1 .. SPDX-License-Identifier: BSD-3-Clause
2 Copyright(c) 2010-2016 Intel Corporation.
6 Testpmd Runtime Functions
7 =========================
9 Where the testpmd application is started in interactive mode, (``-i|--interactive``),
10 it displays a prompt that can be used to start and stop forwarding,
11 configure the application, display statistics (including the extended NIC
12 statistics aka xstats) , set the Flow Director and other tasks::
16 The testpmd prompt has some, limited, readline support.
17 Common bash command-line functions such as ``Ctrl+a`` and ``Ctrl+e`` to go to the start and end of the prompt line are supported
18 as well as access to the command history via the up-arrow.
20 There is also support for tab completion.
21 If you type a partial command and hit ``<TAB>`` you get a list of the available completions:
23 .. code-block:: console
25 testpmd> show port <TAB>
27 info [Mul-choice STRING]: show|clear port info|stats|xstats|fdir|dcb_tc|cap X
28 info [Mul-choice STRING]: show|clear port info|stats|xstats|fdir|dcb_tc|cap all
29 stats [Mul-choice STRING]: show|clear port info|stats|xstats|fdir|dcb_tc|cap X
30 stats [Mul-choice STRING]: show|clear port info|stats|xstats|fdir|dcb_tc|cap all
36 Some examples in this document are too long to fit on one line are shown wrapped at `"\\"` for display purposes::
38 testpmd> set flow_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \
39 (pause_time) (send_xon) (port_id)
41 In the real ``testpmd>`` prompt these commands should be on a single line.
46 The testpmd has on-line help for the functions that are available at runtime.
47 These are divided into sections and can be accessed using help, help section or help all:
49 .. code-block:: console
53 help control : Start and stop forwarding.
54 help display : Displaying port, stats and config information.
55 help config : Configuration information.
56 help ports : Configuring ports.
57 help registers : Reading and setting port registers.
58 help filters : Filters configuration help.
59 help all : All of the above sections.
62 Command File Functions
63 ----------------------
65 To facilitate loading large number of commands or to avoid cutting and pasting where not
66 practical or possible testpmd supports alternative methods for executing commands.
68 * If started with the ``--cmdline-file=FILENAME`` command line argument testpmd
69 will execute all CLI commands contained within the file immediately before
70 starting packet forwarding or entering interactive mode.
72 .. code-block:: console
74 ./dpdk-testpmd -n4 -r2 ... -- -i --cmdline-file=/home/ubuntu/flow-create-commands.txt
75 Interactive-mode selected
76 CLI commands to be read from /home/ubuntu/flow-create-commands.txt
77 Configuring Port 0 (socket 0)
78 Port 0: 7C:FE:90:CB:74:CE
79 Configuring Port 1 (socket 0)
80 Port 1: 7C:FE:90:CB:74:CA
81 Checking link statuses...
82 Port 0 Link Up - speed 10000 Mbps - full-duplex
83 Port 1 Link Up - speed 10000 Mbps - full-duplex
89 Flow rule #498 created
90 Flow rule #499 created
91 Read all CLI commands from /home/ubuntu/flow-create-commands.txt
95 * At run-time additional commands can be loaded in bulk by invoking the ``load FILENAME``
98 .. code-block:: console
100 testpmd> load /home/ubuntu/flow-create-commands.txt
105 Flow rule #498 created
106 Flow rule #499 created
107 Read all CLI commands from /home/ubuntu/flow-create-commands.txt
111 In all cases output from any included command will be displayed as standard output.
112 Execution will continue until the end of the file is reached regardless of
113 whether any errors occur. The end user must examine the output to determine if
114 any failures occurred.
123 Start packet forwarding with current configuration::
130 Start packet forwarding with current configuration after sending specified number of bursts of packets::
132 testpmd> start tx_first (""|burst_num)
134 The default burst number is 1 when ``burst_num`` not presented.
139 Stop packet forwarding, and display accumulated statistics::
154 The functions in the following sections are used to display information about the
155 testpmd configuration or the NIC status.
160 Display information for a given port or all ports::
162 testpmd> show port (info|summary|stats|xstats|fdir|dcb_tc|cap) (port_id|all)
164 The available information categories are:
166 * ``info``: General port information such as MAC address.
168 * ``summary``: Brief port summary such as Device Name, Driver Name etc.
170 * ``stats``: RX/TX statistics.
172 * ``xstats``: RX/TX extended NIC statistics.
174 * ``fdir``: Flow Director information and statistics.
176 * ``dcb_tc``: DCB information such as TC mapping.
180 .. code-block:: console
182 testpmd> show port info 0
184 ********************* Infos for port 0 *********************
186 MAC address: XX:XX:XX:XX:XX:XX
188 memory allocation on the socket: 0
190 Link speed: 40000 Mbps
191 Link duplex: full-duplex
192 Promiscuous mode: enabled
193 Allmulticast mode: disabled
194 Maximum number of MAC addresses: 64
195 Maximum number of MAC addresses of hash filtering: 0
197 strip on, filter on, extend off, qinq strip off
198 Redirection table size: 512
199 Supported flow types:
217 show port (module_eeprom|eeprom)
218 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
220 Display the EEPROM information of a port::
222 testpmd> show port (port_id) (module_eeprom|eeprom)
227 Display the rss redirection table entry indicated by masks on port X::
229 testpmd> show port (port_id) rss reta (size) (mask0, mask1...)
231 size is used to indicate the hardware supported reta size
236 Display the RSS hash functions and RSS hash key of a port::
238 testpmd> show port (port_id) rss-hash [key]
243 Clear the port statistics and forward engine statistics for a given port or for all ports::
245 testpmd> clear port (info|stats|xstats|fdir) (port_id|all)
249 testpmd> clear port stats all
254 Display information for a given port's RX/TX queue::
256 testpmd> show (rxq|txq) info (port_id) (queue_id)
258 show desc status(rxq|txq)
259 ~~~~~~~~~~~~~~~~~~~~~~~~~
261 Display information for a given port's RX/TX descriptor status::
263 testpmd> show port (port_id) (rxq|txq) (queue_id) desc (desc_id) status
265 show rxq desc used count
266 ~~~~~~~~~~~~~~~~~~~~~~~~
268 Display the number of receive packet descriptors currently filled by hardware
269 and ready to be processed by the driver on a given RX queue::
271 testpmd> show port (port_id) rxq (queue_id) desc used count
276 Displays the configuration of the application.
277 The configuration comes from the command-line, the runtime or the application defaults::
279 testpmd> show config (rxtx|cores|fwd|rxoffs|rxpkts|txpkts|txtimes)
281 The available information categories are:
283 * ``rxtx``: RX/TX configuration items.
285 * ``cores``: List of forwarding cores.
287 * ``fwd``: Packet forwarding configuration.
289 * ``rxoffs``: Packet offsets for RX split.
291 * ``rxpkts``: Packets to RX split configuration.
293 * ``txpkts``: Packets to TX configuration.
295 * ``txtimes``: Burst time pattern for Tx only mode.
299 .. code-block:: console
301 testpmd> show config rxtx
303 io packet forwarding - CRC stripping disabled - packets/burst=16
304 nb forwarding cores=2 - nb forwarding ports=1
305 RX queues=1 - RX desc=128 - RX free threshold=0
306 RX threshold registers: pthresh=8 hthresh=8 wthresh=4
307 TX queues=1 - TX desc=512 - TX free threshold=0
308 TX threshold registers: pthresh=36 hthresh=0 wthresh=0
309 TX RS bit threshold=0 - TXQ flags=0x0
314 Set the packet forwarding mode::
316 testpmd> set fwd (io|mac|macswap|flowgen| \
317 rxonly|txonly|csum|icmpecho|noisy|5tswap) (""|retry)
319 ``retry`` can be specified for forwarding engines except ``rx_only``.
321 The available information categories are:
323 * ``io``: Forwards packets "as-is" in I/O mode.
324 This is the fastest possible forwarding operation as it does not access packets data.
325 This is the default mode.
327 * ``mac``: Changes the source and the destination Ethernet addresses of packets before forwarding them.
328 Default application behavior is to set source Ethernet address to that of the transmitting interface, and destination
329 address to a dummy value (set during init). The user may specify a target destination Ethernet address via the 'eth-peer' or
330 'eth-peers-configfile' command-line options. It is not currently possible to specify a specific source Ethernet address.
332 * ``macswap``: MAC swap forwarding mode.
333 Swaps the source and the destination Ethernet addresses of packets before forwarding them.
335 * ``flowgen``: Multi-flow generation mode.
336 Originates a number of flows (with varying destination IP addresses), and terminate receive traffic.
338 * ``rxonly``: Receives packets but doesn't transmit them.
340 * ``txonly``: Generates and transmits packets without receiving any.
342 * ``csum``: Changes the checksum field with hardware or software methods depending on the offload flags on the packet.
344 * ``icmpecho``: Receives a burst of packets, lookup for ICMP echo requests and, if any, send back ICMP echo replies.
346 * ``ieee1588``: Demonstrate L2 IEEE1588 V2 PTP timestamping for RX and TX.
348 * ``noisy``: Noisy neighbor simulation.
349 Simulate more realistic behavior of a guest machine engaged in receiving
350 and sending packets performing Virtual Network Function (VNF).
352 * ``5tswap``: Swap the source and destination of L2,L3,L4 if they exist.
354 L2 swaps the source address and destination address of Ethernet, as same as ``macswap``.
356 L3 swaps the source address and destination address of IP (v4 and v6).
358 L4 swaps the source port and destination port of transport layer (TCP and UDP).
362 testpmd> set fwd rxonly
364 Set rxonly packet forwarding mode
370 When running, forwarding engines maintain statistics from the time they have been started.
371 Example for the io forwarding engine, with some packet drops on the tx side::
373 testpmd> show fwd stats all
375 ------- Forward Stats for RX Port= 0/Queue= 0 -> TX Port= 1/Queue= 0 -------
376 RX-packets: 274293770 TX-packets: 274293642 TX-dropped: 128
378 ------- Forward Stats for RX Port= 1/Queue= 0 -> TX Port= 0/Queue= 0 -------
379 RX-packets: 274301850 TX-packets: 274301850 TX-dropped: 0
381 ---------------------- Forward statistics for port 0 ----------------------
382 RX-packets: 274293802 RX-dropped: 0 RX-total: 274293802
383 TX-packets: 274301862 TX-dropped: 0 TX-total: 274301862
384 ----------------------------------------------------------------------------
386 ---------------------- Forward statistics for port 1 ----------------------
387 RX-packets: 274301894 RX-dropped: 0 RX-total: 274301894
388 TX-packets: 274293706 TX-dropped: 128 TX-total: 274293834
389 ----------------------------------------------------------------------------
391 +++++++++++++++ Accumulated forward statistics for all ports+++++++++++++++
392 RX-packets: 548595696 RX-dropped: 0 RX-total: 548595696
393 TX-packets: 548595568 TX-dropped: 128 TX-total: 548595696
394 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
400 Clear the forwarding engines statistics::
402 testpmd> clear fwd stats all
407 Display an RX descriptor for a port RX queue::
409 testpmd> read rxd (port_id) (queue_id) (rxd_id)
413 testpmd> read rxd 0 0 4
414 0x0000000B - 0x001D0180 / 0x0000000B - 0x001D0180
419 Display a TX descriptor for a port TX queue::
421 testpmd> read txd (port_id) (queue_id) (txd_id)
425 testpmd> read txd 0 0 4
426 0x00000001 - 0x24C3C440 / 0x000F0000 - 0x2330003C
431 Get loaded dynamic device personalization (DDP) package info list::
433 testpmd> ddp get list (port_id)
438 Display information about dynamic device personalization (DDP) profile::
440 testpmd> ddp get info (profile_path)
445 Display VF statistics::
447 testpmd> show vf stats (port_id) (vf_id)
452 Reset VF statistics::
454 testpmd> clear vf stats (port_id) (vf_id)
456 show port pctype mapping
457 ~~~~~~~~~~~~~~~~~~~~~~~~
459 List all items from the pctype mapping table::
461 testpmd> show port (port_id) pctype mapping
463 show rx offloading capabilities
464 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
466 List all per queue and per port Rx offloading capabilities of a port::
468 testpmd> show port (port_id) rx_offload capabilities
470 show rx offloading configuration
471 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
473 List port level and all queue level Rx offloading configuration::
475 testpmd> show port (port_id) rx_offload configuration
477 show tx offloading capabilities
478 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
480 List all per queue and per port Tx offloading capabilities of a port::
482 testpmd> show port (port_id) tx_offload capabilities
484 show tx offloading configuration
485 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
487 List port level and all queue level Tx offloading configuration::
489 testpmd> show port (port_id) tx_offload configuration
491 show tx metadata setting
492 ~~~~~~~~~~~~~~~~~~~~~~~~
494 Show Tx metadata value set for a specific port::
496 testpmd> show port (port_id) tx_metadata
498 show port supported ptypes
499 ~~~~~~~~~~~~~~~~~~~~~~~~~~
501 Show ptypes supported for a specific port::
503 testpmd> show port (port_id) ptypes
505 set port supported ptypes
506 ~~~~~~~~~~~~~~~~~~~~~~~~~
508 set packet types classification for a specific port::
510 testpmd> set port (port_id) ptypes_mask (mask)
512 show port mac addresses info
513 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
515 Show mac addresses added for a specific port::
517 testpmd> show port (port_id) macs
520 show port multicast mac addresses info
521 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
523 Show multicast mac addresses added for a specific port::
525 testpmd> show port (port_id) mcast_macs
530 Show general information about devices probed::
532 testpmd> show device info (<identifier>|all)
536 .. code-block:: console
538 testpmd> show device info net_pcap0
540 ********************* Infos for device net_pcap0 *********************
542 Driver name: net_pcap
543 Devargs: iface=enP2p6s0,phy_mac=1
544 Connect to socket: -1
547 MAC address: 1E:37:93:28:04:B8
548 Device name: net_pcap0
553 Dumps all physical memory segment layouts::
555 testpmd> dump_physmem
560 Dumps the layout of all memory zones::
562 testpmd> dump_memzone
567 Dumps the memory usage of all sockets::
569 testpmd> dump_socket_mem
574 Dumps the size of all memory structures::
576 testpmd> dump_struct_sizes
581 Dumps the status of all or specific element in DPDK rings::
583 testpmd> dump_ring [ring_name]
588 Dumps the statistics of all or specific memory pool::
590 testpmd> dump_mempool [mempool_name]
595 Dumps the user device list::
597 testpmd> dump_devargs
602 Dumps the log level for all the dpdk modules::
604 testpmd> dump_log_types
606 show (raw_encap|raw_decap)
607 ~~~~~~~~~~~~~~~~~~~~~~~~~~
609 Display content of raw_encap/raw_decap buffers in hex::
611 testpmd> show <raw_encap|raw_decap> <index>
612 testpmd> show <raw_encap|raw_decap> all
616 testpmd> show raw_encap 6
618 index: 6 at [0x1c565b0], len=50
619 00000000: 00 00 00 00 00 00 16 26 36 46 56 66 08 00 45 00 | .......&6FVf..E.
620 00000010: 00 00 00 00 00 00 00 11 00 00 C0 A8 01 06 C0 A8 | ................
621 00000020: 03 06 00 00 00 FA 00 00 00 00 08 00 00 00 00 00 | ................
624 show fec capabilities
625 ~~~~~~~~~~~~~~~~~~~~~
627 Show fec capabilities of a port::
629 testpmd> show port (port_id) fec capabilities
634 Show fec mode of a port::
636 testpmd> show port (port_id) fec_mode
639 Configuration Functions
640 -----------------------
642 The testpmd application can be configured from the runtime as well as from the command-line.
644 This section details the available configuration functions that are available.
648 Configuration changes only become active when forwarding is started/restarted.
653 Reset forwarding to the default configuration::
660 Set the debug verbosity level::
662 testpmd> set verbose (level)
664 Available levels are as following:
666 * ``0`` silent except for error.
667 * ``1`` fully verbose except for Tx packets.
668 * ``2`` fully verbose except for Rx packets.
669 * ``> 2`` fully verbose.
674 Set the log level for a log type::
676 testpmd> set log global|(type) (level)
680 * ``type`` is the log name.
682 * ``level`` is the log level.
684 For example, to change the global log level::
686 testpmd> set log global (level)
688 Regexes can also be used for type. To change log level of user1, user2 and user3::
690 testpmd> set log user[1-3] (level)
695 Set the number of ports used by the application:
699 This is equivalent to the ``--nb-ports`` command-line option.
704 Set the number of cores used by the application::
706 testpmd> set nbcore (num)
708 This is equivalent to the ``--nb-cores`` command-line option.
712 The number of cores used must not be greater than number of ports used multiplied by the number of queues per port.
717 Set the forwarding cores hexadecimal mask::
719 testpmd> set coremask (mask)
721 This is equivalent to the ``--coremask`` command-line option.
725 The main lcore is reserved for command line parsing only and cannot be masked on for packet forwarding.
730 Set the forwarding ports hexadecimal mask::
732 testpmd> set portmask (mask)
734 This is equivalent to the ``--portmask`` command-line option.
736 set record-core-cycles
737 ~~~~~~~~~~~~~~~~~~~~~~
739 Set the recording of CPU cycles::
741 testpmd> set record-core-cycles (on|off)
745 * ``on`` enables measurement of CPU cycles per packet.
747 * ``off`` disables measurement of CPU cycles per packet.
749 This is equivalent to the ``--record-core-cycles command-line`` option.
751 set record-burst-stats
752 ~~~~~~~~~~~~~~~~~~~~~~
754 Set the displaying of RX and TX bursts::
756 testpmd> set record-burst-stats (on|off)
760 * ``on`` enables display of RX and TX bursts.
762 * ``off`` disables display of RX and TX bursts.
764 This is equivalent to the ``--record-burst-stats command-line`` option.
769 Set number of packets per burst::
771 testpmd> set burst (num)
773 This is equivalent to the ``--burst command-line`` option.
775 When retry is enabled, the transmit delay time and number of retries can also be set::
777 testpmd> set burst tx delay (microseconds) retry (num)
782 Set the offsets of segments relating to the data buffer beginning on receiving
783 if split feature is engaged. Affects only the queues configured with split
784 offloads (currently BUFFER_SPLIT is supported only).
786 testpmd> set rxoffs (x[,y]*)
788 Where x[,y]* represents a CSV list of values, without white space. If the list
789 of offsets is shorter than the list of segments the zero offsets will be used
790 for the remaining segments.
795 Set the length of segments to scatter packets on receiving if split
796 feature is engaged. Affects only the queues configured with split offloads
797 (currently BUFFER_SPLIT is supported only). Optionally the multiple memory
798 pools can be specified with --mbuf-size command line parameter and the mbufs
799 to receive will be allocated sequentially from these extra memory pools (the
800 mbuf for the first segment is allocated from the first pool, the second one
801 from the second pool, and so on, if segment number is greater then pool's the
802 mbuf for remaining segments will be allocated from the last valid pool).
804 testpmd> set rxpkts (x[,y]*)
806 Where x[,y]* represents a CSV list of values, without white space. Zero value
807 means to use the corresponding memory pool data buffer size.
812 Set the length of each segment of the TX-ONLY packets or length of packet for FLOWGEN mode::
814 testpmd> set txpkts (x[,y]*)
816 Where x[,y]* represents a CSV list of values, without white space.
821 Configure the timing burst pattern for Tx only mode. This command enables
822 the packet send scheduling on dynamic timestamp mbuf field and configures
823 timing pattern in Tx only mode. In this mode, if scheduling is enabled
824 application provides timestamps in the packets being sent. It is possible
825 to configure delay (in unspecified device clock units) between bursts
826 and between the packets within the burst::
828 testpmd> set txtimes (inter),(intra)
832 * ``inter`` is the delay between the bursts in the device clock units.
833 If ``intra`` is zero, this is the time between the beginnings of the
834 first packets in the neighbour bursts, if ``intra`` is not zero,
835 ``inter`` specifies the time between the beginning of the first packet
836 of the current burst and the beginning of the last packet of the
837 previous burst. If ``inter`` parameter is zero the send scheduling
838 on timestamps is disabled (default).
840 * ``intra`` is the delay between the packets within the burst specified
841 in the device clock units. The number of packets in the burst is defined
842 by regular burst setting. If ``intra`` parameter is zero no timestamps
843 provided in the packets excepting the first one in the burst.
845 As the result the bursts of packet will be transmitted with specific
846 delays between the packets within the burst and specific delay between
847 the bursts. The rte_eth_read_clock() must be supported by the device(s)
848 and is supposed to be engaged to get the current device clock value
849 and provide the reference for the timestamps. If there is no supported
850 rte_eth_read_clock() there will be no send scheduling provided on the port.
855 Set the split policy for the TX packets, applicable for TX-ONLY and CSUM forwarding modes::
857 testpmd> set txsplit (off|on|rand)
861 * ``off`` disable packet copy & split for CSUM mode.
863 * ``on`` split outgoing packet into multiple segments. Size of each segment
864 and number of segments per packet is determined by ``set txpkts`` command
867 * ``rand`` same as 'on', but number of segments per each packet is a random value between 1 and total number of segments.
872 Set the list of forwarding cores::
874 testpmd> set corelist (x[,y]*)
876 For example, to change the forwarding cores:
878 .. code-block:: console
880 testpmd> set corelist 3,1
881 testpmd> show config fwd
883 io packet forwarding - ports=2 - cores=2 - streams=2 - NUMA support disabled
884 Logical Core 3 (socket 0) forwards packets on 1 streams:
885 RX P=0/Q=0 (socket 0) -> TX P=1/Q=0 (socket 0) peer=02:00:00:00:00:01
886 Logical Core 1 (socket 0) forwards packets on 1 streams:
887 RX P=1/Q=0 (socket 0) -> TX P=0/Q=0 (socket 0) peer=02:00:00:00:00:00
891 The cores are used in the same order as specified on the command line.
896 Set the list of forwarding ports::
898 testpmd> set portlist (x[,y]*)
900 For example, to change the port forwarding:
902 .. code-block:: console
904 testpmd> set portlist 0,2,1,3
905 testpmd> show config fwd
907 io packet forwarding - ports=4 - cores=1 - streams=4
908 Logical Core 3 (socket 0) forwards packets on 4 streams:
909 RX P=0/Q=0 (socket 0) -> TX P=2/Q=0 (socket 0) peer=02:00:00:00:00:01
910 RX P=2/Q=0 (socket 0) -> TX P=0/Q=0 (socket 0) peer=02:00:00:00:00:00
911 RX P=1/Q=0 (socket 0) -> TX P=3/Q=0 (socket 0) peer=02:00:00:00:00:03
912 RX P=3/Q=0 (socket 0) -> TX P=1/Q=0 (socket 0) peer=02:00:00:00:00:02
917 Select how to retrieve new ports created after "port attach" command::
919 testpmd> set port setup on (iterator|event)
921 For each new port, a setup is done.
922 It will find the probed ports via RTE_ETH_FOREACH_MATCHING_DEV loop
923 in iterator mode, or via RTE_ETH_EVENT_NEW in event mode.
928 Enable/disable tx loopback::
930 testpmd> set tx loopback (port_id) (on|off)
935 set drop enable bit for all queues::
937 testpmd> set all queues drop (port_id) (on|off)
939 set split drop enable (for VF)
940 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
942 set split drop enable bit for VF from PF::
944 testpmd> set vf split drop (port_id) (vf_id) (on|off)
946 set mac antispoof (for VF)
947 ~~~~~~~~~~~~~~~~~~~~~~~~~~
949 Set mac antispoof for a VF from the PF::
951 testpmd> set vf mac antispoof (port_id) (vf_id) (on|off)
956 Enable/disable MACsec offload::
958 testpmd> set macsec offload (port_id) on encrypt (on|off) replay-protect (on|off)
959 testpmd> set macsec offload (port_id) off
964 Configure MACsec secure connection (SC)::
966 testpmd> set macsec sc (tx|rx) (port_id) (mac) (pi)
970 The pi argument is ignored for tx.
971 Check the NIC Datasheet for hardware limits.
976 Configure MACsec secure association (SA)::
978 testpmd> set macsec sa (tx|rx) (port_id) (idx) (an) (pn) (key)
982 The IDX value must be 0 or 1.
983 Check the NIC Datasheet for hardware limits.
985 set broadcast mode (for VF)
986 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
988 Set broadcast mode for a VF from the PF::
990 testpmd> set vf broadcast (port_id) (vf_id) (on|off)
995 Set the VLAN strip for a queue on a port::
997 testpmd> vlan set stripq (on|off) (port_id,queue_id)
999 vlan set stripq (for VF)
1000 ~~~~~~~~~~~~~~~~~~~~~~~~
1002 Set VLAN strip for all queues in a pool for a VF from the PF::
1004 testpmd> set vf vlan stripq (port_id) (vf_id) (on|off)
1006 vlan set insert (for VF)
1007 ~~~~~~~~~~~~~~~~~~~~~~~~
1009 Set VLAN insert for a VF from the PF::
1011 testpmd> set vf vlan insert (port_id) (vf_id) (vlan_id)
1013 vlan set tag (for VF)
1014 ~~~~~~~~~~~~~~~~~~~~~
1016 Set VLAN tag for a VF from the PF::
1018 testpmd> set vf vlan tag (port_id) (vf_id) (on|off)
1020 vlan set antispoof (for VF)
1021 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
1023 Set VLAN antispoof for a VF from the PF::
1025 testpmd> set vf vlan antispoof (port_id) (vf_id) (on|off)
1027 vlan set (strip|filter|qinq_strip|extend)
1028 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1029 Set the VLAN strip/filter/QinQ strip/extend on for a port::
1031 testpmd> vlan set (strip|filter|qinq_strip|extend) (on|off) (port_id)
1036 Set the inner or outer VLAN TPID for packet filtering on a port::
1038 testpmd> vlan set (inner|outer) tpid (value) (port_id)
1042 TPID value must be a 16-bit number (value <= 65536).
1047 Add a VLAN ID, or all identifiers, to the set of VLAN identifiers filtered by port ID::
1049 testpmd> rx_vlan add (vlan_id|all) (port_id)
1053 VLAN filter must be set on that port. VLAN ID < 4096.
1054 Depending on the NIC used, number of vlan_ids may be limited to the maximum entries
1055 in VFTA table. This is important if enabling all vlan_ids.
1060 Remove a VLAN ID, or all identifiers, from the set of VLAN identifiers filtered by port ID::
1062 testpmd> rx_vlan rm (vlan_id|all) (port_id)
1064 rx_vlan add (for VF)
1065 ~~~~~~~~~~~~~~~~~~~~
1067 Add a VLAN ID, to the set of VLAN identifiers filtered for VF(s) for port ID::
1069 testpmd> rx_vlan add (vlan_id) port (port_id) vf (vf_mask)
1074 Remove a VLAN ID, from the set of VLAN identifiers filtered for VF(s) for port ID::
1076 testpmd> rx_vlan rm (vlan_id) port (port_id) vf (vf_mask)
1081 Add an UDP port for VXLAN packet filter on a port::
1083 testpmd> rx_vxlan_port add (udp_port) (port_id)
1085 rx_vxlan_port remove
1086 ~~~~~~~~~~~~~~~~~~~~
1088 Remove an UDP port for VXLAN packet filter on a port::
1090 testpmd> rx_vxlan_port rm (udp_port) (port_id)
1095 Set hardware insertion of VLAN IDs in packets sent on a port::
1097 testpmd> tx_vlan set (port_id) vlan_id[, vlan_id_outer]
1099 For example, set a single VLAN ID (5) insertion on port 0::
1103 Or, set double VLAN ID (inner: 2, outer: 3) insertion on port 1::
1111 Set port based hardware insertion of VLAN ID in packets sent on a port::
1113 testpmd> tx_vlan set pvid (port_id) (vlan_id) (on|off)
1118 Disable hardware insertion of a VLAN header in packets sent on a port::
1120 testpmd> tx_vlan reset (port_id)
1125 Select hardware or software calculation of the checksum when
1126 transmitting a packet using the ``csum`` forwarding engine::
1128 testpmd> csum set (ip|udp|tcp|sctp|outer-ip|outer-udp) (hw|sw) (port_id)
1132 * ``ip|udp|tcp|sctp`` always relate to the inner layer.
1134 * ``outer-ip`` relates to the outer IP layer (only for IPv4) in the case where the packet is recognized
1135 as a tunnel packet by the forwarding engine (geneve, gre, gtp, ipip, vxlan and vxlan-gpe are
1136 supported). See also the ``csum parse-tunnel`` command.
1138 * ``outer-udp`` relates to the outer UDP layer in the case where the packet is recognized
1139 as a tunnel packet by the forwarding engine (geneve, gtp, vxlan and vxlan-gpe are
1140 supported). See also the ``csum parse-tunnel`` command.
1144 Check the NIC Datasheet for hardware limits.
1149 Set RSS queue region span on a port::
1151 testpmd> set port (port_id) queue-region region_id (value) \
1152 queue_start_index (value) queue_num (value)
1154 Set flowtype mapping on a RSS queue region on a port::
1156 testpmd> set port (port_id) queue-region region_id (value) flowtype (value)
1160 * For the flowtype(pctype) of packet,the specific index for each type has
1161 been defined in file i40e_type.h as enum i40e_filter_pctype.
1163 Set user priority mapping on a RSS queue region on a port::
1165 testpmd> set port (port_id) queue-region UP (value) region_id (value)
1167 Flush all queue region related configuration on a port::
1169 testpmd> set port (port_id) queue-region flush (on|off)
1173 * ``on``: is just an enable function which server for other configuration,
1174 it is for all configuration about queue region from up layer,
1175 at first will only keep in DPDK software stored in driver,
1176 only after "flush on", it commit all configuration to HW.
1178 * ``"off``: is just clean all configuration about queue region just now,
1179 and restore all to DPDK i40e driver default config when start up.
1181 Show all queue region related configuration info on a port::
1183 testpmd> show port (port_id) queue-region
1187 Queue region only support on PF by now, so these command is
1188 only for configuration of queue region on PF port.
1193 Define how tunneled packets should be handled by the csum forward
1196 testpmd> csum parse-tunnel (on|off) (tx_port_id)
1198 If enabled, the csum forward engine will try to recognize supported
1199 tunnel headers (geneve, gtp, gre, ipip, vxlan, vxlan-gpe).
1201 If disabled, treat tunnel packets as non-tunneled packets (a inner
1202 header is handled as a packet payload).
1206 The port argument is the TX port like in the ``csum set`` command.
1210 Consider a packet in packet like the following::
1212 eth_out/ipv4_out/udp_out/vxlan/eth_in/ipv4_in/tcp_in
1214 * If parse-tunnel is enabled, the ``ip|udp|tcp|sctp`` parameters of ``csum set``
1215 command relate to the inner headers (here ``ipv4_in`` and ``tcp_in``), and the
1216 ``outer-ip|outer-udp`` parameter relates to the outer headers (here ``ipv4_out`` and ``udp_out``).
1218 * If parse-tunnel is disabled, the ``ip|udp|tcp|sctp`` parameters of ``csum set``
1219 command relate to the outer headers, here ``ipv4_out`` and ``udp_out``.
1224 Display tx checksum offload configuration::
1226 testpmd> csum show (port_id)
1231 Enable TCP Segmentation Offload (TSO) in the ``csum`` forwarding engine::
1233 testpmd> tso set (segsize) (port_id)
1237 Check the NIC datasheet for hardware limits.
1242 Display the status of TCP Segmentation Offload::
1244 testpmd> tso show (port_id)
1249 Set tso segment size of tunneled packets for a port in csum engine::
1251 testpmd> tunnel_tso set (tso_segsz) (port_id)
1256 Display the status of tunneled TCP Segmentation Offload for a port::
1258 testpmd> tunnel_tso show (port_id)
1263 Enable or disable GRO in ``csum`` forwarding engine::
1265 testpmd> set port <port_id> gro on|off
1267 If enabled, the csum forwarding engine will perform GRO on the TCP/IPv4
1268 packets received from the given port.
1270 If disabled, packets received from the given port won't be performed
1271 GRO. By default, GRO is disabled for all ports.
1275 When enable GRO for a port, TCP/IPv4 packets received from the port
1276 will be performed GRO. After GRO, all merged packets have bad
1277 checksums, since the GRO library doesn't re-calculate checksums for
1278 the merged packets. Therefore, if users want the merged packets to
1279 have correct checksums, please select HW IP checksum calculation and
1280 HW TCP checksum calculation for the port which the merged packets are
1286 Display GRO configuration for a given port::
1288 testpmd> show port <port_id> gro
1293 Set the cycle to flush the GROed packets from reassembly tables::
1295 testpmd> set gro flush <cycles>
1297 When enable GRO, the csum forwarding engine performs GRO on received
1298 packets, and the GROed packets are stored in reassembly tables. Users
1299 can use this command to determine when the GROed packets are flushed
1300 from the reassembly tables.
1302 The ``cycles`` is measured in GRO operation times. The csum forwarding
1303 engine flushes the GROed packets from the tables every ``cycles`` GRO
1306 By default, the value of ``cycles`` is 1, which means flush GROed packets
1307 from the reassembly tables as soon as one GRO operation finishes. The value
1308 of ``cycles`` should be in the range of 1 to ``GRO_MAX_FLUSH_CYCLES``.
1310 Please note that the large value of ``cycles`` may cause the poor TCP/IP
1311 stack performance. Because the GROed packets are delayed to arrive the
1312 stack, thus causing more duplicated ACKs and TCP retransmissions.
1317 Toggle per-port GSO support in ``csum`` forwarding engine::
1319 testpmd> set port <port_id> gso on|off
1321 If enabled, the csum forwarding engine will perform GSO on supported IPv4
1322 packets, transmitted on the given port.
1324 If disabled, packets transmitted on the given port will not undergo GSO.
1325 By default, GSO is disabled for all ports.
1329 When GSO is enabled on a port, supported IPv4 packets transmitted on that
1330 port undergo GSO. Afterwards, the segmented packets are represented by
1331 multi-segment mbufs; however, the csum forwarding engine doesn't calculation
1332 of checksums for GSO'd segments in SW. As a result, if users want correct
1333 checksums in GSO segments, they should enable HW checksum calculation for
1336 For example, HW checksum calculation for VxLAN GSO'd packets may be enabled
1337 by setting the following options in the csum forwarding engine:
1339 testpmd> csum set outer_ip hw <port_id>
1341 testpmd> csum set ip hw <port_id>
1343 testpmd> csum set tcp hw <port_id>
1345 UDP GSO is the same as IP fragmentation, which treats the UDP header
1346 as the payload and does not modify it during segmentation. That is,
1347 after UDP GSO, only the first output fragment has the original UDP
1348 header. Therefore, users need to enable HW IP checksum calculation
1349 and SW UDP checksum calculation for GSO-enabled ports, if they want
1350 correct checksums for UDP/IPv4 packets.
1355 Set the maximum GSO segment size (measured in bytes), which includes the
1356 packet header and the packet payload for GSO-enabled ports (global)::
1358 testpmd> set gso segsz <length>
1363 Display the status of Generic Segmentation Offload for a given port::
1365 testpmd> show port <port_id> gso
1370 Add an alternative MAC address to a port::
1372 testpmd> mac_addr add (port_id) (XX:XX:XX:XX:XX:XX)
1377 Remove a MAC address from a port::
1379 testpmd> mac_addr remove (port_id) (XX:XX:XX:XX:XX:XX)
1384 To add the multicast MAC address to/from the set of multicast addresses
1387 testpmd> mcast_addr add (port_id) (mcast_addr)
1392 To remove the multicast MAC address to/from the set of multicast addresses
1395 testpmd> mcast_addr remove (port_id) (mcast_addr)
1397 mac_addr add (for VF)
1398 ~~~~~~~~~~~~~~~~~~~~~
1400 Add an alternative MAC address for a VF to a port::
1402 testpmd> mac_add add port (port_id) vf (vf_id) (XX:XX:XX:XX:XX:XX)
1407 Set the default MAC address for a port::
1409 testpmd> mac_addr set (port_id) (XX:XX:XX:XX:XX:XX)
1411 mac_addr set (for VF)
1412 ~~~~~~~~~~~~~~~~~~~~~
1414 Set the MAC address for a VF from the PF::
1416 testpmd> set vf mac addr (port_id) (vf_id) (XX:XX:XX:XX:XX:XX)
1421 Set the forwarding peer address for certain port::
1423 testpmd> set eth-peer (port_id) (peer_addr)
1425 This is equivalent to the ``--eth-peer`` command-line option.
1430 Set the unicast hash filter(s) on/off for a port::
1432 testpmd> set port (port_id) uta (XX:XX:XX:XX:XX:XX|all) (on|off)
1437 Set the promiscuous mode on for a port or for all ports.
1438 In promiscuous mode packets are not dropped if they aren't for the specified MAC address::
1440 testpmd> set promisc (port_id|all) (on|off)
1445 Set the allmulti mode for a port or for all ports::
1447 testpmd> set allmulti (port_id|all) (on|off)
1449 Same as the ifconfig (8) option. Controls how multicast packets are handled.
1451 set promisc (for VF)
1452 ~~~~~~~~~~~~~~~~~~~~
1454 Set the unicast promiscuous mode for a VF from PF.
1455 It's supported by Intel i40e NICs now.
1456 In promiscuous mode packets are not dropped if they aren't for the specified MAC address::
1458 testpmd> set vf promisc (port_id) (vf_id) (on|off)
1460 set allmulticast (for VF)
1461 ~~~~~~~~~~~~~~~~~~~~~~~~~
1463 Set the multicast promiscuous mode for a VF from PF.
1464 It's supported by Intel i40e NICs now.
1465 In promiscuous mode packets are not dropped if they aren't for the specified MAC address::
1467 testpmd> set vf allmulti (port_id) (vf_id) (on|off)
1469 set tx max bandwidth (for VF)
1470 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1472 Set TX max absolute bandwidth (Mbps) for a VF from PF::
1474 testpmd> set vf tx max-bandwidth (port_id) (vf_id) (max_bandwidth)
1476 set tc tx min bandwidth (for VF)
1477 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1479 Set all TCs' TX min relative bandwidth (%) for a VF from PF::
1481 testpmd> set vf tc tx min-bandwidth (port_id) (vf_id) (bw1, bw2, ...)
1483 set tc tx max bandwidth (for VF)
1484 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1486 Set a TC's TX max absolute bandwidth (Mbps) for a VF from PF::
1488 testpmd> set vf tc tx max-bandwidth (port_id) (vf_id) (tc_no) (max_bandwidth)
1490 set tc strict link priority mode
1491 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1493 Set some TCs' strict link priority mode on a physical port::
1495 testpmd> set tx strict-link-priority (port_id) (tc_bitmap)
1497 set tc tx min bandwidth
1498 ~~~~~~~~~~~~~~~~~~~~~~~
1500 Set all TCs' TX min relative bandwidth (%) globally for all PF and VFs::
1502 testpmd> set tc tx min-bandwidth (port_id) (bw1, bw2, ...)
1507 Set the link flow control parameter on a port::
1509 testpmd> set flow_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \
1510 (pause_time) (send_xon) mac_ctrl_frame_fwd (on|off) \
1511 autoneg (on|off) (port_id)
1515 * ``high_water`` (integer): High threshold value to trigger XOFF.
1517 * ``low_water`` (integer): Low threshold value to trigger XON.
1519 * ``pause_time`` (integer): Pause quota in the Pause frame.
1521 * ``send_xon`` (0/1): Send XON frame.
1523 * ``mac_ctrl_frame_fwd``: Enable receiving MAC control frames.
1525 * ``autoneg``: Change the auto-negotiation parameter.
1530 Set the priority flow control parameter on a port::
1532 testpmd> set pfc_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \
1533 (pause_time) (priority) (port_id)
1537 * ``high_water`` (integer): High threshold value.
1539 * ``low_water`` (integer): Low threshold value.
1541 * ``pause_time`` (integer): Pause quota in the Pause frame.
1543 * ``priority`` (0-7): VLAN User Priority.
1548 Set statistics mapping (qmapping 0..15) for RX/TX queue on port::
1550 testpmd> set stat_qmap (tx|rx) (port_id) (queue_id) (qmapping)
1552 For example, to set rx queue 2 on port 0 to mapping 5::
1554 testpmd>set stat_qmap rx 0 2 5
1556 set xstats-hide-zero
1557 ~~~~~~~~~~~~~~~~~~~~
1559 Set the option to hide zero values for xstats display::
1561 testpmd> set xstats-hide-zero on|off
1565 By default, the zero values are displayed for xstats.
1567 set port - rx/tx (for VF)
1568 ~~~~~~~~~~~~~~~~~~~~~~~~~
1570 Set VF receive/transmit from a port::
1572 testpmd> set port (port_id) vf (vf_id) (rx|tx) (on|off)
1574 set port - rx mode(for VF)
1575 ~~~~~~~~~~~~~~~~~~~~~~~~~~
1577 Set the VF receive mode of a port::
1579 testpmd> set port (port_id) vf (vf_id) \
1580 rxmode (AUPE|ROPE|BAM|MPE) (on|off)
1582 The available receive modes are:
1584 * ``AUPE``: Accepts untagged VLAN.
1586 * ``ROPE``: Accepts unicast hash.
1588 * ``BAM``: Accepts broadcast packets.
1590 * ``MPE``: Accepts all multicast packets.
1592 set port - tx_rate (for Queue)
1593 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1595 Set TX rate limitation for a queue on a port::
1597 testpmd> set port (port_id) queue (queue_id) rate (rate_value)
1599 set port - tx_rate (for VF)
1600 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
1602 Set TX rate limitation for queues in VF on a port::
1604 testpmd> set port (port_id) vf (vf_id) rate (rate_value) queue_mask (queue_mask)
1606 set port - mirror rule
1607 ~~~~~~~~~~~~~~~~~~~~~~
1609 Set pool or vlan type mirror rule for a port::
1611 testpmd> set port (port_id) mirror-rule (rule_id) \
1612 (pool-mirror-up|pool-mirror-down|vlan-mirror) \
1613 (poolmask|vlanid[,vlanid]*) dst-pool (pool_id) (on|off)
1615 Set link mirror rule for a port::
1617 testpmd> set port (port_id) mirror-rule (rule_id) \
1618 (uplink-mirror|downlink-mirror) dst-pool (pool_id) (on|off)
1620 For example to enable mirror traffic with vlan 0,1 to pool 0::
1622 set port 0 mirror-rule 0 vlan-mirror 0,1 dst-pool 0 on
1624 reset port - mirror rule
1625 ~~~~~~~~~~~~~~~~~~~~~~~~
1627 Reset a mirror rule for a port::
1629 testpmd> reset port (port_id) mirror-rule (rule_id)
1634 Set the flush on RX streams before forwarding.
1635 The default is flush ``on``.
1636 Mainly used with PCAP drivers to turn off the default behavior of flushing the first 512 packets on RX streams::
1638 testpmd> set flush_rx off
1643 Set the bypass mode for the lowest port on bypass enabled NIC::
1645 testpmd> set bypass mode (normal|bypass|isolate) (port_id)
1650 Set the event required to initiate specified bypass mode for the lowest port on a bypass enabled::
1652 testpmd> set bypass event (timeout|os_on|os_off|power_on|power_off) \
1653 mode (normal|bypass|isolate) (port_id)
1657 * ``timeout``: Enable bypass after watchdog timeout.
1659 * ``os_on``: Enable bypass when OS/board is powered on.
1661 * ``os_off``: Enable bypass when OS/board is powered off.
1663 * ``power_on``: Enable bypass when power supply is turned on.
1665 * ``power_off``: Enable bypass when power supply is turned off.
1671 Set the bypass watchdog timeout to ``n`` seconds where 0 = instant::
1673 testpmd> set bypass timeout (0|1.5|2|3|4|8|16|32)
1678 Show the bypass configuration for a bypass enabled NIC using the lowest port on the NIC::
1680 testpmd> show bypass config (port_id)
1685 Set link up for a port::
1687 testpmd> set link-up port (port id)
1692 Set link down for a port::
1694 testpmd> set link-down port (port id)
1699 Enable E-tag insertion for a VF on a port::
1701 testpmd> E-tag set insertion on port-tag-id (value) port (port_id) vf (vf_id)
1703 Disable E-tag insertion for a VF on a port::
1705 testpmd> E-tag set insertion off port (port_id) vf (vf_id)
1707 Enable/disable E-tag stripping on a port::
1709 testpmd> E-tag set stripping (on|off) port (port_id)
1711 Enable/disable E-tag based forwarding on a port::
1713 testpmd> E-tag set forwarding (on|off) port (port_id)
1718 Load a dynamic device personalization (DDP) profile and store backup profile::
1720 testpmd> ddp add (port_id) (profile_path[,backup_profile_path])
1725 Delete a dynamic device personalization profile and restore backup profile::
1727 testpmd> ddp del (port_id) (backup_profile_path)
1732 List all items from the ptype mapping table::
1734 testpmd> ptype mapping get (port_id) (valid_only)
1738 * ``valid_only``: A flag indicates if only list valid items(=1) or all itemss(=0).
1740 Replace a specific or a group of software defined ptype with a new one::
1742 testpmd> ptype mapping replace (port_id) (target) (mask) (pkt_type)
1746 * ``target``: A specific software ptype or a mask to represent a group of software ptypes.
1748 * ``mask``: A flag indicate if "target" is a specific software ptype(=0) or a ptype mask(=1).
1750 * ``pkt_type``: The new software ptype to replace the old ones.
1752 Update hardware defined ptype to software defined packet type mapping table::
1754 testpmd> ptype mapping update (port_id) (hw_ptype) (sw_ptype)
1758 * ``hw_ptype``: hardware ptype as the index of the ptype mapping table.
1760 * ``sw_ptype``: software ptype as the value of the ptype mapping table.
1762 Reset ptype mapping table::
1764 testpmd> ptype mapping reset (port_id)
1766 config per port Rx offloading
1767 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1769 Enable or disable a per port Rx offloading on all Rx queues of a port::
1771 testpmd> port config (port_id) rx_offload (offloading) on|off
1773 * ``offloading``: can be any of these offloading capability:
1774 vlan_strip, ipv4_cksum, udp_cksum, tcp_cksum, tcp_lro,
1775 qinq_strip, outer_ipv4_cksum, macsec_strip,
1776 header_split, vlan_filter, vlan_extend, jumbo_frame,
1777 scatter, timestamp, security, keep_crc, rss_hash
1779 This command should be run when the port is stopped, or else it will fail.
1781 config per queue Rx offloading
1782 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1784 Enable or disable a per queue Rx offloading only on a specific Rx queue::
1786 testpmd> port (port_id) rxq (queue_id) rx_offload (offloading) on|off
1788 * ``offloading``: can be any of these offloading capability:
1789 vlan_strip, ipv4_cksum, udp_cksum, tcp_cksum, tcp_lro,
1790 qinq_strip, outer_ipv4_cksum, macsec_strip,
1791 header_split, vlan_filter, vlan_extend, jumbo_frame,
1792 scatter, timestamp, security, keep_crc
1794 This command should be run when the port is stopped, or else it will fail.
1796 config per port Tx offloading
1797 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1799 Enable or disable a per port Tx offloading on all Tx queues of a port::
1801 testpmd> port config (port_id) tx_offload (offloading) on|off
1803 * ``offloading``: can be any of these offloading capability:
1804 vlan_insert, ipv4_cksum, udp_cksum, tcp_cksum,
1805 sctp_cksum, tcp_tso, udp_tso, outer_ipv4_cksum,
1806 qinq_insert, vxlan_tnl_tso, gre_tnl_tso,
1807 ipip_tnl_tso, geneve_tnl_tso, macsec_insert,
1808 mt_lockfree, multi_segs, mbuf_fast_free, security
1810 This command should be run when the port is stopped, or else it will fail.
1812 config per queue Tx offloading
1813 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1815 Enable or disable a per queue Tx offloading only on a specific Tx queue::
1817 testpmd> port (port_id) txq (queue_id) tx_offload (offloading) on|off
1819 * ``offloading``: can be any of these offloading capability:
1820 vlan_insert, ipv4_cksum, udp_cksum, tcp_cksum,
1821 sctp_cksum, tcp_tso, udp_tso, outer_ipv4_cksum,
1822 qinq_insert, vxlan_tnl_tso, gre_tnl_tso,
1823 ipip_tnl_tso, geneve_tnl_tso, macsec_insert,
1824 mt_lockfree, multi_segs, mbuf_fast_free, security
1826 This command should be run when the port is stopped, or else it will fail.
1828 Config VXLAN Encap outer layers
1829 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1831 Configure the outer layer to encapsulate a packet inside a VXLAN tunnel::
1833 set vxlan ip-version (ipv4|ipv6) vni (vni) udp-src (udp-src) \
1834 udp-dst (udp-dst) ip-src (ip-src) ip-dst (ip-dst) eth-src (eth-src) \
1837 set vxlan-with-vlan ip-version (ipv4|ipv6) vni (vni) udp-src (udp-src) \
1838 udp-dst (udp-dst) ip-src (ip-src) ip-dst (ip-dst) vlan-tci (vlan-tci) \
1839 eth-src (eth-src) eth-dst (eth-dst)
1841 set vxlan-tos-ttl ip-version (ipv4|ipv6) vni (vni) udp-src (udp-src) \
1842 udp-dst (udp-dst) ip-tos (ip-tos) ip-ttl (ip-ttl) ip-src (ip-src) \
1843 ip-dst (ip-dst) eth-src (eth-src) eth-dst (eth-dst)
1845 These commands will set an internal configuration inside testpmd, any following
1846 flow rule using the action vxlan_encap will use the last configuration set.
1847 To have a different encapsulation header, one of those commands must be called
1848 before the flow rule creation.
1850 Config NVGRE Encap outer layers
1851 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1853 Configure the outer layer to encapsulate a packet inside a NVGRE tunnel::
1855 set nvgre ip-version (ipv4|ipv6) tni (tni) ip-src (ip-src) ip-dst (ip-dst) \
1856 eth-src (eth-src) eth-dst (eth-dst)
1857 set nvgre-with-vlan ip-version (ipv4|ipv6) tni (tni) ip-src (ip-src) \
1858 ip-dst (ip-dst) vlan-tci (vlan-tci) eth-src (eth-src) eth-dst (eth-dst)
1860 These commands will set an internal configuration inside testpmd, any following
1861 flow rule using the action nvgre_encap will use the last configuration set.
1862 To have a different encapsulation header, one of those commands must be called
1863 before the flow rule creation.
1868 Configure the l2 to be used when encapsulating a packet with L2::
1870 set l2_encap ip-version (ipv4|ipv6) eth-src (eth-src) eth-dst (eth-dst)
1871 set l2_encap-with-vlan ip-version (ipv4|ipv6) vlan-tci (vlan-tci) \
1872 eth-src (eth-src) eth-dst (eth-dst)
1874 Those commands will set an internal configuration inside testpmd, any following
1875 flow rule using the action l2_encap will use the last configuration set.
1876 To have a different encapsulation header, one of those commands must be called
1877 before the flow rule creation.
1882 Configure the l2 to be removed when decapsulating a packet with L2::
1884 set l2_decap ip-version (ipv4|ipv6)
1885 set l2_decap-with-vlan ip-version (ipv4|ipv6)
1887 Those commands will set an internal configuration inside testpmd, any following
1888 flow rule using the action l2_decap will use the last configuration set.
1889 To have a different encapsulation header, one of those commands must be called
1890 before the flow rule creation.
1892 Config MPLSoGRE Encap outer layers
1893 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1895 Configure the outer layer to encapsulate a packet inside a MPLSoGRE tunnel::
1897 set mplsogre_encap ip-version (ipv4|ipv6) label (label) \
1898 ip-src (ip-src) ip-dst (ip-dst) eth-src (eth-src) eth-dst (eth-dst)
1899 set mplsogre_encap-with-vlan ip-version (ipv4|ipv6) label (label) \
1900 ip-src (ip-src) ip-dst (ip-dst) vlan-tci (vlan-tci) \
1901 eth-src (eth-src) eth-dst (eth-dst)
1903 These commands will set an internal configuration inside testpmd, any following
1904 flow rule using the action mplsogre_encap will use the last configuration set.
1905 To have a different encapsulation header, one of those commands must be called
1906 before the flow rule creation.
1908 Config MPLSoGRE Decap outer layers
1909 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1911 Configure the outer layer to decapsulate MPLSoGRE packet::
1913 set mplsogre_decap ip-version (ipv4|ipv6)
1914 set mplsogre_decap-with-vlan ip-version (ipv4|ipv6)
1916 These commands will set an internal configuration inside testpmd, any following
1917 flow rule using the action mplsogre_decap will use the last configuration set.
1918 To have a different decapsulation header, one of those commands must be called
1919 before the flow rule creation.
1921 Config MPLSoUDP Encap outer layers
1922 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1924 Configure the outer layer to encapsulate a packet inside a MPLSoUDP tunnel::
1926 set mplsoudp_encap ip-version (ipv4|ipv6) label (label) udp-src (udp-src) \
1927 udp-dst (udp-dst) ip-src (ip-src) ip-dst (ip-dst) \
1928 eth-src (eth-src) eth-dst (eth-dst)
1929 set mplsoudp_encap-with-vlan ip-version (ipv4|ipv6) label (label) \
1930 udp-src (udp-src) udp-dst (udp-dst) ip-src (ip-src) ip-dst (ip-dst) \
1931 vlan-tci (vlan-tci) eth-src (eth-src) eth-dst (eth-dst)
1933 These commands will set an internal configuration inside testpmd, any following
1934 flow rule using the action mplsoudp_encap will use the last configuration set.
1935 To have a different encapsulation header, one of those commands must be called
1936 before the flow rule creation.
1938 Config MPLSoUDP Decap outer layers
1939 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1941 Configure the outer layer to decapsulate MPLSoUDP packet::
1943 set mplsoudp_decap ip-version (ipv4|ipv6)
1944 set mplsoudp_decap-with-vlan ip-version (ipv4|ipv6)
1946 These commands will set an internal configuration inside testpmd, any following
1947 flow rule using the action mplsoudp_decap will use the last configuration set.
1948 To have a different decapsulation header, one of those commands must be called
1949 before the flow rule creation.
1951 Config Raw Encapsulation
1952 ~~~~~~~~~~~~~~~~~~~~~~~~~
1954 Configure the raw data to be used when encapsulating a packet by
1955 rte_flow_action_raw_encap::
1957 set raw_encap {index} {item} [/ {item} [...]] / end_set
1959 There are multiple global buffers for ``raw_encap``, this command will set one
1960 internal buffer index by ``{index}``.
1961 If there is no ``{index}`` specified::
1963 set raw_encap {item} [/ {item} [...]] / end_set
1965 the default index ``0`` is used.
1966 In order to use different encapsulating header, ``index`` must be specified
1967 during the flow rule creation::
1969 testpmd> flow create 0 egress pattern eth / ipv4 / end actions
1970 raw_encap index 2 / end
1972 Otherwise the default index ``0`` is used.
1974 Config Raw Decapsulation
1975 ~~~~~~~~~~~~~~~~~~~~~~~~
1977 Configure the raw data to be used when decapsulating a packet by
1978 rte_flow_action_raw_decap::
1980 set raw_decap {index} {item} [/ {item} [...]] / end_set
1982 There are multiple global buffers for ``raw_decap``, this command will set
1983 one internal buffer index by ``{index}``.
1984 If there is no ``{index}`` specified::
1986 set raw_decap {item} [/ {item} [...]] / end_set
1988 the default index ``0`` is used.
1989 In order to use different decapsulating header, ``index`` must be specified
1990 during the flow rule creation::
1992 testpmd> flow create 0 egress pattern eth / ipv4 / end actions
1993 raw_encap index 3 / end
1995 Otherwise the default index ``0`` is used.
2000 Set fec mode for a specific port::
2002 testpmd> set port (port_id) fec_mode auto|off|rs|baser
2004 Config Sample actions list
2005 ~~~~~~~~~~~~~~~~~~~~~~~~~~
2007 Configure the sample actions list to be used when sampling a packet by
2008 rte_flow_action_sample::
2010 set sample_actions {index} {action} [/ {action} [...]] / end
2012 There are multiple global buffers for ``sample_actions``, this command will set
2013 one internal buffer index by ``{index}``.
2015 In order to use different sample actions list, ``index`` must be specified
2016 during the flow rule creation::
2018 testpmd> flow create 0 ingress pattern eth / ipv4 / end actions
2019 sample ratio 2 index 2 / end
2021 Otherwise the default index ``0`` is used.
2026 The following sections show functions for configuring ports.
2030 Port configuration changes only become active when forwarding is started/restarted.
2035 Attach a port specified by pci address or virtual device args::
2037 testpmd> port attach (identifier)
2039 To attach a new pci device, the device should be recognized by kernel first.
2040 Then it should be moved under DPDK management.
2041 Finally the port can be attached to testpmd.
2043 For example, to move a pci device using ixgbe under DPDK management:
2045 .. code-block:: console
2047 # Check the status of the available devices.
2048 ./usertools/dpdk-devbind.py --status
2050 Network devices using DPDK-compatible driver
2051 ============================================
2054 Network devices using kernel driver
2055 ===================================
2056 0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused=
2059 # Bind the device to igb_uio.
2060 sudo ./usertools/dpdk-devbind.py -b igb_uio 0000:0a:00.0
2063 # Recheck the status of the devices.
2064 ./usertools/dpdk-devbind.py --status
2065 Network devices using DPDK-compatible driver
2066 ============================================
2067 0000:0a:00.0 '82599ES 10-Gigabit' drv=igb_uio unused=
2069 To attach a port created by virtual device, above steps are not needed.
2071 For example, to attach a port whose pci address is 0000:0a:00.0.
2073 .. code-block:: console
2075 testpmd> port attach 0000:0a:00.0
2076 Attaching a new port...
2077 EAL: PCI device 0000:0a:00.0 on NUMA socket -1
2078 EAL: probe driver: 8086:10fb rte_ixgbe_pmd
2079 EAL: PCI memory mapped at 0x7f83bfa00000
2080 EAL: PCI memory mapped at 0x7f83bfa80000
2081 PMD: eth_ixgbe_dev_init(): MAC: 2, PHY: 18, SFP+: 5
2082 PMD: eth_ixgbe_dev_init(): port 0 vendorID=0x8086 deviceID=0x10fb
2083 Port 0 is attached. Now total ports is 1
2086 For example, to attach a port created by pcap PMD.
2088 .. code-block:: console
2090 testpmd> port attach net_pcap0
2091 Attaching a new port...
2092 PMD: Initializing pmd_pcap for net_pcap0
2093 PMD: Creating pcap-backed ethdev on numa socket 0
2094 Port 0 is attached. Now total ports is 1
2097 In this case, identifier is ``net_pcap0``.
2098 This identifier format is the same as ``--vdev`` format of DPDK applications.
2100 For example, to re-attach a bonded port which has been previously detached,
2101 the mode and slave parameters must be given.
2103 .. code-block:: console
2105 testpmd> port attach net_bond_0,mode=0,slave=1
2106 Attaching a new port...
2107 EAL: Initializing pmd_bond for net_bond_0
2108 EAL: Create bonded device net_bond_0 on port 0 in mode 0 on socket 0.
2109 Port 0 is attached. Now total ports is 1
2116 Detach a specific port::
2118 testpmd> port detach (port_id)
2120 Before detaching a port, the port should be stopped and closed.
2122 For example, to detach a pci device port 0.
2124 .. code-block:: console
2126 testpmd> port stop 0
2129 testpmd> port close 0
2133 testpmd> port detach 0
2135 EAL: PCI device 0000:0a:00.0 on NUMA socket -1
2136 EAL: remove driver: 8086:10fb rte_ixgbe_pmd
2137 EAL: PCI memory unmapped at 0x7f83bfa00000
2138 EAL: PCI memory unmapped at 0x7f83bfa80000
2142 For example, to detach a virtual device port 0.
2144 .. code-block:: console
2146 testpmd> port stop 0
2149 testpmd> port close 0
2153 testpmd> port detach 0
2155 PMD: Closing pcap ethdev on numa socket 0
2156 Port 'net_pcap0' is detached. Now total ports is 0
2159 To remove a pci device completely from the system, first detach the port from testpmd.
2160 Then the device should be moved under kernel management.
2161 Finally the device can be removed using kernel pci hotplug functionality.
2163 For example, to move a pci device under kernel management:
2165 .. code-block:: console
2167 sudo ./usertools/dpdk-devbind.py -b ixgbe 0000:0a:00.0
2169 ./usertools/dpdk-devbind.py --status
2171 Network devices using DPDK-compatible driver
2172 ============================================
2175 Network devices using kernel driver
2176 ===================================
2177 0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused=igb_uio
2179 To remove a port created by a virtual device, above steps are not needed.
2184 Start all ports or a specific port::
2186 testpmd> port start (port_id|all)
2191 Stop all ports or a specific port::
2193 testpmd> port stop (port_id|all)
2198 Close all ports or a specific port::
2200 testpmd> port close (port_id|all)
2205 Reset all ports or a specific port::
2207 testpmd> port reset (port_id|all)
2209 User should stop port(s) before resetting and (re-)start after reset.
2211 port config - queue ring size
2212 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2214 Configure a rx/tx queue ring size::
2216 testpmd> port (port_id) (rxq|txq) (queue_id) ring_size (value)
2218 Only take effect after command that (re-)start the port or command that setup specific queue.
2220 port start/stop queue
2221 ~~~~~~~~~~~~~~~~~~~~~
2223 Start/stop a rx/tx queue on a specific port::
2225 testpmd> port (port_id) (rxq|txq) (queue_id) (start|stop)
2227 port config - queue deferred start
2228 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2230 Switch on/off deferred start of a specific port queue::
2232 testpmd> port (port_id) (rxq|txq) (queue_id) deferred_start (on|off)
2235 ~~~~~~~~~~~~~~~~~~~~~
2237 Setup a rx/tx queue on a specific port::
2239 testpmd> port (port_id) (rxq|txq) (queue_id) setup
2241 Only take effect when port is started.
2246 Set the speed and duplex mode for all ports or a specific port::
2248 testpmd> port config (port_id|all) speed (10|100|1000|10000|25000|40000|50000|100000|200000|auto) \
2249 duplex (half|full|auto)
2251 port config - queues/descriptors
2252 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2254 Set number of queues/descriptors for rxq, txq, rxd and txd::
2256 testpmd> port config all (rxq|txq|rxd|txd) (value)
2258 This is equivalent to the ``--rxq``, ``--txq``, ``--rxd`` and ``--txd`` command-line options.
2260 port config - max-pkt-len
2261 ~~~~~~~~~~~~~~~~~~~~~~~~~
2263 Set the maximum packet length::
2265 testpmd> port config all max-pkt-len (value)
2267 This is equivalent to the ``--max-pkt-len`` command-line option.
2269 port config - max-lro-pkt-size
2270 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2272 Set the maximum LRO aggregated packet size::
2274 testpmd> port config all max-lro-pkt-size (value)
2276 This is equivalent to the ``--max-lro-pkt-size`` command-line option.
2278 port config - Drop Packets
2279 ~~~~~~~~~~~~~~~~~~~~~~~~~~
2281 Enable or disable packet drop on all RX queues of all ports when no receive buffers available::
2283 testpmd> port config all drop-en (on|off)
2285 Packet dropping when no receive buffers available is off by default.
2287 The ``on`` option is equivalent to the ``--enable-drop-en`` command-line option.
2292 Set the RSS (Receive Side Scaling) mode on or off::
2294 testpmd> port config all rss (all|default|eth|vlan|ip|tcp|udp|sctp|ether|port|vxlan|geneve|nvgre|vxlan-gpe|l2tpv3|esp|ah|pfcp|ecpri|mpls|none)
2296 RSS is on by default.
2298 The ``all`` option is equivalent to eth|vlan|ip|tcp|udp|sctp|ether|l2tpv3|esp|ah|pfcp.
2300 The ``default`` option enables all supported RSS types reported by device info.
2302 The ``none`` option is equivalent to the ``--disable-rss`` command-line option.
2304 port config - RSS Reta
2305 ~~~~~~~~~~~~~~~~~~~~~~
2307 Set the RSS (Receive Side Scaling) redirection table::
2309 testpmd> port config all rss reta (hash,queue)[,(hash,queue)]
2314 Set the DCB mode for an individual port::
2316 testpmd> port config (port_id) dcb vt (on|off) (traffic_class) pfc (on|off)
2318 The traffic class should be 4 or 8.
2323 Set the number of packets per burst::
2325 testpmd> port config all burst (value)
2327 This is equivalent to the ``--burst`` command-line option.
2329 port config - Threshold
2330 ~~~~~~~~~~~~~~~~~~~~~~~
2332 Set thresholds for TX/RX queues::
2334 testpmd> port config all (threshold) (value)
2336 Where the threshold type can be:
2338 * ``txpt:`` Set the prefetch threshold register of the TX rings, 0 <= value <= 255.
2340 * ``txht:`` Set the host threshold register of the TX rings, 0 <= value <= 255.
2342 * ``txwt:`` Set the write-back threshold register of the TX rings, 0 <= value <= 255.
2344 * ``rxpt:`` Set the prefetch threshold register of the RX rings, 0 <= value <= 255.
2346 * ``rxht:`` Set the host threshold register of the RX rings, 0 <= value <= 255.
2348 * ``rxwt:`` Set the write-back threshold register of the RX rings, 0 <= value <= 255.
2350 * ``txfreet:`` Set the transmit free threshold of the TX rings, 0 <= value <= txd.
2352 * ``rxfreet:`` Set the transmit free threshold of the RX rings, 0 <= value <= rxd.
2354 * ``txrst:`` Set the transmit RS bit threshold of TX rings, 0 <= value <= txd.
2356 These threshold options are also available from the command-line.
2358 port config pctype mapping
2359 ~~~~~~~~~~~~~~~~~~~~~~~~~~
2361 Reset pctype mapping table::
2363 testpmd> port config (port_id) pctype mapping reset
2365 Update hardware defined pctype to software defined flow type mapping table::
2367 testpmd> port config (port_id) pctype mapping update (pctype_id_0[,pctype_id_1]*) (flow_type_id)
2371 * ``pctype_id_x``: hardware pctype id as index of bit in bitmask value of the pctype mapping table.
2373 * ``flow_type_id``: software flow type id as the index of the pctype mapping table.
2375 port config input set
2376 ~~~~~~~~~~~~~~~~~~~~~
2378 Config RSS/FDIR/FDIR flexible payload input set for some pctype::
2380 testpmd> port config (port_id) pctype (pctype_id) \
2381 (hash_inset|fdir_inset|fdir_flx_inset) \
2382 (get|set|clear) field (field_idx)
2384 Clear RSS/FDIR/FDIR flexible payload input set for some pctype::
2386 testpmd> port config (port_id) pctype (pctype_id) \
2387 (hash_inset|fdir_inset|fdir_flx_inset) clear all
2391 * ``pctype_id``: hardware packet classification types.
2392 * ``field_idx``: hardware field index.
2394 port config udp_tunnel_port
2395 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
2397 Add/remove UDP tunnel port for VXLAN/GENEVE tunneling protocols::
2399 testpmd> port config (port_id) udp_tunnel_port add|rm vxlan|geneve|vxlan-gpe|ecpri (udp_port)
2401 port config tx_metadata
2402 ~~~~~~~~~~~~~~~~~~~~~~~
2404 Set Tx metadata value per port.
2405 testpmd will add this value to any Tx packet sent from this port::
2407 testpmd> port config (port_id) tx_metadata (value)
2412 Set/clear dynamic flag per port.
2413 testpmd will register this flag in the mbuf (same registration
2414 for both Tx and Rx). Then set/clear this flag for each Tx
2415 packet sent from this port. The set bit only works for Tx packet::
2417 testpmd> port config (port_id) dynf (name) (set|clear)
2422 To configure MTU(Maximum Transmission Unit) on devices using testpmd::
2424 testpmd> port config mtu (port_id) (value)
2426 port config rss hash key
2427 ~~~~~~~~~~~~~~~~~~~~~~~~
2429 To configure the RSS hash key used to compute the RSS
2430 hash of input [IP] packets received on port::
2432 testpmd> port config <port_id> rss-hash-key (ipv4|ipv4-frag|\
2433 ipv4-tcp|ipv4-udp|ipv4-sctp|ipv4-other|\
2434 ipv6|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|\
2435 ipv6-other|l2-payload|ipv6-ex|ipv6-tcp-ex|\
2436 ipv6-udp-ex <string of hex digits \
2437 (variable length, NIC dependent)>)
2439 port cleanup txq mbufs
2440 ~~~~~~~~~~~~~~~~~~~~~~
2442 To cleanup txq mbufs currently cached by driver::
2444 testpmd> port cleanup (port_id) txq (queue_id) (free_cnt)
2446 If the value of ``free_cnt`` is 0, driver should free all cached mbufs.
2451 The following sections show functions for device operations.
2456 Detach a device specified by pci address or virtual device args::
2458 testpmd> device detach (identifier)
2460 Before detaching a device associated with ports, the ports should be stopped and closed.
2462 For example, to detach a pci device whose address is 0002:03:00.0.
2464 .. code-block:: console
2466 testpmd> device detach 0002:03:00.0
2467 Removing a device...
2468 Port 1 is now closed
2469 EAL: Releasing pci mapped resource for 0002:03:00.0
2470 EAL: Calling pci_unmap_resource for 0002:03:00.0 at 0x218a050000
2471 EAL: Calling pci_unmap_resource for 0002:03:00.0 at 0x218c050000
2472 Device 0002:03:00.0 is detached
2473 Now total ports is 1
2475 For example, to detach a port created by pcap PMD.
2477 .. code-block:: console
2479 testpmd> device detach net_pcap0
2480 Removing a device...
2481 Port 0 is now closed
2482 Device net_pcap0 is detached
2483 Now total ports is 0
2486 In this case, identifier is ``net_pcap0``.
2487 This identifier format is the same as ``--vdev`` format of DPDK applications.
2489 Link Bonding Functions
2490 ----------------------
2492 The Link Bonding functions make it possible to dynamically create and
2493 manage link bonding devices from within testpmd interactive prompt.
2495 create bonded device
2496 ~~~~~~~~~~~~~~~~~~~~
2498 Create a new bonding device::
2500 testpmd> create bonded device (mode) (socket)
2502 For example, to create a bonded device in mode 1 on socket 0::
2504 testpmd> create bonded device 1 0
2505 created new bonded device (port X)
2510 Adds Ethernet device to a Link Bonding device::
2512 testpmd> add bonding slave (slave id) (port id)
2514 For example, to add Ethernet device (port 6) to a Link Bonding device (port 10)::
2516 testpmd> add bonding slave 6 10
2519 remove bonding slave
2520 ~~~~~~~~~~~~~~~~~~~~
2522 Removes an Ethernet slave device from a Link Bonding device::
2524 testpmd> remove bonding slave (slave id) (port id)
2526 For example, to remove Ethernet slave device (port 6) to a Link Bonding device (port 10)::
2528 testpmd> remove bonding slave 6 10
2533 Set the Link Bonding mode of a Link Bonding device::
2535 testpmd> set bonding mode (value) (port id)
2537 For example, to set the bonding mode of a Link Bonding device (port 10) to broadcast (mode 3)::
2539 testpmd> set bonding mode 3 10
2544 Set an Ethernet slave device as the primary device on a Link Bonding device::
2546 testpmd> set bonding primary (slave id) (port id)
2548 For example, to set the Ethernet slave device (port 6) as the primary port of a Link Bonding device (port 10)::
2550 testpmd> set bonding primary 6 10
2555 Set the MAC address of a Link Bonding device::
2557 testpmd> set bonding mac (port id) (mac)
2559 For example, to set the MAC address of a Link Bonding device (port 10) to 00:00:00:00:00:01::
2561 testpmd> set bonding mac 10 00:00:00:00:00:01
2563 set bonding balance_xmit_policy
2564 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2566 Set the transmission policy for a Link Bonding device when it is in Balance XOR mode::
2568 testpmd> set bonding balance_xmit_policy (port_id) (l2|l23|l34)
2570 For example, set a Link Bonding device (port 10) to use a balance policy of layer 3+4 (IP addresses & UDP ports)::
2572 testpmd> set bonding balance_xmit_policy 10 l34
2575 set bonding mon_period
2576 ~~~~~~~~~~~~~~~~~~~~~~
2578 Set the link status monitoring polling period in milliseconds for a bonding device.
2580 This adds support for PMD slave devices which do not support link status interrupts.
2581 When the mon_period is set to a value greater than 0 then all PMD's which do not support
2582 link status ISR will be queried every polling interval to check if their link status has changed::
2584 testpmd> set bonding mon_period (port_id) (value)
2586 For example, to set the link status monitoring polling period of bonded device (port 5) to 150ms::
2588 testpmd> set bonding mon_period 5 150
2591 set bonding lacp dedicated_queue
2592 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2594 Enable dedicated tx/rx queues on bonding devices slaves to handle LACP control plane traffic
2595 when in mode 4 (link-aggregation-802.3ad)::
2597 testpmd> set bonding lacp dedicated_queues (port_id) (enable|disable)
2600 set bonding agg_mode
2601 ~~~~~~~~~~~~~~~~~~~~
2603 Enable one of the specific aggregators mode when in mode 4 (link-aggregation-802.3ad)::
2605 testpmd> set bonding agg_mode (port_id) (bandwidth|count|stable)
2611 Show the current configuration of a Link Bonding device::
2613 testpmd> show bonding config (port id)
2616 to show the configuration a Link Bonding device (port 9) with 3 slave devices (1, 3, 4)
2617 in balance mode with a transmission policy of layer 2+3::
2619 testpmd> show bonding config 9
2621 Balance Xmit Policy: BALANCE_XMIT_POLICY_LAYER23
2623 Active Slaves (3): [1 3 4]
2630 The Register Functions can be used to read from and write to registers on the network card referenced by a port number.
2631 This is mainly useful for debugging purposes.
2632 Reference should be made to the appropriate datasheet for the network card for details on the register addresses
2633 and fields that can be accessed.
2638 Display the value of a port register::
2640 testpmd> read reg (port_id) (address)
2642 For example, to examine the Flow Director control register (FDIRCTL, 0x0000EE000) on an Intel 82599 10 GbE Controller::
2644 testpmd> read reg 0 0xEE00
2645 port 0 PCI register at offset 0xEE00: 0x4A060029 (1241907241)
2650 Display a port register bit field::
2652 testpmd> read regfield (port_id) (address) (bit_x) (bit_y)
2654 For example, reading the lowest two bits from the register in the example above::
2656 testpmd> read regfield 0 0xEE00 0 1
2657 port 0 PCI register at offset 0xEE00: bits[0, 1]=0x1 (1)
2662 Display a single port register bit::
2664 testpmd> read regbit (port_id) (address) (bit_x)
2666 For example, reading the lowest bit from the register in the example above::
2668 testpmd> read regbit 0 0xEE00 0
2669 port 0 PCI register at offset 0xEE00: bit 0=1
2674 Set the value of a port register::
2676 testpmd> write reg (port_id) (address) (value)
2678 For example, to clear a register::
2680 testpmd> write reg 0 0xEE00 0x0
2681 port 0 PCI register at offset 0xEE00: 0x00000000 (0)
2686 Set bit field of a port register::
2688 testpmd> write regfield (port_id) (address) (bit_x) (bit_y) (value)
2690 For example, writing to the register cleared in the example above::
2692 testpmd> write regfield 0 0xEE00 0 1 2
2693 port 0 PCI register at offset 0xEE00: 0x00000002 (2)
2698 Set single bit value of a port register::
2700 testpmd> write regbit (port_id) (address) (bit_x) (value)
2702 For example, to set the high bit in the register from the example above::
2704 testpmd> write regbit 0 0xEE00 31 1
2705 port 0 PCI register at offset 0xEE00: 0x8000000A (2147483658)
2707 Traffic Metering and Policing
2708 -----------------------------
2710 The following section shows functions for configuring traffic metering and
2711 policing on the ethernet device through the use of generic ethdev API.
2713 show port traffic management capability
2714 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2716 Show traffic metering and policing capability of the port::
2718 testpmd> show port meter cap (port_id)
2720 add port meter profile (srTCM rfc2967)
2721 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2723 Add meter profile (srTCM rfc2697) to the ethernet device::
2725 testpmd> add port meter profile srtcm_rfc2697 (port_id) (profile_id) \
2726 (cir) (cbs) (ebs) (packet_mode)
2730 * ``profile_id``: ID for the meter profile.
2731 * ``cir``: Committed Information Rate (CIR) (bytes per second or packets per second).
2732 * ``cbs``: Committed Burst Size (CBS) (bytes or packets).
2733 * ``ebs``: Excess Burst Size (EBS) (bytes or packets).
2734 * ``packet_mode``: Packets mode for meter profile.
2736 add port meter profile (trTCM rfc2968)
2737 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2739 Add meter profile (srTCM rfc2698) to the ethernet device::
2741 testpmd> add port meter profile trtcm_rfc2698 (port_id) (profile_id) \
2742 (cir) (pir) (cbs) (pbs) (packet_mode)
2746 * ``profile_id``: ID for the meter profile.
2747 * ``cir``: Committed information rate (bytes per second or packets per second).
2748 * ``pir``: Peak information rate (bytes per second or packets per second).
2749 * ``cbs``: Committed burst size (bytes or packets).
2750 * ``pbs``: Peak burst size (bytes or packets).
2751 * ``packet_mode``: Packets mode for meter profile.
2753 add port meter profile (trTCM rfc4115)
2754 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2756 Add meter profile (trTCM rfc4115) to the ethernet device::
2758 testpmd> add port meter profile trtcm_rfc4115 (port_id) (profile_id) \
2759 (cir) (eir) (cbs) (ebs) (packet_mode)
2763 * ``profile_id``: ID for the meter profile.
2764 * ``cir``: Committed information rate (bytes per second or packets per second).
2765 * ``eir``: Excess information rate (bytes per second or packets per second).
2766 * ``cbs``: Committed burst size (bytes or packets).
2767 * ``ebs``: Excess burst size (bytes or packets).
2768 * ``packet_mode``: Packets mode for meter profile.
2770 delete port meter profile
2771 ~~~~~~~~~~~~~~~~~~~~~~~~~
2773 Delete meter profile from the ethernet device::
2775 testpmd> del port meter profile (port_id) (profile_id)
2780 Create new policy object for the ethernet device::
2782 testpmd> add port meter policy (port_id) (policy_id) g_actions \
2783 {action} y_actions {action} r_actions {action}
2787 * ``policy_id``: policy ID.
2788 * ``action``: action lists for green/yellow/red colors.
2793 Delete policy object for the ethernet device::
2795 testpmd> del port meter policy (port_id) (policy_id)
2799 * ``policy_id``: policy ID.
2804 Create new meter object for the ethernet device::
2806 testpmd> create port meter (port_id) (mtr_id) (profile_id) \
2807 (policy_id) (meter_enable) (stats_mask) (shared) \
2808 (use_pre_meter_color) [(dscp_tbl_entry0) (dscp_tbl_entry1)...\
2813 * ``mtr_id``: meter object ID.
2814 * ``profile_id``: ID for the meter profile.
2815 * ``policy_id``: ID for the policy.
2816 * ``meter_enable``: When this parameter has a non-zero value, the meter object
2817 gets enabled at the time of creation, otherwise remains disabled.
2818 * ``stats_mask``: Mask of statistics counter types to be enabled for the
2820 * ``shared``: When this parameter has a non-zero value, the meter object is
2821 shared by multiple flows. Otherwise, meter object is used by single flow.
2822 * ``use_pre_meter_color``: When this parameter has a non-zero value, the
2823 input color for the current meter object is determined by the latest meter
2824 object in the same flow. Otherwise, the current meter object uses the
2825 *dscp_table* to determine the input color.
2826 * ``dscp_tbl_entryx``: DSCP table entry x providing meter providing input
2827 color, 0 <= x <= 63.
2832 Enable meter for the ethernet device::
2834 testpmd> enable port meter (port_id) (mtr_id)
2839 Disable meter for the ethernet device::
2841 testpmd> disable port meter (port_id) (mtr_id)
2846 Delete meter for the ethernet device::
2848 testpmd> del port meter (port_id) (mtr_id)
2850 Set port meter profile
2851 ~~~~~~~~~~~~~~~~~~~~~~
2853 Set meter profile for the ethernet device::
2855 testpmd> set port meter profile (port_id) (mtr_id) (profile_id)
2857 set port meter dscp table
2858 ~~~~~~~~~~~~~~~~~~~~~~~~~
2860 Set meter dscp table for the ethernet device::
2862 testpmd> set port meter dscp table (port_id) (mtr_id) [(dscp_tbl_entry0) \
2863 (dscp_tbl_entry1)...(dscp_tbl_entry63)]
2865 set port meter stats mask
2866 ~~~~~~~~~~~~~~~~~~~~~~~~~
2868 Set meter stats mask for the ethernet device::
2870 testpmd> set port meter stats mask (port_id) (mtr_id) (stats_mask)
2874 * ``stats_mask``: Bit mask indicating statistics counter types to be enabled.
2876 show port meter stats
2877 ~~~~~~~~~~~~~~~~~~~~~
2879 Show meter stats of the ethernet device::
2881 testpmd> show port meter stats (port_id) (mtr_id) (clear)
2885 * ``clear``: Flag that indicates whether the statistics counters should
2886 be cleared (i.e. set to zero) immediately after they have been read or not.
2891 The following section shows functions for configuring traffic management on
2892 the ethernet device through the use of generic TM API.
2894 show port traffic management capability
2895 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2897 Show traffic management capability of the port::
2899 testpmd> show port tm cap (port_id)
2901 show port traffic management capability (hierarchy level)
2902 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2904 Show traffic management hierarchy level capability of the port::
2906 testpmd> show port tm level cap (port_id) (level_id)
2908 show port traffic management capability (hierarchy node level)
2909 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2911 Show the traffic management hierarchy node capability of the port::
2913 testpmd> show port tm node cap (port_id) (node_id)
2915 show port traffic management hierarchy node type
2916 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2918 Show the port traffic management hierarchy node type::
2920 testpmd> show port tm node type (port_id) (node_id)
2922 show port traffic management hierarchy node stats
2923 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2925 Show the port traffic management hierarchy node statistics::
2927 testpmd> show port tm node stats (port_id) (node_id) (clear)
2931 * ``clear``: When this parameter has a non-zero value, the statistics counters
2932 are cleared (i.e. set to zero) immediately after they have been read,
2933 otherwise the statistics counters are left untouched.
2935 Add port traffic management private shaper profile
2936 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2938 Add the port traffic management private shaper profile::
2940 testpmd> add port tm node shaper profile (port_id) (shaper_profile_id) \
2941 (cmit_tb_rate) (cmit_tb_size) (peak_tb_rate) (peak_tb_size) \
2942 (packet_length_adjust) (packet_mode)
2946 * ``shaper_profile id``: Shaper profile ID for the new profile.
2947 * ``cmit_tb_rate``: Committed token bucket rate (bytes per second or packets per second).
2948 * ``cmit_tb_size``: Committed token bucket size (bytes or packets).
2949 * ``peak_tb_rate``: Peak token bucket rate (bytes per second or packets per second).
2950 * ``peak_tb_size``: Peak token bucket size (bytes or packets).
2951 * ``packet_length_adjust``: The value (bytes) to be added to the length of
2952 each packet for the purpose of shaping. This parameter value can be used to
2953 correct the packet length with the framing overhead bytes that are consumed
2955 * ``packet_mode``: Shaper configured in packet mode. This parameter value if
2956 zero, configures shaper in byte mode and if non-zero configures it in packet
2959 Delete port traffic management private shaper profile
2960 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2962 Delete the port traffic management private shaper::
2964 testpmd> del port tm node shaper profile (port_id) (shaper_profile_id)
2968 * ``shaper_profile id``: Shaper profile ID that needs to be deleted.
2970 Add port traffic management shared shaper
2971 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2973 Create the port traffic management shared shaper::
2975 testpmd> add port tm node shared shaper (port_id) (shared_shaper_id) \
2980 * ``shared_shaper_id``: Shared shaper ID to be created.
2981 * ``shaper_profile id``: Shaper profile ID for shared shaper.
2983 Set port traffic management shared shaper
2984 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2986 Update the port traffic management shared shaper::
2988 testpmd> set port tm node shared shaper (port_id) (shared_shaper_id) \
2993 * ``shared_shaper_id``: Shared shaper ID to be update.
2994 * ``shaper_profile id``: Shaper profile ID for shared shaper.
2996 Delete port traffic management shared shaper
2997 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2999 Delete the port traffic management shared shaper::
3001 testpmd> del port tm node shared shaper (port_id) (shared_shaper_id)
3005 * ``shared_shaper_id``: Shared shaper ID to be deleted.
3007 Set port traffic management hierarchy node private shaper
3008 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3010 set the port traffic management hierarchy node private shaper::
3012 testpmd> set port tm node shaper profile (port_id) (node_id) \
3017 * ``shaper_profile id``: Private shaper profile ID to be enabled on the
3020 Add port traffic management WRED profile
3021 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3023 Create a new WRED profile::
3025 testpmd> add port tm node wred profile (port_id) (wred_profile_id) \
3026 (color_g) (min_th_g) (max_th_g) (maxp_inv_g) (wq_log2_g) \
3027 (color_y) (min_th_y) (max_th_y) (maxp_inv_y) (wq_log2_y) \
3028 (color_r) (min_th_r) (max_th_r) (maxp_inv_r) (wq_log2_r)
3032 * ``wred_profile id``: Identifier for the newly create WRED profile
3033 * ``color_g``: Packet color (green)
3034 * ``min_th_g``: Minimum queue threshold for packet with green color
3035 * ``max_th_g``: Minimum queue threshold for packet with green color
3036 * ``maxp_inv_g``: Inverse of packet marking probability maximum value (maxp)
3037 * ``wq_log2_g``: Negated log2 of queue weight (wq)
3038 * ``color_y``: Packet color (yellow)
3039 * ``min_th_y``: Minimum queue threshold for packet with yellow color
3040 * ``max_th_y``: Minimum queue threshold for packet with yellow color
3041 * ``maxp_inv_y``: Inverse of packet marking probability maximum value (maxp)
3042 * ``wq_log2_y``: Negated log2 of queue weight (wq)
3043 * ``color_r``: Packet color (red)
3044 * ``min_th_r``: Minimum queue threshold for packet with yellow color
3045 * ``max_th_r``: Minimum queue threshold for packet with yellow color
3046 * ``maxp_inv_r``: Inverse of packet marking probability maximum value (maxp)
3047 * ``wq_log2_r``: Negated log2 of queue weight (wq)
3049 Delete port traffic management WRED profile
3050 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3052 Delete the WRED profile::
3054 testpmd> del port tm node wred profile (port_id) (wred_profile_id)
3056 Add port traffic management hierarchy nonleaf node
3057 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3059 Add nonleaf node to port traffic management hierarchy::
3061 testpmd> add port tm nonleaf node (port_id) (node_id) (parent_node_id) \
3062 (priority) (weight) (level_id) (shaper_profile_id) \
3063 (n_sp_priorities) (stats_mask) (n_shared_shapers) \
3064 [(shared_shaper_0) (shared_shaper_1) ...] \
3068 * ``parent_node_id``: Node ID of the parent.
3069 * ``priority``: Node priority (highest node priority is zero). This is used by
3070 the SP algorithm running on the parent node for scheduling this node.
3071 * ``weight``: Node weight (lowest weight is one). The node weight is relative
3072 to the weight sum of all siblings that have the same priority. It is used by
3073 the WFQ algorithm running on the parent node for scheduling this node.
3074 * ``level_id``: Hierarchy level of the node.
3075 * ``shaper_profile_id``: Shaper profile ID of the private shaper to be used by
3077 * ``n_sp_priorities``: Number of strict priorities.
3078 * ``stats_mask``: Mask of statistics counter types to be enabled for this node.
3079 * ``n_shared_shapers``: Number of shared shapers.
3080 * ``shared_shaper_id``: Shared shaper id.
3082 Add port traffic management hierarchy nonleaf node with packet mode
3083 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3085 Add nonleaf node with packet mode to port traffic management hierarchy::
3087 testpmd> add port tm nonleaf node pktmode (port_id) (node_id) (parent_node_id) \
3088 (priority) (weight) (level_id) (shaper_profile_id) \
3089 (n_sp_priorities) (stats_mask) (n_shared_shapers) \
3090 [(shared_shaper_0) (shared_shaper_1) ...] \
3094 * ``parent_node_id``: Node ID of the parent.
3095 * ``priority``: Node priority (highest node priority is zero). This is used by
3096 the SP algorithm running on the parent node for scheduling this node.
3097 * ``weight``: Node weight (lowest weight is one). The node weight is relative
3098 to the weight sum of all siblings that have the same priority. It is used by
3099 the WFQ algorithm running on the parent node for scheduling this node.
3100 * ``level_id``: Hierarchy level of the node.
3101 * ``shaper_profile_id``: Shaper profile ID of the private shaper to be used by
3103 * ``n_sp_priorities``: Number of strict priorities. Packet mode is enabled on
3105 * ``stats_mask``: Mask of statistics counter types to be enabled for this node.
3106 * ``n_shared_shapers``: Number of shared shapers.
3107 * ``shared_shaper_id``: Shared shaper id.
3109 Add port traffic management hierarchy leaf node
3110 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3112 Add leaf node to port traffic management hierarchy::
3114 testpmd> add port tm leaf node (port_id) (node_id) (parent_node_id) \
3115 (priority) (weight) (level_id) (shaper_profile_id) \
3116 (cman_mode) (wred_profile_id) (stats_mask) (n_shared_shapers) \
3117 [(shared_shaper_id) (shared_shaper_id) ...] \
3121 * ``parent_node_id``: Node ID of the parent.
3122 * ``priority``: Node priority (highest node priority is zero). This is used by
3123 the SP algorithm running on the parent node for scheduling this node.
3124 * ``weight``: Node weight (lowest weight is one). The node weight is relative
3125 to the weight sum of all siblings that have the same priority. It is used by
3126 the WFQ algorithm running on the parent node for scheduling this node.
3127 * ``level_id``: Hierarchy level of the node.
3128 * ``shaper_profile_id``: Shaper profile ID of the private shaper to be used by
3130 * ``cman_mode``: Congestion management mode to be enabled for this node.
3131 * ``wred_profile_id``: WRED profile id to be enabled for this node.
3132 * ``stats_mask``: Mask of statistics counter types to be enabled for this node.
3133 * ``n_shared_shapers``: Number of shared shapers.
3134 * ``shared_shaper_id``: Shared shaper id.
3136 Delete port traffic management hierarchy node
3137 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3139 Delete node from port traffic management hierarchy::
3141 testpmd> del port tm node (port_id) (node_id)
3143 Update port traffic management hierarchy parent node
3144 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3146 Update port traffic management hierarchy parent node::
3148 testpmd> set port tm node parent (port_id) (node_id) (parent_node_id) \
3151 This function can only be called after the hierarchy commit invocation. Its
3152 success depends on the port support for this operation, as advertised through
3153 the port capability set. This function is valid for all nodes of the traffic
3154 management hierarchy except root node.
3156 Suspend port traffic management hierarchy node
3157 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3159 testpmd> suspend port tm node (port_id) (node_id)
3161 Resume port traffic management hierarchy node
3162 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3164 testpmd> resume port tm node (port_id) (node_id)
3166 Commit port traffic management hierarchy
3167 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3169 Commit the traffic management hierarchy on the port::
3171 testpmd> port tm hierarchy commit (port_id) (clean_on_fail)
3175 * ``clean_on_fail``: When set to non-zero, hierarchy is cleared on function
3176 call failure. On the other hand, hierarchy is preserved when this parameter
3179 Set port traffic management mark VLAN dei
3180 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3182 Enables/Disables the traffic management marking on the port for VLAN packets::
3184 testpmd> set port tm mark vlan_dei <port_id> <green> <yellow> <red>
3188 * ``port_id``: The port which on which VLAN packets marked as ``green`` or
3189 ``yellow`` or ``red`` will have dei bit enabled
3191 * ``green`` enable 1, disable 0 marking for dei bit of VLAN packets marked as green
3193 * ``yellow`` enable 1, disable 0 marking for dei bit of VLAN packets marked as yellow
3195 * ``red`` enable 1, disable 0 marking for dei bit of VLAN packets marked as red
3197 Set port traffic management mark IP dscp
3198 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3200 Enables/Disables the traffic management marking on the port for IP dscp packets::
3202 testpmd> set port tm mark ip_dscp <port_id> <green> <yellow> <red>
3206 * ``port_id``: The port which on which IP packets marked as ``green`` or
3207 ``yellow`` or ``red`` will have IP dscp bits updated
3209 * ``green`` enable 1, disable 0 marking IP dscp to low drop precedence for green packets
3211 * ``yellow`` enable 1, disable 0 marking IP dscp to medium drop precedence for yellow packets
3213 * ``red`` enable 1, disable 0 marking IP dscp to high drop precedence for red packets
3215 Set port traffic management mark IP ecn
3216 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3218 Enables/Disables the traffic management marking on the port for IP ecn packets::
3220 testpmd> set port tm mark ip_ecn <port_id> <green> <yellow> <red>
3224 * ``port_id``: The port which on which IP packets marked as ``green`` or
3225 ``yellow`` or ``red`` will have IP ecn bits updated
3227 * ``green`` enable 1, disable 0 marking IP ecn for green marked packets with ecn of 2'b01 or 2'b10
3228 to ecn of 2'b11 when IP is caring TCP or SCTP
3230 * ``yellow`` enable 1, disable 0 marking IP ecn for yellow marked packets with ecn of 2'b01 or 2'b10
3231 to ecn of 2'b11 when IP is caring TCP or SCTP
3233 * ``red`` enable 1, disable 0 marking IP ecn for yellow marked packets with ecn of 2'b01 or 2'b10
3234 to ecn of 2'b11 when IP is caring TCP or SCTP
3239 This section details the available filter functions that are available.
3241 Note these functions interface the deprecated legacy filtering framework,
3242 superseded by *rte_flow*. See `Flow rules management`_.
3244 .. _testpmd_flow_director:
3249 Set flow director's input masks::
3251 flow_director_mask (port_id) mode IP vlan (vlan_value) \
3252 src_mask (ipv4_src) (ipv6_src) (src_port) \
3253 dst_mask (ipv4_dst) (ipv6_dst) (dst_port)
3255 flow_director_mask (port_id) mode MAC-VLAN vlan (vlan_value)
3257 flow_director_mask (port_id) mode Tunnel vlan (vlan_value) \
3258 mac (mac_value) tunnel-type (tunnel_type_value) \
3259 tunnel-id (tunnel_id_value)
3261 Example, to set flow director mask on port 0::
3263 testpmd> flow_director_mask 0 mode IP vlan 0xefff \
3264 src_mask 255.255.255.255 \
3265 FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF \
3266 dst_mask 255.255.255.255 \
3267 FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF
3269 flow_director_flex_payload
3270 ~~~~~~~~~~~~~~~~~~~~~~~~~~
3272 Configure flexible payload selection::
3274 flow_director_flex_payload (port_id) (raw|l2|l3|l4) (config)
3276 For example, to select the first 16 bytes from the offset 4 (bytes) of packet's payload as flexible payload::
3278 testpmd> flow_director_flex_payload 0 l4 \
3279 (4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19)
3282 .. _testpmd_rte_flow:
3284 Flow rules management
3285 ---------------------
3287 Control of the generic flow API (*rte_flow*) is fully exposed through the
3288 ``flow`` command (validation, creation, destruction, queries and operation
3291 Considering *rte_flow* overlaps with all `Filter Functions`_, using both
3292 features simultaneously may cause undefined side-effects and is therefore
3298 Because the ``flow`` command uses dynamic tokens to handle the large number
3299 of possible flow rules combinations, its behavior differs slightly from
3300 other commands, in particular:
3302 - Pressing *?* or the *<tab>* key displays contextual help for the current
3303 token, not that of the entire command.
3305 - Optional and repeated parameters are supported (provided they are listed
3306 in the contextual help).
3308 The first parameter stands for the operation mode. Possible operations and
3309 their general syntax are described below. They are covered in detail in the
3312 - Check whether a flow rule can be created::
3314 flow validate {port_id}
3315 [group {group_id}] [priority {level}] [ingress] [egress] [transfer]
3316 pattern {item} [/ {item} [...]] / end
3317 actions {action} [/ {action} [...]] / end
3319 - Create a flow rule::
3321 flow create {port_id}
3322 [group {group_id}] [priority {level}] [ingress] [egress] [transfer]
3323 pattern {item} [/ {item} [...]] / end
3324 actions {action} [/ {action} [...]] / end
3326 - Destroy specific flow rules::
3328 flow destroy {port_id} rule {rule_id} [...]
3330 - Destroy all flow rules::
3332 flow flush {port_id}
3334 - Query an existing flow rule::
3336 flow query {port_id} {rule_id} {action}
3338 - List existing flow rules sorted by priority, filtered by group
3341 flow list {port_id} [group {group_id}] [...]
3343 - Restrict ingress traffic to the defined flow rules::
3345 flow isolate {port_id} {boolean}
3347 - Dump internal representation information of all flows in hardware::
3349 flow dump {port_id} all {output_file}
3353 flow dump {port_id} rule {rule_id} {output_file}
3355 - List and destroy aged flow rules::
3357 flow aged {port_id} [destroy]
3359 - Tunnel offload - create a tunnel stub::
3361 flow tunnel create {port_id} type {tunnel_type}
3363 - Tunnel offload - destroy a tunnel stub::
3365 flow tunnel destroy {port_id} id {tunnel_id}
3367 - Tunnel offload - list port tunnel stubs::
3369 flow tunnel list {port_id}
3371 Creating a tunnel stub for offload
3372 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3374 ``flow tunnel create`` setup a tunnel stub for tunnel offload flow rules::
3376 flow tunnel create {port_id} type {tunnel_type}
3378 If successful, it will return a tunnel stub ID usable with other commands::
3380 port [...]: flow tunnel #[...] type [...]
3382 Tunnel stub ID is relative to a port.
3384 Destroying tunnel offload stub
3385 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3387 ``flow tunnel destroy`` destroy port tunnel stub::
3389 flow tunnel destroy {port_id} id {tunnel_id}
3391 Listing tunnel offload stubs
3392 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3394 ``flow tunnel list`` list port tunnel offload stubs::
3396 flow tunnel list {port_id}
3398 Validating flow rules
3399 ~~~~~~~~~~~~~~~~~~~~~
3401 ``flow validate`` reports whether a flow rule would be accepted by the
3402 underlying device in its current state but stops short of creating it. It is
3403 bound to ``rte_flow_validate()``::
3405 flow validate {port_id}
3406 [group {group_id}] [priority {level}] [ingress] [egress] [transfer]
3407 pattern {item} [/ {item} [...]] / end
3408 actions {action} [/ {action} [...]] / end
3410 If successful, it will show::
3414 Otherwise it will show an error message of the form::
3416 Caught error type [...] ([...]): [...]
3418 This command uses the same parameters as ``flow create``, their format is
3419 described in `Creating flow rules`_.
3421 Check whether redirecting any Ethernet packet received on port 0 to RX queue
3422 index 6 is supported::
3424 testpmd> flow validate 0 ingress pattern eth / end
3425 actions queue index 6 / end
3429 Port 0 does not support TCPv6 rules::
3431 testpmd> flow validate 0 ingress pattern eth / ipv6 / tcp / end
3433 Caught error type 9 (specific pattern item): Invalid argument
3439 ``flow create`` validates and creates the specified flow rule. It is bound
3440 to ``rte_flow_create()``::
3442 flow create {port_id}
3443 [group {group_id}] [priority {level}] [ingress] [egress] [transfer]
3444 [tunnel_set {tunnel_id}] [tunnel_match {tunnel_id}]
3445 pattern {item} [/ {item} [...]] / end
3446 actions {action} [/ {action} [...]] / end
3448 If successful, it will return a flow rule ID usable with other commands::
3450 Flow rule #[...] created
3452 Otherwise it will show an error message of the form::
3454 Caught error type [...] ([...]): [...]
3456 Parameters describe in the following order:
3458 - Attributes (*group*, *priority*, *ingress*, *egress*, *transfer* tokens).
3459 - Tunnel offload specification (tunnel_set, tunnel_match)
3460 - A matching pattern, starting with the *pattern* token and terminated by an
3462 - Actions, starting with the *actions* token and terminated by an *end*
3465 These translate directly to *rte_flow* objects provided as-is to the
3466 underlying functions.
3468 The shortest valid definition only comprises mandatory tokens::
3470 testpmd> flow create 0 pattern end actions end
3472 Note that PMDs may refuse rules that essentially do nothing such as this
3475 **All unspecified object values are automatically initialized to 0.**
3480 These tokens affect flow rule attributes (``struct rte_flow_attr``) and are
3481 specified before the ``pattern`` token.
3483 - ``group {group id}``: priority group.
3484 - ``priority {level}``: priority level within group.
3485 - ``ingress``: rule applies to ingress traffic.
3486 - ``egress``: rule applies to egress traffic.
3487 - ``transfer``: apply rule directly to endpoints found in pattern.
3489 Each instance of an attribute specified several times overrides the previous
3490 value as shown below (group 4 is used)::
3492 testpmd> flow create 0 group 42 group 24 group 4 [...]
3494 Note that once enabled, ``ingress`` and ``egress`` cannot be disabled.
3496 While not specifying a direction is an error, some rules may allow both
3499 Most rules affect RX therefore contain the ``ingress`` token::
3501 testpmd> flow create 0 ingress pattern [...]
3506 Indicate tunnel offload rule type
3508 - ``tunnel_set {tunnel_id}``: mark rule as tunnel offload decap_set type.
3509 - ``tunnel_match {tunnel_id}``: mark rule as tunel offload match type.
3514 A matching pattern starts after the ``pattern`` token. It is made of pattern
3515 items and is terminated by a mandatory ``end`` item.
3517 Items are named after their type (*RTE_FLOW_ITEM_TYPE_* from ``enum
3518 rte_flow_item_type``).
3520 The ``/`` token is used as a separator between pattern items as shown
3523 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end [...]
3525 Note that protocol items like these must be stacked from lowest to highest
3526 layer to make sense. For instance, the following rule is either invalid or
3527 unlikely to match any packet::
3529 testpmd> flow create 0 ingress pattern eth / udp / ipv4 / end [...]
3531 More information on these restrictions can be found in the *rte_flow*
3534 Several items support additional specification structures, for example
3535 ``ipv4`` allows specifying source and destination addresses as follows::
3537 testpmd> flow create 0 ingress pattern eth / ipv4 src is 10.1.1.1
3538 dst is 10.2.0.0 / end [...]
3540 This rule matches all IPv4 traffic with the specified properties.
3542 In this example, ``src`` and ``dst`` are field names of the underlying
3543 ``struct rte_flow_item_ipv4`` object. All item properties can be specified
3544 in a similar fashion.
3546 The ``is`` token means that the subsequent value must be matched exactly,
3547 and assigns ``spec`` and ``mask`` fields in ``struct rte_flow_item``
3548 accordingly. Possible assignment tokens are:
3550 - ``is``: match value perfectly (with full bit-mask).
3551 - ``spec``: match value according to configured bit-mask.
3552 - ``last``: specify upper bound to establish a range.
3553 - ``mask``: specify bit-mask with relevant bits set to one.
3554 - ``prefix``: generate bit-mask with <prefix-length> most-significant bits set to one.
3556 These yield identical results::
3558 ipv4 src is 10.1.1.1
3562 ipv4 src spec 10.1.1.1 src mask 255.255.255.255
3566 ipv4 src spec 10.1.1.1 src prefix 32
3570 ipv4 src is 10.1.1.1 src last 10.1.1.1 # range with a single value
3574 ipv4 src is 10.1.1.1 src last 0 # 0 disables range
3576 Inclusive ranges can be defined with ``last``::
3578 ipv4 src is 10.1.1.1 src last 10.2.3.4 # 10.1.1.1 to 10.2.3.4
3580 Note that ``mask`` affects both ``spec`` and ``last``::
3582 ipv4 src is 10.1.1.1 src last 10.2.3.4 src mask 255.255.0.0
3583 # matches 10.1.0.0 to 10.2.255.255
3585 Properties can be modified multiple times::
3587 ipv4 src is 10.1.1.1 src is 10.1.2.3 src is 10.2.3.4 # matches 10.2.3.4
3591 ipv4 src is 10.1.1.1 src prefix 24 src prefix 16 # matches 10.1.0.0/16
3596 This section lists supported pattern items and their attributes, if any.
3598 - ``end``: end list of pattern items.
3600 - ``void``: no-op pattern item.
3602 - ``invert``: perform actions when pattern does not match.
3604 - ``any``: match any protocol for the current layer.
3606 - ``num {unsigned}``: number of layers covered.
3608 - ``pf``: match traffic from/to the physical function.
3610 - ``vf``: match traffic from/to a virtual function ID.
3612 - ``id {unsigned}``: VF ID.
3614 - ``phy_port``: match traffic from/to a specific physical port.
3616 - ``index {unsigned}``: physical port index.
3618 - ``port_id``: match traffic from/to a given DPDK port ID.
3620 - ``id {unsigned}``: DPDK port ID.
3622 - ``mark``: match value set in previously matched flow rule using the mark action.
3624 - ``id {unsigned}``: arbitrary integer value.
3626 - ``raw``: match an arbitrary byte string.
3628 - ``relative {boolean}``: look for pattern after the previous item.
3629 - ``search {boolean}``: search pattern from offset (see also limit).
3630 - ``offset {integer}``: absolute or relative offset for pattern.
3631 - ``limit {unsigned}``: search area limit for start of pattern.
3632 - ``pattern {string}``: byte string to look for.
3634 - ``eth``: match Ethernet header.
3636 - ``dst {MAC-48}``: destination MAC.
3637 - ``src {MAC-48}``: source MAC.
3638 - ``type {unsigned}``: EtherType or TPID.
3640 - ``vlan``: match 802.1Q/ad VLAN tag.
3642 - ``tci {unsigned}``: tag control information.
3643 - ``pcp {unsigned}``: priority code point.
3644 - ``dei {unsigned}``: drop eligible indicator.
3645 - ``vid {unsigned}``: VLAN identifier.
3646 - ``inner_type {unsigned}``: inner EtherType or TPID.
3648 - ``ipv4``: match IPv4 header.
3650 - ``tos {unsigned}``: type of service.
3651 - ``ttl {unsigned}``: time to live.
3652 - ``proto {unsigned}``: next protocol ID.
3653 - ``src {ipv4 address}``: source address.
3654 - ``dst {ipv4 address}``: destination address.
3656 - ``ipv6``: match IPv6 header.
3658 - ``tc {unsigned}``: traffic class.
3659 - ``flow {unsigned}``: flow label.
3660 - ``proto {unsigned}``: protocol (next header).
3661 - ``hop {unsigned}``: hop limit.
3662 - ``src {ipv6 address}``: source address.
3663 - ``dst {ipv6 address}``: destination address.
3665 - ``icmp``: match ICMP header.
3667 - ``type {unsigned}``: ICMP packet type.
3668 - ``code {unsigned}``: ICMP packet code.
3670 - ``udp``: match UDP header.
3672 - ``src {unsigned}``: UDP source port.
3673 - ``dst {unsigned}``: UDP destination port.
3675 - ``tcp``: match TCP header.
3677 - ``src {unsigned}``: TCP source port.
3678 - ``dst {unsigned}``: TCP destination port.
3680 - ``sctp``: match SCTP header.
3682 - ``src {unsigned}``: SCTP source port.
3683 - ``dst {unsigned}``: SCTP destination port.
3684 - ``tag {unsigned}``: validation tag.
3685 - ``cksum {unsigned}``: checksum.
3687 - ``vxlan``: match VXLAN header.
3689 - ``vni {unsigned}``: VXLAN identifier.
3691 - ``e_tag``: match IEEE 802.1BR E-Tag header.
3693 - ``grp_ecid_b {unsigned}``: GRP and E-CID base.
3695 - ``nvgre``: match NVGRE header.
3697 - ``tni {unsigned}``: virtual subnet ID.
3699 - ``mpls``: match MPLS header.
3701 - ``label {unsigned}``: MPLS label.
3703 - ``gre``: match GRE header.
3705 - ``protocol {unsigned}``: protocol type.
3707 - ``gre_key``: match GRE optional key field.
3709 - ``value {unsigned}``: key value.
3711 - ``fuzzy``: fuzzy pattern match, expect faster than default.
3713 - ``thresh {unsigned}``: accuracy threshold.
3715 - ``gtp``, ``gtpc``, ``gtpu``: match GTPv1 header.
3717 - ``teid {unsigned}``: tunnel endpoint identifier.
3719 - ``geneve``: match GENEVE header.
3721 - ``vni {unsigned}``: virtual network identifier.
3722 - ``protocol {unsigned}``: protocol type.
3724 - ``geneve-opt``: match GENEVE header option.
3726 - ``class {unsigned}``: GENEVE option class.
3727 - ``type {unsigned}``: GENEVE option type.
3728 - ``length {unsigned}``: GENEVE option length in 32-bit words.
3729 - ``data {hex string}``: GENEVE option data, the length is defined by
3732 - ``vxlan-gpe``: match VXLAN-GPE header.
3734 - ``vni {unsigned}``: VXLAN-GPE identifier.
3736 - ``arp_eth_ipv4``: match ARP header for Ethernet/IPv4.
3738 - ``sha {MAC-48}``: sender hardware address.
3739 - ``spa {ipv4 address}``: sender IPv4 address.
3740 - ``tha {MAC-48}``: target hardware address.
3741 - ``tpa {ipv4 address}``: target IPv4 address.
3743 - ``ipv6_ext``: match presence of any IPv6 extension header.
3745 - ``next_hdr {unsigned}``: next header.
3747 - ``icmp6``: match any ICMPv6 header.
3749 - ``type {unsigned}``: ICMPv6 type.
3750 - ``code {unsigned}``: ICMPv6 code.
3752 - ``icmp6_nd_ns``: match ICMPv6 neighbor discovery solicitation.
3754 - ``target_addr {ipv6 address}``: target address.
3756 - ``icmp6_nd_na``: match ICMPv6 neighbor discovery advertisement.
3758 - ``target_addr {ipv6 address}``: target address.
3760 - ``icmp6_nd_opt``: match presence of any ICMPv6 neighbor discovery option.
3762 - ``type {unsigned}``: ND option type.
3764 - ``icmp6_nd_opt_sla_eth``: match ICMPv6 neighbor discovery source Ethernet
3765 link-layer address option.
3767 - ``sla {MAC-48}``: source Ethernet LLA.
3769 - ``icmp6_nd_opt_tla_eth``: match ICMPv6 neighbor discovery target Ethernet
3770 link-layer address option.
3772 - ``tla {MAC-48}``: target Ethernet LLA.
3774 - ``meta``: match application specific metadata.
3776 - ``data {unsigned}``: metadata value.
3778 - ``gtp_psc``: match GTP PDU extension header with type 0x85.
3780 - ``pdu_type {unsigned}``: PDU type.
3782 - ``qfi {unsigned}``: PPP, RQI and QoS flow identifier.
3784 - ``pppoes``, ``pppoed``: match PPPoE header.
3786 - ``session_id {unsigned}``: session identifier.
3788 - ``pppoe_proto_id``: match PPPoE session protocol identifier.
3790 - ``proto_id {unsigned}``: PPP protocol identifier.
3792 - ``l2tpv3oip``: match L2TPv3 over IP header.
3794 - ``session_id {unsigned}``: L2TPv3 over IP session identifier.
3796 - ``ah``: match AH header.
3798 - ``spi {unsigned}``: security parameters index.
3800 - ``pfcp``: match PFCP header.
3802 - ``s_field {unsigned}``: S field.
3803 - ``seid {unsigned}``: session endpoint identifier.
3805 - ``integrity``: match packet integrity.
3807 - ``level {unsigned}``: Packet encapsulation level the item should
3808 apply to. See rte_flow_action_rss for details.
3809 - ``value {unsigned}``: A bitmask that specify what packet elements
3810 must be matched for integrity.
3812 - ``conntrack``: match conntrack state.
3817 A list of actions starts after the ``actions`` token in the same fashion as
3818 `Matching pattern`_; actions are separated by ``/`` tokens and the list is
3819 terminated by a mandatory ``end`` action.
3821 Actions are named after their type (*RTE_FLOW_ACTION_TYPE_* from ``enum
3822 rte_flow_action_type``).
3824 Dropping all incoming UDPv4 packets can be expressed as follows::
3826 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
3829 Several actions have configurable properties which must be specified when
3830 there is no valid default value. For example, ``queue`` requires a target
3833 This rule redirects incoming UDPv4 traffic to queue index 6::
3835 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
3836 actions queue index 6 / end
3838 While this one could be rejected by PMDs (unspecified queue index)::
3840 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
3843 As defined by *rte_flow*, the list is not ordered, all actions of a given
3844 rule are performed simultaneously. These are equivalent::
3846 queue index 6 / void / mark id 42 / end
3850 void / mark id 42 / queue index 6 / end
3852 All actions in a list should have different types, otherwise only the last
3853 action of a given type is taken into account::
3855 queue index 4 / queue index 5 / queue index 6 / end # will use queue 6
3859 drop / drop / drop / end # drop is performed only once
3863 mark id 42 / queue index 3 / mark id 24 / end # mark will be 24
3865 Considering they are performed simultaneously, opposite and overlapping
3866 actions can sometimes be combined when the end result is unambiguous::
3868 drop / queue index 6 / end # drop has no effect
3872 queue index 6 / rss queues 6 7 8 / end # queue has no effect
3876 drop / passthru / end # drop has no effect
3878 Note that PMDs may still refuse such combinations.
3883 This section lists supported actions and their attributes, if any.
3885 - ``end``: end list of actions.
3887 - ``void``: no-op action.
3889 - ``passthru``: let subsequent rule process matched packets.
3891 - ``jump``: redirect traffic to group on device.
3893 - ``group {unsigned}``: group to redirect to.
3895 - ``mark``: attach 32 bit value to packets.
3897 - ``id {unsigned}``: 32 bit value to return with packets.
3899 - ``flag``: flag packets.
3901 - ``queue``: assign packets to a given queue index.
3903 - ``index {unsigned}``: queue index to use.
3905 - ``drop``: drop packets (note: passthru has priority).
3907 - ``count``: enable counters for this rule.
3909 - ``rss``: spread packets among several queues.
3911 - ``func {hash function}``: RSS hash function to apply, allowed tokens are
3912 ``toeplitz``, ``simple_xor``, ``symmetric_toeplitz`` and ``default``.
3914 - ``level {unsigned}``: encapsulation level for ``types``.
3916 - ``types [{RSS hash type} [...]] end``: specific RSS hash types.
3917 Note that an empty list does not disable RSS but instead requests
3918 unspecified "best-effort" settings.
3920 - ``key {string}``: RSS hash key, overrides ``key_len``.
3922 - ``key_len {unsigned}``: RSS hash key length in bytes, can be used in
3923 conjunction with ``key`` to pad or truncate it.
3925 - ``queues [{unsigned} [...]] end``: queue indices to use.
3927 - ``pf``: direct traffic to physical function.
3929 - ``vf``: direct traffic to a virtual function ID.
3931 - ``original {boolean}``: use original VF ID if possible.
3932 - ``id {unsigned}``: VF ID.
3934 - ``phy_port``: direct packets to physical port index.
3936 - ``original {boolean}``: use original port index if possible.
3937 - ``index {unsigned}``: physical port index.
3939 - ``port_id``: direct matching traffic to a given DPDK port ID.
3941 - ``original {boolean}``: use original DPDK port ID if possible.
3942 - ``id {unsigned}``: DPDK port ID.
3944 - ``of_set_mpls_ttl``: OpenFlow's ``OFPAT_SET_MPLS_TTL``.
3946 - ``mpls_ttl``: MPLS TTL.
3948 - ``of_dec_mpls_ttl``: OpenFlow's ``OFPAT_DEC_MPLS_TTL``.
3950 - ``of_set_nw_ttl``: OpenFlow's ``OFPAT_SET_NW_TTL``.
3952 - ``nw_ttl``: IP TTL.
3954 - ``of_dec_nw_ttl``: OpenFlow's ``OFPAT_DEC_NW_TTL``.
3956 - ``of_copy_ttl_out``: OpenFlow's ``OFPAT_COPY_TTL_OUT``.
3958 - ``of_copy_ttl_in``: OpenFlow's ``OFPAT_COPY_TTL_IN``.
3960 - ``of_pop_vlan``: OpenFlow's ``OFPAT_POP_VLAN``.
3962 - ``of_push_vlan``: OpenFlow's ``OFPAT_PUSH_VLAN``.
3964 - ``ethertype``: Ethertype.
3966 - ``of_set_vlan_vid``: OpenFlow's ``OFPAT_SET_VLAN_VID``.
3968 - ``vlan_vid``: VLAN id.
3970 - ``of_set_vlan_pcp``: OpenFlow's ``OFPAT_SET_VLAN_PCP``.
3972 - ``vlan_pcp``: VLAN priority.
3974 - ``of_pop_mpls``: OpenFlow's ``OFPAT_POP_MPLS``.
3976 - ``ethertype``: Ethertype.
3978 - ``of_push_mpls``: OpenFlow's ``OFPAT_PUSH_MPLS``.
3980 - ``ethertype``: Ethertype.
3982 - ``vxlan_encap``: Performs a VXLAN encapsulation, outer layer configuration
3983 is done through `Config VXLAN Encap outer layers`_.
3985 - ``vxlan_decap``: Performs a decapsulation action by stripping all headers of
3986 the VXLAN tunnel network overlay from the matched flow.
3988 - ``nvgre_encap``: Performs a NVGRE encapsulation, outer layer configuration
3989 is done through `Config NVGRE Encap outer layers`_.
3991 - ``nvgre_decap``: Performs a decapsulation action by stripping all headers of
3992 the NVGRE tunnel network overlay from the matched flow.
3994 - ``l2_encap``: Performs a L2 encapsulation, L2 configuration
3995 is done through `Config L2 Encap`_.
3997 - ``l2_decap``: Performs a L2 decapsulation, L2 configuration
3998 is done through `Config L2 Decap`_.
4000 - ``mplsogre_encap``: Performs a MPLSoGRE encapsulation, outer layer
4001 configuration is done through `Config MPLSoGRE Encap outer layers`_.
4003 - ``mplsogre_decap``: Performs a MPLSoGRE decapsulation, outer layer
4004 configuration is done through `Config MPLSoGRE Decap outer layers`_.
4006 - ``mplsoudp_encap``: Performs a MPLSoUDP encapsulation, outer layer
4007 configuration is done through `Config MPLSoUDP Encap outer layers`_.
4009 - ``mplsoudp_decap``: Performs a MPLSoUDP decapsulation, outer layer
4010 configuration is done through `Config MPLSoUDP Decap outer layers`_.
4012 - ``set_ipv4_src``: Set a new IPv4 source address in the outermost IPv4 header.
4014 - ``ipv4_addr``: New IPv4 source address.
4016 - ``set_ipv4_dst``: Set a new IPv4 destination address in the outermost IPv4
4019 - ``ipv4_addr``: New IPv4 destination address.
4021 - ``set_ipv6_src``: Set a new IPv6 source address in the outermost IPv6 header.
4023 - ``ipv6_addr``: New IPv6 source address.
4025 - ``set_ipv6_dst``: Set a new IPv6 destination address in the outermost IPv6
4028 - ``ipv6_addr``: New IPv6 destination address.
4030 - ``set_tp_src``: Set a new source port number in the outermost TCP/UDP
4033 - ``port``: New TCP/UDP source port number.
4035 - ``set_tp_dst``: Set a new destination port number in the outermost TCP/UDP
4038 - ``port``: New TCP/UDP destination port number.
4040 - ``mac_swap``: Swap the source and destination MAC addresses in the outermost
4043 - ``dec_ttl``: Performs a decrease TTL value action
4045 - ``set_ttl``: Set TTL value with specified value
4046 - ``ttl_value {unsigned}``: The new TTL value to be set
4048 - ``set_mac_src``: set source MAC address
4050 - ``mac_addr {MAC-48}``: new source MAC address
4052 - ``set_mac_dst``: set destination MAC address
4054 - ``mac_addr {MAC-48}``: new destination MAC address
4056 - ``inc_tcp_seq``: Increase sequence number in the outermost TCP header.
4058 - ``value {unsigned}``: Value to increase TCP sequence number by.
4060 - ``dec_tcp_seq``: Decrease sequence number in the outermost TCP header.
4062 - ``value {unsigned}``: Value to decrease TCP sequence number by.
4064 - ``inc_tcp_ack``: Increase acknowledgment number in the outermost TCP header.
4066 - ``value {unsigned}``: Value to increase TCP acknowledgment number by.
4068 - ``dec_tcp_ack``: Decrease acknowledgment number in the outermost TCP header.
4070 - ``value {unsigned}``: Value to decrease TCP acknowledgment number by.
4072 - ``set_ipv4_dscp``: Set IPv4 DSCP value with specified value
4074 - ``dscp_value {unsigned}``: The new DSCP value to be set
4076 - ``set_ipv6_dscp``: Set IPv6 DSCP value with specified value
4078 - ``dscp_value {unsigned}``: The new DSCP value to be set
4080 - ``indirect``: Use indirect action created via
4081 ``flow indirect_action {port_id} create``
4083 - ``indirect_action_id {unsigned}``: Indirect action ID to use
4085 Destroying flow rules
4086 ~~~~~~~~~~~~~~~~~~~~~
4088 ``flow destroy`` destroys one or more rules from their rule ID (as returned
4089 by ``flow create``), this command calls ``rte_flow_destroy()`` as many
4090 times as necessary::
4092 flow destroy {port_id} rule {rule_id} [...]
4094 If successful, it will show::
4096 Flow rule #[...] destroyed
4098 It does not report anything for rule IDs that do not exist. The usual error
4099 message is shown when a rule cannot be destroyed::
4101 Caught error type [...] ([...]): [...]
4103 ``flow flush`` destroys all rules on a device and does not take extra
4104 arguments. It is bound to ``rte_flow_flush()``::
4106 flow flush {port_id}
4108 Any errors are reported as above.
4110 Creating several rules and destroying them::
4112 testpmd> flow create 0 ingress pattern eth / ipv6 / end
4113 actions queue index 2 / end
4114 Flow rule #0 created
4115 testpmd> flow create 0 ingress pattern eth / ipv4 / end
4116 actions queue index 3 / end
4117 Flow rule #1 created
4118 testpmd> flow destroy 0 rule 0 rule 1
4119 Flow rule #1 destroyed
4120 Flow rule #0 destroyed
4123 The same result can be achieved using ``flow flush``::
4125 testpmd> flow create 0 ingress pattern eth / ipv6 / end
4126 actions queue index 2 / end
4127 Flow rule #0 created
4128 testpmd> flow create 0 ingress pattern eth / ipv4 / end
4129 actions queue index 3 / end
4130 Flow rule #1 created
4131 testpmd> flow flush 0
4134 Non-existent rule IDs are ignored::
4136 testpmd> flow create 0 ingress pattern eth / ipv6 / end
4137 actions queue index 2 / end
4138 Flow rule #0 created
4139 testpmd> flow create 0 ingress pattern eth / ipv4 / end
4140 actions queue index 3 / end
4141 Flow rule #1 created
4142 testpmd> flow destroy 0 rule 42 rule 10 rule 2
4144 testpmd> flow destroy 0 rule 0
4145 Flow rule #0 destroyed
4151 ``flow query`` queries a specific action of a flow rule having that
4152 ability. Such actions collect information that can be reported using this
4153 command. It is bound to ``rte_flow_query()``::
4155 flow query {port_id} {rule_id} {action}
4157 If successful, it will display either the retrieved data for known actions
4158 or the following message::
4160 Cannot display result for action type [...] ([...])
4162 Otherwise, it will complain either that the rule does not exist or that some
4165 Flow rule #[...] not found
4169 Caught error type [...] ([...]): [...]
4171 Currently only the ``count`` action is supported. This action reports the
4172 number of packets that hit the flow rule and the total number of bytes. Its
4173 output has the following format::
4176 hits_set: [...] # whether "hits" contains a valid value
4177 bytes_set: [...] # whether "bytes" contains a valid value
4178 hits: [...] # number of packets
4179 bytes: [...] # number of bytes
4181 Querying counters for TCPv6 packets redirected to queue 6::
4183 testpmd> flow create 0 ingress pattern eth / ipv6 / tcp / end
4184 actions queue index 6 / count / end
4185 Flow rule #4 created
4186 testpmd> flow query 0 4 count
4197 ``flow list`` lists existing flow rules sorted by priority and optionally
4198 filtered by group identifiers::
4200 flow list {port_id} [group {group_id}] [...]
4202 This command only fails with the following message if the device does not
4207 Output consists of a header line followed by a short description of each
4208 flow rule, one per line. There is no output at all when no flow rules are
4209 configured on the device::
4211 ID Group Prio Attr Rule
4212 [...] [...] [...] [...] [...]
4214 ``Attr`` column flags:
4216 - ``i`` for ``ingress``.
4217 - ``e`` for ``egress``.
4219 Creating several flow rules and listing them::
4221 testpmd> flow create 0 ingress pattern eth / ipv4 / end
4222 actions queue index 6 / end
4223 Flow rule #0 created
4224 testpmd> flow create 0 ingress pattern eth / ipv6 / end
4225 actions queue index 2 / end
4226 Flow rule #1 created
4227 testpmd> flow create 0 priority 5 ingress pattern eth / ipv4 / udp / end
4228 actions rss queues 6 7 8 end / end
4229 Flow rule #2 created
4230 testpmd> flow list 0
4231 ID Group Prio Attr Rule
4232 0 0 0 i- ETH IPV4 => QUEUE
4233 1 0 0 i- ETH IPV6 => QUEUE
4234 2 0 5 i- ETH IPV4 UDP => RSS
4237 Rules are sorted by priority (i.e. group ID first, then priority level)::
4239 testpmd> flow list 1
4240 ID Group Prio Attr Rule
4241 0 0 0 i- ETH => COUNT
4242 6 0 500 i- ETH IPV6 TCP => DROP COUNT
4243 5 0 1000 i- ETH IPV6 ICMP => QUEUE
4244 1 24 0 i- ETH IPV4 UDP => QUEUE
4245 4 24 10 i- ETH IPV4 TCP => DROP
4246 3 24 20 i- ETH IPV4 => DROP
4247 2 24 42 i- ETH IPV4 UDP => QUEUE
4248 7 63 0 i- ETH IPV6 UDP VXLAN => MARK QUEUE
4251 Output can be limited to specific groups::
4253 testpmd> flow list 1 group 0 group 63
4254 ID Group Prio Attr Rule
4255 0 0 0 i- ETH => COUNT
4256 6 0 500 i- ETH IPV6 TCP => DROP COUNT
4257 5 0 1000 i- ETH IPV6 ICMP => QUEUE
4258 7 63 0 i- ETH IPV6 UDP VXLAN => MARK QUEUE
4261 Toggling isolated mode
4262 ~~~~~~~~~~~~~~~~~~~~~~
4264 ``flow isolate`` can be used to tell the underlying PMD that ingress traffic
4265 must only be injected from the defined flow rules; that no default traffic
4266 is expected outside those rules and the driver is free to assign more
4267 resources to handle them. It is bound to ``rte_flow_isolate()``::
4269 flow isolate {port_id} {boolean}
4271 If successful, enabling or disabling isolated mode shows either::
4273 Ingress traffic on port [...]
4274 is now restricted to the defined flow rules
4278 Ingress traffic on port [...]
4279 is not restricted anymore to the defined flow rules
4281 Otherwise, in case of error::
4283 Caught error type [...] ([...]): [...]
4285 Mainly due to its side effects, PMDs supporting this mode may not have the
4286 ability to toggle it more than once without reinitializing affected ports
4287 first (e.g. by exiting testpmd).
4289 Enabling isolated mode::
4291 testpmd> flow isolate 0 true
4292 Ingress traffic on port 0 is now restricted to the defined flow rules
4295 Disabling isolated mode::
4297 testpmd> flow isolate 0 false
4298 Ingress traffic on port 0 is not restricted anymore to the defined flow rules
4301 Dumping HW internal information
4302 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4304 ``flow dump`` dumps the hardware's internal representation information of
4305 all flows. It is bound to ``rte_flow_dev_dump()``::
4307 flow dump {port_id} {output_file}
4309 If successful, it will show::
4313 Otherwise, it will complain error occurred::
4315 Caught error type [...] ([...]): [...]
4317 Listing and destroying aged flow rules
4318 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4320 ``flow aged`` simply lists aged flow rules be get from api ``rte_flow_get_aged_flows``,
4321 and ``destroy`` parameter can be used to destroy those flow rules in PMD.
4323 flow aged {port_id} [destroy]
4325 Listing current aged flow rules::
4327 testpmd> flow aged 0
4328 Port 0 total aged flows: 0
4329 testpmd> flow create 0 ingress pattern eth / ipv4 src is 2.2.2.14 / end
4330 actions age timeout 5 / queue index 0 / end
4331 Flow rule #0 created
4332 testpmd> flow create 0 ingress pattern eth / ipv4 src is 2.2.2.15 / end
4333 actions age timeout 4 / queue index 0 / end
4334 Flow rule #1 created
4335 testpmd> flow create 0 ingress pattern eth / ipv4 src is 2.2.2.16 / end
4336 actions age timeout 2 / queue index 0 / end
4337 Flow rule #2 created
4338 testpmd> flow create 0 ingress pattern eth / ipv4 src is 2.2.2.17 / end
4339 actions age timeout 3 / queue index 0 / end
4340 Flow rule #3 created
4343 Aged Rules are simply list as command ``flow list {port_id}``, but strip the detail rule
4344 information, all the aged flows are sorted by the longest timeout time. For example, if
4345 those rules be configured in the same time, ID 2 will be the first aged out rule, the next
4346 will be ID 3, ID 1, ID 0::
4348 testpmd> flow aged 0
4349 Port 0 total aged flows: 4
4356 If attach ``destroy`` parameter, the command will destroy all the list aged flow rules.
4358 testpmd> flow aged 0 destroy
4359 Port 0 total aged flows: 4
4366 Flow rule #2 destroyed
4367 Flow rule #3 destroyed
4368 Flow rule #1 destroyed
4369 Flow rule #0 destroyed
4370 4 flows be destroyed
4371 testpmd> flow aged 0
4372 Port 0 total aged flows: 0
4374 Creating indirect actions
4375 ~~~~~~~~~~~~~~~~~~~~~~~~~
4377 ``flow indirect_action {port_id} create`` creates indirect action with optional
4378 indirect action ID. It is bound to ``rte_flow_action_handle_create()``::
4380 flow indirect_action {port_id} create [action_id {indirect_action_id}]
4381 [ingress] [egress] [transfer] action {action} / end
4383 If successful, it will show::
4385 Indirect action #[...] created
4387 Otherwise, it will complain either that indirect action already exists or that
4388 some error occurred::
4390 Indirect action #[...] is already assigned, delete it first
4394 Caught error type [...] ([...]): [...]
4396 Create indirect rss action with id 100 to queues 1 and 2 on port 0::
4398 testpmd> flow indirect_action 0 create action_id 100 \
4399 ingress action rss queues 1 2 end / end
4401 Create indirect rss action with id assigned by testpmd to queues 1 and 2 on
4404 testpmd> flow indirect_action 0 create action_id \
4405 ingress action rss queues 0 1 end / end
4407 Updating indirect actions
4408 ~~~~~~~~~~~~~~~~~~~~~~~~~
4410 ``flow indirect_action {port_id} update`` updates configuration of the indirect
4411 action from its indirect action ID (as returned by
4412 ``flow indirect_action {port_id} create``). It is bound to
4413 ``rte_flow_action_handle_update()``::
4415 flow indirect_action {port_id} update {indirect_action_id}
4416 action {action} / end
4418 If successful, it will show::
4420 Indirect action #[...] updated
4422 Otherwise, it will complain either that indirect action not found or that some
4425 Failed to find indirect action #[...] on port [...]
4429 Caught error type [...] ([...]): [...]
4431 Update indirect rss action having id 100 on port 0 with rss to queues 0 and 3
4432 (in create example above rss queues were 1 and 2)::
4434 testpmd> flow indirect_action 0 update 100 action rss queues 0 3 end / end
4436 Destroying indirect actions
4437 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
4439 ``flow indirect_action {port_id} destroy`` destroys one or more indirect actions
4440 from their indirect action IDs (as returned by
4441 ``flow indirect_action {port_id} create``). It is bound to
4442 ``rte_flow_action_handle_destroy()``::
4444 flow indirect_action {port_id} destroy action_id {indirect_action_id} [...]
4446 If successful, it will show::
4448 Indirect action #[...] destroyed
4450 It does not report anything for indirect action IDs that do not exist.
4451 The usual error message is shown when a indirect action cannot be destroyed::
4453 Caught error type [...] ([...]): [...]
4455 Destroy indirect actions having id 100 & 101::
4457 testpmd> flow indirect_action 0 destroy action_id 100 action_id 101
4459 Query indirect actions
4460 ~~~~~~~~~~~~~~~~~~~~~~
4462 ``flow indirect_action {port_id} query`` queries the indirect action from its
4463 indirect action ID (as returned by ``flow indirect_action {port_id} create``).
4464 It is bound to ``rte_flow_action_handle_query()``::
4466 flow indirect_action {port_id} query {indirect_action_id}
4468 Currently only rss indirect action supported. If successful, it will show::
4470 Indirect RSS action:
4473 Otherwise, it will complain either that indirect action not found or that some
4476 Failed to find indirect action #[...] on port [...]
4480 Caught error type [...] ([...]): [...]
4482 Query indirect action having id 100::
4484 testpmd> flow indirect_action 0 query 100
4486 Sample QinQ flow rules
4487 ~~~~~~~~~~~~~~~~~~~~~~
4489 Before creating QinQ rule(s) the following commands should be issued to enable QinQ::
4491 testpmd> port stop 0
4492 testpmd> vlan set extend on 0
4494 The above command sets the inner and outer TPID's to 0x8100.
4496 To change the TPID's the following commands should be used::
4498 testpmd> vlan set outer tpid 0x88A8 0
4499 testpmd> vlan set inner tpid 0x8100 0
4500 testpmd> port start 0
4502 Validate and create a QinQ rule on port 0 to steer traffic to a VF queue in a VM.
4506 testpmd> flow validate 0 ingress pattern eth / vlan tci is 123 /
4507 vlan tci is 456 / end actions vf id 1 / queue index 0 / end
4508 Flow rule #0 validated
4510 testpmd> flow create 0 ingress pattern eth / vlan tci is 4 /
4511 vlan tci is 456 / end actions vf id 123 / queue index 0 / end
4512 Flow rule #0 created
4514 testpmd> flow list 0
4515 ID Group Prio Attr Rule
4516 0 0 0 i- ETH VLAN VLAN=>VF QUEUE
4518 Validate and create a QinQ rule on port 0 to steer traffic to a queue on the host.
4522 testpmd> flow validate 0 ingress pattern eth / vlan tci is 321 /
4523 vlan tci is 654 / end actions pf / queue index 0 / end
4524 Flow rule #1 validated
4526 testpmd> flow create 0 ingress pattern eth / vlan tci is 321 /
4527 vlan tci is 654 / end actions pf / queue index 1 / end
4528 Flow rule #1 created
4530 testpmd> flow list 0
4531 ID Group Prio Attr Rule
4532 0 0 0 i- ETH VLAN VLAN=>VF QUEUE
4533 1 0 0 i- ETH VLAN VLAN=>PF QUEUE
4535 Sample VXLAN flow rules
4536 ~~~~~~~~~~~~~~~~~~~~~~~
4538 Before creating VXLAN rule(s), the UDP port should be added for VXLAN packet
4541 testpmd> rx_vxlan_port add 4789 0
4543 Create VXLAN rules on port 0 to steer traffic to PF queues.
4547 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / vxlan /
4548 eth dst is 00:11:22:33:44:55 / end actions pf / queue index 1 / end
4549 Flow rule #0 created
4551 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / vxlan vni is 3 /
4552 eth dst is 00:11:22:33:44:55 / end actions pf / queue index 2 / end
4553 Flow rule #1 created
4555 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / vxlan /
4556 eth dst is 00:11:22:33:44:55 / vlan tci is 10 / end actions pf /
4558 Flow rule #2 created
4560 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / vxlan vni is 5 /
4561 eth dst is 00:11:22:33:44:55 / vlan tci is 20 / end actions pf /
4563 Flow rule #3 created
4565 testpmd> flow create 0 ingress pattern eth dst is 00:00:00:00:01:00 / ipv4 /
4566 udp / vxlan vni is 6 / eth dst is 00:11:22:33:44:55 / end actions pf /
4568 Flow rule #4 created
4570 testpmd> flow list 0
4571 ID Group Prio Attr Rule
4572 0 0 0 i- ETH IPV4 UDP VXLAN ETH => QUEUE
4573 1 0 0 i- ETH IPV4 UDP VXLAN ETH => QUEUE
4574 2 0 0 i- ETH IPV4 UDP VXLAN ETH VLAN => QUEUE
4575 3 0 0 i- ETH IPV4 UDP VXLAN ETH VLAN => QUEUE
4576 4 0 0 i- ETH IPV4 UDP VXLAN ETH => QUEUE
4578 Sample VXLAN encapsulation rule
4579 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4581 VXLAN encapsulation outer layer has default value pre-configured in testpmd
4582 source code, those can be changed by using the following commands
4584 IPv4 VXLAN outer header::
4586 testpmd> set vxlan ip-version ipv4 vni 4 udp-src 4 udp-dst 4 ip-src 127.0.0.1
4587 ip-dst 128.0.0.1 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4588 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4591 testpmd> set vxlan-with-vlan ip-version ipv4 vni 4 udp-src 4 udp-dst 4 ip-src
4592 127.0.0.1 ip-dst 128.0.0.1 vlan-tci 34 eth-src 11:11:11:11:11:11
4593 eth-dst 22:22:22:22:22:22
4594 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4597 testpmd> set vxlan-tos-ttl ip-version ipv4 vni 4 udp-src 4 udp-dst 4 ip-tos 0
4598 ip-ttl 255 ip-src 127.0.0.1 ip-dst 128.0.0.1 eth-src 11:11:11:11:11:11
4599 eth-dst 22:22:22:22:22:22
4600 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4603 IPv6 VXLAN outer header::
4605 testpmd> set vxlan ip-version ipv6 vni 4 udp-src 4 udp-dst 4 ip-src ::1
4606 ip-dst ::2222 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4607 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4610 testpmd> set vxlan-with-vlan ip-version ipv6 vni 4 udp-src 4 udp-dst 4
4611 ip-src ::1 ip-dst ::2222 vlan-tci 34 eth-src 11:11:11:11:11:11
4612 eth-dst 22:22:22:22:22:22
4613 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4616 testpmd> set vxlan-tos-ttl ip-version ipv6 vni 4 udp-src 4 udp-dst 4
4617 ip-tos 0 ip-ttl 255 ::1 ip-dst ::2222 eth-src 11:11:11:11:11:11
4618 eth-dst 22:22:22:22:22:22
4619 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4622 Sample NVGRE encapsulation rule
4623 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4625 NVGRE encapsulation outer layer has default value pre-configured in testpmd
4626 source code, those can be changed by using the following commands
4628 IPv4 NVGRE outer header::
4630 testpmd> set nvgre ip-version ipv4 tni 4 ip-src 127.0.0.1 ip-dst 128.0.0.1
4631 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4632 testpmd> flow create 0 ingress pattern end actions nvgre_encap /
4635 testpmd> set nvgre-with-vlan ip-version ipv4 tni 4 ip-src 127.0.0.1
4636 ip-dst 128.0.0.1 vlan-tci 34 eth-src 11:11:11:11:11:11
4637 eth-dst 22:22:22:22:22:22
4638 testpmd> flow create 0 ingress pattern end actions nvgre_encap /
4641 IPv6 NVGRE outer header::
4643 testpmd> set nvgre ip-version ipv6 tni 4 ip-src ::1 ip-dst ::2222
4644 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4645 testpmd> flow create 0 ingress pattern end actions nvgre_encap /
4648 testpmd> set nvgre-with-vlan ip-version ipv6 tni 4 ip-src ::1 ip-dst ::2222
4649 vlan-tci 34 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4650 testpmd> flow create 0 ingress pattern end actions nvgre_encap /
4653 Sample L2 encapsulation rule
4654 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4656 L2 encapsulation has default value pre-configured in testpmd
4657 source code, those can be changed by using the following commands
4661 testpmd> set l2_encap ip-version ipv4
4662 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4663 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / mpls / end actions
4664 mplsoudp_decap / l2_encap / end
4666 L2 with VXLAN header::
4668 testpmd> set l2_encap-with-vlan ip-version ipv4 vlan-tci 34
4669 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4670 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / mpls / end actions
4671 mplsoudp_decap / l2_encap / end
4673 Sample L2 decapsulation rule
4674 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4676 L2 decapsulation has default value pre-configured in testpmd
4677 source code, those can be changed by using the following commands
4681 testpmd> set l2_decap
4682 testpmd> flow create 0 egress pattern eth / end actions l2_decap / mplsoudp_encap /
4685 L2 with VXLAN header::
4687 testpmd> set l2_encap-with-vlan
4688 testpmd> flow create 0 egress pattern eth / end actions l2_encap / mplsoudp_encap /
4691 Sample MPLSoGRE encapsulation rule
4692 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4694 MPLSoGRE encapsulation outer layer has default value pre-configured in testpmd
4695 source code, those can be changed by using the following commands
4697 IPv4 MPLSoGRE outer header::
4699 testpmd> set mplsogre_encap ip-version ipv4 label 4
4700 ip-src 127.0.0.1 ip-dst 128.0.0.1 eth-src 11:11:11:11:11:11
4701 eth-dst 22:22:22:22:22:22
4702 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4703 mplsogre_encap / end
4705 IPv4 MPLSoGRE with VLAN outer header::
4707 testpmd> set mplsogre_encap-with-vlan ip-version ipv4 label 4
4708 ip-src 127.0.0.1 ip-dst 128.0.0.1 vlan-tci 34
4709 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4710 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4711 mplsogre_encap / end
4713 IPv6 MPLSoGRE outer header::
4715 testpmd> set mplsogre_encap ip-version ipv6 mask 4
4716 ip-src ::1 ip-dst ::2222 eth-src 11:11:11:11:11:11
4717 eth-dst 22:22:22:22:22:22
4718 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4719 mplsogre_encap / end
4721 IPv6 MPLSoGRE with VLAN outer header::
4723 testpmd> set mplsogre_encap-with-vlan ip-version ipv6 mask 4
4724 ip-src ::1 ip-dst ::2222 vlan-tci 34
4725 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4726 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4727 mplsogre_encap / end
4729 Sample MPLSoGRE decapsulation rule
4730 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4732 MPLSoGRE decapsulation outer layer has default value pre-configured in testpmd
4733 source code, those can be changed by using the following commands
4735 IPv4 MPLSoGRE outer header::
4737 testpmd> set mplsogre_decap ip-version ipv4
4738 testpmd> flow create 0 ingress pattern eth / ipv4 / gre / mpls / end actions
4739 mplsogre_decap / l2_encap / end
4741 IPv4 MPLSoGRE with VLAN outer header::
4743 testpmd> set mplsogre_decap-with-vlan ip-version ipv4
4744 testpmd> flow create 0 ingress pattern eth / vlan / ipv4 / gre / mpls / end
4745 actions mplsogre_decap / l2_encap / end
4747 IPv6 MPLSoGRE outer header::
4749 testpmd> set mplsogre_decap ip-version ipv6
4750 testpmd> flow create 0 ingress pattern eth / ipv6 / gre / mpls / end
4751 actions mplsogre_decap / l2_encap / end
4753 IPv6 MPLSoGRE with VLAN outer header::
4755 testpmd> set mplsogre_decap-with-vlan ip-version ipv6
4756 testpmd> flow create 0 ingress pattern eth / vlan / ipv6 / gre / mpls / end
4757 actions mplsogre_decap / l2_encap / end
4759 Sample MPLSoUDP encapsulation rule
4760 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4762 MPLSoUDP encapsulation outer layer has default value pre-configured in testpmd
4763 source code, those can be changed by using the following commands
4765 IPv4 MPLSoUDP outer header::
4767 testpmd> set mplsoudp_encap ip-version ipv4 label 4 udp-src 5 udp-dst 10
4768 ip-src 127.0.0.1 ip-dst 128.0.0.1 eth-src 11:11:11:11:11:11
4769 eth-dst 22:22:22:22:22:22
4770 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4771 mplsoudp_encap / end
4773 IPv4 MPLSoUDP with VLAN outer header::
4775 testpmd> set mplsoudp_encap-with-vlan ip-version ipv4 label 4 udp-src 5
4776 udp-dst 10 ip-src 127.0.0.1 ip-dst 128.0.0.1 vlan-tci 34
4777 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4778 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4779 mplsoudp_encap / end
4781 IPv6 MPLSoUDP outer header::
4783 testpmd> set mplsoudp_encap ip-version ipv6 mask 4 udp-src 5 udp-dst 10
4784 ip-src ::1 ip-dst ::2222 eth-src 11:11:11:11:11:11
4785 eth-dst 22:22:22:22:22:22
4786 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4787 mplsoudp_encap / end
4789 IPv6 MPLSoUDP with VLAN outer header::
4791 testpmd> set mplsoudp_encap-with-vlan ip-version ipv6 mask 4 udp-src 5
4792 udp-dst 10 ip-src ::1 ip-dst ::2222 vlan-tci 34
4793 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4794 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4795 mplsoudp_encap / end
4797 Sample MPLSoUDP decapsulation rule
4798 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4800 MPLSoUDP decapsulation outer layer has default value pre-configured in testpmd
4801 source code, those can be changed by using the following commands
4803 IPv4 MPLSoUDP outer header::
4805 testpmd> set mplsoudp_decap ip-version ipv4
4806 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / mpls / end actions
4807 mplsoudp_decap / l2_encap / end
4809 IPv4 MPLSoUDP with VLAN outer header::
4811 testpmd> set mplsoudp_decap-with-vlan ip-version ipv4
4812 testpmd> flow create 0 ingress pattern eth / vlan / ipv4 / udp / mpls / end
4813 actions mplsoudp_decap / l2_encap / end
4815 IPv6 MPLSoUDP outer header::
4817 testpmd> set mplsoudp_decap ip-version ipv6
4818 testpmd> flow create 0 ingress pattern eth / ipv6 / udp / mpls / end
4819 actions mplsoudp_decap / l2_encap / end
4821 IPv6 MPLSoUDP with VLAN outer header::
4823 testpmd> set mplsoudp_decap-with-vlan ip-version ipv6
4824 testpmd> flow create 0 ingress pattern eth / vlan / ipv6 / udp / mpls / end
4825 actions mplsoudp_decap / l2_encap / end
4827 Sample Raw encapsulation rule
4828 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4830 Raw encapsulation configuration can be set by the following commands
4832 Eecapsulating VxLAN::
4834 testpmd> set raw_encap 4 eth src is 10:11:22:33:44:55 / vlan tci is 1
4835 inner_type is 0x0800 / ipv4 / udp dst is 4789 / vxlan vni
4837 testpmd> flow create 0 egress pattern eth / ipv4 / end actions
4838 raw_encap index 4 / end
4840 Sample Raw decapsulation rule
4841 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4843 Raw decapsulation configuration can be set by the following commands
4845 Decapsulating VxLAN::
4847 testpmd> set raw_decap eth / ipv4 / udp / vxlan / end_set
4848 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / vxlan / eth / ipv4 /
4849 end actions raw_decap / queue index 0 / end
4854 ESP rules can be created by the following commands::
4856 testpmd> flow create 0 ingress pattern eth / ipv4 / esp spi is 1 / end actions
4858 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / esp spi is 1 / end
4859 actions queue index 3 / end
4860 testpmd> flow create 0 ingress pattern eth / ipv6 / esp spi is 1 / end actions
4862 testpmd> flow create 0 ingress pattern eth / ipv6 / udp / esp spi is 1 / end
4863 actions queue index 3 / end
4868 AH rules can be created by the following commands::
4870 testpmd> flow create 0 ingress pattern eth / ipv4 / ah spi is 1 / end actions
4872 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / ah spi is 1 / end
4873 actions queue index 3 / end
4874 testpmd> flow create 0 ingress pattern eth / ipv6 / ah spi is 1 / end actions
4876 testpmd> flow create 0 ingress pattern eth / ipv6 / udp / ah spi is 1 / end
4877 actions queue index 3 / end
4882 PFCP rules can be created by the following commands(s_field need to be 1
4885 testpmd> flow create 0 ingress pattern eth / ipv4 / pfcp s_field is 0 / end
4886 actions queue index 3 / end
4887 testpmd> flow create 0 ingress pattern eth / ipv4 / pfcp s_field is 1
4888 seid is 1 / end actions queue index 3 / end
4889 testpmd> flow create 0 ingress pattern eth / ipv6 / pfcp s_field is 0 / end
4890 actions queue index 3 / end
4891 testpmd> flow create 0 ingress pattern eth / ipv6 / pfcp s_field is 1
4892 seid is 1 / end actions queue index 3 / end
4894 Sample Sampling/Mirroring rules
4895 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4897 Sample/Mirroring rules can be set by the following commands
4899 NIC-RX Sampling rule, the matched ingress packets and sent to the queue 1,
4900 and 50% packets are duplicated and marked with 0x1234 and sent to queue 0.
4904 testpmd> set sample_actions 0 mark id 0x1234 / queue index 0 / end
4905 testpmd> flow create 0 ingress group 1 pattern eth / end actions
4906 sample ratio 2 index 0 / queue index 1 / end
4908 Mirroring rule with port representors (with "transfer" attribute), the matched
4909 ingress packets with encapsulation header are sent to port id 0, and also
4910 mirrored the packets and sent to port id 2.
4914 testpmd> set sample_actions 0 port_id id 2 / end
4915 testpmd> flow create 1 ingress transfer pattern eth / end actions
4916 sample ratio 1 index 0 / raw_encap / port_id id 0 / end
4918 Mirroring rule with port representors (with "transfer" attribute), the matched
4919 ingress packets are sent to port id 2, and also mirrored the packets with
4920 encapsulation header and sent to port id 0.
4924 testpmd> set sample_actions 0 raw_encap / port_id id 0 / end
4925 testpmd> flow create 0 ingress transfer pattern eth / end actions
4926 sample ratio 1 index 0 / port_id id 2 / end
4928 Mirroring rule with port representors (with "transfer" attribute), the matched
4929 ingress packets are sent to port id 2, and also mirrored the packets with
4930 VXLAN encapsulation header and sent to port id 0.
4934 testpmd> set vxlan ip-version ipv4 vni 4 udp-src 4 udp-dst 4 ip-src 127.0.0.1
4935 ip-dst 128.0.0.1 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4936 testpmd> set sample_actions 0 vxlan_encap / port_id id 0 / end
4937 testpmd> flow create 0 ingress transfer pattern eth / end actions
4938 sample ratio 1 index 0 / port_id id 2 / end
4940 Mirroring rule with port representors (with "transfer" attribute), the matched
4941 ingress packets are sent to port id 2, and also mirrored the packets with
4942 NVGRE encapsulation header and sent to port id 0.
4946 testpmd> set nvgre ip-version ipv4 tni 4 ip-src 127.0.0.1 ip-dst 128.0.0.1
4947 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4948 testpmd> set sample_actions 0 nvgre_encap / port_id id 0 / end
4949 testpmd> flow create 0 ingress transfer pattern eth / end actions
4950 sample ratio 1 index 0 / port_id id 2 / end
4952 Sample integrity rules
4953 ~~~~~~~~~~~~~~~~~~~~~~
4955 Integrity rules can be created by the following commands:
4957 Integrity rule that forwards valid TCP packets to group 1.
4958 TCP packet integrity is matched with the ``l4_ok`` bit 3.
4962 testpmd> flow create 0 ingress
4963 pattern eth / ipv4 / tcp / integrity value mask 8 value spec 8 / end
4964 actions jump group 1 / end
4966 Integrity rule that forwards invalid packets to application.
4967 General packet integrity is matched with the ``packet_ok`` bit 0.
4971 testpmd> flow create 0 ingress pattern integrity value mask 1 value spec 0 / end actions queue index 0 / end
4973 Sample conntrack rules
4974 ~~~~~~~~~~~~~~~~~~~~~~
4976 Conntrack rules can be set by the following commands
4978 Need to construct the connection context with provided information.
4979 In the first table, create a flow rule by using conntrack action and jump to
4980 the next table. In the next table, create a rule to check the state.
4984 testpmd> set conntrack com peer 1 is_orig 1 enable 1 live 1 sack 1 cack 0
4985 last_dir 0 liberal 0 state 1 max_ack_win 7 r_lim 5 last_win 510
4986 last_seq 2632987379 last_ack 2532480967 last_end 2632987379
4988 testpmd> set conntrack orig scale 7 fin 0 acked 1 unack_data 0
4989 sent_end 2632987379 reply_end 2633016339 max_win 28960
4991 testpmd> set conntrack rply scale 7 fin 0 acked 1 unack_data 0
4992 sent_end 2532480967 reply_end 2532546247 max_win 65280
4994 testpmd> flow indirect_action 0 create ingress action conntrack / end
4995 testpmd> flow create 0 group 3 ingress pattern eth / ipv4 / tcp / end actions indirect 0 / jump group 5 / end
4996 testpmd> flow create 0 group 5 ingress pattern eth / ipv4 / tcp / conntrack is 1 / end actions queue index 5 / end
4998 Construct the conntrack again with only "is_orig" set to 0 (other fields are
4999 ignored), then use "update" interface to update the direction. Create flow
5000 rules like above for the peer port.
5004 testpmd> flow indirect_action 0 update 0 action conntrack_update dir / end
5009 The following sections show functions to load/unload eBPF based filters.
5014 Load an eBPF program as a callback for particular RX/TX queue::
5016 testpmd> bpf-load rx|tx (portid) (queueid) (load-flags) (bpf-prog-filename)
5018 The available load-flags are:
5020 * ``J``: use JIT generated native code, otherwise BPF interpreter will be used.
5022 * ``M``: assume input parameter is a pointer to rte_mbuf, otherwise assume it is a pointer to first segment's data.
5028 You'll need clang v3.7 or above to build bpf program you'd like to load
5032 .. code-block:: console
5035 clang -O2 -target bpf -c t1.c
5037 Then to load (and JIT compile) t1.o at RX queue 0, port 1:
5039 .. code-block:: console
5041 testpmd> bpf-load rx 1 0 J ./dpdk.org/examples/bpf/t1.o
5043 To load (not JITed) t1.o at TX queue 0, port 0:
5045 .. code-block:: console
5047 testpmd> bpf-load tx 0 0 - ./dpdk.org/examples/bpf/t1.o
5052 Unload previously loaded eBPF program for particular RX/TX queue::
5054 testpmd> bpf-unload rx|tx (portid) (queueid)
5056 For example to unload BPF filter from TX queue 0, port 0:
5058 .. code-block:: console
5060 testpmd> bpf-unload tx 0 0