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)
1715 Load a dynamic device personalization (DDP) profile and store backup profile::
1717 testpmd> ddp add (port_id) (profile_path[,backup_profile_path])
1722 Delete a dynamic device personalization profile and restore backup profile::
1724 testpmd> ddp del (port_id) (backup_profile_path)
1729 List all items from the ptype mapping table::
1731 testpmd> ptype mapping get (port_id) (valid_only)
1735 * ``valid_only``: A flag indicates if only list valid items(=1) or all itemss(=0).
1737 Replace a specific or a group of software defined ptype with a new one::
1739 testpmd> ptype mapping replace (port_id) (target) (mask) (pkt_type)
1743 * ``target``: A specific software ptype or a mask to represent a group of software ptypes.
1745 * ``mask``: A flag indicate if "target" is a specific software ptype(=0) or a ptype mask(=1).
1747 * ``pkt_type``: The new software ptype to replace the old ones.
1749 Update hardware defined ptype to software defined packet type mapping table::
1751 testpmd> ptype mapping update (port_id) (hw_ptype) (sw_ptype)
1755 * ``hw_ptype``: hardware ptype as the index of the ptype mapping table.
1757 * ``sw_ptype``: software ptype as the value of the ptype mapping table.
1759 Reset ptype mapping table::
1761 testpmd> ptype mapping reset (port_id)
1763 config per port Rx offloading
1764 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1766 Enable or disable a per port Rx offloading on all Rx queues of a port::
1768 testpmd> port config (port_id) rx_offload (offloading) on|off
1770 * ``offloading``: can be any of these offloading capability:
1771 vlan_strip, ipv4_cksum, udp_cksum, tcp_cksum, tcp_lro,
1772 qinq_strip, outer_ipv4_cksum, macsec_strip,
1773 header_split, vlan_filter, vlan_extend, jumbo_frame,
1774 scatter, timestamp, security, keep_crc, rss_hash
1776 This command should be run when the port is stopped, or else it will fail.
1778 config per queue Rx offloading
1779 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1781 Enable or disable a per queue Rx offloading only on a specific Rx queue::
1783 testpmd> port (port_id) rxq (queue_id) rx_offload (offloading) on|off
1785 * ``offloading``: can be any of these offloading capability:
1786 vlan_strip, ipv4_cksum, udp_cksum, tcp_cksum, tcp_lro,
1787 qinq_strip, outer_ipv4_cksum, macsec_strip,
1788 header_split, vlan_filter, vlan_extend, jumbo_frame,
1789 scatter, timestamp, security, keep_crc
1791 This command should be run when the port is stopped, or else it will fail.
1793 config per port Tx offloading
1794 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1796 Enable or disable a per port Tx offloading on all Tx queues of a port::
1798 testpmd> port config (port_id) tx_offload (offloading) on|off
1800 * ``offloading``: can be any of these offloading capability:
1801 vlan_insert, ipv4_cksum, udp_cksum, tcp_cksum,
1802 sctp_cksum, tcp_tso, udp_tso, outer_ipv4_cksum,
1803 qinq_insert, vxlan_tnl_tso, gre_tnl_tso,
1804 ipip_tnl_tso, geneve_tnl_tso, macsec_insert,
1805 mt_lockfree, multi_segs, mbuf_fast_free, security
1807 This command should be run when the port is stopped, or else it will fail.
1809 config per queue Tx offloading
1810 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1812 Enable or disable a per queue Tx offloading only on a specific Tx queue::
1814 testpmd> port (port_id) txq (queue_id) tx_offload (offloading) on|off
1816 * ``offloading``: can be any of these offloading capability:
1817 vlan_insert, ipv4_cksum, udp_cksum, tcp_cksum,
1818 sctp_cksum, tcp_tso, udp_tso, outer_ipv4_cksum,
1819 qinq_insert, vxlan_tnl_tso, gre_tnl_tso,
1820 ipip_tnl_tso, geneve_tnl_tso, macsec_insert,
1821 mt_lockfree, multi_segs, mbuf_fast_free, security
1823 This command should be run when the port is stopped, or else it will fail.
1825 Config VXLAN Encap outer layers
1826 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1828 Configure the outer layer to encapsulate a packet inside a VXLAN tunnel::
1830 set vxlan ip-version (ipv4|ipv6) vni (vni) udp-src (udp-src) \
1831 udp-dst (udp-dst) ip-src (ip-src) ip-dst (ip-dst) eth-src (eth-src) \
1834 set vxlan-with-vlan ip-version (ipv4|ipv6) vni (vni) udp-src (udp-src) \
1835 udp-dst (udp-dst) ip-src (ip-src) ip-dst (ip-dst) vlan-tci (vlan-tci) \
1836 eth-src (eth-src) eth-dst (eth-dst)
1838 set vxlan-tos-ttl ip-version (ipv4|ipv6) vni (vni) udp-src (udp-src) \
1839 udp-dst (udp-dst) ip-tos (ip-tos) ip-ttl (ip-ttl) ip-src (ip-src) \
1840 ip-dst (ip-dst) eth-src (eth-src) eth-dst (eth-dst)
1842 These commands will set an internal configuration inside testpmd, any following
1843 flow rule using the action vxlan_encap will use the last configuration set.
1844 To have a different encapsulation header, one of those commands must be called
1845 before the flow rule creation.
1847 Config NVGRE Encap outer layers
1848 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1850 Configure the outer layer to encapsulate a packet inside a NVGRE tunnel::
1852 set nvgre ip-version (ipv4|ipv6) tni (tni) ip-src (ip-src) ip-dst (ip-dst) \
1853 eth-src (eth-src) eth-dst (eth-dst)
1854 set nvgre-with-vlan ip-version (ipv4|ipv6) tni (tni) ip-src (ip-src) \
1855 ip-dst (ip-dst) vlan-tci (vlan-tci) eth-src (eth-src) eth-dst (eth-dst)
1857 These commands will set an internal configuration inside testpmd, any following
1858 flow rule using the action nvgre_encap will use the last configuration set.
1859 To have a different encapsulation header, one of those commands must be called
1860 before the flow rule creation.
1865 Configure the l2 to be used when encapsulating a packet with L2::
1867 set l2_encap ip-version (ipv4|ipv6) eth-src (eth-src) eth-dst (eth-dst)
1868 set l2_encap-with-vlan ip-version (ipv4|ipv6) vlan-tci (vlan-tci) \
1869 eth-src (eth-src) eth-dst (eth-dst)
1871 Those commands will set an internal configuration inside testpmd, any following
1872 flow rule using the action l2_encap will use the last configuration set.
1873 To have a different encapsulation header, one of those commands must be called
1874 before the flow rule creation.
1879 Configure the l2 to be removed when decapsulating a packet with L2::
1881 set l2_decap ip-version (ipv4|ipv6)
1882 set l2_decap-with-vlan ip-version (ipv4|ipv6)
1884 Those commands will set an internal configuration inside testpmd, any following
1885 flow rule using the action l2_decap will use the last configuration set.
1886 To have a different encapsulation header, one of those commands must be called
1887 before the flow rule creation.
1889 Config MPLSoGRE Encap outer layers
1890 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1892 Configure the outer layer to encapsulate a packet inside a MPLSoGRE tunnel::
1894 set mplsogre_encap ip-version (ipv4|ipv6) label (label) \
1895 ip-src (ip-src) ip-dst (ip-dst) eth-src (eth-src) eth-dst (eth-dst)
1896 set mplsogre_encap-with-vlan ip-version (ipv4|ipv6) label (label) \
1897 ip-src (ip-src) ip-dst (ip-dst) vlan-tci (vlan-tci) \
1898 eth-src (eth-src) eth-dst (eth-dst)
1900 These commands will set an internal configuration inside testpmd, any following
1901 flow rule using the action mplsogre_encap will use the last configuration set.
1902 To have a different encapsulation header, one of those commands must be called
1903 before the flow rule creation.
1905 Config MPLSoGRE Decap outer layers
1906 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1908 Configure the outer layer to decapsulate MPLSoGRE packet::
1910 set mplsogre_decap ip-version (ipv4|ipv6)
1911 set mplsogre_decap-with-vlan ip-version (ipv4|ipv6)
1913 These commands will set an internal configuration inside testpmd, any following
1914 flow rule using the action mplsogre_decap will use the last configuration set.
1915 To have a different decapsulation header, one of those commands must be called
1916 before the flow rule creation.
1918 Config MPLSoUDP Encap outer layers
1919 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1921 Configure the outer layer to encapsulate a packet inside a MPLSoUDP tunnel::
1923 set mplsoudp_encap ip-version (ipv4|ipv6) label (label) udp-src (udp-src) \
1924 udp-dst (udp-dst) ip-src (ip-src) ip-dst (ip-dst) \
1925 eth-src (eth-src) eth-dst (eth-dst)
1926 set mplsoudp_encap-with-vlan ip-version (ipv4|ipv6) label (label) \
1927 udp-src (udp-src) udp-dst (udp-dst) ip-src (ip-src) ip-dst (ip-dst) \
1928 vlan-tci (vlan-tci) eth-src (eth-src) eth-dst (eth-dst)
1930 These commands will set an internal configuration inside testpmd, any following
1931 flow rule using the action mplsoudp_encap will use the last configuration set.
1932 To have a different encapsulation header, one of those commands must be called
1933 before the flow rule creation.
1935 Config MPLSoUDP Decap outer layers
1936 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1938 Configure the outer layer to decapsulate MPLSoUDP packet::
1940 set mplsoudp_decap ip-version (ipv4|ipv6)
1941 set mplsoudp_decap-with-vlan ip-version (ipv4|ipv6)
1943 These commands will set an internal configuration inside testpmd, any following
1944 flow rule using the action mplsoudp_decap will use the last configuration set.
1945 To have a different decapsulation header, one of those commands must be called
1946 before the flow rule creation.
1948 Config Raw Encapsulation
1949 ~~~~~~~~~~~~~~~~~~~~~~~~~
1951 Configure the raw data to be used when encapsulating a packet by
1952 rte_flow_action_raw_encap::
1954 set raw_encap {index} {item} [/ {item} [...]] / end_set
1956 There are multiple global buffers for ``raw_encap``, this command will set one
1957 internal buffer index by ``{index}``.
1958 If there is no ``{index}`` specified::
1960 set raw_encap {item} [/ {item} [...]] / end_set
1962 the default index ``0`` is used.
1963 In order to use different encapsulating header, ``index`` must be specified
1964 during the flow rule creation::
1966 testpmd> flow create 0 egress pattern eth / ipv4 / end actions
1967 raw_encap index 2 / end
1969 Otherwise the default index ``0`` is used.
1971 Config Raw Decapsulation
1972 ~~~~~~~~~~~~~~~~~~~~~~~~
1974 Configure the raw data to be used when decapsulating a packet by
1975 rte_flow_action_raw_decap::
1977 set raw_decap {index} {item} [/ {item} [...]] / end_set
1979 There are multiple global buffers for ``raw_decap``, this command will set
1980 one internal buffer index by ``{index}``.
1981 If there is no ``{index}`` specified::
1983 set raw_decap {item} [/ {item} [...]] / end_set
1985 the default index ``0`` is used.
1986 In order to use different decapsulating header, ``index`` must be specified
1987 during the flow rule creation::
1989 testpmd> flow create 0 egress pattern eth / ipv4 / end actions
1990 raw_encap index 3 / end
1992 Otherwise the default index ``0`` is used.
1997 Set fec mode for a specific port::
1999 testpmd> set port (port_id) fec_mode auto|off|rs|baser
2005 The following sections show functions for configuring ports.
2009 Port configuration changes only become active when forwarding is started/restarted.
2014 Attach a port specified by pci address or virtual device args::
2016 testpmd> port attach (identifier)
2018 To attach a new pci device, the device should be recognized by kernel first.
2019 Then it should be moved under DPDK management.
2020 Finally the port can be attached to testpmd.
2022 For example, to move a pci device using ixgbe under DPDK management:
2024 .. code-block:: console
2026 # Check the status of the available devices.
2027 ./usertools/dpdk-devbind.py --status
2029 Network devices using DPDK-compatible driver
2030 ============================================
2033 Network devices using kernel driver
2034 ===================================
2035 0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused=
2038 # Bind the device to igb_uio.
2039 sudo ./usertools/dpdk-devbind.py -b igb_uio 0000:0a:00.0
2042 # Recheck the status of the devices.
2043 ./usertools/dpdk-devbind.py --status
2044 Network devices using DPDK-compatible driver
2045 ============================================
2046 0000:0a:00.0 '82599ES 10-Gigabit' drv=igb_uio unused=
2048 To attach a port created by virtual device, above steps are not needed.
2050 For example, to attach a port whose pci address is 0000:0a:00.0.
2052 .. code-block:: console
2054 testpmd> port attach 0000:0a:00.0
2055 Attaching a new port...
2056 EAL: PCI device 0000:0a:00.0 on NUMA socket -1
2057 EAL: probe driver: 8086:10fb rte_ixgbe_pmd
2058 EAL: PCI memory mapped at 0x7f83bfa00000
2059 EAL: PCI memory mapped at 0x7f83bfa80000
2060 PMD: eth_ixgbe_dev_init(): MAC: 2, PHY: 18, SFP+: 5
2061 PMD: eth_ixgbe_dev_init(): port 0 vendorID=0x8086 deviceID=0x10fb
2062 Port 0 is attached. Now total ports is 1
2065 For example, to attach a port created by pcap PMD.
2067 .. code-block:: console
2069 testpmd> port attach net_pcap0
2070 Attaching a new port...
2071 PMD: Initializing pmd_pcap for net_pcap0
2072 PMD: Creating pcap-backed ethdev on numa socket 0
2073 Port 0 is attached. Now total ports is 1
2076 In this case, identifier is ``net_pcap0``.
2077 This identifier format is the same as ``--vdev`` format of DPDK applications.
2079 For example, to re-attach a bonded port which has been previously detached,
2080 the mode and slave parameters must be given.
2082 .. code-block:: console
2084 testpmd> port attach net_bond_0,mode=0,slave=1
2085 Attaching a new port...
2086 EAL: Initializing pmd_bond for net_bond_0
2087 EAL: Create bonded device net_bond_0 on port 0 in mode 0 on socket 0.
2088 Port 0 is attached. Now total ports is 1
2095 Detach a specific port::
2097 testpmd> port detach (port_id)
2099 Before detaching a port, the port should be stopped and closed.
2101 For example, to detach a pci device port 0.
2103 .. code-block:: console
2105 testpmd> port stop 0
2108 testpmd> port close 0
2112 testpmd> port detach 0
2114 EAL: PCI device 0000:0a:00.0 on NUMA socket -1
2115 EAL: remove driver: 8086:10fb rte_ixgbe_pmd
2116 EAL: PCI memory unmapped at 0x7f83bfa00000
2117 EAL: PCI memory unmapped at 0x7f83bfa80000
2121 For example, to detach a virtual device port 0.
2123 .. code-block:: console
2125 testpmd> port stop 0
2128 testpmd> port close 0
2132 testpmd> port detach 0
2134 PMD: Closing pcap ethdev on numa socket 0
2135 Port 'net_pcap0' is detached. Now total ports is 0
2138 To remove a pci device completely from the system, first detach the port from testpmd.
2139 Then the device should be moved under kernel management.
2140 Finally the device can be removed using kernel pci hotplug functionality.
2142 For example, to move a pci device under kernel management:
2144 .. code-block:: console
2146 sudo ./usertools/dpdk-devbind.py -b ixgbe 0000:0a:00.0
2148 ./usertools/dpdk-devbind.py --status
2150 Network devices using DPDK-compatible driver
2151 ============================================
2154 Network devices using kernel driver
2155 ===================================
2156 0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused=igb_uio
2158 To remove a port created by a virtual device, above steps are not needed.
2163 Start all ports or a specific port::
2165 testpmd> port start (port_id|all)
2170 Stop all ports or a specific port::
2172 testpmd> port stop (port_id|all)
2177 Close all ports or a specific port::
2179 testpmd> port close (port_id|all)
2184 Reset all ports or a specific port::
2186 testpmd> port reset (port_id|all)
2188 User should stop port(s) before resetting and (re-)start after reset.
2190 port config - queue ring size
2191 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2193 Configure a rx/tx queue ring size::
2195 testpmd> port (port_id) (rxq|txq) (queue_id) ring_size (value)
2197 Only take effect after command that (re-)start the port or command that setup specific queue.
2199 port start/stop queue
2200 ~~~~~~~~~~~~~~~~~~~~~
2202 Start/stop a rx/tx queue on a specific port::
2204 testpmd> port (port_id) (rxq|txq) (queue_id) (start|stop)
2206 port config - queue deferred start
2207 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2209 Switch on/off deferred start of a specific port queue::
2211 testpmd> port (port_id) (rxq|txq) (queue_id) deferred_start (on|off)
2214 ~~~~~~~~~~~~~~~~~~~~~
2216 Setup a rx/tx queue on a specific port::
2218 testpmd> port (port_id) (rxq|txq) (queue_id) setup
2220 Only take effect when port is started.
2225 Set the speed and duplex mode for all ports or a specific port::
2227 testpmd> port config (port_id|all) speed (10|100|1000|10000|25000|40000|50000|100000|200000|auto) \
2228 duplex (half|full|auto)
2230 port config - queues/descriptors
2231 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2233 Set number of queues/descriptors for rxq, txq, rxd and txd::
2235 testpmd> port config all (rxq|txq|rxd|txd) (value)
2237 This is equivalent to the ``--rxq``, ``--txq``, ``--rxd`` and ``--txd`` command-line options.
2239 port config - max-pkt-len
2240 ~~~~~~~~~~~~~~~~~~~~~~~~~
2242 Set the maximum packet length::
2244 testpmd> port config all max-pkt-len (value)
2246 This is equivalent to the ``--max-pkt-len`` command-line option.
2248 port config - max-lro-pkt-size
2249 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2251 Set the maximum LRO aggregated packet size::
2253 testpmd> port config all max-lro-pkt-size (value)
2255 This is equivalent to the ``--max-lro-pkt-size`` command-line option.
2257 port config - Drop Packets
2258 ~~~~~~~~~~~~~~~~~~~~~~~~~~
2260 Enable or disable packet drop on all RX queues of all ports when no receive buffers available::
2262 testpmd> port config all drop-en (on|off)
2264 Packet dropping when no receive buffers available is off by default.
2266 The ``on`` option is equivalent to the ``--enable-drop-en`` command-line option.
2271 Set the RSS (Receive Side Scaling) mode on or off::
2273 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)
2275 RSS is on by default.
2277 The ``all`` option is equivalent to eth|vlan|ip|tcp|udp|sctp|ether|l2tpv3|esp|ah|pfcp.
2279 The ``default`` option enables all supported RSS types reported by device info.
2281 The ``none`` option is equivalent to the ``--disable-rss`` command-line option.
2283 port config - RSS Reta
2284 ~~~~~~~~~~~~~~~~~~~~~~
2286 Set the RSS (Receive Side Scaling) redirection table::
2288 testpmd> port config all rss reta (hash,queue)[,(hash,queue)]
2293 Set the DCB mode for an individual port::
2295 testpmd> port config (port_id) dcb vt (on|off) (traffic_class) pfc (on|off)
2297 The traffic class should be 4 or 8.
2302 Set the number of packets per burst::
2304 testpmd> port config all burst (value)
2306 This is equivalent to the ``--burst`` command-line option.
2308 port config - Threshold
2309 ~~~~~~~~~~~~~~~~~~~~~~~
2311 Set thresholds for TX/RX queues::
2313 testpmd> port config all (threshold) (value)
2315 Where the threshold type can be:
2317 * ``txpt:`` Set the prefetch threshold register of the TX rings, 0 <= value <= 255.
2319 * ``txht:`` Set the host threshold register of the TX rings, 0 <= value <= 255.
2321 * ``txwt:`` Set the write-back threshold register of the TX rings, 0 <= value <= 255.
2323 * ``rxpt:`` Set the prefetch threshold register of the RX rings, 0 <= value <= 255.
2325 * ``rxht:`` Set the host threshold register of the RX rings, 0 <= value <= 255.
2327 * ``rxwt:`` Set the write-back threshold register of the RX rings, 0 <= value <= 255.
2329 * ``txfreet:`` Set the transmit free threshold of the TX rings, 0 <= value <= txd.
2331 * ``rxfreet:`` Set the transmit free threshold of the RX rings, 0 <= value <= rxd.
2333 * ``txrst:`` Set the transmit RS bit threshold of TX rings, 0 <= value <= txd.
2335 These threshold options are also available from the command-line.
2340 Set the value of ether-type for E-tag::
2342 testpmd> port config (port_id|all) l2-tunnel E-tag ether-type (value)
2344 Enable/disable the E-tag support::
2346 testpmd> port config (port_id|all) l2-tunnel E-tag (enable|disable)
2348 port config pctype mapping
2349 ~~~~~~~~~~~~~~~~~~~~~~~~~~
2351 Reset pctype mapping table::
2353 testpmd> port config (port_id) pctype mapping reset
2355 Update hardware defined pctype to software defined flow type mapping table::
2357 testpmd> port config (port_id) pctype mapping update (pctype_id_0[,pctype_id_1]*) (flow_type_id)
2361 * ``pctype_id_x``: hardware pctype id as index of bit in bitmask value of the pctype mapping table.
2363 * ``flow_type_id``: software flow type id as the index of the pctype mapping table.
2365 port config input set
2366 ~~~~~~~~~~~~~~~~~~~~~
2368 Config RSS/FDIR/FDIR flexible payload input set for some pctype::
2370 testpmd> port config (port_id) pctype (pctype_id) \
2371 (hash_inset|fdir_inset|fdir_flx_inset) \
2372 (get|set|clear) field (field_idx)
2374 Clear RSS/FDIR/FDIR flexible payload input set for some pctype::
2376 testpmd> port config (port_id) pctype (pctype_id) \
2377 (hash_inset|fdir_inset|fdir_flx_inset) clear all
2381 * ``pctype_id``: hardware packet classification types.
2382 * ``field_idx``: hardware field index.
2384 port config udp_tunnel_port
2385 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
2387 Add/remove UDP tunnel port for VXLAN/GENEVE tunneling protocols::
2389 testpmd> port config (port_id) udp_tunnel_port add|rm vxlan|geneve|vxlan-gpe (udp_port)
2391 port config tx_metadata
2392 ~~~~~~~~~~~~~~~~~~~~~~~
2394 Set Tx metadata value per port.
2395 testpmd will add this value to any Tx packet sent from this port::
2397 testpmd> port config (port_id) tx_metadata (value)
2402 Set/clear dynamic flag per port.
2403 testpmd will register this flag in the mbuf (same registration
2404 for both Tx and Rx). Then set/clear this flag for each Tx
2405 packet sent from this port. The set bit only works for Tx packet::
2407 testpmd> port config (port_id) dynf (name) (set|clear)
2412 To configure MTU(Maximum Transmission Unit) on devices using testpmd::
2414 testpmd> port config mtu (port_id) (value)
2416 port config rss hash key
2417 ~~~~~~~~~~~~~~~~~~~~~~~~
2419 To configure the RSS hash key used to compute the RSS
2420 hash of input [IP] packets received on port::
2422 testpmd> port config <port_id> rss-hash-key (ipv4|ipv4-frag|\
2423 ipv4-tcp|ipv4-udp|ipv4-sctp|ipv4-other|\
2424 ipv6|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|\
2425 ipv6-other|l2-payload|ipv6-ex|ipv6-tcp-ex|\
2426 ipv6-udp-ex <string of hex digits \
2427 (variable length, NIC dependent)>)
2432 The following sections show functions for device operations.
2437 Detach a device specified by pci address or virtual device args::
2439 testpmd> device detach (identifier)
2441 Before detaching a device associated with ports, the ports should be stopped and closed.
2443 For example, to detach a pci device whose address is 0002:03:00.0.
2445 .. code-block:: console
2447 testpmd> device detach 0002:03:00.0
2448 Removing a device...
2449 Port 1 is now closed
2450 EAL: Releasing pci mapped resource for 0002:03:00.0
2451 EAL: Calling pci_unmap_resource for 0002:03:00.0 at 0x218a050000
2452 EAL: Calling pci_unmap_resource for 0002:03:00.0 at 0x218c050000
2453 Device 0002:03:00.0 is detached
2454 Now total ports is 1
2456 For example, to detach a port created by pcap PMD.
2458 .. code-block:: console
2460 testpmd> device detach net_pcap0
2461 Removing a device...
2462 Port 0 is now closed
2463 Device net_pcap0 is detached
2464 Now total ports is 0
2467 In this case, identifier is ``net_pcap0``.
2468 This identifier format is the same as ``--vdev`` format of DPDK applications.
2470 Link Bonding Functions
2471 ----------------------
2473 The Link Bonding functions make it possible to dynamically create and
2474 manage link bonding devices from within testpmd interactive prompt.
2476 create bonded device
2477 ~~~~~~~~~~~~~~~~~~~~
2479 Create a new bonding device::
2481 testpmd> create bonded device (mode) (socket)
2483 For example, to create a bonded device in mode 1 on socket 0::
2485 testpmd> create bonded device 1 0
2486 created new bonded device (port X)
2491 Adds Ethernet device to a Link Bonding device::
2493 testpmd> add bonding slave (slave id) (port id)
2495 For example, to add Ethernet device (port 6) to a Link Bonding device (port 10)::
2497 testpmd> add bonding slave 6 10
2500 remove bonding slave
2501 ~~~~~~~~~~~~~~~~~~~~
2503 Removes an Ethernet slave device from a Link Bonding device::
2505 testpmd> remove bonding slave (slave id) (port id)
2507 For example, to remove Ethernet slave device (port 6) to a Link Bonding device (port 10)::
2509 testpmd> remove bonding slave 6 10
2514 Set the Link Bonding mode of a Link Bonding device::
2516 testpmd> set bonding mode (value) (port id)
2518 For example, to set the bonding mode of a Link Bonding device (port 10) to broadcast (mode 3)::
2520 testpmd> set bonding mode 3 10
2525 Set an Ethernet slave device as the primary device on a Link Bonding device::
2527 testpmd> set bonding primary (slave id) (port id)
2529 For example, to set the Ethernet slave device (port 6) as the primary port of a Link Bonding device (port 10)::
2531 testpmd> set bonding primary 6 10
2536 Set the MAC address of a Link Bonding device::
2538 testpmd> set bonding mac (port id) (mac)
2540 For example, to set the MAC address of a Link Bonding device (port 10) to 00:00:00:00:00:01::
2542 testpmd> set bonding mac 10 00:00:00:00:00:01
2544 set bonding balance_xmit_policy
2545 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2547 Set the transmission policy for a Link Bonding device when it is in Balance XOR mode::
2549 testpmd> set bonding balance_xmit_policy (port_id) (l2|l23|l34)
2551 For example, set a Link Bonding device (port 10) to use a balance policy of layer 3+4 (IP addresses & UDP ports)::
2553 testpmd> set bonding balance_xmit_policy 10 l34
2556 set bonding mon_period
2557 ~~~~~~~~~~~~~~~~~~~~~~
2559 Set the link status monitoring polling period in milliseconds for a bonding device.
2561 This adds support for PMD slave devices which do not support link status interrupts.
2562 When the mon_period is set to a value greater than 0 then all PMD's which do not support
2563 link status ISR will be queried every polling interval to check if their link status has changed::
2565 testpmd> set bonding mon_period (port_id) (value)
2567 For example, to set the link status monitoring polling period of bonded device (port 5) to 150ms::
2569 testpmd> set bonding mon_period 5 150
2572 set bonding lacp dedicated_queue
2573 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2575 Enable dedicated tx/rx queues on bonding devices slaves to handle LACP control plane traffic
2576 when in mode 4 (link-aggregation-802.3ad)::
2578 testpmd> set bonding lacp dedicated_queues (port_id) (enable|disable)
2581 set bonding agg_mode
2582 ~~~~~~~~~~~~~~~~~~~~
2584 Enable one of the specific aggregators mode when in mode 4 (link-aggregation-802.3ad)::
2586 testpmd> set bonding agg_mode (port_id) (bandwidth|count|stable)
2592 Show the current configuration of a Link Bonding device::
2594 testpmd> show bonding config (port id)
2597 to show the configuration a Link Bonding device (port 9) with 3 slave devices (1, 3, 4)
2598 in balance mode with a transmission policy of layer 2+3::
2600 testpmd> show bonding config 9
2602 Balance Xmit Policy: BALANCE_XMIT_POLICY_LAYER23
2604 Active Slaves (3): [1 3 4]
2611 The Register Functions can be used to read from and write to registers on the network card referenced by a port number.
2612 This is mainly useful for debugging purposes.
2613 Reference should be made to the appropriate datasheet for the network card for details on the register addresses
2614 and fields that can be accessed.
2619 Display the value of a port register::
2621 testpmd> read reg (port_id) (address)
2623 For example, to examine the Flow Director control register (FDIRCTL, 0x0000EE000) on an Intel 82599 10 GbE Controller::
2625 testpmd> read reg 0 0xEE00
2626 port 0 PCI register at offset 0xEE00: 0x4A060029 (1241907241)
2631 Display a port register bit field::
2633 testpmd> read regfield (port_id) (address) (bit_x) (bit_y)
2635 For example, reading the lowest two bits from the register in the example above::
2637 testpmd> read regfield 0 0xEE00 0 1
2638 port 0 PCI register at offset 0xEE00: bits[0, 1]=0x1 (1)
2643 Display a single port register bit::
2645 testpmd> read regbit (port_id) (address) (bit_x)
2647 For example, reading the lowest bit from the register in the example above::
2649 testpmd> read regbit 0 0xEE00 0
2650 port 0 PCI register at offset 0xEE00: bit 0=1
2655 Set the value of a port register::
2657 testpmd> write reg (port_id) (address) (value)
2659 For example, to clear a register::
2661 testpmd> write reg 0 0xEE00 0x0
2662 port 0 PCI register at offset 0xEE00: 0x00000000 (0)
2667 Set bit field of a port register::
2669 testpmd> write regfield (port_id) (address) (bit_x) (bit_y) (value)
2671 For example, writing to the register cleared in the example above::
2673 testpmd> write regfield 0 0xEE00 0 1 2
2674 port 0 PCI register at offset 0xEE00: 0x00000002 (2)
2679 Set single bit value of a port register::
2681 testpmd> write regbit (port_id) (address) (bit_x) (value)
2683 For example, to set the high bit in the register from the example above::
2685 testpmd> write regbit 0 0xEE00 31 1
2686 port 0 PCI register at offset 0xEE00: 0x8000000A (2147483658)
2688 Traffic Metering and Policing
2689 -----------------------------
2691 The following section shows functions for configuring traffic metering and
2692 policing on the ethernet device through the use of generic ethdev API.
2694 show port traffic management capability
2695 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2697 Show traffic metering and policing capability of the port::
2699 testpmd> show port meter cap (port_id)
2701 add port meter profile (srTCM rfc2967)
2702 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2704 Add meter profile (srTCM rfc2697) to the ethernet device::
2706 testpmd> add port meter profile srtcm_rfc2697 (port_id) (profile_id) \
2711 * ``profile_id``: ID for the meter profile.
2712 * ``cir``: Committed Information Rate (CIR) (bytes/second).
2713 * ``cbs``: Committed Burst Size (CBS) (bytes).
2714 * ``ebs``: Excess Burst Size (EBS) (bytes).
2716 add port meter profile (trTCM rfc2968)
2717 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2719 Add meter profile (srTCM rfc2698) to the ethernet device::
2721 testpmd> add port meter profile trtcm_rfc2698 (port_id) (profile_id) \
2722 (cir) (pir) (cbs) (pbs)
2726 * ``profile_id``: ID for the meter profile.
2727 * ``cir``: Committed information rate (bytes/second).
2728 * ``pir``: Peak information rate (bytes/second).
2729 * ``cbs``: Committed burst size (bytes).
2730 * ``pbs``: Peak burst size (bytes).
2732 add port meter profile (trTCM rfc4115)
2733 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2735 Add meter profile (trTCM rfc4115) to the ethernet device::
2737 testpmd> add port meter profile trtcm_rfc4115 (port_id) (profile_id) \
2738 (cir) (eir) (cbs) (ebs)
2742 * ``profile_id``: ID for the meter profile.
2743 * ``cir``: Committed information rate (bytes/second).
2744 * ``eir``: Excess information rate (bytes/second).
2745 * ``cbs``: Committed burst size (bytes).
2746 * ``ebs``: Excess burst size (bytes).
2748 delete port meter profile
2749 ~~~~~~~~~~~~~~~~~~~~~~~~~
2751 Delete meter profile from the ethernet device::
2753 testpmd> del port meter profile (port_id) (profile_id)
2758 Create new meter object for the ethernet device::
2760 testpmd> create port meter (port_id) (mtr_id) (profile_id) \
2761 (meter_enable) (g_action) (y_action) (r_action) (stats_mask) (shared) \
2762 (use_pre_meter_color) [(dscp_tbl_entry0) (dscp_tbl_entry1)...\
2767 * ``mtr_id``: meter object ID.
2768 * ``profile_id``: ID for the meter profile.
2769 * ``meter_enable``: When this parameter has a non-zero value, the meter object
2770 gets enabled at the time of creation, otherwise remains disabled.
2771 * ``g_action``: Policer action for the packet with green color.
2772 * ``y_action``: Policer action for the packet with yellow color.
2773 * ``r_action``: Policer action for the packet with red color.
2774 * ``stats_mask``: Mask of statistics counter types to be enabled for the
2776 * ``shared``: When this parameter has a non-zero value, the meter object is
2777 shared by multiple flows. Otherwise, meter object is used by single flow.
2778 * ``use_pre_meter_color``: When this parameter has a non-zero value, the
2779 input color for the current meter object is determined by the latest meter
2780 object in the same flow. Otherwise, the current meter object uses the
2781 *dscp_table* to determine the input color.
2782 * ``dscp_tbl_entryx``: DSCP table entry x providing meter providing input
2783 color, 0 <= x <= 63.
2788 Enable meter for the ethernet device::
2790 testpmd> enable port meter (port_id) (mtr_id)
2795 Disable meter for the ethernet device::
2797 testpmd> disable port meter (port_id) (mtr_id)
2802 Delete meter for the ethernet device::
2804 testpmd> del port meter (port_id) (mtr_id)
2806 Set port meter profile
2807 ~~~~~~~~~~~~~~~~~~~~~~
2809 Set meter profile for the ethernet device::
2811 testpmd> set port meter profile (port_id) (mtr_id) (profile_id)
2813 set port meter dscp table
2814 ~~~~~~~~~~~~~~~~~~~~~~~~~
2816 Set meter dscp table for the ethernet device::
2818 testpmd> set port meter dscp table (port_id) (mtr_id) [(dscp_tbl_entry0) \
2819 (dscp_tbl_entry1)...(dscp_tbl_entry63)]
2821 set port meter policer action
2822 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2824 Set meter policer action for the ethernet device::
2826 testpmd> set port meter policer action (port_id) (mtr_id) (action_mask) \
2827 (action0) [(action1) (action1)]
2831 * ``action_mask``: Bit mask indicating which policer actions need to be
2832 updated. One or more policer actions can be updated in a single function
2833 invocation. To update the policer action associated with color C, bit
2834 (1 << C) needs to be set in *action_mask* and element at position C
2835 in the *actions* array needs to be valid.
2836 * ``actionx``: Policer action for the color x,
2837 RTE_MTR_GREEN <= x < RTE_MTR_COLORS
2839 set port meter stats mask
2840 ~~~~~~~~~~~~~~~~~~~~~~~~~
2842 Set meter stats mask for the ethernet device::
2844 testpmd> set port meter stats mask (port_id) (mtr_id) (stats_mask)
2848 * ``stats_mask``: Bit mask indicating statistics counter types to be enabled.
2850 show port meter stats
2851 ~~~~~~~~~~~~~~~~~~~~~
2853 Show meter stats of the ethernet device::
2855 testpmd> show port meter stats (port_id) (mtr_id) (clear)
2859 * ``clear``: Flag that indicates whether the statistics counters should
2860 be cleared (i.e. set to zero) immediately after they have been read or not.
2865 The following section shows functions for configuring traffic management on
2866 the ethernet device through the use of generic TM API.
2868 show port traffic management capability
2869 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2871 Show traffic management capability of the port::
2873 testpmd> show port tm cap (port_id)
2875 show port traffic management capability (hierarchy level)
2876 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2878 Show traffic management hierarchy level capability of the port::
2880 testpmd> show port tm level cap (port_id) (level_id)
2882 show port traffic management capability (hierarchy node level)
2883 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2885 Show the traffic management hierarchy node capability of the port::
2887 testpmd> show port tm node cap (port_id) (node_id)
2889 show port traffic management hierarchy node type
2890 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2892 Show the port traffic management hierarchy node type::
2894 testpmd> show port tm node type (port_id) (node_id)
2896 show port traffic management hierarchy node stats
2897 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2899 Show the port traffic management hierarchy node statistics::
2901 testpmd> show port tm node stats (port_id) (node_id) (clear)
2905 * ``clear``: When this parameter has a non-zero value, the statistics counters
2906 are cleared (i.e. set to zero) immediately after they have been read,
2907 otherwise the statistics counters are left untouched.
2909 Add port traffic management private shaper profile
2910 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2912 Add the port traffic management private shaper profile::
2914 testpmd> add port tm node shaper profile (port_id) (shaper_profile_id) \
2915 (cmit_tb_rate) (cmit_tb_size) (peak_tb_rate) (peak_tb_size) \
2916 (packet_length_adjust) (packet_mode)
2920 * ``shaper_profile id``: Shaper profile ID for the new profile.
2921 * ``cmit_tb_rate``: Committed token bucket rate (bytes per second or packets per second).
2922 * ``cmit_tb_size``: Committed token bucket size (bytes or packets).
2923 * ``peak_tb_rate``: Peak token bucket rate (bytes per second or packets per second).
2924 * ``peak_tb_size``: Peak token bucket size (bytes or packets).
2925 * ``packet_length_adjust``: The value (bytes) to be added to the length of
2926 each packet for the purpose of shaping. This parameter value can be used to
2927 correct the packet length with the framing overhead bytes that are consumed
2929 * ``packet_mode``: Shaper configured in packet mode. This parameter value if
2930 zero, configures shaper in byte mode and if non-zero configures it in packet
2933 Delete port traffic management private shaper profile
2934 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2936 Delete the port traffic management private shaper::
2938 testpmd> del port tm node shaper profile (port_id) (shaper_profile_id)
2942 * ``shaper_profile id``: Shaper profile ID that needs to be deleted.
2944 Add port traffic management shared shaper
2945 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2947 Create the port traffic management shared shaper::
2949 testpmd> add port tm node shared shaper (port_id) (shared_shaper_id) \
2954 * ``shared_shaper_id``: Shared shaper ID to be created.
2955 * ``shaper_profile id``: Shaper profile ID for shared shaper.
2957 Set port traffic management shared shaper
2958 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2960 Update the port traffic management shared shaper::
2962 testpmd> set port tm node shared shaper (port_id) (shared_shaper_id) \
2967 * ``shared_shaper_id``: Shared shaper ID to be update.
2968 * ``shaper_profile id``: Shaper profile ID for shared shaper.
2970 Delete port traffic management shared shaper
2971 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2973 Delete the port traffic management shared shaper::
2975 testpmd> del port tm node shared shaper (port_id) (shared_shaper_id)
2979 * ``shared_shaper_id``: Shared shaper ID to be deleted.
2981 Set port traffic management hierarchy node private shaper
2982 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2984 set the port traffic management hierarchy node private shaper::
2986 testpmd> set port tm node shaper profile (port_id) (node_id) \
2991 * ``shaper_profile id``: Private shaper profile ID to be enabled on the
2994 Add port traffic management WRED profile
2995 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2997 Create a new WRED profile::
2999 testpmd> add port tm node wred profile (port_id) (wred_profile_id) \
3000 (color_g) (min_th_g) (max_th_g) (maxp_inv_g) (wq_log2_g) \
3001 (color_y) (min_th_y) (max_th_y) (maxp_inv_y) (wq_log2_y) \
3002 (color_r) (min_th_r) (max_th_r) (maxp_inv_r) (wq_log2_r)
3006 * ``wred_profile id``: Identifier for the newly create WRED profile
3007 * ``color_g``: Packet color (green)
3008 * ``min_th_g``: Minimum queue threshold for packet with green color
3009 * ``max_th_g``: Minimum queue threshold for packet with green color
3010 * ``maxp_inv_g``: Inverse of packet marking probability maximum value (maxp)
3011 * ``wq_log2_g``: Negated log2 of queue weight (wq)
3012 * ``color_y``: Packet color (yellow)
3013 * ``min_th_y``: Minimum queue threshold for packet with yellow color
3014 * ``max_th_y``: Minimum queue threshold for packet with yellow color
3015 * ``maxp_inv_y``: Inverse of packet marking probability maximum value (maxp)
3016 * ``wq_log2_y``: Negated log2 of queue weight (wq)
3017 * ``color_r``: Packet color (red)
3018 * ``min_th_r``: Minimum queue threshold for packet with yellow color
3019 * ``max_th_r``: Minimum queue threshold for packet with yellow color
3020 * ``maxp_inv_r``: Inverse of packet marking probability maximum value (maxp)
3021 * ``wq_log2_r``: Negated log2 of queue weight (wq)
3023 Delete port traffic management WRED profile
3024 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3026 Delete the WRED profile::
3028 testpmd> del port tm node wred profile (port_id) (wred_profile_id)
3030 Add port traffic management hierarchy nonleaf node
3031 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3033 Add nonleaf node to port traffic management hierarchy::
3035 testpmd> add port tm nonleaf node (port_id) (node_id) (parent_node_id) \
3036 (priority) (weight) (level_id) (shaper_profile_id) \
3037 (n_sp_priorities) (stats_mask) (n_shared_shapers) \
3038 [(shared_shaper_0) (shared_shaper_1) ...] \
3042 * ``parent_node_id``: Node ID of the parent.
3043 * ``priority``: Node priority (highest node priority is zero). This is used by
3044 the SP algorithm running on the parent node for scheduling this node.
3045 * ``weight``: Node weight (lowest weight is one). The node weight is relative
3046 to the weight sum of all siblings that have the same priority. It is used by
3047 the WFQ algorithm running on the parent node for scheduling this node.
3048 * ``level_id``: Hierarchy level of the node.
3049 * ``shaper_profile_id``: Shaper profile ID of the private shaper to be used by
3051 * ``n_sp_priorities``: Number of strict priorities.
3052 * ``stats_mask``: Mask of statistics counter types to be enabled for this node.
3053 * ``n_shared_shapers``: Number of shared shapers.
3054 * ``shared_shaper_id``: Shared shaper id.
3056 Add port traffic management hierarchy nonleaf node with packet mode
3057 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3059 Add nonleaf node with packet mode to port traffic management hierarchy::
3061 testpmd> add port tm nonleaf node pktmode (port_id) (node_id) (parent_node_id) \
3062 (priority) (weight) (level_id) (shaper_profile_id) \
3063 (n_sp_priorities) (stats_mask) (n_shared_shapers) \
3064 [(shared_shaper_0) (shared_shaper_1) ...] \
3068 * ``parent_node_id``: Node ID of the parent.
3069 * ``priority``: Node priority (highest node priority is zero). This is used by
3070 the SP algorithm running on the parent node for scheduling this node.
3071 * ``weight``: Node weight (lowest weight is one). The node weight is relative
3072 to the weight sum of all siblings that have the same priority. It is used by
3073 the WFQ algorithm running on the parent node for scheduling this node.
3074 * ``level_id``: Hierarchy level of the node.
3075 * ``shaper_profile_id``: Shaper profile ID of the private shaper to be used by
3077 * ``n_sp_priorities``: Number of strict priorities. Packet mode is enabled on
3079 * ``stats_mask``: Mask of statistics counter types to be enabled for this node.
3080 * ``n_shared_shapers``: Number of shared shapers.
3081 * ``shared_shaper_id``: Shared shaper id.
3083 Add port traffic management hierarchy leaf node
3084 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3086 Add leaf node to port traffic management hierarchy::
3088 testpmd> add port tm leaf node (port_id) (node_id) (parent_node_id) \
3089 (priority) (weight) (level_id) (shaper_profile_id) \
3090 (cman_mode) (wred_profile_id) (stats_mask) (n_shared_shapers) \
3091 [(shared_shaper_id) (shared_shaper_id) ...] \
3095 * ``parent_node_id``: Node ID of the parent.
3096 * ``priority``: Node priority (highest node priority is zero). This is used by
3097 the SP algorithm running on the parent node for scheduling this node.
3098 * ``weight``: Node weight (lowest weight is one). The node weight is relative
3099 to the weight sum of all siblings that have the same priority. It is used by
3100 the WFQ algorithm running on the parent node for scheduling this node.
3101 * ``level_id``: Hierarchy level of the node.
3102 * ``shaper_profile_id``: Shaper profile ID of the private shaper to be used by
3104 * ``cman_mode``: Congestion management mode to be enabled for this node.
3105 * ``wred_profile_id``: WRED profile id to be enabled for this node.
3106 * ``stats_mask``: Mask of statistics counter types to be enabled for this node.
3107 * ``n_shared_shapers``: Number of shared shapers.
3108 * ``shared_shaper_id``: Shared shaper id.
3110 Delete port traffic management hierarchy node
3111 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3113 Delete node from port traffic management hierarchy::
3115 testpmd> del port tm node (port_id) (node_id)
3117 Update port traffic management hierarchy parent node
3118 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3120 Update port traffic management hierarchy parent node::
3122 testpmd> set port tm node parent (port_id) (node_id) (parent_node_id) \
3125 This function can only be called after the hierarchy commit invocation. Its
3126 success depends on the port support for this operation, as advertised through
3127 the port capability set. This function is valid for all nodes of the traffic
3128 management hierarchy except root node.
3130 Suspend port traffic management hierarchy node
3131 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3133 testpmd> suspend port tm node (port_id) (node_id)
3135 Resume port traffic management hierarchy node
3136 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3138 testpmd> resume port tm node (port_id) (node_id)
3140 Commit port traffic management hierarchy
3141 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3143 Commit the traffic management hierarchy on the port::
3145 testpmd> port tm hierarchy commit (port_id) (clean_on_fail)
3149 * ``clean_on_fail``: When set to non-zero, hierarchy is cleared on function
3150 call failure. On the other hand, hierarchy is preserved when this parameter
3153 Set port traffic management mark VLAN dei
3154 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3156 Enables/Disables the traffic management marking on the port for VLAN packets::
3158 testpmd> set port tm mark vlan_dei <port_id> <green> <yellow> <red>
3162 * ``port_id``: The port which on which VLAN packets marked as ``green`` or
3163 ``yellow`` or ``red`` will have dei bit enabled
3165 * ``green`` enable 1, disable 0 marking for dei bit of VLAN packets marked as green
3167 * ``yellow`` enable 1, disable 0 marking for dei bit of VLAN packets marked as yellow
3169 * ``red`` enable 1, disable 0 marking for dei bit of VLAN packets marked as red
3171 Set port traffic management mark IP dscp
3172 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3174 Enables/Disables the traffic management marking on the port for IP dscp packets::
3176 testpmd> set port tm mark ip_dscp <port_id> <green> <yellow> <red>
3180 * ``port_id``: The port which on which IP packets marked as ``green`` or
3181 ``yellow`` or ``red`` will have IP dscp bits updated
3183 * ``green`` enable 1, disable 0 marking IP dscp to low drop precedence for green packets
3185 * ``yellow`` enable 1, disable 0 marking IP dscp to medium drop precedence for yellow packets
3187 * ``red`` enable 1, disable 0 marking IP dscp to high drop precedence for red packets
3189 Set port traffic management mark IP ecn
3190 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3192 Enables/Disables the traffic management marking on the port for IP ecn packets::
3194 testpmd> set port tm mark ip_ecn <port_id> <green> <yellow> <red>
3198 * ``port_id``: The port which on which IP packets marked as ``green`` or
3199 ``yellow`` or ``red`` will have IP ecn bits updated
3201 * ``green`` enable 1, disable 0 marking IP ecn for green marked packets with ecn of 2'b01 or 2'b10
3202 to ecn of 2'b11 when IP is caring TCP or SCTP
3204 * ``yellow`` enable 1, disable 0 marking IP ecn for yellow marked packets with ecn of 2'b01 or 2'b10
3205 to ecn of 2'b11 when IP is caring TCP or SCTP
3207 * ``red`` enable 1, disable 0 marking IP ecn for yellow marked packets with ecn of 2'b01 or 2'b10
3208 to ecn of 2'b11 when IP is caring TCP or SCTP
3213 This section details the available filter functions that are available.
3215 Note these functions interface the deprecated legacy filtering framework,
3216 superseded by *rte_flow*. See `Flow rules management`_.
3218 .. _testpmd_flow_director:
3220 flow_director_filter
3221 ~~~~~~~~~~~~~~~~~~~~
3223 The Flow Director works in receive mode to identify specific flows or sets of flows and route them to specific queues.
3225 Four types of filtering are supported which are referred to as Perfect Match, Signature, Perfect-mac-vlan and
3226 Perfect-tunnel filters, the match mode is set by the ``--pkt-filter-mode`` command-line parameter:
3228 * Perfect match filters.
3229 The hardware checks a match between the masked fields of the received packets and the programmed filters.
3230 The masked fields are for IP flow.
3232 * Signature filters.
3233 The hardware checks a match between a hash-based signature of the masked fields of the received packet.
3235 * Perfect-mac-vlan 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 MAC VLAN flow.
3239 * Perfect-tunnel match filters.
3240 The hardware checks a match between the masked fields of the received packets and the programmed filters.
3241 The masked fields are for tunnel flow.
3243 * Perfect-raw-flow-type match filters.
3244 The hardware checks a match between the masked fields of the received packets and pre-loaded raw (template) packet.
3245 The masked fields are specified by input sets.
3247 The Flow Director filters can match the different fields for different type of packet: flow type, specific input set
3248 per flow type and the flexible payload.
3250 The Flow Director can also mask out parts of all of these fields so that filters
3251 are only applied to certain fields or parts of the fields.
3253 Note that for raw flow type mode the source and destination fields in the
3254 raw packet buffer need to be presented in a reversed order with respect
3255 to the expected received packets.
3256 For example: IP source and destination addresses or TCP/UDP/SCTP
3257 source and destination ports
3259 Different NICs may have different capabilities, command show port fdir (port_id) can be used to acquire the information.
3261 # Commands to add flow director filters of different flow types::
3263 flow_director_filter (port_id) mode IP (add|del|update) \
3264 flow (ipv4-other|ipv4-frag|ipv6-other|ipv6-frag) \
3265 src (src_ip_address) dst (dst_ip_address) \
3266 tos (tos_value) proto (proto_value) ttl (ttl_value) \
3267 vlan (vlan_value) flexbytes (flexbytes_value) \
3268 (drop|fwd) pf|vf(vf_id) queue (queue_id) \
3271 flow_director_filter (port_id) mode IP (add|del|update) \
3272 flow (ipv4-tcp|ipv4-udp|ipv6-tcp|ipv6-udp) \
3273 src (src_ip_address) (src_port) \
3274 dst (dst_ip_address) (dst_port) \
3275 tos (tos_value) ttl (ttl_value) \
3276 vlan (vlan_value) flexbytes (flexbytes_value) \
3277 (drop|fwd) queue pf|vf(vf_id) (queue_id) \
3280 flow_director_filter (port_id) mode IP (add|del|update) \
3281 flow (ipv4-sctp|ipv6-sctp) \
3282 src (src_ip_address) (src_port) \
3283 dst (dst_ip_address) (dst_port) \
3284 tos (tos_value) ttl (ttl_value) \
3285 tag (verification_tag) vlan (vlan_value) \
3286 flexbytes (flexbytes_value) (drop|fwd) \
3287 pf|vf(vf_id) queue (queue_id) fd_id (fd_id_value)
3289 flow_director_filter (port_id) mode IP (add|del|update) flow l2_payload \
3290 ether (ethertype) flexbytes (flexbytes_value) \
3291 (drop|fwd) pf|vf(vf_id) queue (queue_id)
3294 flow_director_filter (port_id) mode MAC-VLAN (add|del|update) \
3295 mac (mac_address) vlan (vlan_value) \
3296 flexbytes (flexbytes_value) (drop|fwd) \
3297 queue (queue_id) fd_id (fd_id_value)
3299 flow_director_filter (port_id) mode Tunnel (add|del|update) \
3300 mac (mac_address) vlan (vlan_value) \
3301 tunnel (NVGRE|VxLAN) tunnel-id (tunnel_id_value) \
3302 flexbytes (flexbytes_value) (drop|fwd) \
3303 queue (queue_id) fd_id (fd_id_value)
3305 flow_director_filter (port_id) mode raw (add|del|update) flow (flow_id) \
3306 (drop|fwd) queue (queue_id) fd_id (fd_id_value) \
3307 packet (packet file name)
3309 For example, to add an ipv4-udp flow type filter::
3311 testpmd> flow_director_filter 0 mode IP add flow ipv4-udp src 2.2.2.3 32 \
3312 dst 2.2.2.5 33 tos 2 ttl 40 vlan 0x1 flexbytes (0x88,0x48) \
3313 fwd pf queue 1 fd_id 1
3315 For example, add an ipv4-other flow type filter::
3317 testpmd> flow_director_filter 0 mode IP add flow ipv4-other src 2.2.2.3 \
3318 dst 2.2.2.5 tos 2 proto 20 ttl 40 vlan 0x1 \
3319 flexbytes (0x88,0x48) fwd pf queue 1 fd_id 1
3324 Flush all flow director filters on a device::
3326 testpmd> flush_flow_director (port_id)
3328 Example, to flush all flow director filter on port 0::
3330 testpmd> flush_flow_director 0
3335 Set flow director's input masks::
3337 flow_director_mask (port_id) mode IP vlan (vlan_value) \
3338 src_mask (ipv4_src) (ipv6_src) (src_port) \
3339 dst_mask (ipv4_dst) (ipv6_dst) (dst_port)
3341 flow_director_mask (port_id) mode MAC-VLAN vlan (vlan_value)
3343 flow_director_mask (port_id) mode Tunnel vlan (vlan_value) \
3344 mac (mac_value) tunnel-type (tunnel_type_value) \
3345 tunnel-id (tunnel_id_value)
3347 Example, to set flow director mask on port 0::
3349 testpmd> flow_director_mask 0 mode IP vlan 0xefff \
3350 src_mask 255.255.255.255 \
3351 FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF \
3352 dst_mask 255.255.255.255 \
3353 FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF
3355 flow_director_flex_mask
3356 ~~~~~~~~~~~~~~~~~~~~~~~
3358 set masks of flow director's flexible payload based on certain flow type::
3360 testpmd> flow_director_flex_mask (port_id) \
3361 flow (none|ipv4-other|ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
3362 ipv6-other|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp| \
3363 l2_payload|all) (mask)
3365 Example, to set flow director's flex mask for all flow type on port 0::
3367 testpmd> flow_director_flex_mask 0 flow all \
3368 (0xff,0xff,0,0,0,0,0,0,0,0,0,0,0,0,0,0)
3371 flow_director_flex_payload
3372 ~~~~~~~~~~~~~~~~~~~~~~~~~~
3374 Configure flexible payload selection::
3376 flow_director_flex_payload (port_id) (raw|l2|l3|l4) (config)
3378 For example, to select the first 16 bytes from the offset 4 (bytes) of packet's payload as flexible payload::
3380 testpmd> flow_director_flex_payload 0 l4 \
3381 (4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19)
3386 The Flow Director filters can match the different fields for different type of packet, i.e. specific input set
3387 on per flow type and the flexible payload. This command can be used to change input set for each flow type.
3389 Set the input set for flow director::
3391 set_fdir_input_set (port_id) (ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
3392 ipv4-other|ipv6|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other| \
3393 l2_payload|<flow_id>) (ivlan|ethertype|src-ipv4|dst-ipv4|src-ipv6|dst-ipv6| \
3394 ipv4-tos|ipv4-proto|ipv4-ttl|ipv6-tc|ipv6-next-header|ipv6-hop-limits| \
3395 tudp-src-port|udp-dst-port|cp-src-port|tcp-dst-port|sctp-src-port| \
3396 sctp-dst-port|sctp-veri-tag|none) (select|add)
3398 For example to add source IP to FD input set for flow type of ipv4-udp on port 0::
3400 testpmd> set_fdir_input_set 0 ipv4-udp src-ipv4 add
3403 .. _testpmd_rte_flow:
3405 Flow rules management
3406 ---------------------
3408 Control of the generic flow API (*rte_flow*) is fully exposed through the
3409 ``flow`` command (validation, creation, destruction, queries and operation
3412 Considering *rte_flow* overlaps with all `Filter Functions`_, using both
3413 features simultaneously may cause undefined side-effects and is therefore
3419 Because the ``flow`` command uses dynamic tokens to handle the large number
3420 of possible flow rules combinations, its behavior differs slightly from
3421 other commands, in particular:
3423 - Pressing *?* or the *<tab>* key displays contextual help for the current
3424 token, not that of the entire command.
3426 - Optional and repeated parameters are supported (provided they are listed
3427 in the contextual help).
3429 The first parameter stands for the operation mode. Possible operations and
3430 their general syntax are described below. They are covered in detail in the
3433 - Check whether a flow rule can be created::
3435 flow validate {port_id}
3436 [group {group_id}] [priority {level}] [ingress] [egress] [transfer]
3437 pattern {item} [/ {item} [...]] / end
3438 actions {action} [/ {action} [...]] / end
3440 - Create a flow rule::
3442 flow create {port_id}
3443 [group {group_id}] [priority {level}] [ingress] [egress] [transfer]
3444 pattern {item} [/ {item} [...]] / end
3445 actions {action} [/ {action} [...]] / end
3447 - Destroy specific flow rules::
3449 flow destroy {port_id} rule {rule_id} [...]
3451 - Destroy all flow rules::
3453 flow flush {port_id}
3455 - Query an existing flow rule::
3457 flow query {port_id} {rule_id} {action}
3459 - List existing flow rules sorted by priority, filtered by group
3462 flow list {port_id} [group {group_id}] [...]
3464 - Restrict ingress traffic to the defined flow rules::
3466 flow isolate {port_id} {boolean}
3468 - Dump internal representation information of all flows in hardware::
3470 flow dump {port_id} {output_file}
3472 - List and destroy aged flow rules::
3474 flow aged {port_id} [destroy]
3476 - Tunnel offload - create a tunnel stub::
3478 flow tunnel create {port_id} type {tunnel_type}
3480 - Tunnel offload - destroy a tunnel stub::
3482 flow tunnel destroy {port_id} id {tunnel_id}
3484 - Tunnel offload - list port tunnel stubs::
3486 flow tunnel list {port_id}
3488 Creating a tunnel stub for offload
3489 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3491 ``flow tunnel create`` setup a tunnel stub for tunnel offload flow rules::
3493 flow tunnel create {port_id} type {tunnel_type}
3495 If successful, it will return a tunnel stub ID usable with other commands::
3497 port [...]: flow tunnel #[...] type [...]
3499 Tunnel stub ID is relative to a port.
3501 Destroying tunnel offload stub
3502 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3504 ``flow tunnel destroy`` destroy port tunnel stub::
3506 flow tunnel destroy {port_id} id {tunnel_id}
3508 Listing tunnel offload stubs
3509 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3511 ``flow tunnel list`` list port tunnel offload stubs::
3513 flow tunnel list {port_id}
3515 Validating flow rules
3516 ~~~~~~~~~~~~~~~~~~~~~
3518 ``flow validate`` reports whether a flow rule would be accepted by the
3519 underlying device in its current state but stops short of creating it. It is
3520 bound to ``rte_flow_validate()``::
3522 flow validate {port_id}
3523 [group {group_id}] [priority {level}] [ingress] [egress] [transfer]
3524 pattern {item} [/ {item} [...]] / end
3525 actions {action} [/ {action} [...]] / end
3527 If successful, it will show::
3531 Otherwise it will show an error message of the form::
3533 Caught error type [...] ([...]): [...]
3535 This command uses the same parameters as ``flow create``, their format is
3536 described in `Creating flow rules`_.
3538 Check whether redirecting any Ethernet packet received on port 0 to RX queue
3539 index 6 is supported::
3541 testpmd> flow validate 0 ingress pattern eth / end
3542 actions queue index 6 / end
3546 Port 0 does not support TCPv6 rules::
3548 testpmd> flow validate 0 ingress pattern eth / ipv6 / tcp / end
3550 Caught error type 9 (specific pattern item): Invalid argument
3556 ``flow create`` validates and creates the specified flow rule. It is bound
3557 to ``rte_flow_create()``::
3559 flow create {port_id}
3560 [group {group_id}] [priority {level}] [ingress] [egress] [transfer]
3561 [tunnel_set {tunnel_id}] [tunnel_match {tunnel_id}]
3562 pattern {item} [/ {item} [...]] / end
3563 actions {action} [/ {action} [...]] / end
3565 If successful, it will return a flow rule ID usable with other commands::
3567 Flow rule #[...] created
3569 Otherwise it will show an error message of the form::
3571 Caught error type [...] ([...]): [...]
3573 Parameters describe in the following order:
3575 - Attributes (*group*, *priority*, *ingress*, *egress*, *transfer* tokens).
3576 - Tunnel offload specification (tunnel_set, tunnel_match)
3577 - A matching pattern, starting with the *pattern* token and terminated by an
3579 - Actions, starting with the *actions* token and terminated by an *end*
3582 These translate directly to *rte_flow* objects provided as-is to the
3583 underlying functions.
3585 The shortest valid definition only comprises mandatory tokens::
3587 testpmd> flow create 0 pattern end actions end
3589 Note that PMDs may refuse rules that essentially do nothing such as this
3592 **All unspecified object values are automatically initialized to 0.**
3597 These tokens affect flow rule attributes (``struct rte_flow_attr``) and are
3598 specified before the ``pattern`` token.
3600 - ``group {group id}``: priority group.
3601 - ``priority {level}``: priority level within group.
3602 - ``ingress``: rule applies to ingress traffic.
3603 - ``egress``: rule applies to egress traffic.
3604 - ``transfer``: apply rule directly to endpoints found in pattern.
3606 Each instance of an attribute specified several times overrides the previous
3607 value as shown below (group 4 is used)::
3609 testpmd> flow create 0 group 42 group 24 group 4 [...]
3611 Note that once enabled, ``ingress`` and ``egress`` cannot be disabled.
3613 While not specifying a direction is an error, some rules may allow both
3616 Most rules affect RX therefore contain the ``ingress`` token::
3618 testpmd> flow create 0 ingress pattern [...]
3623 Indicate tunnel offload rule type
3625 - ``tunnel_set {tunnel_id}``: mark rule as tunnel offload decap_set type.
3626 - ``tunnel_match {tunnel_id}``: mark rule as tunel offload match type.
3631 A matching pattern starts after the ``pattern`` token. It is made of pattern
3632 items and is terminated by a mandatory ``end`` item.
3634 Items are named after their type (*RTE_FLOW_ITEM_TYPE_* from ``enum
3635 rte_flow_item_type``).
3637 The ``/`` token is used as a separator between pattern items as shown
3640 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end [...]
3642 Note that protocol items like these must be stacked from lowest to highest
3643 layer to make sense. For instance, the following rule is either invalid or
3644 unlikely to match any packet::
3646 testpmd> flow create 0 ingress pattern eth / udp / ipv4 / end [...]
3648 More information on these restrictions can be found in the *rte_flow*
3651 Several items support additional specification structures, for example
3652 ``ipv4`` allows specifying source and destination addresses as follows::
3654 testpmd> flow create 0 ingress pattern eth / ipv4 src is 10.1.1.1
3655 dst is 10.2.0.0 / end [...]
3657 This rule matches all IPv4 traffic with the specified properties.
3659 In this example, ``src`` and ``dst`` are field names of the underlying
3660 ``struct rte_flow_item_ipv4`` object. All item properties can be specified
3661 in a similar fashion.
3663 The ``is`` token means that the subsequent value must be matched exactly,
3664 and assigns ``spec`` and ``mask`` fields in ``struct rte_flow_item``
3665 accordingly. Possible assignment tokens are:
3667 - ``is``: match value perfectly (with full bit-mask).
3668 - ``spec``: match value according to configured bit-mask.
3669 - ``last``: specify upper bound to establish a range.
3670 - ``mask``: specify bit-mask with relevant bits set to one.
3671 - ``prefix``: generate bit-mask with <prefix-length> most-significant bits set to one.
3673 These yield identical results::
3675 ipv4 src is 10.1.1.1
3679 ipv4 src spec 10.1.1.1 src mask 255.255.255.255
3683 ipv4 src spec 10.1.1.1 src prefix 32
3687 ipv4 src is 10.1.1.1 src last 10.1.1.1 # range with a single value
3691 ipv4 src is 10.1.1.1 src last 0 # 0 disables range
3693 Inclusive ranges can be defined with ``last``::
3695 ipv4 src is 10.1.1.1 src last 10.2.3.4 # 10.1.1.1 to 10.2.3.4
3697 Note that ``mask`` affects both ``spec`` and ``last``::
3699 ipv4 src is 10.1.1.1 src last 10.2.3.4 src mask 255.255.0.0
3700 # matches 10.1.0.0 to 10.2.255.255
3702 Properties can be modified multiple times::
3704 ipv4 src is 10.1.1.1 src is 10.1.2.3 src is 10.2.3.4 # matches 10.2.3.4
3708 ipv4 src is 10.1.1.1 src prefix 24 src prefix 16 # matches 10.1.0.0/16
3713 This section lists supported pattern items and their attributes, if any.
3715 - ``end``: end list of pattern items.
3717 - ``void``: no-op pattern item.
3719 - ``invert``: perform actions when pattern does not match.
3721 - ``any``: match any protocol for the current layer.
3723 - ``num {unsigned}``: number of layers covered.
3725 - ``pf``: match traffic from/to the physical function.
3727 - ``vf``: match traffic from/to a virtual function ID.
3729 - ``id {unsigned}``: VF ID.
3731 - ``phy_port``: match traffic from/to a specific physical port.
3733 - ``index {unsigned}``: physical port index.
3735 - ``port_id``: match traffic from/to a given DPDK port ID.
3737 - ``id {unsigned}``: DPDK port ID.
3739 - ``mark``: match value set in previously matched flow rule using the mark action.
3741 - ``id {unsigned}``: arbitrary integer value.
3743 - ``raw``: match an arbitrary byte string.
3745 - ``relative {boolean}``: look for pattern after the previous item.
3746 - ``search {boolean}``: search pattern from offset (see also limit).
3747 - ``offset {integer}``: absolute or relative offset for pattern.
3748 - ``limit {unsigned}``: search area limit for start of pattern.
3749 - ``pattern {string}``: byte string to look for.
3751 - ``eth``: match Ethernet header.
3753 - ``dst {MAC-48}``: destination MAC.
3754 - ``src {MAC-48}``: source MAC.
3755 - ``type {unsigned}``: EtherType or TPID.
3757 - ``vlan``: match 802.1Q/ad VLAN tag.
3759 - ``tci {unsigned}``: tag control information.
3760 - ``pcp {unsigned}``: priority code point.
3761 - ``dei {unsigned}``: drop eligible indicator.
3762 - ``vid {unsigned}``: VLAN identifier.
3763 - ``inner_type {unsigned}``: inner EtherType or TPID.
3765 - ``ipv4``: match IPv4 header.
3767 - ``tos {unsigned}``: type of service.
3768 - ``ttl {unsigned}``: time to live.
3769 - ``proto {unsigned}``: next protocol ID.
3770 - ``src {ipv4 address}``: source address.
3771 - ``dst {ipv4 address}``: destination address.
3773 - ``ipv6``: match IPv6 header.
3775 - ``tc {unsigned}``: traffic class.
3776 - ``flow {unsigned}``: flow label.
3777 - ``proto {unsigned}``: protocol (next header).
3778 - ``hop {unsigned}``: hop limit.
3779 - ``src {ipv6 address}``: source address.
3780 - ``dst {ipv6 address}``: destination address.
3782 - ``icmp``: match ICMP header.
3784 - ``type {unsigned}``: ICMP packet type.
3785 - ``code {unsigned}``: ICMP packet code.
3787 - ``udp``: match UDP header.
3789 - ``src {unsigned}``: UDP source port.
3790 - ``dst {unsigned}``: UDP destination port.
3792 - ``tcp``: match TCP header.
3794 - ``src {unsigned}``: TCP source port.
3795 - ``dst {unsigned}``: TCP destination port.
3797 - ``sctp``: match SCTP header.
3799 - ``src {unsigned}``: SCTP source port.
3800 - ``dst {unsigned}``: SCTP destination port.
3801 - ``tag {unsigned}``: validation tag.
3802 - ``cksum {unsigned}``: checksum.
3804 - ``vxlan``: match VXLAN header.
3806 - ``vni {unsigned}``: VXLAN identifier.
3808 - ``e_tag``: match IEEE 802.1BR E-Tag header.
3810 - ``grp_ecid_b {unsigned}``: GRP and E-CID base.
3812 - ``nvgre``: match NVGRE header.
3814 - ``tni {unsigned}``: virtual subnet ID.
3816 - ``mpls``: match MPLS header.
3818 - ``label {unsigned}``: MPLS label.
3820 - ``gre``: match GRE header.
3822 - ``protocol {unsigned}``: protocol type.
3824 - ``gre_key``: match GRE optional key field.
3826 - ``value {unsigned}``: key value.
3828 - ``fuzzy``: fuzzy pattern match, expect faster than default.
3830 - ``thresh {unsigned}``: accuracy threshold.
3832 - ``gtp``, ``gtpc``, ``gtpu``: match GTPv1 header.
3834 - ``teid {unsigned}``: tunnel endpoint identifier.
3836 - ``geneve``: match GENEVE header.
3838 - ``vni {unsigned}``: virtual network identifier.
3839 - ``protocol {unsigned}``: protocol type.
3841 - ``vxlan-gpe``: match VXLAN-GPE header.
3843 - ``vni {unsigned}``: VXLAN-GPE identifier.
3845 - ``arp_eth_ipv4``: match ARP header for Ethernet/IPv4.
3847 - ``sha {MAC-48}``: sender hardware address.
3848 - ``spa {ipv4 address}``: sender IPv4 address.
3849 - ``tha {MAC-48}``: target hardware address.
3850 - ``tpa {ipv4 address}``: target IPv4 address.
3852 - ``ipv6_ext``: match presence of any IPv6 extension header.
3854 - ``next_hdr {unsigned}``: next header.
3856 - ``icmp6``: match any ICMPv6 header.
3858 - ``type {unsigned}``: ICMPv6 type.
3859 - ``code {unsigned}``: ICMPv6 code.
3861 - ``icmp6_nd_ns``: match ICMPv6 neighbor discovery solicitation.
3863 - ``target_addr {ipv6 address}``: target address.
3865 - ``icmp6_nd_na``: match ICMPv6 neighbor discovery advertisement.
3867 - ``target_addr {ipv6 address}``: target address.
3869 - ``icmp6_nd_opt``: match presence of any ICMPv6 neighbor discovery option.
3871 - ``type {unsigned}``: ND option type.
3873 - ``icmp6_nd_opt_sla_eth``: match ICMPv6 neighbor discovery source Ethernet
3874 link-layer address option.
3876 - ``sla {MAC-48}``: source Ethernet LLA.
3878 - ``icmp6_nd_opt_tla_eth``: match ICMPv6 neighbor discovery target Ethernet
3879 link-layer address option.
3881 - ``tla {MAC-48}``: target Ethernet LLA.
3883 - ``meta``: match application specific metadata.
3885 - ``data {unsigned}``: metadata value.
3887 - ``gtp_psc``: match GTP PDU extension header with type 0x85.
3889 - ``pdu_type {unsigned}``: PDU type.
3890 - ``qfi {unsigned}``: QoS flow identifier.
3892 - ``pppoes``, ``pppoed``: match PPPoE header.
3894 - ``session_id {unsigned}``: session identifier.
3896 - ``pppoe_proto_id``: match PPPoE session protocol identifier.
3898 - ``proto_id {unsigned}``: PPP protocol identifier.
3900 - ``l2tpv3oip``: match L2TPv3 over IP header.
3902 - ``session_id {unsigned}``: L2TPv3 over IP session identifier.
3904 - ``ah``: match AH header.
3906 - ``spi {unsigned}``: security parameters index.
3908 - ``pfcp``: match PFCP header.
3910 - ``s_field {unsigned}``: S field.
3911 - ``seid {unsigned}``: session endpoint identifier.
3916 A list of actions starts after the ``actions`` token in the same fashion as
3917 `Matching pattern`_; actions are separated by ``/`` tokens and the list is
3918 terminated by a mandatory ``end`` action.
3920 Actions are named after their type (*RTE_FLOW_ACTION_TYPE_* from ``enum
3921 rte_flow_action_type``).
3923 Dropping all incoming UDPv4 packets can be expressed as follows::
3925 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
3928 Several actions have configurable properties which must be specified when
3929 there is no valid default value. For example, ``queue`` requires a target
3932 This rule redirects incoming UDPv4 traffic to queue index 6::
3934 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
3935 actions queue index 6 / end
3937 While this one could be rejected by PMDs (unspecified queue index)::
3939 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
3942 As defined by *rte_flow*, the list is not ordered, all actions of a given
3943 rule are performed simultaneously. These are equivalent::
3945 queue index 6 / void / mark id 42 / end
3949 void / mark id 42 / queue index 6 / end
3951 All actions in a list should have different types, otherwise only the last
3952 action of a given type is taken into account::
3954 queue index 4 / queue index 5 / queue index 6 / end # will use queue 6
3958 drop / drop / drop / end # drop is performed only once
3962 mark id 42 / queue index 3 / mark id 24 / end # mark will be 24
3964 Considering they are performed simultaneously, opposite and overlapping
3965 actions can sometimes be combined when the end result is unambiguous::
3967 drop / queue index 6 / end # drop has no effect
3971 queue index 6 / rss queues 6 7 8 / end # queue has no effect
3975 drop / passthru / end # drop has no effect
3977 Note that PMDs may still refuse such combinations.
3982 This section lists supported actions and their attributes, if any.
3984 - ``end``: end list of actions.
3986 - ``void``: no-op action.
3988 - ``passthru``: let subsequent rule process matched packets.
3990 - ``jump``: redirect traffic to group on device.
3992 - ``group {unsigned}``: group to redirect to.
3994 - ``mark``: attach 32 bit value to packets.
3996 - ``id {unsigned}``: 32 bit value to return with packets.
3998 - ``flag``: flag packets.
4000 - ``queue``: assign packets to a given queue index.
4002 - ``index {unsigned}``: queue index to use.
4004 - ``drop``: drop packets (note: passthru has priority).
4006 - ``count``: enable counters for this rule.
4008 - ``rss``: spread packets among several queues.
4010 - ``func {hash function}``: RSS hash function to apply, allowed tokens are
4011 ``toeplitz``, ``simple_xor``, ``symmetric_toeplitz`` and ``default``.
4013 - ``level {unsigned}``: encapsulation level for ``types``.
4015 - ``types [{RSS hash type} [...]] end``: specific RSS hash types.
4016 Note that an empty list does not disable RSS but instead requests
4017 unspecified "best-effort" settings.
4019 - ``key {string}``: RSS hash key, overrides ``key_len``.
4021 - ``key_len {unsigned}``: RSS hash key length in bytes, can be used in
4022 conjunction with ``key`` to pad or truncate it.
4024 - ``queues [{unsigned} [...]] end``: queue indices to use.
4026 - ``pf``: direct traffic to physical function.
4028 - ``vf``: direct traffic to a virtual function ID.
4030 - ``original {boolean}``: use original VF ID if possible.
4031 - ``id {unsigned}``: VF ID.
4033 - ``phy_port``: direct packets to physical port index.
4035 - ``original {boolean}``: use original port index if possible.
4036 - ``index {unsigned}``: physical port index.
4038 - ``port_id``: direct matching traffic to a given DPDK port ID.
4040 - ``original {boolean}``: use original DPDK port ID if possible.
4041 - ``id {unsigned}``: DPDK port ID.
4043 - ``of_set_mpls_ttl``: OpenFlow's ``OFPAT_SET_MPLS_TTL``.
4045 - ``mpls_ttl``: MPLS TTL.
4047 - ``of_dec_mpls_ttl``: OpenFlow's ``OFPAT_DEC_MPLS_TTL``.
4049 - ``of_set_nw_ttl``: OpenFlow's ``OFPAT_SET_NW_TTL``.
4051 - ``nw_ttl``: IP TTL.
4053 - ``of_dec_nw_ttl``: OpenFlow's ``OFPAT_DEC_NW_TTL``.
4055 - ``of_copy_ttl_out``: OpenFlow's ``OFPAT_COPY_TTL_OUT``.
4057 - ``of_copy_ttl_in``: OpenFlow's ``OFPAT_COPY_TTL_IN``.
4059 - ``of_pop_vlan``: OpenFlow's ``OFPAT_POP_VLAN``.
4061 - ``of_push_vlan``: OpenFlow's ``OFPAT_PUSH_VLAN``.
4063 - ``ethertype``: Ethertype.
4065 - ``of_set_vlan_vid``: OpenFlow's ``OFPAT_SET_VLAN_VID``.
4067 - ``vlan_vid``: VLAN id.
4069 - ``of_set_vlan_pcp``: OpenFlow's ``OFPAT_SET_VLAN_PCP``.
4071 - ``vlan_pcp``: VLAN priority.
4073 - ``of_pop_mpls``: OpenFlow's ``OFPAT_POP_MPLS``.
4075 - ``ethertype``: Ethertype.
4077 - ``of_push_mpls``: OpenFlow's ``OFPAT_PUSH_MPLS``.
4079 - ``ethertype``: Ethertype.
4081 - ``vxlan_encap``: Performs a VXLAN encapsulation, outer layer configuration
4082 is done through `Config VXLAN Encap outer layers`_.
4084 - ``vxlan_decap``: Performs a decapsulation action by stripping all headers of
4085 the VXLAN tunnel network overlay from the matched flow.
4087 - ``nvgre_encap``: Performs a NVGRE encapsulation, outer layer configuration
4088 is done through `Config NVGRE Encap outer layers`_.
4090 - ``nvgre_decap``: Performs a decapsulation action by stripping all headers of
4091 the NVGRE tunnel network overlay from the matched flow.
4093 - ``l2_encap``: Performs a L2 encapsulation, L2 configuration
4094 is done through `Config L2 Encap`_.
4096 - ``l2_decap``: Performs a L2 decapsulation, L2 configuration
4097 is done through `Config L2 Decap`_.
4099 - ``mplsogre_encap``: Performs a MPLSoGRE encapsulation, outer layer
4100 configuration is done through `Config MPLSoGRE Encap outer layers`_.
4102 - ``mplsogre_decap``: Performs a MPLSoGRE decapsulation, outer layer
4103 configuration is done through `Config MPLSoGRE Decap outer layers`_.
4105 - ``mplsoudp_encap``: Performs a MPLSoUDP encapsulation, outer layer
4106 configuration is done through `Config MPLSoUDP Encap outer layers`_.
4108 - ``mplsoudp_decap``: Performs a MPLSoUDP decapsulation, outer layer
4109 configuration is done through `Config MPLSoUDP Decap outer layers`_.
4111 - ``set_ipv4_src``: Set a new IPv4 source address in the outermost IPv4 header.
4113 - ``ipv4_addr``: New IPv4 source address.
4115 - ``set_ipv4_dst``: Set a new IPv4 destination address in the outermost IPv4
4118 - ``ipv4_addr``: New IPv4 destination address.
4120 - ``set_ipv6_src``: Set a new IPv6 source address in the outermost IPv6 header.
4122 - ``ipv6_addr``: New IPv6 source address.
4124 - ``set_ipv6_dst``: Set a new IPv6 destination address in the outermost IPv6
4127 - ``ipv6_addr``: New IPv6 destination address.
4129 - ``set_tp_src``: Set a new source port number in the outermost TCP/UDP
4132 - ``port``: New TCP/UDP source port number.
4134 - ``set_tp_dst``: Set a new destination port number in the outermost TCP/UDP
4137 - ``port``: New TCP/UDP destination port number.
4139 - ``mac_swap``: Swap the source and destination MAC addresses in the outermost
4142 - ``dec_ttl``: Performs a decrease TTL value action
4144 - ``set_ttl``: Set TTL value with specified value
4145 - ``ttl_value {unsigned}``: The new TTL value to be set
4147 - ``set_mac_src``: set source MAC address
4149 - ``mac_addr {MAC-48}``: new source MAC address
4151 - ``set_mac_dst``: set destination MAC address
4153 - ``mac_addr {MAC-48}``: new destination MAC address
4155 - ``inc_tcp_seq``: Increase sequence number in the outermost TCP header.
4157 - ``value {unsigned}``: Value to increase TCP sequence number by.
4159 - ``dec_tcp_seq``: Decrease sequence number in the outermost TCP header.
4161 - ``value {unsigned}``: Value to decrease TCP sequence number by.
4163 - ``inc_tcp_ack``: Increase acknowledgment number in the outermost TCP header.
4165 - ``value {unsigned}``: Value to increase TCP acknowledgment number by.
4167 - ``dec_tcp_ack``: Decrease acknowledgment number in the outermost TCP header.
4169 - ``value {unsigned}``: Value to decrease TCP acknowledgment number by.
4171 - ``set_ipv4_dscp``: Set IPv4 DSCP value with specified value
4173 - ``dscp_value {unsigned}``: The new DSCP value to be set
4175 - ``set_ipv6_dscp``: Set IPv6 DSCP value with specified value
4177 - ``dscp_value {unsigned}``: The new DSCP value to be set
4179 - ``shared``: Use shared action created via
4180 ``flow shared_action {port_id} create``
4182 - ``shared_action_id {unsigned}``: Shared action ID to use
4184 Destroying flow rules
4185 ~~~~~~~~~~~~~~~~~~~~~
4187 ``flow destroy`` destroys one or more rules from their rule ID (as returned
4188 by ``flow create``), this command calls ``rte_flow_destroy()`` as many
4189 times as necessary::
4191 flow destroy {port_id} rule {rule_id} [...]
4193 If successful, it will show::
4195 Flow rule #[...] destroyed
4197 It does not report anything for rule IDs that do not exist. The usual error
4198 message is shown when a rule cannot be destroyed::
4200 Caught error type [...] ([...]): [...]
4202 ``flow flush`` destroys all rules on a device and does not take extra
4203 arguments. It is bound to ``rte_flow_flush()``::
4205 flow flush {port_id}
4207 Any errors are reported as above.
4209 Creating several rules and destroying them::
4211 testpmd> flow create 0 ingress pattern eth / ipv6 / end
4212 actions queue index 2 / end
4213 Flow rule #0 created
4214 testpmd> flow create 0 ingress pattern eth / ipv4 / end
4215 actions queue index 3 / end
4216 Flow rule #1 created
4217 testpmd> flow destroy 0 rule 0 rule 1
4218 Flow rule #1 destroyed
4219 Flow rule #0 destroyed
4222 The same result can be achieved using ``flow flush``::
4224 testpmd> flow create 0 ingress pattern eth / ipv6 / end
4225 actions queue index 2 / end
4226 Flow rule #0 created
4227 testpmd> flow create 0 ingress pattern eth / ipv4 / end
4228 actions queue index 3 / end
4229 Flow rule #1 created
4230 testpmd> flow flush 0
4233 Non-existent rule IDs are ignored::
4235 testpmd> flow create 0 ingress pattern eth / ipv6 / end
4236 actions queue index 2 / end
4237 Flow rule #0 created
4238 testpmd> flow create 0 ingress pattern eth / ipv4 / end
4239 actions queue index 3 / end
4240 Flow rule #1 created
4241 testpmd> flow destroy 0 rule 42 rule 10 rule 2
4243 testpmd> flow destroy 0 rule 0
4244 Flow rule #0 destroyed
4250 ``flow query`` queries a specific action of a flow rule having that
4251 ability. Such actions collect information that can be reported using this
4252 command. It is bound to ``rte_flow_query()``::
4254 flow query {port_id} {rule_id} {action}
4256 If successful, it will display either the retrieved data for known actions
4257 or the following message::
4259 Cannot display result for action type [...] ([...])
4261 Otherwise, it will complain either that the rule does not exist or that some
4264 Flow rule #[...] not found
4268 Caught error type [...] ([...]): [...]
4270 Currently only the ``count`` action is supported. This action reports the
4271 number of packets that hit the flow rule and the total number of bytes. Its
4272 output has the following format::
4275 hits_set: [...] # whether "hits" contains a valid value
4276 bytes_set: [...] # whether "bytes" contains a valid value
4277 hits: [...] # number of packets
4278 bytes: [...] # number of bytes
4280 Querying counters for TCPv6 packets redirected to queue 6::
4282 testpmd> flow create 0 ingress pattern eth / ipv6 / tcp / end
4283 actions queue index 6 / count / end
4284 Flow rule #4 created
4285 testpmd> flow query 0 4 count
4296 ``flow list`` lists existing flow rules sorted by priority and optionally
4297 filtered by group identifiers::
4299 flow list {port_id} [group {group_id}] [...]
4301 This command only fails with the following message if the device does not
4306 Output consists of a header line followed by a short description of each
4307 flow rule, one per line. There is no output at all when no flow rules are
4308 configured on the device::
4310 ID Group Prio Attr Rule
4311 [...] [...] [...] [...] [...]
4313 ``Attr`` column flags:
4315 - ``i`` for ``ingress``.
4316 - ``e`` for ``egress``.
4318 Creating several flow rules and listing them::
4320 testpmd> flow create 0 ingress pattern eth / ipv4 / end
4321 actions queue index 6 / end
4322 Flow rule #0 created
4323 testpmd> flow create 0 ingress pattern eth / ipv6 / end
4324 actions queue index 2 / end
4325 Flow rule #1 created
4326 testpmd> flow create 0 priority 5 ingress pattern eth / ipv4 / udp / end
4327 actions rss queues 6 7 8 end / end
4328 Flow rule #2 created
4329 testpmd> flow list 0
4330 ID Group Prio Attr Rule
4331 0 0 0 i- ETH IPV4 => QUEUE
4332 1 0 0 i- ETH IPV6 => QUEUE
4333 2 0 5 i- ETH IPV4 UDP => RSS
4336 Rules are sorted by priority (i.e. group ID first, then priority level)::
4338 testpmd> flow list 1
4339 ID Group Prio Attr Rule
4340 0 0 0 i- ETH => COUNT
4341 6 0 500 i- ETH IPV6 TCP => DROP COUNT
4342 5 0 1000 i- ETH IPV6 ICMP => QUEUE
4343 1 24 0 i- ETH IPV4 UDP => QUEUE
4344 4 24 10 i- ETH IPV4 TCP => DROP
4345 3 24 20 i- ETH IPV4 => DROP
4346 2 24 42 i- ETH IPV4 UDP => QUEUE
4347 7 63 0 i- ETH IPV6 UDP VXLAN => MARK QUEUE
4350 Output can be limited to specific groups::
4352 testpmd> flow list 1 group 0 group 63
4353 ID Group Prio Attr Rule
4354 0 0 0 i- ETH => COUNT
4355 6 0 500 i- ETH IPV6 TCP => DROP COUNT
4356 5 0 1000 i- ETH IPV6 ICMP => QUEUE
4357 7 63 0 i- ETH IPV6 UDP VXLAN => MARK QUEUE
4360 Toggling isolated mode
4361 ~~~~~~~~~~~~~~~~~~~~~~
4363 ``flow isolate`` can be used to tell the underlying PMD that ingress traffic
4364 must only be injected from the defined flow rules; that no default traffic
4365 is expected outside those rules and the driver is free to assign more
4366 resources to handle them. It is bound to ``rte_flow_isolate()``::
4368 flow isolate {port_id} {boolean}
4370 If successful, enabling or disabling isolated mode shows either::
4372 Ingress traffic on port [...]
4373 is now restricted to the defined flow rules
4377 Ingress traffic on port [...]
4378 is not restricted anymore to the defined flow rules
4380 Otherwise, in case of error::
4382 Caught error type [...] ([...]): [...]
4384 Mainly due to its side effects, PMDs supporting this mode may not have the
4385 ability to toggle it more than once without reinitializing affected ports
4386 first (e.g. by exiting testpmd).
4388 Enabling isolated mode::
4390 testpmd> flow isolate 0 true
4391 Ingress traffic on port 0 is now restricted to the defined flow rules
4394 Disabling isolated mode::
4396 testpmd> flow isolate 0 false
4397 Ingress traffic on port 0 is not restricted anymore to the defined flow rules
4400 Dumping HW internal information
4401 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4403 ``flow dump`` dumps the hardware's internal representation information of
4404 all flows. It is bound to ``rte_flow_dev_dump()``::
4406 flow dump {port_id} {output_file}
4408 If successful, it will show::
4412 Otherwise, it will complain error occurred::
4414 Caught error type [...] ([...]): [...]
4416 Listing and destroying aged flow rules
4417 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4419 ``flow aged`` simply lists aged flow rules be get from api ``rte_flow_get_aged_flows``,
4420 and ``destroy`` parameter can be used to destroy those flow rules in PMD.
4422 flow aged {port_id} [destroy]
4424 Listing current aged flow rules::
4426 testpmd> flow aged 0
4427 Port 0 total aged flows: 0
4428 testpmd> flow create 0 ingress pattern eth / ipv4 src is 2.2.2.14 / end
4429 actions age timeout 5 / queue index 0 / end
4430 Flow rule #0 created
4431 testpmd> flow create 0 ingress pattern eth / ipv4 src is 2.2.2.15 / end
4432 actions age timeout 4 / queue index 0 / end
4433 Flow rule #1 created
4434 testpmd> flow create 0 ingress pattern eth / ipv4 src is 2.2.2.16 / end
4435 actions age timeout 2 / queue index 0 / end
4436 Flow rule #2 created
4437 testpmd> flow create 0 ingress pattern eth / ipv4 src is 2.2.2.17 / end
4438 actions age timeout 3 / queue index 0 / end
4439 Flow rule #3 created
4442 Aged Rules are simply list as command ``flow list {port_id}``, but strip the detail rule
4443 information, all the aged flows are sorted by the longest timeout time. For example, if
4444 those rules be configured in the same time, ID 2 will be the first aged out rule, the next
4445 will be ID 3, ID 1, ID 0::
4447 testpmd> flow aged 0
4448 Port 0 total aged flows: 4
4455 If attach ``destroy`` parameter, the command will destroy all the list aged flow rules.
4457 testpmd> flow aged 0 destroy
4458 Port 0 total aged flows: 4
4465 Flow rule #2 destroyed
4466 Flow rule #3 destroyed
4467 Flow rule #1 destroyed
4468 Flow rule #0 destroyed
4469 4 flows be destroyed
4470 testpmd> flow aged 0
4471 Port 0 total aged flows: 0
4473 Creating shared actions
4474 ~~~~~~~~~~~~~~~~~~~~~~~
4475 ``flow shared_action {port_id} create`` creates shared action with optional
4476 shared action ID. It is bound to ``rte_flow_shared_action_create()``::
4478 flow shared_action {port_id} create [action_id {shared_action_id}]
4479 [ingress] [egress] action {action} / end
4481 If successful, it will show::
4483 Shared action #[...] created
4485 Otherwise, it will complain either that shared action already exists or that
4486 some error occurred::
4488 Shared action #[...] is already assigned, delete it first
4492 Caught error type [...] ([...]): [...]
4494 Create shared rss action with id 100 to queues 1 and 2 on port 0::
4496 testpmd> flow shared_action 0 create action_id 100 \
4497 ingress action rss queues 1 2 end / end
4499 Create shared rss action with id assigned by testpmd to queues 1 and 2 on
4502 testpmd> flow shared_action 0 create action_id \
4503 ingress action rss queues 0 1 end / end
4505 Updating shared actions
4506 ~~~~~~~~~~~~~~~~~~~~~~~
4507 ``flow shared_action {port_id} update`` updates configuration of the shared
4508 action from its shared action ID (as returned by
4509 ``flow shared_action {port_id} create``). It is bound to
4510 ``rte_flow_shared_action_update()``::
4512 flow shared_action {port_id} update {shared_action_id}
4513 action {action} / end
4515 If successful, it will show::
4517 Shared action #[...] updated
4519 Otherwise, it will complain either that shared action not found or that some
4522 Failed to find shared action #[...] on port [...]
4526 Caught error type [...] ([...]): [...]
4528 Update shared rss action having id 100 on port 0 with rss to queues 0 and 3
4529 (in create example above rss queues were 1 and 2)::
4531 testpmd> flow shared_action 0 update 100 action rss queues 0 3 end / end
4533 Destroying shared actions
4534 ~~~~~~~~~~~~~~~~~~~~~~~~~
4535 ``flow shared_action {port_id} update`` destroys one or more shared actions
4536 from their shared action IDs (as returned by
4537 ``flow shared_action {port_id} create``). It is bound to
4538 ``rte_flow_shared_action_destroy()``::
4540 flow shared_action {port_id} destroy action_id {shared_action_id} [...]
4542 If successful, it will show::
4544 Shared action #[...] destroyed
4546 It does not report anything for shared action IDs that do not exist.
4547 The usual error message is shown when a shared action cannot be destroyed::
4549 Caught error type [...] ([...]): [...]
4551 Destroy shared actions having id 100 & 101::
4553 testpmd> flow shared_action 0 destroy action_id 100 action_id 101
4555 Query shared actions
4556 ~~~~~~~~~~~~~~~~~~~~
4557 ``flow shared_action {port_id} query`` queries the shared action from its
4558 shared action ID (as returned by ``flow shared_action {port_id} create``).
4559 It is bound to ``rte_flow_shared_action_query()``::
4561 flow shared_action {port_id} query {shared_action_id}
4563 Currently only rss shared action supported. If successful, it will show::
4568 Otherwise, it will complain either that shared action not found or that some
4571 Failed to find shared action #[...] on port [...]
4575 Caught error type [...] ([...]): [...]
4577 Query shared action having id 100::
4579 testpmd> flow shared_action 0 query 100
4581 Sample QinQ flow rules
4582 ~~~~~~~~~~~~~~~~~~~~~~
4584 Before creating QinQ rule(s) the following commands should be issued to enable QinQ::
4586 testpmd> port stop 0
4587 testpmd> vlan set qinq_strip on 0
4589 The above command sets the inner and outer TPID's to 0x8100.
4591 To change the TPID's the following commands should be used::
4593 testpmd> vlan set outer tpid 0xa100 0
4594 testpmd> vlan set inner tpid 0x9100 0
4595 testpmd> port start 0
4597 Validate and create a QinQ rule on port 0 to steer traffic to a VF queue in a VM.
4601 testpmd> flow validate 0 ingress pattern eth / vlan tci is 123 /
4602 vlan tci is 456 / end actions vf id 1 / queue index 0 / end
4603 Flow rule #0 validated
4605 testpmd> flow create 0 ingress pattern eth / vlan tci is 4 /
4606 vlan tci is 456 / end actions vf id 123 / queue index 0 / end
4607 Flow rule #0 created
4609 testpmd> flow list 0
4610 ID Group Prio Attr Rule
4611 0 0 0 i- ETH VLAN VLAN=>VF QUEUE
4613 Validate and create a QinQ rule on port 0 to steer traffic to a queue on the host.
4617 testpmd> flow validate 0 ingress pattern eth / vlan tci is 321 /
4618 vlan tci is 654 / end actions pf / queue index 0 / end
4619 Flow rule #1 validated
4621 testpmd> flow create 0 ingress pattern eth / vlan tci is 321 /
4622 vlan tci is 654 / end actions pf / queue index 1 / end
4623 Flow rule #1 created
4625 testpmd> flow list 0
4626 ID Group Prio Attr Rule
4627 0 0 0 i- ETH VLAN VLAN=>VF QUEUE
4628 1 0 0 i- ETH VLAN VLAN=>PF QUEUE
4630 Sample VXLAN flow rules
4631 ~~~~~~~~~~~~~~~~~~~~~~~
4633 Before creating VXLAN rule(s), the UDP port should be added for VXLAN packet
4636 testpmd> rx_vxlan_port add 4789 0
4638 Create VXLAN rules on port 0 to steer traffic to PF queues.
4642 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / vxlan /
4643 eth dst is 00:11:22:33:44:55 / end actions pf / queue index 1 / end
4644 Flow rule #0 created
4646 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / vxlan vni is 3 /
4647 eth dst is 00:11:22:33:44:55 / end actions pf / queue index 2 / end
4648 Flow rule #1 created
4650 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / vxlan /
4651 eth dst is 00:11:22:33:44:55 / vlan tci is 10 / end actions pf /
4653 Flow rule #2 created
4655 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / vxlan vni is 5 /
4656 eth dst is 00:11:22:33:44:55 / vlan tci is 20 / end actions pf /
4658 Flow rule #3 created
4660 testpmd> flow create 0 ingress pattern eth dst is 00:00:00:00:01:00 / ipv4 /
4661 udp / vxlan vni is 6 / eth dst is 00:11:22:33:44:55 / end actions pf /
4663 Flow rule #4 created
4665 testpmd> flow list 0
4666 ID Group Prio Attr Rule
4667 0 0 0 i- ETH IPV4 UDP VXLAN ETH => QUEUE
4668 1 0 0 i- ETH IPV4 UDP VXLAN ETH => QUEUE
4669 2 0 0 i- ETH IPV4 UDP VXLAN ETH VLAN => QUEUE
4670 3 0 0 i- ETH IPV4 UDP VXLAN ETH VLAN => QUEUE
4671 4 0 0 i- ETH IPV4 UDP VXLAN ETH => QUEUE
4673 Sample VXLAN encapsulation rule
4674 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4676 VXLAN encapsulation outer layer has default value pre-configured in testpmd
4677 source code, those can be changed by using the following commands
4679 IPv4 VXLAN outer header::
4681 testpmd> set vxlan ip-version ipv4 vni 4 udp-src 4 udp-dst 4 ip-src 127.0.0.1
4682 ip-dst 128.0.0.1 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4683 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4686 testpmd> set vxlan-with-vlan ip-version ipv4 vni 4 udp-src 4 udp-dst 4 ip-src
4687 127.0.0.1 ip-dst 128.0.0.1 vlan-tci 34 eth-src 11:11:11:11:11:11
4688 eth-dst 22:22:22:22:22:22
4689 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4692 testpmd> set vxlan-tos-ttl ip-version ipv4 vni 4 udp-src 4 udp-dst 4 ip-tos 0
4693 ip-ttl 255 ip-src 127.0.0.1 ip-dst 128.0.0.1 eth-src 11:11:11:11:11:11
4694 eth-dst 22:22:22:22:22:22
4695 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4698 IPv6 VXLAN outer header::
4700 testpmd> set vxlan ip-version ipv6 vni 4 udp-src 4 udp-dst 4 ip-src ::1
4701 ip-dst ::2222 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4702 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4705 testpmd> set vxlan-with-vlan ip-version ipv6 vni 4 udp-src 4 udp-dst 4
4706 ip-src ::1 ip-dst ::2222 vlan-tci 34 eth-src 11:11:11:11:11:11
4707 eth-dst 22:22:22:22:22:22
4708 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4711 testpmd> set vxlan-tos-ttl ip-version ipv6 vni 4 udp-src 4 udp-dst 4
4712 ip-tos 0 ip-ttl 255 ::1 ip-dst ::2222 eth-src 11:11:11:11:11:11
4713 eth-dst 22:22:22:22:22:22
4714 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4717 Sample NVGRE encapsulation rule
4718 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4720 NVGRE encapsulation outer layer has default value pre-configured in testpmd
4721 source code, those can be changed by using the following commands
4723 IPv4 NVGRE outer header::
4725 testpmd> set nvgre ip-version ipv4 tni 4 ip-src 127.0.0.1 ip-dst 128.0.0.1
4726 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4727 testpmd> flow create 0 ingress pattern end actions nvgre_encap /
4730 testpmd> set nvgre-with-vlan ip-version ipv4 tni 4 ip-src 127.0.0.1
4731 ip-dst 128.0.0.1 vlan-tci 34 eth-src 11:11:11:11:11:11
4732 eth-dst 22:22:22:22:22:22
4733 testpmd> flow create 0 ingress pattern end actions nvgre_encap /
4736 IPv6 NVGRE outer header::
4738 testpmd> set nvgre ip-version ipv6 tni 4 ip-src ::1 ip-dst ::2222
4739 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4740 testpmd> flow create 0 ingress pattern end actions nvgre_encap /
4743 testpmd> set nvgre-with-vlan ip-version ipv6 tni 4 ip-src ::1 ip-dst ::2222
4744 vlan-tci 34 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4745 testpmd> flow create 0 ingress pattern end actions nvgre_encap /
4748 Sample L2 encapsulation rule
4749 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4751 L2 encapsulation has default value pre-configured in testpmd
4752 source code, those can be changed by using the following commands
4756 testpmd> set l2_encap ip-version ipv4
4757 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4758 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / mpls / end actions
4759 mplsoudp_decap / l2_encap / end
4761 L2 with VXLAN header::
4763 testpmd> set l2_encap-with-vlan ip-version ipv4 vlan-tci 34
4764 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4765 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / mpls / end actions
4766 mplsoudp_decap / l2_encap / end
4768 Sample L2 decapsulation rule
4769 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4771 L2 decapsulation has default value pre-configured in testpmd
4772 source code, those can be changed by using the following commands
4776 testpmd> set l2_decap
4777 testpmd> flow create 0 egress pattern eth / end actions l2_decap / mplsoudp_encap /
4780 L2 with VXLAN header::
4782 testpmd> set l2_encap-with-vlan
4783 testpmd> flow create 0 egress pattern eth / end actions l2_encap / mplsoudp_encap /
4786 Sample MPLSoGRE encapsulation rule
4787 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4789 MPLSoGRE encapsulation outer layer has default value pre-configured in testpmd
4790 source code, those can be changed by using the following commands
4792 IPv4 MPLSoGRE outer header::
4794 testpmd> set mplsogre_encap ip-version ipv4 label 4
4795 ip-src 127.0.0.1 ip-dst 128.0.0.1 eth-src 11:11:11:11:11:11
4796 eth-dst 22:22:22:22:22:22
4797 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4798 mplsogre_encap / end
4800 IPv4 MPLSoGRE with VLAN outer header::
4802 testpmd> set mplsogre_encap-with-vlan ip-version ipv4 label 4
4803 ip-src 127.0.0.1 ip-dst 128.0.0.1 vlan-tci 34
4804 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4805 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4806 mplsogre_encap / end
4808 IPv6 MPLSoGRE outer header::
4810 testpmd> set mplsogre_encap ip-version ipv6 mask 4
4811 ip-src ::1 ip-dst ::2222 eth-src 11:11:11:11:11:11
4812 eth-dst 22:22:22:22:22:22
4813 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4814 mplsogre_encap / end
4816 IPv6 MPLSoGRE with VLAN outer header::
4818 testpmd> set mplsogre_encap-with-vlan ip-version ipv6 mask 4
4819 ip-src ::1 ip-dst ::2222 vlan-tci 34
4820 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4821 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4822 mplsogre_encap / end
4824 Sample MPLSoGRE decapsulation rule
4825 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4827 MPLSoGRE decapsulation outer layer has default value pre-configured in testpmd
4828 source code, those can be changed by using the following commands
4830 IPv4 MPLSoGRE outer header::
4832 testpmd> set mplsogre_decap ip-version ipv4
4833 testpmd> flow create 0 ingress pattern eth / ipv4 / gre / mpls / end actions
4834 mplsogre_decap / l2_encap / end
4836 IPv4 MPLSoGRE with VLAN outer header::
4838 testpmd> set mplsogre_decap-with-vlan ip-version ipv4
4839 testpmd> flow create 0 ingress pattern eth / vlan / ipv4 / gre / mpls / end
4840 actions mplsogre_decap / l2_encap / end
4842 IPv6 MPLSoGRE outer header::
4844 testpmd> set mplsogre_decap ip-version ipv6
4845 testpmd> flow create 0 ingress pattern eth / ipv6 / gre / mpls / end
4846 actions mplsogre_decap / l2_encap / end
4848 IPv6 MPLSoGRE with VLAN outer header::
4850 testpmd> set mplsogre_decap-with-vlan ip-version ipv6
4851 testpmd> flow create 0 ingress pattern eth / vlan / ipv6 / gre / mpls / end
4852 actions mplsogre_decap / l2_encap / end
4854 Sample MPLSoUDP encapsulation rule
4855 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4857 MPLSoUDP encapsulation outer layer has default value pre-configured in testpmd
4858 source code, those can be changed by using the following commands
4860 IPv4 MPLSoUDP outer header::
4862 testpmd> set mplsoudp_encap ip-version ipv4 label 4 udp-src 5 udp-dst 10
4863 ip-src 127.0.0.1 ip-dst 128.0.0.1 eth-src 11:11:11:11:11:11
4864 eth-dst 22:22:22:22:22:22
4865 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4866 mplsoudp_encap / end
4868 IPv4 MPLSoUDP with VLAN outer header::
4870 testpmd> set mplsoudp_encap-with-vlan ip-version ipv4 label 4 udp-src 5
4871 udp-dst 10 ip-src 127.0.0.1 ip-dst 128.0.0.1 vlan-tci 34
4872 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4873 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4874 mplsoudp_encap / end
4876 IPv6 MPLSoUDP outer header::
4878 testpmd> set mplsoudp_encap ip-version ipv6 mask 4 udp-src 5 udp-dst 10
4879 ip-src ::1 ip-dst ::2222 eth-src 11:11:11:11:11:11
4880 eth-dst 22:22:22:22:22:22
4881 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4882 mplsoudp_encap / end
4884 IPv6 MPLSoUDP with VLAN outer header::
4886 testpmd> set mplsoudp_encap-with-vlan ip-version ipv6 mask 4 udp-src 5
4887 udp-dst 10 ip-src ::1 ip-dst ::2222 vlan-tci 34
4888 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4889 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4890 mplsoudp_encap / end
4892 Sample MPLSoUDP decapsulation rule
4893 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4895 MPLSoUDP decapsulation outer layer has default value pre-configured in testpmd
4896 source code, those can be changed by using the following commands
4898 IPv4 MPLSoUDP outer header::
4900 testpmd> set mplsoudp_decap ip-version ipv4
4901 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / mpls / end actions
4902 mplsoudp_decap / l2_encap / end
4904 IPv4 MPLSoUDP with VLAN outer header::
4906 testpmd> set mplsoudp_decap-with-vlan ip-version ipv4
4907 testpmd> flow create 0 ingress pattern eth / vlan / ipv4 / udp / mpls / end
4908 actions mplsoudp_decap / l2_encap / end
4910 IPv6 MPLSoUDP outer header::
4912 testpmd> set mplsoudp_decap ip-version ipv6
4913 testpmd> flow create 0 ingress pattern eth / ipv6 / udp / mpls / end
4914 actions mplsoudp_decap / l2_encap / end
4916 IPv6 MPLSoUDP with VLAN outer header::
4918 testpmd> set mplsoudp_decap-with-vlan ip-version ipv6
4919 testpmd> flow create 0 ingress pattern eth / vlan / ipv6 / udp / mpls / end
4920 actions mplsoudp_decap / l2_encap / end
4922 Sample Raw encapsulation rule
4923 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4925 Raw encapsulation configuration can be set by the following commands
4927 Eecapsulating VxLAN::
4929 testpmd> set raw_encap 4 eth src is 10:11:22:33:44:55 / vlan tci is 1
4930 inner_type is 0x0800 / ipv4 / udp dst is 4789 / vxlan vni
4932 testpmd> flow create 0 egress pattern eth / ipv4 / end actions
4933 raw_encap index 4 / end
4935 Sample Raw decapsulation rule
4936 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4938 Raw decapsulation configuration can be set by the following commands
4940 Decapsulating VxLAN::
4942 testpmd> set raw_decap eth / ipv4 / udp / vxlan / end_set
4943 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / vxlan / eth / ipv4 /
4944 end actions raw_decap / queue index 0 / end
4949 ESP rules can be created by the following commands::
4951 testpmd> flow create 0 ingress pattern eth / ipv4 / esp spi is 1 / end actions
4953 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / esp spi is 1 / end
4954 actions queue index 3 / end
4955 testpmd> flow create 0 ingress pattern eth / ipv6 / esp spi is 1 / end actions
4957 testpmd> flow create 0 ingress pattern eth / ipv6 / udp / esp spi is 1 / end
4958 actions queue index 3 / end
4963 AH rules can be created by the following commands::
4965 testpmd> flow create 0 ingress pattern eth / ipv4 / ah spi is 1 / end actions
4967 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / ah spi is 1 / end
4968 actions queue index 3 / end
4969 testpmd> flow create 0 ingress pattern eth / ipv6 / ah spi is 1 / end actions
4971 testpmd> flow create 0 ingress pattern eth / ipv6 / udp / ah spi is 1 / end
4972 actions queue index 3 / end
4977 PFCP rules can be created by the following commands(s_field need to be 1
4980 testpmd> flow create 0 ingress pattern eth / ipv4 / pfcp s_field is 0 / end
4981 actions queue index 3 / end
4982 testpmd> flow create 0 ingress pattern eth / ipv4 / pfcp s_field is 1
4983 seid is 1 / end actions queue index 3 / end
4984 testpmd> flow create 0 ingress pattern eth / ipv6 / pfcp s_field is 0 / end
4985 actions queue index 3 / end
4986 testpmd> flow create 0 ingress pattern eth / ipv6 / pfcp s_field is 1
4987 seid is 1 / end actions queue index 3 / end
4992 The following sections show functions to load/unload eBPF based filters.
4997 Load an eBPF program as a callback for particular RX/TX queue::
4999 testpmd> bpf-load rx|tx (portid) (queueid) (load-flags) (bpf-prog-filename)
5001 The available load-flags are:
5003 * ``J``: use JIT generated native code, otherwise BPF interpreter will be used.
5005 * ``M``: assume input parameter is a pointer to rte_mbuf, otherwise assume it is a pointer to first segment's data.
5011 You'll need clang v3.7 or above to build bpf program you'd like to load
5015 .. code-block:: console
5018 clang -O2 -target bpf -c t1.c
5020 Then to load (and JIT compile) t1.o at RX queue 0, port 1:
5022 .. code-block:: console
5024 testpmd> bpf-load rx 1 0 J ./dpdk.org/examples/bpf/t1.o
5026 To load (not JITed) t1.o at TX queue 0, port 0:
5028 .. code-block:: console
5030 testpmd> bpf-load tx 0 0 - ./dpdk.org/examples/bpf/t1.o
5035 Unload previously loaded eBPF program for particular RX/TX queue::
5037 testpmd> bpf-unload rx|tx (portid) (queueid)
5039 For example to unload BPF filter from TX queue 0, port 0:
5041 .. code-block:: console
5043 testpmd> bpf-unload tx 0 0