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 are shown wrapped at `"\\"` for display purposes::
38 testpmd> set flow_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \
39 (pause_time) (send_xon) (port_id)
41 In the real ``testpmd>`` prompt these commands should be on a single line.
46 The testpmd has on-line help for the functions that are available at runtime.
47 These are divided into sections and can be accessed using help, help section or help all:
49 .. code-block:: console
53 help control : Start and stop forwarding.
54 help display : Displaying port, stats and config information.
55 help config : Configuration information.
56 help ports : Configuring ports.
57 help registers : Reading and setting port registers.
58 help filters : Filters configuration help.
59 help all : All of the above sections.
62 Command File Functions
63 ----------------------
65 To facilitate loading large number of commands or to avoid cutting and pasting where not
66 practical or possible testpmd supports alternative methods for executing commands.
68 * If started with the ``--cmdline-file=FILENAME`` command line argument testpmd
69 will execute all CLI commands contained within the file immediately before
70 starting packet forwarding or entering interactive mode.
72 .. code-block:: console
74 ./testpmd -n4 -r2 ... -- -i --cmdline-file=/home/ubuntu/flow-create-commands.txt
75 Interactive-mode selected
76 CLI commands to be read from /home/ubuntu/flow-create-commands.txt
77 Configuring Port 0 (socket 0)
78 Port 0: 7C:FE:90:CB:74:CE
79 Configuring Port 1 (socket 0)
80 Port 1: 7C:FE:90:CB:74:CA
81 Checking link statuses...
82 Port 0 Link Up - speed 10000 Mbps - full-duplex
83 Port 1 Link Up - speed 10000 Mbps - full-duplex
89 Flow rule #498 created
90 Flow rule #499 created
91 Read all CLI commands from /home/ubuntu/flow-create-commands.txt
95 * At run-time additional commands can be loaded in bulk by invoking the ``load FILENAME``
98 .. code-block:: console
100 testpmd> load /home/ubuntu/flow-create-commands.txt
105 Flow rule #498 created
106 Flow rule #499 created
107 Read all CLI commands from /home/ubuntu/flow-create-commands.txt
111 In all cases output from any included command will be displayed as standard output.
112 Execution will continue until the end of the file is reached regardless of
113 whether any errors occur. The end user must examine the output to determine if
114 any failures occurred.
123 Start packet forwarding with current configuration::
130 Start packet forwarding with current configuration after sending specified number of bursts of packets::
132 testpmd> start tx_first (""|burst_num)
134 The default burst number is 1 when ``burst_num`` not presented.
139 Stop packet forwarding, and display accumulated statistics::
154 The functions in the following sections are used to display information about the
155 testpmd configuration or the NIC status.
160 Display information for a given port or all ports::
162 testpmd> show port (info|summary|stats|xstats|fdir|stat_qmap|dcb_tc|cap) (port_id|all)
164 The available information categories are:
166 * ``info``: General port information such as MAC address.
168 * ``summary``: Brief port summary such as Device Name, Driver Name etc.
170 * ``stats``: RX/TX statistics.
172 * ``xstats``: RX/TX extended NIC statistics.
174 * ``fdir``: Flow Director information and statistics.
176 * ``stat_qmap``: Queue statistics mapping.
178 * ``dcb_tc``: DCB information such as TC mapping.
180 * ``cap``: Supported offload capabilities.
184 .. code-block:: console
186 testpmd> show port info 0
188 ********************* Infos for port 0 *********************
190 MAC address: XX:XX:XX:XX:XX:XX
192 memory allocation on the socket: 0
194 Link speed: 40000 Mbps
195 Link duplex: full-duplex
196 Promiscuous mode: enabled
197 Allmulticast mode: disabled
198 Maximum number of MAC addresses: 64
199 Maximum number of MAC addresses of hash filtering: 0
204 Redirection table size: 512
205 Supported flow types:
225 Display the rss redirection table entry indicated by masks on port X::
227 testpmd> show port (port_id) rss reta (size) (mask0, mask1...)
229 size is used to indicate the hardware supported reta size
234 Display the RSS hash functions and RSS hash key of a port::
236 testpmd> show port (port_id) rss-hash [key]
241 Clear the port statistics for a given port or for all ports::
243 testpmd> clear port (info|stats|xstats|fdir|stat_qmap) (port_id|all)
247 testpmd> clear port stats all
252 Display information for a given port's RX/TX queue::
254 testpmd> show (rxq|txq) info (port_id) (queue_id)
259 Displays the configuration of the application.
260 The configuration comes from the command-line, the runtime or the application defaults::
262 testpmd> show config (rxtx|cores|fwd|txpkts)
264 The available information categories are:
266 * ``rxtx``: RX/TX configuration items.
268 * ``cores``: List of forwarding cores.
270 * ``fwd``: Packet forwarding configuration.
272 * ``txpkts``: Packets to TX configuration.
276 .. code-block:: console
278 testpmd> show config rxtx
280 io packet forwarding - CRC stripping disabled - packets/burst=16
281 nb forwarding cores=2 - nb forwarding ports=1
282 RX queues=1 - RX desc=128 - RX free threshold=0
283 RX threshold registers: pthresh=8 hthresh=8 wthresh=4
284 TX queues=1 - TX desc=512 - TX free threshold=0
285 TX threshold registers: pthresh=36 hthresh=0 wthresh=0
286 TX RS bit threshold=0 - TXQ flags=0x0
291 Set the packet forwarding mode::
293 testpmd> set fwd (io|mac|macswap|flowgen| \
294 rxonly|txonly|csum|icmpecho|noisy) (""|retry)
296 ``retry`` can be specified for forwarding engines except ``rx_only``.
298 The available information categories are:
300 * ``io``: Forwards packets "as-is" in I/O mode.
301 This is the fastest possible forwarding operation as it does not access packets data.
302 This is the default mode.
304 * ``mac``: Changes the source and the destination Ethernet addresses of packets before forwarding them.
305 Default application behaviour is to set source Ethernet address to that of the transmitting interface, and destination
306 address to a dummy value (set during init). The user may specify a target destination Ethernet address via the 'eth-peer' or
307 'eth-peer-configfile' command-line options. It is not currently possible to specify a specific source Ethernet address.
309 * ``macswap``: MAC swap forwarding mode.
310 Swaps the source and the destination Ethernet addresses of packets before forwarding them.
312 * ``flowgen``: Multi-flow generation mode.
313 Originates a number of flows (with varying destination IP addresses), and terminate receive traffic.
315 * ``rxonly``: Receives packets but doesn't transmit them.
317 * ``txonly``: Generates and transmits packets without receiving any.
319 * ``csum``: Changes the checksum field with hardware or software methods depending on the offload flags on the packet.
321 * ``icmpecho``: Receives a burst of packets, lookup for IMCP echo requests and, if any, send back ICMP echo replies.
323 * ``ieee1588``: Demonstrate L2 IEEE1588 V2 PTP timestamping for RX and TX. Requires ``CONFIG_RTE_LIBRTE_IEEE1588=y``.
325 * ``softnic``: Demonstrates the softnic forwarding operation. In this mode, packet forwarding is
326 similar to I/O mode except for the fact that packets are loopback to the softnic ports only. Therefore, portmask parameter should be set to softnic port only. The various software based custom NIC pipelines specified through the softnic firmware (DPDK packet framework script) can be tested in this mode. Furthermore, it allows to build 5-level hierarchical QoS scheduler as a default option that can be enabled through CLI once testpmd application is initialised. The user can modify the default scheduler hierarchy or can specify the new QoS Scheduler hierarchy through CLI. Requires ``CONFIG_RTE_LIBRTE_PMD_SOFTNIC=y``.
328 * ``noisy``: Noisy neighbour simulation.
329 Simulate more realistic behavior of a guest machine engaged in receiving
330 and sending packets performing Virtual Network Function (VNF).
334 testpmd> set fwd rxonly
336 Set rxonly packet forwarding mode
342 Display an RX descriptor for a port RX queue::
344 testpmd> read rxd (port_id) (queue_id) (rxd_id)
348 testpmd> read rxd 0 0 4
349 0x0000000B - 0x001D0180 / 0x0000000B - 0x001D0180
354 Display a TX descriptor for a port TX queue::
356 testpmd> read txd (port_id) (queue_id) (txd_id)
360 testpmd> read txd 0 0 4
361 0x00000001 - 0x24C3C440 / 0x000F0000 - 0x2330003C
366 Get loaded dynamic device personalization (DDP) package info list::
368 testpmd> ddp get list (port_id)
373 Display information about dynamic device personalization (DDP) profile::
375 testpmd> ddp get info (profile_path)
380 Display VF statistics::
382 testpmd> show vf stats (port_id) (vf_id)
387 Reset VF statistics::
389 testpmd> clear vf stats (port_id) (vf_id)
391 show port pctype mapping
392 ~~~~~~~~~~~~~~~~~~~~~~~~
394 List all items from the pctype mapping table::
396 testpmd> show port (port_id) pctype mapping
398 show rx offloading capabilities
399 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
401 List all per queue and per port Rx offloading capabilities of a port::
403 testpmd> show port (port_id) rx_offload capabilities
405 show rx offloading configuration
406 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
408 List port level and all queue level Rx offloading configuration::
410 testpmd> show port (port_id) rx_offload configuration
412 show tx offloading capabilities
413 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
415 List all per queue and per port Tx offloading capabilities of a port::
417 testpmd> show port (port_id) tx_offload capabilities
419 show tx offloading configuration
420 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
422 List port level and all queue level Tx offloading configuration::
424 testpmd> show port (port_id) tx_offload configuration
426 show tx metadata setting
427 ~~~~~~~~~~~~~~~~~~~~~~~~
429 Show Tx metadata value set for a specific port::
431 testpmd> show port (port_id) tx_metadata
433 Configuration Functions
434 -----------------------
436 The testpmd application can be configured from the runtime as well as from the command-line.
438 This section details the available configuration functions that are available.
442 Configuration changes only become active when forwarding is started/restarted.
447 Reset forwarding to the default configuration::
454 Set the debug verbosity level::
456 testpmd> set verbose (level)
458 Available levels are as following:
460 * ``0`` silent except for error.
461 * ``1`` fully verbose except for Tx packets.
462 * ``2`` fully verbose except for Rx packets.
463 * ``> 2`` fully verbose.
468 Set the log level for a log type::
470 testpmd> set log global|(type) (level)
474 * ``type`` is the log name.
476 * ``level`` is the log level.
478 For example, to change the global log level::
479 testpmd> set log global (level)
481 Regexes can also be used for type. To change log level of user1, user2 and user3::
482 testpmd> set log user[1-3] (level)
487 Set the number of ports used by the application:
491 This is equivalent to the ``--nb-ports`` command-line option.
496 Set the number of cores used by the application::
498 testpmd> set nbcore (num)
500 This is equivalent to the ``--nb-cores`` command-line option.
504 The number of cores used must not be greater than number of ports used multiplied by the number of queues per port.
509 Set the forwarding cores hexadecimal mask::
511 testpmd> set coremask (mask)
513 This is equivalent to the ``--coremask`` command-line option.
517 The master lcore is reserved for command line parsing only and cannot be masked on for packet forwarding.
522 Set the forwarding ports hexadecimal mask::
524 testpmd> set portmask (mask)
526 This is equivalent to the ``--portmask`` command-line option.
531 Set number of packets per burst::
533 testpmd> set burst (num)
535 This is equivalent to the ``--burst command-line`` option.
537 When retry is enabled, the transmit delay time and number of retries can also be set::
539 testpmd> set burst tx delay (microseconds) retry (num)
544 Set the length of each segment of the TX-ONLY packets or length of packet for FLOWGEN mode::
546 testpmd> set txpkts (x[,y]*)
548 Where x[,y]* represents a CSV list of values, without white space.
553 Set the split policy for the TX packets, applicable for TX-ONLY and CSUM forwarding modes::
555 testpmd> set txsplit (off|on|rand)
559 * ``off`` disable packet copy & split for CSUM mode.
561 * ``on`` split outgoing packet into multiple segments. Size of each segment
562 and number of segments per packet is determined by ``set txpkts`` command
565 * ``rand`` same as 'on', but number of segments per each packet is a random value between 1 and total number of segments.
570 Set the list of forwarding cores::
572 testpmd> set corelist (x[,y]*)
574 For example, to change the forwarding cores:
576 .. code-block:: console
578 testpmd> set corelist 3,1
579 testpmd> show config fwd
581 io packet forwarding - ports=2 - cores=2 - streams=2 - NUMA support disabled
582 Logical Core 3 (socket 0) forwards packets on 1 streams:
583 RX P=0/Q=0 (socket 0) -> TX P=1/Q=0 (socket 0) peer=02:00:00:00:00:01
584 Logical Core 1 (socket 0) forwards packets on 1 streams:
585 RX P=1/Q=0 (socket 0) -> TX P=0/Q=0 (socket 0) peer=02:00:00:00:00:00
589 The cores are used in the same order as specified on the command line.
594 Set the list of forwarding ports::
596 testpmd> set portlist (x[,y]*)
598 For example, to change the port forwarding:
600 .. code-block:: console
602 testpmd> set portlist 0,2,1,3
603 testpmd> show config fwd
605 io packet forwarding - ports=4 - cores=1 - streams=4
606 Logical Core 3 (socket 0) forwards packets on 4 streams:
607 RX P=0/Q=0 (socket 0) -> TX P=2/Q=0 (socket 0) peer=02:00:00:00:00:01
608 RX P=2/Q=0 (socket 0) -> TX P=0/Q=0 (socket 0) peer=02:00:00:00:00:00
609 RX P=1/Q=0 (socket 0) -> TX P=3/Q=0 (socket 0) peer=02:00:00:00:00:03
610 RX P=3/Q=0 (socket 0) -> TX P=1/Q=0 (socket 0) peer=02:00:00:00:00:02
615 Select how to retrieve new ports created after "port attach" command::
617 testpmd> set port setup on (iterator|event)
619 For each new port, a setup is done.
620 It will find the probed ports via RTE_ETH_FOREACH_MATCHING_DEV loop
621 in iterator mode, or via RTE_ETH_EVENT_NEW in event mode.
626 Enable/disable tx loopback::
628 testpmd> set tx loopback (port_id) (on|off)
633 set drop enable bit for all queues::
635 testpmd> set all queues drop (port_id) (on|off)
637 set split drop enable (for VF)
638 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
640 set split drop enable bit for VF from PF::
642 testpmd> set vf split drop (port_id) (vf_id) (on|off)
644 set mac antispoof (for VF)
645 ~~~~~~~~~~~~~~~~~~~~~~~~~~
647 Set mac antispoof for a VF from the PF::
649 testpmd> set vf mac antispoof (port_id) (vf_id) (on|off)
654 Enable/disable MACsec offload::
656 testpmd> set macsec offload (port_id) on encrypt (on|off) replay-protect (on|off)
657 testpmd> set macsec offload (port_id) off
662 Configure MACsec secure connection (SC)::
664 testpmd> set macsec sc (tx|rx) (port_id) (mac) (pi)
668 The pi argument is ignored for tx.
669 Check the NIC Datasheet for hardware limits.
674 Configure MACsec secure association (SA)::
676 testpmd> set macsec sa (tx|rx) (port_id) (idx) (an) (pn) (key)
680 The IDX value must be 0 or 1.
681 Check the NIC Datasheet for hardware limits.
683 set broadcast mode (for VF)
684 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
686 Set broadcast mode for a VF from the PF::
688 testpmd> set vf broadcast (port_id) (vf_id) (on|off)
693 Set the VLAN strip on a port::
695 testpmd> vlan set strip (on|off) (port_id)
700 Set the VLAN strip for a queue on a port::
702 testpmd> vlan set stripq (on|off) (port_id,queue_id)
704 vlan set stripq (for VF)
705 ~~~~~~~~~~~~~~~~~~~~~~~~
707 Set VLAN strip for all queues in a pool for a VF from the PF::
709 testpmd> set vf vlan stripq (port_id) (vf_id) (on|off)
711 vlan set insert (for VF)
712 ~~~~~~~~~~~~~~~~~~~~~~~~
714 Set VLAN insert for a VF from the PF::
716 testpmd> set vf vlan insert (port_id) (vf_id) (vlan_id)
718 vlan set tag (for VF)
719 ~~~~~~~~~~~~~~~~~~~~~
721 Set VLAN tag for a VF from the PF::
723 testpmd> set vf vlan tag (port_id) (vf_id) (on|off)
725 vlan set antispoof (for VF)
726 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
728 Set VLAN antispoof for a VF from the PF::
730 testpmd> set vf vlan antispoof (port_id) (vf_id) (on|off)
735 Set the VLAN filter on a port::
737 testpmd> vlan set filter (on|off) (port_id)
742 Set the VLAN QinQ (extended queue in queue) on for a port::
744 testpmd> vlan set qinq (on|off) (port_id)
749 Set the inner or outer VLAN TPID for packet filtering on a port::
751 testpmd> vlan set (inner|outer) tpid (value) (port_id)
755 TPID value must be a 16-bit number (value <= 65536).
760 Add a VLAN ID, or all identifiers, to the set of VLAN identifiers filtered by port ID::
762 testpmd> rx_vlan add (vlan_id|all) (port_id)
766 VLAN filter must be set on that port. VLAN ID < 4096.
767 Depending on the NIC used, number of vlan_ids may be limited to the maximum entries
768 in VFTA table. This is important if enabling all vlan_ids.
773 Remove a VLAN ID, or all identifiers, from the set of VLAN identifiers filtered by port ID::
775 testpmd> rx_vlan rm (vlan_id|all) (port_id)
780 Add a VLAN ID, to the set of VLAN identifiers filtered for VF(s) for port ID::
782 testpmd> rx_vlan add (vlan_id) port (port_id) vf (vf_mask)
787 Remove a VLAN ID, from the set of VLAN identifiers filtered for VF(s) for port ID::
789 testpmd> rx_vlan rm (vlan_id) port (port_id) vf (vf_mask)
794 Add a tunnel filter on a port::
796 testpmd> tunnel_filter add (port_id) (outer_mac) (inner_mac) (ip_addr) \
797 (inner_vlan) (vxlan|nvgre|ipingre) (imac-ivlan|imac-ivlan-tenid|\
798 imac-tenid|imac|omac-imac-tenid|oip|iip) (tenant_id) (queue_id)
800 The available information categories are:
802 * ``vxlan``: Set tunnel type as VXLAN.
804 * ``nvgre``: Set tunnel type as NVGRE.
806 * ``ipingre``: Set tunnel type as IP-in-GRE.
808 * ``imac-ivlan``: Set filter type as Inner MAC and VLAN.
810 * ``imac-ivlan-tenid``: Set filter type as Inner MAC, VLAN and tenant ID.
812 * ``imac-tenid``: Set filter type as Inner MAC and tenant ID.
814 * ``imac``: Set filter type as Inner MAC.
816 * ``omac-imac-tenid``: Set filter type as Outer MAC, Inner MAC and tenant ID.
818 * ``oip``: Set filter type as Outer IP.
820 * ``iip``: Set filter type as Inner IP.
824 testpmd> tunnel_filter add 0 68:05:CA:28:09:82 00:00:00:00:00:00 \
825 192.168.2.2 0 ipingre oip 1 1
827 Set an IP-in-GRE tunnel on port 0, and the filter type is Outer IP.
832 Remove a tunnel filter on a port::
834 testpmd> tunnel_filter rm (port_id) (outer_mac) (inner_mac) (ip_addr) \
835 (inner_vlan) (vxlan|nvgre|ipingre) (imac-ivlan|imac-ivlan-tenid|\
836 imac-tenid|imac|omac-imac-tenid|oip|iip) (tenant_id) (queue_id)
841 Add an UDP port for VXLAN packet filter on a port::
843 testpmd> rx_vxlan_port add (udp_port) (port_id)
848 Remove an UDP port for VXLAN packet filter on a port::
850 testpmd> rx_vxlan_port rm (udp_port) (port_id)
855 Set hardware insertion of VLAN IDs in packets sent on a port::
857 testpmd> tx_vlan set (port_id) vlan_id[, vlan_id_outer]
859 For example, set a single VLAN ID (5) insertion on port 0::
863 Or, set double VLAN ID (inner: 2, outer: 3) insertion on port 1::
871 Set port based hardware insertion of VLAN ID in packets sent on a port::
873 testpmd> tx_vlan set pvid (port_id) (vlan_id) (on|off)
878 Disable hardware insertion of a VLAN header in packets sent on a port::
880 testpmd> tx_vlan reset (port_id)
885 Select hardware or software calculation of the checksum when
886 transmitting a packet using the ``csum`` forwarding engine::
888 testpmd> csum set (ip|udp|tcp|sctp|outer-ip|outer-udp) (hw|sw) (port_id)
892 * ``ip|udp|tcp|sctp`` always relate to the inner layer.
894 * ``outer-ip`` relates to the outer IP layer (only for IPv4) in the case where the packet is recognized
895 as a tunnel packet by the forwarding engine (vxlan, gre and ipip are
896 supported). See also the ``csum parse-tunnel`` command.
898 * ``outer-udp`` relates to the outer UDP layer in the case where the packet is recognized
899 as a tunnel packet by the forwarding engine (vxlan, vxlan-gpe are
900 supported). See also the ``csum parse-tunnel`` command.
904 Check the NIC Datasheet for hardware limits.
909 Set RSS queue region span on a port::
911 testpmd> set port (port_id) queue-region region_id (value) \
912 queue_start_index (value) queue_num (value)
914 Set flowtype mapping on a RSS queue region on a port::
916 testpmd> set port (port_id) queue-region region_id (value) flowtype (value)
920 * For the flowtype(pctype) of packet,the specific index for each type has
921 been defined in file i40e_type.h as enum i40e_filter_pctype.
923 Set user priority mapping on a RSS queue region on a port::
925 testpmd> set port (port_id) queue-region UP (value) region_id (value)
927 Flush all queue region related configuration on a port::
929 testpmd> set port (port_id) queue-region flush (on|off)
933 * "on"is just an enable function which server for other configuration,
934 it is for all configuration about queue region from up layer,
935 at first will only keep in DPDK softwarestored in driver,
936 only after "flush on", it commit all configuration to HW.
937 "off" is just clean all configuration about queue region just now,
938 and restore all to DPDK i40e driver default config when start up.
940 Show all queue region related configuration info on a port::
942 testpmd> show port (port_id) queue-region
946 Queue region only support on PF by now, so these command is
947 only for configuration of queue region on PF port.
952 Define how tunneled packets should be handled by the csum forward
955 testpmd> csum parse-tunnel (on|off) (tx_port_id)
957 If enabled, the csum forward engine will try to recognize supported
958 tunnel headers (vxlan, gre, ipip).
960 If disabled, treat tunnel packets as non-tunneled packets (a inner
961 header is handled as a packet payload).
965 The port argument is the TX port like in the ``csum set`` command.
969 Consider a packet in packet like the following::
971 eth_out/ipv4_out/udp_out/vxlan/eth_in/ipv4_in/tcp_in
973 * If parse-tunnel is enabled, the ``ip|udp|tcp|sctp`` parameters of ``csum set``
974 command relate to the inner headers (here ``ipv4_in`` and ``tcp_in``), and the
975 ``outer-ip|outer-udp`` parameter relates to the outer headers (here ``ipv4_out`` and ``udp_out``).
977 * If parse-tunnel is disabled, the ``ip|udp|tcp|sctp`` parameters of ``csum set``
978 command relate to the outer headers, here ``ipv4_out`` and ``udp_out``.
983 Display tx checksum offload configuration::
985 testpmd> csum show (port_id)
990 Enable TCP Segmentation Offload (TSO) in the ``csum`` forwarding engine::
992 testpmd> tso set (segsize) (port_id)
996 Check the NIC datasheet for hardware limits.
1001 Display the status of TCP Segmentation Offload::
1003 testpmd> tso show (port_id)
1008 Enable or disable GRO in ``csum`` forwarding engine::
1010 testpmd> set port <port_id> gro on|off
1012 If enabled, the csum forwarding engine will perform GRO on the TCP/IPv4
1013 packets received from the given port.
1015 If disabled, packets received from the given port won't be performed
1016 GRO. By default, GRO is disabled for all ports.
1020 When enable GRO for a port, TCP/IPv4 packets received from the port
1021 will be performed GRO. After GRO, all merged packets have bad
1022 checksums, since the GRO library doesn't re-calculate checksums for
1023 the merged packets. Therefore, if users want the merged packets to
1024 have correct checksums, please select HW IP checksum calculation and
1025 HW TCP checksum calculation for the port which the merged packets are
1031 Display GRO configuration for a given port::
1033 testpmd> show port <port_id> gro
1038 Set the cycle to flush the GROed packets from reassembly tables::
1040 testpmd> set gro flush <cycles>
1042 When enable GRO, the csum forwarding engine performs GRO on received
1043 packets, and the GROed packets are stored in reassembly tables. Users
1044 can use this command to determine when the GROed packets are flushed
1045 from the reassembly tables.
1047 The ``cycles`` is measured in GRO operation times. The csum forwarding
1048 engine flushes the GROed packets from the tables every ``cycles`` GRO
1051 By default, the value of ``cycles`` is 1, which means flush GROed packets
1052 from the reassembly tables as soon as one GRO operation finishes. The value
1053 of ``cycles`` should be in the range of 1 to ``GRO_MAX_FLUSH_CYCLES``.
1055 Please note that the large value of ``cycles`` may cause the poor TCP/IP
1056 stack performance. Because the GROed packets are delayed to arrive the
1057 stack, thus causing more duplicated ACKs and TCP retransmissions.
1062 Toggle per-port GSO support in ``csum`` forwarding engine::
1064 testpmd> set port <port_id> gso on|off
1066 If enabled, the csum forwarding engine will perform GSO on supported IPv4
1067 packets, transmitted on the given port.
1069 If disabled, packets transmitted on the given port will not undergo GSO.
1070 By default, GSO is disabled for all ports.
1074 When GSO is enabled on a port, supported IPv4 packets transmitted on that
1075 port undergo GSO. Afterwards, the segmented packets are represented by
1076 multi-segment mbufs; however, the csum forwarding engine doesn't calculation
1077 of checksums for GSO'd segments in SW. As a result, if users want correct
1078 checksums in GSO segments, they should enable HW checksum calculation for
1081 For example, HW checksum calculation for VxLAN GSO'd packets may be enabled
1082 by setting the following options in the csum forwarding engine:
1084 testpmd> csum set outer_ip hw <port_id>
1086 testpmd> csum set ip hw <port_id>
1088 testpmd> csum set tcp hw <port_id>
1090 UDP GSO is the same as IP fragmentation, which treats the UDP header
1091 as the payload and does not modify it during segmentation. That is,
1092 after UDP GSO, only the first output fragment has the original UDP
1093 header. Therefore, users need to enable HW IP checksum calculation
1094 and SW UDP checksum calculation for GSO-enabled ports, if they want
1095 correct checksums for UDP/IPv4 packets.
1100 Set the maximum GSO segment size (measured in bytes), which includes the
1101 packet header and the packet payload for GSO-enabled ports (global)::
1103 testpmd> set gso segsz <length>
1108 Display the status of Generic Segmentation Offload for a given port::
1110 testpmd> show port <port_id> gso
1115 Add an alternative MAC address to a port::
1117 testpmd> mac_addr add (port_id) (XX:XX:XX:XX:XX:XX)
1122 Remove a MAC address from a port::
1124 testpmd> mac_addr remove (port_id) (XX:XX:XX:XX:XX:XX)
1126 mac_addr add (for VF)
1127 ~~~~~~~~~~~~~~~~~~~~~
1129 Add an alternative MAC address for a VF to a port::
1131 testpmd> mac_add add port (port_id) vf (vf_id) (XX:XX:XX:XX:XX:XX)
1136 Set the default MAC address for a port::
1138 testpmd> mac_addr set (port_id) (XX:XX:XX:XX:XX:XX)
1140 mac_addr set (for VF)
1141 ~~~~~~~~~~~~~~~~~~~~~
1143 Set the MAC address for a VF from the PF::
1145 testpmd> set vf mac addr (port_id) (vf_id) (XX:XX:XX:XX:XX:XX)
1150 Set the forwarding peer address for certain port::
1152 testpmd> set eth-peer (port_id) (perr_addr)
1154 This is equivalent to the ``--eth-peer`` command-line option.
1159 Set the unicast hash filter(s) on/off for a port::
1161 testpmd> set port (port_id) uta (XX:XX:XX:XX:XX:XX|all) (on|off)
1166 Set the promiscuous mode on for a port or for all ports.
1167 In promiscuous mode packets are not dropped if they aren't for the specified MAC address::
1169 testpmd> set promisc (port_id|all) (on|off)
1174 Set the allmulti mode for a port or for all ports::
1176 testpmd> set allmulti (port_id|all) (on|off)
1178 Same as the ifconfig (8) option. Controls how multicast packets are handled.
1180 set promisc (for VF)
1181 ~~~~~~~~~~~~~~~~~~~~
1183 Set the unicast promiscuous mode for a VF from PF.
1184 It's supported by Intel i40e NICs now.
1185 In promiscuous mode packets are not dropped if they aren't for the specified MAC address::
1187 testpmd> set vf promisc (port_id) (vf_id) (on|off)
1189 set allmulticast (for VF)
1190 ~~~~~~~~~~~~~~~~~~~~~~~~~
1192 Set the multicast promiscuous mode for a VF from PF.
1193 It's supported by Intel i40e NICs now.
1194 In promiscuous mode packets are not dropped if they aren't for the specified MAC address::
1196 testpmd> set vf allmulti (port_id) (vf_id) (on|off)
1198 set tx max bandwidth (for VF)
1199 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1201 Set TX max absolute bandwidth (Mbps) for a VF from PF::
1203 testpmd> set vf tx max-bandwidth (port_id) (vf_id) (max_bandwidth)
1205 set tc tx min bandwidth (for VF)
1206 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1208 Set all TCs' TX min relative bandwidth (%) for a VF from PF::
1210 testpmd> set vf tc tx min-bandwidth (port_id) (vf_id) (bw1, bw2, ...)
1212 set tc tx max bandwidth (for VF)
1213 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1215 Set a TC's TX max absolute bandwidth (Mbps) for a VF from PF::
1217 testpmd> set vf tc tx max-bandwidth (port_id) (vf_id) (tc_no) (max_bandwidth)
1219 set tc strict link priority mode
1220 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1222 Set some TCs' strict link priority mode on a physical port::
1224 testpmd> set tx strict-link-priority (port_id) (tc_bitmap)
1226 set tc tx min bandwidth
1227 ~~~~~~~~~~~~~~~~~~~~~~~
1229 Set all TCs' TX min relative bandwidth (%) globally for all PF and VFs::
1231 testpmd> set tc tx min-bandwidth (port_id) (bw1, bw2, ...)
1236 Set the link flow control parameter on a port::
1238 testpmd> set flow_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \
1239 (pause_time) (send_xon) mac_ctrl_frame_fwd (on|off) \
1240 autoneg (on|off) (port_id)
1244 * ``high_water`` (integer): High threshold value to trigger XOFF.
1246 * ``low_water`` (integer): Low threshold value to trigger XON.
1248 * ``pause_time`` (integer): Pause quota in the Pause frame.
1250 * ``send_xon`` (0/1): Send XON frame.
1252 * ``mac_ctrl_frame_fwd``: Enable receiving MAC control frames.
1254 * ``autoneg``: Change the auto-negotiation parameter.
1259 Set the priority flow control parameter on a port::
1261 testpmd> set pfc_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \
1262 (pause_time) (priority) (port_id)
1266 * ``high_water`` (integer): High threshold value.
1268 * ``low_water`` (integer): Low threshold value.
1270 * ``pause_time`` (integer): Pause quota in the Pause frame.
1272 * ``priority`` (0-7): VLAN User Priority.
1277 Set statistics mapping (qmapping 0..15) for RX/TX queue on port::
1279 testpmd> set stat_qmap (tx|rx) (port_id) (queue_id) (qmapping)
1281 For example, to set rx queue 2 on port 0 to mapping 5::
1283 testpmd>set stat_qmap rx 0 2 5
1285 set xstats-hide-zero
1286 ~~~~~~~~~~~~~~~~~~~~
1288 Set the option to hide zero values for xstats display::
1290 testpmd> set xstats-hide-zero on|off
1294 By default, the zero values are displayed for xstats.
1296 set port - rx/tx (for VF)
1297 ~~~~~~~~~~~~~~~~~~~~~~~~~
1299 Set VF receive/transmit from a port::
1301 testpmd> set port (port_id) vf (vf_id) (rx|tx) (on|off)
1303 set port - mac address filter (for VF)
1304 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1306 Add/Remove unicast or multicast MAC addr filter for a VF::
1308 testpmd> set port (port_id) vf (vf_id) (mac_addr) \
1309 (exact-mac|exact-mac-vlan|hashmac|hashmac-vlan) (on|off)
1311 set port - rx mode(for VF)
1312 ~~~~~~~~~~~~~~~~~~~~~~~~~~
1314 Set the VF receive mode of a port::
1316 testpmd> set port (port_id) vf (vf_id) \
1317 rxmode (AUPE|ROPE|BAM|MPE) (on|off)
1319 The available receive modes are:
1321 * ``AUPE``: Accepts untagged VLAN.
1323 * ``ROPE``: Accepts unicast hash.
1325 * ``BAM``: Accepts broadcast packets.
1327 * ``MPE``: Accepts all multicast packets.
1329 set port - tx_rate (for Queue)
1330 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1332 Set TX rate limitation for a queue on a port::
1334 testpmd> set port (port_id) queue (queue_id) rate (rate_value)
1336 set port - tx_rate (for VF)
1337 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
1339 Set TX rate limitation for queues in VF on a port::
1341 testpmd> set port (port_id) vf (vf_id) rate (rate_value) queue_mask (queue_mask)
1343 set port - mirror rule
1344 ~~~~~~~~~~~~~~~~~~~~~~
1346 Set pool or vlan type mirror rule for a port::
1348 testpmd> set port (port_id) mirror-rule (rule_id) \
1349 (pool-mirror-up|pool-mirror-down|vlan-mirror) \
1350 (poolmask|vlanid[,vlanid]*) dst-pool (pool_id) (on|off)
1352 Set link mirror rule for a port::
1354 testpmd> set port (port_id) mirror-rule (rule_id) \
1355 (uplink-mirror|downlink-mirror) dst-pool (pool_id) (on|off)
1357 For example to enable mirror traffic with vlan 0,1 to pool 0::
1359 set port 0 mirror-rule 0 vlan-mirror 0,1 dst-pool 0 on
1361 reset port - mirror rule
1362 ~~~~~~~~~~~~~~~~~~~~~~~~
1364 Reset a mirror rule for a port::
1366 testpmd> reset port (port_id) mirror-rule (rule_id)
1371 Set the flush on RX streams before forwarding.
1372 The default is flush ``on``.
1373 Mainly used with PCAP drivers to turn off the default behavior of flushing the first 512 packets on RX streams::
1375 testpmd> set flush_rx off
1380 Set the bypass mode for the lowest port on bypass enabled NIC::
1382 testpmd> set bypass mode (normal|bypass|isolate) (port_id)
1387 Set the event required to initiate specified bypass mode for the lowest port on a bypass enabled::
1389 testpmd> set bypass event (timeout|os_on|os_off|power_on|power_off) \
1390 mode (normal|bypass|isolate) (port_id)
1394 * ``timeout``: Enable bypass after watchdog timeout.
1396 * ``os_on``: Enable bypass when OS/board is powered on.
1398 * ``os_off``: Enable bypass when OS/board is powered off.
1400 * ``power_on``: Enable bypass when power supply is turned on.
1402 * ``power_off``: Enable bypass when power supply is turned off.
1408 Set the bypass watchdog timeout to ``n`` seconds where 0 = instant::
1410 testpmd> set bypass timeout (0|1.5|2|3|4|8|16|32)
1415 Show the bypass configuration for a bypass enabled NIC using the lowest port on the NIC::
1417 testpmd> show bypass config (port_id)
1422 Set link up for a port::
1424 testpmd> set link-up port (port id)
1429 Set link down for a port::
1431 testpmd> set link-down port (port id)
1436 Enable E-tag insertion for a VF on a port::
1438 testpmd> E-tag set insertion on port-tag-id (value) port (port_id) vf (vf_id)
1440 Disable E-tag insertion for a VF on a port::
1442 testpmd> E-tag set insertion off port (port_id) vf (vf_id)
1444 Enable/disable E-tag stripping on a port::
1446 testpmd> E-tag set stripping (on|off) port (port_id)
1448 Enable/disable E-tag based forwarding on a port::
1450 testpmd> E-tag set forwarding (on|off) port (port_id)
1452 Add an E-tag forwarding filter on a port::
1454 testpmd> E-tag set filter add e-tag-id (value) dst-pool (pool_id) port (port_id)
1456 Delete an E-tag forwarding filter on a port::
1457 testpmd> E-tag set filter del e-tag-id (value) port (port_id)
1462 Load a dynamic device personalization (DDP) profile and store backup profile::
1464 testpmd> ddp add (port_id) (profile_path[,backup_profile_path])
1469 Delete a dynamic device personalization profile and restore backup profile::
1471 testpmd> ddp del (port_id) (backup_profile_path)
1476 List all items from the ptype mapping table::
1478 testpmd> ptype mapping get (port_id) (valid_only)
1482 * ``valid_only``: A flag indicates if only list valid items(=1) or all itemss(=0).
1484 Replace a specific or a group of software defined ptype with a new one::
1486 testpmd> ptype mapping replace (port_id) (target) (mask) (pkt_type)
1490 * ``target``: A specific software ptype or a mask to represent a group of software ptypes.
1492 * ``mask``: A flag indicate if "target" is a specific software ptype(=0) or a ptype mask(=1).
1494 * ``pkt_type``: The new software ptype to replace the old ones.
1496 Update hardware defined ptype to software defined packet type mapping table::
1498 testpmd> ptype mapping update (port_id) (hw_ptype) (sw_ptype)
1502 * ``hw_ptype``: hardware ptype as the index of the ptype mapping table.
1504 * ``sw_ptype``: software ptype as the value of the ptype mapping table.
1506 Reset ptype mapping table::
1508 testpmd> ptype mapping reset (port_id)
1510 config per port Rx offloading
1511 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1513 Enable or disable a per port Rx offloading on all Rx queues of a port::
1515 testpmd> port config (port_id) rx_offload (offloading) on|off
1517 * ``offloading``: can be any of these offloading capability:
1518 vlan_strip, ipv4_cksum, udp_cksum, tcp_cksum, tcp_lro,
1519 qinq_strip, outer_ipv4_cksum, macsec_strip,
1520 header_split, vlan_filter, vlan_extend, jumbo_frame,
1521 crc_strip, scatter, timestamp, security, keep_crc
1523 This command should be run when the port is stopped, or else it will fail.
1525 config per queue Rx offloading
1526 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1528 Enable or disable a per queue Rx offloading only on a specific Rx queue::
1530 testpmd> port (port_id) rxq (queue_id) rx_offload (offloading) on|off
1532 * ``offloading``: can be any of these offloading capability:
1533 vlan_strip, ipv4_cksum, udp_cksum, tcp_cksum, tcp_lro,
1534 qinq_strip, outer_ipv4_cksum, macsec_strip,
1535 header_split, vlan_filter, vlan_extend, jumbo_frame,
1536 crc_strip, scatter, timestamp, security, keep_crc
1538 This command should be run when the port is stopped, or else it will fail.
1540 config per port Tx offloading
1541 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1543 Enable or disable a per port Tx offloading on all Tx queues of a port::
1545 testpmd> port config (port_id) tx_offload (offloading) on|off
1547 * ``offloading``: can be any of these offloading capability:
1548 vlan_insert, ipv4_cksum, udp_cksum, tcp_cksum,
1549 sctp_cksum, tcp_tso, udp_tso, outer_ipv4_cksum,
1550 qinq_insert, vxlan_tnl_tso, gre_tnl_tso,
1551 ipip_tnl_tso, geneve_tnl_tso, macsec_insert,
1552 mt_lockfree, multi_segs, mbuf_fast_free, security,
1555 This command should be run when the port is stopped, or else it will fail.
1557 config per queue Tx offloading
1558 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1560 Enable or disable a per queue Tx offloading only on a specific Tx queue::
1562 testpmd> port (port_id) txq (queue_id) tx_offload (offloading) on|off
1564 * ``offloading``: can be any of these offloading capability:
1565 vlan_insert, ipv4_cksum, udp_cksum, tcp_cksum,
1566 sctp_cksum, tcp_tso, udp_tso, outer_ipv4_cksum,
1567 qinq_insert, vxlan_tnl_tso, gre_tnl_tso,
1568 ipip_tnl_tso, geneve_tnl_tso, macsec_insert,
1569 mt_lockfree, multi_segs, mbuf_fast_free, security
1571 This command should be run when the port is stopped, or else it will fail.
1573 Config VXLAN Encap outer layers
1574 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1576 Configure the outer layer to encapsulate a packet inside a VXLAN tunnel::
1578 set vxlan ip-version (ipv4|ipv6) vni (vni) udp-src (udp-src) \
1579 udp-dst (udp-dst) ip-src (ip-src) ip-dst (ip-dst) eth-src (eth-src) \
1582 set vxlan-with-vlan ip-version (ipv4|ipv6) vni (vni) udp-src (udp-src) \
1583 udp-dst (udp-dst) ip-src (ip-src) ip-dst (ip-dst) vlan-tci (vlan-tci) \
1584 eth-src (eth-src) eth-dst (eth-dst)
1586 Those command will set an internal configuration inside testpmd, any following
1587 flow rule using the action vxlan_encap will use the last configuration set.
1588 To have a different encapsulation header, one of those commands must be called
1589 before the flow rule creation.
1591 Config NVGRE Encap outer layers
1592 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1594 Configure the outer layer to encapsulate a packet inside a NVGRE tunnel::
1596 set nvgre ip-version (ipv4|ipv6) tni (tni) ip-src (ip-src) ip-dst (ip-dst) \
1597 eth-src (eth-src) eth-dst (eth-dst)
1598 set nvgre-with-vlan ip-version (ipv4|ipv6) tni (tni) ip-src (ip-src) \
1599 ip-dst (ip-dst) vlan-tci (vlan-tci) eth-src (eth-src) eth-dst (eth-dst)
1601 Those command will set an internal configuration inside testpmd, any following
1602 flow rule using the action nvgre_encap will use the last configuration set.
1603 To have a different encapsulation header, one of those commands must be called
1604 before the flow rule creation.
1609 Configure the l2 to be used when encapsulating a packet with L2::
1611 set l2_encap ip-version (ipv4|ipv6) eth-src (eth-src) eth-dst (eth-dst)
1612 set l2_encap-with-vlan ip-version (ipv4|ipv6) vlan-tci (vlan-tci) \
1613 eth-src (eth-src) eth-dst (eth-dst)
1615 Those commands will set an internal configuration inside testpmd, any following
1616 flow rule using the action l2_encap will use the last configuration set.
1617 To have a different encapsulation header, one of those commands must be called
1618 before the flow rule creation.
1623 Configure the l2 to be removed when decapsulating a packet with L2::
1625 set l2_decap ip-version (ipv4|ipv6)
1626 set l2_decap-with-vlan ip-version (ipv4|ipv6)
1628 Those commands will set an internal configuration inside testpmd, any following
1629 flow rule using the action l2_decap will use the last configuration set.
1630 To have a different encapsulation header, one of those commands must be called
1631 before the flow rule creation.
1633 Config MPLSoGRE Encap outer layers
1634 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1636 Configure the outer layer to encapsulate a packet inside a MPLSoGRE tunnel::
1638 set mplsogre_encap ip-version (ipv4|ipv6) label (label) \
1639 ip-src (ip-src) ip-dst (ip-dst) eth-src (eth-src) eth-dst (eth-dst)
1640 set mplsogre_encap-with-vlan ip-version (ipv4|ipv6) label (label) \
1641 ip-src (ip-src) ip-dst (ip-dst) vlan-tci (vlan-tci) \
1642 eth-src (eth-src) eth-dst (eth-dst)
1644 Those command will set an internal configuration inside testpmd, any following
1645 flow rule using the action mplsogre_encap will use the last configuration set.
1646 To have a different encapsulation header, one of those commands must be called
1647 before the flow rule creation.
1649 Config MPLSoGRE Decap outer layers
1650 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1652 Configure the outer layer to decapsulate MPLSoGRE packet::
1654 set mplsogre_decap ip-version (ipv4|ipv6)
1655 set mplsogre_decap-with-vlan ip-version (ipv4|ipv6)
1657 Those command will set an internal configuration inside testpmd, any following
1658 flow rule using the action mplsogre_decap will use the last configuration set.
1659 To have a different decapsulation header, one of those commands must be called
1660 before the flow rule creation.
1662 Config MPLSoUDP Encap outer layers
1663 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1665 Configure the outer layer to encapsulate a packet inside a MPLSoUDP tunnel::
1667 set mplsoudp_encap ip-version (ipv4|ipv6) label (label) udp-src (udp-src) \
1668 udp-dst (udp-dst) ip-src (ip-src) ip-dst (ip-dst) \
1669 eth-src (eth-src) eth-dst (eth-dst)
1670 set mplsoudp_encap-with-vlan ip-version (ipv4|ipv6) label (label) \
1671 udp-src (udp-src) udp-dst (udp-dst) ip-src (ip-src) ip-dst (ip-dst) \
1672 vlan-tci (vlan-tci) eth-src (eth-src) eth-dst (eth-dst)
1674 Those command will set an internal configuration inside testpmd, any following
1675 flow rule using the action mplsoudp_encap will use the last configuration set.
1676 To have a different encapsulation header, one of those commands must be called
1677 before the flow rule creation.
1679 Config MPLSoUDP Decap outer layers
1680 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1682 Configure the outer layer to decapsulate MPLSoUDP packet::
1684 set mplsoudp_decap ip-version (ipv4|ipv6)
1685 set mplsoudp_decap-with-vlan ip-version (ipv4|ipv6)
1687 Those command will set an internal configuration inside testpmd, any following
1688 flow rule using the action mplsoudp_decap will use the last configuration set.
1689 To have a different decapsulation header, one of those commands must be called
1690 before the flow rule creation.
1695 The following sections show functions for configuring ports.
1699 Port configuration changes only become active when forwarding is started/restarted.
1704 Attach a port specified by pci address or virtual device args::
1706 testpmd> port attach (identifier)
1708 To attach a new pci device, the device should be recognized by kernel first.
1709 Then it should be moved under DPDK management.
1710 Finally the port can be attached to testpmd.
1712 For example, to move a pci device using ixgbe under DPDK management:
1714 .. code-block:: console
1716 # Check the status of the available devices.
1717 ./usertools/dpdk-devbind.py --status
1719 Network devices using DPDK-compatible driver
1720 ============================================
1723 Network devices using kernel driver
1724 ===================================
1725 0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused=
1728 # Bind the device to igb_uio.
1729 sudo ./usertools/dpdk-devbind.py -b igb_uio 0000:0a:00.0
1732 # Recheck the status of the devices.
1733 ./usertools/dpdk-devbind.py --status
1734 Network devices using DPDK-compatible driver
1735 ============================================
1736 0000:0a:00.0 '82599ES 10-Gigabit' drv=igb_uio unused=
1738 To attach a port created by virtual device, above steps are not needed.
1740 For example, to attach a port whose pci address is 0000:0a:00.0.
1742 .. code-block:: console
1744 testpmd> port attach 0000:0a:00.0
1745 Attaching a new port...
1746 EAL: PCI device 0000:0a:00.0 on NUMA socket -1
1747 EAL: probe driver: 8086:10fb rte_ixgbe_pmd
1748 EAL: PCI memory mapped at 0x7f83bfa00000
1749 EAL: PCI memory mapped at 0x7f83bfa80000
1750 PMD: eth_ixgbe_dev_init(): MAC: 2, PHY: 18, SFP+: 5
1751 PMD: eth_ixgbe_dev_init(): port 0 vendorID=0x8086 deviceID=0x10fb
1752 Port 0 is attached. Now total ports is 1
1755 For example, to attach a port created by pcap PMD.
1757 .. code-block:: console
1759 testpmd> port attach net_pcap0
1760 Attaching a new port...
1761 PMD: Initializing pmd_pcap for net_pcap0
1762 PMD: Creating pcap-backed ethdev on numa socket 0
1763 Port 0 is attached. Now total ports is 1
1766 In this case, identifier is ``net_pcap0``.
1767 This identifier format is the same as ``--vdev`` format of DPDK applications.
1769 For example, to re-attach a bonded port which has been previously detached,
1770 the mode and slave parameters must be given.
1772 .. code-block:: console
1774 testpmd> port attach net_bond_0,mode=0,slave=1
1775 Attaching a new port...
1776 EAL: Initializing pmd_bond for net_bond_0
1777 EAL: Create bonded device net_bond_0 on port 0 in mode 0 on socket 0.
1778 Port 0 is attached. Now total ports is 1
1785 Detach a specific port::
1787 testpmd> port detach (port_id)
1789 Before detaching a port, the port should be stopped and closed.
1791 For example, to detach a pci device port 0.
1793 .. code-block:: console
1795 testpmd> port stop 0
1798 testpmd> port close 0
1802 testpmd> port detach 0
1804 EAL: PCI device 0000:0a:00.0 on NUMA socket -1
1805 EAL: remove driver: 8086:10fb rte_ixgbe_pmd
1806 EAL: PCI memory unmapped at 0x7f83bfa00000
1807 EAL: PCI memory unmapped at 0x7f83bfa80000
1811 For example, to detach a virtual device port 0.
1813 .. code-block:: console
1815 testpmd> port stop 0
1818 testpmd> port close 0
1822 testpmd> port detach 0
1824 PMD: Closing pcap ethdev on numa socket 0
1825 Port 'net_pcap0' is detached. Now total ports is 0
1828 To remove a pci device completely from the system, first detach the port from testpmd.
1829 Then the device should be moved under kernel management.
1830 Finally the device can be removed using kernel pci hotplug functionality.
1832 For example, to move a pci device under kernel management:
1834 .. code-block:: console
1836 sudo ./usertools/dpdk-devbind.py -b ixgbe 0000:0a:00.0
1838 ./usertools/dpdk-devbind.py --status
1840 Network devices using DPDK-compatible driver
1841 ============================================
1844 Network devices using kernel driver
1845 ===================================
1846 0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused=igb_uio
1848 To remove a port created by a virtual device, above steps are not needed.
1853 Start all ports or a specific port::
1855 testpmd> port start (port_id|all)
1860 Stop all ports or a specific port::
1862 testpmd> port stop (port_id|all)
1867 Close all ports or a specific port::
1869 testpmd> port close (port_id|all)
1871 port config - queue ring size
1872 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1874 Configure a rx/tx queue ring size::
1876 testpmd> port (port_id) (rxq|txq) (queue_id) ring_size (value)
1878 Only take effect after command that (re-)start the port or command that setup specific queue.
1880 port start/stop queue
1881 ~~~~~~~~~~~~~~~~~~~~~
1883 Start/stop a rx/tx queue on a specific port::
1885 testpmd> port (port_id) (rxq|txq) (queue_id) (start|stop)
1887 port config - queue deferred start
1888 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1890 Switch on/off deferred start of a specific port queue::
1892 testpmd> port (port_id) (rxq|txq) (queue_id) deferred_start (on|off)
1895 ~~~~~~~~~~~~~~~~~~~~~
1897 Setup a rx/tx queue on a specific port::
1899 testpmd> port (port_id) (rxq|txq) (queue_id) setup
1901 Only take effect when port is started.
1906 Set the speed and duplex mode for all ports or a specific port::
1908 testpmd> port config (port_id|all) speed (10|100|1000|10000|25000|40000|50000|100000|auto) \
1909 duplex (half|full|auto)
1911 port config - queues/descriptors
1912 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1914 Set number of queues/descriptors for rxq, txq, rxd and txd::
1916 testpmd> port config all (rxq|txq|rxd|txd) (value)
1918 This is equivalent to the ``--rxq``, ``--txq``, ``--rxd`` and ``--txd`` command-line options.
1920 port config - max-pkt-len
1921 ~~~~~~~~~~~~~~~~~~~~~~~~~
1923 Set the maximum packet length::
1925 testpmd> port config all max-pkt-len (value)
1927 This is equivalent to the ``--max-pkt-len`` command-line option.
1929 port config - CRC Strip
1930 ~~~~~~~~~~~~~~~~~~~~~~~
1932 Set hardware CRC stripping on or off for all ports::
1934 testpmd> port config all crc-strip (on|off)
1936 CRC stripping is on by default.
1938 The ``off`` option is equivalent to the ``--disable-crc-strip`` command-line option.
1940 port config - scatter
1941 ~~~~~~~~~~~~~~~~~~~~~~~
1943 Set RX scatter mode on or off for all ports::
1945 testpmd> port config all scatter (on|off)
1947 RX scatter mode is off by default.
1949 The ``on`` option is equivalent to the ``--enable-scatter`` command-line option.
1951 port config - RX Checksum
1952 ~~~~~~~~~~~~~~~~~~~~~~~~~
1954 Set hardware RX checksum offload to on or off for all ports::
1956 testpmd> port config all rx-cksum (on|off)
1958 Checksum offload is off by default.
1960 The ``on`` option is equivalent to the ``--enable-rx-cksum`` command-line option.
1965 Set hardware VLAN on or off for all ports::
1967 testpmd> port config all hw-vlan (on|off)
1969 Hardware VLAN is off by default.
1971 The ``on`` option is equivalent to the ``--enable-hw-vlan`` command-line option.
1973 port config - VLAN filter
1974 ~~~~~~~~~~~~~~~~~~~~~~~~~
1976 Set hardware VLAN filter on or off for all ports::
1978 testpmd> port config all hw-vlan-filter (on|off)
1980 Hardware VLAN filter is off by default.
1982 The ``on`` option is equivalent to the ``--enable-hw-vlan-filter`` command-line option.
1984 port config - VLAN strip
1985 ~~~~~~~~~~~~~~~~~~~~~~~~
1987 Set hardware VLAN strip on or off for all ports::
1989 testpmd> port config all hw-vlan-strip (on|off)
1991 Hardware VLAN strip is off by default.
1993 The ``on`` option is equivalent to the ``--enable-hw-vlan-strip`` command-line option.
1995 port config - VLAN extend
1996 ~~~~~~~~~~~~~~~~~~~~~~~~~
1998 Set hardware VLAN extend on or off for all ports::
2000 testpmd> port config all hw-vlan-extend (on|off)
2002 Hardware VLAN extend is off by default.
2004 The ``on`` option is equivalent to the ``--enable-hw-vlan-extend`` command-line option.
2006 port config - Drop Packets
2007 ~~~~~~~~~~~~~~~~~~~~~~~~~~
2009 Set packet drop for packets with no descriptors on or off for all ports::
2011 testpmd> port config all drop-en (on|off)
2013 Packet dropping for packets with no descriptors is off by default.
2015 The ``on`` option is equivalent to the ``--enable-drop-en`` command-line option.
2020 Set the RSS (Receive Side Scaling) mode on or off::
2022 testpmd> port config all rss (all|default|ip|tcp|udp|sctp|ether|port|vxlan|geneve|nvgre|none)
2024 RSS is on by default.
2026 The ``all`` option is equivalent to ip|tcp|udp|sctp|ether.
2027 The ``default`` option enables all supported RSS types reported by device info.
2028 The ``none`` option is equivalent to the ``--disable-rss`` command-line option.
2030 port config - RSS Reta
2031 ~~~~~~~~~~~~~~~~~~~~~~
2033 Set the RSS (Receive Side Scaling) redirection table::
2035 testpmd> port config all rss reta (hash,queue)[,(hash,queue)]
2040 Set the DCB mode for an individual port::
2042 testpmd> port config (port_id) dcb vt (on|off) (traffic_class) pfc (on|off)
2044 The traffic class should be 4 or 8.
2049 Set the number of packets per burst::
2051 testpmd> port config all burst (value)
2053 This is equivalent to the ``--burst`` command-line option.
2055 port config - Threshold
2056 ~~~~~~~~~~~~~~~~~~~~~~~
2058 Set thresholds for TX/RX queues::
2060 testpmd> port config all (threshold) (value)
2062 Where the threshold type can be:
2064 * ``txpt:`` Set the prefetch threshold register of the TX rings, 0 <= value <= 255.
2066 * ``txht:`` Set the host threshold register of the TX rings, 0 <= value <= 255.
2068 * ``txwt:`` Set the write-back threshold register of the TX rings, 0 <= value <= 255.
2070 * ``rxpt:`` Set the prefetch threshold register of the RX rings, 0 <= value <= 255.
2072 * ``rxht:`` Set the host threshold register of the RX rings, 0 <= value <= 255.
2074 * ``rxwt:`` Set the write-back threshold register of the RX rings, 0 <= value <= 255.
2076 * ``txfreet:`` Set the transmit free threshold of the TX rings, 0 <= value <= txd.
2078 * ``rxfreet:`` Set the transmit free threshold of the RX rings, 0 <= value <= rxd.
2080 * ``txrst:`` Set the transmit RS bit threshold of TX rings, 0 <= value <= txd.
2082 These threshold options are also available from the command-line.
2087 Set the value of ether-type for E-tag::
2089 testpmd> port config (port_id|all) l2-tunnel E-tag ether-type (value)
2091 Enable/disable the E-tag support::
2093 testpmd> port config (port_id|all) l2-tunnel E-tag (enable|disable)
2095 port config pctype mapping
2096 ~~~~~~~~~~~~~~~~~~~~~~~~~~
2098 Reset pctype mapping table::
2100 testpmd> port config (port_id) pctype mapping reset
2102 Update hardware defined pctype to software defined flow type mapping table::
2104 testpmd> port config (port_id) pctype mapping update (pctype_id_0[,pctype_id_1]*) (flow_type_id)
2108 * ``pctype_id_x``: hardware pctype id as index of bit in bitmask value of the pctype mapping table.
2110 * ``flow_type_id``: software flow type id as the index of the pctype mapping table.
2112 port config input set
2113 ~~~~~~~~~~~~~~~~~~~~~
2115 Config RSS/FDIR/FDIR flexible payload input set for some pctype::
2116 testpmd> port config (port_id) pctype (pctype_id) \
2117 (hash_inset|fdir_inset|fdir_flx_inset) \
2118 (get|set|clear) field (field_idx)
2120 Clear RSS/FDIR/FDIR flexible payload input set for some pctype::
2121 testpmd> port config (port_id) pctype (pctype_id) \
2122 (hash_inset|fdir_inset|fdir_flx_inset) clear all
2126 * ``pctype_id``: hardware packet classification types.
2127 * ``field_idx``: hardware field index.
2129 port config udp_tunnel_port
2130 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
2132 Add/remove UDP tunnel port for VXLAN/GENEVE tunneling protocols::
2133 testpmd> port config (port_id) udp_tunnel_port add|rm vxlan|geneve (udp_port)
2135 port config tx_metadata
2136 ~~~~~~~~~~~~~~~~~~~~~~~
2138 Set Tx metadata value per port.
2139 testpmd will add this value to any Tx packet sent from this port::
2141 testpmd> port config (port_id) tx_metadata (value)
2143 Link Bonding Functions
2144 ----------------------
2146 The Link Bonding functions make it possible to dynamically create and
2147 manage link bonding devices from within testpmd interactive prompt.
2149 create bonded device
2150 ~~~~~~~~~~~~~~~~~~~~
2152 Create a new bonding device::
2154 testpmd> create bonded device (mode) (socket)
2156 For example, to create a bonded device in mode 1 on socket 0::
2158 testpmd> create bonded device 1 0
2159 created new bonded device (port X)
2164 Adds Ethernet device to a Link Bonding device::
2166 testpmd> add bonding slave (slave id) (port id)
2168 For example, to add Ethernet device (port 6) to a Link Bonding device (port 10)::
2170 testpmd> add bonding slave 6 10
2173 remove bonding slave
2174 ~~~~~~~~~~~~~~~~~~~~
2176 Removes an Ethernet slave device from a Link Bonding device::
2178 testpmd> remove bonding slave (slave id) (port id)
2180 For example, to remove Ethernet slave device (port 6) to a Link Bonding device (port 10)::
2182 testpmd> remove bonding slave 6 10
2187 Set the Link Bonding mode of a Link Bonding device::
2189 testpmd> set bonding mode (value) (port id)
2191 For example, to set the bonding mode of a Link Bonding device (port 10) to broadcast (mode 3)::
2193 testpmd> set bonding mode 3 10
2198 Set an Ethernet slave device as the primary device on a Link Bonding device::
2200 testpmd> set bonding primary (slave id) (port id)
2202 For example, to set the Ethernet slave device (port 6) as the primary port of a Link Bonding device (port 10)::
2204 testpmd> set bonding primary 6 10
2209 Set the MAC address of a Link Bonding device::
2211 testpmd> set bonding mac (port id) (mac)
2213 For example, to set the MAC address of a Link Bonding device (port 10) to 00:00:00:00:00:01::
2215 testpmd> set bonding mac 10 00:00:00:00:00:01
2217 set bonding xmit_balance_policy
2218 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2220 Set the transmission policy for a Link Bonding device when it is in Balance XOR mode::
2222 testpmd> set bonding xmit_balance_policy (port_id) (l2|l23|l34)
2224 For example, set a Link Bonding device (port 10) to use a balance policy of layer 3+4 (IP addresses & UDP ports)::
2226 testpmd> set bonding xmit_balance_policy 10 l34
2229 set bonding mon_period
2230 ~~~~~~~~~~~~~~~~~~~~~~
2232 Set the link status monitoring polling period in milliseconds for a bonding device.
2234 This adds support for PMD slave devices which do not support link status interrupts.
2235 When the mon_period is set to a value greater than 0 then all PMD's which do not support
2236 link status ISR will be queried every polling interval to check if their link status has changed::
2238 testpmd> set bonding mon_period (port_id) (value)
2240 For example, to set the link status monitoring polling period of bonded device (port 5) to 150ms::
2242 testpmd> set bonding mon_period 5 150
2245 set bonding lacp dedicated_queue
2246 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2248 Enable dedicated tx/rx queues on bonding devices slaves to handle LACP control plane traffic
2249 when in mode 4 (link-aggregration-802.3ad)::
2251 testpmd> set bonding lacp dedicated_queues (port_id) (enable|disable)
2254 set bonding agg_mode
2255 ~~~~~~~~~~~~~~~~~~~~
2257 Enable one of the specific aggregators mode when in mode 4 (link-aggregration-802.3ad)::
2259 testpmd> set bonding agg_mode (port_id) (bandwidth|count|stable)
2265 Show the current configuration of a Link Bonding device::
2267 testpmd> show bonding config (port id)
2270 to show the configuration a Link Bonding device (port 9) with 3 slave devices (1, 3, 4)
2271 in balance mode with a transmission policy of layer 2+3::
2273 testpmd> show bonding config 9
2275 Balance Xmit Policy: BALANCE_XMIT_POLICY_LAYER23
2277 Active Slaves (3): [1 3 4]
2284 The Register Functions can be used to read from and write to registers on the network card referenced by a port number.
2285 This is mainly useful for debugging purposes.
2286 Reference should be made to the appropriate datasheet for the network card for details on the register addresses
2287 and fields that can be accessed.
2292 Display the value of a port register::
2294 testpmd> read reg (port_id) (address)
2296 For example, to examine the Flow Director control register (FDIRCTL, 0x0000EE000) on an Intel 82599 10 GbE Controller::
2298 testpmd> read reg 0 0xEE00
2299 port 0 PCI register at offset 0xEE00: 0x4A060029 (1241907241)
2304 Display a port register bit field::
2306 testpmd> read regfield (port_id) (address) (bit_x) (bit_y)
2308 For example, reading the lowest two bits from the register in the example above::
2310 testpmd> read regfield 0 0xEE00 0 1
2311 port 0 PCI register at offset 0xEE00: bits[0, 1]=0x1 (1)
2316 Display a single port register bit::
2318 testpmd> read regbit (port_id) (address) (bit_x)
2320 For example, reading the lowest bit from the register in the example above::
2322 testpmd> read regbit 0 0xEE00 0
2323 port 0 PCI register at offset 0xEE00: bit 0=1
2328 Set the value of a port register::
2330 testpmd> write reg (port_id) (address) (value)
2332 For example, to clear a register::
2334 testpmd> write reg 0 0xEE00 0x0
2335 port 0 PCI register at offset 0xEE00: 0x00000000 (0)
2340 Set bit field of a port register::
2342 testpmd> write regfield (port_id) (address) (bit_x) (bit_y) (value)
2344 For example, writing to the register cleared in the example above::
2346 testpmd> write regfield 0 0xEE00 0 1 2
2347 port 0 PCI register at offset 0xEE00: 0x00000002 (2)
2352 Set single bit value of a port register::
2354 testpmd> write regbit (port_id) (address) (bit_x) (value)
2356 For example, to set the high bit in the register from the example above::
2358 testpmd> write regbit 0 0xEE00 31 1
2359 port 0 PCI register at offset 0xEE00: 0x8000000A (2147483658)
2361 Traffic Metering and Policing
2362 -----------------------------
2364 The following section shows functions for configuring traffic metering and
2365 policing on the ethernet device through the use of generic ethdev API.
2367 show port traffic management capability
2368 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2370 Show traffic metering and policing capability of the port::
2372 testpmd> show port meter cap (port_id)
2374 add port meter profile (srTCM rfc2967)
2375 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2377 Add meter profile (srTCM rfc2697) to the ethernet device::
2379 testpmd> add port meter profile srtcm_rfc2697 (port_id) (profile_id) \
2384 * ``profile_id``: ID for the meter profile.
2385 * ``cir``: Committed Information Rate (CIR) (bytes/second).
2386 * ``cbs``: Committed Burst Size (CBS) (bytes).
2387 * ``ebs``: Excess Burst Size (EBS) (bytes).
2389 add port meter profile (trTCM rfc2968)
2390 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2392 Add meter profile (srTCM rfc2698) to the ethernet device::
2394 testpmd> add port meter profile trtcm_rfc2698 (port_id) (profile_id) \
2395 (cir) (pir) (cbs) (pbs)
2399 * ``profile_id``: ID for the meter profile.
2400 * ``cir``: Committed information rate (bytes/second).
2401 * ``pir``: Peak information rate (bytes/second).
2402 * ``cbs``: Committed burst size (bytes).
2403 * ``pbs``: Peak burst size (bytes).
2405 add port meter profile (trTCM rfc4115)
2406 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2408 Add meter profile (trTCM rfc4115) to the ethernet device::
2410 testpmd> add port meter profile trtcm_rfc4115 (port_id) (profile_id) \
2411 (cir) (eir) (cbs) (ebs)
2415 * ``profile_id``: ID for the meter profile.
2416 * ``cir``: Committed information rate (bytes/second).
2417 * ``eir``: Excess information rate (bytes/second).
2418 * ``cbs``: Committed burst size (bytes).
2419 * ``ebs``: Excess burst size (bytes).
2421 delete port meter profile
2422 ~~~~~~~~~~~~~~~~~~~~~~~~~
2424 Delete meter profile from the ethernet device::
2426 testpmd> del port meter profile (port_id) (profile_id)
2431 Create new meter object for the ethernet device::
2433 testpmd> create port meter (port_id) (mtr_id) (profile_id) \
2434 (meter_enable) (g_action) (y_action) (r_action) (stats_mask) (shared) \
2435 (use_pre_meter_color) [(dscp_tbl_entry0) (dscp_tbl_entry1)...\
2440 * ``mtr_id``: meter object ID.
2441 * ``profile_id``: ID for the meter profile.
2442 * ``meter_enable``: When this parameter has a non-zero value, the meter object
2443 gets enabled at the time of creation, otherwise remains disabled.
2444 * ``g_action``: Policer action for the packet with green color.
2445 * ``y_action``: Policer action for the packet with yellow color.
2446 * ``r_action``: Policer action for the packet with red color.
2447 * ``stats_mask``: Mask of statistics counter types to be enabled for the
2449 * ``shared``: When this parameter has a non-zero value, the meter object is
2450 shared by multiple flows. Otherwise, meter object is used by single flow.
2451 * ``use_pre_meter_color``: When this parameter has a non-zero value, the
2452 input color for the current meter object is determined by the latest meter
2453 object in the same flow. Otherwise, the current meter object uses the
2454 *dscp_table* to determine the input color.
2455 * ``dscp_tbl_entryx``: DSCP table entry x providing meter providing input
2456 color, 0 <= x <= 63.
2461 Enable meter for the ethernet device::
2463 testpmd> enable port meter (port_id) (mtr_id)
2468 Disable meter for the ethernet device::
2470 testpmd> disable port meter (port_id) (mtr_id)
2475 Delete meter for the ethernet device::
2477 testpmd> del port meter (port_id) (mtr_id)
2479 Set port meter profile
2480 ~~~~~~~~~~~~~~~~~~~~~~
2482 Set meter profile for the ethernet device::
2484 testpmd> set port meter profile (port_id) (mtr_id) (profile_id)
2486 set port meter dscp table
2487 ~~~~~~~~~~~~~~~~~~~~~~~~~
2489 Set meter dscp table for the ethernet device::
2491 testpmd> set port meter dscp table (port_id) (mtr_id) [(dscp_tbl_entry0) \
2492 (dscp_tbl_entry1)...(dscp_tbl_entry63)]
2494 set port meter policer action
2495 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2497 Set meter policer action for the ethernet device::
2499 testpmd> set port meter policer action (port_id) (mtr_id) (action_mask) \
2500 (action0) [(action1) (action1)]
2504 * ``action_mask``: Bit mask indicating which policer actions need to be
2505 updated. One or more policer actions can be updated in a single function
2506 invocation. To update the policer action associated with color C, bit
2507 (1 << C) needs to be set in *action_mask* and element at position C
2508 in the *actions* array needs to be valid.
2509 * ``actionx``: Policer action for the color x,
2510 RTE_MTR_GREEN <= x < RTE_MTR_COLORS
2512 set port meter stats mask
2513 ~~~~~~~~~~~~~~~~~~~~~~~~~
2515 Set meter stats mask for the ethernet device::
2517 testpmd> set port meter stats mask (port_id) (mtr_id) (stats_mask)
2521 * ``stats_mask``: Bit mask indicating statistics counter types to be enabled.
2523 show port meter stats
2524 ~~~~~~~~~~~~~~~~~~~~~
2526 Show meter stats of the ethernet device::
2528 testpmd> show port meter stats (port_id) (mtr_id) (clear)
2532 * ``clear``: Flag that indicates whether the statistics counters should
2533 be cleared (i.e. set to zero) immediately after they have been read or not.
2538 The following section shows functions for configuring traffic management on
2539 on the ethernet device through the use of generic TM API.
2541 show port traffic management capability
2542 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2544 Show traffic management capability of the port::
2546 testpmd> show port tm cap (port_id)
2548 show port traffic management capability (hierarchy level)
2549 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2551 Show traffic management hierarchy level capability of the port::
2553 testpmd> show port tm level cap (port_id) (level_id)
2555 show port traffic management capability (hierarchy node level)
2556 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2558 Show the traffic management hierarchy node capability of the port::
2560 testpmd> show port tm node cap (port_id) (node_id)
2562 show port traffic management hierarchy node type
2563 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2565 Show the port traffic management hierarchy node type::
2567 testpmd> show port tm node type (port_id) (node_id)
2569 show port traffic management hierarchy node stats
2570 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2572 Show the port traffic management hierarchy node statistics::
2574 testpmd> show port tm node stats (port_id) (node_id) (clear)
2578 * ``clear``: When this parameter has a non-zero value, the statistics counters
2579 are cleared (i.e. set to zero) immediately after they have been read,
2580 otherwise the statistics counters are left untouched.
2582 Add port traffic management private shaper profile
2583 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2585 Add the port traffic management private shaper profile::
2587 testpmd> add port tm node shaper profile (port_id) (shaper_profile_id) \
2588 (tb_rate) (tb_size) (packet_length_adjust)
2592 * ``shaper_profile id``: Shaper profile ID for the new profile.
2593 * ``tb_rate``: Token bucket rate (bytes per second).
2594 * ``tb_size``: Token bucket size (bytes).
2595 * ``packet_length_adjust``: The value (bytes) to be added to the length of
2596 each packet for the purpose of shaping. This parameter value can be used to
2597 correct the packet length with the framing overhead bytes that are consumed
2600 Delete port traffic management private shaper profile
2601 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2603 Delete the port traffic management private shaper::
2605 testpmd> del port tm node shaper profile (port_id) (shaper_profile_id)
2609 * ``shaper_profile id``: Shaper profile ID that needs to be deleted.
2611 Add port traffic management shared shaper
2612 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2614 Create the port traffic management shared shaper::
2616 testpmd> add port tm node shared shaper (port_id) (shared_shaper_id) \
2621 * ``shared_shaper_id``: Shared shaper ID to be created.
2622 * ``shaper_profile id``: Shaper profile ID for shared shaper.
2624 Set port traffic management shared shaper
2625 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2627 Update the port traffic management shared shaper::
2629 testpmd> set port tm node shared shaper (port_id) (shared_shaper_id) \
2634 * ``shared_shaper_id``: Shared shaper ID to be update.
2635 * ``shaper_profile id``: Shaper profile ID for shared shaper.
2637 Delete port traffic management shared shaper
2638 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2640 Delete the port traffic management shared shaper::
2642 testpmd> del port tm node shared shaper (port_id) (shared_shaper_id)
2646 * ``shared_shaper_id``: Shared shaper ID to be deleted.
2648 Set port traffic management hiearchy node private shaper
2649 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2651 set the port traffic management hierarchy node private shaper::
2653 testpmd> set port tm node shaper profile (port_id) (node_id) \
2658 * ``shaper_profile id``: Private shaper profile ID to be enabled on the
2661 Add port traffic management WRED profile
2662 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2664 Create a new WRED profile::
2666 testpmd> add port tm node wred profile (port_id) (wred_profile_id) \
2667 (color_g) (min_th_g) (max_th_g) (maxp_inv_g) (wq_log2_g) \
2668 (color_y) (min_th_y) (max_th_y) (maxp_inv_y) (wq_log2_y) \
2669 (color_r) (min_th_r) (max_th_r) (maxp_inv_r) (wq_log2_r)
2673 * ``wred_profile id``: Identifier for the newly create WRED profile
2674 * ``color_g``: Packet color (green)
2675 * ``min_th_g``: Minimum queue threshold for packet with green color
2676 * ``max_th_g``: Minimum queue threshold for packet with green color
2677 * ``maxp_inv_g``: Inverse of packet marking probability maximum value (maxp)
2678 * ``wq_log2_g``: Negated log2 of queue weight (wq)
2679 * ``color_y``: Packet color (yellow)
2680 * ``min_th_y``: Minimum queue threshold for packet with yellow color
2681 * ``max_th_y``: Minimum queue threshold for packet with yellow color
2682 * ``maxp_inv_y``: Inverse of packet marking probability maximum value (maxp)
2683 * ``wq_log2_y``: Negated log2 of queue weight (wq)
2684 * ``color_r``: Packet color (red)
2685 * ``min_th_r``: Minimum queue threshold for packet with yellow color
2686 * ``max_th_r``: Minimum queue threshold for packet with yellow color
2687 * ``maxp_inv_r``: Inverse of packet marking probability maximum value (maxp)
2688 * ``wq_log2_r``: Negated log2 of queue weight (wq)
2690 Delete port traffic management WRED profile
2691 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2693 Delete the WRED profile::
2695 testpmd> del port tm node wred profile (port_id) (wred_profile_id)
2697 Add port traffic management hierarchy nonleaf node
2698 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2700 Add nonleaf node to port traffic management hiearchy::
2702 testpmd> add port tm nonleaf node (port_id) (node_id) (parent_node_id) \
2703 (priority) (weight) (level_id) (shaper_profile_id) \
2704 (n_sp_priorities) (stats_mask) (n_shared_shapers) \
2705 [(shared_shaper_0) (shared_shaper_1) ...] \
2709 * ``parent_node_id``: Node ID of the parent.
2710 * ``priority``: Node priority (highest node priority is zero). This is used by
2711 the SP algorithm running on the parent node for scheduling this node.
2712 * ``weight``: Node weight (lowest weight is one). The node weight is relative
2713 to the weight sum of all siblings that have the same priority. It is used by
2714 the WFQ algorithm running on the parent node for scheduling this node.
2715 * ``level_id``: Hiearchy level of the node.
2716 * ``shaper_profile_id``: Shaper profile ID of the private shaper to be used by
2718 * ``n_sp_priorities``: Number of strict priorities.
2719 * ``stats_mask``: Mask of statistics counter types to be enabled for this node.
2720 * ``n_shared_shapers``: Number of shared shapers.
2721 * ``shared_shaper_id``: Shared shaper id.
2723 Add port traffic management hierarchy leaf node
2724 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2726 Add leaf node to port traffic management hiearchy::
2728 testpmd> add port tm leaf node (port_id) (node_id) (parent_node_id) \
2729 (priority) (weight) (level_id) (shaper_profile_id) \
2730 (cman_mode) (wred_profile_id) (stats_mask) (n_shared_shapers) \
2731 [(shared_shaper_id) (shared_shaper_id) ...] \
2735 * ``parent_node_id``: Node ID of the parent.
2736 * ``priority``: Node priority (highest node priority is zero). This is used by
2737 the SP algorithm running on the parent node for scheduling this node.
2738 * ``weight``: Node weight (lowest weight is one). The node weight is relative
2739 to the weight sum of all siblings that have the same priority. It is used by
2740 the WFQ algorithm running on the parent node for scheduling this node.
2741 * ``level_id``: Hiearchy level of the node.
2742 * ``shaper_profile_id``: Shaper profile ID of the private shaper to be used by
2744 * ``cman_mode``: Congestion management mode to be enabled for this node.
2745 * ``wred_profile_id``: WRED profile id to be enabled for this node.
2746 * ``stats_mask``: Mask of statistics counter types to be enabled for this node.
2747 * ``n_shared_shapers``: Number of shared shapers.
2748 * ``shared_shaper_id``: Shared shaper id.
2750 Delete port traffic management hierarchy node
2751 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2753 Delete node from port traffic management hiearchy::
2755 testpmd> del port tm node (port_id) (node_id)
2757 Update port traffic management hierarchy parent node
2758 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2760 Update port traffic management hierarchy parent node::
2762 testpmd> set port tm node parent (port_id) (node_id) (parent_node_id) \
2765 This function can only be called after the hierarchy commit invocation. Its
2766 success depends on the port support for this operation, as advertised through
2767 the port capability set. This function is valid for all nodes of the traffic
2768 management hierarchy except root node.
2770 Suspend port traffic management hierarchy node
2771 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2773 testpmd> suspend port tm node (port_id) (node_id)
2775 Resume port traffic management hierarchy node
2776 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2778 testpmd> resume port tm node (port_id) (node_id)
2780 Commit port traffic management hierarchy
2781 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2783 Commit the traffic management hierarchy on the port::
2785 testpmd> port tm hierarchy commit (port_id) (clean_on_fail)
2789 * ``clean_on_fail``: When set to non-zero, hierarchy is cleared on function
2790 call failure. On the other hand, hierarchy is preserved when this parameter
2793 Set port traffic management mark VLAN dei
2794 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2796 Enables/Disables the traffic management marking on the port for VLAN packets::
2798 testpmd> set port tm mark vlan_dei <port_id> <green> <yellow> <red>
2802 * ``port_id``: The port which on which VLAN packets marked as ``green`` or
2803 ``yellow`` or ``red`` will have dei bit enabled
2805 * ``green`` enable 1, disable 0 marking for dei bit of VLAN packets marked as green
2807 * ``yellow`` enable 1, disable 0 marking for dei bit of VLAN packets marked as yellow
2809 * ``red`` enable 1, disable 0 marking for dei bit of VLAN packets marked as red
2811 Set port traffic management mark IP dscp
2812 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2814 Enables/Disables the traffic management marking on the port for IP dscp packets::
2816 testpmd> set port tm mark ip_dscp <port_id> <green> <yellow> <red>
2820 * ``port_id``: The port which on which IP packets marked as ``green`` or
2821 ``yellow`` or ``red`` will have IP dscp bits updated
2823 * ``green`` enable 1, disable 0 marking IP dscp to low drop precedence for green packets
2825 * ``yellow`` enable 1, disable 0 marking IP dscp to medium drop precedence for yellow packets
2827 * ``red`` enable 1, disable 0 marking IP dscp to high drop precedence for red packets
2829 Set port traffic management mark IP ecn
2830 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2832 Enables/Disables the traffic management marking on the port for IP ecn packets::
2834 testpmd> set port tm mark ip_ecn <port_id> <green> <yellow> <red>
2838 * ``port_id``: The port which on which IP packets marked as ``green`` or
2839 ``yellow`` or ``red`` will have IP ecn bits updated
2841 * ``green`` enable 1, disable 0 marking IP ecn for green marked packets with ecn of 2'b01 or 2'b10
2842 to ecn of 2'b11 when IP is caring TCP or SCTP
2844 * ``yellow`` enable 1, disable 0 marking IP ecn for yellow marked packets with ecn of 2'b01 or 2'b10
2845 to ecn of 2'b11 when IP is caring TCP or SCTP
2847 * ``red`` enable 1, disable 0 marking IP ecn for yellow marked packets with ecn of 2'b01 or 2'b10
2848 to ecn of 2'b11 when IP is caring TCP or SCTP
2850 Set port traffic management default hierarchy (softnic forwarding mode)
2851 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2853 set the traffic management default hierarchy on the port::
2855 testpmd> set port tm hierarchy default (port_id)
2860 This section details the available filter functions that are available.
2862 Note these functions interface the deprecated legacy filtering framework,
2863 superseded by *rte_flow*. See `Flow rules management`_.
2866 ~~~~~~~~~~~~~~~~~~~~
2868 Add or delete a L2 Ethertype filter, which identify packets by their L2 Ethertype mainly assign them to a receive queue::
2870 ethertype_filter (port_id) (add|del) (mac_addr|mac_ignr) (mac_address) \
2871 ethertype (ether_type) (drop|fwd) queue (queue_id)
2873 The available information parameters are:
2875 * ``port_id``: The port which the Ethertype filter assigned on.
2877 * ``mac_addr``: Compare destination mac address.
2879 * ``mac_ignr``: Ignore destination mac address match.
2881 * ``mac_address``: Destination mac address to match.
2883 * ``ether_type``: The EtherType value want to match,
2884 for example 0x0806 for ARP packet. 0x0800 (IPv4) and 0x86DD (IPv6) are invalid.
2886 * ``queue_id``: The receive queue associated with this EtherType filter.
2887 It is meaningless when deleting or dropping.
2889 Example, to add/remove an ethertype filter rule::
2891 testpmd> ethertype_filter 0 add mac_ignr 00:11:22:33:44:55 \
2892 ethertype 0x0806 fwd queue 3
2894 testpmd> ethertype_filter 0 del mac_ignr 00:11:22:33:44:55 \
2895 ethertype 0x0806 fwd queue 3
2900 Add or delete a 2-tuple filter,
2901 which identifies packets by specific protocol and destination TCP/UDP port
2902 and forwards packets into one of the receive queues::
2904 2tuple_filter (port_id) (add|del) dst_port (dst_port_value) \
2905 protocol (protocol_value) mask (mask_value) \
2906 tcp_flags (tcp_flags_value) priority (prio_value) \
2909 The available information parameters are:
2911 * ``port_id``: The port which the 2-tuple filter assigned on.
2913 * ``dst_port_value``: Destination port in L4.
2915 * ``protocol_value``: IP L4 protocol.
2917 * ``mask_value``: Participates in the match or not by bit for field above, 1b means participate.
2919 * ``tcp_flags_value``: TCP control bits. The non-zero value is invalid, when the pro_value is not set to 0x06 (TCP).
2921 * ``prio_value``: Priority of this filter.
2923 * ``queue_id``: The receive queue associated with this 2-tuple filter.
2925 Example, to add/remove an 2tuple filter rule::
2927 testpmd> 2tuple_filter 0 add dst_port 32 protocol 0x06 mask 0x03 \
2928 tcp_flags 0x02 priority 3 queue 3
2930 testpmd> 2tuple_filter 0 del dst_port 32 protocol 0x06 mask 0x03 \
2931 tcp_flags 0x02 priority 3 queue 3
2936 Add or delete a 5-tuple filter,
2937 which consists of a 5-tuple (protocol, source and destination IP addresses, source and destination TCP/UDP/SCTP port)
2938 and routes packets into one of the receive queues::
2940 5tuple_filter (port_id) (add|del) dst_ip (dst_address) src_ip \
2941 (src_address) dst_port (dst_port_value) \
2942 src_port (src_port_value) protocol (protocol_value) \
2943 mask (mask_value) tcp_flags (tcp_flags_value) \
2944 priority (prio_value) queue (queue_id)
2946 The available information parameters are:
2948 * ``port_id``: The port which the 5-tuple filter assigned on.
2950 * ``dst_address``: Destination IP address.
2952 * ``src_address``: Source IP address.
2954 * ``dst_port_value``: TCP/UDP destination port.
2956 * ``src_port_value``: TCP/UDP source port.
2958 * ``protocol_value``: L4 protocol.
2960 * ``mask_value``: Participates in the match or not by bit for field above, 1b means participate
2962 * ``tcp_flags_value``: TCP control bits. The non-zero value is invalid, when the protocol_value is not set to 0x06 (TCP).
2964 * ``prio_value``: The priority of this filter.
2966 * ``queue_id``: The receive queue associated with this 5-tuple filter.
2968 Example, to add/remove an 5tuple filter rule::
2970 testpmd> 5tuple_filter 0 add dst_ip 2.2.2.5 src_ip 2.2.2.4 \
2971 dst_port 64 src_port 32 protocol 0x06 mask 0x1F \
2972 flags 0x0 priority 3 queue 3
2974 testpmd> 5tuple_filter 0 del dst_ip 2.2.2.5 src_ip 2.2.2.4 \
2975 dst_port 64 src_port 32 protocol 0x06 mask 0x1F \
2976 flags 0x0 priority 3 queue 3
2981 Using the SYN filter, TCP packets whose *SYN* flag is set can be forwarded to a separate queue::
2983 syn_filter (port_id) (add|del) priority (high|low) queue (queue_id)
2985 The available information parameters are:
2987 * ``port_id``: The port which the SYN filter assigned on.
2989 * ``high``: This SYN filter has higher priority than other filters.
2991 * ``low``: This SYN filter has lower priority than other filters.
2993 * ``queue_id``: The receive queue associated with this SYN filter
2997 testpmd> syn_filter 0 add priority high queue 3
3002 With flex filter, packets can be recognized by any arbitrary pattern within the first 128 bytes of the packet
3003 and routed into one of the receive queues::
3005 flex_filter (port_id) (add|del) len (len_value) bytes (bytes_value) \
3006 mask (mask_value) priority (prio_value) queue (queue_id)
3008 The available information parameters are:
3010 * ``port_id``: The port which the Flex filter is assigned on.
3012 * ``len_value``: Filter length in bytes, no greater than 128.
3014 * ``bytes_value``: A string in hexadecimal, means the value the flex filter needs to match.
3016 * ``mask_value``: A string in hexadecimal, bit 1 means corresponding byte participates in the match.
3018 * ``prio_value``: The priority of this filter.
3020 * ``queue_id``: The receive queue associated with this Flex filter.
3024 testpmd> flex_filter 0 add len 16 bytes 0x00000000000000000000000008060000 \
3025 mask 000C priority 3 queue 3
3027 testpmd> flex_filter 0 del len 16 bytes 0x00000000000000000000000008060000 \
3028 mask 000C priority 3 queue 3
3031 .. _testpmd_flow_director:
3033 flow_director_filter
3034 ~~~~~~~~~~~~~~~~~~~~
3036 The Flow Director works in receive mode to identify specific flows or sets of flows and route them to specific queues.
3038 Four types of filtering are supported which are referred to as Perfect Match, Signature, Perfect-mac-vlan and
3039 Perfect-tunnel filters, the match mode is set by the ``--pkt-filter-mode`` command-line parameter:
3041 * Perfect match filters.
3042 The hardware checks a match between the masked fields of the received packets and the programmed filters.
3043 The masked fields are for IP flow.
3045 * Signature filters.
3046 The hardware checks a match between a hash-based signature of the masked fields of the received packet.
3048 * Perfect-mac-vlan match filters.
3049 The hardware checks a match between the masked fields of the received packets and the programmed filters.
3050 The masked fields are for MAC VLAN flow.
3052 * Perfect-tunnel match filters.
3053 The hardware checks a match between the masked fields of the received packets and the programmed filters.
3054 The masked fields are for tunnel flow.
3056 * Perfect-raw-flow-type match filters.
3057 The hardware checks a match between the masked fields of the received packets and pre-loaded raw (template) packet.
3058 The masked fields are specified by input sets.
3060 The Flow Director filters can match the different fields for different type of packet: flow type, specific input set
3061 per flow type and the flexible payload.
3063 The Flow Director can also mask out parts of all of these fields so that filters
3064 are only applied to certain fields or parts of the fields.
3066 Note that for raw flow type mode the source and destination fields in the
3067 raw packet buffer need to be presented in a reversed order with respect
3068 to the expected received packets.
3069 For example: IP source and destination addresses or TCP/UDP/SCTP
3070 source and destination ports
3072 Different NICs may have different capabilities, command show port fdir (port_id) can be used to acquire the information.
3074 # Commands to add flow director filters of different flow types::
3076 flow_director_filter (port_id) mode IP (add|del|update) \
3077 flow (ipv4-other|ipv4-frag|ipv6-other|ipv6-frag) \
3078 src (src_ip_address) dst (dst_ip_address) \
3079 tos (tos_value) proto (proto_value) ttl (ttl_value) \
3080 vlan (vlan_value) flexbytes (flexbytes_value) \
3081 (drop|fwd) pf|vf(vf_id) queue (queue_id) \
3084 flow_director_filter (port_id) mode IP (add|del|update) \
3085 flow (ipv4-tcp|ipv4-udp|ipv6-tcp|ipv6-udp) \
3086 src (src_ip_address) (src_port) \
3087 dst (dst_ip_address) (dst_port) \
3088 tos (tos_value) ttl (ttl_value) \
3089 vlan (vlan_value) flexbytes (flexbytes_value) \
3090 (drop|fwd) queue pf|vf(vf_id) (queue_id) \
3093 flow_director_filter (port_id) mode IP (add|del|update) \
3094 flow (ipv4-sctp|ipv6-sctp) \
3095 src (src_ip_address) (src_port) \
3096 dst (dst_ip_address) (dst_port) \
3097 tos (tos_value) ttl (ttl_value) \
3098 tag (verification_tag) vlan (vlan_value) \
3099 flexbytes (flexbytes_value) (drop|fwd) \
3100 pf|vf(vf_id) queue (queue_id) fd_id (fd_id_value)
3102 flow_director_filter (port_id) mode IP (add|del|update) flow l2_payload \
3103 ether (ethertype) flexbytes (flexbytes_value) \
3104 (drop|fwd) pf|vf(vf_id) queue (queue_id)
3107 flow_director_filter (port_id) mode MAC-VLAN (add|del|update) \
3108 mac (mac_address) vlan (vlan_value) \
3109 flexbytes (flexbytes_value) (drop|fwd) \
3110 queue (queue_id) fd_id (fd_id_value)
3112 flow_director_filter (port_id) mode Tunnel (add|del|update) \
3113 mac (mac_address) vlan (vlan_value) \
3114 tunnel (NVGRE|VxLAN) tunnel-id (tunnel_id_value) \
3115 flexbytes (flexbytes_value) (drop|fwd) \
3116 queue (queue_id) fd_id (fd_id_value)
3118 flow_director_filter (port_id) mode raw (add|del|update) flow (flow_id) \
3119 (drop|fwd) queue (queue_id) fd_id (fd_id_value) \
3120 packet (packet file name)
3122 For example, to add an ipv4-udp flow type filter::
3124 testpmd> flow_director_filter 0 mode IP add flow ipv4-udp src 2.2.2.3 32 \
3125 dst 2.2.2.5 33 tos 2 ttl 40 vlan 0x1 flexbytes (0x88,0x48) \
3126 fwd pf queue 1 fd_id 1
3128 For example, add an ipv4-other flow type filter::
3130 testpmd> flow_director_filter 0 mode IP add flow ipv4-other src 2.2.2.3 \
3131 dst 2.2.2.5 tos 2 proto 20 ttl 40 vlan 0x1 \
3132 flexbytes (0x88,0x48) fwd pf queue 1 fd_id 1
3137 Flush all flow director filters on a device::
3139 testpmd> flush_flow_director (port_id)
3141 Example, to flush all flow director filter on port 0::
3143 testpmd> flush_flow_director 0
3148 Set flow director's input masks::
3150 flow_director_mask (port_id) mode IP vlan (vlan_value) \
3151 src_mask (ipv4_src) (ipv6_src) (src_port) \
3152 dst_mask (ipv4_dst) (ipv6_dst) (dst_port)
3154 flow_director_mask (port_id) mode MAC-VLAN vlan (vlan_value)
3156 flow_director_mask (port_id) mode Tunnel vlan (vlan_value) \
3157 mac (mac_value) tunnel-type (tunnel_type_value) \
3158 tunnel-id (tunnel_id_value)
3160 Example, to set flow director mask on port 0::
3162 testpmd> flow_director_mask 0 mode IP vlan 0xefff \
3163 src_mask 255.255.255.255 \
3164 FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF \
3165 dst_mask 255.255.255.255 \
3166 FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF
3168 flow_director_flex_mask
3169 ~~~~~~~~~~~~~~~~~~~~~~~
3171 set masks of flow director's flexible payload based on certain flow type::
3173 testpmd> flow_director_flex_mask (port_id) \
3174 flow (none|ipv4-other|ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
3175 ipv6-other|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp| \
3176 l2_payload|all) (mask)
3178 Example, to set flow director's flex mask for all flow type on port 0::
3180 testpmd> flow_director_flex_mask 0 flow all \
3181 (0xff,0xff,0,0,0,0,0,0,0,0,0,0,0,0,0,0)
3184 flow_director_flex_payload
3185 ~~~~~~~~~~~~~~~~~~~~~~~~~~
3187 Configure flexible payload selection::
3189 flow_director_flex_payload (port_id) (raw|l2|l3|l4) (config)
3191 For example, to select the first 16 bytes from the offset 4 (bytes) of packet's payload as flexible payload::
3193 testpmd> flow_director_flex_payload 0 l4 \
3194 (4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19)
3196 get_sym_hash_ena_per_port
3197 ~~~~~~~~~~~~~~~~~~~~~~~~~
3199 Get symmetric hash enable configuration per port::
3201 get_sym_hash_ena_per_port (port_id)
3203 For example, to get symmetric hash enable configuration of port 1::
3205 testpmd> get_sym_hash_ena_per_port 1
3207 set_sym_hash_ena_per_port
3208 ~~~~~~~~~~~~~~~~~~~~~~~~~
3210 Set symmetric hash enable configuration per port to enable or disable::
3212 set_sym_hash_ena_per_port (port_id) (enable|disable)
3214 For example, to set symmetric hash enable configuration of port 1 to enable::
3216 testpmd> set_sym_hash_ena_per_port 1 enable
3218 get_hash_global_config
3219 ~~~~~~~~~~~~~~~~~~~~~~
3221 Get the global configurations of hash filters::
3223 get_hash_global_config (port_id)
3225 For example, to get the global configurations of hash filters of port 1::
3227 testpmd> get_hash_global_config 1
3229 set_hash_global_config
3230 ~~~~~~~~~~~~~~~~~~~~~~
3232 Set the global configurations of hash filters::
3234 set_hash_global_config (port_id) (toeplitz|simple_xor|default) \
3235 (ipv4|ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp|ipv4-other|ipv6|ipv6-frag| \
3236 ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other|l2_payload|<flow_id>) \
3239 For example, to enable simple_xor for flow type of ipv6 on port 2::
3241 testpmd> set_hash_global_config 2 simple_xor ipv6 enable
3246 Set the input set for hash::
3248 set_hash_input_set (port_id) (ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
3249 ipv4-other|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other| \
3250 l2_payload|<flow_id>) (ovlan|ivlan|src-ipv4|dst-ipv4|src-ipv6|dst-ipv6| \
3251 ipv4-tos|ipv4-proto|ipv6-tc|ipv6-next-header|udp-src-port|udp-dst-port| \
3252 tcp-src-port|tcp-dst-port|sctp-src-port|sctp-dst-port|sctp-veri-tag| \
3253 udp-key|gre-key|fld-1st|fld-2nd|fld-3rd|fld-4th|fld-5th|fld-6th|fld-7th| \
3254 fld-8th|none) (select|add)
3256 For example, to add source IP to hash input set for flow type of ipv4-udp on port 0::
3258 testpmd> set_hash_input_set 0 ipv4-udp src-ipv4 add
3263 The Flow Director filters can match the different fields for different type of packet, i.e. specific input set
3264 on per flow type and the flexible payload. This command can be used to change input set for each flow type.
3266 Set the input set for flow director::
3268 set_fdir_input_set (port_id) (ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
3269 ipv4-other|ipv6|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other| \
3270 l2_payload|<flow_id>) (ivlan|ethertype|src-ipv4|dst-ipv4|src-ipv6|dst-ipv6| \
3271 ipv4-tos|ipv4-proto|ipv4-ttl|ipv6-tc|ipv6-next-header|ipv6-hop-limits| \
3272 tudp-src-port|udp-dst-port|cp-src-port|tcp-dst-port|sctp-src-port| \
3273 sctp-dst-port|sctp-veri-tag|none) (select|add)
3275 For example to add source IP to FD input set for flow type of ipv4-udp on port 0::
3277 testpmd> set_fdir_input_set 0 ipv4-udp src-ipv4 add
3282 Set different GRE key length for input set::
3284 global_config (port_id) gre-key-len (number in bytes)
3286 For example to set GRE key length for input set to 4 bytes on port 0::
3288 testpmd> global_config 0 gre-key-len 4
3291 .. _testpmd_rte_flow:
3293 Flow rules management
3294 ---------------------
3296 Control of the generic flow API (*rte_flow*) is fully exposed through the
3297 ``flow`` command (validation, creation, destruction, queries and operation
3300 Considering *rte_flow* overlaps with all `Filter Functions`_, using both
3301 features simultaneously may cause undefined side-effects and is therefore
3307 Because the ``flow`` command uses dynamic tokens to handle the large number
3308 of possible flow rules combinations, its behavior differs slightly from
3309 other commands, in particular:
3311 - Pressing *?* or the *<tab>* key displays contextual help for the current
3312 token, not that of the entire command.
3314 - Optional and repeated parameters are supported (provided they are listed
3315 in the contextual help).
3317 The first parameter stands for the operation mode. Possible operations and
3318 their general syntax are described below. They are covered in detail in the
3321 - Check whether a flow rule can be created::
3323 flow validate {port_id}
3324 [group {group_id}] [priority {level}] [ingress] [egress] [transfer]
3325 pattern {item} [/ {item} [...]] / end
3326 actions {action} [/ {action} [...]] / end
3328 - Create a flow rule::
3330 flow create {port_id}
3331 [group {group_id}] [priority {level}] [ingress] [egress] [transfer]
3332 pattern {item} [/ {item} [...]] / end
3333 actions {action} [/ {action} [...]] / end
3335 - Destroy specific flow rules::
3337 flow destroy {port_id} rule {rule_id} [...]
3339 - Destroy all flow rules::
3341 flow flush {port_id}
3343 - Query an existing flow rule::
3345 flow query {port_id} {rule_id} {action}
3347 - List existing flow rules sorted by priority, filtered by group
3350 flow list {port_id} [group {group_id}] [...]
3352 - Restrict ingress traffic to the defined flow rules::
3354 flow isolate {port_id} {boolean}
3356 Validating flow rules
3357 ~~~~~~~~~~~~~~~~~~~~~
3359 ``flow validate`` reports whether a flow rule would be accepted by the
3360 underlying device in its current state but stops short of creating it. It is
3361 bound to ``rte_flow_validate()``::
3363 flow validate {port_id}
3364 [group {group_id}] [priority {level}] [ingress] [egress] [transfer]
3365 pattern {item} [/ {item} [...]] / end
3366 actions {action} [/ {action} [...]] / end
3368 If successful, it will show::
3372 Otherwise it will show an error message of the form::
3374 Caught error type [...] ([...]): [...]
3376 This command uses the same parameters as ``flow create``, their format is
3377 described in `Creating flow rules`_.
3379 Check whether redirecting any Ethernet packet received on port 0 to RX queue
3380 index 6 is supported::
3382 testpmd> flow validate 0 ingress pattern eth / end
3383 actions queue index 6 / end
3387 Port 0 does not support TCPv6 rules::
3389 testpmd> flow validate 0 ingress pattern eth / ipv6 / tcp / end
3391 Caught error type 9 (specific pattern item): Invalid argument
3397 ``flow create`` validates and creates the specified flow rule. It is bound
3398 to ``rte_flow_create()``::
3400 flow create {port_id}
3401 [group {group_id}] [priority {level}] [ingress] [egress] [transfer]
3402 pattern {item} [/ {item} [...]] / end
3403 actions {action} [/ {action} [...]] / end
3405 If successful, it will return a flow rule ID usable with other commands::
3407 Flow rule #[...] created
3409 Otherwise it will show an error message of the form::
3411 Caught error type [...] ([...]): [...]
3413 Parameters describe in the following order:
3415 - Attributes (*group*, *priority*, *ingress*, *egress*, *transfer* tokens).
3416 - A matching pattern, starting with the *pattern* token and terminated by an
3418 - Actions, starting with the *actions* token and terminated by an *end*
3421 These translate directly to *rte_flow* objects provided as-is to the
3422 underlying functions.
3424 The shortest valid definition only comprises mandatory tokens::
3426 testpmd> flow create 0 pattern end actions end
3428 Note that PMDs may refuse rules that essentially do nothing such as this
3431 **All unspecified object values are automatically initialized to 0.**
3436 These tokens affect flow rule attributes (``struct rte_flow_attr``) and are
3437 specified before the ``pattern`` token.
3439 - ``group {group id}``: priority group.
3440 - ``priority {level}``: priority level within group.
3441 - ``ingress``: rule applies to ingress traffic.
3442 - ``egress``: rule applies to egress traffic.
3443 - ``transfer``: apply rule directly to endpoints found in pattern.
3445 Each instance of an attribute specified several times overrides the previous
3446 value as shown below (group 4 is used)::
3448 testpmd> flow create 0 group 42 group 24 group 4 [...]
3450 Note that once enabled, ``ingress`` and ``egress`` cannot be disabled.
3452 While not specifying a direction is an error, some rules may allow both
3455 Most rules affect RX therefore contain the ``ingress`` token::
3457 testpmd> flow create 0 ingress pattern [...]
3462 A matching pattern starts after the ``pattern`` token. It is made of pattern
3463 items and is terminated by a mandatory ``end`` item.
3465 Items are named after their type (*RTE_FLOW_ITEM_TYPE_* from ``enum
3466 rte_flow_item_type``).
3468 The ``/`` token is used as a separator between pattern items as shown
3471 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end [...]
3473 Note that protocol items like these must be stacked from lowest to highest
3474 layer to make sense. For instance, the following rule is either invalid or
3475 unlikely to match any packet::
3477 testpmd> flow create 0 ingress pattern eth / udp / ipv4 / end [...]
3479 More information on these restrictions can be found in the *rte_flow*
3482 Several items support additional specification structures, for example
3483 ``ipv4`` allows specifying source and destination addresses as follows::
3485 testpmd> flow create 0 ingress pattern eth / ipv4 src is 10.1.1.1
3486 dst is 10.2.0.0 / end [...]
3488 This rule matches all IPv4 traffic with the specified properties.
3490 In this example, ``src`` and ``dst`` are field names of the underlying
3491 ``struct rte_flow_item_ipv4`` object. All item properties can be specified
3492 in a similar fashion.
3494 The ``is`` token means that the subsequent value must be matched exactly,
3495 and assigns ``spec`` and ``mask`` fields in ``struct rte_flow_item``
3496 accordingly. Possible assignment tokens are:
3498 - ``is``: match value perfectly (with full bit-mask).
3499 - ``spec``: match value according to configured bit-mask.
3500 - ``last``: specify upper bound to establish a range.
3501 - ``mask``: specify bit-mask with relevant bits set to one.
3502 - ``prefix``: generate bit-mask from a prefix length.
3504 These yield identical results::
3506 ipv4 src is 10.1.1.1
3510 ipv4 src spec 10.1.1.1 src mask 255.255.255.255
3514 ipv4 src spec 10.1.1.1 src prefix 32
3518 ipv4 src is 10.1.1.1 src last 10.1.1.1 # range with a single value
3522 ipv4 src is 10.1.1.1 src last 0 # 0 disables range
3524 Inclusive ranges can be defined with ``last``::
3526 ipv4 src is 10.1.1.1 src last 10.2.3.4 # 10.1.1.1 to 10.2.3.4
3528 Note that ``mask`` affects both ``spec`` and ``last``::
3530 ipv4 src is 10.1.1.1 src last 10.2.3.4 src mask 255.255.0.0
3531 # matches 10.1.0.0 to 10.2.255.255
3533 Properties can be modified multiple times::
3535 ipv4 src is 10.1.1.1 src is 10.1.2.3 src is 10.2.3.4 # matches 10.2.3.4
3539 ipv4 src is 10.1.1.1 src prefix 24 src prefix 16 # matches 10.1.0.0/16
3544 This section lists supported pattern items and their attributes, if any.
3546 - ``end``: end list of pattern items.
3548 - ``void``: no-op pattern item.
3550 - ``invert``: perform actions when pattern does not match.
3552 - ``any``: match any protocol for the current layer.
3554 - ``num {unsigned}``: number of layers covered.
3556 - ``pf``: match traffic from/to the physical function.
3558 - ``vf``: match traffic from/to a virtual function ID.
3560 - ``id {unsigned}``: VF ID.
3562 - ``phy_port``: match traffic from/to a specific physical port.
3564 - ``index {unsigned}``: physical port index.
3566 - ``port_id``: match traffic from/to a given DPDK port ID.
3568 - ``id {unsigned}``: DPDK port ID.
3570 - ``mark``: match value set in previously matched flow rule using the mark action.
3572 - ``id {unsigned}``: arbitrary integer value.
3574 - ``raw``: match an arbitrary byte string.
3576 - ``relative {boolean}``: look for pattern after the previous item.
3577 - ``search {boolean}``: search pattern from offset (see also limit).
3578 - ``offset {integer}``: absolute or relative offset for pattern.
3579 - ``limit {unsigned}``: search area limit for start of pattern.
3580 - ``pattern {string}``: byte string to look for.
3582 - ``eth``: match Ethernet header.
3584 - ``dst {MAC-48}``: destination MAC.
3585 - ``src {MAC-48}``: source MAC.
3586 - ``type {unsigned}``: EtherType or TPID.
3588 - ``vlan``: match 802.1Q/ad VLAN tag.
3590 - ``tci {unsigned}``: tag control information.
3591 - ``pcp {unsigned}``: priority code point.
3592 - ``dei {unsigned}``: drop eligible indicator.
3593 - ``vid {unsigned}``: VLAN identifier.
3594 - ``inner_type {unsigned}``: inner EtherType or TPID.
3596 - ``ipv4``: match IPv4 header.
3598 - ``tos {unsigned}``: type of service.
3599 - ``ttl {unsigned}``: time to live.
3600 - ``proto {unsigned}``: next protocol ID.
3601 - ``src {ipv4 address}``: source address.
3602 - ``dst {ipv4 address}``: destination address.
3604 - ``ipv6``: match IPv6 header.
3606 - ``tc {unsigned}``: traffic class.
3607 - ``flow {unsigned}``: flow label.
3608 - ``proto {unsigned}``: protocol (next header).
3609 - ``hop {unsigned}``: hop limit.
3610 - ``src {ipv6 address}``: source address.
3611 - ``dst {ipv6 address}``: destination address.
3613 - ``icmp``: match ICMP header.
3615 - ``type {unsigned}``: ICMP packet type.
3616 - ``code {unsigned}``: ICMP packet code.
3618 - ``udp``: match UDP header.
3620 - ``src {unsigned}``: UDP source port.
3621 - ``dst {unsigned}``: UDP destination port.
3623 - ``tcp``: match TCP header.
3625 - ``src {unsigned}``: TCP source port.
3626 - ``dst {unsigned}``: TCP destination port.
3628 - ``sctp``: match SCTP header.
3630 - ``src {unsigned}``: SCTP source port.
3631 - ``dst {unsigned}``: SCTP destination port.
3632 - ``tag {unsigned}``: validation tag.
3633 - ``cksum {unsigned}``: checksum.
3635 - ``vxlan``: match VXLAN header.
3637 - ``vni {unsigned}``: VXLAN identifier.
3639 - ``e_tag``: match IEEE 802.1BR E-Tag header.
3641 - ``grp_ecid_b {unsigned}``: GRP and E-CID base.
3643 - ``nvgre``: match NVGRE header.
3645 - ``tni {unsigned}``: virtual subnet ID.
3647 - ``mpls``: match MPLS header.
3649 - ``label {unsigned}``: MPLS label.
3651 - ``gre``: match GRE header.
3653 - ``protocol {unsigned}``: protocol type.
3655 - ``fuzzy``: fuzzy pattern match, expect faster than default.
3657 - ``thresh {unsigned}``: accuracy threshold.
3659 - ``gtp``, ``gtpc``, ``gtpu``: match GTPv1 header.
3661 - ``teid {unsigned}``: tunnel endpoint identifier.
3663 - ``geneve``: match GENEVE header.
3665 - ``vni {unsigned}``: virtual network identifier.
3666 - ``protocol {unsigned}``: protocol type.
3668 - ``vxlan-gpe``: match VXLAN-GPE header.
3670 - ``vni {unsigned}``: VXLAN-GPE identifier.
3672 - ``arp_eth_ipv4``: match ARP header for Ethernet/IPv4.
3674 - ``sha {MAC-48}``: sender hardware address.
3675 - ``spa {ipv4 address}``: sender IPv4 address.
3676 - ``tha {MAC-48}``: target hardware address.
3677 - ``tpa {ipv4 address}``: target IPv4 address.
3679 - ``ipv6_ext``: match presence of any IPv6 extension header.
3681 - ``next_hdr {unsigned}``: next header.
3683 - ``icmp6``: match any ICMPv6 header.
3685 - ``type {unsigned}``: ICMPv6 type.
3686 - ``code {unsigned}``: ICMPv6 code.
3688 - ``icmp6_nd_ns``: match ICMPv6 neighbor discovery solicitation.
3690 - ``target_addr {ipv6 address}``: target address.
3692 - ``icmp6_nd_na``: match ICMPv6 neighbor discovery advertisement.
3694 - ``target_addr {ipv6 address}``: target address.
3696 - ``icmp6_nd_opt``: match presence of any ICMPv6 neighbor discovery option.
3698 - ``type {unsigned}``: ND option type.
3700 - ``icmp6_nd_opt_sla_eth``: match ICMPv6 neighbor discovery source Ethernet
3701 link-layer address option.
3703 - ``sla {MAC-48}``: source Ethernet LLA.
3705 - ``icmp6_nd_opt_sla_eth``: match ICMPv6 neighbor discovery target Ethernet
3706 link-layer address option.
3708 - ``tla {MAC-48}``: target Ethernet LLA.
3710 - ``meta``: match application specific metadata.
3712 - ``data {unsigned}``: metadata value.
3717 A list of actions starts after the ``actions`` token in the same fashion as
3718 `Matching pattern`_; actions are separated by ``/`` tokens and the list is
3719 terminated by a mandatory ``end`` action.
3721 Actions are named after their type (*RTE_FLOW_ACTION_TYPE_* from ``enum
3722 rte_flow_action_type``).
3724 Dropping all incoming UDPv4 packets can be expressed as follows::
3726 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
3729 Several actions have configurable properties which must be specified when
3730 there is no valid default value. For example, ``queue`` requires a target
3733 This rule redirects incoming UDPv4 traffic to queue index 6::
3735 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
3736 actions queue index 6 / end
3738 While this one could be rejected by PMDs (unspecified queue index)::
3740 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
3743 As defined by *rte_flow*, the list is not ordered, all actions of a given
3744 rule are performed simultaneously. These are equivalent::
3746 queue index 6 / void / mark id 42 / end
3750 void / mark id 42 / queue index 6 / end
3752 All actions in a list should have different types, otherwise only the last
3753 action of a given type is taken into account::
3755 queue index 4 / queue index 5 / queue index 6 / end # will use queue 6
3759 drop / drop / drop / end # drop is performed only once
3763 mark id 42 / queue index 3 / mark id 24 / end # mark will be 24
3765 Considering they are performed simultaneously, opposite and overlapping
3766 actions can sometimes be combined when the end result is unambiguous::
3768 drop / queue index 6 / end # drop has no effect
3772 queue index 6 / rss queues 6 7 8 / end # queue has no effect
3776 drop / passthru / end # drop has no effect
3778 Note that PMDs may still refuse such combinations.
3783 This section lists supported actions and their attributes, if any.
3785 - ``end``: end list of actions.
3787 - ``void``: no-op action.
3789 - ``passthru``: let subsequent rule process matched packets.
3791 - ``jump``: redirect traffic to group on device.
3793 - ``group {unsigned}``: group to redirect to.
3795 - ``mark``: attach 32 bit value to packets.
3797 - ``id {unsigned}``: 32 bit value to return with packets.
3799 - ``flag``: flag packets.
3801 - ``queue``: assign packets to a given queue index.
3803 - ``index {unsigned}``: queue index to use.
3805 - ``drop``: drop packets (note: passthru has priority).
3807 - ``count``: enable counters for this rule.
3809 - ``rss``: spread packets among several queues.
3811 - ``func {hash function}``: RSS hash function to apply, allowed tokens are
3812 the same as `set_hash_global_config`_.
3814 - ``level {unsigned}``: encapsulation level for ``types``.
3816 - ``types [{RSS hash type} [...]] end``: specific RSS hash types, allowed
3817 tokens are the same as `set_hash_input_set`_, except that an empty list
3818 does not disable RSS but instead requests unspecified "best-effort"
3821 - ``key {string}``: RSS hash key, overrides ``key_len``.
3823 - ``key_len {unsigned}``: RSS hash key length in bytes, can be used in
3824 conjunction with ``key`` to pad or truncate it.
3826 - ``queues [{unsigned} [...]] end``: queue indices to use.
3828 - ``pf``: direct traffic to physical function.
3830 - ``vf``: direct traffic to a virtual function ID.
3832 - ``original {boolean}``: use original VF ID if possible.
3833 - ``id {unsigned}``: VF ID.
3835 - ``phy_port``: direct packets to physical port index.
3837 - ``original {boolean}``: use original port index if possible.
3838 - ``index {unsigned}``: physical port index.
3840 - ``port_id``: direct matching traffic to a given DPDK port ID.
3842 - ``original {boolean}``: use original DPDK port ID if possible.
3843 - ``id {unsigned}``: DPDK port ID.
3845 - ``of_set_mpls_ttl``: OpenFlow's ``OFPAT_SET_MPLS_TTL``.
3847 - ``mpls_ttl``: MPLS TTL.
3849 - ``of_dec_mpls_ttl``: OpenFlow's ``OFPAT_DEC_MPLS_TTL``.
3851 - ``of_set_nw_ttl``: OpenFlow's ``OFPAT_SET_NW_TTL``.
3853 - ``nw_ttl``: IP TTL.
3855 - ``of_dec_nw_ttl``: OpenFlow's ``OFPAT_DEC_NW_TTL``.
3857 - ``of_copy_ttl_out``: OpenFlow's ``OFPAT_COPY_TTL_OUT``.
3859 - ``of_copy_ttl_in``: OpenFlow's ``OFPAT_COPY_TTL_IN``.
3861 - ``of_pop_vlan``: OpenFlow's ``OFPAT_POP_VLAN``.
3863 - ``of_push_vlan``: OpenFlow's ``OFPAT_PUSH_VLAN``.
3865 - ``ethertype``: Ethertype.
3867 - ``of_set_vlan_vid``: OpenFlow's ``OFPAT_SET_VLAN_VID``.
3869 - ``vlan_vid``: VLAN id.
3871 - ``of_set_vlan_pcp``: OpenFlow's ``OFPAT_SET_VLAN_PCP``.
3873 - ``vlan_pcp``: VLAN priority.
3875 - ``of_pop_mpls``: OpenFlow's ``OFPAT_POP_MPLS``.
3877 - ``ethertype``: Ethertype.
3879 - ``of_push_mpls``: OpenFlow's ``OFPAT_PUSH_MPLS``.
3881 - ``ethertype``: Ethertype.
3883 - ``vxlan_encap``: Performs a VXLAN encapsulation, outer layer configuration
3884 is done through `Config VXLAN Encap outer layers`_.
3886 - ``vxlan_decap``: Performs a decapsulation action by stripping all headers of
3887 the VXLAN tunnel network overlay from the matched flow.
3889 - ``nvgre_encap``: Performs a NVGRE encapsulation, outer layer configuration
3890 is done through `Config NVGRE Encap outer layers`_.
3892 - ``nvgre_decap``: Performs a decapsulation action by stripping all headers of
3893 the NVGRE tunnel network overlay from the matched flow.
3895 - ``l2_encap``: Performs a L2 encapsulation, L2 configuration
3896 is done through `Config L2 Encap`_.
3898 - ``l2_decap``: Performs a L2 decapsulation, L2 configuration
3899 is done through `Config L2 Decap`_.
3901 - ``mplsogre_encap``: Performs a MPLSoGRE encapsulation, outer layer
3902 configuration is done through `Config MPLSoGRE Encap outer layers`_.
3904 - ``mplsogre_decap``: Performs a MPLSoGRE decapsulation, outer layer
3905 configuration is done through `Config MPLSoGRE Decap outer layers`_.
3907 - ``mplsoudp_encap``: Performs a MPLSoUDP encapsulation, outer layer
3908 configuration is done through `Config MPLSoUDP Encap outer layers`_.
3910 - ``mplsoudp_decap``: Performs a MPLSoUDP decapsulation, outer layer
3911 configuration is done through `Config MPLSoUDP Decap outer layers`_.
3913 - ``set_ipv4_src``: Set a new IPv4 source address in the outermost IPv4 header.
3915 - ``ipv4_addr``: New IPv4 source address.
3917 - ``set_ipv4_dst``: Set a new IPv4 destination address in the outermost IPv4
3920 - ``ipv4_addr``: New IPv4 destination address.
3922 - ``set_ipv6_src``: Set a new IPv6 source address in the outermost IPv6 header.
3924 - ``ipv6_addr``: New IPv6 source address.
3926 - ``set_ipv6_dst``: Set a new IPv6 destination address in the outermost IPv6
3929 - ``ipv6_addr``: New IPv6 destination address.
3931 - ``of_set_tp_src``: Set a new source port number in the outermost TCP/UDP
3934 - ``port``: New TCP/UDP source port number.
3936 - ``of_set_tp_dst``: Set a new destination port number in the outermost TCP/UDP
3939 - ``port``: New TCP/UDP destination port number.
3941 - ``mac_swap``: Swap the source and destination MAC addresses in the outermost
3944 - ``dec_ttl``: Performs a decrease TTL value action
3946 - ``set_ttl``: Set TTL value with specificed value
3947 - ``ttl_value {unsigned}``: The new TTL value to be set
3949 - ``set_mac_src``: set source MAC address
3951 - ``mac_addr {MAC-48}``: new source MAC address
3953 - ``set_mac_dst``: set destination MAC address
3955 - ``mac_addr {MAC-48}``: new destination MAC address
3957 Destroying flow rules
3958 ~~~~~~~~~~~~~~~~~~~~~
3960 ``flow destroy`` destroys one or more rules from their rule ID (as returned
3961 by ``flow create``), this command calls ``rte_flow_destroy()`` as many
3962 times as necessary::
3964 flow destroy {port_id} rule {rule_id} [...]
3966 If successful, it will show::
3968 Flow rule #[...] destroyed
3970 It does not report anything for rule IDs that do not exist. The usual error
3971 message is shown when a rule cannot be destroyed::
3973 Caught error type [...] ([...]): [...]
3975 ``flow flush`` destroys all rules on a device and does not take extra
3976 arguments. It is bound to ``rte_flow_flush()``::
3978 flow flush {port_id}
3980 Any errors are reported as above.
3982 Creating several rules and destroying them::
3984 testpmd> flow create 0 ingress pattern eth / ipv6 / end
3985 actions queue index 2 / end
3986 Flow rule #0 created
3987 testpmd> flow create 0 ingress pattern eth / ipv4 / end
3988 actions queue index 3 / end
3989 Flow rule #1 created
3990 testpmd> flow destroy 0 rule 0 rule 1
3991 Flow rule #1 destroyed
3992 Flow rule #0 destroyed
3995 The same result can be achieved using ``flow flush``::
3997 testpmd> flow create 0 ingress pattern eth / ipv6 / end
3998 actions queue index 2 / end
3999 Flow rule #0 created
4000 testpmd> flow create 0 ingress pattern eth / ipv4 / end
4001 actions queue index 3 / end
4002 Flow rule #1 created
4003 testpmd> flow flush 0
4006 Non-existent rule IDs are ignored::
4008 testpmd> flow create 0 ingress pattern eth / ipv6 / end
4009 actions queue index 2 / end
4010 Flow rule #0 created
4011 testpmd> flow create 0 ingress pattern eth / ipv4 / end
4012 actions queue index 3 / end
4013 Flow rule #1 created
4014 testpmd> flow destroy 0 rule 42 rule 10 rule 2
4016 testpmd> flow destroy 0 rule 0
4017 Flow rule #0 destroyed
4023 ``flow query`` queries a specific action of a flow rule having that
4024 ability. Such actions collect information that can be reported using this
4025 command. It is bound to ``rte_flow_query()``::
4027 flow query {port_id} {rule_id} {action}
4029 If successful, it will display either the retrieved data for known actions
4030 or the following message::
4032 Cannot display result for action type [...] ([...])
4034 Otherwise, it will complain either that the rule does not exist or that some
4037 Flow rule #[...] not found
4041 Caught error type [...] ([...]): [...]
4043 Currently only the ``count`` action is supported. This action reports the
4044 number of packets that hit the flow rule and the total number of bytes. Its
4045 output has the following format::
4048 hits_set: [...] # whether "hits" contains a valid value
4049 bytes_set: [...] # whether "bytes" contains a valid value
4050 hits: [...] # number of packets
4051 bytes: [...] # number of bytes
4053 Querying counters for TCPv6 packets redirected to queue 6::
4055 testpmd> flow create 0 ingress pattern eth / ipv6 / tcp / end
4056 actions queue index 6 / count / end
4057 Flow rule #4 created
4058 testpmd> flow query 0 4 count
4069 ``flow list`` lists existing flow rules sorted by priority and optionally
4070 filtered by group identifiers::
4072 flow list {port_id} [group {group_id}] [...]
4074 This command only fails with the following message if the device does not
4079 Output consists of a header line followed by a short description of each
4080 flow rule, one per line. There is no output at all when no flow rules are
4081 configured on the device::
4083 ID Group Prio Attr Rule
4084 [...] [...] [...] [...] [...]
4086 ``Attr`` column flags:
4088 - ``i`` for ``ingress``.
4089 - ``e`` for ``egress``.
4091 Creating several flow rules and listing them::
4093 testpmd> flow create 0 ingress pattern eth / ipv4 / end
4094 actions queue index 6 / end
4095 Flow rule #0 created
4096 testpmd> flow create 0 ingress pattern eth / ipv6 / end
4097 actions queue index 2 / end
4098 Flow rule #1 created
4099 testpmd> flow create 0 priority 5 ingress pattern eth / ipv4 / udp / end
4100 actions rss queues 6 7 8 end / end
4101 Flow rule #2 created
4102 testpmd> flow list 0
4103 ID Group Prio Attr Rule
4104 0 0 0 i- ETH IPV4 => QUEUE
4105 1 0 0 i- ETH IPV6 => QUEUE
4106 2 0 5 i- ETH IPV4 UDP => RSS
4109 Rules are sorted by priority (i.e. group ID first, then priority level)::
4111 testpmd> flow list 1
4112 ID Group Prio Attr Rule
4113 0 0 0 i- ETH => COUNT
4114 6 0 500 i- ETH IPV6 TCP => DROP COUNT
4115 5 0 1000 i- ETH IPV6 ICMP => QUEUE
4116 1 24 0 i- ETH IPV4 UDP => QUEUE
4117 4 24 10 i- ETH IPV4 TCP => DROP
4118 3 24 20 i- ETH IPV4 => DROP
4119 2 24 42 i- ETH IPV4 UDP => QUEUE
4120 7 63 0 i- ETH IPV6 UDP VXLAN => MARK QUEUE
4123 Output can be limited to specific groups::
4125 testpmd> flow list 1 group 0 group 63
4126 ID Group Prio Attr Rule
4127 0 0 0 i- ETH => COUNT
4128 6 0 500 i- ETH IPV6 TCP => DROP COUNT
4129 5 0 1000 i- ETH IPV6 ICMP => QUEUE
4130 7 63 0 i- ETH IPV6 UDP VXLAN => MARK QUEUE
4133 Toggling isolated mode
4134 ~~~~~~~~~~~~~~~~~~~~~~
4136 ``flow isolate`` can be used to tell the underlying PMD that ingress traffic
4137 must only be injected from the defined flow rules; that no default traffic
4138 is expected outside those rules and the driver is free to assign more
4139 resources to handle them. It is bound to ``rte_flow_isolate()``::
4141 flow isolate {port_id} {boolean}
4143 If successful, enabling or disabling isolated mode shows either::
4145 Ingress traffic on port [...]
4146 is now restricted to the defined flow rules
4150 Ingress traffic on port [...]
4151 is not restricted anymore to the defined flow rules
4153 Otherwise, in case of error::
4155 Caught error type [...] ([...]): [...]
4157 Mainly due to its side effects, PMDs supporting this mode may not have the
4158 ability to toggle it more than once without reinitializing affected ports
4159 first (e.g. by exiting testpmd).
4161 Enabling isolated mode::
4163 testpmd> flow isolate 0 true
4164 Ingress traffic on port 0 is now restricted to the defined flow rules
4167 Disabling isolated mode::
4169 testpmd> flow isolate 0 false
4170 Ingress traffic on port 0 is not restricted anymore to the defined flow rules
4173 Sample QinQ flow rules
4174 ~~~~~~~~~~~~~~~~~~~~~~
4176 Before creating QinQ rule(s) the following commands should be issued to enable QinQ::
4178 testpmd> port stop 0
4179 testpmd> vlan set qinq on 0
4181 The above command sets the inner and outer TPID's to 0x8100.
4183 To change the TPID's the following commands should be used::
4185 testpmd> vlan set outer tpid 0xa100 0
4186 testpmd> vlan set inner tpid 0x9100 0
4187 testpmd> port start 0
4189 Validate and create a QinQ rule on port 0 to steer traffic to a VF queue in a VM.
4193 testpmd> flow validate 0 ingress pattern eth / vlan tci is 123 /
4194 vlan tci is 456 / end actions vf id 1 / queue index 0 / end
4195 Flow rule #0 validated
4197 testpmd> flow create 0 ingress pattern eth / vlan tci is 4 /
4198 vlan tci is 456 / end actions vf id 123 / queue index 0 / end
4199 Flow rule #0 created
4201 testpmd> flow list 0
4202 ID Group Prio Attr Rule
4203 0 0 0 i- ETH VLAN VLAN=>VF QUEUE
4205 Validate and create a QinQ rule on port 0 to steer traffic to a queue on the host.
4209 testpmd> flow validate 0 ingress pattern eth / vlan tci is 321 /
4210 vlan tci is 654 / end actions pf / queue index 0 / end
4211 Flow rule #1 validated
4213 testpmd> flow create 0 ingress pattern eth / vlan tci is 321 /
4214 vlan tci is 654 / end actions pf / queue index 1 / end
4215 Flow rule #1 created
4217 testpmd> flow list 0
4218 ID Group Prio Attr Rule
4219 0 0 0 i- ETH VLAN VLAN=>VF QUEUE
4220 1 0 0 i- ETH VLAN VLAN=>PF QUEUE
4222 Sample VXLAN encapsulation rule
4223 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4225 VXLAN encapsulation outer layer has default value pre-configured in testpmd
4226 source code, those can be changed by using the following commands
4228 IPv4 VXLAN outer header::
4230 testpmd> set vxlan ip-version ipv4 vni 4 udp-src 4 udp-dst 4 ip-src 127.0.0.1
4231 ip-dst 128.0.0.1 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4232 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4235 testpmd> set vxlan-with-vlan ip-version ipv4 vni 4 udp-src 4 udp-dst 4 ip-src
4236 127.0.0.1 ip-dst 128.0.0.1 vlan-tci 34 eth-src 11:11:11:11:11:11
4237 eth-dst 22:22:22:22:22:22
4238 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4241 IPv6 VXLAN outer header::
4243 testpmd> set vxlan ip-version ipv6 vni 4 udp-src 4 udp-dst 4 ip-src ::1
4244 ip-dst ::2222 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4245 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4248 testpmd> set vxlan-with-vlan ip-version ipv6 vni 4 udp-src 4 udp-dst 4
4249 ip-src ::1 ip-dst ::2222 vlan-tci 34 eth-src 11:11:11:11:11:11
4250 eth-dst 22:22:22:22:22:22
4251 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4254 Sample NVGRE encapsulation rule
4255 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4257 NVGRE encapsulation outer layer has default value pre-configured in testpmd
4258 source code, those can be changed by using the following commands
4260 IPv4 NVGRE outer header::
4262 testpmd> set nvgre ip-version ipv4 tni 4 ip-src 127.0.0.1 ip-dst 128.0.0.1
4263 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4264 testpmd> flow create 0 ingress pattern end actions nvgre_encap /
4267 testpmd> set nvgre-with-vlan ip-version ipv4 tni 4 ip-src 127.0.0.1
4268 ip-dst 128.0.0.1 vlan-tci 34 eth-src 11:11:11:11:11:11
4269 eth-dst 22:22:22:22:22:22
4270 testpmd> flow create 0 ingress pattern end actions nvgre_encap /
4273 IPv6 NVGRE outer header::
4275 testpmd> set nvgre ip-version ipv6 tni 4 ip-src ::1 ip-dst ::2222
4276 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4277 testpmd> flow create 0 ingress pattern end actions nvgre_encap /
4280 testpmd> set nvgre-with-vlan ip-version ipv6 tni 4 ip-src ::1 ip-dst ::2222
4281 vlan-tci 34 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4282 testpmd> flow create 0 ingress pattern end actions nvgre_encap /
4285 Sample L2 encapsulation rule
4286 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4288 L2 encapsulation has default value pre-configured in testpmd
4289 source code, those can be changed by using the following commands
4293 testpmd> set l2_encap ip-version ipv4
4294 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4295 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / mpls / end actions
4296 mplsoudp_decap / l2_encap / end
4298 L2 with VXLAN header::
4300 testpmd> set l2_encap-with-vlan ip-version ipv4 vlan-tci 34
4301 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4302 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / mpls / end actions
4303 mplsoudp_decap / l2_encap / end
4305 Sample L2 decapsulation rule
4306 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4308 L2 decapsulation has default value pre-configured in testpmd
4309 source code, those can be changed by using the following commands
4313 testpmd> set l2_decap
4314 testpmd> flow create 0 egress pattern eth / end actions l2_decap / mplsoudp_encap /
4317 L2 with VXLAN header::
4319 testpmd> set l2_encap-with-vlan
4320 testpmd> flow create 0 egress pattern eth / end actions l2_encap / mplsoudp_encap /
4323 Sample MPLSoGRE encapsulation rule
4324 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4326 MPLSoGRE encapsulation outer layer has default value pre-configured in testpmd
4327 source code, those can be changed by using the following commands
4329 IPv4 MPLSoGRE outer header::
4331 testpmd> set mplsogre_encap ip-version ipv4 label 4
4332 ip-src 127.0.0.1 ip-dst 128.0.0.1 eth-src 11:11:11:11:11:11
4333 eth-dst 22:22:22:22:22:22
4334 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4335 mplsogre_encap / end
4337 IPv4 MPLSoGRE with VLAN outer header::
4339 testpmd> set mplsogre_encap-with-vlan ip-version ipv4 label 4
4340 ip-src 127.0.0.1 ip-dst 128.0.0.1 vlan-tci 34
4341 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4342 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4343 mplsogre_encap / end
4345 IPv6 MPLSoGRE outer header::
4347 testpmd> set mplsogre_encap ip-version ipv6 mask 4
4348 ip-src ::1 ip-dst ::2222 eth-src 11:11:11:11:11:11
4349 eth-dst 22:22:22:22:22:22
4350 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4351 mplsogre_encap / end
4353 IPv6 MPLSoGRE with VLAN outer header::
4355 testpmd> set mplsogre_encap-with-vlan ip-version ipv6 mask 4
4356 ip-src ::1 ip-dst ::2222 vlan-tci 34
4357 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4358 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4359 mplsogre_encap / end
4361 Sample MPLSoGRE decapsulation rule
4362 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4364 MPLSoGRE decapsulation outer layer has default value pre-configured in testpmd
4365 source code, those can be changed by using the following commands
4367 IPv4 MPLSoGRE outer header::
4369 testpmd> set mplsogre_decap ip-version ipv4
4370 testpmd> flow create 0 ingress pattern eth / ipv4 / gre / mpls / end actions
4371 mplsogre_decap / l2_encap / end
4373 IPv4 MPLSoGRE with VLAN outer header::
4375 testpmd> set mplsogre_decap-with-vlan ip-version ipv4
4376 testpmd> flow create 0 ingress pattern eth / vlan / ipv4 / gre / mpls / end
4377 actions mplsogre_decap / l2_encap / end
4379 IPv6 MPLSoGRE outer header::
4381 testpmd> set mplsogre_decap ip-version ipv6
4382 testpmd> flow create 0 ingress pattern eth / ipv6 / gre / mpls / end
4383 actions mplsogre_decap / l2_encap / end
4385 IPv6 MPLSoGRE with VLAN outer header::
4387 testpmd> set mplsogre_decap-with-vlan ip-version ipv6
4388 testpmd> flow create 0 ingress pattern eth / vlan / ipv6 / gre / mpls / end
4389 actions mplsogre_decap / l2_encap / end
4391 Sample MPLSoUDP encapsulation rule
4392 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4394 MPLSoUDP encapsulation outer layer has default value pre-configured in testpmd
4395 source code, those can be changed by using the following commands
4397 IPv4 MPLSoUDP outer header::
4399 testpmd> set mplsoudp_encap ip-version ipv4 label 4 udp-src 5 udp-dst 10
4400 ip-src 127.0.0.1 ip-dst 128.0.0.1 eth-src 11:11:11:11:11:11
4401 eth-dst 22:22:22:22:22:22
4402 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4403 mplsoudp_encap / end
4405 IPv4 MPLSoUDP with VLAN outer header::
4407 testpmd> set mplsoudp_encap-with-vlan ip-version ipv4 label 4 udp-src 5
4408 udp-dst 10 ip-src 127.0.0.1 ip-dst 128.0.0.1 vlan-tci 34
4409 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4410 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4411 mplsoudp_encap / end
4413 IPv6 MPLSoUDP outer header::
4415 testpmd> set mplsoudp_encap ip-version ipv6 mask 4 udp-src 5 udp-dst 10
4416 ip-src ::1 ip-dst ::2222 eth-src 11:11:11:11:11:11
4417 eth-dst 22:22:22:22:22:22
4418 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4419 mplsoudp_encap / end
4421 IPv6 MPLSoUDP with VLAN outer header::
4423 testpmd> set mplsoudp_encap-with-vlan ip-version ipv6 mask 4 udp-src 5
4424 udp-dst 10 ip-src ::1 ip-dst ::2222 vlan-tci 34
4425 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4426 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4427 mplsoudp_encap / end
4429 Sample MPLSoUDP decapsulation rule
4430 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4432 MPLSoUDP decapsulation outer layer has default value pre-configured in testpmd
4433 source code, those can be changed by using the following commands
4435 IPv4 MPLSoUDP outer header::
4437 testpmd> set mplsoudp_decap ip-version ipv4
4438 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / mpls / end actions
4439 mplsoudp_decap / l2_encap / end
4441 IPv4 MPLSoUDP with VLAN outer header::
4443 testpmd> set mplsoudp_decap-with-vlan ip-version ipv4
4444 testpmd> flow create 0 ingress pattern eth / vlan / ipv4 / udp / mpls / end
4445 actions mplsoudp_decap / l2_encap / end
4447 IPv6 MPLSoUDP outer header::
4449 testpmd> set mplsoudp_decap ip-version ipv6
4450 testpmd> flow create 0 ingress pattern eth / ipv6 / udp / mpls / end
4451 actions mplsoudp_decap / l2_encap / end
4453 IPv6 MPLSoUDP with VLAN outer header::
4455 testpmd> set mplsoudp_decap-with-vlan ip-version ipv6
4456 testpmd> flow create 0 ingress pattern eth / vlan / ipv6 / udp / mpls / end
4457 actions mplsoudp_decap / l2_encap / end
4462 The following sections show functions to load/unload eBPF based filters.
4467 Load an eBPF program as a callback for partciular RX/TX queue::
4469 testpmd> bpf-load rx|tx (portid) (queueid) (load-flags) (bpf-prog-filename)
4471 The available load-flags are:
4473 * ``J``: use JIT generated native code, otherwise BPF interpreter will be used.
4475 * ``M``: assume input parameter is a pointer to rte_mbuf, otherwise assume it is a pointer to first segment's data.
4481 You'll need clang v3.7 or above to build bpf program you'd like to load
4485 .. code-block:: console
4488 clang -O2 -target bpf -c t1.c
4490 Then to load (and JIT compile) t1.o at RX queue 0, port 1::
4492 .. code-block:: console
4494 testpmd> bpf-load rx 1 0 J ./dpdk.org/test/bpf/t1.o
4496 To load (not JITed) t1.o at TX queue 0, port 0::
4498 .. code-block:: console
4500 testpmd> bpf-load tx 0 0 - ./dpdk.org/test/bpf/t1.o
4505 Unload previously loaded eBPF program for partciular RX/TX queue::
4507 testpmd> bpf-unload rx|tx (portid) (queueid)
4509 For example to unload BPF filter from TX queue 0, port 0:
4511 .. code-block:: console
4513 testpmd> bpf-load tx 0 0 - ./dpdk.org/test/bpf/t1.o