1 /* SPDX-License-Identifier: BSD-3-Clause 2 * Copyright(c) 2010-2017 Intel Corporation 3 */ 4 5 #ifndef _RTE_ETHDEV_H_ 6 #define _RTE_ETHDEV_H_ 7 8 /** 9 * @file 10 * 11 * RTE Ethernet Device API 12 * 13 * The Ethernet Device API is composed of two parts: 14 * 15 * - The application-oriented Ethernet API that includes functions to setup 16 * an Ethernet device (configure it, setup its RX and TX queues and start it), 17 * to get its MAC address, the speed and the status of its physical link, 18 * to receive and to transmit packets, and so on. 19 * 20 * - The driver-oriented Ethernet API that exports functions allowing 21 * an Ethernet Poll Mode Driver (PMD) to allocate an Ethernet device instance, 22 * create memzone for HW rings and process registered callbacks, and so on. 23 * PMDs should include rte_ethdev_driver.h instead of this header. 24 * 25 * By default, all the functions of the Ethernet Device API exported by a PMD 26 * are lock-free functions which assume to not be invoked in parallel on 27 * different logical cores to work on the same target object. For instance, 28 * the receive function of a PMD cannot be invoked in parallel on two logical 29 * cores to poll the same RX queue [of the same port]. Of course, this function 30 * can be invoked in parallel by different logical cores on different RX queues. 31 * It is the responsibility of the upper level application to enforce this rule. 32 * 33 * If needed, parallel accesses by multiple logical cores to shared queues 34 * shall be explicitly protected by dedicated inline lock-aware functions 35 * built on top of their corresponding lock-free functions of the PMD API. 36 * 37 * In all functions of the Ethernet API, the Ethernet device is 38 * designated by an integer >= 0 named the device port identifier. 39 * 40 * At the Ethernet driver level, Ethernet devices are represented by a generic 41 * data structure of type *rte_eth_dev*. 42 * 43 * Ethernet devices are dynamically registered during the PCI probing phase 44 * performed at EAL initialization time. 45 * When an Ethernet device is being probed, an *rte_eth_dev* structure and 46 * a new port identifier are allocated for that device. Then, the eth_dev_init() 47 * function supplied by the Ethernet driver matching the probed PCI 48 * device is invoked to properly initialize the device. 49 * 50 * The role of the device init function consists of resetting the hardware, 51 * checking access to Non-volatile Memory (NVM), reading the MAC address 52 * from NVM etc. 53 * 54 * If the device init operation is successful, the correspondence between 55 * the port identifier assigned to the new device and its associated 56 * *rte_eth_dev* structure is effectively registered. 57 * Otherwise, both the *rte_eth_dev* structure and the port identifier are 58 * freed. 59 * 60 * The functions exported by the application Ethernet API to setup a device 61 * designated by its port identifier must be invoked in the following order: 62 * - rte_eth_dev_configure() 63 * - rte_eth_tx_queue_setup() 64 * - rte_eth_rx_queue_setup() 65 * - rte_eth_dev_start() 66 * 67 * Then, the network application can invoke, in any order, the functions 68 * exported by the Ethernet API to get the MAC address of a given device, to 69 * get the speed and the status of a device physical link, to receive/transmit 70 * [burst of] packets, and so on. 71 * 72 * If the application wants to change the configuration (i.e. call 73 * rte_eth_dev_configure(), rte_eth_tx_queue_setup(), or 74 * rte_eth_rx_queue_setup()), it must call rte_eth_dev_stop() first to stop the 75 * device and then do the reconfiguration before calling rte_eth_dev_start() 76 * again. The transmit and receive functions should not be invoked when the 77 * device is stopped. 78 * 79 * Please note that some configuration is not stored between calls to 80 * rte_eth_dev_stop()/rte_eth_dev_start(). The following configuration will 81 * be retained: 82 * 83 * - MTU 84 * - flow control settings 85 * - receive mode configuration (promiscuous mode, all-multicast mode, 86 * hardware checksum mode, RSS/VMDQ settings etc.) 87 * - VLAN filtering configuration 88 * - default MAC address 89 * - MAC addresses supplied to MAC address array 90 * - flow director filtering mode (but not filtering rules) 91 * - NIC queue statistics mappings 92 * 93 * Any other configuration will not be stored and will need to be re-entered 94 * before a call to rte_eth_dev_start(). 95 * 96 * Finally, a network application can close an Ethernet device by invoking the 97 * rte_eth_dev_close() function. 98 * 99 * Each function of the application Ethernet API invokes a specific function 100 * of the PMD that controls the target device designated by its port 101 * identifier. 102 * For this purpose, all device-specific functions of an Ethernet driver are 103 * supplied through a set of pointers contained in a generic structure of type 104 * *eth_dev_ops*. 105 * The address of the *eth_dev_ops* structure is stored in the *rte_eth_dev* 106 * structure by the device init function of the Ethernet driver, which is 107 * invoked during the PCI probing phase, as explained earlier. 108 * 109 * In other words, each function of the Ethernet API simply retrieves the 110 * *rte_eth_dev* structure associated with the device port identifier and 111 * performs an indirect invocation of the corresponding driver function 112 * supplied in the *eth_dev_ops* structure of the *rte_eth_dev* structure. 113 * 114 * For performance reasons, the address of the burst-oriented RX and TX 115 * functions of the Ethernet driver are not contained in the *eth_dev_ops* 116 * structure. Instead, they are directly stored at the beginning of the 117 * *rte_eth_dev* structure to avoid an extra indirect memory access during 118 * their invocation. 119 * 120 * RTE ethernet device drivers do not use interrupts for transmitting or 121 * receiving. Instead, Ethernet drivers export Poll-Mode receive and transmit 122 * functions to applications. 123 * Both receive and transmit functions are packet-burst oriented to minimize 124 * their cost per packet through the following optimizations: 125 * 126 * - Sharing among multiple packets the incompressible cost of the 127 * invocation of receive/transmit functions. 128 * 129 * - Enabling receive/transmit functions to take advantage of burst-oriented 130 * hardware features (L1 cache, prefetch instructions, NIC head/tail 131 * registers) to minimize the number of CPU cycles per packet, for instance, 132 * by avoiding useless read memory accesses to ring descriptors, or by 133 * systematically using arrays of pointers that exactly fit L1 cache line 134 * boundaries and sizes. 135 * 136 * The burst-oriented receive function does not provide any error notification, 137 * to avoid the corresponding overhead. As a hint, the upper-level application 138 * might check the status of the device link once being systematically returned 139 * a 0 value by the receive function of the driver for a given number of tries. 140 */ 141 142 #ifdef __cplusplus 143 extern "C" { 144 #endif 145 146 #include <stdint.h> 147 148 /* Use this macro to check if LRO API is supported */ 149 #define RTE_ETHDEV_HAS_LRO_SUPPORT 150 151 #include <rte_compat.h> 152 #include <rte_log.h> 153 #include <rte_interrupts.h> 154 #include <rte_dev.h> 155 #include <rte_devargs.h> 156 #include <rte_errno.h> 157 #include <rte_common.h> 158 #include <rte_config.h> 159 #include <rte_ether.h> 160 161 #include "rte_dev_info.h" 162 163 extern int rte_eth_dev_logtype; 164 165 #define RTE_ETHDEV_LOG(level, ...) \ 166 rte_log(RTE_LOG_ ## level, rte_eth_dev_logtype, "" __VA_ARGS__) 167 168 struct rte_mbuf; 169 170 /** 171 * Initializes a device iterator. 172 * 173 * This iterator allows accessing a list of devices matching some devargs. 174 * 175 * @param iter 176 * Device iterator handle initialized by the function. 177 * The fields bus_str and cls_str might be dynamically allocated, 178 * and could be freed by calling rte_eth_iterator_cleanup(). 179 * 180 * @param devargs 181 * Device description string. 182 * 183 * @return 184 * 0 on successful initialization, negative otherwise. 185 */ 186 int rte_eth_iterator_init(struct rte_dev_iterator *iter, const char *devargs); 187 188 /** 189 * Iterates on devices with devargs filter. 190 * The ownership is not checked. 191 * 192 * The next port id is returned, and the iterator is updated. 193 * 194 * @param iter 195 * Device iterator handle initialized by rte_eth_iterator_init(). 196 * Some fields bus_str and cls_str might be freed when no more port is found, 197 * by calling rte_eth_iterator_cleanup(). 198 * 199 * @return 200 * A port id if found, RTE_MAX_ETHPORTS otherwise. 201 */ 202 uint16_t rte_eth_iterator_next(struct rte_dev_iterator *iter); 203 204 /** 205 * Free some allocated fields of the iterator. 206 * 207 * This function is automatically called by rte_eth_iterator_next() 208 * on the last iteration (i.e. when no more matching port is found). 209 * 210 * It is safe to call this function twice; it will do nothing more. 211 * 212 * @param iter 213 * Device iterator handle initialized by rte_eth_iterator_init(). 214 * The fields bus_str and cls_str are freed if needed. 215 */ 216 void rte_eth_iterator_cleanup(struct rte_dev_iterator *iter); 217 218 /** 219 * Macro to iterate over all ethdev ports matching some devargs. 220 * 221 * If a break is done before the end of the loop, 222 * the function rte_eth_iterator_cleanup() must be called. 223 * 224 * @param id 225 * Iterated port id of type uint16_t. 226 * @param devargs 227 * Device parameters input as string of type char*. 228 * @param iter 229 * Iterator handle of type struct rte_dev_iterator, used internally. 230 */ 231 #define RTE_ETH_FOREACH_MATCHING_DEV(id, devargs, iter) \ 232 for (rte_eth_iterator_init(iter, devargs), \ 233 id = rte_eth_iterator_next(iter); \ 234 id != RTE_MAX_ETHPORTS; \ 235 id = rte_eth_iterator_next(iter)) 236 237 /** 238 * A structure used to retrieve statistics for an Ethernet port. 239 * Not all statistics fields in struct rte_eth_stats are supported 240 * by any type of network interface card (NIC). If any statistics 241 * field is not supported, its value is 0. 242 */ 243 struct rte_eth_stats { 244 uint64_t ipackets; /**< Total number of successfully received packets. */ 245 uint64_t opackets; /**< Total number of successfully transmitted packets.*/ 246 uint64_t ibytes; /**< Total number of successfully received bytes. */ 247 uint64_t obytes; /**< Total number of successfully transmitted bytes. */ 248 uint64_t imissed; 249 /**< Total of RX packets dropped by the HW, 250 * because there are no available buffer (i.e. RX queues are full). 251 */ 252 uint64_t ierrors; /**< Total number of erroneous received packets. */ 253 uint64_t oerrors; /**< Total number of failed transmitted packets. */ 254 uint64_t rx_nombuf; /**< Total number of RX mbuf allocation failures. */ 255 uint64_t q_ipackets[RTE_ETHDEV_QUEUE_STAT_CNTRS]; 256 /**< Total number of queue RX packets. */ 257 uint64_t q_opackets[RTE_ETHDEV_QUEUE_STAT_CNTRS]; 258 /**< Total number of queue TX packets. */ 259 uint64_t q_ibytes[RTE_ETHDEV_QUEUE_STAT_CNTRS]; 260 /**< Total number of successfully received queue bytes. */ 261 uint64_t q_obytes[RTE_ETHDEV_QUEUE_STAT_CNTRS]; 262 /**< Total number of successfully transmitted queue bytes. */ 263 uint64_t q_errors[RTE_ETHDEV_QUEUE_STAT_CNTRS]; 264 /**< Total number of queue packets received that are dropped. */ 265 }; 266 267 /** 268 * Device supported speeds bitmap flags 269 */ 270 #define ETH_LINK_SPEED_AUTONEG (0 << 0) /**< Autonegotiate (all speeds) */ 271 #define ETH_LINK_SPEED_FIXED (1 << 0) /**< Disable autoneg (fixed speed) */ 272 #define ETH_LINK_SPEED_10M_HD (1 << 1) /**< 10 Mbps half-duplex */ 273 #define ETH_LINK_SPEED_10M (1 << 2) /**< 10 Mbps full-duplex */ 274 #define ETH_LINK_SPEED_100M_HD (1 << 3) /**< 100 Mbps half-duplex */ 275 #define ETH_LINK_SPEED_100M (1 << 4) /**< 100 Mbps full-duplex */ 276 #define ETH_LINK_SPEED_1G (1 << 5) /**< 1 Gbps */ 277 #define ETH_LINK_SPEED_2_5G (1 << 6) /**< 2.5 Gbps */ 278 #define ETH_LINK_SPEED_5G (1 << 7) /**< 5 Gbps */ 279 #define ETH_LINK_SPEED_10G (1 << 8) /**< 10 Gbps */ 280 #define ETH_LINK_SPEED_20G (1 << 9) /**< 20 Gbps */ 281 #define ETH_LINK_SPEED_25G (1 << 10) /**< 25 Gbps */ 282 #define ETH_LINK_SPEED_40G (1 << 11) /**< 40 Gbps */ 283 #define ETH_LINK_SPEED_50G (1 << 12) /**< 50 Gbps */ 284 #define ETH_LINK_SPEED_56G (1 << 13) /**< 56 Gbps */ 285 #define ETH_LINK_SPEED_100G (1 << 14) /**< 100 Gbps */ 286 287 /** 288 * Ethernet numeric link speeds in Mbps 289 */ 290 #define ETH_SPEED_NUM_NONE 0 /**< Not defined */ 291 #define ETH_SPEED_NUM_10M 10 /**< 10 Mbps */ 292 #define ETH_SPEED_NUM_100M 100 /**< 100 Mbps */ 293 #define ETH_SPEED_NUM_1G 1000 /**< 1 Gbps */ 294 #define ETH_SPEED_NUM_2_5G 2500 /**< 2.5 Gbps */ 295 #define ETH_SPEED_NUM_5G 5000 /**< 5 Gbps */ 296 #define ETH_SPEED_NUM_10G 10000 /**< 10 Gbps */ 297 #define ETH_SPEED_NUM_20G 20000 /**< 20 Gbps */ 298 #define ETH_SPEED_NUM_25G 25000 /**< 25 Gbps */ 299 #define ETH_SPEED_NUM_40G 40000 /**< 40 Gbps */ 300 #define ETH_SPEED_NUM_50G 50000 /**< 50 Gbps */ 301 #define ETH_SPEED_NUM_56G 56000 /**< 56 Gbps */ 302 #define ETH_SPEED_NUM_100G 100000 /**< 100 Gbps */ 303 304 /** 305 * A structure used to retrieve link-level information of an Ethernet port. 306 */ 307 __extension__ 308 struct rte_eth_link { 309 uint32_t link_speed; /**< ETH_SPEED_NUM_ */ 310 uint16_t link_duplex : 1; /**< ETH_LINK_[HALF/FULL]_DUPLEX */ 311 uint16_t link_autoneg : 1; /**< ETH_LINK_[AUTONEG/FIXED] */ 312 uint16_t link_status : 1; /**< ETH_LINK_[DOWN/UP] */ 313 } __attribute__((aligned(8))); /**< aligned for atomic64 read/write */ 314 315 /* Utility constants */ 316 #define ETH_LINK_HALF_DUPLEX 0 /**< Half-duplex connection (see link_duplex). */ 317 #define ETH_LINK_FULL_DUPLEX 1 /**< Full-duplex connection (see link_duplex). */ 318 #define ETH_LINK_DOWN 0 /**< Link is down (see link_status). */ 319 #define ETH_LINK_UP 1 /**< Link is up (see link_status). */ 320 #define ETH_LINK_FIXED 0 /**< No autonegotiation (see link_autoneg). */ 321 #define ETH_LINK_AUTONEG 1 /**< Autonegotiated (see link_autoneg). */ 322 323 /** 324 * A structure used to configure the ring threshold registers of an RX/TX 325 * queue for an Ethernet port. 326 */ 327 struct rte_eth_thresh { 328 uint8_t pthresh; /**< Ring prefetch threshold. */ 329 uint8_t hthresh; /**< Ring host threshold. */ 330 uint8_t wthresh; /**< Ring writeback threshold. */ 331 }; 332 333 /** 334 * Simple flags are used for rte_eth_conf.rxmode.mq_mode. 335 */ 336 #define ETH_MQ_RX_RSS_FLAG 0x1 337 #define ETH_MQ_RX_DCB_FLAG 0x2 338 #define ETH_MQ_RX_VMDQ_FLAG 0x4 339 340 /** 341 * A set of values to identify what method is to be used to route 342 * packets to multiple queues. 343 */ 344 enum rte_eth_rx_mq_mode { 345 /** None of DCB,RSS or VMDQ mode */ 346 ETH_MQ_RX_NONE = 0, 347 348 /** For RX side, only RSS is on */ 349 ETH_MQ_RX_RSS = ETH_MQ_RX_RSS_FLAG, 350 /** For RX side,only DCB is on. */ 351 ETH_MQ_RX_DCB = ETH_MQ_RX_DCB_FLAG, 352 /** Both DCB and RSS enable */ 353 ETH_MQ_RX_DCB_RSS = ETH_MQ_RX_RSS_FLAG | ETH_MQ_RX_DCB_FLAG, 354 355 /** Only VMDQ, no RSS nor DCB */ 356 ETH_MQ_RX_VMDQ_ONLY = ETH_MQ_RX_VMDQ_FLAG, 357 /** RSS mode with VMDQ */ 358 ETH_MQ_RX_VMDQ_RSS = ETH_MQ_RX_RSS_FLAG | ETH_MQ_RX_VMDQ_FLAG, 359 /** Use VMDQ+DCB to route traffic to queues */ 360 ETH_MQ_RX_VMDQ_DCB = ETH_MQ_RX_VMDQ_FLAG | ETH_MQ_RX_DCB_FLAG, 361 /** Enable both VMDQ and DCB in VMDq */ 362 ETH_MQ_RX_VMDQ_DCB_RSS = ETH_MQ_RX_RSS_FLAG | ETH_MQ_RX_DCB_FLAG | 363 ETH_MQ_RX_VMDQ_FLAG, 364 }; 365 366 /** 367 * for rx mq mode backward compatible 368 */ 369 #define ETH_RSS ETH_MQ_RX_RSS 370 #define VMDQ_DCB ETH_MQ_RX_VMDQ_DCB 371 #define ETH_DCB_RX ETH_MQ_RX_DCB 372 373 /** 374 * A set of values to identify what method is to be used to transmit 375 * packets using multi-TCs. 376 */ 377 enum rte_eth_tx_mq_mode { 378 ETH_MQ_TX_NONE = 0, /**< It is in neither DCB nor VT mode. */ 379 ETH_MQ_TX_DCB, /**< For TX side,only DCB is on. */ 380 ETH_MQ_TX_VMDQ_DCB, /**< For TX side,both DCB and VT is on. */ 381 ETH_MQ_TX_VMDQ_ONLY, /**< Only VT on, no DCB */ 382 }; 383 384 /** 385 * for tx mq mode backward compatible 386 */ 387 #define ETH_DCB_NONE ETH_MQ_TX_NONE 388 #define ETH_VMDQ_DCB_TX ETH_MQ_TX_VMDQ_DCB 389 #define ETH_DCB_TX ETH_MQ_TX_DCB 390 391 /** 392 * A structure used to configure the RX features of an Ethernet port. 393 */ 394 struct rte_eth_rxmode { 395 /** The multi-queue packet distribution mode to be used, e.g. RSS. */ 396 enum rte_eth_rx_mq_mode mq_mode; 397 uint32_t max_rx_pkt_len; /**< Only used if JUMBO_FRAME enabled. */ 398 /** Maximum allowed size of LRO aggregated packet. */ 399 uint32_t max_lro_pkt_size; 400 uint16_t split_hdr_size; /**< hdr buf size (header_split enabled).*/ 401 /** 402 * Per-port Rx offloads to be set using DEV_RX_OFFLOAD_* flags. 403 * Only offloads set on rx_offload_capa field on rte_eth_dev_info 404 * structure are allowed to be set. 405 */ 406 uint64_t offloads; 407 408 uint64_t reserved_64s[2]; /**< Reserved for future fields */ 409 void *reserved_ptrs[2]; /**< Reserved for future fields */ 410 }; 411 412 /** 413 * VLAN types to indicate if it is for single VLAN, inner VLAN or outer VLAN. 414 * Note that single VLAN is treated the same as inner VLAN. 415 */ 416 enum rte_vlan_type { 417 ETH_VLAN_TYPE_UNKNOWN = 0, 418 ETH_VLAN_TYPE_INNER, /**< Inner VLAN. */ 419 ETH_VLAN_TYPE_OUTER, /**< Single VLAN, or outer VLAN. */ 420 ETH_VLAN_TYPE_MAX, 421 }; 422 423 /** 424 * A structure used to describe a vlan filter. 425 * If the bit corresponding to a VID is set, such VID is on. 426 */ 427 struct rte_vlan_filter_conf { 428 uint64_t ids[64]; 429 }; 430 431 /** 432 * A structure used to configure the Receive Side Scaling (RSS) feature 433 * of an Ethernet port. 434 * If not NULL, the *rss_key* pointer of the *rss_conf* structure points 435 * to an array holding the RSS key to use for hashing specific header 436 * fields of received packets. The length of this array should be indicated 437 * by *rss_key_len* below. Otherwise, a default random hash key is used by 438 * the device driver. 439 * 440 * The *rss_key_len* field of the *rss_conf* structure indicates the length 441 * in bytes of the array pointed by *rss_key*. To be compatible, this length 442 * will be checked in i40e only. Others assume 40 bytes to be used as before. 443 * 444 * The *rss_hf* field of the *rss_conf* structure indicates the different 445 * types of IPv4/IPv6 packets to which the RSS hashing must be applied. 446 * Supplying an *rss_hf* equal to zero disables the RSS feature. 447 */ 448 struct rte_eth_rss_conf { 449 uint8_t *rss_key; /**< If not NULL, 40-byte hash key. */ 450 uint8_t rss_key_len; /**< hash key length in bytes. */ 451 uint64_t rss_hf; /**< Hash functions to apply - see below. */ 452 }; 453 454 /* 455 * A packet can be identified by hardware as different flow types. Different 456 * NIC hardware may support different flow types. 457 * Basically, the NIC hardware identifies the flow type as deep protocol as 458 * possible, and exclusively. For example, if a packet is identified as 459 * 'RTE_ETH_FLOW_NONFRAG_IPV4_TCP', it will not be any of other flow types, 460 * though it is an actual IPV4 packet. 461 */ 462 #define RTE_ETH_FLOW_UNKNOWN 0 463 #define RTE_ETH_FLOW_RAW 1 464 #define RTE_ETH_FLOW_IPV4 2 465 #define RTE_ETH_FLOW_FRAG_IPV4 3 466 #define RTE_ETH_FLOW_NONFRAG_IPV4_TCP 4 467 #define RTE_ETH_FLOW_NONFRAG_IPV4_UDP 5 468 #define RTE_ETH_FLOW_NONFRAG_IPV4_SCTP 6 469 #define RTE_ETH_FLOW_NONFRAG_IPV4_OTHER 7 470 #define RTE_ETH_FLOW_IPV6 8 471 #define RTE_ETH_FLOW_FRAG_IPV6 9 472 #define RTE_ETH_FLOW_NONFRAG_IPV6_TCP 10 473 #define RTE_ETH_FLOW_NONFRAG_IPV6_UDP 11 474 #define RTE_ETH_FLOW_NONFRAG_IPV6_SCTP 12 475 #define RTE_ETH_FLOW_NONFRAG_IPV6_OTHER 13 476 #define RTE_ETH_FLOW_L2_PAYLOAD 14 477 #define RTE_ETH_FLOW_IPV6_EX 15 478 #define RTE_ETH_FLOW_IPV6_TCP_EX 16 479 #define RTE_ETH_FLOW_IPV6_UDP_EX 17 480 #define RTE_ETH_FLOW_PORT 18 481 /**< Consider device port number as a flow differentiator */ 482 #define RTE_ETH_FLOW_VXLAN 19 /**< VXLAN protocol based flow */ 483 #define RTE_ETH_FLOW_GENEVE 20 /**< GENEVE protocol based flow */ 484 #define RTE_ETH_FLOW_NVGRE 21 /**< NVGRE protocol based flow */ 485 #define RTE_ETH_FLOW_VXLAN_GPE 22 /**< VXLAN-GPE protocol based flow */ 486 #define RTE_ETH_FLOW_GTPU 23 /**< GTPU protocol based flow */ 487 #define RTE_ETH_FLOW_MAX 24 488 489 /* 490 * Below macros are defined for RSS offload types, they can be used to 491 * fill rte_eth_rss_conf.rss_hf or rte_flow_action_rss.types. 492 */ 493 #define ETH_RSS_IPV4 (1ULL << 2) 494 #define ETH_RSS_FRAG_IPV4 (1ULL << 3) 495 #define ETH_RSS_NONFRAG_IPV4_TCP (1ULL << 4) 496 #define ETH_RSS_NONFRAG_IPV4_UDP (1ULL << 5) 497 #define ETH_RSS_NONFRAG_IPV4_SCTP (1ULL << 6) 498 #define ETH_RSS_NONFRAG_IPV4_OTHER (1ULL << 7) 499 #define ETH_RSS_IPV6 (1ULL << 8) 500 #define ETH_RSS_FRAG_IPV6 (1ULL << 9) 501 #define ETH_RSS_NONFRAG_IPV6_TCP (1ULL << 10) 502 #define ETH_RSS_NONFRAG_IPV6_UDP (1ULL << 11) 503 #define ETH_RSS_NONFRAG_IPV6_SCTP (1ULL << 12) 504 #define ETH_RSS_NONFRAG_IPV6_OTHER (1ULL << 13) 505 #define ETH_RSS_L2_PAYLOAD (1ULL << 14) 506 #define ETH_RSS_IPV6_EX (1ULL << 15) 507 #define ETH_RSS_IPV6_TCP_EX (1ULL << 16) 508 #define ETH_RSS_IPV6_UDP_EX (1ULL << 17) 509 #define ETH_RSS_PORT (1ULL << 18) 510 #define ETH_RSS_VXLAN (1ULL << 19) 511 #define ETH_RSS_GENEVE (1ULL << 20) 512 #define ETH_RSS_NVGRE (1ULL << 21) 513 #define ETH_RSS_GTPU (1ULL << 23) 514 515 /* 516 * We use the following macros to combine with above ETH_RSS_* for 517 * more specific input set selection. These bits are defined starting 518 * from the high end of the 64 bits. 519 * Note: If we use above ETH_RSS_* without SRC/DST_ONLY, it represents 520 * both SRC and DST are taken into account. If SRC_ONLY and DST_ONLY of 521 * the same level are used simultaneously, it is the same case as none of 522 * them are added. 523 */ 524 #define ETH_RSS_L3_SRC_ONLY (1ULL << 63) 525 #define ETH_RSS_L3_DST_ONLY (1ULL << 62) 526 #define ETH_RSS_L4_SRC_ONLY (1ULL << 61) 527 #define ETH_RSS_L4_DST_ONLY (1ULL << 60) 528 529 /** 530 * For input set change of hash filter, if SRC_ONLY and DST_ONLY of 531 * the same level are used simultaneously, it is the same case as 532 * none of them are added. 533 * 534 * @param rss_hf 535 * RSS types with SRC/DST_ONLY. 536 * @return 537 * RSS types. 538 */ 539 static inline uint64_t 540 rte_eth_rss_hf_refine(uint64_t rss_hf) 541 { 542 if ((rss_hf & ETH_RSS_L3_SRC_ONLY) && (rss_hf & ETH_RSS_L3_DST_ONLY)) 543 rss_hf &= ~(ETH_RSS_L3_SRC_ONLY | ETH_RSS_L3_DST_ONLY); 544 545 if ((rss_hf & ETH_RSS_L4_SRC_ONLY) && (rss_hf & ETH_RSS_L4_DST_ONLY)) 546 rss_hf &= ~(ETH_RSS_L4_SRC_ONLY | ETH_RSS_L4_DST_ONLY); 547 548 return rss_hf; 549 } 550 551 #define ETH_RSS_IP ( \ 552 ETH_RSS_IPV4 | \ 553 ETH_RSS_FRAG_IPV4 | \ 554 ETH_RSS_NONFRAG_IPV4_OTHER | \ 555 ETH_RSS_IPV6 | \ 556 ETH_RSS_FRAG_IPV6 | \ 557 ETH_RSS_NONFRAG_IPV6_OTHER | \ 558 ETH_RSS_IPV6_EX) 559 560 #define ETH_RSS_UDP ( \ 561 ETH_RSS_NONFRAG_IPV4_UDP | \ 562 ETH_RSS_NONFRAG_IPV6_UDP | \ 563 ETH_RSS_IPV6_UDP_EX) 564 565 #define ETH_RSS_TCP ( \ 566 ETH_RSS_NONFRAG_IPV4_TCP | \ 567 ETH_RSS_NONFRAG_IPV6_TCP | \ 568 ETH_RSS_IPV6_TCP_EX) 569 570 #define ETH_RSS_SCTP ( \ 571 ETH_RSS_NONFRAG_IPV4_SCTP | \ 572 ETH_RSS_NONFRAG_IPV6_SCTP) 573 574 #define ETH_RSS_TUNNEL ( \ 575 ETH_RSS_VXLAN | \ 576 ETH_RSS_GENEVE | \ 577 ETH_RSS_NVGRE) 578 579 /**< Mask of valid RSS hash protocols */ 580 #define ETH_RSS_PROTO_MASK ( \ 581 ETH_RSS_IPV4 | \ 582 ETH_RSS_FRAG_IPV4 | \ 583 ETH_RSS_NONFRAG_IPV4_TCP | \ 584 ETH_RSS_NONFRAG_IPV4_UDP | \ 585 ETH_RSS_NONFRAG_IPV4_SCTP | \ 586 ETH_RSS_NONFRAG_IPV4_OTHER | \ 587 ETH_RSS_IPV6 | \ 588 ETH_RSS_FRAG_IPV6 | \ 589 ETH_RSS_NONFRAG_IPV6_TCP | \ 590 ETH_RSS_NONFRAG_IPV6_UDP | \ 591 ETH_RSS_NONFRAG_IPV6_SCTP | \ 592 ETH_RSS_NONFRAG_IPV6_OTHER | \ 593 ETH_RSS_L2_PAYLOAD | \ 594 ETH_RSS_IPV6_EX | \ 595 ETH_RSS_IPV6_TCP_EX | \ 596 ETH_RSS_IPV6_UDP_EX | \ 597 ETH_RSS_PORT | \ 598 ETH_RSS_VXLAN | \ 599 ETH_RSS_GENEVE | \ 600 ETH_RSS_NVGRE) 601 602 /* 603 * Definitions used for redirection table entry size. 604 * Some RSS RETA sizes may not be supported by some drivers, check the 605 * documentation or the description of relevant functions for more details. 606 */ 607 #define ETH_RSS_RETA_SIZE_64 64 608 #define ETH_RSS_RETA_SIZE_128 128 609 #define ETH_RSS_RETA_SIZE_256 256 610 #define ETH_RSS_RETA_SIZE_512 512 611 #define RTE_RETA_GROUP_SIZE 64 612 613 /* Definitions used for VMDQ and DCB functionality */ 614 #define ETH_VMDQ_MAX_VLAN_FILTERS 64 /**< Maximum nb. of VMDQ vlan filters. */ 615 #define ETH_DCB_NUM_USER_PRIORITIES 8 /**< Maximum nb. of DCB priorities. */ 616 #define ETH_VMDQ_DCB_NUM_QUEUES 128 /**< Maximum nb. of VMDQ DCB queues. */ 617 #define ETH_DCB_NUM_QUEUES 128 /**< Maximum nb. of DCB queues. */ 618 619 /* DCB capability defines */ 620 #define ETH_DCB_PG_SUPPORT 0x00000001 /**< Priority Group(ETS) support. */ 621 #define ETH_DCB_PFC_SUPPORT 0x00000002 /**< Priority Flow Control support. */ 622 623 /* Definitions used for VLAN Offload functionality */ 624 #define ETH_VLAN_STRIP_OFFLOAD 0x0001 /**< VLAN Strip On/Off */ 625 #define ETH_VLAN_FILTER_OFFLOAD 0x0002 /**< VLAN Filter On/Off */ 626 #define ETH_VLAN_EXTEND_OFFLOAD 0x0004 /**< VLAN Extend On/Off */ 627 #define ETH_QINQ_STRIP_OFFLOAD 0x0008 /**< QINQ Strip On/Off */ 628 629 /* Definitions used for mask VLAN setting */ 630 #define ETH_VLAN_STRIP_MASK 0x0001 /**< VLAN Strip setting mask */ 631 #define ETH_VLAN_FILTER_MASK 0x0002 /**< VLAN Filter setting mask*/ 632 #define ETH_VLAN_EXTEND_MASK 0x0004 /**< VLAN Extend setting mask*/ 633 #define ETH_QINQ_STRIP_MASK 0x0008 /**< QINQ Strip setting mask */ 634 #define ETH_VLAN_ID_MAX 0x0FFF /**< VLAN ID is in lower 12 bits*/ 635 636 /* Definitions used for receive MAC address */ 637 #define ETH_NUM_RECEIVE_MAC_ADDR 128 /**< Maximum nb. of receive mac addr. */ 638 639 /* Definitions used for unicast hash */ 640 #define ETH_VMDQ_NUM_UC_HASH_ARRAY 128 /**< Maximum nb. of UC hash array. */ 641 642 /* Definitions used for VMDQ pool rx mode setting */ 643 #define ETH_VMDQ_ACCEPT_UNTAG 0x0001 /**< accept untagged packets. */ 644 #define ETH_VMDQ_ACCEPT_HASH_MC 0x0002 /**< accept packets in multicast table . */ 645 #define ETH_VMDQ_ACCEPT_HASH_UC 0x0004 /**< accept packets in unicast table. */ 646 #define ETH_VMDQ_ACCEPT_BROADCAST 0x0008 /**< accept broadcast packets. */ 647 #define ETH_VMDQ_ACCEPT_MULTICAST 0x0010 /**< multicast promiscuous. */ 648 649 /** Maximum nb. of vlan per mirror rule */ 650 #define ETH_MIRROR_MAX_VLANS 64 651 652 #define ETH_MIRROR_VIRTUAL_POOL_UP 0x01 /**< Virtual Pool uplink Mirroring. */ 653 #define ETH_MIRROR_UPLINK_PORT 0x02 /**< Uplink Port Mirroring. */ 654 #define ETH_MIRROR_DOWNLINK_PORT 0x04 /**< Downlink Port Mirroring. */ 655 #define ETH_MIRROR_VLAN 0x08 /**< VLAN Mirroring. */ 656 #define ETH_MIRROR_VIRTUAL_POOL_DOWN 0x10 /**< Virtual Pool downlink Mirroring. */ 657 658 /** 659 * A structure used to configure VLAN traffic mirror of an Ethernet port. 660 */ 661 struct rte_eth_vlan_mirror { 662 uint64_t vlan_mask; /**< mask for valid VLAN ID. */ 663 /** VLAN ID list for vlan mirroring. */ 664 uint16_t vlan_id[ETH_MIRROR_MAX_VLANS]; 665 }; 666 667 /** 668 * A structure used to configure traffic mirror of an Ethernet port. 669 */ 670 struct rte_eth_mirror_conf { 671 uint8_t rule_type; /**< Mirroring rule type */ 672 uint8_t dst_pool; /**< Destination pool for this mirror rule. */ 673 uint64_t pool_mask; /**< Bitmap of pool for pool mirroring */ 674 /** VLAN ID setting for VLAN mirroring. */ 675 struct rte_eth_vlan_mirror vlan; 676 }; 677 678 /** 679 * A structure used to configure 64 entries of Redirection Table of the 680 * Receive Side Scaling (RSS) feature of an Ethernet port. To configure 681 * more than 64 entries supported by hardware, an array of this structure 682 * is needed. 683 */ 684 struct rte_eth_rss_reta_entry64 { 685 uint64_t mask; 686 /**< Mask bits indicate which entries need to be updated/queried. */ 687 uint16_t reta[RTE_RETA_GROUP_SIZE]; 688 /**< Group of 64 redirection table entries. */ 689 }; 690 691 /** 692 * This enum indicates the possible number of traffic classes 693 * in DCB configurations 694 */ 695 enum rte_eth_nb_tcs { 696 ETH_4_TCS = 4, /**< 4 TCs with DCB. */ 697 ETH_8_TCS = 8 /**< 8 TCs with DCB. */ 698 }; 699 700 /** 701 * This enum indicates the possible number of queue pools 702 * in VMDQ configurations. 703 */ 704 enum rte_eth_nb_pools { 705 ETH_8_POOLS = 8, /**< 8 VMDq pools. */ 706 ETH_16_POOLS = 16, /**< 16 VMDq pools. */ 707 ETH_32_POOLS = 32, /**< 32 VMDq pools. */ 708 ETH_64_POOLS = 64 /**< 64 VMDq pools. */ 709 }; 710 711 /* This structure may be extended in future. */ 712 struct rte_eth_dcb_rx_conf { 713 enum rte_eth_nb_tcs nb_tcs; /**< Possible DCB TCs, 4 or 8 TCs */ 714 /** Traffic class each UP mapped to. */ 715 uint8_t dcb_tc[ETH_DCB_NUM_USER_PRIORITIES]; 716 }; 717 718 struct rte_eth_vmdq_dcb_tx_conf { 719 enum rte_eth_nb_pools nb_queue_pools; /**< With DCB, 16 or 32 pools. */ 720 /** Traffic class each UP mapped to. */ 721 uint8_t dcb_tc[ETH_DCB_NUM_USER_PRIORITIES]; 722 }; 723 724 struct rte_eth_dcb_tx_conf { 725 enum rte_eth_nb_tcs nb_tcs; /**< Possible DCB TCs, 4 or 8 TCs. */ 726 /** Traffic class each UP mapped to. */ 727 uint8_t dcb_tc[ETH_DCB_NUM_USER_PRIORITIES]; 728 }; 729 730 struct rte_eth_vmdq_tx_conf { 731 enum rte_eth_nb_pools nb_queue_pools; /**< VMDq mode, 64 pools. */ 732 }; 733 734 /** 735 * A structure used to configure the VMDQ+DCB feature 736 * of an Ethernet port. 737 * 738 * Using this feature, packets are routed to a pool of queues, based 739 * on the vlan id in the vlan tag, and then to a specific queue within 740 * that pool, using the user priority vlan tag field. 741 * 742 * A default pool may be used, if desired, to route all traffic which 743 * does not match the vlan filter rules. 744 */ 745 struct rte_eth_vmdq_dcb_conf { 746 enum rte_eth_nb_pools nb_queue_pools; /**< With DCB, 16 or 32 pools */ 747 uint8_t enable_default_pool; /**< If non-zero, use a default pool */ 748 uint8_t default_pool; /**< The default pool, if applicable */ 749 uint8_t nb_pool_maps; /**< We can have up to 64 filters/mappings */ 750 struct { 751 uint16_t vlan_id; /**< The vlan id of the received frame */ 752 uint64_t pools; /**< Bitmask of pools for packet rx */ 753 } pool_map[ETH_VMDQ_MAX_VLAN_FILTERS]; /**< VMDq vlan pool maps. */ 754 uint8_t dcb_tc[ETH_DCB_NUM_USER_PRIORITIES]; 755 /**< Selects a queue in a pool */ 756 }; 757 758 /** 759 * A structure used to configure the VMDQ feature of an Ethernet port when 760 * not combined with the DCB feature. 761 * 762 * Using this feature, packets are routed to a pool of queues. By default, 763 * the pool selection is based on the MAC address, the vlan id in the 764 * vlan tag as specified in the pool_map array. 765 * Passing the ETH_VMDQ_ACCEPT_UNTAG in the rx_mode field allows pool 766 * selection using only the MAC address. MAC address to pool mapping is done 767 * using the rte_eth_dev_mac_addr_add function, with the pool parameter 768 * corresponding to the pool id. 769 * 770 * Queue selection within the selected pool will be done using RSS when 771 * it is enabled or revert to the first queue of the pool if not. 772 * 773 * A default pool may be used, if desired, to route all traffic which 774 * does not match the vlan filter rules or any pool MAC address. 775 */ 776 struct rte_eth_vmdq_rx_conf { 777 enum rte_eth_nb_pools nb_queue_pools; /**< VMDq only mode, 8 or 64 pools */ 778 uint8_t enable_default_pool; /**< If non-zero, use a default pool */ 779 uint8_t default_pool; /**< The default pool, if applicable */ 780 uint8_t enable_loop_back; /**< Enable VT loop back */ 781 uint8_t nb_pool_maps; /**< We can have up to 64 filters/mappings */ 782 uint32_t rx_mode; /**< Flags from ETH_VMDQ_ACCEPT_* */ 783 struct { 784 uint16_t vlan_id; /**< The vlan id of the received frame */ 785 uint64_t pools; /**< Bitmask of pools for packet rx */ 786 } pool_map[ETH_VMDQ_MAX_VLAN_FILTERS]; /**< VMDq vlan pool maps. */ 787 }; 788 789 /** 790 * A structure used to configure the TX features of an Ethernet port. 791 */ 792 struct rte_eth_txmode { 793 enum rte_eth_tx_mq_mode mq_mode; /**< TX multi-queues mode. */ 794 /** 795 * Per-port Tx offloads to be set using DEV_TX_OFFLOAD_* flags. 796 * Only offloads set on tx_offload_capa field on rte_eth_dev_info 797 * structure are allowed to be set. 798 */ 799 uint64_t offloads; 800 801 /* For i40e specifically */ 802 uint16_t pvid; 803 __extension__ 804 uint8_t hw_vlan_reject_tagged : 1, 805 /**< If set, reject sending out tagged pkts */ 806 hw_vlan_reject_untagged : 1, 807 /**< If set, reject sending out untagged pkts */ 808 hw_vlan_insert_pvid : 1; 809 /**< If set, enable port based VLAN insertion */ 810 811 uint64_t reserved_64s[2]; /**< Reserved for future fields */ 812 void *reserved_ptrs[2]; /**< Reserved for future fields */ 813 }; 814 815 /** 816 * A structure used to configure an RX ring of an Ethernet port. 817 */ 818 struct rte_eth_rxconf { 819 struct rte_eth_thresh rx_thresh; /**< RX ring threshold registers. */ 820 uint16_t rx_free_thresh; /**< Drives the freeing of RX descriptors. */ 821 uint8_t rx_drop_en; /**< Drop packets if no descriptors are available. */ 822 uint8_t rx_deferred_start; /**< Do not start queue with rte_eth_dev_start(). */ 823 /** 824 * Per-queue Rx offloads to be set using DEV_RX_OFFLOAD_* flags. 825 * Only offloads set on rx_queue_offload_capa or rx_offload_capa 826 * fields on rte_eth_dev_info structure are allowed to be set. 827 */ 828 uint64_t offloads; 829 830 uint64_t reserved_64s[2]; /**< Reserved for future fields */ 831 void *reserved_ptrs[2]; /**< Reserved for future fields */ 832 }; 833 834 /** 835 * A structure used to configure a TX ring of an Ethernet port. 836 */ 837 struct rte_eth_txconf { 838 struct rte_eth_thresh tx_thresh; /**< TX ring threshold registers. */ 839 uint16_t tx_rs_thresh; /**< Drives the setting of RS bit on TXDs. */ 840 uint16_t tx_free_thresh; /**< Start freeing TX buffers if there are 841 less free descriptors than this value. */ 842 843 uint8_t tx_deferred_start; /**< Do not start queue with rte_eth_dev_start(). */ 844 /** 845 * Per-queue Tx offloads to be set using DEV_TX_OFFLOAD_* flags. 846 * Only offloads set on tx_queue_offload_capa or tx_offload_capa 847 * fields on rte_eth_dev_info structure are allowed to be set. 848 */ 849 uint64_t offloads; 850 851 uint64_t reserved_64s[2]; /**< Reserved for future fields */ 852 void *reserved_ptrs[2]; /**< Reserved for future fields */ 853 }; 854 855 /** 856 * @warning 857 * @b EXPERIMENTAL: this API may change, or be removed, without prior notice 858 * 859 * A structure used to return the hairpin capabilities that are supported. 860 */ 861 struct rte_eth_hairpin_cap { 862 /** The max number of hairpin queues (different bindings). */ 863 uint16_t max_nb_queues; 864 /** Max number of Rx queues to be connected to one Tx queue. */ 865 uint16_t max_rx_2_tx; 866 /** Max number of Tx queues to be connected to one Rx queue. */ 867 uint16_t max_tx_2_rx; 868 uint16_t max_nb_desc; /**< The max num of descriptors. */ 869 }; 870 871 #define RTE_ETH_MAX_HAIRPIN_PEERS 32 872 873 /** 874 * @warning 875 * @b EXPERIMENTAL: this API may change, or be removed, without prior notice 876 * 877 * A structure used to hold hairpin peer data. 878 */ 879 struct rte_eth_hairpin_peer { 880 uint16_t port; /**< Peer port. */ 881 uint16_t queue; /**< Peer queue. */ 882 }; 883 884 /** 885 * @warning 886 * @b EXPERIMENTAL: this API may change, or be removed, without prior notice 887 * 888 * A structure used to configure hairpin binding. 889 */ 890 struct rte_eth_hairpin_conf { 891 uint16_t peer_count; /**< The number of peers. */ 892 struct rte_eth_hairpin_peer peers[RTE_ETH_MAX_HAIRPIN_PEERS]; 893 }; 894 895 /** 896 * A structure contains information about HW descriptor ring limitations. 897 */ 898 struct rte_eth_desc_lim { 899 uint16_t nb_max; /**< Max allowed number of descriptors. */ 900 uint16_t nb_min; /**< Min allowed number of descriptors. */ 901 uint16_t nb_align; /**< Number of descriptors should be aligned to. */ 902 903 /** 904 * Max allowed number of segments per whole packet. 905 * 906 * - For TSO packet this is the total number of data descriptors allowed 907 * by device. 908 * 909 * @see nb_mtu_seg_max 910 */ 911 uint16_t nb_seg_max; 912 913 /** 914 * Max number of segments per one MTU. 915 * 916 * - For non-TSO packet, this is the maximum allowed number of segments 917 * in a single transmit packet. 918 * 919 * - For TSO packet each segment within the TSO may span up to this 920 * value. 921 * 922 * @see nb_seg_max 923 */ 924 uint16_t nb_mtu_seg_max; 925 }; 926 927 /** 928 * This enum indicates the flow control mode 929 */ 930 enum rte_eth_fc_mode { 931 RTE_FC_NONE = 0, /**< Disable flow control. */ 932 RTE_FC_RX_PAUSE, /**< RX pause frame, enable flowctrl on TX side. */ 933 RTE_FC_TX_PAUSE, /**< TX pause frame, enable flowctrl on RX side. */ 934 RTE_FC_FULL /**< Enable flow control on both side. */ 935 }; 936 937 /** 938 * A structure used to configure Ethernet flow control parameter. 939 * These parameters will be configured into the register of the NIC. 940 * Please refer to the corresponding data sheet for proper value. 941 */ 942 struct rte_eth_fc_conf { 943 uint32_t high_water; /**< High threshold value to trigger XOFF */ 944 uint32_t low_water; /**< Low threshold value to trigger XON */ 945 uint16_t pause_time; /**< Pause quota in the Pause frame */ 946 uint16_t send_xon; /**< Is XON frame need be sent */ 947 enum rte_eth_fc_mode mode; /**< Link flow control mode */ 948 uint8_t mac_ctrl_frame_fwd; /**< Forward MAC control frames */ 949 uint8_t autoneg; /**< Use Pause autoneg */ 950 }; 951 952 /** 953 * A structure used to configure Ethernet priority flow control parameter. 954 * These parameters will be configured into the register of the NIC. 955 * Please refer to the corresponding data sheet for proper value. 956 */ 957 struct rte_eth_pfc_conf { 958 struct rte_eth_fc_conf fc; /**< General flow control parameter. */ 959 uint8_t priority; /**< VLAN User Priority. */ 960 }; 961 962 /** 963 * Tunneled type. 964 */ 965 enum rte_eth_tunnel_type { 966 RTE_TUNNEL_TYPE_NONE = 0, 967 RTE_TUNNEL_TYPE_VXLAN, 968 RTE_TUNNEL_TYPE_GENEVE, 969 RTE_TUNNEL_TYPE_TEREDO, 970 RTE_TUNNEL_TYPE_NVGRE, 971 RTE_TUNNEL_TYPE_IP_IN_GRE, 972 RTE_L2_TUNNEL_TYPE_E_TAG, 973 RTE_TUNNEL_TYPE_VXLAN_GPE, 974 RTE_TUNNEL_TYPE_MAX, 975 }; 976 977 /* Deprecated API file for rte_eth_dev_filter_* functions */ 978 #include "rte_eth_ctrl.h" 979 980 /** 981 * Memory space that can be configured to store Flow Director filters 982 * in the board memory. 983 */ 984 enum rte_fdir_pballoc_type { 985 RTE_FDIR_PBALLOC_64K = 0, /**< 64k. */ 986 RTE_FDIR_PBALLOC_128K, /**< 128k. */ 987 RTE_FDIR_PBALLOC_256K, /**< 256k. */ 988 }; 989 990 /** 991 * Select report mode of FDIR hash information in RX descriptors. 992 */ 993 enum rte_fdir_status_mode { 994 RTE_FDIR_NO_REPORT_STATUS = 0, /**< Never report FDIR hash. */ 995 RTE_FDIR_REPORT_STATUS, /**< Only report FDIR hash for matching pkts. */ 996 RTE_FDIR_REPORT_STATUS_ALWAYS, /**< Always report FDIR hash. */ 997 }; 998 999 /** 1000 * A structure used to configure the Flow Director (FDIR) feature 1001 * of an Ethernet port. 1002 * 1003 * If mode is RTE_FDIR_MODE_NONE, the pballoc value is ignored. 1004 */ 1005 struct rte_fdir_conf { 1006 enum rte_fdir_mode mode; /**< Flow Director mode. */ 1007 enum rte_fdir_pballoc_type pballoc; /**< Space for FDIR filters. */ 1008 enum rte_fdir_status_mode status; /**< How to report FDIR hash. */ 1009 /** RX queue of packets matching a "drop" filter in perfect mode. */ 1010 uint8_t drop_queue; 1011 struct rte_eth_fdir_masks mask; 1012 struct rte_eth_fdir_flex_conf flex_conf; 1013 /**< Flex payload configuration. */ 1014 }; 1015 1016 /** 1017 * UDP tunneling configuration. 1018 * Used to config the UDP port for a type of tunnel. 1019 * NICs need the UDP port to identify the tunnel type. 1020 * Normally a type of tunnel has a default UDP port, this structure can be used 1021 * in case if the users want to change or support more UDP port. 1022 */ 1023 struct rte_eth_udp_tunnel { 1024 uint16_t udp_port; /**< UDP port used for the tunnel. */ 1025 uint8_t prot_type; /**< Tunnel type. Defined in rte_eth_tunnel_type. */ 1026 }; 1027 1028 /** 1029 * A structure used to enable/disable specific device interrupts. 1030 */ 1031 struct rte_intr_conf { 1032 /** enable/disable lsc interrupt. 0 (default) - disable, 1 enable */ 1033 uint32_t lsc:1; 1034 /** enable/disable rxq interrupt. 0 (default) - disable, 1 enable */ 1035 uint32_t rxq:1; 1036 /** enable/disable rmv interrupt. 0 (default) - disable, 1 enable */ 1037 uint32_t rmv:1; 1038 }; 1039 1040 /** 1041 * A structure used to configure an Ethernet port. 1042 * Depending upon the RX multi-queue mode, extra advanced 1043 * configuration settings may be needed. 1044 */ 1045 struct rte_eth_conf { 1046 uint32_t link_speeds; /**< bitmap of ETH_LINK_SPEED_XXX of speeds to be 1047 used. ETH_LINK_SPEED_FIXED disables link 1048 autonegotiation, and a unique speed shall be 1049 set. Otherwise, the bitmap defines the set of 1050 speeds to be advertised. If the special value 1051 ETH_LINK_SPEED_AUTONEG (0) is used, all speeds 1052 supported are advertised. */ 1053 struct rte_eth_rxmode rxmode; /**< Port RX configuration. */ 1054 struct rte_eth_txmode txmode; /**< Port TX configuration. */ 1055 uint32_t lpbk_mode; /**< Loopback operation mode. By default the value 1056 is 0, meaning the loopback mode is disabled. 1057 Read the datasheet of given ethernet controller 1058 for details. The possible values of this field 1059 are defined in implementation of each driver. */ 1060 struct { 1061 struct rte_eth_rss_conf rss_conf; /**< Port RSS configuration */ 1062 struct rte_eth_vmdq_dcb_conf vmdq_dcb_conf; 1063 /**< Port vmdq+dcb configuration. */ 1064 struct rte_eth_dcb_rx_conf dcb_rx_conf; 1065 /**< Port dcb RX configuration. */ 1066 struct rte_eth_vmdq_rx_conf vmdq_rx_conf; 1067 /**< Port vmdq RX configuration. */ 1068 } rx_adv_conf; /**< Port RX filtering configuration. */ 1069 union { 1070 struct rte_eth_vmdq_dcb_tx_conf vmdq_dcb_tx_conf; 1071 /**< Port vmdq+dcb TX configuration. */ 1072 struct rte_eth_dcb_tx_conf dcb_tx_conf; 1073 /**< Port dcb TX configuration. */ 1074 struct rte_eth_vmdq_tx_conf vmdq_tx_conf; 1075 /**< Port vmdq TX configuration. */ 1076 } tx_adv_conf; /**< Port TX DCB configuration (union). */ 1077 /** Currently,Priority Flow Control(PFC) are supported,if DCB with PFC 1078 is needed,and the variable must be set ETH_DCB_PFC_SUPPORT. */ 1079 uint32_t dcb_capability_en; 1080 struct rte_fdir_conf fdir_conf; /**< FDIR configuration. DEPRECATED */ 1081 struct rte_intr_conf intr_conf; /**< Interrupt mode configuration. */ 1082 }; 1083 1084 /** 1085 * RX offload capabilities of a device. 1086 */ 1087 #define DEV_RX_OFFLOAD_VLAN_STRIP 0x00000001 1088 #define DEV_RX_OFFLOAD_IPV4_CKSUM 0x00000002 1089 #define DEV_RX_OFFLOAD_UDP_CKSUM 0x00000004 1090 #define DEV_RX_OFFLOAD_TCP_CKSUM 0x00000008 1091 #define DEV_RX_OFFLOAD_TCP_LRO 0x00000010 1092 #define DEV_RX_OFFLOAD_QINQ_STRIP 0x00000020 1093 #define DEV_RX_OFFLOAD_OUTER_IPV4_CKSUM 0x00000040 1094 #define DEV_RX_OFFLOAD_MACSEC_STRIP 0x00000080 1095 #define DEV_RX_OFFLOAD_HEADER_SPLIT 0x00000100 1096 #define DEV_RX_OFFLOAD_VLAN_FILTER 0x00000200 1097 #define DEV_RX_OFFLOAD_VLAN_EXTEND 0x00000400 1098 #define DEV_RX_OFFLOAD_JUMBO_FRAME 0x00000800 1099 #define DEV_RX_OFFLOAD_SCATTER 0x00002000 1100 #define DEV_RX_OFFLOAD_TIMESTAMP 0x00004000 1101 #define DEV_RX_OFFLOAD_SECURITY 0x00008000 1102 #define DEV_RX_OFFLOAD_KEEP_CRC 0x00010000 1103 #define DEV_RX_OFFLOAD_SCTP_CKSUM 0x00020000 1104 #define DEV_RX_OFFLOAD_OUTER_UDP_CKSUM 0x00040000 1105 #define DEV_RX_OFFLOAD_RSS_HASH 0x00080000 1106 1107 #define DEV_RX_OFFLOAD_CHECKSUM (DEV_RX_OFFLOAD_IPV4_CKSUM | \ 1108 DEV_RX_OFFLOAD_UDP_CKSUM | \ 1109 DEV_RX_OFFLOAD_TCP_CKSUM) 1110 #define DEV_RX_OFFLOAD_VLAN (DEV_RX_OFFLOAD_VLAN_STRIP | \ 1111 DEV_RX_OFFLOAD_VLAN_FILTER | \ 1112 DEV_RX_OFFLOAD_VLAN_EXTEND | \ 1113 DEV_RX_OFFLOAD_QINQ_STRIP) 1114 1115 /* 1116 * If new Rx offload capabilities are defined, they also must be 1117 * mentioned in rte_rx_offload_names in rte_ethdev.c file. 1118 */ 1119 1120 /** 1121 * TX offload capabilities of a device. 1122 */ 1123 #define DEV_TX_OFFLOAD_VLAN_INSERT 0x00000001 1124 #define DEV_TX_OFFLOAD_IPV4_CKSUM 0x00000002 1125 #define DEV_TX_OFFLOAD_UDP_CKSUM 0x00000004 1126 #define DEV_TX_OFFLOAD_TCP_CKSUM 0x00000008 1127 #define DEV_TX_OFFLOAD_SCTP_CKSUM 0x00000010 1128 #define DEV_TX_OFFLOAD_TCP_TSO 0x00000020 1129 #define DEV_TX_OFFLOAD_UDP_TSO 0x00000040 1130 #define DEV_TX_OFFLOAD_OUTER_IPV4_CKSUM 0x00000080 /**< Used for tunneling packet. */ 1131 #define DEV_TX_OFFLOAD_QINQ_INSERT 0x00000100 1132 #define DEV_TX_OFFLOAD_VXLAN_TNL_TSO 0x00000200 /**< Used for tunneling packet. */ 1133 #define DEV_TX_OFFLOAD_GRE_TNL_TSO 0x00000400 /**< Used for tunneling packet. */ 1134 #define DEV_TX_OFFLOAD_IPIP_TNL_TSO 0x00000800 /**< Used for tunneling packet. */ 1135 #define DEV_TX_OFFLOAD_GENEVE_TNL_TSO 0x00001000 /**< Used for tunneling packet. */ 1136 #define DEV_TX_OFFLOAD_MACSEC_INSERT 0x00002000 1137 #define DEV_TX_OFFLOAD_MT_LOCKFREE 0x00004000 1138 /**< Multiple threads can invoke rte_eth_tx_burst() concurrently on the same 1139 * tx queue without SW lock. 1140 */ 1141 #define DEV_TX_OFFLOAD_MULTI_SEGS 0x00008000 1142 /**< Device supports multi segment send. */ 1143 #define DEV_TX_OFFLOAD_MBUF_FAST_FREE 0x00010000 1144 /**< Device supports optimization for fast release of mbufs. 1145 * When set application must guarantee that per-queue all mbufs comes from 1146 * the same mempool and has refcnt = 1. 1147 */ 1148 #define DEV_TX_OFFLOAD_SECURITY 0x00020000 1149 /** 1150 * Device supports generic UDP tunneled packet TSO. 1151 * Application must set PKT_TX_TUNNEL_UDP and other mbuf fields required 1152 * for tunnel TSO. 1153 */ 1154 #define DEV_TX_OFFLOAD_UDP_TNL_TSO 0x00040000 1155 /** 1156 * Device supports generic IP tunneled packet TSO. 1157 * Application must set PKT_TX_TUNNEL_IP and other mbuf fields required 1158 * for tunnel TSO. 1159 */ 1160 #define DEV_TX_OFFLOAD_IP_TNL_TSO 0x00080000 1161 /** Device supports outer UDP checksum */ 1162 #define DEV_TX_OFFLOAD_OUTER_UDP_CKSUM 0x00100000 1163 1164 #define RTE_ETH_DEV_CAPA_RUNTIME_RX_QUEUE_SETUP 0x00000001 1165 /**< Device supports Rx queue setup after device started*/ 1166 #define RTE_ETH_DEV_CAPA_RUNTIME_TX_QUEUE_SETUP 0x00000002 1167 /**< Device supports Tx queue setup after device started*/ 1168 1169 /* 1170 * If new Tx offload capabilities are defined, they also must be 1171 * mentioned in rte_tx_offload_names in rte_ethdev.c file. 1172 */ 1173 1174 /* 1175 * Fallback default preferred Rx/Tx port parameters. 1176 * These are used if an application requests default parameters 1177 * but the PMD does not provide preferred values. 1178 */ 1179 #define RTE_ETH_DEV_FALLBACK_RX_RINGSIZE 512 1180 #define RTE_ETH_DEV_FALLBACK_TX_RINGSIZE 512 1181 #define RTE_ETH_DEV_FALLBACK_RX_NBQUEUES 1 1182 #define RTE_ETH_DEV_FALLBACK_TX_NBQUEUES 1 1183 1184 /** 1185 * Preferred Rx/Tx port parameters. 1186 * There are separate instances of this structure for transmission 1187 * and reception respectively. 1188 */ 1189 struct rte_eth_dev_portconf { 1190 uint16_t burst_size; /**< Device-preferred burst size */ 1191 uint16_t ring_size; /**< Device-preferred size of queue rings */ 1192 uint16_t nb_queues; /**< Device-preferred number of queues */ 1193 }; 1194 1195 /** 1196 * Default values for switch domain id when ethdev does not support switch 1197 * domain definitions. 1198 */ 1199 #define RTE_ETH_DEV_SWITCH_DOMAIN_ID_INVALID (UINT16_MAX) 1200 1201 /** 1202 * Ethernet device associated switch information 1203 */ 1204 struct rte_eth_switch_info { 1205 const char *name; /**< switch name */ 1206 uint16_t domain_id; /**< switch domain id */ 1207 uint16_t port_id; 1208 /**< 1209 * mapping to the devices physical switch port as enumerated from the 1210 * perspective of the embedded interconnect/switch. For SR-IOV enabled 1211 * device this may correspond to the VF_ID of each virtual function, 1212 * but each driver should explicitly define the mapping of switch 1213 * port identifier to that physical interconnect/switch 1214 */ 1215 }; 1216 1217 /** 1218 * Ethernet device information 1219 */ 1220 1221 /** 1222 * A structure used to retrieve the contextual information of 1223 * an Ethernet device, such as the controlling driver of the 1224 * device, etc... 1225 */ 1226 struct rte_eth_dev_info { 1227 struct rte_device *device; /** Generic device information */ 1228 const char *driver_name; /**< Device Driver name. */ 1229 unsigned int if_index; /**< Index to bound host interface, or 0 if none. 1230 Use if_indextoname() to translate into an interface name. */ 1231 uint16_t min_mtu; /**< Minimum MTU allowed */ 1232 uint16_t max_mtu; /**< Maximum MTU allowed */ 1233 const uint32_t *dev_flags; /**< Device flags */ 1234 uint32_t min_rx_bufsize; /**< Minimum size of RX buffer. */ 1235 uint32_t max_rx_pktlen; /**< Maximum configurable length of RX pkt. */ 1236 /** Maximum configurable size of LRO aggregated packet. */ 1237 uint32_t max_lro_pkt_size; 1238 uint16_t max_rx_queues; /**< Maximum number of RX queues. */ 1239 uint16_t max_tx_queues; /**< Maximum number of TX queues. */ 1240 uint32_t max_mac_addrs; /**< Maximum number of MAC addresses. */ 1241 uint32_t max_hash_mac_addrs; 1242 /** Maximum number of hash MAC addresses for MTA and UTA. */ 1243 uint16_t max_vfs; /**< Maximum number of VFs. */ 1244 uint16_t max_vmdq_pools; /**< Maximum number of VMDq pools. */ 1245 uint64_t rx_offload_capa; 1246 /**< All RX offload capabilities including all per-queue ones */ 1247 uint64_t tx_offload_capa; 1248 /**< All TX offload capabilities including all per-queue ones */ 1249 uint64_t rx_queue_offload_capa; 1250 /**< Device per-queue RX offload capabilities. */ 1251 uint64_t tx_queue_offload_capa; 1252 /**< Device per-queue TX offload capabilities. */ 1253 uint16_t reta_size; 1254 /**< Device redirection table size, the total number of entries. */ 1255 uint8_t hash_key_size; /**< Hash key size in bytes */ 1256 /** Bit mask of RSS offloads, the bit offset also means flow type */ 1257 uint64_t flow_type_rss_offloads; 1258 struct rte_eth_rxconf default_rxconf; /**< Default RX configuration */ 1259 struct rte_eth_txconf default_txconf; /**< Default TX configuration */ 1260 uint16_t vmdq_queue_base; /**< First queue ID for VMDQ pools. */ 1261 uint16_t vmdq_queue_num; /**< Queue number for VMDQ pools. */ 1262 uint16_t vmdq_pool_base; /**< First ID of VMDQ pools. */ 1263 struct rte_eth_desc_lim rx_desc_lim; /**< RX descriptors limits */ 1264 struct rte_eth_desc_lim tx_desc_lim; /**< TX descriptors limits */ 1265 uint32_t speed_capa; /**< Supported speeds bitmap (ETH_LINK_SPEED_). */ 1266 /** Configured number of rx/tx queues */ 1267 uint16_t nb_rx_queues; /**< Number of RX queues. */ 1268 uint16_t nb_tx_queues; /**< Number of TX queues. */ 1269 /** Rx parameter recommendations */ 1270 struct rte_eth_dev_portconf default_rxportconf; 1271 /** Tx parameter recommendations */ 1272 struct rte_eth_dev_portconf default_txportconf; 1273 /** Generic device capabilities (RTE_ETH_DEV_CAPA_). */ 1274 uint64_t dev_capa; 1275 /** 1276 * Switching information for ports on a device with a 1277 * embedded managed interconnect/switch. 1278 */ 1279 struct rte_eth_switch_info switch_info; 1280 1281 uint64_t reserved_64s[2]; /**< Reserved for future fields */ 1282 void *reserved_ptrs[2]; /**< Reserved for future fields */ 1283 }; 1284 1285 /** 1286 * Ethernet device RX queue information structure. 1287 * Used to retrieve information about configured queue. 1288 */ 1289 struct rte_eth_rxq_info { 1290 struct rte_mempool *mp; /**< mempool used by that queue. */ 1291 struct rte_eth_rxconf conf; /**< queue config parameters. */ 1292 uint8_t scattered_rx; /**< scattered packets RX supported. */ 1293 uint16_t nb_desc; /**< configured number of RXDs. */ 1294 } __rte_cache_min_aligned; 1295 1296 /** 1297 * Ethernet device TX queue information structure. 1298 * Used to retrieve information about configured queue. 1299 */ 1300 struct rte_eth_txq_info { 1301 struct rte_eth_txconf conf; /**< queue config parameters. */ 1302 uint16_t nb_desc; /**< configured number of TXDs. */ 1303 } __rte_cache_min_aligned; 1304 1305 /* Generic Burst mode flag definition, values can be ORed. */ 1306 1307 /** 1308 * If the queues have different burst mode description, this bit will be set 1309 * by PMD, then the application can iterate to retrieve burst description for 1310 * all other queues. 1311 */ 1312 #define RTE_ETH_BURST_FLAG_PER_QUEUE (1ULL << 0) 1313 1314 /** 1315 * Ethernet device RX/TX queue packet burst mode information structure. 1316 * Used to retrieve information about packet burst mode setting. 1317 */ 1318 struct rte_eth_burst_mode { 1319 uint64_t flags; /**< The ORed values of RTE_ETH_BURST_FLAG_xxx */ 1320 1321 #define RTE_ETH_BURST_MODE_INFO_SIZE 1024 /**< Maximum size for information */ 1322 char info[RTE_ETH_BURST_MODE_INFO_SIZE]; /**< burst mode information */ 1323 }; 1324 1325 /** Maximum name length for extended statistics counters */ 1326 #define RTE_ETH_XSTATS_NAME_SIZE 64 1327 1328 /** 1329 * An Ethernet device extended statistic structure 1330 * 1331 * This structure is used by rte_eth_xstats_get() to provide 1332 * statistics that are not provided in the generic *rte_eth_stats* 1333 * structure. 1334 * It maps a name id, corresponding to an index in the array returned 1335 * by rte_eth_xstats_get_names(), to a statistic value. 1336 */ 1337 struct rte_eth_xstat { 1338 uint64_t id; /**< The index in xstats name array. */ 1339 uint64_t value; /**< The statistic counter value. */ 1340 }; 1341 1342 /** 1343 * A name element for extended statistics. 1344 * 1345 * An array of this structure is returned by rte_eth_xstats_get_names(). 1346 * It lists the names of extended statistics for a PMD. The *rte_eth_xstat* 1347 * structure references these names by their array index. 1348 */ 1349 struct rte_eth_xstat_name { 1350 char name[RTE_ETH_XSTATS_NAME_SIZE]; /**< The statistic name. */ 1351 }; 1352 1353 #define ETH_DCB_NUM_TCS 8 1354 #define ETH_MAX_VMDQ_POOL 64 1355 1356 /** 1357 * A structure used to get the information of queue and 1358 * TC mapping on both TX and RX paths. 1359 */ 1360 struct rte_eth_dcb_tc_queue_mapping { 1361 /** rx queues assigned to tc per Pool */ 1362 struct { 1363 uint8_t base; 1364 uint8_t nb_queue; 1365 } tc_rxq[ETH_MAX_VMDQ_POOL][ETH_DCB_NUM_TCS]; 1366 /** rx queues assigned to tc per Pool */ 1367 struct { 1368 uint8_t base; 1369 uint8_t nb_queue; 1370 } tc_txq[ETH_MAX_VMDQ_POOL][ETH_DCB_NUM_TCS]; 1371 }; 1372 1373 /** 1374 * A structure used to get the information of DCB. 1375 * It includes TC UP mapping and queue TC mapping. 1376 */ 1377 struct rte_eth_dcb_info { 1378 uint8_t nb_tcs; /**< number of TCs */ 1379 uint8_t prio_tc[ETH_DCB_NUM_USER_PRIORITIES]; /**< Priority to tc */ 1380 uint8_t tc_bws[ETH_DCB_NUM_TCS]; /**< TX BW percentage for each TC */ 1381 /** rx queues assigned to tc */ 1382 struct rte_eth_dcb_tc_queue_mapping tc_queue; 1383 }; 1384 1385 #define RTE_ETH_ALL RTE_MAX_ETHPORTS 1386 1387 /* Macros to check for valid port */ 1388 #define RTE_ETH_VALID_PORTID_OR_ERR_RET(port_id, retval) do { \ 1389 if (!rte_eth_dev_is_valid_port(port_id)) { \ 1390 RTE_ETHDEV_LOG(ERR, "Invalid port_id=%u\n", port_id); \ 1391 return retval; \ 1392 } \ 1393 } while (0) 1394 1395 #define RTE_ETH_VALID_PORTID_OR_RET(port_id) do { \ 1396 if (!rte_eth_dev_is_valid_port(port_id)) { \ 1397 RTE_ETHDEV_LOG(ERR, "Invalid port_id=%u\n", port_id); \ 1398 return; \ 1399 } \ 1400 } while (0) 1401 1402 /** 1403 * l2 tunnel configuration. 1404 */ 1405 1406 /**< l2 tunnel enable mask */ 1407 #define ETH_L2_TUNNEL_ENABLE_MASK 0x00000001 1408 /**< l2 tunnel insertion mask */ 1409 #define ETH_L2_TUNNEL_INSERTION_MASK 0x00000002 1410 /**< l2 tunnel stripping mask */ 1411 #define ETH_L2_TUNNEL_STRIPPING_MASK 0x00000004 1412 /**< l2 tunnel forwarding mask */ 1413 #define ETH_L2_TUNNEL_FORWARDING_MASK 0x00000008 1414 1415 /** 1416 * Function type used for RX packet processing packet callbacks. 1417 * 1418 * The callback function is called on RX with a burst of packets that have 1419 * been received on the given port and queue. 1420 * 1421 * @param port_id 1422 * The Ethernet port on which RX is being performed. 1423 * @param queue 1424 * The queue on the Ethernet port which is being used to receive the packets. 1425 * @param pkts 1426 * The burst of packets that have just been received. 1427 * @param nb_pkts 1428 * The number of packets in the burst pointed to by "pkts". 1429 * @param max_pkts 1430 * The max number of packets that can be stored in the "pkts" array. 1431 * @param user_param 1432 * The arbitrary user parameter passed in by the application when the callback 1433 * was originally configured. 1434 * @return 1435 * The number of packets returned to the user. 1436 */ 1437 typedef uint16_t (*rte_rx_callback_fn)(uint16_t port_id, uint16_t queue, 1438 struct rte_mbuf *pkts[], uint16_t nb_pkts, uint16_t max_pkts, 1439 void *user_param); 1440 1441 /** 1442 * Function type used for TX packet processing packet callbacks. 1443 * 1444 * The callback function is called on TX with a burst of packets immediately 1445 * before the packets are put onto the hardware queue for transmission. 1446 * 1447 * @param port_id 1448 * The Ethernet port on which TX is being performed. 1449 * @param queue 1450 * The queue on the Ethernet port which is being used to transmit the packets. 1451 * @param pkts 1452 * The burst of packets that are about to be transmitted. 1453 * @param nb_pkts 1454 * The number of packets in the burst pointed to by "pkts". 1455 * @param user_param 1456 * The arbitrary user parameter passed in by the application when the callback 1457 * was originally configured. 1458 * @return 1459 * The number of packets to be written to the NIC. 1460 */ 1461 typedef uint16_t (*rte_tx_callback_fn)(uint16_t port_id, uint16_t queue, 1462 struct rte_mbuf *pkts[], uint16_t nb_pkts, void *user_param); 1463 1464 /** 1465 * Possible states of an ethdev port. 1466 */ 1467 enum rte_eth_dev_state { 1468 /** Device is unused before being probed. */ 1469 RTE_ETH_DEV_UNUSED = 0, 1470 /** Device is attached when allocated in probing. */ 1471 RTE_ETH_DEV_ATTACHED, 1472 /** Device is in removed state when plug-out is detected. */ 1473 RTE_ETH_DEV_REMOVED, 1474 }; 1475 1476 struct rte_eth_dev_sriov { 1477 uint8_t active; /**< SRIOV is active with 16, 32 or 64 pools */ 1478 uint8_t nb_q_per_pool; /**< rx queue number per pool */ 1479 uint16_t def_vmdq_idx; /**< Default pool num used for PF */ 1480 uint16_t def_pool_q_idx; /**< Default pool queue start reg index */ 1481 }; 1482 #define RTE_ETH_DEV_SRIOV(dev) ((dev)->data->sriov) 1483 1484 #define RTE_ETH_NAME_MAX_LEN RTE_DEV_NAME_MAX_LEN 1485 1486 #define RTE_ETH_DEV_NO_OWNER 0 1487 1488 #define RTE_ETH_MAX_OWNER_NAME_LEN 64 1489 1490 struct rte_eth_dev_owner { 1491 uint64_t id; /**< The owner unique identifier. */ 1492 char name[RTE_ETH_MAX_OWNER_NAME_LEN]; /**< The owner name. */ 1493 }; 1494 1495 /** 1496 * Port is released (i.e. totally freed and data erased) on close. 1497 * Temporary flag for PMD migration to new rte_eth_dev_close() behaviour. 1498 */ 1499 #define RTE_ETH_DEV_CLOSE_REMOVE 0x0001 1500 /** Device supports link state interrupt */ 1501 #define RTE_ETH_DEV_INTR_LSC 0x0002 1502 /** Device is a bonded slave */ 1503 #define RTE_ETH_DEV_BONDED_SLAVE 0x0004 1504 /** Device supports device removal interrupt */ 1505 #define RTE_ETH_DEV_INTR_RMV 0x0008 1506 /** Device is port representor */ 1507 #define RTE_ETH_DEV_REPRESENTOR 0x0010 1508 /** Device does not support MAC change after started */ 1509 #define RTE_ETH_DEV_NOLIVE_MAC_ADDR 0x0020 1510 1511 /** 1512 * Iterates over valid ethdev ports owned by a specific owner. 1513 * 1514 * @param port_id 1515 * The id of the next possible valid owned port. 1516 * @param owner_id 1517 * The owner identifier. 1518 * RTE_ETH_DEV_NO_OWNER means iterate over all valid ownerless ports. 1519 * @return 1520 * Next valid port id owned by owner_id, RTE_MAX_ETHPORTS if there is none. 1521 */ 1522 uint64_t rte_eth_find_next_owned_by(uint16_t port_id, 1523 const uint64_t owner_id); 1524 1525 /** 1526 * Macro to iterate over all enabled ethdev ports owned by a specific owner. 1527 */ 1528 #define RTE_ETH_FOREACH_DEV_OWNED_BY(p, o) \ 1529 for (p = rte_eth_find_next_owned_by(0, o); \ 1530 (unsigned int)p < (unsigned int)RTE_MAX_ETHPORTS; \ 1531 p = rte_eth_find_next_owned_by(p + 1, o)) 1532 1533 /** 1534 * Iterates over valid ethdev ports. 1535 * 1536 * @param port_id 1537 * The id of the next possible valid port. 1538 * @return 1539 * Next valid port id, RTE_MAX_ETHPORTS if there is none. 1540 */ 1541 uint16_t rte_eth_find_next(uint16_t port_id); 1542 1543 /** 1544 * Macro to iterate over all enabled and ownerless ethdev ports. 1545 */ 1546 #define RTE_ETH_FOREACH_DEV(p) \ 1547 RTE_ETH_FOREACH_DEV_OWNED_BY(p, RTE_ETH_DEV_NO_OWNER) 1548 1549 /** 1550 * @warning 1551 * @b EXPERIMENTAL: this API may change without prior notice. 1552 * 1553 * Iterates over ethdev ports of a specified device. 1554 * 1555 * @param port_id_start 1556 * The id of the next possible valid port. 1557 * @param parent 1558 * The generic device behind the ports to iterate. 1559 * @return 1560 * Next port id of the device, possibly port_id_start, 1561 * RTE_MAX_ETHPORTS if there is none. 1562 */ 1563 __rte_experimental 1564 uint16_t 1565 rte_eth_find_next_of(uint16_t port_id_start, 1566 const struct rte_device *parent); 1567 1568 /** 1569 * Macro to iterate over all ethdev ports of a specified device. 1570 * 1571 * @param port_id 1572 * The id of the matching port being iterated. 1573 * @param parent 1574 * The rte_device pointer matching the iterated ports. 1575 */ 1576 #define RTE_ETH_FOREACH_DEV_OF(port_id, parent) \ 1577 for (port_id = rte_eth_find_next_of(0, parent); \ 1578 port_id < RTE_MAX_ETHPORTS; \ 1579 port_id = rte_eth_find_next_of(port_id + 1, parent)) 1580 1581 /** 1582 * @warning 1583 * @b EXPERIMENTAL: this API may change without prior notice. 1584 * 1585 * Iterates over sibling ethdev ports (i.e. sharing the same rte_device). 1586 * 1587 * @param port_id_start 1588 * The id of the next possible valid sibling port. 1589 * @param ref_port_id 1590 * The id of a reference port to compare rte_device with. 1591 * @return 1592 * Next sibling port id, possibly port_id_start or ref_port_id itself, 1593 * RTE_MAX_ETHPORTS if there is none. 1594 */ 1595 __rte_experimental 1596 uint16_t 1597 rte_eth_find_next_sibling(uint16_t port_id_start, 1598 uint16_t ref_port_id); 1599 1600 /** 1601 * Macro to iterate over all ethdev ports sharing the same rte_device 1602 * as the specified port. 1603 * Note: the specified reference port is part of the loop iterations. 1604 * 1605 * @param port_id 1606 * The id of the matching port being iterated. 1607 * @param ref_port_id 1608 * The id of the port being compared. 1609 */ 1610 #define RTE_ETH_FOREACH_DEV_SIBLING(port_id, ref_port_id) \ 1611 for (port_id = rte_eth_find_next_sibling(0, ref_port_id); \ 1612 port_id < RTE_MAX_ETHPORTS; \ 1613 port_id = rte_eth_find_next_sibling(port_id + 1, ref_port_id)) 1614 1615 /** 1616 * @warning 1617 * @b EXPERIMENTAL: this API may change without prior notice. 1618 * 1619 * Get a new unique owner identifier. 1620 * An owner identifier is used to owns Ethernet devices by only one DPDK entity 1621 * to avoid multiple management of device by different entities. 1622 * 1623 * @param owner_id 1624 * Owner identifier pointer. 1625 * @return 1626 * Negative errno value on error, 0 on success. 1627 */ 1628 __rte_experimental 1629 int rte_eth_dev_owner_new(uint64_t *owner_id); 1630 1631 /** 1632 * @warning 1633 * @b EXPERIMENTAL: this API may change without prior notice. 1634 * 1635 * Set an Ethernet device owner. 1636 * 1637 * @param port_id 1638 * The identifier of the port to own. 1639 * @param owner 1640 * The owner pointer. 1641 * @return 1642 * Negative errno value on error, 0 on success. 1643 */ 1644 __rte_experimental 1645 int rte_eth_dev_owner_set(const uint16_t port_id, 1646 const struct rte_eth_dev_owner *owner); 1647 1648 /** 1649 * @warning 1650 * @b EXPERIMENTAL: this API may change without prior notice. 1651 * 1652 * Unset Ethernet device owner to make the device ownerless. 1653 * 1654 * @param port_id 1655 * The identifier of port to make ownerless. 1656 * @param owner_id 1657 * The owner identifier. 1658 * @return 1659 * 0 on success, negative errno value on error. 1660 */ 1661 __rte_experimental 1662 int rte_eth_dev_owner_unset(const uint16_t port_id, 1663 const uint64_t owner_id); 1664 1665 /** 1666 * @warning 1667 * @b EXPERIMENTAL: this API may change without prior notice. 1668 * 1669 * Remove owner from all Ethernet devices owned by a specific owner. 1670 * 1671 * @param owner_id 1672 * The owner identifier. 1673 * @return 1674 * 0 on success, negative errno value on error. 1675 */ 1676 __rte_experimental 1677 int rte_eth_dev_owner_delete(const uint64_t owner_id); 1678 1679 /** 1680 * @warning 1681 * @b EXPERIMENTAL: this API may change without prior notice. 1682 * 1683 * Get the owner of an Ethernet device. 1684 * 1685 * @param port_id 1686 * The port identifier. 1687 * @param owner 1688 * The owner structure pointer to fill. 1689 * @return 1690 * 0 on success, negative errno value on error.. 1691 */ 1692 __rte_experimental 1693 int rte_eth_dev_owner_get(const uint16_t port_id, 1694 struct rte_eth_dev_owner *owner); 1695 1696 /** 1697 * Get the number of ports which are usable for the application. 1698 * 1699 * These devices must be iterated by using the macro 1700 * ``RTE_ETH_FOREACH_DEV`` or ``RTE_ETH_FOREACH_DEV_OWNED_BY`` 1701 * to deal with non-contiguous ranges of devices. 1702 * 1703 * @return 1704 * The count of available Ethernet devices. 1705 */ 1706 uint16_t rte_eth_dev_count_avail(void); 1707 1708 /** 1709 * Get the total number of ports which are allocated. 1710 * 1711 * Some devices may not be available for the application. 1712 * 1713 * @return 1714 * The total count of Ethernet devices. 1715 */ 1716 uint16_t rte_eth_dev_count_total(void); 1717 1718 /** 1719 * Convert a numerical speed in Mbps to a bitmap flag that can be used in 1720 * the bitmap link_speeds of the struct rte_eth_conf 1721 * 1722 * @param speed 1723 * Numerical speed value in Mbps 1724 * @param duplex 1725 * ETH_LINK_[HALF/FULL]_DUPLEX (only for 10/100M speeds) 1726 * @return 1727 * 0 if the speed cannot be mapped 1728 */ 1729 uint32_t rte_eth_speed_bitflag(uint32_t speed, int duplex); 1730 1731 /** 1732 * Get DEV_RX_OFFLOAD_* flag name. 1733 * 1734 * @param offload 1735 * Offload flag. 1736 * @return 1737 * Offload name or 'UNKNOWN' if the flag cannot be recognised. 1738 */ 1739 const char *rte_eth_dev_rx_offload_name(uint64_t offload); 1740 1741 /** 1742 * Get DEV_TX_OFFLOAD_* flag name. 1743 * 1744 * @param offload 1745 * Offload flag. 1746 * @return 1747 * Offload name or 'UNKNOWN' if the flag cannot be recognised. 1748 */ 1749 const char *rte_eth_dev_tx_offload_name(uint64_t offload); 1750 1751 /** 1752 * Configure an Ethernet device. 1753 * This function must be invoked first before any other function in the 1754 * Ethernet API. This function can also be re-invoked when a device is in the 1755 * stopped state. 1756 * 1757 * @param port_id 1758 * The port identifier of the Ethernet device to configure. 1759 * @param nb_rx_queue 1760 * The number of receive queues to set up for the Ethernet device. 1761 * @param nb_tx_queue 1762 * The number of transmit queues to set up for the Ethernet device. 1763 * @param eth_conf 1764 * The pointer to the configuration data to be used for the Ethernet device. 1765 * The *rte_eth_conf* structure includes: 1766 * - the hardware offload features to activate, with dedicated fields for 1767 * each statically configurable offload hardware feature provided by 1768 * Ethernet devices, such as IP checksum or VLAN tag stripping for 1769 * example. 1770 * The Rx offload bitfield API is obsolete and will be deprecated. 1771 * Applications should set the ignore_bitfield_offloads bit on *rxmode* 1772 * structure and use offloads field to set per-port offloads instead. 1773 * - Any offloading set in eth_conf->[rt]xmode.offloads must be within 1774 * the [rt]x_offload_capa returned from rte_eth_dev_info_get(). 1775 * Any type of device supported offloading set in the input argument 1776 * eth_conf->[rt]xmode.offloads to rte_eth_dev_configure() is enabled 1777 * on all queues and it can't be disabled in rte_eth_[rt]x_queue_setup() 1778 * - the Receive Side Scaling (RSS) configuration when using multiple RX 1779 * queues per port. Any RSS hash function set in eth_conf->rss_conf.rss_hf 1780 * must be within the flow_type_rss_offloads provided by drivers via 1781 * rte_eth_dev_info_get() API. 1782 * 1783 * Embedding all configuration information in a single data structure 1784 * is the more flexible method that allows the addition of new features 1785 * without changing the syntax of the API. 1786 * @return 1787 * - 0: Success, device configured. 1788 * - <0: Error code returned by the driver configuration function. 1789 */ 1790 int rte_eth_dev_configure(uint16_t port_id, uint16_t nb_rx_queue, 1791 uint16_t nb_tx_queue, const struct rte_eth_conf *eth_conf); 1792 1793 /** 1794 * @warning 1795 * @b EXPERIMENTAL: this API may change without prior notice. 1796 * 1797 * Check if an Ethernet device was physically removed. 1798 * 1799 * @param port_id 1800 * The port identifier of the Ethernet device. 1801 * @return 1802 * 1 when the Ethernet device is removed, otherwise 0. 1803 */ 1804 __rte_experimental 1805 int 1806 rte_eth_dev_is_removed(uint16_t port_id); 1807 1808 /** 1809 * Allocate and set up a receive queue for an Ethernet device. 1810 * 1811 * The function allocates a contiguous block of memory for *nb_rx_desc* 1812 * receive descriptors from a memory zone associated with *socket_id* 1813 * and initializes each receive descriptor with a network buffer allocated 1814 * from the memory pool *mb_pool*. 1815 * 1816 * @param port_id 1817 * The port identifier of the Ethernet device. 1818 * @param rx_queue_id 1819 * The index of the receive queue to set up. 1820 * The value must be in the range [0, nb_rx_queue - 1] previously supplied 1821 * to rte_eth_dev_configure(). 1822 * @param nb_rx_desc 1823 * The number of receive descriptors to allocate for the receive ring. 1824 * @param socket_id 1825 * The *socket_id* argument is the socket identifier in case of NUMA. 1826 * The value can be *SOCKET_ID_ANY* if there is no NUMA constraint for 1827 * the DMA memory allocated for the receive descriptors of the ring. 1828 * @param rx_conf 1829 * The pointer to the configuration data to be used for the receive queue. 1830 * NULL value is allowed, in which case default RX configuration 1831 * will be used. 1832 * The *rx_conf* structure contains an *rx_thresh* structure with the values 1833 * of the Prefetch, Host, and Write-Back threshold registers of the receive 1834 * ring. 1835 * In addition it contains the hardware offloads features to activate using 1836 * the DEV_RX_OFFLOAD_* flags. 1837 * If an offloading set in rx_conf->offloads 1838 * hasn't been set in the input argument eth_conf->rxmode.offloads 1839 * to rte_eth_dev_configure(), it is a new added offloading, it must be 1840 * per-queue type and it is enabled for the queue. 1841 * No need to repeat any bit in rx_conf->offloads which has already been 1842 * enabled in rte_eth_dev_configure() at port level. An offloading enabled 1843 * at port level can't be disabled at queue level. 1844 * @param mb_pool 1845 * The pointer to the memory pool from which to allocate *rte_mbuf* network 1846 * memory buffers to populate each descriptor of the receive ring. 1847 * @return 1848 * - 0: Success, receive queue correctly set up. 1849 * - -EIO: if device is removed. 1850 * - -EINVAL: The memory pool pointer is null or the size of network buffers 1851 * which can be allocated from this memory pool does not fit the various 1852 * buffer sizes allowed by the device controller. 1853 * - -ENOMEM: Unable to allocate the receive ring descriptors or to 1854 * allocate network memory buffers from the memory pool when 1855 * initializing receive descriptors. 1856 */ 1857 int rte_eth_rx_queue_setup(uint16_t port_id, uint16_t rx_queue_id, 1858 uint16_t nb_rx_desc, unsigned int socket_id, 1859 const struct rte_eth_rxconf *rx_conf, 1860 struct rte_mempool *mb_pool); 1861 1862 /** 1863 * @warning 1864 * @b EXPERIMENTAL: this API may change, or be removed, without prior notice 1865 * 1866 * Allocate and set up a hairpin receive queue for an Ethernet device. 1867 * 1868 * The function set up the selected queue to be used in hairpin. 1869 * 1870 * @param port_id 1871 * The port identifier of the Ethernet device. 1872 * @param rx_queue_id 1873 * The index of the receive queue to set up. 1874 * The value must be in the range [0, nb_rx_queue - 1] previously supplied 1875 * to rte_eth_dev_configure(). 1876 * @param nb_rx_desc 1877 * The number of receive descriptors to allocate for the receive ring. 1878 * 0 means the PMD will use default value. 1879 * @param conf 1880 * The pointer to the hairpin configuration. 1881 * 1882 * @return 1883 * - (0) if successful. 1884 * - (-ENOTSUP) if hardware doesn't support. 1885 * - (-EINVAL) if bad parameter. 1886 * - (-ENOMEM) if unable to allocate the resources. 1887 */ 1888 __rte_experimental 1889 int rte_eth_rx_hairpin_queue_setup 1890 (uint16_t port_id, uint16_t rx_queue_id, uint16_t nb_rx_desc, 1891 const struct rte_eth_hairpin_conf *conf); 1892 1893 /** 1894 * Allocate and set up a transmit queue for an Ethernet device. 1895 * 1896 * @param port_id 1897 * The port identifier of the Ethernet device. 1898 * @param tx_queue_id 1899 * The index of the transmit queue to set up. 1900 * The value must be in the range [0, nb_tx_queue - 1] previously supplied 1901 * to rte_eth_dev_configure(). 1902 * @param nb_tx_desc 1903 * The number of transmit descriptors to allocate for the transmit ring. 1904 * @param socket_id 1905 * The *socket_id* argument is the socket identifier in case of NUMA. 1906 * Its value can be *SOCKET_ID_ANY* if there is no NUMA constraint for 1907 * the DMA memory allocated for the transmit descriptors of the ring. 1908 * @param tx_conf 1909 * The pointer to the configuration data to be used for the transmit queue. 1910 * NULL value is allowed, in which case default TX configuration 1911 * will be used. 1912 * The *tx_conf* structure contains the following data: 1913 * - The *tx_thresh* structure with the values of the Prefetch, Host, and 1914 * Write-Back threshold registers of the transmit ring. 1915 * When setting Write-Back threshold to the value greater then zero, 1916 * *tx_rs_thresh* value should be explicitly set to one. 1917 * - The *tx_free_thresh* value indicates the [minimum] number of network 1918 * buffers that must be pending in the transmit ring to trigger their 1919 * [implicit] freeing by the driver transmit function. 1920 * - The *tx_rs_thresh* value indicates the [minimum] number of transmit 1921 * descriptors that must be pending in the transmit ring before setting the 1922 * RS bit on a descriptor by the driver transmit function. 1923 * The *tx_rs_thresh* value should be less or equal then 1924 * *tx_free_thresh* value, and both of them should be less then 1925 * *nb_tx_desc* - 3. 1926 * - The *offloads* member contains Tx offloads to be enabled. 1927 * If an offloading set in tx_conf->offloads 1928 * hasn't been set in the input argument eth_conf->txmode.offloads 1929 * to rte_eth_dev_configure(), it is a new added offloading, it must be 1930 * per-queue type and it is enabled for the queue. 1931 * No need to repeat any bit in tx_conf->offloads which has already been 1932 * enabled in rte_eth_dev_configure() at port level. An offloading enabled 1933 * at port level can't be disabled at queue level. 1934 * 1935 * Note that setting *tx_free_thresh* or *tx_rs_thresh* value to 0 forces 1936 * the transmit function to use default values. 1937 * @return 1938 * - 0: Success, the transmit queue is correctly set up. 1939 * - -ENOMEM: Unable to allocate the transmit ring descriptors. 1940 */ 1941 int rte_eth_tx_queue_setup(uint16_t port_id, uint16_t tx_queue_id, 1942 uint16_t nb_tx_desc, unsigned int socket_id, 1943 const struct rte_eth_txconf *tx_conf); 1944 1945 /** 1946 * @warning 1947 * @b EXPERIMENTAL: this API may change, or be removed, without prior notice 1948 * 1949 * Allocate and set up a transmit hairpin queue for an Ethernet device. 1950 * 1951 * @param port_id 1952 * The port identifier of the Ethernet device. 1953 * @param tx_queue_id 1954 * The index of the transmit queue to set up. 1955 * The value must be in the range [0, nb_tx_queue - 1] previously supplied 1956 * to rte_eth_dev_configure(). 1957 * @param nb_tx_desc 1958 * The number of transmit descriptors to allocate for the transmit ring. 1959 * 0 to set default PMD value. 1960 * @param conf 1961 * The hairpin configuration. 1962 * 1963 * @return 1964 * - (0) if successful. 1965 * - (-ENOTSUP) if hardware doesn't support. 1966 * - (-EINVAL) if bad parameter. 1967 * - (-ENOMEM) if unable to allocate the resources. 1968 */ 1969 __rte_experimental 1970 int rte_eth_tx_hairpin_queue_setup 1971 (uint16_t port_id, uint16_t tx_queue_id, uint16_t nb_tx_desc, 1972 const struct rte_eth_hairpin_conf *conf); 1973 1974 /** 1975 * Return the NUMA socket to which an Ethernet device is connected 1976 * 1977 * @param port_id 1978 * The port identifier of the Ethernet device 1979 * @return 1980 * The NUMA socket id to which the Ethernet device is connected or 1981 * a default of zero if the socket could not be determined. 1982 * -1 is returned is the port_id value is out of range. 1983 */ 1984 int rte_eth_dev_socket_id(uint16_t port_id); 1985 1986 /** 1987 * Check if port_id of device is attached 1988 * 1989 * @param port_id 1990 * The port identifier of the Ethernet device 1991 * @return 1992 * - 0 if port is out of range or not attached 1993 * - 1 if device is attached 1994 */ 1995 int rte_eth_dev_is_valid_port(uint16_t port_id); 1996 1997 /** 1998 * Start specified RX queue of a port. It is used when rx_deferred_start 1999 * flag of the specified queue is true. 2000 * 2001 * @param port_id 2002 * The port identifier of the Ethernet device 2003 * @param rx_queue_id 2004 * The index of the rx queue to update the ring. 2005 * The value must be in the range [0, nb_rx_queue - 1] previously supplied 2006 * to rte_eth_dev_configure(). 2007 * @return 2008 * - 0: Success, the receive queue is started. 2009 * - -EINVAL: The port_id or the queue_id out of range or belong to hairpin. 2010 * - -EIO: if device is removed. 2011 * - -ENOTSUP: The function not supported in PMD driver. 2012 */ 2013 int rte_eth_dev_rx_queue_start(uint16_t port_id, uint16_t rx_queue_id); 2014 2015 /** 2016 * Stop specified RX queue of a port 2017 * 2018 * @param port_id 2019 * The port identifier of the Ethernet device 2020 * @param rx_queue_id 2021 * The index of the rx queue to update the ring. 2022 * The value must be in the range [0, nb_rx_queue - 1] previously supplied 2023 * to rte_eth_dev_configure(). 2024 * @return 2025 * - 0: Success, the receive queue is stopped. 2026 * - -EINVAL: The port_id or the queue_id out of range or belong to hairpin. 2027 * - -EIO: if device is removed. 2028 * - -ENOTSUP: The function not supported in PMD driver. 2029 */ 2030 int rte_eth_dev_rx_queue_stop(uint16_t port_id, uint16_t rx_queue_id); 2031 2032 /** 2033 * Start TX for specified queue of a port. It is used when tx_deferred_start 2034 * flag of the specified queue is true. 2035 * 2036 * @param port_id 2037 * The port identifier of the Ethernet device 2038 * @param tx_queue_id 2039 * The index of the tx queue to update the ring. 2040 * The value must be in the range [0, nb_tx_queue - 1] previously supplied 2041 * to rte_eth_dev_configure(). 2042 * @return 2043 * - 0: Success, the transmit queue is started. 2044 * - -EINVAL: The port_id or the queue_id out of range or belong to hairpin. 2045 * - -EIO: if device is removed. 2046 * - -ENOTSUP: The function not supported in PMD driver. 2047 */ 2048 int rte_eth_dev_tx_queue_start(uint16_t port_id, uint16_t tx_queue_id); 2049 2050 /** 2051 * Stop specified TX queue of a port 2052 * 2053 * @param port_id 2054 * The port identifier of the Ethernet device 2055 * @param tx_queue_id 2056 * The index of the tx queue to update the ring. 2057 * The value must be in the range [0, nb_tx_queue - 1] previously supplied 2058 * to rte_eth_dev_configure(). 2059 * @return 2060 * - 0: Success, the transmit queue is stopped. 2061 * - -EINVAL: The port_id or the queue_id out of range or belong to hairpin. 2062 * - -EIO: if device is removed. 2063 * - -ENOTSUP: The function not supported in PMD driver. 2064 */ 2065 int rte_eth_dev_tx_queue_stop(uint16_t port_id, uint16_t tx_queue_id); 2066 2067 /** 2068 * Start an Ethernet device. 2069 * 2070 * The device start step is the last one and consists of setting the configured 2071 * offload features and in starting the transmit and the receive units of the 2072 * device. 2073 * 2074 * Device RTE_ETH_DEV_NOLIVE_MAC_ADDR flag causes MAC address to be set before 2075 * PMD port start callback function is invoked. 2076 * 2077 * On success, all basic functions exported by the Ethernet API (link status, 2078 * receive/transmit, and so on) can be invoked. 2079 * 2080 * @param port_id 2081 * The port identifier of the Ethernet device. 2082 * @return 2083 * - 0: Success, Ethernet device started. 2084 * - <0: Error code of the driver device start function. 2085 */ 2086 int rte_eth_dev_start(uint16_t port_id); 2087 2088 /** 2089 * Stop an Ethernet device. The device can be restarted with a call to 2090 * rte_eth_dev_start() 2091 * 2092 * @param port_id 2093 * The port identifier of the Ethernet device. 2094 */ 2095 void rte_eth_dev_stop(uint16_t port_id); 2096 2097 /** 2098 * Link up an Ethernet device. 2099 * 2100 * Set device link up will re-enable the device rx/tx 2101 * functionality after it is previously set device linked down. 2102 * 2103 * @param port_id 2104 * The port identifier of the Ethernet device. 2105 * @return 2106 * - 0: Success, Ethernet device linked up. 2107 * - <0: Error code of the driver device link up function. 2108 */ 2109 int rte_eth_dev_set_link_up(uint16_t port_id); 2110 2111 /** 2112 * Link down an Ethernet device. 2113 * The device rx/tx functionality will be disabled if success, 2114 * and it can be re-enabled with a call to 2115 * rte_eth_dev_set_link_up() 2116 * 2117 * @param port_id 2118 * The port identifier of the Ethernet device. 2119 */ 2120 int rte_eth_dev_set_link_down(uint16_t port_id); 2121 2122 /** 2123 * Close a stopped Ethernet device. The device cannot be restarted! 2124 * The function frees all port resources if the driver supports 2125 * the flag RTE_ETH_DEV_CLOSE_REMOVE. 2126 * 2127 * @param port_id 2128 * The port identifier of the Ethernet device. 2129 */ 2130 void rte_eth_dev_close(uint16_t port_id); 2131 2132 /** 2133 * Reset a Ethernet device and keep its port id. 2134 * 2135 * When a port has to be reset passively, the DPDK application can invoke 2136 * this function. For example when a PF is reset, all its VFs should also 2137 * be reset. Normally a DPDK application can invoke this function when 2138 * RTE_ETH_EVENT_INTR_RESET event is detected, but can also use it to start 2139 * a port reset in other circumstances. 2140 * 2141 * When this function is called, it first stops the port and then calls the 2142 * PMD specific dev_uninit( ) and dev_init( ) to return the port to initial 2143 * state, in which no Tx and Rx queues are setup, as if the port has been 2144 * reset and not started. The port keeps the port id it had before the 2145 * function call. 2146 * 2147 * After calling rte_eth_dev_reset( ), the application should use 2148 * rte_eth_dev_configure( ), rte_eth_rx_queue_setup( ), 2149 * rte_eth_tx_queue_setup( ), and rte_eth_dev_start( ) 2150 * to reconfigure the device as appropriate. 2151 * 2152 * Note: To avoid unexpected behavior, the application should stop calling 2153 * Tx and Rx functions before calling rte_eth_dev_reset( ). For thread 2154 * safety, all these controlling functions should be called from the same 2155 * thread. 2156 * 2157 * @param port_id 2158 * The port identifier of the Ethernet device. 2159 * 2160 * @return 2161 * - (0) if successful. 2162 * - (-EINVAL) if port identifier is invalid. 2163 * - (-ENOTSUP) if hardware doesn't support this function. 2164 * - (-EPERM) if not ran from the primary process. 2165 * - (-EIO) if re-initialisation failed or device is removed. 2166 * - (-ENOMEM) if the reset failed due to OOM. 2167 * - (-EAGAIN) if the reset temporarily failed and should be retried later. 2168 */ 2169 int rte_eth_dev_reset(uint16_t port_id); 2170 2171 /** 2172 * Enable receipt in promiscuous mode for an Ethernet device. 2173 * 2174 * @param port_id 2175 * The port identifier of the Ethernet device. 2176 * @return 2177 * - (0) if successful. 2178 * - (-ENOTSUP) if support for promiscuous_enable() does not exist 2179 * for the device. 2180 * - (-ENODEV) if *port_id* invalid. 2181 */ 2182 int rte_eth_promiscuous_enable(uint16_t port_id); 2183 2184 /** 2185 * Disable receipt in promiscuous mode for an Ethernet device. 2186 * 2187 * @param port_id 2188 * The port identifier of the Ethernet device. 2189 * @return 2190 * - (0) if successful. 2191 * - (-ENOTSUP) if support for promiscuous_disable() does not exist 2192 * for the device. 2193 * - (-ENODEV) if *port_id* invalid. 2194 */ 2195 int rte_eth_promiscuous_disable(uint16_t port_id); 2196 2197 /** 2198 * Return the value of promiscuous mode for an Ethernet device. 2199 * 2200 * @param port_id 2201 * The port identifier of the Ethernet device. 2202 * @return 2203 * - (1) if promiscuous is enabled 2204 * - (0) if promiscuous is disabled. 2205 * - (-1) on error 2206 */ 2207 int rte_eth_promiscuous_get(uint16_t port_id); 2208 2209 /** 2210 * Enable the receipt of any multicast frame by an Ethernet device. 2211 * 2212 * @param port_id 2213 * The port identifier of the Ethernet device. 2214 * @return 2215 * - (0) if successful. 2216 * - (-ENOTSUP) if support for allmulticast_enable() does not exist 2217 * for the device. 2218 * - (-ENODEV) if *port_id* invalid. 2219 */ 2220 int rte_eth_allmulticast_enable(uint16_t port_id); 2221 2222 /** 2223 * Disable the receipt of all multicast frames by an Ethernet device. 2224 * 2225 * @param port_id 2226 * The port identifier of the Ethernet device. 2227 * @return 2228 * - (0) if successful. 2229 * - (-ENOTSUP) if support for allmulticast_disable() does not exist 2230 * for the device. 2231 * - (-ENODEV) if *port_id* invalid. 2232 */ 2233 int rte_eth_allmulticast_disable(uint16_t port_id); 2234 2235 /** 2236 * Return the value of allmulticast mode for an Ethernet device. 2237 * 2238 * @param port_id 2239 * The port identifier of the Ethernet device. 2240 * @return 2241 * - (1) if allmulticast is enabled 2242 * - (0) if allmulticast is disabled. 2243 * - (-1) on error 2244 */ 2245 int rte_eth_allmulticast_get(uint16_t port_id); 2246 2247 /** 2248 * Retrieve the status (ON/OFF), the speed (in Mbps) and the mode (HALF-DUPLEX 2249 * or FULL-DUPLEX) of the physical link of an Ethernet device. It might need 2250 * to wait up to 9 seconds in it. 2251 * 2252 * @param port_id 2253 * The port identifier of the Ethernet device. 2254 * @param link 2255 * A pointer to an *rte_eth_link* structure to be filled with 2256 * the status, the speed and the mode of the Ethernet device link. 2257 * @return 2258 * - (0) if successful. 2259 * - (-ENOTSUP) if the function is not supported in PMD driver. 2260 * - (-ENODEV) if *port_id* invalid. 2261 */ 2262 int rte_eth_link_get(uint16_t port_id, struct rte_eth_link *link); 2263 2264 /** 2265 * Retrieve the status (ON/OFF), the speed (in Mbps) and the mode (HALF-DUPLEX 2266 * or FULL-DUPLEX) of the physical link of an Ethernet device. It is a no-wait 2267 * version of rte_eth_link_get(). 2268 * 2269 * @param port_id 2270 * The port identifier of the Ethernet device. 2271 * @param link 2272 * A pointer to an *rte_eth_link* structure to be filled with 2273 * the status, the speed and the mode of the Ethernet device link. 2274 * @return 2275 * - (0) if successful. 2276 * - (-ENOTSUP) if the function is not supported in PMD driver. 2277 * - (-ENODEV) if *port_id* invalid. 2278 */ 2279 int rte_eth_link_get_nowait(uint16_t port_id, struct rte_eth_link *link); 2280 2281 /** 2282 * Retrieve the general I/O statistics of an Ethernet device. 2283 * 2284 * @param port_id 2285 * The port identifier of the Ethernet device. 2286 * @param stats 2287 * A pointer to a structure of type *rte_eth_stats* to be filled with 2288 * the values of device counters for the following set of statistics: 2289 * - *ipackets* with the total of successfully received packets. 2290 * - *opackets* with the total of successfully transmitted packets. 2291 * - *ibytes* with the total of successfully received bytes. 2292 * - *obytes* with the total of successfully transmitted bytes. 2293 * - *ierrors* with the total of erroneous received packets. 2294 * - *oerrors* with the total of failed transmitted packets. 2295 * @return 2296 * Zero if successful. Non-zero otherwise. 2297 */ 2298 int rte_eth_stats_get(uint16_t port_id, struct rte_eth_stats *stats); 2299 2300 /** 2301 * Reset the general I/O statistics of an Ethernet device. 2302 * 2303 * @param port_id 2304 * The port identifier of the Ethernet device. 2305 * @return 2306 * - (0) if device notified to reset stats. 2307 * - (-ENOTSUP) if hardware doesn't support. 2308 * - (-ENODEV) if *port_id* invalid. 2309 * - (<0): Error code of the driver stats reset function. 2310 */ 2311 int rte_eth_stats_reset(uint16_t port_id); 2312 2313 /** 2314 * Retrieve names of extended statistics of an Ethernet device. 2315 * 2316 * There is an assumption that 'xstat_names' and 'xstats' arrays are matched 2317 * by array index: 2318 * xstats_names[i].name => xstats[i].value 2319 * 2320 * And the array index is same with id field of 'struct rte_eth_xstat': 2321 * xstats[i].id == i 2322 * 2323 * This assumption makes key-value pair matching less flexible but simpler. 2324 * 2325 * @param port_id 2326 * The port identifier of the Ethernet device. 2327 * @param xstats_names 2328 * An rte_eth_xstat_name array of at least *size* elements to 2329 * be filled. If set to NULL, the function returns the required number 2330 * of elements. 2331 * @param size 2332 * The size of the xstats_names array (number of elements). 2333 * @return 2334 * - A positive value lower or equal to size: success. The return value 2335 * is the number of entries filled in the stats table. 2336 * - A positive value higher than size: error, the given statistics table 2337 * is too small. The return value corresponds to the size that should 2338 * be given to succeed. The entries in the table are not valid and 2339 * shall not be used by the caller. 2340 * - A negative value on error (invalid port id). 2341 */ 2342 int rte_eth_xstats_get_names(uint16_t port_id, 2343 struct rte_eth_xstat_name *xstats_names, 2344 unsigned int size); 2345 2346 /** 2347 * Retrieve extended statistics of an Ethernet device. 2348 * 2349 * There is an assumption that 'xstat_names' and 'xstats' arrays are matched 2350 * by array index: 2351 * xstats_names[i].name => xstats[i].value 2352 * 2353 * And the array index is same with id field of 'struct rte_eth_xstat': 2354 * xstats[i].id == i 2355 * 2356 * This assumption makes key-value pair matching less flexible but simpler. 2357 * 2358 * @param port_id 2359 * The port identifier of the Ethernet device. 2360 * @param xstats 2361 * A pointer to a table of structure of type *rte_eth_xstat* 2362 * to be filled with device statistics ids and values. 2363 * This parameter can be set to NULL if n is 0. 2364 * @param n 2365 * The size of the xstats array (number of elements). 2366 * @return 2367 * - A positive value lower or equal to n: success. The return value 2368 * is the number of entries filled in the stats table. 2369 * - A positive value higher than n: error, the given statistics table 2370 * is too small. The return value corresponds to the size that should 2371 * be given to succeed. The entries in the table are not valid and 2372 * shall not be used by the caller. 2373 * - A negative value on error (invalid port id). 2374 */ 2375 int rte_eth_xstats_get(uint16_t port_id, struct rte_eth_xstat *xstats, 2376 unsigned int n); 2377 2378 /** 2379 * Retrieve names of extended statistics of an Ethernet device. 2380 * 2381 * @param port_id 2382 * The port identifier of the Ethernet device. 2383 * @param xstats_names 2384 * An rte_eth_xstat_name array of at least *size* elements to 2385 * be filled. If set to NULL, the function returns the required number 2386 * of elements. 2387 * @param ids 2388 * IDs array given by app to retrieve specific statistics 2389 * @param size 2390 * The size of the xstats_names array (number of elements). 2391 * @return 2392 * - A positive value lower or equal to size: success. The return value 2393 * is the number of entries filled in the stats table. 2394 * - A positive value higher than size: error, the given statistics table 2395 * is too small. The return value corresponds to the size that should 2396 * be given to succeed. The entries in the table are not valid and 2397 * shall not be used by the caller. 2398 * - A negative value on error (invalid port id). 2399 */ 2400 int 2401 rte_eth_xstats_get_names_by_id(uint16_t port_id, 2402 struct rte_eth_xstat_name *xstats_names, unsigned int size, 2403 uint64_t *ids); 2404 2405 /** 2406 * Retrieve extended statistics of an Ethernet device. 2407 * 2408 * @param port_id 2409 * The port identifier of the Ethernet device. 2410 * @param ids 2411 * A pointer to an ids array passed by application. This tells which 2412 * statistics values function should retrieve. This parameter 2413 * can be set to NULL if size is 0. In this case function will retrieve 2414 * all available statistics. 2415 * @param values 2416 * A pointer to a table to be filled with device statistics values. 2417 * @param size 2418 * The size of the ids array (number of elements). 2419 * @return 2420 * - A positive value lower or equal to size: success. The return value 2421 * is the number of entries filled in the stats table. 2422 * - A positive value higher than size: error, the given statistics table 2423 * is too small. The return value corresponds to the size that should 2424 * be given to succeed. The entries in the table are not valid and 2425 * shall not be used by the caller. 2426 * - A negative value on error (invalid port id). 2427 */ 2428 int rte_eth_xstats_get_by_id(uint16_t port_id, const uint64_t *ids, 2429 uint64_t *values, unsigned int size); 2430 2431 /** 2432 * Gets the ID of a statistic from its name. 2433 * 2434 * This function searches for the statistics using string compares, and 2435 * as such should not be used on the fast-path. For fast-path retrieval of 2436 * specific statistics, store the ID as provided in *id* from this function, 2437 * and pass the ID to rte_eth_xstats_get() 2438 * 2439 * @param port_id The port to look up statistics from 2440 * @param xstat_name The name of the statistic to return 2441 * @param[out] id A pointer to an app-supplied uint64_t which should be 2442 * set to the ID of the stat if the stat exists. 2443 * @return 2444 * 0 on success 2445 * -ENODEV for invalid port_id, 2446 * -EIO if device is removed, 2447 * -EINVAL if the xstat_name doesn't exist in port_id 2448 */ 2449 int rte_eth_xstats_get_id_by_name(uint16_t port_id, const char *xstat_name, 2450 uint64_t *id); 2451 2452 /** 2453 * Reset extended statistics of an Ethernet device. 2454 * 2455 * @param port_id 2456 * The port identifier of the Ethernet device. 2457 * @return 2458 * - (0) if device notified to reset extended stats. 2459 * - (-ENOTSUP) if pmd doesn't support both 2460 * extended stats and basic stats reset. 2461 * - (-ENODEV) if *port_id* invalid. 2462 * - (<0): Error code of the driver xstats reset function. 2463 */ 2464 int rte_eth_xstats_reset(uint16_t port_id); 2465 2466 /** 2467 * Set a mapping for the specified transmit queue to the specified per-queue 2468 * statistics counter. 2469 * 2470 * @param port_id 2471 * The port identifier of the Ethernet device. 2472 * @param tx_queue_id 2473 * The index of the transmit queue for which a queue stats mapping is required. 2474 * The value must be in the range [0, nb_tx_queue - 1] previously supplied 2475 * to rte_eth_dev_configure(). 2476 * @param stat_idx 2477 * The per-queue packet statistics functionality number that the transmit 2478 * queue is to be assigned. 2479 * The value must be in the range [0, RTE_ETHDEV_QUEUE_STAT_CNTRS - 1]. 2480 * @return 2481 * Zero if successful. Non-zero otherwise. 2482 */ 2483 int rte_eth_dev_set_tx_queue_stats_mapping(uint16_t port_id, 2484 uint16_t tx_queue_id, uint8_t stat_idx); 2485 2486 /** 2487 * Set a mapping for the specified receive queue to the specified per-queue 2488 * statistics counter. 2489 * 2490 * @param port_id 2491 * The port identifier of the Ethernet device. 2492 * @param rx_queue_id 2493 * The index of the receive queue for which a queue stats mapping is required. 2494 * The value must be in the range [0, nb_rx_queue - 1] previously supplied 2495 * to rte_eth_dev_configure(). 2496 * @param stat_idx 2497 * The per-queue packet statistics functionality number that the receive 2498 * queue is to be assigned. 2499 * The value must be in the range [0, RTE_ETHDEV_QUEUE_STAT_CNTRS - 1]. 2500 * @return 2501 * Zero if successful. Non-zero otherwise. 2502 */ 2503 int rte_eth_dev_set_rx_queue_stats_mapping(uint16_t port_id, 2504 uint16_t rx_queue_id, 2505 uint8_t stat_idx); 2506 2507 /** 2508 * Retrieve the Ethernet address of an Ethernet device. 2509 * 2510 * @param port_id 2511 * The port identifier of the Ethernet device. 2512 * @param mac_addr 2513 * A pointer to a structure of type *ether_addr* to be filled with 2514 * the Ethernet address of the Ethernet device. 2515 * @return 2516 * - (0) if successful 2517 * - (-ENODEV) if *port_id* invalid. 2518 */ 2519 int rte_eth_macaddr_get(uint16_t port_id, struct rte_ether_addr *mac_addr); 2520 2521 /** 2522 * Retrieve the contextual information of an Ethernet device. 2523 * 2524 * As part of this function, a number of of fields in dev_info will be 2525 * initialized as follows: 2526 * 2527 * rx_desc_lim = lim 2528 * tx_desc_lim = lim 2529 * 2530 * Where lim is defined within the rte_eth_dev_info_get as 2531 * 2532 * const struct rte_eth_desc_lim lim = { 2533 * .nb_max = UINT16_MAX, 2534 * .nb_min = 0, 2535 * .nb_align = 1, 2536 * .nb_seg_max = UINT16_MAX, 2537 * .nb_mtu_seg_max = UINT16_MAX, 2538 * }; 2539 * 2540 * device = dev->device 2541 * min_mtu = RTE_ETHER_MIN_MTU 2542 * max_mtu = UINT16_MAX 2543 * 2544 * The following fields will be populated if support for dev_infos_get() 2545 * exists for the device and the rte_eth_dev 'dev' has been populated 2546 * successfully with a call to it: 2547 * 2548 * driver_name = dev->device->driver->name 2549 * nb_rx_queues = dev->data->nb_rx_queues 2550 * nb_tx_queues = dev->data->nb_tx_queues 2551 * dev_flags = &dev->data->dev_flags 2552 * 2553 * @param port_id 2554 * The port identifier of the Ethernet device. 2555 * @param dev_info 2556 * A pointer to a structure of type *rte_eth_dev_info* to be filled with 2557 * the contextual information of the Ethernet device. 2558 * @return 2559 * - (0) if successful. 2560 * - (-ENOTSUP) if support for dev_infos_get() does not exist for the device. 2561 * - (-ENODEV) if *port_id* invalid. 2562 */ 2563 int rte_eth_dev_info_get(uint16_t port_id, struct rte_eth_dev_info *dev_info); 2564 2565 /** 2566 * Retrieve the firmware version of a device. 2567 * 2568 * @param port_id 2569 * The port identifier of the device. 2570 * @param fw_version 2571 * A pointer to a string array storing the firmware version of a device, 2572 * the string includes terminating null. This pointer is allocated by caller. 2573 * @param fw_size 2574 * The size of the string array pointed by fw_version, which should be 2575 * large enough to store firmware version of the device. 2576 * @return 2577 * - (0) if successful. 2578 * - (-ENOTSUP) if operation is not supported. 2579 * - (-ENODEV) if *port_id* invalid. 2580 * - (-EIO) if device is removed. 2581 * - (>0) if *fw_size* is not enough to store firmware version, return 2582 * the size of the non truncated string. 2583 */ 2584 int rte_eth_dev_fw_version_get(uint16_t port_id, 2585 char *fw_version, size_t fw_size); 2586 2587 /** 2588 * Retrieve the supported packet types of an Ethernet device. 2589 * 2590 * When a packet type is announced as supported, it *must* be recognized by 2591 * the PMD. For instance, if RTE_PTYPE_L2_ETHER, RTE_PTYPE_L2_ETHER_VLAN 2592 * and RTE_PTYPE_L3_IPV4 are announced, the PMD must return the following 2593 * packet types for these packets: 2594 * - Ether/IPv4 -> RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV4 2595 * - Ether/Vlan/IPv4 -> RTE_PTYPE_L2_ETHER_VLAN | RTE_PTYPE_L3_IPV4 2596 * - Ether/[anything else] -> RTE_PTYPE_L2_ETHER 2597 * - Ether/Vlan/[anything else] -> RTE_PTYPE_L2_ETHER_VLAN 2598 * 2599 * When a packet is received by a PMD, the most precise type must be 2600 * returned among the ones supported. However a PMD is allowed to set 2601 * packet type that is not in the supported list, at the condition that it 2602 * is more precise. Therefore, a PMD announcing no supported packet types 2603 * can still set a matching packet type in a received packet. 2604 * 2605 * @note 2606 * Better to invoke this API after the device is already started or rx burst 2607 * function is decided, to obtain correct supported ptypes. 2608 * @note 2609 * if a given PMD does not report what ptypes it supports, then the supported 2610 * ptype count is reported as 0. 2611 * @param port_id 2612 * The port identifier of the Ethernet device. 2613 * @param ptype_mask 2614 * A hint of what kind of packet type which the caller is interested in. 2615 * @param ptypes 2616 * An array pointer to store adequate packet types, allocated by caller. 2617 * @param num 2618 * Size of the array pointed by param ptypes. 2619 * @return 2620 * - (>=0) Number of supported ptypes. If the number of types exceeds num, 2621 * only num entries will be filled into the ptypes array, but the full 2622 * count of supported ptypes will be returned. 2623 * - (-ENODEV) if *port_id* invalid. 2624 */ 2625 int rte_eth_dev_get_supported_ptypes(uint16_t port_id, uint32_t ptype_mask, 2626 uint32_t *ptypes, int num); 2627 /** 2628 * @warning 2629 * @b EXPERIMENTAL: this API may change without prior notice. 2630 * 2631 * Inform Ethernet device about reduced range of packet types to handle. 2632 * 2633 * Application can use this function to set only specific ptypes that it's 2634 * interested. This information can be used by the PMD to optimize Rx path. 2635 * 2636 * The function accepts an array `set_ptypes` allocated by the caller to 2637 * store the packet types set by the driver, the last element of the array 2638 * is set to RTE_PTYPE_UNKNOWN. The size of the `set_ptype` array should be 2639 * `rte_eth_dev_get_supported_ptypes() + 1` else it might only be filled 2640 * partially. 2641 * 2642 * @param port_id 2643 * The port identifier of the Ethernet device. 2644 * @param ptype_mask 2645 * The ptype family that application is interested in should be bitwise OR of 2646 * RTE_PTYPE_*_MASK or 0. 2647 * @param set_ptypes 2648 * An array pointer to store set packet types, allocated by caller. The 2649 * function marks the end of array with RTE_PTYPE_UNKNOWN. 2650 * @param num 2651 * Size of the array pointed by param ptypes. 2652 * Should be rte_eth_dev_get_supported_ptypes() + 1 to accommodate the 2653 * set ptypes. 2654 * @return 2655 * - (0) if Success. 2656 * - (-ENODEV) if *port_id* invalid. 2657 * - (-EINVAL) if *ptype_mask* is invalid (or) set_ptypes is NULL and 2658 * num > 0. 2659 */ 2660 __rte_experimental 2661 int rte_eth_dev_set_ptypes(uint16_t port_id, uint32_t ptype_mask, 2662 uint32_t *set_ptypes, unsigned int num); 2663 2664 /** 2665 * Retrieve the MTU of an Ethernet device. 2666 * 2667 * @param port_id 2668 * The port identifier of the Ethernet device. 2669 * @param mtu 2670 * A pointer to a uint16_t where the retrieved MTU is to be stored. 2671 * @return 2672 * - (0) if successful. 2673 * - (-ENODEV) if *port_id* invalid. 2674 */ 2675 int rte_eth_dev_get_mtu(uint16_t port_id, uint16_t *mtu); 2676 2677 /** 2678 * Change the MTU of an Ethernet device. 2679 * 2680 * @param port_id 2681 * The port identifier of the Ethernet device. 2682 * @param mtu 2683 * A uint16_t for the MTU to be applied. 2684 * @return 2685 * - (0) if successful. 2686 * - (-ENOTSUP) if operation is not supported. 2687 * - (-ENODEV) if *port_id* invalid. 2688 * - (-EIO) if device is removed. 2689 * - (-EINVAL) if *mtu* invalid, validation of mtu can occur within 2690 * rte_eth_dev_set_mtu if dev_infos_get is supported by the device or 2691 * when the mtu is set using dev->dev_ops->mtu_set. 2692 * - (-EBUSY) if operation is not allowed when the port is running 2693 */ 2694 int rte_eth_dev_set_mtu(uint16_t port_id, uint16_t mtu); 2695 2696 /** 2697 * Enable/Disable hardware filtering by an Ethernet device of received 2698 * VLAN packets tagged with a given VLAN Tag Identifier. 2699 * 2700 * @param port_id 2701 * The port identifier of the Ethernet device. 2702 * @param vlan_id 2703 * The VLAN Tag Identifier whose filtering must be enabled or disabled. 2704 * @param on 2705 * If > 0, enable VLAN filtering of VLAN packets tagged with *vlan_id*. 2706 * Otherwise, disable VLAN filtering of VLAN packets tagged with *vlan_id*. 2707 * @return 2708 * - (0) if successful. 2709 * - (-ENOTSUP) if hardware-assisted VLAN filtering not configured. 2710 * - (-ENODEV) if *port_id* invalid. 2711 * - (-EIO) if device is removed. 2712 * - (-ENOSYS) if VLAN filtering on *port_id* disabled. 2713 * - (-EINVAL) if *vlan_id* > 4095. 2714 */ 2715 int rte_eth_dev_vlan_filter(uint16_t port_id, uint16_t vlan_id, int on); 2716 2717 /** 2718 * Enable/Disable hardware VLAN Strip by a rx queue of an Ethernet device. 2719 * 82599/X540/X550 can support VLAN stripping at the rx queue level 2720 * 2721 * @param port_id 2722 * The port identifier of the Ethernet device. 2723 * @param rx_queue_id 2724 * The index of the receive queue for which a queue stats mapping is required. 2725 * The value must be in the range [0, nb_rx_queue - 1] previously supplied 2726 * to rte_eth_dev_configure(). 2727 * @param on 2728 * If 1, Enable VLAN Stripping of the receive queue of the Ethernet port. 2729 * If 0, Disable VLAN Stripping of the receive queue of the Ethernet port. 2730 * @return 2731 * - (0) if successful. 2732 * - (-ENOTSUP) if hardware-assisted VLAN stripping not configured. 2733 * - (-ENODEV) if *port_id* invalid. 2734 * - (-EINVAL) if *rx_queue_id* invalid. 2735 */ 2736 int rte_eth_dev_set_vlan_strip_on_queue(uint16_t port_id, uint16_t rx_queue_id, 2737 int on); 2738 2739 /** 2740 * Set the Outer VLAN Ether Type by an Ethernet device, it can be inserted to 2741 * the VLAN Header. This is a register setup available on some Intel NIC, not 2742 * but all, please check the data sheet for availability. 2743 * 2744 * @param port_id 2745 * The port identifier of the Ethernet device. 2746 * @param vlan_type 2747 * The vlan type. 2748 * @param tag_type 2749 * The Tag Protocol ID 2750 * @return 2751 * - (0) if successful. 2752 * - (-ENOTSUP) if hardware-assisted VLAN TPID setup is not supported. 2753 * - (-ENODEV) if *port_id* invalid. 2754 * - (-EIO) if device is removed. 2755 */ 2756 int rte_eth_dev_set_vlan_ether_type(uint16_t port_id, 2757 enum rte_vlan_type vlan_type, 2758 uint16_t tag_type); 2759 2760 /** 2761 * Set VLAN offload configuration on an Ethernet device 2762 * Enable/Disable Extended VLAN by an Ethernet device, This is a register setup 2763 * available on some Intel NIC, not but all, please check the data sheet for 2764 * availability. 2765 * Enable/Disable VLAN Strip can be done on rx queue for certain NIC, but here 2766 * the configuration is applied on the port level. 2767 * 2768 * @param port_id 2769 * The port identifier of the Ethernet device. 2770 * @param offload_mask 2771 * The VLAN Offload bit mask can be mixed use with "OR" 2772 * ETH_VLAN_STRIP_OFFLOAD 2773 * ETH_VLAN_FILTER_OFFLOAD 2774 * ETH_VLAN_EXTEND_OFFLOAD 2775 * ETH_QINQ_STRIP_OFFLOAD 2776 * @return 2777 * - (0) if successful. 2778 * - (-ENOTSUP) if hardware-assisted VLAN filtering not configured. 2779 * - (-ENODEV) if *port_id* invalid. 2780 * - (-EIO) if device is removed. 2781 */ 2782 int rte_eth_dev_set_vlan_offload(uint16_t port_id, int offload_mask); 2783 2784 /** 2785 * Read VLAN Offload configuration from an Ethernet device 2786 * 2787 * @param port_id 2788 * The port identifier of the Ethernet device. 2789 * @return 2790 * - (>0) if successful. Bit mask to indicate 2791 * ETH_VLAN_STRIP_OFFLOAD 2792 * ETH_VLAN_FILTER_OFFLOAD 2793 * ETH_VLAN_EXTEND_OFFLOAD 2794 * ETH_QINQ_STRIP_OFFLOAD 2795 * - (-ENODEV) if *port_id* invalid. 2796 */ 2797 int rte_eth_dev_get_vlan_offload(uint16_t port_id); 2798 2799 /** 2800 * Set port based TX VLAN insertion on or off. 2801 * 2802 * @param port_id 2803 * The port identifier of the Ethernet device. 2804 * @param pvid 2805 * Port based TX VLAN identifier together with user priority. 2806 * @param on 2807 * Turn on or off the port based TX VLAN insertion. 2808 * 2809 * @return 2810 * - (0) if successful. 2811 * - negative if failed. 2812 */ 2813 int rte_eth_dev_set_vlan_pvid(uint16_t port_id, uint16_t pvid, int on); 2814 2815 typedef void (*buffer_tx_error_fn)(struct rte_mbuf **unsent, uint16_t count, 2816 void *userdata); 2817 2818 /** 2819 * Structure used to buffer packets for future TX 2820 * Used by APIs rte_eth_tx_buffer and rte_eth_tx_buffer_flush 2821 */ 2822 struct rte_eth_dev_tx_buffer { 2823 buffer_tx_error_fn error_callback; 2824 void *error_userdata; 2825 uint16_t size; /**< Size of buffer for buffered tx */ 2826 uint16_t length; /**< Number of packets in the array */ 2827 struct rte_mbuf *pkts[]; 2828 /**< Pending packets to be sent on explicit flush or when full */ 2829 }; 2830 2831 /** 2832 * Calculate the size of the tx buffer. 2833 * 2834 * @param sz 2835 * Number of stored packets. 2836 */ 2837 #define RTE_ETH_TX_BUFFER_SIZE(sz) \ 2838 (sizeof(struct rte_eth_dev_tx_buffer) + (sz) * sizeof(struct rte_mbuf *)) 2839 2840 /** 2841 * Initialize default values for buffered transmitting 2842 * 2843 * @param buffer 2844 * Tx buffer to be initialized. 2845 * @param size 2846 * Buffer size 2847 * @return 2848 * 0 if no error 2849 */ 2850 int 2851 rte_eth_tx_buffer_init(struct rte_eth_dev_tx_buffer *buffer, uint16_t size); 2852 2853 /** 2854 * Configure a callback for buffered packets which cannot be sent 2855 * 2856 * Register a specific callback to be called when an attempt is made to send 2857 * all packets buffered on an ethernet port, but not all packets can 2858 * successfully be sent. The callback registered here will be called only 2859 * from calls to rte_eth_tx_buffer() and rte_eth_tx_buffer_flush() APIs. 2860 * The default callback configured for each queue by default just frees the 2861 * packets back to the calling mempool. If additional behaviour is required, 2862 * for example, to count dropped packets, or to retry transmission of packets 2863 * which cannot be sent, this function should be used to register a suitable 2864 * callback function to implement the desired behaviour. 2865 * The example callback "rte_eth_count_unsent_packet_callback()" is also 2866 * provided as reference. 2867 * 2868 * @param buffer 2869 * The port identifier of the Ethernet device. 2870 * @param callback 2871 * The function to be used as the callback. 2872 * @param userdata 2873 * Arbitrary parameter to be passed to the callback function 2874 * @return 2875 * 0 on success, or -1 on error with rte_errno set appropriately 2876 */ 2877 int 2878 rte_eth_tx_buffer_set_err_callback(struct rte_eth_dev_tx_buffer *buffer, 2879 buffer_tx_error_fn callback, void *userdata); 2880 2881 /** 2882 * Callback function for silently dropping unsent buffered packets. 2883 * 2884 * This function can be passed to rte_eth_tx_buffer_set_err_callback() to 2885 * adjust the default behavior when buffered packets cannot be sent. This 2886 * function drops any unsent packets silently and is used by tx buffered 2887 * operations as default behavior. 2888 * 2889 * NOTE: this function should not be called directly, instead it should be used 2890 * as a callback for packet buffering. 2891 * 2892 * NOTE: when configuring this function as a callback with 2893 * rte_eth_tx_buffer_set_err_callback(), the final, userdata parameter 2894 * should point to an uint64_t value. 2895 * 2896 * @param pkts 2897 * The previously buffered packets which could not be sent 2898 * @param unsent 2899 * The number of unsent packets in the pkts array 2900 * @param userdata 2901 * Not used 2902 */ 2903 void 2904 rte_eth_tx_buffer_drop_callback(struct rte_mbuf **pkts, uint16_t unsent, 2905 void *userdata); 2906 2907 /** 2908 * Callback function for tracking unsent buffered packets. 2909 * 2910 * This function can be passed to rte_eth_tx_buffer_set_err_callback() to 2911 * adjust the default behavior when buffered packets cannot be sent. This 2912 * function drops any unsent packets, but also updates a user-supplied counter 2913 * to track the overall number of packets dropped. The counter should be an 2914 * uint64_t variable. 2915 * 2916 * NOTE: this function should not be called directly, instead it should be used 2917 * as a callback for packet buffering. 2918 * 2919 * NOTE: when configuring this function as a callback with 2920 * rte_eth_tx_buffer_set_err_callback(), the final, userdata parameter 2921 * should point to an uint64_t value. 2922 * 2923 * @param pkts 2924 * The previously buffered packets which could not be sent 2925 * @param unsent 2926 * The number of unsent packets in the pkts array 2927 * @param userdata 2928 * Pointer to an uint64_t value, which will be incremented by unsent 2929 */ 2930 void 2931 rte_eth_tx_buffer_count_callback(struct rte_mbuf **pkts, uint16_t unsent, 2932 void *userdata); 2933 2934 /** 2935 * Request the driver to free mbufs currently cached by the driver. The 2936 * driver will only free the mbuf if it is no longer in use. It is the 2937 * application's responsibility to ensure rte_eth_tx_buffer_flush(..) is 2938 * called if needed. 2939 * 2940 * @param port_id 2941 * The port identifier of the Ethernet device. 2942 * @param queue_id 2943 * The index of the transmit queue through which output packets must be 2944 * sent. 2945 * The value must be in the range [0, nb_tx_queue - 1] previously supplied 2946 * to rte_eth_dev_configure(). 2947 * @param free_cnt 2948 * Maximum number of packets to free. Use 0 to indicate all possible packets 2949 * should be freed. Note that a packet may be using multiple mbufs. 2950 * @return 2951 * Failure: < 0 2952 * -ENODEV: Invalid interface 2953 * -EIO: device is removed 2954 * -ENOTSUP: Driver does not support function 2955 * Success: >= 0 2956 * 0-n: Number of packets freed. More packets may still remain in ring that 2957 * are in use. 2958 */ 2959 int 2960 rte_eth_tx_done_cleanup(uint16_t port_id, uint16_t queue_id, uint32_t free_cnt); 2961 2962 /** 2963 * Subtypes for IPsec offload event(@ref RTE_ETH_EVENT_IPSEC) raised by 2964 * eth device. 2965 */ 2966 enum rte_eth_event_ipsec_subtype { 2967 RTE_ETH_EVENT_IPSEC_UNKNOWN = 0, 2968 /**< Unknown event type */ 2969 RTE_ETH_EVENT_IPSEC_ESN_OVERFLOW, 2970 /**< Sequence number overflow */ 2971 RTE_ETH_EVENT_IPSEC_SA_TIME_EXPIRY, 2972 /**< Soft time expiry of SA */ 2973 RTE_ETH_EVENT_IPSEC_SA_BYTE_EXPIRY, 2974 /**< Soft byte expiry of SA */ 2975 RTE_ETH_EVENT_IPSEC_MAX 2976 /**< Max value of this enum */ 2977 }; 2978 2979 /** 2980 * Descriptor for @ref RTE_ETH_EVENT_IPSEC event. Used by eth dev to send extra 2981 * information of the IPsec offload event. 2982 */ 2983 struct rte_eth_event_ipsec_desc { 2984 enum rte_eth_event_ipsec_subtype subtype; 2985 /**< Type of RTE_ETH_EVENT_IPSEC_* event */ 2986 uint64_t metadata; 2987 /**< Event specific metadata 2988 * 2989 * For the following events, *userdata* registered 2990 * with the *rte_security_session* would be returned 2991 * as metadata, 2992 * 2993 * - @ref RTE_ETH_EVENT_IPSEC_ESN_OVERFLOW 2994 * - @ref RTE_ETH_EVENT_IPSEC_SA_TIME_EXPIRY 2995 * - @ref RTE_ETH_EVENT_IPSEC_SA_BYTE_EXPIRY 2996 * 2997 * @see struct rte_security_session_conf 2998 * 2999 */ 3000 }; 3001 3002 /** 3003 * The eth device event type for interrupt, and maybe others in the future. 3004 */ 3005 enum rte_eth_event_type { 3006 RTE_ETH_EVENT_UNKNOWN, /**< unknown event type */ 3007 RTE_ETH_EVENT_INTR_LSC, /**< lsc interrupt event */ 3008 RTE_ETH_EVENT_QUEUE_STATE, 3009 /**< queue state event (enabled/disabled) */ 3010 RTE_ETH_EVENT_INTR_RESET, 3011 /**< reset interrupt event, sent to VF on PF reset */ 3012 RTE_ETH_EVENT_VF_MBOX, /**< message from the VF received by PF */ 3013 RTE_ETH_EVENT_MACSEC, /**< MACsec offload related event */ 3014 RTE_ETH_EVENT_INTR_RMV, /**< device removal event */ 3015 RTE_ETH_EVENT_NEW, /**< port is probed */ 3016 RTE_ETH_EVENT_DESTROY, /**< port is released */ 3017 RTE_ETH_EVENT_IPSEC, /**< IPsec offload related event */ 3018 RTE_ETH_EVENT_MAX /**< max value of this enum */ 3019 }; 3020 3021 typedef int (*rte_eth_dev_cb_fn)(uint16_t port_id, 3022 enum rte_eth_event_type event, void *cb_arg, void *ret_param); 3023 /**< user application callback to be registered for interrupts */ 3024 3025 /** 3026 * Register a callback function for port event. 3027 * 3028 * @param port_id 3029 * Port id. 3030 * RTE_ETH_ALL means register the event for all port ids. 3031 * @param event 3032 * Event interested. 3033 * @param cb_fn 3034 * User supplied callback function to be called. 3035 * @param cb_arg 3036 * Pointer to the parameters for the registered callback. 3037 * 3038 * @return 3039 * - On success, zero. 3040 * - On failure, a negative value. 3041 */ 3042 int rte_eth_dev_callback_register(uint16_t port_id, 3043 enum rte_eth_event_type event, 3044 rte_eth_dev_cb_fn cb_fn, void *cb_arg); 3045 3046 /** 3047 * Unregister a callback function for port event. 3048 * 3049 * @param port_id 3050 * Port id. 3051 * RTE_ETH_ALL means unregister the event for all port ids. 3052 * @param event 3053 * Event interested. 3054 * @param cb_fn 3055 * User supplied callback function to be called. 3056 * @param cb_arg 3057 * Pointer to the parameters for the registered callback. -1 means to 3058 * remove all for the same callback address and same event. 3059 * 3060 * @return 3061 * - On success, zero. 3062 * - On failure, a negative value. 3063 */ 3064 int rte_eth_dev_callback_unregister(uint16_t port_id, 3065 enum rte_eth_event_type event, 3066 rte_eth_dev_cb_fn cb_fn, void *cb_arg); 3067 3068 /** 3069 * When there is no rx packet coming in Rx Queue for a long time, we can 3070 * sleep lcore related to RX Queue for power saving, and enable rx interrupt 3071 * to be triggered when Rx packet arrives. 3072 * 3073 * The rte_eth_dev_rx_intr_enable() function enables rx queue 3074 * interrupt on specific rx queue of a port. 3075 * 3076 * @param port_id 3077 * The port identifier of the Ethernet device. 3078 * @param queue_id 3079 * The index of the receive queue from which to retrieve input packets. 3080 * The value must be in the range [0, nb_rx_queue - 1] previously supplied 3081 * to rte_eth_dev_configure(). 3082 * @return 3083 * - (0) if successful. 3084 * - (-ENOTSUP) if underlying hardware OR driver doesn't support 3085 * that operation. 3086 * - (-ENODEV) if *port_id* invalid. 3087 * - (-EIO) if device is removed. 3088 */ 3089 int rte_eth_dev_rx_intr_enable(uint16_t port_id, uint16_t queue_id); 3090 3091 /** 3092 * When lcore wakes up from rx interrupt indicating packet coming, disable rx 3093 * interrupt and returns to polling mode. 3094 * 3095 * The rte_eth_dev_rx_intr_disable() function disables rx queue 3096 * interrupt on specific rx queue of a port. 3097 * 3098 * @param port_id 3099 * The port identifier of the Ethernet device. 3100 * @param queue_id 3101 * The index of the receive queue from which to retrieve input packets. 3102 * The value must be in the range [0, nb_rx_queue - 1] previously supplied 3103 * to rte_eth_dev_configure(). 3104 * @return 3105 * - (0) if successful. 3106 * - (-ENOTSUP) if underlying hardware OR driver doesn't support 3107 * that operation. 3108 * - (-ENODEV) if *port_id* invalid. 3109 * - (-EIO) if device is removed. 3110 */ 3111 int rte_eth_dev_rx_intr_disable(uint16_t port_id, uint16_t queue_id); 3112 3113 /** 3114 * RX Interrupt control per port. 3115 * 3116 * @param port_id 3117 * The port identifier of the Ethernet device. 3118 * @param epfd 3119 * Epoll instance fd which the intr vector associated to. 3120 * Using RTE_EPOLL_PER_THREAD allows to use per thread epoll instance. 3121 * @param op 3122 * The operation be performed for the vector. 3123 * Operation type of {RTE_INTR_EVENT_ADD, RTE_INTR_EVENT_DEL}. 3124 * @param data 3125 * User raw data. 3126 * @return 3127 * - On success, zero. 3128 * - On failure, a negative value. 3129 */ 3130 int rte_eth_dev_rx_intr_ctl(uint16_t port_id, int epfd, int op, void *data); 3131 3132 /** 3133 * RX Interrupt control per queue. 3134 * 3135 * @param port_id 3136 * The port identifier of the Ethernet device. 3137 * @param queue_id 3138 * The index of the receive queue from which to retrieve input packets. 3139 * The value must be in the range [0, nb_rx_queue - 1] previously supplied 3140 * to rte_eth_dev_configure(). 3141 * @param epfd 3142 * Epoll instance fd which the intr vector associated to. 3143 * Using RTE_EPOLL_PER_THREAD allows to use per thread epoll instance. 3144 * @param op 3145 * The operation be performed for the vector. 3146 * Operation type of {RTE_INTR_EVENT_ADD, RTE_INTR_EVENT_DEL}. 3147 * @param data 3148 * User raw data. 3149 * @return 3150 * - On success, zero. 3151 * - On failure, a negative value. 3152 */ 3153 int rte_eth_dev_rx_intr_ctl_q(uint16_t port_id, uint16_t queue_id, 3154 int epfd, int op, void *data); 3155 3156 /** 3157 * @warning 3158 * @b EXPERIMENTAL: this API may change without prior notice. 3159 * 3160 * Get interrupt fd per Rx queue. 3161 * 3162 * @param port_id 3163 * The port identifier of the Ethernet device. 3164 * @param queue_id 3165 * The index of the receive queue from which to retrieve input packets. 3166 * The value must be in the range [0, nb_rx_queue - 1] previously supplied 3167 * to rte_eth_dev_configure(). 3168 * @return 3169 * - (>=0) the interrupt fd associated to the requested Rx queue if 3170 * successful. 3171 * - (-1) on error. 3172 */ 3173 __rte_experimental 3174 int 3175 rte_eth_dev_rx_intr_ctl_q_get_fd(uint16_t port_id, uint16_t queue_id); 3176 3177 /** 3178 * Turn on the LED on the Ethernet device. 3179 * This function turns on the LED on the Ethernet device. 3180 * 3181 * @param port_id 3182 * The port identifier of the Ethernet device. 3183 * @return 3184 * - (0) if successful. 3185 * - (-ENOTSUP) if underlying hardware OR driver doesn't support 3186 * that operation. 3187 * - (-ENODEV) if *port_id* invalid. 3188 * - (-EIO) if device is removed. 3189 */ 3190 int rte_eth_led_on(uint16_t port_id); 3191 3192 /** 3193 * Turn off the LED on the Ethernet device. 3194 * This function turns off the LED on the Ethernet device. 3195 * 3196 * @param port_id 3197 * The port identifier of the Ethernet device. 3198 * @return 3199 * - (0) if successful. 3200 * - (-ENOTSUP) if underlying hardware OR driver doesn't support 3201 * that operation. 3202 * - (-ENODEV) if *port_id* invalid. 3203 * - (-EIO) if device is removed. 3204 */ 3205 int rte_eth_led_off(uint16_t port_id); 3206 3207 /** 3208 * Get current status of the Ethernet link flow control for Ethernet device 3209 * 3210 * @param port_id 3211 * The port identifier of the Ethernet device. 3212 * @param fc_conf 3213 * The pointer to the structure where to store the flow control parameters. 3214 * @return 3215 * - (0) if successful. 3216 * - (-ENOTSUP) if hardware doesn't support flow control. 3217 * - (-ENODEV) if *port_id* invalid. 3218 * - (-EIO) if device is removed. 3219 */ 3220 int rte_eth_dev_flow_ctrl_get(uint16_t port_id, 3221 struct rte_eth_fc_conf *fc_conf); 3222 3223 /** 3224 * Configure the Ethernet link flow control for Ethernet device 3225 * 3226 * @param port_id 3227 * The port identifier of the Ethernet device. 3228 * @param fc_conf 3229 * The pointer to the structure of the flow control parameters. 3230 * @return 3231 * - (0) if successful. 3232 * - (-ENOTSUP) if hardware doesn't support flow control mode. 3233 * - (-ENODEV) if *port_id* invalid. 3234 * - (-EINVAL) if bad parameter 3235 * - (-EIO) if flow control setup failure or device is removed. 3236 */ 3237 int rte_eth_dev_flow_ctrl_set(uint16_t port_id, 3238 struct rte_eth_fc_conf *fc_conf); 3239 3240 /** 3241 * Configure the Ethernet priority flow control under DCB environment 3242 * for Ethernet device. 3243 * 3244 * @param port_id 3245 * The port identifier of the Ethernet device. 3246 * @param pfc_conf 3247 * The pointer to the structure of the priority flow control parameters. 3248 * @return 3249 * - (0) if successful. 3250 * - (-ENOTSUP) if hardware doesn't support priority flow control mode. 3251 * - (-ENODEV) if *port_id* invalid. 3252 * - (-EINVAL) if bad parameter 3253 * - (-EIO) if flow control setup failure or device is removed. 3254 */ 3255 int rte_eth_dev_priority_flow_ctrl_set(uint16_t port_id, 3256 struct rte_eth_pfc_conf *pfc_conf); 3257 3258 /** 3259 * Add a MAC address to an internal array of addresses used to enable whitelist 3260 * filtering to accept packets only if the destination MAC address matches. 3261 * 3262 * @param port_id 3263 * The port identifier of the Ethernet device. 3264 * @param mac_addr 3265 * The MAC address to add. 3266 * @param pool 3267 * VMDq pool index to associate address with (if VMDq is enabled). If VMDq is 3268 * not enabled, this should be set to 0. 3269 * @return 3270 * - (0) if successfully added or *mac_addr* was already added. 3271 * - (-ENOTSUP) if hardware doesn't support this feature. 3272 * - (-ENODEV) if *port* is invalid. 3273 * - (-EIO) if device is removed. 3274 * - (-ENOSPC) if no more MAC addresses can be added. 3275 * - (-EINVAL) if MAC address is invalid. 3276 */ 3277 int rte_eth_dev_mac_addr_add(uint16_t port_id, struct rte_ether_addr *mac_addr, 3278 uint32_t pool); 3279 3280 /** 3281 * Remove a MAC address from the internal array of addresses. 3282 * 3283 * @param port_id 3284 * The port identifier of the Ethernet device. 3285 * @param mac_addr 3286 * MAC address to remove. 3287 * @return 3288 * - (0) if successful, or *mac_addr* didn't exist. 3289 * - (-ENOTSUP) if hardware doesn't support. 3290 * - (-ENODEV) if *port* invalid. 3291 * - (-EADDRINUSE) if attempting to remove the default MAC address 3292 */ 3293 int rte_eth_dev_mac_addr_remove(uint16_t port_id, 3294 struct rte_ether_addr *mac_addr); 3295 3296 /** 3297 * Set the default MAC address. 3298 * 3299 * @param port_id 3300 * The port identifier of the Ethernet device. 3301 * @param mac_addr 3302 * New default MAC address. 3303 * @return 3304 * - (0) if successful, or *mac_addr* didn't exist. 3305 * - (-ENOTSUP) if hardware doesn't support. 3306 * - (-ENODEV) if *port* invalid. 3307 * - (-EINVAL) if MAC address is invalid. 3308 */ 3309 int rte_eth_dev_default_mac_addr_set(uint16_t port_id, 3310 struct rte_ether_addr *mac_addr); 3311 3312 /** 3313 * Update Redirection Table(RETA) of Receive Side Scaling of Ethernet device. 3314 * 3315 * @param port_id 3316 * The port identifier of the Ethernet device. 3317 * @param reta_conf 3318 * RETA to update. 3319 * @param reta_size 3320 * Redirection table size. The table size can be queried by 3321 * rte_eth_dev_info_get(). 3322 * @return 3323 * - (0) if successful. 3324 * - (-ENOTSUP) if hardware doesn't support. 3325 * - (-EINVAL) if bad parameter. 3326 * - (-EIO) if device is removed. 3327 */ 3328 int rte_eth_dev_rss_reta_update(uint16_t port_id, 3329 struct rte_eth_rss_reta_entry64 *reta_conf, 3330 uint16_t reta_size); 3331 3332 /** 3333 * Query Redirection Table(RETA) of Receive Side Scaling of Ethernet device. 3334 * 3335 * @param port_id 3336 * The port identifier of the Ethernet device. 3337 * @param reta_conf 3338 * RETA to query. For each requested reta entry, corresponding bit 3339 * in mask must be set. 3340 * @param reta_size 3341 * Redirection table size. The table size can be queried by 3342 * rte_eth_dev_info_get(). 3343 * @return 3344 * - (0) if successful. 3345 * - (-ENOTSUP) if hardware doesn't support. 3346 * - (-EINVAL) if bad parameter. 3347 * - (-EIO) if device is removed. 3348 */ 3349 int rte_eth_dev_rss_reta_query(uint16_t port_id, 3350 struct rte_eth_rss_reta_entry64 *reta_conf, 3351 uint16_t reta_size); 3352 3353 /** 3354 * Updates unicast hash table for receiving packet with the given destination 3355 * MAC address, and the packet is routed to all VFs for which the RX mode is 3356 * accept packets that match the unicast hash table. 3357 * 3358 * @param port_id 3359 * The port identifier of the Ethernet device. 3360 * @param addr 3361 * Unicast MAC address. 3362 * @param on 3363 * 1 - Set an unicast hash bit for receiving packets with the MAC address. 3364 * 0 - Clear an unicast hash bit. 3365 * @return 3366 * - (0) if successful. 3367 * - (-ENOTSUP) if hardware doesn't support. 3368 * - (-ENODEV) if *port_id* invalid. 3369 * - (-EIO) if device is removed. 3370 * - (-EINVAL) if bad parameter. 3371 */ 3372 int rte_eth_dev_uc_hash_table_set(uint16_t port_id, struct rte_ether_addr *addr, 3373 uint8_t on); 3374 3375 /** 3376 * Updates all unicast hash bitmaps for receiving packet with any Unicast 3377 * Ethernet MAC addresses,the packet is routed to all VFs for which the RX 3378 * mode is accept packets that match the unicast hash table. 3379 * 3380 * @param port_id 3381 * The port identifier of the Ethernet device. 3382 * @param on 3383 * 1 - Set all unicast hash bitmaps for receiving all the Ethernet 3384 * MAC addresses 3385 * 0 - Clear all unicast hash bitmaps 3386 * @return 3387 * - (0) if successful. 3388 * - (-ENOTSUP) if hardware doesn't support. 3389 * - (-ENODEV) if *port_id* invalid. 3390 * - (-EIO) if device is removed. 3391 * - (-EINVAL) if bad parameter. 3392 */ 3393 int rte_eth_dev_uc_all_hash_table_set(uint16_t port_id, uint8_t on); 3394 3395 /** 3396 * Set a traffic mirroring rule on an Ethernet device 3397 * 3398 * @param port_id 3399 * The port identifier of the Ethernet device. 3400 * @param mirror_conf 3401 * The pointer to the traffic mirroring structure describing the mirroring rule. 3402 * The *rte_eth_vm_mirror_conf* structure includes the type of mirroring rule, 3403 * destination pool and the value of rule if enable vlan or pool mirroring. 3404 * 3405 * @param rule_id 3406 * The index of traffic mirroring rule, we support four separated rules. 3407 * @param on 3408 * 1 - Enable a mirroring rule. 3409 * 0 - Disable a mirroring rule. 3410 * @return 3411 * - (0) if successful. 3412 * - (-ENOTSUP) if hardware doesn't support this feature. 3413 * - (-ENODEV) if *port_id* invalid. 3414 * - (-EIO) if device is removed. 3415 * - (-EINVAL) if the mr_conf information is not correct. 3416 */ 3417 int rte_eth_mirror_rule_set(uint16_t port_id, 3418 struct rte_eth_mirror_conf *mirror_conf, 3419 uint8_t rule_id, 3420 uint8_t on); 3421 3422 /** 3423 * Reset a traffic mirroring rule on an Ethernet device. 3424 * 3425 * @param port_id 3426 * The port identifier of the Ethernet device. 3427 * @param rule_id 3428 * The index of traffic mirroring rule, we support four separated rules. 3429 * @return 3430 * - (0) if successful. 3431 * - (-ENOTSUP) if hardware doesn't support this feature. 3432 * - (-ENODEV) if *port_id* invalid. 3433 * - (-EIO) if device is removed. 3434 * - (-EINVAL) if bad parameter. 3435 */ 3436 int rte_eth_mirror_rule_reset(uint16_t port_id, 3437 uint8_t rule_id); 3438 3439 /** 3440 * Set the rate limitation for a queue on an Ethernet device. 3441 * 3442 * @param port_id 3443 * The port identifier of the Ethernet device. 3444 * @param queue_idx 3445 * The queue id. 3446 * @param tx_rate 3447 * The tx rate in Mbps. Allocated from the total port link speed. 3448 * @return 3449 * - (0) if successful. 3450 * - (-ENOTSUP) if hardware doesn't support this feature. 3451 * - (-ENODEV) if *port_id* invalid. 3452 * - (-EIO) if device is removed. 3453 * - (-EINVAL) if bad parameter. 3454 */ 3455 int rte_eth_set_queue_rate_limit(uint16_t port_id, uint16_t queue_idx, 3456 uint16_t tx_rate); 3457 3458 /** 3459 * Configuration of Receive Side Scaling hash computation of Ethernet device. 3460 * 3461 * @param port_id 3462 * The port identifier of the Ethernet device. 3463 * @param rss_conf 3464 * The new configuration to use for RSS hash computation on the port. 3465 * @return 3466 * - (0) if successful. 3467 * - (-ENODEV) if port identifier is invalid. 3468 * - (-EIO) if device is removed. 3469 * - (-ENOTSUP) if hardware doesn't support. 3470 * - (-EINVAL) if bad parameter. 3471 */ 3472 int rte_eth_dev_rss_hash_update(uint16_t port_id, 3473 struct rte_eth_rss_conf *rss_conf); 3474 3475 /** 3476 * Retrieve current configuration of Receive Side Scaling hash computation 3477 * of Ethernet device. 3478 * 3479 * @param port_id 3480 * The port identifier of the Ethernet device. 3481 * @param rss_conf 3482 * Where to store the current RSS hash configuration of the Ethernet device. 3483 * @return 3484 * - (0) if successful. 3485 * - (-ENODEV) if port identifier is invalid. 3486 * - (-EIO) if device is removed. 3487 * - (-ENOTSUP) if hardware doesn't support RSS. 3488 */ 3489 int 3490 rte_eth_dev_rss_hash_conf_get(uint16_t port_id, 3491 struct rte_eth_rss_conf *rss_conf); 3492 3493 /** 3494 * Add UDP tunneling port for a specific type of tunnel. 3495 * The packets with this UDP port will be identified as this type of tunnel. 3496 * Before enabling any offloading function for a tunnel, users can call this API 3497 * to change or add more UDP port for the tunnel. So the offloading function 3498 * can take effect on the packets with the specific UDP port. 3499 * 3500 * @param port_id 3501 * The port identifier of the Ethernet device. 3502 * @param tunnel_udp 3503 * UDP tunneling configuration. 3504 * 3505 * @return 3506 * - (0) if successful. 3507 * - (-ENODEV) if port identifier is invalid. 3508 * - (-EIO) if device is removed. 3509 * - (-ENOTSUP) if hardware doesn't support tunnel type. 3510 */ 3511 int 3512 rte_eth_dev_udp_tunnel_port_add(uint16_t port_id, 3513 struct rte_eth_udp_tunnel *tunnel_udp); 3514 3515 /** 3516 * Delete UDP tunneling port a specific type of tunnel. 3517 * The packets with this UDP port will not be identified as this type of tunnel 3518 * any more. 3519 * Before enabling any offloading function for a tunnel, users can call this API 3520 * to delete a UDP port for the tunnel. So the offloading function will not take 3521 * effect on the packets with the specific UDP port. 3522 * 3523 * @param port_id 3524 * The port identifier of the Ethernet device. 3525 * @param tunnel_udp 3526 * UDP tunneling configuration. 3527 * 3528 * @return 3529 * - (0) if successful. 3530 * - (-ENODEV) if port identifier is invalid. 3531 * - (-EIO) if device is removed. 3532 * - (-ENOTSUP) if hardware doesn't support tunnel type. 3533 */ 3534 int 3535 rte_eth_dev_udp_tunnel_port_delete(uint16_t port_id, 3536 struct rte_eth_udp_tunnel *tunnel_udp); 3537 3538 /** 3539 * Check whether the filter type is supported on an Ethernet device. 3540 * All the supported filter types are defined in 'rte_eth_ctrl.h'. 3541 * 3542 * @param port_id 3543 * The port identifier of the Ethernet device. 3544 * @param filter_type 3545 * Filter type. 3546 * @return 3547 * - (0) if successful. 3548 * - (-ENOTSUP) if hardware doesn't support this filter type. 3549 * - (-ENODEV) if *port_id* invalid. 3550 * - (-EIO) if device is removed. 3551 */ 3552 __rte_deprecated 3553 int rte_eth_dev_filter_supported(uint16_t port_id, 3554 enum rte_filter_type filter_type); 3555 3556 /** 3557 * Take operations to assigned filter type on an Ethernet device. 3558 * All the supported operations and filter types are defined in 'rte_eth_ctrl.h'. 3559 * 3560 * @param port_id 3561 * The port identifier of the Ethernet device. 3562 * @param filter_type 3563 * Filter type. 3564 * @param filter_op 3565 * Type of operation. 3566 * @param arg 3567 * A pointer to arguments defined specifically for the operation. 3568 * @return 3569 * - (0) if successful. 3570 * - (-ENOTSUP) if hardware doesn't support. 3571 * - (-ENODEV) if *port_id* invalid. 3572 * - (-EIO) if device is removed. 3573 * - others depends on the specific operations implementation. 3574 */ 3575 __rte_deprecated 3576 int rte_eth_dev_filter_ctrl(uint16_t port_id, enum rte_filter_type filter_type, 3577 enum rte_filter_op filter_op, void *arg); 3578 3579 /** 3580 * Get DCB information on an Ethernet device. 3581 * 3582 * @param port_id 3583 * The port identifier of the Ethernet device. 3584 * @param dcb_info 3585 * dcb information. 3586 * @return 3587 * - (0) if successful. 3588 * - (-ENODEV) if port identifier is invalid. 3589 * - (-EIO) if device is removed. 3590 * - (-ENOTSUP) if hardware doesn't support. 3591 */ 3592 int rte_eth_dev_get_dcb_info(uint16_t port_id, 3593 struct rte_eth_dcb_info *dcb_info); 3594 3595 struct rte_eth_rxtx_callback; 3596 3597 /** 3598 * Add a callback to be called on packet RX on a given port and queue. 3599 * 3600 * This API configures a function to be called for each burst of 3601 * packets received on a given NIC port queue. The return value is a pointer 3602 * that can be used to later remove the callback using 3603 * rte_eth_remove_rx_callback(). 3604 * 3605 * Multiple functions are called in the order that they are added. 3606 * 3607 * @param port_id 3608 * The port identifier of the Ethernet device. 3609 * @param queue_id 3610 * The queue on the Ethernet device on which the callback is to be added. 3611 * @param fn 3612 * The callback function 3613 * @param user_param 3614 * A generic pointer parameter which will be passed to each invocation of the 3615 * callback function on this port and queue. 3616 * 3617 * @return 3618 * NULL on error. 3619 * On success, a pointer value which can later be used to remove the callback. 3620 */ 3621 const struct rte_eth_rxtx_callback * 3622 rte_eth_add_rx_callback(uint16_t port_id, uint16_t queue_id, 3623 rte_rx_callback_fn fn, void *user_param); 3624 3625 /** 3626 * Add a callback that must be called first on packet RX on a given port 3627 * and queue. 3628 * 3629 * This API configures a first function to be called for each burst of 3630 * packets received on a given NIC port queue. The return value is a pointer 3631 * that can be used to later remove the callback using 3632 * rte_eth_remove_rx_callback(). 3633 * 3634 * Multiple functions are called in the order that they are added. 3635 * 3636 * @param port_id 3637 * The port identifier of the Ethernet device. 3638 * @param queue_id 3639 * The queue on the Ethernet device on which the callback is to be added. 3640 * @param fn 3641 * The callback function 3642 * @param user_param 3643 * A generic pointer parameter which will be passed to each invocation of the 3644 * callback function on this port and queue. 3645 * 3646 * @return 3647 * NULL on error. 3648 * On success, a pointer value which can later be used to remove the callback. 3649 */ 3650 const struct rte_eth_rxtx_callback * 3651 rte_eth_add_first_rx_callback(uint16_t port_id, uint16_t queue_id, 3652 rte_rx_callback_fn fn, void *user_param); 3653 3654 /** 3655 * Add a callback to be called on packet TX on a given port and queue. 3656 * 3657 * This API configures a function to be called for each burst of 3658 * packets sent on a given NIC port queue. The return value is a pointer 3659 * that can be used to later remove the callback using 3660 * rte_eth_remove_tx_callback(). 3661 * 3662 * Multiple functions are called in the order that they are added. 3663 * 3664 * @param port_id 3665 * The port identifier of the Ethernet device. 3666 * @param queue_id 3667 * The queue on the Ethernet device on which the callback is to be added. 3668 * @param fn 3669 * The callback function 3670 * @param user_param 3671 * A generic pointer parameter which will be passed to each invocation of the 3672 * callback function on this port and queue. 3673 * 3674 * @return 3675 * NULL on error. 3676 * On success, a pointer value which can later be used to remove the callback. 3677 */ 3678 const struct rte_eth_rxtx_callback * 3679 rte_eth_add_tx_callback(uint16_t port_id, uint16_t queue_id, 3680 rte_tx_callback_fn fn, void *user_param); 3681 3682 /** 3683 * Remove an RX packet callback from a given port and queue. 3684 * 3685 * This function is used to removed callbacks that were added to a NIC port 3686 * queue using rte_eth_add_rx_callback(). 3687 * 3688 * Note: the callback is removed from the callback list but it isn't freed 3689 * since the it may still be in use. The memory for the callback can be 3690 * subsequently freed back by the application by calling rte_free(): 3691 * 3692 * - Immediately - if the port is stopped, or the user knows that no 3693 * callbacks are in flight e.g. if called from the thread doing RX/TX 3694 * on that queue. 3695 * 3696 * - After a short delay - where the delay is sufficient to allow any 3697 * in-flight callbacks to complete. 3698 * 3699 * @param port_id 3700 * The port identifier of the Ethernet device. 3701 * @param queue_id 3702 * The queue on the Ethernet device from which the callback is to be removed. 3703 * @param user_cb 3704 * User supplied callback created via rte_eth_add_rx_callback(). 3705 * 3706 * @return 3707 * - 0: Success. Callback was removed. 3708 * - -ENOTSUP: Callback support is not available. 3709 * - -EINVAL: The port_id or the queue_id is out of range, or the callback 3710 * is NULL or not found for the port/queue. 3711 */ 3712 int rte_eth_remove_rx_callback(uint16_t port_id, uint16_t queue_id, 3713 const struct rte_eth_rxtx_callback *user_cb); 3714 3715 /** 3716 * Remove a TX packet callback from a given port and queue. 3717 * 3718 * This function is used to removed callbacks that were added to a NIC port 3719 * queue using rte_eth_add_tx_callback(). 3720 * 3721 * Note: the callback is removed from the callback list but it isn't freed 3722 * since the it may still be in use. The memory for the callback can be 3723 * subsequently freed back by the application by calling rte_free(): 3724 * 3725 * - Immediately - if the port is stopped, or the user knows that no 3726 * callbacks are in flight e.g. if called from the thread doing RX/TX 3727 * on that queue. 3728 * 3729 * - After a short delay - where the delay is sufficient to allow any 3730 * in-flight callbacks to complete. 3731 * 3732 * @param port_id 3733 * The port identifier of the Ethernet device. 3734 * @param queue_id 3735 * The queue on the Ethernet device from which the callback is to be removed. 3736 * @param user_cb 3737 * User supplied callback created via rte_eth_add_tx_callback(). 3738 * 3739 * @return 3740 * - 0: Success. Callback was removed. 3741 * - -ENOTSUP: Callback support is not available. 3742 * - -EINVAL: The port_id or the queue_id is out of range, or the callback 3743 * is NULL or not found for the port/queue. 3744 */ 3745 int rte_eth_remove_tx_callback(uint16_t port_id, uint16_t queue_id, 3746 const struct rte_eth_rxtx_callback *user_cb); 3747 3748 /** 3749 * Retrieve information about given port's RX queue. 3750 * 3751 * @param port_id 3752 * The port identifier of the Ethernet device. 3753 * @param queue_id 3754 * The RX queue on the Ethernet device for which information 3755 * will be retrieved. 3756 * @param qinfo 3757 * A pointer to a structure of type *rte_eth_rxq_info_info* to be filled with 3758 * the information of the Ethernet device. 3759 * 3760 * @return 3761 * - 0: Success 3762 * - -ENOTSUP: routine is not supported by the device PMD. 3763 * - -EINVAL: The port_id or the queue_id is out of range, or the queue 3764 * is hairpin queue. 3765 */ 3766 int rte_eth_rx_queue_info_get(uint16_t port_id, uint16_t queue_id, 3767 struct rte_eth_rxq_info *qinfo); 3768 3769 /** 3770 * Retrieve information about given port's TX queue. 3771 * 3772 * @param port_id 3773 * The port identifier of the Ethernet device. 3774 * @param queue_id 3775 * The TX queue on the Ethernet device for which information 3776 * will be retrieved. 3777 * @param qinfo 3778 * A pointer to a structure of type *rte_eth_txq_info_info* to be filled with 3779 * the information of the Ethernet device. 3780 * 3781 * @return 3782 * - 0: Success 3783 * - -ENOTSUP: routine is not supported by the device PMD. 3784 * - -EINVAL: The port_id or the queue_id is out of range, or the queue 3785 * is hairpin queue. 3786 */ 3787 int rte_eth_tx_queue_info_get(uint16_t port_id, uint16_t queue_id, 3788 struct rte_eth_txq_info *qinfo); 3789 3790 /** 3791 * Retrieve information about the Rx packet burst mode. 3792 * 3793 * @param port_id 3794 * The port identifier of the Ethernet device. 3795 * @param queue_id 3796 * The Rx queue on the Ethernet device for which information 3797 * will be retrieved. 3798 * @param mode 3799 * A pointer to a structure of type *rte_eth_burst_mode* to be filled 3800 * with the information of the packet burst mode. 3801 * 3802 * @return 3803 * - 0: Success 3804 * - -ENOTSUP: routine is not supported by the device PMD. 3805 * - -EINVAL: The port_id or the queue_id is out of range. 3806 */ 3807 __rte_experimental 3808 int rte_eth_rx_burst_mode_get(uint16_t port_id, uint16_t queue_id, 3809 struct rte_eth_burst_mode *mode); 3810 3811 /** 3812 * Retrieve information about the Tx packet burst mode. 3813 * 3814 * @param port_id 3815 * The port identifier of the Ethernet device. 3816 * @param queue_id 3817 * The Tx queue on the Ethernet device for which information 3818 * will be retrieved. 3819 * @param mode 3820 * A pointer to a structure of type *rte_eth_burst_mode* to be filled 3821 * with the information of the packet burst mode. 3822 * 3823 * @return 3824 * - 0: Success 3825 * - -ENOTSUP: routine is not supported by the device PMD. 3826 * - -EINVAL: The port_id or the queue_id is out of range. 3827 */ 3828 __rte_experimental 3829 int rte_eth_tx_burst_mode_get(uint16_t port_id, uint16_t queue_id, 3830 struct rte_eth_burst_mode *mode); 3831 3832 /** 3833 * Retrieve device registers and register attributes (number of registers and 3834 * register size) 3835 * 3836 * @param port_id 3837 * The port identifier of the Ethernet device. 3838 * @param info 3839 * Pointer to rte_dev_reg_info structure to fill in. If info->data is 3840 * NULL the function fills in the width and length fields. If non-NULL 3841 * the registers are put into the buffer pointed at by the data field. 3842 * @return 3843 * - (0) if successful. 3844 * - (-ENOTSUP) if hardware doesn't support. 3845 * - (-ENODEV) if *port_id* invalid. 3846 * - (-EIO) if device is removed. 3847 * - others depends on the specific operations implementation. 3848 */ 3849 int rte_eth_dev_get_reg_info(uint16_t port_id, struct rte_dev_reg_info *info); 3850 3851 /** 3852 * Retrieve size of device EEPROM 3853 * 3854 * @param port_id 3855 * The port identifier of the Ethernet device. 3856 * @return 3857 * - (>=0) EEPROM size if successful. 3858 * - (-ENOTSUP) if hardware doesn't support. 3859 * - (-ENODEV) if *port_id* invalid. 3860 * - (-EIO) if device is removed. 3861 * - others depends on the specific operations implementation. 3862 */ 3863 int rte_eth_dev_get_eeprom_length(uint16_t port_id); 3864 3865 /** 3866 * Retrieve EEPROM and EEPROM attribute 3867 * 3868 * @param port_id 3869 * The port identifier of the Ethernet device. 3870 * @param info 3871 * The template includes buffer for return EEPROM data and 3872 * EEPROM attributes to be filled. 3873 * @return 3874 * - (0) if successful. 3875 * - (-ENOTSUP) if hardware doesn't support. 3876 * - (-ENODEV) if *port_id* invalid. 3877 * - (-EIO) if device is removed. 3878 * - others depends on the specific operations implementation. 3879 */ 3880 int rte_eth_dev_get_eeprom(uint16_t port_id, struct rte_dev_eeprom_info *info); 3881 3882 /** 3883 * Program EEPROM with provided data 3884 * 3885 * @param port_id 3886 * The port identifier of the Ethernet device. 3887 * @param info 3888 * The template includes EEPROM data for programming and 3889 * EEPROM attributes to be filled 3890 * @return 3891 * - (0) if successful. 3892 * - (-ENOTSUP) if hardware doesn't support. 3893 * - (-ENODEV) if *port_id* invalid. 3894 * - (-EIO) if device is removed. 3895 * - others depends on the specific operations implementation. 3896 */ 3897 int rte_eth_dev_set_eeprom(uint16_t port_id, struct rte_dev_eeprom_info *info); 3898 3899 /** 3900 * @warning 3901 * @b EXPERIMENTAL: this API may change without prior notice. 3902 * 3903 * Retrieve the type and size of plugin module EEPROM 3904 * 3905 * @param port_id 3906 * The port identifier of the Ethernet device. 3907 * @param modinfo 3908 * The type and size of plugin module EEPROM. 3909 * @return 3910 * - (0) if successful. 3911 * - (-ENOTSUP) if hardware doesn't support. 3912 * - (-ENODEV) if *port_id* invalid. 3913 * - (-EIO) if device is removed. 3914 * - others depends on the specific operations implementation. 3915 */ 3916 __rte_experimental 3917 int 3918 rte_eth_dev_get_module_info(uint16_t port_id, 3919 struct rte_eth_dev_module_info *modinfo); 3920 3921 /** 3922 * @warning 3923 * @b EXPERIMENTAL: this API may change without prior notice. 3924 * 3925 * Retrieve the data of plugin module EEPROM 3926 * 3927 * @param port_id 3928 * The port identifier of the Ethernet device. 3929 * @param info 3930 * The template includes the plugin module EEPROM attributes, and the 3931 * buffer for return plugin module EEPROM data. 3932 * @return 3933 * - (0) if successful. 3934 * - (-ENOTSUP) if hardware doesn't support. 3935 * - (-ENODEV) if *port_id* invalid. 3936 * - (-EIO) if device is removed. 3937 * - others depends on the specific operations implementation. 3938 */ 3939 __rte_experimental 3940 int 3941 rte_eth_dev_get_module_eeprom(uint16_t port_id, 3942 struct rte_dev_eeprom_info *info); 3943 3944 /** 3945 * Set the list of multicast addresses to filter on an Ethernet device. 3946 * 3947 * @param port_id 3948 * The port identifier of the Ethernet device. 3949 * @param mc_addr_set 3950 * The array of multicast addresses to set. Equal to NULL when the function 3951 * is invoked to flush the set of filtered addresses. 3952 * @param nb_mc_addr 3953 * The number of multicast addresses in the *mc_addr_set* array. Equal to 0 3954 * when the function is invoked to flush the set of filtered addresses. 3955 * @return 3956 * - (0) if successful. 3957 * - (-ENODEV) if *port_id* invalid. 3958 * - (-EIO) if device is removed. 3959 * - (-ENOTSUP) if PMD of *port_id* doesn't support multicast filtering. 3960 * - (-ENOSPC) if *port_id* has not enough multicast filtering resources. 3961 */ 3962 int rte_eth_dev_set_mc_addr_list(uint16_t port_id, 3963 struct rte_ether_addr *mc_addr_set, 3964 uint32_t nb_mc_addr); 3965 3966 /** 3967 * Enable IEEE1588/802.1AS timestamping for an Ethernet device. 3968 * 3969 * @param port_id 3970 * The port identifier of the Ethernet device. 3971 * 3972 * @return 3973 * - 0: Success. 3974 * - -ENODEV: The port ID is invalid. 3975 * - -EIO: if device is removed. 3976 * - -ENOTSUP: The function is not supported by the Ethernet driver. 3977 */ 3978 int rte_eth_timesync_enable(uint16_t port_id); 3979 3980 /** 3981 * Disable IEEE1588/802.1AS timestamping for an Ethernet device. 3982 * 3983 * @param port_id 3984 * The port identifier of the Ethernet device. 3985 * 3986 * @return 3987 * - 0: Success. 3988 * - -ENODEV: The port ID is invalid. 3989 * - -EIO: if device is removed. 3990 * - -ENOTSUP: The function is not supported by the Ethernet driver. 3991 */ 3992 int rte_eth_timesync_disable(uint16_t port_id); 3993 3994 /** 3995 * Read an IEEE1588/802.1AS RX timestamp from an Ethernet device. 3996 * 3997 * @param port_id 3998 * The port identifier of the Ethernet device. 3999 * @param timestamp 4000 * Pointer to the timestamp struct. 4001 * @param flags 4002 * Device specific flags. Used to pass the RX timesync register index to 4003 * i40e. Unused in igb/ixgbe, pass 0 instead. 4004 * 4005 * @return 4006 * - 0: Success. 4007 * - -EINVAL: No timestamp is available. 4008 * - -ENODEV: The port ID is invalid. 4009 * - -EIO: if device is removed. 4010 * - -ENOTSUP: The function is not supported by the Ethernet driver. 4011 */ 4012 int rte_eth_timesync_read_rx_timestamp(uint16_t port_id, 4013 struct timespec *timestamp, uint32_t flags); 4014 4015 /** 4016 * Read an IEEE1588/802.1AS TX timestamp from an Ethernet device. 4017 * 4018 * @param port_id 4019 * The port identifier of the Ethernet device. 4020 * @param timestamp 4021 * Pointer to the timestamp struct. 4022 * 4023 * @return 4024 * - 0: Success. 4025 * - -EINVAL: No timestamp is available. 4026 * - -ENODEV: The port ID is invalid. 4027 * - -EIO: if device is removed. 4028 * - -ENOTSUP: The function is not supported by the Ethernet driver. 4029 */ 4030 int rte_eth_timesync_read_tx_timestamp(uint16_t port_id, 4031 struct timespec *timestamp); 4032 4033 /** 4034 * Adjust the timesync clock on an Ethernet device. 4035 * 4036 * This is usually used in conjunction with other Ethdev timesync functions to 4037 * synchronize the device time using the IEEE1588/802.1AS protocol. 4038 * 4039 * @param port_id 4040 * The port identifier of the Ethernet device. 4041 * @param delta 4042 * The adjustment in nanoseconds. 4043 * 4044 * @return 4045 * - 0: Success. 4046 * - -ENODEV: The port ID is invalid. 4047 * - -EIO: if device is removed. 4048 * - -ENOTSUP: The function is not supported by the Ethernet driver. 4049 */ 4050 int rte_eth_timesync_adjust_time(uint16_t port_id, int64_t delta); 4051 4052 /** 4053 * Read the time from the timesync clock on an Ethernet device. 4054 * 4055 * This is usually used in conjunction with other Ethdev timesync functions to 4056 * synchronize the device time using the IEEE1588/802.1AS protocol. 4057 * 4058 * @param port_id 4059 * The port identifier of the Ethernet device. 4060 * @param time 4061 * Pointer to the timespec struct that holds the time. 4062 * 4063 * @return 4064 * - 0: Success. 4065 */ 4066 int rte_eth_timesync_read_time(uint16_t port_id, struct timespec *time); 4067 4068 /** 4069 * Set the time of the timesync clock on an Ethernet device. 4070 * 4071 * This is usually used in conjunction with other Ethdev timesync functions to 4072 * synchronize the device time using the IEEE1588/802.1AS protocol. 4073 * 4074 * @param port_id 4075 * The port identifier of the Ethernet device. 4076 * @param time 4077 * Pointer to the timespec struct that holds the time. 4078 * 4079 * @return 4080 * - 0: Success. 4081 * - -EINVAL: No timestamp is available. 4082 * - -ENODEV: The port ID is invalid. 4083 * - -EIO: if device is removed. 4084 * - -ENOTSUP: The function is not supported by the Ethernet driver. 4085 */ 4086 int rte_eth_timesync_write_time(uint16_t port_id, const struct timespec *time); 4087 4088 /** 4089 * @warning 4090 * @b EXPERIMENTAL: this API may change without prior notice. 4091 * 4092 * Read the current clock counter of an Ethernet device 4093 * 4094 * This returns the current raw clock value of an Ethernet device. It is 4095 * a raw amount of ticks, with no given time reference. 4096 * The value returned here is from the same clock than the one 4097 * filling timestamp field of Rx packets when using hardware timestamp 4098 * offload. Therefore it can be used to compute a precise conversion of 4099 * the device clock to the real time. 4100 * 4101 * E.g, a simple heuristic to derivate the frequency would be: 4102 * uint64_t start, end; 4103 * rte_eth_read_clock(port, start); 4104 * rte_delay_ms(100); 4105 * rte_eth_read_clock(port, end); 4106 * double freq = (end - start) * 10; 4107 * 4108 * Compute a common reference with: 4109 * uint64_t base_time_sec = current_time(); 4110 * uint64_t base_clock; 4111 * rte_eth_read_clock(port, base_clock); 4112 * 4113 * Then, convert the raw mbuf timestamp with: 4114 * base_time_sec + (double)(mbuf->timestamp - base_clock) / freq; 4115 * 4116 * This simple example will not provide a very good accuracy. One must 4117 * at least measure multiple times the frequency and do a regression. 4118 * To avoid deviation from the system time, the common reference can 4119 * be repeated from time to time. The integer division can also be 4120 * converted by a multiplication and a shift for better performance. 4121 * 4122 * @param port_id 4123 * The port identifier of the Ethernet device. 4124 * @param clock 4125 * Pointer to the uint64_t that holds the raw clock value. 4126 * 4127 * @return 4128 * - 0: Success. 4129 * - -ENODEV: The port ID is invalid. 4130 * - -ENOTSUP: The function is not supported by the Ethernet driver. 4131 */ 4132 __rte_experimental 4133 int 4134 rte_eth_read_clock(uint16_t port_id, uint64_t *clock); 4135 4136 /** 4137 * Config l2 tunnel ether type of an Ethernet device for filtering specific 4138 * tunnel packets by ether type. 4139 * 4140 * @param port_id 4141 * The port identifier of the Ethernet device. 4142 * @param l2_tunnel 4143 * l2 tunnel configuration. 4144 * 4145 * @return 4146 * - (0) if successful. 4147 * - (-ENODEV) if port identifier is invalid. 4148 * - (-EIO) if device is removed. 4149 * - (-ENOTSUP) if hardware doesn't support tunnel type. 4150 */ 4151 int 4152 rte_eth_dev_l2_tunnel_eth_type_conf(uint16_t port_id, 4153 struct rte_eth_l2_tunnel_conf *l2_tunnel); 4154 4155 /** 4156 * Enable/disable l2 tunnel offload functions. Include, 4157 * 1, The ability of parsing a type of l2 tunnel of an Ethernet device. 4158 * Filtering, forwarding and offloading this type of tunnel packets depend on 4159 * this ability. 4160 * 2, Stripping the l2 tunnel tag. 4161 * 3, Insertion of the l2 tunnel tag. 4162 * 4, Forwarding the packets based on the l2 tunnel tag. 4163 * 4164 * @param port_id 4165 * The port identifier of the Ethernet device. 4166 * @param l2_tunnel 4167 * l2 tunnel parameters. 4168 * @param mask 4169 * Indicate the offload function. 4170 * @param en 4171 * Enable or disable this function. 4172 * 4173 * @return 4174 * - (0) if successful. 4175 * - (-ENODEV) if port identifier is invalid. 4176 * - (-EIO) if device is removed. 4177 * - (-ENOTSUP) if hardware doesn't support tunnel type. 4178 */ 4179 int 4180 rte_eth_dev_l2_tunnel_offload_set(uint16_t port_id, 4181 struct rte_eth_l2_tunnel_conf *l2_tunnel, 4182 uint32_t mask, 4183 uint8_t en); 4184 4185 /** 4186 * Get the port id from device name. The device name should be specified 4187 * as below: 4188 * - PCIe address (Domain:Bus:Device.Function), for example- 0000:2:00.0 4189 * - SoC device name, for example- fsl-gmac0 4190 * - vdev dpdk name, for example- net_[pcap0|null0|tap0] 4191 * 4192 * @param name 4193 * pci address or name of the device 4194 * @param port_id 4195 * pointer to port identifier of the device 4196 * @return 4197 * - (0) if successful and port_id is filled. 4198 * - (-ENODEV or -EINVAL) on failure. 4199 */ 4200 int 4201 rte_eth_dev_get_port_by_name(const char *name, uint16_t *port_id); 4202 4203 /** 4204 * Get the device name from port id. The device name is specified as below: 4205 * - PCIe address (Domain:Bus:Device.Function), for example- 0000:02:00.0 4206 * - SoC device name, for example- fsl-gmac0 4207 * - vdev dpdk name, for example- net_[pcap0|null0|tun0|tap0] 4208 * 4209 * @param port_id 4210 * Port identifier of the device. 4211 * @param name 4212 * Buffer of size RTE_ETH_NAME_MAX_LEN to store the name. 4213 * @return 4214 * - (0) if successful. 4215 * - (-EINVAL) on failure. 4216 */ 4217 int 4218 rte_eth_dev_get_name_by_port(uint16_t port_id, char *name); 4219 4220 /** 4221 * Check that numbers of Rx and Tx descriptors satisfy descriptors limits from 4222 * the ethernet device information, otherwise adjust them to boundaries. 4223 * 4224 * @param port_id 4225 * The port identifier of the Ethernet device. 4226 * @param nb_rx_desc 4227 * A pointer to a uint16_t where the number of receive 4228 * descriptors stored. 4229 * @param nb_tx_desc 4230 * A pointer to a uint16_t where the number of transmit 4231 * descriptors stored. 4232 * @return 4233 * - (0) if successful. 4234 * - (-ENOTSUP, -ENODEV or -EINVAL) on failure. 4235 */ 4236 int rte_eth_dev_adjust_nb_rx_tx_desc(uint16_t port_id, 4237 uint16_t *nb_rx_desc, 4238 uint16_t *nb_tx_desc); 4239 4240 /** 4241 * Test if a port supports specific mempool ops. 4242 * 4243 * @param port_id 4244 * Port identifier of the Ethernet device. 4245 * @param [in] pool 4246 * The name of the pool operations to test. 4247 * @return 4248 * - 0: best mempool ops choice for this port. 4249 * - 1: mempool ops are supported for this port. 4250 * - -ENOTSUP: mempool ops not supported for this port. 4251 * - -ENODEV: Invalid port Identifier. 4252 * - -EINVAL: Pool param is null. 4253 */ 4254 int 4255 rte_eth_dev_pool_ops_supported(uint16_t port_id, const char *pool); 4256 4257 /** 4258 * Get the security context for the Ethernet device. 4259 * 4260 * @param port_id 4261 * Port identifier of the Ethernet device 4262 * @return 4263 * - NULL on error. 4264 * - pointer to security context on success. 4265 */ 4266 void * 4267 rte_eth_dev_get_sec_ctx(uint16_t port_id); 4268 4269 /** 4270 * @warning 4271 * @b EXPERIMENTAL: this API may change, or be removed, without prior notice 4272 * 4273 * Query the device hairpin capabilities. 4274 * 4275 * @param port_id 4276 * The port identifier of the Ethernet device. 4277 * @param cap 4278 * Pointer to a structure that will hold the hairpin capabilities. 4279 * @return 4280 * - (0) if successful. 4281 * - (-ENOTSUP) if hardware doesn't support. 4282 */ 4283 __rte_experimental 4284 int rte_eth_dev_hairpin_capability_get(uint16_t port_id, 4285 struct rte_eth_hairpin_cap *cap); 4286 4287 #include <rte_ethdev_core.h> 4288 4289 /** 4290 * 4291 * Retrieve a burst of input packets from a receive queue of an Ethernet 4292 * device. The retrieved packets are stored in *rte_mbuf* structures whose 4293 * pointers are supplied in the *rx_pkts* array. 4294 * 4295 * The rte_eth_rx_burst() function loops, parsing the RX ring of the 4296 * receive queue, up to *nb_pkts* packets, and for each completed RX 4297 * descriptor in the ring, it performs the following operations: 4298 * 4299 * - Initialize the *rte_mbuf* data structure associated with the 4300 * RX descriptor according to the information provided by the NIC into 4301 * that RX descriptor. 4302 * 4303 * - Store the *rte_mbuf* data structure into the next entry of the 4304 * *rx_pkts* array. 4305 * 4306 * - Replenish the RX descriptor with a new *rte_mbuf* buffer 4307 * allocated from the memory pool associated with the receive queue at 4308 * initialization time. 4309 * 4310 * When retrieving an input packet that was scattered by the controller 4311 * into multiple receive descriptors, the rte_eth_rx_burst() function 4312 * appends the associated *rte_mbuf* buffers to the first buffer of the 4313 * packet. 4314 * 4315 * The rte_eth_rx_burst() function returns the number of packets 4316 * actually retrieved, which is the number of *rte_mbuf* data structures 4317 * effectively supplied into the *rx_pkts* array. 4318 * A return value equal to *nb_pkts* indicates that the RX queue contained 4319 * at least *rx_pkts* packets, and this is likely to signify that other 4320 * received packets remain in the input queue. Applications implementing 4321 * a "retrieve as much received packets as possible" policy can check this 4322 * specific case and keep invoking the rte_eth_rx_burst() function until 4323 * a value less than *nb_pkts* is returned. 4324 * 4325 * This receive method has the following advantages: 4326 * 4327 * - It allows a run-to-completion network stack engine to retrieve and 4328 * to immediately process received packets in a fast burst-oriented 4329 * approach, avoiding the overhead of unnecessary intermediate packet 4330 * queue/dequeue operations. 4331 * 4332 * - Conversely, it also allows an asynchronous-oriented processing 4333 * method to retrieve bursts of received packets and to immediately 4334 * queue them for further parallel processing by another logical core, 4335 * for instance. However, instead of having received packets being 4336 * individually queued by the driver, this approach allows the caller 4337 * of the rte_eth_rx_burst() function to queue a burst of retrieved 4338 * packets at a time and therefore dramatically reduce the cost of 4339 * enqueue/dequeue operations per packet. 4340 * 4341 * - It allows the rte_eth_rx_burst() function of the driver to take 4342 * advantage of burst-oriented hardware features (CPU cache, 4343 * prefetch instructions, and so on) to minimize the number of CPU 4344 * cycles per packet. 4345 * 4346 * To summarize, the proposed receive API enables many 4347 * burst-oriented optimizations in both synchronous and asynchronous 4348 * packet processing environments with no overhead in both cases. 4349 * 4350 * The rte_eth_rx_burst() function does not provide any error 4351 * notification to avoid the corresponding overhead. As a hint, the 4352 * upper-level application might check the status of the device link once 4353 * being systematically returned a 0 value for a given number of tries. 4354 * 4355 * @param port_id 4356 * The port identifier of the Ethernet device. 4357 * @param queue_id 4358 * The index of the receive queue from which to retrieve input packets. 4359 * The value must be in the range [0, nb_rx_queue - 1] previously supplied 4360 * to rte_eth_dev_configure(). 4361 * @param rx_pkts 4362 * The address of an array of pointers to *rte_mbuf* structures that 4363 * must be large enough to store *nb_pkts* pointers in it. 4364 * @param nb_pkts 4365 * The maximum number of packets to retrieve. 4366 * @return 4367 * The number of packets actually retrieved, which is the number 4368 * of pointers to *rte_mbuf* structures effectively supplied to the 4369 * *rx_pkts* array. 4370 */ 4371 static inline uint16_t 4372 rte_eth_rx_burst(uint16_t port_id, uint16_t queue_id, 4373 struct rte_mbuf **rx_pkts, const uint16_t nb_pkts) 4374 { 4375 struct rte_eth_dev *dev = &rte_eth_devices[port_id]; 4376 uint16_t nb_rx; 4377 4378 #ifdef RTE_LIBRTE_ETHDEV_DEBUG 4379 RTE_ETH_VALID_PORTID_OR_ERR_RET(port_id, 0); 4380 RTE_FUNC_PTR_OR_ERR_RET(*dev->rx_pkt_burst, 0); 4381 4382 if (queue_id >= dev->data->nb_rx_queues) { 4383 RTE_ETHDEV_LOG(ERR, "Invalid RX queue_id=%u\n", queue_id); 4384 return 0; 4385 } 4386 #endif 4387 nb_rx = (*dev->rx_pkt_burst)(dev->data->rx_queues[queue_id], 4388 rx_pkts, nb_pkts); 4389 4390 #ifdef RTE_ETHDEV_RXTX_CALLBACKS 4391 if (unlikely(dev->post_rx_burst_cbs[queue_id] != NULL)) { 4392 struct rte_eth_rxtx_callback *cb = 4393 dev->post_rx_burst_cbs[queue_id]; 4394 4395 do { 4396 nb_rx = cb->fn.rx(port_id, queue_id, rx_pkts, nb_rx, 4397 nb_pkts, cb->param); 4398 cb = cb->next; 4399 } while (cb != NULL); 4400 } 4401 #endif 4402 4403 return nb_rx; 4404 } 4405 4406 /** 4407 * Get the number of used descriptors of a rx queue 4408 * 4409 * @param port_id 4410 * The port identifier of the Ethernet device. 4411 * @param queue_id 4412 * The queue id on the specific port. 4413 * @return 4414 * The number of used descriptors in the specific queue, or: 4415 * (-EINVAL) if *port_id* or *queue_id* is invalid 4416 * (-ENOTSUP) if the device does not support this function 4417 */ 4418 static inline int 4419 rte_eth_rx_queue_count(uint16_t port_id, uint16_t queue_id) 4420 { 4421 struct rte_eth_dev *dev; 4422 4423 RTE_ETH_VALID_PORTID_OR_ERR_RET(port_id, -EINVAL); 4424 dev = &rte_eth_devices[port_id]; 4425 RTE_FUNC_PTR_OR_ERR_RET(*dev->dev_ops->rx_queue_count, -ENOTSUP); 4426 if (queue_id >= dev->data->nb_rx_queues) 4427 return -EINVAL; 4428 4429 return (int)(*dev->dev_ops->rx_queue_count)(dev, queue_id); 4430 } 4431 4432 /** 4433 * Check if the DD bit of the specific RX descriptor in the queue has been set 4434 * 4435 * @param port_id 4436 * The port identifier of the Ethernet device. 4437 * @param queue_id 4438 * The queue id on the specific port. 4439 * @param offset 4440 * The offset of the descriptor ID from tail. 4441 * @return 4442 * - (1) if the specific DD bit is set. 4443 * - (0) if the specific DD bit is not set. 4444 * - (-ENODEV) if *port_id* invalid. 4445 * - (-ENOTSUP) if the device does not support this function 4446 */ 4447 static inline int 4448 rte_eth_rx_descriptor_done(uint16_t port_id, uint16_t queue_id, uint16_t offset) 4449 { 4450 struct rte_eth_dev *dev = &rte_eth_devices[port_id]; 4451 RTE_ETH_VALID_PORTID_OR_ERR_RET(port_id, -ENODEV); 4452 RTE_FUNC_PTR_OR_ERR_RET(*dev->dev_ops->rx_descriptor_done, -ENOTSUP); 4453 return (*dev->dev_ops->rx_descriptor_done)( \ 4454 dev->data->rx_queues[queue_id], offset); 4455 } 4456 4457 #define RTE_ETH_RX_DESC_AVAIL 0 /**< Desc available for hw. */ 4458 #define RTE_ETH_RX_DESC_DONE 1 /**< Desc done, filled by hw. */ 4459 #define RTE_ETH_RX_DESC_UNAVAIL 2 /**< Desc used by driver or hw. */ 4460 4461 /** 4462 * Check the status of a Rx descriptor in the queue 4463 * 4464 * It should be called in a similar context than the Rx function: 4465 * - on a dataplane core 4466 * - not concurrently on the same queue 4467 * 4468 * Since it's a dataplane function, no check is performed on port_id and 4469 * queue_id. The caller must therefore ensure that the port is enabled 4470 * and the queue is configured and running. 4471 * 4472 * Note: accessing to a random descriptor in the ring may trigger cache 4473 * misses and have a performance impact. 4474 * 4475 * @param port_id 4476 * A valid port identifier of the Ethernet device which. 4477 * @param queue_id 4478 * A valid Rx queue identifier on this port. 4479 * @param offset 4480 * The offset of the descriptor starting from tail (0 is the next 4481 * packet to be received by the driver). 4482 * 4483 * @return 4484 * - (RTE_ETH_RX_DESC_AVAIL): Descriptor is available for the hardware to 4485 * receive a packet. 4486 * - (RTE_ETH_RX_DESC_DONE): Descriptor is done, it is filled by hw, but 4487 * not yet processed by the driver (i.e. in the receive queue). 4488 * - (RTE_ETH_RX_DESC_UNAVAIL): Descriptor is unavailable, either hold by 4489 * the driver and not yet returned to hw, or reserved by the hw. 4490 * - (-EINVAL) bad descriptor offset. 4491 * - (-ENOTSUP) if the device does not support this function. 4492 * - (-ENODEV) bad port or queue (only if compiled with debug). 4493 */ 4494 static inline int 4495 rte_eth_rx_descriptor_status(uint16_t port_id, uint16_t queue_id, 4496 uint16_t offset) 4497 { 4498 struct rte_eth_dev *dev; 4499 void *rxq; 4500 4501 #ifdef RTE_LIBRTE_ETHDEV_DEBUG 4502 RTE_ETH_VALID_PORTID_OR_ERR_RET(port_id, -ENODEV); 4503 #endif 4504 dev = &rte_eth_devices[port_id]; 4505 #ifdef RTE_LIBRTE_ETHDEV_DEBUG 4506 if (queue_id >= dev->data->nb_rx_queues) 4507 return -ENODEV; 4508 #endif 4509 RTE_FUNC_PTR_OR_ERR_RET(*dev->dev_ops->rx_descriptor_status, -ENOTSUP); 4510 rxq = dev->data->rx_queues[queue_id]; 4511 4512 return (*dev->dev_ops->rx_descriptor_status)(rxq, offset); 4513 } 4514 4515 #define RTE_ETH_TX_DESC_FULL 0 /**< Desc filled for hw, waiting xmit. */ 4516 #define RTE_ETH_TX_DESC_DONE 1 /**< Desc done, packet is transmitted. */ 4517 #define RTE_ETH_TX_DESC_UNAVAIL 2 /**< Desc used by driver or hw. */ 4518 4519 /** 4520 * Check the status of a Tx descriptor in the queue. 4521 * 4522 * It should be called in a similar context than the Tx function: 4523 * - on a dataplane core 4524 * - not concurrently on the same queue 4525 * 4526 * Since it's a dataplane function, no check is performed on port_id and 4527 * queue_id. The caller must therefore ensure that the port is enabled 4528 * and the queue is configured and running. 4529 * 4530 * Note: accessing to a random descriptor in the ring may trigger cache 4531 * misses and have a performance impact. 4532 * 4533 * @param port_id 4534 * A valid port identifier of the Ethernet device which. 4535 * @param queue_id 4536 * A valid Tx queue identifier on this port. 4537 * @param offset 4538 * The offset of the descriptor starting from tail (0 is the place where 4539 * the next packet will be send). 4540 * 4541 * @return 4542 * - (RTE_ETH_TX_DESC_FULL) Descriptor is being processed by the hw, i.e. 4543 * in the transmit queue. 4544 * - (RTE_ETH_TX_DESC_DONE) Hardware is done with this descriptor, it can 4545 * be reused by the driver. 4546 * - (RTE_ETH_TX_DESC_UNAVAIL): Descriptor is unavailable, reserved by the 4547 * driver or the hardware. 4548 * - (-EINVAL) bad descriptor offset. 4549 * - (-ENOTSUP) if the device does not support this function. 4550 * - (-ENODEV) bad port or queue (only if compiled with debug). 4551 */ 4552 static inline int rte_eth_tx_descriptor_status(uint16_t port_id, 4553 uint16_t queue_id, uint16_t offset) 4554 { 4555 struct rte_eth_dev *dev; 4556 void *txq; 4557 4558 #ifdef RTE_LIBRTE_ETHDEV_DEBUG 4559 RTE_ETH_VALID_PORTID_OR_ERR_RET(port_id, -ENODEV); 4560 #endif 4561 dev = &rte_eth_devices[port_id]; 4562 #ifdef RTE_LIBRTE_ETHDEV_DEBUG 4563 if (queue_id >= dev->data->nb_tx_queues) 4564 return -ENODEV; 4565 #endif 4566 RTE_FUNC_PTR_OR_ERR_RET(*dev->dev_ops->tx_descriptor_status, -ENOTSUP); 4567 txq = dev->data->tx_queues[queue_id]; 4568 4569 return (*dev->dev_ops->tx_descriptor_status)(txq, offset); 4570 } 4571 4572 /** 4573 * Send a burst of output packets on a transmit queue of an Ethernet device. 4574 * 4575 * The rte_eth_tx_burst() function is invoked to transmit output packets 4576 * on the output queue *queue_id* of the Ethernet device designated by its 4577 * *port_id*. 4578 * The *nb_pkts* parameter is the number of packets to send which are 4579 * supplied in the *tx_pkts* array of *rte_mbuf* structures, each of them 4580 * allocated from a pool created with rte_pktmbuf_pool_create(). 4581 * The rte_eth_tx_burst() function loops, sending *nb_pkts* packets, 4582 * up to the number of transmit descriptors available in the TX ring of the 4583 * transmit queue. 4584 * For each packet to send, the rte_eth_tx_burst() function performs 4585 * the following operations: 4586 * 4587 * - Pick up the next available descriptor in the transmit ring. 4588 * 4589 * - Free the network buffer previously sent with that descriptor, if any. 4590 * 4591 * - Initialize the transmit descriptor with the information provided 4592 * in the *rte_mbuf data structure. 4593 * 4594 * In the case of a segmented packet composed of a list of *rte_mbuf* buffers, 4595 * the rte_eth_tx_burst() function uses several transmit descriptors 4596 * of the ring. 4597 * 4598 * The rte_eth_tx_burst() function returns the number of packets it 4599 * actually sent. A return value equal to *nb_pkts* means that all packets 4600 * have been sent, and this is likely to signify that other output packets 4601 * could be immediately transmitted again. Applications that implement a 4602 * "send as many packets to transmit as possible" policy can check this 4603 * specific case and keep invoking the rte_eth_tx_burst() function until 4604 * a value less than *nb_pkts* is returned. 4605 * 4606 * It is the responsibility of the rte_eth_tx_burst() function to 4607 * transparently free the memory buffers of packets previously sent. 4608 * This feature is driven by the *tx_free_thresh* value supplied to the 4609 * rte_eth_dev_configure() function at device configuration time. 4610 * When the number of free TX descriptors drops below this threshold, the 4611 * rte_eth_tx_burst() function must [attempt to] free the *rte_mbuf* buffers 4612 * of those packets whose transmission was effectively completed. 4613 * 4614 * If the PMD is DEV_TX_OFFLOAD_MT_LOCKFREE capable, multiple threads can 4615 * invoke this function concurrently on the same tx queue without SW lock. 4616 * @see rte_eth_dev_info_get, struct rte_eth_txconf::offloads 4617 * 4618 * @see rte_eth_tx_prepare to perform some prior checks or adjustments 4619 * for offloads. 4620 * 4621 * @param port_id 4622 * The port identifier of the Ethernet device. 4623 * @param queue_id 4624 * The index of the transmit queue through which output packets must be 4625 * sent. 4626 * The value must be in the range [0, nb_tx_queue - 1] previously supplied 4627 * to rte_eth_dev_configure(). 4628 * @param tx_pkts 4629 * The address of an array of *nb_pkts* pointers to *rte_mbuf* structures 4630 * which contain the output packets. 4631 * @param nb_pkts 4632 * The maximum number of packets to transmit. 4633 * @return 4634 * The number of output packets actually stored in transmit descriptors of 4635 * the transmit ring. The return value can be less than the value of the 4636 * *tx_pkts* parameter when the transmit ring is full or has been filled up. 4637 */ 4638 static inline uint16_t 4639 rte_eth_tx_burst(uint16_t port_id, uint16_t queue_id, 4640 struct rte_mbuf **tx_pkts, uint16_t nb_pkts) 4641 { 4642 struct rte_eth_dev *dev = &rte_eth_devices[port_id]; 4643 4644 #ifdef RTE_LIBRTE_ETHDEV_DEBUG 4645 RTE_ETH_VALID_PORTID_OR_ERR_RET(port_id, 0); 4646 RTE_FUNC_PTR_OR_ERR_RET(*dev->tx_pkt_burst, 0); 4647 4648 if (queue_id >= dev->data->nb_tx_queues) { 4649 RTE_ETHDEV_LOG(ERR, "Invalid TX queue_id=%u\n", queue_id); 4650 return 0; 4651 } 4652 #endif 4653 4654 #ifdef RTE_ETHDEV_RXTX_CALLBACKS 4655 struct rte_eth_rxtx_callback *cb = dev->pre_tx_burst_cbs[queue_id]; 4656 4657 if (unlikely(cb != NULL)) { 4658 do { 4659 nb_pkts = cb->fn.tx(port_id, queue_id, tx_pkts, nb_pkts, 4660 cb->param); 4661 cb = cb->next; 4662 } while (cb != NULL); 4663 } 4664 #endif 4665 4666 return (*dev->tx_pkt_burst)(dev->data->tx_queues[queue_id], tx_pkts, nb_pkts); 4667 } 4668 4669 /** 4670 * Process a burst of output packets on a transmit queue of an Ethernet device. 4671 * 4672 * The rte_eth_tx_prepare() function is invoked to prepare output packets to be 4673 * transmitted on the output queue *queue_id* of the Ethernet device designated 4674 * by its *port_id*. 4675 * The *nb_pkts* parameter is the number of packets to be prepared which are 4676 * supplied in the *tx_pkts* array of *rte_mbuf* structures, each of them 4677 * allocated from a pool created with rte_pktmbuf_pool_create(). 4678 * For each packet to send, the rte_eth_tx_prepare() function performs 4679 * the following operations: 4680 * 4681 * - Check if packet meets devices requirements for tx offloads. 4682 * 4683 * - Check limitations about number of segments. 4684 * 4685 * - Check additional requirements when debug is enabled. 4686 * 4687 * - Update and/or reset required checksums when tx offload is set for packet. 4688 * 4689 * Since this function can modify packet data, provided mbufs must be safely 4690 * writable (e.g. modified data cannot be in shared segment). 4691 * 4692 * The rte_eth_tx_prepare() function returns the number of packets ready to be 4693 * sent. A return value equal to *nb_pkts* means that all packets are valid and 4694 * ready to be sent, otherwise stops processing on the first invalid packet and 4695 * leaves the rest packets untouched. 4696 * 4697 * When this functionality is not implemented in the driver, all packets are 4698 * are returned untouched. 4699 * 4700 * @param port_id 4701 * The port identifier of the Ethernet device. 4702 * The value must be a valid port id. 4703 * @param queue_id 4704 * The index of the transmit queue through which output packets must be 4705 * sent. 4706 * The value must be in the range [0, nb_tx_queue - 1] previously supplied 4707 * to rte_eth_dev_configure(). 4708 * @param tx_pkts 4709 * The address of an array of *nb_pkts* pointers to *rte_mbuf* structures 4710 * which contain the output packets. 4711 * @param nb_pkts 4712 * The maximum number of packets to process. 4713 * @return 4714 * The number of packets correct and ready to be sent. The return value can be 4715 * less than the value of the *tx_pkts* parameter when some packet doesn't 4716 * meet devices requirements with rte_errno set appropriately: 4717 * - EINVAL: offload flags are not correctly set 4718 * - ENOTSUP: the offload feature is not supported by the hardware 4719 * 4720 */ 4721 4722 #ifndef RTE_ETHDEV_TX_PREPARE_NOOP 4723 4724 static inline uint16_t 4725 rte_eth_tx_prepare(uint16_t port_id, uint16_t queue_id, 4726 struct rte_mbuf **tx_pkts, uint16_t nb_pkts) 4727 { 4728 struct rte_eth_dev *dev; 4729 4730 #ifdef RTE_LIBRTE_ETHDEV_DEBUG 4731 if (!rte_eth_dev_is_valid_port(port_id)) { 4732 RTE_ETHDEV_LOG(ERR, "Invalid TX port_id=%u\n", port_id); 4733 rte_errno = EINVAL; 4734 return 0; 4735 } 4736 #endif 4737 4738 dev = &rte_eth_devices[port_id]; 4739 4740 #ifdef RTE_LIBRTE_ETHDEV_DEBUG 4741 if (queue_id >= dev->data->nb_tx_queues) { 4742 RTE_ETHDEV_LOG(ERR, "Invalid TX queue_id=%u\n", queue_id); 4743 rte_errno = EINVAL; 4744 return 0; 4745 } 4746 #endif 4747 4748 if (!dev->tx_pkt_prepare) 4749 return nb_pkts; 4750 4751 return (*dev->tx_pkt_prepare)(dev->data->tx_queues[queue_id], 4752 tx_pkts, nb_pkts); 4753 } 4754 4755 #else 4756 4757 /* 4758 * Native NOOP operation for compilation targets which doesn't require any 4759 * preparations steps, and functional NOOP may introduce unnecessary performance 4760 * drop. 4761 * 4762 * Generally this is not a good idea to turn it on globally and didn't should 4763 * be used if behavior of tx_preparation can change. 4764 */ 4765 4766 static inline uint16_t 4767 rte_eth_tx_prepare(__rte_unused uint16_t port_id, 4768 __rte_unused uint16_t queue_id, 4769 __rte_unused struct rte_mbuf **tx_pkts, uint16_t nb_pkts) 4770 { 4771 return nb_pkts; 4772 } 4773 4774 #endif 4775 4776 /** 4777 * Send any packets queued up for transmission on a port and HW queue 4778 * 4779 * This causes an explicit flush of packets previously buffered via the 4780 * rte_eth_tx_buffer() function. It returns the number of packets successfully 4781 * sent to the NIC, and calls the error callback for any unsent packets. Unless 4782 * explicitly set up otherwise, the default callback simply frees the unsent 4783 * packets back to the owning mempool. 4784 * 4785 * @param port_id 4786 * The port identifier of the Ethernet device. 4787 * @param queue_id 4788 * The index of the transmit queue through which output packets must be 4789 * sent. 4790 * The value must be in the range [0, nb_tx_queue - 1] previously supplied 4791 * to rte_eth_dev_configure(). 4792 * @param buffer 4793 * Buffer of packets to be transmit. 4794 * @return 4795 * The number of packets successfully sent to the Ethernet device. The error 4796 * callback is called for any packets which could not be sent. 4797 */ 4798 static inline uint16_t 4799 rte_eth_tx_buffer_flush(uint16_t port_id, uint16_t queue_id, 4800 struct rte_eth_dev_tx_buffer *buffer) 4801 { 4802 uint16_t sent; 4803 uint16_t to_send = buffer->length; 4804 4805 if (to_send == 0) 4806 return 0; 4807 4808 sent = rte_eth_tx_burst(port_id, queue_id, buffer->pkts, to_send); 4809 4810 buffer->length = 0; 4811 4812 /* All packets sent, or to be dealt with by callback below */ 4813 if (unlikely(sent != to_send)) 4814 buffer->error_callback(&buffer->pkts[sent], 4815 (uint16_t)(to_send - sent), 4816 buffer->error_userdata); 4817 4818 return sent; 4819 } 4820 4821 /** 4822 * Buffer a single packet for future transmission on a port and queue 4823 * 4824 * This function takes a single mbuf/packet and buffers it for later 4825 * transmission on the particular port and queue specified. Once the buffer is 4826 * full of packets, an attempt will be made to transmit all the buffered 4827 * packets. In case of error, where not all packets can be transmitted, a 4828 * callback is called with the unsent packets as a parameter. If no callback 4829 * is explicitly set up, the unsent packets are just freed back to the owning 4830 * mempool. The function returns the number of packets actually sent i.e. 4831 * 0 if no buffer flush occurred, otherwise the number of packets successfully 4832 * flushed 4833 * 4834 * @param port_id 4835 * The port identifier of the Ethernet device. 4836 * @param queue_id 4837 * The index of the transmit queue through which output packets must be 4838 * sent. 4839 * The value must be in the range [0, nb_tx_queue - 1] previously supplied 4840 * to rte_eth_dev_configure(). 4841 * @param buffer 4842 * Buffer used to collect packets to be sent. 4843 * @param tx_pkt 4844 * Pointer to the packet mbuf to be sent. 4845 * @return 4846 * 0 = packet has been buffered for later transmission 4847 * N > 0 = packet has been buffered, and the buffer was subsequently flushed, 4848 * causing N packets to be sent, and the error callback to be called for 4849 * the rest. 4850 */ 4851 static __rte_always_inline uint16_t 4852 rte_eth_tx_buffer(uint16_t port_id, uint16_t queue_id, 4853 struct rte_eth_dev_tx_buffer *buffer, struct rte_mbuf *tx_pkt) 4854 { 4855 buffer->pkts[buffer->length++] = tx_pkt; 4856 if (buffer->length < buffer->size) 4857 return 0; 4858 4859 return rte_eth_tx_buffer_flush(port_id, queue_id, buffer); 4860 } 4861 4862 #ifdef __cplusplus 4863 } 4864 #endif 4865 4866 #endif /* _RTE_ETHDEV_H_ */ 4867