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 The following are the EAL command-line options that can be used in conjunction with the testpmd,
11 or any other DPDK application.
12 See the DPDK Getting Started Guides for more information on these options.
16 Set the hexadecimal bitmask of the cores to run on.
20 List of cores to run on
22 The argument format is ``<c1>[-c2][,c3[-c4],...]``
23 where ``c1``, ``c2``, etc are core indexes between 0 and 128.
25 * ``--lcores COREMAP``
27 Map lcore set to physical cpu set
29 The argument format is::
31 <lcores[@cpus]>[<,lcores[@cpus]>...]
33 Lcore and CPU lists are grouped by ``(`` and ``)`` Within the group.
34 The ``-`` character is used as a range separator and ``,`` is used as a single number separator.
35 The grouping ``()`` can be omitted for single element group.
36 The ``@`` can be omitted if cpus and lcores have the same value.
39 At a given instance only one core option ``--lcores``, ``-l`` or ``-c`` can be used.
42 * ``--master-lcore ID``
44 Core ID that is used as master.
48 Set the number of memory channels to use.
50 * ``-b, --pci-blacklist domain:bus:devid.func``
52 Blacklist a PCI device to prevent EAL from using it. Multiple -b options are allowed.
56 Load an external driver. Multiple -d options are allowed.
58 * ``-w, --pci-whitelist domain:bus:devid:func``
60 Add a PCI device in white list.
64 Memory to allocate. See also ``--socket-mem``.
68 Set the number of memory ranks (auto-detected by default).
72 Display the version information on startup.
76 Set the syslog facility.
80 Set the memory to allocate on specific sockets (use comma separated values).
84 Specify the directory where the hugetlbfs is mounted.
86 * ``mbuf-pool-ops-name``:
88 Pool ops name for mbuf to use.
92 Set the type of the current process.
96 Prefix for hugepage filenames.
100 Use VMware TSC map instead of native RDTSC.
104 Add a virtual device using the format::
106 <driver><id>[,key=val, ...]
110 --vdev 'net_pcap0,rx_pcap=input.pcap,tx_pcap=output.pcap'
112 * ``--base-virtaddr``
114 Specify base virtual address.
116 * ``--create-uio-dev``
118 Create ``/dev/uioX`` (usually done by hotplug).
122 No shared config (mmap-ed files).
134 Use malloc instead of hugetlbfs.
137 Testpmd Command-line Options
138 ----------------------------
140 The following are the command-line options for the testpmd applications.
141 They must be separated from the EAL options, shown in the previous section, with a ``--`` separator:
143 .. code-block:: console
145 sudo ./testpmd -l 0-3 -n 4 -- -i --portmask=0x1 --nb-cores=2
147 The commandline options are:
149 * ``-i, --interactive``
151 Run testpmd in interactive mode.
152 In this mode, the testpmd starts with a prompt that can be used to start and stop forwarding,
153 configure the application and display stats on the current packet processing session.
154 See :ref:`testpmd_runtime` for more details.
156 In non-interactive mode,
157 the application starts with the configuration specified on the command-line and
158 immediately enters forwarding mode.
162 Display a help message and quit.
164 * ``-a, --auto-start``
166 Start forwarding on initialization.
170 Start forwarding, after sending a burst of packets first.
174 This flag should be only used in non-interactive mode.
176 * ``--stats-period PERIOD``
178 Display statistics every PERIOD seconds, if interactive mode is disabled.
179 The default value is 0, which means that the statistics will not be displayed.
183 Set the number of forwarding cores,
184 where 1 <= N <= "number of cores" or ``CONFIG_RTE_MAX_LCORE`` from the configuration file.
185 The default value is 1.
189 Set the number of forwarding ports,
190 where 1 <= N <= "number of ports" on the board or ``CONFIG_RTE_MAX_ETHPORTS`` from the configuration file.
191 The default value is the number of ports on the board.
193 * ``--coremask=0xXX``
195 Set the hexadecimal bitmask of the cores running the packet forwarding test.
196 The master lcore is reserved for command line parsing only and cannot be masked on for packet forwarding.
198 * ``--portmask=0xXX``
200 Set the hexadecimal bitmask of the ports used by the packet forwarding test.
204 Enable NUMA-aware allocation of RX/TX rings and of RX memory buffers
205 (mbufs). [Default setting]
209 Disable NUMA-aware allocation of RX/TX rings and of RX memory buffers (mbufs).
211 * ``--port-numa-config=(port,socket)[,(port,socket)]``
213 Specify the socket on which the memory pool to be used by the port will be allocated.
215 * ``--ring-numa-config=(port,flag,socket)[,(port,flag,socket)]``
217 Specify the socket on which the TX/RX rings for the port will be allocated.
218 Where flag is 1 for RX, 2 for TX, and 3 for RX and TX.
222 Set the socket from which all memory is allocated in NUMA mode,
223 where 0 <= N < number of sockets on the board.
227 Set the data size of the mbufs used to N bytes, where N < 65536. The default value is 2048.
229 * ``--total-num-mbufs=N``
231 Set the number of mbufs to be allocated in the mbuf pools, where N > 1024.
233 * ``--max-pkt-len=N``
235 Set the maximum packet size to N bytes, where N >= 64. The default value is 1518.
237 * ``--eth-peers-configfile=name``
239 Use a configuration file containing the Ethernet addresses of the peer ports.
240 The configuration file should contain the Ethernet addresses on separate lines::
247 * ``--eth-peer=N,XX:XX:XX:XX:XX:XX``
249 Set the MAC address ``XX:XX:XX:XX:XX:XX`` of the peer port N,
250 where 0 <= N < ``CONFIG_RTE_MAX_ETHPORTS`` from the configuration file.
252 * ``--pkt-filter-mode=mode``
254 Set Flow Director mode where mode is either ``none`` (the default), ``signature`` or ``perfect``.
255 See :ref:`testpmd_flow_director` for more details.
257 * ``--pkt-filter-report-hash=mode``
259 Set Flow Director hash match reporting mode where mode is ``none``, ``match`` (the default) or ``always``.
261 * ``--pkt-filter-size=N``
263 Set Flow Director allocated memory size, where N is 64K, 128K or 256K.
264 Sizes are in kilobytes. The default is 64.
266 * ``--pkt-filter-flexbytes-offset=N``
268 Set the flexbytes offset.
269 The offset is defined in words (not bytes) counted from the first byte of the destination Ethernet MAC address,
270 where N is 0 <= N <= 32.
271 The default value is 0x6.
273 * ``--pkt-filter-drop-queue=N``
276 In perfect filter mode, when a rule is added with queue = -1, the packet will be enqueued into the RX drop-queue.
277 If the drop-queue does not exist, the packet is dropped. The default value is N=127.
279 * ``--disable-crc-strip``
281 Disable hardware CRC stripping.
285 Enable large receive offload.
287 * ``--enable-rx-cksum``
289 Enable hardware RX checksum offload.
291 * ``--enable-scatter``
293 Enable scatter (multi-segment) RX.
295 * ``--enable-hw-vlan``
297 Enable hardware VLAN.
299 * ``--enable-hw-vlan-filter``
301 Enable hardware VLAN filter.
303 * ``--enable-hw-vlan-strip``
305 Enable hardware VLAN strip.
307 * ``--enable-hw-vlan-extend``
309 Enable hardware VLAN extend.
311 * ``--enable-drop-en``
313 Enable per-queue packet drop for packets with no descriptors.
317 Disable RSS (Receive Side Scaling).
319 * ``--port-topology=mode``
321 Set port topology, where mode is ``paired`` (the default) or ``chained``.
323 In ``paired`` mode, the forwarding is between pairs of ports, for example: (0,1), (2,3), (4,5).
325 In ``chained`` mode, the forwarding is to the next available port in the port mask, for example: (0,1), (1,2), (2,0).
327 The ordering of the ports can be changed using the portlist testpmd runtime function.
329 * ``--forward-mode=mode``
331 Set the forwarding mode where ``mode`` is one of the following::
346 Set RSS functions for IPv4/IPv6 only.
350 Set RSS functions for IPv4/IPv6 and UDP.
354 Set the number of RX queues per port to N, where 1 <= N <= 65535.
355 The default value is 1.
359 Set the number of descriptors in the RX rings to N, where N > 0.
360 The default value is 128.
364 Set the number of TX queues per port to N, where 1 <= N <= 65535.
365 The default value is 1.
369 Set the number of descriptors in the TX rings to N, where N > 0.
370 The default value is 512.
374 Set the number of packets per burst to N, where 1 <= N <= 512.
375 The default value is 32.
376 If set to 0, driver default is used if defined. Else, if driver
377 default is not defined, default of 32 is used.
381 Set the cache of mbuf memory pools to N, where 0 <= N <= 512.
382 The default value is 16.
386 Set the prefetch threshold register of RX rings to N, where N >= 0.
387 The default value is 8.
391 Set the host threshold register of RX rings to N, where N >= 0.
392 The default value is 8.
396 Set the free threshold of RX descriptors to N, where 0 <= N < value of --rxd.
397 The default value is 0.
401 Set the write-back threshold register of RX rings to N, where N >= 0.
402 The default value is 4.
406 Set the prefetch threshold register of TX rings to N, where N >= 0.
407 The default value is 36.
411 Set the host threshold register of TX rings to N, where N >= 0.
412 The default value is 0.
416 Set the write-back threshold register of TX rings to N, where N >= 0.
417 The default value is 0.
421 Set the transmit free threshold of TX rings to N, where 0 <= N <= value of ``--txd``.
422 The default value is 0.
426 Set the transmit RS bit threshold of TX rings to N, where 0 <= N <= value of ``--txd``.
427 The default value is 0.
429 * ``--rx-queue-stats-mapping=(port,queue,mapping)[,(port,queue,mapping)]``
431 Set the RX queues statistics counters mapping 0 <= mapping <= 15.
433 * ``--tx-queue-stats-mapping=(port,queue,mapping)[,(port,queue,mapping)]``
435 Set the TX queues statistics counters mapping 0 <= mapping <= 15.
439 Don't flush the RX streams before starting forwarding. Used mainly with the PCAP PMD.
443 Set TX segment sizes or total packet length. Valid for ``tx-only``
444 and ``flowgen`` forwarding modes.
446 * ``--disable-link-check``
448 Disable check on link status when starting/stopping ports.
450 * ``--no-lsc-interrupt``
452 Disable LSC interrupts for all ports, even those supporting it.
454 * ``--no-rmv-interrupt``
456 Disable RMV interrupts for all ports, even those supporting it.
458 * ``--bitrate-stats=N``
460 Set the logical core N to perform bitrate calculation.
462 * ``--print-event <unknown|intr_lsc|queue_state|intr_reset|vf_mbox|macsec|intr_rmv|dev_probed|dev_released|all>``
464 Enable printing the occurrence of the designated event. Using all will
467 * ``--mask-event <unknown|intr_lsc|queue_state|intr_reset|vf_mbox|macsec|intr_rmv|dev_probed|dev_released|all>``
469 Disable printing the occurrence of the designated event. Using all will
472 * ``--flow-isolate-all``
474 Providing this parameter requests flow API isolated mode on all ports at
475 initialization time. It ensures all traffic is received through the
476 configured flow rules only (see flow command).
478 Ports that do not support this mode are automatically discarded.
480 * ``--tx-offloads=0xXXXXXXXX``
482 Set the hexadecimal bitmask of TX queue offloads.
483 The default value is 0.
487 Enable device event monitor machenism for hotplug.
489 * ``--vxlan-gpe-port=N``
491 Set the UDP port number of tunnel VXLAN-GPE to N.
492 The default value is 4790.
496 Enable locking all memory.
500 Disable locking all memory.