examples/eventdev: add all type queue option
authorPavan Nikhilesh <pbhagavatula@caviumnetworks.com>
Wed, 10 Jan 2018 11:10:08 +0000 (16:40 +0530)
committerJerin Jacob <jerin.jacob@caviumnetworks.com>
Fri, 19 Jan 2018 15:09:56 +0000 (16:09 +0100)
Added configurable option to make queue type as all type queues i.e.
RTE_EVENT_QUEUE_CFG_ALL_TYPES based on event dev capability
RTE_EVENT_DEV_CAP_QUEUE_ALL_TYPES.

This can be enabled by supplying '-a' as a cmdline argument.

Signed-off-by: Pavan Nikhilesh <pbhagavatula@caviumnetworks.com>
Acked-by: Harry van Haaren <harry.van.haaren@intel.com>
examples/eventdev_pipeline_sw_pmd/main.c
examples/eventdev_pipeline_sw_pmd/pipeline_common.h
examples/eventdev_pipeline_sw_pmd/pipeline_worker_generic.c
examples/eventdev_pipeline_sw_pmd/pipeline_worker_tx.c

index e07f68e..81ba340 100644 (file)
@@ -120,6 +120,7 @@ static struct option long_options[] = {
        {"parallel", no_argument, 0, 'p'},
        {"ordered", no_argument, 0, 'o'},
        {"quiet", no_argument, 0, 'q'},
+       {"use-atq", no_argument, 0, 'a'},
        {"dump", no_argument, 0, 'D'},
        {0, 0, 0, 0}
 };
@@ -143,6 +144,7 @@ usage(void)
                "  -o, --ordered                Use ordered scheduling\n"
                "  -p, --parallel               Use parallel scheduling\n"
                "  -q, --quiet                  Minimize printed output\n"
+               "  -a, --use-atq                Use all type queues\n"
                "  -D, --dump                   Print detailed statistics before exit"
                "\n";
        fprintf(stderr, "%s", usage_str);
@@ -163,7 +165,7 @@ parse_app_args(int argc, char **argv)
        int i;
 
        for (;;) {
-               c = getopt_long(argc, argv, "r:t:e:c:w:n:f:s:poPqDW:",
+               c = getopt_long(argc, argv, "r:t:e:c:w:n:f:s:paoPqDW:",
                                long_options, &option_index);
                if (c == -1)
                        break;
@@ -196,6 +198,9 @@ parse_app_args(int argc, char **argv)
                case 'p':
                        cdata.queue_type = RTE_SCHED_TYPE_PARALLEL;
                        break;
+               case 'a':
+                       cdata.all_type_queues = 1;
+                       break;
                case 'q':
                        cdata.quiet = 1;
                        break;
index e063200..6655303 100644 (file)
@@ -77,6 +77,7 @@ struct config_data {
        int quiet;
        int dump_dev;
        int dump_dev_signal;
+       int all_type_queues;
        unsigned int num_stages;
        unsigned int worker_cq_depth;
        unsigned int rx_stride;
index 90f8770..2c51f4a 100644 (file)
@@ -516,6 +516,11 @@ generic_opt_check(void)
        memset(&eventdev_info, 0, sizeof(struct rte_event_dev_info));
        rte_event_dev_info_get(0, &eventdev_info);
 
+       if (cdata.all_type_queues && !(eventdev_info.event_dev_cap &
+                               RTE_EVENT_DEV_CAP_QUEUE_ALL_TYPES))
+               rte_exit(EXIT_FAILURE,
+                               "Event dev doesn't support all type queues\n");
+
        for (i = 0; i < rte_eth_dev_count(); i++) {
                ret = rte_event_eth_rx_adapter_caps_get(0, i, &cap);
                if (ret)
index 419d8d4..25eabbd 100644 (file)
@@ -94,6 +94,52 @@ worker_do_tx(void *arg)
        return 0;
 }
 
+static int
+worker_do_tx_atq(void *arg)
+{
+       struct rte_event ev;
+
+       struct worker_data *data = (struct worker_data *)arg;
+       const uint8_t dev = data->dev_id;
+       const uint8_t port = data->port_id;
+       const uint8_t lst_qid = cdata.num_stages - 1;
+       size_t fwd = 0, received = 0, tx = 0;
+
+       while (!fdata->done) {
+
+               if (!rte_event_dequeue_burst(dev, port, &ev, 1, 0)) {
+                       rte_pause();
+                       continue;
+               }
+
+               received++;
+               const uint8_t cq_id = ev.sub_event_type % cdata.num_stages;
+
+               if (cq_id == lst_qid) {
+                       if (ev.sched_type == RTE_SCHED_TYPE_ATOMIC) {
+                               worker_tx_pkt(ev.mbuf);
+                               tx++;
+                               continue;
+                       }
+
+                       worker_fwd_event(&ev, RTE_SCHED_TYPE_ATOMIC);
+               } else {
+                       ev.sub_event_type++;
+                       worker_fwd_event(&ev, cdata.queue_type);
+               }
+               work();
+
+               worker_event_enqueue(dev, port, &ev);
+               fwd++;
+       }
+
+       if (!cdata.quiet)
+               printf("  worker %u thread done. RX=%zu FWD=%zu TX=%zu\n",
+                               rte_lcore_id(), received, fwd, tx);
+
+       return 0;
+}
+
 static int
 worker_do_tx_burst(void *arg)
 {
@@ -149,17 +195,81 @@ worker_do_tx_burst(void *arg)
        return 0;
 }
 
+static int
+worker_do_tx_burst_atq(void *arg)
+{
+       struct rte_event ev[BATCH_SIZE];
+
+       struct worker_data *data = (struct worker_data *)arg;
+       uint8_t dev = data->dev_id;
+       uint8_t port = data->port_id;
+       uint8_t lst_qid = cdata.num_stages - 1;
+       size_t fwd = 0, received = 0, tx = 0;
+
+       while (!fdata->done) {
+               uint16_t i;
+
+               const uint16_t nb_rx = rte_event_dequeue_burst(dev, port,
+                               ev, BATCH_SIZE, 0);
+
+               if (nb_rx == 0) {
+                       rte_pause();
+                       continue;
+               }
+               received += nb_rx;
+
+               for (i = 0; i < nb_rx; i++) {
+                       const uint8_t cq_id = ev[i].sub_event_type %
+                               cdata.num_stages;
+
+                       if (cq_id == lst_qid) {
+                               if (ev[i].sched_type == RTE_SCHED_TYPE_ATOMIC) {
+                                       worker_tx_pkt(ev[i].mbuf);
+                                       tx++;
+                                       ev[i].op = RTE_EVENT_OP_RELEASE;
+                                       continue;
+                               }
+
+                               worker_fwd_event(&ev[i], RTE_SCHED_TYPE_ATOMIC);
+                       } else {
+                               ev[i].sub_event_type++;
+                               worker_fwd_event(&ev[i], cdata.queue_type);
+                       }
+                       work();
+               }
+
+               worker_event_enqueue_burst(dev, port, ev, nb_rx);
+               fwd += nb_rx;
+       }
+
+       if (!cdata.quiet)
+               printf("  worker %u thread done. RX=%zu FWD=%zu TX=%zu\n",
+                               rte_lcore_id(), received, fwd, tx);
+
+       return 0;
+}
+
 static int
 setup_eventdev_worker_tx(struct cons_data *cons_data,
                struct worker_data *worker_data)
 {
        RTE_SET_USED(cons_data);
        uint8_t i;
+       const uint8_t atq = cdata.all_type_queues ? 1 : 0;
        const uint8_t dev_id = 0;
        const uint8_t nb_ports = cdata.num_workers;
        uint8_t nb_slots = 0;
-       uint8_t nb_queues = rte_eth_dev_count() * cdata.num_stages;
-       nb_queues += rte_eth_dev_count();
+       uint8_t nb_queues = rte_eth_dev_count();
+
+       /*
+        * In case where all type queues are not enabled, use queues equal to
+        * number of stages * eth_dev_count and one extra queue per pipeline
+        * for Tx.
+        */
+       if (!atq) {
+               nb_queues *= cdata.num_stages;
+               nb_queues += rte_eth_dev_count();
+       }
 
        struct rte_event_dev_config config = {
                        .nb_event_queues = nb_queues,
@@ -211,12 +321,19 @@ setup_eventdev_worker_tx(struct cons_data *cons_data,
        printf("  Stages:\n");
        for (i = 0; i < nb_queues; i++) {
 
-               uint8_t slot;
+               if (atq) {
 
-               nb_slots = cdata.num_stages + 1;
-               slot = i % nb_slots;
-               wkr_q_conf.schedule_type = slot == cdata.num_stages ?
-                       RTE_SCHED_TYPE_ATOMIC : cdata.queue_type;
+                       nb_slots = cdata.num_stages;
+                       wkr_q_conf.event_queue_cfg =
+                               RTE_EVENT_QUEUE_CFG_ALL_TYPES;
+               } else {
+                       uint8_t slot;
+
+                       nb_slots = cdata.num_stages + 1;
+                       slot = i % nb_slots;
+                       wkr_q_conf.schedule_type = slot == cdata.num_stages ?
+                               RTE_SCHED_TYPE_ATOMIC : cdata.queue_type;
+               }
 
                if (rte_event_queue_setup(dev_id, i, &wkr_q_conf) < 0) {
                        printf("%d: error creating qid %d\n", __LINE__, i);
@@ -286,7 +403,7 @@ setup_eventdev_worker_tx(struct cons_data *cons_data,
         *
         * This forms two set of queue pipelines 0->1->2->tx and 3->4->5->tx.
         */
-       cdata.rx_stride = nb_slots;
+       cdata.rx_stride = atq ? 1 : nb_slots;
        ret = rte_event_dev_service_id_get(dev_id,
                                &fdata->evdev_service_id);
        if (ret != -ESRCH && ret != 0) {
@@ -450,6 +567,11 @@ worker_tx_opt_check(void)
        memset(&eventdev_info, 0, sizeof(struct rte_event_dev_info));
        rte_event_dev_info_get(0, &eventdev_info);
 
+       if (cdata.all_type_queues && !(eventdev_info.event_dev_cap &
+                               RTE_EVENT_DEV_CAP_QUEUE_ALL_TYPES))
+               rte_exit(EXIT_FAILURE,
+                               "Event dev doesn't support all type queues\n");
+
        for (i = 0; i < rte_eth_dev_count(); i++) {
                ret = rte_event_eth_rx_adapter_caps_get(0, i, &cap);
                if (ret)
@@ -477,13 +599,33 @@ worker_tx_opt_check(void)
        }
 }
 
+static worker_loop
+get_worker_loop_burst(uint8_t atq)
+{
+       if (atq)
+               return worker_do_tx_burst_atq;
+
+       return worker_do_tx_burst;
+}
+
+static worker_loop
+get_worker_loop_non_burst(uint8_t atq)
+{
+       if (atq)
+               return worker_do_tx_atq;
+
+       return worker_do_tx;
+}
+
 void
 set_worker_tx_setup_data(struct setup_data *caps, bool burst)
 {
+       uint8_t atq = cdata.all_type_queues ? 1 : 0;
+
        if (burst)
-               caps->worker = worker_do_tx_burst;
+               caps->worker = get_worker_loop_burst(atq);
        else
-               caps->worker = worker_do_tx;
+               caps->worker = get_worker_loop_non_burst(atq);
 
        memset(fdata->tx_core, 0, sizeof(unsigned int) * MAX_NUM_CORE);