4 * Copyright(c) 2010-2014 Intel Corporation. All rights reserved.
7 * Redistribution and use in source and binary forms, with or without
8 * modification, are permitted provided that the following conditions
11 * * Redistributions of source code must retain the above copyright
12 * notice, this list of conditions and the following disclaimer.
13 * * Redistributions in binary form must reproduce the above copyright
14 * notice, this list of conditions and the following disclaimer in
15 * the documentation and/or other materials provided with the
17 * * Neither the name of Intel Corporation nor the names of its
18 * contributors may be used to endorse or promote products derived
19 * from this software without specific prior written permission.
21 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
22 * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
23 * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
24 * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
25 * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
26 * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
27 * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
28 * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
29 * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
30 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
31 * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
34 #ifndef _RTE_ETHDEV_H_
35 #define _RTE_ETHDEV_H_
40 * RTE Ethernet Device API
42 * The Ethernet Device API is composed of two parts:
44 * - The application-oriented Ethernet API that includes functions to setup
45 * an Ethernet device (configure it, setup its RX and TX queues and start it),
46 * to get its MAC address, the speed and the status of its physical link,
47 * to receive and to transmit packets, and so on.
49 * - The driver-oriented Ethernet API that exports a function allowing
50 * an Ethernet Poll Mode Driver (PMD) to simultaneously register itself as
51 * an Ethernet device driver and as a PCI driver for a set of matching PCI
52 * [Ethernet] devices classes.
54 * By default, all the functions of the Ethernet Device API exported by a PMD
55 * are lock-free functions which assume to not be invoked in parallel on
56 * different logical cores to work on the same target object. For instance,
57 * the receive function of a PMD cannot be invoked in parallel on two logical
58 * cores to poll the same RX queue [of the same port]. Of course, this function
59 * can be invoked in parallel by different logical cores on different RX queues.
60 * It is the responsibility of the upper level application to enforce this rule.
62 * If needed, parallel accesses by multiple logical cores to shared queues
63 * shall be explicitly protected by dedicated inline lock-aware functions
64 * built on top of their corresponding lock-free functions of the PMD API.
66 * In all functions of the Ethernet API, the Ethernet device is
67 * designated by an integer >= 0 named the device port identifier.
69 * At the Ethernet driver level, Ethernet devices are represented by a generic
70 * data structure of type *rte_eth_dev*.
72 * Ethernet devices are dynamically registered during the PCI probing phase
73 * performed at EAL initialization time.
74 * When an Ethernet device is being probed, an *rte_eth_dev* structure and
75 * a new port identifier are allocated for that device. Then, the eth_dev_init()
76 * function supplied by the Ethernet driver matching the probed PCI
77 * device is invoked to properly initialize the device.
79 * The role of the device init function consists of resetting the hardware,
80 * checking access to Non-volatile Memory (NVM), reading the MAC address
83 * If the device init operation is successful, the correspondence between
84 * the port identifier assigned to the new device and its associated
85 * *rte_eth_dev* structure is effectively registered.
86 * Otherwise, both the *rte_eth_dev* structure and the port identifier are
89 * The functions exported by the application Ethernet API to setup a device
90 * designated by its port identifier must be invoked in the following order:
91 * - rte_eth_dev_configure()
92 * - rte_eth_tx_queue_setup()
93 * - rte_eth_rx_queue_setup()
94 * - rte_eth_dev_start()
96 * Then, the network application can invoke, in any order, the functions
97 * exported by the Ethernet API to get the MAC address of a given device, to
98 * get the speed and the status of a device physical link, to receive/transmit
99 * [burst of] packets, and so on.
101 * If the application wants to change the configuration (i.e. call
102 * rte_eth_dev_configure(), rte_eth_tx_queue_setup(), or
103 * rte_eth_rx_queue_setup()), it must call rte_eth_dev_stop() first to stop the
104 * device and then do the reconfiguration before calling rte_eth_dev_start()
105 * again. The tramsit and receive functions should not be invoked when the
108 * Please note that some configuration is not stored between calls to
109 * rte_eth_dev_stop()/rte_eth_dev_start(). The following configuration will
112 * - flow control settings
113 * - receive mode configuration (promiscuous mode, hardware checksum mode,
114 * RSS/VMDQ settings etc.)
115 * - VLAN filtering configuration
116 * - MAC addresses supplied to MAC address array
117 * - flow director filtering mode (but not filtering rules)
118 * - NIC queue statistics mappings
120 * Any other configuration will not be stored and will need to be re-entered
121 * after a call to rte_eth_dev_start().
123 * Finally, a network application can close an Ethernet device by invoking the
124 * rte_eth_dev_close() function.
126 * Each function of the application Ethernet API invokes a specific function
127 * of the PMD that controls the target device designated by its port
129 * For this purpose, all device-specific functions of an Ethernet driver are
130 * supplied through a set of pointers contained in a generic structure of type
132 * The address of the *eth_dev_ops* structure is stored in the *rte_eth_dev*
133 * structure by the device init function of the Ethernet driver, which is
134 * invoked during the PCI probing phase, as explained earlier.
136 * In other words, each function of the Ethernet API simply retrieves the
137 * *rte_eth_dev* structure associated with the device port identifier and
138 * performs an indirect invocation of the corresponding driver function
139 * supplied in the *eth_dev_ops* structure of the *rte_eth_dev* structure.
141 * For performance reasons, the address of the burst-oriented RX and TX
142 * functions of the Ethernet driver are not contained in the *eth_dev_ops*
143 * structure. Instead, they are directly stored at the beginning of the
144 * *rte_eth_dev* structure to avoid an extra indirect memory access during
147 * RTE ethernet device drivers do not use interrupts for transmitting or
148 * receiving. Instead, Ethernet drivers export Poll-Mode receive and transmit
149 * functions to applications.
150 * Both receive and transmit functions are packet-burst oriented to minimize
151 * their cost per packet through the following optimizations:
153 * - Sharing among multiple packets the incompressible cost of the
154 * invocation of receive/transmit functions.
156 * - Enabling receive/transmit functions to take advantage of burst-oriented
157 * hardware features (L1 cache, prefetch instructions, NIC head/tail
158 * registers) to minimize the number of CPU cycles per packet, for instance,
159 * by avoiding useless read memory accesses to ring descriptors, or by
160 * systematically using arrays of pointers that exactly fit L1 cache line
161 * boundaries and sizes.
163 * The burst-oriented receive function does not provide any error notification,
164 * to avoid the corresponding overhead. As a hint, the upper-level application
165 * might check the status of the device link once being systematically returned
166 * a 0 value by the receive function of the driver for a given number of tries.
176 #include <rte_interrupts.h>
178 #include <rte_mbuf.h>
179 #include "rte_ether.h"
180 #include "rte_eth_ctrl.h"
183 * A structure used to retrieve statistics for an Ethernet port.
185 struct rte_eth_stats {
186 uint64_t ipackets; /**< Total number of successfully received packets. */
187 uint64_t opackets; /**< Total number of successfully transmitted packets.*/
188 uint64_t ibytes; /**< Total number of successfully received bytes. */
189 uint64_t obytes; /**< Total number of successfully transmitted bytes. */
190 uint64_t imissed; /**< Total of RX missed packets (e.g full FIFO). */
191 uint64_t ibadcrc; /**< Total of RX packets with CRC error. */
192 uint64_t ibadlen; /**< Total of RX packets with bad length. */
193 uint64_t ierrors; /**< Total number of erroneous received packets. */
194 uint64_t oerrors; /**< Total number of failed transmitted packets. */
195 uint64_t imcasts; /**< Total number of multicast received packets. */
196 uint64_t rx_nombuf; /**< Total number of RX mbuf allocation failures. */
197 uint64_t fdirmatch; /**< Total number of RX packets matching a filter. */
198 uint64_t fdirmiss; /**< Total number of RX packets not matching any filter. */
199 uint64_t tx_pause_xon; /**< Total nb. of XON pause frame sent. */
200 uint64_t rx_pause_xon; /**< Total nb. of XON pause frame received. */
201 uint64_t tx_pause_xoff; /**< Total nb. of XOFF pause frame sent. */
202 uint64_t rx_pause_xoff; /**< Total nb. of XOFF pause frame received. */
203 uint64_t q_ipackets[RTE_ETHDEV_QUEUE_STAT_CNTRS];
204 /**< Total number of queue RX packets. */
205 uint64_t q_opackets[RTE_ETHDEV_QUEUE_STAT_CNTRS];
206 /**< Total number of queue TX packets. */
207 uint64_t q_ibytes[RTE_ETHDEV_QUEUE_STAT_CNTRS];
208 /**< Total number of successfully received queue bytes. */
209 uint64_t q_obytes[RTE_ETHDEV_QUEUE_STAT_CNTRS];
210 /**< Total number of successfully transmitted queue bytes. */
211 uint64_t q_errors[RTE_ETHDEV_QUEUE_STAT_CNTRS];
212 /**< Total number of queue packets received that are dropped. */
214 /**< Total number of good packets received from loopback,VF Only */
216 /**< Total number of good packets transmitted to loopback,VF Only */
218 /**< Total number of good bytes received from loopback,VF Only */
220 /**< Total number of good bytes transmitted to loopback,VF Only */
224 * A structure used to retrieve link-level information of an Ethernet port.
226 struct rte_eth_link {
227 uint16_t link_speed; /**< ETH_LINK_SPEED_[10, 100, 1000, 10000] */
228 uint16_t link_duplex; /**< ETH_LINK_[HALF_DUPLEX, FULL_DUPLEX] */
229 uint8_t link_status : 1; /**< 1 -> link up, 0 -> link down */
230 }__attribute__((aligned(8))); /**< aligned for atomic64 read/write */
232 #define ETH_LINK_SPEED_AUTONEG 0 /**< Auto-negotiate link speed. */
233 #define ETH_LINK_SPEED_10 10 /**< 10 megabits/second. */
234 #define ETH_LINK_SPEED_100 100 /**< 100 megabits/second. */
235 #define ETH_LINK_SPEED_1000 1000 /**< 1 gigabits/second. */
236 #define ETH_LINK_SPEED_10000 10000 /**< 10 gigabits/second. */
237 #define ETH_LINK_SPEED_10G 10000 /**< alias of 10 gigabits/second. */
238 #define ETH_LINK_SPEED_20G 20000 /**< 20 gigabits/second. */
239 #define ETH_LINK_SPEED_40G 40000 /**< 40 gigabits/second. */
241 #define ETH_LINK_AUTONEG_DUPLEX 0 /**< Auto-negotiate duplex. */
242 #define ETH_LINK_HALF_DUPLEX 1 /**< Half-duplex connection. */
243 #define ETH_LINK_FULL_DUPLEX 2 /**< Full-duplex connection. */
246 * A structure used to configure the ring threshold registers of an RX/TX
247 * queue for an Ethernet port.
249 struct rte_eth_thresh {
250 uint8_t pthresh; /**< Ring prefetch threshold. */
251 uint8_t hthresh; /**< Ring host threshold. */
252 uint8_t wthresh; /**< Ring writeback threshold. */
256 * A set of values to identify what method is to be used to route
257 * packets to multiple queues.
259 enum rte_eth_rx_mq_mode {
260 ETH_MQ_RX_NONE = 0, /**< None of DCB,RSS or VMDQ mode */
262 ETH_MQ_RX_RSS, /**< For RX side, only RSS is on */
263 ETH_MQ_RX_DCB, /**< For RX side,only DCB is on. */
264 ETH_MQ_RX_DCB_RSS, /**< Both DCB and RSS enable */
266 ETH_MQ_RX_VMDQ_ONLY, /**< Only VMDQ, no RSS nor DCB */
267 ETH_MQ_RX_VMDQ_RSS, /**< RSS mode with VMDQ */
268 ETH_MQ_RX_VMDQ_DCB, /**< Use VMDQ+DCB to route traffic to queues */
269 ETH_MQ_RX_VMDQ_DCB_RSS, /**< Enable both VMDQ and DCB in VMDq */
273 * for rx mq mode backward compatible
275 #define ETH_RSS ETH_MQ_RX_RSS
276 #define VMDQ_DCB ETH_MQ_RX_VMDQ_DCB
277 #define ETH_DCB_RX ETH_MQ_RX_DCB
280 * A set of values to identify what method is to be used to transmit
281 * packets using multi-TCs.
283 enum rte_eth_tx_mq_mode {
284 ETH_MQ_TX_NONE = 0, /**< It is in neither DCB nor VT mode. */
285 ETH_MQ_TX_DCB, /**< For TX side,only DCB is on. */
286 ETH_MQ_TX_VMDQ_DCB, /**< For TX side,both DCB and VT is on. */
287 ETH_MQ_TX_VMDQ_ONLY, /**< Only VT on, no DCB */
291 * for tx mq mode backward compatible
293 #define ETH_DCB_NONE ETH_MQ_TX_NONE
294 #define ETH_VMDQ_DCB_TX ETH_MQ_TX_VMDQ_DCB
295 #define ETH_DCB_TX ETH_MQ_TX_DCB
298 * A structure used to configure the RX features of an Ethernet port.
300 struct rte_eth_rxmode {
301 /** The multi-queue packet distribution mode to be used, e.g. RSS. */
302 enum rte_eth_rx_mq_mode mq_mode;
303 uint32_t max_rx_pkt_len; /**< Only used if jumbo_frame enabled. */
304 uint16_t split_hdr_size; /**< hdr buf size (header_split enabled).*/
305 uint8_t header_split : 1, /**< Header Split enable. */
306 hw_ip_checksum : 1, /**< IP/UDP/TCP checksum offload enable. */
307 hw_vlan_filter : 1, /**< VLAN filter enable. */
308 hw_vlan_strip : 1, /**< VLAN strip enable. */
309 hw_vlan_extend : 1, /**< Extended VLAN enable. */
310 jumbo_frame : 1, /**< Jumbo Frame Receipt enable. */
311 hw_strip_crc : 1, /**< Enable CRC stripping by hardware. */
312 enable_scatter : 1; /**< Enable scatter packets rx handler */
316 * A structure used to configure the Receive Side Scaling (RSS) feature
317 * of an Ethernet port.
318 * If not NULL, the *rss_key* pointer of the *rss_conf* structure points
319 * to an array holding the RSS key to use for hashing specific header
320 * fields of received packets. The length of this array should be indicated
321 * by *rss_key_len* below. Otherwise, a default random hash key is used by
324 * The *rss_key_len* field of the *rss_conf* structure indicates the length
325 * in bytes of the array pointed by *rss_key*. To be compatible, this length
326 * will be checked in i40e only. Others assume 40 bytes to be used as before.
328 * The *rss_hf* field of the *rss_conf* structure indicates the different
329 * types of IPv4/IPv6 packets to which the RSS hashing must be applied.
330 * Supplying an *rss_hf* equal to zero disables the RSS feature.
332 struct rte_eth_rss_conf {
333 uint8_t *rss_key; /**< If not NULL, 40-byte hash key. */
334 uint8_t rss_key_len; /**< hash key length in bytes. */
335 uint64_t rss_hf; /**< Hash functions to apply - see below. */
338 /* Supported RSS offloads */
340 #define ETH_RSS_IPV4_SHIFT 0
341 #define ETH_RSS_IPV4_TCP_SHIFT 1
342 #define ETH_RSS_IPV6_SHIFT 2
343 #define ETH_RSS_IPV6_EX_SHIFT 3
344 #define ETH_RSS_IPV6_TCP_SHIFT 4
345 #define ETH_RSS_IPV6_TCP_EX_SHIFT 5
346 #define ETH_RSS_IPV4_UDP_SHIFT 6
347 #define ETH_RSS_IPV6_UDP_SHIFT 7
348 #define ETH_RSS_IPV6_UDP_EX_SHIFT 8
350 #define ETH_RSS_NONF_IPV4_UDP_SHIFT 31
351 #define ETH_RSS_NONF_IPV4_TCP_SHIFT 33
352 #define ETH_RSS_NONF_IPV4_SCTP_SHIFT 34
353 #define ETH_RSS_NONF_IPV4_OTHER_SHIFT 35
354 #define ETH_RSS_FRAG_IPV4_SHIFT 36
355 #define ETH_RSS_NONF_IPV6_UDP_SHIFT 41
356 #define ETH_RSS_NONF_IPV6_TCP_SHIFT 43
357 #define ETH_RSS_NONF_IPV6_SCTP_SHIFT 44
358 #define ETH_RSS_NONF_IPV6_OTHER_SHIFT 45
359 #define ETH_RSS_FRAG_IPV6_SHIFT 46
360 #define ETH_RSS_FCOE_OX_SHIFT 48
361 #define ETH_RSS_FCOE_RX_SHIFT 49
362 #define ETH_RSS_FCOE_OTHER_SHIFT 50
363 #define ETH_RSS_L2_PAYLOAD_SHIFT 63
366 #define ETH_RSS_IPV4 (1 << ETH_RSS_IPV4_SHIFT)
367 #define ETH_RSS_IPV4_TCP (1 << ETH_RSS_IPV4_TCP_SHIFT)
368 #define ETH_RSS_IPV6 (1 << ETH_RSS_IPV6_SHIFT)
369 #define ETH_RSS_IPV6_EX (1 << ETH_RSS_IPV6_EX_SHIFT)
370 #define ETH_RSS_IPV6_TCP (1 << ETH_RSS_IPV6_TCP_SHIFT)
371 #define ETH_RSS_IPV6_TCP_EX (1 << ETH_RSS_IPV6_TCP_EX_SHIFT)
372 #define ETH_RSS_IPV4_UDP (1 << ETH_RSS_IPV4_UDP_SHIFT)
373 #define ETH_RSS_IPV6_UDP (1 << ETH_RSS_IPV6_UDP_SHIFT)
374 #define ETH_RSS_IPV6_UDP_EX (1 << ETH_RSS_IPV6_UDP_EX_SHIFT)
376 #define ETH_RSS_NONF_IPV4_UDP (1ULL << ETH_RSS_NONF_IPV4_UDP_SHIFT)
377 #define ETH_RSS_NONF_IPV4_TCP (1ULL << ETH_RSS_NONF_IPV4_TCP_SHIFT)
378 #define ETH_RSS_NONF_IPV4_SCTP (1ULL << ETH_RSS_NONF_IPV4_SCTP_SHIFT)
379 #define ETH_RSS_NONF_IPV4_OTHER (1ULL << ETH_RSS_NONF_IPV4_OTHER_SHIFT)
380 #define ETH_RSS_FRAG_IPV4 (1ULL << ETH_RSS_FRAG_IPV4_SHIFT)
381 #define ETH_RSS_NONF_IPV6_UDP (1ULL << ETH_RSS_NONF_IPV6_UDP_SHIFT)
382 #define ETH_RSS_NONF_IPV6_TCP (1ULL << ETH_RSS_NONF_IPV6_TCP_SHIFT)
383 #define ETH_RSS_NONF_IPV6_SCTP (1ULL << ETH_RSS_NONF_IPV6_SCTP_SHIFT)
384 #define ETH_RSS_NONF_IPV6_OTHER (1ULL << ETH_RSS_NONF_IPV6_OTHER_SHIFT)
385 #define ETH_RSS_FRAG_IPV6 (1ULL << ETH_RSS_FRAG_IPV6_SHIFT)
386 /* FCOE relevant should not be used */
387 #define ETH_RSS_FCOE_OX (1ULL << ETH_RSS_FCOE_OX_SHIFT)
388 #define ETH_RSS_FCOE_RX (1ULL << ETH_RSS_FCOE_RX_SHIFT)
389 #define ETH_RSS_FCOE_OTHER (1ULL << ETH_RSS_FCOE_OTHER_SHIFT)
390 #define ETH_RSS_L2_PAYLOAD (1ULL << ETH_RSS_L2_PAYLOAD_SHIFT)
392 #define ETH_RSS_IP ( \
395 ETH_RSS_NONF_IPV4_OTHER | \
396 ETH_RSS_FRAG_IPV4 | \
397 ETH_RSS_NONF_IPV6_OTHER | \
399 #define ETH_RSS_UDP ( \
404 ETH_RSS_IPV6_UDP_EX | \
405 ETH_RSS_NONF_IPV4_UDP | \
406 ETH_RSS_NONF_IPV6_UDP)
407 /**< Mask of valid RSS hash protocols */
408 #define ETH_RSS_PROTO_MASK ( \
414 ETH_RSS_IPV6_TCP_EX | \
417 ETH_RSS_IPV6_UDP_EX | \
418 ETH_RSS_NONF_IPV4_UDP | \
419 ETH_RSS_NONF_IPV4_TCP | \
420 ETH_RSS_NONF_IPV4_SCTP | \
421 ETH_RSS_NONF_IPV4_OTHER | \
422 ETH_RSS_FRAG_IPV4 | \
423 ETH_RSS_NONF_IPV6_UDP | \
424 ETH_RSS_NONF_IPV6_TCP | \
425 ETH_RSS_NONF_IPV6_SCTP | \
426 ETH_RSS_NONF_IPV6_OTHER | \
427 ETH_RSS_FRAG_IPV6 | \
430 /* Definitions used for redirection table entry size */
431 #define ETH_RSS_RETA_NUM_ENTRIES 128
432 #define ETH_RSS_RETA_MAX_QUEUE 16
434 /* Definitions used for VMDQ and DCB functionality */
435 #define ETH_VMDQ_MAX_VLAN_FILTERS 64 /**< Maximum nb. of VMDQ vlan filters. */
436 #define ETH_DCB_NUM_USER_PRIORITIES 8 /**< Maximum nb. of DCB priorities. */
437 #define ETH_VMDQ_DCB_NUM_QUEUES 128 /**< Maximum nb. of VMDQ DCB queues. */
438 #define ETH_DCB_NUM_QUEUES 128 /**< Maximum nb. of DCB queues. */
440 /* DCB capability defines */
441 #define ETH_DCB_PG_SUPPORT 0x00000001 /**< Priority Group(ETS) support. */
442 #define ETH_DCB_PFC_SUPPORT 0x00000002 /**< Priority Flow Control support. */
444 /* Definitions used for VLAN Offload functionality */
445 #define ETH_VLAN_STRIP_OFFLOAD 0x0001 /**< VLAN Strip On/Off */
446 #define ETH_VLAN_FILTER_OFFLOAD 0x0002 /**< VLAN Filter On/Off */
447 #define ETH_VLAN_EXTEND_OFFLOAD 0x0004 /**< VLAN Extend On/Off */
449 /* Definitions used for mask VLAN setting */
450 #define ETH_VLAN_STRIP_MASK 0x0001 /**< VLAN Strip setting mask */
451 #define ETH_VLAN_FILTER_MASK 0x0002 /**< VLAN Filter setting mask*/
452 #define ETH_VLAN_EXTEND_MASK 0x0004 /**< VLAN Extend setting mask*/
453 #define ETH_VLAN_ID_MAX 0x0FFF /**< VLAN ID is in lower 12 bits*/
455 /* Definitions used for receive MAC address */
456 #define ETH_NUM_RECEIVE_MAC_ADDR 128 /**< Maximum nb. of receive mac addr. */
458 /* Definitions used for unicast hash */
459 #define ETH_VMDQ_NUM_UC_HASH_ARRAY 128 /**< Maximum nb. of UC hash array. */
461 /* Definitions used for VMDQ pool rx mode setting */
462 #define ETH_VMDQ_ACCEPT_UNTAG 0x0001 /**< accept untagged packets. */
463 #define ETH_VMDQ_ACCEPT_HASH_MC 0x0002 /**< accept packets in multicast table . */
464 #define ETH_VMDQ_ACCEPT_HASH_UC 0x0004 /**< accept packets in unicast table. */
465 #define ETH_VMDQ_ACCEPT_BROADCAST 0x0008 /**< accept broadcast packets. */
466 #define ETH_VMDQ_ACCEPT_MULTICAST 0x0010 /**< multicast promiscuous. */
468 /* Definitions used for VMDQ mirror rules setting */
469 #define ETH_VMDQ_NUM_MIRROR_RULE 4 /**< Maximum nb. of mirror rules. . */
471 #define ETH_VMDQ_POOL_MIRROR 0x0001 /**< Virtual Pool Mirroring. */
472 #define ETH_VMDQ_UPLINK_MIRROR 0x0002 /**< Uplink Port Mirroring. */
473 #define ETH_VMDQ_DOWNLIN_MIRROR 0x0004 /**< Downlink Port Mirroring. */
474 #define ETH_VMDQ_VLAN_MIRROR 0x0008 /**< VLAN Mirroring. */
477 * A structure used to configure VLAN traffic mirror of an Ethernet port.
479 struct rte_eth_vlan_mirror {
480 uint64_t vlan_mask; /**< mask for valid VLAN ID. */
481 uint16_t vlan_id[ETH_VMDQ_MAX_VLAN_FILTERS];
482 /** VLAN ID list for vlan mirror. */
486 * A structure used to configure traffic mirror of an Ethernet port.
488 struct rte_eth_vmdq_mirror_conf {
489 uint8_t rule_type_mask; /**< Mirroring rule type mask we want to set */
490 uint8_t dst_pool; /**< Destination pool for this mirror rule. */
491 uint64_t pool_mask; /**< Bitmap of pool for pool mirroring */
492 struct rte_eth_vlan_mirror vlan; /**< VLAN ID setting for VLAN mirroring */
496 * A structure used to configure Redirection Table of the Receive Side
497 * Scaling (RSS) feature of an Ethernet port.
499 struct rte_eth_rss_reta {
500 /** First 64 mask bits indicate which entry(s) need to updated/queried. */
502 /** Second 64 mask bits indicate which entry(s) need to updated/queried. */
504 uint8_t reta[ETH_RSS_RETA_NUM_ENTRIES]; /**< 128 RETA entries*/
508 * This enum indicates the possible number of traffic classes
509 * in DCB configratioins
511 enum rte_eth_nb_tcs {
512 ETH_4_TCS = 4, /**< 4 TCs with DCB. */
513 ETH_8_TCS = 8 /**< 8 TCs with DCB. */
517 * This enum indicates the possible number of queue pools
518 * in VMDQ configurations.
520 enum rte_eth_nb_pools {
521 ETH_8_POOLS = 8, /**< 8 VMDq pools. */
522 ETH_16_POOLS = 16, /**< 16 VMDq pools. */
523 ETH_32_POOLS = 32, /**< 32 VMDq pools. */
524 ETH_64_POOLS = 64 /**< 64 VMDq pools. */
527 /* This structure may be extended in future. */
528 struct rte_eth_dcb_rx_conf {
529 enum rte_eth_nb_tcs nb_tcs; /**< Possible DCB TCs, 4 or 8 TCs */
530 uint8_t dcb_queue[ETH_DCB_NUM_USER_PRIORITIES];
531 /**< Possible DCB queue,4 or 8. */
534 struct rte_eth_vmdq_dcb_tx_conf {
535 enum rte_eth_nb_pools nb_queue_pools; /**< With DCB, 16 or 32 pools. */
536 uint8_t dcb_queue[ETH_DCB_NUM_USER_PRIORITIES];
537 /**< Possible DCB queue,4 or 8. */
540 struct rte_eth_dcb_tx_conf {
541 enum rte_eth_nb_tcs nb_tcs; /**< Possible DCB TCs, 4 or 8 TCs. */
542 uint8_t dcb_queue[ETH_DCB_NUM_USER_PRIORITIES];
543 /**< Possible DCB queue,4 or 8. */
546 struct rte_eth_vmdq_tx_conf {
547 enum rte_eth_nb_pools nb_queue_pools; /**< VMDq mode, 64 pools. */
551 * A structure used to configure the VMDQ+DCB feature
552 * of an Ethernet port.
554 * Using this feature, packets are routed to a pool of queues, based
555 * on the vlan id in the vlan tag, and then to a specific queue within
556 * that pool, using the user priority vlan tag field.
558 * A default pool may be used, if desired, to route all traffic which
559 * does not match the vlan filter rules.
561 struct rte_eth_vmdq_dcb_conf {
562 enum rte_eth_nb_pools nb_queue_pools; /**< With DCB, 16 or 32 pools */
563 uint8_t enable_default_pool; /**< If non-zero, use a default pool */
564 uint8_t default_pool; /**< The default pool, if applicable */
565 uint8_t nb_pool_maps; /**< We can have up to 64 filters/mappings */
567 uint16_t vlan_id; /**< The vlan id of the received frame */
568 uint64_t pools; /**< Bitmask of pools for packet rx */
569 } pool_map[ETH_VMDQ_MAX_VLAN_FILTERS]; /**< VMDq vlan pool maps. */
570 uint8_t dcb_queue[ETH_DCB_NUM_USER_PRIORITIES];
571 /**< Selects a queue in a pool */
574 struct rte_eth_vmdq_rx_conf {
575 enum rte_eth_nb_pools nb_queue_pools; /**< VMDq only mode, 8 or 64 pools */
576 uint8_t enable_default_pool; /**< If non-zero, use a default pool */
577 uint8_t default_pool; /**< The default pool, if applicable */
578 uint8_t enable_loop_back; /**< Enable VT loop back */
579 uint8_t nb_pool_maps; /**< We can have up to 64 filters/mappings */
581 uint16_t vlan_id; /**< The vlan id of the received frame */
582 uint64_t pools; /**< Bitmask of pools for packet rx */
583 } pool_map[ETH_VMDQ_MAX_VLAN_FILTERS]; /**< VMDq vlan pool maps. */
587 * A structure used to configure the TX features of an Ethernet port.
589 struct rte_eth_txmode {
590 enum rte_eth_tx_mq_mode mq_mode; /**< TX multi-queues mode. */
592 /* For i40e specifically */
594 uint8_t hw_vlan_reject_tagged : 1,
595 /**< If set, reject sending out tagged pkts */
596 hw_vlan_reject_untagged : 1,
597 /**< If set, reject sending out untagged pkts */
598 hw_vlan_insert_pvid : 1;
599 /**< If set, enable port based VLAN insertion */
603 * A structure used to configure an RX ring of an Ethernet port.
605 struct rte_eth_rxconf {
606 struct rte_eth_thresh rx_thresh; /**< RX ring threshold registers. */
607 uint16_t rx_free_thresh; /**< Drives the freeing of RX descriptors. */
608 uint8_t rx_drop_en; /**< Drop packets if no descriptors are available. */
609 uint8_t rx_deferred_start; /**< Do not start queue with rte_eth_dev_start(). */
612 #define ETH_TXQ_FLAGS_NOMULTSEGS 0x0001 /**< nb_segs=1 for all mbufs */
613 #define ETH_TXQ_FLAGS_NOREFCOUNT 0x0002 /**< refcnt can be ignored */
614 #define ETH_TXQ_FLAGS_NOMULTMEMP 0x0004 /**< all bufs come from same mempool */
615 #define ETH_TXQ_FLAGS_NOVLANOFFL 0x0100 /**< disable VLAN offload */
616 #define ETH_TXQ_FLAGS_NOXSUMSCTP 0x0200 /**< disable SCTP checksum offload */
617 #define ETH_TXQ_FLAGS_NOXSUMUDP 0x0400 /**< disable UDP checksum offload */
618 #define ETH_TXQ_FLAGS_NOXSUMTCP 0x0800 /**< disable TCP checksum offload */
619 #define ETH_TXQ_FLAGS_NOOFFLOADS \
620 (ETH_TXQ_FLAGS_NOVLANOFFL | ETH_TXQ_FLAGS_NOXSUMSCTP | \
621 ETH_TXQ_FLAGS_NOXSUMUDP | ETH_TXQ_FLAGS_NOXSUMTCP)
623 * A structure used to configure a TX ring of an Ethernet port.
625 struct rte_eth_txconf {
626 struct rte_eth_thresh tx_thresh; /**< TX ring threshold registers. */
627 uint16_t tx_rs_thresh; /**< Drives the setting of RS bit on TXDs. */
628 uint16_t tx_free_thresh; /**< Drives the freeing of TX buffers. */
629 uint32_t txq_flags; /**< Set flags for the Tx queue */
630 uint8_t tx_deferred_start; /**< Do not start queue with rte_eth_dev_start(). */
634 * This enum indicates the flow control mode
636 enum rte_eth_fc_mode {
637 RTE_FC_NONE = 0, /**< Disable flow control. */
638 RTE_FC_RX_PAUSE, /**< RX pause frame, enable flowctrl on TX side. */
639 RTE_FC_TX_PAUSE, /**< TX pause frame, enable flowctrl on RX side. */
640 RTE_FC_FULL /**< Enable flow control on both side. */
644 * A structure used to configure Ethernet flow control parameter.
645 * These parameters will be configured into the register of the NIC.
646 * Please refer to the corresponding data sheet for proper value.
648 struct rte_eth_fc_conf {
649 uint32_t high_water; /**< High threshold value to trigger XOFF */
650 uint32_t low_water; /**< Low threshold value to trigger XON */
651 uint16_t pause_time; /**< Pause quota in the Pause frame */
652 uint16_t send_xon; /**< Is XON frame need be sent */
653 enum rte_eth_fc_mode mode; /**< Link flow control mode */
654 uint8_t mac_ctrl_frame_fwd; /**< Forward MAC control frames */
655 uint8_t autoneg; /**< Use Pause autoneg */
659 * A structure used to configure Ethernet priority flow control parameter.
660 * These parameters will be configured into the register of the NIC.
661 * Please refer to the corresponding data sheet for proper value.
663 struct rte_eth_pfc_conf {
664 struct rte_eth_fc_conf fc; /**< General flow control parameter. */
665 uint8_t priority; /**< VLAN User Priority. */
669 * Flow Director setting modes: none (default), signature or perfect.
672 RTE_FDIR_MODE_NONE = 0, /**< Disable FDIR support. */
673 RTE_FDIR_MODE_SIGNATURE, /**< Enable FDIR signature filter mode. */
674 RTE_FDIR_MODE_PERFECT, /**< Enable FDIR perfect filter mode. */
678 * Memory space that can be configured to store Flow Director filters
679 * in the board memory.
681 enum rte_fdir_pballoc_type {
682 RTE_FDIR_PBALLOC_64K = 0, /**< 64k. */
683 RTE_FDIR_PBALLOC_128K, /**< 128k. */
684 RTE_FDIR_PBALLOC_256K, /**< 256k. */
688 * Select report mode of FDIR hash information in RX descriptors.
690 enum rte_fdir_status_mode {
691 RTE_FDIR_NO_REPORT_STATUS = 0, /**< Never report FDIR hash. */
692 RTE_FDIR_REPORT_STATUS, /**< Only report FDIR hash for matching pkts. */
693 RTE_FDIR_REPORT_STATUS_ALWAYS, /**< Always report FDIR hash. */
697 * A structure used to configure the Flow Director (FDIR) feature
698 * of an Ethernet port.
700 * If mode is RTE_FDIR_DISABLE, the pballoc value is ignored.
702 struct rte_fdir_conf {
703 enum rte_fdir_mode mode; /**< Flow Director mode. */
704 enum rte_fdir_pballoc_type pballoc; /**< Space for FDIR filters. */
705 enum rte_fdir_status_mode status; /**< How to report FDIR hash. */
706 /** Offset of flexbytes field in RX packets (in 16-bit word units). */
707 uint8_t flexbytes_offset;
708 /** RX queue of packets matching a "drop" filter in perfect mode. */
713 * Possible l4type of FDIR filters.
716 RTE_FDIR_L4TYPE_NONE = 0, /**< None. */
717 RTE_FDIR_L4TYPE_UDP, /**< UDP. */
718 RTE_FDIR_L4TYPE_TCP, /**< TCP. */
719 RTE_FDIR_L4TYPE_SCTP, /**< SCTP. */
723 * Select IPv4 or IPv6 FDIR filters.
726 RTE_FDIR_IPTYPE_IPV4 = 0, /**< IPv4. */
727 RTE_FDIR_IPTYPE_IPV6 , /**< IPv6. */
731 * A structure used to define a FDIR packet filter.
733 struct rte_fdir_filter {
734 uint16_t flex_bytes; /**< Flex bytes value to match. */
735 uint16_t vlan_id; /**< VLAN ID value to match, 0 otherwise. */
736 uint16_t port_src; /**< Source port to match, 0 otherwise. */
737 uint16_t port_dst; /**< Destination port to match, 0 otherwise. */
739 uint32_t ipv4_addr; /**< IPv4 source address to match. */
740 uint32_t ipv6_addr[4]; /**< IPv6 source address to match. */
741 } ip_src; /**< IPv4/IPv6 source address to match (union of above). */
743 uint32_t ipv4_addr; /**< IPv4 destination address to match. */
744 uint32_t ipv6_addr[4]; /**< IPv6 destination address to match */
745 } ip_dst; /**< IPv4/IPv6 destination address to match (union of above). */
746 enum rte_l4type l4type; /**< l4type to match: NONE/UDP/TCP/SCTP. */
747 enum rte_iptype iptype; /**< IP packet type to match: IPv4 or IPv6. */
751 * A structure used to configure FDIR masks that are used by the device
752 * to match the various fields of RX packet headers.
753 * @note The only_ip_flow field has the opposite meaning compared to other
756 struct rte_fdir_masks {
757 /** When set to 1, packet l4type is \b NOT relevant in filters, and
758 source and destination port masks must be set to zero. */
759 uint8_t only_ip_flow;
760 /** If set to 1, vlan_id is relevant in filters. */
762 /** If set to 1, vlan_prio is relevant in filters. */
764 /** If set to 1, flexbytes is relevant in filters. */
766 /** If set to 1, set the IPv6 masks. Otherwise set the IPv4 masks. */
767 uint8_t set_ipv6_mask;
768 /** When set to 1, comparison of destination IPv6 address with IP6AT
769 registers is meaningful. */
770 uint8_t comp_ipv6_dst;
771 /** Mask of Destination IPv4 Address. All bits set to 1 define the
772 relevant bits to use in the destination address of an IPv4 packet
773 when matching it against FDIR filters. */
774 uint32_t dst_ipv4_mask;
775 /** Mask of Source IPv4 Address. All bits set to 1 define
776 the relevant bits to use in the source address of an IPv4 packet
777 when matching it against FDIR filters. */
778 uint32_t src_ipv4_mask;
779 /** Mask of Source IPv6 Address. All bits set to 1 define the
780 relevant BYTES to use in the source address of an IPv6 packet
781 when matching it against FDIR filters. */
782 uint16_t dst_ipv6_mask;
783 /** Mask of Destination IPv6 Address. All bits set to 1 define the
784 relevant BYTES to use in the destination address of an IPv6 packet
785 when matching it against FDIR filters. */
786 uint16_t src_ipv6_mask;
787 /** Mask of Source Port. All bits set to 1 define the relevant
788 bits to use in the source port of an IP packets when matching it
789 against FDIR filters. */
790 uint16_t src_port_mask;
791 /** Mask of Destination Port. All bits set to 1 define the relevant
792 bits to use in the destination port of an IP packet when matching it
793 against FDIR filters. */
794 uint16_t dst_port_mask;
798 * A structure used to report the status of the flow director filters in use.
800 struct rte_eth_fdir {
801 /** Number of filters with collision indication. */
803 /** Number of free (non programmed) filters. */
805 /** The Lookup hash value of the added filter that updated the value
806 of the MAXLEN field */
808 /** Longest linked list of filters in the table. */
810 /** Number of added filters. */
812 /** Number of removed filters. */
814 /** Number of failed added filters (no more space in device). */
816 /** Number of failed removed filters. */
821 * A structure used to enable/disable specific device interrupts.
823 struct rte_intr_conf {
824 /** enable/disable lsc interrupt. 0 (default) - disable, 1 enable */
829 * A structure used to configure an Ethernet port.
830 * Depending upon the RX multi-queue mode, extra advanced
831 * configuration settings may be needed.
833 struct rte_eth_conf {
835 /**< ETH_LINK_SPEED_10[0|00|000], or 0 for autonegotation */
836 uint16_t link_duplex;
837 /**< ETH_LINK_[HALF_DUPLEX|FULL_DUPLEX], or 0 for autonegotation */
838 struct rte_eth_rxmode rxmode; /**< Port RX configuration. */
839 struct rte_eth_txmode txmode; /**< Port TX configuration. */
840 uint32_t lpbk_mode; /**< Loopback operation mode. By default the value
841 is 0, meaning the loopback mode is disabled.
842 Read the datasheet of given ethernet controller
843 for details. The possible values of this field
844 are defined in implementation of each driver. */
846 struct rte_eth_rss_conf rss_conf; /**< Port RSS configuration */
847 struct rte_eth_vmdq_dcb_conf vmdq_dcb_conf;
848 /**< Port vmdq+dcb configuration. */
849 struct rte_eth_dcb_rx_conf dcb_rx_conf;
850 /**< Port dcb RX configuration. */
851 struct rte_eth_vmdq_rx_conf vmdq_rx_conf;
852 /**< Port vmdq RX configuration. */
853 } rx_adv_conf; /**< Port RX filtering configuration (union). */
855 struct rte_eth_vmdq_dcb_tx_conf vmdq_dcb_tx_conf;
856 /**< Port vmdq+dcb TX configuration. */
857 struct rte_eth_dcb_tx_conf dcb_tx_conf;
858 /**< Port dcb TX configuration. */
859 struct rte_eth_vmdq_tx_conf vmdq_tx_conf;
860 /**< Port vmdq TX configuration. */
861 } tx_adv_conf; /**< Port TX DCB configuration (union). */
862 /** Currently,Priority Flow Control(PFC) are supported,if DCB with PFC
863 is needed,and the variable must be set ETH_DCB_PFC_SUPPORT. */
864 uint32_t dcb_capability_en;
865 struct rte_fdir_conf fdir_conf; /**< FDIR configuration. */
866 struct rte_intr_conf intr_conf; /**< Interrupt mode configuration. */
870 * A structure used to retrieve the contextual information of
871 * an Ethernet device, such as the controlling driver of the device,
872 * its PCI context, etc...
876 * RX offload capabilities of a device.
878 #define DEV_RX_OFFLOAD_VLAN_STRIP 0x00000001
879 #define DEV_RX_OFFLOAD_IPV4_CKSUM 0x00000002
880 #define DEV_RX_OFFLOAD_UDP_CKSUM 0x00000004
881 #define DEV_RX_OFFLOAD_TCP_CKSUM 0x00000008
882 #define DEV_RX_OFFLOAD_TCP_LRO 0x00000010
885 * TX offload capabilities of a device.
887 #define DEV_TX_OFFLOAD_VLAN_INSERT 0x00000001
888 #define DEV_TX_OFFLOAD_IPV4_CKSUM 0x00000002
889 #define DEV_TX_OFFLOAD_UDP_CKSUM 0x00000004
890 #define DEV_TX_OFFLOAD_TCP_CKSUM 0x00000008
891 #define DEV_TX_OFFLOAD_SCTP_CKSUM 0x00000010
892 #define DEV_TX_OFFLOAD_TCP_TSO 0x00000020
893 #define DEV_TX_OFFLOAD_UDP_TSO 0x00000040
895 struct rte_eth_dev_info {
896 struct rte_pci_device *pci_dev; /**< Device PCI information. */
897 const char *driver_name; /**< Device Driver name. */
898 unsigned int if_index; /**< Index to bound host interface, or 0 if none.
899 Use if_indextoname() to translate into an interface name. */
900 uint32_t min_rx_bufsize; /**< Minimum size of RX buffer. */
901 uint32_t max_rx_pktlen; /**< Maximum configurable length of RX pkt. */
902 uint16_t max_rx_queues; /**< Maximum number of RX queues. */
903 uint16_t max_tx_queues; /**< Maximum number of TX queues. */
904 uint32_t max_mac_addrs; /**< Maximum number of MAC addresses. */
905 uint32_t max_hash_mac_addrs;
906 /** Maximum number of hash MAC addresses for MTA and UTA. */
907 uint16_t max_vfs; /**< Maximum number of VFs. */
908 uint16_t max_vmdq_pools; /**< Maximum number of VMDq pools. */
909 uint32_t rx_offload_capa; /**< Device RX offload capabilities. */
910 uint32_t tx_offload_capa; /**< Device TX offload capabilities. */
911 struct rte_eth_rxconf default_rxconf; /**< Default RX configuration */
912 struct rte_eth_txconf default_txconf; /**< Default TX configuration */
915 /** Maximum name length for extended statistics counters */
916 #define RTE_ETH_XSTATS_NAME_SIZE 64
919 * An Ethernet device extended statistic structure
921 * This structure is used by ethdev->eth_xstats_get() to provide
922 * statistics that are not provided in the generic rte_eth_stats
925 struct rte_eth_xstats {
926 char name[RTE_ETH_XSTATS_NAME_SIZE];
932 struct rte_eth_dev_callback;
933 /** @internal Structure to keep track of registered callbacks */
934 TAILQ_HEAD(rte_eth_dev_cb_list, rte_eth_dev_callback);
936 #define TCP_UGR_FLAG 0x20
937 #define TCP_ACK_FLAG 0x10
938 #define TCP_PSH_FLAG 0x08
939 #define TCP_RST_FLAG 0x04
940 #define TCP_SYN_FLAG 0x02
941 #define TCP_FIN_FLAG 0x01
942 #define TCP_FLAG_ALL 0x3F
945 * A structure used to define an ethertype filter.
947 struct rte_ethertype_filter {
948 uint16_t ethertype; /**< little endian. */
949 uint8_t priority_en; /**< compare priority enable. */
954 * A structure used to define an syn filter.
956 struct rte_syn_filter {
957 uint8_t hig_pri; /**< 1 means higher pri than 2tuple, 5tupe,
958 and flex filter, 0 means lower pri. */
962 * A structure used to define a 2tuple filter.
964 struct rte_2tuple_filter {
965 uint16_t dst_port; /**< big endian. */
968 uint16_t priority; /**< used when more than one filter matches. */
969 uint8_t dst_port_mask:1, /**< if mask is 1b, means not compare. */
974 * A structure used to define a flex filter.
976 struct rte_flex_filter {
978 uint32_t dwords[32]; /**< flex bytes in big endian. */
979 uint8_t mask[16]; /**< if mask bit is 1b, do not compare
980 corresponding byte in dwords. */
985 * A structure used to define a 5tuple filter.
987 struct rte_5tuple_filter {
988 uint32_t dst_ip; /**< destination IP address in big endian. */
989 uint32_t src_ip; /**< source IP address in big endian. */
990 uint16_t dst_port; /**< destination port in big endian. */
991 uint16_t src_port; /**< source Port big endian. */
992 uint8_t protocol; /**< l4 protocol. */
993 uint8_t tcp_flags; /**< tcp flags. */
994 uint16_t priority; /**< seven evels (001b-111b), 111b is highest,
995 used when more than one filter matches. */
996 uint8_t dst_ip_mask:1, /**< if mask is 1b, do not compare dst ip. */
997 src_ip_mask:1, /**< if mask is 1b, do not compare src ip. */
998 dst_port_mask:1, /**< if mask is 1b, do not compare dst port. */
999 src_port_mask:1, /**< if mask is 1b, do not compare src port. */
1000 protocol_mask:1; /**< if mask is 1b, do not compare protocol. */
1004 * Definitions of all functions exported by an Ethernet driver through the
1005 * the generic structure of type *eth_dev_ops* supplied in the *rte_eth_dev*
1006 * structure associated with an Ethernet device.
1009 typedef int (*eth_dev_configure_t)(struct rte_eth_dev *dev);
1010 /**< @internal Ethernet device configuration. */
1012 typedef int (*eth_dev_start_t)(struct rte_eth_dev *dev);
1013 /**< @internal Function used to start a configured Ethernet device. */
1015 typedef void (*eth_dev_stop_t)(struct rte_eth_dev *dev);
1016 /**< @internal Function used to stop a configured Ethernet device. */
1018 typedef int (*eth_dev_set_link_up_t)(struct rte_eth_dev *dev);
1019 /**< @internal Function used to link up a configured Ethernet device. */
1021 typedef int (*eth_dev_set_link_down_t)(struct rte_eth_dev *dev);
1022 /**< @internal Function used to link down a configured Ethernet device. */
1024 typedef void (*eth_dev_close_t)(struct rte_eth_dev *dev);
1025 /**< @internal Function used to close a configured Ethernet device. */
1027 typedef void (*eth_promiscuous_enable_t)(struct rte_eth_dev *dev);
1028 /**< @internal Function used to enable the RX promiscuous mode of an Ethernet device. */
1030 typedef void (*eth_promiscuous_disable_t)(struct rte_eth_dev *dev);
1031 /**< @internal Function used to disable the RX promiscuous mode of an Ethernet device. */
1033 typedef void (*eth_allmulticast_enable_t)(struct rte_eth_dev *dev);
1034 /**< @internal Enable the receipt of all multicast packets by an Ethernet device. */
1036 typedef void (*eth_allmulticast_disable_t)(struct rte_eth_dev *dev);
1037 /**< @internal Disable the receipt of all multicast packets by an Ethernet device. */
1039 typedef int (*eth_link_update_t)(struct rte_eth_dev *dev,
1040 int wait_to_complete);
1041 /**< @internal Get link speed, duplex mode and state (up/down) of an Ethernet device. */
1043 typedef void (*eth_stats_get_t)(struct rte_eth_dev *dev,
1044 struct rte_eth_stats *igb_stats);
1045 /**< @internal Get global I/O statistics of an Ethernet device. */
1047 typedef void (*eth_stats_reset_t)(struct rte_eth_dev *dev);
1048 /**< @internal Reset global I/O statistics of an Ethernet device to 0. */
1050 typedef int (*eth_xstats_get_t)(struct rte_eth_dev *dev,
1051 struct rte_eth_xstats *stats, unsigned n);
1052 /**< @internal Get extended stats of an Ethernet device. */
1054 typedef void (*eth_xstats_reset_t)(struct rte_eth_dev *dev);
1055 /**< @internal Reset extended stats of an Ethernet device. */
1057 typedef int (*eth_queue_stats_mapping_set_t)(struct rte_eth_dev *dev,
1061 /**< @internal Set a queue statistics mapping for a tx/rx queue of an Ethernet device. */
1063 typedef void (*eth_dev_infos_get_t)(struct rte_eth_dev *dev,
1064 struct rte_eth_dev_info *dev_info);
1065 /**< @internal Get specific informations of an Ethernet device. */
1067 typedef int (*eth_queue_start_t)(struct rte_eth_dev *dev,
1069 /**< @internal Start rx and tx of a queue of an Ethernet device. */
1071 typedef int (*eth_queue_stop_t)(struct rte_eth_dev *dev,
1073 /**< @internal Stop rx and tx of a queue of an Ethernet device. */
1075 typedef int (*eth_rx_queue_setup_t)(struct rte_eth_dev *dev,
1076 uint16_t rx_queue_id,
1077 uint16_t nb_rx_desc,
1078 unsigned int socket_id,
1079 const struct rte_eth_rxconf *rx_conf,
1080 struct rte_mempool *mb_pool);
1081 /**< @internal Set up a receive queue of an Ethernet device. */
1083 typedef int (*eth_tx_queue_setup_t)(struct rte_eth_dev *dev,
1084 uint16_t tx_queue_id,
1085 uint16_t nb_tx_desc,
1086 unsigned int socket_id,
1087 const struct rte_eth_txconf *tx_conf);
1088 /**< @internal Setup a transmit queue of an Ethernet device. */
1090 typedef void (*eth_queue_release_t)(void *queue);
1091 /**< @internal Release memory resources allocated by given RX/TX queue. */
1093 typedef uint32_t (*eth_rx_queue_count_t)(struct rte_eth_dev *dev,
1094 uint16_t rx_queue_id);
1095 /**< @Get number of available descriptors on a receive queue of an Ethernet device. */
1097 typedef int (*eth_rx_descriptor_done_t)(void *rxq, uint16_t offset);
1098 /**< @Check DD bit of specific RX descriptor */
1100 typedef int (*mtu_set_t)(struct rte_eth_dev *dev, uint16_t mtu);
1101 /**< @internal Set MTU. */
1103 typedef int (*vlan_filter_set_t)(struct rte_eth_dev *dev,
1106 /**< @internal filtering of a VLAN Tag Identifier by an Ethernet device. */
1108 typedef void (*vlan_tpid_set_t)(struct rte_eth_dev *dev,
1110 /**< @internal set the outer VLAN-TPID by an Ethernet device. */
1112 typedef void (*vlan_offload_set_t)(struct rte_eth_dev *dev, int mask);
1113 /**< @internal set VLAN offload function by an Ethernet device. */
1115 typedef int (*vlan_pvid_set_t)(struct rte_eth_dev *dev,
1118 /**< @internal set port based TX VLAN insertion by an Ethernet device. */
1120 typedef void (*vlan_strip_queue_set_t)(struct rte_eth_dev *dev,
1121 uint16_t rx_queue_id,
1123 /**< @internal VLAN stripping enable/disable by an queue of Ethernet device. */
1125 typedef uint16_t (*eth_rx_burst_t)(void *rxq,
1126 struct rte_mbuf **rx_pkts,
1128 /**< @internal Retrieve input packets from a receive queue of an Ethernet device. */
1130 typedef uint16_t (*eth_tx_burst_t)(void *txq,
1131 struct rte_mbuf **tx_pkts,
1133 /**< @internal Send output packets on a transmit queue of an Ethernet device. */
1135 typedef int (*fdir_add_signature_filter_t)(struct rte_eth_dev *dev,
1136 struct rte_fdir_filter *fdir_ftr,
1138 /**< @internal Setup a new signature filter rule on an Ethernet device */
1140 typedef int (*fdir_update_signature_filter_t)(struct rte_eth_dev *dev,
1141 struct rte_fdir_filter *fdir_ftr,
1143 /**< @internal Update a signature filter rule on an Ethernet device */
1145 typedef int (*fdir_remove_signature_filter_t)(struct rte_eth_dev *dev,
1146 struct rte_fdir_filter *fdir_ftr);
1147 /**< @internal Remove a signature filter rule on an Ethernet device */
1149 typedef void (*fdir_infos_get_t)(struct rte_eth_dev *dev,
1150 struct rte_eth_fdir *fdir);
1151 /**< @internal Get information about fdir status */
1153 typedef int (*fdir_add_perfect_filter_t)(struct rte_eth_dev *dev,
1154 struct rte_fdir_filter *fdir_ftr,
1155 uint16_t soft_id, uint8_t rx_queue,
1157 /**< @internal Setup a new perfect filter rule on an Ethernet device */
1159 typedef int (*fdir_update_perfect_filter_t)(struct rte_eth_dev *dev,
1160 struct rte_fdir_filter *fdir_ftr,
1161 uint16_t soft_id, uint8_t rx_queue,
1163 /**< @internal Update a perfect filter rule on an Ethernet device */
1165 typedef int (*fdir_remove_perfect_filter_t)(struct rte_eth_dev *dev,
1166 struct rte_fdir_filter *fdir_ftr,
1168 /**< @internal Remove a perfect filter rule on an Ethernet device */
1170 typedef int (*fdir_set_masks_t)(struct rte_eth_dev *dev,
1171 struct rte_fdir_masks *fdir_masks);
1172 /**< @internal Setup flow director masks on an Ethernet device */
1174 typedef int (*flow_ctrl_get_t)(struct rte_eth_dev *dev,
1175 struct rte_eth_fc_conf *fc_conf);
1176 /**< @internal Get current flow control parameter on an Ethernet device */
1178 typedef int (*flow_ctrl_set_t)(struct rte_eth_dev *dev,
1179 struct rte_eth_fc_conf *fc_conf);
1180 /**< @internal Setup flow control parameter on an Ethernet device */
1182 typedef int (*priority_flow_ctrl_set_t)(struct rte_eth_dev *dev,
1183 struct rte_eth_pfc_conf *pfc_conf);
1184 /**< @internal Setup priority flow control parameter on an Ethernet device */
1186 typedef int (*reta_update_t)(struct rte_eth_dev *dev,
1187 struct rte_eth_rss_reta *reta_conf);
1188 /**< @internal Update RSS redirection table on an Ethernet device */
1190 typedef int (*reta_query_t)(struct rte_eth_dev *dev,
1191 struct rte_eth_rss_reta *reta_conf);
1192 /**< @internal Query RSS redirection table on an Ethernet device */
1194 typedef int (*rss_hash_update_t)(struct rte_eth_dev *dev,
1195 struct rte_eth_rss_conf *rss_conf);
1196 /**< @internal Update RSS hash configuration of an Ethernet device */
1198 typedef int (*rss_hash_conf_get_t)(struct rte_eth_dev *dev,
1199 struct rte_eth_rss_conf *rss_conf);
1200 /**< @internal Get current RSS hash configuration of an Ethernet device */
1202 typedef int (*eth_dev_led_on_t)(struct rte_eth_dev *dev);
1203 /**< @internal Turn on SW controllable LED on an Ethernet device */
1205 typedef int (*eth_dev_led_off_t)(struct rte_eth_dev *dev);
1206 /**< @internal Turn off SW controllable LED on an Ethernet device */
1208 typedef void (*eth_mac_addr_remove_t)(struct rte_eth_dev *dev, uint32_t index);
1209 /**< @internal Remove MAC address from receive address register */
1211 typedef void (*eth_mac_addr_add_t)(struct rte_eth_dev *dev,
1212 struct ether_addr *mac_addr,
1215 /**< @internal Set a MAC address into Receive Address Address Register */
1217 typedef int (*eth_uc_hash_table_set_t)(struct rte_eth_dev *dev,
1218 struct ether_addr *mac_addr,
1220 /**< @internal Set a Unicast Hash bitmap */
1222 typedef int (*eth_uc_all_hash_table_set_t)(struct rte_eth_dev *dev,
1224 /**< @internal Set all Unicast Hash bitmap */
1226 typedef int (*eth_set_vf_rx_mode_t)(struct rte_eth_dev *dev,
1230 /**< @internal Set a VF receive mode */
1232 typedef int (*eth_set_vf_rx_t)(struct rte_eth_dev *dev,
1235 /**< @internal Set a VF receive mode */
1237 typedef int (*eth_set_vf_tx_t)(struct rte_eth_dev *dev,
1240 /**< @internal Enable or disable a VF transmit */
1242 typedef int (*eth_set_vf_vlan_filter_t)(struct rte_eth_dev *dev,
1246 /**< @internal Set VF VLAN pool filter */
1248 typedef int (*eth_set_queue_rate_limit_t)(struct rte_eth_dev *dev,
1251 /**< @internal Set queue TX rate */
1253 typedef int (*eth_set_vf_rate_limit_t)(struct rte_eth_dev *dev,
1257 /**< @internal Set VF TX rate */
1259 typedef int (*eth_mirror_rule_set_t)(struct rte_eth_dev *dev,
1260 struct rte_eth_vmdq_mirror_conf *mirror_conf,
1263 /**< @internal Add a traffic mirroring rule on an Ethernet device */
1265 typedef int (*eth_mirror_rule_reset_t)(struct rte_eth_dev *dev,
1267 /**< @internal Remove a traffic mirroring rule on an Ethernet device */
1269 #ifdef RTE_NIC_BYPASS
1272 RTE_BYPASS_MODE_NONE,
1273 RTE_BYPASS_MODE_NORMAL,
1274 RTE_BYPASS_MODE_BYPASS,
1275 RTE_BYPASS_MODE_ISOLATE,
1276 RTE_BYPASS_MODE_NUM,
1279 #define RTE_BYPASS_MODE_VALID(x) \
1280 ((x) > RTE_BYPASS_MODE_NONE && (x) < RTE_BYPASS_MODE_NUM)
1283 RTE_BYPASS_EVENT_NONE,
1284 RTE_BYPASS_EVENT_START,
1285 RTE_BYPASS_EVENT_OS_ON = RTE_BYPASS_EVENT_START,
1286 RTE_BYPASS_EVENT_POWER_ON,
1287 RTE_BYPASS_EVENT_OS_OFF,
1288 RTE_BYPASS_EVENT_POWER_OFF,
1289 RTE_BYPASS_EVENT_TIMEOUT,
1290 RTE_BYPASS_EVENT_NUM
1293 #define RTE_BYPASS_EVENT_VALID(x) \
1294 ((x) > RTE_BYPASS_EVENT_NONE && (x) < RTE_BYPASS_MODE_NUM)
1297 RTE_BYPASS_TMT_OFF, /* timeout disabled. */
1298 RTE_BYPASS_TMT_1_5_SEC, /* timeout for 1.5 seconds */
1299 RTE_BYPASS_TMT_2_SEC, /* timeout for 2 seconds */
1300 RTE_BYPASS_TMT_3_SEC, /* timeout for 3 seconds */
1301 RTE_BYPASS_TMT_4_SEC, /* timeout for 4 seconds */
1302 RTE_BYPASS_TMT_8_SEC, /* timeout for 8 seconds */
1303 RTE_BYPASS_TMT_16_SEC, /* timeout for 16 seconds */
1304 RTE_BYPASS_TMT_32_SEC, /* timeout for 32 seconds */
1308 #define RTE_BYPASS_TMT_VALID(x) \
1309 ((x) == RTE_BYPASS_TMT_OFF || \
1310 ((x) > RTE_BYPASS_TMT_OFF && (x) < RTE_BYPASS_TMT_NUM))
1312 typedef void (*bypass_init_t)(struct rte_eth_dev *dev);
1313 typedef int32_t (*bypass_state_set_t)(struct rte_eth_dev *dev, uint32_t *new_state);
1314 typedef int32_t (*bypass_state_show_t)(struct rte_eth_dev *dev, uint32_t *state);
1315 typedef int32_t (*bypass_event_set_t)(struct rte_eth_dev *dev, uint32_t state, uint32_t event);
1316 typedef int32_t (*bypass_event_show_t)(struct rte_eth_dev *dev, uint32_t event_shift, uint32_t *event);
1317 typedef int32_t (*bypass_wd_timeout_set_t)(struct rte_eth_dev *dev, uint32_t timeout);
1318 typedef int32_t (*bypass_wd_timeout_show_t)(struct rte_eth_dev *dev, uint32_t *wd_timeout);
1319 typedef int32_t (*bypass_ver_show_t)(struct rte_eth_dev *dev, uint32_t *ver);
1320 typedef int32_t (*bypass_wd_reset_t)(struct rte_eth_dev *dev);
1323 typedef int (*eth_add_syn_filter_t)(struct rte_eth_dev *dev,
1324 struct rte_syn_filter *filter, uint16_t rx_queue);
1325 /**< @internal add syn filter rule on an Ethernet device */
1327 typedef int (*eth_remove_syn_filter_t)(struct rte_eth_dev *dev);
1328 /**< @internal remove syn filter rule on an Ethernet device */
1330 typedef int (*eth_get_syn_filter_t)(struct rte_eth_dev *dev,
1331 struct rte_syn_filter *filter, uint16_t *rx_queue);
1332 /**< @internal Get syn filter rule on an Ethernet device */
1334 typedef int (*eth_add_ethertype_filter_t)(struct rte_eth_dev *dev,
1335 uint16_t index, struct rte_ethertype_filter *filter,
1337 /**< @internal Setup a new ethertype filter rule on an Ethernet device */
1339 typedef int (*eth_remove_ethertype_filter_t)(struct rte_eth_dev *dev,
1341 /**< @internal Remove an ethertype filter rule on an Ethernet device */
1343 typedef int (*eth_get_ethertype_filter_t)(struct rte_eth_dev *dev,
1344 uint16_t index, struct rte_ethertype_filter *filter,
1345 uint16_t *rx_queue);
1346 /**< @internal Get an ethertype filter rule on an Ethernet device */
1348 typedef int (*eth_add_2tuple_filter_t)(struct rte_eth_dev *dev,
1349 uint16_t index, struct rte_2tuple_filter *filter,
1351 /**< @internal Setup a new 2tuple filter rule on an Ethernet device */
1353 typedef int (*eth_remove_2tuple_filter_t)(struct rte_eth_dev *dev,
1355 /**< @internal Remove a 2tuple filter rule on an Ethernet device */
1357 typedef int (*eth_get_2tuple_filter_t)(struct rte_eth_dev *dev,
1358 uint16_t index, struct rte_2tuple_filter *filter,
1359 uint16_t *rx_queue);
1360 /**< @internal Get a 2tuple filter rule on an Ethernet device */
1362 typedef int (*eth_add_5tuple_filter_t)(struct rte_eth_dev *dev,
1363 uint16_t index, struct rte_5tuple_filter *filter,
1365 /**< @internal Setup a new 5tuple filter rule on an Ethernet device */
1367 typedef int (*eth_remove_5tuple_filter_t)(struct rte_eth_dev *dev,
1369 /**< @internal Remove a 5tuple filter rule on an Ethernet device */
1371 typedef int (*eth_get_5tuple_filter_t)(struct rte_eth_dev *dev,
1372 uint16_t index, struct rte_5tuple_filter *filter,
1373 uint16_t *rx_queue);
1374 /**< @internal Get a 5tuple filter rule on an Ethernet device */
1376 typedef int (*eth_add_flex_filter_t)(struct rte_eth_dev *dev,
1377 uint16_t index, struct rte_flex_filter *filter,
1379 /**< @internal Setup a new flex filter rule on an Ethernet device */
1381 typedef int (*eth_remove_flex_filter_t)(struct rte_eth_dev *dev,
1383 /**< @internal Remove a flex filter rule on an Ethernet device */
1385 typedef int (*eth_get_flex_filter_t)(struct rte_eth_dev *dev,
1386 uint16_t index, struct rte_flex_filter *filter,
1387 uint16_t *rx_queue);
1388 /**< @internal Get a flex filter rule on an Ethernet device */
1390 typedef int (*eth_filter_ctrl_t)(struct rte_eth_dev *dev,
1391 enum rte_filter_type filter_type,
1392 enum rte_filter_op filter_op,
1394 /**< @internal Take operations to assigned filter type on an Ethernet device */
1397 * @internal A structure containing the functions exported by an Ethernet driver.
1399 struct eth_dev_ops {
1400 eth_dev_configure_t dev_configure; /**< Configure device. */
1401 eth_dev_start_t dev_start; /**< Start device. */
1402 eth_dev_stop_t dev_stop; /**< Stop device. */
1403 eth_dev_set_link_up_t dev_set_link_up; /**< Device link up. */
1404 eth_dev_set_link_down_t dev_set_link_down; /**< Device link down. */
1405 eth_dev_close_t dev_close; /**< Close device. */
1406 eth_promiscuous_enable_t promiscuous_enable; /**< Promiscuous ON. */
1407 eth_promiscuous_disable_t promiscuous_disable;/**< Promiscuous OFF. */
1408 eth_allmulticast_enable_t allmulticast_enable;/**< RX multicast ON. */
1409 eth_allmulticast_disable_t allmulticast_disable;/**< RX multicast OF. */
1410 eth_link_update_t link_update; /**< Get device link state. */
1411 eth_stats_get_t stats_get; /**< Get generic device statistics. */
1412 eth_stats_reset_t stats_reset; /**< Reset generic device statistics. */
1413 eth_xstats_get_t xstats_get; /**< Get extended device statistics. */
1414 eth_xstats_reset_t xstats_reset; /**< Reset extended device statistics. */
1415 eth_queue_stats_mapping_set_t queue_stats_mapping_set;
1416 /**< Configure per queue stat counter mapping. */
1417 eth_dev_infos_get_t dev_infos_get; /**< Get device info. */
1418 mtu_set_t mtu_set; /**< Set MTU. */
1419 vlan_filter_set_t vlan_filter_set; /**< Filter VLAN Setup. */
1420 vlan_tpid_set_t vlan_tpid_set; /**< Outer VLAN TPID Setup. */
1421 vlan_strip_queue_set_t vlan_strip_queue_set; /**< VLAN Stripping on queue. */
1422 vlan_offload_set_t vlan_offload_set; /**< Set VLAN Offload. */
1423 vlan_pvid_set_t vlan_pvid_set; /**< Set port based TX VLAN insertion */
1424 eth_queue_start_t rx_queue_start;/**< Start RX for a queue.*/
1425 eth_queue_stop_t rx_queue_stop;/**< Stop RX for a queue.*/
1426 eth_queue_start_t tx_queue_start;/**< Start TX for a queue.*/
1427 eth_queue_stop_t tx_queue_stop;/**< Stop TX for a queue.*/
1428 eth_rx_queue_setup_t rx_queue_setup;/**< Set up device RX queue.*/
1429 eth_queue_release_t rx_queue_release;/**< Release RX queue.*/
1430 eth_rx_queue_count_t rx_queue_count; /**< Get Rx queue count. */
1431 eth_rx_descriptor_done_t rx_descriptor_done; /**< Check rxd DD bit */
1432 eth_tx_queue_setup_t tx_queue_setup;/**< Set up device TX queue.*/
1433 eth_queue_release_t tx_queue_release;/**< Release TX queue.*/
1434 eth_dev_led_on_t dev_led_on; /**< Turn on LED. */
1435 eth_dev_led_off_t dev_led_off; /**< Turn off LED. */
1436 flow_ctrl_get_t flow_ctrl_get; /**< Get flow control. */
1437 flow_ctrl_set_t flow_ctrl_set; /**< Setup flow control. */
1438 priority_flow_ctrl_set_t priority_flow_ctrl_set; /**< Setup priority flow control.*/
1439 eth_mac_addr_remove_t mac_addr_remove; /**< Remove MAC address */
1440 eth_mac_addr_add_t mac_addr_add; /**< Add a MAC address */
1441 eth_uc_hash_table_set_t uc_hash_table_set; /**< Set Unicast Table Array */
1442 eth_uc_all_hash_table_set_t uc_all_hash_table_set; /**< Set Unicast hash bitmap */
1443 eth_mirror_rule_set_t mirror_rule_set; /**< Add a traffic mirror rule.*/
1444 eth_mirror_rule_reset_t mirror_rule_reset; /**< reset a traffic mirror rule.*/
1445 eth_set_vf_rx_mode_t set_vf_rx_mode; /**< Set VF RX mode */
1446 eth_set_vf_rx_t set_vf_rx; /**< enable/disable a VF receive */
1447 eth_set_vf_tx_t set_vf_tx; /**< enable/disable a VF transmit */
1448 eth_set_vf_vlan_filter_t set_vf_vlan_filter; /**< Set VF VLAN filter */
1449 eth_set_queue_rate_limit_t set_queue_rate_limit; /**< Set queue rate limit */
1450 eth_set_vf_rate_limit_t set_vf_rate_limit; /**< Set VF rate limit */
1452 /** Add a signature filter. */
1453 fdir_add_signature_filter_t fdir_add_signature_filter;
1454 /** Update a signature filter. */
1455 fdir_update_signature_filter_t fdir_update_signature_filter;
1456 /** Remove a signature filter. */
1457 fdir_remove_signature_filter_t fdir_remove_signature_filter;
1458 /** Get information about FDIR status. */
1459 fdir_infos_get_t fdir_infos_get;
1460 /** Add a perfect filter. */
1461 fdir_add_perfect_filter_t fdir_add_perfect_filter;
1462 /** Update a perfect filter. */
1463 fdir_update_perfect_filter_t fdir_update_perfect_filter;
1464 /** Remove a perfect filter. */
1465 fdir_remove_perfect_filter_t fdir_remove_perfect_filter;
1466 /** Setup masks for FDIR filtering. */
1467 fdir_set_masks_t fdir_set_masks;
1468 /** Update redirection table. */
1469 reta_update_t reta_update;
1470 /** Query redirection table. */
1471 reta_query_t reta_query;
1472 /* bypass control */
1473 #ifdef RTE_NIC_BYPASS
1474 bypass_init_t bypass_init;
1475 bypass_state_set_t bypass_state_set;
1476 bypass_state_show_t bypass_state_show;
1477 bypass_event_set_t bypass_event_set;
1478 bypass_event_show_t bypass_event_show;
1479 bypass_wd_timeout_set_t bypass_wd_timeout_set;
1480 bypass_wd_timeout_show_t bypass_wd_timeout_show;
1481 bypass_ver_show_t bypass_ver_show;
1482 bypass_wd_reset_t bypass_wd_reset;
1485 /** Configure RSS hash protocols. */
1486 rss_hash_update_t rss_hash_update;
1487 /** Get current RSS hash configuration. */
1488 rss_hash_conf_get_t rss_hash_conf_get;
1489 eth_add_syn_filter_t add_syn_filter; /**< add syn filter. */
1490 eth_remove_syn_filter_t remove_syn_filter; /**< remove syn filter. */
1491 eth_get_syn_filter_t get_syn_filter; /**< get syn filter. */
1492 eth_add_ethertype_filter_t add_ethertype_filter; /**< add ethertype filter. */
1493 eth_remove_ethertype_filter_t remove_ethertype_filter; /**< remove ethertype filter. */
1494 eth_get_ethertype_filter_t get_ethertype_filter; /**< get ethertype filter. */
1495 eth_add_2tuple_filter_t add_2tuple_filter; /**< add 2tuple filter. */
1496 eth_remove_2tuple_filter_t remove_2tuple_filter; /**< remove 2tuple filter. */
1497 eth_get_2tuple_filter_t get_2tuple_filter; /**< get 2tuple filter. */
1498 eth_add_5tuple_filter_t add_5tuple_filter; /**< add 5tuple filter. */
1499 eth_remove_5tuple_filter_t remove_5tuple_filter; /**< remove 5tuple filter. */
1500 eth_get_5tuple_filter_t get_5tuple_filter; /**< get 5tuple filter. */
1501 eth_add_flex_filter_t add_flex_filter; /**< add flex filter. */
1502 eth_remove_flex_filter_t remove_flex_filter; /**< remove flex filter. */
1503 eth_get_flex_filter_t get_flex_filter; /**< get flex filter. */
1504 eth_filter_ctrl_t filter_ctrl; /**< common filter control*/
1509 * The generic data structure associated with each ethernet device.
1511 * Pointers to burst-oriented packet receive and transmit functions are
1512 * located at the beginning of the structure, along with the pointer to
1513 * where all the data elements for the particular device are stored in shared
1514 * memory. This split allows the function pointer and driver data to be per-
1515 * process, while the actual configuration data for the device is shared.
1517 struct rte_eth_dev {
1518 eth_rx_burst_t rx_pkt_burst; /**< Pointer to PMD receive function. */
1519 eth_tx_burst_t tx_pkt_burst; /**< Pointer to PMD transmit function. */
1520 struct rte_eth_dev_data *data; /**< Pointer to device data */
1521 const struct eth_driver *driver;/**< Driver for this device */
1522 struct eth_dev_ops *dev_ops; /**< Functions exported by PMD */
1523 struct rte_pci_device *pci_dev; /**< PCI info. supplied by probing */
1524 struct rte_eth_dev_cb_list callbacks; /**< User application callbacks */
1527 struct rte_eth_dev_sriov {
1528 uint8_t active; /**< SRIOV is active with 16, 32 or 64 pools */
1529 uint8_t nb_q_per_pool; /**< rx queue number per pool */
1530 uint16_t def_vmdq_idx; /**< Default pool num used for PF */
1531 uint16_t def_pool_q_idx; /**< Default pool queue start reg index */
1533 #define RTE_ETH_DEV_SRIOV(dev) ((dev)->data->sriov)
1535 #define RTE_ETH_NAME_MAX_LEN (32)
1539 * The data part, with no function pointers, associated with each ethernet device.
1541 * This structure is safe to place in shared memory to be common among different
1542 * processes in a multi-process configuration.
1544 struct rte_eth_dev_data {
1545 char name[RTE_ETH_NAME_MAX_LEN]; /**< Unique identifier name */
1547 void **rx_queues; /**< Array of pointers to RX queues. */
1548 void **tx_queues; /**< Array of pointers to TX queues. */
1549 uint16_t nb_rx_queues; /**< Number of RX queues. */
1550 uint16_t nb_tx_queues; /**< Number of TX queues. */
1552 struct rte_eth_dev_sriov sriov; /**< SRIOV data */
1554 void *dev_private; /**< PMD-specific private data */
1556 struct rte_eth_link dev_link;
1557 /**< Link-level information & status */
1559 struct rte_eth_conf dev_conf; /**< Configuration applied to device. */
1560 uint16_t mtu; /**< Maximum Transmission Unit. */
1562 uint32_t min_rx_buf_size;
1563 /**< Common rx buffer size handled by all queues */
1565 uint64_t rx_mbuf_alloc_failed; /**< RX ring mbuf allocation failures. */
1566 struct ether_addr* mac_addrs;/**< Device Ethernet Link address. */
1567 uint64_t mac_pool_sel[ETH_NUM_RECEIVE_MAC_ADDR];
1568 /** bitmap array of associating Ethernet MAC addresses to pools */
1569 struct ether_addr* hash_mac_addrs;
1570 /** Device Ethernet MAC addresses of hash filtering. */
1571 uint8_t port_id; /**< Device [external] port identifier. */
1572 uint8_t promiscuous : 1, /**< RX promiscuous mode ON(1) / OFF(0). */
1573 scattered_rx : 1, /**< RX of scattered packets is ON(1) / OFF(0) */
1574 all_multicast : 1, /**< RX all multicast mode ON(1) / OFF(0). */
1575 dev_started : 1; /**< Device state: STARTED(1) / STOPPED(0). */
1580 * The pool of *rte_eth_dev* structures. The size of the pool
1581 * is configured at compile-time in the <rte_ethdev.c> file.
1583 extern struct rte_eth_dev rte_eth_devices[];
1586 * Get the total number of Ethernet devices that have been successfully
1587 * initialized by the [matching] Ethernet driver during the PCI probing phase.
1588 * All devices whose port identifier is in the range
1589 * [0, rte_eth_dev_count() - 1] can be operated on by network applications.
1592 * - The total number of usable Ethernet devices.
1594 extern uint8_t rte_eth_dev_count(void);
1597 * Function for internal use by dummy drivers primarily, e.g. ring-based
1599 * Allocates a new ethdev slot for an ethernet device and returns the pointer
1600 * to that slot for the driver to use.
1602 * @param name Unique identifier name for each Ethernet device
1604 * - Slot in the rte_dev_devices array for a new device;
1606 struct rte_eth_dev *rte_eth_dev_allocate(const char *name);
1611 * Initialization function of an Ethernet driver invoked for each matching
1612 * Ethernet PCI device detected during the PCI probing phase.
1615 * The pointer to the [matching] Ethernet driver structure supplied by
1616 * the PMD when it registered itself.
1618 * The *eth_dev* pointer is the address of the *rte_eth_dev* structure
1619 * associated with the matching device and which have been [automatically]
1620 * allocated in the *rte_eth_devices* array.
1621 * The *eth_dev* structure is supplied to the driver initialization function
1622 * with the following fields already initialized:
1624 * - *pci_dev*: Holds the pointers to the *rte_pci_device* structure which
1625 * contains the generic PCI information of the matching device.
1627 * - *dev_private*: Holds a pointer to the device private data structure.
1629 * - *max_frame_size*: Contains the default Ethernet maximum frame length
1632 * - *port_id*: Contains the port index of the device (actually the index
1633 * of the *eth_dev* structure in the *rte_eth_devices* array).
1636 * - 0: Success, the device is properly initialized by the driver.
1637 * In particular, the driver MUST have set up the *dev_ops* pointer
1638 * of the *eth_dev* structure.
1639 * - <0: Error code of the device initialization failure.
1641 typedef int (*eth_dev_init_t)(struct eth_driver *eth_drv,
1642 struct rte_eth_dev *eth_dev);
1646 * The structure associated with a PMD Ethernet driver.
1648 * Each Ethernet driver acts as a PCI driver and is represented by a generic
1649 * *eth_driver* structure that holds:
1651 * - An *rte_pci_driver* structure (which must be the first field).
1653 * - The *eth_dev_init* function invoked for each matching PCI device.
1655 * - The size of the private data to allocate for each matching device.
1658 struct rte_pci_driver pci_drv; /**< The PMD is also a PCI driver. */
1659 eth_dev_init_t eth_dev_init; /**< Device init function. */
1660 unsigned int dev_private_size; /**< Size of device private data. */
1665 * A function invoked by the initialization function of an Ethernet driver
1666 * to simultaneously register itself as a PCI driver and as an Ethernet
1667 * Poll Mode Driver (PMD).
1670 * The pointer to the *eth_driver* structure associated with
1671 * the Ethernet driver.
1673 extern void rte_eth_driver_register(struct eth_driver *eth_drv);
1676 * Configure an Ethernet device.
1677 * This function must be invoked first before any other function in the
1678 * Ethernet API. This function can also be re-invoked when a device is in the
1682 * The port identifier of the Ethernet device to configure.
1683 * @param nb_rx_queue
1684 * The number of receive queues to set up for the Ethernet device.
1685 * @param nb_tx_queue
1686 * The number of transmit queues to set up for the Ethernet device.
1688 * The pointer to the configuration data to be used for the Ethernet device.
1689 * The *rte_eth_conf* structure includes:
1690 * - the hardware offload features to activate, with dedicated fields for
1691 * each statically configurable offload hardware feature provided by
1692 * Ethernet devices, such as IP checksum or VLAN tag stripping for
1694 * - the Receive Side Scaling (RSS) configuration when using multiple RX
1697 * Embedding all configuration information in a single data structure
1698 * is the more flexible method that allows the addition of new features
1699 * without changing the syntax of the API.
1701 * - 0: Success, device configured.
1702 * - <0: Error code returned by the driver configuration function.
1704 extern int rte_eth_dev_configure(uint8_t port_id,
1705 uint16_t nb_rx_queue,
1706 uint16_t nb_tx_queue,
1707 const struct rte_eth_conf *eth_conf);
1710 * Allocate and set up a receive queue for an Ethernet device.
1712 * The function allocates a contiguous block of memory for *nb_rx_desc*
1713 * receive descriptors from a memory zone associated with *socket_id*
1714 * and initializes each receive descriptor with a network buffer allocated
1715 * from the memory pool *mb_pool*.
1718 * The port identifier of the Ethernet device.
1719 * @param rx_queue_id
1720 * The index of the receive queue to set up.
1721 * The value must be in the range [0, nb_rx_queue - 1] previously supplied
1722 * to rte_eth_dev_configure().
1724 * The number of receive descriptors to allocate for the receive ring.
1726 * The *socket_id* argument is the socket identifier in case of NUMA.
1727 * The value can be *SOCKET_ID_ANY* if there is no NUMA constraint for
1728 * the DMA memory allocated for the receive descriptors of the ring.
1730 * The pointer to the configuration data to be used for the receive queue.
1731 * NULL value is allowed, in which case default RX configuration
1733 * The *rx_conf* structure contains an *rx_thresh* structure with the values
1734 * of the Prefetch, Host, and Write-Back threshold registers of the receive
1737 * The pointer to the memory pool from which to allocate *rte_mbuf* network
1738 * memory buffers to populate each descriptor of the receive ring.
1740 * - 0: Success, receive queue correctly set up.
1741 * - -EINVAL: The size of network buffers which can be allocated from the
1742 * memory pool does not fit the various buffer sizes allowed by the
1743 * device controller.
1744 * - -ENOMEM: Unable to allocate the receive ring descriptors or to
1745 * allocate network memory buffers from the memory pool when
1746 * initializing receive descriptors.
1748 extern int rte_eth_rx_queue_setup(uint8_t port_id, uint16_t rx_queue_id,
1749 uint16_t nb_rx_desc, unsigned int socket_id,
1750 const struct rte_eth_rxconf *rx_conf,
1751 struct rte_mempool *mb_pool);
1754 * Allocate and set up a transmit queue for an Ethernet device.
1757 * The port identifier of the Ethernet device.
1758 * @param tx_queue_id
1759 * The index of the transmit queue to set up.
1760 * The value must be in the range [0, nb_tx_queue - 1] previously supplied
1761 * to rte_eth_dev_configure().
1763 * The number of transmit descriptors to allocate for the transmit ring.
1765 * The *socket_id* argument is the socket identifier in case of NUMA.
1766 * Its value can be *SOCKET_ID_ANY* if there is no NUMA constraint for
1767 * the DMA memory allocated for the transmit descriptors of the ring.
1769 * The pointer to the configuration data to be used for the transmit queue.
1770 * NULL value is allowed, in which case default RX configuration
1772 * The *tx_conf* structure contains the following data:
1773 * - The *tx_thresh* structure with the values of the Prefetch, Host, and
1774 * Write-Back threshold registers of the transmit ring.
1775 * When setting Write-Back threshold to the value greater then zero,
1776 * *tx_rs_thresh* value should be explicitly set to one.
1777 * - The *tx_free_thresh* value indicates the [minimum] number of network
1778 * buffers that must be pending in the transmit ring to trigger their
1779 * [implicit] freeing by the driver transmit function.
1780 * - The *tx_rs_thresh* value indicates the [minimum] number of transmit
1781 * descriptors that must be pending in the transmit ring before setting the
1782 * RS bit on a descriptor by the driver transmit function.
1783 * The *tx_rs_thresh* value should be less or equal then
1784 * *tx_free_thresh* value, and both of them should be less then
1786 * - The *txq_flags* member contains flags to pass to the TX queue setup
1787 * function to configure the behavior of the TX queue. This should be set
1788 * to 0 if no special configuration is required.
1790 * Note that setting *tx_free_thresh* or *tx_rs_thresh* value to 0 forces
1791 * the transmit function to use default values.
1793 * - 0: Success, the transmit queue is correctly set up.
1794 * - -ENOMEM: Unable to allocate the transmit ring descriptors.
1796 extern int rte_eth_tx_queue_setup(uint8_t port_id, uint16_t tx_queue_id,
1797 uint16_t nb_tx_desc, unsigned int socket_id,
1798 const struct rte_eth_txconf *tx_conf);
1801 * Return the NUMA socket to which an Ethernet device is connected
1804 * The port identifier of the Ethernet device
1806 * The NUMA socket id to which the Ethernet device is connected or
1807 * a default of zero if the socket could not be determined.
1808 * -1 is returned is the port_id value is out of range.
1810 extern int rte_eth_dev_socket_id(uint8_t port_id);
1813 * Allocate mbuf from mempool, setup the DMA physical address
1814 * and then start RX for specified queue of a port. It is used
1815 * when rx_deferred_start flag of the specified queue is true.
1818 * The port identifier of the Ethernet device
1819 * @param rx_queue_id
1820 * The index of the rx queue to update the ring.
1821 * The value must be in the range [0, nb_rx_queue - 1] previously supplied
1822 * to rte_eth_dev_configure().
1824 * - 0: Success, the transmit queue is correctly set up.
1825 * - -EINVAL: The port_id or the queue_id out of range.
1826 * - -ENOTSUP: The function not supported in PMD driver.
1828 extern int rte_eth_dev_rx_queue_start(uint8_t port_id, uint16_t rx_queue_id);
1831 * Stop specified RX queue of a port
1834 * The port identifier of the Ethernet device
1835 * @param rx_queue_id
1836 * The index of the rx queue to update the ring.
1837 * The value must be in the range [0, nb_rx_queue - 1] previously supplied
1838 * to rte_eth_dev_configure().
1840 * - 0: Success, the transmit queue is correctly set up.
1841 * - -EINVAL: The port_id or the queue_id out of range.
1842 * - -ENOTSUP: The function not supported in PMD driver.
1844 extern int rte_eth_dev_rx_queue_stop(uint8_t port_id, uint16_t rx_queue_id);
1847 * Start TX for specified queue of a port. It is used when tx_deferred_start
1848 * flag of the specified queue is true.
1851 * The port identifier of the Ethernet device
1852 * @param tx_queue_id
1853 * The index of the tx queue to update the ring.
1854 * The value must be in the range [0, nb_tx_queue - 1] previously supplied
1855 * to rte_eth_dev_configure().
1857 * - 0: Success, the transmit queue is correctly set up.
1858 * - -EINVAL: The port_id or the queue_id out of range.
1859 * - -ENOTSUP: The function not supported in PMD driver.
1861 extern int rte_eth_dev_tx_queue_start(uint8_t port_id, uint16_t tx_queue_id);
1864 * Stop specified TX queue of a port
1867 * The port identifier of the Ethernet device
1868 * @param tx_queue_id
1869 * The index of the tx queue to update the ring.
1870 * The value must be in the range [0, nb_tx_queue - 1] previously supplied
1871 * to rte_eth_dev_configure().
1873 * - 0: Success, the transmit queue is correctly set up.
1874 * - -EINVAL: The port_id or the queue_id out of range.
1875 * - -ENOTSUP: The function not supported in PMD driver.
1877 extern int rte_eth_dev_tx_queue_stop(uint8_t port_id, uint16_t tx_queue_id);
1882 * Start an Ethernet device.
1884 * The device start step is the last one and consists of setting the configured
1885 * offload features and in starting the transmit and the receive units of the
1887 * On success, all basic functions exported by the Ethernet API (link status,
1888 * receive/transmit, and so on) can be invoked.
1891 * The port identifier of the Ethernet device.
1893 * - 0: Success, Ethernet device started.
1894 * - <0: Error code of the driver device start function.
1896 extern int rte_eth_dev_start(uint8_t port_id);
1899 * Stop an Ethernet device. The device can be restarted with a call to
1900 * rte_eth_dev_start()
1903 * The port identifier of the Ethernet device.
1905 extern void rte_eth_dev_stop(uint8_t port_id);
1909 * Link up an Ethernet device.
1911 * Set device link up will re-enable the device rx/tx
1912 * functionality after it is previously set device linked down.
1915 * The port identifier of the Ethernet device.
1917 * - 0: Success, Ethernet device linked up.
1918 * - <0: Error code of the driver device link up function.
1920 extern int rte_eth_dev_set_link_up(uint8_t port_id);
1923 * Link down an Ethernet device.
1924 * The device rx/tx functionality will be disabled if success,
1925 * and it can be re-enabled with a call to
1926 * rte_eth_dev_set_link_up()
1929 * The port identifier of the Ethernet device.
1931 extern int rte_eth_dev_set_link_down(uint8_t port_id);
1934 * Close an Ethernet device. The device cannot be restarted!
1937 * The port identifier of the Ethernet device.
1939 extern void rte_eth_dev_close(uint8_t port_id);
1942 * Enable receipt in promiscuous mode for an Ethernet device.
1945 * The port identifier of the Ethernet device.
1947 extern void rte_eth_promiscuous_enable(uint8_t port_id);
1950 * Disable receipt in promiscuous mode for an Ethernet device.
1953 * The port identifier of the Ethernet device.
1955 extern void rte_eth_promiscuous_disable(uint8_t port_id);
1958 * Return the value of promiscuous mode for an Ethernet device.
1961 * The port identifier of the Ethernet device.
1963 * - (1) if promiscuous is enabled
1964 * - (0) if promiscuous is disabled.
1967 extern int rte_eth_promiscuous_get(uint8_t port_id);
1970 * Enable the receipt of any multicast frame by an Ethernet device.
1973 * The port identifier of the Ethernet device.
1975 extern void rte_eth_allmulticast_enable(uint8_t port_id);
1978 * Disable the receipt of all multicast frames by an Ethernet device.
1981 * The port identifier of the Ethernet device.
1983 extern void rte_eth_allmulticast_disable(uint8_t port_id);
1986 * Return the value of allmulticast mode for an Ethernet device.
1989 * The port identifier of the Ethernet device.
1991 * - (1) if allmulticast is enabled
1992 * - (0) if allmulticast is disabled.
1995 extern int rte_eth_allmulticast_get(uint8_t port_id);
1998 * Retrieve the status (ON/OFF), the speed (in Mbps) and the mode (HALF-DUPLEX
1999 * or FULL-DUPLEX) of the physical link of an Ethernet device. It might need
2000 * to wait up to 9 seconds in it.
2003 * The port identifier of the Ethernet device.
2005 * A pointer to an *rte_eth_link* structure to be filled with
2006 * the status, the speed and the mode of the Ethernet device link.
2008 extern void rte_eth_link_get(uint8_t port_id, struct rte_eth_link *link);
2011 * Retrieve the status (ON/OFF), the speed (in Mbps) and the mode (HALF-DUPLEX
2012 * or FULL-DUPLEX) of the physical link of an Ethernet device. It is a no-wait
2013 * version of rte_eth_link_get().
2016 * The port identifier of the Ethernet device.
2018 * A pointer to an *rte_eth_link* structure to be filled with
2019 * the status, the speed and the mode of the Ethernet device link.
2021 extern void rte_eth_link_get_nowait(uint8_t port_id,
2022 struct rte_eth_link *link);
2025 * Retrieve the general I/O statistics of an Ethernet device.
2028 * The port identifier of the Ethernet device.
2030 * A pointer to a structure of type *rte_eth_stats* to be filled with
2031 * the values of device counters for the following set of statistics:
2032 * - *ipackets* with the total of successfully received packets.
2033 * - *opackets* with the total of successfully transmitted packets.
2034 * - *ibytes* with the total of successfully received bytes.
2035 * - *obytes* with the total of successfully transmitted bytes.
2036 * - *ierrors* with the total of erroneous received packets.
2037 * - *oerrors* with the total of failed transmitted packets.
2039 extern void rte_eth_stats_get(uint8_t port_id, struct rte_eth_stats *stats);
2042 * Reset the general I/O statistics of an Ethernet device.
2045 * The port identifier of the Ethernet device.
2047 extern void rte_eth_stats_reset(uint8_t port_id);
2050 * Retrieve extended statistics of an Ethernet device.
2053 * The port identifier of the Ethernet device.
2055 * A pointer to a table of structure of type *rte_eth_xstats*
2056 * to be filled with device statistics names and values.
2057 * This parameter can be set to NULL if n is 0.
2059 * The size of the stats table, which should be large enough to store
2060 * all the statistics of the device.
2062 * - positive value lower or equal to n: success. The return value
2063 * is the number of entries filled in the stats table.
2064 * - positive value higher than n: error, the given statistics table
2065 * is too small. The return value corresponds to the size that should
2066 * be given to succeed. The entries in the table are not valid and
2067 * shall not be used by the caller.
2068 * - negative value on error (invalid port id)
2070 extern int rte_eth_xstats_get(uint8_t port_id,
2071 struct rte_eth_xstats *xstats, unsigned n);
2074 * Reset extended statistics of an Ethernet device.
2077 * The port identifier of the Ethernet device.
2079 extern void rte_eth_xstats_reset(uint8_t port_id);
2082 * Set a mapping for the specified transmit queue to the specified per-queue
2083 * statistics counter.
2086 * The port identifier of the Ethernet device.
2087 * @param tx_queue_id
2088 * The index of the transmit queue for which a queue stats mapping is required.
2089 * The value must be in the range [0, nb_tx_queue - 1] previously supplied
2090 * to rte_eth_dev_configure().
2092 * The per-queue packet statistics functionality number that the transmit
2093 * queue is to be assigned.
2094 * The value must be in the range [0, RTE_MAX_ETHPORT_QUEUE_STATS_MAPS - 1].
2096 * Zero if successful. Non-zero otherwise.
2098 extern int rte_eth_dev_set_tx_queue_stats_mapping(uint8_t port_id,
2099 uint16_t tx_queue_id,
2103 * Set a mapping for the specified receive queue to the specified per-queue
2104 * statistics counter.
2107 * The port identifier of the Ethernet device.
2108 * @param rx_queue_id
2109 * The index of the receive queue for which a queue stats mapping is required.
2110 * The value must be in the range [0, nb_rx_queue - 1] previously supplied
2111 * to rte_eth_dev_configure().
2113 * The per-queue packet statistics functionality number that the receive
2114 * queue is to be assigned.
2115 * The value must be in the range [0, RTE_MAX_ETHPORT_QUEUE_STATS_MAPS - 1].
2117 * Zero if successful. Non-zero otherwise.
2119 extern int rte_eth_dev_set_rx_queue_stats_mapping(uint8_t port_id,
2120 uint16_t rx_queue_id,
2124 * Retrieve the Ethernet address of an Ethernet device.
2127 * The port identifier of the Ethernet device.
2129 * A pointer to a structure of type *ether_addr* to be filled with
2130 * the Ethernet address of the Ethernet device.
2132 extern void rte_eth_macaddr_get(uint8_t port_id, struct ether_addr *mac_addr);
2135 * Retrieve the contextual information of an Ethernet device.
2138 * The port identifier of the Ethernet device.
2140 * A pointer to a structure of type *rte_eth_dev_info* to be filled with
2141 * the contextual information of the Ethernet device.
2143 extern void rte_eth_dev_info_get(uint8_t port_id,
2144 struct rte_eth_dev_info *dev_info);
2147 * Retrieve the MTU of an Ethernet device.
2150 * The port identifier of the Ethernet device.
2152 * A pointer to a uint16_t where the retrieved MTU is to be stored.
2154 * - (0) if successful.
2155 * - (-ENODEV) if *port_id* invalid.
2157 extern int rte_eth_dev_get_mtu(uint8_t port_id, uint16_t *mtu);
2160 * Change the MTU of an Ethernet device.
2163 * The port identifier of the Ethernet device.
2165 * A uint16_t for the MTU to be applied.
2167 * - (0) if successful.
2168 * - (-ENOTSUP) if operation is not supported.
2169 * - (-ENODEV) if *port_id* invalid.
2170 * - (-EINVAL) if *mtu* invalid.
2172 extern int rte_eth_dev_set_mtu(uint8_t port_id, uint16_t mtu);
2175 * Enable/Disable hardware filtering by an Ethernet device of received
2176 * VLAN packets tagged with a given VLAN Tag Identifier.
2179 * The port identifier of the Ethernet device.
2181 * The VLAN Tag Identifier whose filtering must be enabled or disabled.
2183 * If > 0, enable VLAN filtering of VLAN packets tagged with *vlan_id*.
2184 * Otherwise, disable VLAN filtering of VLAN packets tagged with *vlan_id*.
2186 * - (0) if successful.
2187 * - (-ENOSUP) if hardware-assisted VLAN filtering not configured.
2188 * - (-ENODEV) if *port_id* invalid.
2189 * - (-ENOSYS) if VLAN filtering on *port_id* disabled.
2190 * - (-EINVAL) if *vlan_id* > 4095.
2192 extern int rte_eth_dev_vlan_filter(uint8_t port_id, uint16_t vlan_id , int on);
2195 * Enable/Disable hardware VLAN Strip by a rx queue of an Ethernet device.
2196 * 82599/X540/X550 can support VLAN stripping at the rx queue level
2199 * The port identifier of the Ethernet device.
2200 * @param rx_queue_id
2201 * The index of the receive queue for which a queue stats mapping is required.
2202 * The value must be in the range [0, nb_rx_queue - 1] previously supplied
2203 * to rte_eth_dev_configure().
2205 * If 1, Enable VLAN Stripping of the receive queue of the Ethernet port.
2206 * If 0, Disable VLAN Stripping of the receive queue of the Ethernet port.
2208 * - (0) if successful.
2209 * - (-ENOSUP) if hardware-assisted VLAN stripping not configured.
2210 * - (-ENODEV) if *port_id* invalid.
2211 * - (-EINVAL) if *rx_queue_id* invalid.
2213 extern int rte_eth_dev_set_vlan_strip_on_queue(uint8_t port_id,
2214 uint16_t rx_queue_id, int on);
2217 * Set the Outer VLAN Ether Type by an Ethernet device, it can be inserted to
2218 * the VLAN Header. This is a register setup available on some Intel NIC, not
2219 * but all, please check the data sheet for availability.
2222 * The port identifier of the Ethernet device.
2224 * The Tag Protocol ID
2226 * - (0) if successful.
2227 * - (-ENOSUP) if hardware-assisted VLAN TPID setup is not supported.
2228 * - (-ENODEV) if *port_id* invalid.
2230 extern int rte_eth_dev_set_vlan_ether_type(uint8_t port_id, uint16_t tag_type);
2233 * Set VLAN offload configuration on an Ethernet device
2234 * Enable/Disable Extended VLAN by an Ethernet device, This is a register setup
2235 * available on some Intel NIC, not but all, please check the data sheet for
2237 * Enable/Disable VLAN Strip can be done on rx queue for certain NIC, but here
2238 * the configuration is applied on the port level.
2241 * The port identifier of the Ethernet device.
2242 * @param offload_mask
2243 * The VLAN Offload bit mask can be mixed use with "OR"
2244 * ETH_VLAN_STRIP_OFFLOAD
2245 * ETH_VLAN_FILTER_OFFLOAD
2246 * ETH_VLAN_EXTEND_OFFLOAD
2248 * - (0) if successful.
2249 * - (-ENOSUP) if hardware-assisted VLAN filtering not configured.
2250 * - (-ENODEV) if *port_id* invalid.
2252 extern int rte_eth_dev_set_vlan_offload(uint8_t port_id, int offload_mask);
2255 * Read VLAN Offload configuration from an Ethernet device
2258 * The port identifier of the Ethernet device.
2260 * - (>0) if successful. Bit mask to indicate
2261 * ETH_VLAN_STRIP_OFFLOAD
2262 * ETH_VLAN_FILTER_OFFLOAD
2263 * ETH_VLAN_EXTEND_OFFLOAD
2264 * - (-ENODEV) if *port_id* invalid.
2266 extern int rte_eth_dev_get_vlan_offload(uint8_t port_id);
2269 * Set port based TX VLAN insersion on or off.
2272 * The port identifier of the Ethernet device.
2274 * Port based TX VLAN identifier togeth with user priority.
2276 * Turn on or off the port based TX VLAN insertion.
2279 * - (0) if successful.
2280 * - negative if failed.
2282 extern int rte_eth_dev_set_vlan_pvid(uint8_t port_id, uint16_t pvid, int on);
2286 * Retrieve a burst of input packets from a receive queue of an Ethernet
2287 * device. The retrieved packets are stored in *rte_mbuf* structures whose
2288 * pointers are supplied in the *rx_pkts* array.
2290 * The rte_eth_rx_burst() function loops, parsing the RX ring of the
2291 * receive queue, up to *nb_pkts* packets, and for each completed RX
2292 * descriptor in the ring, it performs the following operations:
2294 * - Initialize the *rte_mbuf* data structure associated with the
2295 * RX descriptor according to the information provided by the NIC into
2296 * that RX descriptor.
2298 * - Store the *rte_mbuf* data structure into the next entry of the
2301 * - Replenish the RX descriptor with a new *rte_mbuf* buffer
2302 * allocated from the memory pool associated with the receive queue at
2303 * initialization time.
2305 * When retrieving an input packet that was scattered by the controller
2306 * into multiple receive descriptors, the rte_eth_rx_burst() function
2307 * appends the associated *rte_mbuf* buffers to the first buffer of the
2310 * The rte_eth_rx_burst() function returns the number of packets
2311 * actually retrieved, which is the number of *rte_mbuf* data structures
2312 * effectively supplied into the *rx_pkts* array.
2313 * A return value equal to *nb_pkts* indicates that the RX queue contained
2314 * at least *rx_pkts* packets, and this is likely to signify that other
2315 * received packets remain in the input queue. Applications implementing
2316 * a "retrieve as much received packets as possible" policy can check this
2317 * specific case and keep invoking the rte_eth_rx_burst() function until
2318 * a value less than *nb_pkts* is returned.
2320 * This receive method has the following advantages:
2322 * - It allows a run-to-completion network stack engine to retrieve and
2323 * to immediately process received packets in a fast burst-oriented
2324 * approach, avoiding the overhead of unnecessary intermediate packet
2325 * queue/dequeue operations.
2327 * - Conversely, it also allows an asynchronous-oriented processing
2328 * method to retrieve bursts of received packets and to immediately
2329 * queue them for further parallel processing by another logical core,
2330 * for instance. However, instead of having received packets being
2331 * individually queued by the driver, this approach allows the invoker
2332 * of the rte_eth_rx_burst() function to queue a burst of retrieved
2333 * packets at a time and therefore dramatically reduce the cost of
2334 * enqueue/dequeue operations per packet.
2336 * - It allows the rte_eth_rx_burst() function of the driver to take
2337 * advantage of burst-oriented hardware features (CPU cache,
2338 * prefetch instructions, and so on) to minimize the number of CPU
2339 * cycles per packet.
2341 * To summarize, the proposed receive API enables many
2342 * burst-oriented optimizations in both synchronous and asynchronous
2343 * packet processing environments with no overhead in both cases.
2345 * The rte_eth_rx_burst() function does not provide any error
2346 * notification to avoid the corresponding overhead. As a hint, the
2347 * upper-level application might check the status of the device link once
2348 * being systematically returned a 0 value for a given number of tries.
2351 * The port identifier of the Ethernet device.
2353 * The index of the receive queue from which to retrieve input packets.
2354 * The value must be in the range [0, nb_rx_queue - 1] previously supplied
2355 * to rte_eth_dev_configure().
2357 * The address of an array of pointers to *rte_mbuf* structures that
2358 * must be large enough to store *nb_pkts* pointers in it.
2360 * The maximum number of packets to retrieve.
2362 * The number of packets actually retrieved, which is the number
2363 * of pointers to *rte_mbuf* structures effectively supplied to the
2366 #ifdef RTE_LIBRTE_ETHDEV_DEBUG
2367 extern uint16_t rte_eth_rx_burst(uint8_t port_id, uint16_t queue_id,
2368 struct rte_mbuf **rx_pkts, uint16_t nb_pkts);
2370 static inline uint16_t
2371 rte_eth_rx_burst(uint8_t port_id, uint16_t queue_id,
2372 struct rte_mbuf **rx_pkts, uint16_t nb_pkts)
2374 struct rte_eth_dev *dev;
2376 dev = &rte_eth_devices[port_id];
2377 return (*dev->rx_pkt_burst)(dev->data->rx_queues[queue_id], rx_pkts, nb_pkts);
2382 * Get the number of used descriptors in a specific queue
2385 * The port identifier of the Ethernet device.
2387 * The queue id on the specific port.
2389 * The number of used descriptors in the specific queue.
2391 #ifdef RTE_LIBRTE_ETHDEV_DEBUG
2392 extern uint32_t rte_eth_rx_queue_count(uint8_t port_id, uint16_t queue_id);
2394 static inline uint32_t
2395 rte_eth_rx_queue_count(uint8_t port_id, uint16_t queue_id)
2397 struct rte_eth_dev *dev;
2399 dev = &rte_eth_devices[port_id];
2400 return (*dev->dev_ops->rx_queue_count)(dev, queue_id);
2405 * Check if the DD bit of the specific RX descriptor in the queue has been set
2408 * The port identifier of the Ethernet device.
2410 * The queue id on the specific port.
2412 * The offset of the descriptor ID from tail.
2414 * - (1) if the specific DD bit is set.
2415 * - (0) if the specific DD bit is not set.
2416 * - (-ENODEV) if *port_id* invalid.
2418 #ifdef RTE_LIBRTE_ETHDEV_DEBUG
2419 extern int rte_eth_rx_descriptor_done(uint8_t port_id,
2424 rte_eth_rx_descriptor_done(uint8_t port_id, uint16_t queue_id, uint16_t offset)
2426 struct rte_eth_dev *dev;
2428 dev = &rte_eth_devices[port_id];
2429 return (*dev->dev_ops->rx_descriptor_done)( \
2430 dev->data->rx_queues[queue_id], offset);
2435 * Send a burst of output packets on a transmit queue of an Ethernet device.
2437 * The rte_eth_tx_burst() function is invoked to transmit output packets
2438 * on the output queue *queue_id* of the Ethernet device designated by its
2440 * The *nb_pkts* parameter is the number of packets to send which are
2441 * supplied in the *tx_pkts* array of *rte_mbuf* structures.
2442 * The rte_eth_tx_burst() function loops, sending *nb_pkts* packets,
2443 * up to the number of transmit descriptors available in the TX ring of the
2445 * For each packet to send, the rte_eth_tx_burst() function performs
2446 * the following operations:
2448 * - Pick up the next available descriptor in the transmit ring.
2450 * - Free the network buffer previously sent with that descriptor, if any.
2452 * - Initialize the transmit descriptor with the information provided
2453 * in the *rte_mbuf data structure.
2455 * In the case of a segmented packet composed of a list of *rte_mbuf* buffers,
2456 * the rte_eth_tx_burst() function uses several transmit descriptors
2459 * The rte_eth_tx_burst() function returns the number of packets it
2460 * actually sent. A return value equal to *nb_pkts* means that all packets
2461 * have been sent, and this is likely to signify that other output packets
2462 * could be immediately transmitted again. Applications that implement a
2463 * "send as many packets to transmit as possible" policy can check this
2464 * specific case and keep invoking the rte_eth_tx_burst() function until
2465 * a value less than *nb_pkts* is returned.
2467 * It is the responsibility of the rte_eth_tx_burst() function to
2468 * transparently free the memory buffers of packets previously sent.
2469 * This feature is driven by the *tx_free_thresh* value supplied to the
2470 * rte_eth_dev_configure() function at device configuration time.
2471 * When the number of previously sent packets reached the "minimum transmit
2472 * packets to free" threshold, the rte_eth_tx_burst() function must
2473 * [attempt to] free the *rte_mbuf* buffers of those packets whose
2474 * transmission was effectively completed.
2477 * The port identifier of the Ethernet device.
2479 * The index of the transmit queue through which output packets must be
2481 * The value must be in the range [0, nb_tx_queue - 1] previously supplied
2482 * to rte_eth_dev_configure().
2484 * The address of an array of *nb_pkts* pointers to *rte_mbuf* structures
2485 * which contain the output packets.
2487 * The maximum number of packets to transmit.
2489 * The number of output packets actually stored in transmit descriptors of
2490 * the transmit ring. The return value can be less than the value of the
2491 * *tx_pkts* parameter when the transmit ring is full or has been filled up.
2493 #ifdef RTE_LIBRTE_ETHDEV_DEBUG
2494 extern uint16_t rte_eth_tx_burst(uint8_t port_id, uint16_t queue_id,
2495 struct rte_mbuf **tx_pkts, uint16_t nb_pkts);
2497 static inline uint16_t
2498 rte_eth_tx_burst(uint8_t port_id, uint16_t queue_id,
2499 struct rte_mbuf **tx_pkts, uint16_t nb_pkts)
2501 struct rte_eth_dev *dev;
2503 dev = &rte_eth_devices[port_id];
2504 return (*dev->tx_pkt_burst)(dev->data->tx_queues[queue_id], tx_pkts, nb_pkts);
2509 * Setup a new signature filter rule on an Ethernet device
2512 * The port identifier of the Ethernet device.
2513 * @param fdir_filter
2514 * The pointer to the fdir filter structure describing the signature filter
2516 * The *rte_fdir_filter* structure includes the values of the different fields
2517 * to match: source and destination IP addresses, vlan id, flexbytes, source
2518 * and destination ports, and so on.
2520 * The index of the RX queue where to store RX packets matching the added
2521 * signature filter defined in fdir_filter.
2523 * - (0) if successful.
2524 * - (-ENOTSUP) if hardware doesn't support flow director mode.
2525 * - (-ENODEV) if *port_id* invalid.
2526 * - (-ENOSYS) if the FDIR mode is not configured in signature mode
2528 * - (-EINVAL) if the fdir_filter information is not correct.
2530 int rte_eth_dev_fdir_add_signature_filter(uint8_t port_id,
2531 struct rte_fdir_filter *fdir_filter,
2535 * Update a signature filter rule on an Ethernet device.
2536 * If the rule doesn't exits, it is created.
2539 * The port identifier of the Ethernet device.
2541 * The pointer to the structure describing the signature filter rule.
2542 * The *rte_fdir_filter* structure includes the values of the different fields
2543 * to match: source and destination IP addresses, vlan id, flexbytes, source
2544 * and destination ports, and so on.
2546 * The index of the RX queue where to store RX packets matching the added
2547 * signature filter defined in fdir_ftr.
2549 * - (0) if successful.
2550 * - (-ENOTSUP) if hardware doesn't support flow director mode.
2551 * - (-ENODEV) if *port_id* invalid.
2552 * - (-ENOSYS) if the flow director mode is not configured in signature mode
2554 * - (-EINVAL) if the fdir_filter information is not correct.
2556 int rte_eth_dev_fdir_update_signature_filter(uint8_t port_id,
2557 struct rte_fdir_filter *fdir_ftr,
2561 * Remove a signature filter rule on an Ethernet device.
2564 * The port identifier of the Ethernet device.
2566 * The pointer to the structure describing the signature filter rule.
2567 * The *rte_fdir_filter* structure includes the values of the different fields
2568 * to match: source and destination IP addresses, vlan id, flexbytes, source
2569 * and destination ports, and so on.
2571 * - (0) if successful.
2572 * - (-ENOTSUP) if hardware doesn't support flow director mode.
2573 * - (-ENODEV) if *port_id* invalid.
2574 * - (-ENOSYS) if the flow director mode is not configured in signature mode
2576 * - (-EINVAL) if the fdir_filter information is not correct.
2578 int rte_eth_dev_fdir_remove_signature_filter(uint8_t port_id,
2579 struct rte_fdir_filter *fdir_ftr);
2582 * Retrieve the flow director information of an Ethernet device.
2585 * The port identifier of the Ethernet device.
2587 * A pointer to a structure of type *rte_eth_dev_fdir* to be filled with
2588 * the flow director information of the Ethernet device.
2590 * - (0) if successful.
2591 * - (-ENOTSUP) if hardware doesn't support flow director mode.
2592 * - (-ENODEV) if *port_id* invalid.
2593 * - (-ENOSYS) if the flow director mode is not configured on *port_id*.
2595 int rte_eth_dev_fdir_get_infos(uint8_t port_id, struct rte_eth_fdir *fdir);
2598 * Add a new perfect filter rule on an Ethernet device.
2601 * The port identifier of the Ethernet device.
2602 * @param fdir_filter
2603 * The pointer to the structure describing the perfect filter rule.
2604 * The *rte_fdir_filter* structure includes the values of the different fields
2605 * to match: source and destination IP addresses, vlan id, flexbytes, source
2606 * and destination ports, and so on.
2607 * IPv6 are not supported.
2609 * The 16-bit value supplied in the field hash.fdir.id of mbuf for RX
2610 * packets matching the perfect filter.
2612 * The index of the RX queue where to store RX packets matching the added
2613 * perfect filter defined in fdir_filter.
2615 * If drop is set to 1, matching RX packets are stored into the RX drop
2616 * queue defined in the rte_fdir_conf.
2618 * - (0) if successful.
2619 * - (-ENOTSUP) if hardware doesn't support flow director mode.
2620 * - (-ENODEV) if *port_id* invalid.
2621 * - (-ENOSYS) if the flow director mode is not configured in perfect mode
2623 * - (-EINVAL) if the fdir_filter information is not correct.
2625 int rte_eth_dev_fdir_add_perfect_filter(uint8_t port_id,
2626 struct rte_fdir_filter *fdir_filter,
2627 uint16_t soft_id, uint8_t rx_queue,
2631 * Update a perfect filter rule on an Ethernet device.
2632 * If the rule doesn't exits, it is created.
2635 * The port identifier of the Ethernet device.
2636 * @param fdir_filter
2637 * The pointer to the structure describing the perfect filter rule.
2638 * The *rte_fdir_filter* structure includes the values of the different fields
2639 * to match: source and destination IP addresses, vlan id, flexbytes, source
2640 * and destination ports, and so on.
2641 * IPv6 are not supported.
2643 * The 16-bit value supplied in the field hash.fdir.id of mbuf for RX
2644 * packets matching the perfect filter.
2646 * The index of the RX queue where to store RX packets matching the added
2647 * perfect filter defined in fdir_filter.
2649 * If drop is set to 1, matching RX packets are stored into the RX drop
2650 * queue defined in the rte_fdir_conf.
2652 * - (0) if successful.
2653 * - (-ENOTSUP) if hardware doesn't support flow director mode.
2654 * - (-ENODEV) if *port_id* invalid.
2655 * - (-ENOSYS) if the flow director mode is not configured in perfect mode
2657 * - (-EINVAL) if the fdir_filter information is not correct.
2659 int rte_eth_dev_fdir_update_perfect_filter(uint8_t port_id,
2660 struct rte_fdir_filter *fdir_filter,
2661 uint16_t soft_id, uint8_t rx_queue,
2665 * Remove a perfect filter rule on an Ethernet device.
2668 * The port identifier of the Ethernet device.
2669 * @param fdir_filter
2670 * The pointer to the structure describing the perfect filter rule.
2671 * The *rte_fdir_filter* structure includes the values of the different fields
2672 * to match: source and destination IP addresses, vlan id, flexbytes, source
2673 * and destination ports, and so on.
2674 * IPv6 are not supported.
2676 * The soft_id value provided when adding/updating the removed filter.
2678 * - (0) if successful.
2679 * - (-ENOTSUP) if hardware doesn't support flow director mode.
2680 * - (-ENODEV) if *port_id* invalid.
2681 * - (-ENOSYS) if the flow director mode is not configured in perfect mode
2683 * - (-EINVAL) if the fdir_filter information is not correct.
2685 int rte_eth_dev_fdir_remove_perfect_filter(uint8_t port_id,
2686 struct rte_fdir_filter *fdir_filter,
2689 * Configure globally the masks for flow director mode for an Ethernet device.
2690 * For example, the device can match packets with only the first 24 bits of
2691 * the IPv4 source address.
2693 * The following fields can be masked: IPv4 addresses and L4 port numbers.
2694 * The following fields can be either enabled or disabled completely for the
2695 * matching functionality: VLAN ID tag; VLAN Priority + CFI bit; Flexible 2-byte
2697 * IPv6 masks are not supported.
2699 * All filters must comply with the masks previously configured.
2700 * For example, with a mask equal to 255.255.255.0 for the source IPv4 address,
2701 * all IPv4 filters must be created with a source IPv4 address that fits the
2704 * This function flushes all filters that have been previously added in
2708 * The port identifier of the Ethernet device.
2710 * The pointer to the fdir mask structure describing relevant headers fields
2711 * and relevant bits to use when matching packets addresses and ports.
2712 * IPv6 masks are not supported.
2714 * - (0) if successful.
2715 * - (-ENOTSUP) if hardware doesn't support flow director mode.
2716 * - (-ENODEV) if *port_id* invalid.
2717 * - (-ENOSYS) if the flow director mode is not configured in perfect
2718 * mode on *port_id*.
2719 * - (-EINVAL) if the fdir_filter information is not correct
2721 int rte_eth_dev_fdir_set_masks(uint8_t port_id,
2722 struct rte_fdir_masks *fdir_mask);
2725 * The eth device event type for interrupt, and maybe others in the future.
2727 enum rte_eth_event_type {
2728 RTE_ETH_EVENT_UNKNOWN, /**< unknown event type */
2729 RTE_ETH_EVENT_INTR_LSC, /**< lsc interrupt event */
2730 RTE_ETH_EVENT_MAX /**< max value of this enum */
2733 typedef void (*rte_eth_dev_cb_fn)(uint8_t port_id, \
2734 enum rte_eth_event_type event, void *cb_arg);
2735 /**< user application callback to be registered for interrupts */
2740 * Register a callback function for specific port id.
2747 * User supplied callback function to be called.
2749 * Pointer to the parameters for the registered callback.
2752 * - On success, zero.
2753 * - On failure, a negative value.
2755 int rte_eth_dev_callback_register(uint8_t port_id,
2756 enum rte_eth_event_type event,
2757 rte_eth_dev_cb_fn cb_fn, void *cb_arg);
2760 * Unregister a callback function for specific port id.
2767 * User supplied callback function to be called.
2769 * Pointer to the parameters for the registered callback. -1 means to
2770 * remove all for the same callback address and same event.
2773 * - On success, zero.
2774 * - On failure, a negative value.
2776 int rte_eth_dev_callback_unregister(uint8_t port_id,
2777 enum rte_eth_event_type event,
2778 rte_eth_dev_cb_fn cb_fn, void *cb_arg);
2781 * @internal Executes all the user application registered callbacks for
2782 * the specific device. It is for DPDK internal user only. User
2783 * application should not call it directly.
2786 * Pointer to struct rte_eth_dev.
2788 * Eth device interrupt event type.
2793 void _rte_eth_dev_callback_process(struct rte_eth_dev *dev,
2794 enum rte_eth_event_type event);
2797 * Turn on the LED on the Ethernet device.
2798 * This function turns on the LED on the Ethernet device.
2801 * The port identifier of the Ethernet device.
2803 * - (0) if successful.
2804 * - (-ENOTSUP) if underlying hardware OR driver doesn't support
2806 * - (-ENODEV) if *port_id* invalid.
2808 int rte_eth_led_on(uint8_t port_id);
2811 * Turn off the LED on the Ethernet device.
2812 * This function turns off the LED on the Ethernet device.
2815 * The port identifier of the Ethernet device.
2817 * - (0) if successful.
2818 * - (-ENOTSUP) if underlying hardware OR driver doesn't support
2820 * - (-ENODEV) if *port_id* invalid.
2822 int rte_eth_led_off(uint8_t port_id);
2825 * Get current status of the Ethernet link flow control for Ethernet device
2828 * The port identifier of the Ethernet device.
2830 * The pointer to the structure where to store the flow control parameters.
2832 * - (0) if successful.
2833 * - (-ENOTSUP) if hardware doesn't support flow control.
2834 * - (-ENODEV) if *port_id* invalid.
2836 int rte_eth_dev_flow_ctrl_get(uint8_t port_id,
2837 struct rte_eth_fc_conf *fc_conf);
2840 * Configure the Ethernet link flow control for Ethernet device
2843 * The port identifier of the Ethernet device.
2845 * The pointer to the structure of the flow control parameters.
2847 * - (0) if successful.
2848 * - (-ENOTSUP) if hardware doesn't support flow control mode.
2849 * - (-ENODEV) if *port_id* invalid.
2850 * - (-EINVAL) if bad parameter
2851 * - (-EIO) if flow control setup failure
2853 int rte_eth_dev_flow_ctrl_set(uint8_t port_id,
2854 struct rte_eth_fc_conf *fc_conf);
2857 * Configure the Ethernet priority flow control under DCB environment
2858 * for Ethernet device.
2861 * The port identifier of the Ethernet device.
2863 * The pointer to the structure of the priority flow control parameters.
2865 * - (0) if successful.
2866 * - (-ENOTSUP) if hardware doesn't support priority flow control mode.
2867 * - (-ENODEV) if *port_id* invalid.
2868 * - (-EINVAL) if bad parameter
2869 * - (-EIO) if flow control setup failure
2871 int rte_eth_dev_priority_flow_ctrl_set(uint8_t port_id,
2872 struct rte_eth_pfc_conf *pfc_conf);
2875 * Add a MAC address to an internal array of addresses used to enable whitelist
2876 * filtering to accept packets only if the destination MAC address matches.
2879 * The port identifier of the Ethernet device.
2881 * The MAC address to add.
2883 * VMDq pool index to associate address with (if VMDq is enabled). If VMDq is
2884 * not enabled, this should be set to 0.
2886 * - (0) if successfully added or *mac_addr" was already added.
2887 * - (-ENOTSUP) if hardware doesn't support this feature.
2888 * - (-ENODEV) if *port* is invalid.
2889 * - (-ENOSPC) if no more MAC addresses can be added.
2890 * - (-EINVAL) if MAC address is invalid.
2892 int rte_eth_dev_mac_addr_add(uint8_t port, struct ether_addr *mac_addr,
2896 * Remove a MAC address from the internal array of addresses.
2899 * The port identifier of the Ethernet device.
2901 * MAC address to remove.
2903 * - (0) if successful, or *mac_addr* didn't exist.
2904 * - (-ENOTSUP) if hardware doesn't support.
2905 * - (-ENODEV) if *port* invalid.
2906 * - (-EADDRINUSE) if attempting to remove the default MAC address
2908 int rte_eth_dev_mac_addr_remove(uint8_t port, struct ether_addr *mac_addr);
2911 * Update Redirection Table(RETA) of Receive Side Scaling of Ethernet device.
2914 * The port identifier of the Ethernet device.
2918 * - (0) if successful.
2919 * - (-ENOTSUP) if hardware doesn't support.
2920 * - (-EINVAL) if bad parameter.
2922 int rte_eth_dev_rss_reta_update(uint8_t port,
2923 struct rte_eth_rss_reta *reta_conf);
2926 * Query Redirection Table(RETA) of Receive Side Scaling of Ethernet device.
2929 * The port identifier of the Ethernet device.
2933 * - (0) if successful.
2934 * - (-ENOTSUP) if hardware doesn't support.
2935 * - (-EINVAL) if bad parameter.
2937 int rte_eth_dev_rss_reta_query(uint8_t port,
2938 struct rte_eth_rss_reta *reta_conf);
2941 * Updates unicast hash table for receiving packet with the given destination
2942 * MAC address, and the packet is routed to all VFs for which the RX mode is
2943 * accept packets that match the unicast hash table.
2946 * The port identifier of the Ethernet device.
2948 * Unicast MAC address.
2950 * 1 - Set an unicast hash bit for receiving packets with the MAC address.
2951 * 0 - Clear an unicast hash bit.
2953 * - (0) if successful.
2954 * - (-ENOTSUP) if hardware doesn't support.
2955 * - (-ENODEV) if *port_id* invalid.
2956 * - (-EINVAL) if bad parameter.
2958 int rte_eth_dev_uc_hash_table_set(uint8_t port,struct ether_addr *addr,
2962 * Updates all unicast hash bitmaps for receiving packet with any Unicast
2963 * Ethernet MAC addresses,the packet is routed to all VFs for which the RX
2964 * mode is accept packets that match the unicast hash table.
2967 * The port identifier of the Ethernet device.
2969 * 1 - Set all unicast hash bitmaps for receiving all the Ethernet
2971 * 0 - Clear all unicast hash bitmaps
2973 * - (0) if successful.
2974 * - (-ENOTSUP) if hardware doesn't support.
2975 * - (-ENODEV) if *port_id* invalid.
2976 * - (-EINVAL) if bad parameter.
2978 int rte_eth_dev_uc_all_hash_table_set(uint8_t port,uint8_t on);
2981 * Set RX L2 Filtering mode of a VF of an Ethernet device.
2984 * The port identifier of the Ethernet device.
2988 * The RX mode mask, which is one or more of accepting Untagged Packets,
2989 * packets that match the PFUTA table, Broadcast and Multicast Promiscuous.
2990 * ETH_VMDQ_ACCEPT_UNTAG,ETH_VMDQ_ACCEPT_HASH_UC,
2991 * ETH_VMDQ_ACCEPT_BROADCAST and ETH_VMDQ_ACCEPT_MULTICAST will be used
2994 * 1 - Enable a VF RX mode.
2995 * 0 - Disable a VF RX mode.
2997 * - (0) if successful.
2998 * - (-ENOTSUP) if hardware doesn't support.
2999 * - (-ENOTSUP) if hardware doesn't support.
3000 * - (-EINVAL) if bad parameter.
3002 int rte_eth_dev_set_vf_rxmode(uint8_t port, uint16_t vf, uint16_t rx_mode,
3006 * Enable or disable a VF traffic transmit of the Ethernet device.
3009 * The port identifier of the Ethernet device.
3013 * 1 - Enable a VF traffic transmit.
3014 * 0 - Disable a VF traffic transmit.
3016 * - (0) if successful.
3017 * - (-ENODEV) if *port_id* invalid.
3018 * - (-ENOTSUP) if hardware doesn't support.
3019 * - (-EINVAL) if bad parameter.
3022 rte_eth_dev_set_vf_tx(uint8_t port,uint16_t vf, uint8_t on);
3025 * Enable or disable a VF traffic receive of an Ethernet device.
3028 * The port identifier of the Ethernet device.
3032 * 1 - Enable a VF traffic receive.
3033 * 0 - Disable a VF traffic receive.
3035 * - (0) if successful.
3036 * - (-ENOTSUP) if hardware doesn't support.
3037 * - (-ENODEV) if *port_id* invalid.
3038 * - (-EINVAL) if bad parameter.
3041 rte_eth_dev_set_vf_rx(uint8_t port,uint16_t vf, uint8_t on);
3044 * Enable/Disable hardware VF VLAN filtering by an Ethernet device of
3045 * received VLAN packets tagged with a given VLAN Tag Identifier.
3048 * The port identifier of the Ethernet device.
3050 * The VLAN Tag Identifier whose filtering must be enabled or disabled.
3052 * Bitmap listing which VFs participate in the VLAN filtering.
3054 * 1 - Enable VFs VLAN filtering.
3055 * 0 - Disable VFs VLAN filtering.
3057 * - (0) if successful.
3058 * - (-ENOTSUP) if hardware doesn't support.
3059 * - (-ENODEV) if *port_id* invalid.
3060 * - (-EINVAL) if bad parameter.
3063 rte_eth_dev_set_vf_vlan_filter(uint8_t port, uint16_t vlan_id,
3068 * Set a traffic mirroring rule on an Ethernet device
3071 * The port identifier of the Ethernet device.
3072 * @param mirror_conf
3073 * The pointer to the traffic mirroring structure describing the mirroring rule.
3074 * The *rte_eth_vm_mirror_conf* structure includes the type of mirroring rule,
3075 * destination pool and the value of rule if enable vlan or pool mirroring.
3078 * The index of traffic mirroring rule, we support four separated rules.
3080 * 1 - Enable a mirroring rule.
3081 * 0 - Disable a mirroring rule.
3083 * - (0) if successful.
3084 * - (-ENOTSUP) if hardware doesn't support this feature.
3085 * - (-ENODEV) if *port_id* invalid.
3086 * - (-EINVAL) if the mr_conf information is not correct.
3088 int rte_eth_mirror_rule_set(uint8_t port_id,
3089 struct rte_eth_vmdq_mirror_conf *mirror_conf,
3094 * Reset a traffic mirroring rule on an Ethernet device.
3097 * The port identifier of the Ethernet device.
3099 * The index of traffic mirroring rule, we support four separated rules.
3101 * - (0) if successful.
3102 * - (-ENOTSUP) if hardware doesn't support this feature.
3103 * - (-ENODEV) if *port_id* invalid.
3104 * - (-EINVAL) if bad parameter.
3106 int rte_eth_mirror_rule_reset(uint8_t port_id,
3110 * Set the rate limitation for a queue on an Ethernet device.
3113 * The port identifier of the Ethernet device.
3117 * The tx rate allocated from the total link speed for this queue.
3119 * - (0) if successful.
3120 * - (-ENOTSUP) if hardware doesn't support this feature.
3121 * - (-ENODEV) if *port_id* invalid.
3122 * - (-EINVAL) if bad parameter.
3124 int rte_eth_set_queue_rate_limit(uint8_t port_id, uint16_t queue_idx,
3128 * Set the rate limitation for a vf on an Ethernet device.
3131 * The port identifier of the Ethernet device.
3135 * The tx rate allocated from the total link speed for this VF id.
3137 * The queue mask which need to set the rate.
3139 * - (0) if successful.
3140 * - (-ENOTSUP) if hardware doesn't support this feature.
3141 * - (-ENODEV) if *port_id* invalid.
3142 * - (-EINVAL) if bad parameter.
3144 int rte_eth_set_vf_rate_limit(uint8_t port_id, uint16_t vf,
3145 uint16_t tx_rate, uint64_t q_msk);
3148 * Initialize bypass logic. This function needs to be called before
3149 * executing any other bypass API.
3152 * The port identifier of the Ethernet device.
3154 * - (0) if successful.
3155 * - (-ENOTSUP) if hardware doesn't support.
3156 * - (-EINVAL) if bad parameter.
3158 int rte_eth_dev_bypass_init(uint8_t port);
3161 * Return bypass state.
3164 * The port identifier of the Ethernet device.
3166 * The return bypass state.
3169 * - (3) Isolate mode
3171 * - (0) if successful.
3172 * - (-ENOTSUP) if hardware doesn't support.
3173 * - (-EINVAL) if bad parameter.
3175 int rte_eth_dev_bypass_state_show(uint8_t port, uint32_t *state);
3181 * The port identifier of the Ethernet device.
3183 * The current bypass state.
3186 * - (3) Isolate mode
3188 * - (0) if successful.
3189 * - (-ENOTSUP) if hardware doesn't support.
3190 * - (-EINVAL) if bad parameter.
3192 int rte_eth_dev_bypass_state_set(uint8_t port, uint32_t *new_state);
3195 * Return bypass state when given event occurs.
3198 * The port identifier of the Ethernet device.
3201 * - (1) Main power on (power button is pushed)
3202 * - (2) Auxiliary power on (power supply is being plugged)
3203 * - (3) Main power off (system shutdown and power supply is left plugged in)
3204 * - (4) Auxiliary power off (power supply is being unplugged)
3205 * - (5) Display or set the watchdog timer
3207 * The bypass state when given event occurred.
3210 * - (3) Isolate mode
3212 * - (0) if successful.
3213 * - (-ENOTSUP) if hardware doesn't support.
3214 * - (-EINVAL) if bad parameter.
3216 int rte_eth_dev_bypass_event_show(uint8_t port, uint32_t event, uint32_t *state);
3219 * Set bypass state when given event occurs.
3222 * The port identifier of the Ethernet device.
3225 * - (1) Main power on (power button is pushed)
3226 * - (2) Auxiliary power on (power supply is being plugged)
3227 * - (3) Main power off (system shutdown and power supply is left plugged in)
3228 * - (4) Auxiliary power off (power supply is being unplugged)
3229 * - (5) Display or set the watchdog timer
3231 * The assigned state when given event occurs.
3234 * - (3) Isolate mode
3236 * - (0) if successful.
3237 * - (-ENOTSUP) if hardware doesn't support.
3238 * - (-EINVAL) if bad parameter.
3240 int rte_eth_dev_bypass_event_store(uint8_t port, uint32_t event, uint32_t state);
3243 * Set bypass watchdog timeout count.
3246 * The port identifier of the Ethernet device.
3248 * The timeout to be set.
3249 * - (0) 0 seconds (timer is off)
3258 * - (0) if successful.
3259 * - (-ENOTSUP) if hardware doesn't support.
3260 * - (-EINVAL) if bad parameter.
3262 int rte_eth_dev_wd_timeout_store(uint8_t port, uint32_t timeout);
3265 * Get bypass firmware version.
3268 * The port identifier of the Ethernet device.
3270 * The firmware version
3272 * - (0) if successful.
3273 * - (-ENOTSUP) if hardware doesn't support.
3274 * - (-EINVAL) if bad parameter.
3276 int rte_eth_dev_bypass_ver_show(uint8_t port, uint32_t *ver);
3279 * Return bypass watchdog timeout in seconds
3282 * The port identifier of the Ethernet device.
3284 * The return watchdog timeout. "0" represents timer expired
3285 * - (0) 0 seconds (timer is off)
3294 * - (0) if successful.
3295 * - (-ENOTSUP) if hardware doesn't support.
3296 * - (-EINVAL) if bad parameter.
3298 int rte_eth_dev_bypass_wd_timeout_show(uint8_t port, uint32_t *wd_timeout);
3301 * Reset bypass watchdog timer
3304 * The port identifier of the Ethernet device.
3306 * - (0) if successful.
3307 * - (-ENOTSUP) if hardware doesn't support.
3308 * - (-EINVAL) if bad parameter.
3310 int rte_eth_dev_bypass_wd_reset(uint8_t port);
3313 * Configuration of Receive Side Scaling hash computation of Ethernet device.
3316 * The port identifier of the Ethernet device.
3318 * The new configuration to use for RSS hash computation on the port.
3320 * - (0) if successful.
3321 * - (-ENODEV) if port identifier is invalid.
3322 * - (-ENOTSUP) if hardware doesn't support.
3323 * - (-EINVAL) if bad parameter.
3325 int rte_eth_dev_rss_hash_update(uint8_t port_id,
3326 struct rte_eth_rss_conf *rss_conf);
3329 * Retrieve current configuration of Receive Side Scaling hash computation
3330 * of Ethernet device.
3333 * The port identifier of the Ethernet device.
3335 * Where to store the current RSS hash configuration of the Ethernet device.
3337 * - (0) if successful.
3338 * - (-ENODEV) if port identifier is invalid.
3339 * - (-ENOTSUP) if hardware doesn't support RSS.
3342 rte_eth_dev_rss_hash_conf_get(uint8_t port_id,
3343 struct rte_eth_rss_conf *rss_conf);
3349 * The port identifier of the Ethernet device.
3351 * The index of RX queue where to store RX packets matching the syn filter.
3353 * The pointer to the structure describing the syn filter rule.
3355 * - (0) if successful.
3356 * - (-ENOTSUP) if hardware doesn't support.
3357 * - (-EINVAL) if bad parameter.
3359 int rte_eth_dev_add_syn_filter(uint8_t port_id,
3360 struct rte_syn_filter *filter, uint16_t rx_queue);
3366 * The port identifier of the Ethernet device.
3368 * - (0) if successful.
3369 * - (-ENOTSUP) if hardware doesn't support.
3370 * - (-EINVAL) if bad parameter.
3372 int rte_eth_dev_remove_syn_filter(uint8_t port_id);
3378 * The port identifier of the Ethernet device.
3380 * The pointer to the structure describing the syn filter.
3382 * A pointer to get the queue index of syn filter.
3384 * - (0) if successful.
3385 * - (-ENOTSUP) if hardware doesn't support.
3386 * - (-EINVAL) if bad parameter.
3388 int rte_eth_dev_get_syn_filter(uint8_t port_id,
3389 struct rte_syn_filter *filter, uint16_t *rx_queue);
3392 * Add a new ethertype filter rule on an Ethernet device.
3395 * The port identifier of the Ethernet device.
3397 * The identifier of ethertype filter.
3399 * The pointer to the structure describing the ethertype filter rule.
3400 * The *rte_ethertype_filter* structure includes the values of the different
3401 * fields to match: ethertype and priority in vlan tag.
3402 * priority in vlan tag is not supported for E1000 dev.
3404 * The index of the RX queue where to store RX packets matching the added
3407 * - (0) if successful.
3408 * - (-ENOTSUP) if hardware doesn't support ethertype filter.
3409 * - (-ENODEV) if *port_id* invalid.
3410 * - (-EINVAL) if the filter information is not correct.
3412 int rte_eth_dev_add_ethertype_filter(uint8_t port_id, uint16_t index,
3413 struct rte_ethertype_filter *filter, uint16_t rx_queue);
3416 * remove an ethertype filter rule on an Ethernet device.
3419 * The port identifier of the Ethernet device.
3421 * The identifier of ethertype filter.
3423 * - (0) if successful.
3424 * - (-ENOTSUP) if hardware doesn't support ethertype filter.
3425 * - (-ENODEV) if *port_id* invalid.
3426 * - (-EINVAL) if the filter information is not correct.
3428 int rte_eth_dev_remove_ethertype_filter(uint8_t port_id,
3432 * Get an ethertype filter rule on an Ethernet device.
3435 * The port identifier of the Ethernet device.
3437 * The identifier of ethertype filter.
3439 * A pointer to a structure of type *rte_ethertype_filter* to be filled with
3440 * the information of the Ethertype filter.
3442 * A pointer to get the queue index.
3444 * - (0) if successful.
3445 * - (-ENOTSUP) if hardware doesn't support ethertype filter.
3446 * - (-ENODEV) if *port_id* invalid.
3447 * - (-EINVAL) if the filter information is not correct.
3448 * - (-ENOENT) if no enabled filter in this index.
3450 int rte_eth_dev_get_ethertype_filter(uint8_t port_id, uint16_t index,
3451 struct rte_ethertype_filter *filter, uint16_t *rx_queue);
3454 * Add a new 2tuple filter rule on an Ethernet device.
3457 * The port identifier of the Ethernet device.
3459 * The identifier of 2tuple filter.
3461 * The pointer to the structure describing the 2tuple filter rule.
3462 * The *rte_2tuple_filter* structure includes the values of the different
3463 * fields to match: protocol, dst_port and
3464 * tcp_flags if the protocol is tcp type.
3466 * The index of the RX queue where to store RX packets matching the added
3469 * - (0) if successful.
3470 * - (-ENOTSUP) if hardware doesn't support 2tuple filter.
3471 * - (-ENODEV) if *port_id* invalid.
3472 * - (-EINVAL) if the filter information is not correct.
3474 int rte_eth_dev_add_2tuple_filter(uint8_t port_id, uint16_t index,
3475 struct rte_2tuple_filter *filter, uint16_t rx_queue);
3478 * remove a 2tuple filter rule on an Ethernet device.
3481 * The port identifier of the Ethernet device.
3483 * The identifier of 2tuple filter.
3485 * - (0) if successful.
3486 * - (-ENOTSUP) if hardware doesn't support 2tuple filter.
3487 * - (-ENODEV) if *port_id* invalid.
3488 * - (-EINVAL) if the filter information is not correct.
3490 int rte_eth_dev_remove_2tuple_filter(uint8_t port_id, uint16_t index);
3493 * Get an 2tuple filter rule on an Ethernet device.
3496 * The port identifier of the Ethernet device.
3498 * The identifier of 2tuple filter.
3500 * A pointer to a structure of type *rte_2tuple_filter* to be filled with
3501 * the information of the 2tuple filter.
3503 * A pointer to get the queue index.
3505 * - (0) if successful.
3506 * - (-ENOTSUP) if hardware doesn't support 2tuple filter.
3507 * - (-ENODEV) if *port_id* invalid.
3508 * - (-EINVAL) if the filter information is not correct.
3509 * - (-ENOENT) if no enabled filter in this index.
3511 int rte_eth_dev_get_2tuple_filter(uint8_t port_id, uint16_t index,
3512 struct rte_2tuple_filter *filter, uint16_t *rx_queue);
3515 * Add a new 5tuple filter rule on an Ethernet device.
3518 * The port identifier of the Ethernet device.
3520 * The identifier of 5tuple filter.
3522 * The pointer to the structure describing the 5tuple filter rule.
3523 * The *rte_5tuple_filter* structure includes the values of the different
3524 * fields to match: dst src IP, dst src port, protocol and relative masks
3526 * The index of the RX queue where to store RX packets matching the added
3529 * - (0) if successful.
3530 * - (-ENOTSUP) if hardware doesn't support 5tuple filter.
3531 * - (-ENODEV) if *port_id* invalid.
3532 * - (-EINVAL) if the filter information is not correct.
3534 int rte_eth_dev_add_5tuple_filter(uint8_t port_id, uint16_t index,
3535 struct rte_5tuple_filter *filter, uint16_t rx_queue);
3538 * remove a 5tuple filter rule on an Ethernet device.
3541 * The port identifier of the Ethernet device.
3543 * The identifier of 5tuple filter.
3545 * - (0) if successful.
3546 * - (-ENOTSUP) if hardware doesn't support 5tuple filter.
3547 * - (-ENODEV) if *port_id* invalid.
3548 * - (-EINVAL) if the filter information is not correct.
3550 int rte_eth_dev_remove_5tuple_filter(uint8_t port_id, uint16_t index);
3553 * Get an 5tuple filter rule on an Ethernet device.
3556 * The port identifier of the Ethernet device.
3558 * The identifier of 5tuple filter.
3560 * A pointer to a structure of type *rte_5tuple_filter* to be filled with
3561 * the information of the 5tuple filter.
3563 * A pointer to get the queue index.
3565 * - (0) if successful.
3566 * - (-ENOTSUP) if hardware doesn't support 5tuple filter.
3567 * - (-ENODEV) if *port_id* invalid.
3568 * - (-EINVAL) if the filter information is not correct.
3570 int rte_eth_dev_get_5tuple_filter(uint8_t port_id, uint16_t index,
3571 struct rte_5tuple_filter *filter, uint16_t *rx_queue);
3574 * Add a new flex filter rule on an Ethernet device.
3577 * The port identifier of the Ethernet device.
3579 * The identifier of flex filter.
3581 * The pointer to the structure describing the flex filter rule.
3582 * The *rte_flex_filter* structure includes the values of the different fields
3583 * to match: the dwords (first len bytes of packet ) and relative masks.
3585 * The index of the RX queue where to store RX packets matching the added
3588 * - (0) if successful.
3589 * - (-ENOTSUP) if hardware doesn't support flex filter.
3590 * - (-ENODEV) if *port_id* invalid.
3591 * - (-EINVAL) if the filter information is not correct.
3592 * - (-ENOENT) if no enabled filter in this index.
3594 int rte_eth_dev_add_flex_filter(uint8_t port_id, uint16_t index,
3595 struct rte_flex_filter *filter, uint16_t rx_queue);
3598 * remove a flex filter rule on an Ethernet device.
3601 * The port identifier of the Ethernet device.
3603 * The identifier of flex filter.
3605 * - (0) if successful.
3606 * - (-ENOTSUP) if hardware doesn't support flex filter.
3607 * - (-ENODEV) if *port_id* invalid.
3608 * - (-EINVAL) if the filter information is not correct.
3610 int rte_eth_dev_remove_flex_filter(uint8_t port_id, uint16_t index);
3613 * Get an flex filter rule on an Ethernet device.
3616 * The port identifier of the Ethernet device.
3618 * The identifier of flex filter.
3620 * A pointer to a structure of type *rte_flex_filter* to be filled with
3621 * the information of the flex filter.
3623 * A pointer to get the queue index.
3625 * - (0) if successful.
3626 * - (-ENOTSUP) if hardware doesn't support flex filter.
3627 * - (-ENODEV) if *port_id* invalid.
3628 * - (-EINVAL) if the filter information is not correct.
3629 * - (-ENOENT) if no enabled filter in this index.
3631 int rte_eth_dev_get_flex_filter(uint8_t port_id, uint16_t index,
3632 struct rte_flex_filter *filter, uint16_t *rx_queue);
3635 * Check whether the filter type is supported on an Ethernet device.
3636 * All the supported filter types are defined in 'rte_eth_ctrl.h'.
3639 * The port identifier of the Ethernet device.
3640 * @param filter_type
3643 * - (0) if successful.
3644 * - (-ENOTSUP) if hardware doesn't support this filter type.
3645 * - (-ENODEV) if *port_id* invalid.
3647 int rte_eth_dev_filter_supported(uint8_t port_id, enum rte_filter_type filter_type);
3650 * Take operations to assigned filter type on an Ethernet device.
3651 * All the supported operations and filter types are defined in 'rte_eth_ctrl.h'.
3654 * The port identifier of the Ethernet device.
3655 * @param filter_type
3658 * Type of operation.
3660 * A pointer to arguments defined specifically for the operation.
3662 * - (0) if successful.
3663 * - (-ENOTSUP) if hardware doesn't support.
3664 * - (-ENODEV) if *port_id* invalid.
3665 * - others depends on the specific operations implementation.
3667 int rte_eth_dev_filter_ctrl(uint8_t port_id, enum rte_filter_type filter_type,
3668 enum rte_filter_op filter_op, void *arg);
3674 #endif /* _RTE_ETHDEV_H_ */