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_17_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 * **Reorganized the mbuf structure.**
46 * Align fields to facilitate the writing of ``data_off``, ``refcnt``, and
47 ``nb_segs`` in one operation.
48 * Use 2 bytes for port and number of segments.
49 * Move the sequence number in the second cache line.
50 * Add a timestamp field.
51 * Set default value for ``refcnt``, ``next`` and ``nb_segs`` at mbuf free.
53 * **Added mbuf raw free API**
55 Moved ``rte_mbuf_raw_free()`` and ``rte_pktmbuf_prefree_seg()`` functions to
58 * **Added free Tx mbuf on demand API.**
60 Added a new function ``rte_eth_tx_done_cleanup()`` which allows an application
61 to request the driver to release mbufs from their Tx ring that are no longer
62 in use, independent of whether or not the ``tx_rs_thresh`` has been crossed.
64 * **Added EAL dynamic log framework.**
66 Added new APIs to dynamically register named log types, and control
67 the level of each type independently.
69 * **Added descriptor status ethdev API.**
71 Added a new API to get the status of a descriptor.
73 For Rx, it is almost similar to the ``rx_descriptor_done`` API, except
74 it differentiates descriptors which are hold by the driver and not
75 returned to the hardware. For Tx, it is a new API.
77 * **Increased number of next hops for LPM IPv6 to 2^21.**
79 The next_hop field is extended from 8 bits to 21 bits for IPv6.
81 * **Added VFIO hotplug support.**
83 How hotplug supported with UIO and VFIO drivers.
85 * **Added powerpc support in pci probing for vfio-pci devices.**
87 sPAPR IOMMU based pci probing enabled for vfio-pci devices.
89 * **Kept consistent PMD batching behaviour.**
91 Removed the limit of fm10k/i40e/ixgbe TX burst size and vhost RX/TX burst size
92 in order to support the same policy of "make an best effort to RX/TX pkts"
95 * **Updated the ixgbe base driver.**
97 Updated the ixgbe base driver, including the following changes:
99 * Add link block check for KR.
100 * Complete HW initialization even if SFP is not present.
101 * Add VF xcast promiscuous mode.
103 * **Added powerpc support for i40e and its vector PMD .**
105 i40e PMD and its vector PMD enabled by default in powerpc.
107 * **Added VF max bandwidth setting on i40e.**
109 i40e HW supports to set the max bandwidth for a VF. Enable this capability.
111 * **Added VF TC min bandwidth setting on i40e.**
113 i40e HW supports to set the allocated bandwidth for a TC on a VF. Enable this
116 * **Added VF TC max bandwidth setting on i40e.**
118 i40e HW supports to set the max bandwidth for a TC on a VF. Enable this
121 * **Added TC strict priority mode setting on i40e.**
123 There're 2 TX scheduling modes supported for TCs by i40e HW, round ribon mode
124 and strict priority mode. By default it's round robin mode. Enable the
125 capability to change the TX scheduling mode for a TC. It's a global setting
128 * **Added i40e dynamic device personalization support.**
130 * Added dynamic device personalization processing to i40e FW.
132 * **Added TSO support for tunneled and non-tunneled packets on mlx5 driver.**
134 Added support for Hardware TSO for tunneled and non-tunneled packets.
135 Tunneling protocols supported are GRE and VXLAN.
137 * **Added support for Rx interrupts on mlx5 driver.**
139 Rx queues can be armed with an interrupt which will trigger on the
142 * **Updated the sfc_efx driver.**
144 * Generic flow API support for Ethernet, VLAN, IPv4, IPv6, UDP and TCP
145 pattern items with QUEUE action for ingress traffic.
147 * Support virtual functions (VFs)
149 * **Added LiquidIO network PMD.**
151 Added poll mode driver support for Cavium LiquidIO II server adapter VFs.
153 * **Added Atomic Rules Arkville PMD.**
155 Added a new poll mode driver for the Arkville family of
156 devices from Atomic Rules. The net/ark PMD supports line-rate
157 agnostic, multi-queue data movement on Arkville core FPGA instances.
159 * **Added support for NXP DPAA2 - FSLMC bus.**
161 Added the new bus "fslmc" driver for NXP DPAA2 devices. See the
162 "Network Interface Controller Drivers" document for more details on this new
165 * **Added support for NXP DPAA2 Network PMD.**
167 Added the new "dpaa2" net driver for NXP DPAA2 devices. See the
168 "Network Interface Controller Drivers" document for more details on this new
171 * **Added support for the Wind River Systems AVP PMD.**
173 Added a new networking driver for the AVP device type. Theses devices are
174 specific to the Wind River Systems virtualization platforms.
176 * **Added vmxnet3 version 3 support.**
178 Added support for vmxnet3 version 3 which includes several
179 performance enhancements viz. configurable TX data ring, Receive
180 Data Ring, ability to register memory regions.
182 * **Updated the tap driver.**
184 * Support MTU modification.
185 * Support packet type for Rx.
186 * Support segmented packets on Rx and Tx.
187 * Speed up Rx on tap when no packets are available.
188 * Support capturing traffic from another netdevice.
189 * Dynamically change link status when the underlying interface state changes.
190 * Generic flow API support for Ethernet, VLAN, IPv4, IPv6, UDP and TCP pattern
191 items with DROP, QUEUE and PASSTHRU actions for ingress traffic.
193 * **Added MTU feature support to Virtio and Vhost.**
195 Implemented new Virtio MTU feature into Vhost and Virtio:
197 * Add ``rte_vhost_mtu_get()`` API to Vhost library.
198 * Enable Vhost PMD's MTU get feature.
199 * Get max MTU value from host in Virtio PMD
201 * **Added interrupt mode support for virtio-user.**
203 Implemented Rxq interrupt mode and LSC support for virtio-user as a virtual
204 device. Supported cases:
206 * Rxq interrupt for virtio-user + vhost-user as the backend.
207 * Rxq interrupt for virtio-user + vhost-kernel as the backend.
208 * LSC interrupt for virtio-user + vhost-user as the backend.
210 * **Added event driven programming model library (rte_eventdev).**
212 This API introduces event driven programming model.
214 In a polling model, lcores poll ethdev ports and associated
215 rx queues directly to look for packet. In an event driven model,
216 by contrast, lcores call the scheduler that selects packets for
217 them based on programmer-specified criteria. Eventdev library
218 added support for event driven programming model, which offer
219 applications automatic multicore scaling, dynamic load balancing,
220 pipelining, packet ingress order maintenance and
221 synchronization services to simplify application packet processing.
223 By introducing event driven programming model, DPDK can support
224 both polling and event driven programming models for packet processing,
225 and applications are free to choose whatever model
226 (or combination of the two) that best suits their needs.
228 * **Added Software Eventdev PMD.**
230 Added support for the software eventdev PMD. The software eventdev is a
231 software based scheduler device that implements the eventdev API. This
232 PMD allows an application to configure a pipeline using the eventdev
233 library, and run the scheduling workload on a CPU core.
235 * **Added Cavium OCTEONTX Eventdev PMD.**
237 Added the new octeontx ssovf eventdev driver for OCTEONTX devices. See the
238 "Event Device Drivers" document for more details on this new driver.
240 * **Added information metric library.**
242 A library that allows information metrics to be added and updated
243 by producers, typically other libraries, for later retrieval by
244 consumers such as applications. It is intended to provide a
245 reporting mechanism that is independent of other libraries such
248 * **Added bit-rate calculation library.**
250 A library that can be used to calculate device bit-rates. Calculated
251 bitrates are reported using the metrics library.
253 * **Added latency stats library.**
255 A library that measures packet latency. The collected statistics are jitter
256 and latency. For latency the minimum, average, and maximum is measured.
258 * **Updated the Cryptodev Scheduler PMD.**
260 * Added packet-size based distribution mode, which distributes the enqueued
261 crypto operations among two slaves, based on their data lengths.
262 * Added fail-over scheduling mode, which enqueues crypto operations to a
263 primary slave first. Then, any operation that cannot be enqueued is
264 enqueued to a secondary slave.
266 * **Updated the QAT PMD.**
268 The QAT PMD has been updated with additional support for:
270 * AES DOCSIS BPI algorithm.
271 * DES DOCSIS BPI algorithm.
272 * ZUC EEA3/EIA3 algorithms.
274 * **Updated the AESNI MB PMD.**
276 The AESNI MB PMD has been updated with additional support for:
278 * AES DOCSIS BPI algorithm.
280 * **Updated the OpenSSL PMD.**
282 The OpenSSL PMD has been updated with additional support for:
284 * DES DOCSIS BPI algorithm.
289 .. This section should contain bug fixes added to the relevant
290 sections. Sample format:
292 * **code/section Fixed issue in the past tense with a full stop.**
294 Add a short 1-2 sentence description of the resolved issue in the past
297 The title should contain the code/lib section like a commit message.
299 Add the entries in alphabetic order in the relevant sections below.
301 This section is a comment. do not overwrite or remove it.
302 Also, make sure to start the actual text at the margin.
303 =========================================================
329 .. This section should contain new known issues in this release. Sample format:
331 * **Add title in present tense with full stop.**
333 Add a short 1-2 sentence description of the known issue in the present
334 tense. Add information on any known workarounds.
336 This section is a comment. do not overwrite or remove it.
337 Also, make sure to start the actual text at the margin.
338 =========================================================
340 * **LSC interrupt cannot work for virtio-user + vhost-kernel.**
342 LSC interrupt cannot be detected when setting the backend, tap device,
343 up/down as we fail to find a way to monitor such event.
349 .. This section should contain API changes. Sample format:
351 * Add a short 1-2 sentence description of the API change. Use fixed width
352 quotes for ``rte_function_names`` or ``rte_struct_names``. Use the past
355 This section is a comment. do not overwrite or remove it.
356 Also, make sure to start the actual text at the margin.
357 =========================================================
359 * The LPM ``next_hop`` field is extended from 8 bits to 21 bits for IPv6
360 while keeping ABI compatibility.
362 * **Reworked rte_ring library**
364 The rte_ring library has been reworked and updated. The following changes
365 have been made to it:
367 * removed the build-time setting ``CONFIG_RTE_RING_SPLIT_PROD_CONS``
368 * removed the build-time setting ``CONFIG_RTE_LIBRTE_RING_DEBUG``
369 * removed the build-time setting ``CONFIG_RTE_RING_PAUSE_REP_COUNT``
370 * removed the function ``rte_ring_set_water_mark`` as part of a general
371 removal of watermarks support in the library.
372 * added an extra parameter to the burst/bulk enqueue functions to
373 return the number of free spaces in the ring after enqueue. This can
374 be used by an application to implement its own watermark functionality.
375 * added an extra parameter to the burst/bulk dequeue functions to return
376 the number elements remaining in the ring after dequeue.
377 * changed the return value of the enqueue and dequeue bulk functions to
378 match that of the burst equivalents. In all cases, ring functions which
379 operate on multiple packets now return the number of elements enqueued
380 or dequeued, as appropriate. The updated functions are:
382 - ``rte_ring_mp_enqueue_bulk``
383 - ``rte_ring_sp_enqueue_bulk``
384 - ``rte_ring_enqueue_bulk``
385 - ``rte_ring_mc_dequeue_bulk``
386 - ``rte_ring_sc_dequeue_bulk``
387 - ``rte_ring_dequeue_bulk``
389 NOTE: the above functions all have different parameters as well as
390 different return values, due to the other listed changes above. This
391 means that all instances of the functions in existing code will be
392 flagged by the compiler. The return value usage should be checked
393 while fixing the compiler error due to the extra parameter.
395 * **Reworked rte_vhost library**
397 The rte_vhost library has been reworked to make it generic enough so that
398 user could build other vhost-user drivers on top of it. To achieve that,
399 following changes have been made:
401 * The following vhost-pmd APIs are removed:
403 * ``rte_eth_vhost_feature_disable``
404 * ``rte_eth_vhost_feature_enable``
405 * ``rte_eth_vhost_feature_get``
407 * The vhost API ``rte_vhost_driver_callback_register(ops)`` is reworked to
408 be per vhost-user socket file. Thus, it takes one more argument:
409 ``rte_vhost_driver_callback_register(path, ops)``.
411 * The vhost API ``rte_vhost_get_queue_num`` is deprecated, instead,
412 ``rte_vhost_get_vring_num`` should be used.
414 * Following macros are removed in ``rte_virtio_net.h``
420 * Following net specific header files are removed in ``rte_virtio_net.h``
422 * ``linux/virtio_net.h``
427 * The vhost struct ``virtio_net_device_ops`` is renamed to
430 * The vhost API ``rte_vhost_driver_session_start`` is removed. Instead,
431 ``rte_vhost_driver_start`` should be used, and no need to create a
434 * The vhost public header file ``rte_virtio_net.h`` is renamed to
441 .. This section should contain ABI changes. Sample format:
443 * Add a short 1-2 sentence description of the ABI change that was announced
444 in the previous releases and made in this release. Use fixed width quotes
445 for ``rte_function_names`` or ``rte_struct_names``. Use the past tense.
447 This section is a comment. do not overwrite or remove it.
448 Also, make sure to start the actual text at the margin.
449 =========================================================
451 * **Reorganized the mbuf structure.**
453 The order and size of the fields in the ``mbuf`` structure changed,
454 as described in the `New Features`_ section.
456 * The ``rte_cryptodev_info.sym`` structure has new field ``max_nb_sessions_per_qp``
457 to support drivers which may support limited number of sessions per queue_pair.
463 .. This section should contain removed items in this release. Sample format:
465 * Add a short 1-2 sentence description of the removed item in the past
468 This section is a comment. do not overwrite or remove it.
469 Also, make sure to start the actual text at the margin.
470 =========================================================
472 * KNI vhost support removed.
474 * dpdk_qat sample application removed.
476 Shared Library Versions
477 -----------------------
479 .. Update any library version updated in this release and prepend with a ``+``
483 + librte_cfgfile.so.2
486 This section is a comment. do not overwrite or remove it.
487 =========================================================
490 The libraries prepended with a plus sign were incremented in this version.
495 + librte_bitratestats.so.1
498 librte_cryptodev.so.2
499 librte_distributor.so.1
507 + librte_latencystats.so.1
512 + librte_metrics.so.1
531 .. This section should contain a list of platforms that were tested with this
536 * <vendor> platform with <vendor> <type of devices> combinations
541 * Other relevant details...
543 This section is a comment. do not overwrite or remove it.
544 Also, make sure to start the actual text at the margin.
545 =========================================================