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 stats mask
2835 ~~~~~~~~~~~~~~~~~~~~~~~~~
2837 Set meter stats mask for the ethernet device::
2839 testpmd> set port meter stats mask (port_id) (mtr_id) (stats_mask)
2843 * ``stats_mask``: Bit mask indicating statistics counter types to be enabled.
2845 show port meter stats
2846 ~~~~~~~~~~~~~~~~~~~~~
2848 Show meter stats of the ethernet device::
2850 testpmd> show port meter stats (port_id) (mtr_id) (clear)
2854 * ``clear``: Flag that indicates whether the statistics counters should
2855 be cleared (i.e. set to zero) immediately after they have been read or not.
2860 The following section shows functions for configuring traffic management on
2861 the ethernet device through the use of generic TM API.
2863 show port traffic management capability
2864 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2866 Show traffic management capability of the port::
2868 testpmd> show port tm cap (port_id)
2870 show port traffic management capability (hierarchy level)
2871 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2873 Show traffic management hierarchy level capability of the port::
2875 testpmd> show port tm level cap (port_id) (level_id)
2877 show port traffic management capability (hierarchy node level)
2878 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2880 Show the traffic management hierarchy node capability of the port::
2882 testpmd> show port tm node cap (port_id) (node_id)
2884 show port traffic management hierarchy node type
2885 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2887 Show the port traffic management hierarchy node type::
2889 testpmd> show port tm node type (port_id) (node_id)
2891 show port traffic management hierarchy node stats
2892 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2894 Show the port traffic management hierarchy node statistics::
2896 testpmd> show port tm node stats (port_id) (node_id) (clear)
2900 * ``clear``: When this parameter has a non-zero value, the statistics counters
2901 are cleared (i.e. set to zero) immediately after they have been read,
2902 otherwise the statistics counters are left untouched.
2904 Add port traffic management private shaper profile
2905 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2907 Add the port traffic management private shaper profile::
2909 testpmd> add port tm node shaper profile (port_id) (shaper_profile_id) \
2910 (cmit_tb_rate) (cmit_tb_size) (peak_tb_rate) (peak_tb_size) \
2911 (packet_length_adjust) (packet_mode)
2915 * ``shaper_profile id``: Shaper profile ID for the new profile.
2916 * ``cmit_tb_rate``: Committed token bucket rate (bytes per second or packets per second).
2917 * ``cmit_tb_size``: Committed token bucket size (bytes or packets).
2918 * ``peak_tb_rate``: Peak token bucket rate (bytes per second or packets per second).
2919 * ``peak_tb_size``: Peak token bucket size (bytes or packets).
2920 * ``packet_length_adjust``: The value (bytes) to be added to the length of
2921 each packet for the purpose of shaping. This parameter value can be used to
2922 correct the packet length with the framing overhead bytes that are consumed
2924 * ``packet_mode``: Shaper configured in packet mode. This parameter value if
2925 zero, configures shaper in byte mode and if non-zero configures it in packet
2928 Delete port traffic management private shaper profile
2929 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2931 Delete the port traffic management private shaper::
2933 testpmd> del port tm node shaper profile (port_id) (shaper_profile_id)
2937 * ``shaper_profile id``: Shaper profile ID that needs to be deleted.
2939 Add port traffic management shared shaper
2940 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2942 Create the port traffic management shared shaper::
2944 testpmd> add port tm node shared shaper (port_id) (shared_shaper_id) \
2949 * ``shared_shaper_id``: Shared shaper ID to be created.
2950 * ``shaper_profile id``: Shaper profile ID for shared shaper.
2952 Set port traffic management shared shaper
2953 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2955 Update the port traffic management shared shaper::
2957 testpmd> set port tm node shared shaper (port_id) (shared_shaper_id) \
2962 * ``shared_shaper_id``: Shared shaper ID to be update.
2963 * ``shaper_profile id``: Shaper profile ID for shared shaper.
2965 Delete port traffic management shared shaper
2966 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2968 Delete the port traffic management shared shaper::
2970 testpmd> del port tm node shared shaper (port_id) (shared_shaper_id)
2974 * ``shared_shaper_id``: Shared shaper ID to be deleted.
2976 Set port traffic management hierarchy node private shaper
2977 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2979 set the port traffic management hierarchy node private shaper::
2981 testpmd> set port tm node shaper profile (port_id) (node_id) \
2986 * ``shaper_profile id``: Private shaper profile ID to be enabled on the
2989 Add port traffic management WRED profile
2990 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2992 Create a new WRED profile::
2994 testpmd> add port tm node wred profile (port_id) (wred_profile_id) \
2995 (color_g) (min_th_g) (max_th_g) (maxp_inv_g) (wq_log2_g) \
2996 (color_y) (min_th_y) (max_th_y) (maxp_inv_y) (wq_log2_y) \
2997 (color_r) (min_th_r) (max_th_r) (maxp_inv_r) (wq_log2_r)
3001 * ``wred_profile id``: Identifier for the newly create WRED profile
3002 * ``color_g``: Packet color (green)
3003 * ``min_th_g``: Minimum queue threshold for packet with green color
3004 * ``max_th_g``: Minimum queue threshold for packet with green color
3005 * ``maxp_inv_g``: Inverse of packet marking probability maximum value (maxp)
3006 * ``wq_log2_g``: Negated log2 of queue weight (wq)
3007 * ``color_y``: Packet color (yellow)
3008 * ``min_th_y``: Minimum queue threshold for packet with yellow color
3009 * ``max_th_y``: Minimum queue threshold for packet with yellow color
3010 * ``maxp_inv_y``: Inverse of packet marking probability maximum value (maxp)
3011 * ``wq_log2_y``: Negated log2 of queue weight (wq)
3012 * ``color_r``: Packet color (red)
3013 * ``min_th_r``: Minimum queue threshold for packet with yellow color
3014 * ``max_th_r``: Minimum queue threshold for packet with yellow color
3015 * ``maxp_inv_r``: Inverse of packet marking probability maximum value (maxp)
3016 * ``wq_log2_r``: Negated log2 of queue weight (wq)
3018 Delete port traffic management WRED profile
3019 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3021 Delete the WRED profile::
3023 testpmd> del port tm node wred profile (port_id) (wred_profile_id)
3025 Add port traffic management hierarchy nonleaf node
3026 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3028 Add nonleaf node to port traffic management hierarchy::
3030 testpmd> add port tm nonleaf node (port_id) (node_id) (parent_node_id) \
3031 (priority) (weight) (level_id) (shaper_profile_id) \
3032 (n_sp_priorities) (stats_mask) (n_shared_shapers) \
3033 [(shared_shaper_0) (shared_shaper_1) ...] \
3037 * ``parent_node_id``: Node ID of the parent.
3038 * ``priority``: Node priority (highest node priority is zero). This is used by
3039 the SP algorithm running on the parent node for scheduling this node.
3040 * ``weight``: Node weight (lowest weight is one). The node weight is relative
3041 to the weight sum of all siblings that have the same priority. It is used by
3042 the WFQ algorithm running on the parent node for scheduling this node.
3043 * ``level_id``: Hierarchy level of the node.
3044 * ``shaper_profile_id``: Shaper profile ID of the private shaper to be used by
3046 * ``n_sp_priorities``: Number of strict priorities.
3047 * ``stats_mask``: Mask of statistics counter types to be enabled for this node.
3048 * ``n_shared_shapers``: Number of shared shapers.
3049 * ``shared_shaper_id``: Shared shaper id.
3051 Add port traffic management hierarchy nonleaf node with packet mode
3052 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3054 Add nonleaf node with packet mode to port traffic management hierarchy::
3056 testpmd> add port tm nonleaf node pktmode (port_id) (node_id) (parent_node_id) \
3057 (priority) (weight) (level_id) (shaper_profile_id) \
3058 (n_sp_priorities) (stats_mask) (n_shared_shapers) \
3059 [(shared_shaper_0) (shared_shaper_1) ...] \
3063 * ``parent_node_id``: Node ID of the parent.
3064 * ``priority``: Node priority (highest node priority is zero). This is used by
3065 the SP algorithm running on the parent node for scheduling this node.
3066 * ``weight``: Node weight (lowest weight is one). The node weight is relative
3067 to the weight sum of all siblings that have the same priority. It is used by
3068 the WFQ algorithm running on the parent node for scheduling this node.
3069 * ``level_id``: Hierarchy level of the node.
3070 * ``shaper_profile_id``: Shaper profile ID of the private shaper to be used by
3072 * ``n_sp_priorities``: Number of strict priorities. Packet mode is enabled on
3074 * ``stats_mask``: Mask of statistics counter types to be enabled for this node.
3075 * ``n_shared_shapers``: Number of shared shapers.
3076 * ``shared_shaper_id``: Shared shaper id.
3078 Add port traffic management hierarchy leaf node
3079 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3081 Add leaf node to port traffic management hierarchy::
3083 testpmd> add port tm leaf node (port_id) (node_id) (parent_node_id) \
3084 (priority) (weight) (level_id) (shaper_profile_id) \
3085 (cman_mode) (wred_profile_id) (stats_mask) (n_shared_shapers) \
3086 [(shared_shaper_id) (shared_shaper_id) ...] \
3090 * ``parent_node_id``: Node ID of the parent.
3091 * ``priority``: Node priority (highest node priority is zero). This is used by
3092 the SP algorithm running on the parent node for scheduling this node.
3093 * ``weight``: Node weight (lowest weight is one). The node weight is relative
3094 to the weight sum of all siblings that have the same priority. It is used by
3095 the WFQ algorithm running on the parent node for scheduling this node.
3096 * ``level_id``: Hierarchy level of the node.
3097 * ``shaper_profile_id``: Shaper profile ID of the private shaper to be used by
3099 * ``cman_mode``: Congestion management mode to be enabled for this node.
3100 * ``wred_profile_id``: WRED profile id to be enabled for this node.
3101 * ``stats_mask``: Mask of statistics counter types to be enabled for this node.
3102 * ``n_shared_shapers``: Number of shared shapers.
3103 * ``shared_shaper_id``: Shared shaper id.
3105 Delete port traffic management hierarchy node
3106 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3108 Delete node from port traffic management hierarchy::
3110 testpmd> del port tm node (port_id) (node_id)
3112 Update port traffic management hierarchy parent node
3113 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3115 Update port traffic management hierarchy parent node::
3117 testpmd> set port tm node parent (port_id) (node_id) (parent_node_id) \
3120 This function can only be called after the hierarchy commit invocation. Its
3121 success depends on the port support for this operation, as advertised through
3122 the port capability set. This function is valid for all nodes of the traffic
3123 management hierarchy except root node.
3125 Suspend port traffic management hierarchy node
3126 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3128 testpmd> suspend port tm node (port_id) (node_id)
3130 Resume port traffic management hierarchy node
3131 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3133 testpmd> resume port tm node (port_id) (node_id)
3135 Commit port traffic management hierarchy
3136 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3138 Commit the traffic management hierarchy on the port::
3140 testpmd> port tm hierarchy commit (port_id) (clean_on_fail)
3144 * ``clean_on_fail``: When set to non-zero, hierarchy is cleared on function
3145 call failure. On the other hand, hierarchy is preserved when this parameter
3148 Set port traffic management mark VLAN dei
3149 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3151 Enables/Disables the traffic management marking on the port for VLAN packets::
3153 testpmd> set port tm mark vlan_dei <port_id> <green> <yellow> <red>
3157 * ``port_id``: The port which on which VLAN packets marked as ``green`` or
3158 ``yellow`` or ``red`` will have dei bit enabled
3160 * ``green`` enable 1, disable 0 marking for dei bit of VLAN packets marked as green
3162 * ``yellow`` enable 1, disable 0 marking for dei bit of VLAN packets marked as yellow
3164 * ``red`` enable 1, disable 0 marking for dei bit of VLAN packets marked as red
3166 Set port traffic management mark IP dscp
3167 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3169 Enables/Disables the traffic management marking on the port for IP dscp packets::
3171 testpmd> set port tm mark ip_dscp <port_id> <green> <yellow> <red>
3175 * ``port_id``: The port which on which IP packets marked as ``green`` or
3176 ``yellow`` or ``red`` will have IP dscp bits updated
3178 * ``green`` enable 1, disable 0 marking IP dscp to low drop precedence for green packets
3180 * ``yellow`` enable 1, disable 0 marking IP dscp to medium drop precedence for yellow packets
3182 * ``red`` enable 1, disable 0 marking IP dscp to high drop precedence for red packets
3184 Set port traffic management mark IP ecn
3185 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3187 Enables/Disables the traffic management marking on the port for IP ecn packets::
3189 testpmd> set port tm mark ip_ecn <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 ecn bits updated
3196 * ``green`` enable 1, disable 0 marking IP ecn for green marked packets with ecn of 2'b01 or 2'b10
3197 to ecn of 2'b11 when IP is caring TCP or SCTP
3199 * ``yellow`` enable 1, disable 0 marking IP ecn for yellow marked packets with ecn of 2'b01 or 2'b10
3200 to ecn of 2'b11 when IP is caring TCP or SCTP
3202 * ``red`` enable 1, disable 0 marking IP ecn for yellow marked packets with ecn of 2'b01 or 2'b10
3203 to ecn of 2'b11 when IP is caring TCP or SCTP
3208 This section details the available filter functions that are available.
3210 Note these functions interface the deprecated legacy filtering framework,
3211 superseded by *rte_flow*. See `Flow rules management`_.
3213 .. _testpmd_flow_director:
3218 Set flow director's input masks::
3220 flow_director_mask (port_id) mode IP vlan (vlan_value) \
3221 src_mask (ipv4_src) (ipv6_src) (src_port) \
3222 dst_mask (ipv4_dst) (ipv6_dst) (dst_port)
3224 flow_director_mask (port_id) mode MAC-VLAN vlan (vlan_value)
3226 flow_director_mask (port_id) mode Tunnel vlan (vlan_value) \
3227 mac (mac_value) tunnel-type (tunnel_type_value) \
3228 tunnel-id (tunnel_id_value)
3230 Example, to set flow director mask on port 0::
3232 testpmd> flow_director_mask 0 mode IP vlan 0xefff \
3233 src_mask 255.255.255.255 \
3234 FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF \
3235 dst_mask 255.255.255.255 \
3236 FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF
3238 flow_director_flex_payload
3239 ~~~~~~~~~~~~~~~~~~~~~~~~~~
3241 Configure flexible payload selection::
3243 flow_director_flex_payload (port_id) (raw|l2|l3|l4) (config)
3245 For example, to select the first 16 bytes from the offset 4 (bytes) of packet's payload as flexible payload::
3247 testpmd> flow_director_flex_payload 0 l4 \
3248 (4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19)
3251 .. _testpmd_rte_flow:
3253 Flow rules management
3254 ---------------------
3256 Control of the generic flow API (*rte_flow*) is fully exposed through the
3257 ``flow`` command (validation, creation, destruction, queries and operation
3260 Considering *rte_flow* overlaps with all `Filter Functions`_, using both
3261 features simultaneously may cause undefined side-effects and is therefore
3267 Because the ``flow`` command uses dynamic tokens to handle the large number
3268 of possible flow rules combinations, its behavior differs slightly from
3269 other commands, in particular:
3271 - Pressing *?* or the *<tab>* key displays contextual help for the current
3272 token, not that of the entire command.
3274 - Optional and repeated parameters are supported (provided they are listed
3275 in the contextual help).
3277 The first parameter stands for the operation mode. Possible operations and
3278 their general syntax are described below. They are covered in detail in the
3281 - Check whether a flow rule can be created::
3283 flow validate {port_id}
3284 [group {group_id}] [priority {level}] [ingress] [egress] [transfer]
3285 pattern {item} [/ {item} [...]] / end
3286 actions {action} [/ {action} [...]] / end
3288 - Create a flow rule::
3290 flow create {port_id}
3291 [group {group_id}] [priority {level}] [ingress] [egress] [transfer]
3292 pattern {item} [/ {item} [...]] / end
3293 actions {action} [/ {action} [...]] / end
3295 - Destroy specific flow rules::
3297 flow destroy {port_id} rule {rule_id} [...]
3299 - Destroy all flow rules::
3301 flow flush {port_id}
3303 - Query an existing flow rule::
3305 flow query {port_id} {rule_id} {action}
3307 - List existing flow rules sorted by priority, filtered by group
3310 flow list {port_id} [group {group_id}] [...]
3312 - Restrict ingress traffic to the defined flow rules::
3314 flow isolate {port_id} {boolean}
3316 - Dump internal representation information of all flows in hardware::
3318 flow dump {port_id} all {output_file}
3322 flow dump {port_id} rule {rule_id} {output_file}
3324 - List and destroy aged flow rules::
3326 flow aged {port_id} [destroy]
3328 - Tunnel offload - create a tunnel stub::
3330 flow tunnel create {port_id} type {tunnel_type}
3332 - Tunnel offload - destroy a tunnel stub::
3334 flow tunnel destroy {port_id} id {tunnel_id}
3336 - Tunnel offload - list port tunnel stubs::
3338 flow tunnel list {port_id}
3340 Creating a tunnel stub for offload
3341 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3343 ``flow tunnel create`` setup a tunnel stub for tunnel offload flow rules::
3345 flow tunnel create {port_id} type {tunnel_type}
3347 If successful, it will return a tunnel stub ID usable with other commands::
3349 port [...]: flow tunnel #[...] type [...]
3351 Tunnel stub ID is relative to a port.
3353 Destroying tunnel offload stub
3354 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3356 ``flow tunnel destroy`` destroy port tunnel stub::
3358 flow tunnel destroy {port_id} id {tunnel_id}
3360 Listing tunnel offload stubs
3361 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3363 ``flow tunnel list`` list port tunnel offload stubs::
3365 flow tunnel list {port_id}
3367 Validating flow rules
3368 ~~~~~~~~~~~~~~~~~~~~~
3370 ``flow validate`` reports whether a flow rule would be accepted by the
3371 underlying device in its current state but stops short of creating it. It is
3372 bound to ``rte_flow_validate()``::
3374 flow validate {port_id}
3375 [group {group_id}] [priority {level}] [ingress] [egress] [transfer]
3376 pattern {item} [/ {item} [...]] / end
3377 actions {action} [/ {action} [...]] / end
3379 If successful, it will show::
3383 Otherwise it will show an error message of the form::
3385 Caught error type [...] ([...]): [...]
3387 This command uses the same parameters as ``flow create``, their format is
3388 described in `Creating flow rules`_.
3390 Check whether redirecting any Ethernet packet received on port 0 to RX queue
3391 index 6 is supported::
3393 testpmd> flow validate 0 ingress pattern eth / end
3394 actions queue index 6 / end
3398 Port 0 does not support TCPv6 rules::
3400 testpmd> flow validate 0 ingress pattern eth / ipv6 / tcp / end
3402 Caught error type 9 (specific pattern item): Invalid argument
3408 ``flow create`` validates and creates the specified flow rule. It is bound
3409 to ``rte_flow_create()``::
3411 flow create {port_id}
3412 [group {group_id}] [priority {level}] [ingress] [egress] [transfer]
3413 [tunnel_set {tunnel_id}] [tunnel_match {tunnel_id}]
3414 pattern {item} [/ {item} [...]] / end
3415 actions {action} [/ {action} [...]] / end
3417 If successful, it will return a flow rule ID usable with other commands::
3419 Flow rule #[...] created
3421 Otherwise it will show an error message of the form::
3423 Caught error type [...] ([...]): [...]
3425 Parameters describe in the following order:
3427 - Attributes (*group*, *priority*, *ingress*, *egress*, *transfer* tokens).
3428 - Tunnel offload specification (tunnel_set, tunnel_match)
3429 - A matching pattern, starting with the *pattern* token and terminated by an
3431 - Actions, starting with the *actions* token and terminated by an *end*
3434 These translate directly to *rte_flow* objects provided as-is to the
3435 underlying functions.
3437 The shortest valid definition only comprises mandatory tokens::
3439 testpmd> flow create 0 pattern end actions end
3441 Note that PMDs may refuse rules that essentially do nothing such as this
3444 **All unspecified object values are automatically initialized to 0.**
3449 These tokens affect flow rule attributes (``struct rte_flow_attr``) and are
3450 specified before the ``pattern`` token.
3452 - ``group {group id}``: priority group.
3453 - ``priority {level}``: priority level within group.
3454 - ``ingress``: rule applies to ingress traffic.
3455 - ``egress``: rule applies to egress traffic.
3456 - ``transfer``: apply rule directly to endpoints found in pattern.
3458 Each instance of an attribute specified several times overrides the previous
3459 value as shown below (group 4 is used)::
3461 testpmd> flow create 0 group 42 group 24 group 4 [...]
3463 Note that once enabled, ``ingress`` and ``egress`` cannot be disabled.
3465 While not specifying a direction is an error, some rules may allow both
3468 Most rules affect RX therefore contain the ``ingress`` token::
3470 testpmd> flow create 0 ingress pattern [...]
3475 Indicate tunnel offload rule type
3477 - ``tunnel_set {tunnel_id}``: mark rule as tunnel offload decap_set type.
3478 - ``tunnel_match {tunnel_id}``: mark rule as tunel offload match type.
3483 A matching pattern starts after the ``pattern`` token. It is made of pattern
3484 items and is terminated by a mandatory ``end`` item.
3486 Items are named after their type (*RTE_FLOW_ITEM_TYPE_* from ``enum
3487 rte_flow_item_type``).
3489 The ``/`` token is used as a separator between pattern items as shown
3492 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end [...]
3494 Note that protocol items like these must be stacked from lowest to highest
3495 layer to make sense. For instance, the following rule is either invalid or
3496 unlikely to match any packet::
3498 testpmd> flow create 0 ingress pattern eth / udp / ipv4 / end [...]
3500 More information on these restrictions can be found in the *rte_flow*
3503 Several items support additional specification structures, for example
3504 ``ipv4`` allows specifying source and destination addresses as follows::
3506 testpmd> flow create 0 ingress pattern eth / ipv4 src is 10.1.1.1
3507 dst is 10.2.0.0 / end [...]
3509 This rule matches all IPv4 traffic with the specified properties.
3511 In this example, ``src`` and ``dst`` are field names of the underlying
3512 ``struct rte_flow_item_ipv4`` object. All item properties can be specified
3513 in a similar fashion.
3515 The ``is`` token means that the subsequent value must be matched exactly,
3516 and assigns ``spec`` and ``mask`` fields in ``struct rte_flow_item``
3517 accordingly. Possible assignment tokens are:
3519 - ``is``: match value perfectly (with full bit-mask).
3520 - ``spec``: match value according to configured bit-mask.
3521 - ``last``: specify upper bound to establish a range.
3522 - ``mask``: specify bit-mask with relevant bits set to one.
3523 - ``prefix``: generate bit-mask with <prefix-length> most-significant bits set to one.
3525 These yield identical results::
3527 ipv4 src is 10.1.1.1
3531 ipv4 src spec 10.1.1.1 src mask 255.255.255.255
3535 ipv4 src spec 10.1.1.1 src prefix 32
3539 ipv4 src is 10.1.1.1 src last 10.1.1.1 # range with a single value
3543 ipv4 src is 10.1.1.1 src last 0 # 0 disables range
3545 Inclusive ranges can be defined with ``last``::
3547 ipv4 src is 10.1.1.1 src last 10.2.3.4 # 10.1.1.1 to 10.2.3.4
3549 Note that ``mask`` affects both ``spec`` and ``last``::
3551 ipv4 src is 10.1.1.1 src last 10.2.3.4 src mask 255.255.0.0
3552 # matches 10.1.0.0 to 10.2.255.255
3554 Properties can be modified multiple times::
3556 ipv4 src is 10.1.1.1 src is 10.1.2.3 src is 10.2.3.4 # matches 10.2.3.4
3560 ipv4 src is 10.1.1.1 src prefix 24 src prefix 16 # matches 10.1.0.0/16
3565 This section lists supported pattern items and their attributes, if any.
3567 - ``end``: end list of pattern items.
3569 - ``void``: no-op pattern item.
3571 - ``invert``: perform actions when pattern does not match.
3573 - ``any``: match any protocol for the current layer.
3575 - ``num {unsigned}``: number of layers covered.
3577 - ``pf``: match traffic from/to the physical function.
3579 - ``vf``: match traffic from/to a virtual function ID.
3581 - ``id {unsigned}``: VF ID.
3583 - ``phy_port``: match traffic from/to a specific physical port.
3585 - ``index {unsigned}``: physical port index.
3587 - ``port_id``: match traffic from/to a given DPDK port ID.
3589 - ``id {unsigned}``: DPDK port ID.
3591 - ``mark``: match value set in previously matched flow rule using the mark action.
3593 - ``id {unsigned}``: arbitrary integer value.
3595 - ``raw``: match an arbitrary byte string.
3597 - ``relative {boolean}``: look for pattern after the previous item.
3598 - ``search {boolean}``: search pattern from offset (see also limit).
3599 - ``offset {integer}``: absolute or relative offset for pattern.
3600 - ``limit {unsigned}``: search area limit for start of pattern.
3601 - ``pattern {string}``: byte string to look for.
3603 - ``eth``: match Ethernet header.
3605 - ``dst {MAC-48}``: destination MAC.
3606 - ``src {MAC-48}``: source MAC.
3607 - ``type {unsigned}``: EtherType or TPID.
3609 - ``vlan``: match 802.1Q/ad VLAN tag.
3611 - ``tci {unsigned}``: tag control information.
3612 - ``pcp {unsigned}``: priority code point.
3613 - ``dei {unsigned}``: drop eligible indicator.
3614 - ``vid {unsigned}``: VLAN identifier.
3615 - ``inner_type {unsigned}``: inner EtherType or TPID.
3617 - ``ipv4``: match IPv4 header.
3619 - ``tos {unsigned}``: type of service.
3620 - ``ttl {unsigned}``: time to live.
3621 - ``proto {unsigned}``: next protocol ID.
3622 - ``src {ipv4 address}``: source address.
3623 - ``dst {ipv4 address}``: destination address.
3625 - ``ipv6``: match IPv6 header.
3627 - ``tc {unsigned}``: traffic class.
3628 - ``flow {unsigned}``: flow label.
3629 - ``proto {unsigned}``: protocol (next header).
3630 - ``hop {unsigned}``: hop limit.
3631 - ``src {ipv6 address}``: source address.
3632 - ``dst {ipv6 address}``: destination address.
3634 - ``icmp``: match ICMP header.
3636 - ``type {unsigned}``: ICMP packet type.
3637 - ``code {unsigned}``: ICMP packet code.
3639 - ``udp``: match UDP header.
3641 - ``src {unsigned}``: UDP source port.
3642 - ``dst {unsigned}``: UDP destination port.
3644 - ``tcp``: match TCP header.
3646 - ``src {unsigned}``: TCP source port.
3647 - ``dst {unsigned}``: TCP destination port.
3649 - ``sctp``: match SCTP header.
3651 - ``src {unsigned}``: SCTP source port.
3652 - ``dst {unsigned}``: SCTP destination port.
3653 - ``tag {unsigned}``: validation tag.
3654 - ``cksum {unsigned}``: checksum.
3656 - ``vxlan``: match VXLAN header.
3658 - ``vni {unsigned}``: VXLAN identifier.
3660 - ``e_tag``: match IEEE 802.1BR E-Tag header.
3662 - ``grp_ecid_b {unsigned}``: GRP and E-CID base.
3664 - ``nvgre``: match NVGRE header.
3666 - ``tni {unsigned}``: virtual subnet ID.
3668 - ``mpls``: match MPLS header.
3670 - ``label {unsigned}``: MPLS label.
3672 - ``gre``: match GRE header.
3674 - ``protocol {unsigned}``: protocol type.
3676 - ``gre_key``: match GRE optional key field.
3678 - ``value {unsigned}``: key value.
3680 - ``fuzzy``: fuzzy pattern match, expect faster than default.
3682 - ``thresh {unsigned}``: accuracy threshold.
3684 - ``gtp``, ``gtpc``, ``gtpu``: match GTPv1 header.
3686 - ``teid {unsigned}``: tunnel endpoint identifier.
3688 - ``geneve``: match GENEVE header.
3690 - ``vni {unsigned}``: virtual network identifier.
3691 - ``protocol {unsigned}``: protocol type.
3693 - ``geneve-opt``: match GENEVE header option.
3695 - ``class {unsigned}``: GENEVE option class.
3696 - ``type {unsigned}``: GENEVE option type.
3697 - ``length {unsigned}``: GENEVE option length in 32-bit words.
3698 - ``data {hex string}``: GENEVE option data, the length is defined by
3701 - ``vxlan-gpe``: match VXLAN-GPE header.
3703 - ``vni {unsigned}``: VXLAN-GPE identifier.
3705 - ``arp_eth_ipv4``: match ARP header for Ethernet/IPv4.
3707 - ``sha {MAC-48}``: sender hardware address.
3708 - ``spa {ipv4 address}``: sender IPv4 address.
3709 - ``tha {MAC-48}``: target hardware address.
3710 - ``tpa {ipv4 address}``: target IPv4 address.
3712 - ``ipv6_ext``: match presence of any IPv6 extension header.
3714 - ``next_hdr {unsigned}``: next header.
3716 - ``icmp6``: match any ICMPv6 header.
3718 - ``type {unsigned}``: ICMPv6 type.
3719 - ``code {unsigned}``: ICMPv6 code.
3721 - ``icmp6_nd_ns``: match ICMPv6 neighbor discovery solicitation.
3723 - ``target_addr {ipv6 address}``: target address.
3725 - ``icmp6_nd_na``: match ICMPv6 neighbor discovery advertisement.
3727 - ``target_addr {ipv6 address}``: target address.
3729 - ``icmp6_nd_opt``: match presence of any ICMPv6 neighbor discovery option.
3731 - ``type {unsigned}``: ND option type.
3733 - ``icmp6_nd_opt_sla_eth``: match ICMPv6 neighbor discovery source Ethernet
3734 link-layer address option.
3736 - ``sla {MAC-48}``: source Ethernet LLA.
3738 - ``icmp6_nd_opt_tla_eth``: match ICMPv6 neighbor discovery target Ethernet
3739 link-layer address option.
3741 - ``tla {MAC-48}``: target Ethernet LLA.
3743 - ``meta``: match application specific metadata.
3745 - ``data {unsigned}``: metadata value.
3747 - ``gtp_psc``: match GTP PDU extension header with type 0x85.
3749 - ``pdu_type {unsigned}``: PDU type.
3751 - ``qfi {unsigned}``: PPP, RQI and QoS flow identifier.
3753 - ``pppoes``, ``pppoed``: match PPPoE header.
3755 - ``session_id {unsigned}``: session identifier.
3757 - ``pppoe_proto_id``: match PPPoE session protocol identifier.
3759 - ``proto_id {unsigned}``: PPP protocol identifier.
3761 - ``l2tpv3oip``: match L2TPv3 over IP header.
3763 - ``session_id {unsigned}``: L2TPv3 over IP session identifier.
3765 - ``ah``: match AH header.
3767 - ``spi {unsigned}``: security parameters index.
3769 - ``pfcp``: match PFCP header.
3771 - ``s_field {unsigned}``: S field.
3772 - ``seid {unsigned}``: session endpoint identifier.
3774 - ``integrity``: match packet integrity.
3776 - ``level {unsigned}``: Packet encapsulation level the item should
3777 apply to. See rte_flow_action_rss for details.
3778 - ``value {unsigned}``: A bitmask that specify what packet elements
3779 must be matched for integrity.
3781 - ``conntrack``: match conntrack state.
3786 A list of actions starts after the ``actions`` token in the same fashion as
3787 `Matching pattern`_; actions are separated by ``/`` tokens and the list is
3788 terminated by a mandatory ``end`` action.
3790 Actions are named after their type (*RTE_FLOW_ACTION_TYPE_* from ``enum
3791 rte_flow_action_type``).
3793 Dropping all incoming UDPv4 packets can be expressed as follows::
3795 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
3798 Several actions have configurable properties which must be specified when
3799 there is no valid default value. For example, ``queue`` requires a target
3802 This rule redirects incoming UDPv4 traffic to queue index 6::
3804 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
3805 actions queue index 6 / end
3807 While this one could be rejected by PMDs (unspecified queue index)::
3809 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
3812 As defined by *rte_flow*, the list is not ordered, all actions of a given
3813 rule are performed simultaneously. These are equivalent::
3815 queue index 6 / void / mark id 42 / end
3819 void / mark id 42 / queue index 6 / end
3821 All actions in a list should have different types, otherwise only the last
3822 action of a given type is taken into account::
3824 queue index 4 / queue index 5 / queue index 6 / end # will use queue 6
3828 drop / drop / drop / end # drop is performed only once
3832 mark id 42 / queue index 3 / mark id 24 / end # mark will be 24
3834 Considering they are performed simultaneously, opposite and overlapping
3835 actions can sometimes be combined when the end result is unambiguous::
3837 drop / queue index 6 / end # drop has no effect
3841 queue index 6 / rss queues 6 7 8 / end # queue has no effect
3845 drop / passthru / end # drop has no effect
3847 Note that PMDs may still refuse such combinations.
3852 This section lists supported actions and their attributes, if any.
3854 - ``end``: end list of actions.
3856 - ``void``: no-op action.
3858 - ``passthru``: let subsequent rule process matched packets.
3860 - ``jump``: redirect traffic to group on device.
3862 - ``group {unsigned}``: group to redirect to.
3864 - ``mark``: attach 32 bit value to packets.
3866 - ``id {unsigned}``: 32 bit value to return with packets.
3868 - ``flag``: flag packets.
3870 - ``queue``: assign packets to a given queue index.
3872 - ``index {unsigned}``: queue index to use.
3874 - ``drop``: drop packets (note: passthru has priority).
3876 - ``count``: enable counters for this rule.
3878 - ``rss``: spread packets among several queues.
3880 - ``func {hash function}``: RSS hash function to apply, allowed tokens are
3881 ``toeplitz``, ``simple_xor``, ``symmetric_toeplitz`` and ``default``.
3883 - ``level {unsigned}``: encapsulation level for ``types``.
3885 - ``types [{RSS hash type} [...]] end``: specific RSS hash types.
3886 Note that an empty list does not disable RSS but instead requests
3887 unspecified "best-effort" settings.
3889 - ``key {string}``: RSS hash key, overrides ``key_len``.
3891 - ``key_len {unsigned}``: RSS hash key length in bytes, can be used in
3892 conjunction with ``key`` to pad or truncate it.
3894 - ``queues [{unsigned} [...]] end``: queue indices to use.
3896 - ``pf``: direct traffic to physical function.
3898 - ``vf``: direct traffic to a virtual function ID.
3900 - ``original {boolean}``: use original VF ID if possible.
3901 - ``id {unsigned}``: VF ID.
3903 - ``phy_port``: direct packets to physical port index.
3905 - ``original {boolean}``: use original port index if possible.
3906 - ``index {unsigned}``: physical port index.
3908 - ``port_id``: direct matching traffic to a given DPDK port ID.
3910 - ``original {boolean}``: use original DPDK port ID if possible.
3911 - ``id {unsigned}``: DPDK port ID.
3913 - ``of_set_mpls_ttl``: OpenFlow's ``OFPAT_SET_MPLS_TTL``.
3915 - ``mpls_ttl``: MPLS TTL.
3917 - ``of_dec_mpls_ttl``: OpenFlow's ``OFPAT_DEC_MPLS_TTL``.
3919 - ``of_set_nw_ttl``: OpenFlow's ``OFPAT_SET_NW_TTL``.
3921 - ``nw_ttl``: IP TTL.
3923 - ``of_dec_nw_ttl``: OpenFlow's ``OFPAT_DEC_NW_TTL``.
3925 - ``of_copy_ttl_out``: OpenFlow's ``OFPAT_COPY_TTL_OUT``.
3927 - ``of_copy_ttl_in``: OpenFlow's ``OFPAT_COPY_TTL_IN``.
3929 - ``of_pop_vlan``: OpenFlow's ``OFPAT_POP_VLAN``.
3931 - ``of_push_vlan``: OpenFlow's ``OFPAT_PUSH_VLAN``.
3933 - ``ethertype``: Ethertype.
3935 - ``of_set_vlan_vid``: OpenFlow's ``OFPAT_SET_VLAN_VID``.
3937 - ``vlan_vid``: VLAN id.
3939 - ``of_set_vlan_pcp``: OpenFlow's ``OFPAT_SET_VLAN_PCP``.
3941 - ``vlan_pcp``: VLAN priority.
3943 - ``of_pop_mpls``: OpenFlow's ``OFPAT_POP_MPLS``.
3945 - ``ethertype``: Ethertype.
3947 - ``of_push_mpls``: OpenFlow's ``OFPAT_PUSH_MPLS``.
3949 - ``ethertype``: Ethertype.
3951 - ``vxlan_encap``: Performs a VXLAN encapsulation, outer layer configuration
3952 is done through `Config VXLAN Encap outer layers`_.
3954 - ``vxlan_decap``: Performs a decapsulation action by stripping all headers of
3955 the VXLAN tunnel network overlay from the matched flow.
3957 - ``nvgre_encap``: Performs a NVGRE encapsulation, outer layer configuration
3958 is done through `Config NVGRE Encap outer layers`_.
3960 - ``nvgre_decap``: Performs a decapsulation action by stripping all headers of
3961 the NVGRE tunnel network overlay from the matched flow.
3963 - ``l2_encap``: Performs a L2 encapsulation, L2 configuration
3964 is done through `Config L2 Encap`_.
3966 - ``l2_decap``: Performs a L2 decapsulation, L2 configuration
3967 is done through `Config L2 Decap`_.
3969 - ``mplsogre_encap``: Performs a MPLSoGRE encapsulation, outer layer
3970 configuration is done through `Config MPLSoGRE Encap outer layers`_.
3972 - ``mplsogre_decap``: Performs a MPLSoGRE decapsulation, outer layer
3973 configuration is done through `Config MPLSoGRE Decap outer layers`_.
3975 - ``mplsoudp_encap``: Performs a MPLSoUDP encapsulation, outer layer
3976 configuration is done through `Config MPLSoUDP Encap outer layers`_.
3978 - ``mplsoudp_decap``: Performs a MPLSoUDP decapsulation, outer layer
3979 configuration is done through `Config MPLSoUDP Decap outer layers`_.
3981 - ``set_ipv4_src``: Set a new IPv4 source address in the outermost IPv4 header.
3983 - ``ipv4_addr``: New IPv4 source address.
3985 - ``set_ipv4_dst``: Set a new IPv4 destination address in the outermost IPv4
3988 - ``ipv4_addr``: New IPv4 destination address.
3990 - ``set_ipv6_src``: Set a new IPv6 source address in the outermost IPv6 header.
3992 - ``ipv6_addr``: New IPv6 source address.
3994 - ``set_ipv6_dst``: Set a new IPv6 destination address in the outermost IPv6
3997 - ``ipv6_addr``: New IPv6 destination address.
3999 - ``set_tp_src``: Set a new source port number in the outermost TCP/UDP
4002 - ``port``: New TCP/UDP source port number.
4004 - ``set_tp_dst``: Set a new destination port number in the outermost TCP/UDP
4007 - ``port``: New TCP/UDP destination port number.
4009 - ``mac_swap``: Swap the source and destination MAC addresses in the outermost
4012 - ``dec_ttl``: Performs a decrease TTL value action
4014 - ``set_ttl``: Set TTL value with specified value
4015 - ``ttl_value {unsigned}``: The new TTL value to be set
4017 - ``set_mac_src``: set source MAC address
4019 - ``mac_addr {MAC-48}``: new source MAC address
4021 - ``set_mac_dst``: set destination MAC address
4023 - ``mac_addr {MAC-48}``: new destination MAC address
4025 - ``inc_tcp_seq``: Increase sequence number in the outermost TCP header.
4027 - ``value {unsigned}``: Value to increase TCP sequence number by.
4029 - ``dec_tcp_seq``: Decrease sequence number in the outermost TCP header.
4031 - ``value {unsigned}``: Value to decrease TCP sequence number by.
4033 - ``inc_tcp_ack``: Increase acknowledgment number in the outermost TCP header.
4035 - ``value {unsigned}``: Value to increase TCP acknowledgment number by.
4037 - ``dec_tcp_ack``: Decrease acknowledgment number in the outermost TCP header.
4039 - ``value {unsigned}``: Value to decrease TCP acknowledgment number by.
4041 - ``set_ipv4_dscp``: Set IPv4 DSCP value with specified value
4043 - ``dscp_value {unsigned}``: The new DSCP value to be set
4045 - ``set_ipv6_dscp``: Set IPv6 DSCP value with specified value
4047 - ``dscp_value {unsigned}``: The new DSCP value to be set
4049 - ``indirect``: Use indirect action created via
4050 ``flow indirect_action {port_id} create``
4052 - ``indirect_action_id {unsigned}``: Indirect action ID to use
4054 Destroying flow rules
4055 ~~~~~~~~~~~~~~~~~~~~~
4057 ``flow destroy`` destroys one or more rules from their rule ID (as returned
4058 by ``flow create``), this command calls ``rte_flow_destroy()`` as many
4059 times as necessary::
4061 flow destroy {port_id} rule {rule_id} [...]
4063 If successful, it will show::
4065 Flow rule #[...] destroyed
4067 It does not report anything for rule IDs that do not exist. The usual error
4068 message is shown when a rule cannot be destroyed::
4070 Caught error type [...] ([...]): [...]
4072 ``flow flush`` destroys all rules on a device and does not take extra
4073 arguments. It is bound to ``rte_flow_flush()``::
4075 flow flush {port_id}
4077 Any errors are reported as above.
4079 Creating several rules and destroying them::
4081 testpmd> flow create 0 ingress pattern eth / ipv6 / end
4082 actions queue index 2 / end
4083 Flow rule #0 created
4084 testpmd> flow create 0 ingress pattern eth / ipv4 / end
4085 actions queue index 3 / end
4086 Flow rule #1 created
4087 testpmd> flow destroy 0 rule 0 rule 1
4088 Flow rule #1 destroyed
4089 Flow rule #0 destroyed
4092 The same result can be achieved using ``flow flush``::
4094 testpmd> flow create 0 ingress pattern eth / ipv6 / end
4095 actions queue index 2 / end
4096 Flow rule #0 created
4097 testpmd> flow create 0 ingress pattern eth / ipv4 / end
4098 actions queue index 3 / end
4099 Flow rule #1 created
4100 testpmd> flow flush 0
4103 Non-existent rule IDs are ignored::
4105 testpmd> flow create 0 ingress pattern eth / ipv6 / end
4106 actions queue index 2 / end
4107 Flow rule #0 created
4108 testpmd> flow create 0 ingress pattern eth / ipv4 / end
4109 actions queue index 3 / end
4110 Flow rule #1 created
4111 testpmd> flow destroy 0 rule 42 rule 10 rule 2
4113 testpmd> flow destroy 0 rule 0
4114 Flow rule #0 destroyed
4120 ``flow query`` queries a specific action of a flow rule having that
4121 ability. Such actions collect information that can be reported using this
4122 command. It is bound to ``rte_flow_query()``::
4124 flow query {port_id} {rule_id} {action}
4126 If successful, it will display either the retrieved data for known actions
4127 or the following message::
4129 Cannot display result for action type [...] ([...])
4131 Otherwise, it will complain either that the rule does not exist or that some
4134 Flow rule #[...] not found
4138 Caught error type [...] ([...]): [...]
4140 Currently only the ``count`` action is supported. This action reports the
4141 number of packets that hit the flow rule and the total number of bytes. Its
4142 output has the following format::
4145 hits_set: [...] # whether "hits" contains a valid value
4146 bytes_set: [...] # whether "bytes" contains a valid value
4147 hits: [...] # number of packets
4148 bytes: [...] # number of bytes
4150 Querying counters for TCPv6 packets redirected to queue 6::
4152 testpmd> flow create 0 ingress pattern eth / ipv6 / tcp / end
4153 actions queue index 6 / count / end
4154 Flow rule #4 created
4155 testpmd> flow query 0 4 count
4166 ``flow list`` lists existing flow rules sorted by priority and optionally
4167 filtered by group identifiers::
4169 flow list {port_id} [group {group_id}] [...]
4171 This command only fails with the following message if the device does not
4176 Output consists of a header line followed by a short description of each
4177 flow rule, one per line. There is no output at all when no flow rules are
4178 configured on the device::
4180 ID Group Prio Attr Rule
4181 [...] [...] [...] [...] [...]
4183 ``Attr`` column flags:
4185 - ``i`` for ``ingress``.
4186 - ``e`` for ``egress``.
4188 Creating several flow rules and listing them::
4190 testpmd> flow create 0 ingress pattern eth / ipv4 / end
4191 actions queue index 6 / end
4192 Flow rule #0 created
4193 testpmd> flow create 0 ingress pattern eth / ipv6 / end
4194 actions queue index 2 / end
4195 Flow rule #1 created
4196 testpmd> flow create 0 priority 5 ingress pattern eth / ipv4 / udp / end
4197 actions rss queues 6 7 8 end / end
4198 Flow rule #2 created
4199 testpmd> flow list 0
4200 ID Group Prio Attr Rule
4201 0 0 0 i- ETH IPV4 => QUEUE
4202 1 0 0 i- ETH IPV6 => QUEUE
4203 2 0 5 i- ETH IPV4 UDP => RSS
4206 Rules are sorted by priority (i.e. group ID first, then priority level)::
4208 testpmd> flow list 1
4209 ID Group Prio Attr Rule
4210 0 0 0 i- ETH => COUNT
4211 6 0 500 i- ETH IPV6 TCP => DROP COUNT
4212 5 0 1000 i- ETH IPV6 ICMP => QUEUE
4213 1 24 0 i- ETH IPV4 UDP => QUEUE
4214 4 24 10 i- ETH IPV4 TCP => DROP
4215 3 24 20 i- ETH IPV4 => DROP
4216 2 24 42 i- ETH IPV4 UDP => QUEUE
4217 7 63 0 i- ETH IPV6 UDP VXLAN => MARK QUEUE
4220 Output can be limited to specific groups::
4222 testpmd> flow list 1 group 0 group 63
4223 ID Group Prio Attr Rule
4224 0 0 0 i- ETH => COUNT
4225 6 0 500 i- ETH IPV6 TCP => DROP COUNT
4226 5 0 1000 i- ETH IPV6 ICMP => QUEUE
4227 7 63 0 i- ETH IPV6 UDP VXLAN => MARK QUEUE
4230 Toggling isolated mode
4231 ~~~~~~~~~~~~~~~~~~~~~~
4233 ``flow isolate`` can be used to tell the underlying PMD that ingress traffic
4234 must only be injected from the defined flow rules; that no default traffic
4235 is expected outside those rules and the driver is free to assign more
4236 resources to handle them. It is bound to ``rte_flow_isolate()``::
4238 flow isolate {port_id} {boolean}
4240 If successful, enabling or disabling isolated mode shows either::
4242 Ingress traffic on port [...]
4243 is now restricted to the defined flow rules
4247 Ingress traffic on port [...]
4248 is not restricted anymore to the defined flow rules
4250 Otherwise, in case of error::
4252 Caught error type [...] ([...]): [...]
4254 Mainly due to its side effects, PMDs supporting this mode may not have the
4255 ability to toggle it more than once without reinitializing affected ports
4256 first (e.g. by exiting testpmd).
4258 Enabling isolated mode::
4260 testpmd> flow isolate 0 true
4261 Ingress traffic on port 0 is now restricted to the defined flow rules
4264 Disabling isolated mode::
4266 testpmd> flow isolate 0 false
4267 Ingress traffic on port 0 is not restricted anymore to the defined flow rules
4270 Dumping HW internal information
4271 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4273 ``flow dump`` dumps the hardware's internal representation information of
4274 all flows. It is bound to ``rte_flow_dev_dump()``::
4276 flow dump {port_id} {output_file}
4278 If successful, it will show::
4282 Otherwise, it will complain error occurred::
4284 Caught error type [...] ([...]): [...]
4286 Listing and destroying aged flow rules
4287 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4289 ``flow aged`` simply lists aged flow rules be get from api ``rte_flow_get_aged_flows``,
4290 and ``destroy`` parameter can be used to destroy those flow rules in PMD.
4292 flow aged {port_id} [destroy]
4294 Listing current aged flow rules::
4296 testpmd> flow aged 0
4297 Port 0 total aged flows: 0
4298 testpmd> flow create 0 ingress pattern eth / ipv4 src is 2.2.2.14 / end
4299 actions age timeout 5 / queue index 0 / end
4300 Flow rule #0 created
4301 testpmd> flow create 0 ingress pattern eth / ipv4 src is 2.2.2.15 / end
4302 actions age timeout 4 / queue index 0 / end
4303 Flow rule #1 created
4304 testpmd> flow create 0 ingress pattern eth / ipv4 src is 2.2.2.16 / end
4305 actions age timeout 2 / queue index 0 / end
4306 Flow rule #2 created
4307 testpmd> flow create 0 ingress pattern eth / ipv4 src is 2.2.2.17 / end
4308 actions age timeout 3 / queue index 0 / end
4309 Flow rule #3 created
4312 Aged Rules are simply list as command ``flow list {port_id}``, but strip the detail rule
4313 information, all the aged flows are sorted by the longest timeout time. For example, if
4314 those rules be configured in the same time, ID 2 will be the first aged out rule, the next
4315 will be ID 3, ID 1, ID 0::
4317 testpmd> flow aged 0
4318 Port 0 total aged flows: 4
4325 If attach ``destroy`` parameter, the command will destroy all the list aged flow rules.
4327 testpmd> flow aged 0 destroy
4328 Port 0 total aged flows: 4
4335 Flow rule #2 destroyed
4336 Flow rule #3 destroyed
4337 Flow rule #1 destroyed
4338 Flow rule #0 destroyed
4339 4 flows be destroyed
4340 testpmd> flow aged 0
4341 Port 0 total aged flows: 0
4343 Creating indirect actions
4344 ~~~~~~~~~~~~~~~~~~~~~~~~~
4346 ``flow indirect_action {port_id} create`` creates indirect action with optional
4347 indirect action ID. It is bound to ``rte_flow_action_handle_create()``::
4349 flow indirect_action {port_id} create [action_id {indirect_action_id}]
4350 [ingress] [egress] [transfer] action {action} / end
4352 If successful, it will show::
4354 Indirect action #[...] created
4356 Otherwise, it will complain either that indirect action already exists or that
4357 some error occurred::
4359 Indirect action #[...] is already assigned, delete it first
4363 Caught error type [...] ([...]): [...]
4365 Create indirect rss action with id 100 to queues 1 and 2 on port 0::
4367 testpmd> flow indirect_action 0 create action_id 100 \
4368 ingress action rss queues 1 2 end / end
4370 Create indirect rss action with id assigned by testpmd to queues 1 and 2 on
4373 testpmd> flow indirect_action 0 create action_id \
4374 ingress action rss queues 0 1 end / end
4376 Updating indirect actions
4377 ~~~~~~~~~~~~~~~~~~~~~~~~~
4379 ``flow indirect_action {port_id} update`` updates configuration of the indirect
4380 action from its indirect action ID (as returned by
4381 ``flow indirect_action {port_id} create``). It is bound to
4382 ``rte_flow_action_handle_update()``::
4384 flow indirect_action {port_id} update {indirect_action_id}
4385 action {action} / end
4387 If successful, it will show::
4389 Indirect action #[...] updated
4391 Otherwise, it will complain either that indirect action not found or that some
4394 Failed to find indirect action #[...] on port [...]
4398 Caught error type [...] ([...]): [...]
4400 Update indirect rss action having id 100 on port 0 with rss to queues 0 and 3
4401 (in create example above rss queues were 1 and 2)::
4403 testpmd> flow indirect_action 0 update 100 action rss queues 0 3 end / end
4405 Destroying indirect actions
4406 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
4408 ``flow indirect_action {port_id} destroy`` destroys one or more indirect actions
4409 from their indirect action IDs (as returned by
4410 ``flow indirect_action {port_id} create``). It is bound to
4411 ``rte_flow_action_handle_destroy()``::
4413 flow indirect_action {port_id} destroy action_id {indirect_action_id} [...]
4415 If successful, it will show::
4417 Indirect action #[...] destroyed
4419 It does not report anything for indirect action IDs that do not exist.
4420 The usual error message is shown when a indirect action cannot be destroyed::
4422 Caught error type [...] ([...]): [...]
4424 Destroy indirect actions having id 100 & 101::
4426 testpmd> flow indirect_action 0 destroy action_id 100 action_id 101
4428 Query indirect actions
4429 ~~~~~~~~~~~~~~~~~~~~~~
4431 ``flow indirect_action {port_id} query`` queries the indirect action from its
4432 indirect action ID (as returned by ``flow indirect_action {port_id} create``).
4433 It is bound to ``rte_flow_action_handle_query()``::
4435 flow indirect_action {port_id} query {indirect_action_id}
4437 Currently only rss indirect action supported. If successful, it will show::
4439 Indirect RSS action:
4442 Otherwise, it will complain either that indirect action not found or that some
4445 Failed to find indirect action #[...] on port [...]
4449 Caught error type [...] ([...]): [...]
4451 Query indirect action having id 100::
4453 testpmd> flow indirect_action 0 query 100
4455 Sample QinQ flow rules
4456 ~~~~~~~~~~~~~~~~~~~~~~
4458 Before creating QinQ rule(s) the following commands should be issued to enable QinQ::
4460 testpmd> port stop 0
4461 testpmd> vlan set extend on 0
4463 The above command sets the inner and outer TPID's to 0x8100.
4465 To change the TPID's the following commands should be used::
4467 testpmd> vlan set outer tpid 0x88A8 0
4468 testpmd> vlan set inner tpid 0x8100 0
4469 testpmd> port start 0
4471 Validate and create a QinQ rule on port 0 to steer traffic to a VF queue in a VM.
4475 testpmd> flow validate 0 ingress pattern eth / vlan tci is 123 /
4476 vlan tci is 456 / end actions vf id 1 / queue index 0 / end
4477 Flow rule #0 validated
4479 testpmd> flow create 0 ingress pattern eth / vlan tci is 4 /
4480 vlan tci is 456 / end actions vf id 123 / queue index 0 / end
4481 Flow rule #0 created
4483 testpmd> flow list 0
4484 ID Group Prio Attr Rule
4485 0 0 0 i- ETH VLAN VLAN=>VF QUEUE
4487 Validate and create a QinQ rule on port 0 to steer traffic to a queue on the host.
4491 testpmd> flow validate 0 ingress pattern eth / vlan tci is 321 /
4492 vlan tci is 654 / end actions pf / queue index 0 / end
4493 Flow rule #1 validated
4495 testpmd> flow create 0 ingress pattern eth / vlan tci is 321 /
4496 vlan tci is 654 / end actions pf / queue index 1 / end
4497 Flow rule #1 created
4499 testpmd> flow list 0
4500 ID Group Prio Attr Rule
4501 0 0 0 i- ETH VLAN VLAN=>VF QUEUE
4502 1 0 0 i- ETH VLAN VLAN=>PF QUEUE
4504 Sample VXLAN flow rules
4505 ~~~~~~~~~~~~~~~~~~~~~~~
4507 Before creating VXLAN rule(s), the UDP port should be added for VXLAN packet
4510 testpmd> rx_vxlan_port add 4789 0
4512 Create VXLAN rules on port 0 to steer traffic to PF queues.
4516 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / vxlan /
4517 eth dst is 00:11:22:33:44:55 / end actions pf / queue index 1 / end
4518 Flow rule #0 created
4520 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / vxlan vni is 3 /
4521 eth dst is 00:11:22:33:44:55 / end actions pf / queue index 2 / end
4522 Flow rule #1 created
4524 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / vxlan /
4525 eth dst is 00:11:22:33:44:55 / vlan tci is 10 / end actions pf /
4527 Flow rule #2 created
4529 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / vxlan vni is 5 /
4530 eth dst is 00:11:22:33:44:55 / vlan tci is 20 / end actions pf /
4532 Flow rule #3 created
4534 testpmd> flow create 0 ingress pattern eth dst is 00:00:00:00:01:00 / ipv4 /
4535 udp / vxlan vni is 6 / eth dst is 00:11:22:33:44:55 / end actions pf /
4537 Flow rule #4 created
4539 testpmd> flow list 0
4540 ID Group Prio Attr Rule
4541 0 0 0 i- ETH IPV4 UDP VXLAN ETH => QUEUE
4542 1 0 0 i- ETH IPV4 UDP VXLAN ETH => QUEUE
4543 2 0 0 i- ETH IPV4 UDP VXLAN ETH VLAN => QUEUE
4544 3 0 0 i- ETH IPV4 UDP VXLAN ETH VLAN => QUEUE
4545 4 0 0 i- ETH IPV4 UDP VXLAN ETH => QUEUE
4547 Sample VXLAN encapsulation rule
4548 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4550 VXLAN encapsulation outer layer has default value pre-configured in testpmd
4551 source code, those can be changed by using the following commands
4553 IPv4 VXLAN outer header::
4555 testpmd> set vxlan ip-version ipv4 vni 4 udp-src 4 udp-dst 4 ip-src 127.0.0.1
4556 ip-dst 128.0.0.1 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4557 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4560 testpmd> set vxlan-with-vlan ip-version ipv4 vni 4 udp-src 4 udp-dst 4 ip-src
4561 127.0.0.1 ip-dst 128.0.0.1 vlan-tci 34 eth-src 11:11:11:11:11:11
4562 eth-dst 22:22:22:22:22:22
4563 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4566 testpmd> set vxlan-tos-ttl ip-version ipv4 vni 4 udp-src 4 udp-dst 4 ip-tos 0
4567 ip-ttl 255 ip-src 127.0.0.1 ip-dst 128.0.0.1 eth-src 11:11:11:11:11:11
4568 eth-dst 22:22:22:22:22:22
4569 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4572 IPv6 VXLAN outer header::
4574 testpmd> set vxlan ip-version ipv6 vni 4 udp-src 4 udp-dst 4 ip-src ::1
4575 ip-dst ::2222 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4576 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4579 testpmd> set vxlan-with-vlan ip-version ipv6 vni 4 udp-src 4 udp-dst 4
4580 ip-src ::1 ip-dst ::2222 vlan-tci 34 eth-src 11:11:11:11:11:11
4581 eth-dst 22:22:22:22:22:22
4582 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4585 testpmd> set vxlan-tos-ttl ip-version ipv6 vni 4 udp-src 4 udp-dst 4
4586 ip-tos 0 ip-ttl 255 ::1 ip-dst ::2222 eth-src 11:11:11:11:11:11
4587 eth-dst 22:22:22:22:22:22
4588 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4591 Sample NVGRE encapsulation rule
4592 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4594 NVGRE encapsulation outer layer has default value pre-configured in testpmd
4595 source code, those can be changed by using the following commands
4597 IPv4 NVGRE outer header::
4599 testpmd> set nvgre ip-version ipv4 tni 4 ip-src 127.0.0.1 ip-dst 128.0.0.1
4600 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4601 testpmd> flow create 0 ingress pattern end actions nvgre_encap /
4604 testpmd> set nvgre-with-vlan ip-version ipv4 tni 4 ip-src 127.0.0.1
4605 ip-dst 128.0.0.1 vlan-tci 34 eth-src 11:11:11:11:11:11
4606 eth-dst 22:22:22:22:22:22
4607 testpmd> flow create 0 ingress pattern end actions nvgre_encap /
4610 IPv6 NVGRE outer header::
4612 testpmd> set nvgre ip-version ipv6 tni 4 ip-src ::1 ip-dst ::2222
4613 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4614 testpmd> flow create 0 ingress pattern end actions nvgre_encap /
4617 testpmd> set nvgre-with-vlan ip-version ipv6 tni 4 ip-src ::1 ip-dst ::2222
4618 vlan-tci 34 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 Sample L2 encapsulation rule
4623 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4625 L2 encapsulation has default value pre-configured in testpmd
4626 source code, those can be changed by using the following commands
4630 testpmd> set l2_encap ip-version ipv4
4631 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4632 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / mpls / end actions
4633 mplsoudp_decap / l2_encap / end
4635 L2 with VXLAN header::
4637 testpmd> set l2_encap-with-vlan ip-version ipv4 vlan-tci 34
4638 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4639 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / mpls / end actions
4640 mplsoudp_decap / l2_encap / end
4642 Sample L2 decapsulation rule
4643 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4645 L2 decapsulation has default value pre-configured in testpmd
4646 source code, those can be changed by using the following commands
4650 testpmd> set l2_decap
4651 testpmd> flow create 0 egress pattern eth / end actions l2_decap / mplsoudp_encap /
4654 L2 with VXLAN header::
4656 testpmd> set l2_encap-with-vlan
4657 testpmd> flow create 0 egress pattern eth / end actions l2_encap / mplsoudp_encap /
4660 Sample MPLSoGRE encapsulation rule
4661 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4663 MPLSoGRE encapsulation outer layer has default value pre-configured in testpmd
4664 source code, those can be changed by using the following commands
4666 IPv4 MPLSoGRE outer header::
4668 testpmd> set mplsogre_encap ip-version ipv4 label 4
4669 ip-src 127.0.0.1 ip-dst 128.0.0.1 eth-src 11:11:11:11:11:11
4670 eth-dst 22:22:22:22:22:22
4671 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4672 mplsogre_encap / end
4674 IPv4 MPLSoGRE with VLAN outer header::
4676 testpmd> set mplsogre_encap-with-vlan ip-version ipv4 label 4
4677 ip-src 127.0.0.1 ip-dst 128.0.0.1 vlan-tci 34
4678 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4679 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4680 mplsogre_encap / end
4682 IPv6 MPLSoGRE outer header::
4684 testpmd> set mplsogre_encap ip-version ipv6 mask 4
4685 ip-src ::1 ip-dst ::2222 eth-src 11:11:11:11:11:11
4686 eth-dst 22:22:22:22:22:22
4687 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4688 mplsogre_encap / end
4690 IPv6 MPLSoGRE with VLAN outer header::
4692 testpmd> set mplsogre_encap-with-vlan ip-version ipv6 mask 4
4693 ip-src ::1 ip-dst ::2222 vlan-tci 34
4694 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4695 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4696 mplsogre_encap / end
4698 Sample MPLSoGRE decapsulation rule
4699 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4701 MPLSoGRE decapsulation outer layer has default value pre-configured in testpmd
4702 source code, those can be changed by using the following commands
4704 IPv4 MPLSoGRE outer header::
4706 testpmd> set mplsogre_decap ip-version ipv4
4707 testpmd> flow create 0 ingress pattern eth / ipv4 / gre / mpls / end actions
4708 mplsogre_decap / l2_encap / end
4710 IPv4 MPLSoGRE with VLAN outer header::
4712 testpmd> set mplsogre_decap-with-vlan ip-version ipv4
4713 testpmd> flow create 0 ingress pattern eth / vlan / ipv4 / gre / mpls / end
4714 actions mplsogre_decap / l2_encap / end
4716 IPv6 MPLSoGRE outer header::
4718 testpmd> set mplsogre_decap ip-version ipv6
4719 testpmd> flow create 0 ingress pattern eth / ipv6 / gre / mpls / end
4720 actions mplsogre_decap / l2_encap / end
4722 IPv6 MPLSoGRE with VLAN outer header::
4724 testpmd> set mplsogre_decap-with-vlan ip-version ipv6
4725 testpmd> flow create 0 ingress pattern eth / vlan / ipv6 / gre / mpls / end
4726 actions mplsogre_decap / l2_encap / end
4728 Sample MPLSoUDP encapsulation rule
4729 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4731 MPLSoUDP encapsulation outer layer has default value pre-configured in testpmd
4732 source code, those can be changed by using the following commands
4734 IPv4 MPLSoUDP outer header::
4736 testpmd> set mplsoudp_encap ip-version ipv4 label 4 udp-src 5 udp-dst 10
4737 ip-src 127.0.0.1 ip-dst 128.0.0.1 eth-src 11:11:11:11:11:11
4738 eth-dst 22:22:22:22:22:22
4739 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4740 mplsoudp_encap / end
4742 IPv4 MPLSoUDP with VLAN outer header::
4744 testpmd> set mplsoudp_encap-with-vlan ip-version ipv4 label 4 udp-src 5
4745 udp-dst 10 ip-src 127.0.0.1 ip-dst 128.0.0.1 vlan-tci 34
4746 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4747 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4748 mplsoudp_encap / end
4750 IPv6 MPLSoUDP outer header::
4752 testpmd> set mplsoudp_encap ip-version ipv6 mask 4 udp-src 5 udp-dst 10
4753 ip-src ::1 ip-dst ::2222 eth-src 11:11:11:11:11:11
4754 eth-dst 22:22:22:22:22:22
4755 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4756 mplsoudp_encap / end
4758 IPv6 MPLSoUDP with VLAN outer header::
4760 testpmd> set mplsoudp_encap-with-vlan ip-version ipv6 mask 4 udp-src 5
4761 udp-dst 10 ip-src ::1 ip-dst ::2222 vlan-tci 34
4762 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4763 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4764 mplsoudp_encap / end
4766 Sample MPLSoUDP decapsulation rule
4767 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4769 MPLSoUDP decapsulation outer layer has default value pre-configured in testpmd
4770 source code, those can be changed by using the following commands
4772 IPv4 MPLSoUDP outer header::
4774 testpmd> set mplsoudp_decap ip-version ipv4
4775 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / mpls / end actions
4776 mplsoudp_decap / l2_encap / end
4778 IPv4 MPLSoUDP with VLAN outer header::
4780 testpmd> set mplsoudp_decap-with-vlan ip-version ipv4
4781 testpmd> flow create 0 ingress pattern eth / vlan / ipv4 / udp / mpls / end
4782 actions mplsoudp_decap / l2_encap / end
4784 IPv6 MPLSoUDP outer header::
4786 testpmd> set mplsoudp_decap ip-version ipv6
4787 testpmd> flow create 0 ingress pattern eth / ipv6 / udp / mpls / end
4788 actions mplsoudp_decap / l2_encap / end
4790 IPv6 MPLSoUDP with VLAN outer header::
4792 testpmd> set mplsoudp_decap-with-vlan ip-version ipv6
4793 testpmd> flow create 0 ingress pattern eth / vlan / ipv6 / udp / mpls / end
4794 actions mplsoudp_decap / l2_encap / end
4796 Sample Raw encapsulation rule
4797 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4799 Raw encapsulation configuration can be set by the following commands
4801 Eecapsulating VxLAN::
4803 testpmd> set raw_encap 4 eth src is 10:11:22:33:44:55 / vlan tci is 1
4804 inner_type is 0x0800 / ipv4 / udp dst is 4789 / vxlan vni
4806 testpmd> flow create 0 egress pattern eth / ipv4 / end actions
4807 raw_encap index 4 / end
4809 Sample Raw decapsulation rule
4810 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4812 Raw decapsulation configuration can be set by the following commands
4814 Decapsulating VxLAN::
4816 testpmd> set raw_decap eth / ipv4 / udp / vxlan / end_set
4817 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / vxlan / eth / ipv4 /
4818 end actions raw_decap / queue index 0 / end
4823 ESP rules can be created by the following commands::
4825 testpmd> flow create 0 ingress pattern eth / ipv4 / esp spi is 1 / end actions
4827 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / esp spi is 1 / end
4828 actions queue index 3 / end
4829 testpmd> flow create 0 ingress pattern eth / ipv6 / esp spi is 1 / end actions
4831 testpmd> flow create 0 ingress pattern eth / ipv6 / udp / esp spi is 1 / end
4832 actions queue index 3 / end
4837 AH rules can be created by the following commands::
4839 testpmd> flow create 0 ingress pattern eth / ipv4 / ah spi is 1 / end actions
4841 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / ah spi is 1 / end
4842 actions queue index 3 / end
4843 testpmd> flow create 0 ingress pattern eth / ipv6 / ah spi is 1 / end actions
4845 testpmd> flow create 0 ingress pattern eth / ipv6 / udp / ah spi is 1 / end
4846 actions queue index 3 / end
4851 PFCP rules can be created by the following commands(s_field need to be 1
4854 testpmd> flow create 0 ingress pattern eth / ipv4 / pfcp s_field is 0 / end
4855 actions queue index 3 / end
4856 testpmd> flow create 0 ingress pattern eth / ipv4 / pfcp s_field is 1
4857 seid is 1 / end actions queue index 3 / end
4858 testpmd> flow create 0 ingress pattern eth / ipv6 / pfcp s_field is 0 / end
4859 actions queue index 3 / end
4860 testpmd> flow create 0 ingress pattern eth / ipv6 / pfcp s_field is 1
4861 seid is 1 / end actions queue index 3 / end
4863 Sample Sampling/Mirroring rules
4864 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4866 Sample/Mirroring rules can be set by the following commands
4868 NIC-RX Sampling rule, the matched ingress packets and sent to the queue 1,
4869 and 50% packets are duplicated and marked with 0x1234 and sent to queue 0.
4873 testpmd> set sample_actions 0 mark id 0x1234 / queue index 0 / end
4874 testpmd> flow create 0 ingress group 1 pattern eth / end actions
4875 sample ratio 2 index 0 / queue index 1 / end
4877 Mirroring rule with port representors (with "transfer" attribute), the matched
4878 ingress packets with encapsulation header are sent to port id 0, and also
4879 mirrored the packets and sent to port id 2.
4883 testpmd> set sample_actions 0 port_id id 2 / end
4884 testpmd> flow create 1 ingress transfer pattern eth / end actions
4885 sample ratio 1 index 0 / raw_encap / port_id id 0 / end
4887 Mirroring rule with port representors (with "transfer" attribute), the matched
4888 ingress packets are sent to port id 2, and also mirrored the packets with
4889 encapsulation header and sent to port id 0.
4893 testpmd> set sample_actions 0 raw_encap / port_id id 0 / end
4894 testpmd> flow create 0 ingress transfer pattern eth / end actions
4895 sample ratio 1 index 0 / port_id id 2 / end
4897 Mirroring rule with port representors (with "transfer" attribute), the matched
4898 ingress packets are sent to port id 2, and also mirrored the packets with
4899 VXLAN encapsulation header and sent to port id 0.
4903 testpmd> set vxlan ip-version ipv4 vni 4 udp-src 4 udp-dst 4 ip-src 127.0.0.1
4904 ip-dst 128.0.0.1 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4905 testpmd> set sample_actions 0 vxlan_encap / port_id id 0 / end
4906 testpmd> flow create 0 ingress transfer pattern eth / end actions
4907 sample ratio 1 index 0 / port_id id 2 / end
4909 Mirroring rule with port representors (with "transfer" attribute), the matched
4910 ingress packets are sent to port id 2, and also mirrored the packets with
4911 NVGRE encapsulation header and sent to port id 0.
4915 testpmd> set nvgre ip-version ipv4 tni 4 ip-src 127.0.0.1 ip-dst 128.0.0.1
4916 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4917 testpmd> set sample_actions 0 nvgre_encap / port_id id 0 / end
4918 testpmd> flow create 0 ingress transfer pattern eth / end actions
4919 sample ratio 1 index 0 / port_id id 2 / end
4921 Sample integrity rules
4922 ~~~~~~~~~~~~~~~~~~~~~~
4924 Integrity rules can be created by the following commands:
4926 Integrity rule that forwards valid TCP packets to group 1.
4927 TCP packet integrity is matched with the ``l4_ok`` bit 3.
4931 testpmd> flow create 0 ingress
4932 pattern eth / ipv4 / tcp / integrity value mask 8 value spec 8 / end
4933 actions jump group 1 / end
4935 Integrity rule that forwards invalid packets to application.
4936 General packet integrity is matched with the ``packet_ok`` bit 0.
4940 testpmd> flow create 0 ingress pattern integrity value mask 1 value spec 0 / end actions queue index 0 / end
4942 Sample conntrack rules
4943 ~~~~~~~~~~~~~~~~~~~~~~
4945 Conntrack rules can be set by the following commands
4947 Need to construct the connection context with provided information.
4948 In the first table, create a flow rule by using conntrack action and jump to
4949 the next table. In the next table, create a rule to check the state.
4953 testpmd> set conntrack com peer 1 is_orig 1 enable 1 live 1 sack 1 cack 0
4954 last_dir 0 liberal 0 state 1 max_ack_win 7 r_lim 5 last_win 510
4955 last_seq 2632987379 last_ack 2532480967 last_end 2632987379
4957 testpmd> set conntrack orig scale 7 fin 0 acked 1 unack_data 0
4958 sent_end 2632987379 reply_end 2633016339 max_win 28960
4960 testpmd> set conntrack rply scale 7 fin 0 acked 1 unack_data 0
4961 sent_end 2532480967 reply_end 2532546247 max_win 65280
4963 testpmd> flow indirect_action 0 create ingress action conntrack / end
4964 testpmd> flow create 0 group 3 ingress pattern eth / ipv4 / tcp / end actions indirect 0 / jump group 5 / end
4965 testpmd> flow create 0 group 5 ingress pattern eth / ipv4 / tcp / conntrack is 1 / end actions queue index 5 / end
4967 Construct the conntrack again with only "is_orig" set to 0 (other fields are
4968 ignored), then use "update" interface to update the direction. Create flow
4969 rules like above for the peer port.
4973 testpmd> flow indirect_action 0 update 0 action conntrack_update dir / end
4978 The following sections show functions to load/unload eBPF based filters.
4983 Load an eBPF program as a callback for particular RX/TX queue::
4985 testpmd> bpf-load rx|tx (portid) (queueid) (load-flags) (bpf-prog-filename)
4987 The available load-flags are:
4989 * ``J``: use JIT generated native code, otherwise BPF interpreter will be used.
4991 * ``M``: assume input parameter is a pointer to rte_mbuf, otherwise assume it is a pointer to first segment's data.
4997 You'll need clang v3.7 or above to build bpf program you'd like to load
5001 .. code-block:: console
5004 clang -O2 -target bpf -c t1.c
5006 Then to load (and JIT compile) t1.o at RX queue 0, port 1:
5008 .. code-block:: console
5010 testpmd> bpf-load rx 1 0 J ./dpdk.org/examples/bpf/t1.o
5012 To load (not JITed) t1.o at TX queue 0, port 0:
5014 .. code-block:: console
5016 testpmd> bpf-load tx 0 0 - ./dpdk.org/examples/bpf/t1.o
5021 Unload previously loaded eBPF program for particular RX/TX queue::
5023 testpmd> bpf-unload rx|tx (portid) (queueid)
5025 For example to unload BPF filter from TX queue 0, port 0:
5027 .. code-block:: console
5029 testpmd> bpf-unload tx 0 0