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 Set the VLAN strip on a port::
515 testpmd> vlan set strip (on|off) (port_id)
520 Set the VLAN strip for a queue on a port::
522 testpmd> vlan set stripq (on|off) (port_id,queue_id)
524 vlan set stripq (for VF)
525 ~~~~~~~~~~~~~~~~~~~~~~~~
527 Set VLAN strip for all queues in a pool for a VF from the PF::
529 testpmd> set vf vlan stripq (port_id) (vf_id) (on|off)
531 vlan set insert (for VF)
532 ~~~~~~~~~~~~~~~~~~~~~~~~
534 Set VLAN insert for a VF from the PF::
536 testpmd> set vf vlan insert (port_id) (vf_id) (vlan_id)
538 vlan set antispoof (for VF)
539 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
541 Set VLAN antispoof for a VF from the PF::
543 testpmd> set vf vlan antispoof (port_id) (vf_id) (on|off)
548 Set the VLAN filter on a port::
550 testpmd> vlan set filter (on|off) (port_id)
555 Set the VLAN QinQ (extended queue in queue) on for a port::
557 testpmd> vlan set qinq (on|off) (port_id)
562 Set the inner or outer VLAN TPID for packet filtering on a port::
564 testpmd> vlan set (inner|outer) tpid (value) (port_id)
568 TPID value must be a 16-bit number (value <= 65536).
573 Add a VLAN ID, or all identifiers, to the set of VLAN identifiers filtered by port ID::
575 testpmd> rx_vlan add (vlan_id|all) (port_id)
579 VLAN filter must be set on that port. VLAN ID < 4096.
580 Depending on the NIC used, number of vlan_ids may be limited to the maximum entries
581 in VFTA table. This is important if enabling all vlan_ids.
586 Remove a VLAN ID, or all identifiers, from the set of VLAN identifiers filtered by port ID::
588 testpmd> rx_vlan rm (vlan_id|all) (port_id)
593 Add a VLAN ID, to the set of VLAN identifiers filtered for VF(s) for port ID::
595 testpmd> rx_vlan add (vlan_id) port (port_id) vf (vf_mask)
600 Remove a VLAN ID, from the set of VLAN identifiers filtered for VF(s) for port ID::
602 testpmd> rx_vlan rm (vlan_id) port (port_id) vf (vf_mask)
607 Add a tunnel filter on a port::
609 testpmd> tunnel_filter add (port_id) (outer_mac) (inner_mac) (ip_addr) \
610 (inner_vlan) (vxlan|nvgre|ipingre) (imac-ivlan|imac-ivlan-tenid|\
611 imac-tenid|imac|omac-imac-tenid|oip|iip) (tenant_id) (queue_id)
613 The available information categories are:
615 * ``vxlan``: Set tunnel type as VXLAN.
617 * ``nvgre``: Set tunnel type as NVGRE.
619 * ``ipingre``: Set tunnel type as IP-in-GRE.
621 * ``imac-ivlan``: Set filter type as Inner MAC and VLAN.
623 * ``imac-ivlan-tenid``: Set filter type as Inner MAC, VLAN and tenant ID.
625 * ``imac-tenid``: Set filter type as Inner MAC and tenant ID.
627 * ``imac``: Set filter type as Inner MAC.
629 * ``omac-imac-tenid``: Set filter type as Outer MAC, Inner MAC and tenant ID.
631 * ``oip``: Set filter type as Outer IP.
633 * ``iip``: Set filter type as Inner IP.
637 testpmd> tunnel_filter add 0 68:05:CA:28:09:82 00:00:00:00:00:00 \
638 192.168.2.2 0 ipingre oip 1 1
640 Set an IP-in-GRE tunnel on port 0, and the filter type is Outer IP.
645 Remove a tunnel filter on a port::
647 testpmd> tunnel_filter rm (port_id) (outer_mac) (inner_mac) (ip_addr) \
648 (inner_vlan) (vxlan|nvgre|ipingre) (imac-ivlan|imac-ivlan-tenid|\
649 imac-tenid|imac|omac-imac-tenid|oip|iip) (tenant_id) (queue_id)
654 Add an UDP port for VXLAN packet filter on a port::
656 testpmd> rx_vxlan_port add (udp_port) (port_id)
661 Remove an UDP port for VXLAN packet filter on a port::
663 testpmd> rx_vxlan_port rm (udp_port) (port_id)
668 Set hardware insertion of VLAN IDs in packets sent on a port::
670 testpmd> tx_vlan set (port_id) vlan_id[, vlan_id_outer]
672 For example, set a single VLAN ID (5) insertion on port 0::
676 Or, set double VLAN ID (inner: 2, outer: 3) insertion on port 1::
684 Set port based hardware insertion of VLAN ID in packets sent on a port::
686 testpmd> tx_vlan set pvid (port_id) (vlan_id) (on|off)
691 Disable hardware insertion of a VLAN header in packets sent on a port::
693 testpmd> tx_vlan reset (port_id)
698 Select hardware or software calculation of the checksum when
699 transmitting a packet using the ``csum`` forwarding engine::
701 testpmd> csum set (ip|udp|tcp|sctp|outer-ip) (hw|sw) (port_id)
705 * ``ip|udp|tcp|sctp`` always relate to the inner layer.
707 * ``outer-ip`` relates to the outer IP layer (only for IPv4) in the case where the packet is recognized
708 as a tunnel packet by the forwarding engine (vxlan, gre and ipip are
709 supported). See also the ``csum parse-tunnel`` command.
713 Check the NIC Datasheet for hardware limits.
718 Define how tunneled packets should be handled by the csum forward
721 testpmd> csum parse-tunnel (on|off) (tx_port_id)
723 If enabled, the csum forward engine will try to recognize supported
724 tunnel headers (vxlan, gre, ipip).
726 If disabled, treat tunnel packets as non-tunneled packets (a inner
727 header is handled as a packet payload).
731 The port argument is the TX port like in the ``csum set`` command.
735 Consider a packet in packet like the following::
737 eth_out/ipv4_out/udp_out/vxlan/eth_in/ipv4_in/tcp_in
739 * If parse-tunnel is enabled, the ``ip|udp|tcp|sctp`` parameters of ``csum set``
740 command relate to the inner headers (here ``ipv4_in`` and ``tcp_in``), and the
741 ``outer-ip parameter`` relates to the outer headers (here ``ipv4_out``).
743 * If parse-tunnel is disabled, the ``ip|udp|tcp|sctp`` parameters of ``csum set``
744 command relate to the outer headers, here ``ipv4_out`` and ``udp_out``.
749 Display tx checksum offload configuration::
751 testpmd> csum show (port_id)
756 Enable TCP Segmentation Offload (TSO) in the ``csum`` forwarding engine::
758 testpmd> tso set (segsize) (port_id)
762 Check the NIC datasheet for hardware limits.
767 Display the status of TCP Segmentation Offload::
769 testpmd> tso show (port_id)
774 Add an alternative MAC address to a port::
776 testpmd> mac_addr add (port_id) (XX:XX:XX:XX:XX:XX)
781 Remove a MAC address from a port::
783 testpmd> mac_addr remove (port_id) (XX:XX:XX:XX:XX:XX)
785 mac_addr add (for VF)
786 ~~~~~~~~~~~~~~~~~~~~~
788 Add an alternative MAC address for a VF to a port::
790 testpmd> mac_add add port (port_id) vf (vf_id) (XX:XX:XX:XX:XX:XX)
792 mac_addr set (for VF)
793 ~~~~~~~~~~~~~~~~~~~~~
795 Set the MAC address for a VF from the PF::
797 testpmd> set vf mac addr (port_id) (vf_id) (XX:XX:XX:XX:XX:XX)
802 Set the unicast hash filter(s) on/off for a port::
804 testpmd> set port (port_id) uta (XX:XX:XX:XX:XX:XX|all) (on|off)
809 Set the promiscuous mode on for a port or for all ports.
810 In promiscuous mode packets are not dropped if they aren't for the specified MAC address::
812 testpmd> set promisc (port_id|all) (on|off)
817 Set the allmulti mode for a port or for all ports::
819 testpmd> set allmulti (port_id|all) (on|off)
821 Same as the ifconfig (8) option. Controls how multicast packets are handled.
826 Set the link flow control parameter on a port::
828 testpmd> set flow_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \
829 (pause_time) (send_xon) mac_ctrl_frame_fwd (on|off) \
830 autoneg (on|off) (port_id)
834 * ``high_water`` (integer): High threshold value to trigger XOFF.
836 * ``low_water`` (integer): Low threshold value to trigger XON.
838 * ``pause_time`` (integer): Pause quota in the Pause frame.
840 * ``send_xon`` (0/1): Send XON frame.
842 * ``mac_ctrl_frame_fwd``: Enable receiving MAC control frames.
844 * ``autoneg``: Change the auto-negotiation para mete.
849 Set the priority flow control parameter on a port::
851 testpmd> set pfc_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \
852 (pause_time) (priority) (port_id)
856 * ``high_water`` (integer): High threshold value.
858 * ``low_water`` (integer): Low threshold value.
860 * ``pause_time`` (integer): Pause quota in the Pause frame.
862 * ``priority`` (0-7): VLAN User Priority.
867 Set statistics mapping (qmapping 0..15) for RX/TX queue on port::
869 testpmd> set stat_qmap (tx|rx) (port_id) (queue_id) (qmapping)
871 For example, to set rx queue 2 on port 0 to mapping 5::
873 testpmd>set stat_qmap rx 0 2 5
875 set port - rx/tx (for VF)
876 ~~~~~~~~~~~~~~~~~~~~~~~~~
878 Set VF receive/transmit from a port::
880 testpmd> set port (port_id) vf (vf_id) (rx|tx) (on|off)
882 set port - mac address filter (for VF)
883 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
885 Add/Remove unicast or multicast MAC addr filter for a VF::
887 testpmd> set port (port_id) vf (vf_id) (mac_addr) \
888 (exact-mac|exact-mac-vlan|hashmac|hashmac-vlan) (on|off)
890 set port - rx mode(for VF)
891 ~~~~~~~~~~~~~~~~~~~~~~~~~~
893 Set the VF receive mode of a port::
895 testpmd> set port (port_id) vf (vf_id) \
896 rxmode (AUPE|ROPE|BAM|MPE) (on|off)
898 The available receive modes are:
900 * ``AUPE``: Accepts untagged VLAN.
902 * ``ROPE``: Accepts unicast hash.
904 * ``BAM``: Accepts broadcast packets.
906 * ``MPE``: Accepts all multicast packets.
908 set port - tx_rate (for Queue)
909 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
911 Set TX rate limitation for a queue on a port::
913 testpmd> set port (port_id) queue (queue_id) rate (rate_value)
915 set port - tx_rate (for VF)
916 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
918 Set TX rate limitation for queues in VF on a port::
920 testpmd> set port (port_id) vf (vf_id) rate (rate_value) queue_mask (queue_mask)
922 set port - mirror rule
923 ~~~~~~~~~~~~~~~~~~~~~~
925 Set pool or vlan type mirror rule for a port::
927 testpmd> set port (port_id) mirror-rule (rule_id) \
928 (pool-mirror-up|pool-mirror-down|vlan-mirror) \
929 (poolmask|vlanid[,vlanid]*) dst-pool (pool_id) (on|off)
931 Set link mirror rule for a port::
933 testpmd> set port (port_id) mirror-rule (rule_id) \
934 (uplink-mirror|downlink-mirror) dst-pool (pool_id) (on|off)
936 For example to enable mirror traffic with vlan 0,1 to pool 0::
938 set port 0 mirror-rule 0 vlan-mirror 0,1 dst-pool 0 on
940 reset port - mirror rule
941 ~~~~~~~~~~~~~~~~~~~~~~~~
943 Reset a mirror rule for a port::
945 testpmd> reset port (port_id) mirror-rule (rule_id)
950 Set the flush on RX streams before forwarding.
951 The default is flush ``on``.
952 Mainly used with PCAP drivers to turn off the default behavior of flushing the first 512 packets on RX streams::
954 testpmd> set flush_rx off
959 Set the bypass mode for the lowest port on bypass enabled NIC::
961 testpmd> set bypass mode (normal|bypass|isolate) (port_id)
966 Set the event required to initiate specified bypass mode for the lowest port on a bypass enabled::
968 testpmd> set bypass event (timeout|os_on|os_off|power_on|power_off) \
969 mode (normal|bypass|isolate) (port_id)
973 * ``timeout``: Enable bypass after watchdog timeout.
975 * ``os_on``: Enable bypass when OS/board is powered on.
977 * ``os_off``: Enable bypass when OS/board is powered off.
979 * ``power_on``: Enable bypass when power supply is turned on.
981 * ``power_off``: Enable bypass when power supply is turned off.
987 Set the bypass watchdog timeout to ``n`` seconds where 0 = instant::
989 testpmd> set bypass timeout (0|1.5|2|3|4|8|16|32)
994 Show the bypass configuration for a bypass enabled NIC using the lowest port on the NIC::
996 testpmd> show bypass config (port_id)
1001 Set link up for a port::
1003 testpmd> set link-up port (port id)
1008 Set link down for a port::
1010 testpmd> set link-down port (port id)
1015 Enable E-tag insertion for a VF on a port::
1017 testpmd> E-tag set insertion on port-tag-id (value) port (port_id) vf (vf_id)
1019 Disable E-tag insertion for a VF on a port::
1021 testpmd> E-tag set insertion off port (port_id) vf (vf_id)
1023 Enable/disable E-tag stripping on a port::
1025 testpmd> E-tag set stripping (on|off) port (port_id)
1027 Enable/disable E-tag based forwarding on a port::
1029 testpmd> E-tag set forwarding (on|off) port (port_id)
1031 Add an E-tag forwarding filter on a port::
1033 testpmd> E-tag set filter add e-tag-id (value) dst-pool (pool_id) port (port_id)
1035 Delete an E-tag forwarding filter on a port::
1036 testpmd> E-tag set filter del e-tag-id (value) port (port_id)
1042 The following sections show functions for configuring ports.
1046 Port configuration changes only become active when forwarding is started/restarted.
1051 Attach a port specified by pci address or virtual device args::
1053 testpmd> port attach (identifier)
1055 To attach a new pci device, the device should be recognized by kernel first.
1056 Then it should be moved under DPDK management.
1057 Finally the port can be attached to testpmd.
1059 For example, to move a pci device using ixgbe under DPDK management:
1061 .. code-block:: console
1063 # Check the status of the available devices.
1064 ./tools/dpdk-devbind.py --status
1066 Network devices using DPDK-compatible driver
1067 ============================================
1070 Network devices using kernel driver
1071 ===================================
1072 0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused=
1075 # Bind the device to igb_uio.
1076 sudo ./tools/dpdk-devbind.py -b igb_uio 0000:0a:00.0
1079 # Recheck the status of the devices.
1080 ./tools/dpdk-devbind.py --status
1081 Network devices using DPDK-compatible driver
1082 ============================================
1083 0000:0a:00.0 '82599ES 10-Gigabit' drv=igb_uio unused=
1085 To attach a port created by virtual device, above steps are not needed.
1087 For example, to attach a port whose pci address is 0000:0a:00.0.
1089 .. code-block:: console
1091 testpmd> port attach 0000:0a:00.0
1092 Attaching a new port...
1093 EAL: PCI device 0000:0a:00.0 on NUMA socket -1
1094 EAL: probe driver: 8086:10fb rte_ixgbe_pmd
1095 EAL: PCI memory mapped at 0x7f83bfa00000
1096 EAL: PCI memory mapped at 0x7f83bfa80000
1097 PMD: eth_ixgbe_dev_init(): MAC: 2, PHY: 18, SFP+: 5
1098 PMD: eth_ixgbe_dev_init(): port 0 vendorID=0x8086 deviceID=0x10fb
1099 Port 0 is attached. Now total ports is 1
1102 For example, to attach a port created by pcap PMD.
1104 .. code-block:: console
1106 testpmd> port attach net_pcap0
1107 Attaching a new port...
1108 PMD: Initializing pmd_pcap for net_pcap0
1109 PMD: Creating pcap-backed ethdev on numa socket 0
1110 Port 0 is attached. Now total ports is 1
1113 In this case, identifier is ``net_pcap0``.
1114 This identifier format is the same as ``--vdev`` format of DPDK applications.
1116 For example, to re-attach a bonded port which has been previously detached,
1117 the mode and slave parameters must be given.
1119 .. code-block:: console
1121 testpmd> port attach net_bond_0,mode=0,slave=1
1122 Attaching a new port...
1123 EAL: Initializing pmd_bond for net_bond_0
1124 EAL: Create bonded device net_bond_0 on port 0 in mode 0 on socket 0.
1125 Port 0 is attached. Now total ports is 1
1132 Detach a specific port::
1134 testpmd> port detach (port_id)
1136 Before detaching a port, the port should be stopped and closed.
1138 For example, to detach a pci device port 0.
1140 .. code-block:: console
1142 testpmd> port stop 0
1145 testpmd> port close 0
1149 testpmd> port detach 0
1151 EAL: PCI device 0000:0a:00.0 on NUMA socket -1
1152 EAL: remove driver: 8086:10fb rte_ixgbe_pmd
1153 EAL: PCI memory unmapped at 0x7f83bfa00000
1154 EAL: PCI memory unmapped at 0x7f83bfa80000
1158 For example, to detach a virtual device port 0.
1160 .. code-block:: console
1162 testpmd> port stop 0
1165 testpmd> port close 0
1169 testpmd> port detach 0
1171 PMD: Closing pcap ethdev on numa socket 0
1172 Port 'net_pcap0' is detached. Now total ports is 0
1175 To remove a pci device completely from the system, first detach the port from testpmd.
1176 Then the device should be moved under kernel management.
1177 Finally the device can be removed using kernel pci hotplug functionality.
1179 For example, to move a pci device under kernel management:
1181 .. code-block:: console
1183 sudo ./tools/dpdk-devbind.py -b ixgbe 0000:0a:00.0
1185 ./tools/dpdk-devbind.py --status
1187 Network devices using DPDK-compatible driver
1188 ============================================
1191 Network devices using kernel driver
1192 ===================================
1193 0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused=igb_uio
1195 To remove a port created by a virtual device, above steps are not needed.
1200 Start all ports or a specific port::
1202 testpmd> port start (port_id|all)
1207 Stop all ports or a specific port::
1209 testpmd> port stop (port_id|all)
1214 Close all ports or a specific port::
1216 testpmd> port close (port_id|all)
1218 port start/stop queue
1219 ~~~~~~~~~~~~~~~~~~~~~
1221 Start/stop a rx/tx queue on a specific port::
1223 testpmd> port (port_id) (rxq|txq) (queue_id) (start|stop)
1225 Only take effect when port is started.
1230 Set the speed and duplex mode for all ports or a specific port::
1232 testpmd> port config (port_id|all) speed (10|100|1000|10000|25000|40000|50000|100000|auto) \
1233 duplex (half|full|auto)
1235 port config - queues/descriptors
1236 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1238 Set number of queues/descriptors for rxq, txq, rxd and txd::
1240 testpmd> port config all (rxq|txq|rxd|txd) (value)
1242 This is equivalent to the ``--rxq``, ``--txq``, ``--rxd`` and ``--txd`` command-line options.
1244 port config - max-pkt-len
1245 ~~~~~~~~~~~~~~~~~~~~~~~~~
1247 Set the maximum packet length::
1249 testpmd> port config all max-pkt-len (value)
1251 This is equivalent to the ``--max-pkt-len`` command-line option.
1253 port config - CRC Strip
1254 ~~~~~~~~~~~~~~~~~~~~~~~
1256 Set hardware CRC stripping on or off for all ports::
1258 testpmd> port config all crc-strip (on|off)
1260 CRC stripping is off by default.
1262 The ``on`` option is equivalent to the ``--crc-strip`` command-line option.
1264 port config - scatter
1265 ~~~~~~~~~~~~~~~~~~~~~~~
1267 Set RX scatter mode on or off for all ports::
1269 testpmd> port config all scatter (on|off)
1271 RX scatter mode is off by default.
1273 The ``on`` option is equivalent to the ``--enable-scatter`` command-line option.
1275 port config - TX queue flags
1276 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1278 Set a hexadecimal bitmap of TX queue flags for all ports::
1280 testpmd> port config all txqflags value
1282 This command is equivalent to the ``--txqflags`` command-line option.
1284 port config - RX Checksum
1285 ~~~~~~~~~~~~~~~~~~~~~~~~~
1287 Set hardware RX checksum offload to on or off for all ports::
1289 testpmd> port config all rx-cksum (on|off)
1291 Checksum offload is off by default.
1293 The ``on`` option is equivalent to the ``--enable-rx-cksum`` command-line option.
1298 Set hardware VLAN on or off for all ports::
1300 testpmd> port config all hw-vlan (on|off)
1302 Hardware VLAN is on by default.
1304 The ``off`` option is equivalent to the ``--disable-hw-vlan`` command-line option.
1306 port config - VLAN filter
1307 ~~~~~~~~~~~~~~~~~~~~~~~~~
1309 Set hardware VLAN filter on or off for all ports::
1311 testpmd> port config all hw-vlan-filter (on|off)
1313 Hardware VLAN filter is on by default.
1315 The ``off`` option is equivalent to the ``--disable-hw-vlan-filter`` command-line option.
1317 port config - VLAN strip
1318 ~~~~~~~~~~~~~~~~~~~~~~~~
1320 Set hardware VLAN strip on or off for all ports::
1322 testpmd> port config all hw-vlan-strip (on|off)
1324 Hardware VLAN strip is on by default.
1326 The ``off`` option is equivalent to the ``--disable-hw-vlan-strip`` command-line option.
1328 port config - VLAN extend
1329 ~~~~~~~~~~~~~~~~~~~~~~~~~
1331 Set hardware VLAN extend on or off for all ports::
1333 testpmd> port config all hw-vlan-extend (on|off)
1335 Hardware VLAN extend is off by default.
1337 The ``off`` option is equivalent to the ``--disable-hw-vlan-extend`` command-line option.
1339 port config - Drop Packets
1340 ~~~~~~~~~~~~~~~~~~~~~~~~~~
1342 Set packet drop for packets with no descriptors on or off for all ports::
1344 testpmd> port config all drop-en (on|off)
1346 Packet dropping for packets with no descriptors is off by default.
1348 The ``on`` option is equivalent to the ``--enable-drop-en`` command-line option.
1353 Set the RSS (Receive Side Scaling) mode on or off::
1355 testpmd> port config all rss (all|ip|tcp|udp|sctp|ether|port|vxlan|geneve|nvgre|none)
1357 RSS is on by default.
1359 The ``none`` option is equivalent to the ``--disable-rss`` command-line option.
1361 port config - RSS Reta
1362 ~~~~~~~~~~~~~~~~~~~~~~
1364 Set the RSS (Receive Side Scaling) redirection table::
1366 testpmd> port config all rss reta (hash,queue)[,(hash,queue)]
1371 Set the DCB mode for an individual port::
1373 testpmd> port config (port_id) dcb vt (on|off) (traffic_class) pfc (on|off)
1375 The traffic class should be 4 or 8.
1380 Set the number of packets per burst::
1382 testpmd> port config all burst (value)
1384 This is equivalent to the ``--burst`` command-line option.
1386 port config - Threshold
1387 ~~~~~~~~~~~~~~~~~~~~~~~
1389 Set thresholds for TX/RX queues::
1391 testpmd> port config all (threshold) (value)
1393 Where the threshold type can be:
1395 * ``txpt:`` Set the prefetch threshold register of the TX rings, 0 <= value <= 255.
1397 * ``txht:`` Set the host threshold register of the TX rings, 0 <= value <= 255.
1399 * ``txwt:`` Set the write-back threshold register of the TX rings, 0 <= value <= 255.
1401 * ``rxpt:`` Set the prefetch threshold register of the RX rings, 0 <= value <= 255.
1403 * ``rxht:`` Set the host threshold register of the RX rings, 0 <= value <= 255.
1405 * ``rxwt:`` Set the write-back threshold register of the RX rings, 0 <= value <= 255.
1407 * ``txfreet:`` Set the transmit free threshold of the TX rings, 0 <= value <= txd.
1409 * ``rxfreet:`` Set the transmit free threshold of the RX rings, 0 <= value <= rxd.
1411 * ``txrst:`` Set the transmit RS bit threshold of TX rings, 0 <= value <= txd.
1413 These threshold options are also available from the command-line.
1418 Set the value of ether-type for E-tag::
1420 testpmd> port config (port_id|all) l2-tunnel E-tag ether-type (value)
1422 Enable/disable the E-tag support::
1424 testpmd> port config (port_id|all) l2-tunnel E-tag (enable|disable)
1427 Link Bonding Functions
1428 ----------------------
1430 The Link Bonding functions make it possible to dynamically create and
1431 manage link bonding devices from within testpmd interactive prompt.
1433 create bonded device
1434 ~~~~~~~~~~~~~~~~~~~~
1436 Create a new bonding device::
1438 testpmd> create bonded device (mode) (socket)
1440 For example, to create a bonded device in mode 1 on socket 0::
1442 testpmd> create bonded 1 0
1443 created new bonded device (port X)
1448 Adds Ethernet device to a Link Bonding device::
1450 testpmd> add bonding slave (slave id) (port id)
1452 For example, to add Ethernet device (port 6) to a Link Bonding device (port 10)::
1454 testpmd> add bonding slave 6 10
1457 remove bonding slave
1458 ~~~~~~~~~~~~~~~~~~~~
1460 Removes an Ethernet slave device from a Link Bonding device::
1462 testpmd> remove bonding slave (slave id) (port id)
1464 For example, to remove Ethernet slave device (port 6) to a Link Bonding device (port 10)::
1466 testpmd> remove bonding slave 6 10
1471 Set the Link Bonding mode of a Link Bonding device::
1473 testpmd> set bonding mode (value) (port id)
1475 For example, to set the bonding mode of a Link Bonding device (port 10) to broadcast (mode 3)::
1477 testpmd> set bonding mode 3 10
1482 Set an Ethernet slave device as the primary device on a Link Bonding device::
1484 testpmd> set bonding primary (slave id) (port id)
1486 For example, to set the Ethernet slave device (port 6) as the primary port of a Link Bonding device (port 10)::
1488 testpmd> set bonding primary 6 10
1493 Set the MAC address of a Link Bonding device::
1495 testpmd> set bonding mac (port id) (mac)
1497 For example, to set the MAC address of a Link Bonding device (port 10) to 00:00:00:00:00:01::
1499 testpmd> set bonding mac 10 00:00:00:00:00:01
1501 set bonding xmit_balance_policy
1502 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1504 Set the transmission policy for a Link Bonding device when it is in Balance XOR mode::
1506 testpmd> set bonding xmit_balance_policy (port_id) (l2|l23|l34)
1508 For example, set a Link Bonding device (port 10) to use a balance policy of layer 3+4 (IP addresses & UDP ports)::
1510 testpmd> set bonding xmit_balance_policy 10 l34
1513 set bonding mon_period
1514 ~~~~~~~~~~~~~~~~~~~~~~
1516 Set the link status monitoring polling period in milliseconds for a bonding device.
1518 This adds support for PMD slave devices which do not support link status interrupts.
1519 When the mon_period is set to a value greater than 0 then all PMD's which do not support
1520 link status ISR will be queried every polling interval to check if their link status has changed::
1522 testpmd> set bonding mon_period (port_id) (value)
1524 For example, to set the link status monitoring polling period of bonded device (port 5) to 150ms::
1526 testpmd> set bonding mon_period 5 150
1532 Show the current configuration of a Link Bonding device::
1534 testpmd> show bonding config (port id)
1537 to show the configuration a Link Bonding device (port 9) with 3 slave devices (1, 3, 4)
1538 in balance mode with a transmission policy of layer 2+3::
1540 testpmd> show bonding config 9
1542 Balance Xmit Policy: BALANCE_XMIT_POLICY_LAYER23
1544 Active Slaves (3): [1 3 4]
1551 The Register Functions can be used to read from and write to registers on the network card referenced by a port number.
1552 This is mainly useful for debugging purposes.
1553 Reference should be made to the appropriate datasheet for the network card for details on the register addresses
1554 and fields that can be accessed.
1559 Display the value of a port register::
1561 testpmd> read reg (port_id) (address)
1563 For example, to examine the Flow Director control register (FDIRCTL, 0x0000EE000) on an Intel 82599 10 GbE Controller::
1565 testpmd> read reg 0 0xEE00
1566 port 0 PCI register at offset 0xEE00: 0x4A060029 (1241907241)
1571 Display a port register bit field::
1573 testpmd> read regfield (port_id) (address) (bit_x) (bit_y)
1575 For example, reading the lowest two bits from the register in the example above::
1577 testpmd> read regfield 0 0xEE00 0 1
1578 port 0 PCI register at offset 0xEE00: bits[0, 1]=0x1 (1)
1583 Display a single port register bit::
1585 testpmd> read regbit (port_id) (address) (bit_x)
1587 For example, reading the lowest bit from the register in the example above::
1589 testpmd> read regbit 0 0xEE00 0
1590 port 0 PCI register at offset 0xEE00: bit 0=1
1595 Set the value of a port register::
1597 testpmd> write reg (port_id) (address) (value)
1599 For example, to clear a register::
1601 testpmd> write reg 0 0xEE00 0x0
1602 port 0 PCI register at offset 0xEE00: 0x00000000 (0)
1607 Set bit field of a port register::
1609 testpmd> write regfield (port_id) (address) (bit_x) (bit_y) (value)
1611 For example, writing to the register cleared in the example above::
1613 testpmd> write regfield 0 0xEE00 0 1 2
1614 port 0 PCI register at offset 0xEE00: 0x00000002 (2)
1619 Set single bit value of a port register::
1621 testpmd> write regbit (port_id) (address) (bit_x) (value)
1623 For example, to set the high bit in the register from the example above::
1625 testpmd> write regbit 0 0xEE00 31 1
1626 port 0 PCI register at offset 0xEE00: 0x8000000A (2147483658)
1632 This section details the available filter functions that are available.
1634 Note these functions interface the deprecated legacy filtering framework,
1635 superseded by *rte_flow*. See `Flow rules management`_.
1638 ~~~~~~~~~~~~~~~~~~~~
1640 Add or delete a L2 Ethertype filter, which identify packets by their L2 Ethertype mainly assign them to a receive queue::
1642 ethertype_filter (port_id) (add|del) (mac_addr|mac_ignr) (mac_address) \
1643 ethertype (ether_type) (drop|fwd) queue (queue_id)
1645 The available information parameters are:
1647 * ``port_id``: The port which the Ethertype filter assigned on.
1649 * ``mac_addr``: Compare destination mac address.
1651 * ``mac_ignr``: Ignore destination mac address match.
1653 * ``mac_address``: Destination mac address to match.
1655 * ``ether_type``: The EtherType value want to match,
1656 for example 0x0806 for ARP packet. 0x0800 (IPv4) and 0x86DD (IPv6) are invalid.
1658 * ``queue_id``: The receive queue associated with this EtherType filter.
1659 It is meaningless when deleting or dropping.
1661 Example, to add/remove an ethertype filter rule::
1663 testpmd> ethertype_filter 0 add mac_ignr 00:11:22:33:44:55 \
1664 ethertype 0x0806 fwd queue 3
1666 testpmd> ethertype_filter 0 del mac_ignr 00:11:22:33:44:55 \
1667 ethertype 0x0806 fwd queue 3
1672 Add or delete a 2-tuple filter,
1673 which identifies packets by specific protocol and destination TCP/UDP port
1674 and forwards packets into one of the receive queues::
1676 2tuple_filter (port_id) (add|del) dst_port (dst_port_value) \
1677 protocol (protocol_value) mask (mask_value) \
1678 tcp_flags (tcp_flags_value) priority (prio_value) \
1681 The available information parameters are:
1683 * ``port_id``: The port which the 2-tuple filter assigned on.
1685 * ``dst_port_value``: Destination port in L4.
1687 * ``protocol_value``: IP L4 protocol.
1689 * ``mask_value``: Participates in the match or not by bit for field above, 1b means participate.
1691 * ``tcp_flags_value``: TCP control bits. The non-zero value is invalid, when the pro_value is not set to 0x06 (TCP).
1693 * ``prio_value``: Priority of this filter.
1695 * ``queue_id``: The receive queue associated with this 2-tuple filter.
1697 Example, to add/remove an 2tuple filter rule::
1699 testpmd> 2tuple_filter 0 add dst_port 32 protocol 0x06 mask 0x03 \
1700 tcp_flags 0x02 priority 3 queue 3
1702 testpmd> 2tuple_filter 0 del dst_port 32 protocol 0x06 mask 0x03 \
1703 tcp_flags 0x02 priority 3 queue 3
1708 Add or delete a 5-tuple filter,
1709 which consists of a 5-tuple (protocol, source and destination IP addresses, source and destination TCP/UDP/SCTP port)
1710 and routes packets into one of the receive queues::
1712 5tuple_filter (port_id) (add|del) dst_ip (dst_address) src_ip \
1713 (src_address) dst_port (dst_port_value) \
1714 src_port (src_port_value) protocol (protocol_value) \
1715 mask (mask_value) tcp_flags (tcp_flags_value) \
1716 priority (prio_value) queue (queue_id)
1718 The available information parameters are:
1720 * ``port_id``: The port which the 5-tuple filter assigned on.
1722 * ``dst_address``: Destination IP address.
1724 * ``src_address``: Source IP address.
1726 * ``dst_port_value``: TCP/UDP destination port.
1728 * ``src_port_value``: TCP/UDP source port.
1730 * ``protocol_value``: L4 protocol.
1732 * ``mask_value``: Participates in the match or not by bit for field above, 1b means participate
1734 * ``tcp_flags_value``: TCP control bits. The non-zero value is invalid, when the protocol_value is not set to 0x06 (TCP).
1736 * ``prio_value``: The priority of this filter.
1738 * ``queue_id``: The receive queue associated with this 5-tuple filter.
1740 Example, to add/remove an 5tuple filter rule::
1742 testpmd> 5tuple_filter 0 add dst_ip 2.2.2.5 src_ip 2.2.2.4 \
1743 dst_port 64 src_port 32 protocol 0x06 mask 0x1F \
1744 flags 0x0 priority 3 queue 3
1746 testpmd> 5tuple_filter 0 del dst_ip 2.2.2.5 src_ip 2.2.2.4 \
1747 dst_port 64 src_port 32 protocol 0x06 mask 0x1F \
1748 flags 0x0 priority 3 queue 3
1753 Using the SYN filter, TCP packets whose *SYN* flag is set can be forwarded to a separate queue::
1755 syn_filter (port_id) (add|del) priority (high|low) queue (queue_id)
1757 The available information parameters are:
1759 * ``port_id``: The port which the SYN filter assigned on.
1761 * ``high``: This SYN filter has higher priority than other filters.
1763 * ``low``: This SYN filter has lower priority than other filters.
1765 * ``queue_id``: The receive queue associated with this SYN filter
1769 testpmd> syn_filter 0 add priority high queue 3
1774 With flex filter, packets can be recognized by any arbitrary pattern within the first 128 bytes of the packet
1775 and routed into one of the receive queues::
1777 flex_filter (port_id) (add|del) len (len_value) bytes (bytes_value) \
1778 mask (mask_value) priority (prio_value) queue (queue_id)
1780 The available information parameters are:
1782 * ``port_id``: The port which the Flex filter is assigned on.
1784 * ``len_value``: Filter length in bytes, no greater than 128.
1786 * ``bytes_value``: A string in hexadecimal, means the value the flex filter needs to match.
1788 * ``mask_value``: A string in hexadecimal, bit 1 means corresponding byte participates in the match.
1790 * ``prio_value``: The priority of this filter.
1792 * ``queue_id``: The receive queue associated with this Flex filter.
1796 testpmd> flex_filter 0 add len 16 bytes 0x00000000000000000000000008060000 \
1797 mask 000C priority 3 queue 3
1799 testpmd> flex_filter 0 del len 16 bytes 0x00000000000000000000000008060000 \
1800 mask 000C priority 3 queue 3
1803 .. _testpmd_flow_director:
1805 flow_director_filter
1806 ~~~~~~~~~~~~~~~~~~~~
1808 The Flow Director works in receive mode to identify specific flows or sets of flows and route them to specific queues.
1810 Four types of filtering are supported which are referred to as Perfect Match, Signature, Perfect-mac-vlan and
1811 Perfect-tunnel filters, the match mode is set by the ``--pkt-filter-mode`` command-line parameter:
1813 * Perfect match filters.
1814 The hardware checks a match between the masked fields of the received packets and the programmed filters.
1815 The masked fields are for IP flow.
1817 * Signature filters.
1818 The hardware checks a match between a hash-based signature of the masked fields of the received packet.
1820 * Perfect-mac-vlan match filters.
1821 The hardware checks a match between the masked fields of the received packets and the programmed filters.
1822 The masked fields are for MAC VLAN flow.
1824 * Perfect-tunnel match filters.
1825 The hardware checks a match between the masked fields of the received packets and the programmed filters.
1826 The masked fields are for tunnel flow.
1828 The Flow Director filters can match the different fields for different type of packet: flow type, specific input set
1829 per flow type and the flexible payload.
1831 The Flow Director can also mask out parts of all of these fields so that filters
1832 are only applied to certain fields or parts of the fields.
1834 Different NICs may have different capabilities, command show port fdir (port_id) can be used to acquire the information.
1836 # Commands to add flow director filters of different flow types::
1838 flow_director_filter (port_id) mode IP (add|del|update) \
1839 flow (ipv4-other|ipv4-frag|ipv6-other|ipv6-frag) \
1840 src (src_ip_address) dst (dst_ip_address) \
1841 tos (tos_value) proto (proto_value) ttl (ttl_value) \
1842 vlan (vlan_value) flexbytes (flexbytes_value) \
1843 (drop|fwd) pf|vf(vf_id) queue (queue_id) \
1846 flow_director_filter (port_id) mode IP (add|del|update) \
1847 flow (ipv4-tcp|ipv4-udp|ipv6-tcp|ipv6-udp) \
1848 src (src_ip_address) (src_port) \
1849 dst (dst_ip_address) (dst_port) \
1850 tos (tos_value) ttl (ttl_value) \
1851 vlan (vlan_value) flexbytes (flexbytes_value) \
1852 (drop|fwd) queue pf|vf(vf_id) (queue_id) \
1855 flow_director_filter (port_id) mode IP (add|del|update) \
1856 flow (ipv4-sctp|ipv6-sctp) \
1857 src (src_ip_address) (src_port) \
1858 dst (dst_ip_address) (dst_port) \
1859 tos (tos_value) ttl (ttl_value) \
1860 tag (verification_tag) vlan (vlan_value) \
1861 flexbytes (flexbytes_value) (drop|fwd) \
1862 pf|vf(vf_id) queue (queue_id) fd_id (fd_id_value)
1864 flow_director_filter (port_id) mode IP (add|del|update) flow l2_payload \
1865 ether (ethertype) flexbytes (flexbytes_value) \
1866 (drop|fwd) pf|vf(vf_id) queue (queue_id)
1869 flow_director_filter (port_id) mode MAC-VLAN (add|del|update) \
1870 mac (mac_address) vlan (vlan_value) \
1871 flexbytes (flexbytes_value) (drop|fwd) \
1872 queue (queue_id) fd_id (fd_id_value)
1874 flow_director_filter (port_id) mode Tunnel (add|del|update) \
1875 mac (mac_address) vlan (vlan_value) \
1876 tunnel (NVGRE|VxLAN) tunnel-id (tunnel_id_value) \
1877 flexbytes (flexbytes_value) (drop|fwd) \
1878 queue (queue_id) fd_id (fd_id_value)
1880 For example, to add an ipv4-udp flow type filter::
1882 testpmd> flow_director_filter 0 mode IP add flow ipv4-udp src 2.2.2.3 32 \
1883 dst 2.2.2.5 33 tos 2 ttl 40 vlan 0x1 flexbytes (0x88,0x48) \
1884 fwd pf queue 1 fd_id 1
1886 For example, add an ipv4-other flow type filter::
1888 testpmd> flow_director_filter 0 mode IP add flow ipv4-other src 2.2.2.3 \
1889 dst 2.2.2.5 tos 2 proto 20 ttl 40 vlan 0x1 \
1890 flexbytes (0x88,0x48) fwd pf queue 1 fd_id 1
1895 Flush all flow director filters on a device::
1897 testpmd> flush_flow_director (port_id)
1899 Example, to flush all flow director filter on port 0::
1901 testpmd> flush_flow_director 0
1906 Set flow director's input masks::
1908 flow_director_mask (port_id) mode IP vlan (vlan_value) \
1909 src_mask (ipv4_src) (ipv6_src) (src_port) \
1910 dst_mask (ipv4_dst) (ipv6_dst) (dst_port)
1912 flow_director_mask (port_id) mode MAC-VLAN vlan (vlan_value)
1914 flow_director_mask (port_id) mode Tunnel vlan (vlan_value) \
1915 mac (mac_value) tunnel-type (tunnel_type_value) \
1916 tunnel-id (tunnel_id_value)
1918 Example, to set flow director mask on port 0::
1920 testpmd> flow_director_mask 0 mode IP vlan 0xefff \
1921 src_mask 255.255.255.255 \
1922 FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF \
1923 dst_mask 255.255.255.255 \
1924 FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF
1926 flow_director_flex_mask
1927 ~~~~~~~~~~~~~~~~~~~~~~~
1929 set masks of flow director's flexible payload based on certain flow type::
1931 testpmd> flow_director_flex_mask (port_id) \
1932 flow (none|ipv4-other|ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
1933 ipv6-other|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp| \
1934 l2_payload|all) (mask)
1936 Example, to set flow director's flex mask for all flow type on port 0::
1938 testpmd> flow_director_flex_mask 0 flow all \
1939 (0xff,0xff,0,0,0,0,0,0,0,0,0,0,0,0,0,0)
1942 flow_director_flex_payload
1943 ~~~~~~~~~~~~~~~~~~~~~~~~~~
1945 Configure flexible payload selection::
1947 flow_director_flex_payload (port_id) (raw|l2|l3|l4) (config)
1949 For example, to select the first 16 bytes from the offset 4 (bytes) of packet's payload as flexible payload::
1951 testpmd> flow_director_flex_payload 0 l4 \
1952 (4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19)
1954 get_sym_hash_ena_per_port
1955 ~~~~~~~~~~~~~~~~~~~~~~~~~
1957 Get symmetric hash enable configuration per port::
1959 get_sym_hash_ena_per_port (port_id)
1961 For example, to get symmetric hash enable configuration of port 1::
1963 testpmd> get_sym_hash_ena_per_port 1
1965 set_sym_hash_ena_per_port
1966 ~~~~~~~~~~~~~~~~~~~~~~~~~
1968 Set symmetric hash enable configuration per port to enable or disable::
1970 set_sym_hash_ena_per_port (port_id) (enable|disable)
1972 For example, to set symmetric hash enable configuration of port 1 to enable::
1974 testpmd> set_sym_hash_ena_per_port 1 enable
1976 get_hash_global_config
1977 ~~~~~~~~~~~~~~~~~~~~~~
1979 Get the global configurations of hash filters::
1981 get_hash_global_config (port_id)
1983 For example, to get the global configurations of hash filters of port 1::
1985 testpmd> get_hash_global_config 1
1987 set_hash_global_config
1988 ~~~~~~~~~~~~~~~~~~~~~~
1990 Set the global configurations of hash filters::
1992 set_hash_global_config (port_id) (toeplitz|simple_xor|default) \
1993 (ipv4|ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp|ipv4-other|ipv6|ipv6-frag| \
1994 ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other|l2_payload) \
1997 For example, to enable simple_xor for flow type of ipv6 on port 2::
1999 testpmd> set_hash_global_config 2 simple_xor ipv6 enable
2004 Set the input set for hash::
2006 set_hash_input_set (port_id) (ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
2007 ipv4-other|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other| \
2008 l2_payload) (ovlan|ivlan|src-ipv4|dst-ipv4|src-ipv6|dst-ipv6|ipv4-tos| \
2009 ipv4-proto|ipv6-tc|ipv6-next-header|udp-src-port|udp-dst-port| \
2010 tcp-src-port|tcp-dst-port|sctp-src-port|sctp-dst-port|sctp-veri-tag| \
2011 udp-key|gre-key|fld-1st|fld-2nd|fld-3rd|fld-4th|fld-5th|fld-6th|fld-7th| \
2012 fld-8th|none) (select|add)
2014 For example, to add source IP to hash input set for flow type of ipv4-udp on port 0::
2016 testpmd> set_hash_input_set 0 ipv4-udp src-ipv4 add
2021 The Flow Director filters can match the different fields for different type of packet, i.e. specific input set
2022 on per flow type and the flexible payload. This command can be used to change input set for each flow type.
2024 Set the input set for flow director::
2026 set_fdir_input_set (port_id) (ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
2027 ipv4-other|ipv6|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other| \
2028 l2_payload) (ivlan|ethertype|src-ipv4|dst-ipv4|src-ipv6|dst-ipv6|ipv4-tos| \
2029 ipv4-proto|ipv4-ttl|ipv6-tc|ipv6-next-header|ipv6-hop-limits| \
2030 tudp-src-port|udp-dst-port|cp-src-port|tcp-dst-port|sctp-src-port| \
2031 sctp-dst-port|sctp-veri-tag|none) (select|add)
2033 For example to add source IP to FD input set for flow type of ipv4-udp on port 0::
2035 testpmd> set_fdir_input_set 0 ipv4-udp src-ipv4 add
2040 Set different GRE key length for input set::
2042 global_config (port_id) gre-key-len (number in bytes)
2044 For example to set GRE key length for input set to 4 bytes on port 0::
2046 testpmd> global_config 0 gre-key-len 4
2049 .. _testpmd_rte_flow:
2051 Flow rules management
2052 ---------------------
2054 Control of the generic flow API (*rte_flow*) is fully exposed through the
2055 ``flow`` command (validation, creation, destruction and queries).
2057 Considering *rte_flow* overlaps with all `Filter Functions`_, using both
2058 features simultaneously may cause undefined side-effects and is therefore
2064 Because the ``flow`` command uses dynamic tokens to handle the large number
2065 of possible flow rules combinations, its behavior differs slightly from
2066 other commands, in particular:
2068 - Pressing *?* or the *<tab>* key displays contextual help for the current
2069 token, not that of the entire command.
2071 - Optional and repeated parameters are supported (provided they are listed
2072 in the contextual help).
2074 The first parameter stands for the operation mode. Possible operations and
2075 their general syntax are described below. They are covered in detail in the
2078 - Check whether a flow rule can be created::
2080 flow validate {port_id}
2081 [group {group_id}] [priority {level}] [ingress] [egress]
2082 pattern {item} [/ {item} [...]] / end
2083 actions {action} [/ {action} [...]] / end
2085 - Create a flow rule::
2087 flow create {port_id}
2088 [group {group_id}] [priority {level}] [ingress] [egress]
2089 pattern {item} [/ {item} [...]] / end
2090 actions {action} [/ {action} [...]] / end
2092 - Destroy specific flow rules::
2094 flow destroy {port_id} rule {rule_id} [...]
2096 - Destroy all flow rules::
2098 flow flush {port_id}
2100 - Query an existing flow rule::
2102 flow query {port_id} {rule_id} {action}
2104 - List existing flow rules sorted by priority, filtered by group
2107 flow list {port_id} [group {group_id}] [...]
2109 Validating flow rules
2110 ~~~~~~~~~~~~~~~~~~~~~
2112 ``flow validate`` reports whether a flow rule would be accepted by the
2113 underlying device in its current state but stops short of creating it. It is
2114 bound to ``rte_flow_validate()``::
2116 flow validate {port_id}
2117 [group {group_id}] [priority {level}] [ingress] [egress]
2118 pattern {item} [/ {item} [...]] / end
2119 actions {action} [/ {action} [...]] / end
2121 If successful, it will show::
2125 Otherwise it will show an error message of the form::
2127 Caught error type [...] ([...]): [...]
2129 This command uses the same parameters as ``flow create``, their format is
2130 described in `Creating flow rules`_.
2132 Check whether redirecting any Ethernet packet received on port 0 to RX queue
2133 index 6 is supported::
2135 testpmd> flow validate 0 ingress pattern eth / end
2136 actions queue index 6 / end
2140 Port 0 does not support TCPv6 rules::
2142 testpmd> flow validate 0 ingress pattern eth / ipv6 / tcp / end
2144 Caught error type 9 (specific pattern item): Invalid argument
2150 ``flow create`` validates and creates the specified flow rule. It is bound
2151 to ``rte_flow_create()``::
2153 flow create {port_id}
2154 [group {group_id}] [priority {level}] [ingress] [egress]
2155 pattern {item} [/ {item} [...]] / end
2156 actions {action} [/ {action} [...]] / end
2158 If successful, it will return a flow rule ID usable with other commands::
2160 Flow rule #[...] created
2162 Otherwise it will show an error message of the form::
2164 Caught error type [...] ([...]): [...]
2166 Parameters describe in the following order:
2168 - Attributes (*group*, *priority*, *ingress*, *egress* tokens).
2169 - A matching pattern, starting with the *pattern* token and terminated by an
2171 - Actions, starting with the *actions* token and terminated by an *end*
2174 These translate directly to *rte_flow* objects provided as-is to the
2175 underlying functions.
2177 The shortest valid definition only comprises mandatory tokens::
2179 testpmd> flow create 0 pattern end actions end
2181 Note that PMDs may refuse rules that essentially do nothing such as this
2184 **All unspecified object values are automatically initialized to 0.**
2189 These tokens affect flow rule attributes (``struct rte_flow_attr``) and are
2190 specified before the ``pattern`` token.
2192 - ``group {group id}``: priority group.
2193 - ``priority {level}``: priority level within group.
2194 - ``ingress``: rule applies to ingress traffic.
2195 - ``egress``: rule applies to egress traffic.
2197 Each instance of an attribute specified several times overrides the previous
2198 value as shown below (group 4 is used)::
2200 testpmd> flow create 0 group 42 group 24 group 4 [...]
2202 Note that once enabled, ``ingress`` and ``egress`` cannot be disabled.
2204 While not specifying a direction is an error, some rules may allow both
2207 Most rules affect RX therefore contain the ``ingress`` token::
2209 testpmd> flow create 0 ingress pattern [...]
2214 A matching pattern starts after the ``pattern`` token. It is made of pattern
2215 items and is terminated by a mandatory ``end`` item.
2217 Items are named after their type (*RTE_FLOW_ITEM_TYPE_* from ``enum
2218 rte_flow_item_type``).
2220 The ``/`` token is used as a separator between pattern items as shown
2223 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end [...]
2225 Note that protocol items like these must be stacked from lowest to highest
2226 layer to make sense. For instance, the following rule is either invalid or
2227 unlikely to match any packet::
2229 testpmd> flow create 0 ingress pattern eth / udp / ipv4 / end [...]
2231 More information on these restrictions can be found in the *rte_flow*
2234 Several items support additional specification structures, for example
2235 ``ipv4`` allows specifying source and destination addresses as follows::
2237 testpmd> flow create 0 ingress pattern eth / ipv4 src is 10.1.1.1
2238 dst is 10.2.0.0 / end [...]
2240 This rule matches all IPv4 traffic with the specified properties.
2242 In this example, ``src`` and ``dst`` are field names of the underlying
2243 ``struct rte_flow_item_ipv4`` object. All item properties can be specified
2244 in a similar fashion.
2246 The ``is`` token means that the subsequent value must be matched exactly,
2247 and assigns ``spec`` and ``mask`` fields in ``struct rte_flow_item``
2248 accordingly. Possible assignment tokens are:
2250 - ``is``: match value perfectly (with full bit-mask).
2251 - ``spec``: match value according to configured bit-mask.
2252 - ``last``: specify upper bound to establish a range.
2253 - ``mask``: specify bit-mask with relevant bits set to one.
2254 - ``prefix``: generate bit-mask from a prefix length.
2256 These yield identical results::
2258 ipv4 src is 10.1.1.1
2262 ipv4 src spec 10.1.1.1 src mask 255.255.255.255
2266 ipv4 src spec 10.1.1.1 src prefix 32
2270 ipv4 src is 10.1.1.1 src last 10.1.1.1 # range with a single value
2274 ipv4 src is 10.1.1.1 src last 0 # 0 disables range
2276 Inclusive ranges can be defined with ``last``::
2278 ipv4 src is 10.1.1.1 src last 10.2.3.4 # 10.1.1.1 to 10.2.3.4
2280 Note that ``mask`` affects both ``spec`` and ``last``::
2282 ipv4 src is 10.1.1.1 src last 10.2.3.4 src mask 255.255.0.0
2283 # matches 10.1.0.0 to 10.2.255.255
2285 Properties can be modified multiple times::
2287 ipv4 src is 10.1.1.1 src is 10.1.2.3 src is 10.2.3.4 # matches 10.2.3.4
2291 ipv4 src is 10.1.1.1 src prefix 24 src prefix 16 # matches 10.1.0.0/16
2296 This section lists supported pattern items and their attributes, if any.
2298 - ``end``: end list of pattern items.
2300 - ``void``: no-op pattern item.
2302 - ``invert``: perform actions when pattern does not match.
2304 - ``any``: match any protocol for the current layer.
2306 - ``num {unsigned}``: number of layers covered.
2308 - ``pf``: match packets addressed to the physical function.
2310 - ``vf``: match packets addressed to a virtual function ID.
2312 - ``id {unsigned}``: destination VF ID.
2314 - ``port``: device-specific physical port index to use.
2316 - ``index {unsigned}``: physical port index.
2318 - ``raw``: match an arbitrary byte string.
2320 - ``relative {boolean}``: look for pattern after the previous item.
2321 - ``search {boolean}``: search pattern from offset (see also limit).
2322 - ``offset {integer}``: absolute or relative offset for pattern.
2323 - ``limit {unsigned}``: search area limit for start of pattern.
2324 - ``pattern {string}``: byte string to look for.
2326 - ``eth``: match Ethernet header.
2328 - ``dst {MAC-48}``: destination MAC.
2329 - ``src {MAC-48}``: source MAC.
2330 - ``type {unsigned}``: EtherType.
2332 - ``vlan``: match 802.1Q/ad VLAN tag.
2334 - ``tpid {unsigned}``: tag protocol identifier.
2335 - ``tci {unsigned}``: tag control information.
2336 - ``pcp {unsigned}``: priority code point.
2337 - ``dei {unsigned}``: drop eligible indicator.
2338 - ``vid {unsigned}``: VLAN identifier.
2340 - ``ipv4``: match IPv4 header.
2342 - ``tos {unsigned}``: type of service.
2343 - ``ttl {unsigned}``: time to live.
2344 - ``proto {unsigned}``: next protocol ID.
2345 - ``src {ipv4 address}``: source address.
2346 - ``dst {ipv4 address}``: destination address.
2348 - ``ipv6``: match IPv6 header.
2350 - ``tc {unsigned}``: traffic class.
2351 - ``flow {unsigned}``: flow label.
2352 - ``proto {unsigned}``: protocol (next header).
2353 - ``hop {unsigned}``: hop limit.
2354 - ``src {ipv6 address}``: source address.
2355 - ``dst {ipv6 address}``: destination address.
2357 - ``icmp``: match ICMP header.
2359 - ``type {unsigned}``: ICMP packet type.
2360 - ``code {unsigned}``: ICMP packet code.
2362 - ``udp``: match UDP header.
2364 - ``src {unsigned}``: UDP source port.
2365 - ``dst {unsigned}``: UDP destination port.
2367 - ``tcp``: match TCP header.
2369 - ``src {unsigned}``: TCP source port.
2370 - ``dst {unsigned}``: TCP destination port.
2372 - ``sctp``: match SCTP header.
2374 - ``src {unsigned}``: SCTP source port.
2375 - ``dst {unsigned}``: SCTP destination port.
2376 - ``tag {unsigned}``: validation tag.
2377 - ``cksum {unsigned}``: checksum.
2379 - ``vxlan``: match VXLAN header.
2381 - ``vni {unsigned}``: VXLAN identifier.
2386 A list of actions starts after the ``actions`` token in the same fashion as
2387 `Matching pattern`_; actions are separated by ``/`` tokens and the list is
2388 terminated by a mandatory ``end`` action.
2390 Actions are named after their type (*RTE_FLOW_ACTION_TYPE_* from ``enum
2391 rte_flow_action_type``).
2393 Dropping all incoming UDPv4 packets can be expressed as follows::
2395 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
2398 Several actions have configurable properties which must be specified when
2399 there is no valid default value. For example, ``queue`` requires a target
2402 This rule redirects incoming UDPv4 traffic to queue index 6::
2404 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
2405 actions queue index 6 / end
2407 While this one could be rejected by PMDs (unspecified queue index)::
2409 testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
2412 As defined by *rte_flow*, the list is not ordered, all actions of a given
2413 rule are performed simultaneously. These are equivalent::
2415 queue index 6 / void / mark id 42 / end
2419 void / mark id 42 / queue index 6 / end
2421 All actions in a list should have different types, otherwise only the last
2422 action of a given type is taken into account::
2424 queue index 4 / queue index 5 / queue index 6 / end # will use queue 6
2428 drop / drop / drop / end # drop is performed only once
2432 mark id 42 / queue index 3 / mark id 24 / end # mark will be 24
2434 Considering they are performed simultaneously, opposite and overlapping
2435 actions can sometimes be combined when the end result is unambiguous::
2437 drop / queue index 6 / end # drop has no effect
2441 drop / dup index 6 / end # same as above
2445 queue index 6 / rss queues 6 7 8 / end # queue has no effect
2449 drop / passthru / end # drop has no effect
2451 Note that PMDs may still refuse such combinations.
2456 This section lists supported actions and their attributes, if any.
2458 - ``end``: end list of actions.
2460 - ``void``: no-op action.
2462 - ``passthru``: let subsequent rule process matched packets.
2464 - ``mark``: attach 32 bit value to packets.
2466 - ``id {unsigned}``: 32 bit value to return with packets.
2468 - ``flag``: flag packets.
2470 - ``queue``: assign packets to a given queue index.
2472 - ``index {unsigned}``: queue index to use.
2474 - ``drop``: drop packets (note: passthru has priority).
2476 - ``count``: enable counters for this rule.
2478 - ``dup``: duplicate packets to a given queue index.
2480 - ``index {unsigned}``: queue index to duplicate packets to.
2482 - ``rss``: spread packets among several queues.
2484 - ``queues [{unsigned} [...]] end``: queue indices to use.
2486 - ``pf``: redirect packets to physical device function.
2488 - ``vf``: redirect packets to virtual device function.
2490 - ``original {boolean}``: use original VF ID if possible.
2491 - ``id {unsigned}``: VF ID to redirect packets to.
2493 Destroying flow rules
2494 ~~~~~~~~~~~~~~~~~~~~~
2496 ``flow destroy`` destroys one or more rules from their rule ID (as returned
2497 by ``flow create``), this command calls ``rte_flow_destroy()`` as many
2498 times as necessary::
2500 flow destroy {port_id} rule {rule_id} [...]
2502 If successful, it will show::
2504 Flow rule #[...] destroyed
2506 It does not report anything for rule IDs that do not exist. The usual error
2507 message is shown when a rule cannot be destroyed::
2509 Caught error type [...] ([...]): [...]
2511 ``flow flush`` destroys all rules on a device and does not take extra
2512 arguments. It is bound to ``rte_flow_flush()``::
2514 flow flush {port_id}
2516 Any errors are reported as above.
2518 Creating several rules and destroying them::
2520 testpmd> flow create 0 ingress pattern eth / ipv6 / end
2521 actions queue index 2 / end
2522 Flow rule #0 created
2523 testpmd> flow create 0 ingress pattern eth / ipv4 / end
2524 actions queue index 3 / end
2525 Flow rule #1 created
2526 testpmd> flow destroy 0 rule 0 rule 1
2527 Flow rule #1 destroyed
2528 Flow rule #0 destroyed
2531 The same result can be achieved using ``flow flush``::
2533 testpmd> flow create 0 ingress pattern eth / ipv6 / end
2534 actions queue index 2 / end
2535 Flow rule #0 created
2536 testpmd> flow create 0 ingress pattern eth / ipv4 / end
2537 actions queue index 3 / end
2538 Flow rule #1 created
2539 testpmd> flow flush 0
2542 Non-existent rule IDs are ignored::
2544 testpmd> flow create 0 ingress pattern eth / ipv6 / end
2545 actions queue index 2 / end
2546 Flow rule #0 created
2547 testpmd> flow create 0 ingress pattern eth / ipv4 / end
2548 actions queue index 3 / end
2549 Flow rule #1 created
2550 testpmd> flow destroy 0 rule 42 rule 10 rule 2
2552 testpmd> flow destroy 0 rule 0
2553 Flow rule #0 destroyed
2559 ``flow query`` queries a specific action of a flow rule having that
2560 ability. Such actions collect information that can be reported using this
2561 command. It is bound to ``rte_flow_query()``::
2563 flow query {port_id} {rule_id} {action}
2565 If successful, it will display either the retrieved data for known actions
2566 or the following message::
2568 Cannot display result for action type [...] ([...])
2570 Otherwise, it will complain either that the rule does not exist or that some
2573 Flow rule #[...] not found
2577 Caught error type [...] ([...]): [...]
2579 Currently only the ``count`` action is supported. This action reports the
2580 number of packets that hit the flow rule and the total number of bytes. Its
2581 output has the following format::
2584 hits_set: [...] # whether "hits" contains a valid value
2585 bytes_set: [...] # whether "bytes" contains a valid value
2586 hits: [...] # number of packets
2587 bytes: [...] # number of bytes
2589 Querying counters for TCPv6 packets redirected to queue 6::
2591 testpmd> flow create 0 ingress pattern eth / ipv6 / tcp / end
2592 actions queue index 6 / count / end
2593 Flow rule #4 created
2594 testpmd> flow query 0 4 count
2605 ``flow list`` lists existing flow rules sorted by priority and optionally
2606 filtered by group identifiers::
2608 flow list {port_id} [group {group_id}] [...]
2610 This command only fails with the following message if the device does not
2615 Output consists of a header line followed by a short description of each
2616 flow rule, one per line. There is no output at all when no flow rules are
2617 configured on the device::
2619 ID Group Prio Attr Rule
2620 [...] [...] [...] [...] [...]
2622 ``Attr`` column flags:
2624 - ``i`` for ``ingress``.
2625 - ``e`` for ``egress``.
2627 Creating several flow rules and listing them::
2629 testpmd> flow create 0 ingress pattern eth / ipv4 / end
2630 actions queue index 6 / end
2631 Flow rule #0 created
2632 testpmd> flow create 0 ingress pattern eth / ipv6 / end
2633 actions queue index 2 / end
2634 Flow rule #1 created
2635 testpmd> flow create 0 priority 5 ingress pattern eth / ipv4 / udp / end
2636 actions rss queues 6 7 8 end / end
2637 Flow rule #2 created
2638 testpmd> flow list 0
2639 ID Group Prio Attr Rule
2640 0 0 0 i- ETH IPV4 => QUEUE
2641 1 0 0 i- ETH IPV6 => QUEUE
2642 2 0 5 i- ETH IPV4 UDP => RSS
2645 Rules are sorted by priority (i.e. group ID first, then priority level)::
2647 testpmd> flow list 1
2648 ID Group Prio Attr Rule
2649 0 0 0 i- ETH => COUNT
2650 6 0 500 i- ETH IPV6 TCP => DROP COUNT
2651 5 0 1000 i- ETH IPV6 ICMP => QUEUE
2652 1 24 0 i- ETH IPV4 UDP => QUEUE
2653 4 24 10 i- ETH IPV4 TCP => DROP
2654 3 24 20 i- ETH IPV4 => DROP
2655 2 24 42 i- ETH IPV4 UDP => QUEUE
2656 7 63 0 i- ETH IPV6 UDP VXLAN => MARK QUEUE
2659 Output can be limited to specific groups::
2661 testpmd> flow list 1 group 0 group 63
2662 ID Group Prio Attr Rule
2663 0 0 0 i- ETH => COUNT
2664 6 0 500 i- ETH IPV6 TCP => DROP COUNT
2665 5 0 1000 i- ETH IPV6 ICMP => QUEUE
2666 7 63 0 i- ETH IPV6 UDP VXLAN => MARK QUEUE