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 to free 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 * **Support for using virtio-user without hugepages**
74 The --no-huge mode was augmented to use memfd-backed memory (on systems that
75 support memfd), to allow using virtio-user-based NICs without hugepages.
77 * **Release of the ENA PMD v2.0.0**
79 * Added Low Latency Queue v2 (LLQv2). This feature reduces the latency
80 of the packets by pushing the header directly through the PCI to the
81 device. This allows the NIC to start handle packet right after the doorbell
82 without waiting for DMA.
83 * Added independent configuration of HW Tx and Rx ring depths.
84 * Added support for up to 8k Rx descriptors per ring.
85 * Added additional doorbell check on Tx, to handle Tx more efficiently for big
87 * Added per queue statistics.
88 * Added extended statistics using xstats DPDK API.
89 * The reset routine was aligned with the DPDK API, so now it can be
90 handled as in other PMDs.
91 * Fixed out of order (OOO) completion.
92 * Fixed memory leaks due to port stops and starts in the middle of
94 * Updated documentation and features list of the PMD.
96 * **Updated the enic driver.**
98 * Added support for ``RTE_ETH_DEV_CLOSE_REMOVE`` flag.
99 * Added the handler to get firmware version string.
100 * Added support for multicast filtering.
102 * **Added dynamic queues allocation support for i40e VF.**
104 Previously, available queues of VF is reserved by PF at initialize stage.
105 Now both DPDK PF and Kernel PF (>=2.1.14) will support dynamic queue
106 allocation. At runtime, when VF request more queue number exceed the initial
107 reserved amount, PF can allocate up to 16 queues as the request after a VF
110 * **Added ICE net PMD**
112 Added the new ``ice`` net driver for IntelĀ® Ethernet Network Adapters E810.
113 See the :doc:`../nics/ice` NIC guide for more details on this new driver.
115 * **Added support for SW-assisted VDPA live migration.**
117 This SW-assisted VDPA live migration facility helps VDPA devices without
118 logging capability to perform live migration, a mediated SW relay can help
119 devices to track dirty pages caused by DMA. IFC driver has enabled this
120 SW-assisted live migration mode.
122 * **Added security checks to cryptodev symmetric session operations.**
124 Added a set of security checks to the access cryptodev symmetric session.
125 The checks include the session's user data read/write check and the
126 session private data referencing status check while freeing a session.
128 * **Updated the AESNI-MB PMD.**
130 * Add support for intel-ipsec-mb version 0.52.
131 * Add AES-GMAC algorithm support.
132 * Add Plain SHA1, SHA224, SHA256, SHA384, and SHA512 algorithms support.
134 * **Added IPsec Library.**
136 Added an experimental library ``librte_ipsec`` to provide ESP tunnel and
137 transport support for IPv4 and IPv6 packets.
139 The library provides support for AES-CBC ciphering and AES-CBC with HMAC-SHA1
140 algorithm-chaining, and AES-GCM and NULL algorithms only at present. It is
141 planned to add more algorithms in future releases.
143 See :doc:`../prog_guide/ipsec_lib` for more information.
145 * **Updated the ipsec-secgw sample application.**
147 The ``ipsec-secgw`` sample application has been updated to use the new
148 ``librte_ipsec`` library also added in this release.
149 The original functionality of ipsec-secgw is retained, a new command line
150 parameter ``-l`` has been added to ipsec-secgw to use the IPsec library,
151 instead of the existing IPsec code in the application.
153 The IPsec library does not support all the functionality of the existing
154 ipsec-secgw application, its is planned to add the outstanding functionality
157 See :doc:`../sample_app_ug/ipsec_secgw` for more information.
159 * **Enabled checksum support in the ISA-L compressdev driver.**
161 Added support for both adler and crc32 checksums in the ISA-L PMD.
162 This aids data integrity across both compression and decompression.
164 * **Added a compression performance test tool.**
166 Added a new performance test tool to test the compressdev PMD. The tool tests
167 compression ratio and compression throughput.
169 * **Added intel_pstate support to Power Management library.**
171 Previously, using the power management library required the
172 disabling of the intel_pstate kernel driver, and the enabling of the
173 acpi_cpufreq kernel driver. This is no longer the case, as the use of
174 the intel_pstate kernel driver is now supported, and automatically
175 detected by the library.
181 .. This section should contain removed items in this release. Sample format:
183 * Add a short 1-2 sentence description of the removed item
186 This section is a comment. Do not overwrite or remove it.
187 Also, make sure to start the actual text at the margin.
188 =========================================================
194 .. This section should contain API changes. Sample format:
196 * sample: Add a short 1-2 sentence description of the API change
197 which was announced in the previous releases and made in this release.
198 Start with a scope label like "ethdev:".
199 Use fixed width quotes for ``function_names`` or ``struct_names``.
202 This section is a comment. Do not overwrite or remove it.
203 Also, make sure to start the actual text at the margin.
204 =========================================================
206 * eal: Function ``rte_bsf64`` in ``rte_bitmap.h`` has been renamed to
207 ``rte_bsf64_safe`` and moved to ``rte_common.h``. A new ``rte_bsf64`` function
208 has been added in ``rte_common.h`` that follows convention set by existing
209 ``rte_bsf32`` function.
211 * eal: Segment fd API on Linux now sets error code to ``ENOTSUP`` in more cases
212 where segment fd API is not expected to be supported:
214 - On attempt to get segment fd for an externally allocated memory segment
215 - In cases where memfd support would have been required to provide segment
216 fd's (such as in-memory or no-huge mode)
218 * eal: Functions ``rte_malloc_dump_stats()``, ``rte_malloc_dump_heaps()`` and
219 ``rte_malloc_get_socket_stats()`` are no longer safe to call concurrently with
220 ``rte_malloc_heap_create()`` or ``rte_malloc_heap_destroy()`` function calls.
222 * mbuf: ``RTE_MBUF_INDIRECT()``, which was deprecated in 18.05, was replaced
223 with ``RTE_MBUF_CLONED()`` and removed in 19.02.
225 * sched: As result of the new format of the mbuf sched field, the
226 functions ``rte_sched_port_pkt_write()`` and
227 ``rte_sched_port_pkt_read_tree_path()`` got an additional parameter of
228 type ``struct rte_sched_port``.
230 * pdump: The ``rte_pdump_set_socket_dir()``, the parameter ``path`` of
231 ``rte_pdump_init()`` and enum ``rte_pdump_socktype`` were deprecated
232 since 18.05 and are removed in this release.
234 * cryptodev: The parameter ``session_pool`` in the function
235 ``rte_cryptodev_queue_pair_setup()`` is removed.
237 * cryptodev: a new function ``rte_cryptodev_sym_session_pool_create()`` is
238 introduced. This function is now mandatory when creating symmetric session
239 header mempool. Please note all crypto applications are required to use this
240 function from now on. Failed to do so will cause
241 ``rte_cryptodev_sym_session_create()`` function call return error.
247 .. This section should contain ABI changes. Sample format:
249 * sample: Add a short 1-2 sentence description of the ABI change
250 which was announced in the previous releases and made in this release.
251 Start with a scope label like "ethdev:".
252 Use fixed width quotes for ``function_names`` or ``struct_names``.
255 This section is a comment. Do not overwrite or remove it.
256 Also, make sure to start the actual text at the margin.
257 =========================================================
259 * mbuf: The format of the sched field of ``rte_mbuf`` has been changed
260 to include the following fields: ``queue ID``, ``traffic class``, ``color``.
262 * cryptodev: as shown in the the 18.11 deprecation notice, the structure
263 ``rte_cryptodev_qp_conf`` has been added two parameters of symmetric session
264 mempool and symmetric session private data mempool.
266 * cryptodev: as shown in the the 18.11 deprecation notice, the structure
267 ``rte_cryptodev_sym_session`` has been updated to contain more information
268 to ensure safely accessing the session and session private data.
270 * security: New field ``uint64_t opaque_data`` is added into
271 ``rte_security_session`` structure. That would allow upper layer to easily
272 associate/de-associate some user defined data with the security session.
275 Shared Library Versions
276 -----------------------
278 .. Update any library version updated in this release
279 and prepend with a ``+`` sign, like this:
285 This section is a comment. Do not overwrite or remove it.
286 =========================================================
288 The libraries prepended with a plus sign were incremented in this version.
294 librte_bitratestats.so.2
297 librte_bus_fslmc.so.2
298 librte_bus_ifpga.so.2
301 librte_bus_vmbus.so.2
304 librte_compressdev.so.1
305 + librte_cryptodev.so.6
306 librte_distributor.so.1
311 librte_flow_classify.so.1
319 librte_latencystats.so.1
333 librte_pmd_ixgbe.so.2
334 librte_pmd_dpaa2_qdma.so.1
336 librte_pmd_softnic.so.1
337 librte_pmd_vhost.so.2
344 + librte_security.so.2
353 .. This section should contain new known issues in this release. Sample format:
355 * **Add title in present tense with full stop.**
357 Add a short 1-2 sentence description of the known issue
358 in the present tense. Add information on any known workarounds.
360 This section is a comment. Do not overwrite or remove it.
361 Also, make sure to start the actual text at the margin.
362 =========================================================
364 * ``AVX-512`` support has been disabled for ``GCC`` builds when ``binutils 2.30``
365 is detected [1] because of a crash [2]. This can affect ``native`` machine type
366 build targets on the platforms that support ``AVX512F`` like ``Intel Skylake``
367 processors, and can cause a possible performance drop. The immediate workaround
368 is to use ``clang`` compiler on these platforms.
369 Initial workaround in DPDK v18.11 was to disable ``AVX-512`` support for ``GCC``
370 completely, but based on information on defect submitted to GCC community [3],
371 issue has been identified as ``binutils 2.30`` issue. Since currently only GCC
372 generates ``AVX-512`` instructions, the scope is limited to ``GCC`` and
375 - [1]: Commit ("mk: fix scope of disabling AVX512F support")
376 - [2]: https://bugs.dpdk.org/show_bug.cgi?id=97
377 - [3]: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=88096
383 .. This section should contain a list of platforms that were tested
388 * <vendor> platform with <vendor> <type of devices> combinations
393 * Other relevant details...
395 This section is a comment. Do not overwrite or remove it.
396 Also, make sure to start the actual text at the margin.
397 =========================================================
399 * Intel(R) platforms with Intel(R) NICs combinations
403 * Intel(R) Atom(TM) CPU C3758 @ 2.20GHz
404 * Intel(R) Xeon(R) CPU D-1541 @ 2.10GHz
405 * Intel(R) Xeon(R) CPU E5-2680 v2 @ 2.80GHz
406 * Intel(R) Xeon(R) CPU E5-2699 v3 @ 2.30GHz
407 * Intel(R) Xeon(R) CPU E5-2699 v4 @ 2.20GHz
408 * Intel(R) Xeon(R) Platinum 8180 CPU @ 2.50GHz
409 * Intel(R) Xeon(R) Gold 6139 CPU @ 2.30GHz
419 * Red Hat Enterprise Linux Server release 7.4
420 * Red Hat Enterprise Linux Server release 7.5
431 * Intel(R) 82599ES 10 Gigabit Ethernet Controller
433 * Firmware version: 0x61bf0001
434 * Device id (pf/vf): 8086:10fb / 8086:10ed
435 * Driver version: 5.2.3 (ixgbe)
437 * Intel(R) Corporation Ethernet Connection X552/X557-AT 10GBASE-T
439 * Firmware version: 0x800003e7
440 * Device id (pf/vf): 8086:15ad / 8086:15a8
441 * Driver version: 4.4.6 (ixgbe)
443 * Intel(R) Ethernet Converged Network Adapter X710-DA4 (4x10G)
445 * Firmware version: 6.80 0x80003cc1
446 * Device id (pf/vf): 8086:1572 / 8086:154c
447 * Driver version: 2.7.26 (i40e)
449 * Intel(R) Corporation Ethernet Connection X722 for 10GbE SFP+ (4x10G)
451 * Firmware version: 3.33 0x80000fd5 0.0.0
452 * Device id (pf/vf): 8086:37d0 / 8086:37cd
453 * Driver version: 2.7.26 (i40e)
455 * Intel(R) Ethernet Converged Network Adapter XXV710-DA2 (2x25G)
457 * Firmware version: 6.80 0x80003d05
458 * Device id (pf/vf): 8086:158b / 8086:154c
459 * Driver version: 2.7.26 (i40e)
461 * Intel(R) Ethernet Converged Network Adapter XL710-QDA2 (2X40G)
463 * Firmware version: 6.80 0x80003cfb
464 * Device id (pf/vf): 8086:1583 / 8086:154c
465 * Driver version: 2.7.26 (i40e)
467 * Intel(R) Corporation I350 Gigabit Network Connection
469 * Firmware version: 1.63, 0x80000dda
470 * Device id (pf/vf): 8086:1521 / 8086:1520
471 * Driver version: 5.4.0-k (igb)