2 Copyright(c) 2010-2016 Intel Corporation. All rights reserved.
5 Redistribution and use in source and binary forms, with or without
6 modification, are permitted provided that the following conditions
9 * Redistributions of source code must retain the above copyright
10 notice, this list of conditions and the following disclaimer.
11 * Redistributions in binary form must reproduce the above copyright
12 notice, this list of conditions and the following disclaimer in
13 the documentation and/or other materials provided with the
15 * Neither the name of Intel Corporation nor the names of its
16 contributors may be used to endorse or promote products derived
17 from this software without specific prior written permission.
19 THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
20 "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
21 LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
22 A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
23 OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
24 SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
25 LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
26 DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
27 THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
28 (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
29 OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
33 Testpmd Runtime Functions
34 =========================
36 Where the testpmd application is started in interactive mode, (``-i|--interactive``),
37 it displays a prompt that can be used to start and stop forwarding,
38 configure the application, display statistics (including the extended NIC
39 statistics aka xstats) , set the Flow Director and other tasks::
43 The testpmd prompt has some, limited, readline support.
44 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
45 as well as access to the command history via the up-arrow.
47 There is also support for tab completion.
48 If you type a partial command and hit ``<TAB>`` you get a list of the available completions:
50 .. code-block:: console
52 testpmd> show port <TAB>
54 info [Mul-choice STRING]: show|clear port info|stats|xstats|fdir|stat_qmap|dcb_tc X
55 info [Mul-choice STRING]: show|clear port info|stats|xstats|fdir|stat_qmap|dcb_tc all
56 stats [Mul-choice STRING]: show|clear port info|stats|xstats|fdir|stat_qmap|dcb_tc X
57 stats [Mul-choice STRING]: show|clear port info|stats|xstats|fdir|stat_qmap|dcb_tc all
63 Some examples in this document are too long to fit on one line are are shown wrapped at `"\\"` for display purposes::
65 testpmd> set flow_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \
66 (pause_time) (send_xon) (port_id)
68 In the real ``testpmd>`` prompt these commands should be on a single line.
73 The testpmd has on-line help for the functions that are available at runtime.
74 These are divided into sections and can be accessed using help, help section or help all:
76 .. code-block:: console
80 help control : Start and stop forwarding.
81 help display : Displaying port, stats and config information.
82 help config : Configuration information.
83 help ports : Configuring ports.
84 help registers : Reading and setting port registers.
85 help filters : Filters configuration help.
86 help all : All of the above sections.
95 Start packet forwarding with current configuration::
102 Start packet forwarding with current configuration after sending specified number of bursts of packets::
104 testpmd> start tx_first (""|burst_num)
106 The default burst number is 1 when ``burst_num`` not presented.
111 Stop packet forwarding, and display accumulated statistics::
126 The functions in the following sections are used to display information about the
127 testpmd configuration or the NIC status.
132 Display information for a given port or all ports::
134 testpmd> show port (info|stats|xstats|fdir|stat_qmap|dcb_tc) (port_id|all)
136 The available information categories are:
138 * ``info``: General port information such as MAC address.
140 * ``stats``: RX/TX statistics.
142 * ``xstats``: RX/TX extended NIC statistics.
144 * ``fdir``: Flow Director information and statistics.
146 * ``stat_qmap``: Queue statistics mapping.
148 * ``dcb_tc``: DCB information such as TC mapping.
152 .. code-block:: console
154 testpmd> show port info 0
156 ********************* Infos for port 0 *********************
158 MAC address: XX:XX:XX:XX:XX:XX
160 memory allocation on the socket: 0
162 Link speed: 40000 Mbps
163 Link duplex: full-duplex
164 Promiscuous mode: enabled
165 Allmulticast mode: disabled
166 Maximum number of MAC addresses: 64
167 Maximum number of MAC addresses of hash filtering: 0
172 Redirection table size: 512
173 Supported flow types:
193 Display the rss redirection table entry indicated by masks on port X::
195 testpmd> show port (port_id) rss reta (size) (mask0, mask1...)
197 size is used to indicate the hardware supported reta size
202 Display the RSS hash functions and RSS hash key of a port::
204 testpmd> show port (port_id) rss-hash ipv4|ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp|ipv4-other|ipv6|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other|l2-payload|ipv6-ex|ipv6-tcp-ex|ipv6-udp-ex [key]
209 Clear the port statistics for a given port or for all ports::
211 testpmd> clear port (info|stats|xstats|fdir|stat_qmap) (port_id|all)
215 testpmd> clear port stats all
220 Display information for a given port's RX/TX queue::
222 testpmd> show (rxq|txq) info (port_id) (queue_id)
227 Displays the configuration of the application.
228 The configuration comes from the command-line, the runtime or the application defaults::
230 testpmd> show config (rxtx|cores|fwd|txpkts)
232 The available information categories are:
234 * ``rxtx``: RX/TX configuration items.
236 * ``cores``: List of forwarding cores.
238 * ``fwd``: Packet forwarding configuration.
240 * ``txpkts``: Packets to TX configuration.
244 .. code-block:: console
246 testpmd> show config rxtx
248 io packet forwarding - CRC stripping disabled - packets/burst=16
249 nb forwarding cores=2 - nb forwarding ports=1
250 RX queues=1 - RX desc=128 - RX free threshold=0
251 RX threshold registers: pthresh=8 hthresh=8 wthresh=4
252 TX queues=1 - TX desc=512 - TX free threshold=0
253 TX threshold registers: pthresh=36 hthresh=0 wthresh=0
254 TX RS bit threshold=0 - TXQ flags=0x0
259 Set the packet forwarding mode::
261 testpmd> set fwd (io|mac|macswap|flowgen| \
262 rxonly|txonly|csum|icmpecho) (""|retry)
264 ``retry`` can be specified for forwarding engines except ``rx_only``.
266 The available information categories are:
268 * ``io``: Forwards packets "as-is" in I/O mode.
269 This is the fastest possible forwarding operation as it does not access packets data.
270 This is the default mode.
272 * ``mac``: Changes the source and the destination Ethernet addresses of packets before forwarding them.
273 Default application behaviour is to set source Ethernet address to that of the transmitting interface, and destination
274 address to a dummy value (set during init). The user may specify a target destination Ethernet address via the 'eth-peer' or
275 'eth-peer-configfile' command-line options. It is not currently possible to specify a specific source Ethernet address.
277 * ``macswap``: MAC swap forwarding mode.
278 Swaps the source and the destination Ethernet addresses of packets before forwarding them.
280 * ``flowgen``: Multi-flow generation mode.
281 Originates a number of flows (with varying destination IP addresses), and terminate receive traffic.
283 * ``rxonly``: Receives packets but doesn't transmit them.
285 * ``txonly``: Generates and transmits packets without receiving any.
287 * ``csum``: Changes the checksum field with hardware or software methods depending on the offload flags on the packet.
289 * ``icmpecho``: Receives a burst of packets, lookup for IMCP echo requests and, if any, send back ICMP echo replies.
291 * ``ieee1588``: Demonstrate L2 IEEE1588 V2 PTP timestamping for RX and TX. Requires ``CONFIG_RTE_LIBRTE_IEEE1588=y``.
293 Note: TX timestamping is only available in the "Full Featured" TX path. To force ``testpmd`` into this mode set ``--txqflags=0``.
297 testpmd> set fwd rxonly
299 Set rxonly packet forwarding mode
305 Display an RX descriptor for a port RX queue::
307 testpmd> read rxd (port_id) (queue_id) (rxd_id)
311 testpmd> read rxd 0 0 4
312 0x0000000B - 0x001D0180 / 0x0000000B - 0x001D0180
317 Display a TX descriptor for a port TX queue::
319 testpmd> read txd (port_id) (queue_id) (txd_id)
323 testpmd> read txd 0 0 4
324 0x00000001 - 0x24C3C440 / 0x000F0000 - 0x2330003C
327 Configuration Functions
328 -----------------------
330 The testpmd application can be configured from the runtime as well as from the command-line.
332 This section details the available configuration functions that are available.
336 Configuration changes only become active when forwarding is started/restarted.
341 Reset forwarding to the default configuration::
348 Set the debug verbosity level::
350 testpmd> set verbose (level)
352 Currently the only available levels are 0 (silent except for error) and 1 (fully verbose).
357 Set the number of ports used by the application:
361 This is equivalent to the ``--nb-ports`` command-line option.
366 Set the number of cores used by the application::
368 testpmd> set nbcore (num)
370 This is equivalent to the ``--nb-cores`` command-line option.
374 The number of cores used must not be greater than number of ports used multiplied by the number of queues per port.
379 Set the forwarding cores hexadecimal mask::
381 testpmd> set coremask (mask)
383 This is equivalent to the ``--coremask`` command-line option.
387 The master lcore is reserved for command line parsing only and cannot be masked on for packet forwarding.
392 Set the forwarding ports hexadecimal mask::
394 testpmd> set portmask (mask)
396 This is equivalent to the ``--portmask`` command-line option.
401 Set number of packets per burst::
403 testpmd> set burst (num)
405 This is equivalent to the ``--burst command-line`` option.
407 When retry is enabled, the transmit delay time and number of retries can also be set::
409 testpmd> set burst tx delay (microseconds) retry (num)
414 Set the length of each segment of the TX-ONLY packets or length of packet for FLOWGEN mode::
416 testpmd> set txpkts (x[,y]*)
418 Where x[,y]* represents a CSV list of values, without white space.
423 Set the split policy for the TX packets, applicable for TX-ONLY and CSUM forwarding modes::
425 testpmd> set txsplit (off|on|rand)
429 * ``off`` disable packet copy & split for CSUM mode.
431 * ``on`` split outgoing packet into multiple segments. Size of each segment
432 and number of segments per packet is determined by ``set txpkts`` command
435 * ``rand`` same as 'on', but number of segments per each packet is a random value between 1 and total number of segments.
440 Set the list of forwarding cores::
442 testpmd> set corelist (x[,y]*)
444 For example, to change the forwarding cores:
446 .. code-block:: console
448 testpmd> set corelist 3,1
449 testpmd> show config fwd
451 io packet forwarding - ports=2 - cores=2 - streams=2 - NUMA support disabled
452 Logical Core 3 (socket 0) forwards packets on 1 streams:
453 RX P=0/Q=0 (socket 0) -> TX P=1/Q=0 (socket 0) peer=02:00:00:00:00:01
454 Logical Core 1 (socket 0) forwards packets on 1 streams:
455 RX P=1/Q=0 (socket 0) -> TX P=0/Q=0 (socket 0) peer=02:00:00:00:00:00
459 The cores are used in the same order as specified on the command line.
464 Set the list of forwarding ports::
466 testpmd> set portlist (x[,y]*)
468 For example, to change the port forwarding:
470 .. code-block:: console
472 testpmd> set portlist 0,2,1,3
473 testpmd> show config fwd
475 io packet forwarding - ports=4 - cores=1 - streams=4
476 Logical Core 3 (socket 0) forwards packets on 4 streams:
477 RX P=0/Q=0 (socket 0) -> TX P=2/Q=0 (socket 0) peer=02:00:00:00:00:01
478 RX P=2/Q=0 (socket 0) -> TX P=0/Q=0 (socket 0) peer=02:00:00:00:00:00
479 RX P=1/Q=0 (socket 0) -> TX P=3/Q=0 (socket 0) peer=02:00:00:00:00:03
480 RX P=3/Q=0 (socket 0) -> TX P=1/Q=0 (socket 0) peer=02:00:00:00:00:02
485 Enable/disable tx loopback::
487 testpmd> set tx loopback (port_id) (on|off)
492 set drop enable bit for all queues::
494 testpmd> set all queues drop (port_id) (on|off)
496 set split drop enable (for VF)
497 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
499 set split drop enable bit for VF from PF::
501 testpmd> set vf split drop (port_id) (vf_id) (on|off)
503 set mac antispoof (for VF)
504 ~~~~~~~~~~~~~~~~~~~~~~~~~~
506 Set mac antispoof for a VF from the PF::
508 testpmd> set vf mac antispoof (port_id) (vf_id) (on|off)
513 Enable/disable MACsec offload::
515 testpmd> set macsec offload (port_id) on encrypt (on|off) replay-protect (on|off)
516 testpmd> set macsec offload (port_id) off
521 Configure MACsec secure connection (SC)::
523 testpmd> set macsec sc (tx|rx) (port_id) (mac) (pi)
527 The pi argument is ignored for tx.
528 Check the NIC Datasheet for hardware limits.
533 Configure MACsec secure association (SA)::
535 testpmd> set macsec sa (tx|rx) (port_id) (idx) (an) (pn) (key)
539 The IDX value must be 0 or 1.
540 Check the NIC Datasheet for hardware limits.
545 Set the VLAN strip on a port::
547 testpmd> vlan set strip (on|off) (port_id)
552 Set the VLAN strip for a queue on a port::
554 testpmd> vlan set stripq (on|off) (port_id,queue_id)
556 vlan set stripq (for VF)
557 ~~~~~~~~~~~~~~~~~~~~~~~~
559 Set VLAN strip for all queues in a pool for a VF from the PF::
561 testpmd> set vf vlan stripq (port_id) (vf_id) (on|off)
563 vlan set insert (for VF)
564 ~~~~~~~~~~~~~~~~~~~~~~~~
566 Set VLAN insert for a VF from the PF::
568 testpmd> set vf vlan insert (port_id) (vf_id) (vlan_id)
570 vlan set antispoof (for VF)
571 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
573 Set VLAN antispoof for a VF from the PF::
575 testpmd> set vf vlan antispoof (port_id) (vf_id) (on|off)
580 Set the VLAN filter on a port::
582 testpmd> vlan set filter (on|off) (port_id)
587 Set the VLAN QinQ (extended queue in queue) on for a port::
589 testpmd> vlan set qinq (on|off) (port_id)
594 Set the inner or outer VLAN TPID for packet filtering on a port::
596 testpmd> vlan set (inner|outer) tpid (value) (port_id)
600 TPID value must be a 16-bit number (value <= 65536).
605 Add a VLAN ID, or all identifiers, to the set of VLAN identifiers filtered by port ID::
607 testpmd> rx_vlan add (vlan_id|all) (port_id)
611 VLAN filter must be set on that port. VLAN ID < 4096.
612 Depending on the NIC used, number of vlan_ids may be limited to the maximum entries
613 in VFTA table. This is important if enabling all vlan_ids.
618 Remove a VLAN ID, or all identifiers, from the set of VLAN identifiers filtered by port ID::
620 testpmd> rx_vlan rm (vlan_id|all) (port_id)
625 Add a VLAN ID, to the set of VLAN identifiers filtered for VF(s) for port ID::
627 testpmd> rx_vlan add (vlan_id) port (port_id) vf (vf_mask)
632 Remove a VLAN ID, from the set of VLAN identifiers filtered for VF(s) for port ID::
634 testpmd> rx_vlan rm (vlan_id) port (port_id) vf (vf_mask)
639 Add a tunnel filter on a port::
641 testpmd> tunnel_filter add (port_id) (outer_mac) (inner_mac) (ip_addr) \
642 (inner_vlan) (vxlan|nvgre|ipingre) (imac-ivlan|imac-ivlan-tenid|\
643 imac-tenid|imac|omac-imac-tenid|oip|iip) (tenant_id) (queue_id)
645 The available information categories are:
647 * ``vxlan``: Set tunnel type as VXLAN.
649 * ``nvgre``: Set tunnel type as NVGRE.
651 * ``ipingre``: Set tunnel type as IP-in-GRE.
653 * ``imac-ivlan``: Set filter type as Inner MAC and VLAN.
655 * ``imac-ivlan-tenid``: Set filter type as Inner MAC, VLAN and tenant ID.
657 * ``imac-tenid``: Set filter type as Inner MAC and tenant ID.
659 * ``imac``: Set filter type as Inner MAC.
661 * ``omac-imac-tenid``: Set filter type as Outer MAC, Inner MAC and tenant ID.
663 * ``oip``: Set filter type as Outer IP.
665 * ``iip``: Set filter type as Inner IP.
669 testpmd> tunnel_filter add 0 68:05:CA:28:09:82 00:00:00:00:00:00 \
670 192.168.2.2 0 ipingre oip 1 1
672 Set an IP-in-GRE tunnel on port 0, and the filter type is Outer IP.
677 Remove a tunnel filter on a port::
679 testpmd> tunnel_filter rm (port_id) (outer_mac) (inner_mac) (ip_addr) \
680 (inner_vlan) (vxlan|nvgre|ipingre) (imac-ivlan|imac-ivlan-tenid|\
681 imac-tenid|imac|omac-imac-tenid|oip|iip) (tenant_id) (queue_id)
686 Add an UDP port for VXLAN packet filter on a port::
688 testpmd> rx_vxlan_port add (udp_port) (port_id)
693 Remove an UDP port for VXLAN packet filter on a port::
695 testpmd> rx_vxlan_port rm (udp_port) (port_id)
700 Set hardware insertion of VLAN IDs in packets sent on a port::
702 testpmd> tx_vlan set (port_id) vlan_id[, vlan_id_outer]
704 For example, set a single VLAN ID (5) insertion on port 0::
708 Or, set double VLAN ID (inner: 2, outer: 3) insertion on port 1::
716 Set port based hardware insertion of VLAN ID in packets sent on a port::
718 testpmd> tx_vlan set pvid (port_id) (vlan_id) (on|off)
723 Disable hardware insertion of a VLAN header in packets sent on a port::
725 testpmd> tx_vlan reset (port_id)
730 Select hardware or software calculation of the checksum when
731 transmitting a packet using the ``csum`` forwarding engine::
733 testpmd> csum set (ip|udp|tcp|sctp|outer-ip) (hw|sw) (port_id)
737 * ``ip|udp|tcp|sctp`` always relate to the inner layer.
739 * ``outer-ip`` relates to the outer IP layer (only for IPv4) in the case where the packet is recognized
740 as a tunnel packet by the forwarding engine (vxlan, gre and ipip are
741 supported). See also the ``csum parse-tunnel`` command.
745 Check the NIC Datasheet for hardware limits.
750 Define how tunneled packets should be handled by the csum forward
753 testpmd> csum parse-tunnel (on|off) (tx_port_id)
755 If enabled, the csum forward engine will try to recognize supported
756 tunnel headers (vxlan, gre, ipip).
758 If disabled, treat tunnel packets as non-tunneled packets (a inner
759 header is handled as a packet payload).
763 The port argument is the TX port like in the ``csum set`` command.
767 Consider a packet in packet like the following::
769 eth_out/ipv4_out/udp_out/vxlan/eth_in/ipv4_in/tcp_in
771 * If parse-tunnel is enabled, the ``ip|udp|tcp|sctp`` parameters of ``csum set``
772 command relate to the inner headers (here ``ipv4_in`` and ``tcp_in``), and the
773 ``outer-ip parameter`` relates to the outer headers (here ``ipv4_out``).
775 * If parse-tunnel is disabled, the ``ip|udp|tcp|sctp`` parameters of ``csum set``
776 command relate to the outer headers, here ``ipv4_out`` and ``udp_out``.
781 Display tx checksum offload configuration::
783 testpmd> csum show (port_id)
788 Enable TCP Segmentation Offload (TSO) in the ``csum`` forwarding engine::
790 testpmd> tso set (segsize) (port_id)
794 Check the NIC datasheet for hardware limits.
799 Display the status of TCP Segmentation Offload::
801 testpmd> tso show (port_id)
806 Add an alternative MAC address to a port::
808 testpmd> mac_addr add (port_id) (XX:XX:XX:XX:XX:XX)
813 Remove a MAC address from a port::
815 testpmd> mac_addr remove (port_id) (XX:XX:XX:XX:XX:XX)
817 mac_addr add (for VF)
818 ~~~~~~~~~~~~~~~~~~~~~
820 Add an alternative MAC address for a VF to a port::
822 testpmd> mac_add add port (port_id) vf (vf_id) (XX:XX:XX:XX:XX:XX)
824 mac_addr set (for VF)
825 ~~~~~~~~~~~~~~~~~~~~~
827 Set the MAC address for a VF from the PF::
829 testpmd> set vf mac addr (port_id) (vf_id) (XX:XX:XX:XX:XX:XX)
834 Set the unicast hash filter(s) on/off for a port::
836 testpmd> set port (port_id) uta (XX:XX:XX:XX:XX:XX|all) (on|off)
841 Set the promiscuous mode on for a port or for all ports.
842 In promiscuous mode packets are not dropped if they aren't for the specified MAC address::
844 testpmd> set promisc (port_id|all) (on|off)
849 Set the allmulti mode for a port or for all ports::
851 testpmd> set allmulti (port_id|all) (on|off)
853 Same as the ifconfig (8) option. Controls how multicast packets are handled.
858 Set the link flow control parameter on a port::
860 testpmd> set flow_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \
861 (pause_time) (send_xon) mac_ctrl_frame_fwd (on|off) \
862 autoneg (on|off) (port_id)
866 * ``high_water`` (integer): High threshold value to trigger XOFF.
868 * ``low_water`` (integer): Low threshold value to trigger XON.
870 * ``pause_time`` (integer): Pause quota in the Pause frame.
872 * ``send_xon`` (0/1): Send XON frame.
874 * ``mac_ctrl_frame_fwd``: Enable receiving MAC control frames.
876 * ``autoneg``: Change the auto-negotiation parameter.
881 Set the priority flow control parameter on a port::
883 testpmd> set pfc_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \
884 (pause_time) (priority) (port_id)
888 * ``high_water`` (integer): High threshold value.
890 * ``low_water`` (integer): Low threshold value.
892 * ``pause_time`` (integer): Pause quota in the Pause frame.
894 * ``priority`` (0-7): VLAN User Priority.
899 Set statistics mapping (qmapping 0..15) for RX/TX queue on port::
901 testpmd> set stat_qmap (tx|rx) (port_id) (queue_id) (qmapping)
903 For example, to set rx queue 2 on port 0 to mapping 5::
905 testpmd>set stat_qmap rx 0 2 5
907 set port - rx/tx (for VF)
908 ~~~~~~~~~~~~~~~~~~~~~~~~~
910 Set VF receive/transmit from a port::
912 testpmd> set port (port_id) vf (vf_id) (rx|tx) (on|off)
914 set port - mac address filter (for VF)
915 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
917 Add/Remove unicast or multicast MAC addr filter for a VF::
919 testpmd> set port (port_id) vf (vf_id) (mac_addr) \
920 (exact-mac|exact-mac-vlan|hashmac|hashmac-vlan) (on|off)
922 set port - rx mode(for VF)
923 ~~~~~~~~~~~~~~~~~~~~~~~~~~
925 Set the VF receive mode of a port::
927 testpmd> set port (port_id) vf (vf_id) \
928 rxmode (AUPE|ROPE|BAM|MPE) (on|off)
930 The available receive modes are:
932 * ``AUPE``: Accepts untagged VLAN.
934 * ``ROPE``: Accepts unicast hash.
936 * ``BAM``: Accepts broadcast packets.
938 * ``MPE``: Accepts all multicast packets.
940 set port - tx_rate (for Queue)
941 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
943 Set TX rate limitation for a queue on a port::
945 testpmd> set port (port_id) queue (queue_id) rate (rate_value)
947 set port - tx_rate (for VF)
948 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
950 Set TX rate limitation for queues in VF on a port::
952 testpmd> set port (port_id) vf (vf_id) rate (rate_value) queue_mask (queue_mask)
954 set port - mirror rule
955 ~~~~~~~~~~~~~~~~~~~~~~
957 Set pool or vlan type mirror rule for a port::
959 testpmd> set port (port_id) mirror-rule (rule_id) \
960 (pool-mirror-up|pool-mirror-down|vlan-mirror) \
961 (poolmask|vlanid[,vlanid]*) dst-pool (pool_id) (on|off)
963 Set link mirror rule for a port::
965 testpmd> set port (port_id) mirror-rule (rule_id) \
966 (uplink-mirror|downlink-mirror) dst-pool (pool_id) (on|off)
968 For example to enable mirror traffic with vlan 0,1 to pool 0::
970 set port 0 mirror-rule 0 vlan-mirror 0,1 dst-pool 0 on
972 reset port - mirror rule
973 ~~~~~~~~~~~~~~~~~~~~~~~~
975 Reset a mirror rule for a port::
977 testpmd> reset port (port_id) mirror-rule (rule_id)
982 Set the flush on RX streams before forwarding.
983 The default is flush ``on``.
984 Mainly used with PCAP drivers to turn off the default behavior of flushing the first 512 packets on RX streams::
986 testpmd> set flush_rx off
991 Set the bypass mode for the lowest port on bypass enabled NIC::
993 testpmd> set bypass mode (normal|bypass|isolate) (port_id)
998 Set the event required to initiate specified bypass mode for the lowest port on a bypass enabled::
1000 testpmd> set bypass event (timeout|os_on|os_off|power_on|power_off) \
1001 mode (normal|bypass|isolate) (port_id)
1005 * ``timeout``: Enable bypass after watchdog timeout.
1007 * ``os_on``: Enable bypass when OS/board is powered on.
1009 * ``os_off``: Enable bypass when OS/board is powered off.
1011 * ``power_on``: Enable bypass when power supply is turned on.
1013 * ``power_off``: Enable bypass when power supply is turned off.
1019 Set the bypass watchdog timeout to ``n`` seconds where 0 = instant::
1021 testpmd> set bypass timeout (0|1.5|2|3|4|8|16|32)
1026 Show the bypass configuration for a bypass enabled NIC using the lowest port on the NIC::
1028 testpmd> show bypass config (port_id)
1033 Set link up for a port::
1035 testpmd> set link-up port (port id)
1040 Set link down for a port::
1042 testpmd> set link-down port (port id)
1047 Enable E-tag insertion for a VF on a port::
1049 testpmd> E-tag set insertion on port-tag-id (value) port (port_id) vf (vf_id)
1051 Disable E-tag insertion for a VF on a port::
1053 testpmd> E-tag set insertion off port (port_id) vf (vf_id)
1055 Enable/disable E-tag stripping on a port::
1057 testpmd> E-tag set stripping (on|off) port (port_id)
1059 Enable/disable E-tag based forwarding on a port::
1061 testpmd> E-tag set forwarding (on|off) port (port_id)
1063 Add an E-tag forwarding filter on a port::
1065 testpmd> E-tag set filter add e-tag-id (value) dst-pool (pool_id) port (port_id)
1067 Delete an E-tag forwarding filter on a port::
1068 testpmd> E-tag set filter del e-tag-id (value) port (port_id)
1074 The following sections show functions for configuring ports.
1078 Port configuration changes only become active when forwarding is started/restarted.
1083 Attach a port specified by pci address or virtual device args::
1085 testpmd> port attach (identifier)
1087 To attach a new pci device, the device should be recognized by kernel first.
1088 Then it should be moved under DPDK management.
1089 Finally the port can be attached to testpmd.
1091 For example, to move a pci device using ixgbe under DPDK management:
1093 .. code-block:: console
1095 # Check the status of the available devices.
1096 ./usertools/dpdk-devbind.py --status
1098 Network devices using DPDK-compatible driver
1099 ============================================
1102 Network devices using kernel driver
1103 ===================================
1104 0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused=
1107 # Bind the device to igb_uio.
1108 sudo ./usertools/dpdk-devbind.py -b igb_uio 0000:0a:00.0
1111 # Recheck the status of the devices.
1112 ./usertools/dpdk-devbind.py --status
1113 Network devices using DPDK-compatible driver
1114 ============================================
1115 0000:0a:00.0 '82599ES 10-Gigabit' drv=igb_uio unused=
1117 To attach a port created by virtual device, above steps are not needed.
1119 For example, to attach a port whose pci address is 0000:0a:00.0.
1121 .. code-block:: console
1123 testpmd> port attach 0000:0a:00.0
1124 Attaching a new port...
1125 EAL: PCI device 0000:0a:00.0 on NUMA socket -1
1126 EAL: probe driver: 8086:10fb rte_ixgbe_pmd
1127 EAL: PCI memory mapped at 0x7f83bfa00000
1128 EAL: PCI memory mapped at 0x7f83bfa80000
1129 PMD: eth_ixgbe_dev_init(): MAC: 2, PHY: 18, SFP+: 5
1130 PMD: eth_ixgbe_dev_init(): port 0 vendorID=0x8086 deviceID=0x10fb
1131 Port 0 is attached. Now total ports is 1
1134 For example, to attach a port created by pcap PMD.
1136 .. code-block:: console
1138 testpmd> port attach net_pcap0
1139 Attaching a new port...
1140 PMD: Initializing pmd_pcap for net_pcap0
1141 PMD: Creating pcap-backed ethdev on numa socket 0
1142 Port 0 is attached. Now total ports is 1
1145 In this case, identifier is ``net_pcap0``.
1146 This identifier format is the same as ``--vdev`` format of DPDK applications.
1148 For example, to re-attach a bonded port which has been previously detached,
1149 the mode and slave parameters must be given.
1151 .. code-block:: console
1153 testpmd> port attach net_bond_0,mode=0,slave=1
1154 Attaching a new port...
1155 EAL: Initializing pmd_bond for net_bond_0
1156 EAL: Create bonded device net_bond_0 on port 0 in mode 0 on socket 0.
1157 Port 0 is attached. Now total ports is 1
1164 Detach a specific port::
1166 testpmd> port detach (port_id)
1168 Before detaching a port, the port should be stopped and closed.
1170 For example, to detach a pci device port 0.
1172 .. code-block:: console
1174 testpmd> port stop 0
1177 testpmd> port close 0
1181 testpmd> port detach 0
1183 EAL: PCI device 0000:0a:00.0 on NUMA socket -1
1184 EAL: remove driver: 8086:10fb rte_ixgbe_pmd
1185 EAL: PCI memory unmapped at 0x7f83bfa00000
1186 EAL: PCI memory unmapped at 0x7f83bfa80000
1190 For example, to detach a virtual device port 0.
1192 .. code-block:: console
1194 testpmd> port stop 0
1197 testpmd> port close 0
1201 testpmd> port detach 0
1203 PMD: Closing pcap ethdev on numa socket 0
1204 Port 'net_pcap0' is detached. Now total ports is 0
1207 To remove a pci device completely from the system, first detach the port from testpmd.
1208 Then the device should be moved under kernel management.
1209 Finally the device can be removed using kernel pci hotplug functionality.
1211 For example, to move a pci device under kernel management:
1213 .. code-block:: console
1215 sudo ./usertools/dpdk-devbind.py -b ixgbe 0000:0a:00.0
1217 ./usertools/dpdk-devbind.py --status
1219 Network devices using DPDK-compatible driver
1220 ============================================
1223 Network devices using kernel driver
1224 ===================================
1225 0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused=igb_uio
1227 To remove a port created by a virtual device, above steps are not needed.
1232 Start all ports or a specific port::
1234 testpmd> port start (port_id|all)
1239 Stop all ports or a specific port::
1241 testpmd> port stop (port_id|all)
1246 Close all ports or a specific port::
1248 testpmd> port close (port_id|all)
1250 port start/stop queue
1251 ~~~~~~~~~~~~~~~~~~~~~
1253 Start/stop a rx/tx queue on a specific port::
1255 testpmd> port (port_id) (rxq|txq) (queue_id) (start|stop)
1257 Only take effect when port is started.
1262 Set the speed and duplex mode for all ports or a specific port::
1264 testpmd> port config (port_id|all) speed (10|100|1000|10000|25000|40000|50000|100000|auto) \
1265 duplex (half|full|auto)
1267 port config - queues/descriptors
1268 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1270 Set number of queues/descriptors for rxq, txq, rxd and txd::
1272 testpmd> port config all (rxq|txq|rxd|txd) (value)
1274 This is equivalent to the ``--rxq``, ``--txq``, ``--rxd`` and ``--txd`` command-line options.
1276 port config - max-pkt-len
1277 ~~~~~~~~~~~~~~~~~~~~~~~~~
1279 Set the maximum packet length::
1281 testpmd> port config all max-pkt-len (value)
1283 This is equivalent to the ``--max-pkt-len`` command-line option.
1285 port config - CRC Strip
1286 ~~~~~~~~~~~~~~~~~~~~~~~
1288 Set hardware CRC stripping on or off for all ports::
1290 testpmd> port config all crc-strip (on|off)
1292 CRC stripping is off by default.
1294 The ``on`` option is equivalent to the ``--crc-strip`` command-line option.
1296 port config - scatter
1297 ~~~~~~~~~~~~~~~~~~~~~~~
1299 Set RX scatter mode on or off for all ports::
1301 testpmd> port config all scatter (on|off)
1303 RX scatter mode is off by default.
1305 The ``on`` option is equivalent to the ``--enable-scatter`` command-line option.
1307 port config - TX queue flags
1308 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1310 Set a hexadecimal bitmap of TX queue flags for all ports::
1312 testpmd> port config all txqflags value
1314 This command is equivalent to the ``--txqflags`` command-line option.
1316 port config - RX Checksum
1317 ~~~~~~~~~~~~~~~~~~~~~~~~~
1319 Set hardware RX checksum offload to on or off for all ports::
1321 testpmd> port config all rx-cksum (on|off)
1323 Checksum offload is off by default.
1325 The ``on`` option is equivalent to the ``--enable-rx-cksum`` command-line option.
1330 Set hardware VLAN on or off for all ports::
1332 testpmd> port config all hw-vlan (on|off)
1334 Hardware VLAN is on by default.
1336 The ``off`` option is equivalent to the ``--disable-hw-vlan`` command-line option.
1338 port config - VLAN filter
1339 ~~~~~~~~~~~~~~~~~~~~~~~~~
1341 Set hardware VLAN filter on or off for all ports::
1343 testpmd> port config all hw-vlan-filter (on|off)
1345 Hardware VLAN filter is on by default.
1347 The ``off`` option is equivalent to the ``--disable-hw-vlan-filter`` command-line option.
1349 port config - VLAN strip
1350 ~~~~~~~~~~~~~~~~~~~~~~~~
1352 Set hardware VLAN strip on or off for all ports::
1354 testpmd> port config all hw-vlan-strip (on|off)
1356 Hardware VLAN strip is on by default.
1358 The ``off`` option is equivalent to the ``--disable-hw-vlan-strip`` command-line option.
1360 port config - VLAN extend
1361 ~~~~~~~~~~~~~~~~~~~~~~~~~
1363 Set hardware VLAN extend on or off for all ports::
1365 testpmd> port config all hw-vlan-extend (on|off)
1367 Hardware VLAN extend is off by default.
1369 The ``off`` option is equivalent to the ``--disable-hw-vlan-extend`` command-line option.
1371 port config - Drop Packets
1372 ~~~~~~~~~~~~~~~~~~~~~~~~~~
1374 Set packet drop for packets with no descriptors on or off for all ports::
1376 testpmd> port config all drop-en (on|off)
1378 Packet dropping for packets with no descriptors is off by default.
1380 The ``on`` option is equivalent to the ``--enable-drop-en`` command-line option.
1385 Set the RSS (Receive Side Scaling) mode on or off::
1387 testpmd> port config all rss (all|ip|tcp|udp|sctp|ether|port|vxlan|geneve|nvgre|none)
1389 RSS is on by default.
1391 The ``none`` option is equivalent to the ``--disable-rss`` command-line option.
1393 port config - RSS Reta
1394 ~~~~~~~~~~~~~~~~~~~~~~
1396 Set the RSS (Receive Side Scaling) redirection table::
1398 testpmd> port config all rss reta (hash,queue)[,(hash,queue)]
1403 Set the DCB mode for an individual port::
1405 testpmd> port config (port_id) dcb vt (on|off) (traffic_class) pfc (on|off)
1407 The traffic class should be 4 or 8.
1412 Set the number of packets per burst::
1414 testpmd> port config all burst (value)
1416 This is equivalent to the ``--burst`` command-line option.
1418 port config - Threshold
1419 ~~~~~~~~~~~~~~~~~~~~~~~
1421 Set thresholds for TX/RX queues::
1423 testpmd> port config all (threshold) (value)
1425 Where the threshold type can be:
1427 * ``txpt:`` Set the prefetch threshold register of the TX rings, 0 <= value <= 255.
1429 * ``txht:`` Set the host threshold register of the TX rings, 0 <= value <= 255.
1431 * ``txwt:`` Set the write-back threshold register of the TX rings, 0 <= value <= 255.
1433 * ``rxpt:`` Set the prefetch threshold register of the RX rings, 0 <= value <= 255.
1435 * ``rxht:`` Set the host threshold register of the RX rings, 0 <= value <= 255.
1437 * ``rxwt:`` Set the write-back threshold register of the RX rings, 0 <= value <= 255.
1439 * ``txfreet:`` Set the transmit free threshold of the TX rings, 0 <= value <= txd.
1441 * ``rxfreet:`` Set the transmit free threshold of the RX rings, 0 <= value <= rxd.
1443 * ``txrst:`` Set the transmit RS bit threshold of TX rings, 0 <= value <= txd.
1445 These threshold options are also available from the command-line.
1450 Set the value of ether-type for E-tag::
1452 testpmd> port config (port_id|all) l2-tunnel E-tag ether-type (value)
1454 Enable/disable the E-tag support::
1456 testpmd> port config (port_id|all) l2-tunnel E-tag (enable|disable)
1459 Link Bonding Functions
1460 ----------------------
1462 The Link Bonding functions make it possible to dynamically create and
1463 manage link bonding devices from within testpmd interactive prompt.
1465 create bonded device
1466 ~~~~~~~~~~~~~~~~~~~~
1468 Create a new bonding device::
1470 testpmd> create bonded device (mode) (socket)
1472 For example, to create a bonded device in mode 1 on socket 0::
1474 testpmd> create bonded 1 0
1475 created new bonded device (port X)
1480 Adds Ethernet device to a Link Bonding device::
1482 testpmd> add bonding slave (slave id) (port id)
1484 For example, to add Ethernet device (port 6) to a Link Bonding device (port 10)::
1486 testpmd> add bonding slave 6 10
1489 remove bonding slave
1490 ~~~~~~~~~~~~~~~~~~~~
1492 Removes an Ethernet slave device from a Link Bonding device::
1494 testpmd> remove bonding slave (slave id) (port id)
1496 For example, to remove Ethernet slave device (port 6) to a Link Bonding device (port 10)::
1498 testpmd> remove bonding slave 6 10
1503 Set the Link Bonding mode of a Link Bonding device::
1505 testpmd> set bonding mode (value) (port id)
1507 For example, to set the bonding mode of a Link Bonding device (port 10) to broadcast (mode 3)::
1509 testpmd> set bonding mode 3 10
1514 Set an Ethernet slave device as the primary device on a Link Bonding device::
1516 testpmd> set bonding primary (slave id) (port id)
1518 For example, to set the Ethernet slave device (port 6) as the primary port of a Link Bonding device (port 10)::
1520 testpmd> set bonding primary 6 10
1525 Set the MAC address of a Link Bonding device::
1527 testpmd> set bonding mac (port id) (mac)
1529 For example, to set the MAC address of a Link Bonding device (port 10) to 00:00:00:00:00:01::
1531 testpmd> set bonding mac 10 00:00:00:00:00:01
1533 set bonding xmit_balance_policy
1534 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1536 Set the transmission policy for a Link Bonding device when it is in Balance XOR mode::
1538 testpmd> set bonding xmit_balance_policy (port_id) (l2|l23|l34)
1540 For example, set a Link Bonding device (port 10) to use a balance policy of layer 3+4 (IP addresses & UDP ports)::
1542 testpmd> set bonding xmit_balance_policy 10 l34
1545 set bonding mon_period
1546 ~~~~~~~~~~~~~~~~~~~~~~
1548 Set the link status monitoring polling period in milliseconds for a bonding device.
1550 This adds support for PMD slave devices which do not support link status interrupts.
1551 When the mon_period is set to a value greater than 0 then all PMD's which do not support
1552 link status ISR will be queried every polling interval to check if their link status has changed::
1554 testpmd> set bonding mon_period (port_id) (value)
1556 For example, to set the link status monitoring polling period of bonded device (port 5) to 150ms::
1558 testpmd> set bonding mon_period 5 150
1564 Show the current configuration of a Link Bonding device::
1566 testpmd> show bonding config (port id)
1569 to show the configuration a Link Bonding device (port 9) with 3 slave devices (1, 3, 4)
1570 in balance mode with a transmission policy of layer 2+3::
1572 testpmd> show bonding config 9
1574 Balance Xmit Policy: BALANCE_XMIT_POLICY_LAYER23
1576 Active Slaves (3): [1 3 4]
1583 The Register Functions can be used to read from and write to registers on the network card referenced by a port number.
1584 This is mainly useful for debugging purposes.
1585 Reference should be made to the appropriate datasheet for the network card for details on the register addresses
1586 and fields that can be accessed.
1591 Display the value of a port register::
1593 testpmd> read reg (port_id) (address)
1595 For example, to examine the Flow Director control register (FDIRCTL, 0x0000EE000) on an Intel 82599 10 GbE Controller::
1597 testpmd> read reg 0 0xEE00
1598 port 0 PCI register at offset 0xEE00: 0x4A060029 (1241907241)
1603 Display a port register bit field::
1605 testpmd> read regfield (port_id) (address) (bit_x) (bit_y)
1607 For example, reading the lowest two bits from the register in the example above::
1609 testpmd> read regfield 0 0xEE00 0 1
1610 port 0 PCI register at offset 0xEE00: bits[0, 1]=0x1 (1)
1615 Display a single port register bit::
1617 testpmd> read regbit (port_id) (address) (bit_x)
1619 For example, reading the lowest bit from the register in the example above::
1621 testpmd> read regbit 0 0xEE00 0
1622 port 0 PCI register at offset 0xEE00: bit 0=1
1627 Set the value of a port register::
1629 testpmd> write reg (port_id) (address) (value)
1631 For example, to clear a register::
1633 testpmd> write reg 0 0xEE00 0x0
1634 port 0 PCI register at offset 0xEE00: 0x00000000 (0)
1639 Set bit field of a port register::
1641 testpmd> write regfield (port_id) (address) (bit_x) (bit_y) (value)
1643 For example, writing to the register cleared in the example above::
1645 testpmd> write regfield 0 0xEE00 0 1 2
1646 port 0 PCI register at offset 0xEE00: 0x00000002 (2)
1651 Set single bit value of a port register::
1653 testpmd> write regbit (port_id) (address) (bit_x) (value)
1655 For example, to set the high bit in the register from the example above::
1657 testpmd> write regbit 0 0xEE00 31 1
1658 port 0 PCI register at offset 0xEE00: 0x8000000A (2147483658)
1664 This section details the available filter functions that are available.
1666 Note these functions interface the deprecated legacy filtering framework,
1667 superseded by *rte_flow*. See `Flow rules management`_.
1670 ~~~~~~~~~~~~~~~~~~~~
1672 Add or delete a L2 Ethertype filter, which identify packets by their L2 Ethertype mainly assign them to a receive queue::
1674 ethertype_filter (port_id) (add|del) (mac_addr|mac_ignr) (mac_address) \
1675 ethertype (ether_type) (drop|fwd) queue (queue_id)
1677 The available information parameters are:
1679 * ``port_id``: The port which the Ethertype filter assigned on.
1681 * ``mac_addr``: Compare destination mac address.
1683 * ``mac_ignr``: Ignore destination mac address match.
1685 * ``mac_address``: Destination mac address to match.
1687 * ``ether_type``: The EtherType value want to match,
1688 for example 0x0806 for ARP packet. 0x0800 (IPv4) and 0x86DD (IPv6) are invalid.
1690 * ``queue_id``: The receive queue associated with this EtherType filter.
1691 It is meaningless when deleting or dropping.
1693 Example, to add/remove an ethertype filter rule::
1695 testpmd> ethertype_filter 0 add mac_ignr 00:11:22:33:44:55 \
1696 ethertype 0x0806 fwd queue 3
1698 testpmd> ethertype_filter 0 del mac_ignr 00:11:22:33:44:55 \
1699 ethertype 0x0806 fwd queue 3
1704 Add or delete a 2-tuple filter,
1705 which identifies packets by specific protocol and destination TCP/UDP port
1706 and forwards packets into one of the receive queues::
1708 2tuple_filter (port_id) (add|del) dst_port (dst_port_value) \
1709 protocol (protocol_value) mask (mask_value) \
1710 tcp_flags (tcp_flags_value) priority (prio_value) \
1713 The available information parameters are:
1715 * ``port_id``: The port which the 2-tuple filter assigned on.
1717 * ``dst_port_value``: Destination port in L4.
1719 * ``protocol_value``: IP L4 protocol.
1721 * ``mask_value``: Participates in the match or not by bit for field above, 1b means participate.
1723 * ``tcp_flags_value``: TCP control bits. The non-zero value is invalid, when the pro_value is not set to 0x06 (TCP).
1725 * ``prio_value``: Priority of this filter.
1727 * ``queue_id``: The receive queue associated with this 2-tuple filter.
1729 Example, to add/remove an 2tuple filter rule::
1731 testpmd> 2tuple_filter 0 add dst_port 32 protocol 0x06 mask 0x03 \
1732 tcp_flags 0x02 priority 3 queue 3
1734 testpmd> 2tuple_filter 0 del dst_port 32 protocol 0x06 mask 0x03 \
1735 tcp_flags 0x02 priority 3 queue 3
1740 Add or delete a 5-tuple filter,
1741 which consists of a 5-tuple (protocol, source and destination IP addresses, source and destination TCP/UDP/SCTP port)
1742 and routes packets into one of the receive queues::
1744 5tuple_filter (port_id) (add|del) dst_ip (dst_address) src_ip \
1745 (src_address) dst_port (dst_port_value) \
1746 src_port (src_port_value) protocol (protocol_value) \
1747 mask (mask_value) tcp_flags (tcp_flags_value) \
1748 priority (prio_value) queue (queue_id)
1750 The available information parameters are:
1752 * ``port_id``: The port which the 5-tuple filter assigned on.
1754 * ``dst_address``: Destination IP address.
1756 * ``src_address``: Source IP address.
1758 * ``dst_port_value``: TCP/UDP destination port.
1760 * ``src_port_value``: TCP/UDP source port.
1762 * ``protocol_value``: L4 protocol.
1764 * ``mask_value``: Participates in the match or not by bit for field above, 1b means participate
1766 * ``tcp_flags_value``: TCP control bits. The non-zero value is invalid, when the protocol_value is not set to 0x06 (TCP).
1768 * ``prio_value``: The priority of this filter.
1770 * ``queue_id``: The receive queue associated with this 5-tuple filter.
1772 Example, to add/remove an 5tuple filter rule::
1774 testpmd> 5tuple_filter 0 add dst_ip 2.2.2.5 src_ip 2.2.2.4 \
1775 dst_port 64 src_port 32 protocol 0x06 mask 0x1F \
1776 flags 0x0 priority 3 queue 3
1778 testpmd> 5tuple_filter 0 del dst_ip 2.2.2.5 src_ip 2.2.2.4 \
1779 dst_port 64 src_port 32 protocol 0x06 mask 0x1F \
1780 flags 0x0 priority 3 queue 3
1785 Using the SYN filter, TCP packets whose *SYN* flag is set can be forwarded to a separate queue::
1787 syn_filter (port_id) (add|del) priority (high|low) queue (queue_id)
1789 The available information parameters are:
1791 * ``port_id``: The port which the SYN filter assigned on.
1793 * ``high``: This SYN filter has higher priority than other filters.
1795 * ``low``: This SYN filter has lower priority than other filters.
1797 * ``queue_id``: The receive queue associated with this SYN filter
1801 testpmd> syn_filter 0 add priority high queue 3
1806 With flex filter, packets can be recognized by any arbitrary pattern within the first 128 bytes of the packet
1807 and routed into one of the receive queues::
1809 flex_filter (port_id) (add|del) len (len_value) bytes (bytes_value) \
1810 mask (mask_value) priority (prio_value) queue (queue_id)
1812 The available information parameters are:
1814 * ``port_id``: The port which the Flex filter is assigned on.
1816 * ``len_value``: Filter length in bytes, no greater than 128.
1818 * ``bytes_value``: A string in hexadecimal, means the value the flex filter needs to match.
1820 * ``mask_value``: A string in hexadecimal, bit 1 means corresponding byte participates in the match.
1822 * ``prio_value``: The priority of this filter.
1824 * ``queue_id``: The receive queue associated with this Flex filter.
1828 testpmd> flex_filter 0 add len 16 bytes 0x00000000000000000000000008060000 \
1829 mask 000C priority 3 queue 3
1831 testpmd> flex_filter 0 del len 16 bytes 0x00000000000000000000000008060000 \
1832 mask 000C priority 3 queue 3
1835 .. _testpmd_flow_director:
1837 flow_director_filter
1838 ~~~~~~~~~~~~~~~~~~~~
1840 The Flow Director works in receive mode to identify specific flows or sets of flows and route them to specific queues.
1842 Four types of filtering are supported which are referred to as Perfect Match, Signature, Perfect-mac-vlan and
1843 Perfect-tunnel filters, the match mode is set by the ``--pkt-filter-mode`` command-line parameter:
1845 * Perfect match filters.
1846 The hardware checks a match between the masked fields of the received packets and the programmed filters.
1847 The masked fields are for IP flow.
1849 * Signature filters.
1850 The hardware checks a match between a hash-based signature of the masked fields of the received packet.
1852 * Perfect-mac-vlan match filters.
1853 The hardware checks a match between the masked fields of the received packets and the programmed filters.
1854 The masked fields are for MAC VLAN flow.
1856 * Perfect-tunnel match filters.
1857 The hardware checks a match between the masked fields of the received packets and the programmed filters.
1858 The masked fields are for tunnel flow.
1860 The Flow Director filters can match the different fields for different type of packet: flow type, specific input set
1861 per flow type and the flexible payload.
1863 The Flow Director can also mask out parts of all of these fields so that filters
1864 are only applied to certain fields or parts of the fields.
1866 Different NICs may have different capabilities, command show port fdir (port_id) can be used to acquire the information.
1868 # Commands to add flow director filters of different flow types::
1870 flow_director_filter (port_id) mode IP (add|del|update) \
1871 flow (ipv4-other|ipv4-frag|ipv6-other|ipv6-frag) \
1872 src (src_ip_address) dst (dst_ip_address) \
1873 tos (tos_value) proto (proto_value) ttl (ttl_value) \
1874 vlan (vlan_value) flexbytes (flexbytes_value) \
1875 (drop|fwd) pf|vf(vf_id) queue (queue_id) \
1878 flow_director_filter (port_id) mode IP (add|del|update) \
1879 flow (ipv4-tcp|ipv4-udp|ipv6-tcp|ipv6-udp) \
1880 src (src_ip_address) (src_port) \
1881 dst (dst_ip_address) (dst_port) \
1882 tos (tos_value) ttl (ttl_value) \
1883 vlan (vlan_value) flexbytes (flexbytes_value) \
1884 (drop|fwd) queue pf|vf(vf_id) (queue_id) \
1887 flow_director_filter (port_id) mode IP (add|del|update) \
1888 flow (ipv4-sctp|ipv6-sctp) \
1889 src (src_ip_address) (src_port) \
1890 dst (dst_ip_address) (dst_port) \
1891 tos (tos_value) ttl (ttl_value) \
1892 tag (verification_tag) vlan (vlan_value) \
1893 flexbytes (flexbytes_value) (drop|fwd) \
1894 pf|vf(vf_id) queue (queue_id) fd_id (fd_id_value)
1896 flow_director_filter (port_id) mode IP (add|del|update) flow l2_payload \
1897 ether (ethertype) flexbytes (flexbytes_value) \
1898 (drop|fwd) pf|vf(vf_id) queue (queue_id)
1901 flow_director_filter (port_id) mode MAC-VLAN (add|del|update) \
1902 mac (mac_address) vlan (vlan_value) \
1903 flexbytes (flexbytes_value) (drop|fwd) \
1904 queue (queue_id) fd_id (fd_id_value)
1906 flow_director_filter (port_id) mode Tunnel (add|del|update) \
1907 mac (mac_address) vlan (vlan_value) \
1908 tunnel (NVGRE|VxLAN) tunnel-id (tunnel_id_value) \
1909 flexbytes (flexbytes_value) (drop|fwd) \
1910 queue (queue_id) fd_id (fd_id_value)
1912 For example, to add an ipv4-udp flow type filter::
1914 testpmd> flow_director_filter 0 mode IP add flow ipv4-udp src 2.2.2.3 32 \
1915 dst 2.2.2.5 33 tos 2 ttl 40 vlan 0x1 flexbytes (0x88,0x48) \
1916 fwd pf queue 1 fd_id 1
1918 For example, add an ipv4-other flow type filter::
1920 testpmd> flow_director_filter 0 mode IP add flow ipv4-other src 2.2.2.3 \
1921 dst 2.2.2.5 tos 2 proto 20 ttl 40 vlan 0x1 \
1922 flexbytes (0x88,0x48) fwd pf queue 1 fd_id 1
1927 Flush all flow director filters on a device::
1929 testpmd> flush_flow_director (port_id)
1931 Example, to flush all flow director filter on port 0::
1933 testpmd> flush_flow_director 0
1938 Set flow director's input masks::
1940 flow_director_mask (port_id) mode IP vlan (vlan_value) \
1941 src_mask (ipv4_src) (ipv6_src) (src_port) \
1942 dst_mask (ipv4_dst) (ipv6_dst) (dst_port)
1944 flow_director_mask (port_id) mode MAC-VLAN vlan (vlan_value)
1946 flow_director_mask (port_id) mode Tunnel vlan (vlan_value) \
1947 mac (mac_value) tunnel-type (tunnel_type_value) \
1948 tunnel-id (tunnel_id_value)
1950 Example, to set flow director mask on port 0::
1952 testpmd> flow_director_mask 0 mode IP vlan 0xefff \
1953 src_mask 255.255.255.255 \
1954 FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF \
1955 dst_mask 255.255.255.255 \
1956 FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF
1958 flow_director_flex_mask
1959 ~~~~~~~~~~~~~~~~~~~~~~~
1961 set masks of flow director's flexible payload based on certain flow type::
1963 testpmd> flow_director_flex_mask (port_id) \
1964 flow (none|ipv4-other|ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
1965 ipv6-other|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp| \
1966 l2_payload|all) (mask)
1968 Example, to set flow director's flex mask for all flow type on port 0::
1970 testpmd> flow_director_flex_mask 0 flow all \
1971 (0xff,0xff,0,0,0,0,0,0,0,0,0,0,0,0,0,0)
1974 flow_director_flex_payload
1975 ~~~~~~~~~~~~~~~~~~~~~~~~~~
1977 Configure flexible payload selection::
1979 flow_director_flex_payload (port_id) (raw|l2|l3|l4) (config)
1981 For example, to select the first 16 bytes from the offset 4 (bytes) of packet's payload as flexible payload::
1983 testpmd> flow_director_flex_payload 0 l4 \
1984 (4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19)
1986 get_sym_hash_ena_per_port
1987 ~~~~~~~~~~~~~~~~~~~~~~~~~
1989 Get symmetric hash enable configuration per port::
1991 get_sym_hash_ena_per_port (port_id)
1993 For example, to get symmetric hash enable configuration of port 1::
1995 testpmd> get_sym_hash_ena_per_port 1
1997 set_sym_hash_ena_per_port
1998 ~~~~~~~~~~~~~~~~~~~~~~~~~
2000 Set symmetric hash enable configuration per port to enable or disable::
2002 set_sym_hash_ena_per_port (port_id) (enable|disable)
2004 For example, to set symmetric hash enable configuration of port 1 to enable::
2006 testpmd> set_sym_hash_ena_per_port 1 enable
2008 get_hash_global_config
2009 ~~~~~~~~~~~~~~~~~~~~~~
2011 Get the global configurations of hash filters::
2013 get_hash_global_config (port_id)
2015 For example, to get the global configurations of hash filters of port 1::
2017 testpmd> get_hash_global_config 1
2019 set_hash_global_config
2020 ~~~~~~~~~~~~~~~~~~~~~~
2022 Set the global configurations of hash filters::
2024 set_hash_global_config (port_id) (toeplitz|simple_xor|default) \
2025 (ipv4|ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp|ipv4-other|ipv6|ipv6-frag| \
2026 ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other|l2_payload) \
2029 For example, to enable simple_xor for flow type of ipv6 on port 2::
2031 testpmd> set_hash_global_config 2 simple_xor ipv6 enable
2036 Set the input set for hash::
2038 set_hash_input_set (port_id) (ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
2039 ipv4-other|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other| \
2040 l2_payload) (ovlan|ivlan|src-ipv4|dst-ipv4|src-ipv6|dst-ipv6|ipv4-tos| \
2041 ipv4-proto|ipv6-tc|ipv6-next-header|udp-src-port|udp-dst-port| \
2042 tcp-src-port|tcp-dst-port|sctp-src-port|sctp-dst-port|sctp-veri-tag| \
2043 udp-key|gre-key|fld-1st|fld-2nd|fld-3rd|fld-4th|fld-5th|fld-6th|fld-7th| \
2044 fld-8th|none) (select|add)
2046 For example, to add source IP to hash input set for flow type of ipv4-udp on port 0::
2048 testpmd> set_hash_input_set 0 ipv4-udp src-ipv4 add
2053 The Flow Director filters can match the different fields for different type of packet, i.e. specific input set
2054 on per flow type and the flexible payload. This command can be used to change input set for each flow type.
2056 Set the input set for flow director::
2058 set_fdir_input_set (port_id) (ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
2059 ipv4-other|ipv6|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other| \
2060 l2_payload) (ivlan|ethertype|src-ipv4|dst-ipv4|src-ipv6|dst-ipv6|ipv4-tos| \
2061 ipv4-proto|ipv4-ttl|ipv6-tc|ipv6-next-header|ipv6-hop-limits| \
2062 tudp-src-port|udp-dst-port|cp-src-port|tcp-dst-port|sctp-src-port| \
2063 sctp-dst-port|sctp-veri-tag|none) (select|add)
2065 For example to add source IP to FD input set for flow type of ipv4-udp on port 0::
2067 testpmd> set_fdir_input_set 0 ipv4-udp src-ipv4 add
2072 Set different GRE key length for input set::
2074 global_config (port_id) gre-key-len (number in bytes)
2076 For example to set GRE key length for input set to 4 bytes on port 0::
2078 testpmd> global_config 0 gre-key-len 4
2081 .. _testpmd_rte_flow:
2083 Flow rules management
2084 ---------------------
2086 Control of the generic flow API (*rte_flow*) is fully exposed through the
2087 ``flow`` command (validation, creation, destruction and queries).
2089 Considering *rte_flow* overlaps with all `Filter Functions`_, using both
2090 features simultaneously may cause undefined side-effects and is therefore
2096 Because the ``flow`` command uses dynamic tokens to handle the large number
2097 of possible flow rules combinations, its behavior differs slightly from
2098 other commands, in particular:
2100 - Pressing *?* or the *<tab>* key displays contextual help for the current
2101 token, not that of the entire command.
2103 - Optional and repeated parameters are supported (provided they are listed
2104 in the contextual help).
2106 The first parameter stands for the operation mode. Possible operations and
2107 their general syntax are described below. They are covered in detail in the
2110 - Check whether a flow rule can be created::
2112 flow validate {port_id}
2113 [group {group_id}] [priority {level}] [ingress] [egress]
2114 pattern {item} [/ {item} [...]] / end
2115 actions {action} [/ {action} [...]] / end
2117 - Create a flow rule::
2119 flow create {port_id}
2120 [group {group_id}] [priority {level}] [ingress] [egress]
2121 pattern {item} [/ {item} [...]] / end
2122 actions {action} [/ {action} [...]] / end
2124 - Destroy specific flow rules::
2126 flow destroy {port_id} rule {rule_id} [...]
2128 - Destroy all flow rules::
2130 flow flush {port_id}
2132 - Query an existing flow rule::
2134 flow query {port_id} {rule_id} {action}
2136 - List existing flow rules sorted by priority, filtered by group
2139 flow list {port_id} [group {group_id}] [...]
2141 Validating flow rules
2142 ~~~~~~~~~~~~~~~~~~~~~
2144 ``flow validate`` reports whether a flow rule would be accepted by the
2145 underlying device in its current state but stops short of creating it. It is
2146 bound to ``rte_flow_validate()``::
2148 flow validate {port_id}
2149 [group {group_id}] [priority {level}] [ingress] [egress]
2150 pattern {item} [/ {item} [...]] / end
2151 actions {action} [/ {action} [...]] / end
2153 If successful, it will show::
2157 Otherwise it will show an error message of the form::
2159 Caught error type [...] ([...]): [...]
2161 This command uses the same parameters as ``flow create``, their format is
2162 described in `Creating flow rules`_.
2164 Check whether redirecting any Ethernet packet received on port 0 to RX queue
2165 index 6 is supported::
2167 testpmd> flow validate 0 ingress pattern eth / end
2168 actions queue index 6 / end
2172 Port 0 does not support TCPv6 rules::
2174 testpmd> flow validate 0 ingress pattern eth / ipv6 / tcp / end
2176 Caught error type 9 (specific pattern item): Invalid argument
2182 ``flow create`` validates and creates the specified flow rule. It is bound
2183 to ``rte_flow_create()``::
2185 flow create {port_id}
2186 [group {group_id}] [priority {level}] [ingress] [egress]
2187 pattern {item} [/ {item} [...]] / end
2188 actions {action} [/ {action} [...]] / end
2190 If successful, it will return a flow rule ID usable with other commands::
2192 Flow rule #[...] created
2194 Otherwise it will show an error message of the form::
2196 Caught error type [...] ([...]): [...]
2198 Parameters describe in the following order:
2200 - Attributes (*group*, *priority*, *ingress*, *egress* tokens).
2201 - A matching pattern, starting with the *pattern* token and terminated by an
2203 - Actions, starting with the *actions* token and terminated by an *end*
2206 These translate directly to *rte_flow* objects provided as-is to the
2207 underlying functions.
2209 The shortest valid definition only comprises mandatory tokens::
2211 testpmd> flow create 0 pattern end actions end
2213 Note that PMDs may refuse rules that essentially do nothing such as this
2216 **All unspecified object values are automatically initialized to 0.**
2221 These tokens affect flow rule attributes (``struct rte_flow_attr``) and are
2222 specified before the ``pattern`` token.
2224 - ``group {group id}``: priority group.
2225 - ``priority {level}``: priority level within group.
2226 - ``ingress``: rule applies to ingress traffic.
2227 - ``egress``: rule applies to egress traffic.
2229 Each instance of an attribute specified several times overrides the previous
2230 value as shown below (group 4 is used)::
2232 testpmd> flow create 0 group 42 group 24 group 4 [...]
2234 Note that once enabled, ``ingress`` and ``egress`` cannot be disabled.
2236 While not specifying a direction is an error, some rules may allow both
2239 Most rules affect RX therefore contain the ``ingress`` token::
2241 testpmd> flow create 0 ingress pattern [...]
2246 A matching pattern starts after the ``pattern`` token. It is made of pattern
2247 items and is terminated by a mandatory ``end`` item.
2249 Items are named after their type (*RTE_FLOW_ITEM_TYPE_* from ``enum
2250 rte_flow_item_type``).
2252 The ``/`` token is used as a separator between pattern items as shown
2255 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end [...]
2257 Note that protocol items like these must be stacked from lowest to highest
2258 layer to make sense. For instance, the following rule is either invalid or
2259 unlikely to match any packet::
2261 testpmd> flow create 0 ingress pattern eth / udp / ipv4 / end [...]
2263 More information on these restrictions can be found in the *rte_flow*
2266 Several items support additional specification structures, for example
2267 ``ipv4`` allows specifying source and destination addresses as follows::
2269 testpmd> flow create 0 ingress pattern eth / ipv4 src is 10.1.1.1
2270 dst is 10.2.0.0 / end [...]
2272 This rule matches all IPv4 traffic with the specified properties.
2274 In this example, ``src`` and ``dst`` are field names of the underlying
2275 ``struct rte_flow_item_ipv4`` object. All item properties can be specified
2276 in a similar fashion.
2278 The ``is`` token means that the subsequent value must be matched exactly,
2279 and assigns ``spec`` and ``mask`` fields in ``struct rte_flow_item``
2280 accordingly. Possible assignment tokens are:
2282 - ``is``: match value perfectly (with full bit-mask).
2283 - ``spec``: match value according to configured bit-mask.
2284 - ``last``: specify upper bound to establish a range.
2285 - ``mask``: specify bit-mask with relevant bits set to one.
2286 - ``prefix``: generate bit-mask from a prefix length.
2288 These yield identical results::
2290 ipv4 src is 10.1.1.1
2294 ipv4 src spec 10.1.1.1 src mask 255.255.255.255
2298 ipv4 src spec 10.1.1.1 src prefix 32
2302 ipv4 src is 10.1.1.1 src last 10.1.1.1 # range with a single value
2306 ipv4 src is 10.1.1.1 src last 0 # 0 disables range
2308 Inclusive ranges can be defined with ``last``::
2310 ipv4 src is 10.1.1.1 src last 10.2.3.4 # 10.1.1.1 to 10.2.3.4
2312 Note that ``mask`` affects both ``spec`` and ``last``::
2314 ipv4 src is 10.1.1.1 src last 10.2.3.4 src mask 255.255.0.0
2315 # matches 10.1.0.0 to 10.2.255.255
2317 Properties can be modified multiple times::
2319 ipv4 src is 10.1.1.1 src is 10.1.2.3 src is 10.2.3.4 # matches 10.2.3.4
2323 ipv4 src is 10.1.1.1 src prefix 24 src prefix 16 # matches 10.1.0.0/16
2328 This section lists supported pattern items and their attributes, if any.
2330 - ``end``: end list of pattern items.
2332 - ``void``: no-op pattern item.
2334 - ``invert``: perform actions when pattern does not match.
2336 - ``any``: match any protocol for the current layer.
2338 - ``num {unsigned}``: number of layers covered.
2340 - ``pf``: match packets addressed to the physical function.
2342 - ``vf``: match packets addressed to a virtual function ID.
2344 - ``id {unsigned}``: destination VF ID.
2346 - ``port``: device-specific physical port index to use.
2348 - ``index {unsigned}``: physical port index.
2350 - ``raw``: match an arbitrary byte string.
2352 - ``relative {boolean}``: look for pattern after the previous item.
2353 - ``search {boolean}``: search pattern from offset (see also limit).
2354 - ``offset {integer}``: absolute or relative offset for pattern.
2355 - ``limit {unsigned}``: search area limit for start of pattern.
2356 - ``pattern {string}``: byte string to look for.
2358 - ``eth``: match Ethernet header.
2360 - ``dst {MAC-48}``: destination MAC.
2361 - ``src {MAC-48}``: source MAC.
2362 - ``type {unsigned}``: EtherType.
2364 - ``vlan``: match 802.1Q/ad VLAN tag.
2366 - ``tpid {unsigned}``: tag protocol identifier.
2367 - ``tci {unsigned}``: tag control information.
2368 - ``pcp {unsigned}``: priority code point.
2369 - ``dei {unsigned}``: drop eligible indicator.
2370 - ``vid {unsigned}``: VLAN identifier.
2372 - ``ipv4``: match IPv4 header.
2374 - ``tos {unsigned}``: type of service.
2375 - ``ttl {unsigned}``: time to live.
2376 - ``proto {unsigned}``: next protocol ID.
2377 - ``src {ipv4 address}``: source address.
2378 - ``dst {ipv4 address}``: destination address.
2380 - ``ipv6``: match IPv6 header.
2382 - ``tc {unsigned}``: traffic class.
2383 - ``flow {unsigned}``: flow label.
2384 - ``proto {unsigned}``: protocol (next header).
2385 - ``hop {unsigned}``: hop limit.
2386 - ``src {ipv6 address}``: source address.
2387 - ``dst {ipv6 address}``: destination address.
2389 - ``icmp``: match ICMP header.
2391 - ``type {unsigned}``: ICMP packet type.
2392 - ``code {unsigned}``: ICMP packet code.
2394 - ``udp``: match UDP header.
2396 - ``src {unsigned}``: UDP source port.
2397 - ``dst {unsigned}``: UDP destination port.
2399 - ``tcp``: match TCP header.
2401 - ``src {unsigned}``: TCP source port.
2402 - ``dst {unsigned}``: TCP destination port.
2404 - ``sctp``: match SCTP header.
2406 - ``src {unsigned}``: SCTP source port.
2407 - ``dst {unsigned}``: SCTP destination port.
2408 - ``tag {unsigned}``: validation tag.
2409 - ``cksum {unsigned}``: checksum.
2411 - ``vxlan``: match VXLAN header.
2413 - ``vni {unsigned}``: VXLAN identifier.
2418 A list of actions starts after the ``actions`` token in the same fashion as
2419 `Matching pattern`_; actions are separated by ``/`` tokens and the list is
2420 terminated by a mandatory ``end`` action.
2422 Actions are named after their type (*RTE_FLOW_ACTION_TYPE_* from ``enum
2423 rte_flow_action_type``).
2425 Dropping all incoming UDPv4 packets can be expressed as follows::
2427 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
2430 Several actions have configurable properties which must be specified when
2431 there is no valid default value. For example, ``queue`` requires a target
2434 This rule redirects incoming UDPv4 traffic to queue index 6::
2436 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
2437 actions queue index 6 / end
2439 While this one could be rejected by PMDs (unspecified queue index)::
2441 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
2444 As defined by *rte_flow*, the list is not ordered, all actions of a given
2445 rule are performed simultaneously. These are equivalent::
2447 queue index 6 / void / mark id 42 / end
2451 void / mark id 42 / queue index 6 / end
2453 All actions in a list should have different types, otherwise only the last
2454 action of a given type is taken into account::
2456 queue index 4 / queue index 5 / queue index 6 / end # will use queue 6
2460 drop / drop / drop / end # drop is performed only once
2464 mark id 42 / queue index 3 / mark id 24 / end # mark will be 24
2466 Considering they are performed simultaneously, opposite and overlapping
2467 actions can sometimes be combined when the end result is unambiguous::
2469 drop / queue index 6 / end # drop has no effect
2473 drop / dup index 6 / end # same as above
2477 queue index 6 / rss queues 6 7 8 / end # queue has no effect
2481 drop / passthru / end # drop has no effect
2483 Note that PMDs may still refuse such combinations.
2488 This section lists supported actions and their attributes, if any.
2490 - ``end``: end list of actions.
2492 - ``void``: no-op action.
2494 - ``passthru``: let subsequent rule process matched packets.
2496 - ``mark``: attach 32 bit value to packets.
2498 - ``id {unsigned}``: 32 bit value to return with packets.
2500 - ``flag``: flag packets.
2502 - ``queue``: assign packets to a given queue index.
2504 - ``index {unsigned}``: queue index to use.
2506 - ``drop``: drop packets (note: passthru has priority).
2508 - ``count``: enable counters for this rule.
2510 - ``dup``: duplicate packets to a given queue index.
2512 - ``index {unsigned}``: queue index to duplicate packets to.
2514 - ``rss``: spread packets among several queues.
2516 - ``queues [{unsigned} [...]] end``: queue indices to use.
2518 - ``pf``: redirect packets to physical device function.
2520 - ``vf``: redirect packets to virtual device function.
2522 - ``original {boolean}``: use original VF ID if possible.
2523 - ``id {unsigned}``: VF ID to redirect packets to.
2525 Destroying flow rules
2526 ~~~~~~~~~~~~~~~~~~~~~
2528 ``flow destroy`` destroys one or more rules from their rule ID (as returned
2529 by ``flow create``), this command calls ``rte_flow_destroy()`` as many
2530 times as necessary::
2532 flow destroy {port_id} rule {rule_id} [...]
2534 If successful, it will show::
2536 Flow rule #[...] destroyed
2538 It does not report anything for rule IDs that do not exist. The usual error
2539 message is shown when a rule cannot be destroyed::
2541 Caught error type [...] ([...]): [...]
2543 ``flow flush`` destroys all rules on a device and does not take extra
2544 arguments. It is bound to ``rte_flow_flush()``::
2546 flow flush {port_id}
2548 Any errors are reported as above.
2550 Creating several rules and destroying them::
2552 testpmd> flow create 0 ingress pattern eth / ipv6 / end
2553 actions queue index 2 / end
2554 Flow rule #0 created
2555 testpmd> flow create 0 ingress pattern eth / ipv4 / end
2556 actions queue index 3 / end
2557 Flow rule #1 created
2558 testpmd> flow destroy 0 rule 0 rule 1
2559 Flow rule #1 destroyed
2560 Flow rule #0 destroyed
2563 The same result can be achieved using ``flow flush``::
2565 testpmd> flow create 0 ingress pattern eth / ipv6 / end
2566 actions queue index 2 / end
2567 Flow rule #0 created
2568 testpmd> flow create 0 ingress pattern eth / ipv4 / end
2569 actions queue index 3 / end
2570 Flow rule #1 created
2571 testpmd> flow flush 0
2574 Non-existent rule IDs are ignored::
2576 testpmd> flow create 0 ingress pattern eth / ipv6 / end
2577 actions queue index 2 / end
2578 Flow rule #0 created
2579 testpmd> flow create 0 ingress pattern eth / ipv4 / end
2580 actions queue index 3 / end
2581 Flow rule #1 created
2582 testpmd> flow destroy 0 rule 42 rule 10 rule 2
2584 testpmd> flow destroy 0 rule 0
2585 Flow rule #0 destroyed
2591 ``flow query`` queries a specific action of a flow rule having that
2592 ability. Such actions collect information that can be reported using this
2593 command. It is bound to ``rte_flow_query()``::
2595 flow query {port_id} {rule_id} {action}
2597 If successful, it will display either the retrieved data for known actions
2598 or the following message::
2600 Cannot display result for action type [...] ([...])
2602 Otherwise, it will complain either that the rule does not exist or that some
2605 Flow rule #[...] not found
2609 Caught error type [...] ([...]): [...]
2611 Currently only the ``count`` action is supported. This action reports the
2612 number of packets that hit the flow rule and the total number of bytes. Its
2613 output has the following format::
2616 hits_set: [...] # whether "hits" contains a valid value
2617 bytes_set: [...] # whether "bytes" contains a valid value
2618 hits: [...] # number of packets
2619 bytes: [...] # number of bytes
2621 Querying counters for TCPv6 packets redirected to queue 6::
2623 testpmd> flow create 0 ingress pattern eth / ipv6 / tcp / end
2624 actions queue index 6 / count / end
2625 Flow rule #4 created
2626 testpmd> flow query 0 4 count
2637 ``flow list`` lists existing flow rules sorted by priority and optionally
2638 filtered by group identifiers::
2640 flow list {port_id} [group {group_id}] [...]
2642 This command only fails with the following message if the device does not
2647 Output consists of a header line followed by a short description of each
2648 flow rule, one per line. There is no output at all when no flow rules are
2649 configured on the device::
2651 ID Group Prio Attr Rule
2652 [...] [...] [...] [...] [...]
2654 ``Attr`` column flags:
2656 - ``i`` for ``ingress``.
2657 - ``e`` for ``egress``.
2659 Creating several flow rules and listing them::
2661 testpmd> flow create 0 ingress pattern eth / ipv4 / end
2662 actions queue index 6 / end
2663 Flow rule #0 created
2664 testpmd> flow create 0 ingress pattern eth / ipv6 / end
2665 actions queue index 2 / end
2666 Flow rule #1 created
2667 testpmd> flow create 0 priority 5 ingress pattern eth / ipv4 / udp / end
2668 actions rss queues 6 7 8 end / end
2669 Flow rule #2 created
2670 testpmd> flow list 0
2671 ID Group Prio Attr Rule
2672 0 0 0 i- ETH IPV4 => QUEUE
2673 1 0 0 i- ETH IPV6 => QUEUE
2674 2 0 5 i- ETH IPV4 UDP => RSS
2677 Rules are sorted by priority (i.e. group ID first, then priority level)::
2679 testpmd> flow list 1
2680 ID Group Prio Attr Rule
2681 0 0 0 i- ETH => COUNT
2682 6 0 500 i- ETH IPV6 TCP => DROP COUNT
2683 5 0 1000 i- ETH IPV6 ICMP => QUEUE
2684 1 24 0 i- ETH IPV4 UDP => QUEUE
2685 4 24 10 i- ETH IPV4 TCP => DROP
2686 3 24 20 i- ETH IPV4 => DROP
2687 2 24 42 i- ETH IPV4 UDP => QUEUE
2688 7 63 0 i- ETH IPV6 UDP VXLAN => MARK QUEUE
2691 Output can be limited to specific groups::
2693 testpmd> flow list 1 group 0 group 63
2694 ID Group Prio Attr Rule
2695 0 0 0 i- ETH => COUNT
2696 6 0 500 i- ETH IPV6 TCP => DROP COUNT
2697 5 0 1000 i- ETH IPV6 ICMP => QUEUE
2698 7 63 0 i- ETH IPV6 UDP VXLAN => MARK QUEUE