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_19_02.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 freeing hugepages exactly as originally allocated.**
59 Some applications using memory event callbacks (especially for managing
60 RDMA memory regions) require that memory be freed back to the system
61 exactly as it was originally allocated. These applications typically
62 also require that a malloc allocation not span across two separate
63 hugepage allocations. A new ``--match-allocations`` EAL init flag has
64 been added to fulfill both of these requirements.
66 * **Added API to register external memory in DPDK.**
68 A new ``rte_extmem_register``/``rte_extmem_unregister`` API was added to allow
69 chunks of external memory to be registered with DPDK without adding them to
72 * **Added support for using virtio-user without hugepages.**
74 The ``--no-huge`` mode was augmented to use memfd-backed memory (on systems
75 that support memfd), to allow using virtio-user-based NICs without
78 * **Release of the ENA PMD v2.0.0.**
80 Version 2.0.0 of the ENA PMD was added with the following additions:
82 * Added Low Latency Queue v2 (LLQv2). This feature reduces the latency
83 of the packets by pushing the header directly through the PCI to the
84 device. This allows the NIC to start handle packets right after the doorbell
85 without waiting for DMA.
86 * Added independent configuration of HW Tx and Rx ring depths.
87 * Added support for up to 8k Rx descriptors per ring.
88 * Added additional doorbell check on Tx, to handle Tx more efficiently for big
90 * Added per queue statistics.
91 * Added extended statistics using xstats DPDK API.
92 * The reset routine was aligned with the DPDK API, so now it can be
93 handled as in other PMDs.
94 * Fixed out of order (OOO) completion.
95 * Fixed memory leaks due to port stops and starts in the middle of
97 * Updated documentation and features list of the PMD.
99 * **Updated mlx5 driver.**
101 Updated the mlx5 driver including the following changes:
103 * Fixed ``imissed`` counter to be reported through ``rte_eth_stats`` instead
104 of ``rte_eth_xstats``.
105 * Added packet header modification through Direct Verbs flow driver.
106 * Added ConnectX-6 PCI device ID to be proved by ``mlx5`` driver.
107 * Added flow counter support to Direct Verbs flow driver though DevX.
108 * Renamed build options for the glue layer to
109 ``CONFIG_RTE_IBVERBS_LINK_DLOPEN`` for make and ``ibverbs_link`` for meson.
110 * Added static linkage of ``mlx`` dependency.
111 * Improved stability of E-Switch flow driver.
112 * Added new make build configuration to set the cacheline size for BlueField
113 correctly - ``arm64-bluefield-linux-gcc``.
115 * **Updated the enic driver.**
117 * Added support for the ``RTE_ETH_DEV_CLOSE_REMOVE`` flag.
118 * Added a handler to get the firmware version string.
119 * Added support for multicast filtering.
121 * **Added dynamic queues allocation support for i40e VF.**
123 Previously, the available VF queues were reserved by PF at initialization
124 stage. Now both DPDK PF and Kernel PF (>=2.1.14) will support dynamic queue
125 allocation. At runtime, when VF requests for more queue exceed the initial
126 reserved amount, the PF can allocate up to 16 queues as the request after a
129 * **Added ICE net PMD.**
131 Added the new ``ice`` net driver for Intel(R) Ethernet Network Adapters E810.
132 See the :doc:`../nics/ice` NIC guide for more details on this new driver.
134 * **Added support for SW-assisted VDPA live migration.**
136 This SW-assisted VDPA live migration facility helps VDPA devices without
137 logging capability to perform live migration, a mediated SW relay can help
138 devices to track dirty pages caused by DMA. the IFC driver has enabled this
139 SW-assisted live migration mode.
141 * **Added security checks to the cryptodev symmetric session operations.**
143 Added a set of security checks to the access cryptodev symmetric session.
144 The checks include the session's user data read/write check and the
145 session private data referencing status check while freeing a session.
147 * **Updated the AESNI-MB PMD.**
149 * Added support for intel-ipsec-mb version 0.52.
150 * Added AES-GMAC algorithm support.
151 * Added Plain SHA1, SHA224, SHA256, SHA384, and SHA512 algorithms support.
153 * **Added IPsec Library.**
155 Added an experimental library ``librte_ipsec`` to provide ESP tunnel and
156 transport support for IPv4 and IPv6 packets.
158 The library provides support for AES-CBC ciphering and AES-CBC with HMAC-SHA1
159 algorithm-chaining, and AES-GCM and NULL algorithms only at present. It is
160 planned to add more algorithms in future releases.
162 See :doc:`../prog_guide/ipsec_lib` for more information.
164 * **Updated the ipsec-secgw sample application.**
166 The ``ipsec-secgw`` sample application has been updated to use the new
167 ``librte_ipsec`` library, which has also been added in this release.
168 The original functionality of ipsec-secgw is retained, a new command line
169 parameter ``-l`` has been added to ipsec-secgw to use the IPsec library,
170 instead of the existing IPsec code in the application.
172 The IPsec library does not support all the functionality of the existing
173 ipsec-secgw application. It is planned to add the outstanding functionality
176 See :doc:`../sample_app_ug/ipsec_secgw` for more information.
178 * **Enabled checksum support in the ISA-L compressdev driver.**
180 Added support for both adler and crc32 checksums in the ISA-L PMD.
181 This aids data integrity across both compression and decompression.
183 * **Added a compression performance test tool.**
185 Added a new performance test tool to test the compressdev PMD. The tool tests
186 compression ratio and compression throughput.
188 * **Added intel_pstate support to Power Management library.**
190 Previously, using the power management library required the
191 disabling of the intel_pstate kernel driver, and the enabling of the
192 acpi_cpufreq kernel driver. This is no longer the case, as the use of
193 the intel_pstate kernel driver is now supported, and automatically
194 detected by the library.
200 .. This section should contain API changes. Sample format:
202 * sample: Add a short 1-2 sentence description of the API change
203 which was announced in the previous releases and made in this release.
204 Start with a scope label like "ethdev:".
205 Use fixed width quotes for ``function_names`` or ``struct_names``.
208 This section is a comment. Do not overwrite or remove it.
209 Also, make sure to start the actual text at the margin.
210 =========================================================
212 * eal: Function ``rte_bsf64`` in ``rte_bitmap.h`` has been renamed to
213 ``rte_bsf64_safe`` and moved to ``rte_common.h``. A new ``rte_bsf64``
214 function has been added in ``rte_common.h`` that follows the convention set
215 by the existing ``rte_bsf32`` function.
217 * eal: Segment fd API on Linux now sets error code to ``ENOTSUP`` in more cases
218 where segment the fd API is not expected to be supported:
220 - On attempt to get a segment fd for an externally allocated memory segment
221 - In cases where memfd support would have been required to provide segment
222 fds (such as in-memory or no-huge mode)
224 * eal: Functions ``rte_malloc_dump_stats()``, ``rte_malloc_dump_heaps()`` and
225 ``rte_malloc_get_socket_stats()`` are no longer safe to call concurrently with
226 ``rte_malloc_heap_create()`` or ``rte_malloc_heap_destroy()`` function calls.
228 * mbuf: ``RTE_MBUF_INDIRECT()``, which was deprecated in 18.05, was replaced
229 with ``RTE_MBUF_CLONED()`` and removed in 19.02.
231 * sched: As result of the new format of the mbuf sched field, the
232 functions ``rte_sched_port_pkt_write()`` and
233 ``rte_sched_port_pkt_read_tree_path()`` got an additional parameter of
234 type ``struct rte_sched_port``.
236 * pdump: The ``rte_pdump_set_socket_dir()``, the parameter ``path`` of
237 ``rte_pdump_init()`` and enum ``rte_pdump_socktype`` were deprecated
238 since 18.05 and are removed in this release.
240 * cryptodev: The parameter ``session_pool`` in the function
241 ``rte_cryptodev_queue_pair_setup()`` is removed.
243 * cryptodev: a new function ``rte_cryptodev_sym_session_pool_create()`` has been
244 introduced. This function is now mandatory when creating symmetric session
245 header mempool. Please note all crypto applications are required to use this
246 function from now on. Failed to do so will cause a
247 ``rte_cryptodev_sym_session_create()`` function call return error.
253 .. This section should contain ABI changes. Sample format:
255 * sample: Add a short 1-2 sentence description of the ABI change
256 which was announced in the previous releases and made in this release.
257 Start with a scope label like "ethdev:".
258 Use fixed width quotes for ``function_names`` or ``struct_names``.
261 This section is a comment. Do not overwrite or remove it.
262 Also, make sure to start the actual text at the margin.
263 =========================================================
265 * mbuf: The format of the sched field of ``rte_mbuf`` has been changed
266 to include the following fields: ``queue ID``, ``traffic class``, ``color``.
268 * cryptodev: as shown in the 18.11 deprecation notice, the structure
269 ``rte_cryptodev_qp_conf`` has added two parameters for symmetric session
270 mempool and symmetric session private data mempool.
272 * cryptodev: as shown in the 18.11 deprecation notice, the structure
273 ``rte_cryptodev_sym_session`` has been updated to contain more information
274 to ensure safely accessing the session and session private data.
276 * security: A new field ``uint64_t opaque_data`` has been added to
277 ``rte_security_session`` structure. That would allow upper layer to easily
278 associate/de-associate some user defined data with the security session.
281 Shared Library Versions
282 -----------------------
284 .. Update any library version updated in this release
285 and prepend with a ``+`` sign, like this:
291 This section is a comment. Do not overwrite or remove it.
292 =========================================================
294 The libraries prepended with a plus sign were incremented in this version.
300 librte_bitratestats.so.2
303 librte_bus_fslmc.so.2
304 librte_bus_ifpga.so.2
307 librte_bus_vmbus.so.2
310 librte_compressdev.so.1
311 + librte_cryptodev.so.6
312 librte_distributor.so.1
317 librte_flow_classify.so.1
325 librte_latencystats.so.1
339 librte_pmd_ixgbe.so.2
340 librte_pmd_dpaa2_qdma.so.1
342 librte_pmd_softnic.so.1
343 librte_pmd_vhost.so.2
350 + librte_security.so.2
359 .. This section should contain new known issues in this release. Sample format:
361 * **Add title in present tense with full stop.**
363 Add a short 1-2 sentence description of the known issue
364 in the present tense. Add information on any known workarounds.
366 This section is a comment. Do not overwrite or remove it.
367 Also, make sure to start the actual text at the margin.
368 =========================================================
370 * ``AVX-512`` support has been disabled for ``GCC`` builds when ``binutils 2.30``
371 is detected [1] because of a crash [2]. This can affect ``native`` machine type
372 build targets on the platforms that support ``AVX512F`` like ``Intel Skylake``
373 processors, and can cause a possible performance drop. The immediate workaround
374 is to use ``clang`` compiler on these platforms.
375 Initial workaround in DPDK v18.11 was to disable ``AVX-512`` support for ``GCC``
376 completely, but based on information on defect submitted to GCC community [3],
377 issue has been identified as ``binutils 2.30`` issue. Since currently only GCC
378 generates ``AVX-512`` instructions, the scope is limited to ``GCC`` and
381 - [1]: Commit ("mk: fix scope of disabling AVX512F support")
382 - [2]: https://bugs.dpdk.org/show_bug.cgi?id=97
383 - [3]: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=88096
389 .. This section should contain a list of platforms that were tested
394 * <vendor> platform with <vendor> <type of devices> combinations
399 * Other relevant details...
401 This section is a comment. Do not overwrite or remove it.
402 Also, make sure to start the actual text at the margin.
403 =========================================================
405 * Intel(R) platforms with Intel(R) NICs combinations
409 * Intel(R) Atom(TM) CPU C3758 @ 2.20GHz
410 * Intel(R) Xeon(R) CPU D-1541 @ 2.10GHz
411 * Intel(R) Xeon(R) CPU E5-2680 v2 @ 2.80GHz
412 * Intel(R) Xeon(R) CPU E5-2699 v3 @ 2.30GHz
413 * Intel(R) Xeon(R) CPU E5-2699 v4 @ 2.20GHz
414 * Intel(R) Xeon(R) Platinum 8180 CPU @ 2.50GHz
415 * Intel(R) Xeon(R) Gold 6139 CPU @ 2.30GHz
425 * Red Hat Enterprise Linux Server release 7.4
426 * Red Hat Enterprise Linux Server release 7.5
437 * Intel(R) 82599ES 10 Gigabit Ethernet Controller
439 * Firmware version: 0x61bf0001
440 * Device id (pf/vf): 8086:10fb / 8086:10ed
441 * Driver version: 5.2.3 (ixgbe)
443 * Intel(R) Corporation Ethernet Connection X552/X557-AT 10GBASE-T
445 * Firmware version: 0x800003e7
446 * Device id (pf/vf): 8086:15ad / 8086:15a8
447 * Driver version: 4.4.6 (ixgbe)
449 * Intel(R) Ethernet Converged Network Adapter X710-DA4 (4x10G)
451 * Firmware version: 6.80 0x80003cc1
452 * Device id (pf/vf): 8086:1572 / 8086:154c
453 * Driver version: 2.7.26 (i40e)
455 * Intel(R) Corporation Ethernet Connection X722 for 10GbE SFP+ (4x10G)
457 * Firmware version: 3.33 0x80000fd5 0.0.0
458 * Device id (pf/vf): 8086:37d0 / 8086:37cd
459 * Driver version: 2.7.26 (i40e)
461 * Intel(R) Ethernet Converged Network Adapter XXV710-DA2 (2x25G)
463 * Firmware version: 6.80 0x80003d05
464 * Device id (pf/vf): 8086:158b / 8086:154c
465 * Driver version: 2.7.26 (i40e)
467 * Intel(R) Ethernet Converged Network Adapter XL710-QDA2 (2X40G)
469 * Firmware version: 6.80 0x80003cfb
470 * Device id (pf/vf): 8086:1583 / 8086:154c
471 * Driver version: 2.7.26 (i40e)
473 * Intel(R) Corporation I350 Gigabit Network Connection
475 * Firmware version: 1.63, 0x80000dda
476 * Device id (pf/vf): 8086:1521 / 8086:1520
477 * Driver version: 5.4.0-k (igb)
479 * Intel(R) platforms with Mellanox(R) NICs combinations
483 * Intel(R) Xeon(R) Gold 6154 CPU @ 3.00GHz
484 * Intel(R) Xeon(R) CPU E5-2697A v4 @ 2.60GHz
485 * Intel(R) Xeon(R) CPU E5-2697 v3 @ 2.60GHz
486 * Intel(R) Xeon(R) CPU E5-2680 v2 @ 2.80GHz
487 * Intel(R) Xeon(R) CPU E5-2650 v4 @ 2.20GHz
488 * Intel(R) Xeon(R) CPU E5-2640 @ 2.50GHz
489 * Intel(R) Xeon(R) CPU E5-2620 v4 @ 2.10GHz
493 * Red Hat Enterprise Linux Server release 7.6 (Maipo)
494 * Red Hat Enterprise Linux Server release 7.5 (Maipo)
495 * Red Hat Enterprise Linux Server release 7.4 (Maipo)
496 * Red Hat Enterprise Linux Server release 7.3 (Maipo)
497 * Red Hat Enterprise Linux Server release 7.2 (Maipo)
502 * SUSE Linux Enterprise Server 15
504 * MLNX_OFED: 4.4-2.0.1.0
505 * MLNX_OFED: 4.5-1.0.1.0
509 * Mellanox(R) ConnectX(R)-3 Pro 40G MCX354A-FCC_Ax (2x40G)
511 * Host interface: PCI Express 3.0 x8
512 * Device ID: 15b3:1007
513 * Firmware version: 2.42.5000
515 * Mellanox(R) ConnectX(R)-4 10G MCX4111A-XCAT (1x10G)
517 * Host interface: PCI Express 3.0 x8
518 * Device ID: 15b3:1013
519 * Firmware version: 12.24.1000 and above
521 * Mellanox(R) ConnectX(R)-4 10G MCX4121A-XCAT (2x10G)
523 * Host interface: PCI Express 3.0 x8
524 * Device ID: 15b3:1013
525 * Firmware version: 12.24.1000 and above
527 * Mellanox(R) ConnectX(R)-4 25G MCX4111A-ACAT (1x25G)
529 * Host interface: PCI Express 3.0 x8
530 * Device ID: 15b3:1013
531 * Firmware version: 12.24.1000 and above
533 * Mellanox(R) ConnectX(R)-4 25G MCX4121A-ACAT (2x25G)
535 * Host interface: PCI Express 3.0 x8
536 * Device ID: 15b3:1013
537 * Firmware version: 12.24.1000 and above
539 * Mellanox(R) ConnectX(R)-4 40G MCX4131A-BCAT/MCX413A-BCAT (1x40G)
541 * Host interface: PCI Express 3.0 x8
542 * Device ID: 15b3:1013
543 * Firmware version: 12.24.1000 and above
545 * Mellanox(R) ConnectX(R)-4 40G MCX415A-BCAT (1x40G)
547 * Host interface: PCI Express 3.0 x16
548 * Device ID: 15b3:1013
549 * Firmware version: 12.24.1000 and above
551 * Mellanox(R) ConnectX(R)-4 50G MCX4131A-GCAT/MCX413A-GCAT (1x50G)
553 * Host interface: PCI Express 3.0 x8
554 * Device ID: 15b3:1013
555 * Firmware version: 12.24.1000 and above
557 * Mellanox(R) ConnectX(R)-4 50G MCX414A-BCAT (2x50G)
559 * Host interface: PCI Express 3.0 x8
560 * Device ID: 15b3:1013
561 * Firmware version: 12.24.1000 and above
563 * Mellanox(R) ConnectX(R)-4 50G MCX415A-GCAT/MCX416A-BCAT/MCX416A-GCAT (2x50G)
565 * Host interface: PCI Express 3.0 x16
566 * Device ID: 15b3:1013
567 * Firmware version: 12.24.1000 and above
568 * Firmware version: 12.24.1000 and above
570 * Mellanox(R) ConnectX(R)-4 50G MCX415A-CCAT (1x100G)
572 * Host interface: PCI Express 3.0 x16
573 * Device ID: 15b3:1013
574 * Firmware version: 12.24.1000 and above
576 * Mellanox(R) ConnectX(R)-4 100G MCX416A-CCAT (2x100G)
578 * Host interface: PCI Express 3.0 x16
579 * Device ID: 15b3:1013
580 * Firmware version: 12.24.1000 and above
582 * Mellanox(R) ConnectX(R)-4 Lx 10G MCX4121A-XCAT (2x10G)
584 * Host interface: PCI Express 3.0 x8
585 * Device ID: 15b3:1015
586 * Firmware version: 14.24.1000 and above
588 * Mellanox(R) ConnectX(R)-4 Lx 25G MCX4121A-ACAT (2x25G)
590 * Host interface: PCI Express 3.0 x8
591 * Device ID: 15b3:1015
592 * Firmware version: 14.24.1000 and above
594 * Mellanox(R) ConnectX(R)-5 100G MCX556A-ECAT (2x100G)
596 * Host interface: PCI Express 3.0 x16
597 * Device ID: 15b3:1017
598 * Firmware version: 16.24.1000 and above
600 * Mellanox(R) ConnectX(R)-5 Ex EN 100G MCX516A-CDAT (2x100G)
602 * Host interface: PCI Express 4.0 x16
603 * Device ID: 15b3:1019
604 * Firmware version: 16.24.1000 and above
606 * ARM platforms with Mellanox(R) NICs combinations
610 * Qualcomm ARM 1.1 2500MHz
614 * Red Hat Enterprise Linux Server release 7.5 (Maipo)
618 * Mellanox(R) ConnectX(R)-4 Lx 25G MCX4121A-ACAT (2x25G)
620 * Host interface: PCI Express 3.0 x8
621 * Device ID: 15b3:1015
622 * Firmware version: 14.24.0220
624 * Mellanox(R) ConnectX(R)-5 100G MCX556A-ECAT (2x100G)
626 * Host interface: PCI Express 3.0 x16
627 * Device ID: 15b3:1017
628 * Firmware version: 16.24.0220
630 * Mellanox(R) BlueField SmartNIC
632 * Mellanox(R) BlueField SmartNIC MT416842 (2x25G)
634 * Host interface: PCI Express 3.0 x16
635 * Device ID: 15b3:a2d2
636 * Firmware version: 18.24.0328
638 * SoC ARM cores running OS:
640 * CentOS Linux release 7.4.1708 (AltArch)
641 * MLNX_OFED 4.4-2.5.9.0
643 * DPDK application running on ARM cores inside SmartNIC
645 * Power 9 platforms with Mellanox(R) NICs combinations
649 * POWER9 2.2 (pvr 004e 1202) 2300MHz
653 * Ubuntu 18.04.1 LTS (Bionic Beaver)
657 * Mellanox(R) ConnectX(R)-5 100G MCX556A-ECAT (2x100G)
659 * Host interface: PCI Express 3.0 x16
660 * Device ID: 15b3:1017
661 * Firmware version: 16.23.1020
665 * MLNX_OFED_LINUX-4.5-1.0.1.0