app/testeventdev: add order queue test
authorJerin Jacob <jerin.jacob@caviumnetworks.com>
Tue, 4 Jul 2017 04:53:11 +0000 (10:23 +0530)
committerJerin Jacob <jerin.jacob@caviumnetworks.com>
Fri, 7 Jul 2017 07:33:32 +0000 (09:33 +0200)
The order queue test configures the eventdev with two queues
and an event producer to inject the events to q0(ordered) queue.
Both q0(ordered) and q1(atomic) are linked to all the workers.

The event producer maintains a sequence number per flow and
injects the events to the ordered queue.

The worker receives the events from ordered queue and
forwards to atomic queue. Since the events from an ordered queue can
be processed in parallel on the different workers, the
ingress order of events might have changed on the downsteam
atomic queue enqueue. On enqueue to the atomic queue, the eventdev PMD
driver reorders the event to the original ingress order
i.e producer ingress order).

When the event is dequeued from the atomic queue by the worker,
this test verifies the expected
sequence number of associated event per flow by comparing
the free running expected sequence number per flow.

Example command to run order queue test:

sudo build/app/dpdk-test-eventdev --vdev=event_sw0 --\
--test=order_queue --plcores 1 --wlcores 2,3

Signed-off-by: Jerin Jacob <jerin.jacob@caviumnetworks.com>
Acked-by: Harry van Haaren <harry.van.haaren@intel.com>
app/test-eventdev/Makefile
app/test-eventdev/test_order_queue.c [new file with mode: 0644]

index acbf74c..37c04c2 100644 (file)
@@ -44,5 +44,6 @@ SRCS-y += evt_test.c
 SRCS-y += parser.c
 
 SRCS-y += test_order_common.c
+SRCS-y += test_order_queue.c
 
 include $(RTE_SDK)/mk/rte.app.mk
diff --git a/app/test-eventdev/test_order_queue.c b/app/test-eventdev/test_order_queue.c
new file mode 100644 (file)
index 0000000..c4003ef
--- /dev/null
@@ -0,0 +1,142 @@
+/*
+ *   BSD LICENSE
+ *
+ *   Copyright (C) Cavium 2017.
+ *
+ *   Redistribution and use in source and binary forms, with or without
+ *   modification, are permitted provided that the following conditions
+ *   are met:
+ *
+ *     * Redistributions of source code must retain the above copyright
+ *       notice, this list of conditions and the following disclaimer.
+ *     * Redistributions in binary form must reproduce the above copyright
+ *       notice, this list of conditions and the following disclaimer in
+ *       the documentation and/or other materials provided with the
+ *       distribution.
+ *     * Neither the name of Cavium nor the names of its
+ *       contributors may be used to endorse or promote products derived
+ *       from this software without specific prior written permission.
+ *
+ *   THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ *   "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ *   LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ *   A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ *   OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ *   SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ *   LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ *   DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ *   THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ *   (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ *   OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#include <stdio.h>
+#include <unistd.h>
+
+#include "test_order_common.h"
+
+/* See http://dpdk.org/doc/guides/tools/testeventdev.html for test details */
+
+#define NB_QUEUES 2
+static int
+order_queue_eventdev_setup(struct evt_test *test, struct evt_options *opt)
+{
+       int ret;
+
+       const uint8_t nb_workers = evt_nr_active_lcores(opt->wlcores);
+       /* number of active worker cores + 1 producer */
+       const uint8_t nb_ports = nb_workers + 1;
+
+       const struct rte_event_dev_config config = {
+                       .nb_event_queues = NB_QUEUES,/* q0 ordered, q1 atomic */
+                       .nb_event_ports = nb_ports,
+                       .nb_events_limit  = 4096,
+                       .nb_event_queue_flows = opt->nb_flows,
+                       .nb_event_port_dequeue_depth = 128,
+                       .nb_event_port_enqueue_depth = 128,
+       };
+
+       ret = rte_event_dev_configure(opt->dev_id, &config);
+       if (ret) {
+               evt_err("failed to configure eventdev %d", opt->dev_id);
+               return ret;
+       }
+
+       /* q0 (ordered queue) configuration */
+       struct rte_event_queue_conf q0_ordered_conf = {
+                       .priority = RTE_EVENT_DEV_PRIORITY_NORMAL,
+                       .event_queue_cfg = RTE_EVENT_QUEUE_CFG_ORDERED_ONLY,
+                       .nb_atomic_flows = opt->nb_flows,
+                       .nb_atomic_order_sequences = opt->nb_flows,
+       };
+       ret = rte_event_queue_setup(opt->dev_id, 0, &q0_ordered_conf);
+       if (ret) {
+               evt_err("failed to setup queue0 eventdev %d", opt->dev_id);
+               return ret;
+       }
+
+       /* q1 (atomic queue) configuration */
+       struct rte_event_queue_conf q1_atomic_conf = {
+                       .priority = RTE_EVENT_DEV_PRIORITY_NORMAL,
+                       .event_queue_cfg = RTE_EVENT_QUEUE_CFG_ATOMIC_ONLY,
+                       .nb_atomic_flows = opt->nb_flows,
+                       .nb_atomic_order_sequences = opt->nb_flows,
+       };
+       ret = rte_event_queue_setup(opt->dev_id, 1, &q1_atomic_conf);
+       if (ret) {
+               evt_err("failed to setup queue1 eventdev %d", opt->dev_id);
+               return ret;
+       }
+
+       /* setup one port per worker, linking to all queues */
+       ret = order_event_dev_port_setup(test, opt, nb_workers, NB_QUEUES);
+       if (ret)
+               return ret;
+
+       ret = rte_event_dev_start(opt->dev_id);
+       if (ret) {
+               evt_err("failed to start eventdev %d", opt->dev_id);
+               return ret;
+       }
+
+       return 0;
+}
+
+static void
+order_queue_opt_dump(struct evt_options *opt)
+{
+       order_opt_dump(opt);
+       evt_dump("nb_evdev_queues", "%d", NB_QUEUES);
+}
+
+static bool
+order_queue_capability_check(struct evt_options *opt)
+{
+       struct rte_event_dev_info dev_info;
+
+       rte_event_dev_info_get(opt->dev_id, &dev_info);
+       if (dev_info.max_event_queues < NB_QUEUES || dev_info.max_event_ports <
+                       order_nb_event_ports(opt)) {
+               evt_err("not enough eventdev queues=%d/%d or ports=%d/%d",
+                       NB_QUEUES, dev_info.max_event_queues,
+                       order_nb_event_ports(opt), dev_info.max_event_ports);
+               return false;
+       }
+
+       return true;
+}
+
+static const struct evt_test_ops order_queue =  {
+       .cap_check          = order_queue_capability_check,
+       .opt_check          = order_opt_check,
+       .opt_dump           = order_queue_opt_dump,
+       .test_setup         = order_test_setup,
+       .mempool_setup      = order_mempool_setup,
+       .eventdev_setup     = order_queue_eventdev_setup,
+       .eventdev_destroy   = order_eventdev_destroy,
+       .mempool_destroy    = order_mempool_destroy,
+       .test_result        = order_test_result,
+       .test_destroy       = order_test_destroy,
+};
+
+EVT_TEST_REGISTER(order_queue);