examples/ip_pipeline: add application thread
authorMaciej Gajdzica <maciejx.t.gajdzica@intel.com>
Tue, 7 Jul 2015 08:09:30 +0000 (10:09 +0200)
committerThomas Monjalon <thomas.monjalon@6wind.com>
Thu, 9 Jul 2015 15:25:46 +0000 (17:25 +0200)
Application thread runs pipelines on assigned cores.

Signed-off-by: Maciej Gajdzica <maciejx.t.gajdzica@intel.com>
Acked-by: Cristian Dumitrescu <cristian.dumitrescu@intel.com>
examples/ip_pipeline/Makefile
examples/ip_pipeline/main.c
examples/ip_pipeline/thread.c [new file with mode: 0644]

index 9ce80a8..f255338 100644 (file)
@@ -53,6 +53,7 @@ SRCS-$(CONFIG_RTE_LIBRTE_PIPELINE) += config_parse.c
 SRCS-$(CONFIG_RTE_LIBRTE_PIPELINE) += config_parse_tm.c
 SRCS-$(CONFIG_RTE_LIBRTE_PIPELINE) += config_check.c
 SRCS-$(CONFIG_RTE_LIBRTE_PIPELINE) += init.c
+SRCS-$(CONFIG_RTE_LIBRTE_PIPELINE) += thread.c
 SRCS-$(CONFIG_RTE_LIBRTE_PIPELINE) += cpu_core_map.c
 
 SRCS-$(CONFIG_RTE_LIBRTE_PIPELINE) += pipeline_common_be.c
index ef68c86..862e2f2 100644 (file)
@@ -52,5 +52,11 @@ main(int argc, char **argv)
        /* Init */
        app_init(&app);
 
+       /* Run-time */
+       rte_eal_mp_remote_launch(
+               app_thread,
+               (void *) &app,
+               CALL_MASTER);
+
        return 0;
 }
diff --git a/examples/ip_pipeline/thread.c b/examples/ip_pipeline/thread.c
new file mode 100644 (file)
index 0000000..b2a8656
--- /dev/null
@@ -0,0 +1,110 @@
+/*-
+ *   BSD LICENSE
+ *
+ *   Copyright(c) 2010-2015 Intel Corporation. All rights reserved.
+ *   All rights reserved.
+ *
+ *   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 Intel Corporation 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 <rte_common.h>
+#include <rte_cycles.h>
+#include <rte_pipeline.h>
+
+#include "pipeline_common_be.h"
+#include "app.h"
+
+int app_thread(void *arg)
+{
+       struct app_params *app = (struct app_params *) arg;
+       uint32_t core_id = rte_lcore_id(), i, j;
+       struct app_thread_data *t = &app->thread_data[core_id];
+       uint32_t n_regular = RTE_MIN(t->n_regular, RTE_DIM(t->regular));
+       uint32_t n_custom = RTE_MIN(t->n_custom, RTE_DIM(t->custom));
+
+       for (i = 0; ; i++) {
+               /* Run regular pipelines */
+               for (j = 0; j < n_regular; j++) {
+                       struct app_thread_pipeline_data *data = &t->regular[j];
+                       struct pipeline *p = data->be;
+
+                       rte_pipeline_run(p->p);
+               }
+
+               /* Run custom pipelines */
+               for (j = 0; j < n_custom; j++) {
+                       struct app_thread_pipeline_data *data = &t->custom[j];
+
+                       data->f_run(data->be);
+               }
+
+               /* Timer */
+               if ((i & 0xF) == 0) {
+                       uint64_t time = rte_get_tsc_cycles();
+                       uint64_t t_deadline = UINT64_MAX;
+
+                       if (time < t->deadline)
+                               continue;
+
+                       /* Timer for regular pipelines */
+                       for (j = 0; j < n_regular; j++) {
+                               struct app_thread_pipeline_data *data =
+                                       &t->regular[j];
+                               uint64_t p_deadline = data->deadline;
+
+                               if (p_deadline <= time) {
+                                       data->f_timer(data->be);
+                                       p_deadline = time + data->timer_period;
+                                       data->deadline = p_deadline;
+                               }
+
+                               if (p_deadline < t_deadline)
+                                       t_deadline = p_deadline;
+                       }
+
+                       /* Timer for custom pipelines */
+                       for (j = 0; j < n_custom; j++) {
+                               struct app_thread_pipeline_data *data =
+                                       &t->custom[j];
+                               uint64_t p_deadline = data->deadline;
+
+                               if (p_deadline <= time) {
+                                       data->f_timer(data->be);
+                                       p_deadline = time + data->timer_period;
+                                       data->deadline = p_deadline;
+                               }
+
+                               if (p_deadline < t_deadline)
+                                       t_deadline = p_deadline;
+                       }
+
+                       t->deadline = t_deadline;
+               }
+       }
+
+       return 0;
+}