1 .. SPDX-License-Identifier: BSD-3-Clause
2 Copyright(c) 2010-2014 Intel Corporation.
6 L2 Forwarding Eventdev Sample Application
7 =========================================
9 The L2 Forwarding eventdev sample application is a simple example of packet
10 processing using the Data Plane Development Kit (DPDK) to demonstrate usage of
11 poll and event mode packet I/O mechanism.
16 The L2 Forwarding eventdev sample application, performs L2 forwarding for each
17 packet that is received on an RX_PORT. The destination port is the adjacent port
18 from the enabled portmask, that is, if the first four ports are enabled (portmask=0x0f),
19 ports 1 and 2 forward into each other, and ports 3 and 4 forward into each other.
20 Also, if MAC addresses updating is enabled, the MAC addresses are affected as follows:
22 * The source MAC address is replaced by the TX_PORT MAC address
24 * The destination MAC address is replaced by 02:00:00:00:00:TX_PORT_ID
26 Application receives packets from RX_PORT using below mentioned methods:
30 * Eventdev mode (default)
32 This application can be used to benchmark performance using a traffic-generator,
33 as shown in the :numref:`figure_l2fwd_event_benchmark_setup`.
35 .. _figure_l2fwd_event_benchmark_setup:
37 .. figure:: img/l2_fwd_benchmark_setup.*
39 Performance Benchmark Setup (Basic Environment)
41 Compiling the Application
42 -------------------------
44 To compile the sample application see :doc:`compiling`.
46 The application is located in the ``l2fwd-event`` sub-directory.
48 Running the Application
49 -----------------------
51 The application requires a number of command line options:
53 .. code-block:: console
55 ./build/l2fwd-event [EAL options] -- -p PORTMASK [-q NQ] --[no-]mac-updating --mode=MODE --eventq-sched=SCHED_MODE
59 * p PORTMASK: A hexadecimal bitmask of the ports to configure
61 * q NQ: A number of queues (=ports) per lcore (default is 1)
63 * --[no-]mac-updating: Enable or disable MAC addresses updating (enabled by default).
65 * --mode=MODE: Packet transfer mode for I/O, poll or eventdev. Eventdev by default.
67 * --eventq-sched=SCHED_MODE: Event queue schedule mode, Ordered, Atomic or Parallel. Atomic by default.
69 * --config: Configure forwarding port pair mapping. Alternate port pairs by default.
71 Sample usage commands are given below to run the application into different mode:
73 Poll mode with 4 lcores, 16 ports and 8 RX queues per lcore and MAC address updating enabled,
76 .. code-block:: console
78 ./build/l2fwd-event -l 0-3 -n 4 -- -q 8 -p ffff --mode=poll
80 Eventdev mode with 4 lcores, 16 ports , sched method ordered and MAC address updating enabled,
83 .. code-block:: console
85 ./build/l2fwd-event -l 0-3 -n 4 -- -p ffff --eventq-sched=ordered
89 .. code-block:: console
91 ./build/l2fwd-event -l 0-3 -n 4 -- -q 8 -p ffff --mode=eventdev --eventq-sched=ordered
93 Refer to the *DPDK Getting Started Guide* for general information on running
94 applications and the Environment Abstraction Layer (EAL) options.
96 To run application with S/W scheduler, it uses following DPDK services:
99 * Rx adapter service function
100 * Tx adapter service function
102 Application needs service cores to run above mentioned services. Service cores
103 must be provided as EAL parameters along with the --vdev=event_sw0 to enable S/W
104 scheduler. Following is the sample command:
106 .. code-block:: console
108 ./build/l2fwd-event -l 0-7 -s 0-3 -n 4 --vdev event_sw0 -- -q 8 -p ffff --mode=eventdev --eventq-sched=ordered
113 The following sections provide some explanation of the code.
115 .. _l2_fwd_event_app_cmd_arguments:
117 Command Line Arguments
118 ~~~~~~~~~~~~~~~~~~~~~~
120 The L2 Forwarding eventdev sample application takes specific parameters,
121 in addition to Environment Abstraction Layer (EAL) arguments.
122 The preferred way to parse parameters is to use the getopt() function,
123 since it is part of a well-defined and portable library.
125 The parsing of arguments is done in the **l2fwd_parse_args()** function for non
126 eventdev parameters and in **parse_eventdev_args()** for eventdev parameters.
127 The method of argument parsing is not described here. Refer to the
128 *glibc getopt(3)* man page for details.
130 EAL arguments are parsed first, then application-specific arguments.
131 This is done at the beginning of the main() function and eventdev parameters
132 are parsed in eventdev_resource_setup() function during eventdev setup:
138 ret = rte_eal_init(argc, argv);
140 rte_panic("Invalid EAL arguments\n");
145 /* parse application arguments (after the EAL ones) */
147 ret = l2fwd_parse_args(argc, argv);
149 rte_panic("Invalid L2FWD arguments\n");
154 /* Parse eventdev command line options */
155 ret = parse_eventdev_args(argc, argv);
162 .. _l2_fwd_event_app_mbuf_init:
164 Mbuf Pool Initialization
165 ~~~~~~~~~~~~~~~~~~~~~~~~
167 Once the arguments are parsed, the mbuf pool is created.
168 The mbuf pool contains a set of mbuf objects that will be used by the driver
169 and the application to store network packet data:
173 /* create the mbuf pool */
175 l2fwd_pktmbuf_pool = rte_pktmbuf_pool_create("mbuf_pool", NB_MBUF,
176 MEMPOOL_CACHE_SIZE, 0,
177 RTE_MBUF_DEFAULT_BUF_SIZE,
179 if (l2fwd_pktmbuf_pool == NULL)
180 rte_panic("Cannot init mbuf pool\n");
182 The rte_mempool is a generic structure used to handle pools of objects.
183 In this case, it is necessary to create a pool that will be used by the driver.
184 The number of allocated pkt mbufs is NB_MBUF, with a data room size of
185 RTE_MBUF_DEFAULT_BUF_SIZE each.
186 A per-lcore cache of 32 mbufs is kept.
187 The memory is allocated in NUMA socket 0,
188 but it is possible to extend this code to allocate one mbuf pool per socket.
190 The rte_pktmbuf_pool_create() function uses the default mbuf pool and mbuf
191 initializers, respectively rte_pktmbuf_pool_init() and rte_pktmbuf_init().
192 An advanced application may want to use the mempool API to create the
193 mbuf pool with more control.
195 .. _l2_fwd_event_app_drv_init:
197 Driver Initialization
198 ~~~~~~~~~~~~~~~~~~~~~
200 The main part of the code in the main() function relates to the initialization
201 of the driver. To fully understand this code, it is recommended to study the
202 chapters that related to the Poll Mode and Event mode Driver in the
203 *DPDK Programmer's Guide* - Rel 1.4 EAR and the *DPDK API Reference*.
207 /* reset l2fwd_dst_ports */
209 for (portid = 0; portid < RTE_MAX_ETHPORTS; portid++)
210 l2fwd_dst_ports[portid] = 0;
215 * Each logical core is assigned a dedicated TX queue on each port.
218 RTE_ETH_FOREACH_DEV(portid) {
219 /* skip ports that are not enabled */
221 if ((l2fwd_enabled_port_mask & (1 << portid)) == 0)
224 if (nb_ports_in_mask % 2) {
225 l2fwd_dst_ports[portid] = last_port;
226 l2fwd_dst_ports[last_port] = portid;
233 rte_eth_dev_info_get((uint8_t) portid, &dev_info);
236 The next step is to configure the RX and TX queues. For each port, there is only
237 one RX queue (only one lcore is able to poll a given port). The number of TX
238 queues depends on the number of available lcores. The rte_eth_dev_configure()
239 function is used to configure the number of queues for a port:
243 ret = rte_eth_dev_configure((uint8_t)portid, 1, 1, &port_conf);
245 rte_panic("Cannot configure device: err=%d, port=%u\n",
248 .. _l2_fwd_event_app_rx_init:
250 RX Queue Initialization
251 ~~~~~~~~~~~~~~~~~~~~~~~
253 The application uses one lcore to poll one or several ports, depending on the -q
254 option, which specifies the number of queues per lcore.
256 For example, if the user specifies -q 4, the application is able to poll four
257 ports with one lcore. If there are 16 ports on the target (and if the portmask
258 argument is -p ffff ), the application will need four lcores to poll all the
263 ret = rte_eth_rx_queue_setup((uint8_t) portid, 0, nb_rxd, SOCKET0,
264 &rx_conf, l2fwd_pktmbuf_pool);
267 rte_panic("rte_eth_rx_queue_setup: err=%d, port=%u\n",
270 The list of queues that must be polled for a given lcore is stored in a private
271 structure called struct lcore_queue_conf.
275 struct lcore_queue_conf {
277 unsigned rx_port_list[MAX_RX_QUEUE_PER_LCORE];
278 struct mbuf_table tx_mbufs[L2FWD_MAX_PORTS];
281 struct lcore_queue_conf lcore_queue_conf[RTE_MAX_LCORE];
283 The values n_rx_port and rx_port_list[] are used in the main packet processing
284 loop (see :ref:`l2_fwd_event_app_rx_tx_packets`).
286 .. _l2_fwd_event_app_tx_init:
288 TX Queue Initialization
289 ~~~~~~~~~~~~~~~~~~~~~~~
291 Each lcore should be able to transmit on any port. For every port, a single TX
292 queue is initialized.
296 /* init one TX queue on each port */
300 ret = rte_eth_tx_queue_setup((uint8_t) portid, 0, nb_txd,
301 rte_eth_dev_socket_id(portid), &tx_conf);
303 rte_panic("rte_eth_tx_queue_setup:err=%d, port=%u\n",
304 ret, (unsigned) portid);
306 To configure eventdev support, application setups following components:
314 .. _l2_fwd_event_app_event_dev_init:
316 Event device Initialization
317 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
318 Application can use either H/W or S/W based event device scheduler
319 implementation and supports single instance of event device. It configures event
320 device as per below configuration
324 struct rte_event_dev_config event_d_conf = {
325 .nb_event_queues = ethdev_count, /* Dedicated to each Ethernet port */
326 .nb_event_ports = num_workers, /* Dedicated to each lcore */
327 .nb_events_limit = 4096,
328 .nb_event_queue_flows = 1024,
329 .nb_event_port_dequeue_depth = 128,
330 .nb_event_port_enqueue_depth = 128
333 ret = rte_event_dev_configure(event_d_id, &event_d_conf);
335 rte_panic("Error in configuring event device\n");
337 In case of S/W scheduler, application runs eventdev scheduler service on service
338 core. Application retrieves service id and finds the best possible service core to
343 rte_event_dev_info_get(evt_rsrc->event_d_id, &evdev_info);
344 if (evdev_info.event_dev_cap & RTE_EVENT_DEV_CAP_DISTRIBUTED_SCHED) {
345 ret = rte_event_dev_service_id_get(evt_rsrc->event_d_id,
347 if (ret != -ESRCH && ret != 0)
348 rte_panic("Error in starting eventdev service\n");
349 l2fwd_event_service_enable(service_id);
352 .. _l2_fwd_app_event_queue_init:
354 Event queue Initialization
355 ~~~~~~~~~~~~~~~~~~~~~~~~~~
356 Each Ethernet device is assigned a dedicated event queue which will be linked
357 to all available event ports i.e. each lcore can dequeue packets from any of the
362 struct rte_event_queue_conf event_q_conf = {
363 .nb_atomic_flows = 1024,
364 .nb_atomic_order_sequences = 1024,
365 .event_queue_cfg = 0,
366 .schedule_type = RTE_SCHED_TYPE_ATOMIC,
367 .priority = RTE_EVENT_DEV_PRIORITY_HIGHEST
370 /* User requested sched mode */
371 event_q_conf.schedule_type = eventq_sched_mode;
372 for (event_q_id = 0; event_q_id < ethdev_count; event_q_id++) {
373 ret = rte_event_queue_setup(event_d_id, event_q_id,
376 rte_panic("Error in configuring event queue\n");
379 In case of S/W scheduler, an extra event queue is created which will be used for
380 Tx adapter service function for enqueue operation.
382 .. _l2_fwd_app_event_port_init:
384 Event port Initialization
385 ~~~~~~~~~~~~~~~~~~~~~~~~~
386 Each worker thread is assigned a dedicated event port for enq/deq operations
387 to/from an event device. All event ports are linked with all available event
392 struct rte_event_port_conf event_p_conf = {
395 .new_event_threshold = 4096
398 for (event_p_id = 0; event_p_id < num_workers; event_p_id++) {
399 ret = rte_event_port_setup(event_d_id, event_p_id,
402 rte_panic("Error in configuring event port %d\n", event_p_id);
404 ret = rte_event_port_link(event_d_id, event_p_id, NULL,
407 rte_panic("Error in linking event port %d to queue\n",
411 In case of S/W scheduler, an extra event port is created by DPDK library which
412 is retrieved by the application and same will be used by Tx adapter service.
416 ret = rte_event_eth_tx_adapter_event_port_get(tx_adptr_id, &tx_port_id);
418 rte_panic("Failed to get Tx adapter port id: %d\n", ret);
420 ret = rte_event_port_link(event_d_id, tx_port_id,
421 &evt_rsrc.evq.event_q_id[
422 evt_rsrc.evq.nb_queues - 1],
425 rte_panic("Unable to link Tx adapter port to Tx queue:err=%d\n",
428 .. _l2_fwd_event_app_adapter_init:
430 Rx/Tx adapter Initialization
431 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
432 Each Ethernet port is assigned a dedicated Rx/Tx adapter for H/W scheduler. Each
433 Ethernet port's Rx queues are connected to its respective event queue at
434 priority 0 via Rx adapter configuration and Ethernet port's tx queues are
435 connected via Tx adapter.
439 RTE_ETH_FOREACH_DEV(port_id) {
440 if ((rsrc->enabled_port_mask & (1 << port_id)) == 0)
442 ret = rte_event_eth_rx_adapter_create(adapter_id, event_d_id,
443 &evt_rsrc->def_p_conf);
445 rte_panic("Failed to create rx adapter[%d]\n",
448 /* Configure user requested sched type*/
449 eth_q_conf.ev.sched_type = rsrc->sched_type;
450 eth_q_conf.ev.queue_id = evt_rsrc->evq.event_q_id[q_id];
451 ret = rte_event_eth_rx_adapter_queue_add(adapter_id, port_id,
454 rte_panic("Failed to add queues to Rx adapter\n");
456 ret = rte_event_eth_rx_adapter_start(adapter_id);
458 rte_panic("Rx adapter[%d] start Failed\n", adapter_id);
460 evt_rsrc->rx_adptr.rx_adptr[adapter_id] = adapter_id;
462 if (q_id < evt_rsrc->evq.nb_queues)
467 RTE_ETH_FOREACH_DEV(port_id) {
468 if ((rsrc->enabled_port_mask & (1 << port_id)) == 0)
470 ret = rte_event_eth_tx_adapter_create(adapter_id, event_d_id,
471 &evt_rsrc->def_p_conf);
473 rte_panic("Failed to create tx adapter[%d]\n",
476 ret = rte_event_eth_tx_adapter_queue_add(adapter_id, port_id,
479 rte_panic("Failed to add queues to Tx adapter\n");
481 ret = rte_event_eth_tx_adapter_start(adapter_id);
483 rte_panic("Tx adapter[%d] start Failed\n", adapter_id);
485 evt_rsrc->tx_adptr.tx_adptr[adapter_id] = adapter_id;
489 For S/W scheduler instead of dedicated adapters, common Rx/Tx adapters are
490 configured which will be shared among all the Ethernet ports. Also DPDK library
491 need service cores to run internal services for Rx/Tx adapters. Application gets
492 service id for Rx/Tx adapters and after successful setup it runs the services
493 on dedicated service cores.
497 for (i = 0; i < evt_rsrc->rx_adptr.nb_rx_adptr; i++) {
498 ret = rte_event_eth_rx_adapter_caps_get(evt_rsrc->event_d_id,
499 evt_rsrc->rx_adptr.rx_adptr[i], &caps);
501 rte_panic("Failed to get Rx adapter[%d] caps\n",
502 evt_rsrc->rx_adptr.rx_adptr[i]);
503 ret = rte_event_eth_rx_adapter_service_id_get(
504 evt_rsrc->event_d_id,
506 if (ret != -ESRCH && ret != 0)
507 rte_panic("Error in starting Rx adapter[%d] service\n",
508 evt_rsrc->rx_adptr.rx_adptr[i]);
509 l2fwd_event_service_enable(service_id);
512 for (i = 0; i < evt_rsrc->tx_adptr.nb_tx_adptr; i++) {
513 ret = rte_event_eth_tx_adapter_caps_get(evt_rsrc->event_d_id,
514 evt_rsrc->tx_adptr.tx_adptr[i], &caps);
516 rte_panic("Failed to get Rx adapter[%d] caps\n",
517 evt_rsrc->tx_adptr.tx_adptr[i]);
518 ret = rte_event_eth_tx_adapter_service_id_get(
519 evt_rsrc->event_d_id,
521 if (ret != -ESRCH && ret != 0)
522 rte_panic("Error in starting Rx adapter[%d] service\n",
523 evt_rsrc->tx_adptr.tx_adptr[i]);
524 l2fwd_event_service_enable(service_id);
527 .. _l2_fwd_event_app_rx_tx_packets:
529 Receive, Process and Transmit Packets
530 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
532 In the **l2fwd_main_loop()** function, the main task is to read ingress packets from
533 the RX queues. This is done using the following code:
538 * Read packet from RX queues
541 for (i = 0; i < qconf->n_rx_port; i++) {
542 portid = qconf->rx_port_list[i];
543 nb_rx = rte_eth_rx_burst((uint8_t) portid, 0, pkts_burst,
546 for (j = 0; j < nb_rx; j++) {
548 rte_prefetch0(rte_pktmbuf_mtod(m, void *));
549 l2fwd_simple_forward(m, portid);
553 Packets are read in a burst of size MAX_PKT_BURST. The rte_eth_rx_burst()
554 function writes the mbuf pointers in a local table and returns the number of
555 available mbufs in the table.
557 Then, each mbuf in the table is processed by the l2fwd_simple_forward()
558 function. The processing is very simple: process the TX port from the RX port,
559 then replace the source and destination MAC addresses if MAC addresses updating
562 During the initialization process, a static array of destination ports
563 (l2fwd_dst_ports[]) is filled such that for each source port, a destination port
564 is assigned that is either the next or previous enabled port from the portmask.
565 If number of ports are odd in portmask then packet from last port will be
566 forwarded to first port i.e. if portmask=0x07, then forwarding will take place
567 like p0--->p1, p1--->p2, p2--->p0.
569 Also to optimize enqueue operation, l2fwd_simple_forward() stores incoming mbufs
570 up to MAX_PKT_BURST. Once it reaches up to limit, all packets are transmitted to
576 l2fwd_simple_forward(struct rte_mbuf *m, uint32_t portid)
580 struct rte_eth_dev_tx_buffer *buffer;
582 dst_port = l2fwd_dst_ports[portid];
585 l2fwd_mac_updating(m, dst_port);
587 buffer = tx_buffer[dst_port];
588 sent = rte_eth_tx_buffer(dst_port, 0, buffer, m);
590 port_statistics[dst_port].tx += sent;
593 For this test application, the processing is exactly the same for all packets
594 arriving on the same RX port. Therefore, it would have been possible to call
595 the rte_eth_tx_buffer() function directly from the main loop to send all the
596 received packets on the same TX port, using the burst-oriented send function,
597 which is more efficient.
599 However, in real-life applications (such as, L3 routing),
600 packet N is not necessarily forwarded on the same port as packet N-1.
601 The application is implemented to illustrate that, so the same approach can be
602 reused in a more complex application.
604 To ensure that no packets remain in the tables, each lcore does a draining of TX
605 queue in its main loop. This technique introduces some latency when there are
606 not many packets to send, however it improves performance:
610 cur_tsc = rte_rdtsc();
613 * TX burst queue drain
615 diff_tsc = cur_tsc - prev_tsc;
616 if (unlikely(diff_tsc > drain_tsc)) {
617 for (i = 0; i < qconf->n_rx_port; i++) {
618 portid = l2fwd_dst_ports[qconf->rx_port_list[i]];
619 buffer = tx_buffer[portid];
620 sent = rte_eth_tx_buffer_flush(portid, 0,
623 port_statistics[portid].tx += sent;
626 /* if timer is enabled */
627 if (timer_period > 0) {
628 /* advance the timer */
629 timer_tsc += diff_tsc;
631 /* if timer has reached its timeout */
632 if (unlikely(timer_tsc >= timer_period)) {
633 /* do this only on master core */
634 if (lcore_id == rte_get_master_lcore()) {
636 /* reset the timer */
645 In the **l2fwd_event_loop()** function, the main task is to read ingress
646 packets from the event ports. This is done using the following code:
650 /* Read packet from eventdev */
651 nb_rx = rte_event_dequeue_burst(event_d_id, event_p_id,
658 for (i = 0; i < nb_rx; i++) {
659 mbuf[i] = events[i].mbuf;
660 rte_prefetch0(rte_pktmbuf_mtod(mbuf[i], void *));
664 Before reading packets, deq_len is fetched to ensure correct allowed deq length
666 The rte_event_dequeue_burst() function writes the mbuf pointers in a local table
667 and returns the number of available mbufs in the table.
669 Then, each mbuf in the table is processed by the l2fwd_eventdev_forward()
670 function. The processing is very simple: process the TX port from the RX port,
671 then replace the source and destination MAC addresses if MAC addresses updating
674 During the initialization process, a static array of destination ports
675 (l2fwd_dst_ports[]) is filled such that for each source port, a destination port
676 is assigned that is either the next or previous enabled port from the portmask.
677 If number of ports are odd in portmask then packet from last port will be
678 forwarded to first port i.e. if portmask=0x07, then forwarding will take place
679 like p0--->p1, p1--->p2, p2--->p0.
681 l2fwd_eventdev_forward() does not stores incoming mbufs. Packet will forwarded
682 be to destination ports via Tx adapter or generic event dev enqueue API
683 depending H/W or S/W scheduler is used.
687 nb_tx = rte_event_eth_tx_adapter_enqueue(event_d_id, port_id, ev,
689 while (nb_tx < nb_rx && !rsrc->force_quit)
690 nb_tx += rte_event_eth_tx_adapter_enqueue(
692 ev + nb_tx, nb_rx - nb_tx);