xref: /dpdk/doc/guides/sample_app_ug/skeleton.rst (revision c7e9729d)
1..  SPDX-License-Identifier: BSD-3-Clause
2    Copyright(c) 2015 Intel Corporation.
3
4Basic Forwarding Sample Application
5===================================
6
7The Basic Forwarding sample application is a simple *skeleton* example of a
8forwarding application.
9
10It is intended as a demonstration of the basic components of a DPDK forwarding
11application. For more detailed implementations see the L2 and L3 forwarding
12sample applications.
13
14Compiling the Application
15-------------------------
16
17To compile the sample application see :doc:`compiling`.
18
19The application is located in the ``skeleton`` sub-directory.
20
21Running the Application
22-----------------------
23
24To run the example in a ``linuxapp`` environment:
25
26.. code-block:: console
27
28    ./build/basicfwd -l 1 -n 4
29
30Refer to *DPDK Getting Started Guide* for general information on running
31applications and the Environment Abstraction Layer (EAL) options.
32
33
34Explanation
35-----------
36
37The following sections provide an explanation of the main components of the
38code.
39
40All DPDK library functions used in the sample code are prefixed with ``rte_``
41and are explained in detail in the *DPDK API Documentation*.
42
43
44The Main Function
45~~~~~~~~~~~~~~~~~
46
47The ``main()`` function performs the initialization and calls the execution
48threads for each lcore.
49
50The first task is to initialize the Environment Abstraction Layer (EAL).  The
51``argc`` and ``argv`` arguments are provided to the ``rte_eal_init()``
52function. The value returned is the number of parsed arguments:
53
54.. code-block:: c
55
56    int ret = rte_eal_init(argc, argv);
57    if (ret < 0)
58        rte_exit(EXIT_FAILURE, "Error with EAL initialization\n");
59
60
61The ``main()`` also allocates a mempool to hold the mbufs (Message Buffers)
62used by the application:
63
64.. code-block:: c
65
66    mbuf_pool = rte_mempool_create("MBUF_POOL",
67                                   NUM_MBUFS * nb_ports,
68                                   MBUF_SIZE,
69                                   MBUF_CACHE_SIZE,
70                                   sizeof(struct rte_pktmbuf_pool_private),
71                                   rte_pktmbuf_pool_init, NULL,
72                                   rte_pktmbuf_init,      NULL,
73                                   rte_socket_id(),
74                                   0);
75
76Mbufs are the packet buffer structure used by DPDK. They are explained in
77detail in the "Mbuf Library" section of the *DPDK Programmer's Guide*.
78
79The ``main()`` function also initializes all the ports using the user defined
80``port_init()`` function which is explained in the next section:
81
82.. code-block:: c
83
84    for (portid = 0; portid < nb_ports; portid++) {
85        if (port_init(portid, mbuf_pool) != 0) {
86            rte_exit(EXIT_FAILURE,
87                     "Cannot init port %" PRIu8 "\n", portid);
88        }
89    }
90
91
92Once the initialization is complete, the application is ready to launch a
93function on an lcore. In this example ``lcore_main()`` is called on a single
94lcore.
95
96
97.. code-block:: c
98
99	lcore_main();
100
101The ``lcore_main()`` function is explained below.
102
103
104
105The Port Initialization  Function
106~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
107
108The main functional part of the port initialization used in the Basic
109Forwarding application is shown below:
110
111.. code-block:: c
112
113    static inline int
114    port_init(uint16_t port, struct rte_mempool *mbuf_pool)
115    {
116        struct rte_eth_conf port_conf = port_conf_default;
117        const uint16_t rx_rings = 1, tx_rings = 1;
118        struct ether_addr addr;
119        int retval;
120        uint16_t q;
121
122        if (port >= rte_eth_dev_count())
123            return -1;
124
125        /* Configure the Ethernet device. */
126        retval = rte_eth_dev_configure(port, rx_rings, tx_rings, &port_conf);
127        if (retval != 0)
128            return retval;
129
130        /* Allocate and set up 1 RX queue per Ethernet port. */
131        for (q = 0; q < rx_rings; q++) {
132            retval = rte_eth_rx_queue_setup(port, q, RX_RING_SIZE,
133                    rte_eth_dev_socket_id(port), NULL, mbuf_pool);
134            if (retval < 0)
135                return retval;
136        }
137
138        /* Allocate and set up 1 TX queue per Ethernet port. */
139        for (q = 0; q < tx_rings; q++) {
140            retval = rte_eth_tx_queue_setup(port, q, TX_RING_SIZE,
141                    rte_eth_dev_socket_id(port), NULL);
142            if (retval < 0)
143                return retval;
144        }
145
146        /* Start the Ethernet port. */
147        retval = rte_eth_dev_start(port);
148        if (retval < 0)
149            return retval;
150
151        /* Enable RX in promiscuous mode for the Ethernet device. */
152        rte_eth_promiscuous_enable(port);
153
154        return 0;
155    }
156
157The Ethernet ports are configured with default settings using the
158``rte_eth_dev_configure()`` function and the ``port_conf_default`` struct:
159
160.. code-block:: c
161
162    static const struct rte_eth_conf port_conf_default = {
163        .rxmode = { .max_rx_pkt_len = ETHER_MAX_LEN }
164    };
165
166For this example the ports are set up with 1 RX and 1 TX queue using the
167``rte_eth_rx_queue_setup()`` and ``rte_eth_tx_queue_setup()`` functions.
168
169The Ethernet port is then started:
170
171.. code-block:: c
172
173        retval  = rte_eth_dev_start(port);
174
175
176Finally the RX port is set in promiscuous mode:
177
178.. code-block:: c
179
180        rte_eth_promiscuous_enable(port);
181
182
183The Lcores Main
184~~~~~~~~~~~~~~~
185
186As we saw above the ``main()`` function calls an application function on the
187available lcores. For the Basic Forwarding application the lcore function
188looks like the following:
189
190.. code-block:: c
191
192    static __attribute__((noreturn)) void
193    lcore_main(void)
194    {
195        const uint16_t nb_ports = rte_eth_dev_count();
196        uint16_t port;
197
198        /*
199         * Check that the port is on the same NUMA node as the polling thread
200         * for best performance.
201         */
202        for (port = 0; port < nb_ports; port++)
203            if (rte_eth_dev_socket_id(port) > 0 &&
204                    rte_eth_dev_socket_id(port) !=
205                            (int)rte_socket_id())
206                printf("WARNING, port %u is on remote NUMA node to "
207                        "polling thread.\n\tPerformance will "
208                        "not be optimal.\n", port);
209
210        printf("\nCore %u forwarding packets. [Ctrl+C to quit]\n",
211                rte_lcore_id());
212
213        /* Run until the application is quit or killed. */
214        for (;;) {
215            /*
216             * Receive packets on a port and forward them on the paired
217             * port. The mapping is 0 -> 1, 1 -> 0, 2 -> 3, 3 -> 2, etc.
218             */
219            for (port = 0; port < nb_ports; port++) {
220
221                /* Get burst of RX packets, from first port of pair. */
222                struct rte_mbuf *bufs[BURST_SIZE];
223                const uint16_t nb_rx = rte_eth_rx_burst(port, 0,
224                        bufs, BURST_SIZE);
225
226                if (unlikely(nb_rx == 0))
227                    continue;
228
229                /* Send burst of TX packets, to second port of pair. */
230                const uint16_t nb_tx = rte_eth_tx_burst(port ^ 1, 0,
231                        bufs, nb_rx);
232
233                /* Free any unsent packets. */
234                if (unlikely(nb_tx < nb_rx)) {
235                    uint16_t buf;
236                    for (buf = nb_tx; buf < nb_rx; buf++)
237                        rte_pktmbuf_free(bufs[buf]);
238                }
239            }
240        }
241    }
242
243
244The main work of the application is done within the loop:
245
246.. code-block:: c
247
248        for (;;) {
249            for (port = 0; port < nb_ports; port++) {
250
251                /* Get burst of RX packets, from first port of pair. */
252                struct rte_mbuf *bufs[BURST_SIZE];
253                const uint16_t nb_rx = rte_eth_rx_burst(port, 0,
254                        bufs, BURST_SIZE);
255
256                if (unlikely(nb_rx == 0))
257                    continue;
258
259                /* Send burst of TX packets, to second port of pair. */
260                const uint16_t nb_tx = rte_eth_tx_burst(port ^ 1, 0,
261                        bufs, nb_rx);
262
263                /* Free any unsent packets. */
264                if (unlikely(nb_tx < nb_rx)) {
265                    uint16_t buf;
266                    for (buf = nb_tx; buf < nb_rx; buf++)
267                        rte_pktmbuf_free(bufs[buf]);
268                }
269            }
270        }
271
272Packets are received in bursts on the RX ports and transmitted in bursts on
273the TX ports. The ports are grouped in pairs with a simple mapping scheme
274using the an XOR on the port number::
275
276    0 -> 1
277    1 -> 0
278
279    2 -> 3
280    3 -> 2
281
282    etc.
283
284The ``rte_eth_tx_burst()`` function frees the memory buffers of packets that
285are transmitted. If packets fail to transmit, ``(nb_tx < nb_rx)``, then they
286must be freed explicitly using ``rte_pktmbuf_free()``.
287
288The forwarding loop can be interrupted and the application closed using
289``Ctrl-C``.
290