4 * Copyright(c) 2010-2012 Intel Corporation. All rights reserved.
7 * Redistribution and use in source and binary forms, with or without
8 * modification, are permitted provided that the following conditions
11 * * Redistributions of source code must retain the above copyright
12 * notice, this list of conditions and the following disclaimer.
13 * * Redistributions in binary form must reproduce the above copyright
14 * notice, this list of conditions and the following disclaimer in
15 * the documentation and/or other materials provided with the
17 * * Neither the name of Intel Corporation nor the names of its
18 * contributors may be used to endorse or promote products derived
19 * from this software without specific prior written permission.
21 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
22 * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
23 * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
24 * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
25 * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
26 * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
27 * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
28 * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
29 * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
30 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
31 * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
35 #ifndef _RTE_ETHDEV_H_
36 #define _RTE_ETHDEV_H_
41 * RTE Ethernet Device API
43 * The Ethernet Device API is composed of two parts:
45 * - The application-oriented Ethernet API that includes functions to setup
46 * an Ethernet device (configure it, setup its RX and TX queues and start it),
47 * to get its MAC address, the speed and the status of its physical link,
48 * to receive and to transmit packets, and so on.
50 * - The driver-oriented Ethernet API that exports a function allowing
51 * an Ethernet Poll Mode Driver (PMD) to simultaneously register itself as
52 * an Ethernet device driver and as a PCI driver for a set of matching PCI
53 * [Ethernet] devices classes.
55 * By default, all the functions of the Ethernet Device API exported by a PMD
56 * are lock-free functions which assume to not be invoked in parallel on
57 * different logical cores to work on the same target object. For instance,
58 * the receive function of a PMD cannot be invoked in parallel on two logical
59 * cores to poll the same RX queue [of the same port]. Of course, this function
60 * can be invoked in parallel by different logical cores on different RX queues.
61 * It is the responsibility of the upper level application to enforce this rule.
63 * If needed, parallel accesses by multiple logical cores to shared queues
64 * shall be explicitly protected by dedicated inline lock-aware functions
65 * built on top of their corresponding lock-free functions of the PMD API.
67 * In all functions of the Ethernet API, the Ethernet device is
68 * designated by an integer >= 0 named the device port identifier.
70 * At the Ethernet driver level, Ethernet devices are represented by a generic
71 * data structure of type *rte_eth_dev*.
73 * Ethernet devices are dynamically registered during the PCI probing phase
74 * performed at EAL initialization time.
75 * When an Ethernet device is being probed, an *rte_eth_dev* structure and
76 * a new port identifier are allocated for that device. Then, the eth_dev_init()
77 * function supplied by the Ethernet driver matching the probed PCI
78 * device is invoked to properly initialize the device.
80 * The role of the device init function consists of resetting the hardware,
81 * checking access to Non-volatile Memory (NVM), reading the MAC address
84 * If the device init operation is successful, the correspondence between
85 * the port identifier assigned to the new device and its associated
86 * *rte_eth_dev* structure is effectively registered.
87 * Otherwise, both the *rte_eth_dev* structure and the port identifier are
90 * The functions exported by the application Ethernet API to setup a device
91 * designated by its port identifier must be invoked in the following order:
92 * - rte_eth_dev_configure()
93 * - rte_eth_tx_queue_setup()
94 * - rte_eth_rx_queue_setup()
95 * - rte_eth_dev_start()
97 * Then, the network application can invoke, in any order, the functions
98 * exported by the Ethernet API to get the MAC address of a given device, to
99 * get the speed and the status of a device physical link, to receive/transmit
100 * [burst of] packets, and so on.
102 * If the application wants to change the configuration (i.e. call
103 * rte_eth_dev_configure(), rte_eth_tx_queue_setup(), or
104 * rte_eth_rx_queue_setup()), it must call rte_eth_dev_stop() first to stop the
105 * device and then do the reconfiguration before calling rte_eth_dev_start()
106 * again. The tramsit and receive functions should not be invoked when the
109 * Please note that some configuration is not stored between calls to
110 * rte_eth_dev_stop()/rte_eth_dev_start(). The following configuration will
113 * - flow control settings
114 * - receive mode configuration (promiscuous mode, hardware checksum mode,
115 * RSS/VMDQ settings etc.)
116 * - VLAN filtering configuration
117 * - MAC addresses supplied to MAC address array
118 * - flow director filtering mode (but not filtering rules)
119 * - NIC queue statistics mappings
121 * Any other configuration will not be stored and will need to be re-entered
122 * after a call to rte_eth_dev_start().
124 * Finally, a network application can close an Ethernet device by invoking the
125 * rte_eth_dev_close() function.
127 * Each function of the application Ethernet API invokes a specific function
128 * of the PMD that controls the target device designated by its port
130 * For this purpose, all device-specific functions of an Ethernet driver are
131 * supplied through a set of pointers contained in a generic structure of type
133 * The address of the *eth_dev_ops* structure is stored in the *rte_eth_dev*
134 * structure by the device init function of the Ethernet driver, which is
135 * invoked during the PCI probing phase, as explained earlier.
137 * In other words, each function of the Ethernet API simply retrieves the
138 * *rte_eth_dev* structure associated with the device port identifier and
139 * performs an indirect invocation of the corresponding driver function
140 * supplied in the *eth_dev_ops* structure of the *rte_eth_dev* structure.
142 * For performance reasons, the address of the burst-oriented RX and TX
143 * functions of the Ethernet driver are not contained in the *eth_dev_ops*
144 * structure. Instead, they are directly stored at the beginning of the
145 * *rte_eth_dev* structure to avoid an extra indirect memory access during
148 * RTE ethernet device drivers do not use interrupts for transmitting or
149 * receiving. Instead, Ethernet drivers export Poll-Mode receive and transmit
150 * functions to applications.
151 * Both receive and transmit functions are packet-burst oriented to minimize
152 * their cost per packet through the following optimizations:
154 * - Sharing among multiple packets the incompressible cost of the
155 * invocation of receive/transmit functions.
157 * - Enabling receive/transmit functions to take advantage of burst-oriented
158 * hardware features (L1 cache, prefetch instructions, NIC head/tail
159 * registers) to minimize the number of CPU cycles per packet, for instance,
160 * by avoiding useless read memory accesses to ring descriptors, or by
161 * systematically using arrays of pointers that exactly fit L1 cache line
162 * boundaries and sizes.
164 * The burst-oriented receive function does not provide any error notification,
165 * to avoid the corresponding overhead. As a hint, the upper-level application
166 * might check the status of the device link once being systematically returned
167 * a 0 value by the receive function of the driver for a given number of tries.
177 #include <rte_interrupts.h>
179 #include <rte_mbuf.h>
180 #include "rte_ether.h"
183 * A structure used to retrieve statistics for an Ethernet port.
185 struct rte_eth_stats {
186 uint64_t ipackets; /**< Total number of successfully received packets. */
187 uint64_t opackets; /**< Total number of successfully transmitted packets.*/
188 uint64_t ibytes; /**< Total number of successfully received bytes. */
189 uint64_t obytes; /**< Total number of successfully transmitted bytes. */
190 uint64_t ierrors; /**< Total number of erroneous received packets. */
191 uint64_t oerrors; /**< Total number of failed transmitted packets. */
192 uint64_t imcasts; /**< Total number of multicast received packets. */
193 uint64_t rx_nombuf; /**< Total number of RX mbuf allocation failures. */
194 uint64_t fdirmatch; /**< Total number of RX packets matching a filter. */
195 uint64_t fdirmiss; /**< Total number of RX packets not matching any filter. */
196 uint64_t q_ipackets[RTE_ETHDEV_QUEUE_STAT_CNTRS];
197 /**< Total number of queue RX packets. */
198 uint64_t q_opackets[RTE_ETHDEV_QUEUE_STAT_CNTRS];
199 /**< Total number of queue TX packets. */
200 uint64_t q_ibytes[RTE_ETHDEV_QUEUE_STAT_CNTRS];
201 /**< Total number of successfully received queue bytes. */
202 uint64_t q_obytes[RTE_ETHDEV_QUEUE_STAT_CNTRS];
203 /**< Total number of successfully transmitted queue bytes. */
204 uint64_t q_errors[RTE_ETHDEV_QUEUE_STAT_CNTRS];
205 /**< Total number of queue packets received that are dropped. */
209 * A structure used to retrieve link-level information of an Ethernet port.
211 struct rte_eth_link {
212 uint16_t link_speed; /**< ETH_LINK_SPEED_[10, 100, 1000, 10000] */
213 uint16_t link_duplex; /**< ETH_LINK_[HALF_DUPLEX, FULL_DUPLEX] */
214 uint8_t link_status : 1; /**< 1 -> link up, 0 -> link down */
215 }__attribute__((aligned(8))); /**< aligned for atomic64 read/write */
217 #define ETH_LINK_SPEED_AUTONEG 0 /**< Auto-negotiate link speed. */
218 #define ETH_LINK_SPEED_10 10 /**< 10 megabits/second. */
219 #define ETH_LINK_SPEED_100 100 /**< 100 megabits/second. */
220 #define ETH_LINK_SPEED_1000 1000 /**< 1 gigabits/second. */
221 #define ETH_LINK_SPEED_10000 10000 /**< 10 gigabits/second. */
223 #define ETH_LINK_AUTONEG_DUPLEX 0 /**< Auto-negotiate duplex. */
224 #define ETH_LINK_HALF_DUPLEX 1 /**< Half-duplex connection. */
225 #define ETH_LINK_FULL_DUPLEX 2 /**< Full-duplex connection. */
228 * A structure used to configure the ring threshold registers of an RX/TX
229 * queue for an Ethernet port.
231 struct rte_eth_thresh {
232 uint8_t pthresh; /**< Ring prefetch threshold. */
233 uint8_t hthresh; /**< Ring host threshold. */
234 uint8_t wthresh; /**< Ring writeback threshold. */
238 * A set of values to identify what method is to be used to route
239 * packets to multiple queues.
241 enum rte_eth_rx_mq_mode {
242 ETH_RSS = 0, /**< Default to RSS mode */
243 ETH_VMDQ_DCB, /**< Use VMDQ+DCB to route traffic to queues */
244 ETH_DCB_RX /**< For RX side,only DCB is on. */
248 * A set of values to identify what method is to be used to transmit
249 * packets using multi-TCs.
251 enum rte_eth_tx_mq_mode {
252 ETH_DCB_NONE = 0, /**< It is not in DCB mode. */
253 ETH_VMDQ_DCB_TX, /**< For TX side,both DCB and VT is on. */
254 ETH_DCB_TX /**< For TX side,only DCB is on. */
258 * A structure used to configure the RX features of an Ethernet port.
260 struct rte_eth_rxmode {
261 /** The multi-queue packet distribution mode to be used, e.g. RSS. */
262 enum rte_eth_rx_mq_mode mq_mode;
263 uint32_t max_rx_pkt_len; /**< Only used if jumbo_frame enabled. */
264 uint16_t split_hdr_size; /**< hdr buf size (header_split enabled).*/
265 uint8_t header_split : 1, /**< Header Split enable. */
266 hw_ip_checksum : 1, /**< IP/UDP/TCP checksum offload enable. */
267 hw_vlan_filter : 1, /**< VLAN filter enable. */
268 jumbo_frame : 1, /**< Jumbo Frame Receipt enable. */
269 hw_strip_crc : 1; /**< Enable CRC stripping by hardware. */
273 * A structure used to configure the Receive Side Scaling (RSS) feature
274 * of an Ethernet port.
275 * If not NULL, the *rss_key* pointer of the *rss_conf* structure points
276 * to an array of 40 bytes holding the RSS key to use for hashing specific
277 * header fields of received packets.
278 * Otherwise, a default random hash key is used by the device driver.
280 * The *rss_hf* field of the *rss_conf* structure indicates the different
281 * types of IPv4/IPv6 packets to which the RSS hashing must be applied.
282 * Supplying an *rss_hf* equal to zero disables the RSS feature.
284 struct rte_eth_rss_conf {
285 uint8_t *rss_key; /**< If not NULL, 40-byte hash key. */
286 uint16_t rss_hf; /**< Hash functions to apply - see below. */
289 #define ETH_RSS_IPV4 0x0001 /**< IPv4 packet. */
290 #define ETH_RSS_IPV4_TCP 0x0002 /**< IPv4/TCP packet. */
291 #define ETH_RSS_IPV6 0x0004 /**< IPv6 packet. */
292 #define ETH_RSS_IPV6_EX 0x0008 /**< IPv6 packet with extension headers.*/
293 #define ETH_RSS_IPV6_TCP 0x0010 /**< IPv6/TCP packet. */
294 #define ETH_RSS_IPV6_TCP_EX 0x0020 /**< IPv6/TCP with extension headers. */
295 /* Intel RSS extensions to UDP packets */
296 #define ETH_RSS_IPV4_UDP 0x0040 /**< IPv4/UDP packet. */
297 #define ETH_RSS_IPV6_UDP 0x0080 /**< IPv6/UDP packet. */
298 #define ETH_RSS_IPV6_UDP_EX 0x0100 /**< IPv6/UDP with extension headers. */
300 /* Definitions used for VMDQ and DCB functionality */
301 #define ETH_VMDQ_MAX_VLAN_FILTERS 64 /**< Maximum nb. of VMDQ vlan filters. */
302 #define ETH_DCB_NUM_USER_PRIORITIES 8 /**< Maximum nb. of DCB priorities. */
303 #define ETH_VMDQ_DCB_NUM_QUEUES 128 /**< Maximum nb. of VMDQ DCB queues. */
304 #define ETH_DCB_NUM_QUEUES 128 /**< Maximum nb. of DCB queues. */
306 /* DCB capability defines */
307 #define ETH_DCB_PG_SUPPORT 0x00000001 /**< Priority Group(ETS) support. */
308 #define ETH_DCB_PFC_SUPPORT 0x00000002 /**< Priority Flow Control support. */
311 * This enum indicates the possible number of traffic classes
312 * in DCB configratioins
314 enum rte_eth_nb_tcs {
315 ETH_4_TCS = 4, /**< 4 TCs with DCB. */
316 ETH_8_TCS = 8 /**< 8 TCs with DCB. */
320 * This enum indicates the possible number of queue pools
321 * in VMDQ+DCB configurations.
323 enum rte_eth_nb_pools {
324 ETH_16_POOLS = 16, /**< 16 pools with DCB. */
325 ETH_32_POOLS = 32 /**< 32 pools with DCB. */
328 /* This structure may be extended in future. */
329 struct rte_eth_dcb_rx_conf {
330 enum rte_eth_nb_tcs nb_tcs; /**< Possible DCB TCs, 4 or 8 TCs */
331 uint8_t dcb_queue[ETH_DCB_NUM_USER_PRIORITIES];
332 /**< Possible DCB queue,4 or 8. */
335 struct rte_eth_vmdq_dcb_tx_conf {
336 enum rte_eth_nb_pools nb_queue_pools; /**< With DCB, 16 or 32 pools. */
337 uint8_t dcb_queue[ETH_DCB_NUM_USER_PRIORITIES];
338 /**< Possible DCB queue,4 or 8. */
341 struct rte_eth_dcb_tx_conf {
342 enum rte_eth_nb_tcs nb_tcs; /**< possible DCB TCs, 4 or 8 TCs. */
343 uint8_t dcb_queue[ETH_DCB_NUM_USER_PRIORITIES];
344 /**< Possible DCB queue,4 or 8. */
348 * A structure used to configure the VMDQ+DCB feature
349 * of an Ethernet port.
351 * Using this feature, packets are routed to a pool of queues, based
352 * on the vlan id in the vlan tag, and then to a specific queue within
353 * that pool, using the user priority vlan tag field.
355 * A default pool may be used, if desired, to route all traffic which
356 * does not match the vlan filter rules.
358 struct rte_eth_vmdq_dcb_conf {
359 enum rte_eth_nb_pools nb_queue_pools; /**< With DCB, 16 or 32 pools */
360 uint8_t enable_default_pool; /**< If non-zero, use a default pool */
361 uint8_t default_pool; /**< The default pool, if applicable */
362 uint8_t nb_pool_maps; /**< We can have up to 64 filters/mappings */
364 uint16_t vlan_id; /**< The vlan id of the received frame */
365 uint64_t pools; /**< Bitmask of pools for packet rx */
366 } pool_map[ETH_VMDQ_MAX_VLAN_FILTERS]; /**< VMDq vlan pool maps. */
367 uint8_t dcb_queue[ETH_DCB_NUM_USER_PRIORITIES];
368 /**< Selects a queue in a pool */
372 * A structure used to configure the TX features of an Ethernet port.
374 struct rte_eth_txmode {
375 enum rte_eth_tx_mq_mode mq_mode; /**< TX multi-queues mode. */
379 * A structure used to configure an RX ring of an Ethernet port.
381 struct rte_eth_rxconf {
382 struct rte_eth_thresh rx_thresh; /**< RX ring threshold registers. */
383 uint16_t rx_free_thresh; /**< Drives the freeing of RX descriptors. */
384 uint8_t rx_drop_en; /**< Drop packets if no descriptors are available. */
387 #define ETH_TXQ_FLAGS_NOMULTSEGS 0x0001 /**< nb_segs=1 for all mbufs */
388 #define ETH_TXQ_FLAGS_NOREFCOUNT 0x0002 /**< refcnt can be ignored */
389 #define ETH_TXQ_FLAGS_NOMULTMEMP 0x0004 /**< all bufs come from same mempool */
390 #define ETH_TXQ_FLAGS_NOVLANOFFL 0x0100 /**< disable VLAN offload */
391 #define ETH_TXQ_FLAGS_NOXSUMSCTP 0x0200 /**< disable SCTP checksum offload */
392 #define ETH_TXQ_FLAGS_NOXSUMUDP 0x0400 /**< disable UDP checksum offload */
393 #define ETH_TXQ_FLAGS_NOXSUMTCP 0x0800 /**< disable TCP checksum offload */
394 #define ETH_TXQ_FLAGS_NOOFFLOADS \
395 (ETH_TXQ_FLAGS_NOVLANOFFL | ETH_TXQ_FLAGS_NOXSUMSCTP | \
396 ETH_TXQ_FLAGS_NOXSUMUDP | ETH_TXQ_FLAGS_NOXSUMTCP)
398 * A structure used to configure a TX ring of an Ethernet port.
400 struct rte_eth_txconf {
401 struct rte_eth_thresh tx_thresh; /**< TX ring threshold registers. */
402 uint16_t tx_rs_thresh; /**< Drives the setting of RS bit on TXDs. */
403 uint16_t tx_free_thresh; /**< Drives the freeing of TX buffers. */
404 uint32_t txq_flags; /**< Set flags for the Tx queue */
408 * This enum indicates the flow control mode
410 enum rte_eth_fc_mode {
411 RTE_FC_NONE = 0, /**< Disable flow control. */
412 RTE_FC_RX_PAUSE, /**< RX pause frame, enable flowctrl on TX side. */
413 RTE_FC_TX_PAUSE, /**< TX pause frame, enable flowctrl on RX side. */
414 RTE_FC_FULL /**< Enable flow control on both side. */
418 * A structure used to configure Ethernet flow control parameter.
419 * These parameters will be configured into the register of the NIC.
420 * Please refer to the corresponding data sheet for proper value.
422 struct rte_eth_fc_conf {
423 uint32_t high_water; /**< High threshold value to trigger XOFF */
424 uint32_t low_water; /**< Low threshold value to trigger XON */
425 uint16_t pause_time; /**< Pause quota in the Pause frame */
426 uint16_t send_xon; /**< Is XON frame need be sent */
427 enum rte_eth_fc_mode mode; /**< Link flow control mode */
431 * Flow Director setting modes: none (default), signature or perfect.
434 RTE_FDIR_MODE_NONE = 0, /**< Disable FDIR support. */
435 RTE_FDIR_MODE_SIGNATURE, /**< Enable FDIR signature filter mode. */
436 RTE_FDIR_MODE_PERFECT, /**< Enable FDIR perfect filter mode. */
440 * Memory space that can be configured to store Flow Director filters
441 * in the board memory.
443 enum rte_fdir_pballoc_type {
444 RTE_FDIR_PBALLOC_64K = 0, /**< 64k. */
445 RTE_FDIR_PBALLOC_128K, /**< 128k. */
446 RTE_FDIR_PBALLOC_256K, /**< 256k. */
450 * Select report mode of FDIR hash information in RX descriptors.
452 enum rte_fdir_status_mode {
453 RTE_FDIR_NO_REPORT_STATUS = 0, /**< Never report FDIR hash. */
454 RTE_FDIR_REPORT_STATUS, /**< Only report FDIR hash for matching pkts. */
455 RTE_FDIR_REPORT_STATUS_ALWAYS, /**< Always report FDIR hash. */
459 * A structure used to configure the Flow Director (FDIR) feature
460 * of an Ethernet port.
462 * If mode is RTE_FDIR_DISABLE, the pballoc value is ignored.
464 struct rte_fdir_conf {
465 enum rte_fdir_mode mode; /**< Flow Director mode. */
466 enum rte_fdir_pballoc_type pballoc; /**< Space for FDIR filters. */
467 enum rte_fdir_status_mode status; /**< How to report FDIR hash. */
468 /* Offset of flexbytes field in RX packets (in 16-bit word units). */
469 uint8_t flexbytes_offset;
470 /* RX queue of packets matching a "drop" filter in perfect mode. */
475 * Possible l4type of FDIR filters.
478 RTE_FDIR_L4TYPE_NONE = 0, /**< None. */
479 RTE_FDIR_L4TYPE_UDP, /**< UDP. */
480 RTE_FDIR_L4TYPE_TCP, /**< TCP. */
481 RTE_FDIR_L4TYPE_SCTP, /**< SCTP. */
485 * Select IPv4 or IPv6 FDIR filters.
488 RTE_FDIR_IPTYPE_IPV4 = 0, /**< IPv4. */
489 RTE_FDIR_IPTYPE_IPV6 , /**< IPv6. */
493 * A structure used to define a FDIR packet filter.
495 struct rte_fdir_filter {
496 uint16_t flex_bytes; /**< Flex bytes value to match. */
497 uint16_t vlan_id; /**< VLAN ID value to match, 0 otherwise. */
498 uint16_t port_src; /**< Source port to match, 0 otherwise. */
499 uint16_t port_dst; /**< Destination port to match, 0 otherwise. */
501 uint32_t ipv4_addr; /**< IPv4 source address to match. */
502 uint32_t ipv6_addr[4]; /**< IPv6 source address to match. */
503 } ip_src; /**< IPv4/IPv6 source address to match (union of above). */
505 uint32_t ipv4_addr; /**< IPv4 destination address to match. */
506 uint32_t ipv6_addr[4]; /**< IPv6 destination address to match */
507 } ip_dst; /**< IPv4/IPv6 destination address to match (union of above). */
508 enum rte_l4type l4type; /**< l4type to match: NONE/UDP/TCP/SCTP. */
509 enum rte_iptype iptype; /**< IP packet type to match: IPv4 or IPv6. */
513 * A structure used to configure FDIR masks that are used by the device
514 * to match the various fields of RX packet headers.
515 * @note The only_ip_flow field has the opposite meaning compared to other
518 struct rte_fdir_masks {
519 /** When set to 1, packet l4type is \b NOT relevant in filters, and
520 source and destination port masks must be set to zero. */
521 uint8_t only_ip_flow;
522 /** If set to 1, vlan_id is relevant in filters. */
524 /** If set to 1, vlan_prio is relevant in filters. */
526 /** If set to 1, flexbytes is relevant in filters. */
528 /** Mask of Destination IPv4 Address. All bits set to 1 define the
529 relevant bits to use in the destination address of an IPv4 packet
530 when matching it against FDIR filters. */
531 uint32_t dst_ipv4_mask;
532 /** Mask of Source IPv4 Address. All bits set to 1 define
533 the relevant bits to use in the source address of an IPv4 packet
534 when matching it against FDIR filters. */
535 uint32_t src_ipv4_mask;
536 /** Mask of Source IPv6 Address. All bits set to 1 define the
537 relevant BYTES to use in the source address of an IPv6 packet
538 when matching it against FDIR filters. */
539 uint16_t src_ipv6_mask;
540 /** Mask of Source Port. All bits set to 1 define the relevant
541 bits to use in the source port of an IP packets when matching it
542 against FDIR filters. */
543 uint16_t src_port_mask;
544 /** Mask of Destination Port. All bits set to 1 define the relevant
545 bits to use in the destination port of an IP packet when matching it
546 against FDIR filters. */
547 uint16_t dst_port_mask;
551 * A structure used to report the status of the flow director filters in use.
553 struct rte_eth_fdir {
554 /** Number of filters with collision indication. */
556 /** Number of free (non programmed) filters. */
558 /** The Lookup hash value of the added filter that updated the value
559 of the MAXLEN field */
561 /** Longest linked list of filters in the table. */
563 /** Number of added filters. */
565 /** Number of removed filters. */
567 /** Number of failed added filters (no more space in device). */
569 /** Number of failed removed filters. */
574 * A structure used to enable/disable specific device interrupts.
576 struct rte_intr_conf {
577 /** enable/disable lsc interrupt. 0 (default) - disable, 1 enable */
582 * A structure used to configure an Ethernet port.
583 * Depending upon the RX multi-queue mode, extra advanced
584 * configuration settings may be needed.
586 struct rte_eth_conf {
588 /**< ETH_LINK_SPEED_10[0|00|000], or 0 for autonegotation */
589 uint16_t link_duplex;
590 /**< ETH_LINK_[HALF_DUPLEX|FULL_DUPLEX], or 0 for autonegotation */
591 struct rte_eth_rxmode rxmode; /**< Port RX configuration. */
592 struct rte_eth_txmode txmode; /**< Port TX configuration. */
594 struct rte_eth_rss_conf rss_conf; /**< Port RSS configuration */
595 struct rte_eth_vmdq_dcb_conf vmdq_dcb_conf;
596 /**< Port vmdq+dcb configuration. */
597 struct rte_eth_dcb_rx_conf dcb_rx_conf;
598 /**< Port dcb RX configuration. */
599 } rx_adv_conf; /**< Port RX filtering configuration (union). */
601 struct rte_eth_vmdq_dcb_tx_conf vmdq_dcb_tx_conf;
602 /**< Port vmdq+dcb TX configuration. */
603 struct rte_eth_dcb_tx_conf dcb_tx_conf;
604 /**< Port dcb TX configuration. */
605 } tx_adv_conf; /**< Port TX DCB configuration (union). */
606 /** Currently,Priority Flow Control(PFC) are supported,if DCB with PFC
607 is needed,and the variable must be set ETH_DCB_PFC_SUPPORT. */
608 uint32_t dcb_capability_en;
609 struct rte_fdir_conf fdir_conf; /**< FDIR configuration. */
610 struct rte_intr_conf intr_conf; /**< Interrupt mode configuration. */
614 * A structure used to retrieve the contextual information of
615 * an Ethernet device, such as the controlling driver of the device,
616 * its PCI context, etc...
618 struct rte_eth_dev_info {
619 struct rte_pci_device *pci_dev; /**< Device PCI information. */
620 const char *driver_name; /**< Device Driver name. */
621 uint32_t min_rx_bufsize; /**< Minimum size of RX buffer. */
622 uint32_t max_rx_pktlen; /**< Maximum configurable length of RX pkt. */
623 uint16_t max_rx_queues; /**< Maximum number of RX queues. */
624 uint16_t max_tx_queues; /**< Maximum number of TX queues. */
625 uint32_t max_mac_addrs; /**< Maximum number of MAC addresses. */
630 struct rte_eth_dev_callback;
631 /** @internal Structure to keep track of registered callbacks */
632 TAILQ_HEAD(rte_eth_dev_cb_list, rte_eth_dev_callback);
635 * Definitions of all functions exported by an Ethernet driver through the
636 * the generic structure of type *eth_dev_ops* supplied in the *rte_eth_dev*
637 * structure associated with an Ethernet device.
640 typedef int (*eth_dev_configure_t)(struct rte_eth_dev *dev);
641 /**< @internal Ethernet device configuration. */
643 typedef int (*eth_dev_start_t)(struct rte_eth_dev *dev);
644 /**< @internal Function used to start a configured Ethernet device. */
646 typedef void (*eth_dev_stop_t)(struct rte_eth_dev *dev);
647 /**< @internal Function used to stop a configured Ethernet device. */
649 typedef void (*eth_dev_close_t)(struct rte_eth_dev *dev);
650 /**< @internal Function used to close a configured Ethernet device. */
652 typedef void (*eth_promiscuous_enable_t)(struct rte_eth_dev *dev);
653 /**< @internal Function used to enable the RX promiscuous mode of an Ethernet device. */
655 typedef void (*eth_promiscuous_disable_t)(struct rte_eth_dev *dev);
656 /**< @internal Function used to disable the RX promiscuous mode of an Ethernet device. */
658 typedef void (*eth_allmulticast_enable_t)(struct rte_eth_dev *dev);
659 /**< @internal Enable the receipt of all multicast packets by an Ethernet device. */
661 typedef void (*eth_allmulticast_disable_t)(struct rte_eth_dev *dev);
662 /**< @internal Disable the receipt of all multicast packets by an Ethernet device. */
664 typedef int (*eth_link_update_t)(struct rte_eth_dev *dev,
665 int wait_to_complete);
666 /**< @internal Get link speed, duplex mode and state (up/down) of an Ethernet device. */
668 typedef void (*eth_stats_get_t)(struct rte_eth_dev *dev,
669 struct rte_eth_stats *igb_stats);
670 /**< @internal Get global I/O statistics of an Ethernet device. */
672 typedef void (*eth_stats_reset_t)(struct rte_eth_dev *dev);
673 /**< @internal Reset global I/O statistics of an Ethernet device to 0. */
675 typedef int (*eth_queue_stats_mapping_set_t)(struct rte_eth_dev *dev,
679 /**< @internal Set a queue statistics mapping for a tx/rx queue of an Ethernet device. */
681 typedef void (*eth_dev_infos_get_t)(struct rte_eth_dev *dev,
682 struct rte_eth_dev_info *dev_info);
683 /**< @internal Get specific informations of an Ethernet device. */
685 typedef int (*eth_rx_queue_setup_t)(struct rte_eth_dev *dev,
686 uint16_t rx_queue_id,
688 unsigned int socket_id,
689 const struct rte_eth_rxconf *rx_conf,
690 struct rte_mempool *mb_pool);
691 /**< @internal Set up a receive queue of an Ethernet device. */
693 typedef int (*eth_tx_queue_setup_t)(struct rte_eth_dev *dev,
694 uint16_t tx_queue_id,
696 unsigned int socket_id,
697 const struct rte_eth_txconf *tx_conf);
698 /**< @internal Setup a transmit queue of an Ethernet device. */
700 typedef void (*eth_queue_release_t)(void *queue);
701 /**< @internal Release memory resources allocated by given RX/TX queue. */
703 typedef void (*vlan_filter_set_t)(struct rte_eth_dev *dev,
706 /**< @internal filtering of a VLAN Tag Identifier by an Ethernet device. */
708 typedef uint16_t (*eth_rx_burst_t)(void *rxq,
709 struct rte_mbuf **rx_pkts,
711 /**< @internal Retrieve input packets from a receive queue of an Ethernet device. */
713 typedef uint16_t (*eth_tx_burst_t)(void *txq,
714 struct rte_mbuf **tx_pkts,
716 /**< @internal Send output packets on a transmit queue of an Ethernet device. */
718 typedef int (*fdir_add_signature_filter_t)(struct rte_eth_dev *dev,
719 struct rte_fdir_filter *fdir_ftr,
721 /**< @internal Setup a new signature filter rule on an Ethernet device */
723 typedef int (*fdir_update_signature_filter_t)(struct rte_eth_dev *dev,
724 struct rte_fdir_filter *fdir_ftr,
726 /**< @internal Update a signature filter rule on an Ethernet device */
728 typedef int (*fdir_remove_signature_filter_t)(struct rte_eth_dev *dev,
729 struct rte_fdir_filter *fdir_ftr);
730 /**< @internal Remove a signature filter rule on an Ethernet device */
732 typedef void (*fdir_infos_get_t)(struct rte_eth_dev *dev,
733 struct rte_eth_fdir *fdir);
734 /**< @internal Get information about fdir status */
736 typedef int (*fdir_add_perfect_filter_t)(struct rte_eth_dev *dev,
737 struct rte_fdir_filter *fdir_ftr,
738 uint16_t soft_id, uint8_t rx_queue,
740 /**< @internal Setup a new perfect filter rule on an Ethernet device */
742 typedef int (*fdir_update_perfect_filter_t)(struct rte_eth_dev *dev,
743 struct rte_fdir_filter *fdir_ftr,
744 uint16_t soft_id, uint8_t rx_queue,
746 /**< @internal Update a perfect filter rule on an Ethernet device */
748 typedef int (*fdir_remove_perfect_filter_t)(struct rte_eth_dev *dev,
749 struct rte_fdir_filter *fdir_ftr,
751 /**< @internal Remove a perfect filter rule on an Ethernet device */
753 typedef int (*fdir_set_masks_t)(struct rte_eth_dev *dev,
754 struct rte_fdir_masks *fdir_masks);
755 /**< @internal Setup flow director masks on an Ethernet device */
757 typedef int (*flow_ctrl_set_t)(struct rte_eth_dev *dev,
758 struct rte_eth_fc_conf *fc_conf);
759 /**< @internal Setup flow control parameter on an Ethernet device */
761 typedef int (*eth_dev_led_on_t)(struct rte_eth_dev *dev);
762 /**< @internal Turn on SW controllable LED on an Ethernet device */
764 typedef int (*eth_dev_led_off_t)(struct rte_eth_dev *dev);
765 /**< @internal Turn off SW controllable LED on an Ethernet device */
767 typedef void (*eth_mac_addr_remove_t)(struct rte_eth_dev *dev, uint32_t index);
768 /**< @internal Remove MAC address from receive address register */
770 typedef void (*eth_mac_addr_add_t)(struct rte_eth_dev *dev,
771 struct ether_addr *mac_addr,
774 /**< @internal Set a MAC address into Receive Address Address Register */
777 * @internal A structure containing the functions exported by an Ethernet driver.
780 eth_dev_configure_t dev_configure; /**< Configure device. */
781 eth_dev_start_t dev_start; /**< Start device. */
782 eth_dev_stop_t dev_stop; /**< Stop device. */
783 eth_dev_close_t dev_close; /**< Close device. */
784 eth_promiscuous_enable_t promiscuous_enable; /**< Promiscuous ON. */
785 eth_promiscuous_disable_t promiscuous_disable;/**< Promiscuous OFF. */
786 eth_allmulticast_enable_t allmulticast_enable;/**< RX multicast ON. */
787 eth_allmulticast_disable_t allmulticast_disable;/**< RX multicast OF. */
788 eth_link_update_t link_update; /**< Get device link state. */
789 eth_stats_get_t stats_get; /**< Get device statistics. */
790 eth_stats_reset_t stats_reset; /**< Reset device statistics. */
791 eth_queue_stats_mapping_set_t queue_stats_mapping_set;
792 /**< Configure per queue stat counter mapping. */
793 eth_dev_infos_get_t dev_infos_get; /**< Get device info. */
794 vlan_filter_set_t vlan_filter_set; /**< Filter VLAN Setup. */
795 eth_rx_queue_setup_t rx_queue_setup;/**< Set up device RX queue.*/
796 eth_queue_release_t rx_queue_release;/**< Release RX queue.*/
797 eth_tx_queue_setup_t tx_queue_setup;/**< Set up device TX queue.*/
798 eth_queue_release_t tx_queue_release;/**< Release TX queue.*/
799 eth_dev_led_on_t dev_led_on; /**< Turn on LED. */
800 eth_dev_led_off_t dev_led_off; /**< Turn off LED. */
801 flow_ctrl_set_t flow_ctrl_set; /**< Setup flow control. */
802 eth_mac_addr_remove_t mac_addr_remove; /**< Remove MAC address */
803 eth_mac_addr_add_t mac_addr_add; /**< Add a MAC address */
805 /** Add a signature filter. */
806 fdir_add_signature_filter_t fdir_add_signature_filter;
807 /** Update a signature filter. */
808 fdir_update_signature_filter_t fdir_update_signature_filter;
809 /** Remove a signature filter. */
810 fdir_remove_signature_filter_t fdir_remove_signature_filter;
811 /** Get information about FDIR status. */
812 fdir_infos_get_t fdir_infos_get;
813 /** Add a perfect filter. */
814 fdir_add_perfect_filter_t fdir_add_perfect_filter;
815 /** Update a perfect filter. */
816 fdir_update_perfect_filter_t fdir_update_perfect_filter;
817 /** Remove a perfect filter. */
818 fdir_remove_perfect_filter_t fdir_remove_perfect_filter;
819 /** Setup masks for FDIR filtering. */
820 fdir_set_masks_t fdir_set_masks;
825 * The generic data structure associated with each ethernet device.
827 * Pointers to burst-oriented packet receive and transmit functions are
828 * located at the beginning of the structure, along with the pointer to
829 * where all the data elements for the particular device are stored in shared
830 * memory. This split allows the function pointer and driver data to be per-
831 * process, while the actual configuration data for the device is shared.
834 eth_rx_burst_t rx_pkt_burst; /**< Pointer to PMD receive function. */
835 eth_tx_burst_t tx_pkt_burst; /**< Pointer to PMD transmit function. */
836 struct rte_eth_dev_data *data; /**< Pointer to device data */
837 const struct eth_driver *driver;/**< Driver for this device */
838 struct eth_dev_ops *dev_ops; /**< Functions exported by PMD */
839 struct rte_pci_device *pci_dev; /**< PCI info. supplied by probing */
840 struct rte_eth_dev_cb_list callbacks; /**< User application callbacks */
845 * The data part, with no function pointers, associated with each ethernet device.
847 * This structure is safe to place in shared memory to be common among different
848 * processes in a multi-process configuration.
850 struct rte_eth_dev_data {
851 void **rx_queues; /**< Array of pointers to RX queues. */
852 void **tx_queues; /**< Array of pointers to TX queues. */
853 uint16_t nb_rx_queues; /**< Number of RX queues. */
854 uint16_t nb_tx_queues; /**< Number of TX queues. */
856 void *dev_private; /**< PMD-specific private data */
858 struct rte_eth_link dev_link;
859 /**< Link-level information & status */
861 struct rte_eth_conf dev_conf; /**< Configuration applied to device. */
862 uint16_t max_frame_size; /**< Default is ETHER_MAX_LEN (1518). */
864 uint64_t rx_mbuf_alloc_failed; /**< RX ring mbuf allocation failures. */
865 struct ether_addr* mac_addrs;/**< Device Ethernet Link address. */
866 uint8_t port_id; /**< Device [external] port identifier. */
867 uint8_t promiscuous : 1, /**< RX promiscuous mode ON(1) / OFF(0). */
868 scattered_rx : 1, /**< RX of scattered packets is ON(1) / OFF(0) */
869 all_multicast : 1, /**< RX all multicast mode ON(1) / OFF(0). */
870 dev_started : 1; /**< Device state: STARTED(1) / STOPPED(0). */
875 * The pool of *rte_eth_dev* structures. The size of the pool
876 * is configured at compile-time in the <rte_ethdev.c> file.
878 extern struct rte_eth_dev rte_eth_devices[];
881 * Get the total number of Ethernet devices that have been successfully
882 * initialized by the [matching] Ethernet driver during the PCI probing phase.
883 * All devices whose port identifier is in the range
884 * [0, rte_eth_dev_count() - 1] can be operated on by network applications.
887 * - The total number of usable Ethernet devices.
889 extern uint8_t rte_eth_dev_count(void);
894 * Initialization function of an Ethernet driver invoked for each matching
895 * Ethernet PCI device detected during the PCI probing phase.
898 * The pointer to the [matching] Ethernet driver structure supplied by
899 * the PMD when it registered itself.
901 * The *eth_dev* pointer is the address of the *rte_eth_dev* structure
902 * associated with the matching device and which have been [automatically]
903 * allocated in the *rte_eth_devices* array.
904 * The *eth_dev* structure is supplied to the driver initialization function
905 * with the following fields already initialized:
907 * - *pci_dev*: Holds the pointers to the *rte_pci_device* structure which
908 * contains the generic PCI information of the matching device.
910 * - *dev_private*: Holds a pointer to the device private data structure.
912 * - *max_frame_size*: Contains the default Ethernet maximum frame length
915 * - *port_id*: Contains the port index of the device (actually the index
916 * of the *eth_dev* structure in the *rte_eth_devices* array).
919 * - 0: Success, the device is properly initialized by the driver.
920 * In particular, the driver MUST have set up the *dev_ops* pointer
921 * of the *eth_dev* structure.
922 * - <0: Error code of the device initialization failure.
924 typedef int (*eth_dev_init_t)(struct eth_driver *eth_drv,
925 struct rte_eth_dev *eth_dev);
929 * The structure associated with a PMD Ethernet driver.
931 * Each Ethernet driver acts as a PCI driver and is represented by a generic
932 * *eth_driver* structure that holds:
934 * - An *rte_pci_driver* structure (which must be the first field).
936 * - The *eth_dev_init* function invoked for each matching PCI device.
938 * - The size of the private data to allocate for each matching device.
941 struct rte_pci_driver pci_drv; /**< The PMD is also a PCI driver. */
942 eth_dev_init_t eth_dev_init; /**< Device init function. */
943 unsigned int dev_private_size; /**< Size of device private data. */
948 * A function invoked by the initialization function of an Ethernet driver
949 * to simultaneously register itself as a PCI driver and as an Ethernet
950 * Poll Mode Driver (PMD).
953 * The pointer to the *eth_driver* structure associated with
954 * the Ethernet driver.
956 extern void rte_eth_driver_register(struct eth_driver *eth_drv);
959 * The initialization function of the driver for
960 * Intel(r) IGB Gigabit Ethernet Controller devices.
961 * This function is invoked once at EAL start time.
965 extern int rte_igb_pmd_init(void);
968 * The initialization function of the driver for
969 * Intel(r) EM Gigabit Ethernet Controller devices.
970 * This function is invoked once at EAL start time.
974 extern int rte_em_pmd_init(void);
977 * The initialization function of the driver for 1Gbps Intel IGB_VF
979 * Invoked once at EAL start time.
983 extern int rte_igbvf_pmd_init(void);
986 * The initialization function of the driver for 10Gbps Intel IXGBE
988 * Invoked once at EAL start time.
992 extern int rte_ixgbe_pmd_init(void);
995 * The initialization function of the driver for 10Gbps Intel IXGBE_VF
997 * Invoked once at EAL start time.
1001 extern int rte_ixgbevf_pmd_init(void);
1004 * The initialization function of *all* supported and enabled drivers.
1005 * Right now, the following PMDs are supported:
1011 * This function is invoked once at EAL start time.
1014 * Error code of the device initialization failure,
1015 * -ENODEV if there are no drivers available
1016 * (e.g. if all driver config options are = n).
1019 int rte_pmd_init_all(void)
1023 #ifdef RTE_LIBRTE_IGB_PMD
1024 if ((ret = rte_igb_pmd_init()) != 0) {
1025 RTE_LOG(ERR, PMD, "Cannot init igb PMD\n");
1028 if ((ret = rte_igbvf_pmd_init()) != 0) {
1029 RTE_LOG(ERR, PMD, "Cannot init igbvf PMD\n");
1032 #endif /* RTE_LIBRTE_IGB_PMD */
1034 #ifdef RTE_LIBRTE_EM_PMD
1035 if ((ret = rte_em_pmd_init()) != 0) {
1036 RTE_LOG(ERR, PMD, "Cannot init em PMD\n");
1039 #endif /* RTE_LIBRTE_EM_PMD */
1041 #ifdef RTE_LIBRTE_IXGBE_PMD
1042 if ((ret = rte_ixgbe_pmd_init()) != 0) {
1043 RTE_LOG(ERR, PMD, "Cannot init ixgbe PMD\n");
1046 if ((ret = rte_ixgbevf_pmd_init()) != 0) {
1047 RTE_LOG(ERR, PMD, "Cannot init ixgbevf PMD\n");
1050 #endif /* RTE_LIBRTE_IXGBE_PMD */
1053 RTE_LOG(ERR, PMD, "No PMD(s) are configured\n");
1058 * Configure an Ethernet device.
1059 * This function must be invoked first before any other function in the
1060 * Ethernet API. This function can also be re-invoked when a device is in the
1064 * The port identifier of the Ethernet device to configure.
1065 * @param nb_rx_queue
1066 * The number of receive queues to set up for the Ethernet device.
1067 * @param nb_tx_queue
1068 * The number of transmit queues to set up for the Ethernet device.
1070 * The pointer to the configuration data to be used for the Ethernet device.
1071 * The *rte_eth_conf* structure includes:
1072 * - the hardware offload features to activate, with dedicated fields for
1073 * each statically configurable offload hardware feature provided by
1074 * Ethernet devices, such as IP checksum or VLAN tag stripping for
1076 * - the Receive Side Scaling (RSS) configuration when using multiple RX
1079 * Embedding all configuration information in a single data structure
1080 * is the more flexible method that allows the addition of new features
1081 * without changing the syntax of the API.
1083 * - 0: Success, device configured.
1084 * - <0: Error code returned by the driver configuration function.
1086 extern int rte_eth_dev_configure(uint8_t port_id,
1087 uint16_t nb_rx_queue,
1088 uint16_t nb_tx_queue,
1089 const struct rte_eth_conf *eth_conf);
1092 * Allocate and set up a receive queue for an Ethernet device.
1094 * The function allocates a contiguous block of memory for *nb_rx_desc*
1095 * receive descriptors from a memory zone associated with *socket_id*
1096 * and initializes each receive descriptor with a network buffer allocated
1097 * from the memory pool *mb_pool*.
1100 * The port identifier of the Ethernet device.
1101 * @param rx_queue_id
1102 * The index of the receive queue to set up.
1103 * The value must be in the range [0, nb_rx_queue - 1] previously supplied
1104 * to rte_eth_dev_configure().
1106 * The number of receive descriptors to allocate for the receive ring.
1108 * The *socket_id* argument is the socket identifier in case of NUMA.
1109 * The value can be *SOCKET_ID_ANY* if there is no NUMA constraint for
1110 * the DMA memory allocated for the receive descriptors of the ring.
1112 * The pointer to the configuration data to be used for the receive queue.
1113 * The *rx_conf* structure contains an *rx_thresh* structure with the values
1114 * of the Prefetch, Host, and Write-Back threshold registers of the receive
1117 * The pointer to the memory pool from which to allocate *rte_mbuf* network
1118 * memory buffers to populate each descriptor of the receive ring.
1120 * - 0: Success, receive queue correctly set up.
1121 * - -EINVAL: The size of network buffers which can be allocated from the
1122 * memory pool does not fit the various buffer sizes allowed by the
1123 * device controller.
1124 * - -ENOMEM: Unable to allocate the receive ring descriptors or to
1125 * allocate network memory buffers from the memory pool when
1126 * initializing receive descriptors.
1128 extern int rte_eth_rx_queue_setup(uint8_t port_id, uint16_t rx_queue_id,
1129 uint16_t nb_rx_desc, unsigned int socket_id,
1130 const struct rte_eth_rxconf *rx_conf,
1131 struct rte_mempool *mb_pool);
1134 * Allocate and set up a transmit queue for an Ethernet device.
1137 * The port identifier of the Ethernet device.
1138 * @param tx_queue_id
1139 * The index of the transmit queue to set up.
1140 * The value must be in the range [0, nb_tx_queue - 1] previously supplied
1141 * to rte_eth_dev_configure().
1143 * The number of transmit descriptors to allocate for the transmit ring.
1145 * The *socket_id* argument is the socket identifier in case of NUMA.
1146 * Its value can be *SOCKET_ID_ANY* if there is no NUMA constraint for
1147 * the DMA memory allocated for the transmit descriptors of the ring.
1149 * The pointer to the configuration data to be used for the transmit queue.
1150 * The *tx_conf* structure contains the following data:
1151 * - The *tx_thresh* structure with the values of the Prefetch, Host, and
1152 * Write-Back threshold registers of the transmit ring.
1153 * When setting Write-Back threshold to the value greater then zero,
1154 * *tx_rs_thresh* value should be explicitly set to one.
1155 * - The *tx_free_thresh* value indicates the [minimum] number of network
1156 * buffers that must be pending in the transmit ring to trigger their
1157 * [implicit] freeing by the driver transmit function.
1158 * - The *tx_rs_thresh* value indicates the [minimum] number of transmit
1159 * descriptors that must be pending in the transmit ring before setting the
1160 * RS bit on a descriptor by the driver transmit function.
1161 * The *tx_rs_thresh* value should be less or equal then
1162 * *tx_free_thresh* value, and both of them should be less then
1164 * - The *txq_flags* member contains flags to pass to the TX queue setup
1165 * function to configure the behavior of the TX queue. This should be set
1166 * to 0 if no special configuration is required.
1168 * Note that setting *tx_free_thresh* or *tx_rs_thresh* value to 0 forces
1169 * the transmit function to use default values.
1171 * - 0: Success, the transmit queue is correctly set up.
1172 * - -ENOMEM: Unable to allocate the transmit ring descriptors.
1174 extern int rte_eth_tx_queue_setup(uint8_t port_id, uint16_t tx_queue_id,
1175 uint16_t nb_tx_desc, unsigned int socket_id,
1176 const struct rte_eth_txconf *tx_conf);
1179 * Start an Ethernet device.
1181 * The device start step is the last one and consists of setting the configured
1182 * offload features and in starting the transmit and the receive units of the
1184 * On success, all basic functions exported by the Ethernet API (link status,
1185 * receive/transmit, and so on) can be invoked.
1188 * The port identifier of the Ethernet device.
1190 * - 0: Success, Ethernet device started.
1191 * - <0: Error code of the driver device start function.
1193 extern int rte_eth_dev_start(uint8_t port_id);
1196 * Stop an Ethernet device. The device can be restarted with a call to
1197 * rte_eth_dev_start()
1200 * The port identifier of the Ethernet device.
1202 extern void rte_eth_dev_stop(uint8_t port_id);
1205 * Close an Ethernet device. The device cannot be restarted!
1208 * The port identifier of the Ethernet device.
1210 extern void rte_eth_dev_close(uint8_t port_id);
1213 * Enable receipt in promiscuous mode for an Ethernet device.
1216 * The port identifier of the Ethernet device.
1218 extern void rte_eth_promiscuous_enable(uint8_t port_id);
1221 * Disable receipt in promiscuous mode for an Ethernet device.
1224 * The port identifier of the Ethernet device.
1226 extern void rte_eth_promiscuous_disable(uint8_t port_id);
1229 * Return the value of promiscuous mode for an Ethernet device.
1232 * The port identifier of the Ethernet device.
1234 * - (1) if promiscuous is enabled
1235 * - (0) if promiscuous is disabled.
1238 extern int rte_eth_promiscuous_get(uint8_t port_id);
1241 * Enable the receipt of any multicast frame by an Ethernet device.
1244 * The port identifier of the Ethernet device.
1246 extern void rte_eth_allmulticast_enable(uint8_t port_id);
1249 * Disable the receipt of all multicast frames by an Ethernet device.
1252 * The port identifier of the Ethernet device.
1254 extern void rte_eth_allmulticast_disable(uint8_t port_id);
1257 * Return the value of allmulticast mode for an Ethernet device.
1260 * The port identifier of the Ethernet device.
1262 * - (1) if allmulticast is enabled
1263 * - (0) if allmulticast is disabled.
1266 extern int rte_eth_allmulticast_get(uint8_t port_id);
1269 * Retrieve the status (ON/OFF), the speed (in Mbps) and the mode (HALF-DUPLEX
1270 * or FULL-DUPLEX) of the physical link of an Ethernet device. It might need
1271 * to wait up to 9 seconds in it.
1274 * The port identifier of the Ethernet device.
1276 * A pointer to an *rte_eth_link* structure to be filled with
1277 * the status, the speed and the mode of the Ethernet device link.
1279 extern void rte_eth_link_get(uint8_t port_id, struct rte_eth_link *link);
1282 * Retrieve the status (ON/OFF), the speed (in Mbps) and the mode (HALF-DUPLEX
1283 * or FULL-DUPLEX) of the physical link of an Ethernet device. It is a no-wait
1284 * version of rte_eth_link_get().
1287 * The port identifier of the Ethernet device.
1289 * A pointer to an *rte_eth_link* structure to be filled with
1290 * the status, the speed and the mode of the Ethernet device link.
1292 extern void rte_eth_link_get_nowait(uint8_t port_id,
1293 struct rte_eth_link *link);
1296 * Retrieve the general I/O statistics of an Ethernet device.
1299 * The port identifier of the Ethernet device.
1301 * A pointer to a structure of type *rte_eth_stats* to be filled with
1302 * the values of device counters for the following set of statistics:
1303 * - *ipackets* with the total of successfully received packets.
1304 * - *opackets* with the total of successfully transmitted packets.
1305 * - *ibytes* with the total of successfully received bytes.
1306 * - *obytes* with the total of successfully transmitted bytes.
1307 * - *ierrors* with the total of erroneous received packets.
1308 * - *oerrors* with the total of failed transmitted packets.
1310 extern void rte_eth_stats_get(uint8_t port_id, struct rte_eth_stats *stats);
1313 * Reset the general I/O statistics of an Ethernet device.
1316 * The port identifier of the Ethernet device.
1318 extern void rte_eth_stats_reset(uint8_t port_id);
1321 * Set a mapping for the specified transmit queue to the specified per-queue
1322 * statistics counter.
1325 * The port identifier of the Ethernet device.
1326 * @param tx_queue_id
1327 * The index of the transmit queue for which a queue stats mapping is required.
1328 * The value must be in the range [0, nb_tx_queue - 1] previously supplied
1329 * to rte_eth_dev_configure().
1331 * The per-queue packet statistics functionality number that the transmit
1332 * queue is to be assigned.
1333 * The value must be in the range [0, RTE_MAX_ETHPORT_QUEUE_STATS_MAPS - 1].
1335 * Zero if successful. Non-zero otherwise.
1337 extern int rte_eth_dev_set_tx_queue_stats_mapping(uint8_t port_id,
1338 uint16_t tx_queue_id,
1342 * Set a mapping for the specified receive queue to the specified per-queue
1343 * statistics counter.
1346 * The port identifier of the Ethernet device.
1347 * @param rx_queue_id
1348 * The index of the receive queue for which a queue stats mapping is required.
1349 * The value must be in the range [0, nb_rx_queue - 1] previously supplied
1350 * to rte_eth_dev_configure().
1352 * The per-queue packet statistics functionality number that the receive
1353 * queue is to be assigned.
1354 * The value must be in the range [0, RTE_MAX_ETHPORT_QUEUE_STATS_MAPS - 1].
1356 * Zero if successful. Non-zero otherwise.
1358 extern int rte_eth_dev_set_rx_queue_stats_mapping(uint8_t port_id,
1359 uint16_t rx_queue_id,
1363 * Retrieve the Ethernet address of an Ethernet device.
1366 * The port identifier of the Ethernet device.
1368 * A pointer to a structure of type *ether_addr* to be filled with
1369 * the Ethernet address of the Ethernet device.
1371 extern void rte_eth_macaddr_get(uint8_t port_id, struct ether_addr *mac_addr);
1374 * Retrieve the contextual information of an Ethernet device.
1377 * The port identifier of the Ethernet device.
1379 * A pointer to a structure of type *rte_eth_dev_info* to be filled with
1380 * the contextual information of the Ethernet device.
1382 extern void rte_eth_dev_info_get(uint8_t port_id,
1383 struct rte_eth_dev_info *dev_info);
1386 * Enable/Disable hardware filtering by an Ethernet device of received
1387 * VLAN packets tagged with a given VLAN Tag Identifier.
1390 * The port identifier of the Ethernet device.
1392 * The VLAN Tag Identifier whose filtering must be enabled or disabled.
1394 * If > 0, enable VLAN filtering of VLAN packets tagged with *vlan_id*.
1395 * Otherwise, disable VLAN filtering of VLAN packets tagged with *vlan_id*.
1397 * - (0) if successful.
1398 * - (-ENOSUP) if hardware-assisted VLAN filtering not configured.
1399 * - (-ENODEV) if *port_id* invalid.
1400 * - (-ENOSYS) if VLAN filtering on *port_id* disabled.
1401 * - (-EINVAL) if *vlan_id* > 4095.
1403 extern int rte_eth_dev_vlan_filter(uint8_t port_id, uint16_t vlan_id, int on);
1407 * Retrieve a burst of input packets from a receive queue of an Ethernet
1408 * device. The retrieved packets are stored in *rte_mbuf* structures whose
1409 * pointers are supplied in the *rx_pkts* array.
1411 * The rte_eth_rx_burst() function loops, parsing the RX ring of the
1412 * receive queue, up to *nb_pkts* packets, and for each completed RX
1413 * descriptor in the ring, it performs the following operations:
1415 * - Initialize the *rte_mbuf* data structure associated with the
1416 * RX descriptor according to the information provided by the NIC into
1417 * that RX descriptor.
1419 * - Store the *rte_mbuf* data structure into the next entry of the
1422 * - Replenish the RX descriptor with a new *rte_mbuf* buffer
1423 * allocated from the memory pool associated with the receive queue at
1424 * initialization time.
1426 * When retrieving an input packet that was scattered by the controller
1427 * into multiple receive descriptors, the rte_eth_rx_burst() function
1428 * appends the associated *rte_mbuf* buffers to the first buffer of the
1431 * The rte_eth_rx_burst() function returns the number of packets
1432 * actually retrieved, which is the number of *rte_mbuf* data structures
1433 * effectively supplied into the *rx_pkts* array.
1434 * A return value equal to *nb_pkts* indicates that the RX queue contained
1435 * at least *rx_pkts* packets, and this is likely to signify that other
1436 * received packets remain in the input queue. Applications implementing
1437 * a "retrieve as much received packets as possible" policy can check this
1438 * specific case and keep invoking the rte_eth_rx_burst() function until
1439 * a value less than *nb_pkts* is returned.
1441 * This receive method has the following advantages:
1443 * - It allows a run-to-completion network stack engine to retrieve and
1444 * to immediately process received packets in a fast burst-oriented
1445 * approach, avoiding the overhead of unnecessary intermediate packet
1446 * queue/dequeue operations.
1448 * - Conversely, it also allows an asynchronous-oriented processing
1449 * method to retrieve bursts of received packets and to immediately
1450 * queue them for further parallel processing by another logical core,
1451 * for instance. However, instead of having received packets being
1452 * individually queued by the driver, this approach allows the invoker
1453 * of the rte_eth_rx_burst() function to queue a burst of retrieved
1454 * packets at a time and therefore dramatically reduce the cost of
1455 * enqueue/dequeue operations per packet.
1457 * - It allows the rte_eth_rx_burst() function of the driver to take
1458 * advantage of burst-oriented hardware features (CPU cache,
1459 * prefetch instructions, and so on) to minimize the number of CPU
1460 * cycles per packet.
1462 * To summarize, the proposed receive API enables many
1463 * burst-oriented optimizations in both synchronous and asynchronous
1464 * packet processing environments with no overhead in both cases.
1466 * The rte_eth_rx_burst() function does not provide any error
1467 * notification to avoid the corresponding overhead. As a hint, the
1468 * upper-level application might check the status of the device link once
1469 * being systematically returned a 0 value for a given number of tries.
1472 * The port identifier of the Ethernet device.
1474 * The index of the receive queue from which to retrieve input packets.
1475 * The value must be in the range [0, nb_rx_queue - 1] previously supplied
1476 * to rte_eth_dev_configure().
1478 * The address of an array of pointers to *rte_mbuf* structures that
1479 * must be large enough to store *nb_pkts* pointers in it.
1481 * The maximum number of packets to retrieve.
1483 * The number of packets actually retrieved, which is the number
1484 * of pointers to *rte_mbuf* structures effectively supplied to the
1487 #ifdef RTE_LIBRTE_ETHDEV_DEBUG
1488 extern uint16_t rte_eth_rx_burst(uint8_t port_id, uint16_t queue_id,
1489 struct rte_mbuf **rx_pkts, uint16_t nb_pkts);
1491 static inline uint16_t
1492 rte_eth_rx_burst(uint8_t port_id, uint16_t queue_id,
1493 struct rte_mbuf **rx_pkts, uint16_t nb_pkts)
1495 struct rte_eth_dev *dev;
1497 dev = &rte_eth_devices[port_id];
1498 return (*dev->rx_pkt_burst)(dev->data->rx_queues[queue_id], rx_pkts, nb_pkts);
1503 * Send a burst of output packets on a transmit queue of an Ethernet device.
1505 * The rte_eth_tx_burst() function is invoked to transmit output packets
1506 * on the output queue *queue_id* of the Ethernet device designated by its
1508 * The *nb_pkts* parameter is the number of packets to send which are
1509 * supplied in the *tx_pkts* array of *rte_mbuf* structures.
1510 * The rte_eth_tx_burst() function loops, sending *nb_pkts* packets,
1511 * up to the number of transmit descriptors available in the TX ring of the
1513 * For each packet to send, the rte_eth_tx_burst() function performs
1514 * the following operations:
1516 * - Pick up the next available descriptor in the transmit ring.
1518 * - Free the network buffer previously sent with that descriptor, if any.
1520 * - Initialize the transmit descriptor with the information provided
1521 * in the *rte_mbuf data structure.
1523 * In the case of a segmented packet composed of a list of *rte_mbuf* buffers,
1524 * the rte_eth_tx_burst() function uses several transmit descriptors
1527 * The rte_eth_tx_burst() function returns the number of packets it
1528 * actually sent. A return value equal to *nb_pkts* means that all packets
1529 * have been sent, and this is likely to signify that other output packets
1530 * could be immediately transmitted again. Applications that implement a
1531 * "send as many packets to transmit as possible" policy can check this
1532 * specific case and keep invoking the rte_eth_tx_burst() function until
1533 * a value less than *nb_pkts* is returned.
1535 * It is the responsibility of the rte_eth_tx_burst() function to
1536 * transparently free the memory buffers of packets previously sent.
1537 * This feature is driven by the *tx_free_thresh* value supplied to the
1538 * rte_eth_dev_configure() function at device configuration time.
1539 * When the number of previously sent packets reached the "minimum transmit
1540 * packets to free" threshold, the rte_eth_tx_burst() function must
1541 * [attempt to] free the *rte_mbuf* buffers of those packets whose
1542 * transmission was effectively completed.
1545 * The port identifier of the Ethernet device.
1547 * The index of the transmit queue through which output packets must be
1549 * The value must be in the range [0, nb_tx_queue - 1] previously supplied
1550 * to rte_eth_dev_configure().
1552 * The address of an array of *nb_pkts* pointers to *rte_mbuf* structures
1553 * which contain the output packets.
1555 * The maximum number of packets to transmit.
1557 * The number of output packets actually stored in transmit descriptors of
1558 * the transmit ring. The return value can be less than the value of the
1559 * *tx_pkts* parameter when the transmit ring is full or has been filled up.
1561 #ifdef RTE_LIBRTE_ETHDEV_DEBUG
1562 extern uint16_t rte_eth_tx_burst(uint8_t port_id, uint16_t queue_id,
1563 struct rte_mbuf **tx_pkts, uint16_t nb_pkts);
1565 static inline uint16_t
1566 rte_eth_tx_burst(uint8_t port_id, uint16_t queue_id,
1567 struct rte_mbuf **tx_pkts, uint16_t nb_pkts)
1569 struct rte_eth_dev *dev;
1571 dev = &rte_eth_devices[port_id];
1572 return (*dev->tx_pkt_burst)(dev->data->tx_queues[queue_id], tx_pkts, nb_pkts);
1577 * Setup a new signature filter rule on an Ethernet device
1580 * The port identifier of the Ethernet device.
1581 * @param fdir_filter
1582 * The pointer to the fdir filter structure describing the signature filter
1584 * The *rte_fdir_filter* structure includes the values of the different fields
1585 * to match: source and destination IP addresses, vlan id, flexbytes, source
1586 * and destination ports, and so on.
1588 * The index of the RX queue where to store RX packets matching the added
1589 * signature filter defined in fdir_filter.
1591 * - (0) if successful.
1592 * - (-ENOTSUP) if hardware doesn't support flow director mode.
1593 * - (-ENODEV) if *port_id* invalid.
1594 * - (-ENOSYS) if the FDIR mode is not configured in signature mode
1596 * - (-EINVAL) if the fdir_filter information is not correct.
1598 int rte_eth_dev_fdir_add_signature_filter(uint8_t port_id,
1599 struct rte_fdir_filter *fdir_filter,
1603 * Update a signature filter rule on an Ethernet device.
1604 * If the rule doesn't exits, it is created.
1607 * The port identifier of the Ethernet device.
1609 * The pointer to the structure describing the signature filter rule.
1610 * The *rte_fdir_filter* structure includes the values of the different fields
1611 * to match: source and destination IP addresses, vlan id, flexbytes, source
1612 * and destination ports, and so on.
1614 * The index of the RX queue where to store RX packets matching the added
1615 * signature filter defined in fdir_ftr.
1617 * - (0) if successful.
1618 * - (-ENOTSUP) if hardware doesn't support flow director mode.
1619 * - (-ENODEV) if *port_id* invalid.
1620 * - (-ENOSYS) if the flow director mode is not configured in signature mode
1622 * - (-EINVAL) if the fdir_filter information is not correct.
1624 int rte_eth_dev_fdir_update_signature_filter(uint8_t port_id,
1625 struct rte_fdir_filter *fdir_ftr,
1629 * Remove a signature filter rule on an Ethernet device.
1632 * The port identifier of the Ethernet device.
1634 * The pointer to the structure describing the signature filter rule.
1635 * The *rte_fdir_filter* structure includes the values of the different fields
1636 * to match: source and destination IP addresses, vlan id, flexbytes, source
1637 * and destination ports, and so on.
1639 * - (0) if successful.
1640 * - (-ENOTSUP) if hardware doesn't support flow director mode.
1641 * - (-ENODEV) if *port_id* invalid.
1642 * - (-ENOSYS) if the flow director mode is not configured in signature mode
1644 * - (-EINVAL) if the fdir_filter information is not correct.
1646 int rte_eth_dev_fdir_remove_signature_filter(uint8_t port_id,
1647 struct rte_fdir_filter *fdir_ftr);
1650 * Retrieve the flow director information of an Ethernet device.
1653 * The port identifier of the Ethernet device.
1655 * A pointer to a structure of type *rte_eth_dev_fdir* to be filled with
1656 * the flow director information of the Ethernet device.
1658 * - (0) if successful.
1659 * - (-ENOTSUP) if hardware doesn't support flow director mode.
1660 * - (-ENODEV) if *port_id* invalid.
1661 * - (-ENOSYS) if the flow director mode is not configured on *port_id*.
1663 int rte_eth_dev_fdir_get_infos(uint8_t port_id, struct rte_eth_fdir *fdir);
1666 * Add a new perfect filter rule on an Ethernet device.
1669 * The port identifier of the Ethernet device.
1670 * @param fdir_filter
1671 * The pointer to the structure describing the perfect filter rule.
1672 * The *rte_fdir_filter* structure includes the values of the different fields
1673 * to match: source and destination IP addresses, vlan id, flexbytes, source
1674 * and destination ports, and so on.
1675 * IPv6 are not supported.
1677 * The 16-bit value supplied in the field hash.fdir.id of mbuf for RX
1678 * packets matching the perfect filter.
1680 * The index of the RX queue where to store RX packets matching the added
1681 * perfect filter defined in fdir_filter.
1683 * If drop is set to 1, matching RX packets are stored into the RX drop
1684 * queue defined in the rte_fdir_conf.
1686 * - (0) if successful.
1687 * - (-ENOTSUP) if hardware doesn't support flow director mode.
1688 * - (-ENODEV) if *port_id* invalid.
1689 * - (-ENOSYS) if the flow director mode is not configured in perfect mode
1691 * - (-EINVAL) if the fdir_filter information is not correct.
1693 int rte_eth_dev_fdir_add_perfect_filter(uint8_t port_id,
1694 struct rte_fdir_filter *fdir_filter,
1695 uint16_t soft_id, uint8_t rx_queue,
1699 * Update a perfect filter rule on an Ethernet device.
1700 * If the rule doesn't exits, it is created.
1703 * The port identifier of the Ethernet device.
1704 * @param fdir_filter
1705 * The pointer to the structure describing the perfect filter rule.
1706 * The *rte_fdir_filter* structure includes the values of the different fields
1707 * to match: source and destination IP addresses, vlan id, flexbytes, source
1708 * and destination ports, and so on.
1709 * IPv6 are not supported.
1711 * The 16-bit value supplied in the field hash.fdir.id of mbuf for RX
1712 * packets matching the perfect filter.
1714 * The index of the RX queue where to store RX packets matching the added
1715 * perfect filter defined in fdir_filter.
1717 * If drop is set to 1, matching RX packets are stored into the RX drop
1718 * queue defined in the rte_fdir_conf.
1720 * - (0) if successful.
1721 * - (-ENOTSUP) if hardware doesn't support flow director mode.
1722 * - (-ENODEV) if *port_id* invalid.
1723 * - (-ENOSYS) if the flow director mode is not configured in perfect mode
1725 * - (-EINVAL) if the fdir_filter information is not correct.
1727 int rte_eth_dev_fdir_update_perfect_filter(uint8_t port_id,
1728 struct rte_fdir_filter *fdir_filter,
1729 uint16_t soft_id, uint8_t rx_queue,
1733 * Remove a perfect filter rule on an Ethernet device.
1736 * The port identifier of the Ethernet device.
1737 * @param fdir_filter
1738 * The pointer to the structure describing the perfect filter rule.
1739 * The *rte_fdir_filter* structure includes the values of the different fields
1740 * to match: source and destination IP addresses, vlan id, flexbytes, source
1741 * and destination ports, and so on.
1742 * IPv6 are not supported.
1744 * The soft_id value provided when adding/updating the removed filter.
1746 * - (0) if successful.
1747 * - (-ENOTSUP) if hardware doesn't support flow director mode.
1748 * - (-ENODEV) if *port_id* invalid.
1749 * - (-ENOSYS) if the flow director mode is not configured in perfect mode
1751 * - (-EINVAL) if the fdir_filter information is not correct.
1753 int rte_eth_dev_fdir_remove_perfect_filter(uint8_t port_id,
1754 struct rte_fdir_filter *fdir_filter,
1757 * Configure globally the masks for flow director mode for an Ethernet device.
1758 * For example, the device can match packets with only the first 24 bits of
1759 * the IPv4 source address.
1761 * The following fields can be masked: IPv4 addresses and L4 port numbers.
1762 * The following fields can be either enabled or disabled completely for the
1763 * matching functionality: VLAN ID tag; VLAN Priority + CFI bit; Flexible 2-byte
1765 * IPv6 masks are not supported.
1767 * All filters must comply with the masks previously configured.
1768 * For example, with a mask equal to 255.255.255.0 for the source IPv4 address,
1769 * all IPv4 filters must be created with a source IPv4 address that fits the
1772 * This function flushes all filters that have been previously added in
1776 * The port identifier of the Ethernet device.
1778 * The pointer to the fdir mask structure describing relevant headers fields
1779 * and relevant bits to use when matching packets addresses and ports.
1780 * IPv6 masks are not supported.
1782 * - (0) if successful.
1783 * - (-ENOTSUP) if hardware doesn't support flow director mode.
1784 * - (-ENODEV) if *port_id* invalid.
1785 * - (-ENOSYS) if the flow director mode is not configured in perfect
1786 * mode on *port_id*.
1787 * - (-EINVAL) if the fdir_filter information is not correct
1789 int rte_eth_dev_fdir_set_masks(uint8_t port_id,
1790 struct rte_fdir_masks *fdir_mask);
1793 * The eth device event type for interrupt, and maybe others in the future.
1795 enum rte_eth_event_type {
1796 RTE_ETH_EVENT_UNKNOWN, /**< unknown event type */
1797 RTE_ETH_EVENT_INTR_LSC, /**< lsc interrupt event */
1798 RTE_ETH_EVENT_MAX /**< max value of this enum */
1801 typedef void (*rte_eth_dev_cb_fn)(uint8_t port_id, \
1802 enum rte_eth_event_type event, void *cb_arg);
1803 /**< user application callback to be registered for interrupts */
1808 * Register a callback function for specific port id.
1815 * User supplied callback function to be called.
1817 * Pointer to the parameters for the registered callback.
1820 * - On success, zero.
1821 * - On failure, a negative value.
1823 int rte_eth_dev_callback_register(uint8_t port_id,
1824 enum rte_eth_event_type event,
1825 rte_eth_dev_cb_fn cb_fn, void *cb_arg);
1828 * Unregister a callback function for specific port id.
1835 * User supplied callback function to be called.
1837 * Pointer to the parameters for the registered callback. -1 means to
1838 * remove all for the same callback address and same event.
1841 * - On success, zero.
1842 * - On failure, a negative value.
1844 int rte_eth_dev_callback_unregister(uint8_t port_id,
1845 enum rte_eth_event_type event,
1846 rte_eth_dev_cb_fn cb_fn, void *cb_arg);
1849 * @internal Executes all the user application registered callbacks for
1850 * the specific device. It is for DPDK internal user only. User
1851 * application should not call it directly.
1854 * Pointer to struct rte_eth_dev.
1856 * Eth device interrupt event type.
1861 void _rte_eth_dev_callback_process(struct rte_eth_dev *dev,
1862 enum rte_eth_event_type event);
1865 * Turn on the LED on the Ethernet device.
1866 * This function turns on the LED on the Ethernet device.
1869 * The port identifier of the Ethernet device.
1871 * - (0) if successful.
1872 * - (-ENOTSUP) if underlying hardware OR driver doesn't support
1874 * - (-ENODEV) if *port_id* invalid.
1876 int rte_eth_led_on(uint8_t port_id);
1879 * Turn off the LED on the Ethernet device.
1880 * This function turns off the LED on the Ethernet device.
1883 * The port identifier of the Ethernet device.
1885 * - (0) if successful.
1886 * - (-ENOTSUP) if underlying hardware OR driver doesn't support
1888 * - (-ENODEV) if *port_id* invalid.
1890 int rte_eth_led_off(uint8_t port_id);
1893 * Configure the Ethernet link flow control for Ethernet device
1896 * The port identifier of the Ethernet device.
1898 * The pointer to the structure of the flow control parameters.
1900 * - (0) if successful.
1901 * - (-ENOTSUP) if hardware doesn't support flow control mode.
1902 * - (-ENODEV) if *port_id* invalid.
1903 * - (-EINVAL) if bad parameter
1904 * - (-EIO) if flow control setup failure
1906 int rte_eth_dev_flow_ctrl_set(uint8_t port_id,
1907 struct rte_eth_fc_conf *fc_conf);
1910 * Add a MAC address to an internal array of addresses used to enable whitelist
1911 * filtering to accept packets only if the destination MAC address matches.
1914 * The port identifier of the Ethernet device.
1916 * The MAC address to add.
1918 * VMDq pool index to associate address with (if VMDq is enabled). If VMDq is
1919 * not enabled, this should be set to 0.
1921 * - (0) if successfully added or *mac_addr" was already added.
1922 * - (-ENOTSUP) if hardware doesn't support this feature.
1923 * - (-ENODEV) if *port* is invalid.
1924 * - (-ENOSPC) if no more MAC addresses can be added.
1925 * - (-EINVAL) if MAC address is invalid.
1927 int rte_eth_dev_mac_addr_add(uint8_t port, struct ether_addr *mac_addr,
1931 * Remove a MAC address from the internal array of addresses.
1934 * The port identifier of the Ethernet device.
1936 * MAC address to remove.
1938 * - (0) if successful, or *mac_addr* didn't exist.
1939 * - (-ENOTSUP) if hardware doesn't support.
1940 * - (-ENODEV) if *port* invalid.
1941 * - (-EADDRINUSE) if attempting to remove the default MAC address
1943 int rte_eth_dev_mac_addr_remove(uint8_t port, struct ether_addr *mac_addr);
1950 #endif /* _RTE_ETHDEV_H_ */