1 .. SPDX-License-Identifier: BSD-3-Clause
2 Copyright(c) 2010-2016 Intel Corporation.
6 Testpmd Runtime Functions
7 =========================
9 Where the testpmd application is started in interactive mode, (``-i|--interactive``),
10 it displays a prompt that can be used to start and stop forwarding,
11 configure the application, display statistics (including the extended NIC
12 statistics aka xstats) , set the Flow Director and other tasks::
16 The testpmd prompt has some, limited, readline support.
17 Common bash command-line functions such as ``Ctrl+a`` and ``Ctrl+e`` to go to the start and end of the prompt line are supported
18 as well as access to the command history via the up-arrow.
20 There is also support for tab completion.
21 If you type a partial command and hit ``<TAB>`` you get a list of the available completions:
23 .. code-block:: console
25 testpmd> show port <TAB>
27 info [Mul-choice STRING]: show|clear port info|stats|xstats|fdir|stat_qmap|dcb_tc|cap X
28 info [Mul-choice STRING]: show|clear port info|stats|xstats|fdir|stat_qmap|dcb_tc|cap all
29 stats [Mul-choice STRING]: show|clear port info|stats|xstats|fdir|stat_qmap|dcb_tc|cap X
30 stats [Mul-choice STRING]: show|clear port info|stats|xstats|fdir|stat_qmap|dcb_tc|cap all
36 Some examples in this document are too long to fit on one line are shown wrapped at `"\\"` for display purposes::
38 testpmd> set flow_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \
39 (pause_time) (send_xon) (port_id)
41 In the real ``testpmd>`` prompt these commands should be on a single line.
46 The testpmd has on-line help for the functions that are available at runtime.
47 These are divided into sections and can be accessed using help, help section or help all:
49 .. code-block:: console
53 help control : Start and stop forwarding.
54 help display : Displaying port, stats and config information.
55 help config : Configuration information.
56 help ports : Configuring ports.
57 help registers : Reading and setting port registers.
58 help filters : Filters configuration help.
59 help all : All of the above sections.
62 Command File Functions
63 ----------------------
65 To facilitate loading large number of commands or to avoid cutting and pasting where not
66 practical or possible testpmd supports alternative methods for executing commands.
68 * If started with the ``--cmdline-file=FILENAME`` command line argument testpmd
69 will execute all CLI commands contained within the file immediately before
70 starting packet forwarding or entering interactive mode.
72 .. code-block:: console
74 ./testpmd -n4 -r2 ... -- -i --cmdline-file=/home/ubuntu/flow-create-commands.txt
75 Interactive-mode selected
76 CLI commands to be read from /home/ubuntu/flow-create-commands.txt
77 Configuring Port 0 (socket 0)
78 Port 0: 7C:FE:90:CB:74:CE
79 Configuring Port 1 (socket 0)
80 Port 1: 7C:FE:90:CB:74:CA
81 Checking link statuses...
82 Port 0 Link Up - speed 10000 Mbps - full-duplex
83 Port 1 Link Up - speed 10000 Mbps - full-duplex
89 Flow rule #498 created
90 Flow rule #499 created
91 Read all CLI commands from /home/ubuntu/flow-create-commands.txt
95 * At run-time additional commands can be loaded in bulk by invoking the ``load FILENAME``
98 .. code-block:: console
100 testpmd> load /home/ubuntu/flow-create-commands.txt
105 Flow rule #498 created
106 Flow rule #499 created
107 Read all CLI commands from /home/ubuntu/flow-create-commands.txt
111 In all cases output from any included command will be displayed as standard output.
112 Execution will continue until the end of the file is reached regardless of
113 whether any errors occur. The end user must examine the output to determine if
114 any failures occurred.
123 Start packet forwarding with current configuration::
130 Start packet forwarding with current configuration after sending specified number of bursts of packets::
132 testpmd> start tx_first (""|burst_num)
134 The default burst number is 1 when ``burst_num`` not presented.
139 Stop packet forwarding, and display accumulated statistics::
154 The functions in the following sections are used to display information about the
155 testpmd configuration or the NIC status.
160 Display information for a given port or all ports::
162 testpmd> show port (info|summary|stats|xstats|fdir|stat_qmap|dcb_tc|cap) (port_id|all)
164 The available information categories are:
166 * ``info``: General port information such as MAC address.
168 * ``summary``: Brief port summary such as Device Name, Driver Name etc.
170 * ``stats``: RX/TX statistics.
172 * ``xstats``: RX/TX extended NIC statistics.
174 * ``fdir``: Flow Director information and statistics.
176 * ``stat_qmap``: Queue statistics mapping.
178 * ``dcb_tc``: DCB information such as TC mapping.
180 * ``cap``: Supported offload capabilities.
184 .. code-block:: console
186 testpmd> show port info 0
188 ********************* Infos for port 0 *********************
190 MAC address: XX:XX:XX:XX:XX:XX
192 memory allocation on the socket: 0
194 Link speed: 40000 Mbps
195 Link duplex: full-duplex
196 Promiscuous mode: enabled
197 Allmulticast mode: disabled
198 Maximum number of MAC addresses: 64
199 Maximum number of MAC addresses of hash filtering: 0
201 strip on, filter on, extend off, qinq strip off
202 Redirection table size: 512
203 Supported flow types:
221 show port (module_eeprom|eeprom)
222 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
224 Display the EEPROM information of a port::
226 testpmd> show port (port_id) (module_eeprom|eeprom)
231 Display the rss redirection table entry indicated by masks on port X::
233 testpmd> show port (port_id) rss reta (size) (mask0, mask1...)
235 size is used to indicate the hardware supported reta size
240 Display the RSS hash functions and RSS hash key of a port::
242 testpmd> show port (port_id) rss-hash [key]
247 Clear the port statistics and forward engine statistics for a given port or for all ports::
249 testpmd> clear port (info|stats|xstats|fdir|stat_qmap) (port_id|all)
253 testpmd> clear port stats all
258 Display information for a given port's RX/TX queue::
260 testpmd> show (rxq|txq) info (port_id) (queue_id)
262 show desc status(rxq|txq)
263 ~~~~~~~~~~~~~~~~~~~~~~~~~
265 Display information for a given port's RX/TX descriptor status::
267 testpmd> show port (port_id) (rxq|txq) (queue_id) desc (desc_id) status
273 Displays the configuration of the application.
274 The configuration comes from the command-line, the runtime or the application defaults::
276 testpmd> show config (rxtx|cores|fwd|txpkts|txtimes)
278 The available information categories are:
280 * ``rxtx``: RX/TX configuration items.
282 * ``cores``: List of forwarding cores.
284 * ``fwd``: Packet forwarding configuration.
286 * ``txpkts``: Packets to TX configuration.
288 * ``txtimes``: Burst time pattern for Tx only mode.
292 .. code-block:: console
294 testpmd> show config rxtx
296 io packet forwarding - CRC stripping disabled - packets/burst=16
297 nb forwarding cores=2 - nb forwarding ports=1
298 RX queues=1 - RX desc=128 - RX free threshold=0
299 RX threshold registers: pthresh=8 hthresh=8 wthresh=4
300 TX queues=1 - TX desc=512 - TX free threshold=0
301 TX threshold registers: pthresh=36 hthresh=0 wthresh=0
302 TX RS bit threshold=0 - TXQ flags=0x0
307 Set the packet forwarding mode::
309 testpmd> set fwd (io|mac|macswap|flowgen| \
310 rxonly|txonly|csum|icmpecho|noisy|5tswap) (""|retry)
312 ``retry`` can be specified for forwarding engines except ``rx_only``.
314 The available information categories are:
316 * ``io``: Forwards packets "as-is" in I/O mode.
317 This is the fastest possible forwarding operation as it does not access packets data.
318 This is the default mode.
320 * ``mac``: Changes the source and the destination Ethernet addresses of packets before forwarding them.
321 Default application behavior is to set source Ethernet address to that of the transmitting interface, and destination
322 address to a dummy value (set during init). The user may specify a target destination Ethernet address via the 'eth-peer' or
323 'eth-peers-configfile' command-line options. It is not currently possible to specify a specific source Ethernet address.
325 * ``macswap``: MAC swap forwarding mode.
326 Swaps the source and the destination Ethernet addresses of packets before forwarding them.
328 * ``flowgen``: Multi-flow generation mode.
329 Originates a number of flows (with varying destination IP addresses), and terminate receive traffic.
331 * ``rxonly``: Receives packets but doesn't transmit them.
333 * ``txonly``: Generates and transmits packets without receiving any.
335 * ``csum``: Changes the checksum field with hardware or software methods depending on the offload flags on the packet.
337 * ``icmpecho``: Receives a burst of packets, lookup for ICMP echo requests and, if any, send back ICMP echo replies.
339 * ``ieee1588``: Demonstrate L2 IEEE1588 V2 PTP timestamping for RX and TX. Requires ``CONFIG_RTE_LIBRTE_IEEE1588=y``.
341 * ``noisy``: Noisy neighbor simulation.
342 Simulate more realistic behavior of a guest machine engaged in receiving
343 and sending packets performing Virtual Network Function (VNF).
345 * ``5tswap``: Swap the source and destination of L2,L3,L4 if they exist.
347 L2 swaps the source address and destination address of Ethernet, as same as ``macswap``.
349 L3 swaps the source address and destination address of IP (v4 and v6).
351 L4 swaps the source port and destination port of transport layer (TCP and UDP).
355 testpmd> set fwd rxonly
357 Set rxonly packet forwarding mode
363 When running, forwarding engines maintain statistics from the time they have been started.
364 Example for the io forwarding engine, with some packet drops on the tx side::
366 testpmd> show fwd stats all
368 ------- Forward Stats for RX Port= 0/Queue= 0 -> TX Port= 1/Queue= 0 -------
369 RX-packets: 274293770 TX-packets: 274293642 TX-dropped: 128
371 ------- Forward Stats for RX Port= 1/Queue= 0 -> TX Port= 0/Queue= 0 -------
372 RX-packets: 274301850 TX-packets: 274301850 TX-dropped: 0
374 ---------------------- Forward statistics for port 0 ----------------------
375 RX-packets: 274293802 RX-dropped: 0 RX-total: 274293802
376 TX-packets: 274301862 TX-dropped: 0 TX-total: 274301862
377 ----------------------------------------------------------------------------
379 ---------------------- Forward statistics for port 1 ----------------------
380 RX-packets: 274301894 RX-dropped: 0 RX-total: 274301894
381 TX-packets: 274293706 TX-dropped: 128 TX-total: 274293834
382 ----------------------------------------------------------------------------
384 +++++++++++++++ Accumulated forward statistics for all ports+++++++++++++++
385 RX-packets: 548595696 RX-dropped: 0 RX-total: 548595696
386 TX-packets: 548595568 TX-dropped: 128 TX-total: 548595696
387 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
391 Enabling CONFIG_RTE_TEST_PMD_RECORD_CORE_CYCLES appends "CPU cycles/packet" stats, like:
393 CPU cycles/packet=xx.dd (total cycles=xxxx / total RX packets=xxxx) at xxx MHz clock
398 Clear the forwarding engines statistics::
400 testpmd> clear fwd stats all
405 Display an RX descriptor for a port RX queue::
407 testpmd> read rxd (port_id) (queue_id) (rxd_id)
411 testpmd> read rxd 0 0 4
412 0x0000000B - 0x001D0180 / 0x0000000B - 0x001D0180
417 Display a TX descriptor for a port TX queue::
419 testpmd> read txd (port_id) (queue_id) (txd_id)
423 testpmd> read txd 0 0 4
424 0x00000001 - 0x24C3C440 / 0x000F0000 - 0x2330003C
429 Get loaded dynamic device personalization (DDP) package info list::
431 testpmd> ddp get list (port_id)
436 Display information about dynamic device personalization (DDP) profile::
438 testpmd> ddp get info (profile_path)
443 Display VF statistics::
445 testpmd> show vf stats (port_id) (vf_id)
450 Reset VF statistics::
452 testpmd> clear vf stats (port_id) (vf_id)
454 show port pctype mapping
455 ~~~~~~~~~~~~~~~~~~~~~~~~
457 List all items from the pctype mapping table::
459 testpmd> show port (port_id) pctype mapping
461 show rx offloading capabilities
462 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
464 List all per queue and per port Rx offloading capabilities of a port::
466 testpmd> show port (port_id) rx_offload capabilities
468 show rx offloading configuration
469 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
471 List port level and all queue level Rx offloading configuration::
473 testpmd> show port (port_id) rx_offload configuration
475 show tx offloading capabilities
476 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
478 List all per queue and per port Tx offloading capabilities of a port::
480 testpmd> show port (port_id) tx_offload capabilities
482 show tx offloading configuration
483 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
485 List port level and all queue level Tx offloading configuration::
487 testpmd> show port (port_id) tx_offload configuration
489 show tx metadata setting
490 ~~~~~~~~~~~~~~~~~~~~~~~~
492 Show Tx metadata value set for a specific port::
494 testpmd> show port (port_id) tx_metadata
496 show port supported ptypes
497 ~~~~~~~~~~~~~~~~~~~~~~~~~~
499 Show ptypes supported for a specific port::
501 testpmd> show port (port_id) ptypes
503 set port supported ptypes
504 ~~~~~~~~~~~~~~~~~~~~~~~~~
506 set packet types classification for a specific port::
508 testpmd> set port (port_id) ptypes_mask (mask)
510 show port mac addresses info
511 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
513 Show mac addresses added for a specific port::
515 testpmd> show port (port_id) macs
518 show port multicast mac addresses info
519 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
521 Show multicast mac addresses added for a specific port::
523 testpmd> show port (port_id) mcast_macs
528 Show general information about devices probed::
530 testpmd> show device info (<identifier>|all)
534 .. code-block:: console
536 testpmd> show device info net_pcap0
538 ********************* Infos for device net_pcap0 *********************
540 Driver name: net_pcap
541 Devargs: iface=enP2p6s0,phy_mac=1
542 Connect to socket: -1
545 MAC address: 1E:37:93:28:04:B8
546 Device name: net_pcap0
551 Dumps all physical memory segment layouts::
553 testpmd> dump_physmem
558 Dumps the layout of all memory zones::
560 testpmd> dump_memzone
565 Dumps the memory usage of all sockets::
567 testpmd> dump_socket_mem
572 Dumps the size of all memory structures::
574 testpmd> dump_struct_sizes
579 Dumps the status of all or specific element in DPDK rings::
581 testpmd> dump_ring [ring_name]
586 Dumps the statistics of all or specific memory pool::
588 testpmd> dump_mempool [mempool_name]
593 Dumps the user device list::
595 testpmd> dump_devargs
600 Dumps the log level for all the dpdk modules::
602 testpmd> dump_log_types
604 show (raw_encap|raw_decap)
605 ~~~~~~~~~~~~~~~~~~~~~~~~~~
607 Display content of raw_encap/raw_decap buffers in hex::
609 testpmd> show <raw_encap|raw_decap> <index>
610 testpmd> show <raw_encap|raw_decap> all
614 testpmd> show raw_encap 6
616 index: 6 at [0x1c565b0], len=50
617 00000000: 00 00 00 00 00 00 16 26 36 46 56 66 08 00 45 00 | .......&6FVf..E.
618 00000010: 00 00 00 00 00 00 00 11 00 00 C0 A8 01 06 C0 A8 | ................
619 00000020: 03 06 00 00 00 FA 00 00 00 00 08 00 00 00 00 00 | ................
622 show fec capabilities
623 ~~~~~~~~~~~~~~~~~~~~~
625 Show fec capabilities of a port::
627 testpmd> show port (port_id) fec capabilities
632 Show fec mode of a port::
634 testpmd> show port (port_id) fec_mode
637 Configuration Functions
638 -----------------------
640 The testpmd application can be configured from the runtime as well as from the command-line.
642 This section details the available configuration functions that are available.
646 Configuration changes only become active when forwarding is started/restarted.
651 Reset forwarding to the default configuration::
658 Set the debug verbosity level::
660 testpmd> set verbose (level)
662 Available levels are as following:
664 * ``0`` silent except for error.
665 * ``1`` fully verbose except for Tx packets.
666 * ``2`` fully verbose except for Rx packets.
667 * ``> 2`` fully verbose.
672 Set the log level for a log type::
674 testpmd> set log global|(type) (level)
678 * ``type`` is the log name.
680 * ``level`` is the log level.
682 For example, to change the global log level::
684 testpmd> set log global (level)
686 Regexes can also be used for type. To change log level of user1, user2 and user3::
688 testpmd> set log user[1-3] (level)
693 Set the number of ports used by the application:
697 This is equivalent to the ``--nb-ports`` command-line option.
702 Set the number of cores used by the application::
704 testpmd> set nbcore (num)
706 This is equivalent to the ``--nb-cores`` command-line option.
710 The number of cores used must not be greater than number of ports used multiplied by the number of queues per port.
715 Set the forwarding cores hexadecimal mask::
717 testpmd> set coremask (mask)
719 This is equivalent to the ``--coremask`` command-line option.
723 The master lcore is reserved for command line parsing only and cannot be masked on for packet forwarding.
728 Set the forwarding ports hexadecimal mask::
730 testpmd> set portmask (mask)
732 This is equivalent to the ``--portmask`` command-line option.
734 set record-core-cycles
735 ~~~~~~~~~~~~~~~~~~~~~~
737 Set the recording of CPU cycles::
739 testpmd> set record-core-cycles (on|off)
743 * ``on`` enables measurement of CPU cycles per packet.
745 * ``off`` disables measurement of CPU cycles per packet.
747 This is equivalent to the ``--record-core-cycles command-line`` option.
749 set record-burst-stats
750 ~~~~~~~~~~~~~~~~~~~~~~
752 Set the displaying of RX and TX bursts::
754 testpmd> set record-burst-stats (on|off)
758 * ``on`` enables display of RX and TX bursts.
760 * ``off`` disables display of RX and TX bursts.
762 This is equivalent to the ``--record-burst-stats command-line`` option.
767 Set number of packets per burst::
769 testpmd> set burst (num)
771 This is equivalent to the ``--burst command-line`` option.
773 When retry is enabled, the transmit delay time and number of retries can also be set::
775 testpmd> set burst tx delay (microseconds) retry (num)
780 Set the length of each segment of the TX-ONLY packets or length of packet for FLOWGEN mode::
782 testpmd> set txpkts (x[,y]*)
784 Where x[,y]* represents a CSV list of values, without white space.
789 Configure the timing burst pattern for Tx only mode. This command enables
790 the packet send scheduling on dynamic timestamp mbuf field and configures
791 timing pattern in Tx only mode. In this mode, if scheduling is enabled
792 application provides timestamps in the packets being sent. It is possible
793 to configure delay (in unspecified device clock units) between bursts
794 and between the packets within the burst::
796 testpmd> set txtimes (inter),(intra)
800 * ``inter`` is the delay between the bursts in the device clock units.
801 If ``intra`` is zero, this is the time between the beginnings of the
802 first packets in the neighbour bursts, if ``intra`` is not zero,
803 ``inter`` specifies the time between the beginning of the first packet
804 of the current burst and the beginning of the last packet of the
805 previous burst. If ``inter`` parameter is zero the send scheduling
806 on timestamps is disabled (default).
808 * ``intra`` is the delay between the packets within the burst specified
809 in the device clock units. The number of packets in the burst is defined
810 by regular burst setting. If ``intra`` parameter is zero no timestamps
811 provided in the packets excepting the first one in the burst.
813 As the result the bursts of packet will be transmitted with specific
814 delays between the packets within the burst and specific delay between
815 the bursts. The rte_eth_read_clock() must be supported by the device(s)
816 and is supposed to be engaged to get the current device clock value
817 and provide the reference for the timestamps. If there is no supported
818 rte_eth_read_clock() there will be no send scheduling provided on the port.
823 Set the split policy for the TX packets, applicable for TX-ONLY and CSUM forwarding modes::
825 testpmd> set txsplit (off|on|rand)
829 * ``off`` disable packet copy & split for CSUM mode.
831 * ``on`` split outgoing packet into multiple segments. Size of each segment
832 and number of segments per packet is determined by ``set txpkts`` command
835 * ``rand`` same as 'on', but number of segments per each packet is a random value between 1 and total number of segments.
840 Set the list of forwarding cores::
842 testpmd> set corelist (x[,y]*)
844 For example, to change the forwarding cores:
846 .. code-block:: console
848 testpmd> set corelist 3,1
849 testpmd> show config fwd
851 io packet forwarding - ports=2 - cores=2 - streams=2 - NUMA support disabled
852 Logical Core 3 (socket 0) forwards packets on 1 streams:
853 RX P=0/Q=0 (socket 0) -> TX P=1/Q=0 (socket 0) peer=02:00:00:00:00:01
854 Logical Core 1 (socket 0) forwards packets on 1 streams:
855 RX P=1/Q=0 (socket 0) -> TX P=0/Q=0 (socket 0) peer=02:00:00:00:00:00
859 The cores are used in the same order as specified on the command line.
864 Set the list of forwarding ports::
866 testpmd> set portlist (x[,y]*)
868 For example, to change the port forwarding:
870 .. code-block:: console
872 testpmd> set portlist 0,2,1,3
873 testpmd> show config fwd
875 io packet forwarding - ports=4 - cores=1 - streams=4
876 Logical Core 3 (socket 0) forwards packets on 4 streams:
877 RX P=0/Q=0 (socket 0) -> TX P=2/Q=0 (socket 0) peer=02:00:00:00:00:01
878 RX P=2/Q=0 (socket 0) -> TX P=0/Q=0 (socket 0) peer=02:00:00:00:00:00
879 RX P=1/Q=0 (socket 0) -> TX P=3/Q=0 (socket 0) peer=02:00:00:00:00:03
880 RX P=3/Q=0 (socket 0) -> TX P=1/Q=0 (socket 0) peer=02:00:00:00:00:02
885 Select how to retrieve new ports created after "port attach" command::
887 testpmd> set port setup on (iterator|event)
889 For each new port, a setup is done.
890 It will find the probed ports via RTE_ETH_FOREACH_MATCHING_DEV loop
891 in iterator mode, or via RTE_ETH_EVENT_NEW in event mode.
896 Enable/disable tx loopback::
898 testpmd> set tx loopback (port_id) (on|off)
903 set drop enable bit for all queues::
905 testpmd> set all queues drop (port_id) (on|off)
907 set split drop enable (for VF)
908 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
910 set split drop enable bit for VF from PF::
912 testpmd> set vf split drop (port_id) (vf_id) (on|off)
914 set mac antispoof (for VF)
915 ~~~~~~~~~~~~~~~~~~~~~~~~~~
917 Set mac antispoof for a VF from the PF::
919 testpmd> set vf mac antispoof (port_id) (vf_id) (on|off)
924 Enable/disable MACsec offload::
926 testpmd> set macsec offload (port_id) on encrypt (on|off) replay-protect (on|off)
927 testpmd> set macsec offload (port_id) off
932 Configure MACsec secure connection (SC)::
934 testpmd> set macsec sc (tx|rx) (port_id) (mac) (pi)
938 The pi argument is ignored for tx.
939 Check the NIC Datasheet for hardware limits.
944 Configure MACsec secure association (SA)::
946 testpmd> set macsec sa (tx|rx) (port_id) (idx) (an) (pn) (key)
950 The IDX value must be 0 or 1.
951 Check the NIC Datasheet for hardware limits.
953 set broadcast mode (for VF)
954 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
956 Set broadcast mode for a VF from the PF::
958 testpmd> set vf broadcast (port_id) (vf_id) (on|off)
963 Set the VLAN strip for a queue on a port::
965 testpmd> vlan set stripq (on|off) (port_id,queue_id)
967 vlan set stripq (for VF)
968 ~~~~~~~~~~~~~~~~~~~~~~~~
970 Set VLAN strip for all queues in a pool for a VF from the PF::
972 testpmd> set vf vlan stripq (port_id) (vf_id) (on|off)
974 vlan set insert (for VF)
975 ~~~~~~~~~~~~~~~~~~~~~~~~
977 Set VLAN insert for a VF from the PF::
979 testpmd> set vf vlan insert (port_id) (vf_id) (vlan_id)
981 vlan set tag (for VF)
982 ~~~~~~~~~~~~~~~~~~~~~
984 Set VLAN tag for a VF from the PF::
986 testpmd> set vf vlan tag (port_id) (vf_id) (on|off)
988 vlan set antispoof (for VF)
989 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
991 Set VLAN antispoof for a VF from the PF::
993 testpmd> set vf vlan antispoof (port_id) (vf_id) (on|off)
995 vlan set (strip|filter|qinq_strip|extend)
996 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
997 Set the VLAN strip/filter/QinQ strip/extend on for a port::
999 testpmd> vlan set (strip|filter|qinq_strip|extend) (on|off) (port_id)
1004 Set the inner or outer VLAN TPID for packet filtering on a port::
1006 testpmd> vlan set (inner|outer) tpid (value) (port_id)
1010 TPID value must be a 16-bit number (value <= 65536).
1015 Add a VLAN ID, or all identifiers, to the set of VLAN identifiers filtered by port ID::
1017 testpmd> rx_vlan add (vlan_id|all) (port_id)
1021 VLAN filter must be set on that port. VLAN ID < 4096.
1022 Depending on the NIC used, number of vlan_ids may be limited to the maximum entries
1023 in VFTA table. This is important if enabling all vlan_ids.
1028 Remove a VLAN ID, or all identifiers, from the set of VLAN identifiers filtered by port ID::
1030 testpmd> rx_vlan rm (vlan_id|all) (port_id)
1032 rx_vlan add (for VF)
1033 ~~~~~~~~~~~~~~~~~~~~
1035 Add a VLAN ID, to the set of VLAN identifiers filtered for VF(s) for port ID::
1037 testpmd> rx_vlan add (vlan_id) port (port_id) vf (vf_mask)
1042 Remove a VLAN ID, from the set of VLAN identifiers filtered for VF(s) for port ID::
1044 testpmd> rx_vlan rm (vlan_id) port (port_id) vf (vf_mask)
1049 Add a tunnel filter on a port::
1051 testpmd> tunnel_filter add (port_id) (outer_mac) (inner_mac) (ip_addr) \
1052 (inner_vlan) (vxlan|nvgre|ipingre|vxlan-gpe) (imac-ivlan|imac-ivlan-tenid|\
1053 imac-tenid|imac|omac-imac-tenid|oip|iip) (tenant_id) (queue_id)
1055 The available information categories are:
1057 * ``vxlan``: Set tunnel type as VXLAN.
1059 * ``nvgre``: Set tunnel type as NVGRE.
1061 * ``ipingre``: Set tunnel type as IP-in-GRE.
1063 * ``vxlan-gpe``: Set tunnel type as VXLAN-GPE
1065 * ``imac-ivlan``: Set filter type as Inner MAC and VLAN.
1067 * ``imac-ivlan-tenid``: Set filter type as Inner MAC, VLAN and tenant ID.
1069 * ``imac-tenid``: Set filter type as Inner MAC and tenant ID.
1071 * ``imac``: Set filter type as Inner MAC.
1073 * ``omac-imac-tenid``: Set filter type as Outer MAC, Inner MAC and tenant ID.
1075 * ``oip``: Set filter type as Outer IP.
1077 * ``iip``: Set filter type as Inner IP.
1081 testpmd> tunnel_filter add 0 68:05:CA:28:09:82 00:00:00:00:00:00 \
1082 192.168.2.2 0 ipingre oip 1 1
1084 Set an IP-in-GRE tunnel on port 0, and the filter type is Outer IP.
1086 tunnel_filter remove
1087 ~~~~~~~~~~~~~~~~~~~~
1089 Remove a tunnel filter on a port::
1091 testpmd> tunnel_filter rm (port_id) (outer_mac) (inner_mac) (ip_addr) \
1092 (inner_vlan) (vxlan|nvgre|ipingre|vxlan-gpe) (imac-ivlan|imac-ivlan-tenid|\
1093 imac-tenid|imac|omac-imac-tenid|oip|iip) (tenant_id) (queue_id)
1098 Add an UDP port for VXLAN packet filter on a port::
1100 testpmd> rx_vxlan_port add (udp_port) (port_id)
1102 rx_vxlan_port remove
1103 ~~~~~~~~~~~~~~~~~~~~
1105 Remove an UDP port for VXLAN packet filter on a port::
1107 testpmd> rx_vxlan_port rm (udp_port) (port_id)
1112 Set hardware insertion of VLAN IDs in packets sent on a port::
1114 testpmd> tx_vlan set (port_id) vlan_id[, vlan_id_outer]
1116 For example, set a single VLAN ID (5) insertion on port 0::
1120 Or, set double VLAN ID (inner: 2, outer: 3) insertion on port 1::
1128 Set port based hardware insertion of VLAN ID in packets sent on a port::
1130 testpmd> tx_vlan set pvid (port_id) (vlan_id) (on|off)
1135 Disable hardware insertion of a VLAN header in packets sent on a port::
1137 testpmd> tx_vlan reset (port_id)
1142 Select hardware or software calculation of the checksum when
1143 transmitting a packet using the ``csum`` forwarding engine::
1145 testpmd> csum set (ip|udp|tcp|sctp|outer-ip|outer-udp) (hw|sw) (port_id)
1149 * ``ip|udp|tcp|sctp`` always relate to the inner layer.
1151 * ``outer-ip`` relates to the outer IP layer (only for IPv4) in the case where the packet is recognized
1152 as a tunnel packet by the forwarding engine (geneve, gre, gtp, ipip, vxlan and vxlan-gpe are
1153 supported). See also the ``csum parse-tunnel`` command.
1155 * ``outer-udp`` relates to the outer UDP layer in the case where the packet is recognized
1156 as a tunnel packet by the forwarding engine (geneve, gtp, vxlan and vxlan-gpe are
1157 supported). See also the ``csum parse-tunnel`` command.
1161 Check the NIC Datasheet for hardware limits.
1166 Set RSS queue region span on a port::
1168 testpmd> set port (port_id) queue-region region_id (value) \
1169 queue_start_index (value) queue_num (value)
1171 Set flowtype mapping on a RSS queue region on a port::
1173 testpmd> set port (port_id) queue-region region_id (value) flowtype (value)
1177 * For the flowtype(pctype) of packet,the specific index for each type has
1178 been defined in file i40e_type.h as enum i40e_filter_pctype.
1180 Set user priority mapping on a RSS queue region on a port::
1182 testpmd> set port (port_id) queue-region UP (value) region_id (value)
1184 Flush all queue region related configuration on a port::
1186 testpmd> set port (port_id) queue-region flush (on|off)
1190 * ``on``: is just an enable function which server for other configuration,
1191 it is for all configuration about queue region from up layer,
1192 at first will only keep in DPDK software stored in driver,
1193 only after "flush on", it commit all configuration to HW.
1195 * ``"off``: is just clean all configuration about queue region just now,
1196 and restore all to DPDK i40e driver default config when start up.
1198 Show all queue region related configuration info on a port::
1200 testpmd> show port (port_id) queue-region
1204 Queue region only support on PF by now, so these command is
1205 only for configuration of queue region on PF port.
1210 Define how tunneled packets should be handled by the csum forward
1213 testpmd> csum parse-tunnel (on|off) (tx_port_id)
1215 If enabled, the csum forward engine will try to recognize supported
1216 tunnel headers (geneve, gtp, gre, ipip, vxlan, vxlan-gpe).
1218 If disabled, treat tunnel packets as non-tunneled packets (a inner
1219 header is handled as a packet payload).
1223 The port argument is the TX port like in the ``csum set`` command.
1227 Consider a packet in packet like the following::
1229 eth_out/ipv4_out/udp_out/vxlan/eth_in/ipv4_in/tcp_in
1231 * If parse-tunnel is enabled, the ``ip|udp|tcp|sctp`` parameters of ``csum set``
1232 command relate to the inner headers (here ``ipv4_in`` and ``tcp_in``), and the
1233 ``outer-ip|outer-udp`` parameter relates to the outer headers (here ``ipv4_out`` and ``udp_out``).
1235 * If parse-tunnel is disabled, the ``ip|udp|tcp|sctp`` parameters of ``csum set``
1236 command relate to the outer headers, here ``ipv4_out`` and ``udp_out``.
1241 Display tx checksum offload configuration::
1243 testpmd> csum show (port_id)
1248 Enable TCP Segmentation Offload (TSO) in the ``csum`` forwarding engine::
1250 testpmd> tso set (segsize) (port_id)
1254 Check the NIC datasheet for hardware limits.
1259 Display the status of TCP Segmentation Offload::
1261 testpmd> tso show (port_id)
1266 Set tso segment size of tunneled packets for a port in csum engine::
1268 testpmd> tunnel_tso set (tso_segsz) (port_id)
1273 Display the status of tunneled TCP Segmentation Offload for a port::
1275 testpmd> tunnel_tso show (port_id)
1280 Enable or disable GRO in ``csum`` forwarding engine::
1282 testpmd> set port <port_id> gro on|off
1284 If enabled, the csum forwarding engine will perform GRO on the TCP/IPv4
1285 packets received from the given port.
1287 If disabled, packets received from the given port won't be performed
1288 GRO. By default, GRO is disabled for all ports.
1292 When enable GRO for a port, TCP/IPv4 packets received from the port
1293 will be performed GRO. After GRO, all merged packets have bad
1294 checksums, since the GRO library doesn't re-calculate checksums for
1295 the merged packets. Therefore, if users want the merged packets to
1296 have correct checksums, please select HW IP checksum calculation and
1297 HW TCP checksum calculation for the port which the merged packets are
1303 Display GRO configuration for a given port::
1305 testpmd> show port <port_id> gro
1310 Set the cycle to flush the GROed packets from reassembly tables::
1312 testpmd> set gro flush <cycles>
1314 When enable GRO, the csum forwarding engine performs GRO on received
1315 packets, and the GROed packets are stored in reassembly tables. Users
1316 can use this command to determine when the GROed packets are flushed
1317 from the reassembly tables.
1319 The ``cycles`` is measured in GRO operation times. The csum forwarding
1320 engine flushes the GROed packets from the tables every ``cycles`` GRO
1323 By default, the value of ``cycles`` is 1, which means flush GROed packets
1324 from the reassembly tables as soon as one GRO operation finishes. The value
1325 of ``cycles`` should be in the range of 1 to ``GRO_MAX_FLUSH_CYCLES``.
1327 Please note that the large value of ``cycles`` may cause the poor TCP/IP
1328 stack performance. Because the GROed packets are delayed to arrive the
1329 stack, thus causing more duplicated ACKs and TCP retransmissions.
1334 Toggle per-port GSO support in ``csum`` forwarding engine::
1336 testpmd> set port <port_id> gso on|off
1338 If enabled, the csum forwarding engine will perform GSO on supported IPv4
1339 packets, transmitted on the given port.
1341 If disabled, packets transmitted on the given port will not undergo GSO.
1342 By default, GSO is disabled for all ports.
1346 When GSO is enabled on a port, supported IPv4 packets transmitted on that
1347 port undergo GSO. Afterwards, the segmented packets are represented by
1348 multi-segment mbufs; however, the csum forwarding engine doesn't calculation
1349 of checksums for GSO'd segments in SW. As a result, if users want correct
1350 checksums in GSO segments, they should enable HW checksum calculation for
1353 For example, HW checksum calculation for VxLAN GSO'd packets may be enabled
1354 by setting the following options in the csum forwarding engine:
1356 testpmd> csum set outer_ip hw <port_id>
1358 testpmd> csum set ip hw <port_id>
1360 testpmd> csum set tcp hw <port_id>
1362 UDP GSO is the same as IP fragmentation, which treats the UDP header
1363 as the payload and does not modify it during segmentation. That is,
1364 after UDP GSO, only the first output fragment has the original UDP
1365 header. Therefore, users need to enable HW IP checksum calculation
1366 and SW UDP checksum calculation for GSO-enabled ports, if they want
1367 correct checksums for UDP/IPv4 packets.
1372 Set the maximum GSO segment size (measured in bytes), which includes the
1373 packet header and the packet payload for GSO-enabled ports (global)::
1375 testpmd> set gso segsz <length>
1380 Display the status of Generic Segmentation Offload for a given port::
1382 testpmd> show port <port_id> gso
1387 Add an alternative MAC address to a port::
1389 testpmd> mac_addr add (port_id) (XX:XX:XX:XX:XX:XX)
1394 Remove a MAC address from a port::
1396 testpmd> mac_addr remove (port_id) (XX:XX:XX:XX:XX:XX)
1401 To add the multicast MAC address to/from the set of multicast addresses
1404 testpmd> mcast_addr add (port_id) (mcast_addr)
1409 To remove the multicast MAC address to/from the set of multicast addresses
1412 testpmd> mcast_addr remove (port_id) (mcast_addr)
1414 mac_addr add (for VF)
1415 ~~~~~~~~~~~~~~~~~~~~~
1417 Add an alternative MAC address for a VF to a port::
1419 testpmd> mac_add add port (port_id) vf (vf_id) (XX:XX:XX:XX:XX:XX)
1424 Set the default MAC address for a port::
1426 testpmd> mac_addr set (port_id) (XX:XX:XX:XX:XX:XX)
1428 mac_addr set (for VF)
1429 ~~~~~~~~~~~~~~~~~~~~~
1431 Set the MAC address for a VF from the PF::
1433 testpmd> set vf mac addr (port_id) (vf_id) (XX:XX:XX:XX:XX:XX)
1438 Set the forwarding peer address for certain port::
1440 testpmd> set eth-peer (port_id) (peer_addr)
1442 This is equivalent to the ``--eth-peer`` command-line option.
1447 Set the unicast hash filter(s) on/off for a port::
1449 testpmd> set port (port_id) uta (XX:XX:XX:XX:XX:XX|all) (on|off)
1454 Set the promiscuous mode on for a port or for all ports.
1455 In promiscuous mode packets are not dropped if they aren't for the specified MAC address::
1457 testpmd> set promisc (port_id|all) (on|off)
1462 Set the allmulti mode for a port or for all ports::
1464 testpmd> set allmulti (port_id|all) (on|off)
1466 Same as the ifconfig (8) option. Controls how multicast packets are handled.
1468 set promisc (for VF)
1469 ~~~~~~~~~~~~~~~~~~~~
1471 Set the unicast promiscuous mode for a VF from PF.
1472 It's supported by Intel i40e NICs now.
1473 In promiscuous mode packets are not dropped if they aren't for the specified MAC address::
1475 testpmd> set vf promisc (port_id) (vf_id) (on|off)
1477 set allmulticast (for VF)
1478 ~~~~~~~~~~~~~~~~~~~~~~~~~
1480 Set the multicast promiscuous mode for a VF from PF.
1481 It's supported by Intel i40e NICs now.
1482 In promiscuous mode packets are not dropped if they aren't for the specified MAC address::
1484 testpmd> set vf allmulti (port_id) (vf_id) (on|off)
1486 set tx max bandwidth (for VF)
1487 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1489 Set TX max absolute bandwidth (Mbps) for a VF from PF::
1491 testpmd> set vf tx max-bandwidth (port_id) (vf_id) (max_bandwidth)
1493 set tc tx min bandwidth (for VF)
1494 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1496 Set all TCs' TX min relative bandwidth (%) for a VF from PF::
1498 testpmd> set vf tc tx min-bandwidth (port_id) (vf_id) (bw1, bw2, ...)
1500 set tc tx max bandwidth (for VF)
1501 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1503 Set a TC's TX max absolute bandwidth (Mbps) for a VF from PF::
1505 testpmd> set vf tc tx max-bandwidth (port_id) (vf_id) (tc_no) (max_bandwidth)
1507 set tc strict link priority mode
1508 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1510 Set some TCs' strict link priority mode on a physical port::
1512 testpmd> set tx strict-link-priority (port_id) (tc_bitmap)
1514 set tc tx min bandwidth
1515 ~~~~~~~~~~~~~~~~~~~~~~~
1517 Set all TCs' TX min relative bandwidth (%) globally for all PF and VFs::
1519 testpmd> set tc tx min-bandwidth (port_id) (bw1, bw2, ...)
1524 Set the link flow control parameter on a port::
1526 testpmd> set flow_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \
1527 (pause_time) (send_xon) mac_ctrl_frame_fwd (on|off) \
1528 autoneg (on|off) (port_id)
1532 * ``high_water`` (integer): High threshold value to trigger XOFF.
1534 * ``low_water`` (integer): Low threshold value to trigger XON.
1536 * ``pause_time`` (integer): Pause quota in the Pause frame.
1538 * ``send_xon`` (0/1): Send XON frame.
1540 * ``mac_ctrl_frame_fwd``: Enable receiving MAC control frames.
1542 * ``autoneg``: Change the auto-negotiation parameter.
1547 Set the priority flow control parameter on a port::
1549 testpmd> set pfc_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \
1550 (pause_time) (priority) (port_id)
1554 * ``high_water`` (integer): High threshold value.
1556 * ``low_water`` (integer): Low threshold value.
1558 * ``pause_time`` (integer): Pause quota in the Pause frame.
1560 * ``priority`` (0-7): VLAN User Priority.
1565 Set statistics mapping (qmapping 0..15) for RX/TX queue on port::
1567 testpmd> set stat_qmap (tx|rx) (port_id) (queue_id) (qmapping)
1569 For example, to set rx queue 2 on port 0 to mapping 5::
1571 testpmd>set stat_qmap rx 0 2 5
1573 set xstats-hide-zero
1574 ~~~~~~~~~~~~~~~~~~~~
1576 Set the option to hide zero values for xstats display::
1578 testpmd> set xstats-hide-zero on|off
1582 By default, the zero values are displayed for xstats.
1584 set port - rx/tx (for VF)
1585 ~~~~~~~~~~~~~~~~~~~~~~~~~
1587 Set VF receive/transmit from a port::
1589 testpmd> set port (port_id) vf (vf_id) (rx|tx) (on|off)
1591 set port - mac address filter (for VF)
1592 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1594 Add/Remove unicast or multicast MAC addr filter for a VF::
1596 testpmd> set port (port_id) vf (vf_id) (mac_addr) \
1597 (exact-mac|exact-mac-vlan|hashmac|hashmac-vlan) (on|off)
1599 set port - rx mode(for VF)
1600 ~~~~~~~~~~~~~~~~~~~~~~~~~~
1602 Set the VF receive mode of a port::
1604 testpmd> set port (port_id) vf (vf_id) \
1605 rxmode (AUPE|ROPE|BAM|MPE) (on|off)
1607 The available receive modes are:
1609 * ``AUPE``: Accepts untagged VLAN.
1611 * ``ROPE``: Accepts unicast hash.
1613 * ``BAM``: Accepts broadcast packets.
1615 * ``MPE``: Accepts all multicast packets.
1617 set port - tx_rate (for Queue)
1618 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1620 Set TX rate limitation for a queue on a port::
1622 testpmd> set port (port_id) queue (queue_id) rate (rate_value)
1624 set port - tx_rate (for VF)
1625 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
1627 Set TX rate limitation for queues in VF on a port::
1629 testpmd> set port (port_id) vf (vf_id) rate (rate_value) queue_mask (queue_mask)
1631 set port - mirror rule
1632 ~~~~~~~~~~~~~~~~~~~~~~
1634 Set pool or vlan type mirror rule for a port::
1636 testpmd> set port (port_id) mirror-rule (rule_id) \
1637 (pool-mirror-up|pool-mirror-down|vlan-mirror) \
1638 (poolmask|vlanid[,vlanid]*) dst-pool (pool_id) (on|off)
1640 Set link mirror rule for a port::
1642 testpmd> set port (port_id) mirror-rule (rule_id) \
1643 (uplink-mirror|downlink-mirror) dst-pool (pool_id) (on|off)
1645 For example to enable mirror traffic with vlan 0,1 to pool 0::
1647 set port 0 mirror-rule 0 vlan-mirror 0,1 dst-pool 0 on
1649 reset port - mirror rule
1650 ~~~~~~~~~~~~~~~~~~~~~~~~
1652 Reset a mirror rule for a port::
1654 testpmd> reset port (port_id) mirror-rule (rule_id)
1659 Set the flush on RX streams before forwarding.
1660 The default is flush ``on``.
1661 Mainly used with PCAP drivers to turn off the default behavior of flushing the first 512 packets on RX streams::
1663 testpmd> set flush_rx off
1668 Set the bypass mode for the lowest port on bypass enabled NIC::
1670 testpmd> set bypass mode (normal|bypass|isolate) (port_id)
1675 Set the event required to initiate specified bypass mode for the lowest port on a bypass enabled::
1677 testpmd> set bypass event (timeout|os_on|os_off|power_on|power_off) \
1678 mode (normal|bypass|isolate) (port_id)
1682 * ``timeout``: Enable bypass after watchdog timeout.
1684 * ``os_on``: Enable bypass when OS/board is powered on.
1686 * ``os_off``: Enable bypass when OS/board is powered off.
1688 * ``power_on``: Enable bypass when power supply is turned on.
1690 * ``power_off``: Enable bypass when power supply is turned off.
1696 Set the bypass watchdog timeout to ``n`` seconds where 0 = instant::
1698 testpmd> set bypass timeout (0|1.5|2|3|4|8|16|32)
1703 Show the bypass configuration for a bypass enabled NIC using the lowest port on the NIC::
1705 testpmd> show bypass config (port_id)
1710 Set link up for a port::
1712 testpmd> set link-up port (port id)
1717 Set link down for a port::
1719 testpmd> set link-down port (port id)
1724 Enable E-tag insertion for a VF on a port::
1726 testpmd> E-tag set insertion on port-tag-id (value) port (port_id) vf (vf_id)
1728 Disable E-tag insertion for a VF on a port::
1730 testpmd> E-tag set insertion off port (port_id) vf (vf_id)
1732 Enable/disable E-tag stripping on a port::
1734 testpmd> E-tag set stripping (on|off) port (port_id)
1736 Enable/disable E-tag based forwarding on a port::
1738 testpmd> E-tag set forwarding (on|off) port (port_id)
1740 Add an E-tag forwarding filter on a port::
1742 testpmd> E-tag set filter add e-tag-id (value) dst-pool (pool_id) port (port_id)
1744 Delete an E-tag forwarding filter on a port::
1745 testpmd> E-tag set filter del e-tag-id (value) port (port_id)
1750 Load a dynamic device personalization (DDP) profile and store backup profile::
1752 testpmd> ddp add (port_id) (profile_path[,backup_profile_path])
1757 Delete a dynamic device personalization profile and restore backup profile::
1759 testpmd> ddp del (port_id) (backup_profile_path)
1764 List all items from the ptype mapping table::
1766 testpmd> ptype mapping get (port_id) (valid_only)
1770 * ``valid_only``: A flag indicates if only list valid items(=1) or all itemss(=0).
1772 Replace a specific or a group of software defined ptype with a new one::
1774 testpmd> ptype mapping replace (port_id) (target) (mask) (pkt_type)
1778 * ``target``: A specific software ptype or a mask to represent a group of software ptypes.
1780 * ``mask``: A flag indicate if "target" is a specific software ptype(=0) or a ptype mask(=1).
1782 * ``pkt_type``: The new software ptype to replace the old ones.
1784 Update hardware defined ptype to software defined packet type mapping table::
1786 testpmd> ptype mapping update (port_id) (hw_ptype) (sw_ptype)
1790 * ``hw_ptype``: hardware ptype as the index of the ptype mapping table.
1792 * ``sw_ptype``: software ptype as the value of the ptype mapping table.
1794 Reset ptype mapping table::
1796 testpmd> ptype mapping reset (port_id)
1798 config per port Rx offloading
1799 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1801 Enable or disable a per port Rx offloading on all Rx queues of a port::
1803 testpmd> port config (port_id) rx_offload (offloading) on|off
1805 * ``offloading``: can be any of these offloading capability:
1806 vlan_strip, ipv4_cksum, udp_cksum, tcp_cksum, tcp_lro,
1807 qinq_strip, outer_ipv4_cksum, macsec_strip,
1808 header_split, vlan_filter, vlan_extend, jumbo_frame,
1809 scatter, timestamp, security, keep_crc, rss_hash
1811 This command should be run when the port is stopped, or else it will fail.
1813 config per queue Rx offloading
1814 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1816 Enable or disable a per queue Rx offloading only on a specific Rx queue::
1818 testpmd> port (port_id) rxq (queue_id) rx_offload (offloading) on|off
1820 * ``offloading``: can be any of these offloading capability:
1821 vlan_strip, ipv4_cksum, udp_cksum, tcp_cksum, tcp_lro,
1822 qinq_strip, outer_ipv4_cksum, macsec_strip,
1823 header_split, vlan_filter, vlan_extend, jumbo_frame,
1824 scatter, timestamp, security, keep_crc
1826 This command should be run when the port is stopped, or else it will fail.
1828 config per port Tx offloading
1829 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1831 Enable or disable a per port Tx offloading on all Tx queues of a port::
1833 testpmd> port config (port_id) tx_offload (offloading) on|off
1835 * ``offloading``: can be any of these offloading capability:
1836 vlan_insert, ipv4_cksum, udp_cksum, tcp_cksum,
1837 sctp_cksum, tcp_tso, udp_tso, outer_ipv4_cksum,
1838 qinq_insert, vxlan_tnl_tso, gre_tnl_tso,
1839 ipip_tnl_tso, geneve_tnl_tso, macsec_insert,
1840 mt_lockfree, multi_segs, mbuf_fast_free, security
1842 This command should be run when the port is stopped, or else it will fail.
1844 config per queue Tx offloading
1845 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1847 Enable or disable a per queue Tx offloading only on a specific Tx queue::
1849 testpmd> port (port_id) txq (queue_id) tx_offload (offloading) on|off
1851 * ``offloading``: can be any of these offloading capability:
1852 vlan_insert, ipv4_cksum, udp_cksum, tcp_cksum,
1853 sctp_cksum, tcp_tso, udp_tso, outer_ipv4_cksum,
1854 qinq_insert, vxlan_tnl_tso, gre_tnl_tso,
1855 ipip_tnl_tso, geneve_tnl_tso, macsec_insert,
1856 mt_lockfree, multi_segs, mbuf_fast_free, security
1858 This command should be run when the port is stopped, or else it will fail.
1860 Config VXLAN Encap outer layers
1861 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1863 Configure the outer layer to encapsulate a packet inside a VXLAN tunnel::
1865 set vxlan ip-version (ipv4|ipv6) vni (vni) udp-src (udp-src) \
1866 udp-dst (udp-dst) ip-src (ip-src) ip-dst (ip-dst) eth-src (eth-src) \
1869 set vxlan-with-vlan ip-version (ipv4|ipv6) vni (vni) udp-src (udp-src) \
1870 udp-dst (udp-dst) ip-src (ip-src) ip-dst (ip-dst) vlan-tci (vlan-tci) \
1871 eth-src (eth-src) eth-dst (eth-dst)
1873 set vxlan-tos-ttl ip-version (ipv4|ipv6) vni (vni) udp-src (udp-src) \
1874 udp-dst (udp-dst) ip-tos (ip-tos) ip-ttl (ip-ttl) ip-src (ip-src) \
1875 ip-dst (ip-dst) eth-src (eth-src) eth-dst (eth-dst)
1877 These commands will set an internal configuration inside testpmd, any following
1878 flow rule using the action vxlan_encap will use the last configuration set.
1879 To have a different encapsulation header, one of those commands must be called
1880 before the flow rule creation.
1882 Config NVGRE Encap outer layers
1883 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1885 Configure the outer layer to encapsulate a packet inside a NVGRE tunnel::
1887 set nvgre ip-version (ipv4|ipv6) tni (tni) ip-src (ip-src) ip-dst (ip-dst) \
1888 eth-src (eth-src) eth-dst (eth-dst)
1889 set nvgre-with-vlan ip-version (ipv4|ipv6) tni (tni) ip-src (ip-src) \
1890 ip-dst (ip-dst) vlan-tci (vlan-tci) eth-src (eth-src) eth-dst (eth-dst)
1892 These commands will set an internal configuration inside testpmd, any following
1893 flow rule using the action nvgre_encap will use the last configuration set.
1894 To have a different encapsulation header, one of those commands must be called
1895 before the flow rule creation.
1900 Configure the l2 to be used when encapsulating a packet with L2::
1902 set l2_encap ip-version (ipv4|ipv6) eth-src (eth-src) eth-dst (eth-dst)
1903 set l2_encap-with-vlan ip-version (ipv4|ipv6) vlan-tci (vlan-tci) \
1904 eth-src (eth-src) eth-dst (eth-dst)
1906 Those commands will set an internal configuration inside testpmd, any following
1907 flow rule using the action l2_encap will use the last configuration set.
1908 To have a different encapsulation header, one of those commands must be called
1909 before the flow rule creation.
1914 Configure the l2 to be removed when decapsulating a packet with L2::
1916 set l2_decap ip-version (ipv4|ipv6)
1917 set l2_decap-with-vlan ip-version (ipv4|ipv6)
1919 Those commands will set an internal configuration inside testpmd, any following
1920 flow rule using the action l2_decap will use the last configuration set.
1921 To have a different encapsulation header, one of those commands must be called
1922 before the flow rule creation.
1924 Config MPLSoGRE Encap outer layers
1925 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1927 Configure the outer layer to encapsulate a packet inside a MPLSoGRE tunnel::
1929 set mplsogre_encap ip-version (ipv4|ipv6) label (label) \
1930 ip-src (ip-src) ip-dst (ip-dst) eth-src (eth-src) eth-dst (eth-dst)
1931 set mplsogre_encap-with-vlan ip-version (ipv4|ipv6) label (label) \
1932 ip-src (ip-src) ip-dst (ip-dst) vlan-tci (vlan-tci) \
1933 eth-src (eth-src) eth-dst (eth-dst)
1935 These commands will set an internal configuration inside testpmd, any following
1936 flow rule using the action mplsogre_encap will use the last configuration set.
1937 To have a different encapsulation header, one of those commands must be called
1938 before the flow rule creation.
1940 Config MPLSoGRE Decap outer layers
1941 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1943 Configure the outer layer to decapsulate MPLSoGRE packet::
1945 set mplsogre_decap ip-version (ipv4|ipv6)
1946 set mplsogre_decap-with-vlan ip-version (ipv4|ipv6)
1948 These commands will set an internal configuration inside testpmd, any following
1949 flow rule using the action mplsogre_decap will use the last configuration set.
1950 To have a different decapsulation header, one of those commands must be called
1951 before the flow rule creation.
1953 Config MPLSoUDP Encap outer layers
1954 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1956 Configure the outer layer to encapsulate a packet inside a MPLSoUDP tunnel::
1958 set mplsoudp_encap ip-version (ipv4|ipv6) label (label) udp-src (udp-src) \
1959 udp-dst (udp-dst) ip-src (ip-src) ip-dst (ip-dst) \
1960 eth-src (eth-src) eth-dst (eth-dst)
1961 set mplsoudp_encap-with-vlan ip-version (ipv4|ipv6) label (label) \
1962 udp-src (udp-src) udp-dst (udp-dst) ip-src (ip-src) ip-dst (ip-dst) \
1963 vlan-tci (vlan-tci) eth-src (eth-src) eth-dst (eth-dst)
1965 These commands will set an internal configuration inside testpmd, any following
1966 flow rule using the action mplsoudp_encap will use the last configuration set.
1967 To have a different encapsulation header, one of those commands must be called
1968 before the flow rule creation.
1970 Config MPLSoUDP Decap outer layers
1971 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1973 Configure the outer layer to decapsulate MPLSoUDP packet::
1975 set mplsoudp_decap ip-version (ipv4|ipv6)
1976 set mplsoudp_decap-with-vlan ip-version (ipv4|ipv6)
1978 These commands will set an internal configuration inside testpmd, any following
1979 flow rule using the action mplsoudp_decap will use the last configuration set.
1980 To have a different decapsulation header, one of those commands must be called
1981 before the flow rule creation.
1983 Config Raw Encapsulation
1984 ~~~~~~~~~~~~~~~~~~~~~~~~~
1986 Configure the raw data to be used when encapsulating a packet by
1987 rte_flow_action_raw_encap::
1989 set raw_encap {index} {item} [/ {item} [...]] / end_set
1991 There are multiple global buffers for ``raw_encap``, this command will set one
1992 internal buffer index by ``{index}``.
1993 If there is no ``{index}`` specified::
1995 set raw_encap {item} [/ {item} [...]] / end_set
1997 the default index ``0`` is used.
1998 In order to use different encapsulating header, ``index`` must be specified
1999 during the flow rule creation::
2001 testpmd> flow create 0 egress pattern eth / ipv4 / end actions
2002 raw_encap index 2 / end
2004 Otherwise the default index ``0`` is used.
2006 Config Raw Decapsulation
2007 ~~~~~~~~~~~~~~~~~~~~~~~~
2009 Configure the raw data to be used when decapsulating a packet by
2010 rte_flow_action_raw_decap::
2012 set raw_decap {index} {item} [/ {item} [...]] / end_set
2014 There are multiple global buffers for ``raw_decap``, this command will set
2015 one internal buffer index by ``{index}``.
2016 If there is no ``{index}`` specified::
2018 set raw_decap {item} [/ {item} [...]] / end_set
2020 the default index ``0`` is used.
2021 In order to use different decapsulating header, ``index`` must be specified
2022 during the flow rule creation::
2024 testpmd> flow create 0 egress pattern eth / ipv4 / end actions
2025 raw_encap index 3 / end
2027 Otherwise the default index ``0`` is used.
2032 Set fec mode for a specific port::
2034 testpmd> set port (port_id) fec_mode auto|off|rs|baser
2040 The following sections show functions for configuring ports.
2044 Port configuration changes only become active when forwarding is started/restarted.
2049 Attach a port specified by pci address or virtual device args::
2051 testpmd> port attach (identifier)
2053 To attach a new pci device, the device should be recognized by kernel first.
2054 Then it should be moved under DPDK management.
2055 Finally the port can be attached to testpmd.
2057 For example, to move a pci device using ixgbe under DPDK management:
2059 .. code-block:: console
2061 # Check the status of the available devices.
2062 ./usertools/dpdk-devbind.py --status
2064 Network devices using DPDK-compatible driver
2065 ============================================
2068 Network devices using kernel driver
2069 ===================================
2070 0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused=
2073 # Bind the device to igb_uio.
2074 sudo ./usertools/dpdk-devbind.py -b igb_uio 0000:0a:00.0
2077 # Recheck the status of the devices.
2078 ./usertools/dpdk-devbind.py --status
2079 Network devices using DPDK-compatible driver
2080 ============================================
2081 0000:0a:00.0 '82599ES 10-Gigabit' drv=igb_uio unused=
2083 To attach a port created by virtual device, above steps are not needed.
2085 For example, to attach a port whose pci address is 0000:0a:00.0.
2087 .. code-block:: console
2089 testpmd> port attach 0000:0a:00.0
2090 Attaching a new port...
2091 EAL: PCI device 0000:0a:00.0 on NUMA socket -1
2092 EAL: probe driver: 8086:10fb rte_ixgbe_pmd
2093 EAL: PCI memory mapped at 0x7f83bfa00000
2094 EAL: PCI memory mapped at 0x7f83bfa80000
2095 PMD: eth_ixgbe_dev_init(): MAC: 2, PHY: 18, SFP+: 5
2096 PMD: eth_ixgbe_dev_init(): port 0 vendorID=0x8086 deviceID=0x10fb
2097 Port 0 is attached. Now total ports is 1
2100 For example, to attach a port created by pcap PMD.
2102 .. code-block:: console
2104 testpmd> port attach net_pcap0
2105 Attaching a new port...
2106 PMD: Initializing pmd_pcap for net_pcap0
2107 PMD: Creating pcap-backed ethdev on numa socket 0
2108 Port 0 is attached. Now total ports is 1
2111 In this case, identifier is ``net_pcap0``.
2112 This identifier format is the same as ``--vdev`` format of DPDK applications.
2114 For example, to re-attach a bonded port which has been previously detached,
2115 the mode and slave parameters must be given.
2117 .. code-block:: console
2119 testpmd> port attach net_bond_0,mode=0,slave=1
2120 Attaching a new port...
2121 EAL: Initializing pmd_bond for net_bond_0
2122 EAL: Create bonded device net_bond_0 on port 0 in mode 0 on socket 0.
2123 Port 0 is attached. Now total ports is 1
2130 Detach a specific port::
2132 testpmd> port detach (port_id)
2134 Before detaching a port, the port should be stopped and closed.
2136 For example, to detach a pci device port 0.
2138 .. code-block:: console
2140 testpmd> port stop 0
2143 testpmd> port close 0
2147 testpmd> port detach 0
2149 EAL: PCI device 0000:0a:00.0 on NUMA socket -1
2150 EAL: remove driver: 8086:10fb rte_ixgbe_pmd
2151 EAL: PCI memory unmapped at 0x7f83bfa00000
2152 EAL: PCI memory unmapped at 0x7f83bfa80000
2156 For example, to detach a virtual device port 0.
2158 .. code-block:: console
2160 testpmd> port stop 0
2163 testpmd> port close 0
2167 testpmd> port detach 0
2169 PMD: Closing pcap ethdev on numa socket 0
2170 Port 'net_pcap0' is detached. Now total ports is 0
2173 To remove a pci device completely from the system, first detach the port from testpmd.
2174 Then the device should be moved under kernel management.
2175 Finally the device can be removed using kernel pci hotplug functionality.
2177 For example, to move a pci device under kernel management:
2179 .. code-block:: console
2181 sudo ./usertools/dpdk-devbind.py -b ixgbe 0000:0a:00.0
2183 ./usertools/dpdk-devbind.py --status
2185 Network devices using DPDK-compatible driver
2186 ============================================
2189 Network devices using kernel driver
2190 ===================================
2191 0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused=igb_uio
2193 To remove a port created by a virtual device, above steps are not needed.
2198 Start all ports or a specific port::
2200 testpmd> port start (port_id|all)
2205 Stop all ports or a specific port::
2207 testpmd> port stop (port_id|all)
2212 Close all ports or a specific port::
2214 testpmd> port close (port_id|all)
2219 Reset all ports or a specific port::
2221 testpmd> port reset (port_id|all)
2223 User should stop port(s) before resetting and (re-)start after reset.
2225 port config - queue ring size
2226 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2228 Configure a rx/tx queue ring size::
2230 testpmd> port (port_id) (rxq|txq) (queue_id) ring_size (value)
2232 Only take effect after command that (re-)start the port or command that setup specific queue.
2234 port start/stop queue
2235 ~~~~~~~~~~~~~~~~~~~~~
2237 Start/stop a rx/tx queue on a specific port::
2239 testpmd> port (port_id) (rxq|txq) (queue_id) (start|stop)
2241 port config - queue deferred start
2242 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2244 Switch on/off deferred start of a specific port queue::
2246 testpmd> port (port_id) (rxq|txq) (queue_id) deferred_start (on|off)
2249 ~~~~~~~~~~~~~~~~~~~~~
2251 Setup a rx/tx queue on a specific port::
2253 testpmd> port (port_id) (rxq|txq) (queue_id) setup
2255 Only take effect when port is started.
2260 Set the speed and duplex mode for all ports or a specific port::
2262 testpmd> port config (port_id|all) speed (10|100|1000|10000|25000|40000|50000|100000|200000|auto) \
2263 duplex (half|full|auto)
2265 port config - queues/descriptors
2266 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2268 Set number of queues/descriptors for rxq, txq, rxd and txd::
2270 testpmd> port config all (rxq|txq|rxd|txd) (value)
2272 This is equivalent to the ``--rxq``, ``--txq``, ``--rxd`` and ``--txd`` command-line options.
2274 port config - max-pkt-len
2275 ~~~~~~~~~~~~~~~~~~~~~~~~~
2277 Set the maximum packet length::
2279 testpmd> port config all max-pkt-len (value)
2281 This is equivalent to the ``--max-pkt-len`` command-line option.
2283 port config - max-lro-pkt-size
2284 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2286 Set the maximum LRO aggregated packet size::
2288 testpmd> port config all max-lro-pkt-size (value)
2290 This is equivalent to the ``--max-lro-pkt-size`` command-line option.
2292 port config - Drop Packets
2293 ~~~~~~~~~~~~~~~~~~~~~~~~~~
2295 Enable or disable packet drop on all RX queues of all ports when no receive buffers available::
2297 testpmd> port config all drop-en (on|off)
2299 Packet dropping when no receive buffers available is off by default.
2301 The ``on`` option is equivalent to the ``--enable-drop-en`` command-line option.
2306 Set the RSS (Receive Side Scaling) mode on or off::
2308 testpmd> port config all rss (all|default|eth|vlan|ip|tcp|udp|sctp|ether|port|vxlan|geneve|nvgre|vxlan-gpe|l2tpv3|esp|ah|pfcp|none)
2310 RSS is on by default.
2312 The ``all`` option is equivalent to eth|vlan|ip|tcp|udp|sctp|ether|l2tpv3|esp|ah|pfcp.
2314 The ``default`` option enables all supported RSS types reported by device info.
2316 The ``none`` option is equivalent to the ``--disable-rss`` command-line option.
2318 port config - RSS Reta
2319 ~~~~~~~~~~~~~~~~~~~~~~
2321 Set the RSS (Receive Side Scaling) redirection table::
2323 testpmd> port config all rss reta (hash,queue)[,(hash,queue)]
2328 Set the DCB mode for an individual port::
2330 testpmd> port config (port_id) dcb vt (on|off) (traffic_class) pfc (on|off)
2332 The traffic class should be 4 or 8.
2337 Set the number of packets per burst::
2339 testpmd> port config all burst (value)
2341 This is equivalent to the ``--burst`` command-line option.
2343 port config - Threshold
2344 ~~~~~~~~~~~~~~~~~~~~~~~
2346 Set thresholds for TX/RX queues::
2348 testpmd> port config all (threshold) (value)
2350 Where the threshold type can be:
2352 * ``txpt:`` Set the prefetch threshold register of the TX rings, 0 <= value <= 255.
2354 * ``txht:`` Set the host threshold register of the TX rings, 0 <= value <= 255.
2356 * ``txwt:`` Set the write-back threshold register of the TX rings, 0 <= value <= 255.
2358 * ``rxpt:`` Set the prefetch threshold register of the RX rings, 0 <= value <= 255.
2360 * ``rxht:`` Set the host threshold register of the RX rings, 0 <= value <= 255.
2362 * ``rxwt:`` Set the write-back threshold register of the RX rings, 0 <= value <= 255.
2364 * ``txfreet:`` Set the transmit free threshold of the TX rings, 0 <= value <= txd.
2366 * ``rxfreet:`` Set the transmit free threshold of the RX rings, 0 <= value <= rxd.
2368 * ``txrst:`` Set the transmit RS bit threshold of TX rings, 0 <= value <= txd.
2370 These threshold options are also available from the command-line.
2375 Set the value of ether-type for E-tag::
2377 testpmd> port config (port_id|all) l2-tunnel E-tag ether-type (value)
2379 Enable/disable the E-tag support::
2381 testpmd> port config (port_id|all) l2-tunnel E-tag (enable|disable)
2383 port config pctype mapping
2384 ~~~~~~~~~~~~~~~~~~~~~~~~~~
2386 Reset pctype mapping table::
2388 testpmd> port config (port_id) pctype mapping reset
2390 Update hardware defined pctype to software defined flow type mapping table::
2392 testpmd> port config (port_id) pctype mapping update (pctype_id_0[,pctype_id_1]*) (flow_type_id)
2396 * ``pctype_id_x``: hardware pctype id as index of bit in bitmask value of the pctype mapping table.
2398 * ``flow_type_id``: software flow type id as the index of the pctype mapping table.
2400 port config input set
2401 ~~~~~~~~~~~~~~~~~~~~~
2403 Config RSS/FDIR/FDIR flexible payload input set for some pctype::
2405 testpmd> port config (port_id) pctype (pctype_id) \
2406 (hash_inset|fdir_inset|fdir_flx_inset) \
2407 (get|set|clear) field (field_idx)
2409 Clear RSS/FDIR/FDIR flexible payload input set for some pctype::
2411 testpmd> port config (port_id) pctype (pctype_id) \
2412 (hash_inset|fdir_inset|fdir_flx_inset) clear all
2416 * ``pctype_id``: hardware packet classification types.
2417 * ``field_idx``: hardware field index.
2419 port config udp_tunnel_port
2420 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
2422 Add/remove UDP tunnel port for VXLAN/GENEVE tunneling protocols::
2424 testpmd> port config (port_id) udp_tunnel_port add|rm vxlan|geneve|vxlan-gpe (udp_port)
2426 port config tx_metadata
2427 ~~~~~~~~~~~~~~~~~~~~~~~
2429 Set Tx metadata value per port.
2430 testpmd will add this value to any Tx packet sent from this port::
2432 testpmd> port config (port_id) tx_metadata (value)
2437 Set/clear dynamic flag per port.
2438 testpmd will register this flag in the mbuf (same registration
2439 for both Tx and Rx). Then set/clear this flag for each Tx
2440 packet sent from this port. The set bit only works for Tx packet::
2442 testpmd> port config (port_id) dynf (name) (set|clear)
2447 To configure MTU(Maximum Transmission Unit) on devices using testpmd::
2449 testpmd> port config mtu (port_id) (value)
2451 port config rss hash key
2452 ~~~~~~~~~~~~~~~~~~~~~~~~
2454 To configure the RSS hash key used to compute the RSS
2455 hash of input [IP] packets received on port::
2457 testpmd> port config <port_id> rss-hash-key (ipv4|ipv4-frag|\
2458 ipv4-tcp|ipv4-udp|ipv4-sctp|ipv4-other|\
2459 ipv6|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|\
2460 ipv6-other|l2-payload|ipv6-ex|ipv6-tcp-ex|\
2461 ipv6-udp-ex <string of hex digits \
2462 (variable length, NIC dependent)>)
2467 The following sections show functions for device operations.
2472 Detach a device specified by pci address or virtual device args::
2474 testpmd> device detach (identifier)
2476 Before detaching a device associated with ports, the ports should be stopped and closed.
2478 For example, to detach a pci device whose address is 0002:03:00.0.
2480 .. code-block:: console
2482 testpmd> device detach 0002:03:00.0
2483 Removing a device...
2484 Port 1 is now closed
2485 EAL: Releasing pci mapped resource for 0002:03:00.0
2486 EAL: Calling pci_unmap_resource for 0002:03:00.0 at 0x218a050000
2487 EAL: Calling pci_unmap_resource for 0002:03:00.0 at 0x218c050000
2488 Device 0002:03:00.0 is detached
2489 Now total ports is 1
2491 For example, to detach a port created by pcap PMD.
2493 .. code-block:: console
2495 testpmd> device detach net_pcap0
2496 Removing a device...
2497 Port 0 is now closed
2498 Device net_pcap0 is detached
2499 Now total ports is 0
2502 In this case, identifier is ``net_pcap0``.
2503 This identifier format is the same as ``--vdev`` format of DPDK applications.
2505 Link Bonding Functions
2506 ----------------------
2508 The Link Bonding functions make it possible to dynamically create and
2509 manage link bonding devices from within testpmd interactive prompt.
2511 create bonded device
2512 ~~~~~~~~~~~~~~~~~~~~
2514 Create a new bonding device::
2516 testpmd> create bonded device (mode) (socket)
2518 For example, to create a bonded device in mode 1 on socket 0::
2520 testpmd> create bonded device 1 0
2521 created new bonded device (port X)
2526 Adds Ethernet device to a Link Bonding device::
2528 testpmd> add bonding slave (slave id) (port id)
2530 For example, to add Ethernet device (port 6) to a Link Bonding device (port 10)::
2532 testpmd> add bonding slave 6 10
2535 remove bonding slave
2536 ~~~~~~~~~~~~~~~~~~~~
2538 Removes an Ethernet slave device from a Link Bonding device::
2540 testpmd> remove bonding slave (slave id) (port id)
2542 For example, to remove Ethernet slave device (port 6) to a Link Bonding device (port 10)::
2544 testpmd> remove bonding slave 6 10
2549 Set the Link Bonding mode of a Link Bonding device::
2551 testpmd> set bonding mode (value) (port id)
2553 For example, to set the bonding mode of a Link Bonding device (port 10) to broadcast (mode 3)::
2555 testpmd> set bonding mode 3 10
2560 Set an Ethernet slave device as the primary device on a Link Bonding device::
2562 testpmd> set bonding primary (slave id) (port id)
2564 For example, to set the Ethernet slave device (port 6) as the primary port of a Link Bonding device (port 10)::
2566 testpmd> set bonding primary 6 10
2571 Set the MAC address of a Link Bonding device::
2573 testpmd> set bonding mac (port id) (mac)
2575 For example, to set the MAC address of a Link Bonding device (port 10) to 00:00:00:00:00:01::
2577 testpmd> set bonding mac 10 00:00:00:00:00:01
2579 set bonding xmit_balance_policy
2580 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2582 Set the transmission policy for a Link Bonding device when it is in Balance XOR mode::
2584 testpmd> set bonding xmit_balance_policy (port_id) (l2|l23|l34)
2586 For example, set a Link Bonding device (port 10) to use a balance policy of layer 3+4 (IP addresses & UDP ports)::
2588 testpmd> set bonding xmit_balance_policy 10 l34
2591 set bonding mon_period
2592 ~~~~~~~~~~~~~~~~~~~~~~
2594 Set the link status monitoring polling period in milliseconds for a bonding device.
2596 This adds support for PMD slave devices which do not support link status interrupts.
2597 When the mon_period is set to a value greater than 0 then all PMD's which do not support
2598 link status ISR will be queried every polling interval to check if their link status has changed::
2600 testpmd> set bonding mon_period (port_id) (value)
2602 For example, to set the link status monitoring polling period of bonded device (port 5) to 150ms::
2604 testpmd> set bonding mon_period 5 150
2607 set bonding lacp dedicated_queue
2608 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2610 Enable dedicated tx/rx queues on bonding devices slaves to handle LACP control plane traffic
2611 when in mode 4 (link-aggregation-802.3ad)::
2613 testpmd> set bonding lacp dedicated_queues (port_id) (enable|disable)
2616 set bonding agg_mode
2617 ~~~~~~~~~~~~~~~~~~~~
2619 Enable one of the specific aggregators mode when in mode 4 (link-aggregation-802.3ad)::
2621 testpmd> set bonding agg_mode (port_id) (bandwidth|count|stable)
2627 Show the current configuration of a Link Bonding device::
2629 testpmd> show bonding config (port id)
2632 to show the configuration a Link Bonding device (port 9) with 3 slave devices (1, 3, 4)
2633 in balance mode with a transmission policy of layer 2+3::
2635 testpmd> show bonding config 9
2637 Balance Xmit Policy: BALANCE_XMIT_POLICY_LAYER23
2639 Active Slaves (3): [1 3 4]
2646 The Register Functions can be used to read from and write to registers on the network card referenced by a port number.
2647 This is mainly useful for debugging purposes.
2648 Reference should be made to the appropriate datasheet for the network card for details on the register addresses
2649 and fields that can be accessed.
2654 Display the value of a port register::
2656 testpmd> read reg (port_id) (address)
2658 For example, to examine the Flow Director control register (FDIRCTL, 0x0000EE000) on an Intel 82599 10 GbE Controller::
2660 testpmd> read reg 0 0xEE00
2661 port 0 PCI register at offset 0xEE00: 0x4A060029 (1241907241)
2666 Display a port register bit field::
2668 testpmd> read regfield (port_id) (address) (bit_x) (bit_y)
2670 For example, reading the lowest two bits from the register in the example above::
2672 testpmd> read regfield 0 0xEE00 0 1
2673 port 0 PCI register at offset 0xEE00: bits[0, 1]=0x1 (1)
2678 Display a single port register bit::
2680 testpmd> read regbit (port_id) (address) (bit_x)
2682 For example, reading the lowest bit from the register in the example above::
2684 testpmd> read regbit 0 0xEE00 0
2685 port 0 PCI register at offset 0xEE00: bit 0=1
2690 Set the value of a port register::
2692 testpmd> write reg (port_id) (address) (value)
2694 For example, to clear a register::
2696 testpmd> write reg 0 0xEE00 0x0
2697 port 0 PCI register at offset 0xEE00: 0x00000000 (0)
2702 Set bit field of a port register::
2704 testpmd> write regfield (port_id) (address) (bit_x) (bit_y) (value)
2706 For example, writing to the register cleared in the example above::
2708 testpmd> write regfield 0 0xEE00 0 1 2
2709 port 0 PCI register at offset 0xEE00: 0x00000002 (2)
2714 Set single bit value of a port register::
2716 testpmd> write regbit (port_id) (address) (bit_x) (value)
2718 For example, to set the high bit in the register from the example above::
2720 testpmd> write regbit 0 0xEE00 31 1
2721 port 0 PCI register at offset 0xEE00: 0x8000000A (2147483658)
2723 Traffic Metering and Policing
2724 -----------------------------
2726 The following section shows functions for configuring traffic metering and
2727 policing on the ethernet device through the use of generic ethdev API.
2729 show port traffic management capability
2730 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2732 Show traffic metering and policing capability of the port::
2734 testpmd> show port meter cap (port_id)
2736 add port meter profile (srTCM rfc2967)
2737 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2739 Add meter profile (srTCM rfc2697) to the ethernet device::
2741 testpmd> add port meter profile srtcm_rfc2697 (port_id) (profile_id) \
2746 * ``profile_id``: ID for the meter profile.
2747 * ``cir``: Committed Information Rate (CIR) (bytes/second).
2748 * ``cbs``: Committed Burst Size (CBS) (bytes).
2749 * ``ebs``: Excess Burst Size (EBS) (bytes).
2751 add port meter profile (trTCM rfc2968)
2752 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2754 Add meter profile (srTCM rfc2698) to the ethernet device::
2756 testpmd> add port meter profile trtcm_rfc2698 (port_id) (profile_id) \
2757 (cir) (pir) (cbs) (pbs)
2761 * ``profile_id``: ID for the meter profile.
2762 * ``cir``: Committed information rate (bytes/second).
2763 * ``pir``: Peak information rate (bytes/second).
2764 * ``cbs``: Committed burst size (bytes).
2765 * ``pbs``: Peak burst size (bytes).
2767 add port meter profile (trTCM rfc4115)
2768 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2770 Add meter profile (trTCM rfc4115) to the ethernet device::
2772 testpmd> add port meter profile trtcm_rfc4115 (port_id) (profile_id) \
2773 (cir) (eir) (cbs) (ebs)
2777 * ``profile_id``: ID for the meter profile.
2778 * ``cir``: Committed information rate (bytes/second).
2779 * ``eir``: Excess information rate (bytes/second).
2780 * ``cbs``: Committed burst size (bytes).
2781 * ``ebs``: Excess burst size (bytes).
2783 delete port meter profile
2784 ~~~~~~~~~~~~~~~~~~~~~~~~~
2786 Delete meter profile from the ethernet device::
2788 testpmd> del port meter profile (port_id) (profile_id)
2793 Create new meter object for the ethernet device::
2795 testpmd> create port meter (port_id) (mtr_id) (profile_id) \
2796 (meter_enable) (g_action) (y_action) (r_action) (stats_mask) (shared) \
2797 (use_pre_meter_color) [(dscp_tbl_entry0) (dscp_tbl_entry1)...\
2802 * ``mtr_id``: meter object ID.
2803 * ``profile_id``: ID for the meter profile.
2804 * ``meter_enable``: When this parameter has a non-zero value, the meter object
2805 gets enabled at the time of creation, otherwise remains disabled.
2806 * ``g_action``: Policer action for the packet with green color.
2807 * ``y_action``: Policer action for the packet with yellow color.
2808 * ``r_action``: Policer action for the packet with red color.
2809 * ``stats_mask``: Mask of statistics counter types to be enabled for the
2811 * ``shared``: When this parameter has a non-zero value, the meter object is
2812 shared by multiple flows. Otherwise, meter object is used by single flow.
2813 * ``use_pre_meter_color``: When this parameter has a non-zero value, the
2814 input color for the current meter object is determined by the latest meter
2815 object in the same flow. Otherwise, the current meter object uses the
2816 *dscp_table* to determine the input color.
2817 * ``dscp_tbl_entryx``: DSCP table entry x providing meter providing input
2818 color, 0 <= x <= 63.
2823 Enable meter for the ethernet device::
2825 testpmd> enable port meter (port_id) (mtr_id)
2830 Disable meter for the ethernet device::
2832 testpmd> disable port meter (port_id) (mtr_id)
2837 Delete meter for the ethernet device::
2839 testpmd> del port meter (port_id) (mtr_id)
2841 Set port meter profile
2842 ~~~~~~~~~~~~~~~~~~~~~~
2844 Set meter profile for the ethernet device::
2846 testpmd> set port meter profile (port_id) (mtr_id) (profile_id)
2848 set port meter dscp table
2849 ~~~~~~~~~~~~~~~~~~~~~~~~~
2851 Set meter dscp table for the ethernet device::
2853 testpmd> set port meter dscp table (port_id) (mtr_id) [(dscp_tbl_entry0) \
2854 (dscp_tbl_entry1)...(dscp_tbl_entry63)]
2856 set port meter policer action
2857 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2859 Set meter policer action for the ethernet device::
2861 testpmd> set port meter policer action (port_id) (mtr_id) (action_mask) \
2862 (action0) [(action1) (action1)]
2866 * ``action_mask``: Bit mask indicating which policer actions need to be
2867 updated. One or more policer actions can be updated in a single function
2868 invocation. To update the policer action associated with color C, bit
2869 (1 << C) needs to be set in *action_mask* and element at position C
2870 in the *actions* array needs to be valid.
2871 * ``actionx``: Policer action for the color x,
2872 RTE_MTR_GREEN <= x < RTE_MTR_COLORS
2874 set port meter stats mask
2875 ~~~~~~~~~~~~~~~~~~~~~~~~~
2877 Set meter stats mask for the ethernet device::
2879 testpmd> set port meter stats mask (port_id) (mtr_id) (stats_mask)
2883 * ``stats_mask``: Bit mask indicating statistics counter types to be enabled.
2885 show port meter stats
2886 ~~~~~~~~~~~~~~~~~~~~~
2888 Show meter stats of the ethernet device::
2890 testpmd> show port meter stats (port_id) (mtr_id) (clear)
2894 * ``clear``: Flag that indicates whether the statistics counters should
2895 be cleared (i.e. set to zero) immediately after they have been read or not.
2900 The following section shows functions for configuring traffic management on
2901 the ethernet device through the use of generic TM API.
2903 show port traffic management capability
2904 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2906 Show traffic management capability of the port::
2908 testpmd> show port tm cap (port_id)
2910 show port traffic management capability (hierarchy level)
2911 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2913 Show traffic management hierarchy level capability of the port::
2915 testpmd> show port tm level cap (port_id) (level_id)
2917 show port traffic management capability (hierarchy node level)
2918 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2920 Show the traffic management hierarchy node capability of the port::
2922 testpmd> show port tm node cap (port_id) (node_id)
2924 show port traffic management hierarchy node type
2925 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2927 Show the port traffic management hierarchy node type::
2929 testpmd> show port tm node type (port_id) (node_id)
2931 show port traffic management hierarchy node stats
2932 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2934 Show the port traffic management hierarchy node statistics::
2936 testpmd> show port tm node stats (port_id) (node_id) (clear)
2940 * ``clear``: When this parameter has a non-zero value, the statistics counters
2941 are cleared (i.e. set to zero) immediately after they have been read,
2942 otherwise the statistics counters are left untouched.
2944 Add port traffic management private shaper profile
2945 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2947 Add the port traffic management private shaper profile::
2949 testpmd> add port tm node shaper profile (port_id) (shaper_profile_id) \
2950 (cmit_tb_rate) (cmit_tb_size) (peak_tb_rate) (peak_tb_size) \
2951 (packet_length_adjust) (packet_mode)
2955 * ``shaper_profile id``: Shaper profile ID for the new profile.
2956 * ``cmit_tb_rate``: Committed token bucket rate (bytes per second or packets per second).
2957 * ``cmit_tb_size``: Committed token bucket size (bytes or packets).
2958 * ``peak_tb_rate``: Peak token bucket rate (bytes per second or packets per second).
2959 * ``peak_tb_size``: Peak token bucket size (bytes or packets).
2960 * ``packet_length_adjust``: The value (bytes) to be added to the length of
2961 each packet for the purpose of shaping. This parameter value can be used to
2962 correct the packet length with the framing overhead bytes that are consumed
2964 * ``packet_mode``: Shaper configured in packet mode. This parameter value if
2965 zero, configures shaper in byte mode and if non-zero configures it in packet
2968 Delete port traffic management private shaper profile
2969 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2971 Delete the port traffic management private shaper::
2973 testpmd> del port tm node shaper profile (port_id) (shaper_profile_id)
2977 * ``shaper_profile id``: Shaper profile ID that needs to be deleted.
2979 Add port traffic management shared shaper
2980 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2982 Create the port traffic management shared shaper::
2984 testpmd> add port tm node shared shaper (port_id) (shared_shaper_id) \
2989 * ``shared_shaper_id``: Shared shaper ID to be created.
2990 * ``shaper_profile id``: Shaper profile ID for shared shaper.
2992 Set port traffic management shared shaper
2993 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2995 Update the port traffic management shared shaper::
2997 testpmd> set port tm node shared shaper (port_id) (shared_shaper_id) \
3002 * ``shared_shaper_id``: Shared shaper ID to be update.
3003 * ``shaper_profile id``: Shaper profile ID for shared shaper.
3005 Delete port traffic management shared shaper
3006 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3008 Delete the port traffic management shared shaper::
3010 testpmd> del port tm node shared shaper (port_id) (shared_shaper_id)
3014 * ``shared_shaper_id``: Shared shaper ID to be deleted.
3016 Set port traffic management hierarchy node private shaper
3017 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3019 set the port traffic management hierarchy node private shaper::
3021 testpmd> set port tm node shaper profile (port_id) (node_id) \
3026 * ``shaper_profile id``: Private shaper profile ID to be enabled on the
3029 Add port traffic management WRED profile
3030 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3032 Create a new WRED profile::
3034 testpmd> add port tm node wred profile (port_id) (wred_profile_id) \
3035 (color_g) (min_th_g) (max_th_g) (maxp_inv_g) (wq_log2_g) \
3036 (color_y) (min_th_y) (max_th_y) (maxp_inv_y) (wq_log2_y) \
3037 (color_r) (min_th_r) (max_th_r) (maxp_inv_r) (wq_log2_r)
3041 * ``wred_profile id``: Identifier for the newly create WRED profile
3042 * ``color_g``: Packet color (green)
3043 * ``min_th_g``: Minimum queue threshold for packet with green color
3044 * ``max_th_g``: Minimum queue threshold for packet with green color
3045 * ``maxp_inv_g``: Inverse of packet marking probability maximum value (maxp)
3046 * ``wq_log2_g``: Negated log2 of queue weight (wq)
3047 * ``color_y``: Packet color (yellow)
3048 * ``min_th_y``: Minimum queue threshold for packet with yellow color
3049 * ``max_th_y``: Minimum queue threshold for packet with yellow color
3050 * ``maxp_inv_y``: Inverse of packet marking probability maximum value (maxp)
3051 * ``wq_log2_y``: Negated log2 of queue weight (wq)
3052 * ``color_r``: Packet color (red)
3053 * ``min_th_r``: Minimum queue threshold for packet with yellow color
3054 * ``max_th_r``: Minimum queue threshold for packet with yellow color
3055 * ``maxp_inv_r``: Inverse of packet marking probability maximum value (maxp)
3056 * ``wq_log2_r``: Negated log2 of queue weight (wq)
3058 Delete port traffic management WRED profile
3059 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3061 Delete the WRED profile::
3063 testpmd> del port tm node wred profile (port_id) (wred_profile_id)
3065 Add port traffic management hierarchy nonleaf node
3066 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3068 Add nonleaf node to port traffic management hierarchy::
3070 testpmd> add port tm nonleaf node (port_id) (node_id) (parent_node_id) \
3071 (priority) (weight) (level_id) (shaper_profile_id) \
3072 (n_sp_priorities) (stats_mask) (n_shared_shapers) \
3073 [(shared_shaper_0) (shared_shaper_1) ...] \
3077 * ``parent_node_id``: Node ID of the parent.
3078 * ``priority``: Node priority (highest node priority is zero). This is used by
3079 the SP algorithm running on the parent node for scheduling this node.
3080 * ``weight``: Node weight (lowest weight is one). The node weight is relative
3081 to the weight sum of all siblings that have the same priority. It is used by
3082 the WFQ algorithm running on the parent node for scheduling this node.
3083 * ``level_id``: Hierarchy level of the node.
3084 * ``shaper_profile_id``: Shaper profile ID of the private shaper to be used by
3086 * ``n_sp_priorities``: Number of strict priorities.
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 nonleaf node with packet mode
3092 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3094 Add nonleaf node with packet mode to port traffic management hierarchy::
3096 testpmd> add port tm nonleaf node pktmode (port_id) (node_id) (parent_node_id) \
3097 (priority) (weight) (level_id) (shaper_profile_id) \
3098 (n_sp_priorities) (stats_mask) (n_shared_shapers) \
3099 [(shared_shaper_0) (shared_shaper_1) ...] \
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 * ``n_sp_priorities``: Number of strict priorities. Packet mode is enabled on
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 Add port traffic management hierarchy leaf node
3119 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3121 Add leaf node to port traffic management hierarchy::
3123 testpmd> add port tm leaf node (port_id) (node_id) (parent_node_id) \
3124 (priority) (weight) (level_id) (shaper_profile_id) \
3125 (cman_mode) (wred_profile_id) (stats_mask) (n_shared_shapers) \
3126 [(shared_shaper_id) (shared_shaper_id) ...] \
3130 * ``parent_node_id``: Node ID of the parent.
3131 * ``priority``: Node priority (highest node priority is zero). This is used by
3132 the SP algorithm running on the parent node for scheduling this node.
3133 * ``weight``: Node weight (lowest weight is one). The node weight is relative
3134 to the weight sum of all siblings that have the same priority. It is used by
3135 the WFQ algorithm running on the parent node for scheduling this node.
3136 * ``level_id``: Hierarchy level of the node.
3137 * ``shaper_profile_id``: Shaper profile ID of the private shaper to be used by
3139 * ``cman_mode``: Congestion management mode to be enabled for this node.
3140 * ``wred_profile_id``: WRED profile id to be enabled for this node.
3141 * ``stats_mask``: Mask of statistics counter types to be enabled for this node.
3142 * ``n_shared_shapers``: Number of shared shapers.
3143 * ``shared_shaper_id``: Shared shaper id.
3145 Delete port traffic management hierarchy node
3146 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3148 Delete node from port traffic management hierarchy::
3150 testpmd> del port tm node (port_id) (node_id)
3152 Update port traffic management hierarchy parent node
3153 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3155 Update port traffic management hierarchy parent node::
3157 testpmd> set port tm node parent (port_id) (node_id) (parent_node_id) \
3160 This function can only be called after the hierarchy commit invocation. Its
3161 success depends on the port support for this operation, as advertised through
3162 the port capability set. This function is valid for all nodes of the traffic
3163 management hierarchy except root node.
3165 Suspend port traffic management hierarchy node
3166 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3168 testpmd> suspend port tm node (port_id) (node_id)
3170 Resume port traffic management hierarchy node
3171 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3173 testpmd> resume port tm node (port_id) (node_id)
3175 Commit port traffic management hierarchy
3176 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3178 Commit the traffic management hierarchy on the port::
3180 testpmd> port tm hierarchy commit (port_id) (clean_on_fail)
3184 * ``clean_on_fail``: When set to non-zero, hierarchy is cleared on function
3185 call failure. On the other hand, hierarchy is preserved when this parameter
3188 Set port traffic management mark VLAN dei
3189 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3191 Enables/Disables the traffic management marking on the port for VLAN packets::
3193 testpmd> set port tm mark vlan_dei <port_id> <green> <yellow> <red>
3197 * ``port_id``: The port which on which VLAN packets marked as ``green`` or
3198 ``yellow`` or ``red`` will have dei bit enabled
3200 * ``green`` enable 1, disable 0 marking for dei bit of VLAN packets marked as green
3202 * ``yellow`` enable 1, disable 0 marking for dei bit of VLAN packets marked as yellow
3204 * ``red`` enable 1, disable 0 marking for dei bit of VLAN packets marked as red
3206 Set port traffic management mark IP dscp
3207 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3209 Enables/Disables the traffic management marking on the port for IP dscp packets::
3211 testpmd> set port tm mark ip_dscp <port_id> <green> <yellow> <red>
3215 * ``port_id``: The port which on which IP packets marked as ``green`` or
3216 ``yellow`` or ``red`` will have IP dscp bits updated
3218 * ``green`` enable 1, disable 0 marking IP dscp to low drop precedence for green packets
3220 * ``yellow`` enable 1, disable 0 marking IP dscp to medium drop precedence for yellow packets
3222 * ``red`` enable 1, disable 0 marking IP dscp to high drop precedence for red packets
3224 Set port traffic management mark IP ecn
3225 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3227 Enables/Disables the traffic management marking on the port for IP ecn packets::
3229 testpmd> set port tm mark ip_ecn <port_id> <green> <yellow> <red>
3233 * ``port_id``: The port which on which IP packets marked as ``green`` or
3234 ``yellow`` or ``red`` will have IP ecn bits updated
3236 * ``green`` enable 1, disable 0 marking IP ecn for green marked packets with ecn of 2'b01 or 2'b10
3237 to ecn of 2'b11 when IP is caring TCP or SCTP
3239 * ``yellow`` enable 1, disable 0 marking IP ecn for yellow marked packets with ecn of 2'b01 or 2'b10
3240 to ecn of 2'b11 when IP is caring TCP or SCTP
3242 * ``red`` enable 1, disable 0 marking IP ecn for yellow marked packets with ecn of 2'b01 or 2'b10
3243 to ecn of 2'b11 when IP is caring TCP or SCTP
3248 This section details the available filter functions that are available.
3250 Note these functions interface the deprecated legacy filtering framework,
3251 superseded by *rte_flow*. See `Flow rules management`_.
3254 ~~~~~~~~~~~~~~~~~~~~
3256 Add or delete a L2 Ethertype filter, which identify packets by their L2 Ethertype mainly assign them to a receive queue::
3258 ethertype_filter (port_id) (add|del) (mac_addr|mac_ignr) (mac_address) \
3259 ethertype (ether_type) (drop|fwd) queue (queue_id)
3261 The available information parameters are:
3263 * ``port_id``: The port which the Ethertype filter assigned on.
3265 * ``mac_addr``: Compare destination mac address.
3267 * ``mac_ignr``: Ignore destination mac address match.
3269 * ``mac_address``: Destination mac address to match.
3271 * ``ether_type``: The EtherType value want to match,
3272 for example 0x0806 for ARP packet. 0x0800 (IPv4) and 0x86DD (IPv6) are invalid.
3274 * ``queue_id``: The receive queue associated with this EtherType filter.
3275 It is meaningless when deleting or dropping.
3277 Example, to add/remove an ethertype filter rule::
3279 testpmd> ethertype_filter 0 add mac_ignr 00:11:22:33:44:55 \
3280 ethertype 0x0806 fwd queue 3
3282 testpmd> ethertype_filter 0 del mac_ignr 00:11:22:33:44:55 \
3283 ethertype 0x0806 fwd queue 3
3288 Add or delete a 2-tuple filter,
3289 which identifies packets by specific protocol and destination TCP/UDP port
3290 and forwards packets into one of the receive queues::
3292 2tuple_filter (port_id) (add|del) dst_port (dst_port_value) \
3293 protocol (protocol_value) mask (mask_value) \
3294 tcp_flags (tcp_flags_value) priority (prio_value) \
3297 The available information parameters are:
3299 * ``port_id``: The port which the 2-tuple filter assigned on.
3301 * ``dst_port_value``: Destination port in L4.
3303 * ``protocol_value``: IP L4 protocol.
3305 * ``mask_value``: Participates in the match or not by bit for field above, 1b means participate.
3307 * ``tcp_flags_value``: TCP control bits. The non-zero value is invalid, when the pro_value is not set to 0x06 (TCP).
3309 * ``prio_value``: Priority of this filter.
3311 * ``queue_id``: The receive queue associated with this 2-tuple filter.
3313 Example, to add/remove an 2tuple filter rule::
3315 testpmd> 2tuple_filter 0 add dst_port 32 protocol 0x06 mask 0x03 \
3316 tcp_flags 0x02 priority 3 queue 3
3318 testpmd> 2tuple_filter 0 del dst_port 32 protocol 0x06 mask 0x03 \
3319 tcp_flags 0x02 priority 3 queue 3
3324 Add or delete a 5-tuple filter,
3325 which consists of a 5-tuple (protocol, source and destination IP addresses, source and destination TCP/UDP/SCTP port)
3326 and routes packets into one of the receive queues::
3328 5tuple_filter (port_id) (add|del) dst_ip (dst_address) src_ip \
3329 (src_address) dst_port (dst_port_value) \
3330 src_port (src_port_value) protocol (protocol_value) \
3331 mask (mask_value) tcp_flags (tcp_flags_value) \
3332 priority (prio_value) queue (queue_id)
3334 The available information parameters are:
3336 * ``port_id``: The port which the 5-tuple filter assigned on.
3338 * ``dst_address``: Destination IP address.
3340 * ``src_address``: Source IP address.
3342 * ``dst_port_value``: TCP/UDP destination port.
3344 * ``src_port_value``: TCP/UDP source port.
3346 * ``protocol_value``: L4 protocol.
3348 * ``mask_value``: Participates in the match or not by bit for field above, 1b means participate
3350 * ``tcp_flags_value``: TCP control bits. The non-zero value is invalid, when the protocol_value is not set to 0x06 (TCP).
3352 * ``prio_value``: The priority of this filter.
3354 * ``queue_id``: The receive queue associated with this 5-tuple filter.
3356 Example, to add/remove an 5tuple filter rule::
3358 testpmd> 5tuple_filter 0 add dst_ip 2.2.2.5 src_ip 2.2.2.4 \
3359 dst_port 64 src_port 32 protocol 0x06 mask 0x1F \
3360 flags 0x0 priority 3 queue 3
3362 testpmd> 5tuple_filter 0 del dst_ip 2.2.2.5 src_ip 2.2.2.4 \
3363 dst_port 64 src_port 32 protocol 0x06 mask 0x1F \
3364 flags 0x0 priority 3 queue 3
3369 Using the SYN filter, TCP packets whose *SYN* flag is set can be forwarded to a separate queue::
3371 syn_filter (port_id) (add|del) priority (high|low) queue (queue_id)
3373 The available information parameters are:
3375 * ``port_id``: The port which the SYN filter assigned on.
3377 * ``high``: This SYN filter has higher priority than other filters.
3379 * ``low``: This SYN filter has lower priority than other filters.
3381 * ``queue_id``: The receive queue associated with this SYN filter
3385 testpmd> syn_filter 0 add priority high queue 3
3390 With flex filter, packets can be recognized by any arbitrary pattern within the first 128 bytes of the packet
3391 and routed into one of the receive queues::
3393 flex_filter (port_id) (add|del) len (len_value) bytes (bytes_value) \
3394 mask (mask_value) priority (prio_value) queue (queue_id)
3396 The available information parameters are:
3398 * ``port_id``: The port which the Flex filter is assigned on.
3400 * ``len_value``: Filter length in bytes, no greater than 128.
3402 * ``bytes_value``: A string in hexadecimal, means the value the flex filter needs to match.
3404 * ``mask_value``: A string in hexadecimal, bit 1 means corresponding byte participates in the match.
3406 * ``prio_value``: The priority of this filter.
3408 * ``queue_id``: The receive queue associated with this Flex filter.
3412 testpmd> flex_filter 0 add len 16 bytes 0x00000000000000000000000008060000 \
3413 mask 000C priority 3 queue 3
3415 testpmd> flex_filter 0 del len 16 bytes 0x00000000000000000000000008060000 \
3416 mask 000C priority 3 queue 3
3419 .. _testpmd_flow_director:
3421 flow_director_filter
3422 ~~~~~~~~~~~~~~~~~~~~
3424 The Flow Director works in receive mode to identify specific flows or sets of flows and route them to specific queues.
3426 Four types of filtering are supported which are referred to as Perfect Match, Signature, Perfect-mac-vlan and
3427 Perfect-tunnel filters, the match mode is set by the ``--pkt-filter-mode`` command-line parameter:
3429 * Perfect match filters.
3430 The hardware checks a match between the masked fields of the received packets and the programmed filters.
3431 The masked fields are for IP flow.
3433 * Signature filters.
3434 The hardware checks a match between a hash-based signature of the masked fields of the received packet.
3436 * Perfect-mac-vlan match filters.
3437 The hardware checks a match between the masked fields of the received packets and the programmed filters.
3438 The masked fields are for MAC VLAN flow.
3440 * Perfect-tunnel match filters.
3441 The hardware checks a match between the masked fields of the received packets and the programmed filters.
3442 The masked fields are for tunnel flow.
3444 * Perfect-raw-flow-type match filters.
3445 The hardware checks a match between the masked fields of the received packets and pre-loaded raw (template) packet.
3446 The masked fields are specified by input sets.
3448 The Flow Director filters can match the different fields for different type of packet: flow type, specific input set
3449 per flow type and the flexible payload.
3451 The Flow Director can also mask out parts of all of these fields so that filters
3452 are only applied to certain fields or parts of the fields.
3454 Note that for raw flow type mode the source and destination fields in the
3455 raw packet buffer need to be presented in a reversed order with respect
3456 to the expected received packets.
3457 For example: IP source and destination addresses or TCP/UDP/SCTP
3458 source and destination ports
3460 Different NICs may have different capabilities, command show port fdir (port_id) can be used to acquire the information.
3462 # Commands to add flow director filters of different flow types::
3464 flow_director_filter (port_id) mode IP (add|del|update) \
3465 flow (ipv4-other|ipv4-frag|ipv6-other|ipv6-frag) \
3466 src (src_ip_address) dst (dst_ip_address) \
3467 tos (tos_value) proto (proto_value) ttl (ttl_value) \
3468 vlan (vlan_value) flexbytes (flexbytes_value) \
3469 (drop|fwd) pf|vf(vf_id) queue (queue_id) \
3472 flow_director_filter (port_id) mode IP (add|del|update) \
3473 flow (ipv4-tcp|ipv4-udp|ipv6-tcp|ipv6-udp) \
3474 src (src_ip_address) (src_port) \
3475 dst (dst_ip_address) (dst_port) \
3476 tos (tos_value) ttl (ttl_value) \
3477 vlan (vlan_value) flexbytes (flexbytes_value) \
3478 (drop|fwd) queue pf|vf(vf_id) (queue_id) \
3481 flow_director_filter (port_id) mode IP (add|del|update) \
3482 flow (ipv4-sctp|ipv6-sctp) \
3483 src (src_ip_address) (src_port) \
3484 dst (dst_ip_address) (dst_port) \
3485 tos (tos_value) ttl (ttl_value) \
3486 tag (verification_tag) vlan (vlan_value) \
3487 flexbytes (flexbytes_value) (drop|fwd) \
3488 pf|vf(vf_id) queue (queue_id) fd_id (fd_id_value)
3490 flow_director_filter (port_id) mode IP (add|del|update) flow l2_payload \
3491 ether (ethertype) flexbytes (flexbytes_value) \
3492 (drop|fwd) pf|vf(vf_id) queue (queue_id)
3495 flow_director_filter (port_id) mode MAC-VLAN (add|del|update) \
3496 mac (mac_address) vlan (vlan_value) \
3497 flexbytes (flexbytes_value) (drop|fwd) \
3498 queue (queue_id) fd_id (fd_id_value)
3500 flow_director_filter (port_id) mode Tunnel (add|del|update) \
3501 mac (mac_address) vlan (vlan_value) \
3502 tunnel (NVGRE|VxLAN) tunnel-id (tunnel_id_value) \
3503 flexbytes (flexbytes_value) (drop|fwd) \
3504 queue (queue_id) fd_id (fd_id_value)
3506 flow_director_filter (port_id) mode raw (add|del|update) flow (flow_id) \
3507 (drop|fwd) queue (queue_id) fd_id (fd_id_value) \
3508 packet (packet file name)
3510 For example, to add an ipv4-udp flow type filter::
3512 testpmd> flow_director_filter 0 mode IP add flow ipv4-udp src 2.2.2.3 32 \
3513 dst 2.2.2.5 33 tos 2 ttl 40 vlan 0x1 flexbytes (0x88,0x48) \
3514 fwd pf queue 1 fd_id 1
3516 For example, add an ipv4-other flow type filter::
3518 testpmd> flow_director_filter 0 mode IP add flow ipv4-other src 2.2.2.3 \
3519 dst 2.2.2.5 tos 2 proto 20 ttl 40 vlan 0x1 \
3520 flexbytes (0x88,0x48) fwd pf queue 1 fd_id 1
3525 Flush all flow director filters on a device::
3527 testpmd> flush_flow_director (port_id)
3529 Example, to flush all flow director filter on port 0::
3531 testpmd> flush_flow_director 0
3536 Set flow director's input masks::
3538 flow_director_mask (port_id) mode IP vlan (vlan_value) \
3539 src_mask (ipv4_src) (ipv6_src) (src_port) \
3540 dst_mask (ipv4_dst) (ipv6_dst) (dst_port)
3542 flow_director_mask (port_id) mode MAC-VLAN vlan (vlan_value)
3544 flow_director_mask (port_id) mode Tunnel vlan (vlan_value) \
3545 mac (mac_value) tunnel-type (tunnel_type_value) \
3546 tunnel-id (tunnel_id_value)
3548 Example, to set flow director mask on port 0::
3550 testpmd> flow_director_mask 0 mode IP vlan 0xefff \
3551 src_mask 255.255.255.255 \
3552 FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF \
3553 dst_mask 255.255.255.255 \
3554 FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF
3556 flow_director_flex_mask
3557 ~~~~~~~~~~~~~~~~~~~~~~~
3559 set masks of flow director's flexible payload based on certain flow type::
3561 testpmd> flow_director_flex_mask (port_id) \
3562 flow (none|ipv4-other|ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
3563 ipv6-other|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp| \
3564 l2_payload|all) (mask)
3566 Example, to set flow director's flex mask for all flow type on port 0::
3568 testpmd> flow_director_flex_mask 0 flow all \
3569 (0xff,0xff,0,0,0,0,0,0,0,0,0,0,0,0,0,0)
3572 flow_director_flex_payload
3573 ~~~~~~~~~~~~~~~~~~~~~~~~~~
3575 Configure flexible payload selection::
3577 flow_director_flex_payload (port_id) (raw|l2|l3|l4) (config)
3579 For example, to select the first 16 bytes from the offset 4 (bytes) of packet's payload as flexible payload::
3581 testpmd> flow_director_flex_payload 0 l4 \
3582 (4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19)
3584 get_sym_hash_ena_per_port
3585 ~~~~~~~~~~~~~~~~~~~~~~~~~
3587 Get symmetric hash enable configuration per port::
3589 get_sym_hash_ena_per_port (port_id)
3591 For example, to get symmetric hash enable configuration of port 1::
3593 testpmd> get_sym_hash_ena_per_port 1
3595 set_sym_hash_ena_per_port
3596 ~~~~~~~~~~~~~~~~~~~~~~~~~
3598 Set symmetric hash enable configuration per port to enable or disable::
3600 set_sym_hash_ena_per_port (port_id) (enable|disable)
3602 For example, to set symmetric hash enable configuration of port 1 to enable::
3604 testpmd> set_sym_hash_ena_per_port 1 enable
3606 get_hash_global_config
3607 ~~~~~~~~~~~~~~~~~~~~~~
3609 Get the global configurations of hash filters::
3611 get_hash_global_config (port_id)
3613 For example, to get the global configurations of hash filters of port 1::
3615 testpmd> get_hash_global_config 1
3617 set_hash_global_config
3618 ~~~~~~~~~~~~~~~~~~~~~~
3620 Set the global configurations of hash filters::
3622 set_hash_global_config (port_id) (toeplitz|simple_xor|symmetric_toeplitz|default) \
3623 (ipv4|ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp|ipv4-other|ipv6|ipv6-frag| \
3624 ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other|l2_payload|<flow_id>) \
3627 For example, to enable simple_xor for flow type of ipv6 on port 2::
3629 testpmd> set_hash_global_config 2 simple_xor ipv6 enable
3634 Set the input set for hash::
3636 set_hash_input_set (port_id) (ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
3637 ipv4-other|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other| \
3638 l2_payload|<flow_id>) (ovlan|ivlan|src-ipv4|dst-ipv4|src-ipv6|dst-ipv6| \
3639 ipv4-tos|ipv4-proto|ipv6-tc|ipv6-next-header|udp-src-port|udp-dst-port| \
3640 tcp-src-port|tcp-dst-port|sctp-src-port|sctp-dst-port|sctp-veri-tag| \
3641 udp-key|gre-key|fld-1st|fld-2nd|fld-3rd|fld-4th|fld-5th|fld-6th|fld-7th| \
3642 fld-8th|none) (select|add)
3644 For example, to add source IP to hash input set for flow type of ipv4-udp on port 0::
3646 testpmd> set_hash_input_set 0 ipv4-udp src-ipv4 add
3651 The Flow Director filters can match the different fields for different type of packet, i.e. specific input set
3652 on per flow type and the flexible payload. This command can be used to change input set for each flow type.
3654 Set the input set for flow director::
3656 set_fdir_input_set (port_id) (ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
3657 ipv4-other|ipv6|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other| \
3658 l2_payload|<flow_id>) (ivlan|ethertype|src-ipv4|dst-ipv4|src-ipv6|dst-ipv6| \
3659 ipv4-tos|ipv4-proto|ipv4-ttl|ipv6-tc|ipv6-next-header|ipv6-hop-limits| \
3660 tudp-src-port|udp-dst-port|cp-src-port|tcp-dst-port|sctp-src-port| \
3661 sctp-dst-port|sctp-veri-tag|none) (select|add)
3663 For example to add source IP to FD input set for flow type of ipv4-udp on port 0::
3665 testpmd> set_fdir_input_set 0 ipv4-udp src-ipv4 add
3670 Set different GRE key length for input set::
3672 global_config (port_id) gre-key-len (number in bytes)
3674 For example to set GRE key length for input set to 4 bytes on port 0::
3676 testpmd> global_config 0 gre-key-len 4
3679 .. _testpmd_rte_flow:
3681 Flow rules management
3682 ---------------------
3684 Control of the generic flow API (*rte_flow*) is fully exposed through the
3685 ``flow`` command (validation, creation, destruction, queries and operation
3688 Considering *rte_flow* overlaps with all `Filter Functions`_, using both
3689 features simultaneously may cause undefined side-effects and is therefore
3695 Because the ``flow`` command uses dynamic tokens to handle the large number
3696 of possible flow rules combinations, its behavior differs slightly from
3697 other commands, in particular:
3699 - Pressing *?* or the *<tab>* key displays contextual help for the current
3700 token, not that of the entire command.
3702 - Optional and repeated parameters are supported (provided they are listed
3703 in the contextual help).
3705 The first parameter stands for the operation mode. Possible operations and
3706 their general syntax are described below. They are covered in detail in the
3709 - Check whether a flow rule can be created::
3711 flow validate {port_id}
3712 [group {group_id}] [priority {level}] [ingress] [egress] [transfer]
3713 pattern {item} [/ {item} [...]] / end
3714 actions {action} [/ {action} [...]] / end
3716 - Create a flow rule::
3718 flow create {port_id}
3719 [group {group_id}] [priority {level}] [ingress] [egress] [transfer]
3720 pattern {item} [/ {item} [...]] / end
3721 actions {action} [/ {action} [...]] / end
3723 - Destroy specific flow rules::
3725 flow destroy {port_id} rule {rule_id} [...]
3727 - Destroy all flow rules::
3729 flow flush {port_id}
3731 - Query an existing flow rule::
3733 flow query {port_id} {rule_id} {action}
3735 - List existing flow rules sorted by priority, filtered by group
3738 flow list {port_id} [group {group_id}] [...]
3740 - Restrict ingress traffic to the defined flow rules::
3742 flow isolate {port_id} {boolean}
3744 - Dump internal representation information of all flows in hardware::
3746 flow dump {port_id} {output_file}
3748 - List and destroy aged flow rules::
3750 flow aged {port_id} [destroy]
3752 Validating flow rules
3753 ~~~~~~~~~~~~~~~~~~~~~
3755 ``flow validate`` reports whether a flow rule would be accepted by the
3756 underlying device in its current state but stops short of creating it. It is
3757 bound to ``rte_flow_validate()``::
3759 flow validate {port_id}
3760 [group {group_id}] [priority {level}] [ingress] [egress] [transfer]
3761 pattern {item} [/ {item} [...]] / end
3762 actions {action} [/ {action} [...]] / end
3764 If successful, it will show::
3768 Otherwise it will show an error message of the form::
3770 Caught error type [...] ([...]): [...]
3772 This command uses the same parameters as ``flow create``, their format is
3773 described in `Creating flow rules`_.
3775 Check whether redirecting any Ethernet packet received on port 0 to RX queue
3776 index 6 is supported::
3778 testpmd> flow validate 0 ingress pattern eth / end
3779 actions queue index 6 / end
3783 Port 0 does not support TCPv6 rules::
3785 testpmd> flow validate 0 ingress pattern eth / ipv6 / tcp / end
3787 Caught error type 9 (specific pattern item): Invalid argument
3793 ``flow create`` validates and creates the specified flow rule. It is bound
3794 to ``rte_flow_create()``::
3796 flow create {port_id}
3797 [group {group_id}] [priority {level}] [ingress] [egress] [transfer]
3798 pattern {item} [/ {item} [...]] / end
3799 actions {action} [/ {action} [...]] / end
3801 If successful, it will return a flow rule ID usable with other commands::
3803 Flow rule #[...] created
3805 Otherwise it will show an error message of the form::
3807 Caught error type [...] ([...]): [...]
3809 Parameters describe in the following order:
3811 - Attributes (*group*, *priority*, *ingress*, *egress*, *transfer* tokens).
3812 - A matching pattern, starting with the *pattern* token and terminated by an
3814 - Actions, starting with the *actions* token and terminated by an *end*
3817 These translate directly to *rte_flow* objects provided as-is to the
3818 underlying functions.
3820 The shortest valid definition only comprises mandatory tokens::
3822 testpmd> flow create 0 pattern end actions end
3824 Note that PMDs may refuse rules that essentially do nothing such as this
3827 **All unspecified object values are automatically initialized to 0.**
3832 These tokens affect flow rule attributes (``struct rte_flow_attr``) and are
3833 specified before the ``pattern`` token.
3835 - ``group {group id}``: priority group.
3836 - ``priority {level}``: priority level within group.
3837 - ``ingress``: rule applies to ingress traffic.
3838 - ``egress``: rule applies to egress traffic.
3839 - ``transfer``: apply rule directly to endpoints found in pattern.
3841 Each instance of an attribute specified several times overrides the previous
3842 value as shown below (group 4 is used)::
3844 testpmd> flow create 0 group 42 group 24 group 4 [...]
3846 Note that once enabled, ``ingress`` and ``egress`` cannot be disabled.
3848 While not specifying a direction is an error, some rules may allow both
3851 Most rules affect RX therefore contain the ``ingress`` token::
3853 testpmd> flow create 0 ingress pattern [...]
3858 A matching pattern starts after the ``pattern`` token. It is made of pattern
3859 items and is terminated by a mandatory ``end`` item.
3861 Items are named after their type (*RTE_FLOW_ITEM_TYPE_* from ``enum
3862 rte_flow_item_type``).
3864 The ``/`` token is used as a separator between pattern items as shown
3867 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end [...]
3869 Note that protocol items like these must be stacked from lowest to highest
3870 layer to make sense. For instance, the following rule is either invalid or
3871 unlikely to match any packet::
3873 testpmd> flow create 0 ingress pattern eth / udp / ipv4 / end [...]
3875 More information on these restrictions can be found in the *rte_flow*
3878 Several items support additional specification structures, for example
3879 ``ipv4`` allows specifying source and destination addresses as follows::
3881 testpmd> flow create 0 ingress pattern eth / ipv4 src is 10.1.1.1
3882 dst is 10.2.0.0 / end [...]
3884 This rule matches all IPv4 traffic with the specified properties.
3886 In this example, ``src`` and ``dst`` are field names of the underlying
3887 ``struct rte_flow_item_ipv4`` object. All item properties can be specified
3888 in a similar fashion.
3890 The ``is`` token means that the subsequent value must be matched exactly,
3891 and assigns ``spec`` and ``mask`` fields in ``struct rte_flow_item``
3892 accordingly. Possible assignment tokens are:
3894 - ``is``: match value perfectly (with full bit-mask).
3895 - ``spec``: match value according to configured bit-mask.
3896 - ``last``: specify upper bound to establish a range.
3897 - ``mask``: specify bit-mask with relevant bits set to one.
3898 - ``prefix``: generate bit-mask with <prefix-length> most-significant bits set to one.
3900 These yield identical results::
3902 ipv4 src is 10.1.1.1
3906 ipv4 src spec 10.1.1.1 src mask 255.255.255.255
3910 ipv4 src spec 10.1.1.1 src prefix 32
3914 ipv4 src is 10.1.1.1 src last 10.1.1.1 # range with a single value
3918 ipv4 src is 10.1.1.1 src last 0 # 0 disables range
3920 Inclusive ranges can be defined with ``last``::
3922 ipv4 src is 10.1.1.1 src last 10.2.3.4 # 10.1.1.1 to 10.2.3.4
3924 Note that ``mask`` affects both ``spec`` and ``last``::
3926 ipv4 src is 10.1.1.1 src last 10.2.3.4 src mask 255.255.0.0
3927 # matches 10.1.0.0 to 10.2.255.255
3929 Properties can be modified multiple times::
3931 ipv4 src is 10.1.1.1 src is 10.1.2.3 src is 10.2.3.4 # matches 10.2.3.4
3935 ipv4 src is 10.1.1.1 src prefix 24 src prefix 16 # matches 10.1.0.0/16
3940 This section lists supported pattern items and their attributes, if any.
3942 - ``end``: end list of pattern items.
3944 - ``void``: no-op pattern item.
3946 - ``invert``: perform actions when pattern does not match.
3948 - ``any``: match any protocol for the current layer.
3950 - ``num {unsigned}``: number of layers covered.
3952 - ``pf``: match traffic from/to the physical function.
3954 - ``vf``: match traffic from/to a virtual function ID.
3956 - ``id {unsigned}``: VF ID.
3958 - ``phy_port``: match traffic from/to a specific physical port.
3960 - ``index {unsigned}``: physical port index.
3962 - ``port_id``: match traffic from/to a given DPDK port ID.
3964 - ``id {unsigned}``: DPDK port ID.
3966 - ``mark``: match value set in previously matched flow rule using the mark action.
3968 - ``id {unsigned}``: arbitrary integer value.
3970 - ``raw``: match an arbitrary byte string.
3972 - ``relative {boolean}``: look for pattern after the previous item.
3973 - ``search {boolean}``: search pattern from offset (see also limit).
3974 - ``offset {integer}``: absolute or relative offset for pattern.
3975 - ``limit {unsigned}``: search area limit for start of pattern.
3976 - ``pattern {string}``: byte string to look for.
3978 - ``eth``: match Ethernet header.
3980 - ``dst {MAC-48}``: destination MAC.
3981 - ``src {MAC-48}``: source MAC.
3982 - ``type {unsigned}``: EtherType or TPID.
3984 - ``vlan``: match 802.1Q/ad VLAN tag.
3986 - ``tci {unsigned}``: tag control information.
3987 - ``pcp {unsigned}``: priority code point.
3988 - ``dei {unsigned}``: drop eligible indicator.
3989 - ``vid {unsigned}``: VLAN identifier.
3990 - ``inner_type {unsigned}``: inner EtherType or TPID.
3992 - ``ipv4``: match IPv4 header.
3994 - ``tos {unsigned}``: type of service.
3995 - ``ttl {unsigned}``: time to live.
3996 - ``proto {unsigned}``: next protocol ID.
3997 - ``src {ipv4 address}``: source address.
3998 - ``dst {ipv4 address}``: destination address.
4000 - ``ipv6``: match IPv6 header.
4002 - ``tc {unsigned}``: traffic class.
4003 - ``flow {unsigned}``: flow label.
4004 - ``proto {unsigned}``: protocol (next header).
4005 - ``hop {unsigned}``: hop limit.
4006 - ``src {ipv6 address}``: source address.
4007 - ``dst {ipv6 address}``: destination address.
4009 - ``icmp``: match ICMP header.
4011 - ``type {unsigned}``: ICMP packet type.
4012 - ``code {unsigned}``: ICMP packet code.
4014 - ``udp``: match UDP header.
4016 - ``src {unsigned}``: UDP source port.
4017 - ``dst {unsigned}``: UDP destination port.
4019 - ``tcp``: match TCP header.
4021 - ``src {unsigned}``: TCP source port.
4022 - ``dst {unsigned}``: TCP destination port.
4024 - ``sctp``: match SCTP header.
4026 - ``src {unsigned}``: SCTP source port.
4027 - ``dst {unsigned}``: SCTP destination port.
4028 - ``tag {unsigned}``: validation tag.
4029 - ``cksum {unsigned}``: checksum.
4031 - ``vxlan``: match VXLAN header.
4033 - ``vni {unsigned}``: VXLAN identifier.
4035 - ``e_tag``: match IEEE 802.1BR E-Tag header.
4037 - ``grp_ecid_b {unsigned}``: GRP and E-CID base.
4039 - ``nvgre``: match NVGRE header.
4041 - ``tni {unsigned}``: virtual subnet ID.
4043 - ``mpls``: match MPLS header.
4045 - ``label {unsigned}``: MPLS label.
4047 - ``gre``: match GRE header.
4049 - ``protocol {unsigned}``: protocol type.
4051 - ``gre_key``: match GRE optional key field.
4053 - ``value {unsigned}``: key value.
4055 - ``fuzzy``: fuzzy pattern match, expect faster than default.
4057 - ``thresh {unsigned}``: accuracy threshold.
4059 - ``gtp``, ``gtpc``, ``gtpu``: match GTPv1 header.
4061 - ``teid {unsigned}``: tunnel endpoint identifier.
4063 - ``geneve``: match GENEVE header.
4065 - ``vni {unsigned}``: virtual network identifier.
4066 - ``protocol {unsigned}``: protocol type.
4068 - ``vxlan-gpe``: match VXLAN-GPE header.
4070 - ``vni {unsigned}``: VXLAN-GPE identifier.
4072 - ``arp_eth_ipv4``: match ARP header for Ethernet/IPv4.
4074 - ``sha {MAC-48}``: sender hardware address.
4075 - ``spa {ipv4 address}``: sender IPv4 address.
4076 - ``tha {MAC-48}``: target hardware address.
4077 - ``tpa {ipv4 address}``: target IPv4 address.
4079 - ``ipv6_ext``: match presence of any IPv6 extension header.
4081 - ``next_hdr {unsigned}``: next header.
4083 - ``icmp6``: match any ICMPv6 header.
4085 - ``type {unsigned}``: ICMPv6 type.
4086 - ``code {unsigned}``: ICMPv6 code.
4088 - ``icmp6_nd_ns``: match ICMPv6 neighbor discovery solicitation.
4090 - ``target_addr {ipv6 address}``: target address.
4092 - ``icmp6_nd_na``: match ICMPv6 neighbor discovery advertisement.
4094 - ``target_addr {ipv6 address}``: target address.
4096 - ``icmp6_nd_opt``: match presence of any ICMPv6 neighbor discovery option.
4098 - ``type {unsigned}``: ND option type.
4100 - ``icmp6_nd_opt_sla_eth``: match ICMPv6 neighbor discovery source Ethernet
4101 link-layer address option.
4103 - ``sla {MAC-48}``: source Ethernet LLA.
4105 - ``icmp6_nd_opt_tla_eth``: match ICMPv6 neighbor discovery target Ethernet
4106 link-layer address option.
4108 - ``tla {MAC-48}``: target Ethernet LLA.
4110 - ``meta``: match application specific metadata.
4112 - ``data {unsigned}``: metadata value.
4114 - ``gtp_psc``: match GTP PDU extension header with type 0x85.
4116 - ``pdu_type {unsigned}``: PDU type.
4117 - ``qfi {unsigned}``: QoS flow identifier.
4119 - ``pppoes``, ``pppoed``: match PPPoE header.
4121 - ``session_id {unsigned}``: session identifier.
4123 - ``pppoe_proto_id``: match PPPoE session protocol identifier.
4125 - ``proto_id {unsigned}``: PPP protocol identifier.
4127 - ``l2tpv3oip``: match L2TPv3 over IP header.
4129 - ``session_id {unsigned}``: L2TPv3 over IP session identifier.
4131 - ``ah``: match AH header.
4133 - ``spi {unsigned}``: security parameters index.
4135 - ``pfcp``: match PFCP header.
4137 - ``s_field {unsigned}``: S field.
4138 - ``seid {unsigned}``: session endpoint identifier.
4143 A list of actions starts after the ``actions`` token in the same fashion as
4144 `Matching pattern`_; actions are separated by ``/`` tokens and the list is
4145 terminated by a mandatory ``end`` action.
4147 Actions are named after their type (*RTE_FLOW_ACTION_TYPE_* from ``enum
4148 rte_flow_action_type``).
4150 Dropping all incoming UDPv4 packets can be expressed as follows::
4152 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
4155 Several actions have configurable properties which must be specified when
4156 there is no valid default value. For example, ``queue`` requires a target
4159 This rule redirects incoming UDPv4 traffic to queue index 6::
4161 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
4162 actions queue index 6 / end
4164 While this one could be rejected by PMDs (unspecified queue index)::
4166 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
4169 As defined by *rte_flow*, the list is not ordered, all actions of a given
4170 rule are performed simultaneously. These are equivalent::
4172 queue index 6 / void / mark id 42 / end
4176 void / mark id 42 / queue index 6 / end
4178 All actions in a list should have different types, otherwise only the last
4179 action of a given type is taken into account::
4181 queue index 4 / queue index 5 / queue index 6 / end # will use queue 6
4185 drop / drop / drop / end # drop is performed only once
4189 mark id 42 / queue index 3 / mark id 24 / end # mark will be 24
4191 Considering they are performed simultaneously, opposite and overlapping
4192 actions can sometimes be combined when the end result is unambiguous::
4194 drop / queue index 6 / end # drop has no effect
4198 queue index 6 / rss queues 6 7 8 / end # queue has no effect
4202 drop / passthru / end # drop has no effect
4204 Note that PMDs may still refuse such combinations.
4209 This section lists supported actions and their attributes, if any.
4211 - ``end``: end list of actions.
4213 - ``void``: no-op action.
4215 - ``passthru``: let subsequent rule process matched packets.
4217 - ``jump``: redirect traffic to group on device.
4219 - ``group {unsigned}``: group to redirect to.
4221 - ``mark``: attach 32 bit value to packets.
4223 - ``id {unsigned}``: 32 bit value to return with packets.
4225 - ``flag``: flag packets.
4227 - ``queue``: assign packets to a given queue index.
4229 - ``index {unsigned}``: queue index to use.
4231 - ``drop``: drop packets (note: passthru has priority).
4233 - ``count``: enable counters for this rule.
4235 - ``rss``: spread packets among several queues.
4237 - ``func {hash function}``: RSS hash function to apply, allowed tokens are
4238 the same as `set_hash_global_config`_.
4240 - ``level {unsigned}``: encapsulation level for ``types``.
4242 - ``types [{RSS hash type} [...]] end``: specific RSS hash types, allowed
4243 tokens are the same as `set_hash_input_set`_, except that an empty list
4244 does not disable RSS but instead requests unspecified "best-effort"
4247 - ``key {string}``: RSS hash key, overrides ``key_len``.
4249 - ``key_len {unsigned}``: RSS hash key length in bytes, can be used in
4250 conjunction with ``key`` to pad or truncate it.
4252 - ``queues [{unsigned} [...]] end``: queue indices to use.
4254 - ``pf``: direct traffic to physical function.
4256 - ``vf``: direct traffic to a virtual function ID.
4258 - ``original {boolean}``: use original VF ID if possible.
4259 - ``id {unsigned}``: VF ID.
4261 - ``phy_port``: direct packets to physical port index.
4263 - ``original {boolean}``: use original port index if possible.
4264 - ``index {unsigned}``: physical port index.
4266 - ``port_id``: direct matching traffic to a given DPDK port ID.
4268 - ``original {boolean}``: use original DPDK port ID if possible.
4269 - ``id {unsigned}``: DPDK port ID.
4271 - ``of_set_mpls_ttl``: OpenFlow's ``OFPAT_SET_MPLS_TTL``.
4273 - ``mpls_ttl``: MPLS TTL.
4275 - ``of_dec_mpls_ttl``: OpenFlow's ``OFPAT_DEC_MPLS_TTL``.
4277 - ``of_set_nw_ttl``: OpenFlow's ``OFPAT_SET_NW_TTL``.
4279 - ``nw_ttl``: IP TTL.
4281 - ``of_dec_nw_ttl``: OpenFlow's ``OFPAT_DEC_NW_TTL``.
4283 - ``of_copy_ttl_out``: OpenFlow's ``OFPAT_COPY_TTL_OUT``.
4285 - ``of_copy_ttl_in``: OpenFlow's ``OFPAT_COPY_TTL_IN``.
4287 - ``of_pop_vlan``: OpenFlow's ``OFPAT_POP_VLAN``.
4289 - ``of_push_vlan``: OpenFlow's ``OFPAT_PUSH_VLAN``.
4291 - ``ethertype``: Ethertype.
4293 - ``of_set_vlan_vid``: OpenFlow's ``OFPAT_SET_VLAN_VID``.
4295 - ``vlan_vid``: VLAN id.
4297 - ``of_set_vlan_pcp``: OpenFlow's ``OFPAT_SET_VLAN_PCP``.
4299 - ``vlan_pcp``: VLAN priority.
4301 - ``of_pop_mpls``: OpenFlow's ``OFPAT_POP_MPLS``.
4303 - ``ethertype``: Ethertype.
4305 - ``of_push_mpls``: OpenFlow's ``OFPAT_PUSH_MPLS``.
4307 - ``ethertype``: Ethertype.
4309 - ``vxlan_encap``: Performs a VXLAN encapsulation, outer layer configuration
4310 is done through `Config VXLAN Encap outer layers`_.
4312 - ``vxlan_decap``: Performs a decapsulation action by stripping all headers of
4313 the VXLAN tunnel network overlay from the matched flow.
4315 - ``nvgre_encap``: Performs a NVGRE encapsulation, outer layer configuration
4316 is done through `Config NVGRE Encap outer layers`_.
4318 - ``nvgre_decap``: Performs a decapsulation action by stripping all headers of
4319 the NVGRE tunnel network overlay from the matched flow.
4321 - ``l2_encap``: Performs a L2 encapsulation, L2 configuration
4322 is done through `Config L2 Encap`_.
4324 - ``l2_decap``: Performs a L2 decapsulation, L2 configuration
4325 is done through `Config L2 Decap`_.
4327 - ``mplsogre_encap``: Performs a MPLSoGRE encapsulation, outer layer
4328 configuration is done through `Config MPLSoGRE Encap outer layers`_.
4330 - ``mplsogre_decap``: Performs a MPLSoGRE decapsulation, outer layer
4331 configuration is done through `Config MPLSoGRE Decap outer layers`_.
4333 - ``mplsoudp_encap``: Performs a MPLSoUDP encapsulation, outer layer
4334 configuration is done through `Config MPLSoUDP Encap outer layers`_.
4336 - ``mplsoudp_decap``: Performs a MPLSoUDP decapsulation, outer layer
4337 configuration is done through `Config MPLSoUDP Decap outer layers`_.
4339 - ``set_ipv4_src``: Set a new IPv4 source address in the outermost IPv4 header.
4341 - ``ipv4_addr``: New IPv4 source address.
4343 - ``set_ipv4_dst``: Set a new IPv4 destination address in the outermost IPv4
4346 - ``ipv4_addr``: New IPv4 destination address.
4348 - ``set_ipv6_src``: Set a new IPv6 source address in the outermost IPv6 header.
4350 - ``ipv6_addr``: New IPv6 source address.
4352 - ``set_ipv6_dst``: Set a new IPv6 destination address in the outermost IPv6
4355 - ``ipv6_addr``: New IPv6 destination address.
4357 - ``set_tp_src``: Set a new source port number in the outermost TCP/UDP
4360 - ``port``: New TCP/UDP source port number.
4362 - ``set_tp_dst``: Set a new destination port number in the outermost TCP/UDP
4365 - ``port``: New TCP/UDP destination port number.
4367 - ``mac_swap``: Swap the source and destination MAC addresses in the outermost
4370 - ``dec_ttl``: Performs a decrease TTL value action
4372 - ``set_ttl``: Set TTL value with specified value
4373 - ``ttl_value {unsigned}``: The new TTL value to be set
4375 - ``set_mac_src``: set source MAC address
4377 - ``mac_addr {MAC-48}``: new source MAC address
4379 - ``set_mac_dst``: set destination MAC address
4381 - ``mac_addr {MAC-48}``: new destination MAC address
4383 - ``inc_tcp_seq``: Increase sequence number in the outermost TCP header.
4385 - ``value {unsigned}``: Value to increase TCP sequence number by.
4387 - ``dec_tcp_seq``: Decrease sequence number in the outermost TCP header.
4389 - ``value {unsigned}``: Value to decrease TCP sequence number by.
4391 - ``inc_tcp_ack``: Increase acknowledgment number in the outermost TCP header.
4393 - ``value {unsigned}``: Value to increase TCP acknowledgment number by.
4395 - ``dec_tcp_ack``: Decrease acknowledgment number in the outermost TCP header.
4397 - ``value {unsigned}``: Value to decrease TCP acknowledgment number by.
4399 - ``set_ipv4_dscp``: Set IPv4 DSCP value with specified value
4401 - ``dscp_value {unsigned}``: The new DSCP value to be set
4403 - ``set_ipv6_dscp``: Set IPv6 DSCP value with specified value
4405 - ``dscp_value {unsigned}``: The new DSCP value to be set
4407 Destroying flow rules
4408 ~~~~~~~~~~~~~~~~~~~~~
4410 ``flow destroy`` destroys one or more rules from their rule ID (as returned
4411 by ``flow create``), this command calls ``rte_flow_destroy()`` as many
4412 times as necessary::
4414 flow destroy {port_id} rule {rule_id} [...]
4416 If successful, it will show::
4418 Flow rule #[...] destroyed
4420 It does not report anything for rule IDs that do not exist. The usual error
4421 message is shown when a rule cannot be destroyed::
4423 Caught error type [...] ([...]): [...]
4425 ``flow flush`` destroys all rules on a device and does not take extra
4426 arguments. It is bound to ``rte_flow_flush()``::
4428 flow flush {port_id}
4430 Any errors are reported as above.
4432 Creating several rules and destroying them::
4434 testpmd> flow create 0 ingress pattern eth / ipv6 / end
4435 actions queue index 2 / end
4436 Flow rule #0 created
4437 testpmd> flow create 0 ingress pattern eth / ipv4 / end
4438 actions queue index 3 / end
4439 Flow rule #1 created
4440 testpmd> flow destroy 0 rule 0 rule 1
4441 Flow rule #1 destroyed
4442 Flow rule #0 destroyed
4445 The same result can be achieved using ``flow flush``::
4447 testpmd> flow create 0 ingress pattern eth / ipv6 / end
4448 actions queue index 2 / end
4449 Flow rule #0 created
4450 testpmd> flow create 0 ingress pattern eth / ipv4 / end
4451 actions queue index 3 / end
4452 Flow rule #1 created
4453 testpmd> flow flush 0
4456 Non-existent rule IDs are ignored::
4458 testpmd> flow create 0 ingress pattern eth / ipv6 / end
4459 actions queue index 2 / end
4460 Flow rule #0 created
4461 testpmd> flow create 0 ingress pattern eth / ipv4 / end
4462 actions queue index 3 / end
4463 Flow rule #1 created
4464 testpmd> flow destroy 0 rule 42 rule 10 rule 2
4466 testpmd> flow destroy 0 rule 0
4467 Flow rule #0 destroyed
4473 ``flow query`` queries a specific action of a flow rule having that
4474 ability. Such actions collect information that can be reported using this
4475 command. It is bound to ``rte_flow_query()``::
4477 flow query {port_id} {rule_id} {action}
4479 If successful, it will display either the retrieved data for known actions
4480 or the following message::
4482 Cannot display result for action type [...] ([...])
4484 Otherwise, it will complain either that the rule does not exist or that some
4487 Flow rule #[...] not found
4491 Caught error type [...] ([...]): [...]
4493 Currently only the ``count`` action is supported. This action reports the
4494 number of packets that hit the flow rule and the total number of bytes. Its
4495 output has the following format::
4498 hits_set: [...] # whether "hits" contains a valid value
4499 bytes_set: [...] # whether "bytes" contains a valid value
4500 hits: [...] # number of packets
4501 bytes: [...] # number of bytes
4503 Querying counters for TCPv6 packets redirected to queue 6::
4505 testpmd> flow create 0 ingress pattern eth / ipv6 / tcp / end
4506 actions queue index 6 / count / end
4507 Flow rule #4 created
4508 testpmd> flow query 0 4 count
4519 ``flow list`` lists existing flow rules sorted by priority and optionally
4520 filtered by group identifiers::
4522 flow list {port_id} [group {group_id}] [...]
4524 This command only fails with the following message if the device does not
4529 Output consists of a header line followed by a short description of each
4530 flow rule, one per line. There is no output at all when no flow rules are
4531 configured on the device::
4533 ID Group Prio Attr Rule
4534 [...] [...] [...] [...] [...]
4536 ``Attr`` column flags:
4538 - ``i`` for ``ingress``.
4539 - ``e`` for ``egress``.
4541 Creating several flow rules and listing them::
4543 testpmd> flow create 0 ingress pattern eth / ipv4 / end
4544 actions queue index 6 / end
4545 Flow rule #0 created
4546 testpmd> flow create 0 ingress pattern eth / ipv6 / end
4547 actions queue index 2 / end
4548 Flow rule #1 created
4549 testpmd> flow create 0 priority 5 ingress pattern eth / ipv4 / udp / end
4550 actions rss queues 6 7 8 end / end
4551 Flow rule #2 created
4552 testpmd> flow list 0
4553 ID Group Prio Attr Rule
4554 0 0 0 i- ETH IPV4 => QUEUE
4555 1 0 0 i- ETH IPV6 => QUEUE
4556 2 0 5 i- ETH IPV4 UDP => RSS
4559 Rules are sorted by priority (i.e. group ID first, then priority level)::
4561 testpmd> flow list 1
4562 ID Group Prio Attr Rule
4563 0 0 0 i- ETH => COUNT
4564 6 0 500 i- ETH IPV6 TCP => DROP COUNT
4565 5 0 1000 i- ETH IPV6 ICMP => QUEUE
4566 1 24 0 i- ETH IPV4 UDP => QUEUE
4567 4 24 10 i- ETH IPV4 TCP => DROP
4568 3 24 20 i- ETH IPV4 => DROP
4569 2 24 42 i- ETH IPV4 UDP => QUEUE
4570 7 63 0 i- ETH IPV6 UDP VXLAN => MARK QUEUE
4573 Output can be limited to specific groups::
4575 testpmd> flow list 1 group 0 group 63
4576 ID Group Prio Attr Rule
4577 0 0 0 i- ETH => COUNT
4578 6 0 500 i- ETH IPV6 TCP => DROP COUNT
4579 5 0 1000 i- ETH IPV6 ICMP => QUEUE
4580 7 63 0 i- ETH IPV6 UDP VXLAN => MARK QUEUE
4583 Toggling isolated mode
4584 ~~~~~~~~~~~~~~~~~~~~~~
4586 ``flow isolate`` can be used to tell the underlying PMD that ingress traffic
4587 must only be injected from the defined flow rules; that no default traffic
4588 is expected outside those rules and the driver is free to assign more
4589 resources to handle them. It is bound to ``rte_flow_isolate()``::
4591 flow isolate {port_id} {boolean}
4593 If successful, enabling or disabling isolated mode shows either::
4595 Ingress traffic on port [...]
4596 is now restricted to the defined flow rules
4600 Ingress traffic on port [...]
4601 is not restricted anymore to the defined flow rules
4603 Otherwise, in case of error::
4605 Caught error type [...] ([...]): [...]
4607 Mainly due to its side effects, PMDs supporting this mode may not have the
4608 ability to toggle it more than once without reinitializing affected ports
4609 first (e.g. by exiting testpmd).
4611 Enabling isolated mode::
4613 testpmd> flow isolate 0 true
4614 Ingress traffic on port 0 is now restricted to the defined flow rules
4617 Disabling isolated mode::
4619 testpmd> flow isolate 0 false
4620 Ingress traffic on port 0 is not restricted anymore to the defined flow rules
4623 Dumping HW internal information
4624 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4626 ``flow dump`` dumps the hardware's internal representation information of
4627 all flows. It is bound to ``rte_flow_dev_dump()``::
4629 flow dump {port_id} {output_file}
4631 If successful, it will show::
4635 Otherwise, it will complain error occurred::
4637 Caught error type [...] ([...]): [...]
4639 Listing and destroying aged flow rules
4640 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4642 ``flow aged`` simply lists aged flow rules be get from api ``rte_flow_get_aged_flows``,
4643 and ``destroy`` parameter can be used to destroy those flow rules in PMD.
4645 flow aged {port_id} [destroy]
4647 Listing current aged flow rules::
4649 testpmd> flow aged 0
4650 Port 0 total aged flows: 0
4651 testpmd> flow create 0 ingress pattern eth / ipv4 src is 2.2.2.14 / end
4652 actions age timeout 5 / queue index 0 / end
4653 Flow rule #0 created
4654 testpmd> flow create 0 ingress pattern eth / ipv4 src is 2.2.2.15 / end
4655 actions age timeout 4 / queue index 0 / end
4656 Flow rule #1 created
4657 testpmd> flow create 0 ingress pattern eth / ipv4 src is 2.2.2.16 / end
4658 actions age timeout 2 / queue index 0 / end
4659 Flow rule #2 created
4660 testpmd> flow create 0 ingress pattern eth / ipv4 src is 2.2.2.17 / end
4661 actions age timeout 3 / queue index 0 / end
4662 Flow rule #3 created
4665 Aged Rules are simply list as command ``flow list {port_id}``, but strip the detail rule
4666 information, all the aged flows are sorted by the longest timeout time. For example, if
4667 those rules be configured in the same time, ID 2 will be the first aged out rule, the next
4668 will be ID 3, ID 1, ID 0::
4670 testpmd> flow aged 0
4671 Port 0 total aged flows: 4
4678 If attach ``destroy`` parameter, the command will destroy all the list aged flow rules.
4680 testpmd> flow aged 0 destroy
4681 Port 0 total aged flows: 4
4688 Flow rule #2 destroyed
4689 Flow rule #3 destroyed
4690 Flow rule #1 destroyed
4691 Flow rule #0 destroyed
4692 4 flows be destroyed
4693 testpmd> flow aged 0
4694 Port 0 total aged flows: 0
4697 Sample QinQ flow rules
4698 ~~~~~~~~~~~~~~~~~~~~~~
4700 Before creating QinQ rule(s) the following commands should be issued to enable QinQ::
4702 testpmd> port stop 0
4703 testpmd> vlan set qinq_strip on 0
4705 The above command sets the inner and outer TPID's to 0x8100.
4707 To change the TPID's the following commands should be used::
4709 testpmd> vlan set outer tpid 0xa100 0
4710 testpmd> vlan set inner tpid 0x9100 0
4711 testpmd> port start 0
4713 Validate and create a QinQ rule on port 0 to steer traffic to a VF queue in a VM.
4717 testpmd> flow validate 0 ingress pattern eth / vlan tci is 123 /
4718 vlan tci is 456 / end actions vf id 1 / queue index 0 / end
4719 Flow rule #0 validated
4721 testpmd> flow create 0 ingress pattern eth / vlan tci is 4 /
4722 vlan tci is 456 / end actions vf id 123 / queue index 0 / end
4723 Flow rule #0 created
4725 testpmd> flow list 0
4726 ID Group Prio Attr Rule
4727 0 0 0 i- ETH VLAN VLAN=>VF QUEUE
4729 Validate and create a QinQ rule on port 0 to steer traffic to a queue on the host.
4733 testpmd> flow validate 0 ingress pattern eth / vlan tci is 321 /
4734 vlan tci is 654 / end actions pf / queue index 0 / end
4735 Flow rule #1 validated
4737 testpmd> flow create 0 ingress pattern eth / vlan tci is 321 /
4738 vlan tci is 654 / end actions pf / queue index 1 / end
4739 Flow rule #1 created
4741 testpmd> flow list 0
4742 ID Group Prio Attr Rule
4743 0 0 0 i- ETH VLAN VLAN=>VF QUEUE
4744 1 0 0 i- ETH VLAN VLAN=>PF QUEUE
4746 Sample VXLAN encapsulation rule
4747 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4749 VXLAN encapsulation outer layer has default value pre-configured in testpmd
4750 source code, those can be changed by using the following commands
4752 IPv4 VXLAN outer header::
4754 testpmd> set vxlan ip-version ipv4 vni 4 udp-src 4 udp-dst 4 ip-src 127.0.0.1
4755 ip-dst 128.0.0.1 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4756 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4759 testpmd> set vxlan-with-vlan ip-version ipv4 vni 4 udp-src 4 udp-dst 4 ip-src
4760 127.0.0.1 ip-dst 128.0.0.1 vlan-tci 34 eth-src 11:11:11:11:11:11
4761 eth-dst 22:22:22:22:22:22
4762 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4765 testpmd> set vxlan-tos-ttl ip-version ipv4 vni 4 udp-src 4 udp-dst 4 ip-tos 0
4766 ip-ttl 255 ip-src 127.0.0.1 ip-dst 128.0.0.1 eth-src 11:11:11:11:11:11
4767 eth-dst 22:22:22:22:22:22
4768 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4771 IPv6 VXLAN outer header::
4773 testpmd> set vxlan ip-version ipv6 vni 4 udp-src 4 udp-dst 4 ip-src ::1
4774 ip-dst ::2222 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4775 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4778 testpmd> set vxlan-with-vlan ip-version ipv6 vni 4 udp-src 4 udp-dst 4
4779 ip-src ::1 ip-dst ::2222 vlan-tci 34 eth-src 11:11:11:11:11:11
4780 eth-dst 22:22:22:22:22:22
4781 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4784 testpmd> set vxlan-tos-ttl ip-version ipv6 vni 4 udp-src 4 udp-dst 4
4785 ip-tos 0 ip-ttl 255 ::1 ip-dst ::2222 eth-src 11:11:11:11:11:11
4786 eth-dst 22:22:22:22:22:22
4787 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4790 Sample NVGRE encapsulation rule
4791 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4793 NVGRE encapsulation outer layer has default value pre-configured in testpmd
4794 source code, those can be changed by using the following commands
4796 IPv4 NVGRE outer header::
4798 testpmd> set nvgre ip-version ipv4 tni 4 ip-src 127.0.0.1 ip-dst 128.0.0.1
4799 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4800 testpmd> flow create 0 ingress pattern end actions nvgre_encap /
4803 testpmd> set nvgre-with-vlan ip-version ipv4 tni 4 ip-src 127.0.0.1
4804 ip-dst 128.0.0.1 vlan-tci 34 eth-src 11:11:11:11:11:11
4805 eth-dst 22:22:22:22:22:22
4806 testpmd> flow create 0 ingress pattern end actions nvgre_encap /
4809 IPv6 NVGRE outer header::
4811 testpmd> set nvgre ip-version ipv6 tni 4 ip-src ::1 ip-dst ::2222
4812 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4813 testpmd> flow create 0 ingress pattern end actions nvgre_encap /
4816 testpmd> set nvgre-with-vlan ip-version ipv6 tni 4 ip-src ::1 ip-dst ::2222
4817 vlan-tci 34 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4818 testpmd> flow create 0 ingress pattern end actions nvgre_encap /
4821 Sample L2 encapsulation rule
4822 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4824 L2 encapsulation has default value pre-configured in testpmd
4825 source code, those can be changed by using the following commands
4829 testpmd> set l2_encap ip-version ipv4
4830 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4831 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / mpls / end actions
4832 mplsoudp_decap / l2_encap / end
4834 L2 with VXLAN header::
4836 testpmd> set l2_encap-with-vlan ip-version ipv4 vlan-tci 34
4837 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4838 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / mpls / end actions
4839 mplsoudp_decap / l2_encap / end
4841 Sample L2 decapsulation rule
4842 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4844 L2 decapsulation has default value pre-configured in testpmd
4845 source code, those can be changed by using the following commands
4849 testpmd> set l2_decap
4850 testpmd> flow create 0 egress pattern eth / end actions l2_decap / mplsoudp_encap /
4853 L2 with VXLAN header::
4855 testpmd> set l2_encap-with-vlan
4856 testpmd> flow create 0 egress pattern eth / end actions l2_encap / mplsoudp_encap /
4859 Sample MPLSoGRE encapsulation rule
4860 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4862 MPLSoGRE encapsulation outer layer has default value pre-configured in testpmd
4863 source code, those can be changed by using the following commands
4865 IPv4 MPLSoGRE outer header::
4867 testpmd> set mplsogre_encap ip-version ipv4 label 4
4868 ip-src 127.0.0.1 ip-dst 128.0.0.1 eth-src 11:11:11:11:11:11
4869 eth-dst 22:22:22:22:22:22
4870 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4871 mplsogre_encap / end
4873 IPv4 MPLSoGRE with VLAN outer header::
4875 testpmd> set mplsogre_encap-with-vlan ip-version ipv4 label 4
4876 ip-src 127.0.0.1 ip-dst 128.0.0.1 vlan-tci 34
4877 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4878 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4879 mplsogre_encap / end
4881 IPv6 MPLSoGRE outer header::
4883 testpmd> set mplsogre_encap ip-version ipv6 mask 4
4884 ip-src ::1 ip-dst ::2222 eth-src 11:11:11:11:11:11
4885 eth-dst 22:22:22:22:22:22
4886 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4887 mplsogre_encap / end
4889 IPv6 MPLSoGRE with VLAN outer header::
4891 testpmd> set mplsogre_encap-with-vlan ip-version ipv6 mask 4
4892 ip-src ::1 ip-dst ::2222 vlan-tci 34
4893 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4894 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4895 mplsogre_encap / end
4897 Sample MPLSoGRE decapsulation rule
4898 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4900 MPLSoGRE decapsulation outer layer has default value pre-configured in testpmd
4901 source code, those can be changed by using the following commands
4903 IPv4 MPLSoGRE outer header::
4905 testpmd> set mplsogre_decap ip-version ipv4
4906 testpmd> flow create 0 ingress pattern eth / ipv4 / gre / mpls / end actions
4907 mplsogre_decap / l2_encap / end
4909 IPv4 MPLSoGRE with VLAN outer header::
4911 testpmd> set mplsogre_decap-with-vlan ip-version ipv4
4912 testpmd> flow create 0 ingress pattern eth / vlan / ipv4 / gre / mpls / end
4913 actions mplsogre_decap / l2_encap / end
4915 IPv6 MPLSoGRE outer header::
4917 testpmd> set mplsogre_decap ip-version ipv6
4918 testpmd> flow create 0 ingress pattern eth / ipv6 / gre / mpls / end
4919 actions mplsogre_decap / l2_encap / end
4921 IPv6 MPLSoGRE with VLAN outer header::
4923 testpmd> set mplsogre_decap-with-vlan ip-version ipv6
4924 testpmd> flow create 0 ingress pattern eth / vlan / ipv6 / gre / mpls / end
4925 actions mplsogre_decap / l2_encap / end
4927 Sample MPLSoUDP encapsulation rule
4928 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4930 MPLSoUDP encapsulation outer layer has default value pre-configured in testpmd
4931 source code, those can be changed by using the following commands
4933 IPv4 MPLSoUDP outer header::
4935 testpmd> set mplsoudp_encap ip-version ipv4 label 4 udp-src 5 udp-dst 10
4936 ip-src 127.0.0.1 ip-dst 128.0.0.1 eth-src 11:11:11:11:11:11
4937 eth-dst 22:22:22:22:22:22
4938 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4939 mplsoudp_encap / end
4941 IPv4 MPLSoUDP with VLAN outer header::
4943 testpmd> set mplsoudp_encap-with-vlan ip-version ipv4 label 4 udp-src 5
4944 udp-dst 10 ip-src 127.0.0.1 ip-dst 128.0.0.1 vlan-tci 34
4945 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4946 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4947 mplsoudp_encap / end
4949 IPv6 MPLSoUDP outer header::
4951 testpmd> set mplsoudp_encap ip-version ipv6 mask 4 udp-src 5 udp-dst 10
4952 ip-src ::1 ip-dst ::2222 eth-src 11:11:11:11:11:11
4953 eth-dst 22:22:22:22:22:22
4954 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4955 mplsoudp_encap / end
4957 IPv6 MPLSoUDP with VLAN outer header::
4959 testpmd> set mplsoudp_encap-with-vlan ip-version ipv6 mask 4 udp-src 5
4960 udp-dst 10 ip-src ::1 ip-dst ::2222 vlan-tci 34
4961 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4962 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4963 mplsoudp_encap / end
4965 Sample MPLSoUDP decapsulation rule
4966 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4968 MPLSoUDP decapsulation outer layer has default value pre-configured in testpmd
4969 source code, those can be changed by using the following commands
4971 IPv4 MPLSoUDP outer header::
4973 testpmd> set mplsoudp_decap ip-version ipv4
4974 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / mpls / end actions
4975 mplsoudp_decap / l2_encap / end
4977 IPv4 MPLSoUDP with VLAN outer header::
4979 testpmd> set mplsoudp_decap-with-vlan ip-version ipv4
4980 testpmd> flow create 0 ingress pattern eth / vlan / ipv4 / udp / mpls / end
4981 actions mplsoudp_decap / l2_encap / end
4983 IPv6 MPLSoUDP outer header::
4985 testpmd> set mplsoudp_decap ip-version ipv6
4986 testpmd> flow create 0 ingress pattern eth / ipv6 / udp / mpls / end
4987 actions mplsoudp_decap / l2_encap / end
4989 IPv6 MPLSoUDP with VLAN outer header::
4991 testpmd> set mplsoudp_decap-with-vlan ip-version ipv6
4992 testpmd> flow create 0 ingress pattern eth / vlan / ipv6 / udp / mpls / end
4993 actions mplsoudp_decap / l2_encap / end
4995 Sample Raw encapsulation rule
4996 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4998 Raw encapsulation configuration can be set by the following commands
5000 Eecapsulating VxLAN::
5002 testpmd> set raw_encap 4 eth src is 10:11:22:33:44:55 / vlan tci is 1
5003 inner_type is 0x0800 / ipv4 / udp dst is 4789 / vxlan vni
5005 testpmd> flow create 0 egress pattern eth / ipv4 / end actions
5006 raw_encap index 4 / end
5008 Sample Raw decapsulation rule
5009 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
5011 Raw decapsulation configuration can be set by the following commands
5013 Decapsulating VxLAN::
5015 testpmd> set raw_decap eth / ipv4 / udp / vxlan / end_set
5016 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / vxlan / eth / ipv4 /
5017 end actions raw_decap / queue index 0 / end
5022 ESP rules can be created by the following commands::
5024 testpmd> flow create 0 ingress pattern eth / ipv4 / esp spi is 1 / end actions
5026 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / esp spi is 1 / end
5027 actions queue index 3 / end
5028 testpmd> flow create 0 ingress pattern eth / ipv6 / esp spi is 1 / end actions
5030 testpmd> flow create 0 ingress pattern eth / ipv6 / udp / esp spi is 1 / end
5031 actions queue index 3 / end
5036 AH rules can be created by the following commands::
5038 testpmd> flow create 0 ingress pattern eth / ipv4 / ah spi is 1 / end actions
5040 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / ah spi is 1 / end
5041 actions queue index 3 / end
5042 testpmd> flow create 0 ingress pattern eth / ipv6 / ah spi is 1 / end actions
5044 testpmd> flow create 0 ingress pattern eth / ipv6 / udp / ah spi is 1 / end
5045 actions queue index 3 / end
5050 PFCP rules can be created by the following commands(s_field need to be 1
5053 testpmd> flow create 0 ingress pattern eth / ipv4 / pfcp s_field is 0 / end
5054 actions queue index 3 / end
5055 testpmd> flow create 0 ingress pattern eth / ipv4 / pfcp s_field is 1
5056 seid is 1 / end actions queue index 3 / end
5057 testpmd> flow create 0 ingress pattern eth / ipv6 / pfcp s_field is 0 / end
5058 actions queue index 3 / end
5059 testpmd> flow create 0 ingress pattern eth / ipv6 / pfcp s_field is 1
5060 seid is 1 / end actions queue index 3 / end
5065 The following sections show functions to load/unload eBPF based filters.
5070 Load an eBPF program as a callback for particular RX/TX queue::
5072 testpmd> bpf-load rx|tx (portid) (queueid) (load-flags) (bpf-prog-filename)
5074 The available load-flags are:
5076 * ``J``: use JIT generated native code, otherwise BPF interpreter will be used.
5078 * ``M``: assume input parameter is a pointer to rte_mbuf, otherwise assume it is a pointer to first segment's data.
5084 You'll need clang v3.7 or above to build bpf program you'd like to load
5088 .. code-block:: console
5091 clang -O2 -target bpf -c t1.c
5093 Then to load (and JIT compile) t1.o at RX queue 0, port 1:
5095 .. code-block:: console
5097 testpmd> bpf-load rx 1 0 J ./dpdk.org/examples/bpf/t1.o
5099 To load (not JITed) t1.o at TX queue 0, port 0:
5101 .. code-block:: console
5103 testpmd> bpf-load tx 0 0 - ./dpdk.org/examples/bpf/t1.o
5108 Unload previously loaded eBPF program for particular RX/TX queue::
5110 testpmd> bpf-unload rx|tx (portid) (queueid)
5112 For example to unload BPF filter from TX queue 0, port 0:
5114 .. code-block:: console
5116 testpmd> bpf-unload tx 0 0