xref: /dpdk/doc/guides/testpmd_app_ug/testpmd_funcs.rst (revision 909fd9ac45011ad986422083a8d02a687fe57e03)
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 flow_ctrl rx
918~~~~~~~~~~~~~~~~
919
920Set the link flow control parameter on a port::
921
922   testpmd> set flow_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \
923            (pause_time) (send_xon) mac_ctrl_frame_fwd (on|off) \
924	    autoneg (on|off) (port_id)
925
926Where:
927
928* ``high_water`` (integer): High threshold value to trigger XOFF.
929
930* ``low_water`` (integer): Low threshold value to trigger XON.
931
932* ``pause_time`` (integer): Pause quota in the Pause frame.
933
934* ``send_xon`` (0/1): Send XON frame.
935
936* ``mac_ctrl_frame_fwd``: Enable receiving MAC control frames.
937
938* ``autoneg``: Change the auto-negotiation parameter.
939
940set pfc_ctrl rx
941~~~~~~~~~~~~~~~
942
943Set the priority flow control parameter on a port::
944
945   testpmd> set pfc_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \
946            (pause_time) (priority) (port_id)
947
948Where:
949
950* ``high_water`` (integer): High threshold value.
951
952* ``low_water`` (integer): Low threshold value.
953
954* ``pause_time`` (integer): Pause quota in the Pause frame.
955
956* ``priority`` (0-7): VLAN User Priority.
957
958set stat_qmap
959~~~~~~~~~~~~~
960
961Set statistics mapping (qmapping 0..15) for RX/TX queue on port::
962
963   testpmd> set stat_qmap (tx|rx) (port_id) (queue_id) (qmapping)
964
965For example, to set rx queue 2 on port 0 to mapping 5::
966
967   testpmd>set stat_qmap rx 0 2 5
968
969set port - rx/tx (for VF)
970~~~~~~~~~~~~~~~~~~~~~~~~~
971
972Set VF receive/transmit from a port::
973
974   testpmd> set port (port_id) vf (vf_id) (rx|tx) (on|off)
975
976set port - mac address filter (for VF)
977~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
978
979Add/Remove unicast or multicast MAC addr filter for a VF::
980
981   testpmd> set port (port_id) vf (vf_id) (mac_addr) \
982            (exact-mac|exact-mac-vlan|hashmac|hashmac-vlan) (on|off)
983
984set port - rx mode(for VF)
985~~~~~~~~~~~~~~~~~~~~~~~~~~
986
987Set the VF receive mode of a port::
988
989   testpmd> set port (port_id) vf (vf_id) \
990            rxmode (AUPE|ROPE|BAM|MPE) (on|off)
991
992The available receive modes are:
993
994* ``AUPE``: Accepts untagged VLAN.
995
996* ``ROPE``: Accepts unicast hash.
997
998* ``BAM``: Accepts broadcast packets.
999
1000* ``MPE``: Accepts all multicast packets.
1001
1002set port - tx_rate (for Queue)
1003~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1004
1005Set TX rate limitation for a queue on a port::
1006
1007   testpmd> set port (port_id) queue (queue_id) rate (rate_value)
1008
1009set port - tx_rate (for VF)
1010~~~~~~~~~~~~~~~~~~~~~~~~~~~
1011
1012Set TX rate limitation for queues in VF on a port::
1013
1014   testpmd> set port (port_id) vf (vf_id) rate (rate_value) queue_mask (queue_mask)
1015
1016set port - mirror rule
1017~~~~~~~~~~~~~~~~~~~~~~
1018
1019Set pool or vlan type mirror rule for a port::
1020
1021   testpmd> set port (port_id) mirror-rule (rule_id) \
1022            (pool-mirror-up|pool-mirror-down|vlan-mirror) \
1023            (poolmask|vlanid[,vlanid]*) dst-pool (pool_id) (on|off)
1024
1025Set link mirror rule for a port::
1026
1027   testpmd> set port (port_id) mirror-rule (rule_id) \
1028           (uplink-mirror|downlink-mirror) dst-pool (pool_id) (on|off)
1029
1030For example to enable mirror traffic with vlan 0,1 to pool 0::
1031
1032   set port 0 mirror-rule 0 vlan-mirror 0,1 dst-pool 0 on
1033
1034reset port - mirror rule
1035~~~~~~~~~~~~~~~~~~~~~~~~
1036
1037Reset a mirror rule for a port::
1038
1039   testpmd> reset port (port_id) mirror-rule (rule_id)
1040
1041set flush_rx
1042~~~~~~~~~~~~
1043
1044Set the flush on RX streams before forwarding.
1045The default is flush ``on``.
1046Mainly used with PCAP drivers to turn off the default behavior of flushing the first 512 packets on RX streams::
1047
1048   testpmd> set flush_rx off
1049
1050set bypass mode
1051~~~~~~~~~~~~~~~
1052
1053Set the bypass mode for the lowest port on bypass enabled NIC::
1054
1055   testpmd> set bypass mode (normal|bypass|isolate) (port_id)
1056
1057set bypass event
1058~~~~~~~~~~~~~~~~
1059
1060Set the event required to initiate specified bypass mode for the lowest port on a bypass enabled::
1061
1062   testpmd> set bypass event (timeout|os_on|os_off|power_on|power_off) \
1063            mode (normal|bypass|isolate) (port_id)
1064
1065Where:
1066
1067* ``timeout``: Enable bypass after watchdog timeout.
1068
1069* ``os_on``: Enable bypass when OS/board is powered on.
1070
1071* ``os_off``: Enable bypass when OS/board is powered off.
1072
1073* ``power_on``: Enable bypass when power supply is turned on.
1074
1075* ``power_off``: Enable bypass when power supply is turned off.
1076
1077
1078set bypass timeout
1079~~~~~~~~~~~~~~~~~~
1080
1081Set the bypass watchdog timeout to ``n`` seconds where 0 = instant::
1082
1083   testpmd> set bypass timeout (0|1.5|2|3|4|8|16|32)
1084
1085show bypass config
1086~~~~~~~~~~~~~~~~~~
1087
1088Show the bypass configuration for a bypass enabled NIC using the lowest port on the NIC::
1089
1090   testpmd> show bypass config (port_id)
1091
1092set link up
1093~~~~~~~~~~~
1094
1095Set link up for a port::
1096
1097   testpmd> set link-up port (port id)
1098
1099set link down
1100~~~~~~~~~~~~~
1101
1102Set link down for a port::
1103
1104   testpmd> set link-down port (port id)
1105
1106E-tag set
1107~~~~~~~~~
1108
1109Enable E-tag insertion for a VF on a port::
1110
1111   testpmd> E-tag set insertion on port-tag-id (value) port (port_id) vf (vf_id)
1112
1113Disable E-tag insertion for a VF on a port::
1114
1115   testpmd> E-tag set insertion off port (port_id) vf (vf_id)
1116
1117Enable/disable E-tag stripping on a port::
1118
1119   testpmd> E-tag set stripping (on|off) port (port_id)
1120
1121Enable/disable E-tag based forwarding on a port::
1122
1123   testpmd> E-tag set forwarding (on|off) port (port_id)
1124
1125Add an E-tag forwarding filter on a port::
1126
1127   testpmd> E-tag set filter add e-tag-id (value) dst-pool (pool_id) port (port_id)
1128
1129Delete an E-tag forwarding filter on a port::
1130   testpmd> E-tag set filter del e-tag-id (value) port (port_id)
1131
1132
1133Port Functions
1134--------------
1135
1136The following sections show functions for configuring ports.
1137
1138.. note::
1139
1140   Port configuration changes only become active when forwarding is started/restarted.
1141
1142port attach
1143~~~~~~~~~~~
1144
1145Attach a port specified by pci address or virtual device args::
1146
1147   testpmd> port attach (identifier)
1148
1149To attach a new pci device, the device should be recognized by kernel first.
1150Then it should be moved under DPDK management.
1151Finally the port can be attached to testpmd.
1152
1153For example, to move a pci device using ixgbe under DPDK management:
1154
1155.. code-block:: console
1156
1157   # Check the status of the available devices.
1158   ./usertools/dpdk-devbind.py --status
1159
1160   Network devices using DPDK-compatible driver
1161   ============================================
1162   <none>
1163
1164   Network devices using kernel driver
1165   ===================================
1166   0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused=
1167
1168
1169   # Bind the device to igb_uio.
1170   sudo ./usertools/dpdk-devbind.py -b igb_uio 0000:0a:00.0
1171
1172
1173   # Recheck the status of the devices.
1174   ./usertools/dpdk-devbind.py --status
1175   Network devices using DPDK-compatible driver
1176   ============================================
1177   0000:0a:00.0 '82599ES 10-Gigabit' drv=igb_uio unused=
1178
1179To attach a port created by virtual device, above steps are not needed.
1180
1181For example, to attach a port whose pci address is 0000:0a:00.0.
1182
1183.. code-block:: console
1184
1185   testpmd> port attach 0000:0a:00.0
1186   Attaching a new port...
1187   EAL: PCI device 0000:0a:00.0 on NUMA socket -1
1188   EAL:   probe driver: 8086:10fb rte_ixgbe_pmd
1189   EAL:   PCI memory mapped at 0x7f83bfa00000
1190   EAL:   PCI memory mapped at 0x7f83bfa80000
1191   PMD: eth_ixgbe_dev_init(): MAC: 2, PHY: 18, SFP+: 5
1192   PMD: eth_ixgbe_dev_init(): port 0 vendorID=0x8086 deviceID=0x10fb
1193   Port 0 is attached. Now total ports is 1
1194   Done
1195
1196For example, to attach a port created by pcap PMD.
1197
1198.. code-block:: console
1199
1200   testpmd> port attach net_pcap0
1201   Attaching a new port...
1202   PMD: Initializing pmd_pcap for net_pcap0
1203   PMD: Creating pcap-backed ethdev on numa socket 0
1204   Port 0 is attached. Now total ports is 1
1205   Done
1206
1207In this case, identifier is ``net_pcap0``.
1208This identifier format is the same as ``--vdev`` format of DPDK applications.
1209
1210For example, to re-attach a bonded port which has been previously detached,
1211the mode and slave parameters must be given.
1212
1213.. code-block:: console
1214
1215   testpmd> port attach net_bond_0,mode=0,slave=1
1216   Attaching a new port...
1217   EAL: Initializing pmd_bond for net_bond_0
1218   EAL: Create bonded device net_bond_0 on port 0 in mode 0 on socket 0.
1219   Port 0 is attached. Now total ports is 1
1220   Done
1221
1222
1223port detach
1224~~~~~~~~~~~
1225
1226Detach a specific port::
1227
1228   testpmd> port detach (port_id)
1229
1230Before detaching a port, the port should be stopped and closed.
1231
1232For example, to detach a pci device port 0.
1233
1234.. code-block:: console
1235
1236   testpmd> port stop 0
1237   Stopping ports...
1238   Done
1239   testpmd> port close 0
1240   Closing ports...
1241   Done
1242
1243   testpmd> port detach 0
1244   Detaching a port...
1245   EAL: PCI device 0000:0a:00.0 on NUMA socket -1
1246   EAL:   remove driver: 8086:10fb rte_ixgbe_pmd
1247   EAL:   PCI memory unmapped at 0x7f83bfa00000
1248   EAL:   PCI memory unmapped at 0x7f83bfa80000
1249   Done
1250
1251
1252For example, to detach a virtual device port 0.
1253
1254.. code-block:: console
1255
1256   testpmd> port stop 0
1257   Stopping ports...
1258   Done
1259   testpmd> port close 0
1260   Closing ports...
1261   Done
1262
1263   testpmd> port detach 0
1264   Detaching a port...
1265   PMD: Closing pcap ethdev on numa socket 0
1266   Port 'net_pcap0' is detached. Now total ports is 0
1267   Done
1268
1269To remove a pci device completely from the system, first detach the port from testpmd.
1270Then the device should be moved under kernel management.
1271Finally the device can be removed using kernel pci hotplug functionality.
1272
1273For example, to move a pci device under kernel management:
1274
1275.. code-block:: console
1276
1277   sudo ./usertools/dpdk-devbind.py -b ixgbe 0000:0a:00.0
1278
1279   ./usertools/dpdk-devbind.py --status
1280
1281   Network devices using DPDK-compatible driver
1282   ============================================
1283   <none>
1284
1285   Network devices using kernel driver
1286   ===================================
1287   0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused=igb_uio
1288
1289To remove a port created by a virtual device, above steps are not needed.
1290
1291port start
1292~~~~~~~~~~
1293
1294Start all ports or a specific port::
1295
1296   testpmd> port start (port_id|all)
1297
1298port stop
1299~~~~~~~~~
1300
1301Stop all ports or a specific port::
1302
1303   testpmd> port stop (port_id|all)
1304
1305port close
1306~~~~~~~~~~
1307
1308Close all ports or a specific port::
1309
1310   testpmd> port close (port_id|all)
1311
1312port start/stop queue
1313~~~~~~~~~~~~~~~~~~~~~
1314
1315Start/stop a rx/tx queue on a specific port::
1316
1317   testpmd> port (port_id) (rxq|txq) (queue_id) (start|stop)
1318
1319Only take effect when port is started.
1320
1321port config - speed
1322~~~~~~~~~~~~~~~~~~~
1323
1324Set the speed and duplex mode for all ports or a specific port::
1325
1326   testpmd> port config (port_id|all) speed (10|100|1000|10000|25000|40000|50000|100000|auto) \
1327            duplex (half|full|auto)
1328
1329port config - queues/descriptors
1330~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1331
1332Set number of queues/descriptors for rxq, txq, rxd and txd::
1333
1334   testpmd> port config all (rxq|txq|rxd|txd) (value)
1335
1336This is equivalent to the ``--rxq``, ``--txq``, ``--rxd`` and ``--txd`` command-line options.
1337
1338port config - max-pkt-len
1339~~~~~~~~~~~~~~~~~~~~~~~~~
1340
1341Set the maximum packet length::
1342
1343   testpmd> port config all max-pkt-len (value)
1344
1345This is equivalent to the ``--max-pkt-len`` command-line option.
1346
1347port config - CRC Strip
1348~~~~~~~~~~~~~~~~~~~~~~~
1349
1350Set hardware CRC stripping on or off for all ports::
1351
1352   testpmd> port config all crc-strip (on|off)
1353
1354CRC stripping is off by default.
1355
1356The ``on`` option is equivalent to the ``--crc-strip`` command-line option.
1357
1358port config - scatter
1359~~~~~~~~~~~~~~~~~~~~~~~
1360
1361Set RX scatter mode on or off for all ports::
1362
1363   testpmd> port config all scatter (on|off)
1364
1365RX scatter mode is off by default.
1366
1367The ``on`` option is equivalent to the ``--enable-scatter`` command-line option.
1368
1369port config - TX queue flags
1370~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1371
1372Set a hexadecimal bitmap of TX queue flags for all ports::
1373
1374   testpmd> port config all txqflags value
1375
1376This command is equivalent to the ``--txqflags`` command-line option.
1377
1378port config - RX Checksum
1379~~~~~~~~~~~~~~~~~~~~~~~~~
1380
1381Set hardware RX checksum offload to on or off for all ports::
1382
1383   testpmd> port config all rx-cksum (on|off)
1384
1385Checksum offload is off by default.
1386
1387The ``on`` option is equivalent to the ``--enable-rx-cksum`` command-line option.
1388
1389port config - VLAN
1390~~~~~~~~~~~~~~~~~~
1391
1392Set hardware VLAN on or off for all ports::
1393
1394   testpmd> port config all hw-vlan (on|off)
1395
1396Hardware VLAN is on by default.
1397
1398The ``off`` option is equivalent to the ``--disable-hw-vlan`` command-line option.
1399
1400port config - VLAN filter
1401~~~~~~~~~~~~~~~~~~~~~~~~~
1402
1403Set hardware VLAN filter on or off for all ports::
1404
1405   testpmd> port config all hw-vlan-filter (on|off)
1406
1407Hardware VLAN filter is on by default.
1408
1409The ``off`` option is equivalent to the ``--disable-hw-vlan-filter`` command-line option.
1410
1411port config - VLAN strip
1412~~~~~~~~~~~~~~~~~~~~~~~~
1413
1414Set hardware VLAN strip on or off for all ports::
1415
1416   testpmd> port config all hw-vlan-strip (on|off)
1417
1418Hardware VLAN strip is on by default.
1419
1420The ``off`` option is equivalent to the ``--disable-hw-vlan-strip`` command-line option.
1421
1422port config - VLAN extend
1423~~~~~~~~~~~~~~~~~~~~~~~~~
1424
1425Set hardware VLAN extend on or off for all ports::
1426
1427   testpmd> port config all hw-vlan-extend (on|off)
1428
1429Hardware VLAN extend is off by default.
1430
1431The ``off`` option is equivalent to the ``--disable-hw-vlan-extend`` command-line option.
1432
1433port config - Drop Packets
1434~~~~~~~~~~~~~~~~~~~~~~~~~~
1435
1436Set packet drop for packets with no descriptors on or off for all ports::
1437
1438   testpmd> port config all drop-en (on|off)
1439
1440Packet dropping for packets with no descriptors is off by default.
1441
1442The ``on`` option is equivalent to the ``--enable-drop-en`` command-line option.
1443
1444port config - RSS
1445~~~~~~~~~~~~~~~~~
1446
1447Set the RSS (Receive Side Scaling) mode on or off::
1448
1449   testpmd> port config all rss (all|ip|tcp|udp|sctp|ether|port|vxlan|geneve|nvgre|none)
1450
1451RSS is on by default.
1452
1453The ``none`` option is equivalent to the ``--disable-rss`` command-line option.
1454
1455port config - RSS Reta
1456~~~~~~~~~~~~~~~~~~~~~~
1457
1458Set the RSS (Receive Side Scaling) redirection table::
1459
1460   testpmd> port config all rss reta (hash,queue)[,(hash,queue)]
1461
1462port config - DCB
1463~~~~~~~~~~~~~~~~~
1464
1465Set the DCB mode for an individual port::
1466
1467   testpmd> port config (port_id) dcb vt (on|off) (traffic_class) pfc (on|off)
1468
1469The traffic class should be 4 or 8.
1470
1471port config - Burst
1472~~~~~~~~~~~~~~~~~~~
1473
1474Set the number of packets per burst::
1475
1476   testpmd> port config all burst (value)
1477
1478This is equivalent to the ``--burst`` command-line option.
1479
1480port config - Threshold
1481~~~~~~~~~~~~~~~~~~~~~~~
1482
1483Set thresholds for TX/RX queues::
1484
1485   testpmd> port config all (threshold) (value)
1486
1487Where the threshold type can be:
1488
1489* ``txpt:`` Set the prefetch threshold register of the TX rings, 0 <= value <= 255.
1490
1491* ``txht:`` Set the host threshold register of the TX rings, 0 <= value <= 255.
1492
1493* ``txwt:`` Set the write-back threshold register of the TX rings, 0 <= value <= 255.
1494
1495* ``rxpt:`` Set the prefetch threshold register of the RX rings, 0 <= value <= 255.
1496
1497* ``rxht:`` Set the host threshold register of the RX rings, 0 <= value <= 255.
1498
1499* ``rxwt:`` Set the write-back threshold register of the RX rings, 0 <= value <= 255.
1500
1501* ``txfreet:`` Set the transmit free threshold of the TX rings, 0 <= value <= txd.
1502
1503* ``rxfreet:`` Set the transmit free threshold of the RX rings, 0 <= value <= rxd.
1504
1505* ``txrst:`` Set the transmit RS bit threshold of TX rings, 0 <= value <= txd.
1506
1507These threshold options are also available from the command-line.
1508
1509port config - E-tag
1510~~~~~~~~~~~~~~~~~~~
1511
1512Set the value of ether-type for E-tag::
1513
1514   testpmd> port config (port_id|all) l2-tunnel E-tag ether-type (value)
1515
1516Enable/disable the E-tag support::
1517
1518   testpmd> port config (port_id|all) l2-tunnel E-tag (enable|disable)
1519
1520
1521Link Bonding Functions
1522----------------------
1523
1524The Link Bonding functions make it possible to dynamically create and
1525manage link bonding devices from within testpmd interactive prompt.
1526
1527create bonded device
1528~~~~~~~~~~~~~~~~~~~~
1529
1530Create a new bonding device::
1531
1532   testpmd> create bonded device (mode) (socket)
1533
1534For example, to create a bonded device in mode 1 on socket 0::
1535
1536   testpmd> create bonded 1 0
1537   created new bonded device (port X)
1538
1539add bonding slave
1540~~~~~~~~~~~~~~~~~
1541
1542Adds Ethernet device to a Link Bonding device::
1543
1544   testpmd> add bonding slave (slave id) (port id)
1545
1546For example, to add Ethernet device (port 6) to a Link Bonding device (port 10)::
1547
1548   testpmd> add bonding slave 6 10
1549
1550
1551remove bonding slave
1552~~~~~~~~~~~~~~~~~~~~
1553
1554Removes an Ethernet slave device from a Link Bonding device::
1555
1556   testpmd> remove bonding slave (slave id) (port id)
1557
1558For example, to remove Ethernet slave device (port 6) to a Link Bonding device (port 10)::
1559
1560   testpmd> remove bonding slave 6 10
1561
1562set bonding mode
1563~~~~~~~~~~~~~~~~
1564
1565Set the Link Bonding mode of a Link Bonding device::
1566
1567   testpmd> set bonding mode (value) (port id)
1568
1569For example, to set the bonding mode of a Link Bonding device (port 10) to broadcast (mode 3)::
1570
1571   testpmd> set bonding mode 3 10
1572
1573set bonding primary
1574~~~~~~~~~~~~~~~~~~~
1575
1576Set an Ethernet slave device as the primary device on a Link Bonding device::
1577
1578   testpmd> set bonding primary (slave id) (port id)
1579
1580For example, to set the Ethernet slave device (port 6) as the primary port of a Link Bonding device (port 10)::
1581
1582   testpmd> set bonding primary 6 10
1583
1584set bonding mac
1585~~~~~~~~~~~~~~~
1586
1587Set the MAC address of a Link Bonding device::
1588
1589   testpmd> set bonding mac (port id) (mac)
1590
1591For example, to set the MAC address of a Link Bonding device (port 10) to 00:00:00:00:00:01::
1592
1593   testpmd> set bonding mac 10 00:00:00:00:00:01
1594
1595set bonding xmit_balance_policy
1596~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1597
1598Set the transmission policy for a Link Bonding device when it is in Balance XOR mode::
1599
1600   testpmd> set bonding xmit_balance_policy (port_id) (l2|l23|l34)
1601
1602For example, set a Link Bonding device (port 10) to use a balance policy of layer 3+4 (IP addresses & UDP ports)::
1603
1604   testpmd> set bonding xmit_balance_policy 10 l34
1605
1606
1607set bonding mon_period
1608~~~~~~~~~~~~~~~~~~~~~~
1609
1610Set the link status monitoring polling period in milliseconds for a bonding device.
1611
1612This adds support for PMD slave devices which do not support link status interrupts.
1613When the mon_period is set to a value greater than 0 then all PMD's which do not support
1614link status ISR will be queried every polling interval to check if their link status has changed::
1615
1616   testpmd> set bonding mon_period (port_id) (value)
1617
1618For example, to set the link status monitoring polling period of bonded device (port 5) to 150ms::
1619
1620   testpmd> set bonding mon_period 5 150
1621
1622
1623show bonding config
1624~~~~~~~~~~~~~~~~~~~
1625
1626Show the current configuration of a Link Bonding device::
1627
1628   testpmd> show bonding config (port id)
1629
1630For example,
1631to show the configuration a Link Bonding device (port 9) with 3 slave devices (1, 3, 4)
1632in balance mode with a transmission policy of layer 2+3::
1633
1634   testpmd> show bonding config 9
1635        Bonding mode: 2
1636        Balance Xmit Policy: BALANCE_XMIT_POLICY_LAYER23
1637        Slaves (3): [1 3 4]
1638        Active Slaves (3): [1 3 4]
1639        Primary: [3]
1640
1641
1642Register Functions
1643------------------
1644
1645The Register Functions can be used to read from and write to registers on the network card referenced by a port number.
1646This is mainly useful for debugging purposes.
1647Reference should be made to the appropriate datasheet for the network card for details on the register addresses
1648and fields that can be accessed.
1649
1650read reg
1651~~~~~~~~
1652
1653Display the value of a port register::
1654
1655   testpmd> read reg (port_id) (address)
1656
1657For example, to examine the Flow Director control register (FDIRCTL, 0x0000EE000) on an Intel 82599 10 GbE Controller::
1658
1659   testpmd> read reg 0 0xEE00
1660   port 0 PCI register at offset 0xEE00: 0x4A060029 (1241907241)
1661
1662read regfield
1663~~~~~~~~~~~~~
1664
1665Display a port register bit field::
1666
1667   testpmd> read regfield (port_id) (address) (bit_x) (bit_y)
1668
1669For example, reading the lowest two bits from the register in the example above::
1670
1671   testpmd> read regfield 0 0xEE00 0 1
1672   port 0 PCI register at offset 0xEE00: bits[0, 1]=0x1 (1)
1673
1674read regbit
1675~~~~~~~~~~~
1676
1677Display a single port register bit::
1678
1679   testpmd> read regbit (port_id) (address) (bit_x)
1680
1681For example, reading the lowest bit from the register in the example above::
1682
1683   testpmd> read regbit 0 0xEE00 0
1684   port 0 PCI register at offset 0xEE00: bit 0=1
1685
1686write reg
1687~~~~~~~~~
1688
1689Set the value of a port register::
1690
1691   testpmd> write reg (port_id) (address) (value)
1692
1693For example, to clear a register::
1694
1695   testpmd> write reg 0 0xEE00 0x0
1696   port 0 PCI register at offset 0xEE00: 0x00000000 (0)
1697
1698write regfield
1699~~~~~~~~~~~~~~
1700
1701Set bit field of a port register::
1702
1703   testpmd> write regfield (port_id) (address) (bit_x) (bit_y) (value)
1704
1705For example, writing to the register cleared in the example above::
1706
1707   testpmd> write regfield 0 0xEE00 0 1 2
1708   port 0 PCI register at offset 0xEE00: 0x00000002 (2)
1709
1710write regbit
1711~~~~~~~~~~~~
1712
1713Set single bit value of a port register::
1714
1715   testpmd> write regbit (port_id) (address) (bit_x) (value)
1716
1717For example, to set the high bit in the register from the example above::
1718
1719   testpmd> write regbit 0 0xEE00 31 1
1720   port 0 PCI register at offset 0xEE00: 0x8000000A (2147483658)
1721
1722
1723Filter Functions
1724----------------
1725
1726This section details the available filter functions that are available.
1727
1728Note these functions interface the deprecated legacy filtering framework,
1729superseded by *rte_flow*. See `Flow rules management`_.
1730
1731ethertype_filter
1732~~~~~~~~~~~~~~~~~~~~
1733
1734Add or delete a L2 Ethertype filter, which identify packets by their L2 Ethertype mainly assign them to a receive queue::
1735
1736   ethertype_filter (port_id) (add|del) (mac_addr|mac_ignr) (mac_address) \
1737                    ethertype (ether_type) (drop|fwd) queue (queue_id)
1738
1739The available information parameters are:
1740
1741* ``port_id``: The port which the Ethertype filter assigned on.
1742
1743* ``mac_addr``: Compare destination mac address.
1744
1745* ``mac_ignr``: Ignore destination mac address match.
1746
1747* ``mac_address``: Destination mac address to match.
1748
1749* ``ether_type``: The EtherType value want to match,
1750  for example 0x0806 for ARP packet. 0x0800 (IPv4) and 0x86DD (IPv6) are invalid.
1751
1752* ``queue_id``: The receive queue associated with this EtherType filter.
1753  It is meaningless when deleting or dropping.
1754
1755Example, to add/remove an ethertype filter rule::
1756
1757   testpmd> ethertype_filter 0 add mac_ignr 00:11:22:33:44:55 \
1758                             ethertype 0x0806 fwd queue 3
1759
1760   testpmd> ethertype_filter 0 del mac_ignr 00:11:22:33:44:55 \
1761                             ethertype 0x0806 fwd queue 3
1762
17632tuple_filter
1764~~~~~~~~~~~~~~~~~
1765
1766Add or delete a 2-tuple filter,
1767which identifies packets by specific protocol and destination TCP/UDP port
1768and forwards packets into one of the receive queues::
1769
1770   2tuple_filter (port_id) (add|del) dst_port (dst_port_value) \
1771                 protocol (protocol_value) mask (mask_value) \
1772                 tcp_flags (tcp_flags_value) priority (prio_value) \
1773                 queue (queue_id)
1774
1775The available information parameters are:
1776
1777* ``port_id``: The port which the 2-tuple filter assigned on.
1778
1779* ``dst_port_value``: Destination port in L4.
1780
1781* ``protocol_value``: IP L4 protocol.
1782
1783* ``mask_value``: Participates in the match or not by bit for field above, 1b means participate.
1784
1785* ``tcp_flags_value``: TCP control bits. The non-zero value is invalid, when the pro_value is not set to 0x06 (TCP).
1786
1787* ``prio_value``: Priority of this filter.
1788
1789* ``queue_id``: The receive queue associated with this 2-tuple filter.
1790
1791Example, to add/remove an 2tuple filter rule::
1792
1793   testpmd> 2tuple_filter 0 add dst_port 32 protocol 0x06 mask 0x03 \
1794                          tcp_flags 0x02 priority 3 queue 3
1795
1796   testpmd> 2tuple_filter 0 del dst_port 32 protocol 0x06 mask 0x03 \
1797                          tcp_flags 0x02 priority 3 queue 3
1798
17995tuple_filter
1800~~~~~~~~~~~~~~~~~
1801
1802Add or delete a 5-tuple filter,
1803which consists of a 5-tuple (protocol, source and destination IP addresses, source and destination TCP/UDP/SCTP port)
1804and routes packets into one of the receive queues::
1805
1806   5tuple_filter (port_id) (add|del) dst_ip (dst_address) src_ip \
1807                 (src_address) dst_port (dst_port_value) \
1808                 src_port (src_port_value) protocol (protocol_value) \
1809                 mask (mask_value) tcp_flags (tcp_flags_value) \
1810                 priority (prio_value) queue (queue_id)
1811
1812The available information parameters are:
1813
1814* ``port_id``: The port which the 5-tuple filter assigned on.
1815
1816* ``dst_address``: Destination IP address.
1817
1818* ``src_address``: Source IP address.
1819
1820* ``dst_port_value``: TCP/UDP destination port.
1821
1822* ``src_port_value``: TCP/UDP source port.
1823
1824* ``protocol_value``: L4 protocol.
1825
1826* ``mask_value``: Participates in the match or not by bit for field above, 1b means participate
1827
1828* ``tcp_flags_value``: TCP control bits. The non-zero value is invalid, when the protocol_value is not set to 0x06 (TCP).
1829
1830* ``prio_value``: The priority of this filter.
1831
1832* ``queue_id``: The receive queue associated with this 5-tuple filter.
1833
1834Example, to add/remove an 5tuple filter rule::
1835
1836   testpmd> 5tuple_filter 0 add dst_ip 2.2.2.5 src_ip 2.2.2.4 \
1837            dst_port 64 src_port 32 protocol 0x06 mask 0x1F \
1838            flags 0x0 priority 3 queue 3
1839
1840   testpmd> 5tuple_filter 0 del dst_ip 2.2.2.5 src_ip 2.2.2.4 \
1841            dst_port 64 src_port 32 protocol 0x06 mask 0x1F \
1842            flags 0x0 priority 3 queue 3
1843
1844syn_filter
1845~~~~~~~~~~
1846
1847Using the  SYN filter, TCP packets whose *SYN* flag is set can be forwarded to a separate queue::
1848
1849   syn_filter (port_id) (add|del) priority (high|low) queue (queue_id)
1850
1851The available information parameters are:
1852
1853* ``port_id``: The port which the SYN filter assigned on.
1854
1855* ``high``: This SYN filter has higher priority than other filters.
1856
1857* ``low``: This SYN filter has lower priority than other filters.
1858
1859* ``queue_id``: The receive queue associated with this SYN filter
1860
1861Example::
1862
1863   testpmd> syn_filter 0 add priority high queue 3
1864
1865flex_filter
1866~~~~~~~~~~~
1867
1868With flex filter, packets can be recognized by any arbitrary pattern within the first 128 bytes of the packet
1869and routed into one of the receive queues::
1870
1871   flex_filter (port_id) (add|del) len (len_value) bytes (bytes_value) \
1872               mask (mask_value) priority (prio_value) queue (queue_id)
1873
1874The available information parameters are:
1875
1876* ``port_id``: The port which the Flex filter is assigned on.
1877
1878* ``len_value``: Filter length in bytes, no greater than 128.
1879
1880* ``bytes_value``: A string in hexadecimal, means the value the flex filter needs to match.
1881
1882* ``mask_value``: A string in hexadecimal, bit 1 means corresponding byte participates in the match.
1883
1884* ``prio_value``: The priority of this filter.
1885
1886* ``queue_id``: The receive queue associated with this Flex filter.
1887
1888Example::
1889
1890   testpmd> flex_filter 0 add len 16 bytes 0x00000000000000000000000008060000 \
1891                          mask 000C priority 3 queue 3
1892
1893   testpmd> flex_filter 0 del len 16 bytes 0x00000000000000000000000008060000 \
1894                          mask 000C priority 3 queue 3
1895
1896
1897.. _testpmd_flow_director:
1898
1899flow_director_filter
1900~~~~~~~~~~~~~~~~~~~~
1901
1902The Flow Director works in receive mode to identify specific flows or sets of flows and route them to specific queues.
1903
1904Four types of filtering are supported which are referred to as Perfect Match, Signature, Perfect-mac-vlan and
1905Perfect-tunnel filters, the match mode is set by the ``--pkt-filter-mode`` command-line parameter:
1906
1907* Perfect 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 IP flow.
1910
1911* Signature filters.
1912  The hardware checks a match between a hash-based signature of the masked fields of the received packet.
1913
1914* Perfect-mac-vlan 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 MAC VLAN flow.
1917
1918* Perfect-tunnel match filters.
1919  The hardware checks a match between the masked fields of the received packets and the programmed filters.
1920  The masked fields are for tunnel flow.
1921
1922The Flow Director filters can match the different fields for different type of packet: flow type, specific input set
1923per flow type and the flexible payload.
1924
1925The Flow Director can also mask out parts of all of these fields so that filters
1926are only applied to certain fields or parts of the fields.
1927
1928Different NICs may have different capabilities, command show port fdir (port_id) can be used to acquire the information.
1929
1930# Commands to add flow director filters of different flow types::
1931
1932   flow_director_filter (port_id) mode IP (add|del|update) \
1933                        flow (ipv4-other|ipv4-frag|ipv6-other|ipv6-frag) \
1934                        src (src_ip_address) dst (dst_ip_address) \
1935                        tos (tos_value) proto (proto_value) ttl (ttl_value) \
1936                        vlan (vlan_value) flexbytes (flexbytes_value) \
1937                        (drop|fwd) pf|vf(vf_id) queue (queue_id) \
1938                        fd_id (fd_id_value)
1939
1940   flow_director_filter (port_id) mode IP (add|del|update) \
1941                        flow (ipv4-tcp|ipv4-udp|ipv6-tcp|ipv6-udp) \
1942                        src (src_ip_address) (src_port) \
1943                        dst (dst_ip_address) (dst_port) \
1944                        tos (tos_value) ttl (ttl_value) \
1945                        vlan (vlan_value) flexbytes (flexbytes_value) \
1946                        (drop|fwd) queue pf|vf(vf_id) (queue_id) \
1947                        fd_id (fd_id_value)
1948
1949   flow_director_filter (port_id) mode IP (add|del|update) \
1950                        flow (ipv4-sctp|ipv6-sctp) \
1951                        src (src_ip_address) (src_port) \
1952                        dst (dst_ip_address) (dst_port) \
1953                        tos (tos_value) ttl (ttl_value) \
1954                        tag (verification_tag) vlan (vlan_value) \
1955                        flexbytes (flexbytes_value) (drop|fwd) \
1956                        pf|vf(vf_id) queue (queue_id) fd_id (fd_id_value)
1957
1958   flow_director_filter (port_id) mode IP (add|del|update) flow l2_payload \
1959                        ether (ethertype) flexbytes (flexbytes_value) \
1960                        (drop|fwd) pf|vf(vf_id) queue (queue_id)
1961                        fd_id (fd_id_value)
1962
1963   flow_director_filter (port_id) mode MAC-VLAN (add|del|update) \
1964                        mac (mac_address) vlan (vlan_value) \
1965                        flexbytes (flexbytes_value) (drop|fwd) \
1966                        queue (queue_id) fd_id (fd_id_value)
1967
1968   flow_director_filter (port_id) mode Tunnel (add|del|update) \
1969                        mac (mac_address) vlan (vlan_value) \
1970                        tunnel (NVGRE|VxLAN) tunnel-id (tunnel_id_value) \
1971                        flexbytes (flexbytes_value) (drop|fwd) \
1972                        queue (queue_id) fd_id (fd_id_value)
1973
1974For example, to add an ipv4-udp flow type filter::
1975
1976   testpmd> flow_director_filter 0 mode IP add flow ipv4-udp src 2.2.2.3 32 \
1977            dst 2.2.2.5 33 tos 2 ttl 40 vlan 0x1 flexbytes (0x88,0x48) \
1978            fwd pf queue 1 fd_id 1
1979
1980For example, add an ipv4-other flow type filter::
1981
1982   testpmd> flow_director_filter 0 mode IP add flow ipv4-other src 2.2.2.3 \
1983             dst 2.2.2.5 tos 2 proto 20 ttl 40 vlan 0x1 \
1984             flexbytes (0x88,0x48) fwd pf queue 1 fd_id 1
1985
1986flush_flow_director
1987~~~~~~~~~~~~~~~~~~~
1988
1989Flush all flow director filters on a device::
1990
1991   testpmd> flush_flow_director (port_id)
1992
1993Example, to flush all flow director filter on port 0::
1994
1995   testpmd> flush_flow_director 0
1996
1997flow_director_mask
1998~~~~~~~~~~~~~~~~~~
1999
2000Set flow director's input masks::
2001
2002   flow_director_mask (port_id) mode IP vlan (vlan_value) \
2003                      src_mask (ipv4_src) (ipv6_src) (src_port) \
2004                      dst_mask (ipv4_dst) (ipv6_dst) (dst_port)
2005
2006   flow_director_mask (port_id) mode MAC-VLAN vlan (vlan_value)
2007
2008   flow_director_mask (port_id) mode Tunnel vlan (vlan_value) \
2009                      mac (mac_value) tunnel-type (tunnel_type_value) \
2010                      tunnel-id (tunnel_id_value)
2011
2012Example, to set flow director mask on port 0::
2013
2014   testpmd> flow_director_mask 0 mode IP vlan 0xefff \
2015            src_mask 255.255.255.255 \
2016                FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF \
2017            dst_mask 255.255.255.255 \
2018                FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF
2019
2020flow_director_flex_mask
2021~~~~~~~~~~~~~~~~~~~~~~~
2022
2023set masks of flow director's flexible payload based on certain flow type::
2024
2025   testpmd> flow_director_flex_mask (port_id) \
2026            flow (none|ipv4-other|ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
2027                  ipv6-other|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp| \
2028                  l2_payload|all) (mask)
2029
2030Example, to set flow director's flex mask for all flow type on port 0::
2031
2032   testpmd> flow_director_flex_mask 0 flow all \
2033            (0xff,0xff,0,0,0,0,0,0,0,0,0,0,0,0,0,0)
2034
2035
2036flow_director_flex_payload
2037~~~~~~~~~~~~~~~~~~~~~~~~~~
2038
2039Configure flexible payload selection::
2040
2041   flow_director_flex_payload (port_id) (raw|l2|l3|l4) (config)
2042
2043For example, to select the first 16 bytes from the offset 4 (bytes) of packet's payload as flexible payload::
2044
2045   testpmd> flow_director_flex_payload 0 l4 \
2046            (4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19)
2047
2048get_sym_hash_ena_per_port
2049~~~~~~~~~~~~~~~~~~~~~~~~~
2050
2051Get symmetric hash enable configuration per port::
2052
2053   get_sym_hash_ena_per_port (port_id)
2054
2055For example, to get symmetric hash enable configuration of port 1::
2056
2057   testpmd> get_sym_hash_ena_per_port 1
2058
2059set_sym_hash_ena_per_port
2060~~~~~~~~~~~~~~~~~~~~~~~~~
2061
2062Set symmetric hash enable configuration per port to enable or disable::
2063
2064   set_sym_hash_ena_per_port (port_id) (enable|disable)
2065
2066For example, to set symmetric hash enable configuration of port 1 to enable::
2067
2068   testpmd> set_sym_hash_ena_per_port 1 enable
2069
2070get_hash_global_config
2071~~~~~~~~~~~~~~~~~~~~~~
2072
2073Get the global configurations of hash filters::
2074
2075   get_hash_global_config (port_id)
2076
2077For example, to get the global configurations of hash filters of port 1::
2078
2079   testpmd> get_hash_global_config 1
2080
2081set_hash_global_config
2082~~~~~~~~~~~~~~~~~~~~~~
2083
2084Set the global configurations of hash filters::
2085
2086   set_hash_global_config (port_id) (toeplitz|simple_xor|default) \
2087   (ipv4|ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp|ipv4-other|ipv6|ipv6-frag| \
2088   ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other|l2_payload) \
2089   (enable|disable)
2090
2091For example, to enable simple_xor for flow type of ipv6 on port 2::
2092
2093   testpmd> set_hash_global_config 2 simple_xor ipv6 enable
2094
2095set_hash_input_set
2096~~~~~~~~~~~~~~~~~~
2097
2098Set the input set for hash::
2099
2100   set_hash_input_set (port_id) (ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
2101   ipv4-other|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other| \
2102   l2_payload) (ovlan|ivlan|src-ipv4|dst-ipv4|src-ipv6|dst-ipv6|ipv4-tos| \
2103   ipv4-proto|ipv6-tc|ipv6-next-header|udp-src-port|udp-dst-port| \
2104   tcp-src-port|tcp-dst-port|sctp-src-port|sctp-dst-port|sctp-veri-tag| \
2105   udp-key|gre-key|fld-1st|fld-2nd|fld-3rd|fld-4th|fld-5th|fld-6th|fld-7th| \
2106   fld-8th|none) (select|add)
2107
2108For example, to add source IP to hash input set for flow type of ipv4-udp on port 0::
2109
2110   testpmd> set_hash_input_set 0 ipv4-udp src-ipv4 add
2111
2112set_fdir_input_set
2113~~~~~~~~~~~~~~~~~~
2114
2115The Flow Director filters can match the different fields for different type of packet, i.e. specific input set
2116on per flow type and the flexible payload. This command can be used to change input set for each flow type.
2117
2118Set the input set for flow director::
2119
2120   set_fdir_input_set (port_id) (ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
2121   ipv4-other|ipv6|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other| \
2122   l2_payload) (ivlan|ethertype|src-ipv4|dst-ipv4|src-ipv6|dst-ipv6|ipv4-tos| \
2123   ipv4-proto|ipv4-ttl|ipv6-tc|ipv6-next-header|ipv6-hop-limits| \
2124   tudp-src-port|udp-dst-port|cp-src-port|tcp-dst-port|sctp-src-port| \
2125   sctp-dst-port|sctp-veri-tag|none) (select|add)
2126
2127For example to add source IP to FD input set for flow type of ipv4-udp on port 0::
2128
2129   testpmd> set_fdir_input_set 0 ipv4-udp src-ipv4 add
2130
2131global_config
2132~~~~~~~~~~~~~
2133
2134Set different GRE key length for input set::
2135
2136   global_config (port_id) gre-key-len (number in bytes)
2137
2138For example to set GRE key length for input set to 4 bytes on port 0::
2139
2140   testpmd> global_config 0 gre-key-len 4
2141
2142
2143.. _testpmd_rte_flow:
2144
2145Flow rules management
2146---------------------
2147
2148Control of the generic flow API (*rte_flow*) is fully exposed through the
2149``flow`` command (validation, creation, destruction and queries).
2150
2151Considering *rte_flow* overlaps with all `Filter Functions`_, using both
2152features simultaneously may cause undefined side-effects and is therefore
2153not recommended.
2154
2155``flow`` syntax
2156~~~~~~~~~~~~~~~
2157
2158Because the ``flow`` command uses dynamic tokens to handle the large number
2159of possible flow rules combinations, its behavior differs slightly from
2160other commands, in particular:
2161
2162- Pressing *?* or the *<tab>* key displays contextual help for the current
2163  token, not that of the entire command.
2164
2165- Optional and repeated parameters are supported (provided they are listed
2166  in the contextual help).
2167
2168The first parameter stands for the operation mode. Possible operations and
2169their general syntax are described below. They are covered in detail in the
2170following sections.
2171
2172- Check whether a flow rule can be created::
2173
2174   flow validate {port_id}
2175       [group {group_id}] [priority {level}] [ingress] [egress]
2176       pattern {item} [/ {item} [...]] / end
2177       actions {action} [/ {action} [...]] / end
2178
2179- Create a flow rule::
2180
2181   flow create {port_id}
2182       [group {group_id}] [priority {level}] [ingress] [egress]
2183       pattern {item} [/ {item} [...]] / end
2184       actions {action} [/ {action} [...]] / end
2185
2186- Destroy specific flow rules::
2187
2188   flow destroy {port_id} rule {rule_id} [...]
2189
2190- Destroy all flow rules::
2191
2192   flow flush {port_id}
2193
2194- Query an existing flow rule::
2195
2196   flow query {port_id} {rule_id} {action}
2197
2198- List existing flow rules sorted by priority, filtered by group
2199  identifiers::
2200
2201   flow list {port_id} [group {group_id}] [...]
2202
2203Validating flow rules
2204~~~~~~~~~~~~~~~~~~~~~
2205
2206``flow validate`` reports whether a flow rule would be accepted by the
2207underlying device in its current state but stops short of creating it. It is
2208bound to ``rte_flow_validate()``::
2209
2210   flow validate {port_id}
2211      [group {group_id}] [priority {level}] [ingress] [egress]
2212      pattern {item} [/ {item} [...]] / end
2213      actions {action} [/ {action} [...]] / end
2214
2215If successful, it will show::
2216
2217   Flow rule validated
2218
2219Otherwise it will show an error message of the form::
2220
2221   Caught error type [...] ([...]): [...]
2222
2223This command uses the same parameters as ``flow create``, their format is
2224described in `Creating flow rules`_.
2225
2226Check whether redirecting any Ethernet packet received on port 0 to RX queue
2227index 6 is supported::
2228
2229   testpmd> flow validate 0 ingress pattern eth / end
2230      actions queue index 6 / end
2231   Flow rule validated
2232   testpmd>
2233
2234Port 0 does not support TCPv6 rules::
2235
2236   testpmd> flow validate 0 ingress pattern eth / ipv6 / tcp / end
2237      actions drop / end
2238   Caught error type 9 (specific pattern item): Invalid argument
2239   testpmd>
2240
2241Creating flow rules
2242~~~~~~~~~~~~~~~~~~~
2243
2244``flow create`` validates and creates the specified flow rule. It is bound
2245to ``rte_flow_create()``::
2246
2247   flow create {port_id}
2248      [group {group_id}] [priority {level}] [ingress] [egress]
2249      pattern {item} [/ {item} [...]] / end
2250      actions {action} [/ {action} [...]] / end
2251
2252If successful, it will return a flow rule ID usable with other commands::
2253
2254   Flow rule #[...] created
2255
2256Otherwise it will show an error message of the form::
2257
2258   Caught error type [...] ([...]): [...]
2259
2260Parameters describe in the following order:
2261
2262- Attributes (*group*, *priority*, *ingress*, *egress* tokens).
2263- A matching pattern, starting with the *pattern* token and terminated by an
2264  *end* pattern item.
2265- Actions, starting with the *actions* token and terminated by an *end*
2266  action.
2267
2268These translate directly to *rte_flow* objects provided as-is to the
2269underlying functions.
2270
2271The shortest valid definition only comprises mandatory tokens::
2272
2273   testpmd> flow create 0 pattern end actions end
2274
2275Note that PMDs may refuse rules that essentially do nothing such as this
2276one.
2277
2278**All unspecified object values are automatically initialized to 0.**
2279
2280Attributes
2281^^^^^^^^^^
2282
2283These tokens affect flow rule attributes (``struct rte_flow_attr``) and are
2284specified before the ``pattern`` token.
2285
2286- ``group {group id}``: priority group.
2287- ``priority {level}``: priority level within group.
2288- ``ingress``: rule applies to ingress traffic.
2289- ``egress``: rule applies to egress traffic.
2290
2291Each instance of an attribute specified several times overrides the previous
2292value as shown below (group 4 is used)::
2293
2294   testpmd> flow create 0 group 42 group 24 group 4 [...]
2295
2296Note that once enabled, ``ingress`` and ``egress`` cannot be disabled.
2297
2298While not specifying a direction is an error, some rules may allow both
2299simultaneously.
2300
2301Most rules affect RX therefore contain the ``ingress`` token::
2302
2303   testpmd> flow create 0 ingress pattern [...]
2304
2305Matching pattern
2306^^^^^^^^^^^^^^^^
2307
2308A matching pattern starts after the ``pattern`` token. It is made of pattern
2309items and is terminated by a mandatory ``end`` item.
2310
2311Items are named after their type (*RTE_FLOW_ITEM_TYPE_* from ``enum
2312rte_flow_item_type``).
2313
2314The ``/`` token is used as a separator between pattern items as shown
2315below::
2316
2317   testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end [...]
2318
2319Note that protocol items like these must be stacked from lowest to highest
2320layer to make sense. For instance, the following rule is either invalid or
2321unlikely to match any packet::
2322
2323   testpmd> flow create 0 ingress pattern eth / udp / ipv4 / end [...]
2324
2325More information on these restrictions can be found in the *rte_flow*
2326documentation.
2327
2328Several items support additional specification structures, for example
2329``ipv4`` allows specifying source and destination addresses as follows::
2330
2331   testpmd> flow create 0 ingress pattern eth / ipv4 src is 10.1.1.1
2332      dst is 10.2.0.0 / end [...]
2333
2334This rule matches all IPv4 traffic with the specified properties.
2335
2336In this example, ``src`` and ``dst`` are field names of the underlying
2337``struct rte_flow_item_ipv4`` object. All item properties can be specified
2338in a similar fashion.
2339
2340The ``is`` token means that the subsequent value must be matched exactly,
2341and assigns ``spec`` and ``mask`` fields in ``struct rte_flow_item``
2342accordingly. Possible assignment tokens are:
2343
2344- ``is``: match value perfectly (with full bit-mask).
2345- ``spec``: match value according to configured bit-mask.
2346- ``last``: specify upper bound to establish a range.
2347- ``mask``: specify bit-mask with relevant bits set to one.
2348- ``prefix``: generate bit-mask from a prefix length.
2349
2350These yield identical results::
2351
2352   ipv4 src is 10.1.1.1
2353
2354::
2355
2356   ipv4 src spec 10.1.1.1 src mask 255.255.255.255
2357
2358::
2359
2360   ipv4 src spec 10.1.1.1 src prefix 32
2361
2362::
2363
2364   ipv4 src is 10.1.1.1 src last 10.1.1.1 # range with a single value
2365
2366::
2367
2368   ipv4 src is 10.1.1.1 src last 0 # 0 disables range
2369
2370Inclusive ranges can be defined with ``last``::
2371
2372   ipv4 src is 10.1.1.1 src last 10.2.3.4 # 10.1.1.1 to 10.2.3.4
2373
2374Note that ``mask`` affects both ``spec`` and ``last``::
2375
2376   ipv4 src is 10.1.1.1 src last 10.2.3.4 src mask 255.255.0.0
2377      # matches 10.1.0.0 to 10.2.255.255
2378
2379Properties can be modified multiple times::
2380
2381   ipv4 src is 10.1.1.1 src is 10.1.2.3 src is 10.2.3.4 # matches 10.2.3.4
2382
2383::
2384
2385   ipv4 src is 10.1.1.1 src prefix 24 src prefix 16 # matches 10.1.0.0/16
2386
2387Pattern items
2388^^^^^^^^^^^^^
2389
2390This section lists supported pattern items and their attributes, if any.
2391
2392- ``end``: end list of pattern items.
2393
2394- ``void``: no-op pattern item.
2395
2396- ``invert``: perform actions when pattern does not match.
2397
2398- ``any``: match any protocol for the current layer.
2399
2400  - ``num {unsigned}``: number of layers covered.
2401
2402- ``pf``: match packets addressed to the physical function.
2403
2404- ``vf``: match packets addressed to a virtual function ID.
2405
2406  - ``id {unsigned}``: destination VF ID.
2407
2408- ``port``: device-specific physical port index to use.
2409
2410  - ``index {unsigned}``: physical port index.
2411
2412- ``raw``: match an arbitrary byte string.
2413
2414  - ``relative {boolean}``: look for pattern after the previous item.
2415  - ``search {boolean}``: search pattern from offset (see also limit).
2416  - ``offset {integer}``: absolute or relative offset for pattern.
2417  - ``limit {unsigned}``: search area limit for start of pattern.
2418  - ``pattern {string}``: byte string to look for.
2419
2420- ``eth``: match Ethernet header.
2421
2422  - ``dst {MAC-48}``: destination MAC.
2423  - ``src {MAC-48}``: source MAC.
2424  - ``type {unsigned}``: EtherType.
2425
2426- ``vlan``: match 802.1Q/ad VLAN tag.
2427
2428  - ``tpid {unsigned}``: tag protocol identifier.
2429  - ``tci {unsigned}``: tag control information.
2430  - ``pcp {unsigned}``: priority code point.
2431  - ``dei {unsigned}``: drop eligible indicator.
2432  - ``vid {unsigned}``: VLAN identifier.
2433
2434- ``ipv4``: match IPv4 header.
2435
2436  - ``tos {unsigned}``: type of service.
2437  - ``ttl {unsigned}``: time to live.
2438  - ``proto {unsigned}``: next protocol ID.
2439  - ``src {ipv4 address}``: source address.
2440  - ``dst {ipv4 address}``: destination address.
2441
2442- ``ipv6``: match IPv6 header.
2443
2444  - ``tc {unsigned}``: traffic class.
2445  - ``flow {unsigned}``: flow label.
2446  - ``proto {unsigned}``: protocol (next header).
2447  - ``hop {unsigned}``: hop limit.
2448  - ``src {ipv6 address}``: source address.
2449  - ``dst {ipv6 address}``: destination address.
2450
2451- ``icmp``: match ICMP header.
2452
2453  - ``type {unsigned}``: ICMP packet type.
2454  - ``code {unsigned}``: ICMP packet code.
2455
2456- ``udp``: match UDP header.
2457
2458  - ``src {unsigned}``: UDP source port.
2459  - ``dst {unsigned}``: UDP destination port.
2460
2461- ``tcp``: match TCP header.
2462
2463  - ``src {unsigned}``: TCP source port.
2464  - ``dst {unsigned}``: TCP destination port.
2465
2466- ``sctp``: match SCTP header.
2467
2468  - ``src {unsigned}``: SCTP source port.
2469  - ``dst {unsigned}``: SCTP destination port.
2470  - ``tag {unsigned}``: validation tag.
2471  - ``cksum {unsigned}``: checksum.
2472
2473- ``vxlan``: match VXLAN header.
2474
2475  - ``vni {unsigned}``: VXLAN identifier.
2476
2477Actions list
2478^^^^^^^^^^^^
2479
2480A list of actions starts after the ``actions`` token in the same fashion as
2481`Matching pattern`_; actions are separated by ``/`` tokens and the list is
2482terminated by a mandatory ``end`` action.
2483
2484Actions are named after their type (*RTE_FLOW_ACTION_TYPE_* from ``enum
2485rte_flow_action_type``).
2486
2487Dropping all incoming UDPv4 packets can be expressed as follows::
2488
2489   testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
2490      actions drop / end
2491
2492Several actions have configurable properties which must be specified when
2493there is no valid default value. For example, ``queue`` requires a target
2494queue index.
2495
2496This rule redirects incoming UDPv4 traffic to queue index 6::
2497
2498   testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
2499      actions queue index 6 / end
2500
2501While this one could be rejected by PMDs (unspecified queue index)::
2502
2503   testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
2504      actions queue / end
2505
2506As defined by *rte_flow*, the list is not ordered, all actions of a given
2507rule are performed simultaneously. These are equivalent::
2508
2509   queue index 6 / void / mark id 42 / end
2510
2511::
2512
2513   void / mark id 42 / queue index 6 / end
2514
2515All actions in a list should have different types, otherwise only the last
2516action of a given type is taken into account::
2517
2518   queue index 4 / queue index 5 / queue index 6 / end # will use queue 6
2519
2520::
2521
2522   drop / drop / drop / end # drop is performed only once
2523
2524::
2525
2526   mark id 42 / queue index 3 / mark id 24 / end # mark will be 24
2527
2528Considering they are performed simultaneously, opposite and overlapping
2529actions can sometimes be combined when the end result is unambiguous::
2530
2531   drop / queue index 6 / end # drop has no effect
2532
2533::
2534
2535   drop / dup index 6 / end # same as above
2536
2537::
2538
2539   queue index 6 / rss queues 6 7 8 / end # queue has no effect
2540
2541::
2542
2543   drop / passthru / end # drop has no effect
2544
2545Note that PMDs may still refuse such combinations.
2546
2547Actions
2548^^^^^^^
2549
2550This section lists supported actions and their attributes, if any.
2551
2552- ``end``: end list of actions.
2553
2554- ``void``: no-op action.
2555
2556- ``passthru``: let subsequent rule process matched packets.
2557
2558- ``mark``: attach 32 bit value to packets.
2559
2560  - ``id {unsigned}``: 32 bit value to return with packets.
2561
2562- ``flag``: flag packets.
2563
2564- ``queue``: assign packets to a given queue index.
2565
2566  - ``index {unsigned}``: queue index to use.
2567
2568- ``drop``: drop packets (note: passthru has priority).
2569
2570- ``count``: enable counters for this rule.
2571
2572- ``dup``: duplicate packets to a given queue index.
2573
2574  - ``index {unsigned}``: queue index to duplicate packets to.
2575
2576- ``rss``: spread packets among several queues.
2577
2578  - ``queues [{unsigned} [...]] end``: queue indices to use.
2579
2580- ``pf``: redirect packets to physical device function.
2581
2582- ``vf``: redirect packets to virtual device function.
2583
2584  - ``original {boolean}``: use original VF ID if possible.
2585  - ``id {unsigned}``: VF ID to redirect packets to.
2586
2587Destroying flow rules
2588~~~~~~~~~~~~~~~~~~~~~
2589
2590``flow destroy`` destroys one or more rules from their rule ID (as returned
2591by ``flow create``), this command calls ``rte_flow_destroy()`` as many
2592times as necessary::
2593
2594   flow destroy {port_id} rule {rule_id} [...]
2595
2596If successful, it will show::
2597
2598   Flow rule #[...] destroyed
2599
2600It does not report anything for rule IDs that do not exist. The usual error
2601message is shown when a rule cannot be destroyed::
2602
2603   Caught error type [...] ([...]): [...]
2604
2605``flow flush`` destroys all rules on a device and does not take extra
2606arguments. It is bound to ``rte_flow_flush()``::
2607
2608   flow flush {port_id}
2609
2610Any errors are reported as above.
2611
2612Creating several rules and destroying them::
2613
2614   testpmd> flow create 0 ingress pattern eth / ipv6 / end
2615      actions queue index 2 / end
2616   Flow rule #0 created
2617   testpmd> flow create 0 ingress pattern eth / ipv4 / end
2618      actions queue index 3 / end
2619   Flow rule #1 created
2620   testpmd> flow destroy 0 rule 0 rule 1
2621   Flow rule #1 destroyed
2622   Flow rule #0 destroyed
2623   testpmd>
2624
2625The same result can be achieved using ``flow flush``::
2626
2627   testpmd> flow create 0 ingress pattern eth / ipv6 / end
2628      actions queue index 2 / end
2629   Flow rule #0 created
2630   testpmd> flow create 0 ingress pattern eth / ipv4 / end
2631      actions queue index 3 / end
2632   Flow rule #1 created
2633   testpmd> flow flush 0
2634   testpmd>
2635
2636Non-existent rule IDs are ignored::
2637
2638   testpmd> flow create 0 ingress pattern eth / ipv6 / end
2639      actions queue index 2 / end
2640   Flow rule #0 created
2641   testpmd> flow create 0 ingress pattern eth / ipv4 / end
2642      actions queue index 3 / end
2643   Flow rule #1 created
2644   testpmd> flow destroy 0 rule 42 rule 10 rule 2
2645   testpmd>
2646   testpmd> flow destroy 0 rule 0
2647   Flow rule #0 destroyed
2648   testpmd>
2649
2650Querying flow rules
2651~~~~~~~~~~~~~~~~~~~
2652
2653``flow query`` queries a specific action of a flow rule having that
2654ability. Such actions collect information that can be reported using this
2655command. It is bound to ``rte_flow_query()``::
2656
2657   flow query {port_id} {rule_id} {action}
2658
2659If successful, it will display either the retrieved data for known actions
2660or the following message::
2661
2662   Cannot display result for action type [...] ([...])
2663
2664Otherwise, it will complain either that the rule does not exist or that some
2665error occurred::
2666
2667   Flow rule #[...] not found
2668
2669::
2670
2671   Caught error type [...] ([...]): [...]
2672
2673Currently only the ``count`` action is supported. This action reports the
2674number of packets that hit the flow rule and the total number of bytes. Its
2675output has the following format::
2676
2677   count:
2678    hits_set: [...] # whether "hits" contains a valid value
2679    bytes_set: [...] # whether "bytes" contains a valid value
2680    hits: [...] # number of packets
2681    bytes: [...] # number of bytes
2682
2683Querying counters for TCPv6 packets redirected to queue 6::
2684
2685   testpmd> flow create 0 ingress pattern eth / ipv6 / tcp / end
2686      actions queue index 6 / count / end
2687   Flow rule #4 created
2688   testpmd> flow query 0 4 count
2689   count:
2690    hits_set: 1
2691    bytes_set: 0
2692    hits: 386446
2693    bytes: 0
2694   testpmd>
2695
2696Listing flow rules
2697~~~~~~~~~~~~~~~~~~
2698
2699``flow list`` lists existing flow rules sorted by priority and optionally
2700filtered by group identifiers::
2701
2702   flow list {port_id} [group {group_id}] [...]
2703
2704This command only fails with the following message if the device does not
2705exist::
2706
2707   Invalid port [...]
2708
2709Output consists of a header line followed by a short description of each
2710flow rule, one per line. There is no output at all when no flow rules are
2711configured on the device::
2712
2713   ID      Group   Prio    Attr    Rule
2714   [...]   [...]   [...]   [...]   [...]
2715
2716``Attr`` column flags:
2717
2718- ``i`` for ``ingress``.
2719- ``e`` for ``egress``.
2720
2721Creating several flow rules and listing them::
2722
2723   testpmd> flow create 0 ingress pattern eth / ipv4 / end
2724      actions queue index 6 / end
2725   Flow rule #0 created
2726   testpmd> flow create 0 ingress pattern eth / ipv6 / end
2727      actions queue index 2 / end
2728   Flow rule #1 created
2729   testpmd> flow create 0 priority 5 ingress pattern eth / ipv4 / udp / end
2730      actions rss queues 6 7 8 end / end
2731   Flow rule #2 created
2732   testpmd> flow list 0
2733   ID      Group   Prio    Attr    Rule
2734   0       0       0       i-      ETH IPV4 => QUEUE
2735   1       0       0       i-      ETH IPV6 => QUEUE
2736   2       0       5       i-      ETH IPV4 UDP => RSS
2737   testpmd>
2738
2739Rules are sorted by priority (i.e. group ID first, then priority level)::
2740
2741   testpmd> flow list 1
2742   ID      Group   Prio    Attr    Rule
2743   0       0       0       i-      ETH => COUNT
2744   6       0       500     i-      ETH IPV6 TCP => DROP COUNT
2745   5       0       1000    i-      ETH IPV6 ICMP => QUEUE
2746   1       24      0       i-      ETH IPV4 UDP => QUEUE
2747   4       24      10      i-      ETH IPV4 TCP => DROP
2748   3       24      20      i-      ETH IPV4 => DROP
2749   2       24      42      i-      ETH IPV4 UDP => QUEUE
2750   7       63      0       i-      ETH IPV6 UDP VXLAN => MARK QUEUE
2751   testpmd>
2752
2753Output can be limited to specific groups::
2754
2755   testpmd> flow list 1 group 0 group 63
2756   ID      Group   Prio    Attr    Rule
2757   0       0       0       i-      ETH => COUNT
2758   6       0       500     i-      ETH IPV6 TCP => DROP COUNT
2759   5       0       1000    i-      ETH IPV6 ICMP => QUEUE
2760   7       63      0       i-      ETH IPV6 UDP VXLAN => MARK QUEUE
2761   testpmd>
2762