1 .. SPDX-License-Identifier: BSD-3-Clause
2 Copyright(c) 2010-2016 Intel Corporation.
6 Testpmd Runtime Functions
7 =========================
9 Where the testpmd application is started in interactive mode, (``-i|--interactive``),
10 it displays a prompt that can be used to start and stop forwarding,
11 configure the application, display statistics (including the extended NIC
12 statistics aka xstats) , set the Flow Director and other tasks::
16 The testpmd prompt has some, limited, readline support.
17 Common bash command-line functions such as ``Ctrl+a`` and ``Ctrl+e`` to go to the start and end of the prompt line are supported
18 as well as access to the command history via the up-arrow.
20 There is also support for tab completion.
21 If you type a partial command and hit ``<TAB>`` you get a list of the available completions:
23 .. code-block:: console
25 testpmd> show port <TAB>
27 info [Mul-choice STRING]: show|clear port info|stats|xstats|fdir|dcb_tc|cap X
28 info [Mul-choice STRING]: show|clear port info|stats|xstats|fdir|dcb_tc|cap all
29 stats [Mul-choice STRING]: show|clear port info|stats|xstats|fdir|dcb_tc|cap X
30 stats [Mul-choice STRING]: show|clear port info|stats|xstats|fdir|dcb_tc|cap all
36 Some examples in this document are too long to fit on one line are shown wrapped at `"\\"` for display purposes::
38 testpmd> set flow_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \
39 (pause_time) (send_xon) (port_id)
41 In the real ``testpmd>`` prompt these commands should be on a single line.
46 The testpmd has on-line help for the functions that are available at runtime.
47 These are divided into sections and can be accessed using help, help section or help all:
49 .. code-block:: console
53 help control : Start and stop forwarding.
54 help display : Displaying port, stats and config information.
55 help config : Configuration information.
56 help ports : Configuring ports.
57 help registers : Reading and setting port registers.
58 help filters : Filters configuration help.
59 help all : All of the above sections.
62 Command File Functions
63 ----------------------
65 To facilitate loading large number of commands or to avoid cutting and pasting where not
66 practical or possible testpmd supports alternative methods for executing commands.
68 * If started with the ``--cmdline-file=FILENAME`` command line argument testpmd
69 will execute all CLI commands contained within the file immediately before
70 starting packet forwarding or entering interactive mode.
72 .. code-block:: console
74 ./dpdk-testpmd -n4 -r2 ... -- -i --cmdline-file=/home/ubuntu/flow-create-commands.txt
75 Interactive-mode selected
76 CLI commands to be read from /home/ubuntu/flow-create-commands.txt
77 Configuring Port 0 (socket 0)
78 Port 0: 7C:FE:90:CB:74:CE
79 Configuring Port 1 (socket 0)
80 Port 1: 7C:FE:90:CB:74:CA
81 Checking link statuses...
82 Port 0 Link Up - speed 10000 Mbps - full-duplex
83 Port 1 Link Up - speed 10000 Mbps - full-duplex
89 Flow rule #498 created
90 Flow rule #499 created
91 Read all CLI commands from /home/ubuntu/flow-create-commands.txt
95 * At run-time additional commands can be loaded in bulk by invoking the ``load FILENAME``
98 .. code-block:: console
100 testpmd> load /home/ubuntu/flow-create-commands.txt
105 Flow rule #498 created
106 Flow rule #499 created
107 Read all CLI commands from /home/ubuntu/flow-create-commands.txt
111 In all cases output from any included command will be displayed as standard output.
112 Execution will continue until the end of the file is reached regardless of
113 whether any errors occur. The end user must examine the output to determine if
114 any failures occurred.
123 Start packet forwarding with current configuration::
130 Start packet forwarding with current configuration after sending specified number of bursts of packets::
132 testpmd> start tx_first (""|burst_num)
134 The default burst number is 1 when ``burst_num`` not presented.
139 Stop packet forwarding, and display accumulated statistics::
154 The functions in the following sections are used to display information about the
155 testpmd configuration or the NIC status.
160 Display information for a given port or all ports::
162 testpmd> show port (info|summary|stats|xstats|fdir|dcb_tc|cap) (port_id|all)
164 The available information categories are:
166 * ``info``: General port information such as MAC address.
168 * ``summary``: Brief port summary such as Device Name, Driver Name etc.
170 * ``stats``: RX/TX statistics.
172 * ``xstats``: RX/TX extended NIC statistics.
174 * ``fdir``: Flow Director information and statistics.
176 * ``dcb_tc``: DCB information such as TC mapping.
180 .. code-block:: console
182 testpmd> show port info 0
184 ********************* Infos for port 0 *********************
186 MAC address: XX:XX:XX:XX:XX:XX
188 memory allocation on the socket: 0
190 Link speed: 40000 Mbps
191 Link duplex: full-duplex
192 Promiscuous mode: enabled
193 Allmulticast mode: disabled
194 Maximum number of MAC addresses: 64
195 Maximum number of MAC addresses of hash filtering: 0
197 strip on, filter on, extend off, qinq strip off
198 Redirection table size: 512
199 Supported flow types:
217 show port (module_eeprom|eeprom)
218 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
220 Display the EEPROM information of a port::
222 testpmd> show port (port_id) (module_eeprom|eeprom)
227 Display the rss redirection table entry indicated by masks on port X::
229 testpmd> show port (port_id) rss reta (size) (mask0, mask1...)
231 size is used to indicate the hardware supported reta size
236 Display the RSS hash functions and RSS hash key of a port::
238 testpmd> show port (port_id) rss-hash [key]
243 Clear the port statistics and forward engine statistics for a given port or for all ports::
245 testpmd> clear port (info|stats|xstats|fdir) (port_id|all)
249 testpmd> clear port stats all
254 Display information for a given port's RX/TX queue::
256 testpmd> show (rxq|txq) info (port_id) (queue_id)
258 show desc status(rxq|txq)
259 ~~~~~~~~~~~~~~~~~~~~~~~~~
261 Display information for a given port's RX/TX descriptor status::
263 testpmd> show port (port_id) (rxq|txq) (queue_id) desc (desc_id) status
265 show rxq desc used count
266 ~~~~~~~~~~~~~~~~~~~~~~~~
268 Display the number of receive packet descriptors currently filled by hardware
269 and ready to be processed by the driver on a given RX queue::
271 testpmd> show port (port_id) rxq (queue_id) desc used count
276 Displays the configuration of the application.
277 The configuration comes from the command-line, the runtime or the application defaults::
279 testpmd> show config (rxtx|cores|fwd|rxoffs|rxpkts|txpkts|txtimes)
281 The available information categories are:
283 * ``rxtx``: RX/TX configuration items.
285 * ``cores``: List of forwarding cores.
287 * ``fwd``: Packet forwarding configuration.
289 * ``rxoffs``: Packet offsets for RX split.
291 * ``rxpkts``: Packets to RX split configuration.
293 * ``txpkts``: Packets to TX configuration.
295 * ``txtimes``: Burst time pattern for Tx only mode.
299 .. code-block:: console
301 testpmd> show config rxtx
303 io packet forwarding - CRC stripping disabled - packets/burst=16
304 nb forwarding cores=2 - nb forwarding ports=1
305 RX queues=1 - RX desc=128 - RX free threshold=0
306 RX threshold registers: pthresh=8 hthresh=8 wthresh=4
307 TX queues=1 - TX desc=512 - TX free threshold=0
308 TX threshold registers: pthresh=36 hthresh=0 wthresh=0
309 TX RS bit threshold=0 - TXQ flags=0x0
314 Set the packet forwarding mode::
316 testpmd> set fwd (io|mac|macswap|flowgen| \
317 rxonly|txonly|csum|icmpecho|noisy|5tswap) (""|retry)
319 ``retry`` can be specified for forwarding engines except ``rx_only``.
321 The available information categories are:
323 * ``io``: Forwards packets "as-is" in I/O mode.
324 This is the fastest possible forwarding operation as it does not access packets data.
325 This is the default mode.
327 * ``mac``: Changes the source and the destination Ethernet addresses of packets before forwarding them.
328 Default application behavior is to set source Ethernet address to that of the transmitting interface, and destination
329 address to a dummy value (set during init). The user may specify a target destination Ethernet address via the 'eth-peer' or
330 'eth-peers-configfile' command-line options. It is not currently possible to specify a specific source Ethernet address.
332 * ``macswap``: MAC swap forwarding mode.
333 Swaps the source and the destination Ethernet addresses of packets before forwarding them.
335 * ``flowgen``: Multi-flow generation mode.
336 Originates a number of flows (with varying destination IP addresses), and terminate receive traffic.
338 * ``rxonly``: Receives packets but doesn't transmit them.
340 * ``txonly``: Generates and transmits packets without receiving any.
342 * ``csum``: Changes the checksum field with hardware or software methods depending on the offload flags on the packet.
344 * ``icmpecho``: Receives a burst of packets, lookup for ICMP echo requests and, if any, send back ICMP echo replies.
346 * ``ieee1588``: Demonstrate L2 IEEE1588 V2 PTP timestamping for RX and TX.
348 * ``noisy``: Noisy neighbor simulation.
349 Simulate more realistic behavior of a guest machine engaged in receiving
350 and sending packets performing Virtual Network Function (VNF).
352 * ``5tswap``: Swap the source and destination of L2,L3,L4 if they exist.
354 L2 swaps the source address and destination address of Ethernet, as same as ``macswap``.
356 L3 swaps the source address and destination address of IP (v4 and v6).
358 L4 swaps the source port and destination port of transport layer (TCP and UDP).
362 testpmd> set fwd rxonly
364 Set rxonly packet forwarding mode
370 When running, forwarding engines maintain statistics from the time they have been started.
371 Example for the io forwarding engine, with some packet drops on the tx side::
373 testpmd> show fwd stats all
375 ------- Forward Stats for RX Port= 0/Queue= 0 -> TX Port= 1/Queue= 0 -------
376 RX-packets: 274293770 TX-packets: 274293642 TX-dropped: 128
378 ------- Forward Stats for RX Port= 1/Queue= 0 -> TX Port= 0/Queue= 0 -------
379 RX-packets: 274301850 TX-packets: 274301850 TX-dropped: 0
381 ---------------------- Forward statistics for port 0 ----------------------
382 RX-packets: 274293802 RX-dropped: 0 RX-total: 274293802
383 TX-packets: 274301862 TX-dropped: 0 TX-total: 274301862
384 ----------------------------------------------------------------------------
386 ---------------------- Forward statistics for port 1 ----------------------
387 RX-packets: 274301894 RX-dropped: 0 RX-total: 274301894
388 TX-packets: 274293706 TX-dropped: 128 TX-total: 274293834
389 ----------------------------------------------------------------------------
391 +++++++++++++++ Accumulated forward statistics for all ports+++++++++++++++
392 RX-packets: 548595696 RX-dropped: 0 RX-total: 548595696
393 TX-packets: 548595568 TX-dropped: 128 TX-total: 548595696
394 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
400 Clear the forwarding engines statistics::
402 testpmd> clear fwd stats all
407 Display an RX descriptor for a port RX queue::
409 testpmd> read rxd (port_id) (queue_id) (rxd_id)
413 testpmd> read rxd 0 0 4
414 0x0000000B - 0x001D0180 / 0x0000000B - 0x001D0180
419 Display a TX descriptor for a port TX queue::
421 testpmd> read txd (port_id) (queue_id) (txd_id)
425 testpmd> read txd 0 0 4
426 0x00000001 - 0x24C3C440 / 0x000F0000 - 0x2330003C
431 Get loaded dynamic device personalization (DDP) package info list::
433 testpmd> ddp get list (port_id)
438 Display information about dynamic device personalization (DDP) profile::
440 testpmd> ddp get info (profile_path)
445 Display VF statistics::
447 testpmd> show vf stats (port_id) (vf_id)
452 Reset VF statistics::
454 testpmd> clear vf stats (port_id) (vf_id)
456 show port pctype mapping
457 ~~~~~~~~~~~~~~~~~~~~~~~~
459 List all items from the pctype mapping table::
461 testpmd> show port (port_id) pctype mapping
463 show rx offloading capabilities
464 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
466 List all per queue and per port Rx offloading capabilities of a port::
468 testpmd> show port (port_id) rx_offload capabilities
470 show rx offloading configuration
471 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
473 List port level and all queue level Rx offloading configuration::
475 testpmd> show port (port_id) rx_offload configuration
477 show tx offloading capabilities
478 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
480 List all per queue and per port Tx offloading capabilities of a port::
482 testpmd> show port (port_id) tx_offload capabilities
484 show tx offloading configuration
485 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
487 List port level and all queue level Tx offloading configuration::
489 testpmd> show port (port_id) tx_offload configuration
491 show tx metadata setting
492 ~~~~~~~~~~~~~~~~~~~~~~~~
494 Show Tx metadata value set for a specific port::
496 testpmd> show port (port_id) tx_metadata
498 show port supported ptypes
499 ~~~~~~~~~~~~~~~~~~~~~~~~~~
501 Show ptypes supported for a specific port::
503 testpmd> show port (port_id) ptypes
505 set port supported ptypes
506 ~~~~~~~~~~~~~~~~~~~~~~~~~
508 set packet types classification for a specific port::
510 testpmd> set port (port_id) ptypes_mask (mask)
512 show port mac addresses info
513 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
515 Show mac addresses added for a specific port::
517 testpmd> show port (port_id) macs
520 show port multicast mac addresses info
521 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
523 Show multicast mac addresses added for a specific port::
525 testpmd> show port (port_id) mcast_macs
530 Show general information about devices probed::
532 testpmd> show device info (<identifier>|all)
536 .. code-block:: console
538 testpmd> show device info net_pcap0
540 ********************* Infos for device net_pcap0 *********************
542 Driver name: net_pcap
543 Devargs: iface=enP2p6s0,phy_mac=1
544 Connect to socket: -1
547 MAC address: 1E:37:93:28:04:B8
548 Device name: net_pcap0
553 Dumps all physical memory segment layouts::
555 testpmd> dump_physmem
560 Dumps the layout of all memory zones::
562 testpmd> dump_memzone
567 Dumps the memory usage of all sockets::
569 testpmd> dump_socket_mem
574 Dumps the size of all memory structures::
576 testpmd> dump_struct_sizes
581 Dumps the status of all or specific element in DPDK rings::
583 testpmd> dump_ring [ring_name]
588 Dumps the statistics of all or specific memory pool::
590 testpmd> dump_mempool [mempool_name]
595 Dumps the user device list::
597 testpmd> dump_devargs
602 Dumps the log level for all the dpdk modules::
604 testpmd> dump_log_types
606 show (raw_encap|raw_decap)
607 ~~~~~~~~~~~~~~~~~~~~~~~~~~
609 Display content of raw_encap/raw_decap buffers in hex::
611 testpmd> show <raw_encap|raw_decap> <index>
612 testpmd> show <raw_encap|raw_decap> all
616 testpmd> show raw_encap 6
618 index: 6 at [0x1c565b0], len=50
619 00000000: 00 00 00 00 00 00 16 26 36 46 56 66 08 00 45 00 | .......&6FVf..E.
620 00000010: 00 00 00 00 00 00 00 11 00 00 C0 A8 01 06 C0 A8 | ................
621 00000020: 03 06 00 00 00 FA 00 00 00 00 08 00 00 00 00 00 | ................
624 show fec capabilities
625 ~~~~~~~~~~~~~~~~~~~~~
627 Show fec capabilities of a port::
629 testpmd> show port (port_id) fec capabilities
634 Show fec mode of a port::
636 testpmd> show port (port_id) fec_mode
639 Configuration Functions
640 -----------------------
642 The testpmd application can be configured from the runtime as well as from the command-line.
644 This section details the available configuration functions that are available.
648 Configuration changes only become active when forwarding is started/restarted.
653 Reset forwarding to the default configuration::
660 Set the debug verbosity level::
662 testpmd> set verbose (level)
664 Available levels are as following:
666 * ``0`` silent except for error.
667 * ``1`` fully verbose except for Tx packets.
668 * ``2`` fully verbose except for Rx packets.
669 * ``> 2`` fully verbose.
674 Set the log level for a log type::
676 testpmd> set log global|(type) (level)
680 * ``type`` is the log name.
682 * ``level`` is the log level.
684 For example, to change the global log level::
686 testpmd> set log global (level)
688 Regexes can also be used for type. To change log level of user1, user2 and user3::
690 testpmd> set log user[1-3] (level)
695 Set the number of ports used by the application:
699 This is equivalent to the ``--nb-ports`` command-line option.
704 Set the number of cores used by the application::
706 testpmd> set nbcore (num)
708 This is equivalent to the ``--nb-cores`` command-line option.
712 The number of cores used must not be greater than number of ports used multiplied by the number of queues per port.
717 Set the forwarding cores hexadecimal mask::
719 testpmd> set coremask (mask)
721 This is equivalent to the ``--coremask`` command-line option.
725 The main lcore is reserved for command line parsing only and cannot be masked on for packet forwarding.
730 Set the forwarding ports hexadecimal mask::
732 testpmd> set portmask (mask)
734 This is equivalent to the ``--portmask`` command-line option.
736 set record-core-cycles
737 ~~~~~~~~~~~~~~~~~~~~~~
739 Set the recording of CPU cycles::
741 testpmd> set record-core-cycles (on|off)
745 * ``on`` enables measurement of CPU cycles per packet.
747 * ``off`` disables measurement of CPU cycles per packet.
749 This is equivalent to the ``--record-core-cycles command-line`` option.
751 set record-burst-stats
752 ~~~~~~~~~~~~~~~~~~~~~~
754 Set the displaying of RX and TX bursts::
756 testpmd> set record-burst-stats (on|off)
760 * ``on`` enables display of RX and TX bursts.
762 * ``off`` disables display of RX and TX bursts.
764 This is equivalent to the ``--record-burst-stats command-line`` option.
769 Set number of packets per burst::
771 testpmd> set burst (num)
773 This is equivalent to the ``--burst command-line`` option.
775 When retry is enabled, the transmit delay time and number of retries can also be set::
777 testpmd> set burst tx delay (microseconds) retry (num)
782 Set the offsets of segments relating to the data buffer beginning on receiving
783 if split feature is engaged. Affects only the queues configured with split
784 offloads (currently BUFFER_SPLIT is supported only).
786 testpmd> set rxoffs (x[,y]*)
788 Where x[,y]* represents a CSV list of values, without white space. If the list
789 of offsets is shorter than the list of segments the zero offsets will be used
790 for the remaining segments.
795 Set the length of segments to scatter packets on receiving if split
796 feature is engaged. Affects only the queues configured with split offloads
797 (currently BUFFER_SPLIT is supported only). Optionally the multiple memory
798 pools can be specified with --mbuf-size command line parameter and the mbufs
799 to receive will be allocated sequentially from these extra memory pools (the
800 mbuf for the first segment is allocated from the first pool, the second one
801 from the second pool, and so on, if segment number is greater then pool's the
802 mbuf for remaining segments will be allocated from the last valid pool).
804 testpmd> set rxpkts (x[,y]*)
806 Where x[,y]* represents a CSV list of values, without white space. Zero value
807 means to use the corresponding memory pool data buffer size.
812 Set the length of each segment of the TX-ONLY packets or length of packet for FLOWGEN mode::
814 testpmd> set txpkts (x[,y]*)
816 Where x[,y]* represents a CSV list of values, without white space.
821 Configure the timing burst pattern for Tx only mode. This command enables
822 the packet send scheduling on dynamic timestamp mbuf field and configures
823 timing pattern in Tx only mode. In this mode, if scheduling is enabled
824 application provides timestamps in the packets being sent. It is possible
825 to configure delay (in unspecified device clock units) between bursts
826 and between the packets within the burst::
828 testpmd> set txtimes (inter),(intra)
832 * ``inter`` is the delay between the bursts in the device clock units.
833 If ``intra`` is zero, this is the time between the beginnings of the
834 first packets in the neighbour bursts, if ``intra`` is not zero,
835 ``inter`` specifies the time between the beginning of the first packet
836 of the current burst and the beginning of the last packet of the
837 previous burst. If ``inter`` parameter is zero the send scheduling
838 on timestamps is disabled (default).
840 * ``intra`` is the delay between the packets within the burst specified
841 in the device clock units. The number of packets in the burst is defined
842 by regular burst setting. If ``intra`` parameter is zero no timestamps
843 provided in the packets excepting the first one in the burst.
845 As the result the bursts of packet will be transmitted with specific
846 delays between the packets within the burst and specific delay between
847 the bursts. The rte_eth_read_clock() must be supported by the device(s)
848 and is supposed to be engaged to get the current device clock value
849 and provide the reference for the timestamps. If there is no supported
850 rte_eth_read_clock() there will be no send scheduling provided on the port.
855 Set the split policy for the TX packets, applicable for TX-ONLY and CSUM forwarding modes::
857 testpmd> set txsplit (off|on|rand)
861 * ``off`` disable packet copy & split for CSUM mode.
863 * ``on`` split outgoing packet into multiple segments. Size of each segment
864 and number of segments per packet is determined by ``set txpkts`` command
867 * ``rand`` same as 'on', but number of segments per each packet is a random value between 1 and total number of segments.
872 Set the list of forwarding cores::
874 testpmd> set corelist (x[,y]*)
876 For example, to change the forwarding cores:
878 .. code-block:: console
880 testpmd> set corelist 3,1
881 testpmd> show config fwd
883 io packet forwarding - ports=2 - cores=2 - streams=2 - NUMA support disabled
884 Logical Core 3 (socket 0) forwards packets on 1 streams:
885 RX P=0/Q=0 (socket 0) -> TX P=1/Q=0 (socket 0) peer=02:00:00:00:00:01
886 Logical Core 1 (socket 0) forwards packets on 1 streams:
887 RX P=1/Q=0 (socket 0) -> TX P=0/Q=0 (socket 0) peer=02:00:00:00:00:00
891 The cores are used in the same order as specified on the command line.
896 Set the list of forwarding ports::
898 testpmd> set portlist (x[,y]*)
900 For example, to change the port forwarding:
902 .. code-block:: console
904 testpmd> set portlist 0,2,1,3
905 testpmd> show config fwd
907 io packet forwarding - ports=4 - cores=1 - streams=4
908 Logical Core 3 (socket 0) forwards packets on 4 streams:
909 RX P=0/Q=0 (socket 0) -> TX P=2/Q=0 (socket 0) peer=02:00:00:00:00:01
910 RX P=2/Q=0 (socket 0) -> TX P=0/Q=0 (socket 0) peer=02:00:00:00:00:00
911 RX P=1/Q=0 (socket 0) -> TX P=3/Q=0 (socket 0) peer=02:00:00:00:00:03
912 RX P=3/Q=0 (socket 0) -> TX P=1/Q=0 (socket 0) peer=02:00:00:00:00:02
917 Select how to retrieve new ports created after "port attach" command::
919 testpmd> set port setup on (iterator|event)
921 For each new port, a setup is done.
922 It will find the probed ports via RTE_ETH_FOREACH_MATCHING_DEV loop
923 in iterator mode, or via RTE_ETH_EVENT_NEW in event mode.
928 Enable/disable tx loopback::
930 testpmd> set tx loopback (port_id) (on|off)
935 set drop enable bit for all queues::
937 testpmd> set all queues drop (port_id) (on|off)
939 set split drop enable (for VF)
940 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
942 set split drop enable bit for VF from PF::
944 testpmd> set vf split drop (port_id) (vf_id) (on|off)
946 set mac antispoof (for VF)
947 ~~~~~~~~~~~~~~~~~~~~~~~~~~
949 Set mac antispoof for a VF from the PF::
951 testpmd> set vf mac antispoof (port_id) (vf_id) (on|off)
956 Enable/disable MACsec offload::
958 testpmd> set macsec offload (port_id) on encrypt (on|off) replay-protect (on|off)
959 testpmd> set macsec offload (port_id) off
964 Configure MACsec secure connection (SC)::
966 testpmd> set macsec sc (tx|rx) (port_id) (mac) (pi)
970 The pi argument is ignored for tx.
971 Check the NIC Datasheet for hardware limits.
976 Configure MACsec secure association (SA)::
978 testpmd> set macsec sa (tx|rx) (port_id) (idx) (an) (pn) (key)
982 The IDX value must be 0 or 1.
983 Check the NIC Datasheet for hardware limits.
985 set broadcast mode (for VF)
986 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
988 Set broadcast mode for a VF from the PF::
990 testpmd> set vf broadcast (port_id) (vf_id) (on|off)
995 Set the VLAN strip for a queue on a port::
997 testpmd> vlan set stripq (on|off) (port_id,queue_id)
999 vlan set stripq (for VF)
1000 ~~~~~~~~~~~~~~~~~~~~~~~~
1002 Set VLAN strip for all queues in a pool for a VF from the PF::
1004 testpmd> set vf vlan stripq (port_id) (vf_id) (on|off)
1006 vlan set insert (for VF)
1007 ~~~~~~~~~~~~~~~~~~~~~~~~
1009 Set VLAN insert for a VF from the PF::
1011 testpmd> set vf vlan insert (port_id) (vf_id) (vlan_id)
1013 vlan set tag (for VF)
1014 ~~~~~~~~~~~~~~~~~~~~~
1016 Set VLAN tag for a VF from the PF::
1018 testpmd> set vf vlan tag (port_id) (vf_id) (on|off)
1020 vlan set antispoof (for VF)
1021 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
1023 Set VLAN antispoof for a VF from the PF::
1025 testpmd> set vf vlan antispoof (port_id) (vf_id) (on|off)
1027 vlan set (strip|filter|qinq_strip|extend)
1028 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1029 Set the VLAN strip/filter/QinQ strip/extend on for a port::
1031 testpmd> vlan set (strip|filter|qinq_strip|extend) (on|off) (port_id)
1036 Set the inner or outer VLAN TPID for packet filtering on a port::
1038 testpmd> vlan set (inner|outer) tpid (value) (port_id)
1042 TPID value must be a 16-bit number (value <= 65536).
1047 Add a VLAN ID, or all identifiers, to the set of VLAN identifiers filtered by port ID::
1049 testpmd> rx_vlan add (vlan_id|all) (port_id)
1053 VLAN filter must be set on that port. VLAN ID < 4096.
1054 Depending on the NIC used, number of vlan_ids may be limited to the maximum entries
1055 in VFTA table. This is important if enabling all vlan_ids.
1060 Remove a VLAN ID, or all identifiers, from the set of VLAN identifiers filtered by port ID::
1062 testpmd> rx_vlan rm (vlan_id|all) (port_id)
1064 rx_vlan add (for VF)
1065 ~~~~~~~~~~~~~~~~~~~~
1067 Add a VLAN ID, to the set of VLAN identifiers filtered for VF(s) for port ID::
1069 testpmd> rx_vlan add (vlan_id) port (port_id) vf (vf_mask)
1074 Remove a VLAN ID, from the set of VLAN identifiers filtered for VF(s) for port ID::
1076 testpmd> rx_vlan rm (vlan_id) port (port_id) vf (vf_mask)
1081 Add an UDP port for VXLAN packet filter on a port::
1083 testpmd> rx_vxlan_port add (udp_port) (port_id)
1085 rx_vxlan_port remove
1086 ~~~~~~~~~~~~~~~~~~~~
1088 Remove an UDP port for VXLAN packet filter on a port::
1090 testpmd> rx_vxlan_port rm (udp_port) (port_id)
1095 Set hardware insertion of VLAN IDs in packets sent on a port::
1097 testpmd> tx_vlan set (port_id) vlan_id[, vlan_id_outer]
1099 For example, set a single VLAN ID (5) insertion on port 0::
1103 Or, set double VLAN ID (inner: 2, outer: 3) insertion on port 1::
1111 Set port based hardware insertion of VLAN ID in packets sent on a port::
1113 testpmd> tx_vlan set pvid (port_id) (vlan_id) (on|off)
1118 Disable hardware insertion of a VLAN header in packets sent on a port::
1120 testpmd> tx_vlan reset (port_id)
1125 Select hardware or software calculation of the checksum when
1126 transmitting a packet using the ``csum`` forwarding engine::
1128 testpmd> csum set (ip|udp|tcp|sctp|outer-ip|outer-udp) (hw|sw) (port_id)
1132 * ``ip|udp|tcp|sctp`` always relate to the inner layer.
1134 * ``outer-ip`` relates to the outer IP layer (only for IPv4) in the case where the packet is recognized
1135 as a tunnel packet by the forwarding engine (geneve, gre, gtp, ipip, vxlan and vxlan-gpe are
1136 supported). See also the ``csum parse-tunnel`` command.
1138 * ``outer-udp`` relates to the outer UDP layer in the case where the packet is recognized
1139 as a tunnel packet by the forwarding engine (geneve, gtp, vxlan and vxlan-gpe are
1140 supported). See also the ``csum parse-tunnel`` command.
1144 Check the NIC Datasheet for hardware limits.
1149 Set RSS queue region span on a port::
1151 testpmd> set port (port_id) queue-region region_id (value) \
1152 queue_start_index (value) queue_num (value)
1154 Set flowtype mapping on a RSS queue region on a port::
1156 testpmd> set port (port_id) queue-region region_id (value) flowtype (value)
1160 * For the flowtype(pctype) of packet,the specific index for each type has
1161 been defined in file i40e_type.h as enum i40e_filter_pctype.
1163 Set user priority mapping on a RSS queue region on a port::
1165 testpmd> set port (port_id) queue-region UP (value) region_id (value)
1167 Flush all queue region related configuration on a port::
1169 testpmd> set port (port_id) queue-region flush (on|off)
1173 * ``on``: is just an enable function which server for other configuration,
1174 it is for all configuration about queue region from up layer,
1175 at first will only keep in DPDK software stored in driver,
1176 only after "flush on", it commit all configuration to HW.
1178 * ``"off``: is just clean all configuration about queue region just now,
1179 and restore all to DPDK i40e driver default config when start up.
1181 Show all queue region related configuration info on a port::
1183 testpmd> show port (port_id) queue-region
1187 Queue region only support on PF by now, so these command is
1188 only for configuration of queue region on PF port.
1193 Define how tunneled packets should be handled by the csum forward
1196 testpmd> csum parse-tunnel (on|off) (tx_port_id)
1198 If enabled, the csum forward engine will try to recognize supported
1199 tunnel headers (geneve, gtp, gre, ipip, vxlan, vxlan-gpe).
1201 If disabled, treat tunnel packets as non-tunneled packets (a inner
1202 header is handled as a packet payload).
1206 The port argument is the TX port like in the ``csum set`` command.
1210 Consider a packet in packet like the following::
1212 eth_out/ipv4_out/udp_out/vxlan/eth_in/ipv4_in/tcp_in
1214 * If parse-tunnel is enabled, the ``ip|udp|tcp|sctp`` parameters of ``csum set``
1215 command relate to the inner headers (here ``ipv4_in`` and ``tcp_in``), and the
1216 ``outer-ip|outer-udp`` parameter relates to the outer headers (here ``ipv4_out`` and ``udp_out``).
1218 * If parse-tunnel is disabled, the ``ip|udp|tcp|sctp`` parameters of ``csum set``
1219 command relate to the outer headers, here ``ipv4_out`` and ``udp_out``.
1224 Display tx checksum offload configuration::
1226 testpmd> csum show (port_id)
1231 Enable TCP Segmentation Offload (TSO) in the ``csum`` forwarding engine::
1233 testpmd> tso set (segsize) (port_id)
1237 Check the NIC datasheet for hardware limits.
1242 Display the status of TCP Segmentation Offload::
1244 testpmd> tso show (port_id)
1249 Set tso segment size of tunneled packets for a port in csum engine::
1251 testpmd> tunnel_tso set (tso_segsz) (port_id)
1256 Display the status of tunneled TCP Segmentation Offload for a port::
1258 testpmd> tunnel_tso show (port_id)
1263 Enable or disable GRO in ``csum`` forwarding engine::
1265 testpmd> set port <port_id> gro on|off
1267 If enabled, the csum forwarding engine will perform GRO on the TCP/IPv4
1268 packets received from the given port.
1270 If disabled, packets received from the given port won't be performed
1271 GRO. By default, GRO is disabled for all ports.
1275 When enable GRO for a port, TCP/IPv4 packets received from the port
1276 will be performed GRO. After GRO, all merged packets have bad
1277 checksums, since the GRO library doesn't re-calculate checksums for
1278 the merged packets. Therefore, if users want the merged packets to
1279 have correct checksums, please select HW IP checksum calculation and
1280 HW TCP checksum calculation for the port which the merged packets are
1286 Display GRO configuration for a given port::
1288 testpmd> show port <port_id> gro
1293 Set the cycle to flush the GROed packets from reassembly tables::
1295 testpmd> set gro flush <cycles>
1297 When enable GRO, the csum forwarding engine performs GRO on received
1298 packets, and the GROed packets are stored in reassembly tables. Users
1299 can use this command to determine when the GROed packets are flushed
1300 from the reassembly tables.
1302 The ``cycles`` is measured in GRO operation times. The csum forwarding
1303 engine flushes the GROed packets from the tables every ``cycles`` GRO
1306 By default, the value of ``cycles`` is 1, which means flush GROed packets
1307 from the reassembly tables as soon as one GRO operation finishes. The value
1308 of ``cycles`` should be in the range of 1 to ``GRO_MAX_FLUSH_CYCLES``.
1310 Please note that the large value of ``cycles`` may cause the poor TCP/IP
1311 stack performance. Because the GROed packets are delayed to arrive the
1312 stack, thus causing more duplicated ACKs and TCP retransmissions.
1317 Toggle per-port GSO support in ``csum`` forwarding engine::
1319 testpmd> set port <port_id> gso on|off
1321 If enabled, the csum forwarding engine will perform GSO on supported IPv4
1322 packets, transmitted on the given port.
1324 If disabled, packets transmitted on the given port will not undergo GSO.
1325 By default, GSO is disabled for all ports.
1329 When GSO is enabled on a port, supported IPv4 packets transmitted on that
1330 port undergo GSO. Afterwards, the segmented packets are represented by
1331 multi-segment mbufs; however, the csum forwarding engine doesn't calculation
1332 of checksums for GSO'd segments in SW. As a result, if users want correct
1333 checksums in GSO segments, they should enable HW checksum calculation for
1336 For example, HW checksum calculation for VxLAN GSO'd packets may be enabled
1337 by setting the following options in the csum forwarding engine:
1339 testpmd> csum set outer_ip hw <port_id>
1341 testpmd> csum set ip hw <port_id>
1343 testpmd> csum set tcp hw <port_id>
1345 UDP GSO is the same as IP fragmentation, which treats the UDP header
1346 as the payload and does not modify it during segmentation. That is,
1347 after UDP GSO, only the first output fragment has the original UDP
1348 header. Therefore, users need to enable HW IP checksum calculation
1349 and SW UDP checksum calculation for GSO-enabled ports, if they want
1350 correct checksums for UDP/IPv4 packets.
1355 Set the maximum GSO segment size (measured in bytes), which includes the
1356 packet header and the packet payload for GSO-enabled ports (global)::
1358 testpmd> set gso segsz <length>
1363 Display the status of Generic Segmentation Offload for a given port::
1365 testpmd> show port <port_id> gso
1370 Add an alternative MAC address to a port::
1372 testpmd> mac_addr add (port_id) (XX:XX:XX:XX:XX:XX)
1377 Remove a MAC address from a port::
1379 testpmd> mac_addr remove (port_id) (XX:XX:XX:XX:XX:XX)
1384 To add the multicast MAC address to/from the set of multicast addresses
1387 testpmd> mcast_addr add (port_id) (mcast_addr)
1392 To remove the multicast MAC address to/from the set of multicast addresses
1395 testpmd> mcast_addr remove (port_id) (mcast_addr)
1397 mac_addr add (for VF)
1398 ~~~~~~~~~~~~~~~~~~~~~
1400 Add an alternative MAC address for a VF to a port::
1402 testpmd> mac_add add port (port_id) vf (vf_id) (XX:XX:XX:XX:XX:XX)
1407 Set the default MAC address for a port::
1409 testpmd> mac_addr set (port_id) (XX:XX:XX:XX:XX:XX)
1411 mac_addr set (for VF)
1412 ~~~~~~~~~~~~~~~~~~~~~
1414 Set the MAC address for a VF from the PF::
1416 testpmd> set vf mac addr (port_id) (vf_id) (XX:XX:XX:XX:XX:XX)
1421 Set the forwarding peer address for certain port::
1423 testpmd> set eth-peer (port_id) (peer_addr)
1425 This is equivalent to the ``--eth-peer`` command-line option.
1430 Set the unicast hash filter(s) on/off for a port::
1432 testpmd> set port (port_id) uta (XX:XX:XX:XX:XX:XX|all) (on|off)
1437 Set the promiscuous mode on for a port or for all ports.
1438 In promiscuous mode packets are not dropped if they aren't for the specified MAC address::
1440 testpmd> set promisc (port_id|all) (on|off)
1445 Set the allmulti mode for a port or for all ports::
1447 testpmd> set allmulti (port_id|all) (on|off)
1449 Same as the ifconfig (8) option. Controls how multicast packets are handled.
1451 set promisc (for VF)
1452 ~~~~~~~~~~~~~~~~~~~~
1454 Set the unicast promiscuous mode for a VF from PF.
1455 It's supported by Intel i40e NICs now.
1456 In promiscuous mode packets are not dropped if they aren't for the specified MAC address::
1458 testpmd> set vf promisc (port_id) (vf_id) (on|off)
1460 set allmulticast (for VF)
1461 ~~~~~~~~~~~~~~~~~~~~~~~~~
1463 Set the multicast promiscuous mode for a VF from PF.
1464 It's supported by Intel i40e NICs now.
1465 In promiscuous mode packets are not dropped if they aren't for the specified MAC address::
1467 testpmd> set vf allmulti (port_id) (vf_id) (on|off)
1469 set tx max bandwidth (for VF)
1470 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1472 Set TX max absolute bandwidth (Mbps) for a VF from PF::
1474 testpmd> set vf tx max-bandwidth (port_id) (vf_id) (max_bandwidth)
1476 set tc tx min bandwidth (for VF)
1477 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1479 Set all TCs' TX min relative bandwidth (%) for a VF from PF::
1481 testpmd> set vf tc tx min-bandwidth (port_id) (vf_id) (bw1, bw2, ...)
1483 set tc tx max bandwidth (for VF)
1484 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1486 Set a TC's TX max absolute bandwidth (Mbps) for a VF from PF::
1488 testpmd> set vf tc tx max-bandwidth (port_id) (vf_id) (tc_no) (max_bandwidth)
1490 set tc strict link priority mode
1491 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1493 Set some TCs' strict link priority mode on a physical port::
1495 testpmd> set tx strict-link-priority (port_id) (tc_bitmap)
1497 set tc tx min bandwidth
1498 ~~~~~~~~~~~~~~~~~~~~~~~
1500 Set all TCs' TX min relative bandwidth (%) globally for all PF and VFs::
1502 testpmd> set tc tx min-bandwidth (port_id) (bw1, bw2, ...)
1507 Set the link flow control parameter on a port::
1509 testpmd> set flow_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \
1510 (pause_time) (send_xon) mac_ctrl_frame_fwd (on|off) \
1511 autoneg (on|off) (port_id)
1515 * ``high_water`` (integer): High threshold value to trigger XOFF.
1517 * ``low_water`` (integer): Low threshold value to trigger XON.
1519 * ``pause_time`` (integer): Pause quota in the Pause frame.
1521 * ``send_xon`` (0/1): Send XON frame.
1523 * ``mac_ctrl_frame_fwd``: Enable receiving MAC control frames.
1525 * ``autoneg``: Change the auto-negotiation parameter.
1530 show the link flow control parameter on a port::
1532 testpmd> show port <port_id> flow_ctrl
1537 Set the priority flow control parameter on a port::
1539 testpmd> set pfc_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \
1540 (pause_time) (priority) (port_id)
1544 * ``high_water`` (integer): High threshold value.
1546 * ``low_water`` (integer): Low threshold value.
1548 * ``pause_time`` (integer): Pause quota in the Pause frame.
1550 * ``priority`` (0-7): VLAN User Priority.
1555 Set statistics mapping (qmapping 0..15) for RX/TX queue on port::
1557 testpmd> set stat_qmap (tx|rx) (port_id) (queue_id) (qmapping)
1559 For example, to set rx queue 2 on port 0 to mapping 5::
1561 testpmd>set stat_qmap rx 0 2 5
1563 set xstats-hide-zero
1564 ~~~~~~~~~~~~~~~~~~~~
1566 Set the option to hide zero values for xstats display::
1568 testpmd> set xstats-hide-zero on|off
1572 By default, the zero values are displayed for xstats.
1574 set port - rx/tx (for VF)
1575 ~~~~~~~~~~~~~~~~~~~~~~~~~
1577 Set VF receive/transmit from a port::
1579 testpmd> set port (port_id) vf (vf_id) (rx|tx) (on|off)
1581 set port - rx mode(for VF)
1582 ~~~~~~~~~~~~~~~~~~~~~~~~~~
1584 Set the VF receive mode of a port::
1586 testpmd> set port (port_id) vf (vf_id) \
1587 rxmode (AUPE|ROPE|BAM|MPE) (on|off)
1589 The available receive modes are:
1591 * ``AUPE``: Accepts untagged VLAN.
1593 * ``ROPE``: Accepts unicast hash.
1595 * ``BAM``: Accepts broadcast packets.
1597 * ``MPE``: Accepts all multicast packets.
1599 set port - tx_rate (for Queue)
1600 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1602 Set TX rate limitation for a queue on a port::
1604 testpmd> set port (port_id) queue (queue_id) rate (rate_value)
1606 set port - tx_rate (for VF)
1607 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
1609 Set TX rate limitation for queues in VF on a port::
1611 testpmd> set port (port_id) vf (vf_id) rate (rate_value) queue_mask (queue_mask)
1616 Set the flush on RX streams before forwarding.
1617 The default is flush ``on``.
1618 Mainly used with PCAP drivers to turn off the default behavior of flushing the first 512 packets on RX streams::
1620 testpmd> set flush_rx off
1625 Set the bypass mode for the lowest port on bypass enabled NIC::
1627 testpmd> set bypass mode (normal|bypass|isolate) (port_id)
1632 Set the event required to initiate specified bypass mode for the lowest port on a bypass enabled::
1634 testpmd> set bypass event (timeout|os_on|os_off|power_on|power_off) \
1635 mode (normal|bypass|isolate) (port_id)
1639 * ``timeout``: Enable bypass after watchdog timeout.
1641 * ``os_on``: Enable bypass when OS/board is powered on.
1643 * ``os_off``: Enable bypass when OS/board is powered off.
1645 * ``power_on``: Enable bypass when power supply is turned on.
1647 * ``power_off``: Enable bypass when power supply is turned off.
1653 Set the bypass watchdog timeout to ``n`` seconds where 0 = instant::
1655 testpmd> set bypass timeout (0|1.5|2|3|4|8|16|32)
1660 Show the bypass configuration for a bypass enabled NIC using the lowest port on the NIC::
1662 testpmd> show bypass config (port_id)
1667 Set link up for a port::
1669 testpmd> set link-up port (port id)
1674 Set link down for a port::
1676 testpmd> set link-down port (port id)
1681 Enable E-tag insertion for a VF on a port::
1683 testpmd> E-tag set insertion on port-tag-id (value) port (port_id) vf (vf_id)
1685 Disable E-tag insertion for a VF on a port::
1687 testpmd> E-tag set insertion off port (port_id) vf (vf_id)
1689 Enable/disable E-tag stripping on a port::
1691 testpmd> E-tag set stripping (on|off) port (port_id)
1693 Enable/disable E-tag based forwarding on a port::
1695 testpmd> E-tag set forwarding (on|off) port (port_id)
1700 Load a dynamic device personalization (DDP) profile and store backup profile::
1702 testpmd> ddp add (port_id) (profile_path[,backup_profile_path])
1707 Delete a dynamic device personalization profile and restore backup profile::
1709 testpmd> ddp del (port_id) (backup_profile_path)
1714 List all items from the ptype mapping table::
1716 testpmd> ptype mapping get (port_id) (valid_only)
1720 * ``valid_only``: A flag indicates if only list valid items(=1) or all items(=0).
1722 Replace a specific or a group of software defined ptype with a new one::
1724 testpmd> ptype mapping replace (port_id) (target) (mask) (pkt_type)
1728 * ``target``: A specific software ptype or a mask to represent a group of software ptypes.
1730 * ``mask``: A flag indicate if "target" is a specific software ptype(=0) or a ptype mask(=1).
1732 * ``pkt_type``: The new software ptype to replace the old ones.
1734 Update hardware defined ptype to software defined packet type mapping table::
1736 testpmd> ptype mapping update (port_id) (hw_ptype) (sw_ptype)
1740 * ``hw_ptype``: hardware ptype as the index of the ptype mapping table.
1742 * ``sw_ptype``: software ptype as the value of the ptype mapping table.
1744 Reset ptype mapping table::
1746 testpmd> ptype mapping reset (port_id)
1748 config per port Rx offloading
1749 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1751 Enable or disable a per port Rx offloading on all Rx queues of a port::
1753 testpmd> port config (port_id) rx_offload (offloading) on|off
1755 * ``offloading``: can be any of these offloading capability:
1756 vlan_strip, ipv4_cksum, udp_cksum, tcp_cksum, tcp_lro,
1757 qinq_strip, outer_ipv4_cksum, macsec_strip,
1758 header_split, vlan_filter, vlan_extend, jumbo_frame,
1759 scatter, timestamp, security, keep_crc, rss_hash
1761 This command should be run when the port is stopped, or else it will fail.
1763 config per queue Rx offloading
1764 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1766 Enable or disable a per queue Rx offloading only on a specific Rx queue::
1768 testpmd> port (port_id) rxq (queue_id) rx_offload (offloading) on|off
1770 * ``offloading``: can be any of these offloading capability:
1771 vlan_strip, ipv4_cksum, udp_cksum, tcp_cksum, tcp_lro,
1772 qinq_strip, outer_ipv4_cksum, macsec_strip,
1773 header_split, vlan_filter, vlan_extend, jumbo_frame,
1774 scatter, timestamp, security, keep_crc
1776 This command should be run when the port is stopped, or else it will fail.
1778 config per port Tx offloading
1779 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1781 Enable or disable a per port Tx offloading on all Tx queues of a port::
1783 testpmd> port config (port_id) tx_offload (offloading) on|off
1785 * ``offloading``: can be any of these offloading capability:
1786 vlan_insert, ipv4_cksum, udp_cksum, tcp_cksum,
1787 sctp_cksum, tcp_tso, udp_tso, outer_ipv4_cksum,
1788 qinq_insert, vxlan_tnl_tso, gre_tnl_tso,
1789 ipip_tnl_tso, geneve_tnl_tso, macsec_insert,
1790 mt_lockfree, multi_segs, mbuf_fast_free, security
1792 This command should be run when the port is stopped, or else it will fail.
1794 config per queue Tx offloading
1795 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1797 Enable or disable a per queue Tx offloading only on a specific Tx queue::
1799 testpmd> port (port_id) txq (queue_id) tx_offload (offloading) on|off
1801 * ``offloading``: can be any of these offloading capability:
1802 vlan_insert, ipv4_cksum, udp_cksum, tcp_cksum,
1803 sctp_cksum, tcp_tso, udp_tso, outer_ipv4_cksum,
1804 qinq_insert, vxlan_tnl_tso, gre_tnl_tso,
1805 ipip_tnl_tso, geneve_tnl_tso, macsec_insert,
1806 mt_lockfree, multi_segs, mbuf_fast_free, security
1808 This command should be run when the port is stopped, or else it will fail.
1810 Config VXLAN Encap outer layers
1811 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1813 Configure the outer layer to encapsulate a packet inside a VXLAN tunnel::
1815 set vxlan ip-version (ipv4|ipv6) vni (vni) udp-src (udp-src) \
1816 udp-dst (udp-dst) ip-src (ip-src) ip-dst (ip-dst) eth-src (eth-src) \
1819 set vxlan-with-vlan ip-version (ipv4|ipv6) vni (vni) udp-src (udp-src) \
1820 udp-dst (udp-dst) ip-src (ip-src) ip-dst (ip-dst) vlan-tci (vlan-tci) \
1821 eth-src (eth-src) eth-dst (eth-dst)
1823 set vxlan-tos-ttl ip-version (ipv4|ipv6) vni (vni) udp-src (udp-src) \
1824 udp-dst (udp-dst) ip-tos (ip-tos) ip-ttl (ip-ttl) ip-src (ip-src) \
1825 ip-dst (ip-dst) eth-src (eth-src) eth-dst (eth-dst)
1827 These commands will set an internal configuration inside testpmd, any following
1828 flow rule using the action vxlan_encap will use the last configuration set.
1829 To have a different encapsulation header, one of those commands must be called
1830 before the flow rule creation.
1832 Config NVGRE Encap outer layers
1833 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1835 Configure the outer layer to encapsulate a packet inside a NVGRE tunnel::
1837 set nvgre ip-version (ipv4|ipv6) tni (tni) ip-src (ip-src) ip-dst (ip-dst) \
1838 eth-src (eth-src) eth-dst (eth-dst)
1839 set nvgre-with-vlan ip-version (ipv4|ipv6) tni (tni) ip-src (ip-src) \
1840 ip-dst (ip-dst) vlan-tci (vlan-tci) eth-src (eth-src) eth-dst (eth-dst)
1842 These commands will set an internal configuration inside testpmd, any following
1843 flow rule using the action nvgre_encap will use the last configuration set.
1844 To have a different encapsulation header, one of those commands must be called
1845 before the flow rule creation.
1850 Configure the l2 to be used when encapsulating a packet with L2::
1852 set l2_encap ip-version (ipv4|ipv6) eth-src (eth-src) eth-dst (eth-dst)
1853 set l2_encap-with-vlan ip-version (ipv4|ipv6) vlan-tci (vlan-tci) \
1854 eth-src (eth-src) eth-dst (eth-dst)
1856 Those commands will set an internal configuration inside testpmd, any following
1857 flow rule using the action l2_encap will use the last configuration set.
1858 To have a different encapsulation header, one of those commands must be called
1859 before the flow rule creation.
1864 Configure the l2 to be removed when decapsulating a packet with L2::
1866 set l2_decap ip-version (ipv4|ipv6)
1867 set l2_decap-with-vlan ip-version (ipv4|ipv6)
1869 Those commands will set an internal configuration inside testpmd, any following
1870 flow rule using the action l2_decap will use the last configuration set.
1871 To have a different encapsulation header, one of those commands must be called
1872 before the flow rule creation.
1874 Config MPLSoGRE Encap outer layers
1875 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1877 Configure the outer layer to encapsulate a packet inside a MPLSoGRE tunnel::
1879 set mplsogre_encap ip-version (ipv4|ipv6) label (label) \
1880 ip-src (ip-src) ip-dst (ip-dst) eth-src (eth-src) eth-dst (eth-dst)
1881 set mplsogre_encap-with-vlan ip-version (ipv4|ipv6) label (label) \
1882 ip-src (ip-src) ip-dst (ip-dst) vlan-tci (vlan-tci) \
1883 eth-src (eth-src) eth-dst (eth-dst)
1885 These commands will set an internal configuration inside testpmd, any following
1886 flow rule using the action mplsogre_encap will use the last configuration set.
1887 To have a different encapsulation header, one of those commands must be called
1888 before the flow rule creation.
1890 Config MPLSoGRE Decap outer layers
1891 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1893 Configure the outer layer to decapsulate MPLSoGRE packet::
1895 set mplsogre_decap ip-version (ipv4|ipv6)
1896 set mplsogre_decap-with-vlan ip-version (ipv4|ipv6)
1898 These commands will set an internal configuration inside testpmd, any following
1899 flow rule using the action mplsogre_decap will use the last configuration set.
1900 To have a different decapsulation header, one of those commands must be called
1901 before the flow rule creation.
1903 Config MPLSoUDP Encap outer layers
1904 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1906 Configure the outer layer to encapsulate a packet inside a MPLSoUDP tunnel::
1908 set mplsoudp_encap ip-version (ipv4|ipv6) label (label) udp-src (udp-src) \
1909 udp-dst (udp-dst) ip-src (ip-src) ip-dst (ip-dst) \
1910 eth-src (eth-src) eth-dst (eth-dst)
1911 set mplsoudp_encap-with-vlan ip-version (ipv4|ipv6) label (label) \
1912 udp-src (udp-src) udp-dst (udp-dst) ip-src (ip-src) ip-dst (ip-dst) \
1913 vlan-tci (vlan-tci) eth-src (eth-src) eth-dst (eth-dst)
1915 These commands will set an internal configuration inside testpmd, any following
1916 flow rule using the action mplsoudp_encap will use the last configuration set.
1917 To have a different encapsulation header, one of those commands must be called
1918 before the flow rule creation.
1920 Config MPLSoUDP Decap outer layers
1921 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1923 Configure the outer layer to decapsulate MPLSoUDP packet::
1925 set mplsoudp_decap ip-version (ipv4|ipv6)
1926 set mplsoudp_decap-with-vlan ip-version (ipv4|ipv6)
1928 These commands will set an internal configuration inside testpmd, any following
1929 flow rule using the action mplsoudp_decap will use the last configuration set.
1930 To have a different decapsulation header, one of those commands must be called
1931 before the flow rule creation.
1933 Config Raw Encapsulation
1934 ~~~~~~~~~~~~~~~~~~~~~~~~~
1936 Configure the raw data to be used when encapsulating a packet by
1937 rte_flow_action_raw_encap::
1939 set raw_encap {index} {item} [/ {item} [...]] / end_set
1941 There are multiple global buffers for ``raw_encap``, this command will set one
1942 internal buffer index by ``{index}``.
1943 If there is no ``{index}`` specified::
1945 set raw_encap {item} [/ {item} [...]] / end_set
1947 the default index ``0`` is used.
1948 In order to use different encapsulating header, ``index`` must be specified
1949 during the flow rule creation::
1951 testpmd> flow create 0 egress pattern eth / ipv4 / end actions
1952 raw_encap index 2 / end
1954 Otherwise the default index ``0`` is used.
1956 Config Raw Decapsulation
1957 ~~~~~~~~~~~~~~~~~~~~~~~~
1959 Configure the raw data to be used when decapsulating a packet by
1960 rte_flow_action_raw_decap::
1962 set raw_decap {index} {item} [/ {item} [...]] / end_set
1964 There are multiple global buffers for ``raw_decap``, this command will set
1965 one internal buffer index by ``{index}``.
1966 If there is no ``{index}`` specified::
1968 set raw_decap {item} [/ {item} [...]] / end_set
1970 the default index ``0`` is used.
1971 In order to use different decapsulating header, ``index`` must be specified
1972 during the flow rule creation::
1974 testpmd> flow create 0 egress pattern eth / ipv4 / end actions
1975 raw_encap index 3 / end
1977 Otherwise the default index ``0`` is used.
1982 Set fec mode for a specific port::
1984 testpmd> set port (port_id) fec_mode auto|off|rs|baser
1986 Config Sample actions list
1987 ~~~~~~~~~~~~~~~~~~~~~~~~~~
1989 Configure the sample actions list to be used when sampling a packet by
1990 rte_flow_action_sample::
1992 set sample_actions {index} {action} [/ {action} [...]] / end
1994 There are multiple global buffers for ``sample_actions``, this command will set
1995 one internal buffer index by ``{index}``.
1997 In order to use different sample actions list, ``index`` must be specified
1998 during the flow rule creation::
2000 testpmd> flow create 0 ingress pattern eth / ipv4 / end actions
2001 sample ratio 2 index 2 / end
2003 Otherwise the default index ``0`` is used.
2008 The following sections show functions for configuring ports.
2012 Port configuration changes only become active when forwarding is started/restarted.
2017 Attach a port specified by pci address or virtual device args::
2019 testpmd> port attach (identifier)
2021 To attach a new pci device, the device should be recognized by kernel first.
2022 Then it should be moved under DPDK management.
2023 Finally the port can be attached to testpmd.
2025 For example, to move a pci device using ixgbe under DPDK management:
2027 .. code-block:: console
2029 # Check the status of the available devices.
2030 ./usertools/dpdk-devbind.py --status
2032 Network devices using DPDK-compatible driver
2033 ============================================
2036 Network devices using kernel driver
2037 ===================================
2038 0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused=
2041 # Bind the device to igb_uio.
2042 sudo ./usertools/dpdk-devbind.py -b igb_uio 0000:0a:00.0
2045 # Recheck the status of the devices.
2046 ./usertools/dpdk-devbind.py --status
2047 Network devices using DPDK-compatible driver
2048 ============================================
2049 0000:0a:00.0 '82599ES 10-Gigabit' drv=igb_uio unused=
2051 To attach a port created by virtual device, above steps are not needed.
2053 For example, to attach a port whose pci address is 0000:0a:00.0.
2055 .. code-block:: console
2057 testpmd> port attach 0000:0a:00.0
2058 Attaching a new port...
2059 EAL: PCI device 0000:0a:00.0 on NUMA socket -1
2060 EAL: probe driver: 8086:10fb rte_ixgbe_pmd
2061 EAL: PCI memory mapped at 0x7f83bfa00000
2062 EAL: PCI memory mapped at 0x7f83bfa80000
2063 PMD: eth_ixgbe_dev_init(): MAC: 2, PHY: 18, SFP+: 5
2064 PMD: eth_ixgbe_dev_init(): port 0 vendorID=0x8086 deviceID=0x10fb
2065 Port 0 is attached. Now total ports is 1
2068 For example, to attach a port created by pcap PMD.
2070 .. code-block:: console
2072 testpmd> port attach net_pcap0
2073 Attaching a new port...
2074 PMD: Initializing pmd_pcap for net_pcap0
2075 PMD: Creating pcap-backed ethdev on numa socket 0
2076 Port 0 is attached. Now total ports is 1
2079 In this case, identifier is ``net_pcap0``.
2080 This identifier format is the same as ``--vdev`` format of DPDK applications.
2082 For example, to re-attach a bonded port which has been previously detached,
2083 the mode and slave parameters must be given.
2085 .. code-block:: console
2087 testpmd> port attach net_bond_0,mode=0,slave=1
2088 Attaching a new port...
2089 EAL: Initializing pmd_bond for net_bond_0
2090 EAL: Create bonded device net_bond_0 on port 0 in mode 0 on socket 0.
2091 Port 0 is attached. Now total ports is 1
2098 Detach a specific port::
2100 testpmd> port detach (port_id)
2102 Before detaching a port, the port should be stopped and closed.
2104 For example, to detach a pci device port 0.
2106 .. code-block:: console
2108 testpmd> port stop 0
2111 testpmd> port close 0
2115 testpmd> port detach 0
2117 EAL: PCI device 0000:0a:00.0 on NUMA socket -1
2118 EAL: remove driver: 8086:10fb rte_ixgbe_pmd
2119 EAL: PCI memory unmapped at 0x7f83bfa00000
2120 EAL: PCI memory unmapped at 0x7f83bfa80000
2124 For example, to detach a virtual device port 0.
2126 .. code-block:: console
2128 testpmd> port stop 0
2131 testpmd> port close 0
2135 testpmd> port detach 0
2137 PMD: Closing pcap ethdev on numa socket 0
2138 Port 'net_pcap0' is detached. Now total ports is 0
2141 To remove a pci device completely from the system, first detach the port from testpmd.
2142 Then the device should be moved under kernel management.
2143 Finally the device can be removed using kernel pci hotplug functionality.
2145 For example, to move a pci device under kernel management:
2147 .. code-block:: console
2149 sudo ./usertools/dpdk-devbind.py -b ixgbe 0000:0a:00.0
2151 ./usertools/dpdk-devbind.py --status
2153 Network devices using DPDK-compatible driver
2154 ============================================
2157 Network devices using kernel driver
2158 ===================================
2159 0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused=igb_uio
2161 To remove a port created by a virtual device, above steps are not needed.
2166 Start all ports or a specific port::
2168 testpmd> port start (port_id|all)
2173 Stop all ports or a specific port::
2175 testpmd> port stop (port_id|all)
2180 Close all ports or a specific port::
2182 testpmd> port close (port_id|all)
2187 Reset all ports or a specific port::
2189 testpmd> port reset (port_id|all)
2191 User should stop port(s) before resetting and (re-)start after reset.
2193 port config - queue ring size
2194 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2196 Configure a rx/tx queue ring size::
2198 testpmd> port (port_id) (rxq|txq) (queue_id) ring_size (value)
2200 Only take effect after command that (re-)start the port or command that setup specific queue.
2202 port start/stop queue
2203 ~~~~~~~~~~~~~~~~~~~~~
2205 Start/stop a rx/tx queue on a specific port::
2207 testpmd> port (port_id) (rxq|txq) (queue_id) (start|stop)
2209 port config - queue deferred start
2210 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2212 Switch on/off deferred start of a specific port queue::
2214 testpmd> port (port_id) (rxq|txq) (queue_id) deferred_start (on|off)
2217 ~~~~~~~~~~~~~~~~~~~~~
2219 Setup a rx/tx queue on a specific port::
2221 testpmd> port (port_id) (rxq|txq) (queue_id) setup
2223 Only take effect when port is started.
2228 Set the speed and duplex mode for all ports or a specific port::
2230 testpmd> port config (port_id|all) speed (10|100|1000|10000|25000|40000|50000|100000|200000|auto) \
2231 duplex (half|full|auto)
2233 port config - queues/descriptors
2234 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2236 Set number of queues/descriptors for rxq, txq, rxd and txd::
2238 testpmd> port config all (rxq|txq|rxd|txd) (value)
2240 This is equivalent to the ``--rxq``, ``--txq``, ``--rxd`` and ``--txd`` command-line options.
2242 port config - max-pkt-len
2243 ~~~~~~~~~~~~~~~~~~~~~~~~~
2245 Set the maximum packet length::
2247 testpmd> port config all max-pkt-len (value)
2249 This is equivalent to the ``--max-pkt-len`` command-line option.
2251 port config - max-lro-pkt-size
2252 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2254 Set the maximum LRO aggregated packet size::
2256 testpmd> port config all max-lro-pkt-size (value)
2258 This is equivalent to the ``--max-lro-pkt-size`` command-line option.
2260 port config - Drop Packets
2261 ~~~~~~~~~~~~~~~~~~~~~~~~~~
2263 Enable or disable packet drop on all RX queues of all ports when no receive buffers available::
2265 testpmd> port config all drop-en (on|off)
2267 Packet dropping when no receive buffers available is off by default.
2269 The ``on`` option is equivalent to the ``--enable-drop-en`` command-line option.
2274 Set the RSS (Receive Side Scaling) mode on or off::
2276 testpmd> port config all rss (all|default|eth|vlan|ip|tcp|udp|sctp|ether|port|vxlan|geneve|nvgre|vxlan-gpe|l2tpv3|esp|ah|pfcp|ecpri|mpls|none)
2278 RSS is on by default.
2280 The ``all`` option is equivalent to eth|vlan|ip|tcp|udp|sctp|ether|l2tpv3|esp|ah|pfcp.
2282 The ``default`` option enables all supported RSS types reported by device info.
2284 The ``none`` option is equivalent to the ``--disable-rss`` command-line option.
2286 port config - RSS Reta
2287 ~~~~~~~~~~~~~~~~~~~~~~
2289 Set the RSS (Receive Side Scaling) redirection table::
2291 testpmd> port config all rss reta (hash,queue)[,(hash,queue)]
2296 Set the DCB mode for an individual port::
2298 testpmd> port config (port_id) dcb vt (on|off) (traffic_class) pfc (on|off)
2300 The traffic class should be 4 or 8.
2305 Set the number of packets per burst::
2307 testpmd> port config all burst (value)
2309 This is equivalent to the ``--burst`` command-line option.
2311 port config - Threshold
2312 ~~~~~~~~~~~~~~~~~~~~~~~
2314 Set thresholds for TX/RX queues::
2316 testpmd> port config all (threshold) (value)
2318 Where the threshold type can be:
2320 * ``txpt:`` Set the prefetch threshold register of the TX rings, 0 <= value <= 255.
2322 * ``txht:`` Set the host threshold register of the TX rings, 0 <= value <= 255.
2324 * ``txwt:`` Set the write-back threshold register of the TX rings, 0 <= value <= 255.
2326 * ``rxpt:`` Set the prefetch threshold register of the RX rings, 0 <= value <= 255.
2328 * ``rxht:`` Set the host threshold register of the RX rings, 0 <= value <= 255.
2330 * ``rxwt:`` Set the write-back threshold register of the RX rings, 0 <= value <= 255.
2332 * ``txfreet:`` Set the transmit free threshold of the TX rings, 0 <= value <= txd.
2334 * ``rxfreet:`` Set the transmit free threshold of the RX rings, 0 <= value <= rxd.
2336 * ``txrst:`` Set the transmit RS bit threshold of TX rings, 0 <= value <= txd.
2338 These threshold options are also available from the command-line.
2340 port config pctype mapping
2341 ~~~~~~~~~~~~~~~~~~~~~~~~~~
2343 Reset pctype mapping table::
2345 testpmd> port config (port_id) pctype mapping reset
2347 Update hardware defined pctype to software defined flow type mapping table::
2349 testpmd> port config (port_id) pctype mapping update (pctype_id_0[,pctype_id_1]*) (flow_type_id)
2353 * ``pctype_id_x``: hardware pctype id as index of bit in bitmask value of the pctype mapping table.
2355 * ``flow_type_id``: software flow type id as the index of the pctype mapping table.
2357 port config input set
2358 ~~~~~~~~~~~~~~~~~~~~~
2360 Config RSS/FDIR/FDIR flexible payload input set for some pctype::
2362 testpmd> port config (port_id) pctype (pctype_id) \
2363 (hash_inset|fdir_inset|fdir_flx_inset) \
2364 (get|set|clear) field (field_idx)
2366 Clear RSS/FDIR/FDIR flexible payload input set for some pctype::
2368 testpmd> port config (port_id) pctype (pctype_id) \
2369 (hash_inset|fdir_inset|fdir_flx_inset) clear all
2373 * ``pctype_id``: hardware packet classification types.
2374 * ``field_idx``: hardware field index.
2376 port config udp_tunnel_port
2377 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
2379 Add/remove UDP tunnel port for VXLAN/GENEVE tunneling protocols::
2381 testpmd> port config (port_id) udp_tunnel_port add|rm vxlan|geneve|vxlan-gpe|ecpri (udp_port)
2383 port config tx_metadata
2384 ~~~~~~~~~~~~~~~~~~~~~~~
2386 Set Tx metadata value per port.
2387 testpmd will add this value to any Tx packet sent from this port::
2389 testpmd> port config (port_id) tx_metadata (value)
2394 Set/clear dynamic flag per port.
2395 testpmd will register this flag in the mbuf (same registration
2396 for both Tx and Rx). Then set/clear this flag for each Tx
2397 packet sent from this port. The set bit only works for Tx packet::
2399 testpmd> port config (port_id) dynf (name) (set|clear)
2404 To configure MTU(Maximum Transmission Unit) on devices using testpmd::
2406 testpmd> port config mtu (port_id) (value)
2408 port config rss hash key
2409 ~~~~~~~~~~~~~~~~~~~~~~~~
2411 To configure the RSS hash key used to compute the RSS
2412 hash of input [IP] packets received on port::
2414 testpmd> port config <port_id> rss-hash-key (ipv4|ipv4-frag|\
2415 ipv4-tcp|ipv4-udp|ipv4-sctp|ipv4-other|\
2416 ipv6|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|\
2417 ipv6-other|l2-payload|ipv6-ex|ipv6-tcp-ex|\
2418 ipv6-udp-ex <string of hex digits \
2419 (variable length, NIC dependent)>)
2421 port cleanup txq mbufs
2422 ~~~~~~~~~~~~~~~~~~~~~~
2424 To cleanup txq mbufs currently cached by driver::
2426 testpmd> port cleanup (port_id) txq (queue_id) (free_cnt)
2428 If the value of ``free_cnt`` is 0, driver should free all cached mbufs.
2433 The following sections show functions for device operations.
2438 Detach a device specified by pci address or virtual device args::
2440 testpmd> device detach (identifier)
2442 Before detaching a device associated with ports, the ports should be stopped and closed.
2444 For example, to detach a pci device whose address is 0002:03:00.0.
2446 .. code-block:: console
2448 testpmd> device detach 0002:03:00.0
2449 Removing a device...
2450 Port 1 is now closed
2451 EAL: Releasing pci mapped resource for 0002:03:00.0
2452 EAL: Calling pci_unmap_resource for 0002:03:00.0 at 0x218a050000
2453 EAL: Calling pci_unmap_resource for 0002:03:00.0 at 0x218c050000
2454 Device 0002:03:00.0 is detached
2455 Now total ports is 1
2457 For example, to detach a port created by pcap PMD.
2459 .. code-block:: console
2461 testpmd> device detach net_pcap0
2462 Removing a device...
2463 Port 0 is now closed
2464 Device net_pcap0 is detached
2465 Now total ports is 0
2468 In this case, identifier is ``net_pcap0``.
2469 This identifier format is the same as ``--vdev`` format of DPDK applications.
2471 Link Bonding Functions
2472 ----------------------
2474 The Link Bonding functions make it possible to dynamically create and
2475 manage link bonding devices from within testpmd interactive prompt.
2477 create bonded device
2478 ~~~~~~~~~~~~~~~~~~~~
2480 Create a new bonding device::
2482 testpmd> create bonded device (mode) (socket)
2484 For example, to create a bonded device in mode 1 on socket 0::
2486 testpmd> create bonded device 1 0
2487 created new bonded device (port X)
2492 Adds Ethernet device to a Link Bonding device::
2494 testpmd> add bonding slave (slave id) (port id)
2496 For example, to add Ethernet device (port 6) to a Link Bonding device (port 10)::
2498 testpmd> add bonding slave 6 10
2501 remove bonding slave
2502 ~~~~~~~~~~~~~~~~~~~~
2504 Removes an Ethernet slave device from a Link Bonding device::
2506 testpmd> remove bonding slave (slave id) (port id)
2508 For example, to remove Ethernet slave device (port 6) to a Link Bonding device (port 10)::
2510 testpmd> remove bonding slave 6 10
2515 Set the Link Bonding mode of a Link Bonding device::
2517 testpmd> set bonding mode (value) (port id)
2519 For example, to set the bonding mode of a Link Bonding device (port 10) to broadcast (mode 3)::
2521 testpmd> set bonding mode 3 10
2526 Set an Ethernet slave device as the primary device on a Link Bonding device::
2528 testpmd> set bonding primary (slave id) (port id)
2530 For example, to set the Ethernet slave device (port 6) as the primary port of a Link Bonding device (port 10)::
2532 testpmd> set bonding primary 6 10
2537 Set the MAC address of a Link Bonding device::
2539 testpmd> set bonding mac (port id) (mac)
2541 For example, to set the MAC address of a Link Bonding device (port 10) to 00:00:00:00:00:01::
2543 testpmd> set bonding mac 10 00:00:00:00:00:01
2545 set bonding balance_xmit_policy
2546 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2548 Set the transmission policy for a Link Bonding device when it is in Balance XOR mode::
2550 testpmd> set bonding balance_xmit_policy (port_id) (l2|l23|l34)
2552 For example, set a Link Bonding device (port 10) to use a balance policy of layer 3+4 (IP addresses & UDP ports)::
2554 testpmd> set bonding balance_xmit_policy 10 l34
2557 set bonding mon_period
2558 ~~~~~~~~~~~~~~~~~~~~~~
2560 Set the link status monitoring polling period in milliseconds for a bonding device.
2562 This adds support for PMD slave devices which do not support link status interrupts.
2563 When the mon_period is set to a value greater than 0 then all PMD's which do not support
2564 link status ISR will be queried every polling interval to check if their link status has changed::
2566 testpmd> set bonding mon_period (port_id) (value)
2568 For example, to set the link status monitoring polling period of bonded device (port 5) to 150ms::
2570 testpmd> set bonding mon_period 5 150
2573 set bonding lacp dedicated_queue
2574 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2576 Enable dedicated tx/rx queues on bonding devices slaves to handle LACP control plane traffic
2577 when in mode 4 (link-aggregation-802.3ad)::
2579 testpmd> set bonding lacp dedicated_queues (port_id) (enable|disable)
2582 set bonding agg_mode
2583 ~~~~~~~~~~~~~~~~~~~~
2585 Enable one of the specific aggregators mode when in mode 4 (link-aggregation-802.3ad)::
2587 testpmd> set bonding agg_mode (port_id) (bandwidth|count|stable)
2593 Show the current configuration of a Link Bonding device::
2595 testpmd> show bonding config (port id)
2598 to show the configuration a Link Bonding device (port 9) with 3 slave devices (1, 3, 4)
2599 in balance mode with a transmission policy of layer 2+3::
2601 testpmd> show bonding config 9
2603 Balance Xmit Policy: BALANCE_XMIT_POLICY_LAYER23
2605 Active Slaves (3): [1 3 4]
2612 The Register Functions can be used to read from and write to registers on the network card referenced by a port number.
2613 This is mainly useful for debugging purposes.
2614 Reference should be made to the appropriate datasheet for the network card for details on the register addresses
2615 and fields that can be accessed.
2620 Display the value of a port register::
2622 testpmd> read reg (port_id) (address)
2624 For example, to examine the Flow Director control register (FDIRCTL, 0x0000EE000) on an Intel 82599 10 GbE Controller::
2626 testpmd> read reg 0 0xEE00
2627 port 0 PCI register at offset 0xEE00: 0x4A060029 (1241907241)
2632 Display a port register bit field::
2634 testpmd> read regfield (port_id) (address) (bit_x) (bit_y)
2636 For example, reading the lowest two bits from the register in the example above::
2638 testpmd> read regfield 0 0xEE00 0 1
2639 port 0 PCI register at offset 0xEE00: bits[0, 1]=0x1 (1)
2644 Display a single port register bit::
2646 testpmd> read regbit (port_id) (address) (bit_x)
2648 For example, reading the lowest bit from the register in the example above::
2650 testpmd> read regbit 0 0xEE00 0
2651 port 0 PCI register at offset 0xEE00: bit 0=1
2656 Set the value of a port register::
2658 testpmd> write reg (port_id) (address) (value)
2660 For example, to clear a register::
2662 testpmd> write reg 0 0xEE00 0x0
2663 port 0 PCI register at offset 0xEE00: 0x00000000 (0)
2668 Set bit field of a port register::
2670 testpmd> write regfield (port_id) (address) (bit_x) (bit_y) (value)
2672 For example, writing to the register cleared in the example above::
2674 testpmd> write regfield 0 0xEE00 0 1 2
2675 port 0 PCI register at offset 0xEE00: 0x00000002 (2)
2680 Set single bit value of a port register::
2682 testpmd> write regbit (port_id) (address) (bit_x) (value)
2684 For example, to set the high bit in the register from the example above::
2686 testpmd> write regbit 0 0xEE00 31 1
2687 port 0 PCI register at offset 0xEE00: 0x8000000A (2147483658)
2689 Traffic Metering and Policing
2690 -----------------------------
2692 The following section shows functions for configuring traffic metering and
2693 policing on the ethernet device through the use of generic ethdev API.
2695 show port traffic management capability
2696 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2698 Show traffic metering and policing capability of the port::
2700 testpmd> show port meter cap (port_id)
2702 add port meter profile (srTCM rfc2967)
2703 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2705 Add meter profile (srTCM rfc2697) to the ethernet device::
2707 testpmd> add port meter profile srtcm_rfc2697 (port_id) (profile_id) \
2708 (cir) (cbs) (ebs) (packet_mode)
2712 * ``profile_id``: ID for the meter profile.
2713 * ``cir``: Committed Information Rate (CIR) (bytes per second or packets per second).
2714 * ``cbs``: Committed Burst Size (CBS) (bytes or packets).
2715 * ``ebs``: Excess Burst Size (EBS) (bytes or packets).
2716 * ``packet_mode``: Packets mode for meter profile.
2718 add port meter profile (trTCM rfc2968)
2719 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2721 Add meter profile (srTCM rfc2698) to the ethernet device::
2723 testpmd> add port meter profile trtcm_rfc2698 (port_id) (profile_id) \
2724 (cir) (pir) (cbs) (pbs) (packet_mode)
2728 * ``profile_id``: ID for the meter profile.
2729 * ``cir``: Committed information rate (bytes per second or packets per second).
2730 * ``pir``: Peak information rate (bytes per second or packets per second).
2731 * ``cbs``: Committed burst size (bytes or packets).
2732 * ``pbs``: Peak burst size (bytes or packets).
2733 * ``packet_mode``: Packets mode for meter profile.
2735 add port meter profile (trTCM rfc4115)
2736 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2738 Add meter profile (trTCM rfc4115) to the ethernet device::
2740 testpmd> add port meter profile trtcm_rfc4115 (port_id) (profile_id) \
2741 (cir) (eir) (cbs) (ebs) (packet_mode)
2745 * ``profile_id``: ID for the meter profile.
2746 * ``cir``: Committed information rate (bytes per second or packets per second).
2747 * ``eir``: Excess information rate (bytes per second or packets per second).
2748 * ``cbs``: Committed burst size (bytes or packets).
2749 * ``ebs``: Excess burst size (bytes or packets).
2750 * ``packet_mode``: Packets mode for meter profile.
2752 delete port meter profile
2753 ~~~~~~~~~~~~~~~~~~~~~~~~~
2755 Delete meter profile from the ethernet device::
2757 testpmd> del port meter profile (port_id) (profile_id)
2762 Create new policy object for the ethernet device::
2764 testpmd> add port meter policy (port_id) (policy_id) g_actions \
2765 {action} y_actions {action} r_actions {action}
2769 * ``policy_id``: policy ID.
2770 * ``action``: action lists for green/yellow/red colors.
2775 Delete policy object for the ethernet device::
2777 testpmd> del port meter policy (port_id) (policy_id)
2781 * ``policy_id``: policy ID.
2786 Create new meter object for the ethernet device::
2788 testpmd> create port meter (port_id) (mtr_id) (profile_id) \
2789 (policy_id) (meter_enable) (stats_mask) (shared) \
2790 (use_pre_meter_color) [(dscp_tbl_entry0) (dscp_tbl_entry1)...\
2795 * ``mtr_id``: meter object ID.
2796 * ``profile_id``: ID for the meter profile.
2797 * ``policy_id``: ID for the policy.
2798 * ``meter_enable``: When this parameter has a non-zero value, the meter object
2799 gets enabled at the time of creation, otherwise remains disabled.
2800 * ``stats_mask``: Mask of statistics counter types to be enabled for the
2802 * ``shared``: When this parameter has a non-zero value, the meter object is
2803 shared by multiple flows. Otherwise, meter object is used by single flow.
2804 * ``use_pre_meter_color``: When this parameter has a non-zero value, the
2805 input color for the current meter object is determined by the latest meter
2806 object in the same flow. Otherwise, the current meter object uses the
2807 *dscp_table* to determine the input color.
2808 * ``dscp_tbl_entryx``: DSCP table entry x providing meter providing input
2809 color, 0 <= x <= 63.
2814 Enable meter for the ethernet device::
2816 testpmd> enable port meter (port_id) (mtr_id)
2821 Disable meter for the ethernet device::
2823 testpmd> disable port meter (port_id) (mtr_id)
2828 Delete meter for the ethernet device::
2830 testpmd> del port meter (port_id) (mtr_id)
2832 Set port meter profile
2833 ~~~~~~~~~~~~~~~~~~~~~~
2835 Set meter profile for the ethernet device::
2837 testpmd> set port meter profile (port_id) (mtr_id) (profile_id)
2839 set port meter dscp table
2840 ~~~~~~~~~~~~~~~~~~~~~~~~~
2842 Set meter dscp table for the ethernet device::
2844 testpmd> set port meter dscp table (port_id) (mtr_id) [(dscp_tbl_entry0) \
2845 (dscp_tbl_entry1)...(dscp_tbl_entry63)]
2847 set port meter stats mask
2848 ~~~~~~~~~~~~~~~~~~~~~~~~~
2850 Set meter stats mask for the ethernet device::
2852 testpmd> set port meter stats mask (port_id) (mtr_id) (stats_mask)
2856 * ``stats_mask``: Bit mask indicating statistics counter types to be enabled.
2858 show port meter stats
2859 ~~~~~~~~~~~~~~~~~~~~~
2861 Show meter stats of the ethernet device::
2863 testpmd> show port meter stats (port_id) (mtr_id) (clear)
2867 * ``clear``: Flag that indicates whether the statistics counters should
2868 be cleared (i.e. set to zero) immediately after they have been read or not.
2873 The following section shows functions for configuring traffic management on
2874 the ethernet device through the use of generic TM API.
2876 show port traffic management capability
2877 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2879 Show traffic management capability of the port::
2881 testpmd> show port tm cap (port_id)
2883 show port traffic management capability (hierarchy level)
2884 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2886 Show traffic management hierarchy level capability of the port::
2888 testpmd> show port tm level cap (port_id) (level_id)
2890 show port traffic management capability (hierarchy node level)
2891 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2893 Show the traffic management hierarchy node capability of the port::
2895 testpmd> show port tm node cap (port_id) (node_id)
2897 show port traffic management hierarchy node type
2898 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2900 Show the port traffic management hierarchy node type::
2902 testpmd> show port tm node type (port_id) (node_id)
2904 show port traffic management hierarchy node stats
2905 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2907 Show the port traffic management hierarchy node statistics::
2909 testpmd> show port tm node stats (port_id) (node_id) (clear)
2913 * ``clear``: When this parameter has a non-zero value, the statistics counters
2914 are cleared (i.e. set to zero) immediately after they have been read,
2915 otherwise the statistics counters are left untouched.
2917 Add port traffic management private shaper profile
2918 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2920 Add the port traffic management private shaper profile::
2922 testpmd> add port tm node shaper profile (port_id) (shaper_profile_id) \
2923 (cmit_tb_rate) (cmit_tb_size) (peak_tb_rate) (peak_tb_size) \
2924 (packet_length_adjust) (packet_mode)
2928 * ``shaper_profile id``: Shaper profile ID for the new profile.
2929 * ``cmit_tb_rate``: Committed token bucket rate (bytes per second or packets per second).
2930 * ``cmit_tb_size``: Committed token bucket size (bytes or packets).
2931 * ``peak_tb_rate``: Peak token bucket rate (bytes per second or packets per second).
2932 * ``peak_tb_size``: Peak token bucket size (bytes or packets).
2933 * ``packet_length_adjust``: The value (bytes) to be added to the length of
2934 each packet for the purpose of shaping. This parameter value can be used to
2935 correct the packet length with the framing overhead bytes that are consumed
2937 * ``packet_mode``: Shaper configured in packet mode. This parameter value if
2938 zero, configures shaper in byte mode and if non-zero configures it in packet
2941 Delete port traffic management private shaper profile
2942 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2944 Delete the port traffic management private shaper::
2946 testpmd> del port tm node shaper profile (port_id) (shaper_profile_id)
2950 * ``shaper_profile id``: Shaper profile ID that needs to be deleted.
2952 Add port traffic management shared shaper
2953 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2955 Create the port traffic management shared shaper::
2957 testpmd> add port tm node shared shaper (port_id) (shared_shaper_id) \
2962 * ``shared_shaper_id``: Shared shaper ID to be created.
2963 * ``shaper_profile id``: Shaper profile ID for shared shaper.
2965 Set port traffic management shared shaper
2966 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2968 Update the port traffic management shared shaper::
2970 testpmd> set port tm node shared shaper (port_id) (shared_shaper_id) \
2975 * ``shared_shaper_id``: Shared shaper ID to be update.
2976 * ``shaper_profile id``: Shaper profile ID for shared shaper.
2978 Delete port traffic management shared shaper
2979 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2981 Delete the port traffic management shared shaper::
2983 testpmd> del port tm node shared shaper (port_id) (shared_shaper_id)
2987 * ``shared_shaper_id``: Shared shaper ID to be deleted.
2989 Set port traffic management hierarchy node private shaper
2990 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2992 set the port traffic management hierarchy node private shaper::
2994 testpmd> set port tm node shaper profile (port_id) (node_id) \
2999 * ``shaper_profile id``: Private shaper profile ID to be enabled on the
3002 Add port traffic management WRED profile
3003 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3005 Create a new WRED profile::
3007 testpmd> add port tm node wred profile (port_id) (wred_profile_id) \
3008 (color_g) (min_th_g) (max_th_g) (maxp_inv_g) (wq_log2_g) \
3009 (color_y) (min_th_y) (max_th_y) (maxp_inv_y) (wq_log2_y) \
3010 (color_r) (min_th_r) (max_th_r) (maxp_inv_r) (wq_log2_r)
3014 * ``wred_profile id``: Identifier for the newly create WRED profile
3015 * ``color_g``: Packet color (green)
3016 * ``min_th_g``: Minimum queue threshold for packet with green color
3017 * ``max_th_g``: Minimum queue threshold for packet with green color
3018 * ``maxp_inv_g``: Inverse of packet marking probability maximum value (maxp)
3019 * ``wq_log2_g``: Negated log2 of queue weight (wq)
3020 * ``color_y``: Packet color (yellow)
3021 * ``min_th_y``: Minimum queue threshold for packet with yellow color
3022 * ``max_th_y``: Minimum queue threshold for packet with yellow color
3023 * ``maxp_inv_y``: Inverse of packet marking probability maximum value (maxp)
3024 * ``wq_log2_y``: Negated log2 of queue weight (wq)
3025 * ``color_r``: Packet color (red)
3026 * ``min_th_r``: Minimum queue threshold for packet with yellow color
3027 * ``max_th_r``: Minimum queue threshold for packet with yellow color
3028 * ``maxp_inv_r``: Inverse of packet marking probability maximum value (maxp)
3029 * ``wq_log2_r``: Negated log2 of queue weight (wq)
3031 Delete port traffic management WRED profile
3032 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3034 Delete the WRED profile::
3036 testpmd> del port tm node wred profile (port_id) (wred_profile_id)
3038 Add port traffic management hierarchy nonleaf node
3039 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3041 Add nonleaf node to port traffic management hierarchy::
3043 testpmd> add port tm nonleaf node (port_id) (node_id) (parent_node_id) \
3044 (priority) (weight) (level_id) (shaper_profile_id) \
3045 (n_sp_priorities) (stats_mask) (n_shared_shapers) \
3046 [(shared_shaper_0) (shared_shaper_1) ...] \
3050 * ``parent_node_id``: Node ID of the parent.
3051 * ``priority``: Node priority (highest node priority is zero). This is used by
3052 the SP algorithm running on the parent node for scheduling this node.
3053 * ``weight``: Node weight (lowest weight is one). The node weight is relative
3054 to the weight sum of all siblings that have the same priority. It is used by
3055 the WFQ algorithm running on the parent node for scheduling this node.
3056 * ``level_id``: Hierarchy level of the node.
3057 * ``shaper_profile_id``: Shaper profile ID of the private shaper to be used by
3059 * ``n_sp_priorities``: Number of strict priorities.
3060 * ``stats_mask``: Mask of statistics counter types to be enabled for this node.
3061 * ``n_shared_shapers``: Number of shared shapers.
3062 * ``shared_shaper_id``: Shared shaper id.
3064 Add port traffic management hierarchy nonleaf node with packet mode
3065 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3067 Add nonleaf node with packet mode to port traffic management hierarchy::
3069 testpmd> add port tm nonleaf node pktmode (port_id) (node_id) (parent_node_id) \
3070 (priority) (weight) (level_id) (shaper_profile_id) \
3071 (n_sp_priorities) (stats_mask) (n_shared_shapers) \
3072 [(shared_shaper_0) (shared_shaper_1) ...] \
3076 * ``parent_node_id``: Node ID of the parent.
3077 * ``priority``: Node priority (highest node priority is zero). This is used by
3078 the SP algorithm running on the parent node for scheduling this node.
3079 * ``weight``: Node weight (lowest weight is one). The node weight is relative
3080 to the weight sum of all siblings that have the same priority. It is used by
3081 the WFQ algorithm running on the parent node for scheduling this node.
3082 * ``level_id``: Hierarchy level of the node.
3083 * ``shaper_profile_id``: Shaper profile ID of the private shaper to be used by
3085 * ``n_sp_priorities``: Number of strict priorities. Packet mode is enabled on
3087 * ``stats_mask``: Mask of statistics counter types to be enabled for this node.
3088 * ``n_shared_shapers``: Number of shared shapers.
3089 * ``shared_shaper_id``: Shared shaper id.
3091 Add port traffic management hierarchy leaf node
3092 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3094 Add leaf node to port traffic management hierarchy::
3096 testpmd> add port tm leaf node (port_id) (node_id) (parent_node_id) \
3097 (priority) (weight) (level_id) (shaper_profile_id) \
3098 (cman_mode) (wred_profile_id) (stats_mask) (n_shared_shapers) \
3099 [(shared_shaper_id) (shared_shaper_id) ...] \
3103 * ``parent_node_id``: Node ID of the parent.
3104 * ``priority``: Node priority (highest node priority is zero). This is used by
3105 the SP algorithm running on the parent node for scheduling this node.
3106 * ``weight``: Node weight (lowest weight is one). The node weight is relative
3107 to the weight sum of all siblings that have the same priority. It is used by
3108 the WFQ algorithm running on the parent node for scheduling this node.
3109 * ``level_id``: Hierarchy level of the node.
3110 * ``shaper_profile_id``: Shaper profile ID of the private shaper to be used by
3112 * ``cman_mode``: Congestion management mode to be enabled for this node.
3113 * ``wred_profile_id``: WRED profile id to be enabled for this node.
3114 * ``stats_mask``: Mask of statistics counter types to be enabled for this node.
3115 * ``n_shared_shapers``: Number of shared shapers.
3116 * ``shared_shaper_id``: Shared shaper id.
3118 Delete port traffic management hierarchy node
3119 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3121 Delete node from port traffic management hierarchy::
3123 testpmd> del port tm node (port_id) (node_id)
3125 Update port traffic management hierarchy parent node
3126 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3128 Update port traffic management hierarchy parent node::
3130 testpmd> set port tm node parent (port_id) (node_id) (parent_node_id) \
3133 This function can only be called after the hierarchy commit invocation. Its
3134 success depends on the port support for this operation, as advertised through
3135 the port capability set. This function is valid for all nodes of the traffic
3136 management hierarchy except root node.
3138 Suspend port traffic management hierarchy node
3139 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3141 testpmd> suspend port tm node (port_id) (node_id)
3143 Resume port traffic management hierarchy node
3144 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3146 testpmd> resume port tm node (port_id) (node_id)
3148 Commit port traffic management hierarchy
3149 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3151 Commit the traffic management hierarchy on the port::
3153 testpmd> port tm hierarchy commit (port_id) (clean_on_fail)
3157 * ``clean_on_fail``: When set to non-zero, hierarchy is cleared on function
3158 call failure. On the other hand, hierarchy is preserved when this parameter
3161 Set port traffic management mark VLAN dei
3162 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3164 Enables/Disables the traffic management marking on the port for VLAN packets::
3166 testpmd> set port tm mark vlan_dei <port_id> <green> <yellow> <red>
3170 * ``port_id``: The port which on which VLAN packets marked as ``green`` or
3171 ``yellow`` or ``red`` will have dei bit enabled
3173 * ``green`` enable 1, disable 0 marking for dei bit of VLAN packets marked as green
3175 * ``yellow`` enable 1, disable 0 marking for dei bit of VLAN packets marked as yellow
3177 * ``red`` enable 1, disable 0 marking for dei bit of VLAN packets marked as red
3179 Set port traffic management mark IP dscp
3180 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3182 Enables/Disables the traffic management marking on the port for IP dscp packets::
3184 testpmd> set port tm mark ip_dscp <port_id> <green> <yellow> <red>
3188 * ``port_id``: The port which on which IP packets marked as ``green`` or
3189 ``yellow`` or ``red`` will have IP dscp bits updated
3191 * ``green`` enable 1, disable 0 marking IP dscp to low drop precedence for green packets
3193 * ``yellow`` enable 1, disable 0 marking IP dscp to medium drop precedence for yellow packets
3195 * ``red`` enable 1, disable 0 marking IP dscp to high drop precedence for red packets
3197 Set port traffic management mark IP ecn
3198 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3200 Enables/Disables the traffic management marking on the port for IP ecn packets::
3202 testpmd> set port tm mark ip_ecn <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 ecn bits updated
3209 * ``green`` enable 1, disable 0 marking IP ecn for green marked packets with ecn of 2'b01 or 2'b10
3210 to ecn of 2'b11 when IP is caring TCP or SCTP
3212 * ``yellow`` enable 1, disable 0 marking IP ecn for yellow marked packets with ecn of 2'b01 or 2'b10
3213 to ecn of 2'b11 when IP is caring TCP or SCTP
3215 * ``red`` enable 1, disable 0 marking IP ecn for yellow marked packets with ecn of 2'b01 or 2'b10
3216 to ecn of 2'b11 when IP is caring TCP or SCTP
3221 This section details the available filter functions that are available.
3223 Note these functions interface the deprecated legacy filtering framework,
3224 superseded by *rte_flow*. See `Flow rules management`_.
3226 .. _testpmd_flow_director:
3231 Set flow director's input masks::
3233 flow_director_mask (port_id) mode IP vlan (vlan_value) \
3234 src_mask (ipv4_src) (ipv6_src) (src_port) \
3235 dst_mask (ipv4_dst) (ipv6_dst) (dst_port)
3237 flow_director_mask (port_id) mode MAC-VLAN vlan (vlan_value)
3239 flow_director_mask (port_id) mode Tunnel vlan (vlan_value) \
3240 mac (mac_value) tunnel-type (tunnel_type_value) \
3241 tunnel-id (tunnel_id_value)
3243 Example, to set flow director mask on port 0::
3245 testpmd> flow_director_mask 0 mode IP vlan 0xefff \
3246 src_mask 255.255.255.255 \
3247 FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF \
3248 dst_mask 255.255.255.255 \
3249 FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF
3251 flow_director_flex_payload
3252 ~~~~~~~~~~~~~~~~~~~~~~~~~~
3254 Configure flexible payload selection::
3256 flow_director_flex_payload (port_id) (raw|l2|l3|l4) (config)
3258 For example, to select the first 16 bytes from the offset 4 (bytes) of packet's payload as flexible payload::
3260 testpmd> flow_director_flex_payload 0 l4 \
3261 (4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19)
3264 .. _testpmd_rte_flow:
3266 Flow rules management
3267 ---------------------
3269 Control of the generic flow API (*rte_flow*) is fully exposed through the
3270 ``flow`` command (validation, creation, destruction, queries and operation
3273 Considering *rte_flow* overlaps with all `Filter Functions`_, using both
3274 features simultaneously may cause undefined side-effects and is therefore
3280 Because the ``flow`` command uses dynamic tokens to handle the large number
3281 of possible flow rules combinations, its behavior differs slightly from
3282 other commands, in particular:
3284 - Pressing *?* or the *<tab>* key displays contextual help for the current
3285 token, not that of the entire command.
3287 - Optional and repeated parameters are supported (provided they are listed
3288 in the contextual help).
3290 The first parameter stands for the operation mode. Possible operations and
3291 their general syntax are described below. They are covered in detail in the
3294 - Check whether a flow rule can be created::
3296 flow validate {port_id}
3297 [group {group_id}] [priority {level}] [ingress] [egress] [transfer]
3298 pattern {item} [/ {item} [...]] / end
3299 actions {action} [/ {action} [...]] / end
3301 - Create a flow rule::
3303 flow create {port_id}
3304 [group {group_id}] [priority {level}] [ingress] [egress] [transfer]
3305 pattern {item} [/ {item} [...]] / end
3306 actions {action} [/ {action} [...]] / end
3308 - Destroy specific flow rules::
3310 flow destroy {port_id} rule {rule_id} [...]
3312 - Destroy all flow rules::
3314 flow flush {port_id}
3316 - Query an existing flow rule::
3318 flow query {port_id} {rule_id} {action}
3320 - List existing flow rules sorted by priority, filtered by group
3323 flow list {port_id} [group {group_id}] [...]
3325 - Restrict ingress traffic to the defined flow rules::
3327 flow isolate {port_id} {boolean}
3329 - Dump internal representation information of all flows in hardware::
3331 flow dump {port_id} all {output_file}
3335 flow dump {port_id} rule {rule_id} {output_file}
3337 - List and destroy aged flow rules::
3339 flow aged {port_id} [destroy]
3341 - Tunnel offload - create a tunnel stub::
3343 flow tunnel create {port_id} type {tunnel_type}
3345 - Tunnel offload - destroy a tunnel stub::
3347 flow tunnel destroy {port_id} id {tunnel_id}
3349 - Tunnel offload - list port tunnel stubs::
3351 flow tunnel list {port_id}
3353 Creating a tunnel stub for offload
3354 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3356 ``flow tunnel create`` setup a tunnel stub for tunnel offload flow rules::
3358 flow tunnel create {port_id} type {tunnel_type}
3360 If successful, it will return a tunnel stub ID usable with other commands::
3362 port [...]: flow tunnel #[...] type [...]
3364 Tunnel stub ID is relative to a port.
3366 Destroying tunnel offload stub
3367 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3369 ``flow tunnel destroy`` destroy port tunnel stub::
3371 flow tunnel destroy {port_id} id {tunnel_id}
3373 Listing tunnel offload stubs
3374 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3376 ``flow tunnel list`` list port tunnel offload stubs::
3378 flow tunnel list {port_id}
3380 Validating flow rules
3381 ~~~~~~~~~~~~~~~~~~~~~
3383 ``flow validate`` reports whether a flow rule would be accepted by the
3384 underlying device in its current state but stops short of creating it. It is
3385 bound to ``rte_flow_validate()``::
3387 flow validate {port_id}
3388 [group {group_id}] [priority {level}] [ingress] [egress] [transfer]
3389 pattern {item} [/ {item} [...]] / end
3390 actions {action} [/ {action} [...]] / end
3392 If successful, it will show::
3396 Otherwise it will show an error message of the form::
3398 Caught error type [...] ([...]): [...]
3400 This command uses the same parameters as ``flow create``, their format is
3401 described in `Creating flow rules`_.
3403 Check whether redirecting any Ethernet packet received on port 0 to RX queue
3404 index 6 is supported::
3406 testpmd> flow validate 0 ingress pattern eth / end
3407 actions queue index 6 / end
3411 Port 0 does not support TCPv6 rules::
3413 testpmd> flow validate 0 ingress pattern eth / ipv6 / tcp / end
3415 Caught error type 9 (specific pattern item): Invalid argument
3421 ``flow create`` validates and creates the specified flow rule. It is bound
3422 to ``rte_flow_create()``::
3424 flow create {port_id}
3425 [group {group_id}] [priority {level}] [ingress] [egress] [transfer]
3426 [tunnel_set {tunnel_id}] [tunnel_match {tunnel_id}]
3427 pattern {item} [/ {item} [...]] / end
3428 actions {action} [/ {action} [...]] / end
3430 If successful, it will return a flow rule ID usable with other commands::
3432 Flow rule #[...] created
3434 Otherwise it will show an error message of the form::
3436 Caught error type [...] ([...]): [...]
3438 Parameters describe in the following order:
3440 - Attributes (*group*, *priority*, *ingress*, *egress*, *transfer* tokens).
3441 - Tunnel offload specification (tunnel_set, tunnel_match)
3442 - A matching pattern, starting with the *pattern* token and terminated by an
3444 - Actions, starting with the *actions* token and terminated by an *end*
3447 These translate directly to *rte_flow* objects provided as-is to the
3448 underlying functions.
3450 The shortest valid definition only comprises mandatory tokens::
3452 testpmd> flow create 0 pattern end actions end
3454 Note that PMDs may refuse rules that essentially do nothing such as this
3457 **All unspecified object values are automatically initialized to 0.**
3462 These tokens affect flow rule attributes (``struct rte_flow_attr``) and are
3463 specified before the ``pattern`` token.
3465 - ``group {group id}``: priority group.
3466 - ``priority {level}``: priority level within group.
3467 - ``ingress``: rule applies to ingress traffic.
3468 - ``egress``: rule applies to egress traffic.
3469 - ``transfer``: apply rule directly to endpoints found in pattern.
3471 Each instance of an attribute specified several times overrides the previous
3472 value as shown below (group 4 is used)::
3474 testpmd> flow create 0 group 42 group 24 group 4 [...]
3476 Note that once enabled, ``ingress`` and ``egress`` cannot be disabled.
3478 While not specifying a direction is an error, some rules may allow both
3481 Most rules affect RX therefore contain the ``ingress`` token::
3483 testpmd> flow create 0 ingress pattern [...]
3488 Indicate tunnel offload rule type
3490 - ``tunnel_set {tunnel_id}``: mark rule as tunnel offload decap_set type.
3491 - ``tunnel_match {tunnel_id}``: mark rule as tunel offload match type.
3496 A matching pattern starts after the ``pattern`` token. It is made of pattern
3497 items and is terminated by a mandatory ``end`` item.
3499 Items are named after their type (*RTE_FLOW_ITEM_TYPE_* from ``enum
3500 rte_flow_item_type``).
3502 The ``/`` token is used as a separator between pattern items as shown
3505 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end [...]
3507 Note that protocol items like these must be stacked from lowest to highest
3508 layer to make sense. For instance, the following rule is either invalid or
3509 unlikely to match any packet::
3511 testpmd> flow create 0 ingress pattern eth / udp / ipv4 / end [...]
3513 More information on these restrictions can be found in the *rte_flow*
3516 Several items support additional specification structures, for example
3517 ``ipv4`` allows specifying source and destination addresses as follows::
3519 testpmd> flow create 0 ingress pattern eth / ipv4 src is 10.1.1.1
3520 dst is 10.2.0.0 / end [...]
3522 This rule matches all IPv4 traffic with the specified properties.
3524 In this example, ``src`` and ``dst`` are field names of the underlying
3525 ``struct rte_flow_item_ipv4`` object. All item properties can be specified
3526 in a similar fashion.
3528 The ``is`` token means that the subsequent value must be matched exactly,
3529 and assigns ``spec`` and ``mask`` fields in ``struct rte_flow_item``
3530 accordingly. Possible assignment tokens are:
3532 - ``is``: match value perfectly (with full bit-mask).
3533 - ``spec``: match value according to configured bit-mask.
3534 - ``last``: specify upper bound to establish a range.
3535 - ``mask``: specify bit-mask with relevant bits set to one.
3536 - ``prefix``: generate bit-mask with <prefix-length> most-significant bits set to one.
3538 These yield identical results::
3540 ipv4 src is 10.1.1.1
3544 ipv4 src spec 10.1.1.1 src mask 255.255.255.255
3548 ipv4 src spec 10.1.1.1 src prefix 32
3552 ipv4 src is 10.1.1.1 src last 10.1.1.1 # range with a single value
3556 ipv4 src is 10.1.1.1 src last 0 # 0 disables range
3558 Inclusive ranges can be defined with ``last``::
3560 ipv4 src is 10.1.1.1 src last 10.2.3.4 # 10.1.1.1 to 10.2.3.4
3562 Note that ``mask`` affects both ``spec`` and ``last``::
3564 ipv4 src is 10.1.1.1 src last 10.2.3.4 src mask 255.255.0.0
3565 # matches 10.1.0.0 to 10.2.255.255
3567 Properties can be modified multiple times::
3569 ipv4 src is 10.1.1.1 src is 10.1.2.3 src is 10.2.3.4 # matches 10.2.3.4
3573 ipv4 src is 10.1.1.1 src prefix 24 src prefix 16 # matches 10.1.0.0/16
3578 This section lists supported pattern items and their attributes, if any.
3580 - ``end``: end list of pattern items.
3582 - ``void``: no-op pattern item.
3584 - ``invert``: perform actions when pattern does not match.
3586 - ``any``: match any protocol for the current layer.
3588 - ``num {unsigned}``: number of layers covered.
3590 - ``pf``: match traffic from/to the physical function.
3592 - ``vf``: match traffic from/to a virtual function ID.
3594 - ``id {unsigned}``: VF ID.
3596 - ``phy_port``: match traffic from/to a specific physical port.
3598 - ``index {unsigned}``: physical port index.
3600 - ``port_id``: match traffic from/to a given DPDK port ID.
3602 - ``id {unsigned}``: DPDK port ID.
3604 - ``mark``: match value set in previously matched flow rule using the mark action.
3606 - ``id {unsigned}``: arbitrary integer value.
3608 - ``raw``: match an arbitrary byte string.
3610 - ``relative {boolean}``: look for pattern after the previous item.
3611 - ``search {boolean}``: search pattern from offset (see also limit).
3612 - ``offset {integer}``: absolute or relative offset for pattern.
3613 - ``limit {unsigned}``: search area limit for start of pattern.
3614 - ``pattern {string}``: byte string to look for.
3616 - ``eth``: match Ethernet header.
3618 - ``dst {MAC-48}``: destination MAC.
3619 - ``src {MAC-48}``: source MAC.
3620 - ``type {unsigned}``: EtherType or TPID.
3622 - ``vlan``: match 802.1Q/ad VLAN tag.
3624 - ``tci {unsigned}``: tag control information.
3625 - ``pcp {unsigned}``: priority code point.
3626 - ``dei {unsigned}``: drop eligible indicator.
3627 - ``vid {unsigned}``: VLAN identifier.
3628 - ``inner_type {unsigned}``: inner EtherType or TPID.
3630 - ``ipv4``: match IPv4 header.
3632 - ``version_ihl {unsigned}``: IPv4 version and IP header length.
3633 - ``tos {unsigned}``: type of service.
3634 - ``ttl {unsigned}``: time to live.
3635 - ``proto {unsigned}``: next protocol ID.
3636 - ``src {ipv4 address}``: source address.
3637 - ``dst {ipv4 address}``: destination address.
3639 - ``ipv6``: match IPv6 header.
3641 - ``tc {unsigned}``: traffic class.
3642 - ``flow {unsigned}``: flow label.
3643 - ``proto {unsigned}``: protocol (next header).
3644 - ``hop {unsigned}``: hop limit.
3645 - ``src {ipv6 address}``: source address.
3646 - ``dst {ipv6 address}``: destination address.
3648 - ``icmp``: match ICMP header.
3650 - ``type {unsigned}``: ICMP packet type.
3651 - ``code {unsigned}``: ICMP packet code.
3653 - ``udp``: match UDP header.
3655 - ``src {unsigned}``: UDP source port.
3656 - ``dst {unsigned}``: UDP destination port.
3658 - ``tcp``: match TCP header.
3660 - ``src {unsigned}``: TCP source port.
3661 - ``dst {unsigned}``: TCP destination port.
3663 - ``sctp``: match SCTP header.
3665 - ``src {unsigned}``: SCTP source port.
3666 - ``dst {unsigned}``: SCTP destination port.
3667 - ``tag {unsigned}``: validation tag.
3668 - ``cksum {unsigned}``: checksum.
3670 - ``vxlan``: match VXLAN header.
3672 - ``vni {unsigned}``: VXLAN identifier.
3673 - ``last_rsvd {unsigned}``: VXLAN last reserved 8-bits.
3675 - ``e_tag``: match IEEE 802.1BR E-Tag header.
3677 - ``grp_ecid_b {unsigned}``: GRP and E-CID base.
3679 - ``nvgre``: match NVGRE header.
3681 - ``tni {unsigned}``: virtual subnet ID.
3683 - ``mpls``: match MPLS header.
3685 - ``label {unsigned}``: MPLS label.
3687 - ``gre``: match GRE header.
3689 - ``protocol {unsigned}``: protocol type.
3691 - ``gre_key``: match GRE optional key field.
3693 - ``value {unsigned}``: key value.
3695 - ``fuzzy``: fuzzy pattern match, expect faster than default.
3697 - ``thresh {unsigned}``: accuracy threshold.
3699 - ``gtp``, ``gtpc``, ``gtpu``: match GTPv1 header.
3701 - ``teid {unsigned}``: tunnel endpoint identifier.
3703 - ``geneve``: match GENEVE header.
3705 - ``vni {unsigned}``: virtual network identifier.
3706 - ``protocol {unsigned}``: protocol type.
3708 - ``geneve-opt``: match GENEVE header option.
3710 - ``class {unsigned}``: GENEVE option class.
3711 - ``type {unsigned}``: GENEVE option type.
3712 - ``length {unsigned}``: GENEVE option length in 32-bit words.
3713 - ``data {hex string}``: GENEVE option data, the length is defined by
3716 - ``vxlan-gpe``: match VXLAN-GPE header.
3718 - ``vni {unsigned}``: VXLAN-GPE identifier.
3720 - ``arp_eth_ipv4``: match ARP header for Ethernet/IPv4.
3722 - ``sha {MAC-48}``: sender hardware address.
3723 - ``spa {ipv4 address}``: sender IPv4 address.
3724 - ``tha {MAC-48}``: target hardware address.
3725 - ``tpa {ipv4 address}``: target IPv4 address.
3727 - ``ipv6_ext``: match presence of any IPv6 extension header.
3729 - ``next_hdr {unsigned}``: next header.
3731 - ``icmp6``: match any ICMPv6 header.
3733 - ``type {unsigned}``: ICMPv6 type.
3734 - ``code {unsigned}``: ICMPv6 code.
3736 - ``icmp6_nd_ns``: match ICMPv6 neighbor discovery solicitation.
3738 - ``target_addr {ipv6 address}``: target address.
3740 - ``icmp6_nd_na``: match ICMPv6 neighbor discovery advertisement.
3742 - ``target_addr {ipv6 address}``: target address.
3744 - ``icmp6_nd_opt``: match presence of any ICMPv6 neighbor discovery option.
3746 - ``type {unsigned}``: ND option type.
3748 - ``icmp6_nd_opt_sla_eth``: match ICMPv6 neighbor discovery source Ethernet
3749 link-layer address option.
3751 - ``sla {MAC-48}``: source Ethernet LLA.
3753 - ``icmp6_nd_opt_tla_eth``: match ICMPv6 neighbor discovery target Ethernet
3754 link-layer address option.
3756 - ``tla {MAC-48}``: target Ethernet LLA.
3758 - ``meta``: match application specific metadata.
3760 - ``data {unsigned}``: metadata value.
3762 - ``gtp_psc``: match GTP PDU extension header with type 0x85.
3764 - ``pdu_type {unsigned}``: PDU type.
3766 - ``qfi {unsigned}``: QoS flow identifier.
3768 - ``pppoes``, ``pppoed``: match PPPoE header.
3770 - ``session_id {unsigned}``: session identifier.
3772 - ``pppoe_proto_id``: match PPPoE session protocol identifier.
3774 - ``proto_id {unsigned}``: PPP protocol identifier.
3776 - ``l2tpv3oip``: match L2TPv3 over IP header.
3778 - ``session_id {unsigned}``: L2TPv3 over IP session identifier.
3780 - ``ah``: match AH header.
3782 - ``spi {unsigned}``: security parameters index.
3784 - ``pfcp``: match PFCP header.
3786 - ``s_field {unsigned}``: S field.
3787 - ``seid {unsigned}``: session endpoint identifier.
3789 - ``integrity``: match packet integrity.
3791 - ``level {unsigned}``: Packet encapsulation level the item should
3792 apply to. See rte_flow_action_rss for details.
3793 - ``value {unsigned}``: A bitmask that specify what packet elements
3794 must be matched for integrity.
3796 - ``conntrack``: match conntrack state.
3801 A list of actions starts after the ``actions`` token in the same fashion as
3802 `Matching pattern`_; actions are separated by ``/`` tokens and the list is
3803 terminated by a mandatory ``end`` action.
3805 Actions are named after their type (*RTE_FLOW_ACTION_TYPE_* from ``enum
3806 rte_flow_action_type``).
3808 Dropping all incoming UDPv4 packets can be expressed as follows::
3810 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
3813 Several actions have configurable properties which must be specified when
3814 there is no valid default value. For example, ``queue`` requires a target
3817 This rule redirects incoming UDPv4 traffic to queue index 6::
3819 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
3820 actions queue index 6 / end
3822 While this one could be rejected by PMDs (unspecified queue index)::
3824 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
3827 As defined by *rte_flow*, the list is not ordered, all actions of a given
3828 rule are performed simultaneously. These are equivalent::
3830 queue index 6 / void / mark id 42 / end
3834 void / mark id 42 / queue index 6 / end
3836 All actions in a list should have different types, otherwise only the last
3837 action of a given type is taken into account::
3839 queue index 4 / queue index 5 / queue index 6 / end # will use queue 6
3843 drop / drop / drop / end # drop is performed only once
3847 mark id 42 / queue index 3 / mark id 24 / end # mark will be 24
3849 Considering they are performed simultaneously, opposite and overlapping
3850 actions can sometimes be combined when the end result is unambiguous::
3852 drop / queue index 6 / end # drop has no effect
3856 queue index 6 / rss queues 6 7 8 / end # queue has no effect
3860 drop / passthru / end # drop has no effect
3862 Note that PMDs may still refuse such combinations.
3867 This section lists supported actions and their attributes, if any.
3869 - ``end``: end list of actions.
3871 - ``void``: no-op action.
3873 - ``passthru``: let subsequent rule process matched packets.
3875 - ``jump``: redirect traffic to group on device.
3877 - ``group {unsigned}``: group to redirect to.
3879 - ``mark``: attach 32 bit value to packets.
3881 - ``id {unsigned}``: 32 bit value to return with packets.
3883 - ``flag``: flag packets.
3885 - ``queue``: assign packets to a given queue index.
3887 - ``index {unsigned}``: queue index to use.
3889 - ``drop``: drop packets (note: passthru has priority).
3891 - ``count``: enable counters for this rule.
3893 - ``rss``: spread packets among several queues.
3895 - ``func {hash function}``: RSS hash function to apply, allowed tokens are
3896 ``toeplitz``, ``simple_xor``, ``symmetric_toeplitz`` and ``default``.
3898 - ``level {unsigned}``: encapsulation level for ``types``.
3900 - ``types [{RSS hash type} [...]] end``: specific RSS hash types.
3901 Note that an empty list does not disable RSS but instead requests
3902 unspecified "best-effort" settings.
3904 - ``key {string}``: RSS hash key, overrides ``key_len``.
3906 - ``key_len {unsigned}``: RSS hash key length in bytes, can be used in
3907 conjunction with ``key`` to pad or truncate it.
3909 - ``queues [{unsigned} [...]] end``: queue indices to use.
3911 - ``pf``: direct traffic to physical function.
3913 - ``vf``: direct traffic to a virtual function ID.
3915 - ``original {boolean}``: use original VF ID if possible.
3916 - ``id {unsigned}``: VF ID.
3918 - ``phy_port``: direct packets to physical port index.
3920 - ``original {boolean}``: use original port index if possible.
3921 - ``index {unsigned}``: physical port index.
3923 - ``port_id``: direct matching traffic to a given DPDK port ID.
3925 - ``original {boolean}``: use original DPDK port ID if possible.
3926 - ``id {unsigned}``: DPDK port ID.
3928 - ``of_set_mpls_ttl``: OpenFlow's ``OFPAT_SET_MPLS_TTL``.
3930 - ``mpls_ttl``: MPLS TTL.
3932 - ``of_dec_mpls_ttl``: OpenFlow's ``OFPAT_DEC_MPLS_TTL``.
3934 - ``of_set_nw_ttl``: OpenFlow's ``OFPAT_SET_NW_TTL``.
3936 - ``nw_ttl``: IP TTL.
3938 - ``of_dec_nw_ttl``: OpenFlow's ``OFPAT_DEC_NW_TTL``.
3940 - ``of_copy_ttl_out``: OpenFlow's ``OFPAT_COPY_TTL_OUT``.
3942 - ``of_copy_ttl_in``: OpenFlow's ``OFPAT_COPY_TTL_IN``.
3944 - ``of_pop_vlan``: OpenFlow's ``OFPAT_POP_VLAN``.
3946 - ``of_push_vlan``: OpenFlow's ``OFPAT_PUSH_VLAN``.
3948 - ``ethertype``: Ethertype.
3950 - ``of_set_vlan_vid``: OpenFlow's ``OFPAT_SET_VLAN_VID``.
3952 - ``vlan_vid``: VLAN id.
3954 - ``of_set_vlan_pcp``: OpenFlow's ``OFPAT_SET_VLAN_PCP``.
3956 - ``vlan_pcp``: VLAN priority.
3958 - ``of_pop_mpls``: OpenFlow's ``OFPAT_POP_MPLS``.
3960 - ``ethertype``: Ethertype.
3962 - ``of_push_mpls``: OpenFlow's ``OFPAT_PUSH_MPLS``.
3964 - ``ethertype``: Ethertype.
3966 - ``vxlan_encap``: Performs a VXLAN encapsulation, outer layer configuration
3967 is done through `Config VXLAN Encap outer layers`_.
3969 - ``vxlan_decap``: Performs a decapsulation action by stripping all headers of
3970 the VXLAN tunnel network overlay from the matched flow.
3972 - ``nvgre_encap``: Performs a NVGRE encapsulation, outer layer configuration
3973 is done through `Config NVGRE Encap outer layers`_.
3975 - ``nvgre_decap``: Performs a decapsulation action by stripping all headers of
3976 the NVGRE tunnel network overlay from the matched flow.
3978 - ``l2_encap``: Performs a L2 encapsulation, L2 configuration
3979 is done through `Config L2 Encap`_.
3981 - ``l2_decap``: Performs a L2 decapsulation, L2 configuration
3982 is done through `Config L2 Decap`_.
3984 - ``mplsogre_encap``: Performs a MPLSoGRE encapsulation, outer layer
3985 configuration is done through `Config MPLSoGRE Encap outer layers`_.
3987 - ``mplsogre_decap``: Performs a MPLSoGRE decapsulation, outer layer
3988 configuration is done through `Config MPLSoGRE Decap outer layers`_.
3990 - ``mplsoudp_encap``: Performs a MPLSoUDP encapsulation, outer layer
3991 configuration is done through `Config MPLSoUDP Encap outer layers`_.
3993 - ``mplsoudp_decap``: Performs a MPLSoUDP decapsulation, outer layer
3994 configuration is done through `Config MPLSoUDP Decap outer layers`_.
3996 - ``set_ipv4_src``: Set a new IPv4 source address in the outermost IPv4 header.
3998 - ``ipv4_addr``: New IPv4 source address.
4000 - ``set_ipv4_dst``: Set a new IPv4 destination address in the outermost IPv4
4003 - ``ipv4_addr``: New IPv4 destination address.
4005 - ``set_ipv6_src``: Set a new IPv6 source address in the outermost IPv6 header.
4007 - ``ipv6_addr``: New IPv6 source address.
4009 - ``set_ipv6_dst``: Set a new IPv6 destination address in the outermost IPv6
4012 - ``ipv6_addr``: New IPv6 destination address.
4014 - ``set_tp_src``: Set a new source port number in the outermost TCP/UDP
4017 - ``port``: New TCP/UDP source port number.
4019 - ``set_tp_dst``: Set a new destination port number in the outermost TCP/UDP
4022 - ``port``: New TCP/UDP destination port number.
4024 - ``mac_swap``: Swap the source and destination MAC addresses in the outermost
4027 - ``dec_ttl``: Performs a decrease TTL value action
4029 - ``set_ttl``: Set TTL value with specified value
4030 - ``ttl_value {unsigned}``: The new TTL value to be set
4032 - ``set_mac_src``: set source MAC address
4034 - ``mac_addr {MAC-48}``: new source MAC address
4036 - ``set_mac_dst``: set destination MAC address
4038 - ``mac_addr {MAC-48}``: new destination MAC address
4040 - ``inc_tcp_seq``: Increase sequence number in the outermost TCP header.
4042 - ``value {unsigned}``: Value to increase TCP sequence number by.
4044 - ``dec_tcp_seq``: Decrease sequence number in the outermost TCP header.
4046 - ``value {unsigned}``: Value to decrease TCP sequence number by.
4048 - ``inc_tcp_ack``: Increase acknowledgment number in the outermost TCP header.
4050 - ``value {unsigned}``: Value to increase TCP acknowledgment number by.
4052 - ``dec_tcp_ack``: Decrease acknowledgment number in the outermost TCP header.
4054 - ``value {unsigned}``: Value to decrease TCP acknowledgment number by.
4056 - ``set_ipv4_dscp``: Set IPv4 DSCP value with specified value
4058 - ``dscp_value {unsigned}``: The new DSCP value to be set
4060 - ``set_ipv6_dscp``: Set IPv6 DSCP value with specified value
4062 - ``dscp_value {unsigned}``: The new DSCP value to be set
4064 - ``indirect``: Use indirect action created via
4065 ``flow indirect_action {port_id} create``
4067 - ``indirect_action_id {unsigned}``: Indirect action ID to use
4069 - ``color``: Color the packet to reflect the meter color result
4071 - ``type {value}``: Set color type with specified value(green/yellow/red)
4073 Destroying flow rules
4074 ~~~~~~~~~~~~~~~~~~~~~
4076 ``flow destroy`` destroys one or more rules from their rule ID (as returned
4077 by ``flow create``), this command calls ``rte_flow_destroy()`` as many
4078 times as necessary::
4080 flow destroy {port_id} rule {rule_id} [...]
4082 If successful, it will show::
4084 Flow rule #[...] destroyed
4086 It does not report anything for rule IDs that do not exist. The usual error
4087 message is shown when a rule cannot be destroyed::
4089 Caught error type [...] ([...]): [...]
4091 ``flow flush`` destroys all rules on a device and does not take extra
4092 arguments. It is bound to ``rte_flow_flush()``::
4094 flow flush {port_id}
4096 Any errors are reported as above.
4098 Creating several rules and destroying them::
4100 testpmd> flow create 0 ingress pattern eth / ipv6 / end
4101 actions queue index 2 / end
4102 Flow rule #0 created
4103 testpmd> flow create 0 ingress pattern eth / ipv4 / end
4104 actions queue index 3 / end
4105 Flow rule #1 created
4106 testpmd> flow destroy 0 rule 0 rule 1
4107 Flow rule #1 destroyed
4108 Flow rule #0 destroyed
4111 The same result can be achieved using ``flow flush``::
4113 testpmd> flow create 0 ingress pattern eth / ipv6 / end
4114 actions queue index 2 / end
4115 Flow rule #0 created
4116 testpmd> flow create 0 ingress pattern eth / ipv4 / end
4117 actions queue index 3 / end
4118 Flow rule #1 created
4119 testpmd> flow flush 0
4122 Non-existent rule IDs are ignored::
4124 testpmd> flow create 0 ingress pattern eth / ipv6 / end
4125 actions queue index 2 / end
4126 Flow rule #0 created
4127 testpmd> flow create 0 ingress pattern eth / ipv4 / end
4128 actions queue index 3 / end
4129 Flow rule #1 created
4130 testpmd> flow destroy 0 rule 42 rule 10 rule 2
4132 testpmd> flow destroy 0 rule 0
4133 Flow rule #0 destroyed
4139 ``flow query`` queries a specific action of a flow rule having that
4140 ability. Such actions collect information that can be reported using this
4141 command. It is bound to ``rte_flow_query()``::
4143 flow query {port_id} {rule_id} {action}
4145 If successful, it will display either the retrieved data for known actions
4146 or the following message::
4148 Cannot display result for action type [...] ([...])
4150 Otherwise, it will complain either that the rule does not exist or that some
4153 Flow rule #[...] not found
4157 Caught error type [...] ([...]): [...]
4159 Currently only the ``count`` action is supported. This action reports the
4160 number of packets that hit the flow rule and the total number of bytes. Its
4161 output has the following format::
4164 hits_set: [...] # whether "hits" contains a valid value
4165 bytes_set: [...] # whether "bytes" contains a valid value
4166 hits: [...] # number of packets
4167 bytes: [...] # number of bytes
4169 Querying counters for TCPv6 packets redirected to queue 6::
4171 testpmd> flow create 0 ingress pattern eth / ipv6 / tcp / end
4172 actions queue index 6 / count / end
4173 Flow rule #4 created
4174 testpmd> flow query 0 4 count
4185 ``flow list`` lists existing flow rules sorted by priority and optionally
4186 filtered by group identifiers::
4188 flow list {port_id} [group {group_id}] [...]
4190 This command only fails with the following message if the device does not
4195 Output consists of a header line followed by a short description of each
4196 flow rule, one per line. There is no output at all when no flow rules are
4197 configured on the device::
4199 ID Group Prio Attr Rule
4200 [...] [...] [...] [...] [...]
4202 ``Attr`` column flags:
4204 - ``i`` for ``ingress``.
4205 - ``e`` for ``egress``.
4207 Creating several flow rules and listing them::
4209 testpmd> flow create 0 ingress pattern eth / ipv4 / end
4210 actions queue index 6 / end
4211 Flow rule #0 created
4212 testpmd> flow create 0 ingress pattern eth / ipv6 / end
4213 actions queue index 2 / end
4214 Flow rule #1 created
4215 testpmd> flow create 0 priority 5 ingress pattern eth / ipv4 / udp / end
4216 actions rss queues 6 7 8 end / end
4217 Flow rule #2 created
4218 testpmd> flow list 0
4219 ID Group Prio Attr Rule
4220 0 0 0 i- ETH IPV4 => QUEUE
4221 1 0 0 i- ETH IPV6 => QUEUE
4222 2 0 5 i- ETH IPV4 UDP => RSS
4225 Rules are sorted by priority (i.e. group ID first, then priority level)::
4227 testpmd> flow list 1
4228 ID Group Prio Attr Rule
4229 0 0 0 i- ETH => COUNT
4230 6 0 500 i- ETH IPV6 TCP => DROP COUNT
4231 5 0 1000 i- ETH IPV6 ICMP => QUEUE
4232 1 24 0 i- ETH IPV4 UDP => QUEUE
4233 4 24 10 i- ETH IPV4 TCP => DROP
4234 3 24 20 i- ETH IPV4 => DROP
4235 2 24 42 i- ETH IPV4 UDP => QUEUE
4236 7 63 0 i- ETH IPV6 UDP VXLAN => MARK QUEUE
4239 Output can be limited to specific groups::
4241 testpmd> flow list 1 group 0 group 63
4242 ID Group Prio Attr Rule
4243 0 0 0 i- ETH => COUNT
4244 6 0 500 i- ETH IPV6 TCP => DROP COUNT
4245 5 0 1000 i- ETH IPV6 ICMP => QUEUE
4246 7 63 0 i- ETH IPV6 UDP VXLAN => MARK QUEUE
4249 Toggling isolated mode
4250 ~~~~~~~~~~~~~~~~~~~~~~
4252 ``flow isolate`` can be used to tell the underlying PMD that ingress traffic
4253 must only be injected from the defined flow rules; that no default traffic
4254 is expected outside those rules and the driver is free to assign more
4255 resources to handle them. It is bound to ``rte_flow_isolate()``::
4257 flow isolate {port_id} {boolean}
4259 If successful, enabling or disabling isolated mode shows either::
4261 Ingress traffic on port [...]
4262 is now restricted to the defined flow rules
4266 Ingress traffic on port [...]
4267 is not restricted anymore to the defined flow rules
4269 Otherwise, in case of error::
4271 Caught error type [...] ([...]): [...]
4273 Mainly due to its side effects, PMDs supporting this mode may not have the
4274 ability to toggle it more than once without reinitializing affected ports
4275 first (e.g. by exiting testpmd).
4277 Enabling isolated mode::
4279 testpmd> flow isolate 0 true
4280 Ingress traffic on port 0 is now restricted to the defined flow rules
4283 Disabling isolated mode::
4285 testpmd> flow isolate 0 false
4286 Ingress traffic on port 0 is not restricted anymore to the defined flow rules
4289 Dumping HW internal information
4290 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4292 ``flow dump`` dumps the hardware's internal representation information of
4293 all flows. It is bound to ``rte_flow_dev_dump()``::
4295 flow dump {port_id} {output_file}
4297 If successful, it will show::
4301 Otherwise, it will complain error occurred::
4303 Caught error type [...] ([...]): [...]
4305 Listing and destroying aged flow rules
4306 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4308 ``flow aged`` simply lists aged flow rules be get from api ``rte_flow_get_aged_flows``,
4309 and ``destroy`` parameter can be used to destroy those flow rules in PMD.
4311 flow aged {port_id} [destroy]
4313 Listing current aged flow rules::
4315 testpmd> flow aged 0
4316 Port 0 total aged flows: 0
4317 testpmd> flow create 0 ingress pattern eth / ipv4 src is 2.2.2.14 / end
4318 actions age timeout 5 / queue index 0 / end
4319 Flow rule #0 created
4320 testpmd> flow create 0 ingress pattern eth / ipv4 src is 2.2.2.15 / end
4321 actions age timeout 4 / queue index 0 / end
4322 Flow rule #1 created
4323 testpmd> flow create 0 ingress pattern eth / ipv4 src is 2.2.2.16 / end
4324 actions age timeout 2 / queue index 0 / end
4325 Flow rule #2 created
4326 testpmd> flow create 0 ingress pattern eth / ipv4 src is 2.2.2.17 / end
4327 actions age timeout 3 / queue index 0 / end
4328 Flow rule #3 created
4331 Aged Rules are simply list as command ``flow list {port_id}``, but strip the detail rule
4332 information, all the aged flows are sorted by the longest timeout time. For example, if
4333 those rules be configured in the same time, ID 2 will be the first aged out rule, the next
4334 will be ID 3, ID 1, ID 0::
4336 testpmd> flow aged 0
4337 Port 0 total aged flows: 4
4344 If attach ``destroy`` parameter, the command will destroy all the list aged flow rules.
4346 testpmd> flow aged 0 destroy
4347 Port 0 total aged flows: 4
4354 Flow rule #2 destroyed
4355 Flow rule #3 destroyed
4356 Flow rule #1 destroyed
4357 Flow rule #0 destroyed
4358 4 flows be destroyed
4359 testpmd> flow aged 0
4360 Port 0 total aged flows: 0
4362 Creating indirect actions
4363 ~~~~~~~~~~~~~~~~~~~~~~~~~
4365 ``flow indirect_action {port_id} create`` creates indirect action with optional
4366 indirect action ID. It is bound to ``rte_flow_action_handle_create()``::
4368 flow indirect_action {port_id} create [action_id {indirect_action_id}]
4369 [ingress] [egress] [transfer] action {action} / end
4371 If successful, it will show::
4373 Indirect action #[...] created
4375 Otherwise, it will complain either that indirect action already exists or that
4376 some error occurred::
4378 Indirect action #[...] is already assigned, delete it first
4382 Caught error type [...] ([...]): [...]
4384 Create indirect rss action with id 100 to queues 1 and 2 on port 0::
4386 testpmd> flow indirect_action 0 create action_id 100 \
4387 ingress action rss queues 1 2 end / end
4389 Create indirect rss action with id assigned by testpmd to queues 1 and 2 on
4392 testpmd> flow indirect_action 0 create action_id \
4393 ingress action rss queues 0 1 end / end
4395 Updating indirect actions
4396 ~~~~~~~~~~~~~~~~~~~~~~~~~
4398 ``flow indirect_action {port_id} update`` updates configuration of the indirect
4399 action from its indirect action ID (as returned by
4400 ``flow indirect_action {port_id} create``). It is bound to
4401 ``rte_flow_action_handle_update()``::
4403 flow indirect_action {port_id} update {indirect_action_id}
4404 action {action} / end
4406 If successful, it will show::
4408 Indirect action #[...] updated
4410 Otherwise, it will complain either that indirect action not found or that some
4413 Failed to find indirect action #[...] on port [...]
4417 Caught error type [...] ([...]): [...]
4419 Update indirect rss action having id 100 on port 0 with rss to queues 0 and 3
4420 (in create example above rss queues were 1 and 2)::
4422 testpmd> flow indirect_action 0 update 100 action rss queues 0 3 end / end
4424 Destroying indirect actions
4425 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
4427 ``flow indirect_action {port_id} destroy`` destroys one or more indirect actions
4428 from their indirect action IDs (as returned by
4429 ``flow indirect_action {port_id} create``). It is bound to
4430 ``rte_flow_action_handle_destroy()``::
4432 flow indirect_action {port_id} destroy action_id {indirect_action_id} [...]
4434 If successful, it will show::
4436 Indirect action #[...] destroyed
4438 It does not report anything for indirect action IDs that do not exist.
4439 The usual error message is shown when a indirect action cannot be destroyed::
4441 Caught error type [...] ([...]): [...]
4443 Destroy indirect actions having id 100 & 101::
4445 testpmd> flow indirect_action 0 destroy action_id 100 action_id 101
4447 Query indirect actions
4448 ~~~~~~~~~~~~~~~~~~~~~~
4450 ``flow indirect_action {port_id} query`` queries the indirect action from its
4451 indirect action ID (as returned by ``flow indirect_action {port_id} create``).
4452 It is bound to ``rte_flow_action_handle_query()``::
4454 flow indirect_action {port_id} query {indirect_action_id}
4456 Currently only rss indirect action supported. If successful, it will show::
4458 Indirect RSS action:
4461 Otherwise, it will complain either that indirect action not found or that some
4464 Failed to find indirect action #[...] on port [...]
4468 Caught error type [...] ([...]): [...]
4470 Query indirect action having id 100::
4472 testpmd> flow indirect_action 0 query 100
4474 Sample QinQ flow rules
4475 ~~~~~~~~~~~~~~~~~~~~~~
4477 Before creating QinQ rule(s) the following commands should be issued to enable QinQ::
4479 testpmd> port stop 0
4480 testpmd> vlan set extend on 0
4482 The above command sets the inner and outer TPID's to 0x8100.
4484 To change the TPID's the following commands should be used::
4486 testpmd> vlan set outer tpid 0x88A8 0
4487 testpmd> vlan set inner tpid 0x8100 0
4488 testpmd> port start 0
4490 Validate and create a QinQ rule on port 0 to steer traffic to a VF queue in a VM.
4494 testpmd> flow validate 0 ingress pattern eth / vlan tci is 123 /
4495 vlan tci is 456 / end actions vf id 1 / queue index 0 / end
4496 Flow rule #0 validated
4498 testpmd> flow create 0 ingress pattern eth / vlan tci is 4 /
4499 vlan tci is 456 / end actions vf id 123 / queue index 0 / end
4500 Flow rule #0 created
4502 testpmd> flow list 0
4503 ID Group Prio Attr Rule
4504 0 0 0 i- ETH VLAN VLAN=>VF QUEUE
4506 Validate and create a QinQ rule on port 0 to steer traffic to a queue on the host.
4510 testpmd> flow validate 0 ingress pattern eth / vlan tci is 321 /
4511 vlan tci is 654 / end actions pf / queue index 0 / end
4512 Flow rule #1 validated
4514 testpmd> flow create 0 ingress pattern eth / vlan tci is 321 /
4515 vlan tci is 654 / end actions pf / queue index 1 / end
4516 Flow rule #1 created
4518 testpmd> flow list 0
4519 ID Group Prio Attr Rule
4520 0 0 0 i- ETH VLAN VLAN=>VF QUEUE
4521 1 0 0 i- ETH VLAN VLAN=>PF QUEUE
4523 Sample VXLAN flow rules
4524 ~~~~~~~~~~~~~~~~~~~~~~~
4526 Before creating VXLAN rule(s), the UDP port should be added for VXLAN packet
4529 testpmd> rx_vxlan_port add 4789 0
4531 Create VXLAN rules on port 0 to steer traffic to PF queues.
4535 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / vxlan /
4536 eth dst is 00:11:22:33:44:55 / end actions pf / queue index 1 / end
4537 Flow rule #0 created
4539 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / vxlan vni is 3 /
4540 eth dst is 00:11:22:33:44:55 / end actions pf / queue index 2 / end
4541 Flow rule #1 created
4543 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / vxlan /
4544 eth dst is 00:11:22:33:44:55 / vlan tci is 10 / end actions pf /
4546 Flow rule #2 created
4548 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / vxlan vni is 5 /
4549 eth dst is 00:11:22:33:44:55 / vlan tci is 20 / end actions pf /
4551 Flow rule #3 created
4553 testpmd> flow create 0 ingress pattern eth dst is 00:00:00:00:01:00 / ipv4 /
4554 udp / vxlan vni is 6 / eth dst is 00:11:22:33:44:55 / end actions pf /
4556 Flow rule #4 created
4558 testpmd> flow list 0
4559 ID Group Prio Attr Rule
4560 0 0 0 i- ETH IPV4 UDP VXLAN ETH => QUEUE
4561 1 0 0 i- ETH IPV4 UDP VXLAN ETH => QUEUE
4562 2 0 0 i- ETH IPV4 UDP VXLAN ETH VLAN => QUEUE
4563 3 0 0 i- ETH IPV4 UDP VXLAN ETH VLAN => QUEUE
4564 4 0 0 i- ETH IPV4 UDP VXLAN ETH => QUEUE
4566 Sample VXLAN encapsulation rule
4567 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4569 VXLAN encapsulation outer layer has default value pre-configured in testpmd
4570 source code, those can be changed by using the following commands
4572 IPv4 VXLAN outer header::
4574 testpmd> set vxlan ip-version ipv4 vni 4 udp-src 4 udp-dst 4 ip-src 127.0.0.1
4575 ip-dst 128.0.0.1 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4576 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4579 testpmd> set vxlan-with-vlan ip-version ipv4 vni 4 udp-src 4 udp-dst 4 ip-src
4580 127.0.0.1 ip-dst 128.0.0.1 vlan-tci 34 eth-src 11:11:11:11:11:11
4581 eth-dst 22:22:22:22:22:22
4582 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4585 testpmd> set vxlan-tos-ttl ip-version ipv4 vni 4 udp-src 4 udp-dst 4 ip-tos 0
4586 ip-ttl 255 ip-src 127.0.0.1 ip-dst 128.0.0.1 eth-src 11:11:11:11:11:11
4587 eth-dst 22:22:22:22:22:22
4588 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4591 IPv6 VXLAN outer header::
4593 testpmd> set vxlan ip-version ipv6 vni 4 udp-src 4 udp-dst 4 ip-src ::1
4594 ip-dst ::2222 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4595 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4598 testpmd> set vxlan-with-vlan ip-version ipv6 vni 4 udp-src 4 udp-dst 4
4599 ip-src ::1 ip-dst ::2222 vlan-tci 34 eth-src 11:11:11:11:11:11
4600 eth-dst 22:22:22:22:22:22
4601 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4604 testpmd> set vxlan-tos-ttl ip-version ipv6 vni 4 udp-src 4 udp-dst 4
4605 ip-tos 0 ip-ttl 255 ::1 ip-dst ::2222 eth-src 11:11:11:11:11:11
4606 eth-dst 22:22:22:22:22:22
4607 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4610 Sample NVGRE encapsulation rule
4611 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4613 NVGRE encapsulation outer layer has default value pre-configured in testpmd
4614 source code, those can be changed by using the following commands
4616 IPv4 NVGRE outer header::
4618 testpmd> set nvgre ip-version ipv4 tni 4 ip-src 127.0.0.1 ip-dst 128.0.0.1
4619 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4620 testpmd> flow create 0 ingress pattern end actions nvgre_encap /
4623 testpmd> set nvgre-with-vlan ip-version ipv4 tni 4 ip-src 127.0.0.1
4624 ip-dst 128.0.0.1 vlan-tci 34 eth-src 11:11:11:11:11:11
4625 eth-dst 22:22:22:22:22:22
4626 testpmd> flow create 0 ingress pattern end actions nvgre_encap /
4629 IPv6 NVGRE outer header::
4631 testpmd> set nvgre ip-version ipv6 tni 4 ip-src ::1 ip-dst ::2222
4632 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4633 testpmd> flow create 0 ingress pattern end actions nvgre_encap /
4636 testpmd> set nvgre-with-vlan ip-version ipv6 tni 4 ip-src ::1 ip-dst ::2222
4637 vlan-tci 34 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4638 testpmd> flow create 0 ingress pattern end actions nvgre_encap /
4641 Sample L2 encapsulation rule
4642 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4644 L2 encapsulation has default value pre-configured in testpmd
4645 source code, those can be changed by using the following commands
4649 testpmd> set l2_encap ip-version ipv4
4650 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4651 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / mpls / end actions
4652 mplsoudp_decap / l2_encap / end
4654 L2 with VXLAN header::
4656 testpmd> set l2_encap-with-vlan ip-version ipv4 vlan-tci 34
4657 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4658 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / mpls / end actions
4659 mplsoudp_decap / l2_encap / end
4661 Sample L2 decapsulation rule
4662 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4664 L2 decapsulation has default value pre-configured in testpmd
4665 source code, those can be changed by using the following commands
4669 testpmd> set l2_decap
4670 testpmd> flow create 0 egress pattern eth / end actions l2_decap / mplsoudp_encap /
4673 L2 with VXLAN header::
4675 testpmd> set l2_encap-with-vlan
4676 testpmd> flow create 0 egress pattern eth / end actions l2_encap / mplsoudp_encap /
4679 Sample MPLSoGRE encapsulation rule
4680 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4682 MPLSoGRE encapsulation outer layer has default value pre-configured in testpmd
4683 source code, those can be changed by using the following commands
4685 IPv4 MPLSoGRE outer header::
4687 testpmd> set mplsogre_encap ip-version ipv4 label 4
4688 ip-src 127.0.0.1 ip-dst 128.0.0.1 eth-src 11:11:11:11:11:11
4689 eth-dst 22:22:22:22:22:22
4690 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4691 mplsogre_encap / end
4693 IPv4 MPLSoGRE with VLAN outer header::
4695 testpmd> set mplsogre_encap-with-vlan ip-version ipv4 label 4
4696 ip-src 127.0.0.1 ip-dst 128.0.0.1 vlan-tci 34
4697 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4698 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4699 mplsogre_encap / end
4701 IPv6 MPLSoGRE outer header::
4703 testpmd> set mplsogre_encap ip-version ipv6 mask 4
4704 ip-src ::1 ip-dst ::2222 eth-src 11:11:11:11:11:11
4705 eth-dst 22:22:22:22:22:22
4706 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4707 mplsogre_encap / end
4709 IPv6 MPLSoGRE with VLAN outer header::
4711 testpmd> set mplsogre_encap-with-vlan ip-version ipv6 mask 4
4712 ip-src ::1 ip-dst ::2222 vlan-tci 34
4713 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4714 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4715 mplsogre_encap / end
4717 Sample MPLSoGRE decapsulation rule
4718 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4720 MPLSoGRE decapsulation outer layer has default value pre-configured in testpmd
4721 source code, those can be changed by using the following commands
4723 IPv4 MPLSoGRE outer header::
4725 testpmd> set mplsogre_decap ip-version ipv4
4726 testpmd> flow create 0 ingress pattern eth / ipv4 / gre / mpls / end actions
4727 mplsogre_decap / l2_encap / end
4729 IPv4 MPLSoGRE with VLAN outer header::
4731 testpmd> set mplsogre_decap-with-vlan ip-version ipv4
4732 testpmd> flow create 0 ingress pattern eth / vlan / ipv4 / gre / mpls / end
4733 actions mplsogre_decap / l2_encap / end
4735 IPv6 MPLSoGRE outer header::
4737 testpmd> set mplsogre_decap ip-version ipv6
4738 testpmd> flow create 0 ingress pattern eth / ipv6 / gre / mpls / end
4739 actions mplsogre_decap / l2_encap / end
4741 IPv6 MPLSoGRE with VLAN outer header::
4743 testpmd> set mplsogre_decap-with-vlan ip-version ipv6
4744 testpmd> flow create 0 ingress pattern eth / vlan / ipv6 / gre / mpls / end
4745 actions mplsogre_decap / l2_encap / end
4747 Sample MPLSoUDP encapsulation rule
4748 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4750 MPLSoUDP encapsulation outer layer has default value pre-configured in testpmd
4751 source code, those can be changed by using the following commands
4753 IPv4 MPLSoUDP outer header::
4755 testpmd> set mplsoudp_encap ip-version ipv4 label 4 udp-src 5 udp-dst 10
4756 ip-src 127.0.0.1 ip-dst 128.0.0.1 eth-src 11:11:11:11:11:11
4757 eth-dst 22:22:22:22:22:22
4758 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4759 mplsoudp_encap / end
4761 IPv4 MPLSoUDP with VLAN outer header::
4763 testpmd> set mplsoudp_encap-with-vlan ip-version ipv4 label 4 udp-src 5
4764 udp-dst 10 ip-src 127.0.0.1 ip-dst 128.0.0.1 vlan-tci 34
4765 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4766 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4767 mplsoudp_encap / end
4769 IPv6 MPLSoUDP outer header::
4771 testpmd> set mplsoudp_encap ip-version ipv6 mask 4 udp-src 5 udp-dst 10
4772 ip-src ::1 ip-dst ::2222 eth-src 11:11:11:11:11:11
4773 eth-dst 22:22:22:22:22:22
4774 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4775 mplsoudp_encap / end
4777 IPv6 MPLSoUDP with VLAN outer header::
4779 testpmd> set mplsoudp_encap-with-vlan ip-version ipv6 mask 4 udp-src 5
4780 udp-dst 10 ip-src ::1 ip-dst ::2222 vlan-tci 34
4781 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4782 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4783 mplsoudp_encap / end
4785 Sample MPLSoUDP decapsulation rule
4786 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4788 MPLSoUDP decapsulation outer layer has default value pre-configured in testpmd
4789 source code, those can be changed by using the following commands
4791 IPv4 MPLSoUDP outer header::
4793 testpmd> set mplsoudp_decap ip-version ipv4
4794 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / mpls / end actions
4795 mplsoudp_decap / l2_encap / end
4797 IPv4 MPLSoUDP with VLAN outer header::
4799 testpmd> set mplsoudp_decap-with-vlan ip-version ipv4
4800 testpmd> flow create 0 ingress pattern eth / vlan / ipv4 / udp / mpls / end
4801 actions mplsoudp_decap / l2_encap / end
4803 IPv6 MPLSoUDP outer header::
4805 testpmd> set mplsoudp_decap ip-version ipv6
4806 testpmd> flow create 0 ingress pattern eth / ipv6 / udp / mpls / end
4807 actions mplsoudp_decap / l2_encap / end
4809 IPv6 MPLSoUDP with VLAN outer header::
4811 testpmd> set mplsoudp_decap-with-vlan ip-version ipv6
4812 testpmd> flow create 0 ingress pattern eth / vlan / ipv6 / udp / mpls / end
4813 actions mplsoudp_decap / l2_encap / end
4815 Sample Raw encapsulation rule
4816 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4818 Raw encapsulation configuration can be set by the following commands
4820 Encapsulating VxLAN::
4822 testpmd> set raw_encap 4 eth src is 10:11:22:33:44:55 / vlan tci is 1
4823 inner_type is 0x0800 / ipv4 / udp dst is 4789 / vxlan vni
4825 testpmd> flow create 0 egress pattern eth / ipv4 / end actions
4826 raw_encap index 4 / end
4828 Sample Raw decapsulation rule
4829 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4831 Raw decapsulation configuration can be set by the following commands
4833 Decapsulating VxLAN::
4835 testpmd> set raw_decap eth / ipv4 / udp / vxlan / end_set
4836 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / vxlan / eth / ipv4 /
4837 end actions raw_decap / queue index 0 / end
4842 ESP rules can be created by the following commands::
4844 testpmd> flow create 0 ingress pattern eth / ipv4 / esp spi is 1 / end actions
4846 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / esp spi is 1 / end
4847 actions queue index 3 / end
4848 testpmd> flow create 0 ingress pattern eth / ipv6 / esp spi is 1 / end actions
4850 testpmd> flow create 0 ingress pattern eth / ipv6 / udp / esp spi is 1 / end
4851 actions queue index 3 / end
4856 AH rules can be created by the following commands::
4858 testpmd> flow create 0 ingress pattern eth / ipv4 / ah spi is 1 / end actions
4860 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / ah spi is 1 / end
4861 actions queue index 3 / end
4862 testpmd> flow create 0 ingress pattern eth / ipv6 / ah spi is 1 / end actions
4864 testpmd> flow create 0 ingress pattern eth / ipv6 / udp / ah spi is 1 / end
4865 actions queue index 3 / end
4870 PFCP rules can be created by the following commands(s_field need to be 1
4873 testpmd> flow create 0 ingress pattern eth / ipv4 / pfcp s_field is 0 / end
4874 actions queue index 3 / end
4875 testpmd> flow create 0 ingress pattern eth / ipv4 / pfcp s_field is 1
4876 seid is 1 / end actions queue index 3 / end
4877 testpmd> flow create 0 ingress pattern eth / ipv6 / pfcp s_field is 0 / end
4878 actions queue index 3 / end
4879 testpmd> flow create 0 ingress pattern eth / ipv6 / pfcp s_field is 1
4880 seid is 1 / end actions queue index 3 / end
4882 Sample Sampling/Mirroring rules
4883 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4885 Sample/Mirroring rules can be set by the following commands
4887 NIC-RX Sampling rule, the matched ingress packets and sent to the queue 1,
4888 and 50% packets are duplicated and marked with 0x1234 and sent to queue 0.
4892 testpmd> set sample_actions 0 mark id 0x1234 / queue index 0 / end
4893 testpmd> flow create 0 ingress group 1 pattern eth / end actions
4894 sample ratio 2 index 0 / queue index 1 / end
4896 Mirroring rule with port representors (with "transfer" attribute), the matched
4897 ingress packets with encapsulation header are sent to port id 0, and also
4898 mirrored the packets and sent to port id 2.
4902 testpmd> set sample_actions 0 port_id id 2 / end
4903 testpmd> flow create 1 ingress transfer pattern eth / end actions
4904 sample ratio 1 index 0 / raw_encap / port_id id 0 / end
4906 Mirroring rule with port representors (with "transfer" attribute), the matched
4907 ingress packets are sent to port id 2, and also mirrored the packets with
4908 encapsulation header and sent to port id 0.
4912 testpmd> set sample_actions 0 raw_encap / port_id id 0 / end
4913 testpmd> flow create 0 ingress transfer pattern eth / end actions
4914 sample ratio 1 index 0 / port_id id 2 / end
4916 Mirroring rule with port representors (with "transfer" attribute), the matched
4917 ingress packets are sent to port id 2, and also mirrored the packets with
4918 VXLAN encapsulation header and sent to port id 0.
4922 testpmd> set vxlan ip-version ipv4 vni 4 udp-src 4 udp-dst 4 ip-src 127.0.0.1
4923 ip-dst 128.0.0.1 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4924 testpmd> set sample_actions 0 vxlan_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 NVGRE encapsulation header and sent to port id 0.
4934 testpmd> set nvgre ip-version ipv4 tni 4 ip-src 127.0.0.1 ip-dst 128.0.0.1
4935 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4936 testpmd> set sample_actions 0 nvgre_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 Sample integrity rules
4941 ~~~~~~~~~~~~~~~~~~~~~~
4943 Integrity rules can be created by the following commands:
4945 Integrity rule that forwards valid TCP packets to group 1.
4946 TCP packet integrity is matched with the ``l4_ok`` bit 3.
4950 testpmd> flow create 0 ingress
4951 pattern eth / ipv4 / tcp / integrity value mask 8 value spec 8 / end
4952 actions jump group 1 / end
4954 Integrity rule that forwards invalid packets to application.
4955 General packet integrity is matched with the ``packet_ok`` bit 0.
4959 testpmd> flow create 0 ingress pattern integrity value mask 1 value spec 0 / end actions queue index 0 / end
4961 Sample conntrack rules
4962 ~~~~~~~~~~~~~~~~~~~~~~
4964 Conntrack rules can be set by the following commands
4966 Need to construct the connection context with provided information.
4967 In the first table, create a flow rule by using conntrack action and jump to
4968 the next table. In the next table, create a rule to check the state.
4972 testpmd> set conntrack com peer 1 is_orig 1 enable 1 live 1 sack 1 cack 0
4973 last_dir 0 liberal 0 state 1 max_ack_win 7 r_lim 5 last_win 510
4974 last_seq 2632987379 last_ack 2532480967 last_end 2632987379
4976 testpmd> set conntrack orig scale 7 fin 0 acked 1 unack_data 0
4977 sent_end 2632987379 reply_end 2633016339 max_win 28960
4979 testpmd> set conntrack rply scale 7 fin 0 acked 1 unack_data 0
4980 sent_end 2532480967 reply_end 2532546247 max_win 65280
4982 testpmd> flow indirect_action 0 create ingress action conntrack / end
4983 testpmd> flow create 0 group 3 ingress pattern eth / ipv4 / tcp / end actions indirect 0 / jump group 5 / end
4984 testpmd> flow create 0 group 5 ingress pattern eth / ipv4 / tcp / conntrack is 1 / end actions queue index 5 / end
4986 Construct the conntrack again with only "is_orig" set to 0 (other fields are
4987 ignored), then use "update" interface to update the direction. Create flow
4988 rules like above for the peer port.
4992 testpmd> flow indirect_action 0 update 0 action conntrack_update dir / end
4994 Sample meter with policy rules
4995 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4997 Meter with policy rules can be created by the following commands:
4999 Need to create policy first and actions are set for green/yellow/red colors.
5000 Create meter with policy id. Create flow with meter id.
5002 Example for policy with meter color action. The purpose is to color the packet
5003 to reflect the meter color result.
5004 The meter policy action list: ``green -> green, yellow -> yellow, red -> red``.
5008 testpmd> add port meter profile srtcm_rfc2697 0 13 21504 2688 0 0
5009 testpmd> add port meter policy 0 1 g_actions color type green / end y_actions color type yellow / end
5010 r_actions color type red / end
5011 testpmd> create port meter 0 1 13 1 yes 0xffff 0 0
5012 testpmd> flow create 0 priority 0 ingress group 1 pattern eth / end actions meter mtr_id 1 / end
5017 The following sections show functions to load/unload eBPF based filters.
5022 Load an eBPF program as a callback for particular RX/TX queue::
5024 testpmd> bpf-load rx|tx (portid) (queueid) (load-flags) (bpf-prog-filename)
5026 The available load-flags are:
5028 * ``J``: use JIT generated native code, otherwise BPF interpreter will be used.
5030 * ``M``: assume input parameter is a pointer to rte_mbuf, otherwise assume it is a pointer to first segment's data.
5036 You'll need clang v3.7 or above to build bpf program you'd like to load
5040 .. code-block:: console
5043 clang -O2 -target bpf -c t1.c
5045 Then to load (and JIT compile) t1.o at RX queue 0, port 1:
5047 .. code-block:: console
5049 testpmd> bpf-load rx 1 0 J ./dpdk.org/examples/bpf/t1.o
5051 To load (not JITed) t1.o at TX queue 0, port 0:
5053 .. code-block:: console
5055 testpmd> bpf-load tx 0 0 - ./dpdk.org/examples/bpf/t1.o
5060 Unload previously loaded eBPF program for particular RX/TX queue::
5062 testpmd> bpf-unload rx|tx (portid) (queueid)
5064 For example to unload BPF filter from TX queue 0, port 0:
5066 .. code-block:: console
5068 testpmd> bpf-unload tx 0 0