1 .. SPDX-License-Identifier: BSD-3-Clause
2 Copyright(c) 2010-2014 Intel Corporation.
4 Link Status Interrupt Sample Application
5 ========================================
7 The Link Status Interrupt sample application is a simple example of packet processing using
8 the Data Plane Development Kit (DPDK) that
9 demonstrates how network link status changes for a network port can be captured and
10 used by a DPDK application.
15 The Link Status Interrupt sample application registers a user space callback for the link status interrupt of each port
16 and performs L2 forwarding for each packet that is received on an RX_PORT.
17 The following operations are performed:
19 * RX_PORT and TX_PORT are paired with available ports one-by-one according to the core mask
21 * The source MAC address is replaced by the TX_PORT MAC address
23 * The destination MAC address is replaced by 02:00:00:00:00:TX_PORT_ID
25 This application can be used to demonstrate the usage of link status interrupt and its user space callbacks
26 and the behavior of L2 forwarding each time the link status changes.
28 Compiling the Application
29 -------------------------
31 To compile the sample application see :doc:`compiling`.
33 The application is located in the ``link_status_interrupt`` sub-directory.
35 Running the Application
36 -----------------------
38 The application requires a number of command line options:
40 .. code-block:: console
42 ./build/link_status_interrupt [EAL options] -- -p PORTMASK [-q NQ][-T PERIOD]
46 * -p PORTMASK: A hexadecimal bitmask of the ports to configure
48 * -q NQ: A number of queues (=ports) per lcore (default is 1)
50 * -T PERIOD: statistics will be refreshed each PERIOD seconds (0 to disable, 10 default)
52 To run the application in a linux environment with 4 lcores, 4 memory channels, 16 ports and 8 RX queues per lcore,
55 .. code-block:: console
57 $ ./build/link_status_interrupt -l 0-3 -n 4-- -q 8 -p ffff
59 Refer to the *DPDK Getting Started Guide* for general information on running applications
60 and the Environment Abstraction Layer (EAL) options.
65 The following sections provide some explanation of the code.
67 Command Line Arguments
68 ~~~~~~~~~~~~~~~~~~~~~~
70 The Link Status Interrupt sample application takes specific parameters,
71 in addition to Environment Abstraction Layer (EAL) arguments (see Section `Running the Application`_).
73 Command line parsing is done in the same way as it is done in the L2 Forwarding Sample Application.
74 See :ref:`l2_fwd_app_cmd_arguments` for more information.
76 Mbuf Pool Initialization
77 ~~~~~~~~~~~~~~~~~~~~~~~~
79 Mbuf pool initialization is done in the same way as it is done in the L2 Forwarding Sample Application.
80 See :ref:`l2_fwd_app_mbuf_init` for more information.
85 The main part of the code in the main() function relates to the initialization of the driver.
86 To fully understand this code, it is recommended to study the chapters that related to the Poll Mode Driver in the
87 *DPDK Programmer's Guide and the DPDK API Reference*.
92 * Each logical core is assigned a dedicated TX queue on each port.
95 RTE_ETH_FOREACH_DEV(portid) {
96 /* skip ports that are not enabled */
98 if ((lsi_enabled_port_mask & (1 << portid)) == 0)
101 /* save the destination port id */
103 if (nb_ports_in_mask % 2) {
104 lsi_dst_ports[portid] = portid_last;
105 lsi_dst_ports[portid_last] = portid;
108 portid_last = portid;
112 rte_eth_dev_info_get((uint8_t) portid, &dev_info);
115 The next step is to configure the RX and TX queues.
116 For each port, there is only one RX queue (only one lcore is able to poll a given port).
117 The number of TX queues depends on the number of available lcores.
118 The rte_eth_dev_configure() function is used to configure the number of queues for a port:
122 ret = rte_eth_dev_configure((uint8_t) portid, 1, 1, &port_conf);
124 rte_exit(EXIT_FAILURE, "Cannot configure device: err=%d, port=%u\n", ret, portid);
126 The global configuration is stored in a static structure:
130 static const struct rte_eth_conf port_conf = {
136 .lsc = 1, /**< link status interrupt feature enabled */
140 Configuring lsc to 0 (the default) disables the generation of any link status change interrupts in kernel space
141 and no user space interrupt event is received.
142 The public interface rte_eth_link_get() accesses the NIC registers directly to update the link status.
143 Configuring lsc to non-zero enables the generation of link status change interrupts in kernel space
144 when a link status change is present and calls the user space callbacks registered by the application.
145 The public interface rte_eth_link_get() just reads the link status in a global structure
146 that would be updated in the interrupt host thread only.
148 Interrupt Callback Registration
149 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
151 The application can register one or more callbacks to a specific port and interrupt event.
152 An example callback function that has been written as indicated below.
157 lsi_event_callback(uint16_t port_id, enum rte_eth_event_type type, void *param)
159 struct rte_eth_link link;
161 char link_status[RTE_ETH_LINK_MAX_STR_LEN];
165 printf("\n\nIn registered callback...\n");
167 printf("Event type: %s\n", type == RTE_ETH_EVENT_INTR_LSC ? "LSC interrupt" : "unknown event");
169 ret = rte_eth_link_get_nowait(port_id, &link);
171 printf("Failed to get port %d link status: %s\n\n",
172 port_id, rte_strerror(-ret));
174 rte_eth_link_to_str(link_status, sizeof(link_status), &link);
175 printf("Port %d %s\n", port_id, link_status);
179 This function is called when a link status interrupt is present for the right port.
180 The port_id indicates which port the interrupt applies to.
181 The type parameter identifies the interrupt event type,
182 which currently can be RTE_ETH_EVENT_INTR_LSC only, but other types can be added in the future.
183 The param parameter is the address of the parameter for the callback.
184 This function should be implemented with care since it will be called in the interrupt host thread,
185 which is different from the main thread of its caller.
187 The application registers the lsi_event_callback and a NULL parameter to the link status interrupt event on each port:
191 rte_eth_dev_callback_register((uint8_t)portid, RTE_ETH_EVENT_INTR_LSC, lsi_event_callback, NULL);
193 This registration can be done only after calling the rte_eth_dev_configure() function and before calling any other function.
194 If lsc is initialized with 0, the callback is never called since no interrupt event would ever be present.
196 RX Queue Initialization
197 ~~~~~~~~~~~~~~~~~~~~~~~
199 The application uses one lcore to poll one or several ports, depending on the -q option,
200 which specifies the number of queues per lcore.
202 For example, if the user specifies -q 4, the application is able to poll four ports with one lcore.
203 If there are 16 ports on the target (and if the portmask argument is -p ffff),
204 the application will need four lcores to poll all the ports.
208 ret = rte_eth_rx_queue_setup((uint8_t) portid, 0, nb_rxd, SOCKET0, &rx_conf, lsi_pktmbuf_pool);
210 rte_exit(EXIT_FAILURE, "rte_eth_rx_queue_setup: err=%d, port=%u\n", ret, portid);
212 The list of queues that must be polled for a given lcore is stored in a private structure called struct lcore_queue_conf.
216 struct lcore_queue_conf {
218 unsigned rx_port_list[MAX_RX_QUEUE_PER_LCORE]; unsigned tx_queue_id;
219 struct mbuf_table tx_mbufs[LSI_MAX_PORTS];
222 struct lcore_queue_conf lcore_queue_conf[RTE_MAX_LCORE];
224 The n_rx_port and rx_port_list[] fields are used in the main packet processing loop
225 (see `Receive, Process and Transmit Packets`_).
227 The global configuration for the RX queues is stored in a static structure:
231 static const struct rte_eth_rxconf rx_conf = {
233 .pthresh = RX_PTHRESH,
234 .hthresh = RX_HTHRESH,
235 .wthresh = RX_WTHRESH,
239 TX Queue Initialization
240 ~~~~~~~~~~~~~~~~~~~~~~~
242 Each lcore should be able to transmit on any port.
243 For every port, a single TX queue is initialized.
247 /* init one TX queue logical core on each port */
251 ret = rte_eth_tx_queue_setup(portid, 0, nb_txd, rte_eth_dev_socket_id(portid), &tx_conf);
253 rte_exit(EXIT_FAILURE, "rte_eth_tx_queue_setup: err=%d,port=%u\n", ret, (unsigned) portid);
255 The global configuration for TX queues is stored in a static structure:
259 static const struct rte_eth_txconf tx_conf = {
261 .pthresh = TX_PTHRESH,
262 .hthresh = TX_HTHRESH,
263 .wthresh = TX_WTHRESH,
265 .tx_free_thresh = RTE_TEST_TX_DESC_DEFAULT + 1, /* disable feature */
268 Receive, Process and Transmit Packets
269 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
271 In the lsi_main_loop() function, the main task is to read ingress packets from the RX queues.
272 This is done using the following code:
277 * Read packet from RX queues
280 for (i = 0; i < qconf->n_rx_port; i++) {
281 portid = qconf->rx_port_list[i];
282 nb_rx = rte_eth_rx_burst((uint8_t) portid, 0, pkts_burst, MAX_PKT_BURST);
283 port_statistics[portid].rx += nb_rx;
285 for (j = 0; j < nb_rx; j++) {
287 rte_prefetch0(rte_pktmbuf_mtod(m, void *));
288 lsi_simple_forward(m, portid);
292 Packets are read in a burst of size MAX_PKT_BURST.
293 The rte_eth_rx_burst() function writes the mbuf pointers in a local table and returns the number of available mbufs in the table.
295 Then, each mbuf in the table is processed by the lsi_simple_forward() function.
296 The processing is very simple: processes the TX port from the RX port and then replaces the source and destination MAC addresses.
300 In the following code, the two lines for calculating the output port require some explanation.
301 If portId is even, the first line does nothing (as portid & 1 will be 0), and the second line adds 1.
302 If portId is odd, the first line subtracts one and the second line does nothing.
303 Therefore, 0 goes to 1, and 1 to 0, 2 goes to 3 and 3 to 2, and so on.
308 lsi_simple_forward(struct rte_mbuf *m, unsigned portid)
310 struct rte_ether_hdr *eth;
312 unsigned dst_port = lsi_dst_ports[portid];
314 eth = rte_pktmbuf_mtod(m, struct rte_ether_hdr *);
316 /* 02:00:00:00:00:xx */
318 tmp = ð->d_addr.addr_bytes[0];
320 *((uint64_t *)tmp) = 0x000000000002 + (dst_port << 40);
323 rte_ether_addr_copy(&lsi_ports_eth_addr[dst_port], ð->s_addr);
325 lsi_send_packet(m, dst_port);
328 Then, the packet is sent using the lsi_send_packet(m, dst_port) function.
329 For this test application, the processing is exactly the same for all packets arriving on the same RX port.
330 Therefore, it would have been possible to call the lsi_send_burst() function directly from the main loop
331 to send all the received packets on the same TX port using
332 the burst-oriented send function, which is more efficient.
334 However, in real-life applications (such as, L3 routing),
335 packet N is not necessarily forwarded on the same port as packet N-1.
336 The application is implemented to illustrate that so the same approach can be reused in a more complex application.
338 The lsi_send_packet() function stores the packet in a per-lcore and per-txport table.
339 If the table is full, the whole packets table is transmitted using the lsi_send_burst() function:
343 /* Send the packet on an output interface */
346 lsi_send_packet(struct rte_mbuf *m, uint16_t port)
348 unsigned lcore_id, len;
349 struct lcore_queue_conf *qconf;
351 lcore_id = rte_lcore_id();
352 qconf = &lcore_queue_conf[lcore_id];
353 len = qconf->tx_mbufs[port].len;
354 qconf->tx_mbufs[port].m_table[len] = m;
357 /* enough pkts to be sent */
359 if (unlikely(len == MAX_PKT_BURST)) {
360 lsi_send_burst(qconf, MAX_PKT_BURST, port);
363 qconf->tx_mbufs[port].len = len;
368 To ensure that no packets remain in the tables, each lcore does a draining of the TX queue in its main loop.
369 This technique introduces some latency when there are not many packets to send.
370 However, it improves performance:
374 cur_tsc = rte_rdtsc();
377 * TX burst queue drain
380 diff_tsc = cur_tsc - prev_tsc;
382 if (unlikely(diff_tsc > drain_tsc)) {
383 /* this could be optimized (use queueid instead of * portid), but it is not called so often */
385 for (portid = 0; portid < RTE_MAX_ETHPORTS; portid++) {
386 if (qconf->tx_mbufs[portid].len == 0)
389 lsi_send_burst(&lcore_queue_conf[lcore_id],
390 qconf->tx_mbufs[portid].len, (uint8_t) portid);
391 qconf->tx_mbufs[portid].len = 0;
394 /* if timer is enabled */
396 if (timer_period > 0) {
397 /* advance the timer */
399 timer_tsc += diff_tsc;
401 /* if timer has reached its timeout */
403 if (unlikely(timer_tsc >= (uint64_t) timer_period)) {
404 /* do this only on master core */
406 if (lcore_id == rte_get_master_lcore()) {
409 /* reset the timer */