xref: /dpdk/doc/guides/testpmd_app_ug/testpmd_funcs.rst (revision d58a3f312545d3afbf4fbb0cc2220161f790d007)
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 (for VF)
841~~~~~~~~~~~~~~~~~~~~~
842
843Set the MAC address for a VF from the PF::
844
845   testpmd> set vf mac addr (port_id) (vf_id) (XX:XX:XX:XX:XX:XX)
846
847set port-uta
848~~~~~~~~~~~~
849
850Set the unicast hash filter(s) on/off for a port::
851
852   testpmd> set port (port_id) uta (XX:XX:XX:XX:XX:XX|all) (on|off)
853
854set promisc
855~~~~~~~~~~~
856
857Set the promiscuous mode on for a port or for all ports.
858In promiscuous mode packets are not dropped if they aren't for the specified MAC address::
859
860   testpmd> set promisc (port_id|all) (on|off)
861
862set allmulti
863~~~~~~~~~~~~
864
865Set the allmulti mode for a port or for all ports::
866
867   testpmd> set allmulti (port_id|all) (on|off)
868
869Same as the ifconfig (8) option. Controls how multicast packets are handled.
870
871set promisc (for VF)
872~~~~~~~~~~~~~~~~~~~~
873
874Set the unicast promiscuous mode for a VF from PF.
875It's supported by Intel i40e NICs now.
876In promiscuous mode packets are not dropped if they aren't for the specified MAC address::
877
878   testpmd> set vf promisc (port_id) (vf_id) (on|off)
879
880set allmulticast (for VF)
881~~~~~~~~~~~~~~~~~~~~~~~~~
882
883Set the multicast promiscuous mode for a VF from PF.
884It's supported by Intel i40e NICs now.
885In promiscuous mode packets are not dropped if they aren't for the specified MAC address::
886
887   testpmd> set vf allmulti (port_id) (vf_id) (on|off)
888
889set flow_ctrl rx
890~~~~~~~~~~~~~~~~
891
892Set the link flow control parameter on a port::
893
894   testpmd> set flow_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \
895            (pause_time) (send_xon) mac_ctrl_frame_fwd (on|off) \
896	    autoneg (on|off) (port_id)
897
898Where:
899
900* ``high_water`` (integer): High threshold value to trigger XOFF.
901
902* ``low_water`` (integer): Low threshold value to trigger XON.
903
904* ``pause_time`` (integer): Pause quota in the Pause frame.
905
906* ``send_xon`` (0/1): Send XON frame.
907
908* ``mac_ctrl_frame_fwd``: Enable receiving MAC control frames.
909
910* ``autoneg``: Change the auto-negotiation parameter.
911
912set pfc_ctrl rx
913~~~~~~~~~~~~~~~
914
915Set the priority flow control parameter on a port::
916
917   testpmd> set pfc_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \
918            (pause_time) (priority) (port_id)
919
920Where:
921
922* ``high_water`` (integer): High threshold value.
923
924* ``low_water`` (integer): Low threshold value.
925
926* ``pause_time`` (integer): Pause quota in the Pause frame.
927
928* ``priority`` (0-7): VLAN User Priority.
929
930set stat_qmap
931~~~~~~~~~~~~~
932
933Set statistics mapping (qmapping 0..15) for RX/TX queue on port::
934
935   testpmd> set stat_qmap (tx|rx) (port_id) (queue_id) (qmapping)
936
937For example, to set rx queue 2 on port 0 to mapping 5::
938
939   testpmd>set stat_qmap rx 0 2 5
940
941set port - rx/tx (for VF)
942~~~~~~~~~~~~~~~~~~~~~~~~~
943
944Set VF receive/transmit from a port::
945
946   testpmd> set port (port_id) vf (vf_id) (rx|tx) (on|off)
947
948set port - mac address filter (for VF)
949~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
950
951Add/Remove unicast or multicast MAC addr filter for a VF::
952
953   testpmd> set port (port_id) vf (vf_id) (mac_addr) \
954            (exact-mac|exact-mac-vlan|hashmac|hashmac-vlan) (on|off)
955
956set port - rx mode(for VF)
957~~~~~~~~~~~~~~~~~~~~~~~~~~
958
959Set the VF receive mode of a port::
960
961   testpmd> set port (port_id) vf (vf_id) \
962            rxmode (AUPE|ROPE|BAM|MPE) (on|off)
963
964The available receive modes are:
965
966* ``AUPE``: Accepts untagged VLAN.
967
968* ``ROPE``: Accepts unicast hash.
969
970* ``BAM``: Accepts broadcast packets.
971
972* ``MPE``: Accepts all multicast packets.
973
974set port - tx_rate (for Queue)
975~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
976
977Set TX rate limitation for a queue on a port::
978
979   testpmd> set port (port_id) queue (queue_id) rate (rate_value)
980
981set port - tx_rate (for VF)
982~~~~~~~~~~~~~~~~~~~~~~~~~~~
983
984Set TX rate limitation for queues in VF on a port::
985
986   testpmd> set port (port_id) vf (vf_id) rate (rate_value) queue_mask (queue_mask)
987
988set port - mirror rule
989~~~~~~~~~~~~~~~~~~~~~~
990
991Set pool or vlan type mirror rule for a port::
992
993   testpmd> set port (port_id) mirror-rule (rule_id) \
994            (pool-mirror-up|pool-mirror-down|vlan-mirror) \
995            (poolmask|vlanid[,vlanid]*) dst-pool (pool_id) (on|off)
996
997Set link mirror rule for a port::
998
999   testpmd> set port (port_id) mirror-rule (rule_id) \
1000           (uplink-mirror|downlink-mirror) dst-pool (pool_id) (on|off)
1001
1002For example to enable mirror traffic with vlan 0,1 to pool 0::
1003
1004   set port 0 mirror-rule 0 vlan-mirror 0,1 dst-pool 0 on
1005
1006reset port - mirror rule
1007~~~~~~~~~~~~~~~~~~~~~~~~
1008
1009Reset a mirror rule for a port::
1010
1011   testpmd> reset port (port_id) mirror-rule (rule_id)
1012
1013set flush_rx
1014~~~~~~~~~~~~
1015
1016Set the flush on RX streams before forwarding.
1017The default is flush ``on``.
1018Mainly used with PCAP drivers to turn off the default behavior of flushing the first 512 packets on RX streams::
1019
1020   testpmd> set flush_rx off
1021
1022set bypass mode
1023~~~~~~~~~~~~~~~
1024
1025Set the bypass mode for the lowest port on bypass enabled NIC::
1026
1027   testpmd> set bypass mode (normal|bypass|isolate) (port_id)
1028
1029set bypass event
1030~~~~~~~~~~~~~~~~
1031
1032Set the event required to initiate specified bypass mode for the lowest port on a bypass enabled::
1033
1034   testpmd> set bypass event (timeout|os_on|os_off|power_on|power_off) \
1035            mode (normal|bypass|isolate) (port_id)
1036
1037Where:
1038
1039* ``timeout``: Enable bypass after watchdog timeout.
1040
1041* ``os_on``: Enable bypass when OS/board is powered on.
1042
1043* ``os_off``: Enable bypass when OS/board is powered off.
1044
1045* ``power_on``: Enable bypass when power supply is turned on.
1046
1047* ``power_off``: Enable bypass when power supply is turned off.
1048
1049
1050set bypass timeout
1051~~~~~~~~~~~~~~~~~~
1052
1053Set the bypass watchdog timeout to ``n`` seconds where 0 = instant::
1054
1055   testpmd> set bypass timeout (0|1.5|2|3|4|8|16|32)
1056
1057show bypass config
1058~~~~~~~~~~~~~~~~~~
1059
1060Show the bypass configuration for a bypass enabled NIC using the lowest port on the NIC::
1061
1062   testpmd> show bypass config (port_id)
1063
1064set link up
1065~~~~~~~~~~~
1066
1067Set link up for a port::
1068
1069   testpmd> set link-up port (port id)
1070
1071set link down
1072~~~~~~~~~~~~~
1073
1074Set link down for a port::
1075
1076   testpmd> set link-down port (port id)
1077
1078E-tag set
1079~~~~~~~~~
1080
1081Enable E-tag insertion for a VF on a port::
1082
1083   testpmd> E-tag set insertion on port-tag-id (value) port (port_id) vf (vf_id)
1084
1085Disable E-tag insertion for a VF on a port::
1086
1087   testpmd> E-tag set insertion off port (port_id) vf (vf_id)
1088
1089Enable/disable E-tag stripping on a port::
1090
1091   testpmd> E-tag set stripping (on|off) port (port_id)
1092
1093Enable/disable E-tag based forwarding on a port::
1094
1095   testpmd> E-tag set forwarding (on|off) port (port_id)
1096
1097Add an E-tag forwarding filter on a port::
1098
1099   testpmd> E-tag set filter add e-tag-id (value) dst-pool (pool_id) port (port_id)
1100
1101Delete an E-tag forwarding filter on a port::
1102   testpmd> E-tag set filter del e-tag-id (value) port (port_id)
1103
1104
1105Port Functions
1106--------------
1107
1108The following sections show functions for configuring ports.
1109
1110.. note::
1111
1112   Port configuration changes only become active when forwarding is started/restarted.
1113
1114port attach
1115~~~~~~~~~~~
1116
1117Attach a port specified by pci address or virtual device args::
1118
1119   testpmd> port attach (identifier)
1120
1121To attach a new pci device, the device should be recognized by kernel first.
1122Then it should be moved under DPDK management.
1123Finally the port can be attached to testpmd.
1124
1125For example, to move a pci device using ixgbe under DPDK management:
1126
1127.. code-block:: console
1128
1129   # Check the status of the available devices.
1130   ./usertools/dpdk-devbind.py --status
1131
1132   Network devices using DPDK-compatible driver
1133   ============================================
1134   <none>
1135
1136   Network devices using kernel driver
1137   ===================================
1138   0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused=
1139
1140
1141   # Bind the device to igb_uio.
1142   sudo ./usertools/dpdk-devbind.py -b igb_uio 0000:0a:00.0
1143
1144
1145   # Recheck the status of the devices.
1146   ./usertools/dpdk-devbind.py --status
1147   Network devices using DPDK-compatible driver
1148   ============================================
1149   0000:0a:00.0 '82599ES 10-Gigabit' drv=igb_uio unused=
1150
1151To attach a port created by virtual device, above steps are not needed.
1152
1153For example, to attach a port whose pci address is 0000:0a:00.0.
1154
1155.. code-block:: console
1156
1157   testpmd> port attach 0000:0a:00.0
1158   Attaching a new port...
1159   EAL: PCI device 0000:0a:00.0 on NUMA socket -1
1160   EAL:   probe driver: 8086:10fb rte_ixgbe_pmd
1161   EAL:   PCI memory mapped at 0x7f83bfa00000
1162   EAL:   PCI memory mapped at 0x7f83bfa80000
1163   PMD: eth_ixgbe_dev_init(): MAC: 2, PHY: 18, SFP+: 5
1164   PMD: eth_ixgbe_dev_init(): port 0 vendorID=0x8086 deviceID=0x10fb
1165   Port 0 is attached. Now total ports is 1
1166   Done
1167
1168For example, to attach a port created by pcap PMD.
1169
1170.. code-block:: console
1171
1172   testpmd> port attach net_pcap0
1173   Attaching a new port...
1174   PMD: Initializing pmd_pcap for net_pcap0
1175   PMD: Creating pcap-backed ethdev on numa socket 0
1176   Port 0 is attached. Now total ports is 1
1177   Done
1178
1179In this case, identifier is ``net_pcap0``.
1180This identifier format is the same as ``--vdev`` format of DPDK applications.
1181
1182For example, to re-attach a bonded port which has been previously detached,
1183the mode and slave parameters must be given.
1184
1185.. code-block:: console
1186
1187   testpmd> port attach net_bond_0,mode=0,slave=1
1188   Attaching a new port...
1189   EAL: Initializing pmd_bond for net_bond_0
1190   EAL: Create bonded device net_bond_0 on port 0 in mode 0 on socket 0.
1191   Port 0 is attached. Now total ports is 1
1192   Done
1193
1194
1195port detach
1196~~~~~~~~~~~
1197
1198Detach a specific port::
1199
1200   testpmd> port detach (port_id)
1201
1202Before detaching a port, the port should be stopped and closed.
1203
1204For example, to detach a pci device port 0.
1205
1206.. code-block:: console
1207
1208   testpmd> port stop 0
1209   Stopping ports...
1210   Done
1211   testpmd> port close 0
1212   Closing ports...
1213   Done
1214
1215   testpmd> port detach 0
1216   Detaching a port...
1217   EAL: PCI device 0000:0a:00.0 on NUMA socket -1
1218   EAL:   remove driver: 8086:10fb rte_ixgbe_pmd
1219   EAL:   PCI memory unmapped at 0x7f83bfa00000
1220   EAL:   PCI memory unmapped at 0x7f83bfa80000
1221   Done
1222
1223
1224For example, to detach a virtual device port 0.
1225
1226.. code-block:: console
1227
1228   testpmd> port stop 0
1229   Stopping ports...
1230   Done
1231   testpmd> port close 0
1232   Closing ports...
1233   Done
1234
1235   testpmd> port detach 0
1236   Detaching a port...
1237   PMD: Closing pcap ethdev on numa socket 0
1238   Port 'net_pcap0' is detached. Now total ports is 0
1239   Done
1240
1241To remove a pci device completely from the system, first detach the port from testpmd.
1242Then the device should be moved under kernel management.
1243Finally the device can be removed using kernel pci hotplug functionality.
1244
1245For example, to move a pci device under kernel management:
1246
1247.. code-block:: console
1248
1249   sudo ./usertools/dpdk-devbind.py -b ixgbe 0000:0a:00.0
1250
1251   ./usertools/dpdk-devbind.py --status
1252
1253   Network devices using DPDK-compatible driver
1254   ============================================
1255   <none>
1256
1257   Network devices using kernel driver
1258   ===================================
1259   0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused=igb_uio
1260
1261To remove a port created by a virtual device, above steps are not needed.
1262
1263port start
1264~~~~~~~~~~
1265
1266Start all ports or a specific port::
1267
1268   testpmd> port start (port_id|all)
1269
1270port stop
1271~~~~~~~~~
1272
1273Stop all ports or a specific port::
1274
1275   testpmd> port stop (port_id|all)
1276
1277port close
1278~~~~~~~~~~
1279
1280Close all ports or a specific port::
1281
1282   testpmd> port close (port_id|all)
1283
1284port start/stop queue
1285~~~~~~~~~~~~~~~~~~~~~
1286
1287Start/stop a rx/tx queue on a specific port::
1288
1289   testpmd> port (port_id) (rxq|txq) (queue_id) (start|stop)
1290
1291Only take effect when port is started.
1292
1293port config - speed
1294~~~~~~~~~~~~~~~~~~~
1295
1296Set the speed and duplex mode for all ports or a specific port::
1297
1298   testpmd> port config (port_id|all) speed (10|100|1000|10000|25000|40000|50000|100000|auto) \
1299            duplex (half|full|auto)
1300
1301port config - queues/descriptors
1302~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1303
1304Set number of queues/descriptors for rxq, txq, rxd and txd::
1305
1306   testpmd> port config all (rxq|txq|rxd|txd) (value)
1307
1308This is equivalent to the ``--rxq``, ``--txq``, ``--rxd`` and ``--txd`` command-line options.
1309
1310port config - max-pkt-len
1311~~~~~~~~~~~~~~~~~~~~~~~~~
1312
1313Set the maximum packet length::
1314
1315   testpmd> port config all max-pkt-len (value)
1316
1317This is equivalent to the ``--max-pkt-len`` command-line option.
1318
1319port config - CRC Strip
1320~~~~~~~~~~~~~~~~~~~~~~~
1321
1322Set hardware CRC stripping on or off for all ports::
1323
1324   testpmd> port config all crc-strip (on|off)
1325
1326CRC stripping is off by default.
1327
1328The ``on`` option is equivalent to the ``--crc-strip`` command-line option.
1329
1330port config - scatter
1331~~~~~~~~~~~~~~~~~~~~~~~
1332
1333Set RX scatter mode on or off for all ports::
1334
1335   testpmd> port config all scatter (on|off)
1336
1337RX scatter mode is off by default.
1338
1339The ``on`` option is equivalent to the ``--enable-scatter`` command-line option.
1340
1341port config - TX queue flags
1342~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1343
1344Set a hexadecimal bitmap of TX queue flags for all ports::
1345
1346   testpmd> port config all txqflags value
1347
1348This command is equivalent to the ``--txqflags`` command-line option.
1349
1350port config - RX Checksum
1351~~~~~~~~~~~~~~~~~~~~~~~~~
1352
1353Set hardware RX checksum offload to on or off for all ports::
1354
1355   testpmd> port config all rx-cksum (on|off)
1356
1357Checksum offload is off by default.
1358
1359The ``on`` option is equivalent to the ``--enable-rx-cksum`` command-line option.
1360
1361port config - VLAN
1362~~~~~~~~~~~~~~~~~~
1363
1364Set hardware VLAN on or off for all ports::
1365
1366   testpmd> port config all hw-vlan (on|off)
1367
1368Hardware VLAN is on by default.
1369
1370The ``off`` option is equivalent to the ``--disable-hw-vlan`` command-line option.
1371
1372port config - VLAN filter
1373~~~~~~~~~~~~~~~~~~~~~~~~~
1374
1375Set hardware VLAN filter on or off for all ports::
1376
1377   testpmd> port config all hw-vlan-filter (on|off)
1378
1379Hardware VLAN filter is on by default.
1380
1381The ``off`` option is equivalent to the ``--disable-hw-vlan-filter`` command-line option.
1382
1383port config - VLAN strip
1384~~~~~~~~~~~~~~~~~~~~~~~~
1385
1386Set hardware VLAN strip on or off for all ports::
1387
1388   testpmd> port config all hw-vlan-strip (on|off)
1389
1390Hardware VLAN strip is on by default.
1391
1392The ``off`` option is equivalent to the ``--disable-hw-vlan-strip`` command-line option.
1393
1394port config - VLAN extend
1395~~~~~~~~~~~~~~~~~~~~~~~~~
1396
1397Set hardware VLAN extend on or off for all ports::
1398
1399   testpmd> port config all hw-vlan-extend (on|off)
1400
1401Hardware VLAN extend is off by default.
1402
1403The ``off`` option is equivalent to the ``--disable-hw-vlan-extend`` command-line option.
1404
1405port config - Drop Packets
1406~~~~~~~~~~~~~~~~~~~~~~~~~~
1407
1408Set packet drop for packets with no descriptors on or off for all ports::
1409
1410   testpmd> port config all drop-en (on|off)
1411
1412Packet dropping for packets with no descriptors is off by default.
1413
1414The ``on`` option is equivalent to the ``--enable-drop-en`` command-line option.
1415
1416port config - RSS
1417~~~~~~~~~~~~~~~~~
1418
1419Set the RSS (Receive Side Scaling) mode on or off::
1420
1421   testpmd> port config all rss (all|ip|tcp|udp|sctp|ether|port|vxlan|geneve|nvgre|none)
1422
1423RSS is on by default.
1424
1425The ``none`` option is equivalent to the ``--disable-rss`` command-line option.
1426
1427port config - RSS Reta
1428~~~~~~~~~~~~~~~~~~~~~~
1429
1430Set the RSS (Receive Side Scaling) redirection table::
1431
1432   testpmd> port config all rss reta (hash,queue)[,(hash,queue)]
1433
1434port config - DCB
1435~~~~~~~~~~~~~~~~~
1436
1437Set the DCB mode for an individual port::
1438
1439   testpmd> port config (port_id) dcb vt (on|off) (traffic_class) pfc (on|off)
1440
1441The traffic class should be 4 or 8.
1442
1443port config - Burst
1444~~~~~~~~~~~~~~~~~~~
1445
1446Set the number of packets per burst::
1447
1448   testpmd> port config all burst (value)
1449
1450This is equivalent to the ``--burst`` command-line option.
1451
1452port config - Threshold
1453~~~~~~~~~~~~~~~~~~~~~~~
1454
1455Set thresholds for TX/RX queues::
1456
1457   testpmd> port config all (threshold) (value)
1458
1459Where the threshold type can be:
1460
1461* ``txpt:`` Set the prefetch threshold register of the TX rings, 0 <= value <= 255.
1462
1463* ``txht:`` Set the host threshold register of the TX rings, 0 <= value <= 255.
1464
1465* ``txwt:`` Set the write-back threshold register of the TX rings, 0 <= value <= 255.
1466
1467* ``rxpt:`` Set the prefetch threshold register of the RX rings, 0 <= value <= 255.
1468
1469* ``rxht:`` Set the host threshold register of the RX rings, 0 <= value <= 255.
1470
1471* ``rxwt:`` Set the write-back threshold register of the RX rings, 0 <= value <= 255.
1472
1473* ``txfreet:`` Set the transmit free threshold of the TX rings, 0 <= value <= txd.
1474
1475* ``rxfreet:`` Set the transmit free threshold of the RX rings, 0 <= value <= rxd.
1476
1477* ``txrst:`` Set the transmit RS bit threshold of TX rings, 0 <= value <= txd.
1478
1479These threshold options are also available from the command-line.
1480
1481port config - E-tag
1482~~~~~~~~~~~~~~~~~~~
1483
1484Set the value of ether-type for E-tag::
1485
1486   testpmd> port config (port_id|all) l2-tunnel E-tag ether-type (value)
1487
1488Enable/disable the E-tag support::
1489
1490   testpmd> port config (port_id|all) l2-tunnel E-tag (enable|disable)
1491
1492
1493Link Bonding Functions
1494----------------------
1495
1496The Link Bonding functions make it possible to dynamically create and
1497manage link bonding devices from within testpmd interactive prompt.
1498
1499create bonded device
1500~~~~~~~~~~~~~~~~~~~~
1501
1502Create a new bonding device::
1503
1504   testpmd> create bonded device (mode) (socket)
1505
1506For example, to create a bonded device in mode 1 on socket 0::
1507
1508   testpmd> create bonded 1 0
1509   created new bonded device (port X)
1510
1511add bonding slave
1512~~~~~~~~~~~~~~~~~
1513
1514Adds Ethernet device to a Link Bonding device::
1515
1516   testpmd> add bonding slave (slave id) (port id)
1517
1518For example, to add Ethernet device (port 6) to a Link Bonding device (port 10)::
1519
1520   testpmd> add bonding slave 6 10
1521
1522
1523remove bonding slave
1524~~~~~~~~~~~~~~~~~~~~
1525
1526Removes an Ethernet slave device from a Link Bonding device::
1527
1528   testpmd> remove bonding slave (slave id) (port id)
1529
1530For example, to remove Ethernet slave device (port 6) to a Link Bonding device (port 10)::
1531
1532   testpmd> remove bonding slave 6 10
1533
1534set bonding mode
1535~~~~~~~~~~~~~~~~
1536
1537Set the Link Bonding mode of a Link Bonding device::
1538
1539   testpmd> set bonding mode (value) (port id)
1540
1541For example, to set the bonding mode of a Link Bonding device (port 10) to broadcast (mode 3)::
1542
1543   testpmd> set bonding mode 3 10
1544
1545set bonding primary
1546~~~~~~~~~~~~~~~~~~~
1547
1548Set an Ethernet slave device as the primary device on a Link Bonding device::
1549
1550   testpmd> set bonding primary (slave id) (port id)
1551
1552For example, to set the Ethernet slave device (port 6) as the primary port of a Link Bonding device (port 10)::
1553
1554   testpmd> set bonding primary 6 10
1555
1556set bonding mac
1557~~~~~~~~~~~~~~~
1558
1559Set the MAC address of a Link Bonding device::
1560
1561   testpmd> set bonding mac (port id) (mac)
1562
1563For example, to set the MAC address of a Link Bonding device (port 10) to 00:00:00:00:00:01::
1564
1565   testpmd> set bonding mac 10 00:00:00:00:00:01
1566
1567set bonding xmit_balance_policy
1568~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1569
1570Set the transmission policy for a Link Bonding device when it is in Balance XOR mode::
1571
1572   testpmd> set bonding xmit_balance_policy (port_id) (l2|l23|l34)
1573
1574For example, set a Link Bonding device (port 10) to use a balance policy of layer 3+4 (IP addresses & UDP ports)::
1575
1576   testpmd> set bonding xmit_balance_policy 10 l34
1577
1578
1579set bonding mon_period
1580~~~~~~~~~~~~~~~~~~~~~~
1581
1582Set the link status monitoring polling period in milliseconds for a bonding device.
1583
1584This adds support for PMD slave devices which do not support link status interrupts.
1585When the mon_period is set to a value greater than 0 then all PMD's which do not support
1586link status ISR will be queried every polling interval to check if their link status has changed::
1587
1588   testpmd> set bonding mon_period (port_id) (value)
1589
1590For example, to set the link status monitoring polling period of bonded device (port 5) to 150ms::
1591
1592   testpmd> set bonding mon_period 5 150
1593
1594
1595show bonding config
1596~~~~~~~~~~~~~~~~~~~
1597
1598Show the current configuration of a Link Bonding device::
1599
1600   testpmd> show bonding config (port id)
1601
1602For example,
1603to show the configuration a Link Bonding device (port 9) with 3 slave devices (1, 3, 4)
1604in balance mode with a transmission policy of layer 2+3::
1605
1606   testpmd> show bonding config 9
1607        Bonding mode: 2
1608        Balance Xmit Policy: BALANCE_XMIT_POLICY_LAYER23
1609        Slaves (3): [1 3 4]
1610        Active Slaves (3): [1 3 4]
1611        Primary: [3]
1612
1613
1614Register Functions
1615------------------
1616
1617The Register Functions can be used to read from and write to registers on the network card referenced by a port number.
1618This is mainly useful for debugging purposes.
1619Reference should be made to the appropriate datasheet for the network card for details on the register addresses
1620and fields that can be accessed.
1621
1622read reg
1623~~~~~~~~
1624
1625Display the value of a port register::
1626
1627   testpmd> read reg (port_id) (address)
1628
1629For example, to examine the Flow Director control register (FDIRCTL, 0x0000EE000) on an Intel 82599 10 GbE Controller::
1630
1631   testpmd> read reg 0 0xEE00
1632   port 0 PCI register at offset 0xEE00: 0x4A060029 (1241907241)
1633
1634read regfield
1635~~~~~~~~~~~~~
1636
1637Display a port register bit field::
1638
1639   testpmd> read regfield (port_id) (address) (bit_x) (bit_y)
1640
1641For example, reading the lowest two bits from the register in the example above::
1642
1643   testpmd> read regfield 0 0xEE00 0 1
1644   port 0 PCI register at offset 0xEE00: bits[0, 1]=0x1 (1)
1645
1646read regbit
1647~~~~~~~~~~~
1648
1649Display a single port register bit::
1650
1651   testpmd> read regbit (port_id) (address) (bit_x)
1652
1653For example, reading the lowest bit from the register in the example above::
1654
1655   testpmd> read regbit 0 0xEE00 0
1656   port 0 PCI register at offset 0xEE00: bit 0=1
1657
1658write reg
1659~~~~~~~~~
1660
1661Set the value of a port register::
1662
1663   testpmd> write reg (port_id) (address) (value)
1664
1665For example, to clear a register::
1666
1667   testpmd> write reg 0 0xEE00 0x0
1668   port 0 PCI register at offset 0xEE00: 0x00000000 (0)
1669
1670write regfield
1671~~~~~~~~~~~~~~
1672
1673Set bit field of a port register::
1674
1675   testpmd> write regfield (port_id) (address) (bit_x) (bit_y) (value)
1676
1677For example, writing to the register cleared in the example above::
1678
1679   testpmd> write regfield 0 0xEE00 0 1 2
1680   port 0 PCI register at offset 0xEE00: 0x00000002 (2)
1681
1682write regbit
1683~~~~~~~~~~~~
1684
1685Set single bit value of a port register::
1686
1687   testpmd> write regbit (port_id) (address) (bit_x) (value)
1688
1689For example, to set the high bit in the register from the example above::
1690
1691   testpmd> write regbit 0 0xEE00 31 1
1692   port 0 PCI register at offset 0xEE00: 0x8000000A (2147483658)
1693
1694
1695Filter Functions
1696----------------
1697
1698This section details the available filter functions that are available.
1699
1700Note these functions interface the deprecated legacy filtering framework,
1701superseded by *rte_flow*. See `Flow rules management`_.
1702
1703ethertype_filter
1704~~~~~~~~~~~~~~~~~~~~
1705
1706Add or delete a L2 Ethertype filter, which identify packets by their L2 Ethertype mainly assign them to a receive queue::
1707
1708   ethertype_filter (port_id) (add|del) (mac_addr|mac_ignr) (mac_address) \
1709                    ethertype (ether_type) (drop|fwd) queue (queue_id)
1710
1711The available information parameters are:
1712
1713* ``port_id``: The port which the Ethertype filter assigned on.
1714
1715* ``mac_addr``: Compare destination mac address.
1716
1717* ``mac_ignr``: Ignore destination mac address match.
1718
1719* ``mac_address``: Destination mac address to match.
1720
1721* ``ether_type``: The EtherType value want to match,
1722  for example 0x0806 for ARP packet. 0x0800 (IPv4) and 0x86DD (IPv6) are invalid.
1723
1724* ``queue_id``: The receive queue associated with this EtherType filter.
1725  It is meaningless when deleting or dropping.
1726
1727Example, to add/remove an ethertype filter rule::
1728
1729   testpmd> ethertype_filter 0 add mac_ignr 00:11:22:33:44:55 \
1730                             ethertype 0x0806 fwd queue 3
1731
1732   testpmd> ethertype_filter 0 del mac_ignr 00:11:22:33:44:55 \
1733                             ethertype 0x0806 fwd queue 3
1734
17352tuple_filter
1736~~~~~~~~~~~~~~~~~
1737
1738Add or delete a 2-tuple filter,
1739which identifies packets by specific protocol and destination TCP/UDP port
1740and forwards packets into one of the receive queues::
1741
1742   2tuple_filter (port_id) (add|del) dst_port (dst_port_value) \
1743                 protocol (protocol_value) mask (mask_value) \
1744                 tcp_flags (tcp_flags_value) priority (prio_value) \
1745                 queue (queue_id)
1746
1747The available information parameters are:
1748
1749* ``port_id``: The port which the 2-tuple filter assigned on.
1750
1751* ``dst_port_value``: Destination port in L4.
1752
1753* ``protocol_value``: IP L4 protocol.
1754
1755* ``mask_value``: Participates in the match or not by bit for field above, 1b means participate.
1756
1757* ``tcp_flags_value``: TCP control bits. The non-zero value is invalid, when the pro_value is not set to 0x06 (TCP).
1758
1759* ``prio_value``: Priority of this filter.
1760
1761* ``queue_id``: The receive queue associated with this 2-tuple filter.
1762
1763Example, to add/remove an 2tuple filter rule::
1764
1765   testpmd> 2tuple_filter 0 add dst_port 32 protocol 0x06 mask 0x03 \
1766                          tcp_flags 0x02 priority 3 queue 3
1767
1768   testpmd> 2tuple_filter 0 del dst_port 32 protocol 0x06 mask 0x03 \
1769                          tcp_flags 0x02 priority 3 queue 3
1770
17715tuple_filter
1772~~~~~~~~~~~~~~~~~
1773
1774Add or delete a 5-tuple filter,
1775which consists of a 5-tuple (protocol, source and destination IP addresses, source and destination TCP/UDP/SCTP port)
1776and routes packets into one of the receive queues::
1777
1778   5tuple_filter (port_id) (add|del) dst_ip (dst_address) src_ip \
1779                 (src_address) dst_port (dst_port_value) \
1780                 src_port (src_port_value) protocol (protocol_value) \
1781                 mask (mask_value) tcp_flags (tcp_flags_value) \
1782                 priority (prio_value) queue (queue_id)
1783
1784The available information parameters are:
1785
1786* ``port_id``: The port which the 5-tuple filter assigned on.
1787
1788* ``dst_address``: Destination IP address.
1789
1790* ``src_address``: Source IP address.
1791
1792* ``dst_port_value``: TCP/UDP destination port.
1793
1794* ``src_port_value``: TCP/UDP source port.
1795
1796* ``protocol_value``: L4 protocol.
1797
1798* ``mask_value``: Participates in the match or not by bit for field above, 1b means participate
1799
1800* ``tcp_flags_value``: TCP control bits. The non-zero value is invalid, when the protocol_value is not set to 0x06 (TCP).
1801
1802* ``prio_value``: The priority of this filter.
1803
1804* ``queue_id``: The receive queue associated with this 5-tuple filter.
1805
1806Example, to add/remove an 5tuple filter rule::
1807
1808   testpmd> 5tuple_filter 0 add dst_ip 2.2.2.5 src_ip 2.2.2.4 \
1809            dst_port 64 src_port 32 protocol 0x06 mask 0x1F \
1810            flags 0x0 priority 3 queue 3
1811
1812   testpmd> 5tuple_filter 0 del dst_ip 2.2.2.5 src_ip 2.2.2.4 \
1813            dst_port 64 src_port 32 protocol 0x06 mask 0x1F \
1814            flags 0x0 priority 3 queue 3
1815
1816syn_filter
1817~~~~~~~~~~
1818
1819Using the  SYN filter, TCP packets whose *SYN* flag is set can be forwarded to a separate queue::
1820
1821   syn_filter (port_id) (add|del) priority (high|low) queue (queue_id)
1822
1823The available information parameters are:
1824
1825* ``port_id``: The port which the SYN filter assigned on.
1826
1827* ``high``: This SYN filter has higher priority than other filters.
1828
1829* ``low``: This SYN filter has lower priority than other filters.
1830
1831* ``queue_id``: The receive queue associated with this SYN filter
1832
1833Example::
1834
1835   testpmd> syn_filter 0 add priority high queue 3
1836
1837flex_filter
1838~~~~~~~~~~~
1839
1840With flex filter, packets can be recognized by any arbitrary pattern within the first 128 bytes of the packet
1841and routed into one of the receive queues::
1842
1843   flex_filter (port_id) (add|del) len (len_value) bytes (bytes_value) \
1844               mask (mask_value) priority (prio_value) queue (queue_id)
1845
1846The available information parameters are:
1847
1848* ``port_id``: The port which the Flex filter is assigned on.
1849
1850* ``len_value``: Filter length in bytes, no greater than 128.
1851
1852* ``bytes_value``: A string in hexadecimal, means the value the flex filter needs to match.
1853
1854* ``mask_value``: A string in hexadecimal, bit 1 means corresponding byte participates in the match.
1855
1856* ``prio_value``: The priority of this filter.
1857
1858* ``queue_id``: The receive queue associated with this Flex filter.
1859
1860Example::
1861
1862   testpmd> flex_filter 0 add len 16 bytes 0x00000000000000000000000008060000 \
1863                          mask 000C priority 3 queue 3
1864
1865   testpmd> flex_filter 0 del len 16 bytes 0x00000000000000000000000008060000 \
1866                          mask 000C priority 3 queue 3
1867
1868
1869.. _testpmd_flow_director:
1870
1871flow_director_filter
1872~~~~~~~~~~~~~~~~~~~~
1873
1874The Flow Director works in receive mode to identify specific flows or sets of flows and route them to specific queues.
1875
1876Four types of filtering are supported which are referred to as Perfect Match, Signature, Perfect-mac-vlan and
1877Perfect-tunnel filters, the match mode is set by the ``--pkt-filter-mode`` command-line parameter:
1878
1879* Perfect match filters.
1880  The hardware checks a match between the masked fields of the received packets and the programmed filters.
1881  The masked fields are for IP flow.
1882
1883* Signature filters.
1884  The hardware checks a match between a hash-based signature of the masked fields of the received packet.
1885
1886* Perfect-mac-vlan match filters.
1887  The hardware checks a match between the masked fields of the received packets and the programmed filters.
1888  The masked fields are for MAC VLAN flow.
1889
1890* Perfect-tunnel match filters.
1891  The hardware checks a match between the masked fields of the received packets and the programmed filters.
1892  The masked fields are for tunnel flow.
1893
1894The Flow Director filters can match the different fields for different type of packet: flow type, specific input set
1895per flow type and the flexible payload.
1896
1897The Flow Director can also mask out parts of all of these fields so that filters
1898are only applied to certain fields or parts of the fields.
1899
1900Different NICs may have different capabilities, command show port fdir (port_id) can be used to acquire the information.
1901
1902# Commands to add flow director filters of different flow types::
1903
1904   flow_director_filter (port_id) mode IP (add|del|update) \
1905                        flow (ipv4-other|ipv4-frag|ipv6-other|ipv6-frag) \
1906                        src (src_ip_address) dst (dst_ip_address) \
1907                        tos (tos_value) proto (proto_value) ttl (ttl_value) \
1908                        vlan (vlan_value) flexbytes (flexbytes_value) \
1909                        (drop|fwd) pf|vf(vf_id) queue (queue_id) \
1910                        fd_id (fd_id_value)
1911
1912   flow_director_filter (port_id) mode IP (add|del|update) \
1913                        flow (ipv4-tcp|ipv4-udp|ipv6-tcp|ipv6-udp) \
1914                        src (src_ip_address) (src_port) \
1915                        dst (dst_ip_address) (dst_port) \
1916                        tos (tos_value) ttl (ttl_value) \
1917                        vlan (vlan_value) flexbytes (flexbytes_value) \
1918                        (drop|fwd) queue pf|vf(vf_id) (queue_id) \
1919                        fd_id (fd_id_value)
1920
1921   flow_director_filter (port_id) mode IP (add|del|update) \
1922                        flow (ipv4-sctp|ipv6-sctp) \
1923                        src (src_ip_address) (src_port) \
1924                        dst (dst_ip_address) (dst_port) \
1925                        tos (tos_value) ttl (ttl_value) \
1926                        tag (verification_tag) vlan (vlan_value) \
1927                        flexbytes (flexbytes_value) (drop|fwd) \
1928                        pf|vf(vf_id) queue (queue_id) fd_id (fd_id_value)
1929
1930   flow_director_filter (port_id) mode IP (add|del|update) flow l2_payload \
1931                        ether (ethertype) flexbytes (flexbytes_value) \
1932                        (drop|fwd) pf|vf(vf_id) queue (queue_id)
1933                        fd_id (fd_id_value)
1934
1935   flow_director_filter (port_id) mode MAC-VLAN (add|del|update) \
1936                        mac (mac_address) vlan (vlan_value) \
1937                        flexbytes (flexbytes_value) (drop|fwd) \
1938                        queue (queue_id) fd_id (fd_id_value)
1939
1940   flow_director_filter (port_id) mode Tunnel (add|del|update) \
1941                        mac (mac_address) vlan (vlan_value) \
1942                        tunnel (NVGRE|VxLAN) tunnel-id (tunnel_id_value) \
1943                        flexbytes (flexbytes_value) (drop|fwd) \
1944                        queue (queue_id) fd_id (fd_id_value)
1945
1946For example, to add an ipv4-udp flow type filter::
1947
1948   testpmd> flow_director_filter 0 mode IP add flow ipv4-udp src 2.2.2.3 32 \
1949            dst 2.2.2.5 33 tos 2 ttl 40 vlan 0x1 flexbytes (0x88,0x48) \
1950            fwd pf queue 1 fd_id 1
1951
1952For example, add an ipv4-other flow type filter::
1953
1954   testpmd> flow_director_filter 0 mode IP add flow ipv4-other src 2.2.2.3 \
1955             dst 2.2.2.5 tos 2 proto 20 ttl 40 vlan 0x1 \
1956             flexbytes (0x88,0x48) fwd pf queue 1 fd_id 1
1957
1958flush_flow_director
1959~~~~~~~~~~~~~~~~~~~
1960
1961Flush all flow director filters on a device::
1962
1963   testpmd> flush_flow_director (port_id)
1964
1965Example, to flush all flow director filter on port 0::
1966
1967   testpmd> flush_flow_director 0
1968
1969flow_director_mask
1970~~~~~~~~~~~~~~~~~~
1971
1972Set flow director's input masks::
1973
1974   flow_director_mask (port_id) mode IP vlan (vlan_value) \
1975                      src_mask (ipv4_src) (ipv6_src) (src_port) \
1976                      dst_mask (ipv4_dst) (ipv6_dst) (dst_port)
1977
1978   flow_director_mask (port_id) mode MAC-VLAN vlan (vlan_value)
1979
1980   flow_director_mask (port_id) mode Tunnel vlan (vlan_value) \
1981                      mac (mac_value) tunnel-type (tunnel_type_value) \
1982                      tunnel-id (tunnel_id_value)
1983
1984Example, to set flow director mask on port 0::
1985
1986   testpmd> flow_director_mask 0 mode IP vlan 0xefff \
1987            src_mask 255.255.255.255 \
1988                FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF \
1989            dst_mask 255.255.255.255 \
1990                FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF
1991
1992flow_director_flex_mask
1993~~~~~~~~~~~~~~~~~~~~~~~
1994
1995set masks of flow director's flexible payload based on certain flow type::
1996
1997   testpmd> flow_director_flex_mask (port_id) \
1998            flow (none|ipv4-other|ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
1999                  ipv6-other|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp| \
2000                  l2_payload|all) (mask)
2001
2002Example, to set flow director's flex mask for all flow type on port 0::
2003
2004   testpmd> flow_director_flex_mask 0 flow all \
2005            (0xff,0xff,0,0,0,0,0,0,0,0,0,0,0,0,0,0)
2006
2007
2008flow_director_flex_payload
2009~~~~~~~~~~~~~~~~~~~~~~~~~~
2010
2011Configure flexible payload selection::
2012
2013   flow_director_flex_payload (port_id) (raw|l2|l3|l4) (config)
2014
2015For example, to select the first 16 bytes from the offset 4 (bytes) of packet's payload as flexible payload::
2016
2017   testpmd> flow_director_flex_payload 0 l4 \
2018            (4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19)
2019
2020get_sym_hash_ena_per_port
2021~~~~~~~~~~~~~~~~~~~~~~~~~
2022
2023Get symmetric hash enable configuration per port::
2024
2025   get_sym_hash_ena_per_port (port_id)
2026
2027For example, to get symmetric hash enable configuration of port 1::
2028
2029   testpmd> get_sym_hash_ena_per_port 1
2030
2031set_sym_hash_ena_per_port
2032~~~~~~~~~~~~~~~~~~~~~~~~~
2033
2034Set symmetric hash enable configuration per port to enable or disable::
2035
2036   set_sym_hash_ena_per_port (port_id) (enable|disable)
2037
2038For example, to set symmetric hash enable configuration of port 1 to enable::
2039
2040   testpmd> set_sym_hash_ena_per_port 1 enable
2041
2042get_hash_global_config
2043~~~~~~~~~~~~~~~~~~~~~~
2044
2045Get the global configurations of hash filters::
2046
2047   get_hash_global_config (port_id)
2048
2049For example, to get the global configurations of hash filters of port 1::
2050
2051   testpmd> get_hash_global_config 1
2052
2053set_hash_global_config
2054~~~~~~~~~~~~~~~~~~~~~~
2055
2056Set the global configurations of hash filters::
2057
2058   set_hash_global_config (port_id) (toeplitz|simple_xor|default) \
2059   (ipv4|ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp|ipv4-other|ipv6|ipv6-frag| \
2060   ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other|l2_payload) \
2061   (enable|disable)
2062
2063For example, to enable simple_xor for flow type of ipv6 on port 2::
2064
2065   testpmd> set_hash_global_config 2 simple_xor ipv6 enable
2066
2067set_hash_input_set
2068~~~~~~~~~~~~~~~~~~
2069
2070Set the input set for hash::
2071
2072   set_hash_input_set (port_id) (ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
2073   ipv4-other|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other| \
2074   l2_payload) (ovlan|ivlan|src-ipv4|dst-ipv4|src-ipv6|dst-ipv6|ipv4-tos| \
2075   ipv4-proto|ipv6-tc|ipv6-next-header|udp-src-port|udp-dst-port| \
2076   tcp-src-port|tcp-dst-port|sctp-src-port|sctp-dst-port|sctp-veri-tag| \
2077   udp-key|gre-key|fld-1st|fld-2nd|fld-3rd|fld-4th|fld-5th|fld-6th|fld-7th| \
2078   fld-8th|none) (select|add)
2079
2080For example, to add source IP to hash input set for flow type of ipv4-udp on port 0::
2081
2082   testpmd> set_hash_input_set 0 ipv4-udp src-ipv4 add
2083
2084set_fdir_input_set
2085~~~~~~~~~~~~~~~~~~
2086
2087The Flow Director filters can match the different fields for different type of packet, i.e. specific input set
2088on per flow type and the flexible payload. This command can be used to change input set for each flow type.
2089
2090Set the input set for flow director::
2091
2092   set_fdir_input_set (port_id) (ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
2093   ipv4-other|ipv6|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other| \
2094   l2_payload) (ivlan|ethertype|src-ipv4|dst-ipv4|src-ipv6|dst-ipv6|ipv4-tos| \
2095   ipv4-proto|ipv4-ttl|ipv6-tc|ipv6-next-header|ipv6-hop-limits| \
2096   tudp-src-port|udp-dst-port|cp-src-port|tcp-dst-port|sctp-src-port| \
2097   sctp-dst-port|sctp-veri-tag|none) (select|add)
2098
2099For example to add source IP to FD input set for flow type of ipv4-udp on port 0::
2100
2101   testpmd> set_fdir_input_set 0 ipv4-udp src-ipv4 add
2102
2103global_config
2104~~~~~~~~~~~~~
2105
2106Set different GRE key length for input set::
2107
2108   global_config (port_id) gre-key-len (number in bytes)
2109
2110For example to set GRE key length for input set to 4 bytes on port 0::
2111
2112   testpmd> global_config 0 gre-key-len 4
2113
2114
2115.. _testpmd_rte_flow:
2116
2117Flow rules management
2118---------------------
2119
2120Control of the generic flow API (*rte_flow*) is fully exposed through the
2121``flow`` command (validation, creation, destruction and queries).
2122
2123Considering *rte_flow* overlaps with all `Filter Functions`_, using both
2124features simultaneously may cause undefined side-effects and is therefore
2125not recommended.
2126
2127``flow`` syntax
2128~~~~~~~~~~~~~~~
2129
2130Because the ``flow`` command uses dynamic tokens to handle the large number
2131of possible flow rules combinations, its behavior differs slightly from
2132other commands, in particular:
2133
2134- Pressing *?* or the *<tab>* key displays contextual help for the current
2135  token, not that of the entire command.
2136
2137- Optional and repeated parameters are supported (provided they are listed
2138  in the contextual help).
2139
2140The first parameter stands for the operation mode. Possible operations and
2141their general syntax are described below. They are covered in detail in the
2142following sections.
2143
2144- Check whether a flow rule can be created::
2145
2146   flow validate {port_id}
2147       [group {group_id}] [priority {level}] [ingress] [egress]
2148       pattern {item} [/ {item} [...]] / end
2149       actions {action} [/ {action} [...]] / end
2150
2151- Create a flow rule::
2152
2153   flow create {port_id}
2154       [group {group_id}] [priority {level}] [ingress] [egress]
2155       pattern {item} [/ {item} [...]] / end
2156       actions {action} [/ {action} [...]] / end
2157
2158- Destroy specific flow rules::
2159
2160   flow destroy {port_id} rule {rule_id} [...]
2161
2162- Destroy all flow rules::
2163
2164   flow flush {port_id}
2165
2166- Query an existing flow rule::
2167
2168   flow query {port_id} {rule_id} {action}
2169
2170- List existing flow rules sorted by priority, filtered by group
2171  identifiers::
2172
2173   flow list {port_id} [group {group_id}] [...]
2174
2175Validating flow rules
2176~~~~~~~~~~~~~~~~~~~~~
2177
2178``flow validate`` reports whether a flow rule would be accepted by the
2179underlying device in its current state but stops short of creating it. It is
2180bound to ``rte_flow_validate()``::
2181
2182   flow validate {port_id}
2183      [group {group_id}] [priority {level}] [ingress] [egress]
2184      pattern {item} [/ {item} [...]] / end
2185      actions {action} [/ {action} [...]] / end
2186
2187If successful, it will show::
2188
2189   Flow rule validated
2190
2191Otherwise it will show an error message of the form::
2192
2193   Caught error type [...] ([...]): [...]
2194
2195This command uses the same parameters as ``flow create``, their format is
2196described in `Creating flow rules`_.
2197
2198Check whether redirecting any Ethernet packet received on port 0 to RX queue
2199index 6 is supported::
2200
2201   testpmd> flow validate 0 ingress pattern eth / end
2202      actions queue index 6 / end
2203   Flow rule validated
2204   testpmd>
2205
2206Port 0 does not support TCPv6 rules::
2207
2208   testpmd> flow validate 0 ingress pattern eth / ipv6 / tcp / end
2209      actions drop / end
2210   Caught error type 9 (specific pattern item): Invalid argument
2211   testpmd>
2212
2213Creating flow rules
2214~~~~~~~~~~~~~~~~~~~
2215
2216``flow create`` validates and creates the specified flow rule. It is bound
2217to ``rte_flow_create()``::
2218
2219   flow create {port_id}
2220      [group {group_id}] [priority {level}] [ingress] [egress]
2221      pattern {item} [/ {item} [...]] / end
2222      actions {action} [/ {action} [...]] / end
2223
2224If successful, it will return a flow rule ID usable with other commands::
2225
2226   Flow rule #[...] created
2227
2228Otherwise it will show an error message of the form::
2229
2230   Caught error type [...] ([...]): [...]
2231
2232Parameters describe in the following order:
2233
2234- Attributes (*group*, *priority*, *ingress*, *egress* tokens).
2235- A matching pattern, starting with the *pattern* token and terminated by an
2236  *end* pattern item.
2237- Actions, starting with the *actions* token and terminated by an *end*
2238  action.
2239
2240These translate directly to *rte_flow* objects provided as-is to the
2241underlying functions.
2242
2243The shortest valid definition only comprises mandatory tokens::
2244
2245   testpmd> flow create 0 pattern end actions end
2246
2247Note that PMDs may refuse rules that essentially do nothing such as this
2248one.
2249
2250**All unspecified object values are automatically initialized to 0.**
2251
2252Attributes
2253^^^^^^^^^^
2254
2255These tokens affect flow rule attributes (``struct rte_flow_attr``) and are
2256specified before the ``pattern`` token.
2257
2258- ``group {group id}``: priority group.
2259- ``priority {level}``: priority level within group.
2260- ``ingress``: rule applies to ingress traffic.
2261- ``egress``: rule applies to egress traffic.
2262
2263Each instance of an attribute specified several times overrides the previous
2264value as shown below (group 4 is used)::
2265
2266   testpmd> flow create 0 group 42 group 24 group 4 [...]
2267
2268Note that once enabled, ``ingress`` and ``egress`` cannot be disabled.
2269
2270While not specifying a direction is an error, some rules may allow both
2271simultaneously.
2272
2273Most rules affect RX therefore contain the ``ingress`` token::
2274
2275   testpmd> flow create 0 ingress pattern [...]
2276
2277Matching pattern
2278^^^^^^^^^^^^^^^^
2279
2280A matching pattern starts after the ``pattern`` token. It is made of pattern
2281items and is terminated by a mandatory ``end`` item.
2282
2283Items are named after their type (*RTE_FLOW_ITEM_TYPE_* from ``enum
2284rte_flow_item_type``).
2285
2286The ``/`` token is used as a separator between pattern items as shown
2287below::
2288
2289   testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end [...]
2290
2291Note that protocol items like these must be stacked from lowest to highest
2292layer to make sense. For instance, the following rule is either invalid or
2293unlikely to match any packet::
2294
2295   testpmd> flow create 0 ingress pattern eth / udp / ipv4 / end [...]
2296
2297More information on these restrictions can be found in the *rte_flow*
2298documentation.
2299
2300Several items support additional specification structures, for example
2301``ipv4`` allows specifying source and destination addresses as follows::
2302
2303   testpmd> flow create 0 ingress pattern eth / ipv4 src is 10.1.1.1
2304      dst is 10.2.0.0 / end [...]
2305
2306This rule matches all IPv4 traffic with the specified properties.
2307
2308In this example, ``src`` and ``dst`` are field names of the underlying
2309``struct rte_flow_item_ipv4`` object. All item properties can be specified
2310in a similar fashion.
2311
2312The ``is`` token means that the subsequent value must be matched exactly,
2313and assigns ``spec`` and ``mask`` fields in ``struct rte_flow_item``
2314accordingly. Possible assignment tokens are:
2315
2316- ``is``: match value perfectly (with full bit-mask).
2317- ``spec``: match value according to configured bit-mask.
2318- ``last``: specify upper bound to establish a range.
2319- ``mask``: specify bit-mask with relevant bits set to one.
2320- ``prefix``: generate bit-mask from a prefix length.
2321
2322These yield identical results::
2323
2324   ipv4 src is 10.1.1.1
2325
2326::
2327
2328   ipv4 src spec 10.1.1.1 src mask 255.255.255.255
2329
2330::
2331
2332   ipv4 src spec 10.1.1.1 src prefix 32
2333
2334::
2335
2336   ipv4 src is 10.1.1.1 src last 10.1.1.1 # range with a single value
2337
2338::
2339
2340   ipv4 src is 10.1.1.1 src last 0 # 0 disables range
2341
2342Inclusive ranges can be defined with ``last``::
2343
2344   ipv4 src is 10.1.1.1 src last 10.2.3.4 # 10.1.1.1 to 10.2.3.4
2345
2346Note that ``mask`` affects both ``spec`` and ``last``::
2347
2348   ipv4 src is 10.1.1.1 src last 10.2.3.4 src mask 255.255.0.0
2349      # matches 10.1.0.0 to 10.2.255.255
2350
2351Properties can be modified multiple times::
2352
2353   ipv4 src is 10.1.1.1 src is 10.1.2.3 src is 10.2.3.4 # matches 10.2.3.4
2354
2355::
2356
2357   ipv4 src is 10.1.1.1 src prefix 24 src prefix 16 # matches 10.1.0.0/16
2358
2359Pattern items
2360^^^^^^^^^^^^^
2361
2362This section lists supported pattern items and their attributes, if any.
2363
2364- ``end``: end list of pattern items.
2365
2366- ``void``: no-op pattern item.
2367
2368- ``invert``: perform actions when pattern does not match.
2369
2370- ``any``: match any protocol for the current layer.
2371
2372  - ``num {unsigned}``: number of layers covered.
2373
2374- ``pf``: match packets addressed to the physical function.
2375
2376- ``vf``: match packets addressed to a virtual function ID.
2377
2378  - ``id {unsigned}``: destination VF ID.
2379
2380- ``port``: device-specific physical port index to use.
2381
2382  - ``index {unsigned}``: physical port index.
2383
2384- ``raw``: match an arbitrary byte string.
2385
2386  - ``relative {boolean}``: look for pattern after the previous item.
2387  - ``search {boolean}``: search pattern from offset (see also limit).
2388  - ``offset {integer}``: absolute or relative offset for pattern.
2389  - ``limit {unsigned}``: search area limit for start of pattern.
2390  - ``pattern {string}``: byte string to look for.
2391
2392- ``eth``: match Ethernet header.
2393
2394  - ``dst {MAC-48}``: destination MAC.
2395  - ``src {MAC-48}``: source MAC.
2396  - ``type {unsigned}``: EtherType.
2397
2398- ``vlan``: match 802.1Q/ad VLAN tag.
2399
2400  - ``tpid {unsigned}``: tag protocol identifier.
2401  - ``tci {unsigned}``: tag control information.
2402  - ``pcp {unsigned}``: priority code point.
2403  - ``dei {unsigned}``: drop eligible indicator.
2404  - ``vid {unsigned}``: VLAN identifier.
2405
2406- ``ipv4``: match IPv4 header.
2407
2408  - ``tos {unsigned}``: type of service.
2409  - ``ttl {unsigned}``: time to live.
2410  - ``proto {unsigned}``: next protocol ID.
2411  - ``src {ipv4 address}``: source address.
2412  - ``dst {ipv4 address}``: destination address.
2413
2414- ``ipv6``: match IPv6 header.
2415
2416  - ``tc {unsigned}``: traffic class.
2417  - ``flow {unsigned}``: flow label.
2418  - ``proto {unsigned}``: protocol (next header).
2419  - ``hop {unsigned}``: hop limit.
2420  - ``src {ipv6 address}``: source address.
2421  - ``dst {ipv6 address}``: destination address.
2422
2423- ``icmp``: match ICMP header.
2424
2425  - ``type {unsigned}``: ICMP packet type.
2426  - ``code {unsigned}``: ICMP packet code.
2427
2428- ``udp``: match UDP header.
2429
2430  - ``src {unsigned}``: UDP source port.
2431  - ``dst {unsigned}``: UDP destination port.
2432
2433- ``tcp``: match TCP header.
2434
2435  - ``src {unsigned}``: TCP source port.
2436  - ``dst {unsigned}``: TCP destination port.
2437
2438- ``sctp``: match SCTP header.
2439
2440  - ``src {unsigned}``: SCTP source port.
2441  - ``dst {unsigned}``: SCTP destination port.
2442  - ``tag {unsigned}``: validation tag.
2443  - ``cksum {unsigned}``: checksum.
2444
2445- ``vxlan``: match VXLAN header.
2446
2447  - ``vni {unsigned}``: VXLAN identifier.
2448
2449Actions list
2450^^^^^^^^^^^^
2451
2452A list of actions starts after the ``actions`` token in the same fashion as
2453`Matching pattern`_; actions are separated by ``/`` tokens and the list is
2454terminated by a mandatory ``end`` action.
2455
2456Actions are named after their type (*RTE_FLOW_ACTION_TYPE_* from ``enum
2457rte_flow_action_type``).
2458
2459Dropping all incoming UDPv4 packets can be expressed as follows::
2460
2461   testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
2462      actions drop / end
2463
2464Several actions have configurable properties which must be specified when
2465there is no valid default value. For example, ``queue`` requires a target
2466queue index.
2467
2468This rule redirects incoming UDPv4 traffic to queue index 6::
2469
2470   testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
2471      actions queue index 6 / end
2472
2473While this one could be rejected by PMDs (unspecified queue index)::
2474
2475   testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
2476      actions queue / end
2477
2478As defined by *rte_flow*, the list is not ordered, all actions of a given
2479rule are performed simultaneously. These are equivalent::
2480
2481   queue index 6 / void / mark id 42 / end
2482
2483::
2484
2485   void / mark id 42 / queue index 6 / end
2486
2487All actions in a list should have different types, otherwise only the last
2488action of a given type is taken into account::
2489
2490   queue index 4 / queue index 5 / queue index 6 / end # will use queue 6
2491
2492::
2493
2494   drop / drop / drop / end # drop is performed only once
2495
2496::
2497
2498   mark id 42 / queue index 3 / mark id 24 / end # mark will be 24
2499
2500Considering they are performed simultaneously, opposite and overlapping
2501actions can sometimes be combined when the end result is unambiguous::
2502
2503   drop / queue index 6 / end # drop has no effect
2504
2505::
2506
2507   drop / dup index 6 / end # same as above
2508
2509::
2510
2511   queue index 6 / rss queues 6 7 8 / end # queue has no effect
2512
2513::
2514
2515   drop / passthru / end # drop has no effect
2516
2517Note that PMDs may still refuse such combinations.
2518
2519Actions
2520^^^^^^^
2521
2522This section lists supported actions and their attributes, if any.
2523
2524- ``end``: end list of actions.
2525
2526- ``void``: no-op action.
2527
2528- ``passthru``: let subsequent rule process matched packets.
2529
2530- ``mark``: attach 32 bit value to packets.
2531
2532  - ``id {unsigned}``: 32 bit value to return with packets.
2533
2534- ``flag``: flag packets.
2535
2536- ``queue``: assign packets to a given queue index.
2537
2538  - ``index {unsigned}``: queue index to use.
2539
2540- ``drop``: drop packets (note: passthru has priority).
2541
2542- ``count``: enable counters for this rule.
2543
2544- ``dup``: duplicate packets to a given queue index.
2545
2546  - ``index {unsigned}``: queue index to duplicate packets to.
2547
2548- ``rss``: spread packets among several queues.
2549
2550  - ``queues [{unsigned} [...]] end``: queue indices to use.
2551
2552- ``pf``: redirect packets to physical device function.
2553
2554- ``vf``: redirect packets to virtual device function.
2555
2556  - ``original {boolean}``: use original VF ID if possible.
2557  - ``id {unsigned}``: VF ID to redirect packets to.
2558
2559Destroying flow rules
2560~~~~~~~~~~~~~~~~~~~~~
2561
2562``flow destroy`` destroys one or more rules from their rule ID (as returned
2563by ``flow create``), this command calls ``rte_flow_destroy()`` as many
2564times as necessary::
2565
2566   flow destroy {port_id} rule {rule_id} [...]
2567
2568If successful, it will show::
2569
2570   Flow rule #[...] destroyed
2571
2572It does not report anything for rule IDs that do not exist. The usual error
2573message is shown when a rule cannot be destroyed::
2574
2575   Caught error type [...] ([...]): [...]
2576
2577``flow flush`` destroys all rules on a device and does not take extra
2578arguments. It is bound to ``rte_flow_flush()``::
2579
2580   flow flush {port_id}
2581
2582Any errors are reported as above.
2583
2584Creating several rules and destroying them::
2585
2586   testpmd> flow create 0 ingress pattern eth / ipv6 / end
2587      actions queue index 2 / end
2588   Flow rule #0 created
2589   testpmd> flow create 0 ingress pattern eth / ipv4 / end
2590      actions queue index 3 / end
2591   Flow rule #1 created
2592   testpmd> flow destroy 0 rule 0 rule 1
2593   Flow rule #1 destroyed
2594   Flow rule #0 destroyed
2595   testpmd>
2596
2597The same result can be achieved using ``flow flush``::
2598
2599   testpmd> flow create 0 ingress pattern eth / ipv6 / end
2600      actions queue index 2 / end
2601   Flow rule #0 created
2602   testpmd> flow create 0 ingress pattern eth / ipv4 / end
2603      actions queue index 3 / end
2604   Flow rule #1 created
2605   testpmd> flow flush 0
2606   testpmd>
2607
2608Non-existent rule IDs are ignored::
2609
2610   testpmd> flow create 0 ingress pattern eth / ipv6 / end
2611      actions queue index 2 / end
2612   Flow rule #0 created
2613   testpmd> flow create 0 ingress pattern eth / ipv4 / end
2614      actions queue index 3 / end
2615   Flow rule #1 created
2616   testpmd> flow destroy 0 rule 42 rule 10 rule 2
2617   testpmd>
2618   testpmd> flow destroy 0 rule 0
2619   Flow rule #0 destroyed
2620   testpmd>
2621
2622Querying flow rules
2623~~~~~~~~~~~~~~~~~~~
2624
2625``flow query`` queries a specific action of a flow rule having that
2626ability. Such actions collect information that can be reported using this
2627command. It is bound to ``rte_flow_query()``::
2628
2629   flow query {port_id} {rule_id} {action}
2630
2631If successful, it will display either the retrieved data for known actions
2632or the following message::
2633
2634   Cannot display result for action type [...] ([...])
2635
2636Otherwise, it will complain either that the rule does not exist or that some
2637error occurred::
2638
2639   Flow rule #[...] not found
2640
2641::
2642
2643   Caught error type [...] ([...]): [...]
2644
2645Currently only the ``count`` action is supported. This action reports the
2646number of packets that hit the flow rule and the total number of bytes. Its
2647output has the following format::
2648
2649   count:
2650    hits_set: [...] # whether "hits" contains a valid value
2651    bytes_set: [...] # whether "bytes" contains a valid value
2652    hits: [...] # number of packets
2653    bytes: [...] # number of bytes
2654
2655Querying counters for TCPv6 packets redirected to queue 6::
2656
2657   testpmd> flow create 0 ingress pattern eth / ipv6 / tcp / end
2658      actions queue index 6 / count / end
2659   Flow rule #4 created
2660   testpmd> flow query 0 4 count
2661   count:
2662    hits_set: 1
2663    bytes_set: 0
2664    hits: 386446
2665    bytes: 0
2666   testpmd>
2667
2668Listing flow rules
2669~~~~~~~~~~~~~~~~~~
2670
2671``flow list`` lists existing flow rules sorted by priority and optionally
2672filtered by group identifiers::
2673
2674   flow list {port_id} [group {group_id}] [...]
2675
2676This command only fails with the following message if the device does not
2677exist::
2678
2679   Invalid port [...]
2680
2681Output consists of a header line followed by a short description of each
2682flow rule, one per line. There is no output at all when no flow rules are
2683configured on the device::
2684
2685   ID      Group   Prio    Attr    Rule
2686   [...]   [...]   [...]   [...]   [...]
2687
2688``Attr`` column flags:
2689
2690- ``i`` for ``ingress``.
2691- ``e`` for ``egress``.
2692
2693Creating several flow rules and listing them::
2694
2695   testpmd> flow create 0 ingress pattern eth / ipv4 / end
2696      actions queue index 6 / end
2697   Flow rule #0 created
2698   testpmd> flow create 0 ingress pattern eth / ipv6 / end
2699      actions queue index 2 / end
2700   Flow rule #1 created
2701   testpmd> flow create 0 priority 5 ingress pattern eth / ipv4 / udp / end
2702      actions rss queues 6 7 8 end / end
2703   Flow rule #2 created
2704   testpmd> flow list 0
2705   ID      Group   Prio    Attr    Rule
2706   0       0       0       i-      ETH IPV4 => QUEUE
2707   1       0       0       i-      ETH IPV6 => QUEUE
2708   2       0       5       i-      ETH IPV4 UDP => RSS
2709   testpmd>
2710
2711Rules are sorted by priority (i.e. group ID first, then priority level)::
2712
2713   testpmd> flow list 1
2714   ID      Group   Prio    Attr    Rule
2715   0       0       0       i-      ETH => COUNT
2716   6       0       500     i-      ETH IPV6 TCP => DROP COUNT
2717   5       0       1000    i-      ETH IPV6 ICMP => QUEUE
2718   1       24      0       i-      ETH IPV4 UDP => QUEUE
2719   4       24      10      i-      ETH IPV4 TCP => DROP
2720   3       24      20      i-      ETH IPV4 => DROP
2721   2       24      42      i-      ETH IPV4 UDP => QUEUE
2722   7       63      0       i-      ETH IPV6 UDP VXLAN => MARK QUEUE
2723   testpmd>
2724
2725Output can be limited to specific groups::
2726
2727   testpmd> flow list 1 group 0 group 63
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   7       63      0       i-      ETH IPV6 UDP VXLAN => MARK QUEUE
2733   testpmd>
2734