37. ABI and API Deprecation

See the guidelines document for details of the ABI policy.

With DPDK 23.11, there will be a new major ABI version: 24. This means that during the development of 23.11, new items may be added to structs or enums, even if those additions involve an ABI compatibility breakage.

Other API and ABI deprecation notices are to be posted below.

37.1. Deprecation Notices

  • build: The enable_kmods option is deprecated and will be removed in a future release. Setting/clearing the option has no impact on the build. Instead, kernel modules will be always built for OS’s where out-of-tree kernel modules are required for DPDK operation. Currently, this means that modules will only be built for FreeBSD. No modules are shipped with DPDK for either Linux or Windows.

  • kvargs: The function rte_kvargs_process will get a new parameter for returning key match count. It will ease handling of no-match case.

  • telemetry: The functions rte_tel_data_add_array_u64 and rte_tel_data_add_dict_u64, used by telemetry callbacks for adding unsigned integer values to be returned to the user, are renamed to rte_tel_data_add_array_uint and rte_tel_data_add_dict_uint respectively. As such, the old function names are deprecated and will be removed in a future release.

  • 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.

  • 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.

  • 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.

  • net: A new IPv6 address structure will be introduced in DPDK 24.11. It will replace all ad-hoc uint8_t[16] arrays in all public APIs and structures. The following libraries and symbols are expected to be affected:

    ethdev
    • struct rte_flow_item_icmp6_nd_ns
    • struct rte_flow_item_icmp6_nd_na
    • struct rte_flow_action_set_ipv6
    • struct rte_flow_tunnel
    fib
    • rte_fib6_add()
    • rte_fib6_delete()
    • rte_fib6_lookup_bulk()
    gro
    • struct tcp6_flow_key
    hash
    • struct rte_ipv6_tuple
    ipsec
    • struct rte_ipsec_sadv6_key
    lpm
    • rte_lpm6_add()
    • rte_lpm6_is_rule_present()
    • rte_lpm6_delete()
    • rte_lpm6_delete_bulk_func()
    • rte_lpm6_lookup()
    • rte_lpm6_lookup_bulk_func()
    net
    • struct rte_ipv6_hdr
    node
    • rte_node_ip6_route_add()
    pipeline
    • struct rte_table_action_ipv6_header
    rib
    • rte_rib6_lookup()
    • rte_rib6_lookup_exact()
    • rte_rib6_get_nxt()
    • rte_rib6_insert()
    • rte_rib6_remove()
    • rte_rib6_get_ip()
    table
    • struct rte_table_lpm_ipv6_key

  • net, ethdev: The flow item RTE_FLOW_ITEM_TYPE_VXLAN_GPE is replaced with RTE_FLOW_ITEM_TYPE_VXLAN. The struct rte_flow_item_vxlan_gpe and its mask rte_flow_item_vxlan_gpe_mask are replaced with rte_flow_item_vxlan and rte_flow_item_vxlan_mask. The flow item RTE_FLOW_ITEM_TYPE_VXLAN_GPE, the structs rte_flow_item_vxlan_gpe, rte_flow_item_vxlan_gpe_mask, and the header struct rte_vxlan_gpe_hdr with the macro RTE_ETHER_VXLAN_GPE_HLEN will be removed in DPDK 25.11.

  • ethdev: The flow API matching pattern structures, struct rte_flow_item_*, should start with relevant protocol header structure from lib/net/. The individual protocol header fields and the protocol header struct may be kept together in a union as a first migration step. In future (target is DPDK 23.11), the protocol header fields will be cleaned and only protocol header struct will remain.

    These items are not compliant (not including struct from lib/net/):

    • rte_flow_item_ah
    • rte_flow_item_e_tag
    • rte_flow_item_geneve
    • rte_flow_item_geneve_opt
    • rte_flow_item_gre
    • rte_flow_item_icmp6
    • rte_flow_item_icmp6_nd_na
    • rte_flow_item_icmp6_nd_ns
    • rte_flow_item_icmp6_nd_opt
    • rte_flow_item_icmp6_nd_opt_sla_eth
    • rte_flow_item_icmp6_nd_opt_tla_eth
    • rte_flow_item_igmp
    • rte_flow_item_ipv6_ext
    • rte_flow_item_l2tpv3oip
    • rte_flow_item_mpls
    • rte_flow_item_nsh
    • rte_flow_item_nvgre
    • rte_flow_item_pfcp
    • rte_flow_item_pppoe
    • rte_flow_item_pppoe_proto_id

  • 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.

  • ethdev: Flow actions PF and VF have been deprecated since DPDK 21.11 and are yet to be removed. That still has not happened because there are net drivers which support combined use of either action PF or action VF with action QUEUE, namely, i40e, ixgbe and txgbe (L2 tunnel rule). It is unclear whether it is acceptable to just drop support for such a complex use case, so maintainers of the said drivers should take a closer look at this and provide assistance.

  • 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 RTE_FLOW_ACTION_TYPE_MODIFY_FIELD. The legacy actions should be removed once MODIFY_FIELD alternative is implemented in drivers.

  • fib: A new flag field will be introduced in rte_fib_conf structure in DPDK 24.11. This field will be used to pass extra configuration settings.

  • 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.

  • cryptodev: The structure rte_cryptodev_qp_conf will be updated to have a new parameter to set priority of that particular queue pair.

  • cryptodev: The enum rte_crypto_asym_xform_type and struct rte_crypto_asym_op will be extended to include new values to support EDDSA. This will break ABI compatibility with existing applications that use these data types.

  • cryptodev: The rte_crypto_rsa_xform struct member to hold private key in either exponent or quintuple format is changed from union to struct data type. This change is to support ASN.1 syntax (RFC 3447 Appendix A.1.2). This change will not break existing applications.

  • cryptodev: Some changes may happen to manage RSA padding for virtio-crypto.

  • cryptodev: The Intel IPsec Multi-Buffer version will be bumped to a minimum version of v1.4. This will effect the KASUMI, SNOW3G, ZUC, AESNI GCM, AESNI MB and CHACHAPOLY SW PMDs.

  • ipsec: The IPsec library is updated to support sequence number provided by application. To allow the same, two new API functions are being introduced: rte_ipsec_pkt_crypto_sqn_assign and rte_ipsec_pkt_crypto_xprepare. It separates the sequence number update functionality from the existing rte_ipsec_pkt_crypto_prepare function. Corresponding configure structure changes are being made for the new API. Additionally a new flag RTE_IPSEC_SAFLAG_SQN_ASSIGN_DISABLE is introduced to disable sequence number assignment in IPsec library.

  • eventdev: The single-event (non-burst) enqueue and dequeue operations, used by static inline burst enqueue and dequeue functions in rte_eventdev.h, will be removed in DPDK 23.11. This simplification includes changing the layout and potentially also the size of the public rte_event_fp_ops struct, breaking the ABI. Since these functions are not called directly by the application, the API remains unaffected.

  • pipeline: The pipeline library legacy API (functions rte_pipeline_*) will be deprecated and subsequently removed in DPDK 24.11 release. Before this, the new pipeline library API (functions rte_swx_pipeline_*) will gradually transition from experimental to stable status.

  • table: The table library legacy API (functions rte_table_*) will be deprecated and subsequently removed in DPDK 24.11 release. Before this, the new table library API (functions rte_swx_table_*) will gradually transition from experimental to stable status.

  • port: The port library legacy API (functions rte_port_*) will be deprecated and subsequently removed in DPDK 24.11 release. Before this, the new port library API (functions rte_swx_port_*) will gradually transition from experimental to stable status.

  • graph: The graph library data structures will be modified to support node specific errors. The structures rte_node, rte_node_register and rte_graph_cluster_node_stats will be extended to include node error counters and error description.