4 * Copyright(c) 2010-2014 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.
34 #ifndef _RTE_ETHDEV_H_
35 #define _RTE_ETHDEV_H_
40 * RTE Ethernet Device API
42 * The Ethernet Device API is composed of two parts:
44 * - The application-oriented Ethernet API that includes functions to setup
45 * an Ethernet device (configure it, setup its RX and TX queues and start it),
46 * to get its MAC address, the speed and the status of its physical link,
47 * to receive and to transmit packets, and so on.
49 * - The driver-oriented Ethernet API that exports a function allowing
50 * an Ethernet Poll Mode Driver (PMD) to simultaneously register itself as
51 * an Ethernet device driver and as a PCI driver for a set of matching PCI
52 * [Ethernet] devices classes.
54 * By default, all the functions of the Ethernet Device API exported by a PMD
55 * are lock-free functions which assume to not be invoked in parallel on
56 * different logical cores to work on the same target object. For instance,
57 * the receive function of a PMD cannot be invoked in parallel on two logical
58 * cores to poll the same RX queue [of the same port]. Of course, this function
59 * can be invoked in parallel by different logical cores on different RX queues.
60 * It is the responsibility of the upper level application to enforce this rule.
62 * If needed, parallel accesses by multiple logical cores to shared queues
63 * shall be explicitly protected by dedicated inline lock-aware functions
64 * built on top of their corresponding lock-free functions of the PMD API.
66 * In all functions of the Ethernet API, the Ethernet device is
67 * designated by an integer >= 0 named the device port identifier.
69 * At the Ethernet driver level, Ethernet devices are represented by a generic
70 * data structure of type *rte_eth_dev*.
72 * Ethernet devices are dynamically registered during the PCI probing phase
73 * performed at EAL initialization time.
74 * When an Ethernet device is being probed, an *rte_eth_dev* structure and
75 * a new port identifier are allocated for that device. Then, the eth_dev_init()
76 * function supplied by the Ethernet driver matching the probed PCI
77 * device is invoked to properly initialize the device.
79 * The role of the device init function consists of resetting the hardware,
80 * checking access to Non-volatile Memory (NVM), reading the MAC address
83 * If the device init operation is successful, the correspondence between
84 * the port identifier assigned to the new device and its associated
85 * *rte_eth_dev* structure is effectively registered.
86 * Otherwise, both the *rte_eth_dev* structure and the port identifier are
89 * The functions exported by the application Ethernet API to setup a device
90 * designated by its port identifier must be invoked in the following order:
91 * - rte_eth_dev_configure()
92 * - rte_eth_tx_queue_setup()
93 * - rte_eth_rx_queue_setup()
94 * - rte_eth_dev_start()
96 * Then, the network application can invoke, in any order, the functions
97 * exported by the Ethernet API to get the MAC address of a given device, to
98 * get the speed and the status of a device physical link, to receive/transmit
99 * [burst of] packets, and so on.
101 * If the application wants to change the configuration (i.e. call
102 * rte_eth_dev_configure(), rte_eth_tx_queue_setup(), or
103 * rte_eth_rx_queue_setup()), it must call rte_eth_dev_stop() first to stop the
104 * device and then do the reconfiguration before calling rte_eth_dev_start()
105 * again. The tramsit and receive functions should not be invoked when the
108 * Please note that some configuration is not stored between calls to
109 * rte_eth_dev_stop()/rte_eth_dev_start(). The following configuration will
112 * - flow control settings
113 * - receive mode configuration (promiscuous mode, hardware checksum mode,
114 * RSS/VMDQ settings etc.)
115 * - VLAN filtering configuration
116 * - MAC addresses supplied to MAC address array
117 * - flow director filtering mode (but not filtering rules)
118 * - NIC queue statistics mappings
120 * Any other configuration will not be stored and will need to be re-entered
121 * after a call to rte_eth_dev_start().
123 * Finally, a network application can close an Ethernet device by invoking the
124 * rte_eth_dev_close() function.
126 * Each function of the application Ethernet API invokes a specific function
127 * of the PMD that controls the target device designated by its port
129 * For this purpose, all device-specific functions of an Ethernet driver are
130 * supplied through a set of pointers contained in a generic structure of type
132 * The address of the *eth_dev_ops* structure is stored in the *rte_eth_dev*
133 * structure by the device init function of the Ethernet driver, which is
134 * invoked during the PCI probing phase, as explained earlier.
136 * In other words, each function of the Ethernet API simply retrieves the
137 * *rte_eth_dev* structure associated with the device port identifier and
138 * performs an indirect invocation of the corresponding driver function
139 * supplied in the *eth_dev_ops* structure of the *rte_eth_dev* structure.
141 * For performance reasons, the address of the burst-oriented RX and TX
142 * functions of the Ethernet driver are not contained in the *eth_dev_ops*
143 * structure. Instead, they are directly stored at the beginning of the
144 * *rte_eth_dev* structure to avoid an extra indirect memory access during
147 * RTE ethernet device drivers do not use interrupts for transmitting or
148 * receiving. Instead, Ethernet drivers export Poll-Mode receive and transmit
149 * functions to applications.
150 * Both receive and transmit functions are packet-burst oriented to minimize
151 * their cost per packet through the following optimizations:
153 * - Sharing among multiple packets the incompressible cost of the
154 * invocation of receive/transmit functions.
156 * - Enabling receive/transmit functions to take advantage of burst-oriented
157 * hardware features (L1 cache, prefetch instructions, NIC head/tail
158 * registers) to minimize the number of CPU cycles per packet, for instance,
159 * by avoiding useless read memory accesses to ring descriptors, or by
160 * systematically using arrays of pointers that exactly fit L1 cache line
161 * boundaries and sizes.
163 * The burst-oriented receive function does not provide any error notification,
164 * to avoid the corresponding overhead. As a hint, the upper-level application
165 * might check the status of the device link once being systematically returned
166 * a 0 value by the receive function of the driver for a given number of tries.
175 /* Use this macro to check if LRO API is supported */
176 #define RTE_ETHDEV_HAS_LRO_SUPPORT
179 #include <rte_interrupts.h>
182 #include <rte_devargs.h>
183 #include <rte_mbuf.h>
184 #include "rte_ether.h"
185 #include "rte_eth_ctrl.h"
188 * A structure used to retrieve statistics for an Ethernet port.
190 struct rte_eth_stats {
191 uint64_t ipackets; /**< Total number of successfully received packets. */
192 uint64_t opackets; /**< Total number of successfully transmitted packets.*/
193 uint64_t ibytes; /**< Total number of successfully received bytes. */
194 uint64_t obytes; /**< Total number of successfully transmitted bytes. */
195 uint64_t imissed; /**< Total of RX missed packets (e.g full FIFO). */
196 uint64_t ibadcrc; /**< Total of RX packets with CRC error. */
197 uint64_t ibadlen; /**< Total of RX packets with bad length. */
198 uint64_t ierrors; /**< Total number of erroneous received packets. */
199 uint64_t oerrors; /**< Total number of failed transmitted packets. */
200 uint64_t imcasts; /**< Total number of multicast received packets. */
201 uint64_t rx_nombuf; /**< Total number of RX mbuf allocation failures. */
202 uint64_t fdirmatch; /**< Total number of RX packets matching a filter. */
203 uint64_t fdirmiss; /**< Total number of RX packets not matching any filter. */
204 uint64_t tx_pause_xon; /**< Total nb. of XON pause frame sent. */
205 uint64_t rx_pause_xon; /**< Total nb. of XON pause frame received. */
206 uint64_t tx_pause_xoff; /**< Total nb. of XOFF pause frame sent. */
207 uint64_t rx_pause_xoff; /**< Total nb. of XOFF pause frame received. */
208 uint64_t q_ipackets[RTE_ETHDEV_QUEUE_STAT_CNTRS];
209 /**< Total number of queue RX packets. */
210 uint64_t q_opackets[RTE_ETHDEV_QUEUE_STAT_CNTRS];
211 /**< Total number of queue TX packets. */
212 uint64_t q_ibytes[RTE_ETHDEV_QUEUE_STAT_CNTRS];
213 /**< Total number of successfully received queue bytes. */
214 uint64_t q_obytes[RTE_ETHDEV_QUEUE_STAT_CNTRS];
215 /**< Total number of successfully transmitted queue bytes. */
216 uint64_t q_errors[RTE_ETHDEV_QUEUE_STAT_CNTRS];
217 /**< Total number of queue packets received that are dropped. */
219 /**< Total number of good packets received from loopback,VF Only */
221 /**< Total number of good packets transmitted to loopback,VF Only */
223 /**< Total number of good bytes received from loopback,VF Only */
225 /**< Total number of good bytes transmitted to loopback,VF Only */
229 * A structure used to retrieve link-level information of an Ethernet port.
231 struct rte_eth_link {
232 uint16_t link_speed; /**< ETH_LINK_SPEED_[10, 100, 1000, 10000] */
233 uint16_t link_duplex; /**< ETH_LINK_[HALF_DUPLEX, FULL_DUPLEX] */
234 uint8_t link_status : 1; /**< 1 -> link up, 0 -> link down */
235 }__attribute__((aligned(8))); /**< aligned for atomic64 read/write */
237 #define ETH_LINK_SPEED_AUTONEG 0 /**< Auto-negotiate link speed. */
238 #define ETH_LINK_SPEED_10 10 /**< 10 megabits/second. */
239 #define ETH_LINK_SPEED_100 100 /**< 100 megabits/second. */
240 #define ETH_LINK_SPEED_1000 1000 /**< 1 gigabits/second. */
241 #define ETH_LINK_SPEED_10000 10000 /**< 10 gigabits/second. */
242 #define ETH_LINK_SPEED_10G 10000 /**< alias of 10 gigabits/second. */
243 #define ETH_LINK_SPEED_20G 20000 /**< 20 gigabits/second. */
244 #define ETH_LINK_SPEED_40G 40000 /**< 40 gigabits/second. */
246 #define ETH_LINK_AUTONEG_DUPLEX 0 /**< Auto-negotiate duplex. */
247 #define ETH_LINK_HALF_DUPLEX 1 /**< Half-duplex connection. */
248 #define ETH_LINK_FULL_DUPLEX 2 /**< Full-duplex connection. */
251 * A structure used to configure the ring threshold registers of an RX/TX
252 * queue for an Ethernet port.
254 struct rte_eth_thresh {
255 uint8_t pthresh; /**< Ring prefetch threshold. */
256 uint8_t hthresh; /**< Ring host threshold. */
257 uint8_t wthresh; /**< Ring writeback threshold. */
261 * Simple flags are used for rte_eth_conf.rxmode.mq_mode.
263 #define ETH_MQ_RX_RSS_FLAG 0x1
264 #define ETH_MQ_RX_DCB_FLAG 0x2
265 #define ETH_MQ_RX_VMDQ_FLAG 0x4
268 * A set of values to identify what method is to be used to route
269 * packets to multiple queues.
271 enum rte_eth_rx_mq_mode {
272 /** None of DCB,RSS or VMDQ mode */
275 /** For RX side, only RSS is on */
276 ETH_MQ_RX_RSS = ETH_MQ_RX_RSS_FLAG,
277 /** For RX side,only DCB is on. */
278 ETH_MQ_RX_DCB = ETH_MQ_RX_DCB_FLAG,
279 /** Both DCB and RSS enable */
280 ETH_MQ_RX_DCB_RSS = ETH_MQ_RX_RSS_FLAG | ETH_MQ_RX_DCB_FLAG,
282 /** Only VMDQ, no RSS nor DCB */
283 ETH_MQ_RX_VMDQ_ONLY = ETH_MQ_RX_VMDQ_FLAG,
284 /** RSS mode with VMDQ */
285 ETH_MQ_RX_VMDQ_RSS = ETH_MQ_RX_RSS_FLAG | ETH_MQ_RX_VMDQ_FLAG,
286 /** Use VMDQ+DCB to route traffic to queues */
287 ETH_MQ_RX_VMDQ_DCB = ETH_MQ_RX_VMDQ_FLAG | ETH_MQ_RX_DCB_FLAG,
288 /** Enable both VMDQ and DCB in VMDq */
289 ETH_MQ_RX_VMDQ_DCB_RSS = ETH_MQ_RX_RSS_FLAG | ETH_MQ_RX_DCB_FLAG |
294 * for rx mq mode backward compatible
296 #define ETH_RSS ETH_MQ_RX_RSS
297 #define VMDQ_DCB ETH_MQ_RX_VMDQ_DCB
298 #define ETH_DCB_RX ETH_MQ_RX_DCB
301 * A set of values to identify what method is to be used to transmit
302 * packets using multi-TCs.
304 enum rte_eth_tx_mq_mode {
305 ETH_MQ_TX_NONE = 0, /**< It is in neither DCB nor VT mode. */
306 ETH_MQ_TX_DCB, /**< For TX side,only DCB is on. */
307 ETH_MQ_TX_VMDQ_DCB, /**< For TX side,both DCB and VT is on. */
308 ETH_MQ_TX_VMDQ_ONLY, /**< Only VT on, no DCB */
312 * for tx mq mode backward compatible
314 #define ETH_DCB_NONE ETH_MQ_TX_NONE
315 #define ETH_VMDQ_DCB_TX ETH_MQ_TX_VMDQ_DCB
316 #define ETH_DCB_TX ETH_MQ_TX_DCB
319 * A structure used to configure the RX features of an Ethernet port.
321 struct rte_eth_rxmode {
322 /** The multi-queue packet distribution mode to be used, e.g. RSS. */
323 enum rte_eth_rx_mq_mode mq_mode;
324 uint32_t max_rx_pkt_len; /**< Only used if jumbo_frame enabled. */
325 uint16_t split_hdr_size; /**< hdr buf size (header_split enabled).*/
326 uint16_t header_split : 1, /**< Header Split enable. */
327 hw_ip_checksum : 1, /**< IP/UDP/TCP checksum offload enable. */
328 hw_vlan_filter : 1, /**< VLAN filter enable. */
329 hw_vlan_strip : 1, /**< VLAN strip enable. */
330 hw_vlan_extend : 1, /**< Extended VLAN enable. */
331 jumbo_frame : 1, /**< Jumbo Frame Receipt enable. */
332 hw_strip_crc : 1, /**< Enable CRC stripping by hardware. */
333 enable_scatter : 1, /**< Enable scatter packets rx handler */
334 enable_lro : 1; /**< Enable LRO */
338 * A structure used to configure the Receive Side Scaling (RSS) feature
339 * of an Ethernet port.
340 * If not NULL, the *rss_key* pointer of the *rss_conf* structure points
341 * to an array holding the RSS key to use for hashing specific header
342 * fields of received packets. The length of this array should be indicated
343 * by *rss_key_len* below. Otherwise, a default random hash key is used by
346 * The *rss_key_len* field of the *rss_conf* structure indicates the length
347 * in bytes of the array pointed by *rss_key*. To be compatible, this length
348 * will be checked in i40e only. Others assume 40 bytes to be used as before.
350 * The *rss_hf* field of the *rss_conf* structure indicates the different
351 * types of IPv4/IPv6 packets to which the RSS hashing must be applied.
352 * Supplying an *rss_hf* equal to zero disables the RSS feature.
354 struct rte_eth_rss_conf {
355 uint8_t *rss_key; /**< If not NULL, 40-byte hash key. */
356 uint8_t rss_key_len; /**< hash key length in bytes. */
357 uint64_t rss_hf; /**< Hash functions to apply - see below. */
361 * The RSS offload types are defined based on flow types which are defined
362 * in rte_eth_ctrl.h. Different NIC hardwares may support different RSS offload
363 * types. The supported flow types or RSS offload types can be queried by
364 * rte_eth_dev_info_get().
366 #define ETH_RSS_IPV4 (1ULL << RTE_ETH_FLOW_IPV4)
367 #define ETH_RSS_FRAG_IPV4 (1ULL << RTE_ETH_FLOW_FRAG_IPV4)
368 #define ETH_RSS_NONFRAG_IPV4_TCP (1ULL << RTE_ETH_FLOW_NONFRAG_IPV4_TCP)
369 #define ETH_RSS_NONFRAG_IPV4_UDP (1ULL << RTE_ETH_FLOW_NONFRAG_IPV4_UDP)
370 #define ETH_RSS_NONFRAG_IPV4_SCTP (1ULL << RTE_ETH_FLOW_NONFRAG_IPV4_SCTP)
371 #define ETH_RSS_NONFRAG_IPV4_OTHER (1ULL << RTE_ETH_FLOW_NONFRAG_IPV4_OTHER)
372 #define ETH_RSS_IPV6 (1ULL << RTE_ETH_FLOW_IPV6)
373 #define ETH_RSS_FRAG_IPV6 (1ULL << RTE_ETH_FLOW_FRAG_IPV6)
374 #define ETH_RSS_NONFRAG_IPV6_TCP (1ULL << RTE_ETH_FLOW_NONFRAG_IPV6_TCP)
375 #define ETH_RSS_NONFRAG_IPV6_UDP (1ULL << RTE_ETH_FLOW_NONFRAG_IPV6_UDP)
376 #define ETH_RSS_NONFRAG_IPV6_SCTP (1ULL << RTE_ETH_FLOW_NONFRAG_IPV6_SCTP)
377 #define ETH_RSS_NONFRAG_IPV6_OTHER (1ULL << RTE_ETH_FLOW_NONFRAG_IPV6_OTHER)
378 #define ETH_RSS_L2_PAYLOAD (1ULL << RTE_ETH_FLOW_L2_PAYLOAD)
379 #define ETH_RSS_IPV6_EX (1ULL << RTE_ETH_FLOW_IPV6_EX)
380 #define ETH_RSS_IPV6_TCP_EX (1ULL << RTE_ETH_FLOW_IPV6_TCP_EX)
381 #define ETH_RSS_IPV6_UDP_EX (1ULL << RTE_ETH_FLOW_IPV6_UDP_EX)
383 #define ETH_RSS_IP ( \
385 ETH_RSS_FRAG_IPV4 | \
386 ETH_RSS_NONFRAG_IPV4_OTHER | \
388 ETH_RSS_FRAG_IPV6 | \
389 ETH_RSS_NONFRAG_IPV6_OTHER | \
392 #define ETH_RSS_UDP ( \
393 ETH_RSS_NONFRAG_IPV4_UDP | \
394 ETH_RSS_NONFRAG_IPV6_UDP | \
397 #define ETH_RSS_TCP ( \
398 ETH_RSS_NONFRAG_IPV4_TCP | \
399 ETH_RSS_NONFRAG_IPV6_TCP | \
402 #define ETH_RSS_SCTP ( \
403 ETH_RSS_NONFRAG_IPV4_SCTP | \
404 ETH_RSS_NONFRAG_IPV6_SCTP)
406 /**< Mask of valid RSS hash protocols */
407 #define ETH_RSS_PROTO_MASK ( \
409 ETH_RSS_FRAG_IPV4 | \
410 ETH_RSS_NONFRAG_IPV4_TCP | \
411 ETH_RSS_NONFRAG_IPV4_UDP | \
412 ETH_RSS_NONFRAG_IPV4_SCTP | \
413 ETH_RSS_NONFRAG_IPV4_OTHER | \
415 ETH_RSS_FRAG_IPV6 | \
416 ETH_RSS_NONFRAG_IPV6_TCP | \
417 ETH_RSS_NONFRAG_IPV6_UDP | \
418 ETH_RSS_NONFRAG_IPV6_SCTP | \
419 ETH_RSS_NONFRAG_IPV6_OTHER | \
420 ETH_RSS_L2_PAYLOAD | \
422 ETH_RSS_IPV6_TCP_EX | \
426 * Definitions used for redirection table entry size.
427 * Some RSS RETA sizes may not be supported by some drivers, check the
428 * documentation or the description of relevant functions for more details.
430 #define ETH_RSS_RETA_SIZE_64 64
431 #define ETH_RSS_RETA_SIZE_128 128
432 #define ETH_RSS_RETA_SIZE_512 512
433 #define RTE_RETA_GROUP_SIZE 64
435 /* Definitions used for VMDQ and DCB functionality */
436 #define ETH_VMDQ_MAX_VLAN_FILTERS 64 /**< Maximum nb. of VMDQ vlan filters. */
437 #define ETH_DCB_NUM_USER_PRIORITIES 8 /**< Maximum nb. of DCB priorities. */
438 #define ETH_VMDQ_DCB_NUM_QUEUES 128 /**< Maximum nb. of VMDQ DCB queues. */
439 #define ETH_DCB_NUM_QUEUES 128 /**< Maximum nb. of DCB queues. */
441 /* DCB capability defines */
442 #define ETH_DCB_PG_SUPPORT 0x00000001 /**< Priority Group(ETS) support. */
443 #define ETH_DCB_PFC_SUPPORT 0x00000002 /**< Priority Flow Control support. */
445 /* Definitions used for VLAN Offload functionality */
446 #define ETH_VLAN_STRIP_OFFLOAD 0x0001 /**< VLAN Strip On/Off */
447 #define ETH_VLAN_FILTER_OFFLOAD 0x0002 /**< VLAN Filter On/Off */
448 #define ETH_VLAN_EXTEND_OFFLOAD 0x0004 /**< VLAN Extend On/Off */
450 /* Definitions used for mask VLAN setting */
451 #define ETH_VLAN_STRIP_MASK 0x0001 /**< VLAN Strip setting mask */
452 #define ETH_VLAN_FILTER_MASK 0x0002 /**< VLAN Filter setting mask*/
453 #define ETH_VLAN_EXTEND_MASK 0x0004 /**< VLAN Extend setting mask*/
454 #define ETH_VLAN_ID_MAX 0x0FFF /**< VLAN ID is in lower 12 bits*/
456 /* Definitions used for receive MAC address */
457 #define ETH_NUM_RECEIVE_MAC_ADDR 128 /**< Maximum nb. of receive mac addr. */
459 /* Definitions used for unicast hash */
460 #define ETH_VMDQ_NUM_UC_HASH_ARRAY 128 /**< Maximum nb. of UC hash array. */
462 /* Definitions used for VMDQ pool rx mode setting */
463 #define ETH_VMDQ_ACCEPT_UNTAG 0x0001 /**< accept untagged packets. */
464 #define ETH_VMDQ_ACCEPT_HASH_MC 0x0002 /**< accept packets in multicast table . */
465 #define ETH_VMDQ_ACCEPT_HASH_UC 0x0004 /**< accept packets in unicast table. */
466 #define ETH_VMDQ_ACCEPT_BROADCAST 0x0008 /**< accept broadcast packets. */
467 #define ETH_VMDQ_ACCEPT_MULTICAST 0x0010 /**< multicast promiscuous. */
469 /* Definitions used for VMDQ mirror rules setting */
470 #define ETH_VMDQ_NUM_MIRROR_RULE 4 /**< Maximum nb. of mirror rules. . */
472 #define ETH_VMDQ_POOL_MIRROR 0x0001 /**< Virtual Pool Mirroring. */
473 #define ETH_VMDQ_UPLINK_MIRROR 0x0002 /**< Uplink Port Mirroring. */
474 #define ETH_VMDQ_DOWNLIN_MIRROR 0x0004 /**< Downlink Port Mirroring. */
475 #define ETH_VMDQ_VLAN_MIRROR 0x0008 /**< VLAN Mirroring. */
478 * A structure used to configure VLAN traffic mirror of an Ethernet port.
480 struct rte_eth_vlan_mirror {
481 uint64_t vlan_mask; /**< mask for valid VLAN ID. */
482 uint16_t vlan_id[ETH_VMDQ_MAX_VLAN_FILTERS];
483 /** VLAN ID list for vlan mirror. */
487 * A structure used to configure traffic mirror of an Ethernet port.
489 struct rte_eth_vmdq_mirror_conf {
490 uint8_t rule_type_mask; /**< Mirroring rule type mask we want to set */
491 uint8_t dst_pool; /**< Destination pool for this mirror rule. */
492 uint64_t pool_mask; /**< Bitmap of pool for pool mirroring */
493 struct rte_eth_vlan_mirror vlan; /**< VLAN ID setting for VLAN mirroring */
497 * A structure used to configure 64 entries of Redirection Table of the
498 * Receive Side Scaling (RSS) feature of an Ethernet port. To configure
499 * more than 64 entries supported by hardware, an array of this structure
502 struct rte_eth_rss_reta_entry64 {
504 /**< Mask bits indicate which entries need to be updated/queried. */
505 uint8_t reta[RTE_RETA_GROUP_SIZE];
506 /**< Group of 64 redirection table entries. */
510 * This enum indicates the possible number of traffic classes
511 * in DCB configratioins
513 enum rte_eth_nb_tcs {
514 ETH_4_TCS = 4, /**< 4 TCs with DCB. */
515 ETH_8_TCS = 8 /**< 8 TCs with DCB. */
519 * This enum indicates the possible number of queue pools
520 * in VMDQ configurations.
522 enum rte_eth_nb_pools {
523 ETH_8_POOLS = 8, /**< 8 VMDq pools. */
524 ETH_16_POOLS = 16, /**< 16 VMDq pools. */
525 ETH_32_POOLS = 32, /**< 32 VMDq pools. */
526 ETH_64_POOLS = 64 /**< 64 VMDq pools. */
529 /* This structure may be extended in future. */
530 struct rte_eth_dcb_rx_conf {
531 enum rte_eth_nb_tcs nb_tcs; /**< Possible DCB TCs, 4 or 8 TCs */
532 uint8_t dcb_queue[ETH_DCB_NUM_USER_PRIORITIES];
533 /**< Possible DCB queue,4 or 8. */
536 struct rte_eth_vmdq_dcb_tx_conf {
537 enum rte_eth_nb_pools nb_queue_pools; /**< With DCB, 16 or 32 pools. */
538 uint8_t dcb_queue[ETH_DCB_NUM_USER_PRIORITIES];
539 /**< Possible DCB queue,4 or 8. */
542 struct rte_eth_dcb_tx_conf {
543 enum rte_eth_nb_tcs nb_tcs; /**< Possible DCB TCs, 4 or 8 TCs. */
544 uint8_t dcb_queue[ETH_DCB_NUM_USER_PRIORITIES];
545 /**< Possible DCB queue,4 or 8. */
548 struct rte_eth_vmdq_tx_conf {
549 enum rte_eth_nb_pools nb_queue_pools; /**< VMDq mode, 64 pools. */
553 * A structure used to configure the VMDQ+DCB feature
554 * of an Ethernet port.
556 * Using this feature, packets are routed to a pool of queues, based
557 * on the vlan id in the vlan tag, and then to a specific queue within
558 * that pool, using the user priority vlan tag field.
560 * A default pool may be used, if desired, to route all traffic which
561 * does not match the vlan filter rules.
563 struct rte_eth_vmdq_dcb_conf {
564 enum rte_eth_nb_pools nb_queue_pools; /**< With DCB, 16 or 32 pools */
565 uint8_t enable_default_pool; /**< If non-zero, use a default pool */
566 uint8_t default_pool; /**< The default pool, if applicable */
567 uint8_t nb_pool_maps; /**< We can have up to 64 filters/mappings */
569 uint16_t vlan_id; /**< The vlan id of the received frame */
570 uint64_t pools; /**< Bitmask of pools for packet rx */
571 } pool_map[ETH_VMDQ_MAX_VLAN_FILTERS]; /**< VMDq vlan pool maps. */
572 uint8_t dcb_queue[ETH_DCB_NUM_USER_PRIORITIES];
573 /**< Selects a queue in a pool */
576 struct rte_eth_vmdq_rx_conf {
577 enum rte_eth_nb_pools nb_queue_pools; /**< VMDq only mode, 8 or 64 pools */
578 uint8_t enable_default_pool; /**< If non-zero, use a default pool */
579 uint8_t default_pool; /**< The default pool, if applicable */
580 uint8_t enable_loop_back; /**< Enable VT loop back */
581 uint8_t nb_pool_maps; /**< We can have up to 64 filters/mappings */
582 uint32_t rx_mode; /**< Flags from ETH_VMDQ_ACCEPT_* */
584 uint16_t vlan_id; /**< The vlan id of the received frame */
585 uint64_t pools; /**< Bitmask of pools for packet rx */
586 } pool_map[ETH_VMDQ_MAX_VLAN_FILTERS]; /**< VMDq vlan pool maps. */
590 * A structure used to configure the TX features of an Ethernet port.
592 struct rte_eth_txmode {
593 enum rte_eth_tx_mq_mode mq_mode; /**< TX multi-queues mode. */
595 /* For i40e specifically */
597 uint8_t hw_vlan_reject_tagged : 1,
598 /**< If set, reject sending out tagged pkts */
599 hw_vlan_reject_untagged : 1,
600 /**< If set, reject sending out untagged pkts */
601 hw_vlan_insert_pvid : 1;
602 /**< If set, enable port based VLAN insertion */
606 * A structure used to configure an RX ring of an Ethernet port.
608 struct rte_eth_rxconf {
609 struct rte_eth_thresh rx_thresh; /**< RX ring threshold registers. */
610 uint16_t rx_free_thresh; /**< Drives the freeing of RX descriptors. */
611 uint8_t rx_drop_en; /**< Drop packets if no descriptors are available. */
612 uint8_t rx_deferred_start; /**< Do not start queue with rte_eth_dev_start(). */
615 #define ETH_TXQ_FLAGS_NOMULTSEGS 0x0001 /**< nb_segs=1 for all mbufs */
616 #define ETH_TXQ_FLAGS_NOREFCOUNT 0x0002 /**< refcnt can be ignored */
617 #define ETH_TXQ_FLAGS_NOMULTMEMP 0x0004 /**< all bufs come from same mempool */
618 #define ETH_TXQ_FLAGS_NOVLANOFFL 0x0100 /**< disable VLAN offload */
619 #define ETH_TXQ_FLAGS_NOXSUMSCTP 0x0200 /**< disable SCTP checksum offload */
620 #define ETH_TXQ_FLAGS_NOXSUMUDP 0x0400 /**< disable UDP checksum offload */
621 #define ETH_TXQ_FLAGS_NOXSUMTCP 0x0800 /**< disable TCP checksum offload */
622 #define ETH_TXQ_FLAGS_NOOFFLOADS \
623 (ETH_TXQ_FLAGS_NOVLANOFFL | ETH_TXQ_FLAGS_NOXSUMSCTP | \
624 ETH_TXQ_FLAGS_NOXSUMUDP | ETH_TXQ_FLAGS_NOXSUMTCP)
625 #define ETH_TXQ_FLAGS_NOXSUMS \
626 (ETH_TXQ_FLAGS_NOXSUMSCTP | ETH_TXQ_FLAGS_NOXSUMUDP | \
627 ETH_TXQ_FLAGS_NOXSUMTCP)
629 * A structure used to configure a TX ring of an Ethernet port.
631 struct rte_eth_txconf {
632 struct rte_eth_thresh tx_thresh; /**< TX ring threshold registers. */
633 uint16_t tx_rs_thresh; /**< Drives the setting of RS bit on TXDs. */
634 uint16_t tx_free_thresh; /**< Drives the freeing of TX buffers. */
635 uint32_t txq_flags; /**< Set flags for the Tx queue */
636 uint8_t tx_deferred_start; /**< Do not start queue with rte_eth_dev_start(). */
640 * This enum indicates the flow control mode
642 enum rte_eth_fc_mode {
643 RTE_FC_NONE = 0, /**< Disable flow control. */
644 RTE_FC_RX_PAUSE, /**< RX pause frame, enable flowctrl on TX side. */
645 RTE_FC_TX_PAUSE, /**< TX pause frame, enable flowctrl on RX side. */
646 RTE_FC_FULL /**< Enable flow control on both side. */
650 * A structure used to configure Ethernet flow control parameter.
651 * These parameters will be configured into the register of the NIC.
652 * Please refer to the corresponding data sheet for proper value.
654 struct rte_eth_fc_conf {
655 uint32_t high_water; /**< High threshold value to trigger XOFF */
656 uint32_t low_water; /**< Low threshold value to trigger XON */
657 uint16_t pause_time; /**< Pause quota in the Pause frame */
658 uint16_t send_xon; /**< Is XON frame need be sent */
659 enum rte_eth_fc_mode mode; /**< Link flow control mode */
660 uint8_t mac_ctrl_frame_fwd; /**< Forward MAC control frames */
661 uint8_t autoneg; /**< Use Pause autoneg */
665 * A structure used to configure Ethernet priority flow control parameter.
666 * These parameters will be configured into the register of the NIC.
667 * Please refer to the corresponding data sheet for proper value.
669 struct rte_eth_pfc_conf {
670 struct rte_eth_fc_conf fc; /**< General flow control parameter. */
671 uint8_t priority; /**< VLAN User Priority. */
675 * Memory space that can be configured to store Flow Director filters
676 * in the board memory.
678 enum rte_fdir_pballoc_type {
679 RTE_FDIR_PBALLOC_64K = 0, /**< 64k. */
680 RTE_FDIR_PBALLOC_128K, /**< 128k. */
681 RTE_FDIR_PBALLOC_256K, /**< 256k. */
685 * Select report mode of FDIR hash information in RX descriptors.
687 enum rte_fdir_status_mode {
688 RTE_FDIR_NO_REPORT_STATUS = 0, /**< Never report FDIR hash. */
689 RTE_FDIR_REPORT_STATUS, /**< Only report FDIR hash for matching pkts. */
690 RTE_FDIR_REPORT_STATUS_ALWAYS, /**< Always report FDIR hash. */
694 * A structure used to configure the Flow Director (FDIR) feature
695 * of an Ethernet port.
697 * If mode is RTE_FDIR_DISABLE, the pballoc value is ignored.
699 struct rte_fdir_conf {
700 enum rte_fdir_mode mode; /**< Flow Director mode. */
701 enum rte_fdir_pballoc_type pballoc; /**< Space for FDIR filters. */
702 enum rte_fdir_status_mode status; /**< How to report FDIR hash. */
703 /** RX queue of packets matching a "drop" filter in perfect mode. */
705 struct rte_eth_fdir_masks mask;
706 struct rte_eth_fdir_flex_conf flex_conf;
707 /**< Flex payload configuration. */
711 * UDP tunneling configuration.
713 struct rte_eth_udp_tunnel {
719 * Possible l4type of FDIR filters.
722 RTE_FDIR_L4TYPE_NONE = 0, /**< None. */
723 RTE_FDIR_L4TYPE_UDP, /**< UDP. */
724 RTE_FDIR_L4TYPE_TCP, /**< TCP. */
725 RTE_FDIR_L4TYPE_SCTP, /**< SCTP. */
729 * Select IPv4 or IPv6 FDIR filters.
732 RTE_FDIR_IPTYPE_IPV4 = 0, /**< IPv4. */
733 RTE_FDIR_IPTYPE_IPV6 , /**< IPv6. */
737 * A structure used to define a FDIR packet filter.
739 struct rte_fdir_filter {
740 uint16_t flex_bytes; /**< Flex bytes value to match. */
741 uint16_t vlan_id; /**< VLAN ID value to match, 0 otherwise. */
742 uint16_t port_src; /**< Source port to match, 0 otherwise. */
743 uint16_t port_dst; /**< Destination port to match, 0 otherwise. */
745 uint32_t ipv4_addr; /**< IPv4 source address to match. */
746 uint32_t ipv6_addr[4]; /**< IPv6 source address to match. */
747 } ip_src; /**< IPv4/IPv6 source address to match (union of above). */
749 uint32_t ipv4_addr; /**< IPv4 destination address to match. */
750 uint32_t ipv6_addr[4]; /**< IPv6 destination address to match */
751 } ip_dst; /**< IPv4/IPv6 destination address to match (union of above). */
752 enum rte_l4type l4type; /**< l4type to match: NONE/UDP/TCP/SCTP. */
753 enum rte_iptype iptype; /**< IP packet type to match: IPv4 or IPv6. */
757 * A structure used to configure FDIR masks that are used by the device
758 * to match the various fields of RX packet headers.
759 * @note The only_ip_flow field has the opposite meaning compared to other
762 struct rte_fdir_masks {
763 /** When set to 1, packet l4type is \b NOT relevant in filters, and
764 source and destination port masks must be set to zero. */
765 uint8_t only_ip_flow;
766 /** If set to 1, vlan_id is relevant in filters. */
768 /** If set to 1, vlan_prio is relevant in filters. */
770 /** If set to 1, flexbytes is relevant in filters. */
772 /** If set to 1, set the IPv6 masks. Otherwise set the IPv4 masks. */
773 uint8_t set_ipv6_mask;
774 /** When set to 1, comparison of destination IPv6 address with IP6AT
775 registers is meaningful. */
776 uint8_t comp_ipv6_dst;
777 /** Mask of Destination IPv4 Address. All bits set to 1 define the
778 relevant bits to use in the destination address of an IPv4 packet
779 when matching it against FDIR filters. */
780 uint32_t dst_ipv4_mask;
781 /** Mask of Source IPv4 Address. All bits set to 1 define
782 the relevant bits to use in the source address of an IPv4 packet
783 when matching it against FDIR filters. */
784 uint32_t src_ipv4_mask;
785 /** Mask of Source IPv6 Address. All bits set to 1 define the
786 relevant BYTES to use in the source address of an IPv6 packet
787 when matching it against FDIR filters. */
788 uint16_t dst_ipv6_mask;
789 /** Mask of Destination IPv6 Address. All bits set to 1 define the
790 relevant BYTES to use in the destination address of an IPv6 packet
791 when matching it against FDIR filters. */
792 uint16_t src_ipv6_mask;
793 /** Mask of Source Port. All bits set to 1 define the relevant
794 bits to use in the source port of an IP packets when matching it
795 against FDIR filters. */
796 uint16_t src_port_mask;
797 /** Mask of Destination Port. All bits set to 1 define the relevant
798 bits to use in the destination port of an IP packet when matching it
799 against FDIR filters. */
800 uint16_t dst_port_mask;
804 * A structure used to report the status of the flow director filters in use.
806 struct rte_eth_fdir {
807 /** Number of filters with collision indication. */
809 /** Number of free (non programmed) filters. */
811 /** The Lookup hash value of the added filter that updated the value
812 of the MAXLEN field */
814 /** Longest linked list of filters in the table. */
816 /** Number of added filters. */
818 /** Number of removed filters. */
820 /** Number of failed added filters (no more space in device). */
822 /** Number of failed removed filters. */
827 * A structure used to enable/disable specific device interrupts.
829 struct rte_intr_conf {
830 /** enable/disable lsc interrupt. 0 (default) - disable, 1 enable */
835 * A structure used to configure an Ethernet port.
836 * Depending upon the RX multi-queue mode, extra advanced
837 * configuration settings may be needed.
839 struct rte_eth_conf {
841 /**< ETH_LINK_SPEED_10[0|00|000], or 0 for autonegotation */
842 uint16_t link_duplex;
843 /**< ETH_LINK_[HALF_DUPLEX|FULL_DUPLEX], or 0 for autonegotation */
844 struct rte_eth_rxmode rxmode; /**< Port RX configuration. */
845 struct rte_eth_txmode txmode; /**< Port TX configuration. */
846 uint32_t lpbk_mode; /**< Loopback operation mode. By default the value
847 is 0, meaning the loopback mode is disabled.
848 Read the datasheet of given ethernet controller
849 for details. The possible values of this field
850 are defined in implementation of each driver. */
852 struct rte_eth_rss_conf rss_conf; /**< Port RSS configuration */
853 struct rte_eth_vmdq_dcb_conf vmdq_dcb_conf;
854 /**< Port vmdq+dcb configuration. */
855 struct rte_eth_dcb_rx_conf dcb_rx_conf;
856 /**< Port dcb RX configuration. */
857 struct rte_eth_vmdq_rx_conf vmdq_rx_conf;
858 /**< Port vmdq RX configuration. */
859 } rx_adv_conf; /**< Port RX filtering configuration (union). */
861 struct rte_eth_vmdq_dcb_tx_conf vmdq_dcb_tx_conf;
862 /**< Port vmdq+dcb TX configuration. */
863 struct rte_eth_dcb_tx_conf dcb_tx_conf;
864 /**< Port dcb TX configuration. */
865 struct rte_eth_vmdq_tx_conf vmdq_tx_conf;
866 /**< Port vmdq TX configuration. */
867 } tx_adv_conf; /**< Port TX DCB configuration (union). */
868 /** Currently,Priority Flow Control(PFC) are supported,if DCB with PFC
869 is needed,and the variable must be set ETH_DCB_PFC_SUPPORT. */
870 uint32_t dcb_capability_en;
871 struct rte_fdir_conf fdir_conf; /**< FDIR configuration. */
872 struct rte_intr_conf intr_conf; /**< Interrupt mode configuration. */
876 * A structure used to retrieve the contextual information of
877 * an Ethernet device, such as the controlling driver of the device,
878 * its PCI context, etc...
882 * RX offload capabilities of a device.
884 #define DEV_RX_OFFLOAD_VLAN_STRIP 0x00000001
885 #define DEV_RX_OFFLOAD_IPV4_CKSUM 0x00000002
886 #define DEV_RX_OFFLOAD_UDP_CKSUM 0x00000004
887 #define DEV_RX_OFFLOAD_TCP_CKSUM 0x00000008
888 #define DEV_RX_OFFLOAD_TCP_LRO 0x00000010
891 * TX offload capabilities of a device.
893 #define DEV_TX_OFFLOAD_VLAN_INSERT 0x00000001
894 #define DEV_TX_OFFLOAD_IPV4_CKSUM 0x00000002
895 #define DEV_TX_OFFLOAD_UDP_CKSUM 0x00000004
896 #define DEV_TX_OFFLOAD_TCP_CKSUM 0x00000008
897 #define DEV_TX_OFFLOAD_SCTP_CKSUM 0x00000010
898 #define DEV_TX_OFFLOAD_TCP_TSO 0x00000020
899 #define DEV_TX_OFFLOAD_UDP_TSO 0x00000040
900 #define DEV_TX_OFFLOAD_OUTER_IPV4_CKSUM 0x00000080 /**< Used for tunneling packet. */
902 struct rte_eth_dev_info {
903 struct rte_pci_device *pci_dev; /**< Device PCI information. */
904 const char *driver_name; /**< Device Driver name. */
905 unsigned int if_index; /**< Index to bound host interface, or 0 if none.
906 Use if_indextoname() to translate into an interface name. */
907 uint32_t min_rx_bufsize; /**< Minimum size of RX buffer. */
908 uint32_t max_rx_pktlen; /**< Maximum configurable length of RX pkt. */
909 uint16_t max_rx_queues; /**< Maximum number of RX queues. */
910 uint16_t max_tx_queues; /**< Maximum number of TX queues. */
911 uint32_t max_mac_addrs; /**< Maximum number of MAC addresses. */
912 uint32_t max_hash_mac_addrs;
913 /** Maximum number of hash MAC addresses for MTA and UTA. */
914 uint16_t max_vfs; /**< Maximum number of VFs. */
915 uint16_t max_vmdq_pools; /**< Maximum number of VMDq pools. */
916 uint32_t rx_offload_capa; /**< Device RX offload capabilities. */
917 uint32_t tx_offload_capa; /**< Device TX offload capabilities. */
919 /**< Device redirection table size, the total number of entries. */
920 /** Bit mask of RSS offloads, the bit offset also means flow type */
921 uint64_t flow_type_rss_offloads;
922 struct rte_eth_rxconf default_rxconf; /**< Default RX configuration */
923 struct rte_eth_txconf default_txconf; /**< Default TX configuration */
924 uint16_t vmdq_queue_base; /**< First queue ID for VMDQ pools. */
925 uint16_t vmdq_queue_num; /**< Queue number for VMDQ pools. */
926 uint16_t vmdq_pool_base; /**< First ID of VMDQ pools. */
929 /** Maximum name length for extended statistics counters */
930 #define RTE_ETH_XSTATS_NAME_SIZE 64
933 * An Ethernet device extended statistic structure
935 * This structure is used by ethdev->eth_xstats_get() to provide
936 * statistics that are not provided in the generic rte_eth_stats
939 struct rte_eth_xstats {
940 char name[RTE_ETH_XSTATS_NAME_SIZE];
946 struct rte_eth_dev_callback;
947 /** @internal Structure to keep track of registered callbacks */
948 TAILQ_HEAD(rte_eth_dev_cb_list, rte_eth_dev_callback);
951 * Definitions of all functions exported by an Ethernet driver through the
952 * the generic structure of type *eth_dev_ops* supplied in the *rte_eth_dev*
953 * structure associated with an Ethernet device.
956 typedef int (*eth_dev_configure_t)(struct rte_eth_dev *dev);
957 /**< @internal Ethernet device configuration. */
959 typedef int (*eth_dev_start_t)(struct rte_eth_dev *dev);
960 /**< @internal Function used to start a configured Ethernet device. */
962 typedef void (*eth_dev_stop_t)(struct rte_eth_dev *dev);
963 /**< @internal Function used to stop a configured Ethernet device. */
965 typedef int (*eth_dev_set_link_up_t)(struct rte_eth_dev *dev);
966 /**< @internal Function used to link up a configured Ethernet device. */
968 typedef int (*eth_dev_set_link_down_t)(struct rte_eth_dev *dev);
969 /**< @internal Function used to link down a configured Ethernet device. */
971 typedef void (*eth_dev_close_t)(struct rte_eth_dev *dev);
972 /**< @internal Function used to close a configured Ethernet device. */
974 typedef void (*eth_promiscuous_enable_t)(struct rte_eth_dev *dev);
975 /**< @internal Function used to enable the RX promiscuous mode of an Ethernet device. */
977 typedef void (*eth_promiscuous_disable_t)(struct rte_eth_dev *dev);
978 /**< @internal Function used to disable the RX promiscuous mode of an Ethernet device. */
980 typedef void (*eth_allmulticast_enable_t)(struct rte_eth_dev *dev);
981 /**< @internal Enable the receipt of all multicast packets by an Ethernet device. */
983 typedef void (*eth_allmulticast_disable_t)(struct rte_eth_dev *dev);
984 /**< @internal Disable the receipt of all multicast packets by an Ethernet device. */
986 typedef int (*eth_link_update_t)(struct rte_eth_dev *dev,
987 int wait_to_complete);
988 /**< @internal Get link speed, duplex mode and state (up/down) of an Ethernet device. */
990 typedef void (*eth_stats_get_t)(struct rte_eth_dev *dev,
991 struct rte_eth_stats *igb_stats);
992 /**< @internal Get global I/O statistics of an Ethernet device. */
994 typedef void (*eth_stats_reset_t)(struct rte_eth_dev *dev);
995 /**< @internal Reset global I/O statistics of an Ethernet device to 0. */
997 typedef int (*eth_xstats_get_t)(struct rte_eth_dev *dev,
998 struct rte_eth_xstats *stats, unsigned n);
999 /**< @internal Get extended stats of an Ethernet device. */
1001 typedef void (*eth_xstats_reset_t)(struct rte_eth_dev *dev);
1002 /**< @internal Reset extended stats of an Ethernet device. */
1004 typedef int (*eth_queue_stats_mapping_set_t)(struct rte_eth_dev *dev,
1008 /**< @internal Set a queue statistics mapping for a tx/rx queue of an Ethernet device. */
1010 typedef void (*eth_dev_infos_get_t)(struct rte_eth_dev *dev,
1011 struct rte_eth_dev_info *dev_info);
1012 /**< @internal Get specific informations of an Ethernet device. */
1014 typedef int (*eth_queue_start_t)(struct rte_eth_dev *dev,
1016 /**< @internal Start rx and tx of a queue of an Ethernet device. */
1018 typedef int (*eth_queue_stop_t)(struct rte_eth_dev *dev,
1020 /**< @internal Stop rx and tx of a queue of an Ethernet device. */
1022 typedef int (*eth_rx_queue_setup_t)(struct rte_eth_dev *dev,
1023 uint16_t rx_queue_id,
1024 uint16_t nb_rx_desc,
1025 unsigned int socket_id,
1026 const struct rte_eth_rxconf *rx_conf,
1027 struct rte_mempool *mb_pool);
1028 /**< @internal Set up a receive queue of an Ethernet device. */
1030 typedef int (*eth_tx_queue_setup_t)(struct rte_eth_dev *dev,
1031 uint16_t tx_queue_id,
1032 uint16_t nb_tx_desc,
1033 unsigned int socket_id,
1034 const struct rte_eth_txconf *tx_conf);
1035 /**< @internal Setup a transmit queue of an Ethernet device. */
1037 typedef void (*eth_queue_release_t)(void *queue);
1038 /**< @internal Release memory resources allocated by given RX/TX queue. */
1040 typedef uint32_t (*eth_rx_queue_count_t)(struct rte_eth_dev *dev,
1041 uint16_t rx_queue_id);
1042 /**< @Get number of available descriptors on a receive queue of an Ethernet device. */
1044 typedef int (*eth_rx_descriptor_done_t)(void *rxq, uint16_t offset);
1045 /**< @Check DD bit of specific RX descriptor */
1047 typedef int (*mtu_set_t)(struct rte_eth_dev *dev, uint16_t mtu);
1048 /**< @internal Set MTU. */
1050 typedef int (*vlan_filter_set_t)(struct rte_eth_dev *dev,
1053 /**< @internal filtering of a VLAN Tag Identifier by an Ethernet device. */
1055 typedef void (*vlan_tpid_set_t)(struct rte_eth_dev *dev,
1057 /**< @internal set the outer VLAN-TPID by an Ethernet device. */
1059 typedef void (*vlan_offload_set_t)(struct rte_eth_dev *dev, int mask);
1060 /**< @internal set VLAN offload function by an Ethernet device. */
1062 typedef int (*vlan_pvid_set_t)(struct rte_eth_dev *dev,
1065 /**< @internal set port based TX VLAN insertion by an Ethernet device. */
1067 typedef void (*vlan_strip_queue_set_t)(struct rte_eth_dev *dev,
1068 uint16_t rx_queue_id,
1070 /**< @internal VLAN stripping enable/disable by an queue of Ethernet device. */
1072 typedef uint16_t (*eth_rx_burst_t)(void *rxq,
1073 struct rte_mbuf **rx_pkts,
1075 /**< @internal Retrieve input packets from a receive queue of an Ethernet device. */
1077 typedef uint16_t (*eth_tx_burst_t)(void *txq,
1078 struct rte_mbuf **tx_pkts,
1080 /**< @internal Send output packets on a transmit queue of an Ethernet device. */
1082 typedef int (*fdir_add_signature_filter_t)(struct rte_eth_dev *dev,
1083 struct rte_fdir_filter *fdir_ftr,
1085 /**< @internal Setup a new signature filter rule on an Ethernet device */
1087 typedef int (*fdir_update_signature_filter_t)(struct rte_eth_dev *dev,
1088 struct rte_fdir_filter *fdir_ftr,
1090 /**< @internal Update a signature filter rule on an Ethernet device */
1092 typedef int (*fdir_remove_signature_filter_t)(struct rte_eth_dev *dev,
1093 struct rte_fdir_filter *fdir_ftr);
1094 /**< @internal Remove a signature filter rule on an Ethernet device */
1096 typedef void (*fdir_infos_get_t)(struct rte_eth_dev *dev,
1097 struct rte_eth_fdir *fdir);
1098 /**< @internal Get information about fdir status */
1100 typedef int (*fdir_add_perfect_filter_t)(struct rte_eth_dev *dev,
1101 struct rte_fdir_filter *fdir_ftr,
1102 uint16_t soft_id, uint8_t rx_queue,
1104 /**< @internal Setup a new perfect filter rule on an Ethernet device */
1106 typedef int (*fdir_update_perfect_filter_t)(struct rte_eth_dev *dev,
1107 struct rte_fdir_filter *fdir_ftr,
1108 uint16_t soft_id, uint8_t rx_queue,
1110 /**< @internal Update a perfect filter rule on an Ethernet device */
1112 typedef int (*fdir_remove_perfect_filter_t)(struct rte_eth_dev *dev,
1113 struct rte_fdir_filter *fdir_ftr,
1115 /**< @internal Remove a perfect filter rule on an Ethernet device */
1117 typedef int (*fdir_set_masks_t)(struct rte_eth_dev *dev,
1118 struct rte_fdir_masks *fdir_masks);
1119 /**< @internal Setup flow director masks on an Ethernet device */
1121 typedef int (*flow_ctrl_get_t)(struct rte_eth_dev *dev,
1122 struct rte_eth_fc_conf *fc_conf);
1123 /**< @internal Get current flow control parameter on an Ethernet device */
1125 typedef int (*flow_ctrl_set_t)(struct rte_eth_dev *dev,
1126 struct rte_eth_fc_conf *fc_conf);
1127 /**< @internal Setup flow control parameter on an Ethernet device */
1129 typedef int (*priority_flow_ctrl_set_t)(struct rte_eth_dev *dev,
1130 struct rte_eth_pfc_conf *pfc_conf);
1131 /**< @internal Setup priority flow control parameter on an Ethernet device */
1133 typedef int (*reta_update_t)(struct rte_eth_dev *dev,
1134 struct rte_eth_rss_reta_entry64 *reta_conf,
1135 uint16_t reta_size);
1136 /**< @internal Update RSS redirection table on an Ethernet device */
1138 typedef int (*reta_query_t)(struct rte_eth_dev *dev,
1139 struct rte_eth_rss_reta_entry64 *reta_conf,
1140 uint16_t reta_size);
1141 /**< @internal Query RSS redirection table on an Ethernet device */
1143 typedef int (*rss_hash_update_t)(struct rte_eth_dev *dev,
1144 struct rte_eth_rss_conf *rss_conf);
1145 /**< @internal Update RSS hash configuration of an Ethernet device */
1147 typedef int (*rss_hash_conf_get_t)(struct rte_eth_dev *dev,
1148 struct rte_eth_rss_conf *rss_conf);
1149 /**< @internal Get current RSS hash configuration of an Ethernet device */
1151 typedef int (*eth_dev_led_on_t)(struct rte_eth_dev *dev);
1152 /**< @internal Turn on SW controllable LED on an Ethernet device */
1154 typedef int (*eth_dev_led_off_t)(struct rte_eth_dev *dev);
1155 /**< @internal Turn off SW controllable LED on an Ethernet device */
1157 typedef void (*eth_mac_addr_remove_t)(struct rte_eth_dev *dev, uint32_t index);
1158 /**< @internal Remove MAC address from receive address register */
1160 typedef void (*eth_mac_addr_add_t)(struct rte_eth_dev *dev,
1161 struct ether_addr *mac_addr,
1164 /**< @internal Set a MAC address into Receive Address Address Register */
1166 typedef void (*eth_mac_addr_set_t)(struct rte_eth_dev *dev,
1167 struct ether_addr *mac_addr);
1168 /**< @internal Set a MAC address into Receive Address Address Register */
1170 typedef int (*eth_uc_hash_table_set_t)(struct rte_eth_dev *dev,
1171 struct ether_addr *mac_addr,
1173 /**< @internal Set a Unicast Hash bitmap */
1175 typedef int (*eth_uc_all_hash_table_set_t)(struct rte_eth_dev *dev,
1177 /**< @internal Set all Unicast Hash bitmap */
1179 typedef int (*eth_set_vf_rx_mode_t)(struct rte_eth_dev *dev,
1183 /**< @internal Set a VF receive mode */
1185 typedef int (*eth_set_vf_rx_t)(struct rte_eth_dev *dev,
1188 /**< @internal Set a VF receive mode */
1190 typedef int (*eth_set_vf_tx_t)(struct rte_eth_dev *dev,
1193 /**< @internal Enable or disable a VF transmit */
1195 typedef int (*eth_set_vf_vlan_filter_t)(struct rte_eth_dev *dev,
1199 /**< @internal Set VF VLAN pool filter */
1201 typedef int (*eth_set_queue_rate_limit_t)(struct rte_eth_dev *dev,
1204 /**< @internal Set queue TX rate */
1206 typedef int (*eth_set_vf_rate_limit_t)(struct rte_eth_dev *dev,
1210 /**< @internal Set VF TX rate */
1212 typedef int (*eth_mirror_rule_set_t)(struct rte_eth_dev *dev,
1213 struct rte_eth_vmdq_mirror_conf *mirror_conf,
1216 /**< @internal Add a traffic mirroring rule on an Ethernet device */
1218 typedef int (*eth_mirror_rule_reset_t)(struct rte_eth_dev *dev,
1220 /**< @internal Remove a traffic mirroring rule on an Ethernet device */
1222 typedef int (*eth_udp_tunnel_add_t)(struct rte_eth_dev *dev,
1223 struct rte_eth_udp_tunnel *tunnel_udp);
1224 /**< @internal Add tunneling UDP info */
1226 typedef int (*eth_udp_tunnel_del_t)(struct rte_eth_dev *dev,
1227 struct rte_eth_udp_tunnel *tunnel_udp);
1228 /**< @internal Delete tunneling UDP info */
1231 #ifdef RTE_NIC_BYPASS
1234 RTE_BYPASS_MODE_NONE,
1235 RTE_BYPASS_MODE_NORMAL,
1236 RTE_BYPASS_MODE_BYPASS,
1237 RTE_BYPASS_MODE_ISOLATE,
1238 RTE_BYPASS_MODE_NUM,
1241 #define RTE_BYPASS_MODE_VALID(x) \
1242 ((x) > RTE_BYPASS_MODE_NONE && (x) < RTE_BYPASS_MODE_NUM)
1245 RTE_BYPASS_EVENT_NONE,
1246 RTE_BYPASS_EVENT_START,
1247 RTE_BYPASS_EVENT_OS_ON = RTE_BYPASS_EVENT_START,
1248 RTE_BYPASS_EVENT_POWER_ON,
1249 RTE_BYPASS_EVENT_OS_OFF,
1250 RTE_BYPASS_EVENT_POWER_OFF,
1251 RTE_BYPASS_EVENT_TIMEOUT,
1252 RTE_BYPASS_EVENT_NUM
1255 #define RTE_BYPASS_EVENT_VALID(x) \
1256 ((x) > RTE_BYPASS_EVENT_NONE && (x) < RTE_BYPASS_MODE_NUM)
1259 RTE_BYPASS_TMT_OFF, /* timeout disabled. */
1260 RTE_BYPASS_TMT_1_5_SEC, /* timeout for 1.5 seconds */
1261 RTE_BYPASS_TMT_2_SEC, /* timeout for 2 seconds */
1262 RTE_BYPASS_TMT_3_SEC, /* timeout for 3 seconds */
1263 RTE_BYPASS_TMT_4_SEC, /* timeout for 4 seconds */
1264 RTE_BYPASS_TMT_8_SEC, /* timeout for 8 seconds */
1265 RTE_BYPASS_TMT_16_SEC, /* timeout for 16 seconds */
1266 RTE_BYPASS_TMT_32_SEC, /* timeout for 32 seconds */
1270 #define RTE_BYPASS_TMT_VALID(x) \
1271 ((x) == RTE_BYPASS_TMT_OFF || \
1272 ((x) > RTE_BYPASS_TMT_OFF && (x) < RTE_BYPASS_TMT_NUM))
1274 typedef void (*bypass_init_t)(struct rte_eth_dev *dev);
1275 typedef int32_t (*bypass_state_set_t)(struct rte_eth_dev *dev, uint32_t *new_state);
1276 typedef int32_t (*bypass_state_show_t)(struct rte_eth_dev *dev, uint32_t *state);
1277 typedef int32_t (*bypass_event_set_t)(struct rte_eth_dev *dev, uint32_t state, uint32_t event);
1278 typedef int32_t (*bypass_event_show_t)(struct rte_eth_dev *dev, uint32_t event_shift, uint32_t *event);
1279 typedef int32_t (*bypass_wd_timeout_set_t)(struct rte_eth_dev *dev, uint32_t timeout);
1280 typedef int32_t (*bypass_wd_timeout_show_t)(struct rte_eth_dev *dev, uint32_t *wd_timeout);
1281 typedef int32_t (*bypass_ver_show_t)(struct rte_eth_dev *dev, uint32_t *ver);
1282 typedef int32_t (*bypass_wd_reset_t)(struct rte_eth_dev *dev);
1285 typedef int (*eth_filter_ctrl_t)(struct rte_eth_dev *dev,
1286 enum rte_filter_type filter_type,
1287 enum rte_filter_op filter_op,
1289 /**< @internal Take operations to assigned filter type on an Ethernet device */
1292 * @internal A structure containing the functions exported by an Ethernet driver.
1294 struct eth_dev_ops {
1295 eth_dev_configure_t dev_configure; /**< Configure device. */
1296 eth_dev_start_t dev_start; /**< Start device. */
1297 eth_dev_stop_t dev_stop; /**< Stop device. */
1298 eth_dev_set_link_up_t dev_set_link_up; /**< Device link up. */
1299 eth_dev_set_link_down_t dev_set_link_down; /**< Device link down. */
1300 eth_dev_close_t dev_close; /**< Close device. */
1301 eth_promiscuous_enable_t promiscuous_enable; /**< Promiscuous ON. */
1302 eth_promiscuous_disable_t promiscuous_disable;/**< Promiscuous OFF. */
1303 eth_allmulticast_enable_t allmulticast_enable;/**< RX multicast ON. */
1304 eth_allmulticast_disable_t allmulticast_disable;/**< RX multicast OF. */
1305 eth_link_update_t link_update; /**< Get device link state. */
1306 eth_stats_get_t stats_get; /**< Get generic device statistics. */
1307 eth_stats_reset_t stats_reset; /**< Reset generic device statistics. */
1308 eth_xstats_get_t xstats_get; /**< Get extended device statistics. */
1309 eth_xstats_reset_t xstats_reset; /**< Reset extended device statistics. */
1310 eth_queue_stats_mapping_set_t queue_stats_mapping_set;
1311 /**< Configure per queue stat counter mapping. */
1312 eth_dev_infos_get_t dev_infos_get; /**< Get device info. */
1313 mtu_set_t mtu_set; /**< Set MTU. */
1314 vlan_filter_set_t vlan_filter_set; /**< Filter VLAN Setup. */
1315 vlan_tpid_set_t vlan_tpid_set; /**< Outer VLAN TPID Setup. */
1316 vlan_strip_queue_set_t vlan_strip_queue_set; /**< VLAN Stripping on queue. */
1317 vlan_offload_set_t vlan_offload_set; /**< Set VLAN Offload. */
1318 vlan_pvid_set_t vlan_pvid_set; /**< Set port based TX VLAN insertion */
1319 eth_queue_start_t rx_queue_start;/**< Start RX for a queue.*/
1320 eth_queue_stop_t rx_queue_stop;/**< Stop RX for a queue.*/
1321 eth_queue_start_t tx_queue_start;/**< Start TX for a queue.*/
1322 eth_queue_stop_t tx_queue_stop;/**< Stop TX for a queue.*/
1323 eth_rx_queue_setup_t rx_queue_setup;/**< Set up device RX queue.*/
1324 eth_queue_release_t rx_queue_release;/**< Release RX queue.*/
1325 eth_rx_queue_count_t rx_queue_count; /**< Get Rx queue count. */
1326 eth_rx_descriptor_done_t rx_descriptor_done; /**< Check rxd DD bit */
1327 eth_tx_queue_setup_t tx_queue_setup;/**< Set up device TX queue.*/
1328 eth_queue_release_t tx_queue_release;/**< Release TX queue.*/
1329 eth_dev_led_on_t dev_led_on; /**< Turn on LED. */
1330 eth_dev_led_off_t dev_led_off; /**< Turn off LED. */
1331 flow_ctrl_get_t flow_ctrl_get; /**< Get flow control. */
1332 flow_ctrl_set_t flow_ctrl_set; /**< Setup flow control. */
1333 priority_flow_ctrl_set_t priority_flow_ctrl_set; /**< Setup priority flow control.*/
1334 eth_mac_addr_remove_t mac_addr_remove; /**< Remove MAC address */
1335 eth_mac_addr_add_t mac_addr_add; /**< Add a MAC address */
1336 eth_mac_addr_set_t mac_addr_set; /**< Set a MAC address */
1337 eth_uc_hash_table_set_t uc_hash_table_set; /**< Set Unicast Table Array */
1338 eth_uc_all_hash_table_set_t uc_all_hash_table_set; /**< Set Unicast hash bitmap */
1339 eth_mirror_rule_set_t mirror_rule_set; /**< Add a traffic mirror rule.*/
1340 eth_mirror_rule_reset_t mirror_rule_reset; /**< reset a traffic mirror rule.*/
1341 eth_set_vf_rx_mode_t set_vf_rx_mode; /**< Set VF RX mode */
1342 eth_set_vf_rx_t set_vf_rx; /**< enable/disable a VF receive */
1343 eth_set_vf_tx_t set_vf_tx; /**< enable/disable a VF transmit */
1344 eth_set_vf_vlan_filter_t set_vf_vlan_filter; /**< Set VF VLAN filter */
1345 eth_udp_tunnel_add_t udp_tunnel_add;
1346 eth_udp_tunnel_del_t udp_tunnel_del;
1347 eth_set_queue_rate_limit_t set_queue_rate_limit; /**< Set queue rate limit */
1348 eth_set_vf_rate_limit_t set_vf_rate_limit; /**< Set VF rate limit */
1350 /** Add a signature filter. */
1351 fdir_add_signature_filter_t fdir_add_signature_filter;
1352 /** Update a signature filter. */
1353 fdir_update_signature_filter_t fdir_update_signature_filter;
1354 /** Remove a signature filter. */
1355 fdir_remove_signature_filter_t fdir_remove_signature_filter;
1356 /** Get information about FDIR status. */
1357 fdir_infos_get_t fdir_infos_get;
1358 /** Add a perfect filter. */
1359 fdir_add_perfect_filter_t fdir_add_perfect_filter;
1360 /** Update a perfect filter. */
1361 fdir_update_perfect_filter_t fdir_update_perfect_filter;
1362 /** Remove a perfect filter. */
1363 fdir_remove_perfect_filter_t fdir_remove_perfect_filter;
1364 /** Setup masks for FDIR filtering. */
1365 fdir_set_masks_t fdir_set_masks;
1366 /** Update redirection table. */
1367 reta_update_t reta_update;
1368 /** Query redirection table. */
1369 reta_query_t reta_query;
1370 /* bypass control */
1371 #ifdef RTE_NIC_BYPASS
1372 bypass_init_t bypass_init;
1373 bypass_state_set_t bypass_state_set;
1374 bypass_state_show_t bypass_state_show;
1375 bypass_event_set_t bypass_event_set;
1376 bypass_event_show_t bypass_event_show;
1377 bypass_wd_timeout_set_t bypass_wd_timeout_set;
1378 bypass_wd_timeout_show_t bypass_wd_timeout_show;
1379 bypass_ver_show_t bypass_ver_show;
1380 bypass_wd_reset_t bypass_wd_reset;
1383 /** Configure RSS hash protocols. */
1384 rss_hash_update_t rss_hash_update;
1385 /** Get current RSS hash configuration. */
1386 rss_hash_conf_get_t rss_hash_conf_get;
1387 eth_filter_ctrl_t filter_ctrl; /**< common filter control*/
1391 * Function type used for RX packet processing packet callbacks.
1393 * The callback function is called on RX with a burst of packets that have
1394 * been received on the given port and queue.
1397 * The Ethernet port on which RX is being performed.
1399 * The queue on the Ethernet port which is being used to receive the packets.
1401 * The burst of packets that have just been received.
1403 * The number of packets in the burst pointed to by "pkts".
1405 * The max number of packets that can be stored in the "pkts" array.
1407 * The arbitrary user parameter passed in by the application when the callback
1408 * was originally configured.
1410 * The number of packets returned to the user.
1412 typedef uint16_t (*rte_rx_callback_fn)(uint8_t port, uint16_t queue,
1413 struct rte_mbuf *pkts[], uint16_t nb_pkts, uint16_t max_pkts,
1417 * Function type used for TX packet processing packet callbacks.
1419 * The callback function is called on TX with a burst of packets immediately
1420 * before the packets are put onto the hardware queue for transmission.
1423 * The Ethernet port on which TX is being performed.
1425 * The queue on the Ethernet port which is being used to transmit the packets.
1427 * The burst of packets that are about to be transmitted.
1429 * The number of packets in the burst pointed to by "pkts".
1431 * The arbitrary user parameter passed in by the application when the callback
1432 * was originally configured.
1434 * The number of packets to be written to the NIC.
1436 typedef uint16_t (*rte_tx_callback_fn)(uint8_t port, uint16_t queue,
1437 struct rte_mbuf *pkts[], uint16_t nb_pkts, void *user_param);
1441 * Structure used to hold information about the callbacks to be called for a
1442 * queue on RX and TX.
1444 struct rte_eth_rxtx_callback {
1445 struct rte_eth_rxtx_callback *next;
1447 rte_rx_callback_fn rx;
1448 rte_tx_callback_fn tx;
1454 * The eth device type
1456 enum rte_eth_dev_type {
1457 RTE_ETH_DEV_UNKNOWN, /**< unknown device type */
1459 /**< Physical function and Virtual function of PCI devices */
1460 RTE_ETH_DEV_VIRTUAL, /**< non hardware device */
1461 RTE_ETH_DEV_MAX /**< max value of this enum */
1466 * The generic data structure associated with each ethernet device.
1468 * Pointers to burst-oriented packet receive and transmit functions are
1469 * located at the beginning of the structure, along with the pointer to
1470 * where all the data elements for the particular device are stored in shared
1471 * memory. This split allows the function pointer and driver data to be per-
1472 * process, while the actual configuration data for the device is shared.
1474 struct rte_eth_dev {
1475 eth_rx_burst_t rx_pkt_burst; /**< Pointer to PMD receive function. */
1476 eth_tx_burst_t tx_pkt_burst; /**< Pointer to PMD transmit function. */
1477 struct rte_eth_dev_data *data; /**< Pointer to device data */
1478 const struct eth_driver *driver;/**< Driver for this device */
1479 const struct eth_dev_ops *dev_ops; /**< Functions exported by PMD */
1480 struct rte_pci_device *pci_dev; /**< PCI info. supplied by probing */
1481 /** User application callbacks for NIC interrupts */
1482 struct rte_eth_dev_cb_list link_intr_cbs;
1484 * User-supplied functions called from rx_burst to post-process
1485 * received packets before passing them to the user
1487 struct rte_eth_rxtx_callback *post_rx_burst_cbs[RTE_MAX_QUEUES_PER_PORT];
1489 * User-supplied functions called from tx_burst to pre-process
1490 * received packets before passing them to the driver for transmission.
1492 struct rte_eth_rxtx_callback *pre_tx_burst_cbs[RTE_MAX_QUEUES_PER_PORT];
1493 uint8_t attached; /**< Flag indicating the port is attached */
1494 enum rte_eth_dev_type dev_type; /**< Flag indicating the device type */
1497 struct rte_eth_dev_sriov {
1498 uint8_t active; /**< SRIOV is active with 16, 32 or 64 pools */
1499 uint8_t nb_q_per_pool; /**< rx queue number per pool */
1500 uint16_t def_vmdq_idx; /**< Default pool num used for PF */
1501 uint16_t def_pool_q_idx; /**< Default pool queue start reg index */
1503 #define RTE_ETH_DEV_SRIOV(dev) ((dev)->data->sriov)
1505 #define RTE_ETH_NAME_MAX_LEN (32)
1509 * The data part, with no function pointers, associated with each ethernet device.
1511 * This structure is safe to place in shared memory to be common among different
1512 * processes in a multi-process configuration.
1514 struct rte_eth_dev_data {
1515 char name[RTE_ETH_NAME_MAX_LEN]; /**< Unique identifier name */
1517 void **rx_queues; /**< Array of pointers to RX queues. */
1518 void **tx_queues; /**< Array of pointers to TX queues. */
1519 uint16_t nb_rx_queues; /**< Number of RX queues. */
1520 uint16_t nb_tx_queues; /**< Number of TX queues. */
1522 struct rte_eth_dev_sriov sriov; /**< SRIOV data */
1524 void *dev_private; /**< PMD-specific private data */
1526 struct rte_eth_link dev_link;
1527 /**< Link-level information & status */
1529 struct rte_eth_conf dev_conf; /**< Configuration applied to device. */
1530 uint16_t mtu; /**< Maximum Transmission Unit. */
1532 uint32_t min_rx_buf_size;
1533 /**< Common rx buffer size handled by all queues */
1535 uint64_t rx_mbuf_alloc_failed; /**< RX ring mbuf allocation failures. */
1536 struct ether_addr* mac_addrs;/**< Device Ethernet Link address. */
1537 uint64_t mac_pool_sel[ETH_NUM_RECEIVE_MAC_ADDR];
1538 /** bitmap array of associating Ethernet MAC addresses to pools */
1539 struct ether_addr* hash_mac_addrs;
1540 /** Device Ethernet MAC addresses of hash filtering. */
1541 uint8_t port_id; /**< Device [external] port identifier. */
1542 uint8_t promiscuous : 1, /**< RX promiscuous mode ON(1) / OFF(0). */
1543 scattered_rx : 1, /**< RX of scattered packets is ON(1) / OFF(0) */
1544 lro : 1, /**< RX LRO is ON(1) / OFF(0) */
1545 all_multicast : 1, /**< RX all multicast mode ON(1) / OFF(0). */
1546 dev_started : 1; /**< Device state: STARTED(1) / STOPPED(0). */
1551 * The pool of *rte_eth_dev* structures. The size of the pool
1552 * is configured at compile-time in the <rte_ethdev.c> file.
1554 extern struct rte_eth_dev rte_eth_devices[];
1557 * Get the total number of Ethernet devices that have been successfully
1558 * initialized by the [matching] Ethernet driver during the PCI probing phase.
1559 * All devices whose port identifier is in the range
1560 * [0, rte_eth_dev_count() - 1] can be operated on by network applications
1561 * immediately after invoking rte_eal_init().
1562 * If the application unplugs a port using hotplug function, The enabled port
1563 * numbers may be noncontiguous. In the case, the applications need to manage
1564 * enabled port by themselves.
1567 * - The total number of usable Ethernet devices.
1569 extern uint8_t rte_eth_dev_count(void);
1572 * Function for internal use by port hotplug functions.
1573 * Returns a ethdev slot specified by the unique identifier name.
1575 * The pointer to the Unique identifier name for each Ethernet device
1577 * - The pointer to the ethdev slot, on success. NULL on error
1579 extern struct rte_eth_dev *rte_eth_dev_allocated(const char *name);
1582 * Function for internal use by dummy drivers primarily, e.g. ring-based
1584 * Allocates a new ethdev slot for an ethernet device and returns the pointer
1585 * to that slot for the driver to use.
1587 * @param name Unique identifier name for each Ethernet device
1588 * @param type Device type of this Ethernet device
1590 * - Slot in the rte_dev_devices array for a new device;
1592 struct rte_eth_dev *rte_eth_dev_allocate(const char *name,
1593 enum rte_eth_dev_type type);
1596 * Function for internal use by dummy drivers primarily, e.g. ring-based
1598 * Release the specified ethdev port.
1601 * The *eth_dev* pointer is the address of the *rte_eth_dev* structure.
1603 * - 0 on success, negative on error
1605 int rte_eth_dev_release_port(struct rte_eth_dev *eth_dev);
1608 * Attach a new Ethernet device specified by aruguments.
1611 * A pointer to a strings array describing the new device
1612 * to be attached. The strings should be a pci address like
1613 * '0000:01:00.0' or virtual device name like 'eth_pcap0'.
1615 * A pointer to a port identifier actually attached.
1617 * 0 on success and port_id is filled, negative on error
1619 int rte_eth_dev_attach(const char *devargs, uint8_t *port_id);
1622 * Detach a Ethernet device specified by port identifier.
1625 * The port identifier of the device to detach.
1627 * A pointer to a device name actually detached.
1629 * 0 on success and devname is filled, negative on error
1631 int rte_eth_dev_detach(uint8_t port_id, char *devname);
1636 * Initialization function of an Ethernet driver invoked for each matching
1637 * Ethernet PCI device detected during the PCI probing phase.
1640 * The *eth_dev* pointer is the address of the *rte_eth_dev* structure
1641 * associated with the matching device and which have been [automatically]
1642 * allocated in the *rte_eth_devices* array.
1643 * The *eth_dev* structure is supplied to the driver initialization function
1644 * with the following fields already initialized:
1646 * - *pci_dev*: Holds the pointers to the *rte_pci_device* structure which
1647 * contains the generic PCI information of the matching device.
1649 * - *driver*: Holds the pointer to the *eth_driver* structure.
1651 * - *dev_private*: Holds a pointer to the device private data structure.
1653 * - *mtu*: Contains the default Ethernet maximum frame length (1500).
1655 * - *port_id*: Contains the port index of the device (actually the index
1656 * of the *eth_dev* structure in the *rte_eth_devices* array).
1659 * - 0: Success, the device is properly initialized by the driver.
1660 * In particular, the driver MUST have set up the *dev_ops* pointer
1661 * of the *eth_dev* structure.
1662 * - <0: Error code of the device initialization failure.
1664 typedef int (*eth_dev_init_t)(struct rte_eth_dev *eth_dev);
1668 * Finalization function of an Ethernet driver invoked for each matching
1669 * Ethernet PCI device detected during the PCI closing phase.
1672 * The *eth_dev* pointer is the address of the *rte_eth_dev* structure
1673 * associated with the matching device and which have been [automatically]
1674 * allocated in the *rte_eth_devices* array.
1676 * - 0: Success, the device is properly finalized by the driver.
1677 * In particular, the driver MUST free the *dev_ops* pointer
1678 * of the *eth_dev* structure.
1679 * - <0: Error code of the device initialization failure.
1681 typedef int (*eth_dev_uninit_t)(struct rte_eth_dev *eth_dev);
1685 * The structure associated with a PMD Ethernet driver.
1687 * Each Ethernet driver acts as a PCI driver and is represented by a generic
1688 * *eth_driver* structure that holds:
1690 * - An *rte_pci_driver* structure (which must be the first field).
1692 * - The *eth_dev_init* function invoked for each matching PCI device.
1694 * - The *eth_dev_uninit* function invoked for each matching PCI device.
1696 * - The size of the private data to allocate for each matching device.
1699 struct rte_pci_driver pci_drv; /**< The PMD is also a PCI driver. */
1700 eth_dev_init_t eth_dev_init; /**< Device init function. */
1701 eth_dev_uninit_t eth_dev_uninit; /**< Device uninit function. */
1702 unsigned int dev_private_size; /**< Size of device private data. */
1707 * A function invoked by the initialization function of an Ethernet driver
1708 * to simultaneously register itself as a PCI driver and as an Ethernet
1709 * Poll Mode Driver (PMD).
1712 * The pointer to the *eth_driver* structure associated with
1713 * the Ethernet driver.
1715 extern void rte_eth_driver_register(struct eth_driver *eth_drv);
1718 * Configure an Ethernet device.
1719 * This function must be invoked first before any other function in the
1720 * Ethernet API. This function can also be re-invoked when a device is in the
1724 * The port identifier of the Ethernet device to configure.
1725 * @param nb_rx_queue
1726 * The number of receive queues to set up for the Ethernet device.
1727 * @param nb_tx_queue
1728 * The number of transmit queues to set up for the Ethernet device.
1730 * The pointer to the configuration data to be used for the Ethernet device.
1731 * The *rte_eth_conf* structure includes:
1732 * - the hardware offload features to activate, with dedicated fields for
1733 * each statically configurable offload hardware feature provided by
1734 * Ethernet devices, such as IP checksum or VLAN tag stripping for
1736 * - the Receive Side Scaling (RSS) configuration when using multiple RX
1739 * Embedding all configuration information in a single data structure
1740 * is the more flexible method that allows the addition of new features
1741 * without changing the syntax of the API.
1743 * - 0: Success, device configured.
1744 * - <0: Error code returned by the driver configuration function.
1746 extern int rte_eth_dev_configure(uint8_t port_id,
1747 uint16_t nb_rx_queue,
1748 uint16_t nb_tx_queue,
1749 const struct rte_eth_conf *eth_conf);
1752 * Allocate and set up a receive queue for an Ethernet device.
1754 * The function allocates a contiguous block of memory for *nb_rx_desc*
1755 * receive descriptors from a memory zone associated with *socket_id*
1756 * and initializes each receive descriptor with a network buffer allocated
1757 * from the memory pool *mb_pool*.
1760 * The port identifier of the Ethernet device.
1761 * @param rx_queue_id
1762 * The index of the receive queue to set up.
1763 * The value must be in the range [0, nb_rx_queue - 1] previously supplied
1764 * to rte_eth_dev_configure().
1766 * The number of receive descriptors to allocate for the receive ring.
1768 * The *socket_id* argument is the socket identifier in case of NUMA.
1769 * The value can be *SOCKET_ID_ANY* if there is no NUMA constraint for
1770 * the DMA memory allocated for the receive descriptors of the ring.
1772 * The pointer to the configuration data to be used for the receive queue.
1773 * NULL value is allowed, in which case default RX configuration
1775 * The *rx_conf* structure contains an *rx_thresh* structure with the values
1776 * of the Prefetch, Host, and Write-Back threshold registers of the receive
1779 * The pointer to the memory pool from which to allocate *rte_mbuf* network
1780 * memory buffers to populate each descriptor of the receive ring.
1782 * - 0: Success, receive queue correctly set up.
1783 * - -EINVAL: The size of network buffers which can be allocated from the
1784 * memory pool does not fit the various buffer sizes allowed by the
1785 * device controller.
1786 * - -ENOMEM: Unable to allocate the receive ring descriptors or to
1787 * allocate network memory buffers from the memory pool when
1788 * initializing receive descriptors.
1790 extern int rte_eth_rx_queue_setup(uint8_t port_id, uint16_t rx_queue_id,
1791 uint16_t nb_rx_desc, unsigned int socket_id,
1792 const struct rte_eth_rxconf *rx_conf,
1793 struct rte_mempool *mb_pool);
1796 * Allocate and set up a transmit queue for an Ethernet device.
1799 * The port identifier of the Ethernet device.
1800 * @param tx_queue_id
1801 * The index of the transmit queue to set up.
1802 * The value must be in the range [0, nb_tx_queue - 1] previously supplied
1803 * to rte_eth_dev_configure().
1805 * The number of transmit descriptors to allocate for the transmit ring.
1807 * The *socket_id* argument is the socket identifier in case of NUMA.
1808 * Its value can be *SOCKET_ID_ANY* if there is no NUMA constraint for
1809 * the DMA memory allocated for the transmit descriptors of the ring.
1811 * The pointer to the configuration data to be used for the transmit queue.
1812 * NULL value is allowed, in which case default RX configuration
1814 * The *tx_conf* structure contains the following data:
1815 * - The *tx_thresh* structure with the values of the Prefetch, Host, and
1816 * Write-Back threshold registers of the transmit ring.
1817 * When setting Write-Back threshold to the value greater then zero,
1818 * *tx_rs_thresh* value should be explicitly set to one.
1819 * - The *tx_free_thresh* value indicates the [minimum] number of network
1820 * buffers that must be pending in the transmit ring to trigger their
1821 * [implicit] freeing by the driver transmit function.
1822 * - The *tx_rs_thresh* value indicates the [minimum] number of transmit
1823 * descriptors that must be pending in the transmit ring before setting the
1824 * RS bit on a descriptor by the driver transmit function.
1825 * The *tx_rs_thresh* value should be less or equal then
1826 * *tx_free_thresh* value, and both of them should be less then
1828 * - The *txq_flags* member contains flags to pass to the TX queue setup
1829 * function to configure the behavior of the TX queue. This should be set
1830 * to 0 if no special configuration is required.
1832 * Note that setting *tx_free_thresh* or *tx_rs_thresh* value to 0 forces
1833 * the transmit function to use default values.
1835 * - 0: Success, the transmit queue is correctly set up.
1836 * - -ENOMEM: Unable to allocate the transmit ring descriptors.
1838 extern int rte_eth_tx_queue_setup(uint8_t port_id, uint16_t tx_queue_id,
1839 uint16_t nb_tx_desc, unsigned int socket_id,
1840 const struct rte_eth_txconf *tx_conf);
1843 * Return the NUMA socket to which an Ethernet device is connected
1846 * The port identifier of the Ethernet device
1848 * The NUMA socket id to which the Ethernet device is connected or
1849 * a default of zero if the socket could not be determined.
1850 * -1 is returned is the port_id value is out of range.
1852 extern int rte_eth_dev_socket_id(uint8_t port_id);
1855 * Allocate mbuf from mempool, setup the DMA physical address
1856 * and then start RX for specified queue of a port. It is used
1857 * when rx_deferred_start flag of the specified queue is true.
1860 * The port identifier of the Ethernet device
1861 * @param rx_queue_id
1862 * The index of the rx queue to update the ring.
1863 * The value must be in the range [0, nb_rx_queue - 1] previously supplied
1864 * to rte_eth_dev_configure().
1866 * - 0: Success, the transmit queue is correctly set up.
1867 * - -EINVAL: The port_id or the queue_id out of range.
1868 * - -ENOTSUP: The function not supported in PMD driver.
1870 extern int rte_eth_dev_rx_queue_start(uint8_t port_id, uint16_t rx_queue_id);
1873 * Stop specified RX queue of a port
1876 * The port identifier of the Ethernet device
1877 * @param rx_queue_id
1878 * The index of the rx queue to update the ring.
1879 * The value must be in the range [0, nb_rx_queue - 1] previously supplied
1880 * to rte_eth_dev_configure().
1882 * - 0: Success, the transmit queue is correctly set up.
1883 * - -EINVAL: The port_id or the queue_id out of range.
1884 * - -ENOTSUP: The function not supported in PMD driver.
1886 extern int rte_eth_dev_rx_queue_stop(uint8_t port_id, uint16_t rx_queue_id);
1889 * Start TX for specified queue of a port. It is used when tx_deferred_start
1890 * flag of the specified queue is true.
1893 * The port identifier of the Ethernet device
1894 * @param tx_queue_id
1895 * The index of the tx queue to update the ring.
1896 * The value must be in the range [0, nb_tx_queue - 1] previously supplied
1897 * to rte_eth_dev_configure().
1899 * - 0: Success, the transmit queue is correctly set up.
1900 * - -EINVAL: The port_id or the queue_id out of range.
1901 * - -ENOTSUP: The function not supported in PMD driver.
1903 extern int rte_eth_dev_tx_queue_start(uint8_t port_id, uint16_t tx_queue_id);
1906 * Stop specified TX queue of a port
1909 * The port identifier of the Ethernet device
1910 * @param tx_queue_id
1911 * The index of the tx queue to update the ring.
1912 * The value must be in the range [0, nb_tx_queue - 1] previously supplied
1913 * to rte_eth_dev_configure().
1915 * - 0: Success, the transmit queue is correctly set up.
1916 * - -EINVAL: The port_id or the queue_id out of range.
1917 * - -ENOTSUP: The function not supported in PMD driver.
1919 extern int rte_eth_dev_tx_queue_stop(uint8_t port_id, uint16_t tx_queue_id);
1924 * Start an Ethernet device.
1926 * The device start step is the last one and consists of setting the configured
1927 * offload features and in starting the transmit and the receive units of the
1929 * On success, all basic functions exported by the Ethernet API (link status,
1930 * receive/transmit, and so on) can be invoked.
1933 * The port identifier of the Ethernet device.
1935 * - 0: Success, Ethernet device started.
1936 * - <0: Error code of the driver device start function.
1938 extern int rte_eth_dev_start(uint8_t port_id);
1941 * Stop an Ethernet device. The device can be restarted with a call to
1942 * rte_eth_dev_start()
1945 * The port identifier of the Ethernet device.
1947 extern void rte_eth_dev_stop(uint8_t port_id);
1951 * Link up an Ethernet device.
1953 * Set device link up will re-enable the device rx/tx
1954 * functionality after it is previously set device linked down.
1957 * The port identifier of the Ethernet device.
1959 * - 0: Success, Ethernet device linked up.
1960 * - <0: Error code of the driver device link up function.
1962 extern int rte_eth_dev_set_link_up(uint8_t port_id);
1965 * Link down an Ethernet device.
1966 * The device rx/tx functionality will be disabled if success,
1967 * and it can be re-enabled with a call to
1968 * rte_eth_dev_set_link_up()
1971 * The port identifier of the Ethernet device.
1973 extern int rte_eth_dev_set_link_down(uint8_t port_id);
1976 * Close an Ethernet device. The device cannot be restarted!
1979 * The port identifier of the Ethernet device.
1981 extern void rte_eth_dev_close(uint8_t port_id);
1984 * Enable receipt in promiscuous mode for an Ethernet device.
1987 * The port identifier of the Ethernet device.
1989 extern void rte_eth_promiscuous_enable(uint8_t port_id);
1992 * Disable receipt in promiscuous mode for an Ethernet device.
1995 * The port identifier of the Ethernet device.
1997 extern void rte_eth_promiscuous_disable(uint8_t port_id);
2000 * Return the value of promiscuous mode for an Ethernet device.
2003 * The port identifier of the Ethernet device.
2005 * - (1) if promiscuous is enabled
2006 * - (0) if promiscuous is disabled.
2009 extern int rte_eth_promiscuous_get(uint8_t port_id);
2012 * Enable the receipt of any multicast frame by an Ethernet device.
2015 * The port identifier of the Ethernet device.
2017 extern void rte_eth_allmulticast_enable(uint8_t port_id);
2020 * Disable the receipt of all multicast frames by an Ethernet device.
2023 * The port identifier of the Ethernet device.
2025 extern void rte_eth_allmulticast_disable(uint8_t port_id);
2028 * Return the value of allmulticast mode for an Ethernet device.
2031 * The port identifier of the Ethernet device.
2033 * - (1) if allmulticast is enabled
2034 * - (0) if allmulticast is disabled.
2037 extern int rte_eth_allmulticast_get(uint8_t port_id);
2040 * Retrieve the status (ON/OFF), the speed (in Mbps) and the mode (HALF-DUPLEX
2041 * or FULL-DUPLEX) of the physical link of an Ethernet device. It might need
2042 * to wait up to 9 seconds in it.
2045 * The port identifier of the Ethernet device.
2047 * A pointer to an *rte_eth_link* structure to be filled with
2048 * the status, the speed and the mode of the Ethernet device link.
2050 extern void rte_eth_link_get(uint8_t port_id, struct rte_eth_link *link);
2053 * Retrieve the status (ON/OFF), the speed (in Mbps) and the mode (HALF-DUPLEX
2054 * or FULL-DUPLEX) of the physical link of an Ethernet device. It is a no-wait
2055 * version of rte_eth_link_get().
2058 * The port identifier of the Ethernet device.
2060 * A pointer to an *rte_eth_link* structure to be filled with
2061 * the status, the speed and the mode of the Ethernet device link.
2063 extern void rte_eth_link_get_nowait(uint8_t port_id,
2064 struct rte_eth_link *link);
2067 * Retrieve the general I/O statistics of an Ethernet device.
2070 * The port identifier of the Ethernet device.
2072 * A pointer to a structure of type *rte_eth_stats* to be filled with
2073 * the values of device counters for the following set of statistics:
2074 * - *ipackets* with the total of successfully received packets.
2075 * - *opackets* with the total of successfully transmitted packets.
2076 * - *ibytes* with the total of successfully received bytes.
2077 * - *obytes* with the total of successfully transmitted bytes.
2078 * - *ierrors* with the total of erroneous received packets.
2079 * - *oerrors* with the total of failed transmitted packets.
2081 * Zero if successful. Non-zero otherwise.
2083 extern int rte_eth_stats_get(uint8_t port_id, struct rte_eth_stats *stats);
2086 * Reset the general I/O statistics of an Ethernet device.
2089 * The port identifier of the Ethernet device.
2091 extern void rte_eth_stats_reset(uint8_t port_id);
2094 * Retrieve extended statistics of an Ethernet device.
2097 * The port identifier of the Ethernet device.
2099 * A pointer to a table of structure of type *rte_eth_xstats*
2100 * to be filled with device statistics names and values.
2101 * This parameter can be set to NULL if n is 0.
2103 * The size of the stats table, which should be large enough to store
2104 * all the statistics of the device.
2106 * - positive value lower or equal to n: success. The return value
2107 * is the number of entries filled in the stats table.
2108 * - positive value higher than n: error, the given statistics table
2109 * is too small. The return value corresponds to the size that should
2110 * be given to succeed. The entries in the table are not valid and
2111 * shall not be used by the caller.
2112 * - negative value on error (invalid port id)
2114 extern int rte_eth_xstats_get(uint8_t port_id,
2115 struct rte_eth_xstats *xstats, unsigned n);
2118 * Reset extended statistics of an Ethernet device.
2121 * The port identifier of the Ethernet device.
2123 extern void rte_eth_xstats_reset(uint8_t port_id);
2126 * Set a mapping for the specified transmit queue to the specified per-queue
2127 * statistics counter.
2130 * The port identifier of the Ethernet device.
2131 * @param tx_queue_id
2132 * The index of the transmit queue for which a queue stats mapping is required.
2133 * The value must be in the range [0, nb_tx_queue - 1] previously supplied
2134 * to rte_eth_dev_configure().
2136 * The per-queue packet statistics functionality number that the transmit
2137 * queue is to be assigned.
2138 * The value must be in the range [0, RTE_MAX_ETHPORT_QUEUE_STATS_MAPS - 1].
2140 * Zero if successful. Non-zero otherwise.
2142 extern int rte_eth_dev_set_tx_queue_stats_mapping(uint8_t port_id,
2143 uint16_t tx_queue_id,
2147 * Set a mapping for the specified receive queue to the specified per-queue
2148 * statistics counter.
2151 * The port identifier of the Ethernet device.
2152 * @param rx_queue_id
2153 * The index of the receive queue for which a queue stats mapping is required.
2154 * The value must be in the range [0, nb_rx_queue - 1] previously supplied
2155 * to rte_eth_dev_configure().
2157 * The per-queue packet statistics functionality number that the receive
2158 * queue is to be assigned.
2159 * The value must be in the range [0, RTE_MAX_ETHPORT_QUEUE_STATS_MAPS - 1].
2161 * Zero if successful. Non-zero otherwise.
2163 extern int rte_eth_dev_set_rx_queue_stats_mapping(uint8_t port_id,
2164 uint16_t rx_queue_id,
2168 * Retrieve the Ethernet address of an Ethernet device.
2171 * The port identifier of the Ethernet device.
2173 * A pointer to a structure of type *ether_addr* to be filled with
2174 * the Ethernet address of the Ethernet device.
2176 extern void rte_eth_macaddr_get(uint8_t port_id, struct ether_addr *mac_addr);
2179 * Retrieve the contextual information of an Ethernet device.
2182 * The port identifier of the Ethernet device.
2184 * A pointer to a structure of type *rte_eth_dev_info* to be filled with
2185 * the contextual information of the Ethernet device.
2187 extern void rte_eth_dev_info_get(uint8_t port_id,
2188 struct rte_eth_dev_info *dev_info);
2191 * Retrieve the MTU of an Ethernet device.
2194 * The port identifier of the Ethernet device.
2196 * A pointer to a uint16_t where the retrieved MTU is to be stored.
2198 * - (0) if successful.
2199 * - (-ENODEV) if *port_id* invalid.
2201 extern int rte_eth_dev_get_mtu(uint8_t port_id, uint16_t *mtu);
2204 * Change the MTU of an Ethernet device.
2207 * The port identifier of the Ethernet device.
2209 * A uint16_t for the MTU to be applied.
2211 * - (0) if successful.
2212 * - (-ENOTSUP) if operation is not supported.
2213 * - (-ENODEV) if *port_id* invalid.
2214 * - (-EINVAL) if *mtu* invalid.
2216 extern int rte_eth_dev_set_mtu(uint8_t port_id, uint16_t mtu);
2219 * Enable/Disable hardware filtering by an Ethernet device of received
2220 * VLAN packets tagged with a given VLAN Tag Identifier.
2223 * The port identifier of the Ethernet device.
2225 * The VLAN Tag Identifier whose filtering must be enabled or disabled.
2227 * If > 0, enable VLAN filtering of VLAN packets tagged with *vlan_id*.
2228 * Otherwise, disable VLAN filtering of VLAN packets tagged with *vlan_id*.
2230 * - (0) if successful.
2231 * - (-ENOSUP) if hardware-assisted VLAN filtering not configured.
2232 * - (-ENODEV) if *port_id* invalid.
2233 * - (-ENOSYS) if VLAN filtering on *port_id* disabled.
2234 * - (-EINVAL) if *vlan_id* > 4095.
2236 extern int rte_eth_dev_vlan_filter(uint8_t port_id, uint16_t vlan_id , int on);
2239 * Enable/Disable hardware VLAN Strip by a rx queue of an Ethernet device.
2240 * 82599/X540/X550 can support VLAN stripping at the rx queue level
2243 * The port identifier of the Ethernet device.
2244 * @param rx_queue_id
2245 * The index of the receive queue for which a queue stats mapping is required.
2246 * The value must be in the range [0, nb_rx_queue - 1] previously supplied
2247 * to rte_eth_dev_configure().
2249 * If 1, Enable VLAN Stripping of the receive queue of the Ethernet port.
2250 * If 0, Disable VLAN Stripping of the receive queue of the Ethernet port.
2252 * - (0) if successful.
2253 * - (-ENOSUP) if hardware-assisted VLAN stripping not configured.
2254 * - (-ENODEV) if *port_id* invalid.
2255 * - (-EINVAL) if *rx_queue_id* invalid.
2257 extern int rte_eth_dev_set_vlan_strip_on_queue(uint8_t port_id,
2258 uint16_t rx_queue_id, int on);
2261 * Set the Outer VLAN Ether Type by an Ethernet device, it can be inserted to
2262 * the VLAN Header. This is a register setup available on some Intel NIC, not
2263 * but all, please check the data sheet for availability.
2266 * The port identifier of the Ethernet device.
2268 * The Tag Protocol ID
2270 * - (0) if successful.
2271 * - (-ENOSUP) if hardware-assisted VLAN TPID setup is not supported.
2272 * - (-ENODEV) if *port_id* invalid.
2274 extern int rte_eth_dev_set_vlan_ether_type(uint8_t port_id, uint16_t tag_type);
2277 * Set VLAN offload configuration on an Ethernet device
2278 * Enable/Disable Extended VLAN by an Ethernet device, This is a register setup
2279 * available on some Intel NIC, not but all, please check the data sheet for
2281 * Enable/Disable VLAN Strip can be done on rx queue for certain NIC, but here
2282 * the configuration is applied on the port level.
2285 * The port identifier of the Ethernet device.
2286 * @param offload_mask
2287 * The VLAN Offload bit mask can be mixed use with "OR"
2288 * ETH_VLAN_STRIP_OFFLOAD
2289 * ETH_VLAN_FILTER_OFFLOAD
2290 * ETH_VLAN_EXTEND_OFFLOAD
2292 * - (0) if successful.
2293 * - (-ENOSUP) if hardware-assisted VLAN filtering not configured.
2294 * - (-ENODEV) if *port_id* invalid.
2296 extern int rte_eth_dev_set_vlan_offload(uint8_t port_id, int offload_mask);
2299 * Read VLAN Offload configuration from an Ethernet device
2302 * The port identifier of the Ethernet device.
2304 * - (>0) if successful. Bit mask to indicate
2305 * ETH_VLAN_STRIP_OFFLOAD
2306 * ETH_VLAN_FILTER_OFFLOAD
2307 * ETH_VLAN_EXTEND_OFFLOAD
2308 * - (-ENODEV) if *port_id* invalid.
2310 extern int rte_eth_dev_get_vlan_offload(uint8_t port_id);
2313 * Set port based TX VLAN insersion on or off.
2316 * The port identifier of the Ethernet device.
2318 * Port based TX VLAN identifier togeth with user priority.
2320 * Turn on or off the port based TX VLAN insertion.
2323 * - (0) if successful.
2324 * - negative if failed.
2326 extern int rte_eth_dev_set_vlan_pvid(uint8_t port_id, uint16_t pvid, int on);
2330 * Retrieve a burst of input packets from a receive queue of an Ethernet
2331 * device. The retrieved packets are stored in *rte_mbuf* structures whose
2332 * pointers are supplied in the *rx_pkts* array.
2334 * The rte_eth_rx_burst() function loops, parsing the RX ring of the
2335 * receive queue, up to *nb_pkts* packets, and for each completed RX
2336 * descriptor in the ring, it performs the following operations:
2338 * - Initialize the *rte_mbuf* data structure associated with the
2339 * RX descriptor according to the information provided by the NIC into
2340 * that RX descriptor.
2342 * - Store the *rte_mbuf* data structure into the next entry of the
2345 * - Replenish the RX descriptor with a new *rte_mbuf* buffer
2346 * allocated from the memory pool associated with the receive queue at
2347 * initialization time.
2349 * When retrieving an input packet that was scattered by the controller
2350 * into multiple receive descriptors, the rte_eth_rx_burst() function
2351 * appends the associated *rte_mbuf* buffers to the first buffer of the
2354 * The rte_eth_rx_burst() function returns the number of packets
2355 * actually retrieved, which is the number of *rte_mbuf* data structures
2356 * effectively supplied into the *rx_pkts* array.
2357 * A return value equal to *nb_pkts* indicates that the RX queue contained
2358 * at least *rx_pkts* packets, and this is likely to signify that other
2359 * received packets remain in the input queue. Applications implementing
2360 * a "retrieve as much received packets as possible" policy can check this
2361 * specific case and keep invoking the rte_eth_rx_burst() function until
2362 * a value less than *nb_pkts* is returned.
2364 * This receive method has the following advantages:
2366 * - It allows a run-to-completion network stack engine to retrieve and
2367 * to immediately process received packets in a fast burst-oriented
2368 * approach, avoiding the overhead of unnecessary intermediate packet
2369 * queue/dequeue operations.
2371 * - Conversely, it also allows an asynchronous-oriented processing
2372 * method to retrieve bursts of received packets and to immediately
2373 * queue them for further parallel processing by another logical core,
2374 * for instance. However, instead of having received packets being
2375 * individually queued by the driver, this approach allows the invoker
2376 * of the rte_eth_rx_burst() function to queue a burst of retrieved
2377 * packets at a time and therefore dramatically reduce the cost of
2378 * enqueue/dequeue operations per packet.
2380 * - It allows the rte_eth_rx_burst() function of the driver to take
2381 * advantage of burst-oriented hardware features (CPU cache,
2382 * prefetch instructions, and so on) to minimize the number of CPU
2383 * cycles per packet.
2385 * To summarize, the proposed receive API enables many
2386 * burst-oriented optimizations in both synchronous and asynchronous
2387 * packet processing environments with no overhead in both cases.
2389 * The rte_eth_rx_burst() function does not provide any error
2390 * notification to avoid the corresponding overhead. As a hint, the
2391 * upper-level application might check the status of the device link once
2392 * being systematically returned a 0 value for a given number of tries.
2395 * The port identifier of the Ethernet device.
2397 * The index of the receive queue from which to retrieve input packets.
2398 * The value must be in the range [0, nb_rx_queue - 1] previously supplied
2399 * to rte_eth_dev_configure().
2401 * The address of an array of pointers to *rte_mbuf* structures that
2402 * must be large enough to store *nb_pkts* pointers in it.
2404 * The maximum number of packets to retrieve.
2406 * The number of packets actually retrieved, which is the number
2407 * of pointers to *rte_mbuf* structures effectively supplied to the
2410 #ifdef RTE_LIBRTE_ETHDEV_DEBUG
2411 extern uint16_t rte_eth_rx_burst(uint8_t port_id, uint16_t queue_id,
2412 struct rte_mbuf **rx_pkts, uint16_t nb_pkts);
2414 static inline uint16_t
2415 rte_eth_rx_burst(uint8_t port_id, uint16_t queue_id,
2416 struct rte_mbuf **rx_pkts, const uint16_t nb_pkts)
2418 struct rte_eth_dev *dev;
2420 dev = &rte_eth_devices[port_id];
2422 int16_t nb_rx = (*dev->rx_pkt_burst)(dev->data->rx_queues[queue_id],
2425 #ifdef RTE_ETHDEV_RXTX_CALLBACKS
2426 struct rte_eth_rxtx_callback *cb = dev->post_rx_burst_cbs[queue_id];
2428 if (unlikely(cb != NULL)) {
2430 nb_rx = cb->fn.rx(port_id, queue_id, rx_pkts, nb_rx,
2431 nb_pkts, cb->param);
2433 } while (cb != NULL);
2442 * Get the number of used descriptors in a specific queue
2445 * The port identifier of the Ethernet device.
2447 * The queue id on the specific port.
2449 * The number of used descriptors in the specific queue.
2451 #ifdef RTE_LIBRTE_ETHDEV_DEBUG
2452 extern uint32_t rte_eth_rx_queue_count(uint8_t port_id, uint16_t queue_id);
2454 static inline uint32_t
2455 rte_eth_rx_queue_count(uint8_t port_id, uint16_t queue_id)
2457 struct rte_eth_dev *dev;
2459 dev = &rte_eth_devices[port_id];
2460 return (*dev->dev_ops->rx_queue_count)(dev, queue_id);
2465 * Check if the DD bit of the specific RX descriptor in the queue has been set
2468 * The port identifier of the Ethernet device.
2470 * The queue id on the specific port.
2472 * The offset of the descriptor ID from tail.
2474 * - (1) if the specific DD bit is set.
2475 * - (0) if the specific DD bit is not set.
2476 * - (-ENODEV) if *port_id* invalid.
2478 #ifdef RTE_LIBRTE_ETHDEV_DEBUG
2479 extern int rte_eth_rx_descriptor_done(uint8_t port_id,
2484 rte_eth_rx_descriptor_done(uint8_t port_id, uint16_t queue_id, uint16_t offset)
2486 struct rte_eth_dev *dev;
2488 dev = &rte_eth_devices[port_id];
2489 return (*dev->dev_ops->rx_descriptor_done)( \
2490 dev->data->rx_queues[queue_id], offset);
2495 * Send a burst of output packets on a transmit queue of an Ethernet device.
2497 * The rte_eth_tx_burst() function is invoked to transmit output packets
2498 * on the output queue *queue_id* of the Ethernet device designated by its
2500 * The *nb_pkts* parameter is the number of packets to send which are
2501 * supplied in the *tx_pkts* array of *rte_mbuf* structures.
2502 * The rte_eth_tx_burst() function loops, sending *nb_pkts* packets,
2503 * up to the number of transmit descriptors available in the TX ring of the
2505 * For each packet to send, the rte_eth_tx_burst() function performs
2506 * the following operations:
2508 * - Pick up the next available descriptor in the transmit ring.
2510 * - Free the network buffer previously sent with that descriptor, if any.
2512 * - Initialize the transmit descriptor with the information provided
2513 * in the *rte_mbuf data structure.
2515 * In the case of a segmented packet composed of a list of *rte_mbuf* buffers,
2516 * the rte_eth_tx_burst() function uses several transmit descriptors
2519 * The rte_eth_tx_burst() function returns the number of packets it
2520 * actually sent. A return value equal to *nb_pkts* means that all packets
2521 * have been sent, and this is likely to signify that other output packets
2522 * could be immediately transmitted again. Applications that implement a
2523 * "send as many packets to transmit as possible" policy can check this
2524 * specific case and keep invoking the rte_eth_tx_burst() function until
2525 * a value less than *nb_pkts* is returned.
2527 * It is the responsibility of the rte_eth_tx_burst() function to
2528 * transparently free the memory buffers of packets previously sent.
2529 * This feature is driven by the *tx_free_thresh* value supplied to the
2530 * rte_eth_dev_configure() function at device configuration time.
2531 * When the number of previously sent packets reached the "minimum transmit
2532 * packets to free" threshold, the rte_eth_tx_burst() function must
2533 * [attempt to] free the *rte_mbuf* buffers of those packets whose
2534 * transmission was effectively completed.
2537 * The port identifier of the Ethernet device.
2539 * The index of the transmit queue through which output packets must be
2541 * The value must be in the range [0, nb_tx_queue - 1] previously supplied
2542 * to rte_eth_dev_configure().
2544 * The address of an array of *nb_pkts* pointers to *rte_mbuf* structures
2545 * which contain the output packets.
2547 * The maximum number of packets to transmit.
2549 * The number of output packets actually stored in transmit descriptors of
2550 * the transmit ring. The return value can be less than the value of the
2551 * *tx_pkts* parameter when the transmit ring is full or has been filled up.
2553 #ifdef RTE_LIBRTE_ETHDEV_DEBUG
2554 extern uint16_t rte_eth_tx_burst(uint8_t port_id, uint16_t queue_id,
2555 struct rte_mbuf **tx_pkts, uint16_t nb_pkts);
2557 static inline uint16_t
2558 rte_eth_tx_burst(uint8_t port_id, uint16_t queue_id,
2559 struct rte_mbuf **tx_pkts, uint16_t nb_pkts)
2561 struct rte_eth_dev *dev;
2563 dev = &rte_eth_devices[port_id];
2565 #ifdef RTE_ETHDEV_RXTX_CALLBACKS
2566 struct rte_eth_rxtx_callback *cb = dev->pre_tx_burst_cbs[queue_id];
2568 if (unlikely(cb != NULL)) {
2570 nb_pkts = cb->fn.tx(port_id, queue_id, tx_pkts, nb_pkts,
2573 } while (cb != NULL);
2577 return (*dev->tx_pkt_burst)(dev->data->tx_queues[queue_id], tx_pkts, nb_pkts);
2582 * Setup a new signature filter rule on an Ethernet device
2585 * The port identifier of the Ethernet device.
2586 * @param fdir_filter
2587 * The pointer to the fdir filter structure describing the signature filter
2589 * The *rte_fdir_filter* structure includes the values of the different fields
2590 * to match: source and destination IP addresses, vlan id, flexbytes, source
2591 * and destination ports, and so on.
2593 * The index of the RX queue where to store RX packets matching the added
2594 * signature filter defined in fdir_filter.
2596 * - (0) if successful.
2597 * - (-ENOTSUP) if hardware doesn't support flow director mode.
2598 * - (-ENODEV) if *port_id* invalid.
2599 * - (-ENOSYS) if the FDIR mode is not configured in signature mode
2601 * - (-EINVAL) if the fdir_filter information is not correct.
2603 int rte_eth_dev_fdir_add_signature_filter(uint8_t port_id,
2604 struct rte_fdir_filter *fdir_filter,
2608 * Update a signature filter rule on an Ethernet device.
2609 * If the rule doesn't exits, it is created.
2612 * The port identifier of the Ethernet device.
2614 * The pointer to the structure describing the signature filter rule.
2615 * The *rte_fdir_filter* structure includes the values of the different fields
2616 * to match: source and destination IP addresses, vlan id, flexbytes, source
2617 * and destination ports, and so on.
2619 * The index of the RX queue where to store RX packets matching the added
2620 * signature filter defined in fdir_ftr.
2622 * - (0) if successful.
2623 * - (-ENOTSUP) if hardware doesn't support flow director mode.
2624 * - (-ENODEV) if *port_id* invalid.
2625 * - (-ENOSYS) if the flow director mode is not configured in signature mode
2627 * - (-EINVAL) if the fdir_filter information is not correct.
2629 int rte_eth_dev_fdir_update_signature_filter(uint8_t port_id,
2630 struct rte_fdir_filter *fdir_ftr,
2634 * Remove a signature filter rule on an Ethernet device.
2637 * The port identifier of the Ethernet device.
2639 * The pointer to the structure describing the signature filter rule.
2640 * The *rte_fdir_filter* structure includes the values of the different fields
2641 * to match: source and destination IP addresses, vlan id, flexbytes, source
2642 * and destination ports, and so on.
2644 * - (0) if successful.
2645 * - (-ENOTSUP) if hardware doesn't support flow director mode.
2646 * - (-ENODEV) if *port_id* invalid.
2647 * - (-ENOSYS) if the flow director mode is not configured in signature mode
2649 * - (-EINVAL) if the fdir_filter information is not correct.
2651 int rte_eth_dev_fdir_remove_signature_filter(uint8_t port_id,
2652 struct rte_fdir_filter *fdir_ftr);
2655 * Retrieve the flow director information of an Ethernet device.
2658 * The port identifier of the Ethernet device.
2660 * A pointer to a structure of type *rte_eth_dev_fdir* to be filled with
2661 * the flow director information of the Ethernet device.
2663 * - (0) if successful.
2664 * - (-ENOTSUP) if hardware doesn't support flow director mode.
2665 * - (-ENODEV) if *port_id* invalid.
2666 * - (-ENOSYS) if the flow director mode is not configured on *port_id*.
2668 int rte_eth_dev_fdir_get_infos(uint8_t port_id, struct rte_eth_fdir *fdir);
2671 * Add a new perfect filter rule on an Ethernet device.
2674 * The port identifier of the Ethernet device.
2675 * @param fdir_filter
2676 * The pointer to the structure describing the perfect filter rule.
2677 * The *rte_fdir_filter* structure includes the values of the different fields
2678 * to match: source and destination IP addresses, vlan id, flexbytes, source
2679 * and destination ports, and so on.
2680 * IPv6 are not supported.
2682 * The 16-bit value supplied in the field hash.fdir.id of mbuf for RX
2683 * packets matching the perfect filter.
2685 * The index of the RX queue where to store RX packets matching the added
2686 * perfect filter defined in fdir_filter.
2688 * If drop is set to 1, matching RX packets are stored into the RX drop
2689 * queue defined in the rte_fdir_conf.
2691 * - (0) if successful.
2692 * - (-ENOTSUP) if hardware doesn't support flow director mode.
2693 * - (-ENODEV) if *port_id* invalid.
2694 * - (-ENOSYS) if the flow director mode is not configured in perfect mode
2696 * - (-EINVAL) if the fdir_filter information is not correct.
2698 int rte_eth_dev_fdir_add_perfect_filter(uint8_t port_id,
2699 struct rte_fdir_filter *fdir_filter,
2700 uint16_t soft_id, uint8_t rx_queue,
2704 * Update a perfect filter rule on an Ethernet device.
2705 * If the rule doesn't exits, it is created.
2708 * The port identifier of the Ethernet device.
2709 * @param fdir_filter
2710 * The pointer to the structure describing the perfect filter rule.
2711 * The *rte_fdir_filter* structure includes the values of the different fields
2712 * to match: source and destination IP addresses, vlan id, flexbytes, source
2713 * and destination ports, and so on.
2714 * IPv6 are not supported.
2716 * The 16-bit value supplied in the field hash.fdir.id of mbuf for RX
2717 * packets matching the perfect filter.
2719 * The index of the RX queue where to store RX packets matching the added
2720 * perfect filter defined in fdir_filter.
2722 * If drop is set to 1, matching RX packets are stored into the RX drop
2723 * queue defined in the rte_fdir_conf.
2725 * - (0) if successful.
2726 * - (-ENOTSUP) if hardware doesn't support flow director mode.
2727 * - (-ENODEV) if *port_id* invalid.
2728 * - (-ENOSYS) if the flow director mode is not configured in perfect mode
2730 * - (-EINVAL) if the fdir_filter information is not correct.
2732 int rte_eth_dev_fdir_update_perfect_filter(uint8_t port_id,
2733 struct rte_fdir_filter *fdir_filter,
2734 uint16_t soft_id, uint8_t rx_queue,
2738 * Remove a perfect filter rule on an Ethernet device.
2741 * The port identifier of the Ethernet device.
2742 * @param fdir_filter
2743 * The pointer to the structure describing the perfect filter rule.
2744 * The *rte_fdir_filter* structure includes the values of the different fields
2745 * to match: source and destination IP addresses, vlan id, flexbytes, source
2746 * and destination ports, and so on.
2747 * IPv6 are not supported.
2749 * The soft_id value provided when adding/updating the removed filter.
2751 * - (0) if successful.
2752 * - (-ENOTSUP) if hardware doesn't support flow director mode.
2753 * - (-ENODEV) if *port_id* invalid.
2754 * - (-ENOSYS) if the flow director mode is not configured in perfect mode
2756 * - (-EINVAL) if the fdir_filter information is not correct.
2758 int rte_eth_dev_fdir_remove_perfect_filter(uint8_t port_id,
2759 struct rte_fdir_filter *fdir_filter,
2762 * Configure globally the masks for flow director mode for an Ethernet device.
2763 * For example, the device can match packets with only the first 24 bits of
2764 * the IPv4 source address.
2766 * The following fields can be masked: IPv4 addresses and L4 port numbers.
2767 * The following fields can be either enabled or disabled completely for the
2768 * matching functionality: VLAN ID tag; VLAN Priority + CFI bit; Flexible 2-byte
2770 * IPv6 masks are not supported.
2772 * All filters must comply with the masks previously configured.
2773 * For example, with a mask equal to 255.255.255.0 for the source IPv4 address,
2774 * all IPv4 filters must be created with a source IPv4 address that fits the
2777 * This function flushes all filters that have been previously added in
2781 * The port identifier of the Ethernet device.
2783 * The pointer to the fdir mask structure describing relevant headers fields
2784 * and relevant bits to use when matching packets addresses and ports.
2785 * IPv6 masks are not supported.
2787 * - (0) if successful.
2788 * - (-ENOTSUP) if hardware doesn't support flow director mode.
2789 * - (-ENODEV) if *port_id* invalid.
2790 * - (-ENOSYS) if the flow director mode is not configured in perfect
2791 * mode on *port_id*.
2792 * - (-EINVAL) if the fdir_filter information is not correct
2794 int rte_eth_dev_fdir_set_masks(uint8_t port_id,
2795 struct rte_fdir_masks *fdir_mask);
2798 * The eth device event type for interrupt, and maybe others in the future.
2800 enum rte_eth_event_type {
2801 RTE_ETH_EVENT_UNKNOWN, /**< unknown event type */
2802 RTE_ETH_EVENT_INTR_LSC, /**< lsc interrupt event */
2803 RTE_ETH_EVENT_MAX /**< max value of this enum */
2806 typedef void (*rte_eth_dev_cb_fn)(uint8_t port_id, \
2807 enum rte_eth_event_type event, void *cb_arg);
2808 /**< user application callback to be registered for interrupts */
2813 * Register a callback function for specific port id.
2820 * User supplied callback function to be called.
2822 * Pointer to the parameters for the registered callback.
2825 * - On success, zero.
2826 * - On failure, a negative value.
2828 int rte_eth_dev_callback_register(uint8_t port_id,
2829 enum rte_eth_event_type event,
2830 rte_eth_dev_cb_fn cb_fn, void *cb_arg);
2833 * Unregister a callback function for specific port id.
2840 * User supplied callback function to be called.
2842 * Pointer to the parameters for the registered callback. -1 means to
2843 * remove all for the same callback address and same event.
2846 * - On success, zero.
2847 * - On failure, a negative value.
2849 int rte_eth_dev_callback_unregister(uint8_t port_id,
2850 enum rte_eth_event_type event,
2851 rte_eth_dev_cb_fn cb_fn, void *cb_arg);
2854 * @internal Executes all the user application registered callbacks for
2855 * the specific device. It is for DPDK internal user only. User
2856 * application should not call it directly.
2859 * Pointer to struct rte_eth_dev.
2861 * Eth device interrupt event type.
2866 void _rte_eth_dev_callback_process(struct rte_eth_dev *dev,
2867 enum rte_eth_event_type event);
2870 * Turn on the LED on the Ethernet device.
2871 * This function turns on the LED on the Ethernet device.
2874 * The port identifier of the Ethernet device.
2876 * - (0) if successful.
2877 * - (-ENOTSUP) if underlying hardware OR driver doesn't support
2879 * - (-ENODEV) if *port_id* invalid.
2881 int rte_eth_led_on(uint8_t port_id);
2884 * Turn off the LED on the Ethernet device.
2885 * This function turns off the LED on the Ethernet device.
2888 * The port identifier of the Ethernet device.
2890 * - (0) if successful.
2891 * - (-ENOTSUP) if underlying hardware OR driver doesn't support
2893 * - (-ENODEV) if *port_id* invalid.
2895 int rte_eth_led_off(uint8_t port_id);
2898 * Get current status of the Ethernet link flow control for Ethernet device
2901 * The port identifier of the Ethernet device.
2903 * The pointer to the structure where to store the flow control parameters.
2905 * - (0) if successful.
2906 * - (-ENOTSUP) if hardware doesn't support flow control.
2907 * - (-ENODEV) if *port_id* invalid.
2909 int rte_eth_dev_flow_ctrl_get(uint8_t port_id,
2910 struct rte_eth_fc_conf *fc_conf);
2913 * Configure the Ethernet link flow control for Ethernet device
2916 * The port identifier of the Ethernet device.
2918 * The pointer to the structure of the flow control parameters.
2920 * - (0) if successful.
2921 * - (-ENOTSUP) if hardware doesn't support flow control mode.
2922 * - (-ENODEV) if *port_id* invalid.
2923 * - (-EINVAL) if bad parameter
2924 * - (-EIO) if flow control setup failure
2926 int rte_eth_dev_flow_ctrl_set(uint8_t port_id,
2927 struct rte_eth_fc_conf *fc_conf);
2930 * Configure the Ethernet priority flow control under DCB environment
2931 * for Ethernet device.
2934 * The port identifier of the Ethernet device.
2936 * The pointer to the structure of the priority flow control parameters.
2938 * - (0) if successful.
2939 * - (-ENOTSUP) if hardware doesn't support priority flow control mode.
2940 * - (-ENODEV) if *port_id* invalid.
2941 * - (-EINVAL) if bad parameter
2942 * - (-EIO) if flow control setup failure
2944 int rte_eth_dev_priority_flow_ctrl_set(uint8_t port_id,
2945 struct rte_eth_pfc_conf *pfc_conf);
2948 * Add a MAC address to an internal array of addresses used to enable whitelist
2949 * filtering to accept packets only if the destination MAC address matches.
2952 * The port identifier of the Ethernet device.
2954 * The MAC address to add.
2956 * VMDq pool index to associate address with (if VMDq is enabled). If VMDq is
2957 * not enabled, this should be set to 0.
2959 * - (0) if successfully added or *mac_addr" was already added.
2960 * - (-ENOTSUP) if hardware doesn't support this feature.
2961 * - (-ENODEV) if *port* is invalid.
2962 * - (-ENOSPC) if no more MAC addresses can be added.
2963 * - (-EINVAL) if MAC address is invalid.
2965 int rte_eth_dev_mac_addr_add(uint8_t port, struct ether_addr *mac_addr,
2969 * Remove a MAC address from the internal array of addresses.
2972 * The port identifier of the Ethernet device.
2974 * MAC address to remove.
2976 * - (0) if successful, or *mac_addr* didn't exist.
2977 * - (-ENOTSUP) if hardware doesn't support.
2978 * - (-ENODEV) if *port* invalid.
2979 * - (-EADDRINUSE) if attempting to remove the default MAC address
2981 int rte_eth_dev_mac_addr_remove(uint8_t port, struct ether_addr *mac_addr);
2984 * Update Redirection Table(RETA) of Receive Side Scaling of Ethernet device.
2987 * The port identifier of the Ethernet device.
2991 * Redirection table size. The table size can be queried by
2992 * rte_eth_dev_info_get().
2994 * - (0) if successful.
2995 * - (-ENOTSUP) if hardware doesn't support.
2996 * - (-EINVAL) if bad parameter.
2998 int rte_eth_dev_rss_reta_update(uint8_t port,
2999 struct rte_eth_rss_reta_entry64 *reta_conf,
3000 uint16_t reta_size);
3003 * Query Redirection Table(RETA) of Receive Side Scaling of Ethernet device.
3006 * The port identifier of the Ethernet device.
3010 * Redirection table size. The table size can be queried by
3011 * rte_eth_dev_info_get().
3013 * - (0) if successful.
3014 * - (-ENOTSUP) if hardware doesn't support.
3015 * - (-EINVAL) if bad parameter.
3017 int rte_eth_dev_rss_reta_query(uint8_t port,
3018 struct rte_eth_rss_reta_entry64 *reta_conf,
3019 uint16_t reta_size);
3022 * Updates unicast hash table for receiving packet with the given destination
3023 * MAC address, and the packet is routed to all VFs for which the RX mode is
3024 * accept packets that match the unicast hash table.
3027 * The port identifier of the Ethernet device.
3029 * Unicast MAC address.
3031 * 1 - Set an unicast hash bit for receiving packets with the MAC address.
3032 * 0 - Clear an unicast hash bit.
3034 * - (0) if successful.
3035 * - (-ENOTSUP) if hardware doesn't support.
3036 * - (-ENODEV) if *port_id* invalid.
3037 * - (-EINVAL) if bad parameter.
3039 int rte_eth_dev_uc_hash_table_set(uint8_t port,struct ether_addr *addr,
3043 * Updates all unicast hash bitmaps for receiving packet with any Unicast
3044 * Ethernet MAC addresses,the packet is routed to all VFs for which the RX
3045 * mode is accept packets that match the unicast hash table.
3048 * The port identifier of the Ethernet device.
3050 * 1 - Set all unicast hash bitmaps for receiving all the Ethernet
3052 * 0 - Clear all unicast hash bitmaps
3054 * - (0) if successful.
3055 * - (-ENOTSUP) if hardware doesn't support.
3056 * - (-ENODEV) if *port_id* invalid.
3057 * - (-EINVAL) if bad parameter.
3059 int rte_eth_dev_uc_all_hash_table_set(uint8_t port,uint8_t on);
3062 * Set RX L2 Filtering mode of a VF of an Ethernet device.
3065 * The port identifier of the Ethernet device.
3069 * The RX mode mask, which is one or more of accepting Untagged Packets,
3070 * packets that match the PFUTA table, Broadcast and Multicast Promiscuous.
3071 * ETH_VMDQ_ACCEPT_UNTAG,ETH_VMDQ_ACCEPT_HASH_UC,
3072 * ETH_VMDQ_ACCEPT_BROADCAST and ETH_VMDQ_ACCEPT_MULTICAST will be used
3075 * 1 - Enable a VF RX mode.
3076 * 0 - Disable a VF RX mode.
3078 * - (0) if successful.
3079 * - (-ENOTSUP) if hardware doesn't support.
3080 * - (-ENOTSUP) if hardware doesn't support.
3081 * - (-EINVAL) if bad parameter.
3083 int rte_eth_dev_set_vf_rxmode(uint8_t port, uint16_t vf, uint16_t rx_mode,
3087 * Enable or disable a VF traffic transmit of the Ethernet device.
3090 * The port identifier of the Ethernet device.
3094 * 1 - Enable a VF traffic transmit.
3095 * 0 - Disable a VF traffic transmit.
3097 * - (0) if successful.
3098 * - (-ENODEV) if *port_id* invalid.
3099 * - (-ENOTSUP) if hardware doesn't support.
3100 * - (-EINVAL) if bad parameter.
3103 rte_eth_dev_set_vf_tx(uint8_t port,uint16_t vf, uint8_t on);
3106 * Enable or disable a VF traffic receive of an Ethernet device.
3109 * The port identifier of the Ethernet device.
3113 * 1 - Enable a VF traffic receive.
3114 * 0 - Disable a VF traffic receive.
3116 * - (0) if successful.
3117 * - (-ENOTSUP) if hardware doesn't support.
3118 * - (-ENODEV) if *port_id* invalid.
3119 * - (-EINVAL) if bad parameter.
3122 rte_eth_dev_set_vf_rx(uint8_t port,uint16_t vf, uint8_t on);
3125 * Enable/Disable hardware VF VLAN filtering by an Ethernet device of
3126 * received VLAN packets tagged with a given VLAN Tag Identifier.
3129 * The port identifier of the Ethernet device.
3131 * The VLAN Tag Identifier whose filtering must be enabled or disabled.
3133 * Bitmap listing which VFs participate in the VLAN filtering.
3135 * 1 - Enable VFs VLAN filtering.
3136 * 0 - Disable VFs VLAN filtering.
3138 * - (0) if successful.
3139 * - (-ENOTSUP) if hardware doesn't support.
3140 * - (-ENODEV) if *port_id* invalid.
3141 * - (-EINVAL) if bad parameter.
3144 rte_eth_dev_set_vf_vlan_filter(uint8_t port, uint16_t vlan_id,
3149 * Set a traffic mirroring rule on an Ethernet device
3152 * The port identifier of the Ethernet device.
3153 * @param mirror_conf
3154 * The pointer to the traffic mirroring structure describing the mirroring rule.
3155 * The *rte_eth_vm_mirror_conf* structure includes the type of mirroring rule,
3156 * destination pool and the value of rule if enable vlan or pool mirroring.
3159 * The index of traffic mirroring rule, we support four separated rules.
3161 * 1 - Enable a mirroring rule.
3162 * 0 - Disable a mirroring rule.
3164 * - (0) if successful.
3165 * - (-ENOTSUP) if hardware doesn't support this feature.
3166 * - (-ENODEV) if *port_id* invalid.
3167 * - (-EINVAL) if the mr_conf information is not correct.
3169 int rte_eth_mirror_rule_set(uint8_t port_id,
3170 struct rte_eth_vmdq_mirror_conf *mirror_conf,
3175 * Reset a traffic mirroring rule on an Ethernet device.
3178 * The port identifier of the Ethernet device.
3180 * The index of traffic mirroring rule, we support four separated rules.
3182 * - (0) if successful.
3183 * - (-ENOTSUP) if hardware doesn't support this feature.
3184 * - (-ENODEV) if *port_id* invalid.
3185 * - (-EINVAL) if bad parameter.
3187 int rte_eth_mirror_rule_reset(uint8_t port_id,
3191 * Set the rate limitation for a queue on an Ethernet device.
3194 * The port identifier of the Ethernet device.
3198 * The tx rate allocated from the total link speed for this queue.
3200 * - (0) if successful.
3201 * - (-ENOTSUP) if hardware doesn't support this feature.
3202 * - (-ENODEV) if *port_id* invalid.
3203 * - (-EINVAL) if bad parameter.
3205 int rte_eth_set_queue_rate_limit(uint8_t port_id, uint16_t queue_idx,
3209 * Set the rate limitation for a vf on an Ethernet device.
3212 * The port identifier of the Ethernet device.
3216 * The tx rate allocated from the total link speed for this VF id.
3218 * The queue mask which need to set the rate.
3220 * - (0) if successful.
3221 * - (-ENOTSUP) if hardware doesn't support this feature.
3222 * - (-ENODEV) if *port_id* invalid.
3223 * - (-EINVAL) if bad parameter.
3225 int rte_eth_set_vf_rate_limit(uint8_t port_id, uint16_t vf,
3226 uint16_t tx_rate, uint64_t q_msk);
3229 * Initialize bypass logic. This function needs to be called before
3230 * executing any other bypass API.
3233 * The port identifier of the Ethernet device.
3235 * - (0) if successful.
3236 * - (-ENOTSUP) if hardware doesn't support.
3237 * - (-EINVAL) if bad parameter.
3239 int rte_eth_dev_bypass_init(uint8_t port);
3242 * Return bypass state.
3245 * The port identifier of the Ethernet device.
3247 * The return bypass state.
3250 * - (3) Isolate mode
3252 * - (0) if successful.
3253 * - (-ENOTSUP) if hardware doesn't support.
3254 * - (-EINVAL) if bad parameter.
3256 int rte_eth_dev_bypass_state_show(uint8_t port, uint32_t *state);
3262 * The port identifier of the Ethernet device.
3264 * The current bypass state.
3267 * - (3) Isolate mode
3269 * - (0) if successful.
3270 * - (-ENOTSUP) if hardware doesn't support.
3271 * - (-EINVAL) if bad parameter.
3273 int rte_eth_dev_bypass_state_set(uint8_t port, uint32_t *new_state);
3276 * Return bypass state when given event occurs.
3279 * The port identifier of the Ethernet device.
3282 * - (1) Main power on (power button is pushed)
3283 * - (2) Auxiliary power on (power supply is being plugged)
3284 * - (3) Main power off (system shutdown and power supply is left plugged in)
3285 * - (4) Auxiliary power off (power supply is being unplugged)
3286 * - (5) Display or set the watchdog timer
3288 * The bypass state when given event occurred.
3291 * - (3) Isolate mode
3293 * - (0) if successful.
3294 * - (-ENOTSUP) if hardware doesn't support.
3295 * - (-EINVAL) if bad parameter.
3297 int rte_eth_dev_bypass_event_show(uint8_t port, uint32_t event, uint32_t *state);
3300 * Set bypass state when given event occurs.
3303 * The port identifier of the Ethernet device.
3306 * - (1) Main power on (power button is pushed)
3307 * - (2) Auxiliary power on (power supply is being plugged)
3308 * - (3) Main power off (system shutdown and power supply is left plugged in)
3309 * - (4) Auxiliary power off (power supply is being unplugged)
3310 * - (5) Display or set the watchdog timer
3312 * The assigned state when given event occurs.
3315 * - (3) Isolate mode
3317 * - (0) if successful.
3318 * - (-ENOTSUP) if hardware doesn't support.
3319 * - (-EINVAL) if bad parameter.
3321 int rte_eth_dev_bypass_event_store(uint8_t port, uint32_t event, uint32_t state);
3324 * Set bypass watchdog timeout count.
3327 * The port identifier of the Ethernet device.
3329 * The timeout to be set.
3330 * - (0) 0 seconds (timer is off)
3339 * - (0) if successful.
3340 * - (-ENOTSUP) if hardware doesn't support.
3341 * - (-EINVAL) if bad parameter.
3343 int rte_eth_dev_wd_timeout_store(uint8_t port, uint32_t timeout);
3346 * Get bypass firmware version.
3349 * The port identifier of the Ethernet device.
3351 * The firmware version
3353 * - (0) if successful.
3354 * - (-ENOTSUP) if hardware doesn't support.
3355 * - (-EINVAL) if bad parameter.
3357 int rte_eth_dev_bypass_ver_show(uint8_t port, uint32_t *ver);
3360 * Return bypass watchdog timeout in seconds
3363 * The port identifier of the Ethernet device.
3365 * The return watchdog timeout. "0" represents timer expired
3366 * - (0) 0 seconds (timer is off)
3375 * - (0) if successful.
3376 * - (-ENOTSUP) if hardware doesn't support.
3377 * - (-EINVAL) if bad parameter.
3379 int rte_eth_dev_bypass_wd_timeout_show(uint8_t port, uint32_t *wd_timeout);
3382 * Reset bypass watchdog timer
3385 * The port identifier of the Ethernet device.
3387 * - (0) if successful.
3388 * - (-ENOTSUP) if hardware doesn't support.
3389 * - (-EINVAL) if bad parameter.
3391 int rte_eth_dev_bypass_wd_reset(uint8_t port);
3394 * Configuration of Receive Side Scaling hash computation of Ethernet device.
3397 * The port identifier of the Ethernet device.
3399 * The new configuration to use for RSS hash computation on the port.
3401 * - (0) if successful.
3402 * - (-ENODEV) if port identifier is invalid.
3403 * - (-ENOTSUP) if hardware doesn't support.
3404 * - (-EINVAL) if bad parameter.
3406 int rte_eth_dev_rss_hash_update(uint8_t port_id,
3407 struct rte_eth_rss_conf *rss_conf);
3410 * Retrieve current configuration of Receive Side Scaling hash computation
3411 * of Ethernet device.
3414 * The port identifier of the Ethernet device.
3416 * Where to store the current RSS hash configuration of the Ethernet device.
3418 * - (0) if successful.
3419 * - (-ENODEV) if port identifier is invalid.
3420 * - (-ENOTSUP) if hardware doesn't support RSS.
3423 rte_eth_dev_rss_hash_conf_get(uint8_t port_id,
3424 struct rte_eth_rss_conf *rss_conf);
3427 * Add UDP tunneling port of an Ethernet device for filtering a specific
3428 * tunneling packet by UDP port number.
3431 * The port identifier of the Ethernet device.
3433 * UDP tunneling configuration.
3436 * - (0) if successful.
3437 * - (-ENODEV) if port identifier is invalid.
3438 * - (-ENOTSUP) if hardware doesn't support tunnel type.
3441 rte_eth_dev_udp_tunnel_add(uint8_t port_id,
3442 struct rte_eth_udp_tunnel *tunnel_udp);
3445 * Detete UDP tunneling port configuration of Ethernet device
3448 * The port identifier of the Ethernet device.
3450 * UDP tunneling configuration.
3453 * - (0) if successful.
3454 * - (-ENODEV) if port identifier is invalid.
3455 * - (-ENOTSUP) if hardware doesn't support tunnel type.
3458 rte_eth_dev_udp_tunnel_delete(uint8_t port_id,
3459 struct rte_eth_udp_tunnel *tunnel_udp);
3462 * Check whether the filter type is supported on an Ethernet device.
3463 * All the supported filter types are defined in 'rte_eth_ctrl.h'.
3466 * The port identifier of the Ethernet device.
3467 * @param filter_type
3470 * - (0) if successful.
3471 * - (-ENOTSUP) if hardware doesn't support this filter type.
3472 * - (-ENODEV) if *port_id* invalid.
3474 int rte_eth_dev_filter_supported(uint8_t port_id, enum rte_filter_type filter_type);
3477 * Take operations to assigned filter type on an Ethernet device.
3478 * All the supported operations and filter types are defined in 'rte_eth_ctrl.h'.
3481 * The port identifier of the Ethernet device.
3482 * @param filter_type
3485 * Type of operation.
3487 * A pointer to arguments defined specifically for the operation.
3489 * - (0) if successful.
3490 * - (-ENOTSUP) if hardware doesn't support.
3491 * - (-ENODEV) if *port_id* invalid.
3492 * - others depends on the specific operations implementation.
3494 int rte_eth_dev_filter_ctrl(uint8_t port_id, enum rte_filter_type filter_type,
3495 enum rte_filter_op filter_op, void *arg);
3498 * Add a callback to be called on packet RX on a given port and queue.
3500 * This API configures a function to be called for each burst of
3501 * packets received on a given NIC port queue. The return value is a pointer
3502 * that can be used to later remove the callback using
3503 * rte_eth_remove_rx_callback().
3506 * The port identifier of the Ethernet device.
3508 * The queue on the Ethernet device on which the callback is to be added.
3510 * The callback function
3512 * A generic pointer parameter which will be passed to each invocation of the
3513 * callback function on this port and queue.
3517 * On success, a pointer value which can later be used to remove the callback.
3519 void *rte_eth_add_rx_callback(uint8_t port_id, uint16_t queue_id,
3520 rte_rx_callback_fn fn, void *user_param);
3523 * Add a callback to be called on packet TX on a given port and queue.
3525 * This API configures a function to be called for each burst of
3526 * packets sent on a given NIC port queue. The return value is a pointer
3527 * that can be used to later remove the callback using
3528 * rte_eth_remove_tx_callback().
3531 * The port identifier of the Ethernet device.
3533 * The queue on the Ethernet device on which the callback is to be added.
3535 * The callback function
3537 * A generic pointer parameter which will be passed to each invocation of the
3538 * callback function on this port and queue.
3542 * On success, a pointer value which can later be used to remove the callback.
3544 void *rte_eth_add_tx_callback(uint8_t port_id, uint16_t queue_id,
3545 rte_tx_callback_fn fn, void *user_param);
3548 * Remove an RX packet callback from a given port and queue.
3550 * This function is used to removed callbacks that were added to a NIC port
3551 * queue using rte_eth_add_rx_callback().
3553 * Note: the callback is removed from the callback list but it isn't freed
3554 * since the it may still be in use. The memory for the callback can be
3555 * subsequently freed back by the application by calling rte_free():
3557 * - Immediately - if the port is stopped, or the user knows that no
3558 * callbacks are in flight e.g. if called from the thread doing RX/TX
3561 * - After a short delay - where the delay is sufficient to allow any
3562 * in-flight callbacks to complete.
3565 * The port identifier of the Ethernet device.
3567 * The queue on the Ethernet device from which the callback is to be removed.
3569 * User supplied callback created via rte_eth_add_rx_callback().
3572 * - 0: Success. Callback was removed.
3573 * - -ENOTSUP: Callback support is not available.
3574 * - -EINVAL: The port_id or the queue_id is out of range, or the callback
3575 * is NULL or not found for the port/queue.
3577 int rte_eth_remove_rx_callback(uint8_t port_id, uint16_t queue_id,
3578 struct rte_eth_rxtx_callback *user_cb);
3581 * Remove a TX packet callback from a given port and queue.
3583 * This function is used to removed callbacks that were added to a NIC port
3584 * queue using rte_eth_add_tx_callback().
3586 * Note: the callback is removed from the callback list but it isn't freed
3587 * since the it may still be in use. The memory for the callback can be
3588 * subsequently freed back by the application by calling rte_free():
3590 * - Immediately - if the port is stopped, or the user knows that no
3591 * callbacks are in flight e.g. if called from the thread doing RX/TX
3594 * - After a short delay - where the delay is sufficient to allow any
3595 * in-flight callbacks to complete.
3598 * The port identifier of the Ethernet device.
3600 * The queue on the Ethernet device from which the callback is to be removed.
3602 * User supplied callback created via rte_eth_add_tx_callback().
3605 * - 0: Success. Callback was removed.
3606 * - -ENOTSUP: Callback support is not available.
3607 * - -EINVAL: The port_id or the queue_id is out of range, or the callback
3608 * is NULL or not found for the port/queue.
3610 int rte_eth_remove_tx_callback(uint8_t port_id, uint16_t queue_id,
3611 struct rte_eth_rxtx_callback *user_cb);
3617 #endif /* _RTE_ETHDEV_H_ */