xref: /dpdk/doc/guides/testpmd_app_ug/testpmd_funcs.rst (revision 655c901bf7345e2eb7e2bb603a6c30ac6feff3c9)
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
350* ``tm``: Traffic Management forwarding mode
351  Demonstrates the use of ethdev traffic management APIs and softnic PMD for
352  QoS traffic management. In this mode, 5-level hierarchical QoS scheduler is
353  available as an default option that can be enabled through CLI. The user can
354  also modify the default hierarchy or specify the new hierarchy through CLI for
355  implementing QoS scheduler.  Requires ``CONFIG_RTE_LIBRTE_PMD_SOFTNIC=y`` ``CONFIG_RTE_LIBRTE_SCHED=y``.
356
357Example::
358
359   testpmd> set fwd rxonly
360
361   Set rxonly packet forwarding mode
362
363
364read rxd
365~~~~~~~~
366
367Display an RX descriptor for a port RX queue::
368
369   testpmd> read rxd (port_id) (queue_id) (rxd_id)
370
371For example::
372
373   testpmd> read rxd 0 0 4
374        0x0000000B - 0x001D0180 / 0x0000000B - 0x001D0180
375
376read txd
377~~~~~~~~
378
379Display a TX descriptor for a port TX queue::
380
381   testpmd> read txd (port_id) (queue_id) (txd_id)
382
383For example::
384
385   testpmd> read txd 0 0 4
386        0x00000001 - 0x24C3C440 / 0x000F0000 - 0x2330003C
387
388ddp get list
389~~~~~~~~~~~~
390
391Get loaded dynamic device personalization (DDP) package info list::
392
393   testpmd> ddp get list (port_id)
394
395ddp get info
396~~~~~~~~~~~~
397
398Display information about dynamic device personalization (DDP) profile::
399
400   testpmd> ddp get info (profile_path)
401
402show vf stats
403~~~~~~~~~~~~~
404
405Display VF statistics::
406
407   testpmd> show vf stats (port_id) (vf_id)
408
409clear vf stats
410~~~~~~~~~~~~~~
411
412Reset VF statistics::
413
414   testpmd> clear vf stats (port_id) (vf_id)
415
416show port pctype mapping
417~~~~~~~~~~~~~~~~~~~~~~~~
418
419List all items from the pctype mapping table::
420
421   testpmd> show port (port_id) pctype mapping
422
423
424Configuration Functions
425-----------------------
426
427The testpmd application can be configured from the runtime as well as from the command-line.
428
429This section details the available configuration functions that are available.
430
431.. note::
432
433   Configuration changes only become active when forwarding is started/restarted.
434
435set default
436~~~~~~~~~~~
437
438Reset forwarding to the default configuration::
439
440   testpmd> set default
441
442set verbose
443~~~~~~~~~~~
444
445Set the debug verbosity level::
446
447   testpmd> set verbose (level)
448
449Currently the only available levels are 0 (silent except for error) and 1 (fully verbose).
450
451set nbport
452~~~~~~~~~~
453
454Set the number of ports used by the application:
455
456set nbport (num)
457
458This is equivalent to the ``--nb-ports`` command-line option.
459
460set nbcore
461~~~~~~~~~~
462
463Set the number of cores used by the application::
464
465   testpmd> set nbcore (num)
466
467This is equivalent to the ``--nb-cores`` command-line option.
468
469.. note::
470
471   The number of cores used must not be greater than number of ports used multiplied by the number of queues per port.
472
473set coremask
474~~~~~~~~~~~~
475
476Set the forwarding cores hexadecimal mask::
477
478   testpmd> set coremask (mask)
479
480This is equivalent to the ``--coremask`` command-line option.
481
482.. note::
483
484   The master lcore is reserved for command line parsing only and cannot be masked on for packet forwarding.
485
486set portmask
487~~~~~~~~~~~~
488
489Set the forwarding ports hexadecimal mask::
490
491   testpmd> set portmask (mask)
492
493This is equivalent to the ``--portmask`` command-line option.
494
495set burst
496~~~~~~~~~
497
498Set number of packets per burst::
499
500   testpmd> set burst (num)
501
502This is equivalent to the ``--burst command-line`` option.
503
504When retry is enabled, the transmit delay time and number of retries can also be set::
505
506   testpmd> set burst tx delay (microseconds) retry (num)
507
508set txpkts
509~~~~~~~~~~
510
511Set the length of each segment of the TX-ONLY packets or length of packet for FLOWGEN mode::
512
513   testpmd> set txpkts (x[,y]*)
514
515Where x[,y]* represents a CSV list of values, without white space.
516
517set txsplit
518~~~~~~~~~~~
519
520Set the split policy for the TX packets, applicable for TX-ONLY and CSUM forwarding modes::
521
522   testpmd> set txsplit (off|on|rand)
523
524Where:
525
526* ``off`` disable packet copy & split for CSUM mode.
527
528* ``on`` split outgoing packet into multiple segments. Size of each segment
529  and number of segments per packet is determined by ``set txpkts`` command
530  (see above).
531
532* ``rand`` same as 'on', but number of segments per each packet is a random value between 1 and total number of segments.
533
534set corelist
535~~~~~~~~~~~~
536
537Set the list of forwarding cores::
538
539   testpmd> set corelist (x[,y]*)
540
541For example, to change the forwarding cores:
542
543.. code-block:: console
544
545   testpmd> set corelist 3,1
546   testpmd> show config fwd
547
548   io packet forwarding - ports=2 - cores=2 - streams=2 - NUMA support disabled
549   Logical Core 3 (socket 0) forwards packets on 1 streams:
550   RX P=0/Q=0 (socket 0) -> TX P=1/Q=0 (socket 0) peer=02:00:00:00:00:01
551   Logical Core 1 (socket 0) forwards packets on 1 streams:
552   RX P=1/Q=0 (socket 0) -> TX P=0/Q=0 (socket 0) peer=02:00:00:00:00:00
553
554.. note::
555
556   The cores are used in the same order as specified on the command line.
557
558set portlist
559~~~~~~~~~~~~
560
561Set the list of forwarding ports::
562
563   testpmd> set portlist (x[,y]*)
564
565For example, to change the port forwarding:
566
567.. code-block:: console
568
569   testpmd> set portlist 0,2,1,3
570   testpmd> show config fwd
571
572   io packet forwarding - ports=4 - cores=1 - streams=4
573   Logical Core 3 (socket 0) forwards packets on 4 streams:
574   RX P=0/Q=0 (socket 0) -> TX P=2/Q=0 (socket 0) peer=02:00:00:00:00:01
575   RX P=2/Q=0 (socket 0) -> TX P=0/Q=0 (socket 0) peer=02:00:00:00:00:00
576   RX P=1/Q=0 (socket 0) -> TX P=3/Q=0 (socket 0) peer=02:00:00:00:00:03
577   RX P=3/Q=0 (socket 0) -> TX P=1/Q=0 (socket 0) peer=02:00:00:00:00:02
578
579set tx loopback
580~~~~~~~~~~~~~~~
581
582Enable/disable tx loopback::
583
584   testpmd> set tx loopback (port_id) (on|off)
585
586set drop enable
587~~~~~~~~~~~~~~~
588
589set drop enable bit for all queues::
590
591   testpmd> set all queues drop (port_id) (on|off)
592
593set split drop enable (for VF)
594~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
595
596set split drop enable bit for VF from PF::
597
598   testpmd> set vf split drop (port_id) (vf_id) (on|off)
599
600set mac antispoof (for VF)
601~~~~~~~~~~~~~~~~~~~~~~~~~~
602
603Set mac antispoof for a VF from the PF::
604
605   testpmd> set vf mac antispoof  (port_id) (vf_id) (on|off)
606
607set macsec offload
608~~~~~~~~~~~~~~~~~~
609
610Enable/disable MACsec offload::
611
612   testpmd> set macsec offload (port_id) on encrypt (on|off) replay-protect (on|off)
613   testpmd> set macsec offload (port_id) off
614
615set macsec sc
616~~~~~~~~~~~~~
617
618Configure MACsec secure connection (SC)::
619
620   testpmd> set macsec sc (tx|rx) (port_id) (mac) (pi)
621
622.. note::
623
624   The pi argument is ignored for tx.
625   Check the NIC Datasheet for hardware limits.
626
627set macsec sa
628~~~~~~~~~~~~~
629
630Configure MACsec secure association (SA)::
631
632   testpmd> set macsec sa (tx|rx) (port_id) (idx) (an) (pn) (key)
633
634.. note::
635
636   The IDX value must be 0 or 1.
637   Check the NIC Datasheet for hardware limits.
638
639set broadcast mode (for VF)
640~~~~~~~~~~~~~~~~~~~~~~~~~~~
641
642Set broadcast mode for a VF from the PF::
643
644   testpmd> set vf broadcast (port_id) (vf_id) (on|off)
645
646vlan set strip
647~~~~~~~~~~~~~~
648
649Set the VLAN strip on a port::
650
651   testpmd> vlan set strip (on|off) (port_id)
652
653vlan set stripq
654~~~~~~~~~~~~~~~
655
656Set the VLAN strip for a queue on a port::
657
658   testpmd> vlan set stripq (on|off) (port_id,queue_id)
659
660vlan set stripq (for VF)
661~~~~~~~~~~~~~~~~~~~~~~~~
662
663Set VLAN strip for all queues in a pool for a VF from the PF::
664
665   testpmd> set vf vlan stripq (port_id) (vf_id) (on|off)
666
667vlan set insert (for VF)
668~~~~~~~~~~~~~~~~~~~~~~~~
669
670Set VLAN insert for a VF from the PF::
671
672   testpmd> set vf vlan insert (port_id) (vf_id) (vlan_id)
673
674vlan set tag (for VF)
675~~~~~~~~~~~~~~~~~~~~~
676
677Set VLAN tag for a VF from the PF::
678
679   testpmd> set vf vlan tag (port_id) (vf_id) (on|off)
680
681vlan set antispoof (for VF)
682~~~~~~~~~~~~~~~~~~~~~~~~~~~
683
684Set VLAN antispoof for a VF from the PF::
685
686   testpmd> set vf vlan antispoof (port_id) (vf_id) (on|off)
687
688vlan set filter
689~~~~~~~~~~~~~~~
690
691Set the VLAN filter on a port::
692
693   testpmd> vlan set filter (on|off) (port_id)
694
695vlan set qinq
696~~~~~~~~~~~~~
697
698Set the VLAN QinQ (extended queue in queue) on for a port::
699
700   testpmd> vlan set qinq (on|off) (port_id)
701
702vlan set tpid
703~~~~~~~~~~~~~
704
705Set the inner or outer VLAN TPID for packet filtering on a port::
706
707   testpmd> vlan set (inner|outer) tpid (value) (port_id)
708
709.. note::
710
711   TPID value must be a 16-bit number (value <= 65536).
712
713rx_vlan add
714~~~~~~~~~~~
715
716Add a VLAN ID, or all identifiers, to the set of VLAN identifiers filtered by port ID::
717
718   testpmd> rx_vlan add (vlan_id|all) (port_id)
719
720.. note::
721
722   VLAN filter must be set on that port. VLAN ID < 4096.
723   Depending on the NIC used, number of vlan_ids may be limited to the maximum entries
724   in VFTA table. This is important if enabling all vlan_ids.
725
726rx_vlan rm
727~~~~~~~~~~
728
729Remove a VLAN ID, or all identifiers, from the set of VLAN identifiers filtered by port ID::
730
731   testpmd> rx_vlan rm (vlan_id|all) (port_id)
732
733rx_vlan add (for VF)
734~~~~~~~~~~~~~~~~~~~~
735
736Add a VLAN ID, to the set of VLAN identifiers filtered for VF(s) for port ID::
737
738   testpmd> rx_vlan add (vlan_id) port (port_id) vf (vf_mask)
739
740rx_vlan rm (for VF)
741~~~~~~~~~~~~~~~~~~~
742
743Remove a VLAN ID, from the set of VLAN identifiers filtered for VF(s) for port ID::
744
745   testpmd> rx_vlan rm (vlan_id) port (port_id) vf (vf_mask)
746
747tunnel_filter add
748~~~~~~~~~~~~~~~~~
749
750Add a tunnel filter on a port::
751
752   testpmd> tunnel_filter add (port_id) (outer_mac) (inner_mac) (ip_addr) \
753            (inner_vlan) (vxlan|nvgre|ipingre) (imac-ivlan|imac-ivlan-tenid|\
754            imac-tenid|imac|omac-imac-tenid|oip|iip) (tenant_id) (queue_id)
755
756The available information categories are:
757
758* ``vxlan``: Set tunnel type as VXLAN.
759
760* ``nvgre``: Set tunnel type as NVGRE.
761
762* ``ipingre``: Set tunnel type as IP-in-GRE.
763
764* ``imac-ivlan``: Set filter type as Inner MAC and VLAN.
765
766* ``imac-ivlan-tenid``: Set filter type as Inner MAC, VLAN and tenant ID.
767
768* ``imac-tenid``: Set filter type as Inner MAC and tenant ID.
769
770* ``imac``: Set filter type as Inner MAC.
771
772* ``omac-imac-tenid``: Set filter type as Outer MAC, Inner MAC and tenant ID.
773
774* ``oip``: Set filter type as Outer IP.
775
776* ``iip``: Set filter type as Inner IP.
777
778Example::
779
780   testpmd> tunnel_filter add 0 68:05:CA:28:09:82 00:00:00:00:00:00 \
781            192.168.2.2 0 ipingre oip 1 1
782
783   Set an IP-in-GRE tunnel on port 0, and the filter type is Outer IP.
784
785tunnel_filter remove
786~~~~~~~~~~~~~~~~~~~~
787
788Remove a tunnel filter on a port::
789
790   testpmd> tunnel_filter rm (port_id) (outer_mac) (inner_mac) (ip_addr) \
791            (inner_vlan) (vxlan|nvgre|ipingre) (imac-ivlan|imac-ivlan-tenid|\
792            imac-tenid|imac|omac-imac-tenid|oip|iip) (tenant_id) (queue_id)
793
794rx_vxlan_port add
795~~~~~~~~~~~~~~~~~
796
797Add an UDP port for VXLAN packet filter on a port::
798
799   testpmd> rx_vxlan_port add (udp_port) (port_id)
800
801rx_vxlan_port remove
802~~~~~~~~~~~~~~~~~~~~
803
804Remove an UDP port for VXLAN packet filter on a port::
805
806   testpmd> rx_vxlan_port rm (udp_port) (port_id)
807
808tx_vlan set
809~~~~~~~~~~~
810
811Set hardware insertion of VLAN IDs in packets sent on a port::
812
813   testpmd> tx_vlan set (port_id) vlan_id[, vlan_id_outer]
814
815For example, set a single VLAN ID (5) insertion on port 0::
816
817   tx_vlan set 0 5
818
819Or, set double VLAN ID (inner: 2, outer: 3) insertion on port 1::
820
821   tx_vlan set 1 2 3
822
823
824tx_vlan set pvid
825~~~~~~~~~~~~~~~~
826
827Set port based hardware insertion of VLAN ID in packets sent on a port::
828
829   testpmd> tx_vlan set pvid (port_id) (vlan_id) (on|off)
830
831tx_vlan reset
832~~~~~~~~~~~~~
833
834Disable hardware insertion of a VLAN header in packets sent on a port::
835
836   testpmd> tx_vlan reset (port_id)
837
838csum set
839~~~~~~~~
840
841Select hardware or software calculation of the checksum when
842transmitting a packet using the ``csum`` forwarding engine::
843
844   testpmd> csum set (ip|udp|tcp|sctp|outer-ip) (hw|sw) (port_id)
845
846Where:
847
848* ``ip|udp|tcp|sctp`` always relate to  the inner layer.
849
850* ``outer-ip`` relates to the outer IP layer (only for IPv4) in the case where the packet is recognized
851  as a tunnel packet by the forwarding engine (vxlan, gre and ipip are
852  supported). See also the ``csum parse-tunnel`` command.
853
854.. note::
855
856   Check the NIC Datasheet for hardware limits.
857
858RSS queue region
859~~~~~~~~~~~~~~~~
860
861Set RSS queue region span on a port::
862
863   testpmd> set port (port_id) queue-region region_id (value) \
864		queue_start_index (value) queue_num (value)
865
866Set flowtype mapping on a RSS queue region on a port::
867
868   testpmd> set port (port_id) queue-region region_id (value) flowtype (value)
869
870where:
871
872* For the flowtype(pctype) of packet,the specific index for each type has
873  been defined in file i40e_type.h as enum i40e_filter_pctype.
874
875Set user priority mapping on a RSS queue region on a port::
876
877   testpmd> set port (port_id) queue-region UP (value) region_id (value)
878
879Flush all queue region related configuration on a port::
880
881   testpmd> set port (port_id) queue-region flush (on|off)
882
883where:
884
885* "on"is just an enable function which server for other configuration,
886  it is for all configuration about queue region from up layer,
887  at first will only keep in DPDK softwarestored in driver,
888  only after "flush on", it commit all configuration to HW.
889  "off" is just clean all configuration about queue region just now,
890  and restore all to DPDK i40e driver default config when start up.
891
892Show all queue region related configuration info on a port::
893
894   testpmd> show port (port_id) queue-region
895
896.. note::
897
898  Queue region only support on PF by now, so these command is
899  only for configuration of queue region on PF port.
900
901csum parse-tunnel
902~~~~~~~~~~~~~~~~~
903
904Define how tunneled packets should be handled by the csum forward
905engine::
906
907   testpmd> csum parse-tunnel (on|off) (tx_port_id)
908
909If enabled, the csum forward engine will try to recognize supported
910tunnel headers (vxlan, gre, ipip).
911
912If disabled, treat tunnel packets as non-tunneled packets (a inner
913header is handled as a packet payload).
914
915.. note::
916
917   The port argument is the TX port like in the ``csum set`` command.
918
919Example:
920
921Consider a packet in packet like the following::
922
923   eth_out/ipv4_out/udp_out/vxlan/eth_in/ipv4_in/tcp_in
924
925* If parse-tunnel is enabled, the ``ip|udp|tcp|sctp`` parameters of ``csum set``
926  command relate to the inner headers (here ``ipv4_in`` and ``tcp_in``), and the
927  ``outer-ip parameter`` relates to the outer headers (here ``ipv4_out``).
928
929* If parse-tunnel is disabled, the ``ip|udp|tcp|sctp`` parameters of ``csum  set``
930   command relate to the outer headers, here ``ipv4_out`` and ``udp_out``.
931
932csum show
933~~~~~~~~~
934
935Display tx checksum offload configuration::
936
937   testpmd> csum show (port_id)
938
939tso set
940~~~~~~~
941
942Enable TCP Segmentation Offload (TSO) in the ``csum`` forwarding engine::
943
944   testpmd> tso set (segsize) (port_id)
945
946.. note::
947
948   Check the NIC datasheet for hardware limits.
949
950tso show
951~~~~~~~~
952
953Display the status of TCP Segmentation Offload::
954
955   testpmd> tso show (port_id)
956
957set port - gro
958~~~~~~~~~~~~~~
959
960Enable or disable GRO in ``csum`` forwarding engine::
961
962   testpmd> set port <port_id> gro on|off
963
964If enabled, the csum forwarding engine will perform GRO on the TCP/IPv4
965packets received from the given port.
966
967If disabled, packets received from the given port won't be performed
968GRO. By default, GRO is disabled for all ports.
969
970.. note::
971
972   When enable GRO for a port, TCP/IPv4 packets received from the port
973   will be performed GRO. After GRO, all merged packets have bad
974   checksums, since the GRO library doesn't re-calculate checksums for
975   the merged packets. Therefore, if users want the merged packets to
976   have correct checksums, please select HW IP checksum calculation and
977   HW TCP checksum calculation for the port which the merged packets are
978   transmitted to.
979
980show port - gro
981~~~~~~~~~~~~~~~
982
983Display GRO configuration for a given port::
984
985   testpmd> show port <port_id> gro
986
987set gro flush
988~~~~~~~~~~~~~
989
990Set the cycle to flush the GROed packets from reassembly tables::
991
992   testpmd> set gro flush <cycles>
993
994When enable GRO, the csum forwarding engine performs GRO on received
995packets, and the GROed packets are stored in reassembly tables. Users
996can use this command to determine when the GROed packets are flushed
997from the reassembly tables.
998
999The ``cycles`` is measured in GRO operation times. The csum forwarding
1000engine flushes the GROed packets from the tables every ``cycles`` GRO
1001operations.
1002
1003By default, the value of ``cycles`` is 1, which means flush GROed packets
1004from the reassembly tables as soon as one GRO operation finishes. The value
1005of ``cycles`` should be in the range of 1 to ``GRO_MAX_FLUSH_CYCLES``.
1006
1007Please note that the large value of ``cycles`` may cause the poor TCP/IP
1008stack performance. Because the GROed packets are delayed to arrive the
1009stack, thus causing more duplicated ACKs and TCP retransmissions.
1010
1011set port - gso
1012~~~~~~~~~~~~~~
1013
1014Toggle per-port GSO support in ``csum`` forwarding engine::
1015
1016   testpmd> set port <port_id> gso on|off
1017
1018If enabled, the csum forwarding engine will perform GSO on supported IPv4
1019packets, transmitted on the given port.
1020
1021If disabled, packets transmitted on the given port will not undergo GSO.
1022By default, GSO is disabled for all ports.
1023
1024.. note::
1025
1026   When GSO is enabled on a port, supported IPv4 packets transmitted on that
1027   port undergo GSO. Afterwards, the segmented packets are represented by
1028   multi-segment mbufs; however, the csum forwarding engine doesn't calculation
1029   of checksums for GSO'd segments in SW. As a result, if users want correct
1030   checksums in GSO segments, they should enable HW checksum calculation for
1031   GSO-enabled ports.
1032
1033   For example, HW checksum calculation for VxLAN GSO'd packets may be enabled
1034   by setting the following options in the csum forwarding engine:
1035
1036   testpmd> csum set outer_ip hw <port_id>
1037
1038   testpmd> csum set ip hw <port_id>
1039
1040   testpmd> csum set tcp hw <port_id>
1041
1042set gso segsz
1043~~~~~~~~~~~~~
1044
1045Set the maximum GSO segment size (measured in bytes), which includes the
1046packet header and the packet payload for GSO-enabled ports (global)::
1047
1048   testpmd> set gso segsz <length>
1049
1050show port - gso
1051~~~~~~~~~~~~~~~
1052
1053Display the status of Generic Segmentation Offload for a given port::
1054
1055   testpmd> show port <port_id> gso
1056
1057mac_addr add
1058~~~~~~~~~~~~
1059
1060Add an alternative MAC address to a port::
1061
1062   testpmd> mac_addr add (port_id) (XX:XX:XX:XX:XX:XX)
1063
1064mac_addr remove
1065~~~~~~~~~~~~~~~
1066
1067Remove a MAC address from a port::
1068
1069   testpmd> mac_addr remove (port_id) (XX:XX:XX:XX:XX:XX)
1070
1071mac_addr add (for VF)
1072~~~~~~~~~~~~~~~~~~~~~
1073
1074Add an alternative MAC address for a VF to a port::
1075
1076   testpmd> mac_add add port (port_id) vf (vf_id) (XX:XX:XX:XX:XX:XX)
1077
1078mac_addr set
1079~~~~~~~~~~~~
1080
1081Set the default MAC address for a port::
1082
1083   testpmd> mac_addr set (port_id) (XX:XX:XX:XX:XX:XX)
1084
1085mac_addr set (for VF)
1086~~~~~~~~~~~~~~~~~~~~~
1087
1088Set the MAC address for a VF from the PF::
1089
1090   testpmd> set vf mac addr (port_id) (vf_id) (XX:XX:XX:XX:XX:XX)
1091
1092set eth-peer
1093~~~~~~~~~~~~
1094
1095Set the forwarding peer address for certain port::
1096
1097   testpmd> set eth-peer (port_id) (perr_addr)
1098
1099This is equivalent to the ``--eth-peer`` command-line option.
1100
1101set port-uta
1102~~~~~~~~~~~~
1103
1104Set the unicast hash filter(s) on/off for a port::
1105
1106   testpmd> set port (port_id) uta (XX:XX:XX:XX:XX:XX|all) (on|off)
1107
1108set promisc
1109~~~~~~~~~~~
1110
1111Set the promiscuous mode on for a port or for all ports.
1112In promiscuous mode packets are not dropped if they aren't for the specified MAC address::
1113
1114   testpmd> set promisc (port_id|all) (on|off)
1115
1116set allmulti
1117~~~~~~~~~~~~
1118
1119Set the allmulti mode for a port or for all ports::
1120
1121   testpmd> set allmulti (port_id|all) (on|off)
1122
1123Same as the ifconfig (8) option. Controls how multicast packets are handled.
1124
1125set promisc (for VF)
1126~~~~~~~~~~~~~~~~~~~~
1127
1128Set the unicast promiscuous mode for a VF from PF.
1129It's supported by Intel i40e NICs now.
1130In promiscuous mode packets are not dropped if they aren't for the specified MAC address::
1131
1132   testpmd> set vf promisc (port_id) (vf_id) (on|off)
1133
1134set allmulticast (for VF)
1135~~~~~~~~~~~~~~~~~~~~~~~~~
1136
1137Set the multicast promiscuous mode for a VF from PF.
1138It's supported by Intel i40e NICs now.
1139In promiscuous mode packets are not dropped if they aren't for the specified MAC address::
1140
1141   testpmd> set vf allmulti (port_id) (vf_id) (on|off)
1142
1143set tx max bandwidth (for VF)
1144~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1145
1146Set TX max absolute bandwidth (Mbps) for a VF from PF::
1147
1148   testpmd> set vf tx max-bandwidth (port_id) (vf_id) (max_bandwidth)
1149
1150set tc tx min bandwidth (for VF)
1151~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1152
1153Set all TCs' TX min relative bandwidth (%) for a VF from PF::
1154
1155   testpmd> set vf tc tx min-bandwidth (port_id) (vf_id) (bw1, bw2, ...)
1156
1157set tc tx max bandwidth (for VF)
1158~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1159
1160Set a TC's TX max absolute bandwidth (Mbps) for a VF from PF::
1161
1162   testpmd> set vf tc tx max-bandwidth (port_id) (vf_id) (tc_no) (max_bandwidth)
1163
1164set tc strict link priority mode
1165~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1166
1167Set some TCs' strict link priority mode on a physical port::
1168
1169   testpmd> set tx strict-link-priority (port_id) (tc_bitmap)
1170
1171set tc tx min bandwidth
1172~~~~~~~~~~~~~~~~~~~~~~~
1173
1174Set all TCs' TX min relative bandwidth (%) globally for all PF and VFs::
1175
1176   testpmd> set tc tx min-bandwidth (port_id) (bw1, bw2, ...)
1177
1178set flow_ctrl rx
1179~~~~~~~~~~~~~~~~
1180
1181Set the link flow control parameter on a port::
1182
1183   testpmd> set flow_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \
1184            (pause_time) (send_xon) mac_ctrl_frame_fwd (on|off) \
1185	    autoneg (on|off) (port_id)
1186
1187Where:
1188
1189* ``high_water`` (integer): High threshold value to trigger XOFF.
1190
1191* ``low_water`` (integer): Low threshold value to trigger XON.
1192
1193* ``pause_time`` (integer): Pause quota in the Pause frame.
1194
1195* ``send_xon`` (0/1): Send XON frame.
1196
1197* ``mac_ctrl_frame_fwd``: Enable receiving MAC control frames.
1198
1199* ``autoneg``: Change the auto-negotiation parameter.
1200
1201set pfc_ctrl rx
1202~~~~~~~~~~~~~~~
1203
1204Set the priority flow control parameter on a port::
1205
1206   testpmd> set pfc_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \
1207            (pause_time) (priority) (port_id)
1208
1209Where:
1210
1211* ``high_water`` (integer): High threshold value.
1212
1213* ``low_water`` (integer): Low threshold value.
1214
1215* ``pause_time`` (integer): Pause quota in the Pause frame.
1216
1217* ``priority`` (0-7): VLAN User Priority.
1218
1219set stat_qmap
1220~~~~~~~~~~~~~
1221
1222Set statistics mapping (qmapping 0..15) for RX/TX queue on port::
1223
1224   testpmd> set stat_qmap (tx|rx) (port_id) (queue_id) (qmapping)
1225
1226For example, to set rx queue 2 on port 0 to mapping 5::
1227
1228   testpmd>set stat_qmap rx 0 2 5
1229
1230set xstats-hide-zero
1231~~~~~~~~~~~~~~~~~~~~
1232
1233Set the option to hide zero values for xstats display::
1234
1235	testpmd> set xstats-hide-zero on|off
1236
1237.. note::
1238
1239	By default, the zero values are displayed for xstats.
1240
1241set port - rx/tx (for VF)
1242~~~~~~~~~~~~~~~~~~~~~~~~~
1243
1244Set VF receive/transmit from a port::
1245
1246   testpmd> set port (port_id) vf (vf_id) (rx|tx) (on|off)
1247
1248set port - mac address filter (for VF)
1249~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1250
1251Add/Remove unicast or multicast MAC addr filter for a VF::
1252
1253   testpmd> set port (port_id) vf (vf_id) (mac_addr) \
1254            (exact-mac|exact-mac-vlan|hashmac|hashmac-vlan) (on|off)
1255
1256set port - rx mode(for VF)
1257~~~~~~~~~~~~~~~~~~~~~~~~~~
1258
1259Set the VF receive mode of a port::
1260
1261   testpmd> set port (port_id) vf (vf_id) \
1262            rxmode (AUPE|ROPE|BAM|MPE) (on|off)
1263
1264The available receive modes are:
1265
1266* ``AUPE``: Accepts untagged VLAN.
1267
1268* ``ROPE``: Accepts unicast hash.
1269
1270* ``BAM``: Accepts broadcast packets.
1271
1272* ``MPE``: Accepts all multicast packets.
1273
1274set port - tx_rate (for Queue)
1275~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1276
1277Set TX rate limitation for a queue on a port::
1278
1279   testpmd> set port (port_id) queue (queue_id) rate (rate_value)
1280
1281set port - tx_rate (for VF)
1282~~~~~~~~~~~~~~~~~~~~~~~~~~~
1283
1284Set TX rate limitation for queues in VF on a port::
1285
1286   testpmd> set port (port_id) vf (vf_id) rate (rate_value) queue_mask (queue_mask)
1287
1288set port - mirror rule
1289~~~~~~~~~~~~~~~~~~~~~~
1290
1291Set pool or vlan type mirror rule for a port::
1292
1293   testpmd> set port (port_id) mirror-rule (rule_id) \
1294            (pool-mirror-up|pool-mirror-down|vlan-mirror) \
1295            (poolmask|vlanid[,vlanid]*) dst-pool (pool_id) (on|off)
1296
1297Set link mirror rule for a port::
1298
1299   testpmd> set port (port_id) mirror-rule (rule_id) \
1300           (uplink-mirror|downlink-mirror) dst-pool (pool_id) (on|off)
1301
1302For example to enable mirror traffic with vlan 0,1 to pool 0::
1303
1304   set port 0 mirror-rule 0 vlan-mirror 0,1 dst-pool 0 on
1305
1306reset port - mirror rule
1307~~~~~~~~~~~~~~~~~~~~~~~~
1308
1309Reset a mirror rule for a port::
1310
1311   testpmd> reset port (port_id) mirror-rule (rule_id)
1312
1313set flush_rx
1314~~~~~~~~~~~~
1315
1316Set the flush on RX streams before forwarding.
1317The default is flush ``on``.
1318Mainly used with PCAP drivers to turn off the default behavior of flushing the first 512 packets on RX streams::
1319
1320   testpmd> set flush_rx off
1321
1322set bypass mode
1323~~~~~~~~~~~~~~~
1324
1325Set the bypass mode for the lowest port on bypass enabled NIC::
1326
1327   testpmd> set bypass mode (normal|bypass|isolate) (port_id)
1328
1329set bypass event
1330~~~~~~~~~~~~~~~~
1331
1332Set the event required to initiate specified bypass mode for the lowest port on a bypass enabled::
1333
1334   testpmd> set bypass event (timeout|os_on|os_off|power_on|power_off) \
1335            mode (normal|bypass|isolate) (port_id)
1336
1337Where:
1338
1339* ``timeout``: Enable bypass after watchdog timeout.
1340
1341* ``os_on``: Enable bypass when OS/board is powered on.
1342
1343* ``os_off``: Enable bypass when OS/board is powered off.
1344
1345* ``power_on``: Enable bypass when power supply is turned on.
1346
1347* ``power_off``: Enable bypass when power supply is turned off.
1348
1349
1350set bypass timeout
1351~~~~~~~~~~~~~~~~~~
1352
1353Set the bypass watchdog timeout to ``n`` seconds where 0 = instant::
1354
1355   testpmd> set bypass timeout (0|1.5|2|3|4|8|16|32)
1356
1357show bypass config
1358~~~~~~~~~~~~~~~~~~
1359
1360Show the bypass configuration for a bypass enabled NIC using the lowest port on the NIC::
1361
1362   testpmd> show bypass config (port_id)
1363
1364set link up
1365~~~~~~~~~~~
1366
1367Set link up for a port::
1368
1369   testpmd> set link-up port (port id)
1370
1371set link down
1372~~~~~~~~~~~~~
1373
1374Set link down for a port::
1375
1376   testpmd> set link-down port (port id)
1377
1378E-tag set
1379~~~~~~~~~
1380
1381Enable E-tag insertion for a VF on a port::
1382
1383   testpmd> E-tag set insertion on port-tag-id (value) port (port_id) vf (vf_id)
1384
1385Disable E-tag insertion for a VF on a port::
1386
1387   testpmd> E-tag set insertion off port (port_id) vf (vf_id)
1388
1389Enable/disable E-tag stripping on a port::
1390
1391   testpmd> E-tag set stripping (on|off) port (port_id)
1392
1393Enable/disable E-tag based forwarding on a port::
1394
1395   testpmd> E-tag set forwarding (on|off) port (port_id)
1396
1397Add an E-tag forwarding filter on a port::
1398
1399   testpmd> E-tag set filter add e-tag-id (value) dst-pool (pool_id) port (port_id)
1400
1401Delete an E-tag forwarding filter on a port::
1402   testpmd> E-tag set filter del e-tag-id (value) port (port_id)
1403
1404ddp add
1405~~~~~~~
1406
1407Load a dynamic device personalization (DDP) package::
1408
1409   testpmd> ddp add (port_id) (package_path[,output_path])
1410
1411ddp del
1412~~~~~~~
1413
1414Delete a dynamic device personalization package::
1415
1416   testpmd> ddp del (port_id) (package_path)
1417
1418ptype mapping
1419~~~~~~~~~~~~~
1420
1421List all items from the ptype mapping table::
1422
1423   testpmd> ptype mapping get (port_id) (valid_only)
1424
1425Where:
1426
1427* ``valid_only``: A flag indicates if only list valid items(=1) or all itemss(=0).
1428
1429Replace a specific or a group of software defined ptype with a new one::
1430
1431   testpmd> ptype mapping replace  (port_id) (target) (mask) (pkt_type)
1432
1433where:
1434
1435* ``target``: A specific software ptype or a mask to represent a group of software ptypes.
1436
1437* ``mask``: A flag indicate if "target" is a specific software ptype(=0) or a ptype mask(=1).
1438
1439* ``pkt_type``: The new software ptype to replace the old ones.
1440
1441Update hardware defined ptype to software defined packet type mapping table::
1442
1443   testpmd> ptype mapping update (port_id) (hw_ptype) (sw_ptype)
1444
1445where:
1446
1447* ``hw_ptype``: hardware ptype as the index of the ptype mapping table.
1448
1449* ``sw_ptype``: software ptype as the value of the ptype mapping table.
1450
1451Reset ptype mapping table::
1452
1453   testpmd> ptype mapping reset (port_id)
1454
1455Port Functions
1456--------------
1457
1458The following sections show functions for configuring ports.
1459
1460.. note::
1461
1462   Port configuration changes only become active when forwarding is started/restarted.
1463
1464port attach
1465~~~~~~~~~~~
1466
1467Attach a port specified by pci address or virtual device args::
1468
1469   testpmd> port attach (identifier)
1470
1471To attach a new pci device, the device should be recognized by kernel first.
1472Then it should be moved under DPDK management.
1473Finally the port can be attached to testpmd.
1474
1475For example, to move a pci device using ixgbe under DPDK management:
1476
1477.. code-block:: console
1478
1479   # Check the status of the available devices.
1480   ./usertools/dpdk-devbind.py --status
1481
1482   Network devices using DPDK-compatible driver
1483   ============================================
1484   <none>
1485
1486   Network devices using kernel driver
1487   ===================================
1488   0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused=
1489
1490
1491   # Bind the device to igb_uio.
1492   sudo ./usertools/dpdk-devbind.py -b igb_uio 0000:0a:00.0
1493
1494
1495   # Recheck the status of the devices.
1496   ./usertools/dpdk-devbind.py --status
1497   Network devices using DPDK-compatible driver
1498   ============================================
1499   0000:0a:00.0 '82599ES 10-Gigabit' drv=igb_uio unused=
1500
1501To attach a port created by virtual device, above steps are not needed.
1502
1503For example, to attach a port whose pci address is 0000:0a:00.0.
1504
1505.. code-block:: console
1506
1507   testpmd> port attach 0000:0a:00.0
1508   Attaching a new port...
1509   EAL: PCI device 0000:0a:00.0 on NUMA socket -1
1510   EAL:   probe driver: 8086:10fb rte_ixgbe_pmd
1511   EAL:   PCI memory mapped at 0x7f83bfa00000
1512   EAL:   PCI memory mapped at 0x7f83bfa80000
1513   PMD: eth_ixgbe_dev_init(): MAC: 2, PHY: 18, SFP+: 5
1514   PMD: eth_ixgbe_dev_init(): port 0 vendorID=0x8086 deviceID=0x10fb
1515   Port 0 is attached. Now total ports is 1
1516   Done
1517
1518For example, to attach a port created by pcap PMD.
1519
1520.. code-block:: console
1521
1522   testpmd> port attach net_pcap0
1523   Attaching a new port...
1524   PMD: Initializing pmd_pcap for net_pcap0
1525   PMD: Creating pcap-backed ethdev on numa socket 0
1526   Port 0 is attached. Now total ports is 1
1527   Done
1528
1529In this case, identifier is ``net_pcap0``.
1530This identifier format is the same as ``--vdev`` format of DPDK applications.
1531
1532For example, to re-attach a bonded port which has been previously detached,
1533the mode and slave parameters must be given.
1534
1535.. code-block:: console
1536
1537   testpmd> port attach net_bond_0,mode=0,slave=1
1538   Attaching a new port...
1539   EAL: Initializing pmd_bond for net_bond_0
1540   EAL: Create bonded device net_bond_0 on port 0 in mode 0 on socket 0.
1541   Port 0 is attached. Now total ports is 1
1542   Done
1543
1544
1545port detach
1546~~~~~~~~~~~
1547
1548Detach a specific port::
1549
1550   testpmd> port detach (port_id)
1551
1552Before detaching a port, the port should be stopped and closed.
1553
1554For example, to detach a pci device port 0.
1555
1556.. code-block:: console
1557
1558   testpmd> port stop 0
1559   Stopping ports...
1560   Done
1561   testpmd> port close 0
1562   Closing ports...
1563   Done
1564
1565   testpmd> port detach 0
1566   Detaching a port...
1567   EAL: PCI device 0000:0a:00.0 on NUMA socket -1
1568   EAL:   remove driver: 8086:10fb rte_ixgbe_pmd
1569   EAL:   PCI memory unmapped at 0x7f83bfa00000
1570   EAL:   PCI memory unmapped at 0x7f83bfa80000
1571   Done
1572
1573
1574For example, to detach a virtual device port 0.
1575
1576.. code-block:: console
1577
1578   testpmd> port stop 0
1579   Stopping ports...
1580   Done
1581   testpmd> port close 0
1582   Closing ports...
1583   Done
1584
1585   testpmd> port detach 0
1586   Detaching a port...
1587   PMD: Closing pcap ethdev on numa socket 0
1588   Port 'net_pcap0' is detached. Now total ports is 0
1589   Done
1590
1591To remove a pci device completely from the system, first detach the port from testpmd.
1592Then the device should be moved under kernel management.
1593Finally the device can be removed using kernel pci hotplug functionality.
1594
1595For example, to move a pci device under kernel management:
1596
1597.. code-block:: console
1598
1599   sudo ./usertools/dpdk-devbind.py -b ixgbe 0000:0a:00.0
1600
1601   ./usertools/dpdk-devbind.py --status
1602
1603   Network devices using DPDK-compatible driver
1604   ============================================
1605   <none>
1606
1607   Network devices using kernel driver
1608   ===================================
1609   0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused=igb_uio
1610
1611To remove a port created by a virtual device, above steps are not needed.
1612
1613port start
1614~~~~~~~~~~
1615
1616Start all ports or a specific port::
1617
1618   testpmd> port start (port_id|all)
1619
1620port stop
1621~~~~~~~~~
1622
1623Stop all ports or a specific port::
1624
1625   testpmd> port stop (port_id|all)
1626
1627port close
1628~~~~~~~~~~
1629
1630Close all ports or a specific port::
1631
1632   testpmd> port close (port_id|all)
1633
1634port start/stop queue
1635~~~~~~~~~~~~~~~~~~~~~
1636
1637Start/stop a rx/tx queue on a specific port::
1638
1639   testpmd> port (port_id) (rxq|txq) (queue_id) (start|stop)
1640
1641Only take effect when port is started.
1642
1643port config - speed
1644~~~~~~~~~~~~~~~~~~~
1645
1646Set the speed and duplex mode for all ports or a specific port::
1647
1648   testpmd> port config (port_id|all) speed (10|100|1000|10000|25000|40000|50000|100000|auto) \
1649            duplex (half|full|auto)
1650
1651port config - queues/descriptors
1652~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1653
1654Set number of queues/descriptors for rxq, txq, rxd and txd::
1655
1656   testpmd> port config all (rxq|txq|rxd|txd) (value)
1657
1658This is equivalent to the ``--rxq``, ``--txq``, ``--rxd`` and ``--txd`` command-line options.
1659
1660port config - max-pkt-len
1661~~~~~~~~~~~~~~~~~~~~~~~~~
1662
1663Set the maximum packet length::
1664
1665   testpmd> port config all max-pkt-len (value)
1666
1667This is equivalent to the ``--max-pkt-len`` command-line option.
1668
1669port config - CRC Strip
1670~~~~~~~~~~~~~~~~~~~~~~~
1671
1672Set hardware CRC stripping on or off for all ports::
1673
1674   testpmd> port config all crc-strip (on|off)
1675
1676CRC stripping is on by default.
1677
1678The ``off`` option is equivalent to the ``--disable-crc-strip`` command-line option.
1679
1680port config - scatter
1681~~~~~~~~~~~~~~~~~~~~~~~
1682
1683Set RX scatter mode on or off for all ports::
1684
1685   testpmd> port config all scatter (on|off)
1686
1687RX scatter mode is off by default.
1688
1689The ``on`` option is equivalent to the ``--enable-scatter`` command-line option.
1690
1691port config - RX Checksum
1692~~~~~~~~~~~~~~~~~~~~~~~~~
1693
1694Set hardware RX checksum offload to on or off for all ports::
1695
1696   testpmd> port config all rx-cksum (on|off)
1697
1698Checksum offload is off by default.
1699
1700The ``on`` option is equivalent to the ``--enable-rx-cksum`` command-line option.
1701
1702port config - VLAN
1703~~~~~~~~~~~~~~~~~~
1704
1705Set hardware VLAN on or off for all ports::
1706
1707   testpmd> port config all hw-vlan (on|off)
1708
1709Hardware VLAN is on by default.
1710
1711The ``off`` option is equivalent to the ``--disable-hw-vlan`` command-line option.
1712
1713port config - VLAN filter
1714~~~~~~~~~~~~~~~~~~~~~~~~~
1715
1716Set hardware VLAN filter on or off for all ports::
1717
1718   testpmd> port config all hw-vlan-filter (on|off)
1719
1720Hardware VLAN filter is on by default.
1721
1722The ``off`` option is equivalent to the ``--disable-hw-vlan-filter`` command-line option.
1723
1724port config - VLAN strip
1725~~~~~~~~~~~~~~~~~~~~~~~~
1726
1727Set hardware VLAN strip on or off for all ports::
1728
1729   testpmd> port config all hw-vlan-strip (on|off)
1730
1731Hardware VLAN strip is on by default.
1732
1733The ``off`` option is equivalent to the ``--disable-hw-vlan-strip`` command-line option.
1734
1735port config - VLAN extend
1736~~~~~~~~~~~~~~~~~~~~~~~~~
1737
1738Set hardware VLAN extend on or off for all ports::
1739
1740   testpmd> port config all hw-vlan-extend (on|off)
1741
1742Hardware VLAN extend is off by default.
1743
1744The ``off`` option is equivalent to the ``--disable-hw-vlan-extend`` command-line option.
1745
1746port config - Drop Packets
1747~~~~~~~~~~~~~~~~~~~~~~~~~~
1748
1749Set packet drop for packets with no descriptors on or off for all ports::
1750
1751   testpmd> port config all drop-en (on|off)
1752
1753Packet dropping for packets with no descriptors is off by default.
1754
1755The ``on`` option is equivalent to the ``--enable-drop-en`` command-line option.
1756
1757port config - RSS
1758~~~~~~~~~~~~~~~~~
1759
1760Set the RSS (Receive Side Scaling) mode on or off::
1761
1762   testpmd> port config all rss (all|ip|tcp|udp|sctp|ether|port|vxlan|geneve|nvgre|none)
1763
1764RSS is on by default.
1765
1766The ``none`` option is equivalent to the ``--disable-rss`` command-line option.
1767
1768port config - RSS Reta
1769~~~~~~~~~~~~~~~~~~~~~~
1770
1771Set the RSS (Receive Side Scaling) redirection table::
1772
1773   testpmd> port config all rss reta (hash,queue)[,(hash,queue)]
1774
1775port config - DCB
1776~~~~~~~~~~~~~~~~~
1777
1778Set the DCB mode for an individual port::
1779
1780   testpmd> port config (port_id) dcb vt (on|off) (traffic_class) pfc (on|off)
1781
1782The traffic class should be 4 or 8.
1783
1784port config - Burst
1785~~~~~~~~~~~~~~~~~~~
1786
1787Set the number of packets per burst::
1788
1789   testpmd> port config all burst (value)
1790
1791This is equivalent to the ``--burst`` command-line option.
1792
1793port config - Threshold
1794~~~~~~~~~~~~~~~~~~~~~~~
1795
1796Set thresholds for TX/RX queues::
1797
1798   testpmd> port config all (threshold) (value)
1799
1800Where the threshold type can be:
1801
1802* ``txpt:`` Set the prefetch threshold register of the TX rings, 0 <= value <= 255.
1803
1804* ``txht:`` Set the host threshold register of the TX rings, 0 <= value <= 255.
1805
1806* ``txwt:`` Set the write-back threshold register of the TX rings, 0 <= value <= 255.
1807
1808* ``rxpt:`` Set the prefetch threshold register of the RX rings, 0 <= value <= 255.
1809
1810* ``rxht:`` Set the host threshold register of the RX rings, 0 <= value <= 255.
1811
1812* ``rxwt:`` Set the write-back threshold register of the RX rings, 0 <= value <= 255.
1813
1814* ``txfreet:`` Set the transmit free threshold of the TX rings, 0 <= value <= txd.
1815
1816* ``rxfreet:`` Set the transmit free threshold of the RX rings, 0 <= value <= rxd.
1817
1818* ``txrst:`` Set the transmit RS bit threshold of TX rings, 0 <= value <= txd.
1819
1820These threshold options are also available from the command-line.
1821
1822port config - E-tag
1823~~~~~~~~~~~~~~~~~~~
1824
1825Set the value of ether-type for E-tag::
1826
1827   testpmd> port config (port_id|all) l2-tunnel E-tag ether-type (value)
1828
1829Enable/disable the E-tag support::
1830
1831   testpmd> port config (port_id|all) l2-tunnel E-tag (enable|disable)
1832
1833port config pctype mapping
1834~~~~~~~~~~~~~~~~~~~~~~~~~~
1835
1836Reset pctype mapping table::
1837
1838   testpmd> port config (port_id) pctype mapping reset
1839
1840Update hardware defined pctype to software defined flow type mapping table::
1841
1842   testpmd> port config (port_id) pctype mapping update (pctype_id_0[,pctype_id_1]*) (flow_type_id)
1843
1844where:
1845
1846* ``pctype_id_x``: hardware pctype id as index of bit in bitmask value of the pctype mapping table.
1847
1848* ``flow_type_id``: software flow type id as the index of the pctype mapping table.
1849
1850port config input set
1851~~~~~~~~~~~~~~~~~~~~~
1852
1853Config RSS/FDIR/FDIR flexible payload input set for some pctype::
1854   testpmd> port config (port_id) pctype (pctype_id) \
1855            (hash_inset|fdir_inset|fdir_flx_inset) \
1856	    (get|set|clear) field (field_idx)
1857
1858Clear RSS/FDIR/FDIR flexible payload input set for some pctype::
1859   testpmd> port config (port_id) pctype (pctype_id) \
1860            (hash_inset|fdir_inset|fdir_flx_inset) clear all
1861
1862where:
1863
1864* ``pctype_id``: hardware packet classification types.
1865* ``field_idx``: hardware field index.
1866
1867Link Bonding Functions
1868----------------------
1869
1870The Link Bonding functions make it possible to dynamically create and
1871manage link bonding devices from within testpmd interactive prompt.
1872
1873create bonded device
1874~~~~~~~~~~~~~~~~~~~~
1875
1876Create a new bonding device::
1877
1878   testpmd> create bonded device (mode) (socket)
1879
1880For example, to create a bonded device in mode 1 on socket 0::
1881
1882   testpmd> create bonded 1 0
1883   created new bonded device (port X)
1884
1885add bonding slave
1886~~~~~~~~~~~~~~~~~
1887
1888Adds Ethernet device to a Link Bonding device::
1889
1890   testpmd> add bonding slave (slave id) (port id)
1891
1892For example, to add Ethernet device (port 6) to a Link Bonding device (port 10)::
1893
1894   testpmd> add bonding slave 6 10
1895
1896
1897remove bonding slave
1898~~~~~~~~~~~~~~~~~~~~
1899
1900Removes an Ethernet slave device from a Link Bonding device::
1901
1902   testpmd> remove bonding slave (slave id) (port id)
1903
1904For example, to remove Ethernet slave device (port 6) to a Link Bonding device (port 10)::
1905
1906   testpmd> remove bonding slave 6 10
1907
1908set bonding mode
1909~~~~~~~~~~~~~~~~
1910
1911Set the Link Bonding mode of a Link Bonding device::
1912
1913   testpmd> set bonding mode (value) (port id)
1914
1915For example, to set the bonding mode of a Link Bonding device (port 10) to broadcast (mode 3)::
1916
1917   testpmd> set bonding mode 3 10
1918
1919set bonding primary
1920~~~~~~~~~~~~~~~~~~~
1921
1922Set an Ethernet slave device as the primary device on a Link Bonding device::
1923
1924   testpmd> set bonding primary (slave id) (port id)
1925
1926For example, to set the Ethernet slave device (port 6) as the primary port of a Link Bonding device (port 10)::
1927
1928   testpmd> set bonding primary 6 10
1929
1930set bonding mac
1931~~~~~~~~~~~~~~~
1932
1933Set the MAC address of a Link Bonding device::
1934
1935   testpmd> set bonding mac (port id) (mac)
1936
1937For example, to set the MAC address of a Link Bonding device (port 10) to 00:00:00:00:00:01::
1938
1939   testpmd> set bonding mac 10 00:00:00:00:00:01
1940
1941set bonding xmit_balance_policy
1942~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1943
1944Set the transmission policy for a Link Bonding device when it is in Balance XOR mode::
1945
1946   testpmd> set bonding xmit_balance_policy (port_id) (l2|l23|l34)
1947
1948For example, set a Link Bonding device (port 10) to use a balance policy of layer 3+4 (IP addresses & UDP ports)::
1949
1950   testpmd> set bonding xmit_balance_policy 10 l34
1951
1952
1953set bonding mon_period
1954~~~~~~~~~~~~~~~~~~~~~~
1955
1956Set the link status monitoring polling period in milliseconds for a bonding device.
1957
1958This adds support for PMD slave devices which do not support link status interrupts.
1959When the mon_period is set to a value greater than 0 then all PMD's which do not support
1960link status ISR will be queried every polling interval to check if their link status has changed::
1961
1962   testpmd> set bonding mon_period (port_id) (value)
1963
1964For example, to set the link status monitoring polling period of bonded device (port 5) to 150ms::
1965
1966   testpmd> set bonding mon_period 5 150
1967
1968
1969set bonding lacp dedicated_queue
1970~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1971
1972Enable dedicated tx/rx queues on bonding devices slaves to handle LACP control plane traffic
1973when in mode 4 (link-aggregration-802.3ad)::
1974
1975   testpmd> set bonding lacp dedicated_queues (port_id) (enable|disable)
1976
1977
1978set bonding agg_mode
1979~~~~~~~~~~~~~~~~~~~~
1980
1981Enable one of the specific aggregators mode when in mode 4 (link-aggregration-802.3ad)::
1982
1983   testpmd> set bonding agg_mode (port_id) (bandwidth|count|stable)
1984
1985
1986show bonding config
1987~~~~~~~~~~~~~~~~~~~
1988
1989Show the current configuration of a Link Bonding device::
1990
1991   testpmd> show bonding config (port id)
1992
1993For example,
1994to show the configuration a Link Bonding device (port 9) with 3 slave devices (1, 3, 4)
1995in balance mode with a transmission policy of layer 2+3::
1996
1997   testpmd> show bonding config 9
1998        Bonding mode: 2
1999        Balance Xmit Policy: BALANCE_XMIT_POLICY_LAYER23
2000        Slaves (3): [1 3 4]
2001        Active Slaves (3): [1 3 4]
2002        Primary: [3]
2003
2004
2005Register Functions
2006------------------
2007
2008The Register Functions can be used to read from and write to registers on the network card referenced by a port number.
2009This is mainly useful for debugging purposes.
2010Reference should be made to the appropriate datasheet for the network card for details on the register addresses
2011and fields that can be accessed.
2012
2013read reg
2014~~~~~~~~
2015
2016Display the value of a port register::
2017
2018   testpmd> read reg (port_id) (address)
2019
2020For example, to examine the Flow Director control register (FDIRCTL, 0x0000EE000) on an Intel 82599 10 GbE Controller::
2021
2022   testpmd> read reg 0 0xEE00
2023   port 0 PCI register at offset 0xEE00: 0x4A060029 (1241907241)
2024
2025read regfield
2026~~~~~~~~~~~~~
2027
2028Display a port register bit field::
2029
2030   testpmd> read regfield (port_id) (address) (bit_x) (bit_y)
2031
2032For example, reading the lowest two bits from the register in the example above::
2033
2034   testpmd> read regfield 0 0xEE00 0 1
2035   port 0 PCI register at offset 0xEE00: bits[0, 1]=0x1 (1)
2036
2037read regbit
2038~~~~~~~~~~~
2039
2040Display a single port register bit::
2041
2042   testpmd> read regbit (port_id) (address) (bit_x)
2043
2044For example, reading the lowest bit from the register in the example above::
2045
2046   testpmd> read regbit 0 0xEE00 0
2047   port 0 PCI register at offset 0xEE00: bit 0=1
2048
2049write reg
2050~~~~~~~~~
2051
2052Set the value of a port register::
2053
2054   testpmd> write reg (port_id) (address) (value)
2055
2056For example, to clear a register::
2057
2058   testpmd> write reg 0 0xEE00 0x0
2059   port 0 PCI register at offset 0xEE00: 0x00000000 (0)
2060
2061write regfield
2062~~~~~~~~~~~~~~
2063
2064Set bit field of a port register::
2065
2066   testpmd> write regfield (port_id) (address) (bit_x) (bit_y) (value)
2067
2068For example, writing to the register cleared in the example above::
2069
2070   testpmd> write regfield 0 0xEE00 0 1 2
2071   port 0 PCI register at offset 0xEE00: 0x00000002 (2)
2072
2073write regbit
2074~~~~~~~~~~~~
2075
2076Set single bit value of a port register::
2077
2078   testpmd> write regbit (port_id) (address) (bit_x) (value)
2079
2080For example, to set the high bit in the register from the example above::
2081
2082   testpmd> write regbit 0 0xEE00 31 1
2083   port 0 PCI register at offset 0xEE00: 0x8000000A (2147483658)
2084
2085Traffic Metering and Policing
2086-----------------------------
2087
2088The following section shows functions for configuring traffic metering and
2089policing on the ethernet device through the use of generic ethdev API.
2090
2091show port traffic management capability
2092~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2093
2094Show traffic metering and policing capability of the port::
2095
2096   testpmd> show port meter cap (port_id)
2097
2098add port meter profile (srTCM rfc2967)
2099~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2100
2101Add meter profile (srTCM rfc2697) to the ethernet device::
2102
2103   testpmd> add port meter profile srtcm_rfc2697 (port_id) (profile_id) \
2104   (cir) (cbs) (ebs)
2105
2106where:
2107
2108* ``profile_id``: ID for the meter profile.
2109* ``cir``: Committed Information Rate (CIR) (bytes/second).
2110* ``cbs``: Committed Burst Size (CBS) (bytes).
2111* ``ebs``: Excess Burst Size (EBS) (bytes).
2112
2113add port meter profile (trTCM rfc2968)
2114~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2115
2116Add meter profile (srTCM rfc2698) to the ethernet device::
2117
2118   testpmd> add port meter profile trtcm_rfc2698 (port_id) (profile_id) \
2119   (cir) (pir) (cbs) (pbs)
2120
2121where:
2122
2123* ``profile_id``: ID for the meter profile.
2124* ``cir``: Committed information rate (bytes/second).
2125* ``pir``: Peak information rate (bytes/second).
2126* ``cbs``: Committed burst size (bytes).
2127* ``pbs``: Peak burst size (bytes).
2128
2129add port meter profile (trTCM rfc4115)
2130~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2131
2132Add meter profile (trTCM rfc4115) to the ethernet device::
2133
2134   testpmd> add port meter profile trtcm_rfc4115 (port_id) (profile_id) \
2135   (cir) (eir) (cbs) (ebs)
2136
2137where:
2138
2139* ``profile_id``: ID for the meter profile.
2140* ``cir``: Committed information rate (bytes/second).
2141* ``eir``: Excess information rate (bytes/second).
2142* ``cbs``: Committed burst size (bytes).
2143* ``ebs``: Excess burst size (bytes).
2144
2145delete port meter profile
2146~~~~~~~~~~~~~~~~~~~~~~~~~
2147
2148Delete meter profile from the ethernet device::
2149
2150   testpmd> del port meter profile (port_id) (profile_id)
2151
2152create port meter
2153~~~~~~~~~~~~~~~~~
2154
2155Create new meter object for the ethernet device::
2156
2157   testpmd> create port meter (port_id) (mtr_id) (profile_id) \
2158   (meter_enable) (g_action) (y_action) (r_action) (stats_mask) (shared) \
2159   (use_pre_meter_color) [(dscp_tbl_entry0) (dscp_tbl_entry1)...\
2160   (dscp_tbl_entry63)]
2161
2162where:
2163
2164* ``mtr_id``: meter object ID.
2165* ``profile_id``: ID for the meter profile.
2166* ``meter_enable``: When this parameter has a non-zero value, the meter object
2167  gets enabled at the time of creation, otherwise remains disabled.
2168* ``g_action``: Policer action for the packet with green color.
2169* ``y_action``: Policer action for the packet with yellow color.
2170* ``r_action``: Policer action for the packet with red color.
2171* ``stats_mask``: Mask of statistics counter types to be enabled for the
2172  meter object.
2173* ``shared``:  When this parameter has a non-zero value, the meter object is
2174  shared by multiple flows. Otherwise, meter object is used by single flow.
2175* ``use_pre_meter_color``: When this parameter has a non-zero value, the
2176  input color for the current meter object is determined by the latest meter
2177  object in the same flow. Otherwise, the current meter object uses the
2178  *dscp_table* to determine the input color.
2179* ``dscp_tbl_entryx``: DSCP table entry x providing meter providing input
2180  color, 0 <= x <= 63.
2181
2182enable port meter
2183~~~~~~~~~~~~~~~~~
2184
2185Enable meter for the ethernet device::
2186
2187   testpmd> enable port meter (port_id) (mtr_id)
2188
2189disable port meter
2190~~~~~~~~~~~~~~~~~~
2191
2192Disable meter for the ethernet device::
2193
2194   testpmd> disable port meter (port_id) (mtr_id)
2195
2196delete port meter
2197~~~~~~~~~~~~~~~~~
2198
2199Delete meter for the ethernet device::
2200
2201   testpmd> del port meter (port_id) (mtr_id)
2202
2203Set port meter profile
2204~~~~~~~~~~~~~~~~~~~~~~
2205
2206Set meter profile for the ethernet device::
2207
2208   testpmd> set port meter profile (port_id) (mtr_id) (profile_id)
2209
2210set port meter dscp table
2211~~~~~~~~~~~~~~~~~~~~~~~~~
2212
2213Set meter dscp table for the ethernet device::
2214
2215   testpmd> set port meter dscp table (port_id) (mtr_id) [(dscp_tbl_entry0) \
2216   (dscp_tbl_entry1)...(dscp_tbl_entry63)]
2217
2218set port meter policer action
2219~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2220
2221Set meter policer action for the ethernet device::
2222
2223   testpmd> set port meter policer action (port_id) (mtr_id) (action_mask) \
2224   (action0) [(action1) (action1)]
2225
2226where:
2227
2228* ``action_mask``: Bit mask indicating which policer actions need to be
2229  updated. One or more policer actions can be updated in a single function
2230  invocation. To update the policer action associated with color C, bit
2231  (1 << C) needs to be set in *action_mask* and element at position C
2232  in the *actions* array needs to be valid.
2233* ``actionx``: Policer action for the color x,
2234  RTE_MTR_GREEN <= x < RTE_MTR_COLORS
2235
2236set port meter stats mask
2237~~~~~~~~~~~~~~~~~~~~~~~~~
2238
2239Set meter stats mask for the ethernet device::
2240
2241   testpmd> set port meter stats mask (port_id) (mtr_id) (stats_mask)
2242
2243where:
2244
2245* ``stats_mask``: Bit mask indicating statistics counter types to be enabled.
2246
2247show port meter stats
2248~~~~~~~~~~~~~~~~~~~~~
2249
2250Show meter stats of the ethernet device::
2251
2252   testpmd> show port meter stats (port_id) (mtr_id) (clear)
2253
2254where:
2255
2256* ``clear``: Flag that indicates whether the statistics counters should
2257  be cleared (i.e. set to zero) immediately after they have been read or not.
2258
2259Traffic Management
2260------------------
2261
2262The following section shows functions for configuring traffic management on
2263on the ethernet device through the use of generic TM API.
2264
2265show port traffic management capability
2266~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2267
2268Show traffic management capability of the port::
2269
2270   testpmd> show port tm cap (port_id)
2271
2272show port traffic management capability (hierarchy level)
2273~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2274
2275Show traffic management hierarchy level capability of the port::
2276
2277   testpmd> show port tm cap (port_id) (level_id)
2278
2279show port traffic management capability (hierarchy node level)
2280~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2281
2282Show the traffic management hierarchy node capability of the port::
2283
2284   testpmd> show port tm cap (port_id) (node_id)
2285
2286show port traffic management hierarchy node type
2287~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2288
2289Show the port traffic management hierarchy node type::
2290
2291   testpmd> show port tm node type (port_id) (node_id)
2292
2293show port traffic management hierarchy node stats
2294~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2295
2296Show the port traffic management hierarchy node statistics::
2297
2298   testpmd> show port tm node stats (port_id) (node_id) (clear)
2299
2300where:
2301
2302* ``clear``: When this parameter has a non-zero value, the statistics counters
2303  are cleared (i.e. set to zero) immediately after they have been read,
2304  otherwise the statistics counters are left untouched.
2305
2306Add port traffic management private shaper profile
2307~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2308
2309Add the port traffic management private shaper profile::
2310
2311   testpmd> add port tm node shaper profile (port_id) (shaper_profile_id) \
2312   (tb_rate) (tb_size) (packet_length_adjust)
2313
2314where:
2315
2316* ``shaper_profile id``: Shaper profile ID for the new profile.
2317* ``tb_rate``: Token bucket rate (bytes per second).
2318* ``tb_size``: Token bucket size (bytes).
2319* ``packet_length_adjust``: The value (bytes) to be added to the length of
2320  each packet for the purpose of shaping. This parameter value can be used to
2321  correct the packet length with the framing overhead bytes that are consumed
2322  on the wire.
2323
2324Delete port traffic management private shaper profile
2325~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2326
2327Delete the port traffic management private shaper::
2328
2329   testpmd> del port tm node shaper profile (port_id) (shaper_profile_id)
2330
2331where:
2332
2333* ``shaper_profile id``: Shaper profile ID that needs to be deleted.
2334
2335Add port traffic management shared shaper
2336~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2337
2338Create the port traffic management shared shaper::
2339
2340   testpmd> add port tm node shared shaper (port_id) (shared_shaper_id) \
2341   (shaper_profile_id)
2342
2343where:
2344
2345* ``shared_shaper_id``: Shared shaper ID to be created.
2346* ``shaper_profile id``: Shaper profile ID for shared shaper.
2347
2348Set port traffic management shared shaper
2349~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2350
2351Update the port traffic management shared shaper::
2352
2353   testpmd> set port tm node shared shaper (port_id) (shared_shaper_id) \
2354   (shaper_profile_id)
2355
2356where:
2357
2358* ``shared_shaper_id``: Shared shaper ID to be update.
2359* ``shaper_profile id``: Shaper profile ID for shared shaper.
2360
2361Delete port traffic management shared shaper
2362~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2363
2364Delete the port traffic management shared shaper::
2365
2366   testpmd> del port tm node shared shaper (port_id) (shared_shaper_id)
2367
2368where:
2369
2370* ``shared_shaper_id``: Shared shaper ID to be deleted.
2371
2372Set port traffic management hiearchy node private shaper
2373~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2374
2375set the port traffic management hierarchy node private shaper::
2376
2377   testpmd> set port tm node shaper profile (port_id) (node_id) \
2378   (shaper_profile_id)
2379
2380where:
2381
2382* ``shaper_profile id``: Private shaper profile ID to be enabled on the
2383  hierarchy node.
2384
2385Add port traffic management WRED profile
2386~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2387
2388Create a new WRED profile::
2389
2390   testpmd> add port tm node wred profile (port_id) (wred_profile_id) \
2391   (color_g) (min_th_g) (max_th_g) (maxp_inv_g) (wq_log2_g) \
2392   (color_y) (min_th_y) (max_th_y) (maxp_inv_y) (wq_log2_y) \
2393   (color_r) (min_th_r) (max_th_r) (maxp_inv_r) (wq_log2_r)
2394
2395where:
2396
2397* ``wred_profile id``: Identifier for the newly create WRED profile
2398* ``color_g``: Packet color (green)
2399* ``min_th_g``: Minimum queue threshold for packet with green color
2400* ``max_th_g``: Minimum queue threshold for packet with green color
2401* ``maxp_inv_g``: Inverse of packet marking probability maximum value (maxp)
2402* ``wq_log2_g``: Negated log2 of queue weight (wq)
2403* ``color_y``: Packet color (yellow)
2404* ``min_th_y``: Minimum queue threshold for packet with yellow color
2405* ``max_th_y``: Minimum queue threshold for packet with yellow color
2406* ``maxp_inv_y``: Inverse of packet marking probability maximum value (maxp)
2407* ``wq_log2_y``: Negated log2 of queue weight (wq)
2408* ``color_r``: Packet color (red)
2409* ``min_th_r``: Minimum queue threshold for packet with yellow color
2410* ``max_th_r``: Minimum queue threshold for packet with yellow color
2411* ``maxp_inv_r``: Inverse of packet marking probability maximum value (maxp)
2412* ``wq_log2_r``: Negated log2 of queue weight (wq)
2413
2414Delete port traffic management WRED profile
2415~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2416
2417Delete the WRED profile::
2418
2419   testpmd> del port tm node wred profile (port_id) (wred_profile_id)
2420
2421Add port traffic management hierarchy nonleaf node
2422~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2423
2424Add nonleaf node to port traffic management hiearchy::
2425
2426   testpmd> add port tm nonleaf node (port_id) (node_id) (parent_node_id) \
2427   (priority) (weight) (level_id) (shaper_profile_id) \
2428   (n_sp_priorities) (stats_mask) (n_shared_shapers) \
2429   [(shared_shaper_0) (shared_shaper_1) ...] \
2430
2431where:
2432
2433* ``parent_node_id``: Node ID of the parent.
2434* ``priority``: Node priority (highest node priority is zero). This is used by
2435  the SP algorithm running on the parent node for scheduling this node.
2436* ``weight``: Node weight (lowest weight is one). The node weight is relative
2437  to the weight sum of all siblings that have the same priority. It is used by
2438  the WFQ algorithm running on the parent node for scheduling this node.
2439* ``level_id``: Hiearchy level of the node.
2440* ``shaper_profile_id``: Shaper profile ID of the private shaper to be used by
2441  the node.
2442* ``n_sp_priorities``: Number of strict priorities.
2443* ``stats_mask``: Mask of statistics counter types to be enabled for this node.
2444* ``n_shared_shapers``: Number of shared shapers.
2445* ``shared_shaper_id``: Shared shaper id.
2446
2447Add port traffic management hierarchy leaf node
2448~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2449
2450Add leaf node to port traffic management hiearchy::
2451
2452   testpmd> add port tm leaf node (port_id) (node_id) (parent_node_id) \
2453   (priority) (weight) (level_id) (shaper_profile_id) \
2454   (cman_mode) (wred_profile_id) (stats_mask) (n_shared_shapers) \
2455   [(shared_shaper_id) (shared_shaper_id) ...] \
2456
2457where:
2458
2459* ``parent_node_id``: Node ID of the parent.
2460* ``priority``: Node priority (highest node priority is zero). This is used by
2461  the SP algorithm running on the parent node for scheduling this node.
2462* ``weight``: Node weight (lowest weight is one). The node weight is relative
2463  to the weight sum of all siblings that have the same priority. It is used by
2464  the WFQ algorithm running on the parent node for scheduling this node.
2465* ``level_id``: Hiearchy level of the node.
2466* ``shaper_profile_id``: Shaper profile ID of the private shaper to be used by
2467  the node.
2468* ``cman_mode``: Congestion management mode to be enabled for this node.
2469* ``wred_profile_id``: WRED profile id to be enabled for this node.
2470* ``stats_mask``: Mask of statistics counter types to be enabled for this node.
2471* ``n_shared_shapers``: Number of shared shapers.
2472* ``shared_shaper_id``: Shared shaper id.
2473
2474Delete port traffic management hierarchy node
2475~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2476
2477Delete node from port traffic management hiearchy::
2478
2479   testpmd> del port tm node (port_id) (node_id)
2480
2481Update port traffic management hierarchy parent node
2482~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2483
2484Update port traffic management hierarchy parent node::
2485
2486   testpmd> set port tm node parent (port_id) (node_id) (parent_node_id) \
2487   (priority) (weight)
2488
2489This function can only be called after the hierarchy commit invocation. Its
2490success depends on the port support for this operation, as advertised through
2491the port capability set. This function is valid for all nodes of the traffic
2492management hierarchy except root node.
2493
2494Commit port traffic management hierarchy
2495~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2496
2497Commit the traffic management hierarchy on the port::
2498
2499   testpmd> port tm hierarchy commit (port_id) (clean_on_fail)
2500
2501where:
2502
2503* ``clean_on_fail``: When set to non-zero, hierarchy is cleared on function
2504  call failure. On the other hand, hierarchy is preserved when this parameter
2505  is equal to zero.
2506
2507Set port traffic management default hierarchy (tm forwarding mode)
2508~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2509
2510set the traffic management default hierarchy on the port::
2511
2512   testpmd> set port tm hierarchy default (port_id)
2513
2514Filter Functions
2515----------------
2516
2517This section details the available filter functions that are available.
2518
2519Note these functions interface the deprecated legacy filtering framework,
2520superseded by *rte_flow*. See `Flow rules management`_.
2521
2522ethertype_filter
2523~~~~~~~~~~~~~~~~~~~~
2524
2525Add or delete a L2 Ethertype filter, which identify packets by their L2 Ethertype mainly assign them to a receive queue::
2526
2527   ethertype_filter (port_id) (add|del) (mac_addr|mac_ignr) (mac_address) \
2528                    ethertype (ether_type) (drop|fwd) queue (queue_id)
2529
2530The available information parameters are:
2531
2532* ``port_id``: The port which the Ethertype filter assigned on.
2533
2534* ``mac_addr``: Compare destination mac address.
2535
2536* ``mac_ignr``: Ignore destination mac address match.
2537
2538* ``mac_address``: Destination mac address to match.
2539
2540* ``ether_type``: The EtherType value want to match,
2541  for example 0x0806 for ARP packet. 0x0800 (IPv4) and 0x86DD (IPv6) are invalid.
2542
2543* ``queue_id``: The receive queue associated with this EtherType filter.
2544  It is meaningless when deleting or dropping.
2545
2546Example, to add/remove an ethertype filter rule::
2547
2548   testpmd> ethertype_filter 0 add mac_ignr 00:11:22:33:44:55 \
2549                             ethertype 0x0806 fwd queue 3
2550
2551   testpmd> ethertype_filter 0 del mac_ignr 00:11:22:33:44:55 \
2552                             ethertype 0x0806 fwd queue 3
2553
25542tuple_filter
2555~~~~~~~~~~~~~~~~~
2556
2557Add or delete a 2-tuple filter,
2558which identifies packets by specific protocol and destination TCP/UDP port
2559and forwards packets into one of the receive queues::
2560
2561   2tuple_filter (port_id) (add|del) dst_port (dst_port_value) \
2562                 protocol (protocol_value) mask (mask_value) \
2563                 tcp_flags (tcp_flags_value) priority (prio_value) \
2564                 queue (queue_id)
2565
2566The available information parameters are:
2567
2568* ``port_id``: The port which the 2-tuple filter assigned on.
2569
2570* ``dst_port_value``: Destination port in L4.
2571
2572* ``protocol_value``: IP L4 protocol.
2573
2574* ``mask_value``: Participates in the match or not by bit for field above, 1b means participate.
2575
2576* ``tcp_flags_value``: TCP control bits. The non-zero value is invalid, when the pro_value is not set to 0x06 (TCP).
2577
2578* ``prio_value``: Priority of this filter.
2579
2580* ``queue_id``: The receive queue associated with this 2-tuple filter.
2581
2582Example, to add/remove an 2tuple filter rule::
2583
2584   testpmd> 2tuple_filter 0 add dst_port 32 protocol 0x06 mask 0x03 \
2585                          tcp_flags 0x02 priority 3 queue 3
2586
2587   testpmd> 2tuple_filter 0 del dst_port 32 protocol 0x06 mask 0x03 \
2588                          tcp_flags 0x02 priority 3 queue 3
2589
25905tuple_filter
2591~~~~~~~~~~~~~~~~~
2592
2593Add or delete a 5-tuple filter,
2594which consists of a 5-tuple (protocol, source and destination IP addresses, source and destination TCP/UDP/SCTP port)
2595and routes packets into one of the receive queues::
2596
2597   5tuple_filter (port_id) (add|del) dst_ip (dst_address) src_ip \
2598                 (src_address) dst_port (dst_port_value) \
2599                 src_port (src_port_value) protocol (protocol_value) \
2600                 mask (mask_value) tcp_flags (tcp_flags_value) \
2601                 priority (prio_value) queue (queue_id)
2602
2603The available information parameters are:
2604
2605* ``port_id``: The port which the 5-tuple filter assigned on.
2606
2607* ``dst_address``: Destination IP address.
2608
2609* ``src_address``: Source IP address.
2610
2611* ``dst_port_value``: TCP/UDP destination port.
2612
2613* ``src_port_value``: TCP/UDP source port.
2614
2615* ``protocol_value``: L4 protocol.
2616
2617* ``mask_value``: Participates in the match or not by bit for field above, 1b means participate
2618
2619* ``tcp_flags_value``: TCP control bits. The non-zero value is invalid, when the protocol_value is not set to 0x06 (TCP).
2620
2621* ``prio_value``: The priority of this filter.
2622
2623* ``queue_id``: The receive queue associated with this 5-tuple filter.
2624
2625Example, to add/remove an 5tuple filter rule::
2626
2627   testpmd> 5tuple_filter 0 add dst_ip 2.2.2.5 src_ip 2.2.2.4 \
2628            dst_port 64 src_port 32 protocol 0x06 mask 0x1F \
2629            flags 0x0 priority 3 queue 3
2630
2631   testpmd> 5tuple_filter 0 del dst_ip 2.2.2.5 src_ip 2.2.2.4 \
2632            dst_port 64 src_port 32 protocol 0x06 mask 0x1F \
2633            flags 0x0 priority 3 queue 3
2634
2635syn_filter
2636~~~~~~~~~~
2637
2638Using the  SYN filter, TCP packets whose *SYN* flag is set can be forwarded to a separate queue::
2639
2640   syn_filter (port_id) (add|del) priority (high|low) queue (queue_id)
2641
2642The available information parameters are:
2643
2644* ``port_id``: The port which the SYN filter assigned on.
2645
2646* ``high``: This SYN filter has higher priority than other filters.
2647
2648* ``low``: This SYN filter has lower priority than other filters.
2649
2650* ``queue_id``: The receive queue associated with this SYN filter
2651
2652Example::
2653
2654   testpmd> syn_filter 0 add priority high queue 3
2655
2656flex_filter
2657~~~~~~~~~~~
2658
2659With flex filter, packets can be recognized by any arbitrary pattern within the first 128 bytes of the packet
2660and routed into one of the receive queues::
2661
2662   flex_filter (port_id) (add|del) len (len_value) bytes (bytes_value) \
2663               mask (mask_value) priority (prio_value) queue (queue_id)
2664
2665The available information parameters are:
2666
2667* ``port_id``: The port which the Flex filter is assigned on.
2668
2669* ``len_value``: Filter length in bytes, no greater than 128.
2670
2671* ``bytes_value``: A string in hexadecimal, means the value the flex filter needs to match.
2672
2673* ``mask_value``: A string in hexadecimal, bit 1 means corresponding byte participates in the match.
2674
2675* ``prio_value``: The priority of this filter.
2676
2677* ``queue_id``: The receive queue associated with this Flex filter.
2678
2679Example::
2680
2681   testpmd> flex_filter 0 add len 16 bytes 0x00000000000000000000000008060000 \
2682                          mask 000C priority 3 queue 3
2683
2684   testpmd> flex_filter 0 del len 16 bytes 0x00000000000000000000000008060000 \
2685                          mask 000C priority 3 queue 3
2686
2687
2688.. _testpmd_flow_director:
2689
2690flow_director_filter
2691~~~~~~~~~~~~~~~~~~~~
2692
2693The Flow Director works in receive mode to identify specific flows or sets of flows and route them to specific queues.
2694
2695Four types of filtering are supported which are referred to as Perfect Match, Signature, Perfect-mac-vlan and
2696Perfect-tunnel filters, the match mode is set by the ``--pkt-filter-mode`` command-line parameter:
2697
2698* Perfect match filters.
2699  The hardware checks a match between the masked fields of the received packets and the programmed filters.
2700  The masked fields are for IP flow.
2701
2702* Signature filters.
2703  The hardware checks a match between a hash-based signature of the masked fields of the received packet.
2704
2705* Perfect-mac-vlan match filters.
2706  The hardware checks a match between the masked fields of the received packets and the programmed filters.
2707  The masked fields are for MAC VLAN flow.
2708
2709* Perfect-tunnel match filters.
2710  The hardware checks a match between the masked fields of the received packets and the programmed filters.
2711  The masked fields are for tunnel flow.
2712
2713* Perfect-raw-flow-type match filters.
2714  The hardware checks a match between the masked fields of the received packets and pre-loaded raw (template) packet.
2715  The masked fields are specified by input sets.
2716
2717The Flow Director filters can match the different fields for different type of packet: flow type, specific input set
2718per flow type and the flexible payload.
2719
2720The Flow Director can also mask out parts of all of these fields so that filters
2721are only applied to certain fields or parts of the fields.
2722
2723Note that for raw flow type mode the source and destination fields in the
2724raw packet buffer need to be presented in a reversed order with respect
2725to the expected received packets.
2726For example: IP source and destination addresses or TCP/UDP/SCTP
2727source and destination ports
2728
2729Different NICs may have different capabilities, command show port fdir (port_id) can be used to acquire the information.
2730
2731# Commands to add flow director filters of different flow types::
2732
2733   flow_director_filter (port_id) mode IP (add|del|update) \
2734                        flow (ipv4-other|ipv4-frag|ipv6-other|ipv6-frag) \
2735                        src (src_ip_address) dst (dst_ip_address) \
2736                        tos (tos_value) proto (proto_value) ttl (ttl_value) \
2737                        vlan (vlan_value) flexbytes (flexbytes_value) \
2738                        (drop|fwd) pf|vf(vf_id) queue (queue_id) \
2739                        fd_id (fd_id_value)
2740
2741   flow_director_filter (port_id) mode IP (add|del|update) \
2742                        flow (ipv4-tcp|ipv4-udp|ipv6-tcp|ipv6-udp) \
2743                        src (src_ip_address) (src_port) \
2744                        dst (dst_ip_address) (dst_port) \
2745                        tos (tos_value) ttl (ttl_value) \
2746                        vlan (vlan_value) flexbytes (flexbytes_value) \
2747                        (drop|fwd) queue pf|vf(vf_id) (queue_id) \
2748                        fd_id (fd_id_value)
2749
2750   flow_director_filter (port_id) mode IP (add|del|update) \
2751                        flow (ipv4-sctp|ipv6-sctp) \
2752                        src (src_ip_address) (src_port) \
2753                        dst (dst_ip_address) (dst_port) \
2754                        tos (tos_value) ttl (ttl_value) \
2755                        tag (verification_tag) vlan (vlan_value) \
2756                        flexbytes (flexbytes_value) (drop|fwd) \
2757                        pf|vf(vf_id) queue (queue_id) fd_id (fd_id_value)
2758
2759   flow_director_filter (port_id) mode IP (add|del|update) flow l2_payload \
2760                        ether (ethertype) flexbytes (flexbytes_value) \
2761                        (drop|fwd) pf|vf(vf_id) queue (queue_id)
2762                        fd_id (fd_id_value)
2763
2764   flow_director_filter (port_id) mode MAC-VLAN (add|del|update) \
2765                        mac (mac_address) vlan (vlan_value) \
2766                        flexbytes (flexbytes_value) (drop|fwd) \
2767                        queue (queue_id) fd_id (fd_id_value)
2768
2769   flow_director_filter (port_id) mode Tunnel (add|del|update) \
2770                        mac (mac_address) vlan (vlan_value) \
2771                        tunnel (NVGRE|VxLAN) tunnel-id (tunnel_id_value) \
2772                        flexbytes (flexbytes_value) (drop|fwd) \
2773                        queue (queue_id) fd_id (fd_id_value)
2774
2775   flow_director_filter (port_id) mode raw (add|del|update) flow (flow_id) \
2776                        (drop|fwd) queue (queue_id) fd_id (fd_id_value) \
2777                        packet (packet file name)
2778
2779For example, to add an ipv4-udp flow type filter::
2780
2781   testpmd> flow_director_filter 0 mode IP add flow ipv4-udp src 2.2.2.3 32 \
2782            dst 2.2.2.5 33 tos 2 ttl 40 vlan 0x1 flexbytes (0x88,0x48) \
2783            fwd pf queue 1 fd_id 1
2784
2785For example, add an ipv4-other flow type filter::
2786
2787   testpmd> flow_director_filter 0 mode IP add flow ipv4-other src 2.2.2.3 \
2788             dst 2.2.2.5 tos 2 proto 20 ttl 40 vlan 0x1 \
2789             flexbytes (0x88,0x48) fwd pf queue 1 fd_id 1
2790
2791flush_flow_director
2792~~~~~~~~~~~~~~~~~~~
2793
2794Flush all flow director filters on a device::
2795
2796   testpmd> flush_flow_director (port_id)
2797
2798Example, to flush all flow director filter on port 0::
2799
2800   testpmd> flush_flow_director 0
2801
2802flow_director_mask
2803~~~~~~~~~~~~~~~~~~
2804
2805Set flow director's input masks::
2806
2807   flow_director_mask (port_id) mode IP vlan (vlan_value) \
2808                      src_mask (ipv4_src) (ipv6_src) (src_port) \
2809                      dst_mask (ipv4_dst) (ipv6_dst) (dst_port)
2810
2811   flow_director_mask (port_id) mode MAC-VLAN vlan (vlan_value)
2812
2813   flow_director_mask (port_id) mode Tunnel vlan (vlan_value) \
2814                      mac (mac_value) tunnel-type (tunnel_type_value) \
2815                      tunnel-id (tunnel_id_value)
2816
2817Example, to set flow director mask on port 0::
2818
2819   testpmd> flow_director_mask 0 mode IP vlan 0xefff \
2820            src_mask 255.255.255.255 \
2821                FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF \
2822            dst_mask 255.255.255.255 \
2823                FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF
2824
2825flow_director_flex_mask
2826~~~~~~~~~~~~~~~~~~~~~~~
2827
2828set masks of flow director's flexible payload based on certain flow type::
2829
2830   testpmd> flow_director_flex_mask (port_id) \
2831            flow (none|ipv4-other|ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
2832                  ipv6-other|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp| \
2833                  l2_payload|all) (mask)
2834
2835Example, to set flow director's flex mask for all flow type on port 0::
2836
2837   testpmd> flow_director_flex_mask 0 flow all \
2838            (0xff,0xff,0,0,0,0,0,0,0,0,0,0,0,0,0,0)
2839
2840
2841flow_director_flex_payload
2842~~~~~~~~~~~~~~~~~~~~~~~~~~
2843
2844Configure flexible payload selection::
2845
2846   flow_director_flex_payload (port_id) (raw|l2|l3|l4) (config)
2847
2848For example, to select the first 16 bytes from the offset 4 (bytes) of packet's payload as flexible payload::
2849
2850   testpmd> flow_director_flex_payload 0 l4 \
2851            (4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19)
2852
2853get_sym_hash_ena_per_port
2854~~~~~~~~~~~~~~~~~~~~~~~~~
2855
2856Get symmetric hash enable configuration per port::
2857
2858   get_sym_hash_ena_per_port (port_id)
2859
2860For example, to get symmetric hash enable configuration of port 1::
2861
2862   testpmd> get_sym_hash_ena_per_port 1
2863
2864set_sym_hash_ena_per_port
2865~~~~~~~~~~~~~~~~~~~~~~~~~
2866
2867Set symmetric hash enable configuration per port to enable or disable::
2868
2869   set_sym_hash_ena_per_port (port_id) (enable|disable)
2870
2871For example, to set symmetric hash enable configuration of port 1 to enable::
2872
2873   testpmd> set_sym_hash_ena_per_port 1 enable
2874
2875get_hash_global_config
2876~~~~~~~~~~~~~~~~~~~~~~
2877
2878Get the global configurations of hash filters::
2879
2880   get_hash_global_config (port_id)
2881
2882For example, to get the global configurations of hash filters of port 1::
2883
2884   testpmd> get_hash_global_config 1
2885
2886set_hash_global_config
2887~~~~~~~~~~~~~~~~~~~~~~
2888
2889Set the global configurations of hash filters::
2890
2891   set_hash_global_config (port_id) (toeplitz|simple_xor|default) \
2892   (ipv4|ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp|ipv4-other|ipv6|ipv6-frag| \
2893   ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other|l2_payload|<flow_id>) \
2894   (enable|disable)
2895
2896For example, to enable simple_xor for flow type of ipv6 on port 2::
2897
2898   testpmd> set_hash_global_config 2 simple_xor ipv6 enable
2899
2900set_hash_input_set
2901~~~~~~~~~~~~~~~~~~
2902
2903Set the input set for hash::
2904
2905   set_hash_input_set (port_id) (ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
2906   ipv4-other|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other| \
2907   l2_payload|<flow_id>) (ovlan|ivlan|src-ipv4|dst-ipv4|src-ipv6|dst-ipv6| \
2908   ipv4-tos|ipv4-proto|ipv6-tc|ipv6-next-header|udp-src-port|udp-dst-port| \
2909   tcp-src-port|tcp-dst-port|sctp-src-port|sctp-dst-port|sctp-veri-tag| \
2910   udp-key|gre-key|fld-1st|fld-2nd|fld-3rd|fld-4th|fld-5th|fld-6th|fld-7th| \
2911   fld-8th|none) (select|add)
2912
2913For example, to add source IP to hash input set for flow type of ipv4-udp on port 0::
2914
2915   testpmd> set_hash_input_set 0 ipv4-udp src-ipv4 add
2916
2917set_fdir_input_set
2918~~~~~~~~~~~~~~~~~~
2919
2920The Flow Director filters can match the different fields for different type of packet, i.e. specific input set
2921on per flow type and the flexible payload. This command can be used to change input set for each flow type.
2922
2923Set the input set for flow director::
2924
2925   set_fdir_input_set (port_id) (ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
2926   ipv4-other|ipv6|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other| \
2927   l2_payload|<flow_id>) (ivlan|ethertype|src-ipv4|dst-ipv4|src-ipv6|dst-ipv6| \
2928   ipv4-tos|ipv4-proto|ipv4-ttl|ipv6-tc|ipv6-next-header|ipv6-hop-limits| \
2929   tudp-src-port|udp-dst-port|cp-src-port|tcp-dst-port|sctp-src-port| \
2930   sctp-dst-port|sctp-veri-tag|none) (select|add)
2931
2932For example to add source IP to FD input set for flow type of ipv4-udp on port 0::
2933
2934   testpmd> set_fdir_input_set 0 ipv4-udp src-ipv4 add
2935
2936global_config
2937~~~~~~~~~~~~~
2938
2939Set different GRE key length for input set::
2940
2941   global_config (port_id) gre-key-len (number in bytes)
2942
2943For example to set GRE key length for input set to 4 bytes on port 0::
2944
2945   testpmd> global_config 0 gre-key-len 4
2946
2947
2948.. _testpmd_rte_flow:
2949
2950Flow rules management
2951---------------------
2952
2953Control of the generic flow API (*rte_flow*) is fully exposed through the
2954``flow`` command (validation, creation, destruction, queries and operation
2955modes).
2956
2957Considering *rte_flow* overlaps with all `Filter Functions`_, using both
2958features simultaneously may cause undefined side-effects and is therefore
2959not recommended.
2960
2961``flow`` syntax
2962~~~~~~~~~~~~~~~
2963
2964Because the ``flow`` command uses dynamic tokens to handle the large number
2965of possible flow rules combinations, its behavior differs slightly from
2966other commands, in particular:
2967
2968- Pressing *?* or the *<tab>* key displays contextual help for the current
2969  token, not that of the entire command.
2970
2971- Optional and repeated parameters are supported (provided they are listed
2972  in the contextual help).
2973
2974The first parameter stands for the operation mode. Possible operations and
2975their general syntax are described below. They are covered in detail in the
2976following sections.
2977
2978- Check whether a flow rule can be created::
2979
2980   flow validate {port_id}
2981       [group {group_id}] [priority {level}] [ingress] [egress]
2982       pattern {item} [/ {item} [...]] / end
2983       actions {action} [/ {action} [...]] / end
2984
2985- Create a flow rule::
2986
2987   flow create {port_id}
2988       [group {group_id}] [priority {level}] [ingress] [egress]
2989       pattern {item} [/ {item} [...]] / end
2990       actions {action} [/ {action} [...]] / end
2991
2992- Destroy specific flow rules::
2993
2994   flow destroy {port_id} rule {rule_id} [...]
2995
2996- Destroy all flow rules::
2997
2998   flow flush {port_id}
2999
3000- Query an existing flow rule::
3001
3002   flow query {port_id} {rule_id} {action}
3003
3004- List existing flow rules sorted by priority, filtered by group
3005  identifiers::
3006
3007   flow list {port_id} [group {group_id}] [...]
3008
3009- Restrict ingress traffic to the defined flow rules::
3010
3011   flow isolate {port_id} {boolean}
3012
3013Validating flow rules
3014~~~~~~~~~~~~~~~~~~~~~
3015
3016``flow validate`` reports whether a flow rule would be accepted by the
3017underlying device in its current state but stops short of creating it. It is
3018bound to ``rte_flow_validate()``::
3019
3020   flow validate {port_id}
3021      [group {group_id}] [priority {level}] [ingress] [egress]
3022      pattern {item} [/ {item} [...]] / end
3023      actions {action} [/ {action} [...]] / end
3024
3025If successful, it will show::
3026
3027   Flow rule validated
3028
3029Otherwise it will show an error message of the form::
3030
3031   Caught error type [...] ([...]): [...]
3032
3033This command uses the same parameters as ``flow create``, their format is
3034described in `Creating flow rules`_.
3035
3036Check whether redirecting any Ethernet packet received on port 0 to RX queue
3037index 6 is supported::
3038
3039   testpmd> flow validate 0 ingress pattern eth / end
3040      actions queue index 6 / end
3041   Flow rule validated
3042   testpmd>
3043
3044Port 0 does not support TCPv6 rules::
3045
3046   testpmd> flow validate 0 ingress pattern eth / ipv6 / tcp / end
3047      actions drop / end
3048   Caught error type 9 (specific pattern item): Invalid argument
3049   testpmd>
3050
3051Creating flow rules
3052~~~~~~~~~~~~~~~~~~~
3053
3054``flow create`` validates and creates the specified flow rule. It is bound
3055to ``rte_flow_create()``::
3056
3057   flow create {port_id}
3058      [group {group_id}] [priority {level}] [ingress] [egress]
3059      pattern {item} [/ {item} [...]] / end
3060      actions {action} [/ {action} [...]] / end
3061
3062If successful, it will return a flow rule ID usable with other commands::
3063
3064   Flow rule #[...] created
3065
3066Otherwise it will show an error message of the form::
3067
3068   Caught error type [...] ([...]): [...]
3069
3070Parameters describe in the following order:
3071
3072- Attributes (*group*, *priority*, *ingress*, *egress* tokens).
3073- A matching pattern, starting with the *pattern* token and terminated by an
3074  *end* pattern item.
3075- Actions, starting with the *actions* token and terminated by an *end*
3076  action.
3077
3078These translate directly to *rte_flow* objects provided as-is to the
3079underlying functions.
3080
3081The shortest valid definition only comprises mandatory tokens::
3082
3083   testpmd> flow create 0 pattern end actions end
3084
3085Note that PMDs may refuse rules that essentially do nothing such as this
3086one.
3087
3088**All unspecified object values are automatically initialized to 0.**
3089
3090Attributes
3091^^^^^^^^^^
3092
3093These tokens affect flow rule attributes (``struct rte_flow_attr``) and are
3094specified before the ``pattern`` token.
3095
3096- ``group {group id}``: priority group.
3097- ``priority {level}``: priority level within group.
3098- ``ingress``: rule applies to ingress traffic.
3099- ``egress``: rule applies to egress traffic.
3100
3101Each instance of an attribute specified several times overrides the previous
3102value as shown below (group 4 is used)::
3103
3104   testpmd> flow create 0 group 42 group 24 group 4 [...]
3105
3106Note that once enabled, ``ingress`` and ``egress`` cannot be disabled.
3107
3108While not specifying a direction is an error, some rules may allow both
3109simultaneously.
3110
3111Most rules affect RX therefore contain the ``ingress`` token::
3112
3113   testpmd> flow create 0 ingress pattern [...]
3114
3115Matching pattern
3116^^^^^^^^^^^^^^^^
3117
3118A matching pattern starts after the ``pattern`` token. It is made of pattern
3119items and is terminated by a mandatory ``end`` item.
3120
3121Items are named after their type (*RTE_FLOW_ITEM_TYPE_* from ``enum
3122rte_flow_item_type``).
3123
3124The ``/`` token is used as a separator between pattern items as shown
3125below::
3126
3127   testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end [...]
3128
3129Note that protocol items like these must be stacked from lowest to highest
3130layer to make sense. For instance, the following rule is either invalid or
3131unlikely to match any packet::
3132
3133   testpmd> flow create 0 ingress pattern eth / udp / ipv4 / end [...]
3134
3135More information on these restrictions can be found in the *rte_flow*
3136documentation.
3137
3138Several items support additional specification structures, for example
3139``ipv4`` allows specifying source and destination addresses as follows::
3140
3141   testpmd> flow create 0 ingress pattern eth / ipv4 src is 10.1.1.1
3142      dst is 10.2.0.0 / end [...]
3143
3144This rule matches all IPv4 traffic with the specified properties.
3145
3146In this example, ``src`` and ``dst`` are field names of the underlying
3147``struct rte_flow_item_ipv4`` object. All item properties can be specified
3148in a similar fashion.
3149
3150The ``is`` token means that the subsequent value must be matched exactly,
3151and assigns ``spec`` and ``mask`` fields in ``struct rte_flow_item``
3152accordingly. Possible assignment tokens are:
3153
3154- ``is``: match value perfectly (with full bit-mask).
3155- ``spec``: match value according to configured bit-mask.
3156- ``last``: specify upper bound to establish a range.
3157- ``mask``: specify bit-mask with relevant bits set to one.
3158- ``prefix``: generate bit-mask from a prefix length.
3159
3160These yield identical results::
3161
3162   ipv4 src is 10.1.1.1
3163
3164::
3165
3166   ipv4 src spec 10.1.1.1 src mask 255.255.255.255
3167
3168::
3169
3170   ipv4 src spec 10.1.1.1 src prefix 32
3171
3172::
3173
3174   ipv4 src is 10.1.1.1 src last 10.1.1.1 # range with a single value
3175
3176::
3177
3178   ipv4 src is 10.1.1.1 src last 0 # 0 disables range
3179
3180Inclusive ranges can be defined with ``last``::
3181
3182   ipv4 src is 10.1.1.1 src last 10.2.3.4 # 10.1.1.1 to 10.2.3.4
3183
3184Note that ``mask`` affects both ``spec`` and ``last``::
3185
3186   ipv4 src is 10.1.1.1 src last 10.2.3.4 src mask 255.255.0.0
3187      # matches 10.1.0.0 to 10.2.255.255
3188
3189Properties can be modified multiple times::
3190
3191   ipv4 src is 10.1.1.1 src is 10.1.2.3 src is 10.2.3.4 # matches 10.2.3.4
3192
3193::
3194
3195   ipv4 src is 10.1.1.1 src prefix 24 src prefix 16 # matches 10.1.0.0/16
3196
3197Pattern items
3198^^^^^^^^^^^^^
3199
3200This section lists supported pattern items and their attributes, if any.
3201
3202- ``end``: end list of pattern items.
3203
3204- ``void``: no-op pattern item.
3205
3206- ``invert``: perform actions when pattern does not match.
3207
3208- ``any``: match any protocol for the current layer.
3209
3210  - ``num {unsigned}``: number of layers covered.
3211
3212- ``pf``: match packets addressed to the physical function.
3213
3214- ``vf``: match packets addressed to a virtual function ID.
3215
3216  - ``id {unsigned}``: destination VF ID.
3217
3218- ``port``: device-specific physical port index to use.
3219
3220  - ``index {unsigned}``: physical port index.
3221
3222- ``raw``: match an arbitrary byte string.
3223
3224  - ``relative {boolean}``: look for pattern after the previous item.
3225  - ``search {boolean}``: search pattern from offset (see also limit).
3226  - ``offset {integer}``: absolute or relative offset for pattern.
3227  - ``limit {unsigned}``: search area limit for start of pattern.
3228  - ``pattern {string}``: byte string to look for.
3229
3230- ``eth``: match Ethernet header.
3231
3232  - ``dst {MAC-48}``: destination MAC.
3233  - ``src {MAC-48}``: source MAC.
3234  - ``type {unsigned}``: EtherType.
3235
3236- ``vlan``: match 802.1Q/ad VLAN tag.
3237
3238  - ``tpid {unsigned}``: tag protocol identifier.
3239  - ``tci {unsigned}``: tag control information.
3240  - ``pcp {unsigned}``: priority code point.
3241  - ``dei {unsigned}``: drop eligible indicator.
3242  - ``vid {unsigned}``: VLAN identifier.
3243
3244- ``ipv4``: match IPv4 header.
3245
3246  - ``tos {unsigned}``: type of service.
3247  - ``ttl {unsigned}``: time to live.
3248  - ``proto {unsigned}``: next protocol ID.
3249  - ``src {ipv4 address}``: source address.
3250  - ``dst {ipv4 address}``: destination address.
3251
3252- ``ipv6``: match IPv6 header.
3253
3254  - ``tc {unsigned}``: traffic class.
3255  - ``flow {unsigned}``: flow label.
3256  - ``proto {unsigned}``: protocol (next header).
3257  - ``hop {unsigned}``: hop limit.
3258  - ``src {ipv6 address}``: source address.
3259  - ``dst {ipv6 address}``: destination address.
3260
3261- ``icmp``: match ICMP header.
3262
3263  - ``type {unsigned}``: ICMP packet type.
3264  - ``code {unsigned}``: ICMP packet code.
3265
3266- ``udp``: match UDP header.
3267
3268  - ``src {unsigned}``: UDP source port.
3269  - ``dst {unsigned}``: UDP destination port.
3270
3271- ``tcp``: match TCP header.
3272
3273  - ``src {unsigned}``: TCP source port.
3274  - ``dst {unsigned}``: TCP destination port.
3275
3276- ``sctp``: match SCTP header.
3277
3278  - ``src {unsigned}``: SCTP source port.
3279  - ``dst {unsigned}``: SCTP destination port.
3280  - ``tag {unsigned}``: validation tag.
3281  - ``cksum {unsigned}``: checksum.
3282
3283- ``vxlan``: match VXLAN header.
3284
3285  - ``vni {unsigned}``: VXLAN identifier.
3286
3287- ``e_tag``: match IEEE 802.1BR E-Tag header.
3288
3289  - ``grp_ecid_b {unsigned}``: GRP and E-CID base.
3290
3291- ``nvgre``: match NVGRE header.
3292
3293  - ``tni {unsigned}``: virtual subnet ID.
3294
3295- ``mpls``: match MPLS header.
3296
3297  - ``label {unsigned}``: MPLS label.
3298
3299- ``gre``: match GRE header.
3300
3301  - ``protocol {unsigned}``: protocol type.
3302
3303- ``fuzzy``: fuzzy pattern match, expect faster than default.
3304
3305  - ``thresh {unsigned}``: accuracy threshold.
3306
3307- ``gtp``, ``gtpc``, ``gtpu``: match GTPv1 header.
3308
3309  - ``teid {unsigned}``: tunnel endpoint identifier.
3310
3311- ``geneve``: match GENEVE header.
3312
3313  - ``vni {unsigned}``: virtual network identifier.
3314  - ``protocol {unsigned}``: protocol type.
3315
3316Actions list
3317^^^^^^^^^^^^
3318
3319A list of actions starts after the ``actions`` token in the same fashion as
3320`Matching pattern`_; actions are separated by ``/`` tokens and the list is
3321terminated by a mandatory ``end`` action.
3322
3323Actions are named after their type (*RTE_FLOW_ACTION_TYPE_* from ``enum
3324rte_flow_action_type``).
3325
3326Dropping all incoming UDPv4 packets can be expressed as follows::
3327
3328   testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
3329      actions drop / end
3330
3331Several actions have configurable properties which must be specified when
3332there is no valid default value. For example, ``queue`` requires a target
3333queue index.
3334
3335This rule redirects incoming UDPv4 traffic to queue index 6::
3336
3337   testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
3338      actions queue index 6 / end
3339
3340While this one could be rejected by PMDs (unspecified queue index)::
3341
3342   testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
3343      actions queue / end
3344
3345As defined by *rte_flow*, the list is not ordered, all actions of a given
3346rule are performed simultaneously. These are equivalent::
3347
3348   queue index 6 / void / mark id 42 / end
3349
3350::
3351
3352   void / mark id 42 / queue index 6 / end
3353
3354All actions in a list should have different types, otherwise only the last
3355action of a given type is taken into account::
3356
3357   queue index 4 / queue index 5 / queue index 6 / end # will use queue 6
3358
3359::
3360
3361   drop / drop / drop / end # drop is performed only once
3362
3363::
3364
3365   mark id 42 / queue index 3 / mark id 24 / end # mark will be 24
3366
3367Considering they are performed simultaneously, opposite and overlapping
3368actions can sometimes be combined when the end result is unambiguous::
3369
3370   drop / queue index 6 / end # drop has no effect
3371
3372::
3373
3374   drop / dup index 6 / end # same as above
3375
3376::
3377
3378   queue index 6 / rss queues 6 7 8 / end # queue has no effect
3379
3380::
3381
3382   drop / passthru / end # drop has no effect
3383
3384Note that PMDs may still refuse such combinations.
3385
3386Actions
3387^^^^^^^
3388
3389This section lists supported actions and their attributes, if any.
3390
3391- ``end``: end list of actions.
3392
3393- ``void``: no-op action.
3394
3395- ``passthru``: let subsequent rule process matched packets.
3396
3397- ``mark``: attach 32 bit value to packets.
3398
3399  - ``id {unsigned}``: 32 bit value to return with packets.
3400
3401- ``flag``: flag packets.
3402
3403- ``queue``: assign packets to a given queue index.
3404
3405  - ``index {unsigned}``: queue index to use.
3406
3407- ``drop``: drop packets (note: passthru has priority).
3408
3409- ``count``: enable counters for this rule.
3410
3411- ``dup``: duplicate packets to a given queue index.
3412
3413  - ``index {unsigned}``: queue index to duplicate packets to.
3414
3415- ``rss``: spread packets among several queues.
3416
3417  - ``queues [{unsigned} [...]] end``: queue indices to use.
3418
3419- ``pf``: redirect packets to physical device function.
3420
3421- ``vf``: redirect packets to virtual device function.
3422
3423  - ``original {boolean}``: use original VF ID if possible.
3424  - ``id {unsigned}``: VF ID to redirect packets to.
3425
3426Destroying flow rules
3427~~~~~~~~~~~~~~~~~~~~~
3428
3429``flow destroy`` destroys one or more rules from their rule ID (as returned
3430by ``flow create``), this command calls ``rte_flow_destroy()`` as many
3431times as necessary::
3432
3433   flow destroy {port_id} rule {rule_id} [...]
3434
3435If successful, it will show::
3436
3437   Flow rule #[...] destroyed
3438
3439It does not report anything for rule IDs that do not exist. The usual error
3440message is shown when a rule cannot be destroyed::
3441
3442   Caught error type [...] ([...]): [...]
3443
3444``flow flush`` destroys all rules on a device and does not take extra
3445arguments. It is bound to ``rte_flow_flush()``::
3446
3447   flow flush {port_id}
3448
3449Any errors are reported as above.
3450
3451Creating several rules and destroying them::
3452
3453   testpmd> flow create 0 ingress pattern eth / ipv6 / end
3454      actions queue index 2 / end
3455   Flow rule #0 created
3456   testpmd> flow create 0 ingress pattern eth / ipv4 / end
3457      actions queue index 3 / end
3458   Flow rule #1 created
3459   testpmd> flow destroy 0 rule 0 rule 1
3460   Flow rule #1 destroyed
3461   Flow rule #0 destroyed
3462   testpmd>
3463
3464The same result can be achieved using ``flow flush``::
3465
3466   testpmd> flow create 0 ingress pattern eth / ipv6 / end
3467      actions queue index 2 / end
3468   Flow rule #0 created
3469   testpmd> flow create 0 ingress pattern eth / ipv4 / end
3470      actions queue index 3 / end
3471   Flow rule #1 created
3472   testpmd> flow flush 0
3473   testpmd>
3474
3475Non-existent rule IDs are ignored::
3476
3477   testpmd> flow create 0 ingress pattern eth / ipv6 / end
3478      actions queue index 2 / end
3479   Flow rule #0 created
3480   testpmd> flow create 0 ingress pattern eth / ipv4 / end
3481      actions queue index 3 / end
3482   Flow rule #1 created
3483   testpmd> flow destroy 0 rule 42 rule 10 rule 2
3484   testpmd>
3485   testpmd> flow destroy 0 rule 0
3486   Flow rule #0 destroyed
3487   testpmd>
3488
3489Querying flow rules
3490~~~~~~~~~~~~~~~~~~~
3491
3492``flow query`` queries a specific action of a flow rule having that
3493ability. Such actions collect information that can be reported using this
3494command. It is bound to ``rte_flow_query()``::
3495
3496   flow query {port_id} {rule_id} {action}
3497
3498If successful, it will display either the retrieved data for known actions
3499or the following message::
3500
3501   Cannot display result for action type [...] ([...])
3502
3503Otherwise, it will complain either that the rule does not exist or that some
3504error occurred::
3505
3506   Flow rule #[...] not found
3507
3508::
3509
3510   Caught error type [...] ([...]): [...]
3511
3512Currently only the ``count`` action is supported. This action reports the
3513number of packets that hit the flow rule and the total number of bytes. Its
3514output has the following format::
3515
3516   count:
3517    hits_set: [...] # whether "hits" contains a valid value
3518    bytes_set: [...] # whether "bytes" contains a valid value
3519    hits: [...] # number of packets
3520    bytes: [...] # number of bytes
3521
3522Querying counters for TCPv6 packets redirected to queue 6::
3523
3524   testpmd> flow create 0 ingress pattern eth / ipv6 / tcp / end
3525      actions queue index 6 / count / end
3526   Flow rule #4 created
3527   testpmd> flow query 0 4 count
3528   count:
3529    hits_set: 1
3530    bytes_set: 0
3531    hits: 386446
3532    bytes: 0
3533   testpmd>
3534
3535Listing flow rules
3536~~~~~~~~~~~~~~~~~~
3537
3538``flow list`` lists existing flow rules sorted by priority and optionally
3539filtered by group identifiers::
3540
3541   flow list {port_id} [group {group_id}] [...]
3542
3543This command only fails with the following message if the device does not
3544exist::
3545
3546   Invalid port [...]
3547
3548Output consists of a header line followed by a short description of each
3549flow rule, one per line. There is no output at all when no flow rules are
3550configured on the device::
3551
3552   ID      Group   Prio    Attr    Rule
3553   [...]   [...]   [...]   [...]   [...]
3554
3555``Attr`` column flags:
3556
3557- ``i`` for ``ingress``.
3558- ``e`` for ``egress``.
3559
3560Creating several flow rules and listing them::
3561
3562   testpmd> flow create 0 ingress pattern eth / ipv4 / end
3563      actions queue index 6 / end
3564   Flow rule #0 created
3565   testpmd> flow create 0 ingress pattern eth / ipv6 / end
3566      actions queue index 2 / end
3567   Flow rule #1 created
3568   testpmd> flow create 0 priority 5 ingress pattern eth / ipv4 / udp / end
3569      actions rss queues 6 7 8 end / end
3570   Flow rule #2 created
3571   testpmd> flow list 0
3572   ID      Group   Prio    Attr    Rule
3573   0       0       0       i-      ETH IPV4 => QUEUE
3574   1       0       0       i-      ETH IPV6 => QUEUE
3575   2       0       5       i-      ETH IPV4 UDP => RSS
3576   testpmd>
3577
3578Rules are sorted by priority (i.e. group ID first, then priority level)::
3579
3580   testpmd> flow list 1
3581   ID      Group   Prio    Attr    Rule
3582   0       0       0       i-      ETH => COUNT
3583   6       0       500     i-      ETH IPV6 TCP => DROP COUNT
3584   5       0       1000    i-      ETH IPV6 ICMP => QUEUE
3585   1       24      0       i-      ETH IPV4 UDP => QUEUE
3586   4       24      10      i-      ETH IPV4 TCP => DROP
3587   3       24      20      i-      ETH IPV4 => DROP
3588   2       24      42      i-      ETH IPV4 UDP => QUEUE
3589   7       63      0       i-      ETH IPV6 UDP VXLAN => MARK QUEUE
3590   testpmd>
3591
3592Output can be limited to specific groups::
3593
3594   testpmd> flow list 1 group 0 group 63
3595   ID      Group   Prio    Attr    Rule
3596   0       0       0       i-      ETH => COUNT
3597   6       0       500     i-      ETH IPV6 TCP => DROP COUNT
3598   5       0       1000    i-      ETH IPV6 ICMP => QUEUE
3599   7       63      0       i-      ETH IPV6 UDP VXLAN => MARK QUEUE
3600   testpmd>
3601
3602Toggling isolated mode
3603~~~~~~~~~~~~~~~~~~~~~~
3604
3605``flow isolate`` can be used to tell the underlying PMD that ingress traffic
3606must only be injected from the defined flow rules; that no default traffic
3607is expected outside those rules and the driver is free to assign more
3608resources to handle them. It is bound to ``rte_flow_isolate()``::
3609
3610 flow isolate {port_id} {boolean}
3611
3612If successful, enabling or disabling isolated mode shows either::
3613
3614 Ingress traffic on port [...]
3615    is now restricted to the defined flow rules
3616
3617Or::
3618
3619 Ingress traffic on port [...]
3620    is not restricted anymore to the defined flow rules
3621
3622Otherwise, in case of error::
3623
3624   Caught error type [...] ([...]): [...]
3625
3626Mainly due to its side effects, PMDs supporting this mode may not have the
3627ability to toggle it more than once without reinitializing affected ports
3628first (e.g. by exiting testpmd).
3629
3630Enabling isolated mode::
3631
3632 testpmd> flow isolate 0 true
3633 Ingress traffic on port 0 is now restricted to the defined flow rules
3634 testpmd>
3635
3636Disabling isolated mode::
3637
3638 testpmd> flow isolate 0 false
3639 Ingress traffic on port 0 is not restricted anymore to the defined flow rules
3640 testpmd>
3641
3642Sample QinQ flow rules
3643~~~~~~~~~~~~~~~~~~~~~~
3644
3645Before creating QinQ rule(s) the following commands should be issued to enable QinQ::
3646
3647   testpmd> port stop 0
3648   testpmd> vlan set qinq on 0
3649
3650The above command sets the inner and outer TPID's to 0x8100.
3651
3652To change the TPID's the following commands should be used::
3653
3654   testpmd> vlan set outer tpid 0xa100 0
3655   testpmd> vlan set inner tpid 0x9100 0
3656   testpmd> port start 0
3657
3658Validate and create a QinQ rule on port 0 to steer traffic to a VF queue in a VM.
3659
3660::
3661
3662   testpmd> flow validate 0 ingress pattern eth / vlan tci is 123 /
3663       vlan tci is 456 / end actions vf id 1 / queue index 0 / end
3664   Flow rule #0 validated
3665
3666   testpmd> flow create 0 ingress pattern eth / vlan tci is 4 /
3667       vlan tci is 456 / end actions vf id 123 / queue index 0 / end
3668   Flow rule #0 created
3669
3670   testpmd> flow list 0
3671   ID      Group   Prio    Attr    Rule
3672   0       0       0       i-      ETH VLAN VLAN=>VF QUEUE
3673
3674Validate and create a QinQ rule on port 0 to steer traffic to a queue on the host.
3675
3676::
3677
3678   testpmd> flow validate 0 ingress pattern eth / vlan tci is 321 /
3679        vlan tci is 654 / end actions pf / queue index 0 / end
3680   Flow rule #1 validated
3681
3682   testpmd> flow create 0 ingress pattern eth / vlan tci is 321 /
3683        vlan tci is 654 / end actions pf / queue index 1 / end
3684   Flow rule #1 created
3685
3686   testpmd> flow list 0
3687   ID      Group   Prio    Attr    Rule
3688   0       0       0       i-      ETH VLAN VLAN=>VF QUEUE
3689   1       0       0       i-      ETH VLAN VLAN=>PF QUEUE
3690