xref: /dpdk/doc/guides/testpmd_app_ug/testpmd_funcs.rst (revision c6dab2a873f65c5a4ea9735aa24d9539426adba4)
1..  BSD LICENSE
2    Copyright(c) 2010-2016 Intel Corporation. All rights reserved.
3    All rights reserved.
4
5    Redistribution and use in source and binary forms, with or without
6    modification, are permitted provided that the following conditions
7    are met:
8
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
14    distribution.
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.
18
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.
30
31.. _testpmd_runtime:
32
33Testpmd Runtime Functions
34=========================
35
36Where the testpmd application is started in interactive mode, (``-i|--interactive``),
37it displays a prompt that can be used to start and stop forwarding,
38configure the application, display statistics (including the extended NIC
39statistics aka xstats) , set the Flow Director and other tasks::
40
41   testpmd>
42
43The testpmd prompt has some, limited, readline support.
44Common bash command-line functions such as ``Ctrl+a`` and ``Ctrl+e`` to go to the start and end of the prompt line are supported
45as well as access to the command history via the up-arrow.
46
47There is also support for tab completion.
48If you type a partial command and hit ``<TAB>`` you get a list of the available completions:
49
50.. code-block:: console
51
52   testpmd> show port <TAB>
53
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
58       ...
59
60
61.. note::
62
63   Some examples in this document are too long to fit on one line are are shown wrapped at `"\\"` for display purposes::
64
65      testpmd> set flow_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \
66               (pause_time) (send_xon) (port_id)
67
68In the real ``testpmd>`` prompt these commands should be on a single line.
69
70Help Functions
71--------------
72
73The testpmd has on-line help for the functions that are available at runtime.
74These are divided into sections and can be accessed using help, help section or help all:
75
76.. code-block:: console
77
78   testpmd> help
79
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.
87
88
89Control Functions
90-----------------
91
92start
93~~~~~
94
95Start packet forwarding with current configuration::
96
97   testpmd> start
98
99start tx_first
100~~~~~~~~~~~~~~
101
102Start packet forwarding with current configuration after sending specified number of bursts of packets::
103
104   testpmd> start tx_first (""|burst_num)
105
106The default burst number is 1 when ``burst_num`` not presented.
107
108stop
109~~~~
110
111Stop packet forwarding, and display accumulated statistics::
112
113   testpmd> stop
114
115quit
116~~~~
117
118Quit to prompt::
119
120   testpmd> quit
121
122
123Display Functions
124-----------------
125
126The functions in the following sections are used to display information about the
127testpmd configuration or the NIC status.
128
129show port
130~~~~~~~~~
131
132Display information for a given port or all ports::
133
134   testpmd> show port (info|stats|xstats|fdir|stat_qmap|dcb_tc) (port_id|all)
135
136The available information categories are:
137
138* ``info``: General port information such as MAC address.
139
140* ``stats``: RX/TX statistics.
141
142* ``xstats``: RX/TX extended NIC statistics.
143
144* ``fdir``: Flow Director information and statistics.
145
146* ``stat_qmap``: Queue statistics mapping.
147
148* ``dcb_tc``: DCB information such as TC mapping.
149
150For example:
151
152.. code-block:: console
153
154   testpmd> show port info 0
155
156   ********************* Infos for port 0 *********************
157
158   MAC address: XX:XX:XX:XX:XX:XX
159   Connect to socket: 0
160   memory allocation on the socket: 0
161   Link status: up
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
168   VLAN offload:
169       strip on
170       filter on
171       qinq(extend) off
172   Redirection table size: 512
173   Supported flow types:
174     ipv4-frag
175     ipv4-tcp
176     ipv4-udp
177     ipv4-sctp
178     ipv4-other
179     ipv6-frag
180     ipv6-tcp
181     ipv6-udp
182     ipv6-sctp
183     ipv6-other
184     l2_payload
185     port
186     vxlan
187     geneve
188     nvgre
189
190show port rss reta
191~~~~~~~~~~~~~~~~~~
192
193Display the rss redirection table entry indicated by masks on port X::
194
195   testpmd> show port (port_id) rss reta (size) (mask0, mask1...)
196
197size is used to indicate the hardware supported reta size
198
199show port rss-hash
200~~~~~~~~~~~~~~~~~~
201
202Display the RSS hash functions and RSS hash key of a port::
203
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]
205
206clear port
207~~~~~~~~~~
208
209Clear the port statistics for a given port or for all ports::
210
211   testpmd> clear port (info|stats|xstats|fdir|stat_qmap) (port_id|all)
212
213For example::
214
215   testpmd> clear port stats all
216
217show (rxq|txq)
218~~~~~~~~~~~~~~
219
220Display information for a given port's RX/TX queue::
221
222   testpmd> show (rxq|txq) info (port_id) (queue_id)
223
224show config
225~~~~~~~~~~~
226
227Displays the configuration of the application.
228The configuration comes from the command-line, the runtime or the application defaults::
229
230   testpmd> show config (rxtx|cores|fwd|txpkts)
231
232The available information categories are:
233
234* ``rxtx``: RX/TX configuration items.
235
236* ``cores``: List of forwarding cores.
237
238* ``fwd``: Packet forwarding configuration.
239
240* ``txpkts``: Packets to TX configuration.
241
242For example:
243
244.. code-block:: console
245
246   testpmd> show config rxtx
247
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
255
256set fwd
257~~~~~~~
258
259Set the packet forwarding mode::
260
261   testpmd> set fwd (io|mac|macswap|flowgen| \
262                     rxonly|txonly|csum|icmpecho) (""|retry)
263
264``retry`` can be specified for forwarding engines except ``rx_only``.
265
266The available information categories are:
267
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.
271
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.
276
277* ``macswap``: MAC swap forwarding mode.
278  Swaps the source and the destination Ethernet addresses of packets before forwarding them.
279
280* ``flowgen``: Multi-flow generation mode.
281  Originates a number of flows (with varying destination IP addresses), and terminate receive traffic.
282
283* ``rxonly``: Receives packets but doesn't transmit them.
284
285* ``txonly``: Generates and transmits packets without receiving any.
286
287* ``csum``: Changes the checksum field with hardware or software methods depending on the offload flags on the packet.
288
289* ``icmpecho``: Receives a burst of packets, lookup for IMCP echo requests and, if any, send back ICMP echo replies.
290
291* ``ieee1588``: Demonstrate L2 IEEE1588 V2 PTP timestamping for RX and TX. Requires ``CONFIG_RTE_LIBRTE_IEEE1588=y``.
292
293Note: TX timestamping is only available in the "Full Featured" TX path. To force ``testpmd`` into this mode set ``--txqflags=0``.
294
295Example::
296
297   testpmd> set fwd rxonly
298
299   Set rxonly packet forwarding mode
300
301
302read rxd
303~~~~~~~~
304
305Display an RX descriptor for a port RX queue::
306
307   testpmd> read rxd (port_id) (queue_id) (rxd_id)
308
309For example::
310
311   testpmd> read rxd 0 0 4
312        0x0000000B - 0x001D0180 / 0x0000000B - 0x001D0180
313
314read txd
315~~~~~~~~
316
317Display a TX descriptor for a port TX queue::
318
319   testpmd> read txd (port_id) (queue_id) (txd_id)
320
321For example::
322
323   testpmd> read txd 0 0 4
324        0x00000001 - 0x24C3C440 / 0x000F0000 - 0x2330003C
325
326
327Configuration Functions
328-----------------------
329
330The testpmd application can be configured from the runtime as well as from the command-line.
331
332This section details the available configuration functions that are available.
333
334.. note::
335
336   Configuration changes only become active when forwarding is started/restarted.
337
338set default
339~~~~~~~~~~~
340
341Reset forwarding to the default configuration::
342
343   testpmd> set default
344
345set verbose
346~~~~~~~~~~~
347
348Set the debug verbosity level::
349
350   testpmd> set verbose (level)
351
352Currently the only available levels are 0 (silent except for error) and 1 (fully verbose).
353
354set nbport
355~~~~~~~~~~
356
357Set the number of ports used by the application:
358
359set nbport (num)
360
361This is equivalent to the ``--nb-ports`` command-line option.
362
363set nbcore
364~~~~~~~~~~
365
366Set the number of cores used by the application::
367
368   testpmd> set nbcore (num)
369
370This is equivalent to the ``--nb-cores`` command-line option.
371
372.. note::
373
374   The number of cores used must not be greater than number of ports used multiplied by the number of queues per port.
375
376set coremask
377~~~~~~~~~~~~
378
379Set the forwarding cores hexadecimal mask::
380
381   testpmd> set coremask (mask)
382
383This is equivalent to the ``--coremask`` command-line option.
384
385.. note::
386
387   The master lcore is reserved for command line parsing only and cannot be masked on for packet forwarding.
388
389set portmask
390~~~~~~~~~~~~
391
392Set the forwarding ports hexadecimal mask::
393
394   testpmd> set portmask (mask)
395
396This is equivalent to the ``--portmask`` command-line option.
397
398set burst
399~~~~~~~~~
400
401Set number of packets per burst::
402
403   testpmd> set burst (num)
404
405This is equivalent to the ``--burst command-line`` option.
406
407When retry is enabled, the transmit delay time and number of retries can also be set::
408
409   testpmd> set burst tx delay (microseconds) retry (num)
410
411set txpkts
412~~~~~~~~~~
413
414Set the length of each segment of the TX-ONLY packets or length of packet for FLOWGEN mode::
415
416   testpmd> set txpkts (x[,y]*)
417
418Where x[,y]* represents a CSV list of values, without white space.
419
420set txsplit
421~~~~~~~~~~~
422
423Set the split policy for the TX packets, applicable for TX-ONLY and CSUM forwarding modes::
424
425   testpmd> set txsplit (off|on|rand)
426
427Where:
428
429* ``off`` disable packet copy & split for CSUM mode.
430
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
433  (see above).
434
435* ``rand`` same as 'on', but number of segments per each packet is a random value between 1 and total number of segments.
436
437set corelist
438~~~~~~~~~~~~
439
440Set the list of forwarding cores::
441
442   testpmd> set corelist (x[,y]*)
443
444For example, to change the forwarding cores:
445
446.. code-block:: console
447
448   testpmd> set corelist 3,1
449   testpmd> show config fwd
450
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
456
457.. note::
458
459   The cores are used in the same order as specified on the command line.
460
461set portlist
462~~~~~~~~~~~~
463
464Set the list of forwarding ports::
465
466   testpmd> set portlist (x[,y]*)
467
468For example, to change the port forwarding:
469
470.. code-block:: console
471
472   testpmd> set portlist 0,2,1,3
473   testpmd> show config fwd
474
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
481
482set tx loopback
483~~~~~~~~~~~~~~~
484
485Enable/disable tx loopback::
486
487   testpmd> set tx loopback (port_id) (on|off)
488
489set drop enable
490~~~~~~~~~~~~~~~
491
492set drop enable bit for all queues::
493
494   testpmd> set all queues drop (port_id) (on|off)
495
496set split drop enable (for VF)
497~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
498
499set split drop enable bit for VF from PF::
500
501   testpmd> set vf split drop (port_id) (vf_id) (on|off)
502
503set mac antispoof (for VF)
504~~~~~~~~~~~~~~~~~~~~~~~~~~
505
506Set mac antispoof for a VF from the PF::
507
508   testpmd> set vf mac antispoof  (port_id) (vf_id) (on|off)
509
510vlan set strip
511~~~~~~~~~~~~~~
512
513Set the VLAN strip on a port::
514
515   testpmd> vlan set strip (on|off) (port_id)
516
517vlan set stripq
518~~~~~~~~~~~~~~~
519
520Set the VLAN strip for a queue on a port::
521
522   testpmd> vlan set stripq (on|off) (port_id,queue_id)
523
524vlan set stripq (for VF)
525~~~~~~~~~~~~~~~~~~~~~~~~
526
527Set VLAN strip for all queues in a pool for a VF from the PF::
528
529   testpmd> set vf vlan stripq (port_id) (vf_id) (on|off)
530
531vlan set insert (for VF)
532~~~~~~~~~~~~~~~~~~~~~~~~
533
534Set VLAN insert for a VF from the PF::
535
536   testpmd> set vf vlan insert (port_id) (vf_id) (vlan_id)
537
538vlan set antispoof (for VF)
539~~~~~~~~~~~~~~~~~~~~~~~~~~~
540
541Set VLAN antispoof for a VF from the PF::
542
543   testpmd> set vf vlan antispoof (port_id) (vf_id) (on|off)
544
545vlan set filter
546~~~~~~~~~~~~~~~
547
548Set the VLAN filter on a port::
549
550   testpmd> vlan set filter (on|off) (port_id)
551
552vlan set qinq
553~~~~~~~~~~~~~
554
555Set the VLAN QinQ (extended queue in queue) on for a port::
556
557   testpmd> vlan set qinq (on|off) (port_id)
558
559vlan set tpid
560~~~~~~~~~~~~~
561
562Set the inner or outer VLAN TPID for packet filtering on a port::
563
564   testpmd> vlan set (inner|outer) tpid (value) (port_id)
565
566.. note::
567
568   TPID value must be a 16-bit number (value <= 65536).
569
570rx_vlan add
571~~~~~~~~~~~
572
573Add a VLAN ID, or all identifiers, to the set of VLAN identifiers filtered by port ID::
574
575   testpmd> rx_vlan add (vlan_id|all) (port_id)
576
577.. note::
578
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.
582
583rx_vlan rm
584~~~~~~~~~~
585
586Remove a VLAN ID, or all identifiers, from the set of VLAN identifiers filtered by port ID::
587
588   testpmd> rx_vlan rm (vlan_id|all) (port_id)
589
590rx_vlan add (for VF)
591~~~~~~~~~~~~~~~~~~~~
592
593Add a VLAN ID, to the set of VLAN identifiers filtered for VF(s) for port ID::
594
595   testpmd> rx_vlan add (vlan_id) port (port_id) vf (vf_mask)
596
597rx_vlan rm (for VF)
598~~~~~~~~~~~~~~~~~~~
599
600Remove a VLAN ID, from the set of VLAN identifiers filtered for VF(s) for port ID::
601
602   testpmd> rx_vlan rm (vlan_id) port (port_id) vf (vf_mask)
603
604tunnel_filter add
605~~~~~~~~~~~~~~~~~
606
607Add a tunnel filter on a port::
608
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)
612
613The available information categories are:
614
615* ``vxlan``: Set tunnel type as VXLAN.
616
617* ``nvgre``: Set tunnel type as NVGRE.
618
619* ``ipingre``: Set tunnel type as IP-in-GRE.
620
621* ``imac-ivlan``: Set filter type as Inner MAC and VLAN.
622
623* ``imac-ivlan-tenid``: Set filter type as Inner MAC, VLAN and tenant ID.
624
625* ``imac-tenid``: Set filter type as Inner MAC and tenant ID.
626
627* ``imac``: Set filter type as Inner MAC.
628
629* ``omac-imac-tenid``: Set filter type as Outer MAC, Inner MAC and tenant ID.
630
631* ``oip``: Set filter type as Outer IP.
632
633* ``iip``: Set filter type as Inner IP.
634
635Example::
636
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
639
640   Set an IP-in-GRE tunnel on port 0, and the filter type is Outer IP.
641
642tunnel_filter remove
643~~~~~~~~~~~~~~~~~~~~
644
645Remove a tunnel filter on a port::
646
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)
650
651rx_vxlan_port add
652~~~~~~~~~~~~~~~~~
653
654Add an UDP port for VXLAN packet filter on a port::
655
656   testpmd> rx_vxlan_port add (udp_port) (port_id)
657
658rx_vxlan_port remove
659~~~~~~~~~~~~~~~~~~~~
660
661Remove an UDP port for VXLAN packet filter on a port::
662
663   testpmd> rx_vxlan_port rm (udp_port) (port_id)
664
665tx_vlan set
666~~~~~~~~~~~
667
668Set hardware insertion of VLAN IDs in packets sent on a port::
669
670   testpmd> tx_vlan set (port_id) vlan_id[, vlan_id_outer]
671
672For example, set a single VLAN ID (5) insertion on port 0::
673
674   tx_vlan set 0 5
675
676Or, set double VLAN ID (inner: 2, outer: 3) insertion on port 1::
677
678   tx_vlan set 1 2 3
679
680
681tx_vlan set pvid
682~~~~~~~~~~~~~~~~
683
684Set port based hardware insertion of VLAN ID in packets sent on a port::
685
686   testpmd> tx_vlan set pvid (port_id) (vlan_id) (on|off)
687
688tx_vlan reset
689~~~~~~~~~~~~~
690
691Disable hardware insertion of a VLAN header in packets sent on a port::
692
693   testpmd> tx_vlan reset (port_id)
694
695csum set
696~~~~~~~~
697
698Select hardware or software calculation of the checksum when
699transmitting a packet using the ``csum`` forwarding engine::
700
701   testpmd> csum set (ip|udp|tcp|sctp|outer-ip) (hw|sw) (port_id)
702
703Where:
704
705* ``ip|udp|tcp|sctp`` always relate to  the inner layer.
706
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.
710
711.. note::
712
713   Check the NIC Datasheet for hardware limits.
714
715csum parse-tunnel
716~~~~~~~~~~~~~~~~~
717
718Define how tunneled packets should be handled by the csum forward
719engine::
720
721   testpmd> csum parse-tunnel (on|off) (tx_port_id)
722
723If enabled, the csum forward engine will try to recognize supported
724tunnel headers (vxlan, gre, ipip).
725
726If disabled, treat tunnel packets as non-tunneled packets (a inner
727header is handled as a packet payload).
728
729.. note::
730
731   The port argument is the TX port like in the ``csum set`` command.
732
733Example:
734
735Consider a packet in packet like the following::
736
737   eth_out/ipv4_out/udp_out/vxlan/eth_in/ipv4_in/tcp_in
738
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``).
742
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``.
745
746csum show
747~~~~~~~~~
748
749Display tx checksum offload configuration::
750
751   testpmd> csum show (port_id)
752
753tso set
754~~~~~~~
755
756Enable TCP Segmentation Offload (TSO) in the ``csum`` forwarding engine::
757
758   testpmd> tso set (segsize) (port_id)
759
760.. note::
761
762   Check the NIC datasheet for hardware limits.
763
764tso show
765~~~~~~~~
766
767Display the status of TCP Segmentation Offload::
768
769   testpmd> tso show (port_id)
770
771mac_addr add
772~~~~~~~~~~~~
773
774Add an alternative MAC address to a port::
775
776   testpmd> mac_addr add (port_id) (XX:XX:XX:XX:XX:XX)
777
778mac_addr remove
779~~~~~~~~~~~~~~~
780
781Remove a MAC address from a port::
782
783   testpmd> mac_addr remove (port_id) (XX:XX:XX:XX:XX:XX)
784
785mac_addr add (for VF)
786~~~~~~~~~~~~~~~~~~~~~
787
788Add an alternative MAC address for a VF to a port::
789
790   testpmd> mac_add add port (port_id) vf (vf_id) (XX:XX:XX:XX:XX:XX)
791
792mac_addr set (for VF)
793~~~~~~~~~~~~~~~~~~~~~
794
795Set the MAC address for a VF from the PF::
796
797   testpmd> set vf mac addr (port_id) (vf_id) (XX:XX:XX:XX:XX:XX)
798
799set port-uta
800~~~~~~~~~~~~
801
802Set the unicast hash filter(s) on/off for a port::
803
804   testpmd> set port (port_id) uta (XX:XX:XX:XX:XX:XX|all) (on|off)
805
806set promisc
807~~~~~~~~~~~
808
809Set the promiscuous mode on for a port or for all ports.
810In promiscuous mode packets are not dropped if they aren't for the specified MAC address::
811
812   testpmd> set promisc (port_id|all) (on|off)
813
814set allmulti
815~~~~~~~~~~~~
816
817Set the allmulti mode for a port or for all ports::
818
819   testpmd> set allmulti (port_id|all) (on|off)
820
821Same as the ifconfig (8) option. Controls how multicast packets are handled.
822
823set flow_ctrl rx
824~~~~~~~~~~~~~~~~
825
826Set the link flow control parameter on a port::
827
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)
831
832Where:
833
834* ``high_water`` (integer): High threshold value to trigger XOFF.
835
836* ``low_water`` (integer): Low threshold value to trigger XON.
837
838* ``pause_time`` (integer): Pause quota in the Pause frame.
839
840* ``send_xon`` (0/1): Send XON frame.
841
842* ``mac_ctrl_frame_fwd``: Enable receiving MAC control frames.
843
844* ``autoneg``: Change the auto-negotiation para mete.
845
846set pfc_ctrl rx
847~~~~~~~~~~~~~~~
848
849Set the priority flow control parameter on a port::
850
851   testpmd> set pfc_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \
852            (pause_time) (priority) (port_id)
853
854Where:
855
856* ``high_water`` (integer): High threshold value.
857
858* ``low_water`` (integer): Low threshold value.
859
860* ``pause_time`` (integer): Pause quota in the Pause frame.
861
862* ``priority`` (0-7): VLAN User Priority.
863
864set stat_qmap
865~~~~~~~~~~~~~
866
867Set statistics mapping (qmapping 0..15) for RX/TX queue on port::
868
869   testpmd> set stat_qmap (tx|rx) (port_id) (queue_id) (qmapping)
870
871For example, to set rx queue 2 on port 0 to mapping 5::
872
873   testpmd>set stat_qmap rx 0 2 5
874
875set port - rx/tx (for VF)
876~~~~~~~~~~~~~~~~~~~~~~~~~
877
878Set VF receive/transmit from a port::
879
880   testpmd> set port (port_id) vf (vf_id) (rx|tx) (on|off)
881
882set port - mac address filter (for VF)
883~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
884
885Add/Remove unicast or multicast MAC addr filter for a VF::
886
887   testpmd> set port (port_id) vf (vf_id) (mac_addr) \
888            (exact-mac|exact-mac-vlan|hashmac|hashmac-vlan) (on|off)
889
890set port - rx mode(for VF)
891~~~~~~~~~~~~~~~~~~~~~~~~~~
892
893Set the VF receive mode of a port::
894
895   testpmd> set port (port_id) vf (vf_id) \
896            rxmode (AUPE|ROPE|BAM|MPE) (on|off)
897
898The available receive modes are:
899
900* ``AUPE``: Accepts untagged VLAN.
901
902* ``ROPE``: Accepts unicast hash.
903
904* ``BAM``: Accepts broadcast packets.
905
906* ``MPE``: Accepts all multicast packets.
907
908set port - tx_rate (for Queue)
909~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
910
911Set TX rate limitation for a queue on a port::
912
913   testpmd> set port (port_id) queue (queue_id) rate (rate_value)
914
915set port - tx_rate (for VF)
916~~~~~~~~~~~~~~~~~~~~~~~~~~~
917
918Set TX rate limitation for queues in VF on a port::
919
920   testpmd> set port (port_id) vf (vf_id) rate (rate_value) queue_mask (queue_mask)
921
922set port - mirror rule
923~~~~~~~~~~~~~~~~~~~~~~
924
925Set pool or vlan type mirror rule for a port::
926
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)
930
931Set link mirror rule for a port::
932
933   testpmd> set port (port_id) mirror-rule (rule_id) \
934           (uplink-mirror|downlink-mirror) dst-pool (pool_id) (on|off)
935
936For example to enable mirror traffic with vlan 0,1 to pool 0::
937
938   set port 0 mirror-rule 0 vlan-mirror 0,1 dst-pool 0 on
939
940reset port - mirror rule
941~~~~~~~~~~~~~~~~~~~~~~~~
942
943Reset a mirror rule for a port::
944
945   testpmd> reset port (port_id) mirror-rule (rule_id)
946
947set flush_rx
948~~~~~~~~~~~~
949
950Set the flush on RX streams before forwarding.
951The default is flush ``on``.
952Mainly used with PCAP drivers to turn off the default behavior of flushing the first 512 packets on RX streams::
953
954   testpmd> set flush_rx off
955
956set bypass mode
957~~~~~~~~~~~~~~~
958
959Set the bypass mode for the lowest port on bypass enabled NIC::
960
961   testpmd> set bypass mode (normal|bypass|isolate) (port_id)
962
963set bypass event
964~~~~~~~~~~~~~~~~
965
966Set the event required to initiate specified bypass mode for the lowest port on a bypass enabled::
967
968   testpmd> set bypass event (timeout|os_on|os_off|power_on|power_off) \
969            mode (normal|bypass|isolate) (port_id)
970
971Where:
972
973* ``timeout``: Enable bypass after watchdog timeout.
974
975* ``os_on``: Enable bypass when OS/board is powered on.
976
977* ``os_off``: Enable bypass when OS/board is powered off.
978
979* ``power_on``: Enable bypass when power supply is turned on.
980
981* ``power_off``: Enable bypass when power supply is turned off.
982
983
984set bypass timeout
985~~~~~~~~~~~~~~~~~~
986
987Set the bypass watchdog timeout to ``n`` seconds where 0 = instant::
988
989   testpmd> set bypass timeout (0|1.5|2|3|4|8|16|32)
990
991show bypass config
992~~~~~~~~~~~~~~~~~~
993
994Show the bypass configuration for a bypass enabled NIC using the lowest port on the NIC::
995
996   testpmd> show bypass config (port_id)
997
998set link up
999~~~~~~~~~~~
1000
1001Set link up for a port::
1002
1003   testpmd> set link-up port (port id)
1004
1005set link down
1006~~~~~~~~~~~~~
1007
1008Set link down for a port::
1009
1010   testpmd> set link-down port (port id)
1011
1012E-tag set
1013~~~~~~~~~
1014
1015Enable E-tag insertion for a VF on a port::
1016
1017   testpmd> E-tag set insertion on port-tag-id (value) port (port_id) vf (vf_id)
1018
1019Disable E-tag insertion for a VF on a port::
1020
1021   testpmd> E-tag set insertion off port (port_id) vf (vf_id)
1022
1023Enable/disable E-tag stripping on a port::
1024
1025   testpmd> E-tag set stripping (on|off) port (port_id)
1026
1027Enable/disable E-tag based forwarding on a port::
1028
1029   testpmd> E-tag set forwarding (on|off) port (port_id)
1030
1031Add an E-tag forwarding filter on a port::
1032
1033   testpmd> E-tag set filter add e-tag-id (value) dst-pool (pool_id) port (port_id)
1034
1035Delete an E-tag forwarding filter on a port::
1036   testpmd> E-tag set filter del e-tag-id (value) port (port_id)
1037
1038
1039Port Functions
1040--------------
1041
1042The following sections show functions for configuring ports.
1043
1044.. note::
1045
1046   Port configuration changes only become active when forwarding is started/restarted.
1047
1048port attach
1049~~~~~~~~~~~
1050
1051Attach a port specified by pci address or virtual device args::
1052
1053   testpmd> port attach (identifier)
1054
1055To attach a new pci device, the device should be recognized by kernel first.
1056Then it should be moved under DPDK management.
1057Finally the port can be attached to testpmd.
1058
1059For example, to move a pci device using ixgbe under DPDK management:
1060
1061.. code-block:: console
1062
1063   # Check the status of the available devices.
1064   ./usertools/dpdk-devbind.py --status
1065
1066   Network devices using DPDK-compatible driver
1067   ============================================
1068   <none>
1069
1070   Network devices using kernel driver
1071   ===================================
1072   0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused=
1073
1074
1075   # Bind the device to igb_uio.
1076   sudo ./usertools/dpdk-devbind.py -b igb_uio 0000:0a:00.0
1077
1078
1079   # Recheck the status of the devices.
1080   ./usertools/dpdk-devbind.py --status
1081   Network devices using DPDK-compatible driver
1082   ============================================
1083   0000:0a:00.0 '82599ES 10-Gigabit' drv=igb_uio unused=
1084
1085To attach a port created by virtual device, above steps are not needed.
1086
1087For example, to attach a port whose pci address is 0000:0a:00.0.
1088
1089.. code-block:: console
1090
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
1100   Done
1101
1102For example, to attach a port created by pcap PMD.
1103
1104.. code-block:: console
1105
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
1111   Done
1112
1113In this case, identifier is ``net_pcap0``.
1114This identifier format is the same as ``--vdev`` format of DPDK applications.
1115
1116For example, to re-attach a bonded port which has been previously detached,
1117the mode and slave parameters must be given.
1118
1119.. code-block:: console
1120
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
1126   Done
1127
1128
1129port detach
1130~~~~~~~~~~~
1131
1132Detach a specific port::
1133
1134   testpmd> port detach (port_id)
1135
1136Before detaching a port, the port should be stopped and closed.
1137
1138For example, to detach a pci device port 0.
1139
1140.. code-block:: console
1141
1142   testpmd> port stop 0
1143   Stopping ports...
1144   Done
1145   testpmd> port close 0
1146   Closing ports...
1147   Done
1148
1149   testpmd> port detach 0
1150   Detaching a port...
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
1155   Done
1156
1157
1158For example, to detach a virtual device port 0.
1159
1160.. code-block:: console
1161
1162   testpmd> port stop 0
1163   Stopping ports...
1164   Done
1165   testpmd> port close 0
1166   Closing ports...
1167   Done
1168
1169   testpmd> port detach 0
1170   Detaching a port...
1171   PMD: Closing pcap ethdev on numa socket 0
1172   Port 'net_pcap0' is detached. Now total ports is 0
1173   Done
1174
1175To remove a pci device completely from the system, first detach the port from testpmd.
1176Then the device should be moved under kernel management.
1177Finally the device can be removed using kernel pci hotplug functionality.
1178
1179For example, to move a pci device under kernel management:
1180
1181.. code-block:: console
1182
1183   sudo ./usertools/dpdk-devbind.py -b ixgbe 0000:0a:00.0
1184
1185   ./usertools/dpdk-devbind.py --status
1186
1187   Network devices using DPDK-compatible driver
1188   ============================================
1189   <none>
1190
1191   Network devices using kernel driver
1192   ===================================
1193   0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused=igb_uio
1194
1195To remove a port created by a virtual device, above steps are not needed.
1196
1197port start
1198~~~~~~~~~~
1199
1200Start all ports or a specific port::
1201
1202   testpmd> port start (port_id|all)
1203
1204port stop
1205~~~~~~~~~
1206
1207Stop all ports or a specific port::
1208
1209   testpmd> port stop (port_id|all)
1210
1211port close
1212~~~~~~~~~~
1213
1214Close all ports or a specific port::
1215
1216   testpmd> port close (port_id|all)
1217
1218port start/stop queue
1219~~~~~~~~~~~~~~~~~~~~~
1220
1221Start/stop a rx/tx queue on a specific port::
1222
1223   testpmd> port (port_id) (rxq|txq) (queue_id) (start|stop)
1224
1225Only take effect when port is started.
1226
1227port config - speed
1228~~~~~~~~~~~~~~~~~~~
1229
1230Set the speed and duplex mode for all ports or a specific port::
1231
1232   testpmd> port config (port_id|all) speed (10|100|1000|10000|25000|40000|50000|100000|auto) \
1233            duplex (half|full|auto)
1234
1235port config - queues/descriptors
1236~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1237
1238Set number of queues/descriptors for rxq, txq, rxd and txd::
1239
1240   testpmd> port config all (rxq|txq|rxd|txd) (value)
1241
1242This is equivalent to the ``--rxq``, ``--txq``, ``--rxd`` and ``--txd`` command-line options.
1243
1244port config - max-pkt-len
1245~~~~~~~~~~~~~~~~~~~~~~~~~
1246
1247Set the maximum packet length::
1248
1249   testpmd> port config all max-pkt-len (value)
1250
1251This is equivalent to the ``--max-pkt-len`` command-line option.
1252
1253port config - CRC Strip
1254~~~~~~~~~~~~~~~~~~~~~~~
1255
1256Set hardware CRC stripping on or off for all ports::
1257
1258   testpmd> port config all crc-strip (on|off)
1259
1260CRC stripping is off by default.
1261
1262The ``on`` option is equivalent to the ``--crc-strip`` command-line option.
1263
1264port config - scatter
1265~~~~~~~~~~~~~~~~~~~~~~~
1266
1267Set RX scatter mode on or off for all ports::
1268
1269   testpmd> port config all scatter (on|off)
1270
1271RX scatter mode is off by default.
1272
1273The ``on`` option is equivalent to the ``--enable-scatter`` command-line option.
1274
1275port config - TX queue flags
1276~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1277
1278Set a hexadecimal bitmap of TX queue flags for all ports::
1279
1280   testpmd> port config all txqflags value
1281
1282This command is equivalent to the ``--txqflags`` command-line option.
1283
1284port config - RX Checksum
1285~~~~~~~~~~~~~~~~~~~~~~~~~
1286
1287Set hardware RX checksum offload to on or off for all ports::
1288
1289   testpmd> port config all rx-cksum (on|off)
1290
1291Checksum offload is off by default.
1292
1293The ``on`` option is equivalent to the ``--enable-rx-cksum`` command-line option.
1294
1295port config - VLAN
1296~~~~~~~~~~~~~~~~~~
1297
1298Set hardware VLAN on or off for all ports::
1299
1300   testpmd> port config all hw-vlan (on|off)
1301
1302Hardware VLAN is on by default.
1303
1304The ``off`` option is equivalent to the ``--disable-hw-vlan`` command-line option.
1305
1306port config - VLAN filter
1307~~~~~~~~~~~~~~~~~~~~~~~~~
1308
1309Set hardware VLAN filter on or off for all ports::
1310
1311   testpmd> port config all hw-vlan-filter (on|off)
1312
1313Hardware VLAN filter is on by default.
1314
1315The ``off`` option is equivalent to the ``--disable-hw-vlan-filter`` command-line option.
1316
1317port config - VLAN strip
1318~~~~~~~~~~~~~~~~~~~~~~~~
1319
1320Set hardware VLAN strip on or off for all ports::
1321
1322   testpmd> port config all hw-vlan-strip (on|off)
1323
1324Hardware VLAN strip is on by default.
1325
1326The ``off`` option is equivalent to the ``--disable-hw-vlan-strip`` command-line option.
1327
1328port config - VLAN extend
1329~~~~~~~~~~~~~~~~~~~~~~~~~
1330
1331Set hardware VLAN extend on or off for all ports::
1332
1333   testpmd> port config all hw-vlan-extend (on|off)
1334
1335Hardware VLAN extend is off by default.
1336
1337The ``off`` option is equivalent to the ``--disable-hw-vlan-extend`` command-line option.
1338
1339port config - Drop Packets
1340~~~~~~~~~~~~~~~~~~~~~~~~~~
1341
1342Set packet drop for packets with no descriptors on or off for all ports::
1343
1344   testpmd> port config all drop-en (on|off)
1345
1346Packet dropping for packets with no descriptors is off by default.
1347
1348The ``on`` option is equivalent to the ``--enable-drop-en`` command-line option.
1349
1350port config - RSS
1351~~~~~~~~~~~~~~~~~
1352
1353Set the RSS (Receive Side Scaling) mode on or off::
1354
1355   testpmd> port config all rss (all|ip|tcp|udp|sctp|ether|port|vxlan|geneve|nvgre|none)
1356
1357RSS is on by default.
1358
1359The ``none`` option is equivalent to the ``--disable-rss`` command-line option.
1360
1361port config - RSS Reta
1362~~~~~~~~~~~~~~~~~~~~~~
1363
1364Set the RSS (Receive Side Scaling) redirection table::
1365
1366   testpmd> port config all rss reta (hash,queue)[,(hash,queue)]
1367
1368port config - DCB
1369~~~~~~~~~~~~~~~~~
1370
1371Set the DCB mode for an individual port::
1372
1373   testpmd> port config (port_id) dcb vt (on|off) (traffic_class) pfc (on|off)
1374
1375The traffic class should be 4 or 8.
1376
1377port config - Burst
1378~~~~~~~~~~~~~~~~~~~
1379
1380Set the number of packets per burst::
1381
1382   testpmd> port config all burst (value)
1383
1384This is equivalent to the ``--burst`` command-line option.
1385
1386port config - Threshold
1387~~~~~~~~~~~~~~~~~~~~~~~
1388
1389Set thresholds for TX/RX queues::
1390
1391   testpmd> port config all (threshold) (value)
1392
1393Where the threshold type can be:
1394
1395* ``txpt:`` Set the prefetch threshold register of the TX rings, 0 <= value <= 255.
1396
1397* ``txht:`` Set the host threshold register of the TX rings, 0 <= value <= 255.
1398
1399* ``txwt:`` Set the write-back threshold register of the TX rings, 0 <= value <= 255.
1400
1401* ``rxpt:`` Set the prefetch threshold register of the RX rings, 0 <= value <= 255.
1402
1403* ``rxht:`` Set the host threshold register of the RX rings, 0 <= value <= 255.
1404
1405* ``rxwt:`` Set the write-back threshold register of the RX rings, 0 <= value <= 255.
1406
1407* ``txfreet:`` Set the transmit free threshold of the TX rings, 0 <= value <= txd.
1408
1409* ``rxfreet:`` Set the transmit free threshold of the RX rings, 0 <= value <= rxd.
1410
1411* ``txrst:`` Set the transmit RS bit threshold of TX rings, 0 <= value <= txd.
1412
1413These threshold options are also available from the command-line.
1414
1415port config - E-tag
1416~~~~~~~~~~~~~~~~~~~
1417
1418Set the value of ether-type for E-tag::
1419
1420   testpmd> port config (port_id|all) l2-tunnel E-tag ether-type (value)
1421
1422Enable/disable the E-tag support::
1423
1424   testpmd> port config (port_id|all) l2-tunnel E-tag (enable|disable)
1425
1426
1427Link Bonding Functions
1428----------------------
1429
1430The Link Bonding functions make it possible to dynamically create and
1431manage link bonding devices from within testpmd interactive prompt.
1432
1433create bonded device
1434~~~~~~~~~~~~~~~~~~~~
1435
1436Create a new bonding device::
1437
1438   testpmd> create bonded device (mode) (socket)
1439
1440For example, to create a bonded device in mode 1 on socket 0::
1441
1442   testpmd> create bonded 1 0
1443   created new bonded device (port X)
1444
1445add bonding slave
1446~~~~~~~~~~~~~~~~~
1447
1448Adds Ethernet device to a Link Bonding device::
1449
1450   testpmd> add bonding slave (slave id) (port id)
1451
1452For example, to add Ethernet device (port 6) to a Link Bonding device (port 10)::
1453
1454   testpmd> add bonding slave 6 10
1455
1456
1457remove bonding slave
1458~~~~~~~~~~~~~~~~~~~~
1459
1460Removes an Ethernet slave device from a Link Bonding device::
1461
1462   testpmd> remove bonding slave (slave id) (port id)
1463
1464For example, to remove Ethernet slave device (port 6) to a Link Bonding device (port 10)::
1465
1466   testpmd> remove bonding slave 6 10
1467
1468set bonding mode
1469~~~~~~~~~~~~~~~~
1470
1471Set the Link Bonding mode of a Link Bonding device::
1472
1473   testpmd> set bonding mode (value) (port id)
1474
1475For example, to set the bonding mode of a Link Bonding device (port 10) to broadcast (mode 3)::
1476
1477   testpmd> set bonding mode 3 10
1478
1479set bonding primary
1480~~~~~~~~~~~~~~~~~~~
1481
1482Set an Ethernet slave device as the primary device on a Link Bonding device::
1483
1484   testpmd> set bonding primary (slave id) (port id)
1485
1486For example, to set the Ethernet slave device (port 6) as the primary port of a Link Bonding device (port 10)::
1487
1488   testpmd> set bonding primary 6 10
1489
1490set bonding mac
1491~~~~~~~~~~~~~~~
1492
1493Set the MAC address of a Link Bonding device::
1494
1495   testpmd> set bonding mac (port id) (mac)
1496
1497For example, to set the MAC address of a Link Bonding device (port 10) to 00:00:00:00:00:01::
1498
1499   testpmd> set bonding mac 10 00:00:00:00:00:01
1500
1501set bonding xmit_balance_policy
1502~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1503
1504Set the transmission policy for a Link Bonding device when it is in Balance XOR mode::
1505
1506   testpmd> set bonding xmit_balance_policy (port_id) (l2|l23|l34)
1507
1508For example, set a Link Bonding device (port 10) to use a balance policy of layer 3+4 (IP addresses & UDP ports)::
1509
1510   testpmd> set bonding xmit_balance_policy 10 l34
1511
1512
1513set bonding mon_period
1514~~~~~~~~~~~~~~~~~~~~~~
1515
1516Set the link status monitoring polling period in milliseconds for a bonding device.
1517
1518This adds support for PMD slave devices which do not support link status interrupts.
1519When the mon_period is set to a value greater than 0 then all PMD's which do not support
1520link status ISR will be queried every polling interval to check if their link status has changed::
1521
1522   testpmd> set bonding mon_period (port_id) (value)
1523
1524For example, to set the link status monitoring polling period of bonded device (port 5) to 150ms::
1525
1526   testpmd> set bonding mon_period 5 150
1527
1528
1529show bonding config
1530~~~~~~~~~~~~~~~~~~~
1531
1532Show the current configuration of a Link Bonding device::
1533
1534   testpmd> show bonding config (port id)
1535
1536For example,
1537to show the configuration a Link Bonding device (port 9) with 3 slave devices (1, 3, 4)
1538in balance mode with a transmission policy of layer 2+3::
1539
1540   testpmd> show bonding config 9
1541        Bonding mode: 2
1542        Balance Xmit Policy: BALANCE_XMIT_POLICY_LAYER23
1543        Slaves (3): [1 3 4]
1544        Active Slaves (3): [1 3 4]
1545        Primary: [3]
1546
1547
1548Register Functions
1549------------------
1550
1551The Register Functions can be used to read from and write to registers on the network card referenced by a port number.
1552This is mainly useful for debugging purposes.
1553Reference should be made to the appropriate datasheet for the network card for details on the register addresses
1554and fields that can be accessed.
1555
1556read reg
1557~~~~~~~~
1558
1559Display the value of a port register::
1560
1561   testpmd> read reg (port_id) (address)
1562
1563For example, to examine the Flow Director control register (FDIRCTL, 0x0000EE000) on an Intel 82599 10 GbE Controller::
1564
1565   testpmd> read reg 0 0xEE00
1566   port 0 PCI register at offset 0xEE00: 0x4A060029 (1241907241)
1567
1568read regfield
1569~~~~~~~~~~~~~
1570
1571Display a port register bit field::
1572
1573   testpmd> read regfield (port_id) (address) (bit_x) (bit_y)
1574
1575For example, reading the lowest two bits from the register in the example above::
1576
1577   testpmd> read regfield 0 0xEE00 0 1
1578   port 0 PCI register at offset 0xEE00: bits[0, 1]=0x1 (1)
1579
1580read regbit
1581~~~~~~~~~~~
1582
1583Display a single port register bit::
1584
1585   testpmd> read regbit (port_id) (address) (bit_x)
1586
1587For example, reading the lowest bit from the register in the example above::
1588
1589   testpmd> read regbit 0 0xEE00 0
1590   port 0 PCI register at offset 0xEE00: bit 0=1
1591
1592write reg
1593~~~~~~~~~
1594
1595Set the value of a port register::
1596
1597   testpmd> write reg (port_id) (address) (value)
1598
1599For example, to clear a register::
1600
1601   testpmd> write reg 0 0xEE00 0x0
1602   port 0 PCI register at offset 0xEE00: 0x00000000 (0)
1603
1604write regfield
1605~~~~~~~~~~~~~~
1606
1607Set bit field of a port register::
1608
1609   testpmd> write regfield (port_id) (address) (bit_x) (bit_y) (value)
1610
1611For example, writing to the register cleared in the example above::
1612
1613   testpmd> write regfield 0 0xEE00 0 1 2
1614   port 0 PCI register at offset 0xEE00: 0x00000002 (2)
1615
1616write regbit
1617~~~~~~~~~~~~
1618
1619Set single bit value of a port register::
1620
1621   testpmd> write regbit (port_id) (address) (bit_x) (value)
1622
1623For example, to set the high bit in the register from the example above::
1624
1625   testpmd> write regbit 0 0xEE00 31 1
1626   port 0 PCI register at offset 0xEE00: 0x8000000A (2147483658)
1627
1628
1629Filter Functions
1630----------------
1631
1632This section details the available filter functions that are available.
1633
1634Note these functions interface the deprecated legacy filtering framework,
1635superseded by *rte_flow*. See `Flow rules management`_.
1636
1637ethertype_filter
1638~~~~~~~~~~~~~~~~~~~~
1639
1640Add or delete a L2 Ethertype filter, which identify packets by their L2 Ethertype mainly assign them to a receive queue::
1641
1642   ethertype_filter (port_id) (add|del) (mac_addr|mac_ignr) (mac_address) \
1643                    ethertype (ether_type) (drop|fwd) queue (queue_id)
1644
1645The available information parameters are:
1646
1647* ``port_id``: The port which the Ethertype filter assigned on.
1648
1649* ``mac_addr``: Compare destination mac address.
1650
1651* ``mac_ignr``: Ignore destination mac address match.
1652
1653* ``mac_address``: Destination mac address to match.
1654
1655* ``ether_type``: The EtherType value want to match,
1656  for example 0x0806 for ARP packet. 0x0800 (IPv4) and 0x86DD (IPv6) are invalid.
1657
1658* ``queue_id``: The receive queue associated with this EtherType filter.
1659  It is meaningless when deleting or dropping.
1660
1661Example, to add/remove an ethertype filter rule::
1662
1663   testpmd> ethertype_filter 0 add mac_ignr 00:11:22:33:44:55 \
1664                             ethertype 0x0806 fwd queue 3
1665
1666   testpmd> ethertype_filter 0 del mac_ignr 00:11:22:33:44:55 \
1667                             ethertype 0x0806 fwd queue 3
1668
16692tuple_filter
1670~~~~~~~~~~~~~~~~~
1671
1672Add or delete a 2-tuple filter,
1673which identifies packets by specific protocol and destination TCP/UDP port
1674and forwards packets into one of the receive queues::
1675
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) \
1679                 queue (queue_id)
1680
1681The available information parameters are:
1682
1683* ``port_id``: The port which the 2-tuple filter assigned on.
1684
1685* ``dst_port_value``: Destination port in L4.
1686
1687* ``protocol_value``: IP L4 protocol.
1688
1689* ``mask_value``: Participates in the match or not by bit for field above, 1b means participate.
1690
1691* ``tcp_flags_value``: TCP control bits. The non-zero value is invalid, when the pro_value is not set to 0x06 (TCP).
1692
1693* ``prio_value``: Priority of this filter.
1694
1695* ``queue_id``: The receive queue associated with this 2-tuple filter.
1696
1697Example, to add/remove an 2tuple filter rule::
1698
1699   testpmd> 2tuple_filter 0 add dst_port 32 protocol 0x06 mask 0x03 \
1700                          tcp_flags 0x02 priority 3 queue 3
1701
1702   testpmd> 2tuple_filter 0 del dst_port 32 protocol 0x06 mask 0x03 \
1703                          tcp_flags 0x02 priority 3 queue 3
1704
17055tuple_filter
1706~~~~~~~~~~~~~~~~~
1707
1708Add or delete a 5-tuple filter,
1709which consists of a 5-tuple (protocol, source and destination IP addresses, source and destination TCP/UDP/SCTP port)
1710and routes packets into one of the receive queues::
1711
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)
1717
1718The available information parameters are:
1719
1720* ``port_id``: The port which the 5-tuple filter assigned on.
1721
1722* ``dst_address``: Destination IP address.
1723
1724* ``src_address``: Source IP address.
1725
1726* ``dst_port_value``: TCP/UDP destination port.
1727
1728* ``src_port_value``: TCP/UDP source port.
1729
1730* ``protocol_value``: L4 protocol.
1731
1732* ``mask_value``: Participates in the match or not by bit for field above, 1b means participate
1733
1734* ``tcp_flags_value``: TCP control bits. The non-zero value is invalid, when the protocol_value is not set to 0x06 (TCP).
1735
1736* ``prio_value``: The priority of this filter.
1737
1738* ``queue_id``: The receive queue associated with this 5-tuple filter.
1739
1740Example, to add/remove an 5tuple filter rule::
1741
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
1745
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
1749
1750syn_filter
1751~~~~~~~~~~
1752
1753Using the  SYN filter, TCP packets whose *SYN* flag is set can be forwarded to a separate queue::
1754
1755   syn_filter (port_id) (add|del) priority (high|low) queue (queue_id)
1756
1757The available information parameters are:
1758
1759* ``port_id``: The port which the SYN filter assigned on.
1760
1761* ``high``: This SYN filter has higher priority than other filters.
1762
1763* ``low``: This SYN filter has lower priority than other filters.
1764
1765* ``queue_id``: The receive queue associated with this SYN filter
1766
1767Example::
1768
1769   testpmd> syn_filter 0 add priority high queue 3
1770
1771flex_filter
1772~~~~~~~~~~~
1773
1774With flex filter, packets can be recognized by any arbitrary pattern within the first 128 bytes of the packet
1775and routed into one of the receive queues::
1776
1777   flex_filter (port_id) (add|del) len (len_value) bytes (bytes_value) \
1778               mask (mask_value) priority (prio_value) queue (queue_id)
1779
1780The available information parameters are:
1781
1782* ``port_id``: The port which the Flex filter is assigned on.
1783
1784* ``len_value``: Filter length in bytes, no greater than 128.
1785
1786* ``bytes_value``: A string in hexadecimal, means the value the flex filter needs to match.
1787
1788* ``mask_value``: A string in hexadecimal, bit 1 means corresponding byte participates in the match.
1789
1790* ``prio_value``: The priority of this filter.
1791
1792* ``queue_id``: The receive queue associated with this Flex filter.
1793
1794Example::
1795
1796   testpmd> flex_filter 0 add len 16 bytes 0x00000000000000000000000008060000 \
1797                          mask 000C priority 3 queue 3
1798
1799   testpmd> flex_filter 0 del len 16 bytes 0x00000000000000000000000008060000 \
1800                          mask 000C priority 3 queue 3
1801
1802
1803.. _testpmd_flow_director:
1804
1805flow_director_filter
1806~~~~~~~~~~~~~~~~~~~~
1807
1808The Flow Director works in receive mode to identify specific flows or sets of flows and route them to specific queues.
1809
1810Four types of filtering are supported which are referred to as Perfect Match, Signature, Perfect-mac-vlan and
1811Perfect-tunnel filters, the match mode is set by the ``--pkt-filter-mode`` command-line parameter:
1812
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.
1816
1817* Signature filters.
1818  The hardware checks a match between a hash-based signature of the masked fields of the received packet.
1819
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.
1823
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.
1827
1828The Flow Director filters can match the different fields for different type of packet: flow type, specific input set
1829per flow type and the flexible payload.
1830
1831The Flow Director can also mask out parts of all of these fields so that filters
1832are only applied to certain fields or parts of the fields.
1833
1834Different NICs may have different capabilities, command show port fdir (port_id) can be used to acquire the information.
1835
1836# Commands to add flow director filters of different flow types::
1837
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) \
1844                        fd_id (fd_id_value)
1845
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) \
1853                        fd_id (fd_id_value)
1854
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)
1863
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)
1867                        fd_id (fd_id_value)
1868
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)
1873
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)
1879
1880For example, to add an ipv4-udp flow type filter::
1881
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
1885
1886For example, add an ipv4-other flow type filter::
1887
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
1891
1892flush_flow_director
1893~~~~~~~~~~~~~~~~~~~
1894
1895Flush all flow director filters on a device::
1896
1897   testpmd> flush_flow_director (port_id)
1898
1899Example, to flush all flow director filter on port 0::
1900
1901   testpmd> flush_flow_director 0
1902
1903flow_director_mask
1904~~~~~~~~~~~~~~~~~~
1905
1906Set flow director's input masks::
1907
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)
1911
1912   flow_director_mask (port_id) mode MAC-VLAN vlan (vlan_value)
1913
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)
1917
1918Example, to set flow director mask on port 0::
1919
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
1925
1926flow_director_flex_mask
1927~~~~~~~~~~~~~~~~~~~~~~~
1928
1929set masks of flow director's flexible payload based on certain flow type::
1930
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)
1935
1936Example, to set flow director's flex mask for all flow type on port 0::
1937
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)
1940
1941
1942flow_director_flex_payload
1943~~~~~~~~~~~~~~~~~~~~~~~~~~
1944
1945Configure flexible payload selection::
1946
1947   flow_director_flex_payload (port_id) (raw|l2|l3|l4) (config)
1948
1949For example, to select the first 16 bytes from the offset 4 (bytes) of packet's payload as flexible payload::
1950
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)
1953
1954get_sym_hash_ena_per_port
1955~~~~~~~~~~~~~~~~~~~~~~~~~
1956
1957Get symmetric hash enable configuration per port::
1958
1959   get_sym_hash_ena_per_port (port_id)
1960
1961For example, to get symmetric hash enable configuration of port 1::
1962
1963   testpmd> get_sym_hash_ena_per_port 1
1964
1965set_sym_hash_ena_per_port
1966~~~~~~~~~~~~~~~~~~~~~~~~~
1967
1968Set symmetric hash enable configuration per port to enable or disable::
1969
1970   set_sym_hash_ena_per_port (port_id) (enable|disable)
1971
1972For example, to set symmetric hash enable configuration of port 1 to enable::
1973
1974   testpmd> set_sym_hash_ena_per_port 1 enable
1975
1976get_hash_global_config
1977~~~~~~~~~~~~~~~~~~~~~~
1978
1979Get the global configurations of hash filters::
1980
1981   get_hash_global_config (port_id)
1982
1983For example, to get the global configurations of hash filters of port 1::
1984
1985   testpmd> get_hash_global_config 1
1986
1987set_hash_global_config
1988~~~~~~~~~~~~~~~~~~~~~~
1989
1990Set the global configurations of hash filters::
1991
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) \
1995   (enable|disable)
1996
1997For example, to enable simple_xor for flow type of ipv6 on port 2::
1998
1999   testpmd> set_hash_global_config 2 simple_xor ipv6 enable
2000
2001set_hash_input_set
2002~~~~~~~~~~~~~~~~~~
2003
2004Set the input set for hash::
2005
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)
2013
2014For example, to add source IP to hash input set for flow type of ipv4-udp on port 0::
2015
2016   testpmd> set_hash_input_set 0 ipv4-udp src-ipv4 add
2017
2018set_fdir_input_set
2019~~~~~~~~~~~~~~~~~~
2020
2021The Flow Director filters can match the different fields for different type of packet, i.e. specific input set
2022on per flow type and the flexible payload. This command can be used to change input set for each flow type.
2023
2024Set the input set for flow director::
2025
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)
2032
2033For example to add source IP to FD input set for flow type of ipv4-udp on port 0::
2034
2035   testpmd> set_fdir_input_set 0 ipv4-udp src-ipv4 add
2036
2037global_config
2038~~~~~~~~~~~~~
2039
2040Set different GRE key length for input set::
2041
2042   global_config (port_id) gre-key-len (number in bytes)
2043
2044For example to set GRE key length for input set to 4 bytes on port 0::
2045
2046   testpmd> global_config 0 gre-key-len 4
2047
2048
2049.. _testpmd_rte_flow:
2050
2051Flow rules management
2052---------------------
2053
2054Control of the generic flow API (*rte_flow*) is fully exposed through the
2055``flow`` command (validation, creation, destruction and queries).
2056
2057Considering *rte_flow* overlaps with all `Filter Functions`_, using both
2058features simultaneously may cause undefined side-effects and is therefore
2059not recommended.
2060
2061``flow`` syntax
2062~~~~~~~~~~~~~~~
2063
2064Because the ``flow`` command uses dynamic tokens to handle the large number
2065of possible flow rules combinations, its behavior differs slightly from
2066other commands, in particular:
2067
2068- Pressing *?* or the *<tab>* key displays contextual help for the current
2069  token, not that of the entire command.
2070
2071- Optional and repeated parameters are supported (provided they are listed
2072  in the contextual help).
2073
2074The first parameter stands for the operation mode. Possible operations and
2075their general syntax are described below. They are covered in detail in the
2076following sections.
2077
2078- Check whether a flow rule can be created::
2079
2080   flow validate {port_id}
2081       [group {group_id}] [priority {level}] [ingress] [egress]
2082       pattern {item} [/ {item} [...]] / end
2083       actions {action} [/ {action} [...]] / end
2084
2085- Create a flow rule::
2086
2087   flow create {port_id}
2088       [group {group_id}] [priority {level}] [ingress] [egress]
2089       pattern {item} [/ {item} [...]] / end
2090       actions {action} [/ {action} [...]] / end
2091
2092- Destroy specific flow rules::
2093
2094   flow destroy {port_id} rule {rule_id} [...]
2095
2096- Destroy all flow rules::
2097
2098   flow flush {port_id}
2099
2100- Query an existing flow rule::
2101
2102   flow query {port_id} {rule_id} {action}
2103
2104- List existing flow rules sorted by priority, filtered by group
2105  identifiers::
2106
2107   flow list {port_id} [group {group_id}] [...]
2108
2109Validating flow rules
2110~~~~~~~~~~~~~~~~~~~~~
2111
2112``flow validate`` reports whether a flow rule would be accepted by the
2113underlying device in its current state but stops short of creating it. It is
2114bound to ``rte_flow_validate()``::
2115
2116   flow validate {port_id}
2117      [group {group_id}] [priority {level}] [ingress] [egress]
2118      pattern {item} [/ {item} [...]] / end
2119      actions {action} [/ {action} [...]] / end
2120
2121If successful, it will show::
2122
2123   Flow rule validated
2124
2125Otherwise it will show an error message of the form::
2126
2127   Caught error type [...] ([...]): [...]
2128
2129This command uses the same parameters as ``flow create``, their format is
2130described in `Creating flow rules`_.
2131
2132Check whether redirecting any Ethernet packet received on port 0 to RX queue
2133index 6 is supported::
2134
2135   testpmd> flow validate 0 ingress pattern eth / end
2136      actions queue index 6 / end
2137   Flow rule validated
2138   testpmd>
2139
2140Port 0 does not support TCPv6 rules::
2141
2142   testpmd> flow validate 0 ingress pattern eth / ipv6 / tcp / end
2143      actions drop / end
2144   Caught error type 9 (specific pattern item): Invalid argument
2145   testpmd>
2146
2147Creating flow rules
2148~~~~~~~~~~~~~~~~~~~
2149
2150``flow create`` validates and creates the specified flow rule. It is bound
2151to ``rte_flow_create()``::
2152
2153   flow create {port_id}
2154      [group {group_id}] [priority {level}] [ingress] [egress]
2155      pattern {item} [/ {item} [...]] / end
2156      actions {action} [/ {action} [...]] / end
2157
2158If successful, it will return a flow rule ID usable with other commands::
2159
2160   Flow rule #[...] created
2161
2162Otherwise it will show an error message of the form::
2163
2164   Caught error type [...] ([...]): [...]
2165
2166Parameters describe in the following order:
2167
2168- Attributes (*group*, *priority*, *ingress*, *egress* tokens).
2169- A matching pattern, starting with the *pattern* token and terminated by an
2170  *end* pattern item.
2171- Actions, starting with the *actions* token and terminated by an *end*
2172  action.
2173
2174These translate directly to *rte_flow* objects provided as-is to the
2175underlying functions.
2176
2177The shortest valid definition only comprises mandatory tokens::
2178
2179   testpmd> flow create 0 pattern end actions end
2180
2181Note that PMDs may refuse rules that essentially do nothing such as this
2182one.
2183
2184**All unspecified object values are automatically initialized to 0.**
2185
2186Attributes
2187^^^^^^^^^^
2188
2189These tokens affect flow rule attributes (``struct rte_flow_attr``) and are
2190specified before the ``pattern`` token.
2191
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.
2196
2197Each instance of an attribute specified several times overrides the previous
2198value as shown below (group 4 is used)::
2199
2200   testpmd> flow create 0 group 42 group 24 group 4 [...]
2201
2202Note that once enabled, ``ingress`` and ``egress`` cannot be disabled.
2203
2204While not specifying a direction is an error, some rules may allow both
2205simultaneously.
2206
2207Most rules affect RX therefore contain the ``ingress`` token::
2208
2209   testpmd> flow create 0 ingress pattern [...]
2210
2211Matching pattern
2212^^^^^^^^^^^^^^^^
2213
2214A matching pattern starts after the ``pattern`` token. It is made of pattern
2215items and is terminated by a mandatory ``end`` item.
2216
2217Items are named after their type (*RTE_FLOW_ITEM_TYPE_* from ``enum
2218rte_flow_item_type``).
2219
2220The ``/`` token is used as a separator between pattern items as shown
2221below::
2222
2223   testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end [...]
2224
2225Note that protocol items like these must be stacked from lowest to highest
2226layer to make sense. For instance, the following rule is either invalid or
2227unlikely to match any packet::
2228
2229   testpmd> flow create 0 ingress pattern eth / udp / ipv4 / end [...]
2230
2231More information on these restrictions can be found in the *rte_flow*
2232documentation.
2233
2234Several items support additional specification structures, for example
2235``ipv4`` allows specifying source and destination addresses as follows::
2236
2237   testpmd> flow create 0 ingress pattern eth / ipv4 src is 10.1.1.1
2238      dst is 10.2.0.0 / end [...]
2239
2240This rule matches all IPv4 traffic with the specified properties.
2241
2242In this example, ``src`` and ``dst`` are field names of the underlying
2243``struct rte_flow_item_ipv4`` object. All item properties can be specified
2244in a similar fashion.
2245
2246The ``is`` token means that the subsequent value must be matched exactly,
2247and assigns ``spec`` and ``mask`` fields in ``struct rte_flow_item``
2248accordingly. Possible assignment tokens are:
2249
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.
2255
2256These yield identical results::
2257
2258   ipv4 src is 10.1.1.1
2259
2260::
2261
2262   ipv4 src spec 10.1.1.1 src mask 255.255.255.255
2263
2264::
2265
2266   ipv4 src spec 10.1.1.1 src prefix 32
2267
2268::
2269
2270   ipv4 src is 10.1.1.1 src last 10.1.1.1 # range with a single value
2271
2272::
2273
2274   ipv4 src is 10.1.1.1 src last 0 # 0 disables range
2275
2276Inclusive ranges can be defined with ``last``::
2277
2278   ipv4 src is 10.1.1.1 src last 10.2.3.4 # 10.1.1.1 to 10.2.3.4
2279
2280Note that ``mask`` affects both ``spec`` and ``last``::
2281
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
2284
2285Properties can be modified multiple times::
2286
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
2288
2289::
2290
2291   ipv4 src is 10.1.1.1 src prefix 24 src prefix 16 # matches 10.1.0.0/16
2292
2293Pattern items
2294^^^^^^^^^^^^^
2295
2296This section lists supported pattern items and their attributes, if any.
2297
2298- ``end``: end list of pattern items.
2299
2300- ``void``: no-op pattern item.
2301
2302- ``invert``: perform actions when pattern does not match.
2303
2304- ``any``: match any protocol for the current layer.
2305
2306  - ``num {unsigned}``: number of layers covered.
2307
2308- ``pf``: match packets addressed to the physical function.
2309
2310- ``vf``: match packets addressed to a virtual function ID.
2311
2312  - ``id {unsigned}``: destination VF ID.
2313
2314- ``port``: device-specific physical port index to use.
2315
2316  - ``index {unsigned}``: physical port index.
2317
2318- ``raw``: match an arbitrary byte string.
2319
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.
2325
2326- ``eth``: match Ethernet header.
2327
2328  - ``dst {MAC-48}``: destination MAC.
2329  - ``src {MAC-48}``: source MAC.
2330  - ``type {unsigned}``: EtherType.
2331
2332- ``vlan``: match 802.1Q/ad VLAN tag.
2333
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.
2339
2340- ``ipv4``: match IPv4 header.
2341
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.
2347
2348- ``ipv6``: match IPv6 header.
2349
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.
2356
2357- ``icmp``: match ICMP header.
2358
2359  - ``type {unsigned}``: ICMP packet type.
2360  - ``code {unsigned}``: ICMP packet code.
2361
2362- ``udp``: match UDP header.
2363
2364  - ``src {unsigned}``: UDP source port.
2365  - ``dst {unsigned}``: UDP destination port.
2366
2367- ``tcp``: match TCP header.
2368
2369  - ``src {unsigned}``: TCP source port.
2370  - ``dst {unsigned}``: TCP destination port.
2371
2372- ``sctp``: match SCTP header.
2373
2374  - ``src {unsigned}``: SCTP source port.
2375  - ``dst {unsigned}``: SCTP destination port.
2376  - ``tag {unsigned}``: validation tag.
2377  - ``cksum {unsigned}``: checksum.
2378
2379- ``vxlan``: match VXLAN header.
2380
2381  - ``vni {unsigned}``: VXLAN identifier.
2382
2383Actions list
2384^^^^^^^^^^^^
2385
2386A 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
2388terminated by a mandatory ``end`` action.
2389
2390Actions are named after their type (*RTE_FLOW_ACTION_TYPE_* from ``enum
2391rte_flow_action_type``).
2392
2393Dropping all incoming UDPv4 packets can be expressed as follows::
2394
2395   testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
2396      actions drop / end
2397
2398Several actions have configurable properties which must be specified when
2399there is no valid default value. For example, ``queue`` requires a target
2400queue index.
2401
2402This rule redirects incoming UDPv4 traffic to queue index 6::
2403
2404   testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
2405      actions queue index 6 / end
2406
2407While this one could be rejected by PMDs (unspecified queue index)::
2408
2409   testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
2410      actions queue / end
2411
2412As defined by *rte_flow*, the list is not ordered, all actions of a given
2413rule are performed simultaneously. These are equivalent::
2414
2415   queue index 6 / void / mark id 42 / end
2416
2417::
2418
2419   void / mark id 42 / queue index 6 / end
2420
2421All actions in a list should have different types, otherwise only the last
2422action of a given type is taken into account::
2423
2424   queue index 4 / queue index 5 / queue index 6 / end # will use queue 6
2425
2426::
2427
2428   drop / drop / drop / end # drop is performed only once
2429
2430::
2431
2432   mark id 42 / queue index 3 / mark id 24 / end # mark will be 24
2433
2434Considering they are performed simultaneously, opposite and overlapping
2435actions can sometimes be combined when the end result is unambiguous::
2436
2437   drop / queue index 6 / end # drop has no effect
2438
2439::
2440
2441   drop / dup index 6 / end # same as above
2442
2443::
2444
2445   queue index 6 / rss queues 6 7 8 / end # queue has no effect
2446
2447::
2448
2449   drop / passthru / end # drop has no effect
2450
2451Note that PMDs may still refuse such combinations.
2452
2453Actions
2454^^^^^^^
2455
2456This section lists supported actions and their attributes, if any.
2457
2458- ``end``: end list of actions.
2459
2460- ``void``: no-op action.
2461
2462- ``passthru``: let subsequent rule process matched packets.
2463
2464- ``mark``: attach 32 bit value to packets.
2465
2466  - ``id {unsigned}``: 32 bit value to return with packets.
2467
2468- ``flag``: flag packets.
2469
2470- ``queue``: assign packets to a given queue index.
2471
2472  - ``index {unsigned}``: queue index to use.
2473
2474- ``drop``: drop packets (note: passthru has priority).
2475
2476- ``count``: enable counters for this rule.
2477
2478- ``dup``: duplicate packets to a given queue index.
2479
2480  - ``index {unsigned}``: queue index to duplicate packets to.
2481
2482- ``rss``: spread packets among several queues.
2483
2484  - ``queues [{unsigned} [...]] end``: queue indices to use.
2485
2486- ``pf``: redirect packets to physical device function.
2487
2488- ``vf``: redirect packets to virtual device function.
2489
2490  - ``original {boolean}``: use original VF ID if possible.
2491  - ``id {unsigned}``: VF ID to redirect packets to.
2492
2493Destroying flow rules
2494~~~~~~~~~~~~~~~~~~~~~
2495
2496``flow destroy`` destroys one or more rules from their rule ID (as returned
2497by ``flow create``), this command calls ``rte_flow_destroy()`` as many
2498times as necessary::
2499
2500   flow destroy {port_id} rule {rule_id} [...]
2501
2502If successful, it will show::
2503
2504   Flow rule #[...] destroyed
2505
2506It does not report anything for rule IDs that do not exist. The usual error
2507message is shown when a rule cannot be destroyed::
2508
2509   Caught error type [...] ([...]): [...]
2510
2511``flow flush`` destroys all rules on a device and does not take extra
2512arguments. It is bound to ``rte_flow_flush()``::
2513
2514   flow flush {port_id}
2515
2516Any errors are reported as above.
2517
2518Creating several rules and destroying them::
2519
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
2529   testpmd>
2530
2531The same result can be achieved using ``flow flush``::
2532
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
2540   testpmd>
2541
2542Non-existent rule IDs are ignored::
2543
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
2551   testpmd>
2552   testpmd> flow destroy 0 rule 0
2553   Flow rule #0 destroyed
2554   testpmd>
2555
2556Querying flow rules
2557~~~~~~~~~~~~~~~~~~~
2558
2559``flow query`` queries a specific action of a flow rule having that
2560ability. Such actions collect information that can be reported using this
2561command. It is bound to ``rte_flow_query()``::
2562
2563   flow query {port_id} {rule_id} {action}
2564
2565If successful, it will display either the retrieved data for known actions
2566or the following message::
2567
2568   Cannot display result for action type [...] ([...])
2569
2570Otherwise, it will complain either that the rule does not exist or that some
2571error occurred::
2572
2573   Flow rule #[...] not found
2574
2575::
2576
2577   Caught error type [...] ([...]): [...]
2578
2579Currently only the ``count`` action is supported. This action reports the
2580number of packets that hit the flow rule and the total number of bytes. Its
2581output has the following format::
2582
2583   count:
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
2588
2589Querying counters for TCPv6 packets redirected to queue 6::
2590
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
2595   count:
2596    hits_set: 1
2597    bytes_set: 0
2598    hits: 386446
2599    bytes: 0
2600   testpmd>
2601
2602Listing flow rules
2603~~~~~~~~~~~~~~~~~~
2604
2605``flow list`` lists existing flow rules sorted by priority and optionally
2606filtered by group identifiers::
2607
2608   flow list {port_id} [group {group_id}] [...]
2609
2610This command only fails with the following message if the device does not
2611exist::
2612
2613   Invalid port [...]
2614
2615Output consists of a header line followed by a short description of each
2616flow rule, one per line. There is no output at all when no flow rules are
2617configured on the device::
2618
2619   ID      Group   Prio    Attr    Rule
2620   [...]   [...]   [...]   [...]   [...]
2621
2622``Attr`` column flags:
2623
2624- ``i`` for ``ingress``.
2625- ``e`` for ``egress``.
2626
2627Creating several flow rules and listing them::
2628
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
2643   testpmd>
2644
2645Rules are sorted by priority (i.e. group ID first, then priority level)::
2646
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
2657   testpmd>
2658
2659Output can be limited to specific groups::
2660
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
2667   testpmd>
2668