4 .. **Read this first.**
6 The text in the sections below explains how to update the release notes.
8 Use proper spelling, capitalization and punctuation in all sections.
10 Variable and config names should be quoted as fixed width text:
13 Build the docs and view the output file to ensure the changes are correct::
17 xdg-open build/doc/html/guides/rel_notes/release_18_05.html
23 .. This section should contain new features added in this release. Sample
26 * **Add a title in the past tense with a full stop.**
28 Add a short 1-2 sentence description in the past tense. The description
29 should be enough to allow someone scanning the release notes to
30 understand the new feature.
32 If the feature adds a lot of sub-features you can use a bullet list like
35 * Added feature foo to do something.
36 * Enhanced feature bar to do something else.
38 Refer to the previous release notes for examples.
40 This section is a comment. Do not overwrite or remove it.
41 Also, make sure to start the actual text at the margin.
42 =========================================================
44 * **Added bucket mempool driver.**
46 Added bucket mempool driver which provides a way to allocate contiguous
48 Number of objects in the block depends on how many objects fit in
49 RTE_DRIVER_MEMPOOL_BUCKET_SIZE_KB memory chunk which is build time option.
50 The number may be obtained using rte_mempool_ops_get_info() API.
51 Contiguous blocks may be allocated using rte_mempool_get_contig_blocks() API.
53 * **Added PMD-recommended Tx and Rx parameters**
55 Applications can now query drivers for device-tuned values of
56 ring sizes, burst sizes, and number of queues.
58 * **Added RSS hash and key update to CXGBE PMD.**
60 Support to update RSS hash and key has been added to CXGBE PMD.
62 * **Added CXGBE VF PMD.**
64 CXGBE VF Poll Mode Driver has been added to run DPDK over Chelsio
65 T5/T6 NIC VF instances.
67 * **Updated mlx5 driver.**
69 Updated the mlx5 driver including the following changes:
71 * Introduced Multi-packet Rx. With it, achieved 100Gb/sec with 64B frames.
72 * Supported to be run by non-root users given reduced set of capabilities
73 CAP_NET_ADMIN, CAP_NET_RAW and CAP_IPC_LOCK.
74 * Supported TSO and checksum for generic UDP and IP tunnels.
75 * Supported inner checksum and RSS for GRE, VXLAN-GPE, MPLSoGRE
77 * Accommodate to the new memory hotplug model.
78 * Supported non virtually contiguous mempools.
79 * Supported MAC adding along with allmulti and promiscuous modes from VF.
80 * Supported Mellanox BlueField SoC device.
81 * Supported PMD defaults for queue number and depth to improve the out
82 of the box performance.
84 * **Updated mlx4 driver.**
86 Updated the mlx4 driver including the following changes:
88 * Supported to be run by non-root users given reduced set of capabilities
89 CAP_NET_ADMIN, CAP_NET_RAW and CAP_IPC_LOCK.
90 * Supported CRC strip toggling.
91 * Accommodate to the new memory hotplug model.
92 * Supported non virtually contiguous mempools.
94 * **Updated Solarflare network PMD.**
96 Updated the sfc_efx driver including the following changes:
98 * Added support for Solarflare XtremeScale X2xxx family adapters.
99 * Added support for NVGRE, VXLAN and GENEVE filters in flow API.
100 * Added support for DROP action in flow API.
101 * Added support for equal stride super-buffer Rx mode (X2xxx only).
102 * Added support for MARK and FLAG actions in flow API (X2xxx only).
104 * **Added Ethernet poll mode driver for AMD XGBE devices.**
106 Added the new ``axgbe`` ethernet poll mode driver for AMD XGBE devices.
107 See the :doc:`../nics/axgbe` nic driver guide for more details on this
110 * **Updated szedata2 PMD.**
112 Added support for new NFB-200G2QL card.
113 New API was introduced in the libsze2 library which the szedata2 PMD depends
114 on thus the new version of the library was needed.
115 New versions of the packages are available and the minimum required version
118 * **Added support for Broadcom NetXtreme-S (BCM58800) family of controllers (aka Stingray)**
120 The BCM58800 devices feature a NetXtreme E-Series advanced network controller, a high-performance
121 ARM CPU block, PCI Express (PCIe) Gen3 interfaces, key accelerators for compute offload and a high-
122 speed memory subsystem including L3 cache and DDR4 interfaces, all interconnected by a coherent
123 Network-on-chip (NOC) fabric.
125 The ARM CPU subsystem features eight ARMv8 Cortex-A72 CPUs at 3.0 GHz, arranged in a multi-cluster
128 * **Added IFCVF vDPA driver.**
130 Added IFCVF vDPA driver to support Intel FPGA 100G VF device. IFCVF works
131 as a HW vhost data path accelerator, it supports live migration and is
132 compatible with virtio 0.95 and 1.0. This driver registers ifcvf vDPA driver
133 to vhost lib, when virtio connected, with the help of the registered vDPA
134 driver the assigned VF gets configured to Rx/Tx directly to VM's virtio
137 * **Added support for virtio-user server mode.**
138 In a container environment if the vhost-user backend restarts, there's no way
139 for it to reconnect to virtio-user. To address this, support for server mode
140 is added. In this mode the socket file is created by virtio-user, which the
141 backend connects to. This means that if the backend restarts, it can reconnect
142 to virtio-user and continue communications.
144 * **Added crypto workload support to vhost library.**
146 New APIs are introduced in vhost library to enable virtio crypto support
147 including session creation/deletion handling and translating virtio-crypto
148 request into DPDK crypto operations. A sample application is also introduced.
150 * **Added virtio crypto PMD.**
152 Added a new poll mode driver for virtio crypto devices, which provides
153 AES-CBC ciphering and AES-CBC with HMAC-SHA1 algorithm-chaining. See the
154 :doc:`../cryptodevs/virtio` crypto driver guide for more details on
157 * **Added AMD CCP Crypto PMD.**
159 Added the new ``ccp`` crypto driver for AMD CCP devices. See the
160 :doc:`../cryptodevs/ccp` crypto driver guide for more details on
163 * **Updated AESNI MB PMD.**
165 The AESNI MB PMD has been updated with additional support for:
167 * AES-CMAC (128-bit key).
169 * **Added Compressdev Library, a generic compression service library.**
171 The compressdev library provides an API for offload of compression and
172 decompression operations to hardware or software accelerator devices.
174 * **Added a new compression poll mode driver using Intels ISA-L.**
176 Added the new ``ISA-L`` compression driver, for compression and decompression
177 operations in software. See the :doc:`../compressdevs/isal` compression driver
178 guide for details on this new driver.
180 * **Added the Event Timer Adapter Library.**
182 The Event Timer Adapter Library extends the event-based model by introducing
183 APIs that allow applications to arm/cancel event timers that generate
184 timer expiry events. This new type of event is scheduled by an event device
185 along with existing types of events.
187 * **Added OcteonTx TIM Driver (Event timer adapter).**
189 The OcteonTx Timer block enables software to schedule events for a future
190 time, it is exposed to an application via Event timer adapter library.
192 See the :doc:`../eventdevs/octeontx` guide for more details
194 * **Added Event Crypto Adapter Library.**
196 Added the Event Crypto Adapter Library. This library extends the
197 event-based model by introducing APIs that allow applications to
198 enqueue/dequeue crypto operations to/from cryptodev as events scheduled
201 * **Added Ifpga Bus, a generic Intel FPGA Bus library.**
203 The Ifpga Bus library provides support for integrating any Intel FPGA device with
204 the DPDK framework. It provides Intel FPGA Partial Bit Stream AFU (Accelerated
205 Function Unit) scan and drivers probe.
207 * **Added IFPGA (Intel FPGA) Rawdev Driver.**
209 Added a new Rawdev driver called IFPGA(Intel FPGA) Rawdev Driver, which cooperates
210 with OPAE (Open Programmable Acceleration Engine) share code provides common FPGA
211 management ops for FPGA operation.
213 See the :doc:`../rawdevs/ifpga_rawdev` programmer's guide for more details.
215 * **Added DPAA2 QDMA Driver (in rawdev).**
217 The DPAA2 QDMA is an implementation of the rawdev API, that provide means
218 to initiate a DMA transaction from CPU. The initiated DMA is performed
219 without CPU being involved in the actual DMA transaction.
221 See the :doc:`../rawdevs/dpaa2_qdma` guide for more details.
223 * **Added DPAA2 Command Interface Driver (in rawdev).**
225 The DPAA2 CMDIF is an implementation of the rawdev API, that provides
226 communication between the GPP and NXP's QorIQ based AIOP Block (Firmware).
227 Advanced IO Processor i.e. AIOP is clusters of programmable RISC engines
228 optimised for flexible networking and I/O operations. The communication
229 between GPP and AIOP is achieved via using DPCI devices exposed by MC for
230 GPP <--> AIOP interaction.
232 See the :doc:`../rawdevs/dpaa2_cmdif` guide for more details.
234 * **Added device event monitor framework.**
236 Added a general device event monitor framework at EAL, for device dynamic management.
237 Such as device hotplug awareness and actions adopted accordingly. The list of new APIs:
239 * ``rte_dev_event_monitor_start`` and ``rte_dev_event_monitor_stop`` are for
240 the event monitor enable and disable.
241 * ``rte_dev_event_callback_register`` and ``rte_dev_event_callback_unregister``
242 are for the user's callbacks register and unregister.
244 Linux uevent is supported as backend of this device event notification framework.
246 * **Added support for procinfo and pdump on eth vdev.**
248 For ethernet virtual devices (like tap, pcap, etc), with this feature, we can get
249 stats/xstats on shared memory from secondary process, and also pdump packets on
250 those virtual devices.
252 * **Added the BPF Library.**
254 The BPF Library provides the ability to load and execute
255 Enhanced Berkeley Packet Filter (eBPF) within user-space dpdk application.
256 Also it introduces basic framework to load/unload BPF-based filters
257 on eth devices (right now only via SW RX/TX callbacks).
258 It also adds dependency on libelf.
264 .. This section should contain API changes. Sample format:
266 * Add a short 1-2 sentence description of the API change. Use fixed width
267 quotes for ``rte_function_names`` or ``rte_struct_names``. Use the past
270 This section is a comment. Do not overwrite or remove it.
271 Also, make sure to start the actual text at the margin.
272 =========================================================
274 * service cores: no longer marked as experimental.
276 The service cores functions are no longer marked as experimental, and have
277 become part of the normal DPDK API and ABI. Any future ABI changes will be
278 announced at least one release before the ABI change is made. There are no
279 ABI breaking changes planned.
281 * eal: ``rte_lcore_has_role()`` return value changed.
283 This function now returns true or false, respectively,
284 rather than 0 or <0 for success or failure.
285 It makes use of the function more intuitive.
287 * mempool: capability flags and related functions have been removed.
289 Flags ``MEMPOOL_F_CAPA_PHYS_CONTIG`` and
290 ``MEMPOOL_F_CAPA_BLK_ALIGNED_OBJECTS`` were used by octeontx mempool
291 driver to customize generic mempool library behaviour.
292 Now the new driver callbacks ``calc_mem_size`` and ``populate`` may be
293 used to achieve it without specific knowledge in the generic code.
295 * mempool: xmem functions have been deprecated:
297 - ``rte_mempool_xmem_create``
298 - ``rte_mempool_xmem_size``
299 - ``rte_mempool_xmem_usage``
300 - ``rte_mempool_populate_iova_tab``
302 * mbuf: The control mbuf API has been removed in v18.05. The impacted
303 functions and macros are:
305 - ``rte_ctrlmbuf_init()``
306 - ``rte_ctrlmbuf_alloc()``
307 - ``rte_ctrlmbuf_free()``
308 - ``rte_ctrlmbuf_data()``
309 - ``rte_ctrlmbuf_len()``
310 - ``rte_is_ctrlmbuf()``
313 The packet mbuf API should be used as a replacement.
315 * meter: updated to accommodate configuration profiles.
317 The meter API is changed to support meter configuration profiles. The
318 configuration profile represents the set of configuration parameters
319 for a given meter object, such as the rates and sizes for the token
320 buckets. These configuration parameters were previously the part of meter
321 object internal data strcuture. The separation of the configuration
322 parameters from meter object data structure results in reducing its
323 memory footprint which helps in better cache utilization when large number
324 of meter objects are used.
326 * ethdev: The function ``rte_eth_dev_count``, often mis-used to iterate
327 over ports, is deprecated and replaced by ``rte_eth_dev_count_avail``.
328 There is also a new function ``rte_eth_dev_count_total`` to get the
329 total number of allocated ports, available or not.
330 The hotplug-proof applications should use ``RTE_ETH_FOREACH_DEV`` or
331 ``RTE_ETH_FOREACH_DEV_OWNED_BY`` as port iterators.
333 * ethdev, in struct ``struct rte_eth_dev_info``, field ``rte_pci_device *pci_dev``
334 replaced with field ``struct rte_device *device``.
336 * **Changes to semantics of rte_eth_dev_configure() parameters.**
338 If both the ``nb_rx_q`` and ``nb_tx_q`` parameters are zero,
339 ``rte_eth_dev_configure`` will now use PMD-recommended queue sizes, or if
340 recommendations are not provided by the PMD the function will use ethdev
341 fall-back values. Previously setting both of the parameters to zero would
342 have resulted in ``-EINVAL`` being returned.
344 * **Changes to semantics of rte_eth_rx_queue_setup() parameters.**
346 If the ``nb_rx_desc`` parameter is zero, ``rte_eth_rx_queue_setup`` will
347 now use the PMD-recommended Rx ring size, or in the case where the PMD
348 does not provide a recommendation, will use an ethdev-provided
349 fall-back value. Previously, setting ``nb_rx_desc`` to zero would have
350 resulted in an error.
352 * **Changes to semantics of rte_eth_tx_queue_setup() parameters.**
354 If the ``nb_tx_desc`` parameter is zero, ``rte_eth_tx_queue_setup`` will
355 now use the PMD-recommended Tx ring size, or in the case where the PMD
356 does not provide a recoomendation, will use an ethdev-provided
357 fall-back value. Previously, setting ``nb_tx_desc`` to zero would have
358 resulted in an error.
360 * ethdev: several changes were made to the flow API.
362 * Unused DUP action was removed.
363 * Actions semantics in flow rules: list order now matters ("first
364 to last" instead of "all simultaneously"), repeated actions are now
365 all performed, and they do not individually have (non-)terminating
367 * Flow rules are now always terminating unless a PASSTHRU action is
369 * C99-style flexible arrays were replaced with standard pointers in RSS
370 action and in RAW pattern item structures due to compatibility issues.
371 * The RSS action was modified to not rely on external
372 ``struct rte_eth_rss_conf`` anymore to instead expose its own and more
373 appropriately named configuration fields directly
374 (``rss_conf->rss_key`` => ``key``,
375 ``rss_conf->rss_key_len`` => ``key_len``,
376 ``rss_conf->rss_hf`` => ``types``,
377 ``num`` => ``queue_num``), and the addition of missing RSS parameters
378 (``func`` for RSS hash function to apply and ``level`` for the
379 encapsulation level).
380 * The VLAN pattern item (``struct rte_flow_item_vlan``) was modified to
381 include inner EtherType instead of outer TPID. Its default mask was also
382 modified to cover the VID part (lower 12 bits) of TCI only.
383 * A new transfer attribute was added to ``struct rte_flow_attr`` in order
384 to clarify the behavior of some pattern items.
385 * PF and VF pattern items are now only accepted by PMDs that implement
386 them (bnxt and i40e) when the transfer attribute is also present for
388 * Pattern item PORT was renamed PHY_PORT to avoid confusion with DPDK port
390 * An action counterpart to the PHY_PORT pattern item was added in order to
391 redirect matching traffic to a specific physical port.
392 * PORT_ID pattern item and actions were added to match and target DPDK
393 port IDs at a higher level than PHY_PORT.
395 * ethdev: change flow APIs regarding count action:
396 * ``rte_flow_create()`` API count action now requires the ``struct rte_flow_action_count``.
397 * ``rte_flow_query()`` API parameter changed from action type to action structure.
399 * ethdev: changes to offload API
401 A pure per-port offloading isn't requested to be repeated in [rt]x_conf->offloads to
402 ``rte_eth_[rt]x_queue_setup()``. Now any offloading enabled in ``rte_eth_dev_configure()``
403 can't be disabled by ``rte_eth_[rt]x_queue_setup()``. Any new added offloading which has
404 not been enabled in ``rte_eth_dev_configure()`` and is requested to be enabled in
405 ``rte_eth_[rt]x_queue_setup()`` must be per-queue type, otherwise trigger an error log.
407 * ethdev: runtime queue setup:
409 ``rte_eth_rx_queue_setup`` and ``rte_eth_tx_queue_setup`` can be called after
410 ``rte_eth_dev_start`` if device support runtime queue setup. Device driver can
411 expose this capability through ``rte_eth_dev_info_get``. A Rx or Tx queue be
412 setup at runtime need to be started explicitly by ``rte_eth_dev_rx_queue_start``
413 or ``rte_eth_dev_tx_queue_start``.
419 .. This section should contain ABI changes. Sample format:
421 * Add a short 1-2 sentence description of the ABI change that was announced
422 in the previous releases and made in this release. Use fixed width quotes
423 for ``rte_function_names`` or ``rte_struct_names``. Use the past tense.
425 This section is a comment. Do not overwrite or remove it.
426 Also, make sure to start the actual text at the margin.
427 =========================================================
429 * ring: the alignment constraints on the ring structure has been relaxed
430 to one cache line instead of two, and an empty cache line padding is
431 added between the producer and consumer structures. The size of the
432 structure and the offset of the fields remains the same on platforms
433 with 64B cache line, but change on other platforms.
435 * mempool: ops have changed.
437 A new callback ``calc_mem_size`` has been added to ``rte_mempool_ops``
438 to allow to customize required memory size calculation.
439 A new callback ``populate`` has been added to ``rte_mempool_ops``
440 to allow to customize objects population.
441 Callback ``get_capabilities`` has been removed from ``rte_mempool_ops``
442 since its features are covered by ``calc_mem_size`` and ``populate``
444 Callback ``register_memory_area`` has been removed from ``rte_mempool_ops``
445 since the new callback ``populate`` may be used instead of it.
447 * **Additional fields in rte_eth_dev_info.**
449 The ``rte_eth_dev_info`` structure has had two extra entries appended to the
450 end of it: ``default_rxportconf`` and ``default_txportconf``. Each of these
451 in turn are ``rte_eth_dev_portconf`` structures containing three fields of
452 type ``uint16_t``: ``burst_size``, ``ring_size``, and ``nb_queues``. These
453 are parameter values recommended for use by the PMD.
455 * ethdev: ABI for all flow API functions was updated.
457 This includes functions ``rte_flow_copy``, ``rte_flow_create``,
458 ``rte_flow_destroy``, ``rte_flow_error_set``, ``rte_flow_flush``,
459 ``rte_flow_isolate``, ``rte_flow_query`` and ``rte_flow_validate``, due to
460 changes in error type definitions (``enum rte_flow_error_type``), removal
461 of the unused DUP action (``enum rte_flow_action_type``), modified
462 behavior for flow rule actions (see API changes), removal of C99 flexible
463 array from RAW pattern item (``struct rte_flow_item_raw``), complete
464 rework of the RSS action definition (``struct rte_flow_action_rss``),
465 sanity fix in the VLAN pattern item (``struct rte_flow_item_vlan``) and
466 new transfer attribute (``struct rte_flow_attr``).
468 **New parameter added to rte_bbdev_op_cap_turbo_dec.**
470 A new parameter ``max_llr_modulus`` has been added to
471 ``rte_bbdev_op_cap_turbo_dec`` structure to specify maximal LLR (likelihood
472 ratio) absolute value.
474 * **BBdev Queue Groups split into UL/DL Groups**
476 Queue Groups have been split into UL/DL Groups in Turbo Software Driver.
477 They are independent for Decode/Encode. ``rte_bbdev_driver_info`` reflects
484 .. This section should contain removed items in this release. Sample format:
486 * Add a short 1-2 sentence description of the removed item in the past
489 This section is a comment. Do not overwrite or remove it.
490 Also, make sure to start the actual text at the margin.
491 =========================================================
497 .. This section should contain new known issues in this release. Sample format:
499 * **Add title in present tense with full stop.**
501 Add a short 1-2 sentence description of the known issue in the present
502 tense. Add information on any known workarounds.
504 This section is a comment. Do not overwrite or remove it.
505 Also, make sure to start the actual text at the margin.
506 =========================================================
508 * **pdump is not compatible with old applications.**
510 As we changed to use generic multi-process communication for pdump negotiation
511 instead of previous dedicated unix socket way, pdump applications, including
512 dpdk-pdump example and any other applications using librte_pdump, cannot work
513 with older version DPDK primary applications.
516 Shared Library Versions
517 -----------------------
519 .. Update any library version updated in this release and prepend with a ``+``
523 + librte_cfgfile.so.2
526 This section is a comment. Do not overwrite or remove it.
527 =========================================================
530 The libraries prepended with a plus sign were incremented in this version.
536 librte_bitratestats.so.2
539 librte_bus_fslmc.so.1
544 + librte_common_octeontx.so.1
545 + librte_compressdev.so.1
546 librte_cryptodev.so.4
547 librte_distributor.so.1
550 + librte_eventdev.so.4
551 librte_flow_classify.so.1
559 librte_latencystats.so.1
562 + librte_mempool.so.4
572 librte_pmd_ixgbe.so.2
573 + librte_pmd_dpaa2_cmdif.so.1
574 + librte_pmd_dpaa2_qdma.so.1
576 librte_pmd_softnic.so.1
577 librte_pmd_vhost.so.2
593 .. This section should contain a list of platforms that were tested with this
598 * <vendor> platform with <vendor> <type of devices> combinations
603 * Other relevant details...
605 This section is a comment. Do not overwrite or remove it.
606 Also, make sure to start the actual text at the margin.
607 =========================================================
609 * Intel(R) platforms with Intel(R) NICs combinations
613 * Intel(R) Atom(TM) CPU C2758 @ 2.40GHz
614 * Intel(R) Xeon(R) CPU D-1541 @ 2.10GHz
615 * Intel(R) Xeon(R) CPU E5-4667 v3 @ 2.00GHz
616 * Intel(R) Xeon(R) CPU E5-2680 v2 @ 2.80GHz
617 * Intel(R) Xeon(R) CPU E5-2699 v4 @ 2.20GHz
618 * Intel(R) Xeon(R) CPU E5-2695 v4 @ 2.10GHz
619 * Intel(R) Xeon(R) CPU E5-2658 v2 @ 2.40GHz
620 * Intel(R) Xeon(R) CPU E5-2658 v3 @ 2.20GHz
621 * Intel(R) Xeon(R) Platinum 8180 CPU @ 2.50GHz
630 * Red Hat Enterprise Linux Server release 7.3
631 * SUSE Enterprise Linux 12
640 * Intel(R) 82599ES 10 Gigabit Ethernet Controller
642 * Firmware version: 0x61bf0001
643 * Device id (pf/vf): 8086:10fb / 8086:10ed
644 * Driver version: 5.2.3 (ixgbe)
646 * Intel(R) Corporation Ethernet Connection X552/X557-AT 10GBASE-T
648 * Firmware version: 0x800003e7
649 * Device id (pf/vf): 8086:15ad / 8086:15a8
650 * Driver version: 4.4.6 (ixgbe)
652 * Intel(R) Ethernet Converged Network Adapter X710-DA4 (4x10G)
654 * Firmware version: 6.01 0x80003221
655 * Device id (pf/vf): 8086:1572 / 8086:154c
656 * Driver version: 2.4.6 (i40e)
658 * Intel Corporation Ethernet Connection X722 for 10GbE SFP+ (4x10G)
660 * Firmware version: 3.33 0x80000fd5 0.0.0
661 * Device id (pf/vf): 8086:37d0 / 8086:37cd
662 * Driver version: 2.4.3 (i40e)
664 * Intel(R) Ethernet Converged Network Adapter XXV710-DA2 (2x25G)
666 * Firmware version: 6.01 0x80003221
667 * Device id (pf/vf): 8086:158b / 8086:154c
668 * Driver version: 2.4.6 (i40e)
670 * Intel(R) Ethernet Converged Network Adapter XL710-QDA2 (2X40G)
672 * Firmware version: 6.01 0x8000321c
673 * Device id (pf/vf): 8086:1583 / 8086:154c
674 * Driver version: 2.4.6 (i40e)
676 * Intel(R) Corporation I350 Gigabit Network Connection
678 * Firmware version: 1.63, 0x80000dda
679 * Device id (pf/vf): 8086:1521 / 8086:1520
680 * Driver version: 5.4.0-k (igb)