xref: /dpdk/doc/guides/testpmd_app_ug/testpmd_funcs.rst (revision 1315219a22a32aa438de3483eaacc104890ee71f)
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 tc tx max bandwidth (for VF)
911~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
912
913Set a TC's TX max absolute bandwidth (Mbps) for a VF from PF::
914
915   testpmd> set vf tc tx max-bandwidth (port_id) (vf_id) (tc_no) (max_bandwidth)
916
917set tc strict link priority mode
918~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
919
920Set some TCs' strict link priority mode on a physical port::
921
922   testpmd> set tx strict-link-priority (port_id) (tc_bitmap)
923
924set flow_ctrl rx
925~~~~~~~~~~~~~~~~
926
927Set the link flow control parameter on a port::
928
929   testpmd> set flow_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \
930            (pause_time) (send_xon) mac_ctrl_frame_fwd (on|off) \
931	    autoneg (on|off) (port_id)
932
933Where:
934
935* ``high_water`` (integer): High threshold value to trigger XOFF.
936
937* ``low_water`` (integer): Low threshold value to trigger XON.
938
939* ``pause_time`` (integer): Pause quota in the Pause frame.
940
941* ``send_xon`` (0/1): Send XON frame.
942
943* ``mac_ctrl_frame_fwd``: Enable receiving MAC control frames.
944
945* ``autoneg``: Change the auto-negotiation parameter.
946
947set pfc_ctrl rx
948~~~~~~~~~~~~~~~
949
950Set the priority flow control parameter on a port::
951
952   testpmd> set pfc_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \
953            (pause_time) (priority) (port_id)
954
955Where:
956
957* ``high_water`` (integer): High threshold value.
958
959* ``low_water`` (integer): Low threshold value.
960
961* ``pause_time`` (integer): Pause quota in the Pause frame.
962
963* ``priority`` (0-7): VLAN User Priority.
964
965set stat_qmap
966~~~~~~~~~~~~~
967
968Set statistics mapping (qmapping 0..15) for RX/TX queue on port::
969
970   testpmd> set stat_qmap (tx|rx) (port_id) (queue_id) (qmapping)
971
972For example, to set rx queue 2 on port 0 to mapping 5::
973
974   testpmd>set stat_qmap rx 0 2 5
975
976set port - rx/tx (for VF)
977~~~~~~~~~~~~~~~~~~~~~~~~~
978
979Set VF receive/transmit from a port::
980
981   testpmd> set port (port_id) vf (vf_id) (rx|tx) (on|off)
982
983set port - mac address filter (for VF)
984~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
985
986Add/Remove unicast or multicast MAC addr filter for a VF::
987
988   testpmd> set port (port_id) vf (vf_id) (mac_addr) \
989            (exact-mac|exact-mac-vlan|hashmac|hashmac-vlan) (on|off)
990
991set port - rx mode(for VF)
992~~~~~~~~~~~~~~~~~~~~~~~~~~
993
994Set the VF receive mode of a port::
995
996   testpmd> set port (port_id) vf (vf_id) \
997            rxmode (AUPE|ROPE|BAM|MPE) (on|off)
998
999The available receive modes are:
1000
1001* ``AUPE``: Accepts untagged VLAN.
1002
1003* ``ROPE``: Accepts unicast hash.
1004
1005* ``BAM``: Accepts broadcast packets.
1006
1007* ``MPE``: Accepts all multicast packets.
1008
1009set port - tx_rate (for Queue)
1010~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1011
1012Set TX rate limitation for a queue on a port::
1013
1014   testpmd> set port (port_id) queue (queue_id) rate (rate_value)
1015
1016set port - tx_rate (for VF)
1017~~~~~~~~~~~~~~~~~~~~~~~~~~~
1018
1019Set TX rate limitation for queues in VF on a port::
1020
1021   testpmd> set port (port_id) vf (vf_id) rate (rate_value) queue_mask (queue_mask)
1022
1023set port - mirror rule
1024~~~~~~~~~~~~~~~~~~~~~~
1025
1026Set pool or vlan type mirror rule for a port::
1027
1028   testpmd> set port (port_id) mirror-rule (rule_id) \
1029            (pool-mirror-up|pool-mirror-down|vlan-mirror) \
1030            (poolmask|vlanid[,vlanid]*) dst-pool (pool_id) (on|off)
1031
1032Set link mirror rule for a port::
1033
1034   testpmd> set port (port_id) mirror-rule (rule_id) \
1035           (uplink-mirror|downlink-mirror) dst-pool (pool_id) (on|off)
1036
1037For example to enable mirror traffic with vlan 0,1 to pool 0::
1038
1039   set port 0 mirror-rule 0 vlan-mirror 0,1 dst-pool 0 on
1040
1041reset port - mirror rule
1042~~~~~~~~~~~~~~~~~~~~~~~~
1043
1044Reset a mirror rule for a port::
1045
1046   testpmd> reset port (port_id) mirror-rule (rule_id)
1047
1048set flush_rx
1049~~~~~~~~~~~~
1050
1051Set the flush on RX streams before forwarding.
1052The default is flush ``on``.
1053Mainly used with PCAP drivers to turn off the default behavior of flushing the first 512 packets on RX streams::
1054
1055   testpmd> set flush_rx off
1056
1057set bypass mode
1058~~~~~~~~~~~~~~~
1059
1060Set the bypass mode for the lowest port on bypass enabled NIC::
1061
1062   testpmd> set bypass mode (normal|bypass|isolate) (port_id)
1063
1064set bypass event
1065~~~~~~~~~~~~~~~~
1066
1067Set the event required to initiate specified bypass mode for the lowest port on a bypass enabled::
1068
1069   testpmd> set bypass event (timeout|os_on|os_off|power_on|power_off) \
1070            mode (normal|bypass|isolate) (port_id)
1071
1072Where:
1073
1074* ``timeout``: Enable bypass after watchdog timeout.
1075
1076* ``os_on``: Enable bypass when OS/board is powered on.
1077
1078* ``os_off``: Enable bypass when OS/board is powered off.
1079
1080* ``power_on``: Enable bypass when power supply is turned on.
1081
1082* ``power_off``: Enable bypass when power supply is turned off.
1083
1084
1085set bypass timeout
1086~~~~~~~~~~~~~~~~~~
1087
1088Set the bypass watchdog timeout to ``n`` seconds where 0 = instant::
1089
1090   testpmd> set bypass timeout (0|1.5|2|3|4|8|16|32)
1091
1092show bypass config
1093~~~~~~~~~~~~~~~~~~
1094
1095Show the bypass configuration for a bypass enabled NIC using the lowest port on the NIC::
1096
1097   testpmd> show bypass config (port_id)
1098
1099set link up
1100~~~~~~~~~~~
1101
1102Set link up for a port::
1103
1104   testpmd> set link-up port (port id)
1105
1106set link down
1107~~~~~~~~~~~~~
1108
1109Set link down for a port::
1110
1111   testpmd> set link-down port (port id)
1112
1113E-tag set
1114~~~~~~~~~
1115
1116Enable E-tag insertion for a VF on a port::
1117
1118   testpmd> E-tag set insertion on port-tag-id (value) port (port_id) vf (vf_id)
1119
1120Disable E-tag insertion for a VF on a port::
1121
1122   testpmd> E-tag set insertion off port (port_id) vf (vf_id)
1123
1124Enable/disable E-tag stripping on a port::
1125
1126   testpmd> E-tag set stripping (on|off) port (port_id)
1127
1128Enable/disable E-tag based forwarding on a port::
1129
1130   testpmd> E-tag set forwarding (on|off) port (port_id)
1131
1132Add an E-tag forwarding filter on a port::
1133
1134   testpmd> E-tag set filter add e-tag-id (value) dst-pool (pool_id) port (port_id)
1135
1136Delete an E-tag forwarding filter on a port::
1137   testpmd> E-tag set filter del e-tag-id (value) port (port_id)
1138
1139
1140Port Functions
1141--------------
1142
1143The following sections show functions for configuring ports.
1144
1145.. note::
1146
1147   Port configuration changes only become active when forwarding is started/restarted.
1148
1149port attach
1150~~~~~~~~~~~
1151
1152Attach a port specified by pci address or virtual device args::
1153
1154   testpmd> port attach (identifier)
1155
1156To attach a new pci device, the device should be recognized by kernel first.
1157Then it should be moved under DPDK management.
1158Finally the port can be attached to testpmd.
1159
1160For example, to move a pci device using ixgbe under DPDK management:
1161
1162.. code-block:: console
1163
1164   # Check the status of the available devices.
1165   ./usertools/dpdk-devbind.py --status
1166
1167   Network devices using DPDK-compatible driver
1168   ============================================
1169   <none>
1170
1171   Network devices using kernel driver
1172   ===================================
1173   0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused=
1174
1175
1176   # Bind the device to igb_uio.
1177   sudo ./usertools/dpdk-devbind.py -b igb_uio 0000:0a:00.0
1178
1179
1180   # Recheck the status of the devices.
1181   ./usertools/dpdk-devbind.py --status
1182   Network devices using DPDK-compatible driver
1183   ============================================
1184   0000:0a:00.0 '82599ES 10-Gigabit' drv=igb_uio unused=
1185
1186To attach a port created by virtual device, above steps are not needed.
1187
1188For example, to attach a port whose pci address is 0000:0a:00.0.
1189
1190.. code-block:: console
1191
1192   testpmd> port attach 0000:0a:00.0
1193   Attaching a new port...
1194   EAL: PCI device 0000:0a:00.0 on NUMA socket -1
1195   EAL:   probe driver: 8086:10fb rte_ixgbe_pmd
1196   EAL:   PCI memory mapped at 0x7f83bfa00000
1197   EAL:   PCI memory mapped at 0x7f83bfa80000
1198   PMD: eth_ixgbe_dev_init(): MAC: 2, PHY: 18, SFP+: 5
1199   PMD: eth_ixgbe_dev_init(): port 0 vendorID=0x8086 deviceID=0x10fb
1200   Port 0 is attached. Now total ports is 1
1201   Done
1202
1203For example, to attach a port created by pcap PMD.
1204
1205.. code-block:: console
1206
1207   testpmd> port attach net_pcap0
1208   Attaching a new port...
1209   PMD: Initializing pmd_pcap for net_pcap0
1210   PMD: Creating pcap-backed ethdev on numa socket 0
1211   Port 0 is attached. Now total ports is 1
1212   Done
1213
1214In this case, identifier is ``net_pcap0``.
1215This identifier format is the same as ``--vdev`` format of DPDK applications.
1216
1217For example, to re-attach a bonded port which has been previously detached,
1218the mode and slave parameters must be given.
1219
1220.. code-block:: console
1221
1222   testpmd> port attach net_bond_0,mode=0,slave=1
1223   Attaching a new port...
1224   EAL: Initializing pmd_bond for net_bond_0
1225   EAL: Create bonded device net_bond_0 on port 0 in mode 0 on socket 0.
1226   Port 0 is attached. Now total ports is 1
1227   Done
1228
1229
1230port detach
1231~~~~~~~~~~~
1232
1233Detach a specific port::
1234
1235   testpmd> port detach (port_id)
1236
1237Before detaching a port, the port should be stopped and closed.
1238
1239For example, to detach a pci device port 0.
1240
1241.. code-block:: console
1242
1243   testpmd> port stop 0
1244   Stopping ports...
1245   Done
1246   testpmd> port close 0
1247   Closing ports...
1248   Done
1249
1250   testpmd> port detach 0
1251   Detaching a port...
1252   EAL: PCI device 0000:0a:00.0 on NUMA socket -1
1253   EAL:   remove driver: 8086:10fb rte_ixgbe_pmd
1254   EAL:   PCI memory unmapped at 0x7f83bfa00000
1255   EAL:   PCI memory unmapped at 0x7f83bfa80000
1256   Done
1257
1258
1259For example, to detach a virtual device port 0.
1260
1261.. code-block:: console
1262
1263   testpmd> port stop 0
1264   Stopping ports...
1265   Done
1266   testpmd> port close 0
1267   Closing ports...
1268   Done
1269
1270   testpmd> port detach 0
1271   Detaching a port...
1272   PMD: Closing pcap ethdev on numa socket 0
1273   Port 'net_pcap0' is detached. Now total ports is 0
1274   Done
1275
1276To remove a pci device completely from the system, first detach the port from testpmd.
1277Then the device should be moved under kernel management.
1278Finally the device can be removed using kernel pci hotplug functionality.
1279
1280For example, to move a pci device under kernel management:
1281
1282.. code-block:: console
1283
1284   sudo ./usertools/dpdk-devbind.py -b ixgbe 0000:0a:00.0
1285
1286   ./usertools/dpdk-devbind.py --status
1287
1288   Network devices using DPDK-compatible driver
1289   ============================================
1290   <none>
1291
1292   Network devices using kernel driver
1293   ===================================
1294   0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused=igb_uio
1295
1296To remove a port created by a virtual device, above steps are not needed.
1297
1298port start
1299~~~~~~~~~~
1300
1301Start all ports or a specific port::
1302
1303   testpmd> port start (port_id|all)
1304
1305port stop
1306~~~~~~~~~
1307
1308Stop all ports or a specific port::
1309
1310   testpmd> port stop (port_id|all)
1311
1312port close
1313~~~~~~~~~~
1314
1315Close all ports or a specific port::
1316
1317   testpmd> port close (port_id|all)
1318
1319port start/stop queue
1320~~~~~~~~~~~~~~~~~~~~~
1321
1322Start/stop a rx/tx queue on a specific port::
1323
1324   testpmd> port (port_id) (rxq|txq) (queue_id) (start|stop)
1325
1326Only take effect when port is started.
1327
1328port config - speed
1329~~~~~~~~~~~~~~~~~~~
1330
1331Set the speed and duplex mode for all ports or a specific port::
1332
1333   testpmd> port config (port_id|all) speed (10|100|1000|10000|25000|40000|50000|100000|auto) \
1334            duplex (half|full|auto)
1335
1336port config - queues/descriptors
1337~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1338
1339Set number of queues/descriptors for rxq, txq, rxd and txd::
1340
1341   testpmd> port config all (rxq|txq|rxd|txd) (value)
1342
1343This is equivalent to the ``--rxq``, ``--txq``, ``--rxd`` and ``--txd`` command-line options.
1344
1345port config - max-pkt-len
1346~~~~~~~~~~~~~~~~~~~~~~~~~
1347
1348Set the maximum packet length::
1349
1350   testpmd> port config all max-pkt-len (value)
1351
1352This is equivalent to the ``--max-pkt-len`` command-line option.
1353
1354port config - CRC Strip
1355~~~~~~~~~~~~~~~~~~~~~~~
1356
1357Set hardware CRC stripping on or off for all ports::
1358
1359   testpmd> port config all crc-strip (on|off)
1360
1361CRC stripping is off by default.
1362
1363The ``on`` option is equivalent to the ``--crc-strip`` command-line option.
1364
1365port config - scatter
1366~~~~~~~~~~~~~~~~~~~~~~~
1367
1368Set RX scatter mode on or off for all ports::
1369
1370   testpmd> port config all scatter (on|off)
1371
1372RX scatter mode is off by default.
1373
1374The ``on`` option is equivalent to the ``--enable-scatter`` command-line option.
1375
1376port config - TX queue flags
1377~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1378
1379Set a hexadecimal bitmap of TX queue flags for all ports::
1380
1381   testpmd> port config all txqflags value
1382
1383This command is equivalent to the ``--txqflags`` command-line option.
1384
1385port config - RX Checksum
1386~~~~~~~~~~~~~~~~~~~~~~~~~
1387
1388Set hardware RX checksum offload to on or off for all ports::
1389
1390   testpmd> port config all rx-cksum (on|off)
1391
1392Checksum offload is off by default.
1393
1394The ``on`` option is equivalent to the ``--enable-rx-cksum`` command-line option.
1395
1396port config - VLAN
1397~~~~~~~~~~~~~~~~~~
1398
1399Set hardware VLAN on or off for all ports::
1400
1401   testpmd> port config all hw-vlan (on|off)
1402
1403Hardware VLAN is on by default.
1404
1405The ``off`` option is equivalent to the ``--disable-hw-vlan`` command-line option.
1406
1407port config - VLAN filter
1408~~~~~~~~~~~~~~~~~~~~~~~~~
1409
1410Set hardware VLAN filter on or off for all ports::
1411
1412   testpmd> port config all hw-vlan-filter (on|off)
1413
1414Hardware VLAN filter is on by default.
1415
1416The ``off`` option is equivalent to the ``--disable-hw-vlan-filter`` command-line option.
1417
1418port config - VLAN strip
1419~~~~~~~~~~~~~~~~~~~~~~~~
1420
1421Set hardware VLAN strip on or off for all ports::
1422
1423   testpmd> port config all hw-vlan-strip (on|off)
1424
1425Hardware VLAN strip is on by default.
1426
1427The ``off`` option is equivalent to the ``--disable-hw-vlan-strip`` command-line option.
1428
1429port config - VLAN extend
1430~~~~~~~~~~~~~~~~~~~~~~~~~
1431
1432Set hardware VLAN extend on or off for all ports::
1433
1434   testpmd> port config all hw-vlan-extend (on|off)
1435
1436Hardware VLAN extend is off by default.
1437
1438The ``off`` option is equivalent to the ``--disable-hw-vlan-extend`` command-line option.
1439
1440port config - Drop Packets
1441~~~~~~~~~~~~~~~~~~~~~~~~~~
1442
1443Set packet drop for packets with no descriptors on or off for all ports::
1444
1445   testpmd> port config all drop-en (on|off)
1446
1447Packet dropping for packets with no descriptors is off by default.
1448
1449The ``on`` option is equivalent to the ``--enable-drop-en`` command-line option.
1450
1451port config - RSS
1452~~~~~~~~~~~~~~~~~
1453
1454Set the RSS (Receive Side Scaling) mode on or off::
1455
1456   testpmd> port config all rss (all|ip|tcp|udp|sctp|ether|port|vxlan|geneve|nvgre|none)
1457
1458RSS is on by default.
1459
1460The ``none`` option is equivalent to the ``--disable-rss`` command-line option.
1461
1462port config - RSS Reta
1463~~~~~~~~~~~~~~~~~~~~~~
1464
1465Set the RSS (Receive Side Scaling) redirection table::
1466
1467   testpmd> port config all rss reta (hash,queue)[,(hash,queue)]
1468
1469port config - DCB
1470~~~~~~~~~~~~~~~~~
1471
1472Set the DCB mode for an individual port::
1473
1474   testpmd> port config (port_id) dcb vt (on|off) (traffic_class) pfc (on|off)
1475
1476The traffic class should be 4 or 8.
1477
1478port config - Burst
1479~~~~~~~~~~~~~~~~~~~
1480
1481Set the number of packets per burst::
1482
1483   testpmd> port config all burst (value)
1484
1485This is equivalent to the ``--burst`` command-line option.
1486
1487port config - Threshold
1488~~~~~~~~~~~~~~~~~~~~~~~
1489
1490Set thresholds for TX/RX queues::
1491
1492   testpmd> port config all (threshold) (value)
1493
1494Where the threshold type can be:
1495
1496* ``txpt:`` Set the prefetch threshold register of the TX rings, 0 <= value <= 255.
1497
1498* ``txht:`` Set the host threshold register of the TX rings, 0 <= value <= 255.
1499
1500* ``txwt:`` Set the write-back threshold register of the TX rings, 0 <= value <= 255.
1501
1502* ``rxpt:`` Set the prefetch threshold register of the RX rings, 0 <= value <= 255.
1503
1504* ``rxht:`` Set the host threshold register of the RX rings, 0 <= value <= 255.
1505
1506* ``rxwt:`` Set the write-back threshold register of the RX rings, 0 <= value <= 255.
1507
1508* ``txfreet:`` Set the transmit free threshold of the TX rings, 0 <= value <= txd.
1509
1510* ``rxfreet:`` Set the transmit free threshold of the RX rings, 0 <= value <= rxd.
1511
1512* ``txrst:`` Set the transmit RS bit threshold of TX rings, 0 <= value <= txd.
1513
1514These threshold options are also available from the command-line.
1515
1516port config - E-tag
1517~~~~~~~~~~~~~~~~~~~
1518
1519Set the value of ether-type for E-tag::
1520
1521   testpmd> port config (port_id|all) l2-tunnel E-tag ether-type (value)
1522
1523Enable/disable the E-tag support::
1524
1525   testpmd> port config (port_id|all) l2-tunnel E-tag (enable|disable)
1526
1527
1528Link Bonding Functions
1529----------------------
1530
1531The Link Bonding functions make it possible to dynamically create and
1532manage link bonding devices from within testpmd interactive prompt.
1533
1534create bonded device
1535~~~~~~~~~~~~~~~~~~~~
1536
1537Create a new bonding device::
1538
1539   testpmd> create bonded device (mode) (socket)
1540
1541For example, to create a bonded device in mode 1 on socket 0::
1542
1543   testpmd> create bonded 1 0
1544   created new bonded device (port X)
1545
1546add bonding slave
1547~~~~~~~~~~~~~~~~~
1548
1549Adds Ethernet device to a Link Bonding device::
1550
1551   testpmd> add bonding slave (slave id) (port id)
1552
1553For example, to add Ethernet device (port 6) to a Link Bonding device (port 10)::
1554
1555   testpmd> add bonding slave 6 10
1556
1557
1558remove bonding slave
1559~~~~~~~~~~~~~~~~~~~~
1560
1561Removes an Ethernet slave device from a Link Bonding device::
1562
1563   testpmd> remove bonding slave (slave id) (port id)
1564
1565For example, to remove Ethernet slave device (port 6) to a Link Bonding device (port 10)::
1566
1567   testpmd> remove bonding slave 6 10
1568
1569set bonding mode
1570~~~~~~~~~~~~~~~~
1571
1572Set the Link Bonding mode of a Link Bonding device::
1573
1574   testpmd> set bonding mode (value) (port id)
1575
1576For example, to set the bonding mode of a Link Bonding device (port 10) to broadcast (mode 3)::
1577
1578   testpmd> set bonding mode 3 10
1579
1580set bonding primary
1581~~~~~~~~~~~~~~~~~~~
1582
1583Set an Ethernet slave device as the primary device on a Link Bonding device::
1584
1585   testpmd> set bonding primary (slave id) (port id)
1586
1587For example, to set the Ethernet slave device (port 6) as the primary port of a Link Bonding device (port 10)::
1588
1589   testpmd> set bonding primary 6 10
1590
1591set bonding mac
1592~~~~~~~~~~~~~~~
1593
1594Set the MAC address of a Link Bonding device::
1595
1596   testpmd> set bonding mac (port id) (mac)
1597
1598For example, to set the MAC address of a Link Bonding device (port 10) to 00:00:00:00:00:01::
1599
1600   testpmd> set bonding mac 10 00:00:00:00:00:01
1601
1602set bonding xmit_balance_policy
1603~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1604
1605Set the transmission policy for a Link Bonding device when it is in Balance XOR mode::
1606
1607   testpmd> set bonding xmit_balance_policy (port_id) (l2|l23|l34)
1608
1609For example, set a Link Bonding device (port 10) to use a balance policy of layer 3+4 (IP addresses & UDP ports)::
1610
1611   testpmd> set bonding xmit_balance_policy 10 l34
1612
1613
1614set bonding mon_period
1615~~~~~~~~~~~~~~~~~~~~~~
1616
1617Set the link status monitoring polling period in milliseconds for a bonding device.
1618
1619This adds support for PMD slave devices which do not support link status interrupts.
1620When the mon_period is set to a value greater than 0 then all PMD's which do not support
1621link status ISR will be queried every polling interval to check if their link status has changed::
1622
1623   testpmd> set bonding mon_period (port_id) (value)
1624
1625For example, to set the link status monitoring polling period of bonded device (port 5) to 150ms::
1626
1627   testpmd> set bonding mon_period 5 150
1628
1629
1630show bonding config
1631~~~~~~~~~~~~~~~~~~~
1632
1633Show the current configuration of a Link Bonding device::
1634
1635   testpmd> show bonding config (port id)
1636
1637For example,
1638to show the configuration a Link Bonding device (port 9) with 3 slave devices (1, 3, 4)
1639in balance mode with a transmission policy of layer 2+3::
1640
1641   testpmd> show bonding config 9
1642        Bonding mode: 2
1643        Balance Xmit Policy: BALANCE_XMIT_POLICY_LAYER23
1644        Slaves (3): [1 3 4]
1645        Active Slaves (3): [1 3 4]
1646        Primary: [3]
1647
1648
1649Register Functions
1650------------------
1651
1652The Register Functions can be used to read from and write to registers on the network card referenced by a port number.
1653This is mainly useful for debugging purposes.
1654Reference should be made to the appropriate datasheet for the network card for details on the register addresses
1655and fields that can be accessed.
1656
1657read reg
1658~~~~~~~~
1659
1660Display the value of a port register::
1661
1662   testpmd> read reg (port_id) (address)
1663
1664For example, to examine the Flow Director control register (FDIRCTL, 0x0000EE000) on an Intel 82599 10 GbE Controller::
1665
1666   testpmd> read reg 0 0xEE00
1667   port 0 PCI register at offset 0xEE00: 0x4A060029 (1241907241)
1668
1669read regfield
1670~~~~~~~~~~~~~
1671
1672Display a port register bit field::
1673
1674   testpmd> read regfield (port_id) (address) (bit_x) (bit_y)
1675
1676For example, reading the lowest two bits from the register in the example above::
1677
1678   testpmd> read regfield 0 0xEE00 0 1
1679   port 0 PCI register at offset 0xEE00: bits[0, 1]=0x1 (1)
1680
1681read regbit
1682~~~~~~~~~~~
1683
1684Display a single port register bit::
1685
1686   testpmd> read regbit (port_id) (address) (bit_x)
1687
1688For example, reading the lowest bit from the register in the example above::
1689
1690   testpmd> read regbit 0 0xEE00 0
1691   port 0 PCI register at offset 0xEE00: bit 0=1
1692
1693write reg
1694~~~~~~~~~
1695
1696Set the value of a port register::
1697
1698   testpmd> write reg (port_id) (address) (value)
1699
1700For example, to clear a register::
1701
1702   testpmd> write reg 0 0xEE00 0x0
1703   port 0 PCI register at offset 0xEE00: 0x00000000 (0)
1704
1705write regfield
1706~~~~~~~~~~~~~~
1707
1708Set bit field of a port register::
1709
1710   testpmd> write regfield (port_id) (address) (bit_x) (bit_y) (value)
1711
1712For example, writing to the register cleared in the example above::
1713
1714   testpmd> write regfield 0 0xEE00 0 1 2
1715   port 0 PCI register at offset 0xEE00: 0x00000002 (2)
1716
1717write regbit
1718~~~~~~~~~~~~
1719
1720Set single bit value of a port register::
1721
1722   testpmd> write regbit (port_id) (address) (bit_x) (value)
1723
1724For example, to set the high bit in the register from the example above::
1725
1726   testpmd> write regbit 0 0xEE00 31 1
1727   port 0 PCI register at offset 0xEE00: 0x8000000A (2147483658)
1728
1729
1730Filter Functions
1731----------------
1732
1733This section details the available filter functions that are available.
1734
1735Note these functions interface the deprecated legacy filtering framework,
1736superseded by *rte_flow*. See `Flow rules management`_.
1737
1738ethertype_filter
1739~~~~~~~~~~~~~~~~~~~~
1740
1741Add or delete a L2 Ethertype filter, which identify packets by their L2 Ethertype mainly assign them to a receive queue::
1742
1743   ethertype_filter (port_id) (add|del) (mac_addr|mac_ignr) (mac_address) \
1744                    ethertype (ether_type) (drop|fwd) queue (queue_id)
1745
1746The available information parameters are:
1747
1748* ``port_id``: The port which the Ethertype filter assigned on.
1749
1750* ``mac_addr``: Compare destination mac address.
1751
1752* ``mac_ignr``: Ignore destination mac address match.
1753
1754* ``mac_address``: Destination mac address to match.
1755
1756* ``ether_type``: The EtherType value want to match,
1757  for example 0x0806 for ARP packet. 0x0800 (IPv4) and 0x86DD (IPv6) are invalid.
1758
1759* ``queue_id``: The receive queue associated with this EtherType filter.
1760  It is meaningless when deleting or dropping.
1761
1762Example, to add/remove an ethertype filter rule::
1763
1764   testpmd> ethertype_filter 0 add mac_ignr 00:11:22:33:44:55 \
1765                             ethertype 0x0806 fwd queue 3
1766
1767   testpmd> ethertype_filter 0 del mac_ignr 00:11:22:33:44:55 \
1768                             ethertype 0x0806 fwd queue 3
1769
17702tuple_filter
1771~~~~~~~~~~~~~~~~~
1772
1773Add or delete a 2-tuple filter,
1774which identifies packets by specific protocol and destination TCP/UDP port
1775and forwards packets into one of the receive queues::
1776
1777   2tuple_filter (port_id) (add|del) dst_port (dst_port_value) \
1778                 protocol (protocol_value) mask (mask_value) \
1779                 tcp_flags (tcp_flags_value) priority (prio_value) \
1780                 queue (queue_id)
1781
1782The available information parameters are:
1783
1784* ``port_id``: The port which the 2-tuple filter assigned on.
1785
1786* ``dst_port_value``: Destination port in L4.
1787
1788* ``protocol_value``: IP L4 protocol.
1789
1790* ``mask_value``: Participates in the match or not by bit for field above, 1b means participate.
1791
1792* ``tcp_flags_value``: TCP control bits. The non-zero value is invalid, when the pro_value is not set to 0x06 (TCP).
1793
1794* ``prio_value``: Priority of this filter.
1795
1796* ``queue_id``: The receive queue associated with this 2-tuple filter.
1797
1798Example, to add/remove an 2tuple filter rule::
1799
1800   testpmd> 2tuple_filter 0 add dst_port 32 protocol 0x06 mask 0x03 \
1801                          tcp_flags 0x02 priority 3 queue 3
1802
1803   testpmd> 2tuple_filter 0 del dst_port 32 protocol 0x06 mask 0x03 \
1804                          tcp_flags 0x02 priority 3 queue 3
1805
18065tuple_filter
1807~~~~~~~~~~~~~~~~~
1808
1809Add or delete a 5-tuple filter,
1810which consists of a 5-tuple (protocol, source and destination IP addresses, source and destination TCP/UDP/SCTP port)
1811and routes packets into one of the receive queues::
1812
1813   5tuple_filter (port_id) (add|del) dst_ip (dst_address) src_ip \
1814                 (src_address) dst_port (dst_port_value) \
1815                 src_port (src_port_value) protocol (protocol_value) \
1816                 mask (mask_value) tcp_flags (tcp_flags_value) \
1817                 priority (prio_value) queue (queue_id)
1818
1819The available information parameters are:
1820
1821* ``port_id``: The port which the 5-tuple filter assigned on.
1822
1823* ``dst_address``: Destination IP address.
1824
1825* ``src_address``: Source IP address.
1826
1827* ``dst_port_value``: TCP/UDP destination port.
1828
1829* ``src_port_value``: TCP/UDP source port.
1830
1831* ``protocol_value``: L4 protocol.
1832
1833* ``mask_value``: Participates in the match or not by bit for field above, 1b means participate
1834
1835* ``tcp_flags_value``: TCP control bits. The non-zero value is invalid, when the protocol_value is not set to 0x06 (TCP).
1836
1837* ``prio_value``: The priority of this filter.
1838
1839* ``queue_id``: The receive queue associated with this 5-tuple filter.
1840
1841Example, to add/remove an 5tuple filter rule::
1842
1843   testpmd> 5tuple_filter 0 add dst_ip 2.2.2.5 src_ip 2.2.2.4 \
1844            dst_port 64 src_port 32 protocol 0x06 mask 0x1F \
1845            flags 0x0 priority 3 queue 3
1846
1847   testpmd> 5tuple_filter 0 del dst_ip 2.2.2.5 src_ip 2.2.2.4 \
1848            dst_port 64 src_port 32 protocol 0x06 mask 0x1F \
1849            flags 0x0 priority 3 queue 3
1850
1851syn_filter
1852~~~~~~~~~~
1853
1854Using the  SYN filter, TCP packets whose *SYN* flag is set can be forwarded to a separate queue::
1855
1856   syn_filter (port_id) (add|del) priority (high|low) queue (queue_id)
1857
1858The available information parameters are:
1859
1860* ``port_id``: The port which the SYN filter assigned on.
1861
1862* ``high``: This SYN filter has higher priority than other filters.
1863
1864* ``low``: This SYN filter has lower priority than other filters.
1865
1866* ``queue_id``: The receive queue associated with this SYN filter
1867
1868Example::
1869
1870   testpmd> syn_filter 0 add priority high queue 3
1871
1872flex_filter
1873~~~~~~~~~~~
1874
1875With flex filter, packets can be recognized by any arbitrary pattern within the first 128 bytes of the packet
1876and routed into one of the receive queues::
1877
1878   flex_filter (port_id) (add|del) len (len_value) bytes (bytes_value) \
1879               mask (mask_value) priority (prio_value) queue (queue_id)
1880
1881The available information parameters are:
1882
1883* ``port_id``: The port which the Flex filter is assigned on.
1884
1885* ``len_value``: Filter length in bytes, no greater than 128.
1886
1887* ``bytes_value``: A string in hexadecimal, means the value the flex filter needs to match.
1888
1889* ``mask_value``: A string in hexadecimal, bit 1 means corresponding byte participates in the match.
1890
1891* ``prio_value``: The priority of this filter.
1892
1893* ``queue_id``: The receive queue associated with this Flex filter.
1894
1895Example::
1896
1897   testpmd> flex_filter 0 add len 16 bytes 0x00000000000000000000000008060000 \
1898                          mask 000C priority 3 queue 3
1899
1900   testpmd> flex_filter 0 del len 16 bytes 0x00000000000000000000000008060000 \
1901                          mask 000C priority 3 queue 3
1902
1903
1904.. _testpmd_flow_director:
1905
1906flow_director_filter
1907~~~~~~~~~~~~~~~~~~~~
1908
1909The Flow Director works in receive mode to identify specific flows or sets of flows and route them to specific queues.
1910
1911Four types of filtering are supported which are referred to as Perfect Match, Signature, Perfect-mac-vlan and
1912Perfect-tunnel filters, the match mode is set by the ``--pkt-filter-mode`` command-line parameter:
1913
1914* Perfect match filters.
1915  The hardware checks a match between the masked fields of the received packets and the programmed filters.
1916  The masked fields are for IP flow.
1917
1918* Signature filters.
1919  The hardware checks a match between a hash-based signature of the masked fields of the received packet.
1920
1921* Perfect-mac-vlan match filters.
1922  The hardware checks a match between the masked fields of the received packets and the programmed filters.
1923  The masked fields are for MAC VLAN flow.
1924
1925* Perfect-tunnel match filters.
1926  The hardware checks a match between the masked fields of the received packets and the programmed filters.
1927  The masked fields are for tunnel flow.
1928
1929The Flow Director filters can match the different fields for different type of packet: flow type, specific input set
1930per flow type and the flexible payload.
1931
1932The Flow Director can also mask out parts of all of these fields so that filters
1933are only applied to certain fields or parts of the fields.
1934
1935Different NICs may have different capabilities, command show port fdir (port_id) can be used to acquire the information.
1936
1937# Commands to add flow director filters of different flow types::
1938
1939   flow_director_filter (port_id) mode IP (add|del|update) \
1940                        flow (ipv4-other|ipv4-frag|ipv6-other|ipv6-frag) \
1941                        src (src_ip_address) dst (dst_ip_address) \
1942                        tos (tos_value) proto (proto_value) ttl (ttl_value) \
1943                        vlan (vlan_value) flexbytes (flexbytes_value) \
1944                        (drop|fwd) pf|vf(vf_id) queue (queue_id) \
1945                        fd_id (fd_id_value)
1946
1947   flow_director_filter (port_id) mode IP (add|del|update) \
1948                        flow (ipv4-tcp|ipv4-udp|ipv6-tcp|ipv6-udp) \
1949                        src (src_ip_address) (src_port) \
1950                        dst (dst_ip_address) (dst_port) \
1951                        tos (tos_value) ttl (ttl_value) \
1952                        vlan (vlan_value) flexbytes (flexbytes_value) \
1953                        (drop|fwd) queue pf|vf(vf_id) (queue_id) \
1954                        fd_id (fd_id_value)
1955
1956   flow_director_filter (port_id) mode IP (add|del|update) \
1957                        flow (ipv4-sctp|ipv6-sctp) \
1958                        src (src_ip_address) (src_port) \
1959                        dst (dst_ip_address) (dst_port) \
1960                        tos (tos_value) ttl (ttl_value) \
1961                        tag (verification_tag) vlan (vlan_value) \
1962                        flexbytes (flexbytes_value) (drop|fwd) \
1963                        pf|vf(vf_id) queue (queue_id) fd_id (fd_id_value)
1964
1965   flow_director_filter (port_id) mode IP (add|del|update) flow l2_payload \
1966                        ether (ethertype) flexbytes (flexbytes_value) \
1967                        (drop|fwd) pf|vf(vf_id) queue (queue_id)
1968                        fd_id (fd_id_value)
1969
1970   flow_director_filter (port_id) mode MAC-VLAN (add|del|update) \
1971                        mac (mac_address) vlan (vlan_value) \
1972                        flexbytes (flexbytes_value) (drop|fwd) \
1973                        queue (queue_id) fd_id (fd_id_value)
1974
1975   flow_director_filter (port_id) mode Tunnel (add|del|update) \
1976                        mac (mac_address) vlan (vlan_value) \
1977                        tunnel (NVGRE|VxLAN) tunnel-id (tunnel_id_value) \
1978                        flexbytes (flexbytes_value) (drop|fwd) \
1979                        queue (queue_id) fd_id (fd_id_value)
1980
1981For example, to add an ipv4-udp flow type filter::
1982
1983   testpmd> flow_director_filter 0 mode IP add flow ipv4-udp src 2.2.2.3 32 \
1984            dst 2.2.2.5 33 tos 2 ttl 40 vlan 0x1 flexbytes (0x88,0x48) \
1985            fwd pf queue 1 fd_id 1
1986
1987For example, add an ipv4-other flow type filter::
1988
1989   testpmd> flow_director_filter 0 mode IP add flow ipv4-other src 2.2.2.3 \
1990             dst 2.2.2.5 tos 2 proto 20 ttl 40 vlan 0x1 \
1991             flexbytes (0x88,0x48) fwd pf queue 1 fd_id 1
1992
1993flush_flow_director
1994~~~~~~~~~~~~~~~~~~~
1995
1996Flush all flow director filters on a device::
1997
1998   testpmd> flush_flow_director (port_id)
1999
2000Example, to flush all flow director filter on port 0::
2001
2002   testpmd> flush_flow_director 0
2003
2004flow_director_mask
2005~~~~~~~~~~~~~~~~~~
2006
2007Set flow director's input masks::
2008
2009   flow_director_mask (port_id) mode IP vlan (vlan_value) \
2010                      src_mask (ipv4_src) (ipv6_src) (src_port) \
2011                      dst_mask (ipv4_dst) (ipv6_dst) (dst_port)
2012
2013   flow_director_mask (port_id) mode MAC-VLAN vlan (vlan_value)
2014
2015   flow_director_mask (port_id) mode Tunnel vlan (vlan_value) \
2016                      mac (mac_value) tunnel-type (tunnel_type_value) \
2017                      tunnel-id (tunnel_id_value)
2018
2019Example, to set flow director mask on port 0::
2020
2021   testpmd> flow_director_mask 0 mode IP vlan 0xefff \
2022            src_mask 255.255.255.255 \
2023                FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF \
2024            dst_mask 255.255.255.255 \
2025                FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF
2026
2027flow_director_flex_mask
2028~~~~~~~~~~~~~~~~~~~~~~~
2029
2030set masks of flow director's flexible payload based on certain flow type::
2031
2032   testpmd> flow_director_flex_mask (port_id) \
2033            flow (none|ipv4-other|ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
2034                  ipv6-other|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp| \
2035                  l2_payload|all) (mask)
2036
2037Example, to set flow director's flex mask for all flow type on port 0::
2038
2039   testpmd> flow_director_flex_mask 0 flow all \
2040            (0xff,0xff,0,0,0,0,0,0,0,0,0,0,0,0,0,0)
2041
2042
2043flow_director_flex_payload
2044~~~~~~~~~~~~~~~~~~~~~~~~~~
2045
2046Configure flexible payload selection::
2047
2048   flow_director_flex_payload (port_id) (raw|l2|l3|l4) (config)
2049
2050For example, to select the first 16 bytes from the offset 4 (bytes) of packet's payload as flexible payload::
2051
2052   testpmd> flow_director_flex_payload 0 l4 \
2053            (4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19)
2054
2055get_sym_hash_ena_per_port
2056~~~~~~~~~~~~~~~~~~~~~~~~~
2057
2058Get symmetric hash enable configuration per port::
2059
2060   get_sym_hash_ena_per_port (port_id)
2061
2062For example, to get symmetric hash enable configuration of port 1::
2063
2064   testpmd> get_sym_hash_ena_per_port 1
2065
2066set_sym_hash_ena_per_port
2067~~~~~~~~~~~~~~~~~~~~~~~~~
2068
2069Set symmetric hash enable configuration per port to enable or disable::
2070
2071   set_sym_hash_ena_per_port (port_id) (enable|disable)
2072
2073For example, to set symmetric hash enable configuration of port 1 to enable::
2074
2075   testpmd> set_sym_hash_ena_per_port 1 enable
2076
2077get_hash_global_config
2078~~~~~~~~~~~~~~~~~~~~~~
2079
2080Get the global configurations of hash filters::
2081
2082   get_hash_global_config (port_id)
2083
2084For example, to get the global configurations of hash filters of port 1::
2085
2086   testpmd> get_hash_global_config 1
2087
2088set_hash_global_config
2089~~~~~~~~~~~~~~~~~~~~~~
2090
2091Set the global configurations of hash filters::
2092
2093   set_hash_global_config (port_id) (toeplitz|simple_xor|default) \
2094   (ipv4|ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp|ipv4-other|ipv6|ipv6-frag| \
2095   ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other|l2_payload) \
2096   (enable|disable)
2097
2098For example, to enable simple_xor for flow type of ipv6 on port 2::
2099
2100   testpmd> set_hash_global_config 2 simple_xor ipv6 enable
2101
2102set_hash_input_set
2103~~~~~~~~~~~~~~~~~~
2104
2105Set the input set for hash::
2106
2107   set_hash_input_set (port_id) (ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
2108   ipv4-other|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other| \
2109   l2_payload) (ovlan|ivlan|src-ipv4|dst-ipv4|src-ipv6|dst-ipv6|ipv4-tos| \
2110   ipv4-proto|ipv6-tc|ipv6-next-header|udp-src-port|udp-dst-port| \
2111   tcp-src-port|tcp-dst-port|sctp-src-port|sctp-dst-port|sctp-veri-tag| \
2112   udp-key|gre-key|fld-1st|fld-2nd|fld-3rd|fld-4th|fld-5th|fld-6th|fld-7th| \
2113   fld-8th|none) (select|add)
2114
2115For example, to add source IP to hash input set for flow type of ipv4-udp on port 0::
2116
2117   testpmd> set_hash_input_set 0 ipv4-udp src-ipv4 add
2118
2119set_fdir_input_set
2120~~~~~~~~~~~~~~~~~~
2121
2122The Flow Director filters can match the different fields for different type of packet, i.e. specific input set
2123on per flow type and the flexible payload. This command can be used to change input set for each flow type.
2124
2125Set the input set for flow director::
2126
2127   set_fdir_input_set (port_id) (ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
2128   ipv4-other|ipv6|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other| \
2129   l2_payload) (ivlan|ethertype|src-ipv4|dst-ipv4|src-ipv6|dst-ipv6|ipv4-tos| \
2130   ipv4-proto|ipv4-ttl|ipv6-tc|ipv6-next-header|ipv6-hop-limits| \
2131   tudp-src-port|udp-dst-port|cp-src-port|tcp-dst-port|sctp-src-port| \
2132   sctp-dst-port|sctp-veri-tag|none) (select|add)
2133
2134For example to add source IP to FD input set for flow type of ipv4-udp on port 0::
2135
2136   testpmd> set_fdir_input_set 0 ipv4-udp src-ipv4 add
2137
2138global_config
2139~~~~~~~~~~~~~
2140
2141Set different GRE key length for input set::
2142
2143   global_config (port_id) gre-key-len (number in bytes)
2144
2145For example to set GRE key length for input set to 4 bytes on port 0::
2146
2147   testpmd> global_config 0 gre-key-len 4
2148
2149
2150.. _testpmd_rte_flow:
2151
2152Flow rules management
2153---------------------
2154
2155Control of the generic flow API (*rte_flow*) is fully exposed through the
2156``flow`` command (validation, creation, destruction and queries).
2157
2158Considering *rte_flow* overlaps with all `Filter Functions`_, using both
2159features simultaneously may cause undefined side-effects and is therefore
2160not recommended.
2161
2162``flow`` syntax
2163~~~~~~~~~~~~~~~
2164
2165Because the ``flow`` command uses dynamic tokens to handle the large number
2166of possible flow rules combinations, its behavior differs slightly from
2167other commands, in particular:
2168
2169- Pressing *?* or the *<tab>* key displays contextual help for the current
2170  token, not that of the entire command.
2171
2172- Optional and repeated parameters are supported (provided they are listed
2173  in the contextual help).
2174
2175The first parameter stands for the operation mode. Possible operations and
2176their general syntax are described below. They are covered in detail in the
2177following sections.
2178
2179- Check whether a flow rule can be created::
2180
2181   flow validate {port_id}
2182       [group {group_id}] [priority {level}] [ingress] [egress]
2183       pattern {item} [/ {item} [...]] / end
2184       actions {action} [/ {action} [...]] / end
2185
2186- Create a flow rule::
2187
2188   flow create {port_id}
2189       [group {group_id}] [priority {level}] [ingress] [egress]
2190       pattern {item} [/ {item} [...]] / end
2191       actions {action} [/ {action} [...]] / end
2192
2193- Destroy specific flow rules::
2194
2195   flow destroy {port_id} rule {rule_id} [...]
2196
2197- Destroy all flow rules::
2198
2199   flow flush {port_id}
2200
2201- Query an existing flow rule::
2202
2203   flow query {port_id} {rule_id} {action}
2204
2205- List existing flow rules sorted by priority, filtered by group
2206  identifiers::
2207
2208   flow list {port_id} [group {group_id}] [...]
2209
2210Validating flow rules
2211~~~~~~~~~~~~~~~~~~~~~
2212
2213``flow validate`` reports whether a flow rule would be accepted by the
2214underlying device in its current state but stops short of creating it. It is
2215bound to ``rte_flow_validate()``::
2216
2217   flow validate {port_id}
2218      [group {group_id}] [priority {level}] [ingress] [egress]
2219      pattern {item} [/ {item} [...]] / end
2220      actions {action} [/ {action} [...]] / end
2221
2222If successful, it will show::
2223
2224   Flow rule validated
2225
2226Otherwise it will show an error message of the form::
2227
2228   Caught error type [...] ([...]): [...]
2229
2230This command uses the same parameters as ``flow create``, their format is
2231described in `Creating flow rules`_.
2232
2233Check whether redirecting any Ethernet packet received on port 0 to RX queue
2234index 6 is supported::
2235
2236   testpmd> flow validate 0 ingress pattern eth / end
2237      actions queue index 6 / end
2238   Flow rule validated
2239   testpmd>
2240
2241Port 0 does not support TCPv6 rules::
2242
2243   testpmd> flow validate 0 ingress pattern eth / ipv6 / tcp / end
2244      actions drop / end
2245   Caught error type 9 (specific pattern item): Invalid argument
2246   testpmd>
2247
2248Creating flow rules
2249~~~~~~~~~~~~~~~~~~~
2250
2251``flow create`` validates and creates the specified flow rule. It is bound
2252to ``rte_flow_create()``::
2253
2254   flow create {port_id}
2255      [group {group_id}] [priority {level}] [ingress] [egress]
2256      pattern {item} [/ {item} [...]] / end
2257      actions {action} [/ {action} [...]] / end
2258
2259If successful, it will return a flow rule ID usable with other commands::
2260
2261   Flow rule #[...] created
2262
2263Otherwise it will show an error message of the form::
2264
2265   Caught error type [...] ([...]): [...]
2266
2267Parameters describe in the following order:
2268
2269- Attributes (*group*, *priority*, *ingress*, *egress* tokens).
2270- A matching pattern, starting with the *pattern* token and terminated by an
2271  *end* pattern item.
2272- Actions, starting with the *actions* token and terminated by an *end*
2273  action.
2274
2275These translate directly to *rte_flow* objects provided as-is to the
2276underlying functions.
2277
2278The shortest valid definition only comprises mandatory tokens::
2279
2280   testpmd> flow create 0 pattern end actions end
2281
2282Note that PMDs may refuse rules that essentially do nothing such as this
2283one.
2284
2285**All unspecified object values are automatically initialized to 0.**
2286
2287Attributes
2288^^^^^^^^^^
2289
2290These tokens affect flow rule attributes (``struct rte_flow_attr``) and are
2291specified before the ``pattern`` token.
2292
2293- ``group {group id}``: priority group.
2294- ``priority {level}``: priority level within group.
2295- ``ingress``: rule applies to ingress traffic.
2296- ``egress``: rule applies to egress traffic.
2297
2298Each instance of an attribute specified several times overrides the previous
2299value as shown below (group 4 is used)::
2300
2301   testpmd> flow create 0 group 42 group 24 group 4 [...]
2302
2303Note that once enabled, ``ingress`` and ``egress`` cannot be disabled.
2304
2305While not specifying a direction is an error, some rules may allow both
2306simultaneously.
2307
2308Most rules affect RX therefore contain the ``ingress`` token::
2309
2310   testpmd> flow create 0 ingress pattern [...]
2311
2312Matching pattern
2313^^^^^^^^^^^^^^^^
2314
2315A matching pattern starts after the ``pattern`` token. It is made of pattern
2316items and is terminated by a mandatory ``end`` item.
2317
2318Items are named after their type (*RTE_FLOW_ITEM_TYPE_* from ``enum
2319rte_flow_item_type``).
2320
2321The ``/`` token is used as a separator between pattern items as shown
2322below::
2323
2324   testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end [...]
2325
2326Note that protocol items like these must be stacked from lowest to highest
2327layer to make sense. For instance, the following rule is either invalid or
2328unlikely to match any packet::
2329
2330   testpmd> flow create 0 ingress pattern eth / udp / ipv4 / end [...]
2331
2332More information on these restrictions can be found in the *rte_flow*
2333documentation.
2334
2335Several items support additional specification structures, for example
2336``ipv4`` allows specifying source and destination addresses as follows::
2337
2338   testpmd> flow create 0 ingress pattern eth / ipv4 src is 10.1.1.1
2339      dst is 10.2.0.0 / end [...]
2340
2341This rule matches all IPv4 traffic with the specified properties.
2342
2343In this example, ``src`` and ``dst`` are field names of the underlying
2344``struct rte_flow_item_ipv4`` object. All item properties can be specified
2345in a similar fashion.
2346
2347The ``is`` token means that the subsequent value must be matched exactly,
2348and assigns ``spec`` and ``mask`` fields in ``struct rte_flow_item``
2349accordingly. Possible assignment tokens are:
2350
2351- ``is``: match value perfectly (with full bit-mask).
2352- ``spec``: match value according to configured bit-mask.
2353- ``last``: specify upper bound to establish a range.
2354- ``mask``: specify bit-mask with relevant bits set to one.
2355- ``prefix``: generate bit-mask from a prefix length.
2356
2357These yield identical results::
2358
2359   ipv4 src is 10.1.1.1
2360
2361::
2362
2363   ipv4 src spec 10.1.1.1 src mask 255.255.255.255
2364
2365::
2366
2367   ipv4 src spec 10.1.1.1 src prefix 32
2368
2369::
2370
2371   ipv4 src is 10.1.1.1 src last 10.1.1.1 # range with a single value
2372
2373::
2374
2375   ipv4 src is 10.1.1.1 src last 0 # 0 disables range
2376
2377Inclusive ranges can be defined with ``last``::
2378
2379   ipv4 src is 10.1.1.1 src last 10.2.3.4 # 10.1.1.1 to 10.2.3.4
2380
2381Note that ``mask`` affects both ``spec`` and ``last``::
2382
2383   ipv4 src is 10.1.1.1 src last 10.2.3.4 src mask 255.255.0.0
2384      # matches 10.1.0.0 to 10.2.255.255
2385
2386Properties can be modified multiple times::
2387
2388   ipv4 src is 10.1.1.1 src is 10.1.2.3 src is 10.2.3.4 # matches 10.2.3.4
2389
2390::
2391
2392   ipv4 src is 10.1.1.1 src prefix 24 src prefix 16 # matches 10.1.0.0/16
2393
2394Pattern items
2395^^^^^^^^^^^^^
2396
2397This section lists supported pattern items and their attributes, if any.
2398
2399- ``end``: end list of pattern items.
2400
2401- ``void``: no-op pattern item.
2402
2403- ``invert``: perform actions when pattern does not match.
2404
2405- ``any``: match any protocol for the current layer.
2406
2407  - ``num {unsigned}``: number of layers covered.
2408
2409- ``pf``: match packets addressed to the physical function.
2410
2411- ``vf``: match packets addressed to a virtual function ID.
2412
2413  - ``id {unsigned}``: destination VF ID.
2414
2415- ``port``: device-specific physical port index to use.
2416
2417  - ``index {unsigned}``: physical port index.
2418
2419- ``raw``: match an arbitrary byte string.
2420
2421  - ``relative {boolean}``: look for pattern after the previous item.
2422  - ``search {boolean}``: search pattern from offset (see also limit).
2423  - ``offset {integer}``: absolute or relative offset for pattern.
2424  - ``limit {unsigned}``: search area limit for start of pattern.
2425  - ``pattern {string}``: byte string to look for.
2426
2427- ``eth``: match Ethernet header.
2428
2429  - ``dst {MAC-48}``: destination MAC.
2430  - ``src {MAC-48}``: source MAC.
2431  - ``type {unsigned}``: EtherType.
2432
2433- ``vlan``: match 802.1Q/ad VLAN tag.
2434
2435  - ``tpid {unsigned}``: tag protocol identifier.
2436  - ``tci {unsigned}``: tag control information.
2437  - ``pcp {unsigned}``: priority code point.
2438  - ``dei {unsigned}``: drop eligible indicator.
2439  - ``vid {unsigned}``: VLAN identifier.
2440
2441- ``ipv4``: match IPv4 header.
2442
2443  - ``tos {unsigned}``: type of service.
2444  - ``ttl {unsigned}``: time to live.
2445  - ``proto {unsigned}``: next protocol ID.
2446  - ``src {ipv4 address}``: source address.
2447  - ``dst {ipv4 address}``: destination address.
2448
2449- ``ipv6``: match IPv6 header.
2450
2451  - ``tc {unsigned}``: traffic class.
2452  - ``flow {unsigned}``: flow label.
2453  - ``proto {unsigned}``: protocol (next header).
2454  - ``hop {unsigned}``: hop limit.
2455  - ``src {ipv6 address}``: source address.
2456  - ``dst {ipv6 address}``: destination address.
2457
2458- ``icmp``: match ICMP header.
2459
2460  - ``type {unsigned}``: ICMP packet type.
2461  - ``code {unsigned}``: ICMP packet code.
2462
2463- ``udp``: match UDP header.
2464
2465  - ``src {unsigned}``: UDP source port.
2466  - ``dst {unsigned}``: UDP destination port.
2467
2468- ``tcp``: match TCP header.
2469
2470  - ``src {unsigned}``: TCP source port.
2471  - ``dst {unsigned}``: TCP destination port.
2472
2473- ``sctp``: match SCTP header.
2474
2475  - ``src {unsigned}``: SCTP source port.
2476  - ``dst {unsigned}``: SCTP destination port.
2477  - ``tag {unsigned}``: validation tag.
2478  - ``cksum {unsigned}``: checksum.
2479
2480- ``vxlan``: match VXLAN header.
2481
2482  - ``vni {unsigned}``: VXLAN identifier.
2483
2484- ``mpls``: match MPLS header.
2485
2486  - ``label {unsigned}``: MPLS label.
2487
2488- ``gre``: match GRE header.
2489
2490  - ``protocol {unsigned}``: protocol type.
2491
2492Actions list
2493^^^^^^^^^^^^
2494
2495A list of actions starts after the ``actions`` token in the same fashion as
2496`Matching pattern`_; actions are separated by ``/`` tokens and the list is
2497terminated by a mandatory ``end`` action.
2498
2499Actions are named after their type (*RTE_FLOW_ACTION_TYPE_* from ``enum
2500rte_flow_action_type``).
2501
2502Dropping all incoming UDPv4 packets can be expressed as follows::
2503
2504   testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
2505      actions drop / end
2506
2507Several actions have configurable properties which must be specified when
2508there is no valid default value. For example, ``queue`` requires a target
2509queue index.
2510
2511This rule redirects incoming UDPv4 traffic to queue index 6::
2512
2513   testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
2514      actions queue index 6 / end
2515
2516While this one could be rejected by PMDs (unspecified queue index)::
2517
2518   testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
2519      actions queue / end
2520
2521As defined by *rte_flow*, the list is not ordered, all actions of a given
2522rule are performed simultaneously. These are equivalent::
2523
2524   queue index 6 / void / mark id 42 / end
2525
2526::
2527
2528   void / mark id 42 / queue index 6 / end
2529
2530All actions in a list should have different types, otherwise only the last
2531action of a given type is taken into account::
2532
2533   queue index 4 / queue index 5 / queue index 6 / end # will use queue 6
2534
2535::
2536
2537   drop / drop / drop / end # drop is performed only once
2538
2539::
2540
2541   mark id 42 / queue index 3 / mark id 24 / end # mark will be 24
2542
2543Considering they are performed simultaneously, opposite and overlapping
2544actions can sometimes be combined when the end result is unambiguous::
2545
2546   drop / queue index 6 / end # drop has no effect
2547
2548::
2549
2550   drop / dup index 6 / end # same as above
2551
2552::
2553
2554   queue index 6 / rss queues 6 7 8 / end # queue has no effect
2555
2556::
2557
2558   drop / passthru / end # drop has no effect
2559
2560Note that PMDs may still refuse such combinations.
2561
2562Actions
2563^^^^^^^
2564
2565This section lists supported actions and their attributes, if any.
2566
2567- ``end``: end list of actions.
2568
2569- ``void``: no-op action.
2570
2571- ``passthru``: let subsequent rule process matched packets.
2572
2573- ``mark``: attach 32 bit value to packets.
2574
2575  - ``id {unsigned}``: 32 bit value to return with packets.
2576
2577- ``flag``: flag packets.
2578
2579- ``queue``: assign packets to a given queue index.
2580
2581  - ``index {unsigned}``: queue index to use.
2582
2583- ``drop``: drop packets (note: passthru has priority).
2584
2585- ``count``: enable counters for this rule.
2586
2587- ``dup``: duplicate packets to a given queue index.
2588
2589  - ``index {unsigned}``: queue index to duplicate packets to.
2590
2591- ``rss``: spread packets among several queues.
2592
2593  - ``queues [{unsigned} [...]] end``: queue indices to use.
2594
2595- ``pf``: redirect packets to physical device function.
2596
2597- ``vf``: redirect packets to virtual device function.
2598
2599  - ``original {boolean}``: use original VF ID if possible.
2600  - ``id {unsigned}``: VF ID to redirect packets to.
2601
2602Destroying flow rules
2603~~~~~~~~~~~~~~~~~~~~~
2604
2605``flow destroy`` destroys one or more rules from their rule ID (as returned
2606by ``flow create``), this command calls ``rte_flow_destroy()`` as many
2607times as necessary::
2608
2609   flow destroy {port_id} rule {rule_id} [...]
2610
2611If successful, it will show::
2612
2613   Flow rule #[...] destroyed
2614
2615It does not report anything for rule IDs that do not exist. The usual error
2616message is shown when a rule cannot be destroyed::
2617
2618   Caught error type [...] ([...]): [...]
2619
2620``flow flush`` destroys all rules on a device and does not take extra
2621arguments. It is bound to ``rte_flow_flush()``::
2622
2623   flow flush {port_id}
2624
2625Any errors are reported as above.
2626
2627Creating several rules and destroying them::
2628
2629   testpmd> flow create 0 ingress pattern eth / ipv6 / end
2630      actions queue index 2 / end
2631   Flow rule #0 created
2632   testpmd> flow create 0 ingress pattern eth / ipv4 / end
2633      actions queue index 3 / end
2634   Flow rule #1 created
2635   testpmd> flow destroy 0 rule 0 rule 1
2636   Flow rule #1 destroyed
2637   Flow rule #0 destroyed
2638   testpmd>
2639
2640The same result can be achieved using ``flow flush``::
2641
2642   testpmd> flow create 0 ingress pattern eth / ipv6 / end
2643      actions queue index 2 / end
2644   Flow rule #0 created
2645   testpmd> flow create 0 ingress pattern eth / ipv4 / end
2646      actions queue index 3 / end
2647   Flow rule #1 created
2648   testpmd> flow flush 0
2649   testpmd>
2650
2651Non-existent rule IDs are ignored::
2652
2653   testpmd> flow create 0 ingress pattern eth / ipv6 / end
2654      actions queue index 2 / end
2655   Flow rule #0 created
2656   testpmd> flow create 0 ingress pattern eth / ipv4 / end
2657      actions queue index 3 / end
2658   Flow rule #1 created
2659   testpmd> flow destroy 0 rule 42 rule 10 rule 2
2660   testpmd>
2661   testpmd> flow destroy 0 rule 0
2662   Flow rule #0 destroyed
2663   testpmd>
2664
2665Querying flow rules
2666~~~~~~~~~~~~~~~~~~~
2667
2668``flow query`` queries a specific action of a flow rule having that
2669ability. Such actions collect information that can be reported using this
2670command. It is bound to ``rte_flow_query()``::
2671
2672   flow query {port_id} {rule_id} {action}
2673
2674If successful, it will display either the retrieved data for known actions
2675or the following message::
2676
2677   Cannot display result for action type [...] ([...])
2678
2679Otherwise, it will complain either that the rule does not exist or that some
2680error occurred::
2681
2682   Flow rule #[...] not found
2683
2684::
2685
2686   Caught error type [...] ([...]): [...]
2687
2688Currently only the ``count`` action is supported. This action reports the
2689number of packets that hit the flow rule and the total number of bytes. Its
2690output has the following format::
2691
2692   count:
2693    hits_set: [...] # whether "hits" contains a valid value
2694    bytes_set: [...] # whether "bytes" contains a valid value
2695    hits: [...] # number of packets
2696    bytes: [...] # number of bytes
2697
2698Querying counters for TCPv6 packets redirected to queue 6::
2699
2700   testpmd> flow create 0 ingress pattern eth / ipv6 / tcp / end
2701      actions queue index 6 / count / end
2702   Flow rule #4 created
2703   testpmd> flow query 0 4 count
2704   count:
2705    hits_set: 1
2706    bytes_set: 0
2707    hits: 386446
2708    bytes: 0
2709   testpmd>
2710
2711Listing flow rules
2712~~~~~~~~~~~~~~~~~~
2713
2714``flow list`` lists existing flow rules sorted by priority and optionally
2715filtered by group identifiers::
2716
2717   flow list {port_id} [group {group_id}] [...]
2718
2719This command only fails with the following message if the device does not
2720exist::
2721
2722   Invalid port [...]
2723
2724Output consists of a header line followed by a short description of each
2725flow rule, one per line. There is no output at all when no flow rules are
2726configured on the device::
2727
2728   ID      Group   Prio    Attr    Rule
2729   [...]   [...]   [...]   [...]   [...]
2730
2731``Attr`` column flags:
2732
2733- ``i`` for ``ingress``.
2734- ``e`` for ``egress``.
2735
2736Creating several flow rules and listing them::
2737
2738   testpmd> flow create 0 ingress pattern eth / ipv4 / end
2739      actions queue index 6 / end
2740   Flow rule #0 created
2741   testpmd> flow create 0 ingress pattern eth / ipv6 / end
2742      actions queue index 2 / end
2743   Flow rule #1 created
2744   testpmd> flow create 0 priority 5 ingress pattern eth / ipv4 / udp / end
2745      actions rss queues 6 7 8 end / end
2746   Flow rule #2 created
2747   testpmd> flow list 0
2748   ID      Group   Prio    Attr    Rule
2749   0       0       0       i-      ETH IPV4 => QUEUE
2750   1       0       0       i-      ETH IPV6 => QUEUE
2751   2       0       5       i-      ETH IPV4 UDP => RSS
2752   testpmd>
2753
2754Rules are sorted by priority (i.e. group ID first, then priority level)::
2755
2756   testpmd> flow list 1
2757   ID      Group   Prio    Attr    Rule
2758   0       0       0       i-      ETH => COUNT
2759   6       0       500     i-      ETH IPV6 TCP => DROP COUNT
2760   5       0       1000    i-      ETH IPV6 ICMP => QUEUE
2761   1       24      0       i-      ETH IPV4 UDP => QUEUE
2762   4       24      10      i-      ETH IPV4 TCP => DROP
2763   3       24      20      i-      ETH IPV4 => DROP
2764   2       24      42      i-      ETH IPV4 UDP => QUEUE
2765   7       63      0       i-      ETH IPV6 UDP VXLAN => MARK QUEUE
2766   testpmd>
2767
2768Output can be limited to specific groups::
2769
2770   testpmd> flow list 1 group 0 group 63
2771   ID      Group   Prio    Attr    Rule
2772   0       0       0       i-      ETH => COUNT
2773   6       0       500     i-      ETH IPV6 TCP => DROP COUNT
2774   5       0       1000    i-      ETH IPV6 ICMP => QUEUE
2775   7       63      0       i-      ETH IPV6 UDP VXLAN => MARK QUEUE
2776   testpmd>
2777