bonding lacp dedicated_queue
2586 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2588 Enable dedicated tx/rx queues on bonding devices slaves to handle LACP control plane traffic
2589 when in mode 4 (link-aggregation-802.3ad)::
2591 testpmd> set bonding lacp dedicated_queues (port_id) (enable|disable)
2594 set bonding agg_mode
2595 ~~~~~~~~~~~~~~~~~~~~
2597 Enable one of the specific aggregators mode when in mode 4 (link-aggregation-802.3ad)::
2599 testpmd> set bonding agg_mode (port_id) (bandwidth|count|stable)
2605 Show the current configuration of a Link Bonding device::
2607 testpmd> show bonding config (port id)
2610 to show the configuration a Link Bonding device (port 9) with 3 slave devices (1, 3, 4)
2611 in balance mode with a transmission policy of layer 2+3::
2613 testpmd> show bonding config 9
2615 Balance Xmit Policy: BALANCE_XMIT_POLICY_LAYER23
2617 Active Slaves (3): [1 3 4]
2624 The Register Functions can be used to read from and write to registers on the network card referenced by a port number.
2625 This is mainly useful for debugging purposes.
2626 Reference should be made to the appropriate datasheet for the network card for details on the register addresses
2627 and fields that can be accessed.
2632 Display the value of a port register::
2634 testpmd> read reg (port_id) (address)
2636 For example, to examine the Flow Director control register (FDIRCTL, 0x0000EE000) on an Intel 82599 10 GbE Controller::
2638 testpmd> read reg 0 0xEE00
2639 port 0 PCI register at offset 0xEE00: 0x4A060029 (1241907241)
2644 Display a port register bit field::
2646 testpmd> read regfield (port_id) (address) (bit_x) (bit_y)
2648 For example, reading the lowest two bits from the register in the example above::
2650 testpmd> read regfield 0 0xEE00 0 1
2651 port 0 PCI register at offset 0xEE00: bits[0, 1]=0x1 (1)
2656 Display a single port register bit::
2658 testpmd> read regbit (port_id) (address) (bit_x)
2660 For example, reading the lowest bit from the register in the example above::
2662 testpmd> read regbit 0 0xEE00 0
2663 port 0 PCI register at offset 0xEE00: bit 0=1
2668 Set the value of a port register::
2670 testpmd> write reg (port_id) (address) (value)
2672 For example, to clear a register::
2674 testpmd> write reg 0 0xEE00 0x0
2675 port 0 PCI register at offset 0xEE00: 0x00000000 (0)
2680 Set bit field of a port register::
2682 testpmd> write regfield (port_id) (address) (bit_x) (bit_y) (value)
2684 For example, writing to the register cleared in the example above::
2686 testpmd> write regfield 0 0xEE00 0 1 2
2687 port 0 PCI register at offset 0xEE00: 0x00000002 (2)
2692 Set single bit value of a port register::
2694 testpmd> write regbit (port_id) (address) (bit_x) (value)
2696 For example, to set the high bit in the register from the example above::
2698 testpmd> write regbit 0 0xEE00 31 1
2699 port 0 PCI register at offset 0xEE00: 0x8000000A (2147483658)
2701 Traffic Metering and Policing
2702 -----------------------------
2704 The following section shows functions for configuring traffic metering and
2705 policing on the ethernet device through the use of generic ethdev API.
2707 show port traffic management capability
2708 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2710 Show traffic metering and policing capability of the port::
2712 testpmd> show port meter cap (port_id)
2714 add port meter profile (srTCM rfc2967)
2715 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2717 Add meter profile (srTCM rfc2697) to the ethernet device::
2719 testpmd> add port meter profile srtcm_rfc2697 (port_id) (profile_id) \
2724 * ``profile_id``: ID for the meter profile.
2725 * ``cir``: Committed Information Rate (CIR) (bytes/second).
2726 * ``cbs``: Committed Burst Size (CBS) (bytes).
2727 * ``ebs``: Excess Burst Size (EBS) (bytes).
2729 add port meter profile (trTCM rfc2968)
2730 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2732 Add meter profile (srTCM rfc2698) to the ethernet device::
2734 testpmd> add port meter profile trtcm_rfc2698 (port_id) (profile_id) \
2735 (cir) (pir) (cbs) (pbs)
2739 * ``profile_id``: ID for the meter profile.
2740 * ``cir``: Committed information rate (bytes/second).
2741 * ``pir``: Peak information rate (bytes/second).
2742 * ``cbs``: Committed burst size (bytes).
2743 * ``pbs``: Peak burst size (bytes).
2745 add port meter profile (trTCM rfc4115)
2746 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2748 Add meter profile (trTCM rfc4115) to the ethernet device::
2750 testpmd> add port meter profile trtcm_rfc4115 (port_id) (profile_id) \
2751 (cir) (eir) (cbs) (ebs)
2755 * ``profile_id``: ID for the meter profile.
2756 * ``cir``: Committed information rate (bytes/second).
2757 * ``eir``: Excess information rate (bytes/second).
2758 * ``cbs``: Committed burst size (bytes).
2759 * ``ebs``: Excess burst size (bytes).
2761 delete port meter profile
2762 ~~~~~~~~~~~~~~~~~~~~~~~~~
2764 Delete meter profile from the ethernet device::
2766 testpmd> del port meter profile (port_id) (profile_id)
2771 Create new meter object for the ethernet device::
2773 testpmd> create port meter (port_id) (mtr_id) (profile_id) \
2774 (meter_enable) (g_action) (y_action) (r_action) (stats_mask) (shared) \
2775 (use_pre_meter_color) [(dscp_tbl_entry0) (dscp_tbl_entry1)...\
2780 * ``mtr_id``: meter object ID.
2781 * ``profile_id``: ID for the meter profile.
2782 * ``meter_enable``: When this parameter has a non-zero value, the meter object
2783 gets enabled at the time of creation, otherwise remains disabled.
2784 * ``g_action``: Policer action for the packet with green color.
2785 * ``y_action``: Policer action for the packet with yellow color.
2786 * ``r_action``: Policer action for the packet with red color.
2787 * ``stats_mask``: Mask of statistics counter types to be enabled for the
2789 * ``shared``: When this parameter has a non-zero value, the meter object is
2790 shared by multiple flows. Otherwise, meter object is used by single flow.
2791 * ``use_pre_meter_color``: When this parameter has a non-zero value, the
2792 input color for the current meter object is determined by the latest meter
2793 object in the same flow. Otherwise, the current meter object uses the
2794 *dscp_table* to determine the input color.
2795 * ``dscp_tbl_entryx``: DSCP table entry x providing meter providing input
2796 color, 0 <= x <= 63.
2801 Enable meter for the ethernet device::
2803 testpmd> enable port meter (port_id) (mtr_id)
2808 Disable meter for the ethernet device::
2810 testpmd> disable port meter (port_id) (mtr_id)
2815 Delete meter for the ethernet device::
2817 testpmd> del port meter (port_id) (mtr_id)
2819 Set port meter profile
2820 ~~~~~~~~~~~~~~~~~~~~~~
2822 Set meter profile for the ethernet device::
2824 testpmd> set port meter profile (port_id) (mtr_id) (profile_id)
2826 set port meter dscp table
2827 ~~~~~~~~~~~~~~~~~~~~~~~~~
2829 Set meter dscp table for the ethernet device::
2831 testpmd> set port meter dscp table (port_id) (mtr_id) [(dscp_tbl_entry0) \
2832 (dscp_tbl_entry1)...(dscp_tbl_entry63)]
2834 set port meter policer action
2835 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2837 Set meter policer action for the ethernet device::
2839 testpmd> set port meter policer action (port_id) (mtr_id) (action_mask) \
2840 (action0) [(action1) (action1)]
2844 * ``action_mask``: Bit mask indicating which policer actions need to be
2845 updated. One or more policer actions can be updated in a single function
2846 invocation. To update the policer action associated with color C, bit
2847 (1 << C) needs to be set in *action_mask* and element at position C
2848 in the *actions* array needs to be valid.
2849 * ``actionx``: Policer action for the color x,
2850 RTE_MTR_GREEN <= x < RTE_MTR_COLORS
2852 set port meter stats mask
2853 ~~~~~~~~~~~~~~~~~~~~~~~~~
2855 Set meter stats mask for the ethernet device::
2857 testpmd> set port meter stats mask (port_id) (mtr_id) (stats_mask)
2861 * ``stats_mask``: Bit mask indicating statistics counter types to be enabled.
2863 show port meter stats
2864 ~~~~~~~~~~~~~~~~~~~~~
2866 Show meter stats of the ethernet device::
2868 testpmd> show port meter stats (port_id) (mtr_id) (clear)
2872 * ``clear``: Flag that indicates whether the statistics counters should
2873 be cleared (i.e. set to zero) immediately after they have been read or not.
2878 The following section shows functions for configuring traffic management on
2879 the ethernet device through the use of generic TM API.
2881 show port traffic management capability
2882 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2884 Show traffic management capability of the port::
2886 testpmd> show port tm cap (port_id)
2888 show port traffic management capability (hierarchy level)
2889 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2891 Show traffic management hierarchy level capability of the port::
2893 testpmd> show port tm level cap (port_id) (level_id)
2895 show port traffic management capability (hierarchy node level)
2896 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2898 Show the traffic management hierarchy node capability of the port::
2900 testpmd> show port tm node cap (port_id) (node_id)
2902 show port traffic management hierarchy node type
2903 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2905 Show the port traffic management hierarchy node type::
2907 testpmd> show port tm node type (port_id) (node_id)
2909 show port traffic management hierarchy node stats
2910 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2912 Show the port traffic management hierarchy node statistics::
2914 testpmd> show port tm node stats (port_id) (node_id) (clear)
2918 * ``clear``: When this parameter has a non-zero value, the statistics counters
2919 are cleared (i.e. set to zero) immediately after they have been read,
2920 otherwise the statistics counters are left untouched.
2922 Add port traffic management private shaper profile
2923 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2925 Add the port traffic management private shaper profile::
2927 testpmd> add port tm node shaper profile (port_id) (shaper_profile_id) \
2928 (cmit_tb_rate) (cmit_tb_size) (peak_tb_rate) (peak_tb_size) \
2929 (packet_length_adjust) (packet_mode)
2933 * ``shaper_profile id``: Shaper profile ID for the new profile.
2934 * ``cmit_tb_rate``: Committed token bucket rate (bytes per second or packets per second).
2935 * ``cmit_tb_size``: Committed token bucket size (bytes or packets).
2936 * ``peak_tb_rate``: Peak token bucket rate (bytes per second or packets per second).
2937 * ``peak_tb_size``: Peak token bucket size (bytes or packets).
2938 * ``packet_length_adjust``: The value (bytes) to be added to the length of
2939 each packet for the purpose of shaping. This parameter value can be used to
2940 correct the packet length with the framing overhead bytes that are consumed
2942 * ``packet_mode``: Shaper configured in packet mode. This parameter value if
2943 zero, configures shaper in byte mode and if non-zero configures it in packet
2946 Delete port traffic management private shaper profile
2947 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2949 Delete the port traffic management private shaper::
2951 testpmd> del port tm node shaper profile (port_id) (shaper_profile_id)
2955 * ``shaper_profile id``: Shaper profile ID that needs to be deleted.
2957 Add port traffic management shared shaper
2958 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2960 Create the port traffic management shared shaper::
2962 testpmd> add port tm node shared shaper (port_id) (shared_shaper_id) \
2967 * ``shared_shaper_id``: Shared shaper ID to be created.
2968 * ``shaper_profile id``: Shaper profile ID for shared shaper.
2970 Set port traffic management shared shaper
2971 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2973 Update the port traffic management shared shaper::
2975 testpmd> set port tm node shared shaper (port_id) (shared_shaper_id) \
2980 * ``shared_shaper_id``: Shared shaper ID to be update.
2981 * ``shaper_profile id``: Shaper profile ID for shared shaper.
2983 Delete port traffic management shared shaper
2984 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2986 Delete the port traffic management shared shaper::
2988 testpmd> del port tm node shared shaper (port_id) (shared_shaper_id)
2992 * ``shared_shaper_id``: Shared shaper ID to be deleted.
2994 Set port traffic management hierarchy node private shaper
2995 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2997 set the port traffic management hierarchy node private shaper::
2999 testpmd> set port tm node shaper profile (port_id) (node_id) \
3004 * ``shaper_profile id``: Private shaper profile ID to be enabled on the
3007 Add port traffic management WRED profile
3008 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3010 Create a new WRED profile::
3012 testpmd> add port tm node wred profile (port_id) (wred_profile_id) \
3013 (color_g) (min_th_g) (max_th_g) (maxp_inv_g) (wq_log2_g) \
3014 (color_y) (min_th_y) (max_th_y) (maxp_inv_y) (wq_log2_y) \
3015 (color_r) (min_th_r) (max_th_r) (maxp_inv_r) (wq_log2_r)
3019 * ``wred_profile id``: Identifier for the newly create WRED profile
3020 * ``color_g``: Packet color (green)
3021 * ``min_th_g``: Minimum queue threshold for packet with green color
3022 * ``max_th_g``: Minimum queue threshold for packet with green color
3023 * ``maxp_inv_g``: Inverse of packet marking probability maximum value (maxp)
3024 * ``wq_log2_g``: Negated log2 of queue weight (wq)
3025 * ``color_y``: Packet color (yellow)
3026 * ``min_th_y``: Minimum queue threshold for packet with yellow color
3027 * ``max_th_y``: Minimum queue threshold for packet with yellow color
3028 * ``maxp_inv_y``: Inverse of packet marking probability maximum value (maxp)
3029 * ``wq_log2_y``: Negated log2 of queue weight (wq)
3030 * ``color_r``: Packet color (red)
3031 * ``min_th_r``: Minimum queue threshold for packet with yellow color
3032 * ``max_th_r``: Minimum queue threshold for packet with yellow color
3033 * ``maxp_inv_r``: Inverse of packet marking probability maximum value (maxp)
3034 * ``wq_log2_r``: Negated log2 of queue weight (wq)
3036 Delete port traffic management WRED profile
3037 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3039 Delete the WRED profile::
3041 testpmd> del port tm node wred profile (port_id) (wred_profile_id)
3043 Add port traffic management hierarchy nonleaf node
3044 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3046 Add nonleaf node to port traffic management hierarchy::
3048 testpmd> add port tm nonleaf node (port_id) (node_id) (parent_node_id) \
3049 (priority) (weight) (level_id) (shaper_profile_id) \
3050 (n_sp_priorities) (stats_mask) (n_shared_shapers) \
3051 [(shared_shaper_0) (shared_shaper_1) ...] \
3055 * ``parent_node_id``: Node ID of the parent.
3056 * ``priority``: Node priority (highest node priority is zero). This is used by
3057 the SP algorithm running on the parent node for scheduling this node.
3058 * ``weight``: Node weight (lowest weight is one). The node weight is relative
3059 to the weight sum of all siblings that have the same priority. It is used by
3060 the WFQ algorithm running on the parent node for scheduling this node.
3061 * ``level_id``: Hierarchy level of the node.
3062 * ``shaper_profile_id``: Shaper profile ID of the private shaper to be used by
3064 * ``n_sp_priorities``: Number of strict priorities.
3065 * ``stats_mask``: Mask of statistics counter types to be enabled for this node.
3066 * ``n_shared_shapers``: Number of shared shapers.
3067 * ``shared_shaper_id``: Shared shaper id.
3069 Add port traffic management hierarchy nonleaf node with packet mode
3070 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3072 Add nonleaf node with packet mode to port traffic management hierarchy::
3074 testpmd> add port tm nonleaf node pktmode (port_id) (node_id) (parent_node_id) \
3075 (priority) (weight) (level_id) (shaper_profile_id) \
3076 (n_sp_priorities) (stats_mask) (n_shared_shapers) \
3077 [(shared_shaper_0) (shared_shaper_1) ...] \
3081 * ``parent_node_id``: Node ID of the parent.
3082 * ``priority``: Node priority (highest node priority is zero). This is used by
3083 the SP algorithm running on the parent node for scheduling this node.
3084 * ``weight``: Node weight (lowest weight is one). The node weight is relative
3085 to the weight sum of all siblings that have the same priority. It is used by
3086 the WFQ algorithm running on the parent node for scheduling this node.
3087 * ``level_id``: Hierarchy level of the node.
3088 * ``shaper_profile_id``: Shaper profile ID of the private shaper to be used by
3090 * ``n_sp_priorities``: Number of strict priorities. Packet mode is enabled on
3092 * ``stats_mask``: Mask of statistics counter types to be enabled for this node.
3093 * ``n_shared_shapers``: Number of shared shapers.
3094 * ``shared_shaper_id``: Shared shaper id.
3096 Add port traffic management hierarchy leaf node
3097 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3099 Add leaf node to port traffic management hierarchy::
3101 testpmd> add port tm leaf node (port_id) (node_id) (parent_node_id) \
3102 (priority) (weight) (level_id) (shaper_profile_id) \
3103 (cman_mode) (wred_profile_id) (stats_mask) (n_shared_shapers) \
3104 [(shared_shaper_id) (shared_shaper_id) ...] \
3108 * ``parent_node_id``: Node ID of the parent.
3109 * ``priority``: Node priority (highest node priority is zero). This is used by
3110 the SP algorithm running on the parent node for scheduling this node.
3111 * ``weight``: Node weight (lowest weight is one). The node weight is relative
3112 to the weight sum of all siblings that have the same priority. It is used by
3113 the WFQ algorithm running on the parent node for scheduling this node.
3114 * ``level_id``: Hierarchy level of the node.
3115 * ``shaper_profile_id``: Shaper profile ID of the private shaper to be used by
3117 * ``cman_mode``: Congestion management mode to be enabled for this node.
3118 * ``wred_profile_id``: WRED profile id to be enabled for this node.
3119 * ``stats_mask``: Mask of statistics counter types to be enabled for this node.
3120 * ``n_shared_shapers``: Number of shared shapers.
3121 * ``shared_shaper_id``: Shared shaper id.
3123 Delete port traffic management hierarchy node
3124 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3126 Delete node from port traffic management hierarchy::
3128 testpmd> del port tm node (port_id) (node_id)
3130 Update port traffic management hierarchy parent node
3131 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3133 Update port traffic management hierarchy parent node::
3135 testpmd> set port tm node parent (port_id) (node_id) (parent_node_id) \
3138 This function can only be called after the hierarchy commit invocation. Its
3139 success depends on the port support for this operation, as advertised through
3140 the port capability set. This function is valid for all nodes of the traffic
3141 management hierarchy except root node.
3143 Suspend port traffic management hierarchy node
3144 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3146 testpmd> suspend port tm node (port_id) (node_id)
3148 Resume port traffic management hierarchy node
3149 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3151 testpmd> resume port tm node (port_id) (node_id)
3153 Commit port traffic management hierarchy
3154 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3156 Commit the traffic management hierarchy on the port::
3158 testpmd> port tm hierarchy commit (port_id) (clean_on_fail)
3162 * ``clean_on_fail``: When set to non-zero, hierarchy is cleared on function
3163 call failure. On the other hand, hierarchy is preserved when this parameter
3166 Set port traffic management mark VLAN dei
3167 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3169 Enables/Disables the traffic management marking on the port for VLAN packets::
3171 testpmd> set port tm mark vlan_dei <port_id> <green> <yellow> <red>
3175 * ``port_id``: The port which on which VLAN packets marked as ``green`` or
3176 ``yellow`` or ``red`` will have dei bit enabled
3178 * ``green`` enable 1, disable 0 marking for dei bit of VLAN packets marked as green
3180 * ``yellow`` enable 1, disable 0 marking for dei bit of VLAN packets marked as yellow
3182 * ``red`` enable 1, disable 0 marking for dei bit of VLAN packets marked as red
3184 Set port traffic management mark IP dscp
3185 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3187 Enables/Disables the traffic management marking on the port for IP dscp packets::
3189 testpmd> set port tm mark ip_dscp <port_id> <green> <yellow> <red>
3193 * ``port_id``: The port which on which IP packets marked as ``green`` or
3194 ``yellow`` or ``red`` will have IP dscp bits updated
3196 * ``green`` enable 1, disable 0 marking IP dscp to low drop precedence for green packets
3198 * ``yellow`` enable 1, disable 0 marking IP dscp to medium drop precedence for yellow packets
3200 * ``red`` enable 1, disable 0 marking IP dscp to high drop precedence for red packets
3202 Set port traffic management mark IP ecn
3203 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3205 Enables/Disables the traffic management marking on the port for IP ecn packets::
3207 testpmd> set port tm mark ip_ecn <port_id> <green> <yellow> <red>
3211 * ``port_id``: The port which on which IP packets marked as ``green`` or
3212 ``yellow`` or ``red`` will have IP ecn bits updated
3214 * ``green`` enable 1, disable 0 marking IP ecn for green marked packets with ecn of 2'b01 or 2'b10
3215 to ecn of 2'b11 when IP is caring TCP or SCTP
3217 * ``yellow`` enable 1, disable 0 marking IP ecn for yellow marked packets with ecn of 2'b01 or 2'b10
3218 to ecn of 2'b11 when IP is caring TCP or SCTP
3220 * ``red`` enable 1, disable 0 marking IP ecn for yellow marked packets with ecn of 2'b01 or 2'b10
3221 to ecn of 2'b11 when IP is caring TCP or SCTP
3226 This section details the available filter functions that are available.
3228 Note these functions interface the deprecated legacy filtering framework,
3229 superseded by *rte_flow*. See `Flow rules management`_.
3232 ~~~~~~~~~~~~~~~~~~~~
3234 Add or delete a L2 Ethertype filter, which identify packets by their L2 Ethertype mainly assign them to a receive queue::
3236 ethertype_filter (port_id) (add|del) (mac_addr|mac_ignr) (mac_address) \
3237 ethertype (ether_type) (drop|fwd) queue (queue_id)
3239 The available information parameters are:
3241 * ``port_id``: The port which the Ethertype filter assigned on.
3243 * ``mac_addr``: Compare destination mac address.
3245 * ``mac_ignr``: Ignore destination mac address match.
3247 * ``mac_address``: Destination mac address to match.
3249 * ``ether_type``: The EtherType value want to match,
3250 for example 0x0806 for ARP packet. 0x0800 (IPv4) and 0x86DD (IPv6) are invalid.
3252 * ``queue_id``: The receive queue associated with this EtherType filter.
3253 It is meaningless when deleting or dropping.
3255 Example, to add/remove an ethertype filter rule::
3257 testpmd> ethertype_filter 0 add mac_ignr 00:11:22:33:44:55 \
3258 ethertype 0x0806 fwd queue 3
3260 testpmd> ethertype_filter 0 del mac_ignr 00:11:22:33:44:55 \
3261 ethertype 0x0806 fwd queue 3
3266 Add or delete a 2-tuple filter,
3267 which identifies packets by specific protocol and destination TCP/UDP port
3268 and forwards packets into one of the receive queues::
3270 2tuple_filter (port_id) (add|del) dst_port (dst_port_value) \
3271 protocol (protocol_value) mask (mask_value) \
3272 tcp_flags (tcp_flags_value) priority (prio_value) \
3275 The available information parameters are:
3277 * ``port_id``: The port which the 2-tuple filter assigned on.
3279 * ``dst_port_value``: Destination port in L4.
3281 * ``protocol_value``: IP L4 protocol.
3283 * ``mask_value``: Participates in the match or not by bit for field above, 1b means participate.
3285 * ``tcp_flags_value``: TCP control bits. The non-zero value is invalid, when the pro_value is not set to 0x06 (TCP).
3287 * ``prio_value``: Priority of this filter.
3289 * ``queue_id``: The receive queue associated with this 2-tuple filter.
3291 Example, to add/remove an 2tuple filter rule::
3293 testpmd> 2tuple_filter 0 add dst_port 32 protocol 0x06 mask 0x03 \
3294 tcp_flags 0x02 priority 3 queue 3
3296 testpmd> 2tuple_filter 0 del dst_port 32 protocol 0x06 mask 0x03 \
3297 tcp_flags 0x02 priority 3 queue 3
3302 Add or delete a 5-tuple filter,
3303 which consists of a 5-tuple (protocol, source and destination IP addresses, source and destination TCP/UDP/SCTP port)
3304 and routes packets into one of the receive queues::
3306 5tuple_filter (port_id) (add|del) dst_ip (dst_address) src_ip \
3307 (src_address) dst_port (dst_port_value) \
3308 src_port (src_port_value) protocol (protocol_value) \
3309 mask (mask_value) tcp_flags (tcp_flags_value) \
3310 priority (prio_value) queue (queue_id)
3312 The available information parameters are:
3314 * ``port_id``: The port which the 5-tuple filter assigned on.
3316 * ``dst_address``: Destination IP address.
3318 * ``src_address``: Source IP address.
3320 * ``dst_port_value``: TCP/UDP destination port.
3322 * ``src_port_value``: TCP/UDP source port.
3324 * ``protocol_value``: L4 protocol.
3326 * ``mask_value``: Participates in the match or not by bit for field above, 1b means participate
3328 * ``tcp_flags_value``: TCP control bits. The non-zero value is invalid, when the protocol_value is not set to 0x06 (TCP).
3330 * ``prio_value``: The priority of this filter.
3332 * ``queue_id``: The receive queue associated with this 5-tuple filter.
3334 Example, to add/remove an 5tuple filter rule::
3336 testpmd> 5tuple_filter 0 add dst_ip 2.2.2.5 src_ip 2.2.2.4 \
3337 dst_port 64 src_port 32 protocol 0x06 mask 0x1F \
3338 flags 0x0 priority 3 queue 3
3340 testpmd> 5tuple_filter 0 del dst_ip 2.2.2.5 src_ip 2.2.2.4 \
3341 dst_port 64 src_port 32 protocol 0x06 mask 0x1F \
3342 flags 0x0 priority 3 queue 3
3347 Using the SYN filter, TCP packets whose *SYN* flag is set can be forwarded to a separate queue::
3349 syn_filter (port_id) (add|del) priority (high|low) queue (queue_id)
3351 The available information parameters are:
3353 * ``port_id``: The port which the SYN filter assigned on.
3355 * ``high``: This SYN filter has higher priority than other filters.
3357 * ``low``: This SYN filter has lower priority than other filters.
3359 * ``queue_id``: The receive queue associated with this SYN filter
3363 testpmd> syn_filter 0 add priority high queue 3
3368 With flex filter, packets can be recognized by any arbitrary pattern within the first 128 bytes of the packet
3369 and routed into one of the receive queues::
3371 flex_filter (port_id) (add|del) len (len_value) bytes (bytes_value) \
3372 mask (mask_value) priority (prio_value) queue (queue_id)
3374 The available information parameters are:
3376 * ``port_id``: The port which the Flex filter is assigned on.
3378 * ``len_value``: Filter length in bytes, no greater than 128.
3380 * ``bytes_value``: A string in hexadecimal, means the value the flex filter needs to match.
3382 * ``mask_value``: A string in hexadecimal, bit 1 means corresponding byte participates in the match.
3384 * ``prio_value``: The priority of this filter.
3386 * ``queue_id``: The receive queue associated with this Flex filter.
3390 testpmd> flex_filter 0 add len 16 bytes 0x00000000000000000000000008060000 \
3391 mask 000C priority 3 queue 3
3393 testpmd> flex_filter 0 del len 16 bytes 0x00000000000000000000000008060000 \
3394 mask 000C priority 3 queue 3
3397 .. _testpmd_flow_director:
3399 flow_director_filter
3400 ~~~~~~~~~~~~~~~~~~~~
3402 The Flow Director works in receive mode to identify specific flows or sets of flows and route them to specific queues.
3404 Four types of filtering are supported which are referred to as Perfect Match, Signature, Perfect-mac-vlan and
3405 Perfect-tunnel filters, the match mode is set by the ``--pkt-filter-mode`` command-line parameter:
3407 * Perfect match filters.
3408 The hardware checks a match between the masked fields of the received packets and the programmed filters.
3409 The masked fields are for IP flow.
3411 * Signature filters.
3412 The hardware checks a match between a hash-based signature of the masked fields of the received packet.
3414 * Perfect-mac-vlan match filters.
3415 The hardware checks a match between the masked fields of the received packets and the programmed filters.
3416 The masked fields are for MAC VLAN flow.
3418 * Perfect-tunnel match filters.
3419 The hardware checks a match between the masked fields of the received packets and the programmed filters.
3420 The masked fields are for tunnel flow.
3422 * Perfect-raw-flow-type match filters.
3423 The hardware checks a match between the masked fields of the received packets and pre-loaded raw (template) packet.
3424 The masked fields are specified by input sets.
3426 The Flow Director filters can match the different fields for different type of packet: flow type, specific input set
3427 per flow type and the flexible payload.
3429 The Flow Director can also mask out parts of all of these fields so that filters
3430 are only applied to certain fields or parts of the fields.
3432 Note that for raw flow type mode the source and destination fields in the
3433 raw packet buffer need to be presented in a reversed order with respect
3434 to the expected received packets.
3435 For example: IP source and destination addresses or TCP/UDP/SCTP
3436 source and destination ports
3438 Different NICs may have different capabilities, command show port fdir (port_id) can be used to acquire the information.
3440 # Commands to add flow director filters of different flow types::
3442 flow_director_filter (port_id) mode IP (add|del|update) \
3443 flow (ipv4-other|ipv4-frag|ipv6-other|ipv6-frag) \
3444 src (src_ip_address) dst (dst_ip_address) \
3445 tos (tos_value) proto (proto_value) ttl (ttl_value) \
3446 vlan (vlan_value) flexbytes (flexbytes_value) \
3447 (drop|fwd) pf|vf(vf_id) queue (queue_id) \
3450 flow_director_filter (port_id) mode IP (add|del|update) \
3451 flow (ipv4-tcp|ipv4-udp|ipv6-tcp|ipv6-udp) \
3452 src (src_ip_address) (src_port) \
3453 dst (dst_ip_address) (dst_port) \
3454 tos (tos_value) ttl (ttl_value) \
3455 vlan (vlan_value) flexbytes (flexbytes_value) \
3456 (drop|fwd) queue pf|vf(vf_id) (queue_id) \
3459 flow_director_filter (port_id) mode IP (add|del|update) \
3460 flow (ipv4-sctp|ipv6-sctp) \
3461 src (src_ip_address) (src_port) \
3462 dst (dst_ip_address) (dst_port) \
3463 tos (tos_value) ttl (ttl_value) \
3464 tag (verification_tag) vlan (vlan_value) \
3465 flexbytes (flexbytes_value) (drop|fwd) \
3466 pf|vf(vf_id) queue (queue_id) fd_id (fd_id_value)
3468 flow_director_filter (port_id) mode IP (add|del|update) flow l2_payload \
3469 ether (ethertype) flexbytes (flexbytes_value) \
3470 (drop|fwd) pf|vf(vf_id) queue (queue_id)
3473 flow_director_filter (port_id) mode MAC-VLAN (add|del|update) \
3474 mac (mac_address) vlan (vlan_value) \
3475 flexbytes (flexbytes_value) (drop|fwd) \
3476 queue (queue_id) fd_id (fd_id_value)
3478 flow_director_filter (port_id) mode Tunnel (add|del|update) \
3479 mac (mac_address) vlan (vlan_value) \
3480 tunnel (NVGRE|VxLAN) tunnel-id (tunnel_id_value) \
3481 flexbytes (flexbytes_value) (drop|fwd) \
3482 queue (queue_id) fd_id (fd_id_value)
3484 flow_director_filter (port_id) mode raw (add|del|update) flow (flow_id) \
3485 (drop|fwd) queue (queue_id) fd_id (fd_id_value) \
3486 packet (packet file name)
3488 For example, to add an ipv4-udp flow type filter::
3490 testpmd> flow_director_filter 0 mode IP add flow ipv4-udp src 2.2.2.3 32 \
3491 dst 2.2.2.5 33 tos 2 ttl 40 vlan 0x1 flexbytes (0x88,0x48) \
3492 fwd pf queue 1 fd_id 1
3494 For example, add an ipv4-other flow type filter::
3496 testpmd> flow_director_filter 0 mode IP add flow ipv4-other src 2.2.2.3 \
3497 dst 2.2.2.5 tos 2 proto 20 ttl 40 vlan 0x1 \
3498 flexbytes (0x88,0x48) fwd pf queue 1 fd_id 1
3503 Flush all flow director filters on a device::
3505 testpmd> flush_flow_director (port_id)
3507 Example, to flush all flow director filter on port 0::
3509 testpmd> flush_flow_director 0
3514 Set flow director's input masks::
3516 flow_director_mask (port_id) mode IP vlan (vlan_value) \
3517 src_mask (ipv4_src) (ipv6_src) (src_port) \
3518 dst_mask (ipv4_dst) (ipv6_dst) (dst_port)
3520 flow_director_mask (port_id) mode MAC-VLAN vlan (vlan_value)
3522 flow_director_mask (port_id) mode Tunnel vlan (vlan_value) \
3523 mac (mac_value) tunnel-type (tunnel_type_value) \
3524 tunnel-id (tunnel_id_value)
3526 Example, to set flow director mask on port 0::
3528 testpmd> flow_director_mask 0 mode IP vlan 0xefff \
3529 src_mask 255.255.255.255 \
3530 FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF \
3531 dst_mask 255.255.255.255 \
3532 FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF
3534 flow_director_flex_mask
3535 ~~~~~~~~~~~~~~~~~~~~~~~
3537 set masks of flow director's flexible payload based on certain flow type::
3539 testpmd> flow_director_flex_mask (port_id) \
3540 flow (none|ipv4-other|ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
3541 ipv6-other|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp| \
3542 l2_payload|all) (mask)
3544 Example, to set flow director's flex mask for all flow type on port 0::
3546 testpmd> flow_director_flex_mask 0 flow all \
3547 (0xff,0xff,0,0,0,0,0,0,0,0,0,0,0,0,0,0)
3550 flow_director_flex_payload
3551 ~~~~~~~~~~~~~~~~~~~~~~~~~~
3553 Configure flexible payload selection::
3555 flow_director_flex_payload (port_id) (raw|l2|l3|l4) (config)
3557 For example, to select the first 16 bytes from the offset 4 (bytes) of packet's payload as flexible payload::
3559 testpmd> flow_director_flex_payload 0 l4 \
3560 (4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19)
3562 get_sym_hash_ena_per_port
3563 ~~~~~~~~~~~~~~~~~~~~~~~~~
3565 Get symmetric hash enable configuration per port::
3567 get_sym_hash_ena_per_port (port_id)
3569 For example, to get symmetric hash enable configuration of port 1::
3571 testpmd> get_sym_hash_ena_per_port 1
3573 set_sym_hash_ena_per_port
3574 ~~~~~~~~~~~~~~~~~~~~~~~~~
3576 Set symmetric hash enable configuration per port to enable or disable::
3578 set_sym_hash_ena_per_port (port_id) (enable|disable)
3580 For example, to set symmetric hash enable configuration of port 1 to enable::
3582 testpmd> set_sym_hash_ena_per_port 1 enable
3584 get_hash_global_config
3585 ~~~~~~~~~~~~~~~~~~~~~~
3587 Get the global configurations of hash filters::
3589 get_hash_global_config (port_id)
3591 For example, to get the global configurations of hash filters of port 1::
3593 testpmd> get_hash_global_config 1
3595 set_hash_global_config
3596 ~~~~~~~~~~~~~~~~~~~~~~
3598 Set the global configurations of hash filters::
3600 set_hash_global_config (port_id) (toeplitz|simple_xor|symmetric_toeplitz|default) \
3601 (ipv4|ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp|ipv4-other|ipv6|ipv6-frag| \
3602 ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other|l2_payload|<flow_id>) \
3605 For example, to enable simple_xor for flow type of ipv6 on port 2::
3607 testpmd> set_hash_global_config 2 simple_xor ipv6 enable
3612 Set the input set for hash::
3614 set_hash_input_set (port_id) (ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
3615 ipv4-other|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other| \
3616 l2_payload|<flow_id>) (ovlan|ivlan|src-ipv4|dst-ipv4|src-ipv6|dst-ipv6| \
3617 ipv4-tos|ipv4-proto|ipv6-tc|ipv6-next-header|udp-src-port|udp-dst-port| \
3618 tcp-src-port|tcp-dst-port|sctp-src-port|sctp-dst-port|sctp-veri-tag| \
3619 udp-key|gre-key|fld-1st|fld-2nd|fld-3rd|fld-4th|fld-5th|fld-6th|fld-7th| \
3620 fld-8th|none) (select|add)
3622 For example, to add source IP to hash input set for flow type of ipv4-udp on port 0::
3624 testpmd> set_hash_input_set 0 ipv4-udp src-ipv4 add
3629 The Flow Director filters can match the different fields for different type of packet, i.e. specific input set
3630 on per flow type and the flexible payload. This command can be used to change input set for each flow type.
3632 Set the input set for flow director::
3634 set_fdir_input_set (port_id) (ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
3635 ipv4-other|ipv6|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other| \
3636 l2_payload|<flow_id>) (ivlan|ethertype|src-ipv4|dst-ipv4|src-ipv6|dst-ipv6| \
3637 ipv4-tos|ipv4-proto|ipv4-ttl|ipv6-tc|ipv6-next-header|ipv6-hop-limits| \
3638 tudp-src-port|udp-dst-port|cp-src-port|tcp-dst-port|sctp-src-port| \
3639 sctp-dst-port|sctp-veri-tag|none) (select|add)
3641 For example to add source IP to FD input set for flow type of ipv4-udp on port 0::
3643 testpmd> set_fdir_input_set 0 ipv4-udp src-ipv4 add
3648 Set different GRE key length for input set::
3650 global_config (port_id) gre-key-len (number in bytes)
3652 For example to set GRE key length for input set to 4 bytes on port 0::
3654 testpmd> global_config 0 gre-key-len 4
3657 .. _testpmd_rte_flow:
3659 Flow rules management
3660 ---------------------
3662 Control of the generic flow API (*rte_flow*) is fully exposed through the
3663 ``flow`` command (validation, creation, destruction, queries and operation
3666 Considering *rte_flow* overlaps with all `Filter Functions`_, using both
3667 features simultaneously may cause undefined side-effects and is therefore
3673 Because the ``flow`` command uses dynamic tokens to handle the large number
3674 of possible flow rules combinations, its behavior differs slightly from
3675 other commands, in particular:
3677 - Pressing *?* or the *<tab>* key displays contextual help for the current
3678 token, not that of the entire command.
3680 - Optional and repeated parameters are supported (provided they are listed
3681 in the contextual help).
3683 The first parameter stands for the operation mode. Possible operations and
3684 their general syntax are described below. They are covered in detail in the
3687 - Check whether a flow rule can be created::
3689 flow validate {port_id}
3690 [group {group_id}] [priority {level}] [ingress] [egress] [transfer]
3691 pattern {item} [/ {item} [...]] / end
3692 actions {action} [/ {action} [...]] / end
3694 - Create a flow rule::
3696 flow create {port_id}
3697 [group {group_id}] [priority {level}] [ingress] [egress] [transfer]
3698 pattern {item} [/ {item} [...]] / end
3699 actions {action} [/ {action} [...]] / end
3701 - Destroy specific flow rules::
3703 flow destroy {port_id} rule {rule_id} [...]
3705 - Destroy all flow rules::
3707 flow flush {port_id}
3709 - Query an existing flow rule::
3711 flow query {port_id} {rule_id} {action}
3713 - List existing flow rules sorted by priority, filtered by group
3716 flow list {port_id} [group {group_id}] [...]
3718 - Restrict ingress traffic to the defined flow rules::
3720 flow isolate {port_id} {boolean}
3722 - Dump internal representation information of all flows in hardware::
3724 flow dump {port_id} {output_file}
3726 - List and destroy aged flow rules::
3728 flow aged {port_id} [destroy]
3730 Validating flow rules
3731 ~~~~~~~~~~~~~~~~~~~~~
3733 ``flow validate`` reports whether a flow rule would be accepted by the
3734 underlying device in its current state but stops short of creating it. It is
3735 bound to ``rte_flow_validate()``::
3737 flow validate {port_id}
3738 [group {group_id}] [priority {level}] [ingress] [egress] [transfer]
3739 pattern {item} [/ {item} [...]] / end
3740 actions {action} [/ {action} [...]] / end
3742 If successful, it will show::
3746 Otherwise it will show an error message of the form::
3748 Caught error type [...] ([...]): [...]
3750 This command uses the same parameters as ``flow create``, their format is
3751 described in `Creating flow rules`_.
3753 Check whether redirecting any Ethernet packet received on port 0 to RX queue
3754 index 6 is supported::
3756 testpmd> flow validate 0 ingress pattern eth / end
3757 actions queue index 6 / end
3761 Port 0 does not support TCPv6 rules::
3763 testpmd> flow validate 0 ingress pattern eth / ipv6 / tcp / end
3765 Caught error type 9 (specific pattern item): Invalid argument
3771 ``flow create`` validates and creates the specified flow rule. It is bound
3772 to ``rte_flow_create()``::
3774 flow create {port_id}
3775 [group {group_id}] [priority {level}] [ingress] [egress] [transfer]
3776 pattern {item} [/ {item} [...]] / end
3777 actions {action} [/ {action} [...]] / end
3779 If successful, it will return a flow rule ID usable with other commands::
3781 Flow rule #[...] created
3783 Otherwise it will show an error message of the form::
3785 Caught error type [...] ([...]): [...]
3787 Parameters describe in the following order:
3789 - Attributes (*group*, *priority*, *ingress*, *egress*, *transfer* tokens).
3790 - A matching pattern, starting with the *pattern* token and terminated by an
3792 - Actions, starting with the *actions* token and terminated by an *end*
3795 These translate directly to *rte_flow* objects provided as-is to the
3796 underlying functions.
3798 The shortest valid definition only comprises mandatory tokens::
3800 testpmd> flow create 0 pattern end actions end
3802 Note that PMDs may refuse rules that essentially do nothing such as this
3805 **All unspecified object values are automatically initialized to 0.**
3810 These tokens affect flow rule attributes (``struct rte_flow_attr``) and are
3811 specified before the ``pattern`` token.
3813 - ``group {group id}``: priority group.
3814 - ``priority {level}``: priority level within group.
3815 - ``ingress``: rule applies to ingress traffic.
3816 - ``egress``: rule applies to egress traffic.
3817 - ``transfer``: apply rule directly to endpoints found in pattern.
3819 Each instance of an attribute specified several times overrides the previous
3820 value as shown below (group 4 is used)::
3822 testpmd> flow create 0 group 42 group 24 group 4 [...]
3824 Note that once enabled, ``ingress`` and ``egress`` cannot be disabled.
3826 While not specifying a direction is an error, some rules may allow both
3829 Most rules affect RX therefore contain the ``ingress`` token::
3831 testpmd> flow create 0 ingress pattern [...]
3836 A matching pattern starts after the ``pattern`` token. It is made of pattern
3837 items and is terminated by a mandatory ``end`` item.
3839 Items are named after their type (*RTE_FLOW_ITEM_TYPE_* from ``enum
3840 rte_flow_item_type``).
3842 The ``/`` token is used as a separator between pattern items as shown
3845 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end [...]
3847 Note that protocol items like these must be stacked from lowest to highest
3848 layer to make sense. For instance, the following rule is either invalid or
3849 unlikely to match any packet::
3851 testpmd> flow create 0 ingress pattern eth / udp / ipv4 / end [...]
3853 More information on these restrictions can be found in the *rte_flow*
3856 Several items support additional specification structures, for example
3857 ``ipv4`` allows specifying source and destination addresses as follows::
3859 testpmd> flow create 0 ingress pattern eth / ipv4 src is 10.1.1.1
3860 dst is 10.2.0.0 / end [...]
3862 This rule matches all IPv4 traffic with the specified properties.
3864 In this example, ``src`` and ``dst`` are field names of the underlying
3865 ``struct rte_flow_item_ipv4`` object. All item properties can be specified
3866 in a similar fashion.
3868 The ``is`` token means that the subsequent value must be matched exactly,
3869 and assigns ``spec`` and ``mask`` fields in ``struct rte_flow_item``
3870 accordingly. Possible assignment tokens are:
3872 - ``is``: match value perfectly (with full bit-mask).
3873 - ``spec``: match value according to configured bit-mask.
3874 - ``last``: specify upper bound to establish a range.
3875 - ``mask``: specify bit-mask with relevant bits set to one.
3876 - ``prefix``: generate bit-mask with <prefix-length> most-significant bits set to one.
3878 These yield identical results::
3880 ipv4 src is 10.1.1.1
3884 ipv4 src spec 10.1.1.1 src mask 255.255.255.255
3888 ipv4 src spec 10.1.1.1 src prefix 32
3892 ipv4 src is 10.1.1.1 src last 10.1.1.1 # range with a single value
3896 ipv4 src is 10.1.1.1 src last 0 # 0 disables range
3898 Inclusive ranges can be defined with ``last``::
3900 ipv4 src is 10.1.1.1 src last 10.2.3.4 # 10.1.1.1 to 10.2.3.4
3902 Note that ``mask`` affects both ``spec`` and ``last``::
3904 ipv4 src is 10.1.1.1 src last 10.2.3.4 src mask 255.255.0.0
3905 # matches 10.1.0.0 to 10.2.255.255
3907 Properties can be modified multiple times::
3909 ipv4 src is 10.1.1.1 src is 10.1.2.3 src is 10.2.3.4 # matches 10.2.3.4
3913 ipv4 src is 10.1.1.1 src prefix 24 src prefix 16 # matches 10.1.0.0/16
3918 This section lists supported pattern items and their attributes, if any.
3920 - ``end``: end list of pattern items.
3922 - ``void``: no-op pattern item.
3924 - ``invert``: perform actions when pattern does not match.
3926 - ``any``: match any protocol for the current layer.
3928 - ``num {unsigned}``: number of layers covered.
3930 - ``pf``: match traffic from/to the physical function.
3932 - ``vf``: match traffic from/to a virtual function ID.
3934 - ``id {unsigned}``: VF ID.
3936 - ``phy_port``: match traffic from/to a specific physical port.
3938 - ``index {unsigned}``: physical port index.
3940 - ``port_id``: match traffic from/to a given DPDK port ID.
3942 - ``id {unsigned}``: DPDK port ID.
3944 - ``mark``: match value set in previously matched flow rule using the mark action.
3946 - ``id {unsigned}``: arbitrary integer value.
3948 - ``raw``: match an arbitrary byte string.
3950 - ``relative {boolean}``: look for pattern after the previous item.
3951 - ``search {boolean}``: search pattern from offset (see also limit).
3952 - ``offset {integer}``: absolute or relative offset for pattern.
3953 - ``limit {unsigned}``: search area limit for start of pattern.
3954 - ``pattern {string}``: byte string to look for.
3956 - ``eth``: match Ethernet header.
3958 - ``dst {MAC-48}``: destination MAC.
3959 - ``src {MAC-48}``: source MAC.
3960 - ``type {unsigned}``: EtherType or TPID.
3962 - ``vlan``: match 802.1Q/ad VLAN tag.
3964 - ``tci {unsigned}``: tag control information.
3965 - ``pcp {unsigned}``: priority code point.
3966 - ``dei {unsigned}``: drop eligible indicator.
3967 - ``vid {unsigned}``: VLAN identifier.
3968 - ``inner_type {unsigned}``: inner EtherType or TPID.
3970 - ``ipv4``: match IPv4 header.
3972 - ``tos {unsigned}``: type of service.
3973 - ``ttl {unsigned}``: time to live.
3974 - ``proto {unsigned}``: next protocol ID.
3975 - ``src {ipv4 address}``: source address.
3976 - ``dst {ipv4 address}``: destination address.
3978 - ``ipv6``: match IPv6 header.
3980 - ``tc {unsigned}``: traffic class.
3981 - ``flow {unsigned}``: flow label.
3982 - ``proto {unsigned}``: protocol (next header).
3983 - ``hop {unsigned}``: hop limit.
3984 - ``src {ipv6 address}``: source address.
3985 - ``dst {ipv6 address}``: destination address.
3987 - ``icmp``: match ICMP header.
3989 - ``type {unsigned}``: ICMP packet type.
3990 - ``code {unsigned}``: ICMP packet code.
3992 - ``udp``: match UDP header.
3994 - ``src {unsigned}``: UDP source port.
3995 - ``dst {unsigned}``: UDP destination port.
3997 - ``tcp``: match TCP header.
3999 - ``src {unsigned}``: TCP source port.
4000 - ``dst {unsigned}``: TCP destination port.
4002 - ``sctp``: match SCTP header.
4004 - ``src {unsigned}``: SCTP source port.
4005 - ``dst {unsigned}``: SCTP destination port.
4006 - ``tag {unsigned}``: validation tag.
4007 - ``cksum {unsigned}``: checksum.
4009 - ``vxlan``: match VXLAN header.
4011 - ``vni {unsigned}``: VXLAN identifier.
4013 - ``e_tag``: match IEEE 802.1BR E-Tag header.
4015 - ``grp_ecid_b {unsigned}``: GRP and E-CID base.
4017 - ``nvgre``: match NVGRE header.
4019 - ``tni {unsigned}``: virtual subnet ID.
4021 - ``mpls``: match MPLS header.
4023 - ``label {unsigned}``: MPLS label.
4025 - ``gre``: match GRE header.
4027 - ``protocol {unsigned}``: protocol type.
4029 - ``gre_key``: match GRE optional key field.
4031 - ``value {unsigned}``: key value.
4033 - ``fuzzy``: fuzzy pattern match, expect faster than default.
4035 - ``thresh {unsigned}``: accuracy threshold.
4037 - ``gtp``, ``gtpc``, ``gtpu``: match GTPv1 header.
4039 - ``teid {unsigned}``: tunnel endpoint identifier.
4041 - ``geneve``: match GENEVE header.
4043 - ``vni {unsigned}``: virtual network identifier.
4044 - ``protocol {unsigned}``: protocol type.
4046 - ``vxlan-gpe``: match VXLAN-GPE header.
4048 - ``vni {unsigned}``: VXLAN-GPE identifier.
4050 - ``arp_eth_ipv4``: match ARP header for Ethernet/IPv4.
4052 - ``sha {MAC-48}``: sender hardware address.
4053 - ``spa {ipv4 address}``: sender IPv4 address.
4054 - ``tha {MAC-48}``: target hardware address.
4055 - ``tpa {ipv4 address}``: target IPv4 address.
4057 - ``ipv6_ext``: match presence of any IPv6 extension header.
4059 - ``next_hdr {unsigned}``: next header.
4061 - ``icmp6``: match any ICMPv6 header.
4063 - ``type {unsigned}``: ICMPv6 type.
4064 - ``code {unsigned}``: ICMPv6 code.
4066 - ``icmp6_nd_ns``: match ICMPv6 neighbor discovery solicitation.
4068 - ``target_addr {ipv6 address}``: target address.
4070 - ``icmp6_nd_na``: match ICMPv6 neighbor discovery advertisement.
4072 - ``target_addr {ipv6 address}``: target address.
4074 - ``icmp6_nd_opt``: match presence of any ICMPv6 neighbor discovery option.
4076 - ``type {unsigned}``: ND option type.
4078 - ``icmp6_nd_opt_sla_eth``: match ICMPv6 neighbor discovery source Ethernet
4079 link-layer address option.
4081 - ``sla {MAC-48}``: source Ethernet LLA.
4083 - ``icmp6_nd_opt_tla_eth``: match ICMPv6 neighbor discovery target Ethernet
4084 link-layer address option.
4086 - ``tla {MAC-48}``: target Ethernet LLA.
4088 - ``meta``: match application specific metadata.
4090 - ``data {unsigned}``: metadata value.
4092 - ``gtp_psc``: match GTP PDU extension header with type 0x85.
4094 - ``pdu_type {unsigned}``: PDU type.
4095 - ``qfi {unsigned}``: QoS flow identifier.
4097 - ``pppoes``, ``pppoed``: match PPPoE header.
4099 - ``session_id {unsigned}``: session identifier.
4101 - ``pppoe_proto_id``: match PPPoE session protocol identifier.
4103 - ``proto_id {unsigned}``: PPP protocol identifier.
4105 - ``l2tpv3oip``: match L2TPv3 over IP header.
4107 - ``session_id {unsigned}``: L2TPv3 over IP session identifier.
4109 - ``ah``: match AH header.
4111 - ``spi {unsigned}``: security parameters index.
4113 - ``pfcp``: match PFCP header.
4115 - ``s_field {unsigned}``: S field.
4116 - ``seid {unsigned}``: session endpoint identifier.
4121 A list of actions starts after the ``actions`` token in the same fashion as
4122 `Matching pattern`_; actions are separated by ``/`` tokens and the list is
4123 terminated by a mandatory ``end`` action.
4125 Actions are named after their type (*RTE_FLOW_ACTION_TYPE_* from ``enum
4126 rte_flow_action_type``).
4128 Dropping all incoming UDPv4 packets can be expressed as follows::
4130 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
4133 Several actions have configurable properties which must be specified when
4134 there is no valid default value. For example, ``queue`` requires a target
4137 This rule redirects incoming UDPv4 traffic to queue index 6::
4139 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
4140 actions queue index 6 / end
4142 While this one could be rejected by PMDs (unspecified queue index)::
4144 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
4147 As defined by *rte_flow*, the list is not ordered, all actions of a given
4148 rule are performed simultaneously. These are equivalent::
4150 queue index 6 / void / mark id 42 / end
4154 void / mark id 42 / queue index 6 / end
4156 All actions in a list should have different types, otherwise only the last
4157 action of a given type is taken into account::
4159 queue index 4 / queue index 5 / queue index 6 / end # will use queue 6
4163 drop / drop / drop / end # drop is performed only once
4167 mark id 42 / queue index 3 / mark id 24 / end # mark will be 24
4169 Considering they are performed simultaneously, opposite and overlapping
4170 actions can sometimes be combined when the end result is unambiguous::
4172 drop / queue index 6 / end # drop has no effect
4176 queue index 6 / rss queues 6 7 8 / end # queue has no effect
4180 drop / passthru / end # drop has no effect
4182 Note that PMDs may still refuse such combinations.
4187 This section lists supported actions and their attributes, if any.
4189 - ``end``: end list of actions.
4191 - ``void``: no-op action.
4193 - ``passthru``: let subsequent rule process matched packets.
4195 - ``jump``: redirect traffic to group on device.
4197 - ``group {unsigned}``: group to redirect to.
4199 - ``mark``: attach 32 bit value to packets.
4201 - ``id {unsigned}``: 32 bit value to return with packets.
4203 - ``flag``: flag packets.
4205 - ``queue``: assign packets to a given queue index.
4207 - ``index {unsigned}``: queue index to use.
4209 - ``drop``: drop packets (note: passthru has priority).
4211 - ``count``: enable counters for this rule.
4213 - ``rss``: spread packets among several queues.
4215 - ``func {hash function}``: RSS hash function to apply, allowed tokens are
4216 the same as `set_hash_global_config`_.
4218 - ``level {unsigned}``: encapsulation level for ``types``.
4220 - ``types [{RSS hash type} [...]] end``: specific RSS hash types, allowed
4221 tokens are the same as `set_hash_input_set`_, except that an empty list
4222 does not disable RSS but instead requests unspecified "best-effort"
4225 - ``key {string}``: RSS hash key, overrides ``key_len``.
4227 - ``key_len {unsigned}``: RSS hash key length in bytes, can be used in
4228 conjunction with ``key`` to pad or truncate it.
4230 - ``queues [{unsigned} [...]] end``: queue indices to use.
4232 - ``pf``: direct traffic to physical function.
4234 - ``vf``: direct traffic to a virtual function ID.
4236 - ``original {boolean}``: use original VF ID if possible.
4237 - ``id {unsigned}``: VF ID.
4239 - ``phy_port``: direct packets to physical port index.
4241 - ``original {boolean}``: use original port index if possible.
4242 - ``index {unsigned}``: physical port index.
4244 - ``port_id``: direct matching traffic to a given DPDK port ID.
4246 - ``original {boolean}``: use original DPDK port ID if possible.
4247 - ``id {unsigned}``: DPDK port ID.
4249 - ``of_set_mpls_ttl``: OpenFlow's ``OFPAT_SET_MPLS_TTL``.
4251 - ``mpls_ttl``: MPLS TTL.
4253 - ``of_dec_mpls_ttl``: OpenFlow's ``OFPAT_DEC_MPLS_TTL``.
4255 - ``of_set_nw_ttl``: OpenFlow's ``OFPAT_SET_NW_TTL``.
4257 - ``nw_ttl``: IP TTL.
4259 - ``of_dec_nw_ttl``: OpenFlow's ``OFPAT_DEC_NW_TTL``.
4261 - ``of_copy_ttl_out``: OpenFlow's ``OFPAT_COPY_TTL_OUT``.
4263 - ``of_copy_ttl_in``: OpenFlow's ``OFPAT_COPY_TTL_IN``.
4265 - ``of_pop_vlan``: OpenFlow's ``OFPAT_POP_VLAN``.
4267 - ``of_push_vlan``: OpenFlow's ``OFPAT_PUSH_VLAN``.
4269 - ``ethertype``: Ethertype.
4271 - ``of_set_vlan_vid``: OpenFlow's ``OFPAT_SET_VLAN_VID``.
4273 - ``vlan_vid``: VLAN id.
4275 - ``of_set_vlan_pcp``: OpenFlow's ``OFPAT_SET_VLAN_PCP``.
4277 - ``vlan_pcp``: VLAN priority.
4279 - ``of_pop_mpls``: OpenFlow's ``OFPAT_POP_MPLS``.
4281 - ``ethertype``: Ethertype.
4283 - ``of_push_mpls``: OpenFlow's ``OFPAT_PUSH_MPLS``.
4285 - ``ethertype``: Ethertype.
4287 - ``vxlan_encap``: Performs a VXLAN encapsulation, outer layer configuration
4288 is done through `Config VXLAN Encap outer layers`_.
4290 - ``vxlan_decap``: Performs a decapsulation action by stripping all headers of
4291 the VXLAN tunnel network overlay from the matched flow.
4293 - ``nvgre_encap``: Performs a NVGRE encapsulation, outer layer configuration
4294 is done through `Config NVGRE Encap outer layers`_.
4296 - ``nvgre_decap``: Performs a decapsulation action by stripping all headers of
4297 the NVGRE tunnel network overlay from the matched flow.
4299 - ``l2_encap``: Performs a L2 encapsulation, L2 configuration
4300 is done through `Config L2 Encap`_.
4302 - ``l2_decap``: Performs a L2 decapsulation, L2 configuration
4303 is done through `Config L2 Decap`_.
4305 - ``mplsogre_encap``: Performs a MPLSoGRE encapsulation, outer layer
4306 configuration is done through `Config MPLSoGRE Encap outer layers`_.
4308 - ``mplsogre_decap``: Performs a MPLSoGRE decapsulation, outer layer
4309 configuration is done through `Config MPLSoGRE Decap outer layers`_.
4311 - ``mplsoudp_encap``: Performs a MPLSoUDP encapsulation, outer layer
4312 configuration is done through `Config MPLSoUDP Encap outer layers`_.
4314 - ``mplsoudp_decap``: Performs a MPLSoUDP decapsulation, outer layer
4315 configuration is done through `Config MPLSoUDP Decap outer layers`_.
4317 - ``set_ipv4_src``: Set a new IPv4 source address in the outermost IPv4 header.
4319 - ``ipv4_addr``: New IPv4 source address.
4321 - ``set_ipv4_dst``: Set a new IPv4 destination address in the outermost IPv4
4324 - ``ipv4_addr``: New IPv4 destination address.
4326 - ``set_ipv6_src``: Set a new IPv6 source address in the outermost IPv6 header.
4328 - ``ipv6_addr``: New IPv6 source address.
4330 - ``set_ipv6_dst``: Set a new IPv6 destination address in the outermost IPv6
4333 - ``ipv6_addr``: New IPv6 destination address.
4335 - ``set_tp_src``: Set a new source port number in the outermost TCP/UDP
4338 - ``port``: New TCP/UDP source port number.
4340 - ``set_tp_dst``: Set a new destination port number in the outermost TCP/UDP
4343 - ``port``: New TCP/UDP destination port number.
4345 - ``mac_swap``: Swap the source and destination MAC addresses in the outermost
4348 - ``dec_ttl``: Performs a decrease TTL value action
4350 - ``set_ttl``: Set TTL value with specified value
4351 - ``ttl_value {unsigned}``: The new TTL value to be set
4353 - ``set_mac_src``: set source MAC address
4355 - ``mac_addr {MAC-48}``: new source MAC address
4357 - ``set_mac_dst``: set destination MAC address
4359 - ``mac_addr {MAC-48}``: new destination MAC address
4361 - ``inc_tcp_seq``: Increase sequence number in the outermost TCP header.
4363 - ``value {unsigned}``: Value to increase TCP sequence number by.
4365 - ``dec_tcp_seq``: Decrease sequence number in the outermost TCP header.
4367 - ``value {unsigned}``: Value to decrease TCP sequence number by.
4369 - ``inc_tcp_ack``: Increase acknowledgment number in the outermost TCP header.
4371 - ``value {unsigned}``: Value to increase TCP acknowledgment number by.
4373 - ``dec_tcp_ack``: Decrease acknowledgment number in the outermost TCP header.
4375 - ``value {unsigned}``: Value to decrease TCP acknowledgment number by.
4377 - ``set_ipv4_dscp``: Set IPv4 DSCP value with specified value
4379 - ``dscp_value {unsigned}``: The new DSCP value to be set
4381 - ``set_ipv6_dscp``: Set IPv6 DSCP value with specified value
4383 - ``dscp_value {unsigned}``: The new DSCP value to be set
4385 Destroying flow rules
4386 ~~~~~~~~~~~~~~~~~~~~~
4388 ``flow destroy`` destroys one or more rules from their rule ID (as returned
4389 by ``flow create``), this command calls ``rte_flow_destroy()`` as many
4390 times as necessary::
4392 flow destroy {port_id} rule {rule_id} [...]
4394 If successful, it will show::
4396 Flow rule #[...] destroyed
4398 It does not report anything for rule IDs that do not exist. The usual error
4399 message is shown when a rule cannot be destroyed::
4401 Caught error type [...] ([...]): [...]
4403 ``flow flush`` destroys all rules on a device and does not take extra
4404 arguments. It is bound to ``rte_flow_flush()``::
4406 flow flush {port_id}
4408 Any errors are reported as above.
4410 Creating several rules and destroying them::
4412 testpmd> flow create 0 ingress pattern eth / ipv6 / end
4413 actions queue index 2 / end
4414 Flow rule #0 created
4415 testpmd> flow create 0 ingress pattern eth / ipv4 / end
4416 actions queue index 3 / end
4417 Flow rule #1 created
4418 testpmd> flow destroy 0 rule 0 rule 1
4419 Flow rule #1 destroyed
4420 Flow rule #0 destroyed
4423 The same result can be achieved using ``flow flush``::
4425 testpmd> flow create 0 ingress pattern eth / ipv6 / end
4426 actions queue index 2 / end
4427 Flow rule #0 created
4428 testpmd> flow create 0 ingress pattern eth / ipv4 / end
4429 actions queue index 3 / end
4430 Flow rule #1 created
4431 testpmd> flow flush 0
4434 Non-existent rule IDs are ignored::
4436 testpmd> flow create 0 ingress pattern eth / ipv6 / end
4437 actions queue index 2 / end
4438 Flow rule #0 created
4439 testpmd> flow create 0 ingress pattern eth / ipv4 / end
4440 actions queue index 3 / end
4441 Flow rule #1 created
4442 testpmd> flow destroy 0 rule 42 rule 10 rule 2
4444 testpmd> flow destroy 0 rule 0
4445 Flow rule #0 destroyed
4451 ``flow query`` queries a specific action of a flow rule having that
4452 ability. Such actions collect information that can be reported using this
4453 command. It is bound to ``rte_flow_query()``::
4455 flow query {port_id} {rule_id} {action}
4457 If successful, it will display either the retrieved data for known actions
4458 or the following message::
4460 Cannot display result for action type [...] ([...])
4462 Otherwise, it will complain either that the rule does not exist or that some
4465 Flow rule #[...] not found
4469 Caught error type [...] ([...]): [...]
4471 Currently only the ``count`` action is supported. This action reports the
4472 number of packets that hit the flow rule and the total number of bytes. Its
4473 output has the following format::
4476 hits_set: [...] # whether "hits" contains a valid value
4477 bytes_set: [...] # whether "bytes" contains a valid value
4478 hits: [...] # number of packets
4479 bytes: [...] # number of bytes
4481 Querying counters for TCPv6 packets redirected to queue 6::
4483 testpmd> flow create 0 ingress pattern eth / ipv6 / tcp / end
4484 actions queue index 6 / count / end
4485 Flow rule #4 created
4486 testpmd> flow query 0 4 count
4497 ``flow list`` lists existing flow rules sorted by priority and optionally
4498 filtered by group identifiers::
4500 flow list {port_id} [group {group_id}] [...]
4502 This command only fails with the following message if the device does not
4507 Output consists of a header line followed by a short description of each
4508 flow rule, one per line. There is no output at all when no flow rules are
4509 configured on the device::
4511 ID Group Prio Attr Rule
4512 [...] [...] [...] [...] [...]
4514 ``Attr`` column flags:
4516 - ``i`` for ``ingress``.
4517 - ``e`` for ``egress``.
4519 Creating several flow rules and listing them::
4521 testpmd> flow create 0 ingress pattern eth / ipv4 / end
4522 actions queue index 6 / end
4523 Flow rule #0 created
4524 testpmd> flow create 0 ingress pattern eth / ipv6 / end
4525 actions queue index 2 / end
4526 Flow rule #1 created
4527 testpmd> flow create 0 priority 5 ingress pattern eth / ipv4 / udp / end
4528 actions rss queues 6 7 8 end / end
4529 Flow rule #2 created
4530 testpmd> flow list 0
4531 ID Group Prio Attr Rule
4532 0 0 0 i- ETH IPV4 => QUEUE
4533 1 0 0 i- ETH IPV6 => QUEUE
4534 2 0 5 i- ETH IPV4 UDP => RSS
4537 Rules are sorted by priority (i.e. group ID first, then priority level)::
4539 testpmd> flow list 1
4540 ID Group Prio Attr Rule
4541 0 0 0 i- ETH => COUNT
4542 6 0 500 i- ETH IPV6 TCP => DROP COUNT
4543 5 0 1000 i- ETH IPV6 ICMP => QUEUE
4544 1 24 0 i- ETH IPV4 UDP => QUEUE
4545 4 24 10 i- ETH IPV4 TCP => DROP
4546 3 24 20 i- ETH IPV4 => DROP
4547 2 24 42 i- ETH IPV4 UDP => QUEUE
4548 7 63 0 i- ETH IPV6 UDP VXLAN => MARK QUEUE
4551 Output can be limited to specific groups::
4553 testpmd> flow list 1 group 0 group 63
4554 ID Group Prio Attr Rule
4555 0 0 0 i- ETH => COUNT
4556 6 0 500 i- ETH IPV6 TCP => DROP COUNT
4557 5 0 1000 i- ETH IPV6 ICMP => QUEUE
4558 7 63 0 i- ETH IPV6 UDP VXLAN => MARK QUEUE
4561 Toggling isolated mode
4562 ~~~~~~~~~~~~~~~~~~~~~~
4564 ``flow isolate`` can be used to tell the underlying PMD that ingress traffic
4565 must only be injected from the defined flow rules; that no default traffic
4566 is expected outside those rules and the driver is free to assign more
4567 resources to handle them. It is bound to ``rte_flow_isolate()``::
4569 flow isolate {port_id} {boolean}
4571 If successful, enabling or disabling isolated mode shows either::
4573 Ingress traffic on port [...]
4574 is now restricted to the defined flow rules
4578 Ingress traffic on port [...]
4579 is not restricted anymore to the defined flow rules
4581 Otherwise, in case of error::
4583 Caught error type [...] ([...]): [...]
4585 Mainly due to its side effects, PMDs supporting this mode may not have the
4586 ability to toggle it more than once without reinitializing affected ports
4587 first (e.g. by exiting testpmd).
4589 Enabling isolated mode::
4591 testpmd> flow isolate 0 true
4592 Ingress traffic on port 0 is now restricted to the defined flow rules
4595 Disabling isolated mode::
4597 testpmd> flow isolate 0 false
4598 Ingress traffic on port 0 is not restricted anymore to the defined flow rules
4601 Dumping HW internal information
4602 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4604 ``flow dump`` dumps the hardware's internal representation information of
4605 all flows. It is bound to ``rte_flow_dev_dump()``::
4607 flow dump {port_id} {output_file}
4609 If successful, it will show::
4613 Otherwise, it will complain error occurred::
4615 Caught error type [...] ([...]): [...]
4617 Listing and destroying aged flow rules
4618 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4620 ``flow aged`` simply lists aged flow rules be get from api ``rte_flow_get_aged_flows``,
4621 and ``destroy`` parameter can be used to destroy those flow rules in PMD.
4623 flow aged {port_id} [destroy]
4625 Listing current aged flow rules::
4627 testpmd> flow aged 0
4628 Port 0 total aged flows: 0
4629 testpmd> flow create 0 ingress pattern eth / ipv4 src is 2.2.2.14 / end
4630 actions age timeout 5 / queue index 0 / end
4631 Flow rule #0 created
4632 testpmd> flow create 0 ingress pattern eth / ipv4 src is 2.2.2.15 / end
4633 actions age timeout 4 / queue index 0 / end
4634 Flow rule #1 created
4635 testpmd> flow create 0 ingress pattern eth / ipv4 src is 2.2.2.16 / end
4636 actions age timeout 2 / queue index 0 / end
4637 Flow rule #2 created
4638 testpmd> flow create 0 ingress pattern eth / ipv4 src is 2.2.2.17 / end
4639 actions age timeout 3 / queue index 0 / end
4640 Flow rule #3 created
4643 Aged Rules are simply list as command ``flow list {port_id}``, but strip the detail rule
4644 information, all the aged flows are sorted by the longest timeout time. For example, if
4645 those rules be configured in the same time, ID 2 will be the first aged out rule, the next
4646 will be ID 3, ID 1, ID 0::
4648 testpmd> flow aged 0
4649 Port 0 total aged flows: 4
4656 If attach ``destroy`` parameter, the command will destroy all the list aged flow rules.
4658 testpmd> flow aged 0 destroy
4659 Port 0 total aged flows: 4
4666 Flow rule #2 destroyed
4667 Flow rule #3 destroyed
4668 Flow rule #1 destroyed
4669 Flow rule #0 destroyed
4670 4 flows be destroyed
4671 testpmd> flow aged 0
4672 Port 0 total aged flows: 0
4675 Sample QinQ flow rules
4676 ~~~~~~~~~~~~~~~~~~~~~~
4678 Before creating QinQ rule(s) the following commands should be issued to enable QinQ::
4680 testpmd> port stop 0
4681 testpmd> vlan set qinq_strip on 0
4683 The above command sets the inner and outer TPID's to 0x8100.
4685 To change the TPID's the following commands should be used::
4687 testpmd> vlan set outer tpid 0xa100 0
4688 testpmd> vlan set inner tpid 0x9100 0
4689 testpmd> port start 0
4691 Validate and create a QinQ rule on port 0 to steer traffic to a VF queue in a VM.
4695 testpmd> flow validate 0 ingress pattern eth / vlan tci is 123 /
4696 vlan tci is 456 / end actions vf id 1 / queue index 0 / end
4697 Flow rule #0 validated
4699 testpmd> flow create 0 ingress pattern eth / vlan tci is 4 /
4700 vlan tci is 456 / end actions vf id 123 / queue index 0 / end
4701 Flow rule #0 created
4703 testpmd> flow list 0
4704 ID Group Prio Attr Rule
4705 0 0 0 i- ETH VLAN VLAN=>VF QUEUE
4707 Validate and create a QinQ rule on port 0 to steer traffic to a queue on the host.
4711 testpmd> flow validate 0 ingress pattern eth / vlan tci is 321 /
4712 vlan tci is 654 / end actions pf / queue index 0 / end
4713 Flow rule #1 validated
4715 testpmd> flow create 0 ingress pattern eth / vlan tci is 321 /
4716 vlan tci is 654 / end actions pf / queue index 1 / end
4717 Flow rule #1 created
4719 testpmd> flow list 0
4720 ID Group Prio Attr Rule
4721 0 0 0 i- ETH VLAN VLAN=>VF QUEUE
4722 1 0 0 i- ETH VLAN VLAN=>PF QUEUE
4724 Sample VXLAN encapsulation rule
4725 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4727 VXLAN encapsulation outer layer has default value pre-configured in testpmd
4728 source code, those can be changed by using the following commands
4730 IPv4 VXLAN outer header::
4732 testpmd> set vxlan ip-version ipv4 vni 4 udp-src 4 udp-dst 4 ip-src 127.0.0.1
4733 ip-dst 128.0.0.1 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4734 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4737 testpmd> set vxlan-with-vlan ip-version ipv4 vni 4 udp-src 4 udp-dst 4 ip-src
4738 127.0.0.1 ip-dst 128.0.0.1 vlan-tci 34 eth-src 11:11:11:11:11:11
4739 eth-dst 22:22:22:22:22:22
4740 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4743 testpmd> set vxlan-tos-ttl ip-version ipv4 vni 4 udp-src 4 udp-dst 4 ip-tos 0
4744 ip-ttl 255 ip-src 127.0.0.1 ip-dst 128.0.0.1 eth-src 11:11:11:11:11:11
4745 eth-dst 22:22:22:22:22:22
4746 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4749 IPv6 VXLAN outer header::
4751 testpmd> set vxlan ip-version ipv6 vni 4 udp-src 4 udp-dst 4 ip-src ::1
4752 ip-dst ::2222 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4753 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4756 testpmd> set vxlan-with-vlan ip-version ipv6 vni 4 udp-src 4 udp-dst 4
4757 ip-src ::1 ip-dst ::2222 vlan-tci 34 eth-src 11:11:11:11:11:11
4758 eth-dst 22:22:22:22:22:22
4759 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4762 testpmd> set vxlan-tos-ttl ip-version ipv6 vni 4 udp-src 4 udp-dst 4
4763 ip-tos 0 ip-ttl 255 ::1 ip-dst ::2222 eth-src 11:11:11:11:11:11
4764 eth-dst 22:22:22:22:22:22
4765 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4768 Sample NVGRE encapsulation rule
4769 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4771 NVGRE encapsulation outer layer has default value pre-configured in testpmd
4772 source code, those can be changed by using the following commands
4774 IPv4 NVGRE outer header::
4776 testpmd> set nvgre ip-version ipv4 tni 4 ip-src 127.0.0.1 ip-dst 128.0.0.1
4777 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4778 testpmd> flow create 0 ingress pattern end actions nvgre_encap /
4781 testpmd> set nvgre-with-vlan ip-version ipv4 tni 4 ip-src 127.0.0.1
4782 ip-dst 128.0.0.1 vlan-tci 34 eth-src 11:11:11:11:11:11
4783 eth-dst 22:22:22:22:22:22
4784 testpmd> flow create 0 ingress pattern end actions nvgre_encap /
4787 IPv6 NVGRE outer header::
4789 testpmd> set nvgre ip-version ipv6 tni 4 ip-src ::1 ip-dst ::2222
4790 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4791 testpmd> flow create 0 ingress pattern end actions nvgre_encap /
4794 testpmd> set nvgre-with-vlan ip-version ipv6 tni 4 ip-src ::1 ip-dst ::2222
4795 vlan-tci 34 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4796 testpmd> flow create 0 ingress pattern end actions nvgre_encap /
4799 Sample L2 encapsulation rule
4800 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4802 L2 encapsulation has default value pre-configured in testpmd
4803 source code, those can be changed by using the following commands
4807 testpmd> set l2_encap ip-version ipv4
4808 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4809 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / mpls / end actions
4810 mplsoudp_decap / l2_encap / end
4812 L2 with VXLAN header::
4814 testpmd> set l2_encap-with-vlan ip-version ipv4 vlan-tci 34
4815 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4816 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / mpls / end actions
4817 mplsoudp_decap / l2_encap / end
4819 Sample L2 decapsulation rule
4820 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4822 L2 decapsulation has default value pre-configured in testpmd
4823 source code, those can be changed by using the following commands
4827 testpmd> set l2_decap
4828 testpmd> flow create 0 egress pattern eth / end actions l2_decap / mplsoudp_encap /
4831 L2 with VXLAN header::
4833 testpmd> set l2_encap-with-vlan
4834 testpmd> flow create 0 egress pattern eth / end actions l2_encap / mplsoudp_encap /
4837 Sample MPLSoGRE encapsulation rule
4838 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4840 MPLSoGRE encapsulation outer layer has default value pre-configured in testpmd
4841 source code, those can be changed by using the following commands
4843 IPv4 MPLSoGRE outer header::
4845 testpmd> set mplsogre_encap ip-version ipv4 label 4
4846 ip-src 127.0.0.1 ip-dst 128.0.0.1 eth-src 11:11:11:11:11:11
4847 eth-dst 22:22:22:22:22:22
4848 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4849 mplsogre_encap / end
4851 IPv4 MPLSoGRE with VLAN outer header::
4853 testpmd> set mplsogre_encap-with-vlan ip-version ipv4 label 4
4854 ip-src 127.0.0.1 ip-dst 128.0.0.1 vlan-tci 34
4855 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4856 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4857 mplsogre_encap / end
4859 IPv6 MPLSoGRE outer header::
4861 testpmd> set mplsogre_encap ip-version ipv6 mask 4
4862 ip-src ::1 ip-dst ::2222 eth-src 11:11:11:11:11:11
4863 eth-dst 22:22:22:22:22:22
4864 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4865 mplsogre_encap / end
4867 IPv6 MPLSoGRE with VLAN outer header::
4869 testpmd> set mplsogre_encap-with-vlan ip-version ipv6 mask 4
4870 ip-src ::1 ip-dst ::2222 vlan-tci 34
4871 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4872 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4873 mplsogre_encap / end
4875 Sample MPLSoGRE decapsulation rule
4876 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4878 MPLSoGRE decapsulation outer layer has default value pre-configured in testpmd
4879 source code, those can be changed by using the following commands
4881 IPv4 MPLSoGRE outer header::
4883 testpmd> set mplsogre_decap ip-version ipv4
4884 testpmd> flow create 0 ingress pattern eth / ipv4 / gre / mpls / end actions
4885 mplsogre_decap / l2_encap / end
4887 IPv4 MPLSoGRE with VLAN outer header::
4889 testpmd> set mplsogre_decap-with-vlan ip-version ipv4
4890 testpmd> flow create 0 ingress pattern eth / vlan / ipv4 / gre / mpls / end
4891 actions mplsogre_decap / l2_encap / end
4893 IPv6 MPLSoGRE outer header::
4895 testpmd> set mplsogre_decap ip-version ipv6
4896 testpmd> flow create 0 ingress pattern eth / ipv6 / gre / mpls / end
4897 actions mplsogre_decap / l2_encap / end
4899 IPv6 MPLSoGRE with VLAN outer header::
4901 testpmd> set mplsogre_decap-with-vlan ip-version ipv6
4902 testpmd> flow create 0 ingress pattern eth / vlan / ipv6 / gre / mpls / end
4903 actions mplsogre_decap / l2_encap / end
4905 Sample MPLSoUDP encapsulation rule
4906 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4908 MPLSoUDP encapsulation outer layer has default value pre-configured in testpmd
4909 source code, those can be changed by using the following commands
4911 IPv4 MPLSoUDP outer header::
4913 testpmd> set mplsoudp_encap ip-version ipv4 label 4 udp-src 5 udp-dst 10
4914 ip-src 127.0.0.1 ip-dst 128.0.0.1 eth-src 11:11:11:11:11:11
4915 eth-dst 22:22:22:22:22:22
4916 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4917 mplsoudp_encap / end
4919 IPv4 MPLSoUDP with VLAN outer header::
4921 testpmd> set mplsoudp_encap-with-vlan ip-version ipv4 label 4 udp-src 5
4922 udp-dst 10 ip-src 127.0.0.1 ip-dst 128.0.0.1 vlan-tci 34
4923 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4924 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4925 mplsoudp_encap / end
4927 IPv6 MPLSoUDP outer header::
4929 testpmd> set mplsoudp_encap ip-version ipv6 mask 4 udp-src 5 udp-dst 10
4930 ip-src ::1 ip-dst ::2222 eth-src 11:11:11:11:11:11
4931 eth-dst 22:22:22:22:22:22
4932 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4933 mplsoudp_encap / end
4935 IPv6 MPLSoUDP with VLAN outer header::
4937 testpmd> set mplsoudp_encap-with-vlan ip-version ipv6 mask 4 udp-src 5
4938 udp-dst 10 ip-src ::1 ip-dst ::2222 vlan-tci 34
4939 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4940 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4941 mplsoudp_encap / end
4943 Sample MPLSoUDP decapsulation rule
4944 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4946 MPLSoUDP decapsulation outer layer has default value pre-configured in testpmd
4947 source code, those can be changed by using the following commands
4949 IPv4 MPLSoUDP outer header::
4951 testpmd> set mplsoudp_decap ip-version ipv4
4952 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / mpls / end actions
4953 mplsoudp_decap / l2_encap / end
4955 IPv4 MPLSoUDP with VLAN outer header::
4957 testpmd> set mplsoudp_decap-with-vlan ip-version ipv4
4958 testpmd> flow create 0 ingress pattern eth / vlan / ipv4 / udp / mpls / end
4959 actions mplsoudp_decap / l2_encap / end
4961 IPv6 MPLSoUDP outer header::
4963 testpmd> set mplsoudp_decap ip-version ipv6
4964 testpmd> flow create 0 ingress pattern eth / ipv6 / udp / mpls / end
4965 actions mplsoudp_decap / l2_encap / end
4967 IPv6 MPLSoUDP with VLAN outer header::
4969 testpmd> set mplsoudp_decap-with-vlan ip-version ipv6
4970 testpmd> flow create 0 ingress pattern eth / vlan / ipv6 / udp / mpls / end
4971 actions mplsoudp_decap / l2_encap / end
4973 Sample Raw encapsulation rule
4974 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4976 Raw encapsulation configuration can be set by the following commands
4978 Eecapsulating VxLAN::
4980 testpmd> set raw_encap 4 eth src is 10:11:22:33:44:55 / vlan tci is 1
4981 inner_type is 0x0800 / ipv4 / udp dst is 4789 / vxlan vni
4983 testpmd> flow create 0 egress pattern eth / ipv4 / end actions
4984 raw_encap index 4 / end
4986 Sample Raw decapsulation rule
4987 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4989 Raw decapsulation configuration can be set by the following commands
4991 Decapsulating VxLAN::
4993 testpmd> set raw_decap eth / ipv4 / udp / vxlan / end_set
4994 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / vxlan / eth / ipv4 /
4995 end actions raw_decap / queue index 0 / end
5000 ESP rules can be created by the following commands::
5002 testpmd> flow create 0 ingress pattern eth / ipv4 / esp spi is 1 / end actions
5004 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / esp spi is 1 / end
5005 actions queue index 3 / end
5006 testpmd> flow create 0 ingress pattern eth / ipv6 / esp spi is 1 / end actions
5008 testpmd> flow create 0 ingress pattern eth / ipv6 / udp / esp spi is 1 / end
5009 actions queue index 3 / end
5014 AH rules can be created by the following commands::
5016 testpmd> flow create 0 ingress pattern eth / ipv4 / ah spi is 1 / end actions
5018 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / ah spi is 1 / end
5019 actions queue index 3 / end
5020 testpmd> flow create 0 ingress pattern eth / ipv6 / ah spi is 1 / end actions
5022 testpmd> flow create 0 ingress pattern eth / ipv6 / udp / ah spi is 1 / end
5023 actions queue index 3 / end
5028 PFCP rules can be created by the following commands(s_field need to be 1
5031 testpmd> flow create 0 ingress pattern eth / ipv4 / pfcp s_field is 0 / end
5032 actions queue index 3 / end
5033 testpmd> flow create 0 ingress pattern eth / ipv4 / pfcp s_field is 1
5034 seid is 1 / end actions queue index 3 / end
5035 testpmd> flow create 0 ingress pattern eth / ipv6 / pfcp s_field is 0 / end
5036 actions queue index 3 / end
5037 testpmd> flow create 0 ingress pattern eth / ipv6 / pfcp s_field is 1
5038 seid is 1 / end actions queue index 3 / end
5043 The following sections show functions to load/unload eBPF based filters.
5048 Load an eBPF program as a callback for particular RX/TX queue::
5050 testpmd> bpf-load rx|tx (portid) (queueid) (load-flags) (bpf-prog-filename)
5052 The available load-flags are:
5054 * ``J``: use JIT generated native code, otherwise BPF interpreter will be used.
5056 * ``M``: assume input parameter is a pointer to rte_mbuf, otherwise assume it is a pointer to first segment's data.
5062 You'll need clang v3.7 or above to build bpf program you'd like to load
5066 .. code-block:: console
5069 clang -O2 -target bpf -c t1.c
5071 Then to load (and JIT compile) t1.o at RX queue 0, port 1:
5073 .. code-block:: console
5075 testpmd> bpf-load rx 1 0 J ./dpdk.org/examples/bpf/t1.o
5077 To load (not JITed) t1.o at TX queue 0, port 0:
5079 .. code-block:: console
5081 testpmd> bpf-load tx 0 0 - ./dpdk.org/examples/bpf/t1.o
5086 Unload previously loaded eBPF program for particular RX/TX queue::
5088 testpmd> bpf-unload rx|tx (portid) (queueid)
5090 For example to unload BPF filter from TX queue 0, port 0:
5092 .. code-block:: console
5094 testpmd> bpf-unload tx 0 0