1 /* SPDX-License-Identifier: BSD-3-Clause
2 * Copyright(c) 2010-2017 Intel Corporation
11 * RTE Ethernet Device API
13 * The Ethernet Device API is composed of two parts:
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.
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 ethdev_driver.h instead of this header.
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.
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.
37 * In all functions of the Ethernet API, the Ethernet device is
38 * designated by an integer >= 0 named the device port identifier.
40 * At the Ethernet driver level, Ethernet devices are represented by a generic
41 * data structure of type *rte_eth_dev*.
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.
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
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
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()
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.
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 or the queue is stopped.
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
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
93 * The following configuration may be retained or not
94 * depending on the device capabilities:
97 * - flow-related shared objects, e.g. indirect actions
99 * Any other configuration will not be stored and will need to be re-entered
100 * before a call to rte_eth_dev_start().
102 * Finally, a network application can close an Ethernet device by invoking the
103 * rte_eth_dev_close() function.
105 * Each function of the application Ethernet API invokes a specific function
106 * of the PMD that controls the target device designated by its port
108 * For this purpose, all device-specific functions of an Ethernet driver are
109 * supplied through a set of pointers contained in a generic structure of type
111 * The address of the *eth_dev_ops* structure is stored in the *rte_eth_dev*
112 * structure by the device init function of the Ethernet driver, which is
113 * invoked during the PCI probing phase, as explained earlier.
115 * In other words, each function of the Ethernet API simply retrieves the
116 * *rte_eth_dev* structure associated with the device port identifier and
117 * performs an indirect invocation of the corresponding driver function
118 * supplied in the *eth_dev_ops* structure of the *rte_eth_dev* structure.
120 * For performance reasons, the address of the burst-oriented Rx and Tx
121 * functions of the Ethernet driver are not contained in the *eth_dev_ops*
122 * structure. Instead, they are directly stored at the beginning of the
123 * *rte_eth_dev* structure to avoid an extra indirect memory access during
126 * RTE Ethernet device drivers do not use interrupts for transmitting or
127 * receiving. Instead, Ethernet drivers export Poll-Mode receive and transmit
128 * functions to applications.
129 * Both receive and transmit functions are packet-burst oriented to minimize
130 * their cost per packet through the following optimizations:
132 * - Sharing among multiple packets the incompressible cost of the
133 * invocation of receive/transmit functions.
135 * - Enabling receive/transmit functions to take advantage of burst-oriented
136 * hardware features (L1 cache, prefetch instructions, NIC head/tail
137 * registers) to minimize the number of CPU cycles per packet, for instance,
138 * by avoiding useless read memory accesses to ring descriptors, or by
139 * systematically using arrays of pointers that exactly fit L1 cache line
140 * boundaries and sizes.
142 * The burst-oriented receive function does not provide any error notification,
143 * to avoid the corresponding overhead. As a hint, the upper-level application
144 * might check the status of the device link once being systematically returned
145 * a 0 value by the receive function of the driver for a given number of tries.
154 /* Use this macro to check if LRO API is supported */
155 #define RTE_ETHDEV_HAS_LRO_SUPPORT
157 /* Alias RTE_LIBRTE_ETHDEV_DEBUG for backward compatibility. */
158 #ifdef RTE_LIBRTE_ETHDEV_DEBUG
159 #define RTE_ETHDEV_DEBUG_RX
160 #define RTE_ETHDEV_DEBUG_TX
163 #include <rte_compat.h>
165 #include <rte_interrupts.h>
167 #include <rte_devargs.h>
168 #include <rte_bitops.h>
169 #include <rte_errno.h>
170 #include <rte_common.h>
171 #include <rte_config.h>
172 #include <rte_ether.h>
173 #include <rte_power_intrinsics.h>
175 #include "rte_ethdev_trace_fp.h"
176 #include "rte_dev_info.h"
178 extern int rte_eth_dev_logtype;
180 #define RTE_ETHDEV_LOG(level, ...) \
181 rte_log(RTE_LOG_ ## level, rte_eth_dev_logtype, "" __VA_ARGS__)
186 * Initializes a device iterator.
188 * This iterator allows accessing a list of devices matching some devargs.
191 * Device iterator handle initialized by the function.
192 * The fields bus_str and cls_str might be dynamically allocated,
193 * and could be freed by calling rte_eth_iterator_cleanup().
196 * Device description string.
199 * 0 on successful initialization, negative otherwise.
201 int rte_eth_iterator_init(struct rte_dev_iterator *iter, const char *devargs);
204 * Iterates on devices with devargs filter.
205 * The ownership is not checked.
207 * The next port ID is returned, and the iterator is updated.
210 * Device iterator handle initialized by rte_eth_iterator_init().
211 * Some fields bus_str and cls_str might be freed when no more port is found,
212 * by calling rte_eth_iterator_cleanup().
215 * A port ID if found, RTE_MAX_ETHPORTS otherwise.
217 uint16_t rte_eth_iterator_next(struct rte_dev_iterator *iter);
220 * Free some allocated fields of the iterator.
222 * This function is automatically called by rte_eth_iterator_next()
223 * on the last iteration (i.e. when no more matching port is found).
225 * It is safe to call this function twice; it will do nothing more.
228 * Device iterator handle initialized by rte_eth_iterator_init().
229 * The fields bus_str and cls_str are freed if needed.
231 void rte_eth_iterator_cleanup(struct rte_dev_iterator *iter);
234 * Macro to iterate over all ethdev ports matching some devargs.
236 * If a break is done before the end of the loop,
237 * the function rte_eth_iterator_cleanup() must be called.
240 * Iterated port ID of type uint16_t.
242 * Device parameters input as string of type char*.
244 * Iterator handle of type struct rte_dev_iterator, used internally.
246 #define RTE_ETH_FOREACH_MATCHING_DEV(id, devargs, iter) \
247 for (rte_eth_iterator_init(iter, devargs), \
248 id = rte_eth_iterator_next(iter); \
249 id != RTE_MAX_ETHPORTS; \
250 id = rte_eth_iterator_next(iter))
253 * A structure used to retrieve statistics for an Ethernet port.
254 * Not all statistics fields in struct rte_eth_stats are supported
255 * by any type of network interface card (NIC). If any statistics
256 * field is not supported, its value is 0.
257 * All byte-related statistics do not include Ethernet FCS regardless
258 * of whether these bytes have been delivered to the application
259 * (see RTE_ETH_RX_OFFLOAD_KEEP_CRC).
261 struct rte_eth_stats {
262 uint64_t ipackets; /**< Total number of successfully received packets. */
263 uint64_t opackets; /**< Total number of successfully transmitted packets.*/
264 uint64_t ibytes; /**< Total number of successfully received bytes. */
265 uint64_t obytes; /**< Total number of successfully transmitted bytes. */
267 * Total of Rx packets dropped by the HW,
268 * because there are no available buffer (i.e. Rx queues are full).
271 uint64_t ierrors; /**< Total number of erroneous received packets. */
272 uint64_t oerrors; /**< Total number of failed transmitted packets. */
273 uint64_t rx_nombuf; /**< Total number of Rx mbuf allocation failures. */
274 /* Queue stats are limited to max 256 queues */
275 /** Total number of queue Rx packets. */
276 uint64_t q_ipackets[RTE_ETHDEV_QUEUE_STAT_CNTRS];
277 /** Total number of queue Tx packets. */
278 uint64_t q_opackets[RTE_ETHDEV_QUEUE_STAT_CNTRS];
279 /** Total number of successfully received queue bytes. */
280 uint64_t q_ibytes[RTE_ETHDEV_QUEUE_STAT_CNTRS];
281 /** Total number of successfully transmitted queue bytes. */
282 uint64_t q_obytes[RTE_ETHDEV_QUEUE_STAT_CNTRS];
283 /** Total number of queue packets received that are dropped. */
284 uint64_t q_errors[RTE_ETHDEV_QUEUE_STAT_CNTRS];
287 /**@{@name Link speed capabilities
288 * Device supported speeds bitmap flags
290 #define RTE_ETH_LINK_SPEED_AUTONEG 0 /**< Autonegotiate (all speeds) */
291 #define RTE_ETH_LINK_SPEED_FIXED RTE_BIT32(0) /**< Disable autoneg (fixed speed) */
292 #define RTE_ETH_LINK_SPEED_10M_HD RTE_BIT32(1) /**< 10 Mbps half-duplex */
293 #define RTE_ETH_LINK_SPEED_10M RTE_BIT32(2) /**< 10 Mbps full-duplex */
294 #define RTE_ETH_LINK_SPEED_100M_HD RTE_BIT32(3) /**< 100 Mbps half-duplex */
295 #define RTE_ETH_LINK_SPEED_100M RTE_BIT32(4) /**< 100 Mbps full-duplex */
296 #define RTE_ETH_LINK_SPEED_1G RTE_BIT32(5) /**< 1 Gbps */
297 #define RTE_ETH_LINK_SPEED_2_5G RTE_BIT32(6) /**< 2.5 Gbps */
298 #define RTE_ETH_LINK_SPEED_5G RTE_BIT32(7) /**< 5 Gbps */
299 #define RTE_ETH_LINK_SPEED_10G RTE_BIT32(8) /**< 10 Gbps */
300 #define RTE_ETH_LINK_SPEED_20G RTE_BIT32(9) /**< 20 Gbps */
301 #define RTE_ETH_LINK_SPEED_25G RTE_BIT32(10) /**< 25 Gbps */
302 #define RTE_ETH_LINK_SPEED_40G RTE_BIT32(11) /**< 40 Gbps */
303 #define RTE_ETH_LINK_SPEED_50G RTE_BIT32(12) /**< 50 Gbps */
304 #define RTE_ETH_LINK_SPEED_56G RTE_BIT32(13) /**< 56 Gbps */
305 #define RTE_ETH_LINK_SPEED_100G RTE_BIT32(14) /**< 100 Gbps */
306 #define RTE_ETH_LINK_SPEED_200G RTE_BIT32(15) /**< 200 Gbps */
309 #define ETH_LINK_SPEED_AUTONEG RTE_DEPRECATED(ETH_LINK_SPEED_AUTONEG) RTE_ETH_LINK_SPEED_AUTONEG
310 #define ETH_LINK_SPEED_FIXED RTE_DEPRECATED(ETH_LINK_SPEED_FIXED) RTE_ETH_LINK_SPEED_FIXED
311 #define ETH_LINK_SPEED_10M_HD RTE_DEPRECATED(ETH_LINK_SPEED_10M_HD) RTE_ETH_LINK_SPEED_10M_HD
312 #define ETH_LINK_SPEED_10M RTE_DEPRECATED(ETH_LINK_SPEED_10M) RTE_ETH_LINK_SPEED_10M
313 #define ETH_LINK_SPEED_100M_HD RTE_DEPRECATED(ETH_LINK_SPEED_100M_HD) RTE_ETH_LINK_SPEED_100M_HD
314 #define ETH_LINK_SPEED_100M RTE_DEPRECATED(ETH_LINK_SPEED_100M) RTE_ETH_LINK_SPEED_100M
315 #define ETH_LINK_SPEED_1G RTE_DEPRECATED(ETH_LINK_SPEED_1G) RTE_ETH_LINK_SPEED_1G
316 #define ETH_LINK_SPEED_2_5G RTE_DEPRECATED(ETH_LINK_SPEED_2_5G) RTE_ETH_LINK_SPEED_2_5G
317 #define ETH_LINK_SPEED_5G RTE_DEPRECATED(ETH_LINK_SPEED_5G) RTE_ETH_LINK_SPEED_5G
318 #define ETH_LINK_SPEED_10G RTE_DEPRECATED(ETH_LINK_SPEED_10G) RTE_ETH_LINK_SPEED_10G
319 #define ETH_LINK_SPEED_20G RTE_DEPRECATED(ETH_LINK_SPEED_20G) RTE_ETH_LINK_SPEED_20G
320 #define ETH_LINK_SPEED_25G RTE_DEPRECATED(ETH_LINK_SPEED_25G) RTE_ETH_LINK_SPEED_25G
321 #define ETH_LINK_SPEED_40G RTE_DEPRECATED(ETH_LINK_SPEED_40G) RTE_ETH_LINK_SPEED_40G
322 #define ETH_LINK_SPEED_50G RTE_DEPRECATED(ETH_LINK_SPEED_50G) RTE_ETH_LINK_SPEED_50G
323 #define ETH_LINK_SPEED_56G RTE_DEPRECATED(ETH_LINK_SPEED_56G) RTE_ETH_LINK_SPEED_56G
324 #define ETH_LINK_SPEED_100G RTE_DEPRECATED(ETH_LINK_SPEED_100G) RTE_ETH_LINK_SPEED_100G
325 #define ETH_LINK_SPEED_200G RTE_DEPRECATED(ETH_LINK_SPEED_200G) RTE_ETH_LINK_SPEED_200G
327 /**@{@name Link speed
328 * Ethernet numeric link speeds in Mbps
330 #define RTE_ETH_SPEED_NUM_NONE 0 /**< Not defined */
331 #define RTE_ETH_SPEED_NUM_10M 10 /**< 10 Mbps */
332 #define RTE_ETH_SPEED_NUM_100M 100 /**< 100 Mbps */
333 #define RTE_ETH_SPEED_NUM_1G 1000 /**< 1 Gbps */
334 #define RTE_ETH_SPEED_NUM_2_5G 2500 /**< 2.5 Gbps */
335 #define RTE_ETH_SPEED_NUM_5G 5000 /**< 5 Gbps */
336 #define RTE_ETH_SPEED_NUM_10G 10000 /**< 10 Gbps */
337 #define RTE_ETH_SPEED_NUM_20G 20000 /**< 20 Gbps */
338 #define RTE_ETH_SPEED_NUM_25G 25000 /**< 25 Gbps */
339 #define RTE_ETH_SPEED_NUM_40G 40000 /**< 40 Gbps */
340 #define RTE_ETH_SPEED_NUM_50G 50000 /**< 50 Gbps */
341 #define RTE_ETH_SPEED_NUM_56G 56000 /**< 56 Gbps */
342 #define RTE_ETH_SPEED_NUM_100G 100000 /**< 100 Gbps */
343 #define RTE_ETH_SPEED_NUM_200G 200000 /**< 200 Gbps */
344 #define RTE_ETH_SPEED_NUM_UNKNOWN UINT32_MAX /**< Unknown */
347 #define ETH_SPEED_NUM_NONE RTE_DEPRECATED(ETH_SPEED_NUM_NONE) RTE_ETH_SPEED_NUM_NONE
348 #define ETH_SPEED_NUM_10M RTE_DEPRECATED(ETH_SPEED_NUM_10M) RTE_ETH_SPEED_NUM_10M
349 #define ETH_SPEED_NUM_100M RTE_DEPRECATED(ETH_SPEED_NUM_100M) RTE_ETH_SPEED_NUM_100M
350 #define ETH_SPEED_NUM_1G RTE_DEPRECATED(ETH_SPEED_NUM_1G) RTE_ETH_SPEED_NUM_1G
351 #define ETH_SPEED_NUM_2_5G RTE_DEPRECATED(ETH_SPEED_NUM_2_5G) RTE_ETH_SPEED_NUM_2_5G
352 #define ETH_SPEED_NUM_5G RTE_DEPRECATED(ETH_SPEED_NUM_5G) RTE_ETH_SPEED_NUM_5G
353 #define ETH_SPEED_NUM_10G RTE_DEPRECATED(ETH_SPEED_NUM_10G) RTE_ETH_SPEED_NUM_10G
354 #define ETH_SPEED_NUM_20G RTE_DEPRECATED(ETH_SPEED_NUM_20G) RTE_ETH_SPEED_NUM_20G
355 #define ETH_SPEED_NUM_25G RTE_DEPRECATED(ETH_SPEED_NUM_25G) RTE_ETH_SPEED_NUM_25G
356 #define ETH_SPEED_NUM_40G RTE_DEPRECATED(ETH_SPEED_NUM_40G) RTE_ETH_SPEED_NUM_40G
357 #define ETH_SPEED_NUM_50G RTE_DEPRECATED(ETH_SPEED_NUM_50G) RTE_ETH_SPEED_NUM_50G
358 #define ETH_SPEED_NUM_56G RTE_DEPRECATED(ETH_SPEED_NUM_56G) RTE_ETH_SPEED_NUM_56G
359 #define ETH_SPEED_NUM_100G RTE_DEPRECATED(ETH_SPEED_NUM_100G) RTE_ETH_SPEED_NUM_100G
360 #define ETH_SPEED_NUM_200G RTE_DEPRECATED(ETH_SPEED_NUM_200G) RTE_ETH_SPEED_NUM_200G
361 #define ETH_SPEED_NUM_UNKNOWN RTE_DEPRECATED(ETH_SPEED_NUM_UNKNOWN) RTE_ETH_SPEED_NUM_UNKNOWN
364 * A structure used to retrieve link-level information of an Ethernet port.
367 struct rte_eth_link {
368 uint32_t link_speed; /**< RTE_ETH_SPEED_NUM_ */
369 uint16_t link_duplex : 1; /**< RTE_ETH_LINK_[HALF/FULL]_DUPLEX */
370 uint16_t link_autoneg : 1; /**< RTE_ETH_LINK_[AUTONEG/FIXED] */
371 uint16_t link_status : 1; /**< RTE_ETH_LINK_[DOWN/UP] */
372 } __rte_aligned(8); /**< aligned for atomic64 read/write */
374 /**@{@name Link negotiation
375 * Constants used in link management.
377 #define RTE_ETH_LINK_HALF_DUPLEX 0 /**< Half-duplex connection (see link_duplex). */
378 #define RTE_ETH_LINK_FULL_DUPLEX 1 /**< Full-duplex connection (see link_duplex). */
379 #define RTE_ETH_LINK_DOWN 0 /**< Link is down (see link_status). */
380 #define RTE_ETH_LINK_UP 1 /**< Link is up (see link_status). */
381 #define RTE_ETH_LINK_FIXED 0 /**< No autonegotiation (see link_autoneg). */
382 #define RTE_ETH_LINK_AUTONEG 1 /**< Autonegotiated (see link_autoneg). */
383 #define RTE_ETH_LINK_MAX_STR_LEN 40 /**< Max length of default link string. */
386 #define ETH_LINK_HALF_DUPLEX RTE_DEPRECATED(ETH_LINK_HALF_DUPLEX) RTE_ETH_LINK_HALF_DUPLEX
387 #define ETH_LINK_FULL_DUPLEX RTE_DEPRECATED(ETH_LINK_FULL_DUPLEX) RTE_ETH_LINK_FULL_DUPLEX
388 #define ETH_LINK_DOWN RTE_DEPRECATED(ETH_LINK_DOWN) RTE_ETH_LINK_DOWN
389 #define ETH_LINK_UP RTE_DEPRECATED(ETH_LINK_UP) RTE_ETH_LINK_UP
390 #define ETH_LINK_FIXED RTE_DEPRECATED(ETH_LINK_FIXED) RTE_ETH_LINK_FIXED
391 #define ETH_LINK_AUTONEG RTE_DEPRECATED(ETH_LINK_AUTONEG) RTE_ETH_LINK_AUTONEG
394 * A structure used to configure the ring threshold registers of an Rx/Tx
395 * queue for an Ethernet port.
397 struct rte_eth_thresh {
398 uint8_t pthresh; /**< Ring prefetch threshold. */
399 uint8_t hthresh; /**< Ring host threshold. */
400 uint8_t wthresh; /**< Ring writeback threshold. */
403 /**@{@name Multi-queue mode
404 * @see rte_eth_conf.rxmode.mq_mode.
406 #define RTE_ETH_MQ_RX_RSS_FLAG RTE_BIT32(0) /**< Enable RSS. @see rte_eth_rss_conf */
407 #define RTE_ETH_MQ_RX_DCB_FLAG RTE_BIT32(1) /**< Enable DCB. */
408 #define RTE_ETH_MQ_RX_VMDQ_FLAG RTE_BIT32(2) /**< Enable VMDq. */
411 #define ETH_MQ_RX_RSS_FLAG RTE_DEPRECATED(ETH_MQ_RX_RSS_FLAG) RTE_ETH_MQ_RX_RSS_FLAG
412 #define ETH_MQ_RX_DCB_FLAG RTE_DEPRECATED(ETH_MQ_RX_DCB_FLAG) RTE_ETH_MQ_RX_DCB_FLAG
413 #define ETH_MQ_RX_VMDQ_FLAG RTE_DEPRECATED(ETH_MQ_RX_VMDQ_FLAG) RTE_ETH_MQ_RX_VMDQ_FLAG
416 * A set of values to identify what method is to be used to route
417 * packets to multiple queues.
419 enum rte_eth_rx_mq_mode {
420 /** None of DCB, RSS or VMDq mode */
421 RTE_ETH_MQ_RX_NONE = 0,
423 /** For Rx side, only RSS is on */
424 RTE_ETH_MQ_RX_RSS = RTE_ETH_MQ_RX_RSS_FLAG,
425 /** For Rx side,only DCB is on. */
426 RTE_ETH_MQ_RX_DCB = RTE_ETH_MQ_RX_DCB_FLAG,
427 /** Both DCB and RSS enable */
428 RTE_ETH_MQ_RX_DCB_RSS = RTE_ETH_MQ_RX_RSS_FLAG | RTE_ETH_MQ_RX_DCB_FLAG,
430 /** Only VMDq, no RSS nor DCB */
431 RTE_ETH_MQ_RX_VMDQ_ONLY = RTE_ETH_MQ_RX_VMDQ_FLAG,
432 /** RSS mode with VMDq */
433 RTE_ETH_MQ_RX_VMDQ_RSS = RTE_ETH_MQ_RX_RSS_FLAG | RTE_ETH_MQ_RX_VMDQ_FLAG,
434 /** Use VMDq+DCB to route traffic to queues */
435 RTE_ETH_MQ_RX_VMDQ_DCB = RTE_ETH_MQ_RX_VMDQ_FLAG | RTE_ETH_MQ_RX_DCB_FLAG,
436 /** Enable both VMDq and DCB in VMDq */
437 RTE_ETH_MQ_RX_VMDQ_DCB_RSS = RTE_ETH_MQ_RX_RSS_FLAG | RTE_ETH_MQ_RX_DCB_FLAG |
438 RTE_ETH_MQ_RX_VMDQ_FLAG,
441 #define ETH_MQ_RX_NONE RTE_DEPRECATED(ETH_MQ_RX_NONE) RTE_ETH_MQ_RX_NONE
442 #define ETH_MQ_RX_RSS RTE_DEPRECATED(ETH_MQ_RX_RSS) RTE_ETH_MQ_RX_RSS
443 #define ETH_MQ_RX_DCB RTE_DEPRECATED(ETH_MQ_RX_DCB) RTE_ETH_MQ_RX_DCB
444 #define ETH_MQ_RX_DCB_RSS RTE_DEPRECATED(ETH_MQ_RX_DCB_RSS) RTE_ETH_MQ_RX_DCB_RSS
445 #define ETH_MQ_RX_VMDQ_ONLY RTE_DEPRECATED(ETH_MQ_RX_VMDQ_ONLY) RTE_ETH_MQ_RX_VMDQ_ONLY
446 #define ETH_MQ_RX_VMDQ_RSS RTE_DEPRECATED(ETH_MQ_RX_VMDQ_RSS) RTE_ETH_MQ_RX_VMDQ_RSS
447 #define ETH_MQ_RX_VMDQ_DCB RTE_DEPRECATED(ETH_MQ_RX_VMDQ_DCB) RTE_ETH_MQ_RX_VMDQ_DCB
448 #define ETH_MQ_RX_VMDQ_DCB_RSS RTE_DEPRECATED(ETH_MQ_RX_VMDQ_DCB_RSS) RTE_ETH_MQ_RX_VMDQ_DCB_RSS
451 * A set of values to identify what method is to be used to transmit
452 * packets using multi-TCs.
454 enum rte_eth_tx_mq_mode {
455 RTE_ETH_MQ_TX_NONE = 0, /**< It is in neither DCB nor VT mode. */
456 RTE_ETH_MQ_TX_DCB, /**< For Tx side,only DCB is on. */
457 RTE_ETH_MQ_TX_VMDQ_DCB, /**< For Tx side,both DCB and VT is on. */
458 RTE_ETH_MQ_TX_VMDQ_ONLY, /**< Only VT on, no DCB */
461 #define ETH_MQ_TX_NONE RTE_DEPRECATED(ETH_MQ_TX_NONE) RTE_ETH_MQ_TX_NONE
462 #define ETH_MQ_TX_DCB RTE_DEPRECATED(ETH_MQ_TX_DCB) RTE_ETH_MQ_TX_DCB
463 #define ETH_MQ_TX_VMDQ_DCB RTE_DEPRECATED(ETH_MQ_TX_VMDQ_DCB) RTE_ETH_MQ_TX_VMDQ_DCB
464 #define ETH_MQ_TX_VMDQ_ONLY RTE_DEPRECATED(ETH_MQ_TX_VMDQ_ONLY) RTE_ETH_MQ_TX_VMDQ_ONLY
467 * A structure used to configure the Rx features of an Ethernet port.
469 struct rte_eth_rxmode {
470 /** The multi-queue packet distribution mode to be used, e.g. RSS. */
471 enum rte_eth_rx_mq_mode mq_mode;
472 uint32_t mtu; /**< Requested MTU. */
473 /** Maximum allowed size of LRO aggregated packet. */
474 uint32_t max_lro_pkt_size;
475 uint16_t split_hdr_size; /**< hdr buf size (header_split enabled).*/
477 * Per-port Rx offloads to be set using RTE_ETH_RX_OFFLOAD_* flags.
478 * Only offloads set on rx_offload_capa field on rte_eth_dev_info
479 * structure are allowed to be set.
483 uint64_t reserved_64s[2]; /**< Reserved for future fields */
484 void *reserved_ptrs[2]; /**< Reserved for future fields */
488 * VLAN types to indicate if it is for single VLAN, inner VLAN or outer VLAN.
489 * Note that single VLAN is treated the same as inner VLAN.
492 RTE_ETH_VLAN_TYPE_UNKNOWN = 0,
493 RTE_ETH_VLAN_TYPE_INNER, /**< Inner VLAN. */
494 RTE_ETH_VLAN_TYPE_OUTER, /**< Single VLAN, or outer VLAN. */
495 RTE_ETH_VLAN_TYPE_MAX,
498 #define ETH_VLAN_TYPE_UNKNOWN RTE_DEPRECATED(ETH_VLAN_TYPE_UNKNOWN) RTE_ETH_VLAN_TYPE_UNKNOWN
499 #define ETH_VLAN_TYPE_INNER RTE_DEPRECATED(ETH_VLAN_TYPE_INNER) RTE_ETH_VLAN_TYPE_INNER
500 #define ETH_VLAN_TYPE_OUTER RTE_DEPRECATED(ETH_VLAN_TYPE_OUTER) RTE_ETH_VLAN_TYPE_OUTER
501 #define ETH_VLAN_TYPE_MAX RTE_DEPRECATED(ETH_VLAN_TYPE_MAX) RTE_ETH_VLAN_TYPE_MAX
504 * A structure used to describe a VLAN filter.
505 * If the bit corresponding to a VID is set, such VID is on.
507 struct rte_vlan_filter_conf {
512 * A structure used to configure the Receive Side Scaling (RSS) feature
513 * of an Ethernet port.
514 * If not NULL, the *rss_key* pointer of the *rss_conf* structure points
515 * to an array holding the RSS key to use for hashing specific header
516 * fields of received packets. The length of this array should be indicated
517 * by *rss_key_len* below. Otherwise, a default random hash key is used by
520 * The *rss_key_len* field of the *rss_conf* structure indicates the length
521 * in bytes of the array pointed by *rss_key*. To be compatible, this length
522 * will be checked in i40e only. Others assume 40 bytes to be used as before.
524 * The *rss_hf* field of the *rss_conf* structure indicates the different
525 * types of IPv4/IPv6 packets to which the RSS hashing must be applied.
526 * Supplying an *rss_hf* equal to zero disables the RSS feature.
528 struct rte_eth_rss_conf {
529 uint8_t *rss_key; /**< If not NULL, 40-byte hash key. */
530 uint8_t rss_key_len; /**< hash key length in bytes. */
531 uint64_t rss_hf; /**< Hash functions to apply - see below. */
535 * A packet can be identified by hardware as different flow types. Different
536 * NIC hardware may support different flow types.
537 * Basically, the NIC hardware identifies the flow type as deep protocol as
538 * possible, and exclusively. For example, if a packet is identified as
539 * 'RTE_ETH_FLOW_NONFRAG_IPV4_TCP', it will not be any of other flow types,
540 * though it is an actual IPV4 packet.
542 #define RTE_ETH_FLOW_UNKNOWN 0
543 #define RTE_ETH_FLOW_RAW 1
544 #define RTE_ETH_FLOW_IPV4 2
545 #define RTE_ETH_FLOW_FRAG_IPV4 3
546 #define RTE_ETH_FLOW_NONFRAG_IPV4_TCP 4
547 #define RTE_ETH_FLOW_NONFRAG_IPV4_UDP 5
548 #define RTE_ETH_FLOW_NONFRAG_IPV4_SCTP 6
549 #define RTE_ETH_FLOW_NONFRAG_IPV4_OTHER 7
550 #define RTE_ETH_FLOW_IPV6 8
551 #define RTE_ETH_FLOW_FRAG_IPV6 9
552 #define RTE_ETH_FLOW_NONFRAG_IPV6_TCP 10
553 #define RTE_ETH_FLOW_NONFRAG_IPV6_UDP 11
554 #define RTE_ETH_FLOW_NONFRAG_IPV6_SCTP 12
555 #define RTE_ETH_FLOW_NONFRAG_IPV6_OTHER 13
556 #define RTE_ETH_FLOW_L2_PAYLOAD 14
557 #define RTE_ETH_FLOW_IPV6_EX 15
558 #define RTE_ETH_FLOW_IPV6_TCP_EX 16
559 #define RTE_ETH_FLOW_IPV6_UDP_EX 17
560 /** Consider device port number as a flow differentiator */
561 #define RTE_ETH_FLOW_PORT 18
562 #define RTE_ETH_FLOW_VXLAN 19 /**< VXLAN protocol based flow */
563 #define RTE_ETH_FLOW_GENEVE 20 /**< GENEVE protocol based flow */
564 #define RTE_ETH_FLOW_NVGRE 21 /**< NVGRE protocol based flow */
565 #define RTE_ETH_FLOW_VXLAN_GPE 22 /**< VXLAN-GPE protocol based flow */
566 #define RTE_ETH_FLOW_GTPU 23 /**< GTPU protocol based flow */
567 #define RTE_ETH_FLOW_MAX 24
570 * Below macros are defined for RSS offload types, they can be used to
571 * fill rte_eth_rss_conf.rss_hf or rte_flow_action_rss.types.
573 #define RTE_ETH_RSS_IPV4 RTE_BIT64(2)
574 #define RTE_ETH_RSS_FRAG_IPV4 RTE_BIT64(3)
575 #define RTE_ETH_RSS_NONFRAG_IPV4_TCP RTE_BIT64(4)
576 #define RTE_ETH_RSS_NONFRAG_IPV4_UDP RTE_BIT64(5)
577 #define RTE_ETH_RSS_NONFRAG_IPV4_SCTP RTE_BIT64(6)
578 #define RTE_ETH_RSS_NONFRAG_IPV4_OTHER RTE_BIT64(7)
579 #define RTE_ETH_RSS_IPV6 RTE_BIT64(8)
580 #define RTE_ETH_RSS_FRAG_IPV6 RTE_BIT64(9)
581 #define RTE_ETH_RSS_NONFRAG_IPV6_TCP RTE_BIT64(10)
582 #define RTE_ETH_RSS_NONFRAG_IPV6_UDP RTE_BIT64(11)
583 #define RTE_ETH_RSS_NONFRAG_IPV6_SCTP RTE_BIT64(12)
584 #define RTE_ETH_RSS_NONFRAG_IPV6_OTHER RTE_BIT64(13)
585 #define RTE_ETH_RSS_L2_PAYLOAD RTE_BIT64(14)
586 #define RTE_ETH_RSS_IPV6_EX RTE_BIT64(15)
587 #define RTE_ETH_RSS_IPV6_TCP_EX RTE_BIT64(16)
588 #define RTE_ETH_RSS_IPV6_UDP_EX RTE_BIT64(17)
589 #define RTE_ETH_RSS_PORT RTE_BIT64(18)
590 #define RTE_ETH_RSS_VXLAN RTE_BIT64(19)
591 #define RTE_ETH_RSS_GENEVE RTE_BIT64(20)
592 #define RTE_ETH_RSS_NVGRE RTE_BIT64(21)
593 #define RTE_ETH_RSS_GTPU RTE_BIT64(23)
594 #define RTE_ETH_RSS_ETH RTE_BIT64(24)
595 #define RTE_ETH_RSS_S_VLAN RTE_BIT64(25)
596 #define RTE_ETH_RSS_C_VLAN RTE_BIT64(26)
597 #define RTE_ETH_RSS_ESP RTE_BIT64(27)
598 #define RTE_ETH_RSS_AH RTE_BIT64(28)
599 #define RTE_ETH_RSS_L2TPV3 RTE_BIT64(29)
600 #define RTE_ETH_RSS_PFCP RTE_BIT64(30)
601 #define RTE_ETH_RSS_PPPOE RTE_BIT64(31)
602 #define RTE_ETH_RSS_ECPRI RTE_BIT64(32)
603 #define RTE_ETH_RSS_MPLS RTE_BIT64(33)
604 #define RTE_ETH_RSS_IPV4_CHKSUM RTE_BIT64(34)
606 #define ETH_RSS_IPV4 RTE_DEPRECATED(ETH_RSS_IPV4) RTE_ETH_RSS_IPV4
607 #define ETH_RSS_FRAG_IPV4 RTE_DEPRECATED(ETH_RSS_FRAG_IPV4) RTE_ETH_RSS_FRAG_IPV4
608 #define ETH_RSS_NONFRAG_IPV4_TCP RTE_DEPRECATED(ETH_RSS_NONFRAG_IPV4_TCP) RTE_ETH_RSS_NONFRAG_IPV4_TCP
609 #define ETH_RSS_NONFRAG_IPV4_UDP RTE_DEPRECATED(ETH_RSS_NONFRAG_IPV4_UDP) RTE_ETH_RSS_NONFRAG_IPV4_UDP
610 #define ETH_RSS_NONFRAG_IPV4_SCTP RTE_DEPRECATED(ETH_RSS_NONFRAG_IPV4_SCTP) RTE_ETH_RSS_NONFRAG_IPV4_SCTP
611 #define ETH_RSS_NONFRAG_IPV4_OTHER RTE_DEPRECATED(ETH_RSS_NONFRAG_IPV4_OTHER) RTE_ETH_RSS_NONFRAG_IPV4_OTHER
612 #define ETH_RSS_IPV6 RTE_DEPRECATED(ETH_RSS_IPV6) RTE_ETH_RSS_IPV6
613 #define ETH_RSS_FRAG_IPV6 RTE_DEPRECATED(ETH_RSS_FRAG_IPV6) RTE_ETH_RSS_FRAG_IPV6
614 #define ETH_RSS_NONFRAG_IPV6_TCP RTE_DEPRECATED(ETH_RSS_NONFRAG_IPV6_TCP) RTE_ETH_RSS_NONFRAG_IPV6_TCP
615 #define ETH_RSS_NONFRAG_IPV6_UDP RTE_DEPRECATED(ETH_RSS_NONFRAG_IPV6_UDP) RTE_ETH_RSS_NONFRAG_IPV6_UDP
616 #define ETH_RSS_NONFRAG_IPV6_SCTP RTE_DEPRECATED(ETH_RSS_NONFRAG_IPV6_SCTP) RTE_ETH_RSS_NONFRAG_IPV6_SCTP
617 #define ETH_RSS_NONFRAG_IPV6_OTHER RTE_DEPRECATED(ETH_RSS_NONFRAG_IPV6_OTHER) RTE_ETH_RSS_NONFRAG_IPV6_OTHER
618 #define ETH_RSS_L2_PAYLOAD RTE_DEPRECATED(ETH_RSS_L2_PAYLOAD) RTE_ETH_RSS_L2_PAYLOAD
619 #define ETH_RSS_IPV6_EX RTE_DEPRECATED(ETH_RSS_IPV6_EX) RTE_ETH_RSS_IPV6_EX
620 #define ETH_RSS_IPV6_TCP_EX RTE_DEPRECATED(ETH_RSS_IPV6_TCP_EX) RTE_ETH_RSS_IPV6_TCP_EX
621 #define ETH_RSS_IPV6_UDP_EX RTE_DEPRECATED(ETH_RSS_IPV6_UDP_EX) RTE_ETH_RSS_IPV6_UDP_EX
622 #define ETH_RSS_PORT RTE_DEPRECATED(ETH_RSS_PORT) RTE_ETH_RSS_PORT
623 #define ETH_RSS_VXLAN RTE_DEPRECATED(ETH_RSS_VXLAN) RTE_ETH_RSS_VXLAN
624 #define ETH_RSS_GENEVE RTE_DEPRECATED(ETH_RSS_GENEVE) RTE_ETH_RSS_GENEVE
625 #define ETH_RSS_NVGRE RTE_DEPRECATED(ETH_RSS_NVGRE) RTE_ETH_RSS_NVGRE
626 #define ETH_RSS_GTPU RTE_DEPRECATED(ETH_RSS_GTPU) RTE_ETH_RSS_GTPU
627 #define ETH_RSS_ETH RTE_DEPRECATED(ETH_RSS_ETH) RTE_ETH_RSS_ETH
628 #define ETH_RSS_S_VLAN RTE_DEPRECATED(ETH_RSS_S_VLAN) RTE_ETH_RSS_S_VLAN
629 #define ETH_RSS_C_VLAN RTE_DEPRECATED(ETH_RSS_C_VLAN) RTE_ETH_RSS_C_VLAN
630 #define ETH_RSS_ESP RTE_DEPRECATED(ETH_RSS_ESP) RTE_ETH_RSS_ESP
631 #define ETH_RSS_AH RTE_DEPRECATED(ETH_RSS_AH) RTE_ETH_RSS_AH
632 #define ETH_RSS_L2TPV3 RTE_DEPRECATED(ETH_RSS_L2TPV3) RTE_ETH_RSS_L2TPV3
633 #define ETH_RSS_PFCP RTE_DEPRECATED(ETH_RSS_PFCP) RTE_ETH_RSS_PFCP
634 #define ETH_RSS_PPPOE RTE_DEPRECATED(ETH_RSS_PPPOE) RTE_ETH_RSS_PPPOE
635 #define ETH_RSS_ECPRI RTE_DEPRECATED(ETH_RSS_ECPRI) RTE_ETH_RSS_ECPRI
636 #define ETH_RSS_MPLS RTE_DEPRECATED(ETH_RSS_MPLS) RTE_ETH_RSS_MPLS
637 #define ETH_RSS_IPV4_CHKSUM RTE_DEPRECATED(ETH_RSS_IPV4_CHKSUM) RTE_ETH_RSS_IPV4_CHKSUM
640 * The ETH_RSS_L4_CHKSUM works on checksum field of any L4 header.
641 * It is similar to ETH_RSS_PORT that they don't specify the specific type of
642 * L4 header. This macro is defined to replace some specific L4 (TCP/UDP/SCTP)
643 * checksum type for constructing the use of RSS offload bits.
645 * Due to above reason, some old APIs (and configuration) don't support
646 * RTE_ETH_RSS_L4_CHKSUM. The rte_flow RSS API supports it.
648 * For the case that checksum is not used in an UDP header,
649 * it takes the reserved value 0 as input for the hash function.
651 #define RTE_ETH_RSS_L4_CHKSUM RTE_BIT64(35)
652 #define ETH_RSS_L4_CHKSUM RTE_DEPRECATED(ETH_RSS_L4_CHKSUM) RTE_ETH_RSS_L4_CHKSUM
654 #define RTE_ETH_RSS_L2TPV2 RTE_BIT64(36)
657 * We use the following macros to combine with above RTE_ETH_RSS_* for
658 * more specific input set selection. These bits are defined starting
659 * from the high end of the 64 bits.
660 * Note: If we use above RTE_ETH_RSS_* without SRC/DST_ONLY, it represents
661 * both SRC and DST are taken into account. If SRC_ONLY and DST_ONLY of
662 * the same level are used simultaneously, it is the same case as none of
665 #define RTE_ETH_RSS_L3_SRC_ONLY RTE_BIT64(63)
666 #define RTE_ETH_RSS_L3_DST_ONLY RTE_BIT64(62)
667 #define RTE_ETH_RSS_L4_SRC_ONLY RTE_BIT64(61)
668 #define RTE_ETH_RSS_L4_DST_ONLY RTE_BIT64(60)
669 #define RTE_ETH_RSS_L2_SRC_ONLY RTE_BIT64(59)
670 #define RTE_ETH_RSS_L2_DST_ONLY RTE_BIT64(58)
672 #define ETH_RSS_L3_SRC_ONLY RTE_DEPRECATED(ETH_RSS_L3_SRC_ONLY) RTE_ETH_RSS_L3_SRC_ONLY
673 #define ETH_RSS_L3_DST_ONLY RTE_DEPRECATED(ETH_RSS_L3_DST_ONLY) RTE_ETH_RSS_L3_DST_ONLY
674 #define ETH_RSS_L4_SRC_ONLY RTE_DEPRECATED(ETH_RSS_L4_SRC_ONLY) RTE_ETH_RSS_L4_SRC_ONLY
675 #define ETH_RSS_L4_DST_ONLY RTE_DEPRECATED(ETH_RSS_L4_DST_ONLY) RTE_ETH_RSS_L4_DST_ONLY
676 #define ETH_RSS_L2_SRC_ONLY RTE_DEPRECATED(ETH_RSS_L2_SRC_ONLY) RTE_ETH_RSS_L2_SRC_ONLY
677 #define ETH_RSS_L2_DST_ONLY RTE_DEPRECATED(ETH_RSS_L2_DST_ONLY) RTE_ETH_RSS_L2_DST_ONLY
680 * Only select IPV6 address prefix as RSS input set according to
681 * https://tools.ietf.org/html/rfc6052
682 * Must be combined with RTE_ETH_RSS_IPV6, RTE_ETH_RSS_NONFRAG_IPV6_UDP,
683 * RTE_ETH_RSS_NONFRAG_IPV6_TCP, RTE_ETH_RSS_NONFRAG_IPV6_SCTP.
685 #define RTE_ETH_RSS_L3_PRE32 RTE_BIT64(57)
686 #define RTE_ETH_RSS_L3_PRE40 RTE_BIT64(56)
687 #define RTE_ETH_RSS_L3_PRE48 RTE_BIT64(55)
688 #define RTE_ETH_RSS_L3_PRE56 RTE_BIT64(54)
689 #define RTE_ETH_RSS_L3_PRE64 RTE_BIT64(53)
690 #define RTE_ETH_RSS_L3_PRE96 RTE_BIT64(52)
693 * Use the following macros to combine with the above layers
694 * to choose inner and outer layers or both for RSS computation.
695 * Bits 50 and 51 are reserved for this.
699 * level 0, requests the default behavior.
700 * Depending on the packet type, it can mean outermost, innermost,
701 * anything in between or even no RSS.
702 * It basically stands for the innermost encapsulation level RSS
703 * can be performed on according to PMD and device capabilities.
705 #define RTE_ETH_RSS_LEVEL_PMD_DEFAULT (UINT64_C(0) << 50)
706 #define ETH_RSS_LEVEL_PMD_DEFAULT RTE_DEPRECATED(ETH_RSS_LEVEL_PMD_DEFAULT) RTE_ETH_RSS_LEVEL_PMD_DEFAULT
709 * level 1, requests RSS to be performed on the outermost packet
710 * encapsulation level.
712 #define RTE_ETH_RSS_LEVEL_OUTERMOST (UINT64_C(1) << 50)
713 #define ETH_RSS_LEVEL_OUTERMOST RTE_DEPRECATED(ETH_RSS_LEVEL_OUTERMOST) RTE_ETH_RSS_LEVEL_OUTERMOST
716 * level 2, requests RSS to be performed on the specified inner packet
717 * encapsulation level, from outermost to innermost (lower to higher values).
719 #define RTE_ETH_RSS_LEVEL_INNERMOST (UINT64_C(2) << 50)
720 #define RTE_ETH_RSS_LEVEL_MASK (UINT64_C(3) << 50)
722 #define ETH_RSS_LEVEL_INNERMOST RTE_DEPRECATED(ETH_RSS_LEVEL_INNERMOST) RTE_ETH_RSS_LEVEL_INNERMOST
723 #define ETH_RSS_LEVEL_MASK RTE_DEPRECATED(ETH_RSS_LEVEL_MASK) RTE_ETH_RSS_LEVEL_MASK
725 #define RTE_ETH_RSS_LEVEL(rss_hf) ((rss_hf & RTE_ETH_RSS_LEVEL_MASK) >> 50)
726 #define ETH_RSS_LEVEL(rss_hf) RTE_DEPRECATED(ETH_RSS_LEVEL(rss_hf)) RTE_ETH_RSS_LEVEL(rss_hf)
729 * For input set change of hash filter, if SRC_ONLY and DST_ONLY of
730 * the same level are used simultaneously, it is the same case as
731 * none of them are added.
734 * RSS types with SRC/DST_ONLY.
738 static inline uint64_t
739 rte_eth_rss_hf_refine(uint64_t rss_hf)
741 if ((rss_hf & RTE_ETH_RSS_L3_SRC_ONLY) && (rss_hf & RTE_ETH_RSS_L3_DST_ONLY))
742 rss_hf &= ~(RTE_ETH_RSS_L3_SRC_ONLY | RTE_ETH_RSS_L3_DST_ONLY);
744 if ((rss_hf & RTE_ETH_RSS_L4_SRC_ONLY) && (rss_hf & RTE_ETH_RSS_L4_DST_ONLY))
745 rss_hf &= ~(RTE_ETH_RSS_L4_SRC_ONLY | RTE_ETH_RSS_L4_DST_ONLY);
750 #define RTE_ETH_RSS_IPV6_PRE32 ( \
752 RTE_ETH_RSS_L3_PRE32)
753 #define ETH_RSS_IPV6_PRE32 RTE_DEPRECATED(ETH_RSS_IPV6_PRE32) RTE_ETH_RSS_IPV6_PRE32
755 #define RTE_ETH_RSS_IPV6_PRE40 ( \
757 RTE_ETH_RSS_L3_PRE40)
758 #define ETH_RSS_IPV6_PRE40 RTE_DEPRECATED(ETH_RSS_IPV6_PRE40) RTE_ETH_RSS_IPV6_PRE40
760 #define RTE_ETH_RSS_IPV6_PRE48 ( \
762 RTE_ETH_RSS_L3_PRE48)
763 #define ETH_RSS_IPV6_PRE48 RTE_DEPRECATED(ETH_RSS_IPV6_PRE48) RTE_ETH_RSS_IPV6_PRE48
765 #define RTE_ETH_RSS_IPV6_PRE56 ( \
767 RTE_ETH_RSS_L3_PRE56)
768 #define ETH_RSS_IPV6_PRE56 RTE_DEPRECATED(ETH_RSS_IPV6_PRE56) RTE_ETH_RSS_IPV6_PRE56
770 #define RTE_ETH_RSS_IPV6_PRE64 ( \
772 RTE_ETH_RSS_L3_PRE64)
773 #define ETH_RSS_IPV6_PRE64 RTE_DEPRECATED(ETH_RSS_IPV6_PRE64) RTE_ETH_RSS_IPV6_PRE64
775 #define RTE_ETH_RSS_IPV6_PRE96 ( \
777 RTE_ETH_RSS_L3_PRE96)
778 #define ETH_RSS_IPV6_PRE96 RTE_DEPRECATED(ETH_RSS_IPV6_PRE96) RTE_ETH_RSS_IPV6_PRE96
780 #define RTE_ETH_RSS_IPV6_PRE32_UDP ( \
781 RTE_ETH_RSS_NONFRAG_IPV6_UDP | \
782 RTE_ETH_RSS_L3_PRE32)
783 #define ETH_RSS_IPV6_PRE32_UDP RTE_DEPRECATED(ETH_RSS_IPV6_PRE32_UDP) RTE_ETH_RSS_IPV6_PRE32_UDP
785 #define RTE_ETH_RSS_IPV6_PRE40_UDP ( \
786 RTE_ETH_RSS_NONFRAG_IPV6_UDP | \
787 RTE_ETH_RSS_L3_PRE40)
788 #define ETH_RSS_IPV6_PRE40_UDP RTE_DEPRECATED(ETH_RSS_IPV6_PRE40_UDP) RTE_ETH_RSS_IPV6_PRE40_UDP
790 #define RTE_ETH_RSS_IPV6_PRE48_UDP ( \
791 RTE_ETH_RSS_NONFRAG_IPV6_UDP | \
792 RTE_ETH_RSS_L3_PRE48)
793 #define ETH_RSS_IPV6_PRE48_UDP RTE_DEPRECATED(ETH_RSS_IPV6_PRE48_UDP) RTE_ETH_RSS_IPV6_PRE48_UDP
795 #define RTE_ETH_RSS_IPV6_PRE56_UDP ( \
796 RTE_ETH_RSS_NONFRAG_IPV6_UDP | \
797 RTE_ETH_RSS_L3_PRE56)
798 #define ETH_RSS_IPV6_PRE56_UDP RTE_DEPRECATED(ETH_RSS_IPV6_PRE56_UDP) RTE_ETH_RSS_IPV6_PRE56_UDP
800 #define RTE_ETH_RSS_IPV6_PRE64_UDP ( \
801 RTE_ETH_RSS_NONFRAG_IPV6_UDP | \
802 RTE_ETH_RSS_L3_PRE64)
803 #define ETH_RSS_IPV6_PRE64_UDP RTE_DEPRECATED(ETH_RSS_IPV6_PRE64_UDP) RTE_ETH_RSS_IPV6_PRE64_UDP
805 #define RTE_ETH_RSS_IPV6_PRE96_UDP ( \
806 RTE_ETH_RSS_NONFRAG_IPV6_UDP | \
807 RTE_ETH_RSS_L3_PRE96)
808 #define ETH_RSS_IPV6_PRE96_UDP RTE_DEPRECATED(ETH_RSS_IPV6_PRE96_UDP) RTE_ETH_RSS_IPV6_PRE96_UDP
810 #define RTE_ETH_RSS_IPV6_PRE32_TCP ( \
811 RTE_ETH_RSS_NONFRAG_IPV6_TCP | \
812 RTE_ETH_RSS_L3_PRE32)
813 #define ETH_RSS_IPV6_PRE32_TCP RTE_DEPRECATED(ETH_RSS_IPV6_PRE32_TCP) RTE_ETH_RSS_IPV6_PRE32_TCP
815 #define RTE_ETH_RSS_IPV6_PRE40_TCP ( \
816 RTE_ETH_RSS_NONFRAG_IPV6_TCP | \
817 RTE_ETH_RSS_L3_PRE40)
818 #define ETH_RSS_IPV6_PRE40_TCP RTE_DEPRECATED(ETH_RSS_IPV6_PRE40_TCP) RTE_ETH_RSS_IPV6_PRE40_TCP
820 #define RTE_ETH_RSS_IPV6_PRE48_TCP ( \
821 RTE_ETH_RSS_NONFRAG_IPV6_TCP | \
822 RTE_ETH_RSS_L3_PRE48)
823 #define ETH_RSS_IPV6_PRE48_TCP RTE_DEPRECATED(ETH_RSS_IPV6_PRE48_TCP) RTE_ETH_RSS_IPV6_PRE48_TCP
825 #define RTE_ETH_RSS_IPV6_PRE56_TCP ( \
826 RTE_ETH_RSS_NONFRAG_IPV6_TCP | \
827 RTE_ETH_RSS_L3_PRE56)
828 #define ETH_RSS_IPV6_PRE56_TCP RTE_DEPRECATED(ETH_RSS_IPV6_PRE56_TCP) RTE_ETH_RSS_IPV6_PRE56_TCP
830 #define RTE_ETH_RSS_IPV6_PRE64_TCP ( \
831 RTE_ETH_RSS_NONFRAG_IPV6_TCP | \
832 RTE_ETH_RSS_L3_PRE64)
833 #define ETH_RSS_IPV6_PRE64_TCP RTE_DEPRECATED(ETH_RSS_IPV6_PRE64_TCP) RTE_ETH_RSS_IPV6_PRE64_TCP
835 #define RTE_ETH_RSS_IPV6_PRE96_TCP ( \
836 RTE_ETH_RSS_NONFRAG_IPV6_TCP | \
837 RTE_ETH_RSS_L3_PRE96)
838 #define ETH_RSS_IPV6_PRE96_TCP RTE_DEPRECATED(ETH_RSS_IPV6_PRE96_TCP) RTE_ETH_RSS_IPV6_PRE96_TCP
840 #define RTE_ETH_RSS_IPV6_PRE32_SCTP ( \
841 RTE_ETH_RSS_NONFRAG_IPV6_SCTP | \
842 RTE_ETH_RSS_L3_PRE32)
843 #define ETH_RSS_IPV6_PRE32_SCTP RTE_DEPRECATED(ETH_RSS_IPV6_PRE32_SCTP) RTE_ETH_RSS_IPV6_PRE32_SCTP
845 #define RTE_ETH_RSS_IPV6_PRE40_SCTP ( \
846 RTE_ETH_RSS_NONFRAG_IPV6_SCTP | \
847 RTE_ETH_RSS_L3_PRE40)
848 #define ETH_RSS_IPV6_PRE40_SCTP RTE_DEPRECATED(ETH_RSS_IPV6_PRE40_SCTP) RTE_ETH_RSS_IPV6_PRE40_SCTP
850 #define RTE_ETH_RSS_IPV6_PRE48_SCTP ( \
851 RTE_ETH_RSS_NONFRAG_IPV6_SCTP | \
852 RTE_ETH_RSS_L3_PRE48)
853 #define ETH_RSS_IPV6_PRE48_SCTP RTE_DEPRECATED(ETH_RSS_IPV6_PRE48_SCTP) RTE_ETH_RSS_IPV6_PRE48_SCTP
855 #define RTE_ETH_RSS_IPV6_PRE56_SCTP ( \
856 RTE_ETH_RSS_NONFRAG_IPV6_SCTP | \
857 RTE_ETH_RSS_L3_PRE56)
858 #define ETH_RSS_IPV6_PRE56_SCTP RTE_DEPRECATED(ETH_RSS_IPV6_PRE56_SCTP) RTE_ETH_RSS_IPV6_PRE56_SCTP
860 #define RTE_ETH_RSS_IPV6_PRE64_SCTP ( \
861 RTE_ETH_RSS_NONFRAG_IPV6_SCTP | \
862 RTE_ETH_RSS_L3_PRE64)
863 #define ETH_RSS_IPV6_PRE64_SCTP RTE_DEPRECATED(ETH_RSS_IPV6_PRE64_SCTP) RTE_ETH_RSS_IPV6_PRE64_SCTP
865 #define RTE_ETH_RSS_IPV6_PRE96_SCTP ( \
866 RTE_ETH_RSS_NONFRAG_IPV6_SCTP | \
867 RTE_ETH_RSS_L3_PRE96)
868 #define ETH_RSS_IPV6_PRE96_SCTP RTE_DEPRECATED(ETH_RSS_IPV6_PRE96_SCTP) RTE_ETH_RSS_IPV6_PRE96_SCTP
870 #define RTE_ETH_RSS_IP ( \
872 RTE_ETH_RSS_FRAG_IPV4 | \
873 RTE_ETH_RSS_NONFRAG_IPV4_OTHER | \
875 RTE_ETH_RSS_FRAG_IPV6 | \
876 RTE_ETH_RSS_NONFRAG_IPV6_OTHER | \
878 #define ETH_RSS_IP RTE_DEPRECATED(ETH_RSS_IP) RTE_ETH_RSS_IP
880 #define RTE_ETH_RSS_UDP ( \
881 RTE_ETH_RSS_NONFRAG_IPV4_UDP | \
882 RTE_ETH_RSS_NONFRAG_IPV6_UDP | \
883 RTE_ETH_RSS_IPV6_UDP_EX)
884 #define ETH_RSS_UDP RTE_DEPRECATED(ETH_RSS_UDP) RTE_ETH_RSS_UDP
886 #define RTE_ETH_RSS_TCP ( \
887 RTE_ETH_RSS_NONFRAG_IPV4_TCP | \
888 RTE_ETH_RSS_NONFRAG_IPV6_TCP | \
889 RTE_ETH_RSS_IPV6_TCP_EX)
890 #define ETH_RSS_TCP RTE_DEPRECATED(ETH_RSS_TCP) RTE_ETH_RSS_TCP
892 #define RTE_ETH_RSS_SCTP ( \
893 RTE_ETH_RSS_NONFRAG_IPV4_SCTP | \
894 RTE_ETH_RSS_NONFRAG_IPV6_SCTP)
895 #define ETH_RSS_SCTP RTE_DEPRECATED(ETH_RSS_SCTP) RTE_ETH_RSS_SCTP
897 #define RTE_ETH_RSS_TUNNEL ( \
898 RTE_ETH_RSS_VXLAN | \
899 RTE_ETH_RSS_GENEVE | \
901 #define ETH_RSS_TUNNEL RTE_DEPRECATED(ETH_RSS_TUNNEL) RTE_ETH_RSS_TUNNEL
903 #define RTE_ETH_RSS_VLAN ( \
904 RTE_ETH_RSS_S_VLAN | \
906 #define ETH_RSS_VLAN RTE_DEPRECATED(ETH_RSS_VLAN) RTE_ETH_RSS_VLAN
908 /** Mask of valid RSS hash protocols */
909 #define RTE_ETH_RSS_PROTO_MASK ( \
911 RTE_ETH_RSS_FRAG_IPV4 | \
912 RTE_ETH_RSS_NONFRAG_IPV4_TCP | \
913 RTE_ETH_RSS_NONFRAG_IPV4_UDP | \
914 RTE_ETH_RSS_NONFRAG_IPV4_SCTP | \
915 RTE_ETH_RSS_NONFRAG_IPV4_OTHER | \
917 RTE_ETH_RSS_FRAG_IPV6 | \
918 RTE_ETH_RSS_NONFRAG_IPV6_TCP | \
919 RTE_ETH_RSS_NONFRAG_IPV6_UDP | \
920 RTE_ETH_RSS_NONFRAG_IPV6_SCTP | \
921 RTE_ETH_RSS_NONFRAG_IPV6_OTHER | \
922 RTE_ETH_RSS_L2_PAYLOAD | \
923 RTE_ETH_RSS_IPV6_EX | \
924 RTE_ETH_RSS_IPV6_TCP_EX | \
925 RTE_ETH_RSS_IPV6_UDP_EX | \
927 RTE_ETH_RSS_VXLAN | \
928 RTE_ETH_RSS_GENEVE | \
929 RTE_ETH_RSS_NVGRE | \
931 #define ETH_RSS_PROTO_MASK RTE_DEPRECATED(ETH_RSS_PROTO_MASK) RTE_ETH_RSS_PROTO_MASK
934 * Definitions used for redirection table entry size.
935 * Some RSS RETA sizes may not be supported by some drivers, check the
936 * documentation or the description of relevant functions for more details.
938 #define RTE_ETH_RSS_RETA_SIZE_64 64
939 #define RTE_ETH_RSS_RETA_SIZE_128 128
940 #define RTE_ETH_RSS_RETA_SIZE_256 256
941 #define RTE_ETH_RSS_RETA_SIZE_512 512
942 #define RTE_ETH_RETA_GROUP_SIZE 64
944 #define ETH_RSS_RETA_SIZE_64 RTE_DEPRECATED(ETH_RSS_RETA_SIZE_64) RTE_ETH_RSS_RETA_SIZE_64
945 #define ETH_RSS_RETA_SIZE_128 RTE_DEPRECATED(ETH_RSS_RETA_SIZE_128) RTE_ETH_RSS_RETA_SIZE_128
946 #define ETH_RSS_RETA_SIZE_256 RTE_DEPRECATED(ETH_RSS_RETA_SIZE_256) RTE_ETH_RSS_RETA_SIZE_256
947 #define ETH_RSS_RETA_SIZE_512 RTE_DEPRECATED(ETH_RSS_RETA_SIZE_512) RTE_ETH_RSS_RETA_SIZE_512
948 #define RTE_RETA_GROUP_SIZE RTE_DEPRECATED(RTE_RETA_GROUP_SIZE) RTE_ETH_RETA_GROUP_SIZE
950 /**@{@name VMDq and DCB maximums */
951 #define RTE_ETH_VMDQ_MAX_VLAN_FILTERS 64 /**< Maximum nb. of VMDq VLAN filters. */
952 #define RTE_ETH_DCB_NUM_USER_PRIORITIES 8 /**< Maximum nb. of DCB priorities. */
953 #define RTE_ETH_VMDQ_DCB_NUM_QUEUES 128 /**< Maximum nb. of VMDq DCB queues. */
954 #define RTE_ETH_DCB_NUM_QUEUES 128 /**< Maximum nb. of DCB queues. */
957 #define ETH_VMDQ_MAX_VLAN_FILTERS RTE_DEPRECATED(ETH_VMDQ_MAX_VLAN_FILTERS) RTE_ETH_VMDQ_MAX_VLAN_FILTERS
958 #define ETH_DCB_NUM_USER_PRIORITIES RTE_DEPRECATED(ETH_DCB_NUM_USER_PRIORITIES) RTE_ETH_DCB_NUM_USER_PRIORITIES
959 #define ETH_VMDQ_DCB_NUM_QUEUES RTE_DEPRECATED(ETH_VMDQ_DCB_NUM_QUEUES) RTE_ETH_VMDQ_DCB_NUM_QUEUES
960 #define ETH_DCB_NUM_QUEUES RTE_DEPRECATED(ETH_DCB_NUM_QUEUES) RTE_ETH_DCB_NUM_QUEUES
962 /**@{@name DCB capabilities */
963 #define RTE_ETH_DCB_PG_SUPPORT RTE_BIT32(0) /**< Priority Group(ETS) support. */
964 #define RTE_ETH_DCB_PFC_SUPPORT RTE_BIT32(1) /**< Priority Flow Control support. */
967 #define ETH_DCB_PG_SUPPORT RTE_DEPRECATED(ETH_DCB_PG_SUPPORT) RTE_ETH_DCB_PG_SUPPORT
968 #define ETH_DCB_PFC_SUPPORT RTE_DEPRECATED(ETH_DCB_PFC_SUPPORT) RTE_ETH_DCB_PFC_SUPPORT
970 /**@{@name VLAN offload bits */
971 #define RTE_ETH_VLAN_STRIP_OFFLOAD 0x0001 /**< VLAN Strip On/Off */
972 #define RTE_ETH_VLAN_FILTER_OFFLOAD 0x0002 /**< VLAN Filter On/Off */
973 #define RTE_ETH_VLAN_EXTEND_OFFLOAD 0x0004 /**< VLAN Extend On/Off */
974 #define RTE_ETH_QINQ_STRIP_OFFLOAD 0x0008 /**< QINQ Strip On/Off */
976 #define ETH_VLAN_STRIP_OFFLOAD RTE_DEPRECATED(ETH_VLAN_STRIP_OFFLOAD) RTE_ETH_VLAN_STRIP_OFFLOAD
977 #define ETH_VLAN_FILTER_OFFLOAD RTE_DEPRECATED(ETH_VLAN_FILTER_OFFLOAD) RTE_ETH_VLAN_FILTER_OFFLOAD
978 #define ETH_VLAN_EXTEND_OFFLOAD RTE_DEPRECATED(ETH_VLAN_EXTEND_OFFLOAD) RTE_ETH_VLAN_EXTEND_OFFLOAD
979 #define ETH_QINQ_STRIP_OFFLOAD RTE_DEPRECATED(ETH_QINQ_STRIP_OFFLOAD) RTE_ETH_QINQ_STRIP_OFFLOAD
981 #define RTE_ETH_VLAN_STRIP_MASK 0x0001 /**< VLAN Strip setting mask */
982 #define RTE_ETH_VLAN_FILTER_MASK 0x0002 /**< VLAN Filter setting mask*/
983 #define RTE_ETH_VLAN_EXTEND_MASK 0x0004 /**< VLAN Extend setting mask*/
984 #define RTE_ETH_QINQ_STRIP_MASK 0x0008 /**< QINQ Strip setting mask */
985 #define RTE_ETH_VLAN_ID_MAX 0x0FFF /**< VLAN ID is in lower 12 bits*/
988 #define ETH_VLAN_STRIP_MASK RTE_DEPRECATED(ETH_VLAN_STRIP_MASK) RTE_ETH_VLAN_STRIP_MASK
989 #define ETH_VLAN_FILTER_MASK RTE_DEPRECATED(ETH_VLAN_FILTER_MASK) RTE_ETH_VLAN_FILTER_MASK
990 #define ETH_VLAN_EXTEND_MASK RTE_DEPRECATED(ETH_VLAN_EXTEND_MASK) RTE_ETH_VLAN_EXTEND_MASK
991 #define ETH_QINQ_STRIP_MASK RTE_DEPRECATED(ETH_QINQ_STRIP_MASK) RTE_ETH_QINQ_STRIP_MASK
992 #define ETH_VLAN_ID_MAX RTE_DEPRECATED(ETH_VLAN_ID_MAX) RTE_ETH_VLAN_ID_MAX
994 /* Definitions used for receive MAC address */
995 #define RTE_ETH_NUM_RECEIVE_MAC_ADDR 128 /**< Maximum nb. of receive mac addr. */
996 #define ETH_NUM_RECEIVE_MAC_ADDR RTE_DEPRECATED(ETH_NUM_RECEIVE_MAC_ADDR) RTE_ETH_NUM_RECEIVE_MAC_ADDR
998 /* Definitions used for unicast hash */
999 #define RTE_ETH_VMDQ_NUM_UC_HASH_ARRAY 128 /**< Maximum nb. of UC hash array. */
1000 #define ETH_VMDQ_NUM_UC_HASH_ARRAY RTE_DEPRECATED(ETH_VMDQ_NUM_UC_HASH_ARRAY) RTE_ETH_VMDQ_NUM_UC_HASH_ARRAY
1002 /**@{@name VMDq Rx mode
1003 * @see rte_eth_vmdq_rx_conf.rx_mode
1005 /** Accept untagged packets. */
1006 #define RTE_ETH_VMDQ_ACCEPT_UNTAG RTE_BIT32(0)
1007 /** Accept packets in multicast table. */
1008 #define RTE_ETH_VMDQ_ACCEPT_HASH_MC RTE_BIT32(1)
1009 /** Accept packets in unicast table. */
1010 #define RTE_ETH_VMDQ_ACCEPT_HASH_UC RTE_BIT32(2)
1011 /** Accept broadcast packets. */
1012 #define RTE_ETH_VMDQ_ACCEPT_BROADCAST RTE_BIT32(3)
1013 /** Multicast promiscuous. */
1014 #define RTE_ETH_VMDQ_ACCEPT_MULTICAST RTE_BIT32(4)
1017 #define ETH_VMDQ_ACCEPT_UNTAG RTE_DEPRECATED(ETH_VMDQ_ACCEPT_UNTAG) RTE_ETH_VMDQ_ACCEPT_UNTAG
1018 #define ETH_VMDQ_ACCEPT_HASH_MC RTE_DEPRECATED(ETH_VMDQ_ACCEPT_HASH_MC) RTE_ETH_VMDQ_ACCEPT_HASH_MC
1019 #define ETH_VMDQ_ACCEPT_HASH_UC RTE_DEPRECATED(ETH_VMDQ_ACCEPT_HASH_UC) RTE_ETH_VMDQ_ACCEPT_HASH_UC
1020 #define ETH_VMDQ_ACCEPT_BROADCAST RTE_DEPRECATED(ETH_VMDQ_ACCEPT_BROADCAST) RTE_ETH_VMDQ_ACCEPT_BROADCAST
1021 #define ETH_VMDQ_ACCEPT_MULTICAST RTE_DEPRECATED(ETH_VMDQ_ACCEPT_MULTICAST) RTE_ETH_VMDQ_ACCEPT_MULTICAST
1024 * A structure used to configure 64 entries of Redirection Table of the
1025 * Receive Side Scaling (RSS) feature of an Ethernet port. To configure
1026 * more than 64 entries supported by hardware, an array of this structure
1029 struct rte_eth_rss_reta_entry64 {
1030 /** Mask bits indicate which entries need to be updated/queried. */
1032 /** Group of 64 redirection table entries. */
1033 uint16_t reta[RTE_ETH_RETA_GROUP_SIZE];
1037 * This enum indicates the possible number of traffic classes
1038 * in DCB configurations
1040 enum rte_eth_nb_tcs {
1041 RTE_ETH_4_TCS = 4, /**< 4 TCs with DCB. */
1042 RTE_ETH_8_TCS = 8 /**< 8 TCs with DCB. */
1044 #define ETH_4_TCS RTE_DEPRECATED(ETH_4_TCS) RTE_ETH_4_TCS
1045 #define ETH_8_TCS RTE_DEPRECATED(ETH_8_TCS) RTE_ETH_8_TCS
1048 * This enum indicates the possible number of queue pools
1049 * in VMDq configurations.
1051 enum rte_eth_nb_pools {
1052 RTE_ETH_8_POOLS = 8, /**< 8 VMDq pools. */
1053 RTE_ETH_16_POOLS = 16, /**< 16 VMDq pools. */
1054 RTE_ETH_32_POOLS = 32, /**< 32 VMDq pools. */
1055 RTE_ETH_64_POOLS = 64 /**< 64 VMDq pools. */
1057 #define ETH_8_POOLS RTE_DEPRECATED(ETH_8_POOLS) RTE_ETH_8_POOLS
1058 #define ETH_16_POOLS RTE_DEPRECATED(ETH_16_POOLS) RTE_ETH_16_POOLS
1059 #define ETH_32_POOLS RTE_DEPRECATED(ETH_32_POOLS) RTE_ETH_32_POOLS
1060 #define ETH_64_POOLS RTE_DEPRECATED(ETH_64_POOLS) RTE_ETH_64_POOLS
1062 /* This structure may be extended in future. */
1063 struct rte_eth_dcb_rx_conf {
1064 enum rte_eth_nb_tcs nb_tcs; /**< Possible DCB TCs, 4 or 8 TCs */
1065 /** Traffic class each UP mapped to. */
1066 uint8_t dcb_tc[RTE_ETH_DCB_NUM_USER_PRIORITIES];
1069 struct rte_eth_vmdq_dcb_tx_conf {
1070 enum rte_eth_nb_pools nb_queue_pools; /**< With DCB, 16 or 32 pools. */
1071 /** Traffic class each UP mapped to. */
1072 uint8_t dcb_tc[RTE_ETH_DCB_NUM_USER_PRIORITIES];
1075 struct rte_eth_dcb_tx_conf {
1076 enum rte_eth_nb_tcs nb_tcs; /**< Possible DCB TCs, 4 or 8 TCs. */
1077 /** Traffic class each UP mapped to. */
1078 uint8_t dcb_tc[RTE_ETH_DCB_NUM_USER_PRIORITIES];
1081 struct rte_eth_vmdq_tx_conf {
1082 enum rte_eth_nb_pools nb_queue_pools; /**< VMDq mode, 64 pools. */
1086 * A structure used to configure the VMDq+DCB feature
1087 * of an Ethernet port.
1089 * Using this feature, packets are routed to a pool of queues, based
1090 * on the VLAN ID in the VLAN tag, and then to a specific queue within
1091 * that pool, using the user priority VLAN tag field.
1093 * A default pool may be used, if desired, to route all traffic which
1094 * does not match the VLAN filter rules.
1096 struct rte_eth_vmdq_dcb_conf {
1097 enum rte_eth_nb_pools nb_queue_pools; /**< With DCB, 16 or 32 pools */
1098 uint8_t enable_default_pool; /**< If non-zero, use a default pool */
1099 uint8_t default_pool; /**< The default pool, if applicable */
1100 uint8_t nb_pool_maps; /**< We can have up to 64 filters/mappings */
1102 uint16_t vlan_id; /**< The VLAN ID of the received frame */
1103 uint64_t pools; /**< Bitmask of pools for packet Rx */
1104 } pool_map[RTE_ETH_VMDQ_MAX_VLAN_FILTERS]; /**< VMDq VLAN pool maps. */
1105 /** Selects a queue in a pool */
1106 uint8_t dcb_tc[RTE_ETH_DCB_NUM_USER_PRIORITIES];
1110 * A structure used to configure the VMDq feature of an Ethernet port when
1111 * not combined with the DCB feature.
1113 * Using this feature, packets are routed to a pool of queues. By default,
1114 * the pool selection is based on the MAC address, the VLAN ID in the
1115 * VLAN tag as specified in the pool_map array.
1116 * Passing the RTE_ETH_VMDQ_ACCEPT_UNTAG in the rx_mode field allows pool
1117 * selection using only the MAC address. MAC address to pool mapping is done
1118 * using the rte_eth_dev_mac_addr_add function, with the pool parameter
1119 * corresponding to the pool ID.
1121 * Queue selection within the selected pool will be done using RSS when
1122 * it is enabled or revert to the first queue of the pool if not.
1124 * A default pool may be used, if desired, to route all traffic which
1125 * does not match the VLAN filter rules or any pool MAC address.
1127 struct rte_eth_vmdq_rx_conf {
1128 enum rte_eth_nb_pools nb_queue_pools; /**< VMDq only mode, 8 or 64 pools */
1129 uint8_t enable_default_pool; /**< If non-zero, use a default pool */
1130 uint8_t default_pool; /**< The default pool, if applicable */
1131 uint8_t enable_loop_back; /**< Enable VT loop back */
1132 uint8_t nb_pool_maps; /**< We can have up to 64 filters/mappings */
1133 uint32_t rx_mode; /**< Flags from ETH_VMDQ_ACCEPT_* */
1135 uint16_t vlan_id; /**< The VLAN ID of the received frame */
1136 uint64_t pools; /**< Bitmask of pools for packet Rx */
1137 } pool_map[RTE_ETH_VMDQ_MAX_VLAN_FILTERS]; /**< VMDq VLAN pool maps. */
1141 * A structure used to configure the Tx features of an Ethernet port.
1143 struct rte_eth_txmode {
1144 enum rte_eth_tx_mq_mode mq_mode; /**< Tx multi-queues mode. */
1146 * Per-port Tx offloads to be set using RTE_ETH_TX_OFFLOAD_* flags.
1147 * Only offloads set on tx_offload_capa field on rte_eth_dev_info
1148 * structure are allowed to be set.
1154 uint8_t /** If set, reject sending out tagged pkts */
1155 hw_vlan_reject_tagged : 1,
1156 /** If set, reject sending out untagged pkts */
1157 hw_vlan_reject_untagged : 1,
1158 /** If set, enable port based VLAN insertion */
1159 hw_vlan_insert_pvid : 1;
1161 uint64_t reserved_64s[2]; /**< Reserved for future fields */
1162 void *reserved_ptrs[2]; /**< Reserved for future fields */
1167 * @b EXPERIMENTAL: this structure may change without prior notice.
1169 * A structure used to configure an Rx packet segment to split.
1171 * If RTE_ETH_RX_OFFLOAD_BUFFER_SPLIT flag is set in offloads field,
1172 * the PMD will split the received packets into multiple segments
1173 * according to the specification in the description array:
1175 * - The first network buffer will be allocated from the memory pool,
1176 * specified in the first array element, the second buffer, from the
1177 * pool in the second element, and so on.
1179 * - The offsets from the segment description elements specify
1180 * the data offset from the buffer beginning except the first mbuf.
1181 * The first segment offset is added with RTE_PKTMBUF_HEADROOM.
1183 * - The lengths in the elements define the maximal data amount
1184 * being received to each segment. The receiving starts with filling
1185 * up the first mbuf data buffer up to specified length. If the
1186 * there are data remaining (packet is longer than buffer in the first
1187 * mbuf) the following data will be pushed to the next segment
1188 * up to its own length, and so on.
1190 * - If the length in the segment description element is zero
1191 * the actual buffer size will be deduced from the appropriate
1192 * memory pool properties.
1194 * - If there is not enough elements to describe the buffer for entire
1195 * packet of maximal length the following parameters will be used
1196 * for the all remaining segments:
1197 * - pool from the last valid element
1198 * - the buffer size from this pool
1201 struct rte_eth_rxseg_split {
1202 struct rte_mempool *mp; /**< Memory pool to allocate segment from. */
1203 uint16_t length; /**< Segment data length, configures split point. */
1204 uint16_t offset; /**< Data offset from beginning of mbuf data buffer. */
1205 uint32_t reserved; /**< Reserved field. */
1210 * @b EXPERIMENTAL: this structure may change without prior notice.
1212 * A common structure used to describe Rx packet segment properties.
1214 union rte_eth_rxseg {
1215 /* The settings for buffer split offload. */
1216 struct rte_eth_rxseg_split split;
1217 /* The other features settings should be added here. */
1221 * A structure used to configure an Rx ring of an Ethernet port.
1223 struct rte_eth_rxconf {
1224 struct rte_eth_thresh rx_thresh; /**< Rx ring threshold registers. */
1225 uint16_t rx_free_thresh; /**< Drives the freeing of Rx descriptors. */
1226 uint8_t rx_drop_en; /**< Drop packets if no descriptors are available. */
1227 uint8_t rx_deferred_start; /**< Do not start queue with rte_eth_dev_start(). */
1228 uint16_t rx_nseg; /**< Number of descriptions in rx_seg array. */
1230 * Share group index in Rx domain and switch domain.
1231 * Non-zero value to enable Rx queue share, zero value disable share.
1232 * PMD is responsible for Rx queue consistency checks to avoid member
1233 * port's configuration contradict to each other.
1235 uint16_t share_group;
1236 uint16_t share_qid; /**< Shared Rx queue ID in group */
1238 * Per-queue Rx offloads to be set using RTE_ETH_RX_OFFLOAD_* flags.
1239 * Only offloads set on rx_queue_offload_capa or rx_offload_capa
1240 * fields on rte_eth_dev_info structure are allowed to be set.
1244 * Points to the array of segment descriptions for an entire packet.
1245 * Array elements are properties for consecutive Rx segments.
1247 * The supported capabilities of receiving segmentation is reported
1248 * in rte_eth_dev_info.rx_seg_capa field.
1250 union rte_eth_rxseg *rx_seg;
1252 uint64_t reserved_64s[2]; /**< Reserved for future fields */
1253 void *reserved_ptrs[2]; /**< Reserved for future fields */
1257 * A structure used to configure a Tx ring of an Ethernet port.
1259 struct rte_eth_txconf {
1260 struct rte_eth_thresh tx_thresh; /**< Tx ring threshold registers. */
1261 uint16_t tx_rs_thresh; /**< Drives the setting of RS bit on TXDs. */
1262 uint16_t tx_free_thresh; /**< Start freeing Tx buffers if there are
1263 less free descriptors than this value. */
1265 uint8_t tx_deferred_start; /**< Do not start queue with rte_eth_dev_start(). */
1267 * Per-queue Tx offloads to be set using RTE_ETH_TX_OFFLOAD_* flags.
1268 * Only offloads set on tx_queue_offload_capa or tx_offload_capa
1269 * fields on rte_eth_dev_info structure are allowed to be set.
1273 uint64_t reserved_64s[2]; /**< Reserved for future fields */
1274 void *reserved_ptrs[2]; /**< Reserved for future fields */
1279 * @b EXPERIMENTAL: this API may change, or be removed, without prior notice
1281 * A structure used to return the hairpin capabilities that are supported.
1283 struct rte_eth_hairpin_cap {
1284 /** The max number of hairpin queues (different bindings). */
1285 uint16_t max_nb_queues;
1286 /** Max number of Rx queues to be connected to one Tx queue. */
1287 uint16_t max_rx_2_tx;
1288 /** Max number of Tx queues to be connected to one Rx queue. */
1289 uint16_t max_tx_2_rx;
1290 uint16_t max_nb_desc; /**< The max num of descriptors. */
1293 #define RTE_ETH_MAX_HAIRPIN_PEERS 32
1297 * @b EXPERIMENTAL: this API may change, or be removed, without prior notice
1299 * A structure used to hold hairpin peer data.
1301 struct rte_eth_hairpin_peer {
1302 uint16_t port; /**< Peer port. */
1303 uint16_t queue; /**< Peer queue. */
1308 * @b EXPERIMENTAL: this API may change, or be removed, without prior notice
1310 * A structure used to configure hairpin binding.
1312 struct rte_eth_hairpin_conf {
1313 uint32_t peer_count:16; /**< The number of peers. */
1316 * Explicit Tx flow rule mode.
1317 * One hairpin pair of queues should have the same attribute.
1319 * - When set, the user should be responsible for inserting the hairpin
1320 * Tx part flows and removing them.
1321 * - When clear, the PMD will try to handle the Tx part of the flows,
1322 * e.g., by splitting one flow into two parts.
1324 uint32_t tx_explicit:1;
1327 * Manually bind hairpin queues.
1328 * One hairpin pair of queues should have the same attribute.
1330 * - When set, to enable hairpin, the user should call the hairpin bind
1331 * function after all the queues are set up properly and the ports are
1332 * started. Also, the hairpin unbind function should be called
1333 * accordingly before stopping a port that with hairpin configured.
1334 * - When clear, the PMD will try to enable the hairpin with the queues
1335 * configured automatically during port start.
1337 uint32_t manual_bind:1;
1338 uint32_t reserved:14; /**< Reserved bits. */
1339 struct rte_eth_hairpin_peer peers[RTE_ETH_MAX_HAIRPIN_PEERS];
1343 * A structure contains information about HW descriptor ring limitations.
1345 struct rte_eth_desc_lim {
1346 uint16_t nb_max; /**< Max allowed number of descriptors. */
1347 uint16_t nb_min; /**< Min allowed number of descriptors. */
1348 uint16_t nb_align; /**< Number of descriptors should be aligned to. */
1351 * Max allowed number of segments per whole packet.
1353 * - For TSO packet this is the total number of data descriptors allowed
1356 * @see nb_mtu_seg_max
1358 uint16_t nb_seg_max;
1361 * Max number of segments per one MTU.
1363 * - For non-TSO packet, this is the maximum allowed number of segments
1364 * in a single transmit packet.
1366 * - For TSO packet each segment within the TSO may span up to this
1371 uint16_t nb_mtu_seg_max;
1375 * This enum indicates the flow control mode
1377 enum rte_eth_fc_mode {
1378 RTE_ETH_FC_NONE = 0, /**< Disable flow control. */
1379 RTE_ETH_FC_RX_PAUSE, /**< Rx pause frame, enable flowctrl on Tx side. */
1380 RTE_ETH_FC_TX_PAUSE, /**< Tx pause frame, enable flowctrl on Rx side. */
1381 RTE_ETH_FC_FULL /**< Enable flow control on both side. */
1383 #define RTE_FC_NONE RTE_DEPRECATED(RTE_FC_NONE) RTE_ETH_FC_NONE
1384 #define RTE_FC_RX_PAUSE RTE_DEPRECATED(RTE_FC_RX_PAUSE) RTE_ETH_FC_RX_PAUSE
1385 #define RTE_FC_TX_PAUSE RTE_DEPRECATED(RTE_FC_TX_PAUSE) RTE_ETH_FC_TX_PAUSE
1386 #define RTE_FC_FULL RTE_DEPRECATED(RTE_FC_FULL) RTE_ETH_FC_FULL
1389 * A structure used to configure Ethernet flow control parameter.
1390 * These parameters will be configured into the register of the NIC.
1391 * Please refer to the corresponding data sheet for proper value.
1393 struct rte_eth_fc_conf {
1394 uint32_t high_water; /**< High threshold value to trigger XOFF */
1395 uint32_t low_water; /**< Low threshold value to trigger XON */
1396 uint16_t pause_time; /**< Pause quota in the Pause frame */
1397 uint16_t send_xon; /**< Is XON frame need be sent */
1398 enum rte_eth_fc_mode mode; /**< Link flow control mode */
1399 uint8_t mac_ctrl_frame_fwd; /**< Forward MAC control frames */
1400 uint8_t autoneg; /**< Use Pause autoneg */
1404 * A structure used to configure Ethernet priority flow control parameter.
1405 * These parameters will be configured into the register of the NIC.
1406 * Please refer to the corresponding data sheet for proper value.
1408 struct rte_eth_pfc_conf {
1409 struct rte_eth_fc_conf fc; /**< General flow control parameter. */
1410 uint8_t priority; /**< VLAN User Priority. */
1415 * @b EXPERIMENTAL: this API may change, or be removed, without prior notice
1417 * A structure used to retrieve information of queue based PFC.
1419 struct rte_eth_pfc_queue_info {
1421 * Maximum supported traffic class as per PFC (802.1Qbb) specification.
1424 /** PFC queue mode capabilities. */
1425 enum rte_eth_fc_mode mode_capa;
1430 * @b EXPERIMENTAL: this API may change, or be removed, without prior notice
1432 * A structure used to configure Ethernet priority flow control parameters for
1435 * rte_eth_pfc_queue_conf::rx_pause structure shall be used to configure given
1436 * tx_qid with corresponding tc. When ethdev device receives PFC frame with
1437 * rte_eth_pfc_queue_conf::rx_pause::tc, traffic will be paused on
1438 * rte_eth_pfc_queue_conf::rx_pause::tx_qid for that tc.
1440 * rte_eth_pfc_queue_conf::tx_pause structure shall be used to configure given
1441 * rx_qid. When rx_qid is congested, PFC frames are generated with
1442 * rte_eth_pfc_queue_conf::rx_pause::tc and
1443 * rte_eth_pfc_queue_conf::rx_pause::pause_time to the peer.
1445 struct rte_eth_pfc_queue_conf {
1446 enum rte_eth_fc_mode mode; /**< Link flow control mode */
1449 uint16_t tx_qid; /**< Tx queue ID */
1450 /** Traffic class as per PFC (802.1Qbb) spec. The value must be
1451 * in the range [0, rte_eth_pfc_queue_info::tx_max - 1]
1454 } rx_pause; /* Valid when (mode == FC_RX_PAUSE || mode == FC_FULL) */
1457 uint16_t pause_time; /**< Pause quota in the Pause frame */
1458 uint16_t rx_qid; /**< Rx queue ID */
1459 /** Traffic class as per PFC (802.1Qbb) spec. The value must be
1460 * in the range [0, rte_eth_pfc_queue_info::tx_max - 1]
1463 } tx_pause; /* Valid when (mode == FC_TX_PAUSE || mode == FC_FULL) */
1467 * Tunnel type for device-specific classifier configuration.
1468 * @see rte_eth_udp_tunnel
1470 enum rte_eth_tunnel_type {
1471 RTE_ETH_TUNNEL_TYPE_NONE = 0,
1472 RTE_ETH_TUNNEL_TYPE_VXLAN,
1473 RTE_ETH_TUNNEL_TYPE_GENEVE,
1474 RTE_ETH_TUNNEL_TYPE_TEREDO,
1475 RTE_ETH_TUNNEL_TYPE_NVGRE,
1476 RTE_ETH_TUNNEL_TYPE_IP_IN_GRE,
1477 RTE_ETH_L2_TUNNEL_TYPE_E_TAG,
1478 RTE_ETH_TUNNEL_TYPE_VXLAN_GPE,
1479 RTE_ETH_TUNNEL_TYPE_ECPRI,
1480 RTE_ETH_TUNNEL_TYPE_MAX,
1482 #define RTE_TUNNEL_TYPE_NONE RTE_DEPRECATED(RTE_TUNNEL_TYPE_NONE) RTE_ETH_TUNNEL_TYPE_NONE
1483 #define RTE_TUNNEL_TYPE_VXLAN RTE_DEPRECATED(RTE_TUNNEL_TYPE_VXLAN) RTE_ETH_TUNNEL_TYPE_VXLAN
1484 #define RTE_TUNNEL_TYPE_GENEVE RTE_DEPRECATED(RTE_TUNNEL_TYPE_GENEVE) RTE_ETH_TUNNEL_TYPE_GENEVE
1485 #define RTE_TUNNEL_TYPE_TEREDO RTE_DEPRECATED(RTE_TUNNEL_TYPE_TEREDO) RTE_ETH_TUNNEL_TYPE_TEREDO
1486 #define RTE_TUNNEL_TYPE_NVGRE RTE_DEPRECATED(RTE_TUNNEL_TYPE_NVGRE) RTE_ETH_TUNNEL_TYPE_NVGRE
1487 #define RTE_TUNNEL_TYPE_IP_IN_GRE RTE_DEPRECATED(RTE_TUNNEL_TYPE_IP_IN_GRE) RTE_ETH_TUNNEL_TYPE_IP_IN_GRE
1488 #define RTE_L2_TUNNEL_TYPE_E_TAG RTE_DEPRECATED(RTE_L2_TUNNEL_TYPE_E_TAG) RTE_ETH_L2_TUNNEL_TYPE_E_TAG
1489 #define RTE_TUNNEL_TYPE_VXLAN_GPE RTE_DEPRECATED(RTE_TUNNEL_TYPE_VXLAN_GPE) RTE_ETH_TUNNEL_TYPE_VXLAN_GPE
1490 #define RTE_TUNNEL_TYPE_ECPRI RTE_DEPRECATED(RTE_TUNNEL_TYPE_ECPRI) RTE_ETH_TUNNEL_TYPE_ECPRI
1491 #define RTE_TUNNEL_TYPE_MAX RTE_DEPRECATED(RTE_TUNNEL_TYPE_MAX) RTE_ETH_TUNNEL_TYPE_MAX
1493 /* Deprecated API file for rte_eth_dev_filter_* functions */
1494 #include "rte_eth_ctrl.h"
1497 * Memory space that can be configured to store Flow Director filters
1498 * in the board memory.
1500 enum rte_eth_fdir_pballoc_type {
1501 RTE_ETH_FDIR_PBALLOC_64K = 0, /**< 64k. */
1502 RTE_ETH_FDIR_PBALLOC_128K, /**< 128k. */
1503 RTE_ETH_FDIR_PBALLOC_256K, /**< 256k. */
1505 #define rte_fdir_pballoc_type rte_eth_fdir_pballoc_type
1507 #define RTE_FDIR_PBALLOC_64K RTE_DEPRECATED(RTE_FDIR_PBALLOC_64K) RTE_ETH_FDIR_PBALLOC_64K
1508 #define RTE_FDIR_PBALLOC_128K RTE_DEPRECATED(RTE_FDIR_PBALLOC_128K) RTE_ETH_FDIR_PBALLOC_128K
1509 #define RTE_FDIR_PBALLOC_256K RTE_DEPRECATED(RTE_FDIR_PBALLOC_256K) RTE_ETH_FDIR_PBALLOC_256K
1512 * Select report mode of FDIR hash information in Rx descriptors.
1514 enum rte_fdir_status_mode {
1515 RTE_FDIR_NO_REPORT_STATUS = 0, /**< Never report FDIR hash. */
1516 RTE_FDIR_REPORT_STATUS, /**< Only report FDIR hash for matching pkts. */
1517 RTE_FDIR_REPORT_STATUS_ALWAYS, /**< Always report FDIR hash. */
1521 * A structure used to configure the Flow Director (FDIR) feature
1522 * of an Ethernet port.
1524 * If mode is RTE_FDIR_MODE_NONE, the pballoc value is ignored.
1526 struct rte_eth_fdir_conf {
1527 enum rte_fdir_mode mode; /**< Flow Director mode. */
1528 enum rte_eth_fdir_pballoc_type pballoc; /**< Space for FDIR filters. */
1529 enum rte_fdir_status_mode status; /**< How to report FDIR hash. */
1530 /** Rx queue of packets matching a "drop" filter in perfect mode. */
1532 struct rte_eth_fdir_masks mask;
1533 /** Flex payload configuration. */
1534 struct rte_eth_fdir_flex_conf flex_conf;
1536 #define rte_fdir_conf rte_eth_fdir_conf
1539 * UDP tunneling configuration.
1541 * Used to configure the classifier of a device,
1542 * associating an UDP port with a type of tunnel.
1544 * Some NICs may need such configuration to properly parse a tunnel
1545 * with any standard or custom UDP port.
1547 struct rte_eth_udp_tunnel {
1548 uint16_t udp_port; /**< UDP port used for the tunnel. */
1549 uint8_t prot_type; /**< Tunnel type. @see rte_eth_tunnel_type */
1553 * A structure used to enable/disable specific device interrupts.
1555 struct rte_eth_intr_conf {
1556 /** enable/disable lsc interrupt. 0 (default) - disable, 1 enable */
1558 /** enable/disable rxq interrupt. 0 (default) - disable, 1 enable */
1560 /** enable/disable rmv interrupt. 0 (default) - disable, 1 enable */
1564 #define rte_intr_conf rte_eth_intr_conf
1567 * A structure used to configure an Ethernet port.
1568 * Depending upon the Rx multi-queue mode, extra advanced
1569 * configuration settings may be needed.
1571 struct rte_eth_conf {
1572 uint32_t link_speeds; /**< bitmap of RTE_ETH_LINK_SPEED_XXX of speeds to be
1573 used. RTE_ETH_LINK_SPEED_FIXED disables link
1574 autonegotiation, and a unique speed shall be
1575 set. Otherwise, the bitmap defines the set of
1576 speeds to be advertised. If the special value
1577 RTE_ETH_LINK_SPEED_AUTONEG (0) is used, all speeds
1578 supported are advertised. */
1579 struct rte_eth_rxmode rxmode; /**< Port Rx configuration. */
1580 struct rte_eth_txmode txmode; /**< Port Tx configuration. */
1581 uint32_t lpbk_mode; /**< Loopback operation mode. By default the value
1582 is 0, meaning the loopback mode is disabled.
1583 Read the datasheet of given Ethernet controller
1584 for details. The possible values of this field
1585 are defined in implementation of each driver. */
1587 struct rte_eth_rss_conf rss_conf; /**< Port RSS configuration */
1588 /** Port VMDq+DCB configuration. */
1589 struct rte_eth_vmdq_dcb_conf vmdq_dcb_conf;
1590 /** Port DCB Rx configuration. */
1591 struct rte_eth_dcb_rx_conf dcb_rx_conf;
1592 /** Port VMDq Rx configuration. */
1593 struct rte_eth_vmdq_rx_conf vmdq_rx_conf;
1594 } rx_adv_conf; /**< Port Rx filtering configuration. */
1596 /** Port VMDq+DCB Tx configuration. */
1597 struct rte_eth_vmdq_dcb_tx_conf vmdq_dcb_tx_conf;
1598 /** Port DCB Tx configuration. */
1599 struct rte_eth_dcb_tx_conf dcb_tx_conf;
1600 /** Port VMDq Tx configuration. */
1601 struct rte_eth_vmdq_tx_conf vmdq_tx_conf;
1602 } tx_adv_conf; /**< Port Tx DCB configuration (union). */
1603 /** Currently,Priority Flow Control(PFC) are supported,if DCB with PFC
1604 is needed,and the variable must be set RTE_ETH_DCB_PFC_SUPPORT. */
1605 uint32_t dcb_capability_en;
1606 struct rte_eth_fdir_conf fdir_conf; /**< FDIR configuration. DEPRECATED */
1607 struct rte_eth_intr_conf intr_conf; /**< Interrupt mode configuration. */
1611 * Rx offload capabilities of a device.
1613 #define RTE_ETH_RX_OFFLOAD_VLAN_STRIP RTE_BIT64(0)
1614 #define RTE_ETH_RX_OFFLOAD_IPV4_CKSUM RTE_BIT64(1)
1615 #define RTE_ETH_RX_OFFLOAD_UDP_CKSUM RTE_BIT64(2)
1616 #define RTE_ETH_RX_OFFLOAD_TCP_CKSUM RTE_BIT64(3)
1617 #define RTE_ETH_RX_OFFLOAD_TCP_LRO RTE_BIT64(4)
1618 #define RTE_ETH_RX_OFFLOAD_QINQ_STRIP RTE_BIT64(5)
1619 #define RTE_ETH_RX_OFFLOAD_OUTER_IPV4_CKSUM RTE_BIT64(6)
1620 #define RTE_ETH_RX_OFFLOAD_MACSEC_STRIP RTE_BIT64(7)
1621 #define RTE_ETH_RX_OFFLOAD_HEADER_SPLIT RTE_BIT64(8)
1622 #define RTE_ETH_RX_OFFLOAD_VLAN_FILTER RTE_BIT64(9)
1623 #define RTE_ETH_RX_OFFLOAD_VLAN_EXTEND RTE_BIT64(10)
1624 #define RTE_ETH_RX_OFFLOAD_SCATTER RTE_BIT64(13)
1626 * Timestamp is set by the driver in RTE_MBUF_DYNFIELD_TIMESTAMP_NAME
1627 * and RTE_MBUF_DYNFLAG_RX_TIMESTAMP_NAME is set in ol_flags.
1628 * The mbuf field and flag are registered when the offload is configured.
1630 #define RTE_ETH_RX_OFFLOAD_TIMESTAMP RTE_BIT64(14)
1631 #define RTE_ETH_RX_OFFLOAD_SECURITY RTE_BIT64(15)
1632 #define RTE_ETH_RX_OFFLOAD_KEEP_CRC RTE_BIT64(16)
1633 #define RTE_ETH_RX_OFFLOAD_SCTP_CKSUM RTE_BIT64(17)
1634 #define RTE_ETH_RX_OFFLOAD_OUTER_UDP_CKSUM RTE_BIT64(18)
1635 #define RTE_ETH_RX_OFFLOAD_RSS_HASH RTE_BIT64(19)
1636 #define RTE_ETH_RX_OFFLOAD_BUFFER_SPLIT RTE_BIT64(20)
1638 #define DEV_RX_OFFLOAD_VLAN_STRIP RTE_DEPRECATED(DEV_RX_OFFLOAD_VLAN_STRIP) RTE_ETH_RX_OFFLOAD_VLAN_STRIP
1639 #define DEV_RX_OFFLOAD_IPV4_CKSUM RTE_DEPRECATED(DEV_RX_OFFLOAD_IPV4_CKSUM) RTE_ETH_RX_OFFLOAD_IPV4_CKSUM
1640 #define DEV_RX_OFFLOAD_UDP_CKSUM RTE_DEPRECATED(DEV_RX_OFFLOAD_UDP_CKSUM) RTE_ETH_RX_OFFLOAD_UDP_CKSUM
1641 #define DEV_RX_OFFLOAD_TCP_CKSUM RTE_DEPRECATED(DEV_RX_OFFLOAD_TCP_CKSUM) RTE_ETH_RX_OFFLOAD_TCP_CKSUM
1642 #define DEV_RX_OFFLOAD_TCP_LRO RTE_DEPRECATED(DEV_RX_OFFLOAD_TCP_LRO) RTE_ETH_RX_OFFLOAD_TCP_LRO
1643 #define DEV_RX_OFFLOAD_QINQ_STRIP RTE_DEPRECATED(DEV_RX_OFFLOAD_QINQ_STRIP) RTE_ETH_RX_OFFLOAD_QINQ_STRIP
1644 #define DEV_RX_OFFLOAD_OUTER_IPV4_CKSUM RTE_DEPRECATED(DEV_RX_OFFLOAD_OUTER_IPV4_CKSUM) RTE_ETH_RX_OFFLOAD_OUTER_IPV4_CKSUM
1645 #define DEV_RX_OFFLOAD_MACSEC_STRIP RTE_DEPRECATED(DEV_RX_OFFLOAD_MACSEC_STRIP) RTE_ETH_RX_OFFLOAD_MACSEC_STRIP
1646 #define DEV_RX_OFFLOAD_HEADER_SPLIT RTE_DEPRECATED(DEV_RX_OFFLOAD_HEADER_SPLIT) RTE_ETH_RX_OFFLOAD_HEADER_SPLIT
1647 #define DEV_RX_OFFLOAD_VLAN_FILTER RTE_DEPRECATED(DEV_RX_OFFLOAD_VLAN_FILTER) RTE_ETH_RX_OFFLOAD_VLAN_FILTER
1648 #define DEV_RX_OFFLOAD_VLAN_EXTEND RTE_DEPRECATED(DEV_RX_OFFLOAD_VLAN_EXTEND) RTE_ETH_RX_OFFLOAD_VLAN_EXTEND
1649 #define DEV_RX_OFFLOAD_SCATTER RTE_DEPRECATED(DEV_RX_OFFLOAD_SCATTER) RTE_ETH_RX_OFFLOAD_SCATTER
1650 #define DEV_RX_OFFLOAD_TIMESTAMP RTE_DEPRECATED(DEV_RX_OFFLOAD_TIMESTAMP) RTE_ETH_RX_OFFLOAD_TIMESTAMP
1651 #define DEV_RX_OFFLOAD_SECURITY RTE_DEPRECATED(DEV_RX_OFFLOAD_SECURITY) RTE_ETH_RX_OFFLOAD_SECURITY
1652 #define DEV_RX_OFFLOAD_KEEP_CRC RTE_DEPRECATED(DEV_RX_OFFLOAD_KEEP_CRC) RTE_ETH_RX_OFFLOAD_KEEP_CRC
1653 #define DEV_RX_OFFLOAD_SCTP_CKSUM RTE_DEPRECATED(DEV_RX_OFFLOAD_SCTP_CKSUM) RTE_ETH_RX_OFFLOAD_SCTP_CKSUM
1654 #define DEV_RX_OFFLOAD_OUTER_UDP_CKSUM RTE_DEPRECATED(DEV_RX_OFFLOAD_OUTER_UDP_CKSUM) RTE_ETH_RX_OFFLOAD_OUTER_UDP_CKSUM
1655 #define DEV_RX_OFFLOAD_RSS_HASH RTE_DEPRECATED(DEV_RX_OFFLOAD_RSS_HASH) RTE_ETH_RX_OFFLOAD_RSS_HASH
1657 #define RTE_ETH_RX_OFFLOAD_CHECKSUM (RTE_ETH_RX_OFFLOAD_IPV4_CKSUM | \
1658 RTE_ETH_RX_OFFLOAD_UDP_CKSUM | \
1659 RTE_ETH_RX_OFFLOAD_TCP_CKSUM)
1660 #define DEV_RX_OFFLOAD_CHECKSUM RTE_DEPRECATED(DEV_RX_OFFLOAD_CHECKSUM) RTE_ETH_RX_OFFLOAD_CHECKSUM
1661 #define RTE_ETH_RX_OFFLOAD_VLAN (RTE_ETH_RX_OFFLOAD_VLAN_STRIP | \
1662 RTE_ETH_RX_OFFLOAD_VLAN_FILTER | \
1663 RTE_ETH_RX_OFFLOAD_VLAN_EXTEND | \
1664 RTE_ETH_RX_OFFLOAD_QINQ_STRIP)
1665 #define DEV_RX_OFFLOAD_VLAN RTE_DEPRECATED(DEV_RX_OFFLOAD_VLAN) RTE_ETH_RX_OFFLOAD_VLAN
1668 * If new Rx offload capabilities are defined, they also must be
1669 * mentioned in rte_rx_offload_names in rte_ethdev.c file.
1673 * Tx offload capabilities of a device.
1675 #define RTE_ETH_TX_OFFLOAD_VLAN_INSERT RTE_BIT64(0)
1676 #define RTE_ETH_TX_OFFLOAD_IPV4_CKSUM RTE_BIT64(1)
1677 #define RTE_ETH_TX_OFFLOAD_UDP_CKSUM RTE_BIT64(2)
1678 #define RTE_ETH_TX_OFFLOAD_TCP_CKSUM RTE_BIT64(3)
1679 #define RTE_ETH_TX_OFFLOAD_SCTP_CKSUM RTE_BIT64(4)
1680 #define RTE_ETH_TX_OFFLOAD_TCP_TSO RTE_BIT64(5)
1681 #define RTE_ETH_TX_OFFLOAD_UDP_TSO RTE_BIT64(6)
1682 #define RTE_ETH_TX_OFFLOAD_OUTER_IPV4_CKSUM RTE_BIT64(7) /**< Used for tunneling packet. */
1683 #define RTE_ETH_TX_OFFLOAD_QINQ_INSERT RTE_BIT64(8)
1684 #define RTE_ETH_TX_OFFLOAD_VXLAN_TNL_TSO RTE_BIT64(9) /**< Used for tunneling packet. */
1685 #define RTE_ETH_TX_OFFLOAD_GRE_TNL_TSO RTE_BIT64(10) /**< Used for tunneling packet. */
1686 #define RTE_ETH_TX_OFFLOAD_IPIP_TNL_TSO RTE_BIT64(11) /**< Used for tunneling packet. */
1687 #define RTE_ETH_TX_OFFLOAD_GENEVE_TNL_TSO RTE_BIT64(12) /**< Used for tunneling packet. */
1688 #define RTE_ETH_TX_OFFLOAD_MACSEC_INSERT RTE_BIT64(13)
1690 * Multiple threads can invoke rte_eth_tx_burst() concurrently on the same
1691 * Tx queue without SW lock.
1693 #define RTE_ETH_TX_OFFLOAD_MT_LOCKFREE RTE_BIT64(14)
1694 /** Device supports multi segment send. */
1695 #define RTE_ETH_TX_OFFLOAD_MULTI_SEGS RTE_BIT64(15)
1697 * Device supports optimization for fast release of mbufs.
1698 * When set application must guarantee that per-queue all mbufs comes from
1699 * the same mempool and has refcnt = 1.
1701 #define RTE_ETH_TX_OFFLOAD_MBUF_FAST_FREE RTE_BIT64(16)
1702 #define RTE_ETH_TX_OFFLOAD_SECURITY RTE_BIT64(17)
1704 * Device supports generic UDP tunneled packet TSO.
1705 * Application must set RTE_MBUF_F_TX_TUNNEL_UDP and other mbuf fields required
1708 #define RTE_ETH_TX_OFFLOAD_UDP_TNL_TSO RTE_BIT64(18)
1710 * Device supports generic IP tunneled packet TSO.
1711 * Application must set RTE_MBUF_F_TX_TUNNEL_IP and other mbuf fields required
1714 #define RTE_ETH_TX_OFFLOAD_IP_TNL_TSO RTE_BIT64(19)
1715 /** Device supports outer UDP checksum */
1716 #define RTE_ETH_TX_OFFLOAD_OUTER_UDP_CKSUM RTE_BIT64(20)
1718 * Device sends on time read from RTE_MBUF_DYNFIELD_TIMESTAMP_NAME
1719 * if RTE_MBUF_DYNFLAG_TX_TIMESTAMP_NAME is set in ol_flags.
1720 * The mbuf field and flag are registered when the offload is configured.
1722 #define RTE_ETH_TX_OFFLOAD_SEND_ON_TIMESTAMP RTE_BIT64(21)
1724 * If new Tx offload capabilities are defined, they also must be
1725 * mentioned in rte_tx_offload_names in rte_ethdev.c file.
1728 #define DEV_TX_OFFLOAD_VLAN_INSERT RTE_DEPRECATED(DEV_TX_OFFLOAD_VLAN_INSERT) RTE_ETH_TX_OFFLOAD_VLAN_INSERT
1729 #define DEV_TX_OFFLOAD_IPV4_CKSUM RTE_DEPRECATED(DEV_TX_OFFLOAD_IPV4_CKSUM) RTE_ETH_TX_OFFLOAD_IPV4_CKSUM
1730 #define DEV_TX_OFFLOAD_UDP_CKSUM RTE_DEPRECATED(DEV_TX_OFFLOAD_UDP_CKSUM) RTE_ETH_TX_OFFLOAD_UDP_CKSUM
1731 #define DEV_TX_OFFLOAD_TCP_CKSUM RTE_DEPRECATED(DEV_TX_OFFLOAD_TCP_CKSUM) RTE_ETH_TX_OFFLOAD_TCP_CKSUM
1732 #define DEV_TX_OFFLOAD_SCTP_CKSUM RTE_DEPRECATED(DEV_TX_OFFLOAD_SCTP_CKSUM) RTE_ETH_TX_OFFLOAD_SCTP_CKSUM
1733 #define DEV_TX_OFFLOAD_TCP_TSO RTE_DEPRECATED(DEV_TX_OFFLOAD_TCP_TSO) RTE_ETH_TX_OFFLOAD_TCP_TSO
1734 #define DEV_TX_OFFLOAD_UDP_TSO RTE_DEPRECATED(DEV_TX_OFFLOAD_UDP_TSO) RTE_ETH_TX_OFFLOAD_UDP_TSO
1735 #define DEV_TX_OFFLOAD_OUTER_IPV4_CKSUM RTE_DEPRECATED(DEV_TX_OFFLOAD_OUTER_IPV4_CKSUM) RTE_ETH_TX_OFFLOAD_OUTER_IPV4_CKSUM
1736 #define DEV_TX_OFFLOAD_QINQ_INSERT RTE_DEPRECATED(DEV_TX_OFFLOAD_QINQ_INSERT) RTE_ETH_TX_OFFLOAD_QINQ_INSERT
1737 #define DEV_TX_OFFLOAD_VXLAN_TNL_TSO RTE_DEPRECATED(DEV_TX_OFFLOAD_VXLAN_TNL_TSO) RTE_ETH_TX_OFFLOAD_VXLAN_TNL_TSO
1738 #define DEV_TX_OFFLOAD_GRE_TNL_TSO RTE_DEPRECATED(DEV_TX_OFFLOAD_GRE_TNL_TSO) RTE_ETH_TX_OFFLOAD_GRE_TNL_TSO
1739 #define DEV_TX_OFFLOAD_IPIP_TNL_TSO RTE_DEPRECATED(DEV_TX_OFFLOAD_IPIP_TNL_TSO) RTE_ETH_TX_OFFLOAD_IPIP_TNL_TSO
1740 #define DEV_TX_OFFLOAD_GENEVE_TNL_TSO RTE_DEPRECATED(DEV_TX_OFFLOAD_GENEVE_TNL_TSO) RTE_ETH_TX_OFFLOAD_GENEVE_TNL_TSO
1741 #define DEV_TX_OFFLOAD_MACSEC_INSERT RTE_DEPRECATED(DEV_TX_OFFLOAD_MACSEC_INSERT) RTE_ETH_TX_OFFLOAD_MACSEC_INSERT
1742 #define DEV_TX_OFFLOAD_MT_LOCKFREE RTE_DEPRECATED(DEV_TX_OFFLOAD_MT_LOCKFREE) RTE_ETH_TX_OFFLOAD_MT_LOCKFREE
1743 #define DEV_TX_OFFLOAD_MULTI_SEGS RTE_DEPRECATED(DEV_TX_OFFLOAD_MULTI_SEGS) RTE_ETH_TX_OFFLOAD_MULTI_SEGS
1744 #define DEV_TX_OFFLOAD_MBUF_FAST_FREE RTE_DEPRECATED(DEV_TX_OFFLOAD_MBUF_FAST_FREE) RTE_ETH_TX_OFFLOAD_MBUF_FAST_FREE
1745 #define DEV_TX_OFFLOAD_SECURITY RTE_DEPRECATED(DEV_TX_OFFLOAD_SECURITY) RTE_ETH_TX_OFFLOAD_SECURITY
1746 #define DEV_TX_OFFLOAD_UDP_TNL_TSO RTE_DEPRECATED(DEV_TX_OFFLOAD_UDP_TNL_TSO) RTE_ETH_TX_OFFLOAD_UDP_TNL_TSO
1747 #define DEV_TX_OFFLOAD_IP_TNL_TSO RTE_DEPRECATED(DEV_TX_OFFLOAD_IP_TNL_TSO) RTE_ETH_TX_OFFLOAD_IP_TNL_TSO
1748 #define DEV_TX_OFFLOAD_OUTER_UDP_CKSUM RTE_DEPRECATED(DEV_TX_OFFLOAD_OUTER_UDP_CKSUM) RTE_ETH_TX_OFFLOAD_OUTER_UDP_CKSUM
1749 #define DEV_TX_OFFLOAD_SEND_ON_TIMESTAMP RTE_DEPRECATED(DEV_TX_OFFLOAD_SEND_ON_TIMESTAMP) RTE_ETH_TX_OFFLOAD_SEND_ON_TIMESTAMP
1751 /**@{@name Device capabilities
1752 * Non-offload capabilities reported in rte_eth_dev_info.dev_capa.
1754 /** Device supports Rx queue setup after device started. */
1755 #define RTE_ETH_DEV_CAPA_RUNTIME_RX_QUEUE_SETUP RTE_BIT64(0)
1756 /** Device supports Tx queue setup after device started. */
1757 #define RTE_ETH_DEV_CAPA_RUNTIME_TX_QUEUE_SETUP RTE_BIT64(1)
1759 * Device supports shared Rx queue among ports within Rx domain and
1760 * switch domain. Mbufs are consumed by shared Rx queue instead of
1761 * each queue. Multiple groups are supported by share_group of Rx
1762 * queue configuration. Shared Rx queue is identified by PMD using
1763 * share_qid of Rx queue configuration. Polling any port in the group
1764 * receive packets of all member ports, source port identified by
1767 #define RTE_ETH_DEV_CAPA_RXQ_SHARE RTE_BIT64(2)
1768 /** Device supports keeping flow rules across restart. */
1769 #define RTE_ETH_DEV_CAPA_FLOW_RULE_KEEP RTE_BIT64(3)
1770 /** Device supports keeping shared flow objects across restart. */
1771 #define RTE_ETH_DEV_CAPA_FLOW_SHARED_OBJECT_KEEP RTE_BIT64(4)
1775 * Fallback default preferred Rx/Tx port parameters.
1776 * These are used if an application requests default parameters
1777 * but the PMD does not provide preferred values.
1779 #define RTE_ETH_DEV_FALLBACK_RX_RINGSIZE 512
1780 #define RTE_ETH_DEV_FALLBACK_TX_RINGSIZE 512
1781 #define RTE_ETH_DEV_FALLBACK_RX_NBQUEUES 1
1782 #define RTE_ETH_DEV_FALLBACK_TX_NBQUEUES 1
1785 * Preferred Rx/Tx port parameters.
1786 * There are separate instances of this structure for transmission
1787 * and reception respectively.
1789 struct rte_eth_dev_portconf {
1790 uint16_t burst_size; /**< Device-preferred burst size */
1791 uint16_t ring_size; /**< Device-preferred size of queue rings */
1792 uint16_t nb_queues; /**< Device-preferred number of queues */
1796 * Default values for switch domain ID when ethdev does not support switch
1797 * domain definitions.
1799 #define RTE_ETH_DEV_SWITCH_DOMAIN_ID_INVALID (UINT16_MAX)
1802 * Ethernet device associated switch information
1804 struct rte_eth_switch_info {
1805 const char *name; /**< switch name */
1806 uint16_t domain_id; /**< switch domain ID */
1808 * Mapping to the devices physical switch port as enumerated from the
1809 * perspective of the embedded interconnect/switch. For SR-IOV enabled
1810 * device this may correspond to the VF_ID of each virtual function,
1811 * but each driver should explicitly define the mapping of switch
1812 * port identifier to that physical interconnect/switch
1816 * Shared Rx queue sub-domain boundary. Only ports in same Rx domain
1817 * and switch domain can share Rx queue. Valid only if device advertised
1818 * RTE_ETH_DEV_CAPA_RXQ_SHARE capability.
1825 * @b EXPERIMENTAL: this structure may change without prior notice.
1827 * Ethernet device Rx buffer segmentation capabilities.
1829 struct rte_eth_rxseg_capa {
1831 uint32_t multi_pools:1; /**< Supports receiving to multiple pools.*/
1832 uint32_t offset_allowed:1; /**< Supports buffer offsets. */
1833 uint32_t offset_align_log2:4; /**< Required offset alignment. */
1834 uint16_t max_nseg; /**< Maximum amount of segments to split. */
1835 uint16_t reserved; /**< Reserved field. */
1839 * Ethernet device information
1843 * Ethernet device representor port type.
1845 enum rte_eth_representor_type {
1846 RTE_ETH_REPRESENTOR_NONE, /**< not a representor. */
1847 RTE_ETH_REPRESENTOR_VF, /**< representor of Virtual Function. */
1848 RTE_ETH_REPRESENTOR_SF, /**< representor of Sub Function. */
1849 RTE_ETH_REPRESENTOR_PF, /**< representor of Physical Function. */
1853 * A structure used to retrieve the contextual information of
1854 * an Ethernet device, such as the controlling driver of the
1857 struct rte_eth_dev_info {
1858 struct rte_device *device; /**< Generic device information */
1859 const char *driver_name; /**< Device Driver name. */
1860 unsigned int if_index; /**< Index to bound host interface, or 0 if none.
1861 Use if_indextoname() to translate into an interface name. */
1862 uint16_t min_mtu; /**< Minimum MTU allowed */
1863 uint16_t max_mtu; /**< Maximum MTU allowed */
1864 const uint32_t *dev_flags; /**< Device flags */
1865 uint32_t min_rx_bufsize; /**< Minimum size of Rx buffer. */
1866 uint32_t max_rx_pktlen; /**< Maximum configurable length of Rx pkt. */
1867 /** Maximum configurable size of LRO aggregated packet. */
1868 uint32_t max_lro_pkt_size;
1869 uint16_t max_rx_queues; /**< Maximum number of Rx queues. */
1870 uint16_t max_tx_queues; /**< Maximum number of Tx queues. */
1871 uint32_t max_mac_addrs; /**< Maximum number of MAC addresses. */
1872 /** Maximum number of hash MAC addresses for MTA and UTA. */
1873 uint32_t max_hash_mac_addrs;
1874 uint16_t max_vfs; /**< Maximum number of VFs. */
1875 uint16_t max_vmdq_pools; /**< Maximum number of VMDq pools. */
1876 struct rte_eth_rxseg_capa rx_seg_capa; /**< Segmentation capability.*/
1877 /** All Rx offload capabilities including all per-queue ones */
1878 uint64_t rx_offload_capa;
1879 /** All Tx offload capabilities including all per-queue ones */
1880 uint64_t tx_offload_capa;
1881 /** Device per-queue Rx offload capabilities. */
1882 uint64_t rx_queue_offload_capa;
1883 /** Device per-queue Tx offload capabilities. */
1884 uint64_t tx_queue_offload_capa;
1885 /** Device redirection table size, the total number of entries. */
1887 uint8_t hash_key_size; /**< Hash key size in bytes */
1888 /** Bit mask of RSS offloads, the bit offset also means flow type */
1889 uint64_t flow_type_rss_offloads;
1890 struct rte_eth_rxconf default_rxconf; /**< Default Rx configuration */
1891 struct rte_eth_txconf default_txconf; /**< Default Tx configuration */
1892 uint16_t vmdq_queue_base; /**< First queue ID for VMDq pools. */
1893 uint16_t vmdq_queue_num; /**< Queue number for VMDq pools. */
1894 uint16_t vmdq_pool_base; /**< First ID of VMDq pools. */
1895 struct rte_eth_desc_lim rx_desc_lim; /**< Rx descriptors limits */
1896 struct rte_eth_desc_lim tx_desc_lim; /**< Tx descriptors limits */
1897 uint32_t speed_capa; /**< Supported speeds bitmap (RTE_ETH_LINK_SPEED_). */
1898 /** Configured number of Rx/Tx queues */
1899 uint16_t nb_rx_queues; /**< Number of Rx queues. */
1900 uint16_t nb_tx_queues; /**< Number of Tx queues. */
1901 /** Rx parameter recommendations */
1902 struct rte_eth_dev_portconf default_rxportconf;
1903 /** Tx parameter recommendations */
1904 struct rte_eth_dev_portconf default_txportconf;
1905 /** Generic device capabilities (RTE_ETH_DEV_CAPA_). */
1908 * Switching information for ports on a device with a
1909 * embedded managed interconnect/switch.
1911 struct rte_eth_switch_info switch_info;
1913 uint64_t reserved_64s[2]; /**< Reserved for future fields */
1914 void *reserved_ptrs[2]; /**< Reserved for future fields */
1917 /**@{@name Rx/Tx queue states */
1918 #define RTE_ETH_QUEUE_STATE_STOPPED 0 /**< Queue stopped. */
1919 #define RTE_ETH_QUEUE_STATE_STARTED 1 /**< Queue started. */
1920 #define RTE_ETH_QUEUE_STATE_HAIRPIN 2 /**< Queue used for hairpin. */
1924 * Ethernet device Rx queue information structure.
1925 * Used to retrieve information about configured queue.
1927 struct rte_eth_rxq_info {
1928 struct rte_mempool *mp; /**< mempool used by that queue. */
1929 struct rte_eth_rxconf conf; /**< queue config parameters. */
1930 uint8_t scattered_rx; /**< scattered packets Rx supported. */
1931 uint8_t queue_state; /**< one of RTE_ETH_QUEUE_STATE_*. */
1932 uint16_t nb_desc; /**< configured number of RXDs. */
1933 uint16_t rx_buf_size; /**< hardware receive buffer size. */
1934 } __rte_cache_min_aligned;
1937 * Ethernet device Tx queue information structure.
1938 * Used to retrieve information about configured queue.
1940 struct rte_eth_txq_info {
1941 struct rte_eth_txconf conf; /**< queue config parameters. */
1942 uint16_t nb_desc; /**< configured number of TXDs. */
1943 uint8_t queue_state; /**< one of RTE_ETH_QUEUE_STATE_*. */
1944 } __rte_cache_min_aligned;
1946 /* Generic Burst mode flag definition, values can be ORed. */
1949 * If the queues have different burst mode description, this bit will be set
1950 * by PMD, then the application can iterate to retrieve burst description for
1953 #define RTE_ETH_BURST_FLAG_PER_QUEUE RTE_BIT64(0)
1956 * Ethernet device Rx/Tx queue packet burst mode information structure.
1957 * Used to retrieve information about packet burst mode setting.
1959 struct rte_eth_burst_mode {
1960 uint64_t flags; /**< The ORed values of RTE_ETH_BURST_FLAG_xxx */
1962 #define RTE_ETH_BURST_MODE_INFO_SIZE 1024 /**< Maximum size for information */
1963 char info[RTE_ETH_BURST_MODE_INFO_SIZE]; /**< burst mode information */
1966 /** Maximum name length for extended statistics counters */
1967 #define RTE_ETH_XSTATS_NAME_SIZE 64
1970 * An Ethernet device extended statistic structure
1972 * This structure is used by rte_eth_xstats_get() to provide
1973 * statistics that are not provided in the generic *rte_eth_stats*
1975 * It maps a name ID, corresponding to an index in the array returned
1976 * by rte_eth_xstats_get_names(), to a statistic value.
1978 struct rte_eth_xstat {
1979 uint64_t id; /**< The index in xstats name array. */
1980 uint64_t value; /**< The statistic counter value. */
1984 * A name element for extended statistics.
1986 * An array of this structure is returned by rte_eth_xstats_get_names().
1987 * It lists the names of extended statistics for a PMD. The *rte_eth_xstat*
1988 * structure references these names by their array index.
1990 * The xstats should follow a common naming scheme.
1991 * Some names are standardized in rte_stats_strings.
1993 * - rx_missed_errors
1995 * - tx_size_128_to_255_packets
1997 struct rte_eth_xstat_name {
1998 char name[RTE_ETH_XSTATS_NAME_SIZE]; /**< The statistic name. */
2001 #define RTE_ETH_DCB_NUM_TCS 8
2002 #define RTE_ETH_MAX_VMDQ_POOL 64
2004 #define ETH_DCB_NUM_TCS RTE_DEPRECATED(ETH_DCB_NUM_TCS) RTE_ETH_DCB_NUM_TCS
2005 #define ETH_MAX_VMDQ_POOL RTE_DEPRECATED(ETH_MAX_VMDQ_POOL) RTE_ETH_MAX_VMDQ_POOL
2008 * A structure used to get the information of queue and
2009 * TC mapping on both Tx and Rx paths.
2011 struct rte_eth_dcb_tc_queue_mapping {
2012 /** Rx queues assigned to tc per Pool */
2016 } tc_rxq[RTE_ETH_MAX_VMDQ_POOL][RTE_ETH_DCB_NUM_TCS];
2017 /** Rx queues assigned to tc per Pool */
2021 } tc_txq[RTE_ETH_MAX_VMDQ_POOL][RTE_ETH_DCB_NUM_TCS];
2025 * A structure used to get the information of DCB.
2026 * It includes TC UP mapping and queue TC mapping.
2028 struct rte_eth_dcb_info {
2029 uint8_t nb_tcs; /**< number of TCs */
2030 uint8_t prio_tc[RTE_ETH_DCB_NUM_USER_PRIORITIES]; /**< Priority to tc */
2031 uint8_t tc_bws[RTE_ETH_DCB_NUM_TCS]; /**< Tx BW percentage for each TC */
2032 /** Rx queues assigned to tc */
2033 struct rte_eth_dcb_tc_queue_mapping tc_queue;
2037 * This enum indicates the possible Forward Error Correction (FEC) modes
2038 * of an ethdev port.
2040 enum rte_eth_fec_mode {
2041 RTE_ETH_FEC_NOFEC = 0, /**< FEC is off */
2042 RTE_ETH_FEC_AUTO, /**< FEC autonegotiation modes */
2043 RTE_ETH_FEC_BASER, /**< FEC using common algorithm */
2044 RTE_ETH_FEC_RS, /**< FEC using RS algorithm */
2047 /* Translate from FEC mode to FEC capa */
2048 #define RTE_ETH_FEC_MODE_TO_CAPA(x) RTE_BIT32(x)
2050 /* This macro indicates FEC capa mask */
2051 #define RTE_ETH_FEC_MODE_CAPA_MASK(x) RTE_BIT32(RTE_ETH_FEC_ ## x)
2053 /* A structure used to get capabilities per link speed */
2054 struct rte_eth_fec_capa {
2055 uint32_t speed; /**< Link speed (see RTE_ETH_SPEED_NUM_*) */
2056 uint32_t capa; /**< FEC capabilities bitmask */
2059 #define RTE_ETH_ALL RTE_MAX_ETHPORTS
2061 /* Macros to check for valid port */
2062 #define RTE_ETH_VALID_PORTID_OR_ERR_RET(port_id, retval) do { \
2063 if (!rte_eth_dev_is_valid_port(port_id)) { \
2064 RTE_ETHDEV_LOG(ERR, "Invalid port_id=%u\n", port_id); \
2069 #define RTE_ETH_VALID_PORTID_OR_RET(port_id) do { \
2070 if (!rte_eth_dev_is_valid_port(port_id)) { \
2071 RTE_ETHDEV_LOG(ERR, "Invalid port_id=%u\n", port_id); \
2077 * Function type used for Rx packet processing packet callbacks.
2079 * The callback function is called on Rx with a burst of packets that have
2080 * been received on the given port and queue.
2083 * The Ethernet port on which Rx is being performed.
2085 * The queue on the Ethernet port which is being used to receive the packets.
2087 * The burst of packets that have just been received.
2089 * The number of packets in the burst pointed to by "pkts".
2091 * The max number of packets that can be stored in the "pkts" array.
2093 * The arbitrary user parameter passed in by the application when the callback
2094 * was originally configured.
2096 * The number of packets returned to the user.
2098 typedef uint16_t (*rte_rx_callback_fn)(uint16_t port_id, uint16_t queue,
2099 struct rte_mbuf *pkts[], uint16_t nb_pkts, uint16_t max_pkts,
2103 * Function type used for Tx packet processing packet callbacks.
2105 * The callback function is called on Tx with a burst of packets immediately
2106 * before the packets are put onto the hardware queue for transmission.
2109 * The Ethernet port on which Tx is being performed.
2111 * The queue on the Ethernet port which is being used to transmit the packets.
2113 * The burst of packets that are about to be transmitted.
2115 * The number of packets in the burst pointed to by "pkts".
2117 * The arbitrary user parameter passed in by the application when the callback
2118 * was originally configured.
2120 * The number of packets to be written to the NIC.
2122 typedef uint16_t (*rte_tx_callback_fn)(uint16_t port_id, uint16_t queue,
2123 struct rte_mbuf *pkts[], uint16_t nb_pkts, void *user_param);
2126 * Possible states of an ethdev port.
2128 enum rte_eth_dev_state {
2129 /** Device is unused before being probed. */
2130 RTE_ETH_DEV_UNUSED = 0,
2131 /** Device is attached when allocated in probing. */
2132 RTE_ETH_DEV_ATTACHED,
2133 /** Device is in removed state when plug-out is detected. */
2134 RTE_ETH_DEV_REMOVED,
2137 struct rte_eth_dev_sriov {
2138 uint8_t active; /**< SRIOV is active with 16, 32 or 64 pools */
2139 uint8_t nb_q_per_pool; /**< Rx queue number per pool */
2140 uint16_t def_vmdq_idx; /**< Default pool num used for PF */
2141 uint16_t def_pool_q_idx; /**< Default pool queue start reg index */
2143 #define RTE_ETH_DEV_SRIOV(dev) ((dev)->data->sriov)
2145 #define RTE_ETH_NAME_MAX_LEN RTE_DEV_NAME_MAX_LEN
2147 #define RTE_ETH_DEV_NO_OWNER 0
2149 #define RTE_ETH_MAX_OWNER_NAME_LEN 64
2151 struct rte_eth_dev_owner {
2152 uint64_t id; /**< The owner unique identifier. */
2153 char name[RTE_ETH_MAX_OWNER_NAME_LEN]; /**< The owner name. */
2156 /**@{@name Device flags
2157 * Flags internally saved in rte_eth_dev_data.dev_flags
2158 * and reported in rte_eth_dev_info.dev_flags.
2160 /** PMD supports thread-safe flow operations */
2161 #define RTE_ETH_DEV_FLOW_OPS_THREAD_SAFE RTE_BIT32(0)
2162 /** Device supports link state interrupt */
2163 #define RTE_ETH_DEV_INTR_LSC RTE_BIT32(1)
2164 /** Device is a bonded slave */
2165 #define RTE_ETH_DEV_BONDED_SLAVE RTE_BIT32(2)
2166 /** Device supports device removal interrupt */
2167 #define RTE_ETH_DEV_INTR_RMV RTE_BIT32(3)
2168 /** Device is port representor */
2169 #define RTE_ETH_DEV_REPRESENTOR RTE_BIT32(4)
2170 /** Device does not support MAC change after started */
2171 #define RTE_ETH_DEV_NOLIVE_MAC_ADDR RTE_BIT32(5)
2173 * Queue xstats filled automatically by ethdev layer.
2174 * PMDs filling the queue xstats themselves should not set this flag
2176 #define RTE_ETH_DEV_AUTOFILL_QUEUE_XSTATS RTE_BIT32(6)
2180 * Iterates over valid ethdev ports owned by a specific owner.
2183 * The ID of the next possible valid owned port.
2185 * The owner identifier.
2186 * RTE_ETH_DEV_NO_OWNER means iterate over all valid ownerless ports.
2188 * Next valid port ID owned by owner_id, RTE_MAX_ETHPORTS if there is none.
2190 uint64_t rte_eth_find_next_owned_by(uint16_t port_id,
2191 const uint64_t owner_id);
2194 * Macro to iterate over all enabled ethdev ports owned by a specific owner.
2196 #define RTE_ETH_FOREACH_DEV_OWNED_BY(p, o) \
2197 for (p = rte_eth_find_next_owned_by(0, o); \
2198 (unsigned int)p < (unsigned int)RTE_MAX_ETHPORTS; \
2199 p = rte_eth_find_next_owned_by(p + 1, o))
2202 * Iterates over valid ethdev ports.
2205 * The ID of the next possible valid port.
2207 * Next valid port ID, RTE_MAX_ETHPORTS if there is none.
2209 uint16_t rte_eth_find_next(uint16_t port_id);
2212 * Macro to iterate over all enabled and ownerless ethdev ports.
2214 #define RTE_ETH_FOREACH_DEV(p) \
2215 RTE_ETH_FOREACH_DEV_OWNED_BY(p, RTE_ETH_DEV_NO_OWNER)
2218 * Iterates over ethdev ports of a specified device.
2220 * @param port_id_start
2221 * The ID of the next possible valid port.
2223 * The generic device behind the ports to iterate.
2225 * Next port ID of the device, possibly port_id_start,
2226 * RTE_MAX_ETHPORTS if there is none.
2229 rte_eth_find_next_of(uint16_t port_id_start,
2230 const struct rte_device *parent);
2233 * Macro to iterate over all ethdev ports of a specified device.
2236 * The ID of the matching port being iterated.
2238 * The rte_device pointer matching the iterated ports.
2240 #define RTE_ETH_FOREACH_DEV_OF(port_id, parent) \
2241 for (port_id = rte_eth_find_next_of(0, parent); \
2242 port_id < RTE_MAX_ETHPORTS; \
2243 port_id = rte_eth_find_next_of(port_id + 1, parent))
2246 * Iterates over sibling ethdev ports (i.e. sharing the same rte_device).
2248 * @param port_id_start
2249 * The ID of the next possible valid sibling port.
2250 * @param ref_port_id
2251 * The ID of a reference port to compare rte_device with.
2253 * Next sibling port ID, possibly port_id_start or ref_port_id itself,
2254 * RTE_MAX_ETHPORTS if there is none.
2257 rte_eth_find_next_sibling(uint16_t port_id_start, uint16_t ref_port_id);
2260 * Macro to iterate over all ethdev ports sharing the same rte_device
2261 * as the specified port.
2262 * Note: the specified reference port is part of the loop iterations.
2265 * The ID of the matching port being iterated.
2266 * @param ref_port_id
2267 * The ID of the port being compared.
2269 #define RTE_ETH_FOREACH_DEV_SIBLING(port_id, ref_port_id) \
2270 for (port_id = rte_eth_find_next_sibling(0, ref_port_id); \
2271 port_id < RTE_MAX_ETHPORTS; \
2272 port_id = rte_eth_find_next_sibling(port_id + 1, ref_port_id))
2276 * @b EXPERIMENTAL: this API may change without prior notice.
2278 * Get a new unique owner identifier.
2279 * An owner identifier is used to owns Ethernet devices by only one DPDK entity
2280 * to avoid multiple management of device by different entities.
2283 * Owner identifier pointer.
2285 * Negative errno value on error, 0 on success.
2288 int rte_eth_dev_owner_new(uint64_t *owner_id);
2292 * @b EXPERIMENTAL: this API may change without prior notice.
2294 * Set an Ethernet device owner.
2297 * The identifier of the port to own.
2299 * The owner pointer.
2301 * Negative errno value on error, 0 on success.
2304 int rte_eth_dev_owner_set(const uint16_t port_id,
2305 const struct rte_eth_dev_owner *owner);
2309 * @b EXPERIMENTAL: this API may change without prior notice.
2311 * Unset Ethernet device owner to make the device ownerless.
2314 * The identifier of port to make ownerless.
2316 * The owner identifier.
2318 * 0 on success, negative errno value on error.
2321 int rte_eth_dev_owner_unset(const uint16_t port_id,
2322 const uint64_t owner_id);
2326 * @b EXPERIMENTAL: this API may change without prior notice.
2328 * Remove owner from all Ethernet devices owned by a specific owner.
2331 * The owner identifier.
2333 * 0 on success, negative errno value on error.
2336 int rte_eth_dev_owner_delete(const uint64_t owner_id);
2340 * @b EXPERIMENTAL: this API may change without prior notice.
2342 * Get the owner of an Ethernet device.
2345 * The port identifier.
2347 * The owner structure pointer to fill.
2349 * 0 on success, negative errno value on error..
2352 int rte_eth_dev_owner_get(const uint16_t port_id,
2353 struct rte_eth_dev_owner *owner);
2356 * Get the number of ports which are usable for the application.
2358 * These devices must be iterated by using the macro
2359 * ``RTE_ETH_FOREACH_DEV`` or ``RTE_ETH_FOREACH_DEV_OWNED_BY``
2360 * to deal with non-contiguous ranges of devices.
2363 * The count of available Ethernet devices.
2365 uint16_t rte_eth_dev_count_avail(void);
2368 * Get the total number of ports which are allocated.
2370 * Some devices may not be available for the application.
2373 * The total count of Ethernet devices.
2375 uint16_t rte_eth_dev_count_total(void);
2378 * Convert a numerical speed in Mbps to a bitmap flag that can be used in
2379 * the bitmap link_speeds of the struct rte_eth_conf
2382 * Numerical speed value in Mbps
2384 * RTE_ETH_LINK_[HALF/FULL]_DUPLEX (only for 10/100M speeds)
2386 * 0 if the speed cannot be mapped
2388 uint32_t rte_eth_speed_bitflag(uint32_t speed, int duplex);
2391 * Get RTE_ETH_RX_OFFLOAD_* flag name.
2396 * Offload name or 'UNKNOWN' if the flag cannot be recognised.
2398 const char *rte_eth_dev_rx_offload_name(uint64_t offload);
2401 * Get RTE_ETH_TX_OFFLOAD_* flag name.
2406 * Offload name or 'UNKNOWN' if the flag cannot be recognised.
2408 const char *rte_eth_dev_tx_offload_name(uint64_t offload);
2412 * @b EXPERIMENTAL: this API may change without prior notice.
2414 * Get RTE_ETH_DEV_CAPA_* flag name.
2419 * Capability name or 'UNKNOWN' if the flag cannot be recognized.
2422 const char *rte_eth_dev_capability_name(uint64_t capability);
2425 * Configure an Ethernet device.
2426 * This function must be invoked first before any other function in the
2427 * Ethernet API. This function can also be re-invoked when a device is in the
2431 * The port identifier of the Ethernet device to configure.
2432 * @param nb_rx_queue
2433 * The number of receive queues to set up for the Ethernet device.
2434 * @param nb_tx_queue
2435 * The number of transmit queues to set up for the Ethernet device.
2437 * The pointer to the configuration data to be used for the Ethernet device.
2438 * The *rte_eth_conf* structure includes:
2439 * - the hardware offload features to activate, with dedicated fields for
2440 * each statically configurable offload hardware feature provided by
2441 * Ethernet devices, such as IP checksum or VLAN tag stripping for
2443 * The Rx offload bitfield API is obsolete and will be deprecated.
2444 * Applications should set the ignore_bitfield_offloads bit on *rxmode*
2445 * structure and use offloads field to set per-port offloads instead.
2446 * - Any offloading set in eth_conf->[rt]xmode.offloads must be within
2447 * the [rt]x_offload_capa returned from rte_eth_dev_info_get().
2448 * Any type of device supported offloading set in the input argument
2449 * eth_conf->[rt]xmode.offloads to rte_eth_dev_configure() is enabled
2450 * on all queues and it can't be disabled in rte_eth_[rt]x_queue_setup()
2451 * - the Receive Side Scaling (RSS) configuration when using multiple Rx
2452 * queues per port. Any RSS hash function set in eth_conf->rss_conf.rss_hf
2453 * must be within the flow_type_rss_offloads provided by drivers via
2454 * rte_eth_dev_info_get() API.
2456 * Embedding all configuration information in a single data structure
2457 * is the more flexible method that allows the addition of new features
2458 * without changing the syntax of the API.
2460 * - 0: Success, device configured.
2461 * - <0: Error code returned by the driver configuration function.
2463 int rte_eth_dev_configure(uint16_t port_id, uint16_t nb_rx_queue,
2464 uint16_t nb_tx_queue, const struct rte_eth_conf *eth_conf);
2467 * Check if an Ethernet device was physically removed.
2470 * The port identifier of the Ethernet device.
2472 * 1 when the Ethernet device is removed, otherwise 0.
2475 rte_eth_dev_is_removed(uint16_t port_id);
2478 * Allocate and set up a receive queue for an Ethernet device.
2480 * The function allocates a contiguous block of memory for *nb_rx_desc*
2481 * receive descriptors from a memory zone associated with *socket_id*
2482 * and initializes each receive descriptor with a network buffer allocated
2483 * from the memory pool *mb_pool*.
2486 * The port identifier of the Ethernet device.
2487 * @param rx_queue_id
2488 * The index of the receive queue to set up.
2489 * The value must be in the range [0, nb_rx_queue - 1] previously supplied
2490 * to rte_eth_dev_configure().
2492 * The number of receive descriptors to allocate for the receive ring.
2494 * The *socket_id* argument is the socket identifier in case of NUMA.
2495 * The value can be *SOCKET_ID_ANY* if there is no NUMA constraint for
2496 * the DMA memory allocated for the receive descriptors of the ring.
2498 * The pointer to the configuration data to be used for the receive queue.
2499 * NULL value is allowed, in which case default Rx configuration
2501 * The *rx_conf* structure contains an *rx_thresh* structure with the values
2502 * of the Prefetch, Host, and Write-Back threshold registers of the receive
2504 * In addition it contains the hardware offloads features to activate using
2505 * the RTE_ETH_RX_OFFLOAD_* flags.
2506 * If an offloading set in rx_conf->offloads
2507 * hasn't been set in the input argument eth_conf->rxmode.offloads
2508 * to rte_eth_dev_configure(), it is a new added offloading, it must be
2509 * per-queue type and it is enabled for the queue.
2510 * No need to repeat any bit in rx_conf->offloads which has already been
2511 * enabled in rte_eth_dev_configure() at port level. An offloading enabled
2512 * at port level can't be disabled at queue level.
2513 * The configuration structure also contains the pointer to the array
2514 * of the receiving buffer segment descriptions, see rx_seg and rx_nseg
2515 * fields, this extended configuration might be used by split offloads like
2516 * RTE_ETH_RX_OFFLOAD_BUFFER_SPLIT. If mb_pool is not NULL,
2517 * the extended configuration fields must be set to NULL and zero.
2519 * The pointer to the memory pool from which to allocate *rte_mbuf* network
2520 * memory buffers to populate each descriptor of the receive ring. There are
2521 * two options to provide Rx buffer configuration:
2523 * mb_pool is not NULL, rx_conf.rx_nseg is 0.
2524 * - multiple segments description:
2525 * mb_pool is NULL, rx_conf.rx_seg is not NULL, rx_conf.rx_nseg is not 0.
2526 * Taken only if flag RTE_ETH_RX_OFFLOAD_BUFFER_SPLIT is set in offloads.
2529 * - 0: Success, receive queue correctly set up.
2530 * - -EIO: if device is removed.
2531 * - -ENODEV: if *port_id* is invalid.
2532 * - -EINVAL: The memory pool pointer is null or the size of network buffers
2533 * which can be allocated from this memory pool does not fit the various
2534 * buffer sizes allowed by the device controller.
2535 * - -ENOMEM: Unable to allocate the receive ring descriptors or to
2536 * allocate network memory buffers from the memory pool when
2537 * initializing receive descriptors.
2539 int rte_eth_rx_queue_setup(uint16_t port_id, uint16_t rx_queue_id,
2540 uint16_t nb_rx_desc, unsigned int socket_id,
2541 const struct rte_eth_rxconf *rx_conf,
2542 struct rte_mempool *mb_pool);
2546 * @b EXPERIMENTAL: this API may change, or be removed, without prior notice
2548 * Allocate and set up a hairpin receive queue for an Ethernet device.
2550 * The function set up the selected queue to be used in hairpin.
2553 * The port identifier of the Ethernet device.
2554 * @param rx_queue_id
2555 * The index of the receive queue to set up.
2556 * The value must be in the range [0, nb_rx_queue - 1] previously supplied
2557 * to rte_eth_dev_configure().
2559 * The number of receive descriptors to allocate for the receive ring.
2560 * 0 means the PMD will use default value.
2562 * The pointer to the hairpin configuration.
2565 * - (0) if successful.
2566 * - (-ENODEV) if *port_id* is invalid.
2567 * - (-ENOTSUP) if hardware doesn't support.
2568 * - (-EINVAL) if bad parameter.
2569 * - (-ENOMEM) if unable to allocate the resources.
2572 int rte_eth_rx_hairpin_queue_setup
2573 (uint16_t port_id, uint16_t rx_queue_id, uint16_t nb_rx_desc,
2574 const struct rte_eth_hairpin_conf *conf);
2577 * Allocate and set up a transmit queue for an Ethernet device.
2580 * The port identifier of the Ethernet device.
2581 * @param tx_queue_id
2582 * The index of the transmit queue to set up.
2583 * The value must be in the range [0, nb_tx_queue - 1] previously supplied
2584 * to rte_eth_dev_configure().
2586 * The number of transmit descriptors to allocate for the transmit ring.
2588 * The *socket_id* argument is the socket identifier in case of NUMA.
2589 * Its value can be *SOCKET_ID_ANY* if there is no NUMA constraint for
2590 * the DMA memory allocated for the transmit descriptors of the ring.
2592 * The pointer to the configuration data to be used for the transmit queue.
2593 * NULL value is allowed, in which case default Tx configuration
2595 * The *tx_conf* structure contains the following data:
2596 * - The *tx_thresh* structure with the values of the Prefetch, Host, and
2597 * Write-Back threshold registers of the transmit ring.
2598 * When setting Write-Back threshold to the value greater then zero,
2599 * *tx_rs_thresh* value should be explicitly set to one.
2600 * - The *tx_free_thresh* value indicates the [minimum] number of network
2601 * buffers that must be pending in the transmit ring to trigger their
2602 * [implicit] freeing by the driver transmit function.
2603 * - The *tx_rs_thresh* value indicates the [minimum] number of transmit
2604 * descriptors that must be pending in the transmit ring before setting the
2605 * RS bit on a descriptor by the driver transmit function.
2606 * The *tx_rs_thresh* value should be less or equal then
2607 * *tx_free_thresh* value, and both of them should be less then
2609 * - The *offloads* member contains Tx offloads to be enabled.
2610 * If an offloading set in tx_conf->offloads
2611 * hasn't been set in the input argument eth_conf->txmode.offloads
2612 * to rte_eth_dev_configure(), it is a new added offloading, it must be
2613 * per-queue type and it is enabled for the queue.
2614 * No need to repeat any bit in tx_conf->offloads which has already been
2615 * enabled in rte_eth_dev_configure() at port level. An offloading enabled
2616 * at port level can't be disabled at queue level.
2618 * Note that setting *tx_free_thresh* or *tx_rs_thresh* value to 0 forces
2619 * the transmit function to use default values.
2621 * - 0: Success, the transmit queue is correctly set up.
2622 * - -ENOMEM: Unable to allocate the transmit ring descriptors.
2624 int rte_eth_tx_queue_setup(uint16_t port_id, uint16_t tx_queue_id,
2625 uint16_t nb_tx_desc, unsigned int socket_id,
2626 const struct rte_eth_txconf *tx_conf);
2630 * @b EXPERIMENTAL: this API may change, or be removed, without prior notice
2632 * Allocate and set up a transmit hairpin queue for an Ethernet device.
2635 * The port identifier of the Ethernet device.
2636 * @param tx_queue_id
2637 * The index of the transmit queue to set up.
2638 * The value must be in the range [0, nb_tx_queue - 1] previously supplied
2639 * to rte_eth_dev_configure().
2641 * The number of transmit descriptors to allocate for the transmit ring.
2642 * 0 to set default PMD value.
2644 * The hairpin configuration.
2647 * - (0) if successful.
2648 * - (-ENODEV) if *port_id* is invalid.
2649 * - (-ENOTSUP) if hardware doesn't support.
2650 * - (-EINVAL) if bad parameter.
2651 * - (-ENOMEM) if unable to allocate the resources.
2654 int rte_eth_tx_hairpin_queue_setup
2655 (uint16_t port_id, uint16_t tx_queue_id, uint16_t nb_tx_desc,
2656 const struct rte_eth_hairpin_conf *conf);
2660 * @b EXPERIMENTAL: this API may change, or be removed, without prior notice
2662 * Get all the hairpin peer Rx / Tx ports of the current port.
2663 * The caller should ensure that the array is large enough to save the ports
2667 * The port identifier of the Ethernet device.
2669 * Pointer to the array to store the peer ports list.
2671 * Length of the array to store the port identifiers.
2673 * Current port to peer port direction
2674 * positive - current used as Tx to get all peer Rx ports.
2675 * zero - current used as Rx to get all peer Tx ports.
2678 * - (0 or positive) actual peer ports number.
2679 * - (-EINVAL) if bad parameter.
2680 * - (-ENODEV) if *port_id* invalid
2681 * - (-ENOTSUP) if hardware doesn't support.
2682 * - Others detailed errors from PMDs.
2685 int rte_eth_hairpin_get_peer_ports(uint16_t port_id, uint16_t *peer_ports,
2686 size_t len, uint32_t direction);
2690 * @b EXPERIMENTAL: this API may change, or be removed, without prior notice
2692 * Bind all hairpin Tx queues of one port to the Rx queues of the peer port.
2693 * It is only allowed to call this function after all hairpin queues are
2694 * configured properly and the devices are in started state.
2697 * The identifier of the Tx port.
2699 * The identifier of peer Rx port.
2700 * RTE_MAX_ETHPORTS is allowed for the traversal of all devices.
2701 * Rx port ID could have the same value as Tx port ID.
2704 * - (0) if successful.
2705 * - (-ENODEV) if Tx port ID is invalid.
2706 * - (-EBUSY) if device is not in started state.
2707 * - (-ENOTSUP) if hardware doesn't support.
2708 * - Others detailed errors from PMDs.
2711 int rte_eth_hairpin_bind(uint16_t tx_port, uint16_t rx_port);
2715 * @b EXPERIMENTAL: this API may change, or be removed, without prior notice
2717 * Unbind all hairpin Tx queues of one port from the Rx queues of the peer port.
2718 * This should be called before closing the Tx or Rx devices, if the bind
2719 * function is called before.
2720 * After unbinding the hairpin ports pair, it is allowed to bind them again.
2721 * Changing queues configuration should be after stopping the device(s).
2724 * The identifier of the Tx port.
2726 * The identifier of peer Rx port.
2727 * RTE_MAX_ETHPORTS is allowed for traversal of all devices.
2728 * Rx port ID could have the same value as Tx port ID.
2731 * - (0) if successful.
2732 * - (-ENODEV) if Tx port ID is invalid.
2733 * - (-EBUSY) if device is in stopped state.
2734 * - (-ENOTSUP) if hardware doesn't support.
2735 * - Others detailed errors from PMDs.
2738 int rte_eth_hairpin_unbind(uint16_t tx_port, uint16_t rx_port);
2741 * Return the NUMA socket to which an Ethernet device is connected
2744 * The port identifier of the Ethernet device
2746 * The NUMA socket ID to which the Ethernet device is connected or
2747 * a default of zero if the socket could not be determined.
2748 * -1 is returned is the port_id value is out of range.
2750 int rte_eth_dev_socket_id(uint16_t port_id);
2753 * Check if port_id of device is attached
2756 * The port identifier of the Ethernet device
2758 * - 0 if port is out of range or not attached
2759 * - 1 if device is attached
2761 int rte_eth_dev_is_valid_port(uint16_t port_id);
2764 * Start specified Rx queue of a port. It is used when rx_deferred_start
2765 * flag of the specified queue is true.
2768 * The port identifier of the Ethernet device
2769 * @param rx_queue_id
2770 * The index of the Rx queue to update the ring.
2771 * The value must be in the range [0, nb_rx_queue - 1] previously supplied
2772 * to rte_eth_dev_configure().
2774 * - 0: Success, the receive queue is started.
2775 * - -ENODEV: if *port_id* is invalid.
2776 * - -EINVAL: The queue_id out of range or belong to hairpin.
2777 * - -EIO: if device is removed.
2778 * - -ENOTSUP: The function not supported in PMD.
2780 int rte_eth_dev_rx_queue_start(uint16_t port_id, uint16_t rx_queue_id);
2783 * Stop specified Rx queue of a port
2786 * The port identifier of the Ethernet device
2787 * @param rx_queue_id
2788 * The index of the Rx queue to update the ring.
2789 * The value must be in the range [0, nb_rx_queue - 1] previously supplied
2790 * to rte_eth_dev_configure().
2792 * - 0: Success, the receive queue is stopped.
2793 * - -ENODEV: if *port_id* is invalid.
2794 * - -EINVAL: The queue_id out of range or belong to hairpin.
2795 * - -EIO: if device is removed.
2796 * - -ENOTSUP: The function not supported in PMD.
2798 int rte_eth_dev_rx_queue_stop(uint16_t port_id, uint16_t rx_queue_id);
2801 * Start Tx for specified queue of a port. It is used when tx_deferred_start
2802 * flag of the specified queue is true.
2805 * The port identifier of the Ethernet device
2806 * @param tx_queue_id
2807 * The index of the Tx queue to update the ring.
2808 * The value must be in the range [0, nb_tx_queue - 1] previously supplied
2809 * to rte_eth_dev_configure().
2811 * - 0: Success, the transmit queue is started.
2812 * - -ENODEV: if *port_id* is invalid.
2813 * - -EINVAL: The queue_id out of range or belong to hairpin.
2814 * - -EIO: if device is removed.
2815 * - -ENOTSUP: The function not supported in PMD.
2817 int rte_eth_dev_tx_queue_start(uint16_t port_id, uint16_t tx_queue_id);
2820 * Stop specified Tx queue of a port
2823 * The port identifier of the Ethernet device
2824 * @param tx_queue_id
2825 * The index of the Tx queue to update the ring.
2826 * The value must be in the range [0, nb_tx_queue - 1] previously supplied
2827 * to rte_eth_dev_configure().
2829 * - 0: Success, the transmit queue is stopped.
2830 * - -ENODEV: if *port_id* is invalid.
2831 * - -EINVAL: The queue_id out of range or belong to hairpin.
2832 * - -EIO: if device is removed.
2833 * - -ENOTSUP: The function not supported in PMD.
2835 int rte_eth_dev_tx_queue_stop(uint16_t port_id, uint16_t tx_queue_id);
2838 * Start an Ethernet device.
2840 * The device start step is the last one and consists of setting the configured
2841 * offload features and in starting the transmit and the receive units of the
2844 * Device RTE_ETH_DEV_NOLIVE_MAC_ADDR flag causes MAC address to be set before
2845 * PMD port start callback function is invoked.
2847 * On success, all basic functions exported by the Ethernet API (link status,
2848 * receive/transmit, and so on) can be invoked.
2851 * The port identifier of the Ethernet device.
2853 * - 0: Success, Ethernet device started.
2854 * - <0: Error code of the driver device start function.
2856 int rte_eth_dev_start(uint16_t port_id);
2859 * Stop an Ethernet device. The device can be restarted with a call to
2860 * rte_eth_dev_start()
2863 * The port identifier of the Ethernet device.
2865 * - 0: Success, Ethernet device stopped.
2866 * - <0: Error code of the driver device stop function.
2868 int rte_eth_dev_stop(uint16_t port_id);
2871 * Link up an Ethernet device.
2873 * Set device link up will re-enable the device Rx/Tx
2874 * functionality after it is previously set device linked down.
2877 * The port identifier of the Ethernet device.
2879 * - 0: Success, Ethernet device linked up.
2880 * - <0: Error code of the driver device link up function.
2882 int rte_eth_dev_set_link_up(uint16_t port_id);
2885 * Link down an Ethernet device.
2886 * The device Rx/Tx functionality will be disabled if success,
2887 * and it can be re-enabled with a call to
2888 * rte_eth_dev_set_link_up()
2891 * The port identifier of the Ethernet device.
2893 int rte_eth_dev_set_link_down(uint16_t port_id);
2896 * Close a stopped Ethernet device. The device cannot be restarted!
2897 * The function frees all port resources.
2900 * The port identifier of the Ethernet device.
2902 * - Zero if the port is closed successfully.
2903 * - Negative if something went wrong.
2905 int rte_eth_dev_close(uint16_t port_id);
2908 * Reset a Ethernet device and keep its port ID.
2910 * When a port has to be reset passively, the DPDK application can invoke
2911 * this function. For example when a PF is reset, all its VFs should also
2912 * be reset. Normally a DPDK application can invoke this function when
2913 * RTE_ETH_EVENT_INTR_RESET event is detected, but can also use it to start
2914 * a port reset in other circumstances.
2916 * When this function is called, it first stops the port and then calls the
2917 * PMD specific dev_uninit( ) and dev_init( ) to return the port to initial
2918 * state, in which no Tx and Rx queues are setup, as if the port has been
2919 * reset and not started. The port keeps the port ID it had before the
2922 * After calling rte_eth_dev_reset( ), the application should use
2923 * rte_eth_dev_configure( ), rte_eth_rx_queue_setup( ),
2924 * rte_eth_tx_queue_setup( ), and rte_eth_dev_start( )
2925 * to reconfigure the device as appropriate.
2927 * Note: To avoid unexpected behavior, the application should stop calling
2928 * Tx and Rx functions before calling rte_eth_dev_reset( ). For thread
2929 * safety, all these controlling functions should be called from the same
2933 * The port identifier of the Ethernet device.
2936 * - (0) if successful.
2937 * - (-ENODEV) if *port_id* is invalid.
2938 * - (-ENOTSUP) if hardware doesn't support this function.
2939 * - (-EPERM) if not ran from the primary process.
2940 * - (-EIO) if re-initialisation failed or device is removed.
2941 * - (-ENOMEM) if the reset failed due to OOM.
2942 * - (-EAGAIN) if the reset temporarily failed and should be retried later.
2944 int rte_eth_dev_reset(uint16_t port_id);
2947 * Enable receipt in promiscuous mode for an Ethernet device.
2950 * The port identifier of the Ethernet device.
2952 * - (0) if successful.
2953 * - (-ENOTSUP) if support for promiscuous_enable() does not exist
2955 * - (-ENODEV) if *port_id* invalid.
2957 int rte_eth_promiscuous_enable(uint16_t port_id);
2960 * Disable receipt in promiscuous mode for an Ethernet device.
2963 * The port identifier of the Ethernet device.
2965 * - (0) if successful.
2966 * - (-ENOTSUP) if support for promiscuous_disable() does not exist
2968 * - (-ENODEV) if *port_id* invalid.
2970 int rte_eth_promiscuous_disable(uint16_t port_id);
2973 * Return the value of promiscuous mode for an Ethernet device.
2976 * The port identifier of the Ethernet device.
2978 * - (1) if promiscuous is enabled
2979 * - (0) if promiscuous is disabled.
2982 int rte_eth_promiscuous_get(uint16_t port_id);
2985 * Enable the receipt of any multicast frame by an Ethernet device.
2988 * The port identifier of the Ethernet device.
2990 * - (0) if successful.
2991 * - (-ENOTSUP) if support for allmulticast_enable() does not exist
2993 * - (-ENODEV) if *port_id* invalid.
2995 int rte_eth_allmulticast_enable(uint16_t port_id);
2998 * Disable the receipt of all multicast frames by an Ethernet device.
3001 * The port identifier of the Ethernet device.
3003 * - (0) if successful.
3004 * - (-ENOTSUP) if support for allmulticast_disable() does not exist
3006 * - (-ENODEV) if *port_id* invalid.
3008 int rte_eth_allmulticast_disable(uint16_t port_id);
3011 * Return the value of allmulticast mode for an Ethernet device.
3014 * The port identifier of the Ethernet device.
3016 * - (1) if allmulticast is enabled
3017 * - (0) if allmulticast is disabled.
3020 int rte_eth_allmulticast_get(uint16_t port_id);
3023 * Retrieve the link status (up/down), the duplex mode (half/full),
3024 * the negotiation (auto/fixed), and if available, the speed (Mbps).
3026 * It might need to wait up to 9 seconds.
3027 * @see rte_eth_link_get_nowait.
3030 * The port identifier of the Ethernet device.
3032 * Link information written back.
3034 * - (0) if successful.
3035 * - (-ENOTSUP) if the function is not supported in PMD.
3036 * - (-ENODEV) if *port_id* invalid.
3037 * - (-EINVAL) if bad parameter.
3039 int rte_eth_link_get(uint16_t port_id, struct rte_eth_link *link);
3042 * Retrieve the link status (up/down), the duplex mode (half/full),
3043 * the negotiation (auto/fixed), and if available, the speed (Mbps).
3046 * The port identifier of the Ethernet device.
3048 * Link information written back.
3050 * - (0) if successful.
3051 * - (-ENOTSUP) if the function is not supported in PMD.
3052 * - (-ENODEV) if *port_id* invalid.
3053 * - (-EINVAL) if bad parameter.
3055 int rte_eth_link_get_nowait(uint16_t port_id, struct rte_eth_link *link);
3059 * @b EXPERIMENTAL: this API may change without prior notice.
3061 * The function converts a link_speed to a string. It handles all special
3062 * values like unknown or none speed.
3065 * link_speed of rte_eth_link struct
3067 * Link speed in textual format. It's pointer to immutable memory.
3068 * No free is required.
3071 const char *rte_eth_link_speed_to_str(uint32_t link_speed);
3075 * @b EXPERIMENTAL: this API may change without prior notice.
3077 * The function converts a rte_eth_link struct representing a link status to
3081 * A pointer to a string to be filled with textual representation of
3082 * device status. At least RTE_ETH_LINK_MAX_STR_LEN bytes should be allocated to
3083 * store default link status text.
3085 * Length of available memory at 'str' string.
3087 * Link status returned by rte_eth_link_get function
3089 * Number of bytes written to str array or -EINVAL if bad parameter.
3092 int rte_eth_link_to_str(char *str, size_t len,
3093 const struct rte_eth_link *eth_link);
3096 * Retrieve the general I/O statistics of an Ethernet device.
3099 * The port identifier of the Ethernet device.
3101 * A pointer to a structure of type *rte_eth_stats* to be filled with
3102 * the values of device counters for the following set of statistics:
3103 * - *ipackets* with the total of successfully received packets.
3104 * - *opackets* with the total of successfully transmitted packets.
3105 * - *ibytes* with the total of successfully received bytes.
3106 * - *obytes* with the total of successfully transmitted bytes.
3107 * - *ierrors* with the total of erroneous received packets.
3108 * - *oerrors* with the total of failed transmitted packets.
3110 * Zero if successful. Non-zero otherwise.
3112 int rte_eth_stats_get(uint16_t port_id, struct rte_eth_stats *stats);
3115 * Reset the general I/O statistics of an Ethernet device.
3118 * The port identifier of the Ethernet device.
3120 * - (0) if device notified to reset stats.
3121 * - (-ENOTSUP) if hardware doesn't support.
3122 * - (-ENODEV) if *port_id* invalid.
3123 * - (<0): Error code of the driver stats reset function.
3125 int rte_eth_stats_reset(uint16_t port_id);
3128 * Retrieve names of extended statistics of an Ethernet device.
3130 * There is an assumption that 'xstat_names' and 'xstats' arrays are matched
3132 * xstats_names[i].name => xstats[i].value
3134 * And the array index is same with id field of 'struct rte_eth_xstat':
3137 * This assumption makes key-value pair matching less flexible but simpler.
3140 * The port identifier of the Ethernet device.
3141 * @param xstats_names
3142 * An rte_eth_xstat_name array of at least *size* elements to
3143 * be filled. If set to NULL, the function returns the required number
3146 * The size of the xstats_names array (number of elements).
3148 * - A positive value lower or equal to size: success. The return value
3149 * is the number of entries filled in the stats table.
3150 * - A positive value higher than size: error, the given statistics table
3151 * is too small. The return value corresponds to the size that should
3152 * be given to succeed. The entries in the table are not valid and
3153 * shall not be used by the caller.
3154 * - A negative value on error (invalid port ID).
3156 int rte_eth_xstats_get_names(uint16_t port_id,
3157 struct rte_eth_xstat_name *xstats_names,
3161 * Retrieve extended statistics of an Ethernet device.
3163 * There is an assumption that 'xstat_names' and 'xstats' arrays are matched
3165 * xstats_names[i].name => xstats[i].value
3167 * And the array index is same with id field of 'struct rte_eth_xstat':
3170 * This assumption makes key-value pair matching less flexible but simpler.
3173 * The port identifier of the Ethernet device.
3175 * A pointer to a table of structure of type *rte_eth_xstat*
3176 * to be filled with device statistics ids and values.
3177 * This parameter can be set to NULL if and only if n is 0.
3179 * The size of the xstats array (number of elements).
3180 * If lower than the required number of elements, the function returns
3181 * the required number of elements.
3182 * If equal to zero, the xstats must be NULL, the function returns the
3183 * required number of elements.
3185 * - A positive value lower or equal to n: success. The return value
3186 * is the number of entries filled in the stats table.
3187 * - A positive value higher than n: error, the given statistics table
3188 * is too small. The return value corresponds to the size that should
3189 * be given to succeed. The entries in the table are not valid and
3190 * shall not be used by the caller.
3191 * - A negative value on error (invalid port ID).
3193 int rte_eth_xstats_get(uint16_t port_id, struct rte_eth_xstat *xstats,
3197 * Retrieve names of extended statistics of an Ethernet device.
3200 * The port identifier of the Ethernet device.
3201 * @param xstats_names
3202 * Array to be filled in with names of requested device statistics.
3203 * Must not be NULL if @p ids are specified (not NULL).
3205 * Number of elements in @p xstats_names array (if not NULL) and in
3206 * @p ids array (if not NULL). Must be 0 if both array pointers are NULL.
3208 * IDs array given by app to retrieve specific statistics. May be NULL to
3209 * retrieve names of all available statistics or, if @p xstats_names is
3210 * NULL as well, just the number of available statistics.
3212 * - A positive value lower or equal to size: success. The return value
3213 * is the number of entries filled in the stats table.
3214 * - A positive value higher than size: success. The given statistics table
3215 * is too small. The return value corresponds to the size that should
3216 * be given to succeed. The entries in the table are not valid and
3217 * shall not be used by the caller.
3218 * - A negative value on error.
3221 rte_eth_xstats_get_names_by_id(uint16_t port_id,
3222 struct rte_eth_xstat_name *xstats_names, unsigned int size,
3226 * Retrieve extended statistics of an Ethernet device.
3229 * The port identifier of the Ethernet device.
3231 * IDs array given by app to retrieve specific statistics. May be NULL to
3232 * retrieve all available statistics or, if @p values is NULL as well,
3233 * just the number of available statistics.
3235 * Array to be filled in with requested device statistics.
3236 * Must not be NULL if ids are specified (not NULL).
3238 * Number of elements in @p values array (if not NULL) and in @p ids
3239 * array (if not NULL). Must be 0 if both array pointers are NULL.
3241 * - A positive value lower or equal to size: success. The return value
3242 * is the number of entries filled in the stats table.
3243 * - A positive value higher than size: success: The given statistics table
3244 * is too small. The return value corresponds to the size that should
3245 * be given to succeed. The entries in the table are not valid and
3246 * shall not be used by the caller.
3247 * - A negative value on error.
3249 int rte_eth_xstats_get_by_id(uint16_t port_id, const uint64_t *ids,
3250 uint64_t *values, unsigned int size);
3253 * Gets the ID of a statistic from its name.
3255 * This function searches for the statistics using string compares, and
3256 * as such should not be used on the fast-path. For fast-path retrieval of
3257 * specific statistics, store the ID as provided in *id* from this function,
3258 * and pass the ID to rte_eth_xstats_get()
3260 * @param port_id The port to look up statistics from
3261 * @param xstat_name The name of the statistic to return
3262 * @param[out] id A pointer to an app-supplied uint64_t which should be
3263 * set to the ID of the stat if the stat exists.
3266 * -ENODEV for invalid port_id,
3267 * -EIO if device is removed,
3268 * -EINVAL if the xstat_name doesn't exist in port_id
3269 * -ENOMEM if bad parameter.
3271 int rte_eth_xstats_get_id_by_name(uint16_t port_id, const char *xstat_name,
3275 * Reset extended statistics of an Ethernet device.
3278 * The port identifier of the Ethernet device.
3280 * - (0) if device notified to reset extended stats.
3281 * - (-ENOTSUP) if pmd doesn't support both
3282 * extended stats and basic stats reset.
3283 * - (-ENODEV) if *port_id* invalid.
3284 * - (<0): Error code of the driver xstats reset function.
3286 int rte_eth_xstats_reset(uint16_t port_id);
3289 * Set a mapping for the specified transmit queue to the specified per-queue
3290 * statistics counter.
3293 * The port identifier of the Ethernet device.
3294 * @param tx_queue_id
3295 * The index of the transmit queue for which a queue stats mapping is required.
3296 * The value must be in the range [0, nb_tx_queue - 1] previously supplied
3297 * to rte_eth_dev_configure().
3299 * The per-queue packet statistics functionality number that the transmit
3300 * queue is to be assigned.
3301 * The value must be in the range [0, RTE_ETHDEV_QUEUE_STAT_CNTRS - 1].
3302 * Max RTE_ETHDEV_QUEUE_STAT_CNTRS being 256.
3304 * Zero if successful. Non-zero otherwise.
3306 int rte_eth_dev_set_tx_queue_stats_mapping(uint16_t port_id,
3307 uint16_t tx_queue_id, uint8_t stat_idx);
3310 * Set a mapping for the specified receive queue to the specified per-queue
3311 * statistics counter.
3314 * The port identifier of the Ethernet device.
3315 * @param rx_queue_id
3316 * The index of the receive queue for which a queue stats mapping is required.
3317 * The value must be in the range [0, nb_rx_queue - 1] previously supplied
3318 * to rte_eth_dev_configure().
3320 * The per-queue packet statistics functionality number that the receive
3321 * queue is to be assigned.
3322 * The value must be in the range [0, RTE_ETHDEV_QUEUE_STAT_CNTRS - 1].
3323 * Max RTE_ETHDEV_QUEUE_STAT_CNTRS being 256.
3325 * Zero if successful. Non-zero otherwise.
3327 int rte_eth_dev_set_rx_queue_stats_mapping(uint16_t port_id,
3328 uint16_t rx_queue_id,
3332 * Retrieve the Ethernet address of an Ethernet device.
3335 * The port identifier of the Ethernet device.
3337 * A pointer to a structure of type *ether_addr* to be filled with
3338 * the Ethernet address of the Ethernet device.
3340 * - (0) if successful
3341 * - (-ENODEV) if *port_id* invalid.
3342 * - (-EINVAL) if bad parameter.
3344 int rte_eth_macaddr_get(uint16_t port_id, struct rte_ether_addr *mac_addr);
3348 * @b EXPERIMENTAL: this API may change without prior notice
3350 * Retrieve the Ethernet addresses of an Ethernet device.
3353 * The port identifier of the Ethernet device.
3355 * A pointer to an array of structures of type *ether_addr* to be filled with
3356 * the Ethernet addresses of the Ethernet device.
3358 * Number of elements in the @p ma array.
3359 * Note that rte_eth_dev_info::max_mac_addrs can be used to retrieve
3360 * max number of Ethernet addresses for given port.
3362 * - number of retrieved addresses if successful
3363 * - (-ENODEV) if *port_id* invalid.
3364 * - (-EINVAL) if bad parameter.
3367 int rte_eth_macaddrs_get(uint16_t port_id, struct rte_ether_addr *ma,
3371 * Retrieve the contextual information of an Ethernet device.
3373 * As part of this function, a number of of fields in dev_info will be
3374 * initialized as follows:
3379 * Where lim is defined within the rte_eth_dev_info_get as
3381 * const struct rte_eth_desc_lim lim = {
3382 * .nb_max = UINT16_MAX,
3385 * .nb_seg_max = UINT16_MAX,
3386 * .nb_mtu_seg_max = UINT16_MAX,
3389 * device = dev->device
3390 * min_mtu = RTE_ETHER_MIN_LEN - RTE_ETHER_HDR_LEN - RTE_ETHER_CRC_LEN
3391 * max_mtu = UINT16_MAX
3393 * The following fields will be populated if support for dev_infos_get()
3394 * exists for the device and the rte_eth_dev 'dev' has been populated
3395 * successfully with a call to it:
3397 * driver_name = dev->device->driver->name
3398 * nb_rx_queues = dev->data->nb_rx_queues
3399 * nb_tx_queues = dev->data->nb_tx_queues
3400 * dev_flags = &dev->data->dev_flags
3403 * The port identifier of the Ethernet device.
3405 * A pointer to a structure of type *rte_eth_dev_info* to be filled with
3406 * the contextual information of the Ethernet device.
3408 * - (0) if successful.
3409 * - (-ENOTSUP) if support for dev_infos_get() does not exist for the device.
3410 * - (-ENODEV) if *port_id* invalid.
3411 * - (-EINVAL) if bad parameter.
3413 int rte_eth_dev_info_get(uint16_t port_id, struct rte_eth_dev_info *dev_info);
3417 * @b EXPERIMENTAL: this API may change without prior notice.
3419 * Retrieve the configuration of an Ethernet device.
3422 * The port identifier of the Ethernet device.
3424 * Location for Ethernet device configuration to be filled in.
3426 * - (0) if successful.
3427 * - (-ENODEV) if *port_id* invalid.
3428 * - (-EINVAL) if bad parameter.
3431 int rte_eth_dev_conf_get(uint16_t port_id, struct rte_eth_conf *dev_conf);
3434 * Retrieve the firmware version of a device.
3437 * The port identifier of the device.
3439 * A pointer to a string array storing the firmware version of a device,
3440 * the string includes terminating null. This pointer is allocated by caller.
3442 * The size of the string array pointed by fw_version, which should be
3443 * large enough to store firmware version of the device.
3445 * - (0) if successful.
3446 * - (-ENOTSUP) if operation is not supported.
3447 * - (-ENODEV) if *port_id* invalid.
3448 * - (-EIO) if device is removed.
3449 * - (-EINVAL) if bad parameter.
3450 * - (>0) if *fw_size* is not enough to store firmware version, return
3451 * the size of the non truncated string.
3453 int rte_eth_dev_fw_version_get(uint16_t port_id,
3454 char *fw_version, size_t fw_size);
3457 * Retrieve the supported packet types of an Ethernet device.
3459 * When a packet type is announced as supported, it *must* be recognized by
3460 * the PMD. For instance, if RTE_PTYPE_L2_ETHER, RTE_PTYPE_L2_ETHER_VLAN
3461 * and RTE_PTYPE_L3_IPV4 are announced, the PMD must return the following
3462 * packet types for these packets:
3463 * - Ether/IPv4 -> RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV4
3464 * - Ether/VLAN/IPv4 -> RTE_PTYPE_L2_ETHER_VLAN | RTE_PTYPE_L3_IPV4
3465 * - Ether/[anything else] -> RTE_PTYPE_L2_ETHER
3466 * - Ether/VLAN/[anything else] -> RTE_PTYPE_L2_ETHER_VLAN
3468 * When a packet is received by a PMD, the most precise type must be
3469 * returned among the ones supported. However a PMD is allowed to set
3470 * packet type that is not in the supported list, at the condition that it
3471 * is more precise. Therefore, a PMD announcing no supported packet types
3472 * can still set a matching packet type in a received packet.
3475 * Better to invoke this API after the device is already started or Rx burst
3476 * function is decided, to obtain correct supported ptypes.
3478 * if a given PMD does not report what ptypes it supports, then the supported
3479 * ptype count is reported as 0.
3481 * The port identifier of the Ethernet device.
3483 * A hint of what kind of packet type which the caller is interested in.
3485 * An array pointer to store adequate packet types, allocated by caller.
3487 * Size of the array pointed by param ptypes.
3489 * - (>=0) Number of supported ptypes. If the number of types exceeds num,
3490 * only num entries will be filled into the ptypes array, but the full
3491 * count of supported ptypes will be returned.
3492 * - (-ENODEV) if *port_id* invalid.
3493 * - (-EINVAL) if bad parameter.
3495 int rte_eth_dev_get_supported_ptypes(uint16_t port_id, uint32_t ptype_mask,
3496 uint32_t *ptypes, int num);
3498 * Inform Ethernet device about reduced range of packet types to handle.
3500 * Application can use this function to set only specific ptypes that it's
3501 * interested. This information can be used by the PMD to optimize Rx path.
3503 * The function accepts an array `set_ptypes` allocated by the caller to
3504 * store the packet types set by the driver, the last element of the array
3505 * is set to RTE_PTYPE_UNKNOWN. The size of the `set_ptype` array should be
3506 * `rte_eth_dev_get_supported_ptypes() + 1` else it might only be filled
3510 * The port identifier of the Ethernet device.
3512 * The ptype family that application is interested in should be bitwise OR of
3513 * RTE_PTYPE_*_MASK or 0.
3515 * An array pointer to store set packet types, allocated by caller. The
3516 * function marks the end of array with RTE_PTYPE_UNKNOWN.
3518 * Size of the array pointed by param ptypes.
3519 * Should be rte_eth_dev_get_supported_ptypes() + 1 to accommodate the
3523 * - (-ENODEV) if *port_id* invalid.
3524 * - (-EINVAL) if *ptype_mask* is invalid (or) set_ptypes is NULL and
3527 int rte_eth_dev_set_ptypes(uint16_t port_id, uint32_t ptype_mask,
3528 uint32_t *set_ptypes, unsigned int num);
3531 * Retrieve the MTU of an Ethernet device.
3534 * The port identifier of the Ethernet device.
3536 * A pointer to a uint16_t where the retrieved MTU is to be stored.
3538 * - (0) if successful.
3539 * - (-ENODEV) if *port_id* invalid.
3540 * - (-EINVAL) if bad parameter.
3542 int rte_eth_dev_get_mtu(uint16_t port_id, uint16_t *mtu);
3545 * Change the MTU of an Ethernet device.
3548 * The port identifier of the Ethernet device.
3550 * A uint16_t for the MTU to be applied.
3552 * - (0) if successful.
3553 * - (-ENOTSUP) if operation is not supported.
3554 * - (-ENODEV) if *port_id* invalid.
3555 * - (-EIO) if device is removed.
3556 * - (-EINVAL) if *mtu* invalid, validation of mtu can occur within
3557 * rte_eth_dev_set_mtu if dev_infos_get is supported by the device or
3558 * when the mtu is set using dev->dev_ops->mtu_set.
3559 * - (-EBUSY) if operation is not allowed when the port is running
3561 int rte_eth_dev_set_mtu(uint16_t port_id, uint16_t mtu);
3564 * Enable/Disable hardware filtering by an Ethernet device of received
3565 * VLAN packets tagged with a given VLAN Tag Identifier.
3568 * The port identifier of the Ethernet device.
3570 * The VLAN Tag Identifier whose filtering must be enabled or disabled.
3572 * If > 0, enable VLAN filtering of VLAN packets tagged with *vlan_id*.
3573 * Otherwise, disable VLAN filtering of VLAN packets tagged with *vlan_id*.
3575 * - (0) if successful.
3576 * - (-ENOTSUP) if hardware-assisted VLAN filtering not configured.
3577 * - (-ENODEV) if *port_id* invalid.
3578 * - (-EIO) if device is removed.
3579 * - (-ENOSYS) if VLAN filtering on *port_id* disabled.
3580 * - (-EINVAL) if *vlan_id* > 4095.
3582 int rte_eth_dev_vlan_filter(uint16_t port_id, uint16_t vlan_id, int on);
3585 * Enable/Disable hardware VLAN Strip by a Rx queue of an Ethernet device.
3588 * The port identifier of the Ethernet device.
3589 * @param rx_queue_id
3590 * The index of the receive queue for which a queue stats mapping is required.
3591 * The value must be in the range [0, nb_rx_queue - 1] previously supplied
3592 * to rte_eth_dev_configure().
3594 * If 1, Enable VLAN Stripping of the receive queue of the Ethernet port.
3595 * If 0, Disable VLAN Stripping of the receive queue of the Ethernet port.
3597 * - (0) if successful.
3598 * - (-ENOTSUP) if hardware-assisted VLAN stripping not configured.
3599 * - (-ENODEV) if *port_id* invalid.
3600 * - (-EINVAL) if *rx_queue_id* invalid.
3602 int rte_eth_dev_set_vlan_strip_on_queue(uint16_t port_id, uint16_t rx_queue_id,
3606 * Set the Outer VLAN Ether Type by an Ethernet device, it can be inserted to
3610 * The port identifier of the Ethernet device.
3614 * The Tag Protocol ID
3616 * - (0) if successful.
3617 * - (-ENOTSUP) if hardware-assisted VLAN TPID setup is not supported.
3618 * - (-ENODEV) if *port_id* invalid.
3619 * - (-EIO) if device is removed.
3621 int rte_eth_dev_set_vlan_ether_type(uint16_t port_id,
3622 enum rte_vlan_type vlan_type,
3626 * Set VLAN offload configuration on an Ethernet device.
3629 * The port identifier of the Ethernet device.
3630 * @param offload_mask
3631 * The VLAN Offload bit mask can be mixed use with "OR"
3632 * RTE_ETH_VLAN_STRIP_OFFLOAD
3633 * RTE_ETH_VLAN_FILTER_OFFLOAD
3634 * RTE_ETH_VLAN_EXTEND_OFFLOAD
3635 * RTE_ETH_QINQ_STRIP_OFFLOAD
3637 * - (0) if successful.
3638 * - (-ENOTSUP) if hardware-assisted VLAN filtering not configured.
3639 * - (-ENODEV) if *port_id* invalid.
3640 * - (-EIO) if device is removed.
3642 int rte_eth_dev_set_vlan_offload(uint16_t port_id, int offload_mask);
3645 * Read VLAN Offload configuration from an Ethernet device
3648 * The port identifier of the Ethernet device.
3650 * - (>0) if successful. Bit mask to indicate
3651 * RTE_ETH_VLAN_STRIP_OFFLOAD
3652 * RTE_ETH_VLAN_FILTER_OFFLOAD
3653 * RTE_ETH_VLAN_EXTEND_OFFLOAD
3654 * RTE_ETH_QINQ_STRIP_OFFLOAD
3655 * - (-ENODEV) if *port_id* invalid.
3657 int rte_eth_dev_get_vlan_offload(uint16_t port_id);
3660 * Set port based Tx VLAN insertion on or off.
3663 * The port identifier of the Ethernet device.
3665 * Port based Tx VLAN identifier together with user priority.
3667 * Turn on or off the port based Tx VLAN insertion.
3670 * - (0) if successful.
3671 * - negative if failed.
3673 int rte_eth_dev_set_vlan_pvid(uint16_t port_id, uint16_t pvid, int on);
3675 typedef void (*buffer_tx_error_fn)(struct rte_mbuf **unsent, uint16_t count,
3679 * Structure used to buffer packets for future Tx
3680 * Used by APIs rte_eth_tx_buffer and rte_eth_tx_buffer_flush
3682 struct rte_eth_dev_tx_buffer {
3683 buffer_tx_error_fn error_callback;
3684 void *error_userdata;
3685 uint16_t size; /**< Size of buffer for buffered Tx */
3686 uint16_t length; /**< Number of packets in the array */
3687 /** Pending packets to be sent on explicit flush or when full */
3688 struct rte_mbuf *pkts[];
3692 * Calculate the size of the Tx buffer.
3695 * Number of stored packets.
3697 #define RTE_ETH_TX_BUFFER_SIZE(sz) \
3698 (sizeof(struct rte_eth_dev_tx_buffer) + (sz) * sizeof(struct rte_mbuf *))
3701 * Initialize default values for buffered transmitting
3704 * Tx buffer to be initialized.
3711 rte_eth_tx_buffer_init(struct rte_eth_dev_tx_buffer *buffer, uint16_t size);
3714 * Configure a callback for buffered packets which cannot be sent
3716 * Register a specific callback to be called when an attempt is made to send
3717 * all packets buffered on an Ethernet port, but not all packets can
3718 * successfully be sent. The callback registered here will be called only
3719 * from calls to rte_eth_tx_buffer() and rte_eth_tx_buffer_flush() APIs.
3720 * The default callback configured for each queue by default just frees the
3721 * packets back to the calling mempool. If additional behaviour is required,
3722 * for example, to count dropped packets, or to retry transmission of packets
3723 * which cannot be sent, this function should be used to register a suitable
3724 * callback function to implement the desired behaviour.
3725 * The example callback "rte_eth_count_unsent_packet_callback()" is also
3726 * provided as reference.
3729 * The port identifier of the Ethernet device.
3731 * The function to be used as the callback.
3733 * Arbitrary parameter to be passed to the callback function
3735 * 0 on success, or -EINVAL if bad parameter
3738 rte_eth_tx_buffer_set_err_callback(struct rte_eth_dev_tx_buffer *buffer,
3739 buffer_tx_error_fn callback, void *userdata);
3742 * Callback function for silently dropping unsent buffered packets.
3744 * This function can be passed to rte_eth_tx_buffer_set_err_callback() to
3745 * adjust the default behavior when buffered packets cannot be sent. This
3746 * function drops any unsent packets silently and is used by Tx buffered
3747 * operations as default behavior.
3749 * NOTE: this function should not be called directly, instead it should be used
3750 * as a callback for packet buffering.
3752 * NOTE: when configuring this function as a callback with
3753 * rte_eth_tx_buffer_set_err_callback(), the final, userdata parameter
3754 * should point to an uint64_t value.
3757 * The previously buffered packets which could not be sent
3759 * The number of unsent packets in the pkts array
3764 rte_eth_tx_buffer_drop_callback(struct rte_mbuf **pkts, uint16_t unsent,
3768 * Callback function for tracking unsent buffered packets.
3770 * This function can be passed to rte_eth_tx_buffer_set_err_callback() to
3771 * adjust the default behavior when buffered packets cannot be sent. This
3772 * function drops any unsent packets, but also updates a user-supplied counter
3773 * to track the overall number of packets dropped. The counter should be an
3774 * uint64_t variable.
3776 * NOTE: this function should not be called directly, instead it should be used
3777 * as a callback for packet buffering.
3779 * NOTE: when configuring this function as a callback with
3780 * rte_eth_tx_buffer_set_err_callback(), the final, userdata parameter
3781 * should point to an uint64_t value.
3784 * The previously buffered packets which could not be sent
3786 * The number of unsent packets in the pkts array
3788 * Pointer to an uint64_t value, which will be incremented by unsent
3791 rte_eth_tx_buffer_count_callback(struct rte_mbuf **pkts, uint16_t unsent,
3795 * Request the driver to free mbufs currently cached by the driver. The
3796 * driver will only free the mbuf if it is no longer in use. It is the
3797 * application's responsibility to ensure rte_eth_tx_buffer_flush(..) is
3801 * The port identifier of the Ethernet device.
3803 * The index of the transmit queue through which output packets must be
3805 * The value must be in the range [0, nb_tx_queue - 1] previously supplied
3806 * to rte_eth_dev_configure().
3808 * Maximum number of packets to free. Use 0 to indicate all possible packets
3809 * should be freed. Note that a packet may be using multiple mbufs.
3812 * -ENODEV: Invalid interface
3813 * -EIO: device is removed
3814 * -ENOTSUP: Driver does not support function
3816 * 0-n: Number of packets freed. More packets may still remain in ring that
3820 rte_eth_tx_done_cleanup(uint16_t port_id, uint16_t queue_id, uint32_t free_cnt);
3823 * Subtypes for IPsec offload event(@ref RTE_ETH_EVENT_IPSEC) raised by
3826 enum rte_eth_event_ipsec_subtype {
3827 /** Unknown event type */
3828 RTE_ETH_EVENT_IPSEC_UNKNOWN = 0,
3829 /** Sequence number overflow */
3830 RTE_ETH_EVENT_IPSEC_ESN_OVERFLOW,
3831 /** Soft time expiry of SA */
3832 RTE_ETH_EVENT_IPSEC_SA_TIME_EXPIRY,
3833 /** Soft byte expiry of SA */
3834 RTE_ETH_EVENT_IPSEC_SA_BYTE_EXPIRY,
3835 /** Max value of this enum */
3836 RTE_ETH_EVENT_IPSEC_MAX
3840 * Descriptor for @ref RTE_ETH_EVENT_IPSEC event. Used by eth dev to send extra
3841 * information of the IPsec offload event.
3843 struct rte_eth_event_ipsec_desc {
3844 /** Type of RTE_ETH_EVENT_IPSEC_* event */
3845 enum rte_eth_event_ipsec_subtype subtype;
3847 * Event specific metadata.
3849 * For the following events, *userdata* registered
3850 * with the *rte_security_session* would be returned
3853 * - @ref RTE_ETH_EVENT_IPSEC_ESN_OVERFLOW
3854 * - @ref RTE_ETH_EVENT_IPSEC_SA_TIME_EXPIRY
3855 * - @ref RTE_ETH_EVENT_IPSEC_SA_BYTE_EXPIRY
3857 * @see struct rte_security_session_conf
3864 * The eth device event type for interrupt, and maybe others in the future.
3866 enum rte_eth_event_type {
3867 RTE_ETH_EVENT_UNKNOWN, /**< unknown event type */
3868 RTE_ETH_EVENT_INTR_LSC, /**< lsc interrupt event */
3869 /** queue state event (enabled/disabled) */
3870 RTE_ETH_EVENT_QUEUE_STATE,
3871 /** reset interrupt event, sent to VF on PF reset */
3872 RTE_ETH_EVENT_INTR_RESET,
3873 RTE_ETH_EVENT_VF_MBOX, /**< message from the VF received by PF */
3874 RTE_ETH_EVENT_MACSEC, /**< MACsec offload related event */
3875 RTE_ETH_EVENT_INTR_RMV, /**< device removal event */
3876 RTE_ETH_EVENT_NEW, /**< port is probed */
3877 RTE_ETH_EVENT_DESTROY, /**< port is released */
3878 RTE_ETH_EVENT_IPSEC, /**< IPsec offload related event */
3879 RTE_ETH_EVENT_FLOW_AGED,/**< New aged-out flows is detected */
3880 RTE_ETH_EVENT_MAX /**< max value of this enum */
3883 /** User application callback to be registered for interrupts. */
3884 typedef int (*rte_eth_dev_cb_fn)(uint16_t port_id,
3885 enum rte_eth_event_type event, void *cb_arg, void *ret_param);
3888 * Register a callback function for port event.
3892 * RTE_ETH_ALL means register the event for all port ids.
3896 * User supplied callback function to be called.
3898 * Pointer to the parameters for the registered callback.
3901 * - On success, zero.
3902 * - On failure, a negative value.
3904 int rte_eth_dev_callback_register(uint16_t port_id,
3905 enum rte_eth_event_type event,
3906 rte_eth_dev_cb_fn cb_fn, void *cb_arg);
3909 * Unregister a callback function for port event.
3913 * RTE_ETH_ALL means unregister the event for all port ids.
3917 * User supplied callback function to be called.
3919 * Pointer to the parameters for the registered callback. -1 means to
3920 * remove all for the same callback address and same event.
3923 * - On success, zero.
3924 * - On failure, a negative value.
3926 int rte_eth_dev_callback_unregister(uint16_t port_id,
3927 enum rte_eth_event_type event,
3928 rte_eth_dev_cb_fn cb_fn, void *cb_arg);
3931 * When there is no Rx packet coming in Rx Queue for a long time, we can
3932 * sleep lcore related to Rx Queue for power saving, and enable Rx interrupt
3933 * to be triggered when Rx packet arrives.
3935 * The rte_eth_dev_rx_intr_enable() function enables Rx queue
3936 * interrupt on specific Rx queue of a port.
3939 * The port identifier of the Ethernet device.
3941 * The index of the receive queue from which to retrieve input packets.
3942 * The value must be in the range [0, nb_rx_queue - 1] previously supplied
3943 * to rte_eth_dev_configure().
3945 * - (0) if successful.
3946 * - (-ENOTSUP) if underlying hardware OR driver doesn't support
3948 * - (-ENODEV) if *port_id* invalid.
3949 * - (-EIO) if device is removed.
3951 int rte_eth_dev_rx_intr_enable(uint16_t port_id, uint16_t queue_id);
3954 * When lcore wakes up from Rx interrupt indicating packet coming, disable Rx
3955 * interrupt and returns to polling mode.
3957 * The rte_eth_dev_rx_intr_disable() function disables Rx queue
3958 * interrupt on specific Rx queue of a port.
3961 * The port identifier of the Ethernet device.
3963 * The index of the receive queue from which to retrieve input packets.
3964 * The value must be in the range [0, nb_rx_queue - 1] previously supplied
3965 * to rte_eth_dev_configure().
3967 * - (0) if successful.
3968 * - (-ENOTSUP) if underlying hardware OR driver doesn't support
3970 * - (-ENODEV) if *port_id* invalid.
3971 * - (-EIO) if device is removed.
3973 int rte_eth_dev_rx_intr_disable(uint16_t port_id, uint16_t queue_id);
3976 * Rx Interrupt control per port.
3979 * The port identifier of the Ethernet device.
3981 * Epoll instance fd which the intr vector associated to.
3982 * Using RTE_EPOLL_PER_THREAD allows to use per thread epoll instance.
3984 * The operation be performed for the vector.
3985 * Operation type of {RTE_INTR_EVENT_ADD, RTE_INTR_EVENT_DEL}.
3989 * - On success, zero.
3990 * - On failure, a negative value.
3992 int rte_eth_dev_rx_intr_ctl(uint16_t port_id, int epfd, int op, void *data);
3995 * Rx Interrupt control per queue.
3998 * The port identifier of the Ethernet device.
4000 * The index of the receive queue from which to retrieve input packets.
4001 * The value must be in the range [0, nb_rx_queue - 1] previously supplied
4002 * to rte_eth_dev_configure().
4004 * Epoll instance fd which the intr vector associated to.
4005 * Using RTE_EPOLL_PER_THREAD allows to use per thread epoll instance.
4007 * The operation be performed for the vector.
4008 * Operation type of {RTE_INTR_EVENT_ADD, RTE_INTR_EVENT_DEL}.
4012 * - On success, zero.
4013 * - On failure, a negative value.
4015 int rte_eth_dev_rx_intr_ctl_q(uint16_t port_id, uint16_t queue_id,
4016 int epfd, int op, void *data);
4019 * Get interrupt fd per Rx queue.
4022 * The port identifier of the Ethernet device.
4024 * The index of the receive queue from which to retrieve input packets.
4025 * The value must be in the range [0, nb_rx_queue - 1] previously supplied
4026 * to rte_eth_dev_configure().
4028 * - (>=0) the interrupt fd associated to the requested Rx queue if
4033 rte_eth_dev_rx_intr_ctl_q_get_fd(uint16_t port_id, uint16_t queue_id);
4036 * Turn on the LED on the Ethernet device.
4037 * This function turns on the LED on the Ethernet device.
4040 * The port identifier of the Ethernet device.
4042 * - (0) if successful.
4043 * - (-ENOTSUP) if underlying hardware OR driver doesn't support
4045 * - (-ENODEV) if *port_id* invalid.
4046 * - (-EIO) if device is removed.
4048 int rte_eth_led_on(uint16_t port_id);
4051 * Turn off the LED on the Ethernet device.
4052 * This function turns off the LED on the Ethernet device.
4055 * The port identifier of the Ethernet device.
4057 * - (0) if successful.
4058 * - (-ENOTSUP) if underlying hardware OR driver doesn't support
4060 * - (-ENODEV) if *port_id* invalid.
4061 * - (-EIO) if device is removed.
4063 int rte_eth_led_off(uint16_t port_id);
4067 * @b EXPERIMENTAL: this API may change, or be removed, without prior notice
4069 * Get Forward Error Correction(FEC) capability.
4072 * The port identifier of the Ethernet device.
4073 * @param speed_fec_capa
4074 * speed_fec_capa is out only with per-speed capabilities.
4075 * If set to NULL, the function returns the required number
4076 * of required array entries.
4078 * a number of elements in an speed_fec_capa array.
4081 * - A non-negative value lower or equal to num: success. The return value
4082 * is the number of entries filled in the fec capa array.
4083 * - A non-negative value higher than num: error, the given fec capa array
4084 * is too small. The return value corresponds to the num that should
4085 * be given to succeed. The entries in fec capa array are not valid and
4086 * shall not be used by the caller.
4087 * - (-ENOTSUP) if underlying hardware OR driver doesn't support.
4089 * - (-EIO) if device is removed.
4090 * - (-ENODEV) if *port_id* invalid.
4091 * - (-EINVAL) if *num* or *speed_fec_capa* invalid
4094 int rte_eth_fec_get_capability(uint16_t port_id,
4095 struct rte_eth_fec_capa *speed_fec_capa,
4100 * @b EXPERIMENTAL: this API may change, or be removed, without prior notice
4102 * Get current Forward Error Correction(FEC) mode.
4103 * If link is down and AUTO is enabled, AUTO is returned, otherwise,
4104 * configured FEC mode is returned.
4105 * If link is up, current FEC mode is returned.
4108 * The port identifier of the Ethernet device.
4110 * A bitmask of enabled FEC modes. If AUTO bit is set, other
4111 * bits specify FEC modes which may be negotiated. If AUTO
4112 * bit is clear, specify FEC modes to be used (only one valid
4113 * mode per speed may be set).
4115 * - (0) if successful.
4116 * - (-ENOTSUP) if underlying hardware OR driver doesn't support.
4118 * - (-EIO) if device is removed.
4119 * - (-ENODEV) if *port_id* invalid.
4122 int rte_eth_fec_get(uint16_t port_id, uint32_t *fec_capa);
4126 * @b EXPERIMENTAL: this API may change, or be removed, without prior notice
4128 * Set Forward Error Correction(FEC) mode.
4131 * The port identifier of the Ethernet device.
4133 * A bitmask of allowed FEC modes. If AUTO bit is set, other
4134 * bits specify FEC modes which may be negotiated. If AUTO
4135 * bit is clear, specify FEC modes to be used (only one valid
4136 * mode per speed may be set).
4138 * - (0) if successful.
4139 * - (-EINVAL) if the FEC mode is not valid.
4140 * - (-ENOTSUP) if underlying hardware OR driver doesn't support.
4141 * - (-EIO) if device is removed.
4142 * - (-ENODEV) if *port_id* invalid.
4145 int rte_eth_fec_set(uint16_t port_id, uint32_t fec_capa);
4148 * Get current status of the Ethernet link flow control for Ethernet device
4151 * The port identifier of the Ethernet device.
4153 * The pointer to the structure where to store the flow control parameters.
4155 * - (0) if successful.
4156 * - (-ENOTSUP) if hardware doesn't support flow control.
4157 * - (-ENODEV) if *port_id* invalid.
4158 * - (-EIO) if device is removed.
4159 * - (-EINVAL) if bad parameter.
4161 int rte_eth_dev_flow_ctrl_get(uint16_t port_id,
4162 struct rte_eth_fc_conf *fc_conf);
4165 * Configure the Ethernet link flow control for Ethernet device
4168 * The port identifier of the Ethernet device.
4170 * The pointer to the structure of the flow control parameters.
4172 * - (0) if successful.
4173 * - (-ENOTSUP) if hardware doesn't support flow control mode.
4174 * - (-ENODEV) if *port_id* invalid.
4175 * - (-EINVAL) if bad parameter
4176 * - (-EIO) if flow control setup failure or device is removed.
4178 int rte_eth_dev_flow_ctrl_set(uint16_t port_id,
4179 struct rte_eth_fc_conf *fc_conf);
4182 * Configure the Ethernet priority flow control under DCB environment
4183 * for Ethernet device.
4186 * The port identifier of the Ethernet device.
4188 * The pointer to the structure of the priority flow control parameters.
4190 * - (0) if successful.
4191 * - (-ENOTSUP) if hardware doesn't support priority flow control mode.
4192 * - (-ENODEV) if *port_id* invalid.
4193 * - (-EINVAL) if bad parameter
4194 * - (-EIO) if flow control setup failure or device is removed.
4196 int rte_eth_dev_priority_flow_ctrl_set(uint16_t port_id,
4197 struct rte_eth_pfc_conf *pfc_conf);
4200 * Add a MAC address to the set used for filtering incoming packets.
4203 * The port identifier of the Ethernet device.
4205 * The MAC address to add.
4207 * VMDq pool index to associate address with (if VMDq is enabled). If VMDq is
4208 * not enabled, this should be set to 0.
4210 * - (0) if successfully added or *mac_addr* was already added.
4211 * - (-ENOTSUP) if hardware doesn't support this feature.
4212 * - (-ENODEV) if *port* is invalid.
4213 * - (-EIO) if device is removed.
4214 * - (-ENOSPC) if no more MAC addresses can be added.
4215 * - (-EINVAL) if MAC address is invalid.
4217 int rte_eth_dev_mac_addr_add(uint16_t port_id, struct rte_ether_addr *mac_addr,
4222 * @b EXPERIMENTAL: this API may change without prior notice.
4224 * Retrieve the information for queue based PFC.
4227 * The port identifier of the Ethernet device.
4228 * @param pfc_queue_info
4229 * A pointer to a structure of type *rte_eth_pfc_queue_info* to be filled with
4230 * the information about queue based PFC.
4232 * - (0) if successful.
4233 * - (-ENOTSUP) if support for priority_flow_ctrl_queue_info_get does not exist.
4234 * - (-ENODEV) if *port_id* invalid.
4235 * - (-EINVAL) if bad parameter.
4238 int rte_eth_dev_priority_flow_ctrl_queue_info_get(uint16_t port_id,
4239 struct rte_eth_pfc_queue_info *pfc_queue_info);
4243 * @b EXPERIMENTAL: this API may change without prior notice.
4245 * Configure the queue based priority flow control for a given queue
4246 * for Ethernet device.
4248 * @note When an ethdev port switches to queue based PFC mode, the
4249 * unconfigured queues shall be configured by the driver with
4250 * default values such as lower priority value for TC etc.
4253 * The port identifier of the Ethernet device.
4254 * @param pfc_queue_conf
4255 * The pointer to the structure of the priority flow control parameters
4258 * - (0) if successful.
4259 * - (-ENOTSUP) if hardware doesn't support queue based PFC mode.
4260 * - (-ENODEV) if *port_id* invalid.
4261 * - (-EINVAL) if bad parameter
4262 * - (-EIO) if flow control setup queue failure
4265 int rte_eth_dev_priority_flow_ctrl_queue_configure(uint16_t port_id,
4266 struct rte_eth_pfc_queue_conf *pfc_queue_conf);
4269 * Remove a MAC address from the internal array of addresses.
4272 * The port identifier of the Ethernet device.
4274 * MAC address to remove.
4276 * - (0) if successful, or *mac_addr* didn't exist.
4277 * - (-ENOTSUP) if hardware doesn't support.
4278 * - (-ENODEV) if *port* invalid.
4279 * - (-EADDRINUSE) if attempting to remove the default MAC address.
4280 * - (-EINVAL) if MAC address is invalid.
4282 int rte_eth_dev_mac_addr_remove(uint16_t port_id,
4283 struct rte_ether_addr *mac_addr);
4286 * Set the default MAC address.
4289 * The port identifier of the Ethernet device.
4291 * New default MAC address.
4293 * - (0) if successful, or *mac_addr* didn't exist.
4294 * - (-ENOTSUP) if hardware doesn't support.
4295 * - (-ENODEV) if *port* invalid.
4296 * - (-EINVAL) if MAC address is invalid.
4298 int rte_eth_dev_default_mac_addr_set(uint16_t port_id,
4299 struct rte_ether_addr *mac_addr);
4302 * Update Redirection Table(RETA) of Receive Side Scaling of Ethernet device.
4305 * The port identifier of the Ethernet device.
4309 * Redirection table size. The table size can be queried by
4310 * rte_eth_dev_info_get().
4312 * - (0) if successful.
4313 * - (-ENODEV) if *port_id* is invalid.
4314 * - (-ENOTSUP) if hardware doesn't support.
4315 * - (-EINVAL) if bad parameter.
4316 * - (-EIO) if device is removed.
4318 int rte_eth_dev_rss_reta_update(uint16_t port_id,
4319 struct rte_eth_rss_reta_entry64 *reta_conf,
4320 uint16_t reta_size);
4323 * Query Redirection Table(RETA) of Receive Side Scaling of Ethernet device.
4326 * The port identifier of the Ethernet device.
4328 * RETA to query. For each requested reta entry, corresponding bit
4329 * in mask must be set.
4331 * Redirection table size. The table size can be queried by
4332 * rte_eth_dev_info_get().
4334 * - (0) if successful.
4335 * - (-ENODEV) if *port_id* is invalid.
4336 * - (-ENOTSUP) if hardware doesn't support.
4337 * - (-EINVAL) if bad parameter.
4338 * - (-EIO) if device is removed.
4340 int rte_eth_dev_rss_reta_query(uint16_t port_id,
4341 struct rte_eth_rss_reta_entry64 *reta_conf,
4342 uint16_t reta_size);
4345 * Updates unicast hash table for receiving packet with the given destination
4346 * MAC address, and the packet is routed to all VFs for which the Rx mode is
4347 * accept packets that match the unicast hash table.
4350 * The port identifier of the Ethernet device.
4352 * Unicast MAC address.
4354 * 1 - Set an unicast hash bit for receiving packets with the MAC address.
4355 * 0 - Clear an unicast hash bit.
4357 * - (0) if successful.
4358 * - (-ENOTSUP) if hardware doesn't support.
4359 * - (-ENODEV) if *port_id* invalid.
4360 * - (-EIO) if device is removed.
4361 * - (-EINVAL) if bad parameter.
4363 int rte_eth_dev_uc_hash_table_set(uint16_t port_id, struct rte_ether_addr *addr,
4367 * Updates all unicast hash bitmaps for receiving packet with any Unicast
4368 * Ethernet MAC addresses,the packet is routed to all VFs for which the Rx
4369 * mode is accept packets that match the unicast hash table.
4372 * The port identifier of the Ethernet device.
4374 * 1 - Set all unicast hash bitmaps for receiving all the Ethernet
4376 * 0 - Clear all unicast hash bitmaps
4378 * - (0) if successful.
4379 * - (-ENOTSUP) if hardware doesn't support.
4380 * - (-ENODEV) if *port_id* invalid.
4381 * - (-EIO) if device is removed.
4382 * - (-EINVAL) if bad parameter.
4384 int rte_eth_dev_uc_all_hash_table_set(uint16_t port_id, uint8_t on);
4387 * Set the rate limitation for a queue on an Ethernet device.
4390 * The port identifier of the Ethernet device.
4394 * The Tx rate in Mbps. Allocated from the total port link speed.
4396 * - (0) if successful.
4397 * - (-ENOTSUP) if hardware doesn't support this feature.
4398 * - (-ENODEV) if *port_id* invalid.
4399 * - (-EIO) if device is removed.
4400 * - (-EINVAL) if bad parameter.
4402 int rte_eth_set_queue_rate_limit(uint16_t port_id, uint16_t queue_idx,
4406 * Configuration of Receive Side Scaling hash computation of Ethernet device.
4409 * The port identifier of the Ethernet device.
4411 * The new configuration to use for RSS hash computation on the port.
4413 * - (0) if successful.
4414 * - (-ENODEV) if port identifier is invalid.
4415 * - (-EIO) if device is removed.
4416 * - (-ENOTSUP) if hardware doesn't support.
4417 * - (-EINVAL) if bad parameter.
4419 int rte_eth_dev_rss_hash_update(uint16_t port_id,
4420 struct rte_eth_rss_conf *rss_conf);
4423 * Retrieve current configuration of Receive Side Scaling hash computation
4424 * of Ethernet device.
4427 * The port identifier of the Ethernet device.
4429 * Where to store the current RSS hash configuration of the Ethernet device.
4431 * - (0) if successful.
4432 * - (-ENODEV) if port identifier is invalid.
4433 * - (-EIO) if device is removed.
4434 * - (-ENOTSUP) if hardware doesn't support RSS.
4435 * - (-EINVAL) if bad parameter.
4438 rte_eth_dev_rss_hash_conf_get(uint16_t port_id,
4439 struct rte_eth_rss_conf *rss_conf);
4442 * Add UDP tunneling port for a type of tunnel.
4444 * Some NICs may require such configuration to properly parse a tunnel
4445 * with any standard or custom UDP port.
4446 * The packets with this UDP port will be parsed for this type of tunnel.
4447 * The device parser will also check the rest of the tunnel headers
4448 * before classifying the packet.
4450 * With some devices, this API will affect packet classification, i.e.:
4451 * - mbuf.packet_type reported on Rx
4452 * - rte_flow rules with tunnel items
4455 * The port identifier of the Ethernet device.
4457 * UDP tunneling configuration.
4460 * - (0) if successful.
4461 * - (-ENODEV) if port identifier is invalid.
4462 * - (-EIO) if device is removed.
4463 * - (-ENOTSUP) if hardware doesn't support tunnel type.
4466 rte_eth_dev_udp_tunnel_port_add(uint16_t port_id,
4467 struct rte_eth_udp_tunnel *tunnel_udp);
4470 * Delete UDP tunneling port for a type of tunnel.
4472 * The packets with this UDP port will not be classified as this type of tunnel
4473 * anymore if the device use such mapping for tunnel packet classification.
4475 * @see rte_eth_dev_udp_tunnel_port_add
4478 * The port identifier of the Ethernet device.
4480 * UDP tunneling configuration.
4483 * - (0) if successful.
4484 * - (-ENODEV) if port identifier is invalid.
4485 * - (-EIO) if device is removed.
4486 * - (-ENOTSUP) if hardware doesn't support tunnel type.
4489 rte_eth_dev_udp_tunnel_port_delete(uint16_t port_id,
4490 struct rte_eth_udp_tunnel *tunnel_udp);
4493 * Get DCB information on an Ethernet device.
4496 * The port identifier of the Ethernet device.
4500 * - (0) if successful.
4501 * - (-ENODEV) if port identifier is invalid.
4502 * - (-EIO) if device is removed.
4503 * - (-ENOTSUP) if hardware doesn't support.
4504 * - (-EINVAL) if bad parameter.
4506 int rte_eth_dev_get_dcb_info(uint16_t port_id,
4507 struct rte_eth_dcb_info *dcb_info);
4509 struct rte_eth_rxtx_callback;
4512 * Add a callback to be called on packet Rx on a given port and queue.
4514 * This API configures a function to be called for each burst of
4515 * packets received on a given NIC port queue. The return value is a pointer
4516 * that can be used to later remove the callback using
4517 * rte_eth_remove_rx_callback().
4519 * Multiple functions are called in the order that they are added.
4522 * The port identifier of the Ethernet device.
4524 * The queue on the Ethernet device on which the callback is to be added.
4526 * The callback function
4528 * A generic pointer parameter which will be passed to each invocation of the
4529 * callback function on this port and queue. Inter-thread synchronization
4530 * of any user data changes is the responsibility of the user.
4534 * On success, a pointer value which can later be used to remove the callback.
4536 const struct rte_eth_rxtx_callback *
4537 rte_eth_add_rx_callback(uint16_t port_id, uint16_t queue_id,
4538 rte_rx_callback_fn fn, void *user_param);
4541 * Add a callback that must be called first on packet Rx on a given port
4544 * This API configures a first function to be called for each burst of
4545 * packets received on a given NIC port queue. The return value is a pointer
4546 * that can be used to later remove the callback using
4547 * rte_eth_remove_rx_callback().
4549 * Multiple functions are called in the order that they are added.
4552 * The port identifier of the Ethernet device.
4554 * The queue on the Ethernet device on which the callback is to be added.
4556 * The callback function
4558 * A generic pointer parameter which will be passed to each invocation of the
4559 * callback function on this port and queue. Inter-thread synchronization
4560 * of any user data changes is the responsibility of the user.
4564 * On success, a pointer value which can later be used to remove the callback.
4566 const struct rte_eth_rxtx_callback *
4567 rte_eth_add_first_rx_callback(uint16_t port_id, uint16_t queue_id,
4568 rte_rx_callback_fn fn, void *user_param);
4571 * Add a callback to be called on packet Tx on a given port and queue.
4573 * This API configures a function to be called for each burst of
4574 * packets sent on a given NIC port queue. The return value is a pointer
4575 * that can be used to later remove the callback using
4576 * rte_eth_remove_tx_callback().
4578 * Multiple functions are called in the order that they are added.
4581 * The port identifier of the Ethernet device.
4583 * The queue on the Ethernet device on which the callback is to be added.
4585 * The callback function
4587 * A generic pointer parameter which will be passed to each invocation of the
4588 * callback function on this port and queue. Inter-thread synchronization
4589 * of any user data changes is the responsibility of the user.
4593 * On success, a pointer value which can later be used to remove the callback.
4595 const struct rte_eth_rxtx_callback *
4596 rte_eth_add_tx_callback(uint16_t port_id, uint16_t queue_id,
4597 rte_tx_callback_fn fn, void *user_param);
4600 * Remove an Rx packet callback from a given port and queue.
4602 * This function is used to removed callbacks that were added to a NIC port
4603 * queue using rte_eth_add_rx_callback().
4605 * Note: the callback is removed from the callback list but it isn't freed
4606 * since the it may still be in use. The memory for the callback can be
4607 * subsequently freed back by the application by calling rte_free():
4609 * - Immediately - if the port is stopped, or the user knows that no
4610 * callbacks are in flight e.g. if called from the thread doing Rx/Tx
4613 * - After a short delay - where the delay is sufficient to allow any
4614 * in-flight callbacks to complete. Alternately, the RCU mechanism can be
4615 * used to detect when data plane threads have ceased referencing the
4619 * The port identifier of the Ethernet device.
4621 * The queue on the Ethernet device from which the callback is to be removed.
4623 * User supplied callback created via rte_eth_add_rx_callback().
4626 * - 0: Success. Callback was removed.
4627 * - -ENODEV: If *port_id* is invalid.
4628 * - -ENOTSUP: Callback support is not available.
4629 * - -EINVAL: The queue_id is out of range, or the callback
4630 * is NULL or not found for the port/queue.
4632 int rte_eth_remove_rx_callback(uint16_t port_id, uint16_t queue_id,
4633 const struct rte_eth_rxtx_callback *user_cb);
4636 * Remove a Tx packet callback from a given port and queue.
4638 * This function is used to removed callbacks that were added to a NIC port
4639 * queue using rte_eth_add_tx_callback().
4641 * Note: the callback is removed from the callback list but it isn't freed
4642 * since the it may still be in use. The memory for the callback can be
4643 * subsequently freed back by the application by calling rte_free():
4645 * - Immediately - if the port is stopped, or the user knows that no
4646 * callbacks are in flight e.g. if called from the thread doing Rx/Tx
4649 * - After a short delay - where the delay is sufficient to allow any
4650 * in-flight callbacks to complete. Alternately, the RCU mechanism can be
4651 * used to detect when data plane threads have ceased referencing the
4655 * The port identifier of the Ethernet device.
4657 * The queue on the Ethernet device from which the callback is to be removed.
4659 * User supplied callback created via rte_eth_add_tx_callback().
4662 * - 0: Success. Callback was removed.
4663 * - -ENODEV: If *port_id* is invalid.
4664 * - -ENOTSUP: Callback support is not available.
4665 * - -EINVAL: The queue_id is out of range, or the callback
4666 * is NULL or not found for the port/queue.
4668 int rte_eth_remove_tx_callback(uint16_t port_id, uint16_t queue_id,
4669 const struct rte_eth_rxtx_callback *user_cb);
4672 * Retrieve information about given port's Rx queue.
4675 * The port identifier of the Ethernet device.
4677 * The Rx queue on the Ethernet device for which information
4678 * will be retrieved.
4680 * A pointer to a structure of type *rte_eth_rxq_info_info* to be filled with
4681 * the information of the Ethernet device.
4685 * - -ENODEV: If *port_id* is invalid.
4686 * - -ENOTSUP: routine is not supported by the device PMD.
4687 * - -EINVAL: The queue_id is out of range, or the queue
4690 int rte_eth_rx_queue_info_get(uint16_t port_id, uint16_t queue_id,
4691 struct rte_eth_rxq_info *qinfo);
4694 * Retrieve information about given port's Tx queue.
4697 * The port identifier of the Ethernet device.
4699 * The Tx queue on the Ethernet device for which information
4700 * will be retrieved.
4702 * A pointer to a structure of type *rte_eth_txq_info_info* to be filled with
4703 * the information of the Ethernet device.
4707 * - -ENODEV: If *port_id* is invalid.
4708 * - -ENOTSUP: routine is not supported by the device PMD.
4709 * - -EINVAL: The queue_id is out of range, or the queue
4712 int rte_eth_tx_queue_info_get(uint16_t port_id, uint16_t queue_id,
4713 struct rte_eth_txq_info *qinfo);
4716 * Retrieve information about the Rx packet burst mode.
4719 * The port identifier of the Ethernet device.
4721 * The Rx queue on the Ethernet device for which information
4722 * will be retrieved.
4724 * A pointer to a structure of type *rte_eth_burst_mode* to be filled
4725 * with the information of the packet burst mode.
4729 * - -ENODEV: If *port_id* is invalid.
4730 * - -ENOTSUP: routine is not supported by the device PMD.
4731 * - -EINVAL: The queue_id is out of range.
4733 int rte_eth_rx_burst_mode_get(uint16_t port_id, uint16_t queue_id,
4734 struct rte_eth_burst_mode *mode);
4737 * Retrieve information about the Tx packet burst mode.
4740 * The port identifier of the Ethernet device.
4742 * The Tx queue on the Ethernet device for which information
4743 * will be retrieved.
4745 * A pointer to a structure of type *rte_eth_burst_mode* to be filled
4746 * with the information of the packet burst mode.
4750 * - -ENODEV: If *port_id* is invalid.
4751 * - -ENOTSUP: routine is not supported by the device PMD.
4752 * - -EINVAL: The queue_id is out of range.
4754 int rte_eth_tx_burst_mode_get(uint16_t port_id, uint16_t queue_id,
4755 struct rte_eth_burst_mode *mode);
4759 * @b EXPERIMENTAL: this API may change without prior notice.
4761 * Retrieve the monitor condition for a given receive queue.
4764 * The port identifier of the Ethernet device.
4766 * The Rx queue on the Ethernet device for which information
4767 * will be retrieved.
4769 * The pointer to power-optimized monitoring condition structure.
4773 * -ENOTSUP: Operation not supported.
4774 * -EINVAL: Invalid parameters.
4775 * -ENODEV: Invalid port ID.
4778 int rte_eth_get_monitor_addr(uint16_t port_id, uint16_t queue_id,
4779 struct rte_power_monitor_cond *pmc);
4782 * Retrieve device registers and register attributes (number of registers and
4786 * The port identifier of the Ethernet device.
4788 * Pointer to rte_dev_reg_info structure to fill in. If info->data is
4789 * NULL the function fills in the width and length fields. If non-NULL
4790 * the registers are put into the buffer pointed at by the data field.
4792 * - (0) if successful.
4793 * - (-ENOTSUP) if hardware doesn't support.
4794 * - (-EINVAL) if bad parameter.
4795 * - (-ENODEV) if *port_id* invalid.
4796 * - (-EIO) if device is removed.
4797 * - others depends on the specific operations implementation.
4799 int rte_eth_dev_get_reg_info(uint16_t port_id, struct rte_dev_reg_info *info);
4802 * Retrieve size of device EEPROM
4805 * The port identifier of the Ethernet device.
4807 * - (>=0) EEPROM size if successful.
4808 * - (-ENOTSUP) if hardware doesn't support.
4809 * - (-ENODEV) if *port_id* invalid.
4810 * - (-EIO) if device is removed.
4811 * - others depends on the specific operations implementation.
4813 int rte_eth_dev_get_eeprom_length(uint16_t port_id);
4816 * Retrieve EEPROM and EEPROM attribute
4819 * The port identifier of the Ethernet device.
4821 * The template includes buffer for return EEPROM data and
4822 * EEPROM attributes to be filled.
4824 * - (0) if successful.
4825 * - (-ENOTSUP) if hardware doesn't support.
4826 * - (-EINVAL) if bad parameter.
4827 * - (-ENODEV) if *port_id* invalid.
4828 * - (-EIO) if device is removed.
4829 * - others depends on the specific operations implementation.
4831 int rte_eth_dev_get_eeprom(uint16_t port_id, struct rte_dev_eeprom_info *info);
4834 * Program EEPROM with provided data
4837 * The port identifier of the Ethernet device.
4839 * The template includes EEPROM data for programming and
4840 * EEPROM attributes to be filled
4842 * - (0) if successful.
4843 * - (-ENOTSUP) if hardware doesn't support.
4844 * - (-ENODEV) if *port_id* invalid.
4845 * - (-EINVAL) if bad parameter.
4846 * - (-EIO) if device is removed.
4847 * - others depends on the specific operations implementation.
4849 int rte_eth_dev_set_eeprom(uint16_t port_id, struct rte_dev_eeprom_info *info);
4853 * @b EXPERIMENTAL: this API may change without prior notice.
4855 * Retrieve the type and size of plugin module EEPROM
4858 * The port identifier of the Ethernet device.
4860 * The type and size of plugin module EEPROM.
4862 * - (0) if successful.
4863 * - (-ENOTSUP) if hardware doesn't support.
4864 * - (-ENODEV) if *port_id* invalid.
4865 * - (-EINVAL) if bad parameter.
4866 * - (-EIO) if device is removed.
4867 * - others depends on the specific operations implementation.
4871 rte_eth_dev_get_module_info(uint16_t port_id,
4872 struct rte_eth_dev_module_info *modinfo);
4876 * @b EXPERIMENTAL: this API may change without prior notice.
4878 * Retrieve the data of plugin module EEPROM
4881 * The port identifier of the Ethernet device.
4883 * The template includes the plugin module EEPROM attributes, and the
4884 * buffer for return plugin module EEPROM data.
4886 * - (0) if successful.
4887 * - (-ENOTSUP) if hardware doesn't support.
4888 * - (-EINVAL) if bad parameter.
4889 * - (-ENODEV) if *port_id* invalid.
4890 * - (-EIO) if device is removed.
4891 * - others depends on the specific operations implementation.
4895 rte_eth_dev_get_module_eeprom(uint16_t port_id,
4896 struct rte_dev_eeprom_info *info);
4899 * Set the list of multicast addresses to filter on an Ethernet device.
4902 * The port identifier of the Ethernet device.
4903 * @param mc_addr_set
4904 * The array of multicast addresses to set. Equal to NULL when the function
4905 * is invoked to flush the set of filtered addresses.
4907 * The number of multicast addresses in the *mc_addr_set* array. Equal to 0
4908 * when the function is invoked to flush the set of filtered addresses.
4910 * - (0) if successful.
4911 * - (-ENODEV) if *port_id* invalid.
4912 * - (-EIO) if device is removed.
4913 * - (-ENOTSUP) if PMD of *port_id* doesn't support multicast filtering.
4914 * - (-ENOSPC) if *port_id* has not enough multicast filtering resources.
4915 * - (-EINVAL) if bad parameter.
4917 int rte_eth_dev_set_mc_addr_list(uint16_t port_id,
4918 struct rte_ether_addr *mc_addr_set,
4919 uint32_t nb_mc_addr);
4922 * Enable IEEE1588/802.1AS timestamping for an Ethernet device.
4925 * The port identifier of the Ethernet device.
4929 * - -ENODEV: The port ID is invalid.
4930 * - -EIO: if device is removed.
4931 * - -ENOTSUP: The function is not supported by the Ethernet driver.
4933 int rte_eth_timesync_enable(uint16_t port_id);
4936 * Disable IEEE1588/802.1AS timestamping for an Ethernet device.
4939 * The port identifier of the Ethernet device.
4943 * - -ENODEV: The port ID is invalid.
4944 * - -EIO: if device is removed.
4945 * - -ENOTSUP: The function is not supported by the Ethernet driver.
4947 int rte_eth_timesync_disable(uint16_t port_id);
4950 * Read an IEEE1588/802.1AS Rx timestamp from an Ethernet device.
4953 * The port identifier of the Ethernet device.
4955 * Pointer to the timestamp struct.
4957 * Device specific flags. Used to pass the Rx timesync register index to
4958 * i40e. Unused in igb/ixgbe, pass 0 instead.
4962 * - -EINVAL: No timestamp is available.
4963 * - -ENODEV: The port ID is invalid.
4964 * - -EIO: if device is removed.
4965 * - -ENOTSUP: The function is not supported by the Ethernet driver.
4967 int rte_eth_timesync_read_rx_timestamp(uint16_t port_id,
4968 struct timespec *timestamp, uint32_t flags);
4971 * Read an IEEE1588/802.1AS Tx timestamp from an Ethernet device.
4974 * The port identifier of the Ethernet device.
4976 * Pointer to the timestamp struct.
4980 * - -EINVAL: No timestamp is available.
4981 * - -ENODEV: The port ID is invalid.
4982 * - -EIO: if device is removed.
4983 * - -ENOTSUP: The function is not supported by the Ethernet driver.
4985 int rte_eth_timesync_read_tx_timestamp(uint16_t port_id,
4986 struct timespec *timestamp);
4989 * Adjust the timesync clock on an Ethernet device.
4991 * This is usually used in conjunction with other Ethdev timesync functions to
4992 * synchronize the device time using the IEEE1588/802.1AS protocol.
4995 * The port identifier of the Ethernet device.
4997 * The adjustment in nanoseconds.
5001 * - -ENODEV: The port ID is invalid.
5002 * - -EIO: if device is removed.
5003 * - -ENOTSUP: The function is not supported by the Ethernet driver.
5005 int rte_eth_timesync_adjust_time(uint16_t port_id, int64_t delta);
5008 * Read the time from the timesync clock on an Ethernet device.
5010 * This is usually used in conjunction with other Ethdev timesync functions to
5011 * synchronize the device time using the IEEE1588/802.1AS protocol.
5014 * The port identifier of the Ethernet device.
5016 * Pointer to the timespec struct that holds the time.
5020 * - -EINVAL: Bad parameter.
5022 int rte_eth_timesync_read_time(uint16_t port_id, struct timespec *time);
5025 * Set the time of the timesync clock on an Ethernet device.
5027 * This is usually used in conjunction with other Ethdev timesync functions to
5028 * synchronize the device time using the IEEE1588/802.1AS protocol.
5031 * The port identifier of the Ethernet device.
5033 * Pointer to the timespec struct that holds the time.
5037 * - -EINVAL: No timestamp is available.
5038 * - -ENODEV: The port ID is invalid.
5039 * - -EIO: if device is removed.
5040 * - -ENOTSUP: The function is not supported by the Ethernet driver.
5042 int rte_eth_timesync_write_time(uint16_t port_id, const struct timespec *time);
5046 * @b EXPERIMENTAL: this API may change without prior notice.
5048 * Read the current clock counter of an Ethernet device
5050 * This returns the current raw clock value of an Ethernet device. It is
5051 * a raw amount of ticks, with no given time reference.
5052 * The value returned here is from the same clock than the one
5053 * filling timestamp field of Rx packets when using hardware timestamp
5054 * offload. Therefore it can be used to compute a precise conversion of
5055 * the device clock to the real time.
5057 * E.g, a simple heuristic to derivate the frequency would be:
5058 * uint64_t start, end;
5059 * rte_eth_read_clock(port, start);
5060 * rte_delay_ms(100);
5061 * rte_eth_read_clock(port, end);
5062 * double freq = (end - start) * 10;
5064 * Compute a common reference with:
5065 * uint64_t base_time_sec = current_time();
5066 * uint64_t base_clock;
5067 * rte_eth_read_clock(port, base_clock);
5069 * Then, convert the raw mbuf timestamp with:
5070 * base_time_sec + (double)(*timestamp_dynfield(mbuf) - base_clock) / freq;
5072 * This simple example will not provide a very good accuracy. One must
5073 * at least measure multiple times the frequency and do a regression.
5074 * To avoid deviation from the system time, the common reference can
5075 * be repeated from time to time. The integer division can also be
5076 * converted by a multiplication and a shift for better performance.
5079 * The port identifier of the Ethernet device.
5081 * Pointer to the uint64_t that holds the raw clock value.
5085 * - -ENODEV: The port ID is invalid.
5086 * - -ENOTSUP: The function is not supported by the Ethernet driver.
5087 * - -EINVAL: if bad parameter.
5091 rte_eth_read_clock(uint16_t port_id, uint64_t *clock);
5094 * Get the port ID from device name. The device name should be specified
5096 * - PCIe address (Domain:Bus:Device.Function), for example- 0000:2:00.0
5097 * - SoC device name, for example- fsl-gmac0
5098 * - vdev dpdk name, for example- net_[pcap0|null0|tap0]
5101 * pci address or name of the device
5103 * pointer to port identifier of the device
5105 * - (0) if successful and port_id is filled.
5106 * - (-ENODEV or -EINVAL) on failure.
5109 rte_eth_dev_get_port_by_name(const char *name, uint16_t *port_id);
5112 * Get the device name from port ID. The device name is specified as below:
5113 * - PCIe address (Domain:Bus:Device.Function), for example- 0000:02:00.0
5114 * - SoC device name, for example- fsl-gmac0
5115 * - vdev dpdk name, for example- net_[pcap0|null0|tun0|tap0]
5118 * Port identifier of the device.
5120 * Buffer of size RTE_ETH_NAME_MAX_LEN to store the name.
5122 * - (0) if successful.
5123 * - (-ENODEV) if *port_id* is invalid.
5124 * - (-EINVAL) on failure.
5127 rte_eth_dev_get_name_by_port(uint16_t port_id, char *name);
5130 * Check that numbers of Rx and Tx descriptors satisfy descriptors limits from
5131 * the Ethernet device information, otherwise adjust them to boundaries.
5134 * The port identifier of the Ethernet device.
5136 * A pointer to a uint16_t where the number of receive
5137 * descriptors stored.
5139 * A pointer to a uint16_t where the number of transmit
5140 * descriptors stored.
5142 * - (0) if successful.
5143 * - (-ENOTSUP, -ENODEV or -EINVAL) on failure.
5145 int rte_eth_dev_adjust_nb_rx_tx_desc(uint16_t port_id,
5146 uint16_t *nb_rx_desc,
5147 uint16_t *nb_tx_desc);
5150 * Test if a port supports specific mempool ops.
5153 * Port identifier of the Ethernet device.
5155 * The name of the pool operations to test.
5157 * - 0: best mempool ops choice for this port.
5158 * - 1: mempool ops are supported for this port.
5159 * - -ENOTSUP: mempool ops not supported for this port.
5160 * - -ENODEV: Invalid port Identifier.
5161 * - -EINVAL: Pool param is null.
5164 rte_eth_dev_pool_ops_supported(uint16_t port_id, const char *pool);
5167 * Get the security context for the Ethernet device.
5170 * Port identifier of the Ethernet device
5173 * - pointer to security context on success.
5176 rte_eth_dev_get_sec_ctx(uint16_t port_id);
5180 * @b EXPERIMENTAL: this API may change, or be removed, without prior notice
5182 * Query the device hairpin capabilities.
5185 * The port identifier of the Ethernet device.
5187 * Pointer to a structure that will hold the hairpin capabilities.
5189 * - (0) if successful.
5190 * - (-ENOTSUP) if hardware doesn't support.
5191 * - (-EINVAL) if bad parameter.
5194 int rte_eth_dev_hairpin_capability_get(uint16_t port_id,
5195 struct rte_eth_hairpin_cap *cap);
5199 * @b EXPERIMENTAL: this structure may change without prior notice.
5201 * Ethernet device representor ID range entry
5203 struct rte_eth_representor_range {
5204 enum rte_eth_representor_type type; /**< Representor type */
5205 int controller; /**< Controller index */
5206 int pf; /**< Physical function index */
5209 int vf; /**< VF start index */
5210 int sf; /**< SF start index */
5212 uint32_t id_base; /**< Representor ID start index */
5213 uint32_t id_end; /**< Representor ID end index */
5214 char name[RTE_DEV_NAME_MAX_LEN]; /**< Representor name */
5219 * @b EXPERIMENTAL: this structure may change without prior notice.
5221 * Ethernet device representor information
5223 struct rte_eth_representor_info {
5224 uint16_t controller; /**< Controller ID of caller device. */
5225 uint16_t pf; /**< Physical function ID of caller device. */
5226 uint32_t nb_ranges_alloc; /**< Size of the ranges array. */
5227 uint32_t nb_ranges; /**< Number of initialized ranges. */
5228 struct rte_eth_representor_range ranges[];/**< Representor ID range. */
5232 * Retrieve the representor info of the device.
5234 * Get device representor info to be able to calculate a unique
5235 * representor ID. @see rte_eth_representor_id_get helper.
5238 * The port identifier of the device.
5240 * A pointer to a representor info structure.
5241 * NULL to return number of range entries and allocate memory
5242 * for next call to store detail.
5243 * The number of ranges that were written into this structure
5244 * will be placed into its nb_ranges field. This number cannot be
5245 * larger than the nb_ranges_alloc that by the user before calling
5246 * this function. It can be smaller than the value returned by the
5247 * function, however.
5249 * - (-ENOTSUP) if operation is not supported.
5250 * - (-ENODEV) if *port_id* invalid.
5251 * - (-EIO) if device is removed.
5252 * - (>=0) number of available representor range entries.
5255 int rte_eth_representor_info_get(uint16_t port_id,
5256 struct rte_eth_representor_info *info);
5258 /** The NIC is able to deliver flag (if set) with packets to the PMD. */
5259 #define RTE_ETH_RX_METADATA_USER_FLAG RTE_BIT64(0)
5261 /** The NIC is able to deliver mark ID with packets to the PMD. */
5262 #define RTE_ETH_RX_METADATA_USER_MARK RTE_BIT64(1)
5264 /** The NIC is able to deliver tunnel ID with packets to the PMD. */
5265 #define RTE_ETH_RX_METADATA_TUNNEL_ID RTE_BIT64(2)
5269 * @b EXPERIMENTAL: this API may change without prior notice
5271 * Negotiate the NIC's ability to deliver specific kinds of metadata to the PMD.
5273 * Invoke this API before the first rte_eth_dev_configure() invocation
5274 * to let the PMD make preparations that are inconvenient to do later.
5276 * The negotiation process is as follows:
5278 * - the application requests features intending to use at least some of them;
5279 * - the PMD responds with the guaranteed subset of the requested feature set;
5280 * - the application can retry negotiation with another set of features;
5281 * - the application can pass zero to clear the negotiation result;
5282 * - the last negotiated result takes effect upon
5283 * the ethdev configure and start.
5286 * The PMD is supposed to first consider enabling the requested feature set
5287 * in its entirety. Only if it fails to do so, does it have the right to
5288 * respond with a smaller set of the originally requested features.
5291 * Return code (-ENOTSUP) does not necessarily mean that the requested
5292 * features are unsupported. In this case, the application should just
5293 * assume that these features can be used without prior negotiations.
5296 * Port (ethdev) identifier
5298 * @param[inout] features
5299 * Feature selection buffer
5302 * - (-EBUSY) if the port can't handle this in its current state;
5303 * - (-ENOTSUP) if the method itself is not supported by the PMD;
5304 * - (-ENODEV) if *port_id* is invalid;
5305 * - (-EINVAL) if *features* is NULL;
5306 * - (-EIO) if the device is removed;
5310 int rte_eth_rx_metadata_negotiate(uint16_t port_id, uint64_t *features);
5312 /** Flag to offload IP reassembly for IPv4 packets. */
5313 #define RTE_ETH_DEV_REASSEMBLY_F_IPV4 (RTE_BIT32(0))
5314 /** Flag to offload IP reassembly for IPv6 packets. */
5315 #define RTE_ETH_DEV_REASSEMBLY_F_IPV6 (RTE_BIT32(1))
5318 * A structure used to get/set IP reassembly configuration. It is also used
5319 * to get the maximum capability values that a PMD can support.
5321 * If rte_eth_ip_reassembly_capability_get() returns 0, IP reassembly can be
5322 * enabled using rte_eth_ip_reassembly_conf_set() and params values lower than
5323 * capability params can be set in the PMD.
5325 struct rte_eth_ip_reassembly_params {
5326 /** Maximum time in ms which PMD can wait for other fragments. */
5327 uint32_t timeout_ms;
5328 /** Maximum number of fragments that can be reassembled. */
5331 * Flags to enable reassembly of packet types -
5332 * RTE_ETH_DEV_REASSEMBLY_F_xxx.
5339 * @b EXPERIMENTAL: this API may change without prior notice
5341 * Get IP reassembly capabilities supported by the PMD. This is the first API
5342 * to be called for enabling the IP reassembly offload feature. PMD will return
5343 * the maximum values of parameters that PMD can support and user can call
5344 * rte_eth_ip_reassembly_conf_set() with param values lower than capability.
5347 * The port identifier of the device.
5349 * A pointer to rte_eth_ip_reassembly_params structure.
5351 * - (-ENOTSUP) if offload configuration is not supported by device.
5352 * - (-ENODEV) if *port_id* invalid.
5353 * - (-EIO) if device is removed.
5354 * - (-EINVAL) if device is not configured or *capa* passed is NULL.
5358 int rte_eth_ip_reassembly_capability_get(uint16_t port_id,
5359 struct rte_eth_ip_reassembly_params *capa);
5363 * @b EXPERIMENTAL: this API may change without prior notice
5365 * Get IP reassembly configuration parameters currently set in PMD.
5366 * The API will return error if the configuration is not already
5367 * set using rte_eth_ip_reassembly_conf_set() before calling this API or if
5368 * the device is not configured.
5371 * The port identifier of the device.
5373 * A pointer to rte_eth_ip_reassembly_params structure.
5375 * - (-ENOTSUP) if offload configuration is not supported by device.
5376 * - (-ENODEV) if *port_id* invalid.
5377 * - (-EIO) if device is removed.
5378 * - (-EINVAL) if device is not configured or if *conf* passed is NULL or if
5379 * configuration is not set using rte_eth_ip_reassembly_conf_set().
5383 int rte_eth_ip_reassembly_conf_get(uint16_t port_id,
5384 struct rte_eth_ip_reassembly_params *conf);
5388 * @b EXPERIMENTAL: this API may change without prior notice
5390 * Set IP reassembly configuration parameters if the PMD supports IP reassembly
5391 * offload. User should first call rte_eth_ip_reassembly_capability_get() to
5392 * check the maximum values supported by the PMD before setting the
5393 * configuration. The use of this API is mandatory to enable this feature and
5394 * should be called before rte_eth_dev_start().
5396 * In datapath, PMD cannot guarantee that IP reassembly is always successful.
5397 * Hence, PMD shall register mbuf dynamic field and dynamic flag using
5398 * rte_eth_ip_reassembly_dynfield_register() to denote incomplete IP reassembly.
5399 * If dynfield is not successfully registered, error will be returned and
5400 * IP reassembly offload cannot be used.
5403 * The port identifier of the device.
5405 * A pointer to rte_eth_ip_reassembly_params structure.
5407 * - (-ENOTSUP) if offload configuration is not supported by device.
5408 * - (-ENODEV) if *port_id* invalid.
5409 * - (-EIO) if device is removed.
5410 * - (-EINVAL) if device is not configured or if device is already started or
5411 * if *conf* passed is NULL or if mbuf dynfield is not registered
5412 * successfully by the PMD.
5416 int rte_eth_ip_reassembly_conf_set(uint16_t port_id,
5417 const struct rte_eth_ip_reassembly_params *conf);
5420 * In case of IP reassembly offload failure, packet will be updated with
5421 * dynamic flag - RTE_MBUF_DYNFLAG_IP_REASSEMBLY_INCOMPLETE_NAME and packets
5422 * will be returned without alteration.
5423 * The application can retrieve the attached fragments using mbuf dynamic field
5424 * RTE_MBUF_DYNFIELD_IP_REASSEMBLY_NAME.
5428 * Next fragment packet. Application should fetch dynamic field of
5429 * each fragment until a NULL is received and nb_frags is 0.
5431 struct rte_mbuf *next_frag;
5432 /** Time spent(in ms) by HW in waiting for further fragments. */
5433 uint16_t time_spent;
5434 /** Number of more fragments attached in mbuf dynamic fields. */
5436 } rte_eth_ip_reassembly_dynfield_t;
5440 * @b EXPERIMENTAL: this API may change, or be removed, without prior notice
5442 * Dump private info from device to a file. Provided data and the order depends
5446 * The port identifier of the Ethernet device.
5448 * A pointer to a file for output.
5451 * - (-ENODEV) if *port_id* is invalid.
5452 * - (-EINVAL) if null file.
5453 * - (-ENOTSUP) if the device does not support this function.
5454 * - (-EIO) if device is removed.
5457 int rte_eth_dev_priv_dump(uint16_t port_id, FILE *file);
5459 #include <rte_ethdev_core.h>
5463 * Helper routine for rte_eth_rx_burst().
5464 * Should be called at exit from PMD's rte_eth_rx_bulk implementation.
5465 * Does necessary post-processing - invokes Rx callbacks if any, etc.
5468 * The port identifier of the Ethernet device.
5470 * The index of the receive queue from which to retrieve input packets.
5472 * The address of an array of pointers to *rte_mbuf* structures that
5473 * have been retrieved from the device.
5475 * The number of packets that were retrieved from the device.
5477 * The number of elements in @p rx_pkts array.
5479 * Opaque pointer of Rx queue callback related data.
5482 * The number of packets effectively supplied to the @p rx_pkts array.
5484 uint16_t rte_eth_call_rx_callbacks(uint16_t port_id, uint16_t queue_id,
5485 struct rte_mbuf **rx_pkts, uint16_t nb_rx, uint16_t nb_pkts,
5490 * Retrieve a burst of input packets from a receive queue of an Ethernet
5491 * device. The retrieved packets are stored in *rte_mbuf* structures whose
5492 * pointers are supplied in the *rx_pkts* array.
5494 * The rte_eth_rx_burst() function loops, parsing the Rx ring of the
5495 * receive queue, up to *nb_pkts* packets, and for each completed Rx
5496 * descriptor in the ring, it performs the following operations:
5498 * - Initialize the *rte_mbuf* data structure associated with the
5499 * Rx descriptor according to the information provided by the NIC into
5500 * that Rx descriptor.
5502 * - Store the *rte_mbuf* data structure into the next entry of the
5505 * - Replenish the Rx descriptor with a new *rte_mbuf* buffer
5506 * allocated from the memory pool associated with the receive queue at
5507 * initialization time.
5509 * When retrieving an input packet that was scattered by the controller
5510 * into multiple receive descriptors, the rte_eth_rx_burst() function
5511 * appends the associated *rte_mbuf* buffers to the first buffer of the
5514 * The rte_eth_rx_burst() function returns the number of packets
5515 * actually retrieved, which is the number of *rte_mbuf* data structures
5516 * effectively supplied into the *rx_pkts* array.
5517 * A return value equal to *nb_pkts* indicates that the Rx queue contained
5518 * at least *rx_pkts* packets, and this is likely to signify that other
5519 * received packets remain in the input queue. Applications implementing
5520 * a "retrieve as much received packets as possible" policy can check this
5521 * specific case and keep invoking the rte_eth_rx_burst() function until
5522 * a value less than *nb_pkts* is returned.
5524 * This receive method has the following advantages:
5526 * - It allows a run-to-completion network stack engine to retrieve and
5527 * to immediately process received packets in a fast burst-oriented
5528 * approach, avoiding the overhead of unnecessary intermediate packet
5529 * queue/dequeue operations.
5531 * - Conversely, it also allows an asynchronous-oriented processing
5532 * method to retrieve bursts of received packets and to immediately
5533 * queue them for further parallel processing by another logical core,
5534 * for instance. However, instead of having received packets being
5535 * individually queued by the driver, this approach allows the caller
5536 * of the rte_eth_rx_burst() function to queue a burst of retrieved
5537 * packets at a time and therefore dramatically reduce the cost of
5538 * enqueue/dequeue operations per packet.
5540 * - It allows the rte_eth_rx_burst() function of the driver to take
5541 * advantage of burst-oriented hardware features (CPU cache,
5542 * prefetch instructions, and so on) to minimize the number of CPU
5543 * cycles per packet.
5545 * To summarize, the proposed receive API enables many
5546 * burst-oriented optimizations in both synchronous and asynchronous
5547 * packet processing environments with no overhead in both cases.
5550 * Some drivers using vector instructions require that *nb_pkts* is
5551 * divisible by 4 or 8, depending on the driver implementation.
5553 * The rte_eth_rx_burst() function does not provide any error
5554 * notification to avoid the corresponding overhead. As a hint, the
5555 * upper-level application might check the status of the device link once
5556 * being systematically returned a 0 value for a given number of tries.
5559 * The port identifier of the Ethernet device.
5561 * The index of the receive queue from which to retrieve input packets.
5562 * The value must be in the range [0, nb_rx_queue - 1] previously supplied
5563 * to rte_eth_dev_configure().
5565 * The address of an array of pointers to *rte_mbuf* structures that
5566 * must be large enough to store *nb_pkts* pointers in it.
5568 * The maximum number of packets to retrieve.
5569 * The value must be divisible by 8 in order to work with any driver.
5571 * The number of packets actually retrieved, which is the number
5572 * of pointers to *rte_mbuf* structures effectively supplied to the
5575 static inline uint16_t
5576 rte_eth_rx_burst(uint16_t port_id, uint16_t queue_id,
5577 struct rte_mbuf **rx_pkts, const uint16_t nb_pkts)
5580 struct rte_eth_fp_ops *p;
5583 #ifdef RTE_ETHDEV_DEBUG_RX
5584 if (port_id >= RTE_MAX_ETHPORTS ||
5585 queue_id >= RTE_MAX_QUEUES_PER_PORT) {
5587 "Invalid port_id=%u or queue_id=%u\n",
5593 /* fetch pointer to queue data */
5594 p = &rte_eth_fp_ops[port_id];
5595 qd = p->rxq.data[queue_id];
5597 #ifdef RTE_ETHDEV_DEBUG_RX
5598 RTE_ETH_VALID_PORTID_OR_ERR_RET(port_id, 0);
5601 RTE_ETHDEV_LOG(ERR, "Invalid Rx queue_id=%u for port_id=%u\n",
5607 nb_rx = p->rx_pkt_burst(qd, rx_pkts, nb_pkts);
5609 #ifdef RTE_ETHDEV_RXTX_CALLBACKS
5613 /* __ATOMIC_RELEASE memory order was used when the
5614 * call back was inserted into the list.
5615 * Since there is a clear dependency between loading
5616 * cb and cb->fn/cb->next, __ATOMIC_ACQUIRE memory order is
5619 cb = __atomic_load_n((void **)&p->rxq.clbk[queue_id],
5621 if (unlikely(cb != NULL))
5622 nb_rx = rte_eth_call_rx_callbacks(port_id, queue_id,
5623 rx_pkts, nb_rx, nb_pkts, cb);
5627 rte_ethdev_trace_rx_burst(port_id, queue_id, (void **)rx_pkts, nb_rx);
5632 * Get the number of used descriptors of a Rx queue
5635 * The port identifier of the Ethernet device.
5637 * The queue ID on the specific port.
5639 * The number of used descriptors in the specific queue, or:
5640 * - (-ENODEV) if *port_id* is invalid.
5641 * (-EINVAL) if *queue_id* is invalid
5642 * (-ENOTSUP) if the device does not support this function
5645 rte_eth_rx_queue_count(uint16_t port_id, uint16_t queue_id)
5647 struct rte_eth_fp_ops *p;
5650 if (port_id >= RTE_MAX_ETHPORTS ||
5651 queue_id >= RTE_MAX_QUEUES_PER_PORT) {
5653 "Invalid port_id=%u or queue_id=%u\n",
5658 /* fetch pointer to queue data */
5659 p = &rte_eth_fp_ops[port_id];
5660 qd = p->rxq.data[queue_id];
5662 RTE_ETH_VALID_PORTID_OR_ERR_RET(port_id, -ENODEV);
5663 RTE_FUNC_PTR_OR_ERR_RET(*p->rx_queue_count, -ENOTSUP);
5667 return (int)(*p->rx_queue_count)(qd);
5670 /**@{@name Rx hardware descriptor states
5671 * @see rte_eth_rx_descriptor_status
5673 #define RTE_ETH_RX_DESC_AVAIL 0 /**< Desc available for hw. */
5674 #define RTE_ETH_RX_DESC_DONE 1 /**< Desc done, filled by hw. */
5675 #define RTE_ETH_RX_DESC_UNAVAIL 2 /**< Desc used by driver or hw. */
5679 * Check the status of a Rx descriptor in the queue
5681 * It should be called in a similar context than the Rx function:
5682 * - on a dataplane core
5683 * - not concurrently on the same queue
5685 * Since it's a dataplane function, no check is performed on port_id and
5686 * queue_id. The caller must therefore ensure that the port is enabled
5687 * and the queue is configured and running.
5689 * Note: accessing to a random descriptor in the ring may trigger cache
5690 * misses and have a performance impact.
5693 * A valid port identifier of the Ethernet device which.
5695 * A valid Rx queue identifier on this port.
5697 * The offset of the descriptor starting from tail (0 is the next
5698 * packet to be received by the driver).
5701 * - (RTE_ETH_RX_DESC_AVAIL): Descriptor is available for the hardware to
5703 * - (RTE_ETH_RX_DESC_DONE): Descriptor is done, it is filled by hw, but
5704 * not yet processed by the driver (i.e. in the receive queue).
5705 * - (RTE_ETH_RX_DESC_UNAVAIL): Descriptor is unavailable, either hold by
5706 * the driver and not yet returned to hw, or reserved by the hw.
5707 * - (-EINVAL) bad descriptor offset.
5708 * - (-ENOTSUP) if the device does not support this function.
5709 * - (-ENODEV) bad port or queue (only if compiled with debug).
5712 rte_eth_rx_descriptor_status(uint16_t port_id, uint16_t queue_id,
5715 struct rte_eth_fp_ops *p;
5718 #ifdef RTE_ETHDEV_DEBUG_RX
5719 if (port_id >= RTE_MAX_ETHPORTS ||
5720 queue_id >= RTE_MAX_QUEUES_PER_PORT) {
5722 "Invalid port_id=%u or queue_id=%u\n",
5728 /* fetch pointer to queue data */
5729 p = &rte_eth_fp_ops[port_id];
5730 qd = p->rxq.data[queue_id];
5732 #ifdef RTE_ETHDEV_DEBUG_RX
5733 RTE_ETH_VALID_PORTID_OR_ERR_RET(port_id, -ENODEV);
5737 RTE_FUNC_PTR_OR_ERR_RET(*p->rx_descriptor_status, -ENOTSUP);
5738 return (*p->rx_descriptor_status)(qd, offset);
5741 /**@{@name Tx hardware descriptor states
5742 * @see rte_eth_tx_descriptor_status
5744 #define RTE_ETH_TX_DESC_FULL 0 /**< Desc filled for hw, waiting xmit. */
5745 #define RTE_ETH_TX_DESC_DONE 1 /**< Desc done, packet is transmitted. */
5746 #define RTE_ETH_TX_DESC_UNAVAIL 2 /**< Desc used by driver or hw. */
5750 * Check the status of a Tx descriptor in the queue.
5752 * It should be called in a similar context than the Tx function:
5753 * - on a dataplane core
5754 * - not concurrently on the same queue
5756 * Since it's a dataplane function, no check is performed on port_id and
5757 * queue_id. The caller must therefore ensure that the port is enabled
5758 * and the queue is configured and running.
5760 * Note: accessing to a random descriptor in the ring may trigger cache
5761 * misses and have a performance impact.
5764 * A valid port identifier of the Ethernet device which.
5766 * A valid Tx queue identifier on this port.
5768 * The offset of the descriptor starting from tail (0 is the place where
5769 * the next packet will be send).
5772 * - (RTE_ETH_TX_DESC_FULL) Descriptor is being processed by the hw, i.e.
5773 * in the transmit queue.
5774 * - (RTE_ETH_TX_DESC_DONE) Hardware is done with this descriptor, it can
5775 * be reused by the driver.
5776 * - (RTE_ETH_TX_DESC_UNAVAIL): Descriptor is unavailable, reserved by the
5777 * driver or the hardware.
5778 * - (-EINVAL) bad descriptor offset.
5779 * - (-ENOTSUP) if the device does not support this function.
5780 * - (-ENODEV) bad port or queue (only if compiled with debug).
5782 static inline int rte_eth_tx_descriptor_status(uint16_t port_id,
5783 uint16_t queue_id, uint16_t offset)
5785 struct rte_eth_fp_ops *p;
5788 #ifdef RTE_ETHDEV_DEBUG_TX
5789 if (port_id >= RTE_MAX_ETHPORTS ||
5790 queue_id >= RTE_MAX_QUEUES_PER_PORT) {
5792 "Invalid port_id=%u or queue_id=%u\n",
5798 /* fetch pointer to queue data */
5799 p = &rte_eth_fp_ops[port_id];
5800 qd = p->txq.data[queue_id];
5802 #ifdef RTE_ETHDEV_DEBUG_TX
5803 RTE_ETH_VALID_PORTID_OR_ERR_RET(port_id, -ENODEV);
5807 RTE_FUNC_PTR_OR_ERR_RET(*p->tx_descriptor_status, -ENOTSUP);
5808 return (*p->tx_descriptor_status)(qd, offset);
5813 * Helper routine for rte_eth_tx_burst().
5814 * Should be called before entry PMD's rte_eth_tx_bulk implementation.
5815 * Does necessary pre-processing - invokes Tx callbacks if any, etc.
5818 * The port identifier of the Ethernet device.
5820 * The index of the transmit queue through which output packets must be
5823 * The address of an array of *nb_pkts* pointers to *rte_mbuf* structures
5824 * which contain the output packets.
5826 * The maximum number of packets to transmit.
5828 * The number of output packets to transmit.
5830 uint16_t rte_eth_call_tx_callbacks(uint16_t port_id, uint16_t queue_id,
5831 struct rte_mbuf **tx_pkts, uint16_t nb_pkts, void *opaque);
5834 * Send a burst of output packets on a transmit queue of an Ethernet device.
5836 * The rte_eth_tx_burst() function is invoked to transmit output packets
5837 * on the output queue *queue_id* of the Ethernet device designated by its
5839 * The *nb_pkts* parameter is the number of packets to send which are
5840 * supplied in the *tx_pkts* array of *rte_mbuf* structures, each of them
5841 * allocated from a pool created with rte_pktmbuf_pool_create().
5842 * The rte_eth_tx_burst() function loops, sending *nb_pkts* packets,
5843 * up to the number of transmit descriptors available in the Tx ring of the
5845 * For each packet to send, the rte_eth_tx_burst() function performs
5846 * the following operations:
5848 * - Pick up the next available descriptor in the transmit ring.
5850 * - Free the network buffer previously sent with that descriptor, if any.
5852 * - Initialize the transmit descriptor with the information provided
5853 * in the *rte_mbuf data structure.
5855 * In the case of a segmented packet composed of a list of *rte_mbuf* buffers,
5856 * the rte_eth_tx_burst() function uses several transmit descriptors
5859 * The rte_eth_tx_burst() function returns the number of packets it
5860 * actually sent. A return value equal to *nb_pkts* means that all packets
5861 * have been sent, and this is likely to signify that other output packets
5862 * could be immediately transmitted again. Applications that implement a
5863 * "send as many packets to transmit as possible" policy can check this
5864 * specific case and keep invoking the rte_eth_tx_burst() function until
5865 * a value less than *nb_pkts* is returned.
5867 * It is the responsibility of the rte_eth_tx_burst() function to
5868 * transparently free the memory buffers of packets previously sent.
5869 * This feature is driven by the *tx_free_thresh* value supplied to the
5870 * rte_eth_dev_configure() function at device configuration time.
5871 * When the number of free Tx descriptors drops below this threshold, the
5872 * rte_eth_tx_burst() function must [attempt to] free the *rte_mbuf* buffers
5873 * of those packets whose transmission was effectively completed.
5875 * If the PMD is RTE_ETH_TX_OFFLOAD_MT_LOCKFREE capable, multiple threads can
5876 * invoke this function concurrently on the same Tx queue without SW lock.
5877 * @see rte_eth_dev_info_get, struct rte_eth_txconf::offloads
5879 * @see rte_eth_tx_prepare to perform some prior checks or adjustments
5883 * The port identifier of the Ethernet device.
5885 * The index of the transmit queue through which output packets must be
5887 * The value must be in the range [0, nb_tx_queue - 1] previously supplied
5888 * to rte_eth_dev_configure().
5890 * The address of an array of *nb_pkts* pointers to *rte_mbuf* structures
5891 * which contain the output packets.
5893 * The maximum number of packets to transmit.
5895 * The number of output packets actually stored in transmit descriptors of
5896 * the transmit ring. The return value can be less than the value of the
5897 * *tx_pkts* parameter when the transmit ring is full or has been filled up.
5899 static inline uint16_t
5900 rte_eth_tx_burst(uint16_t port_id, uint16_t queue_id,
5901 struct rte_mbuf **tx_pkts, uint16_t nb_pkts)
5903 struct rte_eth_fp_ops *p;
5906 #ifdef RTE_ETHDEV_DEBUG_TX
5907 if (port_id >= RTE_MAX_ETHPORTS ||
5908 queue_id >= RTE_MAX_QUEUES_PER_PORT) {
5910 "Invalid port_id=%u or queue_id=%u\n",
5916 /* fetch pointer to queue data */
5917 p = &rte_eth_fp_ops[port_id];
5918 qd = p->txq.data[queue_id];
5920 #ifdef RTE_ETHDEV_DEBUG_TX
5921 RTE_ETH_VALID_PORTID_OR_ERR_RET(port_id, 0);
5924 RTE_ETHDEV_LOG(ERR, "Invalid Tx queue_id=%u for port_id=%u\n",
5930 #ifdef RTE_ETHDEV_RXTX_CALLBACKS
5934 /* __ATOMIC_RELEASE memory order was used when the
5935 * call back was inserted into the list.
5936 * Since there is a clear dependency between loading
5937 * cb and cb->fn/cb->next, __ATOMIC_ACQUIRE memory order is
5940 cb = __atomic_load_n((void **)&p->txq.clbk[queue_id],
5942 if (unlikely(cb != NULL))
5943 nb_pkts = rte_eth_call_tx_callbacks(port_id, queue_id,
5944 tx_pkts, nb_pkts, cb);
5948 nb_pkts = p->tx_pkt_burst(qd, tx_pkts, nb_pkts);
5950 rte_ethdev_trace_tx_burst(port_id, queue_id, (void **)tx_pkts, nb_pkts);
5955 * Process a burst of output packets on a transmit queue of an Ethernet device.
5957 * The rte_eth_tx_prepare() function is invoked to prepare output packets to be
5958 * transmitted on the output queue *queue_id* of the Ethernet device designated
5960 * The *nb_pkts* parameter is the number of packets to be prepared which are
5961 * supplied in the *tx_pkts* array of *rte_mbuf* structures, each of them
5962 * allocated from a pool created with rte_pktmbuf_pool_create().
5963 * For each packet to send, the rte_eth_tx_prepare() function performs
5964 * the following operations:
5966 * - Check if packet meets devices requirements for Tx offloads.
5968 * - Check limitations about number of segments.
5970 * - Check additional requirements when debug is enabled.
5972 * - Update and/or reset required checksums when Tx offload is set for packet.
5974 * Since this function can modify packet data, provided mbufs must be safely
5975 * writable (e.g. modified data cannot be in shared segment).
5977 * The rte_eth_tx_prepare() function returns the number of packets ready to be
5978 * sent. A return value equal to *nb_pkts* means that all packets are valid and
5979 * ready to be sent, otherwise stops processing on the first invalid packet and
5980 * leaves the rest packets untouched.
5982 * When this functionality is not implemented in the driver, all packets are
5983 * are returned untouched.
5986 * The port identifier of the Ethernet device.
5987 * The value must be a valid port ID.
5989 * The index of the transmit queue through which output packets must be
5991 * The value must be in the range [0, nb_tx_queue - 1] previously supplied
5992 * to rte_eth_dev_configure().
5994 * The address of an array of *nb_pkts* pointers to *rte_mbuf* structures
5995 * which contain the output packets.
5997 * The maximum number of packets to process.
5999 * The number of packets correct and ready to be sent. The return value can be
6000 * less than the value of the *tx_pkts* parameter when some packet doesn't
6001 * meet devices requirements with rte_errno set appropriately:
6002 * - EINVAL: offload flags are not correctly set
6003 * - ENOTSUP: the offload feature is not supported by the hardware
6004 * - ENODEV: if *port_id* is invalid (with debug enabled only)
6008 #ifndef RTE_ETHDEV_TX_PREPARE_NOOP
6010 static inline uint16_t
6011 rte_eth_tx_prepare(uint16_t port_id, uint16_t queue_id,
6012 struct rte_mbuf **tx_pkts, uint16_t nb_pkts)
6014 struct rte_eth_fp_ops *p;
6017 #ifdef RTE_ETHDEV_DEBUG_TX
6018 if (port_id >= RTE_MAX_ETHPORTS ||
6019 queue_id >= RTE_MAX_QUEUES_PER_PORT) {
6021 "Invalid port_id=%u or queue_id=%u\n",
6028 /* fetch pointer to queue data */
6029 p = &rte_eth_fp_ops[port_id];
6030 qd = p->txq.data[queue_id];
6032 #ifdef RTE_ETHDEV_DEBUG_TX
6033 if (!rte_eth_dev_is_valid_port(port_id)) {
6034 RTE_ETHDEV_LOG(ERR, "Invalid Tx port_id=%u\n", port_id);
6039 RTE_ETHDEV_LOG(ERR, "Invalid Tx queue_id=%u for port_id=%u\n",
6046 if (!p->tx_pkt_prepare)
6049 return p->tx_pkt_prepare(qd, tx_pkts, nb_pkts);
6055 * Native NOOP operation for compilation targets which doesn't require any
6056 * preparations steps, and functional NOOP may introduce unnecessary performance
6059 * Generally this is not a good idea to turn it on globally and didn't should
6060 * be used if behavior of tx_preparation can change.
6063 static inline uint16_t
6064 rte_eth_tx_prepare(__rte_unused uint16_t port_id,
6065 __rte_unused uint16_t queue_id,
6066 __rte_unused struct rte_mbuf **tx_pkts, uint16_t nb_pkts)
6074 * Send any packets queued up for transmission on a port and HW queue
6076 * This causes an explicit flush of packets previously buffered via the
6077 * rte_eth_tx_buffer() function. It returns the number of packets successfully
6078 * sent to the NIC, and calls the error callback for any unsent packets. Unless
6079 * explicitly set up otherwise, the default callback simply frees the unsent
6080 * packets back to the owning mempool.
6083 * The port identifier of the Ethernet device.
6085 * The index of the transmit queue through which output packets must be
6087 * The value must be in the range [0, nb_tx_queue - 1] previously supplied
6088 * to rte_eth_dev_configure().
6090 * Buffer of packets to be transmit.
6092 * The number of packets successfully sent to the Ethernet device. The error
6093 * callback is called for any packets which could not be sent.
6095 static inline uint16_t
6096 rte_eth_tx_buffer_flush(uint16_t port_id, uint16_t queue_id,
6097 struct rte_eth_dev_tx_buffer *buffer)
6100 uint16_t to_send = buffer->length;
6105 sent = rte_eth_tx_burst(port_id, queue_id, buffer->pkts, to_send);
6109 /* All packets sent, or to be dealt with by callback below */
6110 if (unlikely(sent != to_send))
6111 buffer->error_callback(&buffer->pkts[sent],
6112 (uint16_t)(to_send - sent),
6113 buffer->error_userdata);
6119 * Buffer a single packet for future transmission on a port and queue
6121 * This function takes a single mbuf/packet and buffers it for later
6122 * transmission on the particular port and queue specified. Once the buffer is
6123 * full of packets, an attempt will be made to transmit all the buffered
6124 * packets. In case of error, where not all packets can be transmitted, a
6125 * callback is called with the unsent packets as a parameter. If no callback
6126 * is explicitly set up, the unsent packets are just freed back to the owning
6127 * mempool. The function returns the number of packets actually sent i.e.
6128 * 0 if no buffer flush occurred, otherwise the number of packets successfully
6132 * The port identifier of the Ethernet device.
6134 * The index of the transmit queue through which output packets must be
6136 * The value must be in the range [0, nb_tx_queue - 1] previously supplied
6137 * to rte_eth_dev_configure().
6139 * Buffer used to collect packets to be sent.
6141 * Pointer to the packet mbuf to be sent.
6143 * 0 = packet has been buffered for later transmission
6144 * N > 0 = packet has been buffered, and the buffer was subsequently flushed,
6145 * causing N packets to be sent, and the error callback to be called for
6148 static __rte_always_inline uint16_t
6149 rte_eth_tx_buffer(uint16_t port_id, uint16_t queue_id,
6150 struct rte_eth_dev_tx_buffer *buffer, struct rte_mbuf *tx_pkt)
6152 buffer->pkts[buffer->length++] = tx_pkt;
6153 if (buffer->length < buffer->size)
6156 return rte_eth_tx_buffer_flush(port_id, queue_id, buffer);
6163 #endif /* _RTE_ETHDEV_H_ */