* DPDK Deprecation Notice Review Meeting 2022-09-30
@ 2022-10-06 15:14 Mcnamara, John
0 siblings, 0 replies; only message in thread
From: Mcnamara, John @ 2022-10-06 15:14 UTC (permalink / raw)
To: dev; +Cc: thomas, david.marchand, orika
[-- Attachment #1: Type: text/plain, Size: 12272 bytes --]
DPDK Deprecation Notice Review Meeting 2022-09-30
=================================================
DPDK Community member met and reviewed the status of the DPDK
deprecation notices prior to the RC1 deadline. The source document
is https://git.dpdk.org/dpdk/tree/doc/guides/rel_notes/deprecation.rst
copied below.
Key messages:
* We should do periodic community reviews of the deprecation notices
and in particular at the start of the yy.11 cycle.
* We could group long standing deprecations notices at the start of
the doc in a section that is separate from the notices that are
specific to the current yy.11.
* Notes to the notices are inline below and prefixed with ">".
Deprecation Notices
-------------------
* kvargs: The function ``rte_kvargs_process`` will get a new parameter
for returning key match count. It will ease handling of no-match case.
> Won't make it into this release.
* eal: RTE_FUNC_PTR_OR_* macros have been marked deprecated and will be removed
in the future. Applications can use ``devtools/cocci/func_or_ret.cocci``
to update their code.
> This is still relevant to current and future releases.
* eal: The function ``rte_eal_remote_launch`` will return new error codes
after read or write error on the pipe, instead of calling ``rte_panic``.
> partially done may need other rte_panic removed
* rte_atomicNN_xxx: These APIs do not take memory order parameter. This does
not allow for writing optimized code for all the CPU architectures supported
in DPDK. DPDK has adopted the atomic operations from
https://gcc.gnu.org/onlinedocs/gcc/_005f_005fatomic-Builtins.html. These
operations must be used for patches that need to be merged in 20.08 onwards.
This change will not introduce any performance degradation.
> This is still relevant to current and future releases.
* rte_smp_*mb: These APIs provide full barrier functionality. However, many
use cases do not require full barriers. To support such use cases, DPDK has
adopted atomic operations from
https://gcc.gnu.org/onlinedocs/gcc/_005f_005fatomic-Builtins.html. These
operations and a new wrapper ``rte_atomic_thread_fence`` instead of
``__atomic_thread_fence`` must be used for patches that need to be merged in
20.08 onwards. This change will not introduce any performance degradation.
> This is still relevant to current and future releases.
* bus: The ``dev->device.numa_node`` field is set by each bus driver for
every device it manages to indicate on which NUMA node this device lies.
When this information is unknown, the assigned value is not consistent
across the bus drivers.
In DPDK 22.11, the default value will be set to -1 by all bus drivers
when the NUMA information is unavailable.
> This is in progress.
* kni: The KNI kernel module and library are not recommended for use by new
applications - other technologies such as virtio-user are recommended instead.
Following the DPDK technical board
`decision <https://mails.dpdk.org/archives/dev/2021-January/197077.html>`_
and `refinement <http://mails.dpdk.org/archives/dev/2022-June/243596.html>`_:
* Some deprecation warnings will be added in DPDK 22.11.
* The KNI example application will be removed from DPDK 22.11.
* The KNI kernel module, library and PMD will be removed from the DPDK 23.11.
> We should add a compilation and/or dmesg deprecation warning @ferruh
* lib: will fix extending some enum/define breaking the ABI. There are multiple
samples in DPDK that enum/define terminated with a ``.*MAX.*`` value which is
used by iterators, and arrays holding these values are sized with this
``.*MAX.*`` value. So extending this enum/define increases the ``.*MAX.*``
value which increases the size of the array and depending on how/where the
array is used this may break the ABI.
``RTE_ETH_FLOW_MAX`` is one sample of the mentioned case, adding a new flow
type will break the ABI because of ``flex_mask[RTE_ETH_FLOW_MAX]`` array
usage in following public struct hierarchy:
``rte_eth_fdir_flex_conf -> rte_eth_fdir_conf -> rte_eth_conf (in the middle)``.
Need to identify this kind of usages and fix in 20.11, otherwise this blocks
us extending existing enum/define.
One solution can be using a fixed size array instead of ``.*MAX.*`` value.
> We need some documentation, in the contributors guide or the programmers guide
on how to deal with this. There is general agreement that having a MAX value
in an enum is a bad idea for ABI but developers like using this as a sentinal
or array size value so some people have been using "reserved" padding values
in enums to workaround this restrictions. The community should agree, and
document, what is acceptable.
* ethdev: The function ``rte_eth_set_queue_rate_limit`` takes ``rate`` in Mbps.
The queue rate is limited to 64 Gbps because declared as ``uint16_t``.
The ``rate`` parameter will be modified to ``uint32_t`` in DPDK 22.11
so that it can work for more than 64 Gbps.
> Patch submitted need ack from Intel and NXT.
> https://patches.dpdk.org/project/dpdk/patch/1664344318-3594-1-git-send-email-skoteshwar@marvell.com/
* ethdev: Since no single PMD supports ``RTE_ETH_RX_OFFLOAD_HEADER_SPLIT``
offload and the ``split_hdr_size`` field in structure ``rte_eth_rxmode``
to enable per-port header split, they will be removed in DPDK 22.11.
The per-queue Rx packet split offload ``RTE_ETH_RX_OFFLOAD_BUFFER_SPLIT``
can still be used, and it is configured by ``rte_eth_rxseg_split``.
> Acked and will be merge
> https://patches.dpdk.org/project/dpdk/patch/20220812031313.87385-1-xuan.ding@intel.com/
* ethdev: Announce moving from dedicated modify function for each field,
to using the general ``rte_flow_modify_field`` action.
> This is still relevant to current and future releases.
* ethdev: The flow API matching pattern structures, ``struct rte_flow_item_*``,
should start with relevant protocol header.
Some matching pattern structures implements this by duplicating protocol header
fields in the struct. To clarify the intention and to be sure protocol header
is intact, will replace those fields with relevant protocol header struct.
In v21.02 both individual protocol header fields and the protocol header struct
will be added as union, target is switch usage to the protocol header by time.
In v21.11 LTS, protocol header fields will be cleaned and only protocol header
struct will remain.
> Some work still ongoing, may need to go into RC2 but should be in RC1
> Best effort for RC1, and then list any missing structs in the docs.
* ethdev: Queue specific stats fields will be removed from ``struct rte_eth_stats``.
Mentioned fields are: ``q_ipackets``, ``q_opackets``, ``q_ibytes``, ``q_obytes``,
``q_errors``.
Instead queue stats will be received via xstats API. Current method support
will be limited to maximum 256 queues.
Also compile time flag ``RTE_ETHDEV_QUEUE_STAT_CNTRS`` will be removed.
> This deprecation notice will stay.
* ethdev: Items and actions ``PF``, ``VF``, ``PHY_PORT``, ``PORT_ID`` are
deprecated as hard-to-use / ambiguous and will be removed in DPDK 22.11.
> Some have been removed as part of a patch series. The deprecation notice will
need to change to reflect.
* ethdev: The use of attributes ``ingress`` / ``egress`` in "transfer" flows
is deprecated as ambiguous with respect to the embedded switch. The use of
these attributes will become invalid starting from DPDK 22.11.
> There is a patch in progress for this.
* ethdev: Actions ``OF_SET_MPLS_TTL``, ``OF_DEC_MPLS_TTL``, ``OF_SET_NW_TTL``,
``OF_COPY_TTL_OUT``, ``OF_COPY_TTL_IN`` are deprecated as not supported by
any PMD, so they will be removed in DPDK 22.11.
> Patch has been submitted
* ethdev: Actions ``OF_DEC_NW_TTL``, ``SET_IPV4_SRC``, ``SET_IPV4_DST``,
``SET_IPV6_SRC``, ``SET_IPV6_DST``, ``SET_TP_SRC``, ``SET_TP_DST``,
``DEC_TTL``, ``SET_TTL``, ``SET_MAC_SRC``, ``SET_MAC_DST``, ``INC_TCP_SEQ``,
``DEC_TCP_SEQ``, ``INC_TCP_ACK``, ``DEC_TCP_ACK``, ``SET_IPV4_DSCP``,
``SET_IPV6_DSCP``, ``SET_TAG``, ``SET_META`` are marked as legacy and
superseded by the generic MODIFY_FIELD action.
The legacy actions should be deprecated in 22.07, once MODIFY_FIELD
alternative is implemented.
The legacy actions should be removed in DPDK 22.11.
> Should be grouped with ??? notice above, and the specific drivers named.
> There was a proposal to have an earlier removal date, at the very start
> of the xx.11 period so that code is changed/tested in time for RC1.
* ethdev: The enum ``rte_eth_event_ipsec_subtype`` will be extended to add
new subtype values ``RTE_ETH_EVENT_IPSEC_SA_PKT_EXPIRY``,
``RTE_ETH_EVENT_IPSEC_SA_BYTE_HARD_EXPIRY`` and
``RTE_ETH_EVENT_IPSEC_SA_PKT_HARD_EXPIRY`` in DPDK 22.11.
> Merged
* bbdev: ``RTE_BBDEV_OP_TYPE_COUNT`` terminating the ``rte_bbdev_op_type``
enum will be deprecated and instead use fixed array size when required
to allow for future enum extension.
Will extend API to support new operation type ``RTE_BBDEV_OP_FFT`` as per
this `RFC <https://patches.dpdk.org/project/dpdk/list/?series=22111>`__.
New members will be added in ``rte_bbdev_driver_info`` to expose
PMD queue topology inspired by
this `RFC <https://patches.dpdk.org/project/dpdk/list/?series=22076>`__.
New member will be added in ``rte_bbdev_driver_info`` to expose
the device status as per
this `RFC <https://patches.dpdk.org/project/dpdk/list/?series=23367>`__.
This should be updated in DPDK 22.11.
> This involves several patches, some will be merged
* cryptodev: Hide structures ``rte_cryptodev_sym_session`` and
``rte_cryptodev_asym_session`` to remove unnecessary indirection between
session and the private data of session. An opaque pointer can be exposed
directly to application which can be attached to the ``rte_crypto_op``.
> Patches submitted. One issue with security below.
* cryptodev: The function ``rte_cryptodev_cb_fn`` will be updated
to have another parameter ``qp_id`` to return the queue pair ID
which got error interrupt to the application,
so that application can reset that particular queue pair.
> Will be removed without any objections(? note unclear)
* security: Hide structure ``rte_security_session`` and expose an opaque
pointer for the private data to the application which can be attached
to the packet while enqueuing.
> Kontantin has a comment (should be merged in RC2)
* eventdev: The function ``rte_event_crypto_adapter_queue_pair_add`` will
accept configuration of type ``rte_event_crypto_adapter_queue_conf`` instead
of ``rte_event``, similar to ``rte_event_eth_rx_adapter_queue_add`` signature.
Event will be one of the configuration fields,
together with additional vector parameters.
> Patches ready. Comment required from Abhinandan
* eventdev: The function pointer declaration ``eventdev_stop_flush_t``
will be renamed to ``rte_eventdev_stop_flush_t`` in DPDK 22.11.
> Patch available and will be merged today
* eventdev: The element ``*u64s`` in the structure ``rte_event_vector``
is deprecated and will be replaced with ``u64s`` in DPDK 22.11.
> Patch available and will be merged today
* eventdev: The structure ``rte_event_vector`` will be modified to include
``elem_offset:12`` bits taken from ``rsvd:15``. The ``elem_offset`` defines
the offset into the vector array from which valid elements are present.
The difference between ``rte_event_vector::nb_elem`` and
``rte_event_vector::elem_offset`` gives the number of valid elements left
to process from the ``rte_event_vector::elem_offset``.
> Patch is merged. Should remove deprecation docs.
* metrics: The function ``rte_metrics_init`` will have a non-void return
in order to notify errors instead of calling ``rte_exit``.
> Patch submitted by Bruce
* raw/dpaa2_cmdif: The ``dpaa2_cmdif`` rawdev driver will be deprecated
in DPDK 22.11, as it is no longer in use, no active user known.
> Drop from deprecations.
[-- Attachment #2: Type: text/html, Size: 35323 bytes --]
^ permalink raw reply [flat|nested] only message in thread
only message in thread, other threads:[~2022-10-06 15:14 UTC | newest]
Thread overview: (only message) (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2022-10-06 15:14 DPDK Deprecation Notice Review Meeting 2022-09-30 Mcnamara, John
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).