2 Copyright(c) 2010-2014 Intel Corporation. All rights reserved.
5 Redistribution and use in source and binary forms, with or without
6 modification, are permitted provided that the following conditions
9 * Redistributions of source code must retain the above copyright
10 notice, this list of conditions and the following disclaimer.
11 * Redistributions in binary form must reproduce the above copyright
12 notice, this list of conditions and the following disclaimer in
13 the documentation and/or other materials provided with the
15 * Neither the name of Intel Corporation nor the names of its
16 contributors may be used to endorse or promote products derived
17 from this software without specific prior written permission.
19 THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
20 "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
21 LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
22 A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
23 OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
24 SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
25 LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
26 DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
27 THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
28 (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
29 OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
31 L2 Forwarding Sample Application (in Real and Virtualized Environments)
32 =======================================================================
34 The L2 Forwarding sample application is a simple example of packet processing using
35 the Data Plane Development Kit (DPDK) which
36 also takes advantage of Single Root I/O Virtualization (SR-IOV) features in a virtualized environment.
40 Please note that previously a separate L2 Forwarding in Virtualized Environments sample application was used,
41 however, in later DPDK versions these sample applications have been merged.
46 The L2 Forwarding sample application, which can operate in real and virtualized environments,
47 performs L2 forwarding for each packet that is received on an RX_PORT.
48 The destination port is the adjacent port from the enabled portmask, that is,
49 if the first four ports are enabled (portmask 0xf),
50 ports 1 and 2 forward into each other, and ports 3 and 4 forward into each other.
51 Also, the MAC addresses are affected as follows:
53 * The source MAC address is replaced by the TX_PORT MAC address
55 * The destination MAC address is replaced by 02:00:00:00:00:TX_PORT_ID
57 This application can be used to benchmark performance using a traffic-generator, as shown in the :numref:`figure_l2_fwd_benchmark_setup`.
59 The application can also be used in a virtualized environment as shown in :numref:`figure_l2_fwd_virtenv_benchmark_setup`.
61 The L2 Forwarding application can also be used as a starting point for developing a new application based on the DPDK.
63 .. _figure_l2_fwd_benchmark_setup:
65 .. figure:: img/l2_fwd_benchmark_setup.*
67 Performance Benchmark Setup (Basic Environment)
70 .. _figure_l2_fwd_virtenv_benchmark_setup:
72 .. figure:: img/l2_fwd_virtenv_benchmark_setup.*
74 Performance Benchmark Setup (Virtualized Environment)
77 Virtual Function Setup Instructions
78 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
80 This application can use the virtual function available in the system and
81 therefore can be used in a virtual machine without passing through
82 the whole Network Device into a guest machine in a virtualized scenario.
83 The virtual functions can be enabled in the host machine or the hypervisor with the respective physical function driver.
85 For example, in a Linux* host machine, it is possible to enable a virtual function using the following command:
87 .. code-block:: console
89 modprobe ixgbe max_vfs=2,2
91 This command enables two Virtual Functions on each of Physical Function of the NIC,
92 with two physical ports in the PCI configuration space.
93 It is important to note that enabled Virtual Function 0 and 2 would belong to Physical Function 0
94 and Virtual Function 1 and 3 would belong to Physical Function 1,
95 in this case enabling a total of four Virtual Functions.
97 Compiling the Application
98 -------------------------
100 #. Go to the example directory:
102 .. code-block:: console
104 export RTE_SDK=/path/to/rte_sdk cd ${RTE_SDK}/examples/l2fwd
106 #. Set the target (a default target is used if not specified). For example:
108 .. code-block:: console
110 export RTE_TARGET=x86_64-native-linuxapp-gcc
112 *See the DPDK Getting Started Guide* for possible RTE_TARGET values.
114 #. Build the application:
116 .. code-block:: console
120 Running the Application
121 -----------------------
123 The application requires a number of command line options:
125 .. code-block:: console
127 ./build/l2fwd [EAL options] -- -p PORTMASK [-q NQ]
131 * p PORTMASK: A hexadecimal bitmask of the ports to configure
133 * q NQ: A number of queues (=ports) per lcore (default is 1)
135 To run the application in linuxapp environment with 4 lcores, 16 ports and 8 RX queues per lcore, issue the command:
137 .. code-block:: console
139 $ ./build/l2fwd -c f -n 4 -- -q 8 -p ffff
141 Refer to the *DPDK Getting Started Guide* for general information on running applications
142 and the Environment Abstraction Layer (EAL) options.
147 The following sections provide some explanation of the code.
149 Command Line Arguments
150 ~~~~~~~~~~~~~~~~~~~~~~
152 The L2 Forwarding sample application takes specific parameters,
153 in addition to Environment Abstraction Layer (EAL) arguments (see Section 9.3).
154 The preferred way to parse parameters is to use the getopt() function,
155 since it is part of a well-defined and portable library.
157 The parsing of arguments is done in the l2fwd_parse_args() function.
158 The method of argument parsing is not described here.
159 Refer to the *glibc getopt(3)* man page for details.
161 EAL arguments are parsed first, then application-specific arguments.
162 This is done at the beginning of the main() function:
168 ret = rte_eal_init(argc, argv);
170 rte_exit(EXIT_FAILURE, "Invalid EAL arguments\n");
175 /* parse application arguments (after the EAL ones) */
177 ret = l2fwd_parse_args(argc, argv);
179 rte_exit(EXIT_FAILURE, "Invalid L2FWD arguments\n");
181 Mbuf Pool Initialization
182 ~~~~~~~~~~~~~~~~~~~~~~~~
184 Once the arguments are parsed, the mbuf pool is created.
185 The mbuf pool contains a set of mbuf objects that will be used by the driver
186 and the application to store network packet data:
190 /* create the mbuf pool */
192 l2fwd_pktmbuf_pool = rte_mempool_create("mbuf_pool", NB_MBUF, MBUF_SIZE, 32, sizeof(struct rte_pktmbuf_pool_private),
193 rte_pktmbuf_pool_init, NULL, rte_pktmbuf_init, NULL, SOCKET0, 0);
195 if (l2fwd_pktmbuf_pool == NULL)
196 rte_panic("Cannot init mbuf pool\n");
198 The rte_mempool is a generic structure used to handle pools of objects.
199 In this case, it is necessary to create a pool that will be used by the driver,
200 which expects to have some reserved space in the mempool structure,
201 sizeof(struct rte_pktmbuf_pool_private) bytes.
202 The number of allocated pkt mbufs is NB_MBUF, with a size of MBUF_SIZE each.
203 A per-lcore cache of 32 mbufs is kept.
204 The memory is allocated in NUMA socket 0,
205 but it is possible to extend this code to allocate one mbuf pool per socket.
207 Two callback pointers are also given to the rte_mempool_create() function:
209 * The first callback pointer is to rte_pktmbuf_pool_init() and is used
210 to initialize the private data of the mempool, which is needed by the driver.
211 This function is provided by the mbuf API, but can be copied and extended by the developer.
213 * The second callback pointer given to rte_mempool_create() is the mbuf initializer.
214 The default is used, that is, rte_pktmbuf_init(), which is provided in the rte_mbuf library.
215 If a more complex application wants to extend the rte_pktmbuf structure for its own needs,
216 a new function derived from rte_pktmbuf_init( ) can be created.
218 Driver Initialization
219 ~~~~~~~~~~~~~~~~~~~~~
221 The main part of the code in the main() function relates to the initialization of the driver.
222 To fully understand this code, it is recommended to study the chapters that related to the Poll Mode Driver
223 in the *DPDK Programmer's Guide* - Rel 1.4 EAR and the *DPDK API Reference*.
227 if (rte_eal_pci_probe() < 0)
228 rte_exit(EXIT_FAILURE, "Cannot probe PCI\n");
230 nb_ports = rte_eth_dev_count();
233 rte_exit(EXIT_FAILURE, "No Ethernet ports - bye\n");
235 if (nb_ports > RTE_MAX_ETHPORTS)
236 nb_ports = RTE_MAX_ETHPORTS;
238 /* reset l2fwd_dst_ports */
240 for (portid = 0; portid < RTE_MAX_ETHPORTS; portid++)
241 l2fwd_dst_ports[portid] = 0;
246 * Each logical core is assigned a dedicated TX queue on each port.
249 for (portid = 0; portid < nb_ports; portid++) {
250 /* skip ports that are not enabled */
252 if ((l2fwd_enabled_port_mask & (1 << portid)) == 0)
255 if (nb_ports_in_mask % 2) {
256 l2fwd_dst_ports[portid] = last_port;
257 l2fwd_dst_ports[last_port] = portid;
264 rte_eth_dev_info_get((uint8_t) portid, &dev_info);
269 * rte_igb_pmd_init_all() simultaneously registers the driver as a PCI driver and as an Ethernet* Poll Mode Driver.
271 * rte_eal_pci_probe() parses the devices on the PCI bus and initializes recognized devices.
273 The next step is to configure the RX and TX queues.
274 For each port, there is only one RX queue (only one lcore is able to poll a given port).
275 The number of TX queues depends on the number of available lcores.
276 The rte_eth_dev_configure() function is used to configure the number of queues for a port:
280 ret = rte_eth_dev_configure((uint8_t)portid, 1, 1, &port_conf);
282 rte_exit(EXIT_FAILURE, "Cannot configure device: "
286 The global configuration is stored in a static structure:
290 static const struct rte_eth_conf port_conf = {
293 .header_split = 0, /**< Header Split disabled */
294 .hw_ip_checksum = 0, /**< IP checksum offload disabled */
295 .hw_vlan_filter = 0, /**< VLAN filtering disabled */
296 .jumbo_frame = 0, /**< Jumbo Frame Support disabled */
297 .hw_strip_crc= 0, /**< CRC stripped by hardware */
301 .mq_mode = ETH_DCB_NONE
305 RX Queue Initialization
306 ~~~~~~~~~~~~~~~~~~~~~~~
308 The application uses one lcore to poll one or several ports, depending on the -q option,
309 which specifies the number of queues per lcore.
311 For example, if the user specifies -q 4, the application is able to poll four ports with one lcore.
312 If there are 16 ports on the target (and if the portmask argument is -p ffff ),
313 the application will need four lcores to poll all the ports.
317 ret = rte_eth_rx_queue_setup((uint8_t) portid, 0, nb_rxd, SOCKET0, &rx_conf, l2fwd_pktmbuf_pool);
320 rte_exit(EXIT_FAILURE, "rte_eth_rx_queue_setup: "
324 The list of queues that must be polled for a given lcore is stored in a private structure called struct lcore_queue_conf.
328 struct lcore_queue_conf {
330 unsigned rx_port_list[MAX_RX_QUEUE_PER_LCORE];
331 struct mbuf_table tx_mbufs[L2FWD_MAX_PORTS];
334 struct lcore_queue_conf lcore_queue_conf[RTE_MAX_LCORE];
336 The values n_rx_port and rx_port_list[] are used in the main packet processing loop
337 (see Section 9.4.6 "Receive, Process and Transmit Packets" later in this chapter).
339 The global configuration for the RX queues is stored in a static structure:
343 static const struct rte_eth_rxconf rx_conf = {
345 .pthresh = RX_PTHRESH,
346 .hthresh = RX_HTHRESH,
347 .wthresh = RX_WTHRESH,
351 TX Queue Initialization
352 ~~~~~~~~~~~~~~~~~~~~~~~
354 Each lcore should be able to transmit on any port. For every port, a single TX queue is initialized.
358 /* init one TX queue on each port */
362 ret = rte_eth_tx_queue_setup((uint8_t) portid, 0, nb_txd, rte_eth_dev_socket_id(portid), &tx_conf);
364 rte_exit(EXIT_FAILURE, "rte_eth_tx_queue_setup:err=%d, port=%u\n", ret, (unsigned) portid);
366 The global configuration for TX queues is stored in a static structure:
370 static const struct rte_eth_txconf tx_conf = {
372 .pthresh = TX_PTHRESH,
373 .hthresh = TX_HTHRESH,
374 .wthresh = TX_WTHRESH,
376 .tx_free_thresh = RTE_TEST_TX_DESC_DEFAULT + 1, /* disable feature */
379 Receive, Process and Transmit Packets
380 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
382 In the l2fwd_main_loop() function, the main task is to read ingress packets from the RX queues.
383 This is done using the following code:
388 * Read packet from RX queues
391 for (i = 0; i < qconf->n_rx_port; i++) {
392 portid = qconf->rx_port_list[i];
393 nb_rx = rte_eth_rx_burst((uint8_t) portid, 0, pkts_burst, MAX_PKT_BURST);
395 for (j = 0; j < nb_rx; j++) {
397 rte_prefetch0[rte_pktmbuf_mtod(m, void *)); l2fwd_simple_forward(m, portid);
401 Packets are read in a burst of size MAX_PKT_BURST.
402 The rte_eth_rx_burst() function writes the mbuf pointers in a local table and returns the number of available mbufs in the table.
404 Then, each mbuf in the table is processed by the l2fwd_simple_forward() function.
405 The processing is very simple: process the TX port from the RX port, then replace the source and destination MAC addresses.
409 In the following code, one line for getting the output port requires some explanation.
411 During the initialization process, a static array of destination ports (l2fwd_dst_ports[]) is filled such that for each source port,
412 a destination port is assigned that is either the next or previous enabled port from the portmask.
413 Naturally, the number of ports in the portmask must be even, otherwise, the application exits.
418 l2fwd_simple_forward(struct rte_mbuf *m, unsigned portid)
420 struct ether_hdr *eth;
424 dst_port = l2fwd_dst_ports[portid];
426 eth = rte_pktmbuf_mtod(m, struct ether_hdr *);
428 /* 02:00:00:00:00:xx */
430 tmp = ð->d_addr.addr_bytes[0];
432 *((uint64_t *)tmp) = 0x000000000002 + ((uint64_t) dst_port << 40);
436 ether_addr_copy(&l2fwd_ports_eth_addr[dst_port], ð->s_addr);
438 l2fwd_send_packet(m, (uint8_t) dst_port);
441 Then, the packet is sent using the l2fwd_send_packet (m, dst_port) function.
442 For this test application, the processing is exactly the same for all packets arriving on the same RX port.
443 Therefore, it would have been possible to call the l2fwd_send_burst() function directly from the main loop
444 to send all the received packets on the same TX port,
445 using the burst-oriented send function, which is more efficient.
447 However, in real-life applications (such as, L3 routing),
448 packet N is not necessarily forwarded on the same port as packet N-1.
449 The application is implemented to illustrate that, so the same approach can be reused in a more complex application.
451 The l2fwd_send_packet() function stores the packet in a per-lcore and per-txport table.
452 If the table is full, the whole packets table is transmitted using the l2fwd_send_burst() function:
456 /* Send the packet on an output interface */
459 l2fwd_send_packet(struct rte_mbuf *m, uint8_t port)
461 unsigned lcore_id, len;
462 struct lcore_queue_conf \*qconf;
464 lcore_id = rte_lcore_id();
465 qconf = &lcore_queue_conf[lcore_id];
466 len = qconf->tx_mbufs[port].len;
467 qconf->tx_mbufs[port].m_table[len] = m;
470 /* enough pkts to be sent */
472 if (unlikely(len == MAX_PKT_BURST)) {
473 l2fwd_send_burst(qconf, MAX_PKT_BURST, port);
477 qconf->tx_mbufs[port].len = len; return 0;
480 To ensure that no packets remain in the tables, each lcore does a draining of TX queue in its main loop.
481 This technique introduces some latency when there are not many packets to send,
482 however it improves performance:
486 cur_tsc = rte_rdtsc();
489 * TX burst queue drain
492 diff_tsc = cur_tsc - prev_tsc;
494 if (unlikely(diff_tsc > drain_tsc)) {
495 for (portid = 0; portid < RTE_MAX_ETHPORTS; portid++) {
496 if (qconf->tx_mbufs[portid].len == 0)
499 l2fwd_send_burst(&lcore_queue_conf[lcore_id], qconf->tx_mbufs[portid].len, (uint8_t) portid);
501 qconf->tx_mbufs[portid].len = 0;
504 /* if timer is enabled */
506 if (timer_period > 0) {
507 /* advance the timer */
509 timer_tsc += diff_tsc;
511 /* if timer has reached its timeout */
513 if (unlikely(timer_tsc >= (uint64_t) timer_period)) {
514 /* do this only on master core */
516 if (lcore_id == rte_get_master_lcore()) {
519 /* reset the timer */