[dpdk-dev] Guidelines for moving PMDs to new ethdev offloads API

[Shahaf Shuler]

In 17.11 the ethdev offloads API has changed:

                commit cba7f53b717d ("ethdev: introduce Tx queue offloads API")

                commit ce17eddefc20 ("ethdev: introduce Rx queue offloads API") The new API is documented in the programmer's guide:

                https://emea01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fdpdk.org%2Fdoc%2Fguides%2Fprog_guide%2Fpoll_mode_drv.html%23hardware-offload&data=02%7C01%7Cshahafs%40mellanox.com%7Cc365451c3f934b73db5e08d53f6fe6fd%7Ca652971c7d2e4d9ba6a4d149256f461b%7C0%7C0%7C636484672216209911&sdata=l0OkZ8m78BMsw%2F0d7bA8m80gVqlSL%2Bpcs9UplVSmlOA%3D&reserved=0



As announced in the deprecation notice, the old API is planned to be removed in 18.05:

                https://emea01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fdpdk.org%2Fdoc%2Fguides%2Frel_notes%2Fdeprecation.html&data=02%7C01%7Cshahafs%40mellanox.com%7Cc365451c3f934b73db5e08d53f6fe6fd%7Ca652971c7d2e4d9ba6a4d149256f461b%7C0%7C0%7C636484672216209911&sdata=85wB%2BoU3um%2BXegYiLAJIDYWIkvpb9WQtPSgA73zsLHI%3D&reserved=0

As agreed, all the PMDs must be converted to the new API by their maintainers.

This email will better explain the transition process.



The main concepts in the new API are:

                - All offloads are disabled by default

                - Distinction between per port and per queue offloads.

                Per port offloads needs to be set also for each queue of the port.



The transition period is using an API translation in rte_ethdev.c:

                - The old API is translated to allow using converted PMDs

                - The new API is translated to allow using not converted PMDs

                - The use of the new API is flagged to trigger the right translation



* Device capabilities - rte_eth_dev_info_get()

                - Not changing: use DEV_RX/TX_OFFLOAD_* flags

                - Old API: per-port fields rte_eth_dev_info.rx/tx_offload_capa

                - New API: added per-queue rte_eth_dev_info.rx/tx_queue_offload_capa



* Rx offloads

                - Not changing: offload disabled by default

                - Old API: per-port rte_eth_conf.rxmode.[bit-fields]

                - New API: per-port rte_eth_conf.rxmode.offloads and per queue rte_eth_rxconf.offloads  using DEV_RX_OFFLOAD_*

                - Transition: rte_eth_conf.rxmode.ignore_offload_bitfield must be set

                              when using the new API (Done by application)

                - To be removed: rte_eth_conf.rxmode.[

                                header_split, hw_ip_checksum, hw_vlan_filter, hw_vlan_strip,

                                hw_vlan_extend, jumbo_frame, hw_strip_crc, enable_scatter,

                                enable_lro, hw_timestamp, security]



* Tx offloads

                - Old API: enabled by default

                           per-queue rte_eth_txconf.txq_flags using ETH_TXQ_FLAGS_NO*

                - New API: disabled by default

                           per-port rte_eth_conf.txmode.offloads and per queue rte_eth_txconf.offloads  using DEV_TX_OFFLOAD_*

                - Transition: ETH_TXQ_FLAGS_IGNORE must be set when using the new API (Done by application)

                              PMD should ignore port offloads not set on the device configuration when using old API

                - To be removed: rte_eth_txconf.txq_flags and ETH_TXQ_FLAGS_NO*


--Shahaf