1 .. SPDX-License-Identifier: BSD-3-Clause
2 Copyright 2018 The DPDK contributors
4 ABI and API Deprecation
5 =======================
7 See the guidelines document for details of the :doc:`ABI policy
8 <../contributing/abi_policy>`. API and ABI deprecation notices are to be posted
14 * meson: The minimum supported version of meson for configuring and building
15 DPDK will be increased to v0.47.1 (from 0.41) from DPDK 19.05 onwards. For
16 those users with a version earlier than 0.47.1, an updated copy of meson
17 can be got using the ``pip``, or ``pip3``, tool for downloading python
20 * build macros: The macros defining RTE_MACHINE_CPUFLAG_* will be removed
21 from the build. The information provided by these macros is available
22 through standard compiler macros. For example, RTE_MACHINE_CPUFLAG_SSE3
23 duplicates the compiler-provided macro __SSE3__.
25 * kvargs: The function ``rte_kvargs_process`` will get a new parameter
26 for returning key match count. It will ease handling of no-match case.
28 * eal: To be more inclusive in choice of naming, the DPDK project
29 will replace uses of master/slave in the API's and command line arguments.
31 References to master/slave in relation to lcore will be renamed
32 to initial/worker. The function ``rte_get_master_lcore()``
33 will be renamed to ``rte_get_initial_lcore()``.
34 For the 20.11 release, both names will be present and the
35 old function will be marked with the deprecated tag.
36 The old function will be removed in a future version.
38 The iterator for worker lcores will also change:
39 ``RTE_LCORE_FOREACH_SLAVE`` will be replaced with
40 ``RTE_LCORE_FOREACH_WORKER``.
42 The ``master-lcore`` argument to testpmd will be replaced
43 with ``initial-lcore``. The old ``master-lcore`` argument
44 will produce a runtime notification in 20.11 release, and
45 be removed completely in a future release.
47 * eal: The terms blacklist and whitelist to describe devices used
48 by DPDK will be replaced in the 20.11 relase.
49 This will apply to command line arguments as well as macros.
51 The macro ``RTE_DEV_BLACKLISTED`` will be replaced with ``RTE_DEV_EXCLUDED``
52 and ``RTE_DEV_WHITELISTED`` will be replaced with ``RTE_DEV_INCLUDED``
53 ``RTE_BUS_SCAN_BLACKLIST`` and ``RTE_BUS_SCAN_WHITELIST`` will be
54 replaced with ``RTE_BUS_SCAN_EXCLUDED`` and ``RTE_BUS_SCAN_INCLUDED``
55 respectively. Likewise ``RTE_DEVTYPE_BLACKLISTED_PCI`` and
56 ``RTE_DEVTYPE_WHITELISTED_PCI`` will be replaced with
57 ``RTE_DEVTYPE_EXCLUDED`` and ``RTE_DEVTYPE_INCLUDED``.
59 The old macros will be marked as deprecated in 20.11 and any
60 usage will cause a compile warning. They will be removed in
63 The command line arguments to ``rte_eal_init`` will change from
64 ``-b, --pci-blacklist`` to ``-x, --exclude`` and
65 ``-w, --pci-whitelist`` to ``-i, --include``.
66 The old command line arguments will continue to be accepted in 20.11
67 but will cause a runtime warning message. The old arguments will
68 be removed in a future release.
70 * eal: The function ``rte_eal_remote_launch`` will return new error codes
71 after read or write error on the pipe, instead of calling ``rte_panic``.
73 * rte_atomicNN_xxx: These APIs do not take memory order parameter. This does
74 not allow for writing optimized code for all the CPU architectures supported
75 in DPDK. DPDK will adopt C11 atomic operations semantics and provide wrappers
76 using C11 atomic built-ins. These wrappers must be used for patches that
77 need to be merged in 20.08 onwards. This change will not introduce any
78 performance degradation.
80 * rte_smp_*mb: These APIs provide full barrier functionality. However, many
81 use cases do not require full barriers. To support such use cases, DPDK will
82 adopt C11 barrier semantics and provide wrappers using C11 atomic built-ins.
83 These wrappers must be used for patches that need to be merged in 20.08
84 onwards. This change will not introduce any performance degradation.
86 * rte_cio_*mb: Since the IO barriers for ARMv8 platforms are relaxed from DSB
87 to DMB, rte_cio_*mb APIs provide the same functionality as rte_io_*mb
88 APIs (taking all platforms into consideration). rte_io_*mb APIs should be
89 used in the place of rte_cio_*mb APIs. The rte_cio_*mb APIs will be
90 deprecated in 20.11 release.
92 * igb_uio: In the view of reducing the kernel dependency from the main tree,
93 as a first step, the Technical Board decided to move ``igb_uio``
94 kernel module to the dpdk-kmods repository in the /linux/igb_uio/ directory
96 Minutes of Technical Board Meeting of `2019-11-06
97 <https://mails.dpdk.org/archives/dev/2019-November/151763.html>`_.
99 * lib: will fix extending some enum/define breaking the ABI. There are multiple
100 samples in DPDK that enum/define terminated with a ``.*MAX.*`` value which is
101 used by iterators, and arrays holding these values are sized with this
102 ``.*MAX.*`` value. So extending this enum/define increases the ``.*MAX.*``
103 value which increases the size of the array and depending on how/where the
104 array is used this may break the ABI.
105 ``RTE_ETH_FLOW_MAX`` is one sample of the mentioned case, adding a new flow
106 type will break the ABI because of ``flex_mask[RTE_ETH_FLOW_MAX]`` array
107 usage in following public struct hierarchy:
108 ``rte_eth_fdir_flex_conf -> rte_fdir_conf -> rte_eth_conf (in the middle)``.
109 Need to identify this kind of usages and fix in 20.11, otherwise this blocks
110 us extending existing enum/define.
111 One solution can be using a fixed size array instead of ``.*MAX.*`` value.
113 * mbuf: Some fields will be converted to dynamic API in DPDK 20.11
114 in order to reserve more space for the dynamic fields, as explained in
115 `this presentation <https://www.youtube.com/watch?v=Ttl6MlhmzWY>`_.
116 The following static fields will be moved as dynamic:
119 - ``userdata`` / ``udata64``
122 As a consequence, the layout of the ``struct rte_mbuf`` will be re-arranged,
123 avoiding impact on vectorized implementation of the driver datapaths,
124 while evaluating performance gains of a better use of the first cache line.
126 * ethdev: Split the ``struct eth_dev_ops`` struct to hide it as much as possible
127 will be done in 20.11.
128 Currently the ``struct eth_dev_ops`` struct is accessible by the application
129 because some inline functions, like ``rte_eth_tx_descriptor_status()``,
130 access the struct directly.
131 The struct will be separate in two, the ops used by inline functions will be
132 moved next to Rx/Tx burst functions, rest of the ``struct eth_dev_ops`` struct
133 will be moved to header file for drivers to hide it from applications.
135 * ethdev: the legacy filter API, including
136 ``rte_eth_dev_filter_supported()``, ``rte_eth_dev_filter_ctrl()`` as well
137 as filter types MACVLAN, ETHERTYPE, FLEXIBLE, SYN, NTUPLE, TUNNEL, FDIR,
138 HASH and L2_TUNNEL, is superseded by the generic flow API (rte_flow) in
139 PMDs that implement the latter.
140 The legacy API will be removed in DPDK 20.11.
142 * ethdev: The flow director API, including ``rte_eth_conf.fdir_conf`` field,
143 and the related structures (``rte_fdir_*`` and ``rte_eth_fdir_*``),
144 will be removed in DPDK 20.11.
146 * ethdev: The legacy L2 tunnel filtering API is deprecated as the rest of
147 the legacy filtering API.
148 The functions ``rte_eth_dev_l2_tunnel_eth_type_conf`` and
149 ``rte_eth_dev_l2_tunnel_offload_set`` which were not marked as deprecated,
150 will be removed in DPDK 20.11.
152 * ethdev: Update API functions returning ``void`` to return ``int`` with
153 negative errno values to indicate various error conditions (e.g.
154 invalid port ID, unsupported operation, failed operation):
156 - ``rte_eth_dev_stop``
157 - ``rte_eth_dev_close``
159 * ethdev: The temporary flag RTE_ETH_DEV_CLOSE_REMOVE will be removed in 20.11.
160 As a consequence, the new behaviour introduced in 18.11 will be effective
161 for all drivers: generic port resources are freed on close operation.
162 Private resources are expected to be released in the ``dev_close`` callback.
163 More details in http://inbox.dpdk.org/dev/5248162.j6AOsuQRmx@thomas/
165 * ethdev: New offload flags ``DEV_RX_OFFLOAD_FLOW_MARK`` will be added in 19.11.
166 This will allow application to enable or disable PMDs from updating
167 ``rte_mbuf::hash::fdir``.
168 This scheme will allow PMDs to avoid writes to ``rte_mbuf`` fields on Rx and
169 thereby improve Rx performance if application wishes do so.
170 In 19.11 PMDs will still update the field even when the offload is not
173 * ethdev: Add new fields to ``rte_eth_rxconf`` to configure the receiving
174 queues to split ingress packets into multiple segments according to the
175 specified lengths into the buffers allocated from the specified
176 memory pools. The backward compatibility to existing API is preserved.
178 * ethdev: The ``struct rte_eth_rxq_info`` will be modified to include
179 a new optional field, indicating the buffer size used in receiving packets
180 for HW. This change is planned for 20.11. For more details:
181 https://mails.dpdk.org/archives/dev/2020-July/176135.html.
183 * ethdev: ``rx_descriptor_done`` dev_ops and ``rte_eth_rx_descriptor_done``
184 will be deprecated in 20.11 and will be removed in 21.11.
185 Existing ``rte_eth_rx_descriptor_status`` and ``rte_eth_tx_descriptor_status``
186 APIs can be used as replacement.
188 * ethdev: The port mirroring API can be replaced with a more fine grain flow API.
189 The structs ``rte_eth_mirror_conf``, ``rte_eth_vlan_mirror`` and the functions
190 ``rte_eth_mirror_rule_set``, ``rte_eth_mirror_rule_reset`` will be marked
191 as deprecated in DPDK 20.11, along with the associated macros ``ETH_MIRROR_*``.
192 This API will be fully removed in DPDK 21.11.
194 * ethdev: The ``struct rte_flow_item_eth`` and ``struct rte_flow_item_vlan``
195 structs will be modified, to include an additional value, indicating existence
196 or absence of a VLAN header following the current header, as proposed in RFC
197 https://mails.dpdk.org/archives/dev/2020-August/177536.html.
199 * ethdev: The ``struct rte_flow_item_ipv6`` struct will be modified to include
200 additional values, indicating existence or absence of IPv6 extension headers
201 following the IPv6 header, as proposed in RFC
202 https://mails.dpdk.org/archives/dev/2020-August/177257.html.
204 * ethdev: Some internal APIs for driver usage are exported in the .map file.
205 Now DPDK has ``__rte_internal`` marker so we can mark internal APIs and move
206 them to the INTERNAL block in .map. Although these APIs are internal it will
207 break the ABI checks, that is why change is planned for 20.11.
208 The list of internal APIs are mainly ones listed in ``rte_ethdev_driver.h``.
210 * pmd_dpaa: The API ``rte_pmd_dpaa_set_tx_loopback`` will have extended
211 ``port_id`` definition from ``uint8_t`` to ``uint16_t``.
213 * vhost: Vhost-user dequeue zero-copy support will be removed in 20.11.
214 The only known user is OVS where the feature is still experimental,
215 and has not received any update for 2.5 years.
216 This feature faces reliability issues and is often conflicting with
217 new features being implemented.
219 * security: The API ``rte_security_session_create`` takes only single mempool
220 for session and session private data. So the application need to create
221 mempool for twice the number of sessions needed and will also lead to
222 wastage of memory as session private data need more memory compared to session.
223 Hence the API will be modified to take two mempool pointers - one for session
224 and one for private data.
226 * cryptodev: ``RTE_CRYPTO_AEAD_LIST_END`` from ``enum rte_crypto_aead_algorithm``,
227 ``RTE_CRYPTO_CIPHER_LIST_END`` from ``enum rte_crypto_cipher_algorithm`` and
228 ``RTE_CRYPTO_AUTH_LIST_END`` from ``enum rte_crypto_auth_algorithm``
231 * cryptodev: support for using IV with all sizes is added, J0 still can
232 be used but only when IV length in following structs ``rte_crypto_auth_xform``,
233 ``rte_crypto_aead_xform`` is set to zero. When IV length is greater or equal
234 to one it means it represents IV, when is set to zero it means J0 is used
235 directly, in this case 16 bytes of J0 need to be passed.
237 * scheduler: The functions ``rte_cryptodev_scheduler_slave_attach``,
238 ``rte_cryptodev_scheduler_slave_detach`` and
239 ``rte_cryptodev_scheduler_slaves_get`` will be replaced in 20.11 by
240 ``rte_cryptodev_scheduler_worker_attach``,
241 ``rte_cryptodev_scheduler_worker_detach`` and
242 ``rte_cryptodev_scheduler_workers_get`` accordingly.
244 * eventdev: Following structures will be modified to support DLB PMD
245 and future extensions:
247 - ``rte_event_dev_info``
248 - ``rte_event_dev_config``
249 - ``rte_event_port_conf``
251 Patches containing justification, documentation, and proposed modifications
254 - https://patches.dpdk.org/patch/71457/
255 - https://patches.dpdk.org/patch/71456/
257 * acl: ``RTE_ACL_CLASSIFY_NUM`` enum value will be removed.
258 This enum value is not used inside DPDK, while it prevents to add new
259 classify algorithms without causing an ABI breakage.
261 * sched: To allow more traffic classes, flexible mapping of pipe queues to
262 traffic classes, and subport level configuration of pipes and queues
263 changes will be made to macros, data structures and API functions defined
264 in "rte_sched.h". These changes are aligned to improvements suggested in the
265 RFC https://mails.dpdk.org/archives/dev/2018-November/120035.html.
267 * sched: To allow dynamic configuration of the subport bandwidth profile,
268 changes will be made to data structures ``rte_sched_subport_params``,
269 ``rte_sched_port_params`` and new data structure, API functions will be
270 defined in ``rte_sched.h``. These changes are aligned as suggested in the
271 RFC https://mails.dpdk.org/archives/dev/2020-July/175161.html
273 * metrics: The function ``rte_metrics_init`` will have a non-void return
274 in order to notify errors instead of calling ``rte_exit``.
276 * power: ``rte_power_set_env`` function will no longer return 0 on attempt
277 to set new power environment if power environment was already initialized.
278 In this case the function will return -1 unless the environment is unset first
279 (using ``rte_power_unset_env``). Other function usage scenarios will not change.
281 * python: Since the beginning of 2020, Python 2 has officially reached
282 end-of-support: https://www.python.org/doc/sunset-python-2/.
283 Python 2 support will be completely removed in 20.11.
284 In 20.08, explicit deprecation warnings will be displayed when running
285 scripts with Python 2.
287 * dpdk-setup.sh: This old script relies on deprecated stuff, and especially
288 ``make``. Given environments are too much variables for such a simple script,
289 it will be removed in DPDK 20.11.
290 Some useful parts may be converted into specific scripts.