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-peers-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 ICMP 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) (peer_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 set vxlan-tos-ttl ip-version (ipv4|ipv6) vni (vni) udp-src (udp-src) \
1587 udp-dst (udp-dst) ip-tos (ip-tos) ip-ttl (ip-ttl) ip-src (ip-src) \
1588 ip-dst (ip-dst) eth-src (eth-src) eth-dst (eth-dst)
1590 Those command will set an internal configuration inside testpmd, any following
1591 flow rule using the action vxlan_encap will use the last configuration set.
1592 To have a different encapsulation header, one of those commands must be called
1593 before the flow rule creation.
1595 Config NVGRE Encap outer layers
1596 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1598 Configure the outer layer to encapsulate a packet inside a NVGRE tunnel::
1600 set nvgre ip-version (ipv4|ipv6) tni (tni) ip-src (ip-src) ip-dst (ip-dst) \
1601 eth-src (eth-src) eth-dst (eth-dst)
1602 set nvgre-with-vlan ip-version (ipv4|ipv6) tni (tni) ip-src (ip-src) \
1603 ip-dst (ip-dst) vlan-tci (vlan-tci) eth-src (eth-src) eth-dst (eth-dst)
1605 Those command will set an internal configuration inside testpmd, any following
1606 flow rule using the action nvgre_encap will use the last configuration set.
1607 To have a different encapsulation header, one of those commands must be called
1608 before the flow rule creation.
1613 Configure the l2 to be used when encapsulating a packet with L2::
1615 set l2_encap ip-version (ipv4|ipv6) eth-src (eth-src) eth-dst (eth-dst)
1616 set l2_encap-with-vlan ip-version (ipv4|ipv6) vlan-tci (vlan-tci) \
1617 eth-src (eth-src) eth-dst (eth-dst)
1619 Those commands will set an internal configuration inside testpmd, any following
1620 flow rule using the action l2_encap will use the last configuration set.
1621 To have a different encapsulation header, one of those commands must be called
1622 before the flow rule creation.
1627 Configure the l2 to be removed when decapsulating a packet with L2::
1629 set l2_decap ip-version (ipv4|ipv6)
1630 set l2_decap-with-vlan ip-version (ipv4|ipv6)
1632 Those commands will set an internal configuration inside testpmd, any following
1633 flow rule using the action l2_decap will use the last configuration set.
1634 To have a different encapsulation header, one of those commands must be called
1635 before the flow rule creation.
1637 Config MPLSoGRE Encap outer layers
1638 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1640 Configure the outer layer to encapsulate a packet inside a MPLSoGRE tunnel::
1642 set mplsogre_encap ip-version (ipv4|ipv6) label (label) \
1643 ip-src (ip-src) ip-dst (ip-dst) eth-src (eth-src) eth-dst (eth-dst)
1644 set mplsogre_encap-with-vlan ip-version (ipv4|ipv6) label (label) \
1645 ip-src (ip-src) ip-dst (ip-dst) vlan-tci (vlan-tci) \
1646 eth-src (eth-src) eth-dst (eth-dst)
1648 Those command will set an internal configuration inside testpmd, any following
1649 flow rule using the action mplsogre_encap will use the last configuration set.
1650 To have a different encapsulation header, one of those commands must be called
1651 before the flow rule creation.
1653 Config MPLSoGRE Decap outer layers
1654 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1656 Configure the outer layer to decapsulate MPLSoGRE packet::
1658 set mplsogre_decap ip-version (ipv4|ipv6)
1659 set mplsogre_decap-with-vlan ip-version (ipv4|ipv6)
1661 Those command will set an internal configuration inside testpmd, any following
1662 flow rule using the action mplsogre_decap will use the last configuration set.
1663 To have a different decapsulation header, one of those commands must be called
1664 before the flow rule creation.
1666 Config MPLSoUDP Encap outer layers
1667 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1669 Configure the outer layer to encapsulate a packet inside a MPLSoUDP tunnel::
1671 set mplsoudp_encap ip-version (ipv4|ipv6) label (label) udp-src (udp-src) \
1672 udp-dst (udp-dst) ip-src (ip-src) ip-dst (ip-dst) \
1673 eth-src (eth-src) eth-dst (eth-dst)
1674 set mplsoudp_encap-with-vlan ip-version (ipv4|ipv6) label (label) \
1675 udp-src (udp-src) udp-dst (udp-dst) ip-src (ip-src) ip-dst (ip-dst) \
1676 vlan-tci (vlan-tci) eth-src (eth-src) eth-dst (eth-dst)
1678 Those command will set an internal configuration inside testpmd, any following
1679 flow rule using the action mplsoudp_encap will use the last configuration set.
1680 To have a different encapsulation header, one of those commands must be called
1681 before the flow rule creation.
1683 Config MPLSoUDP Decap outer layers
1684 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1686 Configure the outer layer to decapsulate MPLSoUDP packet::
1688 set mplsoudp_decap ip-version (ipv4|ipv6)
1689 set mplsoudp_decap-with-vlan ip-version (ipv4|ipv6)
1691 Those command will set an internal configuration inside testpmd, any following
1692 flow rule using the action mplsoudp_decap will use the last configuration set.
1693 To have a different decapsulation header, one of those commands must be called
1694 before the flow rule creation.
1699 The following sections show functions for configuring ports.
1703 Port configuration changes only become active when forwarding is started/restarted.
1708 Attach a port specified by pci address or virtual device args::
1710 testpmd> port attach (identifier)
1712 To attach a new pci device, the device should be recognized by kernel first.
1713 Then it should be moved under DPDK management.
1714 Finally the port can be attached to testpmd.
1716 For example, to move a pci device using ixgbe under DPDK management:
1718 .. code-block:: console
1720 # Check the status of the available devices.
1721 ./usertools/dpdk-devbind.py --status
1723 Network devices using DPDK-compatible driver
1724 ============================================
1727 Network devices using kernel driver
1728 ===================================
1729 0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused=
1732 # Bind the device to igb_uio.
1733 sudo ./usertools/dpdk-devbind.py -b igb_uio 0000:0a:00.0
1736 # Recheck the status of the devices.
1737 ./usertools/dpdk-devbind.py --status
1738 Network devices using DPDK-compatible driver
1739 ============================================
1740 0000:0a:00.0 '82599ES 10-Gigabit' drv=igb_uio unused=
1742 To attach a port created by virtual device, above steps are not needed.
1744 For example, to attach a port whose pci address is 0000:0a:00.0.
1746 .. code-block:: console
1748 testpmd> port attach 0000:0a:00.0
1749 Attaching a new port...
1750 EAL: PCI device 0000:0a:00.0 on NUMA socket -1
1751 EAL: probe driver: 8086:10fb rte_ixgbe_pmd
1752 EAL: PCI memory mapped at 0x7f83bfa00000
1753 EAL: PCI memory mapped at 0x7f83bfa80000
1754 PMD: eth_ixgbe_dev_init(): MAC: 2, PHY: 18, SFP+: 5
1755 PMD: eth_ixgbe_dev_init(): port 0 vendorID=0x8086 deviceID=0x10fb
1756 Port 0 is attached. Now total ports is 1
1759 For example, to attach a port created by pcap PMD.
1761 .. code-block:: console
1763 testpmd> port attach net_pcap0
1764 Attaching a new port...
1765 PMD: Initializing pmd_pcap for net_pcap0
1766 PMD: Creating pcap-backed ethdev on numa socket 0
1767 Port 0 is attached. Now total ports is 1
1770 In this case, identifier is ``net_pcap0``.
1771 This identifier format is the same as ``--vdev`` format of DPDK applications.
1773 For example, to re-attach a bonded port which has been previously detached,
1774 the mode and slave parameters must be given.
1776 .. code-block:: console
1778 testpmd> port attach net_bond_0,mode=0,slave=1
1779 Attaching a new port...
1780 EAL: Initializing pmd_bond for net_bond_0
1781 EAL: Create bonded device net_bond_0 on port 0 in mode 0 on socket 0.
1782 Port 0 is attached. Now total ports is 1
1789 Detach a specific port::
1791 testpmd> port detach (port_id)
1793 Before detaching a port, the port should be stopped and closed.
1795 For example, to detach a pci device port 0.
1797 .. code-block:: console
1799 testpmd> port stop 0
1802 testpmd> port close 0
1806 testpmd> port detach 0
1808 EAL: PCI device 0000:0a:00.0 on NUMA socket -1
1809 EAL: remove driver: 8086:10fb rte_ixgbe_pmd
1810 EAL: PCI memory unmapped at 0x7f83bfa00000
1811 EAL: PCI memory unmapped at 0x7f83bfa80000
1815 For example, to detach a virtual device port 0.
1817 .. code-block:: console
1819 testpmd> port stop 0
1822 testpmd> port close 0
1826 testpmd> port detach 0
1828 PMD: Closing pcap ethdev on numa socket 0
1829 Port 'net_pcap0' is detached. Now total ports is 0
1832 To remove a pci device completely from the system, first detach the port from testpmd.
1833 Then the device should be moved under kernel management.
1834 Finally the device can be removed using kernel pci hotplug functionality.
1836 For example, to move a pci device under kernel management:
1838 .. code-block:: console
1840 sudo ./usertools/dpdk-devbind.py -b ixgbe 0000:0a:00.0
1842 ./usertools/dpdk-devbind.py --status
1844 Network devices using DPDK-compatible driver
1845 ============================================
1848 Network devices using kernel driver
1849 ===================================
1850 0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused=igb_uio
1852 To remove a port created by a virtual device, above steps are not needed.
1857 Start all ports or a specific port::
1859 testpmd> port start (port_id|all)
1864 Stop all ports or a specific port::
1866 testpmd> port stop (port_id|all)
1871 Close all ports or a specific port::
1873 testpmd> port close (port_id|all)
1875 port config - queue ring size
1876 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1878 Configure a rx/tx queue ring size::
1880 testpmd> port (port_id) (rxq|txq) (queue_id) ring_size (value)
1882 Only take effect after command that (re-)start the port or command that setup specific queue.
1884 port start/stop queue
1885 ~~~~~~~~~~~~~~~~~~~~~
1887 Start/stop a rx/tx queue on a specific port::
1889 testpmd> port (port_id) (rxq|txq) (queue_id) (start|stop)
1891 port config - queue deferred start
1892 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1894 Switch on/off deferred start of a specific port queue::
1896 testpmd> port (port_id) (rxq|txq) (queue_id) deferred_start (on|off)
1899 ~~~~~~~~~~~~~~~~~~~~~
1901 Setup a rx/tx queue on a specific port::
1903 testpmd> port (port_id) (rxq|txq) (queue_id) setup
1905 Only take effect when port is started.
1910 Set the speed and duplex mode for all ports or a specific port::
1912 testpmd> port config (port_id|all) speed (10|100|1000|10000|25000|40000|50000|100000|auto) \
1913 duplex (half|full|auto)
1915 port config - queues/descriptors
1916 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1918 Set number of queues/descriptors for rxq, txq, rxd and txd::
1920 testpmd> port config all (rxq|txq|rxd|txd) (value)
1922 This is equivalent to the ``--rxq``, ``--txq``, ``--rxd`` and ``--txd`` command-line options.
1924 port config - max-pkt-len
1925 ~~~~~~~~~~~~~~~~~~~~~~~~~
1927 Set the maximum packet length::
1929 testpmd> port config all max-pkt-len (value)
1931 This is equivalent to the ``--max-pkt-len`` command-line option.
1933 port config - CRC Strip
1934 ~~~~~~~~~~~~~~~~~~~~~~~
1936 Set hardware CRC stripping on or off for all ports::
1938 testpmd> port config all crc-strip (on|off)
1940 CRC stripping is on by default.
1942 The ``off`` option is equivalent to the ``--disable-crc-strip`` command-line option.
1944 port config - scatter
1945 ~~~~~~~~~~~~~~~~~~~~~~~
1947 Set RX scatter mode on or off for all ports::
1949 testpmd> port config all scatter (on|off)
1951 RX scatter mode is off by default.
1953 The ``on`` option is equivalent to the ``--enable-scatter`` command-line option.
1955 port config - RX Checksum
1956 ~~~~~~~~~~~~~~~~~~~~~~~~~
1958 Set hardware RX checksum offload to on or off for all ports::
1960 testpmd> port config all rx-cksum (on|off)
1962 Checksum offload is off by default.
1964 The ``on`` option is equivalent to the ``--enable-rx-cksum`` command-line option.
1969 Set hardware VLAN on or off for all ports::
1971 testpmd> port config all hw-vlan (on|off)
1973 Hardware VLAN is off by default.
1975 The ``on`` option is equivalent to the ``--enable-hw-vlan`` command-line option.
1977 port config - VLAN filter
1978 ~~~~~~~~~~~~~~~~~~~~~~~~~
1980 Set hardware VLAN filter on or off for all ports::
1982 testpmd> port config all hw-vlan-filter (on|off)
1984 Hardware VLAN filter is off by default.
1986 The ``on`` option is equivalent to the ``--enable-hw-vlan-filter`` command-line option.
1988 port config - VLAN strip
1989 ~~~~~~~~~~~~~~~~~~~~~~~~
1991 Set hardware VLAN strip on or off for all ports::
1993 testpmd> port config all hw-vlan-strip (on|off)
1995 Hardware VLAN strip is off by default.
1997 The ``on`` option is equivalent to the ``--enable-hw-vlan-strip`` command-line option.
1999 port config - VLAN extend
2000 ~~~~~~~~~~~~~~~~~~~~~~~~~
2002 Set hardware VLAN extend on or off for all ports::
2004 testpmd> port config all hw-vlan-extend (on|off)
2006 Hardware VLAN extend is off by default.
2008 The ``on`` option is equivalent to the ``--enable-hw-vlan-extend`` command-line option.
2010 port config - Drop Packets
2011 ~~~~~~~~~~~~~~~~~~~~~~~~~~
2013 Set packet drop for packets with no descriptors on or off for all ports::
2015 testpmd> port config all drop-en (on|off)
2017 Packet dropping for packets with no descriptors is off by default.
2019 The ``on`` option is equivalent to the ``--enable-drop-en`` command-line option.
2024 Set the RSS (Receive Side Scaling) mode on or off::
2026 testpmd> port config all rss (all|default|ip|tcp|udp|sctp|ether|port|vxlan|geneve|nvgre|none)
2028 RSS is on by default.
2030 The ``all`` option is equivalent to ip|tcp|udp|sctp|ether.
2031 The ``default`` option enables all supported RSS types reported by device info.
2032 The ``none`` option is equivalent to the ``--disable-rss`` command-line option.
2034 port config - RSS Reta
2035 ~~~~~~~~~~~~~~~~~~~~~~
2037 Set the RSS (Receive Side Scaling) redirection table::
2039 testpmd> port config all rss reta (hash,queue)[,(hash,queue)]
2044 Set the DCB mode for an individual port::
2046 testpmd> port config (port_id) dcb vt (on|off) (traffic_class) pfc (on|off)
2048 The traffic class should be 4 or 8.
2053 Set the number of packets per burst::
2055 testpmd> port config all burst (value)
2057 This is equivalent to the ``--burst`` command-line option.
2059 port config - Threshold
2060 ~~~~~~~~~~~~~~~~~~~~~~~
2062 Set thresholds for TX/RX queues::
2064 testpmd> port config all (threshold) (value)
2066 Where the threshold type can be:
2068 * ``txpt:`` Set the prefetch threshold register of the TX rings, 0 <= value <= 255.
2070 * ``txht:`` Set the host threshold register of the TX rings, 0 <= value <= 255.
2072 * ``txwt:`` Set the write-back threshold register of the TX rings, 0 <= value <= 255.
2074 * ``rxpt:`` Set the prefetch threshold register of the RX rings, 0 <= value <= 255.
2076 * ``rxht:`` Set the host threshold register of the RX rings, 0 <= value <= 255.
2078 * ``rxwt:`` Set the write-back threshold register of the RX rings, 0 <= value <= 255.
2080 * ``txfreet:`` Set the transmit free threshold of the TX rings, 0 <= value <= txd.
2082 * ``rxfreet:`` Set the transmit free threshold of the RX rings, 0 <= value <= rxd.
2084 * ``txrst:`` Set the transmit RS bit threshold of TX rings, 0 <= value <= txd.
2086 These threshold options are also available from the command-line.
2091 Set the value of ether-type for E-tag::
2093 testpmd> port config (port_id|all) l2-tunnel E-tag ether-type (value)
2095 Enable/disable the E-tag support::
2097 testpmd> port config (port_id|all) l2-tunnel E-tag (enable|disable)
2099 port config pctype mapping
2100 ~~~~~~~~~~~~~~~~~~~~~~~~~~
2102 Reset pctype mapping table::
2104 testpmd> port config (port_id) pctype mapping reset
2106 Update hardware defined pctype to software defined flow type mapping table::
2108 testpmd> port config (port_id) pctype mapping update (pctype_id_0[,pctype_id_1]*) (flow_type_id)
2112 * ``pctype_id_x``: hardware pctype id as index of bit in bitmask value of the pctype mapping table.
2114 * ``flow_type_id``: software flow type id as the index of the pctype mapping table.
2116 port config input set
2117 ~~~~~~~~~~~~~~~~~~~~~
2119 Config RSS/FDIR/FDIR flexible payload input set for some pctype::
2120 testpmd> port config (port_id) pctype (pctype_id) \
2121 (hash_inset|fdir_inset|fdir_flx_inset) \
2122 (get|set|clear) field (field_idx)
2124 Clear RSS/FDIR/FDIR flexible payload input set for some pctype::
2125 testpmd> port config (port_id) pctype (pctype_id) \
2126 (hash_inset|fdir_inset|fdir_flx_inset) clear all
2130 * ``pctype_id``: hardware packet classification types.
2131 * ``field_idx``: hardware field index.
2133 port config udp_tunnel_port
2134 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
2136 Add/remove UDP tunnel port for VXLAN/GENEVE tunneling protocols::
2137 testpmd> port config (port_id) udp_tunnel_port add|rm vxlan|geneve (udp_port)
2139 port config tx_metadata
2140 ~~~~~~~~~~~~~~~~~~~~~~~
2142 Set Tx metadata value per port.
2143 testpmd will add this value to any Tx packet sent from this port::
2145 testpmd> port config (port_id) tx_metadata (value)
2147 Link Bonding Functions
2148 ----------------------
2150 The Link Bonding functions make it possible to dynamically create and
2151 manage link bonding devices from within testpmd interactive prompt.
2153 create bonded device
2154 ~~~~~~~~~~~~~~~~~~~~
2156 Create a new bonding device::
2158 testpmd> create bonded device (mode) (socket)
2160 For example, to create a bonded device in mode 1 on socket 0::
2162 testpmd> create bonded device 1 0
2163 created new bonded device (port X)
2168 Adds Ethernet device to a Link Bonding device::
2170 testpmd> add bonding slave (slave id) (port id)
2172 For example, to add Ethernet device (port 6) to a Link Bonding device (port 10)::
2174 testpmd> add bonding slave 6 10
2177 remove bonding slave
2178 ~~~~~~~~~~~~~~~~~~~~
2180 Removes an Ethernet slave device from a Link Bonding device::
2182 testpmd> remove bonding slave (slave id) (port id)
2184 For example, to remove Ethernet slave device (port 6) to a Link Bonding device (port 10)::
2186 testpmd> remove bonding slave 6 10
2191 Set the Link Bonding mode of a Link Bonding device::
2193 testpmd> set bonding mode (value) (port id)
2195 For example, to set the bonding mode of a Link Bonding device (port 10) to broadcast (mode 3)::
2197 testpmd> set bonding mode 3 10
2202 Set an Ethernet slave device as the primary device on a Link Bonding device::
2204 testpmd> set bonding primary (slave id) (port id)
2206 For example, to set the Ethernet slave device (port 6) as the primary port of a Link Bonding device (port 10)::
2208 testpmd> set bonding primary 6 10
2213 Set the MAC address of a Link Bonding device::
2215 testpmd> set bonding mac (port id) (mac)
2217 For example, to set the MAC address of a Link Bonding device (port 10) to 00:00:00:00:00:01::
2219 testpmd> set bonding mac 10 00:00:00:00:00:01
2221 set bonding xmit_balance_policy
2222 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2224 Set the transmission policy for a Link Bonding device when it is in Balance XOR mode::
2226 testpmd> set bonding xmit_balance_policy (port_id) (l2|l23|l34)
2228 For example, set a Link Bonding device (port 10) to use a balance policy of layer 3+4 (IP addresses & UDP ports)::
2230 testpmd> set bonding xmit_balance_policy 10 l34
2233 set bonding mon_period
2234 ~~~~~~~~~~~~~~~~~~~~~~
2236 Set the link status monitoring polling period in milliseconds for a bonding device.
2238 This adds support for PMD slave devices which do not support link status interrupts.
2239 When the mon_period is set to a value greater than 0 then all PMD's which do not support
2240 link status ISR will be queried every polling interval to check if their link status has changed::
2242 testpmd> set bonding mon_period (port_id) (value)
2244 For example, to set the link status monitoring polling period of bonded device (port 5) to 150ms::
2246 testpmd> set bonding mon_period 5 150
2249 set bonding lacp dedicated_queue
2250 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2252 Enable dedicated tx/rx queues on bonding devices slaves to handle LACP control plane traffic
2253 when in mode 4 (link-aggregration-802.3ad)::
2255 testpmd> set bonding lacp dedicated_queues (port_id) (enable|disable)
2258 set bonding agg_mode
2259 ~~~~~~~~~~~~~~~~~~~~
2261 Enable one of the specific aggregators mode when in mode 4 (link-aggregration-802.3ad)::
2263 testpmd> set bonding agg_mode (port_id) (bandwidth|count|stable)
2269 Show the current configuration of a Link Bonding device::
2271 testpmd> show bonding config (port id)
2274 to show the configuration a Link Bonding device (port 9) with 3 slave devices (1, 3, 4)
2275 in balance mode with a transmission policy of layer 2+3::
2277 testpmd> show bonding config 9
2279 Balance Xmit Policy: BALANCE_XMIT_POLICY_LAYER23
2281 Active Slaves (3): [1 3 4]
2288 The Register Functions can be used to read from and write to registers on the network card referenced by a port number.
2289 This is mainly useful for debugging purposes.
2290 Reference should be made to the appropriate datasheet for the network card for details on the register addresses
2291 and fields that can be accessed.
2296 Display the value of a port register::
2298 testpmd> read reg (port_id) (address)
2300 For example, to examine the Flow Director control register (FDIRCTL, 0x0000EE000) on an Intel 82599 10 GbE Controller::
2302 testpmd> read reg 0 0xEE00
2303 port 0 PCI register at offset 0xEE00: 0x4A060029 (1241907241)
2308 Display a port register bit field::
2310 testpmd> read regfield (port_id) (address) (bit_x) (bit_y)
2312 For example, reading the lowest two bits from the register in the example above::
2314 testpmd> read regfield 0 0xEE00 0 1
2315 port 0 PCI register at offset 0xEE00: bits[0, 1]=0x1 (1)
2320 Display a single port register bit::
2322 testpmd> read regbit (port_id) (address) (bit_x)
2324 For example, reading the lowest bit from the register in the example above::
2326 testpmd> read regbit 0 0xEE00 0
2327 port 0 PCI register at offset 0xEE00: bit 0=1
2332 Set the value of a port register::
2334 testpmd> write reg (port_id) (address) (value)
2336 For example, to clear a register::
2338 testpmd> write reg 0 0xEE00 0x0
2339 port 0 PCI register at offset 0xEE00: 0x00000000 (0)
2344 Set bit field of a port register::
2346 testpmd> write regfield (port_id) (address) (bit_x) (bit_y) (value)
2348 For example, writing to the register cleared in the example above::
2350 testpmd> write regfield 0 0xEE00 0 1 2
2351 port 0 PCI register at offset 0xEE00: 0x00000002 (2)
2356 Set single bit value of a port register::
2358 testpmd> write regbit (port_id) (address) (bit_x) (value)
2360 For example, to set the high bit in the register from the example above::
2362 testpmd> write regbit 0 0xEE00 31 1
2363 port 0 PCI register at offset 0xEE00: 0x8000000A (2147483658)
2365 Traffic Metering and Policing
2366 -----------------------------
2368 The following section shows functions for configuring traffic metering and
2369 policing on the ethernet device through the use of generic ethdev API.
2371 show port traffic management capability
2372 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2374 Show traffic metering and policing capability of the port::
2376 testpmd> show port meter cap (port_id)
2378 add port meter profile (srTCM rfc2967)
2379 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2381 Add meter profile (srTCM rfc2697) to the ethernet device::
2383 testpmd> add port meter profile srtcm_rfc2697 (port_id) (profile_id) \
2388 * ``profile_id``: ID for the meter profile.
2389 * ``cir``: Committed Information Rate (CIR) (bytes/second).
2390 * ``cbs``: Committed Burst Size (CBS) (bytes).
2391 * ``ebs``: Excess Burst Size (EBS) (bytes).
2393 add port meter profile (trTCM rfc2968)
2394 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2396 Add meter profile (srTCM rfc2698) to the ethernet device::
2398 testpmd> add port meter profile trtcm_rfc2698 (port_id) (profile_id) \
2399 (cir) (pir) (cbs) (pbs)
2403 * ``profile_id``: ID for the meter profile.
2404 * ``cir``: Committed information rate (bytes/second).
2405 * ``pir``: Peak information rate (bytes/second).
2406 * ``cbs``: Committed burst size (bytes).
2407 * ``pbs``: Peak burst size (bytes).
2409 add port meter profile (trTCM rfc4115)
2410 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2412 Add meter profile (trTCM rfc4115) to the ethernet device::
2414 testpmd> add port meter profile trtcm_rfc4115 (port_id) (profile_id) \
2415 (cir) (eir) (cbs) (ebs)
2419 * ``profile_id``: ID for the meter profile.
2420 * ``cir``: Committed information rate (bytes/second).
2421 * ``eir``: Excess information rate (bytes/second).
2422 * ``cbs``: Committed burst size (bytes).
2423 * ``ebs``: Excess burst size (bytes).
2425 delete port meter profile
2426 ~~~~~~~~~~~~~~~~~~~~~~~~~
2428 Delete meter profile from the ethernet device::
2430 testpmd> del port meter profile (port_id) (profile_id)
2435 Create new meter object for the ethernet device::
2437 testpmd> create port meter (port_id) (mtr_id) (profile_id) \
2438 (meter_enable) (g_action) (y_action) (r_action) (stats_mask) (shared) \
2439 (use_pre_meter_color) [(dscp_tbl_entry0) (dscp_tbl_entry1)...\
2444 * ``mtr_id``: meter object ID.
2445 * ``profile_id``: ID for the meter profile.
2446 * ``meter_enable``: When this parameter has a non-zero value, the meter object
2447 gets enabled at the time of creation, otherwise remains disabled.
2448 * ``g_action``: Policer action for the packet with green color.
2449 * ``y_action``: Policer action for the packet with yellow color.
2450 * ``r_action``: Policer action for the packet with red color.
2451 * ``stats_mask``: Mask of statistics counter types to be enabled for the
2453 * ``shared``: When this parameter has a non-zero value, the meter object is
2454 shared by multiple flows. Otherwise, meter object is used by single flow.
2455 * ``use_pre_meter_color``: When this parameter has a non-zero value, the
2456 input color for the current meter object is determined by the latest meter
2457 object in the same flow. Otherwise, the current meter object uses the
2458 *dscp_table* to determine the input color.
2459 * ``dscp_tbl_entryx``: DSCP table entry x providing meter providing input
2460 color, 0 <= x <= 63.
2465 Enable meter for the ethernet device::
2467 testpmd> enable port meter (port_id) (mtr_id)
2472 Disable meter for the ethernet device::
2474 testpmd> disable port meter (port_id) (mtr_id)
2479 Delete meter for the ethernet device::
2481 testpmd> del port meter (port_id) (mtr_id)
2483 Set port meter profile
2484 ~~~~~~~~~~~~~~~~~~~~~~
2486 Set meter profile for the ethernet device::
2488 testpmd> set port meter profile (port_id) (mtr_id) (profile_id)
2490 set port meter dscp table
2491 ~~~~~~~~~~~~~~~~~~~~~~~~~
2493 Set meter dscp table for the ethernet device::
2495 testpmd> set port meter dscp table (port_id) (mtr_id) [(dscp_tbl_entry0) \
2496 (dscp_tbl_entry1)...(dscp_tbl_entry63)]
2498 set port meter policer action
2499 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2501 Set meter policer action for the ethernet device::
2503 testpmd> set port meter policer action (port_id) (mtr_id) (action_mask) \
2504 (action0) [(action1) (action1)]
2508 * ``action_mask``: Bit mask indicating which policer actions need to be
2509 updated. One or more policer actions can be updated in a single function
2510 invocation. To update the policer action associated with color C, bit
2511 (1 << C) needs to be set in *action_mask* and element at position C
2512 in the *actions* array needs to be valid.
2513 * ``actionx``: Policer action for the color x,
2514 RTE_MTR_GREEN <= x < RTE_MTR_COLORS
2516 set port meter stats mask
2517 ~~~~~~~~~~~~~~~~~~~~~~~~~
2519 Set meter stats mask for the ethernet device::
2521 testpmd> set port meter stats mask (port_id) (mtr_id) (stats_mask)
2525 * ``stats_mask``: Bit mask indicating statistics counter types to be enabled.
2527 show port meter stats
2528 ~~~~~~~~~~~~~~~~~~~~~
2530 Show meter stats of the ethernet device::
2532 testpmd> show port meter stats (port_id) (mtr_id) (clear)
2536 * ``clear``: Flag that indicates whether the statistics counters should
2537 be cleared (i.e. set to zero) immediately after they have been read or not.
2542 The following section shows functions for configuring traffic management on
2543 on the ethernet device through the use of generic TM API.
2545 show port traffic management capability
2546 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2548 Show traffic management capability of the port::
2550 testpmd> show port tm cap (port_id)
2552 show port traffic management capability (hierarchy level)
2553 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2555 Show traffic management hierarchy level capability of the port::
2557 testpmd> show port tm level cap (port_id) (level_id)
2559 show port traffic management capability (hierarchy node level)
2560 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2562 Show the traffic management hierarchy node capability of the port::
2564 testpmd> show port tm node cap (port_id) (node_id)
2566 show port traffic management hierarchy node type
2567 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2569 Show the port traffic management hierarchy node type::
2571 testpmd> show port tm node type (port_id) (node_id)
2573 show port traffic management hierarchy node stats
2574 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2576 Show the port traffic management hierarchy node statistics::
2578 testpmd> show port tm node stats (port_id) (node_id) (clear)
2582 * ``clear``: When this parameter has a non-zero value, the statistics counters
2583 are cleared (i.e. set to zero) immediately after they have been read,
2584 otherwise the statistics counters are left untouched.
2586 Add port traffic management private shaper profile
2587 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2589 Add the port traffic management private shaper profile::
2591 testpmd> add port tm node shaper profile (port_id) (shaper_profile_id) \
2592 (cmit_tb_rate) (cmit_tb_size) (peak_tb_rate) (peak_tb_size) \
2593 (packet_length_adjust)
2597 * ``shaper_profile id``: Shaper profile ID for the new profile.
2598 * ``cmit_tb_rate``: Committed token bucket rate (bytes per second).
2599 * ``cmit_tb_size``: Committed token bucket size (bytes).
2600 * ``peak_tb_rate``: Peak token bucket rate (bytes per second).
2601 * ``peak_tb_size``: Peak token bucket size (bytes).
2602 * ``packet_length_adjust``: The value (bytes) to be added to the length of
2603 each packet for the purpose of shaping. This parameter value can be used to
2604 correct the packet length with the framing overhead bytes that are consumed
2607 Delete port traffic management private shaper profile
2608 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2610 Delete the port traffic management private shaper::
2612 testpmd> del port tm node shaper profile (port_id) (shaper_profile_id)
2616 * ``shaper_profile id``: Shaper profile ID that needs to be deleted.
2618 Add port traffic management shared shaper
2619 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2621 Create the port traffic management shared shaper::
2623 testpmd> add port tm node shared shaper (port_id) (shared_shaper_id) \
2628 * ``shared_shaper_id``: Shared shaper ID to be created.
2629 * ``shaper_profile id``: Shaper profile ID for shared shaper.
2631 Set port traffic management shared shaper
2632 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2634 Update the port traffic management shared shaper::
2636 testpmd> set port tm node shared shaper (port_id) (shared_shaper_id) \
2641 * ``shared_shaper_id``: Shared shaper ID to be update.
2642 * ``shaper_profile id``: Shaper profile ID for shared shaper.
2644 Delete port traffic management shared shaper
2645 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2647 Delete the port traffic management shared shaper::
2649 testpmd> del port tm node shared shaper (port_id) (shared_shaper_id)
2653 * ``shared_shaper_id``: Shared shaper ID to be deleted.
2655 Set port traffic management hiearchy node private shaper
2656 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2658 set the port traffic management hierarchy node private shaper::
2660 testpmd> set port tm node shaper profile (port_id) (node_id) \
2665 * ``shaper_profile id``: Private shaper profile ID to be enabled on the
2668 Add port traffic management WRED profile
2669 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2671 Create a new WRED profile::
2673 testpmd> add port tm node wred profile (port_id) (wred_profile_id) \
2674 (color_g) (min_th_g) (max_th_g) (maxp_inv_g) (wq_log2_g) \
2675 (color_y) (min_th_y) (max_th_y) (maxp_inv_y) (wq_log2_y) \
2676 (color_r) (min_th_r) (max_th_r) (maxp_inv_r) (wq_log2_r)
2680 * ``wred_profile id``: Identifier for the newly create WRED profile
2681 * ``color_g``: Packet color (green)
2682 * ``min_th_g``: Minimum queue threshold for packet with green color
2683 * ``max_th_g``: Minimum queue threshold for packet with green color
2684 * ``maxp_inv_g``: Inverse of packet marking probability maximum value (maxp)
2685 * ``wq_log2_g``: Negated log2 of queue weight (wq)
2686 * ``color_y``: Packet color (yellow)
2687 * ``min_th_y``: Minimum queue threshold for packet with yellow color
2688 * ``max_th_y``: Minimum queue threshold for packet with yellow color
2689 * ``maxp_inv_y``: Inverse of packet marking probability maximum value (maxp)
2690 * ``wq_log2_y``: Negated log2 of queue weight (wq)
2691 * ``color_r``: Packet color (red)
2692 * ``min_th_r``: Minimum queue threshold for packet with yellow color
2693 * ``max_th_r``: Minimum queue threshold for packet with yellow color
2694 * ``maxp_inv_r``: Inverse of packet marking probability maximum value (maxp)
2695 * ``wq_log2_r``: Negated log2 of queue weight (wq)
2697 Delete port traffic management WRED profile
2698 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2700 Delete the WRED profile::
2702 testpmd> del port tm node wred profile (port_id) (wred_profile_id)
2704 Add port traffic management hierarchy nonleaf node
2705 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2707 Add nonleaf node to port traffic management hiearchy::
2709 testpmd> add port tm nonleaf node (port_id) (node_id) (parent_node_id) \
2710 (priority) (weight) (level_id) (shaper_profile_id) \
2711 (n_sp_priorities) (stats_mask) (n_shared_shapers) \
2712 [(shared_shaper_0) (shared_shaper_1) ...] \
2716 * ``parent_node_id``: Node ID of the parent.
2717 * ``priority``: Node priority (highest node priority is zero). This is used by
2718 the SP algorithm running on the parent node for scheduling this node.
2719 * ``weight``: Node weight (lowest weight is one). The node weight is relative
2720 to the weight sum of all siblings that have the same priority. It is used by
2721 the WFQ algorithm running on the parent node for scheduling this node.
2722 * ``level_id``: Hiearchy level of the node.
2723 * ``shaper_profile_id``: Shaper profile ID of the private shaper to be used by
2725 * ``n_sp_priorities``: Number of strict priorities.
2726 * ``stats_mask``: Mask of statistics counter types to be enabled for this node.
2727 * ``n_shared_shapers``: Number of shared shapers.
2728 * ``shared_shaper_id``: Shared shaper id.
2730 Add port traffic management hierarchy leaf node
2731 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2733 Add leaf node to port traffic management hiearchy::
2735 testpmd> add port tm leaf node (port_id) (node_id) (parent_node_id) \
2736 (priority) (weight) (level_id) (shaper_profile_id) \
2737 (cman_mode) (wred_profile_id) (stats_mask) (n_shared_shapers) \
2738 [(shared_shaper_id) (shared_shaper_id) ...] \
2742 * ``parent_node_id``: Node ID of the parent.
2743 * ``priority``: Node priority (highest node priority is zero). This is used by
2744 the SP algorithm running on the parent node for scheduling this node.
2745 * ``weight``: Node weight (lowest weight is one). The node weight is relative
2746 to the weight sum of all siblings that have the same priority. It is used by
2747 the WFQ algorithm running on the parent node for scheduling this node.
2748 * ``level_id``: Hiearchy level of the node.
2749 * ``shaper_profile_id``: Shaper profile ID of the private shaper to be used by
2751 * ``cman_mode``: Congestion management mode to be enabled for this node.
2752 * ``wred_profile_id``: WRED profile id to be enabled for this node.
2753 * ``stats_mask``: Mask of statistics counter types to be enabled for this node.
2754 * ``n_shared_shapers``: Number of shared shapers.
2755 * ``shared_shaper_id``: Shared shaper id.
2757 Delete port traffic management hierarchy node
2758 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2760 Delete node from port traffic management hiearchy::
2762 testpmd> del port tm node (port_id) (node_id)
2764 Update port traffic management hierarchy parent node
2765 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2767 Update port traffic management hierarchy parent node::
2769 testpmd> set port tm node parent (port_id) (node_id) (parent_node_id) \
2772 This function can only be called after the hierarchy commit invocation. Its
2773 success depends on the port support for this operation, as advertised through
2774 the port capability set. This function is valid for all nodes of the traffic
2775 management hierarchy except root node.
2777 Suspend port traffic management hierarchy node
2778 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2780 testpmd> suspend port tm node (port_id) (node_id)
2782 Resume port traffic management hierarchy node
2783 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2785 testpmd> resume port tm node (port_id) (node_id)
2787 Commit port traffic management hierarchy
2788 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2790 Commit the traffic management hierarchy on the port::
2792 testpmd> port tm hierarchy commit (port_id) (clean_on_fail)
2796 * ``clean_on_fail``: When set to non-zero, hierarchy is cleared on function
2797 call failure. On the other hand, hierarchy is preserved when this parameter
2800 Set port traffic management mark VLAN dei
2801 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2803 Enables/Disables the traffic management marking on the port for VLAN packets::
2805 testpmd> set port tm mark vlan_dei <port_id> <green> <yellow> <red>
2809 * ``port_id``: The port which on which VLAN packets marked as ``green`` or
2810 ``yellow`` or ``red`` will have dei bit enabled
2812 * ``green`` enable 1, disable 0 marking for dei bit of VLAN packets marked as green
2814 * ``yellow`` enable 1, disable 0 marking for dei bit of VLAN packets marked as yellow
2816 * ``red`` enable 1, disable 0 marking for dei bit of VLAN packets marked as red
2818 Set port traffic management mark IP dscp
2819 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2821 Enables/Disables the traffic management marking on the port for IP dscp packets::
2823 testpmd> set port tm mark ip_dscp <port_id> <green> <yellow> <red>
2827 * ``port_id``: The port which on which IP packets marked as ``green`` or
2828 ``yellow`` or ``red`` will have IP dscp bits updated
2830 * ``green`` enable 1, disable 0 marking IP dscp to low drop precedence for green packets
2832 * ``yellow`` enable 1, disable 0 marking IP dscp to medium drop precedence for yellow packets
2834 * ``red`` enable 1, disable 0 marking IP dscp to high drop precedence for red packets
2836 Set port traffic management mark IP ecn
2837 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2839 Enables/Disables the traffic management marking on the port for IP ecn packets::
2841 testpmd> set port tm mark ip_ecn <port_id> <green> <yellow> <red>
2845 * ``port_id``: The port which on which IP packets marked as ``green`` or
2846 ``yellow`` or ``red`` will have IP ecn bits updated
2848 * ``green`` enable 1, disable 0 marking IP ecn for green marked packets with ecn of 2'b01 or 2'b10
2849 to ecn of 2'b11 when IP is caring TCP or SCTP
2851 * ``yellow`` enable 1, disable 0 marking IP ecn for yellow marked packets with ecn of 2'b01 or 2'b10
2852 to ecn of 2'b11 when IP is caring TCP or SCTP
2854 * ``red`` enable 1, disable 0 marking IP ecn for yellow marked packets with ecn of 2'b01 or 2'b10
2855 to ecn of 2'b11 when IP is caring TCP or SCTP
2857 Set port traffic management default hierarchy (softnic forwarding mode)
2858 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2860 set the traffic management default hierarchy on the port::
2862 testpmd> set port tm hierarchy default (port_id)
2867 This section details the available filter functions that are available.
2869 Note these functions interface the deprecated legacy filtering framework,
2870 superseded by *rte_flow*. See `Flow rules management`_.
2873 ~~~~~~~~~~~~~~~~~~~~
2875 Add or delete a L2 Ethertype filter, which identify packets by their L2 Ethertype mainly assign them to a receive queue::
2877 ethertype_filter (port_id) (add|del) (mac_addr|mac_ignr) (mac_address) \
2878 ethertype (ether_type) (drop|fwd) queue (queue_id)
2880 The available information parameters are:
2882 * ``port_id``: The port which the Ethertype filter assigned on.
2884 * ``mac_addr``: Compare destination mac address.
2886 * ``mac_ignr``: Ignore destination mac address match.
2888 * ``mac_address``: Destination mac address to match.
2890 * ``ether_type``: The EtherType value want to match,
2891 for example 0x0806 for ARP packet. 0x0800 (IPv4) and 0x86DD (IPv6) are invalid.
2893 * ``queue_id``: The receive queue associated with this EtherType filter.
2894 It is meaningless when deleting or dropping.
2896 Example, to add/remove an ethertype filter rule::
2898 testpmd> ethertype_filter 0 add mac_ignr 00:11:22:33:44:55 \
2899 ethertype 0x0806 fwd queue 3
2901 testpmd> ethertype_filter 0 del mac_ignr 00:11:22:33:44:55 \
2902 ethertype 0x0806 fwd queue 3
2907 Add or delete a 2-tuple filter,
2908 which identifies packets by specific protocol and destination TCP/UDP port
2909 and forwards packets into one of the receive queues::
2911 2tuple_filter (port_id) (add|del) dst_port (dst_port_value) \
2912 protocol (protocol_value) mask (mask_value) \
2913 tcp_flags (tcp_flags_value) priority (prio_value) \
2916 The available information parameters are:
2918 * ``port_id``: The port which the 2-tuple filter assigned on.
2920 * ``dst_port_value``: Destination port in L4.
2922 * ``protocol_value``: IP L4 protocol.
2924 * ``mask_value``: Participates in the match or not by bit for field above, 1b means participate.
2926 * ``tcp_flags_value``: TCP control bits. The non-zero value is invalid, when the pro_value is not set to 0x06 (TCP).
2928 * ``prio_value``: Priority of this filter.
2930 * ``queue_id``: The receive queue associated with this 2-tuple filter.
2932 Example, to add/remove an 2tuple filter rule::
2934 testpmd> 2tuple_filter 0 add dst_port 32 protocol 0x06 mask 0x03 \
2935 tcp_flags 0x02 priority 3 queue 3
2937 testpmd> 2tuple_filter 0 del dst_port 32 protocol 0x06 mask 0x03 \
2938 tcp_flags 0x02 priority 3 queue 3
2943 Add or delete a 5-tuple filter,
2944 which consists of a 5-tuple (protocol, source and destination IP addresses, source and destination TCP/UDP/SCTP port)
2945 and routes packets into one of the receive queues::
2947 5tuple_filter (port_id) (add|del) dst_ip (dst_address) src_ip \
2948 (src_address) dst_port (dst_port_value) \
2949 src_port (src_port_value) protocol (protocol_value) \
2950 mask (mask_value) tcp_flags (tcp_flags_value) \
2951 priority (prio_value) queue (queue_id)
2953 The available information parameters are:
2955 * ``port_id``: The port which the 5-tuple filter assigned on.
2957 * ``dst_address``: Destination IP address.
2959 * ``src_address``: Source IP address.
2961 * ``dst_port_value``: TCP/UDP destination port.
2963 * ``src_port_value``: TCP/UDP source port.
2965 * ``protocol_value``: L4 protocol.
2967 * ``mask_value``: Participates in the match or not by bit for field above, 1b means participate
2969 * ``tcp_flags_value``: TCP control bits. The non-zero value is invalid, when the protocol_value is not set to 0x06 (TCP).
2971 * ``prio_value``: The priority of this filter.
2973 * ``queue_id``: The receive queue associated with this 5-tuple filter.
2975 Example, to add/remove an 5tuple filter rule::
2977 testpmd> 5tuple_filter 0 add dst_ip 2.2.2.5 src_ip 2.2.2.4 \
2978 dst_port 64 src_port 32 protocol 0x06 mask 0x1F \
2979 flags 0x0 priority 3 queue 3
2981 testpmd> 5tuple_filter 0 del dst_ip 2.2.2.5 src_ip 2.2.2.4 \
2982 dst_port 64 src_port 32 protocol 0x06 mask 0x1F \
2983 flags 0x0 priority 3 queue 3
2988 Using the SYN filter, TCP packets whose *SYN* flag is set can be forwarded to a separate queue::
2990 syn_filter (port_id) (add|del) priority (high|low) queue (queue_id)
2992 The available information parameters are:
2994 * ``port_id``: The port which the SYN filter assigned on.
2996 * ``high``: This SYN filter has higher priority than other filters.
2998 * ``low``: This SYN filter has lower priority than other filters.
3000 * ``queue_id``: The receive queue associated with this SYN filter
3004 testpmd> syn_filter 0 add priority high queue 3
3009 With flex filter, packets can be recognized by any arbitrary pattern within the first 128 bytes of the packet
3010 and routed into one of the receive queues::
3012 flex_filter (port_id) (add|del) len (len_value) bytes (bytes_value) \
3013 mask (mask_value) priority (prio_value) queue (queue_id)
3015 The available information parameters are:
3017 * ``port_id``: The port which the Flex filter is assigned on.
3019 * ``len_value``: Filter length in bytes, no greater than 128.
3021 * ``bytes_value``: A string in hexadecimal, means the value the flex filter needs to match.
3023 * ``mask_value``: A string in hexadecimal, bit 1 means corresponding byte participates in the match.
3025 * ``prio_value``: The priority of this filter.
3027 * ``queue_id``: The receive queue associated with this Flex filter.
3031 testpmd> flex_filter 0 add len 16 bytes 0x00000000000000000000000008060000 \
3032 mask 000C priority 3 queue 3
3034 testpmd> flex_filter 0 del len 16 bytes 0x00000000000000000000000008060000 \
3035 mask 000C priority 3 queue 3
3038 .. _testpmd_flow_director:
3040 flow_director_filter
3041 ~~~~~~~~~~~~~~~~~~~~
3043 The Flow Director works in receive mode to identify specific flows or sets of flows and route them to specific queues.
3045 Four types of filtering are supported which are referred to as Perfect Match, Signature, Perfect-mac-vlan and
3046 Perfect-tunnel filters, the match mode is set by the ``--pkt-filter-mode`` command-line parameter:
3048 * Perfect 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 IP flow.
3052 * Signature filters.
3053 The hardware checks a match between a hash-based signature of the masked fields of the received packet.
3055 * Perfect-mac-vlan match filters.
3056 The hardware checks a match between the masked fields of the received packets and the programmed filters.
3057 The masked fields are for MAC VLAN flow.
3059 * Perfect-tunnel match filters.
3060 The hardware checks a match between the masked fields of the received packets and the programmed filters.
3061 The masked fields are for tunnel flow.
3063 * Perfect-raw-flow-type match filters.
3064 The hardware checks a match between the masked fields of the received packets and pre-loaded raw (template) packet.
3065 The masked fields are specified by input sets.
3067 The Flow Director filters can match the different fields for different type of packet: flow type, specific input set
3068 per flow type and the flexible payload.
3070 The Flow Director can also mask out parts of all of these fields so that filters
3071 are only applied to certain fields or parts of the fields.
3073 Note that for raw flow type mode the source and destination fields in the
3074 raw packet buffer need to be presented in a reversed order with respect
3075 to the expected received packets.
3076 For example: IP source and destination addresses or TCP/UDP/SCTP
3077 source and destination ports
3079 Different NICs may have different capabilities, command show port fdir (port_id) can be used to acquire the information.
3081 # Commands to add flow director filters of different flow types::
3083 flow_director_filter (port_id) mode IP (add|del|update) \
3084 flow (ipv4-other|ipv4-frag|ipv6-other|ipv6-frag) \
3085 src (src_ip_address) dst (dst_ip_address) \
3086 tos (tos_value) proto (proto_value) ttl (ttl_value) \
3087 vlan (vlan_value) flexbytes (flexbytes_value) \
3088 (drop|fwd) pf|vf(vf_id) queue (queue_id) \
3091 flow_director_filter (port_id) mode IP (add|del|update) \
3092 flow (ipv4-tcp|ipv4-udp|ipv6-tcp|ipv6-udp) \
3093 src (src_ip_address) (src_port) \
3094 dst (dst_ip_address) (dst_port) \
3095 tos (tos_value) ttl (ttl_value) \
3096 vlan (vlan_value) flexbytes (flexbytes_value) \
3097 (drop|fwd) queue pf|vf(vf_id) (queue_id) \
3100 flow_director_filter (port_id) mode IP (add|del|update) \
3101 flow (ipv4-sctp|ipv6-sctp) \
3102 src (src_ip_address) (src_port) \
3103 dst (dst_ip_address) (dst_port) \
3104 tos (tos_value) ttl (ttl_value) \
3105 tag (verification_tag) vlan (vlan_value) \
3106 flexbytes (flexbytes_value) (drop|fwd) \
3107 pf|vf(vf_id) queue (queue_id) fd_id (fd_id_value)
3109 flow_director_filter (port_id) mode IP (add|del|update) flow l2_payload \
3110 ether (ethertype) flexbytes (flexbytes_value) \
3111 (drop|fwd) pf|vf(vf_id) queue (queue_id)
3114 flow_director_filter (port_id) mode MAC-VLAN (add|del|update) \
3115 mac (mac_address) vlan (vlan_value) \
3116 flexbytes (flexbytes_value) (drop|fwd) \
3117 queue (queue_id) fd_id (fd_id_value)
3119 flow_director_filter (port_id) mode Tunnel (add|del|update) \
3120 mac (mac_address) vlan (vlan_value) \
3121 tunnel (NVGRE|VxLAN) tunnel-id (tunnel_id_value) \
3122 flexbytes (flexbytes_value) (drop|fwd) \
3123 queue (queue_id) fd_id (fd_id_value)
3125 flow_director_filter (port_id) mode raw (add|del|update) flow (flow_id) \
3126 (drop|fwd) queue (queue_id) fd_id (fd_id_value) \
3127 packet (packet file name)
3129 For example, to add an ipv4-udp flow type filter::
3131 testpmd> flow_director_filter 0 mode IP add flow ipv4-udp src 2.2.2.3 32 \
3132 dst 2.2.2.5 33 tos 2 ttl 40 vlan 0x1 flexbytes (0x88,0x48) \
3133 fwd pf queue 1 fd_id 1
3135 For example, add an ipv4-other flow type filter::
3137 testpmd> flow_director_filter 0 mode IP add flow ipv4-other src 2.2.2.3 \
3138 dst 2.2.2.5 tos 2 proto 20 ttl 40 vlan 0x1 \
3139 flexbytes (0x88,0x48) fwd pf queue 1 fd_id 1
3144 Flush all flow director filters on a device::
3146 testpmd> flush_flow_director (port_id)
3148 Example, to flush all flow director filter on port 0::
3150 testpmd> flush_flow_director 0
3155 Set flow director's input masks::
3157 flow_director_mask (port_id) mode IP vlan (vlan_value) \
3158 src_mask (ipv4_src) (ipv6_src) (src_port) \
3159 dst_mask (ipv4_dst) (ipv6_dst) (dst_port)
3161 flow_director_mask (port_id) mode MAC-VLAN vlan (vlan_value)
3163 flow_director_mask (port_id) mode Tunnel vlan (vlan_value) \
3164 mac (mac_value) tunnel-type (tunnel_type_value) \
3165 tunnel-id (tunnel_id_value)
3167 Example, to set flow director mask on port 0::
3169 testpmd> flow_director_mask 0 mode IP vlan 0xefff \
3170 src_mask 255.255.255.255 \
3171 FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF \
3172 dst_mask 255.255.255.255 \
3173 FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF
3175 flow_director_flex_mask
3176 ~~~~~~~~~~~~~~~~~~~~~~~
3178 set masks of flow director's flexible payload based on certain flow type::
3180 testpmd> flow_director_flex_mask (port_id) \
3181 flow (none|ipv4-other|ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
3182 ipv6-other|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp| \
3183 l2_payload|all) (mask)
3185 Example, to set flow director's flex mask for all flow type on port 0::
3187 testpmd> flow_director_flex_mask 0 flow all \
3188 (0xff,0xff,0,0,0,0,0,0,0,0,0,0,0,0,0,0)
3191 flow_director_flex_payload
3192 ~~~~~~~~~~~~~~~~~~~~~~~~~~
3194 Configure flexible payload selection::
3196 flow_director_flex_payload (port_id) (raw|l2|l3|l4) (config)
3198 For example, to select the first 16 bytes from the offset 4 (bytes) of packet's payload as flexible payload::
3200 testpmd> flow_director_flex_payload 0 l4 \
3201 (4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19)
3203 get_sym_hash_ena_per_port
3204 ~~~~~~~~~~~~~~~~~~~~~~~~~
3206 Get symmetric hash enable configuration per port::
3208 get_sym_hash_ena_per_port (port_id)
3210 For example, to get symmetric hash enable configuration of port 1::
3212 testpmd> get_sym_hash_ena_per_port 1
3214 set_sym_hash_ena_per_port
3215 ~~~~~~~~~~~~~~~~~~~~~~~~~
3217 Set symmetric hash enable configuration per port to enable or disable::
3219 set_sym_hash_ena_per_port (port_id) (enable|disable)
3221 For example, to set symmetric hash enable configuration of port 1 to enable::
3223 testpmd> set_sym_hash_ena_per_port 1 enable
3225 get_hash_global_config
3226 ~~~~~~~~~~~~~~~~~~~~~~
3228 Get the global configurations of hash filters::
3230 get_hash_global_config (port_id)
3232 For example, to get the global configurations of hash filters of port 1::
3234 testpmd> get_hash_global_config 1
3236 set_hash_global_config
3237 ~~~~~~~~~~~~~~~~~~~~~~
3239 Set the global configurations of hash filters::
3241 set_hash_global_config (port_id) (toeplitz|simple_xor|default) \
3242 (ipv4|ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp|ipv4-other|ipv6|ipv6-frag| \
3243 ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other|l2_payload|<flow_id>) \
3246 For example, to enable simple_xor for flow type of ipv6 on port 2::
3248 testpmd> set_hash_global_config 2 simple_xor ipv6 enable
3253 Set the input set for hash::
3255 set_hash_input_set (port_id) (ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
3256 ipv4-other|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other| \
3257 l2_payload|<flow_id>) (ovlan|ivlan|src-ipv4|dst-ipv4|src-ipv6|dst-ipv6| \
3258 ipv4-tos|ipv4-proto|ipv6-tc|ipv6-next-header|udp-src-port|udp-dst-port| \
3259 tcp-src-port|tcp-dst-port|sctp-src-port|sctp-dst-port|sctp-veri-tag| \
3260 udp-key|gre-key|fld-1st|fld-2nd|fld-3rd|fld-4th|fld-5th|fld-6th|fld-7th| \
3261 fld-8th|none) (select|add)
3263 For example, to add source IP to hash input set for flow type of ipv4-udp on port 0::
3265 testpmd> set_hash_input_set 0 ipv4-udp src-ipv4 add
3270 The Flow Director filters can match the different fields for different type of packet, i.e. specific input set
3271 on per flow type and the flexible payload. This command can be used to change input set for each flow type.
3273 Set the input set for flow director::
3275 set_fdir_input_set (port_id) (ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
3276 ipv4-other|ipv6|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other| \
3277 l2_payload|<flow_id>) (ivlan|ethertype|src-ipv4|dst-ipv4|src-ipv6|dst-ipv6| \
3278 ipv4-tos|ipv4-proto|ipv4-ttl|ipv6-tc|ipv6-next-header|ipv6-hop-limits| \
3279 tudp-src-port|udp-dst-port|cp-src-port|tcp-dst-port|sctp-src-port| \
3280 sctp-dst-port|sctp-veri-tag|none) (select|add)
3282 For example to add source IP to FD input set for flow type of ipv4-udp on port 0::
3284 testpmd> set_fdir_input_set 0 ipv4-udp src-ipv4 add
3289 Set different GRE key length for input set::
3291 global_config (port_id) gre-key-len (number in bytes)
3293 For example to set GRE key length for input set to 4 bytes on port 0::
3295 testpmd> global_config 0 gre-key-len 4
3298 .. _testpmd_rte_flow:
3300 Flow rules management
3301 ---------------------
3303 Control of the generic flow API (*rte_flow*) is fully exposed through the
3304 ``flow`` command (validation, creation, destruction, queries and operation
3307 Considering *rte_flow* overlaps with all `Filter Functions`_, using both
3308 features simultaneously may cause undefined side-effects and is therefore
3314 Because the ``flow`` command uses dynamic tokens to handle the large number
3315 of possible flow rules combinations, its behavior differs slightly from
3316 other commands, in particular:
3318 - Pressing *?* or the *<tab>* key displays contextual help for the current
3319 token, not that of the entire command.
3321 - Optional and repeated parameters are supported (provided they are listed
3322 in the contextual help).
3324 The first parameter stands for the operation mode. Possible operations and
3325 their general syntax are described below. They are covered in detail in the
3328 - Check whether a flow rule can be created::
3330 flow validate {port_id}
3331 [group {group_id}] [priority {level}] [ingress] [egress] [transfer]
3332 pattern {item} [/ {item} [...]] / end
3333 actions {action} [/ {action} [...]] / end
3335 - Create a flow rule::
3337 flow create {port_id}
3338 [group {group_id}] [priority {level}] [ingress] [egress] [transfer]
3339 pattern {item} [/ {item} [...]] / end
3340 actions {action} [/ {action} [...]] / end
3342 - Destroy specific flow rules::
3344 flow destroy {port_id} rule {rule_id} [...]
3346 - Destroy all flow rules::
3348 flow flush {port_id}
3350 - Query an existing flow rule::
3352 flow query {port_id} {rule_id} {action}
3354 - List existing flow rules sorted by priority, filtered by group
3357 flow list {port_id} [group {group_id}] [...]
3359 - Restrict ingress traffic to the defined flow rules::
3361 flow isolate {port_id} {boolean}
3363 Validating flow rules
3364 ~~~~~~~~~~~~~~~~~~~~~
3366 ``flow validate`` reports whether a flow rule would be accepted by the
3367 underlying device in its current state but stops short of creating it. It is
3368 bound to ``rte_flow_validate()``::
3370 flow validate {port_id}
3371 [group {group_id}] [priority {level}] [ingress] [egress] [transfer]
3372 pattern {item} [/ {item} [...]] / end
3373 actions {action} [/ {action} [...]] / end
3375 If successful, it will show::
3379 Otherwise it will show an error message of the form::
3381 Caught error type [...] ([...]): [...]
3383 This command uses the same parameters as ``flow create``, their format is
3384 described in `Creating flow rules`_.
3386 Check whether redirecting any Ethernet packet received on port 0 to RX queue
3387 index 6 is supported::
3389 testpmd> flow validate 0 ingress pattern eth / end
3390 actions queue index 6 / end
3394 Port 0 does not support TCPv6 rules::
3396 testpmd> flow validate 0 ingress pattern eth / ipv6 / tcp / end
3398 Caught error type 9 (specific pattern item): Invalid argument
3404 ``flow create`` validates and creates the specified flow rule. It is bound
3405 to ``rte_flow_create()``::
3407 flow create {port_id}
3408 [group {group_id}] [priority {level}] [ingress] [egress] [transfer]
3409 pattern {item} [/ {item} [...]] / end
3410 actions {action} [/ {action} [...]] / end
3412 If successful, it will return a flow rule ID usable with other commands::
3414 Flow rule #[...] created
3416 Otherwise it will show an error message of the form::
3418 Caught error type [...] ([...]): [...]
3420 Parameters describe in the following order:
3422 - Attributes (*group*, *priority*, *ingress*, *egress*, *transfer* tokens).
3423 - A matching pattern, starting with the *pattern* token and terminated by an
3425 - Actions, starting with the *actions* token and terminated by an *end*
3428 These translate directly to *rte_flow* objects provided as-is to the
3429 underlying functions.
3431 The shortest valid definition only comprises mandatory tokens::
3433 testpmd> flow create 0 pattern end actions end
3435 Note that PMDs may refuse rules that essentially do nothing such as this
3438 **All unspecified object values are automatically initialized to 0.**
3443 These tokens affect flow rule attributes (``struct rte_flow_attr``) and are
3444 specified before the ``pattern`` token.
3446 - ``group {group id}``: priority group.
3447 - ``priority {level}``: priority level within group.
3448 - ``ingress``: rule applies to ingress traffic.
3449 - ``egress``: rule applies to egress traffic.
3450 - ``transfer``: apply rule directly to endpoints found in pattern.
3452 Each instance of an attribute specified several times overrides the previous
3453 value as shown below (group 4 is used)::
3455 testpmd> flow create 0 group 42 group 24 group 4 [...]
3457 Note that once enabled, ``ingress`` and ``egress`` cannot be disabled.
3459 While not specifying a direction is an error, some rules may allow both
3462 Most rules affect RX therefore contain the ``ingress`` token::
3464 testpmd> flow create 0 ingress pattern [...]
3469 A matching pattern starts after the ``pattern`` token. It is made of pattern
3470 items and is terminated by a mandatory ``end`` item.
3472 Items are named after their type (*RTE_FLOW_ITEM_TYPE_* from ``enum
3473 rte_flow_item_type``).
3475 The ``/`` token is used as a separator between pattern items as shown
3478 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end [...]
3480 Note that protocol items like these must be stacked from lowest to highest
3481 layer to make sense. For instance, the following rule is either invalid or
3482 unlikely to match any packet::
3484 testpmd> flow create 0 ingress pattern eth / udp / ipv4 / end [...]
3486 More information on these restrictions can be found in the *rte_flow*
3489 Several items support additional specification structures, for example
3490 ``ipv4`` allows specifying source and destination addresses as follows::
3492 testpmd> flow create 0 ingress pattern eth / ipv4 src is 10.1.1.1
3493 dst is 10.2.0.0 / end [...]
3495 This rule matches all IPv4 traffic with the specified properties.
3497 In this example, ``src`` and ``dst`` are field names of the underlying
3498 ``struct rte_flow_item_ipv4`` object. All item properties can be specified
3499 in a similar fashion.
3501 The ``is`` token means that the subsequent value must be matched exactly,
3502 and assigns ``spec`` and ``mask`` fields in ``struct rte_flow_item``
3503 accordingly. Possible assignment tokens are:
3505 - ``is``: match value perfectly (with full bit-mask).
3506 - ``spec``: match value according to configured bit-mask.
3507 - ``last``: specify upper bound to establish a range.
3508 - ``mask``: specify bit-mask with relevant bits set to one.
3509 - ``prefix``: generate bit-mask with <prefix-length> most-significant bits set to one.
3511 These yield identical results::
3513 ipv4 src is 10.1.1.1
3517 ipv4 src spec 10.1.1.1 src mask 255.255.255.255
3521 ipv4 src spec 10.1.1.1 src prefix 32
3525 ipv4 src is 10.1.1.1 src last 10.1.1.1 # range with a single value
3529 ipv4 src is 10.1.1.1 src last 0 # 0 disables range
3531 Inclusive ranges can be defined with ``last``::
3533 ipv4 src is 10.1.1.1 src last 10.2.3.4 # 10.1.1.1 to 10.2.3.4
3535 Note that ``mask`` affects both ``spec`` and ``last``::
3537 ipv4 src is 10.1.1.1 src last 10.2.3.4 src mask 255.255.0.0
3538 # matches 10.1.0.0 to 10.2.255.255
3540 Properties can be modified multiple times::
3542 ipv4 src is 10.1.1.1 src is 10.1.2.3 src is 10.2.3.4 # matches 10.2.3.4
3546 ipv4 src is 10.1.1.1 src prefix 24 src prefix 16 # matches 10.1.0.0/16
3551 This section lists supported pattern items and their attributes, if any.
3553 - ``end``: end list of pattern items.
3555 - ``void``: no-op pattern item.
3557 - ``invert``: perform actions when pattern does not match.
3559 - ``any``: match any protocol for the current layer.
3561 - ``num {unsigned}``: number of layers covered.
3563 - ``pf``: match traffic from/to the physical function.
3565 - ``vf``: match traffic from/to a virtual function ID.
3567 - ``id {unsigned}``: VF ID.
3569 - ``phy_port``: match traffic from/to a specific physical port.
3571 - ``index {unsigned}``: physical port index.
3573 - ``port_id``: match traffic from/to a given DPDK port ID.
3575 - ``id {unsigned}``: DPDK port ID.
3577 - ``mark``: match value set in previously matched flow rule using the mark action.
3579 - ``id {unsigned}``: arbitrary integer value.
3581 - ``raw``: match an arbitrary byte string.
3583 - ``relative {boolean}``: look for pattern after the previous item.
3584 - ``search {boolean}``: search pattern from offset (see also limit).
3585 - ``offset {integer}``: absolute or relative offset for pattern.
3586 - ``limit {unsigned}``: search area limit for start of pattern.
3587 - ``pattern {string}``: byte string to look for.
3589 - ``eth``: match Ethernet header.
3591 - ``dst {MAC-48}``: destination MAC.
3592 - ``src {MAC-48}``: source MAC.
3593 - ``type {unsigned}``: EtherType or TPID.
3595 - ``vlan``: match 802.1Q/ad VLAN tag.
3597 - ``tci {unsigned}``: tag control information.
3598 - ``pcp {unsigned}``: priority code point.
3599 - ``dei {unsigned}``: drop eligible indicator.
3600 - ``vid {unsigned}``: VLAN identifier.
3601 - ``inner_type {unsigned}``: inner EtherType or TPID.
3603 - ``ipv4``: match IPv4 header.
3605 - ``tos {unsigned}``: type of service.
3606 - ``ttl {unsigned}``: time to live.
3607 - ``proto {unsigned}``: next protocol ID.
3608 - ``src {ipv4 address}``: source address.
3609 - ``dst {ipv4 address}``: destination address.
3611 - ``ipv6``: match IPv6 header.
3613 - ``tc {unsigned}``: traffic class.
3614 - ``flow {unsigned}``: flow label.
3615 - ``proto {unsigned}``: protocol (next header).
3616 - ``hop {unsigned}``: hop limit.
3617 - ``src {ipv6 address}``: source address.
3618 - ``dst {ipv6 address}``: destination address.
3620 - ``icmp``: match ICMP header.
3622 - ``type {unsigned}``: ICMP packet type.
3623 - ``code {unsigned}``: ICMP packet code.
3625 - ``udp``: match UDP header.
3627 - ``src {unsigned}``: UDP source port.
3628 - ``dst {unsigned}``: UDP destination port.
3630 - ``tcp``: match TCP header.
3632 - ``src {unsigned}``: TCP source port.
3633 - ``dst {unsigned}``: TCP destination port.
3635 - ``sctp``: match SCTP header.
3637 - ``src {unsigned}``: SCTP source port.
3638 - ``dst {unsigned}``: SCTP destination port.
3639 - ``tag {unsigned}``: validation tag.
3640 - ``cksum {unsigned}``: checksum.
3642 - ``vxlan``: match VXLAN header.
3644 - ``vni {unsigned}``: VXLAN identifier.
3646 - ``e_tag``: match IEEE 802.1BR E-Tag header.
3648 - ``grp_ecid_b {unsigned}``: GRP and E-CID base.
3650 - ``nvgre``: match NVGRE header.
3652 - ``tni {unsigned}``: virtual subnet ID.
3654 - ``mpls``: match MPLS header.
3656 - ``label {unsigned}``: MPLS label.
3658 - ``gre``: match GRE header.
3660 - ``protocol {unsigned}``: protocol type.
3662 - ``fuzzy``: fuzzy pattern match, expect faster than default.
3664 - ``thresh {unsigned}``: accuracy threshold.
3666 - ``gtp``, ``gtpc``, ``gtpu``: match GTPv1 header.
3668 - ``teid {unsigned}``: tunnel endpoint identifier.
3670 - ``geneve``: match GENEVE header.
3672 - ``vni {unsigned}``: virtual network identifier.
3673 - ``protocol {unsigned}``: protocol type.
3675 - ``vxlan-gpe``: match VXLAN-GPE header.
3677 - ``vni {unsigned}``: VXLAN-GPE identifier.
3679 - ``arp_eth_ipv4``: match ARP header for Ethernet/IPv4.
3681 - ``sha {MAC-48}``: sender hardware address.
3682 - ``spa {ipv4 address}``: sender IPv4 address.
3683 - ``tha {MAC-48}``: target hardware address.
3684 - ``tpa {ipv4 address}``: target IPv4 address.
3686 - ``ipv6_ext``: match presence of any IPv6 extension header.
3688 - ``next_hdr {unsigned}``: next header.
3690 - ``icmp6``: match any ICMPv6 header.
3692 - ``type {unsigned}``: ICMPv6 type.
3693 - ``code {unsigned}``: ICMPv6 code.
3695 - ``icmp6_nd_ns``: match ICMPv6 neighbor discovery solicitation.
3697 - ``target_addr {ipv6 address}``: target address.
3699 - ``icmp6_nd_na``: match ICMPv6 neighbor discovery advertisement.
3701 - ``target_addr {ipv6 address}``: target address.
3703 - ``icmp6_nd_opt``: match presence of any ICMPv6 neighbor discovery option.
3705 - ``type {unsigned}``: ND option type.
3707 - ``icmp6_nd_opt_sla_eth``: match ICMPv6 neighbor discovery source Ethernet
3708 link-layer address option.
3710 - ``sla {MAC-48}``: source Ethernet LLA.
3712 - ``icmp6_nd_opt_sla_eth``: match ICMPv6 neighbor discovery target Ethernet
3713 link-layer address option.
3715 - ``tla {MAC-48}``: target Ethernet LLA.
3717 - ``meta``: match application specific metadata.
3719 - ``data {unsigned}``: metadata value.
3724 A list of actions starts after the ``actions`` token in the same fashion as
3725 `Matching pattern`_; actions are separated by ``/`` tokens and the list is
3726 terminated by a mandatory ``end`` action.
3728 Actions are named after their type (*RTE_FLOW_ACTION_TYPE_* from ``enum
3729 rte_flow_action_type``).
3731 Dropping all incoming UDPv4 packets can be expressed as follows::
3733 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
3736 Several actions have configurable properties which must be specified when
3737 there is no valid default value. For example, ``queue`` requires a target
3740 This rule redirects incoming UDPv4 traffic to queue index 6::
3742 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
3743 actions queue index 6 / end
3745 While this one could be rejected by PMDs (unspecified queue index)::
3747 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
3750 As defined by *rte_flow*, the list is not ordered, all actions of a given
3751 rule are performed simultaneously. These are equivalent::
3753 queue index 6 / void / mark id 42 / end
3757 void / mark id 42 / queue index 6 / end
3759 All actions in a list should have different types, otherwise only the last
3760 action of a given type is taken into account::
3762 queue index 4 / queue index 5 / queue index 6 / end # will use queue 6
3766 drop / drop / drop / end # drop is performed only once
3770 mark id 42 / queue index 3 / mark id 24 / end # mark will be 24
3772 Considering they are performed simultaneously, opposite and overlapping
3773 actions can sometimes be combined when the end result is unambiguous::
3775 drop / queue index 6 / end # drop has no effect
3779 queue index 6 / rss queues 6 7 8 / end # queue has no effect
3783 drop / passthru / end # drop has no effect
3785 Note that PMDs may still refuse such combinations.
3790 This section lists supported actions and their attributes, if any.
3792 - ``end``: end list of actions.
3794 - ``void``: no-op action.
3796 - ``passthru``: let subsequent rule process matched packets.
3798 - ``jump``: redirect traffic to group on device.
3800 - ``group {unsigned}``: group to redirect to.
3802 - ``mark``: attach 32 bit value to packets.
3804 - ``id {unsigned}``: 32 bit value to return with packets.
3806 - ``flag``: flag packets.
3808 - ``queue``: assign packets to a given queue index.
3810 - ``index {unsigned}``: queue index to use.
3812 - ``drop``: drop packets (note: passthru has priority).
3814 - ``count``: enable counters for this rule.
3816 - ``rss``: spread packets among several queues.
3818 - ``func {hash function}``: RSS hash function to apply, allowed tokens are
3819 the same as `set_hash_global_config`_.
3821 - ``level {unsigned}``: encapsulation level for ``types``.
3823 - ``types [{RSS hash type} [...]] end``: specific RSS hash types, allowed
3824 tokens are the same as `set_hash_input_set`_, except that an empty list
3825 does not disable RSS but instead requests unspecified "best-effort"
3828 - ``key {string}``: RSS hash key, overrides ``key_len``.
3830 - ``key_len {unsigned}``: RSS hash key length in bytes, can be used in
3831 conjunction with ``key`` to pad or truncate it.
3833 - ``queues [{unsigned} [...]] end``: queue indices to use.
3835 - ``pf``: direct traffic to physical function.
3837 - ``vf``: direct traffic to a virtual function ID.
3839 - ``original {boolean}``: use original VF ID if possible.
3840 - ``id {unsigned}``: VF ID.
3842 - ``phy_port``: direct packets to physical port index.
3844 - ``original {boolean}``: use original port index if possible.
3845 - ``index {unsigned}``: physical port index.
3847 - ``port_id``: direct matching traffic to a given DPDK port ID.
3849 - ``original {boolean}``: use original DPDK port ID if possible.
3850 - ``id {unsigned}``: DPDK port ID.
3852 - ``of_set_mpls_ttl``: OpenFlow's ``OFPAT_SET_MPLS_TTL``.
3854 - ``mpls_ttl``: MPLS TTL.
3856 - ``of_dec_mpls_ttl``: OpenFlow's ``OFPAT_DEC_MPLS_TTL``.
3858 - ``of_set_nw_ttl``: OpenFlow's ``OFPAT_SET_NW_TTL``.
3860 - ``nw_ttl``: IP TTL.
3862 - ``of_dec_nw_ttl``: OpenFlow's ``OFPAT_DEC_NW_TTL``.
3864 - ``of_copy_ttl_out``: OpenFlow's ``OFPAT_COPY_TTL_OUT``.
3866 - ``of_copy_ttl_in``: OpenFlow's ``OFPAT_COPY_TTL_IN``.
3868 - ``of_pop_vlan``: OpenFlow's ``OFPAT_POP_VLAN``.
3870 - ``of_push_vlan``: OpenFlow's ``OFPAT_PUSH_VLAN``.
3872 - ``ethertype``: Ethertype.
3874 - ``of_set_vlan_vid``: OpenFlow's ``OFPAT_SET_VLAN_VID``.
3876 - ``vlan_vid``: VLAN id.
3878 - ``of_set_vlan_pcp``: OpenFlow's ``OFPAT_SET_VLAN_PCP``.
3880 - ``vlan_pcp``: VLAN priority.
3882 - ``of_pop_mpls``: OpenFlow's ``OFPAT_POP_MPLS``.
3884 - ``ethertype``: Ethertype.
3886 - ``of_push_mpls``: OpenFlow's ``OFPAT_PUSH_MPLS``.
3888 - ``ethertype``: Ethertype.
3890 - ``vxlan_encap``: Performs a VXLAN encapsulation, outer layer configuration
3891 is done through `Config VXLAN Encap outer layers`_.
3893 - ``vxlan_decap``: Performs a decapsulation action by stripping all headers of
3894 the VXLAN tunnel network overlay from the matched flow.
3896 - ``nvgre_encap``: Performs a NVGRE encapsulation, outer layer configuration
3897 is done through `Config NVGRE Encap outer layers`_.
3899 - ``nvgre_decap``: Performs a decapsulation action by stripping all headers of
3900 the NVGRE tunnel network overlay from the matched flow.
3902 - ``l2_encap``: Performs a L2 encapsulation, L2 configuration
3903 is done through `Config L2 Encap`_.
3905 - ``l2_decap``: Performs a L2 decapsulation, L2 configuration
3906 is done through `Config L2 Decap`_.
3908 - ``mplsogre_encap``: Performs a MPLSoGRE encapsulation, outer layer
3909 configuration is done through `Config MPLSoGRE Encap outer layers`_.
3911 - ``mplsogre_decap``: Performs a MPLSoGRE decapsulation, outer layer
3912 configuration is done through `Config MPLSoGRE Decap outer layers`_.
3914 - ``mplsoudp_encap``: Performs a MPLSoUDP encapsulation, outer layer
3915 configuration is done through `Config MPLSoUDP Encap outer layers`_.
3917 - ``mplsoudp_decap``: Performs a MPLSoUDP decapsulation, outer layer
3918 configuration is done through `Config MPLSoUDP Decap outer layers`_.
3920 - ``set_ipv4_src``: Set a new IPv4 source address in the outermost IPv4 header.
3922 - ``ipv4_addr``: New IPv4 source address.
3924 - ``set_ipv4_dst``: Set a new IPv4 destination address in the outermost IPv4
3927 - ``ipv4_addr``: New IPv4 destination address.
3929 - ``set_ipv6_src``: Set a new IPv6 source address in the outermost IPv6 header.
3931 - ``ipv6_addr``: New IPv6 source address.
3933 - ``set_ipv6_dst``: Set a new IPv6 destination address in the outermost IPv6
3936 - ``ipv6_addr``: New IPv6 destination address.
3938 - ``set_tp_src``: Set a new source port number in the outermost TCP/UDP
3941 - ``port``: New TCP/UDP source port number.
3943 - ``set_tp_dst``: Set a new destination port number in the outermost TCP/UDP
3946 - ``port``: New TCP/UDP destination port number.
3948 - ``mac_swap``: Swap the source and destination MAC addresses in the outermost
3951 - ``dec_ttl``: Performs a decrease TTL value action
3953 - ``set_ttl``: Set TTL value with specificed value
3954 - ``ttl_value {unsigned}``: The new TTL value to be set
3956 - ``set_mac_src``: set source MAC address
3958 - ``mac_addr {MAC-48}``: new source MAC address
3960 - ``set_mac_dst``: set destination MAC address
3962 - ``mac_addr {MAC-48}``: new destination MAC address
3964 Destroying flow rules
3965 ~~~~~~~~~~~~~~~~~~~~~
3967 ``flow destroy`` destroys one or more rules from their rule ID (as returned
3968 by ``flow create``), this command calls ``rte_flow_destroy()`` as many
3969 times as necessary::
3971 flow destroy {port_id} rule {rule_id} [...]
3973 If successful, it will show::
3975 Flow rule #[...] destroyed
3977 It does not report anything for rule IDs that do not exist. The usual error
3978 message is shown when a rule cannot be destroyed::
3980 Caught error type [...] ([...]): [...]
3982 ``flow flush`` destroys all rules on a device and does not take extra
3983 arguments. It is bound to ``rte_flow_flush()``::
3985 flow flush {port_id}
3987 Any errors are reported as above.
3989 Creating several rules and destroying them::
3991 testpmd> flow create 0 ingress pattern eth / ipv6 / end
3992 actions queue index 2 / end
3993 Flow rule #0 created
3994 testpmd> flow create 0 ingress pattern eth / ipv4 / end
3995 actions queue index 3 / end
3996 Flow rule #1 created
3997 testpmd> flow destroy 0 rule 0 rule 1
3998 Flow rule #1 destroyed
3999 Flow rule #0 destroyed
4002 The same result can be achieved using ``flow flush``::
4004 testpmd> flow create 0 ingress pattern eth / ipv6 / end
4005 actions queue index 2 / end
4006 Flow rule #0 created
4007 testpmd> flow create 0 ingress pattern eth / ipv4 / end
4008 actions queue index 3 / end
4009 Flow rule #1 created
4010 testpmd> flow flush 0
4013 Non-existent rule IDs are ignored::
4015 testpmd> flow create 0 ingress pattern eth / ipv6 / end
4016 actions queue index 2 / end
4017 Flow rule #0 created
4018 testpmd> flow create 0 ingress pattern eth / ipv4 / end
4019 actions queue index 3 / end
4020 Flow rule #1 created
4021 testpmd> flow destroy 0 rule 42 rule 10 rule 2
4023 testpmd> flow destroy 0 rule 0
4024 Flow rule #0 destroyed
4030 ``flow query`` queries a specific action of a flow rule having that
4031 ability. Such actions collect information that can be reported using this
4032 command. It is bound to ``rte_flow_query()``::
4034 flow query {port_id} {rule_id} {action}
4036 If successful, it will display either the retrieved data for known actions
4037 or the following message::
4039 Cannot display result for action type [...] ([...])
4041 Otherwise, it will complain either that the rule does not exist or that some
4044 Flow rule #[...] not found
4048 Caught error type [...] ([...]): [...]
4050 Currently only the ``count`` action is supported. This action reports the
4051 number of packets that hit the flow rule and the total number of bytes. Its
4052 output has the following format::
4055 hits_set: [...] # whether "hits" contains a valid value
4056 bytes_set: [...] # whether "bytes" contains a valid value
4057 hits: [...] # number of packets
4058 bytes: [...] # number of bytes
4060 Querying counters for TCPv6 packets redirected to queue 6::
4062 testpmd> flow create 0 ingress pattern eth / ipv6 / tcp / end
4063 actions queue index 6 / count / end
4064 Flow rule #4 created
4065 testpmd> flow query 0 4 count
4076 ``flow list`` lists existing flow rules sorted by priority and optionally
4077 filtered by group identifiers::
4079 flow list {port_id} [group {group_id}] [...]
4081 This command only fails with the following message if the device does not
4086 Output consists of a header line followed by a short description of each
4087 flow rule, one per line. There is no output at all when no flow rules are
4088 configured on the device::
4090 ID Group Prio Attr Rule
4091 [...] [...] [...] [...] [...]
4093 ``Attr`` column flags:
4095 - ``i`` for ``ingress``.
4096 - ``e`` for ``egress``.
4098 Creating several flow rules and listing them::
4100 testpmd> flow create 0 ingress pattern eth / ipv4 / end
4101 actions queue index 6 / end
4102 Flow rule #0 created
4103 testpmd> flow create 0 ingress pattern eth / ipv6 / end
4104 actions queue index 2 / end
4105 Flow rule #1 created
4106 testpmd> flow create 0 priority 5 ingress pattern eth / ipv4 / udp / end
4107 actions rss queues 6 7 8 end / end
4108 Flow rule #2 created
4109 testpmd> flow list 0
4110 ID Group Prio Attr Rule
4111 0 0 0 i- ETH IPV4 => QUEUE
4112 1 0 0 i- ETH IPV6 => QUEUE
4113 2 0 5 i- ETH IPV4 UDP => RSS
4116 Rules are sorted by priority (i.e. group ID first, then priority level)::
4118 testpmd> flow list 1
4119 ID Group Prio Attr Rule
4120 0 0 0 i- ETH => COUNT
4121 6 0 500 i- ETH IPV6 TCP => DROP COUNT
4122 5 0 1000 i- ETH IPV6 ICMP => QUEUE
4123 1 24 0 i- ETH IPV4 UDP => QUEUE
4124 4 24 10 i- ETH IPV4 TCP => DROP
4125 3 24 20 i- ETH IPV4 => DROP
4126 2 24 42 i- ETH IPV4 UDP => QUEUE
4127 7 63 0 i- ETH IPV6 UDP VXLAN => MARK QUEUE
4130 Output can be limited to specific groups::
4132 testpmd> flow list 1 group 0 group 63
4133 ID Group Prio Attr Rule
4134 0 0 0 i- ETH => COUNT
4135 6 0 500 i- ETH IPV6 TCP => DROP COUNT
4136 5 0 1000 i- ETH IPV6 ICMP => QUEUE
4137 7 63 0 i- ETH IPV6 UDP VXLAN => MARK QUEUE
4140 Toggling isolated mode
4141 ~~~~~~~~~~~~~~~~~~~~~~
4143 ``flow isolate`` can be used to tell the underlying PMD that ingress traffic
4144 must only be injected from the defined flow rules; that no default traffic
4145 is expected outside those rules and the driver is free to assign more
4146 resources to handle them. It is bound to ``rte_flow_isolate()``::
4148 flow isolate {port_id} {boolean}
4150 If successful, enabling or disabling isolated mode shows either::
4152 Ingress traffic on port [...]
4153 is now restricted to the defined flow rules
4157 Ingress traffic on port [...]
4158 is not restricted anymore to the defined flow rules
4160 Otherwise, in case of error::
4162 Caught error type [...] ([...]): [...]
4164 Mainly due to its side effects, PMDs supporting this mode may not have the
4165 ability to toggle it more than once without reinitializing affected ports
4166 first (e.g. by exiting testpmd).
4168 Enabling isolated mode::
4170 testpmd> flow isolate 0 true
4171 Ingress traffic on port 0 is now restricted to the defined flow rules
4174 Disabling isolated mode::
4176 testpmd> flow isolate 0 false
4177 Ingress traffic on port 0 is not restricted anymore to the defined flow rules
4180 Sample QinQ flow rules
4181 ~~~~~~~~~~~~~~~~~~~~~~
4183 Before creating QinQ rule(s) the following commands should be issued to enable QinQ::
4185 testpmd> port stop 0
4186 testpmd> vlan set qinq on 0
4188 The above command sets the inner and outer TPID's to 0x8100.
4190 To change the TPID's the following commands should be used::
4192 testpmd> vlan set outer tpid 0xa100 0
4193 testpmd> vlan set inner tpid 0x9100 0
4194 testpmd> port start 0
4196 Validate and create a QinQ rule on port 0 to steer traffic to a VF queue in a VM.
4200 testpmd> flow validate 0 ingress pattern eth / vlan tci is 123 /
4201 vlan tci is 456 / end actions vf id 1 / queue index 0 / end
4202 Flow rule #0 validated
4204 testpmd> flow create 0 ingress pattern eth / vlan tci is 4 /
4205 vlan tci is 456 / end actions vf id 123 / queue index 0 / end
4206 Flow rule #0 created
4208 testpmd> flow list 0
4209 ID Group Prio Attr Rule
4210 0 0 0 i- ETH VLAN VLAN=>VF QUEUE
4212 Validate and create a QinQ rule on port 0 to steer traffic to a queue on the host.
4216 testpmd> flow validate 0 ingress pattern eth / vlan tci is 321 /
4217 vlan tci is 654 / end actions pf / queue index 0 / end
4218 Flow rule #1 validated
4220 testpmd> flow create 0 ingress pattern eth / vlan tci is 321 /
4221 vlan tci is 654 / end actions pf / queue index 1 / end
4222 Flow rule #1 created
4224 testpmd> flow list 0
4225 ID Group Prio Attr Rule
4226 0 0 0 i- ETH VLAN VLAN=>VF QUEUE
4227 1 0 0 i- ETH VLAN VLAN=>PF QUEUE
4229 Sample VXLAN encapsulation rule
4230 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4232 VXLAN encapsulation outer layer has default value pre-configured in testpmd
4233 source code, those can be changed by using the following commands
4235 IPv4 VXLAN outer header::
4237 testpmd> set vxlan ip-version ipv4 vni 4 udp-src 4 udp-dst 4 ip-src 127.0.0.1
4238 ip-dst 128.0.0.1 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4239 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4242 testpmd> set vxlan-with-vlan ip-version ipv4 vni 4 udp-src 4 udp-dst 4 ip-src
4243 127.0.0.1 ip-dst 128.0.0.1 vlan-tci 34 eth-src 11:11:11:11:11:11
4244 eth-dst 22:22:22:22:22:22
4245 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4248 testpmd> set vxlan-tos-ttl ip-version ipv4 vni 4 udp-src 4 udp-dst 4 ip-tos 0
4249 ip-ttl 255 ip-src 127.0.0.1 ip-dst 128.0.0.1 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 IPv6 VXLAN outer header::
4256 testpmd> set vxlan ip-version ipv6 vni 4 udp-src 4 udp-dst 4 ip-src ::1
4257 ip-dst ::2222 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4258 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4261 testpmd> set vxlan-with-vlan ip-version ipv6 vni 4 udp-src 4 udp-dst 4
4262 ip-src ::1 ip-dst ::2222 vlan-tci 34 eth-src 11:11:11:11:11:11
4263 eth-dst 22:22:22:22:22:22
4264 testpmd> flow create 0 ingress pattern end actions vxlan_encap /
4267 testpmd> set vxlan-tos-ttl ip-version ipv6 vni 4 udp-src 4 udp-dst 4
4268 ip-tos 0 ip-ttl 255 ::1 ip-dst ::2222 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 vxlan_encap /
4273 Sample NVGRE encapsulation rule
4274 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4276 NVGRE encapsulation outer layer has default value pre-configured in testpmd
4277 source code, those can be changed by using the following commands
4279 IPv4 NVGRE outer header::
4281 testpmd> set nvgre ip-version ipv4 tni 4 ip-src 127.0.0.1 ip-dst 128.0.0.1
4282 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4283 testpmd> flow create 0 ingress pattern end actions nvgre_encap /
4286 testpmd> set nvgre-with-vlan ip-version ipv4 tni 4 ip-src 127.0.0.1
4287 ip-dst 128.0.0.1 vlan-tci 34 eth-src 11:11:11:11:11:11
4288 eth-dst 22:22:22:22:22:22
4289 testpmd> flow create 0 ingress pattern end actions nvgre_encap /
4292 IPv6 NVGRE outer header::
4294 testpmd> set nvgre ip-version ipv6 tni 4 ip-src ::1 ip-dst ::2222
4295 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4296 testpmd> flow create 0 ingress pattern end actions nvgre_encap /
4299 testpmd> set nvgre-with-vlan ip-version ipv6 tni 4 ip-src ::1 ip-dst ::2222
4300 vlan-tci 34 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4301 testpmd> flow create 0 ingress pattern end actions nvgre_encap /
4304 Sample L2 encapsulation rule
4305 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4307 L2 encapsulation has default value pre-configured in testpmd
4308 source code, those can be changed by using the following commands
4312 testpmd> set l2_encap ip-version ipv4
4313 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4314 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / mpls / end actions
4315 mplsoudp_decap / l2_encap / end
4317 L2 with VXLAN header::
4319 testpmd> set l2_encap-with-vlan ip-version ipv4 vlan-tci 34
4320 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4321 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / mpls / end actions
4322 mplsoudp_decap / l2_encap / end
4324 Sample L2 decapsulation rule
4325 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4327 L2 decapsulation has default value pre-configured in testpmd
4328 source code, those can be changed by using the following commands
4332 testpmd> set l2_decap
4333 testpmd> flow create 0 egress pattern eth / end actions l2_decap / mplsoudp_encap /
4336 L2 with VXLAN header::
4338 testpmd> set l2_encap-with-vlan
4339 testpmd> flow create 0 egress pattern eth / end actions l2_encap / mplsoudp_encap /
4342 Sample MPLSoGRE encapsulation rule
4343 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4345 MPLSoGRE encapsulation outer layer has default value pre-configured in testpmd
4346 source code, those can be changed by using the following commands
4348 IPv4 MPLSoGRE outer header::
4350 testpmd> set mplsogre_encap ip-version ipv4 label 4
4351 ip-src 127.0.0.1 ip-dst 128.0.0.1 eth-src 11:11:11:11:11:11
4352 eth-dst 22:22:22:22:22:22
4353 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4354 mplsogre_encap / end
4356 IPv4 MPLSoGRE with VLAN outer header::
4358 testpmd> set mplsogre_encap-with-vlan ip-version ipv4 label 4
4359 ip-src 127.0.0.1 ip-dst 128.0.0.1 vlan-tci 34
4360 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4361 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4362 mplsogre_encap / end
4364 IPv6 MPLSoGRE outer header::
4366 testpmd> set mplsogre_encap ip-version ipv6 mask 4
4367 ip-src ::1 ip-dst ::2222 eth-src 11:11:11:11:11:11
4368 eth-dst 22:22:22:22:22:22
4369 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4370 mplsogre_encap / end
4372 IPv6 MPLSoGRE with VLAN outer header::
4374 testpmd> set mplsogre_encap-with-vlan ip-version ipv6 mask 4
4375 ip-src ::1 ip-dst ::2222 vlan-tci 34
4376 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4377 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4378 mplsogre_encap / end
4380 Sample MPLSoGRE decapsulation rule
4381 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4383 MPLSoGRE decapsulation outer layer has default value pre-configured in testpmd
4384 source code, those can be changed by using the following commands
4386 IPv4 MPLSoGRE outer header::
4388 testpmd> set mplsogre_decap ip-version ipv4
4389 testpmd> flow create 0 ingress pattern eth / ipv4 / gre / mpls / end actions
4390 mplsogre_decap / l2_encap / end
4392 IPv4 MPLSoGRE with VLAN outer header::
4394 testpmd> set mplsogre_decap-with-vlan ip-version ipv4
4395 testpmd> flow create 0 ingress pattern eth / vlan / ipv4 / gre / mpls / end
4396 actions mplsogre_decap / l2_encap / end
4398 IPv6 MPLSoGRE outer header::
4400 testpmd> set mplsogre_decap ip-version ipv6
4401 testpmd> flow create 0 ingress pattern eth / ipv6 / gre / mpls / end
4402 actions mplsogre_decap / l2_encap / end
4404 IPv6 MPLSoGRE with VLAN outer header::
4406 testpmd> set mplsogre_decap-with-vlan ip-version ipv6
4407 testpmd> flow create 0 ingress pattern eth / vlan / ipv6 / gre / mpls / end
4408 actions mplsogre_decap / l2_encap / end
4410 Sample MPLSoUDP encapsulation rule
4411 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4413 MPLSoUDP encapsulation outer layer has default value pre-configured in testpmd
4414 source code, those can be changed by using the following commands
4416 IPv4 MPLSoUDP outer header::
4418 testpmd> set mplsoudp_encap ip-version ipv4 label 4 udp-src 5 udp-dst 10
4419 ip-src 127.0.0.1 ip-dst 128.0.0.1 eth-src 11:11:11:11:11:11
4420 eth-dst 22:22:22:22:22:22
4421 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4422 mplsoudp_encap / end
4424 IPv4 MPLSoUDP with VLAN outer header::
4426 testpmd> set mplsoudp_encap-with-vlan ip-version ipv4 label 4 udp-src 5
4427 udp-dst 10 ip-src 127.0.0.1 ip-dst 128.0.0.1 vlan-tci 34
4428 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4429 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4430 mplsoudp_encap / end
4432 IPv6 MPLSoUDP outer header::
4434 testpmd> set mplsoudp_encap ip-version ipv6 mask 4 udp-src 5 udp-dst 10
4435 ip-src ::1 ip-dst ::2222 eth-src 11:11:11:11:11:11
4436 eth-dst 22:22:22:22:22:22
4437 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4438 mplsoudp_encap / end
4440 IPv6 MPLSoUDP with VLAN outer header::
4442 testpmd> set mplsoudp_encap-with-vlan ip-version ipv6 mask 4 udp-src 5
4443 udp-dst 10 ip-src ::1 ip-dst ::2222 vlan-tci 34
4444 eth-src 11:11:11:11:11:11 eth-dst 22:22:22:22:22:22
4445 testpmd> flow create 0 egress pattern eth / end actions l2_decap /
4446 mplsoudp_encap / end
4448 Sample MPLSoUDP decapsulation rule
4449 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4451 MPLSoUDP decapsulation outer layer has default value pre-configured in testpmd
4452 source code, those can be changed by using the following commands
4454 IPv4 MPLSoUDP outer header::
4456 testpmd> set mplsoudp_decap ip-version ipv4
4457 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / mpls / end actions
4458 mplsoudp_decap / l2_encap / end
4460 IPv4 MPLSoUDP with VLAN outer header::
4462 testpmd> set mplsoudp_decap-with-vlan ip-version ipv4
4463 testpmd> flow create 0 ingress pattern eth / vlan / ipv4 / udp / mpls / end
4464 actions mplsoudp_decap / l2_encap / end
4466 IPv6 MPLSoUDP outer header::
4468 testpmd> set mplsoudp_decap ip-version ipv6
4469 testpmd> flow create 0 ingress pattern eth / ipv6 / udp / mpls / end
4470 actions mplsoudp_decap / l2_encap / end
4472 IPv6 MPLSoUDP with VLAN outer header::
4474 testpmd> set mplsoudp_decap-with-vlan ip-version ipv6
4475 testpmd> flow create 0 ingress pattern eth / vlan / ipv6 / udp / mpls / end
4476 actions mplsoudp_decap / l2_encap / end
4481 The following sections show functions to load/unload eBPF based filters.
4486 Load an eBPF program as a callback for partciular RX/TX queue::
4488 testpmd> bpf-load rx|tx (portid) (queueid) (load-flags) (bpf-prog-filename)
4490 The available load-flags are:
4492 * ``J``: use JIT generated native code, otherwise BPF interpreter will be used.
4494 * ``M``: assume input parameter is a pointer to rte_mbuf, otherwise assume it is a pointer to first segment's data.
4500 You'll need clang v3.7 or above to build bpf program you'd like to load
4504 .. code-block:: console
4507 clang -O2 -target bpf -c t1.c
4509 Then to load (and JIT compile) t1.o at RX queue 0, port 1::
4511 .. code-block:: console
4513 testpmd> bpf-load rx 1 0 J ./dpdk.org/test/bpf/t1.o
4515 To load (not JITed) t1.o at TX queue 0, port 0::
4517 .. code-block:: console
4519 testpmd> bpf-load tx 0 0 - ./dpdk.org/test/bpf/t1.o
4524 Unload previously loaded eBPF program for partciular RX/TX queue::
4526 testpmd> bpf-unload rx|tx (portid) (queueid)
4528 For example to unload BPF filter from TX queue 0, port 0:
4530 .. code-block:: console
4532 testpmd> bpf-load tx 0 0 - ./dpdk.org/test/bpf/t1.o