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"
182 * A structure used to retrieve statistics for an Ethernet port.
184 struct rte_eth_stats {
185 uint64_t ipackets; /**< Total number of successfully received packets. */
186 uint64_t opackets; /**< Total number of successfully transmitted packets.*/
187 uint64_t ibytes; /**< Total number of successfully received bytes. */
188 uint64_t obytes; /**< Total number of successfully transmitted bytes. */
189 uint64_t imissed; /**< Total of RX missed packets (e.g full FIFO). */
190 uint64_t ibadcrc; /**< Total of RX packets with CRC error. */
191 uint64_t ibadlen; /**< Total of RX packets with bad length. */
192 uint64_t ierrors; /**< Total number of erroneous received packets. */
193 uint64_t oerrors; /**< Total number of failed transmitted packets. */
194 uint64_t imcasts; /**< Total number of multicast received packets. */
195 uint64_t rx_nombuf; /**< Total number of RX mbuf allocation failures. */
196 uint64_t fdirmatch; /**< Total number of RX packets matching a filter. */
197 uint64_t fdirmiss; /**< Total number of RX packets not matching any filter. */
198 uint64_t tx_pause_xon; /**< Total nb. of XON pause frame sent. */
199 uint64_t rx_pause_xon; /**< Total nb. of XON pause frame received. */
200 uint64_t tx_pause_xoff; /**< Total nb. of XOFF pause frame sent. */
201 uint64_t rx_pause_xoff; /**< Total nb. of XOFF pause frame received. */
202 uint64_t q_ipackets[RTE_ETHDEV_QUEUE_STAT_CNTRS];
203 /**< Total number of queue RX packets. */
204 uint64_t q_opackets[RTE_ETHDEV_QUEUE_STAT_CNTRS];
205 /**< Total number of queue TX packets. */
206 uint64_t q_ibytes[RTE_ETHDEV_QUEUE_STAT_CNTRS];
207 /**< Total number of successfully received queue bytes. */
208 uint64_t q_obytes[RTE_ETHDEV_QUEUE_STAT_CNTRS];
209 /**< Total number of successfully transmitted queue bytes. */
210 uint64_t q_errors[RTE_ETHDEV_QUEUE_STAT_CNTRS];
211 /**< Total number of queue packets received that are dropped. */
213 /**< Total number of good packets received from loopback,VF Only */
215 /**< Total number of good packets transmitted to loopback,VF Only */
217 /**< Total number of good bytes received from loopback,VF Only */
219 /**< Total number of good bytes transmitted to loopback,VF Only */
223 * A structure used to retrieve link-level information of an Ethernet port.
225 struct rte_eth_link {
226 uint16_t link_speed; /**< ETH_LINK_SPEED_[10, 100, 1000, 10000] */
227 uint16_t link_duplex; /**< ETH_LINK_[HALF_DUPLEX, FULL_DUPLEX] */
228 uint8_t link_status : 1; /**< 1 -> link up, 0 -> link down */
229 }__attribute__((aligned(8))); /**< aligned for atomic64 read/write */
231 #define ETH_LINK_SPEED_AUTONEG 0 /**< Auto-negotiate link speed. */
232 #define ETH_LINK_SPEED_10 10 /**< 10 megabits/second. */
233 #define ETH_LINK_SPEED_100 100 /**< 100 megabits/second. */
234 #define ETH_LINK_SPEED_1000 1000 /**< 1 gigabits/second. */
235 #define ETH_LINK_SPEED_10000 10000 /**< 10 gigabits/second. */
236 #define ETH_LINK_SPEED_10G 10000 /**< alias of 10 gigabits/second. */
237 #define ETH_LINK_SPEED_20G 20000 /**< 20 gigabits/second. */
238 #define ETH_LINK_SPEED_40G 40000 /**< 40 gigabits/second. */
240 #define ETH_LINK_AUTONEG_DUPLEX 0 /**< Auto-negotiate duplex. */
241 #define ETH_LINK_HALF_DUPLEX 1 /**< Half-duplex connection. */
242 #define ETH_LINK_FULL_DUPLEX 2 /**< Full-duplex connection. */
245 * A structure used to configure the ring threshold registers of an RX/TX
246 * queue for an Ethernet port.
248 struct rte_eth_thresh {
249 uint8_t pthresh; /**< Ring prefetch threshold. */
250 uint8_t hthresh; /**< Ring host threshold. */
251 uint8_t wthresh; /**< Ring writeback threshold. */
255 * A set of values to identify what method is to be used to route
256 * packets to multiple queues.
258 enum rte_eth_rx_mq_mode {
259 ETH_MQ_RX_NONE = 0, /**< None of DCB,RSS or VMDQ mode */
261 ETH_MQ_RX_RSS, /**< For RX side, only RSS is on */
262 ETH_MQ_RX_DCB, /**< For RX side,only DCB is on. */
263 ETH_MQ_RX_DCB_RSS, /**< Both DCB and RSS enable */
265 ETH_MQ_RX_VMDQ_ONLY, /**< Only VMDQ, no RSS nor DCB */
266 ETH_MQ_RX_VMDQ_RSS, /**< RSS mode with VMDQ */
267 ETH_MQ_RX_VMDQ_DCB, /**< Use VMDQ+DCB to route traffic to queues */
268 ETH_MQ_RX_VMDQ_DCB_RSS, /**< Enable both VMDQ and DCB in VMDq */
272 * for rx mq mode backward compatible
274 #define ETH_RSS ETH_MQ_RX_RSS
275 #define VMDQ_DCB ETH_MQ_RX_VMDQ_DCB
276 #define ETH_DCB_RX ETH_MQ_RX_DCB
279 * A set of values to identify what method is to be used to transmit
280 * packets using multi-TCs.
282 enum rte_eth_tx_mq_mode {
283 ETH_MQ_TX_NONE = 0, /**< It is in neither DCB nor VT mode. */
284 ETH_MQ_TX_DCB, /**< For TX side,only DCB is on. */
285 ETH_MQ_TX_VMDQ_DCB, /**< For TX side,both DCB and VT is on. */
286 ETH_MQ_TX_VMDQ_ONLY, /**< Only VT on, no DCB */
290 * for tx mq mode backward compatible
292 #define ETH_DCB_NONE ETH_MQ_TX_NONE
293 #define ETH_VMDQ_DCB_TX ETH_MQ_TX_VMDQ_DCB
294 #define ETH_DCB_TX ETH_MQ_TX_DCB
297 * A structure used to configure the RX features of an Ethernet port.
299 struct rte_eth_rxmode {
300 /** The multi-queue packet distribution mode to be used, e.g. RSS. */
301 enum rte_eth_rx_mq_mode mq_mode;
302 uint32_t max_rx_pkt_len; /**< Only used if jumbo_frame enabled. */
303 uint16_t split_hdr_size; /**< hdr buf size (header_split enabled).*/
304 uint8_t header_split : 1, /**< Header Split enable. */
305 hw_ip_checksum : 1, /**< IP/UDP/TCP checksum offload enable. */
306 hw_vlan_filter : 1, /**< VLAN filter enable. */
307 hw_vlan_strip : 1, /**< VLAN strip enable. */
308 hw_vlan_extend : 1, /**< Extended VLAN enable. */
309 jumbo_frame : 1, /**< Jumbo Frame Receipt enable. */
310 hw_strip_crc : 1; /**< Enable CRC stripping by hardware. */
314 * A structure used to configure the Receive Side Scaling (RSS) feature
315 * of an Ethernet port.
316 * If not NULL, the *rss_key* pointer of the *rss_conf* structure points
317 * to an array holding the RSS key to use for hashing specific header
318 * fields of received packets. The length of this array should be indicated
319 * by *rss_key_len* below. Otherwise, a default random hash key is used by
322 * The *rss_key_len* field of the *rss_conf* structure indicates the length
323 * in bytes of the array pointed by *rss_key*. To be compatible, this length
324 * will be checked in i40e only. Others assume 40 bytes to be used as before.
326 * The *rss_hf* field of the *rss_conf* structure indicates the different
327 * types of IPv4/IPv6 packets to which the RSS hashing must be applied.
328 * Supplying an *rss_hf* equal to zero disables the RSS feature.
330 struct rte_eth_rss_conf {
331 uint8_t *rss_key; /**< If not NULL, 40-byte hash key. */
332 uint8_t rss_key_len; /**< hash key length in bytes. */
333 uint64_t rss_hf; /**< Hash functions to apply - see below. */
336 /* Supported RSS offloads */
338 #define ETH_RSS_IPV4_SHIFT 0
339 #define ETH_RSS_IPV4_TCP_SHIFT 1
340 #define ETH_RSS_IPV6_SHIFT 2
341 #define ETH_RSS_IPV6_EX_SHIFT 3
342 #define ETH_RSS_IPV6_TCP_SHIFT 4
343 #define ETH_RSS_IPV6_TCP_EX_SHIFT 5
344 #define ETH_RSS_IPV4_UDP_SHIFT 6
345 #define ETH_RSS_IPV6_UDP_SHIFT 7
346 #define ETH_RSS_IPV6_UDP_EX_SHIFT 8
348 #define ETH_RSS_NONF_IPV4_UDP_SHIFT 31
349 #define ETH_RSS_NONF_IPV4_TCP_SHIFT 33
350 #define ETH_RSS_NONF_IPV4_SCTP_SHIFT 34
351 #define ETH_RSS_NONF_IPV4_OTHER_SHIFT 35
352 #define ETH_RSS_FRAG_IPV4_SHIFT 36
353 #define ETH_RSS_NONF_IPV6_UDP_SHIFT 41
354 #define ETH_RSS_NONF_IPV6_TCP_SHIFT 43
355 #define ETH_RSS_NONF_IPV6_SCTP_SHIFT 44
356 #define ETH_RSS_NONF_IPV6_OTHER_SHIFT 45
357 #define ETH_RSS_FRAG_IPV6_SHIFT 46
358 #define ETH_RSS_FCOE_OX_SHIFT 48
359 #define ETH_RSS_FCOE_RX_SHIFT 49
360 #define ETH_RSS_FCOE_OTHER_SHIFT 50
361 #define ETH_RSS_L2_PAYLOAD_SHIFT 63
364 #define ETH_RSS_IPV4 ((uint16_t)1 << ETH_RSS_IPV4_SHIFT)
365 #define ETH_RSS_IPV4_TCP ((uint16_t)1 << ETH_RSS_IPV4_TCP_SHIFT)
366 #define ETH_RSS_IPV6 ((uint16_t)1 << ETH_RSS_IPV6_SHIFT)
367 #define ETH_RSS_IPV6_EX ((uint16_t)1 << ETH_RSS_IPV6_EX_SHIFT)
368 #define ETH_RSS_IPV6_TCP ((uint16_t)1 << ETH_RSS_IPV6_TCP_SHIFT)
369 #define ETH_RSS_IPV6_TCP_EX ((uint16_t)1 << ETH_RSS_IPV6_TCP_EX_SHIFT)
370 #define ETH_RSS_IPV4_UDP ((uint16_t)1 << ETH_RSS_IPV4_UDP_SHIFT)
371 #define ETH_RSS_IPV6_UDP ((uint16_t)1 << ETH_RSS_IPV6_UDP_SHIFT)
372 #define ETH_RSS_IPV6_UDP_EX ((uint16_t)1 << ETH_RSS_IPV6_UDP_EX_SHIFT)
374 #define ETH_RSS_NONF_IPV4_UDP ((uint64_t)1 << ETH_RSS_NONF_IPV4_UDP_SHIFT)
375 #define ETH_RSS_NONF_IPV4_TCP ((uint64_t)1 << ETH_RSS_NONF_IPV4_TCP_SHIFT)
376 #define ETH_RSS_NONF_IPV4_SCTP ((uint64_t)1 << ETH_RSS_NONF_IPV4_SCTP_SHIFT)
377 #define ETH_RSS_NONF_IPV4_OTHER ((uint64_t)1 << ETH_RSS_NONF_IPV4_OTHER_SHIFT)
378 #define ETH_RSS_FRAG_IPV4 ((uint64_t)1 << ETH_RSS_FRAG_IPV4_SHIFT)
379 #define ETH_RSS_NONF_IPV6_UDP ((uint64_t)1 << ETH_RSS_NONF_IPV6_UDP_SHIFT)
380 #define ETH_RSS_NONF_IPV6_TCP ((uint64_t)1 << ETH_RSS_NONF_IPV6_TCP_SHIFT)
381 #define ETH_RSS_NONF_IPV6_SCTP ((uint64_t)1 << ETH_RSS_NONF_IPV6_SCTP_SHIFT)
382 #define ETH_RSS_NONF_IPV6_OTHER ((uint64_t)1 << ETH_RSS_NONF_IPV6_OTHER_SHIFT)
383 #define ETH_RSS_FRAG_IPV6 ((uint64_t)1 << ETH_RSS_FRAG_IPV6_SHIFT)
384 #define ETH_RSS_FCOE_OX ((uint64_t)1 << ETH_RSS_FCOE_OX_SHIFT) /* not used */
385 #define ETH_RSS_FCOE_RX ((uint64_t)1 << ETH_RSS_FCOE_RX_SHIFT) /* not used */
386 #define ETH_RSS_FCOE_OTHER ((uint64_t)1 << ETH_RSS_FCOE_OTHER_SHIFT) /* not used */
387 #define ETH_RSS_L2_PAYLOAD ((uint64_t)1 << ETH_RSS_L2_PAYLOAD_SHIFT)
389 #define ETH_RSS_IP ( \
392 ETH_RSS_NONF_IPV4_OTHER | \
393 ETH_RSS_FRAG_IPV4 | \
394 ETH_RSS_NONF_IPV6_OTHER | \
396 #define ETH_RSS_UDP ( \
401 ETH_RSS_IPV6_UDP_EX | \
402 ETH_RSS_NONF_IPV4_UDP | \
403 ETH_RSS_NONF_IPV6_UDP)
404 /**< Mask of valid RSS hash protocols */
405 #define ETH_RSS_PROTO_MASK ( \
411 ETH_RSS_IPV6_TCP_EX | \
414 ETH_RSS_IPV6_UDP_EX | \
415 ETH_RSS_NONF_IPV4_UDP | \
416 ETH_RSS_NONF_IPV4_TCP | \
417 ETH_RSS_NONF_IPV4_SCTP | \
418 ETH_RSS_NONF_IPV4_OTHER | \
419 ETH_RSS_FRAG_IPV4 | \
420 ETH_RSS_NONF_IPV6_UDP | \
421 ETH_RSS_NONF_IPV6_TCP | \
422 ETH_RSS_NONF_IPV6_SCTP | \
423 ETH_RSS_NONF_IPV6_OTHER | \
424 ETH_RSS_FRAG_IPV6 | \
427 /* Definitions used for redirection table entry size */
428 #define ETH_RSS_RETA_NUM_ENTRIES 128
429 #define ETH_RSS_RETA_MAX_QUEUE 16
431 /* Definitions used for VMDQ and DCB functionality */
432 #define ETH_VMDQ_MAX_VLAN_FILTERS 64 /**< Maximum nb. of VMDQ vlan filters. */
433 #define ETH_DCB_NUM_USER_PRIORITIES 8 /**< Maximum nb. of DCB priorities. */
434 #define ETH_VMDQ_DCB_NUM_QUEUES 128 /**< Maximum nb. of VMDQ DCB queues. */
435 #define ETH_DCB_NUM_QUEUES 128 /**< Maximum nb. of DCB queues. */
437 /* DCB capability defines */
438 #define ETH_DCB_PG_SUPPORT 0x00000001 /**< Priority Group(ETS) support. */
439 #define ETH_DCB_PFC_SUPPORT 0x00000002 /**< Priority Flow Control support. */
441 /* Definitions used for VLAN Offload functionality */
442 #define ETH_VLAN_STRIP_OFFLOAD 0x0001 /**< VLAN Strip On/Off */
443 #define ETH_VLAN_FILTER_OFFLOAD 0x0002 /**< VLAN Filter On/Off */
444 #define ETH_VLAN_EXTEND_OFFLOAD 0x0004 /**< VLAN Extend On/Off */
446 /* Definitions used for mask VLAN setting */
447 #define ETH_VLAN_STRIP_MASK 0x0001 /**< VLAN Strip setting mask */
448 #define ETH_VLAN_FILTER_MASK 0x0002 /**< VLAN Filter setting mask*/
449 #define ETH_VLAN_EXTEND_MASK 0x0004 /**< VLAN Extend setting mask*/
450 #define ETH_VLAN_ID_MAX 0x0FFF /**< VLAN ID is in lower 12 bits*/
452 /* Definitions used for receive MAC address */
453 #define ETH_NUM_RECEIVE_MAC_ADDR 128 /**< Maximum nb. of receive mac addr. */
456 /* Definitions used for unicast hash */
457 #define ETH_VMDQ_NUM_UC_HASH_ARRAY 128 /**< Maximum nb. of UC hash array. */
459 /* Definitions used for VMDQ pool rx mode setting */
460 #define ETH_VMDQ_ACCEPT_UNTAG 0x0001 /**< accept untagged packets. */
461 #define ETH_VMDQ_ACCEPT_HASH_MC 0x0002 /**< accept packets in multicast table . */
462 #define ETH_VMDQ_ACCEPT_HASH_UC 0x0004 /**< accept packets in unicast table. */
463 #define ETH_VMDQ_ACCEPT_BROADCAST 0x0008 /**< accept broadcast packets. */
464 #define ETH_VMDQ_ACCEPT_MULTICAST 0x0010 /**< multicast promiscuous. */
466 /* Definitions used for VMDQ mirror rules setting */
467 #define ETH_VMDQ_NUM_MIRROR_RULE 4 /**< Maximum nb. of mirror rules. . */
469 #define ETH_VMDQ_POOL_MIRROR 0x0001 /**< Virtual Pool Mirroring. */
470 #define ETH_VMDQ_UPLINK_MIRROR 0x0002 /**< Uplink Port Mirroring. */
471 #define ETH_VMDQ_DOWNLIN_MIRROR 0x0004 /**< Downlink Port Mirroring. */
472 #define ETH_VMDQ_VLAN_MIRROR 0x0008 /**< VLAN Mirroring. */
475 * A structure used to configure VLAN traffic mirror of an Ethernet port.
477 struct rte_eth_vlan_mirror {
478 uint64_t vlan_mask; /**< mask for valid VLAN ID. */
479 uint16_t vlan_id[ETH_VMDQ_MAX_VLAN_FILTERS];
480 /** VLAN ID list for vlan mirror. */
484 * A structure used to configure traffic mirror of an Ethernet port.
486 struct rte_eth_vmdq_mirror_conf {
487 uint8_t rule_type_mask; /**< Mirroring rule type mask we want to set */
488 uint8_t dst_pool; /**< Destination pool for this mirror rule. */
489 uint64_t pool_mask; /**< Bitmap of pool for pool mirroring */
490 struct rte_eth_vlan_mirror vlan; /**< VLAN ID setting for VLAN mirroring */
494 * A structure used to configure Redirection Table of the Receive Side
495 * Scaling (RSS) feature of an Ethernet port.
497 struct rte_eth_rss_reta {
498 /** First 64 mask bits indicate which entry(s) need to updated/queried. */
500 /** Second 64 mask bits indicate which entry(s) need to updated/queried. */
502 uint8_t reta[ETH_RSS_RETA_NUM_ENTRIES]; /**< 128 RETA entries*/
506 * This enum indicates the possible number of traffic classes
507 * in DCB configratioins
509 enum rte_eth_nb_tcs {
510 ETH_4_TCS = 4, /**< 4 TCs with DCB. */
511 ETH_8_TCS = 8 /**< 8 TCs with DCB. */
515 * This enum indicates the possible number of queue pools
516 * in VMDQ configurations.
518 enum rte_eth_nb_pools {
519 ETH_8_POOLS = 8, /**< 8 VMDq pools. */
520 ETH_16_POOLS = 16, /**< 16 VMDq pools. */
521 ETH_32_POOLS = 32, /**< 32 VMDq pools. */
522 ETH_64_POOLS = 64 /**< 64 VMDq pools. */
525 /* This structure may be extended in future. */
526 struct rte_eth_dcb_rx_conf {
527 enum rte_eth_nb_tcs nb_tcs; /**< Possible DCB TCs, 4 or 8 TCs */
528 uint8_t dcb_queue[ETH_DCB_NUM_USER_PRIORITIES];
529 /**< Possible DCB queue,4 or 8. */
532 struct rte_eth_vmdq_dcb_tx_conf {
533 enum rte_eth_nb_pools nb_queue_pools; /**< With DCB, 16 or 32 pools. */
534 uint8_t dcb_queue[ETH_DCB_NUM_USER_PRIORITIES];
535 /**< Possible DCB queue,4 or 8. */
538 struct rte_eth_dcb_tx_conf {
539 enum rte_eth_nb_tcs nb_tcs; /**< Possible DCB TCs, 4 or 8 TCs. */
540 uint8_t dcb_queue[ETH_DCB_NUM_USER_PRIORITIES];
541 /**< Possible DCB queue,4 or 8. */
544 struct rte_eth_vmdq_tx_conf {
545 enum rte_eth_nb_pools nb_queue_pools; /**< VMDq mode, 64 pools. */
549 * A structure used to configure the VMDQ+DCB feature
550 * of an Ethernet port.
552 * Using this feature, packets are routed to a pool of queues, based
553 * on the vlan id in the vlan tag, and then to a specific queue within
554 * that pool, using the user priority vlan tag field.
556 * A default pool may be used, if desired, to route all traffic which
557 * does not match the vlan filter rules.
559 struct rte_eth_vmdq_dcb_conf {
560 enum rte_eth_nb_pools nb_queue_pools; /**< With DCB, 16 or 32 pools */
561 uint8_t enable_default_pool; /**< If non-zero, use a default pool */
562 uint8_t default_pool; /**< The default pool, if applicable */
563 uint8_t nb_pool_maps; /**< We can have up to 64 filters/mappings */
565 uint16_t vlan_id; /**< The vlan id of the received frame */
566 uint64_t pools; /**< Bitmask of pools for packet rx */
567 } pool_map[ETH_VMDQ_MAX_VLAN_FILTERS]; /**< VMDq vlan pool maps. */
568 uint8_t dcb_queue[ETH_DCB_NUM_USER_PRIORITIES];
569 /**< Selects a queue in a pool */
572 struct rte_eth_vmdq_rx_conf {
573 enum rte_eth_nb_pools nb_queue_pools; /**< VMDq only mode, 8 or 64 pools */
574 uint8_t enable_default_pool; /**< If non-zero, use a default pool */
575 uint8_t default_pool; /**< The default pool, if applicable */
576 uint8_t enable_loop_back; /**< Enable VT loop back */
577 uint8_t nb_pool_maps; /**< We can have up to 64 filters/mappings */
579 uint16_t vlan_id; /**< The vlan id of the received frame */
580 uint64_t pools; /**< Bitmask of pools for packet rx */
581 } pool_map[ETH_VMDQ_MAX_VLAN_FILTERS]; /**< VMDq vlan pool maps. */
585 * A structure used to configure the TX features of an Ethernet port.
587 struct rte_eth_txmode {
588 enum rte_eth_tx_mq_mode mq_mode; /**< TX multi-queues mode. */
592 * A structure used to configure an RX ring of an Ethernet port.
594 struct rte_eth_rxconf {
595 struct rte_eth_thresh rx_thresh; /**< RX ring threshold registers. */
596 uint16_t rx_free_thresh; /**< Drives the freeing of RX descriptors. */
597 uint8_t rx_drop_en; /**< Drop packets if no descriptors are available. */
598 uint8_t start_rx_per_q; /**< start rx per queue. */
601 #define ETH_TXQ_FLAGS_NOMULTSEGS 0x0001 /**< nb_segs=1 for all mbufs */
602 #define ETH_TXQ_FLAGS_NOREFCOUNT 0x0002 /**< refcnt can be ignored */
603 #define ETH_TXQ_FLAGS_NOMULTMEMP 0x0004 /**< all bufs come from same mempool */
604 #define ETH_TXQ_FLAGS_NOVLANOFFL 0x0100 /**< disable VLAN offload */
605 #define ETH_TXQ_FLAGS_NOXSUMSCTP 0x0200 /**< disable SCTP checksum offload */
606 #define ETH_TXQ_FLAGS_NOXSUMUDP 0x0400 /**< disable UDP checksum offload */
607 #define ETH_TXQ_FLAGS_NOXSUMTCP 0x0800 /**< disable TCP checksum offload */
608 #define ETH_TXQ_FLAGS_NOOFFLOADS \
609 (ETH_TXQ_FLAGS_NOVLANOFFL | ETH_TXQ_FLAGS_NOXSUMSCTP | \
610 ETH_TXQ_FLAGS_NOXSUMUDP | ETH_TXQ_FLAGS_NOXSUMTCP)
612 * A structure used to configure a TX ring of an Ethernet port.
614 struct rte_eth_txconf {
615 struct rte_eth_thresh tx_thresh; /**< TX ring threshold registers. */
616 uint16_t tx_rs_thresh; /**< Drives the setting of RS bit on TXDs. */
617 uint16_t tx_free_thresh; /**< Drives the freeing of TX buffers. */
618 uint32_t txq_flags; /**< Set flags for the Tx queue */
619 uint8_t start_tx_per_q; /**< start tx per queue. */
623 * This enum indicates the flow control mode
625 enum rte_eth_fc_mode {
626 RTE_FC_NONE = 0, /**< Disable flow control. */
627 RTE_FC_RX_PAUSE, /**< RX pause frame, enable flowctrl on TX side. */
628 RTE_FC_TX_PAUSE, /**< TX pause frame, enable flowctrl on RX side. */
629 RTE_FC_FULL /**< Enable flow control on both side. */
633 * A structure used to configure Ethernet flow control parameter.
634 * These parameters will be configured into the register of the NIC.
635 * Please refer to the corresponding data sheet for proper value.
637 struct rte_eth_fc_conf {
638 uint32_t high_water; /**< High threshold value to trigger XOFF */
639 uint32_t low_water; /**< Low threshold value to trigger XON */
640 uint16_t pause_time; /**< Pause quota in the Pause frame */
641 uint16_t send_xon; /**< Is XON frame need be sent */
642 enum rte_eth_fc_mode mode; /**< Link flow control mode */
643 uint8_t mac_ctrl_frame_fwd; /**< Forward MAC control frames */
647 * A structure used to configure Ethernet priority flow control parameter.
648 * These parameters will be configured into the register of the NIC.
649 * Please refer to the corresponding data sheet for proper value.
651 struct rte_eth_pfc_conf {
652 struct rte_eth_fc_conf fc; /**< General flow control parameter. */
653 uint8_t priority; /**< VLAN User Priority. */
657 * Flow Director setting modes: none (default), signature or perfect.
660 RTE_FDIR_MODE_NONE = 0, /**< Disable FDIR support. */
661 RTE_FDIR_MODE_SIGNATURE, /**< Enable FDIR signature filter mode. */
662 RTE_FDIR_MODE_PERFECT, /**< Enable FDIR perfect filter mode. */
666 * Memory space that can be configured to store Flow Director filters
667 * in the board memory.
669 enum rte_fdir_pballoc_type {
670 RTE_FDIR_PBALLOC_64K = 0, /**< 64k. */
671 RTE_FDIR_PBALLOC_128K, /**< 128k. */
672 RTE_FDIR_PBALLOC_256K, /**< 256k. */
676 * Select report mode of FDIR hash information in RX descriptors.
678 enum rte_fdir_status_mode {
679 RTE_FDIR_NO_REPORT_STATUS = 0, /**< Never report FDIR hash. */
680 RTE_FDIR_REPORT_STATUS, /**< Only report FDIR hash for matching pkts. */
681 RTE_FDIR_REPORT_STATUS_ALWAYS, /**< Always report FDIR hash. */
685 * A structure used to configure the Flow Director (FDIR) feature
686 * of an Ethernet port.
688 * If mode is RTE_FDIR_DISABLE, the pballoc value is ignored.
690 struct rte_fdir_conf {
691 enum rte_fdir_mode mode; /**< Flow Director mode. */
692 enum rte_fdir_pballoc_type pballoc; /**< Space for FDIR filters. */
693 enum rte_fdir_status_mode status; /**< How to report FDIR hash. */
694 /** Offset of flexbytes field in RX packets (in 16-bit word units). */
695 uint8_t flexbytes_offset;
696 /** RX queue of packets matching a "drop" filter in perfect mode. */
701 * Possible l4type of FDIR filters.
704 RTE_FDIR_L4TYPE_NONE = 0, /**< None. */
705 RTE_FDIR_L4TYPE_UDP, /**< UDP. */
706 RTE_FDIR_L4TYPE_TCP, /**< TCP. */
707 RTE_FDIR_L4TYPE_SCTP, /**< SCTP. */
711 * Select IPv4 or IPv6 FDIR filters.
714 RTE_FDIR_IPTYPE_IPV4 = 0, /**< IPv4. */
715 RTE_FDIR_IPTYPE_IPV6 , /**< IPv6. */
719 * A structure used to define a FDIR packet filter.
721 struct rte_fdir_filter {
722 uint16_t flex_bytes; /**< Flex bytes value to match. */
723 uint16_t vlan_id; /**< VLAN ID value to match, 0 otherwise. */
724 uint16_t port_src; /**< Source port to match, 0 otherwise. */
725 uint16_t port_dst; /**< Destination port to match, 0 otherwise. */
727 uint32_t ipv4_addr; /**< IPv4 source address to match. */
728 uint32_t ipv6_addr[4]; /**< IPv6 source address to match. */
729 } ip_src; /**< IPv4/IPv6 source address to match (union of above). */
731 uint32_t ipv4_addr; /**< IPv4 destination address to match. */
732 uint32_t ipv6_addr[4]; /**< IPv6 destination address to match */
733 } ip_dst; /**< IPv4/IPv6 destination address to match (union of above). */
734 enum rte_l4type l4type; /**< l4type to match: NONE/UDP/TCP/SCTP. */
735 enum rte_iptype iptype; /**< IP packet type to match: IPv4 or IPv6. */
739 * A structure used to configure FDIR masks that are used by the device
740 * to match the various fields of RX packet headers.
741 * @note The only_ip_flow field has the opposite meaning compared to other
744 struct rte_fdir_masks {
745 /** When set to 1, packet l4type is \b NOT relevant in filters, and
746 source and destination port masks must be set to zero. */
747 uint8_t only_ip_flow;
748 /** If set to 1, vlan_id is relevant in filters. */
750 /** If set to 1, vlan_prio is relevant in filters. */
752 /** If set to 1, flexbytes is relevant in filters. */
754 /** If set to 1, set the IPv6 masks. Otherwise set the IPv4 masks. */
755 uint8_t set_ipv6_mask;
756 /** When set to 1, comparison of destination IPv6 address with IP6AT
757 registers is meaningful. */
758 uint8_t comp_ipv6_dst;
759 /** Mask of Destination IPv4 Address. All bits set to 1 define the
760 relevant bits to use in the destination address of an IPv4 packet
761 when matching it against FDIR filters. */
762 uint32_t dst_ipv4_mask;
763 /** Mask of Source IPv4 Address. All bits set to 1 define
764 the relevant bits to use in the source address of an IPv4 packet
765 when matching it against FDIR filters. */
766 uint32_t src_ipv4_mask;
767 /** Mask of Source IPv6 Address. All bits set to 1 define the
768 relevant BYTES to use in the source address of an IPv6 packet
769 when matching it against FDIR filters. */
770 uint16_t dst_ipv6_mask;
771 /** Mask of Destination IPv6 Address. All bits set to 1 define the
772 relevant BYTES to use in the destination address of an IPv6 packet
773 when matching it against FDIR filters. */
774 uint16_t src_ipv6_mask;
775 /** Mask of Source Port. All bits set to 1 define the relevant
776 bits to use in the source port of an IP packets when matching it
777 against FDIR filters. */
778 uint16_t src_port_mask;
779 /** Mask of Destination Port. All bits set to 1 define the relevant
780 bits to use in the destination port of an IP packet when matching it
781 against FDIR filters. */
782 uint16_t dst_port_mask;
786 * A structure used to report the status of the flow director filters in use.
788 struct rte_eth_fdir {
789 /** Number of filters with collision indication. */
791 /** Number of free (non programmed) filters. */
793 /** The Lookup hash value of the added filter that updated the value
794 of the MAXLEN field */
796 /** Longest linked list of filters in the table. */
798 /** Number of added filters. */
800 /** Number of removed filters. */
802 /** Number of failed added filters (no more space in device). */
804 /** Number of failed removed filters. */
809 * A structure used to enable/disable specific device interrupts.
811 struct rte_intr_conf {
812 /** enable/disable lsc interrupt. 0 (default) - disable, 1 enable */
817 * A structure used to configure an Ethernet port.
818 * Depending upon the RX multi-queue mode, extra advanced
819 * configuration settings may be needed.
821 struct rte_eth_conf {
823 /**< ETH_LINK_SPEED_10[0|00|000], or 0 for autonegotation */
824 uint16_t link_duplex;
825 /**< ETH_LINK_[HALF_DUPLEX|FULL_DUPLEX], or 0 for autonegotation */
826 struct rte_eth_rxmode rxmode; /**< Port RX configuration. */
827 struct rte_eth_txmode txmode; /**< Port TX configuration. */
828 uint32_t lpbk_mode; /**< Loopback operation mode. By default the value
829 is 0, meaning the loopback mode is disabled.
830 Read the datasheet of given ethernet controller
831 for details. The possible values of this field
832 are defined in implementation of each driver. */
834 struct rte_eth_rss_conf rss_conf; /**< Port RSS configuration */
835 struct rte_eth_vmdq_dcb_conf vmdq_dcb_conf;
836 /**< Port vmdq+dcb configuration. */
837 struct rte_eth_dcb_rx_conf dcb_rx_conf;
838 /**< Port dcb RX configuration. */
839 struct rte_eth_vmdq_rx_conf vmdq_rx_conf;
840 /**< Port vmdq RX configuration. */
841 } rx_adv_conf; /**< Port RX filtering configuration (union). */
843 struct rte_eth_vmdq_dcb_tx_conf vmdq_dcb_tx_conf;
844 /**< Port vmdq+dcb TX configuration. */
845 struct rte_eth_dcb_tx_conf dcb_tx_conf;
846 /**< Port dcb TX configuration. */
847 struct rte_eth_vmdq_tx_conf vmdq_tx_conf;
848 /**< Port vmdq TX configuration. */
849 } tx_adv_conf; /**< Port TX DCB configuration (union). */
850 /** Currently,Priority Flow Control(PFC) are supported,if DCB with PFC
851 is needed,and the variable must be set ETH_DCB_PFC_SUPPORT. */
852 uint32_t dcb_capability_en;
853 struct rte_fdir_conf fdir_conf; /**< FDIR configuration. */
854 struct rte_intr_conf intr_conf; /**< Interrupt mode configuration. */
858 * A structure used to retrieve the contextual information of
859 * an Ethernet device, such as the controlling driver of the device,
860 * its PCI context, etc...
864 * RX offload capabilities of a device.
866 #define DEV_RX_OFFLOAD_VLAN_STRIP 0x00000001
867 #define DEV_RX_OFFLOAD_IPV4_CKSUM 0x00000002
868 #define DEV_RX_OFFLOAD_UDP_CKSUM 0x00000004
869 #define DEV_RX_OFFLOAD_TCP_CKSUM 0x00000008
870 #define DEV_RX_OFFLOAD_TCP_LRO 0x00000010
873 * TX offload capabilities of a device.
875 #define DEV_TX_OFFLOAD_VLAN_INSERT 0x00000001
876 #define DEV_TX_OFFLOAD_IPV4_CKSUM 0x00000002
877 #define DEV_TX_OFFLOAD_UDP_CKSUM 0x00000004
878 #define DEV_TX_OFFLOAD_TCP_CKSUM 0x00000008
879 #define DEV_TX_OFFLOAD_SCTP_CKSUM 0x00000010
880 #define DEV_TX_OFFLOAD_TCP_TSO 0x00000020
881 #define DEV_TX_OFFLOAD_UDP_TSO 0x00000040
883 struct rte_eth_dev_info {
884 struct rte_pci_device *pci_dev; /**< Device PCI information. */
885 const char *driver_name; /**< Device Driver name. */
886 unsigned int if_index; /**< Index to bound host interface, or 0 if none.
887 Use if_indextoname() to translate into an interface name. */
888 uint32_t min_rx_bufsize; /**< Minimum size of RX buffer. */
889 uint32_t max_rx_pktlen; /**< Maximum configurable length of RX pkt. */
890 uint16_t max_rx_queues; /**< Maximum number of RX queues. */
891 uint16_t max_tx_queues; /**< Maximum number of TX queues. */
892 uint32_t max_mac_addrs; /**< Maximum number of MAC addresses. */
893 uint32_t max_hash_mac_addrs;
894 /** Maximum number of hash MAC addresses for MTA and UTA. */
895 uint16_t max_vfs; /**< Maximum number of VFs. */
896 uint16_t max_vmdq_pools; /**< Maximum number of VMDq pools. */
897 uint32_t rx_offload_capa; /**< Device RX offload capabilities. */
898 uint32_t tx_offload_capa; /**< Device TX offload capabilities. */
903 struct rte_eth_dev_callback;
904 /** @internal Structure to keep track of registered callbacks */
905 TAILQ_HEAD(rte_eth_dev_cb_list, rte_eth_dev_callback);
907 #define TCP_UGR_FLAG 0x20
908 #define TCP_ACK_FLAG 0x10
909 #define TCP_PSH_FLAG 0x08
910 #define TCP_RST_FLAG 0x04
911 #define TCP_SYN_FLAG 0x02
912 #define TCP_FIN_FLAG 0x01
913 #define TCP_FLAG_ALL 0x3F
916 * A structure used to define an ethertype filter.
918 struct rte_ethertype_filter {
919 uint16_t ethertype; /**< little endian. */
920 uint8_t priority_en; /**< compare priority enable. */
925 * A structure used to define an syn filter.
927 struct rte_syn_filter {
928 uint8_t hig_pri; /**< 1 means higher pri than 2tuple, 5tupe,
929 and flex filter, 0 means lower pri. */
933 * A structure used to define a 2tuple filter.
935 struct rte_2tuple_filter {
936 uint16_t dst_port; /**< big endian. */
939 uint16_t priority; /**< used when more than one filter matches. */
940 uint8_t dst_port_mask:1, /**< if mask is 1b, means not compare. */
945 * A structure used to define a flex filter.
947 struct rte_flex_filter {
949 uint32_t dwords[32]; /**< flex bytes in big endian. */
950 uint8_t mask[16]; /**< if mask bit is 1b, do not compare
951 corresponding byte in dwords. */
956 * A structure used to define a 5tuple filter.
958 struct rte_5tuple_filter {
959 uint32_t dst_ip; /**< destination IP address in big endian. */
960 uint32_t src_ip; /**< source IP address in big endian. */
961 uint16_t dst_port; /**< destination port in big endian. */
962 uint16_t src_port; /**< source Port big endian. */
963 uint8_t protocol; /**< l4 protocol. */
964 uint8_t tcp_flags; /**< tcp flags. */
965 uint16_t priority; /**< seven evels (001b-111b), 111b is highest,
966 used when more than one filter matches. */
967 uint8_t dst_ip_mask:1, /**< if mask is 1b, do not compare dst ip. */
968 src_ip_mask:1, /**< if mask is 1b, do not compare src ip. */
969 dst_port_mask:1, /**< if mask is 1b, do not compare dst port. */
970 src_port_mask:1, /**< if mask is 1b, do not compare src port. */
971 protocol_mask:1; /**< if mask is 1b, do not compare protocol. */
975 * Definitions of all functions exported by an Ethernet driver through the
976 * the generic structure of type *eth_dev_ops* supplied in the *rte_eth_dev*
977 * structure associated with an Ethernet device.
980 typedef int (*eth_dev_configure_t)(struct rte_eth_dev *dev);
981 /**< @internal Ethernet device configuration. */
983 typedef int (*eth_dev_start_t)(struct rte_eth_dev *dev);
984 /**< @internal Function used to start a configured Ethernet device. */
986 typedef void (*eth_dev_stop_t)(struct rte_eth_dev *dev);
987 /**< @internal Function used to stop a configured Ethernet device. */
989 typedef int (*eth_dev_set_link_up_t)(struct rte_eth_dev *dev);
990 /**< @internal Function used to link up a configured Ethernet device. */
992 typedef int (*eth_dev_set_link_down_t)(struct rte_eth_dev *dev);
993 /**< @internal Function used to link down a configured Ethernet device. */
995 typedef void (*eth_dev_close_t)(struct rte_eth_dev *dev);
996 /**< @internal Function used to close a configured Ethernet device. */
998 typedef void (*eth_promiscuous_enable_t)(struct rte_eth_dev *dev);
999 /**< @internal Function used to enable the RX promiscuous mode of an Ethernet device. */
1001 typedef void (*eth_promiscuous_disable_t)(struct rte_eth_dev *dev);
1002 /**< @internal Function used to disable the RX promiscuous mode of an Ethernet device. */
1004 typedef void (*eth_allmulticast_enable_t)(struct rte_eth_dev *dev);
1005 /**< @internal Enable the receipt of all multicast packets by an Ethernet device. */
1007 typedef void (*eth_allmulticast_disable_t)(struct rte_eth_dev *dev);
1008 /**< @internal Disable the receipt of all multicast packets by an Ethernet device. */
1010 typedef int (*eth_link_update_t)(struct rte_eth_dev *dev,
1011 int wait_to_complete);
1012 /**< @internal Get link speed, duplex mode and state (up/down) of an Ethernet device. */
1014 typedef void (*eth_stats_get_t)(struct rte_eth_dev *dev,
1015 struct rte_eth_stats *igb_stats);
1016 /**< @internal Get global I/O statistics of an Ethernet device. */
1018 typedef void (*eth_stats_reset_t)(struct rte_eth_dev *dev);
1019 /**< @internal Reset global I/O statistics of an Ethernet device to 0. */
1021 typedef int (*eth_queue_stats_mapping_set_t)(struct rte_eth_dev *dev,
1025 /**< @internal Set a queue statistics mapping for a tx/rx queue of an Ethernet device. */
1027 typedef void (*eth_dev_infos_get_t)(struct rte_eth_dev *dev,
1028 struct rte_eth_dev_info *dev_info);
1029 /**< @internal Get specific informations of an Ethernet device. */
1031 typedef int (*eth_queue_start_t)(struct rte_eth_dev *dev,
1033 /**< @internal Start rx and tx of a queue of an Ethernet device. */
1035 typedef int (*eth_queue_stop_t)(struct rte_eth_dev *dev,
1037 /**< @internal Stop rx and tx of a queue of an Ethernet device. */
1039 typedef int (*eth_rx_queue_setup_t)(struct rte_eth_dev *dev,
1040 uint16_t rx_queue_id,
1041 uint16_t nb_rx_desc,
1042 unsigned int socket_id,
1043 const struct rte_eth_rxconf *rx_conf,
1044 struct rte_mempool *mb_pool);
1045 /**< @internal Set up a receive queue of an Ethernet device. */
1047 typedef int (*eth_tx_queue_setup_t)(struct rte_eth_dev *dev,
1048 uint16_t tx_queue_id,
1049 uint16_t nb_tx_desc,
1050 unsigned int socket_id,
1051 const struct rte_eth_txconf *tx_conf);
1052 /**< @internal Setup a transmit queue of an Ethernet device. */
1054 typedef void (*eth_queue_release_t)(void *queue);
1055 /**< @internal Release memory resources allocated by given RX/TX queue. */
1057 typedef uint32_t (*eth_rx_queue_count_t)(struct rte_eth_dev *dev,
1058 uint16_t rx_queue_id);
1059 /**< @Get number of available descriptors on a receive queue of an Ethernet device. */
1061 typedef int (*eth_rx_descriptor_done_t)(void *rxq, uint16_t offset);
1062 /**< @Check DD bit of specific RX descriptor */
1064 typedef int (*vlan_filter_set_t)(struct rte_eth_dev *dev,
1067 /**< @internal filtering of a VLAN Tag Identifier by an Ethernet device. */
1069 typedef void (*vlan_tpid_set_t)(struct rte_eth_dev *dev,
1071 /**< @internal set the outer VLAN-TPID by an Ethernet device. */
1073 typedef void (*vlan_offload_set_t)(struct rte_eth_dev *dev, int mask);
1074 /**< @internal set VLAN offload function by an Ethernet device. */
1076 typedef void (*vlan_strip_queue_set_t)(struct rte_eth_dev *dev,
1077 uint16_t rx_queue_id,
1079 /**< @internal VLAN stripping enable/disable by an queue of Ethernet device. */
1081 typedef uint16_t (*eth_rx_burst_t)(void *rxq,
1082 struct rte_mbuf **rx_pkts,
1084 /**< @internal Retrieve input packets from a receive queue of an Ethernet device. */
1086 typedef uint16_t (*eth_tx_burst_t)(void *txq,
1087 struct rte_mbuf **tx_pkts,
1089 /**< @internal Send output packets on a transmit queue of an Ethernet device. */
1091 typedef int (*fdir_add_signature_filter_t)(struct rte_eth_dev *dev,
1092 struct rte_fdir_filter *fdir_ftr,
1094 /**< @internal Setup a new signature filter rule on an Ethernet device */
1096 typedef int (*fdir_update_signature_filter_t)(struct rte_eth_dev *dev,
1097 struct rte_fdir_filter *fdir_ftr,
1099 /**< @internal Update a signature filter rule on an Ethernet device */
1101 typedef int (*fdir_remove_signature_filter_t)(struct rte_eth_dev *dev,
1102 struct rte_fdir_filter *fdir_ftr);
1103 /**< @internal Remove a signature filter rule on an Ethernet device */
1105 typedef void (*fdir_infos_get_t)(struct rte_eth_dev *dev,
1106 struct rte_eth_fdir *fdir);
1107 /**< @internal Get information about fdir status */
1109 typedef int (*fdir_add_perfect_filter_t)(struct rte_eth_dev *dev,
1110 struct rte_fdir_filter *fdir_ftr,
1111 uint16_t soft_id, uint8_t rx_queue,
1113 /**< @internal Setup a new perfect filter rule on an Ethernet device */
1115 typedef int (*fdir_update_perfect_filter_t)(struct rte_eth_dev *dev,
1116 struct rte_fdir_filter *fdir_ftr,
1117 uint16_t soft_id, uint8_t rx_queue,
1119 /**< @internal Update a perfect filter rule on an Ethernet device */
1121 typedef int (*fdir_remove_perfect_filter_t)(struct rte_eth_dev *dev,
1122 struct rte_fdir_filter *fdir_ftr,
1124 /**< @internal Remove a perfect filter rule on an Ethernet device */
1126 typedef int (*fdir_set_masks_t)(struct rte_eth_dev *dev,
1127 struct rte_fdir_masks *fdir_masks);
1128 /**< @internal Setup flow director masks on an Ethernet device */
1130 typedef int (*flow_ctrl_set_t)(struct rte_eth_dev *dev,
1131 struct rte_eth_fc_conf *fc_conf);
1132 /**< @internal Setup flow control parameter on an Ethernet device */
1134 typedef int (*priority_flow_ctrl_set_t)(struct rte_eth_dev *dev,
1135 struct rte_eth_pfc_conf *pfc_conf);
1136 /**< @internal Setup priority flow control parameter on an Ethernet device */
1138 typedef int (*reta_update_t)(struct rte_eth_dev *dev,
1139 struct rte_eth_rss_reta *reta_conf);
1140 /**< @internal Update RSS redirection table on an Ethernet device */
1142 typedef int (*reta_query_t)(struct rte_eth_dev *dev,
1143 struct rte_eth_rss_reta *reta_conf);
1144 /**< @internal Query RSS redirection table on an Ethernet device */
1146 typedef int (*rss_hash_update_t)(struct rte_eth_dev *dev,
1147 struct rte_eth_rss_conf *rss_conf);
1148 /**< @internal Update RSS hash configuration of an Ethernet device */
1150 typedef int (*rss_hash_conf_get_t)(struct rte_eth_dev *dev,
1151 struct rte_eth_rss_conf *rss_conf);
1152 /**< @internal Get current RSS hash configuration of an Ethernet device */
1154 typedef int (*eth_dev_led_on_t)(struct rte_eth_dev *dev);
1155 /**< @internal Turn on SW controllable LED on an Ethernet device */
1157 typedef int (*eth_dev_led_off_t)(struct rte_eth_dev *dev);
1158 /**< @internal Turn off SW controllable LED on an Ethernet device */
1160 typedef void (*eth_mac_addr_remove_t)(struct rte_eth_dev *dev, uint32_t index);
1161 /**< @internal Remove MAC address from receive address register */
1163 typedef void (*eth_mac_addr_add_t)(struct rte_eth_dev *dev,
1164 struct ether_addr *mac_addr,
1167 /**< @internal Set a MAC address into Receive Address Address Register */
1169 typedef int (*eth_uc_hash_table_set_t)(struct rte_eth_dev *dev,
1170 struct ether_addr *mac_addr,
1172 /**< @internal Set a Unicast Hash bitmap */
1174 typedef int (*eth_uc_all_hash_table_set_t)(struct rte_eth_dev *dev,
1176 /**< @internal Set all Unicast Hash bitmap */
1178 typedef int (*eth_set_vf_rx_mode_t)(struct rte_eth_dev *dev,
1182 /**< @internal Set a VF receive mode */
1184 typedef int (*eth_set_vf_rx_t)(struct rte_eth_dev *dev,
1187 /**< @internal Set a VF receive mode */
1189 typedef int (*eth_set_vf_tx_t)(struct rte_eth_dev *dev,
1192 /**< @internal Enable or disable a VF transmit */
1194 typedef int (*eth_set_vf_vlan_filter_t)(struct rte_eth_dev *dev,
1198 /**< @internal Set VF VLAN pool filter */
1200 typedef int (*eth_set_queue_rate_limit_t)(struct rte_eth_dev *dev,
1203 /**< @internal Set queue TX rate */
1205 typedef int (*eth_set_vf_rate_limit_t)(struct rte_eth_dev *dev,
1209 /**< @internal Set VF TX rate */
1211 typedef int (*eth_mirror_rule_set_t)(struct rte_eth_dev *dev,
1212 struct rte_eth_vmdq_mirror_conf *mirror_conf,
1215 /**< @internal Add a traffic mirroring rule on an Ethernet device */
1217 typedef int (*eth_mirror_rule_reset_t)(struct rte_eth_dev *dev,
1219 /**< @internal Remove a traffic mirroring rule on an Ethernet device */
1221 #ifdef RTE_NIC_BYPASS
1224 RTE_BYPASS_MODE_NONE,
1225 RTE_BYPASS_MODE_NORMAL,
1226 RTE_BYPASS_MODE_BYPASS,
1227 RTE_BYPASS_MODE_ISOLATE,
1228 RTE_BYPASS_MODE_NUM,
1231 #define RTE_BYPASS_MODE_VALID(x) \
1232 ((x) > RTE_BYPASS_MODE_NONE && (x) < RTE_BYPASS_MODE_NUM)
1235 RTE_BYPASS_EVENT_NONE,
1236 RTE_BYPASS_EVENT_START,
1237 RTE_BYPASS_EVENT_OS_ON = RTE_BYPASS_EVENT_START,
1238 RTE_BYPASS_EVENT_POWER_ON,
1239 RTE_BYPASS_EVENT_OS_OFF,
1240 RTE_BYPASS_EVENT_POWER_OFF,
1241 RTE_BYPASS_EVENT_TIMEOUT,
1242 RTE_BYPASS_EVENT_NUM
1245 #define RTE_BYPASS_EVENT_VALID(x) \
1246 ((x) > RTE_BYPASS_EVENT_NONE && (x) < RTE_BYPASS_MODE_NUM)
1249 RTE_BYPASS_TMT_OFF, /* timeout disabled. */
1250 RTE_BYPASS_TMT_1_5_SEC, /* timeout for 1.5 seconds */
1251 RTE_BYPASS_TMT_2_SEC, /* timeout for 2 seconds */
1252 RTE_BYPASS_TMT_3_SEC, /* timeout for 3 seconds */
1253 RTE_BYPASS_TMT_4_SEC, /* timeout for 4 seconds */
1254 RTE_BYPASS_TMT_8_SEC, /* timeout for 8 seconds */
1255 RTE_BYPASS_TMT_16_SEC, /* timeout for 16 seconds */
1256 RTE_BYPASS_TMT_32_SEC, /* timeout for 32 seconds */
1260 #define RTE_BYPASS_TMT_VALID(x) \
1261 ((x) == RTE_BYPASS_TMT_OFF || \
1262 ((x) > RTE_BYPASS_TMT_OFF && (x) < RTE_BYPASS_TMT_NUM))
1264 typedef void (*bypass_init_t)(struct rte_eth_dev *dev);
1265 typedef int32_t (*bypass_state_set_t)(struct rte_eth_dev *dev, uint32_t *new_state);
1266 typedef int32_t (*bypass_state_show_t)(struct rte_eth_dev *dev, uint32_t *state);
1267 typedef int32_t (*bypass_event_set_t)(struct rte_eth_dev *dev, uint32_t state, uint32_t event);
1268 typedef int32_t (*bypass_event_show_t)(struct rte_eth_dev *dev, uint32_t event_shift, uint32_t *event);
1269 typedef int32_t (*bypass_wd_timeout_set_t)(struct rte_eth_dev *dev, uint32_t timeout);
1270 typedef int32_t (*bypass_wd_timeout_show_t)(struct rte_eth_dev *dev, uint32_t *wd_timeout);
1271 typedef int32_t (*bypass_ver_show_t)(struct rte_eth_dev *dev, uint32_t *ver);
1272 typedef int32_t (*bypass_wd_reset_t)(struct rte_eth_dev *dev);
1275 typedef int (*eth_add_syn_filter_t)(struct rte_eth_dev *dev,
1276 struct rte_syn_filter *filter, uint16_t rx_queue);
1277 /**< @internal add syn filter rule on an Ethernet device */
1279 typedef int (*eth_remove_syn_filter_t)(struct rte_eth_dev *dev);
1280 /**< @internal remove syn filter rule on an Ethernet device */
1282 typedef int (*eth_get_syn_filter_t)(struct rte_eth_dev *dev,
1283 struct rte_syn_filter *filter, uint16_t *rx_queue);
1284 /**< @internal Get syn filter rule on an Ethernet device */
1286 typedef int (*eth_add_ethertype_filter_t)(struct rte_eth_dev *dev,
1287 uint16_t index, struct rte_ethertype_filter *filter,
1289 /**< @internal Setup a new ethertype filter rule on an Ethernet device */
1291 typedef int (*eth_remove_ethertype_filter_t)(struct rte_eth_dev *dev,
1293 /**< @internal Remove an ethertype filter rule on an Ethernet device */
1295 typedef int (*eth_get_ethertype_filter_t)(struct rte_eth_dev *dev,
1296 uint16_t index, struct rte_ethertype_filter *filter,
1297 uint16_t *rx_queue);
1298 /**< @internal Get an ethertype filter rule on an Ethernet device */
1300 typedef int (*eth_add_2tuple_filter_t)(struct rte_eth_dev *dev,
1301 uint16_t index, struct rte_2tuple_filter *filter,
1303 /**< @internal Setup a new 2tuple filter rule on an Ethernet device */
1305 typedef int (*eth_remove_2tuple_filter_t)(struct rte_eth_dev *dev,
1307 /**< @internal Remove a 2tuple filter rule on an Ethernet device */
1309 typedef int (*eth_get_2tuple_filter_t)(struct rte_eth_dev *dev,
1310 uint16_t index, struct rte_2tuple_filter *filter,
1311 uint16_t *rx_queue);
1312 /**< @internal Get a 2tuple filter rule on an Ethernet device */
1314 typedef int (*eth_add_5tuple_filter_t)(struct rte_eth_dev *dev,
1315 uint16_t index, struct rte_5tuple_filter *filter,
1317 /**< @internal Setup a new 5tuple filter rule on an Ethernet device */
1319 typedef int (*eth_remove_5tuple_filter_t)(struct rte_eth_dev *dev,
1321 /**< @internal Remove a 5tuple filter rule on an Ethernet device */
1323 typedef int (*eth_get_5tuple_filter_t)(struct rte_eth_dev *dev,
1324 uint16_t index, struct rte_5tuple_filter *filter,
1325 uint16_t *rx_queue);
1326 /**< @internal Get a 5tuple filter rule on an Ethernet device */
1328 typedef int (*eth_add_flex_filter_t)(struct rte_eth_dev *dev,
1329 uint16_t index, struct rte_flex_filter *filter,
1331 /**< @internal Setup a new flex filter rule on an Ethernet device */
1333 typedef int (*eth_remove_flex_filter_t)(struct rte_eth_dev *dev,
1335 /**< @internal Remove a flex filter rule on an Ethernet device */
1337 typedef int (*eth_get_flex_filter_t)(struct rte_eth_dev *dev,
1338 uint16_t index, struct rte_flex_filter *filter,
1339 uint16_t *rx_queue);
1340 /**< @internal Get a flex filter rule on an Ethernet device */
1343 * @internal A structure containing the functions exported by an Ethernet driver.
1345 struct eth_dev_ops {
1346 eth_dev_configure_t dev_configure; /**< Configure device. */
1347 eth_dev_start_t dev_start; /**< Start device. */
1348 eth_dev_stop_t dev_stop; /**< Stop device. */
1349 eth_dev_set_link_up_t dev_set_link_up; /**< Device link up. */
1350 eth_dev_set_link_down_t dev_set_link_down; /**< Device link down. */
1351 eth_dev_close_t dev_close; /**< Close device. */
1352 eth_promiscuous_enable_t promiscuous_enable; /**< Promiscuous ON. */
1353 eth_promiscuous_disable_t promiscuous_disable;/**< Promiscuous OFF. */
1354 eth_allmulticast_enable_t allmulticast_enable;/**< RX multicast ON. */
1355 eth_allmulticast_disable_t allmulticast_disable;/**< RX multicast OF. */
1356 eth_link_update_t link_update; /**< Get device link state. */
1357 eth_stats_get_t stats_get; /**< Get device statistics. */
1358 eth_stats_reset_t stats_reset; /**< Reset device statistics. */
1359 eth_queue_stats_mapping_set_t queue_stats_mapping_set;
1360 /**< Configure per queue stat counter mapping. */
1361 eth_dev_infos_get_t dev_infos_get; /**< Get device info. */
1362 vlan_filter_set_t vlan_filter_set; /**< Filter VLAN Setup. */
1363 vlan_tpid_set_t vlan_tpid_set; /**< Outer VLAN TPID Setup. */
1364 vlan_strip_queue_set_t vlan_strip_queue_set; /**< VLAN Stripping on queue. */
1365 vlan_offload_set_t vlan_offload_set; /**< Set VLAN Offload. */
1366 eth_queue_start_t rx_queue_start;/**< Start RX for a queue.*/
1367 eth_queue_stop_t rx_queue_stop;/**< Stop RX for a queue.*/
1368 eth_queue_start_t tx_queue_start;/**< Start TX for a queue.*/
1369 eth_queue_stop_t tx_queue_stop;/**< Stop TX for a queue.*/
1370 eth_rx_queue_setup_t rx_queue_setup;/**< Set up device RX queue.*/
1371 eth_queue_release_t rx_queue_release;/**< Release RX queue.*/
1372 eth_rx_queue_count_t rx_queue_count; /**< Get Rx queue count. */
1373 eth_rx_descriptor_done_t rx_descriptor_done; /**< Check rxd DD bit */
1374 eth_tx_queue_setup_t tx_queue_setup;/**< Set up device TX queue.*/
1375 eth_queue_release_t tx_queue_release;/**< Release TX queue.*/
1376 eth_dev_led_on_t dev_led_on; /**< Turn on LED. */
1377 eth_dev_led_off_t dev_led_off; /**< Turn off LED. */
1378 flow_ctrl_set_t flow_ctrl_set; /**< Setup flow control. */
1379 priority_flow_ctrl_set_t priority_flow_ctrl_set; /**< Setup priority flow control.*/
1380 eth_mac_addr_remove_t mac_addr_remove; /**< Remove MAC address */
1381 eth_mac_addr_add_t mac_addr_add; /**< Add a MAC address */
1382 eth_uc_hash_table_set_t uc_hash_table_set; /**< Set Unicast Table Array */
1383 eth_uc_all_hash_table_set_t uc_all_hash_table_set; /**< Set Unicast hash bitmap */
1384 eth_mirror_rule_set_t mirror_rule_set; /**< Add a traffic mirror rule.*/
1385 eth_mirror_rule_reset_t mirror_rule_reset; /**< reset a traffic mirror rule.*/
1386 eth_set_vf_rx_mode_t set_vf_rx_mode; /**< Set VF RX mode */
1387 eth_set_vf_rx_t set_vf_rx; /**< enable/disable a VF receive */
1388 eth_set_vf_tx_t set_vf_tx; /**< enable/disable a VF transmit */
1389 eth_set_vf_vlan_filter_t set_vf_vlan_filter; /**< Set VF VLAN filter */
1390 eth_set_queue_rate_limit_t set_queue_rate_limit; /**< Set queue rate limit */
1391 eth_set_vf_rate_limit_t set_vf_rate_limit; /**< Set VF rate limit */
1393 /** Add a signature filter. */
1394 fdir_add_signature_filter_t fdir_add_signature_filter;
1395 /** Update a signature filter. */
1396 fdir_update_signature_filter_t fdir_update_signature_filter;
1397 /** Remove a signature filter. */
1398 fdir_remove_signature_filter_t fdir_remove_signature_filter;
1399 /** Get information about FDIR status. */
1400 fdir_infos_get_t fdir_infos_get;
1401 /** Add a perfect filter. */
1402 fdir_add_perfect_filter_t fdir_add_perfect_filter;
1403 /** Update a perfect filter. */
1404 fdir_update_perfect_filter_t fdir_update_perfect_filter;
1405 /** Remove a perfect filter. */
1406 fdir_remove_perfect_filter_t fdir_remove_perfect_filter;
1407 /** Setup masks for FDIR filtering. */
1408 fdir_set_masks_t fdir_set_masks;
1409 /** Update redirection table. */
1410 reta_update_t reta_update;
1411 /** Query redirection table. */
1412 reta_query_t reta_query;
1413 /* bypass control */
1414 #ifdef RTE_NIC_BYPASS
1415 bypass_init_t bypass_init;
1416 bypass_state_set_t bypass_state_set;
1417 bypass_state_show_t bypass_state_show;
1418 bypass_event_set_t bypass_event_set;
1419 bypass_event_show_t bypass_event_show;
1420 bypass_wd_timeout_set_t bypass_wd_timeout_set;
1421 bypass_wd_timeout_show_t bypass_wd_timeout_show;
1422 bypass_ver_show_t bypass_ver_show;
1423 bypass_wd_reset_t bypass_wd_reset;
1426 /** Configure RSS hash protocols. */
1427 rss_hash_update_t rss_hash_update;
1428 /** Get current RSS hash configuration. */
1429 rss_hash_conf_get_t rss_hash_conf_get;
1430 eth_add_syn_filter_t add_syn_filter; /**< add syn filter. */
1431 eth_remove_syn_filter_t remove_syn_filter; /**< remove syn filter. */
1432 eth_get_syn_filter_t get_syn_filter; /**< get syn filter. */
1433 eth_add_ethertype_filter_t add_ethertype_filter; /**< add ethertype filter. */
1434 eth_remove_ethertype_filter_t remove_ethertype_filter; /**< remove ethertype filter. */
1435 eth_get_ethertype_filter_t get_ethertype_filter; /**< get ethertype filter. */
1436 eth_add_2tuple_filter_t add_2tuple_filter; /**< add 2tuple filter. */
1437 eth_remove_2tuple_filter_t remove_2tuple_filter; /**< remove 2tuple filter. */
1438 eth_get_2tuple_filter_t get_2tuple_filter; /**< get 2tuple filter. */
1439 eth_add_5tuple_filter_t add_5tuple_filter; /**< add 5tuple filter. */
1440 eth_remove_5tuple_filter_t remove_5tuple_filter; /**< remove 5tuple filter. */
1441 eth_get_5tuple_filter_t get_5tuple_filter; /**< get 5tuple filter. */
1442 eth_add_flex_filter_t add_flex_filter; /**< add flex filter. */
1443 eth_remove_flex_filter_t remove_flex_filter; /**< remove flex filter. */
1444 eth_get_flex_filter_t get_flex_filter; /**< get flex filter. */
1449 * The generic data structure associated with each ethernet device.
1451 * Pointers to burst-oriented packet receive and transmit functions are
1452 * located at the beginning of the structure, along with the pointer to
1453 * where all the data elements for the particular device are stored in shared
1454 * memory. This split allows the function pointer and driver data to be per-
1455 * process, while the actual configuration data for the device is shared.
1457 struct rte_eth_dev {
1458 eth_rx_burst_t rx_pkt_burst; /**< Pointer to PMD receive function. */
1459 eth_tx_burst_t tx_pkt_burst; /**< Pointer to PMD transmit function. */
1460 struct rte_eth_dev_data *data; /**< Pointer to device data */
1461 const struct eth_driver *driver;/**< Driver for this device */
1462 struct eth_dev_ops *dev_ops; /**< Functions exported by PMD */
1463 struct rte_pci_device *pci_dev; /**< PCI info. supplied by probing */
1464 struct rte_eth_dev_cb_list callbacks; /**< User application callbacks */
1467 struct rte_eth_dev_sriov {
1468 uint8_t active; /**< SRIOV is active with 16, 32 or 64 pools */
1469 uint8_t nb_q_per_pool; /**< rx queue number per pool */
1470 uint16_t def_vmdq_idx; /**< Default pool num used for PF */
1471 uint16_t def_pool_q_idx; /**< Default pool queue start reg index */
1473 #define RTE_ETH_DEV_SRIOV(dev) ((dev)->data->sriov)
1477 * The data part, with no function pointers, associated with each ethernet device.
1479 * This structure is safe to place in shared memory to be common among different
1480 * processes in a multi-process configuration.
1482 struct rte_eth_dev_data {
1483 void **rx_queues; /**< Array of pointers to RX queues. */
1484 void **tx_queues; /**< Array of pointers to TX queues. */
1485 uint16_t nb_rx_queues; /**< Number of RX queues. */
1486 uint16_t nb_tx_queues; /**< Number of TX queues. */
1488 struct rte_eth_dev_sriov sriov; /**< SRIOV data */
1490 void *dev_private; /**< PMD-specific private data */
1492 struct rte_eth_link dev_link;
1493 /**< Link-level information & status */
1495 struct rte_eth_conf dev_conf; /**< Configuration applied to device. */
1496 uint16_t max_frame_size; /**< Default is ETHER_MAX_LEN (1518). */
1498 uint64_t rx_mbuf_alloc_failed; /**< RX ring mbuf allocation failures. */
1499 struct ether_addr* mac_addrs;/**< Device Ethernet Link address. */
1500 uint64_t mac_pool_sel[ETH_NUM_RECEIVE_MAC_ADDR];
1501 /** bitmap array of associating Ethernet MAC addresses to pools */
1502 struct ether_addr* hash_mac_addrs;
1503 /** Device Ethernet MAC addresses of hash filtering. */
1504 uint8_t port_id; /**< Device [external] port identifier. */
1505 uint8_t promiscuous : 1, /**< RX promiscuous mode ON(1) / OFF(0). */
1506 scattered_rx : 1, /**< RX of scattered packets is ON(1) / OFF(0) */
1507 all_multicast : 1, /**< RX all multicast mode ON(1) / OFF(0). */
1508 dev_started : 1; /**< Device state: STARTED(1) / STOPPED(0). */
1513 * The pool of *rte_eth_dev* structures. The size of the pool
1514 * is configured at compile-time in the <rte_ethdev.c> file.
1516 extern struct rte_eth_dev rte_eth_devices[];
1519 * Get the total number of Ethernet devices that have been successfully
1520 * initialized by the [matching] Ethernet driver during the PCI probing phase.
1521 * All devices whose port identifier is in the range
1522 * [0, rte_eth_dev_count() - 1] can be operated on by network applications.
1525 * - The total number of usable Ethernet devices.
1527 extern uint8_t rte_eth_dev_count(void);
1530 * Function for internal use by dummy drivers primarily, e.g. ring-based
1532 * Allocates a new ethdev slot for an ethernet device and returns the pointer
1533 * to that slot for the driver to use.
1536 * - Slot in the rte_dev_devices array for a new device;
1538 struct rte_eth_dev *rte_eth_dev_allocate(void);
1543 * Initialization function of an Ethernet driver invoked for each matching
1544 * Ethernet PCI device detected during the PCI probing phase.
1547 * The pointer to the [matching] Ethernet driver structure supplied by
1548 * the PMD when it registered itself.
1550 * The *eth_dev* pointer is the address of the *rte_eth_dev* structure
1551 * associated with the matching device and which have been [automatically]
1552 * allocated in the *rte_eth_devices* array.
1553 * The *eth_dev* structure is supplied to the driver initialization function
1554 * with the following fields already initialized:
1556 * - *pci_dev*: Holds the pointers to the *rte_pci_device* structure which
1557 * contains the generic PCI information of the matching device.
1559 * - *dev_private*: Holds a pointer to the device private data structure.
1561 * - *max_frame_size*: Contains the default Ethernet maximum frame length
1564 * - *port_id*: Contains the port index of the device (actually the index
1565 * of the *eth_dev* structure in the *rte_eth_devices* array).
1568 * - 0: Success, the device is properly initialized by the driver.
1569 * In particular, the driver MUST have set up the *dev_ops* pointer
1570 * of the *eth_dev* structure.
1571 * - <0: Error code of the device initialization failure.
1573 typedef int (*eth_dev_init_t)(struct eth_driver *eth_drv,
1574 struct rte_eth_dev *eth_dev);
1578 * The structure associated with a PMD Ethernet driver.
1580 * Each Ethernet driver acts as a PCI driver and is represented by a generic
1581 * *eth_driver* structure that holds:
1583 * - An *rte_pci_driver* structure (which must be the first field).
1585 * - The *eth_dev_init* function invoked for each matching PCI device.
1587 * - The size of the private data to allocate for each matching device.
1590 struct rte_pci_driver pci_drv; /**< The PMD is also a PCI driver. */
1591 eth_dev_init_t eth_dev_init; /**< Device init function. */
1592 unsigned int dev_private_size; /**< Size of device private data. */
1597 * A function invoked by the initialization function of an Ethernet driver
1598 * to simultaneously register itself as a PCI driver and as an Ethernet
1599 * Poll Mode Driver (PMD).
1602 * The pointer to the *eth_driver* structure associated with
1603 * the Ethernet driver.
1605 extern void rte_eth_driver_register(struct eth_driver *eth_drv);
1608 * Configure an Ethernet device.
1609 * This function must be invoked first before any other function in the
1610 * Ethernet API. This function can also be re-invoked when a device is in the
1614 * The port identifier of the Ethernet device to configure.
1615 * @param nb_rx_queue
1616 * The number of receive queues to set up for the Ethernet device.
1617 * @param nb_tx_queue
1618 * The number of transmit queues to set up for the Ethernet device.
1620 * The pointer to the configuration data to be used for the Ethernet device.
1621 * The *rte_eth_conf* structure includes:
1622 * - the hardware offload features to activate, with dedicated fields for
1623 * each statically configurable offload hardware feature provided by
1624 * Ethernet devices, such as IP checksum or VLAN tag stripping for
1626 * - the Receive Side Scaling (RSS) configuration when using multiple RX
1629 * Embedding all configuration information in a single data structure
1630 * is the more flexible method that allows the addition of new features
1631 * without changing the syntax of the API.
1633 * - 0: Success, device configured.
1634 * - <0: Error code returned by the driver configuration function.
1636 extern int rte_eth_dev_configure(uint8_t port_id,
1637 uint16_t nb_rx_queue,
1638 uint16_t nb_tx_queue,
1639 const struct rte_eth_conf *eth_conf);
1642 * Allocate and set up a receive queue for an Ethernet device.
1644 * The function allocates a contiguous block of memory for *nb_rx_desc*
1645 * receive descriptors from a memory zone associated with *socket_id*
1646 * and initializes each receive descriptor with a network buffer allocated
1647 * from the memory pool *mb_pool*.
1650 * The port identifier of the Ethernet device.
1651 * @param rx_queue_id
1652 * The index of the receive queue to set up.
1653 * The value must be in the range [0, nb_rx_queue - 1] previously supplied
1654 * to rte_eth_dev_configure().
1656 * The number of receive descriptors to allocate for the receive ring.
1658 * The *socket_id* argument is the socket identifier in case of NUMA.
1659 * The value can be *SOCKET_ID_ANY* if there is no NUMA constraint for
1660 * the DMA memory allocated for the receive descriptors of the ring.
1662 * The pointer to the configuration data to be used for the receive queue.
1663 * The *rx_conf* structure contains an *rx_thresh* structure with the values
1664 * of the Prefetch, Host, and Write-Back threshold registers of the receive
1667 * The pointer to the memory pool from which to allocate *rte_mbuf* network
1668 * memory buffers to populate each descriptor of the receive ring.
1670 * - 0: Success, receive queue correctly set up.
1671 * - -EINVAL: The size of network buffers which can be allocated from the
1672 * memory pool does not fit the various buffer sizes allowed by the
1673 * device controller.
1674 * - -ENOMEM: Unable to allocate the receive ring descriptors or to
1675 * allocate network memory buffers from the memory pool when
1676 * initializing receive descriptors.
1678 extern int rte_eth_rx_queue_setup(uint8_t port_id, uint16_t rx_queue_id,
1679 uint16_t nb_rx_desc, unsigned int socket_id,
1680 const struct rte_eth_rxconf *rx_conf,
1681 struct rte_mempool *mb_pool);
1684 * Allocate and set up a transmit queue for an Ethernet device.
1687 * The port identifier of the Ethernet device.
1688 * @param tx_queue_id
1689 * The index of the transmit queue to set up.
1690 * The value must be in the range [0, nb_tx_queue - 1] previously supplied
1691 * to rte_eth_dev_configure().
1693 * The number of transmit descriptors to allocate for the transmit ring.
1695 * The *socket_id* argument is the socket identifier in case of NUMA.
1696 * Its value can be *SOCKET_ID_ANY* if there is no NUMA constraint for
1697 * the DMA memory allocated for the transmit descriptors of the ring.
1699 * The pointer to the configuration data to be used for the transmit queue.
1700 * The *tx_conf* structure contains the following data:
1701 * - The *tx_thresh* structure with the values of the Prefetch, Host, and
1702 * Write-Back threshold registers of the transmit ring.
1703 * When setting Write-Back threshold to the value greater then zero,
1704 * *tx_rs_thresh* value should be explicitly set to one.
1705 * - The *tx_free_thresh* value indicates the [minimum] number of network
1706 * buffers that must be pending in the transmit ring to trigger their
1707 * [implicit] freeing by the driver transmit function.
1708 * - The *tx_rs_thresh* value indicates the [minimum] number of transmit
1709 * descriptors that must be pending in the transmit ring before setting the
1710 * RS bit on a descriptor by the driver transmit function.
1711 * The *tx_rs_thresh* value should be less or equal then
1712 * *tx_free_thresh* value, and both of them should be less then
1714 * - The *txq_flags* member contains flags to pass to the TX queue setup
1715 * function to configure the behavior of the TX queue. This should be set
1716 * to 0 if no special configuration is required.
1718 * Note that setting *tx_free_thresh* or *tx_rs_thresh* value to 0 forces
1719 * the transmit function to use default values.
1721 * - 0: Success, the transmit queue is correctly set up.
1722 * - -ENOMEM: Unable to allocate the transmit ring descriptors.
1724 extern int rte_eth_tx_queue_setup(uint8_t port_id, uint16_t tx_queue_id,
1725 uint16_t nb_tx_desc, unsigned int socket_id,
1726 const struct rte_eth_txconf *tx_conf);
1729 * Return the NUMA socket to which an Ethernet device is connected
1732 * The port identifier of the Ethernet device
1734 * The NUMA socket id to which the Ethernet device is connected or
1735 * a default of zero if the socket could not be determined.
1736 * -1 is returned is the port_id value is out of range.
1738 extern int rte_eth_dev_socket_id(uint8_t port_id);
1741 * Start specified RX queue of a port
1744 * The port identifier of the Ethernet device
1745 * @param rx_queue_id
1746 * The index of the rx queue to update the ring.
1747 * The value must be in the range [0, nb_rx_queue - 1] previously supplied
1748 * to rte_eth_dev_configure().
1750 * - 0: Success, the transmit queue is correctly set up.
1751 * - -EINVAL: The port_id or the queue_id out of range.
1752 * - -ENOTSUP: The function not supported in PMD driver.
1754 extern int rte_eth_dev_rx_queue_start(uint8_t port_id, uint16_t rx_queue_id);
1757 * Stop specified RX queue of a port
1760 * The port identifier of the Ethernet device
1761 * @param rx_queue_id
1762 * The index of the rx queue to update the ring.
1763 * The value must be in the range [0, nb_rx_queue - 1] previously supplied
1764 * to rte_eth_dev_configure().
1766 * - 0: Success, the transmit queue is correctly set up.
1767 * - -EINVAL: The port_id or the queue_id out of range.
1768 * - -ENOTSUP: The function not supported in PMD driver.
1770 extern int rte_eth_dev_rx_queue_stop(uint8_t port_id, uint16_t rx_queue_id);
1773 * Start specified TX queue of a port
1776 * The port identifier of the Ethernet device
1777 * @param tx_queue_id
1778 * The index of the tx queue to update the ring.
1779 * The value must be in the range [0, nb_tx_queue - 1] previously supplied
1780 * to rte_eth_dev_configure().
1782 * - 0: Success, the transmit queue is correctly set up.
1783 * - -EINVAL: The port_id or the queue_id out of range.
1784 * - -ENOTSUP: The function not supported in PMD driver.
1786 extern int rte_eth_dev_tx_queue_start(uint8_t port_id, uint16_t tx_queue_id);
1789 * Stop specified TX queue of a port
1792 * The port identifier of the Ethernet device
1793 * @param tx_queue_id
1794 * The index of the tx queue to update the ring.
1795 * The value must be in the range [0, nb_tx_queue - 1] previously supplied
1796 * to rte_eth_dev_configure().
1798 * - 0: Success, the transmit queue is correctly set up.
1799 * - -EINVAL: The port_id or the queue_id out of range.
1800 * - -ENOTSUP: The function not supported in PMD driver.
1802 extern int rte_eth_dev_tx_queue_stop(uint8_t port_id, uint16_t tx_queue_id);
1807 * Start an Ethernet device.
1809 * The device start step is the last one and consists of setting the configured
1810 * offload features and in starting the transmit and the receive units of the
1812 * On success, all basic functions exported by the Ethernet API (link status,
1813 * receive/transmit, and so on) can be invoked.
1816 * The port identifier of the Ethernet device.
1818 * - 0: Success, Ethernet device started.
1819 * - <0: Error code of the driver device start function.
1821 extern int rte_eth_dev_start(uint8_t port_id);
1824 * Stop an Ethernet device. The device can be restarted with a call to
1825 * rte_eth_dev_start()
1828 * The port identifier of the Ethernet device.
1830 extern void rte_eth_dev_stop(uint8_t port_id);
1834 * Link up an Ethernet device.
1836 * Set device link up will re-enable the device rx/tx
1837 * functionality after it is previously set device linked down.
1840 * The port identifier of the Ethernet device.
1842 * - 0: Success, Ethernet device linked up.
1843 * - <0: Error code of the driver device link up function.
1845 extern int rte_eth_dev_set_link_up(uint8_t port_id);
1848 * Link down an Ethernet device.
1849 * The device rx/tx functionality will be disabled if success,
1850 * and it can be re-enabled with a call to
1851 * rte_eth_dev_set_link_up()
1854 * The port identifier of the Ethernet device.
1856 extern int rte_eth_dev_set_link_down(uint8_t port_id);
1859 * Close an Ethernet device. The device cannot be restarted!
1862 * The port identifier of the Ethernet device.
1864 extern void rte_eth_dev_close(uint8_t port_id);
1867 * Enable receipt in promiscuous mode for an Ethernet device.
1870 * The port identifier of the Ethernet device.
1872 extern void rte_eth_promiscuous_enable(uint8_t port_id);
1875 * Disable receipt in promiscuous mode for an Ethernet device.
1878 * The port identifier of the Ethernet device.
1880 extern void rte_eth_promiscuous_disable(uint8_t port_id);
1883 * Return the value of promiscuous mode for an Ethernet device.
1886 * The port identifier of the Ethernet device.
1888 * - (1) if promiscuous is enabled
1889 * - (0) if promiscuous is disabled.
1892 extern int rte_eth_promiscuous_get(uint8_t port_id);
1895 * Enable the receipt of any multicast frame by an Ethernet device.
1898 * The port identifier of the Ethernet device.
1900 extern void rte_eth_allmulticast_enable(uint8_t port_id);
1903 * Disable the receipt of all multicast frames by an Ethernet device.
1906 * The port identifier of the Ethernet device.
1908 extern void rte_eth_allmulticast_disable(uint8_t port_id);
1911 * Return the value of allmulticast mode for an Ethernet device.
1914 * The port identifier of the Ethernet device.
1916 * - (1) if allmulticast is enabled
1917 * - (0) if allmulticast is disabled.
1920 extern int rte_eth_allmulticast_get(uint8_t port_id);
1923 * Retrieve the status (ON/OFF), the speed (in Mbps) and the mode (HALF-DUPLEX
1924 * or FULL-DUPLEX) of the physical link of an Ethernet device. It might need
1925 * to wait up to 9 seconds in it.
1928 * The port identifier of the Ethernet device.
1930 * A pointer to an *rte_eth_link* structure to be filled with
1931 * the status, the speed and the mode of the Ethernet device link.
1933 extern void rte_eth_link_get(uint8_t port_id, struct rte_eth_link *link);
1936 * Retrieve the status (ON/OFF), the speed (in Mbps) and the mode (HALF-DUPLEX
1937 * or FULL-DUPLEX) of the physical link of an Ethernet device. It is a no-wait
1938 * version of rte_eth_link_get().
1941 * The port identifier of the Ethernet device.
1943 * A pointer to an *rte_eth_link* structure to be filled with
1944 * the status, the speed and the mode of the Ethernet device link.
1946 extern void rte_eth_link_get_nowait(uint8_t port_id,
1947 struct rte_eth_link *link);
1950 * Retrieve the general I/O statistics of an Ethernet device.
1953 * The port identifier of the Ethernet device.
1955 * A pointer to a structure of type *rte_eth_stats* to be filled with
1956 * the values of device counters for the following set of statistics:
1957 * - *ipackets* with the total of successfully received packets.
1958 * - *opackets* with the total of successfully transmitted packets.
1959 * - *ibytes* with the total of successfully received bytes.
1960 * - *obytes* with the total of successfully transmitted bytes.
1961 * - *ierrors* with the total of erroneous received packets.
1962 * - *oerrors* with the total of failed transmitted packets.
1964 extern void rte_eth_stats_get(uint8_t port_id, struct rte_eth_stats *stats);
1967 * Reset the general I/O statistics of an Ethernet device.
1970 * The port identifier of the Ethernet device.
1972 extern void rte_eth_stats_reset(uint8_t port_id);
1975 * Set a mapping for the specified transmit queue to the specified per-queue
1976 * statistics counter.
1979 * The port identifier of the Ethernet device.
1980 * @param tx_queue_id
1981 * The index of the transmit queue for which a queue stats mapping is required.
1982 * The value must be in the range [0, nb_tx_queue - 1] previously supplied
1983 * to rte_eth_dev_configure().
1985 * The per-queue packet statistics functionality number that the transmit
1986 * queue is to be assigned.
1987 * The value must be in the range [0, RTE_MAX_ETHPORT_QUEUE_STATS_MAPS - 1].
1989 * Zero if successful. Non-zero otherwise.
1991 extern int rte_eth_dev_set_tx_queue_stats_mapping(uint8_t port_id,
1992 uint16_t tx_queue_id,
1996 * Set a mapping for the specified receive queue to the specified per-queue
1997 * statistics counter.
2000 * The port identifier of the Ethernet device.
2001 * @param rx_queue_id
2002 * The index of the receive queue for which a queue stats mapping is required.
2003 * The value must be in the range [0, nb_rx_queue - 1] previously supplied
2004 * to rte_eth_dev_configure().
2006 * The per-queue packet statistics functionality number that the receive
2007 * queue is to be assigned.
2008 * The value must be in the range [0, RTE_MAX_ETHPORT_QUEUE_STATS_MAPS - 1].
2010 * Zero if successful. Non-zero otherwise.
2012 extern int rte_eth_dev_set_rx_queue_stats_mapping(uint8_t port_id,
2013 uint16_t rx_queue_id,
2017 * Retrieve the Ethernet address of an Ethernet device.
2020 * The port identifier of the Ethernet device.
2022 * A pointer to a structure of type *ether_addr* to be filled with
2023 * the Ethernet address of the Ethernet device.
2025 extern void rte_eth_macaddr_get(uint8_t port_id, struct ether_addr *mac_addr);
2028 * Retrieve the contextual information of an Ethernet device.
2031 * The port identifier of the Ethernet device.
2033 * A pointer to a structure of type *rte_eth_dev_info* to be filled with
2034 * the contextual information of the Ethernet device.
2036 extern void rte_eth_dev_info_get(uint8_t port_id,
2037 struct rte_eth_dev_info *dev_info);
2040 * Enable/Disable hardware filtering by an Ethernet device of received
2041 * VLAN packets tagged with a given VLAN Tag Identifier.
2044 * The port identifier of the Ethernet device.
2046 * The VLAN Tag Identifier whose filtering must be enabled or disabled.
2048 * If > 0, enable VLAN filtering of VLAN packets tagged with *vlan_id*.
2049 * Otherwise, disable VLAN filtering of VLAN packets tagged with *vlan_id*.
2051 * - (0) if successful.
2052 * - (-ENOSUP) if hardware-assisted VLAN filtering not configured.
2053 * - (-ENODEV) if *port_id* invalid.
2054 * - (-ENOSYS) if VLAN filtering on *port_id* disabled.
2055 * - (-EINVAL) if *vlan_id* > 4095.
2057 extern int rte_eth_dev_vlan_filter(uint8_t port_id, uint16_t vlan_id , int on);
2060 * Enable/Disable hardware VLAN Strip by a rx queue of an Ethernet device.
2061 * 82599/X540 can support VLAN stripping at the rx queue level
2064 * The port identifier of the Ethernet device.
2065 * @param rx_queue_id
2066 * The index of the receive queue for which a queue stats mapping is required.
2067 * The value must be in the range [0, nb_rx_queue - 1] previously supplied
2068 * to rte_eth_dev_configure().
2070 * If 1, Enable VLAN Stripping of the receive queue of the Ethernet port.
2071 * If 0, Disable VLAN Stripping of the receive queue of the Ethernet port.
2073 * - (0) if successful.
2074 * - (-ENOSUP) if hardware-assisted VLAN stripping not configured.
2075 * - (-ENODEV) if *port_id* invalid.
2076 * - (-EINVAL) if *rx_queue_id* invalid.
2078 extern int rte_eth_dev_set_vlan_strip_on_queue(uint8_t port_id,
2079 uint16_t rx_queue_id, int on);
2082 * Set the Outer VLAN Ether Type by an Ethernet device, it can be inserted to
2083 * the VLAN Header. This is a register setup available on some Intel NIC, not
2084 * but all, please check the data sheet for availability.
2087 * The port identifier of the Ethernet device.
2089 * The Tag Protocol ID
2091 * - (0) if successful.
2092 * - (-ENOSUP) if hardware-assisted VLAN TPID setup is not supported.
2093 * - (-ENODEV) if *port_id* invalid.
2095 extern int rte_eth_dev_set_vlan_ether_type(uint8_t port_id, uint16_t tag_type);
2098 * Set VLAN offload configuration on an Ethernet device
2099 * Enable/Disable Extended VLAN by an Ethernet device, This is a register setup
2100 * available on some Intel NIC, not but all, please check the data sheet for
2102 * Enable/Disable VLAN Strip can be done on rx queue for certain NIC, but here
2103 * the configuration is applied on the port level.
2106 * The port identifier of the Ethernet device.
2107 * @param offload_mask
2108 * The VLAN Offload bit mask can be mixed use with "OR"
2109 * ETH_VLAN_STRIP_OFFLOAD
2110 * ETH_VLAN_FILTER_OFFLOAD
2111 * ETH_VLAN_EXTEND_OFFLOAD
2113 * - (0) if successful.
2114 * - (-ENOSUP) if hardware-assisted VLAN filtering not configured.
2115 * - (-ENODEV) if *port_id* invalid.
2117 extern int rte_eth_dev_set_vlan_offload(uint8_t port_id, int offload_mask);
2120 * Read VLAN Offload configuration from an Ethernet device
2123 * The port identifier of the Ethernet device.
2125 * - (>0) if successful. Bit mask to indicate
2126 * ETH_VLAN_STRIP_OFFLOAD
2127 * ETH_VLAN_FILTER_OFFLOAD
2128 * ETH_VLAN_EXTEND_OFFLOAD
2129 * - (-ENODEV) if *port_id* invalid.
2131 extern int rte_eth_dev_get_vlan_offload(uint8_t port_id);
2135 * Retrieve a burst of input packets from a receive queue of an Ethernet
2136 * device. The retrieved packets are stored in *rte_mbuf* structures whose
2137 * pointers are supplied in the *rx_pkts* array.
2139 * The rte_eth_rx_burst() function loops, parsing the RX ring of the
2140 * receive queue, up to *nb_pkts* packets, and for each completed RX
2141 * descriptor in the ring, it performs the following operations:
2143 * - Initialize the *rte_mbuf* data structure associated with the
2144 * RX descriptor according to the information provided by the NIC into
2145 * that RX descriptor.
2147 * - Store the *rte_mbuf* data structure into the next entry of the
2150 * - Replenish the RX descriptor with a new *rte_mbuf* buffer
2151 * allocated from the memory pool associated with the receive queue at
2152 * initialization time.
2154 * When retrieving an input packet that was scattered by the controller
2155 * into multiple receive descriptors, the rte_eth_rx_burst() function
2156 * appends the associated *rte_mbuf* buffers to the first buffer of the
2159 * The rte_eth_rx_burst() function returns the number of packets
2160 * actually retrieved, which is the number of *rte_mbuf* data structures
2161 * effectively supplied into the *rx_pkts* array.
2162 * A return value equal to *nb_pkts* indicates that the RX queue contained
2163 * at least *rx_pkts* packets, and this is likely to signify that other
2164 * received packets remain in the input queue. Applications implementing
2165 * a "retrieve as much received packets as possible" policy can check this
2166 * specific case and keep invoking the rte_eth_rx_burst() function until
2167 * a value less than *nb_pkts* is returned.
2169 * This receive method has the following advantages:
2171 * - It allows a run-to-completion network stack engine to retrieve and
2172 * to immediately process received packets in a fast burst-oriented
2173 * approach, avoiding the overhead of unnecessary intermediate packet
2174 * queue/dequeue operations.
2176 * - Conversely, it also allows an asynchronous-oriented processing
2177 * method to retrieve bursts of received packets and to immediately
2178 * queue them for further parallel processing by another logical core,
2179 * for instance. However, instead of having received packets being
2180 * individually queued by the driver, this approach allows the invoker
2181 * of the rte_eth_rx_burst() function to queue a burst of retrieved
2182 * packets at a time and therefore dramatically reduce the cost of
2183 * enqueue/dequeue operations per packet.
2185 * - It allows the rte_eth_rx_burst() function of the driver to take
2186 * advantage of burst-oriented hardware features (CPU cache,
2187 * prefetch instructions, and so on) to minimize the number of CPU
2188 * cycles per packet.
2190 * To summarize, the proposed receive API enables many
2191 * burst-oriented optimizations in both synchronous and asynchronous
2192 * packet processing environments with no overhead in both cases.
2194 * The rte_eth_rx_burst() function does not provide any error
2195 * notification to avoid the corresponding overhead. As a hint, the
2196 * upper-level application might check the status of the device link once
2197 * being systematically returned a 0 value for a given number of tries.
2200 * The port identifier of the Ethernet device.
2202 * The index of the receive queue from which to retrieve input packets.
2203 * The value must be in the range [0, nb_rx_queue - 1] previously supplied
2204 * to rte_eth_dev_configure().
2206 * The address of an array of pointers to *rte_mbuf* structures that
2207 * must be large enough to store *nb_pkts* pointers in it.
2209 * The maximum number of packets to retrieve.
2211 * The number of packets actually retrieved, which is the number
2212 * of pointers to *rte_mbuf* structures effectively supplied to the
2215 #ifdef RTE_LIBRTE_ETHDEV_DEBUG
2216 extern uint16_t rte_eth_rx_burst(uint8_t port_id, uint16_t queue_id,
2217 struct rte_mbuf **rx_pkts, uint16_t nb_pkts);
2219 static inline uint16_t
2220 rte_eth_rx_burst(uint8_t port_id, uint16_t queue_id,
2221 struct rte_mbuf **rx_pkts, uint16_t nb_pkts)
2223 struct rte_eth_dev *dev;
2225 dev = &rte_eth_devices[port_id];
2226 return (*dev->rx_pkt_burst)(dev->data->rx_queues[queue_id], rx_pkts, nb_pkts);
2231 * Get the number of used descriptors in a specific queue
2234 * The port identifier of the Ethernet device.
2236 * The queue id on the specific port.
2238 * The number of used descriptors in the specific queue.
2240 #ifdef RTE_LIBRTE_ETHDEV_DEBUG
2241 extern uint32_t rte_eth_rx_queue_count(uint8_t port_id, uint16_t queue_id);
2243 static inline uint32_t
2244 rte_eth_rx_queue_count(uint8_t port_id, uint16_t queue_id)
2246 struct rte_eth_dev *dev;
2248 dev = &rte_eth_devices[port_id];
2249 return (*dev->dev_ops->rx_queue_count)(dev, queue_id);
2254 * Check if the DD bit of the specific RX descriptor in the queue has been set
2257 * The port identifier of the Ethernet device.
2259 * The queue id on the specific port.
2261 * The offset of the descriptor ID from tail.
2263 * - (1) if the specific DD bit is set.
2264 * - (0) if the specific DD bit is not set.
2265 * - (-ENODEV) if *port_id* invalid.
2267 #ifdef RTE_LIBRTE_ETHDEV_DEBUG
2268 extern int rte_eth_rx_descriptor_done(uint8_t port_id,
2273 rte_eth_rx_descriptor_done(uint8_t port_id, uint16_t queue_id, uint16_t offset)
2275 struct rte_eth_dev *dev;
2277 dev = &rte_eth_devices[port_id];
2278 return (*dev->dev_ops->rx_descriptor_done)( \
2279 dev->data->rx_queues[queue_id], offset);
2284 * Send a burst of output packets on a transmit queue of an Ethernet device.
2286 * The rte_eth_tx_burst() function is invoked to transmit output packets
2287 * on the output queue *queue_id* of the Ethernet device designated by its
2289 * The *nb_pkts* parameter is the number of packets to send which are
2290 * supplied in the *tx_pkts* array of *rte_mbuf* structures.
2291 * The rte_eth_tx_burst() function loops, sending *nb_pkts* packets,
2292 * up to the number of transmit descriptors available in the TX ring of the
2294 * For each packet to send, the rte_eth_tx_burst() function performs
2295 * the following operations:
2297 * - Pick up the next available descriptor in the transmit ring.
2299 * - Free the network buffer previously sent with that descriptor, if any.
2301 * - Initialize the transmit descriptor with the information provided
2302 * in the *rte_mbuf data structure.
2304 * In the case of a segmented packet composed of a list of *rte_mbuf* buffers,
2305 * the rte_eth_tx_burst() function uses several transmit descriptors
2308 * The rte_eth_tx_burst() function returns the number of packets it
2309 * actually sent. A return value equal to *nb_pkts* means that all packets
2310 * have been sent, and this is likely to signify that other output packets
2311 * could be immediately transmitted again. Applications that implement a
2312 * "send as many packets to transmit as possible" policy can check this
2313 * specific case and keep invoking the rte_eth_tx_burst() function until
2314 * a value less than *nb_pkts* is returned.
2316 * It is the responsibility of the rte_eth_tx_burst() function to
2317 * transparently free the memory buffers of packets previously sent.
2318 * This feature is driven by the *tx_free_thresh* value supplied to the
2319 * rte_eth_dev_configure() function at device configuration time.
2320 * When the number of previously sent packets reached the "minimum transmit
2321 * packets to free" threshold, the rte_eth_tx_burst() function must
2322 * [attempt to] free the *rte_mbuf* buffers of those packets whose
2323 * transmission was effectively completed.
2326 * The port identifier of the Ethernet device.
2328 * The index of the transmit queue through which output packets must be
2330 * The value must be in the range [0, nb_tx_queue - 1] previously supplied
2331 * to rte_eth_dev_configure().
2333 * The address of an array of *nb_pkts* pointers to *rte_mbuf* structures
2334 * which contain the output packets.
2336 * The maximum number of packets to transmit.
2338 * The number of output packets actually stored in transmit descriptors of
2339 * the transmit ring. The return value can be less than the value of the
2340 * *tx_pkts* parameter when the transmit ring is full or has been filled up.
2342 #ifdef RTE_LIBRTE_ETHDEV_DEBUG
2343 extern uint16_t rte_eth_tx_burst(uint8_t port_id, uint16_t queue_id,
2344 struct rte_mbuf **tx_pkts, uint16_t nb_pkts);
2346 static inline uint16_t
2347 rte_eth_tx_burst(uint8_t port_id, uint16_t queue_id,
2348 struct rte_mbuf **tx_pkts, uint16_t nb_pkts)
2350 struct rte_eth_dev *dev;
2352 dev = &rte_eth_devices[port_id];
2353 return (*dev->tx_pkt_burst)(dev->data->tx_queues[queue_id], tx_pkts, nb_pkts);
2358 * Setup a new signature filter rule on an Ethernet device
2361 * The port identifier of the Ethernet device.
2362 * @param fdir_filter
2363 * The pointer to the fdir filter structure describing the signature filter
2365 * The *rte_fdir_filter* structure includes the values of the different fields
2366 * to match: source and destination IP addresses, vlan id, flexbytes, source
2367 * and destination ports, and so on.
2369 * The index of the RX queue where to store RX packets matching the added
2370 * signature filter defined in fdir_filter.
2372 * - (0) if successful.
2373 * - (-ENOTSUP) if hardware doesn't support flow director mode.
2374 * - (-ENODEV) if *port_id* invalid.
2375 * - (-ENOSYS) if the FDIR mode is not configured in signature mode
2377 * - (-EINVAL) if the fdir_filter information is not correct.
2379 int rte_eth_dev_fdir_add_signature_filter(uint8_t port_id,
2380 struct rte_fdir_filter *fdir_filter,
2384 * Update a signature filter rule on an Ethernet device.
2385 * If the rule doesn't exits, it is created.
2388 * The port identifier of the Ethernet device.
2390 * The pointer to the structure describing the signature filter rule.
2391 * The *rte_fdir_filter* structure includes the values of the different fields
2392 * to match: source and destination IP addresses, vlan id, flexbytes, source
2393 * and destination ports, and so on.
2395 * The index of the RX queue where to store RX packets matching the added
2396 * signature filter defined in fdir_ftr.
2398 * - (0) if successful.
2399 * - (-ENOTSUP) if hardware doesn't support flow director mode.
2400 * - (-ENODEV) if *port_id* invalid.
2401 * - (-ENOSYS) if the flow director mode is not configured in signature mode
2403 * - (-EINVAL) if the fdir_filter information is not correct.
2405 int rte_eth_dev_fdir_update_signature_filter(uint8_t port_id,
2406 struct rte_fdir_filter *fdir_ftr,
2410 * Remove a signature filter rule on an Ethernet device.
2413 * The port identifier of the Ethernet device.
2415 * The pointer to the structure describing the signature filter rule.
2416 * The *rte_fdir_filter* structure includes the values of the different fields
2417 * to match: source and destination IP addresses, vlan id, flexbytes, source
2418 * and destination ports, and so on.
2420 * - (0) if successful.
2421 * - (-ENOTSUP) if hardware doesn't support flow director mode.
2422 * - (-ENODEV) if *port_id* invalid.
2423 * - (-ENOSYS) if the flow director mode is not configured in signature mode
2425 * - (-EINVAL) if the fdir_filter information is not correct.
2427 int rte_eth_dev_fdir_remove_signature_filter(uint8_t port_id,
2428 struct rte_fdir_filter *fdir_ftr);
2431 * Retrieve the flow director information of an Ethernet device.
2434 * The port identifier of the Ethernet device.
2436 * A pointer to a structure of type *rte_eth_dev_fdir* to be filled with
2437 * the flow director information of the Ethernet device.
2439 * - (0) if successful.
2440 * - (-ENOTSUP) if hardware doesn't support flow director mode.
2441 * - (-ENODEV) if *port_id* invalid.
2442 * - (-ENOSYS) if the flow director mode is not configured on *port_id*.
2444 int rte_eth_dev_fdir_get_infos(uint8_t port_id, struct rte_eth_fdir *fdir);
2447 * Add a new perfect filter rule on an Ethernet device.
2450 * The port identifier of the Ethernet device.
2451 * @param fdir_filter
2452 * The pointer to the structure describing the perfect filter rule.
2453 * The *rte_fdir_filter* structure includes the values of the different fields
2454 * to match: source and destination IP addresses, vlan id, flexbytes, source
2455 * and destination ports, and so on.
2456 * IPv6 are not supported.
2458 * The 16-bit value supplied in the field hash.fdir.id of mbuf for RX
2459 * packets matching the perfect filter.
2461 * The index of the RX queue where to store RX packets matching the added
2462 * perfect filter defined in fdir_filter.
2464 * If drop is set to 1, matching RX packets are stored into the RX drop
2465 * queue defined in the rte_fdir_conf.
2467 * - (0) if successful.
2468 * - (-ENOTSUP) if hardware doesn't support flow director mode.
2469 * - (-ENODEV) if *port_id* invalid.
2470 * - (-ENOSYS) if the flow director mode is not configured in perfect mode
2472 * - (-EINVAL) if the fdir_filter information is not correct.
2474 int rte_eth_dev_fdir_add_perfect_filter(uint8_t port_id,
2475 struct rte_fdir_filter *fdir_filter,
2476 uint16_t soft_id, uint8_t rx_queue,
2480 * Update a perfect filter rule on an Ethernet device.
2481 * If the rule doesn't exits, it is created.
2484 * The port identifier of the Ethernet device.
2485 * @param fdir_filter
2486 * The pointer to the structure describing the perfect filter rule.
2487 * The *rte_fdir_filter* structure includes the values of the different fields
2488 * to match: source and destination IP addresses, vlan id, flexbytes, source
2489 * and destination ports, and so on.
2490 * IPv6 are not supported.
2492 * The 16-bit value supplied in the field hash.fdir.id of mbuf for RX
2493 * packets matching the perfect filter.
2495 * The index of the RX queue where to store RX packets matching the added
2496 * perfect filter defined in fdir_filter.
2498 * If drop is set to 1, matching RX packets are stored into the RX drop
2499 * queue defined in the rte_fdir_conf.
2501 * - (0) if successful.
2502 * - (-ENOTSUP) if hardware doesn't support flow director mode.
2503 * - (-ENODEV) if *port_id* invalid.
2504 * - (-ENOSYS) if the flow director mode is not configured in perfect mode
2506 * - (-EINVAL) if the fdir_filter information is not correct.
2508 int rte_eth_dev_fdir_update_perfect_filter(uint8_t port_id,
2509 struct rte_fdir_filter *fdir_filter,
2510 uint16_t soft_id, uint8_t rx_queue,
2514 * Remove a perfect filter rule on an Ethernet device.
2517 * The port identifier of the Ethernet device.
2518 * @param fdir_filter
2519 * The pointer to the structure describing the perfect 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.
2523 * IPv6 are not supported.
2525 * The soft_id value provided when adding/updating the removed filter.
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 perfect mode
2532 * - (-EINVAL) if the fdir_filter information is not correct.
2534 int rte_eth_dev_fdir_remove_perfect_filter(uint8_t port_id,
2535 struct rte_fdir_filter *fdir_filter,
2538 * Configure globally the masks for flow director mode for an Ethernet device.
2539 * For example, the device can match packets with only the first 24 bits of
2540 * the IPv4 source address.
2542 * The following fields can be masked: IPv4 addresses and L4 port numbers.
2543 * The following fields can be either enabled or disabled completely for the
2544 * matching functionality: VLAN ID tag; VLAN Priority + CFI bit; Flexible 2-byte
2546 * IPv6 masks are not supported.
2548 * All filters must comply with the masks previously configured.
2549 * For example, with a mask equal to 255.255.255.0 for the source IPv4 address,
2550 * all IPv4 filters must be created with a source IPv4 address that fits the
2553 * This function flushes all filters that have been previously added in
2557 * The port identifier of the Ethernet device.
2559 * The pointer to the fdir mask structure describing relevant headers fields
2560 * and relevant bits to use when matching packets addresses and ports.
2561 * IPv6 masks are not supported.
2563 * - (0) if successful.
2564 * - (-ENOTSUP) if hardware doesn't support flow director mode.
2565 * - (-ENODEV) if *port_id* invalid.
2566 * - (-ENOSYS) if the flow director mode is not configured in perfect
2567 * mode on *port_id*.
2568 * - (-EINVAL) if the fdir_filter information is not correct
2570 int rte_eth_dev_fdir_set_masks(uint8_t port_id,
2571 struct rte_fdir_masks *fdir_mask);
2574 * The eth device event type for interrupt, and maybe others in the future.
2576 enum rte_eth_event_type {
2577 RTE_ETH_EVENT_UNKNOWN, /**< unknown event type */
2578 RTE_ETH_EVENT_INTR_LSC, /**< lsc interrupt event */
2579 RTE_ETH_EVENT_MAX /**< max value of this enum */
2582 typedef void (*rte_eth_dev_cb_fn)(uint8_t port_id, \
2583 enum rte_eth_event_type event, void *cb_arg);
2584 /**< user application callback to be registered for interrupts */
2589 * Register a callback function for specific port id.
2596 * User supplied callback function to be called.
2598 * Pointer to the parameters for the registered callback.
2601 * - On success, zero.
2602 * - On failure, a negative value.
2604 int rte_eth_dev_callback_register(uint8_t port_id,
2605 enum rte_eth_event_type event,
2606 rte_eth_dev_cb_fn cb_fn, void *cb_arg);
2609 * Unregister a callback function for specific port id.
2616 * User supplied callback function to be called.
2618 * Pointer to the parameters for the registered callback. -1 means to
2619 * remove all for the same callback address and same event.
2622 * - On success, zero.
2623 * - On failure, a negative value.
2625 int rte_eth_dev_callback_unregister(uint8_t port_id,
2626 enum rte_eth_event_type event,
2627 rte_eth_dev_cb_fn cb_fn, void *cb_arg);
2630 * @internal Executes all the user application registered callbacks for
2631 * the specific device. It is for DPDK internal user only. User
2632 * application should not call it directly.
2635 * Pointer to struct rte_eth_dev.
2637 * Eth device interrupt event type.
2642 void _rte_eth_dev_callback_process(struct rte_eth_dev *dev,
2643 enum rte_eth_event_type event);
2646 * Turn on the LED on the Ethernet device.
2647 * This function turns on the LED on the Ethernet device.
2650 * The port identifier of the Ethernet device.
2652 * - (0) if successful.
2653 * - (-ENOTSUP) if underlying hardware OR driver doesn't support
2655 * - (-ENODEV) if *port_id* invalid.
2657 int rte_eth_led_on(uint8_t port_id);
2660 * Turn off the LED on the Ethernet device.
2661 * This function turns off the LED on the Ethernet device.
2664 * The port identifier of the Ethernet device.
2666 * - (0) if successful.
2667 * - (-ENOTSUP) if underlying hardware OR driver doesn't support
2669 * - (-ENODEV) if *port_id* invalid.
2671 int rte_eth_led_off(uint8_t port_id);
2674 * Configure the Ethernet link flow control for Ethernet device
2677 * The port identifier of the Ethernet device.
2679 * The pointer to the structure of the flow control parameters.
2681 * - (0) if successful.
2682 * - (-ENOTSUP) if hardware doesn't support flow control mode.
2683 * - (-ENODEV) if *port_id* invalid.
2684 * - (-EINVAL) if bad parameter
2685 * - (-EIO) if flow control setup failure
2687 int rte_eth_dev_flow_ctrl_set(uint8_t port_id,
2688 struct rte_eth_fc_conf *fc_conf);
2691 * Configure the Ethernet priority flow control under DCB environment
2692 * for Ethernet device.
2695 * The port identifier of the Ethernet device.
2697 * The pointer to the structure of the priority flow control parameters.
2699 * - (0) if successful.
2700 * - (-ENOTSUP) if hardware doesn't support priority flow control mode.
2701 * - (-ENODEV) if *port_id* invalid.
2702 * - (-EINVAL) if bad parameter
2703 * - (-EIO) if flow control setup failure
2705 int rte_eth_dev_priority_flow_ctrl_set(uint8_t port_id,
2706 struct rte_eth_pfc_conf *pfc_conf);
2709 * Add a MAC address to an internal array of addresses used to enable whitelist
2710 * filtering to accept packets only if the destination MAC address matches.
2713 * The port identifier of the Ethernet device.
2715 * The MAC address to add.
2717 * VMDq pool index to associate address with (if VMDq is enabled). If VMDq is
2718 * not enabled, this should be set to 0.
2720 * - (0) if successfully added or *mac_addr" was already added.
2721 * - (-ENOTSUP) if hardware doesn't support this feature.
2722 * - (-ENODEV) if *port* is invalid.
2723 * - (-ENOSPC) if no more MAC addresses can be added.
2724 * - (-EINVAL) if MAC address is invalid.
2726 int rte_eth_dev_mac_addr_add(uint8_t port, struct ether_addr *mac_addr,
2730 * Remove a MAC address from the internal array of addresses.
2733 * The port identifier of the Ethernet device.
2735 * MAC address to remove.
2737 * - (0) if successful, or *mac_addr* didn't exist.
2738 * - (-ENOTSUP) if hardware doesn't support.
2739 * - (-ENODEV) if *port* invalid.
2740 * - (-EADDRINUSE) if attempting to remove the default MAC address
2742 int rte_eth_dev_mac_addr_remove(uint8_t port, struct ether_addr *mac_addr);
2745 * Update Redirection Table(RETA) of Receive Side Scaling of Ethernet device.
2748 * The port identifier of the Ethernet device.
2752 * - (0) if successful.
2753 * - (-ENOTSUP) if hardware doesn't support.
2754 * - (-EINVAL) if bad parameter.
2756 int rte_eth_dev_rss_reta_update(uint8_t port,
2757 struct rte_eth_rss_reta *reta_conf);
2760 * Query Redirection Table(RETA) of Receive Side Scaling of Ethernet device.
2763 * The port identifier of the Ethernet device.
2767 * - (0) if successful.
2768 * - (-ENOTSUP) if hardware doesn't support.
2769 * - (-EINVAL) if bad parameter.
2771 int rte_eth_dev_rss_reta_query(uint8_t port,
2772 struct rte_eth_rss_reta *reta_conf);
2775 * Updates unicast hash table for receiving packet with the given destination
2776 * MAC address, and the packet is routed to all VFs for which the RX mode is
2777 * accept packets that match the unicast hash table.
2780 * The port identifier of the Ethernet device.
2782 * Unicast MAC address.
2784 * 1 - Set an unicast hash bit for receiving packets with the MAC address.
2785 * 0 - Clear an unicast hash bit.
2787 * - (0) if successful.
2788 * - (-ENOTSUP) if hardware doesn't support.
2789 * - (-ENODEV) if *port_id* invalid.
2790 * - (-EINVAL) if bad parameter.
2792 int rte_eth_dev_uc_hash_table_set(uint8_t port,struct ether_addr *addr,
2796 * Updates all unicast hash bitmaps for receiving packet with any Unicast
2797 * Ethernet MAC addresses,the packet is routed to all VFs for which the RX
2798 * mode is accept packets that match the unicast hash table.
2801 * The port identifier of the Ethernet device.
2803 * 1 - Set all unicast hash bitmaps for receiving all the Ethernet
2805 * 0 - Clear all unicast hash bitmaps
2807 * - (0) if successful.
2808 * - (-ENOTSUP) if hardware doesn't support.
2809 * - (-ENODEV) if *port_id* invalid.
2810 * - (-EINVAL) if bad parameter.
2812 int rte_eth_dev_uc_all_hash_table_set(uint8_t port,uint8_t on);
2815 * Set RX L2 Filtering mode of a VF of an Ethernet device.
2818 * The port identifier of the Ethernet device.
2822 * The RX mode mask, which is one or more of accepting Untagged Packets,
2823 * packets that match the PFUTA table, Broadcast and Multicast Promiscuous.
2824 * ETH_VMDQ_ACCEPT_UNTAG,ETH_VMDQ_ACCEPT_HASH_UC,
2825 * ETH_VMDQ_ACCEPT_BROADCAST and ETH_VMDQ_ACCEPT_MULTICAST will be used
2828 * 1 - Enable a VF RX mode.
2829 * 0 - Disable a VF RX mode.
2831 * - (0) if successful.
2832 * - (-ENOTSUP) if hardware doesn't support.
2833 * - (-ENOTSUP) if hardware doesn't support.
2834 * - (-EINVAL) if bad parameter.
2836 int rte_eth_dev_set_vf_rxmode(uint8_t port, uint16_t vf, uint16_t rx_mode,
2840 * Enable or disable a VF traffic transmit of the Ethernet device.
2843 * The port identifier of the Ethernet device.
2847 * 1 - Enable a VF traffic transmit.
2848 * 0 - Disable a VF traffic transmit.
2850 * - (0) if successful.
2851 * - (-ENODEV) if *port_id* invalid.
2852 * - (-ENOTSUP) if hardware doesn't support.
2853 * - (-EINVAL) if bad parameter.
2856 rte_eth_dev_set_vf_tx(uint8_t port,uint16_t vf, uint8_t on);
2859 * Enable or disable a VF traffic receive of an Ethernet device.
2862 * The port identifier of the Ethernet device.
2866 * 1 - Enable a VF traffic receive.
2867 * 0 - Disable a VF traffic receive.
2869 * - (0) if successful.
2870 * - (-ENOTSUP) if hardware doesn't support.
2871 * - (-ENODEV) if *port_id* invalid.
2872 * - (-EINVAL) if bad parameter.
2875 rte_eth_dev_set_vf_rx(uint8_t port,uint16_t vf, uint8_t on);
2878 * Enable/Disable hardware VF VLAN filtering by an Ethernet device of
2879 * received VLAN packets tagged with a given VLAN Tag Identifier.
2882 * The port identifier of the Ethernet device.
2884 * The VLAN Tag Identifier whose filtering must be enabled or disabled.
2886 * Bitmap listing which VFs participate in the VLAN filtering.
2888 * 1 - Enable VFs VLAN filtering.
2889 * 0 - Disable VFs VLAN filtering.
2891 * - (0) if successful.
2892 * - (-ENOTSUP) if hardware doesn't support.
2893 * - (-ENODEV) if *port_id* invalid.
2894 * - (-EINVAL) if bad parameter.
2897 rte_eth_dev_set_vf_vlan_filter(uint8_t port, uint16_t vlan_id,
2902 * Set a traffic mirroring rule on an Ethernet device
2905 * The port identifier of the Ethernet device.
2906 * @param mirror_conf
2907 * The pointer to the traffic mirroring structure describing the mirroring rule.
2908 * The *rte_eth_vm_mirror_conf* structure includes the type of mirroring rule,
2909 * destination pool and the value of rule if enable vlan or pool mirroring.
2912 * The index of traffic mirroring rule, we support four separated rules.
2914 * 1 - Enable a mirroring rule.
2915 * 0 - Disable a mirroring rule.
2917 * - (0) if successful.
2918 * - (-ENOTSUP) if hardware doesn't support this feature.
2919 * - (-ENODEV) if *port_id* invalid.
2920 * - (-EINVAL) if the mr_conf information is not correct.
2922 int rte_eth_mirror_rule_set(uint8_t port_id,
2923 struct rte_eth_vmdq_mirror_conf *mirror_conf,
2928 * Reset a traffic mirroring rule on an Ethernet device.
2931 * The port identifier of the Ethernet device.
2933 * The index of traffic mirroring rule, we support four separated rules.
2935 * - (0) if successful.
2936 * - (-ENOTSUP) if hardware doesn't support this feature.
2937 * - (-ENODEV) if *port_id* invalid.
2938 * - (-EINVAL) if bad parameter.
2940 int rte_eth_mirror_rule_reset(uint8_t port_id,
2944 * Set the rate limitation for a queue on an Ethernet device.
2947 * The port identifier of the Ethernet device.
2951 * The tx rate allocated from the total link speed for this queue.
2953 * - (0) if successful.
2954 * - (-ENOTSUP) if hardware doesn't support this feature.
2955 * - (-ENODEV) if *port_id* invalid.
2956 * - (-EINVAL) if bad parameter.
2958 int rte_eth_set_queue_rate_limit(uint8_t port_id, uint16_t queue_idx,
2962 * Set the rate limitation for a vf on an Ethernet device.
2965 * The port identifier of the Ethernet device.
2969 * The tx rate allocated from the total link speed for this VF id.
2971 * The queue mask which need to set the rate.
2973 * - (0) if successful.
2974 * - (-ENOTSUP) if hardware doesn't support this feature.
2975 * - (-ENODEV) if *port_id* invalid.
2976 * - (-EINVAL) if bad parameter.
2978 int rte_eth_set_vf_rate_limit(uint8_t port_id, uint16_t vf,
2979 uint16_t tx_rate, uint64_t q_msk);
2982 * Initialize bypass logic. This function needs to be called before
2983 * executing any other bypass API.
2986 * The port identifier of the Ethernet device.
2988 * - (0) if successful.
2989 * - (-ENOTSUP) if hardware doesn't support.
2990 * - (-EINVAL) if bad parameter.
2992 int rte_eth_dev_bypass_init(uint8_t port);
2995 * Return bypass state.
2998 * The port identifier of the Ethernet device.
3000 * The return bypass state.
3003 * - (3) Isolate mode
3005 * - (0) if successful.
3006 * - (-ENOTSUP) if hardware doesn't support.
3007 * - (-EINVAL) if bad parameter.
3009 int rte_eth_dev_bypass_state_show(uint8_t port, uint32_t *state);
3015 * The port identifier of the Ethernet device.
3017 * The current bypass state.
3020 * - (3) Isolate mode
3022 * - (0) if successful.
3023 * - (-ENOTSUP) if hardware doesn't support.
3024 * - (-EINVAL) if bad parameter.
3026 int rte_eth_dev_bypass_state_set(uint8_t port, uint32_t *new_state);
3029 * Return bypass state when given event occurs.
3032 * The port identifier of the Ethernet device.
3035 * - (1) Main power on (power button is pushed)
3036 * - (2) Auxiliary power on (power supply is being plugged)
3037 * - (3) Main power off (system shutdown and power supply is left plugged in)
3038 * - (4) Auxiliary power off (power supply is being unplugged)
3039 * - (5) Display or set the watchdog timer
3041 * The bypass state when given event occurred.
3044 * - (3) Isolate mode
3046 * - (0) if successful.
3047 * - (-ENOTSUP) if hardware doesn't support.
3048 * - (-EINVAL) if bad parameter.
3050 int rte_eth_dev_bypass_event_show(uint8_t port, uint32_t event, uint32_t *state);
3053 * Set bypass state when given event occurs.
3056 * The port identifier of the Ethernet device.
3059 * - (1) Main power on (power button is pushed)
3060 * - (2) Auxiliary power on (power supply is being plugged)
3061 * - (3) Main power off (system shutdown and power supply is left plugged in)
3062 * - (4) Auxiliary power off (power supply is being unplugged)
3063 * - (5) Display or set the watchdog timer
3065 * The assigned state when given event occurs.
3068 * - (3) Isolate mode
3070 * - (0) if successful.
3071 * - (-ENOTSUP) if hardware doesn't support.
3072 * - (-EINVAL) if bad parameter.
3074 int rte_eth_dev_bypass_event_store(uint8_t port, uint32_t event, uint32_t state);
3077 * Set bypass watchdog timeout count.
3080 * The port identifier of the Ethernet device.
3082 * The timeout to be set.
3083 * - (0) 0 seconds (timer is off)
3092 * - (0) if successful.
3093 * - (-ENOTSUP) if hardware doesn't support.
3094 * - (-EINVAL) if bad parameter.
3096 int rte_eth_dev_wd_timeout_store(uint8_t port, uint32_t timeout);
3099 * Get bypass firmware version.
3102 * The port identifier of the Ethernet device.
3104 * The firmware version
3106 * - (0) if successful.
3107 * - (-ENOTSUP) if hardware doesn't support.
3108 * - (-EINVAL) if bad parameter.
3110 int rte_eth_dev_bypass_ver_show(uint8_t port, uint32_t *ver);
3113 * Return bypass watchdog timeout in seconds
3116 * The port identifier of the Ethernet device.
3118 * The return watchdog timeout. "0" represents timer expired
3119 * - (0) 0 seconds (timer is off)
3128 * - (0) if successful.
3129 * - (-ENOTSUP) if hardware doesn't support.
3130 * - (-EINVAL) if bad parameter.
3132 int rte_eth_dev_bypass_wd_timeout_show(uint8_t port, uint32_t *wd_timeout);
3135 * Reset bypass watchdog timer
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_wd_reset(uint8_t port);
3147 * Configuration of Receive Side Scaling hash computation of Ethernet device.
3150 * The port identifier of the Ethernet device.
3152 * The new configuration to use for RSS hash computation on the port.
3154 * - (0) if successful.
3155 * - (-ENODEV) if port identifier is invalid.
3156 * - (-ENOTSUP) if hardware doesn't support.
3157 * - (-EINVAL) if bad parameter.
3159 int rte_eth_dev_rss_hash_update(uint8_t port_id,
3160 struct rte_eth_rss_conf *rss_conf);
3163 * Retrieve current configuration of Receive Side Scaling hash computation
3164 * of Ethernet device.
3167 * The port identifier of the Ethernet device.
3169 * Where to store the current RSS hash configuration of the Ethernet device.
3171 * - (0) if successful.
3172 * - (-ENODEV) if port identifier is invalid.
3173 * - (-ENOTSUP) if hardware doesn't support RSS.
3176 rte_eth_dev_rss_hash_conf_get(uint8_t port_id,
3177 struct rte_eth_rss_conf *rss_conf);
3183 * The port identifier of the Ethernet device.
3185 * The index of RX queue where to store RX packets matching the syn filter.
3187 * The pointer to the structure describing the syn filter rule.
3189 * - (0) if successful.
3190 * - (-ENOTSUP) if hardware doesn't support.
3191 * - (-EINVAL) if bad parameter.
3193 int rte_eth_dev_add_syn_filter(uint8_t port_id,
3194 struct rte_syn_filter *filter, uint16_t rx_queue);
3200 * The port identifier of the Ethernet device.
3202 * - (0) if successful.
3203 * - (-ENOTSUP) if hardware doesn't support.
3204 * - (-EINVAL) if bad parameter.
3206 int rte_eth_dev_remove_syn_filter(uint8_t port_id);
3212 * The port identifier of the Ethernet device.
3214 * The pointer to the structure describing the syn filter.
3216 * A pointer to get the queue index of syn filter.
3218 * - (0) if successful.
3219 * - (-ENOTSUP) if hardware doesn't support.
3220 * - (-EINVAL) if bad parameter.
3222 int rte_eth_dev_get_syn_filter(uint8_t port_id,
3223 struct rte_syn_filter *filter, uint16_t *rx_queue);
3226 * Add a new ethertype filter rule on an Ethernet device.
3229 * The port identifier of the Ethernet device.
3231 * The identifier of ethertype filter.
3233 * The pointer to the structure describing the ethertype filter rule.
3234 * The *rte_ethertype_filter* structure includes the values of the different
3235 * fields to match: ethertype and priority in vlan tag.
3236 * priority in vlan tag is not supported for E1000 dev.
3238 * The index of the RX queue where to store RX packets matching the added
3241 * - (0) if successful.
3242 * - (-ENOTSUP) if hardware doesn't support ethertype filter.
3243 * - (-ENODEV) if *port_id* invalid.
3244 * - (-EINVAL) if the filter information is not correct.
3246 int rte_eth_dev_add_ethertype_filter(uint8_t port_id, uint16_t index,
3247 struct rte_ethertype_filter *filter, uint16_t rx_queue);
3250 * remove an ethertype filter rule on an Ethernet device.
3253 * The port identifier of the Ethernet device.
3255 * The identifier of ethertype filter.
3257 * - (0) if successful.
3258 * - (-ENOTSUP) if hardware doesn't support ethertype filter.
3259 * - (-ENODEV) if *port_id* invalid.
3260 * - (-EINVAL) if the filter information is not correct.
3262 int rte_eth_dev_remove_ethertype_filter(uint8_t port_id,
3266 * Get an ethertype filter rule on an Ethernet device.
3269 * The port identifier of the Ethernet device.
3271 * The identifier of ethertype filter.
3273 * A pointer to a structure of type *rte_ethertype_filter* to be filled with
3274 * the information of the Ethertype filter.
3276 * A pointer to get the queue index.
3278 * - (0) if successful.
3279 * - (-ENOTSUP) if hardware doesn't support ethertype filter.
3280 * - (-ENODEV) if *port_id* invalid.
3281 * - (-EINVAL) if the filter information is not correct.
3282 * - (-ENOENT) if no enabled filter in this index.
3284 int rte_eth_dev_get_ethertype_filter(uint8_t port_id, uint16_t index,
3285 struct rte_ethertype_filter *filter, uint16_t *rx_queue);
3288 * Add a new 2tuple filter rule on an Ethernet device.
3291 * The port identifier of the Ethernet device.
3293 * The identifier of 2tuple filter.
3295 * The pointer to the structure describing the 2tuple filter rule.
3296 * The *rte_2tuple_filter* structure includes the values of the different
3297 * fields to match: protocol, dst_port and
3298 * tcp_flags if the protocol is tcp type.
3300 * The index of the RX queue where to store RX packets matching the added
3303 * - (0) if successful.
3304 * - (-ENOTSUP) if hardware doesn't support 2tuple filter.
3305 * - (-ENODEV) if *port_id* invalid.
3306 * - (-EINVAL) if the filter information is not correct.
3308 int rte_eth_dev_add_2tuple_filter(uint8_t port_id, uint16_t index,
3309 struct rte_2tuple_filter *filter, uint16_t rx_queue);
3312 * remove a 2tuple filter rule on an Ethernet device.
3315 * The port identifier of the Ethernet device.
3317 * The identifier of 2tuple filter.
3319 * - (0) if successful.
3320 * - (-ENOTSUP) if hardware doesn't support 2tuple filter.
3321 * - (-ENODEV) if *port_id* invalid.
3322 * - (-EINVAL) if the filter information is not correct.
3324 int rte_eth_dev_remove_2tuple_filter(uint8_t port_id, uint16_t index);
3327 * Get an 2tuple filter rule on an Ethernet device.
3330 * The port identifier of the Ethernet device.
3332 * The identifier of 2tuple filter.
3334 * A pointer to a structure of type *rte_2tuple_filter* to be filled with
3335 * the information of the 2tuple filter.
3337 * A pointer to get the queue index.
3339 * - (0) if successful.
3340 * - (-ENOTSUP) if hardware doesn't support 2tuple filter.
3341 * - (-ENODEV) if *port_id* invalid.
3342 * - (-EINVAL) if the filter information is not correct.
3343 * - (-ENOENT) if no enabled filter in this index.
3345 int rte_eth_dev_get_2tuple_filter(uint8_t port_id, uint16_t index,
3346 struct rte_2tuple_filter *filter, uint16_t *rx_queue);
3349 * Add a new 5tuple filter rule on an Ethernet device.
3352 * The port identifier of the Ethernet device.
3354 * The identifier of 5tuple filter.
3356 * The pointer to the structure describing the 5tuple filter rule.
3357 * The *rte_5tuple_filter* structure includes the values of the different
3358 * fields to match: dst src IP, dst src port, protocol and relative masks
3360 * The index of the RX queue where to store RX packets matching the added
3363 * - (0) if successful.
3364 * - (-ENOTSUP) if hardware doesn't support 5tuple filter.
3365 * - (-ENODEV) if *port_id* invalid.
3366 * - (-EINVAL) if the filter information is not correct.
3368 int rte_eth_dev_add_5tuple_filter(uint8_t port_id, uint16_t index,
3369 struct rte_5tuple_filter *filter, uint16_t rx_queue);
3372 * remove a 5tuple filter rule on an Ethernet device.
3375 * The port identifier of the Ethernet device.
3377 * The identifier of 5tuple filter.
3379 * - (0) if successful.
3380 * - (-ENOTSUP) if hardware doesn't support 5tuple filter.
3381 * - (-ENODEV) if *port_id* invalid.
3382 * - (-EINVAL) if the filter information is not correct.
3384 int rte_eth_dev_remove_5tuple_filter(uint8_t port_id, uint16_t index);
3387 * Get an 5tuple filter rule on an Ethernet device.
3390 * The port identifier of the Ethernet device.
3392 * The identifier of 5tuple filter.
3394 * A pointer to a structure of type *rte_5tuple_filter* to be filled with
3395 * the information of the 5tuple filter.
3397 * A pointer to get the queue index.
3399 * - (0) if successful.
3400 * - (-ENOTSUP) if hardware doesn't support 5tuple filter.
3401 * - (-ENODEV) if *port_id* invalid.
3402 * - (-EINVAL) if the filter information is not correct.
3404 int rte_eth_dev_get_5tuple_filter(uint8_t port_id, uint16_t index,
3405 struct rte_5tuple_filter *filter, uint16_t *rx_queue);
3408 * Add a new flex filter rule on an Ethernet device.
3411 * The port identifier of the Ethernet device.
3413 * The identifier of flex filter.
3415 * The pointer to the structure describing the flex filter rule.
3416 * The *rte_flex_filter* structure includes the values of the different fields
3417 * to match: the dwords (first len bytes of packet ) and relative masks.
3419 * The index of the RX queue where to store RX packets matching the added
3422 * - (0) if successful.
3423 * - (-ENOTSUP) if hardware doesn't support flex filter.
3424 * - (-ENODEV) if *port_id* invalid.
3425 * - (-EINVAL) if the filter information is not correct.
3426 * - (-ENOENT) if no enabled filter in this index.
3428 int rte_eth_dev_add_flex_filter(uint8_t port_id, uint16_t index,
3429 struct rte_flex_filter *filter, uint16_t rx_queue);
3432 * remove a flex filter rule on an Ethernet device.
3435 * The port identifier of the Ethernet device.
3437 * The identifier of flex filter.
3439 * - (0) if successful.
3440 * - (-ENOTSUP) if hardware doesn't support flex filter.
3441 * - (-ENODEV) if *port_id* invalid.
3442 * - (-EINVAL) if the filter information is not correct.
3444 int rte_eth_dev_remove_flex_filter(uint8_t port_id, uint16_t index);
3447 * Get an flex filter rule on an Ethernet device.
3450 * The port identifier of the Ethernet device.
3452 * The identifier of flex filter.
3454 * A pointer to a structure of type *rte_flex_filter* to be filled with
3455 * the information of the flex filter.
3457 * A pointer to get the queue index.
3459 * - (0) if successful.
3460 * - (-ENOTSUP) if hardware doesn't support flex filter.
3461 * - (-ENODEV) if *port_id* invalid.
3462 * - (-EINVAL) if the filter information is not correct.
3463 * - (-ENOENT) if no enabled filter in this index.
3465 int rte_eth_dev_get_flex_filter(uint8_t port_id, uint16_t index,
3466 struct rte_flex_filter *filter, uint16_t *rx_queue);
3472 #endif /* _RTE_ETHDEV_H_ */