xref: /dpdk/doc/guides/testpmd_app_ug/testpmd_funcs.rst (revision aac6f11f586480f9222dba99910654eda989c649)
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
1850
1851Link Bonding Functions
1852----------------------
1853
1854The Link Bonding functions make it possible to dynamically create and
1855manage link bonding devices from within testpmd interactive prompt.
1856
1857create bonded device
1858~~~~~~~~~~~~~~~~~~~~
1859
1860Create a new bonding device::
1861
1862   testpmd> create bonded device (mode) (socket)
1863
1864For example, to create a bonded device in mode 1 on socket 0::
1865
1866   testpmd> create bonded 1 0
1867   created new bonded device (port X)
1868
1869add bonding slave
1870~~~~~~~~~~~~~~~~~
1871
1872Adds Ethernet device to a Link Bonding device::
1873
1874   testpmd> add bonding slave (slave id) (port id)
1875
1876For example, to add Ethernet device (port 6) to a Link Bonding device (port 10)::
1877
1878   testpmd> add bonding slave 6 10
1879
1880
1881remove bonding slave
1882~~~~~~~~~~~~~~~~~~~~
1883
1884Removes an Ethernet slave device from a Link Bonding device::
1885
1886   testpmd> remove bonding slave (slave id) (port id)
1887
1888For example, to remove Ethernet slave device (port 6) to a Link Bonding device (port 10)::
1889
1890   testpmd> remove bonding slave 6 10
1891
1892set bonding mode
1893~~~~~~~~~~~~~~~~
1894
1895Set the Link Bonding mode of a Link Bonding device::
1896
1897   testpmd> set bonding mode (value) (port id)
1898
1899For example, to set the bonding mode of a Link Bonding device (port 10) to broadcast (mode 3)::
1900
1901   testpmd> set bonding mode 3 10
1902
1903set bonding primary
1904~~~~~~~~~~~~~~~~~~~
1905
1906Set an Ethernet slave device as the primary device on a Link Bonding device::
1907
1908   testpmd> set bonding primary (slave id) (port id)
1909
1910For example, to set the Ethernet slave device (port 6) as the primary port of a Link Bonding device (port 10)::
1911
1912   testpmd> set bonding primary 6 10
1913
1914set bonding mac
1915~~~~~~~~~~~~~~~
1916
1917Set the MAC address of a Link Bonding device::
1918
1919   testpmd> set bonding mac (port id) (mac)
1920
1921For example, to set the MAC address of a Link Bonding device (port 10) to 00:00:00:00:00:01::
1922
1923   testpmd> set bonding mac 10 00:00:00:00:00:01
1924
1925set bonding xmit_balance_policy
1926~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1927
1928Set the transmission policy for a Link Bonding device when it is in Balance XOR mode::
1929
1930   testpmd> set bonding xmit_balance_policy (port_id) (l2|l23|l34)
1931
1932For example, set a Link Bonding device (port 10) to use a balance policy of layer 3+4 (IP addresses & UDP ports)::
1933
1934   testpmd> set bonding xmit_balance_policy 10 l34
1935
1936
1937set bonding mon_period
1938~~~~~~~~~~~~~~~~~~~~~~
1939
1940Set the link status monitoring polling period in milliseconds for a bonding device.
1941
1942This adds support for PMD slave devices which do not support link status interrupts.
1943When the mon_period is set to a value greater than 0 then all PMD's which do not support
1944link status ISR will be queried every polling interval to check if their link status has changed::
1945
1946   testpmd> set bonding mon_period (port_id) (value)
1947
1948For example, to set the link status monitoring polling period of bonded device (port 5) to 150ms::
1949
1950   testpmd> set bonding mon_period 5 150
1951
1952
1953set bonding lacp dedicated_queue
1954~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1955
1956Enable dedicated tx/rx queues on bonding devices slaves to handle LACP control plane traffic
1957when in mode 4 (link-aggregration-802.3ad)::
1958
1959   testpmd> set bonding lacp dedicated_queues (port_id) (enable|disable)
1960
1961
1962set bonding agg_mode
1963~~~~~~~~~~~~~~~~~~~~
1964
1965Enable one of the specific aggregators mode when in mode 4 (link-aggregration-802.3ad)::
1966
1967   testpmd> set bonding agg_mode (port_id) (bandwidth|count|stable)
1968
1969
1970show bonding config
1971~~~~~~~~~~~~~~~~~~~
1972
1973Show the current configuration of a Link Bonding device::
1974
1975   testpmd> show bonding config (port id)
1976
1977For example,
1978to show the configuration a Link Bonding device (port 9) with 3 slave devices (1, 3, 4)
1979in balance mode with a transmission policy of layer 2+3::
1980
1981   testpmd> show bonding config 9
1982        Bonding mode: 2
1983        Balance Xmit Policy: BALANCE_XMIT_POLICY_LAYER23
1984        Slaves (3): [1 3 4]
1985        Active Slaves (3): [1 3 4]
1986        Primary: [3]
1987
1988
1989Register Functions
1990------------------
1991
1992The Register Functions can be used to read from and write to registers on the network card referenced by a port number.
1993This is mainly useful for debugging purposes.
1994Reference should be made to the appropriate datasheet for the network card for details on the register addresses
1995and fields that can be accessed.
1996
1997read reg
1998~~~~~~~~
1999
2000Display the value of a port register::
2001
2002   testpmd> read reg (port_id) (address)
2003
2004For example, to examine the Flow Director control register (FDIRCTL, 0x0000EE000) on an Intel 82599 10 GbE Controller::
2005
2006   testpmd> read reg 0 0xEE00
2007   port 0 PCI register at offset 0xEE00: 0x4A060029 (1241907241)
2008
2009read regfield
2010~~~~~~~~~~~~~
2011
2012Display a port register bit field::
2013
2014   testpmd> read regfield (port_id) (address) (bit_x) (bit_y)
2015
2016For example, reading the lowest two bits from the register in the example above::
2017
2018   testpmd> read regfield 0 0xEE00 0 1
2019   port 0 PCI register at offset 0xEE00: bits[0, 1]=0x1 (1)
2020
2021read regbit
2022~~~~~~~~~~~
2023
2024Display a single port register bit::
2025
2026   testpmd> read regbit (port_id) (address) (bit_x)
2027
2028For example, reading the lowest bit from the register in the example above::
2029
2030   testpmd> read regbit 0 0xEE00 0
2031   port 0 PCI register at offset 0xEE00: bit 0=1
2032
2033write reg
2034~~~~~~~~~
2035
2036Set the value of a port register::
2037
2038   testpmd> write reg (port_id) (address) (value)
2039
2040For example, to clear a register::
2041
2042   testpmd> write reg 0 0xEE00 0x0
2043   port 0 PCI register at offset 0xEE00: 0x00000000 (0)
2044
2045write regfield
2046~~~~~~~~~~~~~~
2047
2048Set bit field of a port register::
2049
2050   testpmd> write regfield (port_id) (address) (bit_x) (bit_y) (value)
2051
2052For example, writing to the register cleared in the example above::
2053
2054   testpmd> write regfield 0 0xEE00 0 1 2
2055   port 0 PCI register at offset 0xEE00: 0x00000002 (2)
2056
2057write regbit
2058~~~~~~~~~~~~
2059
2060Set single bit value of a port register::
2061
2062   testpmd> write regbit (port_id) (address) (bit_x) (value)
2063
2064For example, to set the high bit in the register from the example above::
2065
2066   testpmd> write regbit 0 0xEE00 31 1
2067   port 0 PCI register at offset 0xEE00: 0x8000000A (2147483658)
2068
2069Traffic Metering and Policing
2070-----------------------------
2071
2072The following section shows functions for configuring traffic metering and
2073policing on the ethernet device through the use of generic ethdev API.
2074
2075show port traffic management capability
2076~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2077
2078Show traffic metering and policing capability of the port::
2079
2080   testpmd> show port meter cap (port_id)
2081
2082add port meter profile (srTCM rfc2967)
2083~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2084
2085Add meter profile (srTCM rfc2697) to the ethernet device::
2086
2087   testpmd> add port meter profile srtcm_rfc2697 (port_id) (profile_id) \
2088   (cir) (cbs) (ebs)
2089
2090where:
2091
2092* ``profile_id``: ID for the meter profile.
2093* ``cir``: Committed Information Rate (CIR) (bytes/second).
2094* ``cbs``: Committed Burst Size (CBS) (bytes).
2095* ``ebs``: Excess Burst Size (EBS) (bytes).
2096
2097add port meter profile (trTCM rfc2968)
2098~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2099
2100Add meter profile (srTCM rfc2698) to the ethernet device::
2101
2102   testpmd> add port meter profile trtcm_rfc2698 (port_id) (profile_id) \
2103   (cir) (pir) (cbs) (pbs)
2104
2105where:
2106
2107* ``profile_id``: ID for the meter profile.
2108* ``cir``: Committed information rate (bytes/second).
2109* ``pir``: Peak information rate (bytes/second).
2110* ``cbs``: Committed burst size (bytes).
2111* ``pbs``: Peak burst size (bytes).
2112
2113add port meter profile (trTCM rfc4115)
2114~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2115
2116Add meter profile (trTCM rfc4115) to the ethernet device::
2117
2118   testpmd> add port meter profile trtcm_rfc4115 (port_id) (profile_id) \
2119   (cir) (eir) (cbs) (ebs)
2120
2121where:
2122
2123* ``profile_id``: ID for the meter profile.
2124* ``cir``: Committed information rate (bytes/second).
2125* ``eir``: Excess information rate (bytes/second).
2126* ``cbs``: Committed burst size (bytes).
2127* ``ebs``: Excess burst size (bytes).
2128
2129delete port meter profile
2130~~~~~~~~~~~~~~~~~~~~~~~~~
2131
2132Delete meter profile from the ethernet device::
2133
2134   testpmd> del port meter profile (port_id) (profile_id)
2135
2136create port meter
2137~~~~~~~~~~~~~~~~~
2138
2139Create new meter object for the ethernet device::
2140
2141   testpmd> create port meter (port_id) (mtr_id) (profile_id) \
2142   (meter_enable) (g_action) (y_action) (r_action) (stats_mask) (shared) \
2143   (use_pre_meter_color) [(dscp_tbl_entry0) (dscp_tbl_entry1)...\
2144   (dscp_tbl_entry63)]
2145
2146where:
2147
2148* ``mtr_id``: meter object ID.
2149* ``profile_id``: ID for the meter profile.
2150* ``meter_enable``: When this parameter has a non-zero value, the meter object
2151  gets enabled at the time of creation, otherwise remains disabled.
2152* ``g_action``: Policer action for the packet with green color.
2153* ``y_action``: Policer action for the packet with yellow color.
2154* ``r_action``: Policer action for the packet with red color.
2155* ``stats_mask``: Mask of statistics counter types to be enabled for the
2156  meter object.
2157* ``shared``:  When this parameter has a non-zero value, the meter object is
2158  shared by multiple flows. Otherwise, meter object is used by single flow.
2159* ``use_pre_meter_color``: When this parameter has a non-zero value, the
2160  input color for the current meter object is determined by the latest meter
2161  object in the same flow. Otherwise, the current meter object uses the
2162  *dscp_table* to determine the input color.
2163* ``dscp_tbl_entryx``: DSCP table entry x providing meter providing input
2164  color, 0 <= x <= 63.
2165
2166enable port meter
2167~~~~~~~~~~~~~~~~~
2168
2169Enable meter for the ethernet device::
2170
2171   testpmd> enable port meter (port_id) (mtr_id)
2172
2173disable port meter
2174~~~~~~~~~~~~~~~~~~
2175
2176Disable meter for the ethernet device::
2177
2178   testpmd> disable port meter (port_id) (mtr_id)
2179
2180delete port meter
2181~~~~~~~~~~~~~~~~~
2182
2183Delete meter for the ethernet device::
2184
2185   testpmd> del port meter (port_id) (mtr_id)
2186
2187Set port meter profile
2188~~~~~~~~~~~~~~~~~~~~~~
2189
2190Set meter profile for the ethernet device::
2191
2192   testpmd> set port meter profile (port_id) (mtr_id) (profile_id)
2193
2194set port meter dscp table
2195~~~~~~~~~~~~~~~~~~~~~~~~~
2196
2197Set meter dscp table for the ethernet device::
2198
2199   testpmd> set port meter dscp table (port_id) (mtr_id) [(dscp_tbl_entry0) \
2200   (dscp_tbl_entry1)...(dscp_tbl_entry63)]
2201
2202set port meter policer action
2203~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2204
2205Set meter policer action for the ethernet device::
2206
2207   testpmd> set port meter policer action (port_id) (mtr_id) (action_mask) \
2208   (action0) [(action1) (action1)]
2209
2210where:
2211
2212* ``action_mask``: Bit mask indicating which policer actions need to be
2213  updated. One or more policer actions can be updated in a single function
2214  invocation. To update the policer action associated with color C, bit
2215  (1 << C) needs to be set in *action_mask* and element at position C
2216  in the *actions* array needs to be valid.
2217* ``actionx``: Policer action for the color x,
2218  RTE_MTR_GREEN <= x < RTE_MTR_COLORS
2219
2220set port meter stats mask
2221~~~~~~~~~~~~~~~~~~~~~~~~~
2222
2223Set meter stats mask for the ethernet device::
2224
2225   testpmd> set port meter stats mask (port_id) (mtr_id) (stats_mask)
2226
2227where:
2228
2229* ``stats_mask``: Bit mask indicating statistics counter types to be enabled.
2230
2231show port meter stats
2232~~~~~~~~~~~~~~~~~~~~~
2233
2234Show meter stats of the ethernet device::
2235
2236   testpmd> show port meter stats (port_id) (mtr_id) (clear)
2237
2238where:
2239
2240* ``clear``: Flag that indicates whether the statistics counters should
2241  be cleared (i.e. set to zero) immediately after they have been read or not.
2242
2243Traffic Management
2244------------------
2245
2246The following section shows functions for configuring traffic management on
2247on the ethernet device through the use of generic TM API.
2248
2249show port traffic management capability
2250~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2251
2252Show traffic management capability of the port::
2253
2254   testpmd> show port tm cap (port_id)
2255
2256show port traffic management capability (hierarchy level)
2257~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2258
2259Show traffic management hierarchy level capability of the port::
2260
2261   testpmd> show port tm cap (port_id) (level_id)
2262
2263show port traffic management capability (hierarchy node level)
2264~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2265
2266Show the traffic management hierarchy node capability of the port::
2267
2268   testpmd> show port tm cap (port_id) (node_id)
2269
2270show port traffic management hierarchy node type
2271~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2272
2273Show the port traffic management hierarchy node type::
2274
2275   testpmd> show port tm node type (port_id) (node_id)
2276
2277show port traffic management hierarchy node stats
2278~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2279
2280Show the port traffic management hierarchy node statistics::
2281
2282   testpmd> show port tm node stats (port_id) (node_id) (clear)
2283
2284where:
2285
2286* ``clear``: When this parameter has a non-zero value, the statistics counters
2287  are cleared (i.e. set to zero) immediately after they have been read,
2288  otherwise the statistics counters are left untouched.
2289
2290Add port traffic management private shaper profile
2291~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2292
2293Add the port traffic management private shaper profile::
2294
2295   testpmd> add port tm node shaper profile (port_id) (shaper_profile_id) \
2296   (tb_rate) (tb_size) (packet_length_adjust)
2297
2298where:
2299
2300* ``shaper_profile id``: Shaper profile ID for the new profile.
2301* ``tb_rate``: Token bucket rate (bytes per second).
2302* ``tb_size``: Token bucket size (bytes).
2303* ``packet_length_adjust``: The value (bytes) to be added to the length of
2304  each packet for the purpose of shaping. This parameter value can be used to
2305  correct the packet length with the framing overhead bytes that are consumed
2306  on the wire.
2307
2308Delete port traffic management private shaper profile
2309~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2310
2311Delete the port traffic management private shaper::
2312
2313   testpmd> del port tm node shaper profile (port_id) (shaper_profile_id)
2314
2315where:
2316
2317* ``shaper_profile id``: Shaper profile ID that needs to be deleted.
2318
2319Add port traffic management shared shaper
2320~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2321
2322Create the port traffic management shared shaper::
2323
2324   testpmd> add port tm node shared shaper (port_id) (shared_shaper_id) \
2325   (shaper_profile_id)
2326
2327where:
2328
2329* ``shared_shaper_id``: Shared shaper ID to be created.
2330* ``shaper_profile id``: Shaper profile ID for shared shaper.
2331
2332Set port traffic management shared shaper
2333~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2334
2335Update the port traffic management shared shaper::
2336
2337   testpmd> set port tm node shared shaper (port_id) (shared_shaper_id) \
2338   (shaper_profile_id)
2339
2340where:
2341
2342* ``shared_shaper_id``: Shared shaper ID to be update.
2343* ``shaper_profile id``: Shaper profile ID for shared shaper.
2344
2345Delete port traffic management shared shaper
2346~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2347
2348Delete the port traffic management shared shaper::
2349
2350   testpmd> del port tm node shared shaper (port_id) (shared_shaper_id)
2351
2352where:
2353
2354* ``shared_shaper_id``: Shared shaper ID to be deleted.
2355
2356Set port traffic management hiearchy node private shaper
2357~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2358
2359set the port traffic management hierarchy node private shaper::
2360
2361   testpmd> set port tm node shaper profile (port_id) (node_id) \
2362   (shaper_profile_id)
2363
2364where:
2365
2366* ``shaper_profile id``: Private shaper profile ID to be enabled on the
2367  hierarchy node.
2368
2369Add port traffic management WRED profile
2370~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2371
2372Create a new WRED profile::
2373
2374   testpmd> add port tm node wred profile (port_id) (wred_profile_id) \
2375   (color_g) (min_th_g) (max_th_g) (maxp_inv_g) (wq_log2_g) \
2376   (color_y) (min_th_y) (max_th_y) (maxp_inv_y) (wq_log2_y) \
2377   (color_r) (min_th_r) (max_th_r) (maxp_inv_r) (wq_log2_r)
2378
2379where:
2380
2381* ``wred_profile id``: Identifier for the newly create WRED profile
2382* ``color_g``: Packet color (green)
2383* ``min_th_g``: Minimum queue threshold for packet with green color
2384* ``max_th_g``: Minimum queue threshold for packet with green color
2385* ``maxp_inv_g``: Inverse of packet marking probability maximum value (maxp)
2386* ``wq_log2_g``: Negated log2 of queue weight (wq)
2387* ``color_y``: Packet color (yellow)
2388* ``min_th_y``: Minimum queue threshold for packet with yellow color
2389* ``max_th_y``: Minimum queue threshold for packet with yellow color
2390* ``maxp_inv_y``: Inverse of packet marking probability maximum value (maxp)
2391* ``wq_log2_y``: Negated log2 of queue weight (wq)
2392* ``color_r``: Packet color (red)
2393* ``min_th_r``: Minimum queue threshold for packet with yellow color
2394* ``max_th_r``: Minimum queue threshold for packet with yellow color
2395* ``maxp_inv_r``: Inverse of packet marking probability maximum value (maxp)
2396* ``wq_log2_r``: Negated log2 of queue weight (wq)
2397
2398Delete port traffic management WRED profile
2399~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2400
2401Delete the WRED profile::
2402
2403   testpmd> del port tm node wred profile (port_id) (wred_profile_id)
2404
2405Add port traffic management hierarchy nonleaf node
2406~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2407
2408Add nonleaf node to port traffic management hiearchy::
2409
2410   testpmd> add port tm nonleaf node (port_id) (node_id) (parent_node_id) \
2411   (priority) (weight) (level_id) (shaper_profile_id) \
2412   (n_sp_priorities) (stats_mask) (n_shared_shapers) \
2413   [(shared_shaper_0) (shared_shaper_1) ...] \
2414
2415where:
2416
2417* ``parent_node_id``: Node ID of the parent.
2418* ``priority``: Node priority (highest node priority is zero). This is used by
2419  the SP algorithm running on the parent node for scheduling this node.
2420* ``weight``: Node weight (lowest weight is one). The node weight is relative
2421  to the weight sum of all siblings that have the same priority. It is used by
2422  the WFQ algorithm running on the parent node for scheduling this node.
2423* ``level_id``: Hiearchy level of the node.
2424* ``shaper_profile_id``: Shaper profile ID of the private shaper to be used by
2425  the node.
2426* ``n_sp_priorities``: Number of strict priorities.
2427* ``stats_mask``: Mask of statistics counter types to be enabled for this node.
2428* ``n_shared_shapers``: Number of shared shapers.
2429* ``shared_shaper_id``: Shared shaper id.
2430
2431Add port traffic management hierarchy leaf node
2432~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2433
2434Add leaf node to port traffic management hiearchy::
2435
2436   testpmd> add port tm leaf node (port_id) (node_id) (parent_node_id) \
2437   (priority) (weight) (level_id) (shaper_profile_id) \
2438   (cman_mode) (wred_profile_id) (stats_mask) (n_shared_shapers) \
2439   [(shared_shaper_id) (shared_shaper_id) ...] \
2440
2441where:
2442
2443* ``parent_node_id``: Node ID of the parent.
2444* ``priority``: Node priority (highest node priority is zero). This is used by
2445  the SP algorithm running on the parent node for scheduling this node.
2446* ``weight``: Node weight (lowest weight is one). The node weight is relative
2447  to the weight sum of all siblings that have the same priority. It is used by
2448  the WFQ algorithm running on the parent node for scheduling this node.
2449* ``level_id``: Hiearchy level of the node.
2450* ``shaper_profile_id``: Shaper profile ID of the private shaper to be used by
2451  the node.
2452* ``cman_mode``: Congestion management mode to be enabled for this node.
2453* ``wred_profile_id``: WRED profile id to be enabled for this node.
2454* ``stats_mask``: Mask of statistics counter types to be enabled for this node.
2455* ``n_shared_shapers``: Number of shared shapers.
2456* ``shared_shaper_id``: Shared shaper id.
2457
2458Delete port traffic management hierarchy node
2459~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2460
2461Delete node from port traffic management hiearchy::
2462
2463   testpmd> del port tm node (port_id) (node_id)
2464
2465Update port traffic management hierarchy parent node
2466~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2467
2468Update port traffic management hierarchy parent node::
2469
2470   testpmd> set port tm node parent (port_id) (node_id) (parent_node_id) \
2471   (priority) (weight)
2472
2473This function can only be called after the hierarchy commit invocation. Its
2474success depends on the port support for this operation, as advertised through
2475the port capability set. This function is valid for all nodes of the traffic
2476management hierarchy except root node.
2477
2478Commit port traffic management hierarchy
2479~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2480
2481Commit the traffic management hierarchy on the port::
2482
2483   testpmd> port tm hierarchy commit (port_id) (clean_on_fail)
2484
2485where:
2486
2487* ``clean_on_fail``: When set to non-zero, hierarchy is cleared on function
2488  call failure. On the other hand, hierarchy is preserved when this parameter
2489  is equal to zero.
2490
2491Set port traffic management default hierarchy (tm forwarding mode)
2492~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2493
2494set the traffic management default hierarchy on the port::
2495
2496   testpmd> set port tm hierarchy default (port_id)
2497
2498Filter Functions
2499----------------
2500
2501This section details the available filter functions that are available.
2502
2503Note these functions interface the deprecated legacy filtering framework,
2504superseded by *rte_flow*. See `Flow rules management`_.
2505
2506ethertype_filter
2507~~~~~~~~~~~~~~~~~~~~
2508
2509Add or delete a L2 Ethertype filter, which identify packets by their L2 Ethertype mainly assign them to a receive queue::
2510
2511   ethertype_filter (port_id) (add|del) (mac_addr|mac_ignr) (mac_address) \
2512                    ethertype (ether_type) (drop|fwd) queue (queue_id)
2513
2514The available information parameters are:
2515
2516* ``port_id``: The port which the Ethertype filter assigned on.
2517
2518* ``mac_addr``: Compare destination mac address.
2519
2520* ``mac_ignr``: Ignore destination mac address match.
2521
2522* ``mac_address``: Destination mac address to match.
2523
2524* ``ether_type``: The EtherType value want to match,
2525  for example 0x0806 for ARP packet. 0x0800 (IPv4) and 0x86DD (IPv6) are invalid.
2526
2527* ``queue_id``: The receive queue associated with this EtherType filter.
2528  It is meaningless when deleting or dropping.
2529
2530Example, to add/remove an ethertype filter rule::
2531
2532   testpmd> ethertype_filter 0 add mac_ignr 00:11:22:33:44:55 \
2533                             ethertype 0x0806 fwd queue 3
2534
2535   testpmd> ethertype_filter 0 del mac_ignr 00:11:22:33:44:55 \
2536                             ethertype 0x0806 fwd queue 3
2537
25382tuple_filter
2539~~~~~~~~~~~~~~~~~
2540
2541Add or delete a 2-tuple filter,
2542which identifies packets by specific protocol and destination TCP/UDP port
2543and forwards packets into one of the receive queues::
2544
2545   2tuple_filter (port_id) (add|del) dst_port (dst_port_value) \
2546                 protocol (protocol_value) mask (mask_value) \
2547                 tcp_flags (tcp_flags_value) priority (prio_value) \
2548                 queue (queue_id)
2549
2550The available information parameters are:
2551
2552* ``port_id``: The port which the 2-tuple filter assigned on.
2553
2554* ``dst_port_value``: Destination port in L4.
2555
2556* ``protocol_value``: IP L4 protocol.
2557
2558* ``mask_value``: Participates in the match or not by bit for field above, 1b means participate.
2559
2560* ``tcp_flags_value``: TCP control bits. The non-zero value is invalid, when the pro_value is not set to 0x06 (TCP).
2561
2562* ``prio_value``: Priority of this filter.
2563
2564* ``queue_id``: The receive queue associated with this 2-tuple filter.
2565
2566Example, to add/remove an 2tuple filter rule::
2567
2568   testpmd> 2tuple_filter 0 add dst_port 32 protocol 0x06 mask 0x03 \
2569                          tcp_flags 0x02 priority 3 queue 3
2570
2571   testpmd> 2tuple_filter 0 del dst_port 32 protocol 0x06 mask 0x03 \
2572                          tcp_flags 0x02 priority 3 queue 3
2573
25745tuple_filter
2575~~~~~~~~~~~~~~~~~
2576
2577Add or delete a 5-tuple filter,
2578which consists of a 5-tuple (protocol, source and destination IP addresses, source and destination TCP/UDP/SCTP port)
2579and routes packets into one of the receive queues::
2580
2581   5tuple_filter (port_id) (add|del) dst_ip (dst_address) src_ip \
2582                 (src_address) dst_port (dst_port_value) \
2583                 src_port (src_port_value) protocol (protocol_value) \
2584                 mask (mask_value) tcp_flags (tcp_flags_value) \
2585                 priority (prio_value) queue (queue_id)
2586
2587The available information parameters are:
2588
2589* ``port_id``: The port which the 5-tuple filter assigned on.
2590
2591* ``dst_address``: Destination IP address.
2592
2593* ``src_address``: Source IP address.
2594
2595* ``dst_port_value``: TCP/UDP destination port.
2596
2597* ``src_port_value``: TCP/UDP source port.
2598
2599* ``protocol_value``: L4 protocol.
2600
2601* ``mask_value``: Participates in the match or not by bit for field above, 1b means participate
2602
2603* ``tcp_flags_value``: TCP control bits. The non-zero value is invalid, when the protocol_value is not set to 0x06 (TCP).
2604
2605* ``prio_value``: The priority of this filter.
2606
2607* ``queue_id``: The receive queue associated with this 5-tuple filter.
2608
2609Example, to add/remove an 5tuple filter rule::
2610
2611   testpmd> 5tuple_filter 0 add dst_ip 2.2.2.5 src_ip 2.2.2.4 \
2612            dst_port 64 src_port 32 protocol 0x06 mask 0x1F \
2613            flags 0x0 priority 3 queue 3
2614
2615   testpmd> 5tuple_filter 0 del dst_ip 2.2.2.5 src_ip 2.2.2.4 \
2616            dst_port 64 src_port 32 protocol 0x06 mask 0x1F \
2617            flags 0x0 priority 3 queue 3
2618
2619syn_filter
2620~~~~~~~~~~
2621
2622Using the  SYN filter, TCP packets whose *SYN* flag is set can be forwarded to a separate queue::
2623
2624   syn_filter (port_id) (add|del) priority (high|low) queue (queue_id)
2625
2626The available information parameters are:
2627
2628* ``port_id``: The port which the SYN filter assigned on.
2629
2630* ``high``: This SYN filter has higher priority than other filters.
2631
2632* ``low``: This SYN filter has lower priority than other filters.
2633
2634* ``queue_id``: The receive queue associated with this SYN filter
2635
2636Example::
2637
2638   testpmd> syn_filter 0 add priority high queue 3
2639
2640flex_filter
2641~~~~~~~~~~~
2642
2643With flex filter, packets can be recognized by any arbitrary pattern within the first 128 bytes of the packet
2644and routed into one of the receive queues::
2645
2646   flex_filter (port_id) (add|del) len (len_value) bytes (bytes_value) \
2647               mask (mask_value) priority (prio_value) queue (queue_id)
2648
2649The available information parameters are:
2650
2651* ``port_id``: The port which the Flex filter is assigned on.
2652
2653* ``len_value``: Filter length in bytes, no greater than 128.
2654
2655* ``bytes_value``: A string in hexadecimal, means the value the flex filter needs to match.
2656
2657* ``mask_value``: A string in hexadecimal, bit 1 means corresponding byte participates in the match.
2658
2659* ``prio_value``: The priority of this filter.
2660
2661* ``queue_id``: The receive queue associated with this Flex filter.
2662
2663Example::
2664
2665   testpmd> flex_filter 0 add len 16 bytes 0x00000000000000000000000008060000 \
2666                          mask 000C priority 3 queue 3
2667
2668   testpmd> flex_filter 0 del len 16 bytes 0x00000000000000000000000008060000 \
2669                          mask 000C priority 3 queue 3
2670
2671
2672.. _testpmd_flow_director:
2673
2674flow_director_filter
2675~~~~~~~~~~~~~~~~~~~~
2676
2677The Flow Director works in receive mode to identify specific flows or sets of flows and route them to specific queues.
2678
2679Four types of filtering are supported which are referred to as Perfect Match, Signature, Perfect-mac-vlan and
2680Perfect-tunnel filters, the match mode is set by the ``--pkt-filter-mode`` command-line parameter:
2681
2682* Perfect match filters.
2683  The hardware checks a match between the masked fields of the received packets and the programmed filters.
2684  The masked fields are for IP flow.
2685
2686* Signature filters.
2687  The hardware checks a match between a hash-based signature of the masked fields of the received packet.
2688
2689* Perfect-mac-vlan match filters.
2690  The hardware checks a match between the masked fields of the received packets and the programmed filters.
2691  The masked fields are for MAC VLAN flow.
2692
2693* Perfect-tunnel match filters.
2694  The hardware checks a match between the masked fields of the received packets and the programmed filters.
2695  The masked fields are for tunnel flow.
2696
2697The Flow Director filters can match the different fields for different type of packet: flow type, specific input set
2698per flow type and the flexible payload.
2699
2700The Flow Director can also mask out parts of all of these fields so that filters
2701are only applied to certain fields or parts of the fields.
2702
2703Different NICs may have different capabilities, command show port fdir (port_id) can be used to acquire the information.
2704
2705# Commands to add flow director filters of different flow types::
2706
2707   flow_director_filter (port_id) mode IP (add|del|update) \
2708                        flow (ipv4-other|ipv4-frag|ipv6-other|ipv6-frag) \
2709                        src (src_ip_address) dst (dst_ip_address) \
2710                        tos (tos_value) proto (proto_value) ttl (ttl_value) \
2711                        vlan (vlan_value) flexbytes (flexbytes_value) \
2712                        (drop|fwd) pf|vf(vf_id) queue (queue_id) \
2713                        fd_id (fd_id_value)
2714
2715   flow_director_filter (port_id) mode IP (add|del|update) \
2716                        flow (ipv4-tcp|ipv4-udp|ipv6-tcp|ipv6-udp) \
2717                        src (src_ip_address) (src_port) \
2718                        dst (dst_ip_address) (dst_port) \
2719                        tos (tos_value) ttl (ttl_value) \
2720                        vlan (vlan_value) flexbytes (flexbytes_value) \
2721                        (drop|fwd) queue pf|vf(vf_id) (queue_id) \
2722                        fd_id (fd_id_value)
2723
2724   flow_director_filter (port_id) mode IP (add|del|update) \
2725                        flow (ipv4-sctp|ipv6-sctp) \
2726                        src (src_ip_address) (src_port) \
2727                        dst (dst_ip_address) (dst_port) \
2728                        tos (tos_value) ttl (ttl_value) \
2729                        tag (verification_tag) vlan (vlan_value) \
2730                        flexbytes (flexbytes_value) (drop|fwd) \
2731                        pf|vf(vf_id) queue (queue_id) fd_id (fd_id_value)
2732
2733   flow_director_filter (port_id) mode IP (add|del|update) flow l2_payload \
2734                        ether (ethertype) flexbytes (flexbytes_value) \
2735                        (drop|fwd) pf|vf(vf_id) queue (queue_id)
2736                        fd_id (fd_id_value)
2737
2738   flow_director_filter (port_id) mode MAC-VLAN (add|del|update) \
2739                        mac (mac_address) vlan (vlan_value) \
2740                        flexbytes (flexbytes_value) (drop|fwd) \
2741                        queue (queue_id) fd_id (fd_id_value)
2742
2743   flow_director_filter (port_id) mode Tunnel (add|del|update) \
2744                        mac (mac_address) vlan (vlan_value) \
2745                        tunnel (NVGRE|VxLAN) tunnel-id (tunnel_id_value) \
2746                        flexbytes (flexbytes_value) (drop|fwd) \
2747                        queue (queue_id) fd_id (fd_id_value)
2748
2749For example, to add an ipv4-udp flow type filter::
2750
2751   testpmd> flow_director_filter 0 mode IP add flow ipv4-udp src 2.2.2.3 32 \
2752            dst 2.2.2.5 33 tos 2 ttl 40 vlan 0x1 flexbytes (0x88,0x48) \
2753            fwd pf queue 1 fd_id 1
2754
2755For example, add an ipv4-other flow type filter::
2756
2757   testpmd> flow_director_filter 0 mode IP add flow ipv4-other src 2.2.2.3 \
2758             dst 2.2.2.5 tos 2 proto 20 ttl 40 vlan 0x1 \
2759             flexbytes (0x88,0x48) fwd pf queue 1 fd_id 1
2760
2761flush_flow_director
2762~~~~~~~~~~~~~~~~~~~
2763
2764Flush all flow director filters on a device::
2765
2766   testpmd> flush_flow_director (port_id)
2767
2768Example, to flush all flow director filter on port 0::
2769
2770   testpmd> flush_flow_director 0
2771
2772flow_director_mask
2773~~~~~~~~~~~~~~~~~~
2774
2775Set flow director's input masks::
2776
2777   flow_director_mask (port_id) mode IP vlan (vlan_value) \
2778                      src_mask (ipv4_src) (ipv6_src) (src_port) \
2779                      dst_mask (ipv4_dst) (ipv6_dst) (dst_port)
2780
2781   flow_director_mask (port_id) mode MAC-VLAN vlan (vlan_value)
2782
2783   flow_director_mask (port_id) mode Tunnel vlan (vlan_value) \
2784                      mac (mac_value) tunnel-type (tunnel_type_value) \
2785                      tunnel-id (tunnel_id_value)
2786
2787Example, to set flow director mask on port 0::
2788
2789   testpmd> flow_director_mask 0 mode IP vlan 0xefff \
2790            src_mask 255.255.255.255 \
2791                FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF \
2792            dst_mask 255.255.255.255 \
2793                FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF
2794
2795flow_director_flex_mask
2796~~~~~~~~~~~~~~~~~~~~~~~
2797
2798set masks of flow director's flexible payload based on certain flow type::
2799
2800   testpmd> flow_director_flex_mask (port_id) \
2801            flow (none|ipv4-other|ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
2802                  ipv6-other|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp| \
2803                  l2_payload|all) (mask)
2804
2805Example, to set flow director's flex mask for all flow type on port 0::
2806
2807   testpmd> flow_director_flex_mask 0 flow all \
2808            (0xff,0xff,0,0,0,0,0,0,0,0,0,0,0,0,0,0)
2809
2810
2811flow_director_flex_payload
2812~~~~~~~~~~~~~~~~~~~~~~~~~~
2813
2814Configure flexible payload selection::
2815
2816   flow_director_flex_payload (port_id) (raw|l2|l3|l4) (config)
2817
2818For example, to select the first 16 bytes from the offset 4 (bytes) of packet's payload as flexible payload::
2819
2820   testpmd> flow_director_flex_payload 0 l4 \
2821            (4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19)
2822
2823get_sym_hash_ena_per_port
2824~~~~~~~~~~~~~~~~~~~~~~~~~
2825
2826Get symmetric hash enable configuration per port::
2827
2828   get_sym_hash_ena_per_port (port_id)
2829
2830For example, to get symmetric hash enable configuration of port 1::
2831
2832   testpmd> get_sym_hash_ena_per_port 1
2833
2834set_sym_hash_ena_per_port
2835~~~~~~~~~~~~~~~~~~~~~~~~~
2836
2837Set symmetric hash enable configuration per port to enable or disable::
2838
2839   set_sym_hash_ena_per_port (port_id) (enable|disable)
2840
2841For example, to set symmetric hash enable configuration of port 1 to enable::
2842
2843   testpmd> set_sym_hash_ena_per_port 1 enable
2844
2845get_hash_global_config
2846~~~~~~~~~~~~~~~~~~~~~~
2847
2848Get the global configurations of hash filters::
2849
2850   get_hash_global_config (port_id)
2851
2852For example, to get the global configurations of hash filters of port 1::
2853
2854   testpmd> get_hash_global_config 1
2855
2856set_hash_global_config
2857~~~~~~~~~~~~~~~~~~~~~~
2858
2859Set the global configurations of hash filters::
2860
2861   set_hash_global_config (port_id) (toeplitz|simple_xor|default) \
2862   (ipv4|ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp|ipv4-other|ipv6|ipv6-frag| \
2863   ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other|l2_payload) \
2864   (enable|disable)
2865
2866For example, to enable simple_xor for flow type of ipv6 on port 2::
2867
2868   testpmd> set_hash_global_config 2 simple_xor ipv6 enable
2869
2870set_hash_input_set
2871~~~~~~~~~~~~~~~~~~
2872
2873Set the input set for hash::
2874
2875   set_hash_input_set (port_id) (ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
2876   ipv4-other|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other| \
2877   l2_payload) (ovlan|ivlan|src-ipv4|dst-ipv4|src-ipv6|dst-ipv6|ipv4-tos| \
2878   ipv4-proto|ipv6-tc|ipv6-next-header|udp-src-port|udp-dst-port| \
2879   tcp-src-port|tcp-dst-port|sctp-src-port|sctp-dst-port|sctp-veri-tag| \
2880   udp-key|gre-key|fld-1st|fld-2nd|fld-3rd|fld-4th|fld-5th|fld-6th|fld-7th| \
2881   fld-8th|none) (select|add)
2882
2883For example, to add source IP to hash input set for flow type of ipv4-udp on port 0::
2884
2885   testpmd> set_hash_input_set 0 ipv4-udp src-ipv4 add
2886
2887set_fdir_input_set
2888~~~~~~~~~~~~~~~~~~
2889
2890The Flow Director filters can match the different fields for different type of packet, i.e. specific input set
2891on per flow type and the flexible payload. This command can be used to change input set for each flow type.
2892
2893Set the input set for flow director::
2894
2895   set_fdir_input_set (port_id) (ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
2896   ipv4-other|ipv6|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other| \
2897   l2_payload) (ivlan|ethertype|src-ipv4|dst-ipv4|src-ipv6|dst-ipv6|ipv4-tos| \
2898   ipv4-proto|ipv4-ttl|ipv6-tc|ipv6-next-header|ipv6-hop-limits| \
2899   tudp-src-port|udp-dst-port|cp-src-port|tcp-dst-port|sctp-src-port| \
2900   sctp-dst-port|sctp-veri-tag|none) (select|add)
2901
2902For example to add source IP to FD input set for flow type of ipv4-udp on port 0::
2903
2904   testpmd> set_fdir_input_set 0 ipv4-udp src-ipv4 add
2905
2906global_config
2907~~~~~~~~~~~~~
2908
2909Set different GRE key length for input set::
2910
2911   global_config (port_id) gre-key-len (number in bytes)
2912
2913For example to set GRE key length for input set to 4 bytes on port 0::
2914
2915   testpmd> global_config 0 gre-key-len 4
2916
2917
2918.. _testpmd_rte_flow:
2919
2920Flow rules management
2921---------------------
2922
2923Control of the generic flow API (*rte_flow*) is fully exposed through the
2924``flow`` command (validation, creation, destruction, queries and operation
2925modes).
2926
2927Considering *rte_flow* overlaps with all `Filter Functions`_, using both
2928features simultaneously may cause undefined side-effects and is therefore
2929not recommended.
2930
2931``flow`` syntax
2932~~~~~~~~~~~~~~~
2933
2934Because the ``flow`` command uses dynamic tokens to handle the large number
2935of possible flow rules combinations, its behavior differs slightly from
2936other commands, in particular:
2937
2938- Pressing *?* or the *<tab>* key displays contextual help for the current
2939  token, not that of the entire command.
2940
2941- Optional and repeated parameters are supported (provided they are listed
2942  in the contextual help).
2943
2944The first parameter stands for the operation mode. Possible operations and
2945their general syntax are described below. They are covered in detail in the
2946following sections.
2947
2948- Check whether a flow rule can be created::
2949
2950   flow validate {port_id}
2951       [group {group_id}] [priority {level}] [ingress] [egress]
2952       pattern {item} [/ {item} [...]] / end
2953       actions {action} [/ {action} [...]] / end
2954
2955- Create a flow rule::
2956
2957   flow create {port_id}
2958       [group {group_id}] [priority {level}] [ingress] [egress]
2959       pattern {item} [/ {item} [...]] / end
2960       actions {action} [/ {action} [...]] / end
2961
2962- Destroy specific flow rules::
2963
2964   flow destroy {port_id} rule {rule_id} [...]
2965
2966- Destroy all flow rules::
2967
2968   flow flush {port_id}
2969
2970- Query an existing flow rule::
2971
2972   flow query {port_id} {rule_id} {action}
2973
2974- List existing flow rules sorted by priority, filtered by group
2975  identifiers::
2976
2977   flow list {port_id} [group {group_id}] [...]
2978
2979- Restrict ingress traffic to the defined flow rules::
2980
2981   flow isolate {port_id} {boolean}
2982
2983Validating flow rules
2984~~~~~~~~~~~~~~~~~~~~~
2985
2986``flow validate`` reports whether a flow rule would be accepted by the
2987underlying device in its current state but stops short of creating it. It is
2988bound to ``rte_flow_validate()``::
2989
2990   flow validate {port_id}
2991      [group {group_id}] [priority {level}] [ingress] [egress]
2992      pattern {item} [/ {item} [...]] / end
2993      actions {action} [/ {action} [...]] / end
2994
2995If successful, it will show::
2996
2997   Flow rule validated
2998
2999Otherwise it will show an error message of the form::
3000
3001   Caught error type [...] ([...]): [...]
3002
3003This command uses the same parameters as ``flow create``, their format is
3004described in `Creating flow rules`_.
3005
3006Check whether redirecting any Ethernet packet received on port 0 to RX queue
3007index 6 is supported::
3008
3009   testpmd> flow validate 0 ingress pattern eth / end
3010      actions queue index 6 / end
3011   Flow rule validated
3012   testpmd>
3013
3014Port 0 does not support TCPv6 rules::
3015
3016   testpmd> flow validate 0 ingress pattern eth / ipv6 / tcp / end
3017      actions drop / end
3018   Caught error type 9 (specific pattern item): Invalid argument
3019   testpmd>
3020
3021Creating flow rules
3022~~~~~~~~~~~~~~~~~~~
3023
3024``flow create`` validates and creates the specified flow rule. It is bound
3025to ``rte_flow_create()``::
3026
3027   flow create {port_id}
3028      [group {group_id}] [priority {level}] [ingress] [egress]
3029      pattern {item} [/ {item} [...]] / end
3030      actions {action} [/ {action} [...]] / end
3031
3032If successful, it will return a flow rule ID usable with other commands::
3033
3034   Flow rule #[...] created
3035
3036Otherwise it will show an error message of the form::
3037
3038   Caught error type [...] ([...]): [...]
3039
3040Parameters describe in the following order:
3041
3042- Attributes (*group*, *priority*, *ingress*, *egress* tokens).
3043- A matching pattern, starting with the *pattern* token and terminated by an
3044  *end* pattern item.
3045- Actions, starting with the *actions* token and terminated by an *end*
3046  action.
3047
3048These translate directly to *rte_flow* objects provided as-is to the
3049underlying functions.
3050
3051The shortest valid definition only comprises mandatory tokens::
3052
3053   testpmd> flow create 0 pattern end actions end
3054
3055Note that PMDs may refuse rules that essentially do nothing such as this
3056one.
3057
3058**All unspecified object values are automatically initialized to 0.**
3059
3060Attributes
3061^^^^^^^^^^
3062
3063These tokens affect flow rule attributes (``struct rte_flow_attr``) and are
3064specified before the ``pattern`` token.
3065
3066- ``group {group id}``: priority group.
3067- ``priority {level}``: priority level within group.
3068- ``ingress``: rule applies to ingress traffic.
3069- ``egress``: rule applies to egress traffic.
3070
3071Each instance of an attribute specified several times overrides the previous
3072value as shown below (group 4 is used)::
3073
3074   testpmd> flow create 0 group 42 group 24 group 4 [...]
3075
3076Note that once enabled, ``ingress`` and ``egress`` cannot be disabled.
3077
3078While not specifying a direction is an error, some rules may allow both
3079simultaneously.
3080
3081Most rules affect RX therefore contain the ``ingress`` token::
3082
3083   testpmd> flow create 0 ingress pattern [...]
3084
3085Matching pattern
3086^^^^^^^^^^^^^^^^
3087
3088A matching pattern starts after the ``pattern`` token. It is made of pattern
3089items and is terminated by a mandatory ``end`` item.
3090
3091Items are named after their type (*RTE_FLOW_ITEM_TYPE_* from ``enum
3092rte_flow_item_type``).
3093
3094The ``/`` token is used as a separator between pattern items as shown
3095below::
3096
3097   testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end [...]
3098
3099Note that protocol items like these must be stacked from lowest to highest
3100layer to make sense. For instance, the following rule is either invalid or
3101unlikely to match any packet::
3102
3103   testpmd> flow create 0 ingress pattern eth / udp / ipv4 / end [...]
3104
3105More information on these restrictions can be found in the *rte_flow*
3106documentation.
3107
3108Several items support additional specification structures, for example
3109``ipv4`` allows specifying source and destination addresses as follows::
3110
3111   testpmd> flow create 0 ingress pattern eth / ipv4 src is 10.1.1.1
3112      dst is 10.2.0.0 / end [...]
3113
3114This rule matches all IPv4 traffic with the specified properties.
3115
3116In this example, ``src`` and ``dst`` are field names of the underlying
3117``struct rte_flow_item_ipv4`` object. All item properties can be specified
3118in a similar fashion.
3119
3120The ``is`` token means that the subsequent value must be matched exactly,
3121and assigns ``spec`` and ``mask`` fields in ``struct rte_flow_item``
3122accordingly. Possible assignment tokens are:
3123
3124- ``is``: match value perfectly (with full bit-mask).
3125- ``spec``: match value according to configured bit-mask.
3126- ``last``: specify upper bound to establish a range.
3127- ``mask``: specify bit-mask with relevant bits set to one.
3128- ``prefix``: generate bit-mask from a prefix length.
3129
3130These yield identical results::
3131
3132   ipv4 src is 10.1.1.1
3133
3134::
3135
3136   ipv4 src spec 10.1.1.1 src mask 255.255.255.255
3137
3138::
3139
3140   ipv4 src spec 10.1.1.1 src prefix 32
3141
3142::
3143
3144   ipv4 src is 10.1.1.1 src last 10.1.1.1 # range with a single value
3145
3146::
3147
3148   ipv4 src is 10.1.1.1 src last 0 # 0 disables range
3149
3150Inclusive ranges can be defined with ``last``::
3151
3152   ipv4 src is 10.1.1.1 src last 10.2.3.4 # 10.1.1.1 to 10.2.3.4
3153
3154Note that ``mask`` affects both ``spec`` and ``last``::
3155
3156   ipv4 src is 10.1.1.1 src last 10.2.3.4 src mask 255.255.0.0
3157      # matches 10.1.0.0 to 10.2.255.255
3158
3159Properties can be modified multiple times::
3160
3161   ipv4 src is 10.1.1.1 src is 10.1.2.3 src is 10.2.3.4 # matches 10.2.3.4
3162
3163::
3164
3165   ipv4 src is 10.1.1.1 src prefix 24 src prefix 16 # matches 10.1.0.0/16
3166
3167Pattern items
3168^^^^^^^^^^^^^
3169
3170This section lists supported pattern items and their attributes, if any.
3171
3172- ``end``: end list of pattern items.
3173
3174- ``void``: no-op pattern item.
3175
3176- ``invert``: perform actions when pattern does not match.
3177
3178- ``any``: match any protocol for the current layer.
3179
3180  - ``num {unsigned}``: number of layers covered.
3181
3182- ``pf``: match packets addressed to the physical function.
3183
3184- ``vf``: match packets addressed to a virtual function ID.
3185
3186  - ``id {unsigned}``: destination VF ID.
3187
3188- ``port``: device-specific physical port index to use.
3189
3190  - ``index {unsigned}``: physical port index.
3191
3192- ``raw``: match an arbitrary byte string.
3193
3194  - ``relative {boolean}``: look for pattern after the previous item.
3195  - ``search {boolean}``: search pattern from offset (see also limit).
3196  - ``offset {integer}``: absolute or relative offset for pattern.
3197  - ``limit {unsigned}``: search area limit for start of pattern.
3198  - ``pattern {string}``: byte string to look for.
3199
3200- ``eth``: match Ethernet header.
3201
3202  - ``dst {MAC-48}``: destination MAC.
3203  - ``src {MAC-48}``: source MAC.
3204  - ``type {unsigned}``: EtherType.
3205
3206- ``vlan``: match 802.1Q/ad VLAN tag.
3207
3208  - ``tpid {unsigned}``: tag protocol identifier.
3209  - ``tci {unsigned}``: tag control information.
3210  - ``pcp {unsigned}``: priority code point.
3211  - ``dei {unsigned}``: drop eligible indicator.
3212  - ``vid {unsigned}``: VLAN identifier.
3213
3214- ``ipv4``: match IPv4 header.
3215
3216  - ``tos {unsigned}``: type of service.
3217  - ``ttl {unsigned}``: time to live.
3218  - ``proto {unsigned}``: next protocol ID.
3219  - ``src {ipv4 address}``: source address.
3220  - ``dst {ipv4 address}``: destination address.
3221
3222- ``ipv6``: match IPv6 header.
3223
3224  - ``tc {unsigned}``: traffic class.
3225  - ``flow {unsigned}``: flow label.
3226  - ``proto {unsigned}``: protocol (next header).
3227  - ``hop {unsigned}``: hop limit.
3228  - ``src {ipv6 address}``: source address.
3229  - ``dst {ipv6 address}``: destination address.
3230
3231- ``icmp``: match ICMP header.
3232
3233  - ``type {unsigned}``: ICMP packet type.
3234  - ``code {unsigned}``: ICMP packet code.
3235
3236- ``udp``: match UDP header.
3237
3238  - ``src {unsigned}``: UDP source port.
3239  - ``dst {unsigned}``: UDP destination port.
3240
3241- ``tcp``: match TCP header.
3242
3243  - ``src {unsigned}``: TCP source port.
3244  - ``dst {unsigned}``: TCP destination port.
3245
3246- ``sctp``: match SCTP header.
3247
3248  - ``src {unsigned}``: SCTP source port.
3249  - ``dst {unsigned}``: SCTP destination port.
3250  - ``tag {unsigned}``: validation tag.
3251  - ``cksum {unsigned}``: checksum.
3252
3253- ``vxlan``: match VXLAN header.
3254
3255  - ``vni {unsigned}``: VXLAN identifier.
3256
3257- ``e_tag``: match IEEE 802.1BR E-Tag header.
3258
3259  - ``grp_ecid_b {unsigned}``: GRP and E-CID base.
3260
3261- ``nvgre``: match NVGRE header.
3262
3263  - ``tni {unsigned}``: virtual subnet ID.
3264
3265- ``mpls``: match MPLS header.
3266
3267  - ``label {unsigned}``: MPLS label.
3268
3269- ``gre``: match GRE header.
3270
3271  - ``protocol {unsigned}``: protocol type.
3272
3273- ``fuzzy``: fuzzy pattern match, expect faster than default.
3274
3275  - ``thresh {unsigned}``: accuracy threshold.
3276
3277- ``gtp``, ``gtpc``, ``gtpu``: match GTPv1 header.
3278
3279  - ``teid {unsigned}``: tunnel endpoint identifier.
3280
3281Actions list
3282^^^^^^^^^^^^
3283
3284A list of actions starts after the ``actions`` token in the same fashion as
3285`Matching pattern`_; actions are separated by ``/`` tokens and the list is
3286terminated by a mandatory ``end`` action.
3287
3288Actions are named after their type (*RTE_FLOW_ACTION_TYPE_* from ``enum
3289rte_flow_action_type``).
3290
3291Dropping all incoming UDPv4 packets can be expressed as follows::
3292
3293   testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
3294      actions drop / end
3295
3296Several actions have configurable properties which must be specified when
3297there is no valid default value. For example, ``queue`` requires a target
3298queue index.
3299
3300This rule redirects incoming UDPv4 traffic to queue index 6::
3301
3302   testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
3303      actions queue index 6 / end
3304
3305While this one could be rejected by PMDs (unspecified queue index)::
3306
3307   testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
3308      actions queue / end
3309
3310As defined by *rte_flow*, the list is not ordered, all actions of a given
3311rule are performed simultaneously. These are equivalent::
3312
3313   queue index 6 / void / mark id 42 / end
3314
3315::
3316
3317   void / mark id 42 / queue index 6 / end
3318
3319All actions in a list should have different types, otherwise only the last
3320action of a given type is taken into account::
3321
3322   queue index 4 / queue index 5 / queue index 6 / end # will use queue 6
3323
3324::
3325
3326   drop / drop / drop / end # drop is performed only once
3327
3328::
3329
3330   mark id 42 / queue index 3 / mark id 24 / end # mark will be 24
3331
3332Considering they are performed simultaneously, opposite and overlapping
3333actions can sometimes be combined when the end result is unambiguous::
3334
3335   drop / queue index 6 / end # drop has no effect
3336
3337::
3338
3339   drop / dup index 6 / end # same as above
3340
3341::
3342
3343   queue index 6 / rss queues 6 7 8 / end # queue has no effect
3344
3345::
3346
3347   drop / passthru / end # drop has no effect
3348
3349Note that PMDs may still refuse such combinations.
3350
3351Actions
3352^^^^^^^
3353
3354This section lists supported actions and their attributes, if any.
3355
3356- ``end``: end list of actions.
3357
3358- ``void``: no-op action.
3359
3360- ``passthru``: let subsequent rule process matched packets.
3361
3362- ``mark``: attach 32 bit value to packets.
3363
3364  - ``id {unsigned}``: 32 bit value to return with packets.
3365
3366- ``flag``: flag packets.
3367
3368- ``queue``: assign packets to a given queue index.
3369
3370  - ``index {unsigned}``: queue index to use.
3371
3372- ``drop``: drop packets (note: passthru has priority).
3373
3374- ``count``: enable counters for this rule.
3375
3376- ``dup``: duplicate packets to a given queue index.
3377
3378  - ``index {unsigned}``: queue index to duplicate packets to.
3379
3380- ``rss``: spread packets among several queues.
3381
3382  - ``queues [{unsigned} [...]] end``: queue indices to use.
3383
3384- ``pf``: redirect packets to physical device function.
3385
3386- ``vf``: redirect packets to virtual device function.
3387
3388  - ``original {boolean}``: use original VF ID if possible.
3389  - ``id {unsigned}``: VF ID to redirect packets to.
3390
3391Destroying flow rules
3392~~~~~~~~~~~~~~~~~~~~~
3393
3394``flow destroy`` destroys one or more rules from their rule ID (as returned
3395by ``flow create``), this command calls ``rte_flow_destroy()`` as many
3396times as necessary::
3397
3398   flow destroy {port_id} rule {rule_id} [...]
3399
3400If successful, it will show::
3401
3402   Flow rule #[...] destroyed
3403
3404It does not report anything for rule IDs that do not exist. The usual error
3405message is shown when a rule cannot be destroyed::
3406
3407   Caught error type [...] ([...]): [...]
3408
3409``flow flush`` destroys all rules on a device and does not take extra
3410arguments. It is bound to ``rte_flow_flush()``::
3411
3412   flow flush {port_id}
3413
3414Any errors are reported as above.
3415
3416Creating several rules and destroying them::
3417
3418   testpmd> flow create 0 ingress pattern eth / ipv6 / end
3419      actions queue index 2 / end
3420   Flow rule #0 created
3421   testpmd> flow create 0 ingress pattern eth / ipv4 / end
3422      actions queue index 3 / end
3423   Flow rule #1 created
3424   testpmd> flow destroy 0 rule 0 rule 1
3425   Flow rule #1 destroyed
3426   Flow rule #0 destroyed
3427   testpmd>
3428
3429The same result can be achieved using ``flow flush``::
3430
3431   testpmd> flow create 0 ingress pattern eth / ipv6 / end
3432      actions queue index 2 / end
3433   Flow rule #0 created
3434   testpmd> flow create 0 ingress pattern eth / ipv4 / end
3435      actions queue index 3 / end
3436   Flow rule #1 created
3437   testpmd> flow flush 0
3438   testpmd>
3439
3440Non-existent rule IDs are ignored::
3441
3442   testpmd> flow create 0 ingress pattern eth / ipv6 / end
3443      actions queue index 2 / end
3444   Flow rule #0 created
3445   testpmd> flow create 0 ingress pattern eth / ipv4 / end
3446      actions queue index 3 / end
3447   Flow rule #1 created
3448   testpmd> flow destroy 0 rule 42 rule 10 rule 2
3449   testpmd>
3450   testpmd> flow destroy 0 rule 0
3451   Flow rule #0 destroyed
3452   testpmd>
3453
3454Querying flow rules
3455~~~~~~~~~~~~~~~~~~~
3456
3457``flow query`` queries a specific action of a flow rule having that
3458ability. Such actions collect information that can be reported using this
3459command. It is bound to ``rte_flow_query()``::
3460
3461   flow query {port_id} {rule_id} {action}
3462
3463If successful, it will display either the retrieved data for known actions
3464or the following message::
3465
3466   Cannot display result for action type [...] ([...])
3467
3468Otherwise, it will complain either that the rule does not exist or that some
3469error occurred::
3470
3471   Flow rule #[...] not found
3472
3473::
3474
3475   Caught error type [...] ([...]): [...]
3476
3477Currently only the ``count`` action is supported. This action reports the
3478number of packets that hit the flow rule and the total number of bytes. Its
3479output has the following format::
3480
3481   count:
3482    hits_set: [...] # whether "hits" contains a valid value
3483    bytes_set: [...] # whether "bytes" contains a valid value
3484    hits: [...] # number of packets
3485    bytes: [...] # number of bytes
3486
3487Querying counters for TCPv6 packets redirected to queue 6::
3488
3489   testpmd> flow create 0 ingress pattern eth / ipv6 / tcp / end
3490      actions queue index 6 / count / end
3491   Flow rule #4 created
3492   testpmd> flow query 0 4 count
3493   count:
3494    hits_set: 1
3495    bytes_set: 0
3496    hits: 386446
3497    bytes: 0
3498   testpmd>
3499
3500Listing flow rules
3501~~~~~~~~~~~~~~~~~~
3502
3503``flow list`` lists existing flow rules sorted by priority and optionally
3504filtered by group identifiers::
3505
3506   flow list {port_id} [group {group_id}] [...]
3507
3508This command only fails with the following message if the device does not
3509exist::
3510
3511   Invalid port [...]
3512
3513Output consists of a header line followed by a short description of each
3514flow rule, one per line. There is no output at all when no flow rules are
3515configured on the device::
3516
3517   ID      Group   Prio    Attr    Rule
3518   [...]   [...]   [...]   [...]   [...]
3519
3520``Attr`` column flags:
3521
3522- ``i`` for ``ingress``.
3523- ``e`` for ``egress``.
3524
3525Creating several flow rules and listing them::
3526
3527   testpmd> flow create 0 ingress pattern eth / ipv4 / end
3528      actions queue index 6 / end
3529   Flow rule #0 created
3530   testpmd> flow create 0 ingress pattern eth / ipv6 / end
3531      actions queue index 2 / end
3532   Flow rule #1 created
3533   testpmd> flow create 0 priority 5 ingress pattern eth / ipv4 / udp / end
3534      actions rss queues 6 7 8 end / end
3535   Flow rule #2 created
3536   testpmd> flow list 0
3537   ID      Group   Prio    Attr    Rule
3538   0       0       0       i-      ETH IPV4 => QUEUE
3539   1       0       0       i-      ETH IPV6 => QUEUE
3540   2       0       5       i-      ETH IPV4 UDP => RSS
3541   testpmd>
3542
3543Rules are sorted by priority (i.e. group ID first, then priority level)::
3544
3545   testpmd> flow list 1
3546   ID      Group   Prio    Attr    Rule
3547   0       0       0       i-      ETH => COUNT
3548   6       0       500     i-      ETH IPV6 TCP => DROP COUNT
3549   5       0       1000    i-      ETH IPV6 ICMP => QUEUE
3550   1       24      0       i-      ETH IPV4 UDP => QUEUE
3551   4       24      10      i-      ETH IPV4 TCP => DROP
3552   3       24      20      i-      ETH IPV4 => DROP
3553   2       24      42      i-      ETH IPV4 UDP => QUEUE
3554   7       63      0       i-      ETH IPV6 UDP VXLAN => MARK QUEUE
3555   testpmd>
3556
3557Output can be limited to specific groups::
3558
3559   testpmd> flow list 1 group 0 group 63
3560   ID      Group   Prio    Attr    Rule
3561   0       0       0       i-      ETH => COUNT
3562   6       0       500     i-      ETH IPV6 TCP => DROP COUNT
3563   5       0       1000    i-      ETH IPV6 ICMP => QUEUE
3564   7       63      0       i-      ETH IPV6 UDP VXLAN => MARK QUEUE
3565   testpmd>
3566
3567Toggling isolated mode
3568~~~~~~~~~~~~~~~~~~~~~~
3569
3570``flow isolate`` can be used to tell the underlying PMD that ingress traffic
3571must only be injected from the defined flow rules; that no default traffic
3572is expected outside those rules and the driver is free to assign more
3573resources to handle them. It is bound to ``rte_flow_isolate()``::
3574
3575 flow isolate {port_id} {boolean}
3576
3577If successful, enabling or disabling isolated mode shows either::
3578
3579 Ingress traffic on port [...]
3580    is now restricted to the defined flow rules
3581
3582Or::
3583
3584 Ingress traffic on port [...]
3585    is not restricted anymore to the defined flow rules
3586
3587Otherwise, in case of error::
3588
3589   Caught error type [...] ([...]): [...]
3590
3591Mainly due to its side effects, PMDs supporting this mode may not have the
3592ability to toggle it more than once without reinitializing affected ports
3593first (e.g. by exiting testpmd).
3594
3595Enabling isolated mode::
3596
3597 testpmd> flow isolate 0 true
3598 Ingress traffic on port 0 is now restricted to the defined flow rules
3599 testpmd>
3600
3601Disabling isolated mode::
3602
3603 testpmd> flow isolate 0 false
3604 Ingress traffic on port 0 is not restricted anymore to the defined flow rules
3605 testpmd>
3606
3607Sample QinQ flow rules
3608~~~~~~~~~~~~~~~~~~~~~~
3609
3610Before creating QinQ rule(s) the following commands should be issued to enable QinQ::
3611
3612   testpmd> port stop 0
3613   testpmd> vlan set qinq on 0
3614
3615The above command sets the inner and outer TPID's to 0x8100.
3616
3617To change the TPID's the following commands should be used::
3618
3619   testpmd> vlan set outer tpid 0xa100 0
3620   testpmd> vlan set inner tpid 0x9100 0
3621   testpmd> port start 0
3622
3623Validate and create a QinQ rule on port 0 to steer traffic to a VF queue in a VM.
3624
3625::
3626
3627   testpmd> flow validate 0 ingress pattern eth / vlan tci is 123 /
3628       vlan tci is 456 / end actions vf id 1 / queue index 0 / end
3629   Flow rule #0 validated
3630
3631   testpmd> flow create 0 ingress pattern eth / vlan tci is 4 /
3632       vlan tci is 456 / end actions vf id 123 / queue index 0 / end
3633   Flow rule #0 created
3634
3635   testpmd> flow list 0
3636   ID      Group   Prio    Attr    Rule
3637   0       0       0       i-      ETH VLAN VLAN=>VF QUEUE
3638
3639Validate and create a QinQ rule on port 0 to steer traffic to a queue on the host.
3640
3641::
3642
3643   testpmd> flow validate 0 ingress pattern eth / vlan tci is 321 /
3644        vlan tci is 654 / end actions pf / queue index 0 / end
3645   Flow rule #1 validated
3646
3647   testpmd> flow create 0 ingress pattern eth / vlan tci is 321 /
3648        vlan tci is 654 / end actions pf / queue index 1 / end
3649   Flow rule #1 created
3650
3651   testpmd> flow list 0
3652   ID      Group   Prio    Attr    Rule
3653   0       0       0       i-      ETH VLAN VLAN=>VF QUEUE
3654   1       0       0       i-      ETH VLAN VLAN=>PF QUEUE
3655