1 .. SPDX-License-Identifier: BSD-3-Clause
2 Copyright(c) 2010-2014 Intel Corporation.
4 Running the Application
5 =======================
7 EAL Command-line Options
8 ------------------------
10 Please refer to :doc:`EAL parameters (Linux) <../linux_gsg/linux_eal_parameters>`
11 or :doc:`EAL parameters (FreeBSD) <../freebsd_gsg/freebsd_eal_parameters>` for
12 a list of available EAL command-line options.
15 Testpmd Command-line Options
16 ----------------------------
18 The following are the command-line options for the testpmd applications.
19 They must be separated from the EAL options, shown in the previous section, with a ``--`` separator:
21 .. code-block:: console
23 sudo ./dpdk-testpmd -l 0-3 -n 4 -- -i --portmask=0x1 --nb-cores=2
25 The command line options are:
27 * ``-i, --interactive``
29 Run testpmd in interactive mode.
30 In this mode, the testpmd starts with a prompt that can be used to start and stop forwarding,
31 configure the application and display stats on the current packet processing session.
32 See :ref:`testpmd_runtime` for more details.
34 In non-interactive mode,
35 the application starts with the configuration specified on the command-line and
36 immediately enters forwarding mode.
40 Display a help message and quit.
42 * ``-a, --auto-start``
44 Start forwarding on initialization.
48 Start forwarding, after sending a burst of packets first.
52 This flag should be only used in non-interactive mode.
54 * ``--stats-period PERIOD``
56 Display statistics every PERIOD seconds, if interactive mode is disabled.
57 The default value is 0, which means that the statistics will not be displayed.
61 Set the number of forwarding cores,
62 where 1 <= N <= "number of cores" or ``RTE_MAX_LCORE`` from the configuration file.
63 The default value is 1.
67 Set the number of forwarding ports,
68 where 1 <= N <= "number of ports" on the board or ``RTE_MAX_ETHPORTS`` from the configuration file.
69 The default value is the number of ports on the board.
73 Set the hexadecimal bitmask of the cores running the packet forwarding test.
74 The main lcore is reserved for command line parsing only and cannot be masked on for packet forwarding.
78 Set the hexadecimal bitmask of the ports used by the packet forwarding test.
82 Set the forwarding ports based on the user input used by the packet forwarding test.
83 '-' denotes a range of ports to set including the two specified port IDs
84 ',' separates multiple port values.
85 Possible examples like --portlist=0,1 or --portlist=0-2 or --portlist=0,1-2 etc
89 Enable NUMA-aware allocation of RX/TX rings and of RX memory buffers
90 (mbufs). [Default setting]
94 Disable NUMA-aware allocation of RX/TX rings and of RX memory buffers (mbufs).
96 * ``--port-numa-config=(port,socket)[,(port,socket)]``
98 Specify the socket on which the memory pool to be used by the port will be allocated.
100 * ``--ring-numa-config=(port,flag,socket)[,(port,flag,socket)]``
102 Specify the socket on which the TX/RX rings for the port will be allocated.
103 Where flag is 1 for RX, 2 for TX, and 3 for RX and TX.
107 Set the socket from which all memory is allocated in NUMA mode,
108 where 0 <= N < number of sockets on the board.
110 * ``--mbuf-size=N[,N1[,...Nn]``
112 Set the data size of the mbufs used to N bytes, where N < 65536.
113 The default value is 2048. If multiple mbuf-size values are specified the
114 extra memory pools will be created for allocating mbufs to receive packets
115 with buffer splittling features.
117 * ``--total-num-mbufs=N``
119 Set the number of mbufs to be allocated in the mbuf pools, where N > 1024.
121 * ``--max-pkt-len=N``
123 Set the maximum packet size to N bytes, where N >= 64. The default value is 1518.
125 * ``--max-lro-pkt-size=N``
127 Set the maximum LRO aggregated packet size to N bytes, where N >= 64.
129 * ``--eth-peers-configfile=name``
131 Use a configuration file containing the Ethernet addresses of the peer ports.
132 The configuration file should contain the Ethernet addresses on separate lines::
138 * ``--eth-peer=N,XX:XX:XX:XX:XX:XX``
140 Set the MAC address ``XX:XX:XX:XX:XX:XX`` of the peer port N,
141 where 0 <= N < ``RTE_MAX_ETHPORTS``.
143 * ``--tx-ip=SRC,DST``
145 Set the source and destination IP address used when doing transmit only test.
146 The defaults address values are source 198.18.0.1 and
147 destination 198.18.0.2. These are special purpose addresses
148 reserved for benchmarking (RFC 5735).
150 * ``--tx-udp=SRC[,DST]``
152 Set the source and destination UDP port number for transmit test only test.
153 The default port is the port 9 which is defined for the discard protocol
156 * ``--pkt-filter-mode=mode``
158 Set Flow Director mode where mode is either ``none`` (the default), ``signature`` or ``perfect``.
159 See :ref:`testpmd_flow_director` for more details.
161 * ``--pkt-filter-report-hash=mode``
163 Set Flow Director hash match reporting mode where mode is ``none``, ``match`` (the default) or ``always``.
165 * ``--pkt-filter-size=N``
167 Set Flow Director allocated memory size, where N is 64K, 128K or 256K.
168 Sizes are in kilobytes. The default is 64.
170 * ``--pkt-filter-flexbytes-offset=N``
172 Set the flexbytes offset.
173 The offset is defined in words (not bytes) counted from the first byte of the destination Ethernet MAC address,
174 where N is 0 <= N <= 32.
175 The default value is 0x6.
177 * ``--pkt-filter-drop-queue=N``
180 In perfect filter mode, when a rule is added with queue = -1, the packet will be enqueued into the RX drop-queue.
181 If the drop-queue does not exist, the packet is dropped. The default value is N=127.
183 * ``--disable-crc-strip``
185 Disable hardware CRC stripping.
189 Enable large receive offload.
191 * ``--enable-rx-cksum``
193 Enable hardware RX checksum offload.
195 * ``--enable-scatter``
197 Enable scatter (multi-segment) RX.
199 * ``--enable-hw-vlan``
201 Enable hardware VLAN.
203 * ``--enable-hw-vlan-filter``
205 Enable hardware VLAN filter.
207 * ``--enable-hw-vlan-strip``
209 Enable hardware VLAN strip.
211 * ``--enable-hw-vlan-extend``
213 Enable hardware VLAN extend.
215 * ``--enable-hw-qinq-strip``
217 Enable hardware QINQ strip.
219 * ``--enable-drop-en``
221 Enable per-queue packet drop for packets with no descriptors.
225 Disable RSS (Receive Side Scaling).
227 * ``--port-topology=mode``
229 Set port topology, where mode is ``paired`` (the default), ``chained`` or ``loop``.
231 In ``paired`` mode, the forwarding is between pairs of ports, for example: (0,1), (2,3), (4,5).
233 In ``chained`` mode, the forwarding is to the next available port in the port mask, for example: (0,1), (1,2), (2,0).
235 The ordering of the ports can be changed using the portlist testpmd runtime function.
237 In ``loop`` mode, ingress traffic is simply transmitted back on the same interface.
239 * ``--forward-mode=mode``
241 Set the forwarding mode where ``mode`` is one of the following::
258 Set RSS functions for IPv4/IPv6 only.
262 Set RSS functions for IPv4/IPv6 and UDP.
266 Set the number of RX queues per port to N, where 1 <= N <= 65535.
267 The default value is 1.
271 Set the number of descriptors in the RX rings to N, where N > 0.
272 The default value is 128.
276 Set the number of TX queues per port to N, where 1 <= N <= 65535.
277 The default value is 1.
281 Set the number of descriptors in the TX rings to N, where N > 0.
282 The default value is 512.
286 Set the number of hairpin queues per port to N, where 1 <= N <= 65535.
287 The default value is 0. The number of hairpin queues are added to the
288 number of TX queues and to the number of RX queues. then the first
289 RX hairpin is binded to the first TX hairpin, the second RX hairpin is
290 binded to the second TX hairpin and so on. The index of the first
291 RX hairpin queue is the number of RX queues as configured using --rxq.
292 The index of the first TX hairpin queue is the number of TX queues
293 as configured using --txq.
297 Set the number of packets per burst to N, where 1 <= N <= 512.
298 The default value is 32.
299 If set to 0, driver default is used if defined. Else, if driver
300 default is not defined, default of 32 is used.
302 * ``--flowgen-clones=N``
304 Set the number of each packet clones to be sent in `flowgen` mode.
305 Sending clones reduces host CPU load on creating packets and may help
306 in testing extreme speeds or maxing out Tx packet performance.
307 N should be not zero, but less than 'burst' parameter.
311 Set the cache of mbuf memory pools to N, where 0 <= N <= 512.
312 The default value is 16.
316 Set the prefetch threshold register of RX rings to N, where N >= 0.
317 The default value is 8.
321 Set the host threshold register of RX rings to N, where N >= 0.
322 The default value is 8.
326 Set the free threshold of RX descriptors to N, where 0 <= N < value of --rxd.
327 The default value is 0.
331 Set the write-back threshold register of RX rings to N, where N >= 0.
332 The default value is 4.
336 Set the prefetch threshold register of TX rings to N, where N >= 0.
337 The default value is 36.
341 Set the host threshold register of TX rings to N, where N >= 0.
342 The default value is 0.
346 Set the write-back threshold register of TX rings to N, where N >= 0.
347 The default value is 0.
351 Set the transmit free threshold of TX rings to N, where 0 <= N <= value of ``--txd``.
352 The default value is 0.
356 Set the transmit RS bit threshold of TX rings to N, where 0 <= N <= value of ``--txd``.
357 The default value is 0.
359 * ``--rx-queue-stats-mapping=(port,queue,mapping)[,(port,queue,mapping)]``
361 Set the RX queues statistics counters mapping 0 <= mapping <= 15.
363 * ``--tx-queue-stats-mapping=(port,queue,mapping)[,(port,queue,mapping)]``
365 Set the TX queues statistics counters mapping 0 <= mapping <= 15.
369 Don't flush the RX streams before starting forwarding. Used mainly with the PCAP PMD.
373 Set the offsets of packet segments on receiving if split
374 feature is engaged. Affects only the queues configured
375 with split offloads (currently BUFFER_SPLIT is supported only).
379 Set the length of segments to scatter packets on receiving if split
380 feature is engaged. Affects only the queues configured
381 with split offloads (currently BUFFER_SPLIT is supported only).
382 Optionally the multiple memory pools can be specified with --mbuf-size
383 command line parameter and the mbufs to receive will be allocated
384 sequentially from these extra memory pools.
388 Set TX segment sizes or total packet length. Valid for ``tx-only``
389 and ``flowgen`` forwarding modes.
391 * ``--txonly-multi-flow``
393 Generate multiple flows in txonly mode.
395 * ``--eth-link-speed``
397 Set a forced link speed to the ethernet port::
399 10 - 10Mbps (not supported)
400 100 - 100Mbps (not supported)
410 * ``--disable-link-check``
412 Disable check on link status when starting/stopping ports.
414 * ``--disable-device-start``
416 Do not automatically start all ports. This allows testing
417 configuration of rx and tx queues before device is started
420 * ``--no-lsc-interrupt``
422 Disable LSC interrupts for all ports, even those supporting it.
424 * ``--no-rmv-interrupt``
426 Disable RMV interrupts for all ports, even those supporting it.
428 * ``--bitrate-stats=N``
430 Set the logical core N to perform bitrate calculation.
432 * ``--print-event <unknown|intr_lsc|queue_state|intr_reset|vf_mbox|macsec|intr_rmv|dev_probed|dev_released|flow_aged|all>``
434 Enable printing the occurrence of the designated event. Using all will
437 * ``--mask-event <unknown|intr_lsc|queue_state|intr_reset|vf_mbox|macsec|intr_rmv|dev_probed|dev_released|flow_aged|all>``
439 Disable printing the occurrence of the designated event. Using all will
442 * ``--flow-isolate-all``
444 Providing this parameter requests flow API isolated mode on all ports at
445 initialization time. It ensures all traffic is received through the
446 configured flow rules only (see flow command).
448 Ports that do not support this mode are automatically discarded.
450 * ``--tx-offloads=0xXXXXXXXX``
452 Set the hexadecimal bitmask of TX queue offloads.
453 The default value is 0.
455 * ``--rx-offloads=0xXXXXXXXX``
457 Set the hexadecimal bitmask of RX queue offloads.
458 The default value is 0.
462 Enable device event monitor mechanism for hotplug.
464 * ``--vxlan-gpe-port=N``
466 Set the UDP port number of tunnel VXLAN-GPE to N.
467 The default value is 4790.
469 * ``--geneve-parsed-port=N``
471 Set the UDP port number that is used for parsing the GENEVE protocol to N.
472 HW may be configured with another tunnel Geneve port.
473 The default value is 6081.
477 Enable locking all memory.
481 Disable locking all memory.
483 * ``--mp-alloc <native|anon|xmem|xmemhuge>``
485 Select mempool allocation mode:
487 * native: create and populate mempool using native DPDK memory
488 * anon: create mempool using native DPDK memory, but populate using
490 * xmem: create and populate mempool using externally and anonymously
492 * xmemhuge: create and populate mempool using externally and anonymously
493 allocated hugepage area
495 * ``--noisy-tx-sw-buffer-size``
497 Set the number of maximum elements of the FIFO queue to be created
498 for buffering packets. Only available with the noisy forwarding mode.
499 The default value is 0.
501 * ``--noisy-tx-sw-buffer-flushtime=N``
503 Set the time before packets in the FIFO queue is flushed.
504 Only available with the noisy forwarding mode. The default value is 0.
506 * ``--noisy-lkup-memory=N``
508 Set the size of the noisy neighbor simulation memory buffer in MB to N.
509 Only available with the noisy forwarding mode. The default value is 0.
512 * ``--noisy-lkup-num-reads=N``
514 Set the number of reads to be done in noisy neighbor simulation memory buffer to N.
515 Only available with the noisy forwarding mode. The default value is 0.
517 * ``--noisy-lkup-num-writes=N``
519 Set the number of writes to be done in noisy neighbor simulation memory buffer to N.
520 Only available with the noisy forwarding mode. The default value is 0.
522 * ``--noisy-lkup-num-reads-writes=N``
524 Set the number of r/w accesses to be done in noisy neighbor simulation memory buffer to N.
525 Only available with the noisy forwarding mode. The default value is 0.
527 * ``--no-iova-contig``
529 Enable to create mempool which is not IOVA contiguous. Valid only with --mp-alloc=anon.
530 The default value is 0.
534 Set the hexadecimal bitmask of RX multi queue mode which can be enabled.
535 The default value is 0x7::
537 ETH_MQ_RX_RSS_FLAG | ETH_MQ_RX_DCB_FLAG | ETH_MQ_RX_VMDQ_FLAG
539 * ``--record-core-cycles``
541 Enable measurement of CPU cycles per packet.
543 * ``--record-burst-stats``
545 Enable display of RX and TX burst stats.
547 * ``--hairpin-mode=0xXX``
549 Set the hairpin port mode with bitmask, only valid when hairpin queues number is set.
550 bit 4 - explicit Tx flow rule
551 bit 1 - two hairpin ports paired
552 bit 0 - two hairpin ports loop
553 The default value is 0. Hairpin will use single port mode and implicit Tx flow mode.