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.
176 #include <rte_interrupts.h>
178 #include <rte_mbuf.h>
179 #include "rte_ether.h"
180 #include "rte_eth_ctrl.h"
183 * A structure used to retrieve statistics for an Ethernet port.
185 struct rte_eth_stats {
186 uint64_t ipackets; /**< Total number of successfully received packets. */
187 uint64_t opackets; /**< Total number of successfully transmitted packets.*/
188 uint64_t ibytes; /**< Total number of successfully received bytes. */
189 uint64_t obytes; /**< Total number of successfully transmitted bytes. */
190 uint64_t imissed; /**< Total of RX missed packets (e.g full FIFO). */
191 uint64_t ibadcrc; /**< Total of RX packets with CRC error. */
192 uint64_t ibadlen; /**< Total of RX packets with bad length. */
193 uint64_t ierrors; /**< Total number of erroneous received packets. */
194 uint64_t oerrors; /**< Total number of failed transmitted packets. */
195 uint64_t imcasts; /**< Total number of multicast received packets. */
196 uint64_t rx_nombuf; /**< Total number of RX mbuf allocation failures. */
197 uint64_t fdirmatch; /**< Total number of RX packets matching a filter. */
198 uint64_t fdirmiss; /**< Total number of RX packets not matching any filter. */
199 uint64_t tx_pause_xon; /**< Total nb. of XON pause frame sent. */
200 uint64_t rx_pause_xon; /**< Total nb. of XON pause frame received. */
201 uint64_t tx_pause_xoff; /**< Total nb. of XOFF pause frame sent. */
202 uint64_t rx_pause_xoff; /**< Total nb. of XOFF pause frame received. */
203 uint64_t q_ipackets[RTE_ETHDEV_QUEUE_STAT_CNTRS];
204 /**< Total number of queue RX packets. */
205 uint64_t q_opackets[RTE_ETHDEV_QUEUE_STAT_CNTRS];
206 /**< Total number of queue TX packets. */
207 uint64_t q_ibytes[RTE_ETHDEV_QUEUE_STAT_CNTRS];
208 /**< Total number of successfully received queue bytes. */
209 uint64_t q_obytes[RTE_ETHDEV_QUEUE_STAT_CNTRS];
210 /**< Total number of successfully transmitted queue bytes. */
211 uint64_t q_errors[RTE_ETHDEV_QUEUE_STAT_CNTRS];
212 /**< Total number of queue packets received that are dropped. */
214 /**< Total number of good packets received from loopback,VF Only */
216 /**< Total number of good packets transmitted to loopback,VF Only */
218 /**< Total number of good bytes received from loopback,VF Only */
220 /**< Total number of good bytes transmitted to loopback,VF Only */
224 * A structure used to retrieve link-level information of an Ethernet port.
226 struct rte_eth_link {
227 uint16_t link_speed; /**< ETH_LINK_SPEED_[10, 100, 1000, 10000] */
228 uint16_t link_duplex; /**< ETH_LINK_[HALF_DUPLEX, FULL_DUPLEX] */
229 uint8_t link_status : 1; /**< 1 -> link up, 0 -> link down */
230 }__attribute__((aligned(8))); /**< aligned for atomic64 read/write */
232 #define ETH_LINK_SPEED_AUTONEG 0 /**< Auto-negotiate link speed. */
233 #define ETH_LINK_SPEED_10 10 /**< 10 megabits/second. */
234 #define ETH_LINK_SPEED_100 100 /**< 100 megabits/second. */
235 #define ETH_LINK_SPEED_1000 1000 /**< 1 gigabits/second. */
236 #define ETH_LINK_SPEED_10000 10000 /**< 10 gigabits/second. */
237 #define ETH_LINK_SPEED_10G 10000 /**< alias of 10 gigabits/second. */
238 #define ETH_LINK_SPEED_20G 20000 /**< 20 gigabits/second. */
239 #define ETH_LINK_SPEED_40G 40000 /**< 40 gigabits/second. */
241 #define ETH_LINK_AUTONEG_DUPLEX 0 /**< Auto-negotiate duplex. */
242 #define ETH_LINK_HALF_DUPLEX 1 /**< Half-duplex connection. */
243 #define ETH_LINK_FULL_DUPLEX 2 /**< Full-duplex connection. */
246 * A structure used to configure the ring threshold registers of an RX/TX
247 * queue for an Ethernet port.
249 struct rte_eth_thresh {
250 uint8_t pthresh; /**< Ring prefetch threshold. */
251 uint8_t hthresh; /**< Ring host threshold. */
252 uint8_t wthresh; /**< Ring writeback threshold. */
256 * Simple flags are used for rte_eth_conf.rxmode.mq_mode.
258 #define ETH_MQ_RX_RSS_FLAG 0x1
259 #define ETH_MQ_RX_DCB_FLAG 0x2
260 #define ETH_MQ_RX_VMDQ_FLAG 0x4
263 * A set of values to identify what method is to be used to route
264 * packets to multiple queues.
266 enum rte_eth_rx_mq_mode {
267 /** None of DCB,RSS or VMDQ mode */
270 /** For RX side, only RSS is on */
271 ETH_MQ_RX_RSS = ETH_MQ_RX_RSS_FLAG,
272 /** For RX side,only DCB is on. */
273 ETH_MQ_RX_DCB = ETH_MQ_RX_DCB_FLAG,
274 /** Both DCB and RSS enable */
275 ETH_MQ_RX_DCB_RSS = ETH_MQ_RX_RSS_FLAG | ETH_MQ_RX_DCB_FLAG,
277 /** Only VMDQ, no RSS nor DCB */
278 ETH_MQ_RX_VMDQ_ONLY = ETH_MQ_RX_VMDQ_FLAG,
279 /** RSS mode with VMDQ */
280 ETH_MQ_RX_VMDQ_RSS = ETH_MQ_RX_RSS_FLAG | ETH_MQ_RX_VMDQ_FLAG,
281 /** Use VMDQ+DCB to route traffic to queues */
282 ETH_MQ_RX_VMDQ_DCB = ETH_MQ_RX_VMDQ_FLAG | ETH_MQ_RX_DCB_FLAG,
283 /** Enable both VMDQ and DCB in VMDq */
284 ETH_MQ_RX_VMDQ_DCB_RSS = ETH_MQ_RX_RSS_FLAG | ETH_MQ_RX_DCB_FLAG |
289 * for rx mq mode backward compatible
291 #define ETH_RSS ETH_MQ_RX_RSS
292 #define VMDQ_DCB ETH_MQ_RX_VMDQ_DCB
293 #define ETH_DCB_RX ETH_MQ_RX_DCB
296 * A set of values to identify what method is to be used to transmit
297 * packets using multi-TCs.
299 enum rte_eth_tx_mq_mode {
300 ETH_MQ_TX_NONE = 0, /**< It is in neither DCB nor VT mode. */
301 ETH_MQ_TX_DCB, /**< For TX side,only DCB is on. */
302 ETH_MQ_TX_VMDQ_DCB, /**< For TX side,both DCB and VT is on. */
303 ETH_MQ_TX_VMDQ_ONLY, /**< Only VT on, no DCB */
307 * for tx mq mode backward compatible
309 #define ETH_DCB_NONE ETH_MQ_TX_NONE
310 #define ETH_VMDQ_DCB_TX ETH_MQ_TX_VMDQ_DCB
311 #define ETH_DCB_TX ETH_MQ_TX_DCB
314 * A structure used to configure the RX features of an Ethernet port.
316 struct rte_eth_rxmode {
317 /** The multi-queue packet distribution mode to be used, e.g. RSS. */
318 enum rte_eth_rx_mq_mode mq_mode;
319 uint32_t max_rx_pkt_len; /**< Only used if jumbo_frame enabled. */
320 uint16_t split_hdr_size; /**< hdr buf size (header_split enabled).*/
321 uint8_t header_split : 1, /**< Header Split enable. */
322 hw_ip_checksum : 1, /**< IP/UDP/TCP checksum offload enable. */
323 hw_vlan_filter : 1, /**< VLAN filter enable. */
324 hw_vlan_strip : 1, /**< VLAN strip enable. */
325 hw_vlan_extend : 1, /**< Extended VLAN enable. */
326 jumbo_frame : 1, /**< Jumbo Frame Receipt enable. */
327 hw_strip_crc : 1, /**< Enable CRC stripping by hardware. */
328 enable_scatter : 1; /**< Enable scatter packets rx handler */
332 * A structure used to configure the Receive Side Scaling (RSS) feature
333 * of an Ethernet port.
334 * If not NULL, the *rss_key* pointer of the *rss_conf* structure points
335 * to an array holding the RSS key to use for hashing specific header
336 * fields of received packets. The length of this array should be indicated
337 * by *rss_key_len* below. Otherwise, a default random hash key is used by
340 * The *rss_key_len* field of the *rss_conf* structure indicates the length
341 * in bytes of the array pointed by *rss_key*. To be compatible, this length
342 * will be checked in i40e only. Others assume 40 bytes to be used as before.
344 * The *rss_hf* field of the *rss_conf* structure indicates the different
345 * types of IPv4/IPv6 packets to which the RSS hashing must be applied.
346 * Supplying an *rss_hf* equal to zero disables the RSS feature.
348 struct rte_eth_rss_conf {
349 uint8_t *rss_key; /**< If not NULL, 40-byte hash key. */
350 uint8_t rss_key_len; /**< hash key length in bytes. */
351 uint64_t rss_hf; /**< Hash functions to apply - see below. */
354 /* Supported RSS offloads */
356 #define ETH_RSS_IPV4_SHIFT 0
357 #define ETH_RSS_IPV4_TCP_SHIFT 1
358 #define ETH_RSS_IPV6_SHIFT 2
359 #define ETH_RSS_IPV6_EX_SHIFT 3
360 #define ETH_RSS_IPV6_TCP_SHIFT 4
361 #define ETH_RSS_IPV6_TCP_EX_SHIFT 5
362 #define ETH_RSS_IPV4_UDP_SHIFT 6
363 #define ETH_RSS_IPV6_UDP_SHIFT 7
364 #define ETH_RSS_IPV6_UDP_EX_SHIFT 8
366 #define ETH_RSS_NONF_IPV4_UDP_SHIFT 31
367 #define ETH_RSS_NONF_IPV4_TCP_SHIFT 33
368 #define ETH_RSS_NONF_IPV4_SCTP_SHIFT 34
369 #define ETH_RSS_NONF_IPV4_OTHER_SHIFT 35
370 #define ETH_RSS_FRAG_IPV4_SHIFT 36
371 #define ETH_RSS_NONF_IPV6_UDP_SHIFT 41
372 #define ETH_RSS_NONF_IPV6_TCP_SHIFT 43
373 #define ETH_RSS_NONF_IPV6_SCTP_SHIFT 44
374 #define ETH_RSS_NONF_IPV6_OTHER_SHIFT 45
375 #define ETH_RSS_FRAG_IPV6_SHIFT 46
376 #define ETH_RSS_FCOE_OX_SHIFT 48
377 #define ETH_RSS_FCOE_RX_SHIFT 49
378 #define ETH_RSS_FCOE_OTHER_SHIFT 50
379 #define ETH_RSS_L2_PAYLOAD_SHIFT 63
382 #define ETH_RSS_IPV4 (1 << ETH_RSS_IPV4_SHIFT)
383 #define ETH_RSS_IPV4_TCP (1 << ETH_RSS_IPV4_TCP_SHIFT)
384 #define ETH_RSS_IPV6 (1 << ETH_RSS_IPV6_SHIFT)
385 #define ETH_RSS_IPV6_EX (1 << ETH_RSS_IPV6_EX_SHIFT)
386 #define ETH_RSS_IPV6_TCP (1 << ETH_RSS_IPV6_TCP_SHIFT)
387 #define ETH_RSS_IPV6_TCP_EX (1 << ETH_RSS_IPV6_TCP_EX_SHIFT)
388 #define ETH_RSS_IPV4_UDP (1 << ETH_RSS_IPV4_UDP_SHIFT)
389 #define ETH_RSS_IPV6_UDP (1 << ETH_RSS_IPV6_UDP_SHIFT)
390 #define ETH_RSS_IPV6_UDP_EX (1 << ETH_RSS_IPV6_UDP_EX_SHIFT)
392 #define ETH_RSS_NONF_IPV4_UDP (1ULL << ETH_RSS_NONF_IPV4_UDP_SHIFT)
393 #define ETH_RSS_NONF_IPV4_TCP (1ULL << ETH_RSS_NONF_IPV4_TCP_SHIFT)
394 #define ETH_RSS_NONF_IPV4_SCTP (1ULL << ETH_RSS_NONF_IPV4_SCTP_SHIFT)
395 #define ETH_RSS_NONF_IPV4_OTHER (1ULL << ETH_RSS_NONF_IPV4_OTHER_SHIFT)
396 #define ETH_RSS_FRAG_IPV4 (1ULL << ETH_RSS_FRAG_IPV4_SHIFT)
397 #define ETH_RSS_NONF_IPV6_UDP (1ULL << ETH_RSS_NONF_IPV6_UDP_SHIFT)
398 #define ETH_RSS_NONF_IPV6_TCP (1ULL << ETH_RSS_NONF_IPV6_TCP_SHIFT)
399 #define ETH_RSS_NONF_IPV6_SCTP (1ULL << ETH_RSS_NONF_IPV6_SCTP_SHIFT)
400 #define ETH_RSS_NONF_IPV6_OTHER (1ULL << ETH_RSS_NONF_IPV6_OTHER_SHIFT)
401 #define ETH_RSS_FRAG_IPV6 (1ULL << ETH_RSS_FRAG_IPV6_SHIFT)
402 /* FCOE relevant should not be used */
403 #define ETH_RSS_FCOE_OX (1ULL << ETH_RSS_FCOE_OX_SHIFT)
404 #define ETH_RSS_FCOE_RX (1ULL << ETH_RSS_FCOE_RX_SHIFT)
405 #define ETH_RSS_FCOE_OTHER (1ULL << ETH_RSS_FCOE_OTHER_SHIFT)
406 #define ETH_RSS_L2_PAYLOAD (1ULL << ETH_RSS_L2_PAYLOAD_SHIFT)
408 #define ETH_RSS_IP ( \
411 ETH_RSS_NONF_IPV4_OTHER | \
412 ETH_RSS_FRAG_IPV4 | \
413 ETH_RSS_NONF_IPV6_OTHER | \
415 #define ETH_RSS_UDP ( \
420 ETH_RSS_IPV6_UDP_EX | \
421 ETH_RSS_NONF_IPV4_UDP | \
422 ETH_RSS_NONF_IPV6_UDP)
423 /**< Mask of valid RSS hash protocols */
424 #define ETH_RSS_PROTO_MASK ( \
430 ETH_RSS_IPV6_TCP_EX | \
433 ETH_RSS_IPV6_UDP_EX | \
434 ETH_RSS_NONF_IPV4_UDP | \
435 ETH_RSS_NONF_IPV4_TCP | \
436 ETH_RSS_NONF_IPV4_SCTP | \
437 ETH_RSS_NONF_IPV4_OTHER | \
438 ETH_RSS_FRAG_IPV4 | \
439 ETH_RSS_NONF_IPV6_UDP | \
440 ETH_RSS_NONF_IPV6_TCP | \
441 ETH_RSS_NONF_IPV6_SCTP | \
442 ETH_RSS_NONF_IPV6_OTHER | \
443 ETH_RSS_FRAG_IPV6 | \
447 * Definitions used for redirection table entry size.
448 * Some RSS RETA sizes may not be supported by some drivers, check the
449 * documentation or the description of relevant functions for more details.
451 #define ETH_RSS_RETA_SIZE_64 64
452 #define ETH_RSS_RETA_SIZE_128 128
453 #define ETH_RSS_RETA_SIZE_512 512
454 #define RTE_RETA_GROUP_SIZE 64
456 /* Definitions used for VMDQ and DCB functionality */
457 #define ETH_VMDQ_MAX_VLAN_FILTERS 64 /**< Maximum nb. of VMDQ vlan filters. */
458 #define ETH_DCB_NUM_USER_PRIORITIES 8 /**< Maximum nb. of DCB priorities. */
459 #define ETH_VMDQ_DCB_NUM_QUEUES 128 /**< Maximum nb. of VMDQ DCB queues. */
460 #define ETH_DCB_NUM_QUEUES 128 /**< Maximum nb. of DCB queues. */
462 /* DCB capability defines */
463 #define ETH_DCB_PG_SUPPORT 0x00000001 /**< Priority Group(ETS) support. */
464 #define ETH_DCB_PFC_SUPPORT 0x00000002 /**< Priority Flow Control support. */
466 /* Definitions used for VLAN Offload functionality */
467 #define ETH_VLAN_STRIP_OFFLOAD 0x0001 /**< VLAN Strip On/Off */
468 #define ETH_VLAN_FILTER_OFFLOAD 0x0002 /**< VLAN Filter On/Off */
469 #define ETH_VLAN_EXTEND_OFFLOAD 0x0004 /**< VLAN Extend On/Off */
471 /* Definitions used for mask VLAN setting */
472 #define ETH_VLAN_STRIP_MASK 0x0001 /**< VLAN Strip setting mask */
473 #define ETH_VLAN_FILTER_MASK 0x0002 /**< VLAN Filter setting mask*/
474 #define ETH_VLAN_EXTEND_MASK 0x0004 /**< VLAN Extend setting mask*/
475 #define ETH_VLAN_ID_MAX 0x0FFF /**< VLAN ID is in lower 12 bits*/
477 /* Definitions used for receive MAC address */
478 #define ETH_NUM_RECEIVE_MAC_ADDR 128 /**< Maximum nb. of receive mac addr. */
480 /* Definitions used for unicast hash */
481 #define ETH_VMDQ_NUM_UC_HASH_ARRAY 128 /**< Maximum nb. of UC hash array. */
483 /* Definitions used for VMDQ pool rx mode setting */
484 #define ETH_VMDQ_ACCEPT_UNTAG 0x0001 /**< accept untagged packets. */
485 #define ETH_VMDQ_ACCEPT_HASH_MC 0x0002 /**< accept packets in multicast table . */
486 #define ETH_VMDQ_ACCEPT_HASH_UC 0x0004 /**< accept packets in unicast table. */
487 #define ETH_VMDQ_ACCEPT_BROADCAST 0x0008 /**< accept broadcast packets. */
488 #define ETH_VMDQ_ACCEPT_MULTICAST 0x0010 /**< multicast promiscuous. */
490 /* Definitions used for VMDQ mirror rules setting */
491 #define ETH_VMDQ_NUM_MIRROR_RULE 4 /**< Maximum nb. of mirror rules. . */
493 #define ETH_VMDQ_POOL_MIRROR 0x0001 /**< Virtual Pool Mirroring. */
494 #define ETH_VMDQ_UPLINK_MIRROR 0x0002 /**< Uplink Port Mirroring. */
495 #define ETH_VMDQ_DOWNLIN_MIRROR 0x0004 /**< Downlink Port Mirroring. */
496 #define ETH_VMDQ_VLAN_MIRROR 0x0008 /**< VLAN Mirroring. */
499 * A structure used to configure VLAN traffic mirror of an Ethernet port.
501 struct rte_eth_vlan_mirror {
502 uint64_t vlan_mask; /**< mask for valid VLAN ID. */
503 uint16_t vlan_id[ETH_VMDQ_MAX_VLAN_FILTERS];
504 /** VLAN ID list for vlan mirror. */
508 * A structure used to configure traffic mirror of an Ethernet port.
510 struct rte_eth_vmdq_mirror_conf {
511 uint8_t rule_type_mask; /**< Mirroring rule type mask we want to set */
512 uint8_t dst_pool; /**< Destination pool for this mirror rule. */
513 uint64_t pool_mask; /**< Bitmap of pool for pool mirroring */
514 struct rte_eth_vlan_mirror vlan; /**< VLAN ID setting for VLAN mirroring */
518 * A structure used to configure 64 entries of Redirection Table of the
519 * Receive Side Scaling (RSS) feature of an Ethernet port. To configure
520 * more than 64 entries supported by hardware, an array of this structure
523 struct rte_eth_rss_reta_entry64 {
525 /**< Mask bits indicate which entries need to be updated/queried. */
526 uint8_t reta[RTE_RETA_GROUP_SIZE];
527 /**< Group of 64 redirection table entries. */
531 * This enum indicates the possible number of traffic classes
532 * in DCB configratioins
534 enum rte_eth_nb_tcs {
535 ETH_4_TCS = 4, /**< 4 TCs with DCB. */
536 ETH_8_TCS = 8 /**< 8 TCs with DCB. */
540 * This enum indicates the possible number of queue pools
541 * in VMDQ configurations.
543 enum rte_eth_nb_pools {
544 ETH_8_POOLS = 8, /**< 8 VMDq pools. */
545 ETH_16_POOLS = 16, /**< 16 VMDq pools. */
546 ETH_32_POOLS = 32, /**< 32 VMDq pools. */
547 ETH_64_POOLS = 64 /**< 64 VMDq pools. */
550 /* This structure may be extended in future. */
551 struct rte_eth_dcb_rx_conf {
552 enum rte_eth_nb_tcs nb_tcs; /**< Possible DCB TCs, 4 or 8 TCs */
553 uint8_t dcb_queue[ETH_DCB_NUM_USER_PRIORITIES];
554 /**< Possible DCB queue,4 or 8. */
557 struct rte_eth_vmdq_dcb_tx_conf {
558 enum rte_eth_nb_pools nb_queue_pools; /**< With DCB, 16 or 32 pools. */
559 uint8_t dcb_queue[ETH_DCB_NUM_USER_PRIORITIES];
560 /**< Possible DCB queue,4 or 8. */
563 struct rte_eth_dcb_tx_conf {
564 enum rte_eth_nb_tcs nb_tcs; /**< Possible DCB TCs, 4 or 8 TCs. */
565 uint8_t dcb_queue[ETH_DCB_NUM_USER_PRIORITIES];
566 /**< Possible DCB queue,4 or 8. */
569 struct rte_eth_vmdq_tx_conf {
570 enum rte_eth_nb_pools nb_queue_pools; /**< VMDq mode, 64 pools. */
574 * A structure used to configure the VMDQ+DCB feature
575 * of an Ethernet port.
577 * Using this feature, packets are routed to a pool of queues, based
578 * on the vlan id in the vlan tag, and then to a specific queue within
579 * that pool, using the user priority vlan tag field.
581 * A default pool may be used, if desired, to route all traffic which
582 * does not match the vlan filter rules.
584 struct rte_eth_vmdq_dcb_conf {
585 enum rte_eth_nb_pools nb_queue_pools; /**< With DCB, 16 or 32 pools */
586 uint8_t enable_default_pool; /**< If non-zero, use a default pool */
587 uint8_t default_pool; /**< The default pool, if applicable */
588 uint8_t nb_pool_maps; /**< We can have up to 64 filters/mappings */
590 uint16_t vlan_id; /**< The vlan id of the received frame */
591 uint64_t pools; /**< Bitmask of pools for packet rx */
592 } pool_map[ETH_VMDQ_MAX_VLAN_FILTERS]; /**< VMDq vlan pool maps. */
593 uint8_t dcb_queue[ETH_DCB_NUM_USER_PRIORITIES];
594 /**< Selects a queue in a pool */
597 struct rte_eth_vmdq_rx_conf {
598 enum rte_eth_nb_pools nb_queue_pools; /**< VMDq only mode, 8 or 64 pools */
599 uint8_t enable_default_pool; /**< If non-zero, use a default pool */
600 uint8_t default_pool; /**< The default pool, if applicable */
601 uint8_t enable_loop_back; /**< Enable VT loop back */
602 uint8_t nb_pool_maps; /**< We can have up to 64 filters/mappings */
603 uint32_t rx_mode; /**< Flags from ETH_VMDQ_ACCEPT_* */
605 uint16_t vlan_id; /**< The vlan id of the received frame */
606 uint64_t pools; /**< Bitmask of pools for packet rx */
607 } pool_map[ETH_VMDQ_MAX_VLAN_FILTERS]; /**< VMDq vlan pool maps. */
611 * A structure used to configure the TX features of an Ethernet port.
613 struct rte_eth_txmode {
614 enum rte_eth_tx_mq_mode mq_mode; /**< TX multi-queues mode. */
616 /* For i40e specifically */
618 uint8_t hw_vlan_reject_tagged : 1,
619 /**< If set, reject sending out tagged pkts */
620 hw_vlan_reject_untagged : 1,
621 /**< If set, reject sending out untagged pkts */
622 hw_vlan_insert_pvid : 1;
623 /**< If set, enable port based VLAN insertion */
627 * A structure used to configure an RX ring of an Ethernet port.
629 struct rte_eth_rxconf {
630 struct rte_eth_thresh rx_thresh; /**< RX ring threshold registers. */
631 uint16_t rx_free_thresh; /**< Drives the freeing of RX descriptors. */
632 uint8_t rx_drop_en; /**< Drop packets if no descriptors are available. */
633 uint8_t rx_deferred_start; /**< Do not start queue with rte_eth_dev_start(). */
636 #define ETH_TXQ_FLAGS_NOMULTSEGS 0x0001 /**< nb_segs=1 for all mbufs */
637 #define ETH_TXQ_FLAGS_NOREFCOUNT 0x0002 /**< refcnt can be ignored */
638 #define ETH_TXQ_FLAGS_NOMULTMEMP 0x0004 /**< all bufs come from same mempool */
639 #define ETH_TXQ_FLAGS_NOVLANOFFL 0x0100 /**< disable VLAN offload */
640 #define ETH_TXQ_FLAGS_NOXSUMSCTP 0x0200 /**< disable SCTP checksum offload */
641 #define ETH_TXQ_FLAGS_NOXSUMUDP 0x0400 /**< disable UDP checksum offload */
642 #define ETH_TXQ_FLAGS_NOXSUMTCP 0x0800 /**< disable TCP checksum offload */
643 #define ETH_TXQ_FLAGS_NOOFFLOADS \
644 (ETH_TXQ_FLAGS_NOVLANOFFL | ETH_TXQ_FLAGS_NOXSUMSCTP | \
645 ETH_TXQ_FLAGS_NOXSUMUDP | ETH_TXQ_FLAGS_NOXSUMTCP)
646 #define ETH_TXQ_FLAGS_NOXSUMS \
647 (ETH_TXQ_FLAGS_NOXSUMSCTP | ETH_TXQ_FLAGS_NOXSUMUDP | \
648 ETH_TXQ_FLAGS_NOXSUMTCP)
650 * A structure used to configure a TX ring of an Ethernet port.
652 struct rte_eth_txconf {
653 struct rte_eth_thresh tx_thresh; /**< TX ring threshold registers. */
654 uint16_t tx_rs_thresh; /**< Drives the setting of RS bit on TXDs. */
655 uint16_t tx_free_thresh; /**< Drives the freeing of TX buffers. */
656 uint32_t txq_flags; /**< Set flags for the Tx queue */
657 uint8_t tx_deferred_start; /**< Do not start queue with rte_eth_dev_start(). */
661 * This enum indicates the flow control mode
663 enum rte_eth_fc_mode {
664 RTE_FC_NONE = 0, /**< Disable flow control. */
665 RTE_FC_RX_PAUSE, /**< RX pause frame, enable flowctrl on TX side. */
666 RTE_FC_TX_PAUSE, /**< TX pause frame, enable flowctrl on RX side. */
667 RTE_FC_FULL /**< Enable flow control on both side. */
671 * A structure used to configure Ethernet flow control parameter.
672 * These parameters will be configured into the register of the NIC.
673 * Please refer to the corresponding data sheet for proper value.
675 struct rte_eth_fc_conf {
676 uint32_t high_water; /**< High threshold value to trigger XOFF */
677 uint32_t low_water; /**< Low threshold value to trigger XON */
678 uint16_t pause_time; /**< Pause quota in the Pause frame */
679 uint16_t send_xon; /**< Is XON frame need be sent */
680 enum rte_eth_fc_mode mode; /**< Link flow control mode */
681 uint8_t mac_ctrl_frame_fwd; /**< Forward MAC control frames */
682 uint8_t autoneg; /**< Use Pause autoneg */
686 * A structure used to configure Ethernet priority flow control parameter.
687 * These parameters will be configured into the register of the NIC.
688 * Please refer to the corresponding data sheet for proper value.
690 struct rte_eth_pfc_conf {
691 struct rte_eth_fc_conf fc; /**< General flow control parameter. */
692 uint8_t priority; /**< VLAN User Priority. */
696 * Memory space that can be configured to store Flow Director filters
697 * in the board memory.
699 enum rte_fdir_pballoc_type {
700 RTE_FDIR_PBALLOC_64K = 0, /**< 64k. */
701 RTE_FDIR_PBALLOC_128K, /**< 128k. */
702 RTE_FDIR_PBALLOC_256K, /**< 256k. */
706 * Select report mode of FDIR hash information in RX descriptors.
708 enum rte_fdir_status_mode {
709 RTE_FDIR_NO_REPORT_STATUS = 0, /**< Never report FDIR hash. */
710 RTE_FDIR_REPORT_STATUS, /**< Only report FDIR hash for matching pkts. */
711 RTE_FDIR_REPORT_STATUS_ALWAYS, /**< Always report FDIR hash. */
715 * A structure used to configure the Flow Director (FDIR) feature
716 * of an Ethernet port.
718 * If mode is RTE_FDIR_DISABLE, the pballoc value is ignored.
720 struct rte_fdir_conf {
721 enum rte_fdir_mode mode; /**< Flow Director mode. */
722 enum rte_fdir_pballoc_type pballoc; /**< Space for FDIR filters. */
723 enum rte_fdir_status_mode status; /**< How to report FDIR hash. */
724 /** RX queue of packets matching a "drop" filter in perfect mode. */
726 struct rte_eth_fdir_masks mask;
727 struct rte_eth_fdir_flex_conf flex_conf;
728 /**< Flex payload configuration. */
732 * UDP tunneling configuration.
734 struct rte_eth_udp_tunnel {
740 * Possible l4type of FDIR filters.
743 RTE_FDIR_L4TYPE_NONE = 0, /**< None. */
744 RTE_FDIR_L4TYPE_UDP, /**< UDP. */
745 RTE_FDIR_L4TYPE_TCP, /**< TCP. */
746 RTE_FDIR_L4TYPE_SCTP, /**< SCTP. */
750 * Select IPv4 or IPv6 FDIR filters.
753 RTE_FDIR_IPTYPE_IPV4 = 0, /**< IPv4. */
754 RTE_FDIR_IPTYPE_IPV6 , /**< IPv6. */
758 * A structure used to define a FDIR packet filter.
760 struct rte_fdir_filter {
761 uint16_t flex_bytes; /**< Flex bytes value to match. */
762 uint16_t vlan_id; /**< VLAN ID value to match, 0 otherwise. */
763 uint16_t port_src; /**< Source port to match, 0 otherwise. */
764 uint16_t port_dst; /**< Destination port to match, 0 otherwise. */
766 uint32_t ipv4_addr; /**< IPv4 source address to match. */
767 uint32_t ipv6_addr[4]; /**< IPv6 source address to match. */
768 } ip_src; /**< IPv4/IPv6 source address to match (union of above). */
770 uint32_t ipv4_addr; /**< IPv4 destination address to match. */
771 uint32_t ipv6_addr[4]; /**< IPv6 destination address to match */
772 } ip_dst; /**< IPv4/IPv6 destination address to match (union of above). */
773 enum rte_l4type l4type; /**< l4type to match: NONE/UDP/TCP/SCTP. */
774 enum rte_iptype iptype; /**< IP packet type to match: IPv4 or IPv6. */
778 * A structure used to configure FDIR masks that are used by the device
779 * to match the various fields of RX packet headers.
780 * @note The only_ip_flow field has the opposite meaning compared to other
783 struct rte_fdir_masks {
784 /** When set to 1, packet l4type is \b NOT relevant in filters, and
785 source and destination port masks must be set to zero. */
786 uint8_t only_ip_flow;
787 /** If set to 1, vlan_id is relevant in filters. */
789 /** If set to 1, vlan_prio is relevant in filters. */
791 /** If set to 1, flexbytes is relevant in filters. */
793 /** If set to 1, set the IPv6 masks. Otherwise set the IPv4 masks. */
794 uint8_t set_ipv6_mask;
795 /** When set to 1, comparison of destination IPv6 address with IP6AT
796 registers is meaningful. */
797 uint8_t comp_ipv6_dst;
798 /** Mask of Destination IPv4 Address. All bits set to 1 define the
799 relevant bits to use in the destination address of an IPv4 packet
800 when matching it against FDIR filters. */
801 uint32_t dst_ipv4_mask;
802 /** Mask of Source IPv4 Address. All bits set to 1 define
803 the relevant bits to use in the source address of an IPv4 packet
804 when matching it against FDIR filters. */
805 uint32_t src_ipv4_mask;
806 /** Mask of Source IPv6 Address. All bits set to 1 define the
807 relevant BYTES to use in the source address of an IPv6 packet
808 when matching it against FDIR filters. */
809 uint16_t dst_ipv6_mask;
810 /** Mask of Destination IPv6 Address. All bits set to 1 define the
811 relevant BYTES to use in the destination address of an IPv6 packet
812 when matching it against FDIR filters. */
813 uint16_t src_ipv6_mask;
814 /** Mask of Source Port. All bits set to 1 define the relevant
815 bits to use in the source port of an IP packets when matching it
816 against FDIR filters. */
817 uint16_t src_port_mask;
818 /** Mask of Destination Port. All bits set to 1 define the relevant
819 bits to use in the destination port of an IP packet when matching it
820 against FDIR filters. */
821 uint16_t dst_port_mask;
825 * A structure used to report the status of the flow director filters in use.
827 struct rte_eth_fdir {
828 /** Number of filters with collision indication. */
830 /** Number of free (non programmed) filters. */
832 /** The Lookup hash value of the added filter that updated the value
833 of the MAXLEN field */
835 /** Longest linked list of filters in the table. */
837 /** Number of added filters. */
839 /** Number of removed filters. */
841 /** Number of failed added filters (no more space in device). */
843 /** Number of failed removed filters. */
848 * A structure used to enable/disable specific device interrupts.
850 struct rte_intr_conf {
851 /** enable/disable lsc interrupt. 0 (default) - disable, 1 enable */
856 * A structure used to configure an Ethernet port.
857 * Depending upon the RX multi-queue mode, extra advanced
858 * configuration settings may be needed.
860 struct rte_eth_conf {
862 /**< ETH_LINK_SPEED_10[0|00|000], or 0 for autonegotation */
863 uint16_t link_duplex;
864 /**< ETH_LINK_[HALF_DUPLEX|FULL_DUPLEX], or 0 for autonegotation */
865 struct rte_eth_rxmode rxmode; /**< Port RX configuration. */
866 struct rte_eth_txmode txmode; /**< Port TX configuration. */
867 uint32_t lpbk_mode; /**< Loopback operation mode. By default the value
868 is 0, meaning the loopback mode is disabled.
869 Read the datasheet of given ethernet controller
870 for details. The possible values of this field
871 are defined in implementation of each driver. */
873 struct rte_eth_rss_conf rss_conf; /**< Port RSS configuration */
874 struct rte_eth_vmdq_dcb_conf vmdq_dcb_conf;
875 /**< Port vmdq+dcb configuration. */
876 struct rte_eth_dcb_rx_conf dcb_rx_conf;
877 /**< Port dcb RX configuration. */
878 struct rte_eth_vmdq_rx_conf vmdq_rx_conf;
879 /**< Port vmdq RX configuration. */
880 } rx_adv_conf; /**< Port RX filtering configuration (union). */
882 struct rte_eth_vmdq_dcb_tx_conf vmdq_dcb_tx_conf;
883 /**< Port vmdq+dcb TX configuration. */
884 struct rte_eth_dcb_tx_conf dcb_tx_conf;
885 /**< Port dcb TX configuration. */
886 struct rte_eth_vmdq_tx_conf vmdq_tx_conf;
887 /**< Port vmdq TX configuration. */
888 } tx_adv_conf; /**< Port TX DCB configuration (union). */
889 /** Currently,Priority Flow Control(PFC) are supported,if DCB with PFC
890 is needed,and the variable must be set ETH_DCB_PFC_SUPPORT. */
891 uint32_t dcb_capability_en;
892 struct rte_fdir_conf fdir_conf; /**< FDIR configuration. */
893 struct rte_intr_conf intr_conf; /**< Interrupt mode configuration. */
897 * A structure used to retrieve the contextual information of
898 * an Ethernet device, such as the controlling driver of the device,
899 * its PCI context, etc...
903 * RX offload capabilities of a device.
905 #define DEV_RX_OFFLOAD_VLAN_STRIP 0x00000001
906 #define DEV_RX_OFFLOAD_IPV4_CKSUM 0x00000002
907 #define DEV_RX_OFFLOAD_UDP_CKSUM 0x00000004
908 #define DEV_RX_OFFLOAD_TCP_CKSUM 0x00000008
909 #define DEV_RX_OFFLOAD_TCP_LRO 0x00000010
912 * TX offload capabilities of a device.
914 #define DEV_TX_OFFLOAD_VLAN_INSERT 0x00000001
915 #define DEV_TX_OFFLOAD_IPV4_CKSUM 0x00000002
916 #define DEV_TX_OFFLOAD_UDP_CKSUM 0x00000004
917 #define DEV_TX_OFFLOAD_TCP_CKSUM 0x00000008
918 #define DEV_TX_OFFLOAD_SCTP_CKSUM 0x00000010
919 #define DEV_TX_OFFLOAD_TCP_TSO 0x00000020
920 #define DEV_TX_OFFLOAD_UDP_TSO 0x00000040
921 #define DEV_TX_OFFLOAD_OUTER_IPV4_CKSUM 0x00000080 /**< Used for tunneling packet. */
923 struct rte_eth_dev_info {
924 struct rte_pci_device *pci_dev; /**< Device PCI information. */
925 const char *driver_name; /**< Device Driver name. */
926 unsigned int if_index; /**< Index to bound host interface, or 0 if none.
927 Use if_indextoname() to translate into an interface name. */
928 uint32_t min_rx_bufsize; /**< Minimum size of RX buffer. */
929 uint32_t max_rx_pktlen; /**< Maximum configurable length of RX pkt. */
930 uint16_t max_rx_queues; /**< Maximum number of RX queues. */
931 uint16_t max_tx_queues; /**< Maximum number of TX queues. */
932 uint32_t max_mac_addrs; /**< Maximum number of MAC addresses. */
933 uint32_t max_hash_mac_addrs;
934 /** Maximum number of hash MAC addresses for MTA and UTA. */
935 uint16_t max_vfs; /**< Maximum number of VFs. */
936 uint16_t max_vmdq_pools; /**< Maximum number of VMDq pools. */
937 uint32_t rx_offload_capa; /**< Device RX offload capabilities. */
938 uint32_t tx_offload_capa; /**< Device TX offload capabilities. */
940 /**< Device redirection table size, the total number of entries. */
941 struct rte_eth_rxconf default_rxconf; /**< Default RX configuration */
942 struct rte_eth_txconf default_txconf; /**< Default TX configuration */
943 uint16_t vmdq_queue_base; /**< First queue ID for VMDQ pools. */
944 uint16_t vmdq_queue_num; /**< Queue number for VMDQ pools. */
945 uint16_t vmdq_pool_base; /**< First ID of VMDQ pools. */
948 /** Maximum name length for extended statistics counters */
949 #define RTE_ETH_XSTATS_NAME_SIZE 64
952 * An Ethernet device extended statistic structure
954 * This structure is used by ethdev->eth_xstats_get() to provide
955 * statistics that are not provided in the generic rte_eth_stats
958 struct rte_eth_xstats {
959 char name[RTE_ETH_XSTATS_NAME_SIZE];
965 struct rte_eth_dev_callback;
966 /** @internal Structure to keep track of registered callbacks */
967 TAILQ_HEAD(rte_eth_dev_cb_list, rte_eth_dev_callback);
970 #define TCP_URG_FLAG 0x20
971 #define TCP_UGR_FLAG 0x20
972 #define TCP_ACK_FLAG 0x10
973 #define TCP_PSH_FLAG 0x08
974 #define TCP_RST_FLAG 0x04
975 #define TCP_SYN_FLAG 0x02
976 #define TCP_FIN_FLAG 0x01
977 #define TCP_FLAG_ALL 0x3F
980 * A structure used to define a 2tuple filter.
982 struct rte_2tuple_filter {
983 uint16_t dst_port; /**< big endian. */
986 uint16_t priority; /**< used when more than one filter matches. */
987 uint8_t dst_port_mask:1, /**< if mask is 1b, means not compare. */
992 * A structure used to define a 5tuple filter.
994 struct rte_5tuple_filter {
995 uint32_t dst_ip; /**< destination IP address in big endian. */
996 uint32_t src_ip; /**< source IP address in big endian. */
997 uint16_t dst_port; /**< destination port in big endian. */
998 uint16_t src_port; /**< source Port big endian. */
999 uint8_t protocol; /**< l4 protocol. */
1000 uint8_t tcp_flags; /**< tcp flags. */
1001 uint16_t priority; /**< seven evels (001b-111b), 111b is highest,
1002 used when more than one filter matches. */
1003 uint8_t dst_ip_mask:1, /**< if mask is 1b, do not compare dst ip. */
1004 src_ip_mask:1, /**< if mask is 1b, do not compare src ip. */
1005 dst_port_mask:1, /**< if mask is 1b, do not compare dst port. */
1006 src_port_mask:1, /**< if mask is 1b, do not compare src port. */
1007 protocol_mask:1; /**< if mask is 1b, do not compare protocol. */
1011 * Definitions of all functions exported by an Ethernet driver through the
1012 * the generic structure of type *eth_dev_ops* supplied in the *rte_eth_dev*
1013 * structure associated with an Ethernet device.
1016 typedef int (*eth_dev_configure_t)(struct rte_eth_dev *dev);
1017 /**< @internal Ethernet device configuration. */
1019 typedef int (*eth_dev_start_t)(struct rte_eth_dev *dev);
1020 /**< @internal Function used to start a configured Ethernet device. */
1022 typedef void (*eth_dev_stop_t)(struct rte_eth_dev *dev);
1023 /**< @internal Function used to stop a configured Ethernet device. */
1025 typedef int (*eth_dev_set_link_up_t)(struct rte_eth_dev *dev);
1026 /**< @internal Function used to link up a configured Ethernet device. */
1028 typedef int (*eth_dev_set_link_down_t)(struct rte_eth_dev *dev);
1029 /**< @internal Function used to link down a configured Ethernet device. */
1031 typedef void (*eth_dev_close_t)(struct rte_eth_dev *dev);
1032 /**< @internal Function used to close a configured Ethernet device. */
1034 typedef void (*eth_promiscuous_enable_t)(struct rte_eth_dev *dev);
1035 /**< @internal Function used to enable the RX promiscuous mode of an Ethernet device. */
1037 typedef void (*eth_promiscuous_disable_t)(struct rte_eth_dev *dev);
1038 /**< @internal Function used to disable the RX promiscuous mode of an Ethernet device. */
1040 typedef void (*eth_allmulticast_enable_t)(struct rte_eth_dev *dev);
1041 /**< @internal Enable the receipt of all multicast packets by an Ethernet device. */
1043 typedef void (*eth_allmulticast_disable_t)(struct rte_eth_dev *dev);
1044 /**< @internal Disable the receipt of all multicast packets by an Ethernet device. */
1046 typedef int (*eth_link_update_t)(struct rte_eth_dev *dev,
1047 int wait_to_complete);
1048 /**< @internal Get link speed, duplex mode and state (up/down) of an Ethernet device. */
1050 typedef void (*eth_stats_get_t)(struct rte_eth_dev *dev,
1051 struct rte_eth_stats *igb_stats);
1052 /**< @internal Get global I/O statistics of an Ethernet device. */
1054 typedef void (*eth_stats_reset_t)(struct rte_eth_dev *dev);
1055 /**< @internal Reset global I/O statistics of an Ethernet device to 0. */
1057 typedef int (*eth_xstats_get_t)(struct rte_eth_dev *dev,
1058 struct rte_eth_xstats *stats, unsigned n);
1059 /**< @internal Get extended stats of an Ethernet device. */
1061 typedef void (*eth_xstats_reset_t)(struct rte_eth_dev *dev);
1062 /**< @internal Reset extended stats of an Ethernet device. */
1064 typedef int (*eth_queue_stats_mapping_set_t)(struct rte_eth_dev *dev,
1068 /**< @internal Set a queue statistics mapping for a tx/rx queue of an Ethernet device. */
1070 typedef void (*eth_dev_infos_get_t)(struct rte_eth_dev *dev,
1071 struct rte_eth_dev_info *dev_info);
1072 /**< @internal Get specific informations of an Ethernet device. */
1074 typedef int (*eth_queue_start_t)(struct rte_eth_dev *dev,
1076 /**< @internal Start rx and tx of a queue of an Ethernet device. */
1078 typedef int (*eth_queue_stop_t)(struct rte_eth_dev *dev,
1080 /**< @internal Stop rx and tx of a queue of an Ethernet device. */
1082 typedef int (*eth_rx_queue_setup_t)(struct rte_eth_dev *dev,
1083 uint16_t rx_queue_id,
1084 uint16_t nb_rx_desc,
1085 unsigned int socket_id,
1086 const struct rte_eth_rxconf *rx_conf,
1087 struct rte_mempool *mb_pool);
1088 /**< @internal Set up a receive queue of an Ethernet device. */
1090 typedef int (*eth_tx_queue_setup_t)(struct rte_eth_dev *dev,
1091 uint16_t tx_queue_id,
1092 uint16_t nb_tx_desc,
1093 unsigned int socket_id,
1094 const struct rte_eth_txconf *tx_conf);
1095 /**< @internal Setup a transmit queue of an Ethernet device. */
1097 typedef void (*eth_queue_release_t)(void *queue);
1098 /**< @internal Release memory resources allocated by given RX/TX queue. */
1100 typedef uint32_t (*eth_rx_queue_count_t)(struct rte_eth_dev *dev,
1101 uint16_t rx_queue_id);
1102 /**< @Get number of available descriptors on a receive queue of an Ethernet device. */
1104 typedef int (*eth_rx_descriptor_done_t)(void *rxq, uint16_t offset);
1105 /**< @Check DD bit of specific RX descriptor */
1107 typedef int (*mtu_set_t)(struct rte_eth_dev *dev, uint16_t mtu);
1108 /**< @internal Set MTU. */
1110 typedef int (*vlan_filter_set_t)(struct rte_eth_dev *dev,
1113 /**< @internal filtering of a VLAN Tag Identifier by an Ethernet device. */
1115 typedef void (*vlan_tpid_set_t)(struct rte_eth_dev *dev,
1117 /**< @internal set the outer VLAN-TPID by an Ethernet device. */
1119 typedef void (*vlan_offload_set_t)(struct rte_eth_dev *dev, int mask);
1120 /**< @internal set VLAN offload function by an Ethernet device. */
1122 typedef int (*vlan_pvid_set_t)(struct rte_eth_dev *dev,
1125 /**< @internal set port based TX VLAN insertion by an Ethernet device. */
1127 typedef void (*vlan_strip_queue_set_t)(struct rte_eth_dev *dev,
1128 uint16_t rx_queue_id,
1130 /**< @internal VLAN stripping enable/disable by an queue of Ethernet device. */
1132 typedef uint16_t (*eth_rx_burst_t)(void *rxq,
1133 struct rte_mbuf **rx_pkts,
1135 /**< @internal Retrieve input packets from a receive queue of an Ethernet device. */
1137 typedef uint16_t (*eth_tx_burst_t)(void *txq,
1138 struct rte_mbuf **tx_pkts,
1140 /**< @internal Send output packets on a transmit queue of an Ethernet device. */
1142 typedef int (*fdir_add_signature_filter_t)(struct rte_eth_dev *dev,
1143 struct rte_fdir_filter *fdir_ftr,
1145 /**< @internal Setup a new signature filter rule on an Ethernet device */
1147 typedef int (*fdir_update_signature_filter_t)(struct rte_eth_dev *dev,
1148 struct rte_fdir_filter *fdir_ftr,
1150 /**< @internal Update a signature filter rule on an Ethernet device */
1152 typedef int (*fdir_remove_signature_filter_t)(struct rte_eth_dev *dev,
1153 struct rte_fdir_filter *fdir_ftr);
1154 /**< @internal Remove a signature filter rule on an Ethernet device */
1156 typedef void (*fdir_infos_get_t)(struct rte_eth_dev *dev,
1157 struct rte_eth_fdir *fdir);
1158 /**< @internal Get information about fdir status */
1160 typedef int (*fdir_add_perfect_filter_t)(struct rte_eth_dev *dev,
1161 struct rte_fdir_filter *fdir_ftr,
1162 uint16_t soft_id, uint8_t rx_queue,
1164 /**< @internal Setup a new perfect filter rule on an Ethernet device */
1166 typedef int (*fdir_update_perfect_filter_t)(struct rte_eth_dev *dev,
1167 struct rte_fdir_filter *fdir_ftr,
1168 uint16_t soft_id, uint8_t rx_queue,
1170 /**< @internal Update a perfect filter rule on an Ethernet device */
1172 typedef int (*fdir_remove_perfect_filter_t)(struct rte_eth_dev *dev,
1173 struct rte_fdir_filter *fdir_ftr,
1175 /**< @internal Remove a perfect filter rule on an Ethernet device */
1177 typedef int (*fdir_set_masks_t)(struct rte_eth_dev *dev,
1178 struct rte_fdir_masks *fdir_masks);
1179 /**< @internal Setup flow director masks on an Ethernet device */
1181 typedef int (*flow_ctrl_get_t)(struct rte_eth_dev *dev,
1182 struct rte_eth_fc_conf *fc_conf);
1183 /**< @internal Get current flow control parameter on an Ethernet device */
1185 typedef int (*flow_ctrl_set_t)(struct rte_eth_dev *dev,
1186 struct rte_eth_fc_conf *fc_conf);
1187 /**< @internal Setup flow control parameter on an Ethernet device */
1189 typedef int (*priority_flow_ctrl_set_t)(struct rte_eth_dev *dev,
1190 struct rte_eth_pfc_conf *pfc_conf);
1191 /**< @internal Setup priority flow control parameter on an Ethernet device */
1193 typedef int (*reta_update_t)(struct rte_eth_dev *dev,
1194 struct rte_eth_rss_reta_entry64 *reta_conf,
1195 uint16_t reta_size);
1196 /**< @internal Update RSS redirection table on an Ethernet device */
1198 typedef int (*reta_query_t)(struct rte_eth_dev *dev,
1199 struct rte_eth_rss_reta_entry64 *reta_conf,
1200 uint16_t reta_size);
1201 /**< @internal Query RSS redirection table on an Ethernet device */
1203 typedef int (*rss_hash_update_t)(struct rte_eth_dev *dev,
1204 struct rte_eth_rss_conf *rss_conf);
1205 /**< @internal Update RSS hash configuration of an Ethernet device */
1207 typedef int (*rss_hash_conf_get_t)(struct rte_eth_dev *dev,
1208 struct rte_eth_rss_conf *rss_conf);
1209 /**< @internal Get current RSS hash configuration of an Ethernet device */
1211 typedef int (*eth_dev_led_on_t)(struct rte_eth_dev *dev);
1212 /**< @internal Turn on SW controllable LED on an Ethernet device */
1214 typedef int (*eth_dev_led_off_t)(struct rte_eth_dev *dev);
1215 /**< @internal Turn off SW controllable LED on an Ethernet device */
1217 typedef void (*eth_mac_addr_remove_t)(struct rte_eth_dev *dev, uint32_t index);
1218 /**< @internal Remove MAC address from receive address register */
1220 typedef void (*eth_mac_addr_add_t)(struct rte_eth_dev *dev,
1221 struct ether_addr *mac_addr,
1224 /**< @internal Set a MAC address into Receive Address Address Register */
1226 typedef void (*eth_mac_addr_set_t)(struct rte_eth_dev *dev,
1227 struct ether_addr *mac_addr);
1228 /**< @internal Set a MAC address into Receive Address Address Register */
1230 typedef int (*eth_uc_hash_table_set_t)(struct rte_eth_dev *dev,
1231 struct ether_addr *mac_addr,
1233 /**< @internal Set a Unicast Hash bitmap */
1235 typedef int (*eth_uc_all_hash_table_set_t)(struct rte_eth_dev *dev,
1237 /**< @internal Set all Unicast Hash bitmap */
1239 typedef int (*eth_set_vf_rx_mode_t)(struct rte_eth_dev *dev,
1243 /**< @internal Set a VF receive mode */
1245 typedef int (*eth_set_vf_rx_t)(struct rte_eth_dev *dev,
1248 /**< @internal Set a VF receive mode */
1250 typedef int (*eth_set_vf_tx_t)(struct rte_eth_dev *dev,
1253 /**< @internal Enable or disable a VF transmit */
1255 typedef int (*eth_set_vf_vlan_filter_t)(struct rte_eth_dev *dev,
1259 /**< @internal Set VF VLAN pool filter */
1261 typedef int (*eth_set_queue_rate_limit_t)(struct rte_eth_dev *dev,
1264 /**< @internal Set queue TX rate */
1266 typedef int (*eth_set_vf_rate_limit_t)(struct rte_eth_dev *dev,
1270 /**< @internal Set VF TX rate */
1272 typedef int (*eth_mirror_rule_set_t)(struct rte_eth_dev *dev,
1273 struct rte_eth_vmdq_mirror_conf *mirror_conf,
1276 /**< @internal Add a traffic mirroring rule on an Ethernet device */
1278 typedef int (*eth_mirror_rule_reset_t)(struct rte_eth_dev *dev,
1280 /**< @internal Remove a traffic mirroring rule on an Ethernet device */
1282 typedef int (*eth_udp_tunnel_add_t)(struct rte_eth_dev *dev,
1283 struct rte_eth_udp_tunnel *tunnel_udp);
1284 /**< @internal Add tunneling UDP info */
1286 typedef int (*eth_udp_tunnel_del_t)(struct rte_eth_dev *dev,
1287 struct rte_eth_udp_tunnel *tunnel_udp);
1288 /**< @internal Delete tunneling UDP info */
1291 #ifdef RTE_NIC_BYPASS
1294 RTE_BYPASS_MODE_NONE,
1295 RTE_BYPASS_MODE_NORMAL,
1296 RTE_BYPASS_MODE_BYPASS,
1297 RTE_BYPASS_MODE_ISOLATE,
1298 RTE_BYPASS_MODE_NUM,
1301 #define RTE_BYPASS_MODE_VALID(x) \
1302 ((x) > RTE_BYPASS_MODE_NONE && (x) < RTE_BYPASS_MODE_NUM)
1305 RTE_BYPASS_EVENT_NONE,
1306 RTE_BYPASS_EVENT_START,
1307 RTE_BYPASS_EVENT_OS_ON = RTE_BYPASS_EVENT_START,
1308 RTE_BYPASS_EVENT_POWER_ON,
1309 RTE_BYPASS_EVENT_OS_OFF,
1310 RTE_BYPASS_EVENT_POWER_OFF,
1311 RTE_BYPASS_EVENT_TIMEOUT,
1312 RTE_BYPASS_EVENT_NUM
1315 #define RTE_BYPASS_EVENT_VALID(x) \
1316 ((x) > RTE_BYPASS_EVENT_NONE && (x) < RTE_BYPASS_MODE_NUM)
1319 RTE_BYPASS_TMT_OFF, /* timeout disabled. */
1320 RTE_BYPASS_TMT_1_5_SEC, /* timeout for 1.5 seconds */
1321 RTE_BYPASS_TMT_2_SEC, /* timeout for 2 seconds */
1322 RTE_BYPASS_TMT_3_SEC, /* timeout for 3 seconds */
1323 RTE_BYPASS_TMT_4_SEC, /* timeout for 4 seconds */
1324 RTE_BYPASS_TMT_8_SEC, /* timeout for 8 seconds */
1325 RTE_BYPASS_TMT_16_SEC, /* timeout for 16 seconds */
1326 RTE_BYPASS_TMT_32_SEC, /* timeout for 32 seconds */
1330 #define RTE_BYPASS_TMT_VALID(x) \
1331 ((x) == RTE_BYPASS_TMT_OFF || \
1332 ((x) > RTE_BYPASS_TMT_OFF && (x) < RTE_BYPASS_TMT_NUM))
1334 typedef void (*bypass_init_t)(struct rte_eth_dev *dev);
1335 typedef int32_t (*bypass_state_set_t)(struct rte_eth_dev *dev, uint32_t *new_state);
1336 typedef int32_t (*bypass_state_show_t)(struct rte_eth_dev *dev, uint32_t *state);
1337 typedef int32_t (*bypass_event_set_t)(struct rte_eth_dev *dev, uint32_t state, uint32_t event);
1338 typedef int32_t (*bypass_event_show_t)(struct rte_eth_dev *dev, uint32_t event_shift, uint32_t *event);
1339 typedef int32_t (*bypass_wd_timeout_set_t)(struct rte_eth_dev *dev, uint32_t timeout);
1340 typedef int32_t (*bypass_wd_timeout_show_t)(struct rte_eth_dev *dev, uint32_t *wd_timeout);
1341 typedef int32_t (*bypass_ver_show_t)(struct rte_eth_dev *dev, uint32_t *ver);
1342 typedef int32_t (*bypass_wd_reset_t)(struct rte_eth_dev *dev);
1345 typedef int (*eth_add_2tuple_filter_t)(struct rte_eth_dev *dev,
1346 uint16_t index, struct rte_2tuple_filter *filter,
1348 /**< @internal Setup a new 2tuple filter rule on an Ethernet device */
1350 typedef int (*eth_remove_2tuple_filter_t)(struct rte_eth_dev *dev,
1352 /**< @internal Remove a 2tuple filter rule on an Ethernet device */
1354 typedef int (*eth_get_2tuple_filter_t)(struct rte_eth_dev *dev,
1355 uint16_t index, struct rte_2tuple_filter *filter,
1356 uint16_t *rx_queue);
1357 /**< @internal Get a 2tuple filter rule on an Ethernet device */
1359 typedef int (*eth_add_5tuple_filter_t)(struct rte_eth_dev *dev,
1360 uint16_t index, struct rte_5tuple_filter *filter,
1362 /**< @internal Setup a new 5tuple filter rule on an Ethernet device */
1364 typedef int (*eth_remove_5tuple_filter_t)(struct rte_eth_dev *dev,
1366 /**< @internal Remove a 5tuple filter rule on an Ethernet device */
1368 typedef int (*eth_get_5tuple_filter_t)(struct rte_eth_dev *dev,
1369 uint16_t index, struct rte_5tuple_filter *filter,
1370 uint16_t *rx_queue);
1371 /**< @internal Get a 5tuple filter rule on an Ethernet device */
1373 typedef int (*eth_filter_ctrl_t)(struct rte_eth_dev *dev,
1374 enum rte_filter_type filter_type,
1375 enum rte_filter_op filter_op,
1377 /**< @internal Take operations to assigned filter type on an Ethernet device */
1380 * @internal A structure containing the functions exported by an Ethernet driver.
1382 struct eth_dev_ops {
1383 eth_dev_configure_t dev_configure; /**< Configure device. */
1384 eth_dev_start_t dev_start; /**< Start device. */
1385 eth_dev_stop_t dev_stop; /**< Stop device. */
1386 eth_dev_set_link_up_t dev_set_link_up; /**< Device link up. */
1387 eth_dev_set_link_down_t dev_set_link_down; /**< Device link down. */
1388 eth_dev_close_t dev_close; /**< Close device. */
1389 eth_promiscuous_enable_t promiscuous_enable; /**< Promiscuous ON. */
1390 eth_promiscuous_disable_t promiscuous_disable;/**< Promiscuous OFF. */
1391 eth_allmulticast_enable_t allmulticast_enable;/**< RX multicast ON. */
1392 eth_allmulticast_disable_t allmulticast_disable;/**< RX multicast OF. */
1393 eth_link_update_t link_update; /**< Get device link state. */
1394 eth_stats_get_t stats_get; /**< Get generic device statistics. */
1395 eth_stats_reset_t stats_reset; /**< Reset generic device statistics. */
1396 eth_xstats_get_t xstats_get; /**< Get extended device statistics. */
1397 eth_xstats_reset_t xstats_reset; /**< Reset extended device statistics. */
1398 eth_queue_stats_mapping_set_t queue_stats_mapping_set;
1399 /**< Configure per queue stat counter mapping. */
1400 eth_dev_infos_get_t dev_infos_get; /**< Get device info. */
1401 mtu_set_t mtu_set; /**< Set MTU. */
1402 vlan_filter_set_t vlan_filter_set; /**< Filter VLAN Setup. */
1403 vlan_tpid_set_t vlan_tpid_set; /**< Outer VLAN TPID Setup. */
1404 vlan_strip_queue_set_t vlan_strip_queue_set; /**< VLAN Stripping on queue. */
1405 vlan_offload_set_t vlan_offload_set; /**< Set VLAN Offload. */
1406 vlan_pvid_set_t vlan_pvid_set; /**< Set port based TX VLAN insertion */
1407 eth_queue_start_t rx_queue_start;/**< Start RX for a queue.*/
1408 eth_queue_stop_t rx_queue_stop;/**< Stop RX for a queue.*/
1409 eth_queue_start_t tx_queue_start;/**< Start TX for a queue.*/
1410 eth_queue_stop_t tx_queue_stop;/**< Stop TX for a queue.*/
1411 eth_rx_queue_setup_t rx_queue_setup;/**< Set up device RX queue.*/
1412 eth_queue_release_t rx_queue_release;/**< Release RX queue.*/
1413 eth_rx_queue_count_t rx_queue_count; /**< Get Rx queue count. */
1414 eth_rx_descriptor_done_t rx_descriptor_done; /**< Check rxd DD bit */
1415 eth_tx_queue_setup_t tx_queue_setup;/**< Set up device TX queue.*/
1416 eth_queue_release_t tx_queue_release;/**< Release TX queue.*/
1417 eth_dev_led_on_t dev_led_on; /**< Turn on LED. */
1418 eth_dev_led_off_t dev_led_off; /**< Turn off LED. */
1419 flow_ctrl_get_t flow_ctrl_get; /**< Get flow control. */
1420 flow_ctrl_set_t flow_ctrl_set; /**< Setup flow control. */
1421 priority_flow_ctrl_set_t priority_flow_ctrl_set; /**< Setup priority flow control.*/
1422 eth_mac_addr_remove_t mac_addr_remove; /**< Remove MAC address */
1423 eth_mac_addr_add_t mac_addr_add; /**< Add a MAC address */
1424 eth_mac_addr_set_t mac_addr_set; /**< Set a MAC address */
1425 eth_uc_hash_table_set_t uc_hash_table_set; /**< Set Unicast Table Array */
1426 eth_uc_all_hash_table_set_t uc_all_hash_table_set; /**< Set Unicast hash bitmap */
1427 eth_mirror_rule_set_t mirror_rule_set; /**< Add a traffic mirror rule.*/
1428 eth_mirror_rule_reset_t mirror_rule_reset; /**< reset a traffic mirror rule.*/
1429 eth_set_vf_rx_mode_t set_vf_rx_mode; /**< Set VF RX mode */
1430 eth_set_vf_rx_t set_vf_rx; /**< enable/disable a VF receive */
1431 eth_set_vf_tx_t set_vf_tx; /**< enable/disable a VF transmit */
1432 eth_set_vf_vlan_filter_t set_vf_vlan_filter; /**< Set VF VLAN filter */
1433 eth_udp_tunnel_add_t udp_tunnel_add;
1434 eth_udp_tunnel_del_t udp_tunnel_del;
1435 eth_set_queue_rate_limit_t set_queue_rate_limit; /**< Set queue rate limit */
1436 eth_set_vf_rate_limit_t set_vf_rate_limit; /**< Set VF rate limit */
1438 /** Add a signature filter. */
1439 fdir_add_signature_filter_t fdir_add_signature_filter;
1440 /** Update a signature filter. */
1441 fdir_update_signature_filter_t fdir_update_signature_filter;
1442 /** Remove a signature filter. */
1443 fdir_remove_signature_filter_t fdir_remove_signature_filter;
1444 /** Get information about FDIR status. */
1445 fdir_infos_get_t fdir_infos_get;
1446 /** Add a perfect filter. */
1447 fdir_add_perfect_filter_t fdir_add_perfect_filter;
1448 /** Update a perfect filter. */
1449 fdir_update_perfect_filter_t fdir_update_perfect_filter;
1450 /** Remove a perfect filter. */
1451 fdir_remove_perfect_filter_t fdir_remove_perfect_filter;
1452 /** Setup masks for FDIR filtering. */
1453 fdir_set_masks_t fdir_set_masks;
1454 /** Update redirection table. */
1455 reta_update_t reta_update;
1456 /** Query redirection table. */
1457 reta_query_t reta_query;
1458 /* bypass control */
1459 #ifdef RTE_NIC_BYPASS
1460 bypass_init_t bypass_init;
1461 bypass_state_set_t bypass_state_set;
1462 bypass_state_show_t bypass_state_show;
1463 bypass_event_set_t bypass_event_set;
1464 bypass_event_show_t bypass_event_show;
1465 bypass_wd_timeout_set_t bypass_wd_timeout_set;
1466 bypass_wd_timeout_show_t bypass_wd_timeout_show;
1467 bypass_ver_show_t bypass_ver_show;
1468 bypass_wd_reset_t bypass_wd_reset;
1471 /** Configure RSS hash protocols. */
1472 rss_hash_update_t rss_hash_update;
1473 /** Get current RSS hash configuration. */
1474 rss_hash_conf_get_t rss_hash_conf_get;
1475 eth_add_2tuple_filter_t add_2tuple_filter; /**< add 2tuple filter. */
1476 eth_remove_2tuple_filter_t remove_2tuple_filter; /**< remove 2tuple filter. */
1477 eth_get_2tuple_filter_t get_2tuple_filter; /**< get 2tuple filter. */
1478 eth_add_5tuple_filter_t add_5tuple_filter; /**< add 5tuple filter. */
1479 eth_remove_5tuple_filter_t remove_5tuple_filter; /**< remove 5tuple filter. */
1480 eth_get_5tuple_filter_t get_5tuple_filter; /**< get 5tuple filter. */
1481 eth_filter_ctrl_t filter_ctrl; /**< common filter control*/
1486 * The generic data structure associated with each ethernet device.
1488 * Pointers to burst-oriented packet receive and transmit functions are
1489 * located at the beginning of the structure, along with the pointer to
1490 * where all the data elements for the particular device are stored in shared
1491 * memory. This split allows the function pointer and driver data to be per-
1492 * process, while the actual configuration data for the device is shared.
1494 struct rte_eth_dev {
1495 eth_rx_burst_t rx_pkt_burst; /**< Pointer to PMD receive function. */
1496 eth_tx_burst_t tx_pkt_burst; /**< Pointer to PMD transmit function. */
1497 struct rte_eth_dev_data *data; /**< Pointer to device data */
1498 const struct eth_driver *driver;/**< Driver for this device */
1499 struct eth_dev_ops *dev_ops; /**< Functions exported by PMD */
1500 struct rte_pci_device *pci_dev; /**< PCI info. supplied by probing */
1501 struct rte_eth_dev_cb_list callbacks; /**< User application callbacks */
1504 struct rte_eth_dev_sriov {
1505 uint8_t active; /**< SRIOV is active with 16, 32 or 64 pools */
1506 uint8_t nb_q_per_pool; /**< rx queue number per pool */
1507 uint16_t def_vmdq_idx; /**< Default pool num used for PF */
1508 uint16_t def_pool_q_idx; /**< Default pool queue start reg index */
1510 #define RTE_ETH_DEV_SRIOV(dev) ((dev)->data->sriov)
1512 #define RTE_ETH_NAME_MAX_LEN (32)
1516 * The data part, with no function pointers, associated with each ethernet device.
1518 * This structure is safe to place in shared memory to be common among different
1519 * processes in a multi-process configuration.
1521 struct rte_eth_dev_data {
1522 char name[RTE_ETH_NAME_MAX_LEN]; /**< Unique identifier name */
1524 void **rx_queues; /**< Array of pointers to RX queues. */
1525 void **tx_queues; /**< Array of pointers to TX queues. */
1526 uint16_t nb_rx_queues; /**< Number of RX queues. */
1527 uint16_t nb_tx_queues; /**< Number of TX queues. */
1529 struct rte_eth_dev_sriov sriov; /**< SRIOV data */
1531 void *dev_private; /**< PMD-specific private data */
1533 struct rte_eth_link dev_link;
1534 /**< Link-level information & status */
1536 struct rte_eth_conf dev_conf; /**< Configuration applied to device. */
1537 uint16_t mtu; /**< Maximum Transmission Unit. */
1539 uint32_t min_rx_buf_size;
1540 /**< Common rx buffer size handled by all queues */
1542 uint64_t rx_mbuf_alloc_failed; /**< RX ring mbuf allocation failures. */
1543 struct ether_addr* mac_addrs;/**< Device Ethernet Link address. */
1544 uint64_t mac_pool_sel[ETH_NUM_RECEIVE_MAC_ADDR];
1545 /** bitmap array of associating Ethernet MAC addresses to pools */
1546 struct ether_addr* hash_mac_addrs;
1547 /** Device Ethernet MAC addresses of hash filtering. */
1548 uint8_t port_id; /**< Device [external] port identifier. */
1549 uint8_t promiscuous : 1, /**< RX promiscuous mode ON(1) / OFF(0). */
1550 scattered_rx : 1, /**< RX of scattered packets is ON(1) / OFF(0) */
1551 all_multicast : 1, /**< RX all multicast mode ON(1) / OFF(0). */
1552 dev_started : 1; /**< Device state: STARTED(1) / STOPPED(0). */
1557 * The pool of *rte_eth_dev* structures. The size of the pool
1558 * is configured at compile-time in the <rte_ethdev.c> file.
1560 extern struct rte_eth_dev rte_eth_devices[];
1563 * Get the total number of Ethernet devices that have been successfully
1564 * initialized by the [matching] Ethernet driver during the PCI probing phase.
1565 * All devices whose port identifier is in the range
1566 * [0, rte_eth_dev_count() - 1] can be operated on by network applications.
1569 * - The total number of usable Ethernet devices.
1571 extern uint8_t rte_eth_dev_count(void);
1574 * Function for internal use by dummy drivers primarily, e.g. ring-based
1576 * Allocates a new ethdev slot for an ethernet device and returns the pointer
1577 * to that slot for the driver to use.
1579 * @param name Unique identifier name for each Ethernet device
1581 * - Slot in the rte_dev_devices array for a new device;
1583 struct rte_eth_dev *rte_eth_dev_allocate(const char *name);
1588 * Initialization function of an Ethernet driver invoked for each matching
1589 * Ethernet PCI device detected during the PCI probing phase.
1592 * The pointer to the [matching] Ethernet driver structure supplied by
1593 * the PMD when it registered itself.
1595 * The *eth_dev* pointer is the address of the *rte_eth_dev* structure
1596 * associated with the matching device and which have been [automatically]
1597 * allocated in the *rte_eth_devices* array.
1598 * The *eth_dev* structure is supplied to the driver initialization function
1599 * with the following fields already initialized:
1601 * - *pci_dev*: Holds the pointers to the *rte_pci_device* structure which
1602 * contains the generic PCI information of the matching device.
1604 * - *dev_private*: Holds a pointer to the device private data structure.
1606 * - *mtu*: Contains the default Ethernet maximum frame length (1500).
1608 * - *port_id*: Contains the port index of the device (actually the index
1609 * of the *eth_dev* structure in the *rte_eth_devices* array).
1612 * - 0: Success, the device is properly initialized by the driver.
1613 * In particular, the driver MUST have set up the *dev_ops* pointer
1614 * of the *eth_dev* structure.
1615 * - <0: Error code of the device initialization failure.
1617 typedef int (*eth_dev_init_t)(struct eth_driver *eth_drv,
1618 struct rte_eth_dev *eth_dev);
1622 * The structure associated with a PMD Ethernet driver.
1624 * Each Ethernet driver acts as a PCI driver and is represented by a generic
1625 * *eth_driver* structure that holds:
1627 * - An *rte_pci_driver* structure (which must be the first field).
1629 * - The *eth_dev_init* function invoked for each matching PCI device.
1631 * - The size of the private data to allocate for each matching device.
1634 struct rte_pci_driver pci_drv; /**< The PMD is also a PCI driver. */
1635 eth_dev_init_t eth_dev_init; /**< Device init function. */
1636 unsigned int dev_private_size; /**< Size of device private data. */
1641 * A function invoked by the initialization function of an Ethernet driver
1642 * to simultaneously register itself as a PCI driver and as an Ethernet
1643 * Poll Mode Driver (PMD).
1646 * The pointer to the *eth_driver* structure associated with
1647 * the Ethernet driver.
1649 extern void rte_eth_driver_register(struct eth_driver *eth_drv);
1652 * Configure an Ethernet device.
1653 * This function must be invoked first before any other function in the
1654 * Ethernet API. This function can also be re-invoked when a device is in the
1658 * The port identifier of the Ethernet device to configure.
1659 * @param nb_rx_queue
1660 * The number of receive queues to set up for the Ethernet device.
1661 * @param nb_tx_queue
1662 * The number of transmit queues to set up for the Ethernet device.
1664 * The pointer to the configuration data to be used for the Ethernet device.
1665 * The *rte_eth_conf* structure includes:
1666 * - the hardware offload features to activate, with dedicated fields for
1667 * each statically configurable offload hardware feature provided by
1668 * Ethernet devices, such as IP checksum or VLAN tag stripping for
1670 * - the Receive Side Scaling (RSS) configuration when using multiple RX
1673 * Embedding all configuration information in a single data structure
1674 * is the more flexible method that allows the addition of new features
1675 * without changing the syntax of the API.
1677 * - 0: Success, device configured.
1678 * - <0: Error code returned by the driver configuration function.
1680 extern int rte_eth_dev_configure(uint8_t port_id,
1681 uint16_t nb_rx_queue,
1682 uint16_t nb_tx_queue,
1683 const struct rte_eth_conf *eth_conf);
1686 * Allocate and set up a receive queue for an Ethernet device.
1688 * The function allocates a contiguous block of memory for *nb_rx_desc*
1689 * receive descriptors from a memory zone associated with *socket_id*
1690 * and initializes each receive descriptor with a network buffer allocated
1691 * from the memory pool *mb_pool*.
1694 * The port identifier of the Ethernet device.
1695 * @param rx_queue_id
1696 * The index of the receive queue to set up.
1697 * The value must be in the range [0, nb_rx_queue - 1] previously supplied
1698 * to rte_eth_dev_configure().
1700 * The number of receive descriptors to allocate for the receive ring.
1702 * The *socket_id* argument is the socket identifier in case of NUMA.
1703 * The value can be *SOCKET_ID_ANY* if there is no NUMA constraint for
1704 * the DMA memory allocated for the receive descriptors of the ring.
1706 * The pointer to the configuration data to be used for the receive queue.
1707 * NULL value is allowed, in which case default RX configuration
1709 * The *rx_conf* structure contains an *rx_thresh* structure with the values
1710 * of the Prefetch, Host, and Write-Back threshold registers of the receive
1713 * The pointer to the memory pool from which to allocate *rte_mbuf* network
1714 * memory buffers to populate each descriptor of the receive ring.
1716 * - 0: Success, receive queue correctly set up.
1717 * - -EINVAL: The size of network buffers which can be allocated from the
1718 * memory pool does not fit the various buffer sizes allowed by the
1719 * device controller.
1720 * - -ENOMEM: Unable to allocate the receive ring descriptors or to
1721 * allocate network memory buffers from the memory pool when
1722 * initializing receive descriptors.
1724 extern int rte_eth_rx_queue_setup(uint8_t port_id, uint16_t rx_queue_id,
1725 uint16_t nb_rx_desc, unsigned int socket_id,
1726 const struct rte_eth_rxconf *rx_conf,
1727 struct rte_mempool *mb_pool);
1730 * Allocate and set up a transmit queue for an Ethernet device.
1733 * The port identifier of the Ethernet device.
1734 * @param tx_queue_id
1735 * The index of the transmit queue to set up.
1736 * The value must be in the range [0, nb_tx_queue - 1] previously supplied
1737 * to rte_eth_dev_configure().
1739 * The number of transmit descriptors to allocate for the transmit ring.
1741 * The *socket_id* argument is the socket identifier in case of NUMA.
1742 * Its value can be *SOCKET_ID_ANY* if there is no NUMA constraint for
1743 * the DMA memory allocated for the transmit descriptors of the ring.
1745 * The pointer to the configuration data to be used for the transmit queue.
1746 * NULL value is allowed, in which case default RX configuration
1748 * The *tx_conf* structure contains the following data:
1749 * - The *tx_thresh* structure with the values of the Prefetch, Host, and
1750 * Write-Back threshold registers of the transmit ring.
1751 * When setting Write-Back threshold to the value greater then zero,
1752 * *tx_rs_thresh* value should be explicitly set to one.
1753 * - The *tx_free_thresh* value indicates the [minimum] number of network
1754 * buffers that must be pending in the transmit ring to trigger their
1755 * [implicit] freeing by the driver transmit function.
1756 * - The *tx_rs_thresh* value indicates the [minimum] number of transmit
1757 * descriptors that must be pending in the transmit ring before setting the
1758 * RS bit on a descriptor by the driver transmit function.
1759 * The *tx_rs_thresh* value should be less or equal then
1760 * *tx_free_thresh* value, and both of them should be less then
1762 * - The *txq_flags* member contains flags to pass to the TX queue setup
1763 * function to configure the behavior of the TX queue. This should be set
1764 * to 0 if no special configuration is required.
1766 * Note that setting *tx_free_thresh* or *tx_rs_thresh* value to 0 forces
1767 * the transmit function to use default values.
1769 * - 0: Success, the transmit queue is correctly set up.
1770 * - -ENOMEM: Unable to allocate the transmit ring descriptors.
1772 extern int rte_eth_tx_queue_setup(uint8_t port_id, uint16_t tx_queue_id,
1773 uint16_t nb_tx_desc, unsigned int socket_id,
1774 const struct rte_eth_txconf *tx_conf);
1777 * Return the NUMA socket to which an Ethernet device is connected
1780 * The port identifier of the Ethernet device
1782 * The NUMA socket id to which the Ethernet device is connected or
1783 * a default of zero if the socket could not be determined.
1784 * -1 is returned is the port_id value is out of range.
1786 extern int rte_eth_dev_socket_id(uint8_t port_id);
1789 * Allocate mbuf from mempool, setup the DMA physical address
1790 * and then start RX for specified queue of a port. It is used
1791 * when rx_deferred_start flag of the specified queue is true.
1794 * The port identifier of the Ethernet device
1795 * @param rx_queue_id
1796 * The index of the rx queue to update the ring.
1797 * The value must be in the range [0, nb_rx_queue - 1] previously supplied
1798 * to rte_eth_dev_configure().
1800 * - 0: Success, the transmit queue is correctly set up.
1801 * - -EINVAL: The port_id or the queue_id out of range.
1802 * - -ENOTSUP: The function not supported in PMD driver.
1804 extern int rte_eth_dev_rx_queue_start(uint8_t port_id, uint16_t rx_queue_id);
1807 * Stop specified RX queue of a port
1810 * The port identifier of the Ethernet device
1811 * @param rx_queue_id
1812 * The index of the rx queue to update the ring.
1813 * The value must be in the range [0, nb_rx_queue - 1] previously supplied
1814 * to rte_eth_dev_configure().
1816 * - 0: Success, the transmit queue is correctly set up.
1817 * - -EINVAL: The port_id or the queue_id out of range.
1818 * - -ENOTSUP: The function not supported in PMD driver.
1820 extern int rte_eth_dev_rx_queue_stop(uint8_t port_id, uint16_t rx_queue_id);
1823 * Start TX for specified queue of a port. It is used when tx_deferred_start
1824 * flag of the specified queue is true.
1827 * The port identifier of the Ethernet device
1828 * @param tx_queue_id
1829 * The index of the tx queue to update the ring.
1830 * The value must be in the range [0, nb_tx_queue - 1] previously supplied
1831 * to rte_eth_dev_configure().
1833 * - 0: Success, the transmit queue is correctly set up.
1834 * - -EINVAL: The port_id or the queue_id out of range.
1835 * - -ENOTSUP: The function not supported in PMD driver.
1837 extern int rte_eth_dev_tx_queue_start(uint8_t port_id, uint16_t tx_queue_id);
1840 * Stop specified TX queue of a port
1843 * The port identifier of the Ethernet device
1844 * @param tx_queue_id
1845 * The index of the tx queue to update the ring.
1846 * The value must be in the range [0, nb_tx_queue - 1] previously supplied
1847 * to rte_eth_dev_configure().
1849 * - 0: Success, the transmit queue is correctly set up.
1850 * - -EINVAL: The port_id or the queue_id out of range.
1851 * - -ENOTSUP: The function not supported in PMD driver.
1853 extern int rte_eth_dev_tx_queue_stop(uint8_t port_id, uint16_t tx_queue_id);
1858 * Start an Ethernet device.
1860 * The device start step is the last one and consists of setting the configured
1861 * offload features and in starting the transmit and the receive units of the
1863 * On success, all basic functions exported by the Ethernet API (link status,
1864 * receive/transmit, and so on) can be invoked.
1867 * The port identifier of the Ethernet device.
1869 * - 0: Success, Ethernet device started.
1870 * - <0: Error code of the driver device start function.
1872 extern int rte_eth_dev_start(uint8_t port_id);
1875 * Stop an Ethernet device. The device can be restarted with a call to
1876 * rte_eth_dev_start()
1879 * The port identifier of the Ethernet device.
1881 extern void rte_eth_dev_stop(uint8_t port_id);
1885 * Link up an Ethernet device.
1887 * Set device link up will re-enable the device rx/tx
1888 * functionality after it is previously set device linked down.
1891 * The port identifier of the Ethernet device.
1893 * - 0: Success, Ethernet device linked up.
1894 * - <0: Error code of the driver device link up function.
1896 extern int rte_eth_dev_set_link_up(uint8_t port_id);
1899 * Link down an Ethernet device.
1900 * The device rx/tx functionality will be disabled if success,
1901 * and it can be re-enabled with a call to
1902 * rte_eth_dev_set_link_up()
1905 * The port identifier of the Ethernet device.
1907 extern int rte_eth_dev_set_link_down(uint8_t port_id);
1910 * Close an Ethernet device. The device cannot be restarted!
1913 * The port identifier of the Ethernet device.
1915 extern void rte_eth_dev_close(uint8_t port_id);
1918 * Enable receipt in promiscuous mode for an Ethernet device.
1921 * The port identifier of the Ethernet device.
1923 extern void rte_eth_promiscuous_enable(uint8_t port_id);
1926 * Disable receipt in promiscuous mode for an Ethernet device.
1929 * The port identifier of the Ethernet device.
1931 extern void rte_eth_promiscuous_disable(uint8_t port_id);
1934 * Return the value of promiscuous mode for an Ethernet device.
1937 * The port identifier of the Ethernet device.
1939 * - (1) if promiscuous is enabled
1940 * - (0) if promiscuous is disabled.
1943 extern int rte_eth_promiscuous_get(uint8_t port_id);
1946 * Enable the receipt of any multicast frame by an Ethernet device.
1949 * The port identifier of the Ethernet device.
1951 extern void rte_eth_allmulticast_enable(uint8_t port_id);
1954 * Disable the receipt of all multicast frames by an Ethernet device.
1957 * The port identifier of the Ethernet device.
1959 extern void rte_eth_allmulticast_disable(uint8_t port_id);
1962 * Return the value of allmulticast mode for an Ethernet device.
1965 * The port identifier of the Ethernet device.
1967 * - (1) if allmulticast is enabled
1968 * - (0) if allmulticast is disabled.
1971 extern int rte_eth_allmulticast_get(uint8_t port_id);
1974 * Retrieve the status (ON/OFF), the speed (in Mbps) and the mode (HALF-DUPLEX
1975 * or FULL-DUPLEX) of the physical link of an Ethernet device. It might need
1976 * to wait up to 9 seconds in it.
1979 * The port identifier of the Ethernet device.
1981 * A pointer to an *rte_eth_link* structure to be filled with
1982 * the status, the speed and the mode of the Ethernet device link.
1984 extern void rte_eth_link_get(uint8_t port_id, struct rte_eth_link *link);
1987 * Retrieve the status (ON/OFF), the speed (in Mbps) and the mode (HALF-DUPLEX
1988 * or FULL-DUPLEX) of the physical link of an Ethernet device. It is a no-wait
1989 * version of rte_eth_link_get().
1992 * The port identifier of the Ethernet device.
1994 * A pointer to an *rte_eth_link* structure to be filled with
1995 * the status, the speed and the mode of the Ethernet device link.
1997 extern void rte_eth_link_get_nowait(uint8_t port_id,
1998 struct rte_eth_link *link);
2001 * Retrieve the general I/O statistics of an Ethernet device.
2004 * The port identifier of the Ethernet device.
2006 * A pointer to a structure of type *rte_eth_stats* to be filled with
2007 * the values of device counters for the following set of statistics:
2008 * - *ipackets* with the total of successfully received packets.
2009 * - *opackets* with the total of successfully transmitted packets.
2010 * - *ibytes* with the total of successfully received bytes.
2011 * - *obytes* with the total of successfully transmitted bytes.
2012 * - *ierrors* with the total of erroneous received packets.
2013 * - *oerrors* with the total of failed transmitted packets.
2015 * Zero if successful. Non-zero otherwise.
2017 extern int rte_eth_stats_get(uint8_t port_id, struct rte_eth_stats *stats);
2020 * Reset the general I/O statistics of an Ethernet device.
2023 * The port identifier of the Ethernet device.
2025 extern void rte_eth_stats_reset(uint8_t port_id);
2028 * Retrieve extended statistics of an Ethernet device.
2031 * The port identifier of the Ethernet device.
2033 * A pointer to a table of structure of type *rte_eth_xstats*
2034 * to be filled with device statistics names and values.
2035 * This parameter can be set to NULL if n is 0.
2037 * The size of the stats table, which should be large enough to store
2038 * all the statistics of the device.
2040 * - positive value lower or equal to n: success. The return value
2041 * is the number of entries filled in the stats table.
2042 * - positive value higher than n: error, the given statistics table
2043 * is too small. The return value corresponds to the size that should
2044 * be given to succeed. The entries in the table are not valid and
2045 * shall not be used by the caller.
2046 * - negative value on error (invalid port id)
2048 extern int rte_eth_xstats_get(uint8_t port_id,
2049 struct rte_eth_xstats *xstats, unsigned n);
2052 * Reset extended statistics of an Ethernet device.
2055 * The port identifier of the Ethernet device.
2057 extern void rte_eth_xstats_reset(uint8_t port_id);
2060 * Set a mapping for the specified transmit queue to the specified per-queue
2061 * statistics counter.
2064 * The port identifier of the Ethernet device.
2065 * @param tx_queue_id
2066 * The index of the transmit queue for which a queue stats mapping is required.
2067 * The value must be in the range [0, nb_tx_queue - 1] previously supplied
2068 * to rte_eth_dev_configure().
2070 * The per-queue packet statistics functionality number that the transmit
2071 * queue is to be assigned.
2072 * The value must be in the range [0, RTE_MAX_ETHPORT_QUEUE_STATS_MAPS - 1].
2074 * Zero if successful. Non-zero otherwise.
2076 extern int rte_eth_dev_set_tx_queue_stats_mapping(uint8_t port_id,
2077 uint16_t tx_queue_id,
2081 * Set a mapping for the specified receive queue to the specified per-queue
2082 * statistics counter.
2085 * The port identifier of the Ethernet device.
2086 * @param rx_queue_id
2087 * The index of the receive queue for which a queue stats mapping is required.
2088 * The value must be in the range [0, nb_rx_queue - 1] previously supplied
2089 * to rte_eth_dev_configure().
2091 * The per-queue packet statistics functionality number that the receive
2092 * queue is to be assigned.
2093 * The value must be in the range [0, RTE_MAX_ETHPORT_QUEUE_STATS_MAPS - 1].
2095 * Zero if successful. Non-zero otherwise.
2097 extern int rte_eth_dev_set_rx_queue_stats_mapping(uint8_t port_id,
2098 uint16_t rx_queue_id,
2102 * Retrieve the Ethernet address of an Ethernet device.
2105 * The port identifier of the Ethernet device.
2107 * A pointer to a structure of type *ether_addr* to be filled with
2108 * the Ethernet address of the Ethernet device.
2110 extern void rte_eth_macaddr_get(uint8_t port_id, struct ether_addr *mac_addr);
2113 * Retrieve the contextual information of an Ethernet device.
2116 * The port identifier of the Ethernet device.
2118 * A pointer to a structure of type *rte_eth_dev_info* to be filled with
2119 * the contextual information of the Ethernet device.
2121 extern void rte_eth_dev_info_get(uint8_t port_id,
2122 struct rte_eth_dev_info *dev_info);
2125 * Retrieve the MTU of an Ethernet device.
2128 * The port identifier of the Ethernet device.
2130 * A pointer to a uint16_t where the retrieved MTU is to be stored.
2132 * - (0) if successful.
2133 * - (-ENODEV) if *port_id* invalid.
2135 extern int rte_eth_dev_get_mtu(uint8_t port_id, uint16_t *mtu);
2138 * Change the MTU of an Ethernet device.
2141 * The port identifier of the Ethernet device.
2143 * A uint16_t for the MTU to be applied.
2145 * - (0) if successful.
2146 * - (-ENOTSUP) if operation is not supported.
2147 * - (-ENODEV) if *port_id* invalid.
2148 * - (-EINVAL) if *mtu* invalid.
2150 extern int rte_eth_dev_set_mtu(uint8_t port_id, uint16_t mtu);
2153 * Enable/Disable hardware filtering by an Ethernet device of received
2154 * VLAN packets tagged with a given VLAN Tag Identifier.
2157 * The port identifier of the Ethernet device.
2159 * The VLAN Tag Identifier whose filtering must be enabled or disabled.
2161 * If > 0, enable VLAN filtering of VLAN packets tagged with *vlan_id*.
2162 * Otherwise, disable VLAN filtering of VLAN packets tagged with *vlan_id*.
2164 * - (0) if successful.
2165 * - (-ENOSUP) if hardware-assisted VLAN filtering not configured.
2166 * - (-ENODEV) if *port_id* invalid.
2167 * - (-ENOSYS) if VLAN filtering on *port_id* disabled.
2168 * - (-EINVAL) if *vlan_id* > 4095.
2170 extern int rte_eth_dev_vlan_filter(uint8_t port_id, uint16_t vlan_id , int on);
2173 * Enable/Disable hardware VLAN Strip by a rx queue of an Ethernet device.
2174 * 82599/X540/X550 can support VLAN stripping at the rx queue level
2177 * The port identifier of the Ethernet device.
2178 * @param rx_queue_id
2179 * The index of the receive queue for which a queue stats mapping is required.
2180 * The value must be in the range [0, nb_rx_queue - 1] previously supplied
2181 * to rte_eth_dev_configure().
2183 * If 1, Enable VLAN Stripping of the receive queue of the Ethernet port.
2184 * If 0, Disable VLAN Stripping of the receive queue of the Ethernet port.
2186 * - (0) if successful.
2187 * - (-ENOSUP) if hardware-assisted VLAN stripping not configured.
2188 * - (-ENODEV) if *port_id* invalid.
2189 * - (-EINVAL) if *rx_queue_id* invalid.
2191 extern int rte_eth_dev_set_vlan_strip_on_queue(uint8_t port_id,
2192 uint16_t rx_queue_id, int on);
2195 * Set the Outer VLAN Ether Type by an Ethernet device, it can be inserted to
2196 * the VLAN Header. This is a register setup available on some Intel NIC, not
2197 * but all, please check the data sheet for availability.
2200 * The port identifier of the Ethernet device.
2202 * The Tag Protocol ID
2204 * - (0) if successful.
2205 * - (-ENOSUP) if hardware-assisted VLAN TPID setup is not supported.
2206 * - (-ENODEV) if *port_id* invalid.
2208 extern int rte_eth_dev_set_vlan_ether_type(uint8_t port_id, uint16_t tag_type);
2211 * Set VLAN offload configuration on an Ethernet device
2212 * Enable/Disable Extended VLAN by an Ethernet device, This is a register setup
2213 * available on some Intel NIC, not but all, please check the data sheet for
2215 * Enable/Disable VLAN Strip can be done on rx queue for certain NIC, but here
2216 * the configuration is applied on the port level.
2219 * The port identifier of the Ethernet device.
2220 * @param offload_mask
2221 * The VLAN Offload bit mask can be mixed use with "OR"
2222 * ETH_VLAN_STRIP_OFFLOAD
2223 * ETH_VLAN_FILTER_OFFLOAD
2224 * ETH_VLAN_EXTEND_OFFLOAD
2226 * - (0) if successful.
2227 * - (-ENOSUP) if hardware-assisted VLAN filtering not configured.
2228 * - (-ENODEV) if *port_id* invalid.
2230 extern int rte_eth_dev_set_vlan_offload(uint8_t port_id, int offload_mask);
2233 * Read VLAN Offload configuration from an Ethernet device
2236 * The port identifier of the Ethernet device.
2238 * - (>0) if successful. Bit mask to indicate
2239 * ETH_VLAN_STRIP_OFFLOAD
2240 * ETH_VLAN_FILTER_OFFLOAD
2241 * ETH_VLAN_EXTEND_OFFLOAD
2242 * - (-ENODEV) if *port_id* invalid.
2244 extern int rte_eth_dev_get_vlan_offload(uint8_t port_id);
2247 * Set port based TX VLAN insersion on or off.
2250 * The port identifier of the Ethernet device.
2252 * Port based TX VLAN identifier togeth with user priority.
2254 * Turn on or off the port based TX VLAN insertion.
2257 * - (0) if successful.
2258 * - negative if failed.
2260 extern int rte_eth_dev_set_vlan_pvid(uint8_t port_id, uint16_t pvid, int on);
2264 * Retrieve a burst of input packets from a receive queue of an Ethernet
2265 * device. The retrieved packets are stored in *rte_mbuf* structures whose
2266 * pointers are supplied in the *rx_pkts* array.
2268 * The rte_eth_rx_burst() function loops, parsing the RX ring of the
2269 * receive queue, up to *nb_pkts* packets, and for each completed RX
2270 * descriptor in the ring, it performs the following operations:
2272 * - Initialize the *rte_mbuf* data structure associated with the
2273 * RX descriptor according to the information provided by the NIC into
2274 * that RX descriptor.
2276 * - Store the *rte_mbuf* data structure into the next entry of the
2279 * - Replenish the RX descriptor with a new *rte_mbuf* buffer
2280 * allocated from the memory pool associated with the receive queue at
2281 * initialization time.
2283 * When retrieving an input packet that was scattered by the controller
2284 * into multiple receive descriptors, the rte_eth_rx_burst() function
2285 * appends the associated *rte_mbuf* buffers to the first buffer of the
2288 * The rte_eth_rx_burst() function returns the number of packets
2289 * actually retrieved, which is the number of *rte_mbuf* data structures
2290 * effectively supplied into the *rx_pkts* array.
2291 * A return value equal to *nb_pkts* indicates that the RX queue contained
2292 * at least *rx_pkts* packets, and this is likely to signify that other
2293 * received packets remain in the input queue. Applications implementing
2294 * a "retrieve as much received packets as possible" policy can check this
2295 * specific case and keep invoking the rte_eth_rx_burst() function until
2296 * a value less than *nb_pkts* is returned.
2298 * This receive method has the following advantages:
2300 * - It allows a run-to-completion network stack engine to retrieve and
2301 * to immediately process received packets in a fast burst-oriented
2302 * approach, avoiding the overhead of unnecessary intermediate packet
2303 * queue/dequeue operations.
2305 * - Conversely, it also allows an asynchronous-oriented processing
2306 * method to retrieve bursts of received packets and to immediately
2307 * queue them for further parallel processing by another logical core,
2308 * for instance. However, instead of having received packets being
2309 * individually queued by the driver, this approach allows the invoker
2310 * of the rte_eth_rx_burst() function to queue a burst of retrieved
2311 * packets at a time and therefore dramatically reduce the cost of
2312 * enqueue/dequeue operations per packet.
2314 * - It allows the rte_eth_rx_burst() function of the driver to take
2315 * advantage of burst-oriented hardware features (CPU cache,
2316 * prefetch instructions, and so on) to minimize the number of CPU
2317 * cycles per packet.
2319 * To summarize, the proposed receive API enables many
2320 * burst-oriented optimizations in both synchronous and asynchronous
2321 * packet processing environments with no overhead in both cases.
2323 * The rte_eth_rx_burst() function does not provide any error
2324 * notification to avoid the corresponding overhead. As a hint, the
2325 * upper-level application might check the status of the device link once
2326 * being systematically returned a 0 value for a given number of tries.
2329 * The port identifier of the Ethernet device.
2331 * The index of the receive queue from which to retrieve input packets.
2332 * The value must be in the range [0, nb_rx_queue - 1] previously supplied
2333 * to rte_eth_dev_configure().
2335 * The address of an array of pointers to *rte_mbuf* structures that
2336 * must be large enough to store *nb_pkts* pointers in it.
2338 * The maximum number of packets to retrieve.
2340 * The number of packets actually retrieved, which is the number
2341 * of pointers to *rte_mbuf* structures effectively supplied to the
2344 #ifdef RTE_LIBRTE_ETHDEV_DEBUG
2345 extern uint16_t rte_eth_rx_burst(uint8_t port_id, uint16_t queue_id,
2346 struct rte_mbuf **rx_pkts, uint16_t nb_pkts);
2348 static inline uint16_t
2349 rte_eth_rx_burst(uint8_t port_id, uint16_t queue_id,
2350 struct rte_mbuf **rx_pkts, uint16_t nb_pkts)
2352 struct rte_eth_dev *dev;
2354 dev = &rte_eth_devices[port_id];
2355 return (*dev->rx_pkt_burst)(dev->data->rx_queues[queue_id], rx_pkts, nb_pkts);
2360 * Get the number of used descriptors in a specific queue
2363 * The port identifier of the Ethernet device.
2365 * The queue id on the specific port.
2367 * The number of used descriptors in the specific queue.
2369 #ifdef RTE_LIBRTE_ETHDEV_DEBUG
2370 extern uint32_t rte_eth_rx_queue_count(uint8_t port_id, uint16_t queue_id);
2372 static inline uint32_t
2373 rte_eth_rx_queue_count(uint8_t port_id, uint16_t queue_id)
2375 struct rte_eth_dev *dev;
2377 dev = &rte_eth_devices[port_id];
2378 return (*dev->dev_ops->rx_queue_count)(dev, queue_id);
2383 * Check if the DD bit of the specific RX descriptor in the queue has been set
2386 * The port identifier of the Ethernet device.
2388 * The queue id on the specific port.
2390 * The offset of the descriptor ID from tail.
2392 * - (1) if the specific DD bit is set.
2393 * - (0) if the specific DD bit is not set.
2394 * - (-ENODEV) if *port_id* invalid.
2396 #ifdef RTE_LIBRTE_ETHDEV_DEBUG
2397 extern int rte_eth_rx_descriptor_done(uint8_t port_id,
2402 rte_eth_rx_descriptor_done(uint8_t port_id, uint16_t queue_id, uint16_t offset)
2404 struct rte_eth_dev *dev;
2406 dev = &rte_eth_devices[port_id];
2407 return (*dev->dev_ops->rx_descriptor_done)( \
2408 dev->data->rx_queues[queue_id], offset);
2413 * Send a burst of output packets on a transmit queue of an Ethernet device.
2415 * The rte_eth_tx_burst() function is invoked to transmit output packets
2416 * on the output queue *queue_id* of the Ethernet device designated by its
2418 * The *nb_pkts* parameter is the number of packets to send which are
2419 * supplied in the *tx_pkts* array of *rte_mbuf* structures.
2420 * The rte_eth_tx_burst() function loops, sending *nb_pkts* packets,
2421 * up to the number of transmit descriptors available in the TX ring of the
2423 * For each packet to send, the rte_eth_tx_burst() function performs
2424 * the following operations:
2426 * - Pick up the next available descriptor in the transmit ring.
2428 * - Free the network buffer previously sent with that descriptor, if any.
2430 * - Initialize the transmit descriptor with the information provided
2431 * in the *rte_mbuf data structure.
2433 * In the case of a segmented packet composed of a list of *rte_mbuf* buffers,
2434 * the rte_eth_tx_burst() function uses several transmit descriptors
2437 * The rte_eth_tx_burst() function returns the number of packets it
2438 * actually sent. A return value equal to *nb_pkts* means that all packets
2439 * have been sent, and this is likely to signify that other output packets
2440 * could be immediately transmitted again. Applications that implement a
2441 * "send as many packets to transmit as possible" policy can check this
2442 * specific case and keep invoking the rte_eth_tx_burst() function until
2443 * a value less than *nb_pkts* is returned.
2445 * It is the responsibility of the rte_eth_tx_burst() function to
2446 * transparently free the memory buffers of packets previously sent.
2447 * This feature is driven by the *tx_free_thresh* value supplied to the
2448 * rte_eth_dev_configure() function at device configuration time.
2449 * When the number of previously sent packets reached the "minimum transmit
2450 * packets to free" threshold, the rte_eth_tx_burst() function must
2451 * [attempt to] free the *rte_mbuf* buffers of those packets whose
2452 * transmission was effectively completed.
2455 * The port identifier of the Ethernet device.
2457 * The index of the transmit queue through which output packets must be
2459 * The value must be in the range [0, nb_tx_queue - 1] previously supplied
2460 * to rte_eth_dev_configure().
2462 * The address of an array of *nb_pkts* pointers to *rte_mbuf* structures
2463 * which contain the output packets.
2465 * The maximum number of packets to transmit.
2467 * The number of output packets actually stored in transmit descriptors of
2468 * the transmit ring. The return value can be less than the value of the
2469 * *tx_pkts* parameter when the transmit ring is full or has been filled up.
2471 #ifdef RTE_LIBRTE_ETHDEV_DEBUG
2472 extern uint16_t rte_eth_tx_burst(uint8_t port_id, uint16_t queue_id,
2473 struct rte_mbuf **tx_pkts, uint16_t nb_pkts);
2475 static inline uint16_t
2476 rte_eth_tx_burst(uint8_t port_id, uint16_t queue_id,
2477 struct rte_mbuf **tx_pkts, uint16_t nb_pkts)
2479 struct rte_eth_dev *dev;
2481 dev = &rte_eth_devices[port_id];
2482 return (*dev->tx_pkt_burst)(dev->data->tx_queues[queue_id], tx_pkts, nb_pkts);
2487 * Setup a new signature filter rule on an Ethernet device
2490 * The port identifier of the Ethernet device.
2491 * @param fdir_filter
2492 * The pointer to the fdir filter structure describing the signature filter
2494 * The *rte_fdir_filter* structure includes the values of the different fields
2495 * to match: source and destination IP addresses, vlan id, flexbytes, source
2496 * and destination ports, and so on.
2498 * The index of the RX queue where to store RX packets matching the added
2499 * signature filter defined in fdir_filter.
2501 * - (0) if successful.
2502 * - (-ENOTSUP) if hardware doesn't support flow director mode.
2503 * - (-ENODEV) if *port_id* invalid.
2504 * - (-ENOSYS) if the FDIR mode is not configured in signature mode
2506 * - (-EINVAL) if the fdir_filter information is not correct.
2508 int rte_eth_dev_fdir_add_signature_filter(uint8_t port_id,
2509 struct rte_fdir_filter *fdir_filter,
2513 * Update a signature filter rule on an Ethernet device.
2514 * If the rule doesn't exits, it is created.
2517 * The port identifier of the Ethernet device.
2519 * The pointer to the structure describing the signature filter rule.
2520 * The *rte_fdir_filter* structure includes the values of the different fields
2521 * to match: source and destination IP addresses, vlan id, flexbytes, source
2522 * and destination ports, and so on.
2524 * The index of the RX queue where to store RX packets matching the added
2525 * signature filter defined in fdir_ftr.
2527 * - (0) if successful.
2528 * - (-ENOTSUP) if hardware doesn't support flow director mode.
2529 * - (-ENODEV) if *port_id* invalid.
2530 * - (-ENOSYS) if the flow director mode is not configured in signature mode
2532 * - (-EINVAL) if the fdir_filter information is not correct.
2534 int rte_eth_dev_fdir_update_signature_filter(uint8_t port_id,
2535 struct rte_fdir_filter *fdir_ftr,
2539 * Remove a signature filter rule on an Ethernet device.
2542 * The port identifier of the Ethernet device.
2544 * The pointer to the structure describing the signature filter rule.
2545 * The *rte_fdir_filter* structure includes the values of the different fields
2546 * to match: source and destination IP addresses, vlan id, flexbytes, source
2547 * and destination ports, and so on.
2549 * - (0) if successful.
2550 * - (-ENOTSUP) if hardware doesn't support flow director mode.
2551 * - (-ENODEV) if *port_id* invalid.
2552 * - (-ENOSYS) if the flow director mode is not configured in signature mode
2554 * - (-EINVAL) if the fdir_filter information is not correct.
2556 int rte_eth_dev_fdir_remove_signature_filter(uint8_t port_id,
2557 struct rte_fdir_filter *fdir_ftr);
2560 * Retrieve the flow director information of an Ethernet device.
2563 * The port identifier of the Ethernet device.
2565 * A pointer to a structure of type *rte_eth_dev_fdir* to be filled with
2566 * the flow director information of the Ethernet device.
2568 * - (0) if successful.
2569 * - (-ENOTSUP) if hardware doesn't support flow director mode.
2570 * - (-ENODEV) if *port_id* invalid.
2571 * - (-ENOSYS) if the flow director mode is not configured on *port_id*.
2573 int rte_eth_dev_fdir_get_infos(uint8_t port_id, struct rte_eth_fdir *fdir);
2576 * Add a new perfect filter rule on an Ethernet device.
2579 * The port identifier of the Ethernet device.
2580 * @param fdir_filter
2581 * The pointer to the structure describing the perfect filter rule.
2582 * The *rte_fdir_filter* structure includes the values of the different fields
2583 * to match: source and destination IP addresses, vlan id, flexbytes, source
2584 * and destination ports, and so on.
2585 * IPv6 are not supported.
2587 * The 16-bit value supplied in the field hash.fdir.id of mbuf for RX
2588 * packets matching the perfect filter.
2590 * The index of the RX queue where to store RX packets matching the added
2591 * perfect filter defined in fdir_filter.
2593 * If drop is set to 1, matching RX packets are stored into the RX drop
2594 * queue defined in the rte_fdir_conf.
2596 * - (0) if successful.
2597 * - (-ENOTSUP) if hardware doesn't support flow director mode.
2598 * - (-ENODEV) if *port_id* invalid.
2599 * - (-ENOSYS) if the flow director mode is not configured in perfect mode
2601 * - (-EINVAL) if the fdir_filter information is not correct.
2603 int rte_eth_dev_fdir_add_perfect_filter(uint8_t port_id,
2604 struct rte_fdir_filter *fdir_filter,
2605 uint16_t soft_id, uint8_t rx_queue,
2609 * Update a perfect filter rule on an Ethernet device.
2610 * If the rule doesn't exits, it is created.
2613 * The port identifier of the Ethernet device.
2614 * @param fdir_filter
2615 * The pointer to the structure describing the perfect filter rule.
2616 * The *rte_fdir_filter* structure includes the values of the different fields
2617 * to match: source and destination IP addresses, vlan id, flexbytes, source
2618 * and destination ports, and so on.
2619 * IPv6 are not supported.
2621 * The 16-bit value supplied in the field hash.fdir.id of mbuf for RX
2622 * packets matching the perfect filter.
2624 * The index of the RX queue where to store RX packets matching the added
2625 * perfect filter defined in fdir_filter.
2627 * If drop is set to 1, matching RX packets are stored into the RX drop
2628 * queue defined in the rte_fdir_conf.
2630 * - (0) if successful.
2631 * - (-ENOTSUP) if hardware doesn't support flow director mode.
2632 * - (-ENODEV) if *port_id* invalid.
2633 * - (-ENOSYS) if the flow director mode is not configured in perfect mode
2635 * - (-EINVAL) if the fdir_filter information is not correct.
2637 int rte_eth_dev_fdir_update_perfect_filter(uint8_t port_id,
2638 struct rte_fdir_filter *fdir_filter,
2639 uint16_t soft_id, uint8_t rx_queue,
2643 * Remove a perfect filter rule on an Ethernet device.
2646 * The port identifier of the Ethernet device.
2647 * @param fdir_filter
2648 * The pointer to the structure describing the perfect filter rule.
2649 * The *rte_fdir_filter* structure includes the values of the different fields
2650 * to match: source and destination IP addresses, vlan id, flexbytes, source
2651 * and destination ports, and so on.
2652 * IPv6 are not supported.
2654 * The soft_id value provided when adding/updating the removed filter.
2656 * - (0) if successful.
2657 * - (-ENOTSUP) if hardware doesn't support flow director mode.
2658 * - (-ENODEV) if *port_id* invalid.
2659 * - (-ENOSYS) if the flow director mode is not configured in perfect mode
2661 * - (-EINVAL) if the fdir_filter information is not correct.
2663 int rte_eth_dev_fdir_remove_perfect_filter(uint8_t port_id,
2664 struct rte_fdir_filter *fdir_filter,
2667 * Configure globally the masks for flow director mode for an Ethernet device.
2668 * For example, the device can match packets with only the first 24 bits of
2669 * the IPv4 source address.
2671 * The following fields can be masked: IPv4 addresses and L4 port numbers.
2672 * The following fields can be either enabled or disabled completely for the
2673 * matching functionality: VLAN ID tag; VLAN Priority + CFI bit; Flexible 2-byte
2675 * IPv6 masks are not supported.
2677 * All filters must comply with the masks previously configured.
2678 * For example, with a mask equal to 255.255.255.0 for the source IPv4 address,
2679 * all IPv4 filters must be created with a source IPv4 address that fits the
2682 * This function flushes all filters that have been previously added in
2686 * The port identifier of the Ethernet device.
2688 * The pointer to the fdir mask structure describing relevant headers fields
2689 * and relevant bits to use when matching packets addresses and ports.
2690 * IPv6 masks are not supported.
2692 * - (0) if successful.
2693 * - (-ENOTSUP) if hardware doesn't support flow director mode.
2694 * - (-ENODEV) if *port_id* invalid.
2695 * - (-ENOSYS) if the flow director mode is not configured in perfect
2696 * mode on *port_id*.
2697 * - (-EINVAL) if the fdir_filter information is not correct
2699 int rte_eth_dev_fdir_set_masks(uint8_t port_id,
2700 struct rte_fdir_masks *fdir_mask);
2703 * The eth device event type for interrupt, and maybe others in the future.
2705 enum rte_eth_event_type {
2706 RTE_ETH_EVENT_UNKNOWN, /**< unknown event type */
2707 RTE_ETH_EVENT_INTR_LSC, /**< lsc interrupt event */
2708 RTE_ETH_EVENT_MAX /**< max value of this enum */
2711 typedef void (*rte_eth_dev_cb_fn)(uint8_t port_id, \
2712 enum rte_eth_event_type event, void *cb_arg);
2713 /**< user application callback to be registered for interrupts */
2718 * Register a callback function for specific port id.
2725 * User supplied callback function to be called.
2727 * Pointer to the parameters for the registered callback.
2730 * - On success, zero.
2731 * - On failure, a negative value.
2733 int rte_eth_dev_callback_register(uint8_t port_id,
2734 enum rte_eth_event_type event,
2735 rte_eth_dev_cb_fn cb_fn, void *cb_arg);
2738 * Unregister a callback function for specific port id.
2745 * User supplied callback function to be called.
2747 * Pointer to the parameters for the registered callback. -1 means to
2748 * remove all for the same callback address and same event.
2751 * - On success, zero.
2752 * - On failure, a negative value.
2754 int rte_eth_dev_callback_unregister(uint8_t port_id,
2755 enum rte_eth_event_type event,
2756 rte_eth_dev_cb_fn cb_fn, void *cb_arg);
2759 * @internal Executes all the user application registered callbacks for
2760 * the specific device. It is for DPDK internal user only. User
2761 * application should not call it directly.
2764 * Pointer to struct rte_eth_dev.
2766 * Eth device interrupt event type.
2771 void _rte_eth_dev_callback_process(struct rte_eth_dev *dev,
2772 enum rte_eth_event_type event);
2775 * Turn on the LED on the Ethernet device.
2776 * This function turns on the LED on the Ethernet device.
2779 * The port identifier of the Ethernet device.
2781 * - (0) if successful.
2782 * - (-ENOTSUP) if underlying hardware OR driver doesn't support
2784 * - (-ENODEV) if *port_id* invalid.
2786 int rte_eth_led_on(uint8_t port_id);
2789 * Turn off the LED on the Ethernet device.
2790 * This function turns off the LED on the Ethernet device.
2793 * The port identifier of the Ethernet device.
2795 * - (0) if successful.
2796 * - (-ENOTSUP) if underlying hardware OR driver doesn't support
2798 * - (-ENODEV) if *port_id* invalid.
2800 int rte_eth_led_off(uint8_t port_id);
2803 * Get current status of the Ethernet link flow control for Ethernet device
2806 * The port identifier of the Ethernet device.
2808 * The pointer to the structure where to store the flow control parameters.
2810 * - (0) if successful.
2811 * - (-ENOTSUP) if hardware doesn't support flow control.
2812 * - (-ENODEV) if *port_id* invalid.
2814 int rte_eth_dev_flow_ctrl_get(uint8_t port_id,
2815 struct rte_eth_fc_conf *fc_conf);
2818 * Configure the Ethernet link flow control for Ethernet device
2821 * The port identifier of the Ethernet device.
2823 * The pointer to the structure of the flow control parameters.
2825 * - (0) if successful.
2826 * - (-ENOTSUP) if hardware doesn't support flow control mode.
2827 * - (-ENODEV) if *port_id* invalid.
2828 * - (-EINVAL) if bad parameter
2829 * - (-EIO) if flow control setup failure
2831 int rte_eth_dev_flow_ctrl_set(uint8_t port_id,
2832 struct rte_eth_fc_conf *fc_conf);
2835 * Configure the Ethernet priority flow control under DCB environment
2836 * for Ethernet device.
2839 * The port identifier of the Ethernet device.
2841 * The pointer to the structure of the priority flow control parameters.
2843 * - (0) if successful.
2844 * - (-ENOTSUP) if hardware doesn't support priority flow control mode.
2845 * - (-ENODEV) if *port_id* invalid.
2846 * - (-EINVAL) if bad parameter
2847 * - (-EIO) if flow control setup failure
2849 int rte_eth_dev_priority_flow_ctrl_set(uint8_t port_id,
2850 struct rte_eth_pfc_conf *pfc_conf);
2853 * Add a MAC address to an internal array of addresses used to enable whitelist
2854 * filtering to accept packets only if the destination MAC address matches.
2857 * The port identifier of the Ethernet device.
2859 * The MAC address to add.
2861 * VMDq pool index to associate address with (if VMDq is enabled). If VMDq is
2862 * not enabled, this should be set to 0.
2864 * - (0) if successfully added or *mac_addr" was already added.
2865 * - (-ENOTSUP) if hardware doesn't support this feature.
2866 * - (-ENODEV) if *port* is invalid.
2867 * - (-ENOSPC) if no more MAC addresses can be added.
2868 * - (-EINVAL) if MAC address is invalid.
2870 int rte_eth_dev_mac_addr_add(uint8_t port, struct ether_addr *mac_addr,
2874 * Remove a MAC address from the internal array of addresses.
2877 * The port identifier of the Ethernet device.
2879 * MAC address to remove.
2881 * - (0) if successful, or *mac_addr* didn't exist.
2882 * - (-ENOTSUP) if hardware doesn't support.
2883 * - (-ENODEV) if *port* invalid.
2884 * - (-EADDRINUSE) if attempting to remove the default MAC address
2886 int rte_eth_dev_mac_addr_remove(uint8_t port, struct ether_addr *mac_addr);
2889 * Update Redirection Table(RETA) of Receive Side Scaling of Ethernet device.
2892 * The port identifier of the Ethernet device.
2896 * Redirection table size. The table size can be queried by
2897 * rte_eth_dev_info_get().
2899 * - (0) if successful.
2900 * - (-ENOTSUP) if hardware doesn't support.
2901 * - (-EINVAL) if bad parameter.
2903 int rte_eth_dev_rss_reta_update(uint8_t port,
2904 struct rte_eth_rss_reta_entry64 *reta_conf,
2905 uint16_t reta_size);
2908 * Query Redirection Table(RETA) of Receive Side Scaling of Ethernet device.
2911 * The port identifier of the Ethernet device.
2915 * Redirection table size. The table size can be queried by
2916 * rte_eth_dev_info_get().
2918 * - (0) if successful.
2919 * - (-ENOTSUP) if hardware doesn't support.
2920 * - (-EINVAL) if bad parameter.
2922 int rte_eth_dev_rss_reta_query(uint8_t port,
2923 struct rte_eth_rss_reta_entry64 *reta_conf,
2924 uint16_t reta_size);
2927 * Updates unicast hash table for receiving packet with the given destination
2928 * MAC address, and the packet is routed to all VFs for which the RX mode is
2929 * accept packets that match the unicast hash table.
2932 * The port identifier of the Ethernet device.
2934 * Unicast MAC address.
2936 * 1 - Set an unicast hash bit for receiving packets with the MAC address.
2937 * 0 - Clear an unicast hash bit.
2939 * - (0) if successful.
2940 * - (-ENOTSUP) if hardware doesn't support.
2941 * - (-ENODEV) if *port_id* invalid.
2942 * - (-EINVAL) if bad parameter.
2944 int rte_eth_dev_uc_hash_table_set(uint8_t port,struct ether_addr *addr,
2948 * Updates all unicast hash bitmaps for receiving packet with any Unicast
2949 * Ethernet MAC addresses,the packet is routed to all VFs for which the RX
2950 * mode is accept packets that match the unicast hash table.
2953 * The port identifier of the Ethernet device.
2955 * 1 - Set all unicast hash bitmaps for receiving all the Ethernet
2957 * 0 - Clear all unicast hash bitmaps
2959 * - (0) if successful.
2960 * - (-ENOTSUP) if hardware doesn't support.
2961 * - (-ENODEV) if *port_id* invalid.
2962 * - (-EINVAL) if bad parameter.
2964 int rte_eth_dev_uc_all_hash_table_set(uint8_t port,uint8_t on);
2967 * Set RX L2 Filtering mode of a VF of an Ethernet device.
2970 * The port identifier of the Ethernet device.
2974 * The RX mode mask, which is one or more of accepting Untagged Packets,
2975 * packets that match the PFUTA table, Broadcast and Multicast Promiscuous.
2976 * ETH_VMDQ_ACCEPT_UNTAG,ETH_VMDQ_ACCEPT_HASH_UC,
2977 * ETH_VMDQ_ACCEPT_BROADCAST and ETH_VMDQ_ACCEPT_MULTICAST will be used
2980 * 1 - Enable a VF RX mode.
2981 * 0 - Disable a VF RX mode.
2983 * - (0) if successful.
2984 * - (-ENOTSUP) if hardware doesn't support.
2985 * - (-ENOTSUP) if hardware doesn't support.
2986 * - (-EINVAL) if bad parameter.
2988 int rte_eth_dev_set_vf_rxmode(uint8_t port, uint16_t vf, uint16_t rx_mode,
2992 * Enable or disable a VF traffic transmit of the Ethernet device.
2995 * The port identifier of the Ethernet device.
2999 * 1 - Enable a VF traffic transmit.
3000 * 0 - Disable a VF traffic transmit.
3002 * - (0) if successful.
3003 * - (-ENODEV) if *port_id* invalid.
3004 * - (-ENOTSUP) if hardware doesn't support.
3005 * - (-EINVAL) if bad parameter.
3008 rte_eth_dev_set_vf_tx(uint8_t port,uint16_t vf, uint8_t on);
3011 * Enable or disable a VF traffic receive of an Ethernet device.
3014 * The port identifier of the Ethernet device.
3018 * 1 - Enable a VF traffic receive.
3019 * 0 - Disable a VF traffic receive.
3021 * - (0) if successful.
3022 * - (-ENOTSUP) if hardware doesn't support.
3023 * - (-ENODEV) if *port_id* invalid.
3024 * - (-EINVAL) if bad parameter.
3027 rte_eth_dev_set_vf_rx(uint8_t port,uint16_t vf, uint8_t on);
3030 * Enable/Disable hardware VF VLAN filtering by an Ethernet device of
3031 * received VLAN packets tagged with a given VLAN Tag Identifier.
3034 * The port identifier of the Ethernet device.
3036 * The VLAN Tag Identifier whose filtering must be enabled or disabled.
3038 * Bitmap listing which VFs participate in the VLAN filtering.
3040 * 1 - Enable VFs VLAN filtering.
3041 * 0 - Disable VFs VLAN filtering.
3043 * - (0) if successful.
3044 * - (-ENOTSUP) if hardware doesn't support.
3045 * - (-ENODEV) if *port_id* invalid.
3046 * - (-EINVAL) if bad parameter.
3049 rte_eth_dev_set_vf_vlan_filter(uint8_t port, uint16_t vlan_id,
3054 * Set a traffic mirroring rule on an Ethernet device
3057 * The port identifier of the Ethernet device.
3058 * @param mirror_conf
3059 * The pointer to the traffic mirroring structure describing the mirroring rule.
3060 * The *rte_eth_vm_mirror_conf* structure includes the type of mirroring rule,
3061 * destination pool and the value of rule if enable vlan or pool mirroring.
3064 * The index of traffic mirroring rule, we support four separated rules.
3066 * 1 - Enable a mirroring rule.
3067 * 0 - Disable a mirroring rule.
3069 * - (0) if successful.
3070 * - (-ENOTSUP) if hardware doesn't support this feature.
3071 * - (-ENODEV) if *port_id* invalid.
3072 * - (-EINVAL) if the mr_conf information is not correct.
3074 int rte_eth_mirror_rule_set(uint8_t port_id,
3075 struct rte_eth_vmdq_mirror_conf *mirror_conf,
3080 * Reset a traffic mirroring rule on an Ethernet device.
3083 * The port identifier of the Ethernet device.
3085 * The index of traffic mirroring rule, we support four separated rules.
3087 * - (0) if successful.
3088 * - (-ENOTSUP) if hardware doesn't support this feature.
3089 * - (-ENODEV) if *port_id* invalid.
3090 * - (-EINVAL) if bad parameter.
3092 int rte_eth_mirror_rule_reset(uint8_t port_id,
3096 * Set the rate limitation for a queue on an Ethernet device.
3099 * The port identifier of the Ethernet device.
3103 * The tx rate allocated from the total link speed for this queue.
3105 * - (0) if successful.
3106 * - (-ENOTSUP) if hardware doesn't support this feature.
3107 * - (-ENODEV) if *port_id* invalid.
3108 * - (-EINVAL) if bad parameter.
3110 int rte_eth_set_queue_rate_limit(uint8_t port_id, uint16_t queue_idx,
3114 * Set the rate limitation for a vf on an Ethernet device.
3117 * The port identifier of the Ethernet device.
3121 * The tx rate allocated from the total link speed for this VF id.
3123 * The queue mask which need to set the rate.
3125 * - (0) if successful.
3126 * - (-ENOTSUP) if hardware doesn't support this feature.
3127 * - (-ENODEV) if *port_id* invalid.
3128 * - (-EINVAL) if bad parameter.
3130 int rte_eth_set_vf_rate_limit(uint8_t port_id, uint16_t vf,
3131 uint16_t tx_rate, uint64_t q_msk);
3134 * Initialize bypass logic. This function needs to be called before
3135 * executing any other bypass API.
3138 * The port identifier of the Ethernet device.
3140 * - (0) if successful.
3141 * - (-ENOTSUP) if hardware doesn't support.
3142 * - (-EINVAL) if bad parameter.
3144 int rte_eth_dev_bypass_init(uint8_t port);
3147 * Return bypass state.
3150 * The port identifier of the Ethernet device.
3152 * The return bypass state.
3155 * - (3) Isolate mode
3157 * - (0) if successful.
3158 * - (-ENOTSUP) if hardware doesn't support.
3159 * - (-EINVAL) if bad parameter.
3161 int rte_eth_dev_bypass_state_show(uint8_t port, uint32_t *state);
3167 * The port identifier of the Ethernet device.
3169 * The current bypass state.
3172 * - (3) Isolate mode
3174 * - (0) if successful.
3175 * - (-ENOTSUP) if hardware doesn't support.
3176 * - (-EINVAL) if bad parameter.
3178 int rte_eth_dev_bypass_state_set(uint8_t port, uint32_t *new_state);
3181 * Return bypass state when given event occurs.
3184 * The port identifier of the Ethernet device.
3187 * - (1) Main power on (power button is pushed)
3188 * - (2) Auxiliary power on (power supply is being plugged)
3189 * - (3) Main power off (system shutdown and power supply is left plugged in)
3190 * - (4) Auxiliary power off (power supply is being unplugged)
3191 * - (5) Display or set the watchdog timer
3193 * The bypass state when given event occurred.
3196 * - (3) Isolate mode
3198 * - (0) if successful.
3199 * - (-ENOTSUP) if hardware doesn't support.
3200 * - (-EINVAL) if bad parameter.
3202 int rte_eth_dev_bypass_event_show(uint8_t port, uint32_t event, uint32_t *state);
3205 * Set bypass state when given event occurs.
3208 * The port identifier of the Ethernet device.
3211 * - (1) Main power on (power button is pushed)
3212 * - (2) Auxiliary power on (power supply is being plugged)
3213 * - (3) Main power off (system shutdown and power supply is left plugged in)
3214 * - (4) Auxiliary power off (power supply is being unplugged)
3215 * - (5) Display or set the watchdog timer
3217 * The assigned state when given event occurs.
3220 * - (3) Isolate mode
3222 * - (0) if successful.
3223 * - (-ENOTSUP) if hardware doesn't support.
3224 * - (-EINVAL) if bad parameter.
3226 int rte_eth_dev_bypass_event_store(uint8_t port, uint32_t event, uint32_t state);
3229 * Set bypass watchdog timeout count.
3232 * The port identifier of the Ethernet device.
3234 * The timeout to be set.
3235 * - (0) 0 seconds (timer is off)
3244 * - (0) if successful.
3245 * - (-ENOTSUP) if hardware doesn't support.
3246 * - (-EINVAL) if bad parameter.
3248 int rte_eth_dev_wd_timeout_store(uint8_t port, uint32_t timeout);
3251 * Get bypass firmware version.
3254 * The port identifier of the Ethernet device.
3256 * The firmware version
3258 * - (0) if successful.
3259 * - (-ENOTSUP) if hardware doesn't support.
3260 * - (-EINVAL) if bad parameter.
3262 int rte_eth_dev_bypass_ver_show(uint8_t port, uint32_t *ver);
3265 * Return bypass watchdog timeout in seconds
3268 * The port identifier of the Ethernet device.
3270 * The return watchdog timeout. "0" represents timer expired
3271 * - (0) 0 seconds (timer is off)
3280 * - (0) if successful.
3281 * - (-ENOTSUP) if hardware doesn't support.
3282 * - (-EINVAL) if bad parameter.
3284 int rte_eth_dev_bypass_wd_timeout_show(uint8_t port, uint32_t *wd_timeout);
3287 * Reset bypass watchdog timer
3290 * The port identifier of the Ethernet device.
3292 * - (0) if successful.
3293 * - (-ENOTSUP) if hardware doesn't support.
3294 * - (-EINVAL) if bad parameter.
3296 int rte_eth_dev_bypass_wd_reset(uint8_t port);
3299 * Configuration of Receive Side Scaling hash computation of Ethernet device.
3302 * The port identifier of the Ethernet device.
3304 * The new configuration to use for RSS hash computation on the port.
3306 * - (0) if successful.
3307 * - (-ENODEV) if port identifier is invalid.
3308 * - (-ENOTSUP) if hardware doesn't support.
3309 * - (-EINVAL) if bad parameter.
3311 int rte_eth_dev_rss_hash_update(uint8_t port_id,
3312 struct rte_eth_rss_conf *rss_conf);
3315 * Retrieve current configuration of Receive Side Scaling hash computation
3316 * of Ethernet device.
3319 * The port identifier of the Ethernet device.
3321 * Where to store the current RSS hash configuration of the Ethernet device.
3323 * - (0) if successful.
3324 * - (-ENODEV) if port identifier is invalid.
3325 * - (-ENOTSUP) if hardware doesn't support RSS.
3328 rte_eth_dev_rss_hash_conf_get(uint8_t port_id,
3329 struct rte_eth_rss_conf *rss_conf);
3332 * Add UDP tunneling port of an Ethernet device for filtering a specific
3333 * tunneling packet by UDP port number.
3336 * The port identifier of the Ethernet device.
3338 * UDP tunneling configuration.
3341 * - (0) if successful.
3342 * - (-ENODEV) if port identifier is invalid.
3343 * - (-ENOTSUP) if hardware doesn't support tunnel type.
3346 rte_eth_dev_udp_tunnel_add(uint8_t port_id,
3347 struct rte_eth_udp_tunnel *tunnel_udp);
3350 * Detete UDP tunneling port configuration of Ethernet device
3353 * The port identifier of the Ethernet device.
3355 * UDP tunneling configuration.
3358 * - (0) if successful.
3359 * - (-ENODEV) if port identifier is invalid.
3360 * - (-ENOTSUP) if hardware doesn't support tunnel type.
3363 rte_eth_dev_udp_tunnel_delete(uint8_t port_id,
3364 struct rte_eth_udp_tunnel *tunnel_udp);
3368 * Add a new 2tuple filter rule on an Ethernet device.
3371 * The port identifier of the Ethernet device.
3373 * The identifier of 2tuple filter.
3375 * The pointer to the structure describing the 2tuple filter rule.
3376 * The *rte_2tuple_filter* structure includes the values of the different
3377 * fields to match: protocol, dst_port and
3378 * tcp_flags if the protocol is tcp type.
3380 * The index of the RX queue where to store RX packets matching the added
3383 * - (0) if successful.
3384 * - (-ENOTSUP) if hardware doesn't support 2tuple filter.
3385 * - (-ENODEV) if *port_id* invalid.
3386 * - (-EINVAL) if the filter information is not correct.
3388 int rte_eth_dev_add_2tuple_filter(uint8_t port_id, uint16_t index,
3389 struct rte_2tuple_filter *filter, uint16_t rx_queue);
3392 * remove a 2tuple filter rule on an Ethernet device.
3395 * The port identifier of the Ethernet device.
3397 * The identifier of 2tuple filter.
3399 * - (0) if successful.
3400 * - (-ENOTSUP) if hardware doesn't support 2tuple filter.
3401 * - (-ENODEV) if *port_id* invalid.
3402 * - (-EINVAL) if the filter information is not correct.
3404 int rte_eth_dev_remove_2tuple_filter(uint8_t port_id, uint16_t index);
3407 * Get an 2tuple filter rule on an Ethernet device.
3410 * The port identifier of the Ethernet device.
3412 * The identifier of 2tuple filter.
3414 * A pointer to a structure of type *rte_2tuple_filter* to be filled with
3415 * the information of the 2tuple filter.
3417 * A pointer to get the queue index.
3419 * - (0) if successful.
3420 * - (-ENOTSUP) if hardware doesn't support 2tuple filter.
3421 * - (-ENODEV) if *port_id* invalid.
3422 * - (-EINVAL) if the filter information is not correct.
3423 * - (-ENOENT) if no enabled filter in this index.
3425 int rte_eth_dev_get_2tuple_filter(uint8_t port_id, uint16_t index,
3426 struct rte_2tuple_filter *filter, uint16_t *rx_queue);
3429 * Add a new 5tuple filter rule on an Ethernet device.
3432 * The port identifier of the Ethernet device.
3434 * The identifier of 5tuple filter.
3436 * The pointer to the structure describing the 5tuple filter rule.
3437 * The *rte_5tuple_filter* structure includes the values of the different
3438 * fields to match: dst src IP, dst src port, protocol and relative masks
3440 * The index of the RX queue where to store RX packets matching the added
3443 * - (0) if successful.
3444 * - (-ENOTSUP) if hardware doesn't support 5tuple filter.
3445 * - (-ENODEV) if *port_id* invalid.
3446 * - (-EINVAL) if the filter information is not correct.
3448 int rte_eth_dev_add_5tuple_filter(uint8_t port_id, uint16_t index,
3449 struct rte_5tuple_filter *filter, uint16_t rx_queue);
3452 * remove a 5tuple filter rule on an Ethernet device.
3455 * The port identifier of the Ethernet device.
3457 * The identifier of 5tuple filter.
3459 * - (0) if successful.
3460 * - (-ENOTSUP) if hardware doesn't support 5tuple filter.
3461 * - (-ENODEV) if *port_id* invalid.
3462 * - (-EINVAL) if the filter information is not correct.
3464 int rte_eth_dev_remove_5tuple_filter(uint8_t port_id, uint16_t index);
3467 * Get an 5tuple filter rule on an Ethernet device.
3470 * The port identifier of the Ethernet device.
3472 * The identifier of 5tuple filter.
3474 * A pointer to a structure of type *rte_5tuple_filter* to be filled with
3475 * the information of the 5tuple filter.
3477 * A pointer to get the queue index.
3479 * - (0) if successful.
3480 * - (-ENOTSUP) if hardware doesn't support 5tuple filter.
3481 * - (-ENODEV) if *port_id* invalid.
3482 * - (-EINVAL) if the filter information is not correct.
3484 int rte_eth_dev_get_5tuple_filter(uint8_t port_id, uint16_t index,
3485 struct rte_5tuple_filter *filter, uint16_t *rx_queue);
3488 * Check whether the filter type is supported on an Ethernet device.
3489 * All the supported filter types are defined in 'rte_eth_ctrl.h'.
3492 * The port identifier of the Ethernet device.
3493 * @param filter_type
3496 * - (0) if successful.
3497 * - (-ENOTSUP) if hardware doesn't support this filter type.
3498 * - (-ENODEV) if *port_id* invalid.
3500 int rte_eth_dev_filter_supported(uint8_t port_id, enum rte_filter_type filter_type);
3503 * Take operations to assigned filter type on an Ethernet device.
3504 * All the supported operations and filter types are defined in 'rte_eth_ctrl.h'.
3507 * The port identifier of the Ethernet device.
3508 * @param filter_type
3511 * Type of operation.
3513 * A pointer to arguments defined specifically for the operation.
3515 * - (0) if successful.
3516 * - (-ENOTSUP) if hardware doesn't support.
3517 * - (-ENODEV) if *port_id* invalid.
3518 * - others depends on the specific operations implementation.
3520 int rte_eth_dev_filter_ctrl(uint8_t port_id, enum rte_filter_type filter_type,
3521 enum rte_filter_op filter_op, void *arg);
3527 #endif /* _RTE_ETHDEV_H_ */