1 .. SPDX-License-Identifier: BSD-3-Clause
2 Copyright 2019 The DPDK contributors
4 .. include:: <isonum.txt>
9 .. **Read this first.**
11 The text in the sections below explains how to update the release notes.
13 Use proper spelling, capitalization and punctuation in all sections.
15 Variable and config names should be quoted as fixed width text:
18 Build the docs and view the output file to ensure the changes are correct::
22 xdg-open build/doc/html/guides/rel_notes/release_19_11.html
28 .. This section should contain new features added in this release.
31 * **Add a title in the past tense with a full stop.**
33 Add a short 1-2 sentence description in the past tense.
34 The description should be enough to allow someone scanning
35 the release notes to understand the new feature.
37 If the feature adds a lot of sub-features you can use a bullet list
40 * Added feature foo to do something.
41 * Enhanced feature bar to do something else.
43 Refer to the previous release notes for examples.
45 Suggested order in release notes items:
46 * Core libs (EAL, mempool, ring, mbuf, buses)
47 * Device abstraction libs and PMDs
49 - cryptodev (lib, PMDs)
50 - eventdev (lib, PMDs)
53 * Apps, Examples, Tools (if significant)
55 This section is a comment. Do not overwrite or remove it.
56 Also, make sure to start the actual text at the margin.
57 =========================================================
59 * **FreeBSD now supports `--base-virtaddr` EAL option.**
61 FreeBSD version now also supports setting base virtual address for mapping
62 pages and resources into its address space.
64 * **Added Lock-free Stack for aarch64.**
66 The lock-free stack implementation is enabled for aarch64 platforms.
68 * **Changed mempool allocation behaviour.**
70 Objects are no longer across pages by default.
71 It may consume more memory when using small memory pages.
73 * **Added support of dynamic fields and flags in mbuf.**
75 This new feature adds the ability to dynamically register some room
76 for a field or a flag in the mbuf structure. This is typically used
77 for specific offload features, where adding a static field or flag
78 in the mbuf is not justified.
80 * **Added hairpin queue.**
82 On supported NICs, we can now setup haipin queue which will offload packets
83 from the wire, backto the wire.
85 * **Added flow tag in rte_flow.**
87 SET_TAG action and TAG item have been added to support transient flow tag.
89 * **Extended metadata support in rte_flow.**
91 Flow metadata is extended to both Rx and Tx.
93 * Tx metadata can also be set by SET_META action of rte_flow.
94 * Rx metadata is delivered to host via a dynamic field of ``rte_mbuf`` with
97 * **Added ethdev API to set supported packet types**
99 * Added new API ``rte_eth_dev_set_ptypes`` that allows an application to
100 inform PMD about reduced range of packet types to handle.
101 * This scheme will allow PMDs to avoid lookup to internal ptype table on Rx
102 and thereby improve Rx performance if application wishes do so.
104 * **Added Rx offload flag to enable or disable RSS update**
106 * Added new Rx offload flag `DEV_RX_OFFLOAD_RSS_HASH` which can be used to
107 enable/disable PMDs write to `rte_mbuf::hash::rss`.
108 * PMDs notify the validity of `rte_mbuf::hash:rss` to the application
109 by enabling `PKT_RX_RSS_HASH ` flag in `rte_mbuf::ol_flags`.
111 * **Updated the enic driver.**
113 * Added support for Geneve with options offload.
114 * Added flow API implementation based on VIC Flow Manager API.
116 * **Added Hisilicon hns3 PMD.**
118 Added the new ``hns3`` net driver for the inbuilt Hisilicon Network
119 Subsystem 3(HNS3) network engine found in the Hisilicon Kunpeng 920 SoC.
120 See the :doc:`../nics/hns3` guide for more details on this new driver.
122 * **Added NXP PFE PMD.**
124 Added the new PFE driver for the NXP LS1012A platform. See the
125 :doc:`../nics/pfe` NIC driver guide for more details on this new driver.
127 * **Updated iavf PMD.**
129 Enable AVX2 data path for iavf PMD.
131 * **Updated the Intel e1000 driver.**
133 Added support for the ``RTE_ETH_DEV_CLOSE_REMOVE`` flag.
135 * **Updated the Intel ixgbe driver.**
137 Added support for the ``RTE_ETH_DEV_CLOSE_REMOVE`` flag.
139 * **Updated the Intel i40e driver.**
141 Added support for the ``RTE_ETH_DEV_CLOSE_REMOVE`` flag.
143 * **Updated the Intel fm10k driver.**
145 Added support for the ``RTE_ETH_DEV_CLOSE_REMOVE`` flag.
147 * **Added RX/TX packet burst mode get API.**
149 Added two new functions ``rte_eth_rx_burst_mode_get`` and
150 ``rte_eth_tx_burst_mode_get`` that allow an application
151 to retrieve the mode information about RX/TX packet burst
152 such as Scalar or Vector, and Vector technology like AVX2.
154 * **Updated the Intel ice driver.**
156 Updated the Intel ice driver with new features and improvements, including:
158 * Added support for device-specific DDP package loading.
159 * Added support for handling Receive Flex Descriptor.
160 * Added support for protocol extraction on per Rx queue.
161 * Added support for Flow Director filter based on generic filter framework.
162 * Added support for the ``RTE_ETH_DEV_CLOSE_REMOVE`` flag.
163 * Generic filter enhancement
164 - Supported pipeline mode.
165 - Supported new packet type like PPPoE for switch filter.
166 * Supported input set change and symmetric hash by rte_flow RSS action.
167 * Added support for GTP Tx checksum offload.
168 * Added new device IDs to support E810_XXV devices.
170 * **Added cryptodev asymmetric session-less operation.**
172 Added session-less option to cryptodev asymmetric structure. It works the same
173 way as symmetric crypto, corresponding xform is used directly by the crypto op.
175 * **Updated the Huawei hinic driver.**
177 Updated the Huawei hinic driver with new features and improvements, including:
179 * Enabled SR-IOV - Partially supported at this point, VFIO only.
180 * Supported VLAN filter and VLAN offload.
181 * Supported Unicast MAC filter and Multicast MAC filter.
182 * Supported Flow API for LACP, VRRP, BGP and so on.
183 * Supported FW version get.
185 * **Updated Mellanox mlx5 driver.**
187 Updated Mellanox mlx5 driver with new features and improvements, including:
189 * Added support for VLAN pop flow offload command.
190 * Added support for VLAN push flow offload command.
191 * Added support for VLAN set PCP offload command.
192 * Added support for VLAN set VID offload command.
193 * Added support for matching on packets withe Geneve tunnel header.
194 * Added hairpin support.
195 * Added ConnectX6-DX support.
197 * **Updated the AF_XDP PMD.**
199 Updated the AF_XDP PMD. The new features include:
201 * Enabled zero copy between application mempools and UMEM by enabling the
202 XDP_UMEM_UNALIGNED_CHUNKS UMEM flag.
204 * **Added Marvell NITROX symmetric crypto PMD.**
206 Added a symmetric crypto PMD for Marvell NITROX V security processor.
207 See the :doc:`../cryptodevs/nitrox` guide for more details on this new
209 * **Added asymmetric support to Marvell OCTEON TX crypto PMD.**
211 Added support for asymmetric operations in Marvell OCTEON TX cypto PMD.
212 Supports RSA and modexp operations.
214 * **Added Marvell OCTEON TX2 crypto PMD**
216 Added a new PMD driver for h/w crypto offload block on ``OCTEON TX2`` SoC.
218 See :doc:`../cryptodevs/octeontx2` for more details
220 * **Updated NXP crypto PMDs for PDCP support.**
222 PDCP support is added to DPAA_SEC and DPAA2_SEC PMDs using rte_security APIs.
223 Support is added for all sequence number sizes for control and user plane.
224 Test and test-crypto-perf applications are updated for unit testing.
226 * **Updated the AESNI-MB PMD.**
228 * Added support for intel-ipsec-mb version 0.53.
230 * **Updated the AESNI-GCM PMD.**
232 * Added support for intel-ipsec-mb version 0.53.
233 * Supported in-place chained mbufs on AES-GCM algorithm.
235 * **Enabled Single Pass GCM acceleration on QAT GEN3.**
237 Added support for Single Pass GCM, available on QAT GEN3 only (Intel
238 QuickAssist Technology C4xxx). It is automatically chosen instead of the
239 classic 2-pass mode when running on QAT GEN3, significantly improving
240 the performance of AES GCM operations.
242 * **Updated the Intel QuickAssist Technology (QAT) asymmetric crypto PMD.**
244 * Added support for asymmetric session-less operations.
245 * Added support for RSA algorithm with pair (n, d) private key representation.
246 * Added support for RSA algorithm with quintuple private key representation.
248 * **Updated the Intel QuickAssist Technology (QAT) compression PMD.**
250 Added stateful decompression support in the Intel QuickAssist Technology PMD.
251 Please note that stateful compression is not supported.
253 * **Added external buffers support for dpdk-test-compress-perf tool.**
255 Added a command line option to dpdk-test-compress-perf tool to allocate
256 and use memory zones as external buffers instead of keeping the data directly
259 * **Updated the IPSec library.**
261 * Added SA Database API to ``librte_ipsec``. A new test-sad application is also
262 introduced to evaluate and perform custom functional and performance tests
263 for IPsec SAD implementation.
265 * Support fragmented packets in inline crypto processing mode with fallback
266 ``lookaside-none`` session. Corresponding changes are also added in IPsec
267 Security Gateway application.
269 * **Introduced FIFO for NTB PMD.**
271 Introduced FIFO for NTB (Non-transparent Bridge) PMD to support
272 packet based processing.
274 * **Added eBPF JIT support for arm64.**
276 Added eBPF JIT support for arm64 architecture to improve the eBPF program
279 * **Added RIB and FIB (Routing/Forwarding Information Base) libraries.**
281 RIB and FIB can replace the LPM (Longest Prefix Match) library
282 with better control plane (RIB) performance.
283 The data plane (FIB) can be extended with new algorithms.
285 * **Updated testpmd.**
287 * Added a console command to testpmd app, ``show port (port_id) ptypes`` which
288 gives ability to print port supported ptypes in different protocol layers.
289 * Packet type detection disabled by default for the supported PMDs.
291 * **Added new example l2fwd-event application.**
293 Added an example application `l2fwd-event` that adds event device support to
294 traditional l2fwd example. It demonstrates usage of poll and event mode IO
295 mechanism under a single application.
297 * **Added build support for Link Time Optimization.**
299 LTO is an optimization technique used by the compiler to perform whole
300 program analysis and optimization at link time. In order to do that
301 compilers store their internal representation of the source code that
302 the linker uses at the final stage of compilation process.
304 See :doc:`../prog_guide/lto` for more information:
306 * **Added IOVA as VA support for KNI.**
308 * Added IOVA = VA support for KNI, KNI can operate in IOVA = VA mode when
309 `iova-mode=va` EAL option is passed to the application or when bus IOVA
310 scheme is selected as RTE_IOVA_VA. This mode only works on Linux Kernel
311 versions 4.6.0 and above.
313 * Due to IOVA to KVA address translations, based on the KNI use case there
314 can be a performance impact. For mitigation, forcing IOVA to PA via EAL
315 "--iova-mode=pa" option can be used, IOVA_DC bus iommu scheme can also
316 result in IOVA as PA.
322 .. This section should contain removed items in this release. Sample format:
324 * Add a short 1-2 sentence description of the removed item
327 This section is a comment. Do not overwrite or remove it.
328 Also, make sure to start the actual text at the margin.
329 =========================================================
331 * Removed duplicated set of commands for Rx offload configuration from testpmd::
333 port config all crc-strip|scatter|rx-cksum|rx-timestamp|
334 hw-vlan|hw-vlan-filter|hw-vlan-strip|hw-vlan-extend on|off
336 The testpmd commands set that can be used instead
337 in order to enable or disable Rx offloading on all Rx queues of a port is::
339 port config <port_id> rx_offload crc_strip|scatter|
340 ipv4_cksum|udp_cksum|tcp_cksum|timestamp|
341 vlan_strip|vlan_filter|vlan_extend on|off
343 * Removed AF_XDP pmd_zero copy vdev argument. Support is now auto-detected.
345 * The following sample applications have been removed in this release:
348 * L3 Forwarding in a Virtualization Environment
350 * Netmap Compatibility
351 * Quota and Watermark
354 * Removed arm64-dpaa2-* build config. arm64-dpaa-* can now build for both
355 dpaa and dpaa2 platforms.
361 .. This section should contain API changes. Sample format:
363 * sample: Add a short 1-2 sentence description of the API change
364 which was announced in the previous releases and made in this release.
365 Start with a scope label like "ethdev:".
366 Use fixed width quotes for ``function_names`` or ``struct_names``.
369 This section is a comment. Do not overwrite or remove it.
370 Also, make sure to start the actual text at the margin.
371 =========================================================
373 * eal: made the ``lcore_config`` struct and global symbol private.
375 * eal: removed the ``rte_cpu_check_supported`` function, replaced by
376 ``rte_cpu_is_supported`` since dpdk v17.08.
378 * eal: removed the ``rte_malloc_virt2phy`` function, replaced by
379 ``rte_malloc_virt2iova`` since v17.11.
381 * eal: made the ``rte_config`` struct and ``rte_eal_get_configuration``
384 * mem: hid the internal ``malloc_heap`` structure and the
385 ``rte_malloc_heap.h`` header.
387 * vfio: removed ``rte_vfio_dma_map`` and ``rte_vfio_dma_unmap`` that have
388 been marked as deprecated in release 19.05.
389 ``rte_vfio_container_dma_map`` and ``rte_vfio_container_dma_unmap`` can
390 be used as substitutes.
392 * pci: removed the following functions deprecated since dpdk v17.11:
394 - ``eal_parse_pci_BDF`` replaced by ``rte_pci_addr_parse``
395 - ``eal_parse_pci_DomBDF`` replaced by ``rte_pci_addr_parse``
396 - ``rte_eal_compare_pci_addr`` replaced by ``rte_pci_addr_cmp``
398 * The network structure ``esp_tail`` has been prefixed by ``rte_``.
400 * The network definitions of PPPoE ethertypes have been prefixed by ``RTE_``.
402 * The network structure for MPLS has been prefixed by ``rte_``.
404 * ethdev: changed ``rte_eth_dev_infos_get`` return value from ``void`` to
405 ``int`` to provide a way to report various error conditions.
407 * ethdev: changed ``rte_eth_promiscuous_enable`` and
408 ``rte_eth_promiscuous_disable`` return value from ``void`` to ``int`` to
409 provide a way to report various error conditions.
411 * ethdev: changed ``rte_eth_allmulticast_enable`` and
412 ``rte_eth_allmulticast_disable`` return value from ``void`` to ``int`` to
413 provide a way to report various error conditions.
415 * ethdev: changed ``rte_eth_dev_xstats_reset`` return value from ``void`` to
416 ``int`` to provide a way to report various error conditions.
418 * ethdev: changed ``rte_eth_link_get`` and ``rte_eth_link_get_nowait``
419 return value from ``void`` to ``int`` to provide a way to report various
422 * ethdev: changed ``rte_eth_macaddr_get`` return value from ``void`` to
423 ``int`` to provide a way to report various error conditions.
425 * ethdev: changed ``rte_eth_dev_owner_delete`` return value from ``void`` to
426 ``int`` to provide a way to report various error conditions.
428 * ethdev: The deprecated function ``rte_eth_dev_count`` was removed.
429 The function ``rte_eth_dev_count_avail`` is a drop-in replacement.
430 If the intent is to iterate over ports, ``RTE_ETH_FOREACH_*`` macros
431 are better port iterators.
433 * ethdev: RTE_FLOW_ITEM_TYPE_META data endianness altered to host one.
434 Due to the new dynamic metadata field in mbuf is host-endian either, there
435 is the minor compatibility issue for applications in case of 32-bit values
438 * ethdev: the tx_metadata mbuf field is moved to dymanic one.
439 PKT_TX_METADATA flag is replaced with PKT_TX_DYNF_METADATA.
440 DEV_TX_OFFLOAD_MATCH_METADATA offload flag is removed, now metadata
441 support in PMD is engaged on dynamic field registration.
443 * event: The function ``rte_event_eth_tx_adapter_enqueue`` takes an additional
444 input as ``flags``. Flag ``RTE_EVENT_ETH_TX_ADAPTER_ENQUEUE_SAME_DEST`` which
445 has been introduced in this release is used when used when all the packets
446 enqueued in the tx adapter are destined for the same Ethernet port & Tx queue.
448 * sched: The pipe nodes configuration parameters such as number of pipes,
449 pipe queue sizes, pipe profiles, etc., are moved from port level structure
450 to subport level. This allows different subports of the same port to
451 have different configuration for the pipe nodes.
457 .. This section should contain ABI changes. Sample format:
459 * sample: Add a short 1-2 sentence description of the ABI change
460 which was announced in the previous releases and made in this release.
461 Start with a scope label like "ethdev:".
462 Use fixed width quotes for ``function_names`` or ``struct_names``.
465 This section is a comment. Do not overwrite or remove it.
466 Also, make sure to start the actual text at the margin.
467 =========================================================
469 * net: The Ethernet address and other header definitions have changed
470 attributes. They have been modified to be aligned on 2-byte boundaries.
471 These changes should not impact normal usage because drivers naturally
472 align the Ethernet header on receive and all known encapsulations
473 preserve the alignment of the header.
475 * security: The field ``replay_win_sz`` has been moved from ipsec library
476 based ``rte_ipsec_sa_prm`` structure to security library based structure
477 ``rte_security_ipsec_xform``, which specify the Anti replay window size
478 to enable sequence replay attack handling.
480 * ipsec: The field ``replay_win_sz`` has been removed from the structure
481 ``rte_ipsec_sa_prm`` as it has been added to the security library.
483 * ethdev: Added 32-bit fields for maximum LRO aggregated packet size, in
484 struct ``rte_eth_dev_info`` for the port capability and in struct
485 ``rte_eth_rxmode`` for the port configuration.
486 Application should use the new field in struct ``rte_eth_rxmode`` to configure
488 PMD should use the new field in struct ``rte_eth_dev_info`` to report the
489 supported port capability.
492 Shared Library Versions
493 -----------------------
495 .. Update any library version updated in this release
496 and prepend with a ``+`` sign, like this:
502 This section is a comment. Do not overwrite or remove it.
503 =========================================================
505 The libraries prepended with a plus sign were incremented in this version.
511 librte_bitratestats.so.2
514 librte_bus_fslmc.so.2
515 librte_bus_ifpga.so.2
518 librte_bus_vmbus.so.2
521 librte_compressdev.so.1
522 librte_cryptodev.so.8
523 librte_distributor.so.1
526 + librte_ethdev.so.13
527 + librte_eventdev.so.8
529 librte_flow_classify.so.1
538 librte_latencystats.so.1
552 librte_pmd_ixgbe.so.2
553 librte_pmd_dpaa2_qdma.so.1
555 librte_pmd_softnic.so.1
556 librte_pmd_vhost.so.2
565 + librte_security.so.3
575 .. This section should contain new known issues in this release. Sample format:
577 * **Add title in present tense with full stop.**
579 Add a short 1-2 sentence description of the known issue
580 in the present tense. Add information on any known workarounds.
582 This section is a comment. Do not overwrite or remove it.
583 Also, make sure to start the actual text at the margin.
584 =========================================================
590 .. This section should contain a list of platforms that were tested
595 * <vendor> platform with <vendor> <type of devices> combinations
600 * Other relevant details...
602 This section is a comment. Do not overwrite or remove it.
603 Also, make sure to start the actual text at the margin.
604 =========================================================