1 .. SPDX-License-Identifier: BSD-3-Clause
2 Copyright 2018 The DPDK contributors
7 .. **Read this first.**
9 The text in the sections below explains how to update the release notes.
11 Use proper spelling, capitalization and punctuation in all sections.
13 Variable and config names should be quoted as fixed width text:
16 Build the docs and view the output file to ensure the changes are correct::
20 xdg-open build/doc/html/guides/rel_notes/release_18_11.html
26 .. This section should contain new features added in this release.
29 * **Add a title in the past tense with a full stop.**
31 Add a short 1-2 sentence description in the past tense.
32 The description should be enough to allow someone scanning
33 the release notes to understand the new feature.
35 If the feature adds a lot of sub-features you can use a bullet list
38 * Added feature foo to do something.
39 * Enhanced feature bar to do something else.
41 Refer to the previous release notes for examples.
43 Suggested order in release notes items:
44 * Core libs (EAL, mempool, ring, mbuf, buses)
45 * Device abstraction libs and PMDs
47 - cryptodev (lib, PMDs)
48 - eventdev (lib, PMDs)
51 * Apps, Examples, Tools (if significant)
53 This section is a comment. Do not overwrite or remove it.
54 Also, make sure to start the actual text at the margin.
55 =========================================================
57 * **Added support for using externally allocated memory in DPDK.**
59 DPDK has added support for creating new ``rte_malloc`` heaps referencing
60 memory that was created outside of DPDK's own page allocator, and using that
61 memory natively with any other DPDK library or data structure.
63 * **Added check for ensuring allocated memory is addressable by devices.**
65 Some devices can have addressing limitations so a new function,
66 ``rte_mem_check_dma_mask()``, has been added for checking that allocated
67 memory is not out of the device range. Since memory can now be allocated
68 dynamically after initialization, a DMA mask is stored and any new allocated
69 memory will be checked against it and rejected if it is out of range. If
70 more than one device has addressing limitations, the DMA mask is the more
73 * **Updated the C11 memory model version of the ring library.**
75 Added changes to decrease latency for architectures using the C11 memory
76 model version of the ring library.
78 On Cavium ThunderX2 platform, the changes decreased latency by 27-29%
79 and 3-15% for MPMC and SPSC cases respectively (with 2 lcores). The
80 real improvements may vary with the number of contending lcores and
83 * **Added hot-unplug handle mechanism.**
85 Added ``rte_dev_hotplug_handle_enable()`` and
86 ``rte_dev_hotplug_handle_disable()`` for enabling or disabling the hotplug
89 * **Added support for device multi-process hotplug.**
91 Added support for hotplug and hot-unplug in a multiprocessing scenario. Any
92 ethdev devices created in the primary process will be regarded as shared and
93 will be available for all DPDK processes. Synchronization between processes
94 will be done using DPDK IPC.
96 * **Added new Flow API actions to rewrite fields in packet headers.**
98 Added new Flow API actions to:
100 * Modify source and destination IP addresses in the outermost IPv4/IPv6
102 * Modify source and destination port numbers in the outermost TCP/UDP
105 * **Added new Flow API action to swap MAC addresses in Ethernet header.**
107 Added new Flow API action to swap the source and destination MAC
108 addresses in the outermost Ethernet header.
110 * **Add support to offload more flow match and actions for CXGBE PMD.**
112 The Flow API support has been enhanced for the CXGBE Poll Mode Driver to
115 * Match items: destination MAC address.
116 * Action items: push/pop/rewrite vlan header,
117 rewrite IP addresses in outermost IPv4/IPv6 header,
118 rewrite port numbers in outermost TCP/UDP header,
119 swap MAC addresses in outermost Ethernet header.
121 * **Added a devarg to use the latest supported vector path in i40e.**
123 A new devarg ``use-latest-supported-vec`` was introduced to allow users to
124 choose the latest vector path that the platform supported. For example, users
125 can use AVX2 vector path on BDW/HSW to get better performance.
127 * **Added support for SR-IOV in netvsc PMD.**
129 The ``netvsc`` poll mode driver now supports the Accelerated Networking
130 SR-IOV option in Hyper-V and Azure. This is an alternative to the previous
131 vdev_netvsc, tap, and failsafe drivers combination.
133 * **Added a new net driver for Marvell Armada 3k device.**
135 Added the new ``mvneta`` net driver for Marvell Armada 3k device. See the
136 :doc:`../nics/mvneta` NIC guide for more details on this new driver.
138 * **Added NXP ENETC PMD.**
140 Added the new enetc driver for the NXP enetc platform. See the
141 :doc:`../nics/enetc` NIC driver guide for more details on this new driver.
143 * **Added Ethernet poll mode driver for Aquantia aQtion family of 10G devices.**
145 Added the new ``atlantic`` ethernet poll mode driver for Aquantia XGBE devices.
146 See the :doc:`../nics/atlantic` NIC driver guide for more details on this
149 * **Updated mlx5 driver.**
151 Updated the mlx5 driver including the following changes:
153 * Improved security of PMD to prevent the NIC from getting stuck when
154 the application misbehaves.
155 * Reworked flow engine to supported e-switch flow rules (transfer attribute).
156 * Added support for header re-write(L2-L4), VXLAN encap/decap, count, match
157 on TCP flags and multiple flow groups with e-switch flow rules.
158 * Added support for match on metadata, VXLAN and MPLS encap/decap with flow
160 * Added support for ``RTE_ETH_DEV_CLOSE_REMOVE`` flag to provide better
161 support for representors.
162 * Added support for meson build.
163 * Fixed build issue with PPC.
164 * Added support for BlueField VF.
165 * Added support for externally allocated static memory for DMA.
167 * **Updated Solarflare network PMD.**
169 Updated the sfc_efx driver including the following changes:
171 * Added support for Rx scatter in EF10 datapath implementation.
172 * Added support for Rx descriptor status API in EF10 datapath implementation.
173 * Added support for TSO in EF10 datapath implementation.
174 * Added support for Tx descriptor status API in EF10 (ef10 and ef10_simple)
175 datapaths implementation.
177 * **Updated the enic driver.**
179 * Added AVX2-based vectorized Rx handler.
180 * Added VLAN and checksum offloads to the simple Tx handler.
181 * Added the "count" flow action.
182 * Enabled the virtual address IOVA mode.
184 * **Updated the failsafe driver.**
186 Updated the failsafe driver including the following changes:
188 * Added support for Rx and Tx queues start and stop.
189 * Added support for Rx and Tx queues deferred start.
190 * Added support for runtime Rx and Tx queues setup.
191 * Added support multicast MAC address set.
193 * **Added a devarg to use a PCAP interface physical MAC address.**
195 A new devarg ``phy_mac`` was introduced to allow users to use the physical
196 MAC address of the selected PCAP interface.
198 * **Added TAP Rx/Tx queues sharing with a secondary process.**
200 Added support to allow a secondary process to attach a TAP device created
201 in the primary process, probe the queues, and process Rx/Tx in a secondary
204 * **Added classification and metering support to SoftNIC PMD.**
206 Added support for flow classification (rte_flow API), and metering and
207 policing (rte_mtr API) to the SoftNIC PMD.
209 * **Added Crypto support to the Softnic PMD.**
211 The Softnic is now capable of processing symmetric crypto workloads such
212 as cipher, cipher-authentication chaining, and AEAD encryption and
213 decryption. This is achieved by calling DPDK Cryptodev APIs.
215 * **Added cryptodev port to port library.**
217 Cryptodev port is a shim layer in the port library that interacts with DPDK
218 Cryptodev PMDs including burst enqueuing and dequeuing crypto operations.
220 * **Added symmetric cryptographic actions to the pipeline library.**
222 In the pipeline library support was added for symmetric crypto action
223 parsing and an action handler was implemented. The action allows automatic
224 preparation of the crypto operation with the rules specified such as
225 algorithm, key, and IV, etc. for the cryptodev port to process.
227 * **Updated the AESNI MB PMD.**
229 The AESNI MB PMD has been updated with additional support for the AES-GCM
232 * **Added NXP CAAM JR PMD.**
234 Added the new caam job ring driver for NXP platforms. See the
235 :doc:`../cryptodevs/caam_jr` guide for more details on this new driver.
237 * **Added support for GEN3 devices to Intel QAT driver.**
239 Added support for the third generation of Intel QuickAssist devices.
241 * **Updated the QAT PMD.**
243 The QAT PMD was updated with additional support for:
245 * The AES-CMAC algorithm.
247 * **Added support for Dynamic Huffman Encoding to Intel QAT comp PMD.**
249 The Intel QuickAssist (QAT) compression PMD has been updated with support
250 for Dynamic Huffman Encoding for the Deflate algorithm.
252 * **Added Event Ethernet Tx Adapter.**
254 Added event ethernet Tx adapter library that provides configuration and
255 data path APIs for the ethernet transmit stage of an event driven packet
256 processing application. These APIs abstract the implementation of the
257 transmit stage and allow the application to use eventdev PMD support or
258 a common implementation.
260 * **Added Distributed Software Eventdev PMD.**
262 Added the new Distributed Software Event Device (DSW), which is a
263 pure-software eventdev driver distributing the work of scheduling
264 among all eventdev ports and the lcores using them. DSW, compared to
265 the SW eventdev PMD, sacrifices load balancing performance to
266 gain better event scheduling throughput and scalability.
268 * **Added extendable bucket feature to hash library (rte_hash).**
270 This new "extendable bucket" feature provides 100% insertion guarantee to
271 the capacity specified by the user by extending hash table with extra
272 buckets when needed to accommodate the unlikely event of intensive hash
273 collisions. In addition, the internal hashing algorithm was changed to use
274 partial-key hashing to improve memory efficiency and lookup performance.
276 * **Added lock free reader/writer concurrency to hash library (rte_hash).**
278 Lock free reader/writer concurrency prevents the readers from getting
279 blocked due to a preempted writer thread. This allows the hash library
280 to be used in scenarios where the writer thread runs on the control plane.
282 * **Added Traffic Pattern Aware Power Control Library.**
284 Added an experimental library that extends the Power Library and provides
285 empty_poll APIs. This feature measures how many times empty_polls are
286 executed per core and uses the number of empty polls as a hint for system
289 See the :doc:`../prog_guide/power_man` section of the DPDK Programmers
290 Guide document for more information.
292 * **Added JSON power policy interface for containers.**
294 Extended the Power Library and vm_power_manager sample app to allow power
295 policies to be submitted via a FIFO using JSON formatted strings. Previously
296 limited to Virtual Machines, this feature extends power policy functionality
297 to containers and host applications that need to have their cores frequency
298 controlled based on the rules contained in the policy.
300 * **Added Telemetry API.**
302 Added a new telemetry API which allows applications to transparently expose
303 their telemetry in JSON via a UNIX socket. The JSON can be consumed by any
304 Service Assurance agent, such as CollectD.
306 * **Updated KNI kernel module, rte_kni library, and KNI sample application.**
308 Updated the KNI kernel module with a new kernel module parameter,
309 ``carrier=[on|off]`` to allow the user to control the default carrier
310 state of the KNI kernel network interfaces. The default carrier state
311 is now set to ``off``, so the interfaces cannot be used until the
312 carrier state is set to ``on`` via ``rte_kni_update_link`` or
313 by writing ``1`` to ``/sys/devices/virtual/net/<iface>/carrier``.
314 In previous versions the default carrier state was left undefined.
315 See :doc:`../prog_guide/kernel_nic_interface` for more information.
317 Also added the new API function ``rte_kni_update_link()`` to allow the user
318 to set the carrier state of the KNI kernel network interface.
320 Also added a new command line flag ``-m`` to the KNI sample application to
321 monitor and automatically reflect the physical NIC carrier state to the
322 KNI kernel network interface with the new ``rte_kni_update_link()`` API.
323 See :doc:`../sample_app_ug/kernel_nic_interface` for more information.
325 * **Added ability to switch queue deferred start flag on testpmd app.**
327 Added a console command to testpmd app, giving ability to switch
328 ``rx_deferred_start`` or ``tx_deferred_start`` flag of the specified queue of
329 the specified port. The port must be stopped before the command call in order
330 to reconfigure queues.
332 * **Add a new sample application for vDPA.**
334 The vdpa sample application creates vhost-user sockets by using the
335 vDPA backend. vDPA stands for vhost Data Path Acceleration which utilizes
336 virtio ring compatible devices to serve virtio driver directly to enable
337 datapath acceleration. As vDPA driver can help to set up vhost datapath,
338 this application doesn't need to launch dedicated worker threads for vhost
339 enqueue/dequeue operations.
341 * **Added cryptodev FIPS validation example application.**
343 Added an example application to parse and perform symmetric cryptography
344 computation to the NIST Cryptographic Algorithm Validation Program (CAVP)
347 * **Allow unit test binary to take parameters from the environment.**
349 The unit test "test", or "dpdk-test", binary is often called from scripts,
350 which can make passing additional parameters, such as a coremask,
351 difficult. Support has been added to the application to allow it to take
352 additional command-line parameter values from the ``DPDK_TEST_PARAMS``
353 environment variable to make this application easier to use.
359 .. This section should contain API changes. Sample format:
361 * Add a short 1-2 sentence description of the API change.
362 Use fixed width quotes for ``function_names`` or ``struct_names``.
365 This section is a comment. Do not overwrite or remove it.
366 Also, make sure to start the actual text at the margin.
367 =========================================================
369 * eal: ``rte_memseg_list`` structure now has an additional flag indicating
370 whether the memseg list is externally allocated. This will have implications
371 for any users of memseg-walk-related functions, as they will now have to skip
372 externally allocated segments in most cases if the intent is to only iterate
373 over internal DPDK memory.
375 In addition the ``socket_id`` parameter across the entire DPDK has gained
376 additional meaning, as some socket ID's will now be representing externally
377 allocated memory. No changes will be required for existing code as backwards
378 compatibility will be kept, and those who do not use this feature will not
379 see these extra socket ID's. Any new API's must not check socket ID
380 parameters themselves, and must instead leave it to the memory subsystem to
381 decide whether socket ID is a valid one.
383 * eal: The following devargs functions, which were deprecated in 18.05,
384 were removed in 18.11:
385 ``rte_eal_parse_devargs_str()``, ``rte_eal_devargs_add()``,
386 ``rte_eal_devargs_type_count()``, and ``rte_eal_devargs_dump()``.
388 * eal: The parameters of the function ``rte_devargs_remove()`` have changed
389 from bus and device names to ``struct rte_devargs``.
391 * eal: The deprecated functions attach/detach were removed in 18.11.
392 ``rte_eal_dev_attach`` can be replaced by
393 ``rte_dev_probe`` or ``rte_eal_hotplug_add``.
394 ``rte_eal_dev_detach`` can be replaced by
395 ``rte_dev_remove`` or ``rte_eal_hotplug_remove``.
397 * eal: The scope of ``rte_eal_hotplug_add()``/``rte_dev_probe()``
398 and ``rte_eal_hotplug_remove()``/``rte_dev_remove()`` has been extended.
399 In the multi-process model, they will guarantee that the device is
400 attached or detached on all processes.
402 * mbuf: The ``__rte_mbuf_raw_free()`` and ``__rte_pktmbuf_prefree_seg()``
403 functions were deprecated since 17.05 and are replaced by
404 ``rte_mbuf_raw_free()`` and ``rte_pktmbuf_prefree_seg()``.
406 * ethdev: The deprecated functions attach/detach were removed in 18.11.
407 ``rte_eth_dev_attach()`` can be replaced by ``RTE_ETH_FOREACH_MATCHING_DEV``
408 and ``rte_dev_probe()`` or ``rte_eal_hotplug_add()``.
409 ``rte_eth_dev_detach()`` can be replaced by
410 ``rte_dev_remove()`` or ``rte_eal_hotplug_remove()``.
412 * ethdev: A call to ``rte_eth_dev_release_port()`` has been added in
413 ``rte_eth_dev_close()``. As a consequence, a closed port is freed
414 and seen as invalid because of its state ``RTE_ETH_DEV_UNUSED``.
415 This new behavior is enabled per driver for a migration period.
417 * A new device flag, ``RTE_ETH_DEV_NOLIVE_MAC_ADDR``, changes the order of
418 actions inside ``rte_eth_dev_start()`` regarding MAC set. Some NICs do not
419 support MAC changes once the port has started and with this new device
420 flag the MAC can be properly configured in any case. This is particularly
421 important for bonding.
423 * The default behavior of CRC strip offload has changed in this
424 release. Without any specific Rx offload flag, default behavior by a PMD is
425 now to strip CRC. ``DEV_RX_OFFLOAD_CRC_STRIP`` offload flag has been removed.
426 To request keeping CRC, application should set ``DEV_RX_OFFLOAD_KEEP_CRC``
429 * eventdev: The type of the second parameter to
430 ``rte_event_eth_rx_adapter_caps_get()`` has been changed from uint8_t to
433 * kni: By default, interface carrier status is ``off`` which means there won't
434 be any traffic. It can be set to ``on`` via ``rte_kni_update_link()`` API or
435 via ``sysfs`` interface: ``echo 1 > /sys/class/net/vEth0/carrier``.
437 Note interface should be ``up`` to be able to read/write sysfs interface.
438 When KNI sample application is used, ``-m`` parameter can be used to
439 automatically update the carrier status for the interface.
441 * kni: When ethtool support is enabled (``CONFIG_RTE_KNI_KMOD_ETHTOOL=y``)
442 ethtool commands ``ETHTOOL_GSET & ETHTOOL_SSET`` are no longer supported for
443 kernels that have ``ETHTOOL_GLINKSETTINGS & ETHTOOL_SLINKSETTINGS`` support.
444 This means ``ethtool "-a|--show-pause", "-s|--change"`` won't work, and
445 ``ethtool <iface>`` output will have less information.
451 .. This section should contain ABI changes. Sample format:
453 * Add a short 1-2 sentence description of the ABI change
454 that was announced in the previous releases and made in this release.
455 Use fixed width quotes for ``function_names`` or ``struct_names``.
458 This section is a comment. Do not overwrite or remove it.
459 Also, make sure to start the actual text at the margin.
460 =========================================================
462 * eal: added ``legacy_mem`` and ``single_file_segments`` values to
463 ``rte_config`` structure on account of improving DPDK usability when
464 using either ``--legacy-mem`` or ``--single-file-segments`` flags.
466 * eal: EAL library ABI version was changed due to previously announced work on
467 supporting external memory in DPDK:
469 - Structure ``rte_memseg_list`` now has a new field indicating length
470 of memory addressed by the segment list
471 - Structure ``rte_memseg_list`` now has a new flag indicating whether
472 the memseg list refers to external memory
473 - Structure ``rte_malloc_heap`` now has a new field indicating socket
474 ID the malloc heap belongs to
475 - Structure ``rte_mem_config`` has had its ``malloc_heaps`` array
476 resized from ``RTE_MAX_NUMA_NODES`` to ``RTE_MAX_HEAPS`` value
477 - Structure ``rte_malloc_heap`` now has a ``heap_name`` member
478 - Structure ``rte_eal_memconfig`` has been extended to contain next
479 socket ID for externally allocated segments
481 * eal: Added ``dma_maskbits`` to ``rte_mem_config`` for keeping the most
482 restrictive DMA mask based on the devices addressing limitations.
484 * eal: The structure ``rte_device`` has a new field to reference a
485 ``rte_bus``. It thus changes the size of the ``struct rte_device`` and the
486 inherited device structures of all buses.
489 Shared Library Versions
490 -----------------------
492 .. Update any library version updated in this release
493 and prepend with a ``+`` sign, like this:
496 + librte_cfgfile.so.2
499 This section is a comment. Do not overwrite or remove it.
500 =========================================================
502 The libraries prepended with a plus sign were incremented in this version.
508 librte_bitratestats.so.2
510 + librte_bus_dpaa.so.2
511 + librte_bus_fslmc.so.2
512 + librte_bus_ifpga.so.2
513 + librte_bus_pci.so.2
514 + librte_bus_vdev.so.2
515 + librte_bus_vmbus.so.2
518 librte_compressdev.so.1
519 librte_cryptodev.so.5
520 librte_distributor.so.1
523 + librte_ethdev.so.11
524 + librte_eventdev.so.6
525 librte_flow_classify.so.1
533 librte_latencystats.so.1
547 librte_pmd_ixgbe.so.2
548 librte_pmd_dpaa2_qdma.so.1
550 librte_pmd_softnic.so.1
551 librte_pmd_vhost.so.2
567 .. This section should contain new known issues in this release. Sample format:
569 * **Add title in present tense with full stop.**
571 Add a short 1-2 sentence description of the known issue
572 in the present tense. Add information on any known workarounds.
574 This section is a comment. Do not overwrite or remove it.
575 Also, make sure to start the actual text at the margin.
576 =========================================================
578 * When using SR-IOV (VF) support with netvsc PMD and the Mellanox mlx5
579 bifurcated driver the Linux netvsc device must be brought up before the
580 netvsc device is unbound and passed to the DPDK.
582 * IBM Power8 is not supported in this release of DPDK. IBM Power9 is
585 * ``AVX-512`` support has been disabled for ``GCC`` builds [1] because of a
586 crash [2]. This can affect ``native`` machine type build targets on the
587 platforms that support ``AVX512F`` like ``Intel Skylake`` processors, and
588 can cause a possible performance drop. The immediate workaround is to use
589 ``clang`` compiler on these platforms. The issue has been identified as a
590 GCC defect and reported to the GCC community [3]. Further actions will be
591 taken based on the GCC defect result.
593 - [1]: Commit 8d07c82b239f ("mk: disable gcc AVX512F support")
594 - [2]: https://bugs.dpdk.org/show_bug.cgi?id=97
595 - [3]: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=88096
601 .. This section should contain a list of platforms that were tested
606 * <vendor> platform with <vendor> <type of devices> combinations
611 * Other relevant details...
613 This section is a comment. Do not overwrite or remove it.
614 Also, make sure to start the actual text at the margin.
615 =========================================================
617 * Intel(R) platforms with Intel(R) NICs combinations
621 * Intel(R) Atom(TM) CPU C3758 @ 2.20GHz
622 * Intel(R) Xeon(R) CPU D-1541 @ 2.10GHz
623 * Intel(R) Xeon(R) CPU E5-2680 v2 @ 2.80GHz
624 * Intel(R) Xeon(R) CPU E5-2699 v3 @ 2.30GHz
625 * Intel(R) Xeon(R) CPU E5-2699 v4 @ 2.20GHz
626 * Intel(R) Xeon(R) Platinum 8180 CPU @ 2.50GHz
634 * Red Hat Enterprise Linux Server release 7.5
645 * Intel(R) 82599ES 10 Gigabit Ethernet Controller
647 * Firmware version: 0x61bf0001
648 * Device id (pf/vf): 8086:10fb / 8086:10ed
649 * Driver version: 5.2.3 (ixgbe)
651 * Intel(R) Corporation Ethernet Connection X552/X557-AT 10GBASE-T
653 * Firmware version: 0x800003e7
654 * Device id (pf/vf): 8086:15ad / 8086:15a8
655 * Driver version: 4.4.6 (ixgbe)
657 * Intel(R) Ethernet Converged Network Adapter X710-DA4 (4x10G)
659 * Firmware version: 6.01 0x80003221
660 * Device id (pf/vf): 8086:1572 / 8086:154c
661 * Driver version: 2.4.6 (i40e)
663 * Intel(R) Corporation Ethernet Connection X722 for 10GbE SFP+ (4x10G)
665 * Firmware version: 3.33 0x80000fd5 0.0.0
666 * Device id (pf/vf): 8086:37d0 / 8086:37cd
667 * Driver version: 2.4.6 (i40e)
669 * Intel(R) Ethernet Converged Network Adapter XXV710-DA2 (2x25G)
671 * Firmware version: 6.01 0x80003221
672 * Device id (pf/vf): 8086:158b / 8086:154c
673 * Driver version: 2.4.6 (i40e)
675 * Intel(R) Ethernet Converged Network Adapter XL710-QDA2 (2X40G)
677 * Firmware version: 6.01 0x8000321c
678 * Device id (pf/vf): 8086:1583 / 8086:154c
679 * Driver version: 2.4.6 (i40e)
681 * Intel(R) Corporation I350 Gigabit Network Connection
683 * Firmware version: 1.63, 0x80000dda
684 * Device id (pf/vf): 8086:1521 / 8086:1520
685 * Driver version: 5.4.0-k (igb)
687 * Intel(R) platforms with Mellanox(R) NICs combinations
691 * Intel(R) Xeon(R) Gold 6154 CPU @ 3.00GHz
692 * Intel(R) Xeon(R) CPU E5-2697A v4 @ 2.60GHz
693 * Intel(R) Xeon(R) CPU E5-2697 v3 @ 2.60GHz
694 * Intel(R) Xeon(R) CPU E5-2680 v2 @ 2.80GHz
695 * Intel(R) Xeon(R) CPU E5-2650 v4 @ 2.20GHz
696 * Intel(R) Xeon(R) CPU E5-2640 @ 2.50GHz
697 * Intel(R) Xeon(R) CPU E5-2620 v4 @ 2.10GHz
701 * Red Hat Enterprise Linux Server release 7.6 (Maipo)
702 * Red Hat Enterprise Linux Server release 7.5 (Maipo)
703 * Red Hat Enterprise Linux Server release 7.4 (Maipo)
704 * Red Hat Enterprise Linux Server release 7.3 (Maipo)
705 * Red Hat Enterprise Linux Server release 7.2 (Maipo)
710 * SUSE Linux Enterprise Server 15
712 * MLNX_OFED: 4.4-2.0.1.0
713 * MLNX_OFED: 4.5-0.3.1.0
717 * Mellanox(R) ConnectX(R)-3 Pro 40G MCX354A-FCC_Ax (2x40G)
719 * Host interface: PCI Express 3.0 x8
720 * Device ID: 15b3:1007
721 * Firmware version: 2.42.5000
723 * Mellanox(R) ConnectX(R)-4 10G MCX4111A-XCAT (1x10G)
725 * Host interface: PCI Express 3.0 x8
726 * Device ID: 15b3:1013
727 * Firmware version: 12.23.8022 and above
729 * Mellanox(R) ConnectX(R)-4 10G MCX4121A-XCAT (2x10G)
731 * Host interface: PCI Express 3.0 x8
732 * Device ID: 15b3:1013
733 * Firmware version: 12.23.8022 and above
735 * Mellanox(R) ConnectX(R)-4 25G MCX4111A-ACAT (1x25G)
737 * Host interface: PCI Express 3.0 x8
738 * Device ID: 15b3:1013
739 * Firmware version: 12.23.8022 and above
741 * Mellanox(R) ConnectX(R)-4 25G MCX4121A-ACAT (2x25G)
743 * Host interface: PCI Express 3.0 x8
744 * Device ID: 15b3:1013
745 * Firmware version: 12.23.8022 and above
747 * Mellanox(R) ConnectX(R)-4 40G MCX4131A-BCAT/MCX413A-BCAT (1x40G)
749 * Host interface: PCI Express 3.0 x8
750 * Device ID: 15b3:1013
751 * Firmware version: 12.23.8022 and above
753 * Mellanox(R) ConnectX(R)-4 40G MCX415A-BCAT (1x40G)
755 * Host interface: PCI Express 3.0 x16
756 * Device ID: 15b3:1013
757 * Firmware version: 12.23.8022 and above
759 * Mellanox(R) ConnectX(R)-4 50G MCX4131A-GCAT/MCX413A-GCAT (1x50G)
761 * Host interface: PCI Express 3.0 x8
762 * Device ID: 15b3:1013
763 * Firmware version: 12.23.8022 and above
765 * Mellanox(R) ConnectX(R)-4 50G MCX414A-BCAT (2x50G)
767 * Host interface: PCI Express 3.0 x8
768 * Device ID: 15b3:1013
769 * Firmware version: 12.23.8022 and above
771 * Mellanox(R) ConnectX(R)-4 50G MCX415A-GCAT/MCX416A-BCAT/MCX416A-GCAT (2x50G)
773 * Host interface: PCI Express 3.0 x16
774 * Device ID: 15b3:1013
775 * Firmware version: 12.23.8022 and above
776 * Firmware version: 12.23.8022 and above
778 * Mellanox(R) ConnectX(R)-4 50G MCX415A-CCAT (1x100G)
780 * Host interface: PCI Express 3.0 x16
781 * Device ID: 15b3:1013
782 * Firmware version: 12.23.8022 and above
784 * Mellanox(R) ConnectX(R)-4 100G MCX416A-CCAT (2x100G)
786 * Host interface: PCI Express 3.0 x16
787 * Device ID: 15b3:1013
788 * Firmware version: 12.23.8022 and above
790 * Mellanox(R) ConnectX(R)-4 Lx 10G MCX4121A-XCAT (2x10G)
792 * Host interface: PCI Express 3.0 x8
793 * Device ID: 15b3:1015
794 * Firmware version: 14.23.8022 and above
796 * Mellanox(R) ConnectX(R)-4 Lx 25G MCX4121A-ACAT (2x25G)
798 * Host interface: PCI Express 3.0 x8
799 * Device ID: 15b3:1015
800 * Firmware version: 14.23.8022 and above
802 * Mellanox(R) ConnectX(R)-5 100G MCX556A-ECAT (2x100G)
804 * Host interface: PCI Express 3.0 x16
805 * Device ID: 15b3:1017
806 * Firmware version: 16.23.8022 and above
808 * Mellanox(R) ConnectX(R)-5 Ex EN 100G MCX516A-CDAT (2x100G)
810 * Host interface: PCI Express 4.0 x16
811 * Device ID: 15b3:1019
812 * Firmware version: 16.23.8022 and above
814 * ARM platforms with Mellanox(R) NICs combinations
818 * Qualcomm ARM 1.1 2500MHz
822 * Red Hat Enterprise Linux Server release 7.5 (Maipo)
826 * Mellanox(R) ConnectX(R)-4 Lx 25G MCX4121A-ACAT (2x25G)
828 * Host interface: PCI Express 3.0 x8
829 * Device ID: 15b3:1015
830 * Firmware version: 14.24.0220
832 * Mellanox(R) ConnectX(R)-5 100G MCX556A-ECAT (2x100G)
834 * Host interface: PCI Express 3.0 x16
835 * Device ID: 15b3:1017
836 * Firmware version: 16.24.0220
838 * Mellanox(R) BlueField SmartNIC
840 * Mellanox(R) BlueField SmartNIC MT416842 (2x25G)
842 * Host interface: PCI Express 3.0 x16
843 * Device ID: 15b3:a2d2
844 * Firmware version: 18.24.0246
846 * SoC ARM cores running OS:
848 * CentOS Linux release 7.4.1708 (AltArch)
849 * MLNX_OFED 4.4-2.5.3.0
851 * DPDK application running on ARM cores inside SmartNIC
853 * ARM SoC combinations from NXP (with integrated NICs)
857 * NXP/Freescale QorIQ LS1046A with ARM Cortex A72
858 * NXP/Freescale QorIQ LS2088A with ARM Cortex A72
862 * Ubuntu 18.04.1 LTS with NXP QorIQ LSDK 1809 support packages
863 * Ubuntu 16.04.3 LTS with NXP QorIQ LSDK 1803 support packages