1.. SPDX-License-Identifier: BSD-3-Clause 2 Copyright(c) 2017 Intel Corporation. 3 4Flow Classify Sample Application 5================================ 6 7The Flow Classify sample application is based on the simple *skeleton* example 8of a forwarding application. 9 10It is intended as a demonstration of the basic components of a DPDK forwarding 11application which uses the Flow Classify library API's. 12 13Please refer to the 14:doc:`../prog_guide/flow_classify_lib` 15for more information. 16 17Compiling the Application 18------------------------- 19 20To compile the sample application see :doc:`compiling`. 21 22The application is located in the ``flow_classify`` sub-directory. 23 24Running the Application 25----------------------- 26 27To run the example in a ``linux`` environment: 28 29.. code-block:: console 30 31 ./<build_dir>/examples/dpdk-flow_classify -c 4 -n 4 -- / 32 --rule_ipv4="../ipv4_rules_file.txt" 33 34Please refer to the *DPDK Getting Started Guide*, section 35:doc:`../linux_gsg/build_sample_apps` 36for general information on running applications and the Environment Abstraction 37Layer (EAL) options. 38 39 40Sample ipv4_rules_file.txt 41-------------------------- 42 43.. code-block:: console 44 45 #file format: 46 #src_ip/masklen dst_ip/masklen src_port : mask dst_port : mask proto/mask priority 47 # 48 2.2.2.3/24 2.2.2.7/24 32 : 0xffff 33 : 0xffff 17/0xff 0 49 9.9.9.3/24 9.9.9.7/24 32 : 0xffff 33 : 0xffff 17/0xff 1 50 9.9.9.3/24 9.9.9.7/24 32 : 0xffff 33 : 0xffff 6/0xff 2 51 9.9.8.3/24 9.9.8.7/24 32 : 0xffff 33 : 0xffff 6/0xff 3 52 6.7.8.9/24 2.3.4.5/24 32 : 0x0000 33 : 0x0000 132/0xff 4 53 54Explanation 55----------- 56 57The following sections provide an explanation of the main components of the 58code. 59 60All DPDK library functions used in the sample code are prefixed with ``rte_`` 61and are explained in detail in the *DPDK API Documentation*. 62 63ACL field definitions for the IPv4 5 tuple rule 64~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 65 66The following field definitions are used when creating the ACL table during 67initialisation of the ``Flow Classify`` application.. 68 69.. code-block:: c 70 71 enum { 72 PROTO_FIELD_IPV4, 73 SRC_FIELD_IPV4, 74 DST_FIELD_IPV4, 75 SRCP_FIELD_IPV4, 76 DSTP_FIELD_IPV4, 77 NUM_FIELDS_IPV4 78 }; 79 80 enum { 81 PROTO_INPUT_IPV4, 82 SRC_INPUT_IPV4, 83 DST_INPUT_IPV4, 84 SRCP_DESTP_INPUT_IPV4 85 }; 86 87 static struct rte_acl_field_def ipv4_defs[NUM_FIELDS_IPV4] = { 88 /* first input field - always one byte long. */ 89 { 90 .type = RTE_ACL_FIELD_TYPE_BITMASK, 91 .size = sizeof(uint8_t), 92 .field_index = PROTO_FIELD_IPV4, 93 .input_index = PROTO_INPUT_IPV4, 94 .offset = sizeof(struct rte_ether_hdr) + 95 offsetof(struct rte_ipv4_hdr, next_proto_id), 96 }, 97 /* next input field (IPv4 source address) - 4 consecutive bytes. */ 98 { 99 /* rte_flow uses a bit mask for IPv4 addresses */ 100 .type = RTE_ACL_FIELD_TYPE_BITMASK, 101 .size = sizeof(uint32_t), 102 .field_index = SRC_FIELD_IPV4, 103 .input_index = SRC_INPUT_IPV4, 104 .offset = sizeof(struct rte_ether_hdr) + 105 offsetof(struct rte_ipv4_hdr, src_addr), 106 }, 107 /* next input field (IPv4 destination address) - 4 consecutive bytes. */ 108 { 109 /* rte_flow uses a bit mask for IPv4 addresses */ 110 .type = RTE_ACL_FIELD_TYPE_BITMASK, 111 .size = sizeof(uint32_t), 112 .field_index = DST_FIELD_IPV4, 113 .input_index = DST_INPUT_IPV4, 114 .offset = sizeof(struct rte_ether_hdr) + 115 offsetof(struct rte_ipv4_hdr, dst_addr), 116 }, 117 /* 118 * Next 2 fields (src & dst ports) form 4 consecutive bytes. 119 * They share the same input index. 120 */ 121 { 122 /* rte_flow uses a bit mask for protocol ports */ 123 .type = RTE_ACL_FIELD_TYPE_BITMASK, 124 .size = sizeof(uint16_t), 125 .field_index = SRCP_FIELD_IPV4, 126 .input_index = SRCP_DESTP_INPUT_IPV4, 127 .offset = sizeof(struct rte_ether_hdr) + 128 sizeof(struct rte_ipv4_hdr) + 129 offsetof(struct rte_tcp_hdr, src_port), 130 }, 131 { 132 /* rte_flow uses a bit mask for protocol ports */ 133 .type = RTE_ACL_FIELD_TYPE_BITMASK, 134 .size = sizeof(uint16_t), 135 .field_index = DSTP_FIELD_IPV4, 136 .input_index = SRCP_DESTP_INPUT_IPV4, 137 .offset = sizeof(struct rte_ether_hdr) + 138 sizeof(struct rte_ipv4_hdr) + 139 offsetof(struct rte_tcp_hdr, dst_port), 140 }, 141 }; 142 143The Main Function 144~~~~~~~~~~~~~~~~~ 145 146The ``main()`` function performs the initialization and calls the execution 147threads for each lcore. 148 149The first task is to initialize the Environment Abstraction Layer (EAL). 150The ``argc`` and ``argv`` arguments are provided to the ``rte_eal_init()`` 151function. The value returned is the number of parsed arguments: 152 153.. code-block:: c 154 155 int ret = rte_eal_init(argc, argv); 156 if (ret < 0) 157 rte_exit(EXIT_FAILURE, "Error with EAL initialization\n"); 158 159It then parses the flow_classify application arguments 160 161.. code-block:: c 162 163 ret = parse_args(argc, argv); 164 if (ret < 0) 165 rte_exit(EXIT_FAILURE, "Invalid flow_classify parameters\n"); 166 167The ``main()`` function also allocates a mempool to hold the mbufs 168(Message Buffers) used by the application: 169 170.. code-block:: c 171 172 mbuf_pool = rte_mempool_create("MBUF_POOL", 173 NUM_MBUFS * nb_ports, 174 MBUF_SIZE, 175 MBUF_CACHE_SIZE, 176 sizeof(struct rte_pktmbuf_pool_private), 177 rte_pktmbuf_pool_init, NULL, 178 rte_pktmbuf_init, NULL, 179 rte_socket_id(), 180 0); 181 182mbufs are the packet buffer structure used by DPDK. They are explained in 183detail in the "Mbuf Library" section of the *DPDK Programmer's Guide*. 184 185The ``main()`` function also initializes all the ports using the user defined 186``port_init()`` function which is explained in the next section: 187 188.. code-block:: c 189 190 RTE_ETH_FOREACH_DEV(portid) { 191 if (port_init(portid, mbuf_pool) != 0) { 192 rte_exit(EXIT_FAILURE, 193 "Cannot init port %" PRIu8 "\n", portid); 194 } 195 } 196 197The ``main()`` function creates the ``flow classifier object`` and adds an ``ACL 198table`` to the flow classifier. 199 200.. code-block:: c 201 202 struct flow_classifier { 203 struct rte_flow_classifier *cls; 204 }; 205 206 struct flow_classifier_acl { 207 struct flow_classifier cls; 208 } __rte_cache_aligned; 209 210 /* Memory allocation */ 211 size = RTE_CACHE_LINE_ROUNDUP(sizeof(struct flow_classifier_acl)); 212 cls_app = rte_zmalloc(NULL, size, RTE_CACHE_LINE_SIZE); 213 if (cls_app == NULL) 214 rte_exit(EXIT_FAILURE, "Cannot allocate classifier memory\n"); 215 216 cls_params.name = "flow_classifier"; 217 cls_params.socket_id = socket_id; 218 219 cls_app->cls = rte_flow_classifier_create(&cls_params); 220 if (cls_app->cls == NULL) { 221 rte_free(cls_app); 222 rte_exit(EXIT_FAILURE, "Cannot create classifier\n"); 223 } 224 225 /* initialise ACL table params */ 226 table_acl_params.name = "table_acl_ipv4_5tuple"; 227 table_acl_params.n_rule_fields = RTE_DIM(ipv4_defs); 228 table_acl_params.n_rules = FLOW_CLASSIFY_MAX_RULE_NUM; 229 memcpy(table_acl_params.field_format, ipv4_defs, sizeof(ipv4_defs)); 230 231 /* initialise table create params */ 232 cls_table_params.ops = &rte_table_acl_ops, 233 cls_table_params.arg_create = &table_acl_params, 234 cls_table_params.type = RTE_FLOW_CLASSIFY_TABLE_ACL_IP4_5TUPLE; 235 236 ret = rte_flow_classify_table_create(cls_app->cls, &cls_table_params); 237 if (ret) { 238 rte_flow_classifier_free(cls_app->cls); 239 rte_free(cls); 240 rte_exit(EXIT_FAILURE, "Failed to create classifier table\n"); 241 } 242 243It then reads the ipv4_rules_file.txt file and initialises the parameters for 244the ``rte_flow_classify_table_entry_add`` API. 245This API adds a rule to the ACL table. 246 247.. code-block:: c 248 249 if (add_rules(parm_config.rule_ipv4_name)) { 250 rte_flow_classifier_free(cls_app->cls); 251 rte_free(cls_app); 252 rte_exit(EXIT_FAILURE, "Failed to add rules\n"); 253 } 254 255Once the initialization is complete, the application is ready to launch a 256function on an lcore. In this example ``lcore_main()`` is called on a single 257lcore. 258 259.. code-block:: c 260 261 lcore_main(cls_app); 262 263The ``lcore_main()`` function is explained below. 264 265The Port Initialization Function 266~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 267 268The main functional part of the port initialization used in the Basic 269Forwarding application is shown below: 270 271.. code-block:: c 272 273 static inline int 274 port_init(uint16_t port, struct rte_mempool *mbuf_pool) 275 { 276 struct rte_eth_conf port_conf = port_conf_default; 277 const uint16_t rx_rings = 1, tx_rings = 1; 278 struct rte_ether_addr addr; 279 int retval; 280 uint16_t q; 281 282 /* Configure the Ethernet device. */ 283 retval = rte_eth_dev_configure(port, rx_rings, tx_rings, &port_conf); 284 if (retval != 0) 285 return retval; 286 287 /* Allocate and set up 1 RX queue per Ethernet port. */ 288 for (q = 0; q < rx_rings; q++) { 289 retval = rte_eth_rx_queue_setup(port, q, RX_RING_SIZE, 290 rte_eth_dev_socket_id(port), NULL, mbuf_pool); 291 if (retval < 0) 292 return retval; 293 } 294 295 /* Allocate and set up 1 TX queue per Ethernet port. */ 296 for (q = 0; q < tx_rings; q++) { 297 retval = rte_eth_tx_queue_setup(port, q, TX_RING_SIZE, 298 rte_eth_dev_socket_id(port), NULL); 299 if (retval < 0) 300 return retval; 301 } 302 303 /* Start the Ethernet port. */ 304 retval = rte_eth_dev_start(port); 305 if (retval < 0) 306 return retval; 307 308 /* Display the port MAC address. */ 309 retval = rte_eth_macaddr_get(port, &addr); 310 if (retval < 0) 311 return retval; 312 printf("Port %u MAC: %02" PRIx8 " %02" PRIx8 " %02" PRIx8 313 " %02" PRIx8 " %02" PRIx8 " %02" PRIx8 "\n", 314 port, 315 addr.addr_bytes[0], addr.addr_bytes[1], 316 addr.addr_bytes[2], addr.addr_bytes[3], 317 addr.addr_bytes[4], addr.addr_bytes[5]); 318 319 /* Enable RX in promiscuous mode for the Ethernet device. */ 320 retval = rte_eth_promiscuous_enable(port); 321 if (retval != 0) 322 return retval; 323 324 return 0; 325 } 326 327The Ethernet ports are configured with default settings using the 328``rte_eth_dev_configure()`` function and the ``port_conf_default`` struct. 329 330.. code-block:: c 331 332 static const struct rte_eth_conf port_conf_default = { 333 .rxmode = { .max_rx_pkt_len = RTE_ETHER_MAX_LEN } 334 }; 335 336For this example the ports are set up with 1 RX and 1 TX queue using the 337``rte_eth_rx_queue_setup()`` and ``rte_eth_tx_queue_setup()`` functions. 338 339The Ethernet port is then started: 340 341.. code-block:: c 342 343 retval = rte_eth_dev_start(port); 344 345 346Finally the RX port is set in promiscuous mode: 347 348.. code-block:: c 349 350 retval = rte_eth_promiscuous_enable(port); 351 352The Add Rules function 353~~~~~~~~~~~~~~~~~~~~~~ 354 355The ``add_rules`` function reads the ``ipv4_rules_file.txt`` file and calls the 356``add_classify_rule`` function which calls the 357``rte_flow_classify_table_entry_add`` API. 358 359.. code-block:: c 360 361 static int 362 add_rules(const char *rule_path) 363 { 364 FILE *fh; 365 char buff[LINE_MAX]; 366 unsigned int i = 0; 367 unsigned int total_num = 0; 368 struct rte_eth_ntuple_filter ntuple_filter; 369 370 fh = fopen(rule_path, "rb"); 371 if (fh == NULL) 372 rte_exit(EXIT_FAILURE, "%s: Open %s failed\n", __func__, 373 rule_path); 374 375 fseek(fh, 0, SEEK_SET); 376 377 i = 0; 378 while (fgets(buff, LINE_MAX, fh) != NULL) { 379 i++; 380 381 if (is_bypass_line(buff)) 382 continue; 383 384 if (total_num >= FLOW_CLASSIFY_MAX_RULE_NUM - 1) { 385 printf("\nINFO: classify rule capacity %d reached\n", 386 total_num); 387 break; 388 } 389 390 if (parse_ipv4_5tuple_rule(buff, &ntuple_filter) != 0) 391 rte_exit(EXIT_FAILURE, 392 "%s Line %u: parse rules error\n", 393 rule_path, i); 394 395 if (add_classify_rule(&ntuple_filter) != 0) 396 rte_exit(EXIT_FAILURE, "add rule error\n"); 397 398 total_num++; 399 } 400 401 fclose(fh); 402 return 0; 403 } 404 405 406The Lcore Main function 407~~~~~~~~~~~~~~~~~~~~~~~ 408 409As we saw above the ``main()`` function calls an application function on the 410available lcores. 411The ``lcore_main`` function calls the ``rte_flow_classifier_query`` API. 412For the Basic Forwarding application the ``lcore_main`` function looks like the 413following: 414 415.. code-block:: c 416 417 /* flow classify data */ 418 static int num_classify_rules; 419 static struct rte_flow_classify_rule *rules[MAX_NUM_CLASSIFY]; 420 static struct rte_flow_classify_ipv4_5tuple_stats ntuple_stats; 421 static struct rte_flow_classify_stats classify_stats = { 422 .stats = (void *)&ntuple_stats 423 }; 424 425 static __rte_noreturn void 426 lcore_main(cls_app) 427 { 428 uint16_t port; 429 430 /* 431 * Check that the port is on the same NUMA node as the polling thread 432 * for best performance. 433 */ 434 RTE_ETH_FOREACH_DEV(port) 435 if (rte_eth_dev_socket_id(port) > 0 && 436 rte_eth_dev_socket_id(port) != (int)rte_socket_id()) { 437 printf("\n\n"); 438 printf("WARNING: port %u is on remote NUMA node\n", 439 port); 440 printf("to polling thread.\n"); 441 printf("Performance will not be optimal.\n"); 442 443 printf("\nCore %u forwarding packets. \n", 444 rte_lcore_id()); 445 printf("[Ctrl+C to quit]\n 446 } 447 448 /* Run until the application is quit or killed. */ 449 for (;;) { 450 /* 451 * Receive packets on a port and forward them on the paired 452 * port. The mapping is 0 -> 1, 1 -> 0, 2 -> 3, 3 -> 2, etc. 453 */ 454 RTE_ETH_FOREACH_DEV(port) { 455 456 /* Get burst of RX packets, from first port of pair. */ 457 struct rte_mbuf *bufs[BURST_SIZE]; 458 const uint16_t nb_rx = rte_eth_rx_burst(port, 0, 459 bufs, BURST_SIZE); 460 461 if (unlikely(nb_rx == 0)) 462 continue; 463 464 for (i = 0; i < MAX_NUM_CLASSIFY; i++) { 465 if (rules[i]) { 466 ret = rte_flow_classifier_query( 467 cls_app->cls, 468 bufs, nb_rx, rules[i], 469 &classify_stats); 470 if (ret) 471 printf( 472 "rule [%d] query failed ret [%d]\n\n", 473 i, ret); 474 else { 475 printf( 476 "rule[%d] count=%"PRIu64"\n", 477 i, ntuple_stats.counter1); 478 479 printf("proto = %d\n", 480 ntuple_stats.ipv4_5tuple.proto); 481 } 482 } 483 } 484 485 /* Send burst of TX packets, to second port of pair. */ 486 const uint16_t nb_tx = rte_eth_tx_burst(port ^ 1, 0, 487 bufs, nb_rx); 488 489 /* Free any unsent packets. */ 490 if (unlikely(nb_tx < nb_rx)) { 491 uint16_t buf; 492 for (buf = nb_tx; buf < nb_rx; buf++) 493 rte_pktmbuf_free(bufs[buf]); 494 } 495 } 496 } 497 } 498 499The main work of the application is done within the loop: 500 501.. code-block:: c 502 503 for (;;) { 504 RTE_ETH_FOREACH_DEV(port) { 505 506 /* Get burst of RX packets, from first port of pair. */ 507 struct rte_mbuf *bufs[BURST_SIZE]; 508 const uint16_t nb_rx = rte_eth_rx_burst(port, 0, 509 bufs, BURST_SIZE); 510 511 if (unlikely(nb_rx == 0)) 512 continue; 513 514 /* Send burst of TX packets, to second port of pair. */ 515 const uint16_t nb_tx = rte_eth_tx_burst(port ^ 1, 0, 516 bufs, nb_rx); 517 518 /* Free any unsent packets. */ 519 if (unlikely(nb_tx < nb_rx)) { 520 uint16_t buf; 521 for (buf = nb_tx; buf < nb_rx; buf++) 522 rte_pktmbuf_free(bufs[buf]); 523 } 524 } 525 } 526 527Packets are received in bursts on the RX ports and transmitted in bursts on 528the TX ports. The ports are grouped in pairs with a simple mapping scheme 529using the an XOR on the port number:: 530 531 0 -> 1 532 1 -> 0 533 534 2 -> 3 535 3 -> 2 536 537 etc. 538 539The ``rte_eth_tx_burst()`` function frees the memory buffers of packets that 540are transmitted. If packets fail to transmit, ``(nb_tx < nb_rx)``, then they 541must be freed explicitly using ``rte_pktmbuf_free()``. 542 543The forwarding loop can be interrupted and the application closed using 544``Ctrl-C``. 545