event/octeontx2: add SSO selftest
authorPavan Nikhilesh <pbhagavatula@marvell.com>
Fri, 28 Jun 2019 18:23:36 +0000 (23:53 +0530)
committerJerin Jacob <jerinj@marvell.com>
Wed, 3 Jul 2019 04:56:29 +0000 (06:56 +0200)
Add selftest to verify sanity of SSO.
Can be run by passing devargs to SSO PF as follows:

Example:
--dev "0002:0e:00.0,selftest=1"

Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
Signed-off-by: Jerin Jacob <jerinj@marvell.com>
app/test/test_eventdev.c
doc/guides/eventdevs/octeontx2.rst
drivers/event/octeontx2/Makefile
drivers/event/octeontx2/meson.build
drivers/event/octeontx2/otx2_evdev.c
drivers/event/octeontx2/otx2_evdev.h
drivers/event/octeontx2/otx2_evdev_selftest.c [new file with mode: 0644]

index c745e99..783140d 100644 (file)
@@ -1014,7 +1014,15 @@ test_eventdev_selftest_octeontx(void)
        return test_eventdev_selftest_impl("event_octeontx", "");
 }
 
+static int
+test_eventdev_selftest_octeontx2(void)
+{
+       return test_eventdev_selftest_impl("otx2_eventdev", "");
+}
+
 REGISTER_TEST_COMMAND(eventdev_common_autotest, test_eventdev_common);
 REGISTER_TEST_COMMAND(eventdev_selftest_sw, test_eventdev_selftest_sw);
 REGISTER_TEST_COMMAND(eventdev_selftest_octeontx,
                test_eventdev_selftest_octeontx);
+REGISTER_TEST_COMMAND(eventdev_selftest_octeontx2,
+               test_eventdev_selftest_octeontx2);
index 9b235f2..562a83d 100644 (file)
@@ -81,6 +81,15 @@ Runtime Config Options
 
     --dev "0002:0e:00.0,qos=[1-50-50-50]"
 
+- ``Selftest``
+
+  The functionality of OCTEON TX2 eventdev can be verified using this option,
+  various unit and functional tests are run to verify the sanity.
+  The tests are run once the vdev creation is successfully complete.
+  For example::
+
+    --dev "0002:0e:00.0,selftest=1"
+
 Debugging Options
 ~~~~~~~~~~~~~~~~~
 
index 01c1a98..ea07909 100644 (file)
@@ -36,6 +36,7 @@ LIBABIVER := 1
 SRCS-$(CONFIG_RTE_LIBRTE_PMD_OCTEONTX2_EVENTDEV) += otx2_worker_dual.c
 SRCS-$(CONFIG_RTE_LIBRTE_PMD_OCTEONTX2_EVENTDEV) += otx2_worker.c
 SRCS-$(CONFIG_RTE_LIBRTE_PMD_OCTEONTX2_EVENTDEV) += otx2_evdev.c
+SRCS-$(CONFIG_RTE_LIBRTE_PMD_OCTEONTX2_EVENTDEV) += otx2_evdev_selftest.c
 SRCS-$(CONFIG_RTE_LIBRTE_PMD_OCTEONTX2_EVENTDEV) += otx2_evdev_irq.c
 
 LDLIBS += -lrte_eal -lrte_bus_pci -lrte_pci -lrte_kvargs
index c2a5f3e..470564b 100644 (file)
@@ -6,6 +6,7 @@ sources = files('otx2_worker.c',
                'otx2_worker_dual.c',
                'otx2_evdev.c',
                'otx2_evdev_irq.c',
+               'otx2_evdev_selftest.c',
                )
 
 allow_experimental_apis = true
index 5004fe2..c5a1509 100644 (file)
@@ -1099,11 +1099,13 @@ static struct rte_eventdev_ops otx2_sso_ops = {
        .dev_start        = otx2_sso_start,
        .dev_stop         = otx2_sso_stop,
        .dev_close        = otx2_sso_close,
+       .dev_selftest     = otx2_sso_selftest,
 };
 
 #define OTX2_SSO_XAE_CNT       "xae_cnt"
 #define OTX2_SSO_SINGLE_WS     "single_ws"
 #define OTX2_SSO_GGRP_QOS      "qos"
+#define OTX2_SSO_SELFTEST      "selftest"
 
 static void
 parse_queue_param(char *value, void *opaque)
@@ -1186,6 +1188,8 @@ sso_parse_devargs(struct otx2_sso_evdev *dev, struct rte_devargs *devargs)
        if (kvlist == NULL)
                return;
 
+       rte_kvargs_process(kvlist, OTX2_SSO_SELFTEST, &parse_kvargs_flag,
+                          &dev->selftest);
        rte_kvargs_process(kvlist, OTX2_SSO_XAE_CNT, &parse_kvargs_value,
                           &dev->xae_cnt);
        rte_kvargs_process(kvlist, OTX2_SSO_SINGLE_WS, &parse_kvargs_flag,
@@ -1301,6 +1305,10 @@ otx2_sso_init(struct rte_eventdev *event_dev)
        otx2_sso_dbg("Initializing %s max_queues=%d max_ports=%d",
                     event_dev->data->name, dev->max_event_queues,
                     dev->max_event_ports);
+       if (dev->selftest) {
+               event_dev->dev->driver = &pci_sso.driver;
+               event_dev->dev_ops->dev_selftest();
+       }
 
 
        return 0;
@@ -1347,4 +1355,5 @@ RTE_PMD_REGISTER_PCI_TABLE(event_octeontx2, pci_sso_map);
 RTE_PMD_REGISTER_KMOD_DEP(event_octeontx2, "vfio-pci");
 RTE_PMD_REGISTER_PARAM_STRING(event_octeontx2, OTX2_SSO_XAE_CNT "=<int>"
                              OTX2_SSO_SINGLE_WS "=1"
-                             OTX2_SSO_GGRP_QOS "=<string>");
+                             OTX2_SSO_GGRP_QOS "=<string>"
+                             OTX2_SSO_SELFTEST "=1");
index 2aa7421..fc8dde4 100644 (file)
@@ -130,6 +130,7 @@ struct otx2_sso_evdev {
        struct rte_mempool *xaq_pool;
        /* Dev args */
        uint8_t dual_ws;
+       uint8_t selftest;
        uint32_t xae_cnt;
        uint8_t qos_queue_cnt;
        struct otx2_sso_qos *qos_parse_data;
@@ -247,6 +248,8 @@ typedef void (*otx2_handle_event_t)(void *arg, struct rte_event ev);
 void ssogws_flush_events(struct otx2_ssogws *ws, uint8_t queue_id,
                         uintptr_t base, otx2_handle_event_t fn, void *arg);
 void ssogws_reset(struct otx2_ssogws *ws);
+/* Selftest */
+int otx2_sso_selftest(void);
 /* Init and Fini API's */
 int otx2_sso_init(struct rte_eventdev *event_dev);
 int otx2_sso_fini(struct rte_eventdev *event_dev);
diff --git a/drivers/event/octeontx2/otx2_evdev_selftest.c b/drivers/event/octeontx2/otx2_evdev_selftest.c
new file mode 100644 (file)
index 0000000..8440a50
--- /dev/null
@@ -0,0 +1,1511 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(C) 2019 Marvell International Ltd.
+ */
+
+#include <rte_atomic.h>
+#include <rte_common.h>
+#include <rte_cycles.h>
+#include <rte_debug.h>
+#include <rte_eal.h>
+#include <rte_ethdev.h>
+#include <rte_eventdev.h>
+#include <rte_hexdump.h>
+#include <rte_launch.h>
+#include <rte_lcore.h>
+#include <rte_mbuf.h>
+#include <rte_malloc.h>
+#include <rte_memcpy.h>
+#include <rte_per_lcore.h>
+#include <rte_random.h>
+#include <rte_test.h>
+
+#include "otx2_evdev.h"
+
+#define NUM_PACKETS (1024)
+#define MAX_EVENTS  (1024)
+
+#define OCTEONTX2_TEST_RUN(setup, teardown, test) \
+       octeontx_test_run(setup, teardown, test, #test)
+
+static int total;
+static int passed;
+static int failed;
+static int unsupported;
+
+static int evdev;
+static struct rte_mempool *eventdev_test_mempool;
+
+struct event_attr {
+       uint32_t flow_id;
+       uint8_t event_type;
+       uint8_t sub_event_type;
+       uint8_t sched_type;
+       uint8_t queue;
+       uint8_t port;
+};
+
+static uint32_t seqn_list_index;
+static int seqn_list[NUM_PACKETS];
+
+static inline void
+seqn_list_init(void)
+{
+       RTE_BUILD_BUG_ON(NUM_PACKETS < MAX_EVENTS);
+       memset(seqn_list, 0, sizeof(seqn_list));
+       seqn_list_index = 0;
+}
+
+static inline int
+seqn_list_update(int val)
+{
+       if (seqn_list_index >= NUM_PACKETS)
+               return -1;
+
+       seqn_list[seqn_list_index++] = val;
+       rte_smp_wmb();
+       return 0;
+}
+
+static inline int
+seqn_list_check(int limit)
+{
+       int i;
+
+       for (i = 0; i < limit; i++) {
+               if (seqn_list[i] != i) {
+                       otx2_err("Seqn mismatch %d %d", seqn_list[i], i);
+                       return -1;
+               }
+       }
+       return 0;
+}
+
+struct test_core_param {
+       rte_atomic32_t *total_events;
+       uint64_t dequeue_tmo_ticks;
+       uint8_t port;
+       uint8_t sched_type;
+};
+
+static int
+testsuite_setup(void)
+{
+       const char *eventdev_name = "event_octeontx2";
+
+       evdev = rte_event_dev_get_dev_id(eventdev_name);
+       if (evdev < 0) {
+               otx2_err("%d: Eventdev %s not found", __LINE__, eventdev_name);
+               return -1;
+       }
+       return 0;
+}
+
+static void
+testsuite_teardown(void)
+{
+       rte_event_dev_close(evdev);
+}
+
+static inline void
+devconf_set_default_sane_values(struct rte_event_dev_config *dev_conf,
+                               struct rte_event_dev_info *info)
+{
+       memset(dev_conf, 0, sizeof(struct rte_event_dev_config));
+       dev_conf->dequeue_timeout_ns = info->min_dequeue_timeout_ns;
+       dev_conf->nb_event_ports = info->max_event_ports;
+       dev_conf->nb_event_queues = info->max_event_queues;
+       dev_conf->nb_event_queue_flows = info->max_event_queue_flows;
+       dev_conf->nb_event_port_dequeue_depth =
+                       info->max_event_port_dequeue_depth;
+       dev_conf->nb_event_port_enqueue_depth =
+                       info->max_event_port_enqueue_depth;
+       dev_conf->nb_event_port_enqueue_depth =
+                       info->max_event_port_enqueue_depth;
+       dev_conf->nb_events_limit =
+                       info->max_num_events;
+}
+
+enum {
+       TEST_EVENTDEV_SETUP_DEFAULT,
+       TEST_EVENTDEV_SETUP_PRIORITY,
+       TEST_EVENTDEV_SETUP_DEQUEUE_TIMEOUT,
+};
+
+static inline int
+_eventdev_setup(int mode)
+{
+       const char *pool_name = "evdev_octeontx_test_pool";
+       struct rte_event_dev_config dev_conf;
+       struct rte_event_dev_info info;
+       int i, ret;
+
+       /* Create and destrory pool for each test case to make it standalone */
+       eventdev_test_mempool = rte_pktmbuf_pool_create(pool_name, MAX_EVENTS,
+                                                       0, 0, 512,
+                                                       rte_socket_id());
+       if (!eventdev_test_mempool) {
+               otx2_err("ERROR creating mempool");
+               return -1;
+       }
+
+       ret = rte_event_dev_info_get(evdev, &info);
+       RTE_TEST_ASSERT_SUCCESS(ret, "Failed to get event dev info");
+
+       devconf_set_default_sane_values(&dev_conf, &info);
+       if (mode == TEST_EVENTDEV_SETUP_DEQUEUE_TIMEOUT)
+               dev_conf.event_dev_cfg |= RTE_EVENT_DEV_CFG_PER_DEQUEUE_TIMEOUT;
+
+       ret = rte_event_dev_configure(evdev, &dev_conf);
+       RTE_TEST_ASSERT_SUCCESS(ret, "Failed to configure eventdev");
+
+       uint32_t queue_count;
+       RTE_TEST_ASSERT_SUCCESS(rte_event_dev_attr_get(evdev,
+                               RTE_EVENT_DEV_ATTR_QUEUE_COUNT, &queue_count),
+                               "Queue count get failed");
+
+       if (mode == TEST_EVENTDEV_SETUP_PRIORITY) {
+               if (queue_count > 8)
+                       queue_count = 8;
+
+               /* Configure event queues(0 to n) with
+                * RTE_EVENT_DEV_PRIORITY_HIGHEST to
+                * RTE_EVENT_DEV_PRIORITY_LOWEST
+                */
+               uint8_t step = (RTE_EVENT_DEV_PRIORITY_LOWEST + 1) /
+                               queue_count;
+               for (i = 0; i < (int)queue_count; i++) {
+                       struct rte_event_queue_conf queue_conf;
+
+                       ret = rte_event_queue_default_conf_get(evdev, i,
+                                                              &queue_conf);
+                       RTE_TEST_ASSERT_SUCCESS(ret, "Failed to get def_conf%d",
+                                               i);
+                       queue_conf.priority = i * step;
+                       ret = rte_event_queue_setup(evdev, i, &queue_conf);
+                       RTE_TEST_ASSERT_SUCCESS(ret, "Failed to setup queue=%d",
+                                               i);
+               }
+
+       } else {
+               /* Configure event queues with default priority */
+               for (i = 0; i < (int)queue_count; i++) {
+                       ret = rte_event_queue_setup(evdev, i, NULL);
+                       RTE_TEST_ASSERT_SUCCESS(ret, "Failed to setup queue=%d",
+                                               i);
+               }
+       }
+       /* Configure event ports */
+       uint32_t port_count;
+       RTE_TEST_ASSERT_SUCCESS(rte_event_dev_attr_get(evdev,
+                               RTE_EVENT_DEV_ATTR_PORT_COUNT, &port_count),
+                               "Port count get failed");
+       for (i = 0; i < (int)port_count; i++) {
+               ret = rte_event_port_setup(evdev, i, NULL);
+               RTE_TEST_ASSERT_SUCCESS(ret, "Failed to setup port=%d", i);
+               ret = rte_event_port_link(evdev, i, NULL, NULL, 0);
+               RTE_TEST_ASSERT(ret >= 0, "Failed to link all queues port=%d",
+                               i);
+       }
+
+       ret = rte_event_dev_start(evdev);
+       RTE_TEST_ASSERT_SUCCESS(ret, "Failed to start device");
+
+       return 0;
+}
+
+static inline int
+eventdev_setup(void)
+{
+       return _eventdev_setup(TEST_EVENTDEV_SETUP_DEFAULT);
+}
+
+static inline int
+eventdev_setup_priority(void)
+{
+       return _eventdev_setup(TEST_EVENTDEV_SETUP_PRIORITY);
+}
+
+static inline int
+eventdev_setup_dequeue_timeout(void)
+{
+       return _eventdev_setup(TEST_EVENTDEV_SETUP_DEQUEUE_TIMEOUT);
+}
+
+static inline void
+eventdev_teardown(void)
+{
+       rte_event_dev_stop(evdev);
+       rte_mempool_free(eventdev_test_mempool);
+}
+
+static inline void
+update_event_and_validation_attr(struct rte_mbuf *m, struct rte_event *ev,
+                                uint32_t flow_id, uint8_t event_type,
+                                uint8_t sub_event_type, uint8_t sched_type,
+                                uint8_t queue, uint8_t port)
+{
+       struct event_attr *attr;
+
+       /* Store the event attributes in mbuf for future reference */
+       attr = rte_pktmbuf_mtod(m, struct event_attr *);
+       attr->flow_id = flow_id;
+       attr->event_type = event_type;
+       attr->sub_event_type = sub_event_type;
+       attr->sched_type = sched_type;
+       attr->queue = queue;
+       attr->port = port;
+
+       ev->flow_id = flow_id;
+       ev->sub_event_type = sub_event_type;
+       ev->event_type = event_type;
+       /* Inject the new event */
+       ev->op = RTE_EVENT_OP_NEW;
+       ev->sched_type = sched_type;
+       ev->queue_id = queue;
+       ev->mbuf = m;
+}
+
+static inline int
+inject_events(uint32_t flow_id, uint8_t event_type, uint8_t sub_event_type,
+             uint8_t sched_type, uint8_t queue, uint8_t port,
+             unsigned int events)
+{
+       struct rte_mbuf *m;
+       unsigned int i;
+
+       for (i = 0; i < events; i++) {
+               struct rte_event ev = {.event = 0, .u64 = 0};
+
+               m = rte_pktmbuf_alloc(eventdev_test_mempool);
+               RTE_TEST_ASSERT_NOT_NULL(m, "mempool alloc failed");
+
+               m->seqn = i;
+               update_event_and_validation_attr(m, &ev, flow_id, event_type,
+                                                sub_event_type, sched_type,
+                                                queue, port);
+               rte_event_enqueue_burst(evdev, port, &ev, 1);
+       }
+       return 0;
+}
+
+static inline int
+check_excess_events(uint8_t port)
+{
+       uint16_t valid_event;
+       struct rte_event ev;
+       int i;
+
+       /* Check for excess events, try for a few times and exit */
+       for (i = 0; i < 32; i++) {
+               valid_event = rte_event_dequeue_burst(evdev, port, &ev, 1, 0);
+
+               RTE_TEST_ASSERT_SUCCESS(valid_event,
+                                       "Unexpected valid event=%d",
+                                       ev.mbuf->seqn);
+       }
+       return 0;
+}
+
+static inline int
+generate_random_events(const unsigned int total_events)
+{
+       struct rte_event_dev_info info;
+       uint32_t queue_count;
+       unsigned int i;
+       int ret;
+
+       RTE_TEST_ASSERT_SUCCESS(rte_event_dev_attr_get(evdev,
+                               RTE_EVENT_DEV_ATTR_QUEUE_COUNT, &queue_count),
+                               "Queue count get failed");
+
+       ret = rte_event_dev_info_get(evdev, &info);
+       RTE_TEST_ASSERT_SUCCESS(ret, "Failed to get event dev info");
+       for (i = 0; i < total_events; i++) {
+               ret = inject_events(
+                       rte_rand() % info.max_event_queue_flows /*flow_id */,
+                       RTE_EVENT_TYPE_CPU /* event_type */,
+                       rte_rand() % 256 /* sub_event_type */,
+                       rte_rand() % (RTE_SCHED_TYPE_PARALLEL + 1),
+                       rte_rand() % queue_count /* queue */,
+                       0 /* port */,
+                       1 /* events */);
+               if (ret)
+                       return -1;
+       }
+       return ret;
+}
+
+
+static inline int
+validate_event(struct rte_event *ev)
+{
+       struct event_attr *attr;
+
+       attr = rte_pktmbuf_mtod(ev->mbuf, struct event_attr *);
+       RTE_TEST_ASSERT_EQUAL(attr->flow_id, ev->flow_id,
+                             "flow_id mismatch enq=%d deq =%d",
+                             attr->flow_id, ev->flow_id);
+       RTE_TEST_ASSERT_EQUAL(attr->event_type, ev->event_type,
+                             "event_type mismatch enq=%d deq =%d",
+                             attr->event_type, ev->event_type);
+       RTE_TEST_ASSERT_EQUAL(attr->sub_event_type, ev->sub_event_type,
+                             "sub_event_type mismatch enq=%d deq =%d",
+                             attr->sub_event_type, ev->sub_event_type);
+       RTE_TEST_ASSERT_EQUAL(attr->sched_type, ev->sched_type,
+                             "sched_type mismatch enq=%d deq =%d",
+                             attr->sched_type, ev->sched_type);
+       RTE_TEST_ASSERT_EQUAL(attr->queue, ev->queue_id,
+                             "queue mismatch enq=%d deq =%d",
+                             attr->queue, ev->queue_id);
+       return 0;
+}
+
+typedef int (*validate_event_cb)(uint32_t index, uint8_t port,
+                                struct rte_event *ev);
+
+static inline int
+consume_events(uint8_t port, const uint32_t total_events, validate_event_cb fn)
+{
+       uint32_t events = 0, forward_progress_cnt = 0, index = 0;
+       uint16_t valid_event;
+       struct rte_event ev;
+       int ret;
+
+       while (1) {
+               if (++forward_progress_cnt > UINT16_MAX) {
+                       otx2_err("Detected deadlock");
+                       return -1;
+               }
+
+               valid_event = rte_event_dequeue_burst(evdev, port, &ev, 1, 0);
+               if (!valid_event)
+                       continue;
+
+               forward_progress_cnt = 0;
+               ret = validate_event(&ev);
+               if (ret)
+                       return -1;
+
+               if (fn != NULL) {
+                       ret = fn(index, port, &ev);
+                       RTE_TEST_ASSERT_SUCCESS(ret,
+                               "Failed to validate test specific event");
+               }
+
+               ++index;
+
+               rte_pktmbuf_free(ev.mbuf);
+               if (++events >= total_events)
+                       break;
+       }
+
+       return check_excess_events(port);
+}
+
+static int
+validate_simple_enqdeq(uint32_t index, uint8_t port, struct rte_event *ev)
+{
+       RTE_SET_USED(port);
+       RTE_TEST_ASSERT_EQUAL(index, ev->mbuf->seqn, "index=%d != seqn=%d",
+                             index, ev->mbuf->seqn);
+       return 0;
+}
+
+static inline int
+test_simple_enqdeq(uint8_t sched_type)
+{
+       int ret;
+
+       ret = inject_events(0 /*flow_id */,
+                           RTE_EVENT_TYPE_CPU /* event_type */,
+                           0 /* sub_event_type */,
+                           sched_type,
+                           0 /* queue */,
+                           0 /* port */,
+                           MAX_EVENTS);
+       if (ret)
+               return -1;
+
+       return consume_events(0 /* port */, MAX_EVENTS, validate_simple_enqdeq);
+}
+
+static int
+test_simple_enqdeq_ordered(void)
+{
+       return test_simple_enqdeq(RTE_SCHED_TYPE_ORDERED);
+}
+
+static int
+test_simple_enqdeq_atomic(void)
+{
+       return test_simple_enqdeq(RTE_SCHED_TYPE_ATOMIC);
+}
+
+static int
+test_simple_enqdeq_parallel(void)
+{
+       return test_simple_enqdeq(RTE_SCHED_TYPE_PARALLEL);
+}
+
+/*
+ * Generate a prescribed number of events and spread them across available
+ * queues. On dequeue, using single event port(port 0) verify the enqueued
+ * event attributes
+ */
+static int
+test_multi_queue_enq_single_port_deq(void)
+{
+       int ret;
+
+       ret = generate_random_events(MAX_EVENTS);
+       if (ret)
+               return -1;
+
+       return consume_events(0 /* port */, MAX_EVENTS, NULL);
+}
+
+/*
+ * Inject 0..MAX_EVENTS events over 0..queue_count with modulus
+ * operation
+ *
+ * For example, Inject 32 events over 0..7 queues
+ * enqueue events 0, 8, 16, 24 in queue 0
+ * enqueue events 1, 9, 17, 25 in queue 1
+ * ..
+ * ..
+ * enqueue events 7, 15, 23, 31 in queue 7
+ *
+ * On dequeue, Validate the events comes in 0,8,16,24,1,9,17,25..,7,15,23,31
+ * order from queue0(highest priority) to queue7(lowest_priority)
+ */
+static int
+validate_queue_priority(uint32_t index, uint8_t port, struct rte_event *ev)
+{
+       uint32_t queue_count;
+
+       RTE_TEST_ASSERT_SUCCESS(rte_event_dev_attr_get(evdev,
+                               RTE_EVENT_DEV_ATTR_QUEUE_COUNT, &queue_count),
+                               "Queue count get failed");
+       if (queue_count > 8)
+               queue_count = 8;
+       uint32_t range = MAX_EVENTS / queue_count;
+       uint32_t expected_val = (index % range) * queue_count;
+
+       expected_val += ev->queue_id;
+       RTE_SET_USED(port);
+       RTE_TEST_ASSERT_EQUAL(ev->mbuf->seqn, expected_val,
+       "seqn=%d index=%d expected=%d range=%d nb_queues=%d max_event=%d",
+                             ev->mbuf->seqn, index, expected_val, range,
+                             queue_count, MAX_EVENTS);
+       return 0;
+}
+
+static int
+test_multi_queue_priority(void)
+{
+       int i, max_evts_roundoff;
+       /* See validate_queue_priority() comments for priority validate logic */
+       uint32_t queue_count;
+       struct rte_mbuf *m;
+       uint8_t queue;
+
+       RTE_TEST_ASSERT_SUCCESS(rte_event_dev_attr_get(evdev,
+                               RTE_EVENT_DEV_ATTR_QUEUE_COUNT, &queue_count),
+                               "Queue count get failed");
+       if (queue_count > 8)
+               queue_count = 8;
+       max_evts_roundoff  = MAX_EVENTS / queue_count;
+       max_evts_roundoff *= queue_count;
+
+       for (i = 0; i < max_evts_roundoff; i++) {
+               struct rte_event ev = {.event = 0, .u64 = 0};
+
+               m = rte_pktmbuf_alloc(eventdev_test_mempool);
+               RTE_TEST_ASSERT_NOT_NULL(m, "mempool alloc failed");
+
+               m->seqn = i;
+               queue = i % queue_count;
+               update_event_and_validation_attr(m, &ev, 0, RTE_EVENT_TYPE_CPU,
+                                                0, RTE_SCHED_TYPE_PARALLEL,
+                                                queue, 0);
+               rte_event_enqueue_burst(evdev, 0, &ev, 1);
+       }
+
+       return consume_events(0, max_evts_roundoff, validate_queue_priority);
+}
+
+static int
+worker_multi_port_fn(void *arg)
+{
+       struct test_core_param *param = arg;
+       rte_atomic32_t *total_events = param->total_events;
+       uint8_t port = param->port;
+       uint16_t valid_event;
+       struct rte_event ev;
+       int ret;
+
+       while (rte_atomic32_read(total_events) > 0) {
+               valid_event = rte_event_dequeue_burst(evdev, port, &ev, 1, 0);
+               if (!valid_event)
+                       continue;
+
+               ret = validate_event(&ev);
+               RTE_TEST_ASSERT_SUCCESS(ret, "Failed to validate event");
+               rte_pktmbuf_free(ev.mbuf);
+               rte_atomic32_sub(total_events, 1);
+       }
+
+       return 0;
+}
+
+static inline int
+wait_workers_to_join(const rte_atomic32_t *count)
+{
+       uint64_t cycles, print_cycles;
+
+       cycles = rte_get_timer_cycles();
+       print_cycles = cycles;
+       while (rte_atomic32_read(count)) {
+               uint64_t new_cycles = rte_get_timer_cycles();
+
+               if (new_cycles - print_cycles > rte_get_timer_hz()) {
+                       otx2_err("Events %d", rte_atomic32_read(count));
+                       print_cycles = new_cycles;
+               }
+               if (new_cycles - cycles > rte_get_timer_hz() * 10000000000) {
+                       otx2_err("No schedules for seconds, deadlock (%d)",
+                                rte_atomic32_read(count));
+                       rte_event_dev_dump(evdev, stdout);
+                       cycles = new_cycles;
+                       return -1;
+               }
+       }
+       rte_eal_mp_wait_lcore();
+
+       return 0;
+}
+
+static inline int
+launch_workers_and_wait(int (*master_worker)(void *),
+                       int (*slave_workers)(void *), uint32_t total_events,
+                       uint8_t nb_workers, uint8_t sched_type)
+{
+       rte_atomic32_t atomic_total_events;
+       struct test_core_param *param;
+       uint64_t dequeue_tmo_ticks;
+       uint8_t port = 0;
+       int w_lcore;
+       int ret;
+
+       if (!nb_workers)
+               return 0;
+
+       rte_atomic32_set(&atomic_total_events, total_events);
+       seqn_list_init();
+
+       param = malloc(sizeof(struct test_core_param) * nb_workers);
+       if (!param)
+               return -1;
+
+       ret = rte_event_dequeue_timeout_ticks(evdev,
+                                             rte_rand() % 10000000/* 10ms */,
+                                             &dequeue_tmo_ticks);
+       if (ret) {
+               free(param);
+               return -1;
+       }
+
+       param[0].total_events = &atomic_total_events;
+       param[0].sched_type = sched_type;
+       param[0].port = 0;
+       param[0].dequeue_tmo_ticks = dequeue_tmo_ticks;
+       rte_wmb();
+
+       w_lcore = rte_get_next_lcore(
+                       /* start core */ -1,
+                       /* skip master */ 1,
+                       /* wrap */ 0);
+       rte_eal_remote_launch(master_worker, &param[0], w_lcore);
+
+       for (port = 1; port < nb_workers; port++) {
+               param[port].total_events = &atomic_total_events;
+               param[port].sched_type = sched_type;
+               param[port].port = port;
+               param[port].dequeue_tmo_ticks = dequeue_tmo_ticks;
+               rte_smp_wmb();
+               w_lcore = rte_get_next_lcore(w_lcore, 1, 0);
+               rte_eal_remote_launch(slave_workers, &param[port], w_lcore);
+       }
+
+       rte_smp_wmb();
+       ret = wait_workers_to_join(&atomic_total_events);
+       free(param);
+
+       return ret;
+}
+
+/*
+ * Generate a prescribed number of events and spread them across available
+ * queues. Dequeue the events through multiple ports and verify the enqueued
+ * event attributes
+ */
+static int
+test_multi_queue_enq_multi_port_deq(void)
+{
+       const unsigned int total_events = MAX_EVENTS;
+       uint32_t nr_ports;
+       int ret;
+
+       ret = generate_random_events(total_events);
+       if (ret)
+               return -1;
+
+       RTE_TEST_ASSERT_SUCCESS(rte_event_dev_attr_get(evdev,
+                               RTE_EVENT_DEV_ATTR_PORT_COUNT, &nr_ports),
+                               "Port count get failed");
+       nr_ports = RTE_MIN(nr_ports, rte_lcore_count() - 1);
+
+       if (!nr_ports) {
+               otx2_err("Not enough ports=%d or workers=%d", nr_ports,
+                        rte_lcore_count() - 1);
+               return 0;
+       }
+
+       return launch_workers_and_wait(worker_multi_port_fn,
+                                      worker_multi_port_fn, total_events,
+                                      nr_ports, 0xff /* invalid */);
+}
+
+static
+void flush(uint8_t dev_id, struct rte_event event, void *arg)
+{
+       unsigned int *count = arg;
+
+       RTE_SET_USED(dev_id);
+       if (event.event_type == RTE_EVENT_TYPE_CPU)
+               *count = *count + 1;
+}
+
+static int
+test_dev_stop_flush(void)
+{
+       unsigned int total_events = MAX_EVENTS, count = 0;
+       int ret;
+
+       ret = generate_random_events(total_events);
+       if (ret)
+               return -1;
+
+       ret = rte_event_dev_stop_flush_callback_register(evdev, flush, &count);
+       if (ret)
+               return -2;
+       rte_event_dev_stop(evdev);
+       ret = rte_event_dev_stop_flush_callback_register(evdev, NULL, NULL);
+       if (ret)
+               return -3;
+       RTE_TEST_ASSERT_EQUAL(total_events, count,
+                             "count mismatch total_events=%d count=%d",
+                             total_events, count);
+
+       return 0;
+}
+
+static int
+validate_queue_to_port_single_link(uint32_t index, uint8_t port,
+                                  struct rte_event *ev)
+{
+       RTE_SET_USED(index);
+       RTE_TEST_ASSERT_EQUAL(port, ev->queue_id,
+                             "queue mismatch enq=%d deq =%d",
+                             port, ev->queue_id);
+
+       return 0;
+}
+
+/*
+ * Link queue x to port x and check correctness of link by checking
+ * queue_id == x on dequeue on the specific port x
+ */
+static int
+test_queue_to_port_single_link(void)
+{
+       int i, nr_links, ret;
+       uint32_t queue_count;
+       uint32_t port_count;
+
+       RTE_TEST_ASSERT_SUCCESS(rte_event_dev_attr_get(evdev,
+                               RTE_EVENT_DEV_ATTR_PORT_COUNT, &port_count),
+                               "Port count get failed");
+
+       /* Unlink all connections that created in eventdev_setup */
+       for (i = 0; i < (int)port_count; i++) {
+               ret = rte_event_port_unlink(evdev, i, NULL, 0);
+               RTE_TEST_ASSERT(ret >= 0,
+                               "Failed to unlink all queues port=%d", i);
+       }
+
+       RTE_TEST_ASSERT_SUCCESS(rte_event_dev_attr_get(evdev,
+                               RTE_EVENT_DEV_ATTR_QUEUE_COUNT, &queue_count),
+                               "Queue count get failed");
+
+       nr_links = RTE_MIN(port_count, queue_count);
+       const unsigned int total_events = MAX_EVENTS / nr_links;
+
+       /* Link queue x to port x and inject events to queue x through port x */
+       for (i = 0; i < nr_links; i++) {
+               uint8_t queue = (uint8_t)i;
+
+               ret = rte_event_port_link(evdev, i, &queue, NULL, 1);
+               RTE_TEST_ASSERT(ret == 1, "Failed to link queue to port %d", i);
+
+               ret = inject_events(0x100 /*flow_id */,
+                                   RTE_EVENT_TYPE_CPU /* event_type */,
+                                   rte_rand() % 256 /* sub_event_type */,
+                                   rte_rand() % (RTE_SCHED_TYPE_PARALLEL + 1),
+                                   queue /* queue */, i /* port */,
+                                   total_events /* events */);
+               if (ret)
+                       return -1;
+       }
+
+       /* Verify the events generated from correct queue */
+       for (i = 0; i < nr_links; i++) {
+               ret = consume_events(i /* port */, total_events,
+                                    validate_queue_to_port_single_link);
+               if (ret)
+                       return -1;
+       }
+
+       return 0;
+}
+
+static int
+validate_queue_to_port_multi_link(uint32_t index, uint8_t port,
+                                 struct rte_event *ev)
+{
+       RTE_SET_USED(index);
+       RTE_TEST_ASSERT_EQUAL(port, (ev->queue_id & 0x1),
+                             "queue mismatch enq=%d deq =%d",
+                             port, ev->queue_id);
+
+       return 0;
+}
+
+/*
+ * Link all even number of queues to port 0 and all odd number of queues to
+ * port 1 and verify the link connection on dequeue
+ */
+static int
+test_queue_to_port_multi_link(void)
+{
+       int ret, port0_events = 0, port1_events = 0;
+       uint32_t nr_queues = 0;
+       uint32_t nr_ports = 0;
+       uint8_t queue, port;
+
+       RTE_TEST_ASSERT_SUCCESS(rte_event_dev_attr_get(evdev,
+                               RTE_EVENT_DEV_ATTR_QUEUE_COUNT, &nr_queues),
+                               "Queue count get failed");
+       RTE_TEST_ASSERT_SUCCESS(rte_event_dev_attr_get(evdev,
+                               RTE_EVENT_DEV_ATTR_QUEUE_COUNT, &nr_queues),
+                               "Queue count get failed");
+       RTE_TEST_ASSERT_SUCCESS(rte_event_dev_attr_get(evdev,
+                               RTE_EVENT_DEV_ATTR_PORT_COUNT, &nr_ports),
+                               "Port count get failed");
+
+       if (nr_ports < 2) {
+               otx2_err("Not enough ports to test ports=%d", nr_ports);
+               return 0;
+       }
+
+       /* Unlink all connections that created in eventdev_setup */
+       for (port = 0; port < nr_ports; port++) {
+               ret = rte_event_port_unlink(evdev, port, NULL, 0);
+               RTE_TEST_ASSERT(ret >= 0, "Failed to unlink all queues port=%d",
+                               port);
+       }
+
+       const unsigned int total_events = MAX_EVENTS / nr_queues;
+
+       /* Link all even number of queues to port0 and odd numbers to port 1*/
+       for (queue = 0; queue < nr_queues; queue++) {
+               port = queue & 0x1;
+               ret = rte_event_port_link(evdev, port, &queue, NULL, 1);
+               RTE_TEST_ASSERT(ret == 1, "Failed to link queue=%d to port=%d",
+                               queue, port);
+
+               ret = inject_events(0x100 /*flow_id */,
+                                   RTE_EVENT_TYPE_CPU /* event_type */,
+                                   rte_rand() % 256 /* sub_event_type */,
+                                   rte_rand() % (RTE_SCHED_TYPE_PARALLEL + 1),
+                                   queue /* queue */, port /* port */,
+                                   total_events /* events */);
+               if (ret)
+                       return -1;
+
+               if (port == 0)
+                       port0_events += total_events;
+               else
+                       port1_events += total_events;
+       }
+
+       ret = consume_events(0 /* port */, port0_events,
+                            validate_queue_to_port_multi_link);
+       if (ret)
+               return -1;
+       ret = consume_events(1 /* port */, port1_events,
+                            validate_queue_to_port_multi_link);
+       if (ret)
+               return -1;
+
+       return 0;
+}
+
+static int
+worker_flow_based_pipeline(void *arg)
+{
+       struct test_core_param *param = arg;
+       uint64_t dequeue_tmo_ticks = param->dequeue_tmo_ticks;
+       rte_atomic32_t *total_events = param->total_events;
+       uint8_t new_sched_type = param->sched_type;
+       uint8_t port = param->port;
+       uint16_t valid_event;
+       struct rte_event ev;
+
+       while (rte_atomic32_read(total_events) > 0) {
+               valid_event = rte_event_dequeue_burst(evdev, port, &ev, 1,
+                                                     dequeue_tmo_ticks);
+               if (!valid_event)
+                       continue;
+
+               /* Events from stage 0 */
+               if (ev.sub_event_type == 0) {
+                       /* Move to atomic flow to maintain the ordering */
+                       ev.flow_id = 0x2;
+                       ev.event_type = RTE_EVENT_TYPE_CPU;
+                       ev.sub_event_type = 1; /* stage 1 */
+                       ev.sched_type = new_sched_type;
+                       ev.op = RTE_EVENT_OP_FORWARD;
+                       rte_event_enqueue_burst(evdev, port, &ev, 1);
+               } else if (ev.sub_event_type == 1) { /* Events from stage 1*/
+                       if (seqn_list_update(ev.mbuf->seqn) == 0) {
+                               rte_pktmbuf_free(ev.mbuf);
+                               rte_atomic32_sub(total_events, 1);
+                       } else {
+                               otx2_err("Failed to update seqn_list");
+                               return -1;
+                       }
+               } else {
+                       otx2_err("Invalid ev.sub_event_type = %d",
+                                ev.sub_event_type);
+                       return -1;
+               }
+       }
+       return 0;
+}
+
+static int
+test_multiport_flow_sched_type_test(uint8_t in_sched_type,
+                                   uint8_t out_sched_type)
+{
+       const unsigned int total_events = MAX_EVENTS;
+       uint32_t nr_ports;
+       int ret;
+
+       RTE_TEST_ASSERT_SUCCESS(rte_event_dev_attr_get(evdev,
+                               RTE_EVENT_DEV_ATTR_PORT_COUNT, &nr_ports),
+                               "Port count get failed");
+       nr_ports = RTE_MIN(nr_ports, rte_lcore_count() - 1);
+
+       if (!nr_ports) {
+               otx2_err("Not enough ports=%d or workers=%d", nr_ports,
+                        rte_lcore_count() - 1);
+               return 0;
+       }
+
+       /* Injects events with m->seqn=0 to total_events */
+       ret = inject_events(0x1 /*flow_id */,
+                           RTE_EVENT_TYPE_CPU /* event_type */,
+                           0 /* sub_event_type (stage 0) */,
+                           in_sched_type,
+                           0 /* queue */,
+                           0 /* port */,
+                           total_events /* events */);
+       if (ret)
+               return -1;
+
+       rte_mb();
+       ret = launch_workers_and_wait(worker_flow_based_pipeline,
+                                     worker_flow_based_pipeline, total_events,
+                                     nr_ports, out_sched_type);
+       if (ret)
+               return -1;
+
+       if (in_sched_type != RTE_SCHED_TYPE_PARALLEL &&
+           out_sched_type == RTE_SCHED_TYPE_ATOMIC) {
+               /* Check the events order maintained or not */
+               return seqn_list_check(total_events);
+       }
+
+       return 0;
+}
+
+/* Multi port ordered to atomic transaction */
+static int
+test_multi_port_flow_ordered_to_atomic(void)
+{
+       /* Ingress event order test */
+       return test_multiport_flow_sched_type_test(RTE_SCHED_TYPE_ORDERED,
+                                                  RTE_SCHED_TYPE_ATOMIC);
+}
+
+static int
+test_multi_port_flow_ordered_to_ordered(void)
+{
+       return test_multiport_flow_sched_type_test(RTE_SCHED_TYPE_ORDERED,
+                                                  RTE_SCHED_TYPE_ORDERED);
+}
+
+static int
+test_multi_port_flow_ordered_to_parallel(void)
+{
+       return test_multiport_flow_sched_type_test(RTE_SCHED_TYPE_ORDERED,
+                                                  RTE_SCHED_TYPE_PARALLEL);
+}
+
+static int
+test_multi_port_flow_atomic_to_atomic(void)
+{
+       /* Ingress event order test */
+       return test_multiport_flow_sched_type_test(RTE_SCHED_TYPE_ATOMIC,
+                                                  RTE_SCHED_TYPE_ATOMIC);
+}
+
+static int
+test_multi_port_flow_atomic_to_ordered(void)
+{
+       return test_multiport_flow_sched_type_test(RTE_SCHED_TYPE_ATOMIC,
+                                                  RTE_SCHED_TYPE_ORDERED);
+}
+
+static int
+test_multi_port_flow_atomic_to_parallel(void)
+{
+       return test_multiport_flow_sched_type_test(RTE_SCHED_TYPE_ATOMIC,
+                                                  RTE_SCHED_TYPE_PARALLEL);
+}
+
+static int
+test_multi_port_flow_parallel_to_atomic(void)
+{
+       return test_multiport_flow_sched_type_test(RTE_SCHED_TYPE_PARALLEL,
+                                                  RTE_SCHED_TYPE_ATOMIC);
+}
+
+static int
+test_multi_port_flow_parallel_to_ordered(void)
+{
+       return test_multiport_flow_sched_type_test(RTE_SCHED_TYPE_PARALLEL,
+                                                  RTE_SCHED_TYPE_ORDERED);
+}
+
+static int
+test_multi_port_flow_parallel_to_parallel(void)
+{
+       return test_multiport_flow_sched_type_test(RTE_SCHED_TYPE_PARALLEL,
+                                                  RTE_SCHED_TYPE_PARALLEL);
+}
+
+static int
+worker_group_based_pipeline(void *arg)
+{
+       struct test_core_param *param = arg;
+       uint64_t dequeue_tmo_ticks = param->dequeue_tmo_ticks;
+       rte_atomic32_t *total_events = param->total_events;
+       uint8_t new_sched_type = param->sched_type;
+       uint8_t port = param->port;
+       uint16_t valid_event;
+       struct rte_event ev;
+
+       while (rte_atomic32_read(total_events) > 0) {
+               valid_event = rte_event_dequeue_burst(evdev, port, &ev, 1,
+                                                     dequeue_tmo_ticks);
+               if (!valid_event)
+                       continue;
+
+               /* Events from stage 0(group 0) */
+               if (ev.queue_id == 0) {
+                       /* Move to atomic flow to maintain the ordering */
+                       ev.flow_id = 0x2;
+                       ev.event_type = RTE_EVENT_TYPE_CPU;
+                       ev.sched_type = new_sched_type;
+                       ev.queue_id = 1; /* Stage 1*/
+                       ev.op = RTE_EVENT_OP_FORWARD;
+                       rte_event_enqueue_burst(evdev, port, &ev, 1);
+               } else if (ev.queue_id == 1) { /* Events from stage 1(group 1)*/
+                       if (seqn_list_update(ev.mbuf->seqn) == 0) {
+                               rte_pktmbuf_free(ev.mbuf);
+                               rte_atomic32_sub(total_events, 1);
+                       } else {
+                               otx2_err("Failed to update seqn_list");
+                               return -1;
+                       }
+               } else {
+                       otx2_err("Invalid ev.queue_id = %d", ev.queue_id);
+                       return -1;
+               }
+       }
+
+       return 0;
+}
+
+static int
+test_multiport_queue_sched_type_test(uint8_t in_sched_type,
+                                    uint8_t out_sched_type)
+{
+       const unsigned int total_events = MAX_EVENTS;
+       uint32_t queue_count;
+       uint32_t nr_ports;
+       int ret;
+
+       RTE_TEST_ASSERT_SUCCESS(rte_event_dev_attr_get(evdev,
+                               RTE_EVENT_DEV_ATTR_PORT_COUNT, &nr_ports),
+                               "Port count get failed");
+
+       nr_ports = RTE_MIN(nr_ports, rte_lcore_count() - 1);
+
+       RTE_TEST_ASSERT_SUCCESS(rte_event_dev_attr_get(evdev,
+                               RTE_EVENT_DEV_ATTR_QUEUE_COUNT, &queue_count),
+                               "Queue count get failed");
+       if (queue_count < 2 ||  !nr_ports) {
+               otx2_err("Not enough queues=%d ports=%d or workers=%d",
+                        queue_count, nr_ports,
+                        rte_lcore_count() - 1);
+               return 0;
+       }
+
+       /* Injects events with m->seqn=0 to total_events */
+       ret = inject_events(0x1 /*flow_id */,
+                           RTE_EVENT_TYPE_CPU /* event_type */,
+                           0 /* sub_event_type (stage 0) */,
+                           in_sched_type,
+                           0 /* queue */,
+                           0 /* port */,
+                           total_events /* events */);
+       if (ret)
+               return -1;
+
+       ret = launch_workers_and_wait(worker_group_based_pipeline,
+                                     worker_group_based_pipeline, total_events,
+                                     nr_ports, out_sched_type);
+       if (ret)
+               return -1;
+
+       if (in_sched_type != RTE_SCHED_TYPE_PARALLEL &&
+           out_sched_type == RTE_SCHED_TYPE_ATOMIC) {
+               /* Check the events order maintained or not */
+               return seqn_list_check(total_events);
+       }
+
+       return 0;
+}
+
+static int
+test_multi_port_queue_ordered_to_atomic(void)
+{
+       /* Ingress event order test */
+       return test_multiport_queue_sched_type_test(RTE_SCHED_TYPE_ORDERED,
+                                                   RTE_SCHED_TYPE_ATOMIC);
+}
+
+static int
+test_multi_port_queue_ordered_to_ordered(void)
+{
+       return test_multiport_queue_sched_type_test(RTE_SCHED_TYPE_ORDERED,
+                                                   RTE_SCHED_TYPE_ORDERED);
+}
+
+static int
+test_multi_port_queue_ordered_to_parallel(void)
+{
+       return test_multiport_queue_sched_type_test(RTE_SCHED_TYPE_ORDERED,
+                                                   RTE_SCHED_TYPE_PARALLEL);
+}
+
+static int
+test_multi_port_queue_atomic_to_atomic(void)
+{
+       /* Ingress event order test */
+       return test_multiport_queue_sched_type_test(RTE_SCHED_TYPE_ATOMIC,
+                                                   RTE_SCHED_TYPE_ATOMIC);
+}
+
+static int
+test_multi_port_queue_atomic_to_ordered(void)
+{
+       return test_multiport_queue_sched_type_test(RTE_SCHED_TYPE_ATOMIC,
+                                                   RTE_SCHED_TYPE_ORDERED);
+}
+
+static int
+test_multi_port_queue_atomic_to_parallel(void)
+{
+       return test_multiport_queue_sched_type_test(RTE_SCHED_TYPE_ATOMIC,
+                                                   RTE_SCHED_TYPE_PARALLEL);
+}
+
+static int
+test_multi_port_queue_parallel_to_atomic(void)
+{
+       return test_multiport_queue_sched_type_test(RTE_SCHED_TYPE_PARALLEL,
+                                                   RTE_SCHED_TYPE_ATOMIC);
+}
+
+static int
+test_multi_port_queue_parallel_to_ordered(void)
+{
+       return test_multiport_queue_sched_type_test(RTE_SCHED_TYPE_PARALLEL,
+                                                   RTE_SCHED_TYPE_ORDERED);
+}
+
+static int
+test_multi_port_queue_parallel_to_parallel(void)
+{
+       return test_multiport_queue_sched_type_test(RTE_SCHED_TYPE_PARALLEL,
+                                                   RTE_SCHED_TYPE_PARALLEL);
+}
+
+static int
+worker_flow_based_pipeline_max_stages_rand_sched_type(void *arg)
+{
+       struct test_core_param *param = arg;
+       rte_atomic32_t *total_events = param->total_events;
+       uint8_t port = param->port;
+       uint16_t valid_event;
+       struct rte_event ev;
+
+       while (rte_atomic32_read(total_events) > 0) {
+               valid_event = rte_event_dequeue_burst(evdev, port, &ev, 1, 0);
+               if (!valid_event)
+                       continue;
+
+               if (ev.sub_event_type == 255) { /* last stage */
+                       rte_pktmbuf_free(ev.mbuf);
+                       rte_atomic32_sub(total_events, 1);
+               } else {
+                       ev.event_type = RTE_EVENT_TYPE_CPU;
+                       ev.sub_event_type++;
+                       ev.sched_type =
+                               rte_rand() % (RTE_SCHED_TYPE_PARALLEL + 1);
+                       ev.op = RTE_EVENT_OP_FORWARD;
+                       rte_event_enqueue_burst(evdev, port, &ev, 1);
+               }
+       }
+
+       return 0;
+}
+
+static int
+launch_multi_port_max_stages_random_sched_type(int (*fn)(void *))
+{
+       uint32_t nr_ports;
+       int ret;
+
+       RTE_TEST_ASSERT_SUCCESS(rte_event_dev_attr_get(evdev,
+                               RTE_EVENT_DEV_ATTR_PORT_COUNT, &nr_ports),
+                               "Port count get failed");
+       nr_ports = RTE_MIN(nr_ports, rte_lcore_count() - 1);
+
+       if (!nr_ports) {
+               otx2_err("Not enough ports=%d or workers=%d",
+                        nr_ports, rte_lcore_count() - 1);
+               return 0;
+       }
+
+       /* Injects events with m->seqn=0 to total_events */
+       ret = inject_events(0x1 /*flow_id */,
+                           RTE_EVENT_TYPE_CPU /* event_type */,
+                           0 /* sub_event_type (stage 0) */,
+                           rte_rand() %
+                               (RTE_SCHED_TYPE_PARALLEL + 1) /* sched_type */,
+                           0 /* queue */,
+                           0 /* port */,
+                           MAX_EVENTS /* events */);
+       if (ret)
+               return -1;
+
+       return launch_workers_and_wait(fn, fn, MAX_EVENTS, nr_ports,
+                                      0xff /* invalid */);
+}
+
+/* Flow based pipeline with maximum stages with random sched type */
+static int
+test_multi_port_flow_max_stages_random_sched_type(void)
+{
+       return launch_multi_port_max_stages_random_sched_type(
+               worker_flow_based_pipeline_max_stages_rand_sched_type);
+}
+
+static int
+worker_queue_based_pipeline_max_stages_rand_sched_type(void *arg)
+{
+       struct test_core_param *param = arg;
+       uint8_t port = param->port;
+       uint32_t queue_count;
+       uint16_t valid_event;
+       struct rte_event ev;
+
+       RTE_TEST_ASSERT_SUCCESS(rte_event_dev_attr_get(evdev,
+                               RTE_EVENT_DEV_ATTR_QUEUE_COUNT, &queue_count),
+                               "Queue count get failed");
+       uint8_t nr_queues = queue_count;
+       rte_atomic32_t *total_events = param->total_events;
+
+       while (rte_atomic32_read(total_events) > 0) {
+               valid_event = rte_event_dequeue_burst(evdev, port, &ev, 1, 0);
+               if (!valid_event)
+                       continue;
+
+               if (ev.queue_id == nr_queues - 1) { /* last stage */
+                       rte_pktmbuf_free(ev.mbuf);
+                       rte_atomic32_sub(total_events, 1);
+               } else {
+                       ev.event_type = RTE_EVENT_TYPE_CPU;
+                       ev.queue_id++;
+                       ev.sched_type =
+                               rte_rand() % (RTE_SCHED_TYPE_PARALLEL + 1);
+                       ev.op = RTE_EVENT_OP_FORWARD;
+                       rte_event_enqueue_burst(evdev, port, &ev, 1);
+               }
+       }
+
+       return 0;
+}
+
+/* Queue based pipeline with maximum stages with random sched type */
+static int
+test_multi_port_queue_max_stages_random_sched_type(void)
+{
+       return launch_multi_port_max_stages_random_sched_type(
+               worker_queue_based_pipeline_max_stages_rand_sched_type);
+}
+
+static int
+worker_mixed_pipeline_max_stages_rand_sched_type(void *arg)
+{
+       struct test_core_param *param = arg;
+       uint8_t port = param->port;
+       uint32_t queue_count;
+       uint16_t valid_event;
+       struct rte_event ev;
+
+       RTE_TEST_ASSERT_SUCCESS(rte_event_dev_attr_get(evdev,
+                               RTE_EVENT_DEV_ATTR_QUEUE_COUNT, &queue_count),
+                               "Queue count get failed");
+       uint8_t nr_queues = queue_count;
+       rte_atomic32_t *total_events = param->total_events;
+
+       while (rte_atomic32_read(total_events) > 0) {
+               valid_event = rte_event_dequeue_burst(evdev, port, &ev, 1, 0);
+               if (!valid_event)
+                       continue;
+
+               if (ev.queue_id == nr_queues - 1) { /* Last stage */
+                       rte_pktmbuf_free(ev.mbuf);
+                       rte_atomic32_sub(total_events, 1);
+               } else {
+                       ev.event_type = RTE_EVENT_TYPE_CPU;
+                       ev.queue_id++;
+                       ev.sub_event_type = rte_rand() % 256;
+                       ev.sched_type =
+                               rte_rand() % (RTE_SCHED_TYPE_PARALLEL + 1);
+                       ev.op = RTE_EVENT_OP_FORWARD;
+                       rte_event_enqueue_burst(evdev, port, &ev, 1);
+               }
+       }
+
+       return 0;
+}
+
+/* Queue and flow based pipeline with maximum stages with random sched type */
+static int
+test_multi_port_mixed_max_stages_random_sched_type(void)
+{
+       return launch_multi_port_max_stages_random_sched_type(
+               worker_mixed_pipeline_max_stages_rand_sched_type);
+}
+
+static int
+worker_ordered_flow_producer(void *arg)
+{
+       struct test_core_param *param = arg;
+       uint8_t port = param->port;
+       struct rte_mbuf *m;
+       int counter = 0;
+
+       while (counter < NUM_PACKETS) {
+               m = rte_pktmbuf_alloc(eventdev_test_mempool);
+               if (m == NULL)
+                       continue;
+
+               m->seqn = counter++;
+
+               struct rte_event ev = {.event = 0, .u64 = 0};
+
+               ev.flow_id = 0x1; /* Generate a fat flow */
+               ev.sub_event_type = 0;
+               /* Inject the new event */
+               ev.op = RTE_EVENT_OP_NEW;
+               ev.event_type = RTE_EVENT_TYPE_CPU;
+               ev.sched_type = RTE_SCHED_TYPE_ORDERED;
+               ev.queue_id = 0;
+               ev.mbuf = m;
+               rte_event_enqueue_burst(evdev, port, &ev, 1);
+       }
+
+       return 0;
+}
+
+static inline int
+test_producer_consumer_ingress_order_test(int (*fn)(void *))
+{
+       uint32_t nr_ports;
+
+       RTE_TEST_ASSERT_SUCCESS(rte_event_dev_attr_get(evdev,
+                               RTE_EVENT_DEV_ATTR_PORT_COUNT, &nr_ports),
+                               "Port count get failed");
+       nr_ports = RTE_MIN(nr_ports, rte_lcore_count() - 1);
+
+       if (rte_lcore_count() < 3 || nr_ports < 2) {
+               otx2_err("### Not enough cores for test.");
+               return 0;
+       }
+
+       launch_workers_and_wait(worker_ordered_flow_producer, fn,
+                               NUM_PACKETS, nr_ports, RTE_SCHED_TYPE_ATOMIC);
+       /* Check the events order maintained or not */
+       return seqn_list_check(NUM_PACKETS);
+}
+
+/* Flow based producer consumer ingress order test */
+static int
+test_flow_producer_consumer_ingress_order_test(void)
+{
+       return test_producer_consumer_ingress_order_test(
+                               worker_flow_based_pipeline);
+}
+
+/* Queue based producer consumer ingress order test */
+static int
+test_queue_producer_consumer_ingress_order_test(void)
+{
+       return test_producer_consumer_ingress_order_test(
+                               worker_group_based_pipeline);
+}
+
+static void octeontx_test_run(int (*setup)(void), void (*tdown)(void),
+                             int (*test)(void), const char *name)
+{
+       if (setup() < 0) {
+               printf("Error setting up test %s", name);
+               unsupported++;
+       } else {
+               if (test() < 0) {
+                       failed++;
+                       printf("+ TestCase [%2d] : %s failed\n", total, name);
+               } else {
+                       passed++;
+                       printf("+ TestCase [%2d] : %s succeeded\n", total,
+                              name);
+               }
+       }
+
+       total++;
+       tdown();
+}
+
+int
+otx2_sso_selftest(void)
+{
+       testsuite_setup();
+
+       OCTEONTX2_TEST_RUN(eventdev_setup, eventdev_teardown,
+                          test_simple_enqdeq_ordered);
+       OCTEONTX2_TEST_RUN(eventdev_setup, eventdev_teardown,
+                          test_simple_enqdeq_atomic);
+       OCTEONTX2_TEST_RUN(eventdev_setup, eventdev_teardown,
+                          test_simple_enqdeq_parallel);
+       OCTEONTX2_TEST_RUN(eventdev_setup, eventdev_teardown,
+                          test_multi_queue_enq_single_port_deq);
+       OCTEONTX2_TEST_RUN(eventdev_setup, eventdev_teardown,
+                          test_dev_stop_flush);
+       OCTEONTX2_TEST_RUN(eventdev_setup, eventdev_teardown,
+                          test_multi_queue_enq_multi_port_deq);
+       OCTEONTX2_TEST_RUN(eventdev_setup, eventdev_teardown,
+                          test_queue_to_port_single_link);
+       OCTEONTX2_TEST_RUN(eventdev_setup, eventdev_teardown,
+                          test_queue_to_port_multi_link);
+       OCTEONTX2_TEST_RUN(eventdev_setup, eventdev_teardown,
+                          test_multi_port_flow_ordered_to_atomic);
+       OCTEONTX2_TEST_RUN(eventdev_setup, eventdev_teardown,
+                          test_multi_port_flow_ordered_to_ordered);
+       OCTEONTX2_TEST_RUN(eventdev_setup, eventdev_teardown,
+                          test_multi_port_flow_ordered_to_parallel);
+       OCTEONTX2_TEST_RUN(eventdev_setup, eventdev_teardown,
+                          test_multi_port_flow_atomic_to_atomic);
+       OCTEONTX2_TEST_RUN(eventdev_setup, eventdev_teardown,
+                          test_multi_port_flow_atomic_to_ordered);
+       OCTEONTX2_TEST_RUN(eventdev_setup, eventdev_teardown,
+                          test_multi_port_flow_atomic_to_parallel);
+       OCTEONTX2_TEST_RUN(eventdev_setup, eventdev_teardown,
+                          test_multi_port_flow_parallel_to_atomic);
+       OCTEONTX2_TEST_RUN(eventdev_setup, eventdev_teardown,
+                          test_multi_port_flow_parallel_to_ordered);
+       OCTEONTX2_TEST_RUN(eventdev_setup, eventdev_teardown,
+                          test_multi_port_flow_parallel_to_parallel);
+       OCTEONTX2_TEST_RUN(eventdev_setup, eventdev_teardown,
+                          test_multi_port_queue_ordered_to_atomic);
+       OCTEONTX2_TEST_RUN(eventdev_setup, eventdev_teardown,
+                          test_multi_port_queue_ordered_to_ordered);
+       OCTEONTX2_TEST_RUN(eventdev_setup, eventdev_teardown,
+                          test_multi_port_queue_ordered_to_parallel);
+       OCTEONTX2_TEST_RUN(eventdev_setup, eventdev_teardown,
+                          test_multi_port_queue_atomic_to_atomic);
+       OCTEONTX2_TEST_RUN(eventdev_setup, eventdev_teardown,
+                          test_multi_port_queue_atomic_to_ordered);
+       OCTEONTX2_TEST_RUN(eventdev_setup, eventdev_teardown,
+                          test_multi_port_queue_atomic_to_parallel);
+       OCTEONTX2_TEST_RUN(eventdev_setup, eventdev_teardown,
+                          test_multi_port_queue_parallel_to_atomic);
+       OCTEONTX2_TEST_RUN(eventdev_setup, eventdev_teardown,
+                          test_multi_port_queue_parallel_to_ordered);
+       OCTEONTX2_TEST_RUN(eventdev_setup, eventdev_teardown,
+                          test_multi_port_queue_parallel_to_parallel);
+       OCTEONTX2_TEST_RUN(eventdev_setup, eventdev_teardown,
+                          test_multi_port_flow_max_stages_random_sched_type);
+       OCTEONTX2_TEST_RUN(eventdev_setup, eventdev_teardown,
+                          test_multi_port_queue_max_stages_random_sched_type);
+       OCTEONTX2_TEST_RUN(eventdev_setup, eventdev_teardown,
+                          test_multi_port_mixed_max_stages_random_sched_type);
+       OCTEONTX2_TEST_RUN(eventdev_setup, eventdev_teardown,
+                          test_flow_producer_consumer_ingress_order_test);
+       OCTEONTX2_TEST_RUN(eventdev_setup, eventdev_teardown,
+                          test_queue_producer_consumer_ingress_order_test);
+       OCTEONTX2_TEST_RUN(eventdev_setup_priority, eventdev_teardown,
+                          test_multi_queue_priority);
+       OCTEONTX2_TEST_RUN(eventdev_setup_dequeue_timeout, eventdev_teardown,
+                          test_multi_port_flow_ordered_to_atomic);
+       OCTEONTX2_TEST_RUN(eventdev_setup_dequeue_timeout, eventdev_teardown,
+                          test_multi_port_queue_ordered_to_atomic);
+       printf("Total tests   : %d\n", total);
+       printf("Passed        : %d\n", passed);
+       printf("Failed        : %d\n", failed);
+       printf("Not supported : %d\n", unsupported);
+
+       testsuite_teardown();
+
+       if (failed)
+               return -1;
+
+       return 0;
+}