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 * **Added free Tx mbuf on demand API.**
46 Added a new function ``rte_eth_tx_done_cleanup()`` which allows an application
47 to request the driver to release mbufs from their Tx ring that are no longer
48 in use, independent of whether or not the ``tx_rs_thresh`` has been crossed.
50 * **Added descriptor status ethdev API.**
52 Added a new API to get the status of a descriptor.
54 For Rx, it is almost similar to the ``rx_descriptor_done`` API, except
55 it differentiates descriptors which are hold by the driver and not
56 returned to the hardware. For Tx, it is a new API.
58 * **Increased number of next hops for LPM IPv6 to 2^21.**
60 The next_hop field is extended from 8 bits to 21 bits for IPv6.
62 * **Added VFIO hotplug support.**
64 How hotplug supported with UIO and VFIO drivers.
66 * **Added powerpc support in pci probing for vfio-pci devices.**
68 sPAPR IOMMU based pci probing enabled for vfio-pci devices.
70 * **Kept consistent PMD batching behaviour.**
72 Removed the limit of fm10k/i40e/ixgbe TX burst size and vhost RX/TX burst size
73 in order to support the same policy of "make an best effort to RX/TX pkts"
76 * **Updated the ixgbe base driver.**
78 Updated the ixgbe base driver, including the following changes:
80 * Add link block check for KR.
81 * Complete HW initialization even if SFP is not present.
82 * Add VF xcast promiscuous mode.
84 * **Added powerpc support for i40e and its vector PMD .**
86 i40e PMD and its vector PMD enabled by default in powerpc.
88 * **Added VF max bandwidth setting on i40e.**
90 i40e HW supports to set the max bandwidth for a VF. Enable this capability.
92 * **Added VF TC min bandwidth setting on i40e.**
94 i40e HW supports to set the allocated bandwidth for a TC on a VF. Enable this
97 * **Added VF TC max bandwidth setting on i40e.**
99 i40e HW supports to set the max bandwidth for a TC on a VF. Enable this
102 * **Added TC strict priority mode setting on i40e.**
104 There're 2 TX scheduling modes supported for TCs by i40e HW, round ribon mode
105 and strict priority mode. By default it's round robin mode. Enable the
106 capability to change the TX scheduling mode for a TC. It's a global setting
109 * **Added i40e dynamic device personalization support.**
111 * Added dynamic device personalization processing to i40e FW.
113 * **Added TSO support for tunneled and non-tunneled packets on mlx5 driver.**
115 Added support for Hardware TSO for tunneled and non-tunneled packets.
116 Tunneling protocols supported are GRE and VXLAN.
118 * **Added support for Rx interrupts on mlx5 driver.**
120 Rx queues can be armed with an interrupt which will trigger on the
123 * **Updated the sfc_efx driver.**
125 * Generic flow API support for Ethernet, VLAN, IPv4, IPv6, UDP and TCP
126 pattern items with QUEUE action for ingress traffic.
128 * Support virtual functions (VFs)
130 * **Added LiquidIO network PMD.**
132 Added poll mode driver support for Cavium LiquidIO II server adapter VFs.
134 * **Added support for the Wind River Systems AVP PMD.**
136 Added a new networking driver for the AVP device type. Theses devices are
137 specific to the Wind River Systems virtualization platforms.
139 * **Added vmxnet3 version 3 support.**
141 Added support for vmxnet3 version 3 which includes several
142 performance enhancements viz. configurable TX data ring, Receive
143 Data Ring, ability to register memory regions.
145 * **Updated the tap driver.**
147 * Support MTU modification.
148 * Support packet type for Rx.
149 * Support segmented packets on Rx and Tx.
150 * Speed up Rx on tap when no packets are available.
151 * Support capturing traffic from another netdevice.
152 * Dynamically change link status when the underlying interface state changes.
153 * Generic flow API support for Ethernet, VLAN, IPv4, IPv6, UDP and TCP pattern
154 items with DROP, QUEUE and PASSTHRU actions for ingress traffic.
156 * **Added MTU feature support to Virtio and Vhost.**
158 Implemented new Virtio MTU feature into Vhost and Virtio:
160 * Add ``rte_vhost_mtu_get()`` API to Vhost library.
161 * Enable Vhost PMD's MTU get feature.
162 * Get max MTU value from host in Virtio PMD
164 * **Added interrupt mode support for virtio-user.**
166 Implemented Rxq interrupt mode and LSC support for virtio-user as a virtual
167 device. Supported cases:
169 * Rxq interrupt for virtio-user + vhost-user as the backend.
170 * Rxq interrupt for virtio-user + vhost-kernel as the backend.
171 * LSC interrupt for virtio-user + vhost-user as the backend.
173 * **Added event driven programming model library (rte_eventdev).**
175 This API introduces event driven programming model.
177 In a polling model, lcores poll ethdev ports and associated
178 rx queues directly to look for packet. In an event driven model,
179 by contrast, lcores call the scheduler that selects packets for
180 them based on programmer-specified criteria. Eventdev library
181 added support for event driven programming model, which offer
182 applications automatic multicore scaling, dynamic load balancing,
183 pipelining, packet ingress order maintenance and
184 synchronization services to simplify application packet processing.
186 By introducing event driven programming model, DPDK can support
187 both polling and event driven programming models for packet processing,
188 and applications are free to choose whatever model
189 (or combination of the two) that best suits their needs.
191 * **Added Software Eventdev PMD.**
193 Added support for the software eventdev PMD. The software eventdev is a
194 software based scheduler device that implements the eventdev API. This
195 PMD allows an application to configure a pipeline using the eventdev
196 library, and run the scheduling workload on a CPU core.
202 .. This section should contain bug fixes added to the relevant
203 sections. Sample format:
205 * **code/section Fixed issue in the past tense with a full stop.**
207 Add a short 1-2 sentence description of the resolved issue in the past
210 The title should contain the code/lib section like a commit message.
212 Add the entries in alphabetic order in the relevant sections below.
214 This section is a comment. do not overwrite or remove it.
215 Also, make sure to start the actual text at the margin.
216 =========================================================
242 .. This section should contain new known issues in this release. Sample format:
244 * **Add title in present tense with full stop.**
246 Add a short 1-2 sentence description of the known issue in the present
247 tense. Add information on any known workarounds.
249 This section is a comment. do not overwrite or remove it.
250 Also, make sure to start the actual text at the margin.
251 =========================================================
253 * **LSC interrupt cannot work for virtio-user + vhost-kernel.**
255 LSC interrupt cannot be detected when setting the backend, tap device,
256 up/down as we fail to find a way to monitor such event.
262 .. This section should contain API changes. Sample format:
264 * Add a short 1-2 sentence description of the API change. Use fixed width
265 quotes for ``rte_function_names`` or ``rte_struct_names``. Use the past
268 This section is a comment. do not overwrite or remove it.
269 Also, make sure to start the actual text at the margin.
270 =========================================================
272 * The LPM ``next_hop`` field is extended from 8 bits to 21 bits for IPv6
273 while keeping ABI compatibility.
275 * **Reworked rte_ring library**
277 The rte_ring library has been reworked and updated. The following changes
278 have been made to it:
280 * removed the build-time setting ``CONFIG_RTE_RING_SPLIT_PROD_CONS``
281 * removed the build-time setting ``CONFIG_RTE_LIBRTE_RING_DEBUG``
282 * removed the build-time setting ``CONFIG_RTE_RING_PAUSE_REP_COUNT``
283 * removed the function ``rte_ring_set_water_mark`` as part of a general
284 removal of watermarks support in the library.
285 * added an extra parameter to the burst/bulk enqueue functions to
286 return the number of free spaces in the ring after enqueue. This can
287 be used by an application to implement its own watermark functionality.
288 * added an extra parameter to the burst/bulk dequeue functions to return
289 the number elements remaining in the ring after dequeue.
290 * changed the return value of the enqueue and dequeue bulk functions to
291 match that of the burst equivalents. In all cases, ring functions which
292 operate on multiple packets now return the number of elements enqueued
293 or dequeued, as appropriate. The updated functions are:
295 - ``rte_ring_mp_enqueue_bulk``
296 - ``rte_ring_sp_enqueue_bulk``
297 - ``rte_ring_enqueue_bulk``
298 - ``rte_ring_mc_dequeue_bulk``
299 - ``rte_ring_sc_dequeue_bulk``
300 - ``rte_ring_dequeue_bulk``
302 NOTE: the above functions all have different parameters as well as
303 different return values, due to the other listed changes above. This
304 means that all instances of the functions in existing code will be
305 flagged by the compiler. The return value usage should be checked
306 while fixing the compiler error due to the extra parameter.
308 * **Reworked rte_vhost library**
310 The rte_vhost library has been reworked to make it generic enough so that
311 user could build other vhost-user drivers on top of it. To achieve that,
312 following changes have been made:
314 * The following vhost-pmd APIs are removed:
316 * ``rte_eth_vhost_feature_disable``
317 * ``rte_eth_vhost_feature_enable``
318 * ``rte_eth_vhost_feature_get``
320 * The vhost API ``rte_vhost_driver_callback_register(ops)`` is reworked to
321 be per vhost-user socket file. Thus, it takes one more argument:
322 ``rte_vhost_driver_callback_register(path, ops)``.
324 * The vhost API ``rte_vhost_get_queue_num`` is deprecated, instead,
325 ``rte_vhost_get_vring_num`` should be used.
327 * Following macros are removed in ``rte_virtio_net.h``
333 * Following net specific header files are removed in ``rte_virtio_net.h``
335 * ``linux/virtio_net.h``
340 * The vhost struct ``virtio_net_device_ops`` is renamed to
343 * The vhost API ``rte_vhost_driver_session_start`` is removed. Instead,
344 ``rte_vhost_driver_start`` should be used, and no need to create a
347 * The vhost public header file ``rte_virtio_net.h`` is renamed to
354 .. This section should contain ABI changes. Sample format:
356 * Add a short 1-2 sentence description of the ABI change that was announced
357 in the previous releases and made in this release. Use fixed width quotes
358 for ``rte_function_names`` or ``rte_struct_names``. Use the past tense.
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 =========================================================
368 .. This section should contain removed items in this release. Sample format:
370 * Add a short 1-2 sentence description of the removed item in the past
373 This section is a comment. do not overwrite or remove it.
374 Also, make sure to start the actual text at the margin.
375 =========================================================
377 * KNI vhost support removed.
380 Shared Library Versions
381 -----------------------
383 .. Update any library version updated in this release and prepend with a ``+``
387 + librte_cfgfile.so.2
390 This section is a comment. do not overwrite or remove it.
391 =========================================================
394 The libraries prepended with a plus sign were incremented in this version.
401 librte_cryptodev.so.2
402 librte_distributor.so.1
432 .. This section should contain a list of platforms that were tested with this
437 * <vendor> platform with <vendor> <type of devices> combinations
442 * Other relevant details...
444 This section is a comment. do not overwrite or remove it.
445 Also, make sure to start the actual text at the margin.
446 =========================================================