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 Set the priority flow control parameter on a port::
1532 testpmd> set pfc_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \
1533 (pause_time) (priority) (port_id)
1537 * ``high_water`` (integer): High threshold value.
1539 * ``low_water`` (integer): Low threshold value.
1541 * ``pause_time`` (integer): Pause quota in the Pause frame.
1543 * ``priority`` (0-7): VLAN User Priority.
1548 Set statistics mapping (qmapping 0..15) for RX/TX queue on port::
1550 testpmd> set stat_qmap (tx|rx) (port_id) (queue_id) (qmapping)
1552 For example, to set rx queue 2 on port 0 to mapping 5::
1554 testpmd>set stat_qmap rx 0 2 5
1556 set xstats-hide-zero
1557 ~~~~~~~~~~~~~~~~~~~~
1559 Set the option to hide zero values for xstats display::
1561 testpmd> set xstats-hide-zero on|off
1565 By default, the zero values are displayed for xstats.
1567 set port - rx/tx (for VF)
1568 ~~~~~~~~~~~~~~~~~~~~~~~~~
1570 Set VF receive/transmit from a port::
1572 testpmd> set port (port_id) vf (vf_id) (rx|tx) (on|off)
1574 set port - rx mode(for VF)
1575 ~~~~~~~~~~~~~~~~~~~~~~~~~~
1577 Set the VF receive mode of a port::
1579 testpmd> set port (port_id) vf (vf_id) \
1580 rxmode (AUPE|ROPE|BAM|MPE) (on|off)
1582 The available receive modes are:
1584 * ``AUPE``: Accepts untagged VLAN.
1586 * ``ROPE``: Accepts unicast hash.
1588 * ``BAM``: Accepts broadcast packets.
1590 * ``MPE``: Accepts all multicast packets.
1592 set port - tx_rate (for Queue)
1593 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1595 Set TX rate limitation for a queue on a port::
1597 testpmd> set port (port_id) queue (queue_id) rate (rate_value)
1599 set port - tx_rate (for VF)
1600 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
1602 Set TX rate limitation for queues in VF on a port::
1604 testpmd> set port (port_id) vf (vf_id) rate (rate_value) queue_mask (queue_mask)
1606 set port - mirror rule
1607 ~~~~~~~~~~~~~~~~~~~~~~
1609 Set pool or vlan type mirror rule for a port::
1611 testpmd> set port (port_id) mirror-rule (rule_id) \
1612 (pool-mirror-up|pool-mirror-down|vlan-mirror) \
1613 (poolmask|vlanid[,vlanid]*) dst-pool (pool_id) (on|off)
1615 Set link mirror rule for a port::
1617 testpmd> set port (port_id) mirror-rule (rule_id) \
1618 (uplink-mirror|downlink-mirror) dst-pool (pool_id) (on|off)
1620 For example to enable mirror traffic with vlan 0,1 to pool 0::
1622 set port 0 mirror-rule 0 vlan-mirror 0,1 dst-pool 0 on
1624 reset port - mirror rule
1625 ~~~~~~~~~~~~~~~~~~~~~~~~
1627 Reset a mirror rule for a port::
1629 testpmd> reset port (port_id) mirror-rule (rule_id)
1634 Set the flush on RX streams before forwarding.
1635 The default is flush ``on``.
1636 Mainly used with PCAP drivers to turn off the default behavior of flushing the first 512 packets on RX streams::
1638 testpmd> set flush_rx off
1643 Set the bypass mode for the lowest port on bypass enabled NIC::
1645 testpmd> set bypass mode (normal|bypass|isolate) (port_id)
1650 Set the event required to initiate specified bypass mode for the lowest port on a bypass enabled::
1652 testpmd> set bypass event (timeout|os_on|os_off|power_on|power_off) \
1653 mode (normal|bypass|isolate) (port_id)
1657 * ``timeout``: Enable bypass after watchdog timeout.
1659 * ``os_on``: Enable bypass when OS/board is powered on.
1661 * ``os_off``: Enable bypass when OS/board is powered off.
1663 * ``power_on``: Enable bypass when power supply is turned on.
1665 * ``power_off``: Enable bypass when power supply is turned off.
1671 Set the bypass watchdog timeout to ``n`` seconds where 0 = instant::
1673 testpmd> set bypass timeout (0|1.5|2|3|4|8|16|32)
1678 Show the bypass configuration for a bypass enabled NIC using the lowest port on the NIC::
1680 testpmd> show bypass config (port_id)
1685 Set link up for a port::
1687 testpmd> set link-up port (port id)
1692 Set link down for a port::
1694 testpmd> set link-down port (port id)
1699 Enable E-tag insertion for a VF on a port::
1701 testpmd> E-tag set insertion on port-tag-id (value) port (port_id) vf (vf_id)
1703 Disable E-tag insertion for a VF on a port::
1705 testpmd> E-tag set insertion off port (port_id) vf (vf_id)
1707 Enable/disable E-tag stripping on a port::
1709 testpmd> E-tag set stripping (on|off) port (port_id)
1711 Enable/disable E-tag based forwarding on a port::
1713 testpmd> E-tag set forwarding (on|off) port (port_id)
1718 Load a dynamic device personalization (DDP) profile and store backup profile::
1720 testpmd> ddp add (port_id) (profile_path[,backup_profile_path])
1725 Delete a dynamic device personalization profile and restore backup profile::
1727 testpmd> ddp del (port_id) (backup_profile_path)
1732 List all items from the ptype mapping table::
1734 testpmd> ptype mapping get (port_id) (valid_only)
1738 * ``valid_only``: A flag indicates if only list valid items(=1) or all itemss(=0).
1740 Replace a specific or a group of software defined ptype with a new one::
1742 testpmd> ptype mapping replace (port_id) (target) (mask) (pkt_type)
1746 * ``target``: A specific software ptype or a mask to represent a group of software ptypes.
1748 * ``mask``: A flag indicate if "target" is a specific software ptype(=0) or a ptype mask(=1).
1750 * ``pkt_type``: The new software ptype to replace the old ones.
1752 Update hardware defined ptype to software defined packet type mapping table::
1754 testpmd> ptype mapping update (port_id) (hw_ptype) (sw_ptype)
1758 * ``hw_ptype``: hardware ptype as the index of the ptype mapping table.
1760 * ``sw_ptype``: software ptype as the value of the ptype mapping table.
1762 Reset ptype mapping table::
1764 testpmd> ptype mapping reset (port_id)
1766 config per port Rx offloading
1767 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1769 Enable or disable a per port Rx offloading on all Rx queues of a port::
1771 testpmd> port config (port_id) rx_offload (offloading) on|off
1773 * ``offloading``: can be any of these offloading capability:
1774 vlan_strip, ipv4_cksum, udp_cksum, tcp_cksum, tcp_lro,
1775 qinq_strip, outer_ipv4_cksum, macsec_strip,
1776 header_split, vlan_filter, vlan_extend, jumbo_frame,
1777 scatter, timestamp, security, keep_crc, rss_hash
1779 This command should be run when the port is stopped, or else it will fail.
1781 config per queue Rx offloading
1782 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1784 Enable or disable a per queue Rx offloading only on a specific Rx queue::
1786 testpmd> port (port_id) rxq (queue_id) rx_offload (offloading) on|off
1788 * ``offloading``: can be any of these offloading capability:
1789 vlan_strip, ipv4_cksum, udp_cksum, tcp_cksum, tcp_lro,
1790 qinq_strip, outer_ipv4_cksum, macsec_strip,
1791 header_split, vlan_filter, vlan_extend, jumbo_frame,
1792 scatter, timestamp, security, keep_crc
1794 This command should be run when the port is stopped, or else it will fail.
1796 config per port Tx offloading
1797 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1799 Enable or disable a per port Tx offloading on all Tx queues of a port::
1801 testpmd> port config (port_id) tx_offload (offloading) on|off
1803 * ``offloading``: can be any of these offloading capability:
1804 vlan_insert, ipv4_cksum, udp_cksum, tcp_cksum,
1805 sctp_cksum, tcp_tso, udp_tso, outer_ipv4_cksum,
1806 qinq_insert, vxlan_tnl_tso, gre_tnl_tso,
1807 ipip_tnl_tso, geneve_tnl_tso, macsec_insert,
1808 mt_lockfree, multi_segs, mbuf_fast_free, security
1810 This command should be run when the port is stopped, or else it will fail.
1812 config per queue Tx offloading
1813 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1815 Enable or disable a per queue Tx offloading only on a specific Tx queue::
1817 testpmd> port (port_id) txq (queue_id) tx_offload (offloading) on|off
1819 * ``offloading``: can be any of these offloading capability:
1820 vlan_insert, ipv4_cksum, udp_cksum, tcp_cksum,
1821 sctp_cksum, tcp_tso, udp_tso, outer_ipv4_cksum,
1822 qinq_insert, vxlan_tnl_tso, gre_tnl_tso,
1823 ipip_tnl_tso, geneve_tnl_tso, macsec_insert,
1824 mt_lockfree, multi_segs, mbuf_fast_free, security
1826 This command should be run when the port is stopped, or else it will fail.
1828 Config VXLAN Encap outer layers
1829 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1831 Configure the outer layer to encapsulate a packet inside a VXLAN tunnel::
1833 set vxlan ip-version (ipv4|ipv6) vni (vni) udp-src (udp-src) \
1834 udp-dst (udp-dst) ip-src (ip-src) ip-dst (ip-dst) eth-src (eth-src) \
1837 set vxlan-with-vlan ip-version (ipv4|ipv6) vni (vni) udp-src (udp-src) \
1838 udp-dst (udp-dst) ip-src (ip-src) ip-dst (ip-dst) vlan-tci (vlan-tci) \
1839 eth-src (eth-src) eth-dst (eth-dst)
1841 set vxlan-tos-ttl ip-version (ipv4|ipv6) vni (vni) udp-src (udp-src) \
1842 udp-dst (udp-dst) ip-tos (ip-tos) ip-ttl (ip-ttl) ip-src (ip-src) \
1843 ip-dst (ip-dst) eth-src (eth-src) eth-dst (eth-dst)
1845 These commands will set an internal configuration inside testpmd, any following
1846 flow rule using the action vxlan_encap will use the last configuration set.
1847 To have a different encapsulation header, one of those commands must be called
1848 before the flow rule creation.
1850 Config NVGRE Encap outer layers
1851 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1853 Configure the outer layer to encapsulate a packet inside a NVGRE tunnel::
1855 set nvgre ip-version (ipv4|ipv6) tni (tni) ip-src (ip-src) ip-dst (ip-dst) \
1856 eth-src (eth-src) eth-dst (eth-dst)
1857 set nvgre-with-vlan ip-version (ipv4|ipv6) tni (tni) ip-src (ip-src) \
1858 ip-dst (ip-dst) vlan-tci (vlan-tci) eth-src (eth-src) eth-dst (eth-dst)
1860 These commands will set an internal configuration inside testpmd, any following
1861 flow rule using the action nvgre_encap will use the last configuration set.
1862 To have a different encapsulation header, one of those commands must be called
1863 before the flow rule creation.
1868 Configure the l2 to be used when encapsulating a packet with L2::
1870 set l2_encap ip-version (ipv4|ipv6) eth-src (eth-src) eth-dst (eth-dst)
1871 set l2_encap-with-vlan ip-version (ipv4|ipv6) vlan-tci (vlan-tci) \
1872 eth-src (eth-src) eth-dst (eth-dst)
1874 Those commands will set an internal configuration inside testpmd, any following
1875 flow rule using the action l2_encap will use the last configuration set.
1876 To have a different encapsulation header, one of those commands must be called
1877 before the flow rule creation.
1882 Configure the l2 to be removed when decapsulating a packet with L2::
1884 set l2_decap ip-version (ipv4|ipv6)
1885 set l2_decap-with-vlan ip-version (ipv4|ipv6)
1887 Those commands will set an internal configuration inside testpmd, any following
1888 flow rule using the action l2_decap will use the last configuration set.
1889 To have a different encapsulation header, one of those commands must be called
1890 before the flow rule creation.
1892 Config MPLSoGRE Encap outer layers
1893 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1895 Configure the outer layer to encapsulate a packet inside a MPLSoGRE tunnel::
1897 set mplsogre_encap ip-version (ipv4|ipv6) label (label) \
1898 ip-src (ip-src) ip-dst (ip-dst) eth-src (eth-src) eth-dst (eth-dst)
1899 set mplsogre_encap-with-vlan ip-version (ipv4|ipv6) label (label) \
1900 ip-src (ip-src) ip-dst (ip-dst) vlan-tci (vlan-tci) \
1901 eth-src (eth-src) eth-dst (eth-dst)
1903 These commands will set an internal configuration inside testpmd, any following
1904 flow rule using the action mplsogre_encap will use the last configuration set.
1905 To have a different encapsulation header, one of those commands must be called
1906 before the flow rule creation.
1908 Config MPLSoGRE Decap outer layers
1909 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1911 Configure the outer layer to decapsulate MPLSoGRE packet::
1913 set mplsogre_decap ip-version (ipv4|ipv6)
1914 set mplsogre_decap-with-vlan ip-version (ipv4|ipv6)
1916 These commands will set an internal configuration inside testpmd, any following
1917 flow rule using the action mplsogre_decap will use the last configuration set.
1918 To have a different decapsulation header, one of those commands must be called
1919 before the flow rule creation.
1921 Config MPLSoUDP Encap outer layers
1922 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1924 Configure the outer layer to encapsulate a packet inside a MPLSoUDP tunnel::
1926 set mplsoudp_encap ip-version (ipv4|ipv6) label (label) udp-src (udp-src) \
1927 udp-dst (udp-dst) ip-src (ip-src) ip-dst (ip-dst) \
1928 eth-src (eth-src) eth-dst (eth-dst)
1929 set mplsoudp_encap-with-vlan ip-version (ipv4|ipv6) label (label) \
1930 udp-src (udp-src) udp-dst (udp-dst) ip-src (ip-src) ip-dst (ip-dst) \
1931 vlan-tci (vlan-tci) eth-src (eth-src) eth-dst (eth-dst)
1933 These commands will set an internal configuration inside testpmd, any following
1934 flow rule using the action mplsoudp_encap will use the last configuration set.
1935 To have a different encapsulation header, one of those commands must be called
1936 before the flow rule creation.
1938 Config MPLSoUDP Decap outer layers
1939 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1941 Configure the outer layer to decapsulate MPLSoUDP packet::
1943 set mplsoudp_decap ip-version (ipv4|ipv6)
1944 set mplsoudp_decap-with-vlan ip-version (ipv4|ipv6)
1946 These commands will set an internal configuration inside testpmd, any following
1947 flow rule using the action mplsoudp_decap will use the last configuration set.
1948 To have a different decapsulation header, one of those commands must be called
1949 before the flow rule creation.
1951 Config Raw Encapsulation
1952 ~~~~~~~~~~~~~~~~~~~~~~~~~
1954 Configure the raw data to be used when encapsulating a packet by
1955 rte_flow_action_raw_encap::
1957 set raw_encap {index} {item} [/ {item} [...]] / end_set
1959 There are multiple global buffers for ``raw_encap``, this command will set one
1960 internal buffer index by ``{index}``.
1961 If there is no ``{index}`` specified::
1963 set raw_encap {item} [/ {item} [...]] / end_set
1965 the default index ``0`` is used.
1966 In order to use different encapsulating header, ``index`` must be specified
1967 during the flow rule creation::
1969 testpmd> flow create 0 egress pattern eth / ipv4 / end actions
1970 raw_encap index 2 / end
1972 Otherwise the default index ``0`` is used.
1974 Config Raw Decapsulation
1975 ~~~~~~~~~~~~~~~~~~~~~~~~
1977 Configure the raw data to be used when decapsulating a packet by
1978 rte_flow_action_raw_decap::
1980 set raw_decap {index} {item} [/ {item} [...]] / end_set
1982 There are multiple global buffers for ``raw_decap``, this command will set
1983 one internal buffer index by ``{index}``.
1984 If there is no ``{index}`` specified::
1986 set raw_decap {item} [/ {item} [...]] / end_set
1988 the default index ``0`` is used.
1989 In order to use different decapsulating header, ``index`` must be specified
1990 during the flow rule creation::
1992 testpmd> flow create 0 egress pattern eth / ipv4 / end actions
1993 raw_encap index 3 / end
1995 Otherwise the default index ``0`` is used.
2000 Set fec mode for a specific port::
2002 testpmd> set port (port_id) fec_mode auto|off|rs|baser
2004 Config Sample actions list
2005 ~~~~~~~~~~~~~~~~~~~~~~~~~~
2007 Configure the sample actions list to be used when sampling a packet by
2008 rte_flow_action_sample::
2010 set sample_actions {index} {action} [/ {action} [...]] / end
2012 There are multiple global buffers for ``sample_actions``, this command will set
2013 one internal buffer index by ``{index}``.
2015 In order to use different sample actions list, ``index`` must be specified
2016 during the flow rule creation::
2018 testpmd> flow create 0 ingress pattern eth / ipv4 / end actions
2019 sample ratio 2 index 2 / end
2021 Otherwise the default index ``0`` is used.
2026 The following sections show functions for configuring ports.
2030 Port configuration changes only become active when forwarding is started/restarted.
2035 Attach a port specified by pci address or virtual device args::
2037 testpmd> port attach (identifier)
2039 To attach a new pci device, the device should be recognized by kernel first.
2040 Then it should be moved under DPDK management.
2041 Finally the port can be attached to testpmd.
2043 For example, to move a pci device using ixgbe under DPDK management:
2045 .. code-block:: console
2047 # Check the status of the available devices.
2048 ./usertools/dpdk-devbind.py --status
2050 Network devices using DPDK-compatible driver
2051 ============================================
2054 Network devices using kernel driver
2055 ===================================
2056 0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused=
2059 # Bind the device to igb_uio.
2060 sudo ./usertools/dpdk-devbind.py -b igb_uio 0000:0a:00.0
2063 # Recheck the status of the devices.
2064 ./usertools/dpdk-devbind.py --status
2065 Network devices using DPDK-compatible driver
2066 ============================================
2067 0000:0a:00.0 '82599ES 10-Gigabit' drv=igb_uio unused=
2069 To attach a port created by virtual device, above steps are not needed.
2071 For example, to attach a port whose pci address is 0000:0a:00.0.
2073 .. code-block:: console
2075 testpmd> port attach 0000:0a:00.0
2076 Attaching a new port...
2077 EAL: PCI device 0000:0a:00.0 on NUMA socket -1
2078 EAL: probe driver: 8086:10fb rte_ixgbe_pmd
2079 EAL: PCI memory mapped at 0x7f83bfa00000
2080 EAL: PCI memory mapped at 0x7f83bfa80000
2081 PMD: eth_ixgbe_dev_init(): MAC: 2, PHY: 18, SFP+: 5
2082 PMD: eth_ixgbe_dev_init(): port 0 vendorID=0x8086 deviceID=0x10fb
2083 Port 0 is attached. Now total ports is 1
2086 For example, to attach a port created by pcap PMD.
2088 .. code-block:: console
2090 testpmd> port attach net_pcap0
2091 Attaching a new port...
2092 PMD: Initializing pmd_pcap for net_pcap0
2093 PMD: Creating pcap-backed ethdev on numa socket 0
2094 Port 0 is attached. Now total ports is 1
2097 In this case, identifier is ``net_pcap0``.
2098 This identifier format is the same as ``--vdev`` format of DPDK applications.
2100 For example, to re-attach a bonded port which has been previously detached,
2101 the mode and slave parameters must be given.
2103 .. code-block:: console
2105 testpmd> port attach net_bond_0,mode=0,slave=1
2106 Attaching a new port...
2107 EAL: Initializing pmd_bond for net_bond_0
2108 EAL: Create bonded device net_bond_0 on port 0 in mode 0 on socket 0.
2109 Port 0 is attached. Now total ports is 1
2116 Detach a specific port::
2118 testpmd> port detach (port_id)
2120 Before detaching a port, the port should be stopped and closed.
2122 For example, to detach a pci device port 0.
2124 .. code-block:: console
2126 testpmd> port stop 0
2129 testpmd> port close 0
2133 testpmd> port detach 0
2135 EAL: PCI device 0000:0a:00.0 on NUMA socket -1
2136 EAL: remove driver: 8086:10fb rte_ixgbe_pmd
2137 EAL: PCI memory unmapped at 0x7f83bfa00000
2138 EAL: PCI memory unmapped at 0x7f83bfa80000
2142 For example, to detach a virtual device port 0.
2144 .. code-block:: console
2146 testpmd> port stop 0
2149 testpmd> port close 0
2153 testpmd> port detach 0
2155 PMD: Closing pcap ethdev on numa socket 0
2156 Port 'net_pcap0' is detached. Now total ports is 0
2159 To remove a pci device completely from the system, first detach the port from testpmd.
2160 Then the device should be moved under kernel management.
2161 Finally the device can be removed using kernel pci hotplug functionality.
2163 For example, to move a pci device under kernel management:
2165 .. code-block:: console
2167 sudo ./usertools/dpdk-devbind.py -b ixgbe 0000:0a:00.0
2169 ./usertools/dpdk-devbind.py --status
2171 Network devices using DPDK-compatible driver
2172 ============================================
2175 Network devices using kernel driver
2176 ===================================
2177 0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused=igb_uio
2179 To remove a port created by a virtual device, above steps are not needed.
2184 Start all ports or a specific port::
2186 testpmd> port start (port_id|all)
2191 Stop all ports or a specific port::
2193 testpmd> port stop (port_id|all)
2198 Close all ports or a specific port::
2200 testpmd> port close (port_id|all)
2205 Reset all ports or a specific port::
2207 testpmd> port reset (port_id|all)
2209 User should stop port(s) before resetting and (re-)start after reset.
2211 port config - queue ring size
2212 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2214 Configure a rx/tx queue ring size::
2216 testpmd> port (port_id) (rxq|txq) (queue_id) ring_size (value)
2218 Only take effect after command that (re-)start the port or command that setup specific queue.
2220 port start/stop queue
2221 ~~~~~~~~~~~~~~~~~~~~~
2223 Start/stop a rx/tx queue on a specific port::
2225 testpmd> port (port_id) (rxq|txq) (queue_id) (start|stop)
2227 port config - queue deferred start
2228 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2230 Switch on/off deferred start of a specific port queue::
2232 testpmd> port (port_id) (rxq|txq) (queue_id) deferred_start (on|off)
2235 ~~~~~~~~~~~~~~~~~~~~~
2237 Setup a rx/tx queue on a specific port::
2239 testpmd> port (port_id) (rxq|txq) (queue_id) setup
2241 Only take effect when port is started.
2246 Set the speed and duplex mode for all ports or a specific port::
2248 testpmd> port config (port_id|all) speed (10|100|1000|10000|25000|40000|50000|100000|200000|auto) \
2249 duplex (half|full|auto)
2251 port config - queues/descriptors
2252 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2254 Set number of queues/descriptors for rxq, txq, rxd and txd::
2256 testpmd> port config all (rxq|txq|rxd|txd) (value)
2258 This is equivalent to the ``--rxq``, ``--txq``, ``--rxd`` and ``--txd`` command-line options.
2260 port config - max-pkt-len
2261 ~~~~~~~~~~~~~~~~~~~~~~~~~
2263 Set the maximum packet length::
2265 testpmd> port config all max-pkt-len (value)
2267 This is equivalent to the ``--max-pkt-len`` command-line option.
2269 port config - max-lro-pkt-size
2270 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2272 Set the maximum LRO aggregated packet size::
2274 testpmd> port config all max-lro-pkt-size (value)
2276 This is equivalent to the ``--max-lro-pkt-size`` command-line option.
2278 port config - Drop Packets
2279 ~~~~~~~~~~~~~~~~~~~~~~~~~~
2281 Enable or disable packet drop on all RX queues of all ports when no receive buffers available::
2283 testpmd> port config all drop-en (on|off)
2285 Packet dropping when no receive buffers available is off by default.
2287 The ``on`` option is equivalent to the ``--enable-drop-en`` command-line option.
2292 Set the RSS (Receive Side Scaling) mode on or off::
2294 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)
2296 RSS is on by default.
2298 The ``all`` option is equivalent to eth|vlan|ip|tcp|udp|sctp|ether|l2tpv3|esp|ah|pfcp.
2300 The ``default`` option enables all supported RSS types reported by device info.
2302 The ``none`` option is equivalent to the ``--disable-rss`` command-line option.
2304 port config - RSS Reta
2305 ~~~~~~~~~~~~~~~~~~~~~~
2307 Set the RSS (Receive Side Scaling) redirection table::
2309 testpmd> port config all rss reta (hash,queue)[,(hash,queue)]
2314 Set the DCB mode for an individual port::
2316 testpmd> port config (port_id) dcb vt (on|off) (traffic_class) pfc (on|off)
2318 The traffic class should be 4 or 8.
2323 Set the number of packets per burst::
2325 testpmd> port config all burst (value)
2327 This is equivalent to the ``--burst`` command-line option.
2329 port config - Threshold
2330 ~~~~~~~~~~~~~~~~~~~~~~~
2332 Set thresholds for TX/RX queues::
2334 testpmd> port config all (threshold) (value)
2336 Where the threshold type can be:
2338 * ``txpt:`` Set the prefetch threshold register of the TX rings, 0 <= value <= 255.
2340 * ``txht:`` Set the host threshold register of the TX rings, 0 <= value <= 255.
2342 * ``txwt:`` Set the write-back threshold register of the TX rings, 0 <= value <= 255.
2344 * ``rxpt:`` Set the prefetch threshold register of the RX rings, 0 <= value <= 255.
2346 * ``rxht:`` Set the host threshold register of the RX rings, 0 <= value <= 255.
2348 * ``rxwt:`` Set the write-back threshold register of the RX rings, 0 <= value <= 255.
2350 * ``txfreet:`` Set the transmit free threshold of the TX rings, 0 <= value <= txd.
2352 * ``rxfreet:`` Set the transmit free threshold of the RX rings, 0 <= value <= rxd.
2354 * ``txrst:`` Set the transmit RS bit threshold of TX rings, 0 <= value <= txd.
2356 These threshold options are also available from the command-line.
2358 port config pctype mapping
2359 ~~~~~~~~~~~~~~~~~~~~~~~~~~
2361 Reset pctype mapping table::
2363 testpmd> port config (port_id) pctype mapping reset
2365 Update hardware defined pctype to software defined flow type mapping table::
2367 testpmd> port config (port_id) pctype mapping update (pctype_id_0[,pctype_id_1]*) (flow_type_id)
2371 * ``pctype_id_x``: hardware pctype id as index of bit in bitmask value of the pctype mapping table.
2373 * ``flow_type_id``: software flow type id as the index of the pctype mapping table.
2375 port config input set
2376 ~~~~~~~~~~~~~~~~~~~~~
2378 Config RSS/FDIR/FDIR flexible payload input set for some pctype::
2380 testpmd> port config (port_id) pctype (pctype_id) \
2381 (hash_inset|fdir_inset|fdir_flx_inset) \
2382 (get|set|clear) field (field_idx)
2384 Clear RSS/FDIR/FDIR flexible payload input set for some pctype::
2386 testpmd> port config (port_id) pctype (pctype_id) \
2387 (hash_inset|fdir_inset|fdir_flx_inset) clear all
2391 * ``pctype_id``: hardware packet classification types.
2392 * ``field_idx``: hardware field index.
2394 port config udp_tunnel_port
2395 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
2397 Add/remove UDP tunnel port for VXLAN/GENEVE tunneling protocols::
2399 testpmd> port config (port_id) udp_tunnel_port add|rm vxlan|geneve|vxlan-gpe|ecpri (udp_port)
2401 port config tx_metadata
2402 ~~~~~~~~~~~~~~~~~~~~~~~
2404 Set Tx metadata value per port.
2405 testpmd will add this value to any Tx packet sent from this port::
2407 testpmd> port config (port_id) tx_metadata (value)
2412 Set/clear dynamic flag per port.
2413 testpmd will register this flag in the mbuf (same registration
2414 for both Tx and Rx). Then set/clear this flag for each Tx
2415 packet sent from this port. The set bit only works for Tx packet::
2417 testpmd> port config (port_id) dynf (name) (set|clear)
2422 To configure MTU(Maximum Transmission Unit) on devices using testpmd::
2424 testpmd> port config mtu (port_id) (value)
2426 port config rss hash key
2427 ~~~~~~~~~~~~~~~~~~~~~~~~
2429 To configure the RSS hash key used to compute the RSS
2430 hash of input [IP] packets received on port::
2432 testpmd> port config <port_id> rss-hash-key (ipv4|ipv4-frag|\
2433 ipv4-tcp|ipv4-udp|ipv4-sctp|ipv4-other|\
2434 ipv6|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|\
2435 ipv6-other|l2-payload|ipv6-ex|ipv6-tcp-ex|\
2436 ipv6-udp-ex <string of hex digits \
2437 (variable length, NIC dependent)>)
2442 The following sections show functions for device operations.
2447 Detach a device specified by pci address or virtual device args::
2449 testpmd> device detach (identifier)
2451 Before detaching a device associated with ports, the ports should be stopped and closed.
2453 For example, to detach a pci device whose address is 0002:03:00.0.
2455 .. code-block:: console
2457 testpmd> device detach 0002:03:00.0
2458 Removing a device...
2459 Port 1 is now closed
2460 EAL: Releasing pci mapped resource for 0002:03:00.0
2461 EAL: Calling pci_unmap_resource for 0002:03:00.0 at 0x218a050000
2462 EAL: Calling pci_unmap_resource for 0002:03:00.0 at 0x218c050000
2463 Device 0002:03:00.0 is detached
2464 Now total ports is 1
2466 For example, to detach a port created by pcap PMD.
2468 .. code-block:: console
2470 testpmd> device detach net_pcap0
2471 Removing a device...
2472 Port 0 is now closed
2473 Device net_pcap0 is detached
2474 Now total ports is 0
2477 In this case, identifier is ``net_pcap0``.
2478 This identifier format is the same as ``--vdev`` format of DPDK applications.
2480 Link Bonding Functions
2481 ----------------------
2483 The Link Bonding functions make it possible to dynamically create and
2484 manage link bonding devices from within testpmd interactive prompt.
2486 create bonded device
2487 ~~~~~~~~~~~~~~~~~~~~
2489 Create a new bonding device::
2491 testpmd> create bonded device (mode) (socket)
2493 For example, to create a bonded device in mode 1 on socket 0::
2495 testpmd> create bonded device 1 0
2496 created new bonded device (port X)
2501 Adds Ethernet device to a Link Bonding device::
2503 testpmd> add bonding slave (slave id) (port id)
2505 For example, to add Ethernet device (port 6) to a Link Bonding device (port 10)::
2507 testpmd> add bonding slave 6 10
2510 remove bonding slave
2511 ~~~~~~~~~~~~~~~~~~~~
2513 Removes an Ethernet slave device from a Link Bonding device::
2515 testpmd> remove bonding slave (slave id) (port id)
2517 For example, to remove Ethernet slave device (port 6) to a Link Bonding device (port 10)::
2519 testpmd> remove bonding slave 6 10
2524 Set the Link Bonding mode of a Link Bonding device::
2526 testpmd> set bonding mode (value) (port id)
2528 For example, to set the bonding mode of a Link Bonding device (port 10) to broadcast (mode 3)::
2530 testpmd> set bonding mode 3 10
2535 Set an Ethernet slave device as the primary device on a Link Bonding device::
2537 testpmd> set bonding primary (slave id) (port id)
2539 For example, to set the Ethernet slave device (port 6) as the primary port of a Link Bonding device (port 10)::
2541 testpmd> set bonding primary 6 10
2546 Set the MAC address of a Link Bonding device::
2548 testpmd> set bonding mac (port id) (mac)
2550 For example, to set the MAC address of a Link Bonding device (port 10) to 00:00:00:00:00:01::
2552 testpmd> set bonding mac 10 00:00:00:00:00:01
2554 set bonding balance_xmit_policy
2555 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2557 Set the transmission policy for a Link Bonding device when it is in Balance XOR mode::
2559 testpmd> set bonding balance_xmit_policy (port_id) (l2|l23|l34)
2561 For example, set a Link Bonding device (port 10) to use a balance policy of layer 3+4 (IP addresses & UDP ports)::
2563 testpmd> set bonding balance_xmit_policy 10 l34
2566 set bonding mon_period
2567 ~~~~~~~~~~~~~~~~~~~~~~
2569 Set the link status monitoring polling period in milliseconds for a bonding device.
2571 This adds support for PMD slave devices which do not support link status interrupts.
2572 When the mon_period is set to a value greater than 0 then all PMD's which do not support
2573 link status ISR will be queried every polling interval to check if their link status has changed::
2575 testpmd> set bonding mon_period (port_id) (value)
2577 For example, to set the link status monitoring polling period of bonded device (port 5) to 150ms::
2579 testpmd> set bonding mon_period 5 150
2582 set bonding lacp dedicated_queue
2583 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2585 Enable dedicated tx/rx queues on bonding devices slaves to handle LACP control plane traffic
2586 when in mode 4 (link-aggregation-802.3ad)::
2588 testpmd> set bonding lacp dedicated_queues (port_id) (enable|disable)
2591 set bonding agg_mode
2592 ~~~~~~~~~~~~~~~~~~~~
2594 Enable one of the specific aggregators mode when in mode 4 (link-aggregation-802.3ad)::
2596 testpmd> set bonding agg_mode (port_id) (bandwidth|count|stable)
2602 Show the current configuration of a Link Bonding device::
2604 testpmd> show bonding config (port id)
2607 to show the configuration a Link Bonding device (port 9) with 3 slave devices (1, 3, 4)
2608 in balance mode with a transmission policy of layer 2+3::
2610 testpmd> show bonding config 9
2612 Balance Xmit Policy: BALANCE_XMIT_POLICY_LAYER23
2614 Active Slaves (3): [1 3 4]
2621 The Register Functions can be used to read from and write to registers on the network card referenced by a port number.
2622 This is mainly useful for debugging purposes.
2623 Reference should be made to the appropriate datasheet for the network card for details on the register addresses
2624 and fields that can be accessed.
2629 Display the value of a port register::
2631 testpmd> read reg (port_id) (address)
2633 For example, to examine the Flow Director control register (FDIRCTL, 0x0000EE000) on an Intel 82599 10 GbE Controller::
2635 testpmd> read reg 0 0xEE00
2636 port 0 PCI register at offset 0xEE00: 0x4A060029 (1241907241)
2641 Display a port register bit field::
2643 testpmd> read regfield (port_id) (address) (bit_x) (bit_y)
2645 For example, reading the lowest two bits from the register in the example above::
2647 testpmd> read regfield 0 0xEE00 0 1
2648 port 0 PCI register at offset 0xEE00: bits[0, 1]=0x1 (1)
2653 Display a single port register bit::
2655 testpmd> read regbit (port_id) (address) (bit_x)
2657 For example, reading the lowest bit from the register in the example above::
2659 testpmd> read regbit 0 0xEE00 0
2660 port 0 PCI register at offset 0xEE00: bit 0=1
2665 Set the value of a port register::
2667 testpmd> write reg (port_id) (address) (value)
2669 For example, to clear a register::
2671 testpmd> write reg 0 0xEE00 0x0
2672 port 0 PCI register at offset 0xEE00: 0x00000000 (0)
2677 Set bit field of a port register::
2679 testpmd> write regfield (port_id) (address) (bit_x) (bit_y) (value)
2681 For example, writing to the register cleared in the example above::
2683 testpmd> write regfield 0 0xEE00 0 1 2
2684 port 0 PCI register at offset 0xEE00: 0x00000002 (2)
2689 Set single bit value of a port register::
2691 testpmd> write regbit (port_id) (address) (bit_x) (value)
2693 For example, to set the high bit in the register from the example above::
2695 testpmd> write regbit 0 0xEE00 31 1
2696 port 0 PCI register at offset 0xEE00: 0x8000000A (2147483658)
2698 Traffic Metering and Policing
2699 -----------------------------
2701 The following section shows functions for configuring traffic metering and
2702 policing on the ethernet device through the use of generic ethdev API.
2704 show port traffic management capability
2705 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2707 Show traffic metering and policing capability of the port::
2709 testpmd> show port meter cap (port_id)
2711 add port meter profile (srTCM rfc2967)
2712 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2714 Add meter profile (srTCM rfc2697) to the ethernet device::
2716 testpmd> add port meter profile srtcm_rfc2697 (port_id) (profile_id) \
2717 (cir) (cbs) (ebs) (packet_mode)
2721 * ``profile_id``: ID for the meter profile.
2722 * ``cir``: Committed Information Rate (CIR) (bytes per second or packets per second).
2723 * ``cbs``: Committed Burst Size (CBS) (bytes or packets).
2724 * ``ebs``: Excess Burst Size (EBS) (bytes or packets).
2725 * ``packet_mode``: Packets mode for meter profile.
2727 add port meter profile (trTCM rfc2968)
2728 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2730 Add meter profile (srTCM rfc2698) to the ethernet device::
2732 testpmd> add port meter profile trtcm_rfc2698 (port_id) (profile_id) \
2733 (cir) (pir) (cbs) (pbs) (packet_mode)
2737 * ``profile_id``: ID for the meter profile.
2738 * ``cir``: Committed information rate (bytes per second or packets per second).
2739 * ``pir``: Peak information rate (bytes per second or packets per second).
2740 * ``cbs``: Committed burst size (bytes or packets).
2741 * ``pbs``: Peak burst size (bytes or packets).
2742 * ``packet_mode``: Packets mode for meter profile.
2744 add port meter profile (trTCM rfc4115)
2745 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2747 Add meter profile (trTCM rfc4115) to the ethernet device::
2749 testpmd> add port meter profile trtcm_rfc4115 (port_id) (profile_id) \
2750 (cir) (eir) (cbs) (ebs) (packet_mode)
2754 * ``profile_id``: ID for the meter profile.
2755 * ``cir``: Committed information rate (bytes per second or packets per second).
2756 * ``eir``: Excess information rate (bytes per second or packets per second).
2757 * ``cbs``: Committed burst size (bytes or packets).
2758 * ``ebs``: Excess burst size (bytes or packets).
2759 * ``packet_mode``: Packets mode for meter profile.
2761 delete port meter profile
2762 ~~~~~~~~~~~~~~~~~~~~~~~~~
2764 Delete meter profile from the ethernet device::
2766 testpmd> del port meter profile (port_id) (profile_id)
2771 Create new meter object for the ethernet device::
2773 testpmd> create port meter (port_id) (mtr_id) (profile_id) \
2774 (meter_enable) (g_action) (y_action) (r_action) (stats_mask) (shared) \
2775 (use_pre_meter_color) [(dscp_tbl_entry0) (dscp_tbl_entry1)...\
2780 * ``mtr_id``: meter object ID.
2781 * ``profile_id``: ID for the meter profile.
2782 * ``meter_enable``: When this parameter has a non-zero value, the meter object
2783 gets enabled at the time of creation, otherwise remains disabled.
2784 * ``g_action``: Policer action for the packet with green color.
2785 * ``y_action``: Policer action for the packet with yellow color.
2786 * ``r_action``: Policer action for the packet with red color.
2787 * ``stats_mask``: Mask of statistics counter types to be enabled for the
2789 * ``shared``: When this parameter has a non-zero value, the meter object is
2790 shared by multiple flows. Otherwise, meter object is used by single flow.
2791 * ``use_pre_meter_color``: When this parameter has a non-zero value, the
2792 input color for the current meter object is determined by the latest meter
2793 object in the same flow. Otherwise, the current meter object uses the
2794 *dscp_table* to determine the input color.
2795 * ``dscp_tbl_entryx``: DSCP table entry x providing meter providing input
2796 color, 0 <= x <= 63.
2801 Enable meter for the ethernet device::
2803 testpmd> enable port meter (port_id) (mtr_id)
2808 Disable meter for the ethernet device::
2810 testpmd> disable port meter (port_id) (mtr_id)
2815 Delete meter for the ethernet device::
2817 testpmd> del port meter (port_id) (mtr_id)
2819 Set port meter profile
2820 ~~~~~~~~~~~~~~~~~~~~~~
2822 Set meter profile for the ethernet device::
2824 testpmd> set port meter profile (port_id) (mtr_id) (profile_id)
2826 set port meter dscp table
2827 ~~~~~~~~~~~~~~~~~~~~~~~~~
2829 Set meter dscp table for the ethernet device::
2831 testpmd> set port meter dscp table (port_id) (mtr_id) [(dscp_tbl_entry0) \
2832 (dscp_tbl_entry1)...(dscp_tbl_entry63)]
2834 set port meter policer action
2835 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2837 Set meter policer action for the ethernet device::
2839 testpmd> set port meter policer action (port_id) (mtr_id) (action_mask) \
2840 (action0) [(action1) (action1)]
2844 * ``action_mask``: Bit mask indicating which policer actions need to be
2845 updated. One or more policer actions can be updated in a single function
2846 invocation. To update the policer action associated with color C, bit
2847 (1 << C) needs to be set in *action_mask* and element at position C
2848 in the *actions* array needs to be valid.
2849 * ``actionx``: Policer action for the color x,
2850 RTE_MTR_GREEN <= x < RTE_MTR_COLORS
2852 set port meter stats mask
2853 ~~~~~~~~~~~~~~~~~~~~~~~~~
2855 Set meter stats mask for the ethernet device::
2857 testpmd> set port meter stats mask (port_id) (mtr_id) (stats_mask)
2861 * ``stats_mask``: Bit mask indicating statistics counter types to be enabled.
2863 show port meter stats
2864 ~~~~~~~~~~~~~~~~~~~~~
2866 Show meter stats of the ethernet device::
2868 testpmd> show port meter stats (port_id) (mtr_id) (clear)
2872 * ``clear``: Flag that indicates whether the statistics counters should
2873 be cleared (i.e. set to zero) immediately after they have been read or not.
2878 The following section shows functions for configuring traffic management on
2879 the ethernet device through the use of generic TM API.
2881 show port traffic management capability
2882 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2884 Show traffic management capability of the port::
2886 testpmd> show port tm cap (port_id)
2888 show port traffic management capability (hierarchy level)
2889 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2891 Show traffic management hierarchy level capability of the port::
2893 testpmd> show port tm level cap (port_id) (level_id)
2895 show port traffic management capability (hierarchy node level)
2896 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2898 Show the traffic management hierarchy node capability of the port::
2900 testpmd> show port tm node cap (port_id) (node_id)
2902 show port traffic management hierarchy node type
2903 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2905 Show the port traffic management hierarchy node type::
2907 testpmd> show port tm node type (port_id) (node_id)
2909 show port traffic management hierarchy node stats
2910 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2912 Show the port traffic management hierarchy node statistics::
2914 testpmd> show port tm node stats (port_id) (node_id) (clear)
2918 * ``clear``: When this parameter has a non-zero value, the statistics counters
2919 are cleared (i.e. set to zero) immediately after they have been read,
2920 otherwise the statistics counters are left untouched.
2922 Add port traffic management private shaper profile
2923 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2925 Add the port traffic management private shaper profile::
2927 testpmd> add port tm node shaper profile (port_id) (shaper_profile_id) \
2928 (cmit_tb_rate) (cmit_tb_size) (peak_tb_rate) (peak_tb_size) \
2929 (packet_length_adjust) (packet_mode)
2933 * ``shaper_profile id``: Shaper profile ID for the new profile.
2934 * ``cmit_tb_rate``: Committed token bucket rate (bytes per second or packets per second).
2935 * ``cmit_tb_size``: Committed token bucket size (bytes or packets).
2936 * ``peak_tb_rate``: Peak token bucket rate (bytes per second or packets per second).
2937 * ``peak_tb_size``: Peak token bucket size (bytes or packets).
2938 * ``packet_length_adjust``: The value (bytes) to be added to the length of
2939 each packet for the purpose of shaping. This parameter value can be used to
2940 correct the packet length with the framing overhead bytes that are consumed
2942 * ``packet_mode``: Shaper configured in packet mode. This parameter value if
2943 zero, configures shaper in byte mode and if non-zero configures it in packet
2946 Delete port traffic management private shaper profile
2947 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2949 Delete the port traffic management private shaper::
2951 testpmd> del port tm node shaper profile (port_id) (shaper_profile_id)
2955 * ``shaper_profile id``: Shaper profile ID that needs to be deleted.
2957 Add port traffic management shared shaper
2958 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2960 Create the port traffic management shared shaper::
2962 testpmd> add port tm node shared shaper (port_id) (shared_shaper_id) \
2967 * ``shared_shaper_id``: Shared shaper ID to be created.
2968 * ``shaper_profile id``: Shaper profile ID for shared shaper.
2970 Set port traffic management shared shaper
2971 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2973 Update the port traffic management shared shaper::
2975 testpmd> set port tm node shared shaper (port_id) (shared_shaper_id) \
2980 * ``shared_shaper_id``: Shared shaper ID to be update.
2981 * ``shaper_profile id``: Shaper profile ID for shared shaper.
2983 Delete port traffic management shared shaper
2984 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2986 Delete the port traffic management shared shaper::
2988 testpmd> del port tm node shared shaper (port_id) (shared_shaper_id)
2992 * ``shared_shaper_id``: Shared shaper ID to be deleted.
2994 Set port traffic management hierarchy node private shaper
2995 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2997 set the port traffic management hierarchy node private shaper::
2999 testpmd> set port tm node shaper profile (port_id) (node_id) \
3004 * ``shaper_profile id``: Private shaper profile ID to be enabled on the
3007 Add port traffic management WRED profile
3008 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3010 Create a new WRED profile::
3012 testpmd> add port tm node wred profile (port_id) (wred_profile_id) \
3013 (color_g) (min_th_g) (max_th_g) (maxp_inv_g) (wq_log2_g) \
3014 (color_y) (min_th_y) (max_th_y) (maxp_inv_y) (wq_log2_y) \
3015 (color_r) (min_th_r) (max_th_r) (maxp_inv_r) (wq_log2_r)
3019 * ``wred_profile id``: Identifier for the newly create WRED profile
3020 * ``color_g``: Packet color (green)
3021 * ``min_th_g``: Minimum queue threshold for packet with green color
3022 * ``max_th_g``: Minimum queue threshold for packet with green color
3023 * ``maxp_inv_g``: Inverse of packet marking probability maximum value (maxp)
3024 * ``wq_log2_g``: Negated log2 of queue weight (wq)
3025 * ``color_y``: Packet color (yellow)
3026 * ``min_th_y``: Minimum queue threshold for packet with yellow color
3027 * ``max_th_y``: Minimum queue threshold for packet with yellow color
3028 * ``maxp_inv_y``: Inverse of packet marking probability maximum value (maxp)
3029 * ``wq_log2_y``: Negated log2 of queue weight (wq)
3030 * ``color_r``: Packet color (red)
3031 * ``min_th_r``: Minimum queue threshold for packet with yellow color
3032 * ``max_th_r``: Minimum queue threshold for packet with yellow color
3033 * ``maxp_inv_r``: Inverse of packet marking probability maximum value (maxp)
3034 * ``wq_log2_r``: Negated log2 of queue weight (wq)
3036 Delete port traffic management WRED profile
3037 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3039 Delete the WRED profile::
3041 testpmd> del port tm node wred profile (port_id) (wred_profile_id)
3043 Add port traffic management hierarchy nonleaf node
3044 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3046 Add nonleaf node to port traffic management hierarchy::
3048 testpmd> add port tm nonleaf node (port_id) (node_id) (parent_node_id) \
3049 (priority) (weight) (level_id) (shaper_profile_id) \
3050 (n_sp_priorities) (stats_mask) (n_shared_shapers) \
3051 [(shared_shaper_0) (shared_shaper_1) ...] \
3055 * ``parent_node_id``: Node ID of the parent.
3056 * ``priority``: Node priority (highest node priority is zero). This is used by
3057 the SP algorithm running on the parent node for scheduling this node.
3058 * ``weight``: Node weight (lowest weight is one). The node weight is relative
3059 to the weight sum of all siblings that have the same priority. It is used by
3060 the WFQ algorithm running on the parent node for scheduling this node.
3061 * ``level_id``: Hierarchy level of the node.
3062 * ``shaper_profile_id``: Shaper profile ID of the private shaper to be used by
3064 * ``n_sp_priorities``: Number of strict priorities.
3065 * ``stats_mask``: Mask of statistics counter types to be enabled for this node.
3066 * ``n_shared_shapers``: Number of shared shapers.
3067 * ``shared_shaper_id``: Shared shaper id.
3069 Add port traffic management hierarchy nonleaf node with packet mode
3070 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3072 Add nonleaf node with packet mode to port traffic management hierarchy::
3074 testpmd> add port tm nonleaf node pktmode (port_id) (node_id) (parent_node_id) \
3075 (priority) (weight) (level_id) (shaper_profile_id) \
3076 (n_sp_priorities) (stats_mask) (n_shared_shapers) \
3077 [(shared_shaper_0) (shared_shaper_1) ...] \
3081 * ``parent_node_id``: Node ID of the parent.
3082 * ``priority``: Node priority (highest node priority is zero). This is used by
3083 the SP algorithm running on the parent node for scheduling this node.
3084 * ``weight``: Node weight (lowest weight is one). The node weight is relative
3085 to the weight sum of all siblings that have the same priority. It is used by
3086 the WFQ algorithm running on the parent node for scheduling this node.
3087 * ``level_id``: Hierarchy level of the node.
3088 * ``shaper_profile_id``: Shaper profile ID of the private shaper to be used by
3090 * ``n_sp_priorities``: Number of strict priorities. Packet mode is enabled on
3092 * ``stats_mask``: Mask of statistics counter types to be enabled for this node.
3093 * ``n_shared_shapers``: Number of shared shapers.
3094 * ``shared_shaper_id``: Shared shaper id.
3096 Add port traffic management hierarchy leaf node
3097 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3099 Add leaf node to port traffic management hierarchy::
3101 testpmd> add port tm leaf node (port_id) (node_id) (parent_node_id) \
3102 (priority) (weight) (level_id) (shaper_profile_id) \
3103 (cman_mode) (wred_profile_id) (stats_mask) (n_shared_shapers) \
3104 [(shared_shaper_id) (shared_shaper_id) ...] \
3108 * ``parent_node_id``: Node ID of the parent.
3109 * ``priority``: Node priority (highest node priority is zero). This is used by
3110 the SP algorithm running on the parent node for scheduling this node.
3111 * ``weight``: Node weight (lowest weight is one). The node weight is relative
3112 to the weight sum of all siblings that have the same priority. It is used by
3113 the WFQ algorithm running on the parent node for scheduling this node.
3114 * ``level_id``: Hierarchy level of the node.
3115 * ``shaper_profile_id``: Shaper profile ID of the private shaper to be used by
3117 * ``cman_mode``: Congestion management mode to be enabled for this node.
3118 * ``wred_profile_id``: WRED profile id to be enabled for this node.
3119 * ``stats_mask``: Mask of statistics counter types to be enabled for this node.
3120 * ``n_shared_shapers``: Number of shared shapers.
3121 * ``shared_shaper_id``: Shared shaper id.
3123 Delete port traffic management hierarchy node
3124 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3126 Delete node from port traffic management hierarchy::
3128 testpmd> del port tm node (port_id) (node_id)
3130 Update port traffic management hierarchy parent node
3131 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3133 Update port traffic management hierarchy parent node::
3135 testpmd> set port tm node parent (port_id) (node_id) (parent_node_id) \
3138 This function can only be called after the hierarchy commit invocation. Its
3139 success depends on the port support for this operation, as advertised through
3140 the port capability set. This function is valid for all nodes of the traffic
3141 management hierarchy except root node.
3143 Suspend port traffic management hierarchy node
3144 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3146 testpmd> suspend port tm node (port_id) (node_id)
3148 Resume port traffic management hierarchy node
3149 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3151 testpmd> resume port tm node (port_id) (node_id)
3153 Commit port traffic management hierarchy
3154 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3156 Commit the traffic management hierarchy on the port::
3158 testpmd> port tm hierarchy commit (port_id) (clean_on_fail)
3162 * ``clean_on_fail``: When set to non-zero, hierarchy is cleared on function
3163 call failure. On the other hand, hierarchy is preserved when this parameter
3166 Set port traffic management mark VLAN dei
3167 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3169 Enables/Disables the traffic management marking on the port for VLAN packets::
3171 testpmd> set port tm mark vlan_dei <port_id> <green> <yellow> <red>
3175 * ``port_id``: The port which on which VLAN packets marked as ``green`` or
3176 ``yellow`` or ``red`` will have dei bit enabled
3178 * ``green`` enable 1, disable 0 marking for dei bit of VLAN packets marked as green
3180 * ``yellow`` enable 1, disable 0 marking for dei bit of VLAN packets marked as yellow
3182 * ``red`` enable 1, disable 0 marking for dei bit of VLAN packets marked as red
3184 Set port traffic management mark IP dscp
3185 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3187 Enables/Disables the traffic management marking on the port for IP dscp packets::
3189 testpmd> set port tm mark ip_dscp <port_id> <green> <yellow> <red>
3193 * ``port_id``: The port which on which IP packets marked as ``green`` or
3194 ``yellow`` or ``red`` will have IP dscp bits updated
3196 * ``green`` enable 1, disable 0 marking IP dscp to low drop precedence for green packets
3198 * ``yellow`` enable 1, disable 0 marking IP dscp to medium drop precedence for yellow packets
3200 * ``red`` enable 1, disable 0 marking IP dscp to high drop precedence for red packets
3202 Set port traffic management mark IP ecn
3203 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3205 Enables/Disables the traffic management marking on the port for IP ecn packets::
3207 testpmd> set port tm mark ip_ecn <port_id> <green> <yellow> <red>
3211 * ``port_id``: The port which on which IP packets marked as ``green`` or
3212 ``yellow`` or ``red`` will have IP ecn bits updated
3214 * ``green`` enable 1, disable 0 marking IP ecn for green marked packets with ecn of 2'b01 or 2'b10
3215 to ecn of 2'b11 when IP is caring TCP or SCTP
3217 * ``yellow`` enable 1, disable 0 marking IP ecn for yellow marked packets with ecn of 2'b01 or 2'b10
3218 to ecn of 2'b11 when IP is caring TCP or SCTP
3220 * ``red`` enable 1, disable 0 marking IP ecn for yellow marked packets with ecn of 2'b01 or 2'b10
3221 to ecn of 2'b11 when IP is caring TCP or SCTP
3226 This section details the available filter functions that are available.
3228 Note these functions interface the deprecated legacy filtering framework,
3229 superseded by *rte_flow*. See `Flow rules management`_.
3231 .. _testpmd_flow_director:
3236 Set flow director's input masks::
3238 flow_director_mask (port_id) mode IP vlan (vlan_value) \
3239 src_mask (ipv4_src) (ipv6_src) (src_port) \
3240 dst_mask (ipv4_dst) (ipv6_dst) (dst_port)
3242 flow_director_mask (port_id) mode MAC-VLAN vlan (vlan_value)
3244 flow_director_mask (port_id) mode Tunnel vlan (vlan_value) \
3245 mac (mac_value) tunnel-type (tunnel_type_value) \
3246 tunnel-id (tunnel_id_value)
3248 Example, to set flow director mask on port 0::
3250 testpmd> flow_director_mask 0 mode IP vlan 0xefff \
3251 src_mask 255.255.255.255 \
3252 FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF \
3253 dst_mask 255.255.255.255 \
3254 FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF
3256 flow_director_flex_payload
3257 ~~~~~~~~~~~~~~~~~~~~~~~~~~
3259 Configure flexible payload selection::
3261 flow_director_flex_payload (port_id) (raw|l2|l3|l4) (config)
3263 For example, to select the first 16 bytes from the offset 4 (bytes) of packet's payload as flexible payload::
3265 testpmd> flow_director_flex_payload 0 l4 \
3266 (4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19)
3269 .. _testpmd_rte_flow:
3271 Flow rules management
3272 ---------------------
3274 Control of the generic flow API (*rte_flow*) is fully exposed through the
3275 ``flow`` command (validation, creation, destruction, queries and operation
3278 Considering *rte_flow* overlaps with all `Filter Functions`_, using both
3279 features simultaneously may cause undefined side-effects and is therefore
3285 Because the ``flow`` command uses dynamic tokens to handle the large number
3286 of possible flow rules combinations, its behavior differs slightly from
3287 other commands, in particular:
3289 - Pressing *?* or the *<tab>* key displays contextual help for the current
3290 token, not that of the entire command.
3292 - Optional and repeated parameters are supported (provided they are listed
3293 in the contextual help).
3295 The first parameter stands for the operation mode. Possible operations and
3296 their general syntax are described below. They are covered in detail in the
3299 - Check whether a flow rule can be created::
3301 flow validate {port_id}
3302 [group {group_id}] [priority {level}] [ingress] [egress] [transfer]
3303 pattern {item} [/ {item} [...]] / end
3304 actions {action} [/ {action} [...]] / end
3306 - Create a flow rule::
3308 flow create {port_id}
3309 [group {group_id}] [priority {level}] [ingress] [egress] [transfer]
3310 pattern {item} [/ {item} [...]] / end
3311 actions {action} [/ {action} [...]] / end
3313 - Destroy specific flow rules::
3315 flow destroy {port_id} rule {rule_id} [...]
3317 - Destroy all flow rules::
3319 flow flush {port_id}
3321 - Query an existing flow rule::
3323 flow query {port_id} {rule_id} {action}
3325 - List existing flow rules sorted by priority, filtered by group
3328 flow list {port_id} [group {group_id}] [...]
3330 - Restrict ingress traffic to the defined flow rules::
3332 flow isolate {port_id} {boolean}
3334 - Dump internal representation information of all flows in hardware::
3336 flow dump {port_id} all {output_file}
3340 flow dump {port_id} rule {rule_id} {output_file}
3342 - List and destroy aged flow rules::
3344 flow aged {port_id} [destroy]
3346 - Tunnel offload - create a tunnel stub::
3348 flow tunnel create {port_id} type {tunnel_type}
3350 - Tunnel offload - destroy a tunnel stub::
3352 flow tunnel destroy {port_id} id {tunnel_id}
3354 - Tunnel offload - list port tunnel stubs::
3356 flow tunnel list {port_id}
3358 Creating a tunnel stub for offload
3359 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3361 ``flow tunnel create`` setup a tunnel stub for tunnel offload flow rules::
3363 flow tunnel create {port_id} type {tunnel_type}
3365 If successful, it will return a tunnel stub ID usable with other commands::
3367 port [...]: flow tunnel #[...] type [...]
3369 Tunnel stub ID is relative to a port.
3371 Destroying tunnel offload stub
3372 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3374 ``flow tunnel destroy`` destroy port tunnel stub::
3376 flow tunnel destroy {port_id} id {tunnel_id}
3378 Listing tunnel offload stubs
3379 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3381 ``flow tunnel list`` list port tunnel offload stubs::
3383 flow tunnel list {port_id}
3385 Validating flow rules
3386 ~~~~~~~~~~~~~~~~~~~~~
3388 ``flow validate`` reports whether a flow rule would be accepted by the
3389 underlying device in its current state but stops short of creating it. It is
3390 bound to ``rte_flow_validate()``::
3392 flow validate {port_id}
3393 [group {group_id}] [priority {level}] [ingress] [egress] [transfer]
3394 pattern {item} [/ {item} [...]] / end
3395 actions {action} [/ {action} [...]] / end
3397 If successful, it will show::
3401 Otherwise it will show an error message of the form::
3403 Caught error type [...] ([...]): [...]
3405 This command uses the same parameters as ``flow create``, their format is
3406 described in `Creating flow rules`_.
3408 Check whether redirecting any Ethernet packet received on port 0 to RX queue
3409 index 6 is supported::
3411 testpmd> flow validate 0 ingress pattern eth / end
3412 actions queue index 6 / end
3416 Port 0 does not support TCPv6 rules::
3418 testpmd> flow validate 0 ingress pattern eth / ipv6 / tcp / end
3420 Caught error type 9 (specific pattern item): Invalid argument
3426 ``flow create`` validates and creates the specified flow rule. It is bound
3427 to ``rte_flow_create()``::
3429 flow create {port_id}
3430 [group {group_id}] [priority {level}] [ingress] [egress] [transfer]
3431 [tunnel_set {tunnel_id}] [tunnel_match {tunnel_id}]
3432 pattern {item} [/ {item} [...]] / end
3433 actions {action} [/ {action} [...]] / end
3435 If successful, it will return a flow rule ID usable with other commands::
3437 Flow rule #[...] created
3439 Otherwise it will show an error message of the form::
3441 Caught error type [...] ([...]): [...]
3443 Parameters describe in the following order:
3445 - Attributes (*group*, *priority*, *ingress*, *egress*, *transfer* tokens).
3446 - Tunnel offload specification (tunnel_set, tunnel_match)
3447 - A matching pattern, starting with the *pattern* token and terminated by an
3449 - Actions, starting with the *actions* token and terminated by an *end*
3452 These translate directly to *rte_flow* objects provided as-is to the
3453 underlying functions.
3455 The shortest valid definition only comprises mandatory tokens::
3457 testpmd> flow create 0 pattern end actions end
3459 Note that PMDs may refuse rules that essentially do nothing such as this
3462 **All unspecified object values are automatically initialized to 0.**
3467 These tokens affect flow rule attributes (``struct rte_flow_attr``) and are
3468 specified before the ``pattern`` token.
3470 - ``group {group id}``: priority group.
3471 - ``priority {level}``: priority level within group.
3472 - ``ingress``: rule applies to ingress traffic.
3473 - ``egress``: rule applies to egress traffic.
3474 - ``transfer``: apply rule directly to endpoints found in pattern.
3476 Each instance of an attribute specified several times overrides the previous
3477 value as shown below (group 4 is used)::
3479 testpmd> flow create 0 group 42 group 24 group 4 [...]
3481 Note that once enabled, ``ingress`` and ``egress`` cannot be disabled.
3483 While not specifying a direction is an error, some rules may allow both
3486 Most rules affect RX therefore contain the ``ingress`` token::
3488 testpmd> flow create 0 ingress pattern [...]
3493 Indicate tunnel offload rule type
3495 - ``tunnel_set {tunnel_id}``: mark rule as tunnel offload decap_set type.
3496 - ``tunnel_match {tunnel_id}``: mark rule as tunel offload match type.
3501 A matching pattern starts after the ``pattern`` token. It is made of pattern
3502 items and is terminated by a mandatory ``end`` item.
3504 Items are named after their type (*RTE_FLOW_ITEM_TYPE_* from ``enum
3505 rte_flow_item_type``).
3507 The ``/`` token is used as a separator between pattern items as shown
3510 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end [...]
3512 Note that protocol items like these must be stacked from lowest to highest
3513 layer to make sense. For instance, the following rule is either invalid or
3514 unlikely to match any packet::
3516 testpmd> flow create 0 ingress pattern eth / udp / ipv4 / end [...]
3518 More information on these restrictions can be found in the *rte_flow*
3521 Several items support additional specification structures, for example
3522 ``ipv4`` allows specifying source and destination addresses as follows::
3524 testpmd> flow create 0 ingress pattern eth / ipv4 src is 10.1.1.1
3525 dst is 10.2.0.0 / end [...]
3527 This rule matches all IPv4 traffic with the specified properties.
3529 In this example, ``src`` and ``dst`` are field names of the underlying
3530 ``struct rte_flow_item_ipv4`` object. All item properties can be specified
3531 in a similar fashion.
3533 The ``is`` token means that the subsequent value must be matched exactly,
3534 and assigns ``spec`` and ``mask`` fields in ``struct rte_flow_item``
3535 accordingly. Possible assignment tokens are:
3537 - ``is``: match value perfectly (with full bit-mask).
3538 - ``spec``: match value according to configured bit-mask.
3539 - ``last``: specify upper bound to establish a range.
3540 - ``mask``: specify bit-mask with relevant bits set to one.
3541 - ``prefix``: generate bit-mask with <prefix-length> most-significant bits set to one.
3543 These yield identical results::
3545 ipv4 src is 10.1.1.1
3549 ipv4 src spec 10.1.1.1 src mask 255.255.255.255
3553 ipv4 src spec 10.1.1.1 src prefix 32
3557 ipv4 src is 10.1.1.1 src last 10.1.1.1 # range with a single value
3561 ipv4 src is 10.1.1.1 src last 0 # 0 disables range
3563 Inclusive ranges can be defined with ``last``::
3565 ipv4 src is 10.1.1.1 src last 10.2.3.4 # 10.1.1.1 to 10.2.3.4
3567 Note that ``mask`` affects both ``spec`` and ``last``::
3569 ipv4 src is 10.1.1.1 src last 10.2.3.4 src mask 255.255.0.0
3570 # matches 10.1.0.0 to 10.2.255.255
3572 Properties can be modified multiple times::
3574 ipv4 src is 10.1.1.1 src is 10.1.2.3 src is 10.2.3.4 # matches 10.2.3.4
3578 ipv4 src is 10.1.1.1 src prefix 24 src prefix 16 # matches 10.1.0.0/16
3583 This section lists supported pattern items and their attributes, if any.
3585 - ``end``: end list of pattern items.
3587 - ``void``: no-op pattern item.
3589 - ``invert``: perform actions when pattern does not match.
3591 - ``any``: match any protocol for the current layer.
3593 - ``num {unsigned}``: number of layers covered.
3595 - ``pf``: match traffic from/to the physical function.
3597 - ``vf``: match traffic from/to a virtual function ID.
3599 - ``id {unsigned}``: VF ID.
3601 - ``phy_port``: match traffic from/to a specific physical port.
3603 - ``index {unsigned}``: physical port index.
3605 - ``port_id``: match traffic from/to a given DPDK port ID.
3607 - ``id {unsigned}``: DPDK port ID.
3609 - ``mark``: match value set in previously matched flow rule using the mark action.
3611 - ``id {unsigned}``: arbitrary integer value.
3613 - ``raw``: match an arbitrary byte string.
3615 - ``relative {boolean}``: look for pattern after the previous item.
3616 - ``search {boolean}``: search pattern from offset (see also limit).
3617 - ``offset {integer}``: absolute or relative offset for pattern.
3618 - ``limit {unsigned}``: search area limit for start of pattern.
3619 - ``pattern {string}``: byte string to look for.
3621 - ``eth``: match Ethernet header.
3623 - ``dst {MAC-48}``: destination MAC.
3624 - ``src {MAC-48}``: source MAC.
3625 - ``type {unsigned}``: EtherType or TPID.
3627 - ``vlan``: match 802.1Q/ad VLAN tag.
3629 - ``tci {unsigned}``: tag control information.
3630 - ``pcp {unsigned}``: priority code point.
3631 - ``dei {unsigned}``: drop eligible indicator.
3632 - ``vid {unsigned}``: VLAN identifier.
3633 - ``inner_type {unsigned}``: inner EtherType or TPID.
3635 - ``ipv4``: match IPv4 header.
3637 - ``tos {unsigned}``: type of service.
3638 - ``ttl {unsigned}``: time to live.
3639 - ``proto {unsigned}``: next protocol ID.
3640 - ``src {ipv4 address}``: source address.
3641 - ``dst {ipv4 address}``: destination address.
3643 - ``ipv6``: match IPv6 header.
3645 - ``tc {unsigned}``: traffic class.
3646 - ``flow {unsigned}``: flow label.
3647 - ``proto {unsigned}``: protocol (next header).
3648 - ``hop {unsigned}``: hop limit.
3649 - ``src {ipv6 address}``: source address.
3650 - ``dst {ipv6 address}``: destination address.
3652 - ``icmp``: match ICMP header.
3654 - ``type {unsigned}``: ICMP packet type.
3655 - ``code {unsigned}``: ICMP packet code.
3657 - ``udp``: match UDP header.
3659 - ``src {unsigned}``: UDP source port.
3660 - ``dst {unsigned}``: UDP destination port.
3662 - ``tcp``: match TCP header.
3664 - ``src {unsigned}``: TCP source port.
3665 - ``dst {unsigned}``: TCP destination port.
3667 - ``sctp``: match SCTP header.
3669 - ``src {unsigned}``: SCTP source port.
3670 - ``dst {unsigned}``: SCTP destination port.
3671 - ``tag {unsigned}``: validation tag.
3672 - ``cksum {unsigned}``: checksum.
3674 - ``vxlan``: match VXLAN header.
3676 - ``vni {unsigned}``: VXLAN identifier.
3678 - ``e_tag``: match IEEE 802.1BR E-Tag header.
3680 - ``grp_ecid_b {unsigned}``: GRP and E-CID base.
3682 - ``nvgre``: match NVGRE header.
3684 - ``tni {unsigned}``: virtual subnet ID.
3686 - ``mpls``: match MPLS header.
3688 - ``label {unsigned}``: MPLS label.
3690 - ``gre``: match GRE header.
3692 - ``protocol {unsigned}``: protocol type.
3694 - ``gre_key``: match GRE optional key field.
3696 - ``value {unsigned}``: key value.
3698 - ``fuzzy``: fuzzy pattern match, expect faster than default.
3700 - ``thresh {unsigned}``: accuracy threshold.
3702 - ``gtp``, ``gtpc``, ``gtpu``: match GTPv1 header.
3704 - ``teid {unsigned}``: tunnel endpoint identifier.
3706 - ``geneve``: match GENEVE header.
3708 - ``vni {unsigned}``: virtual network identifier.
3709 - ``protocol {unsigned}``: protocol type.
3711 - ``geneve-opt``: match GENEVE header option.
3713 - ``class {unsigned}``: GENEVE option class.
3714 - ``type {unsigned}``: GENEVE option type.
3715 - ``length {unsigned}``: GENEVE option length in 32-bit words.
3716 - ``data {hex string}``: GENEVE option data, the length is defined by
3719 - ``vxlan-gpe``: match VXLAN-GPE header.
3721 - ``vni {unsigned}``: VXLAN-GPE identifier.
3723 - ``arp_eth_ipv4``: match ARP header for Ethernet/IPv4.
3725 - ``sha {MAC-48}``: sender hardware address.
3726 - ``spa {ipv4 address}``: sender IPv4 address.
3727 - ``tha {MAC-48}``: target hardware address.
3728 - ``tpa {ipv4 address}``: target IPv4 address.
3730 - ``ipv6_ext``: match presence of any IPv6 extension header.
3732 - ``next_hdr {unsigned}``: next header.
3734 - ``icmp6``: match any ICMPv6 header.
3736 - ``type {unsigned}``: ICMPv6 type.
3737 - ``code {unsigned}``: ICMPv6 code.
3739 - ``icmp6_nd_ns``: match ICMPv6 neighbor discovery solicitation.
3741 - ``target_addr {ipv6 address}``: target address.
3743 - ``icmp6_nd_na``: match ICMPv6 neighbor discovery advertisement.
3745 - ``target_addr {ipv6 address}``: target address.
3747 - ``icmp6_nd_opt``: match presence of any ICMPv6 neighbor discovery option.
3749 - ``type {unsigned}``: ND option type.
3751 - ``icmp6_nd_opt_sla_eth``: match ICMPv6 neighbor discovery source Ethernet
3752 link-layer address option.
3754 - ``sla {MAC-48}``: source Ethernet LLA.
3756 - ``icmp6_nd_opt_tla_eth``: match ICMPv6 neighbor discovery target Ethernet
3757 link-layer address option.
3759 - ``tla {MAC-48}``: target Ethernet LLA.
3761 - ``meta``: match application specific metadata.
3763 - ``data {unsigned}``: metadata value.
3765 - ``gtp_psc``: match GTP PDU extension header with type 0x85.
3767 - ``pdu_type {unsigned}``: PDU type.
3769 - ``qfi {unsigned}``: PPP, RQI and QoS flow identifier.
3771 - ``pppoes``, ``pppoed``: match PPPoE header.
3773 - ``session_id {unsigned}``: session identifier.
3775 - ``pppoe_proto_id``: match PPPoE session protocol identifier.
3777 - ``proto_id {unsigned}``: PPP protocol identifier.
3779 - ``l2tpv3oip``: match L2TPv3 over IP header.
3781 - ``session_id {unsigned}``: L2TPv3 over IP session identifier.
3783 - ``ah``: match AH header.
3785 - ``spi {unsigned}``: security parameters index.
3787 - ``pfcp``: match PFCP header.
3789 - ``s_field {unsigned}``: S field.
3790 - ``seid {unsigned}``: session endpoint identifier.
3795 A list of actions starts after the ``actions`` token in the same fashion as
3796 `Matching pattern`_; actions are separated by ``/`` tokens and the list is
3797 terminated by a mandatory ``end`` action.
3799 Actions are named after their type (*RTE_FLOW_ACTION_TYPE_* from ``enum
3800 rte_flow_action_type``).
3802 Dropping all incoming UDPv4 packets can be expressed as follows::
3804 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
3807 Several actions have configurable properties which must be specified when
3808 there is no valid default value. For example, ``queue`` requires a target
3811 This rule redirects incoming UDPv4 traffic to queue index 6::
3813 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
3814 actions queue index 6 / end
3816 While this one could be rejected by PMDs (unspecified queue index)::
3818 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
3821 As defined by *rte_flow*, the list is not ordered, all actions of a given
3822 rule are performed simultaneously. These are equivalent::
3824 queue index 6 / void / mark id 42 / end
3828 void / mark id 42 / queue index 6 / end
3830 All actions in a list should have different types, otherwise only the last
3831 action of a given type is taken into account::
3833 queue index 4 / queue index 5 / queue index 6 / end # will use queue 6
3837 drop / drop / drop / end # drop is performed only once
3841 mark id 42 / queue index 3 / mark id 24 / end # mark will be 24
3843 Considering they are performed simultaneously, opposite and overlapping
3844 actions can sometimes be combined when the end result is unambiguous::
3846 drop / queue index 6 / end # drop has no effect
3850 queue index 6 / rss queues 6 7 8 / end # queue has no effect
3854 drop / passthru / end # drop has no effect
3856 Note that PMDs may still refuse such combinations.
3861 This section lists supported actions and their attributes, if any.
3863 - ``end``: end list of actions.
3865 - ``void``: no-op action.
3867 - ``passthru``: let subsequent rule process matched packets.
3869 - ``jump``: redirect traffic to group on device.
3871 - ``group {unsigned}``: group to redirect to.
3873 - ``mark``: attach 32 bit value to packets.
3875 - ``id {unsigned}``: 32 bit value to return with packets.
3877 - ``flag``: flag packets.
3879 - ``queue``: assign packets to a given queue index.
3881 - ``index {unsigned}``: queue index to use.
3883 - ``drop``: drop packets (note: passthru has priority).
3885 - ``count``: enable counters for this rule.
3887 - ``rss``: spread packets among several queues.
3889 - ``func {hash function}``: RSS hash function to apply, allowed tokens are
3890 ``toeplitz``, ``simple_xor``, ``symmetric_toeplitz`` and ``default``.
3892 - ``level {unsigned}``: encapsulation level for ``types``.
3894 - ``types [{RSS hash type} [...]] end``: specific RSS hash types.
3895 Note that an empty list does not disable RSS but instead requests
3896 unspecified "best-effort" settings.
3898 - ``key {string}``: RSS hash key, overrides ``key_len``.
3900 - ``key_len {unsigned}``: RSS hash key length in bytes, can be used in
3901 conjunction with ``key`` to pad or truncate it.
3903 - ``queues [{unsigned} [...]] end``: queue indices to use.
3905 - ``pf``: direct traffic to physical function.
3907 - ``vf``: direct traffic to a virtual function ID.
3909 - ``original {boolean}``: use original VF ID if possible.
3910 - ``id {unsigned}``: VF ID.
3912 - ``phy_port``: direct packets to physical port index.
3914 - ``original {boolean}``: use original port index if possible.
3915 - ``index {unsigned}``: physical port index.
3917 - ``port_id``: direct matching traffic to a given DPDK port ID.
3919 - ``original {boolean}``: use original DPDK port ID if possible.
3920 - ``id {unsigned}``: DPDK port ID.
3922 - ``of_set_mpls_ttl``: OpenFlow's ``OFPAT_SET_MPLS_TTL``.
3924 - ``mpls_ttl``: MPLS TTL.
3926 - ``of_dec_mpls_ttl``: OpenFlow's ``OFPAT_DEC_MPLS_TTL``.
3928 - ``of_set_nw_ttl``: OpenFlow's ``OFPAT_SET_NW_TTL``.
3930 - ``nw_ttl``: IP TTL.
3932 - ``of_dec_nw_ttl``: OpenFlow's ``OFPAT_DEC_NW_TTL``.
3934 - ``of_copy_ttl_out``: OpenFlow's ``OFPAT_COPY_TTL_OUT``.
3936 - ``of_copy_ttl_in``: OpenFlow's ``OFPAT_COPY_TTL_IN``.
3938 - ``of_pop_vlan``: OpenFlow's ``OFPAT_POP_VLAN``.
3940 - ``of_push_vlan``: OpenFlow's ``OFPAT_PUSH_VLAN``.
3942 - ``ethertype``: Ethertype.
3944 - ``of_set_vlan_vid``: OpenFlow's ``OFPAT_SET_VLAN_VID``.
3946 - ``vlan_vid``: VLAN id.
3948 - ``of_set_vlan_pcp``: OpenFlow's ``OFPAT_SET_VLAN_PCP``.
3950 - ``vlan_pcp``: VLAN priority.
3952 - ``of_pop_mpls``: OpenFlow's ``OFPAT_POP_MPLS``.
3954 - ``ethertype``: Ethertype.
3956 - ``of_push_mpls``: OpenFlow's ``OFPAT_PUSH_MPLS``.
3958 - ``ethertype``: Ethertype.
3960 - ``vxlan_encap``: Performs a VXLAN encapsulation, outer layer configuration
3961 is done through `Config VXLAN Encap outer layers`_.
3963 - ``vxlan_decap``: Performs a decapsulation action by stripping all headers of
3964 the VXLAN tunnel network overlay from the matched flow.
3966 - ``nvgre_encap``: Performs a NVGRE encapsulation, outer layer configuration
3967 is done through `Config NVGRE Encap outer layers`_.
3969 - ``nvgre_decap``: Performs a decapsulation action by stripping all headers of
3970 the NVGRE tunnel network overlay from the matched flow.
3972 - ``l2_encap``: Performs a L2 encapsulation, L2 configuration
3973 is done through `Config L2 Encap`_.
3975 - ``l2_decap``: Performs a L2 decapsulation, L2 configuration
3976 is done through `Config L2 Decap`_.
3978 - ``mplsogre_encap``: Performs a MPLSoGRE encapsulation, outer layer
3979 configuration is done through `Config MPLSoGRE Encap outer layers`_.
3981 - ``mplsogre_decap``: Performs a MPLSoGRE decapsulation, outer layer
3982 configuration is done through `Config MPLSoGRE Decap outer layers`_.
3984 - ``mplsoudp_encap``: Performs a MPLSoUDP encapsulation, outer layer
3985 configuration is done through `Config MPLSoUDP Encap outer layers`_.
3987 - ``mplsoudp_decap``: Performs a MPLSoUDP decapsulation, outer layer
3988 configuration is done through `Config MPLSoUDP Decap outer layers`_.
3990 - ``set_ipv4_src``: Set a new IPv4 source address in the outermost IPv4 header.
3992 - ``ipv4_addr``: New IPv4 source address.
3994 - ``set_ipv4_dst``: Set a new IPv4 destination address in the outermost IPv4
3997 - ``ipv4_addr``: New IPv4 destination address.
3999 - ``set_ipv6_src``: Set a new IPv6 source address in the outermost IPv6 header.
4001 - ``ipv6_addr``: New IPv6 source address.
4003 - ``set_ipv6_dst``: Set a new IPv6 destination address in the outermost IPv6
4006 - ``ipv6_addr``: New IPv6 destination address.
4008 - ``set_tp_src``: Set a new source port number in the outermost TCP/UDP
4011 - ``port``: New TCP/UDP source port number.
4013 - ``set_tp_dst``: Set a new destination port number in the outermost TCP/UDP
4016 - ``port``: New TCP/UDP destination port number.
4018 - ``mac_swap``: Swap the source and destination MAC addresses in the outermost
4021 - ``dec_ttl``: Performs a decrease TTL value action
4023 - ``set_ttl``: Set TTL value with specified value
4024 - ``ttl_value {unsigned}``: The new TTL value to be set
4026 - ``set_mac_src``: set source MAC address
4028 - ``mac_addr {MAC-48}``: new source MAC address
4030 - ``set_mac_dst``: set destination MAC address
4032 - ``mac_addr {MAC-48}``: new destination MAC address
4034 - ``inc_tcp_seq``: Increase sequence number in the outermost TCP header.
4036 - ``value {unsigned}``: Value to increase TCP sequence number by.
4038 - ``dec_tcp_seq``: Decrease sequence number in the outermost TCP header.
4040 - ``value {unsigned}``: Value to decrease TCP sequence number by.
4042 - ``inc_tcp_ack``: Increase acknowledgment number in the outermost TCP header.
4044 - ``value {unsigned}``: Value to increase TCP acknowledgment number by.
4046 - ``dec_tcp_ack``: Decrease acknowledgment number in the outermost TCP header.
4048 - ``value {unsigned}``: Value to decrease TCP acknowledgment number by.
4050 - ``set_ipv4_dscp``: Set IPv4 DSCP value with specified value
4052 - ``dscp_value {unsigned}``: The new DSCP value to be set
4054 - ``set_ipv6_dscp``: Set IPv6 DSCP value with specified value
4056 - ``dscp_value {unsigned}``: The new DSCP value to be set
4058 - ``shared``: Use shared action created via
4059 ``flow shared_action {port_id} create``
4061 - ``shared_action_id {unsigned}``: Shared action ID to use
4063 Destroying flow rules
4064 ~~~~~~~~~~~~~~~~~~~~~
4066 ``flow destroy`` destroys one or more rules from their rule ID (as returned
4067 by ``flow create``), this command calls ``rte_flow_destroy()`` as many
4068 times as necessary::
4070 flow destroy {port_id} rule {rule_id} [...]
4072 If successful, it will show::
4074 Flow rule #[...] destroyed
4076 It does not report anything for rule IDs that do not exist. The usual error
4077 message is shown when a rule cannot be destroyed::
4079 Caught error type [...] ([...]): [...]
4081 ``flow flush`` destroys all rules on a device and does not take extra
4082 arguments. It is bound to ``rte_flow_flush()``::
4084 flow flush {port_id}
4086 Any errors are reported as above.
4088 Creating several rules and destroying them::
4090 testpmd> flow create 0 ingress pattern eth / ipv6 / end
4091 actions queue index 2 / end
4092 Flow rule #0 created
4093 testpmd> flow create 0 ingress pattern eth / ipv4 / end
4094 actions queue index 3 / end
4095 Flow rule #1 created
4096 testpmd> flow destroy 0 rule 0 rule 1
4097 Flow rule #1 destroyed
4098 Flow rule #0 destroyed
4101 The same result can be achieved using ``flow flush``::
4103 testpmd> flow create 0 ingress pattern eth / ipv6 / end
4104 actions queue index 2 / end
4105 Flow rule #0 created
4106 testpmd> flow create 0 ingress pattern eth / ipv4 / end
4107 actions queue index 3 / end
4108 Flow rule #1 created
4109 testpmd> flow flush 0
4112 Non-existent rule IDs are ignored::
4114 testpmd> flow create 0 ingress pattern eth / ipv6 / end
4115 actions queue index 2 / end
4116 Flow rule #0 created
4117 testpmd> flow create 0 ingress pattern eth / ipv4 / end
4118 actions queue index 3 / end
4119 Flow rule #1 created
4120 testpmd> flow destroy 0 rule 42 rule 10 rule 2
4122 testpmd> flow destroy 0 rule 0
4123 Flow rule #0 destroyed
4129 ``flow query`` queries a specific action of a flow rule having that
4130 ability. Such actions collect information that can be reported using this
4131 command. It is bound to ``rte_flow_query()``::
4133 flow query {port_id} {rule_id} {action}
4135 If successful, it will display either the retrieved data for known actions
4136 or the following message::
4138 Cannot display result for action type [...] ([...])
4140 Otherwise, it will complain either that the rule does not exist or that some
4143 Flow rule #[...] not found
4147 Caught error type [...] ([...]): [...]
4149 Currently only the ``count`` action is supported. This action reports the
4150 number of packets that hit the flow rule and the total number of bytes. Its
4151 output has the following format::
4154 hits_set: [...] # whether "hits" contains a valid value
4155 bytes_set: [...] # whether "bytes" contains a valid value
4156 hits: [...] # number of packets
4157 bytes: [...] # number of bytes
4159 Querying counters for TCPv6 packets redirected to queue 6::
4161 testpmd> flow create 0 ingress pattern eth / ipv6 / tcp / end
4162 actions queue index 6 / count / end
4163 Flow rule #4 created
4164 testpmd> flow query 0 4 count
4175 ``flow list`` lists existing flow rules sorted by priority and optionally
4176 filtered by group identifiers::
4178 flow list {port_id} [group {group_id}] [...]
4180 This command only fails with the following message if the device does not
4185 Output consists of a header line followed by a short description of each
4186 flow rule, one per line. There is no output at all when no flow rules are
4187 configured on the device::
4189 ID Group Prio Attr Rule
4190 [...] [...] [...] [...] [...]
4192 ``Attr`` column flags:
4194 - ``i`` for ``ingress``.
4195 - ``e`` for ``egress``.
4197 Creating several flow rules and listing them::
4199 testpmd> flow create 0 ingress pattern eth / ipv4 / end
4200 actions queue index 6 / end
4201 Flow rule #0 created
4202 testpmd> flow create 0 ingress pattern eth / ipv6 / end
4203 actions queue index 2 / end
4204 Flow rule #1 created
4205 testpmd> flow create 0 priority 5 ingress pattern eth / ipv4 / udp / end
4206 actions rss queues 6 7 8 end / end
4207 Flow rule #2 created
4208 testpmd> flow list 0
4209 ID Group Prio Attr Rule
4210 0 0 0 i- ETH IPV4 => QUEUE
4211 1 0 0 i- ETH IPV6 => QUEUE
4212 2 0 5 i- ETH IPV4 UDP => RSS
4215 Rules are sorted by priority (i.e. group ID first, then priority level)::
4217 testpmd> flow list 1
4218 ID Group Prio Attr Rule
4219 0 0 0 i- ETH => COUNT
4220 6 0 500 i- ETH IPV6 TCP => DROP COUNT
4221 5 0 1000 i- ETH IPV6 ICMP => QUEUE
4222 1 24 0 i- ETH IPV4 UDP => QUEUE
4223 4 24 10 i- ETH IPV4 TCP => DROP
4224 3 24 20 i- ETH IPV4 => DROP
4225 2 24 42 i- ETH IPV4 UDP => QUEUE
4226 7 63 0 i- ETH IPV6 UDP VXLAN => MARK QUEUE
4229 Output can be limited to specific groups::
4231 testpmd> flow list 1 group 0 group 63
4232 ID Group Prio Attr Rule
4233 0 0 0 i- ETH => COUNT
4234 6 0 500 i- ETH IPV6 TCP => DROP COUNT
4235 5 0 1000 i- ETH IPV6 ICMP => QUEUE
4236 7 63 0 i- ETH IPV6 UDP VXLAN => MARK QUEUE
4239 Toggling isolated mode
4240 ~~~~~~~~~~~~~~~~~~~~~~
4242 ``flow isolate`` can be used to tell the underlying PMD that ingress traffic
4243 must only be injected from the defined flow rules; that no default traffic
4244 is expected outside those rules and the driver is free to assign more
4245 resources to handle them. It is bound to ``rte_flow_isolate()``::
4247 flow isolate {port_id} {boolean}
4249 If successful, enabling or disabling isolated mode shows either::
4251 Ingress traffic on port [...]
4252 is now restricted to the defined flow rules
4256 Ingress traffic on port [...]
4257 is not restricted anymore to the defined flow rules
4259 Otherwise, in case of error::
4261 Caught error type [...] ([...]): [...]
4263 Mainly due to its side effects, PMDs supporting this mode may not have the
4264 ability to toggle it more than once without reinitializing affected ports
4265 first (e.g. by exiting testpmd).
4267 Enabling isolated mode::
4269 testpmd> flow isolate 0 true
4270 Ingress traffic on port 0 is now restricted to the defined flow rules
4273 Disabling isolated mode::
4275 testpmd> flow isolate 0 false
4276 Ingress traffic on port 0 is not restricted anymore to the defined flow rules
4279 Dumping HW internal information
4280 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4282 ``flow dump`` dumps the hardware's internal representation information of
4283 all flows. It is bound to ``rte_flow_dev_dump()``::
4285 flow dump {port_id} {output_file}
4287 If successful, it will show::
4291 Otherwise, it will complain error occurred::
4293 Caught error type [...] ([...]): [...]
4295 Listing and destroying aged flow rules
4296 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4298 ``flow aged`` simply lists aged flow rules be get from api ``rte_flow_get_aged_flows``,
4299 and ``destroy`` parameter can be used to destroy those flow rules in PMD.
4301 flow aged {port_id} [destroy]
4303 Listing current aged flow rules::
4305 testpmd> flow aged 0
4306 Port 0 total aged flows: 0
4307 testpmd> flow create 0 ingress pattern eth / ipv4 src is 2.2.2.14 / end
4308 actions age timeout 5 / queue index 0 / end
4309 Flow rule #0 created
4310 testpmd> flow create 0 ingress pattern eth / ipv4 src is 2.2.2.15 / end
4311 actions age timeout 4 / queue index 0 / end
4312 Flow rule #1 created
4313 testpmd> flow create 0 ingress pattern eth / ipv4 src is 2.2.2.16 / end
4314 actions age timeout 2 / queue index 0 / end
4315 Flow rule #2 created
4316 testpmd> flow create 0 ingress pattern eth / ipv4 src is 2.2.2.17 / end
4317 actions age timeout 3 / queue index 0 / end
4318 Flow rule #3 created
4321 Aged Rules are simply list as command ``flow list {port_id}``, but strip the detail rule
4322 information, all the aged flows are sorted by the longest timeout time. For example, if
4323 those rules be configured in the same time, ID 2 will be the first aged out rule, the next
4324 will be ID 3, ID 1, ID 0::
4326 testpmd> flow aged 0
4327 Port 0 total aged flows: 4
4334 If attach ``destroy`` parameter, the command will destroy all the list aged flow rules.
4336 testpmd> flow aged 0 destroy
4337 Port 0 total aged flows: 4
4344 Flow rule #2 destroyed
4345 Flow rule #3 destroyed
4346 Flow rule #1 destroyed
4347 Flow rule #0 destroyed
4348 4 flows be destroyed
4349 testpmd> flow aged 0
4350 Port 0 total aged flows: 0
4352 Creating shared actions
4353 ~~~~~~~~~~~~~~~~~~~~~~~
4354 ``flow shared_action {port_id} create`` creates shared action with optional
4355 shared action ID. It is bound to ``rte_flow_shared_action_create()``::
4357 flow shared_action {port_id} create [action_id {shared_action_id}]
4358 [ingress] [egress] [transfer] action {action} / end
4360 If successful, it will show::
4362 Shared action #[...] created
4364 Otherwise, it will complain either that shared action already exists or that
4365 some error occurred::
4367 Shared action #[...] is already assigned, delete it first
4371 Caught error type [...] ([...]): [...]
4373 Create shared rss action with id 100 to queues 1 and 2 on port 0::
4375 testpmd> flow shared_action 0 create action_id 100 \
4376 ingress action rss queues 1 2 end / end
4378 Create shared rss action with id assigned by testpmd to queues 1 and 2 on
4381 testpmd> flow shared_action 0 create action_id \
4382 ingress action rss queues 0 1 end / end
4384 Updating shared actions
4385 ~~~~~~~~~~~~~~~~~~~~~~~
4386 ``flow shared_action {port_id} update`` updates configuration of the shared
4387 action from its shared action ID (as returned by
4388 ``flow shared_action {port_id} create``). It is bound to
4389 ``rte_flow_shared_action_update()``::
4391 flow shared_action {port_id} update {shared_action_id}
4392 action {action} / end
4394 If successful, it will show::
4396 Shared action #[...] updated
4398 Otherwise, it will complain either that shared action not found or that some
4401 Failed to find shared action #[...] on port [...]
4405 Caught error type [...] ([...]): [...]
4407 Update shared rss action having id 100 on port 0 with rss to queues 0 and 3
4408 (in create example above rss queues were 1 and 2)::
4410 testpmd> flow shared_action 0 update 100 action rss queues 0 3 end / end
4412 Destroying shared actions
4413 ~~~~~~~~~~~~~~~~~~~~~~~~~
4414 ``flow shared_action {port_id} update`` destroys one or more shared actions
4415 from their shared action IDs (as returned by
4416 ``flow shared_action {port_id} create``). It is bound to
4417 ``rte_flow_shared_action_destroy()``::
4419 flow shared_action {port_id} destroy action_id {shared_action_id} [...]
4421 If successful, it will show::
4423 Shared action #[...] destroyed
4425 It does not report anything for shared action IDs that do not exist.
4426 The usual error message is shown when a shared action cannot be destroyed::
4428 Caught error type [...] ([...]): [...]
4430 Destroy shared actions having id 100 & 101::
4432 testpmd> flow shared_action 0 destroy action_id 100 action_id 101
4434 Query shared actions
4435 ~~~~~~~~~~~~~~~~~~~~
4436 ``flow shared_action {port_id} query`` queries the shared action from its
4437 shared action ID (as returned by ``flow shared_action {port_id} create``).
4438 It is bound to ``rte_flow_shared_action_query()``::
4440 flow shared_action {port_id} query {shared_action_id}
4442 Currently only rss shared action supported. If successful, it will show::
4447 Otherwise, it will complain either that shared action not found or that some
4450 Failed to find shared action #[...] on port [...]
4454 Caught error type [...] ([...]): [...]
4456 Query shared action having id 100::
4458 testpmd> flow shared_action 0 query 100
4460 Sample QinQ flow rules
4461 ~~~~~~~~~~~~~~~~~~~~~~
4463 Before creating QinQ rule(s) the following commands should be issued to enable QinQ::
4465 testpmd> port stop 0
4466 testpmd> vlan set extend on 0
4468 The above command sets the inner and outer TPID's to 0x8100.
4470 To change the TPID's the following commands should be used::
4472 testpmd> vlan set outer tpid 0x88A8 0
4473 testpmd> vlan set inner tpid 0x8100 0
4474 testpmd> port start 0
4476 Validate and create a QinQ rule on port 0 to steer traffic to a VF queue in a VM.
4480 testpmd> flow validate 0 ingress pattern eth / vlan tci is 123 /
4481 vlan tci is 456 / end actions vf id 1 / queue index 0 / end
4482 Flow rule #0 validated
4484 testpmd> flow create 0 ingress pattern eth / vlan tci is 4 /
4485 vlan tci is 456 / end actions vf id 123 / queue index 0 / end
4486 Flow rule #0 created
4488 testpmd> flow list 0
4489 ID Group Prio Attr Rule
4490 0 0 0 i- ETH VLAN VLAN=>VF QUEUE
4492 Validate and create a QinQ rule on port 0 to steer traffic to a queue on the host.
4496 testpmd> flow validate 0 ingress pattern eth / vlan tci is 321 /
4497 vlan tci is 654 / end actions pf / queue index 0 / end
4498 Flow rule #1 validated
4500 testpmd> flow create 0 ingress pattern eth / vlan tci is 321 /
4501 vlan tci is 654 / end actions pf / queue index 1 / end
4502 Flow rule #1 created
4504 testpmd> flow list 0
4505 ID Group Prio Attr Rule
4506 0 0 0 i- ETH VLAN VLAN=>VF QUEUE
4507 1 0 0 i- ETH VLAN VLAN=>PF QUEUE
4509 Sample VXLAN flow rules
4510 ~~~~~~~~~~~~~~~~~~~~~~~
4512 Before creating VXLAN rule(s), the UDP port should be added for VXLAN packet
4515 testpmd> rx_vxlan_port add 4789 0
4517 Create VXLAN rules on port 0 to steer traffic to PF queues.
4521 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / vxlan /
4522 eth dst is 00:11:22:33:44:55 / end actions pf / queue index 1 / end
4523 Flow rule #0 created
4525 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / vxlan vni is 3 /
4526 eth dst is 00:11:22:33:44:55 / end actions pf / queue index 2 / end
4527 Flow rule #1 created
4529 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / vxlan /
4530 eth dst is 00:11:22:33:44:55 / vlan tci is 10 / end actions pf /
4532 Flow rule #2 created
4534 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / vxlan vni is 5 /
4535 eth dst is 00:11:22:33:44:55 / vlan tci is 20 / end actions pf /
4537 Flow rule #3 created
4539 testpmd> flow create 0 ingress pattern eth dst is 00:00:00:00:01:00 / ipv4 /
4540 udp / vxlan vni is 6 / eth dst is 00:11:22:33:44:55 / end actions pf /
4542 Flow rule #4 created
4544 testpmd> flow list 0
4545 ID Group Prio Attr Rule
4546 0 0 0 i- ETH IPV4 UDP VXLAN ETH => QUEUE
4547 1 0 0 i- ETH IPV4 UDP VXLAN ETH => QUEUE
4548 2 0 0 i- ETH IPV4 UDP VXLAN ETH VLAN => QUEUE
4549 3 0 0 i- ETH IPV4 UDP VXLAN ETH VLAN => QUEUE
4550 4 0 0 i- ETH IPV4 UDP VXLAN ETH => QUEUE
4552 Sample VXLAN encapsulation rule
4553 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4555 VXLAN encapsulation outer layer has default value pre-configured in testpmd
4556 source code, those can be changed by using the following commands
4558 IPv4 VXLAN outer header::
4560 testpmd> set vxlan ip-version ipv4 vni 4 udp-src 4 udp-dst 4 ip-src 127.0.0.1
4561 ip-dst 128.0.0.1 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4562 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4565 testpmd> set vxlan-with-vlan ip-version ipv4 vni 4 udp-src 4 udp-dst 4 ip-src
4566 127.0.0.1 ip-dst 128.0.0.1 vlan-tci 34 eth-src 11:11:11:11:11:11
4567 eth-dst 22:22:22:22:22:22
4568 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4571 testpmd> set vxlan-tos-ttl ip-version ipv4 vni 4 udp-src 4 udp-dst 4 ip-tos 0
4572 ip-ttl 255 ip-src 127.0.0.1 ip-dst 128.0.0.1 eth-src 11:11:11:11:11:11
4573 eth-dst 22:22:22:22:22:22
4574 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4577 IPv6 VXLAN outer header::
4579 testpmd> set vxlan ip-version ipv6 vni 4 udp-src 4 udp-dst 4 ip-src ::1
4580 ip-dst ::2222 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4581 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4584 testpmd> set vxlan-with-vlan ip-version ipv6 vni 4 udp-src 4 udp-dst 4
4585 ip-src ::1 ip-dst ::2222 vlan-tci 34 eth-src 11:11:11:11:11:11
4586 eth-dst 22:22:22:22:22:22
4587 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4590 testpmd> set vxlan-tos-ttl ip-version ipv6 vni 4 udp-src 4 udp-dst 4
4591 ip-tos 0 ip-ttl 255 ::1 ip-dst ::2222 eth-src 11:11:11:11:11:11
4592 eth-dst 22:22:22:22:22:22
4593 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4596 Sample NVGRE encapsulation rule
4597 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4599 NVGRE encapsulation outer layer has default value pre-configured in testpmd
4600 source code, those can be changed by using the following commands
4602 IPv4 NVGRE outer header::
4604 testpmd> set nvgre ip-version ipv4 tni 4 ip-src 127.0.0.1 ip-dst 128.0.0.1
4605 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4606 testpmd> flow create 0 ingress pattern end actions nvgre_encap /
4609 testpmd> set nvgre-with-vlan ip-version ipv4 tni 4 ip-src 127.0.0.1
4610 ip-dst 128.0.0.1 vlan-tci 34 eth-src 11:11:11:11:11:11
4611 eth-dst 22:22:22:22:22:22
4612 testpmd> flow create 0 ingress pattern end actions nvgre_encap /
4615 IPv6 NVGRE outer header::
4617 testpmd> set nvgre ip-version ipv6 tni 4 ip-src ::1 ip-dst ::2222
4618 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4619 testpmd> flow create 0 ingress pattern end actions nvgre_encap /
4622 testpmd> set nvgre-with-vlan ip-version ipv6 tni 4 ip-src ::1 ip-dst ::2222
4623 vlan-tci 34 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4624 testpmd> flow create 0 ingress pattern end actions nvgre_encap /
4627 Sample L2 encapsulation rule
4628 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4630 L2 encapsulation has default value pre-configured in testpmd
4631 source code, those can be changed by using the following commands
4635 testpmd> set l2_encap ip-version ipv4
4636 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4637 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / mpls / end actions
4638 mplsoudp_decap / l2_encap / end
4640 L2 with VXLAN header::
4642 testpmd> set l2_encap-with-vlan ip-version ipv4 vlan-tci 34
4643 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4644 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / mpls / end actions
4645 mplsoudp_decap / l2_encap / end
4647 Sample L2 decapsulation rule
4648 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4650 L2 decapsulation has default value pre-configured in testpmd
4651 source code, those can be changed by using the following commands
4655 testpmd> set l2_decap
4656 testpmd> flow create 0 egress pattern eth / end actions l2_decap / mplsoudp_encap /
4659 L2 with VXLAN header::
4661 testpmd> set l2_encap-with-vlan
4662 testpmd> flow create 0 egress pattern eth / end actions l2_encap / mplsoudp_encap /
4665 Sample MPLSoGRE encapsulation rule
4666 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4668 MPLSoGRE encapsulation outer layer has default value pre-configured in testpmd
4669 source code, those can be changed by using the following commands
4671 IPv4 MPLSoGRE outer header::
4673 testpmd> set mplsogre_encap ip-version ipv4 label 4
4674 ip-src 127.0.0.1 ip-dst 128.0.0.1 eth-src 11:11:11:11:11:11
4675 eth-dst 22:22:22:22:22:22
4676 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4677 mplsogre_encap / end
4679 IPv4 MPLSoGRE with VLAN outer header::
4681 testpmd> set mplsogre_encap-with-vlan ip-version ipv4 label 4
4682 ip-src 127.0.0.1 ip-dst 128.0.0.1 vlan-tci 34
4683 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4684 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4685 mplsogre_encap / end
4687 IPv6 MPLSoGRE outer header::
4689 testpmd> set mplsogre_encap ip-version ipv6 mask 4
4690 ip-src ::1 ip-dst ::2222 eth-src 11:11:11:11:11:11
4691 eth-dst 22:22:22:22:22:22
4692 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4693 mplsogre_encap / end
4695 IPv6 MPLSoGRE with VLAN outer header::
4697 testpmd> set mplsogre_encap-with-vlan ip-version ipv6 mask 4
4698 ip-src ::1 ip-dst ::2222 vlan-tci 34
4699 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4700 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4701 mplsogre_encap / end
4703 Sample MPLSoGRE decapsulation rule
4704 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4706 MPLSoGRE decapsulation outer layer has default value pre-configured in testpmd
4707 source code, those can be changed by using the following commands
4709 IPv4 MPLSoGRE outer header::
4711 testpmd> set mplsogre_decap ip-version ipv4
4712 testpmd> flow create 0 ingress pattern eth / ipv4 / gre / mpls / end actions
4713 mplsogre_decap / l2_encap / end
4715 IPv4 MPLSoGRE with VLAN outer header::
4717 testpmd> set mplsogre_decap-with-vlan ip-version ipv4
4718 testpmd> flow create 0 ingress pattern eth / vlan / ipv4 / gre / mpls / end
4719 actions mplsogre_decap / l2_encap / end
4721 IPv6 MPLSoGRE outer header::
4723 testpmd> set mplsogre_decap ip-version ipv6
4724 testpmd> flow create 0 ingress pattern eth / ipv6 / gre / mpls / end
4725 actions mplsogre_decap / l2_encap / end
4727 IPv6 MPLSoGRE with VLAN outer header::
4729 testpmd> set mplsogre_decap-with-vlan ip-version ipv6
4730 testpmd> flow create 0 ingress pattern eth / vlan / ipv6 / gre / mpls / end
4731 actions mplsogre_decap / l2_encap / end
4733 Sample MPLSoUDP encapsulation rule
4734 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4736 MPLSoUDP encapsulation outer layer has default value pre-configured in testpmd
4737 source code, those can be changed by using the following commands
4739 IPv4 MPLSoUDP outer header::
4741 testpmd> set mplsoudp_encap ip-version ipv4 label 4 udp-src 5 udp-dst 10
4742 ip-src 127.0.0.1 ip-dst 128.0.0.1 eth-src 11:11:11:11:11:11
4743 eth-dst 22:22:22:22:22:22
4744 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4745 mplsoudp_encap / end
4747 IPv4 MPLSoUDP with VLAN outer header::
4749 testpmd> set mplsoudp_encap-with-vlan ip-version ipv4 label 4 udp-src 5
4750 udp-dst 10 ip-src 127.0.0.1 ip-dst 128.0.0.1 vlan-tci 34
4751 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4752 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4753 mplsoudp_encap / end
4755 IPv6 MPLSoUDP outer header::
4757 testpmd> set mplsoudp_encap ip-version ipv6 mask 4 udp-src 5 udp-dst 10
4758 ip-src ::1 ip-dst ::2222 eth-src 11:11:11:11:11:11
4759 eth-dst 22:22:22:22:22:22
4760 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4761 mplsoudp_encap / end
4763 IPv6 MPLSoUDP with VLAN outer header::
4765 testpmd> set mplsoudp_encap-with-vlan ip-version ipv6 mask 4 udp-src 5
4766 udp-dst 10 ip-src ::1 ip-dst ::2222 vlan-tci 34
4767 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4768 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4769 mplsoudp_encap / end
4771 Sample MPLSoUDP decapsulation rule
4772 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4774 MPLSoUDP decapsulation outer layer has default value pre-configured in testpmd
4775 source code, those can be changed by using the following commands
4777 IPv4 MPLSoUDP outer header::
4779 testpmd> set mplsoudp_decap ip-version ipv4
4780 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / mpls / end actions
4781 mplsoudp_decap / l2_encap / end
4783 IPv4 MPLSoUDP with VLAN outer header::
4785 testpmd> set mplsoudp_decap-with-vlan ip-version ipv4
4786 testpmd> flow create 0 ingress pattern eth / vlan / ipv4 / udp / mpls / end
4787 actions mplsoudp_decap / l2_encap / end
4789 IPv6 MPLSoUDP outer header::
4791 testpmd> set mplsoudp_decap ip-version ipv6
4792 testpmd> flow create 0 ingress pattern eth / ipv6 / udp / mpls / end
4793 actions mplsoudp_decap / l2_encap / end
4795 IPv6 MPLSoUDP with VLAN outer header::
4797 testpmd> set mplsoudp_decap-with-vlan ip-version ipv6
4798 testpmd> flow create 0 ingress pattern eth / vlan / ipv6 / udp / mpls / end
4799 actions mplsoudp_decap / l2_encap / end
4801 Sample Raw encapsulation rule
4802 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4804 Raw encapsulation configuration can be set by the following commands
4806 Eecapsulating VxLAN::
4808 testpmd> set raw_encap 4 eth src is 10:11:22:33:44:55 / vlan tci is 1
4809 inner_type is 0x0800 / ipv4 / udp dst is 4789 / vxlan vni
4811 testpmd> flow create 0 egress pattern eth / ipv4 / end actions
4812 raw_encap index 4 / end
4814 Sample Raw decapsulation rule
4815 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4817 Raw decapsulation configuration can be set by the following commands
4819 Decapsulating VxLAN::
4821 testpmd> set raw_decap eth / ipv4 / udp / vxlan / end_set
4822 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / vxlan / eth / ipv4 /
4823 end actions raw_decap / queue index 0 / end
4828 ESP rules can be created by the following commands::
4830 testpmd> flow create 0 ingress pattern eth / ipv4 / esp spi is 1 / end actions
4832 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / esp spi is 1 / end
4833 actions queue index 3 / end
4834 testpmd> flow create 0 ingress pattern eth / ipv6 / esp spi is 1 / end actions
4836 testpmd> flow create 0 ingress pattern eth / ipv6 / udp / esp spi is 1 / end
4837 actions queue index 3 / end
4842 AH rules can be created by the following commands::
4844 testpmd> flow create 0 ingress pattern eth / ipv4 / ah spi is 1 / end actions
4846 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / ah spi is 1 / end
4847 actions queue index 3 / end
4848 testpmd> flow create 0 ingress pattern eth / ipv6 / ah spi is 1 / end actions
4850 testpmd> flow create 0 ingress pattern eth / ipv6 / udp / ah spi is 1 / end
4851 actions queue index 3 / end
4856 PFCP rules can be created by the following commands(s_field need to be 1
4859 testpmd> flow create 0 ingress pattern eth / ipv4 / pfcp s_field is 0 / end
4860 actions queue index 3 / end
4861 testpmd> flow create 0 ingress pattern eth / ipv4 / pfcp s_field is 1
4862 seid is 1 / end actions queue index 3 / end
4863 testpmd> flow create 0 ingress pattern eth / ipv6 / pfcp s_field is 0 / end
4864 actions queue index 3 / end
4865 testpmd> flow create 0 ingress pattern eth / ipv6 / pfcp s_field is 1
4866 seid is 1 / end actions queue index 3 / end
4868 Sample Sampling/Mirroring rules
4869 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4871 Sample/Mirroring rules can be set by the following commands
4873 NIC-RX Sampling rule, the matched ingress packets and sent to the queue 1,
4874 and 50% packets are duplicated and marked with 0x1234 and sent to queue 0.
4878 testpmd> set sample_actions 0 mark id 0x1234 / queue index 0 / end
4879 testpmd> flow create 0 ingress group 1 pattern eth / end actions
4880 sample ratio 2 index 0 / queue index 1 / end
4882 Mirroring rule with port representors (with "transfer" attribute), the matched
4883 ingress packets with encapsulation header are sent to port id 0, and also
4884 mirrored the packets and sent to port id 2.
4888 testpmd> set sample_actions 0 port_id id 2 / end
4889 testpmd> flow create 1 ingress transfer pattern eth / end actions
4890 sample ratio 1 index 0 / raw_encap / port_id id 0 / end
4892 Mirroring rule with port representors (with "transfer" attribute), the matched
4893 ingress packets are sent to port id 2, and also mirrored the packets with
4894 encapsulation header and sent to port id 0.
4898 testpmd> set sample_actions 0 raw_encap / port_id id 0 / end
4899 testpmd> flow create 0 ingress transfer pattern eth / end actions
4900 sample ratio 1 index 0 / port_id id 2 / end
4902 Mirroring rule with port representors (with "transfer" attribute), the matched
4903 ingress packets are sent to port id 2, and also mirrored the packets with
4904 VXLAN encapsulation header and sent to port id 0.
4908 testpmd> set vxlan ip-version ipv4 vni 4 udp-src 4 udp-dst 4 ip-src 127.0.0.1
4909 ip-dst 128.0.0.1 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4910 testpmd> set sample_actions 0 vxlan_encap / port_id id 0 / end
4911 testpmd> flow create 0 ingress transfer pattern eth / end actions
4912 sample ratio 1 index 0 / port_id id 2 / end
4914 Mirroring rule with port representors (with "transfer" attribute), the matched
4915 ingress packets are sent to port id 2, and also mirrored the packets with
4916 NVGRE encapsulation header and sent to port id 0.
4920 testpmd> set nvgre ip-version ipv4 tni 4 ip-src 127.0.0.1 ip-dst 128.0.0.1
4921 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4922 testpmd> set sample_actions 0 nvgre_encap / port_id id 0 / end
4923 testpmd> flow create 0 ingress transfer pattern eth / end actions
4924 sample ratio 1 index 0 / port_id id 2 / end
4929 The following sections show functions to load/unload eBPF based filters.
4934 Load an eBPF program as a callback for particular RX/TX queue::
4936 testpmd> bpf-load rx|tx (portid) (queueid) (load-flags) (bpf-prog-filename)
4938 The available load-flags are:
4940 * ``J``: use JIT generated native code, otherwise BPF interpreter will be used.
4942 * ``M``: assume input parameter is a pointer to rte_mbuf, otherwise assume it is a pointer to first segment's data.
4948 You'll need clang v3.7 or above to build bpf program you'd like to load
4952 .. code-block:: console
4955 clang -O2 -target bpf -c t1.c
4957 Then to load (and JIT compile) t1.o at RX queue 0, port 1:
4959 .. code-block:: console
4961 testpmd> bpf-load rx 1 0 J ./dpdk.org/examples/bpf/t1.o
4963 To load (not JITed) t1.o at TX queue 0, port 0:
4965 .. code-block:: console
4967 testpmd> bpf-load tx 0 0 - ./dpdk.org/examples/bpf/t1.o
4972 Unload previously loaded eBPF program for particular RX/TX queue::
4974 testpmd> bpf-unload rx|tx (portid) (queueid)
4976 For example to unload BPF filter from TX queue 0, port 0:
4978 .. code-block:: console
4980 testpmd> bpf-unload tx 0 0