1 .. SPDX-License-Identifier: BSD-3-Clause
2 Copyright(c) 2010-2016 Intel Corporation.
6 Testpmd Runtime Functions
7 =========================
9 Where the testpmd application is started in interactive mode, (``-i|--interactive``),
10 it displays a prompt that can be used to start and stop forwarding,
11 configure the application, display statistics (including the extended NIC
12 statistics aka xstats) , set the Flow Director and other tasks::
16 The testpmd prompt has some, limited, readline support.
17 Common bash command-line functions such as ``Ctrl+a`` and ``Ctrl+e`` to go to the start and end of the prompt line are supported
18 as well as access to the command history via the up-arrow.
20 There is also support for tab completion.
21 If you type a partial command and hit ``<TAB>`` you get a list of the available completions:
23 .. code-block:: console
25 testpmd> show port <TAB>
27 info [Mul-choice STRING]: show|clear port info|stats|xstats|fdir|stat_qmap|dcb_tc|cap X
28 info [Mul-choice STRING]: show|clear port info|stats|xstats|fdir|stat_qmap|dcb_tc|cap all
29 stats [Mul-choice STRING]: show|clear port info|stats|xstats|fdir|stat_qmap|dcb_tc|cap X
30 stats [Mul-choice STRING]: show|clear port info|stats|xstats|fdir|stat_qmap|dcb_tc|cap all
36 Some examples in this document are too long to fit on one line are shown wrapped at `"\\"` for display purposes::
38 testpmd> set flow_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \
39 (pause_time) (send_xon) (port_id)
41 In the real ``testpmd>`` prompt these commands should be on a single line.
46 The testpmd has on-line help for the functions that are available at runtime.
47 These are divided into sections and can be accessed using help, help section or help all:
49 .. code-block:: console
53 help control : Start and stop forwarding.
54 help display : Displaying port, stats and config information.
55 help config : Configuration information.
56 help ports : Configuring ports.
57 help registers : Reading and setting port registers.
58 help filters : Filters configuration help.
59 help all : All of the above sections.
62 Command File Functions
63 ----------------------
65 To facilitate loading large number of commands or to avoid cutting and pasting where not
66 practical or possible testpmd supports alternative methods for executing commands.
68 * If started with the ``--cmdline-file=FILENAME`` command line argument testpmd
69 will execute all CLI commands contained within the file immediately before
70 starting packet forwarding or entering interactive mode.
72 .. code-block:: console
74 ./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|stat_qmap|dcb_tc|cap) (port_id|all)
164 The available information categories are:
166 * ``info``: General port information such as MAC address.
168 * ``summary``: Brief port summary such as Device Name, Driver Name etc.
170 * ``stats``: RX/TX statistics.
172 * ``xstats``: RX/TX extended NIC statistics.
174 * ``fdir``: Flow Director information and statistics.
176 * ``stat_qmap``: Queue statistics mapping.
178 * ``dcb_tc``: DCB information such as TC mapping.
180 * ``cap``: Supported offload capabilities.
184 .. code-block:: console
186 testpmd> show port info 0
188 ********************* Infos for port 0 *********************
190 MAC address: XX:XX:XX:XX:XX:XX
192 memory allocation on the socket: 0
194 Link speed: 40000 Mbps
195 Link duplex: full-duplex
196 Promiscuous mode: enabled
197 Allmulticast mode: disabled
198 Maximum number of MAC addresses: 64
199 Maximum number of MAC addresses of hash filtering: 0
201 strip on, filter on, extend off, qinq strip off
202 Redirection table size: 512
203 Supported flow types:
221 show port (module_eeprom|eeprom)
222 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
224 Display the EEPROM information of a port::
226 testpmd> show port (port_id) (module_eeprom|eeprom)
231 Display the rss redirection table entry indicated by masks on port X::
233 testpmd> show port (port_id) rss reta (size) (mask0, mask1...)
235 size is used to indicate the hardware supported reta size
240 Display the RSS hash functions and RSS hash key of a port::
242 testpmd> show port (port_id) rss-hash [key]
247 Clear the port statistics and forward engine statistics for a given port or for all ports::
249 testpmd> clear port (info|stats|xstats|fdir|stat_qmap) (port_id|all)
253 testpmd> clear port stats all
258 Display information for a given port's RX/TX queue::
260 testpmd> show (rxq|txq) info (port_id) (queue_id)
262 show desc status(rxq|txq)
263 ~~~~~~~~~~~~~~~~~~~~~~~~~
265 Display information for a given port's RX/TX descriptor status::
267 testpmd> show port (port_id) (rxq|txq) (queue_id) desc (desc_id) status
269 show rxq desc used count
270 ~~~~~~~~~~~~~~~~~~~~~~~~
272 Display the number of receive packet descriptors currently filled by hardware
273 and ready to be processed by the driver on a given RX queue::
275 testpmd> show port (port_id) rxq (queue_id) desc used count
280 Displays the configuration of the application.
281 The configuration comes from the command-line, the runtime or the application defaults::
283 testpmd> show config (rxtx|cores|fwd|rxoffs|rxpkts|txpkts|txtimes)
285 The available information categories are:
287 * ``rxtx``: RX/TX configuration items.
289 * ``cores``: List of forwarding cores.
291 * ``fwd``: Packet forwarding configuration.
293 * ``rxoffs``: Packet offsets for RX split.
295 * ``rxpkts``: Packets to RX split configuration.
297 * ``txpkts``: Packets to TX configuration.
299 * ``txtimes``: Burst time pattern for Tx only mode.
303 .. code-block:: console
305 testpmd> show config rxtx
307 io packet forwarding - CRC stripping disabled - packets/burst=16
308 nb forwarding cores=2 - nb forwarding ports=1
309 RX queues=1 - RX desc=128 - RX free threshold=0
310 RX threshold registers: pthresh=8 hthresh=8 wthresh=4
311 TX queues=1 - TX desc=512 - TX free threshold=0
312 TX threshold registers: pthresh=36 hthresh=0 wthresh=0
313 TX RS bit threshold=0 - TXQ flags=0x0
318 Set the packet forwarding mode::
320 testpmd> set fwd (io|mac|macswap|flowgen| \
321 rxonly|txonly|csum|icmpecho|noisy|5tswap) (""|retry)
323 ``retry`` can be specified for forwarding engines except ``rx_only``.
325 The available information categories are:
327 * ``io``: Forwards packets "as-is" in I/O mode.
328 This is the fastest possible forwarding operation as it does not access packets data.
329 This is the default mode.
331 * ``mac``: Changes the source and the destination Ethernet addresses of packets before forwarding them.
332 Default application behavior is to set source Ethernet address to that of the transmitting interface, and destination
333 address to a dummy value (set during init). The user may specify a target destination Ethernet address via the 'eth-peer' or
334 'eth-peers-configfile' command-line options. It is not currently possible to specify a specific source Ethernet address.
336 * ``macswap``: MAC swap forwarding mode.
337 Swaps the source and the destination Ethernet addresses of packets before forwarding them.
339 * ``flowgen``: Multi-flow generation mode.
340 Originates a number of flows (with varying destination IP addresses), and terminate receive traffic.
342 * ``rxonly``: Receives packets but doesn't transmit them.
344 * ``txonly``: Generates and transmits packets without receiving any.
346 * ``csum``: Changes the checksum field with hardware or software methods depending on the offload flags on the packet.
348 * ``icmpecho``: Receives a burst of packets, lookup for ICMP echo requests and, if any, send back ICMP echo replies.
350 * ``ieee1588``: Demonstrate L2 IEEE1588 V2 PTP timestamping for RX and TX.
352 * ``noisy``: Noisy neighbor simulation.
353 Simulate more realistic behavior of a guest machine engaged in receiving
354 and sending packets performing Virtual Network Function (VNF).
356 * ``5tswap``: Swap the source and destination of L2,L3,L4 if they exist.
358 L2 swaps the source address and destination address of Ethernet, as same as ``macswap``.
360 L3 swaps the source address and destination address of IP (v4 and v6).
362 L4 swaps the source port and destination port of transport layer (TCP and UDP).
366 testpmd> set fwd rxonly
368 Set rxonly packet forwarding mode
374 When running, forwarding engines maintain statistics from the time they have been started.
375 Example for the io forwarding engine, with some packet drops on the tx side::
377 testpmd> show fwd stats all
379 ------- Forward Stats for RX Port= 0/Queue= 0 -> TX Port= 1/Queue= 0 -------
380 RX-packets: 274293770 TX-packets: 274293642 TX-dropped: 128
382 ------- Forward Stats for RX Port= 1/Queue= 0 -> TX Port= 0/Queue= 0 -------
383 RX-packets: 274301850 TX-packets: 274301850 TX-dropped: 0
385 ---------------------- Forward statistics for port 0 ----------------------
386 RX-packets: 274293802 RX-dropped: 0 RX-total: 274293802
387 TX-packets: 274301862 TX-dropped: 0 TX-total: 274301862
388 ----------------------------------------------------------------------------
390 ---------------------- Forward statistics for port 1 ----------------------
391 RX-packets: 274301894 RX-dropped: 0 RX-total: 274301894
392 TX-packets: 274293706 TX-dropped: 128 TX-total: 274293834
393 ----------------------------------------------------------------------------
395 +++++++++++++++ Accumulated forward statistics for all ports+++++++++++++++
396 RX-packets: 548595696 RX-dropped: 0 RX-total: 548595696
397 TX-packets: 548595568 TX-dropped: 128 TX-total: 548595696
398 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
404 Clear the forwarding engines statistics::
406 testpmd> clear fwd stats all
411 Display an RX descriptor for a port RX queue::
413 testpmd> read rxd (port_id) (queue_id) (rxd_id)
417 testpmd> read rxd 0 0 4
418 0x0000000B - 0x001D0180 / 0x0000000B - 0x001D0180
423 Display a TX descriptor for a port TX queue::
425 testpmd> read txd (port_id) (queue_id) (txd_id)
429 testpmd> read txd 0 0 4
430 0x00000001 - 0x24C3C440 / 0x000F0000 - 0x2330003C
435 Get loaded dynamic device personalization (DDP) package info list::
437 testpmd> ddp get list (port_id)
442 Display information about dynamic device personalization (DDP) profile::
444 testpmd> ddp get info (profile_path)
449 Display VF statistics::
451 testpmd> show vf stats (port_id) (vf_id)
456 Reset VF statistics::
458 testpmd> clear vf stats (port_id) (vf_id)
460 show port pctype mapping
461 ~~~~~~~~~~~~~~~~~~~~~~~~
463 List all items from the pctype mapping table::
465 testpmd> show port (port_id) pctype mapping
467 show rx offloading capabilities
468 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
470 List all per queue and per port Rx offloading capabilities of a port::
472 testpmd> show port (port_id) rx_offload capabilities
474 show rx offloading configuration
475 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
477 List port level and all queue level Rx offloading configuration::
479 testpmd> show port (port_id) rx_offload configuration
481 show tx offloading capabilities
482 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
484 List all per queue and per port Tx offloading capabilities of a port::
486 testpmd> show port (port_id) tx_offload capabilities
488 show tx offloading configuration
489 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
491 List port level and all queue level Tx offloading configuration::
493 testpmd> show port (port_id) tx_offload configuration
495 show tx metadata setting
496 ~~~~~~~~~~~~~~~~~~~~~~~~
498 Show Tx metadata value set for a specific port::
500 testpmd> show port (port_id) tx_metadata
502 show port supported ptypes
503 ~~~~~~~~~~~~~~~~~~~~~~~~~~
505 Show ptypes supported for a specific port::
507 testpmd> show port (port_id) ptypes
509 set port supported ptypes
510 ~~~~~~~~~~~~~~~~~~~~~~~~~
512 set packet types classification for a specific port::
514 testpmd> set port (port_id) ptypes_mask (mask)
516 show port mac addresses info
517 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
519 Show mac addresses added for a specific port::
521 testpmd> show port (port_id) macs
524 show port multicast mac addresses info
525 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
527 Show multicast mac addresses added for a specific port::
529 testpmd> show port (port_id) mcast_macs
534 Show general information about devices probed::
536 testpmd> show device info (<identifier>|all)
540 .. code-block:: console
542 testpmd> show device info net_pcap0
544 ********************* Infos for device net_pcap0 *********************
546 Driver name: net_pcap
547 Devargs: iface=enP2p6s0,phy_mac=1
548 Connect to socket: -1
551 MAC address: 1E:37:93:28:04:B8
552 Device name: net_pcap0
557 Dumps all physical memory segment layouts::
559 testpmd> dump_physmem
564 Dumps the layout of all memory zones::
566 testpmd> dump_memzone
571 Dumps the memory usage of all sockets::
573 testpmd> dump_socket_mem
578 Dumps the size of all memory structures::
580 testpmd> dump_struct_sizes
585 Dumps the status of all or specific element in DPDK rings::
587 testpmd> dump_ring [ring_name]
592 Dumps the statistics of all or specific memory pool::
594 testpmd> dump_mempool [mempool_name]
599 Dumps the user device list::
601 testpmd> dump_devargs
606 Dumps the log level for all the dpdk modules::
608 testpmd> dump_log_types
610 show (raw_encap|raw_decap)
611 ~~~~~~~~~~~~~~~~~~~~~~~~~~
613 Display content of raw_encap/raw_decap buffers in hex::
615 testpmd> show <raw_encap|raw_decap> <index>
616 testpmd> show <raw_encap|raw_decap> all
620 testpmd> show raw_encap 6
622 index: 6 at [0x1c565b0], len=50
623 00000000: 00 00 00 00 00 00 16 26 36 46 56 66 08 00 45 00 | .......&6FVf..E.
624 00000010: 00 00 00 00 00 00 00 11 00 00 C0 A8 01 06 C0 A8 | ................
625 00000020: 03 06 00 00 00 FA 00 00 00 00 08 00 00 00 00 00 | ................
628 show fec capabilities
629 ~~~~~~~~~~~~~~~~~~~~~
631 Show fec capabilities of a port::
633 testpmd> show port (port_id) fec capabilities
638 Show fec mode of a port::
640 testpmd> show port (port_id) fec_mode
643 Configuration Functions
644 -----------------------
646 The testpmd application can be configured from the runtime as well as from the command-line.
648 This section details the available configuration functions that are available.
652 Configuration changes only become active when forwarding is started/restarted.
657 Reset forwarding to the default configuration::
664 Set the debug verbosity level::
666 testpmd> set verbose (level)
668 Available levels are as following:
670 * ``0`` silent except for error.
671 * ``1`` fully verbose except for Tx packets.
672 * ``2`` fully verbose except for Rx packets.
673 * ``> 2`` fully verbose.
678 Set the log level for a log type::
680 testpmd> set log global|(type) (level)
684 * ``type`` is the log name.
686 * ``level`` is the log level.
688 For example, to change the global log level::
690 testpmd> set log global (level)
692 Regexes can also be used for type. To change log level of user1, user2 and user3::
694 testpmd> set log user[1-3] (level)
699 Set the number of ports used by the application:
703 This is equivalent to the ``--nb-ports`` command-line option.
708 Set the number of cores used by the application::
710 testpmd> set nbcore (num)
712 This is equivalent to the ``--nb-cores`` command-line option.
716 The number of cores used must not be greater than number of ports used multiplied by the number of queues per port.
721 Set the forwarding cores hexadecimal mask::
723 testpmd> set coremask (mask)
725 This is equivalent to the ``--coremask`` command-line option.
729 The main lcore is reserved for command line parsing only and cannot be masked on for packet forwarding.
734 Set the forwarding ports hexadecimal mask::
736 testpmd> set portmask (mask)
738 This is equivalent to the ``--portmask`` command-line option.
740 set record-core-cycles
741 ~~~~~~~~~~~~~~~~~~~~~~
743 Set the recording of CPU cycles::
745 testpmd> set record-core-cycles (on|off)
749 * ``on`` enables measurement of CPU cycles per packet.
751 * ``off`` disables measurement of CPU cycles per packet.
753 This is equivalent to the ``--record-core-cycles command-line`` option.
755 set record-burst-stats
756 ~~~~~~~~~~~~~~~~~~~~~~
758 Set the displaying of RX and TX bursts::
760 testpmd> set record-burst-stats (on|off)
764 * ``on`` enables display of RX and TX bursts.
766 * ``off`` disables display of RX and TX bursts.
768 This is equivalent to the ``--record-burst-stats command-line`` option.
773 Set number of packets per burst::
775 testpmd> set burst (num)
777 This is equivalent to the ``--burst command-line`` option.
779 When retry is enabled, the transmit delay time and number of retries can also be set::
781 testpmd> set burst tx delay (microseconds) retry (num)
786 Set the offsets of segments relating to the data buffer beginning on receiving
787 if split feature is engaged. Affects only the queues configured with split
788 offloads (currently BUFFER_SPLIT is supported only).
790 testpmd> set rxoffs (x[,y]*)
792 Where x[,y]* represents a CSV list of values, without white space. If the list
793 of offsets is shorter than the list of segments the zero offsets will be used
794 for the remaining segments.
799 Set the length of segments to scatter packets on receiving if split
800 feature is engaged. Affects only the queues configured with split offloads
801 (currently BUFFER_SPLIT is supported only). Optionally the multiple memory
802 pools can be specified with --mbuf-size command line parameter and the mbufs
803 to receive will be allocated sequentially from these extra memory pools (the
804 mbuf for the first segment is allocated from the first pool, the second one
805 from the second pool, and so on, if segment number is greater then pool's the
806 mbuf for remaining segments will be allocated from the last valid pool).
808 testpmd> set rxpkts (x[,y]*)
810 Where x[,y]* represents a CSV list of values, without white space. Zero value
811 means to use the corresponding memory pool data buffer size.
816 Set the length of each segment of the TX-ONLY packets or length of packet for FLOWGEN mode::
818 testpmd> set txpkts (x[,y]*)
820 Where x[,y]* represents a CSV list of values, without white space.
825 Configure the timing burst pattern for Tx only mode. This command enables
826 the packet send scheduling on dynamic timestamp mbuf field and configures
827 timing pattern in Tx only mode. In this mode, if scheduling is enabled
828 application provides timestamps in the packets being sent. It is possible
829 to configure delay (in unspecified device clock units) between bursts
830 and between the packets within the burst::
832 testpmd> set txtimes (inter),(intra)
836 * ``inter`` is the delay between the bursts in the device clock units.
837 If ``intra`` is zero, this is the time between the beginnings of the
838 first packets in the neighbour bursts, if ``intra`` is not zero,
839 ``inter`` specifies the time between the beginning of the first packet
840 of the current burst and the beginning of the last packet of the
841 previous burst. If ``inter`` parameter is zero the send scheduling
842 on timestamps is disabled (default).
844 * ``intra`` is the delay between the packets within the burst specified
845 in the device clock units. The number of packets in the burst is defined
846 by regular burst setting. If ``intra`` parameter is zero no timestamps
847 provided in the packets excepting the first one in the burst.
849 As the result the bursts of packet will be transmitted with specific
850 delays between the packets within the burst and specific delay between
851 the bursts. The rte_eth_read_clock() must be supported by the device(s)
852 and is supposed to be engaged to get the current device clock value
853 and provide the reference for the timestamps. If there is no supported
854 rte_eth_read_clock() there will be no send scheduling provided on the port.
859 Set the split policy for the TX packets, applicable for TX-ONLY and CSUM forwarding modes::
861 testpmd> set txsplit (off|on|rand)
865 * ``off`` disable packet copy & split for CSUM mode.
867 * ``on`` split outgoing packet into multiple segments. Size of each segment
868 and number of segments per packet is determined by ``set txpkts`` command
871 * ``rand`` same as 'on', but number of segments per each packet is a random value between 1 and total number of segments.
876 Set the list of forwarding cores::
878 testpmd> set corelist (x[,y]*)
880 For example, to change the forwarding cores:
882 .. code-block:: console
884 testpmd> set corelist 3,1
885 testpmd> show config fwd
887 io packet forwarding - ports=2 - cores=2 - streams=2 - NUMA support disabled
888 Logical Core 3 (socket 0) forwards packets on 1 streams:
889 RX P=0/Q=0 (socket 0) -> TX P=1/Q=0 (socket 0) peer=02:00:00:00:00:01
890 Logical Core 1 (socket 0) forwards packets on 1 streams:
891 RX P=1/Q=0 (socket 0) -> TX P=0/Q=0 (socket 0) peer=02:00:00:00:00:00
895 The cores are used in the same order as specified on the command line.
900 Set the list of forwarding ports::
902 testpmd> set portlist (x[,y]*)
904 For example, to change the port forwarding:
906 .. code-block:: console
908 testpmd> set portlist 0,2,1,3
909 testpmd> show config fwd
911 io packet forwarding - ports=4 - cores=1 - streams=4
912 Logical Core 3 (socket 0) forwards packets on 4 streams:
913 RX P=0/Q=0 (socket 0) -> TX P=2/Q=0 (socket 0) peer=02:00:00:00:00:01
914 RX P=2/Q=0 (socket 0) -> TX P=0/Q=0 (socket 0) peer=02:00:00:00:00:00
915 RX P=1/Q=0 (socket 0) -> TX P=3/Q=0 (socket 0) peer=02:00:00:00:00:03
916 RX P=3/Q=0 (socket 0) -> TX P=1/Q=0 (socket 0) peer=02:00:00:00:00:02
921 Select how to retrieve new ports created after "port attach" command::
923 testpmd> set port setup on (iterator|event)
925 For each new port, a setup is done.
926 It will find the probed ports via RTE_ETH_FOREACH_MATCHING_DEV loop
927 in iterator mode, or via RTE_ETH_EVENT_NEW in event mode.
932 Enable/disable tx loopback::
934 testpmd> set tx loopback (port_id) (on|off)
939 set drop enable bit for all queues::
941 testpmd> set all queues drop (port_id) (on|off)
943 set split drop enable (for VF)
944 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
946 set split drop enable bit for VF from PF::
948 testpmd> set vf split drop (port_id) (vf_id) (on|off)
950 set mac antispoof (for VF)
951 ~~~~~~~~~~~~~~~~~~~~~~~~~~
953 Set mac antispoof for a VF from the PF::
955 testpmd> set vf mac antispoof (port_id) (vf_id) (on|off)
960 Enable/disable MACsec offload::
962 testpmd> set macsec offload (port_id) on encrypt (on|off) replay-protect (on|off)
963 testpmd> set macsec offload (port_id) off
968 Configure MACsec secure connection (SC)::
970 testpmd> set macsec sc (tx|rx) (port_id) (mac) (pi)
974 The pi argument is ignored for tx.
975 Check the NIC Datasheet for hardware limits.
980 Configure MACsec secure association (SA)::
982 testpmd> set macsec sa (tx|rx) (port_id) (idx) (an) (pn) (key)
986 The IDX value must be 0 or 1.
987 Check the NIC Datasheet for hardware limits.
989 set broadcast mode (for VF)
990 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
992 Set broadcast mode for a VF from the PF::
994 testpmd> set vf broadcast (port_id) (vf_id) (on|off)
999 Set the VLAN strip for a queue on a port::
1001 testpmd> vlan set stripq (on|off) (port_id,queue_id)
1003 vlan set stripq (for VF)
1004 ~~~~~~~~~~~~~~~~~~~~~~~~
1006 Set VLAN strip for all queues in a pool for a VF from the PF::
1008 testpmd> set vf vlan stripq (port_id) (vf_id) (on|off)
1010 vlan set insert (for VF)
1011 ~~~~~~~~~~~~~~~~~~~~~~~~
1013 Set VLAN insert for a VF from the PF::
1015 testpmd> set vf vlan insert (port_id) (vf_id) (vlan_id)
1017 vlan set tag (for VF)
1018 ~~~~~~~~~~~~~~~~~~~~~
1020 Set VLAN tag for a VF from the PF::
1022 testpmd> set vf vlan tag (port_id) (vf_id) (on|off)
1024 vlan set antispoof (for VF)
1025 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
1027 Set VLAN antispoof for a VF from the PF::
1029 testpmd> set vf vlan antispoof (port_id) (vf_id) (on|off)
1031 vlan set (strip|filter|qinq_strip|extend)
1032 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1033 Set the VLAN strip/filter/QinQ strip/extend on for a port::
1035 testpmd> vlan set (strip|filter|qinq_strip|extend) (on|off) (port_id)
1040 Set the inner or outer VLAN TPID for packet filtering on a port::
1042 testpmd> vlan set (inner|outer) tpid (value) (port_id)
1046 TPID value must be a 16-bit number (value <= 65536).
1051 Add a VLAN ID, or all identifiers, to the set of VLAN identifiers filtered by port ID::
1053 testpmd> rx_vlan add (vlan_id|all) (port_id)
1057 VLAN filter must be set on that port. VLAN ID < 4096.
1058 Depending on the NIC used, number of vlan_ids may be limited to the maximum entries
1059 in VFTA table. This is important if enabling all vlan_ids.
1064 Remove a VLAN ID, or all identifiers, from the set of VLAN identifiers filtered by port ID::
1066 testpmd> rx_vlan rm (vlan_id|all) (port_id)
1068 rx_vlan add (for VF)
1069 ~~~~~~~~~~~~~~~~~~~~
1071 Add a VLAN ID, to the set of VLAN identifiers filtered for VF(s) for port ID::
1073 testpmd> rx_vlan add (vlan_id) port (port_id) vf (vf_mask)
1078 Remove a VLAN ID, from the set of VLAN identifiers filtered for VF(s) for port ID::
1080 testpmd> rx_vlan rm (vlan_id) port (port_id) vf (vf_mask)
1085 Add an UDP port for VXLAN packet filter on a port::
1087 testpmd> rx_vxlan_port add (udp_port) (port_id)
1089 rx_vxlan_port remove
1090 ~~~~~~~~~~~~~~~~~~~~
1092 Remove an UDP port for VXLAN packet filter on a port::
1094 testpmd> rx_vxlan_port rm (udp_port) (port_id)
1099 Set hardware insertion of VLAN IDs in packets sent on a port::
1101 testpmd> tx_vlan set (port_id) vlan_id[, vlan_id_outer]
1103 For example, set a single VLAN ID (5) insertion on port 0::
1107 Or, set double VLAN ID (inner: 2, outer: 3) insertion on port 1::
1115 Set port based hardware insertion of VLAN ID in packets sent on a port::
1117 testpmd> tx_vlan set pvid (port_id) (vlan_id) (on|off)
1122 Disable hardware insertion of a VLAN header in packets sent on a port::
1124 testpmd> tx_vlan reset (port_id)
1129 Select hardware or software calculation of the checksum when
1130 transmitting a packet using the ``csum`` forwarding engine::
1132 testpmd> csum set (ip|udp|tcp|sctp|outer-ip|outer-udp) (hw|sw) (port_id)
1136 * ``ip|udp|tcp|sctp`` always relate to the inner layer.
1138 * ``outer-ip`` relates to the outer IP layer (only for IPv4) in the case where the packet is recognized
1139 as a tunnel packet by the forwarding engine (geneve, gre, gtp, ipip, vxlan and vxlan-gpe are
1140 supported). See also the ``csum parse-tunnel`` command.
1142 * ``outer-udp`` relates to the outer UDP layer in the case where the packet is recognized
1143 as a tunnel packet by the forwarding engine (geneve, gtp, vxlan and vxlan-gpe are
1144 supported). See also the ``csum parse-tunnel`` command.
1148 Check the NIC Datasheet for hardware limits.
1153 Set RSS queue region span on a port::
1155 testpmd> set port (port_id) queue-region region_id (value) \
1156 queue_start_index (value) queue_num (value)
1158 Set flowtype mapping on a RSS queue region on a port::
1160 testpmd> set port (port_id) queue-region region_id (value) flowtype (value)
1164 * For the flowtype(pctype) of packet,the specific index for each type has
1165 been defined in file i40e_type.h as enum i40e_filter_pctype.
1167 Set user priority mapping on a RSS queue region on a port::
1169 testpmd> set port (port_id) queue-region UP (value) region_id (value)
1171 Flush all queue region related configuration on a port::
1173 testpmd> set port (port_id) queue-region flush (on|off)
1177 * ``on``: is just an enable function which server for other configuration,
1178 it is for all configuration about queue region from up layer,
1179 at first will only keep in DPDK software stored in driver,
1180 only after "flush on", it commit all configuration to HW.
1182 * ``"off``: is just clean all configuration about queue region just now,
1183 and restore all to DPDK i40e driver default config when start up.
1185 Show all queue region related configuration info on a port::
1187 testpmd> show port (port_id) queue-region
1191 Queue region only support on PF by now, so these command is
1192 only for configuration of queue region on PF port.
1197 Define how tunneled packets should be handled by the csum forward
1200 testpmd> csum parse-tunnel (on|off) (tx_port_id)
1202 If enabled, the csum forward engine will try to recognize supported
1203 tunnel headers (geneve, gtp, gre, ipip, vxlan, vxlan-gpe).
1205 If disabled, treat tunnel packets as non-tunneled packets (a inner
1206 header is handled as a packet payload).
1210 The port argument is the TX port like in the ``csum set`` command.
1214 Consider a packet in packet like the following::
1216 eth_out/ipv4_out/udp_out/vxlan/eth_in/ipv4_in/tcp_in
1218 * If parse-tunnel is enabled, the ``ip|udp|tcp|sctp`` parameters of ``csum set``
1219 command relate to the inner headers (here ``ipv4_in`` and ``tcp_in``), and the
1220 ``outer-ip|outer-udp`` parameter relates to the outer headers (here ``ipv4_out`` and ``udp_out``).
1222 * If parse-tunnel is disabled, the ``ip|udp|tcp|sctp`` parameters of ``csum set``
1223 command relate to the outer headers, here ``ipv4_out`` and ``udp_out``.
1228 Display tx checksum offload configuration::
1230 testpmd> csum show (port_id)
1235 Enable TCP Segmentation Offload (TSO) in the ``csum`` forwarding engine::
1237 testpmd> tso set (segsize) (port_id)
1241 Check the NIC datasheet for hardware limits.
1246 Display the status of TCP Segmentation Offload::
1248 testpmd> tso show (port_id)
1253 Set tso segment size of tunneled packets for a port in csum engine::
1255 testpmd> tunnel_tso set (tso_segsz) (port_id)
1260 Display the status of tunneled TCP Segmentation Offload for a port::
1262 testpmd> tunnel_tso show (port_id)
1267 Enable or disable GRO in ``csum`` forwarding engine::
1269 testpmd> set port <port_id> gro on|off
1271 If enabled, the csum forwarding engine will perform GRO on the TCP/IPv4
1272 packets received from the given port.
1274 If disabled, packets received from the given port won't be performed
1275 GRO. By default, GRO is disabled for all ports.
1279 When enable GRO for a port, TCP/IPv4 packets received from the port
1280 will be performed GRO. After GRO, all merged packets have bad
1281 checksums, since the GRO library doesn't re-calculate checksums for
1282 the merged packets. Therefore, if users want the merged packets to
1283 have correct checksums, please select HW IP checksum calculation and
1284 HW TCP checksum calculation for the port which the merged packets are
1290 Display GRO configuration for a given port::
1292 testpmd> show port <port_id> gro
1297 Set the cycle to flush the GROed packets from reassembly tables::
1299 testpmd> set gro flush <cycles>
1301 When enable GRO, the csum forwarding engine performs GRO on received
1302 packets, and the GROed packets are stored in reassembly tables. Users
1303 can use this command to determine when the GROed packets are flushed
1304 from the reassembly tables.
1306 The ``cycles`` is measured in GRO operation times. The csum forwarding
1307 engine flushes the GROed packets from the tables every ``cycles`` GRO
1310 By default, the value of ``cycles`` is 1, which means flush GROed packets
1311 from the reassembly tables as soon as one GRO operation finishes. The value
1312 of ``cycles`` should be in the range of 1 to ``GRO_MAX_FLUSH_CYCLES``.
1314 Please note that the large value of ``cycles`` may cause the poor TCP/IP
1315 stack performance. Because the GROed packets are delayed to arrive the
1316 stack, thus causing more duplicated ACKs and TCP retransmissions.
1321 Toggle per-port GSO support in ``csum`` forwarding engine::
1323 testpmd> set port <port_id> gso on|off
1325 If enabled, the csum forwarding engine will perform GSO on supported IPv4
1326 packets, transmitted on the given port.
1328 If disabled, packets transmitted on the given port will not undergo GSO.
1329 By default, GSO is disabled for all ports.
1333 When GSO is enabled on a port, supported IPv4 packets transmitted on that
1334 port undergo GSO. Afterwards, the segmented packets are represented by
1335 multi-segment mbufs; however, the csum forwarding engine doesn't calculation
1336 of checksums for GSO'd segments in SW. As a result, if users want correct
1337 checksums in GSO segments, they should enable HW checksum calculation for
1340 For example, HW checksum calculation for VxLAN GSO'd packets may be enabled
1341 by setting the following options in the csum forwarding engine:
1343 testpmd> csum set outer_ip hw <port_id>
1345 testpmd> csum set ip hw <port_id>
1347 testpmd> csum set tcp hw <port_id>
1349 UDP GSO is the same as IP fragmentation, which treats the UDP header
1350 as the payload and does not modify it during segmentation. That is,
1351 after UDP GSO, only the first output fragment has the original UDP
1352 header. Therefore, users need to enable HW IP checksum calculation
1353 and SW UDP checksum calculation for GSO-enabled ports, if they want
1354 correct checksums for UDP/IPv4 packets.
1359 Set the maximum GSO segment size (measured in bytes), which includes the
1360 packet header and the packet payload for GSO-enabled ports (global)::
1362 testpmd> set gso segsz <length>
1367 Display the status of Generic Segmentation Offload for a given port::
1369 testpmd> show port <port_id> gso
1374 Add an alternative MAC address to a port::
1376 testpmd> mac_addr add (port_id) (XX:XX:XX:XX:XX:XX)
1381 Remove a MAC address from a port::
1383 testpmd> mac_addr remove (port_id) (XX:XX:XX:XX:XX:XX)
1388 To add the multicast MAC address to/from the set of multicast addresses
1391 testpmd> mcast_addr add (port_id) (mcast_addr)
1396 To remove the multicast MAC address to/from the set of multicast addresses
1399 testpmd> mcast_addr remove (port_id) (mcast_addr)
1401 mac_addr add (for VF)
1402 ~~~~~~~~~~~~~~~~~~~~~
1404 Add an alternative MAC address for a VF to a port::
1406 testpmd> mac_add add port (port_id) vf (vf_id) (XX:XX:XX:XX:XX:XX)
1411 Set the default MAC address for a port::
1413 testpmd> mac_addr set (port_id) (XX:XX:XX:XX:XX:XX)
1415 mac_addr set (for VF)
1416 ~~~~~~~~~~~~~~~~~~~~~
1418 Set the MAC address for a VF from the PF::
1420 testpmd> set vf mac addr (port_id) (vf_id) (XX:XX:XX:XX:XX:XX)
1425 Set the forwarding peer address for certain port::
1427 testpmd> set eth-peer (port_id) (peer_addr)
1429 This is equivalent to the ``--eth-peer`` command-line option.
1434 Set the unicast hash filter(s) on/off for a port::
1436 testpmd> set port (port_id) uta (XX:XX:XX:XX:XX:XX|all) (on|off)
1441 Set the promiscuous mode on for a port or for all ports.
1442 In promiscuous mode packets are not dropped if they aren't for the specified MAC address::
1444 testpmd> set promisc (port_id|all) (on|off)
1449 Set the allmulti mode for a port or for all ports::
1451 testpmd> set allmulti (port_id|all) (on|off)
1453 Same as the ifconfig (8) option. Controls how multicast packets are handled.
1455 set promisc (for VF)
1456 ~~~~~~~~~~~~~~~~~~~~
1458 Set the unicast promiscuous mode for a VF from PF.
1459 It's supported by Intel i40e NICs now.
1460 In promiscuous mode packets are not dropped if they aren't for the specified MAC address::
1462 testpmd> set vf promisc (port_id) (vf_id) (on|off)
1464 set allmulticast (for VF)
1465 ~~~~~~~~~~~~~~~~~~~~~~~~~
1467 Set the multicast promiscuous mode for a VF from PF.
1468 It's supported by Intel i40e NICs now.
1469 In promiscuous mode packets are not dropped if they aren't for the specified MAC address::
1471 testpmd> set vf allmulti (port_id) (vf_id) (on|off)
1473 set tx max bandwidth (for VF)
1474 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1476 Set TX max absolute bandwidth (Mbps) for a VF from PF::
1478 testpmd> set vf tx max-bandwidth (port_id) (vf_id) (max_bandwidth)
1480 set tc tx min bandwidth (for VF)
1481 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1483 Set all TCs' TX min relative bandwidth (%) for a VF from PF::
1485 testpmd> set vf tc tx min-bandwidth (port_id) (vf_id) (bw1, bw2, ...)
1487 set tc tx max bandwidth (for VF)
1488 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1490 Set a TC's TX max absolute bandwidth (Mbps) for a VF from PF::
1492 testpmd> set vf tc tx max-bandwidth (port_id) (vf_id) (tc_no) (max_bandwidth)
1494 set tc strict link priority mode
1495 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1497 Set some TCs' strict link priority mode on a physical port::
1499 testpmd> set tx strict-link-priority (port_id) (tc_bitmap)
1501 set tc tx min bandwidth
1502 ~~~~~~~~~~~~~~~~~~~~~~~
1504 Set all TCs' TX min relative bandwidth (%) globally for all PF and VFs::
1506 testpmd> set tc tx min-bandwidth (port_id) (bw1, bw2, ...)
1511 Set the link flow control parameter on a port::
1513 testpmd> set flow_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \
1514 (pause_time) (send_xon) mac_ctrl_frame_fwd (on|off) \
1515 autoneg (on|off) (port_id)
1519 * ``high_water`` (integer): High threshold value to trigger XOFF.
1521 * ``low_water`` (integer): Low threshold value to trigger XON.
1523 * ``pause_time`` (integer): Pause quota in the Pause frame.
1525 * ``send_xon`` (0/1): Send XON frame.
1527 * ``mac_ctrl_frame_fwd``: Enable receiving MAC control frames.
1529 * ``autoneg``: Change the auto-negotiation parameter.
1534 Set the priority flow control parameter on a port::
1536 testpmd> set pfc_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \
1537 (pause_time) (priority) (port_id)
1541 * ``high_water`` (integer): High threshold value.
1543 * ``low_water`` (integer): Low threshold value.
1545 * ``pause_time`` (integer): Pause quota in the Pause frame.
1547 * ``priority`` (0-7): VLAN User Priority.
1552 Set statistics mapping (qmapping 0..15) for RX/TX queue on port::
1554 testpmd> set stat_qmap (tx|rx) (port_id) (queue_id) (qmapping)
1556 For example, to set rx queue 2 on port 0 to mapping 5::
1558 testpmd>set stat_qmap rx 0 2 5
1560 set xstats-hide-zero
1561 ~~~~~~~~~~~~~~~~~~~~
1563 Set the option to hide zero values for xstats display::
1565 testpmd> set xstats-hide-zero on|off
1569 By default, the zero values are displayed for xstats.
1571 set port - rx/tx (for VF)
1572 ~~~~~~~~~~~~~~~~~~~~~~~~~
1574 Set VF receive/transmit from a port::
1576 testpmd> set port (port_id) vf (vf_id) (rx|tx) (on|off)
1578 set port - rx mode(for VF)
1579 ~~~~~~~~~~~~~~~~~~~~~~~~~~
1581 Set the VF receive mode of a port::
1583 testpmd> set port (port_id) vf (vf_id) \
1584 rxmode (AUPE|ROPE|BAM|MPE) (on|off)
1586 The available receive modes are:
1588 * ``AUPE``: Accepts untagged VLAN.
1590 * ``ROPE``: Accepts unicast hash.
1592 * ``BAM``: Accepts broadcast packets.
1594 * ``MPE``: Accepts all multicast packets.
1596 set port - tx_rate (for Queue)
1597 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1599 Set TX rate limitation for a queue on a port::
1601 testpmd> set port (port_id) queue (queue_id) rate (rate_value)
1603 set port - tx_rate (for VF)
1604 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
1606 Set TX rate limitation for queues in VF on a port::
1608 testpmd> set port (port_id) vf (vf_id) rate (rate_value) queue_mask (queue_mask)
1610 set port - mirror rule
1611 ~~~~~~~~~~~~~~~~~~~~~~
1613 Set pool or vlan type mirror rule for a port::
1615 testpmd> set port (port_id) mirror-rule (rule_id) \
1616 (pool-mirror-up|pool-mirror-down|vlan-mirror) \
1617 (poolmask|vlanid[,vlanid]*) dst-pool (pool_id) (on|off)
1619 Set link mirror rule for a port::
1621 testpmd> set port (port_id) mirror-rule (rule_id) \
1622 (uplink-mirror|downlink-mirror) dst-pool (pool_id) (on|off)
1624 For example to enable mirror traffic with vlan 0,1 to pool 0::
1626 set port 0 mirror-rule 0 vlan-mirror 0,1 dst-pool 0 on
1628 reset port - mirror rule
1629 ~~~~~~~~~~~~~~~~~~~~~~~~
1631 Reset a mirror rule for a port::
1633 testpmd> reset port (port_id) mirror-rule (rule_id)
1638 Set the flush on RX streams before forwarding.
1639 The default is flush ``on``.
1640 Mainly used with PCAP drivers to turn off the default behavior of flushing the first 512 packets on RX streams::
1642 testpmd> set flush_rx off
1647 Set the bypass mode for the lowest port on bypass enabled NIC::
1649 testpmd> set bypass mode (normal|bypass|isolate) (port_id)
1654 Set the event required to initiate specified bypass mode for the lowest port on a bypass enabled::
1656 testpmd> set bypass event (timeout|os_on|os_off|power_on|power_off) \
1657 mode (normal|bypass|isolate) (port_id)
1661 * ``timeout``: Enable bypass after watchdog timeout.
1663 * ``os_on``: Enable bypass when OS/board is powered on.
1665 * ``os_off``: Enable bypass when OS/board is powered off.
1667 * ``power_on``: Enable bypass when power supply is turned on.
1669 * ``power_off``: Enable bypass when power supply is turned off.
1675 Set the bypass watchdog timeout to ``n`` seconds where 0 = instant::
1677 testpmd> set bypass timeout (0|1.5|2|3|4|8|16|32)
1682 Show the bypass configuration for a bypass enabled NIC using the lowest port on the NIC::
1684 testpmd> show bypass config (port_id)
1689 Set link up for a port::
1691 testpmd> set link-up port (port id)
1696 Set link down for a port::
1698 testpmd> set link-down port (port id)
1703 Enable E-tag insertion for a VF on a port::
1705 testpmd> E-tag set insertion on port-tag-id (value) port (port_id) vf (vf_id)
1707 Disable E-tag insertion for a VF on a port::
1709 testpmd> E-tag set insertion off port (port_id) vf (vf_id)
1711 Enable/disable E-tag stripping on a port::
1713 testpmd> E-tag set stripping (on|off) port (port_id)
1715 Enable/disable E-tag based forwarding on a port::
1717 testpmd> E-tag set forwarding (on|off) port (port_id)
1722 Load a dynamic device personalization (DDP) profile and store backup profile::
1724 testpmd> ddp add (port_id) (profile_path[,backup_profile_path])
1729 Delete a dynamic device personalization profile and restore backup profile::
1731 testpmd> ddp del (port_id) (backup_profile_path)
1736 List all items from the ptype mapping table::
1738 testpmd> ptype mapping get (port_id) (valid_only)
1742 * ``valid_only``: A flag indicates if only list valid items(=1) or all itemss(=0).
1744 Replace a specific or a group of software defined ptype with a new one::
1746 testpmd> ptype mapping replace (port_id) (target) (mask) (pkt_type)
1750 * ``target``: A specific software ptype or a mask to represent a group of software ptypes.
1752 * ``mask``: A flag indicate if "target" is a specific software ptype(=0) or a ptype mask(=1).
1754 * ``pkt_type``: The new software ptype to replace the old ones.
1756 Update hardware defined ptype to software defined packet type mapping table::
1758 testpmd> ptype mapping update (port_id) (hw_ptype) (sw_ptype)
1762 * ``hw_ptype``: hardware ptype as the index of the ptype mapping table.
1764 * ``sw_ptype``: software ptype as the value of the ptype mapping table.
1766 Reset ptype mapping table::
1768 testpmd> ptype mapping reset (port_id)
1770 config per port Rx offloading
1771 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1773 Enable or disable a per port Rx offloading on all Rx queues of a port::
1775 testpmd> port config (port_id) rx_offload (offloading) on|off
1777 * ``offloading``: can be any of these offloading capability:
1778 vlan_strip, ipv4_cksum, udp_cksum, tcp_cksum, tcp_lro,
1779 qinq_strip, outer_ipv4_cksum, macsec_strip,
1780 header_split, vlan_filter, vlan_extend, jumbo_frame,
1781 scatter, timestamp, security, keep_crc, rss_hash
1783 This command should be run when the port is stopped, or else it will fail.
1785 config per queue Rx offloading
1786 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1788 Enable or disable a per queue Rx offloading only on a specific Rx queue::
1790 testpmd> port (port_id) rxq (queue_id) rx_offload (offloading) on|off
1792 * ``offloading``: can be any of these offloading capability:
1793 vlan_strip, ipv4_cksum, udp_cksum, tcp_cksum, tcp_lro,
1794 qinq_strip, outer_ipv4_cksum, macsec_strip,
1795 header_split, vlan_filter, vlan_extend, jumbo_frame,
1796 scatter, timestamp, security, keep_crc
1798 This command should be run when the port is stopped, or else it will fail.
1800 config per port Tx offloading
1801 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1803 Enable or disable a per port Tx offloading on all Tx queues of a port::
1805 testpmd> port config (port_id) tx_offload (offloading) on|off
1807 * ``offloading``: can be any of these offloading capability:
1808 vlan_insert, ipv4_cksum, udp_cksum, tcp_cksum,
1809 sctp_cksum, tcp_tso, udp_tso, outer_ipv4_cksum,
1810 qinq_insert, vxlan_tnl_tso, gre_tnl_tso,
1811 ipip_tnl_tso, geneve_tnl_tso, macsec_insert,
1812 mt_lockfree, multi_segs, mbuf_fast_free, security
1814 This command should be run when the port is stopped, or else it will fail.
1816 config per queue Tx offloading
1817 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1819 Enable or disable a per queue Tx offloading only on a specific Tx queue::
1821 testpmd> port (port_id) txq (queue_id) tx_offload (offloading) on|off
1823 * ``offloading``: can be any of these offloading capability:
1824 vlan_insert, ipv4_cksum, udp_cksum, tcp_cksum,
1825 sctp_cksum, tcp_tso, udp_tso, outer_ipv4_cksum,
1826 qinq_insert, vxlan_tnl_tso, gre_tnl_tso,
1827 ipip_tnl_tso, geneve_tnl_tso, macsec_insert,
1828 mt_lockfree, multi_segs, mbuf_fast_free, security
1830 This command should be run when the port is stopped, or else it will fail.
1832 Config VXLAN Encap outer layers
1833 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1835 Configure the outer layer to encapsulate a packet inside a VXLAN tunnel::
1837 set vxlan ip-version (ipv4|ipv6) vni (vni) udp-src (udp-src) \
1838 udp-dst (udp-dst) ip-src (ip-src) ip-dst (ip-dst) eth-src (eth-src) \
1841 set vxlan-with-vlan ip-version (ipv4|ipv6) vni (vni) udp-src (udp-src) \
1842 udp-dst (udp-dst) ip-src (ip-src) ip-dst (ip-dst) vlan-tci (vlan-tci) \
1843 eth-src (eth-src) eth-dst (eth-dst)
1845 set vxlan-tos-ttl ip-version (ipv4|ipv6) vni (vni) udp-src (udp-src) \
1846 udp-dst (udp-dst) ip-tos (ip-tos) ip-ttl (ip-ttl) ip-src (ip-src) \
1847 ip-dst (ip-dst) eth-src (eth-src) eth-dst (eth-dst)
1849 These commands will set an internal configuration inside testpmd, any following
1850 flow rule using the action vxlan_encap will use the last configuration set.
1851 To have a different encapsulation header, one of those commands must be called
1852 before the flow rule creation.
1854 Config NVGRE Encap outer layers
1855 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1857 Configure the outer layer to encapsulate a packet inside a NVGRE tunnel::
1859 set nvgre ip-version (ipv4|ipv6) tni (tni) ip-src (ip-src) ip-dst (ip-dst) \
1860 eth-src (eth-src) eth-dst (eth-dst)
1861 set nvgre-with-vlan ip-version (ipv4|ipv6) tni (tni) ip-src (ip-src) \
1862 ip-dst (ip-dst) vlan-tci (vlan-tci) eth-src (eth-src) eth-dst (eth-dst)
1864 These commands will set an internal configuration inside testpmd, any following
1865 flow rule using the action nvgre_encap will use the last configuration set.
1866 To have a different encapsulation header, one of those commands must be called
1867 before the flow rule creation.
1872 Configure the l2 to be used when encapsulating a packet with L2::
1874 set l2_encap ip-version (ipv4|ipv6) eth-src (eth-src) eth-dst (eth-dst)
1875 set l2_encap-with-vlan ip-version (ipv4|ipv6) vlan-tci (vlan-tci) \
1876 eth-src (eth-src) eth-dst (eth-dst)
1878 Those commands will set an internal configuration inside testpmd, any following
1879 flow rule using the action l2_encap will use the last configuration set.
1880 To have a different encapsulation header, one of those commands must be called
1881 before the flow rule creation.
1886 Configure the l2 to be removed when decapsulating a packet with L2::
1888 set l2_decap ip-version (ipv4|ipv6)
1889 set l2_decap-with-vlan ip-version (ipv4|ipv6)
1891 Those commands will set an internal configuration inside testpmd, any following
1892 flow rule using the action l2_decap will use the last configuration set.
1893 To have a different encapsulation header, one of those commands must be called
1894 before the flow rule creation.
1896 Config MPLSoGRE Encap outer layers
1897 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1899 Configure the outer layer to encapsulate a packet inside a MPLSoGRE tunnel::
1901 set mplsogre_encap ip-version (ipv4|ipv6) label (label) \
1902 ip-src (ip-src) ip-dst (ip-dst) eth-src (eth-src) eth-dst (eth-dst)
1903 set mplsogre_encap-with-vlan ip-version (ipv4|ipv6) label (label) \
1904 ip-src (ip-src) ip-dst (ip-dst) vlan-tci (vlan-tci) \
1905 eth-src (eth-src) eth-dst (eth-dst)
1907 These commands will set an internal configuration inside testpmd, any following
1908 flow rule using the action mplsogre_encap will use the last configuration set.
1909 To have a different encapsulation header, one of those commands must be called
1910 before the flow rule creation.
1912 Config MPLSoGRE Decap outer layers
1913 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1915 Configure the outer layer to decapsulate MPLSoGRE packet::
1917 set mplsogre_decap ip-version (ipv4|ipv6)
1918 set mplsogre_decap-with-vlan ip-version (ipv4|ipv6)
1920 These commands will set an internal configuration inside testpmd, any following
1921 flow rule using the action mplsogre_decap will use the last configuration set.
1922 To have a different decapsulation header, one of those commands must be called
1923 before the flow rule creation.
1925 Config MPLSoUDP Encap outer layers
1926 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1928 Configure the outer layer to encapsulate a packet inside a MPLSoUDP tunnel::
1930 set mplsoudp_encap ip-version (ipv4|ipv6) label (label) udp-src (udp-src) \
1931 udp-dst (udp-dst) ip-src (ip-src) ip-dst (ip-dst) \
1932 eth-src (eth-src) eth-dst (eth-dst)
1933 set mplsoudp_encap-with-vlan ip-version (ipv4|ipv6) label (label) \
1934 udp-src (udp-src) udp-dst (udp-dst) ip-src (ip-src) ip-dst (ip-dst) \
1935 vlan-tci (vlan-tci) eth-src (eth-src) eth-dst (eth-dst)
1937 These commands will set an internal configuration inside testpmd, any following
1938 flow rule using the action mplsoudp_encap will use the last configuration set.
1939 To have a different encapsulation header, one of those commands must be called
1940 before the flow rule creation.
1942 Config MPLSoUDP Decap outer layers
1943 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1945 Configure the outer layer to decapsulate MPLSoUDP packet::
1947 set mplsoudp_decap ip-version (ipv4|ipv6)
1948 set mplsoudp_decap-with-vlan ip-version (ipv4|ipv6)
1950 These commands will set an internal configuration inside testpmd, any following
1951 flow rule using the action mplsoudp_decap will use the last configuration set.
1952 To have a different decapsulation header, one of those commands must be called
1953 before the flow rule creation.
1955 Config Raw Encapsulation
1956 ~~~~~~~~~~~~~~~~~~~~~~~~~
1958 Configure the raw data to be used when encapsulating a packet by
1959 rte_flow_action_raw_encap::
1961 set raw_encap {index} {item} [/ {item} [...]] / end_set
1963 There are multiple global buffers for ``raw_encap``, this command will set one
1964 internal buffer index by ``{index}``.
1965 If there is no ``{index}`` specified::
1967 set raw_encap {item} [/ {item} [...]] / end_set
1969 the default index ``0`` is used.
1970 In order to use different encapsulating header, ``index`` must be specified
1971 during the flow rule creation::
1973 testpmd> flow create 0 egress pattern eth / ipv4 / end actions
1974 raw_encap index 2 / end
1976 Otherwise the default index ``0`` is used.
1978 Config Raw Decapsulation
1979 ~~~~~~~~~~~~~~~~~~~~~~~~
1981 Configure the raw data to be used when decapsulating a packet by
1982 rte_flow_action_raw_decap::
1984 set raw_decap {index} {item} [/ {item} [...]] / end_set
1986 There are multiple global buffers for ``raw_decap``, this command will set
1987 one internal buffer index by ``{index}``.
1988 If there is no ``{index}`` specified::
1990 set raw_decap {item} [/ {item} [...]] / end_set
1992 the default index ``0`` is used.
1993 In order to use different decapsulating header, ``index`` must be specified
1994 during the flow rule creation::
1996 testpmd> flow create 0 egress pattern eth / ipv4 / end actions
1997 raw_encap index 3 / end
1999 Otherwise the default index ``0`` is used.
2004 Set fec mode for a specific port::
2006 testpmd> set port (port_id) fec_mode auto|off|rs|baser
2012 The following sections show functions for configuring ports.
2016 Port configuration changes only become active when forwarding is started/restarted.
2021 Attach a port specified by pci address or virtual device args::
2023 testpmd> port attach (identifier)
2025 To attach a new pci device, the device should be recognized by kernel first.
2026 Then it should be moved under DPDK management.
2027 Finally the port can be attached to testpmd.
2029 For example, to move a pci device using ixgbe under DPDK management:
2031 .. code-block:: console
2033 # Check the status of the available devices.
2034 ./usertools/dpdk-devbind.py --status
2036 Network devices using DPDK-compatible driver
2037 ============================================
2040 Network devices using kernel driver
2041 ===================================
2042 0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused=
2045 # Bind the device to igb_uio.
2046 sudo ./usertools/dpdk-devbind.py -b igb_uio 0000:0a:00.0
2049 # Recheck the status of the devices.
2050 ./usertools/dpdk-devbind.py --status
2051 Network devices using DPDK-compatible driver
2052 ============================================
2053 0000:0a:00.0 '82599ES 10-Gigabit' drv=igb_uio unused=
2055 To attach a port created by virtual device, above steps are not needed.
2057 For example, to attach a port whose pci address is 0000:0a:00.0.
2059 .. code-block:: console
2061 testpmd> port attach 0000:0a:00.0
2062 Attaching a new port...
2063 EAL: PCI device 0000:0a:00.0 on NUMA socket -1
2064 EAL: probe driver: 8086:10fb rte_ixgbe_pmd
2065 EAL: PCI memory mapped at 0x7f83bfa00000
2066 EAL: PCI memory mapped at 0x7f83bfa80000
2067 PMD: eth_ixgbe_dev_init(): MAC: 2, PHY: 18, SFP+: 5
2068 PMD: eth_ixgbe_dev_init(): port 0 vendorID=0x8086 deviceID=0x10fb
2069 Port 0 is attached. Now total ports is 1
2072 For example, to attach a port created by pcap PMD.
2074 .. code-block:: console
2076 testpmd> port attach net_pcap0
2077 Attaching a new port...
2078 PMD: Initializing pmd_pcap for net_pcap0
2079 PMD: Creating pcap-backed ethdev on numa socket 0
2080 Port 0 is attached. Now total ports is 1
2083 In this case, identifier is ``net_pcap0``.
2084 This identifier format is the same as ``--vdev`` format of DPDK applications.
2086 For example, to re-attach a bonded port which has been previously detached,
2087 the mode and slave parameters must be given.
2089 .. code-block:: console
2091 testpmd> port attach net_bond_0,mode=0,slave=1
2092 Attaching a new port...
2093 EAL: Initializing pmd_bond for net_bond_0
2094 EAL: Create bonded device net_bond_0 on port 0 in mode 0 on socket 0.
2095 Port 0 is attached. Now total ports is 1
2102 Detach a specific port::
2104 testpmd> port detach (port_id)
2106 Before detaching a port, the port should be stopped and closed.
2108 For example, to detach a pci device port 0.
2110 .. code-block:: console
2112 testpmd> port stop 0
2115 testpmd> port close 0
2119 testpmd> port detach 0
2121 EAL: PCI device 0000:0a:00.0 on NUMA socket -1
2122 EAL: remove driver: 8086:10fb rte_ixgbe_pmd
2123 EAL: PCI memory unmapped at 0x7f83bfa00000
2124 EAL: PCI memory unmapped at 0x7f83bfa80000
2128 For example, to detach a virtual device port 0.
2130 .. code-block:: console
2132 testpmd> port stop 0
2135 testpmd> port close 0
2139 testpmd> port detach 0
2141 PMD: Closing pcap ethdev on numa socket 0
2142 Port 'net_pcap0' is detached. Now total ports is 0
2145 To remove a pci device completely from the system, first detach the port from testpmd.
2146 Then the device should be moved under kernel management.
2147 Finally the device can be removed using kernel pci hotplug functionality.
2149 For example, to move a pci device under kernel management:
2151 .. code-block:: console
2153 sudo ./usertools/dpdk-devbind.py -b ixgbe 0000:0a:00.0
2155 ./usertools/dpdk-devbind.py --status
2157 Network devices using DPDK-compatible driver
2158 ============================================
2161 Network devices using kernel driver
2162 ===================================
2163 0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused=igb_uio
2165 To remove a port created by a virtual device, above steps are not needed.
2170 Start all ports or a specific port::
2172 testpmd> port start (port_id|all)
2177 Stop all ports or a specific port::
2179 testpmd> port stop (port_id|all)
2184 Close all ports or a specific port::
2186 testpmd> port close (port_id|all)
2191 Reset all ports or a specific port::
2193 testpmd> port reset (port_id|all)
2195 User should stop port(s) before resetting and (re-)start after reset.
2197 port config - queue ring size
2198 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2200 Configure a rx/tx queue ring size::
2202 testpmd> port (port_id) (rxq|txq) (queue_id) ring_size (value)
2204 Only take effect after command that (re-)start the port or command that setup specific queue.
2206 port start/stop queue
2207 ~~~~~~~~~~~~~~~~~~~~~
2209 Start/stop a rx/tx queue on a specific port::
2211 testpmd> port (port_id) (rxq|txq) (queue_id) (start|stop)
2213 port config - queue deferred start
2214 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2216 Switch on/off deferred start of a specific port queue::
2218 testpmd> port (port_id) (rxq|txq) (queue_id) deferred_start (on|off)
2221 ~~~~~~~~~~~~~~~~~~~~~
2223 Setup a rx/tx queue on a specific port::
2225 testpmd> port (port_id) (rxq|txq) (queue_id) setup
2227 Only take effect when port is started.
2232 Set the speed and duplex mode for all ports or a specific port::
2234 testpmd> port config (port_id|all) speed (10|100|1000|10000|25000|40000|50000|100000|200000|auto) \
2235 duplex (half|full|auto)
2237 port config - queues/descriptors
2238 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2240 Set number of queues/descriptors for rxq, txq, rxd and txd::
2242 testpmd> port config all (rxq|txq|rxd|txd) (value)
2244 This is equivalent to the ``--rxq``, ``--txq``, ``--rxd`` and ``--txd`` command-line options.
2246 port config - max-pkt-len
2247 ~~~~~~~~~~~~~~~~~~~~~~~~~
2249 Set the maximum packet length::
2251 testpmd> port config all max-pkt-len (value)
2253 This is equivalent to the ``--max-pkt-len`` command-line option.
2255 port config - max-lro-pkt-size
2256 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2258 Set the maximum LRO aggregated packet size::
2260 testpmd> port config all max-lro-pkt-size (value)
2262 This is equivalent to the ``--max-lro-pkt-size`` command-line option.
2264 port config - Drop Packets
2265 ~~~~~~~~~~~~~~~~~~~~~~~~~~
2267 Enable or disable packet drop on all RX queues of all ports when no receive buffers available::
2269 testpmd> port config all drop-en (on|off)
2271 Packet dropping when no receive buffers available is off by default.
2273 The ``on`` option is equivalent to the ``--enable-drop-en`` command-line option.
2278 Set the RSS (Receive Side Scaling) mode on or off::
2280 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)
2282 RSS is on by default.
2284 The ``all`` option is equivalent to eth|vlan|ip|tcp|udp|sctp|ether|l2tpv3|esp|ah|pfcp.
2286 The ``default`` option enables all supported RSS types reported by device info.
2288 The ``none`` option is equivalent to the ``--disable-rss`` command-line option.
2290 port config - RSS Reta
2291 ~~~~~~~~~~~~~~~~~~~~~~
2293 Set the RSS (Receive Side Scaling) redirection table::
2295 testpmd> port config all rss reta (hash,queue)[,(hash,queue)]
2300 Set the DCB mode for an individual port::
2302 testpmd> port config (port_id) dcb vt (on|off) (traffic_class) pfc (on|off)
2304 The traffic class should be 4 or 8.
2309 Set the number of packets per burst::
2311 testpmd> port config all burst (value)
2313 This is equivalent to the ``--burst`` command-line option.
2315 port config - Threshold
2316 ~~~~~~~~~~~~~~~~~~~~~~~
2318 Set thresholds for TX/RX queues::
2320 testpmd> port config all (threshold) (value)
2322 Where the threshold type can be:
2324 * ``txpt:`` Set the prefetch threshold register of the TX rings, 0 <= value <= 255.
2326 * ``txht:`` Set the host threshold register of the TX rings, 0 <= value <= 255.
2328 * ``txwt:`` Set the write-back threshold register of the TX rings, 0 <= value <= 255.
2330 * ``rxpt:`` Set the prefetch threshold register of the RX rings, 0 <= value <= 255.
2332 * ``rxht:`` Set the host threshold register of the RX rings, 0 <= value <= 255.
2334 * ``rxwt:`` Set the write-back threshold register of the RX rings, 0 <= value <= 255.
2336 * ``txfreet:`` Set the transmit free threshold of the TX rings, 0 <= value <= txd.
2338 * ``rxfreet:`` Set the transmit free threshold of the RX rings, 0 <= value <= rxd.
2340 * ``txrst:`` Set the transmit RS bit threshold of TX rings, 0 <= value <= txd.
2342 These threshold options are also available from the command-line.
2344 port config pctype mapping
2345 ~~~~~~~~~~~~~~~~~~~~~~~~~~
2347 Reset pctype mapping table::
2349 testpmd> port config (port_id) pctype mapping reset
2351 Update hardware defined pctype to software defined flow type mapping table::
2353 testpmd> port config (port_id) pctype mapping update (pctype_id_0[,pctype_id_1]*) (flow_type_id)
2357 * ``pctype_id_x``: hardware pctype id as index of bit in bitmask value of the pctype mapping table.
2359 * ``flow_type_id``: software flow type id as the index of the pctype mapping table.
2361 port config input set
2362 ~~~~~~~~~~~~~~~~~~~~~
2364 Config RSS/FDIR/FDIR flexible payload input set for some pctype::
2366 testpmd> port config (port_id) pctype (pctype_id) \
2367 (hash_inset|fdir_inset|fdir_flx_inset) \
2368 (get|set|clear) field (field_idx)
2370 Clear RSS/FDIR/FDIR flexible payload input set for some pctype::
2372 testpmd> port config (port_id) pctype (pctype_id) \
2373 (hash_inset|fdir_inset|fdir_flx_inset) clear all
2377 * ``pctype_id``: hardware packet classification types.
2378 * ``field_idx``: hardware field index.
2380 port config udp_tunnel_port
2381 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
2383 Add/remove UDP tunnel port for VXLAN/GENEVE tunneling protocols::
2385 testpmd> port config (port_id) udp_tunnel_port add|rm vxlan|geneve|vxlan-gpe|ecpri (udp_port)
2387 port config tx_metadata
2388 ~~~~~~~~~~~~~~~~~~~~~~~
2390 Set Tx metadata value per port.
2391 testpmd will add this value to any Tx packet sent from this port::
2393 testpmd> port config (port_id) tx_metadata (value)
2398 Set/clear dynamic flag per port.
2399 testpmd will register this flag in the mbuf (same registration
2400 for both Tx and Rx). Then set/clear this flag for each Tx
2401 packet sent from this port. The set bit only works for Tx packet::
2403 testpmd> port config (port_id) dynf (name) (set|clear)
2408 To configure MTU(Maximum Transmission Unit) on devices using testpmd::
2410 testpmd> port config mtu (port_id) (value)
2412 port config rss hash key
2413 ~~~~~~~~~~~~~~~~~~~~~~~~
2415 To configure the RSS hash key used to compute the RSS
2416 hash of input [IP] packets received on port::
2418 testpmd> port config <port_id> rss-hash-key (ipv4|ipv4-frag|\
2419 ipv4-tcp|ipv4-udp|ipv4-sctp|ipv4-other|\
2420 ipv6|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|\
2421 ipv6-other|l2-payload|ipv6-ex|ipv6-tcp-ex|\
2422 ipv6-udp-ex <string of hex digits \
2423 (variable length, NIC dependent)>)
2428 The following sections show functions for device operations.
2433 Detach a device specified by pci address or virtual device args::
2435 testpmd> device detach (identifier)
2437 Before detaching a device associated with ports, the ports should be stopped and closed.
2439 For example, to detach a pci device whose address is 0002:03:00.0.
2441 .. code-block:: console
2443 testpmd> device detach 0002:03:00.0
2444 Removing a device...
2445 Port 1 is now closed
2446 EAL: Releasing pci mapped resource for 0002:03:00.0
2447 EAL: Calling pci_unmap_resource for 0002:03:00.0 at 0x218a050000
2448 EAL: Calling pci_unmap_resource for 0002:03:00.0 at 0x218c050000
2449 Device 0002:03:00.0 is detached
2450 Now total ports is 1
2452 For example, to detach a port created by pcap PMD.
2454 .. code-block:: console
2456 testpmd> device detach net_pcap0
2457 Removing a device...
2458 Port 0 is now closed
2459 Device net_pcap0 is detached
2460 Now total ports is 0
2463 In this case, identifier is ``net_pcap0``.
2464 This identifier format is the same as ``--vdev`` format of DPDK applications.
2466 Link Bonding Functions
2467 ----------------------
2469 The Link Bonding functions make it possible to dynamically create and
2470 manage link bonding devices from within testpmd interactive prompt.
2472 create bonded device
2473 ~~~~~~~~~~~~~~~~~~~~
2475 Create a new bonding device::
2477 testpmd> create bonded device (mode) (socket)
2479 For example, to create a bonded device in mode 1 on socket 0::
2481 testpmd> create bonded device 1 0
2482 created new bonded device (port X)
2487 Adds Ethernet device to a Link Bonding device::
2489 testpmd> add bonding slave (slave id) (port id)
2491 For example, to add Ethernet device (port 6) to a Link Bonding device (port 10)::
2493 testpmd> add bonding slave 6 10
2496 remove bonding slave
2497 ~~~~~~~~~~~~~~~~~~~~
2499 Removes an Ethernet slave device from a Link Bonding device::
2501 testpmd> remove bonding slave (slave id) (port id)
2503 For example, to remove Ethernet slave device (port 6) to a Link Bonding device (port 10)::
2505 testpmd> remove bonding slave 6 10
2510 Set the Link Bonding mode of a Link Bonding device::
2512 testpmd> set bonding mode (value) (port id)
2514 For example, to set the bonding mode of a Link Bonding device (port 10) to broadcast (mode 3)::
2516 testpmd> set bonding mode 3 10
2521 Set an Ethernet slave device as the primary device on a Link Bonding device::
2523 testpmd> set bonding primary (slave id) (port id)
2525 For example, to set the Ethernet slave device (port 6) as the primary port of a Link Bonding device (port 10)::
2527 testpmd> set bonding primary 6 10
2532 Set the MAC address of a Link Bonding device::
2534 testpmd> set bonding mac (port id) (mac)
2536 For example, to set the MAC address of a Link Bonding device (port 10) to 00:00:00:00:00:01::
2538 testpmd> set bonding mac 10 00:00:00:00:00:01
2540 set bonding balance_xmit_policy
2541 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2543 Set the transmission policy for a Link Bonding device when it is in Balance XOR mode::
2545 testpmd> set bonding balance_xmit_policy (port_id) (l2|l23|l34)
2547 For example, set a Link Bonding device (port 10) to use a balance policy of layer 3+4 (IP addresses & UDP ports)::
2549 testpmd> set bonding balance_xmit_policy 10 l34
2552 set bonding mon_period
2553 ~~~~~~~~~~~~~~~~~~~~~~
2555 Set the link status monitoring polling period in milliseconds for a bonding device.
2557 This adds support for PMD slave devices which do not support link status interrupts.
2558 When the mon_period is set to a value greater than 0 then all PMD's which do not support
2559 link status ISR will be queried every polling interval to check if their link status has changed::
2561 testpmd> set bonding mon_period (port_id) (value)
2563 For example, to set the link status monitoring polling period of bonded device (port 5) to 150ms::
2565 testpmd> set bonding mon_period 5 150
2568 set bonding lacp dedicated_queue
2569 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2571 Enable dedicated tx/rx queues on bonding devices slaves to handle LACP control plane traffic
2572 when in mode 4 (link-aggregation-802.3ad)::
2574 testpmd> set bonding lacp dedicated_queues (port_id) (enable|disable)
2577 set bonding agg_mode
2578 ~~~~~~~~~~~~~~~~~~~~
2580 Enable one of the specific aggregators mode when in mode 4 (link-aggregation-802.3ad)::
2582 testpmd> set bonding agg_mode (port_id) (bandwidth|count|stable)
2588 Show the current configuration of a Link Bonding device::
2590 testpmd> show bonding config (port id)
2593 to show the configuration a Link Bonding device (port 9) with 3 slave devices (1, 3, 4)
2594 in balance mode with a transmission policy of layer 2+3::
2596 testpmd> show bonding config 9
2598 Balance Xmit Policy: BALANCE_XMIT_POLICY_LAYER23
2600 Active Slaves (3): [1 3 4]
2607 The Register Functions can be used to read from and write to registers on the network card referenced by a port number.
2608 This is mainly useful for debugging purposes.
2609 Reference should be made to the appropriate datasheet for the network card for details on the register addresses
2610 and fields that can be accessed.
2615 Display the value of a port register::
2617 testpmd> read reg (port_id) (address)
2619 For example, to examine the Flow Director control register (FDIRCTL, 0x0000EE000) on an Intel 82599 10 GbE Controller::
2621 testpmd> read reg 0 0xEE00
2622 port 0 PCI register at offset 0xEE00: 0x4A060029 (1241907241)
2627 Display a port register bit field::
2629 testpmd> read regfield (port_id) (address) (bit_x) (bit_y)
2631 For example, reading the lowest two bits from the register in the example above::
2633 testpmd> read regfield 0 0xEE00 0 1
2634 port 0 PCI register at offset 0xEE00: bits[0, 1]=0x1 (1)
2639 Display a single port register bit::
2641 testpmd> read regbit (port_id) (address) (bit_x)
2643 For example, reading the lowest bit from the register in the example above::
2645 testpmd> read regbit 0 0xEE00 0
2646 port 0 PCI register at offset 0xEE00: bit 0=1
2651 Set the value of a port register::
2653 testpmd> write reg (port_id) (address) (value)
2655 For example, to clear a register::
2657 testpmd> write reg 0 0xEE00 0x0
2658 port 0 PCI register at offset 0xEE00: 0x00000000 (0)
2663 Set bit field of a port register::
2665 testpmd> write regfield (port_id) (address) (bit_x) (bit_y) (value)
2667 For example, writing to the register cleared in the example above::
2669 testpmd> write regfield 0 0xEE00 0 1 2
2670 port 0 PCI register at offset 0xEE00: 0x00000002 (2)
2675 Set single bit value of a port register::
2677 testpmd> write regbit (port_id) (address) (bit_x) (value)
2679 For example, to set the high bit in the register from the example above::
2681 testpmd> write regbit 0 0xEE00 31 1
2682 port 0 PCI register at offset 0xEE00: 0x8000000A (2147483658)
2684 Traffic Metering and Policing
2685 -----------------------------
2687 The following section shows functions for configuring traffic metering and
2688 policing on the ethernet device through the use of generic ethdev API.
2690 show port traffic management capability
2691 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2693 Show traffic metering and policing capability of the port::
2695 testpmd> show port meter cap (port_id)
2697 add port meter profile (srTCM rfc2967)
2698 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2700 Add meter profile (srTCM rfc2697) to the ethernet device::
2702 testpmd> add port meter profile srtcm_rfc2697 (port_id) (profile_id) \
2707 * ``profile_id``: ID for the meter profile.
2708 * ``cir``: Committed Information Rate (CIR) (bytes/second).
2709 * ``cbs``: Committed Burst Size (CBS) (bytes).
2710 * ``ebs``: Excess Burst Size (EBS) (bytes).
2712 add port meter profile (trTCM rfc2968)
2713 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2715 Add meter profile (srTCM rfc2698) to the ethernet device::
2717 testpmd> add port meter profile trtcm_rfc2698 (port_id) (profile_id) \
2718 (cir) (pir) (cbs) (pbs)
2722 * ``profile_id``: ID for the meter profile.
2723 * ``cir``: Committed information rate (bytes/second).
2724 * ``pir``: Peak information rate (bytes/second).
2725 * ``cbs``: Committed burst size (bytes).
2726 * ``pbs``: Peak burst size (bytes).
2728 add port meter profile (trTCM rfc4115)
2729 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2731 Add meter profile (trTCM rfc4115) to the ethernet device::
2733 testpmd> add port meter profile trtcm_rfc4115 (port_id) (profile_id) \
2734 (cir) (eir) (cbs) (ebs)
2738 * ``profile_id``: ID for the meter profile.
2739 * ``cir``: Committed information rate (bytes/second).
2740 * ``eir``: Excess information rate (bytes/second).
2741 * ``cbs``: Committed burst size (bytes).
2742 * ``ebs``: Excess burst size (bytes).
2744 delete port meter profile
2745 ~~~~~~~~~~~~~~~~~~~~~~~~~
2747 Delete meter profile from the ethernet device::
2749 testpmd> del port meter profile (port_id) (profile_id)
2754 Create new meter object for the ethernet device::
2756 testpmd> create port meter (port_id) (mtr_id) (profile_id) \
2757 (meter_enable) (g_action) (y_action) (r_action) (stats_mask) (shared) \
2758 (use_pre_meter_color) [(dscp_tbl_entry0) (dscp_tbl_entry1)...\
2763 * ``mtr_id``: meter object ID.
2764 * ``profile_id``: ID for the meter profile.
2765 * ``meter_enable``: When this parameter has a non-zero value, the meter object
2766 gets enabled at the time of creation, otherwise remains disabled.
2767 * ``g_action``: Policer action for the packet with green color.
2768 * ``y_action``: Policer action for the packet with yellow color.
2769 * ``r_action``: Policer action for the packet with red color.
2770 * ``stats_mask``: Mask of statistics counter types to be enabled for the
2772 * ``shared``: When this parameter has a non-zero value, the meter object is
2773 shared by multiple flows. Otherwise, meter object is used by single flow.
2774 * ``use_pre_meter_color``: When this parameter has a non-zero value, the
2775 input color for the current meter object is determined by the latest meter
2776 object in the same flow. Otherwise, the current meter object uses the
2777 *dscp_table* to determine the input color.
2778 * ``dscp_tbl_entryx``: DSCP table entry x providing meter providing input
2779 color, 0 <= x <= 63.
2784 Enable meter for the ethernet device::
2786 testpmd> enable port meter (port_id) (mtr_id)
2791 Disable meter for the ethernet device::
2793 testpmd> disable port meter (port_id) (mtr_id)
2798 Delete meter for the ethernet device::
2800 testpmd> del port meter (port_id) (mtr_id)
2802 Set port meter profile
2803 ~~~~~~~~~~~~~~~~~~~~~~
2805 Set meter profile for the ethernet device::
2807 testpmd> set port meter profile (port_id) (mtr_id) (profile_id)
2809 set port meter dscp table
2810 ~~~~~~~~~~~~~~~~~~~~~~~~~
2812 Set meter dscp table for the ethernet device::
2814 testpmd> set port meter dscp table (port_id) (mtr_id) [(dscp_tbl_entry0) \
2815 (dscp_tbl_entry1)...(dscp_tbl_entry63)]
2817 set port meter policer action
2818 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2820 Set meter policer action for the ethernet device::
2822 testpmd> set port meter policer action (port_id) (mtr_id) (action_mask) \
2823 (action0) [(action1) (action1)]
2827 * ``action_mask``: Bit mask indicating which policer actions need to be
2828 updated. One or more policer actions can be updated in a single function
2829 invocation. To update the policer action associated with color C, bit
2830 (1 << C) needs to be set in *action_mask* and element at position C
2831 in the *actions* array needs to be valid.
2832 * ``actionx``: Policer action for the color x,
2833 RTE_MTR_GREEN <= x < RTE_MTR_COLORS
2835 set port meter stats mask
2836 ~~~~~~~~~~~~~~~~~~~~~~~~~
2838 Set meter stats mask for the ethernet device::
2840 testpmd> set port meter stats mask (port_id) (mtr_id) (stats_mask)
2844 * ``stats_mask``: Bit mask indicating statistics counter types to be enabled.
2846 show port meter stats
2847 ~~~~~~~~~~~~~~~~~~~~~
2849 Show meter stats of the ethernet device::
2851 testpmd> show port meter stats (port_id) (mtr_id) (clear)
2855 * ``clear``: Flag that indicates whether the statistics counters should
2856 be cleared (i.e. set to zero) immediately after they have been read or not.
2861 The following section shows functions for configuring traffic management on
2862 the ethernet device through the use of generic TM API.
2864 show port traffic management capability
2865 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2867 Show traffic management capability of the port::
2869 testpmd> show port tm cap (port_id)
2871 show port traffic management capability (hierarchy level)
2872 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2874 Show traffic management hierarchy level capability of the port::
2876 testpmd> show port tm level cap (port_id) (level_id)
2878 show port traffic management capability (hierarchy node level)
2879 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2881 Show the traffic management hierarchy node capability of the port::
2883 testpmd> show port tm node cap (port_id) (node_id)
2885 show port traffic management hierarchy node type
2886 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2888 Show the port traffic management hierarchy node type::
2890 testpmd> show port tm node type (port_id) (node_id)
2892 show port traffic management hierarchy node stats
2893 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2895 Show the port traffic management hierarchy node statistics::
2897 testpmd> show port tm node stats (port_id) (node_id) (clear)
2901 * ``clear``: When this parameter has a non-zero value, the statistics counters
2902 are cleared (i.e. set to zero) immediately after they have been read,
2903 otherwise the statistics counters are left untouched.
2905 Add port traffic management private shaper profile
2906 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2908 Add the port traffic management private shaper profile::
2910 testpmd> add port tm node shaper profile (port_id) (shaper_profile_id) \
2911 (cmit_tb_rate) (cmit_tb_size) (peak_tb_rate) (peak_tb_size) \
2912 (packet_length_adjust) (packet_mode)
2916 * ``shaper_profile id``: Shaper profile ID for the new profile.
2917 * ``cmit_tb_rate``: Committed token bucket rate (bytes per second or packets per second).
2918 * ``cmit_tb_size``: Committed token bucket size (bytes or packets).
2919 * ``peak_tb_rate``: Peak token bucket rate (bytes per second or packets per second).
2920 * ``peak_tb_size``: Peak token bucket size (bytes or packets).
2921 * ``packet_length_adjust``: The value (bytes) to be added to the length of
2922 each packet for the purpose of shaping. This parameter value can be used to
2923 correct the packet length with the framing overhead bytes that are consumed
2925 * ``packet_mode``: Shaper configured in packet mode. This parameter value if
2926 zero, configures shaper in byte mode and if non-zero configures it in packet
2929 Delete port traffic management private shaper profile
2930 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2932 Delete the port traffic management private shaper::
2934 testpmd> del port tm node shaper profile (port_id) (shaper_profile_id)
2938 * ``shaper_profile id``: Shaper profile ID that needs to be deleted.
2940 Add port traffic management shared shaper
2941 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2943 Create the port traffic management shared shaper::
2945 testpmd> add port tm node shared shaper (port_id) (shared_shaper_id) \
2950 * ``shared_shaper_id``: Shared shaper ID to be created.
2951 * ``shaper_profile id``: Shaper profile ID for shared shaper.
2953 Set port traffic management shared shaper
2954 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2956 Update the port traffic management shared shaper::
2958 testpmd> set port tm node shared shaper (port_id) (shared_shaper_id) \
2963 * ``shared_shaper_id``: Shared shaper ID to be update.
2964 * ``shaper_profile id``: Shaper profile ID for shared shaper.
2966 Delete port traffic management shared shaper
2967 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2969 Delete the port traffic management shared shaper::
2971 testpmd> del port tm node shared shaper (port_id) (shared_shaper_id)
2975 * ``shared_shaper_id``: Shared shaper ID to be deleted.
2977 Set port traffic management hierarchy node private shaper
2978 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2980 set the port traffic management hierarchy node private shaper::
2982 testpmd> set port tm node shaper profile (port_id) (node_id) \
2987 * ``shaper_profile id``: Private shaper profile ID to be enabled on the
2990 Add port traffic management WRED profile
2991 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2993 Create a new WRED profile::
2995 testpmd> add port tm node wred profile (port_id) (wred_profile_id) \
2996 (color_g) (min_th_g) (max_th_g) (maxp_inv_g) (wq_log2_g) \
2997 (color_y) (min_th_y) (max_th_y) (maxp_inv_y) (wq_log2_y) \
2998 (color_r) (min_th_r) (max_th_r) (maxp_inv_r) (wq_log2_r)
3002 * ``wred_profile id``: Identifier for the newly create WRED profile
3003 * ``color_g``: Packet color (green)
3004 * ``min_th_g``: Minimum queue threshold for packet with green color
3005 * ``max_th_g``: Minimum queue threshold for packet with green color
3006 * ``maxp_inv_g``: Inverse of packet marking probability maximum value (maxp)
3007 * ``wq_log2_g``: Negated log2 of queue weight (wq)
3008 * ``color_y``: Packet color (yellow)
3009 * ``min_th_y``: Minimum queue threshold for packet with yellow color
3010 * ``max_th_y``: Minimum queue threshold for packet with yellow color
3011 * ``maxp_inv_y``: Inverse of packet marking probability maximum value (maxp)
3012 * ``wq_log2_y``: Negated log2 of queue weight (wq)
3013 * ``color_r``: Packet color (red)
3014 * ``min_th_r``: Minimum queue threshold for packet with yellow color
3015 * ``max_th_r``: Minimum queue threshold for packet with yellow color
3016 * ``maxp_inv_r``: Inverse of packet marking probability maximum value (maxp)
3017 * ``wq_log2_r``: Negated log2 of queue weight (wq)
3019 Delete port traffic management WRED profile
3020 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3022 Delete the WRED profile::
3024 testpmd> del port tm node wred profile (port_id) (wred_profile_id)
3026 Add port traffic management hierarchy nonleaf node
3027 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3029 Add nonleaf node to port traffic management hierarchy::
3031 testpmd> add port tm nonleaf node (port_id) (node_id) (parent_node_id) \
3032 (priority) (weight) (level_id) (shaper_profile_id) \
3033 (n_sp_priorities) (stats_mask) (n_shared_shapers) \
3034 [(shared_shaper_0) (shared_shaper_1) ...] \
3038 * ``parent_node_id``: Node ID of the parent.
3039 * ``priority``: Node priority (highest node priority is zero). This is used by
3040 the SP algorithm running on the parent node for scheduling this node.
3041 * ``weight``: Node weight (lowest weight is one). The node weight is relative
3042 to the weight sum of all siblings that have the same priority. It is used by
3043 the WFQ algorithm running on the parent node for scheduling this node.
3044 * ``level_id``: Hierarchy level of the node.
3045 * ``shaper_profile_id``: Shaper profile ID of the private shaper to be used by
3047 * ``n_sp_priorities``: Number of strict priorities.
3048 * ``stats_mask``: Mask of statistics counter types to be enabled for this node.
3049 * ``n_shared_shapers``: Number of shared shapers.
3050 * ``shared_shaper_id``: Shared shaper id.
3052 Add port traffic management hierarchy nonleaf node with packet mode
3053 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3055 Add nonleaf node with packet mode to port traffic management hierarchy::
3057 testpmd> add port tm nonleaf node pktmode (port_id) (node_id) (parent_node_id) \
3058 (priority) (weight) (level_id) (shaper_profile_id) \
3059 (n_sp_priorities) (stats_mask) (n_shared_shapers) \
3060 [(shared_shaper_0) (shared_shaper_1) ...] \
3064 * ``parent_node_id``: Node ID of the parent.
3065 * ``priority``: Node priority (highest node priority is zero). This is used by
3066 the SP algorithm running on the parent node for scheduling this node.
3067 * ``weight``: Node weight (lowest weight is one). The node weight is relative
3068 to the weight sum of all siblings that have the same priority. It is used by
3069 the WFQ algorithm running on the parent node for scheduling this node.
3070 * ``level_id``: Hierarchy level of the node.
3071 * ``shaper_profile_id``: Shaper profile ID of the private shaper to be used by
3073 * ``n_sp_priorities``: Number of strict priorities. Packet mode is enabled on
3075 * ``stats_mask``: Mask of statistics counter types to be enabled for this node.
3076 * ``n_shared_shapers``: Number of shared shapers.
3077 * ``shared_shaper_id``: Shared shaper id.
3079 Add port traffic management hierarchy leaf node
3080 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3082 Add leaf node to port traffic management hierarchy::
3084 testpmd> add port tm leaf node (port_id) (node_id) (parent_node_id) \
3085 (priority) (weight) (level_id) (shaper_profile_id) \
3086 (cman_mode) (wred_profile_id) (stats_mask) (n_shared_shapers) \
3087 [(shared_shaper_id) (shared_shaper_id) ...] \
3091 * ``parent_node_id``: Node ID of the parent.
3092 * ``priority``: Node priority (highest node priority is zero). This is used by
3093 the SP algorithm running on the parent node for scheduling this node.
3094 * ``weight``: Node weight (lowest weight is one). The node weight is relative
3095 to the weight sum of all siblings that have the same priority. It is used by
3096 the WFQ algorithm running on the parent node for scheduling this node.
3097 * ``level_id``: Hierarchy level of the node.
3098 * ``shaper_profile_id``: Shaper profile ID of the private shaper to be used by
3100 * ``cman_mode``: Congestion management mode to be enabled for this node.
3101 * ``wred_profile_id``: WRED profile id to be enabled for this node.
3102 * ``stats_mask``: Mask of statistics counter types to be enabled for this node.
3103 * ``n_shared_shapers``: Number of shared shapers.
3104 * ``shared_shaper_id``: Shared shaper id.
3106 Delete port traffic management hierarchy node
3107 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3109 Delete node from port traffic management hierarchy::
3111 testpmd> del port tm node (port_id) (node_id)
3113 Update port traffic management hierarchy parent node
3114 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3116 Update port traffic management hierarchy parent node::
3118 testpmd> set port tm node parent (port_id) (node_id) (parent_node_id) \
3121 This function can only be called after the hierarchy commit invocation. Its
3122 success depends on the port support for this operation, as advertised through
3123 the port capability set. This function is valid for all nodes of the traffic
3124 management hierarchy except root node.
3126 Suspend port traffic management hierarchy node
3127 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3129 testpmd> suspend port tm node (port_id) (node_id)
3131 Resume port traffic management hierarchy node
3132 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3134 testpmd> resume port tm node (port_id) (node_id)
3136 Commit port traffic management hierarchy
3137 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3139 Commit the traffic management hierarchy on the port::
3141 testpmd> port tm hierarchy commit (port_id) (clean_on_fail)
3145 * ``clean_on_fail``: When set to non-zero, hierarchy is cleared on function
3146 call failure. On the other hand, hierarchy is preserved when this parameter
3149 Set port traffic management mark VLAN dei
3150 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3152 Enables/Disables the traffic management marking on the port for VLAN packets::
3154 testpmd> set port tm mark vlan_dei <port_id> <green> <yellow> <red>
3158 * ``port_id``: The port which on which VLAN packets marked as ``green`` or
3159 ``yellow`` or ``red`` will have dei bit enabled
3161 * ``green`` enable 1, disable 0 marking for dei bit of VLAN packets marked as green
3163 * ``yellow`` enable 1, disable 0 marking for dei bit of VLAN packets marked as yellow
3165 * ``red`` enable 1, disable 0 marking for dei bit of VLAN packets marked as red
3167 Set port traffic management mark IP dscp
3168 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3170 Enables/Disables the traffic management marking on the port for IP dscp packets::
3172 testpmd> set port tm mark ip_dscp <port_id> <green> <yellow> <red>
3176 * ``port_id``: The port which on which IP packets marked as ``green`` or
3177 ``yellow`` or ``red`` will have IP dscp bits updated
3179 * ``green`` enable 1, disable 0 marking IP dscp to low drop precedence for green packets
3181 * ``yellow`` enable 1, disable 0 marking IP dscp to medium drop precedence for yellow packets
3183 * ``red`` enable 1, disable 0 marking IP dscp to high drop precedence for red packets
3185 Set port traffic management mark IP ecn
3186 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3188 Enables/Disables the traffic management marking on the port for IP ecn packets::
3190 testpmd> set port tm mark ip_ecn <port_id> <green> <yellow> <red>
3194 * ``port_id``: The port which on which IP packets marked as ``green`` or
3195 ``yellow`` or ``red`` will have IP ecn bits updated
3197 * ``green`` enable 1, disable 0 marking IP ecn for green marked packets with ecn of 2'b01 or 2'b10
3198 to ecn of 2'b11 when IP is caring TCP or SCTP
3200 * ``yellow`` enable 1, disable 0 marking IP ecn for yellow marked packets with ecn of 2'b01 or 2'b10
3201 to ecn of 2'b11 when IP is caring TCP or SCTP
3203 * ``red`` enable 1, disable 0 marking IP ecn for yellow marked packets with ecn of 2'b01 or 2'b10
3204 to ecn of 2'b11 when IP is caring TCP or SCTP
3209 This section details the available filter functions that are available.
3211 Note these functions interface the deprecated legacy filtering framework,
3212 superseded by *rte_flow*. See `Flow rules management`_.
3214 .. _testpmd_flow_director:
3219 Set flow director's input masks::
3221 flow_director_mask (port_id) mode IP vlan (vlan_value) \
3222 src_mask (ipv4_src) (ipv6_src) (src_port) \
3223 dst_mask (ipv4_dst) (ipv6_dst) (dst_port)
3225 flow_director_mask (port_id) mode MAC-VLAN vlan (vlan_value)
3227 flow_director_mask (port_id) mode Tunnel vlan (vlan_value) \
3228 mac (mac_value) tunnel-type (tunnel_type_value) \
3229 tunnel-id (tunnel_id_value)
3231 Example, to set flow director mask on port 0::
3233 testpmd> flow_director_mask 0 mode IP vlan 0xefff \
3234 src_mask 255.255.255.255 \
3235 FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF \
3236 dst_mask 255.255.255.255 \
3237 FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF
3239 flow_director_flex_payload
3240 ~~~~~~~~~~~~~~~~~~~~~~~~~~
3242 Configure flexible payload selection::
3244 flow_director_flex_payload (port_id) (raw|l2|l3|l4) (config)
3246 For example, to select the first 16 bytes from the offset 4 (bytes) of packet's payload as flexible payload::
3248 testpmd> flow_director_flex_payload 0 l4 \
3249 (4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19)
3252 .. _testpmd_rte_flow:
3254 Flow rules management
3255 ---------------------
3257 Control of the generic flow API (*rte_flow*) is fully exposed through the
3258 ``flow`` command (validation, creation, destruction, queries and operation
3261 Considering *rte_flow* overlaps with all `Filter Functions`_, using both
3262 features simultaneously may cause undefined side-effects and is therefore
3268 Because the ``flow`` command uses dynamic tokens to handle the large number
3269 of possible flow rules combinations, its behavior differs slightly from
3270 other commands, in particular:
3272 - Pressing *?* or the *<tab>* key displays contextual help for the current
3273 token, not that of the entire command.
3275 - Optional and repeated parameters are supported (provided they are listed
3276 in the contextual help).
3278 The first parameter stands for the operation mode. Possible operations and
3279 their general syntax are described below. They are covered in detail in the
3282 - Check whether a flow rule can be created::
3284 flow validate {port_id}
3285 [group {group_id}] [priority {level}] [ingress] [egress] [transfer]
3286 pattern {item} [/ {item} [...]] / end
3287 actions {action} [/ {action} [...]] / end
3289 - Create a flow rule::
3291 flow create {port_id}
3292 [group {group_id}] [priority {level}] [ingress] [egress] [transfer]
3293 pattern {item} [/ {item} [...]] / end
3294 actions {action} [/ {action} [...]] / end
3296 - Destroy specific flow rules::
3298 flow destroy {port_id} rule {rule_id} [...]
3300 - Destroy all flow rules::
3302 flow flush {port_id}
3304 - Query an existing flow rule::
3306 flow query {port_id} {rule_id} {action}
3308 - List existing flow rules sorted by priority, filtered by group
3311 flow list {port_id} [group {group_id}] [...]
3313 - Restrict ingress traffic to the defined flow rules::
3315 flow isolate {port_id} {boolean}
3317 - Dump internal representation information of all flows in hardware::
3319 flow dump {port_id} {output_file}
3321 - List and destroy aged flow rules::
3323 flow aged {port_id} [destroy]
3325 - Tunnel offload - create a tunnel stub::
3327 flow tunnel create {port_id} type {tunnel_type}
3329 - Tunnel offload - destroy a tunnel stub::
3331 flow tunnel destroy {port_id} id {tunnel_id}
3333 - Tunnel offload - list port tunnel stubs::
3335 flow tunnel list {port_id}
3337 Creating a tunnel stub for offload
3338 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3340 ``flow tunnel create`` setup a tunnel stub for tunnel offload flow rules::
3342 flow tunnel create {port_id} type {tunnel_type}
3344 If successful, it will return a tunnel stub ID usable with other commands::
3346 port [...]: flow tunnel #[...] type [...]
3348 Tunnel stub ID is relative to a port.
3350 Destroying tunnel offload stub
3351 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3353 ``flow tunnel destroy`` destroy port tunnel stub::
3355 flow tunnel destroy {port_id} id {tunnel_id}
3357 Listing tunnel offload stubs
3358 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3360 ``flow tunnel list`` list port tunnel offload stubs::
3362 flow tunnel list {port_id}
3364 Validating flow rules
3365 ~~~~~~~~~~~~~~~~~~~~~
3367 ``flow validate`` reports whether a flow rule would be accepted by the
3368 underlying device in its current state but stops short of creating it. It is
3369 bound to ``rte_flow_validate()``::
3371 flow validate {port_id}
3372 [group {group_id}] [priority {level}] [ingress] [egress] [transfer]
3373 pattern {item} [/ {item} [...]] / end
3374 actions {action} [/ {action} [...]] / end
3376 If successful, it will show::
3380 Otherwise it will show an error message of the form::
3382 Caught error type [...] ([...]): [...]
3384 This command uses the same parameters as ``flow create``, their format is
3385 described in `Creating flow rules`_.
3387 Check whether redirecting any Ethernet packet received on port 0 to RX queue
3388 index 6 is supported::
3390 testpmd> flow validate 0 ingress pattern eth / end
3391 actions queue index 6 / end
3395 Port 0 does not support TCPv6 rules::
3397 testpmd> flow validate 0 ingress pattern eth / ipv6 / tcp / end
3399 Caught error type 9 (specific pattern item): Invalid argument
3405 ``flow create`` validates and creates the specified flow rule. It is bound
3406 to ``rte_flow_create()``::
3408 flow create {port_id}
3409 [group {group_id}] [priority {level}] [ingress] [egress] [transfer]
3410 [tunnel_set {tunnel_id}] [tunnel_match {tunnel_id}]
3411 pattern {item} [/ {item} [...]] / end
3412 actions {action} [/ {action} [...]] / end
3414 If successful, it will return a flow rule ID usable with other commands::
3416 Flow rule #[...] created
3418 Otherwise it will show an error message of the form::
3420 Caught error type [...] ([...]): [...]
3422 Parameters describe in the following order:
3424 - Attributes (*group*, *priority*, *ingress*, *egress*, *transfer* tokens).
3425 - Tunnel offload specification (tunnel_set, tunnel_match)
3426 - A matching pattern, starting with the *pattern* token and terminated by an
3428 - Actions, starting with the *actions* token and terminated by an *end*
3431 These translate directly to *rte_flow* objects provided as-is to the
3432 underlying functions.
3434 The shortest valid definition only comprises mandatory tokens::
3436 testpmd> flow create 0 pattern end actions end
3438 Note that PMDs may refuse rules that essentially do nothing such as this
3441 **All unspecified object values are automatically initialized to 0.**
3446 These tokens affect flow rule attributes (``struct rte_flow_attr``) and are
3447 specified before the ``pattern`` token.
3449 - ``group {group id}``: priority group.
3450 - ``priority {level}``: priority level within group.
3451 - ``ingress``: rule applies to ingress traffic.
3452 - ``egress``: rule applies to egress traffic.
3453 - ``transfer``: apply rule directly to endpoints found in pattern.
3455 Each instance of an attribute specified several times overrides the previous
3456 value as shown below (group 4 is used)::
3458 testpmd> flow create 0 group 42 group 24 group 4 [...]
3460 Note that once enabled, ``ingress`` and ``egress`` cannot be disabled.
3462 While not specifying a direction is an error, some rules may allow both
3465 Most rules affect RX therefore contain the ``ingress`` token::
3467 testpmd> flow create 0 ingress pattern [...]
3472 Indicate tunnel offload rule type
3474 - ``tunnel_set {tunnel_id}``: mark rule as tunnel offload decap_set type.
3475 - ``tunnel_match {tunnel_id}``: mark rule as tunel offload match type.
3480 A matching pattern starts after the ``pattern`` token. It is made of pattern
3481 items and is terminated by a mandatory ``end`` item.
3483 Items are named after their type (*RTE_FLOW_ITEM_TYPE_* from ``enum
3484 rte_flow_item_type``).
3486 The ``/`` token is used as a separator between pattern items as shown
3489 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end [...]
3491 Note that protocol items like these must be stacked from lowest to highest
3492 layer to make sense. For instance, the following rule is either invalid or
3493 unlikely to match any packet::
3495 testpmd> flow create 0 ingress pattern eth / udp / ipv4 / end [...]
3497 More information on these restrictions can be found in the *rte_flow*
3500 Several items support additional specification structures, for example
3501 ``ipv4`` allows specifying source and destination addresses as follows::
3503 testpmd> flow create 0 ingress pattern eth / ipv4 src is 10.1.1.1
3504 dst is 10.2.0.0 / end [...]
3506 This rule matches all IPv4 traffic with the specified properties.
3508 In this example, ``src`` and ``dst`` are field names of the underlying
3509 ``struct rte_flow_item_ipv4`` object. All item properties can be specified
3510 in a similar fashion.
3512 The ``is`` token means that the subsequent value must be matched exactly,
3513 and assigns ``spec`` and ``mask`` fields in ``struct rte_flow_item``
3514 accordingly. Possible assignment tokens are:
3516 - ``is``: match value perfectly (with full bit-mask).
3517 - ``spec``: match value according to configured bit-mask.
3518 - ``last``: specify upper bound to establish a range.
3519 - ``mask``: specify bit-mask with relevant bits set to one.
3520 - ``prefix``: generate bit-mask with <prefix-length> most-significant bits set to one.
3522 These yield identical results::
3524 ipv4 src is 10.1.1.1
3528 ipv4 src spec 10.1.1.1 src mask 255.255.255.255
3532 ipv4 src spec 10.1.1.1 src prefix 32
3536 ipv4 src is 10.1.1.1 src last 10.1.1.1 # range with a single value
3540 ipv4 src is 10.1.1.1 src last 0 # 0 disables range
3542 Inclusive ranges can be defined with ``last``::
3544 ipv4 src is 10.1.1.1 src last 10.2.3.4 # 10.1.1.1 to 10.2.3.4
3546 Note that ``mask`` affects both ``spec`` and ``last``::
3548 ipv4 src is 10.1.1.1 src last 10.2.3.4 src mask 255.255.0.0
3549 # matches 10.1.0.0 to 10.2.255.255
3551 Properties can be modified multiple times::
3553 ipv4 src is 10.1.1.1 src is 10.1.2.3 src is 10.2.3.4 # matches 10.2.3.4
3557 ipv4 src is 10.1.1.1 src prefix 24 src prefix 16 # matches 10.1.0.0/16
3562 This section lists supported pattern items and their attributes, if any.
3564 - ``end``: end list of pattern items.
3566 - ``void``: no-op pattern item.
3568 - ``invert``: perform actions when pattern does not match.
3570 - ``any``: match any protocol for the current layer.
3572 - ``num {unsigned}``: number of layers covered.
3574 - ``pf``: match traffic from/to the physical function.
3576 - ``vf``: match traffic from/to a virtual function ID.
3578 - ``id {unsigned}``: VF ID.
3580 - ``phy_port``: match traffic from/to a specific physical port.
3582 - ``index {unsigned}``: physical port index.
3584 - ``port_id``: match traffic from/to a given DPDK port ID.
3586 - ``id {unsigned}``: DPDK port ID.
3588 - ``mark``: match value set in previously matched flow rule using the mark action.
3590 - ``id {unsigned}``: arbitrary integer value.
3592 - ``raw``: match an arbitrary byte string.
3594 - ``relative {boolean}``: look for pattern after the previous item.
3595 - ``search {boolean}``: search pattern from offset (see also limit).
3596 - ``offset {integer}``: absolute or relative offset for pattern.
3597 - ``limit {unsigned}``: search area limit for start of pattern.
3598 - ``pattern {string}``: byte string to look for.
3600 - ``eth``: match Ethernet header.
3602 - ``dst {MAC-48}``: destination MAC.
3603 - ``src {MAC-48}``: source MAC.
3604 - ``type {unsigned}``: EtherType or TPID.
3606 - ``vlan``: match 802.1Q/ad VLAN tag.
3608 - ``tci {unsigned}``: tag control information.
3609 - ``pcp {unsigned}``: priority code point.
3610 - ``dei {unsigned}``: drop eligible indicator.
3611 - ``vid {unsigned}``: VLAN identifier.
3612 - ``inner_type {unsigned}``: inner EtherType or TPID.
3614 - ``ipv4``: match IPv4 header.
3616 - ``tos {unsigned}``: type of service.
3617 - ``ttl {unsigned}``: time to live.
3618 - ``proto {unsigned}``: next protocol ID.
3619 - ``src {ipv4 address}``: source address.
3620 - ``dst {ipv4 address}``: destination address.
3622 - ``ipv6``: match IPv6 header.
3624 - ``tc {unsigned}``: traffic class.
3625 - ``flow {unsigned}``: flow label.
3626 - ``proto {unsigned}``: protocol (next header).
3627 - ``hop {unsigned}``: hop limit.
3628 - ``src {ipv6 address}``: source address.
3629 - ``dst {ipv6 address}``: destination address.
3631 - ``icmp``: match ICMP header.
3633 - ``type {unsigned}``: ICMP packet type.
3634 - ``code {unsigned}``: ICMP packet code.
3636 - ``udp``: match UDP header.
3638 - ``src {unsigned}``: UDP source port.
3639 - ``dst {unsigned}``: UDP destination port.
3641 - ``tcp``: match TCP header.
3643 - ``src {unsigned}``: TCP source port.
3644 - ``dst {unsigned}``: TCP destination port.
3646 - ``sctp``: match SCTP header.
3648 - ``src {unsigned}``: SCTP source port.
3649 - ``dst {unsigned}``: SCTP destination port.
3650 - ``tag {unsigned}``: validation tag.
3651 - ``cksum {unsigned}``: checksum.
3653 - ``vxlan``: match VXLAN header.
3655 - ``vni {unsigned}``: VXLAN identifier.
3657 - ``e_tag``: match IEEE 802.1BR E-Tag header.
3659 - ``grp_ecid_b {unsigned}``: GRP and E-CID base.
3661 - ``nvgre``: match NVGRE header.
3663 - ``tni {unsigned}``: virtual subnet ID.
3665 - ``mpls``: match MPLS header.
3667 - ``label {unsigned}``: MPLS label.
3669 - ``gre``: match GRE header.
3671 - ``protocol {unsigned}``: protocol type.
3673 - ``gre_key``: match GRE optional key field.
3675 - ``value {unsigned}``: key value.
3677 - ``fuzzy``: fuzzy pattern match, expect faster than default.
3679 - ``thresh {unsigned}``: accuracy threshold.
3681 - ``gtp``, ``gtpc``, ``gtpu``: match GTPv1 header.
3683 - ``teid {unsigned}``: tunnel endpoint identifier.
3685 - ``geneve``: match GENEVE header.
3687 - ``vni {unsigned}``: virtual network identifier.
3688 - ``protocol {unsigned}``: protocol type.
3690 - ``geneve-opt``: match GENEVE header option.
3692 - ``class {unsigned}``: GENEVE option class.
3693 - ``type {unsigned}``: GENEVE option type.
3694 - ``length {unsigned}``: GENEVE option length in 32-bit words.
3695 - ``data {hex string}``: GENEVE option data, the length is defined by
3698 - ``vxlan-gpe``: match VXLAN-GPE header.
3700 - ``vni {unsigned}``: VXLAN-GPE identifier.
3702 - ``arp_eth_ipv4``: match ARP header for Ethernet/IPv4.
3704 - ``sha {MAC-48}``: sender hardware address.
3705 - ``spa {ipv4 address}``: sender IPv4 address.
3706 - ``tha {MAC-48}``: target hardware address.
3707 - ``tpa {ipv4 address}``: target IPv4 address.
3709 - ``ipv6_ext``: match presence of any IPv6 extension header.
3711 - ``next_hdr {unsigned}``: next header.
3713 - ``icmp6``: match any ICMPv6 header.
3715 - ``type {unsigned}``: ICMPv6 type.
3716 - ``code {unsigned}``: ICMPv6 code.
3718 - ``icmp6_nd_ns``: match ICMPv6 neighbor discovery solicitation.
3720 - ``target_addr {ipv6 address}``: target address.
3722 - ``icmp6_nd_na``: match ICMPv6 neighbor discovery advertisement.
3724 - ``target_addr {ipv6 address}``: target address.
3726 - ``icmp6_nd_opt``: match presence of any ICMPv6 neighbor discovery option.
3728 - ``type {unsigned}``: ND option type.
3730 - ``icmp6_nd_opt_sla_eth``: match ICMPv6 neighbor discovery source Ethernet
3731 link-layer address option.
3733 - ``sla {MAC-48}``: source Ethernet LLA.
3735 - ``icmp6_nd_opt_tla_eth``: match ICMPv6 neighbor discovery target Ethernet
3736 link-layer address option.
3738 - ``tla {MAC-48}``: target Ethernet LLA.
3740 - ``meta``: match application specific metadata.
3742 - ``data {unsigned}``: metadata value.
3744 - ``gtp_psc``: match GTP PDU extension header with type 0x85.
3746 - ``pdu_type {unsigned}``: PDU type.
3747 - ``qfi {unsigned}``: QoS flow identifier.
3749 - ``pppoes``, ``pppoed``: match PPPoE header.
3751 - ``session_id {unsigned}``: session identifier.
3753 - ``pppoe_proto_id``: match PPPoE session protocol identifier.
3755 - ``proto_id {unsigned}``: PPP protocol identifier.
3757 - ``l2tpv3oip``: match L2TPv3 over IP header.
3759 - ``session_id {unsigned}``: L2TPv3 over IP session identifier.
3761 - ``ah``: match AH header.
3763 - ``spi {unsigned}``: security parameters index.
3765 - ``pfcp``: match PFCP header.
3767 - ``s_field {unsigned}``: S field.
3768 - ``seid {unsigned}``: session endpoint identifier.
3773 A list of actions starts after the ``actions`` token in the same fashion as
3774 `Matching pattern`_; actions are separated by ``/`` tokens and the list is
3775 terminated by a mandatory ``end`` action.
3777 Actions are named after their type (*RTE_FLOW_ACTION_TYPE_* from ``enum
3778 rte_flow_action_type``).
3780 Dropping all incoming UDPv4 packets can be expressed as follows::
3782 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
3785 Several actions have configurable properties which must be specified when
3786 there is no valid default value. For example, ``queue`` requires a target
3789 This rule redirects incoming UDPv4 traffic to queue index 6::
3791 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
3792 actions queue index 6 / end
3794 While this one could be rejected by PMDs (unspecified queue index)::
3796 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
3799 As defined by *rte_flow*, the list is not ordered, all actions of a given
3800 rule are performed simultaneously. These are equivalent::
3802 queue index 6 / void / mark id 42 / end
3806 void / mark id 42 / queue index 6 / end
3808 All actions in a list should have different types, otherwise only the last
3809 action of a given type is taken into account::
3811 queue index 4 / queue index 5 / queue index 6 / end # will use queue 6
3815 drop / drop / drop / end # drop is performed only once
3819 mark id 42 / queue index 3 / mark id 24 / end # mark will be 24
3821 Considering they are performed simultaneously, opposite and overlapping
3822 actions can sometimes be combined when the end result is unambiguous::
3824 drop / queue index 6 / end # drop has no effect
3828 queue index 6 / rss queues 6 7 8 / end # queue has no effect
3832 drop / passthru / end # drop has no effect
3834 Note that PMDs may still refuse such combinations.
3839 This section lists supported actions and their attributes, if any.
3841 - ``end``: end list of actions.
3843 - ``void``: no-op action.
3845 - ``passthru``: let subsequent rule process matched packets.
3847 - ``jump``: redirect traffic to group on device.
3849 - ``group {unsigned}``: group to redirect to.
3851 - ``mark``: attach 32 bit value to packets.
3853 - ``id {unsigned}``: 32 bit value to return with packets.
3855 - ``flag``: flag packets.
3857 - ``queue``: assign packets to a given queue index.
3859 - ``index {unsigned}``: queue index to use.
3861 - ``drop``: drop packets (note: passthru has priority).
3863 - ``count``: enable counters for this rule.
3865 - ``rss``: spread packets among several queues.
3867 - ``func {hash function}``: RSS hash function to apply, allowed tokens are
3868 ``toeplitz``, ``simple_xor``, ``symmetric_toeplitz`` and ``default``.
3870 - ``level {unsigned}``: encapsulation level for ``types``.
3872 - ``types [{RSS hash type} [...]] end``: specific RSS hash types.
3873 Note that an empty list does not disable RSS but instead requests
3874 unspecified "best-effort" settings.
3876 - ``key {string}``: RSS hash key, overrides ``key_len``.
3878 - ``key_len {unsigned}``: RSS hash key length in bytes, can be used in
3879 conjunction with ``key`` to pad or truncate it.
3881 - ``queues [{unsigned} [...]] end``: queue indices to use.
3883 - ``pf``: direct traffic to physical function.
3885 - ``vf``: direct traffic to a virtual function ID.
3887 - ``original {boolean}``: use original VF ID if possible.
3888 - ``id {unsigned}``: VF ID.
3890 - ``phy_port``: direct packets to physical port index.
3892 - ``original {boolean}``: use original port index if possible.
3893 - ``index {unsigned}``: physical port index.
3895 - ``port_id``: direct matching traffic to a given DPDK port ID.
3897 - ``original {boolean}``: use original DPDK port ID if possible.
3898 - ``id {unsigned}``: DPDK port ID.
3900 - ``of_set_mpls_ttl``: OpenFlow's ``OFPAT_SET_MPLS_TTL``.
3902 - ``mpls_ttl``: MPLS TTL.
3904 - ``of_dec_mpls_ttl``: OpenFlow's ``OFPAT_DEC_MPLS_TTL``.
3906 - ``of_set_nw_ttl``: OpenFlow's ``OFPAT_SET_NW_TTL``.
3908 - ``nw_ttl``: IP TTL.
3910 - ``of_dec_nw_ttl``: OpenFlow's ``OFPAT_DEC_NW_TTL``.
3912 - ``of_copy_ttl_out``: OpenFlow's ``OFPAT_COPY_TTL_OUT``.
3914 - ``of_copy_ttl_in``: OpenFlow's ``OFPAT_COPY_TTL_IN``.
3916 - ``of_pop_vlan``: OpenFlow's ``OFPAT_POP_VLAN``.
3918 - ``of_push_vlan``: OpenFlow's ``OFPAT_PUSH_VLAN``.
3920 - ``ethertype``: Ethertype.
3922 - ``of_set_vlan_vid``: OpenFlow's ``OFPAT_SET_VLAN_VID``.
3924 - ``vlan_vid``: VLAN id.
3926 - ``of_set_vlan_pcp``: OpenFlow's ``OFPAT_SET_VLAN_PCP``.
3928 - ``vlan_pcp``: VLAN priority.
3930 - ``of_pop_mpls``: OpenFlow's ``OFPAT_POP_MPLS``.
3932 - ``ethertype``: Ethertype.
3934 - ``of_push_mpls``: OpenFlow's ``OFPAT_PUSH_MPLS``.
3936 - ``ethertype``: Ethertype.
3938 - ``vxlan_encap``: Performs a VXLAN encapsulation, outer layer configuration
3939 is done through `Config VXLAN Encap outer layers`_.
3941 - ``vxlan_decap``: Performs a decapsulation action by stripping all headers of
3942 the VXLAN tunnel network overlay from the matched flow.
3944 - ``nvgre_encap``: Performs a NVGRE encapsulation, outer layer configuration
3945 is done through `Config NVGRE Encap outer layers`_.
3947 - ``nvgre_decap``: Performs a decapsulation action by stripping all headers of
3948 the NVGRE tunnel network overlay from the matched flow.
3950 - ``l2_encap``: Performs a L2 encapsulation, L2 configuration
3951 is done through `Config L2 Encap`_.
3953 - ``l2_decap``: Performs a L2 decapsulation, L2 configuration
3954 is done through `Config L2 Decap`_.
3956 - ``mplsogre_encap``: Performs a MPLSoGRE encapsulation, outer layer
3957 configuration is done through `Config MPLSoGRE Encap outer layers`_.
3959 - ``mplsogre_decap``: Performs a MPLSoGRE decapsulation, outer layer
3960 configuration is done through `Config MPLSoGRE Decap outer layers`_.
3962 - ``mplsoudp_encap``: Performs a MPLSoUDP encapsulation, outer layer
3963 configuration is done through `Config MPLSoUDP Encap outer layers`_.
3965 - ``mplsoudp_decap``: Performs a MPLSoUDP decapsulation, outer layer
3966 configuration is done through `Config MPLSoUDP Decap outer layers`_.
3968 - ``set_ipv4_src``: Set a new IPv4 source address in the outermost IPv4 header.
3970 - ``ipv4_addr``: New IPv4 source address.
3972 - ``set_ipv4_dst``: Set a new IPv4 destination address in the outermost IPv4
3975 - ``ipv4_addr``: New IPv4 destination address.
3977 - ``set_ipv6_src``: Set a new IPv6 source address in the outermost IPv6 header.
3979 - ``ipv6_addr``: New IPv6 source address.
3981 - ``set_ipv6_dst``: Set a new IPv6 destination address in the outermost IPv6
3984 - ``ipv6_addr``: New IPv6 destination address.
3986 - ``set_tp_src``: Set a new source port number in the outermost TCP/UDP
3989 - ``port``: New TCP/UDP source port number.
3991 - ``set_tp_dst``: Set a new destination port number in the outermost TCP/UDP
3994 - ``port``: New TCP/UDP destination port number.
3996 - ``mac_swap``: Swap the source and destination MAC addresses in the outermost
3999 - ``dec_ttl``: Performs a decrease TTL value action
4001 - ``set_ttl``: Set TTL value with specified value
4002 - ``ttl_value {unsigned}``: The new TTL value to be set
4004 - ``set_mac_src``: set source MAC address
4006 - ``mac_addr {MAC-48}``: new source MAC address
4008 - ``set_mac_dst``: set destination MAC address
4010 - ``mac_addr {MAC-48}``: new destination MAC address
4012 - ``inc_tcp_seq``: Increase sequence number in the outermost TCP header.
4014 - ``value {unsigned}``: Value to increase TCP sequence number by.
4016 - ``dec_tcp_seq``: Decrease sequence number in the outermost TCP header.
4018 - ``value {unsigned}``: Value to decrease TCP sequence number by.
4020 - ``inc_tcp_ack``: Increase acknowledgment number in the outermost TCP header.
4022 - ``value {unsigned}``: Value to increase TCP acknowledgment number by.
4024 - ``dec_tcp_ack``: Decrease acknowledgment number in the outermost TCP header.
4026 - ``value {unsigned}``: Value to decrease TCP acknowledgment number by.
4028 - ``set_ipv4_dscp``: Set IPv4 DSCP value with specified value
4030 - ``dscp_value {unsigned}``: The new DSCP value to be set
4032 - ``set_ipv6_dscp``: Set IPv6 DSCP value with specified value
4034 - ``dscp_value {unsigned}``: The new DSCP value to be set
4036 - ``shared``: Use shared action created via
4037 ``flow shared_action {port_id} create``
4039 - ``shared_action_id {unsigned}``: Shared action ID to use
4041 Destroying flow rules
4042 ~~~~~~~~~~~~~~~~~~~~~
4044 ``flow destroy`` destroys one or more rules from their rule ID (as returned
4045 by ``flow create``), this command calls ``rte_flow_destroy()`` as many
4046 times as necessary::
4048 flow destroy {port_id} rule {rule_id} [...]
4050 If successful, it will show::
4052 Flow rule #[...] destroyed
4054 It does not report anything for rule IDs that do not exist. The usual error
4055 message is shown when a rule cannot be destroyed::
4057 Caught error type [...] ([...]): [...]
4059 ``flow flush`` destroys all rules on a device and does not take extra
4060 arguments. It is bound to ``rte_flow_flush()``::
4062 flow flush {port_id}
4064 Any errors are reported as above.
4066 Creating several rules and destroying them::
4068 testpmd> flow create 0 ingress pattern eth / ipv6 / end
4069 actions queue index 2 / end
4070 Flow rule #0 created
4071 testpmd> flow create 0 ingress pattern eth / ipv4 / end
4072 actions queue index 3 / end
4073 Flow rule #1 created
4074 testpmd> flow destroy 0 rule 0 rule 1
4075 Flow rule #1 destroyed
4076 Flow rule #0 destroyed
4079 The same result can be achieved using ``flow flush``::
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 flush 0
4090 Non-existent rule IDs are ignored::
4092 testpmd> flow create 0 ingress pattern eth / ipv6 / end
4093 actions queue index 2 / end
4094 Flow rule #0 created
4095 testpmd> flow create 0 ingress pattern eth / ipv4 / end
4096 actions queue index 3 / end
4097 Flow rule #1 created
4098 testpmd> flow destroy 0 rule 42 rule 10 rule 2
4100 testpmd> flow destroy 0 rule 0
4101 Flow rule #0 destroyed
4107 ``flow query`` queries a specific action of a flow rule having that
4108 ability. Such actions collect information that can be reported using this
4109 command. It is bound to ``rte_flow_query()``::
4111 flow query {port_id} {rule_id} {action}
4113 If successful, it will display either the retrieved data for known actions
4114 or the following message::
4116 Cannot display result for action type [...] ([...])
4118 Otherwise, it will complain either that the rule does not exist or that some
4121 Flow rule #[...] not found
4125 Caught error type [...] ([...]): [...]
4127 Currently only the ``count`` action is supported. This action reports the
4128 number of packets that hit the flow rule and the total number of bytes. Its
4129 output has the following format::
4132 hits_set: [...] # whether "hits" contains a valid value
4133 bytes_set: [...] # whether "bytes" contains a valid value
4134 hits: [...] # number of packets
4135 bytes: [...] # number of bytes
4137 Querying counters for TCPv6 packets redirected to queue 6::
4139 testpmd> flow create 0 ingress pattern eth / ipv6 / tcp / end
4140 actions queue index 6 / count / end
4141 Flow rule #4 created
4142 testpmd> flow query 0 4 count
4153 ``flow list`` lists existing flow rules sorted by priority and optionally
4154 filtered by group identifiers::
4156 flow list {port_id} [group {group_id}] [...]
4158 This command only fails with the following message if the device does not
4163 Output consists of a header line followed by a short description of each
4164 flow rule, one per line. There is no output at all when no flow rules are
4165 configured on the device::
4167 ID Group Prio Attr Rule
4168 [...] [...] [...] [...] [...]
4170 ``Attr`` column flags:
4172 - ``i`` for ``ingress``.
4173 - ``e`` for ``egress``.
4175 Creating several flow rules and listing them::
4177 testpmd> flow create 0 ingress pattern eth / ipv4 / end
4178 actions queue index 6 / end
4179 Flow rule #0 created
4180 testpmd> flow create 0 ingress pattern eth / ipv6 / end
4181 actions queue index 2 / end
4182 Flow rule #1 created
4183 testpmd> flow create 0 priority 5 ingress pattern eth / ipv4 / udp / end
4184 actions rss queues 6 7 8 end / end
4185 Flow rule #2 created
4186 testpmd> flow list 0
4187 ID Group Prio Attr Rule
4188 0 0 0 i- ETH IPV4 => QUEUE
4189 1 0 0 i- ETH IPV6 => QUEUE
4190 2 0 5 i- ETH IPV4 UDP => RSS
4193 Rules are sorted by priority (i.e. group ID first, then priority level)::
4195 testpmd> flow list 1
4196 ID Group Prio Attr Rule
4197 0 0 0 i- ETH => COUNT
4198 6 0 500 i- ETH IPV6 TCP => DROP COUNT
4199 5 0 1000 i- ETH IPV6 ICMP => QUEUE
4200 1 24 0 i- ETH IPV4 UDP => QUEUE
4201 4 24 10 i- ETH IPV4 TCP => DROP
4202 3 24 20 i- ETH IPV4 => DROP
4203 2 24 42 i- ETH IPV4 UDP => QUEUE
4204 7 63 0 i- ETH IPV6 UDP VXLAN => MARK QUEUE
4207 Output can be limited to specific groups::
4209 testpmd> flow list 1 group 0 group 63
4210 ID Group Prio Attr Rule
4211 0 0 0 i- ETH => COUNT
4212 6 0 500 i- ETH IPV6 TCP => DROP COUNT
4213 5 0 1000 i- ETH IPV6 ICMP => QUEUE
4214 7 63 0 i- ETH IPV6 UDP VXLAN => MARK QUEUE
4217 Toggling isolated mode
4218 ~~~~~~~~~~~~~~~~~~~~~~
4220 ``flow isolate`` can be used to tell the underlying PMD that ingress traffic
4221 must only be injected from the defined flow rules; that no default traffic
4222 is expected outside those rules and the driver is free to assign more
4223 resources to handle them. It is bound to ``rte_flow_isolate()``::
4225 flow isolate {port_id} {boolean}
4227 If successful, enabling or disabling isolated mode shows either::
4229 Ingress traffic on port [...]
4230 is now restricted to the defined flow rules
4234 Ingress traffic on port [...]
4235 is not restricted anymore to the defined flow rules
4237 Otherwise, in case of error::
4239 Caught error type [...] ([...]): [...]
4241 Mainly due to its side effects, PMDs supporting this mode may not have the
4242 ability to toggle it more than once without reinitializing affected ports
4243 first (e.g. by exiting testpmd).
4245 Enabling isolated mode::
4247 testpmd> flow isolate 0 true
4248 Ingress traffic on port 0 is now restricted to the defined flow rules
4251 Disabling isolated mode::
4253 testpmd> flow isolate 0 false
4254 Ingress traffic on port 0 is not restricted anymore to the defined flow rules
4257 Dumping HW internal information
4258 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4260 ``flow dump`` dumps the hardware's internal representation information of
4261 all flows. It is bound to ``rte_flow_dev_dump()``::
4263 flow dump {port_id} {output_file}
4265 If successful, it will show::
4269 Otherwise, it will complain error occurred::
4271 Caught error type [...] ([...]): [...]
4273 Listing and destroying aged flow rules
4274 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4276 ``flow aged`` simply lists aged flow rules be get from api ``rte_flow_get_aged_flows``,
4277 and ``destroy`` parameter can be used to destroy those flow rules in PMD.
4279 flow aged {port_id} [destroy]
4281 Listing current aged flow rules::
4283 testpmd> flow aged 0
4284 Port 0 total aged flows: 0
4285 testpmd> flow create 0 ingress pattern eth / ipv4 src is 2.2.2.14 / end
4286 actions age timeout 5 / queue index 0 / end
4287 Flow rule #0 created
4288 testpmd> flow create 0 ingress pattern eth / ipv4 src is 2.2.2.15 / end
4289 actions age timeout 4 / queue index 0 / end
4290 Flow rule #1 created
4291 testpmd> flow create 0 ingress pattern eth / ipv4 src is 2.2.2.16 / end
4292 actions age timeout 2 / queue index 0 / end
4293 Flow rule #2 created
4294 testpmd> flow create 0 ingress pattern eth / ipv4 src is 2.2.2.17 / end
4295 actions age timeout 3 / queue index 0 / end
4296 Flow rule #3 created
4299 Aged Rules are simply list as command ``flow list {port_id}``, but strip the detail rule
4300 information, all the aged flows are sorted by the longest timeout time. For example, if
4301 those rules be configured in the same time, ID 2 will be the first aged out rule, the next
4302 will be ID 3, ID 1, ID 0::
4304 testpmd> flow aged 0
4305 Port 0 total aged flows: 4
4312 If attach ``destroy`` parameter, the command will destroy all the list aged flow rules.
4314 testpmd> flow aged 0 destroy
4315 Port 0 total aged flows: 4
4322 Flow rule #2 destroyed
4323 Flow rule #3 destroyed
4324 Flow rule #1 destroyed
4325 Flow rule #0 destroyed
4326 4 flows be destroyed
4327 testpmd> flow aged 0
4328 Port 0 total aged flows: 0
4330 Creating shared actions
4331 ~~~~~~~~~~~~~~~~~~~~~~~
4332 ``flow shared_action {port_id} create`` creates shared action with optional
4333 shared action ID. It is bound to ``rte_flow_shared_action_create()``::
4335 flow shared_action {port_id} create [action_id {shared_action_id}]
4336 [ingress] [egress] [transfer] action {action} / end
4338 If successful, it will show::
4340 Shared action #[...] created
4342 Otherwise, it will complain either that shared action already exists or that
4343 some error occurred::
4345 Shared action #[...] is already assigned, delete it first
4349 Caught error type [...] ([...]): [...]
4351 Create shared rss action with id 100 to queues 1 and 2 on port 0::
4353 testpmd> flow shared_action 0 create action_id 100 \
4354 ingress action rss queues 1 2 end / end
4356 Create shared rss action with id assigned by testpmd to queues 1 and 2 on
4359 testpmd> flow shared_action 0 create action_id \
4360 ingress action rss queues 0 1 end / end
4362 Updating shared actions
4363 ~~~~~~~~~~~~~~~~~~~~~~~
4364 ``flow shared_action {port_id} update`` updates configuration of the shared
4365 action from its shared action ID (as returned by
4366 ``flow shared_action {port_id} create``). It is bound to
4367 ``rte_flow_shared_action_update()``::
4369 flow shared_action {port_id} update {shared_action_id}
4370 action {action} / end
4372 If successful, it will show::
4374 Shared action #[...] updated
4376 Otherwise, it will complain either that shared action not found or that some
4379 Failed to find shared action #[...] on port [...]
4383 Caught error type [...] ([...]): [...]
4385 Update shared rss action having id 100 on port 0 with rss to queues 0 and 3
4386 (in create example above rss queues were 1 and 2)::
4388 testpmd> flow shared_action 0 update 100 action rss queues 0 3 end / end
4390 Destroying shared actions
4391 ~~~~~~~~~~~~~~~~~~~~~~~~~
4392 ``flow shared_action {port_id} update`` destroys one or more shared actions
4393 from their shared action IDs (as returned by
4394 ``flow shared_action {port_id} create``). It is bound to
4395 ``rte_flow_shared_action_destroy()``::
4397 flow shared_action {port_id} destroy action_id {shared_action_id} [...]
4399 If successful, it will show::
4401 Shared action #[...] destroyed
4403 It does not report anything for shared action IDs that do not exist.
4404 The usual error message is shown when a shared action cannot be destroyed::
4406 Caught error type [...] ([...]): [...]
4408 Destroy shared actions having id 100 & 101::
4410 testpmd> flow shared_action 0 destroy action_id 100 action_id 101
4412 Query shared actions
4413 ~~~~~~~~~~~~~~~~~~~~
4414 ``flow shared_action {port_id} query`` queries the shared action from its
4415 shared action ID (as returned by ``flow shared_action {port_id} create``).
4416 It is bound to ``rte_flow_shared_action_query()``::
4418 flow shared_action {port_id} query {shared_action_id}
4420 Currently only rss shared action supported. If successful, it will show::
4425 Otherwise, it will complain either that shared action not found or that some
4428 Failed to find shared action #[...] on port [...]
4432 Caught error type [...] ([...]): [...]
4434 Query shared action having id 100::
4436 testpmd> flow shared_action 0 query 100
4438 Sample QinQ flow rules
4439 ~~~~~~~~~~~~~~~~~~~~~~
4441 Before creating QinQ rule(s) the following commands should be issued to enable QinQ::
4443 testpmd> port stop 0
4444 testpmd> vlan set extend on 0
4446 The above command sets the inner and outer TPID's to 0x8100.
4448 To change the TPID's the following commands should be used::
4450 testpmd> vlan set outer tpid 0x88A8 0
4451 testpmd> vlan set inner tpid 0x8100 0
4452 testpmd> port start 0
4454 Validate and create a QinQ rule on port 0 to steer traffic to a VF queue in a VM.
4458 testpmd> flow validate 0 ingress pattern eth / vlan tci is 123 /
4459 vlan tci is 456 / end actions vf id 1 / queue index 0 / end
4460 Flow rule #0 validated
4462 testpmd> flow create 0 ingress pattern eth / vlan tci is 4 /
4463 vlan tci is 456 / end actions vf id 123 / queue index 0 / end
4464 Flow rule #0 created
4466 testpmd> flow list 0
4467 ID Group Prio Attr Rule
4468 0 0 0 i- ETH VLAN VLAN=>VF QUEUE
4470 Validate and create a QinQ rule on port 0 to steer traffic to a queue on the host.
4474 testpmd> flow validate 0 ingress pattern eth / vlan tci is 321 /
4475 vlan tci is 654 / end actions pf / queue index 0 / end
4476 Flow rule #1 validated
4478 testpmd> flow create 0 ingress pattern eth / vlan tci is 321 /
4479 vlan tci is 654 / end actions pf / queue index 1 / end
4480 Flow rule #1 created
4482 testpmd> flow list 0
4483 ID Group Prio Attr Rule
4484 0 0 0 i- ETH VLAN VLAN=>VF QUEUE
4485 1 0 0 i- ETH VLAN VLAN=>PF QUEUE
4487 Sample VXLAN flow rules
4488 ~~~~~~~~~~~~~~~~~~~~~~~
4490 Before creating VXLAN rule(s), the UDP port should be added for VXLAN packet
4493 testpmd> rx_vxlan_port add 4789 0
4495 Create VXLAN rules on port 0 to steer traffic to PF queues.
4499 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / vxlan /
4500 eth dst is 00:11:22:33:44:55 / end actions pf / queue index 1 / end
4501 Flow rule #0 created
4503 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / vxlan vni is 3 /
4504 eth dst is 00:11:22:33:44:55 / end actions pf / queue index 2 / end
4505 Flow rule #1 created
4507 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / vxlan /
4508 eth dst is 00:11:22:33:44:55 / vlan tci is 10 / end actions pf /
4510 Flow rule #2 created
4512 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / vxlan vni is 5 /
4513 eth dst is 00:11:22:33:44:55 / vlan tci is 20 / end actions pf /
4515 Flow rule #3 created
4517 testpmd> flow create 0 ingress pattern eth dst is 00:00:00:00:01:00 / ipv4 /
4518 udp / vxlan vni is 6 / eth dst is 00:11:22:33:44:55 / end actions pf /
4520 Flow rule #4 created
4522 testpmd> flow list 0
4523 ID Group Prio Attr Rule
4524 0 0 0 i- ETH IPV4 UDP VXLAN ETH => QUEUE
4525 1 0 0 i- ETH IPV4 UDP VXLAN ETH => QUEUE
4526 2 0 0 i- ETH IPV4 UDP VXLAN ETH VLAN => QUEUE
4527 3 0 0 i- ETH IPV4 UDP VXLAN ETH VLAN => QUEUE
4528 4 0 0 i- ETH IPV4 UDP VXLAN ETH => QUEUE
4530 Sample VXLAN encapsulation rule
4531 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4533 VXLAN encapsulation outer layer has default value pre-configured in testpmd
4534 source code, those can be changed by using the following commands
4536 IPv4 VXLAN outer header::
4538 testpmd> set vxlan ip-version ipv4 vni 4 udp-src 4 udp-dst 4 ip-src 127.0.0.1
4539 ip-dst 128.0.0.1 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4540 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4543 testpmd> set vxlan-with-vlan ip-version ipv4 vni 4 udp-src 4 udp-dst 4 ip-src
4544 127.0.0.1 ip-dst 128.0.0.1 vlan-tci 34 eth-src 11:11:11:11:11:11
4545 eth-dst 22:22:22:22:22:22
4546 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4549 testpmd> set vxlan-tos-ttl ip-version ipv4 vni 4 udp-src 4 udp-dst 4 ip-tos 0
4550 ip-ttl 255 ip-src 127.0.0.1 ip-dst 128.0.0.1 eth-src 11:11:11:11:11:11
4551 eth-dst 22:22:22:22:22:22
4552 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4555 IPv6 VXLAN outer header::
4557 testpmd> set vxlan ip-version ipv6 vni 4 udp-src 4 udp-dst 4 ip-src ::1
4558 ip-dst ::2222 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4559 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4562 testpmd> set vxlan-with-vlan ip-version ipv6 vni 4 udp-src 4 udp-dst 4
4563 ip-src ::1 ip-dst ::2222 vlan-tci 34 eth-src 11:11:11:11:11:11
4564 eth-dst 22:22:22:22:22:22
4565 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4568 testpmd> set vxlan-tos-ttl ip-version ipv6 vni 4 udp-src 4 udp-dst 4
4569 ip-tos 0 ip-ttl 255 ::1 ip-dst ::2222 eth-src 11:11:11:11:11:11
4570 eth-dst 22:22:22:22:22:22
4571 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4574 Sample NVGRE encapsulation rule
4575 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4577 NVGRE encapsulation outer layer has default value pre-configured in testpmd
4578 source code, those can be changed by using the following commands
4580 IPv4 NVGRE outer header::
4582 testpmd> set nvgre ip-version ipv4 tni 4 ip-src 127.0.0.1 ip-dst 128.0.0.1
4583 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4584 testpmd> flow create 0 ingress pattern end actions nvgre_encap /
4587 testpmd> set nvgre-with-vlan ip-version ipv4 tni 4 ip-src 127.0.0.1
4588 ip-dst 128.0.0.1 vlan-tci 34 eth-src 11:11:11:11:11:11
4589 eth-dst 22:22:22:22:22:22
4590 testpmd> flow create 0 ingress pattern end actions nvgre_encap /
4593 IPv6 NVGRE outer header::
4595 testpmd> set nvgre ip-version ipv6 tni 4 ip-src ::1 ip-dst ::2222
4596 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4597 testpmd> flow create 0 ingress pattern end actions nvgre_encap /
4600 testpmd> set nvgre-with-vlan ip-version ipv6 tni 4 ip-src ::1 ip-dst ::2222
4601 vlan-tci 34 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4602 testpmd> flow create 0 ingress pattern end actions nvgre_encap /
4605 Sample L2 encapsulation rule
4606 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4608 L2 encapsulation has default value pre-configured in testpmd
4609 source code, those can be changed by using the following commands
4613 testpmd> set l2_encap ip-version ipv4
4614 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4615 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / mpls / end actions
4616 mplsoudp_decap / l2_encap / end
4618 L2 with VXLAN header::
4620 testpmd> set l2_encap-with-vlan ip-version ipv4 vlan-tci 34
4621 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4622 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / mpls / end actions
4623 mplsoudp_decap / l2_encap / end
4625 Sample L2 decapsulation rule
4626 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4628 L2 decapsulation has default value pre-configured in testpmd
4629 source code, those can be changed by using the following commands
4633 testpmd> set l2_decap
4634 testpmd> flow create 0 egress pattern eth / end actions l2_decap / mplsoudp_encap /
4637 L2 with VXLAN header::
4639 testpmd> set l2_encap-with-vlan
4640 testpmd> flow create 0 egress pattern eth / end actions l2_encap / mplsoudp_encap /
4643 Sample MPLSoGRE encapsulation rule
4644 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4646 MPLSoGRE encapsulation outer layer has default value pre-configured in testpmd
4647 source code, those can be changed by using the following commands
4649 IPv4 MPLSoGRE outer header::
4651 testpmd> set mplsogre_encap ip-version ipv4 label 4
4652 ip-src 127.0.0.1 ip-dst 128.0.0.1 eth-src 11:11:11:11:11:11
4653 eth-dst 22:22:22:22:22:22
4654 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4655 mplsogre_encap / end
4657 IPv4 MPLSoGRE with VLAN outer header::
4659 testpmd> set mplsogre_encap-with-vlan ip-version ipv4 label 4
4660 ip-src 127.0.0.1 ip-dst 128.0.0.1 vlan-tci 34
4661 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4662 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4663 mplsogre_encap / end
4665 IPv6 MPLSoGRE outer header::
4667 testpmd> set mplsogre_encap ip-version ipv6 mask 4
4668 ip-src ::1 ip-dst ::2222 eth-src 11:11:11:11:11:11
4669 eth-dst 22:22:22:22:22:22
4670 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4671 mplsogre_encap / end
4673 IPv6 MPLSoGRE with VLAN outer header::
4675 testpmd> set mplsogre_encap-with-vlan ip-version ipv6 mask 4
4676 ip-src ::1 ip-dst ::2222 vlan-tci 34
4677 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4678 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4679 mplsogre_encap / end
4681 Sample MPLSoGRE decapsulation rule
4682 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4684 MPLSoGRE decapsulation outer layer has default value pre-configured in testpmd
4685 source code, those can be changed by using the following commands
4687 IPv4 MPLSoGRE outer header::
4689 testpmd> set mplsogre_decap ip-version ipv4
4690 testpmd> flow create 0 ingress pattern eth / ipv4 / gre / mpls / end actions
4691 mplsogre_decap / l2_encap / end
4693 IPv4 MPLSoGRE with VLAN outer header::
4695 testpmd> set mplsogre_decap-with-vlan ip-version ipv4
4696 testpmd> flow create 0 ingress pattern eth / vlan / ipv4 / gre / mpls / end
4697 actions mplsogre_decap / l2_encap / end
4699 IPv6 MPLSoGRE outer header::
4701 testpmd> set mplsogre_decap ip-version ipv6
4702 testpmd> flow create 0 ingress pattern eth / ipv6 / gre / mpls / end
4703 actions mplsogre_decap / l2_encap / end
4705 IPv6 MPLSoGRE with VLAN outer header::
4707 testpmd> set mplsogre_decap-with-vlan ip-version ipv6
4708 testpmd> flow create 0 ingress pattern eth / vlan / ipv6 / gre / mpls / end
4709 actions mplsogre_decap / l2_encap / end
4711 Sample MPLSoUDP encapsulation rule
4712 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4714 MPLSoUDP encapsulation outer layer has default value pre-configured in testpmd
4715 source code, those can be changed by using the following commands
4717 IPv4 MPLSoUDP outer header::
4719 testpmd> set mplsoudp_encap ip-version ipv4 label 4 udp-src 5 udp-dst 10
4720 ip-src 127.0.0.1 ip-dst 128.0.0.1 eth-src 11:11:11:11:11:11
4721 eth-dst 22:22:22:22:22:22
4722 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4723 mplsoudp_encap / end
4725 IPv4 MPLSoUDP with VLAN outer header::
4727 testpmd> set mplsoudp_encap-with-vlan ip-version ipv4 label 4 udp-src 5
4728 udp-dst 10 ip-src 127.0.0.1 ip-dst 128.0.0.1 vlan-tci 34
4729 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4730 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4731 mplsoudp_encap / end
4733 IPv6 MPLSoUDP outer header::
4735 testpmd> set mplsoudp_encap ip-version ipv6 mask 4 udp-src 5 udp-dst 10
4736 ip-src ::1 ip-dst ::2222 eth-src 11:11:11:11:11:11
4737 eth-dst 22:22:22:22:22:22
4738 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4739 mplsoudp_encap / end
4741 IPv6 MPLSoUDP with VLAN outer header::
4743 testpmd> set mplsoudp_encap-with-vlan ip-version ipv6 mask 4 udp-src 5
4744 udp-dst 10 ip-src ::1 ip-dst ::2222 vlan-tci 34
4745 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4746 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4747 mplsoudp_encap / end
4749 Sample MPLSoUDP decapsulation rule
4750 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4752 MPLSoUDP decapsulation outer layer has default value pre-configured in testpmd
4753 source code, those can be changed by using the following commands
4755 IPv4 MPLSoUDP outer header::
4757 testpmd> set mplsoudp_decap ip-version ipv4
4758 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / mpls / end actions
4759 mplsoudp_decap / l2_encap / end
4761 IPv4 MPLSoUDP with VLAN outer header::
4763 testpmd> set mplsoudp_decap-with-vlan ip-version ipv4
4764 testpmd> flow create 0 ingress pattern eth / vlan / ipv4 / udp / mpls / end
4765 actions mplsoudp_decap / l2_encap / end
4767 IPv6 MPLSoUDP outer header::
4769 testpmd> set mplsoudp_decap ip-version ipv6
4770 testpmd> flow create 0 ingress pattern eth / ipv6 / udp / mpls / end
4771 actions mplsoudp_decap / l2_encap / end
4773 IPv6 MPLSoUDP with VLAN outer header::
4775 testpmd> set mplsoudp_decap-with-vlan ip-version ipv6
4776 testpmd> flow create 0 ingress pattern eth / vlan / ipv6 / udp / mpls / end
4777 actions mplsoudp_decap / l2_encap / end
4779 Sample Raw encapsulation rule
4780 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4782 Raw encapsulation configuration can be set by the following commands
4784 Eecapsulating VxLAN::
4786 testpmd> set raw_encap 4 eth src is 10:11:22:33:44:55 / vlan tci is 1
4787 inner_type is 0x0800 / ipv4 / udp dst is 4789 / vxlan vni
4789 testpmd> flow create 0 egress pattern eth / ipv4 / end actions
4790 raw_encap index 4 / end
4792 Sample Raw decapsulation rule
4793 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4795 Raw decapsulation configuration can be set by the following commands
4797 Decapsulating VxLAN::
4799 testpmd> set raw_decap eth / ipv4 / udp / vxlan / end_set
4800 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / vxlan / eth / ipv4 /
4801 end actions raw_decap / queue index 0 / end
4806 ESP rules can be created by the following commands::
4808 testpmd> flow create 0 ingress pattern eth / ipv4 / esp spi is 1 / end actions
4810 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / esp spi is 1 / end
4811 actions queue index 3 / end
4812 testpmd> flow create 0 ingress pattern eth / ipv6 / esp spi is 1 / end actions
4814 testpmd> flow create 0 ingress pattern eth / ipv6 / udp / esp spi is 1 / end
4815 actions queue index 3 / end
4820 AH rules can be created by the following commands::
4822 testpmd> flow create 0 ingress pattern eth / ipv4 / ah spi is 1 / end actions
4824 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / ah spi is 1 / end
4825 actions queue index 3 / end
4826 testpmd> flow create 0 ingress pattern eth / ipv6 / ah spi is 1 / end actions
4828 testpmd> flow create 0 ingress pattern eth / ipv6 / udp / ah spi is 1 / end
4829 actions queue index 3 / end
4834 PFCP rules can be created by the following commands(s_field need to be 1
4837 testpmd> flow create 0 ingress pattern eth / ipv4 / pfcp s_field is 0 / end
4838 actions queue index 3 / end
4839 testpmd> flow create 0 ingress pattern eth / ipv4 / pfcp s_field is 1
4840 seid is 1 / end actions queue index 3 / end
4841 testpmd> flow create 0 ingress pattern eth / ipv6 / pfcp s_field is 0 / end
4842 actions queue index 3 / end
4843 testpmd> flow create 0 ingress pattern eth / ipv6 / pfcp s_field is 1
4844 seid is 1 / end actions queue index 3 / end
4849 The following sections show functions to load/unload eBPF based filters.
4854 Load an eBPF program as a callback for particular RX/TX queue::
4856 testpmd> bpf-load rx|tx (portid) (queueid) (load-flags) (bpf-prog-filename)
4858 The available load-flags are:
4860 * ``J``: use JIT generated native code, otherwise BPF interpreter will be used.
4862 * ``M``: assume input parameter is a pointer to rte_mbuf, otherwise assume it is a pointer to first segment's data.
4868 You'll need clang v3.7 or above to build bpf program you'd like to load
4872 .. code-block:: console
4875 clang -O2 -target bpf -c t1.c
4877 Then to load (and JIT compile) t1.o at RX queue 0, port 1:
4879 .. code-block:: console
4881 testpmd> bpf-load rx 1 0 J ./dpdk.org/examples/bpf/t1.o
4883 To load (not JITed) t1.o at TX queue 0, port 0:
4885 .. code-block:: console
4887 testpmd> bpf-load tx 0 0 - ./dpdk.org/examples/bpf/t1.o
4892 Unload previously loaded eBPF program for particular RX/TX queue::
4894 testpmd> bpf-unload rx|tx (portid) (queueid)
4896 For example to unload BPF filter from TX queue 0, port 0:
4898 .. code-block:: console
4900 testpmd> bpf-unload tx 0 0