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
105 cd ${RTE_SDK}/examples/l2fwd
107 #. Set the target (a default target is used if not specified). For example:
109 .. code-block:: console
111 export RTE_TARGET=x86_64-native-linuxapp-gcc
113 *See the DPDK Getting Started Guide* for possible RTE_TARGET values.
115 #. Build the application:
117 .. code-block:: console
121 Running the Application
122 -----------------------
124 The application requires a number of command line options:
126 .. code-block:: console
128 ./build/l2fwd [EAL options] -- -p PORTMASK [-q NQ]
132 * p PORTMASK: A hexadecimal bitmask of the ports to configure
134 * q NQ: A number of queues (=ports) per lcore (default is 1)
136 To run the application in linuxapp environment with 4 lcores, 16 ports and 8 RX queues per lcore, issue the command:
138 .. code-block:: console
140 $ ./build/l2fwd -c f -n 4 -- -q 8 -p ffff
142 Refer to the *DPDK Getting Started Guide* for general information on running applications
143 and the Environment Abstraction Layer (EAL) options.
148 The following sections provide some explanation of the code.
150 Command Line Arguments
151 ~~~~~~~~~~~~~~~~~~~~~~
153 The L2 Forwarding sample application takes specific parameters,
154 in addition to Environment Abstraction Layer (EAL) arguments (see Section 9.3).
155 The preferred way to parse parameters is to use the getopt() function,
156 since it is part of a well-defined and portable library.
158 The parsing of arguments is done in the l2fwd_parse_args() function.
159 The method of argument parsing is not described here.
160 Refer to the *glibc getopt(3)* man page for details.
162 EAL arguments are parsed first, then application-specific arguments.
163 This is done at the beginning of the main() function:
169 ret = rte_eal_init(argc, argv);
171 rte_exit(EXIT_FAILURE, "Invalid EAL arguments\n");
176 /* parse application arguments (after the EAL ones) */
178 ret = l2fwd_parse_args(argc, argv);
180 rte_exit(EXIT_FAILURE, "Invalid L2FWD arguments\n");
182 Mbuf Pool Initialization
183 ~~~~~~~~~~~~~~~~~~~~~~~~
185 Once the arguments are parsed, the mbuf pool is created.
186 The mbuf pool contains a set of mbuf objects that will be used by the driver
187 and the application to store network packet data:
191 /* create the mbuf pool */
193 l2fwd_pktmbuf_pool = rte_mempool_create("mbuf_pool", NB_MBUF, MBUF_SIZE, 32, sizeof(struct rte_pktmbuf_pool_private),
194 rte_pktmbuf_pool_init, NULL, rte_pktmbuf_init, NULL, SOCKET0, 0);
196 if (l2fwd_pktmbuf_pool == NULL)
197 rte_panic("Cannot init mbuf pool\n");
199 The rte_mempool is a generic structure used to handle pools of objects.
200 In this case, it is necessary to create a pool that will be used by the driver,
201 which expects to have some reserved space in the mempool structure,
202 sizeof(struct rte_pktmbuf_pool_private) bytes.
203 The number of allocated pkt mbufs is NB_MBUF, with a size of MBUF_SIZE each.
204 A per-lcore cache of 32 mbufs is kept.
205 The memory is allocated in NUMA socket 0,
206 but it is possible to extend this code to allocate one mbuf pool per socket.
208 Two callback pointers are also given to the rte_mempool_create() function:
210 * The first callback pointer is to rte_pktmbuf_pool_init() and is used
211 to initialize the private data of the mempool, which is needed by the driver.
212 This function is provided by the mbuf API, but can be copied and extended by the developer.
214 * The second callback pointer given to rte_mempool_create() is the mbuf initializer.
215 The default is used, that is, rte_pktmbuf_init(), which is provided in the rte_mbuf library.
216 If a more complex application wants to extend the rte_pktmbuf structure for its own needs,
217 a new function derived from rte_pktmbuf_init( ) can be created.
219 Driver Initialization
220 ~~~~~~~~~~~~~~~~~~~~~
222 The main part of the code in the main() function relates to the initialization of the driver.
223 To fully understand this code, it is recommended to study the chapters that related to the Poll Mode Driver
224 in the *DPDK Programmer's Guide* - Rel 1.4 EAR and the *DPDK API Reference*.
228 if (rte_eal_pci_probe() < 0)
229 rte_exit(EXIT_FAILURE, "Cannot probe PCI\n");
231 nb_ports = rte_eth_dev_count();
234 rte_exit(EXIT_FAILURE, "No Ethernet ports - bye\n");
236 if (nb_ports > RTE_MAX_ETHPORTS)
237 nb_ports = RTE_MAX_ETHPORTS;
239 /* reset l2fwd_dst_ports */
241 for (portid = 0; portid < RTE_MAX_ETHPORTS; portid++)
242 l2fwd_dst_ports[portid] = 0;
247 * Each logical core is assigned a dedicated TX queue on each port.
250 for (portid = 0; portid < nb_ports; portid++) {
251 /* skip ports that are not enabled */
253 if ((l2fwd_enabled_port_mask & (1 << portid)) == 0)
256 if (nb_ports_in_mask % 2) {
257 l2fwd_dst_ports[portid] = last_port;
258 l2fwd_dst_ports[last_port] = portid;
265 rte_eth_dev_info_get((uint8_t) portid, &dev_info);
270 * rte_igb_pmd_init_all() simultaneously registers the driver as a PCI driver and as an Ethernet* Poll Mode Driver.
272 * rte_eal_pci_probe() parses the devices on the PCI bus and initializes recognized devices.
274 The next step is to configure the RX and TX queues.
275 For each port, there is only one RX queue (only one lcore is able to poll a given port).
276 The number of TX queues depends on the number of available lcores.
277 The rte_eth_dev_configure() function is used to configure the number of queues for a port:
281 ret = rte_eth_dev_configure((uint8_t)portid, 1, 1, &port_conf);
283 rte_exit(EXIT_FAILURE, "Cannot configure device: "
287 The global configuration is stored in a static structure:
291 static const struct rte_eth_conf port_conf = {
294 .header_split = 0, /**< Header Split disabled */
295 .hw_ip_checksum = 0, /**< IP checksum offload disabled */
296 .hw_vlan_filter = 0, /**< VLAN filtering disabled */
297 .jumbo_frame = 0, /**< Jumbo Frame Support disabled */
298 .hw_strip_crc= 0, /**< CRC stripped by hardware */
302 .mq_mode = ETH_DCB_NONE
306 RX Queue Initialization
307 ~~~~~~~~~~~~~~~~~~~~~~~
309 The application uses one lcore to poll one or several ports, depending on the -q option,
310 which specifies the number of queues per lcore.
312 For example, if the user specifies -q 4, the application is able to poll four ports with one lcore.
313 If there are 16 ports on the target (and if the portmask argument is -p ffff ),
314 the application will need four lcores to poll all the ports.
318 ret = rte_eth_rx_queue_setup((uint8_t) portid, 0, nb_rxd, SOCKET0, &rx_conf, l2fwd_pktmbuf_pool);
321 rte_exit(EXIT_FAILURE, "rte_eth_rx_queue_setup: "
325 The list of queues that must be polled for a given lcore is stored in a private structure called struct lcore_queue_conf.
329 struct lcore_queue_conf {
331 unsigned rx_port_list[MAX_RX_QUEUE_PER_LCORE];
332 struct mbuf_table tx_mbufs[L2FWD_MAX_PORTS];
335 struct lcore_queue_conf lcore_queue_conf[RTE_MAX_LCORE];
337 The values n_rx_port and rx_port_list[] are used in the main packet processing loop
338 (see Section 9.4.6 "Receive, Process and Transmit Packets" later in this chapter).
340 The global configuration for the RX queues is stored in a static structure:
344 static const struct rte_eth_rxconf rx_conf = {
346 .pthresh = RX_PTHRESH,
347 .hthresh = RX_HTHRESH,
348 .wthresh = RX_WTHRESH,
352 TX Queue Initialization
353 ~~~~~~~~~~~~~~~~~~~~~~~
355 Each lcore should be able to transmit on any port. For every port, a single TX queue is initialized.
359 /* init one TX queue on each port */
363 ret = rte_eth_tx_queue_setup((uint8_t) portid, 0, nb_txd, rte_eth_dev_socket_id(portid), &tx_conf);
365 rte_exit(EXIT_FAILURE, "rte_eth_tx_queue_setup:err=%d, port=%u\n", ret, (unsigned) portid);
367 The global configuration for TX queues is stored in a static structure:
371 static const struct rte_eth_txconf tx_conf = {
373 .pthresh = TX_PTHRESH,
374 .hthresh = TX_HTHRESH,
375 .wthresh = TX_WTHRESH,
377 .tx_free_thresh = RTE_TEST_TX_DESC_DEFAULT + 1, /* disable feature */
380 Receive, Process and Transmit Packets
381 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
383 In the l2fwd_main_loop() function, the main task is to read ingress packets from the RX queues.
384 This is done using the following code:
389 * Read packet from RX queues
392 for (i = 0; i < qconf->n_rx_port; i++) {
393 portid = qconf->rx_port_list[i];
394 nb_rx = rte_eth_rx_burst((uint8_t) portid, 0, pkts_burst, MAX_PKT_BURST);
396 for (j = 0; j < nb_rx; j++) {
398 rte_prefetch0[rte_pktmbuf_mtod(m, void *)); l2fwd_simple_forward(m, portid);
402 Packets are read in a burst of size MAX_PKT_BURST.
403 The rte_eth_rx_burst() function writes the mbuf pointers in a local table and returns the number of available mbufs in the table.
405 Then, each mbuf in the table is processed by the l2fwd_simple_forward() function.
406 The processing is very simple: process the TX port from the RX port, then replace the source and destination MAC addresses.
410 In the following code, one line for getting the output port requires some explanation.
412 During the initialization process, a static array of destination ports (l2fwd_dst_ports[]) is filled such that for each source port,
413 a destination port is assigned that is either the next or previous enabled port from the portmask.
414 Naturally, the number of ports in the portmask must be even, otherwise, the application exits.
419 l2fwd_simple_forward(struct rte_mbuf *m, unsigned portid)
421 struct ether_hdr *eth;
425 dst_port = l2fwd_dst_ports[portid];
427 eth = rte_pktmbuf_mtod(m, struct ether_hdr *);
429 /* 02:00:00:00:00:xx */
431 tmp = ð->d_addr.addr_bytes[0];
433 *((uint64_t *)tmp) = 0x000000000002 + ((uint64_t) dst_port << 40);
437 ether_addr_copy(&l2fwd_ports_eth_addr[dst_port], ð->s_addr);
439 l2fwd_send_packet(m, (uint8_t) dst_port);
442 Then, the packet is sent using the l2fwd_send_packet (m, dst_port) function.
443 For this test application, the processing is exactly the same for all packets arriving on the same RX port.
444 Therefore, it would have been possible to call the l2fwd_send_burst() function directly from the main loop
445 to send all the received packets on the same TX port,
446 using the burst-oriented send function, which is more efficient.
448 However, in real-life applications (such as, L3 routing),
449 packet N is not necessarily forwarded on the same port as packet N-1.
450 The application is implemented to illustrate that, so the same approach can be reused in a more complex application.
452 The l2fwd_send_packet() function stores the packet in a per-lcore and per-txport table.
453 If the table is full, the whole packets table is transmitted using the l2fwd_send_burst() function:
457 /* Send the packet on an output interface */
460 l2fwd_send_packet(struct rte_mbuf *m, uint8_t port)
462 unsigned lcore_id, len;
463 struct lcore_queue_conf \*qconf;
465 lcore_id = rte_lcore_id();
466 qconf = &lcore_queue_conf[lcore_id];
467 len = qconf->tx_mbufs[port].len;
468 qconf->tx_mbufs[port].m_table[len] = m;
471 /* enough pkts to be sent */
473 if (unlikely(len == MAX_PKT_BURST)) {
474 l2fwd_send_burst(qconf, MAX_PKT_BURST, port);
478 qconf->tx_mbufs[port].len = len; return 0;
481 To ensure that no packets remain in the tables, each lcore does a draining of TX queue in its main loop.
482 This technique introduces some latency when there are not many packets to send,
483 however it improves performance:
487 cur_tsc = rte_rdtsc();
490 * TX burst queue drain
493 diff_tsc = cur_tsc - prev_tsc;
495 if (unlikely(diff_tsc > drain_tsc)) {
496 for (portid = 0; portid < RTE_MAX_ETHPORTS; portid++) {
497 if (qconf->tx_mbufs[portid].len == 0)
500 l2fwd_send_burst(&lcore_queue_conf[lcore_id], qconf->tx_mbufs[portid].len, (uint8_t) portid);
502 qconf->tx_mbufs[portid].len = 0;
505 /* if timer is enabled */
507 if (timer_period > 0) {
508 /* advance the timer */
510 timer_tsc += diff_tsc;
512 /* if timer has reached its timeout */
514 if (unlikely(timer_tsc >= (uint64_t) timer_period)) {
515 /* do this only on master core */
517 if (lcore_id == rte_get_master_lcore()) {
520 /* reset the timer */