xref: /dpdk/doc/guides/testpmd_app_ug/testpmd_funcs.rst (revision 42d0baaea576542fadff2008489a6fea16b8afeb)
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 tc tx min bandwidth (for VF)
904~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
905
906Set all TCs' TX min relative bandwidth (%) for a VF from PF::
907
908   testpmd> set vf tc tx min-bandwidth (port_id) (vf_id) (bw1, bw2, ...)
909
910set flow_ctrl rx
911~~~~~~~~~~~~~~~~
912
913Set the link flow control parameter on a port::
914
915   testpmd> set flow_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \
916            (pause_time) (send_xon) mac_ctrl_frame_fwd (on|off) \
917	    autoneg (on|off) (port_id)
918
919Where:
920
921* ``high_water`` (integer): High threshold value to trigger XOFF.
922
923* ``low_water`` (integer): Low threshold value to trigger XON.
924
925* ``pause_time`` (integer): Pause quota in the Pause frame.
926
927* ``send_xon`` (0/1): Send XON frame.
928
929* ``mac_ctrl_frame_fwd``: Enable receiving MAC control frames.
930
931* ``autoneg``: Change the auto-negotiation parameter.
932
933set pfc_ctrl rx
934~~~~~~~~~~~~~~~
935
936Set the priority flow control parameter on a port::
937
938   testpmd> set pfc_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \
939            (pause_time) (priority) (port_id)
940
941Where:
942
943* ``high_water`` (integer): High threshold value.
944
945* ``low_water`` (integer): Low threshold value.
946
947* ``pause_time`` (integer): Pause quota in the Pause frame.
948
949* ``priority`` (0-7): VLAN User Priority.
950
951set stat_qmap
952~~~~~~~~~~~~~
953
954Set statistics mapping (qmapping 0..15) for RX/TX queue on port::
955
956   testpmd> set stat_qmap (tx|rx) (port_id) (queue_id) (qmapping)
957
958For example, to set rx queue 2 on port 0 to mapping 5::
959
960   testpmd>set stat_qmap rx 0 2 5
961
962set port - rx/tx (for VF)
963~~~~~~~~~~~~~~~~~~~~~~~~~
964
965Set VF receive/transmit from a port::
966
967   testpmd> set port (port_id) vf (vf_id) (rx|tx) (on|off)
968
969set port - mac address filter (for VF)
970~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
971
972Add/Remove unicast or multicast MAC addr filter for a VF::
973
974   testpmd> set port (port_id) vf (vf_id) (mac_addr) \
975            (exact-mac|exact-mac-vlan|hashmac|hashmac-vlan) (on|off)
976
977set port - rx mode(for VF)
978~~~~~~~~~~~~~~~~~~~~~~~~~~
979
980Set the VF receive mode of a port::
981
982   testpmd> set port (port_id) vf (vf_id) \
983            rxmode (AUPE|ROPE|BAM|MPE) (on|off)
984
985The available receive modes are:
986
987* ``AUPE``: Accepts untagged VLAN.
988
989* ``ROPE``: Accepts unicast hash.
990
991* ``BAM``: Accepts broadcast packets.
992
993* ``MPE``: Accepts all multicast packets.
994
995set port - tx_rate (for Queue)
996~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
997
998Set TX rate limitation for a queue on a port::
999
1000   testpmd> set port (port_id) queue (queue_id) rate (rate_value)
1001
1002set port - tx_rate (for VF)
1003~~~~~~~~~~~~~~~~~~~~~~~~~~~
1004
1005Set TX rate limitation for queues in VF on a port::
1006
1007   testpmd> set port (port_id) vf (vf_id) rate (rate_value) queue_mask (queue_mask)
1008
1009set port - mirror rule
1010~~~~~~~~~~~~~~~~~~~~~~
1011
1012Set pool or vlan type mirror rule for a port::
1013
1014   testpmd> set port (port_id) mirror-rule (rule_id) \
1015            (pool-mirror-up|pool-mirror-down|vlan-mirror) \
1016            (poolmask|vlanid[,vlanid]*) dst-pool (pool_id) (on|off)
1017
1018Set link mirror rule for a port::
1019
1020   testpmd> set port (port_id) mirror-rule (rule_id) \
1021           (uplink-mirror|downlink-mirror) dst-pool (pool_id) (on|off)
1022
1023For example to enable mirror traffic with vlan 0,1 to pool 0::
1024
1025   set port 0 mirror-rule 0 vlan-mirror 0,1 dst-pool 0 on
1026
1027reset port - mirror rule
1028~~~~~~~~~~~~~~~~~~~~~~~~
1029
1030Reset a mirror rule for a port::
1031
1032   testpmd> reset port (port_id) mirror-rule (rule_id)
1033
1034set flush_rx
1035~~~~~~~~~~~~
1036
1037Set the flush on RX streams before forwarding.
1038The default is flush ``on``.
1039Mainly used with PCAP drivers to turn off the default behavior of flushing the first 512 packets on RX streams::
1040
1041   testpmd> set flush_rx off
1042
1043set bypass mode
1044~~~~~~~~~~~~~~~
1045
1046Set the bypass mode for the lowest port on bypass enabled NIC::
1047
1048   testpmd> set bypass mode (normal|bypass|isolate) (port_id)
1049
1050set bypass event
1051~~~~~~~~~~~~~~~~
1052
1053Set the event required to initiate specified bypass mode for the lowest port on a bypass enabled::
1054
1055   testpmd> set bypass event (timeout|os_on|os_off|power_on|power_off) \
1056            mode (normal|bypass|isolate) (port_id)
1057
1058Where:
1059
1060* ``timeout``: Enable bypass after watchdog timeout.
1061
1062* ``os_on``: Enable bypass when OS/board is powered on.
1063
1064* ``os_off``: Enable bypass when OS/board is powered off.
1065
1066* ``power_on``: Enable bypass when power supply is turned on.
1067
1068* ``power_off``: Enable bypass when power supply is turned off.
1069
1070
1071set bypass timeout
1072~~~~~~~~~~~~~~~~~~
1073
1074Set the bypass watchdog timeout to ``n`` seconds where 0 = instant::
1075
1076   testpmd> set bypass timeout (0|1.5|2|3|4|8|16|32)
1077
1078show bypass config
1079~~~~~~~~~~~~~~~~~~
1080
1081Show the bypass configuration for a bypass enabled NIC using the lowest port on the NIC::
1082
1083   testpmd> show bypass config (port_id)
1084
1085set link up
1086~~~~~~~~~~~
1087
1088Set link up for a port::
1089
1090   testpmd> set link-up port (port id)
1091
1092set link down
1093~~~~~~~~~~~~~
1094
1095Set link down for a port::
1096
1097   testpmd> set link-down port (port id)
1098
1099E-tag set
1100~~~~~~~~~
1101
1102Enable E-tag insertion for a VF on a port::
1103
1104   testpmd> E-tag set insertion on port-tag-id (value) port (port_id) vf (vf_id)
1105
1106Disable E-tag insertion for a VF on a port::
1107
1108   testpmd> E-tag set insertion off port (port_id) vf (vf_id)
1109
1110Enable/disable E-tag stripping on a port::
1111
1112   testpmd> E-tag set stripping (on|off) port (port_id)
1113
1114Enable/disable E-tag based forwarding on a port::
1115
1116   testpmd> E-tag set forwarding (on|off) port (port_id)
1117
1118Add an E-tag forwarding filter on a port::
1119
1120   testpmd> E-tag set filter add e-tag-id (value) dst-pool (pool_id) port (port_id)
1121
1122Delete an E-tag forwarding filter on a port::
1123   testpmd> E-tag set filter del e-tag-id (value) port (port_id)
1124
1125
1126Port Functions
1127--------------
1128
1129The following sections show functions for configuring ports.
1130
1131.. note::
1132
1133   Port configuration changes only become active when forwarding is started/restarted.
1134
1135port attach
1136~~~~~~~~~~~
1137
1138Attach a port specified by pci address or virtual device args::
1139
1140   testpmd> port attach (identifier)
1141
1142To attach a new pci device, the device should be recognized by kernel first.
1143Then it should be moved under DPDK management.
1144Finally the port can be attached to testpmd.
1145
1146For example, to move a pci device using ixgbe under DPDK management:
1147
1148.. code-block:: console
1149
1150   # Check the status of the available devices.
1151   ./usertools/dpdk-devbind.py --status
1152
1153   Network devices using DPDK-compatible driver
1154   ============================================
1155   <none>
1156
1157   Network devices using kernel driver
1158   ===================================
1159   0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused=
1160
1161
1162   # Bind the device to igb_uio.
1163   sudo ./usertools/dpdk-devbind.py -b igb_uio 0000:0a:00.0
1164
1165
1166   # Recheck the status of the devices.
1167   ./usertools/dpdk-devbind.py --status
1168   Network devices using DPDK-compatible driver
1169   ============================================
1170   0000:0a:00.0 '82599ES 10-Gigabit' drv=igb_uio unused=
1171
1172To attach a port created by virtual device, above steps are not needed.
1173
1174For example, to attach a port whose pci address is 0000:0a:00.0.
1175
1176.. code-block:: console
1177
1178   testpmd> port attach 0000:0a:00.0
1179   Attaching a new port...
1180   EAL: PCI device 0000:0a:00.0 on NUMA socket -1
1181   EAL:   probe driver: 8086:10fb rte_ixgbe_pmd
1182   EAL:   PCI memory mapped at 0x7f83bfa00000
1183   EAL:   PCI memory mapped at 0x7f83bfa80000
1184   PMD: eth_ixgbe_dev_init(): MAC: 2, PHY: 18, SFP+: 5
1185   PMD: eth_ixgbe_dev_init(): port 0 vendorID=0x8086 deviceID=0x10fb
1186   Port 0 is attached. Now total ports is 1
1187   Done
1188
1189For example, to attach a port created by pcap PMD.
1190
1191.. code-block:: console
1192
1193   testpmd> port attach net_pcap0
1194   Attaching a new port...
1195   PMD: Initializing pmd_pcap for net_pcap0
1196   PMD: Creating pcap-backed ethdev on numa socket 0
1197   Port 0 is attached. Now total ports is 1
1198   Done
1199
1200In this case, identifier is ``net_pcap0``.
1201This identifier format is the same as ``--vdev`` format of DPDK applications.
1202
1203For example, to re-attach a bonded port which has been previously detached,
1204the mode and slave parameters must be given.
1205
1206.. code-block:: console
1207
1208   testpmd> port attach net_bond_0,mode=0,slave=1
1209   Attaching a new port...
1210   EAL: Initializing pmd_bond for net_bond_0
1211   EAL: Create bonded device net_bond_0 on port 0 in mode 0 on socket 0.
1212   Port 0 is attached. Now total ports is 1
1213   Done
1214
1215
1216port detach
1217~~~~~~~~~~~
1218
1219Detach a specific port::
1220
1221   testpmd> port detach (port_id)
1222
1223Before detaching a port, the port should be stopped and closed.
1224
1225For example, to detach a pci device port 0.
1226
1227.. code-block:: console
1228
1229   testpmd> port stop 0
1230   Stopping ports...
1231   Done
1232   testpmd> port close 0
1233   Closing ports...
1234   Done
1235
1236   testpmd> port detach 0
1237   Detaching a port...
1238   EAL: PCI device 0000:0a:00.0 on NUMA socket -1
1239   EAL:   remove driver: 8086:10fb rte_ixgbe_pmd
1240   EAL:   PCI memory unmapped at 0x7f83bfa00000
1241   EAL:   PCI memory unmapped at 0x7f83bfa80000
1242   Done
1243
1244
1245For example, to detach a virtual device port 0.
1246
1247.. code-block:: console
1248
1249   testpmd> port stop 0
1250   Stopping ports...
1251   Done
1252   testpmd> port close 0
1253   Closing ports...
1254   Done
1255
1256   testpmd> port detach 0
1257   Detaching a port...
1258   PMD: Closing pcap ethdev on numa socket 0
1259   Port 'net_pcap0' is detached. Now total ports is 0
1260   Done
1261
1262To remove a pci device completely from the system, first detach the port from testpmd.
1263Then the device should be moved under kernel management.
1264Finally the device can be removed using kernel pci hotplug functionality.
1265
1266For example, to move a pci device under kernel management:
1267
1268.. code-block:: console
1269
1270   sudo ./usertools/dpdk-devbind.py -b ixgbe 0000:0a:00.0
1271
1272   ./usertools/dpdk-devbind.py --status
1273
1274   Network devices using DPDK-compatible driver
1275   ============================================
1276   <none>
1277
1278   Network devices using kernel driver
1279   ===================================
1280   0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused=igb_uio
1281
1282To remove a port created by a virtual device, above steps are not needed.
1283
1284port start
1285~~~~~~~~~~
1286
1287Start all ports or a specific port::
1288
1289   testpmd> port start (port_id|all)
1290
1291port stop
1292~~~~~~~~~
1293
1294Stop all ports or a specific port::
1295
1296   testpmd> port stop (port_id|all)
1297
1298port close
1299~~~~~~~~~~
1300
1301Close all ports or a specific port::
1302
1303   testpmd> port close (port_id|all)
1304
1305port start/stop queue
1306~~~~~~~~~~~~~~~~~~~~~
1307
1308Start/stop a rx/tx queue on a specific port::
1309
1310   testpmd> port (port_id) (rxq|txq) (queue_id) (start|stop)
1311
1312Only take effect when port is started.
1313
1314port config - speed
1315~~~~~~~~~~~~~~~~~~~
1316
1317Set the speed and duplex mode for all ports or a specific port::
1318
1319   testpmd> port config (port_id|all) speed (10|100|1000|10000|25000|40000|50000|100000|auto) \
1320            duplex (half|full|auto)
1321
1322port config - queues/descriptors
1323~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1324
1325Set number of queues/descriptors for rxq, txq, rxd and txd::
1326
1327   testpmd> port config all (rxq|txq|rxd|txd) (value)
1328
1329This is equivalent to the ``--rxq``, ``--txq``, ``--rxd`` and ``--txd`` command-line options.
1330
1331port config - max-pkt-len
1332~~~~~~~~~~~~~~~~~~~~~~~~~
1333
1334Set the maximum packet length::
1335
1336   testpmd> port config all max-pkt-len (value)
1337
1338This is equivalent to the ``--max-pkt-len`` command-line option.
1339
1340port config - CRC Strip
1341~~~~~~~~~~~~~~~~~~~~~~~
1342
1343Set hardware CRC stripping on or off for all ports::
1344
1345   testpmd> port config all crc-strip (on|off)
1346
1347CRC stripping is off by default.
1348
1349The ``on`` option is equivalent to the ``--crc-strip`` command-line option.
1350
1351port config - scatter
1352~~~~~~~~~~~~~~~~~~~~~~~
1353
1354Set RX scatter mode on or off for all ports::
1355
1356   testpmd> port config all scatter (on|off)
1357
1358RX scatter mode is off by default.
1359
1360The ``on`` option is equivalent to the ``--enable-scatter`` command-line option.
1361
1362port config - TX queue flags
1363~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1364
1365Set a hexadecimal bitmap of TX queue flags for all ports::
1366
1367   testpmd> port config all txqflags value
1368
1369This command is equivalent to the ``--txqflags`` command-line option.
1370
1371port config - RX Checksum
1372~~~~~~~~~~~~~~~~~~~~~~~~~
1373
1374Set hardware RX checksum offload to on or off for all ports::
1375
1376   testpmd> port config all rx-cksum (on|off)
1377
1378Checksum offload is off by default.
1379
1380The ``on`` option is equivalent to the ``--enable-rx-cksum`` command-line option.
1381
1382port config - VLAN
1383~~~~~~~~~~~~~~~~~~
1384
1385Set hardware VLAN on or off for all ports::
1386
1387   testpmd> port config all hw-vlan (on|off)
1388
1389Hardware VLAN is on by default.
1390
1391The ``off`` option is equivalent to the ``--disable-hw-vlan`` command-line option.
1392
1393port config - VLAN filter
1394~~~~~~~~~~~~~~~~~~~~~~~~~
1395
1396Set hardware VLAN filter on or off for all ports::
1397
1398   testpmd> port config all hw-vlan-filter (on|off)
1399
1400Hardware VLAN filter is on by default.
1401
1402The ``off`` option is equivalent to the ``--disable-hw-vlan-filter`` command-line option.
1403
1404port config - VLAN strip
1405~~~~~~~~~~~~~~~~~~~~~~~~
1406
1407Set hardware VLAN strip on or off for all ports::
1408
1409   testpmd> port config all hw-vlan-strip (on|off)
1410
1411Hardware VLAN strip is on by default.
1412
1413The ``off`` option is equivalent to the ``--disable-hw-vlan-strip`` command-line option.
1414
1415port config - VLAN extend
1416~~~~~~~~~~~~~~~~~~~~~~~~~
1417
1418Set hardware VLAN extend on or off for all ports::
1419
1420   testpmd> port config all hw-vlan-extend (on|off)
1421
1422Hardware VLAN extend is off by default.
1423
1424The ``off`` option is equivalent to the ``--disable-hw-vlan-extend`` command-line option.
1425
1426port config - Drop Packets
1427~~~~~~~~~~~~~~~~~~~~~~~~~~
1428
1429Set packet drop for packets with no descriptors on or off for all ports::
1430
1431   testpmd> port config all drop-en (on|off)
1432
1433Packet dropping for packets with no descriptors is off by default.
1434
1435The ``on`` option is equivalent to the ``--enable-drop-en`` command-line option.
1436
1437port config - RSS
1438~~~~~~~~~~~~~~~~~
1439
1440Set the RSS (Receive Side Scaling) mode on or off::
1441
1442   testpmd> port config all rss (all|ip|tcp|udp|sctp|ether|port|vxlan|geneve|nvgre|none)
1443
1444RSS is on by default.
1445
1446The ``none`` option is equivalent to the ``--disable-rss`` command-line option.
1447
1448port config - RSS Reta
1449~~~~~~~~~~~~~~~~~~~~~~
1450
1451Set the RSS (Receive Side Scaling) redirection table::
1452
1453   testpmd> port config all rss reta (hash,queue)[,(hash,queue)]
1454
1455port config - DCB
1456~~~~~~~~~~~~~~~~~
1457
1458Set the DCB mode for an individual port::
1459
1460   testpmd> port config (port_id) dcb vt (on|off) (traffic_class) pfc (on|off)
1461
1462The traffic class should be 4 or 8.
1463
1464port config - Burst
1465~~~~~~~~~~~~~~~~~~~
1466
1467Set the number of packets per burst::
1468
1469   testpmd> port config all burst (value)
1470
1471This is equivalent to the ``--burst`` command-line option.
1472
1473port config - Threshold
1474~~~~~~~~~~~~~~~~~~~~~~~
1475
1476Set thresholds for TX/RX queues::
1477
1478   testpmd> port config all (threshold) (value)
1479
1480Where the threshold type can be:
1481
1482* ``txpt:`` Set the prefetch threshold register of the TX rings, 0 <= value <= 255.
1483
1484* ``txht:`` Set the host threshold register of the TX rings, 0 <= value <= 255.
1485
1486* ``txwt:`` Set the write-back threshold register of the TX rings, 0 <= value <= 255.
1487
1488* ``rxpt:`` Set the prefetch threshold register of the RX rings, 0 <= value <= 255.
1489
1490* ``rxht:`` Set the host threshold register of the RX rings, 0 <= value <= 255.
1491
1492* ``rxwt:`` Set the write-back threshold register of the RX rings, 0 <= value <= 255.
1493
1494* ``txfreet:`` Set the transmit free threshold of the TX rings, 0 <= value <= txd.
1495
1496* ``rxfreet:`` Set the transmit free threshold of the RX rings, 0 <= value <= rxd.
1497
1498* ``txrst:`` Set the transmit RS bit threshold of TX rings, 0 <= value <= txd.
1499
1500These threshold options are also available from the command-line.
1501
1502port config - E-tag
1503~~~~~~~~~~~~~~~~~~~
1504
1505Set the value of ether-type for E-tag::
1506
1507   testpmd> port config (port_id|all) l2-tunnel E-tag ether-type (value)
1508
1509Enable/disable the E-tag support::
1510
1511   testpmd> port config (port_id|all) l2-tunnel E-tag (enable|disable)
1512
1513
1514Link Bonding Functions
1515----------------------
1516
1517The Link Bonding functions make it possible to dynamically create and
1518manage link bonding devices from within testpmd interactive prompt.
1519
1520create bonded device
1521~~~~~~~~~~~~~~~~~~~~
1522
1523Create a new bonding device::
1524
1525   testpmd> create bonded device (mode) (socket)
1526
1527For example, to create a bonded device in mode 1 on socket 0::
1528
1529   testpmd> create bonded 1 0
1530   created new bonded device (port X)
1531
1532add bonding slave
1533~~~~~~~~~~~~~~~~~
1534
1535Adds Ethernet device to a Link Bonding device::
1536
1537   testpmd> add bonding slave (slave id) (port id)
1538
1539For example, to add Ethernet device (port 6) to a Link Bonding device (port 10)::
1540
1541   testpmd> add bonding slave 6 10
1542
1543
1544remove bonding slave
1545~~~~~~~~~~~~~~~~~~~~
1546
1547Removes an Ethernet slave device from a Link Bonding device::
1548
1549   testpmd> remove bonding slave (slave id) (port id)
1550
1551For example, to remove Ethernet slave device (port 6) to a Link Bonding device (port 10)::
1552
1553   testpmd> remove bonding slave 6 10
1554
1555set bonding mode
1556~~~~~~~~~~~~~~~~
1557
1558Set the Link Bonding mode of a Link Bonding device::
1559
1560   testpmd> set bonding mode (value) (port id)
1561
1562For example, to set the bonding mode of a Link Bonding device (port 10) to broadcast (mode 3)::
1563
1564   testpmd> set bonding mode 3 10
1565
1566set bonding primary
1567~~~~~~~~~~~~~~~~~~~
1568
1569Set an Ethernet slave device as the primary device on a Link Bonding device::
1570
1571   testpmd> set bonding primary (slave id) (port id)
1572
1573For example, to set the Ethernet slave device (port 6) as the primary port of a Link Bonding device (port 10)::
1574
1575   testpmd> set bonding primary 6 10
1576
1577set bonding mac
1578~~~~~~~~~~~~~~~
1579
1580Set the MAC address of a Link Bonding device::
1581
1582   testpmd> set bonding mac (port id) (mac)
1583
1584For example, to set the MAC address of a Link Bonding device (port 10) to 00:00:00:00:00:01::
1585
1586   testpmd> set bonding mac 10 00:00:00:00:00:01
1587
1588set bonding xmit_balance_policy
1589~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1590
1591Set the transmission policy for a Link Bonding device when it is in Balance XOR mode::
1592
1593   testpmd> set bonding xmit_balance_policy (port_id) (l2|l23|l34)
1594
1595For example, set a Link Bonding device (port 10) to use a balance policy of layer 3+4 (IP addresses & UDP ports)::
1596
1597   testpmd> set bonding xmit_balance_policy 10 l34
1598
1599
1600set bonding mon_period
1601~~~~~~~~~~~~~~~~~~~~~~
1602
1603Set the link status monitoring polling period in milliseconds for a bonding device.
1604
1605This adds support for PMD slave devices which do not support link status interrupts.
1606When the mon_period is set to a value greater than 0 then all PMD's which do not support
1607link status ISR will be queried every polling interval to check if their link status has changed::
1608
1609   testpmd> set bonding mon_period (port_id) (value)
1610
1611For example, to set the link status monitoring polling period of bonded device (port 5) to 150ms::
1612
1613   testpmd> set bonding mon_period 5 150
1614
1615
1616show bonding config
1617~~~~~~~~~~~~~~~~~~~
1618
1619Show the current configuration of a Link Bonding device::
1620
1621   testpmd> show bonding config (port id)
1622
1623For example,
1624to show the configuration a Link Bonding device (port 9) with 3 slave devices (1, 3, 4)
1625in balance mode with a transmission policy of layer 2+3::
1626
1627   testpmd> show bonding config 9
1628        Bonding mode: 2
1629        Balance Xmit Policy: BALANCE_XMIT_POLICY_LAYER23
1630        Slaves (3): [1 3 4]
1631        Active Slaves (3): [1 3 4]
1632        Primary: [3]
1633
1634
1635Register Functions
1636------------------
1637
1638The Register Functions can be used to read from and write to registers on the network card referenced by a port number.
1639This is mainly useful for debugging purposes.
1640Reference should be made to the appropriate datasheet for the network card for details on the register addresses
1641and fields that can be accessed.
1642
1643read reg
1644~~~~~~~~
1645
1646Display the value of a port register::
1647
1648   testpmd> read reg (port_id) (address)
1649
1650For example, to examine the Flow Director control register (FDIRCTL, 0x0000EE000) on an Intel 82599 10 GbE Controller::
1651
1652   testpmd> read reg 0 0xEE00
1653   port 0 PCI register at offset 0xEE00: 0x4A060029 (1241907241)
1654
1655read regfield
1656~~~~~~~~~~~~~
1657
1658Display a port register bit field::
1659
1660   testpmd> read regfield (port_id) (address) (bit_x) (bit_y)
1661
1662For example, reading the lowest two bits from the register in the example above::
1663
1664   testpmd> read regfield 0 0xEE00 0 1
1665   port 0 PCI register at offset 0xEE00: bits[0, 1]=0x1 (1)
1666
1667read regbit
1668~~~~~~~~~~~
1669
1670Display a single port register bit::
1671
1672   testpmd> read regbit (port_id) (address) (bit_x)
1673
1674For example, reading the lowest bit from the register in the example above::
1675
1676   testpmd> read regbit 0 0xEE00 0
1677   port 0 PCI register at offset 0xEE00: bit 0=1
1678
1679write reg
1680~~~~~~~~~
1681
1682Set the value of a port register::
1683
1684   testpmd> write reg (port_id) (address) (value)
1685
1686For example, to clear a register::
1687
1688   testpmd> write reg 0 0xEE00 0x0
1689   port 0 PCI register at offset 0xEE00: 0x00000000 (0)
1690
1691write regfield
1692~~~~~~~~~~~~~~
1693
1694Set bit field of a port register::
1695
1696   testpmd> write regfield (port_id) (address) (bit_x) (bit_y) (value)
1697
1698For example, writing to the register cleared in the example above::
1699
1700   testpmd> write regfield 0 0xEE00 0 1 2
1701   port 0 PCI register at offset 0xEE00: 0x00000002 (2)
1702
1703write regbit
1704~~~~~~~~~~~~
1705
1706Set single bit value of a port register::
1707
1708   testpmd> write regbit (port_id) (address) (bit_x) (value)
1709
1710For example, to set the high bit in the register from the example above::
1711
1712   testpmd> write regbit 0 0xEE00 31 1
1713   port 0 PCI register at offset 0xEE00: 0x8000000A (2147483658)
1714
1715
1716Filter Functions
1717----------------
1718
1719This section details the available filter functions that are available.
1720
1721Note these functions interface the deprecated legacy filtering framework,
1722superseded by *rte_flow*. See `Flow rules management`_.
1723
1724ethertype_filter
1725~~~~~~~~~~~~~~~~~~~~
1726
1727Add or delete a L2 Ethertype filter, which identify packets by their L2 Ethertype mainly assign them to a receive queue::
1728
1729   ethertype_filter (port_id) (add|del) (mac_addr|mac_ignr) (mac_address) \
1730                    ethertype (ether_type) (drop|fwd) queue (queue_id)
1731
1732The available information parameters are:
1733
1734* ``port_id``: The port which the Ethertype filter assigned on.
1735
1736* ``mac_addr``: Compare destination mac address.
1737
1738* ``mac_ignr``: Ignore destination mac address match.
1739
1740* ``mac_address``: Destination mac address to match.
1741
1742* ``ether_type``: The EtherType value want to match,
1743  for example 0x0806 for ARP packet. 0x0800 (IPv4) and 0x86DD (IPv6) are invalid.
1744
1745* ``queue_id``: The receive queue associated with this EtherType filter.
1746  It is meaningless when deleting or dropping.
1747
1748Example, to add/remove an ethertype filter rule::
1749
1750   testpmd> ethertype_filter 0 add mac_ignr 00:11:22:33:44:55 \
1751                             ethertype 0x0806 fwd queue 3
1752
1753   testpmd> ethertype_filter 0 del mac_ignr 00:11:22:33:44:55 \
1754                             ethertype 0x0806 fwd queue 3
1755
17562tuple_filter
1757~~~~~~~~~~~~~~~~~
1758
1759Add or delete a 2-tuple filter,
1760which identifies packets by specific protocol and destination TCP/UDP port
1761and forwards packets into one of the receive queues::
1762
1763   2tuple_filter (port_id) (add|del) dst_port (dst_port_value) \
1764                 protocol (protocol_value) mask (mask_value) \
1765                 tcp_flags (tcp_flags_value) priority (prio_value) \
1766                 queue (queue_id)
1767
1768The available information parameters are:
1769
1770* ``port_id``: The port which the 2-tuple filter assigned on.
1771
1772* ``dst_port_value``: Destination port in L4.
1773
1774* ``protocol_value``: IP L4 protocol.
1775
1776* ``mask_value``: Participates in the match or not by bit for field above, 1b means participate.
1777
1778* ``tcp_flags_value``: TCP control bits. The non-zero value is invalid, when the pro_value is not set to 0x06 (TCP).
1779
1780* ``prio_value``: Priority of this filter.
1781
1782* ``queue_id``: The receive queue associated with this 2-tuple filter.
1783
1784Example, to add/remove an 2tuple filter rule::
1785
1786   testpmd> 2tuple_filter 0 add dst_port 32 protocol 0x06 mask 0x03 \
1787                          tcp_flags 0x02 priority 3 queue 3
1788
1789   testpmd> 2tuple_filter 0 del dst_port 32 protocol 0x06 mask 0x03 \
1790                          tcp_flags 0x02 priority 3 queue 3
1791
17925tuple_filter
1793~~~~~~~~~~~~~~~~~
1794
1795Add or delete a 5-tuple filter,
1796which consists of a 5-tuple (protocol, source and destination IP addresses, source and destination TCP/UDP/SCTP port)
1797and routes packets into one of the receive queues::
1798
1799   5tuple_filter (port_id) (add|del) dst_ip (dst_address) src_ip \
1800                 (src_address) dst_port (dst_port_value) \
1801                 src_port (src_port_value) protocol (protocol_value) \
1802                 mask (mask_value) tcp_flags (tcp_flags_value) \
1803                 priority (prio_value) queue (queue_id)
1804
1805The available information parameters are:
1806
1807* ``port_id``: The port which the 5-tuple filter assigned on.
1808
1809* ``dst_address``: Destination IP address.
1810
1811* ``src_address``: Source IP address.
1812
1813* ``dst_port_value``: TCP/UDP destination port.
1814
1815* ``src_port_value``: TCP/UDP source port.
1816
1817* ``protocol_value``: L4 protocol.
1818
1819* ``mask_value``: Participates in the match or not by bit for field above, 1b means participate
1820
1821* ``tcp_flags_value``: TCP control bits. The non-zero value is invalid, when the protocol_value is not set to 0x06 (TCP).
1822
1823* ``prio_value``: The priority of this filter.
1824
1825* ``queue_id``: The receive queue associated with this 5-tuple filter.
1826
1827Example, to add/remove an 5tuple filter rule::
1828
1829   testpmd> 5tuple_filter 0 add dst_ip 2.2.2.5 src_ip 2.2.2.4 \
1830            dst_port 64 src_port 32 protocol 0x06 mask 0x1F \
1831            flags 0x0 priority 3 queue 3
1832
1833   testpmd> 5tuple_filter 0 del dst_ip 2.2.2.5 src_ip 2.2.2.4 \
1834            dst_port 64 src_port 32 protocol 0x06 mask 0x1F \
1835            flags 0x0 priority 3 queue 3
1836
1837syn_filter
1838~~~~~~~~~~
1839
1840Using the  SYN filter, TCP packets whose *SYN* flag is set can be forwarded to a separate queue::
1841
1842   syn_filter (port_id) (add|del) priority (high|low) queue (queue_id)
1843
1844The available information parameters are:
1845
1846* ``port_id``: The port which the SYN filter assigned on.
1847
1848* ``high``: This SYN filter has higher priority than other filters.
1849
1850* ``low``: This SYN filter has lower priority than other filters.
1851
1852* ``queue_id``: The receive queue associated with this SYN filter
1853
1854Example::
1855
1856   testpmd> syn_filter 0 add priority high queue 3
1857
1858flex_filter
1859~~~~~~~~~~~
1860
1861With flex filter, packets can be recognized by any arbitrary pattern within the first 128 bytes of the packet
1862and routed into one of the receive queues::
1863
1864   flex_filter (port_id) (add|del) len (len_value) bytes (bytes_value) \
1865               mask (mask_value) priority (prio_value) queue (queue_id)
1866
1867The available information parameters are:
1868
1869* ``port_id``: The port which the Flex filter is assigned on.
1870
1871* ``len_value``: Filter length in bytes, no greater than 128.
1872
1873* ``bytes_value``: A string in hexadecimal, means the value the flex filter needs to match.
1874
1875* ``mask_value``: A string in hexadecimal, bit 1 means corresponding byte participates in the match.
1876
1877* ``prio_value``: The priority of this filter.
1878
1879* ``queue_id``: The receive queue associated with this Flex filter.
1880
1881Example::
1882
1883   testpmd> flex_filter 0 add len 16 bytes 0x00000000000000000000000008060000 \
1884                          mask 000C priority 3 queue 3
1885
1886   testpmd> flex_filter 0 del len 16 bytes 0x00000000000000000000000008060000 \
1887                          mask 000C priority 3 queue 3
1888
1889
1890.. _testpmd_flow_director:
1891
1892flow_director_filter
1893~~~~~~~~~~~~~~~~~~~~
1894
1895The Flow Director works in receive mode to identify specific flows or sets of flows and route them to specific queues.
1896
1897Four types of filtering are supported which are referred to as Perfect Match, Signature, Perfect-mac-vlan and
1898Perfect-tunnel filters, the match mode is set by the ``--pkt-filter-mode`` command-line parameter:
1899
1900* Perfect 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 IP flow.
1903
1904* Signature filters.
1905  The hardware checks a match between a hash-based signature of the masked fields of the received packet.
1906
1907* Perfect-mac-vlan match filters.
1908  The hardware checks a match between the masked fields of the received packets and the programmed filters.
1909  The masked fields are for MAC VLAN flow.
1910
1911* Perfect-tunnel match filters.
1912  The hardware checks a match between the masked fields of the received packets and the programmed filters.
1913  The masked fields are for tunnel flow.
1914
1915The Flow Director filters can match the different fields for different type of packet: flow type, specific input set
1916per flow type and the flexible payload.
1917
1918The Flow Director can also mask out parts of all of these fields so that filters
1919are only applied to certain fields or parts of the fields.
1920
1921Different NICs may have different capabilities, command show port fdir (port_id) can be used to acquire the information.
1922
1923# Commands to add flow director filters of different flow types::
1924
1925   flow_director_filter (port_id) mode IP (add|del|update) \
1926                        flow (ipv4-other|ipv4-frag|ipv6-other|ipv6-frag) \
1927                        src (src_ip_address) dst (dst_ip_address) \
1928                        tos (tos_value) proto (proto_value) ttl (ttl_value) \
1929                        vlan (vlan_value) flexbytes (flexbytes_value) \
1930                        (drop|fwd) pf|vf(vf_id) queue (queue_id) \
1931                        fd_id (fd_id_value)
1932
1933   flow_director_filter (port_id) mode IP (add|del|update) \
1934                        flow (ipv4-tcp|ipv4-udp|ipv6-tcp|ipv6-udp) \
1935                        src (src_ip_address) (src_port) \
1936                        dst (dst_ip_address) (dst_port) \
1937                        tos (tos_value) ttl (ttl_value) \
1938                        vlan (vlan_value) flexbytes (flexbytes_value) \
1939                        (drop|fwd) queue pf|vf(vf_id) (queue_id) \
1940                        fd_id (fd_id_value)
1941
1942   flow_director_filter (port_id) mode IP (add|del|update) \
1943                        flow (ipv4-sctp|ipv6-sctp) \
1944                        src (src_ip_address) (src_port) \
1945                        dst (dst_ip_address) (dst_port) \
1946                        tos (tos_value) ttl (ttl_value) \
1947                        tag (verification_tag) vlan (vlan_value) \
1948                        flexbytes (flexbytes_value) (drop|fwd) \
1949                        pf|vf(vf_id) queue (queue_id) fd_id (fd_id_value)
1950
1951   flow_director_filter (port_id) mode IP (add|del|update) flow l2_payload \
1952                        ether (ethertype) flexbytes (flexbytes_value) \
1953                        (drop|fwd) pf|vf(vf_id) queue (queue_id)
1954                        fd_id (fd_id_value)
1955
1956   flow_director_filter (port_id) mode MAC-VLAN (add|del|update) \
1957                        mac (mac_address) vlan (vlan_value) \
1958                        flexbytes (flexbytes_value) (drop|fwd) \
1959                        queue (queue_id) fd_id (fd_id_value)
1960
1961   flow_director_filter (port_id) mode Tunnel (add|del|update) \
1962                        mac (mac_address) vlan (vlan_value) \
1963                        tunnel (NVGRE|VxLAN) tunnel-id (tunnel_id_value) \
1964                        flexbytes (flexbytes_value) (drop|fwd) \
1965                        queue (queue_id) fd_id (fd_id_value)
1966
1967For example, to add an ipv4-udp flow type filter::
1968
1969   testpmd> flow_director_filter 0 mode IP add flow ipv4-udp src 2.2.2.3 32 \
1970            dst 2.2.2.5 33 tos 2 ttl 40 vlan 0x1 flexbytes (0x88,0x48) \
1971            fwd pf queue 1 fd_id 1
1972
1973For example, add an ipv4-other flow type filter::
1974
1975   testpmd> flow_director_filter 0 mode IP add flow ipv4-other src 2.2.2.3 \
1976             dst 2.2.2.5 tos 2 proto 20 ttl 40 vlan 0x1 \
1977             flexbytes (0x88,0x48) fwd pf queue 1 fd_id 1
1978
1979flush_flow_director
1980~~~~~~~~~~~~~~~~~~~
1981
1982Flush all flow director filters on a device::
1983
1984   testpmd> flush_flow_director (port_id)
1985
1986Example, to flush all flow director filter on port 0::
1987
1988   testpmd> flush_flow_director 0
1989
1990flow_director_mask
1991~~~~~~~~~~~~~~~~~~
1992
1993Set flow director's input masks::
1994
1995   flow_director_mask (port_id) mode IP vlan (vlan_value) \
1996                      src_mask (ipv4_src) (ipv6_src) (src_port) \
1997                      dst_mask (ipv4_dst) (ipv6_dst) (dst_port)
1998
1999   flow_director_mask (port_id) mode MAC-VLAN vlan (vlan_value)
2000
2001   flow_director_mask (port_id) mode Tunnel vlan (vlan_value) \
2002                      mac (mac_value) tunnel-type (tunnel_type_value) \
2003                      tunnel-id (tunnel_id_value)
2004
2005Example, to set flow director mask on port 0::
2006
2007   testpmd> flow_director_mask 0 mode IP vlan 0xefff \
2008            src_mask 255.255.255.255 \
2009                FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF \
2010            dst_mask 255.255.255.255 \
2011                FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF
2012
2013flow_director_flex_mask
2014~~~~~~~~~~~~~~~~~~~~~~~
2015
2016set masks of flow director's flexible payload based on certain flow type::
2017
2018   testpmd> flow_director_flex_mask (port_id) \
2019            flow (none|ipv4-other|ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
2020                  ipv6-other|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp| \
2021                  l2_payload|all) (mask)
2022
2023Example, to set flow director's flex mask for all flow type on port 0::
2024
2025   testpmd> flow_director_flex_mask 0 flow all \
2026            (0xff,0xff,0,0,0,0,0,0,0,0,0,0,0,0,0,0)
2027
2028
2029flow_director_flex_payload
2030~~~~~~~~~~~~~~~~~~~~~~~~~~
2031
2032Configure flexible payload selection::
2033
2034   flow_director_flex_payload (port_id) (raw|l2|l3|l4) (config)
2035
2036For example, to select the first 16 bytes from the offset 4 (bytes) of packet's payload as flexible payload::
2037
2038   testpmd> flow_director_flex_payload 0 l4 \
2039            (4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19)
2040
2041get_sym_hash_ena_per_port
2042~~~~~~~~~~~~~~~~~~~~~~~~~
2043
2044Get symmetric hash enable configuration per port::
2045
2046   get_sym_hash_ena_per_port (port_id)
2047
2048For example, to get symmetric hash enable configuration of port 1::
2049
2050   testpmd> get_sym_hash_ena_per_port 1
2051
2052set_sym_hash_ena_per_port
2053~~~~~~~~~~~~~~~~~~~~~~~~~
2054
2055Set symmetric hash enable configuration per port to enable or disable::
2056
2057   set_sym_hash_ena_per_port (port_id) (enable|disable)
2058
2059For example, to set symmetric hash enable configuration of port 1 to enable::
2060
2061   testpmd> set_sym_hash_ena_per_port 1 enable
2062
2063get_hash_global_config
2064~~~~~~~~~~~~~~~~~~~~~~
2065
2066Get the global configurations of hash filters::
2067
2068   get_hash_global_config (port_id)
2069
2070For example, to get the global configurations of hash filters of port 1::
2071
2072   testpmd> get_hash_global_config 1
2073
2074set_hash_global_config
2075~~~~~~~~~~~~~~~~~~~~~~
2076
2077Set the global configurations of hash filters::
2078
2079   set_hash_global_config (port_id) (toeplitz|simple_xor|default) \
2080   (ipv4|ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp|ipv4-other|ipv6|ipv6-frag| \
2081   ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other|l2_payload) \
2082   (enable|disable)
2083
2084For example, to enable simple_xor for flow type of ipv6 on port 2::
2085
2086   testpmd> set_hash_global_config 2 simple_xor ipv6 enable
2087
2088set_hash_input_set
2089~~~~~~~~~~~~~~~~~~
2090
2091Set the input set for hash::
2092
2093   set_hash_input_set (port_id) (ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
2094   ipv4-other|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other| \
2095   l2_payload) (ovlan|ivlan|src-ipv4|dst-ipv4|src-ipv6|dst-ipv6|ipv4-tos| \
2096   ipv4-proto|ipv6-tc|ipv6-next-header|udp-src-port|udp-dst-port| \
2097   tcp-src-port|tcp-dst-port|sctp-src-port|sctp-dst-port|sctp-veri-tag| \
2098   udp-key|gre-key|fld-1st|fld-2nd|fld-3rd|fld-4th|fld-5th|fld-6th|fld-7th| \
2099   fld-8th|none) (select|add)
2100
2101For example, to add source IP to hash input set for flow type of ipv4-udp on port 0::
2102
2103   testpmd> set_hash_input_set 0 ipv4-udp src-ipv4 add
2104
2105set_fdir_input_set
2106~~~~~~~~~~~~~~~~~~
2107
2108The Flow Director filters can match the different fields for different type of packet, i.e. specific input set
2109on per flow type and the flexible payload. This command can be used to change input set for each flow type.
2110
2111Set the input set for flow director::
2112
2113   set_fdir_input_set (port_id) (ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
2114   ipv4-other|ipv6|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other| \
2115   l2_payload) (ivlan|ethertype|src-ipv4|dst-ipv4|src-ipv6|dst-ipv6|ipv4-tos| \
2116   ipv4-proto|ipv4-ttl|ipv6-tc|ipv6-next-header|ipv6-hop-limits| \
2117   tudp-src-port|udp-dst-port|cp-src-port|tcp-dst-port|sctp-src-port| \
2118   sctp-dst-port|sctp-veri-tag|none) (select|add)
2119
2120For example to add source IP to FD input set for flow type of ipv4-udp on port 0::
2121
2122   testpmd> set_fdir_input_set 0 ipv4-udp src-ipv4 add
2123
2124global_config
2125~~~~~~~~~~~~~
2126
2127Set different GRE key length for input set::
2128
2129   global_config (port_id) gre-key-len (number in bytes)
2130
2131For example to set GRE key length for input set to 4 bytes on port 0::
2132
2133   testpmd> global_config 0 gre-key-len 4
2134
2135
2136.. _testpmd_rte_flow:
2137
2138Flow rules management
2139---------------------
2140
2141Control of the generic flow API (*rte_flow*) is fully exposed through the
2142``flow`` command (validation, creation, destruction and queries).
2143
2144Considering *rte_flow* overlaps with all `Filter Functions`_, using both
2145features simultaneously may cause undefined side-effects and is therefore
2146not recommended.
2147
2148``flow`` syntax
2149~~~~~~~~~~~~~~~
2150
2151Because the ``flow`` command uses dynamic tokens to handle the large number
2152of possible flow rules combinations, its behavior differs slightly from
2153other commands, in particular:
2154
2155- Pressing *?* or the *<tab>* key displays contextual help for the current
2156  token, not that of the entire command.
2157
2158- Optional and repeated parameters are supported (provided they are listed
2159  in the contextual help).
2160
2161The first parameter stands for the operation mode. Possible operations and
2162their general syntax are described below. They are covered in detail in the
2163following sections.
2164
2165- Check whether a flow rule can be created::
2166
2167   flow validate {port_id}
2168       [group {group_id}] [priority {level}] [ingress] [egress]
2169       pattern {item} [/ {item} [...]] / end
2170       actions {action} [/ {action} [...]] / end
2171
2172- Create a flow rule::
2173
2174   flow create {port_id}
2175       [group {group_id}] [priority {level}] [ingress] [egress]
2176       pattern {item} [/ {item} [...]] / end
2177       actions {action} [/ {action} [...]] / end
2178
2179- Destroy specific flow rules::
2180
2181   flow destroy {port_id} rule {rule_id} [...]
2182
2183- Destroy all flow rules::
2184
2185   flow flush {port_id}
2186
2187- Query an existing flow rule::
2188
2189   flow query {port_id} {rule_id} {action}
2190
2191- List existing flow rules sorted by priority, filtered by group
2192  identifiers::
2193
2194   flow list {port_id} [group {group_id}] [...]
2195
2196Validating flow rules
2197~~~~~~~~~~~~~~~~~~~~~
2198
2199``flow validate`` reports whether a flow rule would be accepted by the
2200underlying device in its current state but stops short of creating it. It is
2201bound to ``rte_flow_validate()``::
2202
2203   flow validate {port_id}
2204      [group {group_id}] [priority {level}] [ingress] [egress]
2205      pattern {item} [/ {item} [...]] / end
2206      actions {action} [/ {action} [...]] / end
2207
2208If successful, it will show::
2209
2210   Flow rule validated
2211
2212Otherwise it will show an error message of the form::
2213
2214   Caught error type [...] ([...]): [...]
2215
2216This command uses the same parameters as ``flow create``, their format is
2217described in `Creating flow rules`_.
2218
2219Check whether redirecting any Ethernet packet received on port 0 to RX queue
2220index 6 is supported::
2221
2222   testpmd> flow validate 0 ingress pattern eth / end
2223      actions queue index 6 / end
2224   Flow rule validated
2225   testpmd>
2226
2227Port 0 does not support TCPv6 rules::
2228
2229   testpmd> flow validate 0 ingress pattern eth / ipv6 / tcp / end
2230      actions drop / end
2231   Caught error type 9 (specific pattern item): Invalid argument
2232   testpmd>
2233
2234Creating flow rules
2235~~~~~~~~~~~~~~~~~~~
2236
2237``flow create`` validates and creates the specified flow rule. It is bound
2238to ``rte_flow_create()``::
2239
2240   flow create {port_id}
2241      [group {group_id}] [priority {level}] [ingress] [egress]
2242      pattern {item} [/ {item} [...]] / end
2243      actions {action} [/ {action} [...]] / end
2244
2245If successful, it will return a flow rule ID usable with other commands::
2246
2247   Flow rule #[...] created
2248
2249Otherwise it will show an error message of the form::
2250
2251   Caught error type [...] ([...]): [...]
2252
2253Parameters describe in the following order:
2254
2255- Attributes (*group*, *priority*, *ingress*, *egress* tokens).
2256- A matching pattern, starting with the *pattern* token and terminated by an
2257  *end* pattern item.
2258- Actions, starting with the *actions* token and terminated by an *end*
2259  action.
2260
2261These translate directly to *rte_flow* objects provided as-is to the
2262underlying functions.
2263
2264The shortest valid definition only comprises mandatory tokens::
2265
2266   testpmd> flow create 0 pattern end actions end
2267
2268Note that PMDs may refuse rules that essentially do nothing such as this
2269one.
2270
2271**All unspecified object values are automatically initialized to 0.**
2272
2273Attributes
2274^^^^^^^^^^
2275
2276These tokens affect flow rule attributes (``struct rte_flow_attr``) and are
2277specified before the ``pattern`` token.
2278
2279- ``group {group id}``: priority group.
2280- ``priority {level}``: priority level within group.
2281- ``ingress``: rule applies to ingress traffic.
2282- ``egress``: rule applies to egress traffic.
2283
2284Each instance of an attribute specified several times overrides the previous
2285value as shown below (group 4 is used)::
2286
2287   testpmd> flow create 0 group 42 group 24 group 4 [...]
2288
2289Note that once enabled, ``ingress`` and ``egress`` cannot be disabled.
2290
2291While not specifying a direction is an error, some rules may allow both
2292simultaneously.
2293
2294Most rules affect RX therefore contain the ``ingress`` token::
2295
2296   testpmd> flow create 0 ingress pattern [...]
2297
2298Matching pattern
2299^^^^^^^^^^^^^^^^
2300
2301A matching pattern starts after the ``pattern`` token. It is made of pattern
2302items and is terminated by a mandatory ``end`` item.
2303
2304Items are named after their type (*RTE_FLOW_ITEM_TYPE_* from ``enum
2305rte_flow_item_type``).
2306
2307The ``/`` token is used as a separator between pattern items as shown
2308below::
2309
2310   testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end [...]
2311
2312Note that protocol items like these must be stacked from lowest to highest
2313layer to make sense. For instance, the following rule is either invalid or
2314unlikely to match any packet::
2315
2316   testpmd> flow create 0 ingress pattern eth / udp / ipv4 / end [...]
2317
2318More information on these restrictions can be found in the *rte_flow*
2319documentation.
2320
2321Several items support additional specification structures, for example
2322``ipv4`` allows specifying source and destination addresses as follows::
2323
2324   testpmd> flow create 0 ingress pattern eth / ipv4 src is 10.1.1.1
2325      dst is 10.2.0.0 / end [...]
2326
2327This rule matches all IPv4 traffic with the specified properties.
2328
2329In this example, ``src`` and ``dst`` are field names of the underlying
2330``struct rte_flow_item_ipv4`` object. All item properties can be specified
2331in a similar fashion.
2332
2333The ``is`` token means that the subsequent value must be matched exactly,
2334and assigns ``spec`` and ``mask`` fields in ``struct rte_flow_item``
2335accordingly. Possible assignment tokens are:
2336
2337- ``is``: match value perfectly (with full bit-mask).
2338- ``spec``: match value according to configured bit-mask.
2339- ``last``: specify upper bound to establish a range.
2340- ``mask``: specify bit-mask with relevant bits set to one.
2341- ``prefix``: generate bit-mask from a prefix length.
2342
2343These yield identical results::
2344
2345   ipv4 src is 10.1.1.1
2346
2347::
2348
2349   ipv4 src spec 10.1.1.1 src mask 255.255.255.255
2350
2351::
2352
2353   ipv4 src spec 10.1.1.1 src prefix 32
2354
2355::
2356
2357   ipv4 src is 10.1.1.1 src last 10.1.1.1 # range with a single value
2358
2359::
2360
2361   ipv4 src is 10.1.1.1 src last 0 # 0 disables range
2362
2363Inclusive ranges can be defined with ``last``::
2364
2365   ipv4 src is 10.1.1.1 src last 10.2.3.4 # 10.1.1.1 to 10.2.3.4
2366
2367Note that ``mask`` affects both ``spec`` and ``last``::
2368
2369   ipv4 src is 10.1.1.1 src last 10.2.3.4 src mask 255.255.0.0
2370      # matches 10.1.0.0 to 10.2.255.255
2371
2372Properties can be modified multiple times::
2373
2374   ipv4 src is 10.1.1.1 src is 10.1.2.3 src is 10.2.3.4 # matches 10.2.3.4
2375
2376::
2377
2378   ipv4 src is 10.1.1.1 src prefix 24 src prefix 16 # matches 10.1.0.0/16
2379
2380Pattern items
2381^^^^^^^^^^^^^
2382
2383This section lists supported pattern items and their attributes, if any.
2384
2385- ``end``: end list of pattern items.
2386
2387- ``void``: no-op pattern item.
2388
2389- ``invert``: perform actions when pattern does not match.
2390
2391- ``any``: match any protocol for the current layer.
2392
2393  - ``num {unsigned}``: number of layers covered.
2394
2395- ``pf``: match packets addressed to the physical function.
2396
2397- ``vf``: match packets addressed to a virtual function ID.
2398
2399  - ``id {unsigned}``: destination VF ID.
2400
2401- ``port``: device-specific physical port index to use.
2402
2403  - ``index {unsigned}``: physical port index.
2404
2405- ``raw``: match an arbitrary byte string.
2406
2407  - ``relative {boolean}``: look for pattern after the previous item.
2408  - ``search {boolean}``: search pattern from offset (see also limit).
2409  - ``offset {integer}``: absolute or relative offset for pattern.
2410  - ``limit {unsigned}``: search area limit for start of pattern.
2411  - ``pattern {string}``: byte string to look for.
2412
2413- ``eth``: match Ethernet header.
2414
2415  - ``dst {MAC-48}``: destination MAC.
2416  - ``src {MAC-48}``: source MAC.
2417  - ``type {unsigned}``: EtherType.
2418
2419- ``vlan``: match 802.1Q/ad VLAN tag.
2420
2421  - ``tpid {unsigned}``: tag protocol identifier.
2422  - ``tci {unsigned}``: tag control information.
2423  - ``pcp {unsigned}``: priority code point.
2424  - ``dei {unsigned}``: drop eligible indicator.
2425  - ``vid {unsigned}``: VLAN identifier.
2426
2427- ``ipv4``: match IPv4 header.
2428
2429  - ``tos {unsigned}``: type of service.
2430  - ``ttl {unsigned}``: time to live.
2431  - ``proto {unsigned}``: next protocol ID.
2432  - ``src {ipv4 address}``: source address.
2433  - ``dst {ipv4 address}``: destination address.
2434
2435- ``ipv6``: match IPv6 header.
2436
2437  - ``tc {unsigned}``: traffic class.
2438  - ``flow {unsigned}``: flow label.
2439  - ``proto {unsigned}``: protocol (next header).
2440  - ``hop {unsigned}``: hop limit.
2441  - ``src {ipv6 address}``: source address.
2442  - ``dst {ipv6 address}``: destination address.
2443
2444- ``icmp``: match ICMP header.
2445
2446  - ``type {unsigned}``: ICMP packet type.
2447  - ``code {unsigned}``: ICMP packet code.
2448
2449- ``udp``: match UDP header.
2450
2451  - ``src {unsigned}``: UDP source port.
2452  - ``dst {unsigned}``: UDP destination port.
2453
2454- ``tcp``: match TCP header.
2455
2456  - ``src {unsigned}``: TCP source port.
2457  - ``dst {unsigned}``: TCP destination port.
2458
2459- ``sctp``: match SCTP header.
2460
2461  - ``src {unsigned}``: SCTP source port.
2462  - ``dst {unsigned}``: SCTP destination port.
2463  - ``tag {unsigned}``: validation tag.
2464  - ``cksum {unsigned}``: checksum.
2465
2466- ``vxlan``: match VXLAN header.
2467
2468  - ``vni {unsigned}``: VXLAN identifier.
2469
2470Actions list
2471^^^^^^^^^^^^
2472
2473A list of actions starts after the ``actions`` token in the same fashion as
2474`Matching pattern`_; actions are separated by ``/`` tokens and the list is
2475terminated by a mandatory ``end`` action.
2476
2477Actions are named after their type (*RTE_FLOW_ACTION_TYPE_* from ``enum
2478rte_flow_action_type``).
2479
2480Dropping all incoming UDPv4 packets can be expressed as follows::
2481
2482   testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
2483      actions drop / end
2484
2485Several actions have configurable properties which must be specified when
2486there is no valid default value. For example, ``queue`` requires a target
2487queue index.
2488
2489This rule redirects incoming UDPv4 traffic to queue index 6::
2490
2491   testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
2492      actions queue index 6 / end
2493
2494While this one could be rejected by PMDs (unspecified queue index)::
2495
2496   testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
2497      actions queue / end
2498
2499As defined by *rte_flow*, the list is not ordered, all actions of a given
2500rule are performed simultaneously. These are equivalent::
2501
2502   queue index 6 / void / mark id 42 / end
2503
2504::
2505
2506   void / mark id 42 / queue index 6 / end
2507
2508All actions in a list should have different types, otherwise only the last
2509action of a given type is taken into account::
2510
2511   queue index 4 / queue index 5 / queue index 6 / end # will use queue 6
2512
2513::
2514
2515   drop / drop / drop / end # drop is performed only once
2516
2517::
2518
2519   mark id 42 / queue index 3 / mark id 24 / end # mark will be 24
2520
2521Considering they are performed simultaneously, opposite and overlapping
2522actions can sometimes be combined when the end result is unambiguous::
2523
2524   drop / queue index 6 / end # drop has no effect
2525
2526::
2527
2528   drop / dup index 6 / end # same as above
2529
2530::
2531
2532   queue index 6 / rss queues 6 7 8 / end # queue has no effect
2533
2534::
2535
2536   drop / passthru / end # drop has no effect
2537
2538Note that PMDs may still refuse such combinations.
2539
2540Actions
2541^^^^^^^
2542
2543This section lists supported actions and their attributes, if any.
2544
2545- ``end``: end list of actions.
2546
2547- ``void``: no-op action.
2548
2549- ``passthru``: let subsequent rule process matched packets.
2550
2551- ``mark``: attach 32 bit value to packets.
2552
2553  - ``id {unsigned}``: 32 bit value to return with packets.
2554
2555- ``flag``: flag packets.
2556
2557- ``queue``: assign packets to a given queue index.
2558
2559  - ``index {unsigned}``: queue index to use.
2560
2561- ``drop``: drop packets (note: passthru has priority).
2562
2563- ``count``: enable counters for this rule.
2564
2565- ``dup``: duplicate packets to a given queue index.
2566
2567  - ``index {unsigned}``: queue index to duplicate packets to.
2568
2569- ``rss``: spread packets among several queues.
2570
2571  - ``queues [{unsigned} [...]] end``: queue indices to use.
2572
2573- ``pf``: redirect packets to physical device function.
2574
2575- ``vf``: redirect packets to virtual device function.
2576
2577  - ``original {boolean}``: use original VF ID if possible.
2578  - ``id {unsigned}``: VF ID to redirect packets to.
2579
2580Destroying flow rules
2581~~~~~~~~~~~~~~~~~~~~~
2582
2583``flow destroy`` destroys one or more rules from their rule ID (as returned
2584by ``flow create``), this command calls ``rte_flow_destroy()`` as many
2585times as necessary::
2586
2587   flow destroy {port_id} rule {rule_id} [...]
2588
2589If successful, it will show::
2590
2591   Flow rule #[...] destroyed
2592
2593It does not report anything for rule IDs that do not exist. The usual error
2594message is shown when a rule cannot be destroyed::
2595
2596   Caught error type [...] ([...]): [...]
2597
2598``flow flush`` destroys all rules on a device and does not take extra
2599arguments. It is bound to ``rte_flow_flush()``::
2600
2601   flow flush {port_id}
2602
2603Any errors are reported as above.
2604
2605Creating several rules and destroying them::
2606
2607   testpmd> flow create 0 ingress pattern eth / ipv6 / end
2608      actions queue index 2 / end
2609   Flow rule #0 created
2610   testpmd> flow create 0 ingress pattern eth / ipv4 / end
2611      actions queue index 3 / end
2612   Flow rule #1 created
2613   testpmd> flow destroy 0 rule 0 rule 1
2614   Flow rule #1 destroyed
2615   Flow rule #0 destroyed
2616   testpmd>
2617
2618The same result can be achieved using ``flow flush``::
2619
2620   testpmd> flow create 0 ingress pattern eth / ipv6 / end
2621      actions queue index 2 / end
2622   Flow rule #0 created
2623   testpmd> flow create 0 ingress pattern eth / ipv4 / end
2624      actions queue index 3 / end
2625   Flow rule #1 created
2626   testpmd> flow flush 0
2627   testpmd>
2628
2629Non-existent rule IDs are ignored::
2630
2631   testpmd> flow create 0 ingress pattern eth / ipv6 / end
2632      actions queue index 2 / end
2633   Flow rule #0 created
2634   testpmd> flow create 0 ingress pattern eth / ipv4 / end
2635      actions queue index 3 / end
2636   Flow rule #1 created
2637   testpmd> flow destroy 0 rule 42 rule 10 rule 2
2638   testpmd>
2639   testpmd> flow destroy 0 rule 0
2640   Flow rule #0 destroyed
2641   testpmd>
2642
2643Querying flow rules
2644~~~~~~~~~~~~~~~~~~~
2645
2646``flow query`` queries a specific action of a flow rule having that
2647ability. Such actions collect information that can be reported using this
2648command. It is bound to ``rte_flow_query()``::
2649
2650   flow query {port_id} {rule_id} {action}
2651
2652If successful, it will display either the retrieved data for known actions
2653or the following message::
2654
2655   Cannot display result for action type [...] ([...])
2656
2657Otherwise, it will complain either that the rule does not exist or that some
2658error occurred::
2659
2660   Flow rule #[...] not found
2661
2662::
2663
2664   Caught error type [...] ([...]): [...]
2665
2666Currently only the ``count`` action is supported. This action reports the
2667number of packets that hit the flow rule and the total number of bytes. Its
2668output has the following format::
2669
2670   count:
2671    hits_set: [...] # whether "hits" contains a valid value
2672    bytes_set: [...] # whether "bytes" contains a valid value
2673    hits: [...] # number of packets
2674    bytes: [...] # number of bytes
2675
2676Querying counters for TCPv6 packets redirected to queue 6::
2677
2678   testpmd> flow create 0 ingress pattern eth / ipv6 / tcp / end
2679      actions queue index 6 / count / end
2680   Flow rule #4 created
2681   testpmd> flow query 0 4 count
2682   count:
2683    hits_set: 1
2684    bytes_set: 0
2685    hits: 386446
2686    bytes: 0
2687   testpmd>
2688
2689Listing flow rules
2690~~~~~~~~~~~~~~~~~~
2691
2692``flow list`` lists existing flow rules sorted by priority and optionally
2693filtered by group identifiers::
2694
2695   flow list {port_id} [group {group_id}] [...]
2696
2697This command only fails with the following message if the device does not
2698exist::
2699
2700   Invalid port [...]
2701
2702Output consists of a header line followed by a short description of each
2703flow rule, one per line. There is no output at all when no flow rules are
2704configured on the device::
2705
2706   ID      Group   Prio    Attr    Rule
2707   [...]   [...]   [...]   [...]   [...]
2708
2709``Attr`` column flags:
2710
2711- ``i`` for ``ingress``.
2712- ``e`` for ``egress``.
2713
2714Creating several flow rules and listing them::
2715
2716   testpmd> flow create 0 ingress pattern eth / ipv4 / end
2717      actions queue index 6 / end
2718   Flow rule #0 created
2719   testpmd> flow create 0 ingress pattern eth / ipv6 / end
2720      actions queue index 2 / end
2721   Flow rule #1 created
2722   testpmd> flow create 0 priority 5 ingress pattern eth / ipv4 / udp / end
2723      actions rss queues 6 7 8 end / end
2724   Flow rule #2 created
2725   testpmd> flow list 0
2726   ID      Group   Prio    Attr    Rule
2727   0       0       0       i-      ETH IPV4 => QUEUE
2728   1       0       0       i-      ETH IPV6 => QUEUE
2729   2       0       5       i-      ETH IPV4 UDP => RSS
2730   testpmd>
2731
2732Rules are sorted by priority (i.e. group ID first, then priority level)::
2733
2734   testpmd> flow list 1
2735   ID      Group   Prio    Attr    Rule
2736   0       0       0       i-      ETH => COUNT
2737   6       0       500     i-      ETH IPV6 TCP => DROP COUNT
2738   5       0       1000    i-      ETH IPV6 ICMP => QUEUE
2739   1       24      0       i-      ETH IPV4 UDP => QUEUE
2740   4       24      10      i-      ETH IPV4 TCP => DROP
2741   3       24      20      i-      ETH IPV4 => DROP
2742   2       24      42      i-      ETH IPV4 UDP => QUEUE
2743   7       63      0       i-      ETH IPV6 UDP VXLAN => MARK QUEUE
2744   testpmd>
2745
2746Output can be limited to specific groups::
2747
2748   testpmd> flow list 1 group 0 group 63
2749   ID      Group   Prio    Attr    Rule
2750   0       0       0       i-      ETH => COUNT
2751   6       0       500     i-      ETH IPV6 TCP => DROP COUNT
2752   5       0       1000    i-      ETH IPV6 ICMP => QUEUE
2753   7       63      0       i-      ETH IPV6 UDP VXLAN => MARK QUEUE
2754   testpmd>
2755