1 .. SPDX-License-Identifier: BSD-3-Clause
2 Copyright 2019 The DPDK contributors
4 .. include:: <isonum.txt>
12 * **Added support for --base-virtaddr EAL option to FreeBSD.**
14 The FreeBSD version of DPDK now also supports setting base virtual address
15 for mapping pages and resources into its address space.
17 * **Added Lock-free Stack for aarch64.**
19 Enabled the lock-free stack implementation for aarch64 platforms.
21 * **Extended pktmbuf mempool private structure.**
23 rte_pktmbuf_pool_private structure was extended to include flags field
24 for future compatibility.
25 As per 19.11 release this field is reserved and should be set to 0
28 +* **Changed mempool allocation behavior.**
30 Changed the mempool allocation behaviour so that objects no longer cross
31 pages by default. Note, this may consume more memory when using small memory
34 * **Added support for dynamic fields and flags in mbuf.**
36 This new feature adds the ability to dynamically register some room
37 for a field or a flag in the mbuf structure. This is typically used
38 for specific offload features, where adding a static field or flag
39 in the mbuf is not justified.
41 * **Added support for hairpin queues.**
43 On supported NICs, we can now setup hairpin queues which will offload packets
44 from the wire, back to the wire.
46 * **Added flow tag in rte_flow.**
48 The ``SET_TAG`` action and ``TAG`` item have been added to support transient
51 * **Extended metadata support in rte_flow.**
53 Flow metadata has been extended to both Rx and Tx.
55 * Tx metadata can also be set by SET_META action of rte_flow.
56 * Rx metadata is delivered to the host via a dynamic field of ``rte_mbuf``
57 with ``PKT_RX_DYNF_METADATA``.
59 * **Added ethdev API to set supported packet types.**
61 * Added new API ``rte_eth_dev_set_ptypes`` which allows an application to
62 inform a PMD about a reduced range of packet types to handle.
63 * This scheme will allow PMDs to avoid lookup of internal ptype table on Rx
64 and thereby improve Rx performance if the application wishes to do so.
66 * **Added Rx offload flag to enable or disable RSS update.**
68 * Added new Rx offload flag ``DEV_RX_OFFLOAD_RSS_HASH`` which can be used to
69 enable/disable PMDs write to ``rte_mbuf::hash::rss``.
70 * PMDs notify the validity of ``rte_mbuf::hash:rss`` to the application
71 by enabling ``PKT_RX_RSS_HASH`` flag in ``rte_mbuf::ol_flags``.
73 * **Added Rx/Tx packet burst mode "get" API.**
75 Added two new functions ``rte_eth_rx_burst_mode_get`` and
76 ``rte_eth_tx_burst_mode_get`` that allow an application
77 to retrieve the mode information about Rx/Tx packet burst
78 such as Scalar or Vector, and Vector technology like AVX2.
80 * **Added Hisilicon hns3 PMD.**
82 Added the new ``hns3`` net driver for the inbuilt Hisilicon Network
83 Subsystem 3 (HNS3) network engine found in the Hisilicon Kunpeng 920 SoC.
84 See the :doc:`../nics/hns3` guide for more details on this new driver.
86 * **Added NXP PFE PMD.**
88 Added the new PFE driver for the NXP LS1012A platform. See the
89 :doc:`../nics/pfe` NIC driver guide for more details on this new driver.
91 * **Updated Broadcom bnxt driver.**
93 Updated Broadcom bnxt driver with new features and improvements, including:
95 * Added support for hot firmware upgrade.
96 * Added support for error recovery.
97 * Added support for querying and using COS classification in hardware.
98 * Added LRO support Thor devices.
99 * Update HWRM API to version 1.10.1.6
101 * **Updated the enic driver.**
103 * Added support for Geneve with options offload.
104 * Added flow API implementation based on VIC Flow Manager API.
106 * **Updated iavf PMD.**
108 Enable AVX2 data path for iavf PMD.
110 * **Updated the Intel e1000 driver.**
112 Added support for the ``RTE_ETH_DEV_CLOSE_REMOVE`` flag.
114 * **Updated the Intel ixgbe driver.**
116 Added support for the ``RTE_ETH_DEV_CLOSE_REMOVE`` flag.
118 * **Updated the Intel i40e driver.**
120 Added support for the ``RTE_ETH_DEV_CLOSE_REMOVE`` flag.
122 * **Updated the Intel fm10k driver.**
124 Added support for the ``RTE_ETH_DEV_CLOSE_REMOVE`` flag.
126 * **Updated the Intel ice driver.**
128 Updated the Intel ice driver with new features and improvements, including:
130 * Added support for device-specific DDP package loading.
131 * Added support for handling Receive Flex Descriptor.
132 * Added support for protocol extraction on per Rx queue.
133 * Added support for Flow Director filter based on generic filter framework.
134 * Added support for the ``RTE_ETH_DEV_CLOSE_REMOVE`` flag.
135 * Generic filter enhancement
136 - Supported pipeline mode.
137 - Supported new packet type like PPPoE for switch filter.
138 * Supported input set change and symmetric hash by rte_flow RSS action.
139 * Added support for GTP Tx checksum offload.
140 * Added new device IDs to support E810_XXV devices.
142 * **Updated the Huawei hinic driver.**
144 Updated the Huawei hinic driver with new features and improvements, including:
146 * Enabled SR-IOV - Partially supported at this point, VFIO only.
147 * Supported VLAN filter and VLAN offload.
148 * Supported Unicast MAC filter and Multicast MAC filter.
149 * Supported Flow API for LACP, VRRP, BGP and so on.
150 * Supported FW version get.
152 * **Updated Mellanox mlx5 driver.**
154 Updated Mellanox mlx5 driver with new features and improvements, including:
156 * Added support for VLAN pop flow offload command.
157 * Added support for VLAN push flow offload command.
158 * Added support for VLAN set PCP offload command.
159 * Added support for VLAN set VID offload command.
160 * Added support for matching on packets withe Geneve tunnel header.
161 * Added hairpin support.
162 * Added ConnectX-6 Dx support.
163 * Flow engine selected based on RDMA Core library version.
164 DV flow engine selected if version is rdma-core-24.0 or higher.
165 Verbs flow engine selected otherwise.
167 * **Updated the AF_XDP PMD.**
169 Updated the AF_XDP PMD. The new features include:
171 * Enabled zero copy between application mempools and UMEM by enabling the
172 ``XDP_UMEM_UNALIGNED_CHUNKS UMEM`` flag.
174 * **Added cryptodev asymmetric session-less operation.**
176 Added a session-less option to the cryptodev asymmetric structure. It works
177 the same way as symmetric crypto, and the corresponding transform is used
178 directly by the crypto operation.
180 * **Added Marvell NITROX symmetric crypto PMD.**
182 Added a symmetric crypto PMD for Marvell NITROX V security processor.
183 See the :doc:`../cryptodevs/nitrox` guide for more details on this new PMD.
185 * **Added asymmetric support to Marvell OCTEON TX crypto PMD.**
187 Added support for asymmetric operations to Marvell OCTEON TX crypto PMD.
188 Supports RSA and modexp operations.
190 * **Added Marvell OCTEON TX2 crypto PMD.**
192 Added a new PMD for hardware crypto offload block on ``OCTEON TX2``
195 See ``cryptodevs/octeontx2`` for more details
197 * **Updated NXP crypto PMDs for PDCP support.**
199 Added PDCP support to the DPAA_SEC and DPAA2_SEC PMDs using rte_security
200 APIs. Support has been added for all sequence number sizes for control and
201 user plane. Test and test-crypto-perf applications have been updated for
204 * **Updated the AESNI-MB PMD.**
206 * Added support for intel-ipsec-mb version 0.53.
208 * **Updated the AESNI-GCM PMD.**
210 * Added support for intel-ipsec-mb version 0.53.
211 * Added support for in-place chained mbufs with AES-GCM algorithm.
213 * **Enabled Single Pass GCM acceleration on QAT GEN3.**
215 Added support for Single Pass GCM, available on QAT GEN3 only (Intel
216 QuickAssist Technology P5xxx). It is automatically chosen instead of the
217 classic 2-pass mode when running on QAT GEN3, significantly improving
218 the performance of AES GCM operations.
220 * **Updated the Intel QuickAssist Technology (QAT) asymmetric crypto PMD.**
222 * Added support for asymmetric session-less operations.
223 * Added support for RSA algorithm with pair ``(n, d)`` private key
225 * Added support for RSA algorithm with quintuple private key representation.
227 * **Updated the Intel QuickAssist Technology (QAT) compression PMD.**
229 Added stateful decompression support in the Intel QuickAssist Technology PMD.
230 Please note that stateful compression is not supported.
232 * **Added external buffers support for dpdk-test-compress-perf tool.**
234 Added a command line option to the ``dpdk-test-compress-perf`` tool to
235 allocate and use memory zones as external buffers instead of keeping the
236 data directly in mbuf areas.
238 * **Updated the IPSec library.**
240 * Added Security Associations (SA) Database API to ``librte_ipsec``. A new
241 test-sad application has also been introduced to evaluate and perform
242 custom functional and performance tests for an IPsec SAD implementation.
244 * Support fragmented packets in inline crypto processing mode with fallback
245 ``lookaside-none`` session. Corresponding changes are also added in the
246 IPsec Security Gateway application.
248 * **Introduced FIFO for NTB PMD.**
250 Introduced FIFO for NTB (Non-transparent Bridge) PMD to support
251 packet based processing.
253 * **Added eBPF JIT support for arm64.**
255 Added eBPF JIT support for arm64 architecture to improve the eBPF program
258 * **Added RIB and FIB (Routing/Forwarding Information Base) libraries.**
260 Added Routing and Forwarding Information Base (RIB/FIB) libraries. RIB and
261 FIB can replace the LPM (Longest Prefix Match) library with better control
262 plane (RIB) performance. The data plane (FIB) can be extended with new
265 * **Updated testpmd with a command for ptypes.**
267 * Added a console command to testpmd app, ``show port (port_id) ptypes`` which
268 gives ability to print port supported ptypes in different protocol layers.
269 * Packet type detection disabled by default for the supported PMDs.
271 * **Added new l2fwd-event sample application.**
273 Added an example application ``l2fwd-event`` that adds event device support to
274 the traditional l2fwd example. It demonstrates usage of poll and event mode IO
275 mechanism under a single application.
277 * **Added build support for Link Time Optimization.**
279 LTO is an optimization technique used by the compiler to perform whole
280 program analysis and optimization at link time. In order to do that
281 compilers store their internal representation of the source code that
282 the linker uses at the final stage of the compilation process.
284 See :doc:`../prog_guide/lto` for more information:
286 * **Added IOVA as VA support for KNI.**
288 * Added IOVA = VA support for KNI. KNI can operate in IOVA = VA mode when
289 ``iova-mode=va`` EAL option is passed to the application or when bus IOVA
290 scheme is selected as RTE_IOVA_VA. This mode only works on Linux Kernel
291 versions 4.10.0 and above.
293 * Due to IOVA to KVA address translations, based on the KNI use case there
294 can be a performance impact. For mitigation, forcing IOVA to PA via EAL
295 ``--iova-mode=pa`` option can be used, IOVA_DC bus iommu scheme can also
296 result in IOVA as PA.
302 * Removed library-level ABI versions. These have been replaced with a single
303 project-level ABI version for non-experimental libraries and an ABI version of
304 ``0`` for experimental libraries. Review the :doc:`../contributing/abi_policy`
305 and :doc:`../contributing/abi_versioning` guides for more information.
307 * Removed duplicated set of commands for Rx offload configuration from testpmd::
309 port config all crc-strip|scatter|rx-cksum|rx-timestamp|
310 hw-vlan|hw-vlan-filter|hw-vlan-strip|hw-vlan-extend on|off
312 The testpmd command set that can be used instead in order to enable or
313 disable Rx offloading on all Rx queues of a port is::
315 port config <port_id> rx_offload crc_strip|scatter|
316 ipv4_cksum|udp_cksum|tcp_cksum|timestamp|
317 vlan_strip|vlan_filter|vlan_extend on|off
319 * Removed AF_XDP pmd_zero copy vdev argument. Support is now auto-detected.
321 * The following sample applications have been removed in this release:
324 * L3 Forwarding in a Virtualization Environment
326 * Netmap Compatibility
327 * Quota and Watermark
330 * Removed arm64-dpaa2-* build config. arm64-dpaa-* can now build for both
331 dpaa and dpaa2 platforms.
337 * eal: made the ``lcore_config`` struct and global symbol private.
339 * eal: removed the ``rte_cpu_check_supported`` function, replaced by
340 ``rte_cpu_is_supported`` since dpdk v17.08.
342 * eal: removed the ``rte_malloc_virt2phy`` function, replaced by
343 ``rte_malloc_virt2iova`` since v17.11.
345 * eal: made the ``rte_config`` struct and ``rte_eal_get_configuration``
348 * mem: hid the internal ``malloc_heap`` structure and the
349 ``rte_malloc_heap.h`` header.
351 * vfio: removed ``rte_vfio_dma_map`` and ``rte_vfio_dma_unmap`` that have
352 been marked as deprecated in release 19.05.
353 ``rte_vfio_container_dma_map`` and ``rte_vfio_container_dma_unmap`` can
354 be used as substitutes.
356 * pci: removed the following functions deprecated since dpdk v17.11:
358 - ``eal_parse_pci_BDF`` replaced by ``rte_pci_addr_parse``
359 - ``eal_parse_pci_DomBDF`` replaced by ``rte_pci_addr_parse``
360 - ``rte_eal_compare_pci_addr`` replaced by ``rte_pci_addr_cmp``
362 * The network structure ``esp_tail`` has been prefixed by ``rte_``.
364 * The network definitions of PPPoE ethertypes have been prefixed by ``RTE_``.
366 * The network structure for MPLS has been prefixed by ``rte_``.
368 * ethdev: changed ``rte_eth_dev_infos_get`` return value from ``void`` to
369 ``int`` to provide a way to report various error conditions.
371 * ethdev: changed ``rte_eth_promiscuous_enable`` and
372 ``rte_eth_promiscuous_disable`` return value from ``void`` to ``int`` to
373 provide a way to report various error conditions.
375 * ethdev: changed ``rte_eth_allmulticast_enable`` and
376 ``rte_eth_allmulticast_disable`` return value from ``void`` to ``int`` to
377 provide a way to report various error conditions.
379 * ethdev: changed ``rte_eth_dev_xstats_reset`` return value from ``void`` to
380 ``int`` to provide a way to report various error conditions.
382 * ethdev: changed ``rte_eth_link_get`` and ``rte_eth_link_get_nowait``
383 return value from ``void`` to ``int`` to provide a way to report various
386 * ethdev: changed ``rte_eth_macaddr_get`` return value from ``void`` to
387 ``int`` to provide a way to report various error conditions.
389 * ethdev: changed ``rte_eth_dev_owner_delete`` return value from ``void`` to
390 ``int`` to provide a way to report various error conditions.
392 * ethdev: The deprecated function ``rte_eth_dev_count`` was removed.
393 The function ``rte_eth_dev_count_avail`` is a drop-in replacement.
394 If the intent is to iterate over ports, ``RTE_ETH_FOREACH_*`` macros
395 are better port iterators.
397 * ethdev: ``RTE_FLOW_ITEM_TYPE_META`` data endianness altered to host one.
398 Due to the new dynamic metadata field in mbuf is host-endian either, there
399 is a minor compatibility issue for applications in case of 32-bit values
402 * ethdev: the tx_metadata mbuf field is moved to dynamic one.
403 ``PKT_TX_METADATA`` flag is replaced with ``PKT_TX_DYNF_METADATA``.
404 ``DEV_TX_OFFLOAD_MATCH_METADATA`` offload flag is removed, now metadata
405 support in PMD is engaged on dynamic field registration.
407 * event: The function ``rte_event_eth_tx_adapter_enqueue`` takes an additional
408 input as ``flags``. Flag ``RTE_EVENT_ETH_TX_ADAPTER_ENQUEUE_SAME_DEST`` which
409 has been introduced in this release is used when all the packets enqueued in
410 the Tx adapter are destined for the same Ethernet port and Tx queue.
412 * sched: The pipe nodes configuration parameters such as number of pipes,
413 pipe queue sizes, pipe profiles, etc., are moved from port level structure
414 to subport level. This allows different subports of the same port to
415 have different configuration for the pipe nodes.
421 * policy: Please note the revisions to the :doc:`../contributing/abi_policy`
422 introducing major ABI versions, with DPDK 19.11 becoming the first major
423 version ``v20``. ABI changes to add new features continue to be permitted in
424 subsequent releases, with the condition that ABI compatibility with the major
425 ABI version is maintained.
427 * net: The Ethernet address and other header definitions have changed
428 attributes. They have been modified to be aligned on 2-byte boundaries.
429 These changes should not impact normal usage because drivers naturally
430 align the Ethernet header on receive and all known encapsulations
431 preserve the alignment of the header.
433 * security: The field ``replay_win_sz`` has been moved from the ipsec library
434 based ``rte_ipsec_sa_prm`` structure to security library based structure
435 ``rte_security_ipsec_xform``, which specify the anti-replay window size
436 to enable sequence replay attack handling.
438 * ipsec: The field ``replay_win_sz`` has been removed from the structure
439 ``rte_ipsec_sa_prm`` as it has been added to the security library.
441 * ethdev: Added 32-bit fields for maximum LRO aggregated packet size, in
442 struct ``rte_eth_dev_info`` for the port capability and in struct
443 ``rte_eth_rxmode`` for the port configuration.
444 Application should use the new field in struct ``rte_eth_rxmode`` to configure
446 PMD should use the new field in struct ``rte_eth_dev_info`` to report the
447 supported port capability.
450 Shared Library Versions
451 -----------------------
453 The libraries prepended with a plus sign were incremented in this version.
459 librte_bitratestats.so.2
462 librte_bus_fslmc.so.2
463 librte_bus_ifpga.so.2
466 librte_bus_vmbus.so.2
469 librte_compressdev.so.1
470 librte_cryptodev.so.8
471 librte_distributor.so.1
474 + librte_ethdev.so.13
475 + librte_eventdev.so.8
477 librte_flow_classify.so.1
486 librte_latencystats.so.1
500 librte_pmd_ixgbe.so.2
501 librte_pmd_dpaa2_qdma.so.1
503 librte_pmd_softnic.so.1
504 librte_pmd_vhost.so.2
513 + librte_security.so.3
523 * Intel\ |reg| platforms with Intel\ |reg| NICs combinations
527 * Intel\ |reg| Atom\ |trade| CPU C3758 @ 2.20GHz
528 * Intel\ |reg| Atom\ |trade| CPU C3858 @ 2.00GHz
529 * Intel\ |reg| Atom\ |trade| CPU C3958 @ 2.00GHz
530 * Intel\ |reg| Xeon\ |reg| CPU D-1541 @ 2.10GHz
531 * Intel\ |reg| Xeon\ |reg| CPU D-1553N @ 2.30GHz
532 * Intel\ |reg| Xeon\ |reg| CPU E5-2680 0 @ 2.70GHz
533 * Intel\ |reg| Xeon\ |reg| CPU E5-2680 v2 @ 2.80GHz
534 * Intel\ |reg| Xeon\ |reg| CPU E5-2699 v3 @ 2.30GHz
535 * Intel\ |reg| Xeon\ |reg| CPU E5-2699 v4 @ 2.20GHz
536 * Intel\ |reg| Xeon\ |reg| Gold 6139 CPU @ 2.30GHz
537 * Intel\ |reg| Xeon\ |reg| Gold 6252N CPU @ 2.30GHz
538 * Intel\ |reg| Xeon\ |reg| Platinum 8180 CPU @ 2.50GHz
539 * Intel\ |reg| Xeon\ |reg| Platinum 8280M CPU @ 2.70GHz
546 * Red Hat Enterprise Linux Server release 8.0
547 * Red Hat Enterprise Linux Server release 7.6
557 * Intel\ |reg| Corporation Ethernet Controller E810-C for SFP (2x25G)
559 * Firmware version: 1.02 0x80002084 1.2538.0/1.02 0x80002082 1.2538.0
560 * Device id (pf): 8086:1593
561 * Driver version: 0.12.25 (ice)
563 * Intel\ |reg| Corporation Ethernet Controller E810-C for SFP (2x100G)
565 * Firmware version: 1.02 0x80002081 1.2538.0
566 * Device id (pf): 8086:1592
567 * Driver version: 0.12.25 (ice)
569 * Intel\ |reg| 82599ES 10 Gigabit Ethernet Controller
571 * Firmware version: 0x61bf0001
572 * Device id (pf/vf): 8086:10fb / 8086:10ed
573 * Driver version: 5.6.1 (ixgbe)
575 * Intel\ |reg| Corporation Ethernet Connection X552/X557-AT 10GBASE-T
577 * Firmware version: 0x800003e7
578 * Device id (pf/vf): 8086:15ad / 8086:15a8
579 * Driver version: 5.1.0 (ixgbe)
581 * Intel\ |reg| Corporation Ethernet Controller 10G X550T
583 * Firmware version: 0x80000482
584 * Device id (pf): 8086:1563
585 * Driver version: 5.6.1 (ixgbe)
587 * Intel\ |reg| Ethernet Converged Network Adapter X710-DA4 (4x10G)
589 * Firmware version: 7.00 0x80004cdb
590 * Device id (pf/vf): 8086:1572 / 8086:154c
591 * Driver version: 2.9.21 (i40e)
593 * Intel\ |reg| Corporation Ethernet Connection X722 for 10GbE SFP+ (4x10G)
595 * Firmware version: 4.10 0x80001a3c
596 * Device id (pf/vf): 8086:37d0 / 8086:37cd
597 * Driver version: 2.9.21 (i40e)
599 * Intel\ |reg| Ethernet Converged Network Adapter XXV710-DA2 (2x25G)
601 * Firmware version: 7.00 0x80004cf8
602 * Device id (pf/vf): 8086:158b / 8086:154c
603 * Driver version: 2.9.21 (i40e)
605 * Intel\ |reg| Ethernet Converged Network Adapter XL710-QDA2 (2X40G)
607 * Firmware version: 7.00 0x80004c97
608 * Device id (pf/vf): 8086:1583 / 8086:154c
609 * Driver version: 2.9.21 (i40e)
611 * Intel\ |reg| Corporation I350 Gigabit Network Connection
613 * Firmware version: 1.63, 0x80000cbc
614 * Device id (pf/vf): 8086:1521 / 8086:1520
615 * Driver version: 5.4.0-k (igb)
617 * Intel\ |reg| Corporation I210 Gigabit Network Connection
619 * Firmware version: 3.25, 0x800006eb
620 * Device id (pf): 8086:1533
621 * Driver version: 5.4.0-k(igb)
623 * ARMv8 SoC combinations from Marvell (with integrated NICs)
627 * CN83xx, CN96xx, CN93xx
629 * OS (Based on Marvell OCTEON TX SDK-10.1.2.0):
638 * Intel\ |reg| platforms with Mellanox\ |reg| NICs combinations
642 * Intel\ |reg| Xeon\ |reg| Gold 6154 CPU @ 3.00GHz
643 * Intel\ |reg| Xeon\ |reg| CPU E5-2697A v4 @ 2.60GHz
644 * Intel\ |reg| Xeon\ |reg| CPU E5-2697 v3 @ 2.60GHz
645 * Intel\ |reg| Xeon\ |reg| CPU E5-2680 v2 @ 2.80GHz
646 * Intel\ |reg| Xeon\ |reg| CPU E5-2650 v4 @ 2.20GHz
647 * Intel\ |reg| Xeon\ |reg| CPU E5-2640 @ 2.50GHz
648 * Intel\ |reg| Xeon\ |reg| CPU E5-2620 v4 @ 2.10GHz
652 * Red Hat Enterprise Linux Server release 8.0 (Maipo)
653 * Red Hat Enterprise Linux Server release 7.7 (Maipo)
654 * Red Hat Enterprise Linux Server release 7.6 (Maipo)
655 * Red Hat Enterprise Linux Server release 7.5 (Maipo)
656 * Red Hat Enterprise Linux Server release 7.4 (Maipo)
657 * Red Hat Enterprise Linux Server release 7.3 (Maipo)
658 * Red Hat Enterprise Linux Server release 7.2 (Maipo)
663 * SUSE Linux Enterprise Server 15
667 * MLNX_OFED 4.6-1.0.1.1
668 * MLNX_OFED 4.7-1.0.0.1
669 * MLNX_OFED 4.7-3.1.9.0 and above
673 * Linux 5.3 and above
677 * rdma-core-24.1-1 and above
681 * Mellanox\ |reg| ConnectX\ |reg|-3 Pro 40G MCX354A-FCC_Ax (2x40G)
683 * Host interface: PCI Express 3.0 x8
684 * Device ID: 15b3:1007
685 * Firmware version: 2.42.5000
687 * Mellanox\ |reg| ConnectX\ |reg|-4 10G MCX4111A-XCAT (1x10G)
689 * Host interface: PCI Express 3.0 x8
690 * Device ID: 15b3:1013
691 * Firmware version: 12.26.2032 and above
693 * Mellanox\ |reg| ConnectX\ |reg|-4 10G MCX4121A-XCAT (2x10G)
695 * Host interface: PCI Express 3.0 x8
696 * Device ID: 15b3:1013
697 * Firmware version: 12.26.2032 and above
699 * Mellanox\ |reg| ConnectX\ |reg|-4 25G MCX4111A-ACAT (1x25G)
701 * Host interface: PCI Express 3.0 x8
702 * Device ID: 15b3:1013
703 * Firmware version: 12.26.2032 and above
705 * Mellanox\ |reg| ConnectX\ |reg|-4 25G MCX4121A-ACAT (2x25G)
707 * Host interface: PCI Express 3.0 x8
708 * Device ID: 15b3:1013
709 * Firmware version: 12.26.2032 and above
711 * Mellanox\ |reg| ConnectX\ |reg|-4 40G MCX4131A-BCAT/MCX413A-BCAT (1x40G)
713 * Host interface: PCI Express 3.0 x8
714 * Device ID: 15b3:1013
715 * Firmware version: 12.26.2032 and above
717 * Mellanox\ |reg| ConnectX\ |reg|-4 40G MCX415A-BCAT (1x40G)
719 * Host interface: PCI Express 3.0 x16
720 * Device ID: 15b3:1013
721 * Firmware version: 12.26.2032 and above
723 * Mellanox\ |reg| ConnectX\ |reg|-4 50G MCX4131A-GCAT/MCX413A-GCAT (1x50G)
725 * Host interface: PCI Express 3.0 x8
726 * Device ID: 15b3:1013
727 * Firmware version: 12.26.2032 and above
729 * Mellanox\ |reg| ConnectX\ |reg|-4 50G MCX414A-BCAT (2x50G)
731 * Host interface: PCI Express 3.0 x8
732 * Device ID: 15b3:1013
733 * Firmware version: 12.26.2032 and above
735 * Mellanox\ |reg| ConnectX\ |reg|-4 50G MCX415A-GCAT/MCX416A-BCAT/MCX416A-GCAT (2x50G)
737 * Host interface: PCI Express 3.0 x16
738 * Device ID: 15b3:1013
739 * Firmware version: 12.26.2032 and above
740 * Firmware version: 12.26.2032 and above
742 * Mellanox\ |reg| ConnectX\ |reg|-4 50G MCX415A-CCAT (1x100G)
744 * Host interface: PCI Express 3.0 x16
745 * Device ID: 15b3:1013
746 * Firmware version: 12.26.2032 and above
748 * Mellanox\ |reg| ConnectX\ |reg|-4 100G MCX416A-CCAT (2x100G)
750 * Host interface: PCI Express 3.0 x16
751 * Device ID: 15b3:1013
752 * Firmware version: 12.26.2032 and above
754 * Mellanox\ |reg| ConnectX\ |reg|-4 Lx 10G MCX4121A-XCAT (2x10G)
756 * Host interface: PCI Express 3.0 x8
757 * Device ID: 15b3:1015
758 * Firmware version: 14.26.2032 and above
760 * Mellanox\ |reg| ConnectX\ |reg|-4 Lx 25G MCX4121A-ACAT (2x25G)
762 * Host interface: PCI Express 3.0 x8
763 * Device ID: 15b3:1015
764 * Firmware version: 14.26.2032 and above
766 * Mellanox\ |reg| ConnectX\ |reg|-5 100G MCX556A-ECAT (2x100G)
768 * Host interface: PCI Express 3.0 x16
769 * Device ID: 15b3:1017
770 * Firmware version: 16.26.2032 and above
772 * Mellanox\ |reg| ConnectX\ |reg|-5 Ex EN 100G MCX516A-CDAT (2x100G)
774 * Host interface: PCI Express 4.0 x16
775 * Device ID: 15b3:1019
776 * Firmware version: 16.26.2032 and above
778 * IBM Power 9 platforms with Mellanox\ |reg| NICs combinations
782 * POWER9 2.2 (pvr 004e 1202) 2300MHz
786 * Ubuntu 18.04.1 LTS (Bionic Beaver)
790 * Mellanox\ |reg| ConnectX\ |reg|-5 100G MCX556A-ECAT (2x100G)
792 * Host interface: PCI Express 3.0 x16
793 * Device ID: 15b3:1017
794 * Firmware version: 16.26.1040
798 * MLNX_OFED 4.7-1.0.0.2