xref: /dpdk/doc/guides/testpmd_app_ug/testpmd_funcs.rst (revision cac4821929d4e5132a629e431a5be413989caa13)
1..  BSD LICENSE
2    Copyright(c) 2010-2016 Intel Corporation. All rights reserved.
3    All rights reserved.
4
5    Redistribution and use in source and binary forms, with or without
6    modification, are permitted provided that the following conditions
7    are met:
8
9    * Redistributions of source code must retain the above copyright
10    notice, this list of conditions and the following disclaimer.
11    * Redistributions in binary form must reproduce the above copyright
12    notice, this list of conditions and the following disclaimer in
13    the documentation and/or other materials provided with the
14    distribution.
15    * Neither the name of Intel Corporation nor the names of its
16    contributors may be used to endorse or promote products derived
17    from this software without specific prior written permission.
18
19    THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
20    "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
21    LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
22    A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
23    OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
24    SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
25    LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
26    DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
27    THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
28    (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
29    OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
30
31.. _testpmd_runtime:
32
33Testpmd Runtime Functions
34=========================
35
36Where the testpmd application is started in interactive mode, (``-i|--interactive``),
37it displays a prompt that can be used to start and stop forwarding,
38configure the application, display statistics (including the extended NIC
39statistics aka xstats) , set the Flow Director and other tasks::
40
41   testpmd>
42
43The testpmd prompt has some, limited, readline support.
44Common bash command-line functions such as ``Ctrl+a`` and ``Ctrl+e`` to go to the start and end of the prompt line are supported
45as well as access to the command history via the up-arrow.
46
47There is also support for tab completion.
48If you type a partial command and hit ``<TAB>`` you get a list of the available completions:
49
50.. code-block:: console
51
52   testpmd> show port <TAB>
53
54       info [Mul-choice STRING]: show|clear port info|stats|xstats|fdir|stat_qmap|dcb_tc|cap X
55       info [Mul-choice STRING]: show|clear port info|stats|xstats|fdir|stat_qmap|dcb_tc|cap all
56       stats [Mul-choice STRING]: show|clear port info|stats|xstats|fdir|stat_qmap|dcb_tc|cap X
57       stats [Mul-choice STRING]: show|clear port info|stats|xstats|fdir|stat_qmap|dcb_tc|cap all
58       ...
59
60
61.. note::
62
63   Some examples in this document are too long to fit on one line are are shown wrapped at `"\\"` for display purposes::
64
65      testpmd> set flow_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \
66               (pause_time) (send_xon) (port_id)
67
68In the real ``testpmd>`` prompt these commands should be on a single line.
69
70Help Functions
71--------------
72
73The testpmd has on-line help for the functions that are available at runtime.
74These are divided into sections and can be accessed using help, help section or help all:
75
76.. code-block:: console
77
78   testpmd> help
79
80       help control    : Start and stop forwarding.
81       help display    : Displaying port, stats and config information.
82       help config     : Configuration information.
83       help ports      : Configuring ports.
84       help registers  : Reading and setting port registers.
85       help filters    : Filters configuration help.
86       help all        : All of the above sections.
87
88
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|cap) (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
150* ``cap``: Supported offload capabilities.
151
152For example:
153
154.. code-block:: console
155
156   testpmd> show port info 0
157
158   ********************* Infos for port 0 *********************
159
160   MAC address: XX:XX:XX:XX:XX:XX
161   Connect to socket: 0
162   memory allocation on the socket: 0
163   Link status: up
164   Link speed: 40000 Mbps
165   Link duplex: full-duplex
166   Promiscuous mode: enabled
167   Allmulticast mode: disabled
168   Maximum number of MAC addresses: 64
169   Maximum number of MAC addresses of hash filtering: 0
170   VLAN offload:
171       strip on
172       filter on
173       qinq(extend) off
174   Redirection table size: 512
175   Supported flow types:
176     ipv4-frag
177     ipv4-tcp
178     ipv4-udp
179     ipv4-sctp
180     ipv4-other
181     ipv6-frag
182     ipv6-tcp
183     ipv6-udp
184     ipv6-sctp
185     ipv6-other
186     l2_payload
187     port
188     vxlan
189     geneve
190     nvgre
191
192show port rss reta
193~~~~~~~~~~~~~~~~~~
194
195Display the rss redirection table entry indicated by masks on port X::
196
197   testpmd> show port (port_id) rss reta (size) (mask0, mask1...)
198
199size is used to indicate the hardware supported reta size
200
201show port rss-hash
202~~~~~~~~~~~~~~~~~~
203
204Display the RSS hash functions and RSS hash key of a port::
205
206   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]
207
208clear port
209~~~~~~~~~~
210
211Clear the port statistics for a given port or for all ports::
212
213   testpmd> clear port (info|stats|xstats|fdir|stat_qmap) (port_id|all)
214
215For example::
216
217   testpmd> clear port stats all
218
219show (rxq|txq)
220~~~~~~~~~~~~~~
221
222Display information for a given port's RX/TX queue::
223
224   testpmd> show (rxq|txq) info (port_id) (queue_id)
225
226show config
227~~~~~~~~~~~
228
229Displays the configuration of the application.
230The configuration comes from the command-line, the runtime or the application defaults::
231
232   testpmd> show config (rxtx|cores|fwd|txpkts)
233
234The available information categories are:
235
236* ``rxtx``: RX/TX configuration items.
237
238* ``cores``: List of forwarding cores.
239
240* ``fwd``: Packet forwarding configuration.
241
242* ``txpkts``: Packets to TX configuration.
243
244For example:
245
246.. code-block:: console
247
248   testpmd> show config rxtx
249
250   io packet forwarding - CRC stripping disabled - packets/burst=16
251   nb forwarding cores=2 - nb forwarding ports=1
252   RX queues=1 - RX desc=128 - RX free threshold=0
253   RX threshold registers: pthresh=8 hthresh=8 wthresh=4
254   TX queues=1 - TX desc=512 - TX free threshold=0
255   TX threshold registers: pthresh=36 hthresh=0 wthresh=0
256   TX RS bit threshold=0 - TXQ flags=0x0
257
258set fwd
259~~~~~~~
260
261Set the packet forwarding mode::
262
263   testpmd> set fwd (io|mac|macswap|flowgen| \
264                     rxonly|txonly|csum|icmpecho) (""|retry)
265
266``retry`` can be specified for forwarding engines except ``rx_only``.
267
268The available information categories are:
269
270* ``io``: Forwards packets "as-is" in I/O mode.
271  This is the fastest possible forwarding operation as it does not access packets data.
272  This is the default mode.
273
274* ``mac``: Changes the source and the destination Ethernet addresses of packets before forwarding them.
275  Default application behaviour is to set source Ethernet address to that of the transmitting interface, and destination
276  address to a dummy value (set during init). The user may specify a target destination Ethernet address via the 'eth-peer' or
277  'eth-peer-configfile' command-line options. It is not currently possible to specify a specific source Ethernet address.
278
279* ``macswap``: MAC swap forwarding mode.
280  Swaps the source and the destination Ethernet addresses of packets before forwarding them.
281
282* ``flowgen``: Multi-flow generation mode.
283  Originates a number of flows (with varying destination IP addresses), and terminate receive traffic.
284
285* ``rxonly``: Receives packets but doesn't transmit them.
286
287* ``txonly``: Generates and transmits packets without receiving any.
288
289* ``csum``: Changes the checksum field with hardware or software methods depending on the offload flags on the packet.
290
291* ``icmpecho``: Receives a burst of packets, lookup for IMCP echo requests and, if any, send back ICMP echo replies.
292
293* ``ieee1588``: Demonstrate L2 IEEE1588 V2 PTP timestamping for RX and TX. Requires ``CONFIG_RTE_LIBRTE_IEEE1588=y``.
294
295Note: TX timestamping is only available in the "Full Featured" TX path. To force ``testpmd`` into this mode set ``--txqflags=0``.
296
297Example::
298
299   testpmd> set fwd rxonly
300
301   Set rxonly packet forwarding mode
302
303
304read rxd
305~~~~~~~~
306
307Display an RX descriptor for a port RX queue::
308
309   testpmd> read rxd (port_id) (queue_id) (rxd_id)
310
311For example::
312
313   testpmd> read rxd 0 0 4
314        0x0000000B - 0x001D0180 / 0x0000000B - 0x001D0180
315
316read txd
317~~~~~~~~
318
319Display a TX descriptor for a port TX queue::
320
321   testpmd> read txd (port_id) (queue_id) (txd_id)
322
323For example::
324
325   testpmd> read txd 0 0 4
326        0x00000001 - 0x24C3C440 / 0x000F0000 - 0x2330003C
327
328
329Configuration Functions
330-----------------------
331
332The testpmd application can be configured from the runtime as well as from the command-line.
333
334This section details the available configuration functions that are available.
335
336.. note::
337
338   Configuration changes only become active when forwarding is started/restarted.
339
340set default
341~~~~~~~~~~~
342
343Reset forwarding to the default configuration::
344
345   testpmd> set default
346
347set verbose
348~~~~~~~~~~~
349
350Set the debug verbosity level::
351
352   testpmd> set verbose (level)
353
354Currently the only available levels are 0 (silent except for error) and 1 (fully verbose).
355
356set nbport
357~~~~~~~~~~
358
359Set the number of ports used by the application:
360
361set nbport (num)
362
363This is equivalent to the ``--nb-ports`` command-line option.
364
365set nbcore
366~~~~~~~~~~
367
368Set the number of cores used by the application::
369
370   testpmd> set nbcore (num)
371
372This is equivalent to the ``--nb-cores`` command-line option.
373
374.. note::
375
376   The number of cores used must not be greater than number of ports used multiplied by the number of queues per port.
377
378set coremask
379~~~~~~~~~~~~
380
381Set the forwarding cores hexadecimal mask::
382
383   testpmd> set coremask (mask)
384
385This is equivalent to the ``--coremask`` command-line option.
386
387.. note::
388
389   The master lcore is reserved for command line parsing only and cannot be masked on for packet forwarding.
390
391set portmask
392~~~~~~~~~~~~
393
394Set the forwarding ports hexadecimal mask::
395
396   testpmd> set portmask (mask)
397
398This is equivalent to the ``--portmask`` command-line option.
399
400set burst
401~~~~~~~~~
402
403Set number of packets per burst::
404
405   testpmd> set burst (num)
406
407This is equivalent to the ``--burst command-line`` option.
408
409When retry is enabled, the transmit delay time and number of retries can also be set::
410
411   testpmd> set burst tx delay (microseconds) retry (num)
412
413set txpkts
414~~~~~~~~~~
415
416Set the length of each segment of the TX-ONLY packets or length of packet for FLOWGEN mode::
417
418   testpmd> set txpkts (x[,y]*)
419
420Where x[,y]* represents a CSV list of values, without white space.
421
422set txsplit
423~~~~~~~~~~~
424
425Set the split policy for the TX packets, applicable for TX-ONLY and CSUM forwarding modes::
426
427   testpmd> set txsplit (off|on|rand)
428
429Where:
430
431* ``off`` disable packet copy & split for CSUM mode.
432
433* ``on`` split outgoing packet into multiple segments. Size of each segment
434  and number of segments per packet is determined by ``set txpkts`` command
435  (see above).
436
437* ``rand`` same as 'on', but number of segments per each packet is a random value between 1 and total number of segments.
438
439set corelist
440~~~~~~~~~~~~
441
442Set the list of forwarding cores::
443
444   testpmd> set corelist (x[,y]*)
445
446For example, to change the forwarding cores:
447
448.. code-block:: console
449
450   testpmd> set corelist 3,1
451   testpmd> show config fwd
452
453   io packet forwarding - ports=2 - cores=2 - streams=2 - NUMA support disabled
454   Logical Core 3 (socket 0) forwards packets on 1 streams:
455   RX P=0/Q=0 (socket 0) -> TX P=1/Q=0 (socket 0) peer=02:00:00:00:00:01
456   Logical Core 1 (socket 0) forwards packets on 1 streams:
457   RX P=1/Q=0 (socket 0) -> TX P=0/Q=0 (socket 0) peer=02:00:00:00:00:00
458
459.. note::
460
461   The cores are used in the same order as specified on the command line.
462
463set portlist
464~~~~~~~~~~~~
465
466Set the list of forwarding ports::
467
468   testpmd> set portlist (x[,y]*)
469
470For example, to change the port forwarding:
471
472.. code-block:: console
473
474   testpmd> set portlist 0,2,1,3
475   testpmd> show config fwd
476
477   io packet forwarding - ports=4 - cores=1 - streams=4
478   Logical Core 3 (socket 0) forwards packets on 4 streams:
479   RX P=0/Q=0 (socket 0) -> TX P=2/Q=0 (socket 0) peer=02:00:00:00:00:01
480   RX P=2/Q=0 (socket 0) -> TX P=0/Q=0 (socket 0) peer=02:00:00:00:00:00
481   RX P=1/Q=0 (socket 0) -> TX P=3/Q=0 (socket 0) peer=02:00:00:00:00:03
482   RX P=3/Q=0 (socket 0) -> TX P=1/Q=0 (socket 0) peer=02:00:00:00:00:02
483
484set tx loopback
485~~~~~~~~~~~~~~~
486
487Enable/disable tx loopback::
488
489   testpmd> set tx loopback (port_id) (on|off)
490
491set drop enable
492~~~~~~~~~~~~~~~
493
494set drop enable bit for all queues::
495
496   testpmd> set all queues drop (port_id) (on|off)
497
498set split drop enable (for VF)
499~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
500
501set split drop enable bit for VF from PF::
502
503   testpmd> set vf split drop (port_id) (vf_id) (on|off)
504
505set mac antispoof (for VF)
506~~~~~~~~~~~~~~~~~~~~~~~~~~
507
508Set mac antispoof for a VF from the PF::
509
510   testpmd> set vf mac antispoof  (port_id) (vf_id) (on|off)
511
512set macsec offload
513~~~~~~~~~~~~~~~~~~
514
515Enable/disable MACsec offload::
516
517   testpmd> set macsec offload (port_id) on encrypt (on|off) replay-protect (on|off)
518   testpmd> set macsec offload (port_id) off
519
520set macsec sc
521~~~~~~~~~~~~~
522
523Configure MACsec secure connection (SC)::
524
525   testpmd> set macsec sc (tx|rx) (port_id) (mac) (pi)
526
527.. note::
528
529   The pi argument is ignored for tx.
530   Check the NIC Datasheet for hardware limits.
531
532set macsec sa
533~~~~~~~~~~~~~
534
535Configure MACsec secure association (SA)::
536
537   testpmd> set macsec sa (tx|rx) (port_id) (idx) (an) (pn) (key)
538
539.. note::
540
541   The IDX value must be 0 or 1.
542   Check the NIC Datasheet for hardware limits.
543
544set broadcast mode (for VF)
545~~~~~~~~~~~~~~~~~~~~~~~~~~~
546
547Set broadcast mode for a VF from the PF::
548
549   testpmd> set vf broadcast (port_id) (vf_id) (on|off)
550
551vlan set strip
552~~~~~~~~~~~~~~
553
554Set the VLAN strip on a port::
555
556   testpmd> vlan set strip (on|off) (port_id)
557
558vlan set stripq
559~~~~~~~~~~~~~~~
560
561Set the VLAN strip for a queue on a port::
562
563   testpmd> vlan set stripq (on|off) (port_id,queue_id)
564
565vlan set stripq (for VF)
566~~~~~~~~~~~~~~~~~~~~~~~~
567
568Set VLAN strip for all queues in a pool for a VF from the PF::
569
570   testpmd> set vf vlan stripq (port_id) (vf_id) (on|off)
571
572vlan set insert (for VF)
573~~~~~~~~~~~~~~~~~~~~~~~~
574
575Set VLAN insert for a VF from the PF::
576
577   testpmd> set vf vlan insert (port_id) (vf_id) (vlan_id)
578
579vlan set tag (for VF)
580~~~~~~~~~~~~~~~~~~~~~
581
582Set VLAN tag for a VF from the PF::
583
584   testpmd> set vf vlan tag (port_id) (vf_id) (on|off)
585
586vlan set antispoof (for VF)
587~~~~~~~~~~~~~~~~~~~~~~~~~~~
588
589Set VLAN antispoof for a VF from the PF::
590
591   testpmd> set vf vlan antispoof (port_id) (vf_id) (on|off)
592
593vlan set filter
594~~~~~~~~~~~~~~~
595
596Set the VLAN filter on a port::
597
598   testpmd> vlan set filter (on|off) (port_id)
599
600vlan set qinq
601~~~~~~~~~~~~~
602
603Set the VLAN QinQ (extended queue in queue) on for a port::
604
605   testpmd> vlan set qinq (on|off) (port_id)
606
607vlan set tpid
608~~~~~~~~~~~~~
609
610Set the inner or outer VLAN TPID for packet filtering on a port::
611
612   testpmd> vlan set (inner|outer) tpid (value) (port_id)
613
614.. note::
615
616   TPID value must be a 16-bit number (value <= 65536).
617
618rx_vlan add
619~~~~~~~~~~~
620
621Add a VLAN ID, or all identifiers, to the set of VLAN identifiers filtered by port ID::
622
623   testpmd> rx_vlan add (vlan_id|all) (port_id)
624
625.. note::
626
627   VLAN filter must be set on that port. VLAN ID < 4096.
628   Depending on the NIC used, number of vlan_ids may be limited to the maximum entries
629   in VFTA table. This is important if enabling all vlan_ids.
630
631rx_vlan rm
632~~~~~~~~~~
633
634Remove a VLAN ID, or all identifiers, from the set of VLAN identifiers filtered by port ID::
635
636   testpmd> rx_vlan rm (vlan_id|all) (port_id)
637
638rx_vlan add (for VF)
639~~~~~~~~~~~~~~~~~~~~
640
641Add a VLAN ID, to the set of VLAN identifiers filtered for VF(s) for port ID::
642
643   testpmd> rx_vlan add (vlan_id) port (port_id) vf (vf_mask)
644
645rx_vlan rm (for VF)
646~~~~~~~~~~~~~~~~~~~
647
648Remove a VLAN ID, from the set of VLAN identifiers filtered for VF(s) for port ID::
649
650   testpmd> rx_vlan rm (vlan_id) port (port_id) vf (vf_mask)
651
652tunnel_filter add
653~~~~~~~~~~~~~~~~~
654
655Add a tunnel filter on a port::
656
657   testpmd> tunnel_filter add (port_id) (outer_mac) (inner_mac) (ip_addr) \
658            (inner_vlan) (vxlan|nvgre|ipingre) (imac-ivlan|imac-ivlan-tenid|\
659            imac-tenid|imac|omac-imac-tenid|oip|iip) (tenant_id) (queue_id)
660
661The available information categories are:
662
663* ``vxlan``: Set tunnel type as VXLAN.
664
665* ``nvgre``: Set tunnel type as NVGRE.
666
667* ``ipingre``: Set tunnel type as IP-in-GRE.
668
669* ``imac-ivlan``: Set filter type as Inner MAC and VLAN.
670
671* ``imac-ivlan-tenid``: Set filter type as Inner MAC, VLAN and tenant ID.
672
673* ``imac-tenid``: Set filter type as Inner MAC and tenant ID.
674
675* ``imac``: Set filter type as Inner MAC.
676
677* ``omac-imac-tenid``: Set filter type as Outer MAC, Inner MAC and tenant ID.
678
679* ``oip``: Set filter type as Outer IP.
680
681* ``iip``: Set filter type as Inner IP.
682
683Example::
684
685   testpmd> tunnel_filter add 0 68:05:CA:28:09:82 00:00:00:00:00:00 \
686            192.168.2.2 0 ipingre oip 1 1
687
688   Set an IP-in-GRE tunnel on port 0, and the filter type is Outer IP.
689
690tunnel_filter remove
691~~~~~~~~~~~~~~~~~~~~
692
693Remove a tunnel filter on a port::
694
695   testpmd> tunnel_filter rm (port_id) (outer_mac) (inner_mac) (ip_addr) \
696            (inner_vlan) (vxlan|nvgre|ipingre) (imac-ivlan|imac-ivlan-tenid|\
697            imac-tenid|imac|omac-imac-tenid|oip|iip) (tenant_id) (queue_id)
698
699rx_vxlan_port add
700~~~~~~~~~~~~~~~~~
701
702Add an UDP port for VXLAN packet filter on a port::
703
704   testpmd> rx_vxlan_port add (udp_port) (port_id)
705
706rx_vxlan_port remove
707~~~~~~~~~~~~~~~~~~~~
708
709Remove an UDP port for VXLAN packet filter on a port::
710
711   testpmd> rx_vxlan_port rm (udp_port) (port_id)
712
713tx_vlan set
714~~~~~~~~~~~
715
716Set hardware insertion of VLAN IDs in packets sent on a port::
717
718   testpmd> tx_vlan set (port_id) vlan_id[, vlan_id_outer]
719
720For example, set a single VLAN ID (5) insertion on port 0::
721
722   tx_vlan set 0 5
723
724Or, set double VLAN ID (inner: 2, outer: 3) insertion on port 1::
725
726   tx_vlan set 1 2 3
727
728
729tx_vlan set pvid
730~~~~~~~~~~~~~~~~
731
732Set port based hardware insertion of VLAN ID in packets sent on a port::
733
734   testpmd> tx_vlan set pvid (port_id) (vlan_id) (on|off)
735
736tx_vlan reset
737~~~~~~~~~~~~~
738
739Disable hardware insertion of a VLAN header in packets sent on a port::
740
741   testpmd> tx_vlan reset (port_id)
742
743csum set
744~~~~~~~~
745
746Select hardware or software calculation of the checksum when
747transmitting a packet using the ``csum`` forwarding engine::
748
749   testpmd> csum set (ip|udp|tcp|sctp|outer-ip) (hw|sw) (port_id)
750
751Where:
752
753* ``ip|udp|tcp|sctp`` always relate to  the inner layer.
754
755* ``outer-ip`` relates to the outer IP layer (only for IPv4) in the case where the packet is recognized
756  as a tunnel packet by the forwarding engine (vxlan, gre and ipip are
757  supported). See also the ``csum parse-tunnel`` command.
758
759.. note::
760
761   Check the NIC Datasheet for hardware limits.
762
763csum parse-tunnel
764~~~~~~~~~~~~~~~~~
765
766Define how tunneled packets should be handled by the csum forward
767engine::
768
769   testpmd> csum parse-tunnel (on|off) (tx_port_id)
770
771If enabled, the csum forward engine will try to recognize supported
772tunnel headers (vxlan, gre, ipip).
773
774If disabled, treat tunnel packets as non-tunneled packets (a inner
775header is handled as a packet payload).
776
777.. note::
778
779   The port argument is the TX port like in the ``csum set`` command.
780
781Example:
782
783Consider a packet in packet like the following::
784
785   eth_out/ipv4_out/udp_out/vxlan/eth_in/ipv4_in/tcp_in
786
787* If parse-tunnel is enabled, the ``ip|udp|tcp|sctp`` parameters of ``csum set``
788  command relate to the inner headers (here ``ipv4_in`` and ``tcp_in``), and the
789  ``outer-ip parameter`` relates to the outer headers (here ``ipv4_out``).
790
791* If parse-tunnel is disabled, the ``ip|udp|tcp|sctp`` parameters of ``csum  set``
792   command relate to the outer headers, here ``ipv4_out`` and ``udp_out``.
793
794csum show
795~~~~~~~~~
796
797Display tx checksum offload configuration::
798
799   testpmd> csum show (port_id)
800
801tso set
802~~~~~~~
803
804Enable TCP Segmentation Offload (TSO) in the ``csum`` forwarding engine::
805
806   testpmd> tso set (segsize) (port_id)
807
808.. note::
809
810   Check the NIC datasheet for hardware limits.
811
812tso show
813~~~~~~~~
814
815Display the status of TCP Segmentation Offload::
816
817   testpmd> tso show (port_id)
818
819mac_addr add
820~~~~~~~~~~~~
821
822Add an alternative MAC address to a port::
823
824   testpmd> mac_addr add (port_id) (XX:XX:XX:XX:XX:XX)
825
826mac_addr remove
827~~~~~~~~~~~~~~~
828
829Remove a MAC address from a port::
830
831   testpmd> mac_addr remove (port_id) (XX:XX:XX:XX:XX:XX)
832
833mac_addr add (for VF)
834~~~~~~~~~~~~~~~~~~~~~
835
836Add an alternative MAC address for a VF to a port::
837
838   testpmd> mac_add add port (port_id) vf (vf_id) (XX:XX:XX:XX:XX:XX)
839
840mac_addr set
841~~~~~~~~~~~~
842
843Set the default MAC address for a port::
844
845   testpmd> mac_addr set (port_id) (XX:XX:XX:XX:XX:XX)
846
847mac_addr set (for VF)
848~~~~~~~~~~~~~~~~~~~~~
849
850Set the MAC address for a VF from the PF::
851
852   testpmd> set vf mac addr (port_id) (vf_id) (XX:XX:XX:XX:XX:XX)
853
854set port-uta
855~~~~~~~~~~~~
856
857Set the unicast hash filter(s) on/off for a port::
858
859   testpmd> set port (port_id) uta (XX:XX:XX:XX:XX:XX|all) (on|off)
860
861set promisc
862~~~~~~~~~~~
863
864Set the promiscuous mode on for a port or for all ports.
865In promiscuous mode packets are not dropped if they aren't for the specified MAC address::
866
867   testpmd> set promisc (port_id|all) (on|off)
868
869set allmulti
870~~~~~~~~~~~~
871
872Set the allmulti mode for a port or for all ports::
873
874   testpmd> set allmulti (port_id|all) (on|off)
875
876Same as the ifconfig (8) option. Controls how multicast packets are handled.
877
878set promisc (for VF)
879~~~~~~~~~~~~~~~~~~~~
880
881Set the unicast promiscuous mode for a VF from PF.
882It's supported by Intel i40e NICs now.
883In promiscuous mode packets are not dropped if they aren't for the specified MAC address::
884
885   testpmd> set vf promisc (port_id) (vf_id) (on|off)
886
887set allmulticast (for VF)
888~~~~~~~~~~~~~~~~~~~~~~~~~
889
890Set the multicast promiscuous mode for a VF from PF.
891It's supported by Intel i40e NICs now.
892In promiscuous mode packets are not dropped if they aren't for the specified MAC address::
893
894   testpmd> set vf allmulti (port_id) (vf_id) (on|off)
895
896set tx max bandwidth (for VF)
897~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
898
899Set TX max absolute bandwidth (Mbps) for a VF from PF::
900
901   testpmd> set vf tx max-bandwidth (port_id) (vf_id) (max_bandwidth)
902
903set flow_ctrl rx
904~~~~~~~~~~~~~~~~
905
906Set the link flow control parameter on a port::
907
908   testpmd> set flow_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \
909            (pause_time) (send_xon) mac_ctrl_frame_fwd (on|off) \
910	    autoneg (on|off) (port_id)
911
912Where:
913
914* ``high_water`` (integer): High threshold value to trigger XOFF.
915
916* ``low_water`` (integer): Low threshold value to trigger XON.
917
918* ``pause_time`` (integer): Pause quota in the Pause frame.
919
920* ``send_xon`` (0/1): Send XON frame.
921
922* ``mac_ctrl_frame_fwd``: Enable receiving MAC control frames.
923
924* ``autoneg``: Change the auto-negotiation parameter.
925
926set pfc_ctrl rx
927~~~~~~~~~~~~~~~
928
929Set the priority flow control parameter on a port::
930
931   testpmd> set pfc_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \
932            (pause_time) (priority) (port_id)
933
934Where:
935
936* ``high_water`` (integer): High threshold value.
937
938* ``low_water`` (integer): Low threshold value.
939
940* ``pause_time`` (integer): Pause quota in the Pause frame.
941
942* ``priority`` (0-7): VLAN User Priority.
943
944set stat_qmap
945~~~~~~~~~~~~~
946
947Set statistics mapping (qmapping 0..15) for RX/TX queue on port::
948
949   testpmd> set stat_qmap (tx|rx) (port_id) (queue_id) (qmapping)
950
951For example, to set rx queue 2 on port 0 to mapping 5::
952
953   testpmd>set stat_qmap rx 0 2 5
954
955set port - rx/tx (for VF)
956~~~~~~~~~~~~~~~~~~~~~~~~~
957
958Set VF receive/transmit from a port::
959
960   testpmd> set port (port_id) vf (vf_id) (rx|tx) (on|off)
961
962set port - mac address filter (for VF)
963~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
964
965Add/Remove unicast or multicast MAC addr filter for a VF::
966
967   testpmd> set port (port_id) vf (vf_id) (mac_addr) \
968            (exact-mac|exact-mac-vlan|hashmac|hashmac-vlan) (on|off)
969
970set port - rx mode(for VF)
971~~~~~~~~~~~~~~~~~~~~~~~~~~
972
973Set the VF receive mode of a port::
974
975   testpmd> set port (port_id) vf (vf_id) \
976            rxmode (AUPE|ROPE|BAM|MPE) (on|off)
977
978The available receive modes are:
979
980* ``AUPE``: Accepts untagged VLAN.
981
982* ``ROPE``: Accepts unicast hash.
983
984* ``BAM``: Accepts broadcast packets.
985
986* ``MPE``: Accepts all multicast packets.
987
988set port - tx_rate (for Queue)
989~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
990
991Set TX rate limitation for a queue on a port::
992
993   testpmd> set port (port_id) queue (queue_id) rate (rate_value)
994
995set port - tx_rate (for VF)
996~~~~~~~~~~~~~~~~~~~~~~~~~~~
997
998Set TX rate limitation for queues in VF on a port::
999
1000   testpmd> set port (port_id) vf (vf_id) rate (rate_value) queue_mask (queue_mask)
1001
1002set port - mirror rule
1003~~~~~~~~~~~~~~~~~~~~~~
1004
1005Set pool or vlan type mirror rule for a port::
1006
1007   testpmd> set port (port_id) mirror-rule (rule_id) \
1008            (pool-mirror-up|pool-mirror-down|vlan-mirror) \
1009            (poolmask|vlanid[,vlanid]*) dst-pool (pool_id) (on|off)
1010
1011Set link mirror rule for a port::
1012
1013   testpmd> set port (port_id) mirror-rule (rule_id) \
1014           (uplink-mirror|downlink-mirror) dst-pool (pool_id) (on|off)
1015
1016For example to enable mirror traffic with vlan 0,1 to pool 0::
1017
1018   set port 0 mirror-rule 0 vlan-mirror 0,1 dst-pool 0 on
1019
1020reset port - mirror rule
1021~~~~~~~~~~~~~~~~~~~~~~~~
1022
1023Reset a mirror rule for a port::
1024
1025   testpmd> reset port (port_id) mirror-rule (rule_id)
1026
1027set flush_rx
1028~~~~~~~~~~~~
1029
1030Set the flush on RX streams before forwarding.
1031The default is flush ``on``.
1032Mainly used with PCAP drivers to turn off the default behavior of flushing the first 512 packets on RX streams::
1033
1034   testpmd> set flush_rx off
1035
1036set bypass mode
1037~~~~~~~~~~~~~~~
1038
1039Set the bypass mode for the lowest port on bypass enabled NIC::
1040
1041   testpmd> set bypass mode (normal|bypass|isolate) (port_id)
1042
1043set bypass event
1044~~~~~~~~~~~~~~~~
1045
1046Set the event required to initiate specified bypass mode for the lowest port on a bypass enabled::
1047
1048   testpmd> set bypass event (timeout|os_on|os_off|power_on|power_off) \
1049            mode (normal|bypass|isolate) (port_id)
1050
1051Where:
1052
1053* ``timeout``: Enable bypass after watchdog timeout.
1054
1055* ``os_on``: Enable bypass when OS/board is powered on.
1056
1057* ``os_off``: Enable bypass when OS/board is powered off.
1058
1059* ``power_on``: Enable bypass when power supply is turned on.
1060
1061* ``power_off``: Enable bypass when power supply is turned off.
1062
1063
1064set bypass timeout
1065~~~~~~~~~~~~~~~~~~
1066
1067Set the bypass watchdog timeout to ``n`` seconds where 0 = instant::
1068
1069   testpmd> set bypass timeout (0|1.5|2|3|4|8|16|32)
1070
1071show bypass config
1072~~~~~~~~~~~~~~~~~~
1073
1074Show the bypass configuration for a bypass enabled NIC using the lowest port on the NIC::
1075
1076   testpmd> show bypass config (port_id)
1077
1078set link up
1079~~~~~~~~~~~
1080
1081Set link up for a port::
1082
1083   testpmd> set link-up port (port id)
1084
1085set link down
1086~~~~~~~~~~~~~
1087
1088Set link down for a port::
1089
1090   testpmd> set link-down port (port id)
1091
1092E-tag set
1093~~~~~~~~~
1094
1095Enable E-tag insertion for a VF on a port::
1096
1097   testpmd> E-tag set insertion on port-tag-id (value) port (port_id) vf (vf_id)
1098
1099Disable E-tag insertion for a VF on a port::
1100
1101   testpmd> E-tag set insertion off port (port_id) vf (vf_id)
1102
1103Enable/disable E-tag stripping on a port::
1104
1105   testpmd> E-tag set stripping (on|off) port (port_id)
1106
1107Enable/disable E-tag based forwarding on a port::
1108
1109   testpmd> E-tag set forwarding (on|off) port (port_id)
1110
1111Add an E-tag forwarding filter on a port::
1112
1113   testpmd> E-tag set filter add e-tag-id (value) dst-pool (pool_id) port (port_id)
1114
1115Delete an E-tag forwarding filter on a port::
1116   testpmd> E-tag set filter del e-tag-id (value) port (port_id)
1117
1118
1119Port Functions
1120--------------
1121
1122The following sections show functions for configuring ports.
1123
1124.. note::
1125
1126   Port configuration changes only become active when forwarding is started/restarted.
1127
1128port attach
1129~~~~~~~~~~~
1130
1131Attach a port specified by pci address or virtual device args::
1132
1133   testpmd> port attach (identifier)
1134
1135To attach a new pci device, the device should be recognized by kernel first.
1136Then it should be moved under DPDK management.
1137Finally the port can be attached to testpmd.
1138
1139For example, to move a pci device using ixgbe under DPDK management:
1140
1141.. code-block:: console
1142
1143   # Check the status of the available devices.
1144   ./usertools/dpdk-devbind.py --status
1145
1146   Network devices using DPDK-compatible driver
1147   ============================================
1148   <none>
1149
1150   Network devices using kernel driver
1151   ===================================
1152   0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused=
1153
1154
1155   # Bind the device to igb_uio.
1156   sudo ./usertools/dpdk-devbind.py -b igb_uio 0000:0a:00.0
1157
1158
1159   # Recheck the status of the devices.
1160   ./usertools/dpdk-devbind.py --status
1161   Network devices using DPDK-compatible driver
1162   ============================================
1163   0000:0a:00.0 '82599ES 10-Gigabit' drv=igb_uio unused=
1164
1165To attach a port created by virtual device, above steps are not needed.
1166
1167For example, to attach a port whose pci address is 0000:0a:00.0.
1168
1169.. code-block:: console
1170
1171   testpmd> port attach 0000:0a:00.0
1172   Attaching a new port...
1173   EAL: PCI device 0000:0a:00.0 on NUMA socket -1
1174   EAL:   probe driver: 8086:10fb rte_ixgbe_pmd
1175   EAL:   PCI memory mapped at 0x7f83bfa00000
1176   EAL:   PCI memory mapped at 0x7f83bfa80000
1177   PMD: eth_ixgbe_dev_init(): MAC: 2, PHY: 18, SFP+: 5
1178   PMD: eth_ixgbe_dev_init(): port 0 vendorID=0x8086 deviceID=0x10fb
1179   Port 0 is attached. Now total ports is 1
1180   Done
1181
1182For example, to attach a port created by pcap PMD.
1183
1184.. code-block:: console
1185
1186   testpmd> port attach net_pcap0
1187   Attaching a new port...
1188   PMD: Initializing pmd_pcap for net_pcap0
1189   PMD: Creating pcap-backed ethdev on numa socket 0
1190   Port 0 is attached. Now total ports is 1
1191   Done
1192
1193In this case, identifier is ``net_pcap0``.
1194This identifier format is the same as ``--vdev`` format of DPDK applications.
1195
1196For example, to re-attach a bonded port which has been previously detached,
1197the mode and slave parameters must be given.
1198
1199.. code-block:: console
1200
1201   testpmd> port attach net_bond_0,mode=0,slave=1
1202   Attaching a new port...
1203   EAL: Initializing pmd_bond for net_bond_0
1204   EAL: Create bonded device net_bond_0 on port 0 in mode 0 on socket 0.
1205   Port 0 is attached. Now total ports is 1
1206   Done
1207
1208
1209port detach
1210~~~~~~~~~~~
1211
1212Detach a specific port::
1213
1214   testpmd> port detach (port_id)
1215
1216Before detaching a port, the port should be stopped and closed.
1217
1218For example, to detach a pci device port 0.
1219
1220.. code-block:: console
1221
1222   testpmd> port stop 0
1223   Stopping ports...
1224   Done
1225   testpmd> port close 0
1226   Closing ports...
1227   Done
1228
1229   testpmd> port detach 0
1230   Detaching a port...
1231   EAL: PCI device 0000:0a:00.0 on NUMA socket -1
1232   EAL:   remove driver: 8086:10fb rte_ixgbe_pmd
1233   EAL:   PCI memory unmapped at 0x7f83bfa00000
1234   EAL:   PCI memory unmapped at 0x7f83bfa80000
1235   Done
1236
1237
1238For example, to detach a virtual device port 0.
1239
1240.. code-block:: console
1241
1242   testpmd> port stop 0
1243   Stopping ports...
1244   Done
1245   testpmd> port close 0
1246   Closing ports...
1247   Done
1248
1249   testpmd> port detach 0
1250   Detaching a port...
1251   PMD: Closing pcap ethdev on numa socket 0
1252   Port 'net_pcap0' is detached. Now total ports is 0
1253   Done
1254
1255To remove a pci device completely from the system, first detach the port from testpmd.
1256Then the device should be moved under kernel management.
1257Finally the device can be removed using kernel pci hotplug functionality.
1258
1259For example, to move a pci device under kernel management:
1260
1261.. code-block:: console
1262
1263   sudo ./usertools/dpdk-devbind.py -b ixgbe 0000:0a:00.0
1264
1265   ./usertools/dpdk-devbind.py --status
1266
1267   Network devices using DPDK-compatible driver
1268   ============================================
1269   <none>
1270
1271   Network devices using kernel driver
1272   ===================================
1273   0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused=igb_uio
1274
1275To remove a port created by a virtual device, above steps are not needed.
1276
1277port start
1278~~~~~~~~~~
1279
1280Start all ports or a specific port::
1281
1282   testpmd> port start (port_id|all)
1283
1284port stop
1285~~~~~~~~~
1286
1287Stop all ports or a specific port::
1288
1289   testpmd> port stop (port_id|all)
1290
1291port close
1292~~~~~~~~~~
1293
1294Close all ports or a specific port::
1295
1296   testpmd> port close (port_id|all)
1297
1298port start/stop queue
1299~~~~~~~~~~~~~~~~~~~~~
1300
1301Start/stop a rx/tx queue on a specific port::
1302
1303   testpmd> port (port_id) (rxq|txq) (queue_id) (start|stop)
1304
1305Only take effect when port is started.
1306
1307port config - speed
1308~~~~~~~~~~~~~~~~~~~
1309
1310Set the speed and duplex mode for all ports or a specific port::
1311
1312   testpmd> port config (port_id|all) speed (10|100|1000|10000|25000|40000|50000|100000|auto) \
1313            duplex (half|full|auto)
1314
1315port config - queues/descriptors
1316~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1317
1318Set number of queues/descriptors for rxq, txq, rxd and txd::
1319
1320   testpmd> port config all (rxq|txq|rxd|txd) (value)
1321
1322This is equivalent to the ``--rxq``, ``--txq``, ``--rxd`` and ``--txd`` command-line options.
1323
1324port config - max-pkt-len
1325~~~~~~~~~~~~~~~~~~~~~~~~~
1326
1327Set the maximum packet length::
1328
1329   testpmd> port config all max-pkt-len (value)
1330
1331This is equivalent to the ``--max-pkt-len`` command-line option.
1332
1333port config - CRC Strip
1334~~~~~~~~~~~~~~~~~~~~~~~
1335
1336Set hardware CRC stripping on or off for all ports::
1337
1338   testpmd> port config all crc-strip (on|off)
1339
1340CRC stripping is off by default.
1341
1342The ``on`` option is equivalent to the ``--crc-strip`` command-line option.
1343
1344port config - scatter
1345~~~~~~~~~~~~~~~~~~~~~~~
1346
1347Set RX scatter mode on or off for all ports::
1348
1349   testpmd> port config all scatter (on|off)
1350
1351RX scatter mode is off by default.
1352
1353The ``on`` option is equivalent to the ``--enable-scatter`` command-line option.
1354
1355port config - TX queue flags
1356~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1357
1358Set a hexadecimal bitmap of TX queue flags for all ports::
1359
1360   testpmd> port config all txqflags value
1361
1362This command is equivalent to the ``--txqflags`` command-line option.
1363
1364port config - RX Checksum
1365~~~~~~~~~~~~~~~~~~~~~~~~~
1366
1367Set hardware RX checksum offload to on or off for all ports::
1368
1369   testpmd> port config all rx-cksum (on|off)
1370
1371Checksum offload is off by default.
1372
1373The ``on`` option is equivalent to the ``--enable-rx-cksum`` command-line option.
1374
1375port config - VLAN
1376~~~~~~~~~~~~~~~~~~
1377
1378Set hardware VLAN on or off for all ports::
1379
1380   testpmd> port config all hw-vlan (on|off)
1381
1382Hardware VLAN is on by default.
1383
1384The ``off`` option is equivalent to the ``--disable-hw-vlan`` command-line option.
1385
1386port config - VLAN filter
1387~~~~~~~~~~~~~~~~~~~~~~~~~
1388
1389Set hardware VLAN filter on or off for all ports::
1390
1391   testpmd> port config all hw-vlan-filter (on|off)
1392
1393Hardware VLAN filter is on by default.
1394
1395The ``off`` option is equivalent to the ``--disable-hw-vlan-filter`` command-line option.
1396
1397port config - VLAN strip
1398~~~~~~~~~~~~~~~~~~~~~~~~
1399
1400Set hardware VLAN strip on or off for all ports::
1401
1402   testpmd> port config all hw-vlan-strip (on|off)
1403
1404Hardware VLAN strip is on by default.
1405
1406The ``off`` option is equivalent to the ``--disable-hw-vlan-strip`` command-line option.
1407
1408port config - VLAN extend
1409~~~~~~~~~~~~~~~~~~~~~~~~~
1410
1411Set hardware VLAN extend on or off for all ports::
1412
1413   testpmd> port config all hw-vlan-extend (on|off)
1414
1415Hardware VLAN extend is off by default.
1416
1417The ``off`` option is equivalent to the ``--disable-hw-vlan-extend`` command-line option.
1418
1419port config - Drop Packets
1420~~~~~~~~~~~~~~~~~~~~~~~~~~
1421
1422Set packet drop for packets with no descriptors on or off for all ports::
1423
1424   testpmd> port config all drop-en (on|off)
1425
1426Packet dropping for packets with no descriptors is off by default.
1427
1428The ``on`` option is equivalent to the ``--enable-drop-en`` command-line option.
1429
1430port config - RSS
1431~~~~~~~~~~~~~~~~~
1432
1433Set the RSS (Receive Side Scaling) mode on or off::
1434
1435   testpmd> port config all rss (all|ip|tcp|udp|sctp|ether|port|vxlan|geneve|nvgre|none)
1436
1437RSS is on by default.
1438
1439The ``none`` option is equivalent to the ``--disable-rss`` command-line option.
1440
1441port config - RSS Reta
1442~~~~~~~~~~~~~~~~~~~~~~
1443
1444Set the RSS (Receive Side Scaling) redirection table::
1445
1446   testpmd> port config all rss reta (hash,queue)[,(hash,queue)]
1447
1448port config - DCB
1449~~~~~~~~~~~~~~~~~
1450
1451Set the DCB mode for an individual port::
1452
1453   testpmd> port config (port_id) dcb vt (on|off) (traffic_class) pfc (on|off)
1454
1455The traffic class should be 4 or 8.
1456
1457port config - Burst
1458~~~~~~~~~~~~~~~~~~~
1459
1460Set the number of packets per burst::
1461
1462   testpmd> port config all burst (value)
1463
1464This is equivalent to the ``--burst`` command-line option.
1465
1466port config - Threshold
1467~~~~~~~~~~~~~~~~~~~~~~~
1468
1469Set thresholds for TX/RX queues::
1470
1471   testpmd> port config all (threshold) (value)
1472
1473Where the threshold type can be:
1474
1475* ``txpt:`` Set the prefetch threshold register of the TX rings, 0 <= value <= 255.
1476
1477* ``txht:`` Set the host threshold register of the TX rings, 0 <= value <= 255.
1478
1479* ``txwt:`` Set the write-back threshold register of the TX rings, 0 <= value <= 255.
1480
1481* ``rxpt:`` Set the prefetch threshold register of the RX rings, 0 <= value <= 255.
1482
1483* ``rxht:`` Set the host threshold register of the RX rings, 0 <= value <= 255.
1484
1485* ``rxwt:`` Set the write-back threshold register of the RX rings, 0 <= value <= 255.
1486
1487* ``txfreet:`` Set the transmit free threshold of the TX rings, 0 <= value <= txd.
1488
1489* ``rxfreet:`` Set the transmit free threshold of the RX rings, 0 <= value <= rxd.
1490
1491* ``txrst:`` Set the transmit RS bit threshold of TX rings, 0 <= value <= txd.
1492
1493These threshold options are also available from the command-line.
1494
1495port config - E-tag
1496~~~~~~~~~~~~~~~~~~~
1497
1498Set the value of ether-type for E-tag::
1499
1500   testpmd> port config (port_id|all) l2-tunnel E-tag ether-type (value)
1501
1502Enable/disable the E-tag support::
1503
1504   testpmd> port config (port_id|all) l2-tunnel E-tag (enable|disable)
1505
1506
1507Link Bonding Functions
1508----------------------
1509
1510The Link Bonding functions make it possible to dynamically create and
1511manage link bonding devices from within testpmd interactive prompt.
1512
1513create bonded device
1514~~~~~~~~~~~~~~~~~~~~
1515
1516Create a new bonding device::
1517
1518   testpmd> create bonded device (mode) (socket)
1519
1520For example, to create a bonded device in mode 1 on socket 0::
1521
1522   testpmd> create bonded 1 0
1523   created new bonded device (port X)
1524
1525add bonding slave
1526~~~~~~~~~~~~~~~~~
1527
1528Adds Ethernet device to a Link Bonding device::
1529
1530   testpmd> add bonding slave (slave id) (port id)
1531
1532For example, to add Ethernet device (port 6) to a Link Bonding device (port 10)::
1533
1534   testpmd> add bonding slave 6 10
1535
1536
1537remove bonding slave
1538~~~~~~~~~~~~~~~~~~~~
1539
1540Removes an Ethernet slave device from a Link Bonding device::
1541
1542   testpmd> remove bonding slave (slave id) (port id)
1543
1544For example, to remove Ethernet slave device (port 6) to a Link Bonding device (port 10)::
1545
1546   testpmd> remove bonding slave 6 10
1547
1548set bonding mode
1549~~~~~~~~~~~~~~~~
1550
1551Set the Link Bonding mode of a Link Bonding device::
1552
1553   testpmd> set bonding mode (value) (port id)
1554
1555For example, to set the bonding mode of a Link Bonding device (port 10) to broadcast (mode 3)::
1556
1557   testpmd> set bonding mode 3 10
1558
1559set bonding primary
1560~~~~~~~~~~~~~~~~~~~
1561
1562Set an Ethernet slave device as the primary device on a Link Bonding device::
1563
1564   testpmd> set bonding primary (slave id) (port id)
1565
1566For example, to set the Ethernet slave device (port 6) as the primary port of a Link Bonding device (port 10)::
1567
1568   testpmd> set bonding primary 6 10
1569
1570set bonding mac
1571~~~~~~~~~~~~~~~
1572
1573Set the MAC address of a Link Bonding device::
1574
1575   testpmd> set bonding mac (port id) (mac)
1576
1577For example, to set the MAC address of a Link Bonding device (port 10) to 00:00:00:00:00:01::
1578
1579   testpmd> set bonding mac 10 00:00:00:00:00:01
1580
1581set bonding xmit_balance_policy
1582~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1583
1584Set the transmission policy for a Link Bonding device when it is in Balance XOR mode::
1585
1586   testpmd> set bonding xmit_balance_policy (port_id) (l2|l23|l34)
1587
1588For example, set a Link Bonding device (port 10) to use a balance policy of layer 3+4 (IP addresses & UDP ports)::
1589
1590   testpmd> set bonding xmit_balance_policy 10 l34
1591
1592
1593set bonding mon_period
1594~~~~~~~~~~~~~~~~~~~~~~
1595
1596Set the link status monitoring polling period in milliseconds for a bonding device.
1597
1598This adds support for PMD slave devices which do not support link status interrupts.
1599When the mon_period is set to a value greater than 0 then all PMD's which do not support
1600link status ISR will be queried every polling interval to check if their link status has changed::
1601
1602   testpmd> set bonding mon_period (port_id) (value)
1603
1604For example, to set the link status monitoring polling period of bonded device (port 5) to 150ms::
1605
1606   testpmd> set bonding mon_period 5 150
1607
1608
1609show bonding config
1610~~~~~~~~~~~~~~~~~~~
1611
1612Show the current configuration of a Link Bonding device::
1613
1614   testpmd> show bonding config (port id)
1615
1616For example,
1617to show the configuration a Link Bonding device (port 9) with 3 slave devices (1, 3, 4)
1618in balance mode with a transmission policy of layer 2+3::
1619
1620   testpmd> show bonding config 9
1621        Bonding mode: 2
1622        Balance Xmit Policy: BALANCE_XMIT_POLICY_LAYER23
1623        Slaves (3): [1 3 4]
1624        Active Slaves (3): [1 3 4]
1625        Primary: [3]
1626
1627
1628Register Functions
1629------------------
1630
1631The Register Functions can be used to read from and write to registers on the network card referenced by a port number.
1632This is mainly useful for debugging purposes.
1633Reference should be made to the appropriate datasheet for the network card for details on the register addresses
1634and fields that can be accessed.
1635
1636read reg
1637~~~~~~~~
1638
1639Display the value of a port register::
1640
1641   testpmd> read reg (port_id) (address)
1642
1643For example, to examine the Flow Director control register (FDIRCTL, 0x0000EE000) on an Intel 82599 10 GbE Controller::
1644
1645   testpmd> read reg 0 0xEE00
1646   port 0 PCI register at offset 0xEE00: 0x4A060029 (1241907241)
1647
1648read regfield
1649~~~~~~~~~~~~~
1650
1651Display a port register bit field::
1652
1653   testpmd> read regfield (port_id) (address) (bit_x) (bit_y)
1654
1655For example, reading the lowest two bits from the register in the example above::
1656
1657   testpmd> read regfield 0 0xEE00 0 1
1658   port 0 PCI register at offset 0xEE00: bits[0, 1]=0x1 (1)
1659
1660read regbit
1661~~~~~~~~~~~
1662
1663Display a single port register bit::
1664
1665   testpmd> read regbit (port_id) (address) (bit_x)
1666
1667For example, reading the lowest bit from the register in the example above::
1668
1669   testpmd> read regbit 0 0xEE00 0
1670   port 0 PCI register at offset 0xEE00: bit 0=1
1671
1672write reg
1673~~~~~~~~~
1674
1675Set the value of a port register::
1676
1677   testpmd> write reg (port_id) (address) (value)
1678
1679For example, to clear a register::
1680
1681   testpmd> write reg 0 0xEE00 0x0
1682   port 0 PCI register at offset 0xEE00: 0x00000000 (0)
1683
1684write regfield
1685~~~~~~~~~~~~~~
1686
1687Set bit field of a port register::
1688
1689   testpmd> write regfield (port_id) (address) (bit_x) (bit_y) (value)
1690
1691For example, writing to the register cleared in the example above::
1692
1693   testpmd> write regfield 0 0xEE00 0 1 2
1694   port 0 PCI register at offset 0xEE00: 0x00000002 (2)
1695
1696write regbit
1697~~~~~~~~~~~~
1698
1699Set single bit value of a port register::
1700
1701   testpmd> write regbit (port_id) (address) (bit_x) (value)
1702
1703For example, to set the high bit in the register from the example above::
1704
1705   testpmd> write regbit 0 0xEE00 31 1
1706   port 0 PCI register at offset 0xEE00: 0x8000000A (2147483658)
1707
1708
1709Filter Functions
1710----------------
1711
1712This section details the available filter functions that are available.
1713
1714Note these functions interface the deprecated legacy filtering framework,
1715superseded by *rte_flow*. See `Flow rules management`_.
1716
1717ethertype_filter
1718~~~~~~~~~~~~~~~~~~~~
1719
1720Add or delete a L2 Ethertype filter, which identify packets by their L2 Ethertype mainly assign them to a receive queue::
1721
1722   ethertype_filter (port_id) (add|del) (mac_addr|mac_ignr) (mac_address) \
1723                    ethertype (ether_type) (drop|fwd) queue (queue_id)
1724
1725The available information parameters are:
1726
1727* ``port_id``: The port which the Ethertype filter assigned on.
1728
1729* ``mac_addr``: Compare destination mac address.
1730
1731* ``mac_ignr``: Ignore destination mac address match.
1732
1733* ``mac_address``: Destination mac address to match.
1734
1735* ``ether_type``: The EtherType value want to match,
1736  for example 0x0806 for ARP packet. 0x0800 (IPv4) and 0x86DD (IPv6) are invalid.
1737
1738* ``queue_id``: The receive queue associated with this EtherType filter.
1739  It is meaningless when deleting or dropping.
1740
1741Example, to add/remove an ethertype filter rule::
1742
1743   testpmd> ethertype_filter 0 add mac_ignr 00:11:22:33:44:55 \
1744                             ethertype 0x0806 fwd queue 3
1745
1746   testpmd> ethertype_filter 0 del mac_ignr 00:11:22:33:44:55 \
1747                             ethertype 0x0806 fwd queue 3
1748
17492tuple_filter
1750~~~~~~~~~~~~~~~~~
1751
1752Add or delete a 2-tuple filter,
1753which identifies packets by specific protocol and destination TCP/UDP port
1754and forwards packets into one of the receive queues::
1755
1756   2tuple_filter (port_id) (add|del) dst_port (dst_port_value) \
1757                 protocol (protocol_value) mask (mask_value) \
1758                 tcp_flags (tcp_flags_value) priority (prio_value) \
1759                 queue (queue_id)
1760
1761The available information parameters are:
1762
1763* ``port_id``: The port which the 2-tuple filter assigned on.
1764
1765* ``dst_port_value``: Destination port in L4.
1766
1767* ``protocol_value``: IP L4 protocol.
1768
1769* ``mask_value``: Participates in the match or not by bit for field above, 1b means participate.
1770
1771* ``tcp_flags_value``: TCP control bits. The non-zero value is invalid, when the pro_value is not set to 0x06 (TCP).
1772
1773* ``prio_value``: Priority of this filter.
1774
1775* ``queue_id``: The receive queue associated with this 2-tuple filter.
1776
1777Example, to add/remove an 2tuple filter rule::
1778
1779   testpmd> 2tuple_filter 0 add dst_port 32 protocol 0x06 mask 0x03 \
1780                          tcp_flags 0x02 priority 3 queue 3
1781
1782   testpmd> 2tuple_filter 0 del dst_port 32 protocol 0x06 mask 0x03 \
1783                          tcp_flags 0x02 priority 3 queue 3
1784
17855tuple_filter
1786~~~~~~~~~~~~~~~~~
1787
1788Add or delete a 5-tuple filter,
1789which consists of a 5-tuple (protocol, source and destination IP addresses, source and destination TCP/UDP/SCTP port)
1790and routes packets into one of the receive queues::
1791
1792   5tuple_filter (port_id) (add|del) dst_ip (dst_address) src_ip \
1793                 (src_address) dst_port (dst_port_value) \
1794                 src_port (src_port_value) protocol (protocol_value) \
1795                 mask (mask_value) tcp_flags (tcp_flags_value) \
1796                 priority (prio_value) queue (queue_id)
1797
1798The available information parameters are:
1799
1800* ``port_id``: The port which the 5-tuple filter assigned on.
1801
1802* ``dst_address``: Destination IP address.
1803
1804* ``src_address``: Source IP address.
1805
1806* ``dst_port_value``: TCP/UDP destination port.
1807
1808* ``src_port_value``: TCP/UDP source port.
1809
1810* ``protocol_value``: L4 protocol.
1811
1812* ``mask_value``: Participates in the match or not by bit for field above, 1b means participate
1813
1814* ``tcp_flags_value``: TCP control bits. The non-zero value is invalid, when the protocol_value is not set to 0x06 (TCP).
1815
1816* ``prio_value``: The priority of this filter.
1817
1818* ``queue_id``: The receive queue associated with this 5-tuple filter.
1819
1820Example, to add/remove an 5tuple filter rule::
1821
1822   testpmd> 5tuple_filter 0 add dst_ip 2.2.2.5 src_ip 2.2.2.4 \
1823            dst_port 64 src_port 32 protocol 0x06 mask 0x1F \
1824            flags 0x0 priority 3 queue 3
1825
1826   testpmd> 5tuple_filter 0 del dst_ip 2.2.2.5 src_ip 2.2.2.4 \
1827            dst_port 64 src_port 32 protocol 0x06 mask 0x1F \
1828            flags 0x0 priority 3 queue 3
1829
1830syn_filter
1831~~~~~~~~~~
1832
1833Using the  SYN filter, TCP packets whose *SYN* flag is set can be forwarded to a separate queue::
1834
1835   syn_filter (port_id) (add|del) priority (high|low) queue (queue_id)
1836
1837The available information parameters are:
1838
1839* ``port_id``: The port which the SYN filter assigned on.
1840
1841* ``high``: This SYN filter has higher priority than other filters.
1842
1843* ``low``: This SYN filter has lower priority than other filters.
1844
1845* ``queue_id``: The receive queue associated with this SYN filter
1846
1847Example::
1848
1849   testpmd> syn_filter 0 add priority high queue 3
1850
1851flex_filter
1852~~~~~~~~~~~
1853
1854With flex filter, packets can be recognized by any arbitrary pattern within the first 128 bytes of the packet
1855and routed into one of the receive queues::
1856
1857   flex_filter (port_id) (add|del) len (len_value) bytes (bytes_value) \
1858               mask (mask_value) priority (prio_value) queue (queue_id)
1859
1860The available information parameters are:
1861
1862* ``port_id``: The port which the Flex filter is assigned on.
1863
1864* ``len_value``: Filter length in bytes, no greater than 128.
1865
1866* ``bytes_value``: A string in hexadecimal, means the value the flex filter needs to match.
1867
1868* ``mask_value``: A string in hexadecimal, bit 1 means corresponding byte participates in the match.
1869
1870* ``prio_value``: The priority of this filter.
1871
1872* ``queue_id``: The receive queue associated with this Flex filter.
1873
1874Example::
1875
1876   testpmd> flex_filter 0 add len 16 bytes 0x00000000000000000000000008060000 \
1877                          mask 000C priority 3 queue 3
1878
1879   testpmd> flex_filter 0 del len 16 bytes 0x00000000000000000000000008060000 \
1880                          mask 000C priority 3 queue 3
1881
1882
1883.. _testpmd_flow_director:
1884
1885flow_director_filter
1886~~~~~~~~~~~~~~~~~~~~
1887
1888The Flow Director works in receive mode to identify specific flows or sets of flows and route them to specific queues.
1889
1890Four types of filtering are supported which are referred to as Perfect Match, Signature, Perfect-mac-vlan and
1891Perfect-tunnel filters, the match mode is set by the ``--pkt-filter-mode`` command-line parameter:
1892
1893* Perfect match filters.
1894  The hardware checks a match between the masked fields of the received packets and the programmed filters.
1895  The masked fields are for IP flow.
1896
1897* Signature filters.
1898  The hardware checks a match between a hash-based signature of the masked fields of the received packet.
1899
1900* Perfect-mac-vlan match filters.
1901  The hardware checks a match between the masked fields of the received packets and the programmed filters.
1902  The masked fields are for MAC VLAN flow.
1903
1904* Perfect-tunnel match filters.
1905  The hardware checks a match between the masked fields of the received packets and the programmed filters.
1906  The masked fields are for tunnel flow.
1907
1908The Flow Director filters can match the different fields for different type of packet: flow type, specific input set
1909per flow type and the flexible payload.
1910
1911The Flow Director can also mask out parts of all of these fields so that filters
1912are only applied to certain fields or parts of the fields.
1913
1914Different NICs may have different capabilities, command show port fdir (port_id) can be used to acquire the information.
1915
1916# Commands to add flow director filters of different flow types::
1917
1918   flow_director_filter (port_id) mode IP (add|del|update) \
1919                        flow (ipv4-other|ipv4-frag|ipv6-other|ipv6-frag) \
1920                        src (src_ip_address) dst (dst_ip_address) \
1921                        tos (tos_value) proto (proto_value) ttl (ttl_value) \
1922                        vlan (vlan_value) flexbytes (flexbytes_value) \
1923                        (drop|fwd) pf|vf(vf_id) queue (queue_id) \
1924                        fd_id (fd_id_value)
1925
1926   flow_director_filter (port_id) mode IP (add|del|update) \
1927                        flow (ipv4-tcp|ipv4-udp|ipv6-tcp|ipv6-udp) \
1928                        src (src_ip_address) (src_port) \
1929                        dst (dst_ip_address) (dst_port) \
1930                        tos (tos_value) ttl (ttl_value) \
1931                        vlan (vlan_value) flexbytes (flexbytes_value) \
1932                        (drop|fwd) queue pf|vf(vf_id) (queue_id) \
1933                        fd_id (fd_id_value)
1934
1935   flow_director_filter (port_id) mode IP (add|del|update) \
1936                        flow (ipv4-sctp|ipv6-sctp) \
1937                        src (src_ip_address) (src_port) \
1938                        dst (dst_ip_address) (dst_port) \
1939                        tos (tos_value) ttl (ttl_value) \
1940                        tag (verification_tag) vlan (vlan_value) \
1941                        flexbytes (flexbytes_value) (drop|fwd) \
1942                        pf|vf(vf_id) queue (queue_id) fd_id (fd_id_value)
1943
1944   flow_director_filter (port_id) mode IP (add|del|update) flow l2_payload \
1945                        ether (ethertype) flexbytes (flexbytes_value) \
1946                        (drop|fwd) pf|vf(vf_id) queue (queue_id)
1947                        fd_id (fd_id_value)
1948
1949   flow_director_filter (port_id) mode MAC-VLAN (add|del|update) \
1950                        mac (mac_address) vlan (vlan_value) \
1951                        flexbytes (flexbytes_value) (drop|fwd) \
1952                        queue (queue_id) fd_id (fd_id_value)
1953
1954   flow_director_filter (port_id) mode Tunnel (add|del|update) \
1955                        mac (mac_address) vlan (vlan_value) \
1956                        tunnel (NVGRE|VxLAN) tunnel-id (tunnel_id_value) \
1957                        flexbytes (flexbytes_value) (drop|fwd) \
1958                        queue (queue_id) fd_id (fd_id_value)
1959
1960For example, to add an ipv4-udp flow type filter::
1961
1962   testpmd> flow_director_filter 0 mode IP add flow ipv4-udp src 2.2.2.3 32 \
1963            dst 2.2.2.5 33 tos 2 ttl 40 vlan 0x1 flexbytes (0x88,0x48) \
1964            fwd pf queue 1 fd_id 1
1965
1966For example, add an ipv4-other flow type filter::
1967
1968   testpmd> flow_director_filter 0 mode IP add flow ipv4-other src 2.2.2.3 \
1969             dst 2.2.2.5 tos 2 proto 20 ttl 40 vlan 0x1 \
1970             flexbytes (0x88,0x48) fwd pf queue 1 fd_id 1
1971
1972flush_flow_director
1973~~~~~~~~~~~~~~~~~~~
1974
1975Flush all flow director filters on a device::
1976
1977   testpmd> flush_flow_director (port_id)
1978
1979Example, to flush all flow director filter on port 0::
1980
1981   testpmd> flush_flow_director 0
1982
1983flow_director_mask
1984~~~~~~~~~~~~~~~~~~
1985
1986Set flow director's input masks::
1987
1988   flow_director_mask (port_id) mode IP vlan (vlan_value) \
1989                      src_mask (ipv4_src) (ipv6_src) (src_port) \
1990                      dst_mask (ipv4_dst) (ipv6_dst) (dst_port)
1991
1992   flow_director_mask (port_id) mode MAC-VLAN vlan (vlan_value)
1993
1994   flow_director_mask (port_id) mode Tunnel vlan (vlan_value) \
1995                      mac (mac_value) tunnel-type (tunnel_type_value) \
1996                      tunnel-id (tunnel_id_value)
1997
1998Example, to set flow director mask on port 0::
1999
2000   testpmd> flow_director_mask 0 mode IP vlan 0xefff \
2001            src_mask 255.255.255.255 \
2002                FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF \
2003            dst_mask 255.255.255.255 \
2004                FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF
2005
2006flow_director_flex_mask
2007~~~~~~~~~~~~~~~~~~~~~~~
2008
2009set masks of flow director's flexible payload based on certain flow type::
2010
2011   testpmd> flow_director_flex_mask (port_id) \
2012            flow (none|ipv4-other|ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
2013                  ipv6-other|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp| \
2014                  l2_payload|all) (mask)
2015
2016Example, to set flow director's flex mask for all flow type on port 0::
2017
2018   testpmd> flow_director_flex_mask 0 flow all \
2019            (0xff,0xff,0,0,0,0,0,0,0,0,0,0,0,0,0,0)
2020
2021
2022flow_director_flex_payload
2023~~~~~~~~~~~~~~~~~~~~~~~~~~
2024
2025Configure flexible payload selection::
2026
2027   flow_director_flex_payload (port_id) (raw|l2|l3|l4) (config)
2028
2029For example, to select the first 16 bytes from the offset 4 (bytes) of packet's payload as flexible payload::
2030
2031   testpmd> flow_director_flex_payload 0 l4 \
2032            (4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19)
2033
2034get_sym_hash_ena_per_port
2035~~~~~~~~~~~~~~~~~~~~~~~~~
2036
2037Get symmetric hash enable configuration per port::
2038
2039   get_sym_hash_ena_per_port (port_id)
2040
2041For example, to get symmetric hash enable configuration of port 1::
2042
2043   testpmd> get_sym_hash_ena_per_port 1
2044
2045set_sym_hash_ena_per_port
2046~~~~~~~~~~~~~~~~~~~~~~~~~
2047
2048Set symmetric hash enable configuration per port to enable or disable::
2049
2050   set_sym_hash_ena_per_port (port_id) (enable|disable)
2051
2052For example, to set symmetric hash enable configuration of port 1 to enable::
2053
2054   testpmd> set_sym_hash_ena_per_port 1 enable
2055
2056get_hash_global_config
2057~~~~~~~~~~~~~~~~~~~~~~
2058
2059Get the global configurations of hash filters::
2060
2061   get_hash_global_config (port_id)
2062
2063For example, to get the global configurations of hash filters of port 1::
2064
2065   testpmd> get_hash_global_config 1
2066
2067set_hash_global_config
2068~~~~~~~~~~~~~~~~~~~~~~
2069
2070Set the global configurations of hash filters::
2071
2072   set_hash_global_config (port_id) (toeplitz|simple_xor|default) \
2073   (ipv4|ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp|ipv4-other|ipv6|ipv6-frag| \
2074   ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other|l2_payload) \
2075   (enable|disable)
2076
2077For example, to enable simple_xor for flow type of ipv6 on port 2::
2078
2079   testpmd> set_hash_global_config 2 simple_xor ipv6 enable
2080
2081set_hash_input_set
2082~~~~~~~~~~~~~~~~~~
2083
2084Set the input set for hash::
2085
2086   set_hash_input_set (port_id) (ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
2087   ipv4-other|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other| \
2088   l2_payload) (ovlan|ivlan|src-ipv4|dst-ipv4|src-ipv6|dst-ipv6|ipv4-tos| \
2089   ipv4-proto|ipv6-tc|ipv6-next-header|udp-src-port|udp-dst-port| \
2090   tcp-src-port|tcp-dst-port|sctp-src-port|sctp-dst-port|sctp-veri-tag| \
2091   udp-key|gre-key|fld-1st|fld-2nd|fld-3rd|fld-4th|fld-5th|fld-6th|fld-7th| \
2092   fld-8th|none) (select|add)
2093
2094For example, to add source IP to hash input set for flow type of ipv4-udp on port 0::
2095
2096   testpmd> set_hash_input_set 0 ipv4-udp src-ipv4 add
2097
2098set_fdir_input_set
2099~~~~~~~~~~~~~~~~~~
2100
2101The Flow Director filters can match the different fields for different type of packet, i.e. specific input set
2102on per flow type and the flexible payload. This command can be used to change input set for each flow type.
2103
2104Set the input set for flow director::
2105
2106   set_fdir_input_set (port_id) (ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
2107   ipv4-other|ipv6|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other| \
2108   l2_payload) (ivlan|ethertype|src-ipv4|dst-ipv4|src-ipv6|dst-ipv6|ipv4-tos| \
2109   ipv4-proto|ipv4-ttl|ipv6-tc|ipv6-next-header|ipv6-hop-limits| \
2110   tudp-src-port|udp-dst-port|cp-src-port|tcp-dst-port|sctp-src-port| \
2111   sctp-dst-port|sctp-veri-tag|none) (select|add)
2112
2113For example to add source IP to FD input set for flow type of ipv4-udp on port 0::
2114
2115   testpmd> set_fdir_input_set 0 ipv4-udp src-ipv4 add
2116
2117global_config
2118~~~~~~~~~~~~~
2119
2120Set different GRE key length for input set::
2121
2122   global_config (port_id) gre-key-len (number in bytes)
2123
2124For example to set GRE key length for input set to 4 bytes on port 0::
2125
2126   testpmd> global_config 0 gre-key-len 4
2127
2128
2129.. _testpmd_rte_flow:
2130
2131Flow rules management
2132---------------------
2133
2134Control of the generic flow API (*rte_flow*) is fully exposed through the
2135``flow`` command (validation, creation, destruction and queries).
2136
2137Considering *rte_flow* overlaps with all `Filter Functions`_, using both
2138features simultaneously may cause undefined side-effects and is therefore
2139not recommended.
2140
2141``flow`` syntax
2142~~~~~~~~~~~~~~~
2143
2144Because the ``flow`` command uses dynamic tokens to handle the large number
2145of possible flow rules combinations, its behavior differs slightly from
2146other commands, in particular:
2147
2148- Pressing *?* or the *<tab>* key displays contextual help for the current
2149  token, not that of the entire command.
2150
2151- Optional and repeated parameters are supported (provided they are listed
2152  in the contextual help).
2153
2154The first parameter stands for the operation mode. Possible operations and
2155their general syntax are described below. They are covered in detail in the
2156following sections.
2157
2158- Check whether a flow rule can be created::
2159
2160   flow validate {port_id}
2161       [group {group_id}] [priority {level}] [ingress] [egress]
2162       pattern {item} [/ {item} [...]] / end
2163       actions {action} [/ {action} [...]] / end
2164
2165- Create a flow rule::
2166
2167   flow create {port_id}
2168       [group {group_id}] [priority {level}] [ingress] [egress]
2169       pattern {item} [/ {item} [...]] / end
2170       actions {action} [/ {action} [...]] / end
2171
2172- Destroy specific flow rules::
2173
2174   flow destroy {port_id} rule {rule_id} [...]
2175
2176- Destroy all flow rules::
2177
2178   flow flush {port_id}
2179
2180- Query an existing flow rule::
2181
2182   flow query {port_id} {rule_id} {action}
2183
2184- List existing flow rules sorted by priority, filtered by group
2185  identifiers::
2186
2187   flow list {port_id} [group {group_id}] [...]
2188
2189Validating flow rules
2190~~~~~~~~~~~~~~~~~~~~~
2191
2192``flow validate`` reports whether a flow rule would be accepted by the
2193underlying device in its current state but stops short of creating it. It is
2194bound to ``rte_flow_validate()``::
2195
2196   flow validate {port_id}
2197      [group {group_id}] [priority {level}] [ingress] [egress]
2198      pattern {item} [/ {item} [...]] / end
2199      actions {action} [/ {action} [...]] / end
2200
2201If successful, it will show::
2202
2203   Flow rule validated
2204
2205Otherwise it will show an error message of the form::
2206
2207   Caught error type [...] ([...]): [...]
2208
2209This command uses the same parameters as ``flow create``, their format is
2210described in `Creating flow rules`_.
2211
2212Check whether redirecting any Ethernet packet received on port 0 to RX queue
2213index 6 is supported::
2214
2215   testpmd> flow validate 0 ingress pattern eth / end
2216      actions queue index 6 / end
2217   Flow rule validated
2218   testpmd>
2219
2220Port 0 does not support TCPv6 rules::
2221
2222   testpmd> flow validate 0 ingress pattern eth / ipv6 / tcp / end
2223      actions drop / end
2224   Caught error type 9 (specific pattern item): Invalid argument
2225   testpmd>
2226
2227Creating flow rules
2228~~~~~~~~~~~~~~~~~~~
2229
2230``flow create`` validates and creates the specified flow rule. It is bound
2231to ``rte_flow_create()``::
2232
2233   flow create {port_id}
2234      [group {group_id}] [priority {level}] [ingress] [egress]
2235      pattern {item} [/ {item} [...]] / end
2236      actions {action} [/ {action} [...]] / end
2237
2238If successful, it will return a flow rule ID usable with other commands::
2239
2240   Flow rule #[...] created
2241
2242Otherwise it will show an error message of the form::
2243
2244   Caught error type [...] ([...]): [...]
2245
2246Parameters describe in the following order:
2247
2248- Attributes (*group*, *priority*, *ingress*, *egress* tokens).
2249- A matching pattern, starting with the *pattern* token and terminated by an
2250  *end* pattern item.
2251- Actions, starting with the *actions* token and terminated by an *end*
2252  action.
2253
2254These translate directly to *rte_flow* objects provided as-is to the
2255underlying functions.
2256
2257The shortest valid definition only comprises mandatory tokens::
2258
2259   testpmd> flow create 0 pattern end actions end
2260
2261Note that PMDs may refuse rules that essentially do nothing such as this
2262one.
2263
2264**All unspecified object values are automatically initialized to 0.**
2265
2266Attributes
2267^^^^^^^^^^
2268
2269These tokens affect flow rule attributes (``struct rte_flow_attr``) and are
2270specified before the ``pattern`` token.
2271
2272- ``group {group id}``: priority group.
2273- ``priority {level}``: priority level within group.
2274- ``ingress``: rule applies to ingress traffic.
2275- ``egress``: rule applies to egress traffic.
2276
2277Each instance of an attribute specified several times overrides the previous
2278value as shown below (group 4 is used)::
2279
2280   testpmd> flow create 0 group 42 group 24 group 4 [...]
2281
2282Note that once enabled, ``ingress`` and ``egress`` cannot be disabled.
2283
2284While not specifying a direction is an error, some rules may allow both
2285simultaneously.
2286
2287Most rules affect RX therefore contain the ``ingress`` token::
2288
2289   testpmd> flow create 0 ingress pattern [...]
2290
2291Matching pattern
2292^^^^^^^^^^^^^^^^
2293
2294A matching pattern starts after the ``pattern`` token. It is made of pattern
2295items and is terminated by a mandatory ``end`` item.
2296
2297Items are named after their type (*RTE_FLOW_ITEM_TYPE_* from ``enum
2298rte_flow_item_type``).
2299
2300The ``/`` token is used as a separator between pattern items as shown
2301below::
2302
2303   testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end [...]
2304
2305Note that protocol items like these must be stacked from lowest to highest
2306layer to make sense. For instance, the following rule is either invalid or
2307unlikely to match any packet::
2308
2309   testpmd> flow create 0 ingress pattern eth / udp / ipv4 / end [...]
2310
2311More information on these restrictions can be found in the *rte_flow*
2312documentation.
2313
2314Several items support additional specification structures, for example
2315``ipv4`` allows specifying source and destination addresses as follows::
2316
2317   testpmd> flow create 0 ingress pattern eth / ipv4 src is 10.1.1.1
2318      dst is 10.2.0.0 / end [...]
2319
2320This rule matches all IPv4 traffic with the specified properties.
2321
2322In this example, ``src`` and ``dst`` are field names of the underlying
2323``struct rte_flow_item_ipv4`` object. All item properties can be specified
2324in a similar fashion.
2325
2326The ``is`` token means that the subsequent value must be matched exactly,
2327and assigns ``spec`` and ``mask`` fields in ``struct rte_flow_item``
2328accordingly. Possible assignment tokens are:
2329
2330- ``is``: match value perfectly (with full bit-mask).
2331- ``spec``: match value according to configured bit-mask.
2332- ``last``: specify upper bound to establish a range.
2333- ``mask``: specify bit-mask with relevant bits set to one.
2334- ``prefix``: generate bit-mask from a prefix length.
2335
2336These yield identical results::
2337
2338   ipv4 src is 10.1.1.1
2339
2340::
2341
2342   ipv4 src spec 10.1.1.1 src mask 255.255.255.255
2343
2344::
2345
2346   ipv4 src spec 10.1.1.1 src prefix 32
2347
2348::
2349
2350   ipv4 src is 10.1.1.1 src last 10.1.1.1 # range with a single value
2351
2352::
2353
2354   ipv4 src is 10.1.1.1 src last 0 # 0 disables range
2355
2356Inclusive ranges can be defined with ``last``::
2357
2358   ipv4 src is 10.1.1.1 src last 10.2.3.4 # 10.1.1.1 to 10.2.3.4
2359
2360Note that ``mask`` affects both ``spec`` and ``last``::
2361
2362   ipv4 src is 10.1.1.1 src last 10.2.3.4 src mask 255.255.0.0
2363      # matches 10.1.0.0 to 10.2.255.255
2364
2365Properties can be modified multiple times::
2366
2367   ipv4 src is 10.1.1.1 src is 10.1.2.3 src is 10.2.3.4 # matches 10.2.3.4
2368
2369::
2370
2371   ipv4 src is 10.1.1.1 src prefix 24 src prefix 16 # matches 10.1.0.0/16
2372
2373Pattern items
2374^^^^^^^^^^^^^
2375
2376This section lists supported pattern items and their attributes, if any.
2377
2378- ``end``: end list of pattern items.
2379
2380- ``void``: no-op pattern item.
2381
2382- ``invert``: perform actions when pattern does not match.
2383
2384- ``any``: match any protocol for the current layer.
2385
2386  - ``num {unsigned}``: number of layers covered.
2387
2388- ``pf``: match packets addressed to the physical function.
2389
2390- ``vf``: match packets addressed to a virtual function ID.
2391
2392  - ``id {unsigned}``: destination VF ID.
2393
2394- ``port``: device-specific physical port index to use.
2395
2396  - ``index {unsigned}``: physical port index.
2397
2398- ``raw``: match an arbitrary byte string.
2399
2400  - ``relative {boolean}``: look for pattern after the previous item.
2401  - ``search {boolean}``: search pattern from offset (see also limit).
2402  - ``offset {integer}``: absolute or relative offset for pattern.
2403  - ``limit {unsigned}``: search area limit for start of pattern.
2404  - ``pattern {string}``: byte string to look for.
2405
2406- ``eth``: match Ethernet header.
2407
2408  - ``dst {MAC-48}``: destination MAC.
2409  - ``src {MAC-48}``: source MAC.
2410  - ``type {unsigned}``: EtherType.
2411
2412- ``vlan``: match 802.1Q/ad VLAN tag.
2413
2414  - ``tpid {unsigned}``: tag protocol identifier.
2415  - ``tci {unsigned}``: tag control information.
2416  - ``pcp {unsigned}``: priority code point.
2417  - ``dei {unsigned}``: drop eligible indicator.
2418  - ``vid {unsigned}``: VLAN identifier.
2419
2420- ``ipv4``: match IPv4 header.
2421
2422  - ``tos {unsigned}``: type of service.
2423  - ``ttl {unsigned}``: time to live.
2424  - ``proto {unsigned}``: next protocol ID.
2425  - ``src {ipv4 address}``: source address.
2426  - ``dst {ipv4 address}``: destination address.
2427
2428- ``ipv6``: match IPv6 header.
2429
2430  - ``tc {unsigned}``: traffic class.
2431  - ``flow {unsigned}``: flow label.
2432  - ``proto {unsigned}``: protocol (next header).
2433  - ``hop {unsigned}``: hop limit.
2434  - ``src {ipv6 address}``: source address.
2435  - ``dst {ipv6 address}``: destination address.
2436
2437- ``icmp``: match ICMP header.
2438
2439  - ``type {unsigned}``: ICMP packet type.
2440  - ``code {unsigned}``: ICMP packet code.
2441
2442- ``udp``: match UDP header.
2443
2444  - ``src {unsigned}``: UDP source port.
2445  - ``dst {unsigned}``: UDP destination port.
2446
2447- ``tcp``: match TCP header.
2448
2449  - ``src {unsigned}``: TCP source port.
2450  - ``dst {unsigned}``: TCP destination port.
2451
2452- ``sctp``: match SCTP header.
2453
2454  - ``src {unsigned}``: SCTP source port.
2455  - ``dst {unsigned}``: SCTP destination port.
2456  - ``tag {unsigned}``: validation tag.
2457  - ``cksum {unsigned}``: checksum.
2458
2459- ``vxlan``: match VXLAN header.
2460
2461  - ``vni {unsigned}``: VXLAN identifier.
2462
2463Actions list
2464^^^^^^^^^^^^
2465
2466A list of actions starts after the ``actions`` token in the same fashion as
2467`Matching pattern`_; actions are separated by ``/`` tokens and the list is
2468terminated by a mandatory ``end`` action.
2469
2470Actions are named after their type (*RTE_FLOW_ACTION_TYPE_* from ``enum
2471rte_flow_action_type``).
2472
2473Dropping all incoming UDPv4 packets can be expressed as follows::
2474
2475   testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
2476      actions drop / end
2477
2478Several actions have configurable properties which must be specified when
2479there is no valid default value. For example, ``queue`` requires a target
2480queue index.
2481
2482This rule redirects incoming UDPv4 traffic to queue index 6::
2483
2484   testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
2485      actions queue index 6 / end
2486
2487While this one could be rejected by PMDs (unspecified queue index)::
2488
2489   testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
2490      actions queue / end
2491
2492As defined by *rte_flow*, the list is not ordered, all actions of a given
2493rule are performed simultaneously. These are equivalent::
2494
2495   queue index 6 / void / mark id 42 / end
2496
2497::
2498
2499   void / mark id 42 / queue index 6 / end
2500
2501All actions in a list should have different types, otherwise only the last
2502action of a given type is taken into account::
2503
2504   queue index 4 / queue index 5 / queue index 6 / end # will use queue 6
2505
2506::
2507
2508   drop / drop / drop / end # drop is performed only once
2509
2510::
2511
2512   mark id 42 / queue index 3 / mark id 24 / end # mark will be 24
2513
2514Considering they are performed simultaneously, opposite and overlapping
2515actions can sometimes be combined when the end result is unambiguous::
2516
2517   drop / queue index 6 / end # drop has no effect
2518
2519::
2520
2521   drop / dup index 6 / end # same as above
2522
2523::
2524
2525   queue index 6 / rss queues 6 7 8 / end # queue has no effect
2526
2527::
2528
2529   drop / passthru / end # drop has no effect
2530
2531Note that PMDs may still refuse such combinations.
2532
2533Actions
2534^^^^^^^
2535
2536This section lists supported actions and their attributes, if any.
2537
2538- ``end``: end list of actions.
2539
2540- ``void``: no-op action.
2541
2542- ``passthru``: let subsequent rule process matched packets.
2543
2544- ``mark``: attach 32 bit value to packets.
2545
2546  - ``id {unsigned}``: 32 bit value to return with packets.
2547
2548- ``flag``: flag packets.
2549
2550- ``queue``: assign packets to a given queue index.
2551
2552  - ``index {unsigned}``: queue index to use.
2553
2554- ``drop``: drop packets (note: passthru has priority).
2555
2556- ``count``: enable counters for this rule.
2557
2558- ``dup``: duplicate packets to a given queue index.
2559
2560  - ``index {unsigned}``: queue index to duplicate packets to.
2561
2562- ``rss``: spread packets among several queues.
2563
2564  - ``queues [{unsigned} [...]] end``: queue indices to use.
2565
2566- ``pf``: redirect packets to physical device function.
2567
2568- ``vf``: redirect packets to virtual device function.
2569
2570  - ``original {boolean}``: use original VF ID if possible.
2571  - ``id {unsigned}``: VF ID to redirect packets to.
2572
2573Destroying flow rules
2574~~~~~~~~~~~~~~~~~~~~~
2575
2576``flow destroy`` destroys one or more rules from their rule ID (as returned
2577by ``flow create``), this command calls ``rte_flow_destroy()`` as many
2578times as necessary::
2579
2580   flow destroy {port_id} rule {rule_id} [...]
2581
2582If successful, it will show::
2583
2584   Flow rule #[...] destroyed
2585
2586It does not report anything for rule IDs that do not exist. The usual error
2587message is shown when a rule cannot be destroyed::
2588
2589   Caught error type [...] ([...]): [...]
2590
2591``flow flush`` destroys all rules on a device and does not take extra
2592arguments. It is bound to ``rte_flow_flush()``::
2593
2594   flow flush {port_id}
2595
2596Any errors are reported as above.
2597
2598Creating several rules and destroying them::
2599
2600   testpmd> flow create 0 ingress pattern eth / ipv6 / end
2601      actions queue index 2 / end
2602   Flow rule #0 created
2603   testpmd> flow create 0 ingress pattern eth / ipv4 / end
2604      actions queue index 3 / end
2605   Flow rule #1 created
2606   testpmd> flow destroy 0 rule 0 rule 1
2607   Flow rule #1 destroyed
2608   Flow rule #0 destroyed
2609   testpmd>
2610
2611The same result can be achieved using ``flow flush``::
2612
2613   testpmd> flow create 0 ingress pattern eth / ipv6 / end
2614      actions queue index 2 / end
2615   Flow rule #0 created
2616   testpmd> flow create 0 ingress pattern eth / ipv4 / end
2617      actions queue index 3 / end
2618   Flow rule #1 created
2619   testpmd> flow flush 0
2620   testpmd>
2621
2622Non-existent rule IDs are ignored::
2623
2624   testpmd> flow create 0 ingress pattern eth / ipv6 / end
2625      actions queue index 2 / end
2626   Flow rule #0 created
2627   testpmd> flow create 0 ingress pattern eth / ipv4 / end
2628      actions queue index 3 / end
2629   Flow rule #1 created
2630   testpmd> flow destroy 0 rule 42 rule 10 rule 2
2631   testpmd>
2632   testpmd> flow destroy 0 rule 0
2633   Flow rule #0 destroyed
2634   testpmd>
2635
2636Querying flow rules
2637~~~~~~~~~~~~~~~~~~~
2638
2639``flow query`` queries a specific action of a flow rule having that
2640ability. Such actions collect information that can be reported using this
2641command. It is bound to ``rte_flow_query()``::
2642
2643   flow query {port_id} {rule_id} {action}
2644
2645If successful, it will display either the retrieved data for known actions
2646or the following message::
2647
2648   Cannot display result for action type [...] ([...])
2649
2650Otherwise, it will complain either that the rule does not exist or that some
2651error occurred::
2652
2653   Flow rule #[...] not found
2654
2655::
2656
2657   Caught error type [...] ([...]): [...]
2658
2659Currently only the ``count`` action is supported. This action reports the
2660number of packets that hit the flow rule and the total number of bytes. Its
2661output has the following format::
2662
2663   count:
2664    hits_set: [...] # whether "hits" contains a valid value
2665    bytes_set: [...] # whether "bytes" contains a valid value
2666    hits: [...] # number of packets
2667    bytes: [...] # number of bytes
2668
2669Querying counters for TCPv6 packets redirected to queue 6::
2670
2671   testpmd> flow create 0 ingress pattern eth / ipv6 / tcp / end
2672      actions queue index 6 / count / end
2673   Flow rule #4 created
2674   testpmd> flow query 0 4 count
2675   count:
2676    hits_set: 1
2677    bytes_set: 0
2678    hits: 386446
2679    bytes: 0
2680   testpmd>
2681
2682Listing flow rules
2683~~~~~~~~~~~~~~~~~~
2684
2685``flow list`` lists existing flow rules sorted by priority and optionally
2686filtered by group identifiers::
2687
2688   flow list {port_id} [group {group_id}] [...]
2689
2690This command only fails with the following message if the device does not
2691exist::
2692
2693   Invalid port [...]
2694
2695Output consists of a header line followed by a short description of each
2696flow rule, one per line. There is no output at all when no flow rules are
2697configured on the device::
2698
2699   ID      Group   Prio    Attr    Rule
2700   [...]   [...]   [...]   [...]   [...]
2701
2702``Attr`` column flags:
2703
2704- ``i`` for ``ingress``.
2705- ``e`` for ``egress``.
2706
2707Creating several flow rules and listing them::
2708
2709   testpmd> flow create 0 ingress pattern eth / ipv4 / end
2710      actions queue index 6 / end
2711   Flow rule #0 created
2712   testpmd> flow create 0 ingress pattern eth / ipv6 / end
2713      actions queue index 2 / end
2714   Flow rule #1 created
2715   testpmd> flow create 0 priority 5 ingress pattern eth / ipv4 / udp / end
2716      actions rss queues 6 7 8 end / end
2717   Flow rule #2 created
2718   testpmd> flow list 0
2719   ID      Group   Prio    Attr    Rule
2720   0       0       0       i-      ETH IPV4 => QUEUE
2721   1       0       0       i-      ETH IPV6 => QUEUE
2722   2       0       5       i-      ETH IPV4 UDP => RSS
2723   testpmd>
2724
2725Rules are sorted by priority (i.e. group ID first, then priority level)::
2726
2727   testpmd> flow list 1
2728   ID      Group   Prio    Attr    Rule
2729   0       0       0       i-      ETH => COUNT
2730   6       0       500     i-      ETH IPV6 TCP => DROP COUNT
2731   5       0       1000    i-      ETH IPV6 ICMP => QUEUE
2732   1       24      0       i-      ETH IPV4 UDP => QUEUE
2733   4       24      10      i-      ETH IPV4 TCP => DROP
2734   3       24      20      i-      ETH IPV4 => DROP
2735   2       24      42      i-      ETH IPV4 UDP => QUEUE
2736   7       63      0       i-      ETH IPV6 UDP VXLAN => MARK QUEUE
2737   testpmd>
2738
2739Output can be limited to specific groups::
2740
2741   testpmd> flow list 1 group 0 group 63
2742   ID      Group   Prio    Attr    Rule
2743   0       0       0       i-      ETH => COUNT
2744   6       0       500     i-      ETH IPV6 TCP => DROP COUNT
2745   5       0       1000    i-      ETH IPV6 ICMP => QUEUE
2746   7       63      0       i-      ETH IPV6 UDP VXLAN => MARK QUEUE
2747   testpmd>
2748