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|dcb_tc|cap X
28 info [Mul-choice STRING]: show|clear port info|stats|xstats|fdir|dcb_tc|cap all
29 stats [Mul-choice STRING]: show|clear port info|stats|xstats|fdir|dcb_tc|cap X
30 stats [Mul-choice STRING]: show|clear port info|stats|xstats|fdir|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 ./dpdk-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|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 * ``dcb_tc``: DCB information such as TC mapping.
180 .. code-block:: console
182 testpmd> show port info 0
184 ********************* Infos for port 0 *********************
186 MAC address: XX:XX:XX:XX:XX:XX
188 memory allocation on the socket: 0
190 Link speed: 40000 Mbps
191 Link duplex: full-duplex
192 Promiscuous mode: enabled
193 Allmulticast mode: disabled
194 Maximum number of MAC addresses: 64
195 Maximum number of MAC addresses of hash filtering: 0
197 strip on, filter on, extend off, qinq strip off
198 Redirection table size: 512
199 Supported flow types:
217 show port (module_eeprom|eeprom)
218 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
220 Display the EEPROM information of a port::
222 testpmd> show port (port_id) (module_eeprom|eeprom)
227 Display the rss redirection table entry indicated by masks on port X::
229 testpmd> show port (port_id) rss reta (size) (mask0, mask1...)
231 size is used to indicate the hardware supported reta size
236 Display the RSS hash functions and RSS hash key of a port::
238 testpmd> show port (port_id) rss-hash [key]
243 Clear the port statistics and forward engine statistics for a given port or for all ports::
245 testpmd> clear port (info|stats|xstats|fdir) (port_id|all)
249 testpmd> clear port stats all
254 Display information for a given port's RX/TX queue::
256 testpmd> show (rxq|txq) info (port_id) (queue_id)
258 show desc status(rxq|txq)
259 ~~~~~~~~~~~~~~~~~~~~~~~~~
261 Display information for a given port's RX/TX descriptor status::
263 testpmd> show port (port_id) (rxq|txq) (queue_id) desc (desc_id) status
265 show rxq desc used count
266 ~~~~~~~~~~~~~~~~~~~~~~~~
268 Display the number of receive packet descriptors currently filled by hardware
269 and ready to be processed by the driver on a given RX queue::
271 testpmd> show port (port_id) rxq (queue_id) desc used count
276 Displays the configuration of the application.
277 The configuration comes from the command-line, the runtime or the application defaults::
279 testpmd> show config (rxtx|cores|fwd|rxoffs|rxpkts|txpkts|txtimes)
281 The available information categories are:
283 * ``rxtx``: RX/TX configuration items.
285 * ``cores``: List of forwarding cores.
287 * ``fwd``: Packet forwarding configuration.
289 * ``rxoffs``: Packet offsets for RX split.
291 * ``rxpkts``: Packets to RX split configuration.
293 * ``txpkts``: Packets to TX configuration.
295 * ``txtimes``: Burst time pattern for Tx only mode.
299 .. code-block:: console
301 testpmd> show config rxtx
303 io packet forwarding - CRC stripping disabled - packets/burst=16
304 nb forwarding cores=2 - nb forwarding ports=1
305 RX queues=1 - RX desc=128 - RX free threshold=0
306 RX threshold registers: pthresh=8 hthresh=8 wthresh=4
307 TX queues=1 - TX desc=512 - TX free threshold=0
308 TX threshold registers: pthresh=36 hthresh=0 wthresh=0
309 TX RS bit threshold=0 - TXQ flags=0x0
314 Set the packet forwarding mode::
316 testpmd> set fwd (io|mac|macswap|flowgen| \
317 rxonly|txonly|csum|icmpecho|noisy|5tswap) (""|retry)
319 ``retry`` can be specified for forwarding engines except ``rx_only``.
321 The available information categories are:
323 * ``io``: Forwards packets "as-is" in I/O mode.
324 This is the fastest possible forwarding operation as it does not access packets data.
325 This is the default mode.
327 * ``mac``: Changes the source and the destination Ethernet addresses of packets before forwarding them.
328 Default application behavior is to set source Ethernet address to that of the transmitting interface, and destination
329 address to a dummy value (set during init). The user may specify a target destination Ethernet address via the 'eth-peer' or
330 'eth-peers-configfile' command-line options. It is not currently possible to specify a specific source Ethernet address.
332 * ``macswap``: MAC swap forwarding mode.
333 Swaps the source and the destination Ethernet addresses of packets before forwarding them.
335 * ``flowgen``: Multi-flow generation mode.
336 Originates a number of flows (with varying destination IP addresses), and terminate receive traffic.
338 * ``rxonly``: Receives packets but doesn't transmit them.
340 * ``txonly``: Generates and transmits packets without receiving any.
342 * ``csum``: Changes the checksum field with hardware or software methods depending on the offload flags on the packet.
344 * ``icmpecho``: Receives a burst of packets, lookup for ICMP echo requests and, if any, send back ICMP echo replies.
346 * ``ieee1588``: Demonstrate L2 IEEE1588 V2 PTP timestamping for RX and TX.
348 * ``noisy``: Noisy neighbor simulation.
349 Simulate more realistic behavior of a guest machine engaged in receiving
350 and sending packets performing Virtual Network Function (VNF).
352 * ``5tswap``: Swap the source and destination of L2,L3,L4 if they exist.
354 L2 swaps the source address and destination address of Ethernet, as same as ``macswap``.
356 L3 swaps the source address and destination address of IP (v4 and v6).
358 L4 swaps the source port and destination port of transport layer (TCP and UDP).
362 testpmd> set fwd rxonly
364 Set rxonly packet forwarding mode
370 When running, forwarding engines maintain statistics from the time they have been started.
371 Example for the io forwarding engine, with some packet drops on the tx side::
373 testpmd> show fwd stats all
375 ------- Forward Stats for RX Port= 0/Queue= 0 -> TX Port= 1/Queue= 0 -------
376 RX-packets: 274293770 TX-packets: 274293642 TX-dropped: 128
378 ------- Forward Stats for RX Port= 1/Queue= 0 -> TX Port= 0/Queue= 0 -------
379 RX-packets: 274301850 TX-packets: 274301850 TX-dropped: 0
381 ---------------------- Forward statistics for port 0 ----------------------
382 RX-packets: 274293802 RX-dropped: 0 RX-total: 274293802
383 TX-packets: 274301862 TX-dropped: 0 TX-total: 274301862
384 ----------------------------------------------------------------------------
386 ---------------------- Forward statistics for port 1 ----------------------
387 RX-packets: 274301894 RX-dropped: 0 RX-total: 274301894
388 TX-packets: 274293706 TX-dropped: 128 TX-total: 274293834
389 ----------------------------------------------------------------------------
391 +++++++++++++++ Accumulated forward statistics for all ports+++++++++++++++
392 RX-packets: 548595696 RX-dropped: 0 RX-total: 548595696
393 TX-packets: 548595568 TX-dropped: 128 TX-total: 548595696
394 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
400 Clear the forwarding engines statistics::
402 testpmd> clear fwd stats all
407 Display an RX descriptor for a port RX queue::
409 testpmd> read rxd (port_id) (queue_id) (rxd_id)
413 testpmd> read rxd 0 0 4
414 0x0000000B - 0x001D0180 / 0x0000000B - 0x001D0180
419 Display a TX descriptor for a port TX queue::
421 testpmd> read txd (port_id) (queue_id) (txd_id)
425 testpmd> read txd 0 0 4
426 0x00000001 - 0x24C3C440 / 0x000F0000 - 0x2330003C
431 Get loaded dynamic device personalization (DDP) package info list::
433 testpmd> ddp get list (port_id)
438 Display information about dynamic device personalization (DDP) profile::
440 testpmd> ddp get info (profile_path)
445 Display VF statistics::
447 testpmd> show vf stats (port_id) (vf_id)
452 Reset VF statistics::
454 testpmd> clear vf stats (port_id) (vf_id)
456 show port pctype mapping
457 ~~~~~~~~~~~~~~~~~~~~~~~~
459 List all items from the pctype mapping table::
461 testpmd> show port (port_id) pctype mapping
463 show rx offloading capabilities
464 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
466 List all per queue and per port Rx offloading capabilities of a port::
468 testpmd> show port (port_id) rx_offload capabilities
470 show rx offloading configuration
471 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
473 List port level and all queue level Rx offloading configuration::
475 testpmd> show port (port_id) rx_offload configuration
477 show tx offloading capabilities
478 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
480 List all per queue and per port Tx offloading capabilities of a port::
482 testpmd> show port (port_id) tx_offload capabilities
484 show tx offloading configuration
485 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
487 List port level and all queue level Tx offloading configuration::
489 testpmd> show port (port_id) tx_offload configuration
491 show tx metadata setting
492 ~~~~~~~~~~~~~~~~~~~~~~~~
494 Show Tx metadata value set for a specific port::
496 testpmd> show port (port_id) tx_metadata
498 show port supported ptypes
499 ~~~~~~~~~~~~~~~~~~~~~~~~~~
501 Show ptypes supported for a specific port::
503 testpmd> show port (port_id) ptypes
505 set port supported ptypes
506 ~~~~~~~~~~~~~~~~~~~~~~~~~
508 set packet types classification for a specific port::
510 testpmd> set port (port_id) ptypes_mask (mask)
512 show port mac addresses info
513 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
515 Show mac addresses added for a specific port::
517 testpmd> show port (port_id) macs
520 show port multicast mac addresses info
521 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
523 Show multicast mac addresses added for a specific port::
525 testpmd> show port (port_id) mcast_macs
530 Show general information about devices probed::
532 testpmd> show device info (<identifier>|all)
536 .. code-block:: console
538 testpmd> show device info net_pcap0
540 ********************* Infos for device net_pcap0 *********************
542 Driver name: net_pcap
543 Devargs: iface=enP2p6s0,phy_mac=1
544 Connect to socket: -1
547 MAC address: 1E:37:93:28:04:B8
548 Device name: net_pcap0
553 Dumps all physical memory segment layouts::
555 testpmd> dump_physmem
560 Dumps the layout of all memory zones::
562 testpmd> dump_memzone
567 Dumps the memory usage of all sockets::
569 testpmd> dump_socket_mem
574 Dumps the size of all memory structures::
576 testpmd> dump_struct_sizes
581 Dumps the status of all or specific element in DPDK rings::
583 testpmd> dump_ring [ring_name]
588 Dumps the statistics of all or specific memory pool::
590 testpmd> dump_mempool [mempool_name]
595 Dumps the user device list::
597 testpmd> dump_devargs
602 Dumps the log level for all the dpdk modules::
604 testpmd> dump_log_types
606 show (raw_encap|raw_decap)
607 ~~~~~~~~~~~~~~~~~~~~~~~~~~
609 Display content of raw_encap/raw_decap buffers in hex::
611 testpmd> show <raw_encap|raw_decap> <index>
612 testpmd> show <raw_encap|raw_decap> all
616 testpmd> show raw_encap 6
618 index: 6 at [0x1c565b0], len=50
619 00000000: 00 00 00 00 00 00 16 26 36 46 56 66 08 00 45 00 | .......&6FVf..E.
620 00000010: 00 00 00 00 00 00 00 11 00 00 C0 A8 01 06 C0 A8 | ................
621 00000020: 03 06 00 00 00 FA 00 00 00 00 08 00 00 00 00 00 | ................
624 show fec capabilities
625 ~~~~~~~~~~~~~~~~~~~~~
627 Show fec capabilities of a port::
629 testpmd> show port (port_id) fec capabilities
634 Show fec mode of a port::
636 testpmd> show port (port_id) fec_mode
639 Configuration Functions
640 -----------------------
642 The testpmd application can be configured from the runtime as well as from the command-line.
644 This section details the available configuration functions that are available.
648 Configuration changes only become active when forwarding is started/restarted.
653 Reset forwarding to the default configuration::
660 Set the debug verbosity level::
662 testpmd> set verbose (level)
664 Available levels are as following:
666 * ``0`` silent except for error.
667 * ``1`` fully verbose except for Tx packets.
668 * ``2`` fully verbose except for Rx packets.
669 * ``> 2`` fully verbose.
674 Set the log level for a log type::
676 testpmd> set log global|(type) (level)
680 * ``type`` is the log name.
682 * ``level`` is the log level.
684 For example, to change the global log level::
686 testpmd> set log global (level)
688 Regexes can also be used for type. To change log level of user1, user2 and user3::
690 testpmd> set log user[1-3] (level)
695 Set the number of ports used by the application:
699 This is equivalent to the ``--nb-ports`` command-line option.
704 Set the number of cores used by the application::
706 testpmd> set nbcore (num)
708 This is equivalent to the ``--nb-cores`` command-line option.
712 The number of cores used must not be greater than number of ports used multiplied by the number of queues per port.
717 Set the forwarding cores hexadecimal mask::
719 testpmd> set coremask (mask)
721 This is equivalent to the ``--coremask`` command-line option.
725 The main lcore is reserved for command line parsing only and cannot be masked on for packet forwarding.
730 Set the forwarding ports hexadecimal mask::
732 testpmd> set portmask (mask)
734 This is equivalent to the ``--portmask`` command-line option.
736 set record-core-cycles
737 ~~~~~~~~~~~~~~~~~~~~~~
739 Set the recording of CPU cycles::
741 testpmd> set record-core-cycles (on|off)
745 * ``on`` enables measurement of CPU cycles per packet.
747 * ``off`` disables measurement of CPU cycles per packet.
749 This is equivalent to the ``--record-core-cycles command-line`` option.
751 set record-burst-stats
752 ~~~~~~~~~~~~~~~~~~~~~~
754 Set the displaying of RX and TX bursts::
756 testpmd> set record-burst-stats (on|off)
760 * ``on`` enables display of RX and TX bursts.
762 * ``off`` disables display of RX and TX bursts.
764 This is equivalent to the ``--record-burst-stats command-line`` option.
769 Set number of packets per burst::
771 testpmd> set burst (num)
773 This is equivalent to the ``--burst command-line`` option.
775 When retry is enabled, the transmit delay time and number of retries can also be set::
777 testpmd> set burst tx delay (microseconds) retry (num)
782 Set the offsets of segments relating to the data buffer beginning on receiving
783 if split feature is engaged. Affects only the queues configured with split
784 offloads (currently BUFFER_SPLIT is supported only).
786 testpmd> set rxoffs (x[,y]*)
788 Where x[,y]* represents a CSV list of values, without white space. If the list
789 of offsets is shorter than the list of segments the zero offsets will be used
790 for the remaining segments.
795 Set the length of segments to scatter packets on receiving if split
796 feature is engaged. Affects only the queues configured with split offloads
797 (currently BUFFER_SPLIT is supported only). Optionally the multiple memory
798 pools can be specified with --mbuf-size command line parameter and the mbufs
799 to receive will be allocated sequentially from these extra memory pools (the
800 mbuf for the first segment is allocated from the first pool, the second one
801 from the second pool, and so on, if segment number is greater then pool's the
802 mbuf for remaining segments will be allocated from the last valid pool).
804 testpmd> set rxpkts (x[,y]*)
806 Where x[,y]* represents a CSV list of values, without white space. Zero value
807 means to use the corresponding memory pool data buffer size.
812 Set the length of each segment of the TX-ONLY packets or length of packet for FLOWGEN mode::
814 testpmd> set txpkts (x[,y]*)
816 Where x[,y]* represents a CSV list of values, without white space.
821 Configure the timing burst pattern for Tx only mode. This command enables
822 the packet send scheduling on dynamic timestamp mbuf field and configures
823 timing pattern in Tx only mode. In this mode, if scheduling is enabled
824 application provides timestamps in the packets being sent. It is possible
825 to configure delay (in unspecified device clock units) between bursts
826 and between the packets within the burst::
828 testpmd> set txtimes (inter),(intra)
832 * ``inter`` is the delay between the bursts in the device clock units.
833 If ``intra`` is zero, this is the time between the beginnings of the
834 first packets in the neighbour bursts, if ``intra`` is not zero,
835 ``inter`` specifies the time between the beginning of the first packet
836 of the current burst and the beginning of the last packet of the
837 previous burst. If ``inter`` parameter is zero the send scheduling
838 on timestamps is disabled (default).
840 * ``intra`` is the delay between the packets within the burst specified
841 in the device clock units. The number of packets in the burst is defined
842 by regular burst setting. If ``intra`` parameter is zero no timestamps
843 provided in the packets excepting the first one in the burst.
845 As the result the bursts of packet will be transmitted with specific
846 delays between the packets within the burst and specific delay between
847 the bursts. The rte_eth_read_clock() must be supported by the device(s)
848 and is supposed to be engaged to get the current device clock value
849 and provide the reference for the timestamps. If there is no supported
850 rte_eth_read_clock() there will be no send scheduling provided on the port.
855 Set the split policy for the TX packets, applicable for TX-ONLY and CSUM forwarding modes::
857 testpmd> set txsplit (off|on|rand)
861 * ``off`` disable packet copy & split for CSUM mode.
863 * ``on`` split outgoing packet into multiple segments. Size of each segment
864 and number of segments per packet is determined by ``set txpkts`` command
867 * ``rand`` same as 'on', but number of segments per each packet is a random value between 1 and total number of segments.
872 Set the list of forwarding cores::
874 testpmd> set corelist (x[,y]*)
876 For example, to change the forwarding cores:
878 .. code-block:: console
880 testpmd> set corelist 3,1
881 testpmd> show config fwd
883 io packet forwarding - ports=2 - cores=2 - streams=2 - NUMA support disabled
884 Logical Core 3 (socket 0) forwards packets on 1 streams:
885 RX P=0/Q=0 (socket 0) -> TX P=1/Q=0 (socket 0) peer=02:00:00:00:00:01
886 Logical Core 1 (socket 0) forwards packets on 1 streams:
887 RX P=1/Q=0 (socket 0) -> TX P=0/Q=0 (socket 0) peer=02:00:00:00:00:00
891 The cores are used in the same order as specified on the command line.
896 Set the list of forwarding ports::
898 testpmd> set portlist (x[,y]*)
900 For example, to change the port forwarding:
902 .. code-block:: console
904 testpmd> set portlist 0,2,1,3
905 testpmd> show config fwd
907 io packet forwarding - ports=4 - cores=1 - streams=4
908 Logical Core 3 (socket 0) forwards packets on 4 streams:
909 RX P=0/Q=0 (socket 0) -> TX P=2/Q=0 (socket 0) peer=02:00:00:00:00:01
910 RX P=2/Q=0 (socket 0) -> TX P=0/Q=0 (socket 0) peer=02:00:00:00:00:00
911 RX P=1/Q=0 (socket 0) -> TX P=3/Q=0 (socket 0) peer=02:00:00:00:00:03
912 RX P=3/Q=0 (socket 0) -> TX P=1/Q=0 (socket 0) peer=02:00:00:00:00:02
917 Select how to retrieve new ports created after "port attach" command::
919 testpmd> set port setup on (iterator|event)
921 For each new port, a setup is done.
922 It will find the probed ports via RTE_ETH_FOREACH_MATCHING_DEV loop
923 in iterator mode, or via RTE_ETH_EVENT_NEW in event mode.
928 Enable/disable tx loopback::
930 testpmd> set tx loopback (port_id) (on|off)
935 set drop enable bit for all queues::
937 testpmd> set all queues drop (port_id) (on|off)
939 set split drop enable (for VF)
940 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
942 set split drop enable bit for VF from PF::
944 testpmd> set vf split drop (port_id) (vf_id) (on|off)
946 set mac antispoof (for VF)
947 ~~~~~~~~~~~~~~~~~~~~~~~~~~
949 Set mac antispoof for a VF from the PF::
951 testpmd> set vf mac antispoof (port_id) (vf_id) (on|off)
956 Enable/disable MACsec offload::
958 testpmd> set macsec offload (port_id) on encrypt (on|off) replay-protect (on|off)
959 testpmd> set macsec offload (port_id) off
964 Configure MACsec secure connection (SC)::
966 testpmd> set macsec sc (tx|rx) (port_id) (mac) (pi)
970 The pi argument is ignored for tx.
971 Check the NIC Datasheet for hardware limits.
976 Configure MACsec secure association (SA)::
978 testpmd> set macsec sa (tx|rx) (port_id) (idx) (an) (pn) (key)
982 The IDX value must be 0 or 1.
983 Check the NIC Datasheet for hardware limits.
985 set broadcast mode (for VF)
986 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
988 Set broadcast mode for a VF from the PF::
990 testpmd> set vf broadcast (port_id) (vf_id) (on|off)
995 Set the VLAN strip for a queue on a port::
997 testpmd> vlan set stripq (on|off) (port_id,queue_id)
999 vlan set stripq (for VF)
1000 ~~~~~~~~~~~~~~~~~~~~~~~~
1002 Set VLAN strip for all queues in a pool for a VF from the PF::
1004 testpmd> set vf vlan stripq (port_id) (vf_id) (on|off)
1006 vlan set insert (for VF)
1007 ~~~~~~~~~~~~~~~~~~~~~~~~
1009 Set VLAN insert for a VF from the PF::
1011 testpmd> set vf vlan insert (port_id) (vf_id) (vlan_id)
1013 vlan set tag (for VF)
1014 ~~~~~~~~~~~~~~~~~~~~~
1016 Set VLAN tag for a VF from the PF::
1018 testpmd> set vf vlan tag (port_id) (vf_id) (on|off)
1020 vlan set antispoof (for VF)
1021 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
1023 Set VLAN antispoof for a VF from the PF::
1025 testpmd> set vf vlan antispoof (port_id) (vf_id) (on|off)
1027 vlan set (strip|filter|qinq_strip|extend)
1028 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1029 Set the VLAN strip/filter/QinQ strip/extend on for a port::
1031 testpmd> vlan set (strip|filter|qinq_strip|extend) (on|off) (port_id)
1036 Set the inner or outer VLAN TPID for packet filtering on a port::
1038 testpmd> vlan set (inner|outer) tpid (value) (port_id)
1042 TPID value must be a 16-bit number (value <= 65536).
1047 Add a VLAN ID, or all identifiers, to the set of VLAN identifiers filtered by port ID::
1049 testpmd> rx_vlan add (vlan_id|all) (port_id)
1053 VLAN filter must be set on that port. VLAN ID < 4096.
1054 Depending on the NIC used, number of vlan_ids may be limited to the maximum entries
1055 in VFTA table. This is important if enabling all vlan_ids.
1060 Remove a VLAN ID, or all identifiers, from the set of VLAN identifiers filtered by port ID::
1062 testpmd> rx_vlan rm (vlan_id|all) (port_id)
1064 rx_vlan add (for VF)
1065 ~~~~~~~~~~~~~~~~~~~~
1067 Add a VLAN ID, to the set of VLAN identifiers filtered for VF(s) for port ID::
1069 testpmd> rx_vlan add (vlan_id) port (port_id) vf (vf_mask)
1074 Remove a VLAN ID, from the set of VLAN identifiers filtered for VF(s) for port ID::
1076 testpmd> rx_vlan rm (vlan_id) port (port_id) vf (vf_mask)
1081 Add an UDP port for VXLAN packet filter on a port::
1083 testpmd> rx_vxlan_port add (udp_port) (port_id)
1085 rx_vxlan_port remove
1086 ~~~~~~~~~~~~~~~~~~~~
1088 Remove an UDP port for VXLAN packet filter on a port::
1090 testpmd> rx_vxlan_port rm (udp_port) (port_id)
1095 Set hardware insertion of VLAN IDs in packets sent on a port::
1097 testpmd> tx_vlan set (port_id) vlan_id[, vlan_id_outer]
1099 For example, set a single VLAN ID (5) insertion on port 0::
1103 Or, set double VLAN ID (inner: 2, outer: 3) insertion on port 1::
1111 Set port based hardware insertion of VLAN ID in packets sent on a port::
1113 testpmd> tx_vlan set pvid (port_id) (vlan_id) (on|off)
1118 Disable hardware insertion of a VLAN header in packets sent on a port::
1120 testpmd> tx_vlan reset (port_id)
1125 Select hardware or software calculation of the checksum when
1126 transmitting a packet using the ``csum`` forwarding engine::
1128 testpmd> csum set (ip|udp|tcp|sctp|outer-ip|outer-udp) (hw|sw) (port_id)
1132 * ``ip|udp|tcp|sctp`` always relate to the inner layer.
1134 * ``outer-ip`` relates to the outer IP layer (only for IPv4) in the case where the packet is recognized
1135 as a tunnel packet by the forwarding engine (geneve, gre, gtp, ipip, vxlan and vxlan-gpe are
1136 supported). See also the ``csum parse-tunnel`` command.
1138 * ``outer-udp`` relates to the outer UDP layer in the case where the packet is recognized
1139 as a tunnel packet by the forwarding engine (geneve, gtp, vxlan and vxlan-gpe are
1140 supported). See also the ``csum parse-tunnel`` command.
1144 Check the NIC Datasheet for hardware limits.
1149 Set RSS queue region span on a port::
1151 testpmd> set port (port_id) queue-region region_id (value) \
1152 queue_start_index (value) queue_num (value)
1154 Set flowtype mapping on a RSS queue region on a port::
1156 testpmd> set port (port_id) queue-region region_id (value) flowtype (value)
1160 * For the flowtype(pctype) of packet,the specific index for each type has
1161 been defined in file i40e_type.h as enum i40e_filter_pctype.
1163 Set user priority mapping on a RSS queue region on a port::
1165 testpmd> set port (port_id) queue-region UP (value) region_id (value)
1167 Flush all queue region related configuration on a port::
1169 testpmd> set port (port_id) queue-region flush (on|off)
1173 * ``on``: is just an enable function which server for other configuration,
1174 it is for all configuration about queue region from up layer,
1175 at first will only keep in DPDK software stored in driver,
1176 only after "flush on", it commit all configuration to HW.
1178 * ``"off``: is just clean all configuration about queue region just now,
1179 and restore all to DPDK i40e driver default config when start up.
1181 Show all queue region related configuration info on a port::
1183 testpmd> show port (port_id) queue-region
1187 Queue region only support on PF by now, so these command is
1188 only for configuration of queue region on PF port.
1193 Define how tunneled packets should be handled by the csum forward
1196 testpmd> csum parse-tunnel (on|off) (tx_port_id)
1198 If enabled, the csum forward engine will try to recognize supported
1199 tunnel headers (geneve, gtp, gre, ipip, vxlan, vxlan-gpe).
1201 If disabled, treat tunnel packets as non-tunneled packets (a inner
1202 header is handled as a packet payload).
1206 The port argument is the TX port like in the ``csum set`` command.
1210 Consider a packet in packet like the following::
1212 eth_out/ipv4_out/udp_out/vxlan/eth_in/ipv4_in/tcp_in
1214 * If parse-tunnel is enabled, the ``ip|udp|tcp|sctp`` parameters of ``csum set``
1215 command relate to the inner headers (here ``ipv4_in`` and ``tcp_in``), and the
1216 ``outer-ip|outer-udp`` parameter relates to the outer headers (here ``ipv4_out`` and ``udp_out``).
1218 * If parse-tunnel is disabled, the ``ip|udp|tcp|sctp`` parameters of ``csum set``
1219 command relate to the outer headers, here ``ipv4_out`` and ``udp_out``.
1224 Display tx checksum offload configuration::
1226 testpmd> csum show (port_id)
1231 Enable TCP Segmentation Offload (TSO) in the ``csum`` forwarding engine::
1233 testpmd> tso set (segsize) (port_id)
1237 Check the NIC datasheet for hardware limits.
1242 Display the status of TCP Segmentation Offload::
1244 testpmd> tso show (port_id)
1249 Set tso segment size of tunneled packets for a port in csum engine::
1251 testpmd> tunnel_tso set (tso_segsz) (port_id)
1256 Display the status of tunneled TCP Segmentation Offload for a port::
1258 testpmd> tunnel_tso show (port_id)
1263 Enable or disable GRO in ``csum`` forwarding engine::
1265 testpmd> set port <port_id> gro on|off
1267 If enabled, the csum forwarding engine will perform GRO on the TCP/IPv4
1268 packets received from the given port.
1270 If disabled, packets received from the given port won't be performed
1271 GRO. By default, GRO is disabled for all ports.
1275 When enable GRO for a port, TCP/IPv4 packets received from the port
1276 will be performed GRO. After GRO, all merged packets have bad
1277 checksums, since the GRO library doesn't re-calculate checksums for
1278 the merged packets. Therefore, if users want the merged packets to
1279 have correct checksums, please select HW IP checksum calculation and
1280 HW TCP checksum calculation for the port which the merged packets are
1286 Display GRO configuration for a given port::
1288 testpmd> show port <port_id> gro
1293 Set the cycle to flush the GROed packets from reassembly tables::
1295 testpmd> set gro flush <cycles>
1297 When enable GRO, the csum forwarding engine performs GRO on received
1298 packets, and the GROed packets are stored in reassembly tables. Users
1299 can use this command to determine when the GROed packets are flushed
1300 from the reassembly tables.
1302 The ``cycles`` is measured in GRO operation times. The csum forwarding
1303 engine flushes the GROed packets from the tables every ``cycles`` GRO
1306 By default, the value of ``cycles`` is 1, which means flush GROed packets
1307 from the reassembly tables as soon as one GRO operation finishes. The value
1308 of ``cycles`` should be in the range of 1 to ``GRO_MAX_FLUSH_CYCLES``.
1310 Please note that the large value of ``cycles`` may cause the poor TCP/IP
1311 stack performance. Because the GROed packets are delayed to arrive the
1312 stack, thus causing more duplicated ACKs and TCP retransmissions.
1317 Toggle per-port GSO support in ``csum`` forwarding engine::
1319 testpmd> set port <port_id> gso on|off
1321 If enabled, the csum forwarding engine will perform GSO on supported IPv4
1322 packets, transmitted on the given port.
1324 If disabled, packets transmitted on the given port will not undergo GSO.
1325 By default, GSO is disabled for all ports.
1329 When GSO is enabled on a port, supported IPv4 packets transmitted on that
1330 port undergo GSO. Afterwards, the segmented packets are represented by
1331 multi-segment mbufs; however, the csum forwarding engine doesn't calculation
1332 of checksums for GSO'd segments in SW. As a result, if users want correct
1333 checksums in GSO segments, they should enable HW checksum calculation for
1336 For example, HW checksum calculation for VxLAN GSO'd packets may be enabled
1337 by setting the following options in the csum forwarding engine:
1339 testpmd> csum set outer_ip hw <port_id>
1341 testpmd> csum set ip hw <port_id>
1343 testpmd> csum set tcp hw <port_id>
1345 UDP GSO is the same as IP fragmentation, which treats the UDP header
1346 as the payload and does not modify it during segmentation. That is,
1347 after UDP GSO, only the first output fragment has the original UDP
1348 header. Therefore, users need to enable HW IP checksum calculation
1349 and SW UDP checksum calculation for GSO-enabled ports, if they want
1350 correct checksums for UDP/IPv4 packets.
1355 Set the maximum GSO segment size (measured in bytes), which includes the
1356 packet header and the packet payload for GSO-enabled ports (global)::
1358 testpmd> set gso segsz <length>
1363 Display the status of Generic Segmentation Offload for a given port::
1365 testpmd> show port <port_id> gso
1370 Add an alternative MAC address to a port::
1372 testpmd> mac_addr add (port_id) (XX:XX:XX:XX:XX:XX)
1377 Remove a MAC address from a port::
1379 testpmd> mac_addr remove (port_id) (XX:XX:XX:XX:XX:XX)
1384 To add the multicast MAC address to/from the set of multicast addresses
1387 testpmd> mcast_addr add (port_id) (mcast_addr)
1392 To remove the multicast MAC address to/from the set of multicast addresses
1395 testpmd> mcast_addr remove (port_id) (mcast_addr)
1397 mac_addr add (for VF)
1398 ~~~~~~~~~~~~~~~~~~~~~
1400 Add an alternative MAC address for a VF to a port::
1402 testpmd> mac_add add port (port_id) vf (vf_id) (XX:XX:XX:XX:XX:XX)
1407 Set the default MAC address for a port::
1409 testpmd> mac_addr set (port_id) (XX:XX:XX:XX:XX:XX)
1411 mac_addr set (for VF)
1412 ~~~~~~~~~~~~~~~~~~~~~
1414 Set the MAC address for a VF from the PF::
1416 testpmd> set vf mac addr (port_id) (vf_id) (XX:XX:XX:XX:XX:XX)
1421 Set the forwarding peer address for certain port::
1423 testpmd> set eth-peer (port_id) (peer_addr)
1425 This is equivalent to the ``--eth-peer`` command-line option.
1430 Set the unicast hash filter(s) on/off for a port::
1432 testpmd> set port (port_id) uta (XX:XX:XX:XX:XX:XX|all) (on|off)
1437 Set the promiscuous mode on for a port or for all ports.
1438 In promiscuous mode packets are not dropped if they aren't for the specified MAC address::
1440 testpmd> set promisc (port_id|all) (on|off)
1445 Set the allmulti mode for a port or for all ports::
1447 testpmd> set allmulti (port_id|all) (on|off)
1449 Same as the ifconfig (8) option. Controls how multicast packets are handled.
1451 set promisc (for VF)
1452 ~~~~~~~~~~~~~~~~~~~~
1454 Set the unicast promiscuous mode for a VF from PF.
1455 It's supported by Intel i40e NICs now.
1456 In promiscuous mode packets are not dropped if they aren't for the specified MAC address::
1458 testpmd> set vf promisc (port_id) (vf_id) (on|off)
1460 set allmulticast (for VF)
1461 ~~~~~~~~~~~~~~~~~~~~~~~~~
1463 Set the multicast promiscuous mode for a VF from PF.
1464 It's supported by Intel i40e NICs now.
1465 In promiscuous mode packets are not dropped if they aren't for the specified MAC address::
1467 testpmd> set vf allmulti (port_id) (vf_id) (on|off)
1469 set tx max bandwidth (for VF)
1470 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1472 Set TX max absolute bandwidth (Mbps) for a VF from PF::
1474 testpmd> set vf tx max-bandwidth (port_id) (vf_id) (max_bandwidth)
1476 set tc tx min bandwidth (for VF)
1477 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1479 Set all TCs' TX min relative bandwidth (%) for a VF from PF::
1481 testpmd> set vf tc tx min-bandwidth (port_id) (vf_id) (bw1, bw2, ...)
1483 set tc tx max bandwidth (for VF)
1484 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1486 Set a TC's TX max absolute bandwidth (Mbps) for a VF from PF::
1488 testpmd> set vf tc tx max-bandwidth (port_id) (vf_id) (tc_no) (max_bandwidth)
1490 set tc strict link priority mode
1491 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1493 Set some TCs' strict link priority mode on a physical port::
1495 testpmd> set tx strict-link-priority (port_id) (tc_bitmap)
1497 set tc tx min bandwidth
1498 ~~~~~~~~~~~~~~~~~~~~~~~
1500 Set all TCs' TX min relative bandwidth (%) globally for all PF and VFs::
1502 testpmd> set tc tx min-bandwidth (port_id) (bw1, bw2, ...)
1507 Set the link flow control parameter on a port::
1509 testpmd> set flow_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \
1510 (pause_time) (send_xon) mac_ctrl_frame_fwd (on|off) \
1511 autoneg (on|off) (port_id)
1515 * ``high_water`` (integer): High threshold value to trigger XOFF.
1517 * ``low_water`` (integer): Low threshold value to trigger XON.
1519 * ``pause_time`` (integer): Pause quota in the Pause frame.
1521 * ``send_xon`` (0/1): Send XON frame.
1523 * ``mac_ctrl_frame_fwd``: Enable receiving MAC control frames.
1525 * ``autoneg``: Change the auto-negotiation parameter.
1530 show the link flow control parameter on a port::
1532 testpmd> show port <port_id> flow_ctrl
1537 Set the priority flow control parameter on a port::
1539 testpmd> set pfc_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \
1540 (pause_time) (priority) (port_id)
1544 * ``high_water`` (integer): High threshold value.
1546 * ``low_water`` (integer): Low threshold value.
1548 * ``pause_time`` (integer): Pause quota in the Pause frame.
1550 * ``priority`` (0-7): VLAN User Priority.
1555 Set statistics mapping (qmapping 0..15) for RX/TX queue on port::
1557 testpmd> set stat_qmap (tx|rx) (port_id) (queue_id) (qmapping)
1559 For example, to set rx queue 2 on port 0 to mapping 5::
1561 testpmd>set stat_qmap rx 0 2 5
1563 set xstats-hide-zero
1564 ~~~~~~~~~~~~~~~~~~~~
1566 Set the option to hide zero values for xstats display::
1568 testpmd> set xstats-hide-zero on|off
1572 By default, the zero values are displayed for xstats.
1574 set port - rx/tx (for VF)
1575 ~~~~~~~~~~~~~~~~~~~~~~~~~
1577 Set VF receive/transmit from a port::
1579 testpmd> set port (port_id) vf (vf_id) (rx|tx) (on|off)
1581 set port - rx mode(for VF)
1582 ~~~~~~~~~~~~~~~~~~~~~~~~~~
1584 Set the VF receive mode of a port::
1586 testpmd> set port (port_id) vf (vf_id) \
1587 rxmode (AUPE|ROPE|BAM|MPE) (on|off)
1589 The available receive modes are:
1591 * ``AUPE``: Accepts untagged VLAN.
1593 * ``ROPE``: Accepts unicast hash.
1595 * ``BAM``: Accepts broadcast packets.
1597 * ``MPE``: Accepts all multicast packets.
1599 set port - tx_rate (for Queue)
1600 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1602 Set TX rate limitation for a queue on a port::
1604 testpmd> set port (port_id) queue (queue_id) rate (rate_value)
1606 set port - tx_rate (for VF)
1607 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
1609 Set TX rate limitation for queues in VF on a port::
1611 testpmd> set port (port_id) vf (vf_id) rate (rate_value) queue_mask (queue_mask)
1616 Set the flush on RX streams before forwarding.
1617 The default is flush ``on``.
1618 Mainly used with PCAP drivers to turn off the default behavior of flushing the first 512 packets on RX streams::
1620 testpmd> set flush_rx off
1625 Set the bypass mode for the lowest port on bypass enabled NIC::
1627 testpmd> set bypass mode (normal|bypass|isolate) (port_id)
1632 Set the event required to initiate specified bypass mode for the lowest port on a bypass enabled::
1634 testpmd> set bypass event (timeout|os_on|os_off|power_on|power_off) \
1635 mode (normal|bypass|isolate) (port_id)
1639 * ``timeout``: Enable bypass after watchdog timeout.
1641 * ``os_on``: Enable bypass when OS/board is powered on.
1643 * ``os_off``: Enable bypass when OS/board is powered off.
1645 * ``power_on``: Enable bypass when power supply is turned on.
1647 * ``power_off``: Enable bypass when power supply is turned off.
1653 Set the bypass watchdog timeout to ``n`` seconds where 0 = instant::
1655 testpmd> set bypass timeout (0|1.5|2|3|4|8|16|32)
1660 Show the bypass configuration for a bypass enabled NIC using the lowest port on the NIC::
1662 testpmd> show bypass config (port_id)
1667 Set link up for a port::
1669 testpmd> set link-up port (port id)
1674 Set link down for a port::
1676 testpmd> set link-down port (port id)
1681 Enable E-tag insertion for a VF on a port::
1683 testpmd> E-tag set insertion on port-tag-id (value) port (port_id) vf (vf_id)
1685 Disable E-tag insertion for a VF on a port::
1687 testpmd> E-tag set insertion off port (port_id) vf (vf_id)
1689 Enable/disable E-tag stripping on a port::
1691 testpmd> E-tag set stripping (on|off) port (port_id)
1693 Enable/disable E-tag based forwarding on a port::
1695 testpmd> E-tag set forwarding (on|off) port (port_id)
1700 Load a dynamic device personalization (DDP) profile and store backup profile::
1702 testpmd> ddp add (port_id) (profile_path[,backup_profile_path])
1707 Delete a dynamic device personalization profile and restore backup profile::
1709 testpmd> ddp del (port_id) (backup_profile_path)
1714 List all items from the ptype mapping table::
1716 testpmd> ptype mapping get (port_id) (valid_only)
1720 * ``valid_only``: A flag indicates if only list valid items(=1) or all items(=0).
1722 Replace a specific or a group of software defined ptype with a new one::
1724 testpmd> ptype mapping replace (port_id) (target) (mask) (pkt_type)
1728 * ``target``: A specific software ptype or a mask to represent a group of software ptypes.
1730 * ``mask``: A flag indicate if "target" is a specific software ptype(=0) or a ptype mask(=1).
1732 * ``pkt_type``: The new software ptype to replace the old ones.
1734 Update hardware defined ptype to software defined packet type mapping table::
1736 testpmd> ptype mapping update (port_id) (hw_ptype) (sw_ptype)
1740 * ``hw_ptype``: hardware ptype as the index of the ptype mapping table.
1742 * ``sw_ptype``: software ptype as the value of the ptype mapping table.
1744 Reset ptype mapping table::
1746 testpmd> ptype mapping reset (port_id)
1748 config per port Rx offloading
1749 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1751 Enable or disable a per port Rx offloading on all Rx queues of a port::
1753 testpmd> port config (port_id) rx_offload (offloading) on|off
1755 * ``offloading``: can be any of these offloading capability:
1756 vlan_strip, ipv4_cksum, udp_cksum, tcp_cksum, tcp_lro,
1757 qinq_strip, outer_ipv4_cksum, macsec_strip,
1758 header_split, vlan_filter, vlan_extend, jumbo_frame,
1759 scatter, timestamp, security, keep_crc, rss_hash
1761 This command should be run when the port is stopped, or else it will fail.
1763 config per queue Rx offloading
1764 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1766 Enable or disable a per queue Rx offloading only on a specific Rx queue::
1768 testpmd> port (port_id) rxq (queue_id) rx_offload (offloading) on|off
1770 * ``offloading``: can be any of these offloading capability:
1771 vlan_strip, ipv4_cksum, udp_cksum, tcp_cksum, tcp_lro,
1772 qinq_strip, outer_ipv4_cksum, macsec_strip,
1773 header_split, vlan_filter, vlan_extend, jumbo_frame,
1774 scatter, timestamp, security, keep_crc
1776 This command should be run when the port is stopped, or else it will fail.
1778 config per port Tx offloading
1779 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1781 Enable or disable a per port Tx offloading on all Tx queues of a port::
1783 testpmd> port config (port_id) tx_offload (offloading) on|off
1785 * ``offloading``: can be any of these offloading capability:
1786 vlan_insert, ipv4_cksum, udp_cksum, tcp_cksum,
1787 sctp_cksum, tcp_tso, udp_tso, outer_ipv4_cksum,
1788 qinq_insert, vxlan_tnl_tso, gre_tnl_tso,
1789 ipip_tnl_tso, geneve_tnl_tso, macsec_insert,
1790 mt_lockfree, multi_segs, mbuf_fast_free, security
1792 This command should be run when the port is stopped, or else it will fail.
1794 config per queue Tx offloading
1795 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1797 Enable or disable a per queue Tx offloading only on a specific Tx queue::
1799 testpmd> port (port_id) txq (queue_id) tx_offload (offloading) on|off
1801 * ``offloading``: can be any of these offloading capability:
1802 vlan_insert, ipv4_cksum, udp_cksum, tcp_cksum,
1803 sctp_cksum, tcp_tso, udp_tso, outer_ipv4_cksum,
1804 qinq_insert, vxlan_tnl_tso, gre_tnl_tso,
1805 ipip_tnl_tso, geneve_tnl_tso, macsec_insert,
1806 mt_lockfree, multi_segs, mbuf_fast_free, security
1808 This command should be run when the port is stopped, or else it will fail.
1810 Config VXLAN Encap outer layers
1811 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1813 Configure the outer layer to encapsulate a packet inside a VXLAN tunnel::
1815 set vxlan ip-version (ipv4|ipv6) vni (vni) udp-src (udp-src) \
1816 udp-dst (udp-dst) ip-src (ip-src) ip-dst (ip-dst) eth-src (eth-src) \
1819 set vxlan-with-vlan ip-version (ipv4|ipv6) vni (vni) udp-src (udp-src) \
1820 udp-dst (udp-dst) ip-src (ip-src) ip-dst (ip-dst) vlan-tci (vlan-tci) \
1821 eth-src (eth-src) eth-dst (eth-dst)
1823 set vxlan-tos-ttl ip-version (ipv4|ipv6) vni (vni) udp-src (udp-src) \
1824 udp-dst (udp-dst) ip-tos (ip-tos) ip-ttl (ip-ttl) ip-src (ip-src) \
1825 ip-dst (ip-dst) eth-src (eth-src) eth-dst (eth-dst)
1827 These commands will set an internal configuration inside testpmd, any following
1828 flow rule using the action vxlan_encap will use the last configuration set.
1829 To have a different encapsulation header, one of those commands must be called
1830 before the flow rule creation.
1832 Config NVGRE Encap outer layers
1833 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1835 Configure the outer layer to encapsulate a packet inside a NVGRE tunnel::
1837 set nvgre ip-version (ipv4|ipv6) tni (tni) ip-src (ip-src) ip-dst (ip-dst) \
1838 eth-src (eth-src) eth-dst (eth-dst)
1839 set nvgre-with-vlan ip-version (ipv4|ipv6) tni (tni) ip-src (ip-src) \
1840 ip-dst (ip-dst) vlan-tci (vlan-tci) eth-src (eth-src) eth-dst (eth-dst)
1842 These commands will set an internal configuration inside testpmd, any following
1843 flow rule using the action nvgre_encap will use the last configuration set.
1844 To have a different encapsulation header, one of those commands must be called
1845 before the flow rule creation.
1850 Configure the l2 to be used when encapsulating a packet with L2::
1852 set l2_encap ip-version (ipv4|ipv6) eth-src (eth-src) eth-dst (eth-dst)
1853 set l2_encap-with-vlan ip-version (ipv4|ipv6) vlan-tci (vlan-tci) \
1854 eth-src (eth-src) eth-dst (eth-dst)
1856 Those commands will set an internal configuration inside testpmd, any following
1857 flow rule using the action l2_encap will use the last configuration set.
1858 To have a different encapsulation header, one of those commands must be called
1859 before the flow rule creation.
1864 Configure the l2 to be removed when decapsulating a packet with L2::
1866 set l2_decap ip-version (ipv4|ipv6)
1867 set l2_decap-with-vlan ip-version (ipv4|ipv6)
1869 Those commands will set an internal configuration inside testpmd, any following
1870 flow rule using the action l2_decap will use the last configuration set.
1871 To have a different encapsulation header, one of those commands must be called
1872 before the flow rule creation.
1874 Config MPLSoGRE Encap outer layers
1875 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1877 Configure the outer layer to encapsulate a packet inside a MPLSoGRE tunnel::
1879 set mplsogre_encap ip-version (ipv4|ipv6) label (label) \
1880 ip-src (ip-src) ip-dst (ip-dst) eth-src (eth-src) eth-dst (eth-dst)
1881 set mplsogre_encap-with-vlan ip-version (ipv4|ipv6) label (label) \
1882 ip-src (ip-src) ip-dst (ip-dst) vlan-tci (vlan-tci) \
1883 eth-src (eth-src) eth-dst (eth-dst)
1885 These commands will set an internal configuration inside testpmd, any following
1886 flow rule using the action mplsogre_encap will use the last configuration set.
1887 To have a different encapsulation header, one of those commands must be called
1888 before the flow rule creation.
1890 Config MPLSoGRE Decap outer layers
1891 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1893 Configure the outer layer to decapsulate MPLSoGRE packet::
1895 set mplsogre_decap ip-version (ipv4|ipv6)
1896 set mplsogre_decap-with-vlan ip-version (ipv4|ipv6)
1898 These commands will set an internal configuration inside testpmd, any following
1899 flow rule using the action mplsogre_decap will use the last configuration set.
1900 To have a different decapsulation header, one of those commands must be called
1901 before the flow rule creation.
1903 Config MPLSoUDP Encap outer layers
1904 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1906 Configure the outer layer to encapsulate a packet inside a MPLSoUDP tunnel::
1908 set mplsoudp_encap ip-version (ipv4|ipv6) label (label) udp-src (udp-src) \
1909 udp-dst (udp-dst) ip-src (ip-src) ip-dst (ip-dst) \
1910 eth-src (eth-src) eth-dst (eth-dst)
1911 set mplsoudp_encap-with-vlan ip-version (ipv4|ipv6) label (label) \
1912 udp-src (udp-src) udp-dst (udp-dst) ip-src (ip-src) ip-dst (ip-dst) \
1913 vlan-tci (vlan-tci) eth-src (eth-src) eth-dst (eth-dst)
1915 These commands will set an internal configuration inside testpmd, any following
1916 flow rule using the action mplsoudp_encap will use the last configuration set.
1917 To have a different encapsulation header, one of those commands must be called
1918 before the flow rule creation.
1920 Config MPLSoUDP Decap outer layers
1921 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1923 Configure the outer layer to decapsulate MPLSoUDP packet::
1925 set mplsoudp_decap ip-version (ipv4|ipv6)
1926 set mplsoudp_decap-with-vlan ip-version (ipv4|ipv6)
1928 These commands will set an internal configuration inside testpmd, any following
1929 flow rule using the action mplsoudp_decap will use the last configuration set.
1930 To have a different decapsulation header, one of those commands must be called
1931 before the flow rule creation.
1933 Config Raw Encapsulation
1934 ~~~~~~~~~~~~~~~~~~~~~~~~~
1936 Configure the raw data to be used when encapsulating a packet by
1937 rte_flow_action_raw_encap::
1939 set raw_encap {index} {item} [/ {item} [...]] / end_set
1941 There are multiple global buffers for ``raw_encap``, this command will set one
1942 internal buffer index by ``{index}``.
1943 If there is no ``{index}`` specified::
1945 set raw_encap {item} [/ {item} [...]] / end_set
1947 the default index ``0`` is used.
1948 In order to use different encapsulating header, ``index`` must be specified
1949 during the flow rule creation::
1951 testpmd> flow create 0 egress pattern eth / ipv4 / end actions
1952 raw_encap index 2 / end
1954 Otherwise the default index ``0`` is used.
1956 Config Raw Decapsulation
1957 ~~~~~~~~~~~~~~~~~~~~~~~~
1959 Configure the raw data to be used when decapsulating a packet by
1960 rte_flow_action_raw_decap::
1962 set raw_decap {index} {item} [/ {item} [...]] / end_set
1964 There are multiple global buffers for ``raw_decap``, this command will set
1965 one internal buffer index by ``{index}``.
1966 If there is no ``{index}`` specified::
1968 set raw_decap {item} [/ {item} [...]] / end_set
1970 the default index ``0`` is used.
1971 In order to use different decapsulating header, ``index`` must be specified
1972 during the flow rule creation::
1974 testpmd> flow create 0 egress pattern eth / ipv4 / end actions
1975 raw_encap index 3 / end
1977 Otherwise the default index ``0`` is used.
1982 Set fec mode for a specific port::
1984 testpmd> set port (port_id) fec_mode auto|off|rs|baser
1986 Config Sample actions list
1987 ~~~~~~~~~~~~~~~~~~~~~~~~~~
1989 Configure the sample actions list to be used when sampling a packet by
1990 rte_flow_action_sample::
1992 set sample_actions {index} {action} [/ {action} [...]] / end
1994 There are multiple global buffers for ``sample_actions``, this command will set
1995 one internal buffer index by ``{index}``.
1997 In order to use different sample actions list, ``index`` must be specified
1998 during the flow rule creation::
2000 testpmd> flow create 0 ingress pattern eth / ipv4 / end actions
2001 sample ratio 2 index 2 / end
2003 Otherwise the default index ``0`` is used.
2008 The following sections show functions for configuring ports.
2012 Port configuration changes only become active when forwarding is started/restarted.
2017 Attach a port specified by pci address or virtual device args::
2019 testpmd> port attach (identifier)
2021 To attach a new pci device, the device should be recognized by kernel first.
2022 Then it should be moved under DPDK management.
2023 Finally the port can be attached to testpmd.
2025 For example, to move a pci device using ixgbe under DPDK management:
2027 .. code-block:: console
2029 # Check the status of the available devices.
2030 ./usertools/dpdk-devbind.py --status
2032 Network devices using DPDK-compatible driver
2033 ============================================
2036 Network devices using kernel driver
2037 ===================================
2038 0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused=
2041 # Bind the device to igb_uio.
2042 sudo ./usertools/dpdk-devbind.py -b igb_uio 0000:0a:00.0
2045 # Recheck the status of the devices.
2046 ./usertools/dpdk-devbind.py --status
2047 Network devices using DPDK-compatible driver
2048 ============================================
2049 0000:0a:00.0 '82599ES 10-Gigabit' drv=igb_uio unused=
2051 To attach a port created by virtual device, above steps are not needed.
2053 For example, to attach a port whose pci address is 0000:0a:00.0.
2055 .. code-block:: console
2057 testpmd> port attach 0000:0a:00.0
2058 Attaching a new port...
2059 EAL: PCI device 0000:0a:00.0 on NUMA socket -1
2060 EAL: probe driver: 8086:10fb rte_ixgbe_pmd
2061 EAL: PCI memory mapped at 0x7f83bfa00000
2062 EAL: PCI memory mapped at 0x7f83bfa80000
2063 PMD: eth_ixgbe_dev_init(): MAC: 2, PHY: 18, SFP+: 5
2064 PMD: eth_ixgbe_dev_init(): port 0 vendorID=0x8086 deviceID=0x10fb
2065 Port 0 is attached. Now total ports is 1
2068 For example, to attach a port created by pcap PMD.
2070 .. code-block:: console
2072 testpmd> port attach net_pcap0
2073 Attaching a new port...
2074 PMD: Initializing pmd_pcap for net_pcap0
2075 PMD: Creating pcap-backed ethdev on numa socket 0
2076 Port 0 is attached. Now total ports is 1
2079 In this case, identifier is ``net_pcap0``.
2080 This identifier format is the same as ``--vdev`` format of DPDK applications.
2082 For example, to re-attach a bonded port which has been previously detached,
2083 the mode and slave parameters must be given.
2085 .. code-block:: console
2087 testpmd> port attach net_bond_0,mode=0,slave=1
2088 Attaching a new port...
2089 EAL: Initializing pmd_bond for net_bond_0
2090 EAL: Create bonded device net_bond_0 on port 0 in mode 0 on socket 0.
2091 Port 0 is attached. Now total ports is 1
2098 Detach a specific port::
2100 testpmd> port detach (port_id)
2102 Before detaching a port, the port should be stopped and closed.
2104 For example, to detach a pci device port 0.
2106 .. code-block:: console
2108 testpmd> port stop 0
2111 testpmd> port close 0
2115 testpmd> port detach 0
2117 EAL: PCI device 0000:0a:00.0 on NUMA socket -1
2118 EAL: remove driver: 8086:10fb rte_ixgbe_pmd
2119 EAL: PCI memory unmapped at 0x7f83bfa00000
2120 EAL: PCI memory unmapped at 0x7f83bfa80000
2124 For example, to detach a virtual device port 0.
2126 .. code-block:: console
2128 testpmd> port stop 0
2131 testpmd> port close 0
2135 testpmd> port detach 0
2137 PMD: Closing pcap ethdev on numa socket 0
2138 Port 'net_pcap0' is detached. Now total ports is 0
2141 To remove a pci device completely from the system, first detach the port from testpmd.
2142 Then the device should be moved under kernel management.
2143 Finally the device can be removed using kernel pci hotplug functionality.
2145 For example, to move a pci device under kernel management:
2147 .. code-block:: console
2149 sudo ./usertools/dpdk-devbind.py -b ixgbe 0000:0a:00.0
2151 ./usertools/dpdk-devbind.py --status
2153 Network devices using DPDK-compatible driver
2154 ============================================
2157 Network devices using kernel driver
2158 ===================================
2159 0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused=igb_uio
2161 To remove a port created by a virtual device, above steps are not needed.
2166 Start all ports or a specific port::
2168 testpmd> port start (port_id|all)
2173 Stop all ports or a specific port::
2175 testpmd> port stop (port_id|all)
2180 Close all ports or a specific port::
2182 testpmd> port close (port_id|all)
2187 Reset all ports or a specific port::
2189 testpmd> port reset (port_id|all)
2191 User should stop port(s) before resetting and (re-)start after reset.
2193 port config - queue ring size
2194 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2196 Configure a rx/tx queue ring size::
2198 testpmd> port (port_id) (rxq|txq) (queue_id) ring_size (value)
2200 Only take effect after command that (re-)start the port or command that setup specific queue.
2202 port start/stop queue
2203 ~~~~~~~~~~~~~~~~~~~~~
2205 Start/stop a rx/tx queue on a specific port::
2207 testpmd> port (port_id) (rxq|txq) (queue_id) (start|stop)
2209 port config - queue deferred start
2210 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2212 Switch on/off deferred start of a specific port queue::
2214 testpmd> port (port_id) (rxq|txq) (queue_id) deferred_start (on|off)
2217 ~~~~~~~~~~~~~~~~~~~~~
2219 Setup a rx/tx queue on a specific port::
2221 testpmd> port (port_id) (rxq|txq) (queue_id) setup
2223 Only take effect when port is started.
2228 Set the speed and duplex mode for all ports or a specific port::
2230 testpmd> port config (port_id|all) speed (10|100|1000|10000|25000|40000|50000|100000|200000|auto) \
2231 duplex (half|full|auto)
2233 port config - queues/descriptors
2234 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2236 Set number of queues/descriptors for rxq, txq, rxd and txd::
2238 testpmd> port config all (rxq|txq|rxd|txd) (value)
2240 This is equivalent to the ``--rxq``, ``--txq``, ``--rxd`` and ``--txd`` command-line options.
2242 port config - max-pkt-len
2243 ~~~~~~~~~~~~~~~~~~~~~~~~~
2245 Set the maximum packet length::
2247 testpmd> port config all max-pkt-len (value)
2249 This is equivalent to the ``--max-pkt-len`` command-line option.
2251 port config - max-lro-pkt-size
2252 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2254 Set the maximum LRO aggregated packet size::
2256 testpmd> port config all max-lro-pkt-size (value)
2258 This is equivalent to the ``--max-lro-pkt-size`` command-line option.
2260 port config - Drop Packets
2261 ~~~~~~~~~~~~~~~~~~~~~~~~~~
2263 Enable or disable packet drop on all RX queues of all ports when no receive buffers available::
2265 testpmd> port config all drop-en (on|off)
2267 Packet dropping when no receive buffers available is off by default.
2269 The ``on`` option is equivalent to the ``--enable-drop-en`` command-line option.
2274 Set the RSS (Receive Side Scaling) mode on or off::
2276 testpmd> port config all rss (all|default|eth|vlan|ip|tcp|udp|sctp|ether|port|vxlan|geneve|nvgre|vxlan-gpe|l2tpv3|esp|ah|pfcp|ecpri|mpls|none)
2278 RSS is on by default.
2280 The ``all`` option is equivalent to eth|vlan|ip|tcp|udp|sctp|ether|l2tpv3|esp|ah|pfcp.
2282 The ``default`` option enables all supported RSS types reported by device info.
2284 The ``none`` option is equivalent to the ``--disable-rss`` command-line option.
2286 port config - RSS Reta
2287 ~~~~~~~~~~~~~~~~~~~~~~
2289 Set the RSS (Receive Side Scaling) redirection table::
2291 testpmd> port config all rss reta (hash,queue)[,(hash,queue)]
2296 Set the DCB mode for an individual port::
2298 testpmd> port config (port_id) dcb vt (on|off) (traffic_class) pfc (on|off)
2300 The traffic class should be 4 or 8.
2305 Set the number of packets per burst::
2307 testpmd> port config all burst (value)
2309 This is equivalent to the ``--burst`` command-line option.
2311 port config - Threshold
2312 ~~~~~~~~~~~~~~~~~~~~~~~
2314 Set thresholds for TX/RX queues::
2316 testpmd> port config all (threshold) (value)
2318 Where the threshold type can be:
2320 * ``txpt:`` Set the prefetch threshold register of the TX rings, 0 <= value <= 255.
2322 * ``txht:`` Set the host threshold register of the TX rings, 0 <= value <= 255.
2324 * ``txwt:`` Set the write-back threshold register of the TX rings, 0 <= value <= 255.
2326 * ``rxpt:`` Set the prefetch threshold register of the RX rings, 0 <= value <= 255.
2328 * ``rxht:`` Set the host threshold register of the RX rings, 0 <= value <= 255.
2330 * ``rxwt:`` Set the write-back threshold register of the RX rings, 0 <= value <= 255.
2332 * ``txfreet:`` Set the transmit free threshold of the TX rings, 0 <= value <= txd.
2334 * ``rxfreet:`` Set the transmit free threshold of the RX rings, 0 <= value <= rxd.
2336 * ``txrst:`` Set the transmit RS bit threshold of TX rings, 0 <= value <= txd.
2338 These threshold options are also available from the command-line.
2340 port config pctype mapping
2341 ~~~~~~~~~~~~~~~~~~~~~~~~~~
2343 Reset pctype mapping table::
2345 testpmd> port config (port_id) pctype mapping reset
2347 Update hardware defined pctype to software defined flow type mapping table::
2349 testpmd> port config (port_id) pctype mapping update (pctype_id_0[,pctype_id_1]*) (flow_type_id)
2353 * ``pctype_id_x``: hardware pctype id as index of bit in bitmask value of the pctype mapping table.
2355 * ``flow_type_id``: software flow type id as the index of the pctype mapping table.
2357 port config input set
2358 ~~~~~~~~~~~~~~~~~~~~~
2360 Config RSS/FDIR/FDIR flexible payload input set for some pctype::
2362 testpmd> port config (port_id) pctype (pctype_id) \
2363 (hash_inset|fdir_inset|fdir_flx_inset) \
2364 (get|set|clear) field (field_idx)
2366 Clear RSS/FDIR/FDIR flexible payload input set for some pctype::
2368 testpmd> port config (port_id) pctype (pctype_id) \
2369 (hash_inset|fdir_inset|fdir_flx_inset) clear all
2373 * ``pctype_id``: hardware packet classification types.
2374 * ``field_idx``: hardware field index.
2376 port config udp_tunnel_port
2377 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
2379 Add/remove UDP tunnel port for VXLAN/GENEVE tunneling protocols::
2381 testpmd> port config (port_id) udp_tunnel_port add|rm vxlan|geneve|vxlan-gpe|ecpri (udp_port)
2383 port config tx_metadata
2384 ~~~~~~~~~~~~~~~~~~~~~~~
2386 Set Tx metadata value per port.
2387 testpmd will add this value to any Tx packet sent from this port::
2389 testpmd> port config (port_id) tx_metadata (value)
2394 Set/clear dynamic flag per port.
2395 testpmd will register this flag in the mbuf (same registration
2396 for both Tx and Rx). Then set/clear this flag for each Tx
2397 packet sent from this port. The set bit only works for Tx packet::
2399 testpmd> port config (port_id) dynf (name) (set|clear)
2404 To configure MTU(Maximum Transmission Unit) on devices using testpmd::
2406 testpmd> port config mtu (port_id) (value)
2408 port config rss hash key
2409 ~~~~~~~~~~~~~~~~~~~~~~~~
2411 To configure the RSS hash key used to compute the RSS
2412 hash of input [IP] packets received on port::
2414 testpmd> port config <port_id> rss-hash-key (ipv4|ipv4-frag|\
2415 ipv4-tcp|ipv4-udp|ipv4-sctp|ipv4-other|\
2416 ipv6|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|\
2417 ipv6-other|l2-payload|ipv6-ex|ipv6-tcp-ex|\
2418 ipv6-udp-ex <string of hex digits \
2419 (variable length, NIC dependent)>)
2421 port cleanup txq mbufs
2422 ~~~~~~~~~~~~~~~~~~~~~~
2424 To cleanup txq mbufs currently cached by driver::
2426 testpmd> port cleanup (port_id) txq (queue_id) (free_cnt)
2428 If the value of ``free_cnt`` is 0, driver should free all cached mbufs.
2433 The following sections show functions for device operations.
2438 Detach a device specified by pci address or virtual device args::
2440 testpmd> device detach (identifier)
2442 Before detaching a device associated with ports, the ports should be stopped and closed.
2444 For example, to detach a pci device whose address is 0002:03:00.0.
2446 .. code-block:: console
2448 testpmd> device detach 0002:03:00.0
2449 Removing a device...
2450 Port 1 is now closed
2451 EAL: Releasing pci mapped resource for 0002:03:00.0
2452 EAL: Calling pci_unmap_resource for 0002:03:00.0 at 0x218a050000
2453 EAL: Calling pci_unmap_resource for 0002:03:00.0 at 0x218c050000
2454 Device 0002:03:00.0 is detached
2455 Now total ports is 1
2457 For example, to detach a port created by pcap PMD.
2459 .. code-block:: console
2461 testpmd> device detach net_pcap0
2462 Removing a device...
2463 Port 0 is now closed
2464 Device net_pcap0 is detached
2465 Now total ports is 0
2468 In this case, identifier is ``net_pcap0``.
2469 This identifier format is the same as ``--vdev`` format of DPDK applications.
2471 Link Bonding Functions
2472 ----------------------
2474 The Link Bonding functions make it possible to dynamically create and
2475 manage link bonding devices from within testpmd interactive prompt.
2477 create bonded device
2478 ~~~~~~~~~~~~~~~~~~~~
2480 Create a new bonding device::
2482 testpmd> create bonded device (mode) (socket)
2484 For example, to create a bonded device in mode 1 on socket 0::
2486 testpmd> create bonded device 1 0
2487 created new bonded device (port X)
2492 Adds Ethernet device to a Link Bonding device::
2494 testpmd> add bonding slave (slave id) (port id)
2496 For example, to add Ethernet device (port 6) to a Link Bonding device (port 10)::
2498 testpmd> add bonding slave 6 10
2501 remove bonding slave
2502 ~~~~~~~~~~~~~~~~~~~~
2504 Removes an Ethernet slave device from a Link Bonding device::
2506 testpmd> remove bonding slave (slave id) (port id)
2508 For example, to remove Ethernet slave device (port 6) to a Link Bonding device (port 10)::
2510 testpmd> remove bonding slave 6 10
2515 Set the Link Bonding mode of a Link Bonding device::
2517 testpmd> set bonding mode (value) (port id)
2519 For example, to set the bonding mode of a Link Bonding device (port 10) to broadcast (mode 3)::
2521 testpmd> set bonding mode 3 10
2526 Set an Ethernet slave device as the primary device on a Link Bonding device::
2528 testpmd> set bonding primary (slave id) (port id)
2530 For example, to set the Ethernet slave device (port 6) as the primary port of a Link Bonding device (port 10)::
2532 testpmd> set bonding primary 6 10
2537 Set the MAC address of a Link Bonding device::
2539 testpmd> set bonding mac (port id) (mac)
2541 For example, to set the MAC address of a Link Bonding device (port 10) to 00:00:00:00:00:01::
2543 testpmd> set bonding mac 10 00:00:00:00:00:01
2545 set bonding balance_xmit_policy
2546 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2548 Set the transmission policy for a Link Bonding device when it is in Balance XOR mode::
2550 testpmd> set bonding balance_xmit_policy (port_id) (l2|l23|l34)
2552 For example, set a Link Bonding device (port 10) to use a balance policy of layer 3+4 (IP addresses & UDP ports)::
2554 testpmd> set bonding balance_xmit_policy 10 l34
2557 set bonding mon_period
2558 ~~~~~~~~~~~~~~~~~~~~~~
2560 Set the link status monitoring polling period in milliseconds for a bonding device.
2562 This adds support for PMD slave devices which do not support link status interrupts.
2563 When the mon_period is set to a value greater than 0 then all PMD's which do not support
2564 link status ISR will be queried every polling interval to check if their link status has changed::
2566 testpmd> set bonding mon_period (port_id) (value)
2568 For example, to set the link status monitoring polling period of bonded device (port 5) to 150ms::
2570 testpmd> set bonding mon_period 5 150
2573 set bonding lacp dedicated_queue
2574 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2576 Enable dedicated tx/rx queues on bonding devices slaves to handle LACP control plane traffic
2577 when in mode 4 (link-aggregation-802.3ad)::
2579 testpmd> set bonding lacp dedicated_queues (port_id) (enable|disable)
2582 set bonding agg_mode
2583 ~~~~~~~~~~~~~~~~~~~~
2585 Enable one of the specific aggregators mode when in mode 4 (link-aggregation-802.3ad)::
2587 testpmd> set bonding agg_mode (port_id) (bandwidth|count|stable)
2593 Show the current configuration of a Link Bonding device::
2595 testpmd> show bonding config (port id)
2598 to show the configuration a Link Bonding device (port 9) with 3 slave devices (1, 3, 4)
2599 in balance mode with a transmission policy of layer 2+3::
2601 testpmd> show bonding config 9
2603 Balance Xmit Policy: BALANCE_XMIT_POLICY_LAYER23
2605 Active Slaves (3): [1 3 4]
2608 show bonding lacp info
2609 ~~~~~~~~~~~~~~~~~~~~~~
2611 Show information about the Link Bonding device in mode 4 (link-aggregation-802.3ad)::
2613 testpmd> show bonding lacp info (port_id)
2618 The Register Functions can be used to read from and write to registers on the network card referenced by a port number.
2619 This is mainly useful for debugging purposes.
2620 Reference should be made to the appropriate datasheet for the network card for details on the register addresses
2621 and fields that can be accessed.
2626 Display the value of a port register::
2628 testpmd> read reg (port_id) (address)
2630 For example, to examine the Flow Director control register (FDIRCTL, 0x0000EE000) on an Intel 82599 10 GbE Controller::
2632 testpmd> read reg 0 0xEE00
2633 port 0 PCI register at offset 0xEE00: 0x4A060029 (1241907241)
2638 Display a port register bit field::
2640 testpmd> read regfield (port_id) (address) (bit_x) (bit_y)
2642 For example, reading the lowest two bits from the register in the example above::
2644 testpmd> read regfield 0 0xEE00 0 1
2645 port 0 PCI register at offset 0xEE00: bits[0, 1]=0x1 (1)
2650 Display a single port register bit::
2652 testpmd> read regbit (port_id) (address) (bit_x)
2654 For example, reading the lowest bit from the register in the example above::
2656 testpmd> read regbit 0 0xEE00 0
2657 port 0 PCI register at offset 0xEE00: bit 0=1
2662 Set the value of a port register::
2664 testpmd> write reg (port_id) (address) (value)
2666 For example, to clear a register::
2668 testpmd> write reg 0 0xEE00 0x0
2669 port 0 PCI register at offset 0xEE00: 0x00000000 (0)
2674 Set bit field of a port register::
2676 testpmd> write regfield (port_id) (address) (bit_x) (bit_y) (value)
2678 For example, writing to the register cleared in the example above::
2680 testpmd> write regfield 0 0xEE00 0 1 2
2681 port 0 PCI register at offset 0xEE00: 0x00000002 (2)
2686 Set single bit value of a port register::
2688 testpmd> write regbit (port_id) (address) (bit_x) (value)
2690 For example, to set the high bit in the register from the example above::
2692 testpmd> write regbit 0 0xEE00 31 1
2693 port 0 PCI register at offset 0xEE00: 0x8000000A (2147483658)
2695 Traffic Metering and Policing
2696 -----------------------------
2698 The following section shows functions for configuring traffic metering and
2699 policing on the ethernet device through the use of generic ethdev API.
2701 show port traffic management capability
2702 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2704 Show traffic metering and policing capability of the port::
2706 testpmd> show port meter cap (port_id)
2708 add port meter profile (srTCM rfc2967)
2709 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2711 Add meter profile (srTCM rfc2697) to the ethernet device::
2713 testpmd> add port meter profile srtcm_rfc2697 (port_id) (profile_id) \
2714 (cir) (cbs) (ebs) (packet_mode)
2718 * ``profile_id``: ID for the meter profile.
2719 * ``cir``: Committed Information Rate (CIR) (bytes per second or packets per second).
2720 * ``cbs``: Committed Burst Size (CBS) (bytes or packets).
2721 * ``ebs``: Excess Burst Size (EBS) (bytes or packets).
2722 * ``packet_mode``: Packets mode for meter profile.
2724 add port meter profile (trTCM rfc2968)
2725 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2727 Add meter profile (srTCM rfc2698) to the ethernet device::
2729 testpmd> add port meter profile trtcm_rfc2698 (port_id) (profile_id) \
2730 (cir) (pir) (cbs) (pbs) (packet_mode)
2734 * ``profile_id``: ID for the meter profile.
2735 * ``cir``: Committed information rate (bytes per second or packets per second).
2736 * ``pir``: Peak information rate (bytes per second or packets per second).
2737 * ``cbs``: Committed burst size (bytes or packets).
2738 * ``pbs``: Peak burst size (bytes or packets).
2739 * ``packet_mode``: Packets mode for meter profile.
2741 add port meter profile (trTCM rfc4115)
2742 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2744 Add meter profile (trTCM rfc4115) to the ethernet device::
2746 testpmd> add port meter profile trtcm_rfc4115 (port_id) (profile_id) \
2747 (cir) (eir) (cbs) (ebs) (packet_mode)
2751 * ``profile_id``: ID for the meter profile.
2752 * ``cir``: Committed information rate (bytes per second or packets per second).
2753 * ``eir``: Excess information rate (bytes per second or packets per second).
2754 * ``cbs``: Committed burst size (bytes or packets).
2755 * ``ebs``: Excess burst size (bytes or packets).
2756 * ``packet_mode``: Packets mode for meter profile.
2758 delete port meter profile
2759 ~~~~~~~~~~~~~~~~~~~~~~~~~
2761 Delete meter profile from the ethernet device::
2763 testpmd> del port meter profile (port_id) (profile_id)
2768 Create new policy object for the ethernet device::
2770 testpmd> add port meter policy (port_id) (policy_id) g_actions \
2771 {action} y_actions {action} r_actions {action}
2775 * ``policy_id``: policy ID.
2776 * ``action``: action lists for green/yellow/red colors.
2781 Delete policy object for the ethernet device::
2783 testpmd> del port meter policy (port_id) (policy_id)
2787 * ``policy_id``: policy ID.
2792 Create new meter object for the ethernet device::
2794 testpmd> create port meter (port_id) (mtr_id) (profile_id) \
2795 (policy_id) (meter_enable) (stats_mask) (shared) \
2796 (use_pre_meter_color) [(dscp_tbl_entry0) (dscp_tbl_entry1)...\
2801 * ``mtr_id``: meter object ID.
2802 * ``profile_id``: ID for the meter profile.
2803 * ``policy_id``: ID for the policy.
2804 * ``meter_enable``: When this parameter has a non-zero value, the meter object
2805 gets enabled at the time of creation, otherwise remains disabled.
2806 * ``stats_mask``: Mask of statistics counter types to be enabled for the
2808 * ``shared``: When this parameter has a non-zero value, the meter object is
2809 shared by multiple flows. Otherwise, meter object is used by single flow.
2810 * ``use_pre_meter_color``: When this parameter has a non-zero value, the
2811 input color for the current meter object is determined by the latest meter
2812 object in the same flow. Otherwise, the current meter object uses the
2813 *dscp_table* to determine the input color.
2814 * ``dscp_tbl_entryx``: DSCP table entry x providing meter providing input
2815 color, 0 <= x <= 63.
2820 Enable meter for the ethernet device::
2822 testpmd> enable port meter (port_id) (mtr_id)
2827 Disable meter for the ethernet device::
2829 testpmd> disable port meter (port_id) (mtr_id)
2834 Delete meter for the ethernet device::
2836 testpmd> del port meter (port_id) (mtr_id)
2838 Set port meter profile
2839 ~~~~~~~~~~~~~~~~~~~~~~
2841 Set meter profile for the ethernet device::
2843 testpmd> set port meter profile (port_id) (mtr_id) (profile_id)
2845 set port meter dscp table
2846 ~~~~~~~~~~~~~~~~~~~~~~~~~
2848 Set meter dscp table for the ethernet device::
2850 testpmd> set port meter dscp table (port_id) (mtr_id) [(dscp_tbl_entry0) \
2851 (dscp_tbl_entry1)...(dscp_tbl_entry63)]
2853 set port meter stats mask
2854 ~~~~~~~~~~~~~~~~~~~~~~~~~
2856 Set meter stats mask for the ethernet device::
2858 testpmd> set port meter stats mask (port_id) (mtr_id) (stats_mask)
2862 * ``stats_mask``: Bit mask indicating statistics counter types to be enabled.
2864 show port meter stats
2865 ~~~~~~~~~~~~~~~~~~~~~
2867 Show meter stats of the ethernet device::
2869 testpmd> show port meter stats (port_id) (mtr_id) (clear)
2873 * ``clear``: Flag that indicates whether the statistics counters should
2874 be cleared (i.e. set to zero) immediately after they have been read or not.
2879 The following section shows functions for configuring traffic management on
2880 the ethernet device through the use of generic TM API.
2882 show port traffic management capability
2883 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2885 Show traffic management capability of the port::
2887 testpmd> show port tm cap (port_id)
2889 show port traffic management capability (hierarchy level)
2890 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2892 Show traffic management hierarchy level capability of the port::
2894 testpmd> show port tm level cap (port_id) (level_id)
2896 show port traffic management capability (hierarchy node level)
2897 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2899 Show the traffic management hierarchy node capability of the port::
2901 testpmd> show port tm node cap (port_id) (node_id)
2903 show port traffic management hierarchy node type
2904 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2906 Show the port traffic management hierarchy node type::
2908 testpmd> show port tm node type (port_id) (node_id)
2910 show port traffic management hierarchy node stats
2911 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2913 Show the port traffic management hierarchy node statistics::
2915 testpmd> show port tm node stats (port_id) (node_id) (clear)
2919 * ``clear``: When this parameter has a non-zero value, the statistics counters
2920 are cleared (i.e. set to zero) immediately after they have been read,
2921 otherwise the statistics counters are left untouched.
2923 Add port traffic management private shaper profile
2924 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2926 Add the port traffic management private shaper profile::
2928 testpmd> add port tm node shaper profile (port_id) (shaper_profile_id) \
2929 (cmit_tb_rate) (cmit_tb_size) (peak_tb_rate) (peak_tb_size) \
2930 (packet_length_adjust) (packet_mode)
2934 * ``shaper_profile id``: Shaper profile ID for the new profile.
2935 * ``cmit_tb_rate``: Committed token bucket rate (bytes per second or packets per second).
2936 * ``cmit_tb_size``: Committed token bucket size (bytes or packets).
2937 * ``peak_tb_rate``: Peak token bucket rate (bytes per second or packets per second).
2938 * ``peak_tb_size``: Peak token bucket size (bytes or packets).
2939 * ``packet_length_adjust``: The value (bytes) to be added to the length of
2940 each packet for the purpose of shaping. This parameter value can be used to
2941 correct the packet length with the framing overhead bytes that are consumed
2943 * ``packet_mode``: Shaper configured in packet mode. This parameter value if
2944 zero, configures shaper in byte mode and if non-zero configures it in packet
2947 Delete port traffic management private shaper profile
2948 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2950 Delete the port traffic management private shaper::
2952 testpmd> del port tm node shaper profile (port_id) (shaper_profile_id)
2956 * ``shaper_profile id``: Shaper profile ID that needs to be deleted.
2958 Add port traffic management shared shaper
2959 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2961 Create the port traffic management shared shaper::
2963 testpmd> add port tm node shared shaper (port_id) (shared_shaper_id) \
2968 * ``shared_shaper_id``: Shared shaper ID to be created.
2969 * ``shaper_profile id``: Shaper profile ID for shared shaper.
2971 Set port traffic management shared shaper
2972 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2974 Update the port traffic management shared shaper::
2976 testpmd> set port tm node shared shaper (port_id) (shared_shaper_id) \
2981 * ``shared_shaper_id``: Shared shaper ID to be update.
2982 * ``shaper_profile id``: Shaper profile ID for shared shaper.
2984 Delete port traffic management shared shaper
2985 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2987 Delete the port traffic management shared shaper::
2989 testpmd> del port tm node shared shaper (port_id) (shared_shaper_id)
2993 * ``shared_shaper_id``: Shared shaper ID to be deleted.
2995 Set port traffic management hierarchy node private shaper
2996 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2998 set the port traffic management hierarchy node private shaper::
3000 testpmd> set port tm node shaper profile (port_id) (node_id) \
3005 * ``shaper_profile id``: Private shaper profile ID to be enabled on the
3008 Add port traffic management WRED profile
3009 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3011 Create a new WRED profile::
3013 testpmd> add port tm node wred profile (port_id) (wred_profile_id) \
3014 (color_g) (min_th_g) (max_th_g) (maxp_inv_g) (wq_log2_g) \
3015 (color_y) (min_th_y) (max_th_y) (maxp_inv_y) (wq_log2_y) \
3016 (color_r) (min_th_r) (max_th_r) (maxp_inv_r) (wq_log2_r)
3020 * ``wred_profile id``: Identifier for the newly create WRED profile
3021 * ``color_g``: Packet color (green)
3022 * ``min_th_g``: Minimum queue threshold for packet with green color
3023 * ``max_th_g``: Minimum queue threshold for packet with green color
3024 * ``maxp_inv_g``: Inverse of packet marking probability maximum value (maxp)
3025 * ``wq_log2_g``: Negated log2 of queue weight (wq)
3026 * ``color_y``: Packet color (yellow)
3027 * ``min_th_y``: Minimum queue threshold for packet with yellow color
3028 * ``max_th_y``: Minimum queue threshold for packet with yellow color
3029 * ``maxp_inv_y``: Inverse of packet marking probability maximum value (maxp)
3030 * ``wq_log2_y``: Negated log2 of queue weight (wq)
3031 * ``color_r``: Packet color (red)
3032 * ``min_th_r``: Minimum queue threshold for packet with yellow color
3033 * ``max_th_r``: Minimum queue threshold for packet with yellow color
3034 * ``maxp_inv_r``: Inverse of packet marking probability maximum value (maxp)
3035 * ``wq_log2_r``: Negated log2 of queue weight (wq)
3037 Delete port traffic management WRED profile
3038 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3040 Delete the WRED profile::
3042 testpmd> del port tm node wred profile (port_id) (wred_profile_id)
3044 Add port traffic management hierarchy nonleaf node
3045 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3047 Add nonleaf node to port traffic management hierarchy::
3049 testpmd> add port tm nonleaf node (port_id) (node_id) (parent_node_id) \
3050 (priority) (weight) (level_id) (shaper_profile_id) \
3051 (n_sp_priorities) (stats_mask) (n_shared_shapers) \
3052 [(shared_shaper_0) (shared_shaper_1) ...] \
3056 * ``parent_node_id``: Node ID of the parent.
3057 * ``priority``: Node priority (highest node priority is zero). This is used by
3058 the SP algorithm running on the parent node for scheduling this node.
3059 * ``weight``: Node weight (lowest weight is one). The node weight is relative
3060 to the weight sum of all siblings that have the same priority. It is used by
3061 the WFQ algorithm running on the parent node for scheduling this node.
3062 * ``level_id``: Hierarchy level of the node.
3063 * ``shaper_profile_id``: Shaper profile ID of the private shaper to be used by
3065 * ``n_sp_priorities``: Number of strict priorities.
3066 * ``stats_mask``: Mask of statistics counter types to be enabled for this node.
3067 * ``n_shared_shapers``: Number of shared shapers.
3068 * ``shared_shaper_id``: Shared shaper id.
3070 Add port traffic management hierarchy nonleaf node with packet mode
3071 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3073 Add nonleaf node with packet mode to port traffic management hierarchy::
3075 testpmd> add port tm nonleaf node pktmode (port_id) (node_id) (parent_node_id) \
3076 (priority) (weight) (level_id) (shaper_profile_id) \
3077 (n_sp_priorities) (stats_mask) (n_shared_shapers) \
3078 [(shared_shaper_0) (shared_shaper_1) ...] \
3082 * ``parent_node_id``: Node ID of the parent.
3083 * ``priority``: Node priority (highest node priority is zero). This is used by
3084 the SP algorithm running on the parent node for scheduling this node.
3085 * ``weight``: Node weight (lowest weight is one). The node weight is relative
3086 to the weight sum of all siblings that have the same priority. It is used by
3087 the WFQ algorithm running on the parent node for scheduling this node.
3088 * ``level_id``: Hierarchy level of the node.
3089 * ``shaper_profile_id``: Shaper profile ID of the private shaper to be used by
3091 * ``n_sp_priorities``: Number of strict priorities. Packet mode is enabled on
3093 * ``stats_mask``: Mask of statistics counter types to be enabled for this node.
3094 * ``n_shared_shapers``: Number of shared shapers.
3095 * ``shared_shaper_id``: Shared shaper id.
3097 Add port traffic management hierarchy leaf node
3098 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3100 Add leaf node to port traffic management hierarchy::
3102 testpmd> add port tm leaf node (port_id) (node_id) (parent_node_id) \
3103 (priority) (weight) (level_id) (shaper_profile_id) \
3104 (cman_mode) (wred_profile_id) (stats_mask) (n_shared_shapers) \
3105 [(shared_shaper_id) (shared_shaper_id) ...] \
3109 * ``parent_node_id``: Node ID of the parent.
3110 * ``priority``: Node priority (highest node priority is zero). This is used by
3111 the SP algorithm running on the parent node for scheduling this node.
3112 * ``weight``: Node weight (lowest weight is one). The node weight is relative
3113 to the weight sum of all siblings that have the same priority. It is used by
3114 the WFQ algorithm running on the parent node for scheduling this node.
3115 * ``level_id``: Hierarchy level of the node.
3116 * ``shaper_profile_id``: Shaper profile ID of the private shaper to be used by
3118 * ``cman_mode``: Congestion management mode to be enabled for this node.
3119 * ``wred_profile_id``: WRED profile id to be enabled for this node.
3120 * ``stats_mask``: Mask of statistics counter types to be enabled for this node.
3121 * ``n_shared_shapers``: Number of shared shapers.
3122 * ``shared_shaper_id``: Shared shaper id.
3124 Delete port traffic management hierarchy node
3125 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3127 Delete node from port traffic management hierarchy::
3129 testpmd> del port tm node (port_id) (node_id)
3131 Update port traffic management hierarchy parent node
3132 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3134 Update port traffic management hierarchy parent node::
3136 testpmd> set port tm node parent (port_id) (node_id) (parent_node_id) \
3139 This function can only be called after the hierarchy commit invocation. Its
3140 success depends on the port support for this operation, as advertised through
3141 the port capability set. This function is valid for all nodes of the traffic
3142 management hierarchy except root node.
3144 Suspend port traffic management hierarchy node
3145 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3147 testpmd> suspend port tm node (port_id) (node_id)
3149 Resume port traffic management hierarchy node
3150 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3152 testpmd> resume port tm node (port_id) (node_id)
3154 Commit port traffic management hierarchy
3155 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3157 Commit the traffic management hierarchy on the port::
3159 testpmd> port tm hierarchy commit (port_id) (clean_on_fail)
3163 * ``clean_on_fail``: When set to non-zero, hierarchy is cleared on function
3164 call failure. On the other hand, hierarchy is preserved when this parameter
3167 Set port traffic management mark VLAN dei
3168 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3170 Enables/Disables the traffic management marking on the port for VLAN packets::
3172 testpmd> set port tm mark vlan_dei <port_id> <green> <yellow> <red>
3176 * ``port_id``: The port which on which VLAN packets marked as ``green`` or
3177 ``yellow`` or ``red`` will have dei bit enabled
3179 * ``green`` enable 1, disable 0 marking for dei bit of VLAN packets marked as green
3181 * ``yellow`` enable 1, disable 0 marking for dei bit of VLAN packets marked as yellow
3183 * ``red`` enable 1, disable 0 marking for dei bit of VLAN packets marked as red
3185 Set port traffic management mark IP dscp
3186 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3188 Enables/Disables the traffic management marking on the port for IP dscp packets::
3190 testpmd> set port tm mark ip_dscp <port_id> <green> <yellow> <red>
3194 * ``port_id``: The port which on which IP packets marked as ``green`` or
3195 ``yellow`` or ``red`` will have IP dscp bits updated
3197 * ``green`` enable 1, disable 0 marking IP dscp to low drop precedence for green packets
3199 * ``yellow`` enable 1, disable 0 marking IP dscp to medium drop precedence for yellow packets
3201 * ``red`` enable 1, disable 0 marking IP dscp to high drop precedence for red packets
3203 Set port traffic management mark IP ecn
3204 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3206 Enables/Disables the traffic management marking on the port for IP ecn packets::
3208 testpmd> set port tm mark ip_ecn <port_id> <green> <yellow> <red>
3212 * ``port_id``: The port which on which IP packets marked as ``green`` or
3213 ``yellow`` or ``red`` will have IP ecn bits updated
3215 * ``green`` enable 1, disable 0 marking IP ecn for green marked packets with ecn of 2'b01 or 2'b10
3216 to ecn of 2'b11 when IP is caring TCP or SCTP
3218 * ``yellow`` enable 1, disable 0 marking IP ecn for yellow marked packets with ecn of 2'b01 or 2'b10
3219 to ecn of 2'b11 when IP is caring TCP or SCTP
3221 * ``red`` enable 1, disable 0 marking IP ecn for yellow marked packets with ecn of 2'b01 or 2'b10
3222 to ecn of 2'b11 when IP is caring TCP or SCTP
3227 This section details the available filter functions that are available.
3229 Note these functions interface the deprecated legacy filtering framework,
3230 superseded by *rte_flow*. See `Flow rules management`_.
3232 .. _testpmd_flow_director:
3237 Set flow director's input masks::
3239 flow_director_mask (port_id) mode IP vlan (vlan_value) \
3240 src_mask (ipv4_src) (ipv6_src) (src_port) \
3241 dst_mask (ipv4_dst) (ipv6_dst) (dst_port)
3243 flow_director_mask (port_id) mode MAC-VLAN vlan (vlan_value)
3245 flow_director_mask (port_id) mode Tunnel vlan (vlan_value) \
3246 mac (mac_value) tunnel-type (tunnel_type_value) \
3247 tunnel-id (tunnel_id_value)
3249 Example, to set flow director mask on port 0::
3251 testpmd> flow_director_mask 0 mode IP vlan 0xefff \
3252 src_mask 255.255.255.255 \
3253 FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF \
3254 dst_mask 255.255.255.255 \
3255 FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF
3257 flow_director_flex_payload
3258 ~~~~~~~~~~~~~~~~~~~~~~~~~~
3260 Configure flexible payload selection::
3262 flow_director_flex_payload (port_id) (raw|l2|l3|l4) (config)
3264 For example, to select the first 16 bytes from the offset 4 (bytes) of packet's payload as flexible payload::
3266 testpmd> flow_director_flex_payload 0 l4 \
3267 (4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19)
3270 .. _testpmd_rte_flow:
3272 Flow rules management
3273 ---------------------
3275 Control of the generic flow API (*rte_flow*) is fully exposed through the
3276 ``flow`` command (validation, creation, destruction, queries and operation
3279 Considering *rte_flow* overlaps with all `Filter Functions`_, using both
3280 features simultaneously may cause undefined side-effects and is therefore
3286 Because the ``flow`` command uses dynamic tokens to handle the large number
3287 of possible flow rules combinations, its behavior differs slightly from
3288 other commands, in particular:
3290 - Pressing *?* or the *<tab>* key displays contextual help for the current
3291 token, not that of the entire command.
3293 - Optional and repeated parameters are supported (provided they are listed
3294 in the contextual help).
3296 The first parameter stands for the operation mode. Possible operations and
3297 their general syntax are described below. They are covered in detail in the
3300 - Check whether a flow rule can be created::
3302 flow validate {port_id}
3303 [group {group_id}] [priority {level}] [ingress] [egress] [transfer]
3304 pattern {item} [/ {item} [...]] / end
3305 actions {action} [/ {action} [...]] / end
3307 - Create a flow rule::
3309 flow create {port_id}
3310 [group {group_id}] [priority {level}] [ingress] [egress] [transfer]
3311 pattern {item} [/ {item} [...]] / end
3312 actions {action} [/ {action} [...]] / end
3314 - Destroy specific flow rules::
3316 flow destroy {port_id} rule {rule_id} [...]
3318 - Destroy all flow rules::
3320 flow flush {port_id}
3322 - Query an existing flow rule::
3324 flow query {port_id} {rule_id} {action}
3326 - List existing flow rules sorted by priority, filtered by group
3329 flow list {port_id} [group {group_id}] [...]
3331 - Restrict ingress traffic to the defined flow rules::
3333 flow isolate {port_id} {boolean}
3335 - Dump internal representation information of all flows in hardware::
3337 flow dump {port_id} all {output_file}
3341 flow dump {port_id} rule {rule_id} {output_file}
3343 - List and destroy aged flow rules::
3345 flow aged {port_id} [destroy]
3347 - Tunnel offload - create a tunnel stub::
3349 flow tunnel create {port_id} type {tunnel_type}
3351 - Tunnel offload - destroy a tunnel stub::
3353 flow tunnel destroy {port_id} id {tunnel_id}
3355 - Tunnel offload - list port tunnel stubs::
3357 flow tunnel list {port_id}
3359 Creating a tunnel stub for offload
3360 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3362 ``flow tunnel create`` setup a tunnel stub for tunnel offload flow rules::
3364 flow tunnel create {port_id} type {tunnel_type}
3366 If successful, it will return a tunnel stub ID usable with other commands::
3368 port [...]: flow tunnel #[...] type [...]
3370 Tunnel stub ID is relative to a port.
3372 Destroying tunnel offload stub
3373 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3375 ``flow tunnel destroy`` destroy port tunnel stub::
3377 flow tunnel destroy {port_id} id {tunnel_id}
3379 Listing tunnel offload stubs
3380 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3382 ``flow tunnel list`` list port tunnel offload stubs::
3384 flow tunnel list {port_id}
3386 Validating flow rules
3387 ~~~~~~~~~~~~~~~~~~~~~
3389 ``flow validate`` reports whether a flow rule would be accepted by the
3390 underlying device in its current state but stops short of creating it. It is
3391 bound to ``rte_flow_validate()``::
3393 flow validate {port_id}
3394 [group {group_id}] [priority {level}] [ingress] [egress] [transfer]
3395 pattern {item} [/ {item} [...]] / end
3396 actions {action} [/ {action} [...]] / end
3398 If successful, it will show::
3402 Otherwise it will show an error message of the form::
3404 Caught error type [...] ([...]): [...]
3406 This command uses the same parameters as ``flow create``, their format is
3407 described in `Creating flow rules`_.
3409 Check whether redirecting any Ethernet packet received on port 0 to RX queue
3410 index 6 is supported::
3412 testpmd> flow validate 0 ingress pattern eth / end
3413 actions queue index 6 / end
3417 Port 0 does not support TCPv6 rules::
3419 testpmd> flow validate 0 ingress pattern eth / ipv6 / tcp / end
3421 Caught error type 9 (specific pattern item): Invalid argument
3427 ``flow create`` validates and creates the specified flow rule. It is bound
3428 to ``rte_flow_create()``::
3430 flow create {port_id}
3431 [group {group_id}] [priority {level}] [ingress] [egress] [transfer]
3432 [tunnel_set {tunnel_id}] [tunnel_match {tunnel_id}]
3433 pattern {item} [/ {item} [...]] / end
3434 actions {action} [/ {action} [...]] / end
3436 If successful, it will return a flow rule ID usable with other commands::
3438 Flow rule #[...] created
3440 Otherwise it will show an error message of the form::
3442 Caught error type [...] ([...]): [...]
3444 Parameters describe in the following order:
3446 - Attributes (*group*, *priority*, *ingress*, *egress*, *transfer* tokens).
3447 - Tunnel offload specification (tunnel_set, tunnel_match)
3448 - A matching pattern, starting with the *pattern* token and terminated by an
3450 - Actions, starting with the *actions* token and terminated by an *end*
3453 These translate directly to *rte_flow* objects provided as-is to the
3454 underlying functions.
3456 The shortest valid definition only comprises mandatory tokens::
3458 testpmd> flow create 0 pattern end actions end
3460 Note that PMDs may refuse rules that essentially do nothing such as this
3463 **All unspecified object values are automatically initialized to 0.**
3468 These tokens affect flow rule attributes (``struct rte_flow_attr``) and are
3469 specified before the ``pattern`` token.
3471 - ``group {group id}``: priority group.
3472 - ``priority {level}``: priority level within group.
3473 - ``ingress``: rule applies to ingress traffic.
3474 - ``egress``: rule applies to egress traffic.
3475 - ``transfer``: apply rule directly to endpoints found in pattern.
3477 Each instance of an attribute specified several times overrides the previous
3478 value as shown below (group 4 is used)::
3480 testpmd> flow create 0 group 42 group 24 group 4 [...]
3482 Note that once enabled, ``ingress`` and ``egress`` cannot be disabled.
3484 While not specifying a direction is an error, some rules may allow both
3487 Most rules affect RX therefore contain the ``ingress`` token::
3489 testpmd> flow create 0 ingress pattern [...]
3494 Indicate tunnel offload rule type
3496 - ``tunnel_set {tunnel_id}``: mark rule as tunnel offload decap_set type.
3497 - ``tunnel_match {tunnel_id}``: mark rule as tunel offload match type.
3502 A matching pattern starts after the ``pattern`` token. It is made of pattern
3503 items and is terminated by a mandatory ``end`` item.
3505 Items are named after their type (*RTE_FLOW_ITEM_TYPE_* from ``enum
3506 rte_flow_item_type``).
3508 The ``/`` token is used as a separator between pattern items as shown
3511 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end [...]
3513 Note that protocol items like these must be stacked from lowest to highest
3514 layer to make sense. For instance, the following rule is either invalid or
3515 unlikely to match any packet::
3517 testpmd> flow create 0 ingress pattern eth / udp / ipv4 / end [...]
3519 More information on these restrictions can be found in the *rte_flow*
3522 Several items support additional specification structures, for example
3523 ``ipv4`` allows specifying source and destination addresses as follows::
3525 testpmd> flow create 0 ingress pattern eth / ipv4 src is 10.1.1.1
3526 dst is 10.2.0.0 / end [...]
3528 This rule matches all IPv4 traffic with the specified properties.
3530 In this example, ``src`` and ``dst`` are field names of the underlying
3531 ``struct rte_flow_item_ipv4`` object. All item properties can be specified
3532 in a similar fashion.
3534 The ``is`` token means that the subsequent value must be matched exactly,
3535 and assigns ``spec`` and ``mask`` fields in ``struct rte_flow_item``
3536 accordingly. Possible assignment tokens are:
3538 - ``is``: match value perfectly (with full bit-mask).
3539 - ``spec``: match value according to configured bit-mask.
3540 - ``last``: specify upper bound to establish a range.
3541 - ``mask``: specify bit-mask with relevant bits set to one.
3542 - ``prefix``: generate bit-mask with <prefix-length> most-significant bits set to one.
3544 These yield identical results::
3546 ipv4 src is 10.1.1.1
3550 ipv4 src spec 10.1.1.1 src mask 255.255.255.255
3554 ipv4 src spec 10.1.1.1 src prefix 32
3558 ipv4 src is 10.1.1.1 src last 10.1.1.1 # range with a single value
3562 ipv4 src is 10.1.1.1 src last 0 # 0 disables range
3564 Inclusive ranges can be defined with ``last``::
3566 ipv4 src is 10.1.1.1 src last 10.2.3.4 # 10.1.1.1 to 10.2.3.4
3568 Note that ``mask`` affects both ``spec`` and ``last``::
3570 ipv4 src is 10.1.1.1 src last 10.2.3.4 src mask 255.255.0.0
3571 # matches 10.1.0.0 to 10.2.255.255
3573 Properties can be modified multiple times::
3575 ipv4 src is 10.1.1.1 src is 10.1.2.3 src is 10.2.3.4 # matches 10.2.3.4
3579 ipv4 src is 10.1.1.1 src prefix 24 src prefix 16 # matches 10.1.0.0/16
3584 This section lists supported pattern items and their attributes, if any.
3586 - ``end``: end list of pattern items.
3588 - ``void``: no-op pattern item.
3590 - ``invert``: perform actions when pattern does not match.
3592 - ``any``: match any protocol for the current layer.
3594 - ``num {unsigned}``: number of layers covered.
3596 - ``pf``: match traffic from/to the physical function.
3598 - ``vf``: match traffic from/to a virtual function ID.
3600 - ``id {unsigned}``: VF ID.
3602 - ``phy_port``: match traffic from/to a specific physical port.
3604 - ``index {unsigned}``: physical port index.
3606 - ``port_id``: match traffic from/to a given DPDK port ID.
3608 - ``id {unsigned}``: DPDK port ID.
3610 - ``mark``: match value set in previously matched flow rule using the mark action.
3612 - ``id {unsigned}``: arbitrary integer value.
3614 - ``raw``: match an arbitrary byte string.
3616 - ``relative {boolean}``: look for pattern after the previous item.
3617 - ``search {boolean}``: search pattern from offset (see also limit).
3618 - ``offset {integer}``: absolute or relative offset for pattern.
3619 - ``limit {unsigned}``: search area limit for start of pattern.
3620 - ``pattern {string}``: byte string to look for.
3622 - ``eth``: match Ethernet header.
3624 - ``dst {MAC-48}``: destination MAC.
3625 - ``src {MAC-48}``: source MAC.
3626 - ``type {unsigned}``: EtherType or TPID.
3628 - ``vlan``: match 802.1Q/ad VLAN tag.
3630 - ``tci {unsigned}``: tag control information.
3631 - ``pcp {unsigned}``: priority code point.
3632 - ``dei {unsigned}``: drop eligible indicator.
3633 - ``vid {unsigned}``: VLAN identifier.
3634 - ``inner_type {unsigned}``: inner EtherType or TPID.
3636 - ``ipv4``: match IPv4 header.
3638 - ``version_ihl {unsigned}``: IPv4 version and IP header length.
3639 - ``tos {unsigned}``: type of service.
3640 - ``ttl {unsigned}``: time to live.
3641 - ``proto {unsigned}``: next protocol ID.
3642 - ``src {ipv4 address}``: source address.
3643 - ``dst {ipv4 address}``: destination address.
3645 - ``ipv6``: match IPv6 header.
3647 - ``tc {unsigned}``: traffic class.
3648 - ``flow {unsigned}``: flow label.
3649 - ``proto {unsigned}``: protocol (next header).
3650 - ``hop {unsigned}``: hop limit.
3651 - ``src {ipv6 address}``: source address.
3652 - ``dst {ipv6 address}``: destination address.
3654 - ``icmp``: match ICMP header.
3656 - ``type {unsigned}``: ICMP packet type.
3657 - ``code {unsigned}``: ICMP packet code.
3659 - ``udp``: match UDP header.
3661 - ``src {unsigned}``: UDP source port.
3662 - ``dst {unsigned}``: UDP destination port.
3664 - ``tcp``: match TCP header.
3666 - ``src {unsigned}``: TCP source port.
3667 - ``dst {unsigned}``: TCP destination port.
3669 - ``sctp``: match SCTP header.
3671 - ``src {unsigned}``: SCTP source port.
3672 - ``dst {unsigned}``: SCTP destination port.
3673 - ``tag {unsigned}``: validation tag.
3674 - ``cksum {unsigned}``: checksum.
3676 - ``vxlan``: match VXLAN header.
3678 - ``vni {unsigned}``: VXLAN identifier.
3679 - ``last_rsvd {unsigned}``: VXLAN last reserved 8-bits.
3681 - ``e_tag``: match IEEE 802.1BR E-Tag header.
3683 - ``grp_ecid_b {unsigned}``: GRP and E-CID base.
3685 - ``nvgre``: match NVGRE header.
3687 - ``tni {unsigned}``: virtual subnet ID.
3689 - ``mpls``: match MPLS header.
3691 - ``label {unsigned}``: MPLS label.
3693 - ``gre``: match GRE header.
3695 - ``protocol {unsigned}``: protocol type.
3697 - ``gre_key``: match GRE optional key field.
3699 - ``value {unsigned}``: key value.
3701 - ``fuzzy``: fuzzy pattern match, expect faster than default.
3703 - ``thresh {unsigned}``: accuracy threshold.
3705 - ``gtp``, ``gtpc``, ``gtpu``: match GTPv1 header.
3707 - ``teid {unsigned}``: tunnel endpoint identifier.
3709 - ``geneve``: match GENEVE header.
3711 - ``vni {unsigned}``: virtual network identifier.
3712 - ``protocol {unsigned}``: protocol type.
3714 - ``geneve-opt``: match GENEVE header option.
3716 - ``class {unsigned}``: GENEVE option class.
3717 - ``type {unsigned}``: GENEVE option type.
3718 - ``length {unsigned}``: GENEVE option length in 32-bit words.
3719 - ``data {hex string}``: GENEVE option data, the length is defined by
3722 - ``vxlan-gpe``: match VXLAN-GPE header.
3724 - ``vni {unsigned}``: VXLAN-GPE identifier.
3726 - ``arp_eth_ipv4``: match ARP header for Ethernet/IPv4.
3728 - ``sha {MAC-48}``: sender hardware address.
3729 - ``spa {ipv4 address}``: sender IPv4 address.
3730 - ``tha {MAC-48}``: target hardware address.
3731 - ``tpa {ipv4 address}``: target IPv4 address.
3733 - ``ipv6_ext``: match presence of any IPv6 extension header.
3735 - ``next_hdr {unsigned}``: next header.
3737 - ``icmp6``: match any ICMPv6 header.
3739 - ``type {unsigned}``: ICMPv6 type.
3740 - ``code {unsigned}``: ICMPv6 code.
3742 - ``icmp6_nd_ns``: match ICMPv6 neighbor discovery solicitation.
3744 - ``target_addr {ipv6 address}``: target address.
3746 - ``icmp6_nd_na``: match ICMPv6 neighbor discovery advertisement.
3748 - ``target_addr {ipv6 address}``: target address.
3750 - ``icmp6_nd_opt``: match presence of any ICMPv6 neighbor discovery option.
3752 - ``type {unsigned}``: ND option type.
3754 - ``icmp6_nd_opt_sla_eth``: match ICMPv6 neighbor discovery source Ethernet
3755 link-layer address option.
3757 - ``sla {MAC-48}``: source Ethernet LLA.
3759 - ``icmp6_nd_opt_tla_eth``: match ICMPv6 neighbor discovery target Ethernet
3760 link-layer address option.
3762 - ``tla {MAC-48}``: target Ethernet LLA.
3764 - ``meta``: match application specific metadata.
3766 - ``data {unsigned}``: metadata value.
3768 - ``gtp_psc``: match GTP PDU extension header with type 0x85.
3770 - ``pdu_type {unsigned}``: PDU type.
3772 - ``qfi {unsigned}``: QoS flow identifier.
3774 - ``pppoes``, ``pppoed``: match PPPoE header.
3776 - ``session_id {unsigned}``: session identifier.
3778 - ``pppoe_proto_id``: match PPPoE session protocol identifier.
3780 - ``proto_id {unsigned}``: PPP protocol identifier.
3782 - ``l2tpv3oip``: match L2TPv3 over IP header.
3784 - ``session_id {unsigned}``: L2TPv3 over IP session identifier.
3786 - ``ah``: match AH header.
3788 - ``spi {unsigned}``: security parameters index.
3790 - ``pfcp``: match PFCP header.
3792 - ``s_field {unsigned}``: S field.
3793 - ``seid {unsigned}``: session endpoint identifier.
3795 - ``integrity``: match packet integrity.
3797 - ``level {unsigned}``: Packet encapsulation level the item should
3798 apply to. See rte_flow_action_rss for details.
3799 - ``value {unsigned}``: A bitmask that specify what packet elements
3800 must be matched for integrity.
3802 - ``conntrack``: match conntrack state.
3804 - ``port_representor``: match traffic entering the embedded switch from the given ethdev
3806 - ``port_id {unsigned}``: ethdev port ID
3808 - ``represented_port``: match traffic entering the embedded switch from
3809 the entity represented by the given ethdev
3811 - ``ethdev_port_id {unsigned}``: ethdev port ID
3813 - ``l2tpv2``: match L2TPv2 header.
3815 - ``length {unsigned}``: L2TPv2 option length.
3816 - ``tunnel_id {unsigned}``: L2TPv2 tunnel identifier.
3817 - ``session_id {unsigned}``: L2TPv2 session identifier.
3818 - ``ns {unsigned}``: L2TPv2 option ns.
3819 - ``nr {unsigned}``: L2TPv2 option nr.
3821 - ``ppp``: match PPP header.
3823 - ``addr {unsigned}``: PPP address.
3824 - ``ctrl {unsigned}``: PPP control.
3825 - ``proto_id {unsigned}``: PPP protocol identifier.
3830 A list of actions starts after the ``actions`` token in the same fashion as
3831 `Matching pattern`_; actions are separated by ``/`` tokens and the list is
3832 terminated by a mandatory ``end`` action.
3834 Actions are named after their type (*RTE_FLOW_ACTION_TYPE_* from ``enum
3835 rte_flow_action_type``).
3837 Dropping all incoming UDPv4 packets can be expressed as follows::
3839 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
3842 Several actions have configurable properties which must be specified when
3843 there is no valid default value. For example, ``queue`` requires a target
3846 This rule redirects incoming UDPv4 traffic to queue index 6::
3848 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
3849 actions queue index 6 / end
3851 While this one could be rejected by PMDs (unspecified queue index)::
3853 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
3856 As defined by *rte_flow*, the list is not ordered, all actions of a given
3857 rule are performed simultaneously. These are equivalent::
3859 queue index 6 / void / mark id 42 / end
3863 void / mark id 42 / queue index 6 / end
3865 All actions in a list should have different types, otherwise only the last
3866 action of a given type is taken into account::
3868 queue index 4 / queue index 5 / queue index 6 / end # will use queue 6
3872 drop / drop / drop / end # drop is performed only once
3876 mark id 42 / queue index 3 / mark id 24 / end # mark will be 24
3878 Considering they are performed simultaneously, opposite and overlapping
3879 actions can sometimes be combined when the end result is unambiguous::
3881 drop / queue index 6 / end # drop has no effect
3885 queue index 6 / rss queues 6 7 8 / end # queue has no effect
3889 drop / passthru / end # drop has no effect
3891 Note that PMDs may still refuse such combinations.
3896 This section lists supported actions and their attributes, if any.
3898 - ``end``: end list of actions.
3900 - ``void``: no-op action.
3902 - ``passthru``: let subsequent rule process matched packets.
3904 - ``jump``: redirect traffic to group on device.
3906 - ``group {unsigned}``: group to redirect to.
3908 - ``mark``: attach 32 bit value to packets.
3910 - ``id {unsigned}``: 32 bit value to return with packets.
3912 - ``flag``: flag packets.
3914 - ``queue``: assign packets to a given queue index.
3916 - ``index {unsigned}``: queue index to use.
3918 - ``drop``: drop packets (note: passthru has priority).
3920 - ``count``: enable counters for this rule.
3922 - ``rss``: spread packets among several queues.
3924 - ``func {hash function}``: RSS hash function to apply, allowed tokens are
3925 ``toeplitz``, ``simple_xor``, ``symmetric_toeplitz`` and ``default``.
3927 - ``level {unsigned}``: encapsulation level for ``types``.
3929 - ``types [{RSS hash type} [...]] end``: specific RSS hash types.
3930 Note that an empty list does not disable RSS but instead requests
3931 unspecified "best-effort" settings.
3933 - ``key {string}``: RSS hash key, overrides ``key_len``.
3935 - ``key_len {unsigned}``: RSS hash key length in bytes, can be used in
3936 conjunction with ``key`` to pad or truncate it.
3938 - ``queues [{unsigned} [...]] end``: queue indices to use.
3940 - ``pf``: direct traffic to physical function.
3942 - ``vf``: direct traffic to a virtual function ID.
3944 - ``original {boolean}``: use original VF ID if possible.
3945 - ``id {unsigned}``: VF ID.
3947 - ``phy_port``: direct packets to physical port index.
3949 - ``original {boolean}``: use original port index if possible.
3950 - ``index {unsigned}``: physical port index.
3952 - ``port_id``: direct matching traffic to a given DPDK port ID.
3954 - ``original {boolean}``: use original DPDK port ID if possible.
3955 - ``id {unsigned}``: DPDK port ID.
3957 - ``of_set_mpls_ttl``: OpenFlow's ``OFPAT_SET_MPLS_TTL``.
3959 - ``mpls_ttl``: MPLS TTL.
3961 - ``of_dec_mpls_ttl``: OpenFlow's ``OFPAT_DEC_MPLS_TTL``.
3963 - ``of_set_nw_ttl``: OpenFlow's ``OFPAT_SET_NW_TTL``.
3965 - ``nw_ttl``: IP TTL.
3967 - ``of_dec_nw_ttl``: OpenFlow's ``OFPAT_DEC_NW_TTL``.
3969 - ``of_copy_ttl_out``: OpenFlow's ``OFPAT_COPY_TTL_OUT``.
3971 - ``of_copy_ttl_in``: OpenFlow's ``OFPAT_COPY_TTL_IN``.
3973 - ``of_pop_vlan``: OpenFlow's ``OFPAT_POP_VLAN``.
3975 - ``of_push_vlan``: OpenFlow's ``OFPAT_PUSH_VLAN``.
3977 - ``ethertype``: Ethertype.
3979 - ``of_set_vlan_vid``: OpenFlow's ``OFPAT_SET_VLAN_VID``.
3981 - ``vlan_vid``: VLAN id.
3983 - ``of_set_vlan_pcp``: OpenFlow's ``OFPAT_SET_VLAN_PCP``.
3985 - ``vlan_pcp``: VLAN priority.
3987 - ``of_pop_mpls``: OpenFlow's ``OFPAT_POP_MPLS``.
3989 - ``ethertype``: Ethertype.
3991 - ``of_push_mpls``: OpenFlow's ``OFPAT_PUSH_MPLS``.
3993 - ``ethertype``: Ethertype.
3995 - ``vxlan_encap``: Performs a VXLAN encapsulation, outer layer configuration
3996 is done through `Config VXLAN Encap outer layers`_.
3998 - ``vxlan_decap``: Performs a decapsulation action by stripping all headers of
3999 the VXLAN tunnel network overlay from the matched flow.
4001 - ``nvgre_encap``: Performs a NVGRE encapsulation, outer layer configuration
4002 is done through `Config NVGRE Encap outer layers`_.
4004 - ``nvgre_decap``: Performs a decapsulation action by stripping all headers of
4005 the NVGRE tunnel network overlay from the matched flow.
4007 - ``l2_encap``: Performs a L2 encapsulation, L2 configuration
4008 is done through `Config L2 Encap`_.
4010 - ``l2_decap``: Performs a L2 decapsulation, L2 configuration
4011 is done through `Config L2 Decap`_.
4013 - ``mplsogre_encap``: Performs a MPLSoGRE encapsulation, outer layer
4014 configuration is done through `Config MPLSoGRE Encap outer layers`_.
4016 - ``mplsogre_decap``: Performs a MPLSoGRE decapsulation, outer layer
4017 configuration is done through `Config MPLSoGRE Decap outer layers`_.
4019 - ``mplsoudp_encap``: Performs a MPLSoUDP encapsulation, outer layer
4020 configuration is done through `Config MPLSoUDP Encap outer layers`_.
4022 - ``mplsoudp_decap``: Performs a MPLSoUDP decapsulation, outer layer
4023 configuration is done through `Config MPLSoUDP Decap outer layers`_.
4025 - ``set_ipv4_src``: Set a new IPv4 source address in the outermost IPv4 header.
4027 - ``ipv4_addr``: New IPv4 source address.
4029 - ``set_ipv4_dst``: Set a new IPv4 destination address in the outermost IPv4
4032 - ``ipv4_addr``: New IPv4 destination address.
4034 - ``set_ipv6_src``: Set a new IPv6 source address in the outermost IPv6 header.
4036 - ``ipv6_addr``: New IPv6 source address.
4038 - ``set_ipv6_dst``: Set a new IPv6 destination address in the outermost IPv6
4041 - ``ipv6_addr``: New IPv6 destination address.
4043 - ``set_tp_src``: Set a new source port number in the outermost TCP/UDP
4046 - ``port``: New TCP/UDP source port number.
4048 - ``set_tp_dst``: Set a new destination port number in the outermost TCP/UDP
4051 - ``port``: New TCP/UDP destination port number.
4053 - ``mac_swap``: Swap the source and destination MAC addresses in the outermost
4056 - ``dec_ttl``: Performs a decrease TTL value action
4058 - ``set_ttl``: Set TTL value with specified value
4059 - ``ttl_value {unsigned}``: The new TTL value to be set
4061 - ``set_mac_src``: set source MAC address
4063 - ``mac_addr {MAC-48}``: new source MAC address
4065 - ``set_mac_dst``: set destination MAC address
4067 - ``mac_addr {MAC-48}``: new destination MAC address
4069 - ``inc_tcp_seq``: Increase sequence number in the outermost TCP header.
4071 - ``value {unsigned}``: Value to increase TCP sequence number by.
4073 - ``dec_tcp_seq``: Decrease sequence number in the outermost TCP header.
4075 - ``value {unsigned}``: Value to decrease TCP sequence number by.
4077 - ``inc_tcp_ack``: Increase acknowledgment number in the outermost TCP header.
4079 - ``value {unsigned}``: Value to increase TCP acknowledgment number by.
4081 - ``dec_tcp_ack``: Decrease acknowledgment number in the outermost TCP header.
4083 - ``value {unsigned}``: Value to decrease TCP acknowledgment number by.
4085 - ``set_ipv4_dscp``: Set IPv4 DSCP value with specified value
4087 - ``dscp_value {unsigned}``: The new DSCP value to be set
4089 - ``set_ipv6_dscp``: Set IPv6 DSCP value with specified value
4091 - ``dscp_value {unsigned}``: The new DSCP value to be set
4093 - ``indirect``: Use indirect action created via
4094 ``flow indirect_action {port_id} create``
4096 - ``indirect_action_id {unsigned}``: Indirect action ID to use
4098 - ``color``: Color the packet to reflect the meter color result
4100 - ``type {value}``: Set color type with specified value(green/yellow/red)
4102 - ``port_representor``: at embedded switch level, send matching traffic to
4105 - ``port_id {unsigned}``: ethdev port ID
4107 - ``represented_port``: at embedded switch level, send matching traffic to
4108 the entity represented by the given ethdev
4110 - ``ethdev_port_id {unsigned}``: ethdev port ID
4112 Destroying flow rules
4113 ~~~~~~~~~~~~~~~~~~~~~
4115 ``flow destroy`` destroys one or more rules from their rule ID (as returned
4116 by ``flow create``), this command calls ``rte_flow_destroy()`` as many
4117 times as necessary::
4119 flow destroy {port_id} rule {rule_id} [...]
4121 If successful, it will show::
4123 Flow rule #[...] destroyed
4125 It does not report anything for rule IDs that do not exist. The usual error
4126 message is shown when a rule cannot be destroyed::
4128 Caught error type [...] ([...]): [...]
4130 ``flow flush`` destroys all rules on a device and does not take extra
4131 arguments. It is bound to ``rte_flow_flush()``::
4133 flow flush {port_id}
4135 Any errors are reported as above.
4137 Creating several rules and destroying them::
4139 testpmd> flow create 0 ingress pattern eth / ipv6 / end
4140 actions queue index 2 / end
4141 Flow rule #0 created
4142 testpmd> flow create 0 ingress pattern eth / ipv4 / end
4143 actions queue index 3 / end
4144 Flow rule #1 created
4145 testpmd> flow destroy 0 rule 0 rule 1
4146 Flow rule #1 destroyed
4147 Flow rule #0 destroyed
4150 The same result can be achieved using ``flow flush``::
4152 testpmd> flow create 0 ingress pattern eth / ipv6 / end
4153 actions queue index 2 / end
4154 Flow rule #0 created
4155 testpmd> flow create 0 ingress pattern eth / ipv4 / end
4156 actions queue index 3 / end
4157 Flow rule #1 created
4158 testpmd> flow flush 0
4161 Non-existent rule IDs are ignored::
4163 testpmd> flow create 0 ingress pattern eth / ipv6 / end
4164 actions queue index 2 / end
4165 Flow rule #0 created
4166 testpmd> flow create 0 ingress pattern eth / ipv4 / end
4167 actions queue index 3 / end
4168 Flow rule #1 created
4169 testpmd> flow destroy 0 rule 42 rule 10 rule 2
4171 testpmd> flow destroy 0 rule 0
4172 Flow rule #0 destroyed
4178 ``flow query`` queries a specific action of a flow rule having that
4179 ability. Such actions collect information that can be reported using this
4180 command. It is bound to ``rte_flow_query()``::
4182 flow query {port_id} {rule_id} {action}
4184 If successful, it will display either the retrieved data for known actions
4185 or the following message::
4187 Cannot display result for action type [...] ([...])
4189 Otherwise, it will complain either that the rule does not exist or that some
4192 Flow rule #[...] not found
4196 Caught error type [...] ([...]): [...]
4198 Currently only the ``count`` action is supported. This action reports the
4199 number of packets that hit the flow rule and the total number of bytes. Its
4200 output has the following format::
4203 hits_set: [...] # whether "hits" contains a valid value
4204 bytes_set: [...] # whether "bytes" contains a valid value
4205 hits: [...] # number of packets
4206 bytes: [...] # number of bytes
4208 Querying counters for TCPv6 packets redirected to queue 6::
4210 testpmd> flow create 0 ingress pattern eth / ipv6 / tcp / end
4211 actions queue index 6 / count / end
4212 Flow rule #4 created
4213 testpmd> flow query 0 4 count
4224 ``flow list`` lists existing flow rules sorted by priority and optionally
4225 filtered by group identifiers::
4227 flow list {port_id} [group {group_id}] [...]
4229 This command only fails with the following message if the device does not
4234 Output consists of a header line followed by a short description of each
4235 flow rule, one per line. There is no output at all when no flow rules are
4236 configured on the device::
4238 ID Group Prio Attr Rule
4239 [...] [...] [...] [...] [...]
4241 ``Attr`` column flags:
4243 - ``i`` for ``ingress``.
4244 - ``e`` for ``egress``.
4246 Creating several flow rules and listing them::
4248 testpmd> flow create 0 ingress pattern eth / ipv4 / end
4249 actions queue index 6 / end
4250 Flow rule #0 created
4251 testpmd> flow create 0 ingress pattern eth / ipv6 / end
4252 actions queue index 2 / end
4253 Flow rule #1 created
4254 testpmd> flow create 0 priority 5 ingress pattern eth / ipv4 / udp / end
4255 actions rss queues 6 7 8 end / end
4256 Flow rule #2 created
4257 testpmd> flow list 0
4258 ID Group Prio Attr Rule
4259 0 0 0 i- ETH IPV4 => QUEUE
4260 1 0 0 i- ETH IPV6 => QUEUE
4261 2 0 5 i- ETH IPV4 UDP => RSS
4264 Rules are sorted by priority (i.e. group ID first, then priority level)::
4266 testpmd> flow list 1
4267 ID Group Prio Attr Rule
4268 0 0 0 i- ETH => COUNT
4269 6 0 500 i- ETH IPV6 TCP => DROP COUNT
4270 5 0 1000 i- ETH IPV6 ICMP => QUEUE
4271 1 24 0 i- ETH IPV4 UDP => QUEUE
4272 4 24 10 i- ETH IPV4 TCP => DROP
4273 3 24 20 i- ETH IPV4 => DROP
4274 2 24 42 i- ETH IPV4 UDP => QUEUE
4275 7 63 0 i- ETH IPV6 UDP VXLAN => MARK QUEUE
4278 Output can be limited to specific groups::
4280 testpmd> flow list 1 group 0 group 63
4281 ID Group Prio Attr Rule
4282 0 0 0 i- ETH => COUNT
4283 6 0 500 i- ETH IPV6 TCP => DROP COUNT
4284 5 0 1000 i- ETH IPV6 ICMP => QUEUE
4285 7 63 0 i- ETH IPV6 UDP VXLAN => MARK QUEUE
4288 Toggling isolated mode
4289 ~~~~~~~~~~~~~~~~~~~~~~
4291 ``flow isolate`` can be used to tell the underlying PMD that ingress traffic
4292 must only be injected from the defined flow rules; that no default traffic
4293 is expected outside those rules and the driver is free to assign more
4294 resources to handle them. It is bound to ``rte_flow_isolate()``::
4296 flow isolate {port_id} {boolean}
4298 If successful, enabling or disabling isolated mode shows either::
4300 Ingress traffic on port [...]
4301 is now restricted to the defined flow rules
4305 Ingress traffic on port [...]
4306 is not restricted anymore to the defined flow rules
4308 Otherwise, in case of error::
4310 Caught error type [...] ([...]): [...]
4312 Mainly due to its side effects, PMDs supporting this mode may not have the
4313 ability to toggle it more than once without reinitializing affected ports
4314 first (e.g. by exiting testpmd).
4316 Enabling isolated mode::
4318 testpmd> flow isolate 0 true
4319 Ingress traffic on port 0 is now restricted to the defined flow rules
4322 Disabling isolated mode::
4324 testpmd> flow isolate 0 false
4325 Ingress traffic on port 0 is not restricted anymore to the defined flow rules
4328 Dumping HW internal information
4329 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4331 ``flow dump`` dumps the hardware's internal representation information of
4332 all flows. It is bound to ``rte_flow_dev_dump()``::
4334 flow dump {port_id} {output_file}
4336 If successful, it will show::
4340 Otherwise, it will complain error occurred::
4342 Caught error type [...] ([...]): [...]
4344 Listing and destroying aged flow rules
4345 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4347 ``flow aged`` simply lists aged flow rules be get from api ``rte_flow_get_aged_flows``,
4348 and ``destroy`` parameter can be used to destroy those flow rules in PMD.
4350 flow aged {port_id} [destroy]
4352 Listing current aged flow rules::
4354 testpmd> flow aged 0
4355 Port 0 total aged flows: 0
4356 testpmd> flow create 0 ingress pattern eth / ipv4 src is 2.2.2.14 / end
4357 actions age timeout 5 / queue index 0 / end
4358 Flow rule #0 created
4359 testpmd> flow create 0 ingress pattern eth / ipv4 src is 2.2.2.15 / end
4360 actions age timeout 4 / queue index 0 / end
4361 Flow rule #1 created
4362 testpmd> flow create 0 ingress pattern eth / ipv4 src is 2.2.2.16 / end
4363 actions age timeout 2 / queue index 0 / end
4364 Flow rule #2 created
4365 testpmd> flow create 0 ingress pattern eth / ipv4 src is 2.2.2.17 / end
4366 actions age timeout 3 / queue index 0 / end
4367 Flow rule #3 created
4370 Aged Rules are simply list as command ``flow list {port_id}``, but strip the detail rule
4371 information, all the aged flows are sorted by the longest timeout time. For example, if
4372 those rules be configured in the same time, ID 2 will be the first aged out rule, the next
4373 will be ID 3, ID 1, ID 0::
4375 testpmd> flow aged 0
4376 Port 0 total aged flows: 4
4383 If attach ``destroy`` parameter, the command will destroy all the list aged flow rules.
4385 testpmd> flow aged 0 destroy
4386 Port 0 total aged flows: 4
4393 Flow rule #2 destroyed
4394 Flow rule #3 destroyed
4395 Flow rule #1 destroyed
4396 Flow rule #0 destroyed
4397 4 flows be destroyed
4398 testpmd> flow aged 0
4399 Port 0 total aged flows: 0
4401 Creating indirect actions
4402 ~~~~~~~~~~~~~~~~~~~~~~~~~
4404 ``flow indirect_action {port_id} create`` creates indirect action with optional
4405 indirect action ID. It is bound to ``rte_flow_action_handle_create()``::
4407 flow indirect_action {port_id} create [action_id {indirect_action_id}]
4408 [ingress] [egress] [transfer] action {action} / end
4410 If successful, it will show::
4412 Indirect action #[...] created
4414 Otherwise, it will complain either that indirect action already exists or that
4415 some error occurred::
4417 Indirect action #[...] is already assigned, delete it first
4421 Caught error type [...] ([...]): [...]
4423 Create indirect rss action with id 100 to queues 1 and 2 on port 0::
4425 testpmd> flow indirect_action 0 create action_id 100 \
4426 ingress action rss queues 1 2 end / end
4428 Create indirect rss action with id assigned by testpmd to queues 1 and 2 on
4431 testpmd> flow indirect_action 0 create action_id \
4432 ingress action rss queues 0 1 end / end
4434 Updating indirect actions
4435 ~~~~~~~~~~~~~~~~~~~~~~~~~
4437 ``flow indirect_action {port_id} update`` updates configuration of the indirect
4438 action from its indirect action ID (as returned by
4439 ``flow indirect_action {port_id} create``). It is bound to
4440 ``rte_flow_action_handle_update()``::
4442 flow indirect_action {port_id} update {indirect_action_id}
4443 action {action} / end
4445 If successful, it will show::
4447 Indirect action #[...] updated
4449 Otherwise, it will complain either that indirect action not found or that some
4452 Failed to find indirect action #[...] on port [...]
4456 Caught error type [...] ([...]): [...]
4458 Update indirect rss action having id 100 on port 0 with rss to queues 0 and 3
4459 (in create example above rss queues were 1 and 2)::
4461 testpmd> flow indirect_action 0 update 100 action rss queues 0 3 end / end
4463 Destroying indirect actions
4464 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
4466 ``flow indirect_action {port_id} destroy`` destroys one or more indirect actions
4467 from their indirect action IDs (as returned by
4468 ``flow indirect_action {port_id} create``). It is bound to
4469 ``rte_flow_action_handle_destroy()``::
4471 flow indirect_action {port_id} destroy action_id {indirect_action_id} [...]
4473 If successful, it will show::
4475 Indirect action #[...] destroyed
4477 It does not report anything for indirect action IDs that do not exist.
4478 The usual error message is shown when a indirect action cannot be destroyed::
4480 Caught error type [...] ([...]): [...]
4482 Destroy indirect actions having id 100 & 101::
4484 testpmd> flow indirect_action 0 destroy action_id 100 action_id 101
4486 Query indirect actions
4487 ~~~~~~~~~~~~~~~~~~~~~~
4489 ``flow indirect_action {port_id} query`` queries the indirect action from its
4490 indirect action ID (as returned by ``flow indirect_action {port_id} create``).
4491 It is bound to ``rte_flow_action_handle_query()``::
4493 flow indirect_action {port_id} query {indirect_action_id}
4495 Currently only rss indirect action supported. If successful, it will show::
4497 Indirect RSS action:
4500 Otherwise, it will complain either that indirect action not found or that some
4503 Failed to find indirect action #[...] on port [...]
4507 Caught error type [...] ([...]): [...]
4509 Query indirect action having id 100::
4511 testpmd> flow indirect_action 0 query 100
4513 Sample QinQ flow rules
4514 ~~~~~~~~~~~~~~~~~~~~~~
4516 Before creating QinQ rule(s) the following commands should be issued to enable QinQ::
4518 testpmd> port stop 0
4519 testpmd> vlan set extend on 0
4521 The above command sets the inner and outer TPID's to 0x8100.
4523 To change the TPID's the following commands should be used::
4525 testpmd> vlan set outer tpid 0x88A8 0
4526 testpmd> vlan set inner tpid 0x8100 0
4527 testpmd> port start 0
4529 Validate and create a QinQ rule on port 0 to steer traffic to a VF queue in a VM.
4533 testpmd> flow validate 0 ingress pattern eth / vlan tci is 123 /
4534 vlan tci is 456 / end actions vf id 1 / queue index 0 / end
4535 Flow rule #0 validated
4537 testpmd> flow create 0 ingress pattern eth / vlan tci is 4 /
4538 vlan tci is 456 / end actions vf id 123 / queue index 0 / end
4539 Flow rule #0 created
4541 testpmd> flow list 0
4542 ID Group Prio Attr Rule
4543 0 0 0 i- ETH VLAN VLAN=>VF QUEUE
4545 Validate and create a QinQ rule on port 0 to steer traffic to a queue on the host.
4549 testpmd> flow validate 0 ingress pattern eth / vlan tci is 321 /
4550 vlan tci is 654 / end actions pf / queue index 0 / end
4551 Flow rule #1 validated
4553 testpmd> flow create 0 ingress pattern eth / vlan tci is 321 /
4554 vlan tci is 654 / end actions pf / queue index 1 / end
4555 Flow rule #1 created
4557 testpmd> flow list 0
4558 ID Group Prio Attr Rule
4559 0 0 0 i- ETH VLAN VLAN=>VF QUEUE
4560 1 0 0 i- ETH VLAN VLAN=>PF QUEUE
4562 Sample VXLAN flow rules
4563 ~~~~~~~~~~~~~~~~~~~~~~~
4565 Before creating VXLAN rule(s), the UDP port should be added for VXLAN packet
4568 testpmd> rx_vxlan_port add 4789 0
4570 Create VXLAN rules on port 0 to steer traffic to PF queues.
4574 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / vxlan /
4575 eth dst is 00:11:22:33:44:55 / end actions pf / queue index 1 / end
4576 Flow rule #0 created
4578 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / vxlan vni is 3 /
4579 eth dst is 00:11:22:33:44:55 / end actions pf / queue index 2 / end
4580 Flow rule #1 created
4582 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / vxlan /
4583 eth dst is 00:11:22:33:44:55 / vlan tci is 10 / end actions pf /
4585 Flow rule #2 created
4587 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / vxlan vni is 5 /
4588 eth dst is 00:11:22:33:44:55 / vlan tci is 20 / end actions pf /
4590 Flow rule #3 created
4592 testpmd> flow create 0 ingress pattern eth dst is 00:00:00:00:01:00 / ipv4 /
4593 udp / vxlan vni is 6 / eth dst is 00:11:22:33:44:55 / end actions pf /
4595 Flow rule #4 created
4597 testpmd> flow list 0
4598 ID Group Prio Attr Rule
4599 0 0 0 i- ETH IPV4 UDP VXLAN ETH => QUEUE
4600 1 0 0 i- ETH IPV4 UDP VXLAN ETH => QUEUE
4601 2 0 0 i- ETH IPV4 UDP VXLAN ETH VLAN => QUEUE
4602 3 0 0 i- ETH IPV4 UDP VXLAN ETH VLAN => QUEUE
4603 4 0 0 i- ETH IPV4 UDP VXLAN ETH => QUEUE
4605 Sample VXLAN encapsulation rule
4606 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4608 VXLAN encapsulation outer layer has default value pre-configured in testpmd
4609 source code, those can be changed by using the following commands
4611 IPv4 VXLAN outer header::
4613 testpmd> set vxlan ip-version ipv4 vni 4 udp-src 4 udp-dst 4 ip-src 127.0.0.1
4614 ip-dst 128.0.0.1 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4615 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4618 testpmd> set vxlan-with-vlan ip-version ipv4 vni 4 udp-src 4 udp-dst 4 ip-src
4619 127.0.0.1 ip-dst 128.0.0.1 vlan-tci 34 eth-src 11:11:11:11:11:11
4620 eth-dst 22:22:22:22:22:22
4621 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4624 testpmd> set vxlan-tos-ttl ip-version ipv4 vni 4 udp-src 4 udp-dst 4 ip-tos 0
4625 ip-ttl 255 ip-src 127.0.0.1 ip-dst 128.0.0.1 eth-src 11:11:11:11:11:11
4626 eth-dst 22:22:22:22:22:22
4627 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4630 IPv6 VXLAN outer header::
4632 testpmd> set vxlan ip-version ipv6 vni 4 udp-src 4 udp-dst 4 ip-src ::1
4633 ip-dst ::2222 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4634 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4637 testpmd> set vxlan-with-vlan ip-version ipv6 vni 4 udp-src 4 udp-dst 4
4638 ip-src ::1 ip-dst ::2222 vlan-tci 34 eth-src 11:11:11:11:11:11
4639 eth-dst 22:22:22:22:22:22
4640 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4643 testpmd> set vxlan-tos-ttl ip-version ipv6 vni 4 udp-src 4 udp-dst 4
4644 ip-tos 0 ip-ttl 255 ::1 ip-dst ::2222 eth-src 11:11:11:11:11:11
4645 eth-dst 22:22:22:22:22:22
4646 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4649 Sample NVGRE encapsulation rule
4650 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4652 NVGRE encapsulation outer layer has default value pre-configured in testpmd
4653 source code, those can be changed by using the following commands
4655 IPv4 NVGRE outer header::
4657 testpmd> set nvgre ip-version ipv4 tni 4 ip-src 127.0.0.1 ip-dst 128.0.0.1
4658 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4659 testpmd> flow create 0 ingress pattern end actions nvgre_encap /
4662 testpmd> set nvgre-with-vlan ip-version ipv4 tni 4 ip-src 127.0.0.1
4663 ip-dst 128.0.0.1 vlan-tci 34 eth-src 11:11:11:11:11:11
4664 eth-dst 22:22:22:22:22:22
4665 testpmd> flow create 0 ingress pattern end actions nvgre_encap /
4668 IPv6 NVGRE outer header::
4670 testpmd> set nvgre ip-version ipv6 tni 4 ip-src ::1 ip-dst ::2222
4671 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4672 testpmd> flow create 0 ingress pattern end actions nvgre_encap /
4675 testpmd> set nvgre-with-vlan ip-version ipv6 tni 4 ip-src ::1 ip-dst ::2222
4676 vlan-tci 34 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4677 testpmd> flow create 0 ingress pattern end actions nvgre_encap /
4680 Sample L2 encapsulation rule
4681 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4683 L2 encapsulation has default value pre-configured in testpmd
4684 source code, those can be changed by using the following commands
4688 testpmd> set l2_encap ip-version ipv4
4689 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4690 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / mpls / end actions
4691 mplsoudp_decap / l2_encap / end
4693 L2 with VXLAN header::
4695 testpmd> set l2_encap-with-vlan ip-version ipv4 vlan-tci 34
4696 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4697 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / mpls / end actions
4698 mplsoudp_decap / l2_encap / end
4700 Sample L2 decapsulation rule
4701 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4703 L2 decapsulation has default value pre-configured in testpmd
4704 source code, those can be changed by using the following commands
4708 testpmd> set l2_decap
4709 testpmd> flow create 0 egress pattern eth / end actions l2_decap / mplsoudp_encap /
4712 L2 with VXLAN header::
4714 testpmd> set l2_encap-with-vlan
4715 testpmd> flow create 0 egress pattern eth / end actions l2_encap / mplsoudp_encap /
4718 Sample MPLSoGRE encapsulation rule
4719 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4721 MPLSoGRE encapsulation outer layer has default value pre-configured in testpmd
4722 source code, those can be changed by using the following commands
4724 IPv4 MPLSoGRE outer header::
4726 testpmd> set mplsogre_encap ip-version ipv4 label 4
4727 ip-src 127.0.0.1 ip-dst 128.0.0.1 eth-src 11:11:11:11:11:11
4728 eth-dst 22:22:22:22:22:22
4729 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4730 mplsogre_encap / end
4732 IPv4 MPLSoGRE with VLAN outer header::
4734 testpmd> set mplsogre_encap-with-vlan ip-version ipv4 label 4
4735 ip-src 127.0.0.1 ip-dst 128.0.0.1 vlan-tci 34
4736 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4737 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4738 mplsogre_encap / end
4740 IPv6 MPLSoGRE outer header::
4742 testpmd> set mplsogre_encap ip-version ipv6 mask 4
4743 ip-src ::1 ip-dst ::2222 eth-src 11:11:11:11:11:11
4744 eth-dst 22:22:22:22:22:22
4745 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4746 mplsogre_encap / end
4748 IPv6 MPLSoGRE with VLAN outer header::
4750 testpmd> set mplsogre_encap-with-vlan ip-version ipv6 mask 4
4751 ip-src ::1 ip-dst ::2222 vlan-tci 34
4752 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4753 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4754 mplsogre_encap / end
4756 Sample MPLSoGRE decapsulation rule
4757 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4759 MPLSoGRE decapsulation outer layer has default value pre-configured in testpmd
4760 source code, those can be changed by using the following commands
4762 IPv4 MPLSoGRE outer header::
4764 testpmd> set mplsogre_decap ip-version ipv4
4765 testpmd> flow create 0 ingress pattern eth / ipv4 / gre / mpls / end actions
4766 mplsogre_decap / l2_encap / end
4768 IPv4 MPLSoGRE with VLAN outer header::
4770 testpmd> set mplsogre_decap-with-vlan ip-version ipv4
4771 testpmd> flow create 0 ingress pattern eth / vlan / ipv4 / gre / mpls / end
4772 actions mplsogre_decap / l2_encap / end
4774 IPv6 MPLSoGRE outer header::
4776 testpmd> set mplsogre_decap ip-version ipv6
4777 testpmd> flow create 0 ingress pattern eth / ipv6 / gre / mpls / end
4778 actions mplsogre_decap / l2_encap / end
4780 IPv6 MPLSoGRE with VLAN outer header::
4782 testpmd> set mplsogre_decap-with-vlan ip-version ipv6
4783 testpmd> flow create 0 ingress pattern eth / vlan / ipv6 / gre / mpls / end
4784 actions mplsogre_decap / l2_encap / end
4786 Sample MPLSoUDP encapsulation rule
4787 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4789 MPLSoUDP encapsulation outer layer has default value pre-configured in testpmd
4790 source code, those can be changed by using the following commands
4792 IPv4 MPLSoUDP outer header::
4794 testpmd> set mplsoudp_encap ip-version ipv4 label 4 udp-src 5 udp-dst 10
4795 ip-src 127.0.0.1 ip-dst 128.0.0.1 eth-src 11:11:11:11:11:11
4796 eth-dst 22:22:22:22:22:22
4797 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4798 mplsoudp_encap / end
4800 IPv4 MPLSoUDP with VLAN outer header::
4802 testpmd> set mplsoudp_encap-with-vlan ip-version ipv4 label 4 udp-src 5
4803 udp-dst 10 ip-src 127.0.0.1 ip-dst 128.0.0.1 vlan-tci 34
4804 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4805 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4806 mplsoudp_encap / end
4808 IPv6 MPLSoUDP outer header::
4810 testpmd> set mplsoudp_encap ip-version ipv6 mask 4 udp-src 5 udp-dst 10
4811 ip-src ::1 ip-dst ::2222 eth-src 11:11:11:11:11:11
4812 eth-dst 22:22:22:22:22:22
4813 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4814 mplsoudp_encap / end
4816 IPv6 MPLSoUDP with VLAN outer header::
4818 testpmd> set mplsoudp_encap-with-vlan ip-version ipv6 mask 4 udp-src 5
4819 udp-dst 10 ip-src ::1 ip-dst ::2222 vlan-tci 34
4820 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4821 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4822 mplsoudp_encap / end
4824 Sample MPLSoUDP decapsulation rule
4825 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4827 MPLSoUDP decapsulation outer layer has default value pre-configured in testpmd
4828 source code, those can be changed by using the following commands
4830 IPv4 MPLSoUDP outer header::
4832 testpmd> set mplsoudp_decap ip-version ipv4
4833 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / mpls / end actions
4834 mplsoudp_decap / l2_encap / end
4836 IPv4 MPLSoUDP with VLAN outer header::
4838 testpmd> set mplsoudp_decap-with-vlan ip-version ipv4
4839 testpmd> flow create 0 ingress pattern eth / vlan / ipv4 / udp / mpls / end
4840 actions mplsoudp_decap / l2_encap / end
4842 IPv6 MPLSoUDP outer header::
4844 testpmd> set mplsoudp_decap ip-version ipv6
4845 testpmd> flow create 0 ingress pattern eth / ipv6 / udp / mpls / end
4846 actions mplsoudp_decap / l2_encap / end
4848 IPv6 MPLSoUDP with VLAN outer header::
4850 testpmd> set mplsoudp_decap-with-vlan ip-version ipv6
4851 testpmd> flow create 0 ingress pattern eth / vlan / ipv6 / udp / mpls / end
4852 actions mplsoudp_decap / l2_encap / end
4854 Sample Raw encapsulation rule
4855 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4857 Raw encapsulation configuration can be set by the following commands
4859 Encapsulating VxLAN::
4861 testpmd> set raw_encap 4 eth src is 10:11:22:33:44:55 / vlan tci is 1
4862 inner_type is 0x0800 / ipv4 / udp dst is 4789 / vxlan vni
4864 testpmd> flow create 0 egress pattern eth / ipv4 / end actions
4865 raw_encap index 4 / end
4867 Sample Raw decapsulation rule
4868 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4870 Raw decapsulation configuration can be set by the following commands
4872 Decapsulating VxLAN::
4874 testpmd> set raw_decap eth / ipv4 / udp / vxlan / end_set
4875 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / vxlan / eth / ipv4 /
4876 end actions raw_decap / queue index 0 / end
4881 ESP rules can be created by the following commands::
4883 testpmd> flow create 0 ingress pattern eth / ipv4 / esp spi is 1 / end actions
4885 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / esp spi is 1 / end
4886 actions queue index 3 / end
4887 testpmd> flow create 0 ingress pattern eth / ipv6 / esp spi is 1 / end actions
4889 testpmd> flow create 0 ingress pattern eth / ipv6 / udp / esp spi is 1 / end
4890 actions queue index 3 / end
4895 AH rules can be created by the following commands::
4897 testpmd> flow create 0 ingress pattern eth / ipv4 / ah spi is 1 / end actions
4899 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / ah spi is 1 / end
4900 actions queue index 3 / end
4901 testpmd> flow create 0 ingress pattern eth / ipv6 / ah spi is 1 / end actions
4903 testpmd> flow create 0 ingress pattern eth / ipv6 / udp / ah spi is 1 / end
4904 actions queue index 3 / end
4909 PFCP rules can be created by the following commands(s_field need to be 1
4912 testpmd> flow create 0 ingress pattern eth / ipv4 / pfcp s_field is 0 / end
4913 actions queue index 3 / end
4914 testpmd> flow create 0 ingress pattern eth / ipv4 / pfcp s_field is 1
4915 seid is 1 / end actions queue index 3 / end
4916 testpmd> flow create 0 ingress pattern eth / ipv6 / pfcp s_field is 0 / end
4917 actions queue index 3 / end
4918 testpmd> flow create 0 ingress pattern eth / ipv6 / pfcp s_field is 1
4919 seid is 1 / end actions queue index 3 / end
4921 Sample Sampling/Mirroring rules
4922 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4924 Sample/Mirroring rules can be set by the following commands
4926 NIC-RX Sampling rule, the matched ingress packets and sent to the queue 1,
4927 and 50% packets are duplicated and marked with 0x1234 and sent to queue 0.
4931 testpmd> set sample_actions 0 mark id 0x1234 / queue index 0 / end
4932 testpmd> flow create 0 ingress group 1 pattern eth / end actions
4933 sample ratio 2 index 0 / queue index 1 / end
4935 Mirroring rule with port representors (with "transfer" attribute), the matched
4936 ingress packets with encapsulation header are sent to port id 0, and also
4937 mirrored the packets and sent to port id 2.
4941 testpmd> set sample_actions 0 port_id id 2 / end
4942 testpmd> flow create 1 ingress transfer pattern eth / end actions
4943 sample ratio 1 index 0 / raw_encap / port_id id 0 / end
4945 Mirroring rule with port representors (with "transfer" attribute), the matched
4946 ingress packets are sent to port id 2, and also mirrored the packets with
4947 encapsulation header and sent to port id 0.
4951 testpmd> set sample_actions 0 raw_encap / port_id id 0 / end
4952 testpmd> flow create 0 ingress transfer pattern eth / end actions
4953 sample ratio 1 index 0 / port_id id 2 / end
4955 Mirroring rule with port representors (with "transfer" attribute), the matched
4956 ingress packets are sent to port id 2, and also mirrored the packets with
4957 VXLAN encapsulation header and sent to port id 0.
4961 testpmd> set vxlan ip-version ipv4 vni 4 udp-src 4 udp-dst 4 ip-src 127.0.0.1
4962 ip-dst 128.0.0.1 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4963 testpmd> set sample_actions 0 vxlan_encap / port_id id 0 / end
4964 testpmd> flow create 0 ingress transfer pattern eth / end actions
4965 sample ratio 1 index 0 / port_id id 2 / end
4967 Mirroring rule with port representors (with "transfer" attribute), the matched
4968 ingress packets are sent to port id 2, and also mirrored the packets with
4969 NVGRE encapsulation header and sent to port id 0.
4973 testpmd> set nvgre ip-version ipv4 tni 4 ip-src 127.0.0.1 ip-dst 128.0.0.1
4974 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4975 testpmd> set sample_actions 0 nvgre_encap / port_id id 0 / end
4976 testpmd> flow create 0 ingress transfer pattern eth / end actions
4977 sample ratio 1 index 0 / port_id id 2 / end
4979 Sample integrity rules
4980 ~~~~~~~~~~~~~~~~~~~~~~
4982 Integrity rules can be created by the following commands:
4984 Integrity rule that forwards valid TCP packets to group 1.
4985 TCP packet integrity is matched with the ``l4_ok`` bit 3.
4989 testpmd> flow create 0 ingress
4990 pattern eth / ipv4 / tcp / integrity value mask 8 value spec 8 / end
4991 actions jump group 1 / end
4993 Integrity rule that forwards invalid packets to application.
4994 General packet integrity is matched with the ``packet_ok`` bit 0.
4998 testpmd> flow create 0 ingress pattern integrity value mask 1 value spec 0 / end actions queue index 0 / end
5000 Sample conntrack rules
5001 ~~~~~~~~~~~~~~~~~~~~~~
5003 Conntrack rules can be set by the following commands
5005 Need to construct the connection context with provided information.
5006 In the first table, create a flow rule by using conntrack action and jump to
5007 the next table. In the next table, create a rule to check the state.
5011 testpmd> set conntrack com peer 1 is_orig 1 enable 1 live 1 sack 1 cack 0
5012 last_dir 0 liberal 0 state 1 max_ack_win 7 r_lim 5 last_win 510
5013 last_seq 2632987379 last_ack 2532480967 last_end 2632987379
5015 testpmd> set conntrack orig scale 7 fin 0 acked 1 unack_data 0
5016 sent_end 2632987379 reply_end 2633016339 max_win 28960
5018 testpmd> set conntrack rply scale 7 fin 0 acked 1 unack_data 0
5019 sent_end 2532480967 reply_end 2532546247 max_win 65280
5021 testpmd> flow indirect_action 0 create ingress action conntrack / end
5022 testpmd> flow create 0 group 3 ingress pattern eth / ipv4 / tcp / end actions indirect 0 / jump group 5 / end
5023 testpmd> flow create 0 group 5 ingress pattern eth / ipv4 / tcp / conntrack is 1 / end actions queue index 5 / end
5025 Construct the conntrack again with only "is_orig" set to 0 (other fields are
5026 ignored), then use "update" interface to update the direction. Create flow
5027 rules like above for the peer port.
5031 testpmd> flow indirect_action 0 update 0 action conntrack_update dir / end
5033 Sample meter with policy rules
5034 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
5036 Meter with policy rules can be created by the following commands:
5038 Need to create policy first and actions are set for green/yellow/red colors.
5039 Create meter with policy id. Create flow with meter id.
5041 Example for policy with meter color action. The purpose is to color the packet
5042 to reflect the meter color result.
5043 The meter policy action list: ``green -> green, yellow -> yellow, red -> red``.
5047 testpmd> add port meter profile srtcm_rfc2697 0 13 21504 2688 0 0
5048 testpmd> add port meter policy 0 1 g_actions color type green / end y_actions color type yellow / end
5049 r_actions color type red / end
5050 testpmd> create port meter 0 1 13 1 yes 0xffff 0 0
5051 testpmd> flow create 0 priority 0 ingress group 1 pattern eth / end actions meter mtr_id 1 / end
5053 Sample PPPoL2TPv2oUDP RSS rules
5054 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
5056 PPPoL2TPv2oUDP RSS rules can be created by the following commands::
5058 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / l2tpv2 / ppp / ipv4
5059 / end actions rss types ipv4 end queues end / end
5060 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / l2tpv2 / ppp / ipv6
5061 / udp / end actions rss types ipv6-udp end queues end / end
5062 testpmd> flow create 0 ingress pattern eth / ipv6 / udp / l2tpv2 / ppp / ipv4
5063 / tcp / end actions rss types ipv4-tcp end queues end / end
5064 testpmd> flow create 0 ingress pattern eth / ipv6 / udp / l2tpv2 / ppp / ipv6
5065 / end actions rss types ipv6 end queues end / end
5070 The following sections show functions to load/unload eBPF based filters.
5075 Load an eBPF program as a callback for particular RX/TX queue::
5077 testpmd> bpf-load rx|tx (portid) (queueid) (load-flags) (bpf-prog-filename)
5079 The available load-flags are:
5081 * ``J``: use JIT generated native code, otherwise BPF interpreter will be used.
5083 * ``M``: assume input parameter is a pointer to rte_mbuf, otherwise assume it is a pointer to first segment's data.
5089 You'll need clang v3.7 or above to build bpf program you'd like to load
5093 .. code-block:: console
5096 clang -O2 -target bpf -c t1.c
5098 Then to load (and JIT compile) t1.o at RX queue 0, port 1:
5100 .. code-block:: console
5102 testpmd> bpf-load rx 1 0 J ./dpdk.org/examples/bpf/t1.o
5104 To load (not JITed) t1.o at TX queue 0, port 0:
5106 .. code-block:: console
5108 testpmd> bpf-load tx 0 0 - ./dpdk.org/examples/bpf/t1.o
5113 Unload previously loaded eBPF program for particular RX/TX queue::
5115 testpmd> bpf-unload rx|tx (portid) (queueid)
5117 For example to unload BPF filter from TX queue 0, port 0:
5119 .. code-block:: console
5121 testpmd> bpf-unload tx 0 0
5126 The following sections show functions that configure and create flex item object,
5127 create flex pattern and use it in a flow rule.
5128 The commands will use 20 bytes IPv4 header for examples:
5133 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
5134 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
5135 | ver | IHL | TOS | length | +0
5136 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
5137 | identification | flg | frag. offset | +4
5138 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
5139 | TTL | protocol | checksum | +8
5140 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
5141 | source IP address | +12
5142 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
5143 | destination IP address | +16
5144 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
5150 Flex item object is created by PMD according to a new header configuration. The
5151 header configuration is compiled by the testpmd and stored in
5152 ``rte_flow_item_flex_conf`` type variable.
5156 # flow flex_item create <port> <flex id> <configuration file>
5157 testpmd> flow flex_item init 0 3 ipv4_flex_config.json
5158 port-0: created flex item #3
5160 Flex item configuration is kept in external JSON file.
5161 It describes the following header elements:
5163 **New header length.**
5165 Specify whether the new header has fixed or variable length and the basic/minimal
5166 header length value.
5168 If header length is not fixed, header location with a value that completes header
5169 length calculation and scale/offset function must be added.
5171 Scale function depends on port hardware.
5175 Describes location in the new header that specify following network header type.
5177 **Flow match samples.**
5179 Describes locations in the new header that will be used in flow rules.
5181 Number of flow samples and sample maximal length depend of port hardware.
5185 Describes preceding network header configuration.
5189 Describes conditions that trigger transfer to following network header
5191 .. code-block:: json
5194 "next_header": { "field_mode": "FIELD_MODE_FIXED", "field_size": 20},
5195 "next_protocol": {"field_size": 8, "field_base": 72},
5197 { "field_mode": "FIELD_MODE_FIXED", "field_size": 32, "field_base": 0},
5198 { "field_mode": "FIELD_MODE_FIXED", "field_size": 32, "field_base": 32},
5199 { "field_mode": "FIELD_MODE_FIXED", "field_size": 32, "field_base": 64},
5200 { "field_mode": "FIELD_MODE_FIXED", "field_size": 32, "field_base": 96}
5203 {"item": "eth type is 0x0800"},
5204 {"item": "vlan inner_type is 0x0800"}
5207 {"item": "udp", "next": 17},
5208 {"item": "tcp", "next": 6},
5209 {"item": "icmp", "next": 1}
5214 Flex pattern and flow rules
5215 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
5217 Flex pattern describe parts of network header that will trigger flex flow item hit in a flow rule.
5218 Flex pattern directly related to flex item samples configuration.
5219 Flex pattern can be shared between ports.
5221 **Flex pattern and flow rule to match IPv4 version and 20 bytes length**
5225 # set flex_pattern <pattern_id> is <hex bytes sequence>
5226 testpmd> flow flex_item pattern 5 is 45FF
5229 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / flex item is 3 pattern is 5 / end actions mark id 1 / queue index 0 / end
5230 Flow rule #0 created
5232 **Flex pattern and flow rule to match packets with source address 1.2.3.4**
5236 testpmd> flow flex_item pattern 2 spec 45000000000000000000000001020304 mask FF0000000000000000000000FFFFFFFF
5239 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / flex item is 3 pattern is 2 / end actions mark id 1 / queue index 0 / end
5240 Flow rule #0 created