eventdev: clarify atomic and ordered queue config
authorGage Eads <gage.eads@intel.com>
Wed, 21 Jun 2017 13:32:51 +0000 (19:02 +0530)
committerJerin Jacob <jerin.jacob@caviumnetworks.com>
Wed, 21 Jun 2017 13:33:13 +0000 (15:33 +0200)
The nb_atomic_flows and nb_atomic_order_sequences fields are only inspected
if the queue is configured for atomic or ordered scheduling, respectively.
This commit updates the documentation to reflect that.

Signed-off-by: Gage Eads <gage.eads@intel.com>
Acked-by: Jerin Jacob <jerin.jacob@caviumnetworks.com>
lib/librte_eventdev/rte_eventdev.h

index 20e7293..9428433 100644 (file)
@@ -521,9 +521,11 @@ rte_event_dev_configure(uint8_t dev_id,
 struct rte_event_queue_conf {
        uint32_t nb_atomic_flows;
        /**< The maximum number of active flows this queue can track at any
-        * given time. The value must be in the range of
-        * [1 - nb_event_queue_flows)] which previously provided in
-        * rte_event_dev_info_get().
+        * given time. If the queue is configured for atomic scheduling (by
+        * applying the RTE_EVENT_QUEUE_CFG_ALL_TYPES or
+        * RTE_EVENT_QUEUE_CFG_ATOMIC_ONLY flags to event_queue_cfg), then the
+        * value must be in the range of [1, nb_event_queue_flows], which was
+        * previously provided in rte_event_dev_configure().
         */
        uint32_t nb_atomic_order_sequences;
        /**< The maximum number of outstanding events waiting to be
@@ -533,8 +535,11 @@ struct rte_event_queue_conf {
         * scheduler cannot schedule the events from this queue and invalid
         * event will be returned from dequeue until one or more entries are
         * freed up/released.
-        * The value must be in the range of [1 - nb_event_queue_flows)]
-        * which previously supplied to rte_event_dev_configure().
+        * If the queue is configured for ordered scheduling (by applying the
+        * RTE_EVENT_QUEUE_CFG_ALL_TYPES or RTE_EVENT_QUEUE_CFG_ORDERED_ONLY
+        * flags to event_queue_cfg), then the value must be in the range of
+        * [1, nb_event_queue_flows], which was previously supplied to
+        * rte_event_dev_configure().
         */
        uint32_t event_queue_cfg; /**< Queue cfg flags(EVENT_QUEUE_CFG_) */
        uint8_t priority;