xref: /dpdk/doc/guides/testpmd_app_ug/testpmd_funcs.rst (revision eb6d5a0af9a05bf940ba19ec1ddbe575b5e7540b)
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
357Note: TX timestamping is only available in the "Full Featured" TX path. To force ``testpmd`` into this mode set ``--txqflags=0``.
358
359Example::
360
361   testpmd> set fwd rxonly
362
363   Set rxonly packet forwarding mode
364
365
366read rxd
367~~~~~~~~
368
369Display an RX descriptor for a port RX queue::
370
371   testpmd> read rxd (port_id) (queue_id) (rxd_id)
372
373For example::
374
375   testpmd> read rxd 0 0 4
376        0x0000000B - 0x001D0180 / 0x0000000B - 0x001D0180
377
378read txd
379~~~~~~~~
380
381Display a TX descriptor for a port TX queue::
382
383   testpmd> read txd (port_id) (queue_id) (txd_id)
384
385For example::
386
387   testpmd> read txd 0 0 4
388        0x00000001 - 0x24C3C440 / 0x000F0000 - 0x2330003C
389
390ddp get list
391~~~~~~~~~~~~
392
393Get loaded dynamic device personalization (DDP) package info list::
394
395   testpmd> ddp get list (port_id)
396
397ddp get info
398~~~~~~~~~~~~
399
400Display information about dynamic device personalization (DDP) profile::
401
402   testpmd> ddp get info (profile_path)
403
404show vf stats
405~~~~~~~~~~~~~
406
407Display VF statistics::
408
409   testpmd> show vf stats (port_id) (vf_id)
410
411clear vf stats
412~~~~~~~~~~~~~~
413
414Reset VF statistics::
415
416   testpmd> clear vf stats (port_id) (vf_id)
417
418show port pctype mapping
419~~~~~~~~~~~~~~~~~~~~~~~~
420
421List all items from the pctype mapping table::
422
423   testpmd> show port (port_id) pctype mapping
424
425
426Configuration Functions
427-----------------------
428
429The testpmd application can be configured from the runtime as well as from the command-line.
430
431This section details the available configuration functions that are available.
432
433.. note::
434
435   Configuration changes only become active when forwarding is started/restarted.
436
437set default
438~~~~~~~~~~~
439
440Reset forwarding to the default configuration::
441
442   testpmd> set default
443
444set verbose
445~~~~~~~~~~~
446
447Set the debug verbosity level::
448
449   testpmd> set verbose (level)
450
451Currently the only available levels are 0 (silent except for error) and 1 (fully verbose).
452
453set nbport
454~~~~~~~~~~
455
456Set the number of ports used by the application:
457
458set nbport (num)
459
460This is equivalent to the ``--nb-ports`` command-line option.
461
462set nbcore
463~~~~~~~~~~
464
465Set the number of cores used by the application::
466
467   testpmd> set nbcore (num)
468
469This is equivalent to the ``--nb-cores`` command-line option.
470
471.. note::
472
473   The number of cores used must not be greater than number of ports used multiplied by the number of queues per port.
474
475set coremask
476~~~~~~~~~~~~
477
478Set the forwarding cores hexadecimal mask::
479
480   testpmd> set coremask (mask)
481
482This is equivalent to the ``--coremask`` command-line option.
483
484.. note::
485
486   The master lcore is reserved for command line parsing only and cannot be masked on for packet forwarding.
487
488set portmask
489~~~~~~~~~~~~
490
491Set the forwarding ports hexadecimal mask::
492
493   testpmd> set portmask (mask)
494
495This is equivalent to the ``--portmask`` command-line option.
496
497set burst
498~~~~~~~~~
499
500Set number of packets per burst::
501
502   testpmd> set burst (num)
503
504This is equivalent to the ``--burst command-line`` option.
505
506When retry is enabled, the transmit delay time and number of retries can also be set::
507
508   testpmd> set burst tx delay (microseconds) retry (num)
509
510set txpkts
511~~~~~~~~~~
512
513Set the length of each segment of the TX-ONLY packets or length of packet for FLOWGEN mode::
514
515   testpmd> set txpkts (x[,y]*)
516
517Where x[,y]* represents a CSV list of values, without white space.
518
519set txsplit
520~~~~~~~~~~~
521
522Set the split policy for the TX packets, applicable for TX-ONLY and CSUM forwarding modes::
523
524   testpmd> set txsplit (off|on|rand)
525
526Where:
527
528* ``off`` disable packet copy & split for CSUM mode.
529
530* ``on`` split outgoing packet into multiple segments. Size of each segment
531  and number of segments per packet is determined by ``set txpkts`` command
532  (see above).
533
534* ``rand`` same as 'on', but number of segments per each packet is a random value between 1 and total number of segments.
535
536set corelist
537~~~~~~~~~~~~
538
539Set the list of forwarding cores::
540
541   testpmd> set corelist (x[,y]*)
542
543For example, to change the forwarding cores:
544
545.. code-block:: console
546
547   testpmd> set corelist 3,1
548   testpmd> show config fwd
549
550   io packet forwarding - ports=2 - cores=2 - streams=2 - NUMA support disabled
551   Logical Core 3 (socket 0) forwards packets on 1 streams:
552   RX P=0/Q=0 (socket 0) -> TX P=1/Q=0 (socket 0) peer=02:00:00:00:00:01
553   Logical Core 1 (socket 0) forwards packets on 1 streams:
554   RX P=1/Q=0 (socket 0) -> TX P=0/Q=0 (socket 0) peer=02:00:00:00:00:00
555
556.. note::
557
558   The cores are used in the same order as specified on the command line.
559
560set portlist
561~~~~~~~~~~~~
562
563Set the list of forwarding ports::
564
565   testpmd> set portlist (x[,y]*)
566
567For example, to change the port forwarding:
568
569.. code-block:: console
570
571   testpmd> set portlist 0,2,1,3
572   testpmd> show config fwd
573
574   io packet forwarding - ports=4 - cores=1 - streams=4
575   Logical Core 3 (socket 0) forwards packets on 4 streams:
576   RX P=0/Q=0 (socket 0) -> TX P=2/Q=0 (socket 0) peer=02:00:00:00:00:01
577   RX P=2/Q=0 (socket 0) -> TX P=0/Q=0 (socket 0) peer=02:00:00:00:00:00
578   RX P=1/Q=0 (socket 0) -> TX P=3/Q=0 (socket 0) peer=02:00:00:00:00:03
579   RX P=3/Q=0 (socket 0) -> TX P=1/Q=0 (socket 0) peer=02:00:00:00:00:02
580
581set tx loopback
582~~~~~~~~~~~~~~~
583
584Enable/disable tx loopback::
585
586   testpmd> set tx loopback (port_id) (on|off)
587
588set drop enable
589~~~~~~~~~~~~~~~
590
591set drop enable bit for all queues::
592
593   testpmd> set all queues drop (port_id) (on|off)
594
595set split drop enable (for VF)
596~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
597
598set split drop enable bit for VF from PF::
599
600   testpmd> set vf split drop (port_id) (vf_id) (on|off)
601
602set mac antispoof (for VF)
603~~~~~~~~~~~~~~~~~~~~~~~~~~
604
605Set mac antispoof for a VF from the PF::
606
607   testpmd> set vf mac antispoof  (port_id) (vf_id) (on|off)
608
609set macsec offload
610~~~~~~~~~~~~~~~~~~
611
612Enable/disable MACsec offload::
613
614   testpmd> set macsec offload (port_id) on encrypt (on|off) replay-protect (on|off)
615   testpmd> set macsec offload (port_id) off
616
617set macsec sc
618~~~~~~~~~~~~~
619
620Configure MACsec secure connection (SC)::
621
622   testpmd> set macsec sc (tx|rx) (port_id) (mac) (pi)
623
624.. note::
625
626   The pi argument is ignored for tx.
627   Check the NIC Datasheet for hardware limits.
628
629set macsec sa
630~~~~~~~~~~~~~
631
632Configure MACsec secure association (SA)::
633
634   testpmd> set macsec sa (tx|rx) (port_id) (idx) (an) (pn) (key)
635
636.. note::
637
638   The IDX value must be 0 or 1.
639   Check the NIC Datasheet for hardware limits.
640
641set broadcast mode (for VF)
642~~~~~~~~~~~~~~~~~~~~~~~~~~~
643
644Set broadcast mode for a VF from the PF::
645
646   testpmd> set vf broadcast (port_id) (vf_id) (on|off)
647
648vlan set strip
649~~~~~~~~~~~~~~
650
651Set the VLAN strip on a port::
652
653   testpmd> vlan set strip (on|off) (port_id)
654
655vlan set stripq
656~~~~~~~~~~~~~~~
657
658Set the VLAN strip for a queue on a port::
659
660   testpmd> vlan set stripq (on|off) (port_id,queue_id)
661
662vlan set stripq (for VF)
663~~~~~~~~~~~~~~~~~~~~~~~~
664
665Set VLAN strip for all queues in a pool for a VF from the PF::
666
667   testpmd> set vf vlan stripq (port_id) (vf_id) (on|off)
668
669vlan set insert (for VF)
670~~~~~~~~~~~~~~~~~~~~~~~~
671
672Set VLAN insert for a VF from the PF::
673
674   testpmd> set vf vlan insert (port_id) (vf_id) (vlan_id)
675
676vlan set tag (for VF)
677~~~~~~~~~~~~~~~~~~~~~
678
679Set VLAN tag for a VF from the PF::
680
681   testpmd> set vf vlan tag (port_id) (vf_id) (on|off)
682
683vlan set antispoof (for VF)
684~~~~~~~~~~~~~~~~~~~~~~~~~~~
685
686Set VLAN antispoof for a VF from the PF::
687
688   testpmd> set vf vlan antispoof (port_id) (vf_id) (on|off)
689
690vlan set filter
691~~~~~~~~~~~~~~~
692
693Set the VLAN filter on a port::
694
695   testpmd> vlan set filter (on|off) (port_id)
696
697vlan set qinq
698~~~~~~~~~~~~~
699
700Set the VLAN QinQ (extended queue in queue) on for a port::
701
702   testpmd> vlan set qinq (on|off) (port_id)
703
704vlan set tpid
705~~~~~~~~~~~~~
706
707Set the inner or outer VLAN TPID for packet filtering on a port::
708
709   testpmd> vlan set (inner|outer) tpid (value) (port_id)
710
711.. note::
712
713   TPID value must be a 16-bit number (value <= 65536).
714
715rx_vlan add
716~~~~~~~~~~~
717
718Add a VLAN ID, or all identifiers, to the set of VLAN identifiers filtered by port ID::
719
720   testpmd> rx_vlan add (vlan_id|all) (port_id)
721
722.. note::
723
724   VLAN filter must be set on that port. VLAN ID < 4096.
725   Depending on the NIC used, number of vlan_ids may be limited to the maximum entries
726   in VFTA table. This is important if enabling all vlan_ids.
727
728rx_vlan rm
729~~~~~~~~~~
730
731Remove a VLAN ID, or all identifiers, from the set of VLAN identifiers filtered by port ID::
732
733   testpmd> rx_vlan rm (vlan_id|all) (port_id)
734
735rx_vlan add (for VF)
736~~~~~~~~~~~~~~~~~~~~
737
738Add a VLAN ID, to the set of VLAN identifiers filtered for VF(s) for port ID::
739
740   testpmd> rx_vlan add (vlan_id) port (port_id) vf (vf_mask)
741
742rx_vlan rm (for VF)
743~~~~~~~~~~~~~~~~~~~
744
745Remove a VLAN ID, from the set of VLAN identifiers filtered for VF(s) for port ID::
746
747   testpmd> rx_vlan rm (vlan_id) port (port_id) vf (vf_mask)
748
749tunnel_filter add
750~~~~~~~~~~~~~~~~~
751
752Add a tunnel filter on a port::
753
754   testpmd> tunnel_filter add (port_id) (outer_mac) (inner_mac) (ip_addr) \
755            (inner_vlan) (vxlan|nvgre|ipingre) (imac-ivlan|imac-ivlan-tenid|\
756            imac-tenid|imac|omac-imac-tenid|oip|iip) (tenant_id) (queue_id)
757
758The available information categories are:
759
760* ``vxlan``: Set tunnel type as VXLAN.
761
762* ``nvgre``: Set tunnel type as NVGRE.
763
764* ``ipingre``: Set tunnel type as IP-in-GRE.
765
766* ``imac-ivlan``: Set filter type as Inner MAC and VLAN.
767
768* ``imac-ivlan-tenid``: Set filter type as Inner MAC, VLAN and tenant ID.
769
770* ``imac-tenid``: Set filter type as Inner MAC and tenant ID.
771
772* ``imac``: Set filter type as Inner MAC.
773
774* ``omac-imac-tenid``: Set filter type as Outer MAC, Inner MAC and tenant ID.
775
776* ``oip``: Set filter type as Outer IP.
777
778* ``iip``: Set filter type as Inner IP.
779
780Example::
781
782   testpmd> tunnel_filter add 0 68:05:CA:28:09:82 00:00:00:00:00:00 \
783            192.168.2.2 0 ipingre oip 1 1
784
785   Set an IP-in-GRE tunnel on port 0, and the filter type is Outer IP.
786
787tunnel_filter remove
788~~~~~~~~~~~~~~~~~~~~
789
790Remove a tunnel filter on a port::
791
792   testpmd> tunnel_filter rm (port_id) (outer_mac) (inner_mac) (ip_addr) \
793            (inner_vlan) (vxlan|nvgre|ipingre) (imac-ivlan|imac-ivlan-tenid|\
794            imac-tenid|imac|omac-imac-tenid|oip|iip) (tenant_id) (queue_id)
795
796rx_vxlan_port add
797~~~~~~~~~~~~~~~~~
798
799Add an UDP port for VXLAN packet filter on a port::
800
801   testpmd> rx_vxlan_port add (udp_port) (port_id)
802
803rx_vxlan_port remove
804~~~~~~~~~~~~~~~~~~~~
805
806Remove an UDP port for VXLAN packet filter on a port::
807
808   testpmd> rx_vxlan_port rm (udp_port) (port_id)
809
810tx_vlan set
811~~~~~~~~~~~
812
813Set hardware insertion of VLAN IDs in packets sent on a port::
814
815   testpmd> tx_vlan set (port_id) vlan_id[, vlan_id_outer]
816
817For example, set a single VLAN ID (5) insertion on port 0::
818
819   tx_vlan set 0 5
820
821Or, set double VLAN ID (inner: 2, outer: 3) insertion on port 1::
822
823   tx_vlan set 1 2 3
824
825
826tx_vlan set pvid
827~~~~~~~~~~~~~~~~
828
829Set port based hardware insertion of VLAN ID in packets sent on a port::
830
831   testpmd> tx_vlan set pvid (port_id) (vlan_id) (on|off)
832
833tx_vlan reset
834~~~~~~~~~~~~~
835
836Disable hardware insertion of a VLAN header in packets sent on a port::
837
838   testpmd> tx_vlan reset (port_id)
839
840csum set
841~~~~~~~~
842
843Select hardware or software calculation of the checksum when
844transmitting a packet using the ``csum`` forwarding engine::
845
846   testpmd> csum set (ip|udp|tcp|sctp|outer-ip) (hw|sw) (port_id)
847
848Where:
849
850* ``ip|udp|tcp|sctp`` always relate to  the inner layer.
851
852* ``outer-ip`` relates to the outer IP layer (only for IPv4) in the case where the packet is recognized
853  as a tunnel packet by the forwarding engine (vxlan, gre and ipip are
854  supported). See also the ``csum parse-tunnel`` command.
855
856.. note::
857
858   Check the NIC Datasheet for hardware limits.
859
860csum parse-tunnel
861~~~~~~~~~~~~~~~~~
862
863Define how tunneled packets should be handled by the csum forward
864engine::
865
866   testpmd> csum parse-tunnel (on|off) (tx_port_id)
867
868If enabled, the csum forward engine will try to recognize supported
869tunnel headers (vxlan, gre, ipip).
870
871If disabled, treat tunnel packets as non-tunneled packets (a inner
872header is handled as a packet payload).
873
874.. note::
875
876   The port argument is the TX port like in the ``csum set`` command.
877
878Example:
879
880Consider a packet in packet like the following::
881
882   eth_out/ipv4_out/udp_out/vxlan/eth_in/ipv4_in/tcp_in
883
884* If parse-tunnel is enabled, the ``ip|udp|tcp|sctp`` parameters of ``csum set``
885  command relate to the inner headers (here ``ipv4_in`` and ``tcp_in``), and the
886  ``outer-ip parameter`` relates to the outer headers (here ``ipv4_out``).
887
888* If parse-tunnel is disabled, the ``ip|udp|tcp|sctp`` parameters of ``csum  set``
889   command relate to the outer headers, here ``ipv4_out`` and ``udp_out``.
890
891csum show
892~~~~~~~~~
893
894Display tx checksum offload configuration::
895
896   testpmd> csum show (port_id)
897
898tso set
899~~~~~~~
900
901Enable TCP Segmentation Offload (TSO) in the ``csum`` forwarding engine::
902
903   testpmd> tso set (segsize) (port_id)
904
905.. note::
906
907   Check the NIC datasheet for hardware limits.
908
909tso show
910~~~~~~~~
911
912Display the status of TCP Segmentation Offload::
913
914   testpmd> tso show (port_id)
915
916set port - gro
917~~~~~~~~~~~~~~
918
919Enable or disable GRO in ``csum`` forwarding engine::
920
921   testpmd> set port <port_id> gro on|off
922
923If enabled, the csum forwarding engine will perform GRO on the TCP/IPv4
924packets received from the given port.
925
926If disabled, packets received from the given port won't be performed
927GRO. By default, GRO is disabled for all ports.
928
929.. note::
930
931   When enable GRO for a port, TCP/IPv4 packets received from the port
932   will be performed GRO. After GRO, all merged packets have bad
933   checksums, since the GRO library doesn't re-calculate checksums for
934   the merged packets. Therefore, if users want the merged packets to
935   have correct checksums, please select HW IP checksum calculation and
936   HW TCP checksum calculation for the port which the merged packets are
937   transmitted to.
938
939show port - gro
940~~~~~~~~~~~~~~~
941
942Display GRO configuration for a given port::
943
944   testpmd> show port <port_id> gro
945
946set gro flush
947~~~~~~~~~~~~~
948
949Set the cycle to flush the GROed packets from reassembly tables::
950
951   testpmd> set gro flush <cycles>
952
953When enable GRO, the csum forwarding engine performs GRO on received
954packets, and the GROed packets are stored in reassembly tables. Users
955can use this command to determine when the GROed packets are flushed
956from the reassembly tables.
957
958The ``cycles`` is measured in GRO operation times. The csum forwarding
959engine flushes the GROed packets from the tables every ``cycles`` GRO
960operations.
961
962By default, the value of ``cycles`` is 1, which means flush GROed packets
963from the reassembly tables as soon as one GRO operation finishes. The value
964of ``cycles`` should be in the range of 1 to ``GRO_MAX_FLUSH_CYCLES``.
965
966Please note that the large value of ``cycles`` may cause the poor TCP/IP
967stack performance. Because the GROed packets are delayed to arrive the
968stack, thus causing more duplicated ACKs and TCP retransmissions.
969
970set port - gso
971~~~~~~~~~~~~~~
972
973Toggle per-port GSO support in ``csum`` forwarding engine::
974
975   testpmd> set port <port_id> gso on|off
976
977If enabled, the csum forwarding engine will perform GSO on supported IPv4
978packets, transmitted on the given port.
979
980If disabled, packets transmitted on the given port will not undergo GSO.
981By default, GSO is disabled for all ports.
982
983.. note::
984
985   When GSO is enabled on a port, supported IPv4 packets transmitted on that
986   port undergo GSO. Afterwards, the segmented packets are represented by
987   multi-segment mbufs; however, the csum forwarding engine doesn't calculation
988   of checksums for GSO'd segments in SW. As a result, if users want correct
989   checksums in GSO segments, they should enable HW checksum calculation for
990   GSO-enabled ports.
991
992   For example, HW checksum calculation for VxLAN GSO'd packets may be enabled
993   by setting the following options in the csum forwarding engine:
994
995   testpmd> csum set outer_ip hw <port_id>
996
997   testpmd> csum set ip hw <port_id>
998
999   testpmd> csum set tcp hw <port_id>
1000
1001set gso segsz
1002~~~~~~~~~~~~~
1003
1004Set the maximum GSO segment size (measured in bytes), which includes the
1005packet header and the packet payload for GSO-enabled ports (global)::
1006
1007   testpmd> set gso segsz <length>
1008
1009show port - gso
1010~~~~~~~~~~~~~~~
1011
1012Display the status of Generic Segmentation Offload for a given port::
1013
1014   testpmd> show port <port_id> gso
1015
1016mac_addr add
1017~~~~~~~~~~~~
1018
1019Add an alternative MAC address to a port::
1020
1021   testpmd> mac_addr add (port_id) (XX:XX:XX:XX:XX:XX)
1022
1023mac_addr remove
1024~~~~~~~~~~~~~~~
1025
1026Remove a MAC address from a port::
1027
1028   testpmd> mac_addr remove (port_id) (XX:XX:XX:XX:XX:XX)
1029
1030mac_addr add (for VF)
1031~~~~~~~~~~~~~~~~~~~~~
1032
1033Add an alternative MAC address for a VF to a port::
1034
1035   testpmd> mac_add add port (port_id) vf (vf_id) (XX:XX:XX:XX:XX:XX)
1036
1037mac_addr set
1038~~~~~~~~~~~~
1039
1040Set the default MAC address for a port::
1041
1042   testpmd> mac_addr set (port_id) (XX:XX:XX:XX:XX:XX)
1043
1044mac_addr set (for VF)
1045~~~~~~~~~~~~~~~~~~~~~
1046
1047Set the MAC address for a VF from the PF::
1048
1049   testpmd> set vf mac addr (port_id) (vf_id) (XX:XX:XX:XX:XX:XX)
1050
1051set port-uta
1052~~~~~~~~~~~~
1053
1054Set the unicast hash filter(s) on/off for a port::
1055
1056   testpmd> set port (port_id) uta (XX:XX:XX:XX:XX:XX|all) (on|off)
1057
1058set promisc
1059~~~~~~~~~~~
1060
1061Set the promiscuous mode on for a port or for all ports.
1062In promiscuous mode packets are not dropped if they aren't for the specified MAC address::
1063
1064   testpmd> set promisc (port_id|all) (on|off)
1065
1066set allmulti
1067~~~~~~~~~~~~
1068
1069Set the allmulti mode for a port or for all ports::
1070
1071   testpmd> set allmulti (port_id|all) (on|off)
1072
1073Same as the ifconfig (8) option. Controls how multicast packets are handled.
1074
1075set promisc (for VF)
1076~~~~~~~~~~~~~~~~~~~~
1077
1078Set the unicast promiscuous mode for a VF from PF.
1079It's supported by Intel i40e NICs now.
1080In promiscuous mode packets are not dropped if they aren't for the specified MAC address::
1081
1082   testpmd> set vf promisc (port_id) (vf_id) (on|off)
1083
1084set allmulticast (for VF)
1085~~~~~~~~~~~~~~~~~~~~~~~~~
1086
1087Set the multicast promiscuous mode for a VF from PF.
1088It's supported by Intel i40e NICs now.
1089In promiscuous mode packets are not dropped if they aren't for the specified MAC address::
1090
1091   testpmd> set vf allmulti (port_id) (vf_id) (on|off)
1092
1093set tx max bandwidth (for VF)
1094~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1095
1096Set TX max absolute bandwidth (Mbps) for a VF from PF::
1097
1098   testpmd> set vf tx max-bandwidth (port_id) (vf_id) (max_bandwidth)
1099
1100set tc tx min bandwidth (for VF)
1101~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1102
1103Set all TCs' TX min relative bandwidth (%) for a VF from PF::
1104
1105   testpmd> set vf tc tx min-bandwidth (port_id) (vf_id) (bw1, bw2, ...)
1106
1107set tc tx max bandwidth (for VF)
1108~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1109
1110Set a TC's TX max absolute bandwidth (Mbps) for a VF from PF::
1111
1112   testpmd> set vf tc tx max-bandwidth (port_id) (vf_id) (tc_no) (max_bandwidth)
1113
1114set tc strict link priority mode
1115~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1116
1117Set some TCs' strict link priority mode on a physical port::
1118
1119   testpmd> set tx strict-link-priority (port_id) (tc_bitmap)
1120
1121set tc tx min bandwidth
1122~~~~~~~~~~~~~~~~~~~~~~~
1123
1124Set all TCs' TX min relative bandwidth (%) globally for all PF and VFs::
1125
1126   testpmd> set tc tx min-bandwidth (port_id) (bw1, bw2, ...)
1127
1128set flow_ctrl rx
1129~~~~~~~~~~~~~~~~
1130
1131Set the link flow control parameter on a port::
1132
1133   testpmd> set flow_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \
1134            (pause_time) (send_xon) mac_ctrl_frame_fwd (on|off) \
1135	    autoneg (on|off) (port_id)
1136
1137Where:
1138
1139* ``high_water`` (integer): High threshold value to trigger XOFF.
1140
1141* ``low_water`` (integer): Low threshold value to trigger XON.
1142
1143* ``pause_time`` (integer): Pause quota in the Pause frame.
1144
1145* ``send_xon`` (0/1): Send XON frame.
1146
1147* ``mac_ctrl_frame_fwd``: Enable receiving MAC control frames.
1148
1149* ``autoneg``: Change the auto-negotiation parameter.
1150
1151set pfc_ctrl rx
1152~~~~~~~~~~~~~~~
1153
1154Set the priority flow control parameter on a port::
1155
1156   testpmd> set pfc_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \
1157            (pause_time) (priority) (port_id)
1158
1159Where:
1160
1161* ``high_water`` (integer): High threshold value.
1162
1163* ``low_water`` (integer): Low threshold value.
1164
1165* ``pause_time`` (integer): Pause quota in the Pause frame.
1166
1167* ``priority`` (0-7): VLAN User Priority.
1168
1169set stat_qmap
1170~~~~~~~~~~~~~
1171
1172Set statistics mapping (qmapping 0..15) for RX/TX queue on port::
1173
1174   testpmd> set stat_qmap (tx|rx) (port_id) (queue_id) (qmapping)
1175
1176For example, to set rx queue 2 on port 0 to mapping 5::
1177
1178   testpmd>set stat_qmap rx 0 2 5
1179
1180set port - rx/tx (for VF)
1181~~~~~~~~~~~~~~~~~~~~~~~~~
1182
1183Set VF receive/transmit from a port::
1184
1185   testpmd> set port (port_id) vf (vf_id) (rx|tx) (on|off)
1186
1187set port - mac address filter (for VF)
1188~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1189
1190Add/Remove unicast or multicast MAC addr filter for a VF::
1191
1192   testpmd> set port (port_id) vf (vf_id) (mac_addr) \
1193            (exact-mac|exact-mac-vlan|hashmac|hashmac-vlan) (on|off)
1194
1195set port - rx mode(for VF)
1196~~~~~~~~~~~~~~~~~~~~~~~~~~
1197
1198Set the VF receive mode of a port::
1199
1200   testpmd> set port (port_id) vf (vf_id) \
1201            rxmode (AUPE|ROPE|BAM|MPE) (on|off)
1202
1203The available receive modes are:
1204
1205* ``AUPE``: Accepts untagged VLAN.
1206
1207* ``ROPE``: Accepts unicast hash.
1208
1209* ``BAM``: Accepts broadcast packets.
1210
1211* ``MPE``: Accepts all multicast packets.
1212
1213set port - tx_rate (for Queue)
1214~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1215
1216Set TX rate limitation for a queue on a port::
1217
1218   testpmd> set port (port_id) queue (queue_id) rate (rate_value)
1219
1220set port - tx_rate (for VF)
1221~~~~~~~~~~~~~~~~~~~~~~~~~~~
1222
1223Set TX rate limitation for queues in VF on a port::
1224
1225   testpmd> set port (port_id) vf (vf_id) rate (rate_value) queue_mask (queue_mask)
1226
1227set port - mirror rule
1228~~~~~~~~~~~~~~~~~~~~~~
1229
1230Set pool or vlan type mirror rule for a port::
1231
1232   testpmd> set port (port_id) mirror-rule (rule_id) \
1233            (pool-mirror-up|pool-mirror-down|vlan-mirror) \
1234            (poolmask|vlanid[,vlanid]*) dst-pool (pool_id) (on|off)
1235
1236Set link mirror rule for a port::
1237
1238   testpmd> set port (port_id) mirror-rule (rule_id) \
1239           (uplink-mirror|downlink-mirror) dst-pool (pool_id) (on|off)
1240
1241For example to enable mirror traffic with vlan 0,1 to pool 0::
1242
1243   set port 0 mirror-rule 0 vlan-mirror 0,1 dst-pool 0 on
1244
1245reset port - mirror rule
1246~~~~~~~~~~~~~~~~~~~~~~~~
1247
1248Reset a mirror rule for a port::
1249
1250   testpmd> reset port (port_id) mirror-rule (rule_id)
1251
1252set flush_rx
1253~~~~~~~~~~~~
1254
1255Set the flush on RX streams before forwarding.
1256The default is flush ``on``.
1257Mainly used with PCAP drivers to turn off the default behavior of flushing the first 512 packets on RX streams::
1258
1259   testpmd> set flush_rx off
1260
1261set bypass mode
1262~~~~~~~~~~~~~~~
1263
1264Set the bypass mode for the lowest port on bypass enabled NIC::
1265
1266   testpmd> set bypass mode (normal|bypass|isolate) (port_id)
1267
1268set bypass event
1269~~~~~~~~~~~~~~~~
1270
1271Set the event required to initiate specified bypass mode for the lowest port on a bypass enabled::
1272
1273   testpmd> set bypass event (timeout|os_on|os_off|power_on|power_off) \
1274            mode (normal|bypass|isolate) (port_id)
1275
1276Where:
1277
1278* ``timeout``: Enable bypass after watchdog timeout.
1279
1280* ``os_on``: Enable bypass when OS/board is powered on.
1281
1282* ``os_off``: Enable bypass when OS/board is powered off.
1283
1284* ``power_on``: Enable bypass when power supply is turned on.
1285
1286* ``power_off``: Enable bypass when power supply is turned off.
1287
1288
1289set bypass timeout
1290~~~~~~~~~~~~~~~~~~
1291
1292Set the bypass watchdog timeout to ``n`` seconds where 0 = instant::
1293
1294   testpmd> set bypass timeout (0|1.5|2|3|4|8|16|32)
1295
1296show bypass config
1297~~~~~~~~~~~~~~~~~~
1298
1299Show the bypass configuration for a bypass enabled NIC using the lowest port on the NIC::
1300
1301   testpmd> show bypass config (port_id)
1302
1303set link up
1304~~~~~~~~~~~
1305
1306Set link up for a port::
1307
1308   testpmd> set link-up port (port id)
1309
1310set link down
1311~~~~~~~~~~~~~
1312
1313Set link down for a port::
1314
1315   testpmd> set link-down port (port id)
1316
1317E-tag set
1318~~~~~~~~~
1319
1320Enable E-tag insertion for a VF on a port::
1321
1322   testpmd> E-tag set insertion on port-tag-id (value) port (port_id) vf (vf_id)
1323
1324Disable E-tag insertion for a VF on a port::
1325
1326   testpmd> E-tag set insertion off port (port_id) vf (vf_id)
1327
1328Enable/disable E-tag stripping on a port::
1329
1330   testpmd> E-tag set stripping (on|off) port (port_id)
1331
1332Enable/disable E-tag based forwarding on a port::
1333
1334   testpmd> E-tag set forwarding (on|off) port (port_id)
1335
1336Add an E-tag forwarding filter on a port::
1337
1338   testpmd> E-tag set filter add e-tag-id (value) dst-pool (pool_id) port (port_id)
1339
1340Delete an E-tag forwarding filter on a port::
1341   testpmd> E-tag set filter del e-tag-id (value) port (port_id)
1342
1343ddp add
1344~~~~~~~
1345
1346Load a dynamic device personalization (DDP) package::
1347
1348   testpmd> ddp add (port_id) (package_path[,output_path])
1349
1350ddp del
1351~~~~~~~
1352
1353Delete a dynamic device personalization package::
1354
1355   testpmd> ddp del (port_id) (package_path)
1356
1357ptype mapping
1358~~~~~~~~~~~~~
1359
1360List all items from the ptype mapping table::
1361
1362   testpmd> ptype mapping get (port_id) (valid_only)
1363
1364Where:
1365
1366* ``valid_only``: A flag indicates if only list valid items(=1) or all itemss(=0).
1367
1368Replace a specific or a group of software defined ptype with a new one::
1369
1370   testpmd> ptype mapping replace  (port_id) (target) (mask) (pkt_type)
1371
1372where:
1373
1374* ``target``: A specific software ptype or a mask to represent a group of software ptypes.
1375
1376* ``mask``: A flag indicate if "target" is a specific software ptype(=0) or a ptype mask(=1).
1377
1378* ``pkt_type``: The new software ptype to replace the old ones.
1379
1380Update hardware defined ptype to software defined packet type mapping table::
1381
1382   testpmd> ptype mapping update (port_id) (hw_ptype) (sw_ptype)
1383
1384where:
1385
1386* ``hw_ptype``: hardware ptype as the index of the ptype mapping table.
1387
1388* ``sw_ptype``: software ptype as the value of the ptype mapping table.
1389
1390Reset ptype mapping table::
1391
1392   testpmd> ptype mapping reset (port_id)
1393
1394Port Functions
1395--------------
1396
1397The following sections show functions for configuring ports.
1398
1399.. note::
1400
1401   Port configuration changes only become active when forwarding is started/restarted.
1402
1403port attach
1404~~~~~~~~~~~
1405
1406Attach a port specified by pci address or virtual device args::
1407
1408   testpmd> port attach (identifier)
1409
1410To attach a new pci device, the device should be recognized by kernel first.
1411Then it should be moved under DPDK management.
1412Finally the port can be attached to testpmd.
1413
1414For example, to move a pci device using ixgbe under DPDK management:
1415
1416.. code-block:: console
1417
1418   # Check the status of the available devices.
1419   ./usertools/dpdk-devbind.py --status
1420
1421   Network devices using DPDK-compatible driver
1422   ============================================
1423   <none>
1424
1425   Network devices using kernel driver
1426   ===================================
1427   0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused=
1428
1429
1430   # Bind the device to igb_uio.
1431   sudo ./usertools/dpdk-devbind.py -b igb_uio 0000:0a:00.0
1432
1433
1434   # Recheck the status of the devices.
1435   ./usertools/dpdk-devbind.py --status
1436   Network devices using DPDK-compatible driver
1437   ============================================
1438   0000:0a:00.0 '82599ES 10-Gigabit' drv=igb_uio unused=
1439
1440To attach a port created by virtual device, above steps are not needed.
1441
1442For example, to attach a port whose pci address is 0000:0a:00.0.
1443
1444.. code-block:: console
1445
1446   testpmd> port attach 0000:0a:00.0
1447   Attaching a new port...
1448   EAL: PCI device 0000:0a:00.0 on NUMA socket -1
1449   EAL:   probe driver: 8086:10fb rte_ixgbe_pmd
1450   EAL:   PCI memory mapped at 0x7f83bfa00000
1451   EAL:   PCI memory mapped at 0x7f83bfa80000
1452   PMD: eth_ixgbe_dev_init(): MAC: 2, PHY: 18, SFP+: 5
1453   PMD: eth_ixgbe_dev_init(): port 0 vendorID=0x8086 deviceID=0x10fb
1454   Port 0 is attached. Now total ports is 1
1455   Done
1456
1457For example, to attach a port created by pcap PMD.
1458
1459.. code-block:: console
1460
1461   testpmd> port attach net_pcap0
1462   Attaching a new port...
1463   PMD: Initializing pmd_pcap for net_pcap0
1464   PMD: Creating pcap-backed ethdev on numa socket 0
1465   Port 0 is attached. Now total ports is 1
1466   Done
1467
1468In this case, identifier is ``net_pcap0``.
1469This identifier format is the same as ``--vdev`` format of DPDK applications.
1470
1471For example, to re-attach a bonded port which has been previously detached,
1472the mode and slave parameters must be given.
1473
1474.. code-block:: console
1475
1476   testpmd> port attach net_bond_0,mode=0,slave=1
1477   Attaching a new port...
1478   EAL: Initializing pmd_bond for net_bond_0
1479   EAL: Create bonded device net_bond_0 on port 0 in mode 0 on socket 0.
1480   Port 0 is attached. Now total ports is 1
1481   Done
1482
1483
1484port detach
1485~~~~~~~~~~~
1486
1487Detach a specific port::
1488
1489   testpmd> port detach (port_id)
1490
1491Before detaching a port, the port should be stopped and closed.
1492
1493For example, to detach a pci device port 0.
1494
1495.. code-block:: console
1496
1497   testpmd> port stop 0
1498   Stopping ports...
1499   Done
1500   testpmd> port close 0
1501   Closing ports...
1502   Done
1503
1504   testpmd> port detach 0
1505   Detaching a port...
1506   EAL: PCI device 0000:0a:00.0 on NUMA socket -1
1507   EAL:   remove driver: 8086:10fb rte_ixgbe_pmd
1508   EAL:   PCI memory unmapped at 0x7f83bfa00000
1509   EAL:   PCI memory unmapped at 0x7f83bfa80000
1510   Done
1511
1512
1513For example, to detach a virtual device port 0.
1514
1515.. code-block:: console
1516
1517   testpmd> port stop 0
1518   Stopping ports...
1519   Done
1520   testpmd> port close 0
1521   Closing ports...
1522   Done
1523
1524   testpmd> port detach 0
1525   Detaching a port...
1526   PMD: Closing pcap ethdev on numa socket 0
1527   Port 'net_pcap0' is detached. Now total ports is 0
1528   Done
1529
1530To remove a pci device completely from the system, first detach the port from testpmd.
1531Then the device should be moved under kernel management.
1532Finally the device can be removed using kernel pci hotplug functionality.
1533
1534For example, to move a pci device under kernel management:
1535
1536.. code-block:: console
1537
1538   sudo ./usertools/dpdk-devbind.py -b ixgbe 0000:0a:00.0
1539
1540   ./usertools/dpdk-devbind.py --status
1541
1542   Network devices using DPDK-compatible driver
1543   ============================================
1544   <none>
1545
1546   Network devices using kernel driver
1547   ===================================
1548   0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused=igb_uio
1549
1550To remove a port created by a virtual device, above steps are not needed.
1551
1552port start
1553~~~~~~~~~~
1554
1555Start all ports or a specific port::
1556
1557   testpmd> port start (port_id|all)
1558
1559port stop
1560~~~~~~~~~
1561
1562Stop all ports or a specific port::
1563
1564   testpmd> port stop (port_id|all)
1565
1566port close
1567~~~~~~~~~~
1568
1569Close all ports or a specific port::
1570
1571   testpmd> port close (port_id|all)
1572
1573port start/stop queue
1574~~~~~~~~~~~~~~~~~~~~~
1575
1576Start/stop a rx/tx queue on a specific port::
1577
1578   testpmd> port (port_id) (rxq|txq) (queue_id) (start|stop)
1579
1580Only take effect when port is started.
1581
1582port config - speed
1583~~~~~~~~~~~~~~~~~~~
1584
1585Set the speed and duplex mode for all ports or a specific port::
1586
1587   testpmd> port config (port_id|all) speed (10|100|1000|10000|25000|40000|50000|100000|auto) \
1588            duplex (half|full|auto)
1589
1590port config - queues/descriptors
1591~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1592
1593Set number of queues/descriptors for rxq, txq, rxd and txd::
1594
1595   testpmd> port config all (rxq|txq|rxd|txd) (value)
1596
1597This is equivalent to the ``--rxq``, ``--txq``, ``--rxd`` and ``--txd`` command-line options.
1598
1599port config - max-pkt-len
1600~~~~~~~~~~~~~~~~~~~~~~~~~
1601
1602Set the maximum packet length::
1603
1604   testpmd> port config all max-pkt-len (value)
1605
1606This is equivalent to the ``--max-pkt-len`` command-line option.
1607
1608port config - CRC Strip
1609~~~~~~~~~~~~~~~~~~~~~~~
1610
1611Set hardware CRC stripping on or off for all ports::
1612
1613   testpmd> port config all crc-strip (on|off)
1614
1615CRC stripping is on by default.
1616
1617The ``off`` option is equivalent to the ``--disable-crc-strip`` command-line option.
1618
1619port config - scatter
1620~~~~~~~~~~~~~~~~~~~~~~~
1621
1622Set RX scatter mode on or off for all ports::
1623
1624   testpmd> port config all scatter (on|off)
1625
1626RX scatter mode is off by default.
1627
1628The ``on`` option is equivalent to the ``--enable-scatter`` command-line option.
1629
1630port config - TX queue flags
1631~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1632
1633Set a hexadecimal bitmap of TX queue flags for all ports::
1634
1635   testpmd> port config all txqflags value
1636
1637This command is equivalent to the ``--txqflags`` command-line option.
1638
1639port config - RX Checksum
1640~~~~~~~~~~~~~~~~~~~~~~~~~
1641
1642Set hardware RX checksum offload to on or off for all ports::
1643
1644   testpmd> port config all rx-cksum (on|off)
1645
1646Checksum offload is off by default.
1647
1648The ``on`` option is equivalent to the ``--enable-rx-cksum`` command-line option.
1649
1650port config - VLAN
1651~~~~~~~~~~~~~~~~~~
1652
1653Set hardware VLAN on or off for all ports::
1654
1655   testpmd> port config all hw-vlan (on|off)
1656
1657Hardware VLAN is on by default.
1658
1659The ``off`` option is equivalent to the ``--disable-hw-vlan`` command-line option.
1660
1661port config - VLAN filter
1662~~~~~~~~~~~~~~~~~~~~~~~~~
1663
1664Set hardware VLAN filter on or off for all ports::
1665
1666   testpmd> port config all hw-vlan-filter (on|off)
1667
1668Hardware VLAN filter is on by default.
1669
1670The ``off`` option is equivalent to the ``--disable-hw-vlan-filter`` command-line option.
1671
1672port config - VLAN strip
1673~~~~~~~~~~~~~~~~~~~~~~~~
1674
1675Set hardware VLAN strip on or off for all ports::
1676
1677   testpmd> port config all hw-vlan-strip (on|off)
1678
1679Hardware VLAN strip is on by default.
1680
1681The ``off`` option is equivalent to the ``--disable-hw-vlan-strip`` command-line option.
1682
1683port config - VLAN extend
1684~~~~~~~~~~~~~~~~~~~~~~~~~
1685
1686Set hardware VLAN extend on or off for all ports::
1687
1688   testpmd> port config all hw-vlan-extend (on|off)
1689
1690Hardware VLAN extend is off by default.
1691
1692The ``off`` option is equivalent to the ``--disable-hw-vlan-extend`` command-line option.
1693
1694port config - Drop Packets
1695~~~~~~~~~~~~~~~~~~~~~~~~~~
1696
1697Set packet drop for packets with no descriptors on or off for all ports::
1698
1699   testpmd> port config all drop-en (on|off)
1700
1701Packet dropping for packets with no descriptors is off by default.
1702
1703The ``on`` option is equivalent to the ``--enable-drop-en`` command-line option.
1704
1705port config - RSS
1706~~~~~~~~~~~~~~~~~
1707
1708Set the RSS (Receive Side Scaling) mode on or off::
1709
1710   testpmd> port config all rss (all|ip|tcp|udp|sctp|ether|port|vxlan|geneve|nvgre|none)
1711
1712RSS is on by default.
1713
1714The ``none`` option is equivalent to the ``--disable-rss`` command-line option.
1715
1716port config - RSS Reta
1717~~~~~~~~~~~~~~~~~~~~~~
1718
1719Set the RSS (Receive Side Scaling) redirection table::
1720
1721   testpmd> port config all rss reta (hash,queue)[,(hash,queue)]
1722
1723port config - DCB
1724~~~~~~~~~~~~~~~~~
1725
1726Set the DCB mode for an individual port::
1727
1728   testpmd> port config (port_id) dcb vt (on|off) (traffic_class) pfc (on|off)
1729
1730The traffic class should be 4 or 8.
1731
1732port config - Burst
1733~~~~~~~~~~~~~~~~~~~
1734
1735Set the number of packets per burst::
1736
1737   testpmd> port config all burst (value)
1738
1739This is equivalent to the ``--burst`` command-line option.
1740
1741port config - Threshold
1742~~~~~~~~~~~~~~~~~~~~~~~
1743
1744Set thresholds for TX/RX queues::
1745
1746   testpmd> port config all (threshold) (value)
1747
1748Where the threshold type can be:
1749
1750* ``txpt:`` Set the prefetch threshold register of the TX rings, 0 <= value <= 255.
1751
1752* ``txht:`` Set the host threshold register of the TX rings, 0 <= value <= 255.
1753
1754* ``txwt:`` Set the write-back threshold register of the TX rings, 0 <= value <= 255.
1755
1756* ``rxpt:`` Set the prefetch threshold register of the RX rings, 0 <= value <= 255.
1757
1758* ``rxht:`` Set the host threshold register of the RX rings, 0 <= value <= 255.
1759
1760* ``rxwt:`` Set the write-back threshold register of the RX rings, 0 <= value <= 255.
1761
1762* ``txfreet:`` Set the transmit free threshold of the TX rings, 0 <= value <= txd.
1763
1764* ``rxfreet:`` Set the transmit free threshold of the RX rings, 0 <= value <= rxd.
1765
1766* ``txrst:`` Set the transmit RS bit threshold of TX rings, 0 <= value <= txd.
1767
1768These threshold options are also available from the command-line.
1769
1770port config - E-tag
1771~~~~~~~~~~~~~~~~~~~
1772
1773Set the value of ether-type for E-tag::
1774
1775   testpmd> port config (port_id|all) l2-tunnel E-tag ether-type (value)
1776
1777Enable/disable the E-tag support::
1778
1779   testpmd> port config (port_id|all) l2-tunnel E-tag (enable|disable)
1780
1781port config pctype mapping
1782~~~~~~~~~~~~~~~~~~~~~~~~~~
1783
1784Reset pctype mapping table::
1785
1786   testpmd> port config (port_id) pctype mapping reset
1787
1788Update hardware defined pctype to software defined flow type mapping table::
1789
1790   testpmd> port config (port_id) pctype mapping update (pctype_id_0[,pctype_id_1]*) (flow_type_id)
1791
1792where:
1793
1794* ``pctype_id_x``: hardware pctype id as index of bit in bitmask value of the pctype mapping table.
1795
1796* ``flow_type_id``: software flow type id as the index of the pctype mapping table.
1797
1798
1799Link Bonding Functions
1800----------------------
1801
1802The Link Bonding functions make it possible to dynamically create and
1803manage link bonding devices from within testpmd interactive prompt.
1804
1805create bonded device
1806~~~~~~~~~~~~~~~~~~~~
1807
1808Create a new bonding device::
1809
1810   testpmd> create bonded device (mode) (socket)
1811
1812For example, to create a bonded device in mode 1 on socket 0::
1813
1814   testpmd> create bonded 1 0
1815   created new bonded device (port X)
1816
1817add bonding slave
1818~~~~~~~~~~~~~~~~~
1819
1820Adds Ethernet device to a Link Bonding device::
1821
1822   testpmd> add bonding slave (slave id) (port id)
1823
1824For example, to add Ethernet device (port 6) to a Link Bonding device (port 10)::
1825
1826   testpmd> add bonding slave 6 10
1827
1828
1829remove bonding slave
1830~~~~~~~~~~~~~~~~~~~~
1831
1832Removes an Ethernet slave device from a Link Bonding device::
1833
1834   testpmd> remove bonding slave (slave id) (port id)
1835
1836For example, to remove Ethernet slave device (port 6) to a Link Bonding device (port 10)::
1837
1838   testpmd> remove bonding slave 6 10
1839
1840set bonding mode
1841~~~~~~~~~~~~~~~~
1842
1843Set the Link Bonding mode of a Link Bonding device::
1844
1845   testpmd> set bonding mode (value) (port id)
1846
1847For example, to set the bonding mode of a Link Bonding device (port 10) to broadcast (mode 3)::
1848
1849   testpmd> set bonding mode 3 10
1850
1851set bonding primary
1852~~~~~~~~~~~~~~~~~~~
1853
1854Set an Ethernet slave device as the primary device on a Link Bonding device::
1855
1856   testpmd> set bonding primary (slave id) (port id)
1857
1858For example, to set the Ethernet slave device (port 6) as the primary port of a Link Bonding device (port 10)::
1859
1860   testpmd> set bonding primary 6 10
1861
1862set bonding mac
1863~~~~~~~~~~~~~~~
1864
1865Set the MAC address of a Link Bonding device::
1866
1867   testpmd> set bonding mac (port id) (mac)
1868
1869For example, to set the MAC address of a Link Bonding device (port 10) to 00:00:00:00:00:01::
1870
1871   testpmd> set bonding mac 10 00:00:00:00:00:01
1872
1873set bonding xmit_balance_policy
1874~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1875
1876Set the transmission policy for a Link Bonding device when it is in Balance XOR mode::
1877
1878   testpmd> set bonding xmit_balance_policy (port_id) (l2|l23|l34)
1879
1880For example, set a Link Bonding device (port 10) to use a balance policy of layer 3+4 (IP addresses & UDP ports)::
1881
1882   testpmd> set bonding xmit_balance_policy 10 l34
1883
1884
1885set bonding mon_period
1886~~~~~~~~~~~~~~~~~~~~~~
1887
1888Set the link status monitoring polling period in milliseconds for a bonding device.
1889
1890This adds support for PMD slave devices which do not support link status interrupts.
1891When the mon_period is set to a value greater than 0 then all PMD's which do not support
1892link status ISR will be queried every polling interval to check if their link status has changed::
1893
1894   testpmd> set bonding mon_period (port_id) (value)
1895
1896For example, to set the link status monitoring polling period of bonded device (port 5) to 150ms::
1897
1898   testpmd> set bonding mon_period 5 150
1899
1900
1901set bonding lacp dedicated_queue
1902~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1903
1904Enable dedicated tx/rx queues on bonding devices slaves to handle LACP control plane traffic
1905when in mode 4 (link-aggregration-802.3ad)::
1906
1907   testpmd> set bonding lacp dedicated_queues (port_id) (enable|disable)
1908
1909
1910set bonding agg_mode
1911~~~~~~~~~~~~~~~~~~~~
1912
1913Enable one of the specific aggregators mode when in mode 4 (link-aggregration-802.3ad)::
1914
1915   testpmd> set bonding agg_mode (port_id) (bandwidth|count|stable)
1916
1917
1918show bonding config
1919~~~~~~~~~~~~~~~~~~~
1920
1921Show the current configuration of a Link Bonding device::
1922
1923   testpmd> show bonding config (port id)
1924
1925For example,
1926to show the configuration a Link Bonding device (port 9) with 3 slave devices (1, 3, 4)
1927in balance mode with a transmission policy of layer 2+3::
1928
1929   testpmd> show bonding config 9
1930        Bonding mode: 2
1931        Balance Xmit Policy: BALANCE_XMIT_POLICY_LAYER23
1932        Slaves (3): [1 3 4]
1933        Active Slaves (3): [1 3 4]
1934        Primary: [3]
1935
1936
1937Register Functions
1938------------------
1939
1940The Register Functions can be used to read from and write to registers on the network card referenced by a port number.
1941This is mainly useful for debugging purposes.
1942Reference should be made to the appropriate datasheet for the network card for details on the register addresses
1943and fields that can be accessed.
1944
1945read reg
1946~~~~~~~~
1947
1948Display the value of a port register::
1949
1950   testpmd> read reg (port_id) (address)
1951
1952For example, to examine the Flow Director control register (FDIRCTL, 0x0000EE000) on an Intel 82599 10 GbE Controller::
1953
1954   testpmd> read reg 0 0xEE00
1955   port 0 PCI register at offset 0xEE00: 0x4A060029 (1241907241)
1956
1957read regfield
1958~~~~~~~~~~~~~
1959
1960Display a port register bit field::
1961
1962   testpmd> read regfield (port_id) (address) (bit_x) (bit_y)
1963
1964For example, reading the lowest two bits from the register in the example above::
1965
1966   testpmd> read regfield 0 0xEE00 0 1
1967   port 0 PCI register at offset 0xEE00: bits[0, 1]=0x1 (1)
1968
1969read regbit
1970~~~~~~~~~~~
1971
1972Display a single port register bit::
1973
1974   testpmd> read regbit (port_id) (address) (bit_x)
1975
1976For example, reading the lowest bit from the register in the example above::
1977
1978   testpmd> read regbit 0 0xEE00 0
1979   port 0 PCI register at offset 0xEE00: bit 0=1
1980
1981write reg
1982~~~~~~~~~
1983
1984Set the value of a port register::
1985
1986   testpmd> write reg (port_id) (address) (value)
1987
1988For example, to clear a register::
1989
1990   testpmd> write reg 0 0xEE00 0x0
1991   port 0 PCI register at offset 0xEE00: 0x00000000 (0)
1992
1993write regfield
1994~~~~~~~~~~~~~~
1995
1996Set bit field of a port register::
1997
1998   testpmd> write regfield (port_id) (address) (bit_x) (bit_y) (value)
1999
2000For example, writing to the register cleared in the example above::
2001
2002   testpmd> write regfield 0 0xEE00 0 1 2
2003   port 0 PCI register at offset 0xEE00: 0x00000002 (2)
2004
2005write regbit
2006~~~~~~~~~~~~
2007
2008Set single bit value of a port register::
2009
2010   testpmd> write regbit (port_id) (address) (bit_x) (value)
2011
2012For example, to set the high bit in the register from the example above::
2013
2014   testpmd> write regbit 0 0xEE00 31 1
2015   port 0 PCI register at offset 0xEE00: 0x8000000A (2147483658)
2016
2017
2018Traffic Management
2019------------------
2020
2021The following section shows functions for configuring traffic management on
2022on the ethernet device through the use of generic TM API.
2023
2024show port traffic management capability
2025~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2026
2027Show traffic management capability of the port::
2028
2029   testpmd> show port tm cap (port_id)
2030
2031show port traffic management capability (hierarchy level)
2032~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2033
2034Show traffic management hierarchy level capability of the port::
2035
2036   testpmd> show port tm cap (port_id) (level_id)
2037
2038show port traffic management capability (hierarchy node level)
2039~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2040
2041Show the traffic management hierarchy node capability of the port::
2042
2043   testpmd> show port tm cap (port_id) (node_id)
2044
2045show port traffic management hierarchy node type
2046~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2047
2048Show the port traffic management hierarchy node type::
2049
2050   testpmd> show port tm node type (port_id) (node_id)
2051
2052show port traffic management hierarchy node stats
2053~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2054
2055Show the port traffic management hierarchy node statistics::
2056
2057   testpmd> show port tm node stats (port_id) (node_id) (clear)
2058
2059where:
2060
2061* ``clear``: When this parameter has a non-zero value, the statistics counters
2062  are cleared (i.e. set to zero) immediately after they have been read,
2063  otherwise the statistics counters are left untouched.
2064
2065Add port traffic management private shaper profile
2066~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2067
2068Add the port traffic management private shaper profile::
2069
2070   testpmd> add port tm node shaper profile (port_id) (shaper_profile_id) \
2071   (tb_rate) (tb_size) (packet_length_adjust)
2072
2073where:
2074
2075* ``shaper_profile id``: Shaper profile ID for the new profile.
2076* ``tb_rate``: Token bucket rate (bytes per second).
2077* ``tb_size``: Token bucket size (bytes).
2078* ``packet_length_adjust``: The value (bytes) to be added to the length of
2079  each packet for the purpose of shaping. This parameter value can be used to
2080  correct the packet length with the framing overhead bytes that are consumed
2081  on the wire.
2082
2083Delete port traffic management private shaper profile
2084~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2085
2086Delete the port traffic management private shaper::
2087
2088   testpmd> del port tm node shaper profile (port_id) (shaper_profile_id)
2089
2090where:
2091
2092* ``shaper_profile id``: Shaper profile ID that needs to be deleted.
2093
2094Add port traffic management shared shaper
2095~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2096
2097Create the port traffic management shared shaper::
2098
2099   testpmd> add port tm node shared shaper (port_id) (shared_shaper_id) \
2100   (shaper_profile_id)
2101
2102where:
2103
2104* ``shared_shaper_id``: Shared shaper ID to be created.
2105* ``shaper_profile id``: Shaper profile ID for shared shaper.
2106
2107Set port traffic management shared shaper
2108~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2109
2110Update the port traffic management shared shaper::
2111
2112   testpmd> set port tm node shared shaper (port_id) (shared_shaper_id) \
2113   (shaper_profile_id)
2114
2115where:
2116
2117* ``shared_shaper_id``: Shared shaper ID to be update.
2118* ``shaper_profile id``: Shaper profile ID for shared shaper.
2119
2120Delete port traffic management shared shaper
2121~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2122
2123Delete the port traffic management shared shaper::
2124
2125   testpmd> del port tm node shared shaper (port_id) (shared_shaper_id)
2126
2127where:
2128
2129* ``shared_shaper_id``: Shared shaper ID to be deleted.
2130
2131Set port traffic management hiearchy node private shaper
2132~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2133
2134set the port traffic management hierarchy node private shaper::
2135
2136   testpmd> set port tm node shaper profile (port_id) (node_id) \
2137   (shaper_profile_id)
2138
2139where:
2140
2141* ``shaper_profile id``: Private shaper profile ID to be enabled on the
2142  hierarchy node.
2143
2144Add port traffic management WRED profile
2145~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2146
2147Create a new WRED profile::
2148
2149   testpmd> add port tm node wred profile (port_id) (wred_profile_id) \
2150   (color_g) (min_th_g) (max_th_g) (maxp_inv_g) (wq_log2_g) \
2151   (color_y) (min_th_y) (max_th_y) (maxp_inv_y) (wq_log2_y) \
2152   (color_r) (min_th_r) (max_th_r) (maxp_inv_r) (wq_log2_r)
2153
2154where:
2155
2156* ``wred_profile id``: Identifier for the newly create WRED profile
2157* ``color_g``: Packet color (green)
2158* ``min_th_g``: Minimum queue threshold for packet with green color
2159* ``max_th_g``: Minimum queue threshold for packet with green color
2160* ``maxp_inv_g``: Inverse of packet marking probability maximum value (maxp)
2161* ``wq_log2_g``: Negated log2 of queue weight (wq)
2162* ``color_y``: Packet color (yellow)
2163* ``min_th_y``: Minimum queue threshold for packet with yellow color
2164* ``max_th_y``: Minimum queue threshold for packet with yellow color
2165* ``maxp_inv_y``: Inverse of packet marking probability maximum value (maxp)
2166* ``wq_log2_y``: Negated log2 of queue weight (wq)
2167* ``color_r``: Packet color (red)
2168* ``min_th_r``: Minimum queue threshold for packet with yellow color
2169* ``max_th_r``: Minimum queue threshold for packet with yellow color
2170* ``maxp_inv_r``: Inverse of packet marking probability maximum value (maxp)
2171* ``wq_log2_r``: Negated log2 of queue weight (wq)
2172
2173Delete port traffic management WRED profile
2174~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2175
2176Delete the WRED profile::
2177
2178   testpmd> del port tm node wred profile (port_id) (wred_profile_id)
2179
2180Add port traffic management hierarchy nonleaf node
2181~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2182
2183Add nonleaf node to port traffic management hiearchy::
2184
2185   testpmd> add port tm nonleaf node (port_id) (node_id) (parent_node_id) \
2186   (priority) (weight) (level_id) (shaper_profile_id) \
2187   (shared_shaper_id) (n_shared_shapers) (n_sp_priorities) (stats_mask) \
2188
2189where:
2190
2191* ``parent_node_id``: Node ID of the parent.
2192* ``priority``: Node priority (highest node priority is zero). This is used by
2193  the SP algorithm running on the parent node for scheduling this node.
2194* ``weight``: Node weight (lowest weight is one). The node weight is relative
2195  to the weight sum of all siblings that have the same priority. It is used by
2196  the WFQ algorithm running on the parent node for scheduling this node.
2197* ``level_id``: Hiearchy level of the node.
2198* ``shaper_profile_id``: Shaper profile ID of the private shaper to be used by
2199  the node.
2200* ``shared_shaper_id``: Shared shaper id.
2201* ``n_shared_shapers``: Number of shared shapers.
2202* ``n_sp_priorities``: Number of strict priorities.
2203* ``stats_mask``: Mask of statistics counter types to be enabled for this node.
2204
2205Add port traffic management hierarchy leaf node
2206~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2207
2208Add leaf node to port traffic management hiearchy::
2209
2210   testpmd> add port tm leaf node (port_id) (node_id) (parent_node_id) \
2211   (priority) (weight) (level_id) (shaper_profile_id) \
2212   (shared_shaper_id) (n_shared_shapers) (cman_mode) \
2213   (wred_profile_id) (stats_mask) \
2214
2215where:
2216
2217* ``parent_node_id``: Node ID of the parent.
2218* ``priority``: Node priority (highest node priority is zero). This is used by
2219  the SP algorithm running on the parent node for scheduling this node.
2220* ``weight``: Node weight (lowest weight is one). The node weight is relative
2221  to the weight sum of all siblings that have the same priority. It is used by
2222  the WFQ algorithm running on the parent node for scheduling this node.
2223* ``level_id``: Hiearchy level of the node.
2224* ``shaper_profile_id``: Shaper profile ID of the private shaper to be used by
2225  the node.
2226* ``shared_shaper_id``: Shared shaper id.
2227* ``n_shared_shapers``: Number of shared shapers.
2228* ``cman_mode``: Congestion management mode to be enabled for this node.
2229* ``wred_profile_id``: WRED profile id to be enabled for this node.
2230* ``stats_mask``: Mask of statistics counter types to be enabled for this node.
2231
2232Delete port traffic management hierarchy node
2233~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2234
2235Delete node from port traffic management hiearchy::
2236
2237   testpmd> del port tm node (port_id) (node_id)
2238
2239Update port traffic management hierarchy parent node
2240~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2241
2242Update port traffic management hierarchy parent node::
2243
2244   testpmd> set port tm node parent (port_id) (node_id) (parent_node_id) \
2245   (priority) (weight)
2246
2247This function can only be called after the hierarchy commit invocation. Its
2248success depends on the port support for this operation, as advertised through
2249the port capability set. This function is valid for all nodes of the traffic
2250management hierarchy except root node.
2251
2252Commit port traffic management hierarchy
2253~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2254
2255Commit the traffic management hierarchy on the port::
2256
2257   testpmd> port tm hierarchy commit (port_id) (clean_on_fail)
2258
2259where:
2260
2261* ``clean_on_fail``: When set to non-zero, hierarchy is cleared on function
2262  call failure. On the other hand, hierarchy is preserved when this parameter
2263  is equal to zero.
2264
2265Set port traffic management default hierarchy (tm forwarding mode)
2266~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2267
2268set the traffic management default hierarchy on the port::
2269
2270   testpmd> set port tm hierarchy default (port_id)
2271
2272Filter Functions
2273----------------
2274
2275This section details the available filter functions that are available.
2276
2277Note these functions interface the deprecated legacy filtering framework,
2278superseded by *rte_flow*. See `Flow rules management`_.
2279
2280ethertype_filter
2281~~~~~~~~~~~~~~~~~~~~
2282
2283Add or delete a L2 Ethertype filter, which identify packets by their L2 Ethertype mainly assign them to a receive queue::
2284
2285   ethertype_filter (port_id) (add|del) (mac_addr|mac_ignr) (mac_address) \
2286                    ethertype (ether_type) (drop|fwd) queue (queue_id)
2287
2288The available information parameters are:
2289
2290* ``port_id``: The port which the Ethertype filter assigned on.
2291
2292* ``mac_addr``: Compare destination mac address.
2293
2294* ``mac_ignr``: Ignore destination mac address match.
2295
2296* ``mac_address``: Destination mac address to match.
2297
2298* ``ether_type``: The EtherType value want to match,
2299  for example 0x0806 for ARP packet. 0x0800 (IPv4) and 0x86DD (IPv6) are invalid.
2300
2301* ``queue_id``: The receive queue associated with this EtherType filter.
2302  It is meaningless when deleting or dropping.
2303
2304Example, to add/remove an ethertype filter rule::
2305
2306   testpmd> ethertype_filter 0 add mac_ignr 00:11:22:33:44:55 \
2307                             ethertype 0x0806 fwd queue 3
2308
2309   testpmd> ethertype_filter 0 del mac_ignr 00:11:22:33:44:55 \
2310                             ethertype 0x0806 fwd queue 3
2311
23122tuple_filter
2313~~~~~~~~~~~~~~~~~
2314
2315Add or delete a 2-tuple filter,
2316which identifies packets by specific protocol and destination TCP/UDP port
2317and forwards packets into one of the receive queues::
2318
2319   2tuple_filter (port_id) (add|del) dst_port (dst_port_value) \
2320                 protocol (protocol_value) mask (mask_value) \
2321                 tcp_flags (tcp_flags_value) priority (prio_value) \
2322                 queue (queue_id)
2323
2324The available information parameters are:
2325
2326* ``port_id``: The port which the 2-tuple filter assigned on.
2327
2328* ``dst_port_value``: Destination port in L4.
2329
2330* ``protocol_value``: IP L4 protocol.
2331
2332* ``mask_value``: Participates in the match or not by bit for field above, 1b means participate.
2333
2334* ``tcp_flags_value``: TCP control bits. The non-zero value is invalid, when the pro_value is not set to 0x06 (TCP).
2335
2336* ``prio_value``: Priority of this filter.
2337
2338* ``queue_id``: The receive queue associated with this 2-tuple filter.
2339
2340Example, to add/remove an 2tuple filter rule::
2341
2342   testpmd> 2tuple_filter 0 add dst_port 32 protocol 0x06 mask 0x03 \
2343                          tcp_flags 0x02 priority 3 queue 3
2344
2345   testpmd> 2tuple_filter 0 del dst_port 32 protocol 0x06 mask 0x03 \
2346                          tcp_flags 0x02 priority 3 queue 3
2347
23485tuple_filter
2349~~~~~~~~~~~~~~~~~
2350
2351Add or delete a 5-tuple filter,
2352which consists of a 5-tuple (protocol, source and destination IP addresses, source and destination TCP/UDP/SCTP port)
2353and routes packets into one of the receive queues::
2354
2355   5tuple_filter (port_id) (add|del) dst_ip (dst_address) src_ip \
2356                 (src_address) dst_port (dst_port_value) \
2357                 src_port (src_port_value) protocol (protocol_value) \
2358                 mask (mask_value) tcp_flags (tcp_flags_value) \
2359                 priority (prio_value) queue (queue_id)
2360
2361The available information parameters are:
2362
2363* ``port_id``: The port which the 5-tuple filter assigned on.
2364
2365* ``dst_address``: Destination IP address.
2366
2367* ``src_address``: Source IP address.
2368
2369* ``dst_port_value``: TCP/UDP destination port.
2370
2371* ``src_port_value``: TCP/UDP source port.
2372
2373* ``protocol_value``: L4 protocol.
2374
2375* ``mask_value``: Participates in the match or not by bit for field above, 1b means participate
2376
2377* ``tcp_flags_value``: TCP control bits. The non-zero value is invalid, when the protocol_value is not set to 0x06 (TCP).
2378
2379* ``prio_value``: The priority of this filter.
2380
2381* ``queue_id``: The receive queue associated with this 5-tuple filter.
2382
2383Example, to add/remove an 5tuple filter rule::
2384
2385   testpmd> 5tuple_filter 0 add dst_ip 2.2.2.5 src_ip 2.2.2.4 \
2386            dst_port 64 src_port 32 protocol 0x06 mask 0x1F \
2387            flags 0x0 priority 3 queue 3
2388
2389   testpmd> 5tuple_filter 0 del dst_ip 2.2.2.5 src_ip 2.2.2.4 \
2390            dst_port 64 src_port 32 protocol 0x06 mask 0x1F \
2391            flags 0x0 priority 3 queue 3
2392
2393syn_filter
2394~~~~~~~~~~
2395
2396Using the  SYN filter, TCP packets whose *SYN* flag is set can be forwarded to a separate queue::
2397
2398   syn_filter (port_id) (add|del) priority (high|low) queue (queue_id)
2399
2400The available information parameters are:
2401
2402* ``port_id``: The port which the SYN filter assigned on.
2403
2404* ``high``: This SYN filter has higher priority than other filters.
2405
2406* ``low``: This SYN filter has lower priority than other filters.
2407
2408* ``queue_id``: The receive queue associated with this SYN filter
2409
2410Example::
2411
2412   testpmd> syn_filter 0 add priority high queue 3
2413
2414flex_filter
2415~~~~~~~~~~~
2416
2417With flex filter, packets can be recognized by any arbitrary pattern within the first 128 bytes of the packet
2418and routed into one of the receive queues::
2419
2420   flex_filter (port_id) (add|del) len (len_value) bytes (bytes_value) \
2421               mask (mask_value) priority (prio_value) queue (queue_id)
2422
2423The available information parameters are:
2424
2425* ``port_id``: The port which the Flex filter is assigned on.
2426
2427* ``len_value``: Filter length in bytes, no greater than 128.
2428
2429* ``bytes_value``: A string in hexadecimal, means the value the flex filter needs to match.
2430
2431* ``mask_value``: A string in hexadecimal, bit 1 means corresponding byte participates in the match.
2432
2433* ``prio_value``: The priority of this filter.
2434
2435* ``queue_id``: The receive queue associated with this Flex filter.
2436
2437Example::
2438
2439   testpmd> flex_filter 0 add len 16 bytes 0x00000000000000000000000008060000 \
2440                          mask 000C priority 3 queue 3
2441
2442   testpmd> flex_filter 0 del len 16 bytes 0x00000000000000000000000008060000 \
2443                          mask 000C priority 3 queue 3
2444
2445
2446.. _testpmd_flow_director:
2447
2448flow_director_filter
2449~~~~~~~~~~~~~~~~~~~~
2450
2451The Flow Director works in receive mode to identify specific flows or sets of flows and route them to specific queues.
2452
2453Four types of filtering are supported which are referred to as Perfect Match, Signature, Perfect-mac-vlan and
2454Perfect-tunnel filters, the match mode is set by the ``--pkt-filter-mode`` command-line parameter:
2455
2456* Perfect match filters.
2457  The hardware checks a match between the masked fields of the received packets and the programmed filters.
2458  The masked fields are for IP flow.
2459
2460* Signature filters.
2461  The hardware checks a match between a hash-based signature of the masked fields of the received packet.
2462
2463* Perfect-mac-vlan match filters.
2464  The hardware checks a match between the masked fields of the received packets and the programmed filters.
2465  The masked fields are for MAC VLAN flow.
2466
2467* Perfect-tunnel match filters.
2468  The hardware checks a match between the masked fields of the received packets and the programmed filters.
2469  The masked fields are for tunnel flow.
2470
2471The Flow Director filters can match the different fields for different type of packet: flow type, specific input set
2472per flow type and the flexible payload.
2473
2474The Flow Director can also mask out parts of all of these fields so that filters
2475are only applied to certain fields or parts of the fields.
2476
2477Different NICs may have different capabilities, command show port fdir (port_id) can be used to acquire the information.
2478
2479# Commands to add flow director filters of different flow types::
2480
2481   flow_director_filter (port_id) mode IP (add|del|update) \
2482                        flow (ipv4-other|ipv4-frag|ipv6-other|ipv6-frag) \
2483                        src (src_ip_address) dst (dst_ip_address) \
2484                        tos (tos_value) proto (proto_value) ttl (ttl_value) \
2485                        vlan (vlan_value) flexbytes (flexbytes_value) \
2486                        (drop|fwd) pf|vf(vf_id) queue (queue_id) \
2487                        fd_id (fd_id_value)
2488
2489   flow_director_filter (port_id) mode IP (add|del|update) \
2490                        flow (ipv4-tcp|ipv4-udp|ipv6-tcp|ipv6-udp) \
2491                        src (src_ip_address) (src_port) \
2492                        dst (dst_ip_address) (dst_port) \
2493                        tos (tos_value) ttl (ttl_value) \
2494                        vlan (vlan_value) flexbytes (flexbytes_value) \
2495                        (drop|fwd) queue pf|vf(vf_id) (queue_id) \
2496                        fd_id (fd_id_value)
2497
2498   flow_director_filter (port_id) mode IP (add|del|update) \
2499                        flow (ipv4-sctp|ipv6-sctp) \
2500                        src (src_ip_address) (src_port) \
2501                        dst (dst_ip_address) (dst_port) \
2502                        tos (tos_value) ttl (ttl_value) \
2503                        tag (verification_tag) vlan (vlan_value) \
2504                        flexbytes (flexbytes_value) (drop|fwd) \
2505                        pf|vf(vf_id) queue (queue_id) fd_id (fd_id_value)
2506
2507   flow_director_filter (port_id) mode IP (add|del|update) flow l2_payload \
2508                        ether (ethertype) flexbytes (flexbytes_value) \
2509                        (drop|fwd) pf|vf(vf_id) queue (queue_id)
2510                        fd_id (fd_id_value)
2511
2512   flow_director_filter (port_id) mode MAC-VLAN (add|del|update) \
2513                        mac (mac_address) vlan (vlan_value) \
2514                        flexbytes (flexbytes_value) (drop|fwd) \
2515                        queue (queue_id) fd_id (fd_id_value)
2516
2517   flow_director_filter (port_id) mode Tunnel (add|del|update) \
2518                        mac (mac_address) vlan (vlan_value) \
2519                        tunnel (NVGRE|VxLAN) tunnel-id (tunnel_id_value) \
2520                        flexbytes (flexbytes_value) (drop|fwd) \
2521                        queue (queue_id) fd_id (fd_id_value)
2522
2523For example, to add an ipv4-udp flow type filter::
2524
2525   testpmd> flow_director_filter 0 mode IP add flow ipv4-udp src 2.2.2.3 32 \
2526            dst 2.2.2.5 33 tos 2 ttl 40 vlan 0x1 flexbytes (0x88,0x48) \
2527            fwd pf queue 1 fd_id 1
2528
2529For example, add an ipv4-other flow type filter::
2530
2531   testpmd> flow_director_filter 0 mode IP add flow ipv4-other src 2.2.2.3 \
2532             dst 2.2.2.5 tos 2 proto 20 ttl 40 vlan 0x1 \
2533             flexbytes (0x88,0x48) fwd pf queue 1 fd_id 1
2534
2535flush_flow_director
2536~~~~~~~~~~~~~~~~~~~
2537
2538Flush all flow director filters on a device::
2539
2540   testpmd> flush_flow_director (port_id)
2541
2542Example, to flush all flow director filter on port 0::
2543
2544   testpmd> flush_flow_director 0
2545
2546flow_director_mask
2547~~~~~~~~~~~~~~~~~~
2548
2549Set flow director's input masks::
2550
2551   flow_director_mask (port_id) mode IP vlan (vlan_value) \
2552                      src_mask (ipv4_src) (ipv6_src) (src_port) \
2553                      dst_mask (ipv4_dst) (ipv6_dst) (dst_port)
2554
2555   flow_director_mask (port_id) mode MAC-VLAN vlan (vlan_value)
2556
2557   flow_director_mask (port_id) mode Tunnel vlan (vlan_value) \
2558                      mac (mac_value) tunnel-type (tunnel_type_value) \
2559                      tunnel-id (tunnel_id_value)
2560
2561Example, to set flow director mask on port 0::
2562
2563   testpmd> flow_director_mask 0 mode IP vlan 0xefff \
2564            src_mask 255.255.255.255 \
2565                FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF \
2566            dst_mask 255.255.255.255 \
2567                FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF
2568
2569flow_director_flex_mask
2570~~~~~~~~~~~~~~~~~~~~~~~
2571
2572set masks of flow director's flexible payload based on certain flow type::
2573
2574   testpmd> flow_director_flex_mask (port_id) \
2575            flow (none|ipv4-other|ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
2576                  ipv6-other|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp| \
2577                  l2_payload|all) (mask)
2578
2579Example, to set flow director's flex mask for all flow type on port 0::
2580
2581   testpmd> flow_director_flex_mask 0 flow all \
2582            (0xff,0xff,0,0,0,0,0,0,0,0,0,0,0,0,0,0)
2583
2584
2585flow_director_flex_payload
2586~~~~~~~~~~~~~~~~~~~~~~~~~~
2587
2588Configure flexible payload selection::
2589
2590   flow_director_flex_payload (port_id) (raw|l2|l3|l4) (config)
2591
2592For example, to select the first 16 bytes from the offset 4 (bytes) of packet's payload as flexible payload::
2593
2594   testpmd> flow_director_flex_payload 0 l4 \
2595            (4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19)
2596
2597get_sym_hash_ena_per_port
2598~~~~~~~~~~~~~~~~~~~~~~~~~
2599
2600Get symmetric hash enable configuration per port::
2601
2602   get_sym_hash_ena_per_port (port_id)
2603
2604For example, to get symmetric hash enable configuration of port 1::
2605
2606   testpmd> get_sym_hash_ena_per_port 1
2607
2608set_sym_hash_ena_per_port
2609~~~~~~~~~~~~~~~~~~~~~~~~~
2610
2611Set symmetric hash enable configuration per port to enable or disable::
2612
2613   set_sym_hash_ena_per_port (port_id) (enable|disable)
2614
2615For example, to set symmetric hash enable configuration of port 1 to enable::
2616
2617   testpmd> set_sym_hash_ena_per_port 1 enable
2618
2619get_hash_global_config
2620~~~~~~~~~~~~~~~~~~~~~~
2621
2622Get the global configurations of hash filters::
2623
2624   get_hash_global_config (port_id)
2625
2626For example, to get the global configurations of hash filters of port 1::
2627
2628   testpmd> get_hash_global_config 1
2629
2630set_hash_global_config
2631~~~~~~~~~~~~~~~~~~~~~~
2632
2633Set the global configurations of hash filters::
2634
2635   set_hash_global_config (port_id) (toeplitz|simple_xor|default) \
2636   (ipv4|ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp|ipv4-other|ipv6|ipv6-frag| \
2637   ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other|l2_payload) \
2638   (enable|disable)
2639
2640For example, to enable simple_xor for flow type of ipv6 on port 2::
2641
2642   testpmd> set_hash_global_config 2 simple_xor ipv6 enable
2643
2644set_hash_input_set
2645~~~~~~~~~~~~~~~~~~
2646
2647Set the input set for hash::
2648
2649   set_hash_input_set (port_id) (ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
2650   ipv4-other|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other| \
2651   l2_payload) (ovlan|ivlan|src-ipv4|dst-ipv4|src-ipv6|dst-ipv6|ipv4-tos| \
2652   ipv4-proto|ipv6-tc|ipv6-next-header|udp-src-port|udp-dst-port| \
2653   tcp-src-port|tcp-dst-port|sctp-src-port|sctp-dst-port|sctp-veri-tag| \
2654   udp-key|gre-key|fld-1st|fld-2nd|fld-3rd|fld-4th|fld-5th|fld-6th|fld-7th| \
2655   fld-8th|none) (select|add)
2656
2657For example, to add source IP to hash input set for flow type of ipv4-udp on port 0::
2658
2659   testpmd> set_hash_input_set 0 ipv4-udp src-ipv4 add
2660
2661set_fdir_input_set
2662~~~~~~~~~~~~~~~~~~
2663
2664The Flow Director filters can match the different fields for different type of packet, i.e. specific input set
2665on per flow type and the flexible payload. This command can be used to change input set for each flow type.
2666
2667Set the input set for flow director::
2668
2669   set_fdir_input_set (port_id) (ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
2670   ipv4-other|ipv6|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other| \
2671   l2_payload) (ivlan|ethertype|src-ipv4|dst-ipv4|src-ipv6|dst-ipv6|ipv4-tos| \
2672   ipv4-proto|ipv4-ttl|ipv6-tc|ipv6-next-header|ipv6-hop-limits| \
2673   tudp-src-port|udp-dst-port|cp-src-port|tcp-dst-port|sctp-src-port| \
2674   sctp-dst-port|sctp-veri-tag|none) (select|add)
2675
2676For example to add source IP to FD input set for flow type of ipv4-udp on port 0::
2677
2678   testpmd> set_fdir_input_set 0 ipv4-udp src-ipv4 add
2679
2680global_config
2681~~~~~~~~~~~~~
2682
2683Set different GRE key length for input set::
2684
2685   global_config (port_id) gre-key-len (number in bytes)
2686
2687For example to set GRE key length for input set to 4 bytes on port 0::
2688
2689   testpmd> global_config 0 gre-key-len 4
2690
2691
2692.. _testpmd_rte_flow:
2693
2694Flow rules management
2695---------------------
2696
2697Control of the generic flow API (*rte_flow*) is fully exposed through the
2698``flow`` command (validation, creation, destruction, queries and operation
2699modes).
2700
2701Considering *rte_flow* overlaps with all `Filter Functions`_, using both
2702features simultaneously may cause undefined side-effects and is therefore
2703not recommended.
2704
2705``flow`` syntax
2706~~~~~~~~~~~~~~~
2707
2708Because the ``flow`` command uses dynamic tokens to handle the large number
2709of possible flow rules combinations, its behavior differs slightly from
2710other commands, in particular:
2711
2712- Pressing *?* or the *<tab>* key displays contextual help for the current
2713  token, not that of the entire command.
2714
2715- Optional and repeated parameters are supported (provided they are listed
2716  in the contextual help).
2717
2718The first parameter stands for the operation mode. Possible operations and
2719their general syntax are described below. They are covered in detail in the
2720following sections.
2721
2722- Check whether a flow rule can be created::
2723
2724   flow validate {port_id}
2725       [group {group_id}] [priority {level}] [ingress] [egress]
2726       pattern {item} [/ {item} [...]] / end
2727       actions {action} [/ {action} [...]] / end
2728
2729- Create a flow rule::
2730
2731   flow create {port_id}
2732       [group {group_id}] [priority {level}] [ingress] [egress]
2733       pattern {item} [/ {item} [...]] / end
2734       actions {action} [/ {action} [...]] / end
2735
2736- Destroy specific flow rules::
2737
2738   flow destroy {port_id} rule {rule_id} [...]
2739
2740- Destroy all flow rules::
2741
2742   flow flush {port_id}
2743
2744- Query an existing flow rule::
2745
2746   flow query {port_id} {rule_id} {action}
2747
2748- List existing flow rules sorted by priority, filtered by group
2749  identifiers::
2750
2751   flow list {port_id} [group {group_id}] [...]
2752
2753- Restrict ingress traffic to the defined flow rules::
2754
2755   flow isolate {port_id} {boolean}
2756
2757Validating flow rules
2758~~~~~~~~~~~~~~~~~~~~~
2759
2760``flow validate`` reports whether a flow rule would be accepted by the
2761underlying device in its current state but stops short of creating it. It is
2762bound to ``rte_flow_validate()``::
2763
2764   flow validate {port_id}
2765      [group {group_id}] [priority {level}] [ingress] [egress]
2766      pattern {item} [/ {item} [...]] / end
2767      actions {action} [/ {action} [...]] / end
2768
2769If successful, it will show::
2770
2771   Flow rule validated
2772
2773Otherwise it will show an error message of the form::
2774
2775   Caught error type [...] ([...]): [...]
2776
2777This command uses the same parameters as ``flow create``, their format is
2778described in `Creating flow rules`_.
2779
2780Check whether redirecting any Ethernet packet received on port 0 to RX queue
2781index 6 is supported::
2782
2783   testpmd> flow validate 0 ingress pattern eth / end
2784      actions queue index 6 / end
2785   Flow rule validated
2786   testpmd>
2787
2788Port 0 does not support TCPv6 rules::
2789
2790   testpmd> flow validate 0 ingress pattern eth / ipv6 / tcp / end
2791      actions drop / end
2792   Caught error type 9 (specific pattern item): Invalid argument
2793   testpmd>
2794
2795Creating flow rules
2796~~~~~~~~~~~~~~~~~~~
2797
2798``flow create`` validates and creates the specified flow rule. It is bound
2799to ``rte_flow_create()``::
2800
2801   flow create {port_id}
2802      [group {group_id}] [priority {level}] [ingress] [egress]
2803      pattern {item} [/ {item} [...]] / end
2804      actions {action} [/ {action} [...]] / end
2805
2806If successful, it will return a flow rule ID usable with other commands::
2807
2808   Flow rule #[...] created
2809
2810Otherwise it will show an error message of the form::
2811
2812   Caught error type [...] ([...]): [...]
2813
2814Parameters describe in the following order:
2815
2816- Attributes (*group*, *priority*, *ingress*, *egress* tokens).
2817- A matching pattern, starting with the *pattern* token and terminated by an
2818  *end* pattern item.
2819- Actions, starting with the *actions* token and terminated by an *end*
2820  action.
2821
2822These translate directly to *rte_flow* objects provided as-is to the
2823underlying functions.
2824
2825The shortest valid definition only comprises mandatory tokens::
2826
2827   testpmd> flow create 0 pattern end actions end
2828
2829Note that PMDs may refuse rules that essentially do nothing such as this
2830one.
2831
2832**All unspecified object values are automatically initialized to 0.**
2833
2834Attributes
2835^^^^^^^^^^
2836
2837These tokens affect flow rule attributes (``struct rte_flow_attr``) and are
2838specified before the ``pattern`` token.
2839
2840- ``group {group id}``: priority group.
2841- ``priority {level}``: priority level within group.
2842- ``ingress``: rule applies to ingress traffic.
2843- ``egress``: rule applies to egress traffic.
2844
2845Each instance of an attribute specified several times overrides the previous
2846value as shown below (group 4 is used)::
2847
2848   testpmd> flow create 0 group 42 group 24 group 4 [...]
2849
2850Note that once enabled, ``ingress`` and ``egress`` cannot be disabled.
2851
2852While not specifying a direction is an error, some rules may allow both
2853simultaneously.
2854
2855Most rules affect RX therefore contain the ``ingress`` token::
2856
2857   testpmd> flow create 0 ingress pattern [...]
2858
2859Matching pattern
2860^^^^^^^^^^^^^^^^
2861
2862A matching pattern starts after the ``pattern`` token. It is made of pattern
2863items and is terminated by a mandatory ``end`` item.
2864
2865Items are named after their type (*RTE_FLOW_ITEM_TYPE_* from ``enum
2866rte_flow_item_type``).
2867
2868The ``/`` token is used as a separator between pattern items as shown
2869below::
2870
2871   testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end [...]
2872
2873Note that protocol items like these must be stacked from lowest to highest
2874layer to make sense. For instance, the following rule is either invalid or
2875unlikely to match any packet::
2876
2877   testpmd> flow create 0 ingress pattern eth / udp / ipv4 / end [...]
2878
2879More information on these restrictions can be found in the *rte_flow*
2880documentation.
2881
2882Several items support additional specification structures, for example
2883``ipv4`` allows specifying source and destination addresses as follows::
2884
2885   testpmd> flow create 0 ingress pattern eth / ipv4 src is 10.1.1.1
2886      dst is 10.2.0.0 / end [...]
2887
2888This rule matches all IPv4 traffic with the specified properties.
2889
2890In this example, ``src`` and ``dst`` are field names of the underlying
2891``struct rte_flow_item_ipv4`` object. All item properties can be specified
2892in a similar fashion.
2893
2894The ``is`` token means that the subsequent value must be matched exactly,
2895and assigns ``spec`` and ``mask`` fields in ``struct rte_flow_item``
2896accordingly. Possible assignment tokens are:
2897
2898- ``is``: match value perfectly (with full bit-mask).
2899- ``spec``: match value according to configured bit-mask.
2900- ``last``: specify upper bound to establish a range.
2901- ``mask``: specify bit-mask with relevant bits set to one.
2902- ``prefix``: generate bit-mask from a prefix length.
2903
2904These yield identical results::
2905
2906   ipv4 src is 10.1.1.1
2907
2908::
2909
2910   ipv4 src spec 10.1.1.1 src mask 255.255.255.255
2911
2912::
2913
2914   ipv4 src spec 10.1.1.1 src prefix 32
2915
2916::
2917
2918   ipv4 src is 10.1.1.1 src last 10.1.1.1 # range with a single value
2919
2920::
2921
2922   ipv4 src is 10.1.1.1 src last 0 # 0 disables range
2923
2924Inclusive ranges can be defined with ``last``::
2925
2926   ipv4 src is 10.1.1.1 src last 10.2.3.4 # 10.1.1.1 to 10.2.3.4
2927
2928Note that ``mask`` affects both ``spec`` and ``last``::
2929
2930   ipv4 src is 10.1.1.1 src last 10.2.3.4 src mask 255.255.0.0
2931      # matches 10.1.0.0 to 10.2.255.255
2932
2933Properties can be modified multiple times::
2934
2935   ipv4 src is 10.1.1.1 src is 10.1.2.3 src is 10.2.3.4 # matches 10.2.3.4
2936
2937::
2938
2939   ipv4 src is 10.1.1.1 src prefix 24 src prefix 16 # matches 10.1.0.0/16
2940
2941Pattern items
2942^^^^^^^^^^^^^
2943
2944This section lists supported pattern items and their attributes, if any.
2945
2946- ``end``: end list of pattern items.
2947
2948- ``void``: no-op pattern item.
2949
2950- ``invert``: perform actions when pattern does not match.
2951
2952- ``any``: match any protocol for the current layer.
2953
2954  - ``num {unsigned}``: number of layers covered.
2955
2956- ``pf``: match packets addressed to the physical function.
2957
2958- ``vf``: match packets addressed to a virtual function ID.
2959
2960  - ``id {unsigned}``: destination VF ID.
2961
2962- ``port``: device-specific physical port index to use.
2963
2964  - ``index {unsigned}``: physical port index.
2965
2966- ``raw``: match an arbitrary byte string.
2967
2968  - ``relative {boolean}``: look for pattern after the previous item.
2969  - ``search {boolean}``: search pattern from offset (see also limit).
2970  - ``offset {integer}``: absolute or relative offset for pattern.
2971  - ``limit {unsigned}``: search area limit for start of pattern.
2972  - ``pattern {string}``: byte string to look for.
2973
2974- ``eth``: match Ethernet header.
2975
2976  - ``dst {MAC-48}``: destination MAC.
2977  - ``src {MAC-48}``: source MAC.
2978  - ``type {unsigned}``: EtherType.
2979
2980- ``vlan``: match 802.1Q/ad VLAN tag.
2981
2982  - ``tpid {unsigned}``: tag protocol identifier.
2983  - ``tci {unsigned}``: tag control information.
2984  - ``pcp {unsigned}``: priority code point.
2985  - ``dei {unsigned}``: drop eligible indicator.
2986  - ``vid {unsigned}``: VLAN identifier.
2987
2988- ``ipv4``: match IPv4 header.
2989
2990  - ``tos {unsigned}``: type of service.
2991  - ``ttl {unsigned}``: time to live.
2992  - ``proto {unsigned}``: next protocol ID.
2993  - ``src {ipv4 address}``: source address.
2994  - ``dst {ipv4 address}``: destination address.
2995
2996- ``ipv6``: match IPv6 header.
2997
2998  - ``tc {unsigned}``: traffic class.
2999  - ``flow {unsigned}``: flow label.
3000  - ``proto {unsigned}``: protocol (next header).
3001  - ``hop {unsigned}``: hop limit.
3002  - ``src {ipv6 address}``: source address.
3003  - ``dst {ipv6 address}``: destination address.
3004
3005- ``icmp``: match ICMP header.
3006
3007  - ``type {unsigned}``: ICMP packet type.
3008  - ``code {unsigned}``: ICMP packet code.
3009
3010- ``udp``: match UDP header.
3011
3012  - ``src {unsigned}``: UDP source port.
3013  - ``dst {unsigned}``: UDP destination port.
3014
3015- ``tcp``: match TCP header.
3016
3017  - ``src {unsigned}``: TCP source port.
3018  - ``dst {unsigned}``: TCP destination port.
3019
3020- ``sctp``: match SCTP header.
3021
3022  - ``src {unsigned}``: SCTP source port.
3023  - ``dst {unsigned}``: SCTP destination port.
3024  - ``tag {unsigned}``: validation tag.
3025  - ``cksum {unsigned}``: checksum.
3026
3027- ``vxlan``: match VXLAN header.
3028
3029  - ``vni {unsigned}``: VXLAN identifier.
3030
3031- ``e_tag``: match IEEE 802.1BR E-Tag header.
3032
3033  - ``grp_ecid_b {unsigned}``: GRP and E-CID base.
3034
3035- ``nvgre``: match NVGRE header.
3036
3037  - ``tni {unsigned}``: virtual subnet ID.
3038
3039- ``mpls``: match MPLS header.
3040
3041  - ``label {unsigned}``: MPLS label.
3042
3043- ``gre``: match GRE header.
3044
3045  - ``protocol {unsigned}``: protocol type.
3046
3047- ``fuzzy``: fuzzy pattern match, expect faster than default.
3048
3049  - ``thresh {unsigned}``: accuracy threshold.
3050
3051- ``gtp``, ``gtpc``, ``gtpu``: match GTPv1 header.
3052
3053  - ``teid {unsigned}``: tunnel endpoint identifier.
3054
3055Actions list
3056^^^^^^^^^^^^
3057
3058A list of actions starts after the ``actions`` token in the same fashion as
3059`Matching pattern`_; actions are separated by ``/`` tokens and the list is
3060terminated by a mandatory ``end`` action.
3061
3062Actions are named after their type (*RTE_FLOW_ACTION_TYPE_* from ``enum
3063rte_flow_action_type``).
3064
3065Dropping all incoming UDPv4 packets can be expressed as follows::
3066
3067   testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
3068      actions drop / end
3069
3070Several actions have configurable properties which must be specified when
3071there is no valid default value. For example, ``queue`` requires a target
3072queue index.
3073
3074This rule redirects incoming UDPv4 traffic to queue index 6::
3075
3076   testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
3077      actions queue index 6 / end
3078
3079While this one could be rejected by PMDs (unspecified queue index)::
3080
3081   testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
3082      actions queue / end
3083
3084As defined by *rte_flow*, the list is not ordered, all actions of a given
3085rule are performed simultaneously. These are equivalent::
3086
3087   queue index 6 / void / mark id 42 / end
3088
3089::
3090
3091   void / mark id 42 / queue index 6 / end
3092
3093All actions in a list should have different types, otherwise only the last
3094action of a given type is taken into account::
3095
3096   queue index 4 / queue index 5 / queue index 6 / end # will use queue 6
3097
3098::
3099
3100   drop / drop / drop / end # drop is performed only once
3101
3102::
3103
3104   mark id 42 / queue index 3 / mark id 24 / end # mark will be 24
3105
3106Considering they are performed simultaneously, opposite and overlapping
3107actions can sometimes be combined when the end result is unambiguous::
3108
3109   drop / queue index 6 / end # drop has no effect
3110
3111::
3112
3113   drop / dup index 6 / end # same as above
3114
3115::
3116
3117   queue index 6 / rss queues 6 7 8 / end # queue has no effect
3118
3119::
3120
3121   drop / passthru / end # drop has no effect
3122
3123Note that PMDs may still refuse such combinations.
3124
3125Actions
3126^^^^^^^
3127
3128This section lists supported actions and their attributes, if any.
3129
3130- ``end``: end list of actions.
3131
3132- ``void``: no-op action.
3133
3134- ``passthru``: let subsequent rule process matched packets.
3135
3136- ``mark``: attach 32 bit value to packets.
3137
3138  - ``id {unsigned}``: 32 bit value to return with packets.
3139
3140- ``flag``: flag packets.
3141
3142- ``queue``: assign packets to a given queue index.
3143
3144  - ``index {unsigned}``: queue index to use.
3145
3146- ``drop``: drop packets (note: passthru has priority).
3147
3148- ``count``: enable counters for this rule.
3149
3150- ``dup``: duplicate packets to a given queue index.
3151
3152  - ``index {unsigned}``: queue index to duplicate packets to.
3153
3154- ``rss``: spread packets among several queues.
3155
3156  - ``queues [{unsigned} [...]] end``: queue indices to use.
3157
3158- ``pf``: redirect packets to physical device function.
3159
3160- ``vf``: redirect packets to virtual device function.
3161
3162  - ``original {boolean}``: use original VF ID if possible.
3163  - ``id {unsigned}``: VF ID to redirect packets to.
3164
3165Destroying flow rules
3166~~~~~~~~~~~~~~~~~~~~~
3167
3168``flow destroy`` destroys one or more rules from their rule ID (as returned
3169by ``flow create``), this command calls ``rte_flow_destroy()`` as many
3170times as necessary::
3171
3172   flow destroy {port_id} rule {rule_id} [...]
3173
3174If successful, it will show::
3175
3176   Flow rule #[...] destroyed
3177
3178It does not report anything for rule IDs that do not exist. The usual error
3179message is shown when a rule cannot be destroyed::
3180
3181   Caught error type [...] ([...]): [...]
3182
3183``flow flush`` destroys all rules on a device and does not take extra
3184arguments. It is bound to ``rte_flow_flush()``::
3185
3186   flow flush {port_id}
3187
3188Any errors are reported as above.
3189
3190Creating several rules and destroying them::
3191
3192   testpmd> flow create 0 ingress pattern eth / ipv6 / end
3193      actions queue index 2 / end
3194   Flow rule #0 created
3195   testpmd> flow create 0 ingress pattern eth / ipv4 / end
3196      actions queue index 3 / end
3197   Flow rule #1 created
3198   testpmd> flow destroy 0 rule 0 rule 1
3199   Flow rule #1 destroyed
3200   Flow rule #0 destroyed
3201   testpmd>
3202
3203The same result can be achieved using ``flow flush``::
3204
3205   testpmd> flow create 0 ingress pattern eth / ipv6 / end
3206      actions queue index 2 / end
3207   Flow rule #0 created
3208   testpmd> flow create 0 ingress pattern eth / ipv4 / end
3209      actions queue index 3 / end
3210   Flow rule #1 created
3211   testpmd> flow flush 0
3212   testpmd>
3213
3214Non-existent rule IDs are ignored::
3215
3216   testpmd> flow create 0 ingress pattern eth / ipv6 / end
3217      actions queue index 2 / end
3218   Flow rule #0 created
3219   testpmd> flow create 0 ingress pattern eth / ipv4 / end
3220      actions queue index 3 / end
3221   Flow rule #1 created
3222   testpmd> flow destroy 0 rule 42 rule 10 rule 2
3223   testpmd>
3224   testpmd> flow destroy 0 rule 0
3225   Flow rule #0 destroyed
3226   testpmd>
3227
3228Querying flow rules
3229~~~~~~~~~~~~~~~~~~~
3230
3231``flow query`` queries a specific action of a flow rule having that
3232ability. Such actions collect information that can be reported using this
3233command. It is bound to ``rte_flow_query()``::
3234
3235   flow query {port_id} {rule_id} {action}
3236
3237If successful, it will display either the retrieved data for known actions
3238or the following message::
3239
3240   Cannot display result for action type [...] ([...])
3241
3242Otherwise, it will complain either that the rule does not exist or that some
3243error occurred::
3244
3245   Flow rule #[...] not found
3246
3247::
3248
3249   Caught error type [...] ([...]): [...]
3250
3251Currently only the ``count`` action is supported. This action reports the
3252number of packets that hit the flow rule and the total number of bytes. Its
3253output has the following format::
3254
3255   count:
3256    hits_set: [...] # whether "hits" contains a valid value
3257    bytes_set: [...] # whether "bytes" contains a valid value
3258    hits: [...] # number of packets
3259    bytes: [...] # number of bytes
3260
3261Querying counters for TCPv6 packets redirected to queue 6::
3262
3263   testpmd> flow create 0 ingress pattern eth / ipv6 / tcp / end
3264      actions queue index 6 / count / end
3265   Flow rule #4 created
3266   testpmd> flow query 0 4 count
3267   count:
3268    hits_set: 1
3269    bytes_set: 0
3270    hits: 386446
3271    bytes: 0
3272   testpmd>
3273
3274Listing flow rules
3275~~~~~~~~~~~~~~~~~~
3276
3277``flow list`` lists existing flow rules sorted by priority and optionally
3278filtered by group identifiers::
3279
3280   flow list {port_id} [group {group_id}] [...]
3281
3282This command only fails with the following message if the device does not
3283exist::
3284
3285   Invalid port [...]
3286
3287Output consists of a header line followed by a short description of each
3288flow rule, one per line. There is no output at all when no flow rules are
3289configured on the device::
3290
3291   ID      Group   Prio    Attr    Rule
3292   [...]   [...]   [...]   [...]   [...]
3293
3294``Attr`` column flags:
3295
3296- ``i`` for ``ingress``.
3297- ``e`` for ``egress``.
3298
3299Creating several flow rules and listing them::
3300
3301   testpmd> flow create 0 ingress pattern eth / ipv4 / end
3302      actions queue index 6 / end
3303   Flow rule #0 created
3304   testpmd> flow create 0 ingress pattern eth / ipv6 / end
3305      actions queue index 2 / end
3306   Flow rule #1 created
3307   testpmd> flow create 0 priority 5 ingress pattern eth / ipv4 / udp / end
3308      actions rss queues 6 7 8 end / end
3309   Flow rule #2 created
3310   testpmd> flow list 0
3311   ID      Group   Prio    Attr    Rule
3312   0       0       0       i-      ETH IPV4 => QUEUE
3313   1       0       0       i-      ETH IPV6 => QUEUE
3314   2       0       5       i-      ETH IPV4 UDP => RSS
3315   testpmd>
3316
3317Rules are sorted by priority (i.e. group ID first, then priority level)::
3318
3319   testpmd> flow list 1
3320   ID      Group   Prio    Attr    Rule
3321   0       0       0       i-      ETH => COUNT
3322   6       0       500     i-      ETH IPV6 TCP => DROP COUNT
3323   5       0       1000    i-      ETH IPV6 ICMP => QUEUE
3324   1       24      0       i-      ETH IPV4 UDP => QUEUE
3325   4       24      10      i-      ETH IPV4 TCP => DROP
3326   3       24      20      i-      ETH IPV4 => DROP
3327   2       24      42      i-      ETH IPV4 UDP => QUEUE
3328   7       63      0       i-      ETH IPV6 UDP VXLAN => MARK QUEUE
3329   testpmd>
3330
3331Output can be limited to specific groups::
3332
3333   testpmd> flow list 1 group 0 group 63
3334   ID      Group   Prio    Attr    Rule
3335   0       0       0       i-      ETH => COUNT
3336   6       0       500     i-      ETH IPV6 TCP => DROP COUNT
3337   5       0       1000    i-      ETH IPV6 ICMP => QUEUE
3338   7       63      0       i-      ETH IPV6 UDP VXLAN => MARK QUEUE
3339   testpmd>
3340
3341Toggling isolated mode
3342~~~~~~~~~~~~~~~~~~~~~~
3343
3344``flow isolate`` can be used to tell the underlying PMD that ingress traffic
3345must only be injected from the defined flow rules; that no default traffic
3346is expected outside those rules and the driver is free to assign more
3347resources to handle them. It is bound to ``rte_flow_isolate()``::
3348
3349 flow isolate {port_id} {boolean}
3350
3351If successful, enabling or disabling isolated mode shows either::
3352
3353 Ingress traffic on port [...]
3354    is now restricted to the defined flow rules
3355
3356Or::
3357
3358 Ingress traffic on port [...]
3359    is not restricted anymore to the defined flow rules
3360
3361Otherwise, in case of error::
3362
3363   Caught error type [...] ([...]): [...]
3364
3365Mainly due to its side effects, PMDs supporting this mode may not have the
3366ability to toggle it more than once without reinitializing affected ports
3367first (e.g. by exiting testpmd).
3368
3369Enabling isolated mode::
3370
3371 testpmd> flow isolate 0 true
3372 Ingress traffic on port 0 is now restricted to the defined flow rules
3373 testpmd>
3374
3375Disabling isolated mode::
3376
3377 testpmd> flow isolate 0 false
3378 Ingress traffic on port 0 is not restricted anymore to the defined flow rules
3379 testpmd>
3380
3381Sample QinQ flow rules
3382~~~~~~~~~~~~~~~~~~~~~~
3383
3384Before creating QinQ rule(s) the following commands should be issued to enable QinQ::
3385
3386   testpmd> port stop 0
3387   testpmd> vlan set qinq on 0
3388
3389The above command sets the inner and outer TPID's to 0x8100.
3390
3391To change the TPID's the following commands should be used::
3392
3393   testpmd> vlan set outer tpid 0xa100 0
3394   testpmd> vlan set inner tpid 0x9100 0
3395   testpmd> port start 0
3396
3397Validate and create a QinQ rule on port 0 to steer traffic to a VF queue in a VM.
3398
3399::
3400
3401   testpmd> flow validate 0 ingress pattern eth / vlan tci is 123 /
3402       vlan tci is 456 / end actions vf id 1 / queue index 0 / end
3403   Flow rule #0 validated
3404
3405   testpmd> flow create 0 ingress pattern eth / vlan tci is 4 /
3406       vlan tci is 456 / end actions vf id 123 / queue index 0 / end
3407   Flow rule #0 created
3408
3409   testpmd> flow list 0
3410   ID      Group   Prio    Attr    Rule
3411   0       0       0       i-      ETH VLAN VLAN=>VF QUEUE
3412
3413Validate and create a QinQ rule on port 0 to steer traffic to a queue on the host.
3414
3415::
3416
3417   testpmd> flow validate 0 ingress pattern eth / vlan tci is 321 /
3418        vlan tci is 654 / end actions pf / queue index 0 / end
3419   Flow rule #1 validated
3420
3421   testpmd> flow create 0 ingress pattern eth / vlan tci is 321 /
3422        vlan tci is 654 / end actions pf / queue index 1 / end
3423   Flow rule #1 created
3424
3425   testpmd> flow list 0
3426   ID      Group   Prio    Attr    Rule
3427   0       0       0       i-      ETH VLAN VLAN=>VF QUEUE
3428   1       0       0       i-      ETH VLAN VLAN=>PF QUEUE
3429