xref: /dpdk/doc/guides/testpmd_app_ug/testpmd_funcs.rst (revision 5852bf6ae1fae11d01712a6a0d3138a7a1370e2c)
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 flow_ctrl rx
897~~~~~~~~~~~~~~~~
898
899Set the link flow control parameter on a port::
900
901   testpmd> set flow_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \
902            (pause_time) (send_xon) mac_ctrl_frame_fwd (on|off) \
903	    autoneg (on|off) (port_id)
904
905Where:
906
907* ``high_water`` (integer): High threshold value to trigger XOFF.
908
909* ``low_water`` (integer): Low threshold value to trigger XON.
910
911* ``pause_time`` (integer): Pause quota in the Pause frame.
912
913* ``send_xon`` (0/1): Send XON frame.
914
915* ``mac_ctrl_frame_fwd``: Enable receiving MAC control frames.
916
917* ``autoneg``: Change the auto-negotiation parameter.
918
919set pfc_ctrl rx
920~~~~~~~~~~~~~~~
921
922Set the priority flow control parameter on a port::
923
924   testpmd> set pfc_ctrl rx (on|off) tx (on|off) (high_water) (low_water) \
925            (pause_time) (priority) (port_id)
926
927Where:
928
929* ``high_water`` (integer): High threshold value.
930
931* ``low_water`` (integer): Low threshold value.
932
933* ``pause_time`` (integer): Pause quota in the Pause frame.
934
935* ``priority`` (0-7): VLAN User Priority.
936
937set stat_qmap
938~~~~~~~~~~~~~
939
940Set statistics mapping (qmapping 0..15) for RX/TX queue on port::
941
942   testpmd> set stat_qmap (tx|rx) (port_id) (queue_id) (qmapping)
943
944For example, to set rx queue 2 on port 0 to mapping 5::
945
946   testpmd>set stat_qmap rx 0 2 5
947
948set port - rx/tx (for VF)
949~~~~~~~~~~~~~~~~~~~~~~~~~
950
951Set VF receive/transmit from a port::
952
953   testpmd> set port (port_id) vf (vf_id) (rx|tx) (on|off)
954
955set port - mac address filter (for VF)
956~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
957
958Add/Remove unicast or multicast MAC addr filter for a VF::
959
960   testpmd> set port (port_id) vf (vf_id) (mac_addr) \
961            (exact-mac|exact-mac-vlan|hashmac|hashmac-vlan) (on|off)
962
963set port - rx mode(for VF)
964~~~~~~~~~~~~~~~~~~~~~~~~~~
965
966Set the VF receive mode of a port::
967
968   testpmd> set port (port_id) vf (vf_id) \
969            rxmode (AUPE|ROPE|BAM|MPE) (on|off)
970
971The available receive modes are:
972
973* ``AUPE``: Accepts untagged VLAN.
974
975* ``ROPE``: Accepts unicast hash.
976
977* ``BAM``: Accepts broadcast packets.
978
979* ``MPE``: Accepts all multicast packets.
980
981set port - tx_rate (for Queue)
982~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
983
984Set TX rate limitation for a queue on a port::
985
986   testpmd> set port (port_id) queue (queue_id) rate (rate_value)
987
988set port - tx_rate (for VF)
989~~~~~~~~~~~~~~~~~~~~~~~~~~~
990
991Set TX rate limitation for queues in VF on a port::
992
993   testpmd> set port (port_id) vf (vf_id) rate (rate_value) queue_mask (queue_mask)
994
995set port - mirror rule
996~~~~~~~~~~~~~~~~~~~~~~
997
998Set pool or vlan type mirror rule for a port::
999
1000   testpmd> set port (port_id) mirror-rule (rule_id) \
1001            (pool-mirror-up|pool-mirror-down|vlan-mirror) \
1002            (poolmask|vlanid[,vlanid]*) dst-pool (pool_id) (on|off)
1003
1004Set link mirror rule for a port::
1005
1006   testpmd> set port (port_id) mirror-rule (rule_id) \
1007           (uplink-mirror|downlink-mirror) dst-pool (pool_id) (on|off)
1008
1009For example to enable mirror traffic with vlan 0,1 to pool 0::
1010
1011   set port 0 mirror-rule 0 vlan-mirror 0,1 dst-pool 0 on
1012
1013reset port - mirror rule
1014~~~~~~~~~~~~~~~~~~~~~~~~
1015
1016Reset a mirror rule for a port::
1017
1018   testpmd> reset port (port_id) mirror-rule (rule_id)
1019
1020set flush_rx
1021~~~~~~~~~~~~
1022
1023Set the flush on RX streams before forwarding.
1024The default is flush ``on``.
1025Mainly used with PCAP drivers to turn off the default behavior of flushing the first 512 packets on RX streams::
1026
1027   testpmd> set flush_rx off
1028
1029set bypass mode
1030~~~~~~~~~~~~~~~
1031
1032Set the bypass mode for the lowest port on bypass enabled NIC::
1033
1034   testpmd> set bypass mode (normal|bypass|isolate) (port_id)
1035
1036set bypass event
1037~~~~~~~~~~~~~~~~
1038
1039Set the event required to initiate specified bypass mode for the lowest port on a bypass enabled::
1040
1041   testpmd> set bypass event (timeout|os_on|os_off|power_on|power_off) \
1042            mode (normal|bypass|isolate) (port_id)
1043
1044Where:
1045
1046* ``timeout``: Enable bypass after watchdog timeout.
1047
1048* ``os_on``: Enable bypass when OS/board is powered on.
1049
1050* ``os_off``: Enable bypass when OS/board is powered off.
1051
1052* ``power_on``: Enable bypass when power supply is turned on.
1053
1054* ``power_off``: Enable bypass when power supply is turned off.
1055
1056
1057set bypass timeout
1058~~~~~~~~~~~~~~~~~~
1059
1060Set the bypass watchdog timeout to ``n`` seconds where 0 = instant::
1061
1062   testpmd> set bypass timeout (0|1.5|2|3|4|8|16|32)
1063
1064show bypass config
1065~~~~~~~~~~~~~~~~~~
1066
1067Show the bypass configuration for a bypass enabled NIC using the lowest port on the NIC::
1068
1069   testpmd> show bypass config (port_id)
1070
1071set link up
1072~~~~~~~~~~~
1073
1074Set link up for a port::
1075
1076   testpmd> set link-up port (port id)
1077
1078set link down
1079~~~~~~~~~~~~~
1080
1081Set link down for a port::
1082
1083   testpmd> set link-down port (port id)
1084
1085E-tag set
1086~~~~~~~~~
1087
1088Enable E-tag insertion for a VF on a port::
1089
1090   testpmd> E-tag set insertion on port-tag-id (value) port (port_id) vf (vf_id)
1091
1092Disable E-tag insertion for a VF on a port::
1093
1094   testpmd> E-tag set insertion off port (port_id) vf (vf_id)
1095
1096Enable/disable E-tag stripping on a port::
1097
1098   testpmd> E-tag set stripping (on|off) port (port_id)
1099
1100Enable/disable E-tag based forwarding on a port::
1101
1102   testpmd> E-tag set forwarding (on|off) port (port_id)
1103
1104Add an E-tag forwarding filter on a port::
1105
1106   testpmd> E-tag set filter add e-tag-id (value) dst-pool (pool_id) port (port_id)
1107
1108Delete an E-tag forwarding filter on a port::
1109   testpmd> E-tag set filter del e-tag-id (value) port (port_id)
1110
1111
1112Port Functions
1113--------------
1114
1115The following sections show functions for configuring ports.
1116
1117.. note::
1118
1119   Port configuration changes only become active when forwarding is started/restarted.
1120
1121port attach
1122~~~~~~~~~~~
1123
1124Attach a port specified by pci address or virtual device args::
1125
1126   testpmd> port attach (identifier)
1127
1128To attach a new pci device, the device should be recognized by kernel first.
1129Then it should be moved under DPDK management.
1130Finally the port can be attached to testpmd.
1131
1132For example, to move a pci device using ixgbe under DPDK management:
1133
1134.. code-block:: console
1135
1136   # Check the status of the available devices.
1137   ./usertools/dpdk-devbind.py --status
1138
1139   Network devices using DPDK-compatible driver
1140   ============================================
1141   <none>
1142
1143   Network devices using kernel driver
1144   ===================================
1145   0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused=
1146
1147
1148   # Bind the device to igb_uio.
1149   sudo ./usertools/dpdk-devbind.py -b igb_uio 0000:0a:00.0
1150
1151
1152   # Recheck the status of the devices.
1153   ./usertools/dpdk-devbind.py --status
1154   Network devices using DPDK-compatible driver
1155   ============================================
1156   0000:0a:00.0 '82599ES 10-Gigabit' drv=igb_uio unused=
1157
1158To attach a port created by virtual device, above steps are not needed.
1159
1160For example, to attach a port whose pci address is 0000:0a:00.0.
1161
1162.. code-block:: console
1163
1164   testpmd> port attach 0000:0a:00.0
1165   Attaching a new port...
1166   EAL: PCI device 0000:0a:00.0 on NUMA socket -1
1167   EAL:   probe driver: 8086:10fb rte_ixgbe_pmd
1168   EAL:   PCI memory mapped at 0x7f83bfa00000
1169   EAL:   PCI memory mapped at 0x7f83bfa80000
1170   PMD: eth_ixgbe_dev_init(): MAC: 2, PHY: 18, SFP+: 5
1171   PMD: eth_ixgbe_dev_init(): port 0 vendorID=0x8086 deviceID=0x10fb
1172   Port 0 is attached. Now total ports is 1
1173   Done
1174
1175For example, to attach a port created by pcap PMD.
1176
1177.. code-block:: console
1178
1179   testpmd> port attach net_pcap0
1180   Attaching a new port...
1181   PMD: Initializing pmd_pcap for net_pcap0
1182   PMD: Creating pcap-backed ethdev on numa socket 0
1183   Port 0 is attached. Now total ports is 1
1184   Done
1185
1186In this case, identifier is ``net_pcap0``.
1187This identifier format is the same as ``--vdev`` format of DPDK applications.
1188
1189For example, to re-attach a bonded port which has been previously detached,
1190the mode and slave parameters must be given.
1191
1192.. code-block:: console
1193
1194   testpmd> port attach net_bond_0,mode=0,slave=1
1195   Attaching a new port...
1196   EAL: Initializing pmd_bond for net_bond_0
1197   EAL: Create bonded device net_bond_0 on port 0 in mode 0 on socket 0.
1198   Port 0 is attached. Now total ports is 1
1199   Done
1200
1201
1202port detach
1203~~~~~~~~~~~
1204
1205Detach a specific port::
1206
1207   testpmd> port detach (port_id)
1208
1209Before detaching a port, the port should be stopped and closed.
1210
1211For example, to detach a pci device port 0.
1212
1213.. code-block:: console
1214
1215   testpmd> port stop 0
1216   Stopping ports...
1217   Done
1218   testpmd> port close 0
1219   Closing ports...
1220   Done
1221
1222   testpmd> port detach 0
1223   Detaching a port...
1224   EAL: PCI device 0000:0a:00.0 on NUMA socket -1
1225   EAL:   remove driver: 8086:10fb rte_ixgbe_pmd
1226   EAL:   PCI memory unmapped at 0x7f83bfa00000
1227   EAL:   PCI memory unmapped at 0x7f83bfa80000
1228   Done
1229
1230
1231For example, to detach a virtual device port 0.
1232
1233.. code-block:: console
1234
1235   testpmd> port stop 0
1236   Stopping ports...
1237   Done
1238   testpmd> port close 0
1239   Closing ports...
1240   Done
1241
1242   testpmd> port detach 0
1243   Detaching a port...
1244   PMD: Closing pcap ethdev on numa socket 0
1245   Port 'net_pcap0' is detached. Now total ports is 0
1246   Done
1247
1248To remove a pci device completely from the system, first detach the port from testpmd.
1249Then the device should be moved under kernel management.
1250Finally the device can be removed using kernel pci hotplug functionality.
1251
1252For example, to move a pci device under kernel management:
1253
1254.. code-block:: console
1255
1256   sudo ./usertools/dpdk-devbind.py -b ixgbe 0000:0a:00.0
1257
1258   ./usertools/dpdk-devbind.py --status
1259
1260   Network devices using DPDK-compatible driver
1261   ============================================
1262   <none>
1263
1264   Network devices using kernel driver
1265   ===================================
1266   0000:0a:00.0 '82599ES 10-Gigabit' if=eth2 drv=ixgbe unused=igb_uio
1267
1268To remove a port created by a virtual device, above steps are not needed.
1269
1270port start
1271~~~~~~~~~~
1272
1273Start all ports or a specific port::
1274
1275   testpmd> port start (port_id|all)
1276
1277port stop
1278~~~~~~~~~
1279
1280Stop all ports or a specific port::
1281
1282   testpmd> port stop (port_id|all)
1283
1284port close
1285~~~~~~~~~~
1286
1287Close all ports or a specific port::
1288
1289   testpmd> port close (port_id|all)
1290
1291port start/stop queue
1292~~~~~~~~~~~~~~~~~~~~~
1293
1294Start/stop a rx/tx queue on a specific port::
1295
1296   testpmd> port (port_id) (rxq|txq) (queue_id) (start|stop)
1297
1298Only take effect when port is started.
1299
1300port config - speed
1301~~~~~~~~~~~~~~~~~~~
1302
1303Set the speed and duplex mode for all ports or a specific port::
1304
1305   testpmd> port config (port_id|all) speed (10|100|1000|10000|25000|40000|50000|100000|auto) \
1306            duplex (half|full|auto)
1307
1308port config - queues/descriptors
1309~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1310
1311Set number of queues/descriptors for rxq, txq, rxd and txd::
1312
1313   testpmd> port config all (rxq|txq|rxd|txd) (value)
1314
1315This is equivalent to the ``--rxq``, ``--txq``, ``--rxd`` and ``--txd`` command-line options.
1316
1317port config - max-pkt-len
1318~~~~~~~~~~~~~~~~~~~~~~~~~
1319
1320Set the maximum packet length::
1321
1322   testpmd> port config all max-pkt-len (value)
1323
1324This is equivalent to the ``--max-pkt-len`` command-line option.
1325
1326port config - CRC Strip
1327~~~~~~~~~~~~~~~~~~~~~~~
1328
1329Set hardware CRC stripping on or off for all ports::
1330
1331   testpmd> port config all crc-strip (on|off)
1332
1333CRC stripping is off by default.
1334
1335The ``on`` option is equivalent to the ``--crc-strip`` command-line option.
1336
1337port config - scatter
1338~~~~~~~~~~~~~~~~~~~~~~~
1339
1340Set RX scatter mode on or off for all ports::
1341
1342   testpmd> port config all scatter (on|off)
1343
1344RX scatter mode is off by default.
1345
1346The ``on`` option is equivalent to the ``--enable-scatter`` command-line option.
1347
1348port config - TX queue flags
1349~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1350
1351Set a hexadecimal bitmap of TX queue flags for all ports::
1352
1353   testpmd> port config all txqflags value
1354
1355This command is equivalent to the ``--txqflags`` command-line option.
1356
1357port config - RX Checksum
1358~~~~~~~~~~~~~~~~~~~~~~~~~
1359
1360Set hardware RX checksum offload to on or off for all ports::
1361
1362   testpmd> port config all rx-cksum (on|off)
1363
1364Checksum offload is off by default.
1365
1366The ``on`` option is equivalent to the ``--enable-rx-cksum`` command-line option.
1367
1368port config - VLAN
1369~~~~~~~~~~~~~~~~~~
1370
1371Set hardware VLAN on or off for all ports::
1372
1373   testpmd> port config all hw-vlan (on|off)
1374
1375Hardware VLAN is on by default.
1376
1377The ``off`` option is equivalent to the ``--disable-hw-vlan`` command-line option.
1378
1379port config - VLAN filter
1380~~~~~~~~~~~~~~~~~~~~~~~~~
1381
1382Set hardware VLAN filter on or off for all ports::
1383
1384   testpmd> port config all hw-vlan-filter (on|off)
1385
1386Hardware VLAN filter is on by default.
1387
1388The ``off`` option is equivalent to the ``--disable-hw-vlan-filter`` command-line option.
1389
1390port config - VLAN strip
1391~~~~~~~~~~~~~~~~~~~~~~~~
1392
1393Set hardware VLAN strip on or off for all ports::
1394
1395   testpmd> port config all hw-vlan-strip (on|off)
1396
1397Hardware VLAN strip is on by default.
1398
1399The ``off`` option is equivalent to the ``--disable-hw-vlan-strip`` command-line option.
1400
1401port config - VLAN extend
1402~~~~~~~~~~~~~~~~~~~~~~~~~
1403
1404Set hardware VLAN extend on or off for all ports::
1405
1406   testpmd> port config all hw-vlan-extend (on|off)
1407
1408Hardware VLAN extend is off by default.
1409
1410The ``off`` option is equivalent to the ``--disable-hw-vlan-extend`` command-line option.
1411
1412port config - Drop Packets
1413~~~~~~~~~~~~~~~~~~~~~~~~~~
1414
1415Set packet drop for packets with no descriptors on or off for all ports::
1416
1417   testpmd> port config all drop-en (on|off)
1418
1419Packet dropping for packets with no descriptors is off by default.
1420
1421The ``on`` option is equivalent to the ``--enable-drop-en`` command-line option.
1422
1423port config - RSS
1424~~~~~~~~~~~~~~~~~
1425
1426Set the RSS (Receive Side Scaling) mode on or off::
1427
1428   testpmd> port config all rss (all|ip|tcp|udp|sctp|ether|port|vxlan|geneve|nvgre|none)
1429
1430RSS is on by default.
1431
1432The ``none`` option is equivalent to the ``--disable-rss`` command-line option.
1433
1434port config - RSS Reta
1435~~~~~~~~~~~~~~~~~~~~~~
1436
1437Set the RSS (Receive Side Scaling) redirection table::
1438
1439   testpmd> port config all rss reta (hash,queue)[,(hash,queue)]
1440
1441port config - DCB
1442~~~~~~~~~~~~~~~~~
1443
1444Set the DCB mode for an individual port::
1445
1446   testpmd> port config (port_id) dcb vt (on|off) (traffic_class) pfc (on|off)
1447
1448The traffic class should be 4 or 8.
1449
1450port config - Burst
1451~~~~~~~~~~~~~~~~~~~
1452
1453Set the number of packets per burst::
1454
1455   testpmd> port config all burst (value)
1456
1457This is equivalent to the ``--burst`` command-line option.
1458
1459port config - Threshold
1460~~~~~~~~~~~~~~~~~~~~~~~
1461
1462Set thresholds for TX/RX queues::
1463
1464   testpmd> port config all (threshold) (value)
1465
1466Where the threshold type can be:
1467
1468* ``txpt:`` Set the prefetch threshold register of the TX rings, 0 <= value <= 255.
1469
1470* ``txht:`` Set the host threshold register of the TX rings, 0 <= value <= 255.
1471
1472* ``txwt:`` Set the write-back threshold register of the TX rings, 0 <= value <= 255.
1473
1474* ``rxpt:`` Set the prefetch threshold register of the RX rings, 0 <= value <= 255.
1475
1476* ``rxht:`` Set the host threshold register of the RX rings, 0 <= value <= 255.
1477
1478* ``rxwt:`` Set the write-back threshold register of the RX rings, 0 <= value <= 255.
1479
1480* ``txfreet:`` Set the transmit free threshold of the TX rings, 0 <= value <= txd.
1481
1482* ``rxfreet:`` Set the transmit free threshold of the RX rings, 0 <= value <= rxd.
1483
1484* ``txrst:`` Set the transmit RS bit threshold of TX rings, 0 <= value <= txd.
1485
1486These threshold options are also available from the command-line.
1487
1488port config - E-tag
1489~~~~~~~~~~~~~~~~~~~
1490
1491Set the value of ether-type for E-tag::
1492
1493   testpmd> port config (port_id|all) l2-tunnel E-tag ether-type (value)
1494
1495Enable/disable the E-tag support::
1496
1497   testpmd> port config (port_id|all) l2-tunnel E-tag (enable|disable)
1498
1499
1500Link Bonding Functions
1501----------------------
1502
1503The Link Bonding functions make it possible to dynamically create and
1504manage link bonding devices from within testpmd interactive prompt.
1505
1506create bonded device
1507~~~~~~~~~~~~~~~~~~~~
1508
1509Create a new bonding device::
1510
1511   testpmd> create bonded device (mode) (socket)
1512
1513For example, to create a bonded device in mode 1 on socket 0::
1514
1515   testpmd> create bonded 1 0
1516   created new bonded device (port X)
1517
1518add bonding slave
1519~~~~~~~~~~~~~~~~~
1520
1521Adds Ethernet device to a Link Bonding device::
1522
1523   testpmd> add bonding slave (slave id) (port id)
1524
1525For example, to add Ethernet device (port 6) to a Link Bonding device (port 10)::
1526
1527   testpmd> add bonding slave 6 10
1528
1529
1530remove bonding slave
1531~~~~~~~~~~~~~~~~~~~~
1532
1533Removes an Ethernet slave device from a Link Bonding device::
1534
1535   testpmd> remove bonding slave (slave id) (port id)
1536
1537For example, to remove Ethernet slave device (port 6) to a Link Bonding device (port 10)::
1538
1539   testpmd> remove bonding slave 6 10
1540
1541set bonding mode
1542~~~~~~~~~~~~~~~~
1543
1544Set the Link Bonding mode of a Link Bonding device::
1545
1546   testpmd> set bonding mode (value) (port id)
1547
1548For example, to set the bonding mode of a Link Bonding device (port 10) to broadcast (mode 3)::
1549
1550   testpmd> set bonding mode 3 10
1551
1552set bonding primary
1553~~~~~~~~~~~~~~~~~~~
1554
1555Set an Ethernet slave device as the primary device on a Link Bonding device::
1556
1557   testpmd> set bonding primary (slave id) (port id)
1558
1559For example, to set the Ethernet slave device (port 6) as the primary port of a Link Bonding device (port 10)::
1560
1561   testpmd> set bonding primary 6 10
1562
1563set bonding mac
1564~~~~~~~~~~~~~~~
1565
1566Set the MAC address of a Link Bonding device::
1567
1568   testpmd> set bonding mac (port id) (mac)
1569
1570For example, to set the MAC address of a Link Bonding device (port 10) to 00:00:00:00:00:01::
1571
1572   testpmd> set bonding mac 10 00:00:00:00:00:01
1573
1574set bonding xmit_balance_policy
1575~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1576
1577Set the transmission policy for a Link Bonding device when it is in Balance XOR mode::
1578
1579   testpmd> set bonding xmit_balance_policy (port_id) (l2|l23|l34)
1580
1581For example, set a Link Bonding device (port 10) to use a balance policy of layer 3+4 (IP addresses & UDP ports)::
1582
1583   testpmd> set bonding xmit_balance_policy 10 l34
1584
1585
1586set bonding mon_period
1587~~~~~~~~~~~~~~~~~~~~~~
1588
1589Set the link status monitoring polling period in milliseconds for a bonding device.
1590
1591This adds support for PMD slave devices which do not support link status interrupts.
1592When the mon_period is set to a value greater than 0 then all PMD's which do not support
1593link status ISR will be queried every polling interval to check if their link status has changed::
1594
1595   testpmd> set bonding mon_period (port_id) (value)
1596
1597For example, to set the link status monitoring polling period of bonded device (port 5) to 150ms::
1598
1599   testpmd> set bonding mon_period 5 150
1600
1601
1602show bonding config
1603~~~~~~~~~~~~~~~~~~~
1604
1605Show the current configuration of a Link Bonding device::
1606
1607   testpmd> show bonding config (port id)
1608
1609For example,
1610to show the configuration a Link Bonding device (port 9) with 3 slave devices (1, 3, 4)
1611in balance mode with a transmission policy of layer 2+3::
1612
1613   testpmd> show bonding config 9
1614        Bonding mode: 2
1615        Balance Xmit Policy: BALANCE_XMIT_POLICY_LAYER23
1616        Slaves (3): [1 3 4]
1617        Active Slaves (3): [1 3 4]
1618        Primary: [3]
1619
1620
1621Register Functions
1622------------------
1623
1624The Register Functions can be used to read from and write to registers on the network card referenced by a port number.
1625This is mainly useful for debugging purposes.
1626Reference should be made to the appropriate datasheet for the network card for details on the register addresses
1627and fields that can be accessed.
1628
1629read reg
1630~~~~~~~~
1631
1632Display the value of a port register::
1633
1634   testpmd> read reg (port_id) (address)
1635
1636For example, to examine the Flow Director control register (FDIRCTL, 0x0000EE000) on an Intel 82599 10 GbE Controller::
1637
1638   testpmd> read reg 0 0xEE00
1639   port 0 PCI register at offset 0xEE00: 0x4A060029 (1241907241)
1640
1641read regfield
1642~~~~~~~~~~~~~
1643
1644Display a port register bit field::
1645
1646   testpmd> read regfield (port_id) (address) (bit_x) (bit_y)
1647
1648For example, reading the lowest two bits from the register in the example above::
1649
1650   testpmd> read regfield 0 0xEE00 0 1
1651   port 0 PCI register at offset 0xEE00: bits[0, 1]=0x1 (1)
1652
1653read regbit
1654~~~~~~~~~~~
1655
1656Display a single port register bit::
1657
1658   testpmd> read regbit (port_id) (address) (bit_x)
1659
1660For example, reading the lowest bit from the register in the example above::
1661
1662   testpmd> read regbit 0 0xEE00 0
1663   port 0 PCI register at offset 0xEE00: bit 0=1
1664
1665write reg
1666~~~~~~~~~
1667
1668Set the value of a port register::
1669
1670   testpmd> write reg (port_id) (address) (value)
1671
1672For example, to clear a register::
1673
1674   testpmd> write reg 0 0xEE00 0x0
1675   port 0 PCI register at offset 0xEE00: 0x00000000 (0)
1676
1677write regfield
1678~~~~~~~~~~~~~~
1679
1680Set bit field of a port register::
1681
1682   testpmd> write regfield (port_id) (address) (bit_x) (bit_y) (value)
1683
1684For example, writing to the register cleared in the example above::
1685
1686   testpmd> write regfield 0 0xEE00 0 1 2
1687   port 0 PCI register at offset 0xEE00: 0x00000002 (2)
1688
1689write regbit
1690~~~~~~~~~~~~
1691
1692Set single bit value of a port register::
1693
1694   testpmd> write regbit (port_id) (address) (bit_x) (value)
1695
1696For example, to set the high bit in the register from the example above::
1697
1698   testpmd> write regbit 0 0xEE00 31 1
1699   port 0 PCI register at offset 0xEE00: 0x8000000A (2147483658)
1700
1701
1702Filter Functions
1703----------------
1704
1705This section details the available filter functions that are available.
1706
1707Note these functions interface the deprecated legacy filtering framework,
1708superseded by *rte_flow*. See `Flow rules management`_.
1709
1710ethertype_filter
1711~~~~~~~~~~~~~~~~~~~~
1712
1713Add or delete a L2 Ethertype filter, which identify packets by their L2 Ethertype mainly assign them to a receive queue::
1714
1715   ethertype_filter (port_id) (add|del) (mac_addr|mac_ignr) (mac_address) \
1716                    ethertype (ether_type) (drop|fwd) queue (queue_id)
1717
1718The available information parameters are:
1719
1720* ``port_id``: The port which the Ethertype filter assigned on.
1721
1722* ``mac_addr``: Compare destination mac address.
1723
1724* ``mac_ignr``: Ignore destination mac address match.
1725
1726* ``mac_address``: Destination mac address to match.
1727
1728* ``ether_type``: The EtherType value want to match,
1729  for example 0x0806 for ARP packet. 0x0800 (IPv4) and 0x86DD (IPv6) are invalid.
1730
1731* ``queue_id``: The receive queue associated with this EtherType filter.
1732  It is meaningless when deleting or dropping.
1733
1734Example, to add/remove an ethertype filter rule::
1735
1736   testpmd> ethertype_filter 0 add mac_ignr 00:11:22:33:44:55 \
1737                             ethertype 0x0806 fwd queue 3
1738
1739   testpmd> ethertype_filter 0 del mac_ignr 00:11:22:33:44:55 \
1740                             ethertype 0x0806 fwd queue 3
1741
17422tuple_filter
1743~~~~~~~~~~~~~~~~~
1744
1745Add or delete a 2-tuple filter,
1746which identifies packets by specific protocol and destination TCP/UDP port
1747and forwards packets into one of the receive queues::
1748
1749   2tuple_filter (port_id) (add|del) dst_port (dst_port_value) \
1750                 protocol (protocol_value) mask (mask_value) \
1751                 tcp_flags (tcp_flags_value) priority (prio_value) \
1752                 queue (queue_id)
1753
1754The available information parameters are:
1755
1756* ``port_id``: The port which the 2-tuple filter assigned on.
1757
1758* ``dst_port_value``: Destination port in L4.
1759
1760* ``protocol_value``: IP L4 protocol.
1761
1762* ``mask_value``: Participates in the match or not by bit for field above, 1b means participate.
1763
1764* ``tcp_flags_value``: TCP control bits. The non-zero value is invalid, when the pro_value is not set to 0x06 (TCP).
1765
1766* ``prio_value``: Priority of this filter.
1767
1768* ``queue_id``: The receive queue associated with this 2-tuple filter.
1769
1770Example, to add/remove an 2tuple filter rule::
1771
1772   testpmd> 2tuple_filter 0 add dst_port 32 protocol 0x06 mask 0x03 \
1773                          tcp_flags 0x02 priority 3 queue 3
1774
1775   testpmd> 2tuple_filter 0 del dst_port 32 protocol 0x06 mask 0x03 \
1776                          tcp_flags 0x02 priority 3 queue 3
1777
17785tuple_filter
1779~~~~~~~~~~~~~~~~~
1780
1781Add or delete a 5-tuple filter,
1782which consists of a 5-tuple (protocol, source and destination IP addresses, source and destination TCP/UDP/SCTP port)
1783and routes packets into one of the receive queues::
1784
1785   5tuple_filter (port_id) (add|del) dst_ip (dst_address) src_ip \
1786                 (src_address) dst_port (dst_port_value) \
1787                 src_port (src_port_value) protocol (protocol_value) \
1788                 mask (mask_value) tcp_flags (tcp_flags_value) \
1789                 priority (prio_value) queue (queue_id)
1790
1791The available information parameters are:
1792
1793* ``port_id``: The port which the 5-tuple filter assigned on.
1794
1795* ``dst_address``: Destination IP address.
1796
1797* ``src_address``: Source IP address.
1798
1799* ``dst_port_value``: TCP/UDP destination port.
1800
1801* ``src_port_value``: TCP/UDP source port.
1802
1803* ``protocol_value``: L4 protocol.
1804
1805* ``mask_value``: Participates in the match or not by bit for field above, 1b means participate
1806
1807* ``tcp_flags_value``: TCP control bits. The non-zero value is invalid, when the protocol_value is not set to 0x06 (TCP).
1808
1809* ``prio_value``: The priority of this filter.
1810
1811* ``queue_id``: The receive queue associated with this 5-tuple filter.
1812
1813Example, to add/remove an 5tuple filter rule::
1814
1815   testpmd> 5tuple_filter 0 add dst_ip 2.2.2.5 src_ip 2.2.2.4 \
1816            dst_port 64 src_port 32 protocol 0x06 mask 0x1F \
1817            flags 0x0 priority 3 queue 3
1818
1819   testpmd> 5tuple_filter 0 del dst_ip 2.2.2.5 src_ip 2.2.2.4 \
1820            dst_port 64 src_port 32 protocol 0x06 mask 0x1F \
1821            flags 0x0 priority 3 queue 3
1822
1823syn_filter
1824~~~~~~~~~~
1825
1826Using the  SYN filter, TCP packets whose *SYN* flag is set can be forwarded to a separate queue::
1827
1828   syn_filter (port_id) (add|del) priority (high|low) queue (queue_id)
1829
1830The available information parameters are:
1831
1832* ``port_id``: The port which the SYN filter assigned on.
1833
1834* ``high``: This SYN filter has higher priority than other filters.
1835
1836* ``low``: This SYN filter has lower priority than other filters.
1837
1838* ``queue_id``: The receive queue associated with this SYN filter
1839
1840Example::
1841
1842   testpmd> syn_filter 0 add priority high queue 3
1843
1844flex_filter
1845~~~~~~~~~~~
1846
1847With flex filter, packets can be recognized by any arbitrary pattern within the first 128 bytes of the packet
1848and routed into one of the receive queues::
1849
1850   flex_filter (port_id) (add|del) len (len_value) bytes (bytes_value) \
1851               mask (mask_value) priority (prio_value) queue (queue_id)
1852
1853The available information parameters are:
1854
1855* ``port_id``: The port which the Flex filter is assigned on.
1856
1857* ``len_value``: Filter length in bytes, no greater than 128.
1858
1859* ``bytes_value``: A string in hexadecimal, means the value the flex filter needs to match.
1860
1861* ``mask_value``: A string in hexadecimal, bit 1 means corresponding byte participates in the match.
1862
1863* ``prio_value``: The priority of this filter.
1864
1865* ``queue_id``: The receive queue associated with this Flex filter.
1866
1867Example::
1868
1869   testpmd> flex_filter 0 add len 16 bytes 0x00000000000000000000000008060000 \
1870                          mask 000C priority 3 queue 3
1871
1872   testpmd> flex_filter 0 del len 16 bytes 0x00000000000000000000000008060000 \
1873                          mask 000C priority 3 queue 3
1874
1875
1876.. _testpmd_flow_director:
1877
1878flow_director_filter
1879~~~~~~~~~~~~~~~~~~~~
1880
1881The Flow Director works in receive mode to identify specific flows or sets of flows and route them to specific queues.
1882
1883Four types of filtering are supported which are referred to as Perfect Match, Signature, Perfect-mac-vlan and
1884Perfect-tunnel filters, the match mode is set by the ``--pkt-filter-mode`` command-line parameter:
1885
1886* Perfect match filters.
1887  The hardware checks a match between the masked fields of the received packets and the programmed filters.
1888  The masked fields are for IP flow.
1889
1890* Signature filters.
1891  The hardware checks a match between a hash-based signature of the masked fields of the received packet.
1892
1893* Perfect-mac-vlan match filters.
1894  The hardware checks a match between the masked fields of the received packets and the programmed filters.
1895  The masked fields are for MAC VLAN flow.
1896
1897* Perfect-tunnel match filters.
1898  The hardware checks a match between the masked fields of the received packets and the programmed filters.
1899  The masked fields are for tunnel flow.
1900
1901The Flow Director filters can match the different fields for different type of packet: flow type, specific input set
1902per flow type and the flexible payload.
1903
1904The Flow Director can also mask out parts of all of these fields so that filters
1905are only applied to certain fields or parts of the fields.
1906
1907Different NICs may have different capabilities, command show port fdir (port_id) can be used to acquire the information.
1908
1909# Commands to add flow director filters of different flow types::
1910
1911   flow_director_filter (port_id) mode IP (add|del|update) \
1912                        flow (ipv4-other|ipv4-frag|ipv6-other|ipv6-frag) \
1913                        src (src_ip_address) dst (dst_ip_address) \
1914                        tos (tos_value) proto (proto_value) ttl (ttl_value) \
1915                        vlan (vlan_value) flexbytes (flexbytes_value) \
1916                        (drop|fwd) pf|vf(vf_id) queue (queue_id) \
1917                        fd_id (fd_id_value)
1918
1919   flow_director_filter (port_id) mode IP (add|del|update) \
1920                        flow (ipv4-tcp|ipv4-udp|ipv6-tcp|ipv6-udp) \
1921                        src (src_ip_address) (src_port) \
1922                        dst (dst_ip_address) (dst_port) \
1923                        tos (tos_value) ttl (ttl_value) \
1924                        vlan (vlan_value) flexbytes (flexbytes_value) \
1925                        (drop|fwd) queue pf|vf(vf_id) (queue_id) \
1926                        fd_id (fd_id_value)
1927
1928   flow_director_filter (port_id) mode IP (add|del|update) \
1929                        flow (ipv4-sctp|ipv6-sctp) \
1930                        src (src_ip_address) (src_port) \
1931                        dst (dst_ip_address) (dst_port) \
1932                        tos (tos_value) ttl (ttl_value) \
1933                        tag (verification_tag) vlan (vlan_value) \
1934                        flexbytes (flexbytes_value) (drop|fwd) \
1935                        pf|vf(vf_id) queue (queue_id) fd_id (fd_id_value)
1936
1937   flow_director_filter (port_id) mode IP (add|del|update) flow l2_payload \
1938                        ether (ethertype) flexbytes (flexbytes_value) \
1939                        (drop|fwd) pf|vf(vf_id) queue (queue_id)
1940                        fd_id (fd_id_value)
1941
1942   flow_director_filter (port_id) mode MAC-VLAN (add|del|update) \
1943                        mac (mac_address) vlan (vlan_value) \
1944                        flexbytes (flexbytes_value) (drop|fwd) \
1945                        queue (queue_id) fd_id (fd_id_value)
1946
1947   flow_director_filter (port_id) mode Tunnel (add|del|update) \
1948                        mac (mac_address) vlan (vlan_value) \
1949                        tunnel (NVGRE|VxLAN) tunnel-id (tunnel_id_value) \
1950                        flexbytes (flexbytes_value) (drop|fwd) \
1951                        queue (queue_id) fd_id (fd_id_value)
1952
1953For example, to add an ipv4-udp flow type filter::
1954
1955   testpmd> flow_director_filter 0 mode IP add flow ipv4-udp src 2.2.2.3 32 \
1956            dst 2.2.2.5 33 tos 2 ttl 40 vlan 0x1 flexbytes (0x88,0x48) \
1957            fwd pf queue 1 fd_id 1
1958
1959For example, add an ipv4-other flow type filter::
1960
1961   testpmd> flow_director_filter 0 mode IP add flow ipv4-other src 2.2.2.3 \
1962             dst 2.2.2.5 tos 2 proto 20 ttl 40 vlan 0x1 \
1963             flexbytes (0x88,0x48) fwd pf queue 1 fd_id 1
1964
1965flush_flow_director
1966~~~~~~~~~~~~~~~~~~~
1967
1968Flush all flow director filters on a device::
1969
1970   testpmd> flush_flow_director (port_id)
1971
1972Example, to flush all flow director filter on port 0::
1973
1974   testpmd> flush_flow_director 0
1975
1976flow_director_mask
1977~~~~~~~~~~~~~~~~~~
1978
1979Set flow director's input masks::
1980
1981   flow_director_mask (port_id) mode IP vlan (vlan_value) \
1982                      src_mask (ipv4_src) (ipv6_src) (src_port) \
1983                      dst_mask (ipv4_dst) (ipv6_dst) (dst_port)
1984
1985   flow_director_mask (port_id) mode MAC-VLAN vlan (vlan_value)
1986
1987   flow_director_mask (port_id) mode Tunnel vlan (vlan_value) \
1988                      mac (mac_value) tunnel-type (tunnel_type_value) \
1989                      tunnel-id (tunnel_id_value)
1990
1991Example, to set flow director mask on port 0::
1992
1993   testpmd> flow_director_mask 0 mode IP vlan 0xefff \
1994            src_mask 255.255.255.255 \
1995                FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF \
1996            dst_mask 255.255.255.255 \
1997                FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF 0xFFFF
1998
1999flow_director_flex_mask
2000~~~~~~~~~~~~~~~~~~~~~~~
2001
2002set masks of flow director's flexible payload based on certain flow type::
2003
2004   testpmd> flow_director_flex_mask (port_id) \
2005            flow (none|ipv4-other|ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
2006                  ipv6-other|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp| \
2007                  l2_payload|all) (mask)
2008
2009Example, to set flow director's flex mask for all flow type on port 0::
2010
2011   testpmd> flow_director_flex_mask 0 flow all \
2012            (0xff,0xff,0,0,0,0,0,0,0,0,0,0,0,0,0,0)
2013
2014
2015flow_director_flex_payload
2016~~~~~~~~~~~~~~~~~~~~~~~~~~
2017
2018Configure flexible payload selection::
2019
2020   flow_director_flex_payload (port_id) (raw|l2|l3|l4) (config)
2021
2022For example, to select the first 16 bytes from the offset 4 (bytes) of packet's payload as flexible payload::
2023
2024   testpmd> flow_director_flex_payload 0 l4 \
2025            (4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19)
2026
2027get_sym_hash_ena_per_port
2028~~~~~~~~~~~~~~~~~~~~~~~~~
2029
2030Get symmetric hash enable configuration per port::
2031
2032   get_sym_hash_ena_per_port (port_id)
2033
2034For example, to get symmetric hash enable configuration of port 1::
2035
2036   testpmd> get_sym_hash_ena_per_port 1
2037
2038set_sym_hash_ena_per_port
2039~~~~~~~~~~~~~~~~~~~~~~~~~
2040
2041Set symmetric hash enable configuration per port to enable or disable::
2042
2043   set_sym_hash_ena_per_port (port_id) (enable|disable)
2044
2045For example, to set symmetric hash enable configuration of port 1 to enable::
2046
2047   testpmd> set_sym_hash_ena_per_port 1 enable
2048
2049get_hash_global_config
2050~~~~~~~~~~~~~~~~~~~~~~
2051
2052Get the global configurations of hash filters::
2053
2054   get_hash_global_config (port_id)
2055
2056For example, to get the global configurations of hash filters of port 1::
2057
2058   testpmd> get_hash_global_config 1
2059
2060set_hash_global_config
2061~~~~~~~~~~~~~~~~~~~~~~
2062
2063Set the global configurations of hash filters::
2064
2065   set_hash_global_config (port_id) (toeplitz|simple_xor|default) \
2066   (ipv4|ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp|ipv4-other|ipv6|ipv6-frag| \
2067   ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other|l2_payload) \
2068   (enable|disable)
2069
2070For example, to enable simple_xor for flow type of ipv6 on port 2::
2071
2072   testpmd> set_hash_global_config 2 simple_xor ipv6 enable
2073
2074set_hash_input_set
2075~~~~~~~~~~~~~~~~~~
2076
2077Set the input set for hash::
2078
2079   set_hash_input_set (port_id) (ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
2080   ipv4-other|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other| \
2081   l2_payload) (ovlan|ivlan|src-ipv4|dst-ipv4|src-ipv6|dst-ipv6|ipv4-tos| \
2082   ipv4-proto|ipv6-tc|ipv6-next-header|udp-src-port|udp-dst-port| \
2083   tcp-src-port|tcp-dst-port|sctp-src-port|sctp-dst-port|sctp-veri-tag| \
2084   udp-key|gre-key|fld-1st|fld-2nd|fld-3rd|fld-4th|fld-5th|fld-6th|fld-7th| \
2085   fld-8th|none) (select|add)
2086
2087For example, to add source IP to hash input set for flow type of ipv4-udp on port 0::
2088
2089   testpmd> set_hash_input_set 0 ipv4-udp src-ipv4 add
2090
2091set_fdir_input_set
2092~~~~~~~~~~~~~~~~~~
2093
2094The Flow Director filters can match the different fields for different type of packet, i.e. specific input set
2095on per flow type and the flexible payload. This command can be used to change input set for each flow type.
2096
2097Set the input set for flow director::
2098
2099   set_fdir_input_set (port_id) (ipv4-frag|ipv4-tcp|ipv4-udp|ipv4-sctp| \
2100   ipv4-other|ipv6|ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other| \
2101   l2_payload) (ivlan|ethertype|src-ipv4|dst-ipv4|src-ipv6|dst-ipv6|ipv4-tos| \
2102   ipv4-proto|ipv4-ttl|ipv6-tc|ipv6-next-header|ipv6-hop-limits| \
2103   tudp-src-port|udp-dst-port|cp-src-port|tcp-dst-port|sctp-src-port| \
2104   sctp-dst-port|sctp-veri-tag|none) (select|add)
2105
2106For example to add source IP to FD input set for flow type of ipv4-udp on port 0::
2107
2108   testpmd> set_fdir_input_set 0 ipv4-udp src-ipv4 add
2109
2110global_config
2111~~~~~~~~~~~~~
2112
2113Set different GRE key length for input set::
2114
2115   global_config (port_id) gre-key-len (number in bytes)
2116
2117For example to set GRE key length for input set to 4 bytes on port 0::
2118
2119   testpmd> global_config 0 gre-key-len 4
2120
2121
2122.. _testpmd_rte_flow:
2123
2124Flow rules management
2125---------------------
2126
2127Control of the generic flow API (*rte_flow*) is fully exposed through the
2128``flow`` command (validation, creation, destruction and queries).
2129
2130Considering *rte_flow* overlaps with all `Filter Functions`_, using both
2131features simultaneously may cause undefined side-effects and is therefore
2132not recommended.
2133
2134``flow`` syntax
2135~~~~~~~~~~~~~~~
2136
2137Because the ``flow`` command uses dynamic tokens to handle the large number
2138of possible flow rules combinations, its behavior differs slightly from
2139other commands, in particular:
2140
2141- Pressing *?* or the *<tab>* key displays contextual help for the current
2142  token, not that of the entire command.
2143
2144- Optional and repeated parameters are supported (provided they are listed
2145  in the contextual help).
2146
2147The first parameter stands for the operation mode. Possible operations and
2148their general syntax are described below. They are covered in detail in the
2149following sections.
2150
2151- Check whether a flow rule can be created::
2152
2153   flow validate {port_id}
2154       [group {group_id}] [priority {level}] [ingress] [egress]
2155       pattern {item} [/ {item} [...]] / end
2156       actions {action} [/ {action} [...]] / end
2157
2158- Create a flow rule::
2159
2160   flow create {port_id}
2161       [group {group_id}] [priority {level}] [ingress] [egress]
2162       pattern {item} [/ {item} [...]] / end
2163       actions {action} [/ {action} [...]] / end
2164
2165- Destroy specific flow rules::
2166
2167   flow destroy {port_id} rule {rule_id} [...]
2168
2169- Destroy all flow rules::
2170
2171   flow flush {port_id}
2172
2173- Query an existing flow rule::
2174
2175   flow query {port_id} {rule_id} {action}
2176
2177- List existing flow rules sorted by priority, filtered by group
2178  identifiers::
2179
2180   flow list {port_id} [group {group_id}] [...]
2181
2182Validating flow rules
2183~~~~~~~~~~~~~~~~~~~~~
2184
2185``flow validate`` reports whether a flow rule would be accepted by the
2186underlying device in its current state but stops short of creating it. It is
2187bound to ``rte_flow_validate()``::
2188
2189   flow validate {port_id}
2190      [group {group_id}] [priority {level}] [ingress] [egress]
2191      pattern {item} [/ {item} [...]] / end
2192      actions {action} [/ {action} [...]] / end
2193
2194If successful, it will show::
2195
2196   Flow rule validated
2197
2198Otherwise it will show an error message of the form::
2199
2200   Caught error type [...] ([...]): [...]
2201
2202This command uses the same parameters as ``flow create``, their format is
2203described in `Creating flow rules`_.
2204
2205Check whether redirecting any Ethernet packet received on port 0 to RX queue
2206index 6 is supported::
2207
2208   testpmd> flow validate 0 ingress pattern eth / end
2209      actions queue index 6 / end
2210   Flow rule validated
2211   testpmd>
2212
2213Port 0 does not support TCPv6 rules::
2214
2215   testpmd> flow validate 0 ingress pattern eth / ipv6 / tcp / end
2216      actions drop / end
2217   Caught error type 9 (specific pattern item): Invalid argument
2218   testpmd>
2219
2220Creating flow rules
2221~~~~~~~~~~~~~~~~~~~
2222
2223``flow create`` validates and creates the specified flow rule. It is bound
2224to ``rte_flow_create()``::
2225
2226   flow create {port_id}
2227      [group {group_id}] [priority {level}] [ingress] [egress]
2228      pattern {item} [/ {item} [...]] / end
2229      actions {action} [/ {action} [...]] / end
2230
2231If successful, it will return a flow rule ID usable with other commands::
2232
2233   Flow rule #[...] created
2234
2235Otherwise it will show an error message of the form::
2236
2237   Caught error type [...] ([...]): [...]
2238
2239Parameters describe in the following order:
2240
2241- Attributes (*group*, *priority*, *ingress*, *egress* tokens).
2242- A matching pattern, starting with the *pattern* token and terminated by an
2243  *end* pattern item.
2244- Actions, starting with the *actions* token and terminated by an *end*
2245  action.
2246
2247These translate directly to *rte_flow* objects provided as-is to the
2248underlying functions.
2249
2250The shortest valid definition only comprises mandatory tokens::
2251
2252   testpmd> flow create 0 pattern end actions end
2253
2254Note that PMDs may refuse rules that essentially do nothing such as this
2255one.
2256
2257**All unspecified object values are automatically initialized to 0.**
2258
2259Attributes
2260^^^^^^^^^^
2261
2262These tokens affect flow rule attributes (``struct rte_flow_attr``) and are
2263specified before the ``pattern`` token.
2264
2265- ``group {group id}``: priority group.
2266- ``priority {level}``: priority level within group.
2267- ``ingress``: rule applies to ingress traffic.
2268- ``egress``: rule applies to egress traffic.
2269
2270Each instance of an attribute specified several times overrides the previous
2271value as shown below (group 4 is used)::
2272
2273   testpmd> flow create 0 group 42 group 24 group 4 [...]
2274
2275Note that once enabled, ``ingress`` and ``egress`` cannot be disabled.
2276
2277While not specifying a direction is an error, some rules may allow both
2278simultaneously.
2279
2280Most rules affect RX therefore contain the ``ingress`` token::
2281
2282   testpmd> flow create 0 ingress pattern [...]
2283
2284Matching pattern
2285^^^^^^^^^^^^^^^^
2286
2287A matching pattern starts after the ``pattern`` token. It is made of pattern
2288items and is terminated by a mandatory ``end`` item.
2289
2290Items are named after their type (*RTE_FLOW_ITEM_TYPE_* from ``enum
2291rte_flow_item_type``).
2292
2293The ``/`` token is used as a separator between pattern items as shown
2294below::
2295
2296   testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end [...]
2297
2298Note that protocol items like these must be stacked from lowest to highest
2299layer to make sense. For instance, the following rule is either invalid or
2300unlikely to match any packet::
2301
2302   testpmd> flow create 0 ingress pattern eth / udp / ipv4 / end [...]
2303
2304More information on these restrictions can be found in the *rte_flow*
2305documentation.
2306
2307Several items support additional specification structures, for example
2308``ipv4`` allows specifying source and destination addresses as follows::
2309
2310   testpmd> flow create 0 ingress pattern eth / ipv4 src is 10.1.1.1
2311      dst is 10.2.0.0 / end [...]
2312
2313This rule matches all IPv4 traffic with the specified properties.
2314
2315In this example, ``src`` and ``dst`` are field names of the underlying
2316``struct rte_flow_item_ipv4`` object. All item properties can be specified
2317in a similar fashion.
2318
2319The ``is`` token means that the subsequent value must be matched exactly,
2320and assigns ``spec`` and ``mask`` fields in ``struct rte_flow_item``
2321accordingly. Possible assignment tokens are:
2322
2323- ``is``: match value perfectly (with full bit-mask).
2324- ``spec``: match value according to configured bit-mask.
2325- ``last``: specify upper bound to establish a range.
2326- ``mask``: specify bit-mask with relevant bits set to one.
2327- ``prefix``: generate bit-mask from a prefix length.
2328
2329These yield identical results::
2330
2331   ipv4 src is 10.1.1.1
2332
2333::
2334
2335   ipv4 src spec 10.1.1.1 src mask 255.255.255.255
2336
2337::
2338
2339   ipv4 src spec 10.1.1.1 src prefix 32
2340
2341::
2342
2343   ipv4 src is 10.1.1.1 src last 10.1.1.1 # range with a single value
2344
2345::
2346
2347   ipv4 src is 10.1.1.1 src last 0 # 0 disables range
2348
2349Inclusive ranges can be defined with ``last``::
2350
2351   ipv4 src is 10.1.1.1 src last 10.2.3.4 # 10.1.1.1 to 10.2.3.4
2352
2353Note that ``mask`` affects both ``spec`` and ``last``::
2354
2355   ipv4 src is 10.1.1.1 src last 10.2.3.4 src mask 255.255.0.0
2356      # matches 10.1.0.0 to 10.2.255.255
2357
2358Properties can be modified multiple times::
2359
2360   ipv4 src is 10.1.1.1 src is 10.1.2.3 src is 10.2.3.4 # matches 10.2.3.4
2361
2362::
2363
2364   ipv4 src is 10.1.1.1 src prefix 24 src prefix 16 # matches 10.1.0.0/16
2365
2366Pattern items
2367^^^^^^^^^^^^^
2368
2369This section lists supported pattern items and their attributes, if any.
2370
2371- ``end``: end list of pattern items.
2372
2373- ``void``: no-op pattern item.
2374
2375- ``invert``: perform actions when pattern does not match.
2376
2377- ``any``: match any protocol for the current layer.
2378
2379  - ``num {unsigned}``: number of layers covered.
2380
2381- ``pf``: match packets addressed to the physical function.
2382
2383- ``vf``: match packets addressed to a virtual function ID.
2384
2385  - ``id {unsigned}``: destination VF ID.
2386
2387- ``port``: device-specific physical port index to use.
2388
2389  - ``index {unsigned}``: physical port index.
2390
2391- ``raw``: match an arbitrary byte string.
2392
2393  - ``relative {boolean}``: look for pattern after the previous item.
2394  - ``search {boolean}``: search pattern from offset (see also limit).
2395  - ``offset {integer}``: absolute or relative offset for pattern.
2396  - ``limit {unsigned}``: search area limit for start of pattern.
2397  - ``pattern {string}``: byte string to look for.
2398
2399- ``eth``: match Ethernet header.
2400
2401  - ``dst {MAC-48}``: destination MAC.
2402  - ``src {MAC-48}``: source MAC.
2403  - ``type {unsigned}``: EtherType.
2404
2405- ``vlan``: match 802.1Q/ad VLAN tag.
2406
2407  - ``tpid {unsigned}``: tag protocol identifier.
2408  - ``tci {unsigned}``: tag control information.
2409  - ``pcp {unsigned}``: priority code point.
2410  - ``dei {unsigned}``: drop eligible indicator.
2411  - ``vid {unsigned}``: VLAN identifier.
2412
2413- ``ipv4``: match IPv4 header.
2414
2415  - ``tos {unsigned}``: type of service.
2416  - ``ttl {unsigned}``: time to live.
2417  - ``proto {unsigned}``: next protocol ID.
2418  - ``src {ipv4 address}``: source address.
2419  - ``dst {ipv4 address}``: destination address.
2420
2421- ``ipv6``: match IPv6 header.
2422
2423  - ``tc {unsigned}``: traffic class.
2424  - ``flow {unsigned}``: flow label.
2425  - ``proto {unsigned}``: protocol (next header).
2426  - ``hop {unsigned}``: hop limit.
2427  - ``src {ipv6 address}``: source address.
2428  - ``dst {ipv6 address}``: destination address.
2429
2430- ``icmp``: match ICMP header.
2431
2432  - ``type {unsigned}``: ICMP packet type.
2433  - ``code {unsigned}``: ICMP packet code.
2434
2435- ``udp``: match UDP header.
2436
2437  - ``src {unsigned}``: UDP source port.
2438  - ``dst {unsigned}``: UDP destination port.
2439
2440- ``tcp``: match TCP header.
2441
2442  - ``src {unsigned}``: TCP source port.
2443  - ``dst {unsigned}``: TCP destination port.
2444
2445- ``sctp``: match SCTP header.
2446
2447  - ``src {unsigned}``: SCTP source port.
2448  - ``dst {unsigned}``: SCTP destination port.
2449  - ``tag {unsigned}``: validation tag.
2450  - ``cksum {unsigned}``: checksum.
2451
2452- ``vxlan``: match VXLAN header.
2453
2454  - ``vni {unsigned}``: VXLAN identifier.
2455
2456Actions list
2457^^^^^^^^^^^^
2458
2459A list of actions starts after the ``actions`` token in the same fashion as
2460`Matching pattern`_; actions are separated by ``/`` tokens and the list is
2461terminated by a mandatory ``end`` action.
2462
2463Actions are named after their type (*RTE_FLOW_ACTION_TYPE_* from ``enum
2464rte_flow_action_type``).
2465
2466Dropping all incoming UDPv4 packets can be expressed as follows::
2467
2468   testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
2469      actions drop / end
2470
2471Several actions have configurable properties which must be specified when
2472there is no valid default value. For example, ``queue`` requires a target
2473queue index.
2474
2475This rule redirects incoming UDPv4 traffic to queue index 6::
2476
2477   testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
2478      actions queue index 6 / end
2479
2480While this one could be rejected by PMDs (unspecified queue index)::
2481
2482   testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end
2483      actions queue / end
2484
2485As defined by *rte_flow*, the list is not ordered, all actions of a given
2486rule are performed simultaneously. These are equivalent::
2487
2488   queue index 6 / void / mark id 42 / end
2489
2490::
2491
2492   void / mark id 42 / queue index 6 / end
2493
2494All actions in a list should have different types, otherwise only the last
2495action of a given type is taken into account::
2496
2497   queue index 4 / queue index 5 / queue index 6 / end # will use queue 6
2498
2499::
2500
2501   drop / drop / drop / end # drop is performed only once
2502
2503::
2504
2505   mark id 42 / queue index 3 / mark id 24 / end # mark will be 24
2506
2507Considering they are performed simultaneously, opposite and overlapping
2508actions can sometimes be combined when the end result is unambiguous::
2509
2510   drop / queue index 6 / end # drop has no effect
2511
2512::
2513
2514   drop / dup index 6 / end # same as above
2515
2516::
2517
2518   queue index 6 / rss queues 6 7 8 / end # queue has no effect
2519
2520::
2521
2522   drop / passthru / end # drop has no effect
2523
2524Note that PMDs may still refuse such combinations.
2525
2526Actions
2527^^^^^^^
2528
2529This section lists supported actions and their attributes, if any.
2530
2531- ``end``: end list of actions.
2532
2533- ``void``: no-op action.
2534
2535- ``passthru``: let subsequent rule process matched packets.
2536
2537- ``mark``: attach 32 bit value to packets.
2538
2539  - ``id {unsigned}``: 32 bit value to return with packets.
2540
2541- ``flag``: flag packets.
2542
2543- ``queue``: assign packets to a given queue index.
2544
2545  - ``index {unsigned}``: queue index to use.
2546
2547- ``drop``: drop packets (note: passthru has priority).
2548
2549- ``count``: enable counters for this rule.
2550
2551- ``dup``: duplicate packets to a given queue index.
2552
2553  - ``index {unsigned}``: queue index to duplicate packets to.
2554
2555- ``rss``: spread packets among several queues.
2556
2557  - ``queues [{unsigned} [...]] end``: queue indices to use.
2558
2559- ``pf``: redirect packets to physical device function.
2560
2561- ``vf``: redirect packets to virtual device function.
2562
2563  - ``original {boolean}``: use original VF ID if possible.
2564  - ``id {unsigned}``: VF ID to redirect packets to.
2565
2566Destroying flow rules
2567~~~~~~~~~~~~~~~~~~~~~
2568
2569``flow destroy`` destroys one or more rules from their rule ID (as returned
2570by ``flow create``), this command calls ``rte_flow_destroy()`` as many
2571times as necessary::
2572
2573   flow destroy {port_id} rule {rule_id} [...]
2574
2575If successful, it will show::
2576
2577   Flow rule #[...] destroyed
2578
2579It does not report anything for rule IDs that do not exist. The usual error
2580message is shown when a rule cannot be destroyed::
2581
2582   Caught error type [...] ([...]): [...]
2583
2584``flow flush`` destroys all rules on a device and does not take extra
2585arguments. It is bound to ``rte_flow_flush()``::
2586
2587   flow flush {port_id}
2588
2589Any errors are reported as above.
2590
2591Creating several rules and destroying them::
2592
2593   testpmd> flow create 0 ingress pattern eth / ipv6 / end
2594      actions queue index 2 / end
2595   Flow rule #0 created
2596   testpmd> flow create 0 ingress pattern eth / ipv4 / end
2597      actions queue index 3 / end
2598   Flow rule #1 created
2599   testpmd> flow destroy 0 rule 0 rule 1
2600   Flow rule #1 destroyed
2601   Flow rule #0 destroyed
2602   testpmd>
2603
2604The same result can be achieved using ``flow flush``::
2605
2606   testpmd> flow create 0 ingress pattern eth / ipv6 / end
2607      actions queue index 2 / end
2608   Flow rule #0 created
2609   testpmd> flow create 0 ingress pattern eth / ipv4 / end
2610      actions queue index 3 / end
2611   Flow rule #1 created
2612   testpmd> flow flush 0
2613   testpmd>
2614
2615Non-existent rule IDs are ignored::
2616
2617   testpmd> flow create 0 ingress pattern eth / ipv6 / end
2618      actions queue index 2 / end
2619   Flow rule #0 created
2620   testpmd> flow create 0 ingress pattern eth / ipv4 / end
2621      actions queue index 3 / end
2622   Flow rule #1 created
2623   testpmd> flow destroy 0 rule 42 rule 10 rule 2
2624   testpmd>
2625   testpmd> flow destroy 0 rule 0
2626   Flow rule #0 destroyed
2627   testpmd>
2628
2629Querying flow rules
2630~~~~~~~~~~~~~~~~~~~
2631
2632``flow query`` queries a specific action of a flow rule having that
2633ability. Such actions collect information that can be reported using this
2634command. It is bound to ``rte_flow_query()``::
2635
2636   flow query {port_id} {rule_id} {action}
2637
2638If successful, it will display either the retrieved data for known actions
2639or the following message::
2640
2641   Cannot display result for action type [...] ([...])
2642
2643Otherwise, it will complain either that the rule does not exist or that some
2644error occurred::
2645
2646   Flow rule #[...] not found
2647
2648::
2649
2650   Caught error type [...] ([...]): [...]
2651
2652Currently only the ``count`` action is supported. This action reports the
2653number of packets that hit the flow rule and the total number of bytes. Its
2654output has the following format::
2655
2656   count:
2657    hits_set: [...] # whether "hits" contains a valid value
2658    bytes_set: [...] # whether "bytes" contains a valid value
2659    hits: [...] # number of packets
2660    bytes: [...] # number of bytes
2661
2662Querying counters for TCPv6 packets redirected to queue 6::
2663
2664   testpmd> flow create 0 ingress pattern eth / ipv6 / tcp / end
2665      actions queue index 6 / count / end
2666   Flow rule #4 created
2667   testpmd> flow query 0 4 count
2668   count:
2669    hits_set: 1
2670    bytes_set: 0
2671    hits: 386446
2672    bytes: 0
2673   testpmd>
2674
2675Listing flow rules
2676~~~~~~~~~~~~~~~~~~
2677
2678``flow list`` lists existing flow rules sorted by priority and optionally
2679filtered by group identifiers::
2680
2681   flow list {port_id} [group {group_id}] [...]
2682
2683This command only fails with the following message if the device does not
2684exist::
2685
2686   Invalid port [...]
2687
2688Output consists of a header line followed by a short description of each
2689flow rule, one per line. There is no output at all when no flow rules are
2690configured on the device::
2691
2692   ID      Group   Prio    Attr    Rule
2693   [...]   [...]   [...]   [...]   [...]
2694
2695``Attr`` column flags:
2696
2697- ``i`` for ``ingress``.
2698- ``e`` for ``egress``.
2699
2700Creating several flow rules and listing them::
2701
2702   testpmd> flow create 0 ingress pattern eth / ipv4 / end
2703      actions queue index 6 / end
2704   Flow rule #0 created
2705   testpmd> flow create 0 ingress pattern eth / ipv6 / end
2706      actions queue index 2 / end
2707   Flow rule #1 created
2708   testpmd> flow create 0 priority 5 ingress pattern eth / ipv4 / udp / end
2709      actions rss queues 6 7 8 end / end
2710   Flow rule #2 created
2711   testpmd> flow list 0
2712   ID      Group   Prio    Attr    Rule
2713   0       0       0       i-      ETH IPV4 => QUEUE
2714   1       0       0       i-      ETH IPV6 => QUEUE
2715   2       0       5       i-      ETH IPV4 UDP => RSS
2716   testpmd>
2717
2718Rules are sorted by priority (i.e. group ID first, then priority level)::
2719
2720   testpmd> flow list 1
2721   ID      Group   Prio    Attr    Rule
2722   0       0       0       i-      ETH => COUNT
2723   6       0       500     i-      ETH IPV6 TCP => DROP COUNT
2724   5       0       1000    i-      ETH IPV6 ICMP => QUEUE
2725   1       24      0       i-      ETH IPV4 UDP => QUEUE
2726   4       24      10      i-      ETH IPV4 TCP => DROP
2727   3       24      20      i-      ETH IPV4 => DROP
2728   2       24      42      i-      ETH IPV4 UDP => QUEUE
2729   7       63      0       i-      ETH IPV6 UDP VXLAN => MARK QUEUE
2730   testpmd>
2731
2732Output can be limited to specific groups::
2733
2734   testpmd> flow list 1 group 0 group 63
2735   ID      Group   Prio    Attr    Rule
2736   0       0       0       i-      ETH => COUNT
2737   6       0       500     i-      ETH IPV6 TCP => DROP COUNT
2738   5       0       1000    i-      ETH IPV6 ICMP => QUEUE
2739   7       63      0       i-      ETH IPV6 UDP VXLAN => MARK QUEUE
2740   testpmd>
2741