[dpdk-dev] [PATCH v5 4/7] eventdev: add eth Rx adapter API header
Jerin Jacob
jerin.jacob at caviumnetworks.com
Mon Oct 9 14:27:14 CEST 2017
-----Original Message-----
> Date: Sat, 7 Oct 2017 02:39:58 +0530
> From: Nikhil Rao <nikhil.rao at intel.com>
> To: jerin.jacob at caviumnetworks.com, bruce.richardson at intel.com
> CC: dev at dpdk.org
> Subject: [PATCH v5 4/7] eventdev: add eth Rx adapter API header
> X-Mailer: git-send-email 2.7.4
>
> Add common APIs for configuring packet transfer from ethernet Rx
> queues to event devices across HW & SW packet transfer mechanisms.
> A detailed description of the adapter is contained in the header's
> comments.
>
> Signed-off-by: Nikhil Rao <nikhil.rao at intel.com>
> ---
The EXPERIMENTAL functions follows the following comment in doxygen. Add
that for new Rx adapter functions
* @warning
* @b EXPERIMENTAL: this API may change without prior notice
reference file: lib/librte_eal/common/include/rte_service.h
> +#define RTE_EVENT_ETH_RX_ADAPTER_MAX_INSTANCE 32
> +
> +/* struct rte_event_eth_rx_adapter_queue_conf flags definitions */
> +#define RTE_EVENT_ETH_RX_ADAPTER_QUEUE_FLOW_ID_VALID 0x1
> +/**< This flag indicates the flow identifier is valid
> + * @see rte_event_eth_rx_adapter_queue_conf::rx_queue_flags
> + */
> +
> +struct rte_event_eth_rx_adapter_conf {
Doxygen comment missing for this structure.
> + uint8_t event_port_id;
> + /**< Event port identifier, the adapter enqueues mbuf events to this
> + * port.
> + */
> + uint32_t max_nb_rx;
> + /**< The adapter can return early if it has processed at least
> + * max_nb_rx mbufs. This isn't treated as a requirement; batching may
> + * cause the adapter to process more than max_nb_rx mbufs.
> + */
> +};
> +
> +
> +struct rte_event_eth_rx_adapter_stats {
Doxygen comment missing for this structure.
> + uint64_t rx_poll_count;
> + /**< Receive queue poll count */
> + uint64_t rx_packets;
> + /**< Received packet count */
> + uint64_t rx_enq_count;
> + /**< Eventdev enqueue count */
> + uint64_t rx_enq_retry;
> + /**< Eventdev enqueue retry count */
> + uint64_t rx_enq_start_ts;
> + /**< Rx enqueue start timestamp */
> + uint64_t rx_enq_block_cycles;
> + /**< Cycles for which the service is blocked by the event device,
> + * i.e, the service fails to enqueue to the event device.
> + */
> + uint64_t rx_enq_end_ts;
> + /**< Latest timestamp at which the service is unblocked
> + * by the event device. The start, end timestamps and
> + * block cycles can be used to compute the percentage of
> + * cycles the service is blocked by the event device.
> + */
> +};
> +
> +
> +#ifdef __cplusplus
> +}
> +#endif
> +#endif /* _RTE_EVENT_ETH_RX_ADAPTER_ */
> diff --git a/MAINTAINERS b/MAINTAINERS
> index 8df2a7f2a..53fd50e1f 100644
> --- a/MAINTAINERS
> +++ b/MAINTAINERS
> @@ -273,6 +273,9 @@ F: lib/librte_eventdev/
> F: drivers/event/skeleton/
> F: test/test/test_eventdev.c
>
> +Event Ethdev Rx Adapter API - EXPERIMENTAL
> +M: Nikhil Rao <nikhil.rao at intel.com>
T: git://dpdk.org/next/dpdk-next-eventdev
> +F: lib/librte_eventdev/*eth_rx_adapter*
>
With above changes:
Acked-by: Jerin Jacob <jerin.jacob at caviumnetworks.com>
More information about the dev
mailing list