1 .. SPDX-License-Identifier: BSD-3-Clause
2 Copyright 2015 6WIND S.A.
3 Copyright 2015 Mellanox Technologies, Ltd
5 .. include:: <isonum.txt>
10 The MLX5 poll mode driver library (**librte_net_mlx5**) provides support
11 for **Mellanox ConnectX-4**, **Mellanox ConnectX-4 Lx** , **Mellanox
12 ConnectX-5**, **Mellanox ConnectX-6**, **Mellanox ConnectX-6 Dx**, **Mellanox
13 ConnectX-6 Lx**, **Mellanox BlueField** and **Mellanox BlueField-2** families
14 of 10/25/40/50/100/200 Gb/s adapters as well as their virtual functions (VF)
17 Information and documentation about these adapters can be found on the
18 `Mellanox website <http://www.mellanox.com>`__. Help is also provided by the
19 `Mellanox community <http://community.mellanox.com/welcome>`__.
21 There is also a `section dedicated to this poll mode driver
22 <http://www.mellanox.com/page/products_dyn?product_family=209&mtag=pmd_for_dpdk>`__.
28 Besides its dependency on libibverbs (that implies libmlx5 and associated
29 kernel support), librte_net_mlx5 relies heavily on system calls for control
30 operations such as querying/updating the MTU and flow control parameters.
32 For security reasons and robustness, this driver only deals with virtual
33 memory addresses. The way resources allocations are handled by the kernel,
34 combined with hardware specifications that allow to handle virtual memory
35 addresses directly, ensure that DPDK applications cannot access random
36 physical memory (or memory that does not belong to the current process).
38 This capability allows the PMD to coexist with kernel network interfaces
39 which remain functional, although they stop receiving unicast packets as
40 long as they share the same MAC address.
41 This means legacy linux control tools (for example: ethtool, ifconfig and
42 more) can operate on the same network interfaces that owned by the DPDK
45 The PMD can use libibverbs and libmlx5 to access the device firmware
46 or directly the hardware components.
47 There are different levels of objects and bypassing abilities
48 to get the best performances:
50 - Verbs is a complete high-level generic API
51 - Direct Verbs is a device-specific API
52 - DevX allows to access firmware objects
53 - Direct Rules manages flow steering at low-level hardware layer
55 Enabling librte_net_mlx5 causes DPDK applications to be linked against
61 - Multi arch support: x86_64, POWER8, ARMv8, i686.
62 - Multiple TX and RX queues.
63 - Support for scattered TX frames.
64 - Advanced support for scattered Rx frames with tunable buffer attributes.
65 - IPv4, IPv6, TCPv4, TCPv6, UDPv4 and UDPv6 RSS on any number of queues.
66 - RSS using different combinations of fields: L3 only, L4 only or both,
67 and source only, destination only or both.
68 - Several RSS hash keys, one for each flow type.
69 - Default RSS operation with no hash key specification.
70 - Configurable RETA table.
71 - Link flow control (pause frame).
72 - Support for multiple MAC addresses.
76 - RX CRC stripping configuration.
77 - TX mbuf fast free offload.
78 - Promiscuous mode on PF and VF.
79 - Multicast promiscuous mode on PF and VF.
80 - Hardware checksum offloads.
81 - Flow director (RTE_FDIR_MODE_PERFECT, RTE_FDIR_MODE_PERFECT_MAC_VLAN and
83 - Flow API, including :ref:`flow_isolated_mode`.
85 - KVM and VMware ESX SR-IOV modes are supported.
86 - RSS hash result is supported.
87 - Hardware TSO for generic IP or UDP tunnel, including VXLAN and GRE.
88 - Hardware checksum Tx offload for generic IP or UDP tunnel, including VXLAN and GRE.
90 - Statistics query including Basic, Extended and per queue.
92 - Tunnel types: VXLAN, L3 VXLAN, VXLAN-GPE, GRE, MPLSoGRE, MPLSoUDP, IP-in-IP, Geneve, GTP.
93 - Tunnel HW offloads: packet type, inner/outer RSS, IP and UDP checksum verification.
94 - NIC HW offloads: encapsulation (vxlan, gre, mplsoudp, mplsogre), NAT, routing, TTL
95 increment/decrement, count, drop, mark. For details please see :ref:`mlx5_offloads_support`.
96 - Flow insertion rate of more then million flows per second, when using Direct Rules.
97 - Support for multiple rte_flow groups.
98 - Per packet no-inline hint flag to disable packet data copying into Tx descriptors.
101 - Multiple-thread flow insertion.
102 - Matching on GTP extension header with raw encap/decap action.
103 - Matching on Geneve TLV option header with raw encap/decap action.
104 - RSS support in sample action.
105 - E-Switch mirroring and jump.
106 - E-Switch mirroring and modify.
107 - 21844 flow priorities for ingress or egress flow groups greater than 0 and for any transfer
109 - Flow metering, including meter policy API.
110 - Flow integrity offload API.
111 - Connection tracking.
112 - Sub-Function representors.
119 On Windows, the features are limited:
121 - Promiscuous mode is not supported
122 - The following rules are supported:
124 - IPv4/UDP with CVLAN filtering
125 - Unicast MAC filtering
127 - Additional rules are supported from WinOF2 version 2.70:
129 - IPv4/TCP with CVLAN filtering
130 - L4 steering rules for port RSS of UDP, TCP and IP
132 - For secondary process:
134 - Forked secondary process not supported.
135 - External memory unregistered in EAL memseg list cannot be used for DMA
136 unless such memory has been registered by ``mlx5_mr_update_ext_mp()`` in
137 primary process and remapped to the same virtual address in secondary
138 process. If the external memory is registered by primary process but has
139 different virtual address in secondary process, unexpected error may happen.
141 - When using Verbs flow engine (``dv_flow_en`` = 0), flow pattern without any
142 specific VLAN will match for VLAN packets as well:
144 When VLAN spec is not specified in the pattern, the matching rule will be created with VLAN as a wild card.
145 Meaning, the flow rule::
147 flow create 0 ingress pattern eth / vlan vid is 3 / ipv4 / end ...
149 Will only match vlan packets with vid=3. and the flow rule::
151 flow create 0 ingress pattern eth / ipv4 / end ...
153 Will match any ipv4 packet (VLAN included).
155 - When using Verbs flow engine (``dv_flow_en`` = 0), multi-tagged(QinQ) match is not supported.
157 - When using DV flow engine (``dv_flow_en`` = 1), flow pattern with any VLAN specification will match only single-tagged packets unless the ETH item ``type`` field is 0x88A8 or the VLAN item ``has_more_vlan`` field is 1.
160 flow create 0 ingress pattern eth / ipv4 / end ...
162 Will match any ipv4 packet.
165 flow create 0 ingress pattern eth / vlan / end ...
166 flow create 0 ingress pattern eth has_vlan is 1 / end ...
167 flow create 0 ingress pattern eth type is 0x8100 / end ...
169 Will match single-tagged packets only, with any VLAN ID value.
172 flow create 0 ingress pattern eth type is 0x88A8 / end ...
173 flow create 0 ingress pattern eth / vlan has_more_vlan is 1 / end ...
175 Will match multi-tagged packets only, with any VLAN ID value.
177 - A flow pattern with 2 sequential VLAN items is not supported.
179 - VLAN pop offload command:
181 - Flow rules having a VLAN pop offload command as one of their actions and
182 are lacking a match on VLAN as one of their items are not supported.
183 - The command is not supported on egress traffic in NIC mode.
185 - VLAN push offload is not supported on ingress traffic in NIC mode.
187 - VLAN set PCP offload is not supported on existing headers.
189 - A multi segment packet must have not more segments than reported by dev_infos_get()
190 in tx_desc_lim.nb_seg_max field. This value depends on maximal supported Tx descriptor
191 size and ``txq_inline_min`` settings and may be from 2 (worst case forced by maximal
192 inline settings) to 58.
194 - Flows with a VXLAN Network Identifier equal (or ends to be equal)
195 to 0 are not supported.
197 - L3 VXLAN and VXLAN-GPE tunnels cannot be supported together with MPLSoGRE and MPLSoUDP.
199 - Match on Geneve header supports the following fields only:
206 - Match on Geneve TLV option is supported on the following fields:
213 Only one Class/Type/Length Geneve TLV option is supported per shared device.
214 Class/Type/Length fields must be specified as well as masks.
215 Class/Type/Length specified masks must be full.
216 Matching Geneve TLV option without specifying data is not supported.
217 Matching Geneve TLV option with ``data & mask == 0`` is not supported.
219 - VF: flow rules created on VF devices can only match traffic targeted at the
220 configured MAC addresses (see ``rte_eth_dev_mac_addr_add()``).
222 - Match on GTP tunnel header item supports the following fields only:
224 - v_pt_rsv_flags: E flag, S flag, PN flag
228 - Match on GTP extension header only for GTP PDU session container (next
229 extension header type = 0x85).
230 - Match on GTP extension header is not supported in group 0.
232 - No Tx metadata go to the E-Switch steering domain for the Flow group 0.
233 The flows within group 0 and set metadata action are rejected by hardware.
237 MAC addresses not already present in the bridge table of the associated
238 kernel network device will be added and cleaned up by the PMD when closing
239 the device. In case of ungraceful program termination, some entries may
240 remain present and should be removed manually by other means.
242 - Buffer split offload is supported with regular Rx burst routine only,
243 no MPRQ feature or vectorized code can be engaged.
245 - When Multi-Packet Rx queue is configured (``mprq_en``), a Rx packet can be
246 externally attached to a user-provided mbuf with having EXT_ATTACHED_MBUF in
247 ol_flags. As the mempool for the external buffer is managed by PMD, all the
248 Rx mbufs must be freed before the device is closed. Otherwise, the mempool of
249 the external buffers will be freed by PMD and the application which still
250 holds the external buffers may be corrupted.
252 - If Multi-Packet Rx queue is configured (``mprq_en``) and Rx CQE compression is
253 enabled (``rxq_cqe_comp_en``) at the same time, RSS hash result is not fully
254 supported. Some Rx packets may not have PKT_RX_RSS_HASH.
256 - IPv6 Multicast messages are not supported on VM, while promiscuous mode
257 and allmulticast mode are both set to off.
258 To receive IPv6 Multicast messages on VM, explicitly set the relevant
259 MAC address using rte_eth_dev_mac_addr_add() API.
261 - To support a mixed traffic pattern (some buffers from local host memory, some
262 buffers from other devices) with high bandwidth, a mbuf flag is used.
264 An application hints the PMD whether or not it should try to inline the
265 given mbuf data buffer. PMD should do the best effort to act upon this request.
267 The hint flag ``RTE_PMD_MLX5_FINE_GRANULARITY_INLINE`` is dynamic,
268 registered by application with rte_mbuf_dynflag_register(). This flag is
269 purely driver-specific and declared in PMD specific header ``rte_pmd_mlx5.h``,
270 which is intended to be used by the application.
272 To query the supported specific flags in runtime,
273 the function ``rte_pmd_mlx5_get_dyn_flag_names`` returns the array of
274 currently (over present hardware and configuration) supported specific flags.
275 The "not inline hint" feature operating flow is the following one:
278 - probe the devices, ports are created
279 - query the port capabilities
280 - if port supporting the feature is found
281 - register dynamic flag ``RTE_PMD_MLX5_FINE_GRANULARITY_INLINE``
282 - application starts the ports
283 - on ``dev_start()`` PMD checks whether the feature flag is registered and
284 enables the feature support in datapath
285 - application might set the registered flag bit in ``ol_flags`` field
286 of mbuf being sent and PMD will handle ones appropriately.
288 - The amount of descriptors in Tx queue may be limited by data inline settings.
289 Inline data require the more descriptor building blocks and overall block
290 amount may exceed the hardware supported limits. The application should
291 reduce the requested Tx size or adjust data inline settings with
292 ``txq_inline_max`` and ``txq_inline_mpw`` devargs keys.
294 - To provide the packet send scheduling on mbuf timestamps the ``tx_pp``
295 parameter should be specified.
296 When PMD sees the RTE_MBUF_DYNFLAG_TX_TIMESTAMP_NAME set on the packet
297 being sent it tries to synchronize the time of packet appearing on
298 the wire with the specified packet timestamp. It the specified one
299 is in the past it should be ignored, if one is in the distant future
300 it should be capped with some reasonable value (in range of seconds).
301 These specific cases ("too late" and "distant future") can be optionally
302 reported via device xstats to assist applications to detect the
303 time-related problems.
305 The timestamp upper "too-distant-future" limit
306 at the moment of invoking the Tx burst routine
307 can be estimated as ``tx_pp`` option (in nanoseconds) multiplied by 2^23.
308 Please note, for the testpmd txonly mode,
309 the limit is deduced from the expression::
311 (n_tx_descriptors / burst_size + 1) * inter_burst_gap
313 There is no any packet reordering according timestamps is supposed,
314 neither within packet burst, nor between packets, it is an entirely
315 application responsibility to generate packets and its timestamps
316 in desired order. The timestamps can be put only in the first packet
317 in the burst providing the entire burst scheduling.
319 - E-Switch decapsulation Flow:
321 - can be applied to PF port only.
322 - must specify VF port action (packet redirection from PF to VF).
323 - optionally may specify tunnel inner source and destination MAC addresses.
325 - E-Switch encapsulation Flow:
327 - can be applied to VF ports only.
328 - must specify PF port action (packet redirection from VF to PF).
332 - The input buffer, used as outer header, is not validated.
336 - The decapsulation is always done up to the outermost tunnel detected by the HW.
337 - The input buffer, providing the removal size, is not validated.
338 - The buffer size must match the length of the headers to be removed.
340 - ICMP(code/type/identifier/sequence number) / ICMP6(code/type) matching, IP-in-IP and MPLS flow matching are all
341 mutually exclusive features which cannot be supported together
342 (see :ref:`mlx5_firmware_config`).
346 - Requires DevX and DV flow to be enabled.
347 - KEEP_CRC offload cannot be supported with LRO.
348 - The first mbuf length, without head-room, must be big enough to include the
350 - Rx queue with LRO offload enabled, receiving a non-LRO packet, can forward
351 it with size limited to max LRO size, not to max RX packet length.
352 - LRO can be used with outer header of TCP packets of the standard format:
353 eth (with or without vlan) / ipv4 or ipv6 / tcp / payload
355 Other TCP packets (e.g. with MPLS label) received on Rx queue with LRO enabled, will be received with bad checksum.
356 - LRO packet aggregation is performed by HW only for packet size larger than
357 ``lro_min_mss_size``. This value is reported on device start, when debug
362 - ``DEV_RX_OFFLOAD_KEEP_CRC`` cannot be supported with decapsulation
363 for some NICs (such as ConnectX-6 Dx, ConnectX-6 Lx, and BlueField-2).
364 The capability bit ``scatter_fcs_w_decap_disable`` shows NIC support.
368 - fast free offload assumes the all mbufs being sent are originated from the
369 same memory pool and there is no any extra references to the mbufs (the
370 reference counter for each mbuf is equal 1 on tx_burst call). The latter
371 means there should be no any externally attached buffers in mbufs. It is
372 an application responsibility to provide the correct mbufs if the fast
373 free offload is engaged. The mlx5 PMD implicitly produces the mbufs with
374 externally attached buffers if MPRQ option is enabled, hence, the fast
375 free offload is neither supported nor advertised if there is MPRQ enabled.
379 - Supports ``RTE_FLOW_ACTION_TYPE_SAMPLE`` action only within NIC Rx and
380 E-Switch steering domain.
381 - For E-Switch Sampling flow with sample ratio > 1, additional actions are not
382 supported in the sample actions list.
383 - For ConnectX-5, the ``RTE_FLOW_ACTION_TYPE_SAMPLE`` is typically used as
384 first action in the E-Switch egress flow if with header modify or
385 encapsulation actions.
386 - For NIC Rx flow, supports ``MARK``, ``COUNT``, ``QUEUE``, ``RSS`` in the
388 - For E-Switch mirroring flow, supports ``RAW ENCAP``, ``Port ID``,
389 ``VXLAN ENCAP``, ``NVGRE ENCAP`` in the sample actions list.
393 - Supports the 'set' operation only for ``RTE_FLOW_ACTION_TYPE_MODIFY_FIELD`` action.
394 - Modification of an arbitrary place in a packet via the special ``RTE_FLOW_FIELD_START`` Field ID is not supported.
395 - Modification of the 802.1Q Tag, VXLAN Network or GENEVE Network ID's is not supported.
396 - Encapsulation levels are not supported, can modify outermost header fields only.
397 - Offsets must be 32-bits aligned, cannot skip past the boundary of a field.
399 - IPv6 header item 'proto' field, indicating the next header protocol, should
400 not be set as extension header.
401 In case the next header is an extension header, it should not be specified in
402 IPv6 header item 'proto' field.
403 The last extension header item 'next header' field can specify the following
404 header protocol type.
408 - Hairpin between two ports could only manual binding and explicit Tx flow mode. For single port hairpin, all the combinations of auto/manual binding and explicit/implicit Tx flow mode could be supported.
409 - Hairpin in switchdev SR-IOV mode is not supported till now.
413 - All the meter colors with drop action will be counted only by the global drop statistics.
414 - Green color is not supported with drop action.
415 - Yellow detection is not supported.
416 - Red color must be with drop action.
417 - Meter statistics are supported only for drop case.
418 - Meter yellow color detection is not supported.
419 - A meter action created with pre-defined policy must be the last action in the flow except single case where the policy actions are:
420 - green: NULL or END.
421 - yellow: NULL or END.
423 - The only supported meter policy actions:
424 - green: QUEUE, RSS, PORT_ID, JUMP, MARK and SET_TAG.
425 - yellow: must be empty.
427 - meter profile packet mode is supported.
431 - Integrity offload is enabled for **ConnectX-6** family.
432 - Verification bits provided by the hardware are ``l3_ok``, ``ipv4_csum_ok``, ``l4_ok``, ``l4_csum_ok``.
433 - ``level`` value 0 references outer headers.
434 - Multiple integrity items not supported in a single flow rule.
435 - Flow rule items supplied by application must explicitly specify network headers referred by integrity item.
436 For example, if integrity item mask sets ``l4_ok`` or ``l4_csum_ok`` bits, reference to L4 network header,
437 TCP or UDP, must be in the rule pattern as well::
439 flow create 0 ingress pattern integrity level is 0 value mask l3_ok value spec l3_ok / eth / ipv6 / end …
441 flow create 0 ingress pattern integrity level is 0 value mask l4_ok value spec 0 / eth / ipv4 proto is udp / end …
443 - Connection tracking:
445 - Cannot co-exist with ASO meter, ASO age action in a single flow rule.
446 - Flow rules insertion rate and memory consumption need more optimization.
448 - 4M connections maximum.
453 MLX5 supports various methods to report statistics:
455 Port statistics can be queried using ``rte_eth_stats_get()``. The received and sent statistics are through SW only and counts the number of packets received or sent successfully by the PMD. The imissed counter is the amount of packets that could not be delivered to SW because a queue was full. Packets not received due to congestion in the bus or on the NIC can be queried via the rx_discards_phy xstats counter.
457 Extended statistics can be queried using ``rte_eth_xstats_get()``. The extended statistics expose a wider set of counters counted by the device. The extended port statistics counts the number of packets received or sent successfully by the port. As Mellanox NICs are using the :ref:`Bifurcated Linux Driver <linux_gsg_linux_drivers>` those counters counts also packet received or sent by the Linux kernel. The counters with ``_phy`` suffix counts the total events on the physical port, therefore not valid for VF.
459 Finally per-flow statistics can by queried using ``rte_flow_query`` when attaching a count action for specific flow. The flow counter counts the number of packets received successfully by the port and match the specific flow.
467 The ibverbs libraries can be linked with this PMD in a number of ways,
468 configured by the ``ibverbs_link`` build option:
470 - ``shared`` (default): the PMD depends on some .so files.
472 - ``dlopen``: Split the dependencies glue in a separate library
473 loaded when needed by dlopen.
474 It make dependencies on libibverbs and libmlx4 optional,
475 and has no performance impact.
477 - ``static``: Embed static flavor of the dependencies libibverbs and libmlx4
478 in the PMD shared library or the executable static binary.
480 Environment variables
481 ~~~~~~~~~~~~~~~~~~~~~
485 A list of directories in which to search for the rdma-core "glue" plug-in,
486 separated by colons or semi-colons.
488 - ``MLX5_SHUT_UP_BF``
490 Configures HW Tx doorbell register as IO-mapped.
492 By default, the HW Tx doorbell is configured as a write-combining register.
493 The register would be flushed to HW usually when the write-combining buffer
494 becomes full, but it depends on CPU design.
496 Except for vectorized Tx burst routines, a write memory barrier is enforced
497 after updating the register so that the update can be immediately visible to
500 When vectorized Tx burst is called, the barrier is set only if the burst size
501 is not aligned to MLX5_VPMD_TX_MAX_BURST. However, setting this environmental
502 variable will bring better latency even though the maximum throughput can
505 Run-time configuration
506 ~~~~~~~~~~~~~~~~~~~~~~
508 - librte_net_mlx5 brings kernel network interfaces up during initialization
509 because it is affected by their state. Forcing them down prevents packets
512 - **ethtool** operations on related kernel interfaces also affect the PMD.
517 In order to run as a non-root user,
518 some capabilities must be granted to the application::
520 setcap cap_sys_admin,cap_net_admin,cap_net_raw,cap_ipc_lock+ep <dpdk-app>
522 Below are the reasons of the need for each capability:
525 When using physical addresses (PA mode), with Linux >= 4.0,
526 for access to ``/proc/self/pagemap``.
529 For device configuration.
532 For raw ethernet queue allocation through kernel driver.
535 For DMA memory pinning.
540 - ``rxq_cqe_comp_en`` parameter [int]
542 A nonzero value enables the compression of CQE on RX side. This feature
543 allows to save PCI bandwidth and improve performance. Enabled by default.
544 Different compression formats are supported in order to achieve the best
545 performance for different traffic patterns. Default format depends on
546 Multi-Packet Rx queue configuration: Hash RSS format is used in case
547 MPRQ is disabled, Checksum format is used in case MPRQ is enabled.
549 Specifying 2 as a ``rxq_cqe_comp_en`` value selects Flow Tag format for
550 better compression rate in case of RTE Flow Mark traffic.
551 Specifying 3 as a ``rxq_cqe_comp_en`` value selects Checksum format.
552 Specifying 4 as a ``rxq_cqe_comp_en`` value selects L3/L4 Header format for
553 better compression rate in case of mixed TCP/UDP and IPv4/IPv6 traffic.
554 CQE compression format selection requires DevX to be enabled. If there is
555 no DevX enabled/supported the value is reset to 1 by default.
559 - x86_64 with ConnectX-4, ConnectX-4 Lx, ConnectX-5, ConnectX-6, ConnectX-6 Dx,
560 ConnectX-6 Lx, BlueField and BlueField-2.
561 - POWER9 and ARMv8 with ConnectX-4 Lx, ConnectX-5, ConnectX-6, ConnectX-6 Dx,
562 ConnectX-6 Lx, BlueField and BlueField-2.
564 - ``rxq_pkt_pad_en`` parameter [int]
566 A nonzero value enables padding Rx packet to the size of cacheline on PCI
567 transaction. This feature would waste PCI bandwidth but could improve
568 performance by avoiding partial cacheline write which may cause costly
569 read-modify-copy in memory transaction on some architectures. Disabled by
574 - x86_64 with ConnectX-4, ConnectX-4 Lx, ConnectX-5, ConnectX-6, ConnectX-6 Dx,
575 ConnectX-6 Lx, BlueField and BlueField-2.
576 - POWER8 and ARMv8 with ConnectX-4 Lx, ConnectX-5, ConnectX-6, ConnectX-6 Dx,
577 ConnectX-6 Lx, BlueField and BlueField-2.
579 - ``mprq_en`` parameter [int]
581 A nonzero value enables configuring Multi-Packet Rx queues. Rx queue is
582 configured as Multi-Packet RQ if the total number of Rx queues is
583 ``rxqs_min_mprq`` or more. Disabled by default.
585 Multi-Packet Rx Queue (MPRQ a.k.a Striding RQ) can further save PCIe bandwidth
586 by posting a single large buffer for multiple packets. Instead of posting a
587 buffers per a packet, one large buffer is posted in order to receive multiple
588 packets on the buffer. A MPRQ buffer consists of multiple fixed-size strides
589 and each stride receives one packet. MPRQ can improve throughput for
590 small-packet traffic.
592 When MPRQ is enabled, max_rx_pkt_len can be larger than the size of
593 user-provided mbuf even if DEV_RX_OFFLOAD_SCATTER isn't enabled. PMD will
594 configure large stride size enough to accommodate max_rx_pkt_len as long as
595 device allows. Note that this can waste system memory compared to enabling Rx
596 scatter and multi-segment packet.
598 - ``mprq_log_stride_num`` parameter [int]
600 Log 2 of the number of strides for Multi-Packet Rx queue. Configuring more
601 strides can reduce PCIe traffic further. If configured value is not in the
602 range of device capability, the default value will be set with a warning
603 message. The default value is 4 which is 16 strides per a buffer, valid only
604 if ``mprq_en`` is set.
606 The size of Rx queue should be bigger than the number of strides.
608 - ``mprq_log_stride_size`` parameter [int]
610 Log 2 of the size of a stride for Multi-Packet Rx queue. Configuring a smaller
611 stride size can save some memory and reduce probability of a depletion of all
612 available strides due to unreleased packets by an application. If configured
613 value is not in the range of device capability, the default value will be set
614 with a warning message. The default value is 11 which is 2048 bytes per a
615 stride, valid only if ``mprq_en`` is set. With ``mprq_log_stride_size`` set
616 it is possible for a packet to span across multiple strides. This mode allows
617 support of jumbo frames (9K) with MPRQ. The memcopy of some packets (or part
618 of a packet if Rx scatter is configured) may be required in case there is no
619 space left for a head room at the end of a stride which incurs some
622 - ``mprq_max_memcpy_len`` parameter [int]
624 The maximum length of packet to memcpy in case of Multi-Packet Rx queue. Rx
625 packet is mem-copied to a user-provided mbuf if the size of Rx packet is less
626 than or equal to this parameter. Otherwise, PMD will attach the Rx packet to
627 the mbuf by external buffer attachment - ``rte_pktmbuf_attach_extbuf()``.
628 A mempool for external buffers will be allocated and managed by PMD. If Rx
629 packet is externally attached, ol_flags field of the mbuf will have
630 EXT_ATTACHED_MBUF and this flag must be preserved. ``RTE_MBUF_HAS_EXTBUF()``
631 checks the flag. The default value is 128, valid only if ``mprq_en`` is set.
633 - ``rxqs_min_mprq`` parameter [int]
635 Configure Rx queues as Multi-Packet RQ if the total number of Rx queues is
636 greater or equal to this value. The default value is 12, valid only if
639 - ``txq_inline`` parameter [int]
641 Amount of data to be inlined during TX operations. This parameter is
642 deprecated and converted to the new parameter ``txq_inline_max`` providing
643 partial compatibility.
645 - ``txqs_min_inline`` parameter [int]
647 Enable inline data send only when the number of TX queues is greater or equal
650 This option should be used in combination with ``txq_inline_max`` and
651 ``txq_inline_mpw`` below and does not affect ``txq_inline_min`` settings above.
653 If this option is not specified the default value 16 is used for BlueField
654 and 8 for other platforms
656 The data inlining consumes the CPU cycles, so this option is intended to
657 auto enable inline data if we have enough Tx queues, which means we have
658 enough CPU cores and PCI bandwidth is getting more critical and CPU
659 is not supposed to be bottleneck anymore.
661 The copying data into WQE improves latency and can improve PPS performance
662 when PCI back pressure is detected and may be useful for scenarios involving
663 heavy traffic on many queues.
665 Because additional software logic is necessary to handle this mode, this
666 option should be used with care, as it may lower performance when back
667 pressure is not expected.
669 If inline data are enabled it may affect the maximal size of Tx queue in
670 descriptors because the inline data increase the descriptor size and
671 queue size limits supported by hardware may be exceeded.
673 - ``txq_inline_min`` parameter [int]
675 Minimal amount of data to be inlined into WQE during Tx operations. NICs
676 may require this minimal data amount to operate correctly. The exact value
677 may depend on NIC operation mode, requested offloads, etc. It is strongly
678 recommended to omit this parameter and use the default values. Anyway,
679 applications using this parameter should take into consideration that
680 specifying an inconsistent value may prevent the NIC from sending packets.
682 If ``txq_inline_min`` key is present the specified value (may be aligned
683 by the driver in order not to exceed the limits and provide better descriptor
684 space utilization) will be used by the driver and it is guaranteed that
685 requested amount of data bytes are inlined into the WQE beside other inline
686 settings. This key also may update ``txq_inline_max`` value (default
687 or specified explicitly in devargs) to reserve the space for inline data.
689 If ``txq_inline_min`` key is not present, the value may be queried by the
690 driver from the NIC via DevX if this feature is available. If there is no DevX
691 enabled/supported the value 18 (supposing L2 header including VLAN) is set
692 for ConnectX-4 and ConnectX-4 Lx, and 0 is set by default for ConnectX-5
693 and newer NICs. If packet is shorter the ``txq_inline_min`` value, the entire
696 For ConnectX-4 NIC, driver does not allow specifying value below 18
697 (minimal L2 header, including VLAN), error will be raised.
699 For ConnectX-4 Lx NIC, it is allowed to specify values below 18, but
700 it is not recommended and may prevent NIC from sending packets over
703 Please, note, this minimal data inlining disengages eMPW feature (Enhanced
704 Multi-Packet Write), because last one does not support partial packet inlining.
705 This is not very critical due to minimal data inlining is mostly required
706 by ConnectX-4 and ConnectX-4 Lx, these NICs do not support eMPW feature.
708 - ``txq_inline_max`` parameter [int]
710 Specifies the maximal packet length to be completely inlined into WQE
711 Ethernet Segment for ordinary SEND method. If packet is larger than specified
712 value, the packet data won't be copied by the driver at all, data buffer
713 is addressed with a pointer. If packet length is less or equal all packet
714 data will be copied into WQE. This may improve PCI bandwidth utilization for
715 short packets significantly but requires the extra CPU cycles.
717 The data inline feature is controlled by number of Tx queues, if number of Tx
718 queues is larger than ``txqs_min_inline`` key parameter, the inline feature
719 is engaged, if there are not enough Tx queues (which means not enough CPU cores
720 and CPU resources are scarce), data inline is not performed by the driver.
721 Assigning ``txqs_min_inline`` with zero always enables the data inline.
723 The default ``txq_inline_max`` value is 290. The specified value may be adjusted
724 by the driver in order not to exceed the limit (930 bytes) and to provide better
725 WQE space filling without gaps, the adjustment is reflected in the debug log.
726 Also, the default value (290) may be decreased in run-time if the large transmit
727 queue size is requested and hardware does not support enough descriptor
728 amount, in this case warning is emitted. If ``txq_inline_max`` key is
729 specified and requested inline settings can not be satisfied then error
732 - ``txq_inline_mpw`` parameter [int]
734 Specifies the maximal packet length to be completely inlined into WQE for
735 Enhanced MPW method. If packet is large the specified value, the packet data
736 won't be copied, and data buffer is addressed with pointer. If packet length
737 is less or equal, all packet data will be copied into WQE. This may improve PCI
738 bandwidth utilization for short packets significantly but requires the extra
741 The data inline feature is controlled by number of TX queues, if number of Tx
742 queues is larger than ``txqs_min_inline`` key parameter, the inline feature
743 is engaged, if there are not enough Tx queues (which means not enough CPU cores
744 and CPU resources are scarce), data inline is not performed by the driver.
745 Assigning ``txqs_min_inline`` with zero always enables the data inline.
747 The default ``txq_inline_mpw`` value is 268. The specified value may be adjusted
748 by the driver in order not to exceed the limit (930 bytes) and to provide better
749 WQE space filling without gaps, the adjustment is reflected in the debug log.
750 Due to multiple packets may be included to the same WQE with Enhanced Multi
751 Packet Write Method and overall WQE size is limited it is not recommended to
752 specify large values for the ``txq_inline_mpw``. Also, the default value (268)
753 may be decreased in run-time if the large transmit queue size is requested
754 and hardware does not support enough descriptor amount, in this case warning
755 is emitted. If ``txq_inline_mpw`` key is specified and requested inline
756 settings can not be satisfied then error will be raised.
758 - ``txqs_max_vec`` parameter [int]
760 Enable vectorized Tx only when the number of TX queues is less than or
761 equal to this value. This parameter is deprecated and ignored, kept
762 for compatibility issue to not prevent driver from probing.
764 - ``txq_mpw_hdr_dseg_en`` parameter [int]
766 A nonzero value enables including two pointers in the first block of TX
767 descriptor. The parameter is deprecated and ignored, kept for compatibility
770 - ``txq_max_inline_len`` parameter [int]
772 Maximum size of packet to be inlined. This limits the size of packet to
773 be inlined. If the size of a packet is larger than configured value, the
774 packet isn't inlined even though there's enough space remained in the
775 descriptor. Instead, the packet is included with pointer. This parameter
776 is deprecated and converted directly to ``txq_inline_mpw`` providing full
777 compatibility. Valid only if eMPW feature is engaged.
779 - ``txq_mpw_en`` parameter [int]
781 A nonzero value enables Enhanced Multi-Packet Write (eMPW) for ConnectX-5,
782 ConnectX-6, ConnectX-6 Dx, ConnectX-6 Lx, BlueField, BlueField-2.
783 eMPW allows the Tx burst function to pack up multiple packets
784 in a single descriptor session in order to save PCI bandwidth
785 and improve performance at the cost of a slightly higher CPU usage.
786 When ``txq_inline_mpw`` is set along with ``txq_mpw_en``,
787 Tx burst function copies entire packet data on to Tx descriptor
788 instead of including pointer of packet.
790 The Enhanced Multi-Packet Write feature is enabled by default if NIC supports
791 it, can be disabled by explicit specifying 0 value for ``txq_mpw_en`` option.
792 Also, if minimal data inlining is requested by non-zero ``txq_inline_min``
793 option or reported by the NIC, the eMPW feature is disengaged.
795 - ``tx_db_nc`` parameter [int]
797 The rdma core library can map doorbell register in two ways, depending on the
798 environment variable "MLX5_SHUT_UP_BF":
800 - As regular cached memory (usually with write combining attribute), if the
801 variable is either missing or set to zero.
802 - As non-cached memory, if the variable is present and set to not "0" value.
804 The type of mapping may slightly affect the Tx performance, the optimal choice
805 is strongly relied on the host architecture and should be deduced practically.
807 If ``tx_db_nc`` is set to zero, the doorbell is forced to be mapped to regular
808 memory (with write combining), the PMD will perform the extra write memory barrier
809 after writing to doorbell, it might increase the needed CPU clocks per packet
810 to send, but latency might be improved.
812 If ``tx_db_nc`` is set to one, the doorbell is forced to be mapped to non
813 cached memory, the PMD will not perform the extra write memory barrier
814 after writing to doorbell, on some architectures it might improve the
817 If ``tx_db_nc`` is set to two, the doorbell is forced to be mapped to regular
818 memory, the PMD will use heuristics to decide whether write memory barrier
819 should be performed. For bursts with size multiple of recommended one (64 pkts)
820 it is supposed the next burst is coming and no need to issue the extra memory
821 barrier (it is supposed to be issued in the next coming burst, at least after
822 descriptor writing). It might increase latency (on some hosts till next
823 packets transmit) and should be used with care.
825 If ``tx_db_nc`` is omitted or set to zero, the preset (if any) environment
826 variable "MLX5_SHUT_UP_BF" value is used. If there is no "MLX5_SHUT_UP_BF",
827 the default ``tx_db_nc`` value is zero for ARM64 hosts and one for others.
829 - ``tx_pp`` parameter [int]
831 If a nonzero value is specified the driver creates all necessary internal
832 objects to provide accurate packet send scheduling on mbuf timestamps.
833 The positive value specifies the scheduling granularity in nanoseconds,
834 the packet send will be accurate up to specified digits. The allowed range is
835 from 500 to 1 million of nanoseconds. The negative value specifies the module
836 of granularity and engages the special test mode the check the schedule rate.
837 By default (if the ``tx_pp`` is not specified) send scheduling on timestamps
840 - ``tx_skew`` parameter [int]
842 The parameter adjusts the send packet scheduling on timestamps and represents
843 the average delay between beginning of the transmitting descriptor processing
844 by the hardware and appearance of actual packet data on the wire. The value
845 should be provided in nanoseconds and is valid only if ``tx_pp`` parameter is
846 specified. The default value is zero.
848 - ``tx_vec_en`` parameter [int]
850 A nonzero value enables Tx vector on ConnectX-5, ConnectX-6, ConnectX-6 Dx,
851 ConnectX-6 Lx, BlueField and BlueField-2 NICs
852 if the number of global Tx queues on the port is less than ``txqs_max_vec``.
853 The parameter is deprecated and ignored.
855 - ``rx_vec_en`` parameter [int]
857 A nonzero value enables Rx vector if the port is not configured in
858 multi-segment otherwise this parameter is ignored.
862 - ``vf_nl_en`` parameter [int]
864 A nonzero value enables Netlink requests from the VF to add/remove MAC
865 addresses or/and enable/disable promiscuous/all multicast on the Netdevice.
866 Otherwise the relevant configuration must be run with Linux iproute2 tools.
867 This is a prerequisite to receive this kind of traffic.
869 Enabled by default, valid only on VF devices ignored otherwise.
871 - ``l3_vxlan_en`` parameter [int]
873 A nonzero value allows L3 VXLAN and VXLAN-GPE flow creation. To enable
874 L3 VXLAN or VXLAN-GPE, users has to configure firmware and enable this
875 parameter. This is a prerequisite to receive this kind of traffic.
879 - ``dv_xmeta_en`` parameter [int]
881 A nonzero value enables extensive flow metadata support if device is
882 capable and driver supports it. This can enable extensive support of
883 ``MARK`` and ``META`` item of ``rte_flow``. The newly introduced
884 ``SET_TAG`` and ``SET_META`` actions do not depend on ``dv_xmeta_en``.
886 There are some possible configurations, depending on parameter value:
888 - 0, this is default value, defines the legacy mode, the ``MARK`` and
889 ``META`` related actions and items operate only within NIC Tx and
890 NIC Rx steering domains, no ``MARK`` and ``META`` information crosses
891 the domain boundaries. The ``MARK`` item is 24 bits wide, the ``META``
892 item is 32 bits wide and match supported on egress only.
894 - 1, this engages extensive metadata mode, the ``MARK`` and ``META``
895 related actions and items operate within all supported steering domains,
896 including FDB, ``MARK`` and ``META`` information may cross the domain
897 boundaries. The ``MARK`` item is 24 bits wide, the ``META`` item width
898 depends on kernel and firmware configurations and might be 0, 16 or
899 32 bits. Within NIC Tx domain ``META`` data width is 32 bits for
900 compatibility, the actual width of data transferred to the FDB domain
901 depends on kernel configuration and may be vary. The actual supported
902 width can be retrieved in runtime by series of rte_flow_validate()
905 - 2, this engages extensive metadata mode, the ``MARK`` and ``META``
906 related actions and items operate within all supported steering domains,
907 including FDB, ``MARK`` and ``META`` information may cross the domain
908 boundaries. The ``META`` item is 32 bits wide, the ``MARK`` item width
909 depends on kernel and firmware configurations and might be 0, 16 or
910 24 bits. The actual supported width can be retrieved in runtime by
911 series of rte_flow_validate() trials.
913 - 3, this engages tunnel offload mode. In E-Switch configuration, that
914 mode implicitly activates ``dv_xmeta_en=1``.
916 +------+-----------+-----------+-------------+-------------+
917 | Mode | ``MARK`` | ``META`` | ``META`` Tx | FDB/Through |
918 +======+===========+===========+=============+=============+
919 | 0 | 24 bits | 32 bits | 32 bits | no |
920 +------+-----------+-----------+-------------+-------------+
921 | 1 | 24 bits | vary 0-32 | 32 bits | yes |
922 +------+-----------+-----------+-------------+-------------+
923 | 2 | vary 0-24 | 32 bits | 32 bits | yes |
924 +------+-----------+-----------+-------------+-------------+
926 If there is no E-Switch configuration the ``dv_xmeta_en`` parameter is
927 ignored and the device is configured to operate in legacy mode (0).
929 Disabled by default (set to 0).
931 The Direct Verbs/Rules (engaged with ``dv_flow_en`` = 1) supports all
932 of the extensive metadata features. The legacy Verbs supports FLAG and
933 MARK metadata actions over NIC Rx steering domain only.
935 Setting META value to zero in flow action means there is no item provided
936 and receiving datapath will not report in mbufs the metadata are present.
937 Setting MARK value to zero in flow action means the zero FDIR ID value
938 will be reported on packet receiving.
940 For the MARK action the last 16 values in the full range are reserved for
941 internal PMD purposes (to emulate FLAG action). The valid range for the
942 MARK action values is 0-0xFFEF for the 16-bit mode and 0-xFFFFEF
943 for the 24-bit mode, the flows with the MARK action value outside
944 the specified range will be rejected.
946 - ``dv_flow_en`` parameter [int]
948 A nonzero value enables the DV flow steering assuming it is supported
949 by the driver (RDMA Core library version is rdma-core-24.0 or higher).
951 Enabled by default if supported.
953 - ``dv_esw_en`` parameter [int]
955 A nonzero value enables E-Switch using Direct Rules.
957 Enabled by default if supported.
959 - ``lacp_by_user`` parameter [int]
961 A nonzero value enables the control of LACP traffic by the user application.
962 When a bond exists in the driver, by default it should be managed by the
963 kernel and therefore LACP traffic should be steered to the kernel.
964 If this devarg is set to 1 it will allow the user to manage the bond by
965 itself and not steer LACP traffic to the kernel.
967 Disabled by default (set to 0).
969 - ``mr_ext_memseg_en`` parameter [int]
971 A nonzero value enables extending memseg when registering DMA memory. If
972 enabled, the number of entries in MR (Memory Region) lookup table on datapath
973 is minimized and it benefits performance. On the other hand, it worsens memory
974 utilization because registered memory is pinned by kernel driver. Even if a
975 page in the extended chunk is freed, that doesn't become reusable until the
976 entire memory is freed.
980 - ``representor`` parameter [list]
982 This parameter can be used to instantiate DPDK Ethernet devices from
983 existing port (PF, VF or SF) representors configured on the device.
985 It is a standard parameter whose format is described in
986 :ref:`ethernet_device_standard_device_arguments`.
988 For instance, to probe VF port representors 0 through 2::
990 <PCI_BDF>,representor=vf[0-2]
992 To probe SF port representors 0 through 2::
994 <PCI_BDF>,representor=sf[0-2]
996 To probe VF port representors 0 through 2 on both PFs of bonding device::
998 <Primary_PCI_BDF>,representor=pf[0,1]vf[0-2]
1000 - ``max_dump_files_num`` parameter [int]
1002 The maximum number of files per PMD entity that may be created for debug information.
1003 The files will be created in /var/log directory or in current directory.
1005 set to 128 by default.
1007 - ``lro_timeout_usec`` parameter [int]
1009 The maximum allowed duration of an LRO session, in micro-seconds.
1010 PMD will set the nearest value supported by HW, which is not bigger than
1011 the input ``lro_timeout_usec`` value.
1012 If this parameter is not specified, by default PMD will set
1013 the smallest value supported by HW.
1015 - ``hp_buf_log_sz`` parameter [int]
1017 The total data buffer size of a hairpin queue (logarithmic form), in bytes.
1018 PMD will set the data buffer size to 2 ** ``hp_buf_log_sz``, both for RX & TX.
1019 The capacity of the value is specified by the firmware and the initialization
1020 will get a failure if it is out of scope.
1021 The range of the value is from 11 to 19 right now, and the supported frame
1022 size of a single packet for hairpin is from 512B to 128KB. It might change if
1023 different firmware release is being used. By using a small value, it could
1024 reduce memory consumption but not work with a large frame. If the value is
1025 too large, the memory consumption will be high and some potential performance
1026 degradation will be introduced.
1027 By default, the PMD will set this value to 16, which means that 9KB jumbo
1028 frames will be supported.
1030 - ``reclaim_mem_mode`` parameter [int]
1032 Cache some resources in flow destroy will help flow recreation more efficient.
1033 While some systems may require the all the resources can be reclaimed after
1035 The parameter ``reclaim_mem_mode`` provides the option for user to configure
1036 if the resource cache is needed or not.
1038 There are three options to choose:
1040 - 0. It means the flow resources will be cached as usual. The resources will
1041 be cached, helpful with flow insertion rate.
1043 - 1. It will only enable the DPDK PMD level resources reclaim.
1045 - 2. Both DPDK PMD level and rdma-core low level will be configured as
1048 By default, the PMD will set this value to 0.
1050 - ``sys_mem_en`` parameter [int]
1052 A non-zero value enables the PMD memory management allocating memory
1053 from system by default, without explicit rte memory flag.
1055 By default, the PMD will set this value to 0.
1057 - ``decap_en`` parameter [int]
1059 Some devices do not support FCS (frame checksum) scattering for
1060 tunnel-decapsulated packets.
1061 If set to 0, this option forces the FCS feature and rejects tunnel
1062 decapsulation in the flow engine for such devices.
1064 By default, the PMD will set this value to 1.
1066 .. _mlx5_firmware_config:
1068 Firmware configuration
1069 ~~~~~~~~~~~~~~~~~~~~~~
1071 Firmware features can be configured as key/value pairs.
1073 The command to set a value is::
1075 mlxconfig -d <device> set <key>=<value>
1077 The command to query a value is::
1079 mlxconfig -d <device> query | grep <key>
1081 The device name for the command ``mlxconfig`` can be either the PCI address,
1082 or the mst device name found with::
1086 Below are some firmware configurations listed.
1092 value: 1=Infiniband 2=Ethernet 3=VPI(auto-sense)
1098 - maximum number of SR-IOV virtual functions::
1102 - enable DevX (required by Direct Rules and other features)::
1106 - aggressive CQE zipping::
1110 - L3 VXLAN and VXLAN-GPE destination UDP port::
1113 IP_OVER_VXLAN_PORT=<udp dport>
1115 - enable VXLAN-GPE tunnel flow matching::
1117 FLEX_PARSER_PROFILE_ENABLE=0
1119 FLEX_PARSER_PROFILE_ENABLE=2
1121 - enable IP-in-IP tunnel flow matching::
1123 FLEX_PARSER_PROFILE_ENABLE=0
1125 - enable MPLS flow matching::
1127 FLEX_PARSER_PROFILE_ENABLE=1
1129 - enable ICMP(code/type/identifier/sequence number) / ICMP6(code/type) fields matching::
1131 FLEX_PARSER_PROFILE_ENABLE=2
1133 - enable Geneve flow matching::
1135 FLEX_PARSER_PROFILE_ENABLE=0
1137 FLEX_PARSER_PROFILE_ENABLE=1
1139 - enable Geneve TLV option flow matching::
1141 FLEX_PARSER_PROFILE_ENABLE=0
1143 - enable GTP flow matching::
1145 FLEX_PARSER_PROFILE_ENABLE=3
1147 - enable eCPRI flow matching::
1149 FLEX_PARSER_PROFILE_ENABLE=4
1155 This driver relies on external libraries and kernel drivers for resources
1156 allocations and initialization. The following dependencies are not part of
1157 DPDK and must be installed separately:
1161 User space Verbs framework used by librte_net_mlx5. This library provides
1162 a generic interface between the kernel and low-level user space drivers
1165 It allows slow and privileged operations (context initialization, hardware
1166 resources allocations) to be managed by the kernel and fast operations to
1167 never leave user space.
1171 Low-level user space driver library for Mellanox
1172 ConnectX-4/ConnectX-5/ConnectX-6/BlueField devices, it is automatically loaded
1175 This library basically implements send/receive calls to the hardware
1178 - **Kernel modules**
1180 They provide the kernel-side Verbs API and low level device drivers that
1181 manage actual hardware initialization and resources sharing with user
1184 Unlike most other PMDs, these modules must remain loaded and bound to
1187 - mlx5_core: hardware driver managing Mellanox
1188 ConnectX-4/ConnectX-5/ConnectX-6/BlueField devices and related Ethernet kernel
1190 - mlx5_ib: InifiniBand device driver.
1191 - ib_uverbs: user space driver for Verbs (entry point for libibverbs).
1193 - **Firmware update**
1195 Mellanox OFED/EN releases include firmware updates for
1196 ConnectX-4/ConnectX-5/ConnectX-6/BlueField adapters.
1198 Because each release provides new features, these updates must be applied to
1199 match the kernel modules and libraries they come with.
1203 Both libraries are BSD and GPL licensed. Linux kernel modules are GPL
1209 Either RDMA Core library with a recent enough Linux kernel release
1210 (recommended) or Mellanox OFED/EN, which provides compatibility with older
1213 RDMA Core with Linux Kernel
1214 ^^^^^^^^^^^^^^^^^^^^^^^^^^^
1216 - Minimal kernel version : v4.14 or the most recent 4.14-rc (see `Linux installation documentation`_)
1217 - Minimal rdma-core version: v15+ commit 0c5f5765213a ("Merge pull request #227 from yishaih/tm")
1218 (see `RDMA Core installation documentation`_)
1219 - When building for i686 use:
1221 - rdma-core version 18.0 or above built with 32bit support.
1222 - Kernel version 4.14.41 or above.
1224 - Starting with rdma-core v21, static libraries can be built::
1227 CFLAGS=-fPIC cmake -DIN_PLACE=1 -DENABLE_STATIC=1 -GNinja ..
1230 .. _`Linux installation documentation`: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable.git/plain/Documentation/admin-guide/README.rst
1231 .. _`RDMA Core installation documentation`: https://raw.githubusercontent.com/linux-rdma/rdma-core/master/README.md
1237 - Mellanox OFED version: **4.5** and above /
1238 Mellanox EN version: **4.5** and above
1241 - ConnectX-4: **12.21.1000** and above.
1242 - ConnectX-4 Lx: **14.21.1000** and above.
1243 - ConnectX-5: **16.21.1000** and above.
1244 - ConnectX-5 Ex: **16.21.1000** and above.
1245 - ConnectX-6: **20.27.0090** and above.
1246 - ConnectX-6 Dx: **22.27.0090** and above.
1247 - BlueField: **18.25.1010** and above.
1249 While these libraries and kernel modules are available on OpenFabrics
1250 Alliance's `website <https://www.openfabrics.org/>`__ and provided by package
1251 managers on most distributions, this PMD requires Ethernet extensions that
1252 may not be supported at the moment (this is a work in progress).
1255 <http://www.mellanox.com/page/products_dyn?product_family=26&mtag=linux>`__ and
1257 <http://www.mellanox.com/page/products_dyn?product_family=27&mtag=linux>`__
1258 include the necessary support and should be used in the meantime. For DPDK,
1259 only libibverbs, libmlx5, mlnx-ofed-kernel packages and firmware updates are
1260 required from that distribution.
1264 Several versions of Mellanox OFED/EN are available. Installing the version
1265 this DPDK release was developed and tested against is strongly
1266 recommended. Please check the `linux prerequisites`_.
1268 Windows Prerequisites
1269 ---------------------
1271 This driver relies on external libraries and kernel drivers for resources
1272 allocations and initialization. The dependencies in the following sub-sections
1273 are not part of DPDK, and must be installed separately.
1275 Compilation Prerequisites
1276 ~~~~~~~~~~~~~~~~~~~~~~~~~
1278 DevX SDK installation
1279 ^^^^^^^^^^^^^^^^^^^^^
1281 The DevX SDK must be installed on the machine building the Windows PMD.
1282 Additional information can be found at
1283 `How to Integrate Windows DevX in Your Development Environment
1284 <https://docs.mellanox.com/display/winof2v250/RShim+Drivers+and+Usage#RShimDriversandUsage-DevXInterface>`__.
1286 Runtime Prerequisites
1287 ~~~~~~~~~~~~~~~~~~~~~
1289 WinOF2 version 2.60 or higher must be installed on the machine.
1294 The driver can be downloaded from the following site:
1296 <https://www.mellanox.com/products/adapter-software/ethernet/windows/winof-2>`__
1301 DevX for Windows must be enabled in the Windows registry.
1302 The keys ``DevxEnabled`` and ``DevxFsRules`` must be set.
1303 Additional information can be found in the WinOF2 user manual.
1308 The following Mellanox device families are supported by the same mlx5 driver:
1320 Below are detailed device names:
1322 * Mellanox\ |reg| ConnectX\ |reg|-4 10G MCX4111A-XCAT (1x10G)
1323 * Mellanox\ |reg| ConnectX\ |reg|-4 10G MCX412A-XCAT (2x10G)
1324 * Mellanox\ |reg| ConnectX\ |reg|-4 25G MCX4111A-ACAT (1x25G)
1325 * Mellanox\ |reg| ConnectX\ |reg|-4 25G MCX412A-ACAT (2x25G)
1326 * Mellanox\ |reg| ConnectX\ |reg|-4 40G MCX413A-BCAT (1x40G)
1327 * Mellanox\ |reg| ConnectX\ |reg|-4 40G MCX4131A-BCAT (1x40G)
1328 * Mellanox\ |reg| ConnectX\ |reg|-4 40G MCX415A-BCAT (1x40G)
1329 * Mellanox\ |reg| ConnectX\ |reg|-4 50G MCX413A-GCAT (1x50G)
1330 * Mellanox\ |reg| ConnectX\ |reg|-4 50G MCX4131A-GCAT (1x50G)
1331 * Mellanox\ |reg| ConnectX\ |reg|-4 50G MCX414A-BCAT (2x50G)
1332 * Mellanox\ |reg| ConnectX\ |reg|-4 50G MCX415A-GCAT (1x50G)
1333 * Mellanox\ |reg| ConnectX\ |reg|-4 50G MCX416A-BCAT (2x50G)
1334 * Mellanox\ |reg| ConnectX\ |reg|-4 50G MCX416A-GCAT (2x50G)
1335 * Mellanox\ |reg| ConnectX\ |reg|-4 50G MCX415A-CCAT (1x100G)
1336 * Mellanox\ |reg| ConnectX\ |reg|-4 100G MCX416A-CCAT (2x100G)
1337 * Mellanox\ |reg| ConnectX\ |reg|-4 Lx 10G MCX4111A-XCAT (1x10G)
1338 * Mellanox\ |reg| ConnectX\ |reg|-4 Lx 10G MCX4121A-XCAT (2x10G)
1339 * Mellanox\ |reg| ConnectX\ |reg|-4 Lx 25G MCX4111A-ACAT (1x25G)
1340 * Mellanox\ |reg| ConnectX\ |reg|-4 Lx 25G MCX4121A-ACAT (2x25G)
1341 * Mellanox\ |reg| ConnectX\ |reg|-4 Lx 40G MCX4131A-BCAT (1x40G)
1342 * Mellanox\ |reg| ConnectX\ |reg|-5 100G MCX556A-ECAT (2x100G)
1343 * Mellanox\ |reg| ConnectX\ |reg|-5 Ex EN 100G MCX516A-CDAT (2x100G)
1344 * Mellanox\ |reg| ConnectX\ |reg|-6 200G MCX654106A-HCAT (2x200G)
1345 * Mellanox\ |reg| ConnectX\ |reg|-6 Dx EN 100G MCX623106AN-CDAT (2x100G)
1346 * Mellanox\ |reg| ConnectX\ |reg|-6 Dx EN 200G MCX623105AN-VDAT (1x200G)
1347 * Mellanox\ |reg| ConnectX\ |reg|-6 Lx EN 25G MCX631102AN-ADAT (2x25G)
1349 Quick Start Guide on OFED/EN
1350 ----------------------------
1352 1. Download latest Mellanox OFED/EN. For more info check the `linux prerequisites`_.
1355 2. Install the required libraries and kernel modules either by installing
1356 only the required set, or by installing the entire Mellanox OFED/EN::
1358 ./mlnxofedinstall --upstream-libs --dpdk
1360 3. Verify the firmware is the correct one::
1364 4. Verify all ports links are set to Ethernet::
1366 mlxconfig -d <mst device> query | grep LINK_TYPE
1370 Link types may have to be configured to Ethernet::
1372 mlxconfig -d <mst device> set LINK_TYPE_P1/2=1/2/3
1374 * LINK_TYPE_P1=<1|2|3> , 1=Infiniband 2=Ethernet 3=VPI(auto-sense)
1376 For hypervisors, verify SR-IOV is enabled on the NIC::
1378 mlxconfig -d <mst device> query | grep SRIOV_EN
1381 If needed, configure SR-IOV::
1383 mlxconfig -d <mst device> set SRIOV_EN=1 NUM_OF_VFS=16
1384 mlxfwreset -d <mst device> reset
1386 5. Restart the driver::
1388 /etc/init.d/openibd restart
1392 service openibd restart
1394 If link type was changed, firmware must be reset as well::
1396 mlxfwreset -d <mst device> reset
1398 For hypervisors, after reset write the sysfs number of virtual functions
1401 To dynamically instantiate a given number of virtual functions (VFs)::
1403 echo [num_vfs] > /sys/class/infiniband/mlx5_0/device/sriov_numvfs
1405 6. Install DPDK and you are ready to go.
1406 See :doc:`compilation instructions <../linux_gsg/build_dpdk>`.
1408 Enable switchdev mode
1409 ---------------------
1411 Switchdev mode is a mode in E-Switch, that binds between representor and VF or SF.
1412 Representor is a port in DPDK that is connected to a VF or SF in such a way
1413 that assuming there are no offload flows, each packet that is sent from the VF or SF
1414 will be received by the corresponding representor. While each packet that is or SF
1415 sent to a representor will be received by the VF or SF.
1416 This is very useful in case of SRIOV mode, where the first packet that is sent
1417 by the VF or SF will be received by the DPDK application which will decide if this
1418 flow should be offloaded to the E-Switch. After offloading the flow packet
1419 that the VF or SF that are matching the flow will not be received any more by
1420 the DPDK application.
1422 1. Enable SRIOV mode::
1424 mlxconfig -d <mst device> set SRIOV_EN=true
1426 2. Configure the max number of VFs::
1428 mlxconfig -d <mst device> set NUM_OF_VFS=<num of vfs>
1432 mlxfwreset -d <mst device> reset
1434 3. Configure the actual number of VFs::
1436 echo <num of vfs > /sys/class/net/<net device>/device/sriov_numvfs
1438 4. Unbind the device (can be rebind after the switchdev mode)::
1440 echo -n "<device pci address" > /sys/bus/pci/drivers/mlx5_core/unbind
1442 5. Enbale switchdev mode::
1444 echo switchdev > /sys/class/net/<net device>/compat/devlink/mode
1446 Sub-Function representor
1447 ------------------------
1449 Sub-Function is a portion of the PCI device, a SF netdev has its own
1450 dedicated queues(txq, rxq). A SF netdev supports E-Switch representation
1451 offload similar to existing PF and VF representors. A SF shares PCI
1452 level resources with other SFs and/or with its parent PCI function.
1454 1. Configure SF feature::
1456 mlxconfig -d <mst device> set PF_BAR2_SIZE=<0/1/2/3> PF_BAR2_ENABLE=1
1458 Value of PF_BAR2_SIZE:
1467 mlxfwreset -d <mst device> reset
1469 3. Enable switchdev mode::
1471 echo switchdev > /sys/class/net/<net device>/compat/devlink/mode
1475 mlnx-sf -d <PCI_BDF> -a create
1477 5. Probe SF representor::
1479 testpmd> port attach <PCI_BDF>,representor=sf0,dv_flow_en=1
1484 1. Configure aggressive CQE Zipping for maximum performance::
1486 mlxconfig -d <mst device> s CQE_COMPRESSION=1
1488 To set it back to the default CQE Zipping mode use::
1490 mlxconfig -d <mst device> s CQE_COMPRESSION=0
1492 2. In case of virtualization:
1494 - Make sure that hypervisor kernel is 3.16 or newer.
1495 - Configure boot with ``iommu=pt``.
1496 - Use 1G huge pages.
1497 - Make sure to allocate a VM on huge pages.
1498 - Make sure to set CPU pinning.
1500 3. Use the CPU near local NUMA node to which the PCIe adapter is connected,
1501 for better performance. For VMs, verify that the right CPU
1502 and NUMA node are pinned according to the above. Run::
1504 lstopo-no-graphics --merge
1506 to identify the NUMA node to which the PCIe adapter is connected.
1508 4. If more than one adapter is used, and root complex capabilities allow
1509 to put both adapters on the same NUMA node without PCI bandwidth degradation,
1510 it is recommended to locate both adapters on the same NUMA node.
1511 This in order to forward packets from one to the other without
1512 NUMA performance penalty.
1514 5. Disable pause frames::
1516 ethtool -A <netdev> rx off tx off
1518 6. Verify IO non-posted prefetch is disabled by default. This can be checked
1519 via the BIOS configuration. Please contact you server provider for more
1520 information about the settings.
1524 On some machines, depends on the machine integrator, it is beneficial
1525 to set the PCI max read request parameter to 1K. This can be
1526 done in the following way:
1528 To query the read request size use::
1530 setpci -s <NIC PCI address> 68.w
1532 If the output is different than 3XXX, set it by::
1534 setpci -s <NIC PCI address> 68.w=3XXX
1536 The XXX can be different on different systems. Make sure to configure
1537 according to the setpci output.
1539 7. To minimize overhead of searching Memory Regions:
1541 - '--socket-mem' is recommended to pin memory by predictable amount.
1542 - Configure per-lcore cache when creating Mempools for packet buffer.
1543 - Refrain from dynamically allocating/freeing memory in run-time.
1548 There are multiple Rx burst functions with different advantages and limitations.
1550 .. table:: Rx burst functions
1552 +-------------------+------------------------+---------+-----------------+------+-------+
1553 || Function Name || Enabler || Scatter|| Error Recovery || CQE || Large|
1554 | | | | || comp|| MTU |
1555 +===================+========================+=========+=================+======+=======+
1556 | rx_burst | rx_vec_en=0 | Yes | Yes | Yes | Yes |
1557 +-------------------+------------------------+---------+-----------------+------+-------+
1558 | rx_burst_vec | rx_vec_en=1 (default) | No | if CQE comp off | Yes | No |
1559 +-------------------+------------------------+---------+-----------------+------+-------+
1560 | rx_burst_mprq || mprq_en=1 | No | Yes | Yes | Yes |
1561 | || RxQs >= rxqs_min_mprq | | | | |
1562 +-------------------+------------------------+---------+-----------------+------+-------+
1563 | rx_burst_mprq_vec || rx_vec_en=1 (default) | No | if CQE comp off | Yes | Yes |
1564 | || mprq_en=1 | | | | |
1565 | || RxQs >= rxqs_min_mprq | | | | |
1566 +-------------------+------------------------+---------+-----------------+------+-------+
1568 .. _mlx5_offloads_support:
1570 Supported hardware offloads
1571 ---------------------------
1573 .. table:: Minimal SW/HW versions for queue offloads
1575 ============== ===== ===== ========= ===== ========== =============
1576 Offload DPDK Linux rdma-core OFED firmware hardware
1577 ============== ===== ===== ========= ===== ========== =============
1578 common base 17.11 4.14 16 4.2-1 12.21.1000 ConnectX-4
1579 checksums 17.11 4.14 16 4.2-1 12.21.1000 ConnectX-4
1580 Rx timestamp 17.11 4.14 16 4.2-1 12.21.1000 ConnectX-4
1581 TSO 17.11 4.14 16 4.2-1 12.21.1000 ConnectX-4
1582 LRO 19.08 N/A N/A 4.6-4 16.25.6406 ConnectX-5
1583 Tx scheduling 20.08 N/A N/A 5.1-2 22.28.2006 ConnectX-6 Dx
1584 Buffer Split 20.11 N/A N/A 5.1-2 16.28.2006 ConnectX-5
1585 ============== ===== ===== ========= ===== ========== =============
1587 .. table:: Minimal SW/HW versions for rte_flow offloads
1589 +-----------------------+-----------------+-----------------+
1590 | Offload | with E-Switch | with NIC |
1591 +=======================+=================+=================+
1592 | Count | | DPDK 19.05 | | DPDK 19.02 |
1593 | | | OFED 4.6 | | OFED 4.6 |
1594 | | | rdma-core 24 | | rdma-core 23 |
1595 | | | ConnectX-5 | | ConnectX-5 |
1596 +-----------------------+-----------------+-----------------+
1597 | Drop | | DPDK 19.05 | | DPDK 18.11 |
1598 | | | OFED 4.6 | | OFED 4.5 |
1599 | | | rdma-core 24 | | rdma-core 23 |
1600 | | | ConnectX-5 | | ConnectX-4 |
1601 +-----------------------+-----------------+-----------------+
1602 | Queue / RSS | | | | DPDK 18.11 |
1603 | | | N/A | | OFED 4.5 |
1604 | | | | | rdma-core 23 |
1605 | | | | | ConnectX-4 |
1606 +-----------------------+-----------------+-----------------+
1607 | Shared action | | | | |
1608 | | | :numref:`sact`| | :numref:`sact`|
1611 +-----------------------+-----------------+-----------------+
1612 | | VLAN | | DPDK 19.11 | | DPDK 19.11 |
1613 | | (of_pop_vlan / | | OFED 4.7-1 | | OFED 4.7-1 |
1614 | | of_push_vlan / | | ConnectX-5 | | ConnectX-5 |
1615 | | of_set_vlan_pcp / | | | | |
1616 | | of_set_vlan_vid) | | | | |
1617 +-----------------------+-----------------+-----------------+
1618 | | VLAN | | DPDK 21.05 | | |
1619 | | ingress and / | | OFED 5.3 | | N/A |
1620 | | of_push_vlan / | | ConnectX-6 Dx | | |
1621 +-----------------------+-----------------+-----------------+
1622 | | VLAN | | DPDK 21.05 | | |
1623 | | egress and / | | OFED 5.3 | | N/A |
1624 | | of_pop_vlan / | | ConnectX-6 Dx | | |
1625 +-----------------------+-----------------+-----------------+
1626 | Encapsulation | | DPDK 19.05 | | DPDK 19.02 |
1627 | (VXLAN / NVGRE / RAW) | | OFED 4.7-1 | | OFED 4.6 |
1628 | | | rdma-core 24 | | rdma-core 23 |
1629 | | | ConnectX-5 | | ConnectX-5 |
1630 +-----------------------+-----------------+-----------------+
1631 | Encapsulation | | DPDK 19.11 | | DPDK 19.11 |
1632 | GENEVE | | OFED 4.7-3 | | OFED 4.7-3 |
1633 | | | rdma-core 27 | | rdma-core 27 |
1634 | | | ConnectX-5 | | ConnectX-5 |
1635 +-----------------------+-----------------+-----------------+
1636 | Tunnel Offload | | DPDK 20.11 | | DPDK 20.11 |
1637 | | | OFED 5.1-2 | | OFED 5.1-2 |
1638 | | | rdma-core 32 | | N/A |
1639 | | | ConnectX-5 | | ConnectX-5 |
1640 +-----------------------+-----------------+-----------------+
1641 | | Header rewrite | | DPDK 19.05 | | DPDK 19.02 |
1642 | | (set_ipv4_src / | | OFED 4.7-1 | | OFED 4.7-1 |
1643 | | set_ipv4_dst / | | rdma-core 24 | | rdma-core 24 |
1644 | | set_ipv6_src / | | ConnectX-5 | | ConnectX-5 |
1645 | | set_ipv6_dst / | | | | |
1646 | | set_tp_src / | | | | |
1647 | | set_tp_dst / | | | | |
1648 | | dec_ttl / | | | | |
1649 | | set_ttl / | | | | |
1650 | | set_mac_src / | | | | |
1651 | | set_mac_dst) | | | | |
1652 +-----------------------+-----------------+-----------------+
1653 | | Header rewrite | | DPDK 20.02 | | DPDK 20.02 |
1654 | | (set_dscp) | | OFED 5.0 | | OFED 5.0 |
1655 | | | | rdma-core 24 | | rdma-core 24 |
1656 | | | | ConnectX-5 | | ConnectX-5 |
1657 +-----------------------+-----------------+-----------------+
1658 | Jump | | DPDK 19.05 | | DPDK 19.02 |
1659 | | | OFED 4.7-1 | | OFED 4.7-1 |
1660 | | | rdma-core 24 | | N/A |
1661 | | | ConnectX-5 | | ConnectX-5 |
1662 +-----------------------+-----------------+-----------------+
1663 | Mark / Flag | | DPDK 19.05 | | DPDK 18.11 |
1664 | | | OFED 4.6 | | OFED 4.5 |
1665 | | | rdma-core 24 | | rdma-core 23 |
1666 | | | ConnectX-5 | | ConnectX-4 |
1667 +-----------------------+-----------------+-----------------+
1668 | Meta data | | DPDK 19.11 | | DPDK 19.11 |
1669 | | | OFED 4.7-3 | | OFED 4.7-3 |
1670 | | | rdma-core 26 | | rdma-core 26 |
1671 | | | ConnectX-5 | | ConnectX-5 |
1672 +-----------------------+-----------------+-----------------+
1673 | Port ID | | DPDK 19.05 | | N/A |
1674 | | | OFED 4.7-1 | | N/A |
1675 | | | rdma-core 24 | | N/A |
1676 | | | ConnectX-5 | | N/A |
1677 +-----------------------+-----------------+-----------------+
1678 | Hairpin | | | | DPDK 19.11 |
1679 | | | N/A | | OFED 4.7-3 |
1680 | | | | | rdma-core 26 |
1681 | | | | | ConnectX-5 |
1682 +-----------------------+-----------------+-----------------+
1683 | 2-port Hairpin | | | | DPDK 20.11 |
1684 | | | N/A | | OFED 5.1-2 |
1686 | | | | | ConnectX-5 |
1687 +-----------------------+-----------------+-----------------+
1688 | Metering | | DPDK 19.11 | | DPDK 19.11 |
1689 | | | OFED 4.7-3 | | OFED 4.7-3 |
1690 | | | rdma-core 26 | | rdma-core 26 |
1691 | | | ConnectX-5 | | ConnectX-5 |
1692 +-----------------------+-----------------+-----------------+
1693 | Sampling | | DPDK 20.11 | | DPDK 20.11 |
1694 | | | OFED 5.1-2 | | OFED 5.1-2 |
1695 | | | rdma-core 32 | | N/A |
1696 | | | ConnectX-5 | | ConnectX-5 |
1697 +-----------------------+-----------------+-----------------+
1698 | Encapsulation | | DPDK 21.02 | | DPDK 21.02 |
1699 | GTP PSC | | OFED 5.2 | | OFED 5.2 |
1700 | | | rdma-core 35 | | rdma-core 35 |
1701 | | | ConnectX-6 Dx| | ConnectX-6 Dx |
1702 +-----------------------+-----------------+-----------------+
1703 | Encapsulation | | DPDK 21.02 | | DPDK 21.02 |
1704 | GENEVE TLV option | | OFED 5.2 | | OFED 5.2 |
1705 | | | rdma-core 34 | | rdma-core 34 |
1706 | | | ConnectX-6 Dx | | ConnectX-6 Dx |
1707 +-----------------------+-----------------+-----------------+
1708 | Modify Field | | DPDK 21.02 | | DPDK 21.02 |
1709 | | | OFED 5.2 | | OFED 5.2 |
1710 | | | rdma-core 35 | | rdma-core 35 |
1711 | | | ConnectX-5 | | ConnectX-5 |
1712 +-----------------------+-----------------+-----------------+
1713 | Connection tracking | | | | DPDK 21.05 |
1714 | | | N/A | | OFED 5.3 |
1715 | | | | | rdma-core 35 |
1716 | | | | | ConnectX-6 Dx |
1717 +-----------------------+-----------------+-----------------+
1719 .. table:: Minimal SW/HW versions for shared action offload
1722 +-----------------------+-----------------+-----------------+
1723 | Shared Action | with E-Switch | with NIC |
1724 +=======================+=================+=================+
1725 | RSS | | | | DPDK 20.11 |
1726 | | | N/A | | OFED 5.2 |
1727 | | | | | rdma-core 33 |
1728 | | | | | ConnectX-5 |
1729 +-----------------------+-----------------+-----------------+
1730 | Age | | DPDK 20.11 | | DPDK 20.11 |
1731 | | | OFED 5.2 | | OFED 5.2 |
1732 | | | rdma-core 32 | | rdma-core 32 |
1733 | | | ConnectX-6 Dx | | ConnectX-6 Dx |
1734 +-----------------------+-----------------+-----------------+
1735 | Count | | DPDK 21.05 | | DPDK 21.05 |
1736 | | | OFED 4.6 | | OFED 4.6 |
1737 | | | rdma-core 24 | | rdma-core 23 |
1738 | | | ConnectX-5 | | ConnectX-5 |
1739 +-----------------------+-----------------+-----------------+
1744 MARK and META items are interrelated with datapath - they might move from/to
1745 the applications in mbuf fields. Hence, zero value for these items has the
1746 special meaning - it means "no metadata are provided", not zero values are
1747 treated by applications and PMD as valid ones.
1749 Moreover in the flow engine domain the value zero is acceptable to match and
1750 set, and we should allow to specify zero values as rte_flow parameters for the
1751 META and MARK items and actions. In the same time zero mask has no meaning and
1752 should be rejected on validation stage.
1757 Flows are not cached in the driver.
1758 When stopping a device port, all the flows created on this port from the
1759 application will be flushed automatically in the background.
1760 After stopping the device port, all flows on this port become invalid and
1761 not represented in the system.
1762 All references to these flows held by the application should be discarded
1763 directly but neither destroyed nor flushed.
1765 The application should re-create the flows as required after the port restart.
1770 Compared to librte_net_mlx4 that implements a single RSS configuration per
1771 port, librte_net_mlx5 supports per-protocol RSS configuration.
1773 Since ``testpmd`` defaults to IP RSS mode and there is currently no
1774 command-line parameter to enable additional protocols (UDP and TCP as well
1775 as IP), the following commands must be entered from its CLI to get the same
1776 behavior as librte_net_mlx4::
1779 > port config all rss all
1785 This section demonstrates how to launch **testpmd** with Mellanox
1786 ConnectX-4/ConnectX-5/ConnectX-6/BlueField devices managed by librte_net_mlx5.
1788 #. Load the kernel modules::
1790 modprobe -a ib_uverbs mlx5_core mlx5_ib
1792 Alternatively if MLNX_OFED/MLNX_EN is fully installed, the following script
1795 /etc/init.d/openibd restart
1799 User space I/O kernel modules (uio and igb_uio) are not used and do
1800 not have to be loaded.
1802 #. Make sure Ethernet interfaces are in working order and linked to kernel
1803 verbs. Related sysfs entries should be present::
1805 ls -d /sys/class/net/*/device/infiniband_verbs/uverbs* | cut -d / -f 5
1814 #. Optionally, retrieve their PCI bus addresses for to be used with the allow list::
1817 for intf in eth2 eth3 eth4 eth5;
1819 (cd "/sys/class/net/${intf}/device/" && pwd -P);
1822 sed -n 's,.*/\(.*\),-a \1,p'
1831 #. Request huge pages::
1833 dpdk-hugepages.py --setup 2G
1835 #. Start testpmd with basic parameters::
1837 dpdk-testpmd -l 8-15 -n 4 -a 05:00.0 -a 05:00.1 -a 06:00.0 -a 06:00.1 -- --rxq=2 --txq=2 -i
1842 EAL: PCI device 0000:05:00.0 on NUMA socket 0
1843 EAL: probe driver: 15b3:1013 librte_net_mlx5
1844 PMD: librte_net_mlx5: PCI information matches, using device "mlx5_0" (VF: false)
1845 PMD: librte_net_mlx5: 1 port(s) detected
1846 PMD: librte_net_mlx5: port 1 MAC address is e4:1d:2d:e7:0c:fe
1847 EAL: PCI device 0000:05:00.1 on NUMA socket 0
1848 EAL: probe driver: 15b3:1013 librte_net_mlx5
1849 PMD: librte_net_mlx5: PCI information matches, using device "mlx5_1" (VF: false)
1850 PMD: librte_net_mlx5: 1 port(s) detected
1851 PMD: librte_net_mlx5: port 1 MAC address is e4:1d:2d:e7:0c:ff
1852 EAL: PCI device 0000:06:00.0 on NUMA socket 0
1853 EAL: probe driver: 15b3:1013 librte_net_mlx5
1854 PMD: librte_net_mlx5: PCI information matches, using device "mlx5_2" (VF: false)
1855 PMD: librte_net_mlx5: 1 port(s) detected
1856 PMD: librte_net_mlx5: port 1 MAC address is e4:1d:2d:e7:0c:fa
1857 EAL: PCI device 0000:06:00.1 on NUMA socket 0
1858 EAL: probe driver: 15b3:1013 librte_net_mlx5
1859 PMD: librte_net_mlx5: PCI information matches, using device "mlx5_3" (VF: false)
1860 PMD: librte_net_mlx5: 1 port(s) detected
1861 PMD: librte_net_mlx5: port 1 MAC address is e4:1d:2d:e7:0c:fb
1862 Interactive-mode selected
1863 Configuring Port 0 (socket 0)
1864 PMD: librte_net_mlx5: 0x8cba80: TX queues number update: 0 -> 2
1865 PMD: librte_net_mlx5: 0x8cba80: RX queues number update: 0 -> 2
1866 Port 0: E4:1D:2D:E7:0C:FE
1867 Configuring Port 1 (socket 0)
1868 PMD: librte_net_mlx5: 0x8ccac8: TX queues number update: 0 -> 2
1869 PMD: librte_net_mlx5: 0x8ccac8: RX queues number update: 0 -> 2
1870 Port 1: E4:1D:2D:E7:0C:FF
1871 Configuring Port 2 (socket 0)
1872 PMD: librte_net_mlx5: 0x8cdb10: TX queues number update: 0 -> 2
1873 PMD: librte_net_mlx5: 0x8cdb10: RX queues number update: 0 -> 2
1874 Port 2: E4:1D:2D:E7:0C:FA
1875 Configuring Port 3 (socket 0)
1876 PMD: librte_net_mlx5: 0x8ceb58: TX queues number update: 0 -> 2
1877 PMD: librte_net_mlx5: 0x8ceb58: RX queues number update: 0 -> 2
1878 Port 3: E4:1D:2D:E7:0C:FB
1879 Checking link statuses...
1880 Port 0 Link Up - speed 40000 Mbps - full-duplex
1881 Port 1 Link Up - speed 40000 Mbps - full-duplex
1882 Port 2 Link Up - speed 10000 Mbps - full-duplex
1883 Port 3 Link Up - speed 10000 Mbps - full-duplex
1890 This section demonstrates how to dump flows. Currently, it's possible to dump
1891 all flows with assistance of external tools.
1893 #. 2 ways to get flow raw file:
1895 - Using testpmd CLI:
1897 .. code-block:: console
1900 testpmd> flow dump <port> all <output_file>
1902 testpmd> flow dump <port> rule <rule_id> <output_file>
1904 - call rte_flow_dev_dump api:
1906 .. code-block:: console
1908 rte_flow_dev_dump(port, flow, file, NULL);
1910 #. Dump human-readable flows from raw file:
1912 Get flow parsing tool from: https://github.com/Mellanox/mlx_steering_dump
1914 .. code-block:: console
1916 mlx_steering_dump.py -f <output_file> -flowptr <flow_ptr>