xref: /dpdk/doc/guides/testpmd_app_ug/testpmd_funcs.rst (revision 982a8970717dd66dbd43bb2fa9ea17fb480103c6)
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 allmulticast (for VF)
865~~~~~~~~~~~~~~~~~~~~~~~~~
866
867Set the multicast promiscuous mode for a VF from PF.
868It's supported by Intel i40e NICs now.
869In promiscuous mode packets are not dropped if they aren't for the specified MAC address::
870
871   testpmd> set vf allmulti (port_id) (vf_id) (on|off)
872
873set flow_ctrl rx
874~~~~~~~~~~~~~~~~
875
876Set the link flow control parameter on a port::
877
878   testpmd> set flow_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \
879            (pause_time) (send_xon) mac_ctrl_frame_fwd (on|off) \
880	    autoneg (on|off) (port_id)
881
882Where:
883
884* ``high_water`` (integer): High threshold value to trigger XOFF.
885
886* ``low_water`` (integer): Low threshold value to trigger XON.
887
888* ``pause_time`` (integer): Pause quota in the Pause frame.
889
890* ``send_xon`` (0/1): Send XON frame.
891
892* ``mac_ctrl_frame_fwd``: Enable receiving MAC control frames.
893
894* ``autoneg``: Change the auto-negotiation parameter.
895
896set pfc_ctrl rx
897~~~~~~~~~~~~~~~
898
899Set the priority flow control parameter on a port::
900
901   testpmd> set pfc_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \
902            (pause_time) (priority) (port_id)
903
904Where:
905
906* ``high_water`` (integer): High threshold value.
907
908* ``low_water`` (integer): Low threshold value.
909
910* ``pause_time`` (integer): Pause quota in the Pause frame.
911
912* ``priority`` (0-7): VLAN User Priority.
913
914set stat_qmap
915~~~~~~~~~~~~~
916
917Set statistics mapping (qmapping 0..15) for RX/TX queue on port::
918
919   testpmd> set stat_qmap (tx|rx) (port_id) (queue_id) (qmapping)
920
921For example, to set rx queue 2 on port 0 to mapping 5::
922
923   testpmd>set stat_qmap rx 0 2 5
924
925set port - rx/tx (for VF)
926~~~~~~~~~~~~~~~~~~~~~~~~~
927
928Set VF receive/transmit from a port::
929
930   testpmd> set port (port_id) vf (vf_id) (rx|tx) (on|off)
931
932set port - mac address filter (for VF)
933~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
934
935Add/Remove unicast or multicast MAC addr filter for a VF::
936
937   testpmd> set port (port_id) vf (vf_id) (mac_addr) \
938            (exact-mac|exact-mac-vlan|hashmac|hashmac-vlan) (on|off)
939
940set port - rx mode(for VF)
941~~~~~~~~~~~~~~~~~~~~~~~~~~
942
943Set the VF receive mode of a port::
944
945   testpmd> set port (port_id) vf (vf_id) \
946            rxmode (AUPE|ROPE|BAM|MPE) (on|off)
947
948The available receive modes are:
949
950* ``AUPE``: Accepts untagged VLAN.
951
952* ``ROPE``: Accepts unicast hash.
953
954* ``BAM``: Accepts broadcast packets.
955
956* ``MPE``: Accepts all multicast packets.
957
958set port - tx_rate (for Queue)
959~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
960
961Set TX rate limitation for a queue on a port::
962
963   testpmd> set port (port_id) queue (queue_id) rate (rate_value)
964
965set port - tx_rate (for VF)
966~~~~~~~~~~~~~~~~~~~~~~~~~~~
967
968Set TX rate limitation for queues in VF on a port::
969
970   testpmd> set port (port_id) vf (vf_id) rate (rate_value) queue_mask (queue_mask)
971
972set port - mirror rule
973~~~~~~~~~~~~~~~~~~~~~~
974
975Set pool or vlan type mirror rule for a port::
976
977   testpmd> set port (port_id) mirror-rule (rule_id) \
978            (pool-mirror-up|pool-mirror-down|vlan-mirror) \
979            (poolmask|vlanid[,vlanid]*) dst-pool (pool_id) (on|off)
980
981Set link mirror rule for a port::
982
983   testpmd> set port (port_id) mirror-rule (rule_id) \
984           (uplink-mirror|downlink-mirror) dst-pool (pool_id) (on|off)
985
986For example to enable mirror traffic with vlan 0,1 to pool 0::
987
988   set port 0 mirror-rule 0 vlan-mirror 0,1 dst-pool 0 on
989
990reset port - mirror rule
991~~~~~~~~~~~~~~~~~~~~~~~~
992
993Reset a mirror rule for a port::
994
995   testpmd> reset port (port_id) mirror-rule (rule_id)
996
997set flush_rx
998~~~~~~~~~~~~
999
1000Set the flush on RX streams before forwarding.
1001The default is flush ``on``.
1002Mainly used with PCAP drivers to turn off the default behavior of flushing the first 512 packets on RX streams::
1003
1004   testpmd> set flush_rx off
1005
1006set bypass mode
1007~~~~~~~~~~~~~~~
1008
1009Set the bypass mode for the lowest port on bypass enabled NIC::
1010
1011   testpmd> set bypass mode (normal|bypass|isolate) (port_id)
1012
1013set bypass event
1014~~~~~~~~~~~~~~~~
1015
1016Set the event required to initiate specified bypass mode for the lowest port on a bypass enabled::
1017
1018   testpmd> set bypass event (timeout|os_on|os_off|power_on|power_off) \
1019            mode (normal|bypass|isolate) (port_id)
1020
1021Where:
1022
1023* ``timeout``: Enable bypass after watchdog timeout.
1024
1025* ``os_on``: Enable bypass when OS/board is powered on.
1026
1027* ``os_off``: Enable bypass when OS/board is powered off.
1028
1029* ``power_on``: Enable bypass when power supply is turned on.
1030
1031* ``power_off``: Enable bypass when power supply is turned off.
1032
1033
1034set bypass timeout
1035~~~~~~~~~~~~~~~~~~
1036
1037Set the bypass watchdog timeout to ``n`` seconds where 0 = instant::
1038
1039   testpmd> set bypass timeout (0|1.5|2|3|4|8|16|32)
1040
1041show bypass config
1042~~~~~~~~~~~~~~~~~~
1043
1044Show the bypass configuration for a bypass enabled NIC using the lowest port on the NIC::
1045
1046   testpmd> show bypass config (port_id)
1047
1048set link up
1049~~~~~~~~~~~
1050
1051Set link up for a port::
1052
1053   testpmd> set link-up port (port id)
1054
1055set link down
1056~~~~~~~~~~~~~
1057
1058Set link down for a port::
1059
1060   testpmd> set link-down port (port id)
1061
1062E-tag set
1063~~~~~~~~~
1064
1065Enable E-tag insertion for a VF on a port::
1066
1067   testpmd> E-tag set insertion on port-tag-id (value) port (port_id) vf (vf_id)
1068
1069Disable E-tag insertion for a VF on a port::
1070
1071   testpmd> E-tag set insertion off port (port_id) vf (vf_id)
1072
1073Enable/disable E-tag stripping on a port::
1074
1075   testpmd> E-tag set stripping (on|off) port (port_id)
1076
1077Enable/disable E-tag based forwarding on a port::
1078
1079   testpmd> E-tag set forwarding (on|off) port (port_id)
1080
1081Add an E-tag forwarding filter on a port::
1082
1083   testpmd> E-tag set filter add e-tag-id (value) dst-pool (pool_id) port (port_id)
1084
1085Delete an E-tag forwarding filter on a port::
1086   testpmd> E-tag set filter del e-tag-id (value) port (port_id)
1087
1088
1089Port Functions
1090--------------
1091
1092The following sections show functions for configuring ports.
1093
1094.. note::
1095
1096   Port configuration changes only become active when forwarding is started/restarted.
1097
1098port attach
1099~~~~~~~~~~~
1100
1101Attach a port specified by pci address or virtual device args::
1102
1103   testpmd> port attach (identifier)
1104
1105To attach a new pci device, the device should be recognized by kernel first.
1106Then it should be moved under DPDK management.
1107Finally the port can be attached to testpmd.
1108
1109For example, to move a pci device using ixgbe under DPDK management:
1110
1111.. code-block:: console
1112
1113   # Check the status of the available devices.
1114   ./usertools/dpdk-devbind.py --status
1115
1116   Network devices using DPDK-compatible driver
1117   ============================================
1118   <none>
1119
1120   Network devices using kernel driver
1121   ===================================
1122   0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused=
1123
1124
1125   # Bind the device to igb_uio.
1126   sudo ./usertools/dpdk-devbind.py -b igb_uio 0000:0a:00.0
1127
1128
1129   # Recheck the status of the devices.
1130   ./usertools/dpdk-devbind.py --status
1131   Network devices using DPDK-compatible driver
1132   ============================================
1133   0000:0a:00.0 '82599ES 10-Gigabit' drv=igb_uio unused=
1134
1135To attach a port created by virtual device, above steps are not needed.
1136
1137For example, to attach a port whose pci address is 0000:0a:00.0.
1138
1139.. code-block:: console
1140
1141   testpmd> port attach 0000:0a:00.0
1142   Attaching a new port...
1143   EAL: PCI device 0000:0a:00.0 on NUMA socket -1
1144   EAL:   probe driver: 8086:10fb rte_ixgbe_pmd
1145   EAL:   PCI memory mapped at 0x7f83bfa00000
1146   EAL:   PCI memory mapped at 0x7f83bfa80000
1147   PMD: eth_ixgbe_dev_init(): MAC: 2, PHY: 18, SFP+: 5
1148   PMD: eth_ixgbe_dev_init(): port 0 vendorID=0x8086 deviceID=0x10fb
1149   Port 0 is attached. Now total ports is 1
1150   Done
1151
1152For example, to attach a port created by pcap PMD.
1153
1154.. code-block:: console
1155
1156   testpmd> port attach net_pcap0
1157   Attaching a new port...
1158   PMD: Initializing pmd_pcap for net_pcap0
1159   PMD: Creating pcap-backed ethdev on numa socket 0
1160   Port 0 is attached. Now total ports is 1
1161   Done
1162
1163In this case, identifier is ``net_pcap0``.
1164This identifier format is the same as ``--vdev`` format of DPDK applications.
1165
1166For example, to re-attach a bonded port which has been previously detached,
1167the mode and slave parameters must be given.
1168
1169.. code-block:: console
1170
1171   testpmd> port attach net_bond_0,mode=0,slave=1
1172   Attaching a new port...
1173   EAL: Initializing pmd_bond for net_bond_0
1174   EAL: Create bonded device net_bond_0 on port 0 in mode 0 on socket 0.
1175   Port 0 is attached. Now total ports is 1
1176   Done
1177
1178
1179port detach
1180~~~~~~~~~~~
1181
1182Detach a specific port::
1183
1184   testpmd> port detach (port_id)
1185
1186Before detaching a port, the port should be stopped and closed.
1187
1188For example, to detach a pci device port 0.
1189
1190.. code-block:: console
1191
1192   testpmd> port stop 0
1193   Stopping ports...
1194   Done
1195   testpmd> port close 0
1196   Closing ports...
1197   Done
1198
1199   testpmd> port detach 0
1200   Detaching a port...
1201   EAL: PCI device 0000:0a:00.0 on NUMA socket -1
1202   EAL:   remove driver: 8086:10fb rte_ixgbe_pmd
1203   EAL:   PCI memory unmapped at 0x7f83bfa00000
1204   EAL:   PCI memory unmapped at 0x7f83bfa80000
1205   Done
1206
1207
1208For example, to detach a virtual device port 0.
1209
1210.. code-block:: console
1211
1212   testpmd> port stop 0
1213   Stopping ports...
1214   Done
1215   testpmd> port close 0
1216   Closing ports...
1217   Done
1218
1219   testpmd> port detach 0
1220   Detaching a port...
1221   PMD: Closing pcap ethdev on numa socket 0
1222   Port 'net_pcap0' is detached. Now total ports is 0
1223   Done
1224
1225To remove a pci device completely from the system, first detach the port from testpmd.
1226Then the device should be moved under kernel management.
1227Finally the device can be removed using kernel pci hotplug functionality.
1228
1229For example, to move a pci device under kernel management:
1230
1231.. code-block:: console
1232
1233   sudo ./usertools/dpdk-devbind.py -b ixgbe 0000:0a:00.0
1234
1235   ./usertools/dpdk-devbind.py --status
1236
1237   Network devices using DPDK-compatible driver
1238   ============================================
1239   <none>
1240
1241   Network devices using kernel driver
1242   ===================================
1243   0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused=igb_uio
1244
1245To remove a port created by a virtual device, above steps are not needed.
1246
1247port start
1248~~~~~~~~~~
1249
1250Start all ports or a specific port::
1251
1252   testpmd> port start (port_id|all)
1253
1254port stop
1255~~~~~~~~~
1256
1257Stop all ports or a specific port::
1258
1259   testpmd> port stop (port_id|all)
1260
1261port close
1262~~~~~~~~~~
1263
1264Close all ports or a specific port::
1265
1266   testpmd> port close (port_id|all)
1267
1268port start/stop queue
1269~~~~~~~~~~~~~~~~~~~~~
1270
1271Start/stop a rx/tx queue on a specific port::
1272
1273   testpmd> port (port_id) (rxq|txq) (queue_id) (start|stop)
1274
1275Only take effect when port is started.
1276
1277port config - speed
1278~~~~~~~~~~~~~~~~~~~
1279
1280Set the speed and duplex mode for all ports or a specific port::
1281
1282   testpmd> port config (port_id|all) speed (10|100|1000|10000|25000|40000|50000|100000|auto) \
1283            duplex (half|full|auto)
1284
1285port config - queues/descriptors
1286~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1287
1288Set number of queues/descriptors for rxq, txq, rxd and txd::
1289
1290   testpmd> port config all (rxq|txq|rxd|txd) (value)
1291
1292This is equivalent to the ``--rxq``, ``--txq``, ``--rxd`` and ``--txd`` command-line options.
1293
1294port config - max-pkt-len
1295~~~~~~~~~~~~~~~~~~~~~~~~~
1296
1297Set the maximum packet length::
1298
1299   testpmd> port config all max-pkt-len (value)
1300
1301This is equivalent to the ``--max-pkt-len`` command-line option.
1302
1303port config - CRC Strip
1304~~~~~~~~~~~~~~~~~~~~~~~
1305
1306Set hardware CRC stripping on or off for all ports::
1307
1308   testpmd> port config all crc-strip (on|off)
1309
1310CRC stripping is off by default.
1311
1312The ``on`` option is equivalent to the ``--crc-strip`` command-line option.
1313
1314port config - scatter
1315~~~~~~~~~~~~~~~~~~~~~~~
1316
1317Set RX scatter mode on or off for all ports::
1318
1319   testpmd> port config all scatter (on|off)
1320
1321RX scatter mode is off by default.
1322
1323The ``on`` option is equivalent to the ``--enable-scatter`` command-line option.
1324
1325port config - TX queue flags
1326~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1327
1328Set a hexadecimal bitmap of TX queue flags for all ports::
1329
1330   testpmd> port config all txqflags value
1331
1332This command is equivalent to the ``--txqflags`` command-line option.
1333
1334port config - RX Checksum
1335~~~~~~~~~~~~~~~~~~~~~~~~~
1336
1337Set hardware RX checksum offload to on or off for all ports::
1338
1339   testpmd> port config all rx-cksum (on|off)
1340
1341Checksum offload is off by default.
1342
1343The ``on`` option is equivalent to the ``--enable-rx-cksum`` command-line option.
1344
1345port config - VLAN
1346~~~~~~~~~~~~~~~~~~
1347
1348Set hardware VLAN on or off for all ports::
1349
1350   testpmd> port config all hw-vlan (on|off)
1351
1352Hardware VLAN is on by default.
1353
1354The ``off`` option is equivalent to the ``--disable-hw-vlan`` command-line option.
1355
1356port config - VLAN filter
1357~~~~~~~~~~~~~~~~~~~~~~~~~
1358
1359Set hardware VLAN filter on or off for all ports::
1360
1361   testpmd> port config all hw-vlan-filter (on|off)
1362
1363Hardware VLAN filter is on by default.
1364
1365The ``off`` option is equivalent to the ``--disable-hw-vlan-filter`` command-line option.
1366
1367port config - VLAN strip
1368~~~~~~~~~~~~~~~~~~~~~~~~
1369
1370Set hardware VLAN strip on or off for all ports::
1371
1372   testpmd> port config all hw-vlan-strip (on|off)
1373
1374Hardware VLAN strip is on by default.
1375
1376The ``off`` option is equivalent to the ``--disable-hw-vlan-strip`` command-line option.
1377
1378port config - VLAN extend
1379~~~~~~~~~~~~~~~~~~~~~~~~~
1380
1381Set hardware VLAN extend on or off for all ports::
1382
1383   testpmd> port config all hw-vlan-extend (on|off)
1384
1385Hardware VLAN extend is off by default.
1386
1387The ``off`` option is equivalent to the ``--disable-hw-vlan-extend`` command-line option.
1388
1389port config - Drop Packets
1390~~~~~~~~~~~~~~~~~~~~~~~~~~
1391
1392Set packet drop for packets with no descriptors on or off for all ports::
1393
1394   testpmd> port config all drop-en (on|off)
1395
1396Packet dropping for packets with no descriptors is off by default.
1397
1398The ``on`` option is equivalent to the ``--enable-drop-en`` command-line option.
1399
1400port config - RSS
1401~~~~~~~~~~~~~~~~~
1402
1403Set the RSS (Receive Side Scaling) mode on or off::
1404
1405   testpmd> port config all rss (all|ip|tcp|udp|sctp|ether|port|vxlan|geneve|nvgre|none)
1406
1407RSS is on by default.
1408
1409The ``none`` option is equivalent to the ``--disable-rss`` command-line option.
1410
1411port config - RSS Reta
1412~~~~~~~~~~~~~~~~~~~~~~
1413
1414Set the RSS (Receive Side Scaling) redirection table::
1415
1416   testpmd> port config all rss reta (hash,queue)[,(hash,queue)]
1417
1418port config - DCB
1419~~~~~~~~~~~~~~~~~
1420
1421Set the DCB mode for an individual port::
1422
1423   testpmd> port config (port_id) dcb vt (on|off) (traffic_class) pfc (on|off)
1424
1425The traffic class should be 4 or 8.
1426
1427port config - Burst
1428~~~~~~~~~~~~~~~~~~~
1429
1430Set the number of packets per burst::
1431
1432   testpmd> port config all burst (value)
1433
1434This is equivalent to the ``--burst`` command-line option.
1435
1436port config - Threshold
1437~~~~~~~~~~~~~~~~~~~~~~~
1438
1439Set thresholds for TX/RX queues::
1440
1441   testpmd> port config all (threshold) (value)
1442
1443Where the threshold type can be:
1444
1445* ``txpt:`` Set the prefetch threshold register of the TX rings, 0 <= value <= 255.
1446
1447* ``txht:`` Set the host threshold register of the TX rings, 0 <= value <= 255.
1448
1449* ``txwt:`` Set the write-back threshold register of the TX rings, 0 <= value <= 255.
1450
1451* ``rxpt:`` Set the prefetch threshold register of the RX rings, 0 <= value <= 255.
1452
1453* ``rxht:`` Set the host threshold register of the RX rings, 0 <= value <= 255.
1454
1455* ``rxwt:`` Set the write-back threshold register of the RX rings, 0 <= value <= 255.
1456
1457* ``txfreet:`` Set the transmit free threshold of the TX rings, 0 <= value <= txd.
1458
1459* ``rxfreet:`` Set the transmit free threshold of the RX rings, 0 <= value <= rxd.
1460
1461* ``txrst:`` Set the transmit RS bit threshold of TX rings, 0 <= value <= txd.
1462
1463These threshold options are also available from the command-line.
1464
1465port config - E-tag
1466~~~~~~~~~~~~~~~~~~~
1467
1468Set the value of ether-type for E-tag::
1469
1470   testpmd> port config (port_id|all) l2-tunnel E-tag ether-type (value)
1471
1472Enable/disable the E-tag support::
1473
1474   testpmd> port config (port_id|all) l2-tunnel E-tag (enable|disable)
1475
1476
1477Link Bonding Functions
1478----------------------
1479
1480The Link Bonding functions make it possible to dynamically create and
1481manage link bonding devices from within testpmd interactive prompt.
1482
1483create bonded device
1484~~~~~~~~~~~~~~~~~~~~
1485
1486Create a new bonding device::
1487
1488   testpmd> create bonded device (mode) (socket)
1489
1490For example, to create a bonded device in mode 1 on socket 0::
1491
1492   testpmd> create bonded 1 0
1493   created new bonded device (port X)
1494
1495add bonding slave
1496~~~~~~~~~~~~~~~~~
1497
1498Adds Ethernet device to a Link Bonding device::
1499
1500   testpmd> add bonding slave (slave id) (port id)
1501
1502For example, to add Ethernet device (port 6) to a Link Bonding device (port 10)::
1503
1504   testpmd> add bonding slave 6 10
1505
1506
1507remove bonding slave
1508~~~~~~~~~~~~~~~~~~~~
1509
1510Removes an Ethernet slave device from a Link Bonding device::
1511
1512   testpmd> remove bonding slave (slave id) (port id)
1513
1514For example, to remove Ethernet slave device (port 6) to a Link Bonding device (port 10)::
1515
1516   testpmd> remove bonding slave 6 10
1517
1518set bonding mode
1519~~~~~~~~~~~~~~~~
1520
1521Set the Link Bonding mode of a Link Bonding device::
1522
1523   testpmd> set bonding mode (value) (port id)
1524
1525For example, to set the bonding mode of a Link Bonding device (port 10) to broadcast (mode 3)::
1526
1527   testpmd> set bonding mode 3 10
1528
1529set bonding primary
1530~~~~~~~~~~~~~~~~~~~
1531
1532Set an Ethernet slave device as the primary device on a Link Bonding device::
1533
1534   testpmd> set bonding primary (slave id) (port id)
1535
1536For example, to set the Ethernet slave device (port 6) as the primary port of a Link Bonding device (port 10)::
1537
1538   testpmd> set bonding primary 6 10
1539
1540set bonding mac
1541~~~~~~~~~~~~~~~
1542
1543Set the MAC address of a Link Bonding device::
1544
1545   testpmd> set bonding mac (port id) (mac)
1546
1547For example, to set the MAC address of a Link Bonding device (port 10) to 00:00:00:00:00:01::
1548
1549   testpmd> set bonding mac 10 00:00:00:00:00:01
1550
1551set bonding xmit_balance_policy
1552~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1553
1554Set the transmission policy for a Link Bonding device when it is in Balance XOR mode::
1555
1556   testpmd> set bonding xmit_balance_policy (port_id) (l2|l23|l34)
1557
1558For example, set a Link Bonding device (port 10) to use a balance policy of layer 3+4 (IP addresses & UDP ports)::
1559
1560   testpmd> set bonding xmit_balance_policy 10 l34
1561
1562
1563set bonding mon_period
1564~~~~~~~~~~~~~~~~~~~~~~
1565
1566Set the link status monitoring polling period in milliseconds for a bonding device.
1567
1568This adds support for PMD slave devices which do not support link status interrupts.
1569When the mon_period is set to a value greater than 0 then all PMD's which do not support
1570link status ISR will be queried every polling interval to check if their link status has changed::
1571
1572   testpmd> set bonding mon_period (port_id) (value)
1573
1574For example, to set the link status monitoring polling period of bonded device (port 5) to 150ms::
1575
1576   testpmd> set bonding mon_period 5 150
1577
1578
1579show bonding config
1580~~~~~~~~~~~~~~~~~~~
1581
1582Show the current configuration of a Link Bonding device::
1583
1584   testpmd> show bonding config (port id)
1585
1586For example,
1587to show the configuration a Link Bonding device (port 9) with 3 slave devices (1, 3, 4)
1588in balance mode with a transmission policy of layer 2+3::
1589
1590   testpmd> show bonding config 9
1591        Bonding mode: 2
1592        Balance Xmit Policy: BALANCE_XMIT_POLICY_LAYER23
1593        Slaves (3): [1 3 4]
1594        Active Slaves (3): [1 3 4]
1595        Primary: [3]
1596
1597
1598Register Functions
1599------------------
1600
1601The Register Functions can be used to read from and write to registers on the network card referenced by a port number.
1602This is mainly useful for debugging purposes.
1603Reference should be made to the appropriate datasheet for the network card for details on the register addresses
1604and fields that can be accessed.
1605
1606read reg
1607~~~~~~~~
1608
1609Display the value of a port register::
1610
1611   testpmd> read reg (port_id) (address)
1612
1613For example, to examine the Flow Director control register (FDIRCTL, 0x0000EE000) on an Intel 82599 10 GbE Controller::
1614
1615   testpmd> read reg 0 0xEE00
1616   port 0 PCI register at offset 0xEE00: 0x4A060029 (1241907241)
1617
1618read regfield
1619~~~~~~~~~~~~~
1620
1621Display a port register bit field::
1622
1623   testpmd> read regfield (port_id) (address) (bit_x) (bit_y)
1624
1625For example, reading the lowest two bits from the register in the example above::
1626
1627   testpmd> read regfield 0 0xEE00 0 1
1628   port 0 PCI register at offset 0xEE00: bits[0, 1]=0x1 (1)
1629
1630read regbit
1631~~~~~~~~~~~
1632
1633Display a single port register bit::
1634
1635   testpmd> read regbit (port_id) (address) (bit_x)
1636
1637For example, reading the lowest bit from the register in the example above::
1638
1639   testpmd> read regbit 0 0xEE00 0
1640   port 0 PCI register at offset 0xEE00: bit 0=1
1641
1642write reg
1643~~~~~~~~~
1644
1645Set the value of a port register::
1646
1647   testpmd> write reg (port_id) (address) (value)
1648
1649For example, to clear a register::
1650
1651   testpmd> write reg 0 0xEE00 0x0
1652   port 0 PCI register at offset 0xEE00: 0x00000000 (0)
1653
1654write regfield
1655~~~~~~~~~~~~~~
1656
1657Set bit field of a port register::
1658
1659   testpmd> write regfield (port_id) (address) (bit_x) (bit_y) (value)
1660
1661For example, writing to the register cleared in the example above::
1662
1663   testpmd> write regfield 0 0xEE00 0 1 2
1664   port 0 PCI register at offset 0xEE00: 0x00000002 (2)
1665
1666write regbit
1667~~~~~~~~~~~~
1668
1669Set single bit value of a port register::
1670
1671   testpmd> write regbit (port_id) (address) (bit_x) (value)
1672
1673For example, to set the high bit in the register from the example above::
1674
1675   testpmd> write regbit 0 0xEE00 31 1
1676   port 0 PCI register at offset 0xEE00: 0x8000000A (2147483658)
1677
1678
1679Filter Functions
1680----------------
1681
1682This section details the available filter functions that are available.
1683
1684Note these functions interface the deprecated legacy filtering framework,
1685superseded by *rte_flow*. See `Flow rules management`_.
1686
1687ethertype_filter
1688~~~~~~~~~~~~~~~~~~~~
1689
1690Add or delete a L2 Ethertype filter, which identify packets by their L2 Ethertype mainly assign them to a receive queue::
1691
1692   ethertype_filter (port_id) (add|del) (mac_addr|mac_ignr) (mac_address) \
1693                    ethertype (ether_type) (drop|fwd) queue (queue_id)
1694
1695The available information parameters are:
1696
1697* ``port_id``: The port which the Ethertype filter assigned on.
1698
1699* ``mac_addr``: Compare destination mac address.
1700
1701* ``mac_ignr``: Ignore destination mac address match.
1702
1703* ``mac_address``: Destination mac address to match.
1704
1705* ``ether_type``: The EtherType value want to match,
1706  for example 0x0806 for ARP packet. 0x0800 (IPv4) and 0x86DD (IPv6) are invalid.
1707
1708* ``queue_id``: The receive queue associated with this EtherType filter.
1709  It is meaningless when deleting or dropping.
1710
1711Example, to add/remove an ethertype filter rule::
1712
1713   testpmd> ethertype_filter 0 add mac_ignr 00:11:22:33:44:55 \
1714                             ethertype 0x0806 fwd queue 3
1715
1716   testpmd> ethertype_filter 0 del mac_ignr 00:11:22:33:44:55 \
1717                             ethertype 0x0806 fwd queue 3
1718
17192tuple_filter
1720~~~~~~~~~~~~~~~~~
1721
1722Add or delete a 2-tuple filter,
1723which identifies packets by specific protocol and destination TCP/UDP port
1724and forwards packets into one of the receive queues::
1725
1726   2tuple_filter (port_id) (add|del) dst_port (dst_port_value) \
1727                 protocol (protocol_value) mask (mask_value) \
1728                 tcp_flags (tcp_flags_value) priority (prio_value) \
1729                 queue (queue_id)
1730
1731The available information parameters are:
1732
1733* ``port_id``: The port which the 2-tuple filter assigned on.
1734
1735* ``dst_port_value``: Destination port in L4.
1736
1737* ``protocol_value``: IP L4 protocol.
1738
1739* ``mask_value``: Participates in the match or not by bit for field above, 1b means participate.
1740
1741* ``tcp_flags_value``: TCP control bits. The non-zero value is invalid, when the pro_value is not set to 0x06 (TCP).
1742
1743* ``prio_value``: Priority of this filter.
1744
1745* ``queue_id``: The receive queue associated with this 2-tuple filter.
1746
1747Example, to add/remove an 2tuple filter rule::
1748
1749   testpmd> 2tuple_filter 0 add dst_port 32 protocol 0x06 mask 0x03 \
1750                          tcp_flags 0x02 priority 3 queue 3
1751
1752   testpmd> 2tuple_filter 0 del dst_port 32 protocol 0x06 mask 0x03 \
1753                          tcp_flags 0x02 priority 3 queue 3
1754
17555tuple_filter
1756~~~~~~~~~~~~~~~~~
1757
1758Add or delete a 5-tuple filter,
1759which consists of a 5-tuple (protocol, source and destination IP addresses, source and destination TCP/UDP/SCTP port)
1760and routes packets into one of the receive queues::
1761
1762   5tuple_filter (port_id) (add|del) dst_ip (dst_address) src_ip \
1763                 (src_address) dst_port (dst_port_value) \
1764                 src_port (src_port_value) protocol (protocol_value) \
1765                 mask (mask_value) tcp_flags (tcp_flags_value) \
1766                 priority (prio_value) queue (queue_id)
1767
1768The available information parameters are:
1769
1770* ``port_id``: The port which the 5-tuple filter assigned on.
1771
1772* ``dst_address``: Destination IP address.
1773
1774* ``src_address``: Source IP address.
1775
1776* ``dst_port_value``: TCP/UDP destination port.
1777
1778* ``src_port_value``: TCP/UDP source port.
1779
1780* ``protocol_value``: L4 protocol.
1781
1782* ``mask_value``: Participates in the match or not by bit for field above, 1b means participate
1783
1784* ``tcp_flags_value``: TCP control bits. The non-zero value is invalid, when the protocol_value is not set to 0x06 (TCP).
1785
1786* ``prio_value``: The priority of this filter.
1787
1788* ``queue_id``: The receive queue associated with this 5-tuple filter.
1789
1790Example, to add/remove an 5tuple filter rule::
1791
1792   testpmd> 5tuple_filter 0 add dst_ip 2.2.2.5 src_ip 2.2.2.4 \
1793            dst_port 64 src_port 32 protocol 0x06 mask 0x1F \
1794            flags 0x0 priority 3 queue 3
1795
1796   testpmd> 5tuple_filter 0 del dst_ip 2.2.2.5 src_ip 2.2.2.4 \
1797            dst_port 64 src_port 32 protocol 0x06 mask 0x1F \
1798            flags 0x0 priority 3 queue 3
1799
1800syn_filter
1801~~~~~~~~~~
1802
1803Using the  SYN filter, TCP packets whose *SYN* flag is set can be forwarded to a separate queue::
1804
1805   syn_filter (port_id) (add|del) priority (high|low) queue (queue_id)
1806
1807The available information parameters are:
1808
1809* ``port_id``: The port which the SYN filter assigned on.
1810
1811* ``high``: This SYN filter has higher priority than other filters.
1812
1813* ``low``: This SYN filter has lower priority than other filters.
1814
1815* ``queue_id``: The receive queue associated with this SYN filter
1816
1817Example::
1818
1819   testpmd> syn_filter 0 add priority high queue 3
1820
1821flex_filter
1822~~~~~~~~~~~
1823
1824With flex filter, packets can be recognized by any arbitrary pattern within the first 128 bytes of the packet
1825and routed into one of the receive queues::
1826
1827   flex_filter (port_id) (add|del) len (len_value) bytes (bytes_value) \
1828               mask (mask_value) priority (prio_value) queue (queue_id)
1829
1830The available information parameters are:
1831
1832* ``port_id``: The port which the Flex filter is assigned on.
1833
1834* ``len_value``: Filter length in bytes, no greater than 128.
1835
1836* ``bytes_value``: A string in hexadecimal, means the value the flex filter needs to match.
1837
1838* ``mask_value``: A string in hexadecimal, bit 1 means corresponding byte participates in the match.
1839
1840* ``prio_value``: The priority of this filter.
1841
1842* ``queue_id``: The receive queue associated with this Flex filter.
1843
1844Example::
1845
1846   testpmd> flex_filter 0 add len 16 bytes 0x00000000000000000000000008060000 \
1847                          mask 000C priority 3 queue 3
1848
1849   testpmd> flex_filter 0 del len 16 bytes 0x00000000000000000000000008060000 \
1850                          mask 000C priority 3 queue 3
1851
1852
1853.. _testpmd_flow_director:
1854
1855flow_director_filter
1856~~~~~~~~~~~~~~~~~~~~
1857
1858The Flow Director works in receive mode to identify specific flows or sets of flows and route them to specific queues.
1859
1860Four types of filtering are supported which are referred to as Perfect Match, Signature, Perfect-mac-vlan and
1861Perfect-tunnel filters, the match mode is set by the ``--pkt-filter-mode`` command-line parameter:
1862
1863* Perfect match filters.
1864  The hardware checks a match between the masked fields of the received packets and the programmed filters.
1865  The masked fields are for IP flow.
1866
1867* Signature filters.
1868  The hardware checks a match between a hash-based signature of the masked fields of the received packet.
1869
1870* Perfect-mac-vlan match filters.
1871  The hardware checks a match between the masked fields of the received packets and the programmed filters.
1872  The masked fields are for MAC VLAN flow.
1873
1874* Perfect-tunnel match filters.
1875  The hardware checks a match between the masked fields of the received packets and the programmed filters.
1876  The masked fields are for tunnel flow.
1877
1878The Flow Director filters can match the different fields for different type of packet: flow type, specific input set
1879per flow type and the flexible payload.
1880
1881The Flow Director can also mask out parts of all of these fields so that filters
1882are only applied to certain fields or parts of the fields.
1883
1884Different NICs may have different capabilities, command show port fdir (port_id) can be used to acquire the information.
1885
1886# Commands to add flow director filters of different flow types::
1887
1888   flow_director_filter (port_id) mode IP (add|del|update) \
1889                        flow (ipv4-other|ipv4-frag|ipv6-other|ipv6-frag) \
1890                        src (src_ip_address) dst (dst_ip_address) \
1891                        tos (tos_value) proto (proto_value) ttl (ttl_value) \
1892                        vlan (vlan_value) flexbytes (flexbytes_value) \
1893                        (drop|fwd) pf|vf(vf_id) queue (queue_id) \
1894                        fd_id (fd_id_value)
1895
1896   flow_director_filter (port_id) mode IP (add|del|update) \
1897                        flow (ipv4-tcp|ipv4-udp|ipv6-tcp|ipv6-udp) \
1898                        src (src_ip_address) (src_port) \
1899                        dst (dst_ip_address) (dst_port) \
1900                        tos (tos_value) ttl (ttl_value) \
1901                        vlan (vlan_value) flexbytes (flexbytes_value) \
1902                        (drop|fwd) queue pf|vf(vf_id) (queue_id) \
1903                        fd_id (fd_id_value)
1904
1905   flow_director_filter (port_id) mode IP (add|del|update) \
1906                        flow (ipv4-sctp|ipv6-sctp) \
1907                        src (src_ip_address) (src_port) \
1908                        dst (dst_ip_address) (dst_port) \
1909                        tos (tos_value) ttl (ttl_value) \
1910                        tag (verification_tag) vlan (vlan_value) \
1911                        flexbytes (flexbytes_value) (drop|fwd) \
1912                        pf|vf(vf_id) queue (queue_id) fd_id (fd_id_value)
1913
1914   flow_director_filter (port_id) mode IP (add|del|update) flow l2_payload \
1915                        ether (ethertype) flexbytes (flexbytes_value) \
1916                        (drop|fwd) pf|vf(vf_id) queue (queue_id)
1917                        fd_id (fd_id_value)
1918
1919   flow_director_filter (port_id) mode MAC-VLAN (add|del|update) \
1920                        mac (mac_address) vlan (vlan_value) \
1921                        flexbytes (flexbytes_value) (drop|fwd) \
1922                        queue (queue_id) fd_id (fd_id_value)
1923
1924   flow_director_filter (port_id) mode Tunnel (add|del|update) \
1925                        mac (mac_address) vlan (vlan_value) \
1926                        tunnel (NVGRE|VxLAN) tunnel-id (tunnel_id_value) \
1927                        flexbytes (flexbytes_value) (drop|fwd) \
1928                        queue (queue_id) fd_id (fd_id_value)
1929
1930For example, to add an ipv4-udp flow type filter::
1931
1932   testpmd> flow_director_filter 0 mode IP add flow ipv4-udp src 2.2.2.3 32 \
1933            dst 2.2.2.5 33 tos 2 ttl 40 vlan 0x1 flexbytes (0x88,0x48) \
1934            fwd pf queue 1 fd_id 1
1935
1936For example, add an ipv4-other flow type filter::
1937
1938   testpmd> flow_director_filter 0 mode IP add flow ipv4-other src 2.2.2.3 \
1939             dst 2.2.2.5 tos 2 proto 20 ttl 40 vlan 0x1 \
1940             flexbytes (0x88,0x48) fwd pf queue 1 fd_id 1
1941
1942flush_flow_director
1943~~~~~~~~~~~~~~~~~~~
1944
1945Flush all flow director filters on a device::
1946
1947   testpmd> flush_flow_director (port_id)
1948
1949Example, to flush all flow director filter on port 0::
1950
1951   testpmd> flush_flow_director 0
1952
1953flow_director_mask
1954~~~~~~~~~~~~~~~~~~
1955
1956Set flow director's input masks::
1957
1958   flow_director_mask (port_id) mode IP vlan (vlan_value) \
1959                      src_mask (ipv4_src) (ipv6_src) (src_port) \
1960                      dst_mask (ipv4_dst) (ipv6_dst) (dst_port)
1961
1962   flow_director_mask (port_id) mode MAC-VLAN vlan (vlan_value)
1963
1964   flow_director_mask (port_id) mode Tunnel vlan (vlan_value) \
1965                      mac (mac_value) tunnel-type (tunnel_type_value) \
1966                      tunnel-id (tunnel_id_value)
1967
1968Example, to set flow director mask on port 0::
1969
1970   testpmd> flow_director_mask 0 mode IP vlan 0xefff \
1971            src_mask 255.255.255.255 \
1972                FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF \
1973            dst_mask 255.255.255.255 \
1974                FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF
1975
1976flow_director_flex_mask
1977~~~~~~~~~~~~~~~~~~~~~~~
1978
1979set masks of flow director's flexible payload based on certain flow type::
1980
1981   testpmd> flow_director_flex_mask (port_id) \
1982            flow (none|ipv4-other|ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
1983                  ipv6-other|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp| \
1984                  l2_payload|all) (mask)
1985
1986Example, to set flow director's flex mask for all flow type on port 0::
1987
1988   testpmd> flow_director_flex_mask 0 flow all \
1989            (0xff,0xff,0,0,0,0,0,0,0,0,0,0,0,0,0,0)
1990
1991
1992flow_director_flex_payload
1993~~~~~~~~~~~~~~~~~~~~~~~~~~
1994
1995Configure flexible payload selection::
1996
1997   flow_director_flex_payload (port_id) (raw|l2|l3|l4) (config)
1998
1999For example, to select the first 16 bytes from the offset 4 (bytes) of packet's payload as flexible payload::
2000
2001   testpmd> flow_director_flex_payload 0 l4 \
2002            (4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19)
2003
2004get_sym_hash_ena_per_port
2005~~~~~~~~~~~~~~~~~~~~~~~~~
2006
2007Get symmetric hash enable configuration per port::
2008
2009   get_sym_hash_ena_per_port (port_id)
2010
2011For example, to get symmetric hash enable configuration of port 1::
2012
2013   testpmd> get_sym_hash_ena_per_port 1
2014
2015set_sym_hash_ena_per_port
2016~~~~~~~~~~~~~~~~~~~~~~~~~
2017
2018Set symmetric hash enable configuration per port to enable or disable::
2019
2020   set_sym_hash_ena_per_port (port_id) (enable|disable)
2021
2022For example, to set symmetric hash enable configuration of port 1 to enable::
2023
2024   testpmd> set_sym_hash_ena_per_port 1 enable
2025
2026get_hash_global_config
2027~~~~~~~~~~~~~~~~~~~~~~
2028
2029Get the global configurations of hash filters::
2030
2031   get_hash_global_config (port_id)
2032
2033For example, to get the global configurations of hash filters of port 1::
2034
2035   testpmd> get_hash_global_config 1
2036
2037set_hash_global_config
2038~~~~~~~~~~~~~~~~~~~~~~
2039
2040Set the global configurations of hash filters::
2041
2042   set_hash_global_config (port_id) (toeplitz|simple_xor|default) \
2043   (ipv4|ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp|ipv4-other|ipv6|ipv6-frag| \
2044   ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other|l2_payload) \
2045   (enable|disable)
2046
2047For example, to enable simple_xor for flow type of ipv6 on port 2::
2048
2049   testpmd> set_hash_global_config 2 simple_xor ipv6 enable
2050
2051set_hash_input_set
2052~~~~~~~~~~~~~~~~~~
2053
2054Set the input set for hash::
2055
2056   set_hash_input_set (port_id) (ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
2057   ipv4-other|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other| \
2058   l2_payload) (ovlan|ivlan|src-ipv4|dst-ipv4|src-ipv6|dst-ipv6|ipv4-tos| \
2059   ipv4-proto|ipv6-tc|ipv6-next-header|udp-src-port|udp-dst-port| \
2060   tcp-src-port|tcp-dst-port|sctp-src-port|sctp-dst-port|sctp-veri-tag| \
2061   udp-key|gre-key|fld-1st|fld-2nd|fld-3rd|fld-4th|fld-5th|fld-6th|fld-7th| \
2062   fld-8th|none) (select|add)
2063
2064For example, to add source IP to hash input set for flow type of ipv4-udp on port 0::
2065
2066   testpmd> set_hash_input_set 0 ipv4-udp src-ipv4 add
2067
2068set_fdir_input_set
2069~~~~~~~~~~~~~~~~~~
2070
2071The Flow Director filters can match the different fields for different type of packet, i.e. specific input set
2072on per flow type and the flexible payload. This command can be used to change input set for each flow type.
2073
2074Set the input set for flow director::
2075
2076   set_fdir_input_set (port_id) (ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
2077   ipv4-other|ipv6|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other| \
2078   l2_payload) (ivlan|ethertype|src-ipv4|dst-ipv4|src-ipv6|dst-ipv6|ipv4-tos| \
2079   ipv4-proto|ipv4-ttl|ipv6-tc|ipv6-next-header|ipv6-hop-limits| \
2080   tudp-src-port|udp-dst-port|cp-src-port|tcp-dst-port|sctp-src-port| \
2081   sctp-dst-port|sctp-veri-tag|none) (select|add)
2082
2083For example to add source IP to FD input set for flow type of ipv4-udp on port 0::
2084
2085   testpmd> set_fdir_input_set 0 ipv4-udp src-ipv4 add
2086
2087global_config
2088~~~~~~~~~~~~~
2089
2090Set different GRE key length for input set::
2091
2092   global_config (port_id) gre-key-len (number in bytes)
2093
2094For example to set GRE key length for input set to 4 bytes on port 0::
2095
2096   testpmd> global_config 0 gre-key-len 4
2097
2098
2099.. _testpmd_rte_flow:
2100
2101Flow rules management
2102---------------------
2103
2104Control of the generic flow API (*rte_flow*) is fully exposed through the
2105``flow`` command (validation, creation, destruction and queries).
2106
2107Considering *rte_flow* overlaps with all `Filter Functions`_, using both
2108features simultaneously may cause undefined side-effects and is therefore
2109not recommended.
2110
2111``flow`` syntax
2112~~~~~~~~~~~~~~~
2113
2114Because the ``flow`` command uses dynamic tokens to handle the large number
2115of possible flow rules combinations, its behavior differs slightly from
2116other commands, in particular:
2117
2118- Pressing *?* or the *<tab>* key displays contextual help for the current
2119  token, not that of the entire command.
2120
2121- Optional and repeated parameters are supported (provided they are listed
2122  in the contextual help).
2123
2124The first parameter stands for the operation mode. Possible operations and
2125their general syntax are described below. They are covered in detail in the
2126following sections.
2127
2128- Check whether a flow rule can be created::
2129
2130   flow validate {port_id}
2131       [group {group_id}] [priority {level}] [ingress] [egress]
2132       pattern {item} [/ {item} [...]] / end
2133       actions {action} [/ {action} [...]] / end
2134
2135- Create a flow rule::
2136
2137   flow create {port_id}
2138       [group {group_id}] [priority {level}] [ingress] [egress]
2139       pattern {item} [/ {item} [...]] / end
2140       actions {action} [/ {action} [...]] / end
2141
2142- Destroy specific flow rules::
2143
2144   flow destroy {port_id} rule {rule_id} [...]
2145
2146- Destroy all flow rules::
2147
2148   flow flush {port_id}
2149
2150- Query an existing flow rule::
2151
2152   flow query {port_id} {rule_id} {action}
2153
2154- List existing flow rules sorted by priority, filtered by group
2155  identifiers::
2156
2157   flow list {port_id} [group {group_id}] [...]
2158
2159Validating flow rules
2160~~~~~~~~~~~~~~~~~~~~~
2161
2162``flow validate`` reports whether a flow rule would be accepted by the
2163underlying device in its current state but stops short of creating it. It is
2164bound to ``rte_flow_validate()``::
2165
2166   flow validate {port_id}
2167      [group {group_id}] [priority {level}] [ingress] [egress]
2168      pattern {item} [/ {item} [...]] / end
2169      actions {action} [/ {action} [...]] / end
2170
2171If successful, it will show::
2172
2173   Flow rule validated
2174
2175Otherwise it will show an error message of the form::
2176
2177   Caught error type [...] ([...]): [...]
2178
2179This command uses the same parameters as ``flow create``, their format is
2180described in `Creating flow rules`_.
2181
2182Check whether redirecting any Ethernet packet received on port 0 to RX queue
2183index 6 is supported::
2184
2185   testpmd> flow validate 0 ingress pattern eth / end
2186      actions queue index 6 / end
2187   Flow rule validated
2188   testpmd>
2189
2190Port 0 does not support TCPv6 rules::
2191
2192   testpmd> flow validate 0 ingress pattern eth / ipv6 / tcp / end
2193      actions drop / end
2194   Caught error type 9 (specific pattern item): Invalid argument
2195   testpmd>
2196
2197Creating flow rules
2198~~~~~~~~~~~~~~~~~~~
2199
2200``flow create`` validates and creates the specified flow rule. It is bound
2201to ``rte_flow_create()``::
2202
2203   flow create {port_id}
2204      [group {group_id}] [priority {level}] [ingress] [egress]
2205      pattern {item} [/ {item} [...]] / end
2206      actions {action} [/ {action} [...]] / end
2207
2208If successful, it will return a flow rule ID usable with other commands::
2209
2210   Flow rule #[...] created
2211
2212Otherwise it will show an error message of the form::
2213
2214   Caught error type [...] ([...]): [...]
2215
2216Parameters describe in the following order:
2217
2218- Attributes (*group*, *priority*, *ingress*, *egress* tokens).
2219- A matching pattern, starting with the *pattern* token and terminated by an
2220  *end* pattern item.
2221- Actions, starting with the *actions* token and terminated by an *end*
2222  action.
2223
2224These translate directly to *rte_flow* objects provided as-is to the
2225underlying functions.
2226
2227The shortest valid definition only comprises mandatory tokens::
2228
2229   testpmd> flow create 0 pattern end actions end
2230
2231Note that PMDs may refuse rules that essentially do nothing such as this
2232one.
2233
2234**All unspecified object values are automatically initialized to 0.**
2235
2236Attributes
2237^^^^^^^^^^
2238
2239These tokens affect flow rule attributes (``struct rte_flow_attr``) and are
2240specified before the ``pattern`` token.
2241
2242- ``group {group id}``: priority group.
2243- ``priority {level}``: priority level within group.
2244- ``ingress``: rule applies to ingress traffic.
2245- ``egress``: rule applies to egress traffic.
2246
2247Each instance of an attribute specified several times overrides the previous
2248value as shown below (group 4 is used)::
2249
2250   testpmd> flow create 0 group 42 group 24 group 4 [...]
2251
2252Note that once enabled, ``ingress`` and ``egress`` cannot be disabled.
2253
2254While not specifying a direction is an error, some rules may allow both
2255simultaneously.
2256
2257Most rules affect RX therefore contain the ``ingress`` token::
2258
2259   testpmd> flow create 0 ingress pattern [...]
2260
2261Matching pattern
2262^^^^^^^^^^^^^^^^
2263
2264A matching pattern starts after the ``pattern`` token. It is made of pattern
2265items and is terminated by a mandatory ``end`` item.
2266
2267Items are named after their type (*RTE_FLOW_ITEM_TYPE_* from ``enum
2268rte_flow_item_type``).
2269
2270The ``/`` token is used as a separator between pattern items as shown
2271below::
2272
2273   testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end [...]
2274
2275Note that protocol items like these must be stacked from lowest to highest
2276layer to make sense. For instance, the following rule is either invalid or
2277unlikely to match any packet::
2278
2279   testpmd> flow create 0 ingress pattern eth / udp / ipv4 / end [...]
2280
2281More information on these restrictions can be found in the *rte_flow*
2282documentation.
2283
2284Several items support additional specification structures, for example
2285``ipv4`` allows specifying source and destination addresses as follows::
2286
2287   testpmd> flow create 0 ingress pattern eth / ipv4 src is 10.1.1.1
2288      dst is 10.2.0.0 / end [...]
2289
2290This rule matches all IPv4 traffic with the specified properties.
2291
2292In this example, ``src`` and ``dst`` are field names of the underlying
2293``struct rte_flow_item_ipv4`` object. All item properties can be specified
2294in a similar fashion.
2295
2296The ``is`` token means that the subsequent value must be matched exactly,
2297and assigns ``spec`` and ``mask`` fields in ``struct rte_flow_item``
2298accordingly. Possible assignment tokens are:
2299
2300- ``is``: match value perfectly (with full bit-mask).
2301- ``spec``: match value according to configured bit-mask.
2302- ``last``: specify upper bound to establish a range.
2303- ``mask``: specify bit-mask with relevant bits set to one.
2304- ``prefix``: generate bit-mask from a prefix length.
2305
2306These yield identical results::
2307
2308   ipv4 src is 10.1.1.1
2309
2310::
2311
2312   ipv4 src spec 10.1.1.1 src mask 255.255.255.255
2313
2314::
2315
2316   ipv4 src spec 10.1.1.1 src prefix 32
2317
2318::
2319
2320   ipv4 src is 10.1.1.1 src last 10.1.1.1 # range with a single value
2321
2322::
2323
2324   ipv4 src is 10.1.1.1 src last 0 # 0 disables range
2325
2326Inclusive ranges can be defined with ``last``::
2327
2328   ipv4 src is 10.1.1.1 src last 10.2.3.4 # 10.1.1.1 to 10.2.3.4
2329
2330Note that ``mask`` affects both ``spec`` and ``last``::
2331
2332   ipv4 src is 10.1.1.1 src last 10.2.3.4 src mask 255.255.0.0
2333      # matches 10.1.0.0 to 10.2.255.255
2334
2335Properties can be modified multiple times::
2336
2337   ipv4 src is 10.1.1.1 src is 10.1.2.3 src is 10.2.3.4 # matches 10.2.3.4
2338
2339::
2340
2341   ipv4 src is 10.1.1.1 src prefix 24 src prefix 16 # matches 10.1.0.0/16
2342
2343Pattern items
2344^^^^^^^^^^^^^
2345
2346This section lists supported pattern items and their attributes, if any.
2347
2348- ``end``: end list of pattern items.
2349
2350- ``void``: no-op pattern item.
2351
2352- ``invert``: perform actions when pattern does not match.
2353
2354- ``any``: match any protocol for the current layer.
2355
2356  - ``num {unsigned}``: number of layers covered.
2357
2358- ``pf``: match packets addressed to the physical function.
2359
2360- ``vf``: match packets addressed to a virtual function ID.
2361
2362  - ``id {unsigned}``: destination VF ID.
2363
2364- ``port``: device-specific physical port index to use.
2365
2366  - ``index {unsigned}``: physical port index.
2367
2368- ``raw``: match an arbitrary byte string.
2369
2370  - ``relative {boolean}``: look for pattern after the previous item.
2371  - ``search {boolean}``: search pattern from offset (see also limit).
2372  - ``offset {integer}``: absolute or relative offset for pattern.
2373  - ``limit {unsigned}``: search area limit for start of pattern.
2374  - ``pattern {string}``: byte string to look for.
2375
2376- ``eth``: match Ethernet header.
2377
2378  - ``dst {MAC-48}``: destination MAC.
2379  - ``src {MAC-48}``: source MAC.
2380  - ``type {unsigned}``: EtherType.
2381
2382- ``vlan``: match 802.1Q/ad VLAN tag.
2383
2384  - ``tpid {unsigned}``: tag protocol identifier.
2385  - ``tci {unsigned}``: tag control information.
2386  - ``pcp {unsigned}``: priority code point.
2387  - ``dei {unsigned}``: drop eligible indicator.
2388  - ``vid {unsigned}``: VLAN identifier.
2389
2390- ``ipv4``: match IPv4 header.
2391
2392  - ``tos {unsigned}``: type of service.
2393  - ``ttl {unsigned}``: time to live.
2394  - ``proto {unsigned}``: next protocol ID.
2395  - ``src {ipv4 address}``: source address.
2396  - ``dst {ipv4 address}``: destination address.
2397
2398- ``ipv6``: match IPv6 header.
2399
2400  - ``tc {unsigned}``: traffic class.
2401  - ``flow {unsigned}``: flow label.
2402  - ``proto {unsigned}``: protocol (next header).
2403  - ``hop {unsigned}``: hop limit.
2404  - ``src {ipv6 address}``: source address.
2405  - ``dst {ipv6 address}``: destination address.
2406
2407- ``icmp``: match ICMP header.
2408
2409  - ``type {unsigned}``: ICMP packet type.
2410  - ``code {unsigned}``: ICMP packet code.
2411
2412- ``udp``: match UDP header.
2413
2414  - ``src {unsigned}``: UDP source port.
2415  - ``dst {unsigned}``: UDP destination port.
2416
2417- ``tcp``: match TCP header.
2418
2419  - ``src {unsigned}``: TCP source port.
2420  - ``dst {unsigned}``: TCP destination port.
2421
2422- ``sctp``: match SCTP header.
2423
2424  - ``src {unsigned}``: SCTP source port.
2425  - ``dst {unsigned}``: SCTP destination port.
2426  - ``tag {unsigned}``: validation tag.
2427  - ``cksum {unsigned}``: checksum.
2428
2429- ``vxlan``: match VXLAN header.
2430
2431  - ``vni {unsigned}``: VXLAN identifier.
2432
2433Actions list
2434^^^^^^^^^^^^
2435
2436A list of actions starts after the ``actions`` token in the same fashion as
2437`Matching pattern`_; actions are separated by ``/`` tokens and the list is
2438terminated by a mandatory ``end`` action.
2439
2440Actions are named after their type (*RTE_FLOW_ACTION_TYPE_* from ``enum
2441rte_flow_action_type``).
2442
2443Dropping all incoming UDPv4 packets can be expressed as follows::
2444
2445   testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
2446      actions drop / end
2447
2448Several actions have configurable properties which must be specified when
2449there is no valid default value. For example, ``queue`` requires a target
2450queue index.
2451
2452This rule redirects incoming UDPv4 traffic to queue index 6::
2453
2454   testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
2455      actions queue index 6 / end
2456
2457While this one could be rejected by PMDs (unspecified queue index)::
2458
2459   testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
2460      actions queue / end
2461
2462As defined by *rte_flow*, the list is not ordered, all actions of a given
2463rule are performed simultaneously. These are equivalent::
2464
2465   queue index 6 / void / mark id 42 / end
2466
2467::
2468
2469   void / mark id 42 / queue index 6 / end
2470
2471All actions in a list should have different types, otherwise only the last
2472action of a given type is taken into account::
2473
2474   queue index 4 / queue index 5 / queue index 6 / end # will use queue 6
2475
2476::
2477
2478   drop / drop / drop / end # drop is performed only once
2479
2480::
2481
2482   mark id 42 / queue index 3 / mark id 24 / end # mark will be 24
2483
2484Considering they are performed simultaneously, opposite and overlapping
2485actions can sometimes be combined when the end result is unambiguous::
2486
2487   drop / queue index 6 / end # drop has no effect
2488
2489::
2490
2491   drop / dup index 6 / end # same as above
2492
2493::
2494
2495   queue index 6 / rss queues 6 7 8 / end # queue has no effect
2496
2497::
2498
2499   drop / passthru / end # drop has no effect
2500
2501Note that PMDs may still refuse such combinations.
2502
2503Actions
2504^^^^^^^
2505
2506This section lists supported actions and their attributes, if any.
2507
2508- ``end``: end list of actions.
2509
2510- ``void``: no-op action.
2511
2512- ``passthru``: let subsequent rule process matched packets.
2513
2514- ``mark``: attach 32 bit value to packets.
2515
2516  - ``id {unsigned}``: 32 bit value to return with packets.
2517
2518- ``flag``: flag packets.
2519
2520- ``queue``: assign packets to a given queue index.
2521
2522  - ``index {unsigned}``: queue index to use.
2523
2524- ``drop``: drop packets (note: passthru has priority).
2525
2526- ``count``: enable counters for this rule.
2527
2528- ``dup``: duplicate packets to a given queue index.
2529
2530  - ``index {unsigned}``: queue index to duplicate packets to.
2531
2532- ``rss``: spread packets among several queues.
2533
2534  - ``queues [{unsigned} [...]] end``: queue indices to use.
2535
2536- ``pf``: redirect packets to physical device function.
2537
2538- ``vf``: redirect packets to virtual device function.
2539
2540  - ``original {boolean}``: use original VF ID if possible.
2541  - ``id {unsigned}``: VF ID to redirect packets to.
2542
2543Destroying flow rules
2544~~~~~~~~~~~~~~~~~~~~~
2545
2546``flow destroy`` destroys one or more rules from their rule ID (as returned
2547by ``flow create``), this command calls ``rte_flow_destroy()`` as many
2548times as necessary::
2549
2550   flow destroy {port_id} rule {rule_id} [...]
2551
2552If successful, it will show::
2553
2554   Flow rule #[...] destroyed
2555
2556It does not report anything for rule IDs that do not exist. The usual error
2557message is shown when a rule cannot be destroyed::
2558
2559   Caught error type [...] ([...]): [...]
2560
2561``flow flush`` destroys all rules on a device and does not take extra
2562arguments. It is bound to ``rte_flow_flush()``::
2563
2564   flow flush {port_id}
2565
2566Any errors are reported as above.
2567
2568Creating several rules and destroying them::
2569
2570   testpmd> flow create 0 ingress pattern eth / ipv6 / end
2571      actions queue index 2 / end
2572   Flow rule #0 created
2573   testpmd> flow create 0 ingress pattern eth / ipv4 / end
2574      actions queue index 3 / end
2575   Flow rule #1 created
2576   testpmd> flow destroy 0 rule 0 rule 1
2577   Flow rule #1 destroyed
2578   Flow rule #0 destroyed
2579   testpmd>
2580
2581The same result can be achieved using ``flow flush``::
2582
2583   testpmd> flow create 0 ingress pattern eth / ipv6 / end
2584      actions queue index 2 / end
2585   Flow rule #0 created
2586   testpmd> flow create 0 ingress pattern eth / ipv4 / end
2587      actions queue index 3 / end
2588   Flow rule #1 created
2589   testpmd> flow flush 0
2590   testpmd>
2591
2592Non-existent rule IDs are ignored::
2593
2594   testpmd> flow create 0 ingress pattern eth / ipv6 / end
2595      actions queue index 2 / end
2596   Flow rule #0 created
2597   testpmd> flow create 0 ingress pattern eth / ipv4 / end
2598      actions queue index 3 / end
2599   Flow rule #1 created
2600   testpmd> flow destroy 0 rule 42 rule 10 rule 2
2601   testpmd>
2602   testpmd> flow destroy 0 rule 0
2603   Flow rule #0 destroyed
2604   testpmd>
2605
2606Querying flow rules
2607~~~~~~~~~~~~~~~~~~~
2608
2609``flow query`` queries a specific action of a flow rule having that
2610ability. Such actions collect information that can be reported using this
2611command. It is bound to ``rte_flow_query()``::
2612
2613   flow query {port_id} {rule_id} {action}
2614
2615If successful, it will display either the retrieved data for known actions
2616or the following message::
2617
2618   Cannot display result for action type [...] ([...])
2619
2620Otherwise, it will complain either that the rule does not exist or that some
2621error occurred::
2622
2623   Flow rule #[...] not found
2624
2625::
2626
2627   Caught error type [...] ([...]): [...]
2628
2629Currently only the ``count`` action is supported. This action reports the
2630number of packets that hit the flow rule and the total number of bytes. Its
2631output has the following format::
2632
2633   count:
2634    hits_set: [...] # whether "hits" contains a valid value
2635    bytes_set: [...] # whether "bytes" contains a valid value
2636    hits: [...] # number of packets
2637    bytes: [...] # number of bytes
2638
2639Querying counters for TCPv6 packets redirected to queue 6::
2640
2641   testpmd> flow create 0 ingress pattern eth / ipv6 / tcp / end
2642      actions queue index 6 / count / end
2643   Flow rule #4 created
2644   testpmd> flow query 0 4 count
2645   count:
2646    hits_set: 1
2647    bytes_set: 0
2648    hits: 386446
2649    bytes: 0
2650   testpmd>
2651
2652Listing flow rules
2653~~~~~~~~~~~~~~~~~~
2654
2655``flow list`` lists existing flow rules sorted by priority and optionally
2656filtered by group identifiers::
2657
2658   flow list {port_id} [group {group_id}] [...]
2659
2660This command only fails with the following message if the device does not
2661exist::
2662
2663   Invalid port [...]
2664
2665Output consists of a header line followed by a short description of each
2666flow rule, one per line. There is no output at all when no flow rules are
2667configured on the device::
2668
2669   ID      Group   Prio    Attr    Rule
2670   [...]   [...]   [...]   [...]   [...]
2671
2672``Attr`` column flags:
2673
2674- ``i`` for ``ingress``.
2675- ``e`` for ``egress``.
2676
2677Creating several flow rules and listing them::
2678
2679   testpmd> flow create 0 ingress pattern eth / ipv4 / end
2680      actions queue index 6 / end
2681   Flow rule #0 created
2682   testpmd> flow create 0 ingress pattern eth / ipv6 / end
2683      actions queue index 2 / end
2684   Flow rule #1 created
2685   testpmd> flow create 0 priority 5 ingress pattern eth / ipv4 / udp / end
2686      actions rss queues 6 7 8 end / end
2687   Flow rule #2 created
2688   testpmd> flow list 0
2689   ID      Group   Prio    Attr    Rule
2690   0       0       0       i-      ETH IPV4 => QUEUE
2691   1       0       0       i-      ETH IPV6 => QUEUE
2692   2       0       5       i-      ETH IPV4 UDP => RSS
2693   testpmd>
2694
2695Rules are sorted by priority (i.e. group ID first, then priority level)::
2696
2697   testpmd> flow list 1
2698   ID      Group   Prio    Attr    Rule
2699   0       0       0       i-      ETH => COUNT
2700   6       0       500     i-      ETH IPV6 TCP => DROP COUNT
2701   5       0       1000    i-      ETH IPV6 ICMP => QUEUE
2702   1       24      0       i-      ETH IPV4 UDP => QUEUE
2703   4       24      10      i-      ETH IPV4 TCP => DROP
2704   3       24      20      i-      ETH IPV4 => DROP
2705   2       24      42      i-      ETH IPV4 UDP => QUEUE
2706   7       63      0       i-      ETH IPV6 UDP VXLAN => MARK QUEUE
2707   testpmd>
2708
2709Output can be limited to specific groups::
2710
2711   testpmd> flow list 1 group 0 group 63
2712   ID      Group   Prio    Attr    Rule
2713   0       0       0       i-      ETH => COUNT
2714   6       0       500     i-      ETH IPV6 TCP => DROP COUNT
2715   5       0       1000    i-      ETH IPV6 ICMP => QUEUE
2716   7       63      0       i-      ETH IPV6 UDP VXLAN => MARK QUEUE
2717   testpmd>
2718