xref: /dpdk/doc/guides/testpmd_app_ug/testpmd_funcs.rst (revision 28da7d752bf054e76d32c8ce049f0e6a29874f59)
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 port-uta
1093~~~~~~~~~~~~
1094
1095Set the unicast hash filter(s) on/off for a port::
1096
1097   testpmd> set port (port_id) uta (XX:XX:XX:XX:XX:XX|all) (on|off)
1098
1099set promisc
1100~~~~~~~~~~~
1101
1102Set the promiscuous mode on for a port or for all ports.
1103In promiscuous mode packets are not dropped if they aren't for the specified MAC address::
1104
1105   testpmd> set promisc (port_id|all) (on|off)
1106
1107set allmulti
1108~~~~~~~~~~~~
1109
1110Set the allmulti mode for a port or for all ports::
1111
1112   testpmd> set allmulti (port_id|all) (on|off)
1113
1114Same as the ifconfig (8) option. Controls how multicast packets are handled.
1115
1116set promisc (for VF)
1117~~~~~~~~~~~~~~~~~~~~
1118
1119Set the unicast promiscuous mode for a VF from PF.
1120It's supported by Intel i40e NICs now.
1121In promiscuous mode packets are not dropped if they aren't for the specified MAC address::
1122
1123   testpmd> set vf promisc (port_id) (vf_id) (on|off)
1124
1125set allmulticast (for VF)
1126~~~~~~~~~~~~~~~~~~~~~~~~~
1127
1128Set the multicast 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 allmulti (port_id) (vf_id) (on|off)
1133
1134set tx max bandwidth (for VF)
1135~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1136
1137Set TX max absolute bandwidth (Mbps) for a VF from PF::
1138
1139   testpmd> set vf tx max-bandwidth (port_id) (vf_id) (max_bandwidth)
1140
1141set tc tx min bandwidth (for VF)
1142~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1143
1144Set all TCs' TX min relative bandwidth (%) for a VF from PF::
1145
1146   testpmd> set vf tc tx min-bandwidth (port_id) (vf_id) (bw1, bw2, ...)
1147
1148set tc tx max bandwidth (for VF)
1149~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1150
1151Set a TC's TX max absolute bandwidth (Mbps) for a VF from PF::
1152
1153   testpmd> set vf tc tx max-bandwidth (port_id) (vf_id) (tc_no) (max_bandwidth)
1154
1155set tc strict link priority mode
1156~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1157
1158Set some TCs' strict link priority mode on a physical port::
1159
1160   testpmd> set tx strict-link-priority (port_id) (tc_bitmap)
1161
1162set tc tx min bandwidth
1163~~~~~~~~~~~~~~~~~~~~~~~
1164
1165Set all TCs' TX min relative bandwidth (%) globally for all PF and VFs::
1166
1167   testpmd> set tc tx min-bandwidth (port_id) (bw1, bw2, ...)
1168
1169set flow_ctrl rx
1170~~~~~~~~~~~~~~~~
1171
1172Set the link flow control parameter on a port::
1173
1174   testpmd> set flow_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \
1175            (pause_time) (send_xon) mac_ctrl_frame_fwd (on|off) \
1176	    autoneg (on|off) (port_id)
1177
1178Where:
1179
1180* ``high_water`` (integer): High threshold value to trigger XOFF.
1181
1182* ``low_water`` (integer): Low threshold value to trigger XON.
1183
1184* ``pause_time`` (integer): Pause quota in the Pause frame.
1185
1186* ``send_xon`` (0/1): Send XON frame.
1187
1188* ``mac_ctrl_frame_fwd``: Enable receiving MAC control frames.
1189
1190* ``autoneg``: Change the auto-negotiation parameter.
1191
1192set pfc_ctrl rx
1193~~~~~~~~~~~~~~~
1194
1195Set the priority flow control parameter on a port::
1196
1197   testpmd> set pfc_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \
1198            (pause_time) (priority) (port_id)
1199
1200Where:
1201
1202* ``high_water`` (integer): High threshold value.
1203
1204* ``low_water`` (integer): Low threshold value.
1205
1206* ``pause_time`` (integer): Pause quota in the Pause frame.
1207
1208* ``priority`` (0-7): VLAN User Priority.
1209
1210set stat_qmap
1211~~~~~~~~~~~~~
1212
1213Set statistics mapping (qmapping 0..15) for RX/TX queue on port::
1214
1215   testpmd> set stat_qmap (tx|rx) (port_id) (queue_id) (qmapping)
1216
1217For example, to set rx queue 2 on port 0 to mapping 5::
1218
1219   testpmd>set stat_qmap rx 0 2 5
1220
1221set xstats-hide-zero
1222~~~~~~~~~~~~~~~~~~~~
1223
1224Set the option to hide zero values for xstats display::
1225
1226	testpmd> set xstats-hide-zero on|off
1227
1228.. note::
1229
1230	By default, the zero values are displayed for xstats.
1231
1232set port - rx/tx (for VF)
1233~~~~~~~~~~~~~~~~~~~~~~~~~
1234
1235Set VF receive/transmit from a port::
1236
1237   testpmd> set port (port_id) vf (vf_id) (rx|tx) (on|off)
1238
1239set port - mac address filter (for VF)
1240~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1241
1242Add/Remove unicast or multicast MAC addr filter for a VF::
1243
1244   testpmd> set port (port_id) vf (vf_id) (mac_addr) \
1245            (exact-mac|exact-mac-vlan|hashmac|hashmac-vlan) (on|off)
1246
1247set port - rx mode(for VF)
1248~~~~~~~~~~~~~~~~~~~~~~~~~~
1249
1250Set the VF receive mode of a port::
1251
1252   testpmd> set port (port_id) vf (vf_id) \
1253            rxmode (AUPE|ROPE|BAM|MPE) (on|off)
1254
1255The available receive modes are:
1256
1257* ``AUPE``: Accepts untagged VLAN.
1258
1259* ``ROPE``: Accepts unicast hash.
1260
1261* ``BAM``: Accepts broadcast packets.
1262
1263* ``MPE``: Accepts all multicast packets.
1264
1265set port - tx_rate (for Queue)
1266~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1267
1268Set TX rate limitation for a queue on a port::
1269
1270   testpmd> set port (port_id) queue (queue_id) rate (rate_value)
1271
1272set port - tx_rate (for VF)
1273~~~~~~~~~~~~~~~~~~~~~~~~~~~
1274
1275Set TX rate limitation for queues in VF on a port::
1276
1277   testpmd> set port (port_id) vf (vf_id) rate (rate_value) queue_mask (queue_mask)
1278
1279set port - mirror rule
1280~~~~~~~~~~~~~~~~~~~~~~
1281
1282Set pool or vlan type mirror rule for a port::
1283
1284   testpmd> set port (port_id) mirror-rule (rule_id) \
1285            (pool-mirror-up|pool-mirror-down|vlan-mirror) \
1286            (poolmask|vlanid[,vlanid]*) dst-pool (pool_id) (on|off)
1287
1288Set link mirror rule for a port::
1289
1290   testpmd> set port (port_id) mirror-rule (rule_id) \
1291           (uplink-mirror|downlink-mirror) dst-pool (pool_id) (on|off)
1292
1293For example to enable mirror traffic with vlan 0,1 to pool 0::
1294
1295   set port 0 mirror-rule 0 vlan-mirror 0,1 dst-pool 0 on
1296
1297reset port - mirror rule
1298~~~~~~~~~~~~~~~~~~~~~~~~
1299
1300Reset a mirror rule for a port::
1301
1302   testpmd> reset port (port_id) mirror-rule (rule_id)
1303
1304set flush_rx
1305~~~~~~~~~~~~
1306
1307Set the flush on RX streams before forwarding.
1308The default is flush ``on``.
1309Mainly used with PCAP drivers to turn off the default behavior of flushing the first 512 packets on RX streams::
1310
1311   testpmd> set flush_rx off
1312
1313set bypass mode
1314~~~~~~~~~~~~~~~
1315
1316Set the bypass mode for the lowest port on bypass enabled NIC::
1317
1318   testpmd> set bypass mode (normal|bypass|isolate) (port_id)
1319
1320set bypass event
1321~~~~~~~~~~~~~~~~
1322
1323Set the event required to initiate specified bypass mode for the lowest port on a bypass enabled::
1324
1325   testpmd> set bypass event (timeout|os_on|os_off|power_on|power_off) \
1326            mode (normal|bypass|isolate) (port_id)
1327
1328Where:
1329
1330* ``timeout``: Enable bypass after watchdog timeout.
1331
1332* ``os_on``: Enable bypass when OS/board is powered on.
1333
1334* ``os_off``: Enable bypass when OS/board is powered off.
1335
1336* ``power_on``: Enable bypass when power supply is turned on.
1337
1338* ``power_off``: Enable bypass when power supply is turned off.
1339
1340
1341set bypass timeout
1342~~~~~~~~~~~~~~~~~~
1343
1344Set the bypass watchdog timeout to ``n`` seconds where 0 = instant::
1345
1346   testpmd> set bypass timeout (0|1.5|2|3|4|8|16|32)
1347
1348show bypass config
1349~~~~~~~~~~~~~~~~~~
1350
1351Show the bypass configuration for a bypass enabled NIC using the lowest port on the NIC::
1352
1353   testpmd> show bypass config (port_id)
1354
1355set link up
1356~~~~~~~~~~~
1357
1358Set link up for a port::
1359
1360   testpmd> set link-up port (port id)
1361
1362set link down
1363~~~~~~~~~~~~~
1364
1365Set link down for a port::
1366
1367   testpmd> set link-down port (port id)
1368
1369E-tag set
1370~~~~~~~~~
1371
1372Enable E-tag insertion for a VF on a port::
1373
1374   testpmd> E-tag set insertion on port-tag-id (value) port (port_id) vf (vf_id)
1375
1376Disable E-tag insertion for a VF on a port::
1377
1378   testpmd> E-tag set insertion off port (port_id) vf (vf_id)
1379
1380Enable/disable E-tag stripping on a port::
1381
1382   testpmd> E-tag set stripping (on|off) port (port_id)
1383
1384Enable/disable E-tag based forwarding on a port::
1385
1386   testpmd> E-tag set forwarding (on|off) port (port_id)
1387
1388Add an E-tag forwarding filter on a port::
1389
1390   testpmd> E-tag set filter add e-tag-id (value) dst-pool (pool_id) port (port_id)
1391
1392Delete an E-tag forwarding filter on a port::
1393   testpmd> E-tag set filter del e-tag-id (value) port (port_id)
1394
1395ddp add
1396~~~~~~~
1397
1398Load a dynamic device personalization (DDP) package::
1399
1400   testpmd> ddp add (port_id) (package_path[,output_path])
1401
1402ddp del
1403~~~~~~~
1404
1405Delete a dynamic device personalization package::
1406
1407   testpmd> ddp del (port_id) (package_path)
1408
1409ptype mapping
1410~~~~~~~~~~~~~
1411
1412List all items from the ptype mapping table::
1413
1414   testpmd> ptype mapping get (port_id) (valid_only)
1415
1416Where:
1417
1418* ``valid_only``: A flag indicates if only list valid items(=1) or all itemss(=0).
1419
1420Replace a specific or a group of software defined ptype with a new one::
1421
1422   testpmd> ptype mapping replace  (port_id) (target) (mask) (pkt_type)
1423
1424where:
1425
1426* ``target``: A specific software ptype or a mask to represent a group of software ptypes.
1427
1428* ``mask``: A flag indicate if "target" is a specific software ptype(=0) or a ptype mask(=1).
1429
1430* ``pkt_type``: The new software ptype to replace the old ones.
1431
1432Update hardware defined ptype to software defined packet type mapping table::
1433
1434   testpmd> ptype mapping update (port_id) (hw_ptype) (sw_ptype)
1435
1436where:
1437
1438* ``hw_ptype``: hardware ptype as the index of the ptype mapping table.
1439
1440* ``sw_ptype``: software ptype as the value of the ptype mapping table.
1441
1442Reset ptype mapping table::
1443
1444   testpmd> ptype mapping reset (port_id)
1445
1446Port Functions
1447--------------
1448
1449The following sections show functions for configuring ports.
1450
1451.. note::
1452
1453   Port configuration changes only become active when forwarding is started/restarted.
1454
1455port attach
1456~~~~~~~~~~~
1457
1458Attach a port specified by pci address or virtual device args::
1459
1460   testpmd> port attach (identifier)
1461
1462To attach a new pci device, the device should be recognized by kernel first.
1463Then it should be moved under DPDK management.
1464Finally the port can be attached to testpmd.
1465
1466For example, to move a pci device using ixgbe under DPDK management:
1467
1468.. code-block:: console
1469
1470   # Check the status of the available devices.
1471   ./usertools/dpdk-devbind.py --status
1472
1473   Network devices using DPDK-compatible driver
1474   ============================================
1475   <none>
1476
1477   Network devices using kernel driver
1478   ===================================
1479   0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused=
1480
1481
1482   # Bind the device to igb_uio.
1483   sudo ./usertools/dpdk-devbind.py -b igb_uio 0000:0a:00.0
1484
1485
1486   # Recheck the status of the devices.
1487   ./usertools/dpdk-devbind.py --status
1488   Network devices using DPDK-compatible driver
1489   ============================================
1490   0000:0a:00.0 '82599ES 10-Gigabit' drv=igb_uio unused=
1491
1492To attach a port created by virtual device, above steps are not needed.
1493
1494For example, to attach a port whose pci address is 0000:0a:00.0.
1495
1496.. code-block:: console
1497
1498   testpmd> port attach 0000:0a:00.0
1499   Attaching a new port...
1500   EAL: PCI device 0000:0a:00.0 on NUMA socket -1
1501   EAL:   probe driver: 8086:10fb rte_ixgbe_pmd
1502   EAL:   PCI memory mapped at 0x7f83bfa00000
1503   EAL:   PCI memory mapped at 0x7f83bfa80000
1504   PMD: eth_ixgbe_dev_init(): MAC: 2, PHY: 18, SFP+: 5
1505   PMD: eth_ixgbe_dev_init(): port 0 vendorID=0x8086 deviceID=0x10fb
1506   Port 0 is attached. Now total ports is 1
1507   Done
1508
1509For example, to attach a port created by pcap PMD.
1510
1511.. code-block:: console
1512
1513   testpmd> port attach net_pcap0
1514   Attaching a new port...
1515   PMD: Initializing pmd_pcap for net_pcap0
1516   PMD: Creating pcap-backed ethdev on numa socket 0
1517   Port 0 is attached. Now total ports is 1
1518   Done
1519
1520In this case, identifier is ``net_pcap0``.
1521This identifier format is the same as ``--vdev`` format of DPDK applications.
1522
1523For example, to re-attach a bonded port which has been previously detached,
1524the mode and slave parameters must be given.
1525
1526.. code-block:: console
1527
1528   testpmd> port attach net_bond_0,mode=0,slave=1
1529   Attaching a new port...
1530   EAL: Initializing pmd_bond for net_bond_0
1531   EAL: Create bonded device net_bond_0 on port 0 in mode 0 on socket 0.
1532   Port 0 is attached. Now total ports is 1
1533   Done
1534
1535
1536port detach
1537~~~~~~~~~~~
1538
1539Detach a specific port::
1540
1541   testpmd> port detach (port_id)
1542
1543Before detaching a port, the port should be stopped and closed.
1544
1545For example, to detach a pci device port 0.
1546
1547.. code-block:: console
1548
1549   testpmd> port stop 0
1550   Stopping ports...
1551   Done
1552   testpmd> port close 0
1553   Closing ports...
1554   Done
1555
1556   testpmd> port detach 0
1557   Detaching a port...
1558   EAL: PCI device 0000:0a:00.0 on NUMA socket -1
1559   EAL:   remove driver: 8086:10fb rte_ixgbe_pmd
1560   EAL:   PCI memory unmapped at 0x7f83bfa00000
1561   EAL:   PCI memory unmapped at 0x7f83bfa80000
1562   Done
1563
1564
1565For example, to detach a virtual device port 0.
1566
1567.. code-block:: console
1568
1569   testpmd> port stop 0
1570   Stopping ports...
1571   Done
1572   testpmd> port close 0
1573   Closing ports...
1574   Done
1575
1576   testpmd> port detach 0
1577   Detaching a port...
1578   PMD: Closing pcap ethdev on numa socket 0
1579   Port 'net_pcap0' is detached. Now total ports is 0
1580   Done
1581
1582To remove a pci device completely from the system, first detach the port from testpmd.
1583Then the device should be moved under kernel management.
1584Finally the device can be removed using kernel pci hotplug functionality.
1585
1586For example, to move a pci device under kernel management:
1587
1588.. code-block:: console
1589
1590   sudo ./usertools/dpdk-devbind.py -b ixgbe 0000:0a:00.0
1591
1592   ./usertools/dpdk-devbind.py --status
1593
1594   Network devices using DPDK-compatible driver
1595   ============================================
1596   <none>
1597
1598   Network devices using kernel driver
1599   ===================================
1600   0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused=igb_uio
1601
1602To remove a port created by a virtual device, above steps are not needed.
1603
1604port start
1605~~~~~~~~~~
1606
1607Start all ports or a specific port::
1608
1609   testpmd> port start (port_id|all)
1610
1611port stop
1612~~~~~~~~~
1613
1614Stop all ports or a specific port::
1615
1616   testpmd> port stop (port_id|all)
1617
1618port close
1619~~~~~~~~~~
1620
1621Close all ports or a specific port::
1622
1623   testpmd> port close (port_id|all)
1624
1625port start/stop queue
1626~~~~~~~~~~~~~~~~~~~~~
1627
1628Start/stop a rx/tx queue on a specific port::
1629
1630   testpmd> port (port_id) (rxq|txq) (queue_id) (start|stop)
1631
1632Only take effect when port is started.
1633
1634port config - speed
1635~~~~~~~~~~~~~~~~~~~
1636
1637Set the speed and duplex mode for all ports or a specific port::
1638
1639   testpmd> port config (port_id|all) speed (10|100|1000|10000|25000|40000|50000|100000|auto) \
1640            duplex (half|full|auto)
1641
1642port config - queues/descriptors
1643~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1644
1645Set number of queues/descriptors for rxq, txq, rxd and txd::
1646
1647   testpmd> port config all (rxq|txq|rxd|txd) (value)
1648
1649This is equivalent to the ``--rxq``, ``--txq``, ``--rxd`` and ``--txd`` command-line options.
1650
1651port config - max-pkt-len
1652~~~~~~~~~~~~~~~~~~~~~~~~~
1653
1654Set the maximum packet length::
1655
1656   testpmd> port config all max-pkt-len (value)
1657
1658This is equivalent to the ``--max-pkt-len`` command-line option.
1659
1660port config - CRC Strip
1661~~~~~~~~~~~~~~~~~~~~~~~
1662
1663Set hardware CRC stripping on or off for all ports::
1664
1665   testpmd> port config all crc-strip (on|off)
1666
1667CRC stripping is on by default.
1668
1669The ``off`` option is equivalent to the ``--disable-crc-strip`` command-line option.
1670
1671port config - scatter
1672~~~~~~~~~~~~~~~~~~~~~~~
1673
1674Set RX scatter mode on or off for all ports::
1675
1676   testpmd> port config all scatter (on|off)
1677
1678RX scatter mode is off by default.
1679
1680The ``on`` option is equivalent to the ``--enable-scatter`` command-line option.
1681
1682port config - RX Checksum
1683~~~~~~~~~~~~~~~~~~~~~~~~~
1684
1685Set hardware RX checksum offload to on or off for all ports::
1686
1687   testpmd> port config all rx-cksum (on|off)
1688
1689Checksum offload is off by default.
1690
1691The ``on`` option is equivalent to the ``--enable-rx-cksum`` command-line option.
1692
1693port config - VLAN
1694~~~~~~~~~~~~~~~~~~
1695
1696Set hardware VLAN on or off for all ports::
1697
1698   testpmd> port config all hw-vlan (on|off)
1699
1700Hardware VLAN is on by default.
1701
1702The ``off`` option is equivalent to the ``--disable-hw-vlan`` command-line option.
1703
1704port config - VLAN filter
1705~~~~~~~~~~~~~~~~~~~~~~~~~
1706
1707Set hardware VLAN filter on or off for all ports::
1708
1709   testpmd> port config all hw-vlan-filter (on|off)
1710
1711Hardware VLAN filter is on by default.
1712
1713The ``off`` option is equivalent to the ``--disable-hw-vlan-filter`` command-line option.
1714
1715port config - VLAN strip
1716~~~~~~~~~~~~~~~~~~~~~~~~
1717
1718Set hardware VLAN strip on or off for all ports::
1719
1720   testpmd> port config all hw-vlan-strip (on|off)
1721
1722Hardware VLAN strip is on by default.
1723
1724The ``off`` option is equivalent to the ``--disable-hw-vlan-strip`` command-line option.
1725
1726port config - VLAN extend
1727~~~~~~~~~~~~~~~~~~~~~~~~~
1728
1729Set hardware VLAN extend on or off for all ports::
1730
1731   testpmd> port config all hw-vlan-extend (on|off)
1732
1733Hardware VLAN extend is off by default.
1734
1735The ``off`` option is equivalent to the ``--disable-hw-vlan-extend`` command-line option.
1736
1737port config - Drop Packets
1738~~~~~~~~~~~~~~~~~~~~~~~~~~
1739
1740Set packet drop for packets with no descriptors on or off for all ports::
1741
1742   testpmd> port config all drop-en (on|off)
1743
1744Packet dropping for packets with no descriptors is off by default.
1745
1746The ``on`` option is equivalent to the ``--enable-drop-en`` command-line option.
1747
1748port config - RSS
1749~~~~~~~~~~~~~~~~~
1750
1751Set the RSS (Receive Side Scaling) mode on or off::
1752
1753   testpmd> port config all rss (all|ip|tcp|udp|sctp|ether|port|vxlan|geneve|nvgre|none)
1754
1755RSS is on by default.
1756
1757The ``none`` option is equivalent to the ``--disable-rss`` command-line option.
1758
1759port config - RSS Reta
1760~~~~~~~~~~~~~~~~~~~~~~
1761
1762Set the RSS (Receive Side Scaling) redirection table::
1763
1764   testpmd> port config all rss reta (hash,queue)[,(hash,queue)]
1765
1766port config - DCB
1767~~~~~~~~~~~~~~~~~
1768
1769Set the DCB mode for an individual port::
1770
1771   testpmd> port config (port_id) dcb vt (on|off) (traffic_class) pfc (on|off)
1772
1773The traffic class should be 4 or 8.
1774
1775port config - Burst
1776~~~~~~~~~~~~~~~~~~~
1777
1778Set the number of packets per burst::
1779
1780   testpmd> port config all burst (value)
1781
1782This is equivalent to the ``--burst`` command-line option.
1783
1784port config - Threshold
1785~~~~~~~~~~~~~~~~~~~~~~~
1786
1787Set thresholds for TX/RX queues::
1788
1789   testpmd> port config all (threshold) (value)
1790
1791Where the threshold type can be:
1792
1793* ``txpt:`` Set the prefetch threshold register of the TX rings, 0 <= value <= 255.
1794
1795* ``txht:`` Set the host threshold register of the TX rings, 0 <= value <= 255.
1796
1797* ``txwt:`` Set the write-back threshold register of the TX rings, 0 <= value <= 255.
1798
1799* ``rxpt:`` Set the prefetch threshold register of the RX rings, 0 <= value <= 255.
1800
1801* ``rxht:`` Set the host threshold register of the RX rings, 0 <= value <= 255.
1802
1803* ``rxwt:`` Set the write-back threshold register of the RX rings, 0 <= value <= 255.
1804
1805* ``txfreet:`` Set the transmit free threshold of the TX rings, 0 <= value <= txd.
1806
1807* ``rxfreet:`` Set the transmit free threshold of the RX rings, 0 <= value <= rxd.
1808
1809* ``txrst:`` Set the transmit RS bit threshold of TX rings, 0 <= value <= txd.
1810
1811These threshold options are also available from the command-line.
1812
1813port config - E-tag
1814~~~~~~~~~~~~~~~~~~~
1815
1816Set the value of ether-type for E-tag::
1817
1818   testpmd> port config (port_id|all) l2-tunnel E-tag ether-type (value)
1819
1820Enable/disable the E-tag support::
1821
1822   testpmd> port config (port_id|all) l2-tunnel E-tag (enable|disable)
1823
1824port config pctype mapping
1825~~~~~~~~~~~~~~~~~~~~~~~~~~
1826
1827Reset pctype mapping table::
1828
1829   testpmd> port config (port_id) pctype mapping reset
1830
1831Update hardware defined pctype to software defined flow type mapping table::
1832
1833   testpmd> port config (port_id) pctype mapping update (pctype_id_0[,pctype_id_1]*) (flow_type_id)
1834
1835where:
1836
1837* ``pctype_id_x``: hardware pctype id as index of bit in bitmask value of the pctype mapping table.
1838
1839* ``flow_type_id``: software flow type id as the index of the pctype mapping table.
1840
1841
1842Link Bonding Functions
1843----------------------
1844
1845The Link Bonding functions make it possible to dynamically create and
1846manage link bonding devices from within testpmd interactive prompt.
1847
1848create bonded device
1849~~~~~~~~~~~~~~~~~~~~
1850
1851Create a new bonding device::
1852
1853   testpmd> create bonded device (mode) (socket)
1854
1855For example, to create a bonded device in mode 1 on socket 0::
1856
1857   testpmd> create bonded 1 0
1858   created new bonded device (port X)
1859
1860add bonding slave
1861~~~~~~~~~~~~~~~~~
1862
1863Adds Ethernet device to a Link Bonding device::
1864
1865   testpmd> add bonding slave (slave id) (port id)
1866
1867For example, to add Ethernet device (port 6) to a Link Bonding device (port 10)::
1868
1869   testpmd> add bonding slave 6 10
1870
1871
1872remove bonding slave
1873~~~~~~~~~~~~~~~~~~~~
1874
1875Removes an Ethernet slave device from a Link Bonding device::
1876
1877   testpmd> remove bonding slave (slave id) (port id)
1878
1879For example, to remove Ethernet slave device (port 6) to a Link Bonding device (port 10)::
1880
1881   testpmd> remove bonding slave 6 10
1882
1883set bonding mode
1884~~~~~~~~~~~~~~~~
1885
1886Set the Link Bonding mode of a Link Bonding device::
1887
1888   testpmd> set bonding mode (value) (port id)
1889
1890For example, to set the bonding mode of a Link Bonding device (port 10) to broadcast (mode 3)::
1891
1892   testpmd> set bonding mode 3 10
1893
1894set bonding primary
1895~~~~~~~~~~~~~~~~~~~
1896
1897Set an Ethernet slave device as the primary device on a Link Bonding device::
1898
1899   testpmd> set bonding primary (slave id) (port id)
1900
1901For example, to set the Ethernet slave device (port 6) as the primary port of a Link Bonding device (port 10)::
1902
1903   testpmd> set bonding primary 6 10
1904
1905set bonding mac
1906~~~~~~~~~~~~~~~
1907
1908Set the MAC address of a Link Bonding device::
1909
1910   testpmd> set bonding mac (port id) (mac)
1911
1912For example, to set the MAC address of a Link Bonding device (port 10) to 00:00:00:00:00:01::
1913
1914   testpmd> set bonding mac 10 00:00:00:00:00:01
1915
1916set bonding xmit_balance_policy
1917~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1918
1919Set the transmission policy for a Link Bonding device when it is in Balance XOR mode::
1920
1921   testpmd> set bonding xmit_balance_policy (port_id) (l2|l23|l34)
1922
1923For example, set a Link Bonding device (port 10) to use a balance policy of layer 3+4 (IP addresses & UDP ports)::
1924
1925   testpmd> set bonding xmit_balance_policy 10 l34
1926
1927
1928set bonding mon_period
1929~~~~~~~~~~~~~~~~~~~~~~
1930
1931Set the link status monitoring polling period in milliseconds for a bonding device.
1932
1933This adds support for PMD slave devices which do not support link status interrupts.
1934When the mon_period is set to a value greater than 0 then all PMD's which do not support
1935link status ISR will be queried every polling interval to check if their link status has changed::
1936
1937   testpmd> set bonding mon_period (port_id) (value)
1938
1939For example, to set the link status monitoring polling period of bonded device (port 5) to 150ms::
1940
1941   testpmd> set bonding mon_period 5 150
1942
1943
1944set bonding lacp dedicated_queue
1945~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1946
1947Enable dedicated tx/rx queues on bonding devices slaves to handle LACP control plane traffic
1948when in mode 4 (link-aggregration-802.3ad)::
1949
1950   testpmd> set bonding lacp dedicated_queues (port_id) (enable|disable)
1951
1952
1953set bonding agg_mode
1954~~~~~~~~~~~~~~~~~~~~
1955
1956Enable one of the specific aggregators mode when in mode 4 (link-aggregration-802.3ad)::
1957
1958   testpmd> set bonding agg_mode (port_id) (bandwidth|count|stable)
1959
1960
1961show bonding config
1962~~~~~~~~~~~~~~~~~~~
1963
1964Show the current configuration of a Link Bonding device::
1965
1966   testpmd> show bonding config (port id)
1967
1968For example,
1969to show the configuration a Link Bonding device (port 9) with 3 slave devices (1, 3, 4)
1970in balance mode with a transmission policy of layer 2+3::
1971
1972   testpmd> show bonding config 9
1973        Bonding mode: 2
1974        Balance Xmit Policy: BALANCE_XMIT_POLICY_LAYER23
1975        Slaves (3): [1 3 4]
1976        Active Slaves (3): [1 3 4]
1977        Primary: [3]
1978
1979
1980Register Functions
1981------------------
1982
1983The Register Functions can be used to read from and write to registers on the network card referenced by a port number.
1984This is mainly useful for debugging purposes.
1985Reference should be made to the appropriate datasheet for the network card for details on the register addresses
1986and fields that can be accessed.
1987
1988read reg
1989~~~~~~~~
1990
1991Display the value of a port register::
1992
1993   testpmd> read reg (port_id) (address)
1994
1995For example, to examine the Flow Director control register (FDIRCTL, 0x0000EE000) on an Intel 82599 10 GbE Controller::
1996
1997   testpmd> read reg 0 0xEE00
1998   port 0 PCI register at offset 0xEE00: 0x4A060029 (1241907241)
1999
2000read regfield
2001~~~~~~~~~~~~~
2002
2003Display a port register bit field::
2004
2005   testpmd> read regfield (port_id) (address) (bit_x) (bit_y)
2006
2007For example, reading the lowest two bits from the register in the example above::
2008
2009   testpmd> read regfield 0 0xEE00 0 1
2010   port 0 PCI register at offset 0xEE00: bits[0, 1]=0x1 (1)
2011
2012read regbit
2013~~~~~~~~~~~
2014
2015Display a single port register bit::
2016
2017   testpmd> read regbit (port_id) (address) (bit_x)
2018
2019For example, reading the lowest bit from the register in the example above::
2020
2021   testpmd> read regbit 0 0xEE00 0
2022   port 0 PCI register at offset 0xEE00: bit 0=1
2023
2024write reg
2025~~~~~~~~~
2026
2027Set the value of a port register::
2028
2029   testpmd> write reg (port_id) (address) (value)
2030
2031For example, to clear a register::
2032
2033   testpmd> write reg 0 0xEE00 0x0
2034   port 0 PCI register at offset 0xEE00: 0x00000000 (0)
2035
2036write regfield
2037~~~~~~~~~~~~~~
2038
2039Set bit field of a port register::
2040
2041   testpmd> write regfield (port_id) (address) (bit_x) (bit_y) (value)
2042
2043For example, writing to the register cleared in the example above::
2044
2045   testpmd> write regfield 0 0xEE00 0 1 2
2046   port 0 PCI register at offset 0xEE00: 0x00000002 (2)
2047
2048write regbit
2049~~~~~~~~~~~~
2050
2051Set single bit value of a port register::
2052
2053   testpmd> write regbit (port_id) (address) (bit_x) (value)
2054
2055For example, to set the high bit in the register from the example above::
2056
2057   testpmd> write regbit 0 0xEE00 31 1
2058   port 0 PCI register at offset 0xEE00: 0x8000000A (2147483658)
2059
2060Traffic Metering and Policing
2061-----------------------------
2062
2063The following section shows functions for configuring traffic metering and
2064policing on the ethernet device through the use of generic ethdev API.
2065
2066show port traffic management capability
2067~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2068
2069Show traffic metering and policing capability of the port::
2070
2071   testpmd> show port meter cap (port_id)
2072
2073add port meter profile (srTCM rfc2967)
2074~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2075
2076Add meter profile (srTCM rfc2697) to the ethernet device::
2077
2078   testpmd> add port meter profile srtcm_rfc2697 (port_id) (profile_id) \
2079   (cir) (cbs) (ebs)
2080
2081where:
2082
2083* ``profile_id``: ID for the meter profile.
2084* ``cir``: Committed Information Rate (CIR) (bytes/second).
2085* ``cbs``: Committed Burst Size (CBS) (bytes).
2086* ``ebs``: Excess Burst Size (EBS) (bytes).
2087
2088add port meter profile (trTCM rfc2968)
2089~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2090
2091Add meter profile (srTCM rfc2698) to the ethernet device::
2092
2093   testpmd> add port meter profile trtcm_rfc2698 (port_id) (profile_id) \
2094   (cir) (pir) (cbs) (pbs)
2095
2096where:
2097
2098* ``profile_id``: ID for the meter profile.
2099* ``cir``: Committed information rate (bytes/second).
2100* ``pir``: Peak information rate (bytes/second).
2101* ``cbs``: Committed burst size (bytes).
2102* ``pbs``: Peak burst size (bytes).
2103
2104add port meter profile (trTCM rfc4115)
2105~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2106
2107Add meter profile (trTCM rfc4115) to the ethernet device::
2108
2109   testpmd> add port meter profile trtcm_rfc4115 (port_id) (profile_id) \
2110   (cir) (eir) (cbs) (ebs)
2111
2112where:
2113
2114* ``profile_id``: ID for the meter profile.
2115* ``cir``: Committed information rate (bytes/second).
2116* ``eir``: Excess information rate (bytes/second).
2117* ``cbs``: Committed burst size (bytes).
2118* ``ebs``: Excess burst size (bytes).
2119
2120delete port meter profile
2121~~~~~~~~~~~~~~~~~~~~~~~~~
2122
2123Delete meter profile from the ethernet device::
2124
2125   testpmd> del port meter profile (port_id) (profile_id)
2126
2127create port meter
2128~~~~~~~~~~~~~~~~~
2129
2130Create new meter object for the ethernet device::
2131
2132   testpmd> create port meter (port_id) (mtr_id) (profile_id) \
2133   (meter_enable) (g_action) (y_action) (r_action) (stats_mask) (shared) \
2134   (use_pre_meter_color) [(dscp_tbl_entry0) (dscp_tbl_entry1)...\
2135   (dscp_tbl_entry63)]
2136
2137where:
2138
2139* ``mtr_id``: meter object ID.
2140* ``profile_id``: ID for the meter profile.
2141* ``meter_enable``: When this parameter has a non-zero value, the meter object
2142  gets enabled at the time of creation, otherwise remains disabled.
2143* ``g_action``: Policer action for the packet with green color.
2144* ``y_action``: Policer action for the packet with yellow color.
2145* ``r_action``: Policer action for the packet with red color.
2146* ``stats_mask``: Mask of statistics counter types to be enabled for the
2147  meter object.
2148* ``shared``:  When this parameter has a non-zero value, the meter object is
2149  shared by multiple flows. Otherwise, meter object is used by single flow.
2150* ``use_pre_meter_color``: When this parameter has a non-zero value, the
2151  input color for the current meter object is determined by the latest meter
2152  object in the same flow. Otherwise, the current meter object uses the
2153  *dscp_table* to determine the input color.
2154* ``dscp_tbl_entryx``: DSCP table entry x providing meter providing input
2155  color, 0 <= x <= 63.
2156
2157enable port meter
2158~~~~~~~~~~~~~~~~~
2159
2160Enable meter for the ethernet device::
2161
2162   testpmd> enable port meter (port_id) (mtr_id)
2163
2164disable port meter
2165~~~~~~~~~~~~~~~~~~
2166
2167Disable meter for the ethernet device::
2168
2169   testpmd> disable port meter (port_id) (mtr_id)
2170
2171delete port meter
2172~~~~~~~~~~~~~~~~~
2173
2174Delete meter for the ethernet device::
2175
2176   testpmd> del port meter (port_id) (mtr_id)
2177
2178Set port meter profile
2179~~~~~~~~~~~~~~~~~~~~~~
2180
2181Set meter profile for the ethernet device::
2182
2183   testpmd> set port meter profile (port_id) (mtr_id) (profile_id)
2184
2185set port meter dscp table
2186~~~~~~~~~~~~~~~~~~~~~~~~~
2187
2188Set meter dscp table for the ethernet device::
2189
2190   testpmd> set port meter dscp table (port_id) (mtr_id) [(dscp_tbl_entry0) \
2191   (dscp_tbl_entry1)...(dscp_tbl_entry63)]
2192
2193set port meter policer action
2194~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2195
2196Set meter policer action for the ethernet device::
2197
2198   testpmd> set port meter policer action (port_id) (mtr_id) (action_mask) \
2199   (action0) [(action1) (action1)]
2200
2201where:
2202
2203* ``action_mask``: Bit mask indicating which policer actions need to be
2204  updated. One or more policer actions can be updated in a single function
2205  invocation. To update the policer action associated with color C, bit
2206  (1 << C) needs to be set in *action_mask* and element at position C
2207  in the *actions* array needs to be valid.
2208* ``actionx``: Policer action for the color x,
2209  RTE_MTR_GREEN <= x < RTE_MTR_COLORS
2210
2211set port meter stats mask
2212~~~~~~~~~~~~~~~~~~~~~~~~~
2213
2214Set meter stats mask for the ethernet device::
2215
2216   testpmd> set port meter stats mask (port_id) (mtr_id) (stats_mask)
2217
2218where:
2219
2220* ``stats_mask``: Bit mask indicating statistics counter types to be enabled.
2221
2222show port meter stats
2223~~~~~~~~~~~~~~~~~~~~~
2224
2225Show meter stats of the ethernet device::
2226
2227   testpmd> show port meter stats (port_id) (mtr_id) (clear)
2228
2229where:
2230
2231* ``clear``: Flag that indicates whether the statistics counters should
2232  be cleared (i.e. set to zero) immediately after they have been read or not.
2233
2234Traffic Management
2235------------------
2236
2237The following section shows functions for configuring traffic management on
2238on the ethernet device through the use of generic TM API.
2239
2240show port traffic management capability
2241~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2242
2243Show traffic management capability of the port::
2244
2245   testpmd> show port tm cap (port_id)
2246
2247show port traffic management capability (hierarchy level)
2248~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2249
2250Show traffic management hierarchy level capability of the port::
2251
2252   testpmd> show port tm cap (port_id) (level_id)
2253
2254show port traffic management capability (hierarchy node level)
2255~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2256
2257Show the traffic management hierarchy node capability of the port::
2258
2259   testpmd> show port tm cap (port_id) (node_id)
2260
2261show port traffic management hierarchy node type
2262~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2263
2264Show the port traffic management hierarchy node type::
2265
2266   testpmd> show port tm node type (port_id) (node_id)
2267
2268show port traffic management hierarchy node stats
2269~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2270
2271Show the port traffic management hierarchy node statistics::
2272
2273   testpmd> show port tm node stats (port_id) (node_id) (clear)
2274
2275where:
2276
2277* ``clear``: When this parameter has a non-zero value, the statistics counters
2278  are cleared (i.e. set to zero) immediately after they have been read,
2279  otherwise the statistics counters are left untouched.
2280
2281Add port traffic management private shaper profile
2282~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2283
2284Add the port traffic management private shaper profile::
2285
2286   testpmd> add port tm node shaper profile (port_id) (shaper_profile_id) \
2287   (tb_rate) (tb_size) (packet_length_adjust)
2288
2289where:
2290
2291* ``shaper_profile id``: Shaper profile ID for the new profile.
2292* ``tb_rate``: Token bucket rate (bytes per second).
2293* ``tb_size``: Token bucket size (bytes).
2294* ``packet_length_adjust``: The value (bytes) to be added to the length of
2295  each packet for the purpose of shaping. This parameter value can be used to
2296  correct the packet length with the framing overhead bytes that are consumed
2297  on the wire.
2298
2299Delete port traffic management private shaper profile
2300~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2301
2302Delete the port traffic management private shaper::
2303
2304   testpmd> del port tm node shaper profile (port_id) (shaper_profile_id)
2305
2306where:
2307
2308* ``shaper_profile id``: Shaper profile ID that needs to be deleted.
2309
2310Add port traffic management shared shaper
2311~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2312
2313Create the port traffic management shared shaper::
2314
2315   testpmd> add port tm node shared shaper (port_id) (shared_shaper_id) \
2316   (shaper_profile_id)
2317
2318where:
2319
2320* ``shared_shaper_id``: Shared shaper ID to be created.
2321* ``shaper_profile id``: Shaper profile ID for shared shaper.
2322
2323Set port traffic management shared shaper
2324~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2325
2326Update the port traffic management shared shaper::
2327
2328   testpmd> set port tm node shared shaper (port_id) (shared_shaper_id) \
2329   (shaper_profile_id)
2330
2331where:
2332
2333* ``shared_shaper_id``: Shared shaper ID to be update.
2334* ``shaper_profile id``: Shaper profile ID for shared shaper.
2335
2336Delete port traffic management shared shaper
2337~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2338
2339Delete the port traffic management shared shaper::
2340
2341   testpmd> del port tm node shared shaper (port_id) (shared_shaper_id)
2342
2343where:
2344
2345* ``shared_shaper_id``: Shared shaper ID to be deleted.
2346
2347Set port traffic management hiearchy node private shaper
2348~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2349
2350set the port traffic management hierarchy node private shaper::
2351
2352   testpmd> set port tm node shaper profile (port_id) (node_id) \
2353   (shaper_profile_id)
2354
2355where:
2356
2357* ``shaper_profile id``: Private shaper profile ID to be enabled on the
2358  hierarchy node.
2359
2360Add port traffic management WRED profile
2361~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2362
2363Create a new WRED profile::
2364
2365   testpmd> add port tm node wred profile (port_id) (wred_profile_id) \
2366   (color_g) (min_th_g) (max_th_g) (maxp_inv_g) (wq_log2_g) \
2367   (color_y) (min_th_y) (max_th_y) (maxp_inv_y) (wq_log2_y) \
2368   (color_r) (min_th_r) (max_th_r) (maxp_inv_r) (wq_log2_r)
2369
2370where:
2371
2372* ``wred_profile id``: Identifier for the newly create WRED profile
2373* ``color_g``: Packet color (green)
2374* ``min_th_g``: Minimum queue threshold for packet with green color
2375* ``max_th_g``: Minimum queue threshold for packet with green color
2376* ``maxp_inv_g``: Inverse of packet marking probability maximum value (maxp)
2377* ``wq_log2_g``: Negated log2 of queue weight (wq)
2378* ``color_y``: Packet color (yellow)
2379* ``min_th_y``: Minimum queue threshold for packet with yellow color
2380* ``max_th_y``: Minimum queue threshold for packet with yellow color
2381* ``maxp_inv_y``: Inverse of packet marking probability maximum value (maxp)
2382* ``wq_log2_y``: Negated log2 of queue weight (wq)
2383* ``color_r``: Packet color (red)
2384* ``min_th_r``: Minimum queue threshold for packet with yellow color
2385* ``max_th_r``: Minimum queue threshold for packet with yellow color
2386* ``maxp_inv_r``: Inverse of packet marking probability maximum value (maxp)
2387* ``wq_log2_r``: Negated log2 of queue weight (wq)
2388
2389Delete port traffic management WRED profile
2390~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2391
2392Delete the WRED profile::
2393
2394   testpmd> del port tm node wred profile (port_id) (wred_profile_id)
2395
2396Add port traffic management hierarchy nonleaf node
2397~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2398
2399Add nonleaf node to port traffic management hiearchy::
2400
2401   testpmd> add port tm nonleaf node (port_id) (node_id) (parent_node_id) \
2402   (priority) (weight) (level_id) (shaper_profile_id) \
2403   (n_sp_priorities) (stats_mask) (n_shared_shapers) \
2404   [(shared_shaper_0) (shared_shaper_1) ...] \
2405
2406where:
2407
2408* ``parent_node_id``: Node ID of the parent.
2409* ``priority``: Node priority (highest node priority is zero). This is used by
2410  the SP algorithm running on the parent node for scheduling this node.
2411* ``weight``: Node weight (lowest weight is one). The node weight is relative
2412  to the weight sum of all siblings that have the same priority. It is used by
2413  the WFQ algorithm running on the parent node for scheduling this node.
2414* ``level_id``: Hiearchy level of the node.
2415* ``shaper_profile_id``: Shaper profile ID of the private shaper to be used by
2416  the node.
2417* ``n_sp_priorities``: Number of strict priorities.
2418* ``stats_mask``: Mask of statistics counter types to be enabled for this node.
2419* ``n_shared_shapers``: Number of shared shapers.
2420* ``shared_shaper_id``: Shared shaper id.
2421
2422Add port traffic management hierarchy leaf node
2423~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2424
2425Add leaf node to port traffic management hiearchy::
2426
2427   testpmd> add port tm leaf node (port_id) (node_id) (parent_node_id) \
2428   (priority) (weight) (level_id) (shaper_profile_id) \
2429   (cman_mode) (wred_profile_id) (stats_mask) (n_shared_shapers) \
2430   [(shared_shaper_id) (shared_shaper_id) ...] \
2431
2432where:
2433
2434* ``parent_node_id``: Node ID of the parent.
2435* ``priority``: Node priority (highest node priority is zero). This is used by
2436  the SP algorithm running on the parent node for scheduling this node.
2437* ``weight``: Node weight (lowest weight is one). The node weight is relative
2438  to the weight sum of all siblings that have the same priority. It is used by
2439  the WFQ algorithm running on the parent node for scheduling this node.
2440* ``level_id``: Hiearchy level of the node.
2441* ``shaper_profile_id``: Shaper profile ID of the private shaper to be used by
2442  the node.
2443* ``cman_mode``: Congestion management mode to be enabled for this node.
2444* ``wred_profile_id``: WRED profile id to be enabled for this node.
2445* ``stats_mask``: Mask of statistics counter types to be enabled for this node.
2446* ``n_shared_shapers``: Number of shared shapers.
2447* ``shared_shaper_id``: Shared shaper id.
2448
2449Delete port traffic management hierarchy node
2450~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2451
2452Delete node from port traffic management hiearchy::
2453
2454   testpmd> del port tm node (port_id) (node_id)
2455
2456Update port traffic management hierarchy parent node
2457~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2458
2459Update port traffic management hierarchy parent node::
2460
2461   testpmd> set port tm node parent (port_id) (node_id) (parent_node_id) \
2462   (priority) (weight)
2463
2464This function can only be called after the hierarchy commit invocation. Its
2465success depends on the port support for this operation, as advertised through
2466the port capability set. This function is valid for all nodes of the traffic
2467management hierarchy except root node.
2468
2469Commit port traffic management hierarchy
2470~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2471
2472Commit the traffic management hierarchy on the port::
2473
2474   testpmd> port tm hierarchy commit (port_id) (clean_on_fail)
2475
2476where:
2477
2478* ``clean_on_fail``: When set to non-zero, hierarchy is cleared on function
2479  call failure. On the other hand, hierarchy is preserved when this parameter
2480  is equal to zero.
2481
2482Set port traffic management default hierarchy (tm forwarding mode)
2483~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2484
2485set the traffic management default hierarchy on the port::
2486
2487   testpmd> set port tm hierarchy default (port_id)
2488
2489Filter Functions
2490----------------
2491
2492This section details the available filter functions that are available.
2493
2494Note these functions interface the deprecated legacy filtering framework,
2495superseded by *rte_flow*. See `Flow rules management`_.
2496
2497ethertype_filter
2498~~~~~~~~~~~~~~~~~~~~
2499
2500Add or delete a L2 Ethertype filter, which identify packets by their L2 Ethertype mainly assign them to a receive queue::
2501
2502   ethertype_filter (port_id) (add|del) (mac_addr|mac_ignr) (mac_address) \
2503                    ethertype (ether_type) (drop|fwd) queue (queue_id)
2504
2505The available information parameters are:
2506
2507* ``port_id``: The port which the Ethertype filter assigned on.
2508
2509* ``mac_addr``: Compare destination mac address.
2510
2511* ``mac_ignr``: Ignore destination mac address match.
2512
2513* ``mac_address``: Destination mac address to match.
2514
2515* ``ether_type``: The EtherType value want to match,
2516  for example 0x0806 for ARP packet. 0x0800 (IPv4) and 0x86DD (IPv6) are invalid.
2517
2518* ``queue_id``: The receive queue associated with this EtherType filter.
2519  It is meaningless when deleting or dropping.
2520
2521Example, to add/remove an ethertype filter rule::
2522
2523   testpmd> ethertype_filter 0 add mac_ignr 00:11:22:33:44:55 \
2524                             ethertype 0x0806 fwd queue 3
2525
2526   testpmd> ethertype_filter 0 del mac_ignr 00:11:22:33:44:55 \
2527                             ethertype 0x0806 fwd queue 3
2528
25292tuple_filter
2530~~~~~~~~~~~~~~~~~
2531
2532Add or delete a 2-tuple filter,
2533which identifies packets by specific protocol and destination TCP/UDP port
2534and forwards packets into one of the receive queues::
2535
2536   2tuple_filter (port_id) (add|del) dst_port (dst_port_value) \
2537                 protocol (protocol_value) mask (mask_value) \
2538                 tcp_flags (tcp_flags_value) priority (prio_value) \
2539                 queue (queue_id)
2540
2541The available information parameters are:
2542
2543* ``port_id``: The port which the 2-tuple filter assigned on.
2544
2545* ``dst_port_value``: Destination port in L4.
2546
2547* ``protocol_value``: IP L4 protocol.
2548
2549* ``mask_value``: Participates in the match or not by bit for field above, 1b means participate.
2550
2551* ``tcp_flags_value``: TCP control bits. The non-zero value is invalid, when the pro_value is not set to 0x06 (TCP).
2552
2553* ``prio_value``: Priority of this filter.
2554
2555* ``queue_id``: The receive queue associated with this 2-tuple filter.
2556
2557Example, to add/remove an 2tuple filter rule::
2558
2559   testpmd> 2tuple_filter 0 add dst_port 32 protocol 0x06 mask 0x03 \
2560                          tcp_flags 0x02 priority 3 queue 3
2561
2562   testpmd> 2tuple_filter 0 del dst_port 32 protocol 0x06 mask 0x03 \
2563                          tcp_flags 0x02 priority 3 queue 3
2564
25655tuple_filter
2566~~~~~~~~~~~~~~~~~
2567
2568Add or delete a 5-tuple filter,
2569which consists of a 5-tuple (protocol, source and destination IP addresses, source and destination TCP/UDP/SCTP port)
2570and routes packets into one of the receive queues::
2571
2572   5tuple_filter (port_id) (add|del) dst_ip (dst_address) src_ip \
2573                 (src_address) dst_port (dst_port_value) \
2574                 src_port (src_port_value) protocol (protocol_value) \
2575                 mask (mask_value) tcp_flags (tcp_flags_value) \
2576                 priority (prio_value) queue (queue_id)
2577
2578The available information parameters are:
2579
2580* ``port_id``: The port which the 5-tuple filter assigned on.
2581
2582* ``dst_address``: Destination IP address.
2583
2584* ``src_address``: Source IP address.
2585
2586* ``dst_port_value``: TCP/UDP destination port.
2587
2588* ``src_port_value``: TCP/UDP source port.
2589
2590* ``protocol_value``: L4 protocol.
2591
2592* ``mask_value``: Participates in the match or not by bit for field above, 1b means participate
2593
2594* ``tcp_flags_value``: TCP control bits. The non-zero value is invalid, when the protocol_value is not set to 0x06 (TCP).
2595
2596* ``prio_value``: The priority of this filter.
2597
2598* ``queue_id``: The receive queue associated with this 5-tuple filter.
2599
2600Example, to add/remove an 5tuple filter rule::
2601
2602   testpmd> 5tuple_filter 0 add dst_ip 2.2.2.5 src_ip 2.2.2.4 \
2603            dst_port 64 src_port 32 protocol 0x06 mask 0x1F \
2604            flags 0x0 priority 3 queue 3
2605
2606   testpmd> 5tuple_filter 0 del dst_ip 2.2.2.5 src_ip 2.2.2.4 \
2607            dst_port 64 src_port 32 protocol 0x06 mask 0x1F \
2608            flags 0x0 priority 3 queue 3
2609
2610syn_filter
2611~~~~~~~~~~
2612
2613Using the  SYN filter, TCP packets whose *SYN* flag is set can be forwarded to a separate queue::
2614
2615   syn_filter (port_id) (add|del) priority (high|low) queue (queue_id)
2616
2617The available information parameters are:
2618
2619* ``port_id``: The port which the SYN filter assigned on.
2620
2621* ``high``: This SYN filter has higher priority than other filters.
2622
2623* ``low``: This SYN filter has lower priority than other filters.
2624
2625* ``queue_id``: The receive queue associated with this SYN filter
2626
2627Example::
2628
2629   testpmd> syn_filter 0 add priority high queue 3
2630
2631flex_filter
2632~~~~~~~~~~~
2633
2634With flex filter, packets can be recognized by any arbitrary pattern within the first 128 bytes of the packet
2635and routed into one of the receive queues::
2636
2637   flex_filter (port_id) (add|del) len (len_value) bytes (bytes_value) \
2638               mask (mask_value) priority (prio_value) queue (queue_id)
2639
2640The available information parameters are:
2641
2642* ``port_id``: The port which the Flex filter is assigned on.
2643
2644* ``len_value``: Filter length in bytes, no greater than 128.
2645
2646* ``bytes_value``: A string in hexadecimal, means the value the flex filter needs to match.
2647
2648* ``mask_value``: A string in hexadecimal, bit 1 means corresponding byte participates in the match.
2649
2650* ``prio_value``: The priority of this filter.
2651
2652* ``queue_id``: The receive queue associated with this Flex filter.
2653
2654Example::
2655
2656   testpmd> flex_filter 0 add len 16 bytes 0x00000000000000000000000008060000 \
2657                          mask 000C priority 3 queue 3
2658
2659   testpmd> flex_filter 0 del len 16 bytes 0x00000000000000000000000008060000 \
2660                          mask 000C priority 3 queue 3
2661
2662
2663.. _testpmd_flow_director:
2664
2665flow_director_filter
2666~~~~~~~~~~~~~~~~~~~~
2667
2668The Flow Director works in receive mode to identify specific flows or sets of flows and route them to specific queues.
2669
2670Four types of filtering are supported which are referred to as Perfect Match, Signature, Perfect-mac-vlan and
2671Perfect-tunnel filters, the match mode is set by the ``--pkt-filter-mode`` command-line parameter:
2672
2673* Perfect match filters.
2674  The hardware checks a match between the masked fields of the received packets and the programmed filters.
2675  The masked fields are for IP flow.
2676
2677* Signature filters.
2678  The hardware checks a match between a hash-based signature of the masked fields of the received packet.
2679
2680* Perfect-mac-vlan match filters.
2681  The hardware checks a match between the masked fields of the received packets and the programmed filters.
2682  The masked fields are for MAC VLAN flow.
2683
2684* Perfect-tunnel match filters.
2685  The hardware checks a match between the masked fields of the received packets and the programmed filters.
2686  The masked fields are for tunnel flow.
2687
2688The Flow Director filters can match the different fields for different type of packet: flow type, specific input set
2689per flow type and the flexible payload.
2690
2691The Flow Director can also mask out parts of all of these fields so that filters
2692are only applied to certain fields or parts of the fields.
2693
2694Different NICs may have different capabilities, command show port fdir (port_id) can be used to acquire the information.
2695
2696# Commands to add flow director filters of different flow types::
2697
2698   flow_director_filter (port_id) mode IP (add|del|update) \
2699                        flow (ipv4-other|ipv4-frag|ipv6-other|ipv6-frag) \
2700                        src (src_ip_address) dst (dst_ip_address) \
2701                        tos (tos_value) proto (proto_value) ttl (ttl_value) \
2702                        vlan (vlan_value) flexbytes (flexbytes_value) \
2703                        (drop|fwd) pf|vf(vf_id) queue (queue_id) \
2704                        fd_id (fd_id_value)
2705
2706   flow_director_filter (port_id) mode IP (add|del|update) \
2707                        flow (ipv4-tcp|ipv4-udp|ipv6-tcp|ipv6-udp) \
2708                        src (src_ip_address) (src_port) \
2709                        dst (dst_ip_address) (dst_port) \
2710                        tos (tos_value) ttl (ttl_value) \
2711                        vlan (vlan_value) flexbytes (flexbytes_value) \
2712                        (drop|fwd) queue pf|vf(vf_id) (queue_id) \
2713                        fd_id (fd_id_value)
2714
2715   flow_director_filter (port_id) mode IP (add|del|update) \
2716                        flow (ipv4-sctp|ipv6-sctp) \
2717                        src (src_ip_address) (src_port) \
2718                        dst (dst_ip_address) (dst_port) \
2719                        tos (tos_value) ttl (ttl_value) \
2720                        tag (verification_tag) vlan (vlan_value) \
2721                        flexbytes (flexbytes_value) (drop|fwd) \
2722                        pf|vf(vf_id) queue (queue_id) fd_id (fd_id_value)
2723
2724   flow_director_filter (port_id) mode IP (add|del|update) flow l2_payload \
2725                        ether (ethertype) flexbytes (flexbytes_value) \
2726                        (drop|fwd) pf|vf(vf_id) queue (queue_id)
2727                        fd_id (fd_id_value)
2728
2729   flow_director_filter (port_id) mode MAC-VLAN (add|del|update) \
2730                        mac (mac_address) vlan (vlan_value) \
2731                        flexbytes (flexbytes_value) (drop|fwd) \
2732                        queue (queue_id) fd_id (fd_id_value)
2733
2734   flow_director_filter (port_id) mode Tunnel (add|del|update) \
2735                        mac (mac_address) vlan (vlan_value) \
2736                        tunnel (NVGRE|VxLAN) tunnel-id (tunnel_id_value) \
2737                        flexbytes (flexbytes_value) (drop|fwd) \
2738                        queue (queue_id) fd_id (fd_id_value)
2739
2740For example, to add an ipv4-udp flow type filter::
2741
2742   testpmd> flow_director_filter 0 mode IP add flow ipv4-udp src 2.2.2.3 32 \
2743            dst 2.2.2.5 33 tos 2 ttl 40 vlan 0x1 flexbytes (0x88,0x48) \
2744            fwd pf queue 1 fd_id 1
2745
2746For example, add an ipv4-other flow type filter::
2747
2748   testpmd> flow_director_filter 0 mode IP add flow ipv4-other src 2.2.2.3 \
2749             dst 2.2.2.5 tos 2 proto 20 ttl 40 vlan 0x1 \
2750             flexbytes (0x88,0x48) fwd pf queue 1 fd_id 1
2751
2752flush_flow_director
2753~~~~~~~~~~~~~~~~~~~
2754
2755Flush all flow director filters on a device::
2756
2757   testpmd> flush_flow_director (port_id)
2758
2759Example, to flush all flow director filter on port 0::
2760
2761   testpmd> flush_flow_director 0
2762
2763flow_director_mask
2764~~~~~~~~~~~~~~~~~~
2765
2766Set flow director's input masks::
2767
2768   flow_director_mask (port_id) mode IP vlan (vlan_value) \
2769                      src_mask (ipv4_src) (ipv6_src) (src_port) \
2770                      dst_mask (ipv4_dst) (ipv6_dst) (dst_port)
2771
2772   flow_director_mask (port_id) mode MAC-VLAN vlan (vlan_value)
2773
2774   flow_director_mask (port_id) mode Tunnel vlan (vlan_value) \
2775                      mac (mac_value) tunnel-type (tunnel_type_value) \
2776                      tunnel-id (tunnel_id_value)
2777
2778Example, to set flow director mask on port 0::
2779
2780   testpmd> flow_director_mask 0 mode IP vlan 0xefff \
2781            src_mask 255.255.255.255 \
2782                FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF \
2783            dst_mask 255.255.255.255 \
2784                FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF
2785
2786flow_director_flex_mask
2787~~~~~~~~~~~~~~~~~~~~~~~
2788
2789set masks of flow director's flexible payload based on certain flow type::
2790
2791   testpmd> flow_director_flex_mask (port_id) \
2792            flow (none|ipv4-other|ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
2793                  ipv6-other|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp| \
2794                  l2_payload|all) (mask)
2795
2796Example, to set flow director's flex mask for all flow type on port 0::
2797
2798   testpmd> flow_director_flex_mask 0 flow all \
2799            (0xff,0xff,0,0,0,0,0,0,0,0,0,0,0,0,0,0)
2800
2801
2802flow_director_flex_payload
2803~~~~~~~~~~~~~~~~~~~~~~~~~~
2804
2805Configure flexible payload selection::
2806
2807   flow_director_flex_payload (port_id) (raw|l2|l3|l4) (config)
2808
2809For example, to select the first 16 bytes from the offset 4 (bytes) of packet's payload as flexible payload::
2810
2811   testpmd> flow_director_flex_payload 0 l4 \
2812            (4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19)
2813
2814get_sym_hash_ena_per_port
2815~~~~~~~~~~~~~~~~~~~~~~~~~
2816
2817Get symmetric hash enable configuration per port::
2818
2819   get_sym_hash_ena_per_port (port_id)
2820
2821For example, to get symmetric hash enable configuration of port 1::
2822
2823   testpmd> get_sym_hash_ena_per_port 1
2824
2825set_sym_hash_ena_per_port
2826~~~~~~~~~~~~~~~~~~~~~~~~~
2827
2828Set symmetric hash enable configuration per port to enable or disable::
2829
2830   set_sym_hash_ena_per_port (port_id) (enable|disable)
2831
2832For example, to set symmetric hash enable configuration of port 1 to enable::
2833
2834   testpmd> set_sym_hash_ena_per_port 1 enable
2835
2836get_hash_global_config
2837~~~~~~~~~~~~~~~~~~~~~~
2838
2839Get the global configurations of hash filters::
2840
2841   get_hash_global_config (port_id)
2842
2843For example, to get the global configurations of hash filters of port 1::
2844
2845   testpmd> get_hash_global_config 1
2846
2847set_hash_global_config
2848~~~~~~~~~~~~~~~~~~~~~~
2849
2850Set the global configurations of hash filters::
2851
2852   set_hash_global_config (port_id) (toeplitz|simple_xor|default) \
2853   (ipv4|ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp|ipv4-other|ipv6|ipv6-frag| \
2854   ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other|l2_payload) \
2855   (enable|disable)
2856
2857For example, to enable simple_xor for flow type of ipv6 on port 2::
2858
2859   testpmd> set_hash_global_config 2 simple_xor ipv6 enable
2860
2861set_hash_input_set
2862~~~~~~~~~~~~~~~~~~
2863
2864Set the input set for hash::
2865
2866   set_hash_input_set (port_id) (ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
2867   ipv4-other|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other| \
2868   l2_payload) (ovlan|ivlan|src-ipv4|dst-ipv4|src-ipv6|dst-ipv6|ipv4-tos| \
2869   ipv4-proto|ipv6-tc|ipv6-next-header|udp-src-port|udp-dst-port| \
2870   tcp-src-port|tcp-dst-port|sctp-src-port|sctp-dst-port|sctp-veri-tag| \
2871   udp-key|gre-key|fld-1st|fld-2nd|fld-3rd|fld-4th|fld-5th|fld-6th|fld-7th| \
2872   fld-8th|none) (select|add)
2873
2874For example, to add source IP to hash input set for flow type of ipv4-udp on port 0::
2875
2876   testpmd> set_hash_input_set 0 ipv4-udp src-ipv4 add
2877
2878set_fdir_input_set
2879~~~~~~~~~~~~~~~~~~
2880
2881The Flow Director filters can match the different fields for different type of packet, i.e. specific input set
2882on per flow type and the flexible payload. This command can be used to change input set for each flow type.
2883
2884Set the input set for flow director::
2885
2886   set_fdir_input_set (port_id) (ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
2887   ipv4-other|ipv6|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other| \
2888   l2_payload) (ivlan|ethertype|src-ipv4|dst-ipv4|src-ipv6|dst-ipv6|ipv4-tos| \
2889   ipv4-proto|ipv4-ttl|ipv6-tc|ipv6-next-header|ipv6-hop-limits| \
2890   tudp-src-port|udp-dst-port|cp-src-port|tcp-dst-port|sctp-src-port| \
2891   sctp-dst-port|sctp-veri-tag|none) (select|add)
2892
2893For example to add source IP to FD input set for flow type of ipv4-udp on port 0::
2894
2895   testpmd> set_fdir_input_set 0 ipv4-udp src-ipv4 add
2896
2897global_config
2898~~~~~~~~~~~~~
2899
2900Set different GRE key length for input set::
2901
2902   global_config (port_id) gre-key-len (number in bytes)
2903
2904For example to set GRE key length for input set to 4 bytes on port 0::
2905
2906   testpmd> global_config 0 gre-key-len 4
2907
2908
2909.. _testpmd_rte_flow:
2910
2911Flow rules management
2912---------------------
2913
2914Control of the generic flow API (*rte_flow*) is fully exposed through the
2915``flow`` command (validation, creation, destruction, queries and operation
2916modes).
2917
2918Considering *rte_flow* overlaps with all `Filter Functions`_, using both
2919features simultaneously may cause undefined side-effects and is therefore
2920not recommended.
2921
2922``flow`` syntax
2923~~~~~~~~~~~~~~~
2924
2925Because the ``flow`` command uses dynamic tokens to handle the large number
2926of possible flow rules combinations, its behavior differs slightly from
2927other commands, in particular:
2928
2929- Pressing *?* or the *<tab>* key displays contextual help for the current
2930  token, not that of the entire command.
2931
2932- Optional and repeated parameters are supported (provided they are listed
2933  in the contextual help).
2934
2935The first parameter stands for the operation mode. Possible operations and
2936their general syntax are described below. They are covered in detail in the
2937following sections.
2938
2939- Check whether a flow rule can be created::
2940
2941   flow validate {port_id}
2942       [group {group_id}] [priority {level}] [ingress] [egress]
2943       pattern {item} [/ {item} [...]] / end
2944       actions {action} [/ {action} [...]] / end
2945
2946- Create a flow rule::
2947
2948   flow create {port_id}
2949       [group {group_id}] [priority {level}] [ingress] [egress]
2950       pattern {item} [/ {item} [...]] / end
2951       actions {action} [/ {action} [...]] / end
2952
2953- Destroy specific flow rules::
2954
2955   flow destroy {port_id} rule {rule_id} [...]
2956
2957- Destroy all flow rules::
2958
2959   flow flush {port_id}
2960
2961- Query an existing flow rule::
2962
2963   flow query {port_id} {rule_id} {action}
2964
2965- List existing flow rules sorted by priority, filtered by group
2966  identifiers::
2967
2968   flow list {port_id} [group {group_id}] [...]
2969
2970- Restrict ingress traffic to the defined flow rules::
2971
2972   flow isolate {port_id} {boolean}
2973
2974Validating flow rules
2975~~~~~~~~~~~~~~~~~~~~~
2976
2977``flow validate`` reports whether a flow rule would be accepted by the
2978underlying device in its current state but stops short of creating it. It is
2979bound to ``rte_flow_validate()``::
2980
2981   flow validate {port_id}
2982      [group {group_id}] [priority {level}] [ingress] [egress]
2983      pattern {item} [/ {item} [...]] / end
2984      actions {action} [/ {action} [...]] / end
2985
2986If successful, it will show::
2987
2988   Flow rule validated
2989
2990Otherwise it will show an error message of the form::
2991
2992   Caught error type [...] ([...]): [...]
2993
2994This command uses the same parameters as ``flow create``, their format is
2995described in `Creating flow rules`_.
2996
2997Check whether redirecting any Ethernet packet received on port 0 to RX queue
2998index 6 is supported::
2999
3000   testpmd> flow validate 0 ingress pattern eth / end
3001      actions queue index 6 / end
3002   Flow rule validated
3003   testpmd>
3004
3005Port 0 does not support TCPv6 rules::
3006
3007   testpmd> flow validate 0 ingress pattern eth / ipv6 / tcp / end
3008      actions drop / end
3009   Caught error type 9 (specific pattern item): Invalid argument
3010   testpmd>
3011
3012Creating flow rules
3013~~~~~~~~~~~~~~~~~~~
3014
3015``flow create`` validates and creates the specified flow rule. It is bound
3016to ``rte_flow_create()``::
3017
3018   flow create {port_id}
3019      [group {group_id}] [priority {level}] [ingress] [egress]
3020      pattern {item} [/ {item} [...]] / end
3021      actions {action} [/ {action} [...]] / end
3022
3023If successful, it will return a flow rule ID usable with other commands::
3024
3025   Flow rule #[...] created
3026
3027Otherwise it will show an error message of the form::
3028
3029   Caught error type [...] ([...]): [...]
3030
3031Parameters describe in the following order:
3032
3033- Attributes (*group*, *priority*, *ingress*, *egress* tokens).
3034- A matching pattern, starting with the *pattern* token and terminated by an
3035  *end* pattern item.
3036- Actions, starting with the *actions* token and terminated by an *end*
3037  action.
3038
3039These translate directly to *rte_flow* objects provided as-is to the
3040underlying functions.
3041
3042The shortest valid definition only comprises mandatory tokens::
3043
3044   testpmd> flow create 0 pattern end actions end
3045
3046Note that PMDs may refuse rules that essentially do nothing such as this
3047one.
3048
3049**All unspecified object values are automatically initialized to 0.**
3050
3051Attributes
3052^^^^^^^^^^
3053
3054These tokens affect flow rule attributes (``struct rte_flow_attr``) and are
3055specified before the ``pattern`` token.
3056
3057- ``group {group id}``: priority group.
3058- ``priority {level}``: priority level within group.
3059- ``ingress``: rule applies to ingress traffic.
3060- ``egress``: rule applies to egress traffic.
3061
3062Each instance of an attribute specified several times overrides the previous
3063value as shown below (group 4 is used)::
3064
3065   testpmd> flow create 0 group 42 group 24 group 4 [...]
3066
3067Note that once enabled, ``ingress`` and ``egress`` cannot be disabled.
3068
3069While not specifying a direction is an error, some rules may allow both
3070simultaneously.
3071
3072Most rules affect RX therefore contain the ``ingress`` token::
3073
3074   testpmd> flow create 0 ingress pattern [...]
3075
3076Matching pattern
3077^^^^^^^^^^^^^^^^
3078
3079A matching pattern starts after the ``pattern`` token. It is made of pattern
3080items and is terminated by a mandatory ``end`` item.
3081
3082Items are named after their type (*RTE_FLOW_ITEM_TYPE_* from ``enum
3083rte_flow_item_type``).
3084
3085The ``/`` token is used as a separator between pattern items as shown
3086below::
3087
3088   testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end [...]
3089
3090Note that protocol items like these must be stacked from lowest to highest
3091layer to make sense. For instance, the following rule is either invalid or
3092unlikely to match any packet::
3093
3094   testpmd> flow create 0 ingress pattern eth / udp / ipv4 / end [...]
3095
3096More information on these restrictions can be found in the *rte_flow*
3097documentation.
3098
3099Several items support additional specification structures, for example
3100``ipv4`` allows specifying source and destination addresses as follows::
3101
3102   testpmd> flow create 0 ingress pattern eth / ipv4 src is 10.1.1.1
3103      dst is 10.2.0.0 / end [...]
3104
3105This rule matches all IPv4 traffic with the specified properties.
3106
3107In this example, ``src`` and ``dst`` are field names of the underlying
3108``struct rte_flow_item_ipv4`` object. All item properties can be specified
3109in a similar fashion.
3110
3111The ``is`` token means that the subsequent value must be matched exactly,
3112and assigns ``spec`` and ``mask`` fields in ``struct rte_flow_item``
3113accordingly. Possible assignment tokens are:
3114
3115- ``is``: match value perfectly (with full bit-mask).
3116- ``spec``: match value according to configured bit-mask.
3117- ``last``: specify upper bound to establish a range.
3118- ``mask``: specify bit-mask with relevant bits set to one.
3119- ``prefix``: generate bit-mask from a prefix length.
3120
3121These yield identical results::
3122
3123   ipv4 src is 10.1.1.1
3124
3125::
3126
3127   ipv4 src spec 10.1.1.1 src mask 255.255.255.255
3128
3129::
3130
3131   ipv4 src spec 10.1.1.1 src prefix 32
3132
3133::
3134
3135   ipv4 src is 10.1.1.1 src last 10.1.1.1 # range with a single value
3136
3137::
3138
3139   ipv4 src is 10.1.1.1 src last 0 # 0 disables range
3140
3141Inclusive ranges can be defined with ``last``::
3142
3143   ipv4 src is 10.1.1.1 src last 10.2.3.4 # 10.1.1.1 to 10.2.3.4
3144
3145Note that ``mask`` affects both ``spec`` and ``last``::
3146
3147   ipv4 src is 10.1.1.1 src last 10.2.3.4 src mask 255.255.0.0
3148      # matches 10.1.0.0 to 10.2.255.255
3149
3150Properties can be modified multiple times::
3151
3152   ipv4 src is 10.1.1.1 src is 10.1.2.3 src is 10.2.3.4 # matches 10.2.3.4
3153
3154::
3155
3156   ipv4 src is 10.1.1.1 src prefix 24 src prefix 16 # matches 10.1.0.0/16
3157
3158Pattern items
3159^^^^^^^^^^^^^
3160
3161This section lists supported pattern items and their attributes, if any.
3162
3163- ``end``: end list of pattern items.
3164
3165- ``void``: no-op pattern item.
3166
3167- ``invert``: perform actions when pattern does not match.
3168
3169- ``any``: match any protocol for the current layer.
3170
3171  - ``num {unsigned}``: number of layers covered.
3172
3173- ``pf``: match packets addressed to the physical function.
3174
3175- ``vf``: match packets addressed to a virtual function ID.
3176
3177  - ``id {unsigned}``: destination VF ID.
3178
3179- ``port``: device-specific physical port index to use.
3180
3181  - ``index {unsigned}``: physical port index.
3182
3183- ``raw``: match an arbitrary byte string.
3184
3185  - ``relative {boolean}``: look for pattern after the previous item.
3186  - ``search {boolean}``: search pattern from offset (see also limit).
3187  - ``offset {integer}``: absolute or relative offset for pattern.
3188  - ``limit {unsigned}``: search area limit for start of pattern.
3189  - ``pattern {string}``: byte string to look for.
3190
3191- ``eth``: match Ethernet header.
3192
3193  - ``dst {MAC-48}``: destination MAC.
3194  - ``src {MAC-48}``: source MAC.
3195  - ``type {unsigned}``: EtherType.
3196
3197- ``vlan``: match 802.1Q/ad VLAN tag.
3198
3199  - ``tpid {unsigned}``: tag protocol identifier.
3200  - ``tci {unsigned}``: tag control information.
3201  - ``pcp {unsigned}``: priority code point.
3202  - ``dei {unsigned}``: drop eligible indicator.
3203  - ``vid {unsigned}``: VLAN identifier.
3204
3205- ``ipv4``: match IPv4 header.
3206
3207  - ``tos {unsigned}``: type of service.
3208  - ``ttl {unsigned}``: time to live.
3209  - ``proto {unsigned}``: next protocol ID.
3210  - ``src {ipv4 address}``: source address.
3211  - ``dst {ipv4 address}``: destination address.
3212
3213- ``ipv6``: match IPv6 header.
3214
3215  - ``tc {unsigned}``: traffic class.
3216  - ``flow {unsigned}``: flow label.
3217  - ``proto {unsigned}``: protocol (next header).
3218  - ``hop {unsigned}``: hop limit.
3219  - ``src {ipv6 address}``: source address.
3220  - ``dst {ipv6 address}``: destination address.
3221
3222- ``icmp``: match ICMP header.
3223
3224  - ``type {unsigned}``: ICMP packet type.
3225  - ``code {unsigned}``: ICMP packet code.
3226
3227- ``udp``: match UDP header.
3228
3229  - ``src {unsigned}``: UDP source port.
3230  - ``dst {unsigned}``: UDP destination port.
3231
3232- ``tcp``: match TCP header.
3233
3234  - ``src {unsigned}``: TCP source port.
3235  - ``dst {unsigned}``: TCP destination port.
3236
3237- ``sctp``: match SCTP header.
3238
3239  - ``src {unsigned}``: SCTP source port.
3240  - ``dst {unsigned}``: SCTP destination port.
3241  - ``tag {unsigned}``: validation tag.
3242  - ``cksum {unsigned}``: checksum.
3243
3244- ``vxlan``: match VXLAN header.
3245
3246  - ``vni {unsigned}``: VXLAN identifier.
3247
3248- ``e_tag``: match IEEE 802.1BR E-Tag header.
3249
3250  - ``grp_ecid_b {unsigned}``: GRP and E-CID base.
3251
3252- ``nvgre``: match NVGRE header.
3253
3254  - ``tni {unsigned}``: virtual subnet ID.
3255
3256- ``mpls``: match MPLS header.
3257
3258  - ``label {unsigned}``: MPLS label.
3259
3260- ``gre``: match GRE header.
3261
3262  - ``protocol {unsigned}``: protocol type.
3263
3264- ``fuzzy``: fuzzy pattern match, expect faster than default.
3265
3266  - ``thresh {unsigned}``: accuracy threshold.
3267
3268- ``gtp``, ``gtpc``, ``gtpu``: match GTPv1 header.
3269
3270  - ``teid {unsigned}``: tunnel endpoint identifier.
3271
3272Actions list
3273^^^^^^^^^^^^
3274
3275A list of actions starts after the ``actions`` token in the same fashion as
3276`Matching pattern`_; actions are separated by ``/`` tokens and the list is
3277terminated by a mandatory ``end`` action.
3278
3279Actions are named after their type (*RTE_FLOW_ACTION_TYPE_* from ``enum
3280rte_flow_action_type``).
3281
3282Dropping all incoming UDPv4 packets can be expressed as follows::
3283
3284   testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
3285      actions drop / end
3286
3287Several actions have configurable properties which must be specified when
3288there is no valid default value. For example, ``queue`` requires a target
3289queue index.
3290
3291This rule redirects incoming UDPv4 traffic to queue index 6::
3292
3293   testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
3294      actions queue index 6 / end
3295
3296While this one could be rejected by PMDs (unspecified queue index)::
3297
3298   testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
3299      actions queue / end
3300
3301As defined by *rte_flow*, the list is not ordered, all actions of a given
3302rule are performed simultaneously. These are equivalent::
3303
3304   queue index 6 / void / mark id 42 / end
3305
3306::
3307
3308   void / mark id 42 / queue index 6 / end
3309
3310All actions in a list should have different types, otherwise only the last
3311action of a given type is taken into account::
3312
3313   queue index 4 / queue index 5 / queue index 6 / end # will use queue 6
3314
3315::
3316
3317   drop / drop / drop / end # drop is performed only once
3318
3319::
3320
3321   mark id 42 / queue index 3 / mark id 24 / end # mark will be 24
3322
3323Considering they are performed simultaneously, opposite and overlapping
3324actions can sometimes be combined when the end result is unambiguous::
3325
3326   drop / queue index 6 / end # drop has no effect
3327
3328::
3329
3330   drop / dup index 6 / end # same as above
3331
3332::
3333
3334   queue index 6 / rss queues 6 7 8 / end # queue has no effect
3335
3336::
3337
3338   drop / passthru / end # drop has no effect
3339
3340Note that PMDs may still refuse such combinations.
3341
3342Actions
3343^^^^^^^
3344
3345This section lists supported actions and their attributes, if any.
3346
3347- ``end``: end list of actions.
3348
3349- ``void``: no-op action.
3350
3351- ``passthru``: let subsequent rule process matched packets.
3352
3353- ``mark``: attach 32 bit value to packets.
3354
3355  - ``id {unsigned}``: 32 bit value to return with packets.
3356
3357- ``flag``: flag packets.
3358
3359- ``queue``: assign packets to a given queue index.
3360
3361  - ``index {unsigned}``: queue index to use.
3362
3363- ``drop``: drop packets (note: passthru has priority).
3364
3365- ``count``: enable counters for this rule.
3366
3367- ``dup``: duplicate packets to a given queue index.
3368
3369  - ``index {unsigned}``: queue index to duplicate packets to.
3370
3371- ``rss``: spread packets among several queues.
3372
3373  - ``queues [{unsigned} [...]] end``: queue indices to use.
3374
3375- ``pf``: redirect packets to physical device function.
3376
3377- ``vf``: redirect packets to virtual device function.
3378
3379  - ``original {boolean}``: use original VF ID if possible.
3380  - ``id {unsigned}``: VF ID to redirect packets to.
3381
3382Destroying flow rules
3383~~~~~~~~~~~~~~~~~~~~~
3384
3385``flow destroy`` destroys one or more rules from their rule ID (as returned
3386by ``flow create``), this command calls ``rte_flow_destroy()`` as many
3387times as necessary::
3388
3389   flow destroy {port_id} rule {rule_id} [...]
3390
3391If successful, it will show::
3392
3393   Flow rule #[...] destroyed
3394
3395It does not report anything for rule IDs that do not exist. The usual error
3396message is shown when a rule cannot be destroyed::
3397
3398   Caught error type [...] ([...]): [...]
3399
3400``flow flush`` destroys all rules on a device and does not take extra
3401arguments. It is bound to ``rte_flow_flush()``::
3402
3403   flow flush {port_id}
3404
3405Any errors are reported as above.
3406
3407Creating several rules and destroying them::
3408
3409   testpmd> flow create 0 ingress pattern eth / ipv6 / end
3410      actions queue index 2 / end
3411   Flow rule #0 created
3412   testpmd> flow create 0 ingress pattern eth / ipv4 / end
3413      actions queue index 3 / end
3414   Flow rule #1 created
3415   testpmd> flow destroy 0 rule 0 rule 1
3416   Flow rule #1 destroyed
3417   Flow rule #0 destroyed
3418   testpmd>
3419
3420The same result can be achieved using ``flow flush``::
3421
3422   testpmd> flow create 0 ingress pattern eth / ipv6 / end
3423      actions queue index 2 / end
3424   Flow rule #0 created
3425   testpmd> flow create 0 ingress pattern eth / ipv4 / end
3426      actions queue index 3 / end
3427   Flow rule #1 created
3428   testpmd> flow flush 0
3429   testpmd>
3430
3431Non-existent rule IDs are ignored::
3432
3433   testpmd> flow create 0 ingress pattern eth / ipv6 / end
3434      actions queue index 2 / end
3435   Flow rule #0 created
3436   testpmd> flow create 0 ingress pattern eth / ipv4 / end
3437      actions queue index 3 / end
3438   Flow rule #1 created
3439   testpmd> flow destroy 0 rule 42 rule 10 rule 2
3440   testpmd>
3441   testpmd> flow destroy 0 rule 0
3442   Flow rule #0 destroyed
3443   testpmd>
3444
3445Querying flow rules
3446~~~~~~~~~~~~~~~~~~~
3447
3448``flow query`` queries a specific action of a flow rule having that
3449ability. Such actions collect information that can be reported using this
3450command. It is bound to ``rte_flow_query()``::
3451
3452   flow query {port_id} {rule_id} {action}
3453
3454If successful, it will display either the retrieved data for known actions
3455or the following message::
3456
3457   Cannot display result for action type [...] ([...])
3458
3459Otherwise, it will complain either that the rule does not exist or that some
3460error occurred::
3461
3462   Flow rule #[...] not found
3463
3464::
3465
3466   Caught error type [...] ([...]): [...]
3467
3468Currently only the ``count`` action is supported. This action reports the
3469number of packets that hit the flow rule and the total number of bytes. Its
3470output has the following format::
3471
3472   count:
3473    hits_set: [...] # whether "hits" contains a valid value
3474    bytes_set: [...] # whether "bytes" contains a valid value
3475    hits: [...] # number of packets
3476    bytes: [...] # number of bytes
3477
3478Querying counters for TCPv6 packets redirected to queue 6::
3479
3480   testpmd> flow create 0 ingress pattern eth / ipv6 / tcp / end
3481      actions queue index 6 / count / end
3482   Flow rule #4 created
3483   testpmd> flow query 0 4 count
3484   count:
3485    hits_set: 1
3486    bytes_set: 0
3487    hits: 386446
3488    bytes: 0
3489   testpmd>
3490
3491Listing flow rules
3492~~~~~~~~~~~~~~~~~~
3493
3494``flow list`` lists existing flow rules sorted by priority and optionally
3495filtered by group identifiers::
3496
3497   flow list {port_id} [group {group_id}] [...]
3498
3499This command only fails with the following message if the device does not
3500exist::
3501
3502   Invalid port [...]
3503
3504Output consists of a header line followed by a short description of each
3505flow rule, one per line. There is no output at all when no flow rules are
3506configured on the device::
3507
3508   ID      Group   Prio    Attr    Rule
3509   [...]   [...]   [...]   [...]   [...]
3510
3511``Attr`` column flags:
3512
3513- ``i`` for ``ingress``.
3514- ``e`` for ``egress``.
3515
3516Creating several flow rules and listing them::
3517
3518   testpmd> flow create 0 ingress pattern eth / ipv4 / end
3519      actions queue index 6 / end
3520   Flow rule #0 created
3521   testpmd> flow create 0 ingress pattern eth / ipv6 / end
3522      actions queue index 2 / end
3523   Flow rule #1 created
3524   testpmd> flow create 0 priority 5 ingress pattern eth / ipv4 / udp / end
3525      actions rss queues 6 7 8 end / end
3526   Flow rule #2 created
3527   testpmd> flow list 0
3528   ID      Group   Prio    Attr    Rule
3529   0       0       0       i-      ETH IPV4 => QUEUE
3530   1       0       0       i-      ETH IPV6 => QUEUE
3531   2       0       5       i-      ETH IPV4 UDP => RSS
3532   testpmd>
3533
3534Rules are sorted by priority (i.e. group ID first, then priority level)::
3535
3536   testpmd> flow list 1
3537   ID      Group   Prio    Attr    Rule
3538   0       0       0       i-      ETH => COUNT
3539   6       0       500     i-      ETH IPV6 TCP => DROP COUNT
3540   5       0       1000    i-      ETH IPV6 ICMP => QUEUE
3541   1       24      0       i-      ETH IPV4 UDP => QUEUE
3542   4       24      10      i-      ETH IPV4 TCP => DROP
3543   3       24      20      i-      ETH IPV4 => DROP
3544   2       24      42      i-      ETH IPV4 UDP => QUEUE
3545   7       63      0       i-      ETH IPV6 UDP VXLAN => MARK QUEUE
3546   testpmd>
3547
3548Output can be limited to specific groups::
3549
3550   testpmd> flow list 1 group 0 group 63
3551   ID      Group   Prio    Attr    Rule
3552   0       0       0       i-      ETH => COUNT
3553   6       0       500     i-      ETH IPV6 TCP => DROP COUNT
3554   5       0       1000    i-      ETH IPV6 ICMP => QUEUE
3555   7       63      0       i-      ETH IPV6 UDP VXLAN => MARK QUEUE
3556   testpmd>
3557
3558Toggling isolated mode
3559~~~~~~~~~~~~~~~~~~~~~~
3560
3561``flow isolate`` can be used to tell the underlying PMD that ingress traffic
3562must only be injected from the defined flow rules; that no default traffic
3563is expected outside those rules and the driver is free to assign more
3564resources to handle them. It is bound to ``rte_flow_isolate()``::
3565
3566 flow isolate {port_id} {boolean}
3567
3568If successful, enabling or disabling isolated mode shows either::
3569
3570 Ingress traffic on port [...]
3571    is now restricted to the defined flow rules
3572
3573Or::
3574
3575 Ingress traffic on port [...]
3576    is not restricted anymore to the defined flow rules
3577
3578Otherwise, in case of error::
3579
3580   Caught error type [...] ([...]): [...]
3581
3582Mainly due to its side effects, PMDs supporting this mode may not have the
3583ability to toggle it more than once without reinitializing affected ports
3584first (e.g. by exiting testpmd).
3585
3586Enabling isolated mode::
3587
3588 testpmd> flow isolate 0 true
3589 Ingress traffic on port 0 is now restricted to the defined flow rules
3590 testpmd>
3591
3592Disabling isolated mode::
3593
3594 testpmd> flow isolate 0 false
3595 Ingress traffic on port 0 is not restricted anymore to the defined flow rules
3596 testpmd>
3597
3598Sample QinQ flow rules
3599~~~~~~~~~~~~~~~~~~~~~~
3600
3601Before creating QinQ rule(s) the following commands should be issued to enable QinQ::
3602
3603   testpmd> port stop 0
3604   testpmd> vlan set qinq on 0
3605
3606The above command sets the inner and outer TPID's to 0x8100.
3607
3608To change the TPID's the following commands should be used::
3609
3610   testpmd> vlan set outer tpid 0xa100 0
3611   testpmd> vlan set inner tpid 0x9100 0
3612   testpmd> port start 0
3613
3614Validate and create a QinQ rule on port 0 to steer traffic to a VF queue in a VM.
3615
3616::
3617
3618   testpmd> flow validate 0 ingress pattern eth / vlan tci is 123 /
3619       vlan tci is 456 / end actions vf id 1 / queue index 0 / end
3620   Flow rule #0 validated
3621
3622   testpmd> flow create 0 ingress pattern eth / vlan tci is 4 /
3623       vlan tci is 456 / end actions vf id 123 / queue index 0 / end
3624   Flow rule #0 created
3625
3626   testpmd> flow list 0
3627   ID      Group   Prio    Attr    Rule
3628   0       0       0       i-      ETH VLAN VLAN=>VF QUEUE
3629
3630Validate and create a QinQ rule on port 0 to steer traffic to a queue on the host.
3631
3632::
3633
3634   testpmd> flow validate 0 ingress pattern eth / vlan tci is 321 /
3635        vlan tci is 654 / end actions pf / queue index 0 / end
3636   Flow rule #1 validated
3637
3638   testpmd> flow create 0 ingress pattern eth / vlan tci is 321 /
3639        vlan tci is 654 / end actions pf / queue index 1 / end
3640   Flow rule #1 created
3641
3642   testpmd> flow list 0
3643   ID      Group   Prio    Attr    Rule
3644   0       0       0       i-      ETH VLAN VLAN=>VF QUEUE
3645   1       0       0       i-      ETH VLAN VLAN=>PF QUEUE
3646