xref: /dpdk/doc/guides/testpmd_app_ug/testpmd_funcs.rst (revision 1aa94d0536fce839acb00794027158cb7d3cfa68)
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 X
55       info [Mul-choice STRING]: show|clear port info|stats|xstats|fdir|stat_qmap|dcb_tc all
56       stats [Mul-choice STRING]: show|clear port info|stats|xstats|fdir|stat_qmap|dcb_tc X
57       stats [Mul-choice STRING]: show|clear port info|stats|xstats|fdir|stat_qmap|dcb_tc 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
89Control Functions
90-----------------
91
92start
93~~~~~
94
95Start packet forwarding with current configuration::
96
97   testpmd> start
98
99start tx_first
100~~~~~~~~~~~~~~
101
102Start packet forwarding with current configuration after sending specified number of bursts of packets::
103
104   testpmd> start tx_first (""|burst_num)
105
106The default burst number is 1 when ``burst_num`` not presented.
107
108stop
109~~~~
110
111Stop packet forwarding, and display accumulated statistics::
112
113   testpmd> stop
114
115quit
116~~~~
117
118Quit to prompt::
119
120   testpmd> quit
121
122
123Display Functions
124-----------------
125
126The functions in the following sections are used to display information about the
127testpmd configuration or the NIC status.
128
129show port
130~~~~~~~~~
131
132Display information for a given port or all ports::
133
134   testpmd> show port (info|stats|xstats|fdir|stat_qmap|dcb_tc) (port_id|all)
135
136The available information categories are:
137
138* ``info``: General port information such as MAC address.
139
140* ``stats``: RX/TX statistics.
141
142* ``xstats``: RX/TX extended NIC statistics.
143
144* ``fdir``: Flow Director information and statistics.
145
146* ``stat_qmap``: Queue statistics mapping.
147
148* ``dcb_tc``: DCB information such as TC mapping.
149
150For example:
151
152.. code-block:: console
153
154   testpmd> show port info 0
155
156   ********************* Infos for port 0 *********************
157
158   MAC address: XX:XX:XX:XX:XX:XX
159   Connect to socket: 0
160   memory allocation on the socket: 0
161   Link status: up
162   Link speed: 40000 Mbps
163   Link duplex: full-duplex
164   Promiscuous mode: enabled
165   Allmulticast mode: disabled
166   Maximum number of MAC addresses: 64
167   Maximum number of MAC addresses of hash filtering: 0
168   VLAN offload:
169       strip on
170       filter on
171       qinq(extend) off
172   Redirection table size: 512
173   Supported flow types:
174     ipv4-frag
175     ipv4-tcp
176     ipv4-udp
177     ipv4-sctp
178     ipv4-other
179     ipv6-frag
180     ipv6-tcp
181     ipv6-udp
182     ipv6-sctp
183     ipv6-other
184     l2_payload
185     port
186     vxlan
187     geneve
188     nvgre
189
190show port rss reta
191~~~~~~~~~~~~~~~~~~
192
193Display the rss redirection table entry indicated by masks on port X::
194
195   testpmd> show port (port_id) rss reta (size) (mask0, mask1...)
196
197size is used to indicate the hardware supported reta size
198
199show port rss-hash
200~~~~~~~~~~~~~~~~~~
201
202Display the RSS hash functions and RSS hash key of a port::
203
204   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]
205
206clear port
207~~~~~~~~~~
208
209Clear the port statistics for a given port or for all ports::
210
211   testpmd> clear port (info|stats|xstats|fdir|stat_qmap) (port_id|all)
212
213For example::
214
215   testpmd> clear port stats all
216
217show (rxq|txq)
218~~~~~~~~~~~~~~
219
220Display information for a given port's RX/TX queue::
221
222   testpmd> show (rxq|txq) info (port_id) (queue_id)
223
224show config
225~~~~~~~~~~~
226
227Displays the configuration of the application.
228The configuration comes from the command-line, the runtime or the application defaults::
229
230   testpmd> show config (rxtx|cores|fwd|txpkts)
231
232The available information categories are:
233
234* ``rxtx``: RX/TX configuration items.
235
236* ``cores``: List of forwarding cores.
237
238* ``fwd``: Packet forwarding configuration.
239
240* ``txpkts``: Packets to TX configuration.
241
242For example:
243
244.. code-block:: console
245
246   testpmd> show config rxtx
247
248   io packet forwarding - CRC stripping disabled - packets/burst=16
249   nb forwarding cores=2 - nb forwarding ports=1
250   RX queues=1 - RX desc=128 - RX free threshold=0
251   RX threshold registers: pthresh=8 hthresh=8 wthresh=4
252   TX queues=1 - TX desc=512 - TX free threshold=0
253   TX threshold registers: pthresh=36 hthresh=0 wthresh=0
254   TX RS bit threshold=0 - TXQ flags=0x0
255
256set fwd
257~~~~~~~
258
259Set the packet forwarding mode::
260
261   testpmd> set fwd (io|mac|macswap|flowgen| \
262                     rxonly|txonly|csum|icmpecho) (""|retry)
263
264``retry`` can be specified for forwarding engines except ``rx_only``.
265
266The available information categories are:
267
268* ``io``: Forwards packets "as-is" in I/O mode.
269  This is the fastest possible forwarding operation as it does not access packets data.
270  This is the default mode.
271
272* ``mac``: Changes the source and the destination Ethernet addresses of packets before forwarding them.
273  Default application behaviour is to set source Ethernet address to that of the transmitting interface, and destination
274  address to a dummy value (set during init). The user may specify a target destination Ethernet address via the 'eth-peer' or
275  'eth-peer-configfile' command-line options. It is not currently possible to specify a specific source Ethernet address.
276
277* ``macswap``: MAC swap forwarding mode.
278  Swaps the source and the destination Ethernet addresses of packets before forwarding them.
279
280* ``flowgen``: Multi-flow generation mode.
281  Originates a number of flows (with varying destination IP addresses), and terminate receive traffic.
282
283* ``rxonly``: Receives packets but doesn't transmit them.
284
285* ``txonly``: Generates and transmits packets without receiving any.
286
287* ``csum``: Changes the checksum field with hardware or software methods depending on the offload flags on the packet.
288
289* ``icmpecho``: Receives a burst of packets, lookup for IMCP echo requests and, if any, send back ICMP echo replies.
290
291* ``ieee1588``: Demonstrate L2 IEEE1588 V2 PTP timestamping for RX and TX. Requires ``CONFIG_RTE_LIBRTE_IEEE1588=y``.
292
293Note: TX timestamping is only available in the "Full Featured" TX path. To force ``testpmd`` into this mode set ``--txqflags=0``.
294
295Example::
296
297   testpmd> set fwd rxonly
298
299   Set rxonly packet forwarding mode
300
301
302read rxd
303~~~~~~~~
304
305Display an RX descriptor for a port RX queue::
306
307   testpmd> read rxd (port_id) (queue_id) (rxd_id)
308
309For example::
310
311   testpmd> read rxd 0 0 4
312        0x0000000B - 0x001D0180 / 0x0000000B - 0x001D0180
313
314read txd
315~~~~~~~~
316
317Display a TX descriptor for a port TX queue::
318
319   testpmd> read txd (port_id) (queue_id) (txd_id)
320
321For example::
322
323   testpmd> read txd 0 0 4
324        0x00000001 - 0x24C3C440 / 0x000F0000 - 0x2330003C
325
326
327Configuration Functions
328-----------------------
329
330The testpmd application can be configured from the runtime as well as from the command-line.
331
332This section details the available configuration functions that are available.
333
334.. note::
335
336   Configuration changes only become active when forwarding is started/restarted.
337
338set default
339~~~~~~~~~~~
340
341Reset forwarding to the default configuration::
342
343   testpmd> set default
344
345set verbose
346~~~~~~~~~~~
347
348Set the debug verbosity level::
349
350   testpmd> set verbose (level)
351
352Currently the only available levels are 0 (silent except for error) and 1 (fully verbose).
353
354set nbport
355~~~~~~~~~~
356
357Set the number of ports used by the application:
358
359set nbport (num)
360
361This is equivalent to the ``--nb-ports`` command-line option.
362
363set nbcore
364~~~~~~~~~~
365
366Set the number of cores used by the application::
367
368   testpmd> set nbcore (num)
369
370This is equivalent to the ``--nb-cores`` command-line option.
371
372.. note::
373
374   The number of cores used must not be greater than number of ports used multiplied by the number of queues per port.
375
376set coremask
377~~~~~~~~~~~~
378
379Set the forwarding cores hexadecimal mask::
380
381   testpmd> set coremask (mask)
382
383This is equivalent to the ``--coremask`` command-line option.
384
385.. note::
386
387   The master lcore is reserved for command line parsing only and cannot be masked on for packet forwarding.
388
389set portmask
390~~~~~~~~~~~~
391
392Set the forwarding ports hexadecimal mask::
393
394   testpmd> set portmask (mask)
395
396This is equivalent to the ``--portmask`` command-line option.
397
398set burst
399~~~~~~~~~
400
401Set number of packets per burst::
402
403   testpmd> set burst (num)
404
405This is equivalent to the ``--burst command-line`` option.
406
407When retry is enabled, the transmit delay time and number of retries can also be set::
408
409   testpmd> set burst tx delay (microseconds) retry (num)
410
411set txpkts
412~~~~~~~~~~
413
414Set the length of each segment of the TX-ONLY packets or length of packet for FLOWGEN mode::
415
416   testpmd> set txpkts (x[,y]*)
417
418Where x[,y]* represents a CSV list of values, without white space.
419
420set txsplit
421~~~~~~~~~~~
422
423Set the split policy for the TX packets, applicable for TX-ONLY and CSUM forwarding modes::
424
425   testpmd> set txsplit (off|on|rand)
426
427Where:
428
429* ``off`` disable packet copy & split for CSUM mode.
430
431* ``on`` split outgoing packet into multiple segments. Size of each segment
432  and number of segments per packet is determined by ``set txpkts`` command
433  (see above).
434
435* ``rand`` same as 'on', but number of segments per each packet is a random value between 1 and total number of segments.
436
437set corelist
438~~~~~~~~~~~~
439
440Set the list of forwarding cores::
441
442   testpmd> set corelist (x[,y]*)
443
444For example, to change the forwarding cores:
445
446.. code-block:: console
447
448   testpmd> set corelist 3,1
449   testpmd> show config fwd
450
451   io packet forwarding - ports=2 - cores=2 - streams=2 - NUMA support disabled
452   Logical Core 3 (socket 0) forwards packets on 1 streams:
453   RX P=0/Q=0 (socket 0) -> TX P=1/Q=0 (socket 0) peer=02:00:00:00:00:01
454   Logical Core 1 (socket 0) forwards packets on 1 streams:
455   RX P=1/Q=0 (socket 0) -> TX P=0/Q=0 (socket 0) peer=02:00:00:00:00:00
456
457.. note::
458
459   The cores are used in the same order as specified on the command line.
460
461set portlist
462~~~~~~~~~~~~
463
464Set the list of forwarding ports::
465
466   testpmd> set portlist (x[,y]*)
467
468For example, to change the port forwarding:
469
470.. code-block:: console
471
472   testpmd> set portlist 0,2,1,3
473   testpmd> show config fwd
474
475   io packet forwarding - ports=4 - cores=1 - streams=4
476   Logical Core 3 (socket 0) forwards packets on 4 streams:
477   RX P=0/Q=0 (socket 0) -> TX P=2/Q=0 (socket 0) peer=02:00:00:00:00:01
478   RX P=2/Q=0 (socket 0) -> TX P=0/Q=0 (socket 0) peer=02:00:00:00:00:00
479   RX P=1/Q=0 (socket 0) -> TX P=3/Q=0 (socket 0) peer=02:00:00:00:00:03
480   RX P=3/Q=0 (socket 0) -> TX P=1/Q=0 (socket 0) peer=02:00:00:00:00:02
481
482set tx loopback
483~~~~~~~~~~~~~~~
484
485Enable/disable tx loopback::
486
487   testpmd> set tx loopback (port_id) (on|off)
488
489set drop enable
490~~~~~~~~~~~~~~~
491
492set drop enable bit for all queues::
493
494   testpmd> set all queues drop (port_id) (on|off)
495
496set split drop enable (for VF)
497~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
498
499set split drop enable bit for VF from PF::
500
501   testpmd> set vf split drop (port_id) (vf_id) (on|off)
502
503set mac antispoof (for VF)
504~~~~~~~~~~~~~~~~~~~~~~~~~~
505
506Set mac antispoof for a VF from the PF::
507
508   testpmd> set vf mac antispoof  (port_id) (vf_id) (on|off)
509
510set macsec offload
511~~~~~~~~~~~~~~~~~~
512
513Enable/disable MACsec offload::
514
515   testpmd> set macsec offload (port_id) on encrypt (on|off) replay-protect (on|off)
516   testpmd> set macsec offload (port_id) off
517
518set macsec sc
519~~~~~~~~~~~~~
520
521Configure MACsec secure connection (SC)::
522
523   testpmd> set macsec sc (tx|rx) (port_id) (mac) (pi)
524
525.. note::
526
527   The pi argument is ignored for tx.
528   Check the NIC Datasheet for hardware limits.
529
530set macsec sa
531~~~~~~~~~~~~~
532
533Configure MACsec secure association (SA)::
534
535   testpmd> set macsec sa (tx|rx) (port_id) (idx) (an) (pn) (key)
536
537.. note::
538
539   The IDX value must be 0 or 1.
540   Check the NIC Datasheet for hardware limits.
541
542vlan set strip
543~~~~~~~~~~~~~~
544
545Set the VLAN strip on a port::
546
547   testpmd> vlan set strip (on|off) (port_id)
548
549vlan set stripq
550~~~~~~~~~~~~~~~
551
552Set the VLAN strip for a queue on a port::
553
554   testpmd> vlan set stripq (on|off) (port_id,queue_id)
555
556vlan set stripq (for VF)
557~~~~~~~~~~~~~~~~~~~~~~~~
558
559Set VLAN strip for all queues in a pool for a VF from the PF::
560
561   testpmd> set vf vlan stripq (port_id) (vf_id) (on|off)
562
563vlan set insert (for VF)
564~~~~~~~~~~~~~~~~~~~~~~~~
565
566Set VLAN insert for a VF from the PF::
567
568   testpmd> set vf vlan insert (port_id) (vf_id) (vlan_id)
569
570vlan set antispoof (for VF)
571~~~~~~~~~~~~~~~~~~~~~~~~~~~
572
573Set VLAN antispoof for a VF from the PF::
574
575   testpmd> set vf vlan antispoof (port_id) (vf_id) (on|off)
576
577vlan set filter
578~~~~~~~~~~~~~~~
579
580Set the VLAN filter on a port::
581
582   testpmd> vlan set filter (on|off) (port_id)
583
584vlan set qinq
585~~~~~~~~~~~~~
586
587Set the VLAN QinQ (extended queue in queue) on for a port::
588
589   testpmd> vlan set qinq (on|off) (port_id)
590
591vlan set tpid
592~~~~~~~~~~~~~
593
594Set the inner or outer VLAN TPID for packet filtering on a port::
595
596   testpmd> vlan set (inner|outer) tpid (value) (port_id)
597
598.. note::
599
600   TPID value must be a 16-bit number (value <= 65536).
601
602rx_vlan add
603~~~~~~~~~~~
604
605Add a VLAN ID, or all identifiers, to the set of VLAN identifiers filtered by port ID::
606
607   testpmd> rx_vlan add (vlan_id|all) (port_id)
608
609.. note::
610
611   VLAN filter must be set on that port. VLAN ID < 4096.
612   Depending on the NIC used, number of vlan_ids may be limited to the maximum entries
613   in VFTA table. This is important if enabling all vlan_ids.
614
615rx_vlan rm
616~~~~~~~~~~
617
618Remove a VLAN ID, or all identifiers, from the set of VLAN identifiers filtered by port ID::
619
620   testpmd> rx_vlan rm (vlan_id|all) (port_id)
621
622rx_vlan add (for VF)
623~~~~~~~~~~~~~~~~~~~~
624
625Add a VLAN ID, to the set of VLAN identifiers filtered for VF(s) for port ID::
626
627   testpmd> rx_vlan add (vlan_id) port (port_id) vf (vf_mask)
628
629rx_vlan rm (for VF)
630~~~~~~~~~~~~~~~~~~~
631
632Remove a VLAN ID, from the set of VLAN identifiers filtered for VF(s) for port ID::
633
634   testpmd> rx_vlan rm (vlan_id) port (port_id) vf (vf_mask)
635
636tunnel_filter add
637~~~~~~~~~~~~~~~~~
638
639Add a tunnel filter on a port::
640
641   testpmd> tunnel_filter add (port_id) (outer_mac) (inner_mac) (ip_addr) \
642            (inner_vlan) (vxlan|nvgre|ipingre) (imac-ivlan|imac-ivlan-tenid|\
643            imac-tenid|imac|omac-imac-tenid|oip|iip) (tenant_id) (queue_id)
644
645The available information categories are:
646
647* ``vxlan``: Set tunnel type as VXLAN.
648
649* ``nvgre``: Set tunnel type as NVGRE.
650
651* ``ipingre``: Set tunnel type as IP-in-GRE.
652
653* ``imac-ivlan``: Set filter type as Inner MAC and VLAN.
654
655* ``imac-ivlan-tenid``: Set filter type as Inner MAC, VLAN and tenant ID.
656
657* ``imac-tenid``: Set filter type as Inner MAC and tenant ID.
658
659* ``imac``: Set filter type as Inner MAC.
660
661* ``omac-imac-tenid``: Set filter type as Outer MAC, Inner MAC and tenant ID.
662
663* ``oip``: Set filter type as Outer IP.
664
665* ``iip``: Set filter type as Inner IP.
666
667Example::
668
669   testpmd> tunnel_filter add 0 68:05:CA:28:09:82 00:00:00:00:00:00 \
670            192.168.2.2 0 ipingre oip 1 1
671
672   Set an IP-in-GRE tunnel on port 0, and the filter type is Outer IP.
673
674tunnel_filter remove
675~~~~~~~~~~~~~~~~~~~~
676
677Remove a tunnel filter on a port::
678
679   testpmd> tunnel_filter rm (port_id) (outer_mac) (inner_mac) (ip_addr) \
680            (inner_vlan) (vxlan|nvgre|ipingre) (imac-ivlan|imac-ivlan-tenid|\
681            imac-tenid|imac|omac-imac-tenid|oip|iip) (tenant_id) (queue_id)
682
683rx_vxlan_port add
684~~~~~~~~~~~~~~~~~
685
686Add an UDP port for VXLAN packet filter on a port::
687
688   testpmd> rx_vxlan_port add (udp_port) (port_id)
689
690rx_vxlan_port remove
691~~~~~~~~~~~~~~~~~~~~
692
693Remove an UDP port for VXLAN packet filter on a port::
694
695   testpmd> rx_vxlan_port rm (udp_port) (port_id)
696
697tx_vlan set
698~~~~~~~~~~~
699
700Set hardware insertion of VLAN IDs in packets sent on a port::
701
702   testpmd> tx_vlan set (port_id) vlan_id[, vlan_id_outer]
703
704For example, set a single VLAN ID (5) insertion on port 0::
705
706   tx_vlan set 0 5
707
708Or, set double VLAN ID (inner: 2, outer: 3) insertion on port 1::
709
710   tx_vlan set 1 2 3
711
712
713tx_vlan set pvid
714~~~~~~~~~~~~~~~~
715
716Set port based hardware insertion of VLAN ID in packets sent on a port::
717
718   testpmd> tx_vlan set pvid (port_id) (vlan_id) (on|off)
719
720tx_vlan reset
721~~~~~~~~~~~~~
722
723Disable hardware insertion of a VLAN header in packets sent on a port::
724
725   testpmd> tx_vlan reset (port_id)
726
727csum set
728~~~~~~~~
729
730Select hardware or software calculation of the checksum when
731transmitting a packet using the ``csum`` forwarding engine::
732
733   testpmd> csum set (ip|udp|tcp|sctp|outer-ip) (hw|sw) (port_id)
734
735Where:
736
737* ``ip|udp|tcp|sctp`` always relate to  the inner layer.
738
739* ``outer-ip`` relates to the outer IP layer (only for IPv4) in the case where the packet is recognized
740  as a tunnel packet by the forwarding engine (vxlan, gre and ipip are
741  supported). See also the ``csum parse-tunnel`` command.
742
743.. note::
744
745   Check the NIC Datasheet for hardware limits.
746
747csum parse-tunnel
748~~~~~~~~~~~~~~~~~
749
750Define how tunneled packets should be handled by the csum forward
751engine::
752
753   testpmd> csum parse-tunnel (on|off) (tx_port_id)
754
755If enabled, the csum forward engine will try to recognize supported
756tunnel headers (vxlan, gre, ipip).
757
758If disabled, treat tunnel packets as non-tunneled packets (a inner
759header is handled as a packet payload).
760
761.. note::
762
763   The port argument is the TX port like in the ``csum set`` command.
764
765Example:
766
767Consider a packet in packet like the following::
768
769   eth_out/ipv4_out/udp_out/vxlan/eth_in/ipv4_in/tcp_in
770
771* If parse-tunnel is enabled, the ``ip|udp|tcp|sctp`` parameters of ``csum set``
772  command relate to the inner headers (here ``ipv4_in`` and ``tcp_in``), and the
773  ``outer-ip parameter`` relates to the outer headers (here ``ipv4_out``).
774
775* If parse-tunnel is disabled, the ``ip|udp|tcp|sctp`` parameters of ``csum  set``
776   command relate to the outer headers, here ``ipv4_out`` and ``udp_out``.
777
778csum show
779~~~~~~~~~
780
781Display tx checksum offload configuration::
782
783   testpmd> csum show (port_id)
784
785tso set
786~~~~~~~
787
788Enable TCP Segmentation Offload (TSO) in the ``csum`` forwarding engine::
789
790   testpmd> tso set (segsize) (port_id)
791
792.. note::
793
794   Check the NIC datasheet for hardware limits.
795
796tso show
797~~~~~~~~
798
799Display the status of TCP Segmentation Offload::
800
801   testpmd> tso show (port_id)
802
803mac_addr add
804~~~~~~~~~~~~
805
806Add an alternative MAC address to a port::
807
808   testpmd> mac_addr add (port_id) (XX:XX:XX:XX:XX:XX)
809
810mac_addr remove
811~~~~~~~~~~~~~~~
812
813Remove a MAC address from a port::
814
815   testpmd> mac_addr remove (port_id) (XX:XX:XX:XX:XX:XX)
816
817mac_addr add (for VF)
818~~~~~~~~~~~~~~~~~~~~~
819
820Add an alternative MAC address for a VF to a port::
821
822   testpmd> mac_add add port (port_id) vf (vf_id) (XX:XX:XX:XX:XX:XX)
823
824mac_addr set (for VF)
825~~~~~~~~~~~~~~~~~~~~~
826
827Set the MAC address for a VF from the PF::
828
829   testpmd> set vf mac addr (port_id) (vf_id) (XX:XX:XX:XX:XX:XX)
830
831set port-uta
832~~~~~~~~~~~~
833
834Set the unicast hash filter(s) on/off for a port::
835
836   testpmd> set port (port_id) uta (XX:XX:XX:XX:XX:XX|all) (on|off)
837
838set promisc
839~~~~~~~~~~~
840
841Set the promiscuous mode on for a port or for all ports.
842In promiscuous mode packets are not dropped if they aren't for the specified MAC address::
843
844   testpmd> set promisc (port_id|all) (on|off)
845
846set allmulti
847~~~~~~~~~~~~
848
849Set the allmulti mode for a port or for all ports::
850
851   testpmd> set allmulti (port_id|all) (on|off)
852
853Same as the ifconfig (8) option. Controls how multicast packets are handled.
854
855set promisc (for VF)
856~~~~~~~~~~~~~~~~~~~~
857
858Set the unicast promiscuous mode for a VF from PF.
859It's supported by Intel i40e NICs now.
860In promiscuous mode packets are not dropped if they aren't for the specified MAC address::
861
862   testpmd> set vf promisc (port_id) (vf_id) (on|off)
863
864set flow_ctrl rx
865~~~~~~~~~~~~~~~~
866
867Set the link flow control parameter on a port::
868
869   testpmd> set flow_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \
870            (pause_time) (send_xon) mac_ctrl_frame_fwd (on|off) \
871	    autoneg (on|off) (port_id)
872
873Where:
874
875* ``high_water`` (integer): High threshold value to trigger XOFF.
876
877* ``low_water`` (integer): Low threshold value to trigger XON.
878
879* ``pause_time`` (integer): Pause quota in the Pause frame.
880
881* ``send_xon`` (0/1): Send XON frame.
882
883* ``mac_ctrl_frame_fwd``: Enable receiving MAC control frames.
884
885* ``autoneg``: Change the auto-negotiation parameter.
886
887set pfc_ctrl rx
888~~~~~~~~~~~~~~~
889
890Set the priority flow control parameter on a port::
891
892   testpmd> set pfc_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \
893            (pause_time) (priority) (port_id)
894
895Where:
896
897* ``high_water`` (integer): High threshold value.
898
899* ``low_water`` (integer): Low threshold value.
900
901* ``pause_time`` (integer): Pause quota in the Pause frame.
902
903* ``priority`` (0-7): VLAN User Priority.
904
905set stat_qmap
906~~~~~~~~~~~~~
907
908Set statistics mapping (qmapping 0..15) for RX/TX queue on port::
909
910   testpmd> set stat_qmap (tx|rx) (port_id) (queue_id) (qmapping)
911
912For example, to set rx queue 2 on port 0 to mapping 5::
913
914   testpmd>set stat_qmap rx 0 2 5
915
916set port - rx/tx (for VF)
917~~~~~~~~~~~~~~~~~~~~~~~~~
918
919Set VF receive/transmit from a port::
920
921   testpmd> set port (port_id) vf (vf_id) (rx|tx) (on|off)
922
923set port - mac address filter (for VF)
924~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
925
926Add/Remove unicast or multicast MAC addr filter for a VF::
927
928   testpmd> set port (port_id) vf (vf_id) (mac_addr) \
929            (exact-mac|exact-mac-vlan|hashmac|hashmac-vlan) (on|off)
930
931set port - rx mode(for VF)
932~~~~~~~~~~~~~~~~~~~~~~~~~~
933
934Set the VF receive mode of a port::
935
936   testpmd> set port (port_id) vf (vf_id) \
937            rxmode (AUPE|ROPE|BAM|MPE) (on|off)
938
939The available receive modes are:
940
941* ``AUPE``: Accepts untagged VLAN.
942
943* ``ROPE``: Accepts unicast hash.
944
945* ``BAM``: Accepts broadcast packets.
946
947* ``MPE``: Accepts all multicast packets.
948
949set port - tx_rate (for Queue)
950~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
951
952Set TX rate limitation for a queue on a port::
953
954   testpmd> set port (port_id) queue (queue_id) rate (rate_value)
955
956set port - tx_rate (for VF)
957~~~~~~~~~~~~~~~~~~~~~~~~~~~
958
959Set TX rate limitation for queues in VF on a port::
960
961   testpmd> set port (port_id) vf (vf_id) rate (rate_value) queue_mask (queue_mask)
962
963set port - mirror rule
964~~~~~~~~~~~~~~~~~~~~~~
965
966Set pool or vlan type mirror rule for a port::
967
968   testpmd> set port (port_id) mirror-rule (rule_id) \
969            (pool-mirror-up|pool-mirror-down|vlan-mirror) \
970            (poolmask|vlanid[,vlanid]*) dst-pool (pool_id) (on|off)
971
972Set link mirror rule for a port::
973
974   testpmd> set port (port_id) mirror-rule (rule_id) \
975           (uplink-mirror|downlink-mirror) dst-pool (pool_id) (on|off)
976
977For example to enable mirror traffic with vlan 0,1 to pool 0::
978
979   set port 0 mirror-rule 0 vlan-mirror 0,1 dst-pool 0 on
980
981reset port - mirror rule
982~~~~~~~~~~~~~~~~~~~~~~~~
983
984Reset a mirror rule for a port::
985
986   testpmd> reset port (port_id) mirror-rule (rule_id)
987
988set flush_rx
989~~~~~~~~~~~~
990
991Set the flush on RX streams before forwarding.
992The default is flush ``on``.
993Mainly used with PCAP drivers to turn off the default behavior of flushing the first 512 packets on RX streams::
994
995   testpmd> set flush_rx off
996
997set bypass mode
998~~~~~~~~~~~~~~~
999
1000Set the bypass mode for the lowest port on bypass enabled NIC::
1001
1002   testpmd> set bypass mode (normal|bypass|isolate) (port_id)
1003
1004set bypass event
1005~~~~~~~~~~~~~~~~
1006
1007Set the event required to initiate specified bypass mode for the lowest port on a bypass enabled::
1008
1009   testpmd> set bypass event (timeout|os_on|os_off|power_on|power_off) \
1010            mode (normal|bypass|isolate) (port_id)
1011
1012Where:
1013
1014* ``timeout``: Enable bypass after watchdog timeout.
1015
1016* ``os_on``: Enable bypass when OS/board is powered on.
1017
1018* ``os_off``: Enable bypass when OS/board is powered off.
1019
1020* ``power_on``: Enable bypass when power supply is turned on.
1021
1022* ``power_off``: Enable bypass when power supply is turned off.
1023
1024
1025set bypass timeout
1026~~~~~~~~~~~~~~~~~~
1027
1028Set the bypass watchdog timeout to ``n`` seconds where 0 = instant::
1029
1030   testpmd> set bypass timeout (0|1.5|2|3|4|8|16|32)
1031
1032show bypass config
1033~~~~~~~~~~~~~~~~~~
1034
1035Show the bypass configuration for a bypass enabled NIC using the lowest port on the NIC::
1036
1037   testpmd> show bypass config (port_id)
1038
1039set link up
1040~~~~~~~~~~~
1041
1042Set link up for a port::
1043
1044   testpmd> set link-up port (port id)
1045
1046set link down
1047~~~~~~~~~~~~~
1048
1049Set link down for a port::
1050
1051   testpmd> set link-down port (port id)
1052
1053E-tag set
1054~~~~~~~~~
1055
1056Enable E-tag insertion for a VF on a port::
1057
1058   testpmd> E-tag set insertion on port-tag-id (value) port (port_id) vf (vf_id)
1059
1060Disable E-tag insertion for a VF on a port::
1061
1062   testpmd> E-tag set insertion off port (port_id) vf (vf_id)
1063
1064Enable/disable E-tag stripping on a port::
1065
1066   testpmd> E-tag set stripping (on|off) port (port_id)
1067
1068Enable/disable E-tag based forwarding on a port::
1069
1070   testpmd> E-tag set forwarding (on|off) port (port_id)
1071
1072Add an E-tag forwarding filter on a port::
1073
1074   testpmd> E-tag set filter add e-tag-id (value) dst-pool (pool_id) port (port_id)
1075
1076Delete an E-tag forwarding filter on a port::
1077   testpmd> E-tag set filter del e-tag-id (value) port (port_id)
1078
1079
1080Port Functions
1081--------------
1082
1083The following sections show functions for configuring ports.
1084
1085.. note::
1086
1087   Port configuration changes only become active when forwarding is started/restarted.
1088
1089port attach
1090~~~~~~~~~~~
1091
1092Attach a port specified by pci address or virtual device args::
1093
1094   testpmd> port attach (identifier)
1095
1096To attach a new pci device, the device should be recognized by kernel first.
1097Then it should be moved under DPDK management.
1098Finally the port can be attached to testpmd.
1099
1100For example, to move a pci device using ixgbe under DPDK management:
1101
1102.. code-block:: console
1103
1104   # Check the status of the available devices.
1105   ./usertools/dpdk-devbind.py --status
1106
1107   Network devices using DPDK-compatible driver
1108   ============================================
1109   <none>
1110
1111   Network devices using kernel driver
1112   ===================================
1113   0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused=
1114
1115
1116   # Bind the device to igb_uio.
1117   sudo ./usertools/dpdk-devbind.py -b igb_uio 0000:0a:00.0
1118
1119
1120   # Recheck the status of the devices.
1121   ./usertools/dpdk-devbind.py --status
1122   Network devices using DPDK-compatible driver
1123   ============================================
1124   0000:0a:00.0 '82599ES 10-Gigabit' drv=igb_uio unused=
1125
1126To attach a port created by virtual device, above steps are not needed.
1127
1128For example, to attach a port whose pci address is 0000:0a:00.0.
1129
1130.. code-block:: console
1131
1132   testpmd> port attach 0000:0a:00.0
1133   Attaching a new port...
1134   EAL: PCI device 0000:0a:00.0 on NUMA socket -1
1135   EAL:   probe driver: 8086:10fb rte_ixgbe_pmd
1136   EAL:   PCI memory mapped at 0x7f83bfa00000
1137   EAL:   PCI memory mapped at 0x7f83bfa80000
1138   PMD: eth_ixgbe_dev_init(): MAC: 2, PHY: 18, SFP+: 5
1139   PMD: eth_ixgbe_dev_init(): port 0 vendorID=0x8086 deviceID=0x10fb
1140   Port 0 is attached. Now total ports is 1
1141   Done
1142
1143For example, to attach a port created by pcap PMD.
1144
1145.. code-block:: console
1146
1147   testpmd> port attach net_pcap0
1148   Attaching a new port...
1149   PMD: Initializing pmd_pcap for net_pcap0
1150   PMD: Creating pcap-backed ethdev on numa socket 0
1151   Port 0 is attached. Now total ports is 1
1152   Done
1153
1154In this case, identifier is ``net_pcap0``.
1155This identifier format is the same as ``--vdev`` format of DPDK applications.
1156
1157For example, to re-attach a bonded port which has been previously detached,
1158the mode and slave parameters must be given.
1159
1160.. code-block:: console
1161
1162   testpmd> port attach net_bond_0,mode=0,slave=1
1163   Attaching a new port...
1164   EAL: Initializing pmd_bond for net_bond_0
1165   EAL: Create bonded device net_bond_0 on port 0 in mode 0 on socket 0.
1166   Port 0 is attached. Now total ports is 1
1167   Done
1168
1169
1170port detach
1171~~~~~~~~~~~
1172
1173Detach a specific port::
1174
1175   testpmd> port detach (port_id)
1176
1177Before detaching a port, the port should be stopped and closed.
1178
1179For example, to detach a pci device port 0.
1180
1181.. code-block:: console
1182
1183   testpmd> port stop 0
1184   Stopping ports...
1185   Done
1186   testpmd> port close 0
1187   Closing ports...
1188   Done
1189
1190   testpmd> port detach 0
1191   Detaching a port...
1192   EAL: PCI device 0000:0a:00.0 on NUMA socket -1
1193   EAL:   remove driver: 8086:10fb rte_ixgbe_pmd
1194   EAL:   PCI memory unmapped at 0x7f83bfa00000
1195   EAL:   PCI memory unmapped at 0x7f83bfa80000
1196   Done
1197
1198
1199For example, to detach a virtual device port 0.
1200
1201.. code-block:: console
1202
1203   testpmd> port stop 0
1204   Stopping ports...
1205   Done
1206   testpmd> port close 0
1207   Closing ports...
1208   Done
1209
1210   testpmd> port detach 0
1211   Detaching a port...
1212   PMD: Closing pcap ethdev on numa socket 0
1213   Port 'net_pcap0' is detached. Now total ports is 0
1214   Done
1215
1216To remove a pci device completely from the system, first detach the port from testpmd.
1217Then the device should be moved under kernel management.
1218Finally the device can be removed using kernel pci hotplug functionality.
1219
1220For example, to move a pci device under kernel management:
1221
1222.. code-block:: console
1223
1224   sudo ./usertools/dpdk-devbind.py -b ixgbe 0000:0a:00.0
1225
1226   ./usertools/dpdk-devbind.py --status
1227
1228   Network devices using DPDK-compatible driver
1229   ============================================
1230   <none>
1231
1232   Network devices using kernel driver
1233   ===================================
1234   0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused=igb_uio
1235
1236To remove a port created by a virtual device, above steps are not needed.
1237
1238port start
1239~~~~~~~~~~
1240
1241Start all ports or a specific port::
1242
1243   testpmd> port start (port_id|all)
1244
1245port stop
1246~~~~~~~~~
1247
1248Stop all ports or a specific port::
1249
1250   testpmd> port stop (port_id|all)
1251
1252port close
1253~~~~~~~~~~
1254
1255Close all ports or a specific port::
1256
1257   testpmd> port close (port_id|all)
1258
1259port start/stop queue
1260~~~~~~~~~~~~~~~~~~~~~
1261
1262Start/stop a rx/tx queue on a specific port::
1263
1264   testpmd> port (port_id) (rxq|txq) (queue_id) (start|stop)
1265
1266Only take effect when port is started.
1267
1268port config - speed
1269~~~~~~~~~~~~~~~~~~~
1270
1271Set the speed and duplex mode for all ports or a specific port::
1272
1273   testpmd> port config (port_id|all) speed (10|100|1000|10000|25000|40000|50000|100000|auto) \
1274            duplex (half|full|auto)
1275
1276port config - queues/descriptors
1277~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1278
1279Set number of queues/descriptors for rxq, txq, rxd and txd::
1280
1281   testpmd> port config all (rxq|txq|rxd|txd) (value)
1282
1283This is equivalent to the ``--rxq``, ``--txq``, ``--rxd`` and ``--txd`` command-line options.
1284
1285port config - max-pkt-len
1286~~~~~~~~~~~~~~~~~~~~~~~~~
1287
1288Set the maximum packet length::
1289
1290   testpmd> port config all max-pkt-len (value)
1291
1292This is equivalent to the ``--max-pkt-len`` command-line option.
1293
1294port config - CRC Strip
1295~~~~~~~~~~~~~~~~~~~~~~~
1296
1297Set hardware CRC stripping on or off for all ports::
1298
1299   testpmd> port config all crc-strip (on|off)
1300
1301CRC stripping is off by default.
1302
1303The ``on`` option is equivalent to the ``--crc-strip`` command-line option.
1304
1305port config - scatter
1306~~~~~~~~~~~~~~~~~~~~~~~
1307
1308Set RX scatter mode on or off for all ports::
1309
1310   testpmd> port config all scatter (on|off)
1311
1312RX scatter mode is off by default.
1313
1314The ``on`` option is equivalent to the ``--enable-scatter`` command-line option.
1315
1316port config - TX queue flags
1317~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1318
1319Set a hexadecimal bitmap of TX queue flags for all ports::
1320
1321   testpmd> port config all txqflags value
1322
1323This command is equivalent to the ``--txqflags`` command-line option.
1324
1325port config - RX Checksum
1326~~~~~~~~~~~~~~~~~~~~~~~~~
1327
1328Set hardware RX checksum offload to on or off for all ports::
1329
1330   testpmd> port config all rx-cksum (on|off)
1331
1332Checksum offload is off by default.
1333
1334The ``on`` option is equivalent to the ``--enable-rx-cksum`` command-line option.
1335
1336port config - VLAN
1337~~~~~~~~~~~~~~~~~~
1338
1339Set hardware VLAN on or off for all ports::
1340
1341   testpmd> port config all hw-vlan (on|off)
1342
1343Hardware VLAN is on by default.
1344
1345The ``off`` option is equivalent to the ``--disable-hw-vlan`` command-line option.
1346
1347port config - VLAN filter
1348~~~~~~~~~~~~~~~~~~~~~~~~~
1349
1350Set hardware VLAN filter on or off for all ports::
1351
1352   testpmd> port config all hw-vlan-filter (on|off)
1353
1354Hardware VLAN filter is on by default.
1355
1356The ``off`` option is equivalent to the ``--disable-hw-vlan-filter`` command-line option.
1357
1358port config - VLAN strip
1359~~~~~~~~~~~~~~~~~~~~~~~~
1360
1361Set hardware VLAN strip on or off for all ports::
1362
1363   testpmd> port config all hw-vlan-strip (on|off)
1364
1365Hardware VLAN strip is on by default.
1366
1367The ``off`` option is equivalent to the ``--disable-hw-vlan-strip`` command-line option.
1368
1369port config - VLAN extend
1370~~~~~~~~~~~~~~~~~~~~~~~~~
1371
1372Set hardware VLAN extend on or off for all ports::
1373
1374   testpmd> port config all hw-vlan-extend (on|off)
1375
1376Hardware VLAN extend is off by default.
1377
1378The ``off`` option is equivalent to the ``--disable-hw-vlan-extend`` command-line option.
1379
1380port config - Drop Packets
1381~~~~~~~~~~~~~~~~~~~~~~~~~~
1382
1383Set packet drop for packets with no descriptors on or off for all ports::
1384
1385   testpmd> port config all drop-en (on|off)
1386
1387Packet dropping for packets with no descriptors is off by default.
1388
1389The ``on`` option is equivalent to the ``--enable-drop-en`` command-line option.
1390
1391port config - RSS
1392~~~~~~~~~~~~~~~~~
1393
1394Set the RSS (Receive Side Scaling) mode on or off::
1395
1396   testpmd> port config all rss (all|ip|tcp|udp|sctp|ether|port|vxlan|geneve|nvgre|none)
1397
1398RSS is on by default.
1399
1400The ``none`` option is equivalent to the ``--disable-rss`` command-line option.
1401
1402port config - RSS Reta
1403~~~~~~~~~~~~~~~~~~~~~~
1404
1405Set the RSS (Receive Side Scaling) redirection table::
1406
1407   testpmd> port config all rss reta (hash,queue)[,(hash,queue)]
1408
1409port config - DCB
1410~~~~~~~~~~~~~~~~~
1411
1412Set the DCB mode for an individual port::
1413
1414   testpmd> port config (port_id) dcb vt (on|off) (traffic_class) pfc (on|off)
1415
1416The traffic class should be 4 or 8.
1417
1418port config - Burst
1419~~~~~~~~~~~~~~~~~~~
1420
1421Set the number of packets per burst::
1422
1423   testpmd> port config all burst (value)
1424
1425This is equivalent to the ``--burst`` command-line option.
1426
1427port config - Threshold
1428~~~~~~~~~~~~~~~~~~~~~~~
1429
1430Set thresholds for TX/RX queues::
1431
1432   testpmd> port config all (threshold) (value)
1433
1434Where the threshold type can be:
1435
1436* ``txpt:`` Set the prefetch threshold register of the TX rings, 0 <= value <= 255.
1437
1438* ``txht:`` Set the host threshold register of the TX rings, 0 <= value <= 255.
1439
1440* ``txwt:`` Set the write-back threshold register of the TX rings, 0 <= value <= 255.
1441
1442* ``rxpt:`` Set the prefetch threshold register of the RX rings, 0 <= value <= 255.
1443
1444* ``rxht:`` Set the host threshold register of the RX rings, 0 <= value <= 255.
1445
1446* ``rxwt:`` Set the write-back threshold register of the RX rings, 0 <= value <= 255.
1447
1448* ``txfreet:`` Set the transmit free threshold of the TX rings, 0 <= value <= txd.
1449
1450* ``rxfreet:`` Set the transmit free threshold of the RX rings, 0 <= value <= rxd.
1451
1452* ``txrst:`` Set the transmit RS bit threshold of TX rings, 0 <= value <= txd.
1453
1454These threshold options are also available from the command-line.
1455
1456port config - E-tag
1457~~~~~~~~~~~~~~~~~~~
1458
1459Set the value of ether-type for E-tag::
1460
1461   testpmd> port config (port_id|all) l2-tunnel E-tag ether-type (value)
1462
1463Enable/disable the E-tag support::
1464
1465   testpmd> port config (port_id|all) l2-tunnel E-tag (enable|disable)
1466
1467
1468Link Bonding Functions
1469----------------------
1470
1471The Link Bonding functions make it possible to dynamically create and
1472manage link bonding devices from within testpmd interactive prompt.
1473
1474create bonded device
1475~~~~~~~~~~~~~~~~~~~~
1476
1477Create a new bonding device::
1478
1479   testpmd> create bonded device (mode) (socket)
1480
1481For example, to create a bonded device in mode 1 on socket 0::
1482
1483   testpmd> create bonded 1 0
1484   created new bonded device (port X)
1485
1486add bonding slave
1487~~~~~~~~~~~~~~~~~
1488
1489Adds Ethernet device to a Link Bonding device::
1490
1491   testpmd> add bonding slave (slave id) (port id)
1492
1493For example, to add Ethernet device (port 6) to a Link Bonding device (port 10)::
1494
1495   testpmd> add bonding slave 6 10
1496
1497
1498remove bonding slave
1499~~~~~~~~~~~~~~~~~~~~
1500
1501Removes an Ethernet slave device from a Link Bonding device::
1502
1503   testpmd> remove bonding slave (slave id) (port id)
1504
1505For example, to remove Ethernet slave device (port 6) to a Link Bonding device (port 10)::
1506
1507   testpmd> remove bonding slave 6 10
1508
1509set bonding mode
1510~~~~~~~~~~~~~~~~
1511
1512Set the Link Bonding mode of a Link Bonding device::
1513
1514   testpmd> set bonding mode (value) (port id)
1515
1516For example, to set the bonding mode of a Link Bonding device (port 10) to broadcast (mode 3)::
1517
1518   testpmd> set bonding mode 3 10
1519
1520set bonding primary
1521~~~~~~~~~~~~~~~~~~~
1522
1523Set an Ethernet slave device as the primary device on a Link Bonding device::
1524
1525   testpmd> set bonding primary (slave id) (port id)
1526
1527For example, to set the Ethernet slave device (port 6) as the primary port of a Link Bonding device (port 10)::
1528
1529   testpmd> set bonding primary 6 10
1530
1531set bonding mac
1532~~~~~~~~~~~~~~~
1533
1534Set the MAC address of a Link Bonding device::
1535
1536   testpmd> set bonding mac (port id) (mac)
1537
1538For example, to set the MAC address of a Link Bonding device (port 10) to 00:00:00:00:00:01::
1539
1540   testpmd> set bonding mac 10 00:00:00:00:00:01
1541
1542set bonding xmit_balance_policy
1543~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1544
1545Set the transmission policy for a Link Bonding device when it is in Balance XOR mode::
1546
1547   testpmd> set bonding xmit_balance_policy (port_id) (l2|l23|l34)
1548
1549For example, set a Link Bonding device (port 10) to use a balance policy of layer 3+4 (IP addresses & UDP ports)::
1550
1551   testpmd> set bonding xmit_balance_policy 10 l34
1552
1553
1554set bonding mon_period
1555~~~~~~~~~~~~~~~~~~~~~~
1556
1557Set the link status monitoring polling period in milliseconds for a bonding device.
1558
1559This adds support for PMD slave devices which do not support link status interrupts.
1560When the mon_period is set to a value greater than 0 then all PMD's which do not support
1561link status ISR will be queried every polling interval to check if their link status has changed::
1562
1563   testpmd> set bonding mon_period (port_id) (value)
1564
1565For example, to set the link status monitoring polling period of bonded device (port 5) to 150ms::
1566
1567   testpmd> set bonding mon_period 5 150
1568
1569
1570show bonding config
1571~~~~~~~~~~~~~~~~~~~
1572
1573Show the current configuration of a Link Bonding device::
1574
1575   testpmd> show bonding config (port id)
1576
1577For example,
1578to show the configuration a Link Bonding device (port 9) with 3 slave devices (1, 3, 4)
1579in balance mode with a transmission policy of layer 2+3::
1580
1581   testpmd> show bonding config 9
1582        Bonding mode: 2
1583        Balance Xmit Policy: BALANCE_XMIT_POLICY_LAYER23
1584        Slaves (3): [1 3 4]
1585        Active Slaves (3): [1 3 4]
1586        Primary: [3]
1587
1588
1589Register Functions
1590------------------
1591
1592The Register Functions can be used to read from and write to registers on the network card referenced by a port number.
1593This is mainly useful for debugging purposes.
1594Reference should be made to the appropriate datasheet for the network card for details on the register addresses
1595and fields that can be accessed.
1596
1597read reg
1598~~~~~~~~
1599
1600Display the value of a port register::
1601
1602   testpmd> read reg (port_id) (address)
1603
1604For example, to examine the Flow Director control register (FDIRCTL, 0x0000EE000) on an Intel 82599 10 GbE Controller::
1605
1606   testpmd> read reg 0 0xEE00
1607   port 0 PCI register at offset 0xEE00: 0x4A060029 (1241907241)
1608
1609read regfield
1610~~~~~~~~~~~~~
1611
1612Display a port register bit field::
1613
1614   testpmd> read regfield (port_id) (address) (bit_x) (bit_y)
1615
1616For example, reading the lowest two bits from the register in the example above::
1617
1618   testpmd> read regfield 0 0xEE00 0 1
1619   port 0 PCI register at offset 0xEE00: bits[0, 1]=0x1 (1)
1620
1621read regbit
1622~~~~~~~~~~~
1623
1624Display a single port register bit::
1625
1626   testpmd> read regbit (port_id) (address) (bit_x)
1627
1628For example, reading the lowest bit from the register in the example above::
1629
1630   testpmd> read regbit 0 0xEE00 0
1631   port 0 PCI register at offset 0xEE00: bit 0=1
1632
1633write reg
1634~~~~~~~~~
1635
1636Set the value of a port register::
1637
1638   testpmd> write reg (port_id) (address) (value)
1639
1640For example, to clear a register::
1641
1642   testpmd> write reg 0 0xEE00 0x0
1643   port 0 PCI register at offset 0xEE00: 0x00000000 (0)
1644
1645write regfield
1646~~~~~~~~~~~~~~
1647
1648Set bit field of a port register::
1649
1650   testpmd> write regfield (port_id) (address) (bit_x) (bit_y) (value)
1651
1652For example, writing to the register cleared in the example above::
1653
1654   testpmd> write regfield 0 0xEE00 0 1 2
1655   port 0 PCI register at offset 0xEE00: 0x00000002 (2)
1656
1657write regbit
1658~~~~~~~~~~~~
1659
1660Set single bit value of a port register::
1661
1662   testpmd> write regbit (port_id) (address) (bit_x) (value)
1663
1664For example, to set the high bit in the register from the example above::
1665
1666   testpmd> write regbit 0 0xEE00 31 1
1667   port 0 PCI register at offset 0xEE00: 0x8000000A (2147483658)
1668
1669
1670Filter Functions
1671----------------
1672
1673This section details the available filter functions that are available.
1674
1675Note these functions interface the deprecated legacy filtering framework,
1676superseded by *rte_flow*. See `Flow rules management`_.
1677
1678ethertype_filter
1679~~~~~~~~~~~~~~~~~~~~
1680
1681Add or delete a L2 Ethertype filter, which identify packets by their L2 Ethertype mainly assign them to a receive queue::
1682
1683   ethertype_filter (port_id) (add|del) (mac_addr|mac_ignr) (mac_address) \
1684                    ethertype (ether_type) (drop|fwd) queue (queue_id)
1685
1686The available information parameters are:
1687
1688* ``port_id``: The port which the Ethertype filter assigned on.
1689
1690* ``mac_addr``: Compare destination mac address.
1691
1692* ``mac_ignr``: Ignore destination mac address match.
1693
1694* ``mac_address``: Destination mac address to match.
1695
1696* ``ether_type``: The EtherType value want to match,
1697  for example 0x0806 for ARP packet. 0x0800 (IPv4) and 0x86DD (IPv6) are invalid.
1698
1699* ``queue_id``: The receive queue associated with this EtherType filter.
1700  It is meaningless when deleting or dropping.
1701
1702Example, to add/remove an ethertype filter rule::
1703
1704   testpmd> ethertype_filter 0 add mac_ignr 00:11:22:33:44:55 \
1705                             ethertype 0x0806 fwd queue 3
1706
1707   testpmd> ethertype_filter 0 del mac_ignr 00:11:22:33:44:55 \
1708                             ethertype 0x0806 fwd queue 3
1709
17102tuple_filter
1711~~~~~~~~~~~~~~~~~
1712
1713Add or delete a 2-tuple filter,
1714which identifies packets by specific protocol and destination TCP/UDP port
1715and forwards packets into one of the receive queues::
1716
1717   2tuple_filter (port_id) (add|del) dst_port (dst_port_value) \
1718                 protocol (protocol_value) mask (mask_value) \
1719                 tcp_flags (tcp_flags_value) priority (prio_value) \
1720                 queue (queue_id)
1721
1722The available information parameters are:
1723
1724* ``port_id``: The port which the 2-tuple filter assigned on.
1725
1726* ``dst_port_value``: Destination port in L4.
1727
1728* ``protocol_value``: IP L4 protocol.
1729
1730* ``mask_value``: Participates in the match or not by bit for field above, 1b means participate.
1731
1732* ``tcp_flags_value``: TCP control bits. The non-zero value is invalid, when the pro_value is not set to 0x06 (TCP).
1733
1734* ``prio_value``: Priority of this filter.
1735
1736* ``queue_id``: The receive queue associated with this 2-tuple filter.
1737
1738Example, to add/remove an 2tuple filter rule::
1739
1740   testpmd> 2tuple_filter 0 add dst_port 32 protocol 0x06 mask 0x03 \
1741                          tcp_flags 0x02 priority 3 queue 3
1742
1743   testpmd> 2tuple_filter 0 del dst_port 32 protocol 0x06 mask 0x03 \
1744                          tcp_flags 0x02 priority 3 queue 3
1745
17465tuple_filter
1747~~~~~~~~~~~~~~~~~
1748
1749Add or delete a 5-tuple filter,
1750which consists of a 5-tuple (protocol, source and destination IP addresses, source and destination TCP/UDP/SCTP port)
1751and routes packets into one of the receive queues::
1752
1753   5tuple_filter (port_id) (add|del) dst_ip (dst_address) src_ip \
1754                 (src_address) dst_port (dst_port_value) \
1755                 src_port (src_port_value) protocol (protocol_value) \
1756                 mask (mask_value) tcp_flags (tcp_flags_value) \
1757                 priority (prio_value) queue (queue_id)
1758
1759The available information parameters are:
1760
1761* ``port_id``: The port which the 5-tuple filter assigned on.
1762
1763* ``dst_address``: Destination IP address.
1764
1765* ``src_address``: Source IP address.
1766
1767* ``dst_port_value``: TCP/UDP destination port.
1768
1769* ``src_port_value``: TCP/UDP source port.
1770
1771* ``protocol_value``: L4 protocol.
1772
1773* ``mask_value``: Participates in the match or not by bit for field above, 1b means participate
1774
1775* ``tcp_flags_value``: TCP control bits. The non-zero value is invalid, when the protocol_value is not set to 0x06 (TCP).
1776
1777* ``prio_value``: The priority of this filter.
1778
1779* ``queue_id``: The receive queue associated with this 5-tuple filter.
1780
1781Example, to add/remove an 5tuple filter rule::
1782
1783   testpmd> 5tuple_filter 0 add dst_ip 2.2.2.5 src_ip 2.2.2.4 \
1784            dst_port 64 src_port 32 protocol 0x06 mask 0x1F \
1785            flags 0x0 priority 3 queue 3
1786
1787   testpmd> 5tuple_filter 0 del dst_ip 2.2.2.5 src_ip 2.2.2.4 \
1788            dst_port 64 src_port 32 protocol 0x06 mask 0x1F \
1789            flags 0x0 priority 3 queue 3
1790
1791syn_filter
1792~~~~~~~~~~
1793
1794Using the  SYN filter, TCP packets whose *SYN* flag is set can be forwarded to a separate queue::
1795
1796   syn_filter (port_id) (add|del) priority (high|low) queue (queue_id)
1797
1798The available information parameters are:
1799
1800* ``port_id``: The port which the SYN filter assigned on.
1801
1802* ``high``: This SYN filter has higher priority than other filters.
1803
1804* ``low``: This SYN filter has lower priority than other filters.
1805
1806* ``queue_id``: The receive queue associated with this SYN filter
1807
1808Example::
1809
1810   testpmd> syn_filter 0 add priority high queue 3
1811
1812flex_filter
1813~~~~~~~~~~~
1814
1815With flex filter, packets can be recognized by any arbitrary pattern within the first 128 bytes of the packet
1816and routed into one of the receive queues::
1817
1818   flex_filter (port_id) (add|del) len (len_value) bytes (bytes_value) \
1819               mask (mask_value) priority (prio_value) queue (queue_id)
1820
1821The available information parameters are:
1822
1823* ``port_id``: The port which the Flex filter is assigned on.
1824
1825* ``len_value``: Filter length in bytes, no greater than 128.
1826
1827* ``bytes_value``: A string in hexadecimal, means the value the flex filter needs to match.
1828
1829* ``mask_value``: A string in hexadecimal, bit 1 means corresponding byte participates in the match.
1830
1831* ``prio_value``: The priority of this filter.
1832
1833* ``queue_id``: The receive queue associated with this Flex filter.
1834
1835Example::
1836
1837   testpmd> flex_filter 0 add len 16 bytes 0x00000000000000000000000008060000 \
1838                          mask 000C priority 3 queue 3
1839
1840   testpmd> flex_filter 0 del len 16 bytes 0x00000000000000000000000008060000 \
1841                          mask 000C priority 3 queue 3
1842
1843
1844.. _testpmd_flow_director:
1845
1846flow_director_filter
1847~~~~~~~~~~~~~~~~~~~~
1848
1849The Flow Director works in receive mode to identify specific flows or sets of flows and route them to specific queues.
1850
1851Four types of filtering are supported which are referred to as Perfect Match, Signature, Perfect-mac-vlan and
1852Perfect-tunnel filters, the match mode is set by the ``--pkt-filter-mode`` command-line parameter:
1853
1854* Perfect match filters.
1855  The hardware checks a match between the masked fields of the received packets and the programmed filters.
1856  The masked fields are for IP flow.
1857
1858* Signature filters.
1859  The hardware checks a match between a hash-based signature of the masked fields of the received packet.
1860
1861* Perfect-mac-vlan match filters.
1862  The hardware checks a match between the masked fields of the received packets and the programmed filters.
1863  The masked fields are for MAC VLAN flow.
1864
1865* Perfect-tunnel match filters.
1866  The hardware checks a match between the masked fields of the received packets and the programmed filters.
1867  The masked fields are for tunnel flow.
1868
1869The Flow Director filters can match the different fields for different type of packet: flow type, specific input set
1870per flow type and the flexible payload.
1871
1872The Flow Director can also mask out parts of all of these fields so that filters
1873are only applied to certain fields or parts of the fields.
1874
1875Different NICs may have different capabilities, command show port fdir (port_id) can be used to acquire the information.
1876
1877# Commands to add flow director filters of different flow types::
1878
1879   flow_director_filter (port_id) mode IP (add|del|update) \
1880                        flow (ipv4-other|ipv4-frag|ipv6-other|ipv6-frag) \
1881                        src (src_ip_address) dst (dst_ip_address) \
1882                        tos (tos_value) proto (proto_value) ttl (ttl_value) \
1883                        vlan (vlan_value) flexbytes (flexbytes_value) \
1884                        (drop|fwd) pf|vf(vf_id) queue (queue_id) \
1885                        fd_id (fd_id_value)
1886
1887   flow_director_filter (port_id) mode IP (add|del|update) \
1888                        flow (ipv4-tcp|ipv4-udp|ipv6-tcp|ipv6-udp) \
1889                        src (src_ip_address) (src_port) \
1890                        dst (dst_ip_address) (dst_port) \
1891                        tos (tos_value) ttl (ttl_value) \
1892                        vlan (vlan_value) flexbytes (flexbytes_value) \
1893                        (drop|fwd) queue pf|vf(vf_id) (queue_id) \
1894                        fd_id (fd_id_value)
1895
1896   flow_director_filter (port_id) mode IP (add|del|update) \
1897                        flow (ipv4-sctp|ipv6-sctp) \
1898                        src (src_ip_address) (src_port) \
1899                        dst (dst_ip_address) (dst_port) \
1900                        tos (tos_value) ttl (ttl_value) \
1901                        tag (verification_tag) vlan (vlan_value) \
1902                        flexbytes (flexbytes_value) (drop|fwd) \
1903                        pf|vf(vf_id) queue (queue_id) fd_id (fd_id_value)
1904
1905   flow_director_filter (port_id) mode IP (add|del|update) flow l2_payload \
1906                        ether (ethertype) flexbytes (flexbytes_value) \
1907                        (drop|fwd) pf|vf(vf_id) queue (queue_id)
1908                        fd_id (fd_id_value)
1909
1910   flow_director_filter (port_id) mode MAC-VLAN (add|del|update) \
1911                        mac (mac_address) vlan (vlan_value) \
1912                        flexbytes (flexbytes_value) (drop|fwd) \
1913                        queue (queue_id) fd_id (fd_id_value)
1914
1915   flow_director_filter (port_id) mode Tunnel (add|del|update) \
1916                        mac (mac_address) vlan (vlan_value) \
1917                        tunnel (NVGRE|VxLAN) tunnel-id (tunnel_id_value) \
1918                        flexbytes (flexbytes_value) (drop|fwd) \
1919                        queue (queue_id) fd_id (fd_id_value)
1920
1921For example, to add an ipv4-udp flow type filter::
1922
1923   testpmd> flow_director_filter 0 mode IP add flow ipv4-udp src 2.2.2.3 32 \
1924            dst 2.2.2.5 33 tos 2 ttl 40 vlan 0x1 flexbytes (0x88,0x48) \
1925            fwd pf queue 1 fd_id 1
1926
1927For example, add an ipv4-other flow type filter::
1928
1929   testpmd> flow_director_filter 0 mode IP add flow ipv4-other src 2.2.2.3 \
1930             dst 2.2.2.5 tos 2 proto 20 ttl 40 vlan 0x1 \
1931             flexbytes (0x88,0x48) fwd pf queue 1 fd_id 1
1932
1933flush_flow_director
1934~~~~~~~~~~~~~~~~~~~
1935
1936Flush all flow director filters on a device::
1937
1938   testpmd> flush_flow_director (port_id)
1939
1940Example, to flush all flow director filter on port 0::
1941
1942   testpmd> flush_flow_director 0
1943
1944flow_director_mask
1945~~~~~~~~~~~~~~~~~~
1946
1947Set flow director's input masks::
1948
1949   flow_director_mask (port_id) mode IP vlan (vlan_value) \
1950                      src_mask (ipv4_src) (ipv6_src) (src_port) \
1951                      dst_mask (ipv4_dst) (ipv6_dst) (dst_port)
1952
1953   flow_director_mask (port_id) mode MAC-VLAN vlan (vlan_value)
1954
1955   flow_director_mask (port_id) mode Tunnel vlan (vlan_value) \
1956                      mac (mac_value) tunnel-type (tunnel_type_value) \
1957                      tunnel-id (tunnel_id_value)
1958
1959Example, to set flow director mask on port 0::
1960
1961   testpmd> flow_director_mask 0 mode IP vlan 0xefff \
1962            src_mask 255.255.255.255 \
1963                FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF \
1964            dst_mask 255.255.255.255 \
1965                FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF
1966
1967flow_director_flex_mask
1968~~~~~~~~~~~~~~~~~~~~~~~
1969
1970set masks of flow director's flexible payload based on certain flow type::
1971
1972   testpmd> flow_director_flex_mask (port_id) \
1973            flow (none|ipv4-other|ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
1974                  ipv6-other|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp| \
1975                  l2_payload|all) (mask)
1976
1977Example, to set flow director's flex mask for all flow type on port 0::
1978
1979   testpmd> flow_director_flex_mask 0 flow all \
1980            (0xff,0xff,0,0,0,0,0,0,0,0,0,0,0,0,0,0)
1981
1982
1983flow_director_flex_payload
1984~~~~~~~~~~~~~~~~~~~~~~~~~~
1985
1986Configure flexible payload selection::
1987
1988   flow_director_flex_payload (port_id) (raw|l2|l3|l4) (config)
1989
1990For example, to select the first 16 bytes from the offset 4 (bytes) of packet's payload as flexible payload::
1991
1992   testpmd> flow_director_flex_payload 0 l4 \
1993            (4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19)
1994
1995get_sym_hash_ena_per_port
1996~~~~~~~~~~~~~~~~~~~~~~~~~
1997
1998Get symmetric hash enable configuration per port::
1999
2000   get_sym_hash_ena_per_port (port_id)
2001
2002For example, to get symmetric hash enable configuration of port 1::
2003
2004   testpmd> get_sym_hash_ena_per_port 1
2005
2006set_sym_hash_ena_per_port
2007~~~~~~~~~~~~~~~~~~~~~~~~~
2008
2009Set symmetric hash enable configuration per port to enable or disable::
2010
2011   set_sym_hash_ena_per_port (port_id) (enable|disable)
2012
2013For example, to set symmetric hash enable configuration of port 1 to enable::
2014
2015   testpmd> set_sym_hash_ena_per_port 1 enable
2016
2017get_hash_global_config
2018~~~~~~~~~~~~~~~~~~~~~~
2019
2020Get the global configurations of hash filters::
2021
2022   get_hash_global_config (port_id)
2023
2024For example, to get the global configurations of hash filters of port 1::
2025
2026   testpmd> get_hash_global_config 1
2027
2028set_hash_global_config
2029~~~~~~~~~~~~~~~~~~~~~~
2030
2031Set the global configurations of hash filters::
2032
2033   set_hash_global_config (port_id) (toeplitz|simple_xor|default) \
2034   (ipv4|ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp|ipv4-other|ipv6|ipv6-frag| \
2035   ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other|l2_payload) \
2036   (enable|disable)
2037
2038For example, to enable simple_xor for flow type of ipv6 on port 2::
2039
2040   testpmd> set_hash_global_config 2 simple_xor ipv6 enable
2041
2042set_hash_input_set
2043~~~~~~~~~~~~~~~~~~
2044
2045Set the input set for hash::
2046
2047   set_hash_input_set (port_id) (ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
2048   ipv4-other|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other| \
2049   l2_payload) (ovlan|ivlan|src-ipv4|dst-ipv4|src-ipv6|dst-ipv6|ipv4-tos| \
2050   ipv4-proto|ipv6-tc|ipv6-next-header|udp-src-port|udp-dst-port| \
2051   tcp-src-port|tcp-dst-port|sctp-src-port|sctp-dst-port|sctp-veri-tag| \
2052   udp-key|gre-key|fld-1st|fld-2nd|fld-3rd|fld-4th|fld-5th|fld-6th|fld-7th| \
2053   fld-8th|none) (select|add)
2054
2055For example, to add source IP to hash input set for flow type of ipv4-udp on port 0::
2056
2057   testpmd> set_hash_input_set 0 ipv4-udp src-ipv4 add
2058
2059set_fdir_input_set
2060~~~~~~~~~~~~~~~~~~
2061
2062The Flow Director filters can match the different fields for different type of packet, i.e. specific input set
2063on per flow type and the flexible payload. This command can be used to change input set for each flow type.
2064
2065Set the input set for flow director::
2066
2067   set_fdir_input_set (port_id) (ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
2068   ipv4-other|ipv6|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other| \
2069   l2_payload) (ivlan|ethertype|src-ipv4|dst-ipv4|src-ipv6|dst-ipv6|ipv4-tos| \
2070   ipv4-proto|ipv4-ttl|ipv6-tc|ipv6-next-header|ipv6-hop-limits| \
2071   tudp-src-port|udp-dst-port|cp-src-port|tcp-dst-port|sctp-src-port| \
2072   sctp-dst-port|sctp-veri-tag|none) (select|add)
2073
2074For example to add source IP to FD input set for flow type of ipv4-udp on port 0::
2075
2076   testpmd> set_fdir_input_set 0 ipv4-udp src-ipv4 add
2077
2078global_config
2079~~~~~~~~~~~~~
2080
2081Set different GRE key length for input set::
2082
2083   global_config (port_id) gre-key-len (number in bytes)
2084
2085For example to set GRE key length for input set to 4 bytes on port 0::
2086
2087   testpmd> global_config 0 gre-key-len 4
2088
2089
2090.. _testpmd_rte_flow:
2091
2092Flow rules management
2093---------------------
2094
2095Control of the generic flow API (*rte_flow*) is fully exposed through the
2096``flow`` command (validation, creation, destruction and queries).
2097
2098Considering *rte_flow* overlaps with all `Filter Functions`_, using both
2099features simultaneously may cause undefined side-effects and is therefore
2100not recommended.
2101
2102``flow`` syntax
2103~~~~~~~~~~~~~~~
2104
2105Because the ``flow`` command uses dynamic tokens to handle the large number
2106of possible flow rules combinations, its behavior differs slightly from
2107other commands, in particular:
2108
2109- Pressing *?* or the *<tab>* key displays contextual help for the current
2110  token, not that of the entire command.
2111
2112- Optional and repeated parameters are supported (provided they are listed
2113  in the contextual help).
2114
2115The first parameter stands for the operation mode. Possible operations and
2116their general syntax are described below. They are covered in detail in the
2117following sections.
2118
2119- Check whether a flow rule can be created::
2120
2121   flow validate {port_id}
2122       [group {group_id}] [priority {level}] [ingress] [egress]
2123       pattern {item} [/ {item} [...]] / end
2124       actions {action} [/ {action} [...]] / end
2125
2126- Create a flow rule::
2127
2128   flow create {port_id}
2129       [group {group_id}] [priority {level}] [ingress] [egress]
2130       pattern {item} [/ {item} [...]] / end
2131       actions {action} [/ {action} [...]] / end
2132
2133- Destroy specific flow rules::
2134
2135   flow destroy {port_id} rule {rule_id} [...]
2136
2137- Destroy all flow rules::
2138
2139   flow flush {port_id}
2140
2141- Query an existing flow rule::
2142
2143   flow query {port_id} {rule_id} {action}
2144
2145- List existing flow rules sorted by priority, filtered by group
2146  identifiers::
2147
2148   flow list {port_id} [group {group_id}] [...]
2149
2150Validating flow rules
2151~~~~~~~~~~~~~~~~~~~~~
2152
2153``flow validate`` reports whether a flow rule would be accepted by the
2154underlying device in its current state but stops short of creating it. It is
2155bound to ``rte_flow_validate()``::
2156
2157   flow validate {port_id}
2158      [group {group_id}] [priority {level}] [ingress] [egress]
2159      pattern {item} [/ {item} [...]] / end
2160      actions {action} [/ {action} [...]] / end
2161
2162If successful, it will show::
2163
2164   Flow rule validated
2165
2166Otherwise it will show an error message of the form::
2167
2168   Caught error type [...] ([...]): [...]
2169
2170This command uses the same parameters as ``flow create``, their format is
2171described in `Creating flow rules`_.
2172
2173Check whether redirecting any Ethernet packet received on port 0 to RX queue
2174index 6 is supported::
2175
2176   testpmd> flow validate 0 ingress pattern eth / end
2177      actions queue index 6 / end
2178   Flow rule validated
2179   testpmd>
2180
2181Port 0 does not support TCPv6 rules::
2182
2183   testpmd> flow validate 0 ingress pattern eth / ipv6 / tcp / end
2184      actions drop / end
2185   Caught error type 9 (specific pattern item): Invalid argument
2186   testpmd>
2187
2188Creating flow rules
2189~~~~~~~~~~~~~~~~~~~
2190
2191``flow create`` validates and creates the specified flow rule. It is bound
2192to ``rte_flow_create()``::
2193
2194   flow create {port_id}
2195      [group {group_id}] [priority {level}] [ingress] [egress]
2196      pattern {item} [/ {item} [...]] / end
2197      actions {action} [/ {action} [...]] / end
2198
2199If successful, it will return a flow rule ID usable with other commands::
2200
2201   Flow rule #[...] created
2202
2203Otherwise it will show an error message of the form::
2204
2205   Caught error type [...] ([...]): [...]
2206
2207Parameters describe in the following order:
2208
2209- Attributes (*group*, *priority*, *ingress*, *egress* tokens).
2210- A matching pattern, starting with the *pattern* token and terminated by an
2211  *end* pattern item.
2212- Actions, starting with the *actions* token and terminated by an *end*
2213  action.
2214
2215These translate directly to *rte_flow* objects provided as-is to the
2216underlying functions.
2217
2218The shortest valid definition only comprises mandatory tokens::
2219
2220   testpmd> flow create 0 pattern end actions end
2221
2222Note that PMDs may refuse rules that essentially do nothing such as this
2223one.
2224
2225**All unspecified object values are automatically initialized to 0.**
2226
2227Attributes
2228^^^^^^^^^^
2229
2230These tokens affect flow rule attributes (``struct rte_flow_attr``) and are
2231specified before the ``pattern`` token.
2232
2233- ``group {group id}``: priority group.
2234- ``priority {level}``: priority level within group.
2235- ``ingress``: rule applies to ingress traffic.
2236- ``egress``: rule applies to egress traffic.
2237
2238Each instance of an attribute specified several times overrides the previous
2239value as shown below (group 4 is used)::
2240
2241   testpmd> flow create 0 group 42 group 24 group 4 [...]
2242
2243Note that once enabled, ``ingress`` and ``egress`` cannot be disabled.
2244
2245While not specifying a direction is an error, some rules may allow both
2246simultaneously.
2247
2248Most rules affect RX therefore contain the ``ingress`` token::
2249
2250   testpmd> flow create 0 ingress pattern [...]
2251
2252Matching pattern
2253^^^^^^^^^^^^^^^^
2254
2255A matching pattern starts after the ``pattern`` token. It is made of pattern
2256items and is terminated by a mandatory ``end`` item.
2257
2258Items are named after their type (*RTE_FLOW_ITEM_TYPE_* from ``enum
2259rte_flow_item_type``).
2260
2261The ``/`` token is used as a separator between pattern items as shown
2262below::
2263
2264   testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end [...]
2265
2266Note that protocol items like these must be stacked from lowest to highest
2267layer to make sense. For instance, the following rule is either invalid or
2268unlikely to match any packet::
2269
2270   testpmd> flow create 0 ingress pattern eth / udp / ipv4 / end [...]
2271
2272More information on these restrictions can be found in the *rte_flow*
2273documentation.
2274
2275Several items support additional specification structures, for example
2276``ipv4`` allows specifying source and destination addresses as follows::
2277
2278   testpmd> flow create 0 ingress pattern eth / ipv4 src is 10.1.1.1
2279      dst is 10.2.0.0 / end [...]
2280
2281This rule matches all IPv4 traffic with the specified properties.
2282
2283In this example, ``src`` and ``dst`` are field names of the underlying
2284``struct rte_flow_item_ipv4`` object. All item properties can be specified
2285in a similar fashion.
2286
2287The ``is`` token means that the subsequent value must be matched exactly,
2288and assigns ``spec`` and ``mask`` fields in ``struct rte_flow_item``
2289accordingly. Possible assignment tokens are:
2290
2291- ``is``: match value perfectly (with full bit-mask).
2292- ``spec``: match value according to configured bit-mask.
2293- ``last``: specify upper bound to establish a range.
2294- ``mask``: specify bit-mask with relevant bits set to one.
2295- ``prefix``: generate bit-mask from a prefix length.
2296
2297These yield identical results::
2298
2299   ipv4 src is 10.1.1.1
2300
2301::
2302
2303   ipv4 src spec 10.1.1.1 src mask 255.255.255.255
2304
2305::
2306
2307   ipv4 src spec 10.1.1.1 src prefix 32
2308
2309::
2310
2311   ipv4 src is 10.1.1.1 src last 10.1.1.1 # range with a single value
2312
2313::
2314
2315   ipv4 src is 10.1.1.1 src last 0 # 0 disables range
2316
2317Inclusive ranges can be defined with ``last``::
2318
2319   ipv4 src is 10.1.1.1 src last 10.2.3.4 # 10.1.1.1 to 10.2.3.4
2320
2321Note that ``mask`` affects both ``spec`` and ``last``::
2322
2323   ipv4 src is 10.1.1.1 src last 10.2.3.4 src mask 255.255.0.0
2324      # matches 10.1.0.0 to 10.2.255.255
2325
2326Properties can be modified multiple times::
2327
2328   ipv4 src is 10.1.1.1 src is 10.1.2.3 src is 10.2.3.4 # matches 10.2.3.4
2329
2330::
2331
2332   ipv4 src is 10.1.1.1 src prefix 24 src prefix 16 # matches 10.1.0.0/16
2333
2334Pattern items
2335^^^^^^^^^^^^^
2336
2337This section lists supported pattern items and their attributes, if any.
2338
2339- ``end``: end list of pattern items.
2340
2341- ``void``: no-op pattern item.
2342
2343- ``invert``: perform actions when pattern does not match.
2344
2345- ``any``: match any protocol for the current layer.
2346
2347  - ``num {unsigned}``: number of layers covered.
2348
2349- ``pf``: match packets addressed to the physical function.
2350
2351- ``vf``: match packets addressed to a virtual function ID.
2352
2353  - ``id {unsigned}``: destination VF ID.
2354
2355- ``port``: device-specific physical port index to use.
2356
2357  - ``index {unsigned}``: physical port index.
2358
2359- ``raw``: match an arbitrary byte string.
2360
2361  - ``relative {boolean}``: look for pattern after the previous item.
2362  - ``search {boolean}``: search pattern from offset (see also limit).
2363  - ``offset {integer}``: absolute or relative offset for pattern.
2364  - ``limit {unsigned}``: search area limit for start of pattern.
2365  - ``pattern {string}``: byte string to look for.
2366
2367- ``eth``: match Ethernet header.
2368
2369  - ``dst {MAC-48}``: destination MAC.
2370  - ``src {MAC-48}``: source MAC.
2371  - ``type {unsigned}``: EtherType.
2372
2373- ``vlan``: match 802.1Q/ad VLAN tag.
2374
2375  - ``tpid {unsigned}``: tag protocol identifier.
2376  - ``tci {unsigned}``: tag control information.
2377  - ``pcp {unsigned}``: priority code point.
2378  - ``dei {unsigned}``: drop eligible indicator.
2379  - ``vid {unsigned}``: VLAN identifier.
2380
2381- ``ipv4``: match IPv4 header.
2382
2383  - ``tos {unsigned}``: type of service.
2384  - ``ttl {unsigned}``: time to live.
2385  - ``proto {unsigned}``: next protocol ID.
2386  - ``src {ipv4 address}``: source address.
2387  - ``dst {ipv4 address}``: destination address.
2388
2389- ``ipv6``: match IPv6 header.
2390
2391  - ``tc {unsigned}``: traffic class.
2392  - ``flow {unsigned}``: flow label.
2393  - ``proto {unsigned}``: protocol (next header).
2394  - ``hop {unsigned}``: hop limit.
2395  - ``src {ipv6 address}``: source address.
2396  - ``dst {ipv6 address}``: destination address.
2397
2398- ``icmp``: match ICMP header.
2399
2400  - ``type {unsigned}``: ICMP packet type.
2401  - ``code {unsigned}``: ICMP packet code.
2402
2403- ``udp``: match UDP header.
2404
2405  - ``src {unsigned}``: UDP source port.
2406  - ``dst {unsigned}``: UDP destination port.
2407
2408- ``tcp``: match TCP header.
2409
2410  - ``src {unsigned}``: TCP source port.
2411  - ``dst {unsigned}``: TCP destination port.
2412
2413- ``sctp``: match SCTP header.
2414
2415  - ``src {unsigned}``: SCTP source port.
2416  - ``dst {unsigned}``: SCTP destination port.
2417  - ``tag {unsigned}``: validation tag.
2418  - ``cksum {unsigned}``: checksum.
2419
2420- ``vxlan``: match VXLAN header.
2421
2422  - ``vni {unsigned}``: VXLAN identifier.
2423
2424Actions list
2425^^^^^^^^^^^^
2426
2427A list of actions starts after the ``actions`` token in the same fashion as
2428`Matching pattern`_; actions are separated by ``/`` tokens and the list is
2429terminated by a mandatory ``end`` action.
2430
2431Actions are named after their type (*RTE_FLOW_ACTION_TYPE_* from ``enum
2432rte_flow_action_type``).
2433
2434Dropping all incoming UDPv4 packets can be expressed as follows::
2435
2436   testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
2437      actions drop / end
2438
2439Several actions have configurable properties which must be specified when
2440there is no valid default value. For example, ``queue`` requires a target
2441queue index.
2442
2443This rule redirects incoming UDPv4 traffic to queue index 6::
2444
2445   testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
2446      actions queue index 6 / end
2447
2448While this one could be rejected by PMDs (unspecified queue index)::
2449
2450   testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
2451      actions queue / end
2452
2453As defined by *rte_flow*, the list is not ordered, all actions of a given
2454rule are performed simultaneously. These are equivalent::
2455
2456   queue index 6 / void / mark id 42 / end
2457
2458::
2459
2460   void / mark id 42 / queue index 6 / end
2461
2462All actions in a list should have different types, otherwise only the last
2463action of a given type is taken into account::
2464
2465   queue index 4 / queue index 5 / queue index 6 / end # will use queue 6
2466
2467::
2468
2469   drop / drop / drop / end # drop is performed only once
2470
2471::
2472
2473   mark id 42 / queue index 3 / mark id 24 / end # mark will be 24
2474
2475Considering they are performed simultaneously, opposite and overlapping
2476actions can sometimes be combined when the end result is unambiguous::
2477
2478   drop / queue index 6 / end # drop has no effect
2479
2480::
2481
2482   drop / dup index 6 / end # same as above
2483
2484::
2485
2486   queue index 6 / rss queues 6 7 8 / end # queue has no effect
2487
2488::
2489
2490   drop / passthru / end # drop has no effect
2491
2492Note that PMDs may still refuse such combinations.
2493
2494Actions
2495^^^^^^^
2496
2497This section lists supported actions and their attributes, if any.
2498
2499- ``end``: end list of actions.
2500
2501- ``void``: no-op action.
2502
2503- ``passthru``: let subsequent rule process matched packets.
2504
2505- ``mark``: attach 32 bit value to packets.
2506
2507  - ``id {unsigned}``: 32 bit value to return with packets.
2508
2509- ``flag``: flag packets.
2510
2511- ``queue``: assign packets to a given queue index.
2512
2513  - ``index {unsigned}``: queue index to use.
2514
2515- ``drop``: drop packets (note: passthru has priority).
2516
2517- ``count``: enable counters for this rule.
2518
2519- ``dup``: duplicate packets to a given queue index.
2520
2521  - ``index {unsigned}``: queue index to duplicate packets to.
2522
2523- ``rss``: spread packets among several queues.
2524
2525  - ``queues [{unsigned} [...]] end``: queue indices to use.
2526
2527- ``pf``: redirect packets to physical device function.
2528
2529- ``vf``: redirect packets to virtual device function.
2530
2531  - ``original {boolean}``: use original VF ID if possible.
2532  - ``id {unsigned}``: VF ID to redirect packets to.
2533
2534Destroying flow rules
2535~~~~~~~~~~~~~~~~~~~~~
2536
2537``flow destroy`` destroys one or more rules from their rule ID (as returned
2538by ``flow create``), this command calls ``rte_flow_destroy()`` as many
2539times as necessary::
2540
2541   flow destroy {port_id} rule {rule_id} [...]
2542
2543If successful, it will show::
2544
2545   Flow rule #[...] destroyed
2546
2547It does not report anything for rule IDs that do not exist. The usual error
2548message is shown when a rule cannot be destroyed::
2549
2550   Caught error type [...] ([...]): [...]
2551
2552``flow flush`` destroys all rules on a device and does not take extra
2553arguments. It is bound to ``rte_flow_flush()``::
2554
2555   flow flush {port_id}
2556
2557Any errors are reported as above.
2558
2559Creating several rules and destroying them::
2560
2561   testpmd> flow create 0 ingress pattern eth / ipv6 / end
2562      actions queue index 2 / end
2563   Flow rule #0 created
2564   testpmd> flow create 0 ingress pattern eth / ipv4 / end
2565      actions queue index 3 / end
2566   Flow rule #1 created
2567   testpmd> flow destroy 0 rule 0 rule 1
2568   Flow rule #1 destroyed
2569   Flow rule #0 destroyed
2570   testpmd>
2571
2572The same result can be achieved using ``flow flush``::
2573
2574   testpmd> flow create 0 ingress pattern eth / ipv6 / end
2575      actions queue index 2 / end
2576   Flow rule #0 created
2577   testpmd> flow create 0 ingress pattern eth / ipv4 / end
2578      actions queue index 3 / end
2579   Flow rule #1 created
2580   testpmd> flow flush 0
2581   testpmd>
2582
2583Non-existent rule IDs are ignored::
2584
2585   testpmd> flow create 0 ingress pattern eth / ipv6 / end
2586      actions queue index 2 / end
2587   Flow rule #0 created
2588   testpmd> flow create 0 ingress pattern eth / ipv4 / end
2589      actions queue index 3 / end
2590   Flow rule #1 created
2591   testpmd> flow destroy 0 rule 42 rule 10 rule 2
2592   testpmd>
2593   testpmd> flow destroy 0 rule 0
2594   Flow rule #0 destroyed
2595   testpmd>
2596
2597Querying flow rules
2598~~~~~~~~~~~~~~~~~~~
2599
2600``flow query`` queries a specific action of a flow rule having that
2601ability. Such actions collect information that can be reported using this
2602command. It is bound to ``rte_flow_query()``::
2603
2604   flow query {port_id} {rule_id} {action}
2605
2606If successful, it will display either the retrieved data for known actions
2607or the following message::
2608
2609   Cannot display result for action type [...] ([...])
2610
2611Otherwise, it will complain either that the rule does not exist or that some
2612error occurred::
2613
2614   Flow rule #[...] not found
2615
2616::
2617
2618   Caught error type [...] ([...]): [...]
2619
2620Currently only the ``count`` action is supported. This action reports the
2621number of packets that hit the flow rule and the total number of bytes. Its
2622output has the following format::
2623
2624   count:
2625    hits_set: [...] # whether "hits" contains a valid value
2626    bytes_set: [...] # whether "bytes" contains a valid value
2627    hits: [...] # number of packets
2628    bytes: [...] # number of bytes
2629
2630Querying counters for TCPv6 packets redirected to queue 6::
2631
2632   testpmd> flow create 0 ingress pattern eth / ipv6 / tcp / end
2633      actions queue index 6 / count / end
2634   Flow rule #4 created
2635   testpmd> flow query 0 4 count
2636   count:
2637    hits_set: 1
2638    bytes_set: 0
2639    hits: 386446
2640    bytes: 0
2641   testpmd>
2642
2643Listing flow rules
2644~~~~~~~~~~~~~~~~~~
2645
2646``flow list`` lists existing flow rules sorted by priority and optionally
2647filtered by group identifiers::
2648
2649   flow list {port_id} [group {group_id}] [...]
2650
2651This command only fails with the following message if the device does not
2652exist::
2653
2654   Invalid port [...]
2655
2656Output consists of a header line followed by a short description of each
2657flow rule, one per line. There is no output at all when no flow rules are
2658configured on the device::
2659
2660   ID      Group   Prio    Attr    Rule
2661   [...]   [...]   [...]   [...]   [...]
2662
2663``Attr`` column flags:
2664
2665- ``i`` for ``ingress``.
2666- ``e`` for ``egress``.
2667
2668Creating several flow rules and listing them::
2669
2670   testpmd> flow create 0 ingress pattern eth / ipv4 / end
2671      actions queue index 6 / end
2672   Flow rule #0 created
2673   testpmd> flow create 0 ingress pattern eth / ipv6 / end
2674      actions queue index 2 / end
2675   Flow rule #1 created
2676   testpmd> flow create 0 priority 5 ingress pattern eth / ipv4 / udp / end
2677      actions rss queues 6 7 8 end / end
2678   Flow rule #2 created
2679   testpmd> flow list 0
2680   ID      Group   Prio    Attr    Rule
2681   0       0       0       i-      ETH IPV4 => QUEUE
2682   1       0       0       i-      ETH IPV6 => QUEUE
2683   2       0       5       i-      ETH IPV4 UDP => RSS
2684   testpmd>
2685
2686Rules are sorted by priority (i.e. group ID first, then priority level)::
2687
2688   testpmd> flow list 1
2689   ID      Group   Prio    Attr    Rule
2690   0       0       0       i-      ETH => COUNT
2691   6       0       500     i-      ETH IPV6 TCP => DROP COUNT
2692   5       0       1000    i-      ETH IPV6 ICMP => QUEUE
2693   1       24      0       i-      ETH IPV4 UDP => QUEUE
2694   4       24      10      i-      ETH IPV4 TCP => DROP
2695   3       24      20      i-      ETH IPV4 => DROP
2696   2       24      42      i-      ETH IPV4 UDP => QUEUE
2697   7       63      0       i-      ETH IPV6 UDP VXLAN => MARK QUEUE
2698   testpmd>
2699
2700Output can be limited to specific groups::
2701
2702   testpmd> flow list 1 group 0 group 63
2703   ID      Group   Prio    Attr    Rule
2704   0       0       0       i-      ETH => COUNT
2705   6       0       500     i-      ETH IPV6 TCP => DROP COUNT
2706   5       0       1000    i-      ETH IPV6 ICMP => QUEUE
2707   7       63      0       i-      ETH IPV6 UDP VXLAN => MARK QUEUE
2708   testpmd>
2709