4 * Copyright(c) 2010-2014 Intel Corporation. All rights reserved.
7 * Redistribution and use in source and binary forms, with or without
8 * modification, are permitted provided that the following conditions
11 * * Redistributions of source code must retain the above copyright
12 * notice, this list of conditions and the following disclaimer.
13 * * Redistributions in binary form must reproduce the above copyright
14 * notice, this list of conditions and the following disclaimer in
15 * the documentation and/or other materials provided with the
17 * * Neither the name of Intel Corporation nor the names of its
18 * contributors may be used to endorse or promote products derived
19 * from this software without specific prior written permission.
21 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
22 * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
23 * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
24 * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
25 * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
26 * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
27 * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
28 * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
29 * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
30 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
31 * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
34 #ifndef _RTE_ETHDEV_H_
35 #define _RTE_ETHDEV_H_
40 * RTE Ethernet Device API
42 * The Ethernet Device API is composed of two parts:
44 * - The application-oriented Ethernet API that includes functions to setup
45 * an Ethernet device (configure it, setup its RX and TX queues and start it),
46 * to get its MAC address, the speed and the status of its physical link,
47 * to receive and to transmit packets, and so on.
49 * - The driver-oriented Ethernet API that exports a function allowing
50 * an Ethernet Poll Mode Driver (PMD) to simultaneously register itself as
51 * an Ethernet device driver and as a PCI driver for a set of matching PCI
52 * [Ethernet] devices classes.
54 * By default, all the functions of the Ethernet Device API exported by a PMD
55 * are lock-free functions which assume to not be invoked in parallel on
56 * different logical cores to work on the same target object. For instance,
57 * the receive function of a PMD cannot be invoked in parallel on two logical
58 * cores to poll the same RX queue [of the same port]. Of course, this function
59 * can be invoked in parallel by different logical cores on different RX queues.
60 * It is the responsibility of the upper level application to enforce this rule.
62 * If needed, parallel accesses by multiple logical cores to shared queues
63 * shall be explicitly protected by dedicated inline lock-aware functions
64 * built on top of their corresponding lock-free functions of the PMD API.
66 * In all functions of the Ethernet API, the Ethernet device is
67 * designated by an integer >= 0 named the device port identifier.
69 * At the Ethernet driver level, Ethernet devices are represented by a generic
70 * data structure of type *rte_eth_dev*.
72 * Ethernet devices are dynamically registered during the PCI probing phase
73 * performed at EAL initialization time.
74 * When an Ethernet device is being probed, an *rte_eth_dev* structure and
75 * a new port identifier are allocated for that device. Then, the eth_dev_init()
76 * function supplied by the Ethernet driver matching the probed PCI
77 * device is invoked to properly initialize the device.
79 * The role of the device init function consists of resetting the hardware,
80 * checking access to Non-volatile Memory (NVM), reading the MAC address
83 * If the device init operation is successful, the correspondence between
84 * the port identifier assigned to the new device and its associated
85 * *rte_eth_dev* structure is effectively registered.
86 * Otherwise, both the *rte_eth_dev* structure and the port identifier are
89 * The functions exported by the application Ethernet API to setup a device
90 * designated by its port identifier must be invoked in the following order:
91 * - rte_eth_dev_configure()
92 * - rte_eth_tx_queue_setup()
93 * - rte_eth_rx_queue_setup()
94 * - rte_eth_dev_start()
96 * Then, the network application can invoke, in any order, the functions
97 * exported by the Ethernet API to get the MAC address of a given device, to
98 * get the speed and the status of a device physical link, to receive/transmit
99 * [burst of] packets, and so on.
101 * If the application wants to change the configuration (i.e. call
102 * rte_eth_dev_configure(), rte_eth_tx_queue_setup(), or
103 * rte_eth_rx_queue_setup()), it must call rte_eth_dev_stop() first to stop the
104 * device and then do the reconfiguration before calling rte_eth_dev_start()
105 * again. The tramsit and receive functions should not be invoked when the
108 * Please note that some configuration is not stored between calls to
109 * rte_eth_dev_stop()/rte_eth_dev_start(). The following configuration will
112 * - flow control settings
113 * - receive mode configuration (promiscuous mode, hardware checksum mode,
114 * RSS/VMDQ settings etc.)
115 * - VLAN filtering configuration
116 * - MAC addresses supplied to MAC address array
117 * - flow director filtering mode (but not filtering rules)
118 * - NIC queue statistics mappings
120 * Any other configuration will not be stored and will need to be re-entered
121 * after a call to rte_eth_dev_start().
123 * Finally, a network application can close an Ethernet device by invoking the
124 * rte_eth_dev_close() function.
126 * Each function of the application Ethernet API invokes a specific function
127 * of the PMD that controls the target device designated by its port
129 * For this purpose, all device-specific functions of an Ethernet driver are
130 * supplied through a set of pointers contained in a generic structure of type
132 * The address of the *eth_dev_ops* structure is stored in the *rte_eth_dev*
133 * structure by the device init function of the Ethernet driver, which is
134 * invoked during the PCI probing phase, as explained earlier.
136 * In other words, each function of the Ethernet API simply retrieves the
137 * *rte_eth_dev* structure associated with the device port identifier and
138 * performs an indirect invocation of the corresponding driver function
139 * supplied in the *eth_dev_ops* structure of the *rte_eth_dev* structure.
141 * For performance reasons, the address of the burst-oriented RX and TX
142 * functions of the Ethernet driver are not contained in the *eth_dev_ops*
143 * structure. Instead, they are directly stored at the beginning of the
144 * *rte_eth_dev* structure to avoid an extra indirect memory access during
147 * RTE ethernet device drivers do not use interrupts for transmitting or
148 * receiving. Instead, Ethernet drivers export Poll-Mode receive and transmit
149 * functions to applications.
150 * Both receive and transmit functions are packet-burst oriented to minimize
151 * their cost per packet through the following optimizations:
153 * - Sharing among multiple packets the incompressible cost of the
154 * invocation of receive/transmit functions.
156 * - Enabling receive/transmit functions to take advantage of burst-oriented
157 * hardware features (L1 cache, prefetch instructions, NIC head/tail
158 * registers) to minimize the number of CPU cycles per packet, for instance,
159 * by avoiding useless read memory accesses to ring descriptors, or by
160 * systematically using arrays of pointers that exactly fit L1 cache line
161 * boundaries and sizes.
163 * The burst-oriented receive function does not provide any error notification,
164 * to avoid the corresponding overhead. As a hint, the upper-level application
165 * might check the status of the device link once being systematically returned
166 * a 0 value by the receive function of the driver for a given number of tries.
176 #include <rte_interrupts.h>
178 #include <rte_mbuf.h>
179 #include "rte_ether.h"
182 * A structure used to retrieve statistics for an Ethernet port.
184 struct rte_eth_stats {
185 uint64_t ipackets; /**< Total number of successfully received packets. */
186 uint64_t opackets; /**< Total number of successfully transmitted packets.*/
187 uint64_t ibytes; /**< Total number of successfully received bytes. */
188 uint64_t obytes; /**< Total number of successfully transmitted bytes. */
189 uint64_t ierrors; /**< Total number of erroneous received packets. */
190 uint64_t oerrors; /**< Total number of failed transmitted packets. */
191 uint64_t imcasts; /**< Total number of multicast received packets. */
192 uint64_t rx_nombuf; /**< Total number of RX mbuf allocation failures. */
193 uint64_t fdirmatch; /**< Total number of RX packets matching a filter. */
194 uint64_t fdirmiss; /**< Total number of RX packets not matching any filter. */
195 uint64_t q_ipackets[RTE_ETHDEV_QUEUE_STAT_CNTRS];
196 /**< Total number of queue RX packets. */
197 uint64_t q_opackets[RTE_ETHDEV_QUEUE_STAT_CNTRS];
198 /**< Total number of queue TX packets. */
199 uint64_t q_ibytes[RTE_ETHDEV_QUEUE_STAT_CNTRS];
200 /**< Total number of successfully received queue bytes. */
201 uint64_t q_obytes[RTE_ETHDEV_QUEUE_STAT_CNTRS];
202 /**< Total number of successfully transmitted queue bytes. */
203 uint64_t q_errors[RTE_ETHDEV_QUEUE_STAT_CNTRS];
204 /**< Total number of queue packets received that are dropped. */
206 /**< Total number of good packets received from loopback,VF Only */
208 /**< Total number of good packets transmitted to loopback,VF Only */
210 /**< Total number of good bytes received from loopback,VF Only */
212 /**< Total number of good bytes transmitted to loopback,VF Only */
216 * A structure used to retrieve link-level information of an Ethernet port.
218 struct rte_eth_link {
219 uint16_t link_speed; /**< ETH_LINK_SPEED_[10, 100, 1000, 10000] */
220 uint16_t link_duplex; /**< ETH_LINK_[HALF_DUPLEX, FULL_DUPLEX] */
221 uint8_t link_status : 1; /**< 1 -> link up, 0 -> link down */
222 }__attribute__((aligned(8))); /**< aligned for atomic64 read/write */
224 #define ETH_LINK_SPEED_AUTONEG 0 /**< Auto-negotiate link speed. */
225 #define ETH_LINK_SPEED_10 10 /**< 10 megabits/second. */
226 #define ETH_LINK_SPEED_100 100 /**< 100 megabits/second. */
227 #define ETH_LINK_SPEED_1000 1000 /**< 1 gigabits/second. */
228 #define ETH_LINK_SPEED_10000 10000 /**< 10 gigabits/second. */
230 #define ETH_LINK_AUTONEG_DUPLEX 0 /**< Auto-negotiate duplex. */
231 #define ETH_LINK_HALF_DUPLEX 1 /**< Half-duplex connection. */
232 #define ETH_LINK_FULL_DUPLEX 2 /**< Full-duplex connection. */
235 * A structure used to configure the ring threshold registers of an RX/TX
236 * queue for an Ethernet port.
238 struct rte_eth_thresh {
239 uint8_t pthresh; /**< Ring prefetch threshold. */
240 uint8_t hthresh; /**< Ring host threshold. */
241 uint8_t wthresh; /**< Ring writeback threshold. */
245 * A set of values to identify what method is to be used to route
246 * packets to multiple queues.
248 enum rte_eth_rx_mq_mode {
249 ETH_MQ_RX_NONE = 0, /**< None of DCB,RSS or VMDQ mode */
251 ETH_MQ_RX_RSS, /**< For RX side, only RSS is on */
252 ETH_MQ_RX_DCB, /**< For RX side,only DCB is on. */
253 ETH_MQ_RX_DCB_RSS, /**< Both DCB and RSS enable */
255 ETH_MQ_RX_VMDQ_ONLY, /**< Only VMDQ, no RSS nor DCB */
256 ETH_MQ_RX_VMDQ_RSS, /**< RSS mode with VMDQ */
257 ETH_MQ_RX_VMDQ_DCB, /**< Use VMDQ+DCB to route traffic to queues */
258 ETH_MQ_RX_VMDQ_DCB_RSS, /**< Enable both VMDQ and DCB in VMDq */
262 * for rx mq mode backward compatible
264 #define ETH_RSS ETH_MQ_RX_RSS
265 #define VMDQ_DCB ETH_MQ_RX_VMDQ_DCB
266 #define ETH_DCB_RX ETH_MQ_RX_DCB
269 * A set of values to identify what method is to be used to transmit
270 * packets using multi-TCs.
272 enum rte_eth_tx_mq_mode {
273 ETH_MQ_TX_NONE = 0, /**< It is in neither DCB nor VT mode. */
274 ETH_MQ_TX_DCB, /**< For TX side,only DCB is on. */
275 ETH_MQ_TX_VMDQ_DCB, /**< For TX side,both DCB and VT is on. */
276 ETH_MQ_TX_VMDQ_ONLY, /**< Only VT on, no DCB */
280 * for tx mq mode backward compatible
282 #define ETH_DCB_NONE ETH_MQ_TX_NONE
283 #define ETH_VMDQ_DCB_TX ETH_MQ_TX_VMDQ_DCB
284 #define ETH_DCB_TX ETH_MQ_TX_DCB
287 * A structure used to configure the RX features of an Ethernet port.
289 struct rte_eth_rxmode {
290 /** The multi-queue packet distribution mode to be used, e.g. RSS. */
291 enum rte_eth_rx_mq_mode mq_mode;
292 uint32_t max_rx_pkt_len; /**< Only used if jumbo_frame enabled. */
293 uint16_t split_hdr_size; /**< hdr buf size (header_split enabled).*/
294 uint8_t header_split : 1, /**< Header Split enable. */
295 hw_ip_checksum : 1, /**< IP/UDP/TCP checksum offload enable. */
296 hw_vlan_filter : 1, /**< VLAN filter enable. */
297 hw_vlan_strip : 1, /**< VLAN strip enable. */
298 hw_vlan_extend : 1, /**< Extended VLAN enable. */
299 jumbo_frame : 1, /**< Jumbo Frame Receipt enable. */
300 hw_strip_crc : 1; /**< Enable CRC stripping by hardware. */
304 * A structure used to configure the Receive Side Scaling (RSS) feature
305 * of an Ethernet port.
306 * If not NULL, the *rss_key* pointer of the *rss_conf* structure points
307 * to an array of 40 bytes holding the RSS key to use for hashing specific
308 * header fields of received packets.
309 * Otherwise, a default random hash key is used by the device driver.
311 * The *rss_hf* field of the *rss_conf* structure indicates the different
312 * types of IPv4/IPv6 packets to which the RSS hashing must be applied.
313 * Supplying an *rss_hf* equal to zero disables the RSS feature.
315 struct rte_eth_rss_conf {
316 uint8_t *rss_key; /**< If not NULL, 40-byte hash key. */
317 uint16_t rss_hf; /**< Hash functions to apply - see below. */
320 #define ETH_RSS_IPV4 0x0001 /**< IPv4 packet. */
321 #define ETH_RSS_IPV4_TCP 0x0002 /**< IPv4/TCP packet. */
322 #define ETH_RSS_IPV6 0x0004 /**< IPv6 packet. */
323 #define ETH_RSS_IPV6_EX 0x0008 /**< IPv6 packet with extension headers.*/
324 #define ETH_RSS_IPV6_TCP 0x0010 /**< IPv6/TCP packet. */
325 #define ETH_RSS_IPV6_TCP_EX 0x0020 /**< IPv6/TCP with extension headers. */
326 /* Intel RSS extensions to UDP packets */
327 #define ETH_RSS_IPV4_UDP 0x0040 /**< IPv4/UDP packet. */
328 #define ETH_RSS_IPV6_UDP 0x0080 /**< IPv6/UDP packet. */
329 #define ETH_RSS_IPV6_UDP_EX 0x0100 /**< IPv6/UDP with extension headers. */
330 /* Definitions used for redirection table entry size */
331 #define ETH_RSS_RETA_NUM_ENTRIES 128
332 #define ETH_RSS_RETA_MAX_QUEUE 16
334 /* Definitions used for VMDQ and DCB functionality */
335 #define ETH_VMDQ_MAX_VLAN_FILTERS 64 /**< Maximum nb. of VMDQ vlan filters. */
336 #define ETH_DCB_NUM_USER_PRIORITIES 8 /**< Maximum nb. of DCB priorities. */
337 #define ETH_VMDQ_DCB_NUM_QUEUES 128 /**< Maximum nb. of VMDQ DCB queues. */
338 #define ETH_DCB_NUM_QUEUES 128 /**< Maximum nb. of DCB queues. */
340 /* DCB capability defines */
341 #define ETH_DCB_PG_SUPPORT 0x00000001 /**< Priority Group(ETS) support. */
342 #define ETH_DCB_PFC_SUPPORT 0x00000002 /**< Priority Flow Control support. */
344 /* Definitions used for VLAN Offload functionality */
345 #define ETH_VLAN_STRIP_OFFLOAD 0x0001 /**< VLAN Strip On/Off */
346 #define ETH_VLAN_FILTER_OFFLOAD 0x0002 /**< VLAN Filter On/Off */
347 #define ETH_VLAN_EXTEND_OFFLOAD 0x0004 /**< VLAN Extend On/Off */
349 /* Definitions used for mask VLAN setting */
350 #define ETH_VLAN_STRIP_MASK 0x0001 /**< VLAN Strip setting mask */
351 #define ETH_VLAN_FILTER_MASK 0x0002 /**< VLAN Filter setting mask*/
352 #define ETH_VLAN_EXTEND_MASK 0x0004 /**< VLAN Extend setting mask*/
353 #define ETH_VLAN_ID_MAX 0x0FFF /**< VLAN ID is in lower 12 bits*/
355 /* Definitions used for receive MAC address */
356 #define ETH_NUM_RECEIVE_MAC_ADDR 128 /**< Maximum nb. of receive mac addr. */
359 /* Definitions used for unicast hash */
360 #define ETH_VMDQ_NUM_UC_HASH_ARRAY 128 /**< Maximum nb. of UC hash array. */
362 /* Definitions used for VMDQ pool rx mode setting */
363 #define ETH_VMDQ_ACCEPT_UNTAG 0x0001 /**< accept untagged packets. */
364 #define ETH_VMDQ_ACCEPT_HASH_MC 0x0002 /**< accept packets in multicast table . */
365 #define ETH_VMDQ_ACCEPT_HASH_UC 0x0004 /**< accept packets in unicast table. */
366 #define ETH_VMDQ_ACCEPT_BROADCAST 0x0008 /**< accept broadcast packets. */
367 #define ETH_VMDQ_ACCEPT_MULTICAST 0x0010 /**< multicast promiscuous. */
369 /* Definitions used for VMDQ mirror rules setting */
370 #define ETH_VMDQ_NUM_MIRROR_RULE 4 /**< Maximum nb. of mirror rules. . */
372 #define ETH_VMDQ_POOL_MIRROR 0x0001 /**< Virtual Pool Mirroring. */
373 #define ETH_VMDQ_UPLINK_MIRROR 0x0002 /**< Uplink Port Mirroring. */
374 #define ETH_VMDQ_DOWNLIN_MIRROR 0x0004 /**< Downlink Port Mirroring. */
375 #define ETH_VMDQ_VLAN_MIRROR 0x0008 /**< VLAN Mirroring. */
378 * A structure used to configure VLAN traffic mirror of an Ethernet port.
380 struct rte_eth_vlan_mirror {
381 uint64_t vlan_mask; /**< mask for valid VLAN ID. */
382 uint16_t vlan_id[ETH_VMDQ_MAX_VLAN_FILTERS];
383 /** VLAN ID list for vlan mirror. */
387 * A structure used to configure traffic mirror of an Ethernet port.
389 struct rte_eth_vmdq_mirror_conf {
390 uint8_t rule_type_mask; /**< Mirroring rule type mask we want to set */
391 uint8_t dst_pool; /**< Destination pool for this mirror rule. */
392 uint64_t pool_mask; /**< Bitmap of pool for pool mirroring */
393 struct rte_eth_vlan_mirror vlan; /**< VLAN ID setting for VLAN mirroring */
397 * A structure used to configure Redirection Table of the Receive Side
398 * Scaling (RSS) feature of an Ethernet port.
400 struct rte_eth_rss_reta {
401 /** First 64 mask bits indicate which entry(s) need to updated/queried. */
403 /** Second 64 mask bits indicate which entry(s) need to updated/queried. */
405 uint8_t reta[ETH_RSS_RETA_NUM_ENTRIES]; /**< 128 RETA entries*/
409 * This enum indicates the possible number of traffic classes
410 * in DCB configratioins
412 enum rte_eth_nb_tcs {
413 ETH_4_TCS = 4, /**< 4 TCs with DCB. */
414 ETH_8_TCS = 8 /**< 8 TCs with DCB. */
418 * This enum indicates the possible number of queue pools
419 * in VMDQ configurations.
421 enum rte_eth_nb_pools {
422 ETH_8_POOLS = 8, /**< 8 VMDq pools. */
423 ETH_16_POOLS = 16, /**< 16 VMDq pools. */
424 ETH_32_POOLS = 32, /**< 32 VMDq pools. */
425 ETH_64_POOLS = 64 /**< 64 VMDq pools. */
428 /* This structure may be extended in future. */
429 struct rte_eth_dcb_rx_conf {
430 enum rte_eth_nb_tcs nb_tcs; /**< Possible DCB TCs, 4 or 8 TCs */
431 uint8_t dcb_queue[ETH_DCB_NUM_USER_PRIORITIES];
432 /**< Possible DCB queue,4 or 8. */
435 struct rte_eth_vmdq_dcb_tx_conf {
436 enum rte_eth_nb_pools nb_queue_pools; /**< With DCB, 16 or 32 pools. */
437 uint8_t dcb_queue[ETH_DCB_NUM_USER_PRIORITIES];
438 /**< Possible DCB queue,4 or 8. */
441 struct rte_eth_dcb_tx_conf {
442 enum rte_eth_nb_tcs nb_tcs; /**< Possible DCB TCs, 4 or 8 TCs. */
443 uint8_t dcb_queue[ETH_DCB_NUM_USER_PRIORITIES];
444 /**< Possible DCB queue,4 or 8. */
447 struct rte_eth_vmdq_tx_conf {
448 enum rte_eth_nb_pools nb_queue_pools; /**< VMDq mode, 64 pools. */
452 * A structure used to configure the VMDQ+DCB feature
453 * of an Ethernet port.
455 * Using this feature, packets are routed to a pool of queues, based
456 * on the vlan id in the vlan tag, and then to a specific queue within
457 * that pool, using the user priority vlan tag field.
459 * A default pool may be used, if desired, to route all traffic which
460 * does not match the vlan filter rules.
462 struct rte_eth_vmdq_dcb_conf {
463 enum rte_eth_nb_pools nb_queue_pools; /**< With DCB, 16 or 32 pools */
464 uint8_t enable_default_pool; /**< If non-zero, use a default pool */
465 uint8_t default_pool; /**< The default pool, if applicable */
466 uint8_t nb_pool_maps; /**< We can have up to 64 filters/mappings */
468 uint16_t vlan_id; /**< The vlan id of the received frame */
469 uint64_t pools; /**< Bitmask of pools for packet rx */
470 } pool_map[ETH_VMDQ_MAX_VLAN_FILTERS]; /**< VMDq vlan pool maps. */
471 uint8_t dcb_queue[ETH_DCB_NUM_USER_PRIORITIES];
472 /**< Selects a queue in a pool */
475 struct rte_eth_vmdq_rx_conf {
476 enum rte_eth_nb_pools nb_queue_pools; /**< VMDq only mode, 8 or 64 pools */
477 uint8_t enable_default_pool; /**< If non-zero, use a default pool */
478 uint8_t default_pool; /**< The default pool, if applicable */
479 uint8_t nb_pool_maps; /**< We can have up to 64 filters/mappings */
481 uint16_t vlan_id; /**< The vlan id of the received frame */
482 uint64_t pools; /**< Bitmask of pools for packet rx */
483 } pool_map[ETH_VMDQ_MAX_VLAN_FILTERS]; /**< VMDq vlan pool maps. */
487 * A structure used to configure the TX features of an Ethernet port.
489 struct rte_eth_txmode {
490 enum rte_eth_tx_mq_mode mq_mode; /**< TX multi-queues mode. */
494 * A structure used to configure an RX ring of an Ethernet port.
496 struct rte_eth_rxconf {
497 struct rte_eth_thresh rx_thresh; /**< RX ring threshold registers. */
498 uint16_t rx_free_thresh; /**< Drives the freeing of RX descriptors. */
499 uint8_t rx_drop_en; /**< Drop packets if no descriptors are available. */
502 #define ETH_TXQ_FLAGS_NOMULTSEGS 0x0001 /**< nb_segs=1 for all mbufs */
503 #define ETH_TXQ_FLAGS_NOREFCOUNT 0x0002 /**< refcnt can be ignored */
504 #define ETH_TXQ_FLAGS_NOMULTMEMP 0x0004 /**< all bufs come from same mempool */
505 #define ETH_TXQ_FLAGS_NOVLANOFFL 0x0100 /**< disable VLAN offload */
506 #define ETH_TXQ_FLAGS_NOXSUMSCTP 0x0200 /**< disable SCTP checksum offload */
507 #define ETH_TXQ_FLAGS_NOXSUMUDP 0x0400 /**< disable UDP checksum offload */
508 #define ETH_TXQ_FLAGS_NOXSUMTCP 0x0800 /**< disable TCP checksum offload */
509 #define ETH_TXQ_FLAGS_NOOFFLOADS \
510 (ETH_TXQ_FLAGS_NOVLANOFFL | ETH_TXQ_FLAGS_NOXSUMSCTP | \
511 ETH_TXQ_FLAGS_NOXSUMUDP | ETH_TXQ_FLAGS_NOXSUMTCP)
513 * A structure used to configure a TX ring of an Ethernet port.
515 struct rte_eth_txconf {
516 struct rte_eth_thresh tx_thresh; /**< TX ring threshold registers. */
517 uint16_t tx_rs_thresh; /**< Drives the setting of RS bit on TXDs. */
518 uint16_t tx_free_thresh; /**< Drives the freeing of TX buffers. */
519 uint32_t txq_flags; /**< Set flags for the Tx queue */
523 * This enum indicates the flow control mode
525 enum rte_eth_fc_mode {
526 RTE_FC_NONE = 0, /**< Disable flow control. */
527 RTE_FC_RX_PAUSE, /**< RX pause frame, enable flowctrl on TX side. */
528 RTE_FC_TX_PAUSE, /**< TX pause frame, enable flowctrl on RX side. */
529 RTE_FC_FULL /**< Enable flow control on both side. */
533 * A structure used to configure Ethernet flow control parameter.
534 * These parameters will be configured into the register of the NIC.
535 * Please refer to the corresponding data sheet for proper value.
537 struct rte_eth_fc_conf {
538 uint32_t high_water; /**< High threshold value to trigger XOFF */
539 uint32_t low_water; /**< Low threshold value to trigger XON */
540 uint16_t pause_time; /**< Pause quota in the Pause frame */
541 uint16_t send_xon; /**< Is XON frame need be sent */
542 enum rte_eth_fc_mode mode; /**< Link flow control mode */
543 uint8_t mac_ctrl_frame_fwd; /**< Forward MAC control frames */
547 * A structure used to configure Ethernet priority flow control parameter.
548 * These parameters will be configured into the register of the NIC.
549 * Please refer to the corresponding data sheet for proper value.
551 struct rte_eth_pfc_conf {
552 struct rte_eth_fc_conf fc; /**< General flow control parameter. */
553 uint8_t priority; /**< VLAN User Priority. */
557 * Flow Director setting modes: none (default), signature or perfect.
560 RTE_FDIR_MODE_NONE = 0, /**< Disable FDIR support. */
561 RTE_FDIR_MODE_SIGNATURE, /**< Enable FDIR signature filter mode. */
562 RTE_FDIR_MODE_PERFECT, /**< Enable FDIR perfect filter mode. */
566 * Memory space that can be configured to store Flow Director filters
567 * in the board memory.
569 enum rte_fdir_pballoc_type {
570 RTE_FDIR_PBALLOC_64K = 0, /**< 64k. */
571 RTE_FDIR_PBALLOC_128K, /**< 128k. */
572 RTE_FDIR_PBALLOC_256K, /**< 256k. */
576 * Select report mode of FDIR hash information in RX descriptors.
578 enum rte_fdir_status_mode {
579 RTE_FDIR_NO_REPORT_STATUS = 0, /**< Never report FDIR hash. */
580 RTE_FDIR_REPORT_STATUS, /**< Only report FDIR hash for matching pkts. */
581 RTE_FDIR_REPORT_STATUS_ALWAYS, /**< Always report FDIR hash. */
585 * A structure used to configure the Flow Director (FDIR) feature
586 * of an Ethernet port.
588 * If mode is RTE_FDIR_DISABLE, the pballoc value is ignored.
590 struct rte_fdir_conf {
591 enum rte_fdir_mode mode; /**< Flow Director mode. */
592 enum rte_fdir_pballoc_type pballoc; /**< Space for FDIR filters. */
593 enum rte_fdir_status_mode status; /**< How to report FDIR hash. */
594 /** Offset of flexbytes field in RX packets (in 16-bit word units). */
595 uint8_t flexbytes_offset;
596 /** RX queue of packets matching a "drop" filter in perfect mode. */
601 * Possible l4type of FDIR filters.
604 RTE_FDIR_L4TYPE_NONE = 0, /**< None. */
605 RTE_FDIR_L4TYPE_UDP, /**< UDP. */
606 RTE_FDIR_L4TYPE_TCP, /**< TCP. */
607 RTE_FDIR_L4TYPE_SCTP, /**< SCTP. */
611 * Select IPv4 or IPv6 FDIR filters.
614 RTE_FDIR_IPTYPE_IPV4 = 0, /**< IPv4. */
615 RTE_FDIR_IPTYPE_IPV6 , /**< IPv6. */
619 * A structure used to define a FDIR packet filter.
621 struct rte_fdir_filter {
622 uint16_t flex_bytes; /**< Flex bytes value to match. */
623 uint16_t vlan_id; /**< VLAN ID value to match, 0 otherwise. */
624 uint16_t port_src; /**< Source port to match, 0 otherwise. */
625 uint16_t port_dst; /**< Destination port to match, 0 otherwise. */
627 uint32_t ipv4_addr; /**< IPv4 source address to match. */
628 uint32_t ipv6_addr[4]; /**< IPv6 source address to match. */
629 } ip_src; /**< IPv4/IPv6 source address to match (union of above). */
631 uint32_t ipv4_addr; /**< IPv4 destination address to match. */
632 uint32_t ipv6_addr[4]; /**< IPv6 destination address to match */
633 } ip_dst; /**< IPv4/IPv6 destination address to match (union of above). */
634 enum rte_l4type l4type; /**< l4type to match: NONE/UDP/TCP/SCTP. */
635 enum rte_iptype iptype; /**< IP packet type to match: IPv4 or IPv6. */
639 * A structure used to configure FDIR masks that are used by the device
640 * to match the various fields of RX packet headers.
641 * @note The only_ip_flow field has the opposite meaning compared to other
644 struct rte_fdir_masks {
645 /** When set to 1, packet l4type is \b NOT relevant in filters, and
646 source and destination port masks must be set to zero. */
647 uint8_t only_ip_flow;
648 /** If set to 1, vlan_id is relevant in filters. */
650 /** If set to 1, vlan_prio is relevant in filters. */
652 /** If set to 1, flexbytes is relevant in filters. */
654 /** If set to 1, set the IPv6 masks. Otherwise set the IPv4 masks. */
655 uint8_t set_ipv6_mask;
656 /** When set to 1, comparison of destination IPv6 address with IP6AT
657 registers is meaningful. */
658 uint8_t comp_ipv6_dst;
659 /** Mask of Destination IPv4 Address. All bits set to 1 define the
660 relevant bits to use in the destination address of an IPv4 packet
661 when matching it against FDIR filters. */
662 uint32_t dst_ipv4_mask;
663 /** Mask of Source IPv4 Address. All bits set to 1 define
664 the relevant bits to use in the source address of an IPv4 packet
665 when matching it against FDIR filters. */
666 uint32_t src_ipv4_mask;
667 /** Mask of Source IPv6 Address. All bits set to 1 define the
668 relevant BYTES to use in the source address of an IPv6 packet
669 when matching it against FDIR filters. */
670 uint16_t dst_ipv6_mask;
671 /** Mask of Destination IPv6 Address. All bits set to 1 define the
672 relevant BYTES to use in the destination address of an IPv6 packet
673 when matching it against FDIR filters. */
674 uint16_t src_ipv6_mask;
675 /** Mask of Source Port. All bits set to 1 define the relevant
676 bits to use in the source port of an IP packets when matching it
677 against FDIR filters. */
678 uint16_t src_port_mask;
679 /** Mask of Destination Port. All bits set to 1 define the relevant
680 bits to use in the destination port of an IP packet when matching it
681 against FDIR filters. */
682 uint16_t dst_port_mask;
686 * A structure used to report the status of the flow director filters in use.
688 struct rte_eth_fdir {
689 /** Number of filters with collision indication. */
691 /** Number of free (non programmed) filters. */
693 /** The Lookup hash value of the added filter that updated the value
694 of the MAXLEN field */
696 /** Longest linked list of filters in the table. */
698 /** Number of added filters. */
700 /** Number of removed filters. */
702 /** Number of failed added filters (no more space in device). */
704 /** Number of failed removed filters. */
709 * A structure used to enable/disable specific device interrupts.
711 struct rte_intr_conf {
712 /** enable/disable lsc interrupt. 0 (default) - disable, 1 enable */
717 * A structure used to configure an Ethernet port.
718 * Depending upon the RX multi-queue mode, extra advanced
719 * configuration settings may be needed.
721 struct rte_eth_conf {
723 /**< ETH_LINK_SPEED_10[0|00|000], or 0 for autonegotation */
724 uint16_t link_duplex;
725 /**< ETH_LINK_[HALF_DUPLEX|FULL_DUPLEX], or 0 for autonegotation */
726 struct rte_eth_rxmode rxmode; /**< Port RX configuration. */
727 struct rte_eth_txmode txmode; /**< Port TX configuration. */
728 uint32_t lpbk_mode; /**< Loopback operation mode. By default the value
729 is 0, meaning the loopback mode is disabled.
730 Read the datasheet of given ethernet controller
731 for details. The possible values of this field
732 are defined in implementation of each driver. */
734 struct rte_eth_rss_conf rss_conf; /**< Port RSS configuration */
735 struct rte_eth_vmdq_dcb_conf vmdq_dcb_conf;
736 /**< Port vmdq+dcb configuration. */
737 struct rte_eth_dcb_rx_conf dcb_rx_conf;
738 /**< Port dcb RX configuration. */
739 struct rte_eth_vmdq_rx_conf vmdq_rx_conf;
740 /**< Port vmdq RX configuration. */
741 } rx_adv_conf; /**< Port RX filtering configuration (union). */
743 struct rte_eth_vmdq_dcb_tx_conf vmdq_dcb_tx_conf;
744 /**< Port vmdq+dcb TX configuration. */
745 struct rte_eth_dcb_tx_conf dcb_tx_conf;
746 /**< Port dcb TX configuration. */
747 struct rte_eth_vmdq_tx_conf vmdq_tx_conf;
748 /**< Port vmdq TX configuration. */
749 } tx_adv_conf; /**< Port TX DCB configuration (union). */
750 /** Currently,Priority Flow Control(PFC) are supported,if DCB with PFC
751 is needed,and the variable must be set ETH_DCB_PFC_SUPPORT. */
752 uint32_t dcb_capability_en;
753 struct rte_fdir_conf fdir_conf; /**< FDIR configuration. */
754 struct rte_intr_conf intr_conf; /**< Interrupt mode configuration. */
758 * A structure used to retrieve the contextual information of
759 * an Ethernet device, such as the controlling driver of the device,
760 * its PCI context, etc...
764 * RX offload capabilities of a device.
766 #define DEV_RX_OFFLOAD_VLAN_STRIP 0x00000001
767 #define DEV_RX_OFFLOAD_IPV4_CKSUM 0x00000002
768 #define DEV_RX_OFFLOAD_UDP_CKSUM 0x00000004
769 #define DEV_RX_OFFLOAD_TCP_CKSUM 0x00000008
770 #define DEV_RX_OFFLOAD_TCP_LRO 0x00000010
773 * TX offload capabilities of a device.
775 #define DEV_TX_OFFLOAD_VLAN_INSERT 0x00000001
776 #define DEV_TX_OFFLOAD_IPV4_CKSUM 0x00000002
777 #define DEV_TX_OFFLOAD_UDP_CKSUM 0x00000004
778 #define DEV_TX_OFFLOAD_TCP_CKSUM 0x00000008
779 #define DEV_TX_OFFLOAD_SCTP_CKSUM 0x00000010
780 #define DEV_TX_OFFLOAD_TCP_TSO 0x00000020
781 #define DEV_TX_OFFLOAD_UDP_TSO 0x00000040
783 struct rte_eth_dev_info {
784 struct rte_pci_device *pci_dev; /**< Device PCI information. */
785 const char *driver_name; /**< Device Driver name. */
786 uint32_t min_rx_bufsize; /**< Minimum size of RX buffer. */
787 uint32_t max_rx_pktlen; /**< Maximum configurable length of RX pkt. */
788 uint16_t max_rx_queues; /**< Maximum number of RX queues. */
789 uint16_t max_tx_queues; /**< Maximum number of TX queues. */
790 uint32_t max_mac_addrs; /**< Maximum number of MAC addresses. */
791 uint32_t max_hash_mac_addrs;
792 /** Maximum number of hash MAC addresses for MTA and UTA. */
793 uint16_t max_vfs; /**< Maximum number of VFs. */
794 uint16_t max_vmdq_pools; /**< Maximum number of VMDq pools. */
795 uint32_t rx_offload_capa; /**< Device RX offload capabilities. */
796 uint32_t tx_offload_capa; /**< Device TX offload capabilities. */
801 struct rte_eth_dev_callback;
802 /** @internal Structure to keep track of registered callbacks */
803 TAILQ_HEAD(rte_eth_dev_cb_list, rte_eth_dev_callback);
806 * Definitions of all functions exported by an Ethernet driver through the
807 * the generic structure of type *eth_dev_ops* supplied in the *rte_eth_dev*
808 * structure associated with an Ethernet device.
811 typedef int (*eth_dev_configure_t)(struct rte_eth_dev *dev);
812 /**< @internal Ethernet device configuration. */
814 typedef int (*eth_dev_start_t)(struct rte_eth_dev *dev);
815 /**< @internal Function used to start a configured Ethernet device. */
817 typedef void (*eth_dev_stop_t)(struct rte_eth_dev *dev);
818 /**< @internal Function used to stop a configured Ethernet device. */
820 typedef void (*eth_dev_close_t)(struct rte_eth_dev *dev);
821 /**< @internal Function used to close a configured Ethernet device. */
823 typedef void (*eth_promiscuous_enable_t)(struct rte_eth_dev *dev);
824 /**< @internal Function used to enable the RX promiscuous mode of an Ethernet device. */
826 typedef void (*eth_promiscuous_disable_t)(struct rte_eth_dev *dev);
827 /**< @internal Function used to disable the RX promiscuous mode of an Ethernet device. */
829 typedef void (*eth_allmulticast_enable_t)(struct rte_eth_dev *dev);
830 /**< @internal Enable the receipt of all multicast packets by an Ethernet device. */
832 typedef void (*eth_allmulticast_disable_t)(struct rte_eth_dev *dev);
833 /**< @internal Disable the receipt of all multicast packets by an Ethernet device. */
835 typedef int (*eth_link_update_t)(struct rte_eth_dev *dev,
836 int wait_to_complete);
837 /**< @internal Get link speed, duplex mode and state (up/down) of an Ethernet device. */
839 typedef void (*eth_stats_get_t)(struct rte_eth_dev *dev,
840 struct rte_eth_stats *igb_stats);
841 /**< @internal Get global I/O statistics of an Ethernet device. */
843 typedef void (*eth_stats_reset_t)(struct rte_eth_dev *dev);
844 /**< @internal Reset global I/O statistics of an Ethernet device to 0. */
846 typedef int (*eth_queue_stats_mapping_set_t)(struct rte_eth_dev *dev,
850 /**< @internal Set a queue statistics mapping for a tx/rx queue of an Ethernet device. */
852 typedef void (*eth_dev_infos_get_t)(struct rte_eth_dev *dev,
853 struct rte_eth_dev_info *dev_info);
854 /**< @internal Get specific informations of an Ethernet device. */
856 typedef int (*eth_rx_queue_setup_t)(struct rte_eth_dev *dev,
857 uint16_t rx_queue_id,
859 unsigned int socket_id,
860 const struct rte_eth_rxconf *rx_conf,
861 struct rte_mempool *mb_pool);
862 /**< @internal Set up a receive queue of an Ethernet device. */
864 typedef int (*eth_tx_queue_setup_t)(struct rte_eth_dev *dev,
865 uint16_t tx_queue_id,
867 unsigned int socket_id,
868 const struct rte_eth_txconf *tx_conf);
869 /**< @internal Setup a transmit queue of an Ethernet device. */
871 typedef void (*eth_queue_release_t)(void *queue);
872 /**< @internal Release memory resources allocated by given RX/TX queue. */
874 typedef uint32_t (*eth_rx_queue_count_t)(struct rte_eth_dev *dev,
875 uint16_t rx_queue_id);
876 /**< @Get number of available descriptors on a receive queue of an Ethernet device. */
878 typedef int (*eth_rx_descriptor_done_t)(void *rxq, uint16_t offset);
879 /**< @Check DD bit of specific RX descriptor */
881 typedef int (*vlan_filter_set_t)(struct rte_eth_dev *dev,
884 /**< @internal filtering of a VLAN Tag Identifier by an Ethernet device. */
886 typedef void (*vlan_tpid_set_t)(struct rte_eth_dev *dev,
888 /**< @internal set the outer VLAN-TPID by an Ethernet device. */
890 typedef void (*vlan_offload_set_t)(struct rte_eth_dev *dev, int mask);
891 /**< @internal set VLAN offload function by an Ethernet device. */
893 typedef void (*vlan_strip_queue_set_t)(struct rte_eth_dev *dev,
894 uint16_t rx_queue_id,
896 /**< @internal VLAN stripping enable/disable by an queue of Ethernet device. */
898 typedef uint16_t (*eth_rx_burst_t)(void *rxq,
899 struct rte_mbuf **rx_pkts,
901 /**< @internal Retrieve input packets from a receive queue of an Ethernet device. */
903 typedef uint16_t (*eth_tx_burst_t)(void *txq,
904 struct rte_mbuf **tx_pkts,
906 /**< @internal Send output packets on a transmit queue of an Ethernet device. */
908 typedef int (*fdir_add_signature_filter_t)(struct rte_eth_dev *dev,
909 struct rte_fdir_filter *fdir_ftr,
911 /**< @internal Setup a new signature filter rule on an Ethernet device */
913 typedef int (*fdir_update_signature_filter_t)(struct rte_eth_dev *dev,
914 struct rte_fdir_filter *fdir_ftr,
916 /**< @internal Update a signature filter rule on an Ethernet device */
918 typedef int (*fdir_remove_signature_filter_t)(struct rte_eth_dev *dev,
919 struct rte_fdir_filter *fdir_ftr);
920 /**< @internal Remove a signature filter rule on an Ethernet device */
922 typedef void (*fdir_infos_get_t)(struct rte_eth_dev *dev,
923 struct rte_eth_fdir *fdir);
924 /**< @internal Get information about fdir status */
926 typedef int (*fdir_add_perfect_filter_t)(struct rte_eth_dev *dev,
927 struct rte_fdir_filter *fdir_ftr,
928 uint16_t soft_id, uint8_t rx_queue,
930 /**< @internal Setup a new perfect filter rule on an Ethernet device */
932 typedef int (*fdir_update_perfect_filter_t)(struct rte_eth_dev *dev,
933 struct rte_fdir_filter *fdir_ftr,
934 uint16_t soft_id, uint8_t rx_queue,
936 /**< @internal Update a perfect filter rule on an Ethernet device */
938 typedef int (*fdir_remove_perfect_filter_t)(struct rte_eth_dev *dev,
939 struct rte_fdir_filter *fdir_ftr,
941 /**< @internal Remove a perfect filter rule on an Ethernet device */
943 typedef int (*fdir_set_masks_t)(struct rte_eth_dev *dev,
944 struct rte_fdir_masks *fdir_masks);
945 /**< @internal Setup flow director masks on an Ethernet device */
947 typedef int (*flow_ctrl_set_t)(struct rte_eth_dev *dev,
948 struct rte_eth_fc_conf *fc_conf);
949 /**< @internal Setup flow control parameter on an Ethernet device */
951 typedef int (*priority_flow_ctrl_set_t)(struct rte_eth_dev *dev,
952 struct rte_eth_pfc_conf *pfc_conf);
953 /**< @internal Setup priority flow control parameter on an Ethernet device */
955 typedef int (*reta_update_t)(struct rte_eth_dev *dev,
956 struct rte_eth_rss_reta *reta_conf);
957 /**< @internal Update RSS redirection table on an Ethernet device */
959 typedef int (*reta_query_t)(struct rte_eth_dev *dev,
960 struct rte_eth_rss_reta *reta_conf);
961 /**< @internal Query RSS redirection table on an Ethernet device */
963 typedef int (*eth_dev_led_on_t)(struct rte_eth_dev *dev);
964 /**< @internal Turn on SW controllable LED on an Ethernet device */
966 typedef int (*eth_dev_led_off_t)(struct rte_eth_dev *dev);
967 /**< @internal Turn off SW controllable LED on an Ethernet device */
969 typedef void (*eth_mac_addr_remove_t)(struct rte_eth_dev *dev, uint32_t index);
970 /**< @internal Remove MAC address from receive address register */
972 typedef void (*eth_mac_addr_add_t)(struct rte_eth_dev *dev,
973 struct ether_addr *mac_addr,
976 /**< @internal Set a MAC address into Receive Address Address Register */
978 typedef int (*eth_uc_hash_table_set_t)(struct rte_eth_dev *dev,
979 struct ether_addr *mac_addr,
981 /**< @internal Set a Unicast Hash bitmap */
983 typedef int (*eth_uc_all_hash_table_set_t)(struct rte_eth_dev *dev,
985 /**< @internal Set all Unicast Hash bitmap */
987 typedef int (*eth_set_vf_rx_mode_t)(struct rte_eth_dev *dev,
991 /**< @internal Set a VF receive mode */
993 typedef int (*eth_set_vf_rx_t)(struct rte_eth_dev *dev,
996 /**< @internal Set a VF receive mode */
998 typedef int (*eth_set_vf_tx_t)(struct rte_eth_dev *dev,
1001 /**< @internal Enable or disable a VF transmit */
1003 typedef int (*eth_set_vf_vlan_filter_t)(struct rte_eth_dev *dev,
1007 /**< @internal Set VF VLAN pool filter */
1009 typedef int (*eth_mirror_rule_set_t)(struct rte_eth_dev *dev,
1010 struct rte_eth_vmdq_mirror_conf *mirror_conf,
1013 /**< @internal Add a traffic mirroring rule on an Ethernet device */
1015 typedef int (*eth_mirror_rule_reset_t)(struct rte_eth_dev *dev,
1017 /**< @internal Remove a traffic mirroring rule on an Ethernet device */
1019 #ifdef RTE_NIC_BYPASS
1022 RTE_BYPASS_MODE_NONE,
1023 RTE_BYPASS_MODE_NORMAL,
1024 RTE_BYPASS_MODE_BYPASS,
1025 RTE_BYPASS_MODE_ISOLATE,
1026 RTE_BYPASS_MODE_NUM,
1029 #define RTE_BYPASS_MODE_VALID(x) \
1030 ((x) > RTE_BYPASS_MODE_NONE && (x) < RTE_BYPASS_MODE_NUM)
1033 RTE_BYPASS_EVENT_NONE,
1034 RTE_BYPASS_EVENT_START,
1035 RTE_BYPASS_EVENT_OS_ON = RTE_BYPASS_EVENT_START,
1036 RTE_BYPASS_EVENT_POWER_ON,
1037 RTE_BYPASS_EVENT_OS_OFF,
1038 RTE_BYPASS_EVENT_POWER_OFF,
1039 RTE_BYPASS_EVENT_TIMEOUT,
1040 RTE_BYPASS_EVENT_NUM
1043 #define RTE_BYPASS_EVENT_VALID(x) \
1044 ((x) > RTE_BYPASS_EVENT_NONE && (x) < RTE_BYPASS_MODE_NUM)
1047 RTE_BYPASS_TMT_OFF, /* timeout disabled. */
1048 RTE_BYPASS_TMT_1_5_SEC, /* timeout for 1.5 seconds */
1049 RTE_BYPASS_TMT_2_SEC, /* timeout for 2 seconds */
1050 RTE_BYPASS_TMT_3_SEC, /* timeout for 3 seconds */
1051 RTE_BYPASS_TMT_4_SEC, /* timeout for 4 seconds */
1052 RTE_BYPASS_TMT_8_SEC, /* timeout for 8 seconds */
1053 RTE_BYPASS_TMT_16_SEC, /* timeout for 16 seconds */
1054 RTE_BYPASS_TMT_32_SEC, /* timeout for 32 seconds */
1058 #define RTE_BYPASS_TMT_VALID(x) \
1059 ((x) == RTE_BYPASS_TMT_OFF || \
1060 ((x) > RTE_BYPASS_TMT_OFF && (x) < RTE_BYPASS_TMT_NUM))
1062 typedef void (*bypass_init_t)(struct rte_eth_dev *dev);
1063 typedef int32_t (*bypass_state_set_t)(struct rte_eth_dev *dev, uint32_t *new_state);
1064 typedef int32_t (*bypass_state_show_t)(struct rte_eth_dev *dev, uint32_t *state);
1065 typedef int32_t (*bypass_event_set_t)(struct rte_eth_dev *dev, uint32_t state, uint32_t event);
1066 typedef int32_t (*bypass_event_show_t)(struct rte_eth_dev *dev, uint32_t event_shift, uint32_t *event);
1067 typedef int32_t (*bypass_wd_timeout_set_t)(struct rte_eth_dev *dev, uint32_t timeout);
1068 typedef int32_t (*bypass_wd_timeout_show_t)(struct rte_eth_dev *dev, uint32_t *wd_timeout);
1069 typedef int32_t (*bypass_ver_show_t)(struct rte_eth_dev *dev, uint32_t *ver);
1070 typedef int32_t (*bypass_wd_reset_t)(struct rte_eth_dev *dev);
1075 * @internal A structure containing the functions exported by an Ethernet driver.
1077 struct eth_dev_ops {
1078 eth_dev_configure_t dev_configure; /**< Configure device. */
1079 eth_dev_start_t dev_start; /**< Start device. */
1080 eth_dev_stop_t dev_stop; /**< Stop device. */
1081 eth_dev_close_t dev_close; /**< Close device. */
1082 eth_promiscuous_enable_t promiscuous_enable; /**< Promiscuous ON. */
1083 eth_promiscuous_disable_t promiscuous_disable;/**< Promiscuous OFF. */
1084 eth_allmulticast_enable_t allmulticast_enable;/**< RX multicast ON. */
1085 eth_allmulticast_disable_t allmulticast_disable;/**< RX multicast OF. */
1086 eth_link_update_t link_update; /**< Get device link state. */
1087 eth_stats_get_t stats_get; /**< Get device statistics. */
1088 eth_stats_reset_t stats_reset; /**< Reset device statistics. */
1089 eth_queue_stats_mapping_set_t queue_stats_mapping_set;
1090 /**< Configure per queue stat counter mapping. */
1091 eth_dev_infos_get_t dev_infos_get; /**< Get device info. */
1092 vlan_filter_set_t vlan_filter_set; /**< Filter VLAN Setup. */
1093 vlan_tpid_set_t vlan_tpid_set; /**< Outer VLAN TPID Setup. */
1094 vlan_strip_queue_set_t vlan_strip_queue_set; /**< VLAN Stripping on queue. */
1095 vlan_offload_set_t vlan_offload_set; /**< Set VLAN Offload. */
1096 eth_rx_queue_setup_t rx_queue_setup;/**< Set up device RX queue.*/
1097 eth_queue_release_t rx_queue_release;/**< Release RX queue.*/
1098 eth_rx_queue_count_t rx_queue_count; /**< Get Rx queue count. */
1099 eth_rx_descriptor_done_t rx_descriptor_done; /**< Check rxd DD bit */
1100 eth_tx_queue_setup_t tx_queue_setup;/**< Set up device TX queue.*/
1101 eth_queue_release_t tx_queue_release;/**< Release TX queue.*/
1102 eth_dev_led_on_t dev_led_on; /**< Turn on LED. */
1103 eth_dev_led_off_t dev_led_off; /**< Turn off LED. */
1104 flow_ctrl_set_t flow_ctrl_set; /**< Setup flow control. */
1105 priority_flow_ctrl_set_t priority_flow_ctrl_set; /**< Setup priority flow control.*/
1106 eth_mac_addr_remove_t mac_addr_remove; /**< Remove MAC address */
1107 eth_mac_addr_add_t mac_addr_add; /**< Add a MAC address */
1108 eth_uc_hash_table_set_t uc_hash_table_set; /**< Set Unicast Table Array */
1109 eth_uc_all_hash_table_set_t uc_all_hash_table_set; /**< Set Unicast hash bitmap */
1110 eth_mirror_rule_set_t mirror_rule_set; /**< Add a traffic mirror rule.*/
1111 eth_mirror_rule_reset_t mirror_rule_reset; /**< reset a traffic mirror rule.*/
1112 eth_set_vf_rx_mode_t set_vf_rx_mode; /**< Set VF RX mode */
1113 eth_set_vf_rx_t set_vf_rx; /**< enable/disable a VF receive */
1114 eth_set_vf_tx_t set_vf_tx; /**< enable/disable a VF transmit */
1115 eth_set_vf_vlan_filter_t set_vf_vlan_filter; /**< Set VF VLAN filter */
1117 /** Add a signature filter. */
1118 fdir_add_signature_filter_t fdir_add_signature_filter;
1119 /** Update a signature filter. */
1120 fdir_update_signature_filter_t fdir_update_signature_filter;
1121 /** Remove a signature filter. */
1122 fdir_remove_signature_filter_t fdir_remove_signature_filter;
1123 /** Get information about FDIR status. */
1124 fdir_infos_get_t fdir_infos_get;
1125 /** Add a perfect filter. */
1126 fdir_add_perfect_filter_t fdir_add_perfect_filter;
1127 /** Update a perfect filter. */
1128 fdir_update_perfect_filter_t fdir_update_perfect_filter;
1129 /** Remove a perfect filter. */
1130 fdir_remove_perfect_filter_t fdir_remove_perfect_filter;
1131 /** Setup masks for FDIR filtering. */
1132 fdir_set_masks_t fdir_set_masks;
1133 /** Update redirection table. */
1134 reta_update_t reta_update;
1135 /** Query redirection table. */
1136 reta_query_t reta_query;
1137 /* bypass control */
1138 #ifdef RTE_NIC_BYPASS
1139 bypass_init_t bypass_init;
1140 bypass_state_set_t bypass_state_set;
1141 bypass_state_show_t bypass_state_show;
1142 bypass_event_set_t bypass_event_set;
1143 bypass_event_show_t bypass_event_show;
1144 bypass_wd_timeout_set_t bypass_wd_timeout_set;
1145 bypass_wd_timeout_show_t bypass_wd_timeout_show;
1146 bypass_ver_show_t bypass_ver_show;
1147 bypass_wd_reset_t bypass_wd_reset;
1154 * The generic data structure associated with each ethernet device.
1156 * Pointers to burst-oriented packet receive and transmit functions are
1157 * located at the beginning of the structure, along with the pointer to
1158 * where all the data elements for the particular device are stored in shared
1159 * memory. This split allows the function pointer and driver data to be per-
1160 * process, while the actual configuration data for the device is shared.
1162 struct rte_eth_dev {
1163 eth_rx_burst_t rx_pkt_burst; /**< Pointer to PMD receive function. */
1164 eth_tx_burst_t tx_pkt_burst; /**< Pointer to PMD transmit function. */
1165 struct rte_eth_dev_data *data; /**< Pointer to device data */
1166 const struct eth_driver *driver;/**< Driver for this device */
1167 struct eth_dev_ops *dev_ops; /**< Functions exported by PMD */
1168 struct rte_pci_device *pci_dev; /**< PCI info. supplied by probing */
1169 struct rte_eth_dev_cb_list callbacks; /**< User application callbacks */
1172 struct rte_eth_dev_sriov {
1173 uint8_t active; /**< SRIOV is active with 16, 32 or 64 pools */
1174 uint8_t nb_q_per_pool; /**< rx queue number per pool */
1175 uint16_t def_vmdq_idx; /**< Default pool num used for PF */
1176 uint16_t def_pool_q_idx; /**< Default pool queue start reg index */
1178 #define RTE_ETH_DEV_SRIOV(dev) ((dev)->data->sriov)
1182 * The data part, with no function pointers, associated with each ethernet device.
1184 * This structure is safe to place in shared memory to be common among different
1185 * processes in a multi-process configuration.
1187 struct rte_eth_dev_data {
1188 void **rx_queues; /**< Array of pointers to RX queues. */
1189 void **tx_queues; /**< Array of pointers to TX queues. */
1190 uint16_t nb_rx_queues; /**< Number of RX queues. */
1191 uint16_t nb_tx_queues; /**< Number of TX queues. */
1193 struct rte_eth_dev_sriov sriov; /**< SRIOV data */
1195 void *dev_private; /**< PMD-specific private data */
1197 struct rte_eth_link dev_link;
1198 /**< Link-level information & status */
1200 struct rte_eth_conf dev_conf; /**< Configuration applied to device. */
1201 uint16_t max_frame_size; /**< Default is ETHER_MAX_LEN (1518). */
1203 uint64_t rx_mbuf_alloc_failed; /**< RX ring mbuf allocation failures. */
1204 struct ether_addr* mac_addrs;/**< Device Ethernet Link address. */
1205 uint64_t mac_pool_sel[ETH_NUM_RECEIVE_MAC_ADDR];
1206 /** bitmap array of associating Ethernet MAC addresses to pools */
1207 struct ether_addr* hash_mac_addrs;
1208 /** Device Ethernet MAC addresses of hash filtering. */
1209 uint8_t port_id; /**< Device [external] port identifier. */
1210 uint8_t promiscuous : 1, /**< RX promiscuous mode ON(1) / OFF(0). */
1211 scattered_rx : 1, /**< RX of scattered packets is ON(1) / OFF(0) */
1212 all_multicast : 1, /**< RX all multicast mode ON(1) / OFF(0). */
1213 dev_started : 1; /**< Device state: STARTED(1) / STOPPED(0). */
1218 * The pool of *rte_eth_dev* structures. The size of the pool
1219 * is configured at compile-time in the <rte_ethdev.c> file.
1221 extern struct rte_eth_dev rte_eth_devices[];
1224 * Get the total number of Ethernet devices that have been successfully
1225 * initialized by the [matching] Ethernet driver during the PCI probing phase.
1226 * All devices whose port identifier is in the range
1227 * [0, rte_eth_dev_count() - 1] can be operated on by network applications.
1230 * - The total number of usable Ethernet devices.
1232 extern uint8_t rte_eth_dev_count(void);
1235 * Function for internal use by dummy drivers primarily, e.g. ring-based
1237 * Allocates a new ethdev slot for an ethernet device and returns the pointer
1238 * to that slot for the driver to use.
1241 * - Slot in the rte_dev_devices array for a new device;
1243 struct rte_eth_dev *rte_eth_dev_allocate(void);
1248 * Initialization function of an Ethernet driver invoked for each matching
1249 * Ethernet PCI device detected during the PCI probing phase.
1252 * The pointer to the [matching] Ethernet driver structure supplied by
1253 * the PMD when it registered itself.
1255 * The *eth_dev* pointer is the address of the *rte_eth_dev* structure
1256 * associated with the matching device and which have been [automatically]
1257 * allocated in the *rte_eth_devices* array.
1258 * The *eth_dev* structure is supplied to the driver initialization function
1259 * with the following fields already initialized:
1261 * - *pci_dev*: Holds the pointers to the *rte_pci_device* structure which
1262 * contains the generic PCI information of the matching device.
1264 * - *dev_private*: Holds a pointer to the device private data structure.
1266 * - *max_frame_size*: Contains the default Ethernet maximum frame length
1269 * - *port_id*: Contains the port index of the device (actually the index
1270 * of the *eth_dev* structure in the *rte_eth_devices* array).
1273 * - 0: Success, the device is properly initialized by the driver.
1274 * In particular, the driver MUST have set up the *dev_ops* pointer
1275 * of the *eth_dev* structure.
1276 * - <0: Error code of the device initialization failure.
1278 typedef int (*eth_dev_init_t)(struct eth_driver *eth_drv,
1279 struct rte_eth_dev *eth_dev);
1283 * The structure associated with a PMD Ethernet driver.
1285 * Each Ethernet driver acts as a PCI driver and is represented by a generic
1286 * *eth_driver* structure that holds:
1288 * - An *rte_pci_driver* structure (which must be the first field).
1290 * - The *eth_dev_init* function invoked for each matching PCI device.
1292 * - The size of the private data to allocate for each matching device.
1295 struct rte_pci_driver pci_drv; /**< The PMD is also a PCI driver. */
1296 eth_dev_init_t eth_dev_init; /**< Device init function. */
1297 unsigned int dev_private_size; /**< Size of device private data. */
1302 * A function invoked by the initialization function of an Ethernet driver
1303 * to simultaneously register itself as a PCI driver and as an Ethernet
1304 * Poll Mode Driver (PMD).
1307 * The pointer to the *eth_driver* structure associated with
1308 * the Ethernet driver.
1310 extern void rte_eth_driver_register(struct eth_driver *eth_drv);
1313 * The initialization function of the driver for
1314 * Intel(r) IGB Gigabit Ethernet Controller devices.
1315 * This function is invoked once at EAL start time.
1319 extern int rte_igb_pmd_init(void);
1322 * The initialization function of the driver for
1323 * Intel(r) EM Gigabit Ethernet Controller devices.
1324 * This function is invoked once at EAL start time.
1328 extern int rte_em_pmd_init(void);
1331 * The initialization function of the driver for 1Gbps Intel IGB_VF
1333 * Invoked once at EAL start time.
1337 extern int rte_igbvf_pmd_init(void);
1340 * The initialization function of the driver for 10Gbps Intel IXGBE
1342 * Invoked once at EAL start time.
1346 extern int rte_ixgbe_pmd_init(void);
1349 * The initialization function of the driver for 10Gbps Intel IXGBE_VF
1351 * Invoked once at EAL start time.
1355 extern int rte_ixgbevf_pmd_init(void);
1358 * The initialization function of the driver for Qumranet virtio-net
1360 * Invoked once at EAL start time.
1364 extern int rte_virtio_pmd_init(void);
1367 * The initialization function of the driver for VMware VMXNET3
1369 * Invoked once at EAL start time.
1373 extern int rte_vmxnet3_pmd_init(void);
1377 * The initialization function of *all* supported and enabled drivers.
1378 * Right now, the following PMDs are supported:
1386 * This function is invoked once at EAL start time.
1389 * Error code of the device initialization failure,
1390 * -ENODEV if there are no drivers available
1391 * (e.g. if all driver config options are = n).
1394 int rte_pmd_init_all(void)
1398 #ifdef RTE_LIBRTE_IGB_PMD
1399 if ((ret = rte_igb_pmd_init()) != 0) {
1400 RTE_LOG(ERR, PMD, "Cannot init igb PMD\n");
1403 if ((ret = rte_igbvf_pmd_init()) != 0) {
1404 RTE_LOG(ERR, PMD, "Cannot init igbvf PMD\n");
1407 #endif /* RTE_LIBRTE_IGB_PMD */
1409 #ifdef RTE_LIBRTE_EM_PMD
1410 if ((ret = rte_em_pmd_init()) != 0) {
1411 RTE_LOG(ERR, PMD, "Cannot init em PMD\n");
1414 #endif /* RTE_LIBRTE_EM_PMD */
1416 #ifdef RTE_LIBRTE_IXGBE_PMD
1417 if ((ret = rte_ixgbe_pmd_init()) != 0) {
1418 RTE_LOG(ERR, PMD, "Cannot init ixgbe PMD\n");
1421 if ((ret = rte_ixgbevf_pmd_init()) != 0) {
1422 RTE_LOG(ERR, PMD, "Cannot init ixgbevf PMD\n");
1425 #endif /* RTE_LIBRTE_IXGBE_PMD */
1427 #ifdef RTE_LIBRTE_VIRTIO_PMD
1428 if ((ret = rte_virtio_pmd_init()) != 0) {
1429 RTE_LOG(ERR, PMD, "Cannot init virtio PMD\n");
1432 #endif /* RTE_LIBRTE_VIRTIO_PMD */
1434 #ifdef RTE_LIBRTE_VMXNET3_PMD
1435 if ((ret = rte_vmxnet3_pmd_init()) != 0) {
1436 RTE_LOG(ERR, PMD, "Cannot init vmxnet3 PMD\n");
1439 #endif /* RTE_LIBRTE_VMXNET3_PMD */
1442 RTE_LOG(ERR, PMD, "No PMD(s) are configured\n");
1447 * Configure an Ethernet device.
1448 * This function must be invoked first before any other function in the
1449 * Ethernet API. This function can also be re-invoked when a device is in the
1453 * The port identifier of the Ethernet device to configure.
1454 * @param nb_rx_queue
1455 * The number of receive queues to set up for the Ethernet device.
1456 * @param nb_tx_queue
1457 * The number of transmit queues to set up for the Ethernet device.
1459 * The pointer to the configuration data to be used for the Ethernet device.
1460 * The *rte_eth_conf* structure includes:
1461 * - the hardware offload features to activate, with dedicated fields for
1462 * each statically configurable offload hardware feature provided by
1463 * Ethernet devices, such as IP checksum or VLAN tag stripping for
1465 * - the Receive Side Scaling (RSS) configuration when using multiple RX
1468 * Embedding all configuration information in a single data structure
1469 * is the more flexible method that allows the addition of new features
1470 * without changing the syntax of the API.
1472 * - 0: Success, device configured.
1473 * - <0: Error code returned by the driver configuration function.
1475 extern int rte_eth_dev_configure(uint8_t port_id,
1476 uint16_t nb_rx_queue,
1477 uint16_t nb_tx_queue,
1478 const struct rte_eth_conf *eth_conf);
1481 * Allocate and set up a receive queue for an Ethernet device.
1483 * The function allocates a contiguous block of memory for *nb_rx_desc*
1484 * receive descriptors from a memory zone associated with *socket_id*
1485 * and initializes each receive descriptor with a network buffer allocated
1486 * from the memory pool *mb_pool*.
1489 * The port identifier of the Ethernet device.
1490 * @param rx_queue_id
1491 * The index of the receive queue to set up.
1492 * The value must be in the range [0, nb_rx_queue - 1] previously supplied
1493 * to rte_eth_dev_configure().
1495 * The number of receive descriptors to allocate for the receive ring.
1497 * The *socket_id* argument is the socket identifier in case of NUMA.
1498 * The value can be *SOCKET_ID_ANY* if there is no NUMA constraint for
1499 * the DMA memory allocated for the receive descriptors of the ring.
1501 * The pointer to the configuration data to be used for the receive queue.
1502 * The *rx_conf* structure contains an *rx_thresh* structure with the values
1503 * of the Prefetch, Host, and Write-Back threshold registers of the receive
1506 * The pointer to the memory pool from which to allocate *rte_mbuf* network
1507 * memory buffers to populate each descriptor of the receive ring.
1509 * - 0: Success, receive queue correctly set up.
1510 * - -EINVAL: The size of network buffers which can be allocated from the
1511 * memory pool does not fit the various buffer sizes allowed by the
1512 * device controller.
1513 * - -ENOMEM: Unable to allocate the receive ring descriptors or to
1514 * allocate network memory buffers from the memory pool when
1515 * initializing receive descriptors.
1517 extern int rte_eth_rx_queue_setup(uint8_t port_id, uint16_t rx_queue_id,
1518 uint16_t nb_rx_desc, unsigned int socket_id,
1519 const struct rte_eth_rxconf *rx_conf,
1520 struct rte_mempool *mb_pool);
1523 * Allocate and set up a transmit queue for an Ethernet device.
1526 * The port identifier of the Ethernet device.
1527 * @param tx_queue_id
1528 * The index of the transmit queue to set up.
1529 * The value must be in the range [0, nb_tx_queue - 1] previously supplied
1530 * to rte_eth_dev_configure().
1532 * The number of transmit descriptors to allocate for the transmit ring.
1534 * The *socket_id* argument is the socket identifier in case of NUMA.
1535 * Its value can be *SOCKET_ID_ANY* if there is no NUMA constraint for
1536 * the DMA memory allocated for the transmit descriptors of the ring.
1538 * The pointer to the configuration data to be used for the transmit queue.
1539 * The *tx_conf* structure contains the following data:
1540 * - The *tx_thresh* structure with the values of the Prefetch, Host, and
1541 * Write-Back threshold registers of the transmit ring.
1542 * When setting Write-Back threshold to the value greater then zero,
1543 * *tx_rs_thresh* value should be explicitly set to one.
1544 * - The *tx_free_thresh* value indicates the [minimum] number of network
1545 * buffers that must be pending in the transmit ring to trigger their
1546 * [implicit] freeing by the driver transmit function.
1547 * - The *tx_rs_thresh* value indicates the [minimum] number of transmit
1548 * descriptors that must be pending in the transmit ring before setting the
1549 * RS bit on a descriptor by the driver transmit function.
1550 * The *tx_rs_thresh* value should be less or equal then
1551 * *tx_free_thresh* value, and both of them should be less then
1553 * - The *txq_flags* member contains flags to pass to the TX queue setup
1554 * function to configure the behavior of the TX queue. This should be set
1555 * to 0 if no special configuration is required.
1557 * Note that setting *tx_free_thresh* or *tx_rs_thresh* value to 0 forces
1558 * the transmit function to use default values.
1560 * - 0: Success, the transmit queue is correctly set up.
1561 * - -ENOMEM: Unable to allocate the transmit ring descriptors.
1563 extern int rte_eth_tx_queue_setup(uint8_t port_id, uint16_t tx_queue_id,
1564 uint16_t nb_tx_desc, unsigned int socket_id,
1565 const struct rte_eth_txconf *tx_conf);
1568 * Return the NUMA socket to which an Ethernet device is connected
1571 * The port identifier of the Ethernet device
1573 * The NUMA socket id to which the Ethernet device is connected or
1574 * a default of zero if the socket could not be determined.
1575 * -1 is returned is the port_id value is out of range.
1577 extern int rte_eth_dev_socket_id(uint8_t port_id);
1581 * Start an Ethernet device.
1583 * The device start step is the last one and consists of setting the configured
1584 * offload features and in starting the transmit and the receive units of the
1586 * On success, all basic functions exported by the Ethernet API (link status,
1587 * receive/transmit, and so on) can be invoked.
1590 * The port identifier of the Ethernet device.
1592 * - 0: Success, Ethernet device started.
1593 * - <0: Error code of the driver device start function.
1595 extern int rte_eth_dev_start(uint8_t port_id);
1598 * Stop an Ethernet device. The device can be restarted with a call to
1599 * rte_eth_dev_start()
1602 * The port identifier of the Ethernet device.
1604 extern void rte_eth_dev_stop(uint8_t port_id);
1607 * Close an Ethernet device. The device cannot be restarted!
1610 * The port identifier of the Ethernet device.
1612 extern void rte_eth_dev_close(uint8_t port_id);
1615 * Enable receipt in promiscuous mode for an Ethernet device.
1618 * The port identifier of the Ethernet device.
1620 extern void rte_eth_promiscuous_enable(uint8_t port_id);
1623 * Disable receipt in promiscuous mode for an Ethernet device.
1626 * The port identifier of the Ethernet device.
1628 extern void rte_eth_promiscuous_disable(uint8_t port_id);
1631 * Return the value of promiscuous mode for an Ethernet device.
1634 * The port identifier of the Ethernet device.
1636 * - (1) if promiscuous is enabled
1637 * - (0) if promiscuous is disabled.
1640 extern int rte_eth_promiscuous_get(uint8_t port_id);
1643 * Enable the receipt of any multicast frame by an Ethernet device.
1646 * The port identifier of the Ethernet device.
1648 extern void rte_eth_allmulticast_enable(uint8_t port_id);
1651 * Disable the receipt of all multicast frames by an Ethernet device.
1654 * The port identifier of the Ethernet device.
1656 extern void rte_eth_allmulticast_disable(uint8_t port_id);
1659 * Return the value of allmulticast mode for an Ethernet device.
1662 * The port identifier of the Ethernet device.
1664 * - (1) if allmulticast is enabled
1665 * - (0) if allmulticast is disabled.
1668 extern int rte_eth_allmulticast_get(uint8_t port_id);
1671 * Retrieve the status (ON/OFF), the speed (in Mbps) and the mode (HALF-DUPLEX
1672 * or FULL-DUPLEX) of the physical link of an Ethernet device. It might need
1673 * to wait up to 9 seconds in it.
1676 * The port identifier of the Ethernet device.
1678 * A pointer to an *rte_eth_link* structure to be filled with
1679 * the status, the speed and the mode of the Ethernet device link.
1681 extern void rte_eth_link_get(uint8_t port_id, struct rte_eth_link *link);
1684 * Retrieve the status (ON/OFF), the speed (in Mbps) and the mode (HALF-DUPLEX
1685 * or FULL-DUPLEX) of the physical link of an Ethernet device. It is a no-wait
1686 * version of rte_eth_link_get().
1689 * The port identifier of the Ethernet device.
1691 * A pointer to an *rte_eth_link* structure to be filled with
1692 * the status, the speed and the mode of the Ethernet device link.
1694 extern void rte_eth_link_get_nowait(uint8_t port_id,
1695 struct rte_eth_link *link);
1698 * Retrieve the general I/O statistics of an Ethernet device.
1701 * The port identifier of the Ethernet device.
1703 * A pointer to a structure of type *rte_eth_stats* to be filled with
1704 * the values of device counters for the following set of statistics:
1705 * - *ipackets* with the total of successfully received packets.
1706 * - *opackets* with the total of successfully transmitted packets.
1707 * - *ibytes* with the total of successfully received bytes.
1708 * - *obytes* with the total of successfully transmitted bytes.
1709 * - *ierrors* with the total of erroneous received packets.
1710 * - *oerrors* with the total of failed transmitted packets.
1712 extern void rte_eth_stats_get(uint8_t port_id, struct rte_eth_stats *stats);
1715 * Reset the general I/O statistics of an Ethernet device.
1718 * The port identifier of the Ethernet device.
1720 extern void rte_eth_stats_reset(uint8_t port_id);
1723 * Set a mapping for the specified transmit queue to the specified per-queue
1724 * statistics counter.
1727 * The port identifier of the Ethernet device.
1728 * @param tx_queue_id
1729 * The index of the transmit queue for which a queue stats mapping is required.
1730 * The value must be in the range [0, nb_tx_queue - 1] previously supplied
1731 * to rte_eth_dev_configure().
1733 * The per-queue packet statistics functionality number that the transmit
1734 * queue is to be assigned.
1735 * The value must be in the range [0, RTE_MAX_ETHPORT_QUEUE_STATS_MAPS - 1].
1737 * Zero if successful. Non-zero otherwise.
1739 extern int rte_eth_dev_set_tx_queue_stats_mapping(uint8_t port_id,
1740 uint16_t tx_queue_id,
1744 * Set a mapping for the specified receive queue to the specified per-queue
1745 * statistics counter.
1748 * The port identifier of the Ethernet device.
1749 * @param rx_queue_id
1750 * The index of the receive queue for which a queue stats mapping is required.
1751 * The value must be in the range [0, nb_rx_queue - 1] previously supplied
1752 * to rte_eth_dev_configure().
1754 * The per-queue packet statistics functionality number that the receive
1755 * queue is to be assigned.
1756 * The value must be in the range [0, RTE_MAX_ETHPORT_QUEUE_STATS_MAPS - 1].
1758 * Zero if successful. Non-zero otherwise.
1760 extern int rte_eth_dev_set_rx_queue_stats_mapping(uint8_t port_id,
1761 uint16_t rx_queue_id,
1765 * Retrieve the Ethernet address of an Ethernet device.
1768 * The port identifier of the Ethernet device.
1770 * A pointer to a structure of type *ether_addr* to be filled with
1771 * the Ethernet address of the Ethernet device.
1773 extern void rte_eth_macaddr_get(uint8_t port_id, struct ether_addr *mac_addr);
1776 * Retrieve the contextual information of an Ethernet device.
1779 * The port identifier of the Ethernet device.
1781 * A pointer to a structure of type *rte_eth_dev_info* to be filled with
1782 * the contextual information of the Ethernet device.
1784 extern void rte_eth_dev_info_get(uint8_t port_id,
1785 struct rte_eth_dev_info *dev_info);
1788 * Enable/Disable hardware filtering by an Ethernet device of received
1789 * VLAN packets tagged with a given VLAN Tag Identifier.
1792 * The port identifier of the Ethernet device.
1794 * The VLAN Tag Identifier whose filtering must be enabled or disabled.
1796 * If > 0, enable VLAN filtering of VLAN packets tagged with *vlan_id*.
1797 * Otherwise, disable VLAN filtering of VLAN packets tagged with *vlan_id*.
1799 * - (0) if successful.
1800 * - (-ENOSUP) if hardware-assisted VLAN filtering not configured.
1801 * - (-ENODEV) if *port_id* invalid.
1802 * - (-ENOSYS) if VLAN filtering on *port_id* disabled.
1803 * - (-EINVAL) if *vlan_id* > 4095.
1805 extern int rte_eth_dev_vlan_filter(uint8_t port_id, uint16_t vlan_id , int on);
1808 * Enable/Disable hardware VLAN Strip by a rx queue of an Ethernet device.
1809 * 82599/X540 can support VLAN stripping at the rx queue level
1812 * The port identifier of the Ethernet device.
1813 * @param rx_queue_id
1814 * The index of the receive queue for which a queue stats mapping is required.
1815 * The value must be in the range [0, nb_rx_queue - 1] previously supplied
1816 * to rte_eth_dev_configure().
1818 * If 1, Enable VLAN Stripping of the receive queue of the Ethernet port.
1819 * If 0, Disable VLAN Stripping of the receive queue of the Ethernet port.
1821 * - (0) if successful.
1822 * - (-ENOSUP) if hardware-assisted VLAN stripping not configured.
1823 * - (-ENODEV) if *port_id* invalid.
1824 * - (-EINVAL) if *rx_queue_id* invalid.
1826 extern int rte_eth_dev_set_vlan_strip_on_queue(uint8_t port_id,
1827 uint16_t rx_queue_id, int on);
1830 * Set the Outer VLAN Ether Type by an Ethernet device, it can be inserted to
1831 * the VLAN Header. This is a register setup available on some Intel NIC, not
1832 * but all, please check the data sheet for availability.
1835 * The port identifier of the Ethernet device.
1837 * The Tag Protocol ID
1839 * - (0) if successful.
1840 * - (-ENOSUP) if hardware-assisted VLAN TPID setup is not supported.
1841 * - (-ENODEV) if *port_id* invalid.
1843 extern int rte_eth_dev_set_vlan_ether_type(uint8_t port_id, uint16_t tag_type);
1846 * Set VLAN offload configuration on an Ethernet device
1847 * Enable/Disable Extended VLAN by an Ethernet device, This is a register setup
1848 * available on some Intel NIC, not but all, please check the data sheet for
1850 * Enable/Disable VLAN Strip can be done on rx queue for certain NIC, but here
1851 * the configuration is applied on the port level.
1854 * The port identifier of the Ethernet device.
1855 * @param offload_mask
1856 * The VLAN Offload bit mask can be mixed use with "OR"
1857 * ETH_VLAN_STRIP_OFFLOAD
1858 * ETH_VLAN_FILTER_OFFLOAD
1859 * ETH_VLAN_EXTEND_OFFLOAD
1861 * - (0) if successful.
1862 * - (-ENOSUP) if hardware-assisted VLAN filtering not configured.
1863 * - (-ENODEV) if *port_id* invalid.
1865 extern int rte_eth_dev_set_vlan_offload(uint8_t port_id, int offload_mask);
1868 * Read VLAN Offload configuration from an Ethernet device
1871 * The port identifier of the Ethernet device.
1873 * - (>0) if successful. Bit mask to indicate
1874 * ETH_VLAN_STRIP_OFFLOAD
1875 * ETH_VLAN_FILTER_OFFLOAD
1876 * ETH_VLAN_EXTEND_OFFLOAD
1877 * - (-ENODEV) if *port_id* invalid.
1879 extern int rte_eth_dev_get_vlan_offload(uint8_t port_id);
1883 * Retrieve a burst of input packets from a receive queue of an Ethernet
1884 * device. The retrieved packets are stored in *rte_mbuf* structures whose
1885 * pointers are supplied in the *rx_pkts* array.
1887 * The rte_eth_rx_burst() function loops, parsing the RX ring of the
1888 * receive queue, up to *nb_pkts* packets, and for each completed RX
1889 * descriptor in the ring, it performs the following operations:
1891 * - Initialize the *rte_mbuf* data structure associated with the
1892 * RX descriptor according to the information provided by the NIC into
1893 * that RX descriptor.
1895 * - Store the *rte_mbuf* data structure into the next entry of the
1898 * - Replenish the RX descriptor with a new *rte_mbuf* buffer
1899 * allocated from the memory pool associated with the receive queue at
1900 * initialization time.
1902 * When retrieving an input packet that was scattered by the controller
1903 * into multiple receive descriptors, the rte_eth_rx_burst() function
1904 * appends the associated *rte_mbuf* buffers to the first buffer of the
1907 * The rte_eth_rx_burst() function returns the number of packets
1908 * actually retrieved, which is the number of *rte_mbuf* data structures
1909 * effectively supplied into the *rx_pkts* array.
1910 * A return value equal to *nb_pkts* indicates that the RX queue contained
1911 * at least *rx_pkts* packets, and this is likely to signify that other
1912 * received packets remain in the input queue. Applications implementing
1913 * a "retrieve as much received packets as possible" policy can check this
1914 * specific case and keep invoking the rte_eth_rx_burst() function until
1915 * a value less than *nb_pkts* is returned.
1917 * This receive method has the following advantages:
1919 * - It allows a run-to-completion network stack engine to retrieve and
1920 * to immediately process received packets in a fast burst-oriented
1921 * approach, avoiding the overhead of unnecessary intermediate packet
1922 * queue/dequeue operations.
1924 * - Conversely, it also allows an asynchronous-oriented processing
1925 * method to retrieve bursts of received packets and to immediately
1926 * queue them for further parallel processing by another logical core,
1927 * for instance. However, instead of having received packets being
1928 * individually queued by the driver, this approach allows the invoker
1929 * of the rte_eth_rx_burst() function to queue a burst of retrieved
1930 * packets at a time and therefore dramatically reduce the cost of
1931 * enqueue/dequeue operations per packet.
1933 * - It allows the rte_eth_rx_burst() function of the driver to take
1934 * advantage of burst-oriented hardware features (CPU cache,
1935 * prefetch instructions, and so on) to minimize the number of CPU
1936 * cycles per packet.
1938 * To summarize, the proposed receive API enables many
1939 * burst-oriented optimizations in both synchronous and asynchronous
1940 * packet processing environments with no overhead in both cases.
1942 * The rte_eth_rx_burst() function does not provide any error
1943 * notification to avoid the corresponding overhead. As a hint, the
1944 * upper-level application might check the status of the device link once
1945 * being systematically returned a 0 value for a given number of tries.
1948 * The port identifier of the Ethernet device.
1950 * The index of the receive queue from which to retrieve input packets.
1951 * The value must be in the range [0, nb_rx_queue - 1] previously supplied
1952 * to rte_eth_dev_configure().
1954 * The address of an array of pointers to *rte_mbuf* structures that
1955 * must be large enough to store *nb_pkts* pointers in it.
1957 * The maximum number of packets to retrieve.
1959 * The number of packets actually retrieved, which is the number
1960 * of pointers to *rte_mbuf* structures effectively supplied to the
1963 #ifdef RTE_LIBRTE_ETHDEV_DEBUG
1964 extern uint16_t rte_eth_rx_burst(uint8_t port_id, uint16_t queue_id,
1965 struct rte_mbuf **rx_pkts, uint16_t nb_pkts);
1967 static inline uint16_t
1968 rte_eth_rx_burst(uint8_t port_id, uint16_t queue_id,
1969 struct rte_mbuf **rx_pkts, uint16_t nb_pkts)
1971 struct rte_eth_dev *dev;
1973 dev = &rte_eth_devices[port_id];
1974 return (*dev->rx_pkt_burst)(dev->data->rx_queues[queue_id], rx_pkts, nb_pkts);
1979 * Get the number of used descriptors in a specific queue
1982 * The port identifier of the Ethernet device.
1984 * The queue id on the specific port.
1986 * The number of used descriptors in the specific queue.
1988 #ifdef RTE_LIBRTE_ETHDEV_DEBUG
1989 extern uint32_t rte_eth_rx_queue_count(uint8_t port_id, uint16_t queue_id);
1991 static inline uint32_t
1992 rte_eth_rx_queue_count(uint8_t port_id, uint16_t queue_id)
1994 struct rte_eth_dev *dev;
1996 dev = &rte_eth_devices[port_id];
1997 return (*dev->dev_ops->rx_queue_count)(dev, queue_id);
2002 * Check if the DD bit of the specific RX descriptor in the queue has been set
2005 * The port identifier of the Ethernet device.
2007 * The queue id on the specific port.
2009 * The offset of the descriptor ID from tail.
2011 * - (1) if the specific DD bit is set.
2012 * - (0) if the specific DD bit is not set.
2013 * - (-ENODEV) if *port_id* invalid.
2015 #ifdef RTE_LIBRTE_ETHDEV_DEBUG
2016 extern int rte_eth_rx_descriptor_done(uint8_t port_id,
2021 rte_eth_rx_descriptor_done(uint8_t port_id, uint16_t queue_id, uint16_t offset)
2023 struct rte_eth_dev *dev;
2025 dev = &rte_eth_devices[port_id];
2026 return (*dev->dev_ops->rx_descriptor_done)( \
2027 dev->data->rx_queues[queue_id], offset);
2032 * Send a burst of output packets on a transmit queue of an Ethernet device.
2034 * The rte_eth_tx_burst() function is invoked to transmit output packets
2035 * on the output queue *queue_id* of the Ethernet device designated by its
2037 * The *nb_pkts* parameter is the number of packets to send which are
2038 * supplied in the *tx_pkts* array of *rte_mbuf* structures.
2039 * The rte_eth_tx_burst() function loops, sending *nb_pkts* packets,
2040 * up to the number of transmit descriptors available in the TX ring of the
2042 * For each packet to send, the rte_eth_tx_burst() function performs
2043 * the following operations:
2045 * - Pick up the next available descriptor in the transmit ring.
2047 * - Free the network buffer previously sent with that descriptor, if any.
2049 * - Initialize the transmit descriptor with the information provided
2050 * in the *rte_mbuf data structure.
2052 * In the case of a segmented packet composed of a list of *rte_mbuf* buffers,
2053 * the rte_eth_tx_burst() function uses several transmit descriptors
2056 * The rte_eth_tx_burst() function returns the number of packets it
2057 * actually sent. A return value equal to *nb_pkts* means that all packets
2058 * have been sent, and this is likely to signify that other output packets
2059 * could be immediately transmitted again. Applications that implement a
2060 * "send as many packets to transmit as possible" policy can check this
2061 * specific case and keep invoking the rte_eth_tx_burst() function until
2062 * a value less than *nb_pkts* is returned.
2064 * It is the responsibility of the rte_eth_tx_burst() function to
2065 * transparently free the memory buffers of packets previously sent.
2066 * This feature is driven by the *tx_free_thresh* value supplied to the
2067 * rte_eth_dev_configure() function at device configuration time.
2068 * When the number of previously sent packets reached the "minimum transmit
2069 * packets to free" threshold, the rte_eth_tx_burst() function must
2070 * [attempt to] free the *rte_mbuf* buffers of those packets whose
2071 * transmission was effectively completed.
2074 * The port identifier of the Ethernet device.
2076 * The index of the transmit queue through which output packets must be
2078 * The value must be in the range [0, nb_tx_queue - 1] previously supplied
2079 * to rte_eth_dev_configure().
2081 * The address of an array of *nb_pkts* pointers to *rte_mbuf* structures
2082 * which contain the output packets.
2084 * The maximum number of packets to transmit.
2086 * The number of output packets actually stored in transmit descriptors of
2087 * the transmit ring. The return value can be less than the value of the
2088 * *tx_pkts* parameter when the transmit ring is full or has been filled up.
2090 #ifdef RTE_LIBRTE_ETHDEV_DEBUG
2091 extern uint16_t rte_eth_tx_burst(uint8_t port_id, uint16_t queue_id,
2092 struct rte_mbuf **tx_pkts, uint16_t nb_pkts);
2094 static inline uint16_t
2095 rte_eth_tx_burst(uint8_t port_id, uint16_t queue_id,
2096 struct rte_mbuf **tx_pkts, uint16_t nb_pkts)
2098 struct rte_eth_dev *dev;
2100 dev = &rte_eth_devices[port_id];
2101 return (*dev->tx_pkt_burst)(dev->data->tx_queues[queue_id], tx_pkts, nb_pkts);
2106 * Setup a new signature filter rule on an Ethernet device
2109 * The port identifier of the Ethernet device.
2110 * @param fdir_filter
2111 * The pointer to the fdir filter structure describing the signature filter
2113 * The *rte_fdir_filter* structure includes the values of the different fields
2114 * to match: source and destination IP addresses, vlan id, flexbytes, source
2115 * and destination ports, and so on.
2117 * The index of the RX queue where to store RX packets matching the added
2118 * signature filter defined in fdir_filter.
2120 * - (0) if successful.
2121 * - (-ENOTSUP) if hardware doesn't support flow director mode.
2122 * - (-ENODEV) if *port_id* invalid.
2123 * - (-ENOSYS) if the FDIR mode is not configured in signature mode
2125 * - (-EINVAL) if the fdir_filter information is not correct.
2127 int rte_eth_dev_fdir_add_signature_filter(uint8_t port_id,
2128 struct rte_fdir_filter *fdir_filter,
2132 * Update a signature filter rule on an Ethernet device.
2133 * If the rule doesn't exits, it is created.
2136 * The port identifier of the Ethernet device.
2138 * The pointer to the structure describing the signature filter rule.
2139 * The *rte_fdir_filter* structure includes the values of the different fields
2140 * to match: source and destination IP addresses, vlan id, flexbytes, source
2141 * and destination ports, and so on.
2143 * The index of the RX queue where to store RX packets matching the added
2144 * signature filter defined in fdir_ftr.
2146 * - (0) if successful.
2147 * - (-ENOTSUP) if hardware doesn't support flow director mode.
2148 * - (-ENODEV) if *port_id* invalid.
2149 * - (-ENOSYS) if the flow director mode is not configured in signature mode
2151 * - (-EINVAL) if the fdir_filter information is not correct.
2153 int rte_eth_dev_fdir_update_signature_filter(uint8_t port_id,
2154 struct rte_fdir_filter *fdir_ftr,
2158 * Remove a signature filter rule on an Ethernet device.
2161 * The port identifier of the Ethernet device.
2163 * The pointer to the structure describing the signature filter rule.
2164 * The *rte_fdir_filter* structure includes the values of the different fields
2165 * to match: source and destination IP addresses, vlan id, flexbytes, source
2166 * and destination ports, and so on.
2168 * - (0) if successful.
2169 * - (-ENOTSUP) if hardware doesn't support flow director mode.
2170 * - (-ENODEV) if *port_id* invalid.
2171 * - (-ENOSYS) if the flow director mode is not configured in signature mode
2173 * - (-EINVAL) if the fdir_filter information is not correct.
2175 int rte_eth_dev_fdir_remove_signature_filter(uint8_t port_id,
2176 struct rte_fdir_filter *fdir_ftr);
2179 * Retrieve the flow director information of an Ethernet device.
2182 * The port identifier of the Ethernet device.
2184 * A pointer to a structure of type *rte_eth_dev_fdir* to be filled with
2185 * the flow director information of the Ethernet device.
2187 * - (0) if successful.
2188 * - (-ENOTSUP) if hardware doesn't support flow director mode.
2189 * - (-ENODEV) if *port_id* invalid.
2190 * - (-ENOSYS) if the flow director mode is not configured on *port_id*.
2192 int rte_eth_dev_fdir_get_infos(uint8_t port_id, struct rte_eth_fdir *fdir);
2195 * Add a new perfect filter rule on an Ethernet device.
2198 * The port identifier of the Ethernet device.
2199 * @param fdir_filter
2200 * The pointer to the structure describing the perfect filter rule.
2201 * The *rte_fdir_filter* structure includes the values of the different fields
2202 * to match: source and destination IP addresses, vlan id, flexbytes, source
2203 * and destination ports, and so on.
2204 * IPv6 are not supported.
2206 * The 16-bit value supplied in the field hash.fdir.id of mbuf for RX
2207 * packets matching the perfect filter.
2209 * The index of the RX queue where to store RX packets matching the added
2210 * perfect filter defined in fdir_filter.
2212 * If drop is set to 1, matching RX packets are stored into the RX drop
2213 * queue defined in the rte_fdir_conf.
2215 * - (0) if successful.
2216 * - (-ENOTSUP) if hardware doesn't support flow director mode.
2217 * - (-ENODEV) if *port_id* invalid.
2218 * - (-ENOSYS) if the flow director mode is not configured in perfect mode
2220 * - (-EINVAL) if the fdir_filter information is not correct.
2222 int rte_eth_dev_fdir_add_perfect_filter(uint8_t port_id,
2223 struct rte_fdir_filter *fdir_filter,
2224 uint16_t soft_id, uint8_t rx_queue,
2228 * Update a perfect filter rule on an Ethernet device.
2229 * If the rule doesn't exits, it is created.
2232 * The port identifier of the Ethernet device.
2233 * @param fdir_filter
2234 * The pointer to the structure describing the perfect filter rule.
2235 * The *rte_fdir_filter* structure includes the values of the different fields
2236 * to match: source and destination IP addresses, vlan id, flexbytes, source
2237 * and destination ports, and so on.
2238 * IPv6 are not supported.
2240 * The 16-bit value supplied in the field hash.fdir.id of mbuf for RX
2241 * packets matching the perfect filter.
2243 * The index of the RX queue where to store RX packets matching the added
2244 * perfect filter defined in fdir_filter.
2246 * If drop is set to 1, matching RX packets are stored into the RX drop
2247 * queue defined in the rte_fdir_conf.
2249 * - (0) if successful.
2250 * - (-ENOTSUP) if hardware doesn't support flow director mode.
2251 * - (-ENODEV) if *port_id* invalid.
2252 * - (-ENOSYS) if the flow director mode is not configured in perfect mode
2254 * - (-EINVAL) if the fdir_filter information is not correct.
2256 int rte_eth_dev_fdir_update_perfect_filter(uint8_t port_id,
2257 struct rte_fdir_filter *fdir_filter,
2258 uint16_t soft_id, uint8_t rx_queue,
2262 * Remove a perfect filter rule on an Ethernet device.
2265 * The port identifier of the Ethernet device.
2266 * @param fdir_filter
2267 * The pointer to the structure describing the perfect filter rule.
2268 * The *rte_fdir_filter* structure includes the values of the different fields
2269 * to match: source and destination IP addresses, vlan id, flexbytes, source
2270 * and destination ports, and so on.
2271 * IPv6 are not supported.
2273 * The soft_id value provided when adding/updating the removed filter.
2275 * - (0) if successful.
2276 * - (-ENOTSUP) if hardware doesn't support flow director mode.
2277 * - (-ENODEV) if *port_id* invalid.
2278 * - (-ENOSYS) if the flow director mode is not configured in perfect mode
2280 * - (-EINVAL) if the fdir_filter information is not correct.
2282 int rte_eth_dev_fdir_remove_perfect_filter(uint8_t port_id,
2283 struct rte_fdir_filter *fdir_filter,
2286 * Configure globally the masks for flow director mode for an Ethernet device.
2287 * For example, the device can match packets with only the first 24 bits of
2288 * the IPv4 source address.
2290 * The following fields can be masked: IPv4 addresses and L4 port numbers.
2291 * The following fields can be either enabled or disabled completely for the
2292 * matching functionality: VLAN ID tag; VLAN Priority + CFI bit; Flexible 2-byte
2294 * IPv6 masks are not supported.
2296 * All filters must comply with the masks previously configured.
2297 * For example, with a mask equal to 255.255.255.0 for the source IPv4 address,
2298 * all IPv4 filters must be created with a source IPv4 address that fits the
2301 * This function flushes all filters that have been previously added in
2305 * The port identifier of the Ethernet device.
2307 * The pointer to the fdir mask structure describing relevant headers fields
2308 * and relevant bits to use when matching packets addresses and ports.
2309 * IPv6 masks are not supported.
2311 * - (0) if successful.
2312 * - (-ENOTSUP) if hardware doesn't support flow director mode.
2313 * - (-ENODEV) if *port_id* invalid.
2314 * - (-ENOSYS) if the flow director mode is not configured in perfect
2315 * mode on *port_id*.
2316 * - (-EINVAL) if the fdir_filter information is not correct
2318 int rte_eth_dev_fdir_set_masks(uint8_t port_id,
2319 struct rte_fdir_masks *fdir_mask);
2322 * The eth device event type for interrupt, and maybe others in the future.
2324 enum rte_eth_event_type {
2325 RTE_ETH_EVENT_UNKNOWN, /**< unknown event type */
2326 RTE_ETH_EVENT_INTR_LSC, /**< lsc interrupt event */
2327 RTE_ETH_EVENT_MAX /**< max value of this enum */
2330 typedef void (*rte_eth_dev_cb_fn)(uint8_t port_id, \
2331 enum rte_eth_event_type event, void *cb_arg);
2332 /**< user application callback to be registered for interrupts */
2337 * Register a callback function for specific port id.
2344 * User supplied callback function to be called.
2346 * Pointer to the parameters for the registered callback.
2349 * - On success, zero.
2350 * - On failure, a negative value.
2352 int rte_eth_dev_callback_register(uint8_t port_id,
2353 enum rte_eth_event_type event,
2354 rte_eth_dev_cb_fn cb_fn, void *cb_arg);
2357 * Unregister a callback function for specific port id.
2364 * User supplied callback function to be called.
2366 * Pointer to the parameters for the registered callback. -1 means to
2367 * remove all for the same callback address and same event.
2370 * - On success, zero.
2371 * - On failure, a negative value.
2373 int rte_eth_dev_callback_unregister(uint8_t port_id,
2374 enum rte_eth_event_type event,
2375 rte_eth_dev_cb_fn cb_fn, void *cb_arg);
2378 * @internal Executes all the user application registered callbacks for
2379 * the specific device. It is for DPDK internal user only. User
2380 * application should not call it directly.
2383 * Pointer to struct rte_eth_dev.
2385 * Eth device interrupt event type.
2390 void _rte_eth_dev_callback_process(struct rte_eth_dev *dev,
2391 enum rte_eth_event_type event);
2394 * Turn on the LED on the Ethernet device.
2395 * This function turns on the LED on the Ethernet device.
2398 * The port identifier of the Ethernet device.
2400 * - (0) if successful.
2401 * - (-ENOTSUP) if underlying hardware OR driver doesn't support
2403 * - (-ENODEV) if *port_id* invalid.
2405 int rte_eth_led_on(uint8_t port_id);
2408 * Turn off the LED on the Ethernet device.
2409 * This function turns off the LED on the Ethernet device.
2412 * The port identifier of the Ethernet device.
2414 * - (0) if successful.
2415 * - (-ENOTSUP) if underlying hardware OR driver doesn't support
2417 * - (-ENODEV) if *port_id* invalid.
2419 int rte_eth_led_off(uint8_t port_id);
2422 * Configure the Ethernet link flow control for Ethernet device
2425 * The port identifier of the Ethernet device.
2427 * The pointer to the structure of the flow control parameters.
2429 * - (0) if successful.
2430 * - (-ENOTSUP) if hardware doesn't support flow control mode.
2431 * - (-ENODEV) if *port_id* invalid.
2432 * - (-EINVAL) if bad parameter
2433 * - (-EIO) if flow control setup failure
2435 int rte_eth_dev_flow_ctrl_set(uint8_t port_id,
2436 struct rte_eth_fc_conf *fc_conf);
2439 * Configure the Ethernet priority flow control under DCB environment
2440 * for Ethernet device.
2443 * The port identifier of the Ethernet device.
2445 * The pointer to the structure of the priority flow control parameters.
2447 * - (0) if successful.
2448 * - (-ENOTSUP) if hardware doesn't support priority flow control mode.
2449 * - (-ENODEV) if *port_id* invalid.
2450 * - (-EINVAL) if bad parameter
2451 * - (-EIO) if flow control setup failure
2453 int rte_eth_dev_priority_flow_ctrl_set(uint8_t port_id,
2454 struct rte_eth_pfc_conf *pfc_conf);
2457 * Add a MAC address to an internal array of addresses used to enable whitelist
2458 * filtering to accept packets only if the destination MAC address matches.
2461 * The port identifier of the Ethernet device.
2463 * The MAC address to add.
2465 * VMDq pool index to associate address with (if VMDq is enabled). If VMDq is
2466 * not enabled, this should be set to 0.
2468 * - (0) if successfully added or *mac_addr" was already added.
2469 * - (-ENOTSUP) if hardware doesn't support this feature.
2470 * - (-ENODEV) if *port* is invalid.
2471 * - (-ENOSPC) if no more MAC addresses can be added.
2472 * - (-EINVAL) if MAC address is invalid.
2474 int rte_eth_dev_mac_addr_add(uint8_t port, struct ether_addr *mac_addr,
2478 * Remove a MAC address from the internal array of addresses.
2481 * The port identifier of the Ethernet device.
2483 * MAC address to remove.
2485 * - (0) if successful, or *mac_addr* didn't exist.
2486 * - (-ENOTSUP) if hardware doesn't support.
2487 * - (-ENODEV) if *port* invalid.
2488 * - (-EADDRINUSE) if attempting to remove the default MAC address
2490 int rte_eth_dev_mac_addr_remove(uint8_t port, struct ether_addr *mac_addr);
2493 * Update Redirection Table(RETA) of Receive Side Scaling of Ethernet device.
2496 * The port identifier of the Ethernet device.
2500 * - (0) if successful.
2501 * - (-ENOTSUP) if hardware doesn't support.
2502 * - (-EINVAL) if bad parameter.
2504 int rte_eth_dev_rss_reta_update(uint8_t port,
2505 struct rte_eth_rss_reta *reta_conf);
2508 * Query Redirection Table(RETA) of Receive Side Scaling of Ethernet device.
2511 * The port identifier of the Ethernet device.
2515 * - (0) if successful.
2516 * - (-ENOTSUP) if hardware doesn't support.
2517 * - (-EINVAL) if bad parameter.
2519 int rte_eth_dev_rss_reta_query(uint8_t port,
2520 struct rte_eth_rss_reta *reta_conf);
2523 * Updates unicast hash table for receiving packet with the given destination
2524 * MAC address, and the packet is routed to all VFs for which the RX mode is
2525 * accept packets that match the unicast hash table.
2528 * The port identifier of the Ethernet device.
2530 * Unicast MAC address.
2532 * 1 - Set an unicast hash bit for receiving packets with the MAC address.
2533 * 0 - Clear an unicast hash bit.
2535 * - (0) if successful.
2536 * - (-ENOTSUP) if hardware doesn't support.
2537 * - (-ENODEV) if *port_id* invalid.
2538 * - (-EINVAL) if bad parameter.
2540 int rte_eth_dev_uc_hash_table_set(uint8_t port,struct ether_addr *addr,
2544 * Updates all unicast hash bitmaps for receiving packet with any Unicast
2545 * Ethernet MAC addresses,the packet is routed to all VFs for which the RX
2546 * mode is accept packets that match the unicast hash table.
2549 * The port identifier of the Ethernet device.
2551 * 1 - Set all unicast hash bitmaps for receiving all the Ethernet
2553 * 0 - Clear all unicast hash bitmaps
2555 * - (0) if successful.
2556 * - (-ENOTSUP) if hardware doesn't support.
2557 * - (-ENODEV) if *port_id* invalid.
2558 * - (-EINVAL) if bad parameter.
2560 int rte_eth_dev_uc_all_hash_table_set(uint8_t port,uint8_t on);
2563 * Set RX L2 Filtering mode of a VF of an Ethernet device.
2566 * The port identifier of the Ethernet device.
2570 * The RX mode mask, which is one or more of accepting Untagged Packets,
2571 * packets that match the PFUTA table, Broadcast and Multicast Promiscuous.
2572 * ETH_VMDQ_ACCEPT_UNTAG,ETH_VMDQ_ACCEPT_HASH_UC,
2573 * ETH_VMDQ_ACCEPT_BROADCAST and ETH_VMDQ_ACCEPT_MULTICAST will be used
2576 * 1 - Enable a VF RX mode.
2577 * 0 - Disable a VF RX mode.
2579 * - (0) if successful.
2580 * - (-ENOTSUP) if hardware doesn't support.
2581 * - (-ENOTSUP) if hardware doesn't support.
2582 * - (-EINVAL) if bad parameter.
2584 int rte_eth_dev_set_vf_rxmode(uint8_t port, uint16_t vf, uint16_t rx_mode,
2588 * Enable or disable a VF traffic transmit of the Ethernet device.
2591 * The port identifier of the Ethernet device.
2595 * 1 - Enable a VF traffic transmit.
2596 * 0 - Disable a VF traffic transmit.
2598 * - (0) if successful.
2599 * - (-ENODEV) if *port_id* invalid.
2600 * - (-ENOTSUP) if hardware doesn't support.
2601 * - (-EINVAL) if bad parameter.
2604 rte_eth_dev_set_vf_tx(uint8_t port,uint16_t vf, uint8_t on);
2607 * Enable or disable a VF traffic receive of an Ethernet device.
2610 * The port identifier of the Ethernet device.
2614 * 1 - Enable a VF traffic receive.
2615 * 0 - Disable a VF traffic receive.
2617 * - (0) if successful.
2618 * - (-ENOTSUP) if hardware doesn't support.
2619 * - (-ENODEV) if *port_id* invalid.
2620 * - (-EINVAL) if bad parameter.
2623 rte_eth_dev_set_vf_rx(uint8_t port,uint16_t vf, uint8_t on);
2626 * Enable/Disable hardware VF VLAN filtering by an Ethernet device of
2627 * received VLAN packets tagged with a given VLAN Tag Identifier.
2630 * The port identifier of the Ethernet device.
2632 * The VLAN Tag Identifier whose filtering must be enabled or disabled.
2634 * Bitmap listing which VFs participate in the VLAN filtering.
2636 * 1 - Enable VFs VLAN filtering.
2637 * 0 - Disable VFs VLAN filtering.
2639 * - (0) if successful.
2640 * - (-ENOTSUP) if hardware doesn't support.
2641 * - (-ENODEV) if *port_id* invalid.
2642 * - (-EINVAL) if bad parameter.
2645 rte_eth_dev_set_vf_vlan_filter(uint8_t port, uint16_t vlan_id,
2650 * Set a traffic mirroring rule on an Ethernet device
2653 * The port identifier of the Ethernet device.
2654 * @param mirror_conf
2655 * The pointer to the traffic mirroring structure describing the mirroring rule.
2656 * The *rte_eth_vm_mirror_conf* structure includes the type of mirroring rule,
2657 * destination pool and the value of rule if enable vlan or pool mirroring.
2660 * The index of traffic mirroring rule, we support four separated rules.
2662 * 1 - Enable a mirroring rule.
2663 * 0 - Disable a mirroring rule.
2665 * - (0) if successful.
2666 * - (-ENOTSUP) if hardware doesn't support this feature.
2667 * - (-ENODEV) if *port_id* invalid.
2668 * - (-EINVAL) if the mr_conf information is not correct.
2670 int rte_eth_mirror_rule_set(uint8_t port_id,
2671 struct rte_eth_vmdq_mirror_conf *mirror_conf,
2676 * Reset a traffic mirroring rule on an Ethernet device.
2679 * The port identifier of the Ethernet device.
2681 * The index of traffic mirroring rule, we support four separated rules.
2683 * - (0) if successful.
2684 * - (-ENOTSUP) if hardware doesn't support this feature.
2685 * - (-ENODEV) if *port_id* invalid.
2686 * - (-EINVAL) if bad parameter.
2688 int rte_eth_mirror_rule_reset(uint8_t port_id,
2692 * Initialize bypass logic. This function needs to be called before
2693 * executing any other bypass API.
2696 * The port identifier of the Ethernet device.
2698 * - (0) if successful.
2699 * - (-ENOTSUP) if hardware doesn't support.
2700 * - (-EINVAL) if bad parameter.
2702 int rte_eth_dev_bypass_init(uint8_t port);
2705 * Return bypass state.
2708 * The port identifier of the Ethernet device.
2710 * The return bypass state.
2713 * - (3) Isolate mode
2715 * - (0) if successful.
2716 * - (-ENOTSUP) if hardware doesn't support.
2717 * - (-EINVAL) if bad parameter.
2719 int rte_eth_dev_bypass_state_show(uint8_t port, uint32_t *state);
2725 * The port identifier of the Ethernet device.
2727 * The current bypass state.
2730 * - (3) Isolate mode
2732 * - (0) if successful.
2733 * - (-ENOTSUP) if hardware doesn't support.
2734 * - (-EINVAL) if bad parameter.
2736 int rte_eth_dev_bypass_state_set(uint8_t port, uint32_t *new_state);
2739 * Return bypass state when given event occurs.
2742 * The port identifier of the Ethernet device.
2745 * - (1) Main power on (power button is pushed)
2746 * - (2) Auxiliary power on (power supply is being plugged)
2747 * - (3) Main power off (system shutdown and power supply is left plugged in)
2748 * - (4) Auxiliary power off (power supply is being unplugged)
2749 * - (5) Display or set the watchdog timer
2751 * The bypass state when given event occurred.
2754 * - (3) Isolate mode
2756 * - (0) if successful.
2757 * - (-ENOTSUP) if hardware doesn't support.
2758 * - (-EINVAL) if bad parameter.
2760 int rte_eth_dev_bypass_event_show(uint8_t port, uint32_t event, uint32_t *state);
2763 * Set bypass state when given event occurs.
2766 * The port identifier of the Ethernet device.
2769 * - (1) Main power on (power button is pushed)
2770 * - (2) Auxiliary power on (power supply is being plugged)
2771 * - (3) Main power off (system shutdown and power supply is left plugged in)
2772 * - (4) Auxiliary power off (power supply is being unplugged)
2773 * - (5) Display or set the watchdog timer
2775 * The assigned state when given event occurs.
2778 * - (3) Isolate mode
2780 * - (0) if successful.
2781 * - (-ENOTSUP) if hardware doesn't support.
2782 * - (-EINVAL) if bad parameter.
2784 int rte_eth_dev_bypass_event_store(uint8_t port, uint32_t event, uint32_t state);
2787 * Set bypass watchdog timeout count.
2790 * The port identifier of the Ethernet device.
2792 * The timeout to be set.
2793 * - (0) 0 seconds (timer is off)
2802 * - (0) if successful.
2803 * - (-ENOTSUP) if hardware doesn't support.
2804 * - (-EINVAL) if bad parameter.
2806 int rte_eth_dev_wd_timeout_store(uint8_t port, uint32_t timeout);
2809 * Get bypass firmware version.
2812 * The port identifier of the Ethernet device.
2814 * The firmware version
2816 * - (0) if successful.
2817 * - (-ENOTSUP) if hardware doesn't support.
2818 * - (-EINVAL) if bad parameter.
2820 int rte_eth_dev_bypass_ver_show(uint8_t port, uint32_t *ver);
2823 * Return bypass watchdog timeout in seconds
2826 * The port identifier of the Ethernet device.
2828 * The return watchdog timeout. "0" represents timer expired
2829 * - (0) 0 seconds (timer is off)
2838 * - (0) if successful.
2839 * - (-ENOTSUP) if hardware doesn't support.
2840 * - (-EINVAL) if bad parameter.
2842 int rte_eth_dev_bypass_wd_timeout_show(uint8_t port, uint32_t *wd_timeout);
2845 * Reset bypass watchdog timer
2848 * The port identifier of the Ethernet device.
2850 * - (0) if successful.
2851 * - (-ENOTSUP) if hardware doesn't support.
2852 * - (-EINVAL) if bad parameter.
2854 int rte_eth_dev_bypass_wd_reset(uint8_t port);
2860 #endif /* _RTE_ETHDEV_H_ */