xref: /dpdk/doc/guides/testpmd_app_ug/testpmd_funcs.rst (revision b7091f1dcfbc62f358b4c1882d51c434471d51da)
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|cap X
55       info [Mul-choice STRING]: show|clear port info|stats|xstats|fdir|stat_qmap|dcb_tc|cap all
56       stats [Mul-choice STRING]: show|clear port info|stats|xstats|fdir|stat_qmap|dcb_tc|cap X
57       stats [Mul-choice STRING]: show|clear port info|stats|xstats|fdir|stat_qmap|dcb_tc|cap 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
89Command File Functions
90----------------------
91
92To facilitate loading large number of commands or to avoid cutting and pasting where not
93practical or possible testpmd supports alternative methods for executing commands.
94
95* If started with the ``--cmdline-file=FILENAME`` command line argument testpmd
96  will execute all CLI commands contained within the file immediately before
97  starting packet forwarding or entering interactive mode.
98
99.. code-block:: console
100
101   ./testpmd -n4 -r2 ... -- -i --cmdline-file=/home/ubuntu/flow-create-commands.txt
102   Interactive-mode selected
103   CLI commands to be read from /home/ubuntu/flow-create-commands.txt
104   Configuring Port 0 (socket 0)
105   Port 0: 7C:FE:90:CB:74:CE
106   Configuring Port 1 (socket 0)
107   Port 1: 7C:FE:90:CB:74:CA
108   Checking link statuses...
109   Port 0 Link Up - speed 10000 Mbps - full-duplex
110   Port 1 Link Up - speed 10000 Mbps - full-duplex
111   Done
112   Flow rule #0 created
113   Flow rule #1 created
114   ...
115   ...
116   Flow rule #498 created
117   Flow rule #499 created
118   Read all CLI commands from /home/ubuntu/flow-create-commands.txt
119   testpmd>
120
121
122* At run-time additional commands can be loaded in bulk by invoking the ``load FILENAME``
123  command.
124
125.. code-block:: console
126
127   testpmd> load /home/ubuntu/flow-create-commands.txt
128   Flow rule #0 created
129   Flow rule #1 created
130   ...
131   ...
132   Flow rule #498 created
133   Flow rule #499 created
134   Read all CLI commands from /home/ubuntu/flow-create-commands.txt
135   testpmd>
136
137
138In all cases output from any included command will be displayed as standard output.
139Execution will continue until the end of the file is reached regardless of
140whether any errors occur.  The end user must examine the output to determine if
141any failures occurred.
142
143
144Control Functions
145-----------------
146
147start
148~~~~~
149
150Start packet forwarding with current configuration::
151
152   testpmd> start
153
154start tx_first
155~~~~~~~~~~~~~~
156
157Start packet forwarding with current configuration after sending specified number of bursts of packets::
158
159   testpmd> start tx_first (""|burst_num)
160
161The default burst number is 1 when ``burst_num`` not presented.
162
163stop
164~~~~
165
166Stop packet forwarding, and display accumulated statistics::
167
168   testpmd> stop
169
170quit
171~~~~
172
173Quit to prompt::
174
175   testpmd> quit
176
177
178Display Functions
179-----------------
180
181The functions in the following sections are used to display information about the
182testpmd configuration or the NIC status.
183
184show port
185~~~~~~~~~
186
187Display information for a given port or all ports::
188
189   testpmd> show port (info|stats|xstats|fdir|stat_qmap|dcb_tc|cap) (port_id|all)
190
191The available information categories are:
192
193* ``info``: General port information such as MAC address.
194
195* ``stats``: RX/TX statistics.
196
197* ``xstats``: RX/TX extended NIC statistics.
198
199* ``fdir``: Flow Director information and statistics.
200
201* ``stat_qmap``: Queue statistics mapping.
202
203* ``dcb_tc``: DCB information such as TC mapping.
204
205* ``cap``: Supported offload capabilities.
206
207For example:
208
209.. code-block:: console
210
211   testpmd> show port info 0
212
213   ********************* Infos for port 0 *********************
214
215   MAC address: XX:XX:XX:XX:XX:XX
216   Connect to socket: 0
217   memory allocation on the socket: 0
218   Link status: up
219   Link speed: 40000 Mbps
220   Link duplex: full-duplex
221   Promiscuous mode: enabled
222   Allmulticast mode: disabled
223   Maximum number of MAC addresses: 64
224   Maximum number of MAC addresses of hash filtering: 0
225   VLAN offload:
226       strip on
227       filter on
228       qinq(extend) off
229   Redirection table size: 512
230   Supported flow types:
231     ipv4-frag
232     ipv4-tcp
233     ipv4-udp
234     ipv4-sctp
235     ipv4-other
236     ipv6-frag
237     ipv6-tcp
238     ipv6-udp
239     ipv6-sctp
240     ipv6-other
241     l2_payload
242     port
243     vxlan
244     geneve
245     nvgre
246
247show port rss reta
248~~~~~~~~~~~~~~~~~~
249
250Display the rss redirection table entry indicated by masks on port X::
251
252   testpmd> show port (port_id) rss reta (size) (mask0, mask1...)
253
254size is used to indicate the hardware supported reta size
255
256show port rss-hash
257~~~~~~~~~~~~~~~~~~
258
259Display the RSS hash functions and RSS hash key of a port::
260
261   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]
262
263clear port
264~~~~~~~~~~
265
266Clear the port statistics for a given port or for all ports::
267
268   testpmd> clear port (info|stats|xstats|fdir|stat_qmap) (port_id|all)
269
270For example::
271
272   testpmd> clear port stats all
273
274show (rxq|txq)
275~~~~~~~~~~~~~~
276
277Display information for a given port's RX/TX queue::
278
279   testpmd> show (rxq|txq) info (port_id) (queue_id)
280
281show config
282~~~~~~~~~~~
283
284Displays the configuration of the application.
285The configuration comes from the command-line, the runtime or the application defaults::
286
287   testpmd> show config (rxtx|cores|fwd|txpkts)
288
289The available information categories are:
290
291* ``rxtx``: RX/TX configuration items.
292
293* ``cores``: List of forwarding cores.
294
295* ``fwd``: Packet forwarding configuration.
296
297* ``txpkts``: Packets to TX configuration.
298
299For example:
300
301.. code-block:: console
302
303   testpmd> show config rxtx
304
305   io packet forwarding - CRC stripping disabled - packets/burst=16
306   nb forwarding cores=2 - nb forwarding ports=1
307   RX queues=1 - RX desc=128 - RX free threshold=0
308   RX threshold registers: pthresh=8 hthresh=8 wthresh=4
309   TX queues=1 - TX desc=512 - TX free threshold=0
310   TX threshold registers: pthresh=36 hthresh=0 wthresh=0
311   TX RS bit threshold=0 - TXQ flags=0x0
312
313set fwd
314~~~~~~~
315
316Set the packet forwarding mode::
317
318   testpmd> set fwd (io|mac|macswap|flowgen| \
319                     rxonly|txonly|csum|icmpecho) (""|retry)
320
321``retry`` can be specified for forwarding engines except ``rx_only``.
322
323The available information categories are:
324
325* ``io``: Forwards packets "as-is" in I/O mode.
326  This is the fastest possible forwarding operation as it does not access packets data.
327  This is the default mode.
328
329* ``mac``: Changes the source and the destination Ethernet addresses of packets before forwarding them.
330  Default application behaviour is to set source Ethernet address to that of the transmitting interface, and destination
331  address to a dummy value (set during init). The user may specify a target destination Ethernet address via the 'eth-peer' or
332  'eth-peer-configfile' command-line options. It is not currently possible to specify a specific source Ethernet address.
333
334* ``macswap``: MAC swap forwarding mode.
335  Swaps the source and the destination Ethernet addresses of packets before forwarding them.
336
337* ``flowgen``: Multi-flow generation mode.
338  Originates a number of flows (with varying destination IP addresses), and terminate receive traffic.
339
340* ``rxonly``: Receives packets but doesn't transmit them.
341
342* ``txonly``: Generates and transmits packets without receiving any.
343
344* ``csum``: Changes the checksum field with hardware or software methods depending on the offload flags on the packet.
345
346* ``icmpecho``: Receives a burst of packets, lookup for IMCP echo requests and, if any, send back ICMP echo replies.
347
348* ``ieee1588``: Demonstrate L2 IEEE1588 V2 PTP timestamping for RX and TX. Requires ``CONFIG_RTE_LIBRTE_IEEE1588=y``.
349
350Note: TX timestamping is only available in the "Full Featured" TX path. To force ``testpmd`` into this mode set ``--txqflags=0``.
351
352Example::
353
354   testpmd> set fwd rxonly
355
356   Set rxonly packet forwarding mode
357
358
359read rxd
360~~~~~~~~
361
362Display an RX descriptor for a port RX queue::
363
364   testpmd> read rxd (port_id) (queue_id) (rxd_id)
365
366For example::
367
368   testpmd> read rxd 0 0 4
369        0x0000000B - 0x001D0180 / 0x0000000B - 0x001D0180
370
371read txd
372~~~~~~~~
373
374Display a TX descriptor for a port TX queue::
375
376   testpmd> read txd (port_id) (queue_id) (txd_id)
377
378For example::
379
380   testpmd> read txd 0 0 4
381        0x00000001 - 0x24C3C440 / 0x000F0000 - 0x2330003C
382
383ddp get list
384~~~~~~~~~~~~
385
386Get loaded dynamic device personalization (DDP) package info list::
387
388   testpmd> ddp get list (port_id)
389
390ddp get info
391~~~~~~~~~~~~
392
393Display information about dynamic device personalization (DDP) profile::
394
395   testpmd> ddp get info (profile_path)
396
397show vf stats
398~~~~~~~~~~~~~
399
400Display VF statistics::
401
402   testpmd> show vf stats (port_id) (vf_id)
403
404clear vf stats
405~~~~~~~~~~~~~~
406
407Reset VF statistics::
408
409   testpmd> clear vf stats (port_id) (vf_id)
410
411show port pctype mapping
412~~~~~~~~~~~~~~~~~~~~~~~~
413
414List all items from the pctype mapping table::
415
416   testpmd> show port (port_id) pctype mapping
417
418
419Configuration Functions
420-----------------------
421
422The testpmd application can be configured from the runtime as well as from the command-line.
423
424This section details the available configuration functions that are available.
425
426.. note::
427
428   Configuration changes only become active when forwarding is started/restarted.
429
430set default
431~~~~~~~~~~~
432
433Reset forwarding to the default configuration::
434
435   testpmd> set default
436
437set verbose
438~~~~~~~~~~~
439
440Set the debug verbosity level::
441
442   testpmd> set verbose (level)
443
444Currently the only available levels are 0 (silent except for error) and 1 (fully verbose).
445
446set nbport
447~~~~~~~~~~
448
449Set the number of ports used by the application:
450
451set nbport (num)
452
453This is equivalent to the ``--nb-ports`` command-line option.
454
455set nbcore
456~~~~~~~~~~
457
458Set the number of cores used by the application::
459
460   testpmd> set nbcore (num)
461
462This is equivalent to the ``--nb-cores`` command-line option.
463
464.. note::
465
466   The number of cores used must not be greater than number of ports used multiplied by the number of queues per port.
467
468set coremask
469~~~~~~~~~~~~
470
471Set the forwarding cores hexadecimal mask::
472
473   testpmd> set coremask (mask)
474
475This is equivalent to the ``--coremask`` command-line option.
476
477.. note::
478
479   The master lcore is reserved for command line parsing only and cannot be masked on for packet forwarding.
480
481set portmask
482~~~~~~~~~~~~
483
484Set the forwarding ports hexadecimal mask::
485
486   testpmd> set portmask (mask)
487
488This is equivalent to the ``--portmask`` command-line option.
489
490set burst
491~~~~~~~~~
492
493Set number of packets per burst::
494
495   testpmd> set burst (num)
496
497This is equivalent to the ``--burst command-line`` option.
498
499When retry is enabled, the transmit delay time and number of retries can also be set::
500
501   testpmd> set burst tx delay (microseconds) retry (num)
502
503set txpkts
504~~~~~~~~~~
505
506Set the length of each segment of the TX-ONLY packets or length of packet for FLOWGEN mode::
507
508   testpmd> set txpkts (x[,y]*)
509
510Where x[,y]* represents a CSV list of values, without white space.
511
512set txsplit
513~~~~~~~~~~~
514
515Set the split policy for the TX packets, applicable for TX-ONLY and CSUM forwarding modes::
516
517   testpmd> set txsplit (off|on|rand)
518
519Where:
520
521* ``off`` disable packet copy & split for CSUM mode.
522
523* ``on`` split outgoing packet into multiple segments. Size of each segment
524  and number of segments per packet is determined by ``set txpkts`` command
525  (see above).
526
527* ``rand`` same as 'on', but number of segments per each packet is a random value between 1 and total number of segments.
528
529set corelist
530~~~~~~~~~~~~
531
532Set the list of forwarding cores::
533
534   testpmd> set corelist (x[,y]*)
535
536For example, to change the forwarding cores:
537
538.. code-block:: console
539
540   testpmd> set corelist 3,1
541   testpmd> show config fwd
542
543   io packet forwarding - ports=2 - cores=2 - streams=2 - NUMA support disabled
544   Logical Core 3 (socket 0) forwards packets on 1 streams:
545   RX P=0/Q=0 (socket 0) -> TX P=1/Q=0 (socket 0) peer=02:00:00:00:00:01
546   Logical Core 1 (socket 0) forwards packets on 1 streams:
547   RX P=1/Q=0 (socket 0) -> TX P=0/Q=0 (socket 0) peer=02:00:00:00:00:00
548
549.. note::
550
551   The cores are used in the same order as specified on the command line.
552
553set portlist
554~~~~~~~~~~~~
555
556Set the list of forwarding ports::
557
558   testpmd> set portlist (x[,y]*)
559
560For example, to change the port forwarding:
561
562.. code-block:: console
563
564   testpmd> set portlist 0,2,1,3
565   testpmd> show config fwd
566
567   io packet forwarding - ports=4 - cores=1 - streams=4
568   Logical Core 3 (socket 0) forwards packets on 4 streams:
569   RX P=0/Q=0 (socket 0) -> TX P=2/Q=0 (socket 0) peer=02:00:00:00:00:01
570   RX P=2/Q=0 (socket 0) -> TX P=0/Q=0 (socket 0) peer=02:00:00:00:00:00
571   RX P=1/Q=0 (socket 0) -> TX P=3/Q=0 (socket 0) peer=02:00:00:00:00:03
572   RX P=3/Q=0 (socket 0) -> TX P=1/Q=0 (socket 0) peer=02:00:00:00:00:02
573
574set tx loopback
575~~~~~~~~~~~~~~~
576
577Enable/disable tx loopback::
578
579   testpmd> set tx loopback (port_id) (on|off)
580
581set drop enable
582~~~~~~~~~~~~~~~
583
584set drop enable bit for all queues::
585
586   testpmd> set all queues drop (port_id) (on|off)
587
588set split drop enable (for VF)
589~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
590
591set split drop enable bit for VF from PF::
592
593   testpmd> set vf split drop (port_id) (vf_id) (on|off)
594
595set mac antispoof (for VF)
596~~~~~~~~~~~~~~~~~~~~~~~~~~
597
598Set mac antispoof for a VF from the PF::
599
600   testpmd> set vf mac antispoof  (port_id) (vf_id) (on|off)
601
602set macsec offload
603~~~~~~~~~~~~~~~~~~
604
605Enable/disable MACsec offload::
606
607   testpmd> set macsec offload (port_id) on encrypt (on|off) replay-protect (on|off)
608   testpmd> set macsec offload (port_id) off
609
610set macsec sc
611~~~~~~~~~~~~~
612
613Configure MACsec secure connection (SC)::
614
615   testpmd> set macsec sc (tx|rx) (port_id) (mac) (pi)
616
617.. note::
618
619   The pi argument is ignored for tx.
620   Check the NIC Datasheet for hardware limits.
621
622set macsec sa
623~~~~~~~~~~~~~
624
625Configure MACsec secure association (SA)::
626
627   testpmd> set macsec sa (tx|rx) (port_id) (idx) (an) (pn) (key)
628
629.. note::
630
631   The IDX value must be 0 or 1.
632   Check the NIC Datasheet for hardware limits.
633
634set broadcast mode (for VF)
635~~~~~~~~~~~~~~~~~~~~~~~~~~~
636
637Set broadcast mode for a VF from the PF::
638
639   testpmd> set vf broadcast (port_id) (vf_id) (on|off)
640
641vlan set strip
642~~~~~~~~~~~~~~
643
644Set the VLAN strip on a port::
645
646   testpmd> vlan set strip (on|off) (port_id)
647
648vlan set stripq
649~~~~~~~~~~~~~~~
650
651Set the VLAN strip for a queue on a port::
652
653   testpmd> vlan set stripq (on|off) (port_id,queue_id)
654
655vlan set stripq (for VF)
656~~~~~~~~~~~~~~~~~~~~~~~~
657
658Set VLAN strip for all queues in a pool for a VF from the PF::
659
660   testpmd> set vf vlan stripq (port_id) (vf_id) (on|off)
661
662vlan set insert (for VF)
663~~~~~~~~~~~~~~~~~~~~~~~~
664
665Set VLAN insert for a VF from the PF::
666
667   testpmd> set vf vlan insert (port_id) (vf_id) (vlan_id)
668
669vlan set tag (for VF)
670~~~~~~~~~~~~~~~~~~~~~
671
672Set VLAN tag for a VF from the PF::
673
674   testpmd> set vf vlan tag (port_id) (vf_id) (on|off)
675
676vlan set antispoof (for VF)
677~~~~~~~~~~~~~~~~~~~~~~~~~~~
678
679Set VLAN antispoof for a VF from the PF::
680
681   testpmd> set vf vlan antispoof (port_id) (vf_id) (on|off)
682
683vlan set filter
684~~~~~~~~~~~~~~~
685
686Set the VLAN filter on a port::
687
688   testpmd> vlan set filter (on|off) (port_id)
689
690vlan set qinq
691~~~~~~~~~~~~~
692
693Set the VLAN QinQ (extended queue in queue) on for a port::
694
695   testpmd> vlan set qinq (on|off) (port_id)
696
697vlan set tpid
698~~~~~~~~~~~~~
699
700Set the inner or outer VLAN TPID for packet filtering on a port::
701
702   testpmd> vlan set (inner|outer) tpid (value) (port_id)
703
704.. note::
705
706   TPID value must be a 16-bit number (value <= 65536).
707
708rx_vlan add
709~~~~~~~~~~~
710
711Add a VLAN ID, or all identifiers, to the set of VLAN identifiers filtered by port ID::
712
713   testpmd> rx_vlan add (vlan_id|all) (port_id)
714
715.. note::
716
717   VLAN filter must be set on that port. VLAN ID < 4096.
718   Depending on the NIC used, number of vlan_ids may be limited to the maximum entries
719   in VFTA table. This is important if enabling all vlan_ids.
720
721rx_vlan rm
722~~~~~~~~~~
723
724Remove a VLAN ID, or all identifiers, from the set of VLAN identifiers filtered by port ID::
725
726   testpmd> rx_vlan rm (vlan_id|all) (port_id)
727
728rx_vlan add (for VF)
729~~~~~~~~~~~~~~~~~~~~
730
731Add a VLAN ID, to the set of VLAN identifiers filtered for VF(s) for port ID::
732
733   testpmd> rx_vlan add (vlan_id) port (port_id) vf (vf_mask)
734
735rx_vlan rm (for VF)
736~~~~~~~~~~~~~~~~~~~
737
738Remove a VLAN ID, from the set of VLAN identifiers filtered for VF(s) for port ID::
739
740   testpmd> rx_vlan rm (vlan_id) port (port_id) vf (vf_mask)
741
742tunnel_filter add
743~~~~~~~~~~~~~~~~~
744
745Add a tunnel filter on a port::
746
747   testpmd> tunnel_filter add (port_id) (outer_mac) (inner_mac) (ip_addr) \
748            (inner_vlan) (vxlan|nvgre|ipingre) (imac-ivlan|imac-ivlan-tenid|\
749            imac-tenid|imac|omac-imac-tenid|oip|iip) (tenant_id) (queue_id)
750
751The available information categories are:
752
753* ``vxlan``: Set tunnel type as VXLAN.
754
755* ``nvgre``: Set tunnel type as NVGRE.
756
757* ``ipingre``: Set tunnel type as IP-in-GRE.
758
759* ``imac-ivlan``: Set filter type as Inner MAC and VLAN.
760
761* ``imac-ivlan-tenid``: Set filter type as Inner MAC, VLAN and tenant ID.
762
763* ``imac-tenid``: Set filter type as Inner MAC and tenant ID.
764
765* ``imac``: Set filter type as Inner MAC.
766
767* ``omac-imac-tenid``: Set filter type as Outer MAC, Inner MAC and tenant ID.
768
769* ``oip``: Set filter type as Outer IP.
770
771* ``iip``: Set filter type as Inner IP.
772
773Example::
774
775   testpmd> tunnel_filter add 0 68:05:CA:28:09:82 00:00:00:00:00:00 \
776            192.168.2.2 0 ipingre oip 1 1
777
778   Set an IP-in-GRE tunnel on port 0, and the filter type is Outer IP.
779
780tunnel_filter remove
781~~~~~~~~~~~~~~~~~~~~
782
783Remove a tunnel filter on a port::
784
785   testpmd> tunnel_filter rm (port_id) (outer_mac) (inner_mac) (ip_addr) \
786            (inner_vlan) (vxlan|nvgre|ipingre) (imac-ivlan|imac-ivlan-tenid|\
787            imac-tenid|imac|omac-imac-tenid|oip|iip) (tenant_id) (queue_id)
788
789rx_vxlan_port add
790~~~~~~~~~~~~~~~~~
791
792Add an UDP port for VXLAN packet filter on a port::
793
794   testpmd> rx_vxlan_port add (udp_port) (port_id)
795
796rx_vxlan_port remove
797~~~~~~~~~~~~~~~~~~~~
798
799Remove an UDP port for VXLAN packet filter on a port::
800
801   testpmd> rx_vxlan_port rm (udp_port) (port_id)
802
803tx_vlan set
804~~~~~~~~~~~
805
806Set hardware insertion of VLAN IDs in packets sent on a port::
807
808   testpmd> tx_vlan set (port_id) vlan_id[, vlan_id_outer]
809
810For example, set a single VLAN ID (5) insertion on port 0::
811
812   tx_vlan set 0 5
813
814Or, set double VLAN ID (inner: 2, outer: 3) insertion on port 1::
815
816   tx_vlan set 1 2 3
817
818
819tx_vlan set pvid
820~~~~~~~~~~~~~~~~
821
822Set port based hardware insertion of VLAN ID in packets sent on a port::
823
824   testpmd> tx_vlan set pvid (port_id) (vlan_id) (on|off)
825
826tx_vlan reset
827~~~~~~~~~~~~~
828
829Disable hardware insertion of a VLAN header in packets sent on a port::
830
831   testpmd> tx_vlan reset (port_id)
832
833csum set
834~~~~~~~~
835
836Select hardware or software calculation of the checksum when
837transmitting a packet using the ``csum`` forwarding engine::
838
839   testpmd> csum set (ip|udp|tcp|sctp|outer-ip) (hw|sw) (port_id)
840
841Where:
842
843* ``ip|udp|tcp|sctp`` always relate to  the inner layer.
844
845* ``outer-ip`` relates to the outer IP layer (only for IPv4) in the case where the packet is recognized
846  as a tunnel packet by the forwarding engine (vxlan, gre and ipip are
847  supported). See also the ``csum parse-tunnel`` command.
848
849.. note::
850
851   Check the NIC Datasheet for hardware limits.
852
853csum parse-tunnel
854~~~~~~~~~~~~~~~~~
855
856Define how tunneled packets should be handled by the csum forward
857engine::
858
859   testpmd> csum parse-tunnel (on|off) (tx_port_id)
860
861If enabled, the csum forward engine will try to recognize supported
862tunnel headers (vxlan, gre, ipip).
863
864If disabled, treat tunnel packets as non-tunneled packets (a inner
865header is handled as a packet payload).
866
867.. note::
868
869   The port argument is the TX port like in the ``csum set`` command.
870
871Example:
872
873Consider a packet in packet like the following::
874
875   eth_out/ipv4_out/udp_out/vxlan/eth_in/ipv4_in/tcp_in
876
877* If parse-tunnel is enabled, the ``ip|udp|tcp|sctp`` parameters of ``csum set``
878  command relate to the inner headers (here ``ipv4_in`` and ``tcp_in``), and the
879  ``outer-ip parameter`` relates to the outer headers (here ``ipv4_out``).
880
881* If parse-tunnel is disabled, the ``ip|udp|tcp|sctp`` parameters of ``csum  set``
882   command relate to the outer headers, here ``ipv4_out`` and ``udp_out``.
883
884csum show
885~~~~~~~~~
886
887Display tx checksum offload configuration::
888
889   testpmd> csum show (port_id)
890
891tso set
892~~~~~~~
893
894Enable TCP Segmentation Offload (TSO) in the ``csum`` forwarding engine::
895
896   testpmd> tso set (segsize) (port_id)
897
898.. note::
899
900   Check the NIC datasheet for hardware limits.
901
902tso show
903~~~~~~~~
904
905Display the status of TCP Segmentation Offload::
906
907   testpmd> tso show (port_id)
908
909set port - gro
910~~~~~~~~~~~~~~
911
912Enable or disable GRO in ``csum`` forwarding engine::
913
914   testpmd> set port <port_id> gro on|off
915
916If enabled, the csum forwarding engine will perform GRO on the TCP/IPv4
917packets received from the given port.
918
919If disabled, packets received from the given port won't be performed
920GRO. By default, GRO is disabled for all ports.
921
922.. note::
923
924   When enable GRO for a port, TCP/IPv4 packets received from the port
925   will be performed GRO. After GRO, all merged packets have bad
926   checksums, since the GRO library doesn't re-calculate checksums for
927   the merged packets. Therefore, if users want the merged packets to
928   have correct checksums, please select HW IP checksum calculation and
929   HW TCP checksum calculation for the port which the merged packets are
930   transmitted to.
931
932show port - gro
933~~~~~~~~~~~~~~~
934
935Display GRO configuration for a given port::
936
937   testpmd> show port <port_id> gro
938
939set gro flush
940~~~~~~~~~~~~~
941
942Set the cycle to flush the GROed packets from reassembly tables::
943
944   testpmd> set gro flush <cycles>
945
946When enable GRO, the csum forwarding engine performs GRO on received
947packets, and the GROed packets are stored in reassembly tables. Users
948can use this command to determine when the GROed packets are flushed
949from the reassembly tables.
950
951The ``cycles`` is measured in GRO operation times. The csum forwarding
952engine flushes the GROed packets from the tables every ``cycles`` GRO
953operations.
954
955By default, the value of ``cycles`` is 1, which means flush GROed packets
956from the reassembly tables as soon as one GRO operation finishes. The value
957of ``cycles`` should be in the range of 1 to ``GRO_MAX_FLUSH_CYCLES``.
958
959Please note that the large value of ``cycles`` may cause the poor TCP/IP
960stack performance. Because the GROed packets are delayed to arrive the
961stack, thus causing more duplicated ACKs and TCP retransmissions.
962
963mac_addr add
964~~~~~~~~~~~~
965
966Add an alternative MAC address to a port::
967
968   testpmd> mac_addr add (port_id) (XX:XX:XX:XX:XX:XX)
969
970mac_addr remove
971~~~~~~~~~~~~~~~
972
973Remove a MAC address from a port::
974
975   testpmd> mac_addr remove (port_id) (XX:XX:XX:XX:XX:XX)
976
977mac_addr add (for VF)
978~~~~~~~~~~~~~~~~~~~~~
979
980Add an alternative MAC address for a VF to a port::
981
982   testpmd> mac_add add port (port_id) vf (vf_id) (XX:XX:XX:XX:XX:XX)
983
984mac_addr set
985~~~~~~~~~~~~
986
987Set the default MAC address for a port::
988
989   testpmd> mac_addr set (port_id) (XX:XX:XX:XX:XX:XX)
990
991mac_addr set (for VF)
992~~~~~~~~~~~~~~~~~~~~~
993
994Set the MAC address for a VF from the PF::
995
996   testpmd> set vf mac addr (port_id) (vf_id) (XX:XX:XX:XX:XX:XX)
997
998set port-uta
999~~~~~~~~~~~~
1000
1001Set the unicast hash filter(s) on/off for a port::
1002
1003   testpmd> set port (port_id) uta (XX:XX:XX:XX:XX:XX|all) (on|off)
1004
1005set promisc
1006~~~~~~~~~~~
1007
1008Set the promiscuous mode on for a port or for all ports.
1009In promiscuous mode packets are not dropped if they aren't for the specified MAC address::
1010
1011   testpmd> set promisc (port_id|all) (on|off)
1012
1013set allmulti
1014~~~~~~~~~~~~
1015
1016Set the allmulti mode for a port or for all ports::
1017
1018   testpmd> set allmulti (port_id|all) (on|off)
1019
1020Same as the ifconfig (8) option. Controls how multicast packets are handled.
1021
1022set promisc (for VF)
1023~~~~~~~~~~~~~~~~~~~~
1024
1025Set the unicast promiscuous mode for a VF from PF.
1026It's supported by Intel i40e NICs now.
1027In promiscuous mode packets are not dropped if they aren't for the specified MAC address::
1028
1029   testpmd> set vf promisc (port_id) (vf_id) (on|off)
1030
1031set allmulticast (for VF)
1032~~~~~~~~~~~~~~~~~~~~~~~~~
1033
1034Set the multicast promiscuous mode for a VF from PF.
1035It's supported by Intel i40e NICs now.
1036In promiscuous mode packets are not dropped if they aren't for the specified MAC address::
1037
1038   testpmd> set vf allmulti (port_id) (vf_id) (on|off)
1039
1040set tx max bandwidth (for VF)
1041~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1042
1043Set TX max absolute bandwidth (Mbps) for a VF from PF::
1044
1045   testpmd> set vf tx max-bandwidth (port_id) (vf_id) (max_bandwidth)
1046
1047set tc tx min bandwidth (for VF)
1048~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1049
1050Set all TCs' TX min relative bandwidth (%) for a VF from PF::
1051
1052   testpmd> set vf tc tx min-bandwidth (port_id) (vf_id) (bw1, bw2, ...)
1053
1054set tc tx max bandwidth (for VF)
1055~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1056
1057Set a TC's TX max absolute bandwidth (Mbps) for a VF from PF::
1058
1059   testpmd> set vf tc tx max-bandwidth (port_id) (vf_id) (tc_no) (max_bandwidth)
1060
1061set tc strict link priority mode
1062~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1063
1064Set some TCs' strict link priority mode on a physical port::
1065
1066   testpmd> set tx strict-link-priority (port_id) (tc_bitmap)
1067
1068set tc tx min bandwidth
1069~~~~~~~~~~~~~~~~~~~~~~~
1070
1071Set all TCs' TX min relative bandwidth (%) globally for all PF and VFs::
1072
1073   testpmd> set tc tx min-bandwidth (port_id) (bw1, bw2, ...)
1074
1075set flow_ctrl rx
1076~~~~~~~~~~~~~~~~
1077
1078Set the link flow control parameter on a port::
1079
1080   testpmd> set flow_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \
1081            (pause_time) (send_xon) mac_ctrl_frame_fwd (on|off) \
1082	    autoneg (on|off) (port_id)
1083
1084Where:
1085
1086* ``high_water`` (integer): High threshold value to trigger XOFF.
1087
1088* ``low_water`` (integer): Low threshold value to trigger XON.
1089
1090* ``pause_time`` (integer): Pause quota in the Pause frame.
1091
1092* ``send_xon`` (0/1): Send XON frame.
1093
1094* ``mac_ctrl_frame_fwd``: Enable receiving MAC control frames.
1095
1096* ``autoneg``: Change the auto-negotiation parameter.
1097
1098set pfc_ctrl rx
1099~~~~~~~~~~~~~~~
1100
1101Set the priority flow control parameter on a port::
1102
1103   testpmd> set pfc_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \
1104            (pause_time) (priority) (port_id)
1105
1106Where:
1107
1108* ``high_water`` (integer): High threshold value.
1109
1110* ``low_water`` (integer): Low threshold value.
1111
1112* ``pause_time`` (integer): Pause quota in the Pause frame.
1113
1114* ``priority`` (0-7): VLAN User Priority.
1115
1116set stat_qmap
1117~~~~~~~~~~~~~
1118
1119Set statistics mapping (qmapping 0..15) for RX/TX queue on port::
1120
1121   testpmd> set stat_qmap (tx|rx) (port_id) (queue_id) (qmapping)
1122
1123For example, to set rx queue 2 on port 0 to mapping 5::
1124
1125   testpmd>set stat_qmap rx 0 2 5
1126
1127set port - rx/tx (for VF)
1128~~~~~~~~~~~~~~~~~~~~~~~~~
1129
1130Set VF receive/transmit from a port::
1131
1132   testpmd> set port (port_id) vf (vf_id) (rx|tx) (on|off)
1133
1134set port - mac address filter (for VF)
1135~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1136
1137Add/Remove unicast or multicast MAC addr filter for a VF::
1138
1139   testpmd> set port (port_id) vf (vf_id) (mac_addr) \
1140            (exact-mac|exact-mac-vlan|hashmac|hashmac-vlan) (on|off)
1141
1142set port - rx mode(for VF)
1143~~~~~~~~~~~~~~~~~~~~~~~~~~
1144
1145Set the VF receive mode of a port::
1146
1147   testpmd> set port (port_id) vf (vf_id) \
1148            rxmode (AUPE|ROPE|BAM|MPE) (on|off)
1149
1150The available receive modes are:
1151
1152* ``AUPE``: Accepts untagged VLAN.
1153
1154* ``ROPE``: Accepts unicast hash.
1155
1156* ``BAM``: Accepts broadcast packets.
1157
1158* ``MPE``: Accepts all multicast packets.
1159
1160set port - tx_rate (for Queue)
1161~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1162
1163Set TX rate limitation for a queue on a port::
1164
1165   testpmd> set port (port_id) queue (queue_id) rate (rate_value)
1166
1167set port - tx_rate (for VF)
1168~~~~~~~~~~~~~~~~~~~~~~~~~~~
1169
1170Set TX rate limitation for queues in VF on a port::
1171
1172   testpmd> set port (port_id) vf (vf_id) rate (rate_value) queue_mask (queue_mask)
1173
1174set port - mirror rule
1175~~~~~~~~~~~~~~~~~~~~~~
1176
1177Set pool or vlan type mirror rule for a port::
1178
1179   testpmd> set port (port_id) mirror-rule (rule_id) \
1180            (pool-mirror-up|pool-mirror-down|vlan-mirror) \
1181            (poolmask|vlanid[,vlanid]*) dst-pool (pool_id) (on|off)
1182
1183Set link mirror rule for a port::
1184
1185   testpmd> set port (port_id) mirror-rule (rule_id) \
1186           (uplink-mirror|downlink-mirror) dst-pool (pool_id) (on|off)
1187
1188For example to enable mirror traffic with vlan 0,1 to pool 0::
1189
1190   set port 0 mirror-rule 0 vlan-mirror 0,1 dst-pool 0 on
1191
1192reset port - mirror rule
1193~~~~~~~~~~~~~~~~~~~~~~~~
1194
1195Reset a mirror rule for a port::
1196
1197   testpmd> reset port (port_id) mirror-rule (rule_id)
1198
1199set flush_rx
1200~~~~~~~~~~~~
1201
1202Set the flush on RX streams before forwarding.
1203The default is flush ``on``.
1204Mainly used with PCAP drivers to turn off the default behavior of flushing the first 512 packets on RX streams::
1205
1206   testpmd> set flush_rx off
1207
1208set bypass mode
1209~~~~~~~~~~~~~~~
1210
1211Set the bypass mode for the lowest port on bypass enabled NIC::
1212
1213   testpmd> set bypass mode (normal|bypass|isolate) (port_id)
1214
1215set bypass event
1216~~~~~~~~~~~~~~~~
1217
1218Set the event required to initiate specified bypass mode for the lowest port on a bypass enabled::
1219
1220   testpmd> set bypass event (timeout|os_on|os_off|power_on|power_off) \
1221            mode (normal|bypass|isolate) (port_id)
1222
1223Where:
1224
1225* ``timeout``: Enable bypass after watchdog timeout.
1226
1227* ``os_on``: Enable bypass when OS/board is powered on.
1228
1229* ``os_off``: Enable bypass when OS/board is powered off.
1230
1231* ``power_on``: Enable bypass when power supply is turned on.
1232
1233* ``power_off``: Enable bypass when power supply is turned off.
1234
1235
1236set bypass timeout
1237~~~~~~~~~~~~~~~~~~
1238
1239Set the bypass watchdog timeout to ``n`` seconds where 0 = instant::
1240
1241   testpmd> set bypass timeout (0|1.5|2|3|4|8|16|32)
1242
1243show bypass config
1244~~~~~~~~~~~~~~~~~~
1245
1246Show the bypass configuration for a bypass enabled NIC using the lowest port on the NIC::
1247
1248   testpmd> show bypass config (port_id)
1249
1250set link up
1251~~~~~~~~~~~
1252
1253Set link up for a port::
1254
1255   testpmd> set link-up port (port id)
1256
1257set link down
1258~~~~~~~~~~~~~
1259
1260Set link down for a port::
1261
1262   testpmd> set link-down port (port id)
1263
1264E-tag set
1265~~~~~~~~~
1266
1267Enable E-tag insertion for a VF on a port::
1268
1269   testpmd> E-tag set insertion on port-tag-id (value) port (port_id) vf (vf_id)
1270
1271Disable E-tag insertion for a VF on a port::
1272
1273   testpmd> E-tag set insertion off port (port_id) vf (vf_id)
1274
1275Enable/disable E-tag stripping on a port::
1276
1277   testpmd> E-tag set stripping (on|off) port (port_id)
1278
1279Enable/disable E-tag based forwarding on a port::
1280
1281   testpmd> E-tag set forwarding (on|off) port (port_id)
1282
1283Add an E-tag forwarding filter on a port::
1284
1285   testpmd> E-tag set filter add e-tag-id (value) dst-pool (pool_id) port (port_id)
1286
1287Delete an E-tag forwarding filter on a port::
1288   testpmd> E-tag set filter del e-tag-id (value) port (port_id)
1289
1290ddp add
1291~~~~~~~
1292
1293Load a dynamic device personalization (DDP) package::
1294
1295   testpmd> ddp add (port_id) (package_path[,output_path])
1296
1297ddp del
1298~~~~~~~
1299
1300Delete a dynamic device personalization package::
1301
1302   testpmd> ddp del (port_id) (package_path)
1303
1304ptype mapping
1305~~~~~~~~~~~~~
1306
1307List all items from the ptype mapping table::
1308
1309   testpmd> ptype mapping get (port_id) (valid_only)
1310
1311Where:
1312
1313* ``valid_only``: A flag indicates if only list valid items(=1) or all itemss(=0).
1314
1315Replace a specific or a group of software defined ptype with a new one::
1316
1317   testpmd> ptype mapping replace  (port_id) (target) (mask) (pkt_type)
1318
1319where:
1320
1321* ``target``: A specific software ptype or a mask to represent a group of software ptypes.
1322
1323* ``mask``: A flag indicate if "target" is a specific software ptype(=0) or a ptype mask(=1).
1324
1325* ``pkt_type``: The new software ptype to replace the old ones.
1326
1327Update hardware defined ptype to software defined packet type mapping table::
1328
1329   testpmd> ptype mapping update (port_id) (hw_ptype) (sw_ptype)
1330
1331where:
1332
1333* ``hw_ptype``: hardware ptype as the index of the ptype mapping table.
1334
1335* ``sw_ptype``: software ptype as the value of the ptype mapping table.
1336
1337Reset ptype mapping table::
1338
1339   testpmd> ptype mapping reset (port_id)
1340
1341Port Functions
1342--------------
1343
1344The following sections show functions for configuring ports.
1345
1346.. note::
1347
1348   Port configuration changes only become active when forwarding is started/restarted.
1349
1350port attach
1351~~~~~~~~~~~
1352
1353Attach a port specified by pci address or virtual device args::
1354
1355   testpmd> port attach (identifier)
1356
1357To attach a new pci device, the device should be recognized by kernel first.
1358Then it should be moved under DPDK management.
1359Finally the port can be attached to testpmd.
1360
1361For example, to move a pci device using ixgbe under DPDK management:
1362
1363.. code-block:: console
1364
1365   # Check the status of the available devices.
1366   ./usertools/dpdk-devbind.py --status
1367
1368   Network devices using DPDK-compatible driver
1369   ============================================
1370   <none>
1371
1372   Network devices using kernel driver
1373   ===================================
1374   0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused=
1375
1376
1377   # Bind the device to igb_uio.
1378   sudo ./usertools/dpdk-devbind.py -b igb_uio 0000:0a:00.0
1379
1380
1381   # Recheck the status of the devices.
1382   ./usertools/dpdk-devbind.py --status
1383   Network devices using DPDK-compatible driver
1384   ============================================
1385   0000:0a:00.0 '82599ES 10-Gigabit' drv=igb_uio unused=
1386
1387To attach a port created by virtual device, above steps are not needed.
1388
1389For example, to attach a port whose pci address is 0000:0a:00.0.
1390
1391.. code-block:: console
1392
1393   testpmd> port attach 0000:0a:00.0
1394   Attaching a new port...
1395   EAL: PCI device 0000:0a:00.0 on NUMA socket -1
1396   EAL:   probe driver: 8086:10fb rte_ixgbe_pmd
1397   EAL:   PCI memory mapped at 0x7f83bfa00000
1398   EAL:   PCI memory mapped at 0x7f83bfa80000
1399   PMD: eth_ixgbe_dev_init(): MAC: 2, PHY: 18, SFP+: 5
1400   PMD: eth_ixgbe_dev_init(): port 0 vendorID=0x8086 deviceID=0x10fb
1401   Port 0 is attached. Now total ports is 1
1402   Done
1403
1404For example, to attach a port created by pcap PMD.
1405
1406.. code-block:: console
1407
1408   testpmd> port attach net_pcap0
1409   Attaching a new port...
1410   PMD: Initializing pmd_pcap for net_pcap0
1411   PMD: Creating pcap-backed ethdev on numa socket 0
1412   Port 0 is attached. Now total ports is 1
1413   Done
1414
1415In this case, identifier is ``net_pcap0``.
1416This identifier format is the same as ``--vdev`` format of DPDK applications.
1417
1418For example, to re-attach a bonded port which has been previously detached,
1419the mode and slave parameters must be given.
1420
1421.. code-block:: console
1422
1423   testpmd> port attach net_bond_0,mode=0,slave=1
1424   Attaching a new port...
1425   EAL: Initializing pmd_bond for net_bond_0
1426   EAL: Create bonded device net_bond_0 on port 0 in mode 0 on socket 0.
1427   Port 0 is attached. Now total ports is 1
1428   Done
1429
1430
1431port detach
1432~~~~~~~~~~~
1433
1434Detach a specific port::
1435
1436   testpmd> port detach (port_id)
1437
1438Before detaching a port, the port should be stopped and closed.
1439
1440For example, to detach a pci device port 0.
1441
1442.. code-block:: console
1443
1444   testpmd> port stop 0
1445   Stopping ports...
1446   Done
1447   testpmd> port close 0
1448   Closing ports...
1449   Done
1450
1451   testpmd> port detach 0
1452   Detaching a port...
1453   EAL: PCI device 0000:0a:00.0 on NUMA socket -1
1454   EAL:   remove driver: 8086:10fb rte_ixgbe_pmd
1455   EAL:   PCI memory unmapped at 0x7f83bfa00000
1456   EAL:   PCI memory unmapped at 0x7f83bfa80000
1457   Done
1458
1459
1460For example, to detach a virtual device port 0.
1461
1462.. code-block:: console
1463
1464   testpmd> port stop 0
1465   Stopping ports...
1466   Done
1467   testpmd> port close 0
1468   Closing ports...
1469   Done
1470
1471   testpmd> port detach 0
1472   Detaching a port...
1473   PMD: Closing pcap ethdev on numa socket 0
1474   Port 'net_pcap0' is detached. Now total ports is 0
1475   Done
1476
1477To remove a pci device completely from the system, first detach the port from testpmd.
1478Then the device should be moved under kernel management.
1479Finally the device can be removed using kernel pci hotplug functionality.
1480
1481For example, to move a pci device under kernel management:
1482
1483.. code-block:: console
1484
1485   sudo ./usertools/dpdk-devbind.py -b ixgbe 0000:0a:00.0
1486
1487   ./usertools/dpdk-devbind.py --status
1488
1489   Network devices using DPDK-compatible driver
1490   ============================================
1491   <none>
1492
1493   Network devices using kernel driver
1494   ===================================
1495   0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused=igb_uio
1496
1497To remove a port created by a virtual device, above steps are not needed.
1498
1499port start
1500~~~~~~~~~~
1501
1502Start all ports or a specific port::
1503
1504   testpmd> port start (port_id|all)
1505
1506port stop
1507~~~~~~~~~
1508
1509Stop all ports or a specific port::
1510
1511   testpmd> port stop (port_id|all)
1512
1513port close
1514~~~~~~~~~~
1515
1516Close all ports or a specific port::
1517
1518   testpmd> port close (port_id|all)
1519
1520port start/stop queue
1521~~~~~~~~~~~~~~~~~~~~~
1522
1523Start/stop a rx/tx queue on a specific port::
1524
1525   testpmd> port (port_id) (rxq|txq) (queue_id) (start|stop)
1526
1527Only take effect when port is started.
1528
1529port config - speed
1530~~~~~~~~~~~~~~~~~~~
1531
1532Set the speed and duplex mode for all ports or a specific port::
1533
1534   testpmd> port config (port_id|all) speed (10|100|1000|10000|25000|40000|50000|100000|auto) \
1535            duplex (half|full|auto)
1536
1537port config - queues/descriptors
1538~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1539
1540Set number of queues/descriptors for rxq, txq, rxd and txd::
1541
1542   testpmd> port config all (rxq|txq|rxd|txd) (value)
1543
1544This is equivalent to the ``--rxq``, ``--txq``, ``--rxd`` and ``--txd`` command-line options.
1545
1546port config - max-pkt-len
1547~~~~~~~~~~~~~~~~~~~~~~~~~
1548
1549Set the maximum packet length::
1550
1551   testpmd> port config all max-pkt-len (value)
1552
1553This is equivalent to the ``--max-pkt-len`` command-line option.
1554
1555port config - CRC Strip
1556~~~~~~~~~~~~~~~~~~~~~~~
1557
1558Set hardware CRC stripping on or off for all ports::
1559
1560   testpmd> port config all crc-strip (on|off)
1561
1562CRC stripping is on by default.
1563
1564The ``off`` option is equivalent to the ``--disable-crc-strip`` command-line option.
1565
1566port config - scatter
1567~~~~~~~~~~~~~~~~~~~~~~~
1568
1569Set RX scatter mode on or off for all ports::
1570
1571   testpmd> port config all scatter (on|off)
1572
1573RX scatter mode is off by default.
1574
1575The ``on`` option is equivalent to the ``--enable-scatter`` command-line option.
1576
1577port config - TX queue flags
1578~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1579
1580Set a hexadecimal bitmap of TX queue flags for all ports::
1581
1582   testpmd> port config all txqflags value
1583
1584This command is equivalent to the ``--txqflags`` command-line option.
1585
1586port config - RX Checksum
1587~~~~~~~~~~~~~~~~~~~~~~~~~
1588
1589Set hardware RX checksum offload to on or off for all ports::
1590
1591   testpmd> port config all rx-cksum (on|off)
1592
1593Checksum offload is off by default.
1594
1595The ``on`` option is equivalent to the ``--enable-rx-cksum`` command-line option.
1596
1597port config - VLAN
1598~~~~~~~~~~~~~~~~~~
1599
1600Set hardware VLAN on or off for all ports::
1601
1602   testpmd> port config all hw-vlan (on|off)
1603
1604Hardware VLAN is on by default.
1605
1606The ``off`` option is equivalent to the ``--disable-hw-vlan`` command-line option.
1607
1608port config - VLAN filter
1609~~~~~~~~~~~~~~~~~~~~~~~~~
1610
1611Set hardware VLAN filter on or off for all ports::
1612
1613   testpmd> port config all hw-vlan-filter (on|off)
1614
1615Hardware VLAN filter is on by default.
1616
1617The ``off`` option is equivalent to the ``--disable-hw-vlan-filter`` command-line option.
1618
1619port config - VLAN strip
1620~~~~~~~~~~~~~~~~~~~~~~~~
1621
1622Set hardware VLAN strip on or off for all ports::
1623
1624   testpmd> port config all hw-vlan-strip (on|off)
1625
1626Hardware VLAN strip is on by default.
1627
1628The ``off`` option is equivalent to the ``--disable-hw-vlan-strip`` command-line option.
1629
1630port config - VLAN extend
1631~~~~~~~~~~~~~~~~~~~~~~~~~
1632
1633Set hardware VLAN extend on or off for all ports::
1634
1635   testpmd> port config all hw-vlan-extend (on|off)
1636
1637Hardware VLAN extend is off by default.
1638
1639The ``off`` option is equivalent to the ``--disable-hw-vlan-extend`` command-line option.
1640
1641port config - Drop Packets
1642~~~~~~~~~~~~~~~~~~~~~~~~~~
1643
1644Set packet drop for packets with no descriptors on or off for all ports::
1645
1646   testpmd> port config all drop-en (on|off)
1647
1648Packet dropping for packets with no descriptors is off by default.
1649
1650The ``on`` option is equivalent to the ``--enable-drop-en`` command-line option.
1651
1652port config - RSS
1653~~~~~~~~~~~~~~~~~
1654
1655Set the RSS (Receive Side Scaling) mode on or off::
1656
1657   testpmd> port config all rss (all|ip|tcp|udp|sctp|ether|port|vxlan|geneve|nvgre|none)
1658
1659RSS is on by default.
1660
1661The ``none`` option is equivalent to the ``--disable-rss`` command-line option.
1662
1663port config - RSS Reta
1664~~~~~~~~~~~~~~~~~~~~~~
1665
1666Set the RSS (Receive Side Scaling) redirection table::
1667
1668   testpmd> port config all rss reta (hash,queue)[,(hash,queue)]
1669
1670port config - DCB
1671~~~~~~~~~~~~~~~~~
1672
1673Set the DCB mode for an individual port::
1674
1675   testpmd> port config (port_id) dcb vt (on|off) (traffic_class) pfc (on|off)
1676
1677The traffic class should be 4 or 8.
1678
1679port config - Burst
1680~~~~~~~~~~~~~~~~~~~
1681
1682Set the number of packets per burst::
1683
1684   testpmd> port config all burst (value)
1685
1686This is equivalent to the ``--burst`` command-line option.
1687
1688port config - Threshold
1689~~~~~~~~~~~~~~~~~~~~~~~
1690
1691Set thresholds for TX/RX queues::
1692
1693   testpmd> port config all (threshold) (value)
1694
1695Where the threshold type can be:
1696
1697* ``txpt:`` Set the prefetch threshold register of the TX rings, 0 <= value <= 255.
1698
1699* ``txht:`` Set the host threshold register of the TX rings, 0 <= value <= 255.
1700
1701* ``txwt:`` Set the write-back threshold register of the TX rings, 0 <= value <= 255.
1702
1703* ``rxpt:`` Set the prefetch threshold register of the RX rings, 0 <= value <= 255.
1704
1705* ``rxht:`` Set the host threshold register of the RX rings, 0 <= value <= 255.
1706
1707* ``rxwt:`` Set the write-back threshold register of the RX rings, 0 <= value <= 255.
1708
1709* ``txfreet:`` Set the transmit free threshold of the TX rings, 0 <= value <= txd.
1710
1711* ``rxfreet:`` Set the transmit free threshold of the RX rings, 0 <= value <= rxd.
1712
1713* ``txrst:`` Set the transmit RS bit threshold of TX rings, 0 <= value <= txd.
1714
1715These threshold options are also available from the command-line.
1716
1717port config - E-tag
1718~~~~~~~~~~~~~~~~~~~
1719
1720Set the value of ether-type for E-tag::
1721
1722   testpmd> port config (port_id|all) l2-tunnel E-tag ether-type (value)
1723
1724Enable/disable the E-tag support::
1725
1726   testpmd> port config (port_id|all) l2-tunnel E-tag (enable|disable)
1727
1728port config pctype mapping
1729~~~~~~~~~~~~~~~~~~~~~~~~~~
1730
1731Reset pctype mapping table::
1732
1733   testpmd> port config (port_id) pctype mapping reset
1734
1735Update hardware defined pctype to software defined flow type mapping table::
1736
1737   testpmd> port config (port_id) pctype mapping update (pctype_id_0[,pctype_id_1]*) (flow_type_id)
1738
1739where:
1740
1741* ``pctype_id_x``: hardware pctype id as index of bit in bitmask value of the pctype mapping table.
1742
1743* ``flow_type_id``: software flow type id as the index of the pctype mapping table.
1744
1745
1746Link Bonding Functions
1747----------------------
1748
1749The Link Bonding functions make it possible to dynamically create and
1750manage link bonding devices from within testpmd interactive prompt.
1751
1752create bonded device
1753~~~~~~~~~~~~~~~~~~~~
1754
1755Create a new bonding device::
1756
1757   testpmd> create bonded device (mode) (socket)
1758
1759For example, to create a bonded device in mode 1 on socket 0::
1760
1761   testpmd> create bonded 1 0
1762   created new bonded device (port X)
1763
1764add bonding slave
1765~~~~~~~~~~~~~~~~~
1766
1767Adds Ethernet device to a Link Bonding device::
1768
1769   testpmd> add bonding slave (slave id) (port id)
1770
1771For example, to add Ethernet device (port 6) to a Link Bonding device (port 10)::
1772
1773   testpmd> add bonding slave 6 10
1774
1775
1776remove bonding slave
1777~~~~~~~~~~~~~~~~~~~~
1778
1779Removes an Ethernet slave device from a Link Bonding device::
1780
1781   testpmd> remove bonding slave (slave id) (port id)
1782
1783For example, to remove Ethernet slave device (port 6) to a Link Bonding device (port 10)::
1784
1785   testpmd> remove bonding slave 6 10
1786
1787set bonding mode
1788~~~~~~~~~~~~~~~~
1789
1790Set the Link Bonding mode of a Link Bonding device::
1791
1792   testpmd> set bonding mode (value) (port id)
1793
1794For example, to set the bonding mode of a Link Bonding device (port 10) to broadcast (mode 3)::
1795
1796   testpmd> set bonding mode 3 10
1797
1798set bonding primary
1799~~~~~~~~~~~~~~~~~~~
1800
1801Set an Ethernet slave device as the primary device on a Link Bonding device::
1802
1803   testpmd> set bonding primary (slave id) (port id)
1804
1805For example, to set the Ethernet slave device (port 6) as the primary port of a Link Bonding device (port 10)::
1806
1807   testpmd> set bonding primary 6 10
1808
1809set bonding mac
1810~~~~~~~~~~~~~~~
1811
1812Set the MAC address of a Link Bonding device::
1813
1814   testpmd> set bonding mac (port id) (mac)
1815
1816For example, to set the MAC address of a Link Bonding device (port 10) to 00:00:00:00:00:01::
1817
1818   testpmd> set bonding mac 10 00:00:00:00:00:01
1819
1820set bonding xmit_balance_policy
1821~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1822
1823Set the transmission policy for a Link Bonding device when it is in Balance XOR mode::
1824
1825   testpmd> set bonding xmit_balance_policy (port_id) (l2|l23|l34)
1826
1827For example, set a Link Bonding device (port 10) to use a balance policy of layer 3+4 (IP addresses & UDP ports)::
1828
1829   testpmd> set bonding xmit_balance_policy 10 l34
1830
1831
1832set bonding mon_period
1833~~~~~~~~~~~~~~~~~~~~~~
1834
1835Set the link status monitoring polling period in milliseconds for a bonding device.
1836
1837This adds support for PMD slave devices which do not support link status interrupts.
1838When the mon_period is set to a value greater than 0 then all PMD's which do not support
1839link status ISR will be queried every polling interval to check if their link status has changed::
1840
1841   testpmd> set bonding mon_period (port_id) (value)
1842
1843For example, to set the link status monitoring polling period of bonded device (port 5) to 150ms::
1844
1845   testpmd> set bonding mon_period 5 150
1846
1847
1848set bonding lacp dedicated_queue
1849~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1850
1851Enable dedicated tx/rx queues on bonding devices slaves to handle LACP control plane traffic
1852when in mode 4 (link-aggregration-802.3ad)::
1853
1854   testpmd> set bonding lacp dedicated_queues (port_id) (enable|disable)
1855
1856
1857set bonding agg_mode
1858~~~~~~~~~~~~~~~~~~~~
1859
1860Enable one of the specific aggregators mode when in mode 4 (link-aggregration-802.3ad)::
1861
1862   testpmd> set bonding agg_mode (port_id) (bandwidth|count|stable)
1863
1864
1865show bonding config
1866~~~~~~~~~~~~~~~~~~~
1867
1868Show the current configuration of a Link Bonding device::
1869
1870   testpmd> show bonding config (port id)
1871
1872For example,
1873to show the configuration a Link Bonding device (port 9) with 3 slave devices (1, 3, 4)
1874in balance mode with a transmission policy of layer 2+3::
1875
1876   testpmd> show bonding config 9
1877        Bonding mode: 2
1878        Balance Xmit Policy: BALANCE_XMIT_POLICY_LAYER23
1879        Slaves (3): [1 3 4]
1880        Active Slaves (3): [1 3 4]
1881        Primary: [3]
1882
1883
1884Register Functions
1885------------------
1886
1887The Register Functions can be used to read from and write to registers on the network card referenced by a port number.
1888This is mainly useful for debugging purposes.
1889Reference should be made to the appropriate datasheet for the network card for details on the register addresses
1890and fields that can be accessed.
1891
1892read reg
1893~~~~~~~~
1894
1895Display the value of a port register::
1896
1897   testpmd> read reg (port_id) (address)
1898
1899For example, to examine the Flow Director control register (FDIRCTL, 0x0000EE000) on an Intel 82599 10 GbE Controller::
1900
1901   testpmd> read reg 0 0xEE00
1902   port 0 PCI register at offset 0xEE00: 0x4A060029 (1241907241)
1903
1904read regfield
1905~~~~~~~~~~~~~
1906
1907Display a port register bit field::
1908
1909   testpmd> read regfield (port_id) (address) (bit_x) (bit_y)
1910
1911For example, reading the lowest two bits from the register in the example above::
1912
1913   testpmd> read regfield 0 0xEE00 0 1
1914   port 0 PCI register at offset 0xEE00: bits[0, 1]=0x1 (1)
1915
1916read regbit
1917~~~~~~~~~~~
1918
1919Display a single port register bit::
1920
1921   testpmd> read regbit (port_id) (address) (bit_x)
1922
1923For example, reading the lowest bit from the register in the example above::
1924
1925   testpmd> read regbit 0 0xEE00 0
1926   port 0 PCI register at offset 0xEE00: bit 0=1
1927
1928write reg
1929~~~~~~~~~
1930
1931Set the value of a port register::
1932
1933   testpmd> write reg (port_id) (address) (value)
1934
1935For example, to clear a register::
1936
1937   testpmd> write reg 0 0xEE00 0x0
1938   port 0 PCI register at offset 0xEE00: 0x00000000 (0)
1939
1940write regfield
1941~~~~~~~~~~~~~~
1942
1943Set bit field of a port register::
1944
1945   testpmd> write regfield (port_id) (address) (bit_x) (bit_y) (value)
1946
1947For example, writing to the register cleared in the example above::
1948
1949   testpmd> write regfield 0 0xEE00 0 1 2
1950   port 0 PCI register at offset 0xEE00: 0x00000002 (2)
1951
1952write regbit
1953~~~~~~~~~~~~
1954
1955Set single bit value of a port register::
1956
1957   testpmd> write regbit (port_id) (address) (bit_x) (value)
1958
1959For example, to set the high bit in the register from the example above::
1960
1961   testpmd> write regbit 0 0xEE00 31 1
1962   port 0 PCI register at offset 0xEE00: 0x8000000A (2147483658)
1963
1964
1965Filter Functions
1966----------------
1967
1968This section details the available filter functions that are available.
1969
1970Note these functions interface the deprecated legacy filtering framework,
1971superseded by *rte_flow*. See `Flow rules management`_.
1972
1973ethertype_filter
1974~~~~~~~~~~~~~~~~~~~~
1975
1976Add or delete a L2 Ethertype filter, which identify packets by their L2 Ethertype mainly assign them to a receive queue::
1977
1978   ethertype_filter (port_id) (add|del) (mac_addr|mac_ignr) (mac_address) \
1979                    ethertype (ether_type) (drop|fwd) queue (queue_id)
1980
1981The available information parameters are:
1982
1983* ``port_id``: The port which the Ethertype filter assigned on.
1984
1985* ``mac_addr``: Compare destination mac address.
1986
1987* ``mac_ignr``: Ignore destination mac address match.
1988
1989* ``mac_address``: Destination mac address to match.
1990
1991* ``ether_type``: The EtherType value want to match,
1992  for example 0x0806 for ARP packet. 0x0800 (IPv4) and 0x86DD (IPv6) are invalid.
1993
1994* ``queue_id``: The receive queue associated with this EtherType filter.
1995  It is meaningless when deleting or dropping.
1996
1997Example, to add/remove an ethertype filter rule::
1998
1999   testpmd> ethertype_filter 0 add mac_ignr 00:11:22:33:44:55 \
2000                             ethertype 0x0806 fwd queue 3
2001
2002   testpmd> ethertype_filter 0 del mac_ignr 00:11:22:33:44:55 \
2003                             ethertype 0x0806 fwd queue 3
2004
20052tuple_filter
2006~~~~~~~~~~~~~~~~~
2007
2008Add or delete a 2-tuple filter,
2009which identifies packets by specific protocol and destination TCP/UDP port
2010and forwards packets into one of the receive queues::
2011
2012   2tuple_filter (port_id) (add|del) dst_port (dst_port_value) \
2013                 protocol (protocol_value) mask (mask_value) \
2014                 tcp_flags (tcp_flags_value) priority (prio_value) \
2015                 queue (queue_id)
2016
2017The available information parameters are:
2018
2019* ``port_id``: The port which the 2-tuple filter assigned on.
2020
2021* ``dst_port_value``: Destination port in L4.
2022
2023* ``protocol_value``: IP L4 protocol.
2024
2025* ``mask_value``: Participates in the match or not by bit for field above, 1b means participate.
2026
2027* ``tcp_flags_value``: TCP control bits. The non-zero value is invalid, when the pro_value is not set to 0x06 (TCP).
2028
2029* ``prio_value``: Priority of this filter.
2030
2031* ``queue_id``: The receive queue associated with this 2-tuple filter.
2032
2033Example, to add/remove an 2tuple filter rule::
2034
2035   testpmd> 2tuple_filter 0 add dst_port 32 protocol 0x06 mask 0x03 \
2036                          tcp_flags 0x02 priority 3 queue 3
2037
2038   testpmd> 2tuple_filter 0 del dst_port 32 protocol 0x06 mask 0x03 \
2039                          tcp_flags 0x02 priority 3 queue 3
2040
20415tuple_filter
2042~~~~~~~~~~~~~~~~~
2043
2044Add or delete a 5-tuple filter,
2045which consists of a 5-tuple (protocol, source and destination IP addresses, source and destination TCP/UDP/SCTP port)
2046and routes packets into one of the receive queues::
2047
2048   5tuple_filter (port_id) (add|del) dst_ip (dst_address) src_ip \
2049                 (src_address) dst_port (dst_port_value) \
2050                 src_port (src_port_value) protocol (protocol_value) \
2051                 mask (mask_value) tcp_flags (tcp_flags_value) \
2052                 priority (prio_value) queue (queue_id)
2053
2054The available information parameters are:
2055
2056* ``port_id``: The port which the 5-tuple filter assigned on.
2057
2058* ``dst_address``: Destination IP address.
2059
2060* ``src_address``: Source IP address.
2061
2062* ``dst_port_value``: TCP/UDP destination port.
2063
2064* ``src_port_value``: TCP/UDP source port.
2065
2066* ``protocol_value``: L4 protocol.
2067
2068* ``mask_value``: Participates in the match or not by bit for field above, 1b means participate
2069
2070* ``tcp_flags_value``: TCP control bits. The non-zero value is invalid, when the protocol_value is not set to 0x06 (TCP).
2071
2072* ``prio_value``: The priority of this filter.
2073
2074* ``queue_id``: The receive queue associated with this 5-tuple filter.
2075
2076Example, to add/remove an 5tuple filter rule::
2077
2078   testpmd> 5tuple_filter 0 add dst_ip 2.2.2.5 src_ip 2.2.2.4 \
2079            dst_port 64 src_port 32 protocol 0x06 mask 0x1F \
2080            flags 0x0 priority 3 queue 3
2081
2082   testpmd> 5tuple_filter 0 del dst_ip 2.2.2.5 src_ip 2.2.2.4 \
2083            dst_port 64 src_port 32 protocol 0x06 mask 0x1F \
2084            flags 0x0 priority 3 queue 3
2085
2086syn_filter
2087~~~~~~~~~~
2088
2089Using the  SYN filter, TCP packets whose *SYN* flag is set can be forwarded to a separate queue::
2090
2091   syn_filter (port_id) (add|del) priority (high|low) queue (queue_id)
2092
2093The available information parameters are:
2094
2095* ``port_id``: The port which the SYN filter assigned on.
2096
2097* ``high``: This SYN filter has higher priority than other filters.
2098
2099* ``low``: This SYN filter has lower priority than other filters.
2100
2101* ``queue_id``: The receive queue associated with this SYN filter
2102
2103Example::
2104
2105   testpmd> syn_filter 0 add priority high queue 3
2106
2107flex_filter
2108~~~~~~~~~~~
2109
2110With flex filter, packets can be recognized by any arbitrary pattern within the first 128 bytes of the packet
2111and routed into one of the receive queues::
2112
2113   flex_filter (port_id) (add|del) len (len_value) bytes (bytes_value) \
2114               mask (mask_value) priority (prio_value) queue (queue_id)
2115
2116The available information parameters are:
2117
2118* ``port_id``: The port which the Flex filter is assigned on.
2119
2120* ``len_value``: Filter length in bytes, no greater than 128.
2121
2122* ``bytes_value``: A string in hexadecimal, means the value the flex filter needs to match.
2123
2124* ``mask_value``: A string in hexadecimal, bit 1 means corresponding byte participates in the match.
2125
2126* ``prio_value``: The priority of this filter.
2127
2128* ``queue_id``: The receive queue associated with this Flex filter.
2129
2130Example::
2131
2132   testpmd> flex_filter 0 add len 16 bytes 0x00000000000000000000000008060000 \
2133                          mask 000C priority 3 queue 3
2134
2135   testpmd> flex_filter 0 del len 16 bytes 0x00000000000000000000000008060000 \
2136                          mask 000C priority 3 queue 3
2137
2138
2139.. _testpmd_flow_director:
2140
2141flow_director_filter
2142~~~~~~~~~~~~~~~~~~~~
2143
2144The Flow Director works in receive mode to identify specific flows or sets of flows and route them to specific queues.
2145
2146Four types of filtering are supported which are referred to as Perfect Match, Signature, Perfect-mac-vlan and
2147Perfect-tunnel filters, the match mode is set by the ``--pkt-filter-mode`` command-line parameter:
2148
2149* Perfect match filters.
2150  The hardware checks a match between the masked fields of the received packets and the programmed filters.
2151  The masked fields are for IP flow.
2152
2153* Signature filters.
2154  The hardware checks a match between a hash-based signature of the masked fields of the received packet.
2155
2156* Perfect-mac-vlan match filters.
2157  The hardware checks a match between the masked fields of the received packets and the programmed filters.
2158  The masked fields are for MAC VLAN flow.
2159
2160* Perfect-tunnel match filters.
2161  The hardware checks a match between the masked fields of the received packets and the programmed filters.
2162  The masked fields are for tunnel flow.
2163
2164The Flow Director filters can match the different fields for different type of packet: flow type, specific input set
2165per flow type and the flexible payload.
2166
2167The Flow Director can also mask out parts of all of these fields so that filters
2168are only applied to certain fields or parts of the fields.
2169
2170Different NICs may have different capabilities, command show port fdir (port_id) can be used to acquire the information.
2171
2172# Commands to add flow director filters of different flow types::
2173
2174   flow_director_filter (port_id) mode IP (add|del|update) \
2175                        flow (ipv4-other|ipv4-frag|ipv6-other|ipv6-frag) \
2176                        src (src_ip_address) dst (dst_ip_address) \
2177                        tos (tos_value) proto (proto_value) ttl (ttl_value) \
2178                        vlan (vlan_value) flexbytes (flexbytes_value) \
2179                        (drop|fwd) pf|vf(vf_id) queue (queue_id) \
2180                        fd_id (fd_id_value)
2181
2182   flow_director_filter (port_id) mode IP (add|del|update) \
2183                        flow (ipv4-tcp|ipv4-udp|ipv6-tcp|ipv6-udp) \
2184                        src (src_ip_address) (src_port) \
2185                        dst (dst_ip_address) (dst_port) \
2186                        tos (tos_value) ttl (ttl_value) \
2187                        vlan (vlan_value) flexbytes (flexbytes_value) \
2188                        (drop|fwd) queue pf|vf(vf_id) (queue_id) \
2189                        fd_id (fd_id_value)
2190
2191   flow_director_filter (port_id) mode IP (add|del|update) \
2192                        flow (ipv4-sctp|ipv6-sctp) \
2193                        src (src_ip_address) (src_port) \
2194                        dst (dst_ip_address) (dst_port) \
2195                        tos (tos_value) ttl (ttl_value) \
2196                        tag (verification_tag) vlan (vlan_value) \
2197                        flexbytes (flexbytes_value) (drop|fwd) \
2198                        pf|vf(vf_id) queue (queue_id) fd_id (fd_id_value)
2199
2200   flow_director_filter (port_id) mode IP (add|del|update) flow l2_payload \
2201                        ether (ethertype) flexbytes (flexbytes_value) \
2202                        (drop|fwd) pf|vf(vf_id) queue (queue_id)
2203                        fd_id (fd_id_value)
2204
2205   flow_director_filter (port_id) mode MAC-VLAN (add|del|update) \
2206                        mac (mac_address) vlan (vlan_value) \
2207                        flexbytes (flexbytes_value) (drop|fwd) \
2208                        queue (queue_id) fd_id (fd_id_value)
2209
2210   flow_director_filter (port_id) mode Tunnel (add|del|update) \
2211                        mac (mac_address) vlan (vlan_value) \
2212                        tunnel (NVGRE|VxLAN) tunnel-id (tunnel_id_value) \
2213                        flexbytes (flexbytes_value) (drop|fwd) \
2214                        queue (queue_id) fd_id (fd_id_value)
2215
2216For example, to add an ipv4-udp flow type filter::
2217
2218   testpmd> flow_director_filter 0 mode IP add flow ipv4-udp src 2.2.2.3 32 \
2219            dst 2.2.2.5 33 tos 2 ttl 40 vlan 0x1 flexbytes (0x88,0x48) \
2220            fwd pf queue 1 fd_id 1
2221
2222For example, add an ipv4-other flow type filter::
2223
2224   testpmd> flow_director_filter 0 mode IP add flow ipv4-other src 2.2.2.3 \
2225             dst 2.2.2.5 tos 2 proto 20 ttl 40 vlan 0x1 \
2226             flexbytes (0x88,0x48) fwd pf queue 1 fd_id 1
2227
2228flush_flow_director
2229~~~~~~~~~~~~~~~~~~~
2230
2231Flush all flow director filters on a device::
2232
2233   testpmd> flush_flow_director (port_id)
2234
2235Example, to flush all flow director filter on port 0::
2236
2237   testpmd> flush_flow_director 0
2238
2239flow_director_mask
2240~~~~~~~~~~~~~~~~~~
2241
2242Set flow director's input masks::
2243
2244   flow_director_mask (port_id) mode IP vlan (vlan_value) \
2245                      src_mask (ipv4_src) (ipv6_src) (src_port) \
2246                      dst_mask (ipv4_dst) (ipv6_dst) (dst_port)
2247
2248   flow_director_mask (port_id) mode MAC-VLAN vlan (vlan_value)
2249
2250   flow_director_mask (port_id) mode Tunnel vlan (vlan_value) \
2251                      mac (mac_value) tunnel-type (tunnel_type_value) \
2252                      tunnel-id (tunnel_id_value)
2253
2254Example, to set flow director mask on port 0::
2255
2256   testpmd> flow_director_mask 0 mode IP vlan 0xefff \
2257            src_mask 255.255.255.255 \
2258                FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF \
2259            dst_mask 255.255.255.255 \
2260                FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF
2261
2262flow_director_flex_mask
2263~~~~~~~~~~~~~~~~~~~~~~~
2264
2265set masks of flow director's flexible payload based on certain flow type::
2266
2267   testpmd> flow_director_flex_mask (port_id) \
2268            flow (none|ipv4-other|ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
2269                  ipv6-other|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp| \
2270                  l2_payload|all) (mask)
2271
2272Example, to set flow director's flex mask for all flow type on port 0::
2273
2274   testpmd> flow_director_flex_mask 0 flow all \
2275            (0xff,0xff,0,0,0,0,0,0,0,0,0,0,0,0,0,0)
2276
2277
2278flow_director_flex_payload
2279~~~~~~~~~~~~~~~~~~~~~~~~~~
2280
2281Configure flexible payload selection::
2282
2283   flow_director_flex_payload (port_id) (raw|l2|l3|l4) (config)
2284
2285For example, to select the first 16 bytes from the offset 4 (bytes) of packet's payload as flexible payload::
2286
2287   testpmd> flow_director_flex_payload 0 l4 \
2288            (4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19)
2289
2290get_sym_hash_ena_per_port
2291~~~~~~~~~~~~~~~~~~~~~~~~~
2292
2293Get symmetric hash enable configuration per port::
2294
2295   get_sym_hash_ena_per_port (port_id)
2296
2297For example, to get symmetric hash enable configuration of port 1::
2298
2299   testpmd> get_sym_hash_ena_per_port 1
2300
2301set_sym_hash_ena_per_port
2302~~~~~~~~~~~~~~~~~~~~~~~~~
2303
2304Set symmetric hash enable configuration per port to enable or disable::
2305
2306   set_sym_hash_ena_per_port (port_id) (enable|disable)
2307
2308For example, to set symmetric hash enable configuration of port 1 to enable::
2309
2310   testpmd> set_sym_hash_ena_per_port 1 enable
2311
2312get_hash_global_config
2313~~~~~~~~~~~~~~~~~~~~~~
2314
2315Get the global configurations of hash filters::
2316
2317   get_hash_global_config (port_id)
2318
2319For example, to get the global configurations of hash filters of port 1::
2320
2321   testpmd> get_hash_global_config 1
2322
2323set_hash_global_config
2324~~~~~~~~~~~~~~~~~~~~~~
2325
2326Set the global configurations of hash filters::
2327
2328   set_hash_global_config (port_id) (toeplitz|simple_xor|default) \
2329   (ipv4|ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp|ipv4-other|ipv6|ipv6-frag| \
2330   ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other|l2_payload) \
2331   (enable|disable)
2332
2333For example, to enable simple_xor for flow type of ipv6 on port 2::
2334
2335   testpmd> set_hash_global_config 2 simple_xor ipv6 enable
2336
2337set_hash_input_set
2338~~~~~~~~~~~~~~~~~~
2339
2340Set the input set for hash::
2341
2342   set_hash_input_set (port_id) (ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
2343   ipv4-other|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other| \
2344   l2_payload) (ovlan|ivlan|src-ipv4|dst-ipv4|src-ipv6|dst-ipv6|ipv4-tos| \
2345   ipv4-proto|ipv6-tc|ipv6-next-header|udp-src-port|udp-dst-port| \
2346   tcp-src-port|tcp-dst-port|sctp-src-port|sctp-dst-port|sctp-veri-tag| \
2347   udp-key|gre-key|fld-1st|fld-2nd|fld-3rd|fld-4th|fld-5th|fld-6th|fld-7th| \
2348   fld-8th|none) (select|add)
2349
2350For example, to add source IP to hash input set for flow type of ipv4-udp on port 0::
2351
2352   testpmd> set_hash_input_set 0 ipv4-udp src-ipv4 add
2353
2354set_fdir_input_set
2355~~~~~~~~~~~~~~~~~~
2356
2357The Flow Director filters can match the different fields for different type of packet, i.e. specific input set
2358on per flow type and the flexible payload. This command can be used to change input set for each flow type.
2359
2360Set the input set for flow director::
2361
2362   set_fdir_input_set (port_id) (ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
2363   ipv4-other|ipv6|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other| \
2364   l2_payload) (ivlan|ethertype|src-ipv4|dst-ipv4|src-ipv6|dst-ipv6|ipv4-tos| \
2365   ipv4-proto|ipv4-ttl|ipv6-tc|ipv6-next-header|ipv6-hop-limits| \
2366   tudp-src-port|udp-dst-port|cp-src-port|tcp-dst-port|sctp-src-port| \
2367   sctp-dst-port|sctp-veri-tag|none) (select|add)
2368
2369For example to add source IP to FD input set for flow type of ipv4-udp on port 0::
2370
2371   testpmd> set_fdir_input_set 0 ipv4-udp src-ipv4 add
2372
2373global_config
2374~~~~~~~~~~~~~
2375
2376Set different GRE key length for input set::
2377
2378   global_config (port_id) gre-key-len (number in bytes)
2379
2380For example to set GRE key length for input set to 4 bytes on port 0::
2381
2382   testpmd> global_config 0 gre-key-len 4
2383
2384
2385.. _testpmd_rte_flow:
2386
2387Flow rules management
2388---------------------
2389
2390Control of the generic flow API (*rte_flow*) is fully exposed through the
2391``flow`` command (validation, creation, destruction, queries and operation
2392modes).
2393
2394Considering *rte_flow* overlaps with all `Filter Functions`_, using both
2395features simultaneously may cause undefined side-effects and is therefore
2396not recommended.
2397
2398``flow`` syntax
2399~~~~~~~~~~~~~~~
2400
2401Because the ``flow`` command uses dynamic tokens to handle the large number
2402of possible flow rules combinations, its behavior differs slightly from
2403other commands, in particular:
2404
2405- Pressing *?* or the *<tab>* key displays contextual help for the current
2406  token, not that of the entire command.
2407
2408- Optional and repeated parameters are supported (provided they are listed
2409  in the contextual help).
2410
2411The first parameter stands for the operation mode. Possible operations and
2412their general syntax are described below. They are covered in detail in the
2413following sections.
2414
2415- Check whether a flow rule can be created::
2416
2417   flow validate {port_id}
2418       [group {group_id}] [priority {level}] [ingress] [egress]
2419       pattern {item} [/ {item} [...]] / end
2420       actions {action} [/ {action} [...]] / end
2421
2422- Create a flow rule::
2423
2424   flow create {port_id}
2425       [group {group_id}] [priority {level}] [ingress] [egress]
2426       pattern {item} [/ {item} [...]] / end
2427       actions {action} [/ {action} [...]] / end
2428
2429- Destroy specific flow rules::
2430
2431   flow destroy {port_id} rule {rule_id} [...]
2432
2433- Destroy all flow rules::
2434
2435   flow flush {port_id}
2436
2437- Query an existing flow rule::
2438
2439   flow query {port_id} {rule_id} {action}
2440
2441- List existing flow rules sorted by priority, filtered by group
2442  identifiers::
2443
2444   flow list {port_id} [group {group_id}] [...]
2445
2446- Restrict ingress traffic to the defined flow rules::
2447
2448   flow isolate {port_id} {boolean}
2449
2450Validating flow rules
2451~~~~~~~~~~~~~~~~~~~~~
2452
2453``flow validate`` reports whether a flow rule would be accepted by the
2454underlying device in its current state but stops short of creating it. It is
2455bound to ``rte_flow_validate()``::
2456
2457   flow validate {port_id}
2458      [group {group_id}] [priority {level}] [ingress] [egress]
2459      pattern {item} [/ {item} [...]] / end
2460      actions {action} [/ {action} [...]] / end
2461
2462If successful, it will show::
2463
2464   Flow rule validated
2465
2466Otherwise it will show an error message of the form::
2467
2468   Caught error type [...] ([...]): [...]
2469
2470This command uses the same parameters as ``flow create``, their format is
2471described in `Creating flow rules`_.
2472
2473Check whether redirecting any Ethernet packet received on port 0 to RX queue
2474index 6 is supported::
2475
2476   testpmd> flow validate 0 ingress pattern eth / end
2477      actions queue index 6 / end
2478   Flow rule validated
2479   testpmd>
2480
2481Port 0 does not support TCPv6 rules::
2482
2483   testpmd> flow validate 0 ingress pattern eth / ipv6 / tcp / end
2484      actions drop / end
2485   Caught error type 9 (specific pattern item): Invalid argument
2486   testpmd>
2487
2488Creating flow rules
2489~~~~~~~~~~~~~~~~~~~
2490
2491``flow create`` validates and creates the specified flow rule. It is bound
2492to ``rte_flow_create()``::
2493
2494   flow create {port_id}
2495      [group {group_id}] [priority {level}] [ingress] [egress]
2496      pattern {item} [/ {item} [...]] / end
2497      actions {action} [/ {action} [...]] / end
2498
2499If successful, it will return a flow rule ID usable with other commands::
2500
2501   Flow rule #[...] created
2502
2503Otherwise it will show an error message of the form::
2504
2505   Caught error type [...] ([...]): [...]
2506
2507Parameters describe in the following order:
2508
2509- Attributes (*group*, *priority*, *ingress*, *egress* tokens).
2510- A matching pattern, starting with the *pattern* token and terminated by an
2511  *end* pattern item.
2512- Actions, starting with the *actions* token and terminated by an *end*
2513  action.
2514
2515These translate directly to *rte_flow* objects provided as-is to the
2516underlying functions.
2517
2518The shortest valid definition only comprises mandatory tokens::
2519
2520   testpmd> flow create 0 pattern end actions end
2521
2522Note that PMDs may refuse rules that essentially do nothing such as this
2523one.
2524
2525**All unspecified object values are automatically initialized to 0.**
2526
2527Attributes
2528^^^^^^^^^^
2529
2530These tokens affect flow rule attributes (``struct rte_flow_attr``) and are
2531specified before the ``pattern`` token.
2532
2533- ``group {group id}``: priority group.
2534- ``priority {level}``: priority level within group.
2535- ``ingress``: rule applies to ingress traffic.
2536- ``egress``: rule applies to egress traffic.
2537
2538Each instance of an attribute specified several times overrides the previous
2539value as shown below (group 4 is used)::
2540
2541   testpmd> flow create 0 group 42 group 24 group 4 [...]
2542
2543Note that once enabled, ``ingress`` and ``egress`` cannot be disabled.
2544
2545While not specifying a direction is an error, some rules may allow both
2546simultaneously.
2547
2548Most rules affect RX therefore contain the ``ingress`` token::
2549
2550   testpmd> flow create 0 ingress pattern [...]
2551
2552Matching pattern
2553^^^^^^^^^^^^^^^^
2554
2555A matching pattern starts after the ``pattern`` token. It is made of pattern
2556items and is terminated by a mandatory ``end`` item.
2557
2558Items are named after their type (*RTE_FLOW_ITEM_TYPE_* from ``enum
2559rte_flow_item_type``).
2560
2561The ``/`` token is used as a separator between pattern items as shown
2562below::
2563
2564   testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end [...]
2565
2566Note that protocol items like these must be stacked from lowest to highest
2567layer to make sense. For instance, the following rule is either invalid or
2568unlikely to match any packet::
2569
2570   testpmd> flow create 0 ingress pattern eth / udp / ipv4 / end [...]
2571
2572More information on these restrictions can be found in the *rte_flow*
2573documentation.
2574
2575Several items support additional specification structures, for example
2576``ipv4`` allows specifying source and destination addresses as follows::
2577
2578   testpmd> flow create 0 ingress pattern eth / ipv4 src is 10.1.1.1
2579      dst is 10.2.0.0 / end [...]
2580
2581This rule matches all IPv4 traffic with the specified properties.
2582
2583In this example, ``src`` and ``dst`` are field names of the underlying
2584``struct rte_flow_item_ipv4`` object. All item properties can be specified
2585in a similar fashion.
2586
2587The ``is`` token means that the subsequent value must be matched exactly,
2588and assigns ``spec`` and ``mask`` fields in ``struct rte_flow_item``
2589accordingly. Possible assignment tokens are:
2590
2591- ``is``: match value perfectly (with full bit-mask).
2592- ``spec``: match value according to configured bit-mask.
2593- ``last``: specify upper bound to establish a range.
2594- ``mask``: specify bit-mask with relevant bits set to one.
2595- ``prefix``: generate bit-mask from a prefix length.
2596
2597These yield identical results::
2598
2599   ipv4 src is 10.1.1.1
2600
2601::
2602
2603   ipv4 src spec 10.1.1.1 src mask 255.255.255.255
2604
2605::
2606
2607   ipv4 src spec 10.1.1.1 src prefix 32
2608
2609::
2610
2611   ipv4 src is 10.1.1.1 src last 10.1.1.1 # range with a single value
2612
2613::
2614
2615   ipv4 src is 10.1.1.1 src last 0 # 0 disables range
2616
2617Inclusive ranges can be defined with ``last``::
2618
2619   ipv4 src is 10.1.1.1 src last 10.2.3.4 # 10.1.1.1 to 10.2.3.4
2620
2621Note that ``mask`` affects both ``spec`` and ``last``::
2622
2623   ipv4 src is 10.1.1.1 src last 10.2.3.4 src mask 255.255.0.0
2624      # matches 10.1.0.0 to 10.2.255.255
2625
2626Properties can be modified multiple times::
2627
2628   ipv4 src is 10.1.1.1 src is 10.1.2.3 src is 10.2.3.4 # matches 10.2.3.4
2629
2630::
2631
2632   ipv4 src is 10.1.1.1 src prefix 24 src prefix 16 # matches 10.1.0.0/16
2633
2634Pattern items
2635^^^^^^^^^^^^^
2636
2637This section lists supported pattern items and their attributes, if any.
2638
2639- ``end``: end list of pattern items.
2640
2641- ``void``: no-op pattern item.
2642
2643- ``invert``: perform actions when pattern does not match.
2644
2645- ``any``: match any protocol for the current layer.
2646
2647  - ``num {unsigned}``: number of layers covered.
2648
2649- ``pf``: match packets addressed to the physical function.
2650
2651- ``vf``: match packets addressed to a virtual function ID.
2652
2653  - ``id {unsigned}``: destination VF ID.
2654
2655- ``port``: device-specific physical port index to use.
2656
2657  - ``index {unsigned}``: physical port index.
2658
2659- ``raw``: match an arbitrary byte string.
2660
2661  - ``relative {boolean}``: look for pattern after the previous item.
2662  - ``search {boolean}``: search pattern from offset (see also limit).
2663  - ``offset {integer}``: absolute or relative offset for pattern.
2664  - ``limit {unsigned}``: search area limit for start of pattern.
2665  - ``pattern {string}``: byte string to look for.
2666
2667- ``eth``: match Ethernet header.
2668
2669  - ``dst {MAC-48}``: destination MAC.
2670  - ``src {MAC-48}``: source MAC.
2671  - ``type {unsigned}``: EtherType.
2672
2673- ``vlan``: match 802.1Q/ad VLAN tag.
2674
2675  - ``tpid {unsigned}``: tag protocol identifier.
2676  - ``tci {unsigned}``: tag control information.
2677  - ``pcp {unsigned}``: priority code point.
2678  - ``dei {unsigned}``: drop eligible indicator.
2679  - ``vid {unsigned}``: VLAN identifier.
2680
2681- ``ipv4``: match IPv4 header.
2682
2683  - ``tos {unsigned}``: type of service.
2684  - ``ttl {unsigned}``: time to live.
2685  - ``proto {unsigned}``: next protocol ID.
2686  - ``src {ipv4 address}``: source address.
2687  - ``dst {ipv4 address}``: destination address.
2688
2689- ``ipv6``: match IPv6 header.
2690
2691  - ``tc {unsigned}``: traffic class.
2692  - ``flow {unsigned}``: flow label.
2693  - ``proto {unsigned}``: protocol (next header).
2694  - ``hop {unsigned}``: hop limit.
2695  - ``src {ipv6 address}``: source address.
2696  - ``dst {ipv6 address}``: destination address.
2697
2698- ``icmp``: match ICMP header.
2699
2700  - ``type {unsigned}``: ICMP packet type.
2701  - ``code {unsigned}``: ICMP packet code.
2702
2703- ``udp``: match UDP header.
2704
2705  - ``src {unsigned}``: UDP source port.
2706  - ``dst {unsigned}``: UDP destination port.
2707
2708- ``tcp``: match TCP header.
2709
2710  - ``src {unsigned}``: TCP source port.
2711  - ``dst {unsigned}``: TCP destination port.
2712
2713- ``sctp``: match SCTP header.
2714
2715  - ``src {unsigned}``: SCTP source port.
2716  - ``dst {unsigned}``: SCTP destination port.
2717  - ``tag {unsigned}``: validation tag.
2718  - ``cksum {unsigned}``: checksum.
2719
2720- ``vxlan``: match VXLAN header.
2721
2722  - ``vni {unsigned}``: VXLAN identifier.
2723
2724- ``e_tag``: match IEEE 802.1BR E-Tag header.
2725
2726  - ``grp_ecid_b {unsigned}``: GRP and E-CID base.
2727
2728- ``nvgre``: match NVGRE header.
2729
2730  - ``tni {unsigned}``: virtual subnet ID.
2731
2732- ``mpls``: match MPLS header.
2733
2734  - ``label {unsigned}``: MPLS label.
2735
2736- ``gre``: match GRE header.
2737
2738  - ``protocol {unsigned}``: protocol type.
2739
2740- ``fuzzy``: fuzzy pattern match, expect faster than default.
2741
2742  - ``thresh {unsigned}``: accuracy threshold.
2743
2744- ``gtp``, ``gtpc``, ``gtpu``: match GTPv1 header.
2745
2746  - ``teid {unsigned}``: tunnel endpoint identifier.
2747
2748Actions list
2749^^^^^^^^^^^^
2750
2751A list of actions starts after the ``actions`` token in the same fashion as
2752`Matching pattern`_; actions are separated by ``/`` tokens and the list is
2753terminated by a mandatory ``end`` action.
2754
2755Actions are named after their type (*RTE_FLOW_ACTION_TYPE_* from ``enum
2756rte_flow_action_type``).
2757
2758Dropping all incoming UDPv4 packets can be expressed as follows::
2759
2760   testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
2761      actions drop / end
2762
2763Several actions have configurable properties which must be specified when
2764there is no valid default value. For example, ``queue`` requires a target
2765queue index.
2766
2767This rule redirects incoming UDPv4 traffic to queue index 6::
2768
2769   testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
2770      actions queue index 6 / end
2771
2772While this one could be rejected by PMDs (unspecified queue index)::
2773
2774   testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
2775      actions queue / end
2776
2777As defined by *rte_flow*, the list is not ordered, all actions of a given
2778rule are performed simultaneously. These are equivalent::
2779
2780   queue index 6 / void / mark id 42 / end
2781
2782::
2783
2784   void / mark id 42 / queue index 6 / end
2785
2786All actions in a list should have different types, otherwise only the last
2787action of a given type is taken into account::
2788
2789   queue index 4 / queue index 5 / queue index 6 / end # will use queue 6
2790
2791::
2792
2793   drop / drop / drop / end # drop is performed only once
2794
2795::
2796
2797   mark id 42 / queue index 3 / mark id 24 / end # mark will be 24
2798
2799Considering they are performed simultaneously, opposite and overlapping
2800actions can sometimes be combined when the end result is unambiguous::
2801
2802   drop / queue index 6 / end # drop has no effect
2803
2804::
2805
2806   drop / dup index 6 / end # same as above
2807
2808::
2809
2810   queue index 6 / rss queues 6 7 8 / end # queue has no effect
2811
2812::
2813
2814   drop / passthru / end # drop has no effect
2815
2816Note that PMDs may still refuse such combinations.
2817
2818Actions
2819^^^^^^^
2820
2821This section lists supported actions and their attributes, if any.
2822
2823- ``end``: end list of actions.
2824
2825- ``void``: no-op action.
2826
2827- ``passthru``: let subsequent rule process matched packets.
2828
2829- ``mark``: attach 32 bit value to packets.
2830
2831  - ``id {unsigned}``: 32 bit value to return with packets.
2832
2833- ``flag``: flag packets.
2834
2835- ``queue``: assign packets to a given queue index.
2836
2837  - ``index {unsigned}``: queue index to use.
2838
2839- ``drop``: drop packets (note: passthru has priority).
2840
2841- ``count``: enable counters for this rule.
2842
2843- ``dup``: duplicate packets to a given queue index.
2844
2845  - ``index {unsigned}``: queue index to duplicate packets to.
2846
2847- ``rss``: spread packets among several queues.
2848
2849  - ``queues [{unsigned} [...]] end``: queue indices to use.
2850
2851- ``pf``: redirect packets to physical device function.
2852
2853- ``vf``: redirect packets to virtual device function.
2854
2855  - ``original {boolean}``: use original VF ID if possible.
2856  - ``id {unsigned}``: VF ID to redirect packets to.
2857
2858Destroying flow rules
2859~~~~~~~~~~~~~~~~~~~~~
2860
2861``flow destroy`` destroys one or more rules from their rule ID (as returned
2862by ``flow create``), this command calls ``rte_flow_destroy()`` as many
2863times as necessary::
2864
2865   flow destroy {port_id} rule {rule_id} [...]
2866
2867If successful, it will show::
2868
2869   Flow rule #[...] destroyed
2870
2871It does not report anything for rule IDs that do not exist. The usual error
2872message is shown when a rule cannot be destroyed::
2873
2874   Caught error type [...] ([...]): [...]
2875
2876``flow flush`` destroys all rules on a device and does not take extra
2877arguments. It is bound to ``rte_flow_flush()``::
2878
2879   flow flush {port_id}
2880
2881Any errors are reported as above.
2882
2883Creating several rules and destroying them::
2884
2885   testpmd> flow create 0 ingress pattern eth / ipv6 / end
2886      actions queue index 2 / end
2887   Flow rule #0 created
2888   testpmd> flow create 0 ingress pattern eth / ipv4 / end
2889      actions queue index 3 / end
2890   Flow rule #1 created
2891   testpmd> flow destroy 0 rule 0 rule 1
2892   Flow rule #1 destroyed
2893   Flow rule #0 destroyed
2894   testpmd>
2895
2896The same result can be achieved using ``flow flush``::
2897
2898   testpmd> flow create 0 ingress pattern eth / ipv6 / end
2899      actions queue index 2 / end
2900   Flow rule #0 created
2901   testpmd> flow create 0 ingress pattern eth / ipv4 / end
2902      actions queue index 3 / end
2903   Flow rule #1 created
2904   testpmd> flow flush 0
2905   testpmd>
2906
2907Non-existent rule IDs are ignored::
2908
2909   testpmd> flow create 0 ingress pattern eth / ipv6 / end
2910      actions queue index 2 / end
2911   Flow rule #0 created
2912   testpmd> flow create 0 ingress pattern eth / ipv4 / end
2913      actions queue index 3 / end
2914   Flow rule #1 created
2915   testpmd> flow destroy 0 rule 42 rule 10 rule 2
2916   testpmd>
2917   testpmd> flow destroy 0 rule 0
2918   Flow rule #0 destroyed
2919   testpmd>
2920
2921Querying flow rules
2922~~~~~~~~~~~~~~~~~~~
2923
2924``flow query`` queries a specific action of a flow rule having that
2925ability. Such actions collect information that can be reported using this
2926command. It is bound to ``rte_flow_query()``::
2927
2928   flow query {port_id} {rule_id} {action}
2929
2930If successful, it will display either the retrieved data for known actions
2931or the following message::
2932
2933   Cannot display result for action type [...] ([...])
2934
2935Otherwise, it will complain either that the rule does not exist or that some
2936error occurred::
2937
2938   Flow rule #[...] not found
2939
2940::
2941
2942   Caught error type [...] ([...]): [...]
2943
2944Currently only the ``count`` action is supported. This action reports the
2945number of packets that hit the flow rule and the total number of bytes. Its
2946output has the following format::
2947
2948   count:
2949    hits_set: [...] # whether "hits" contains a valid value
2950    bytes_set: [...] # whether "bytes" contains a valid value
2951    hits: [...] # number of packets
2952    bytes: [...] # number of bytes
2953
2954Querying counters for TCPv6 packets redirected to queue 6::
2955
2956   testpmd> flow create 0 ingress pattern eth / ipv6 / tcp / end
2957      actions queue index 6 / count / end
2958   Flow rule #4 created
2959   testpmd> flow query 0 4 count
2960   count:
2961    hits_set: 1
2962    bytes_set: 0
2963    hits: 386446
2964    bytes: 0
2965   testpmd>
2966
2967Listing flow rules
2968~~~~~~~~~~~~~~~~~~
2969
2970``flow list`` lists existing flow rules sorted by priority and optionally
2971filtered by group identifiers::
2972
2973   flow list {port_id} [group {group_id}] [...]
2974
2975This command only fails with the following message if the device does not
2976exist::
2977
2978   Invalid port [...]
2979
2980Output consists of a header line followed by a short description of each
2981flow rule, one per line. There is no output at all when no flow rules are
2982configured on the device::
2983
2984   ID      Group   Prio    Attr    Rule
2985   [...]   [...]   [...]   [...]   [...]
2986
2987``Attr`` column flags:
2988
2989- ``i`` for ``ingress``.
2990- ``e`` for ``egress``.
2991
2992Creating several flow rules and listing them::
2993
2994   testpmd> flow create 0 ingress pattern eth / ipv4 / end
2995      actions queue index 6 / end
2996   Flow rule #0 created
2997   testpmd> flow create 0 ingress pattern eth / ipv6 / end
2998      actions queue index 2 / end
2999   Flow rule #1 created
3000   testpmd> flow create 0 priority 5 ingress pattern eth / ipv4 / udp / end
3001      actions rss queues 6 7 8 end / end
3002   Flow rule #2 created
3003   testpmd> flow list 0
3004   ID      Group   Prio    Attr    Rule
3005   0       0       0       i-      ETH IPV4 => QUEUE
3006   1       0       0       i-      ETH IPV6 => QUEUE
3007   2       0       5       i-      ETH IPV4 UDP => RSS
3008   testpmd>
3009
3010Rules are sorted by priority (i.e. group ID first, then priority level)::
3011
3012   testpmd> flow list 1
3013   ID      Group   Prio    Attr    Rule
3014   0       0       0       i-      ETH => COUNT
3015   6       0       500     i-      ETH IPV6 TCP => DROP COUNT
3016   5       0       1000    i-      ETH IPV6 ICMP => QUEUE
3017   1       24      0       i-      ETH IPV4 UDP => QUEUE
3018   4       24      10      i-      ETH IPV4 TCP => DROP
3019   3       24      20      i-      ETH IPV4 => DROP
3020   2       24      42      i-      ETH IPV4 UDP => QUEUE
3021   7       63      0       i-      ETH IPV6 UDP VXLAN => MARK QUEUE
3022   testpmd>
3023
3024Output can be limited to specific groups::
3025
3026   testpmd> flow list 1 group 0 group 63
3027   ID      Group   Prio    Attr    Rule
3028   0       0       0       i-      ETH => COUNT
3029   6       0       500     i-      ETH IPV6 TCP => DROP COUNT
3030   5       0       1000    i-      ETH IPV6 ICMP => QUEUE
3031   7       63      0       i-      ETH IPV6 UDP VXLAN => MARK QUEUE
3032   testpmd>
3033
3034Toggling isolated mode
3035~~~~~~~~~~~~~~~~~~~~~~
3036
3037``flow isolate`` can be used to tell the underlying PMD that ingress traffic
3038must only be injected from the defined flow rules; that no default traffic
3039is expected outside those rules and the driver is free to assign more
3040resources to handle them. It is bound to ``rte_flow_isolate()``::
3041
3042 flow isolate {port_id} {boolean}
3043
3044If successful, enabling or disabling isolated mode shows either::
3045
3046 Ingress traffic on port [...]
3047    is now restricted to the defined flow rules
3048
3049Or::
3050
3051 Ingress traffic on port [...]
3052    is not restricted anymore to the defined flow rules
3053
3054Otherwise, in case of error::
3055
3056   Caught error type [...] ([...]): [...]
3057
3058Mainly due to its side effects, PMDs supporting this mode may not have the
3059ability to toggle it more than once without reinitializing affected ports
3060first (e.g. by exiting testpmd).
3061
3062Enabling isolated mode::
3063
3064 testpmd> flow isolate 0 true
3065 Ingress traffic on port 0 is now restricted to the defined flow rules
3066 testpmd>
3067
3068Disabling isolated mode::
3069
3070 testpmd> flow isolate 0 false
3071 Ingress traffic on port 0 is not restricted anymore to the defined flow rules
3072 testpmd>
3073
3074Sample QinQ flow rules
3075~~~~~~~~~~~~~~~~~~~~~~
3076
3077Before creating QinQ rule(s) the following commands should be issued to enable QinQ::
3078
3079   testpmd> port stop 0
3080   testpmd> vlan set qinq on 0
3081
3082The above command sets the inner and outer TPID's to 0x8100.
3083
3084To change the TPID's the following commands should be used::
3085
3086   testpmd> vlan set outer tpid 0xa100 0
3087   testpmd> vlan set inner tpid 0x9100 0
3088   testpmd> port start 0
3089
3090Validate and create a QinQ rule on port 0 to steer traffic to a VF queue in a VM.
3091
3092::
3093
3094   testpmd> flow validate 0 ingress pattern eth / vlan tci is 123 /
3095       vlan tci is 456 / end actions vf id 1 / queue index 0 / end
3096   Flow rule #0 validated
3097
3098   testpmd> flow create 0 ingress pattern eth / vlan tci is 4 /
3099       vlan tci is 456 / end actions vf id 123 / queue index 0 / end
3100   Flow rule #0 created
3101
3102   testpmd> flow list 0
3103   ID      Group   Prio    Attr    Rule
3104   0       0       0       i-      ETH VLAN VLAN=>VF QUEUE
3105
3106Validate and create a QinQ rule on port 0 to steer traffic to a queue on the host.
3107
3108::
3109
3110   testpmd> flow validate 0 ingress pattern eth / vlan tci is 321 /
3111        vlan tci is 654 / end actions pf / queue index 0 / end
3112   Flow rule #1 validated
3113
3114   testpmd> flow create 0 ingress pattern eth / vlan tci is 321 /
3115        vlan tci is 654 / end actions pf / queue index 1 / end
3116   Flow rule #1 created
3117
3118   testpmd> flow list 0
3119   ID      Group   Prio    Attr    Rule
3120   0       0       0       i-      ETH VLAN VLAN=>VF QUEUE
3121   1       0       0       i-      ETH VLAN VLAN=>PF QUEUE
3122