release_16_11.rst revision 6b3e017e
1DPDK Release 16.11
2==================
3
4.. **Read this first.**
5
6   The text below explains how to update the release notes.
7
8   Use proper spelling, capitalization and punctuation in all sections.
9
10   Variable and config names should be quoted as fixed width text: ``LIKE_THIS``.
11
12   Build the docs and view the output file to ensure the changes are correct::
13
14      make doc-guides-html
15
16      firefox build/doc/html/guides/rel_notes/release_16_11.html
17
18
19New Features
20------------
21
22.. This section should contain new features added in this release. Sample format:
23
24   * **Add a title in the past tense with a full stop.**
25
26     Add a short 1-2 sentence description in the past tense. The description
27     should be enough to allow someone scanning the release notes to understand
28     the new feature.
29
30     If the feature adds a lot of sub-features you can use a bullet list like this.
31
32     * Added feature foo to do something.
33     * Enhanced feature bar to do something else.
34
35     Refer to the previous release notes for examples.
36
37     This section is a comment. Make sure to start the actual text at the margin.
38
39
40* **Added software parser for packet type.**
41
42  * Added a new function ``rte_pktmbuf_read()`` to read the packet data from an
43    mbuf chain, linearizing if required.
44  * Added a new function ``rte_net_get_ptype()`` to parse an Ethernet packet
45    in an mbuf chain and retrieve its packet type from software.
46  * Added new functions ``rte_get_ptype_*()`` to dump a packet type as a string.
47
48* **Improved offloads support in mbuf.**
49
50  * Added a new function ``rte_raw_cksum_mbuf()`` to process the checksum of
51    data embedded in an mbuf chain.
52  * Added new Rx checksum flags in mbufs to describe more states: unknown,
53    good, bad, or not present (useful for virtual drivers). This modification
54    was done for IP and L4.
55  * Added a new Rx LRO mbuf flag, used when packets are coalesced. This
56    flag indicates that the segment size of original packets is known.
57
58* **Added vhost-user dequeue zero copy support.**
59
60  The copy in the dequeue path is avoided in order to improve the performance.
61  In the VM2VM case, the boost is quite impressive. The bigger the packet size,
62  the bigger performance boost you may get. However, for the VM2NIC case, there
63  are some limitations, so the boost is not as  impressive as the VM2VM case.
64  It may even drop quite a bit for small packets.
65
66  For that reason, this feature is disabled by default. It can be enabled when
67  the ``RTE_VHOST_USER_DEQUEUE_ZERO_COPY`` flag is set. Check the VHost section
68  of the Programming Guide for more information.
69
70* **Added vhost-user indirect descriptors support.**
71
72  If the indirect descriptor feature is enabled, each packet sent by the guest
73  will take exactly one slot in the enqueue virtqueue. Without this feature, as in
74  the current version, even 64 bytes packets take two slots with Virtio PMD on guest
75  side.
76
77  The main impact is better performance for 0% packet loss use-cases, as it
78  behaves as if the virtqueue size was enlarged, so more packets can be buffered
79  in the case of system perturbations. On the downside, small performance degradations
80  were measured when running micro-benchmarks.
81
82* **Added vhost PMD xstats.**
83
84  Added extended statistics to vhost PMD from a per port perspective.
85
86* **Supported offloads with virtio.**
87
88  Added support for the following offloads in virtio:
89
90  * Rx/Tx checksums.
91  * LRO.
92  * TSO.
93
94* **Added virtio NEON support for ARM.**
95
96  Added NEON support for ARM based virtio.
97
98* **Updated the ixgbe base driver.**
99
100  Updated the ixgbe base driver, including the following changes:
101
102  * Added X550em_a 10G PHY support.
103  * Added support for flow control auto negotiation for X550em_a 1G PHY.
104  * Added X550em_a FW ALEF support.
105  * Increased mailbox version to ``ixgbe_mbox_api_13``.
106  * Added two MAC operations for Hyper-V support.
107
108* **Added APIs for VF management to the ixgbe PMD.**
109
110  Eight new APIs have been added to the ixgbe PMD for VF management from the PF.
111  The declarations for the API's can be found in ``rte_pmd_ixgbe.h``.
112
113* **Updated the enic driver.**
114
115  * Added update to use interrupt for link status checking instead of polling.
116  * Added more flow director modes on UCS Blade with firmware version >= 2.0(13e).
117  * Added full support for MTU update.
118  * Added support for the ``rte_eth_rx_queue_count`` function.
119
120* **Updated the mlx5 driver.**
121
122  * Added support for RSS hash results.
123  * Added several performance improvements.
124  * Added several bug fixes.
125
126* **Updated the QAT PMD.**
127
128  The QAT PMD was updated with additional support for:
129
130  * MD5_HMAC algorithm.
131  * SHA224-HMAC algorithm.
132  * SHA384-HMAC algorithm.
133  * GMAC algorithm.
134  * KASUMI (F8 and F9) algorithm.
135  * 3DES algorithm.
136  * NULL algorithm.
137  * C3XXX device.
138  * C62XX device.
139
140* **Added openssl PMD.**
141
142  A new crypto PMD has been added, which provides several ciphering and hashing algorithms.
143  All cryptography operations use the Openssl library crypto API.
144
145* **Updated the IPsec example.**
146
147  Updated the IPsec example with the following support:
148
149  * Configuration file support.
150  * AES CBC IV generation with cipher forward function.
151  * AES GCM/CTR mode.
152
153* **Added support for new gcc -march option.**
154
155  The GCC 4.9 ``-march`` option supports the Intel processor code names.
156  The config option ``RTE_MACHINE`` can be used to pass code names to the compiler via the ``-march`` flag.
157
158
159Resolved Issues
160---------------
161
162.. This section should contain bug fixes added to the relevant sections. Sample format:
163
164   * **code/section Fixed issue in the past tense with a full stop.**
165
166     Add a short 1-2 sentence description of the resolved issue in the past tense.
167     The title should contain the code/lib section like a commit message.
168     Add the entries in alphabetic order in the relevant sections below.
169
170   This section is a comment. Make sure to start the actual text at the margin.
171
172
173Drivers
174~~~~~~~
175
176* **enic: Fixed several flow director issues.**
177
178* **enic: Fixed inadvertent setting of L4 checksum ptype on ICMP packets.**
179
180* **enic: Fixed high driver overhead when servicing Rx queues beyond the first.**
181
182
183
184Known Issues
185------------
186
187.. This section should contain new known issues in this release. Sample format:
188
189   * **Add title in present tense with full stop.**
190
191     Add a short 1-2 sentence description of the known issue in the present
192     tense. Add information on any known workarounds.
193
194   This section is a comment. Make sure to start the actual text at the margin.
195
196* **L3fwd-power app does not work properly when Rx vector is enabled.**
197
198  The L3fwd-power app doesn't work properly with some drivers in vector mode
199  since the queue monitoring works differently between scalar and vector modes
200  leading to incorrect frequency scaling. In addition, L3fwd-power application
201  requires the mbuf to have correct packet type set but in some drivers the
202  vector mode must be disabled for this.
203
204  Therefore, in order to use L3fwd-power, vector mode should be disabled
205  via the config file.
206
207* **Digest address must be supplied for crypto auth operation on QAT PMD.**
208
209  The cryptodev API specifies that if the rte_crypto_sym_op.digest.data field,
210  and by inference the digest.phys_addr field which points to the same location,
211  is not set for an auth operation the driver is to understand that the digest
212  result is located immediately following the region over which the digest is
213  computed. The QAT PMD doesn't correctly handle this case and reads and writes
214  to an incorrect location.
215
216  Callers can workaround this by always supplying the digest virtual and
217  physical address fields in the rte_crypto_sym_op for an auth operation.
218
219
220API Changes
221-----------
222
223.. This section should contain API changes. Sample format:
224
225   * Add a short 1-2 sentence description of the API change. Use fixed width
226     quotes for ``rte_function_names`` or ``rte_struct_names``. Use the past tense.
227
228   This section is a comment. Make sure to start the actual text at the margin.
229
230* The driver naming convention has been changed to make them more
231  consistent. It especially impacts ``--vdev`` arguments. For example
232  ``eth_pcap`` becomes ``net_pcap`` and ``cryptodev_aesni_mb_pmd`` becomes
233  ``crypto_aesni_mb``.
234
235  For backward compatibility an alias feature has been enabled to support the
236  original names.
237
238* The log history has been removed.
239
240* The ``rte_ivshmem`` feature (including library and EAL code) has been removed
241  in 16.11 because it had some design issues which were not planned to be fixed.
242
243* The ``file_name`` data type of ``struct rte_port_source_params`` and
244  ``struct rte_port_sink_params`` is changed from ``char *`` to ``const char *``.
245
246* **Improved device/driver hierarchy and generalized hotplugging.**
247
248  The device and driver relationship has been restructured by introducing generic
249  classes. This paves the way for having PCI, VDEV and other device types as
250  instantiated objects rather than classes in themselves. Hotplugging has also
251  been generalized into EAL so that Ethernet or crypto devices can use the
252  common infrastructure.
253
254  * Removed ``pmd_type`` as a way of segregation of devices.
255  * Moved ``numa_node`` and ``devargs`` into ``rte_driver`` from
256    ``rte_pci_driver``. These can now be used by any instantiated object of
257    ``rte_driver``.
258  * Added ``rte_device`` class and all PCI and VDEV devices inherit from it
259  * Renamed devinit/devuninit handlers to probe/remove to make it more
260    semantically correct with respect to the device <=> driver relationship.
261  * Moved hotplugging support to EAL. Hereafter, PCI and vdev can use the
262    APIs ``rte_eal_dev_attach`` and ``rte_eal_dev_detach``.
263  * Renamed helpers and support macros to make them more synonymous
264    with their device types
265    (e.g. ``PMD_REGISTER_DRIVER`` => ``RTE_PMD_REGISTER_PCI``).
266  * Device naming functions have been generalized from ethdev and cryptodev
267    to EAL. ``rte_eal_pci_device_name`` has been introduced for obtaining
268    unique device name from PCI Domain-BDF description.
269  * Virtual device registration APIs have been added: ``rte_eal_vdrv_register``
270    and ``rte_eal_vdrv_unregister``.
271
272
273ABI Changes
274-----------
275
276.. This section should contain ABI changes. Sample format:
277
278   * Add a short 1-2 sentence description of the ABI change that was announced in
279     the previous releases and made in this release. Use fixed width quotes for
280     ``rte_function_names`` or ``rte_struct_names``. Use the past tense.
281
282   This section is a comment. Make sure to start the actual text at the margin.
283
284
285
286Shared Library Versions
287-----------------------
288
289.. Update any library version updated in this release and prepend with a ``+``
290   sign, like this:
291
292     libethdev.so.4
293     librte_acl.so.2
294   + librte_cfgfile.so.2
295     librte_cmdline.so.2
296
297
298
299The libraries prepended with a plus sign were incremented in this version.
300
301.. code-block:: diff
302
303     librte_acl.so.2
304     librte_cfgfile.so.2
305     librte_cmdline.so.2
306   + librte_cryptodev.so.2
307     librte_distributor.so.1
308   + librte_eal.so.3
309   + librte_ethdev.so.5
310     librte_hash.so.2
311     librte_ip_frag.so.1
312     librte_jobstats.so.1
313     librte_kni.so.2
314     librte_kvargs.so.1
315     librte_lpm.so.2
316     librte_mbuf.so.2
317     librte_mempool.so.2
318     librte_meter.so.1
319     librte_net.so.1
320     librte_pdump.so.1
321     librte_pipeline.so.3
322     librte_pmd_bond.so.1
323     librte_pmd_ring.so.2
324     librte_port.so.3
325     librte_power.so.1
326     librte_reorder.so.1
327     librte_ring.so.1
328     librte_sched.so.1
329     librte_table.so.2
330     librte_timer.so.1
331     librte_vhost.so.3
332
333
334Tested Platforms
335----------------
336
337.. This section should contain a list of platforms that were tested with this release.
338
339   The format is:
340
341   #. Platform name.
342
343      * Platform details.
344      * Platform details.
345
346   This section is a comment. Make sure to start the actual text at the margin.
347
348#. SuperMicro 1U
349
350   - BIOS: 1.0c
351   - Processor: Intel(R) Atom(TM) CPU C2758 @ 2.40GHz
352
353#. SuperMicro 1U
354
355   - BIOS: 1.0a
356   - Processor: Intel(R) Xeon(R) CPU D-1540 @ 2.00GHz
357   - Onboard NIC: Intel(R) X552/X557-AT (2x10G)
358
359     - Firmware-version: 0x800001cf
360     - Device ID (PF/VF): 8086:15ad /8086:15a8
361
362   - kernel driver version: 4.2.5 (ixgbe)
363
364#. SuperMicro 2U
365
366   - BIOS: 1.0a
367   - Processor: Intel(R) Xeon(R) CPU E5-4667 v3 @ 2.00GHz
368
369#. Intel(R) Server board S2600GZ
370
371   - BIOS: SE5C600.86B.02.02.0002.122320131210
372   - Processor: Intel(R) Xeon(R) CPU E5-2680 v2 @ 2.80GHz
373
374#. Intel(R) Server board W2600CR
375
376   - BIOS: SE5C600.86B.02.01.0002.082220131453
377   - Processor: Intel(R) Xeon(R) CPU E5-2680 v2 @ 2.80GHz
378
379#. Intel(R) Server board S2600CWT
380
381   - BIOS: SE5C610.86B.01.01.0009.060120151350
382   - Processor: Intel(R) Xeon(R) CPU E5-2699 v3 @ 2.30GHz
383
384#. Intel(R) Server board S2600WTT
385
386   - BIOS: SE5C610.86B.01.01.0005.101720141054
387   - Processor: Intel(R) Xeon(R) CPU E5-2699 v3 @ 2.30GHz
388
389#. Intel(R) Server board S2600WTT
390
391   - BIOS: SE5C610.86B.11.01.0044.090120151156
392   - Processor: Intel(R) Xeon(R) CPU E5-2695 v4 @ 2.10GHz
393
394#. Intel(R) Server board S2600WTT
395
396   - Processor: Intel(R) Xeon(R) CPU E5-2697 v2 @ 2.70GHz
397
398#. Intel(R) Server
399
400   - Intel(R) Xeon(R) CPU E5-2697 v3 @ 2.60GHz
401
402#. IBM(R) Power8(R)
403
404   - Machine type-model: 8247-22L
405   - Firmware FW810.21 (SV810_108)
406   - Processor: POWER8E (raw), AltiVec supported
407
408
409Tested NICs
410-----------
411
412.. This section should contain a list of NICs that were tested with this release.
413
414   The format is:
415
416   #. NIC name.
417
418      * NIC details.
419      * NIC details.
420
421   This section is a comment. Make sure to start the actual text at the margin.
422
423#. Intel(R) Ethernet Controller X540-AT2
424
425   - Firmware version: 0x80000389
426   - Device id (pf): 8086:1528
427   - Driver version: 3.23.2 (ixgbe)
428
429#. Intel(R) 82599ES 10 Gigabit Ethernet Controller
430
431   - Firmware version: 0x61bf0001
432   - Device id (pf/vf): 8086:10fb / 8086:10ed
433   - Driver version: 4.0.1-k (ixgbe)
434
435#. Intel(R) Corporation Ethernet Connection X552/X557-AT 10GBASE-T
436
437   - Firmware version: 0x800001cf
438   - Device id (pf/vf): 8086:15ad / 8086:15a8
439   - Driver version: 4.2.5 (ixgbe)
440
441#. Intel(R) Ethernet Converged Network Adapter X710-DA4 (4x10G)
442
443   - Firmware version: 5.05
444   - Device id (pf/vf): 8086:1572 / 8086:154c
445   - Driver version: 1.5.23 (i40e)
446
447#. Intel(R) Ethernet Converged Network Adapter X710-DA2 (2x10G)
448
449   - Firmware version: 5.05
450   - Device id (pf/vf): 8086:1572 / 8086:154c
451   - Driver version: 1.5.23 (i40e)
452
453#. Intel(R) Ethernet Converged Network Adapter XL710-QDA1 (1x40G)
454
455   - Firmware version: 5.05
456   - Device id (pf/vf): 8086:1584 / 8086:154c
457   - Driver version: 1.5.23 (i40e)
458
459#. Intel(R) Ethernet Converged Network Adapter XL710-QDA2 (2X40G)
460
461   - Firmware version: 5.05
462   - Device id (pf/vf): 8086:1583 / 8086:154c
463   - Driver version: 1.5.23 (i40e)
464
465#. Intel(R) Corporation I350 Gigabit Network Connection
466
467   - Firmware version: 1.48, 0x800006e7
468   - Device id (pf/vf): 8086:1521 / 8086:1520
469   - Driver version: 5.2.13-k (igb)
470
471#. Intel(R) Ethernet Multi-host Controller FM10000
472
473   - Firmware version: N/A
474   - Device id (pf/vf): 8086:15d0
475   - Driver version: 0.17.0.9 (fm10k)
476
477#. Mellanox(R) ConnectX(R)-4 10G MCX4111A-XCAT (1x10G)
478
479   * Host interface: PCI Express 3.0 x8
480   * Device ID: 15b3:1013
481   * MLNX_OFED: 3.4-1.0.0.0
482   * Firmware version: 12.17.1010
483
484#. Mellanox(R) ConnectX(R)-4 10G MCX4121A-XCAT (2x10G)
485
486   * Host interface: PCI Express 3.0 x8
487   * Device ID: 15b3:1013
488   * MLNX_OFED: 3.4-1.0.0.0
489   * Firmware version: 12.17.1010
490
491#. Mellanox(R) ConnectX(R)-4 25G MCX4111A-ACAT (1x25G)
492
493   * Host interface: PCI Express 3.0 x8
494   * Device ID: 15b3:1013
495   * MLNX_OFED: 3.4-1.0.0.0
496   * Firmware version: 12.17.1010
497
498#. Mellanox(R) ConnectX(R)-4 25G MCX4121A-ACAT (2x25G)
499
500   * Host interface: PCI Express 3.0 x8
501   * Device ID: 15b3:1013
502   * MLNX_OFED: 3.4-1.0.0.0
503   * Firmware version: 12.17.1010
504
505#. Mellanox(R) ConnectX(R)-4 40G MCX4131A-BCAT/MCX413A-BCAT (1x40G)
506
507   * Host interface: PCI Express 3.0 x8
508   * Device ID: 15b3:1013
509   * MLNX_OFED: 3.4-1.0.0.0
510   * Firmware version: 12.17.1010
511
512#. Mellanox(R) ConnectX(R)-4 40G MCX415A-BCAT (1x40G)
513
514   * Host interface: PCI Express 3.0 x16
515   * Device ID: 15b3:1013
516   * MLNX_OFED: 3.4-1.0.0.0
517   * Firmware version: 12.17.1010
518
519#. Mellanox(R) ConnectX(R)-4 50G MCX4131A-GCAT/MCX413A-GCAT (1x50G)
520
521   * Host interface: PCI Express 3.0 x8
522   * Device ID: 15b3:1013
523   * MLNX_OFED: 3.4-1.0.0.0
524   * Firmware version: 12.17.1010
525
526#. Mellanox(R) ConnectX(R)-4 50G MCX414A-BCAT (2x50G)
527
528   * Host interface: PCI Express 3.0 x8
529   * Device ID: 15b3:1013
530   * MLNX_OFED: 3.4-1.0.0.0
531   * Firmware version: 12.17.1010
532
533#. Mellanox(R) ConnectX(R)-4 50G MCX415A-GCAT/MCX416A-BCAT/MCX416A-GCAT (2x50G)
534
535   * Host interface: PCI Express 3.0 x16
536   * Device ID: 15b3:1013
537   * MLNX_OFED: 3.4-1.0.0.0
538   * Firmware version: 12.17.1010
539
540#. Mellanox(R) ConnectX(R)-4 50G MCX415A-CCAT (1x100G)
541
542   * Host interface: PCI Express 3.0 x16
543   * Device ID: 15b3:1013
544   * MLNX_OFED: 3.4-1.0.0.0
545   * Firmware version: 12.17.1010
546
547#. Mellanox(R) ConnectX(R)-4 100G MCX416A-CCAT (2x100G)
548
549   * Host interface: PCI Express 3.0 x16
550   * Device ID: 15b3:1013
551   * MLNX_OFED: 3.4-1.0.0.0
552   * Firmware version: 12.17.1010
553
554#. Mellanox(R) ConnectX(R)-4 Lx 10G MCX4121A-XCAT (2x10G)
555
556   * Host interface: PCI Express 3.0 x8
557   * Device ID: 15b3:1015
558   * MLNX_OFED: 3.4-1.0.0.0
559   * Firmware version: 14.17.1010
560
561#. Mellanox(R) ConnectX(R)-4 Lx 25G MCX4121A-ACAT (2x25G)
562
563   * Host interface: PCI Express 3.0 x8
564   * Device ID: 15b3:1015
565   * MLNX_OFED: 3.4-1.0.0.0
566   * Firmware version: 14.17.1010
567
568
569Tested OSes
570-----------
571
572.. This section should contain a list of OSes that were tested with this release.
573   The format is as follows, in alphabetical order:
574
575   * CentOS 7.0
576   * Fedora 23
577   * Fedora 24
578   * FreeBSD 10.3
579   * Red Hat Enterprise Linux 7.2
580   * SUSE Enterprise Linux 12
581   * Ubuntu 15.10
582   * Ubuntu 16.04 LTS
583   * Wind River Linux 8
584
585   This section is a comment. Make sure to start the actual text at the margin.
586
587* CentOS 7.2
588* Fedora 23
589* Fedora 24
590* FreeBSD 10.3
591* FreeBSD 11
592* Red Hat Enterprise Linux Server release 6.7 (Santiago)
593* Red Hat Enterprise Linux Server release 7.0 (Maipo)
594* Red Hat Enterprise Linux Server release 7.2 (Maipo)
595* SUSE Enterprise Linux 12
596* Wind River Linux 6.0.0.26
597* Wind River Linux 8
598* Ubuntu 14.04
599* Ubuntu 15.04
600* Ubuntu 16.04
601