1 .. SPDX-License-Identifier: BSD-3-Clause
2 Copyright(c) 2010-2016 Intel Corporation.
6 Testpmd Runtime Functions
7 =========================
9 Where the testpmd application is started in interactive mode, (``-i|--interactive``),
10 it displays a prompt that can be used to start and stop forwarding,
11 configure the application, display statistics (including the extended NIC
12 statistics aka xstats) , set the Flow Director and other tasks::
16 The testpmd prompt has some, limited, readline support.
17 Common bash command-line functions such as ``Ctrl+a`` and ``Ctrl+e`` to go to the start and end of the prompt line are supported
18 as well as access to the command history via the up-arrow.
20 There is also support for tab completion.
21 If you type a partial command and hit ``<TAB>`` you get a list of the available completions:
23 .. code-block:: console
25 testpmd> show port <TAB>
27 info [Mul-choice STRING]: show|clear port info|stats|xstats|fdir|stat_qmap|dcb_tc|cap X
28 info [Mul-choice STRING]: show|clear port info|stats|xstats|fdir|stat_qmap|dcb_tc|cap all
29 stats [Mul-choice STRING]: show|clear port info|stats|xstats|fdir|stat_qmap|dcb_tc|cap X
30 stats [Mul-choice STRING]: show|clear port info|stats|xstats|fdir|stat_qmap|dcb_tc|cap all
36 Some examples in this document are too long to fit on one line are shown wrapped at `"\\"` for display purposes::
38 testpmd> set flow_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \
39 (pause_time) (send_xon) (port_id)
41 In the real ``testpmd>`` prompt these commands should be on a single line.
46 The testpmd has on-line help for the functions that are available at runtime.
47 These are divided into sections and can be accessed using help, help section or help all:
49 .. code-block:: console
53 help control : Start and stop forwarding.
54 help display : Displaying port, stats and config information.
55 help config : Configuration information.
56 help ports : Configuring ports.
57 help registers : Reading and setting port registers.
58 help filters : Filters configuration help.
59 help all : All of the above sections.
62 Command File Functions
63 ----------------------
65 To facilitate loading large number of commands or to avoid cutting and pasting where not
66 practical or possible testpmd supports alternative methods for executing commands.
68 * If started with the ``--cmdline-file=FILENAME`` command line argument testpmd
69 will execute all CLI commands contained within the file immediately before
70 starting packet forwarding or entering interactive mode.
72 .. code-block:: console
74 ./dpdk-testpmd -n4 -r2 ... -- -i --cmdline-file=/home/ubuntu/flow-create-commands.txt
75 Interactive-mode selected
76 CLI commands to be read from /home/ubuntu/flow-create-commands.txt
77 Configuring Port 0 (socket 0)
78 Port 0: 7C:FE:90:CB:74:CE
79 Configuring Port 1 (socket 0)
80 Port 1: 7C:FE:90:CB:74:CA
81 Checking link statuses...
82 Port 0 Link Up - speed 10000 Mbps - full-duplex
83 Port 1 Link Up - speed 10000 Mbps - full-duplex
89 Flow rule #498 created
90 Flow rule #499 created
91 Read all CLI commands from /home/ubuntu/flow-create-commands.txt
95 * At run-time additional commands can be loaded in bulk by invoking the ``load FILENAME``
98 .. code-block:: console
100 testpmd> load /home/ubuntu/flow-create-commands.txt
105 Flow rule #498 created
106 Flow rule #499 created
107 Read all CLI commands from /home/ubuntu/flow-create-commands.txt
111 In all cases output from any included command will be displayed as standard output.
112 Execution will continue until the end of the file is reached regardless of
113 whether any errors occur. The end user must examine the output to determine if
114 any failures occurred.
123 Start packet forwarding with current configuration::
130 Start packet forwarding with current configuration after sending specified number of bursts of packets::
132 testpmd> start tx_first (""|burst_num)
134 The default burst number is 1 when ``burst_num`` not presented.
139 Stop packet forwarding, and display accumulated statistics::
154 The functions in the following sections are used to display information about the
155 testpmd configuration or the NIC status.
160 Display information for a given port or all ports::
162 testpmd> show port (info|summary|stats|xstats|fdir|stat_qmap|dcb_tc|cap) (port_id|all)
164 The available information categories are:
166 * ``info``: General port information such as MAC address.
168 * ``summary``: Brief port summary such as Device Name, Driver Name etc.
170 * ``stats``: RX/TX statistics.
172 * ``xstats``: RX/TX extended NIC statistics.
174 * ``fdir``: Flow Director information and statistics.
176 * ``stat_qmap``: Queue statistics mapping.
178 * ``dcb_tc``: DCB information such as TC mapping.
180 * ``cap``: Supported offload capabilities.
184 .. code-block:: console
186 testpmd> show port info 0
188 ********************* Infos for port 0 *********************
190 MAC address: XX:XX:XX:XX:XX:XX
192 memory allocation on the socket: 0
194 Link speed: 40000 Mbps
195 Link duplex: full-duplex
196 Promiscuous mode: enabled
197 Allmulticast mode: disabled
198 Maximum number of MAC addresses: 64
199 Maximum number of MAC addresses of hash filtering: 0
201 strip on, filter on, extend off, qinq strip off
202 Redirection table size: 512
203 Supported flow types:
221 show port (module_eeprom|eeprom)
222 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
224 Display the EEPROM information of a port::
226 testpmd> show port (port_id) (module_eeprom|eeprom)
231 Display the rss redirection table entry indicated by masks on port X::
233 testpmd> show port (port_id) rss reta (size) (mask0, mask1...)
235 size is used to indicate the hardware supported reta size
240 Display the RSS hash functions and RSS hash key of a port::
242 testpmd> show port (port_id) rss-hash [key]
247 Clear the port statistics and forward engine statistics for a given port or for all ports::
249 testpmd> clear port (info|stats|xstats|fdir|stat_qmap) (port_id|all)
253 testpmd> clear port stats all
258 Display information for a given port's RX/TX queue::
260 testpmd> show (rxq|txq) info (port_id) (queue_id)
262 show desc status(rxq|txq)
263 ~~~~~~~~~~~~~~~~~~~~~~~~~
265 Display information for a given port's RX/TX descriptor status::
267 testpmd> show port (port_id) (rxq|txq) (queue_id) desc (desc_id) status
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|rxoffs|rxpkts|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 * ``rxoffs``: Packet offsets for RX split.
288 * ``rxpkts``: Packets to RX split configuration.
290 * ``txpkts``: Packets to TX configuration.
292 * ``txtimes``: Burst time pattern for Tx only mode.
296 .. code-block:: console
298 testpmd> show config rxtx
300 io packet forwarding - CRC stripping disabled - packets/burst=16
301 nb forwarding cores=2 - nb forwarding ports=1
302 RX queues=1 - RX desc=128 - RX free threshold=0
303 RX threshold registers: pthresh=8 hthresh=8 wthresh=4
304 TX queues=1 - TX desc=512 - TX free threshold=0
305 TX threshold registers: pthresh=36 hthresh=0 wthresh=0
306 TX RS bit threshold=0 - TXQ flags=0x0
311 Set the packet forwarding mode::
313 testpmd> set fwd (io|mac|macswap|flowgen| \
314 rxonly|txonly|csum|icmpecho|noisy|5tswap) (""|retry)
316 ``retry`` can be specified for forwarding engines except ``rx_only``.
318 The available information categories are:
320 * ``io``: Forwards packets "as-is" in I/O mode.
321 This is the fastest possible forwarding operation as it does not access packets data.
322 This is the default mode.
324 * ``mac``: Changes the source and the destination Ethernet addresses of packets before forwarding them.
325 Default application behavior is to set source Ethernet address to that of the transmitting interface, and destination
326 address to a dummy value (set during init). The user may specify a target destination Ethernet address via the 'eth-peer' or
327 'eth-peers-configfile' command-line options. It is not currently possible to specify a specific source Ethernet address.
329 * ``macswap``: MAC swap forwarding mode.
330 Swaps the source and the destination Ethernet addresses of packets before forwarding them.
332 * ``flowgen``: Multi-flow generation mode.
333 Originates a number of flows (with varying destination IP addresses), and terminate receive traffic.
335 * ``rxonly``: Receives packets but doesn't transmit them.
337 * ``txonly``: Generates and transmits packets without receiving any.
339 * ``csum``: Changes the checksum field with hardware or software methods depending on the offload flags on the packet.
341 * ``icmpecho``: Receives a burst of packets, lookup for ICMP echo requests and, if any, send back ICMP echo replies.
343 * ``ieee1588``: Demonstrate L2 IEEE1588 V2 PTP timestamping for RX and TX.
345 * ``noisy``: Noisy neighbor simulation.
346 Simulate more realistic behavior of a guest machine engaged in receiving
347 and sending packets performing Virtual Network Function (VNF).
349 * ``5tswap``: Swap the source and destination of L2,L3,L4 if they exist.
351 L2 swaps the source address and destination address of Ethernet, as same as ``macswap``.
353 L3 swaps the source address and destination address of IP (v4 and v6).
355 L4 swaps the source port and destination port of transport layer (TCP and UDP).
359 testpmd> set fwd rxonly
361 Set rxonly packet forwarding mode
367 When running, forwarding engines maintain statistics from the time they have been started.
368 Example for the io forwarding engine, with some packet drops on the tx side::
370 testpmd> show fwd stats all
372 ------- Forward Stats for RX Port= 0/Queue= 0 -> TX Port= 1/Queue= 0 -------
373 RX-packets: 274293770 TX-packets: 274293642 TX-dropped: 128
375 ------- Forward Stats for RX Port= 1/Queue= 0 -> TX Port= 0/Queue= 0 -------
376 RX-packets: 274301850 TX-packets: 274301850 TX-dropped: 0
378 ---------------------- Forward statistics for port 0 ----------------------
379 RX-packets: 274293802 RX-dropped: 0 RX-total: 274293802
380 TX-packets: 274301862 TX-dropped: 0 TX-total: 274301862
381 ----------------------------------------------------------------------------
383 ---------------------- Forward statistics for port 1 ----------------------
384 RX-packets: 274301894 RX-dropped: 0 RX-total: 274301894
385 TX-packets: 274293706 TX-dropped: 128 TX-total: 274293834
386 ----------------------------------------------------------------------------
388 +++++++++++++++ Accumulated forward statistics for all ports+++++++++++++++
389 RX-packets: 548595696 RX-dropped: 0 RX-total: 548595696
390 TX-packets: 548595568 TX-dropped: 128 TX-total: 548595696
391 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
397 Clear the forwarding engines statistics::
399 testpmd> clear fwd stats all
404 Display an RX descriptor for a port RX queue::
406 testpmd> read rxd (port_id) (queue_id) (rxd_id)
410 testpmd> read rxd 0 0 4
411 0x0000000B - 0x001D0180 / 0x0000000B - 0x001D0180
416 Display a TX descriptor for a port TX queue::
418 testpmd> read txd (port_id) (queue_id) (txd_id)
422 testpmd> read txd 0 0 4
423 0x00000001 - 0x24C3C440 / 0x000F0000 - 0x2330003C
428 Get loaded dynamic device personalization (DDP) package info list::
430 testpmd> ddp get list (port_id)
435 Display information about dynamic device personalization (DDP) profile::
437 testpmd> ddp get info (profile_path)
442 Display VF statistics::
444 testpmd> show vf stats (port_id) (vf_id)
449 Reset VF statistics::
451 testpmd> clear vf stats (port_id) (vf_id)
453 show port pctype mapping
454 ~~~~~~~~~~~~~~~~~~~~~~~~
456 List all items from the pctype mapping table::
458 testpmd> show port (port_id) pctype mapping
460 show rx offloading capabilities
461 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
463 List all per queue and per port Rx offloading capabilities of a port::
465 testpmd> show port (port_id) rx_offload capabilities
467 show rx offloading configuration
468 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
470 List port level and all queue level Rx offloading configuration::
472 testpmd> show port (port_id) rx_offload configuration
474 show tx offloading capabilities
475 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
477 List all per queue and per port Tx offloading capabilities of a port::
479 testpmd> show port (port_id) tx_offload capabilities
481 show tx offloading configuration
482 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
484 List port level and all queue level Tx offloading configuration::
486 testpmd> show port (port_id) tx_offload configuration
488 show tx metadata setting
489 ~~~~~~~~~~~~~~~~~~~~~~~~
491 Show Tx metadata value set for a specific port::
493 testpmd> show port (port_id) tx_metadata
495 show port supported ptypes
496 ~~~~~~~~~~~~~~~~~~~~~~~~~~
498 Show ptypes supported for a specific port::
500 testpmd> show port (port_id) ptypes
502 set port supported ptypes
503 ~~~~~~~~~~~~~~~~~~~~~~~~~
505 set packet types classification for a specific port::
507 testpmd> set port (port_id) ptypes_mask (mask)
509 show port mac addresses info
510 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
512 Show mac addresses added for a specific port::
514 testpmd> show port (port_id) macs
517 show port multicast mac addresses info
518 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
520 Show multicast mac addresses added for a specific port::
522 testpmd> show port (port_id) mcast_macs
527 Show general information about devices probed::
529 testpmd> show device info (<identifier>|all)
533 .. code-block:: console
535 testpmd> show device info net_pcap0
537 ********************* Infos for device net_pcap0 *********************
539 Driver name: net_pcap
540 Devargs: iface=enP2p6s0,phy_mac=1
541 Connect to socket: -1
544 MAC address: 1E:37:93:28:04:B8
545 Device name: net_pcap0
550 Dumps all physical memory segment layouts::
552 testpmd> dump_physmem
557 Dumps the layout of all memory zones::
559 testpmd> dump_memzone
564 Dumps the memory usage of all sockets::
566 testpmd> dump_socket_mem
571 Dumps the size of all memory structures::
573 testpmd> dump_struct_sizes
578 Dumps the status of all or specific element in DPDK rings::
580 testpmd> dump_ring [ring_name]
585 Dumps the statistics of all or specific memory pool::
587 testpmd> dump_mempool [mempool_name]
592 Dumps the user device list::
594 testpmd> dump_devargs
599 Dumps the log level for all the dpdk modules::
601 testpmd> dump_log_types
603 show (raw_encap|raw_decap)
604 ~~~~~~~~~~~~~~~~~~~~~~~~~~
606 Display content of raw_encap/raw_decap buffers in hex::
608 testpmd> show <raw_encap|raw_decap> <index>
609 testpmd> show <raw_encap|raw_decap> all
613 testpmd> show raw_encap 6
615 index: 6 at [0x1c565b0], len=50
616 00000000: 00 00 00 00 00 00 16 26 36 46 56 66 08 00 45 00 | .......&6FVf..E.
617 00000010: 00 00 00 00 00 00 00 11 00 00 C0 A8 01 06 C0 A8 | ................
618 00000020: 03 06 00 00 00 FA 00 00 00 00 08 00 00 00 00 00 | ................
621 show fec capabilities
622 ~~~~~~~~~~~~~~~~~~~~~
624 Show fec capabilities of a port::
626 testpmd> show port (port_id) fec capabilities
631 Show fec mode of a port::
633 testpmd> show port (port_id) fec_mode
636 Configuration Functions
637 -----------------------
639 The testpmd application can be configured from the runtime as well as from the command-line.
641 This section details the available configuration functions that are available.
645 Configuration changes only become active when forwarding is started/restarted.
650 Reset forwarding to the default configuration::
657 Set the debug verbosity level::
659 testpmd> set verbose (level)
661 Available levels are as following:
663 * ``0`` silent except for error.
664 * ``1`` fully verbose except for Tx packets.
665 * ``2`` fully verbose except for Rx packets.
666 * ``> 2`` fully verbose.
671 Set the log level for a log type::
673 testpmd> set log global|(type) (level)
677 * ``type`` is the log name.
679 * ``level`` is the log level.
681 For example, to change the global log level::
683 testpmd> set log global (level)
685 Regexes can also be used for type. To change log level of user1, user2 and user3::
687 testpmd> set log user[1-3] (level)
692 Set the number of ports used by the application:
696 This is equivalent to the ``--nb-ports`` command-line option.
701 Set the number of cores used by the application::
703 testpmd> set nbcore (num)
705 This is equivalent to the ``--nb-cores`` command-line option.
709 The number of cores used must not be greater than number of ports used multiplied by the number of queues per port.
714 Set the forwarding cores hexadecimal mask::
716 testpmd> set coremask (mask)
718 This is equivalent to the ``--coremask`` command-line option.
722 The main lcore is reserved for command line parsing only and cannot be masked on for packet forwarding.
727 Set the forwarding ports hexadecimal mask::
729 testpmd> set portmask (mask)
731 This is equivalent to the ``--portmask`` command-line option.
733 set record-core-cycles
734 ~~~~~~~~~~~~~~~~~~~~~~
736 Set the recording of CPU cycles::
738 testpmd> set record-core-cycles (on|off)
742 * ``on`` enables measurement of CPU cycles per packet.
744 * ``off`` disables measurement of CPU cycles per packet.
746 This is equivalent to the ``--record-core-cycles command-line`` option.
748 set record-burst-stats
749 ~~~~~~~~~~~~~~~~~~~~~~
751 Set the displaying of RX and TX bursts::
753 testpmd> set record-burst-stats (on|off)
757 * ``on`` enables display of RX and TX bursts.
759 * ``off`` disables display of RX and TX bursts.
761 This is equivalent to the ``--record-burst-stats command-line`` option.
766 Set number of packets per burst::
768 testpmd> set burst (num)
770 This is equivalent to the ``--burst command-line`` option.
772 When retry is enabled, the transmit delay time and number of retries can also be set::
774 testpmd> set burst tx delay (microseconds) retry (num)
779 Set the offsets of segments relating to the data buffer beginning on receiving
780 if split feature is engaged. Affects only the queues configured with split
781 offloads (currently BUFFER_SPLIT is supported only).
783 testpmd> set rxoffs (x[,y]*)
785 Where x[,y]* represents a CSV list of values, without white space. If the list
786 of offsets is shorter than the list of segments the zero offsets will be used
787 for the remaining segments.
792 Set the length of segments to scatter packets on receiving if split
793 feature is engaged. Affects only the queues configured with split offloads
794 (currently BUFFER_SPLIT is supported only). Optionally the multiple memory
795 pools can be specified with --mbuf-size command line parameter and the mbufs
796 to receive will be allocated sequentially from these extra memory pools (the
797 mbuf for the first segment is allocated from the first pool, the second one
798 from the second pool, and so on, if segment number is greater then pool's the
799 mbuf for remaining segments will be allocated from the last valid pool).
801 testpmd> set rxpkts (x[,y]*)
803 Where x[,y]* represents a CSV list of values, without white space. Zero value
804 means to use the corresponding memory pool data buffer size.
809 Set the length of each segment of the TX-ONLY packets or length of packet for FLOWGEN mode::
811 testpmd> set txpkts (x[,y]*)
813 Where x[,y]* represents a CSV list of values, without white space.
818 Configure the timing burst pattern for Tx only mode. This command enables
819 the packet send scheduling on dynamic timestamp mbuf field and configures
820 timing pattern in Tx only mode. In this mode, if scheduling is enabled
821 application provides timestamps in the packets being sent. It is possible
822 to configure delay (in unspecified device clock units) between bursts
823 and between the packets within the burst::
825 testpmd> set txtimes (inter),(intra)
829 * ``inter`` is the delay between the bursts in the device clock units.
830 If ``intra`` is zero, this is the time between the beginnings of the
831 first packets in the neighbour bursts, if ``intra`` is not zero,
832 ``inter`` specifies the time between the beginning of the first packet
833 of the current burst and the beginning of the last packet of the
834 previous burst. If ``inter`` parameter is zero the send scheduling
835 on timestamps is disabled (default).
837 * ``intra`` is the delay between the packets within the burst specified
838 in the device clock units. The number of packets in the burst is defined
839 by regular burst setting. If ``intra`` parameter is zero no timestamps
840 provided in the packets excepting the first one in the burst.
842 As the result the bursts of packet will be transmitted with specific
843 delays between the packets within the burst and specific delay between
844 the bursts. The rte_eth_read_clock() must be supported by the device(s)
845 and is supposed to be engaged to get the current device clock value
846 and provide the reference for the timestamps. If there is no supported
847 rte_eth_read_clock() there will be no send scheduling provided on the port.
852 Set the split policy for the TX packets, applicable for TX-ONLY and CSUM forwarding modes::
854 testpmd> set txsplit (off|on|rand)
858 * ``off`` disable packet copy & split for CSUM mode.
860 * ``on`` split outgoing packet into multiple segments. Size of each segment
861 and number of segments per packet is determined by ``set txpkts`` command
864 * ``rand`` same as 'on', but number of segments per each packet is a random value between 1 and total number of segments.
869 Set the list of forwarding cores::
871 testpmd> set corelist (x[,y]*)
873 For example, to change the forwarding cores:
875 .. code-block:: console
877 testpmd> set corelist 3,1
878 testpmd> show config fwd
880 io packet forwarding - ports=2 - cores=2 - streams=2 - NUMA support disabled
881 Logical Core 3 (socket 0) forwards packets on 1 streams:
882 RX P=0/Q=0 (socket 0) -> TX P=1/Q=0 (socket 0) peer=02:00:00:00:00:01
883 Logical Core 1 (socket 0) forwards packets on 1 streams:
884 RX P=1/Q=0 (socket 0) -> TX P=0/Q=0 (socket 0) peer=02:00:00:00:00:00
888 The cores are used in the same order as specified on the command line.
893 Set the list of forwarding ports::
895 testpmd> set portlist (x[,y]*)
897 For example, to change the port forwarding:
899 .. code-block:: console
901 testpmd> set portlist 0,2,1,3
902 testpmd> show config fwd
904 io packet forwarding - ports=4 - cores=1 - streams=4
905 Logical Core 3 (socket 0) forwards packets on 4 streams:
906 RX P=0/Q=0 (socket 0) -> TX P=2/Q=0 (socket 0) peer=02:00:00:00:00:01
907 RX P=2/Q=0 (socket 0) -> TX P=0/Q=0 (socket 0) peer=02:00:00:00:00:00
908 RX P=1/Q=0 (socket 0) -> TX P=3/Q=0 (socket 0) peer=02:00:00:00:00:03
909 RX P=3/Q=0 (socket 0) -> TX P=1/Q=0 (socket 0) peer=02:00:00:00:00:02
914 Select how to retrieve new ports created after "port attach" command::
916 testpmd> set port setup on (iterator|event)
918 For each new port, a setup is done.
919 It will find the probed ports via RTE_ETH_FOREACH_MATCHING_DEV loop
920 in iterator mode, or via RTE_ETH_EVENT_NEW in event mode.
925 Enable/disable tx loopback::
927 testpmd> set tx loopback (port_id) (on|off)
932 set drop enable bit for all queues::
934 testpmd> set all queues drop (port_id) (on|off)
936 set split drop enable (for VF)
937 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
939 set split drop enable bit for VF from PF::
941 testpmd> set vf split drop (port_id) (vf_id) (on|off)
943 set mac antispoof (for VF)
944 ~~~~~~~~~~~~~~~~~~~~~~~~~~
946 Set mac antispoof for a VF from the PF::
948 testpmd> set vf mac antispoof (port_id) (vf_id) (on|off)
953 Enable/disable MACsec offload::
955 testpmd> set macsec offload (port_id) on encrypt (on|off) replay-protect (on|off)
956 testpmd> set macsec offload (port_id) off
961 Configure MACsec secure connection (SC)::
963 testpmd> set macsec sc (tx|rx) (port_id) (mac) (pi)
967 The pi argument is ignored for tx.
968 Check the NIC Datasheet for hardware limits.
973 Configure MACsec secure association (SA)::
975 testpmd> set macsec sa (tx|rx) (port_id) (idx) (an) (pn) (key)
979 The IDX value must be 0 or 1.
980 Check the NIC Datasheet for hardware limits.
982 set broadcast mode (for VF)
983 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
985 Set broadcast mode for a VF from the PF::
987 testpmd> set vf broadcast (port_id) (vf_id) (on|off)
992 Set the VLAN strip for a queue on a port::
994 testpmd> vlan set stripq (on|off) (port_id,queue_id)
996 vlan set stripq (for VF)
997 ~~~~~~~~~~~~~~~~~~~~~~~~
999 Set VLAN strip for all queues in a pool for a VF from the PF::
1001 testpmd> set vf vlan stripq (port_id) (vf_id) (on|off)
1003 vlan set insert (for VF)
1004 ~~~~~~~~~~~~~~~~~~~~~~~~
1006 Set VLAN insert for a VF from the PF::
1008 testpmd> set vf vlan insert (port_id) (vf_id) (vlan_id)
1010 vlan set tag (for VF)
1011 ~~~~~~~~~~~~~~~~~~~~~
1013 Set VLAN tag for a VF from the PF::
1015 testpmd> set vf vlan tag (port_id) (vf_id) (on|off)
1017 vlan set antispoof (for VF)
1018 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
1020 Set VLAN antispoof for a VF from the PF::
1022 testpmd> set vf vlan antispoof (port_id) (vf_id) (on|off)
1024 vlan set (strip|filter|qinq_strip|extend)
1025 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1026 Set the VLAN strip/filter/QinQ strip/extend on for a port::
1028 testpmd> vlan set (strip|filter|qinq_strip|extend) (on|off) (port_id)
1033 Set the inner or outer VLAN TPID for packet filtering on a port::
1035 testpmd> vlan set (inner|outer) tpid (value) (port_id)
1039 TPID value must be a 16-bit number (value <= 65536).
1044 Add a VLAN ID, or all identifiers, to the set of VLAN identifiers filtered by port ID::
1046 testpmd> rx_vlan add (vlan_id|all) (port_id)
1050 VLAN filter must be set on that port. VLAN ID < 4096.
1051 Depending on the NIC used, number of vlan_ids may be limited to the maximum entries
1052 in VFTA table. This is important if enabling all vlan_ids.
1057 Remove a VLAN ID, or all identifiers, from the set of VLAN identifiers filtered by port ID::
1059 testpmd> rx_vlan rm (vlan_id|all) (port_id)
1061 rx_vlan add (for VF)
1062 ~~~~~~~~~~~~~~~~~~~~
1064 Add a VLAN ID, to the set of VLAN identifiers filtered for VF(s) for port ID::
1066 testpmd> rx_vlan add (vlan_id) port (port_id) vf (vf_mask)
1071 Remove a VLAN ID, from the set of VLAN identifiers filtered for VF(s) for port ID::
1073 testpmd> rx_vlan rm (vlan_id) port (port_id) vf (vf_mask)
1078 Add an UDP port for VXLAN packet filter on a port::
1080 testpmd> rx_vxlan_port add (udp_port) (port_id)
1082 rx_vxlan_port remove
1083 ~~~~~~~~~~~~~~~~~~~~
1085 Remove an UDP port for VXLAN packet filter on a port::
1087 testpmd> rx_vxlan_port rm (udp_port) (port_id)
1092 Set hardware insertion of VLAN IDs in packets sent on a port::
1094 testpmd> tx_vlan set (port_id) vlan_id[, vlan_id_outer]
1096 For example, set a single VLAN ID (5) insertion on port 0::
1100 Or, set double VLAN ID (inner: 2, outer: 3) insertion on port 1::
1108 Set port based hardware insertion of VLAN ID in packets sent on a port::
1110 testpmd> tx_vlan set pvid (port_id) (vlan_id) (on|off)
1115 Disable hardware insertion of a VLAN header in packets sent on a port::
1117 testpmd> tx_vlan reset (port_id)
1122 Select hardware or software calculation of the checksum when
1123 transmitting a packet using the ``csum`` forwarding engine::
1125 testpmd> csum set (ip|udp|tcp|sctp|outer-ip|outer-udp) (hw|sw) (port_id)
1129 * ``ip|udp|tcp|sctp`` always relate to the inner layer.
1131 * ``outer-ip`` relates to the outer IP layer (only for IPv4) in the case where the packet is recognized
1132 as a tunnel packet by the forwarding engine (geneve, gre, gtp, ipip, vxlan and vxlan-gpe are
1133 supported). See also the ``csum parse-tunnel`` command.
1135 * ``outer-udp`` relates to the outer UDP layer in the case where the packet is recognized
1136 as a tunnel packet by the forwarding engine (geneve, gtp, vxlan and vxlan-gpe are
1137 supported). See also the ``csum parse-tunnel`` command.
1141 Check the NIC Datasheet for hardware limits.
1146 Set RSS queue region span on a port::
1148 testpmd> set port (port_id) queue-region region_id (value) \
1149 queue_start_index (value) queue_num (value)
1151 Set flowtype mapping on a RSS queue region on a port::
1153 testpmd> set port (port_id) queue-region region_id (value) flowtype (value)
1157 * For the flowtype(pctype) of packet,the specific index for each type has
1158 been defined in file i40e_type.h as enum i40e_filter_pctype.
1160 Set user priority mapping on a RSS queue region on a port::
1162 testpmd> set port (port_id) queue-region UP (value) region_id (value)
1164 Flush all queue region related configuration on a port::
1166 testpmd> set port (port_id) queue-region flush (on|off)
1170 * ``on``: is just an enable function which server for other configuration,
1171 it is for all configuration about queue region from up layer,
1172 at first will only keep in DPDK software stored in driver,
1173 only after "flush on", it commit all configuration to HW.
1175 * ``"off``: is just clean all configuration about queue region just now,
1176 and restore all to DPDK i40e driver default config when start up.
1178 Show all queue region related configuration info on a port::
1180 testpmd> show port (port_id) queue-region
1184 Queue region only support on PF by now, so these command is
1185 only for configuration of queue region on PF port.
1190 Define how tunneled packets should be handled by the csum forward
1193 testpmd> csum parse-tunnel (on|off) (tx_port_id)
1195 If enabled, the csum forward engine will try to recognize supported
1196 tunnel headers (geneve, gtp, gre, ipip, vxlan, vxlan-gpe).
1198 If disabled, treat tunnel packets as non-tunneled packets (a inner
1199 header is handled as a packet payload).
1203 The port argument is the TX port like in the ``csum set`` command.
1207 Consider a packet in packet like the following::
1209 eth_out/ipv4_out/udp_out/vxlan/eth_in/ipv4_in/tcp_in
1211 * If parse-tunnel is enabled, the ``ip|udp|tcp|sctp`` parameters of ``csum set``
1212 command relate to the inner headers (here ``ipv4_in`` and ``tcp_in``), and the
1213 ``outer-ip|outer-udp`` parameter relates to the outer headers (here ``ipv4_out`` and ``udp_out``).
1215 * If parse-tunnel is disabled, the ``ip|udp|tcp|sctp`` parameters of ``csum set``
1216 command relate to the outer headers, here ``ipv4_out`` and ``udp_out``.
1221 Display tx checksum offload configuration::
1223 testpmd> csum show (port_id)
1228 Enable TCP Segmentation Offload (TSO) in the ``csum`` forwarding engine::
1230 testpmd> tso set (segsize) (port_id)
1234 Check the NIC datasheet for hardware limits.
1239 Display the status of TCP Segmentation Offload::
1241 testpmd> tso show (port_id)
1246 Set tso segment size of tunneled packets for a port in csum engine::
1248 testpmd> tunnel_tso set (tso_segsz) (port_id)
1253 Display the status of tunneled TCP Segmentation Offload for a port::
1255 testpmd> tunnel_tso show (port_id)
1260 Enable or disable GRO in ``csum`` forwarding engine::
1262 testpmd> set port <port_id> gro on|off
1264 If enabled, the csum forwarding engine will perform GRO on the TCP/IPv4
1265 packets received from the given port.
1267 If disabled, packets received from the given port won't be performed
1268 GRO. By default, GRO is disabled for all ports.
1272 When enable GRO for a port, TCP/IPv4 packets received from the port
1273 will be performed GRO. After GRO, all merged packets have bad
1274 checksums, since the GRO library doesn't re-calculate checksums for
1275 the merged packets. Therefore, if users want the merged packets to
1276 have correct checksums, please select HW IP checksum calculation and
1277 HW TCP checksum calculation for the port which the merged packets are
1283 Display GRO configuration for a given port::
1285 testpmd> show port <port_id> gro
1290 Set the cycle to flush the GROed packets from reassembly tables::
1292 testpmd> set gro flush <cycles>
1294 When enable GRO, the csum forwarding engine performs GRO on received
1295 packets, and the GROed packets are stored in reassembly tables. Users
1296 can use this command to determine when the GROed packets are flushed
1297 from the reassembly tables.
1299 The ``cycles`` is measured in GRO operation times. The csum forwarding
1300 engine flushes the GROed packets from the tables every ``cycles`` GRO
1303 By default, the value of ``cycles`` is 1, which means flush GROed packets
1304 from the reassembly tables as soon as one GRO operation finishes. The value
1305 of ``cycles`` should be in the range of 1 to ``GRO_MAX_FLUSH_CYCLES``.
1307 Please note that the large value of ``cycles`` may cause the poor TCP/IP
1308 stack performance. Because the GROed packets are delayed to arrive the
1309 stack, thus causing more duplicated ACKs and TCP retransmissions.
1314 Toggle per-port GSO support in ``csum`` forwarding engine::
1316 testpmd> set port <port_id> gso on|off
1318 If enabled, the csum forwarding engine will perform GSO on supported IPv4
1319 packets, transmitted on the given port.
1321 If disabled, packets transmitted on the given port will not undergo GSO.
1322 By default, GSO is disabled for all ports.
1326 When GSO is enabled on a port, supported IPv4 packets transmitted on that
1327 port undergo GSO. Afterwards, the segmented packets are represented by
1328 multi-segment mbufs; however, the csum forwarding engine doesn't calculation
1329 of checksums for GSO'd segments in SW. As a result, if users want correct
1330 checksums in GSO segments, they should enable HW checksum calculation for
1333 For example, HW checksum calculation for VxLAN GSO'd packets may be enabled
1334 by setting the following options in the csum forwarding engine:
1336 testpmd> csum set outer_ip hw <port_id>
1338 testpmd> csum set ip hw <port_id>
1340 testpmd> csum set tcp hw <port_id>
1342 UDP GSO is the same as IP fragmentation, which treats the UDP header
1343 as the payload and does not modify it during segmentation. That is,
1344 after UDP GSO, only the first output fragment has the original UDP
1345 header. Therefore, users need to enable HW IP checksum calculation
1346 and SW UDP checksum calculation for GSO-enabled ports, if they want
1347 correct checksums for UDP/IPv4 packets.
1352 Set the maximum GSO segment size (measured in bytes), which includes the
1353 packet header and the packet payload for GSO-enabled ports (global)::
1355 testpmd> set gso segsz <length>
1360 Display the status of Generic Segmentation Offload for a given port::
1362 testpmd> show port <port_id> gso
1367 Add an alternative MAC address to a port::
1369 testpmd> mac_addr add (port_id) (XX:XX:XX:XX:XX:XX)
1374 Remove a MAC address from a port::
1376 testpmd> mac_addr remove (port_id) (XX:XX:XX:XX:XX:XX)
1381 To add the multicast MAC address to/from the set of multicast addresses
1384 testpmd> mcast_addr add (port_id) (mcast_addr)
1389 To remove the multicast MAC address to/from the set of multicast addresses
1392 testpmd> mcast_addr remove (port_id) (mcast_addr)
1394 mac_addr add (for VF)
1395 ~~~~~~~~~~~~~~~~~~~~~
1397 Add an alternative MAC address for a VF to a port::
1399 testpmd> mac_add add port (port_id) vf (vf_id) (XX:XX:XX:XX:XX:XX)
1404 Set the default MAC address for a port::
1406 testpmd> mac_addr set (port_id) (XX:XX:XX:XX:XX:XX)
1408 mac_addr set (for VF)
1409 ~~~~~~~~~~~~~~~~~~~~~
1411 Set the MAC address for a VF from the PF::
1413 testpmd> set vf mac addr (port_id) (vf_id) (XX:XX:XX:XX:XX:XX)
1418 Set the forwarding peer address for certain port::
1420 testpmd> set eth-peer (port_id) (peer_addr)
1422 This is equivalent to the ``--eth-peer`` command-line option.
1427 Set the unicast hash filter(s) on/off for a port::
1429 testpmd> set port (port_id) uta (XX:XX:XX:XX:XX:XX|all) (on|off)
1434 Set the promiscuous mode on for a port or for all ports.
1435 In promiscuous mode packets are not dropped if they aren't for the specified MAC address::
1437 testpmd> set promisc (port_id|all) (on|off)
1442 Set the allmulti mode for a port or for all ports::
1444 testpmd> set allmulti (port_id|all) (on|off)
1446 Same as the ifconfig (8) option. Controls how multicast packets are handled.
1448 set promisc (for VF)
1449 ~~~~~~~~~~~~~~~~~~~~
1451 Set the unicast promiscuous mode for a VF from PF.
1452 It's supported by Intel i40e NICs now.
1453 In promiscuous mode packets are not dropped if they aren't for the specified MAC address::
1455 testpmd> set vf promisc (port_id) (vf_id) (on|off)
1457 set allmulticast (for VF)
1458 ~~~~~~~~~~~~~~~~~~~~~~~~~
1460 Set the multicast promiscuous mode for a VF from PF.
1461 It's supported by Intel i40e NICs now.
1462 In promiscuous mode packets are not dropped if they aren't for the specified MAC address::
1464 testpmd> set vf allmulti (port_id) (vf_id) (on|off)
1466 set tx max bandwidth (for VF)
1467 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1469 Set TX max absolute bandwidth (Mbps) for a VF from PF::
1471 testpmd> set vf tx max-bandwidth (port_id) (vf_id) (max_bandwidth)
1473 set tc tx min bandwidth (for VF)
1474 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1476 Set all TCs' TX min relative bandwidth (%) for a VF from PF::
1478 testpmd> set vf tc tx min-bandwidth (port_id) (vf_id) (bw1, bw2, ...)
1480 set tc tx max bandwidth (for VF)
1481 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1483 Set a TC's TX max absolute bandwidth (Mbps) for a VF from PF::
1485 testpmd> set vf tc tx max-bandwidth (port_id) (vf_id) (tc_no) (max_bandwidth)
1487 set tc strict link priority mode
1488 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1490 Set some TCs' strict link priority mode on a physical port::
1492 testpmd> set tx strict-link-priority (port_id) (tc_bitmap)
1494 set tc tx min bandwidth
1495 ~~~~~~~~~~~~~~~~~~~~~~~
1497 Set all TCs' TX min relative bandwidth (%) globally for all PF and VFs::
1499 testpmd> set tc tx min-bandwidth (port_id) (bw1, bw2, ...)
1504 Set the link flow control parameter on a port::
1506 testpmd> set flow_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \
1507 (pause_time) (send_xon) mac_ctrl_frame_fwd (on|off) \
1508 autoneg (on|off) (port_id)
1512 * ``high_water`` (integer): High threshold value to trigger XOFF.
1514 * ``low_water`` (integer): Low threshold value to trigger XON.
1516 * ``pause_time`` (integer): Pause quota in the Pause frame.
1518 * ``send_xon`` (0/1): Send XON frame.
1520 * ``mac_ctrl_frame_fwd``: Enable receiving MAC control frames.
1522 * ``autoneg``: Change the auto-negotiation parameter.
1527 Set the priority flow control parameter on a port::
1529 testpmd> set pfc_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \
1530 (pause_time) (priority) (port_id)
1534 * ``high_water`` (integer): High threshold value.
1536 * ``low_water`` (integer): Low threshold value.
1538 * ``pause_time`` (integer): Pause quota in the Pause frame.
1540 * ``priority`` (0-7): VLAN User Priority.
1545 Set statistics mapping (qmapping 0..15) for RX/TX queue on port::
1547 testpmd> set stat_qmap (tx|rx) (port_id) (queue_id) (qmapping)
1549 For example, to set rx queue 2 on port 0 to mapping 5::
1551 testpmd>set stat_qmap rx 0 2 5
1553 set xstats-hide-zero
1554 ~~~~~~~~~~~~~~~~~~~~
1556 Set the option to hide zero values for xstats display::
1558 testpmd> set xstats-hide-zero on|off
1562 By default, the zero values are displayed for xstats.
1564 set port - rx/tx (for VF)
1565 ~~~~~~~~~~~~~~~~~~~~~~~~~
1567 Set VF receive/transmit from a port::
1569 testpmd> set port (port_id) vf (vf_id) (rx|tx) (on|off)
1571 set port - rx mode(for VF)
1572 ~~~~~~~~~~~~~~~~~~~~~~~~~~
1574 Set the VF receive mode of a port::
1576 testpmd> set port (port_id) vf (vf_id) \
1577 rxmode (AUPE|ROPE|BAM|MPE) (on|off)
1579 The available receive modes are:
1581 * ``AUPE``: Accepts untagged VLAN.
1583 * ``ROPE``: Accepts unicast hash.
1585 * ``BAM``: Accepts broadcast packets.
1587 * ``MPE``: Accepts all multicast packets.
1589 set port - tx_rate (for Queue)
1590 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1592 Set TX rate limitation for a queue on a port::
1594 testpmd> set port (port_id) queue (queue_id) rate (rate_value)
1596 set port - tx_rate (for VF)
1597 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
1599 Set TX rate limitation for queues in VF on a port::
1601 testpmd> set port (port_id) vf (vf_id) rate (rate_value) queue_mask (queue_mask)
1603 set port - mirror rule
1604 ~~~~~~~~~~~~~~~~~~~~~~
1606 Set pool or vlan type mirror rule for a port::
1608 testpmd> set port (port_id) mirror-rule (rule_id) \
1609 (pool-mirror-up|pool-mirror-down|vlan-mirror) \
1610 (poolmask|vlanid[,vlanid]*) dst-pool (pool_id) (on|off)
1612 Set link mirror rule for a port::
1614 testpmd> set port (port_id) mirror-rule (rule_id) \
1615 (uplink-mirror|downlink-mirror) dst-pool (pool_id) (on|off)
1617 For example to enable mirror traffic with vlan 0,1 to pool 0::
1619 set port 0 mirror-rule 0 vlan-mirror 0,1 dst-pool 0 on
1621 reset port - mirror rule
1622 ~~~~~~~~~~~~~~~~~~~~~~~~
1624 Reset a mirror rule for a port::
1626 testpmd> reset port (port_id) mirror-rule (rule_id)
1631 Set the flush on RX streams before forwarding.
1632 The default is flush ``on``.
1633 Mainly used with PCAP drivers to turn off the default behavior of flushing the first 512 packets on RX streams::
1635 testpmd> set flush_rx off
1640 Set the bypass mode for the lowest port on bypass enabled NIC::
1642 testpmd> set bypass mode (normal|bypass|isolate) (port_id)
1647 Set the event required to initiate specified bypass mode for the lowest port on a bypass enabled::
1649 testpmd> set bypass event (timeout|os_on|os_off|power_on|power_off) \
1650 mode (normal|bypass|isolate) (port_id)
1654 * ``timeout``: Enable bypass after watchdog timeout.
1656 * ``os_on``: Enable bypass when OS/board is powered on.
1658 * ``os_off``: Enable bypass when OS/board is powered off.
1660 * ``power_on``: Enable bypass when power supply is turned on.
1662 * ``power_off``: Enable bypass when power supply is turned off.
1668 Set the bypass watchdog timeout to ``n`` seconds where 0 = instant::
1670 testpmd> set bypass timeout (0|1.5|2|3|4|8|16|32)
1675 Show the bypass configuration for a bypass enabled NIC using the lowest port on the NIC::
1677 testpmd> show bypass config (port_id)
1682 Set link up for a port::
1684 testpmd> set link-up port (port id)
1689 Set link down for a port::
1691 testpmd> set link-down port (port id)
1696 Enable E-tag insertion for a VF on a port::
1698 testpmd> E-tag set insertion on port-tag-id (value) port (port_id) vf (vf_id)
1700 Disable E-tag insertion for a VF on a port::
1702 testpmd> E-tag set insertion off port (port_id) vf (vf_id)
1704 Enable/disable E-tag stripping on a port::
1706 testpmd> E-tag set stripping (on|off) port (port_id)
1708 Enable/disable E-tag based forwarding on a port::
1710 testpmd> E-tag set forwarding (on|off) port (port_id)
1712 Add an E-tag forwarding filter on a port::
1714 testpmd> E-tag set filter add e-tag-id (value) dst-pool (pool_id) port (port_id)
1716 Delete an E-tag forwarding filter on a port::
1717 testpmd> E-tag set filter del e-tag-id (value) port (port_id)
1722 Load a dynamic device personalization (DDP) profile and store backup profile::
1724 testpmd> ddp add (port_id) (profile_path[,backup_profile_path])
1729 Delete a dynamic device personalization profile and restore backup profile::
1731 testpmd> ddp del (port_id) (backup_profile_path)
1736 List all items from the ptype mapping table::
1738 testpmd> ptype mapping get (port_id) (valid_only)
1742 * ``valid_only``: A flag indicates if only list valid items(=1) or all itemss(=0).
1744 Replace a specific or a group of software defined ptype with a new one::
1746 testpmd> ptype mapping replace (port_id) (target) (mask) (pkt_type)
1750 * ``target``: A specific software ptype or a mask to represent a group of software ptypes.
1752 * ``mask``: A flag indicate if "target" is a specific software ptype(=0) or a ptype mask(=1).
1754 * ``pkt_type``: The new software ptype to replace the old ones.
1756 Update hardware defined ptype to software defined packet type mapping table::
1758 testpmd> ptype mapping update (port_id) (hw_ptype) (sw_ptype)
1762 * ``hw_ptype``: hardware ptype as the index of the ptype mapping table.
1764 * ``sw_ptype``: software ptype as the value of the ptype mapping table.
1766 Reset ptype mapping table::
1768 testpmd> ptype mapping reset (port_id)
1770 config per port Rx offloading
1771 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1773 Enable or disable a per port Rx offloading on all Rx queues of a port::
1775 testpmd> port config (port_id) rx_offload (offloading) on|off
1777 * ``offloading``: can be any of these offloading capability:
1778 vlan_strip, ipv4_cksum, udp_cksum, tcp_cksum, tcp_lro,
1779 qinq_strip, outer_ipv4_cksum, macsec_strip,
1780 header_split, vlan_filter, vlan_extend, jumbo_frame,
1781 scatter, timestamp, security, keep_crc, rss_hash
1783 This command should be run when the port is stopped, or else it will fail.
1785 config per queue Rx offloading
1786 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1788 Enable or disable a per queue Rx offloading only on a specific Rx queue::
1790 testpmd> port (port_id) rxq (queue_id) rx_offload (offloading) on|off
1792 * ``offloading``: can be any of these offloading capability:
1793 vlan_strip, ipv4_cksum, udp_cksum, tcp_cksum, tcp_lro,
1794 qinq_strip, outer_ipv4_cksum, macsec_strip,
1795 header_split, vlan_filter, vlan_extend, jumbo_frame,
1796 scatter, timestamp, security, keep_crc
1798 This command should be run when the port is stopped, or else it will fail.
1800 config per port Tx offloading
1801 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1803 Enable or disable a per port Tx offloading on all Tx queues of a port::
1805 testpmd> port config (port_id) tx_offload (offloading) on|off
1807 * ``offloading``: can be any of these offloading capability:
1808 vlan_insert, ipv4_cksum, udp_cksum, tcp_cksum,
1809 sctp_cksum, tcp_tso, udp_tso, outer_ipv4_cksum,
1810 qinq_insert, vxlan_tnl_tso, gre_tnl_tso,
1811 ipip_tnl_tso, geneve_tnl_tso, macsec_insert,
1812 mt_lockfree, multi_segs, mbuf_fast_free, security
1814 This command should be run when the port is stopped, or else it will fail.
1816 config per queue Tx offloading
1817 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1819 Enable or disable a per queue Tx offloading only on a specific Tx queue::
1821 testpmd> port (port_id) txq (queue_id) tx_offload (offloading) on|off
1823 * ``offloading``: can be any of these offloading capability:
1824 vlan_insert, ipv4_cksum, udp_cksum, tcp_cksum,
1825 sctp_cksum, tcp_tso, udp_tso, outer_ipv4_cksum,
1826 qinq_insert, vxlan_tnl_tso, gre_tnl_tso,
1827 ipip_tnl_tso, geneve_tnl_tso, macsec_insert,
1828 mt_lockfree, multi_segs, mbuf_fast_free, security
1830 This command should be run when the port is stopped, or else it will fail.
1832 Config VXLAN Encap outer layers
1833 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1835 Configure the outer layer to encapsulate a packet inside a VXLAN tunnel::
1837 set vxlan ip-version (ipv4|ipv6) vni (vni) udp-src (udp-src) \
1838 udp-dst (udp-dst) ip-src (ip-src) ip-dst (ip-dst) eth-src (eth-src) \
1841 set vxlan-with-vlan ip-version (ipv4|ipv6) vni (vni) udp-src (udp-src) \
1842 udp-dst (udp-dst) ip-src (ip-src) ip-dst (ip-dst) vlan-tci (vlan-tci) \
1843 eth-src (eth-src) eth-dst (eth-dst)
1845 set vxlan-tos-ttl ip-version (ipv4|ipv6) vni (vni) udp-src (udp-src) \
1846 udp-dst (udp-dst) ip-tos (ip-tos) ip-ttl (ip-ttl) ip-src (ip-src) \
1847 ip-dst (ip-dst) eth-src (eth-src) eth-dst (eth-dst)
1849 These commands will set an internal configuration inside testpmd, any following
1850 flow rule using the action vxlan_encap will use the last configuration set.
1851 To have a different encapsulation header, one of those commands must be called
1852 before the flow rule creation.
1854 Config NVGRE Encap outer layers
1855 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1857 Configure the outer layer to encapsulate a packet inside a NVGRE tunnel::
1859 set nvgre ip-version (ipv4|ipv6) tni (tni) ip-src (ip-src) ip-dst (ip-dst) \
1860 eth-src (eth-src) eth-dst (eth-dst)
1861 set nvgre-with-vlan ip-version (ipv4|ipv6) tni (tni) ip-src (ip-src) \
1862 ip-dst (ip-dst) vlan-tci (vlan-tci) eth-src (eth-src) eth-dst (eth-dst)
1864 These commands will set an internal configuration inside testpmd, any following
1865 flow rule using the action nvgre_encap will use the last configuration set.
1866 To have a different encapsulation header, one of those commands must be called
1867 before the flow rule creation.
1872 Configure the l2 to be used when encapsulating a packet with L2::
1874 set l2_encap ip-version (ipv4|ipv6) eth-src (eth-src) eth-dst (eth-dst)
1875 set l2_encap-with-vlan ip-version (ipv4|ipv6) vlan-tci (vlan-tci) \
1876 eth-src (eth-src) eth-dst (eth-dst)
1878 Those commands will set an internal configuration inside testpmd, any following
1879 flow rule using the action l2_encap will use the last configuration set.
1880 To have a different encapsulation header, one of those commands must be called
1881 before the flow rule creation.
1886 Configure the l2 to be removed when decapsulating a packet with L2::
1888 set l2_decap ip-version (ipv4|ipv6)
1889 set l2_decap-with-vlan ip-version (ipv4|ipv6)
1891 Those commands will set an internal configuration inside testpmd, any following
1892 flow rule using the action l2_decap will use the last configuration set.
1893 To have a different encapsulation header, one of those commands must be called
1894 before the flow rule creation.
1896 Config MPLSoGRE Encap outer layers
1897 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1899 Configure the outer layer to encapsulate a packet inside a MPLSoGRE tunnel::
1901 set mplsogre_encap ip-version (ipv4|ipv6) label (label) \
1902 ip-src (ip-src) ip-dst (ip-dst) eth-src (eth-src) eth-dst (eth-dst)
1903 set mplsogre_encap-with-vlan ip-version (ipv4|ipv6) label (label) \
1904 ip-src (ip-src) ip-dst (ip-dst) vlan-tci (vlan-tci) \
1905 eth-src (eth-src) eth-dst (eth-dst)
1907 These commands will set an internal configuration inside testpmd, any following
1908 flow rule using the action mplsogre_encap will use the last configuration set.
1909 To have a different encapsulation header, one of those commands must be called
1910 before the flow rule creation.
1912 Config MPLSoGRE Decap outer layers
1913 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1915 Configure the outer layer to decapsulate MPLSoGRE packet::
1917 set mplsogre_decap ip-version (ipv4|ipv6)
1918 set mplsogre_decap-with-vlan ip-version (ipv4|ipv6)
1920 These commands will set an internal configuration inside testpmd, any following
1921 flow rule using the action mplsogre_decap will use the last configuration set.
1922 To have a different decapsulation header, one of those commands must be called
1923 before the flow rule creation.
1925 Config MPLSoUDP Encap outer layers
1926 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1928 Configure the outer layer to encapsulate a packet inside a MPLSoUDP tunnel::
1930 set mplsoudp_encap ip-version (ipv4|ipv6) label (label) udp-src (udp-src) \
1931 udp-dst (udp-dst) ip-src (ip-src) ip-dst (ip-dst) \
1932 eth-src (eth-src) eth-dst (eth-dst)
1933 set mplsoudp_encap-with-vlan ip-version (ipv4|ipv6) label (label) \
1934 udp-src (udp-src) udp-dst (udp-dst) ip-src (ip-src) ip-dst (ip-dst) \
1935 vlan-tci (vlan-tci) eth-src (eth-src) eth-dst (eth-dst)
1937 These commands will set an internal configuration inside testpmd, any following
1938 flow rule using the action mplsoudp_encap will use the last configuration set.
1939 To have a different encapsulation header, one of those commands must be called
1940 before the flow rule creation.
1942 Config MPLSoUDP Decap outer layers
1943 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1945 Configure the outer layer to decapsulate MPLSoUDP packet::
1947 set mplsoudp_decap ip-version (ipv4|ipv6)
1948 set mplsoudp_decap-with-vlan ip-version (ipv4|ipv6)
1950 These commands will set an internal configuration inside testpmd, any following
1951 flow rule using the action mplsoudp_decap will use the last configuration set.
1952 To have a different decapsulation header, one of those commands must be called
1953 before the flow rule creation.
1955 Config Raw Encapsulation
1956 ~~~~~~~~~~~~~~~~~~~~~~~~~
1958 Configure the raw data to be used when encapsulating a packet by
1959 rte_flow_action_raw_encap::
1961 set raw_encap {index} {item} [/ {item} [...]] / end_set
1963 There are multiple global buffers for ``raw_encap``, this command will set one
1964 internal buffer index by ``{index}``.
1965 If there is no ``{index}`` specified::
1967 set raw_encap {item} [/ {item} [...]] / end_set
1969 the default index ``0`` is used.
1970 In order to use different encapsulating header, ``index`` must be specified
1971 during the flow rule creation::
1973 testpmd> flow create 0 egress pattern eth / ipv4 / end actions
1974 raw_encap index 2 / end
1976 Otherwise the default index ``0`` is used.
1978 Config Raw Decapsulation
1979 ~~~~~~~~~~~~~~~~~~~~~~~~
1981 Configure the raw data to be used when decapsulating a packet by
1982 rte_flow_action_raw_decap::
1984 set raw_decap {index} {item} [/ {item} [...]] / end_set
1986 There are multiple global buffers for ``raw_decap``, this command will set
1987 one internal buffer index by ``{index}``.
1988 If there is no ``{index}`` specified::
1990 set raw_decap {item} [/ {item} [...]] / end_set
1992 the default index ``0`` is used.
1993 In order to use different decapsulating header, ``index`` must be specified
1994 during the flow rule creation::
1996 testpmd> flow create 0 egress pattern eth / ipv4 / end actions
1997 raw_encap index 3 / end
1999 Otherwise the default index ``0`` is used.
2004 Set fec mode for a specific port::
2006 testpmd> set port (port_id) fec_mode auto|off|rs|baser
2012 The following sections show functions for configuring ports.
2016 Port configuration changes only become active when forwarding is started/restarted.
2021 Attach a port specified by pci address or virtual device args::
2023 testpmd> port attach (identifier)
2025 To attach a new pci device, the device should be recognized by kernel first.
2026 Then it should be moved under DPDK management.
2027 Finally the port can be attached to testpmd.
2029 For example, to move a pci device using ixgbe under DPDK management:
2031 .. code-block:: console
2033 # Check the status of the available devices.
2034 ./usertools/dpdk-devbind.py --status
2036 Network devices using DPDK-compatible driver
2037 ============================================
2040 Network devices using kernel driver
2041 ===================================
2042 0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused=
2045 # Bind the device to igb_uio.
2046 sudo ./usertools/dpdk-devbind.py -b igb_uio 0000:0a:00.0
2049 # Recheck the status of the devices.
2050 ./usertools/dpdk-devbind.py --status
2051 Network devices using DPDK-compatible driver
2052 ============================================
2053 0000:0a:00.0 '82599ES 10-Gigabit' drv=igb_uio unused=
2055 To attach a port created by virtual device, above steps are not needed.
2057 For example, to attach a port whose pci address is 0000:0a:00.0.
2059 .. code-block:: console
2061 testpmd> port attach 0000:0a:00.0
2062 Attaching a new port...
2063 EAL: PCI device 0000:0a:00.0 on NUMA socket -1
2064 EAL: probe driver: 8086:10fb rte_ixgbe_pmd
2065 EAL: PCI memory mapped at 0x7f83bfa00000
2066 EAL: PCI memory mapped at 0x7f83bfa80000
2067 PMD: eth_ixgbe_dev_init(): MAC: 2, PHY: 18, SFP+: 5
2068 PMD: eth_ixgbe_dev_init(): port 0 vendorID=0x8086 deviceID=0x10fb
2069 Port 0 is attached. Now total ports is 1
2072 For example, to attach a port created by pcap PMD.
2074 .. code-block:: console
2076 testpmd> port attach net_pcap0
2077 Attaching a new port...
2078 PMD: Initializing pmd_pcap for net_pcap0
2079 PMD: Creating pcap-backed ethdev on numa socket 0
2080 Port 0 is attached. Now total ports is 1
2083 In this case, identifier is ``net_pcap0``.
2084 This identifier format is the same as ``--vdev`` format of DPDK applications.
2086 For example, to re-attach a bonded port which has been previously detached,
2087 the mode and slave parameters must be given.
2089 .. code-block:: console
2091 testpmd> port attach net_bond_0,mode=0,slave=1
2092 Attaching a new port...
2093 EAL: Initializing pmd_bond for net_bond_0
2094 EAL: Create bonded device net_bond_0 on port 0 in mode 0 on socket 0.
2095 Port 0 is attached. Now total ports is 1
2102 Detach a specific port::
2104 testpmd> port detach (port_id)
2106 Before detaching a port, the port should be stopped and closed.
2108 For example, to detach a pci device port 0.
2110 .. code-block:: console
2112 testpmd> port stop 0
2115 testpmd> port close 0
2119 testpmd> port detach 0
2121 EAL: PCI device 0000:0a:00.0 on NUMA socket -1
2122 EAL: remove driver: 8086:10fb rte_ixgbe_pmd
2123 EAL: PCI memory unmapped at 0x7f83bfa00000
2124 EAL: PCI memory unmapped at 0x7f83bfa80000
2128 For example, to detach a virtual device port 0.
2130 .. code-block:: console
2132 testpmd> port stop 0
2135 testpmd> port close 0
2139 testpmd> port detach 0
2141 PMD: Closing pcap ethdev on numa socket 0
2142 Port 'net_pcap0' is detached. Now total ports is 0
2145 To remove a pci device completely from the system, first detach the port from testpmd.
2146 Then the device should be moved under kernel management.
2147 Finally the device can be removed using kernel pci hotplug functionality.
2149 For example, to move a pci device under kernel management:
2151 .. code-block:: console
2153 sudo ./usertools/dpdk-devbind.py -b ixgbe 0000:0a:00.0
2155 ./usertools/dpdk-devbind.py --status
2157 Network devices using DPDK-compatible driver
2158 ============================================
2161 Network devices using kernel driver
2162 ===================================
2163 0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused=igb_uio
2165 To remove a port created by a virtual device, above steps are not needed.
2170 Start all ports or a specific port::
2172 testpmd> port start (port_id|all)
2177 Stop all ports or a specific port::
2179 testpmd> port stop (port_id|all)
2184 Close all ports or a specific port::
2186 testpmd> port close (port_id|all)
2191 Reset all ports or a specific port::
2193 testpmd> port reset (port_id|all)
2195 User should stop port(s) before resetting and (re-)start after reset.
2197 port config - queue ring size
2198 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2200 Configure a rx/tx queue ring size::
2202 testpmd> port (port_id) (rxq|txq) (queue_id) ring_size (value)
2204 Only take effect after command that (re-)start the port or command that setup specific queue.
2206 port start/stop queue
2207 ~~~~~~~~~~~~~~~~~~~~~
2209 Start/stop a rx/tx queue on a specific port::
2211 testpmd> port (port_id) (rxq|txq) (queue_id) (start|stop)
2213 port config - queue deferred start
2214 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2216 Switch on/off deferred start of a specific port queue::
2218 testpmd> port (port_id) (rxq|txq) (queue_id) deferred_start (on|off)
2221 ~~~~~~~~~~~~~~~~~~~~~
2223 Setup a rx/tx queue on a specific port::
2225 testpmd> port (port_id) (rxq|txq) (queue_id) setup
2227 Only take effect when port is started.
2232 Set the speed and duplex mode for all ports or a specific port::
2234 testpmd> port config (port_id|all) speed (10|100|1000|10000|25000|40000|50000|100000|200000|auto) \
2235 duplex (half|full|auto)
2237 port config - queues/descriptors
2238 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2240 Set number of queues/descriptors for rxq, txq, rxd and txd::
2242 testpmd> port config all (rxq|txq|rxd|txd) (value)
2244 This is equivalent to the ``--rxq``, ``--txq``, ``--rxd`` and ``--txd`` command-line options.
2246 port config - max-pkt-len
2247 ~~~~~~~~~~~~~~~~~~~~~~~~~
2249 Set the maximum packet length::
2251 testpmd> port config all max-pkt-len (value)
2253 This is equivalent to the ``--max-pkt-len`` command-line option.
2255 port config - max-lro-pkt-size
2256 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2258 Set the maximum LRO aggregated packet size::
2260 testpmd> port config all max-lro-pkt-size (value)
2262 This is equivalent to the ``--max-lro-pkt-size`` command-line option.
2264 port config - Drop Packets
2265 ~~~~~~~~~~~~~~~~~~~~~~~~~~
2267 Enable or disable packet drop on all RX queues of all ports when no receive buffers available::
2269 testpmd> port config all drop-en (on|off)
2271 Packet dropping when no receive buffers available is off by default.
2273 The ``on`` option is equivalent to the ``--enable-drop-en`` command-line option.
2278 Set the RSS (Receive Side Scaling) mode on or off::
2280 testpmd> port config all rss (all|default|eth|vlan|ip|tcp|udp|sctp|ether|port|vxlan|geneve|nvgre|vxlan-gpe|l2tpv3|esp|ah|pfcp|none)
2282 RSS is on by default.
2284 The ``all`` option is equivalent to eth|vlan|ip|tcp|udp|sctp|ether|l2tpv3|esp|ah|pfcp.
2286 The ``default`` option enables all supported RSS types reported by device info.
2288 The ``none`` option is equivalent to the ``--disable-rss`` command-line option.
2290 port config - RSS Reta
2291 ~~~~~~~~~~~~~~~~~~~~~~
2293 Set the RSS (Receive Side Scaling) redirection table::
2295 testpmd> port config all rss reta (hash,queue)[,(hash,queue)]
2300 Set the DCB mode for an individual port::
2302 testpmd> port config (port_id) dcb vt (on|off) (traffic_class) pfc (on|off)
2304 The traffic class should be 4 or 8.
2309 Set the number of packets per burst::
2311 testpmd> port config all burst (value)
2313 This is equivalent to the ``--burst`` command-line option.
2315 port config - Threshold
2316 ~~~~~~~~~~~~~~~~~~~~~~~
2318 Set thresholds for TX/RX queues::
2320 testpmd> port config all (threshold) (value)
2322 Where the threshold type can be:
2324 * ``txpt:`` Set the prefetch threshold register of the TX rings, 0 <= value <= 255.
2326 * ``txht:`` Set the host threshold register of the TX rings, 0 <= value <= 255.
2328 * ``txwt:`` Set the write-back threshold register of the TX rings, 0 <= value <= 255.
2330 * ``rxpt:`` Set the prefetch threshold register of the RX rings, 0 <= value <= 255.
2332 * ``rxht:`` Set the host threshold register of the RX rings, 0 <= value <= 255.
2334 * ``rxwt:`` Set the write-back threshold register of the RX rings, 0 <= value <= 255.
2336 * ``txfreet:`` Set the transmit free threshold of the TX rings, 0 <= value <= txd.
2338 * ``rxfreet:`` Set the transmit free threshold of the RX rings, 0 <= value <= rxd.
2340 * ``txrst:`` Set the transmit RS bit threshold of TX rings, 0 <= value <= txd.
2342 These threshold options are also available from the command-line.
2347 Set the value of ether-type for E-tag::
2349 testpmd> port config (port_id|all) l2-tunnel E-tag ether-type (value)
2351 Enable/disable the E-tag support::
2353 testpmd> port config (port_id|all) l2-tunnel E-tag (enable|disable)
2355 port config pctype mapping
2356 ~~~~~~~~~~~~~~~~~~~~~~~~~~
2358 Reset pctype mapping table::
2360 testpmd> port config (port_id) pctype mapping reset
2362 Update hardware defined pctype to software defined flow type mapping table::
2364 testpmd> port config (port_id) pctype mapping update (pctype_id_0[,pctype_id_1]*) (flow_type_id)
2368 * ``pctype_id_x``: hardware pctype id as index of bit in bitmask value of the pctype mapping table.
2370 * ``flow_type_id``: software flow type id as the index of the pctype mapping table.
2372 port config input set
2373 ~~~~~~~~~~~~~~~~~~~~~
2375 Config RSS/FDIR/FDIR flexible payload input set for some pctype::
2377 testpmd> port config (port_id) pctype (pctype_id) \
2378 (hash_inset|fdir_inset|fdir_flx_inset) \
2379 (get|set|clear) field (field_idx)
2381 Clear RSS/FDIR/FDIR flexible payload input set for some pctype::
2383 testpmd> port config (port_id) pctype (pctype_id) \
2384 (hash_inset|fdir_inset|fdir_flx_inset) clear all
2388 * ``pctype_id``: hardware packet classification types.
2389 * ``field_idx``: hardware field index.
2391 port config udp_tunnel_port
2392 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
2394 Add/remove UDP tunnel port for VXLAN/GENEVE tunneling protocols::
2396 testpmd> port config (port_id) udp_tunnel_port add|rm vxlan|geneve|vxlan-gpe (udp_port)
2398 port config tx_metadata
2399 ~~~~~~~~~~~~~~~~~~~~~~~
2401 Set Tx metadata value per port.
2402 testpmd will add this value to any Tx packet sent from this port::
2404 testpmd> port config (port_id) tx_metadata (value)
2409 Set/clear dynamic flag per port.
2410 testpmd will register this flag in the mbuf (same registration
2411 for both Tx and Rx). Then set/clear this flag for each Tx
2412 packet sent from this port. The set bit only works for Tx packet::
2414 testpmd> port config (port_id) dynf (name) (set|clear)
2419 To configure MTU(Maximum Transmission Unit) on devices using testpmd::
2421 testpmd> port config mtu (port_id) (value)
2423 port config rss hash key
2424 ~~~~~~~~~~~~~~~~~~~~~~~~
2426 To configure the RSS hash key used to compute the RSS
2427 hash of input [IP] packets received on port::
2429 testpmd> port config <port_id> rss-hash-key (ipv4|ipv4-frag|\
2430 ipv4-tcp|ipv4-udp|ipv4-sctp|ipv4-other|\
2431 ipv6|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|\
2432 ipv6-other|l2-payload|ipv6-ex|ipv6-tcp-ex|\
2433 ipv6-udp-ex <string of hex digits \
2434 (variable length, NIC dependent)>)
2439 The following sections show functions for device operations.
2444 Detach a device specified by pci address or virtual device args::
2446 testpmd> device detach (identifier)
2448 Before detaching a device associated with ports, the ports should be stopped and closed.
2450 For example, to detach a pci device whose address is 0002:03:00.0.
2452 .. code-block:: console
2454 testpmd> device detach 0002:03:00.0
2455 Removing a device...
2456 Port 1 is now closed
2457 EAL: Releasing pci mapped resource for 0002:03:00.0
2458 EAL: Calling pci_unmap_resource for 0002:03:00.0 at 0x218a050000
2459 EAL: Calling pci_unmap_resource for 0002:03:00.0 at 0x218c050000
2460 Device 0002:03:00.0 is detached
2461 Now total ports is 1
2463 For example, to detach a port created by pcap PMD.
2465 .. code-block:: console
2467 testpmd> device detach net_pcap0
2468 Removing a device...
2469 Port 0 is now closed
2470 Device net_pcap0 is detached
2471 Now total ports is 0
2474 In this case, identifier is ``net_pcap0``.
2475 This identifier format is the same as ``--vdev`` format of DPDK applications.
2477 Link Bonding Functions
2478 ----------------------
2480 The Link Bonding functions make it possible to dynamically create and
2481 manage link bonding devices from within testpmd interactive prompt.
2483 create bonded device
2484 ~~~~~~~~~~~~~~~~~~~~
2486 Create a new bonding device::
2488 testpmd> create bonded device (mode) (socket)
2490 For example, to create a bonded device in mode 1 on socket 0::
2492 testpmd> create bonded device 1 0
2493 created new bonded device (port X)
2498 Adds Ethernet device to a Link Bonding device::
2500 testpmd> add bonding slave (slave id) (port id)
2502 For example, to add Ethernet device (port 6) to a Link Bonding device (port 10)::
2504 testpmd> add bonding slave 6 10
2507 remove bonding slave
2508 ~~~~~~~~~~~~~~~~~~~~
2510 Removes an Ethernet slave device from a Link Bonding device::
2512 testpmd> remove bonding slave (slave id) (port id)
2514 For example, to remove Ethernet slave device (port 6) to a Link Bonding device (port 10)::
2516 testpmd> remove bonding slave 6 10
2521 Set the Link Bonding mode of a Link Bonding device::
2523 testpmd> set bonding mode (value) (port id)
2525 For example, to set the bonding mode of a Link Bonding device (port 10) to broadcast (mode 3)::
2527 testpmd> set bonding mode 3 10
2532 Set an Ethernet slave device as the primary device on a Link Bonding device::
2534 testpmd> set bonding primary (slave id) (port id)
2536 For example, to set the Ethernet slave device (port 6) as the primary port of a Link Bonding device (port 10)::
2538 testpmd> set bonding primary 6 10
2543 Set the MAC address of a Link Bonding device::
2545 testpmd> set bonding mac (port id) (mac)
2547 For example, to set the MAC address of a Link Bonding device (port 10) to 00:00:00:00:00:01::
2549 testpmd> set bonding mac 10 00:00:00:00:00:01
2551 set bonding balance_xmit_policy
2552 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2554 Set the transmission policy for a Link Bonding device when it is in Balance XOR mode::
2556 testpmd> set bonding balance_xmit_policy (port_id) (l2|l23|l34)
2558 For example, set a Link Bonding device (port 10) to use a balance policy of layer 3+4 (IP addresses & UDP ports)::
2560 testpmd> set bonding balance_xmit_policy 10 l34
2563 set bonding mon_period
2564 ~~~~~~~~~~~~~~~~~~~~~~
2566 Set the link status monitoring polling period in milliseconds for a bonding device.
2568 This adds support for PMD slave devices which do not support link status interrupts.
2569 When the mon_period is set to a value greater than 0 then all PMD's which do not support
2570 link status ISR will be queried every polling interval to check if their link status has changed::
2572 testpmd> set bonding mon_period (port_id) (value)
2574 For example, to set the link status monitoring polling period of bonded device (port 5) to 150ms::
2576 testpmd> set bonding mon_period 5 150
2579 set bonding lacp dedicated_queue
2580 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2582 Enable dedicated tx/rx queues on bonding devices slaves to handle LACP control plane traffic
2583 when in mode 4 (link-aggregation-802.3ad)::
2585 testpmd> set bonding lacp dedicated_queues (port_id) (enable|disable)
2588 set bonding agg_mode
2589 ~~~~~~~~~~~~~~~~~~~~
2591 Enable one of the specific aggregators mode when in mode 4 (link-aggregation-802.3ad)::
2593 testpmd> set bonding agg_mode (port_id) (bandwidth|count|stable)
2599 Show the current configuration of a Link Bonding device::
2601 testpmd> show bonding config (port id)
2604 to show the configuration a Link Bonding device (port 9) with 3 slave devices (1, 3, 4)
2605 in balance mode with a transmission policy of layer 2+3::
2607 testpmd> show bonding config 9
2609 Balance Xmit Policy: BALANCE_XMIT_POLICY_LAYER23
2611 Active Slaves (3): [1 3 4]
2618 The Register Functions can be used to read from and write to registers on the network card referenced by a port number.
2619 This is mainly useful for debugging purposes.
2620 Reference should be made to the appropriate datasheet for the network card for details on the register addresses
2621 and fields that can be accessed.
2626 Display the value of a port register::
2628 testpmd> read reg (port_id) (address)
2630 For example, to examine the Flow Director control register (FDIRCTL, 0x0000EE000) on an Intel 82599 10 GbE Controller::
2632 testpmd> read reg 0 0xEE00
2633 port 0 PCI register at offset 0xEE00: 0x4A060029 (1241907241)
2638 Display a port register bit field::
2640 testpmd> read regfield (port_id) (address) (bit_x) (bit_y)
2642 For example, reading the lowest two bits from the register in the example above::
2644 testpmd> read regfield 0 0xEE00 0 1
2645 port 0 PCI register at offset 0xEE00: bits[0, 1]=0x1 (1)
2650 Display a single port register bit::
2652 testpmd> read regbit (port_id) (address) (bit_x)
2654 For example, reading the lowest bit from the register in the example above::
2656 testpmd> read regbit 0 0xEE00 0
2657 port 0 PCI register at offset 0xEE00: bit 0=1
2662 Set the value of a port register::
2664 testpmd> write reg (port_id) (address) (value)
2666 For example, to clear a register::
2668 testpmd> write reg 0 0xEE00 0x0
2669 port 0 PCI register at offset 0xEE00: 0x00000000 (0)
2674 Set bit field of a port register::
2676 testpmd> write regfield (port_id) (address) (bit_x) (bit_y) (value)
2678 For example, writing to the register cleared in the example above::
2680 testpmd> write regfield 0 0xEE00 0 1 2
2681 port 0 PCI register at offset 0xEE00: 0x00000002 (2)
2686 Set single bit value of a port register::
2688 testpmd> write regbit (port_id) (address) (bit_x) (value)
2690 For example, to set the high bit in the register from the example above::
2692 testpmd> write regbit 0 0xEE00 31 1
2693 port 0 PCI register at offset 0xEE00: 0x8000000A (2147483658)
2695 Traffic Metering and Policing
2696 -----------------------------
2698 The following section shows functions for configuring traffic metering and
2699 policing on the ethernet device through the use of generic ethdev API.
2701 show port traffic management capability
2702 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2704 Show traffic metering and policing capability of the port::
2706 testpmd> show port meter cap (port_id)
2708 add port meter profile (srTCM rfc2967)
2709 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2711 Add meter profile (srTCM rfc2697) to the ethernet device::
2713 testpmd> add port meter profile srtcm_rfc2697 (port_id) (profile_id) \
2718 * ``profile_id``: ID for the meter profile.
2719 * ``cir``: Committed Information Rate (CIR) (bytes/second).
2720 * ``cbs``: Committed Burst Size (CBS) (bytes).
2721 * ``ebs``: Excess Burst Size (EBS) (bytes).
2723 add port meter profile (trTCM rfc2968)
2724 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2726 Add meter profile (srTCM rfc2698) to the ethernet device::
2728 testpmd> add port meter profile trtcm_rfc2698 (port_id) (profile_id) \
2729 (cir) (pir) (cbs) (pbs)
2733 * ``profile_id``: ID for the meter profile.
2734 * ``cir``: Committed information rate (bytes/second).
2735 * ``pir``: Peak information rate (bytes/second).
2736 * ``cbs``: Committed burst size (bytes).
2737 * ``pbs``: Peak burst size (bytes).
2739 add port meter profile (trTCM rfc4115)
2740 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2742 Add meter profile (trTCM rfc4115) to the ethernet device::
2744 testpmd> add port meter profile trtcm_rfc4115 (port_id) (profile_id) \
2745 (cir) (eir) (cbs) (ebs)
2749 * ``profile_id``: ID for the meter profile.
2750 * ``cir``: Committed information rate (bytes/second).
2751 * ``eir``: Excess information rate (bytes/second).
2752 * ``cbs``: Committed burst size (bytes).
2753 * ``ebs``: Excess burst size (bytes).
2755 delete port meter profile
2756 ~~~~~~~~~~~~~~~~~~~~~~~~~
2758 Delete meter profile from the ethernet device::
2760 testpmd> del port meter profile (port_id) (profile_id)
2765 Create new meter object for the ethernet device::
2767 testpmd> create port meter (port_id) (mtr_id) (profile_id) \
2768 (meter_enable) (g_action) (y_action) (r_action) (stats_mask) (shared) \
2769 (use_pre_meter_color) [(dscp_tbl_entry0) (dscp_tbl_entry1)...\
2774 * ``mtr_id``: meter object ID.
2775 * ``profile_id``: ID for the meter profile.
2776 * ``meter_enable``: When this parameter has a non-zero value, the meter object
2777 gets enabled at the time of creation, otherwise remains disabled.
2778 * ``g_action``: Policer action for the packet with green color.
2779 * ``y_action``: Policer action for the packet with yellow color.
2780 * ``r_action``: Policer action for the packet with red color.
2781 * ``stats_mask``: Mask of statistics counter types to be enabled for the
2783 * ``shared``: When this parameter has a non-zero value, the meter object is
2784 shared by multiple flows. Otherwise, meter object is used by single flow.
2785 * ``use_pre_meter_color``: When this parameter has a non-zero value, the
2786 input color for the current meter object is determined by the latest meter
2787 object in the same flow. Otherwise, the current meter object uses the
2788 *dscp_table* to determine the input color.
2789 * ``dscp_tbl_entryx``: DSCP table entry x providing meter providing input
2790 color, 0 <= x <= 63.
2795 Enable meter for the ethernet device::
2797 testpmd> enable port meter (port_id) (mtr_id)
2802 Disable meter for the ethernet device::
2804 testpmd> disable port meter (port_id) (mtr_id)
2809 Delete meter for the ethernet device::
2811 testpmd> del port meter (port_id) (mtr_id)
2813 Set port meter profile
2814 ~~~~~~~~~~~~~~~~~~~~~~
2816 Set meter profile for the ethernet device::
2818 testpmd> set port meter profile (port_id) (mtr_id) (profile_id)
2820 set port meter dscp table
2821 ~~~~~~~~~~~~~~~~~~~~~~~~~
2823 Set meter dscp table for the ethernet device::
2825 testpmd> set port meter dscp table (port_id) (mtr_id) [(dscp_tbl_entry0) \
2826 (dscp_tbl_entry1)...(dscp_tbl_entry63)]
2828 set port meter policer action
2829 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2831 Set meter policer action for the ethernet device::
2833 testpmd> set port meter policer action (port_id) (mtr_id) (action_mask) \
2834 (action0) [(action1) (action1)]
2838 * ``action_mask``: Bit mask indicating which policer actions need to be
2839 updated. One or more policer actions can be updated in a single function
2840 invocation. To update the policer action associated with color C, bit
2841 (1 << C) needs to be set in *action_mask* and element at position C
2842 in the *actions* array needs to be valid.
2843 * ``actionx``: Policer action for the color x,
2844 RTE_MTR_GREEN <= x < RTE_MTR_COLORS
2846 set port meter stats mask
2847 ~~~~~~~~~~~~~~~~~~~~~~~~~
2849 Set meter stats mask for the ethernet device::
2851 testpmd> set port meter stats mask (port_id) (mtr_id) (stats_mask)
2855 * ``stats_mask``: Bit mask indicating statistics counter types to be enabled.
2857 show port meter stats
2858 ~~~~~~~~~~~~~~~~~~~~~
2860 Show meter stats of the ethernet device::
2862 testpmd> show port meter stats (port_id) (mtr_id) (clear)
2866 * ``clear``: Flag that indicates whether the statistics counters should
2867 be cleared (i.e. set to zero) immediately after they have been read or not.
2872 The following section shows functions for configuring traffic management on
2873 the ethernet device through the use of generic TM API.
2875 show port traffic management capability
2876 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2878 Show traffic management capability of the port::
2880 testpmd> show port tm cap (port_id)
2882 show port traffic management capability (hierarchy level)
2883 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2885 Show traffic management hierarchy level capability of the port::
2887 testpmd> show port tm level cap (port_id) (level_id)
2889 show port traffic management capability (hierarchy node level)
2890 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2892 Show the traffic management hierarchy node capability of the port::
2894 testpmd> show port tm node cap (port_id) (node_id)
2896 show port traffic management hierarchy node type
2897 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2899 Show the port traffic management hierarchy node type::
2901 testpmd> show port tm node type (port_id) (node_id)
2903 show port traffic management hierarchy node stats
2904 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2906 Show the port traffic management hierarchy node statistics::
2908 testpmd> show port tm node stats (port_id) (node_id) (clear)
2912 * ``clear``: When this parameter has a non-zero value, the statistics counters
2913 are cleared (i.e. set to zero) immediately after they have been read,
2914 otherwise the statistics counters are left untouched.
2916 Add port traffic management private shaper profile
2917 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2919 Add the port traffic management private shaper profile::
2921 testpmd> add port tm node shaper profile (port_id) (shaper_profile_id) \
2922 (cmit_tb_rate) (cmit_tb_size) (peak_tb_rate) (peak_tb_size) \
2923 (packet_length_adjust) (packet_mode)
2927 * ``shaper_profile id``: Shaper profile ID for the new profile.
2928 * ``cmit_tb_rate``: Committed token bucket rate (bytes per second or packets per second).
2929 * ``cmit_tb_size``: Committed token bucket size (bytes or packets).
2930 * ``peak_tb_rate``: Peak token bucket rate (bytes per second or packets per second).
2931 * ``peak_tb_size``: Peak token bucket size (bytes or packets).
2932 * ``packet_length_adjust``: The value (bytes) to be added to the length of
2933 each packet for the purpose of shaping. This parameter value can be used to
2934 correct the packet length with the framing overhead bytes that are consumed
2936 * ``packet_mode``: Shaper configured in packet mode. This parameter value if
2937 zero, configures shaper in byte mode and if non-zero configures it in packet
2940 Delete port traffic management private shaper profile
2941 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2943 Delete the port traffic management private shaper::
2945 testpmd> del port tm node shaper profile (port_id) (shaper_profile_id)
2949 * ``shaper_profile id``: Shaper profile ID that needs to be deleted.
2951 Add port traffic management shared shaper
2952 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2954 Create the port traffic management shared shaper::
2956 testpmd> add port tm node shared shaper (port_id) (shared_shaper_id) \
2961 * ``shared_shaper_id``: Shared shaper ID to be created.
2962 * ``shaper_profile id``: Shaper profile ID for shared shaper.
2964 Set port traffic management shared shaper
2965 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2967 Update the port traffic management shared shaper::
2969 testpmd> set port tm node shared shaper (port_id) (shared_shaper_id) \
2974 * ``shared_shaper_id``: Shared shaper ID to be update.
2975 * ``shaper_profile id``: Shaper profile ID for shared shaper.
2977 Delete port traffic management shared shaper
2978 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2980 Delete the port traffic management shared shaper::
2982 testpmd> del port tm node shared shaper (port_id) (shared_shaper_id)
2986 * ``shared_shaper_id``: Shared shaper ID to be deleted.
2988 Set port traffic management hierarchy node private shaper
2989 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2991 set the port traffic management hierarchy node private shaper::
2993 testpmd> set port tm node shaper profile (port_id) (node_id) \
2998 * ``shaper_profile id``: Private shaper profile ID to be enabled on the
3001 Add port traffic management WRED profile
3002 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3004 Create a new WRED profile::
3006 testpmd> add port tm node wred profile (port_id) (wred_profile_id) \
3007 (color_g) (min_th_g) (max_th_g) (maxp_inv_g) (wq_log2_g) \
3008 (color_y) (min_th_y) (max_th_y) (maxp_inv_y) (wq_log2_y) \
3009 (color_r) (min_th_r) (max_th_r) (maxp_inv_r) (wq_log2_r)
3013 * ``wred_profile id``: Identifier for the newly create WRED profile
3014 * ``color_g``: Packet color (green)
3015 * ``min_th_g``: Minimum queue threshold for packet with green color
3016 * ``max_th_g``: Minimum queue threshold for packet with green color
3017 * ``maxp_inv_g``: Inverse of packet marking probability maximum value (maxp)
3018 * ``wq_log2_g``: Negated log2 of queue weight (wq)
3019 * ``color_y``: Packet color (yellow)
3020 * ``min_th_y``: Minimum queue threshold for packet with yellow color
3021 * ``max_th_y``: Minimum queue threshold for packet with yellow color
3022 * ``maxp_inv_y``: Inverse of packet marking probability maximum value (maxp)
3023 * ``wq_log2_y``: Negated log2 of queue weight (wq)
3024 * ``color_r``: Packet color (red)
3025 * ``min_th_r``: Minimum queue threshold for packet with yellow color
3026 * ``max_th_r``: Minimum queue threshold for packet with yellow color
3027 * ``maxp_inv_r``: Inverse of packet marking probability maximum value (maxp)
3028 * ``wq_log2_r``: Negated log2 of queue weight (wq)
3030 Delete port traffic management WRED profile
3031 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3033 Delete the WRED profile::
3035 testpmd> del port tm node wred profile (port_id) (wred_profile_id)
3037 Add port traffic management hierarchy nonleaf node
3038 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3040 Add nonleaf node to port traffic management hierarchy::
3042 testpmd> add port tm nonleaf node (port_id) (node_id) (parent_node_id) \
3043 (priority) (weight) (level_id) (shaper_profile_id) \
3044 (n_sp_priorities) (stats_mask) (n_shared_shapers) \
3045 [(shared_shaper_0) (shared_shaper_1) ...] \
3049 * ``parent_node_id``: Node ID of the parent.
3050 * ``priority``: Node priority (highest node priority is zero). This is used by
3051 the SP algorithm running on the parent node for scheduling this node.
3052 * ``weight``: Node weight (lowest weight is one). The node weight is relative
3053 to the weight sum of all siblings that have the same priority. It is used by
3054 the WFQ algorithm running on the parent node for scheduling this node.
3055 * ``level_id``: Hierarchy level of the node.
3056 * ``shaper_profile_id``: Shaper profile ID of the private shaper to be used by
3058 * ``n_sp_priorities``: Number of strict priorities.
3059 * ``stats_mask``: Mask of statistics counter types to be enabled for this node.
3060 * ``n_shared_shapers``: Number of shared shapers.
3061 * ``shared_shaper_id``: Shared shaper id.
3063 Add port traffic management hierarchy nonleaf node with packet mode
3064 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3066 Add nonleaf node with packet mode to port traffic management hierarchy::
3068 testpmd> add port tm nonleaf node pktmode (port_id) (node_id) (parent_node_id) \
3069 (priority) (weight) (level_id) (shaper_profile_id) \
3070 (n_sp_priorities) (stats_mask) (n_shared_shapers) \
3071 [(shared_shaper_0) (shared_shaper_1) ...] \
3075 * ``parent_node_id``: Node ID of the parent.
3076 * ``priority``: Node priority (highest node priority is zero). This is used by
3077 the SP algorithm running on the parent node for scheduling this node.
3078 * ``weight``: Node weight (lowest weight is one). The node weight is relative
3079 to the weight sum of all siblings that have the same priority. It is used by
3080 the WFQ algorithm running on the parent node for scheduling this node.
3081 * ``level_id``: Hierarchy level of the node.
3082 * ``shaper_profile_id``: Shaper profile ID of the private shaper to be used by
3084 * ``n_sp_priorities``: Number of strict priorities. Packet mode is enabled on
3086 * ``stats_mask``: Mask of statistics counter types to be enabled for this node.
3087 * ``n_shared_shapers``: Number of shared shapers.
3088 * ``shared_shaper_id``: Shared shaper id.
3090 Add port traffic management hierarchy leaf node
3091 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3093 Add leaf node to port traffic management hierarchy::
3095 testpmd> add port tm leaf node (port_id) (node_id) (parent_node_id) \
3096 (priority) (weight) (level_id) (shaper_profile_id) \
3097 (cman_mode) (wred_profile_id) (stats_mask) (n_shared_shapers) \
3098 [(shared_shaper_id) (shared_shaper_id) ...] \
3102 * ``parent_node_id``: Node ID of the parent.
3103 * ``priority``: Node priority (highest node priority is zero). This is used by
3104 the SP algorithm running on the parent node for scheduling this node.
3105 * ``weight``: Node weight (lowest weight is one). The node weight is relative
3106 to the weight sum of all siblings that have the same priority. It is used by
3107 the WFQ algorithm running on the parent node for scheduling this node.
3108 * ``level_id``: Hierarchy level of the node.
3109 * ``shaper_profile_id``: Shaper profile ID of the private shaper to be used by
3111 * ``cman_mode``: Congestion management mode to be enabled for this node.
3112 * ``wred_profile_id``: WRED profile id to be enabled for this node.
3113 * ``stats_mask``: Mask of statistics counter types to be enabled for this node.
3114 * ``n_shared_shapers``: Number of shared shapers.
3115 * ``shared_shaper_id``: Shared shaper id.
3117 Delete port traffic management hierarchy node
3118 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3120 Delete node from port traffic management hierarchy::
3122 testpmd> del port tm node (port_id) (node_id)
3124 Update port traffic management hierarchy parent node
3125 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3127 Update port traffic management hierarchy parent node::
3129 testpmd> set port tm node parent (port_id) (node_id) (parent_node_id) \
3132 This function can only be called after the hierarchy commit invocation. Its
3133 success depends on the port support for this operation, as advertised through
3134 the port capability set. This function is valid for all nodes of the traffic
3135 management hierarchy except root node.
3137 Suspend port traffic management hierarchy node
3138 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3140 testpmd> suspend port tm node (port_id) (node_id)
3142 Resume port traffic management hierarchy node
3143 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3145 testpmd> resume port tm node (port_id) (node_id)
3147 Commit port traffic management hierarchy
3148 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3150 Commit the traffic management hierarchy on the port::
3152 testpmd> port tm hierarchy commit (port_id) (clean_on_fail)
3156 * ``clean_on_fail``: When set to non-zero, hierarchy is cleared on function
3157 call failure. On the other hand, hierarchy is preserved when this parameter
3160 Set port traffic management mark VLAN dei
3161 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3163 Enables/Disables the traffic management marking on the port for VLAN packets::
3165 testpmd> set port tm mark vlan_dei <port_id> <green> <yellow> <red>
3169 * ``port_id``: The port which on which VLAN packets marked as ``green`` or
3170 ``yellow`` or ``red`` will have dei bit enabled
3172 * ``green`` enable 1, disable 0 marking for dei bit of VLAN packets marked as green
3174 * ``yellow`` enable 1, disable 0 marking for dei bit of VLAN packets marked as yellow
3176 * ``red`` enable 1, disable 0 marking for dei bit of VLAN packets marked as red
3178 Set port traffic management mark IP dscp
3179 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3181 Enables/Disables the traffic management marking on the port for IP dscp packets::
3183 testpmd> set port tm mark ip_dscp <port_id> <green> <yellow> <red>
3187 * ``port_id``: The port which on which IP packets marked as ``green`` or
3188 ``yellow`` or ``red`` will have IP dscp bits updated
3190 * ``green`` enable 1, disable 0 marking IP dscp to low drop precedence for green packets
3192 * ``yellow`` enable 1, disable 0 marking IP dscp to medium drop precedence for yellow packets
3194 * ``red`` enable 1, disable 0 marking IP dscp to high drop precedence for red packets
3196 Set port traffic management mark IP ecn
3197 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3199 Enables/Disables the traffic management marking on the port for IP ecn packets::
3201 testpmd> set port tm mark ip_ecn <port_id> <green> <yellow> <red>
3205 * ``port_id``: The port which on which IP packets marked as ``green`` or
3206 ``yellow`` or ``red`` will have IP ecn bits updated
3208 * ``green`` enable 1, disable 0 marking IP ecn for green marked packets with ecn of 2'b01 or 2'b10
3209 to ecn of 2'b11 when IP is caring TCP or SCTP
3211 * ``yellow`` enable 1, disable 0 marking IP ecn for yellow marked packets with ecn of 2'b01 or 2'b10
3212 to ecn of 2'b11 when IP is caring TCP or SCTP
3214 * ``red`` enable 1, disable 0 marking IP ecn for yellow marked packets with ecn of 2'b01 or 2'b10
3215 to ecn of 2'b11 when IP is caring TCP or SCTP
3220 This section details the available filter functions that are available.
3222 Note these functions interface the deprecated legacy filtering framework,
3223 superseded by *rte_flow*. See `Flow rules management`_.
3225 .. _testpmd_flow_director:
3227 flow_director_filter
3228 ~~~~~~~~~~~~~~~~~~~~
3230 The Flow Director works in receive mode to identify specific flows or sets of flows and route them to specific queues.
3232 Four types of filtering are supported which are referred to as Perfect Match, Signature, Perfect-mac-vlan and
3233 Perfect-tunnel filters, the match mode is set by the ``--pkt-filter-mode`` command-line parameter:
3235 * Perfect match filters.
3236 The hardware checks a match between the masked fields of the received packets and the programmed filters.
3237 The masked fields are for IP flow.
3239 * Signature filters.
3240 The hardware checks a match between a hash-based signature of the masked fields of the received packet.
3242 * Perfect-mac-vlan match filters.
3243 The hardware checks a match between the masked fields of the received packets and the programmed filters.
3244 The masked fields are for MAC VLAN flow.
3246 * Perfect-tunnel match filters.
3247 The hardware checks a match between the masked fields of the received packets and the programmed filters.
3248 The masked fields are for tunnel flow.
3250 * Perfect-raw-flow-type match filters.
3251 The hardware checks a match between the masked fields of the received packets and pre-loaded raw (template) packet.
3252 The masked fields are specified by input sets.
3254 The Flow Director filters can match the different fields for different type of packet: flow type, specific input set
3255 per flow type and the flexible payload.
3257 The Flow Director can also mask out parts of all of these fields so that filters
3258 are only applied to certain fields or parts of the fields.
3260 Note that for raw flow type mode the source and destination fields in the
3261 raw packet buffer need to be presented in a reversed order with respect
3262 to the expected received packets.
3263 For example: IP source and destination addresses or TCP/UDP/SCTP
3264 source and destination ports
3266 Different NICs may have different capabilities, command show port fdir (port_id) can be used to acquire the information.
3268 # Commands to add flow director filters of different flow types::
3270 flow_director_filter (port_id) mode IP (add|del|update) \
3271 flow (ipv4-other|ipv4-frag|ipv6-other|ipv6-frag) \
3272 src (src_ip_address) dst (dst_ip_address) \
3273 tos (tos_value) proto (proto_value) ttl (ttl_value) \
3274 vlan (vlan_value) flexbytes (flexbytes_value) \
3275 (drop|fwd) pf|vf(vf_id) queue (queue_id) \
3278 flow_director_filter (port_id) mode IP (add|del|update) \
3279 flow (ipv4-tcp|ipv4-udp|ipv6-tcp|ipv6-udp) \
3280 src (src_ip_address) (src_port) \
3281 dst (dst_ip_address) (dst_port) \
3282 tos (tos_value) ttl (ttl_value) \
3283 vlan (vlan_value) flexbytes (flexbytes_value) \
3284 (drop|fwd) queue pf|vf(vf_id) (queue_id) \
3287 flow_director_filter (port_id) mode IP (add|del|update) \
3288 flow (ipv4-sctp|ipv6-sctp) \
3289 src (src_ip_address) (src_port) \
3290 dst (dst_ip_address) (dst_port) \
3291 tos (tos_value) ttl (ttl_value) \
3292 tag (verification_tag) vlan (vlan_value) \
3293 flexbytes (flexbytes_value) (drop|fwd) \
3294 pf|vf(vf_id) queue (queue_id) fd_id (fd_id_value)
3296 flow_director_filter (port_id) mode IP (add|del|update) flow l2_payload \
3297 ether (ethertype) flexbytes (flexbytes_value) \
3298 (drop|fwd) pf|vf(vf_id) queue (queue_id)
3301 flow_director_filter (port_id) mode MAC-VLAN (add|del|update) \
3302 mac (mac_address) vlan (vlan_value) \
3303 flexbytes (flexbytes_value) (drop|fwd) \
3304 queue (queue_id) fd_id (fd_id_value)
3306 flow_director_filter (port_id) mode Tunnel (add|del|update) \
3307 mac (mac_address) vlan (vlan_value) \
3308 tunnel (NVGRE|VxLAN) tunnel-id (tunnel_id_value) \
3309 flexbytes (flexbytes_value) (drop|fwd) \
3310 queue (queue_id) fd_id (fd_id_value)
3312 flow_director_filter (port_id) mode raw (add|del|update) flow (flow_id) \
3313 (drop|fwd) queue (queue_id) fd_id (fd_id_value) \
3314 packet (packet file name)
3316 For example, to add an ipv4-udp flow type filter::
3318 testpmd> flow_director_filter 0 mode IP add flow ipv4-udp src 2.2.2.3 32 \
3319 dst 2.2.2.5 33 tos 2 ttl 40 vlan 0x1 flexbytes (0x88,0x48) \
3320 fwd pf queue 1 fd_id 1
3322 For example, add an ipv4-other flow type filter::
3324 testpmd> flow_director_filter 0 mode IP add flow ipv4-other src 2.2.2.3 \
3325 dst 2.2.2.5 tos 2 proto 20 ttl 40 vlan 0x1 \
3326 flexbytes (0x88,0x48) fwd pf queue 1 fd_id 1
3331 Flush all flow director filters on a device::
3333 testpmd> flush_flow_director (port_id)
3335 Example, to flush all flow director filter on port 0::
3337 testpmd> flush_flow_director 0
3342 Set flow director's input masks::
3344 flow_director_mask (port_id) mode IP vlan (vlan_value) \
3345 src_mask (ipv4_src) (ipv6_src) (src_port) \
3346 dst_mask (ipv4_dst) (ipv6_dst) (dst_port)
3348 flow_director_mask (port_id) mode MAC-VLAN vlan (vlan_value)
3350 flow_director_mask (port_id) mode Tunnel vlan (vlan_value) \
3351 mac (mac_value) tunnel-type (tunnel_type_value) \
3352 tunnel-id (tunnel_id_value)
3354 Example, to set flow director mask on port 0::
3356 testpmd> flow_director_mask 0 mode IP vlan 0xefff \
3357 src_mask 255.255.255.255 \
3358 FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF \
3359 dst_mask 255.255.255.255 \
3360 FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF
3362 flow_director_flex_mask
3363 ~~~~~~~~~~~~~~~~~~~~~~~
3365 set masks of flow director's flexible payload based on certain flow type::
3367 testpmd> flow_director_flex_mask (port_id) \
3368 flow (none|ipv4-other|ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
3369 ipv6-other|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp| \
3370 l2_payload|all) (mask)
3372 Example, to set flow director's flex mask for all flow type on port 0::
3374 testpmd> flow_director_flex_mask 0 flow all \
3375 (0xff,0xff,0,0,0,0,0,0,0,0,0,0,0,0,0,0)
3378 flow_director_flex_payload
3379 ~~~~~~~~~~~~~~~~~~~~~~~~~~
3381 Configure flexible payload selection::
3383 flow_director_flex_payload (port_id) (raw|l2|l3|l4) (config)
3385 For example, to select the first 16 bytes from the offset 4 (bytes) of packet's payload as flexible payload::
3387 testpmd> flow_director_flex_payload 0 l4 \
3388 (4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19)
3390 get_sym_hash_ena_per_port
3391 ~~~~~~~~~~~~~~~~~~~~~~~~~
3393 Get symmetric hash enable configuration per port::
3395 get_sym_hash_ena_per_port (port_id)
3397 For example, to get symmetric hash enable configuration of port 1::
3399 testpmd> get_sym_hash_ena_per_port 1
3401 set_sym_hash_ena_per_port
3402 ~~~~~~~~~~~~~~~~~~~~~~~~~
3404 Set symmetric hash enable configuration per port to enable or disable::
3406 set_sym_hash_ena_per_port (port_id) (enable|disable)
3408 For example, to set symmetric hash enable configuration of port 1 to enable::
3410 testpmd> set_sym_hash_ena_per_port 1 enable
3412 get_hash_global_config
3413 ~~~~~~~~~~~~~~~~~~~~~~
3415 Get the global configurations of hash filters::
3417 get_hash_global_config (port_id)
3419 For example, to get the global configurations of hash filters of port 1::
3421 testpmd> get_hash_global_config 1
3423 set_hash_global_config
3424 ~~~~~~~~~~~~~~~~~~~~~~
3426 Set the global configurations of hash filters::
3428 set_hash_global_config (port_id) (toeplitz|simple_xor|symmetric_toeplitz|default) \
3429 (ipv4|ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp|ipv4-other|ipv6|ipv6-frag| \
3430 ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other|l2_payload|<flow_id>) \
3433 For example, to enable simple_xor for flow type of ipv6 on port 2::
3435 testpmd> set_hash_global_config 2 simple_xor ipv6 enable
3440 Set the input set for hash::
3442 set_hash_input_set (port_id) (ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
3443 ipv4-other|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other| \
3444 l2_payload|<flow_id>) (ovlan|ivlan|src-ipv4|dst-ipv4|src-ipv6|dst-ipv6| \
3445 ipv4-tos|ipv4-proto|ipv6-tc|ipv6-next-header|udp-src-port|udp-dst-port| \
3446 tcp-src-port|tcp-dst-port|sctp-src-port|sctp-dst-port|sctp-veri-tag| \
3447 udp-key|gre-key|fld-1st|fld-2nd|fld-3rd|fld-4th|fld-5th|fld-6th|fld-7th| \
3448 fld-8th|none) (select|add)
3450 For example, to add source IP to hash input set for flow type of ipv4-udp on port 0::
3452 testpmd> set_hash_input_set 0 ipv4-udp src-ipv4 add
3457 The Flow Director filters can match the different fields for different type of packet, i.e. specific input set
3458 on per flow type and the flexible payload. This command can be used to change input set for each flow type.
3460 Set the input set for flow director::
3462 set_fdir_input_set (port_id) (ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
3463 ipv4-other|ipv6|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other| \
3464 l2_payload|<flow_id>) (ivlan|ethertype|src-ipv4|dst-ipv4|src-ipv6|dst-ipv6| \
3465 ipv4-tos|ipv4-proto|ipv4-ttl|ipv6-tc|ipv6-next-header|ipv6-hop-limits| \
3466 tudp-src-port|udp-dst-port|cp-src-port|tcp-dst-port|sctp-src-port| \
3467 sctp-dst-port|sctp-veri-tag|none) (select|add)
3469 For example to add source IP to FD input set for flow type of ipv4-udp on port 0::
3471 testpmd> set_fdir_input_set 0 ipv4-udp src-ipv4 add
3476 Set different GRE key length for input set::
3478 global_config (port_id) gre-key-len (number in bytes)
3480 For example to set GRE key length for input set to 4 bytes on port 0::
3482 testpmd> global_config 0 gre-key-len 4
3485 .. _testpmd_rte_flow:
3487 Flow rules management
3488 ---------------------
3490 Control of the generic flow API (*rte_flow*) is fully exposed through the
3491 ``flow`` command (validation, creation, destruction, queries and operation
3494 Considering *rte_flow* overlaps with all `Filter Functions`_, using both
3495 features simultaneously may cause undefined side-effects and is therefore
3501 Because the ``flow`` command uses dynamic tokens to handle the large number
3502 of possible flow rules combinations, its behavior differs slightly from
3503 other commands, in particular:
3505 - Pressing *?* or the *<tab>* key displays contextual help for the current
3506 token, not that of the entire command.
3508 - Optional and repeated parameters are supported (provided they are listed
3509 in the contextual help).
3511 The first parameter stands for the operation mode. Possible operations and
3512 their general syntax are described below. They are covered in detail in the
3515 - Check whether a flow rule can be created::
3517 flow validate {port_id}
3518 [group {group_id}] [priority {level}] [ingress] [egress] [transfer]
3519 pattern {item} [/ {item} [...]] / end
3520 actions {action} [/ {action} [...]] / end
3522 - Create a flow rule::
3524 flow create {port_id}
3525 [group {group_id}] [priority {level}] [ingress] [egress] [transfer]
3526 pattern {item} [/ {item} [...]] / end
3527 actions {action} [/ {action} [...]] / end
3529 - Destroy specific flow rules::
3531 flow destroy {port_id} rule {rule_id} [...]
3533 - Destroy all flow rules::
3535 flow flush {port_id}
3537 - Query an existing flow rule::
3539 flow query {port_id} {rule_id} {action}
3541 - List existing flow rules sorted by priority, filtered by group
3544 flow list {port_id} [group {group_id}] [...]
3546 - Restrict ingress traffic to the defined flow rules::
3548 flow isolate {port_id} {boolean}
3550 - Dump internal representation information of all flows in hardware::
3552 flow dump {port_id} {output_file}
3554 - List and destroy aged flow rules::
3556 flow aged {port_id} [destroy]
3558 - Tunnel offload - create a tunnel stub::
3560 flow tunnel create {port_id} type {tunnel_type}
3562 - Tunnel offload - destroy a tunnel stub::
3564 flow tunnel destroy {port_id} id {tunnel_id}
3566 - Tunnel offload - list port tunnel stubs::
3568 flow tunnel list {port_id}
3570 Creating a tunnel stub for offload
3571 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3573 ``flow tunnel create`` setup a tunnel stub for tunnel offload flow rules::
3575 flow tunnel create {port_id} type {tunnel_type}
3577 If successful, it will return a tunnel stub ID usable with other commands::
3579 port [...]: flow tunnel #[...] type [...]
3581 Tunnel stub ID is relative to a port.
3583 Destroying tunnel offload stub
3584 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3586 ``flow tunnel destroy`` destroy port tunnel stub::
3588 flow tunnel destroy {port_id} id {tunnel_id}
3590 Listing tunnel offload stubs
3591 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3593 ``flow tunnel list`` list port tunnel offload stubs::
3595 flow tunnel list {port_id}
3597 Validating flow rules
3598 ~~~~~~~~~~~~~~~~~~~~~
3600 ``flow validate`` reports whether a flow rule would be accepted by the
3601 underlying device in its current state but stops short of creating it. It is
3602 bound to ``rte_flow_validate()``::
3604 flow validate {port_id}
3605 [group {group_id}] [priority {level}] [ingress] [egress] [transfer]
3606 pattern {item} [/ {item} [...]] / end
3607 actions {action} [/ {action} [...]] / end
3609 If successful, it will show::
3613 Otherwise it will show an error message of the form::
3615 Caught error type [...] ([...]): [...]
3617 This command uses the same parameters as ``flow create``, their format is
3618 described in `Creating flow rules`_.
3620 Check whether redirecting any Ethernet packet received on port 0 to RX queue
3621 index 6 is supported::
3623 testpmd> flow validate 0 ingress pattern eth / end
3624 actions queue index 6 / end
3628 Port 0 does not support TCPv6 rules::
3630 testpmd> flow validate 0 ingress pattern eth / ipv6 / tcp / end
3632 Caught error type 9 (specific pattern item): Invalid argument
3638 ``flow create`` validates and creates the specified flow rule. It is bound
3639 to ``rte_flow_create()``::
3641 flow create {port_id}
3642 [group {group_id}] [priority {level}] [ingress] [egress] [transfer]
3643 [tunnel_set {tunnel_id}] [tunnel_match {tunnel_id}]
3644 pattern {item} [/ {item} [...]] / end
3645 actions {action} [/ {action} [...]] / end
3647 If successful, it will return a flow rule ID usable with other commands::
3649 Flow rule #[...] created
3651 Otherwise it will show an error message of the form::
3653 Caught error type [...] ([...]): [...]
3655 Parameters describe in the following order:
3657 - Attributes (*group*, *priority*, *ingress*, *egress*, *transfer* tokens).
3658 - Tunnel offload specification (tunnel_set, tunnel_match)
3659 - A matching pattern, starting with the *pattern* token and terminated by an
3661 - Actions, starting with the *actions* token and terminated by an *end*
3664 These translate directly to *rte_flow* objects provided as-is to the
3665 underlying functions.
3667 The shortest valid definition only comprises mandatory tokens::
3669 testpmd> flow create 0 pattern end actions end
3671 Note that PMDs may refuse rules that essentially do nothing such as this
3674 **All unspecified object values are automatically initialized to 0.**
3679 These tokens affect flow rule attributes (``struct rte_flow_attr``) and are
3680 specified before the ``pattern`` token.
3682 - ``group {group id}``: priority group.
3683 - ``priority {level}``: priority level within group.
3684 - ``ingress``: rule applies to ingress traffic.
3685 - ``egress``: rule applies to egress traffic.
3686 - ``transfer``: apply rule directly to endpoints found in pattern.
3688 Each instance of an attribute specified several times overrides the previous
3689 value as shown below (group 4 is used)::
3691 testpmd> flow create 0 group 42 group 24 group 4 [...]
3693 Note that once enabled, ``ingress`` and ``egress`` cannot be disabled.
3695 While not specifying a direction is an error, some rules may allow both
3698 Most rules affect RX therefore contain the ``ingress`` token::
3700 testpmd> flow create 0 ingress pattern [...]
3705 Indicate tunnel offload rule type
3707 - ``tunnel_set {tunnel_id}``: mark rule as tunnel offload decap_set type.
3708 - ``tunnel_match {tunnel_id}``: mark rule as tunel offload match type.
3713 A matching pattern starts after the ``pattern`` token. It is made of pattern
3714 items and is terminated by a mandatory ``end`` item.
3716 Items are named after their type (*RTE_FLOW_ITEM_TYPE_* from ``enum
3717 rte_flow_item_type``).
3719 The ``/`` token is used as a separator between pattern items as shown
3722 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end [...]
3724 Note that protocol items like these must be stacked from lowest to highest
3725 layer to make sense. For instance, the following rule is either invalid or
3726 unlikely to match any packet::
3728 testpmd> flow create 0 ingress pattern eth / udp / ipv4 / end [...]
3730 More information on these restrictions can be found in the *rte_flow*
3733 Several items support additional specification structures, for example
3734 ``ipv4`` allows specifying source and destination addresses as follows::
3736 testpmd> flow create 0 ingress pattern eth / ipv4 src is 10.1.1.1
3737 dst is 10.2.0.0 / end [...]
3739 This rule matches all IPv4 traffic with the specified properties.
3741 In this example, ``src`` and ``dst`` are field names of the underlying
3742 ``struct rte_flow_item_ipv4`` object. All item properties can be specified
3743 in a similar fashion.
3745 The ``is`` token means that the subsequent value must be matched exactly,
3746 and assigns ``spec`` and ``mask`` fields in ``struct rte_flow_item``
3747 accordingly. Possible assignment tokens are:
3749 - ``is``: match value perfectly (with full bit-mask).
3750 - ``spec``: match value according to configured bit-mask.
3751 - ``last``: specify upper bound to establish a range.
3752 - ``mask``: specify bit-mask with relevant bits set to one.
3753 - ``prefix``: generate bit-mask with <prefix-length> most-significant bits set to one.
3755 These yield identical results::
3757 ipv4 src is 10.1.1.1
3761 ipv4 src spec 10.1.1.1 src mask 255.255.255.255
3765 ipv4 src spec 10.1.1.1 src prefix 32
3769 ipv4 src is 10.1.1.1 src last 10.1.1.1 # range with a single value
3773 ipv4 src is 10.1.1.1 src last 0 # 0 disables range
3775 Inclusive ranges can be defined with ``last``::
3777 ipv4 src is 10.1.1.1 src last 10.2.3.4 # 10.1.1.1 to 10.2.3.4
3779 Note that ``mask`` affects both ``spec`` and ``last``::
3781 ipv4 src is 10.1.1.1 src last 10.2.3.4 src mask 255.255.0.0
3782 # matches 10.1.0.0 to 10.2.255.255
3784 Properties can be modified multiple times::
3786 ipv4 src is 10.1.1.1 src is 10.1.2.3 src is 10.2.3.4 # matches 10.2.3.4
3790 ipv4 src is 10.1.1.1 src prefix 24 src prefix 16 # matches 10.1.0.0/16
3795 This section lists supported pattern items and their attributes, if any.
3797 - ``end``: end list of pattern items.
3799 - ``void``: no-op pattern item.
3801 - ``invert``: perform actions when pattern does not match.
3803 - ``any``: match any protocol for the current layer.
3805 - ``num {unsigned}``: number of layers covered.
3807 - ``pf``: match traffic from/to the physical function.
3809 - ``vf``: match traffic from/to a virtual function ID.
3811 - ``id {unsigned}``: VF ID.
3813 - ``phy_port``: match traffic from/to a specific physical port.
3815 - ``index {unsigned}``: physical port index.
3817 - ``port_id``: match traffic from/to a given DPDK port ID.
3819 - ``id {unsigned}``: DPDK port ID.
3821 - ``mark``: match value set in previously matched flow rule using the mark action.
3823 - ``id {unsigned}``: arbitrary integer value.
3825 - ``raw``: match an arbitrary byte string.
3827 - ``relative {boolean}``: look for pattern after the previous item.
3828 - ``search {boolean}``: search pattern from offset (see also limit).
3829 - ``offset {integer}``: absolute or relative offset for pattern.
3830 - ``limit {unsigned}``: search area limit for start of pattern.
3831 - ``pattern {string}``: byte string to look for.
3833 - ``eth``: match Ethernet header.
3835 - ``dst {MAC-48}``: destination MAC.
3836 - ``src {MAC-48}``: source MAC.
3837 - ``type {unsigned}``: EtherType or TPID.
3839 - ``vlan``: match 802.1Q/ad VLAN tag.
3841 - ``tci {unsigned}``: tag control information.
3842 - ``pcp {unsigned}``: priority code point.
3843 - ``dei {unsigned}``: drop eligible indicator.
3844 - ``vid {unsigned}``: VLAN identifier.
3845 - ``inner_type {unsigned}``: inner EtherType or TPID.
3847 - ``ipv4``: match IPv4 header.
3849 - ``tos {unsigned}``: type of service.
3850 - ``ttl {unsigned}``: time to live.
3851 - ``proto {unsigned}``: next protocol ID.
3852 - ``src {ipv4 address}``: source address.
3853 - ``dst {ipv4 address}``: destination address.
3855 - ``ipv6``: match IPv6 header.
3857 - ``tc {unsigned}``: traffic class.
3858 - ``flow {unsigned}``: flow label.
3859 - ``proto {unsigned}``: protocol (next header).
3860 - ``hop {unsigned}``: hop limit.
3861 - ``src {ipv6 address}``: source address.
3862 - ``dst {ipv6 address}``: destination address.
3864 - ``icmp``: match ICMP header.
3866 - ``type {unsigned}``: ICMP packet type.
3867 - ``code {unsigned}``: ICMP packet code.
3869 - ``udp``: match UDP header.
3871 - ``src {unsigned}``: UDP source port.
3872 - ``dst {unsigned}``: UDP destination port.
3874 - ``tcp``: match TCP header.
3876 - ``src {unsigned}``: TCP source port.
3877 - ``dst {unsigned}``: TCP destination port.
3879 - ``sctp``: match SCTP header.
3881 - ``src {unsigned}``: SCTP source port.
3882 - ``dst {unsigned}``: SCTP destination port.
3883 - ``tag {unsigned}``: validation tag.
3884 - ``cksum {unsigned}``: checksum.
3886 - ``vxlan``: match VXLAN header.
3888 - ``vni {unsigned}``: VXLAN identifier.
3890 - ``e_tag``: match IEEE 802.1BR E-Tag header.
3892 - ``grp_ecid_b {unsigned}``: GRP and E-CID base.
3894 - ``nvgre``: match NVGRE header.
3896 - ``tni {unsigned}``: virtual subnet ID.
3898 - ``mpls``: match MPLS header.
3900 - ``label {unsigned}``: MPLS label.
3902 - ``gre``: match GRE header.
3904 - ``protocol {unsigned}``: protocol type.
3906 - ``gre_key``: match GRE optional key field.
3908 - ``value {unsigned}``: key value.
3910 - ``fuzzy``: fuzzy pattern match, expect faster than default.
3912 - ``thresh {unsigned}``: accuracy threshold.
3914 - ``gtp``, ``gtpc``, ``gtpu``: match GTPv1 header.
3916 - ``teid {unsigned}``: tunnel endpoint identifier.
3918 - ``geneve``: match GENEVE header.
3920 - ``vni {unsigned}``: virtual network identifier.
3921 - ``protocol {unsigned}``: protocol type.
3923 - ``vxlan-gpe``: match VXLAN-GPE header.
3925 - ``vni {unsigned}``: VXLAN-GPE identifier.
3927 - ``arp_eth_ipv4``: match ARP header for Ethernet/IPv4.
3929 - ``sha {MAC-48}``: sender hardware address.
3930 - ``spa {ipv4 address}``: sender IPv4 address.
3931 - ``tha {MAC-48}``: target hardware address.
3932 - ``tpa {ipv4 address}``: target IPv4 address.
3934 - ``ipv6_ext``: match presence of any IPv6 extension header.
3936 - ``next_hdr {unsigned}``: next header.
3938 - ``icmp6``: match any ICMPv6 header.
3940 - ``type {unsigned}``: ICMPv6 type.
3941 - ``code {unsigned}``: ICMPv6 code.
3943 - ``icmp6_nd_ns``: match ICMPv6 neighbor discovery solicitation.
3945 - ``target_addr {ipv6 address}``: target address.
3947 - ``icmp6_nd_na``: match ICMPv6 neighbor discovery advertisement.
3949 - ``target_addr {ipv6 address}``: target address.
3951 - ``icmp6_nd_opt``: match presence of any ICMPv6 neighbor discovery option.
3953 - ``type {unsigned}``: ND option type.
3955 - ``icmp6_nd_opt_sla_eth``: match ICMPv6 neighbor discovery source Ethernet
3956 link-layer address option.
3958 - ``sla {MAC-48}``: source Ethernet LLA.
3960 - ``icmp6_nd_opt_tla_eth``: match ICMPv6 neighbor discovery target Ethernet
3961 link-layer address option.
3963 - ``tla {MAC-48}``: target Ethernet LLA.
3965 - ``meta``: match application specific metadata.
3967 - ``data {unsigned}``: metadata value.
3969 - ``gtp_psc``: match GTP PDU extension header with type 0x85.
3971 - ``pdu_type {unsigned}``: PDU type.
3972 - ``qfi {unsigned}``: QoS flow identifier.
3974 - ``pppoes``, ``pppoed``: match PPPoE header.
3976 - ``session_id {unsigned}``: session identifier.
3978 - ``pppoe_proto_id``: match PPPoE session protocol identifier.
3980 - ``proto_id {unsigned}``: PPP protocol identifier.
3982 - ``l2tpv3oip``: match L2TPv3 over IP header.
3984 - ``session_id {unsigned}``: L2TPv3 over IP session identifier.
3986 - ``ah``: match AH header.
3988 - ``spi {unsigned}``: security parameters index.
3990 - ``pfcp``: match PFCP header.
3992 - ``s_field {unsigned}``: S field.
3993 - ``seid {unsigned}``: session endpoint identifier.
3998 A list of actions starts after the ``actions`` token in the same fashion as
3999 `Matching pattern`_; actions are separated by ``/`` tokens and the list is
4000 terminated by a mandatory ``end`` action.
4002 Actions are named after their type (*RTE_FLOW_ACTION_TYPE_* from ``enum
4003 rte_flow_action_type``).
4005 Dropping all incoming UDPv4 packets can be expressed as follows::
4007 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
4010 Several actions have configurable properties which must be specified when
4011 there is no valid default value. For example, ``queue`` requires a target
4014 This rule redirects incoming UDPv4 traffic to queue index 6::
4016 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
4017 actions queue index 6 / end
4019 While this one could be rejected by PMDs (unspecified queue index)::
4021 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
4024 As defined by *rte_flow*, the list is not ordered, all actions of a given
4025 rule are performed simultaneously. These are equivalent::
4027 queue index 6 / void / mark id 42 / end
4031 void / mark id 42 / queue index 6 / end
4033 All actions in a list should have different types, otherwise only the last
4034 action of a given type is taken into account::
4036 queue index 4 / queue index 5 / queue index 6 / end # will use queue 6
4040 drop / drop / drop / end # drop is performed only once
4044 mark id 42 / queue index 3 / mark id 24 / end # mark will be 24
4046 Considering they are performed simultaneously, opposite and overlapping
4047 actions can sometimes be combined when the end result is unambiguous::
4049 drop / queue index 6 / end # drop has no effect
4053 queue index 6 / rss queues 6 7 8 / end # queue has no effect
4057 drop / passthru / end # drop has no effect
4059 Note that PMDs may still refuse such combinations.
4064 This section lists supported actions and their attributes, if any.
4066 - ``end``: end list of actions.
4068 - ``void``: no-op action.
4070 - ``passthru``: let subsequent rule process matched packets.
4072 - ``jump``: redirect traffic to group on device.
4074 - ``group {unsigned}``: group to redirect to.
4076 - ``mark``: attach 32 bit value to packets.
4078 - ``id {unsigned}``: 32 bit value to return with packets.
4080 - ``flag``: flag packets.
4082 - ``queue``: assign packets to a given queue index.
4084 - ``index {unsigned}``: queue index to use.
4086 - ``drop``: drop packets (note: passthru has priority).
4088 - ``count``: enable counters for this rule.
4090 - ``rss``: spread packets among several queues.
4092 - ``func {hash function}``: RSS hash function to apply, allowed tokens are
4093 the same as `set_hash_global_config`_.
4095 - ``level {unsigned}``: encapsulation level for ``types``.
4097 - ``types [{RSS hash type} [...]] end``: specific RSS hash types, allowed
4098 tokens are the same as `set_hash_input_set`_, except that an empty list
4099 does not disable RSS but instead requests unspecified "best-effort"
4102 - ``key {string}``: RSS hash key, overrides ``key_len``.
4104 - ``key_len {unsigned}``: RSS hash key length in bytes, can be used in
4105 conjunction with ``key`` to pad or truncate it.
4107 - ``queues [{unsigned} [...]] end``: queue indices to use.
4109 - ``pf``: direct traffic to physical function.
4111 - ``vf``: direct traffic to a virtual function ID.
4113 - ``original {boolean}``: use original VF ID if possible.
4114 - ``id {unsigned}``: VF ID.
4116 - ``phy_port``: direct packets to physical port index.
4118 - ``original {boolean}``: use original port index if possible.
4119 - ``index {unsigned}``: physical port index.
4121 - ``port_id``: direct matching traffic to a given DPDK port ID.
4123 - ``original {boolean}``: use original DPDK port ID if possible.
4124 - ``id {unsigned}``: DPDK port ID.
4126 - ``of_set_mpls_ttl``: OpenFlow's ``OFPAT_SET_MPLS_TTL``.
4128 - ``mpls_ttl``: MPLS TTL.
4130 - ``of_dec_mpls_ttl``: OpenFlow's ``OFPAT_DEC_MPLS_TTL``.
4132 - ``of_set_nw_ttl``: OpenFlow's ``OFPAT_SET_NW_TTL``.
4134 - ``nw_ttl``: IP TTL.
4136 - ``of_dec_nw_ttl``: OpenFlow's ``OFPAT_DEC_NW_TTL``.
4138 - ``of_copy_ttl_out``: OpenFlow's ``OFPAT_COPY_TTL_OUT``.
4140 - ``of_copy_ttl_in``: OpenFlow's ``OFPAT_COPY_TTL_IN``.
4142 - ``of_pop_vlan``: OpenFlow's ``OFPAT_POP_VLAN``.
4144 - ``of_push_vlan``: OpenFlow's ``OFPAT_PUSH_VLAN``.
4146 - ``ethertype``: Ethertype.
4148 - ``of_set_vlan_vid``: OpenFlow's ``OFPAT_SET_VLAN_VID``.
4150 - ``vlan_vid``: VLAN id.
4152 - ``of_set_vlan_pcp``: OpenFlow's ``OFPAT_SET_VLAN_PCP``.
4154 - ``vlan_pcp``: VLAN priority.
4156 - ``of_pop_mpls``: OpenFlow's ``OFPAT_POP_MPLS``.
4158 - ``ethertype``: Ethertype.
4160 - ``of_push_mpls``: OpenFlow's ``OFPAT_PUSH_MPLS``.
4162 - ``ethertype``: Ethertype.
4164 - ``vxlan_encap``: Performs a VXLAN encapsulation, outer layer configuration
4165 is done through `Config VXLAN Encap outer layers`_.
4167 - ``vxlan_decap``: Performs a decapsulation action by stripping all headers of
4168 the VXLAN tunnel network overlay from the matched flow.
4170 - ``nvgre_encap``: Performs a NVGRE encapsulation, outer layer configuration
4171 is done through `Config NVGRE Encap outer layers`_.
4173 - ``nvgre_decap``: Performs a decapsulation action by stripping all headers of
4174 the NVGRE tunnel network overlay from the matched flow.
4176 - ``l2_encap``: Performs a L2 encapsulation, L2 configuration
4177 is done through `Config L2 Encap`_.
4179 - ``l2_decap``: Performs a L2 decapsulation, L2 configuration
4180 is done through `Config L2 Decap`_.
4182 - ``mplsogre_encap``: Performs a MPLSoGRE encapsulation, outer layer
4183 configuration is done through `Config MPLSoGRE Encap outer layers`_.
4185 - ``mplsogre_decap``: Performs a MPLSoGRE decapsulation, outer layer
4186 configuration is done through `Config MPLSoGRE Decap outer layers`_.
4188 - ``mplsoudp_encap``: Performs a MPLSoUDP encapsulation, outer layer
4189 configuration is done through `Config MPLSoUDP Encap outer layers`_.
4191 - ``mplsoudp_decap``: Performs a MPLSoUDP decapsulation, outer layer
4192 configuration is done through `Config MPLSoUDP Decap outer layers`_.
4194 - ``set_ipv4_src``: Set a new IPv4 source address in the outermost IPv4 header.
4196 - ``ipv4_addr``: New IPv4 source address.
4198 - ``set_ipv4_dst``: Set a new IPv4 destination address in the outermost IPv4
4201 - ``ipv4_addr``: New IPv4 destination address.
4203 - ``set_ipv6_src``: Set a new IPv6 source address in the outermost IPv6 header.
4205 - ``ipv6_addr``: New IPv6 source address.
4207 - ``set_ipv6_dst``: Set a new IPv6 destination address in the outermost IPv6
4210 - ``ipv6_addr``: New IPv6 destination address.
4212 - ``set_tp_src``: Set a new source port number in the outermost TCP/UDP
4215 - ``port``: New TCP/UDP source port number.
4217 - ``set_tp_dst``: Set a new destination port number in the outermost TCP/UDP
4220 - ``port``: New TCP/UDP destination port number.
4222 - ``mac_swap``: Swap the source and destination MAC addresses in the outermost
4225 - ``dec_ttl``: Performs a decrease TTL value action
4227 - ``set_ttl``: Set TTL value with specified value
4228 - ``ttl_value {unsigned}``: The new TTL value to be set
4230 - ``set_mac_src``: set source MAC address
4232 - ``mac_addr {MAC-48}``: new source MAC address
4234 - ``set_mac_dst``: set destination MAC address
4236 - ``mac_addr {MAC-48}``: new destination MAC address
4238 - ``inc_tcp_seq``: Increase sequence number in the outermost TCP header.
4240 - ``value {unsigned}``: Value to increase TCP sequence number by.
4242 - ``dec_tcp_seq``: Decrease sequence number in the outermost TCP header.
4244 - ``value {unsigned}``: Value to decrease TCP sequence number by.
4246 - ``inc_tcp_ack``: Increase acknowledgment number in the outermost TCP header.
4248 - ``value {unsigned}``: Value to increase TCP acknowledgment number by.
4250 - ``dec_tcp_ack``: Decrease acknowledgment number in the outermost TCP header.
4252 - ``value {unsigned}``: Value to decrease TCP acknowledgment number by.
4254 - ``set_ipv4_dscp``: Set IPv4 DSCP value with specified value
4256 - ``dscp_value {unsigned}``: The new DSCP value to be set
4258 - ``set_ipv6_dscp``: Set IPv6 DSCP value with specified value
4260 - ``dscp_value {unsigned}``: The new DSCP value to be set
4262 - ``shared``: Use shared action created via
4263 ``flow shared_action {port_id} create``
4265 - ``shared_action_id {unsigned}``: Shared action ID to use
4267 Destroying flow rules
4268 ~~~~~~~~~~~~~~~~~~~~~
4270 ``flow destroy`` destroys one or more rules from their rule ID (as returned
4271 by ``flow create``), this command calls ``rte_flow_destroy()`` as many
4272 times as necessary::
4274 flow destroy {port_id} rule {rule_id} [...]
4276 If successful, it will show::
4278 Flow rule #[...] destroyed
4280 It does not report anything for rule IDs that do not exist. The usual error
4281 message is shown when a rule cannot be destroyed::
4283 Caught error type [...] ([...]): [...]
4285 ``flow flush`` destroys all rules on a device and does not take extra
4286 arguments. It is bound to ``rte_flow_flush()``::
4288 flow flush {port_id}
4290 Any errors are reported as above.
4292 Creating several rules and destroying them::
4294 testpmd> flow create 0 ingress pattern eth / ipv6 / end
4295 actions queue index 2 / end
4296 Flow rule #0 created
4297 testpmd> flow create 0 ingress pattern eth / ipv4 / end
4298 actions queue index 3 / end
4299 Flow rule #1 created
4300 testpmd> flow destroy 0 rule 0 rule 1
4301 Flow rule #1 destroyed
4302 Flow rule #0 destroyed
4305 The same result can be achieved using ``flow flush``::
4307 testpmd> flow create 0 ingress pattern eth / ipv6 / end
4308 actions queue index 2 / end
4309 Flow rule #0 created
4310 testpmd> flow create 0 ingress pattern eth / ipv4 / end
4311 actions queue index 3 / end
4312 Flow rule #1 created
4313 testpmd> flow flush 0
4316 Non-existent rule IDs are ignored::
4318 testpmd> flow create 0 ingress pattern eth / ipv6 / end
4319 actions queue index 2 / end
4320 Flow rule #0 created
4321 testpmd> flow create 0 ingress pattern eth / ipv4 / end
4322 actions queue index 3 / end
4323 Flow rule #1 created
4324 testpmd> flow destroy 0 rule 42 rule 10 rule 2
4326 testpmd> flow destroy 0 rule 0
4327 Flow rule #0 destroyed
4333 ``flow query`` queries a specific action of a flow rule having that
4334 ability. Such actions collect information that can be reported using this
4335 command. It is bound to ``rte_flow_query()``::
4337 flow query {port_id} {rule_id} {action}
4339 If successful, it will display either the retrieved data for known actions
4340 or the following message::
4342 Cannot display result for action type [...] ([...])
4344 Otherwise, it will complain either that the rule does not exist or that some
4347 Flow rule #[...] not found
4351 Caught error type [...] ([...]): [...]
4353 Currently only the ``count`` action is supported. This action reports the
4354 number of packets that hit the flow rule and the total number of bytes. Its
4355 output has the following format::
4358 hits_set: [...] # whether "hits" contains a valid value
4359 bytes_set: [...] # whether "bytes" contains a valid value
4360 hits: [...] # number of packets
4361 bytes: [...] # number of bytes
4363 Querying counters for TCPv6 packets redirected to queue 6::
4365 testpmd> flow create 0 ingress pattern eth / ipv6 / tcp / end
4366 actions queue index 6 / count / end
4367 Flow rule #4 created
4368 testpmd> flow query 0 4 count
4379 ``flow list`` lists existing flow rules sorted by priority and optionally
4380 filtered by group identifiers::
4382 flow list {port_id} [group {group_id}] [...]
4384 This command only fails with the following message if the device does not
4389 Output consists of a header line followed by a short description of each
4390 flow rule, one per line. There is no output at all when no flow rules are
4391 configured on the device::
4393 ID Group Prio Attr Rule
4394 [...] [...] [...] [...] [...]
4396 ``Attr`` column flags:
4398 - ``i`` for ``ingress``.
4399 - ``e`` for ``egress``.
4401 Creating several flow rules and listing them::
4403 testpmd> flow create 0 ingress pattern eth / ipv4 / end
4404 actions queue index 6 / end
4405 Flow rule #0 created
4406 testpmd> flow create 0 ingress pattern eth / ipv6 / end
4407 actions queue index 2 / end
4408 Flow rule #1 created
4409 testpmd> flow create 0 priority 5 ingress pattern eth / ipv4 / udp / end
4410 actions rss queues 6 7 8 end / end
4411 Flow rule #2 created
4412 testpmd> flow list 0
4413 ID Group Prio Attr Rule
4414 0 0 0 i- ETH IPV4 => QUEUE
4415 1 0 0 i- ETH IPV6 => QUEUE
4416 2 0 5 i- ETH IPV4 UDP => RSS
4419 Rules are sorted by priority (i.e. group ID first, then priority level)::
4421 testpmd> flow list 1
4422 ID Group Prio Attr Rule
4423 0 0 0 i- ETH => COUNT
4424 6 0 500 i- ETH IPV6 TCP => DROP COUNT
4425 5 0 1000 i- ETH IPV6 ICMP => QUEUE
4426 1 24 0 i- ETH IPV4 UDP => QUEUE
4427 4 24 10 i- ETH IPV4 TCP => DROP
4428 3 24 20 i- ETH IPV4 => DROP
4429 2 24 42 i- ETH IPV4 UDP => QUEUE
4430 7 63 0 i- ETH IPV6 UDP VXLAN => MARK QUEUE
4433 Output can be limited to specific groups::
4435 testpmd> flow list 1 group 0 group 63
4436 ID Group Prio Attr Rule
4437 0 0 0 i- ETH => COUNT
4438 6 0 500 i- ETH IPV6 TCP => DROP COUNT
4439 5 0 1000 i- ETH IPV6 ICMP => QUEUE
4440 7 63 0 i- ETH IPV6 UDP VXLAN => MARK QUEUE
4443 Toggling isolated mode
4444 ~~~~~~~~~~~~~~~~~~~~~~
4446 ``flow isolate`` can be used to tell the underlying PMD that ingress traffic
4447 must only be injected from the defined flow rules; that no default traffic
4448 is expected outside those rules and the driver is free to assign more
4449 resources to handle them. It is bound to ``rte_flow_isolate()``::
4451 flow isolate {port_id} {boolean}
4453 If successful, enabling or disabling isolated mode shows either::
4455 Ingress traffic on port [...]
4456 is now restricted to the defined flow rules
4460 Ingress traffic on port [...]
4461 is not restricted anymore to the defined flow rules
4463 Otherwise, in case of error::
4465 Caught error type [...] ([...]): [...]
4467 Mainly due to its side effects, PMDs supporting this mode may not have the
4468 ability to toggle it more than once without reinitializing affected ports
4469 first (e.g. by exiting testpmd).
4471 Enabling isolated mode::
4473 testpmd> flow isolate 0 true
4474 Ingress traffic on port 0 is now restricted to the defined flow rules
4477 Disabling isolated mode::
4479 testpmd> flow isolate 0 false
4480 Ingress traffic on port 0 is not restricted anymore to the defined flow rules
4483 Dumping HW internal information
4484 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4486 ``flow dump`` dumps the hardware's internal representation information of
4487 all flows. It is bound to ``rte_flow_dev_dump()``::
4489 flow dump {port_id} {output_file}
4491 If successful, it will show::
4495 Otherwise, it will complain error occurred::
4497 Caught error type [...] ([...]): [...]
4499 Listing and destroying aged flow rules
4500 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4502 ``flow aged`` simply lists aged flow rules be get from api ``rte_flow_get_aged_flows``,
4503 and ``destroy`` parameter can be used to destroy those flow rules in PMD.
4505 flow aged {port_id} [destroy]
4507 Listing current aged flow rules::
4509 testpmd> flow aged 0
4510 Port 0 total aged flows: 0
4511 testpmd> flow create 0 ingress pattern eth / ipv4 src is 2.2.2.14 / end
4512 actions age timeout 5 / queue index 0 / end
4513 Flow rule #0 created
4514 testpmd> flow create 0 ingress pattern eth / ipv4 src is 2.2.2.15 / end
4515 actions age timeout 4 / queue index 0 / end
4516 Flow rule #1 created
4517 testpmd> flow create 0 ingress pattern eth / ipv4 src is 2.2.2.16 / end
4518 actions age timeout 2 / queue index 0 / end
4519 Flow rule #2 created
4520 testpmd> flow create 0 ingress pattern eth / ipv4 src is 2.2.2.17 / end
4521 actions age timeout 3 / queue index 0 / end
4522 Flow rule #3 created
4525 Aged Rules are simply list as command ``flow list {port_id}``, but strip the detail rule
4526 information, all the aged flows are sorted by the longest timeout time. For example, if
4527 those rules be configured in the same time, ID 2 will be the first aged out rule, the next
4528 will be ID 3, ID 1, ID 0::
4530 testpmd> flow aged 0
4531 Port 0 total aged flows: 4
4538 If attach ``destroy`` parameter, the command will destroy all the list aged flow rules.
4540 testpmd> flow aged 0 destroy
4541 Port 0 total aged flows: 4
4548 Flow rule #2 destroyed
4549 Flow rule #3 destroyed
4550 Flow rule #1 destroyed
4551 Flow rule #0 destroyed
4552 4 flows be destroyed
4553 testpmd> flow aged 0
4554 Port 0 total aged flows: 0
4556 Creating shared actions
4557 ~~~~~~~~~~~~~~~~~~~~~~~
4558 ``flow shared_action {port_id} create`` creates shared action with optional
4559 shared action ID. It is bound to ``rte_flow_shared_action_create()``::
4561 flow shared_action {port_id} create [action_id {shared_action_id}]
4562 [ingress] [egress] action {action} / end
4564 If successful, it will show::
4566 Shared action #[...] created
4568 Otherwise, it will complain either that shared action already exists or that
4569 some error occurred::
4571 Shared action #[...] is already assigned, delete it first
4575 Caught error type [...] ([...]): [...]
4577 Create shared rss action with id 100 to queues 1 and 2 on port 0::
4579 testpmd> flow shared_action 0 create action_id 100 \
4580 ingress action rss queues 1 2 end / end
4582 Create shared rss action with id assigned by testpmd to queues 1 and 2 on
4585 testpmd> flow shared_action 0 create action_id \
4586 ingress action rss queues 0 1 end / end
4588 Updating shared actions
4589 ~~~~~~~~~~~~~~~~~~~~~~~
4590 ``flow shared_action {port_id} update`` updates configuration of the shared
4591 action from its shared action ID (as returned by
4592 ``flow shared_action {port_id} create``). It is bound to
4593 ``rte_flow_shared_action_update()``::
4595 flow shared_action {port_id} update {shared_action_id}
4596 action {action} / end
4598 If successful, it will show::
4600 Shared action #[...] updated
4602 Otherwise, it will complain either that shared action not found or that some
4605 Failed to find shared action #[...] on port [...]
4609 Caught error type [...] ([...]): [...]
4611 Update shared rss action having id 100 on port 0 with rss to queues 0 and 3
4612 (in create example above rss queues were 1 and 2)::
4614 testpmd> flow shared_action 0 update 100 action rss queues 0 3 end / end
4616 Destroying shared actions
4617 ~~~~~~~~~~~~~~~~~~~~~~~~~
4618 ``flow shared_action {port_id} update`` destroys one or more shared actions
4619 from their shared action IDs (as returned by
4620 ``flow shared_action {port_id} create``). It is bound to
4621 ``rte_flow_shared_action_destroy()``::
4623 flow shared_action {port_id} destroy action_id {shared_action_id} [...]
4625 If successful, it will show::
4627 Shared action #[...] destroyed
4629 It does not report anything for shared action IDs that do not exist.
4630 The usual error message is shown when a shared action cannot be destroyed::
4632 Caught error type [...] ([...]): [...]
4634 Destroy shared actions having id 100 & 101::
4636 testpmd> flow shared_action 0 destroy action_id 100 action_id 101
4638 Query shared actions
4639 ~~~~~~~~~~~~~~~~~~~~
4640 ``flow shared_action {port_id} query`` queries the shared action from its
4641 shared action ID (as returned by ``flow shared_action {port_id} create``).
4642 It is bound to ``rte_flow_shared_action_query()``::
4644 flow shared_action {port_id} query {shared_action_id}
4646 Currently only rss shared action supported. If successful, it will show::
4651 Otherwise, it will complain either that shared action not found or that some
4654 Failed to find shared action #[...] on port [...]
4658 Caught error type [...] ([...]): [...]
4660 Query shared action having id 100::
4662 testpmd> flow shared_action 0 query 100
4664 Sample QinQ flow rules
4665 ~~~~~~~~~~~~~~~~~~~~~~
4667 Before creating QinQ rule(s) the following commands should be issued to enable QinQ::
4669 testpmd> port stop 0
4670 testpmd> vlan set qinq_strip on 0
4672 The above command sets the inner and outer TPID's to 0x8100.
4674 To change the TPID's the following commands should be used::
4676 testpmd> vlan set outer tpid 0xa100 0
4677 testpmd> vlan set inner tpid 0x9100 0
4678 testpmd> port start 0
4680 Validate and create a QinQ rule on port 0 to steer traffic to a VF queue in a VM.
4684 testpmd> flow validate 0 ingress pattern eth / vlan tci is 123 /
4685 vlan tci is 456 / end actions vf id 1 / queue index 0 / end
4686 Flow rule #0 validated
4688 testpmd> flow create 0 ingress pattern eth / vlan tci is 4 /
4689 vlan tci is 456 / end actions vf id 123 / queue index 0 / end
4690 Flow rule #0 created
4692 testpmd> flow list 0
4693 ID Group Prio Attr Rule
4694 0 0 0 i- ETH VLAN VLAN=>VF QUEUE
4696 Validate and create a QinQ rule on port 0 to steer traffic to a queue on the host.
4700 testpmd> flow validate 0 ingress pattern eth / vlan tci is 321 /
4701 vlan tci is 654 / end actions pf / queue index 0 / end
4702 Flow rule #1 validated
4704 testpmd> flow create 0 ingress pattern eth / vlan tci is 321 /
4705 vlan tci is 654 / end actions pf / queue index 1 / end
4706 Flow rule #1 created
4708 testpmd> flow list 0
4709 ID Group Prio Attr Rule
4710 0 0 0 i- ETH VLAN VLAN=>VF QUEUE
4711 1 0 0 i- ETH VLAN VLAN=>PF QUEUE
4713 Sample VXLAN flow rules
4714 ~~~~~~~~~~~~~~~~~~~~~~~
4716 Before creating VXLAN rule(s), the UDP port should be added for VXLAN packet
4719 testpmd> rx_vxlan_port add 4789 0
4721 Create VXLAN rules on port 0 to steer traffic to PF queues.
4725 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / vxlan /
4726 eth dst is 00:11:22:33:44:55 / end actions pf / queue index 1 / end
4727 Flow rule #0 created
4729 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / vxlan vni is 3 /
4730 eth dst is 00:11:22:33:44:55 / end actions pf / queue index 2 / end
4731 Flow rule #1 created
4733 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / vxlan /
4734 eth dst is 00:11:22:33:44:55 / vlan tci is 10 / end actions pf /
4736 Flow rule #2 created
4738 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / vxlan vni is 5 /
4739 eth dst is 00:11:22:33:44:55 / vlan tci is 20 / end actions pf /
4741 Flow rule #3 created
4743 testpmd> flow create 0 ingress pattern eth dst is 00:00:00:00:01:00 / ipv4 /
4744 udp / vxlan vni is 6 / eth dst is 00:11:22:33:44:55 / end actions pf /
4746 Flow rule #4 created
4748 testpmd> flow list 0
4749 ID Group Prio Attr Rule
4750 0 0 0 i- ETH IPV4 UDP VXLAN ETH => QUEUE
4751 1 0 0 i- ETH IPV4 UDP VXLAN ETH => QUEUE
4752 2 0 0 i- ETH IPV4 UDP VXLAN ETH VLAN => QUEUE
4753 3 0 0 i- ETH IPV4 UDP VXLAN ETH VLAN => QUEUE
4754 4 0 0 i- ETH IPV4 UDP VXLAN ETH => QUEUE
4756 Sample VXLAN encapsulation rule
4757 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4759 VXLAN encapsulation outer layer has default value pre-configured in testpmd
4760 source code, those can be changed by using the following commands
4762 IPv4 VXLAN outer header::
4764 testpmd> set vxlan ip-version ipv4 vni 4 udp-src 4 udp-dst 4 ip-src 127.0.0.1
4765 ip-dst 128.0.0.1 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4766 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4769 testpmd> set vxlan-with-vlan ip-version ipv4 vni 4 udp-src 4 udp-dst 4 ip-src
4770 127.0.0.1 ip-dst 128.0.0.1 vlan-tci 34 eth-src 11:11:11:11:11:11
4771 eth-dst 22:22:22:22:22:22
4772 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4775 testpmd> set vxlan-tos-ttl ip-version ipv4 vni 4 udp-src 4 udp-dst 4 ip-tos 0
4776 ip-ttl 255 ip-src 127.0.0.1 ip-dst 128.0.0.1 eth-src 11:11:11:11:11:11
4777 eth-dst 22:22:22:22:22:22
4778 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4781 IPv6 VXLAN outer header::
4783 testpmd> set vxlan ip-version ipv6 vni 4 udp-src 4 udp-dst 4 ip-src ::1
4784 ip-dst ::2222 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4785 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4788 testpmd> set vxlan-with-vlan ip-version ipv6 vni 4 udp-src 4 udp-dst 4
4789 ip-src ::1 ip-dst ::2222 vlan-tci 34 eth-src 11:11:11:11:11:11
4790 eth-dst 22:22:22:22:22:22
4791 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4794 testpmd> set vxlan-tos-ttl ip-version ipv6 vni 4 udp-src 4 udp-dst 4
4795 ip-tos 0 ip-ttl 255 ::1 ip-dst ::2222 eth-src 11:11:11:11:11:11
4796 eth-dst 22:22:22:22:22:22
4797 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4800 Sample NVGRE encapsulation rule
4801 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4803 NVGRE encapsulation outer layer has default value pre-configured in testpmd
4804 source code, those can be changed by using the following commands
4806 IPv4 NVGRE outer header::
4808 testpmd> set nvgre ip-version ipv4 tni 4 ip-src 127.0.0.1 ip-dst 128.0.0.1
4809 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4810 testpmd> flow create 0 ingress pattern end actions nvgre_encap /
4813 testpmd> set nvgre-with-vlan ip-version ipv4 tni 4 ip-src 127.0.0.1
4814 ip-dst 128.0.0.1 vlan-tci 34 eth-src 11:11:11:11:11:11
4815 eth-dst 22:22:22:22:22:22
4816 testpmd> flow create 0 ingress pattern end actions nvgre_encap /
4819 IPv6 NVGRE outer header::
4821 testpmd> set nvgre ip-version ipv6 tni 4 ip-src ::1 ip-dst ::2222
4822 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4823 testpmd> flow create 0 ingress pattern end actions nvgre_encap /
4826 testpmd> set nvgre-with-vlan ip-version ipv6 tni 4 ip-src ::1 ip-dst ::2222
4827 vlan-tci 34 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4828 testpmd> flow create 0 ingress pattern end actions nvgre_encap /
4831 Sample L2 encapsulation rule
4832 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4834 L2 encapsulation has default value pre-configured in testpmd
4835 source code, those can be changed by using the following commands
4839 testpmd> set l2_encap ip-version ipv4
4840 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4841 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / mpls / end actions
4842 mplsoudp_decap / l2_encap / end
4844 L2 with VXLAN header::
4846 testpmd> set l2_encap-with-vlan ip-version ipv4 vlan-tci 34
4847 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4848 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / mpls / end actions
4849 mplsoudp_decap / l2_encap / end
4851 Sample L2 decapsulation rule
4852 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4854 L2 decapsulation has default value pre-configured in testpmd
4855 source code, those can be changed by using the following commands
4859 testpmd> set l2_decap
4860 testpmd> flow create 0 egress pattern eth / end actions l2_decap / mplsoudp_encap /
4863 L2 with VXLAN header::
4865 testpmd> set l2_encap-with-vlan
4866 testpmd> flow create 0 egress pattern eth / end actions l2_encap / mplsoudp_encap /
4869 Sample MPLSoGRE encapsulation rule
4870 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4872 MPLSoGRE encapsulation outer layer has default value pre-configured in testpmd
4873 source code, those can be changed by using the following commands
4875 IPv4 MPLSoGRE outer header::
4877 testpmd> set mplsogre_encap ip-version ipv4 label 4
4878 ip-src 127.0.0.1 ip-dst 128.0.0.1 eth-src 11:11:11:11:11:11
4879 eth-dst 22:22:22:22:22:22
4880 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4881 mplsogre_encap / end
4883 IPv4 MPLSoGRE with VLAN outer header::
4885 testpmd> set mplsogre_encap-with-vlan ip-version ipv4 label 4
4886 ip-src 127.0.0.1 ip-dst 128.0.0.1 vlan-tci 34
4887 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4888 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4889 mplsogre_encap / end
4891 IPv6 MPLSoGRE outer header::
4893 testpmd> set mplsogre_encap ip-version ipv6 mask 4
4894 ip-src ::1 ip-dst ::2222 eth-src 11:11:11:11:11:11
4895 eth-dst 22:22:22:22:22:22
4896 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4897 mplsogre_encap / end
4899 IPv6 MPLSoGRE with VLAN outer header::
4901 testpmd> set mplsogre_encap-with-vlan ip-version ipv6 mask 4
4902 ip-src ::1 ip-dst ::2222 vlan-tci 34
4903 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4904 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4905 mplsogre_encap / end
4907 Sample MPLSoGRE decapsulation rule
4908 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4910 MPLSoGRE decapsulation outer layer has default value pre-configured in testpmd
4911 source code, those can be changed by using the following commands
4913 IPv4 MPLSoGRE outer header::
4915 testpmd> set mplsogre_decap ip-version ipv4
4916 testpmd> flow create 0 ingress pattern eth / ipv4 / gre / mpls / end actions
4917 mplsogre_decap / l2_encap / end
4919 IPv4 MPLSoGRE with VLAN outer header::
4921 testpmd> set mplsogre_decap-with-vlan ip-version ipv4
4922 testpmd> flow create 0 ingress pattern eth / vlan / ipv4 / gre / mpls / end
4923 actions mplsogre_decap / l2_encap / end
4925 IPv6 MPLSoGRE outer header::
4927 testpmd> set mplsogre_decap ip-version ipv6
4928 testpmd> flow create 0 ingress pattern eth / ipv6 / gre / mpls / end
4929 actions mplsogre_decap / l2_encap / end
4931 IPv6 MPLSoGRE with VLAN outer header::
4933 testpmd> set mplsogre_decap-with-vlan ip-version ipv6
4934 testpmd> flow create 0 ingress pattern eth / vlan / ipv6 / gre / mpls / end
4935 actions mplsogre_decap / l2_encap / end
4937 Sample MPLSoUDP encapsulation rule
4938 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4940 MPLSoUDP encapsulation outer layer has default value pre-configured in testpmd
4941 source code, those can be changed by using the following commands
4943 IPv4 MPLSoUDP outer header::
4945 testpmd> set mplsoudp_encap ip-version ipv4 label 4 udp-src 5 udp-dst 10
4946 ip-src 127.0.0.1 ip-dst 128.0.0.1 eth-src 11:11:11:11:11:11
4947 eth-dst 22:22:22:22:22:22
4948 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4949 mplsoudp_encap / end
4951 IPv4 MPLSoUDP with VLAN outer header::
4953 testpmd> set mplsoudp_encap-with-vlan ip-version ipv4 label 4 udp-src 5
4954 udp-dst 10 ip-src 127.0.0.1 ip-dst 128.0.0.1 vlan-tci 34
4955 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4956 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4957 mplsoudp_encap / end
4959 IPv6 MPLSoUDP outer header::
4961 testpmd> set mplsoudp_encap ip-version ipv6 mask 4 udp-src 5 udp-dst 10
4962 ip-src ::1 ip-dst ::2222 eth-src 11:11:11:11:11:11
4963 eth-dst 22:22:22:22:22:22
4964 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4965 mplsoudp_encap / end
4967 IPv6 MPLSoUDP with VLAN outer header::
4969 testpmd> set mplsoudp_encap-with-vlan ip-version ipv6 mask 4 udp-src 5
4970 udp-dst 10 ip-src ::1 ip-dst ::2222 vlan-tci 34
4971 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4972 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4973 mplsoudp_encap / end
4975 Sample MPLSoUDP decapsulation rule
4976 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4978 MPLSoUDP decapsulation outer layer has default value pre-configured in testpmd
4979 source code, those can be changed by using the following commands
4981 IPv4 MPLSoUDP outer header::
4983 testpmd> set mplsoudp_decap ip-version ipv4
4984 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / mpls / end actions
4985 mplsoudp_decap / l2_encap / end
4987 IPv4 MPLSoUDP with VLAN outer header::
4989 testpmd> set mplsoudp_decap-with-vlan ip-version ipv4
4990 testpmd> flow create 0 ingress pattern eth / vlan / ipv4 / udp / mpls / end
4991 actions mplsoudp_decap / l2_encap / end
4993 IPv6 MPLSoUDP outer header::
4995 testpmd> set mplsoudp_decap ip-version ipv6
4996 testpmd> flow create 0 ingress pattern eth / ipv6 / udp / mpls / end
4997 actions mplsoudp_decap / l2_encap / end
4999 IPv6 MPLSoUDP with VLAN outer header::
5001 testpmd> set mplsoudp_decap-with-vlan ip-version ipv6
5002 testpmd> flow create 0 ingress pattern eth / vlan / ipv6 / udp / mpls / end
5003 actions mplsoudp_decap / l2_encap / end
5005 Sample Raw encapsulation rule
5006 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
5008 Raw encapsulation configuration can be set by the following commands
5010 Eecapsulating VxLAN::
5012 testpmd> set raw_encap 4 eth src is 10:11:22:33:44:55 / vlan tci is 1
5013 inner_type is 0x0800 / ipv4 / udp dst is 4789 / vxlan vni
5015 testpmd> flow create 0 egress pattern eth / ipv4 / end actions
5016 raw_encap index 4 / end
5018 Sample Raw decapsulation rule
5019 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
5021 Raw decapsulation configuration can be set by the following commands
5023 Decapsulating VxLAN::
5025 testpmd> set raw_decap eth / ipv4 / udp / vxlan / end_set
5026 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / vxlan / eth / ipv4 /
5027 end actions raw_decap / queue index 0 / end
5032 ESP rules can be created by the following commands::
5034 testpmd> flow create 0 ingress pattern eth / ipv4 / esp spi is 1 / end actions
5036 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / esp spi is 1 / end
5037 actions queue index 3 / end
5038 testpmd> flow create 0 ingress pattern eth / ipv6 / esp spi is 1 / end actions
5040 testpmd> flow create 0 ingress pattern eth / ipv6 / udp / esp spi is 1 / end
5041 actions queue index 3 / end
5046 AH rules can be created by the following commands::
5048 testpmd> flow create 0 ingress pattern eth / ipv4 / ah spi is 1 / end actions
5050 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / ah spi is 1 / end
5051 actions queue index 3 / end
5052 testpmd> flow create 0 ingress pattern eth / ipv6 / ah spi is 1 / end actions
5054 testpmd> flow create 0 ingress pattern eth / ipv6 / udp / ah spi is 1 / end
5055 actions queue index 3 / end
5060 PFCP rules can be created by the following commands(s_field need to be 1
5063 testpmd> flow create 0 ingress pattern eth / ipv4 / pfcp s_field is 0 / end
5064 actions queue index 3 / end
5065 testpmd> flow create 0 ingress pattern eth / ipv4 / pfcp s_field is 1
5066 seid is 1 / end actions queue index 3 / end
5067 testpmd> flow create 0 ingress pattern eth / ipv6 / pfcp s_field is 0 / end
5068 actions queue index 3 / end
5069 testpmd> flow create 0 ingress pattern eth / ipv6 / pfcp s_field is 1
5070 seid is 1 / end actions queue index 3 / end
5075 The following sections show functions to load/unload eBPF based filters.
5080 Load an eBPF program as a callback for particular RX/TX queue::
5082 testpmd> bpf-load rx|tx (portid) (queueid) (load-flags) (bpf-prog-filename)
5084 The available load-flags are:
5086 * ``J``: use JIT generated native code, otherwise BPF interpreter will be used.
5088 * ``M``: assume input parameter is a pointer to rte_mbuf, otherwise assume it is a pointer to first segment's data.
5094 You'll need clang v3.7 or above to build bpf program you'd like to load
5098 .. code-block:: console
5101 clang -O2 -target bpf -c t1.c
5103 Then to load (and JIT compile) t1.o at RX queue 0, port 1:
5105 .. code-block:: console
5107 testpmd> bpf-load rx 1 0 J ./dpdk.org/examples/bpf/t1.o
5109 To load (not JITed) t1.o at TX queue 0, port 0:
5111 .. code-block:: console
5113 testpmd> bpf-load tx 0 0 - ./dpdk.org/examples/bpf/t1.o
5118 Unload previously loaded eBPF program for particular RX/TX queue::
5120 testpmd> bpf-unload rx|tx (portid) (queueid)
5122 For example to unload BPF filter from TX queue 0, port 0:
5124 .. code-block:: console
5126 testpmd> bpf-unload tx 0 0