1..  SPDX-License-Identifier: BSD-3-Clause
2    Copyright 2016 The DPDK contributors
3
4DPDK Release 16.11
5==================
6
7.. **Read this first.**
8
9   The text below explains how to update the release notes.
10
11   Use proper spelling, capitalization and punctuation in all sections.
12
13   Variable and config names should be quoted as fixed width text: ``LIKE_THIS``.
14
15   Build the docs and view the output file to ensure the changes are correct::
16
17      make doc-guides-html
18
19      firefox build/doc/html/guides/rel_notes/release_16_11.html
20
21
22New Features
23------------
24
25.. This section should contain new features added in this release. Sample format:
26
27   * **Add a title in the past tense with a full stop.**
28
29     Add a short 1-2 sentence description in the past tense. The description
30     should be enough to allow someone scanning the release notes to understand
31     the new feature.
32
33     If the feature adds a lot of sub-features you can use a bullet list like this.
34
35     * Added feature foo to do something.
36     * Enhanced feature bar to do something else.
37
38     Refer to the previous release notes for examples.
39
40     This section is a comment. Make sure to start the actual text at the margin.
41
42
43* **Added software parser for packet type.**
44
45  * Added a new function ``rte_pktmbuf_read()`` to read the packet data from an
46    mbuf chain, linearizing if required.
47  * Added a new function ``rte_net_get_ptype()`` to parse an Ethernet packet
48    in an mbuf chain and retrieve its packet type from software.
49  * Added new functions ``rte_get_ptype_*()`` to dump a packet type as a string.
50
51* **Improved offloads support in mbuf.**
52
53  * Added a new function ``rte_raw_cksum_mbuf()`` to process the checksum of
54    data embedded in an mbuf chain.
55  * Added new Rx checksum flags in mbufs to describe more states: unknown,
56    good, bad, or not present (useful for virtual drivers). This modification
57    was done for IP and L4.
58  * Added a new Rx LRO mbuf flag, used when packets are coalesced. This
59    flag indicates that the segment size of original packets is known.
60
61* **Added vhost-user dequeue zero copy support.**
62
63  The copy in the dequeue path is avoided in order to improve the performance.
64  In the VM2VM case, the boost is quite impressive. The bigger the packet size,
65  the bigger performance boost you may get. However, for the VM2NIC case, there
66  are some limitations, so the boost is not as  impressive as the VM2VM case.
67  It may even drop quite a bit for small packets.
68
69  For that reason, this feature is disabled by default. It can be enabled when
70  the ``RTE_VHOST_USER_DEQUEUE_ZERO_COPY`` flag is set. Check the VHost section
71  of the Programming Guide for more information.
72
73* **Added vhost-user indirect descriptors support.**
74
75  If the indirect descriptor feature is enabled, each packet sent by the guest
76  will take exactly one slot in the enqueue virtqueue. Without this feature, as in
77  the current version, even 64 bytes packets take two slots with Virtio PMD on guest
78  side.
79
80  The main impact is better performance for 0% packet loss use-cases, as it
81  behaves as if the virtqueue size was enlarged, so more packets can be buffered
82  in the case of system perturbations. On the downside, small performance degradations
83  were measured when running micro-benchmarks.
84
85* **Added vhost PMD xstats.**
86
87  Added extended statistics to vhost PMD from a per port perspective.
88
89* **Supported offloads with virtio.**
90
91  Added support for the following offloads in virtio:
92
93  * Rx/Tx checksums.
94  * LRO.
95  * TSO.
96
97* **Added virtio NEON support for ARM.**
98
99  Added NEON support for ARM based virtio.
100
101* **Updated the ixgbe base driver.**
102
103  Updated the ixgbe base driver, including the following changes:
104
105  * Added X550em_a 10G PHY support.
106  * Added support for flow control auto negotiation for X550em_a 1G PHY.
107  * Added X550em_a FW ALEF support.
108  * Increased mailbox version to ``ixgbe_mbox_api_13``.
109  * Added two MAC operations for Hyper-V support.
110
111* **Added APIs for VF management to the ixgbe PMD.**
112
113  Eight new APIs have been added to the ixgbe PMD for VF management from the PF.
114  The declarations for the API's can be found in ``rte_pmd_ixgbe.h``.
115
116* **Updated the enic driver.**
117
118  * Added update to use interrupt for link status checking instead of polling.
119  * Added more flow director modes on UCS Blade with firmware version >= 2.0(13e).
120  * Added full support for MTU update.
121  * Added support for the ``rte_eth_rx_queue_count`` function.
122
123* **Updated the mlx5 driver.**
124
125  * Added support for RSS hash results.
126  * Added several performance improvements.
127  * Added several bug fixes.
128
129* **Updated the QAT PMD.**
130
131  The QAT PMD was updated with additional support for:
132
133  * MD5_HMAC algorithm.
134  * SHA224-HMAC algorithm.
135  * SHA384-HMAC algorithm.
136  * GMAC algorithm.
137  * KASUMI (F8 and F9) algorithm.
138  * 3DES algorithm.
139  * NULL algorithm.
140  * C3XXX device.
141  * C62XX device.
142
143* **Added openssl PMD.**
144
145  A new crypto PMD has been added, which provides several ciphering and hashing algorithms.
146  All cryptography operations use the Openssl library crypto API.
147
148* **Updated the IPsec example.**
149
150  Updated the IPsec example with the following support:
151
152  * Configuration file support.
153  * AES CBC IV generation with cipher forward function.
154  * AES GCM/CTR mode.
155
156* **Added support for new gcc -march option.**
157
158  The GCC 4.9 ``-march`` option supports the Intel processor code names.
159  The config option ``RTE_MACHINE`` can be used to pass code names to the compiler via the ``-march`` flag.
160
161
162Resolved Issues
163---------------
164
165.. This section should contain bug fixes added to the relevant sections. Sample format:
166
167   * **code/section Fixed issue in the past tense with a full stop.**
168
169     Add a short 1-2 sentence description of the resolved issue in the past tense.
170     The title should contain the code/lib section like a commit message.
171     Add the entries in alphabetic order in the relevant sections below.
172
173   This section is a comment. Make sure to start the actual text at the margin.
174
175
176Drivers
177~~~~~~~
178
179* **enic: Fixed several flow director issues.**
180
181* **enic: Fixed inadvertent setting of L4 checksum ptype on ICMP packets.**
182
183* **enic: Fixed high driver overhead when servicing Rx queues beyond the first.**
184
185
186
187Known Issues
188------------
189
190.. This section should contain new known issues in this release. Sample format:
191
192   * **Add title in present tense with full stop.**
193
194     Add a short 1-2 sentence description of the known issue in the present
195     tense. Add information on any known workarounds.
196
197   This section is a comment. Make sure to start the actual text at the margin.
198
199* **L3fwd-power app does not work properly when Rx vector is enabled.**
200
201  The L3fwd-power app doesn't work properly with some drivers in vector mode
202  since the queue monitoring works differently between scalar and vector modes
203  leading to incorrect frequency scaling. In addition, L3fwd-power application
204  requires the mbuf to have correct packet type set but in some drivers the
205  vector mode must be disabled for this.
206
207  Therefore, in order to use L3fwd-power, vector mode should be disabled
208  via the config file.
209
210* **Digest address must be supplied for crypto auth operation on QAT PMD.**
211
212  The cryptodev API specifies that if the rte_crypto_sym_op.digest.data field,
213  and by inference the digest.phys_addr field which points to the same location,
214  is not set for an auth operation the driver is to understand that the digest
215  result is located immediately following the region over which the digest is
216  computed. The QAT PMD doesn't correctly handle this case and reads and writes
217  to an incorrect location.
218
219  Callers can workaround this by always supplying the digest virtual and
220  physical address fields in the rte_crypto_sym_op for an auth operation.
221
222
223API Changes
224-----------
225
226.. This section should contain API changes. Sample format:
227
228   * Add a short 1-2 sentence description of the API change. Use fixed width
229     quotes for ``rte_function_names`` or ``rte_struct_names``. Use the past tense.
230
231   This section is a comment. Make sure to start the actual text at the margin.
232
233* The driver naming convention has been changed to make them more
234  consistent. It especially impacts ``--vdev`` arguments. For example
235  ``eth_pcap`` becomes ``net_pcap`` and ``cryptodev_aesni_mb_pmd`` becomes
236  ``crypto_aesni_mb``.
237
238  For backward compatibility an alias feature has been enabled to support the
239  original names.
240
241* The log history has been removed.
242
243* The ``rte_ivshmem`` feature (including library and EAL code) has been removed
244  in 16.11 because it had some design issues which were not planned to be fixed.
245
246* The ``file_name`` data type of ``struct rte_port_source_params`` and
247  ``struct rte_port_sink_params`` is changed from ``char *`` to ``const char *``.
248
249* **Improved device/driver hierarchy and generalized hotplugging.**
250
251  The device and driver relationship has been restructured by introducing generic
252  classes. This paves the way for having PCI, VDEV and other device types as
253  instantiated objects rather than classes in themselves. Hotplugging has also
254  been generalized into EAL so that Ethernet or crypto devices can use the
255  common infrastructure.
256
257  * Removed ``pmd_type`` as a way of segregation of devices.
258  * Moved ``numa_node`` and ``devargs`` into ``rte_driver`` from
259    ``rte_pci_driver``. These can now be used by any instantiated object of
260    ``rte_driver``.
261  * Added ``rte_device`` class and all PCI and VDEV devices inherit from it
262  * Renamed devinit/devuninit handlers to probe/remove to make it more
263    semantically correct with respect to the device <=> driver relationship.
264  * Moved hotplugging support to EAL. Hereafter, PCI and vdev can use the
265    APIs ``rte_eal_dev_attach`` and ``rte_eal_dev_detach``.
266  * Renamed helpers and support macros to make them more synonymous
267    with their device types
268    (e.g. ``PMD_REGISTER_DRIVER`` => ``RTE_PMD_REGISTER_PCI``).
269  * Device naming functions have been generalized from ethdev and cryptodev
270    to EAL. ``rte_eal_pci_device_name`` has been introduced for obtaining
271    unique device name from PCI Domain-BDF description.
272  * Virtual device registration APIs have been added: ``rte_eal_vdrv_register``
273    and ``rte_eal_vdrv_unregister``.
274
275
276ABI Changes
277-----------
278
279.. This section should contain ABI changes. Sample format:
280
281   * Add a short 1-2 sentence description of the ABI change that was announced in
282     the previous releases and made in this release. Use fixed width quotes for
283     ``rte_function_names`` or ``rte_struct_names``. Use the past tense.
284
285   This section is a comment. Make sure to start the actual text at the margin.
286
287
288
289Shared Library Versions
290-----------------------
291
292.. Update any library version updated in this release and prepend with a ``+``
293   sign, like this:
294
295     libethdev.so.4
296     librte_acl.so.2
297   + librte_cfgfile.so.2
298     librte_cmdline.so.2
299
300
301
302The libraries prepended with a plus sign were incremented in this version.
303
304.. code-block:: diff
305
306     librte_acl.so.2
307     librte_cfgfile.so.2
308     librte_cmdline.so.2
309   + librte_cryptodev.so.2
310     librte_distributor.so.1
311   + librte_eal.so.3
312   + librte_ethdev.so.5
313     librte_hash.so.2
314     librte_ip_frag.so.1
315     librte_jobstats.so.1
316     librte_kni.so.2
317     librte_kvargs.so.1
318     librte_lpm.so.2
319     librte_mbuf.so.2
320     librte_mempool.so.2
321     librte_meter.so.1
322     librte_net.so.1
323     librte_pdump.so.1
324     librte_pipeline.so.3
325     librte_pmd_bond.so.1
326     librte_pmd_ring.so.2
327     librte_port.so.3
328     librte_power.so.1
329     librte_reorder.so.1
330     librte_ring.so.1
331     librte_sched.so.1
332     librte_table.so.2
333     librte_timer.so.1
334     librte_vhost.so.3
335
336
337Tested Platforms
338----------------
339
340.. This section should contain a list of platforms that were tested with this release.
341
342   The format is:
343
344   #. Platform name.
345
346      * Platform details.
347      * Platform details.
348
349   This section is a comment. Make sure to start the actual text at the margin.
350
351#. SuperMicro 1U
352
353   - BIOS: 1.0c
354   - Processor: Intel(R) Atom(TM) CPU C2758 @ 2.40GHz
355
356#. SuperMicro 1U
357
358   - BIOS: 1.0a
359   - Processor: Intel(R) Xeon(R) CPU D-1540 @ 2.00GHz
360   - Onboard NIC: Intel(R) X552/X557-AT (2x10G)
361
362     - Firmware-version: 0x800001cf
363     - Device ID (PF/VF): 8086:15ad /8086:15a8
364
365   - kernel driver version: 4.2.5 (ixgbe)
366
367#. SuperMicro 2U
368
369   - BIOS: 1.0a
370   - Processor: Intel(R) Xeon(R) CPU E5-4667 v3 @ 2.00GHz
371
372#. Intel(R) Server board S2600GZ
373
374   - BIOS: SE5C600.86B.02.02.0002.122320131210
375   - Processor: Intel(R) Xeon(R) CPU E5-2680 v2 @ 2.80GHz
376
377#. Intel(R) Server board W2600CR
378
379   - BIOS: SE5C600.86B.02.01.0002.082220131453
380   - Processor: Intel(R) Xeon(R) CPU E5-2680 v2 @ 2.80GHz
381
382#. Intel(R) Server board S2600CWT
383
384   - BIOS: SE5C610.86B.01.01.0009.060120151350
385   - Processor: Intel(R) Xeon(R) CPU E5-2699 v3 @ 2.30GHz
386
387#. Intel(R) Server board S2600WTT
388
389   - BIOS: SE5C610.86B.01.01.0005.101720141054
390   - Processor: Intel(R) Xeon(R) CPU E5-2699 v3 @ 2.30GHz
391
392#. Intel(R) Server board S2600WTT
393
394   - BIOS: SE5C610.86B.11.01.0044.090120151156
395   - Processor: Intel(R) Xeon(R) CPU E5-2695 v4 @ 2.10GHz
396
397#. Intel(R) Server board S2600WTT
398
399   - Processor: Intel(R) Xeon(R) CPU E5-2697 v2 @ 2.70GHz
400
401#. Intel(R) Server
402
403   - Intel(R) Xeon(R) CPU E5-2697 v3 @ 2.60GHz
404
405#. IBM(R) Power8(R)
406
407   - Machine type-model: 8247-22L
408   - Firmware FW810.21 (SV810_108)
409   - Processor: POWER8E (raw), AltiVec supported
410
411
412Tested NICs
413-----------
414
415.. This section should contain a list of NICs that were tested with this release.
416
417   The format is:
418
419   #. NIC name.
420
421      * NIC details.
422      * NIC details.
423
424   This section is a comment. Make sure to start the actual text at the margin.
425
426#. Intel(R) Ethernet Controller X540-AT2
427
428   - Firmware version: 0x80000389
429   - Device id (pf): 8086:1528
430   - Driver version: 3.23.2 (ixgbe)
431
432#. Intel(R) 82599ES 10 Gigabit Ethernet Controller
433
434   - Firmware version: 0x61bf0001
435   - Device id (pf/vf): 8086:10fb / 8086:10ed
436   - Driver version: 4.0.1-k (ixgbe)
437
438#. Intel(R) Corporation Ethernet Connection X552/X557-AT 10GBASE-T
439
440   - Firmware version: 0x800001cf
441   - Device id (pf/vf): 8086:15ad / 8086:15a8
442   - Driver version: 4.2.5 (ixgbe)
443
444#. Intel(R) Ethernet Converged Network Adapter X710-DA4 (4x10G)
445
446   - Firmware version: 5.05
447   - Device id (pf/vf): 8086:1572 / 8086:154c
448   - Driver version: 1.5.23 (i40e)
449
450#. Intel(R) Ethernet Converged Network Adapter X710-DA2 (2x10G)
451
452   - Firmware version: 5.05
453   - Device id (pf/vf): 8086:1572 / 8086:154c
454   - Driver version: 1.5.23 (i40e)
455
456#. Intel(R) Ethernet Converged Network Adapter XL710-QDA1 (1x40G)
457
458   - Firmware version: 5.05
459   - Device id (pf/vf): 8086:1584 / 8086:154c
460   - Driver version: 1.5.23 (i40e)
461
462#. Intel(R) Ethernet Converged Network Adapter XL710-QDA2 (2X40G)
463
464   - Firmware version: 5.05
465   - Device id (pf/vf): 8086:1583 / 8086:154c
466   - Driver version: 1.5.23 (i40e)
467
468#. Intel(R) Corporation I350 Gigabit Network Connection
469
470   - Firmware version: 1.48, 0x800006e7
471   - Device id (pf/vf): 8086:1521 / 8086:1520
472   - Driver version: 5.2.13-k (igb)
473
474#. Intel(R) Ethernet Multi-host Controller FM10000
475
476   - Firmware version: N/A
477   - Device id (pf/vf): 8086:15d0
478   - Driver version: 0.17.0.9 (fm10k)
479
480#. Mellanox(R) ConnectX(R)-4 10G MCX4111A-XCAT (1x10G)
481
482   * Host interface: PCI Express 3.0 x8
483   * Device ID: 15b3:1013
484   * MLNX_OFED: 3.4-1.0.0.0
485   * Firmware version: 12.17.1010
486
487#. Mellanox(R) ConnectX(R)-4 10G MCX4121A-XCAT (2x10G)
488
489   * Host interface: PCI Express 3.0 x8
490   * Device ID: 15b3:1013
491   * MLNX_OFED: 3.4-1.0.0.0
492   * Firmware version: 12.17.1010
493
494#. Mellanox(R) ConnectX(R)-4 25G MCX4111A-ACAT (1x25G)
495
496   * Host interface: PCI Express 3.0 x8
497   * Device ID: 15b3:1013
498   * MLNX_OFED: 3.4-1.0.0.0
499   * Firmware version: 12.17.1010
500
501#. Mellanox(R) ConnectX(R)-4 25G MCX4121A-ACAT (2x25G)
502
503   * Host interface: PCI Express 3.0 x8
504   * Device ID: 15b3:1013
505   * MLNX_OFED: 3.4-1.0.0.0
506   * Firmware version: 12.17.1010
507
508#. Mellanox(R) ConnectX(R)-4 40G MCX4131A-BCAT/MCX413A-BCAT (1x40G)
509
510   * Host interface: PCI Express 3.0 x8
511   * Device ID: 15b3:1013
512   * MLNX_OFED: 3.4-1.0.0.0
513   * Firmware version: 12.17.1010
514
515#. Mellanox(R) ConnectX(R)-4 40G MCX415A-BCAT (1x40G)
516
517   * Host interface: PCI Express 3.0 x16
518   * Device ID: 15b3:1013
519   * MLNX_OFED: 3.4-1.0.0.0
520   * Firmware version: 12.17.1010
521
522#. Mellanox(R) ConnectX(R)-4 50G MCX4131A-GCAT/MCX413A-GCAT (1x50G)
523
524   * Host interface: PCI Express 3.0 x8
525   * Device ID: 15b3:1013
526   * MLNX_OFED: 3.4-1.0.0.0
527   * Firmware version: 12.17.1010
528
529#. Mellanox(R) ConnectX(R)-4 50G MCX414A-BCAT (2x50G)
530
531   * Host interface: PCI Express 3.0 x8
532   * Device ID: 15b3:1013
533   * MLNX_OFED: 3.4-1.0.0.0
534   * Firmware version: 12.17.1010
535
536#. Mellanox(R) ConnectX(R)-4 50G MCX415A-GCAT/MCX416A-BCAT/MCX416A-GCAT (2x50G)
537
538   * Host interface: PCI Express 3.0 x16
539   * Device ID: 15b3:1013
540   * MLNX_OFED: 3.4-1.0.0.0
541   * Firmware version: 12.17.1010
542
543#. Mellanox(R) ConnectX(R)-4 50G MCX415A-CCAT (1x100G)
544
545   * Host interface: PCI Express 3.0 x16
546   * Device ID: 15b3:1013
547   * MLNX_OFED: 3.4-1.0.0.0
548   * Firmware version: 12.17.1010
549
550#. Mellanox(R) ConnectX(R)-4 100G MCX416A-CCAT (2x100G)
551
552   * Host interface: PCI Express 3.0 x16
553   * Device ID: 15b3:1013
554   * MLNX_OFED: 3.4-1.0.0.0
555   * Firmware version: 12.17.1010
556
557#. Mellanox(R) ConnectX(R)-4 Lx 10G MCX4121A-XCAT (2x10G)
558
559   * Host interface: PCI Express 3.0 x8
560   * Device ID: 15b3:1015
561   * MLNX_OFED: 3.4-1.0.0.0
562   * Firmware version: 14.17.1010
563
564#. Mellanox(R) ConnectX(R)-4 Lx 25G MCX4121A-ACAT (2x25G)
565
566   * Host interface: PCI Express 3.0 x8
567   * Device ID: 15b3:1015
568   * MLNX_OFED: 3.4-1.0.0.0
569   * Firmware version: 14.17.1010
570
571
572Tested OSes
573-----------
574
575.. This section should contain a list of OSes that were tested with this release.
576   The format is as follows, in alphabetical order:
577
578   * CentOS 7.0
579   * Fedora 23
580   * Fedora 24
581   * FreeBSD 10.3
582   * Red Hat Enterprise Linux 7.2
583   * SUSE Enterprise Linux 12
584   * Ubuntu 15.10
585   * Ubuntu 16.04 LTS
586   * Wind River Linux 8
587
588   This section is a comment. Make sure to start the actual text at the margin.
589
590* CentOS 7.2
591* Fedora 23
592* Fedora 24
593* FreeBSD 10.3
594* FreeBSD 11
595* Red Hat Enterprise Linux Server release 6.7 (Santiago)
596* Red Hat Enterprise Linux Server release 7.0 (Maipo)
597* Red Hat Enterprise Linux Server release 7.2 (Maipo)
598* SUSE Enterprise Linux 12
599* Wind River Linux 6.0.0.26
600* Wind River Linux 8
601* Ubuntu 14.04
602* Ubuntu 15.04
603* Ubuntu 16.04
604