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 tx_pause_xon; /**< Total nb. of XON pause frame sent. */
196 uint64_t rx_pause_xon; /**< Total nb. of XON pause frame received. */
197 uint64_t tx_pause_xoff; /**< Total nb. of XOFF pause frame sent. */
198 uint64_t rx_pause_xoff; /**< Total nb. of XOFF pause frame received. */
199 uint64_t q_ipackets[RTE_ETHDEV_QUEUE_STAT_CNTRS];
200 /**< Total number of queue RX packets. */
201 uint64_t q_opackets[RTE_ETHDEV_QUEUE_STAT_CNTRS];
202 /**< Total number of queue TX packets. */
203 uint64_t q_ibytes[RTE_ETHDEV_QUEUE_STAT_CNTRS];
204 /**< Total number of successfully received queue bytes. */
205 uint64_t q_obytes[RTE_ETHDEV_QUEUE_STAT_CNTRS];
206 /**< Total number of successfully transmitted queue bytes. */
207 uint64_t q_errors[RTE_ETHDEV_QUEUE_STAT_CNTRS];
208 /**< Total number of queue packets received that are dropped. */
210 /**< Total number of good packets received from loopback,VF Only */
212 /**< Total number of good packets transmitted to loopback,VF Only */
214 /**< Total number of good bytes received from loopback,VF Only */
216 /**< Total number of good bytes transmitted to loopback,VF Only */
220 * A structure used to retrieve link-level information of an Ethernet port.
222 struct rte_eth_link {
223 uint16_t link_speed; /**< ETH_LINK_SPEED_[10, 100, 1000, 10000] */
224 uint16_t link_duplex; /**< ETH_LINK_[HALF_DUPLEX, FULL_DUPLEX] */
225 uint8_t link_status : 1; /**< 1 -> link up, 0 -> link down */
226 }__attribute__((aligned(8))); /**< aligned for atomic64 read/write */
228 #define ETH_LINK_SPEED_AUTONEG 0 /**< Auto-negotiate link speed. */
229 #define ETH_LINK_SPEED_10 10 /**< 10 megabits/second. */
230 #define ETH_LINK_SPEED_100 100 /**< 100 megabits/second. */
231 #define ETH_LINK_SPEED_1000 1000 /**< 1 gigabits/second. */
232 #define ETH_LINK_SPEED_10000 10000 /**< 10 gigabits/second. */
234 #define ETH_LINK_AUTONEG_DUPLEX 0 /**< Auto-negotiate duplex. */
235 #define ETH_LINK_HALF_DUPLEX 1 /**< Half-duplex connection. */
236 #define ETH_LINK_FULL_DUPLEX 2 /**< Full-duplex connection. */
239 * A structure used to configure the ring threshold registers of an RX/TX
240 * queue for an Ethernet port.
242 struct rte_eth_thresh {
243 uint8_t pthresh; /**< Ring prefetch threshold. */
244 uint8_t hthresh; /**< Ring host threshold. */
245 uint8_t wthresh; /**< Ring writeback threshold. */
249 * A set of values to identify what method is to be used to route
250 * packets to multiple queues.
252 enum rte_eth_rx_mq_mode {
253 ETH_MQ_RX_NONE = 0, /**< None of DCB,RSS or VMDQ mode */
255 ETH_MQ_RX_RSS, /**< For RX side, only RSS is on */
256 ETH_MQ_RX_DCB, /**< For RX side,only DCB is on. */
257 ETH_MQ_RX_DCB_RSS, /**< Both DCB and RSS enable */
259 ETH_MQ_RX_VMDQ_ONLY, /**< Only VMDQ, no RSS nor DCB */
260 ETH_MQ_RX_VMDQ_RSS, /**< RSS mode with VMDQ */
261 ETH_MQ_RX_VMDQ_DCB, /**< Use VMDQ+DCB to route traffic to queues */
262 ETH_MQ_RX_VMDQ_DCB_RSS, /**< Enable both VMDQ and DCB in VMDq */
266 * for rx mq mode backward compatible
268 #define ETH_RSS ETH_MQ_RX_RSS
269 #define VMDQ_DCB ETH_MQ_RX_VMDQ_DCB
270 #define ETH_DCB_RX ETH_MQ_RX_DCB
273 * A set of values to identify what method is to be used to transmit
274 * packets using multi-TCs.
276 enum rte_eth_tx_mq_mode {
277 ETH_MQ_TX_NONE = 0, /**< It is in neither DCB nor VT mode. */
278 ETH_MQ_TX_DCB, /**< For TX side,only DCB is on. */
279 ETH_MQ_TX_VMDQ_DCB, /**< For TX side,both DCB and VT is on. */
280 ETH_MQ_TX_VMDQ_ONLY, /**< Only VT on, no DCB */
284 * for tx mq mode backward compatible
286 #define ETH_DCB_NONE ETH_MQ_TX_NONE
287 #define ETH_VMDQ_DCB_TX ETH_MQ_TX_VMDQ_DCB
288 #define ETH_DCB_TX ETH_MQ_TX_DCB
291 * A structure used to configure the RX features of an Ethernet port.
293 struct rte_eth_rxmode {
294 /** The multi-queue packet distribution mode to be used, e.g. RSS. */
295 enum rte_eth_rx_mq_mode mq_mode;
296 uint32_t max_rx_pkt_len; /**< Only used if jumbo_frame enabled. */
297 uint16_t split_hdr_size; /**< hdr buf size (header_split enabled).*/
298 uint8_t header_split : 1, /**< Header Split enable. */
299 hw_ip_checksum : 1, /**< IP/UDP/TCP checksum offload enable. */
300 hw_vlan_filter : 1, /**< VLAN filter enable. */
301 hw_vlan_strip : 1, /**< VLAN strip enable. */
302 hw_vlan_extend : 1, /**< Extended VLAN enable. */
303 jumbo_frame : 1, /**< Jumbo Frame Receipt enable. */
304 hw_strip_crc : 1; /**< Enable CRC stripping by hardware. */
308 * A structure used to configure the Receive Side Scaling (RSS) feature
309 * of an Ethernet port.
310 * If not NULL, the *rss_key* pointer of the *rss_conf* structure points
311 * to an array of 40 bytes holding the RSS key to use for hashing specific
312 * header fields of received packets.
313 * Otherwise, a default random hash key is used by the device driver.
315 * The *rss_hf* field of the *rss_conf* structure indicates the different
316 * types of IPv4/IPv6 packets to which the RSS hashing must be applied.
317 * Supplying an *rss_hf* equal to zero disables the RSS feature.
319 struct rte_eth_rss_conf {
320 uint8_t *rss_key; /**< If not NULL, 40-byte hash key. */
321 uint16_t rss_hf; /**< Hash functions to apply - see below. */
324 #define ETH_RSS_IPV4 0x0001 /**< IPv4 packet. */
325 #define ETH_RSS_IPV4_TCP 0x0002 /**< IPv4/TCP packet. */
326 #define ETH_RSS_IPV6 0x0004 /**< IPv6 packet. */
327 #define ETH_RSS_IPV6_EX 0x0008 /**< IPv6 packet with extension headers.*/
328 #define ETH_RSS_IPV6_TCP 0x0010 /**< IPv6/TCP packet. */
329 #define ETH_RSS_IPV6_TCP_EX 0x0020 /**< IPv6/TCP with extension headers. */
330 /* Intel RSS extensions to UDP packets */
331 #define ETH_RSS_IPV4_UDP 0x0040 /**< IPv4/UDP packet. */
332 #define ETH_RSS_IPV6_UDP 0x0080 /**< IPv6/UDP packet. */
333 #define ETH_RSS_IPV6_UDP_EX 0x0100 /**< IPv6/UDP with extension headers. */
335 #define ETH_RSS_PROTO_MASK 0x01FF /**< Mask of valid RSS hash protocols */
336 /* Definitions used for redirection table entry size */
337 #define ETH_RSS_RETA_NUM_ENTRIES 128
338 #define ETH_RSS_RETA_MAX_QUEUE 16
340 /* Definitions used for VMDQ and DCB functionality */
341 #define ETH_VMDQ_MAX_VLAN_FILTERS 64 /**< Maximum nb. of VMDQ vlan filters. */
342 #define ETH_DCB_NUM_USER_PRIORITIES 8 /**< Maximum nb. of DCB priorities. */
343 #define ETH_VMDQ_DCB_NUM_QUEUES 128 /**< Maximum nb. of VMDQ DCB queues. */
344 #define ETH_DCB_NUM_QUEUES 128 /**< Maximum nb. of DCB queues. */
346 /* DCB capability defines */
347 #define ETH_DCB_PG_SUPPORT 0x00000001 /**< Priority Group(ETS) support. */
348 #define ETH_DCB_PFC_SUPPORT 0x00000002 /**< Priority Flow Control support. */
350 /* Definitions used for VLAN Offload functionality */
351 #define ETH_VLAN_STRIP_OFFLOAD 0x0001 /**< VLAN Strip On/Off */
352 #define ETH_VLAN_FILTER_OFFLOAD 0x0002 /**< VLAN Filter On/Off */
353 #define ETH_VLAN_EXTEND_OFFLOAD 0x0004 /**< VLAN Extend On/Off */
355 /* Definitions used for mask VLAN setting */
356 #define ETH_VLAN_STRIP_MASK 0x0001 /**< VLAN Strip setting mask */
357 #define ETH_VLAN_FILTER_MASK 0x0002 /**< VLAN Filter setting mask*/
358 #define ETH_VLAN_EXTEND_MASK 0x0004 /**< VLAN Extend setting mask*/
359 #define ETH_VLAN_ID_MAX 0x0FFF /**< VLAN ID is in lower 12 bits*/
361 /* Definitions used for receive MAC address */
362 #define ETH_NUM_RECEIVE_MAC_ADDR 128 /**< Maximum nb. of receive mac addr. */
365 /* Definitions used for unicast hash */
366 #define ETH_VMDQ_NUM_UC_HASH_ARRAY 128 /**< Maximum nb. of UC hash array. */
368 /* Definitions used for VMDQ pool rx mode setting */
369 #define ETH_VMDQ_ACCEPT_UNTAG 0x0001 /**< accept untagged packets. */
370 #define ETH_VMDQ_ACCEPT_HASH_MC 0x0002 /**< accept packets in multicast table . */
371 #define ETH_VMDQ_ACCEPT_HASH_UC 0x0004 /**< accept packets in unicast table. */
372 #define ETH_VMDQ_ACCEPT_BROADCAST 0x0008 /**< accept broadcast packets. */
373 #define ETH_VMDQ_ACCEPT_MULTICAST 0x0010 /**< multicast promiscuous. */
375 /* Definitions used for VMDQ mirror rules setting */
376 #define ETH_VMDQ_NUM_MIRROR_RULE 4 /**< Maximum nb. of mirror rules. . */
378 #define ETH_VMDQ_POOL_MIRROR 0x0001 /**< Virtual Pool Mirroring. */
379 #define ETH_VMDQ_UPLINK_MIRROR 0x0002 /**< Uplink Port Mirroring. */
380 #define ETH_VMDQ_DOWNLIN_MIRROR 0x0004 /**< Downlink Port Mirroring. */
381 #define ETH_VMDQ_VLAN_MIRROR 0x0008 /**< VLAN Mirroring. */
384 * A structure used to configure VLAN traffic mirror of an Ethernet port.
386 struct rte_eth_vlan_mirror {
387 uint64_t vlan_mask; /**< mask for valid VLAN ID. */
388 uint16_t vlan_id[ETH_VMDQ_MAX_VLAN_FILTERS];
389 /** VLAN ID list for vlan mirror. */
393 * A structure used to configure traffic mirror of an Ethernet port.
395 struct rte_eth_vmdq_mirror_conf {
396 uint8_t rule_type_mask; /**< Mirroring rule type mask we want to set */
397 uint8_t dst_pool; /**< Destination pool for this mirror rule. */
398 uint64_t pool_mask; /**< Bitmap of pool for pool mirroring */
399 struct rte_eth_vlan_mirror vlan; /**< VLAN ID setting for VLAN mirroring */
403 * A structure used to configure Redirection Table of the Receive Side
404 * Scaling (RSS) feature of an Ethernet port.
406 struct rte_eth_rss_reta {
407 /** First 64 mask bits indicate which entry(s) need to updated/queried. */
409 /** Second 64 mask bits indicate which entry(s) need to updated/queried. */
411 uint8_t reta[ETH_RSS_RETA_NUM_ENTRIES]; /**< 128 RETA entries*/
415 * This enum indicates the possible number of traffic classes
416 * in DCB configratioins
418 enum rte_eth_nb_tcs {
419 ETH_4_TCS = 4, /**< 4 TCs with DCB. */
420 ETH_8_TCS = 8 /**< 8 TCs with DCB. */
424 * This enum indicates the possible number of queue pools
425 * in VMDQ configurations.
427 enum rte_eth_nb_pools {
428 ETH_8_POOLS = 8, /**< 8 VMDq pools. */
429 ETH_16_POOLS = 16, /**< 16 VMDq pools. */
430 ETH_32_POOLS = 32, /**< 32 VMDq pools. */
431 ETH_64_POOLS = 64 /**< 64 VMDq pools. */
434 /* This structure may be extended in future. */
435 struct rte_eth_dcb_rx_conf {
436 enum rte_eth_nb_tcs nb_tcs; /**< Possible DCB TCs, 4 or 8 TCs */
437 uint8_t dcb_queue[ETH_DCB_NUM_USER_PRIORITIES];
438 /**< Possible DCB queue,4 or 8. */
441 struct rte_eth_vmdq_dcb_tx_conf {
442 enum rte_eth_nb_pools nb_queue_pools; /**< With DCB, 16 or 32 pools. */
443 uint8_t dcb_queue[ETH_DCB_NUM_USER_PRIORITIES];
444 /**< Possible DCB queue,4 or 8. */
447 struct rte_eth_dcb_tx_conf {
448 enum rte_eth_nb_tcs nb_tcs; /**< Possible DCB TCs, 4 or 8 TCs. */
449 uint8_t dcb_queue[ETH_DCB_NUM_USER_PRIORITIES];
450 /**< Possible DCB queue,4 or 8. */
453 struct rte_eth_vmdq_tx_conf {
454 enum rte_eth_nb_pools nb_queue_pools; /**< VMDq mode, 64 pools. */
458 * A structure used to configure the VMDQ+DCB feature
459 * of an Ethernet port.
461 * Using this feature, packets are routed to a pool of queues, based
462 * on the vlan id in the vlan tag, and then to a specific queue within
463 * that pool, using the user priority vlan tag field.
465 * A default pool may be used, if desired, to route all traffic which
466 * does not match the vlan filter rules.
468 struct rte_eth_vmdq_dcb_conf {
469 enum rte_eth_nb_pools nb_queue_pools; /**< With DCB, 16 or 32 pools */
470 uint8_t enable_default_pool; /**< If non-zero, use a default pool */
471 uint8_t default_pool; /**< The default pool, if applicable */
472 uint8_t nb_pool_maps; /**< We can have up to 64 filters/mappings */
474 uint16_t vlan_id; /**< The vlan id of the received frame */
475 uint64_t pools; /**< Bitmask of pools for packet rx */
476 } pool_map[ETH_VMDQ_MAX_VLAN_FILTERS]; /**< VMDq vlan pool maps. */
477 uint8_t dcb_queue[ETH_DCB_NUM_USER_PRIORITIES];
478 /**< Selects a queue in a pool */
481 struct rte_eth_vmdq_rx_conf {
482 enum rte_eth_nb_pools nb_queue_pools; /**< VMDq only mode, 8 or 64 pools */
483 uint8_t enable_default_pool; /**< If non-zero, use a default pool */
484 uint8_t default_pool; /**< The default pool, if applicable */
485 uint8_t enable_loop_back; /**< Enable VT loop back */
486 uint8_t nb_pool_maps; /**< We can have up to 64 filters/mappings */
488 uint16_t vlan_id; /**< The vlan id of the received frame */
489 uint64_t pools; /**< Bitmask of pools for packet rx */
490 } pool_map[ETH_VMDQ_MAX_VLAN_FILTERS]; /**< VMDq vlan pool maps. */
494 * A structure used to configure the TX features of an Ethernet port.
496 struct rte_eth_txmode {
497 enum rte_eth_tx_mq_mode mq_mode; /**< TX multi-queues mode. */
501 * A structure used to configure an RX ring of an Ethernet port.
503 struct rte_eth_rxconf {
504 struct rte_eth_thresh rx_thresh; /**< RX ring threshold registers. */
505 uint16_t rx_free_thresh; /**< Drives the freeing of RX descriptors. */
506 uint8_t rx_drop_en; /**< Drop packets if no descriptors are available. */
507 uint8_t start_rx_per_q; /**< start rx per queue. */
510 #define ETH_TXQ_FLAGS_NOMULTSEGS 0x0001 /**< nb_segs=1 for all mbufs */
511 #define ETH_TXQ_FLAGS_NOREFCOUNT 0x0002 /**< refcnt can be ignored */
512 #define ETH_TXQ_FLAGS_NOMULTMEMP 0x0004 /**< all bufs come from same mempool */
513 #define ETH_TXQ_FLAGS_NOVLANOFFL 0x0100 /**< disable VLAN offload */
514 #define ETH_TXQ_FLAGS_NOXSUMSCTP 0x0200 /**< disable SCTP checksum offload */
515 #define ETH_TXQ_FLAGS_NOXSUMUDP 0x0400 /**< disable UDP checksum offload */
516 #define ETH_TXQ_FLAGS_NOXSUMTCP 0x0800 /**< disable TCP checksum offload */
517 #define ETH_TXQ_FLAGS_NOOFFLOADS \
518 (ETH_TXQ_FLAGS_NOVLANOFFL | ETH_TXQ_FLAGS_NOXSUMSCTP | \
519 ETH_TXQ_FLAGS_NOXSUMUDP | ETH_TXQ_FLAGS_NOXSUMTCP)
521 * A structure used to configure a TX ring of an Ethernet port.
523 struct rte_eth_txconf {
524 struct rte_eth_thresh tx_thresh; /**< TX ring threshold registers. */
525 uint16_t tx_rs_thresh; /**< Drives the setting of RS bit on TXDs. */
526 uint16_t tx_free_thresh; /**< Drives the freeing of TX buffers. */
527 uint32_t txq_flags; /**< Set flags for the Tx queue */
528 uint8_t start_tx_per_q; /**< start tx per queue. */
532 * This enum indicates the flow control mode
534 enum rte_eth_fc_mode {
535 RTE_FC_NONE = 0, /**< Disable flow control. */
536 RTE_FC_RX_PAUSE, /**< RX pause frame, enable flowctrl on TX side. */
537 RTE_FC_TX_PAUSE, /**< TX pause frame, enable flowctrl on RX side. */
538 RTE_FC_FULL /**< Enable flow control on both side. */
542 * A structure used to configure Ethernet flow control parameter.
543 * These parameters will be configured into the register of the NIC.
544 * Please refer to the corresponding data sheet for proper value.
546 struct rte_eth_fc_conf {
547 uint32_t high_water; /**< High threshold value to trigger XOFF */
548 uint32_t low_water; /**< Low threshold value to trigger XON */
549 uint16_t pause_time; /**< Pause quota in the Pause frame */
550 uint16_t send_xon; /**< Is XON frame need be sent */
551 enum rte_eth_fc_mode mode; /**< Link flow control mode */
552 uint8_t mac_ctrl_frame_fwd; /**< Forward MAC control frames */
556 * A structure used to configure Ethernet priority flow control parameter.
557 * These parameters will be configured into the register of the NIC.
558 * Please refer to the corresponding data sheet for proper value.
560 struct rte_eth_pfc_conf {
561 struct rte_eth_fc_conf fc; /**< General flow control parameter. */
562 uint8_t priority; /**< VLAN User Priority. */
566 * Flow Director setting modes: none (default), signature or perfect.
569 RTE_FDIR_MODE_NONE = 0, /**< Disable FDIR support. */
570 RTE_FDIR_MODE_SIGNATURE, /**< Enable FDIR signature filter mode. */
571 RTE_FDIR_MODE_PERFECT, /**< Enable FDIR perfect filter mode. */
575 * Memory space that can be configured to store Flow Director filters
576 * in the board memory.
578 enum rte_fdir_pballoc_type {
579 RTE_FDIR_PBALLOC_64K = 0, /**< 64k. */
580 RTE_FDIR_PBALLOC_128K, /**< 128k. */
581 RTE_FDIR_PBALLOC_256K, /**< 256k. */
585 * Select report mode of FDIR hash information in RX descriptors.
587 enum rte_fdir_status_mode {
588 RTE_FDIR_NO_REPORT_STATUS = 0, /**< Never report FDIR hash. */
589 RTE_FDIR_REPORT_STATUS, /**< Only report FDIR hash for matching pkts. */
590 RTE_FDIR_REPORT_STATUS_ALWAYS, /**< Always report FDIR hash. */
594 * A structure used to configure the Flow Director (FDIR) feature
595 * of an Ethernet port.
597 * If mode is RTE_FDIR_DISABLE, the pballoc value is ignored.
599 struct rte_fdir_conf {
600 enum rte_fdir_mode mode; /**< Flow Director mode. */
601 enum rte_fdir_pballoc_type pballoc; /**< Space for FDIR filters. */
602 enum rte_fdir_status_mode status; /**< How to report FDIR hash. */
603 /** Offset of flexbytes field in RX packets (in 16-bit word units). */
604 uint8_t flexbytes_offset;
605 /** RX queue of packets matching a "drop" filter in perfect mode. */
610 * Possible l4type of FDIR filters.
613 RTE_FDIR_L4TYPE_NONE = 0, /**< None. */
614 RTE_FDIR_L4TYPE_UDP, /**< UDP. */
615 RTE_FDIR_L4TYPE_TCP, /**< TCP. */
616 RTE_FDIR_L4TYPE_SCTP, /**< SCTP. */
620 * Select IPv4 or IPv6 FDIR filters.
623 RTE_FDIR_IPTYPE_IPV4 = 0, /**< IPv4. */
624 RTE_FDIR_IPTYPE_IPV6 , /**< IPv6. */
628 * A structure used to define a FDIR packet filter.
630 struct rte_fdir_filter {
631 uint16_t flex_bytes; /**< Flex bytes value to match. */
632 uint16_t vlan_id; /**< VLAN ID value to match, 0 otherwise. */
633 uint16_t port_src; /**< Source port to match, 0 otherwise. */
634 uint16_t port_dst; /**< Destination port to match, 0 otherwise. */
636 uint32_t ipv4_addr; /**< IPv4 source address to match. */
637 uint32_t ipv6_addr[4]; /**< IPv6 source address to match. */
638 } ip_src; /**< IPv4/IPv6 source address to match (union of above). */
640 uint32_t ipv4_addr; /**< IPv4 destination address to match. */
641 uint32_t ipv6_addr[4]; /**< IPv6 destination address to match */
642 } ip_dst; /**< IPv4/IPv6 destination address to match (union of above). */
643 enum rte_l4type l4type; /**< l4type to match: NONE/UDP/TCP/SCTP. */
644 enum rte_iptype iptype; /**< IP packet type to match: IPv4 or IPv6. */
648 * A structure used to configure FDIR masks that are used by the device
649 * to match the various fields of RX packet headers.
650 * @note The only_ip_flow field has the opposite meaning compared to other
653 struct rte_fdir_masks {
654 /** When set to 1, packet l4type is \b NOT relevant in filters, and
655 source and destination port masks must be set to zero. */
656 uint8_t only_ip_flow;
657 /** If set to 1, vlan_id is relevant in filters. */
659 /** If set to 1, vlan_prio is relevant in filters. */
661 /** If set to 1, flexbytes is relevant in filters. */
663 /** If set to 1, set the IPv6 masks. Otherwise set the IPv4 masks. */
664 uint8_t set_ipv6_mask;
665 /** When set to 1, comparison of destination IPv6 address with IP6AT
666 registers is meaningful. */
667 uint8_t comp_ipv6_dst;
668 /** Mask of Destination IPv4 Address. All bits set to 1 define the
669 relevant bits to use in the destination address of an IPv4 packet
670 when matching it against FDIR filters. */
671 uint32_t dst_ipv4_mask;
672 /** Mask of Source IPv4 Address. All bits set to 1 define
673 the relevant bits to use in the source address of an IPv4 packet
674 when matching it against FDIR filters. */
675 uint32_t src_ipv4_mask;
676 /** Mask of Source IPv6 Address. All bits set to 1 define the
677 relevant BYTES to use in the source address of an IPv6 packet
678 when matching it against FDIR filters. */
679 uint16_t dst_ipv6_mask;
680 /** Mask of Destination IPv6 Address. All bits set to 1 define the
681 relevant BYTES to use in the destination address of an IPv6 packet
682 when matching it against FDIR filters. */
683 uint16_t src_ipv6_mask;
684 /** Mask of Source Port. All bits set to 1 define the relevant
685 bits to use in the source port of an IP packets when matching it
686 against FDIR filters. */
687 uint16_t src_port_mask;
688 /** Mask of Destination Port. All bits set to 1 define the relevant
689 bits to use in the destination port of an IP packet when matching it
690 against FDIR filters. */
691 uint16_t dst_port_mask;
695 * A structure used to report the status of the flow director filters in use.
697 struct rte_eth_fdir {
698 /** Number of filters with collision indication. */
700 /** Number of free (non programmed) filters. */
702 /** The Lookup hash value of the added filter that updated the value
703 of the MAXLEN field */
705 /** Longest linked list of filters in the table. */
707 /** Number of added filters. */
709 /** Number of removed filters. */
711 /** Number of failed added filters (no more space in device). */
713 /** Number of failed removed filters. */
718 * A structure used to enable/disable specific device interrupts.
720 struct rte_intr_conf {
721 /** enable/disable lsc interrupt. 0 (default) - disable, 1 enable */
726 * A structure used to configure an Ethernet port.
727 * Depending upon the RX multi-queue mode, extra advanced
728 * configuration settings may be needed.
730 struct rte_eth_conf {
732 /**< ETH_LINK_SPEED_10[0|00|000], or 0 for autonegotation */
733 uint16_t link_duplex;
734 /**< ETH_LINK_[HALF_DUPLEX|FULL_DUPLEX], or 0 for autonegotation */
735 struct rte_eth_rxmode rxmode; /**< Port RX configuration. */
736 struct rte_eth_txmode txmode; /**< Port TX configuration. */
737 uint32_t lpbk_mode; /**< Loopback operation mode. By default the value
738 is 0, meaning the loopback mode is disabled.
739 Read the datasheet of given ethernet controller
740 for details. The possible values of this field
741 are defined in implementation of each driver. */
743 struct rte_eth_rss_conf rss_conf; /**< Port RSS configuration */
744 struct rte_eth_vmdq_dcb_conf vmdq_dcb_conf;
745 /**< Port vmdq+dcb configuration. */
746 struct rte_eth_dcb_rx_conf dcb_rx_conf;
747 /**< Port dcb RX configuration. */
748 struct rte_eth_vmdq_rx_conf vmdq_rx_conf;
749 /**< Port vmdq RX configuration. */
750 } rx_adv_conf; /**< Port RX filtering configuration (union). */
752 struct rte_eth_vmdq_dcb_tx_conf vmdq_dcb_tx_conf;
753 /**< Port vmdq+dcb TX configuration. */
754 struct rte_eth_dcb_tx_conf dcb_tx_conf;
755 /**< Port dcb TX configuration. */
756 struct rte_eth_vmdq_tx_conf vmdq_tx_conf;
757 /**< Port vmdq TX configuration. */
758 } tx_adv_conf; /**< Port TX DCB configuration (union). */
759 /** Currently,Priority Flow Control(PFC) are supported,if DCB with PFC
760 is needed,and the variable must be set ETH_DCB_PFC_SUPPORT. */
761 uint32_t dcb_capability_en;
762 struct rte_fdir_conf fdir_conf; /**< FDIR configuration. */
763 struct rte_intr_conf intr_conf; /**< Interrupt mode configuration. */
767 * A structure used to retrieve the contextual information of
768 * an Ethernet device, such as the controlling driver of the device,
769 * its PCI context, etc...
773 * RX offload capabilities of a device.
775 #define DEV_RX_OFFLOAD_VLAN_STRIP 0x00000001
776 #define DEV_RX_OFFLOAD_IPV4_CKSUM 0x00000002
777 #define DEV_RX_OFFLOAD_UDP_CKSUM 0x00000004
778 #define DEV_RX_OFFLOAD_TCP_CKSUM 0x00000008
779 #define DEV_RX_OFFLOAD_TCP_LRO 0x00000010
782 * TX offload capabilities of a device.
784 #define DEV_TX_OFFLOAD_VLAN_INSERT 0x00000001
785 #define DEV_TX_OFFLOAD_IPV4_CKSUM 0x00000002
786 #define DEV_TX_OFFLOAD_UDP_CKSUM 0x00000004
787 #define DEV_TX_OFFLOAD_TCP_CKSUM 0x00000008
788 #define DEV_TX_OFFLOAD_SCTP_CKSUM 0x00000010
789 #define DEV_TX_OFFLOAD_TCP_TSO 0x00000020
790 #define DEV_TX_OFFLOAD_UDP_TSO 0x00000040
792 struct rte_eth_dev_info {
793 struct rte_pci_device *pci_dev; /**< Device PCI information. */
794 const char *driver_name; /**< Device Driver name. */
795 unsigned int if_index; /**< Index to bound host interface, or 0 if none.
796 Use if_indextoname() to translate into an interface name. */
797 uint32_t min_rx_bufsize; /**< Minimum size of RX buffer. */
798 uint32_t max_rx_pktlen; /**< Maximum configurable length of RX pkt. */
799 uint16_t max_rx_queues; /**< Maximum number of RX queues. */
800 uint16_t max_tx_queues; /**< Maximum number of TX queues. */
801 uint32_t max_mac_addrs; /**< Maximum number of MAC addresses. */
802 uint32_t max_hash_mac_addrs;
803 /** Maximum number of hash MAC addresses for MTA and UTA. */
804 uint16_t max_vfs; /**< Maximum number of VFs. */
805 uint16_t max_vmdq_pools; /**< Maximum number of VMDq pools. */
806 uint32_t rx_offload_capa; /**< Device RX offload capabilities. */
807 uint32_t tx_offload_capa; /**< Device TX offload capabilities. */
812 struct rte_eth_dev_callback;
813 /** @internal Structure to keep track of registered callbacks */
814 TAILQ_HEAD(rte_eth_dev_cb_list, rte_eth_dev_callback);
816 #define TCP_UGR_FLAG 0x20
817 #define TCP_ACK_FLAG 0x10
818 #define TCP_PSH_FLAG 0x08
819 #define TCP_RST_FLAG 0x04
820 #define TCP_SYN_FLAG 0x02
821 #define TCP_FIN_FLAG 0x01
822 #define TCP_FLAG_ALL 0x3F
825 * A structure used to define an ethertype filter.
827 struct rte_ethertype_filter {
828 uint16_t ethertype; /**< little endian. */
829 uint8_t priority_en; /**< compare priority enable. */
834 * A structure used to define an syn filter.
836 struct rte_syn_filter {
837 uint8_t hig_pri; /**< 1 means higher pri than 2tuple, 5tupe,
838 and flex filter, 0 means lower pri. */
842 * A structure used to define a 2tuple filter.
844 struct rte_2tuple_filter {
845 uint16_t dst_port; /**< big endian. */
848 uint16_t priority; /**< used when more than one filter matches. */
849 uint8_t dst_port_mask:1, /**< if mask is 1b, means not compare. */
854 * A structure used to define a flex filter.
856 struct rte_flex_filter {
858 uint32_t dwords[32]; /**< flex bytes in big endian. */
859 uint8_t mask[16]; /**< if mask bit is 1b, do not compare
860 corresponding byte in dwords. */
865 * A structure used to define a 5tuple filter.
867 struct rte_5tuple_filter {
868 uint32_t dst_ip; /**< destination IP address in big endian. */
869 uint32_t src_ip; /**< source IP address in big endian. */
870 uint16_t dst_port; /**< destination port in big endian. */
871 uint16_t src_port; /**< source Port big endian. */
872 uint8_t protocol; /**< l4 protocol. */
873 uint8_t tcp_flags; /**< tcp flags. */
874 uint16_t priority; /**< seven evels (001b-111b), 111b is highest,
875 used when more than one filter matches. */
876 uint8_t dst_ip_mask:1, /**< if mask is 1b, do not compare dst ip. */
877 src_ip_mask:1, /**< if mask is 1b, do not compare src ip. */
878 dst_port_mask:1, /**< if mask is 1b, do not compare dst port. */
879 src_port_mask:1, /**< if mask is 1b, do not compare src port. */
880 protocol_mask:1; /**< if mask is 1b, do not compare protocol. */
884 * Definitions of all functions exported by an Ethernet driver through the
885 * the generic structure of type *eth_dev_ops* supplied in the *rte_eth_dev*
886 * structure associated with an Ethernet device.
889 typedef int (*eth_dev_configure_t)(struct rte_eth_dev *dev);
890 /**< @internal Ethernet device configuration. */
892 typedef int (*eth_dev_start_t)(struct rte_eth_dev *dev);
893 /**< @internal Function used to start a configured Ethernet device. */
895 typedef void (*eth_dev_stop_t)(struct rte_eth_dev *dev);
896 /**< @internal Function used to stop a configured Ethernet device. */
898 typedef int (*eth_dev_set_link_up_t)(struct rte_eth_dev *dev);
899 /**< @internal Function used to link up a configured Ethernet device. */
901 typedef int (*eth_dev_set_link_down_t)(struct rte_eth_dev *dev);
902 /**< @internal Function used to link down a configured Ethernet device. */
904 typedef void (*eth_dev_close_t)(struct rte_eth_dev *dev);
905 /**< @internal Function used to close a configured Ethernet device. */
907 typedef void (*eth_promiscuous_enable_t)(struct rte_eth_dev *dev);
908 /**< @internal Function used to enable the RX promiscuous mode of an Ethernet device. */
910 typedef void (*eth_promiscuous_disable_t)(struct rte_eth_dev *dev);
911 /**< @internal Function used to disable the RX promiscuous mode of an Ethernet device. */
913 typedef void (*eth_allmulticast_enable_t)(struct rte_eth_dev *dev);
914 /**< @internal Enable the receipt of all multicast packets by an Ethernet device. */
916 typedef void (*eth_allmulticast_disable_t)(struct rte_eth_dev *dev);
917 /**< @internal Disable the receipt of all multicast packets by an Ethernet device. */
919 typedef int (*eth_link_update_t)(struct rte_eth_dev *dev,
920 int wait_to_complete);
921 /**< @internal Get link speed, duplex mode and state (up/down) of an Ethernet device. */
923 typedef void (*eth_stats_get_t)(struct rte_eth_dev *dev,
924 struct rte_eth_stats *igb_stats);
925 /**< @internal Get global I/O statistics of an Ethernet device. */
927 typedef void (*eth_stats_reset_t)(struct rte_eth_dev *dev);
928 /**< @internal Reset global I/O statistics of an Ethernet device to 0. */
930 typedef int (*eth_queue_stats_mapping_set_t)(struct rte_eth_dev *dev,
934 /**< @internal Set a queue statistics mapping for a tx/rx queue of an Ethernet device. */
936 typedef void (*eth_dev_infos_get_t)(struct rte_eth_dev *dev,
937 struct rte_eth_dev_info *dev_info);
938 /**< @internal Get specific informations of an Ethernet device. */
940 typedef int (*eth_queue_start_t)(struct rte_eth_dev *dev,
942 /**< @internal Start rx and tx of a queue of an Ethernet device. */
944 typedef int (*eth_queue_stop_t)(struct rte_eth_dev *dev,
946 /**< @internal Stop rx and tx of a queue of an Ethernet device. */
948 typedef int (*eth_rx_queue_setup_t)(struct rte_eth_dev *dev,
949 uint16_t rx_queue_id,
951 unsigned int socket_id,
952 const struct rte_eth_rxconf *rx_conf,
953 struct rte_mempool *mb_pool);
954 /**< @internal Set up a receive queue of an Ethernet device. */
956 typedef int (*eth_tx_queue_setup_t)(struct rte_eth_dev *dev,
957 uint16_t tx_queue_id,
959 unsigned int socket_id,
960 const struct rte_eth_txconf *tx_conf);
961 /**< @internal Setup a transmit queue of an Ethernet device. */
963 typedef void (*eth_queue_release_t)(void *queue);
964 /**< @internal Release memory resources allocated by given RX/TX queue. */
966 typedef uint32_t (*eth_rx_queue_count_t)(struct rte_eth_dev *dev,
967 uint16_t rx_queue_id);
968 /**< @Get number of available descriptors on a receive queue of an Ethernet device. */
970 typedef int (*eth_rx_descriptor_done_t)(void *rxq, uint16_t offset);
971 /**< @Check DD bit of specific RX descriptor */
973 typedef int (*vlan_filter_set_t)(struct rte_eth_dev *dev,
976 /**< @internal filtering of a VLAN Tag Identifier by an Ethernet device. */
978 typedef void (*vlan_tpid_set_t)(struct rte_eth_dev *dev,
980 /**< @internal set the outer VLAN-TPID by an Ethernet device. */
982 typedef void (*vlan_offload_set_t)(struct rte_eth_dev *dev, int mask);
983 /**< @internal set VLAN offload function by an Ethernet device. */
985 typedef void (*vlan_strip_queue_set_t)(struct rte_eth_dev *dev,
986 uint16_t rx_queue_id,
988 /**< @internal VLAN stripping enable/disable by an queue of Ethernet device. */
990 typedef uint16_t (*eth_rx_burst_t)(void *rxq,
991 struct rte_mbuf **rx_pkts,
993 /**< @internal Retrieve input packets from a receive queue of an Ethernet device. */
995 typedef uint16_t (*eth_tx_burst_t)(void *txq,
996 struct rte_mbuf **tx_pkts,
998 /**< @internal Send output packets on a transmit queue of an Ethernet device. */
1000 typedef int (*fdir_add_signature_filter_t)(struct rte_eth_dev *dev,
1001 struct rte_fdir_filter *fdir_ftr,
1003 /**< @internal Setup a new signature filter rule on an Ethernet device */
1005 typedef int (*fdir_update_signature_filter_t)(struct rte_eth_dev *dev,
1006 struct rte_fdir_filter *fdir_ftr,
1008 /**< @internal Update a signature filter rule on an Ethernet device */
1010 typedef int (*fdir_remove_signature_filter_t)(struct rte_eth_dev *dev,
1011 struct rte_fdir_filter *fdir_ftr);
1012 /**< @internal Remove a signature filter rule on an Ethernet device */
1014 typedef void (*fdir_infos_get_t)(struct rte_eth_dev *dev,
1015 struct rte_eth_fdir *fdir);
1016 /**< @internal Get information about fdir status */
1018 typedef int (*fdir_add_perfect_filter_t)(struct rte_eth_dev *dev,
1019 struct rte_fdir_filter *fdir_ftr,
1020 uint16_t soft_id, uint8_t rx_queue,
1022 /**< @internal Setup a new perfect filter rule on an Ethernet device */
1024 typedef int (*fdir_update_perfect_filter_t)(struct rte_eth_dev *dev,
1025 struct rte_fdir_filter *fdir_ftr,
1026 uint16_t soft_id, uint8_t rx_queue,
1028 /**< @internal Update a perfect filter rule on an Ethernet device */
1030 typedef int (*fdir_remove_perfect_filter_t)(struct rte_eth_dev *dev,
1031 struct rte_fdir_filter *fdir_ftr,
1033 /**< @internal Remove a perfect filter rule on an Ethernet device */
1035 typedef int (*fdir_set_masks_t)(struct rte_eth_dev *dev,
1036 struct rte_fdir_masks *fdir_masks);
1037 /**< @internal Setup flow director masks on an Ethernet device */
1039 typedef int (*flow_ctrl_set_t)(struct rte_eth_dev *dev,
1040 struct rte_eth_fc_conf *fc_conf);
1041 /**< @internal Setup flow control parameter on an Ethernet device */
1043 typedef int (*priority_flow_ctrl_set_t)(struct rte_eth_dev *dev,
1044 struct rte_eth_pfc_conf *pfc_conf);
1045 /**< @internal Setup priority flow control parameter on an Ethernet device */
1047 typedef int (*reta_update_t)(struct rte_eth_dev *dev,
1048 struct rte_eth_rss_reta *reta_conf);
1049 /**< @internal Update RSS redirection table on an Ethernet device */
1051 typedef int (*reta_query_t)(struct rte_eth_dev *dev,
1052 struct rte_eth_rss_reta *reta_conf);
1053 /**< @internal Query RSS redirection table on an Ethernet device */
1055 typedef int (*rss_hash_update_t)(struct rte_eth_dev *dev,
1056 struct rte_eth_rss_conf *rss_conf);
1057 /**< @internal Update RSS hash configuration of an Ethernet device */
1059 typedef int (*rss_hash_conf_get_t)(struct rte_eth_dev *dev,
1060 struct rte_eth_rss_conf *rss_conf);
1061 /**< @internal Get current RSS hash configuration of an Ethernet device */
1063 typedef int (*eth_dev_led_on_t)(struct rte_eth_dev *dev);
1064 /**< @internal Turn on SW controllable LED on an Ethernet device */
1066 typedef int (*eth_dev_led_off_t)(struct rte_eth_dev *dev);
1067 /**< @internal Turn off SW controllable LED on an Ethernet device */
1069 typedef void (*eth_mac_addr_remove_t)(struct rte_eth_dev *dev, uint32_t index);
1070 /**< @internal Remove MAC address from receive address register */
1072 typedef void (*eth_mac_addr_add_t)(struct rte_eth_dev *dev,
1073 struct ether_addr *mac_addr,
1076 /**< @internal Set a MAC address into Receive Address Address Register */
1078 typedef int (*eth_uc_hash_table_set_t)(struct rte_eth_dev *dev,
1079 struct ether_addr *mac_addr,
1081 /**< @internal Set a Unicast Hash bitmap */
1083 typedef int (*eth_uc_all_hash_table_set_t)(struct rte_eth_dev *dev,
1085 /**< @internal Set all Unicast Hash bitmap */
1087 typedef int (*eth_set_vf_rx_mode_t)(struct rte_eth_dev *dev,
1091 /**< @internal Set a VF receive mode */
1093 typedef int (*eth_set_vf_rx_t)(struct rte_eth_dev *dev,
1096 /**< @internal Set a VF receive mode */
1098 typedef int (*eth_set_vf_tx_t)(struct rte_eth_dev *dev,
1101 /**< @internal Enable or disable a VF transmit */
1103 typedef int (*eth_set_vf_vlan_filter_t)(struct rte_eth_dev *dev,
1107 /**< @internal Set VF VLAN pool filter */
1109 typedef int (*eth_set_queue_rate_limit_t)(struct rte_eth_dev *dev,
1112 /**< @internal Set queue TX rate */
1114 typedef int (*eth_set_vf_rate_limit_t)(struct rte_eth_dev *dev,
1118 /**< @internal Set VF TX rate */
1120 typedef int (*eth_mirror_rule_set_t)(struct rte_eth_dev *dev,
1121 struct rte_eth_vmdq_mirror_conf *mirror_conf,
1124 /**< @internal Add a traffic mirroring rule on an Ethernet device */
1126 typedef int (*eth_mirror_rule_reset_t)(struct rte_eth_dev *dev,
1128 /**< @internal Remove a traffic mirroring rule on an Ethernet device */
1130 #ifdef RTE_NIC_BYPASS
1133 RTE_BYPASS_MODE_NONE,
1134 RTE_BYPASS_MODE_NORMAL,
1135 RTE_BYPASS_MODE_BYPASS,
1136 RTE_BYPASS_MODE_ISOLATE,
1137 RTE_BYPASS_MODE_NUM,
1140 #define RTE_BYPASS_MODE_VALID(x) \
1141 ((x) > RTE_BYPASS_MODE_NONE && (x) < RTE_BYPASS_MODE_NUM)
1144 RTE_BYPASS_EVENT_NONE,
1145 RTE_BYPASS_EVENT_START,
1146 RTE_BYPASS_EVENT_OS_ON = RTE_BYPASS_EVENT_START,
1147 RTE_BYPASS_EVENT_POWER_ON,
1148 RTE_BYPASS_EVENT_OS_OFF,
1149 RTE_BYPASS_EVENT_POWER_OFF,
1150 RTE_BYPASS_EVENT_TIMEOUT,
1151 RTE_BYPASS_EVENT_NUM
1154 #define RTE_BYPASS_EVENT_VALID(x) \
1155 ((x) > RTE_BYPASS_EVENT_NONE && (x) < RTE_BYPASS_MODE_NUM)
1158 RTE_BYPASS_TMT_OFF, /* timeout disabled. */
1159 RTE_BYPASS_TMT_1_5_SEC, /* timeout for 1.5 seconds */
1160 RTE_BYPASS_TMT_2_SEC, /* timeout for 2 seconds */
1161 RTE_BYPASS_TMT_3_SEC, /* timeout for 3 seconds */
1162 RTE_BYPASS_TMT_4_SEC, /* timeout for 4 seconds */
1163 RTE_BYPASS_TMT_8_SEC, /* timeout for 8 seconds */
1164 RTE_BYPASS_TMT_16_SEC, /* timeout for 16 seconds */
1165 RTE_BYPASS_TMT_32_SEC, /* timeout for 32 seconds */
1169 #define RTE_BYPASS_TMT_VALID(x) \
1170 ((x) == RTE_BYPASS_TMT_OFF || \
1171 ((x) > RTE_BYPASS_TMT_OFF && (x) < RTE_BYPASS_TMT_NUM))
1173 typedef void (*bypass_init_t)(struct rte_eth_dev *dev);
1174 typedef int32_t (*bypass_state_set_t)(struct rte_eth_dev *dev, uint32_t *new_state);
1175 typedef int32_t (*bypass_state_show_t)(struct rte_eth_dev *dev, uint32_t *state);
1176 typedef int32_t (*bypass_event_set_t)(struct rte_eth_dev *dev, uint32_t state, uint32_t event);
1177 typedef int32_t (*bypass_event_show_t)(struct rte_eth_dev *dev, uint32_t event_shift, uint32_t *event);
1178 typedef int32_t (*bypass_wd_timeout_set_t)(struct rte_eth_dev *dev, uint32_t timeout);
1179 typedef int32_t (*bypass_wd_timeout_show_t)(struct rte_eth_dev *dev, uint32_t *wd_timeout);
1180 typedef int32_t (*bypass_ver_show_t)(struct rte_eth_dev *dev, uint32_t *ver);
1181 typedef int32_t (*bypass_wd_reset_t)(struct rte_eth_dev *dev);
1184 typedef int (*eth_add_syn_filter_t)(struct rte_eth_dev *dev,
1185 struct rte_syn_filter *filter, uint16_t rx_queue);
1186 /**< @internal add syn filter rule on an Ethernet device */
1188 typedef int (*eth_remove_syn_filter_t)(struct rte_eth_dev *dev);
1189 /**< @internal remove syn filter rule on an Ethernet device */
1191 typedef int (*eth_get_syn_filter_t)(struct rte_eth_dev *dev,
1192 struct rte_syn_filter *filter, uint16_t *rx_queue);
1193 /**< @internal Get syn filter rule on an Ethernet device */
1195 typedef int (*eth_add_ethertype_filter_t)(struct rte_eth_dev *dev,
1196 uint16_t index, struct rte_ethertype_filter *filter,
1198 /**< @internal Setup a new ethertype filter rule on an Ethernet device */
1200 typedef int (*eth_remove_ethertype_filter_t)(struct rte_eth_dev *dev,
1202 /**< @internal Remove an ethertype filter rule on an Ethernet device */
1204 typedef int (*eth_get_ethertype_filter_t)(struct rte_eth_dev *dev,
1205 uint16_t index, struct rte_ethertype_filter *filter,
1206 uint16_t *rx_queue);
1207 /**< @internal Get an ethertype filter rule on an Ethernet device */
1209 typedef int (*eth_add_2tuple_filter_t)(struct rte_eth_dev *dev,
1210 uint16_t index, struct rte_2tuple_filter *filter,
1212 /**< @internal Setup a new 2tuple filter rule on an Ethernet device */
1214 typedef int (*eth_remove_2tuple_filter_t)(struct rte_eth_dev *dev,
1216 /**< @internal Remove a 2tuple filter rule on an Ethernet device */
1218 typedef int (*eth_get_2tuple_filter_t)(struct rte_eth_dev *dev,
1219 uint16_t index, struct rte_2tuple_filter *filter,
1220 uint16_t *rx_queue);
1221 /**< @internal Get a 2tuple filter rule on an Ethernet device */
1223 typedef int (*eth_add_5tuple_filter_t)(struct rte_eth_dev *dev,
1224 uint16_t index, struct rte_5tuple_filter *filter,
1226 /**< @internal Setup a new 5tuple filter rule on an Ethernet device */
1228 typedef int (*eth_remove_5tuple_filter_t)(struct rte_eth_dev *dev,
1230 /**< @internal Remove a 5tuple filter rule on an Ethernet device */
1232 typedef int (*eth_get_5tuple_filter_t)(struct rte_eth_dev *dev,
1233 uint16_t index, struct rte_5tuple_filter *filter,
1234 uint16_t *rx_queue);
1235 /**< @internal Get a 5tuple filter rule on an Ethernet device */
1237 typedef int (*eth_add_flex_filter_t)(struct rte_eth_dev *dev,
1238 uint16_t index, struct rte_flex_filter *filter,
1240 /**< @internal Setup a new flex filter rule on an Ethernet device */
1242 typedef int (*eth_remove_flex_filter_t)(struct rte_eth_dev *dev,
1244 /**< @internal Remove a flex filter rule on an Ethernet device */
1246 typedef int (*eth_get_flex_filter_t)(struct rte_eth_dev *dev,
1247 uint16_t index, struct rte_flex_filter *filter,
1248 uint16_t *rx_queue);
1249 /**< @internal Get a flex filter rule on an Ethernet device */
1252 * @internal A structure containing the functions exported by an Ethernet driver.
1254 struct eth_dev_ops {
1255 eth_dev_configure_t dev_configure; /**< Configure device. */
1256 eth_dev_start_t dev_start; /**< Start device. */
1257 eth_dev_stop_t dev_stop; /**< Stop device. */
1258 eth_dev_set_link_up_t dev_set_link_up; /**< Device link up. */
1259 eth_dev_set_link_down_t dev_set_link_down; /**< Device link down. */
1260 eth_dev_close_t dev_close; /**< Close device. */
1261 eth_promiscuous_enable_t promiscuous_enable; /**< Promiscuous ON. */
1262 eth_promiscuous_disable_t promiscuous_disable;/**< Promiscuous OFF. */
1263 eth_allmulticast_enable_t allmulticast_enable;/**< RX multicast ON. */
1264 eth_allmulticast_disable_t allmulticast_disable;/**< RX multicast OF. */
1265 eth_link_update_t link_update; /**< Get device link state. */
1266 eth_stats_get_t stats_get; /**< Get device statistics. */
1267 eth_stats_reset_t stats_reset; /**< Reset device statistics. */
1268 eth_queue_stats_mapping_set_t queue_stats_mapping_set;
1269 /**< Configure per queue stat counter mapping. */
1270 eth_dev_infos_get_t dev_infos_get; /**< Get device info. */
1271 vlan_filter_set_t vlan_filter_set; /**< Filter VLAN Setup. */
1272 vlan_tpid_set_t vlan_tpid_set; /**< Outer VLAN TPID Setup. */
1273 vlan_strip_queue_set_t vlan_strip_queue_set; /**< VLAN Stripping on queue. */
1274 vlan_offload_set_t vlan_offload_set; /**< Set VLAN Offload. */
1275 eth_queue_start_t rx_queue_start;/**< Start RX for a queue.*/
1276 eth_queue_stop_t rx_queue_stop;/**< Stop RX for a queue.*/
1277 eth_queue_start_t tx_queue_start;/**< Start TX for a queue.*/
1278 eth_queue_stop_t tx_queue_stop;/**< Stop TX for a queue.*/
1279 eth_rx_queue_setup_t rx_queue_setup;/**< Set up device RX queue.*/
1280 eth_queue_release_t rx_queue_release;/**< Release RX queue.*/
1281 eth_rx_queue_count_t rx_queue_count; /**< Get Rx queue count. */
1282 eth_rx_descriptor_done_t rx_descriptor_done; /**< Check rxd DD bit */
1283 eth_tx_queue_setup_t tx_queue_setup;/**< Set up device TX queue.*/
1284 eth_queue_release_t tx_queue_release;/**< Release TX queue.*/
1285 eth_dev_led_on_t dev_led_on; /**< Turn on LED. */
1286 eth_dev_led_off_t dev_led_off; /**< Turn off LED. */
1287 flow_ctrl_set_t flow_ctrl_set; /**< Setup flow control. */
1288 priority_flow_ctrl_set_t priority_flow_ctrl_set; /**< Setup priority flow control.*/
1289 eth_mac_addr_remove_t mac_addr_remove; /**< Remove MAC address */
1290 eth_mac_addr_add_t mac_addr_add; /**< Add a MAC address */
1291 eth_uc_hash_table_set_t uc_hash_table_set; /**< Set Unicast Table Array */
1292 eth_uc_all_hash_table_set_t uc_all_hash_table_set; /**< Set Unicast hash bitmap */
1293 eth_mirror_rule_set_t mirror_rule_set; /**< Add a traffic mirror rule.*/
1294 eth_mirror_rule_reset_t mirror_rule_reset; /**< reset a traffic mirror rule.*/
1295 eth_set_vf_rx_mode_t set_vf_rx_mode; /**< Set VF RX mode */
1296 eth_set_vf_rx_t set_vf_rx; /**< enable/disable a VF receive */
1297 eth_set_vf_tx_t set_vf_tx; /**< enable/disable a VF transmit */
1298 eth_set_vf_vlan_filter_t set_vf_vlan_filter; /**< Set VF VLAN filter */
1299 eth_set_queue_rate_limit_t set_queue_rate_limit; /**< Set queue rate limit */
1300 eth_set_vf_rate_limit_t set_vf_rate_limit; /**< Set VF rate limit */
1302 /** Add a signature filter. */
1303 fdir_add_signature_filter_t fdir_add_signature_filter;
1304 /** Update a signature filter. */
1305 fdir_update_signature_filter_t fdir_update_signature_filter;
1306 /** Remove a signature filter. */
1307 fdir_remove_signature_filter_t fdir_remove_signature_filter;
1308 /** Get information about FDIR status. */
1309 fdir_infos_get_t fdir_infos_get;
1310 /** Add a perfect filter. */
1311 fdir_add_perfect_filter_t fdir_add_perfect_filter;
1312 /** Update a perfect filter. */
1313 fdir_update_perfect_filter_t fdir_update_perfect_filter;
1314 /** Remove a perfect filter. */
1315 fdir_remove_perfect_filter_t fdir_remove_perfect_filter;
1316 /** Setup masks for FDIR filtering. */
1317 fdir_set_masks_t fdir_set_masks;
1318 /** Update redirection table. */
1319 reta_update_t reta_update;
1320 /** Query redirection table. */
1321 reta_query_t reta_query;
1322 /* bypass control */
1323 #ifdef RTE_NIC_BYPASS
1324 bypass_init_t bypass_init;
1325 bypass_state_set_t bypass_state_set;
1326 bypass_state_show_t bypass_state_show;
1327 bypass_event_set_t bypass_event_set;
1328 bypass_event_show_t bypass_event_show;
1329 bypass_wd_timeout_set_t bypass_wd_timeout_set;
1330 bypass_wd_timeout_show_t bypass_wd_timeout_show;
1331 bypass_ver_show_t bypass_ver_show;
1332 bypass_wd_reset_t bypass_wd_reset;
1335 /** Configure RSS hash protocols. */
1336 rss_hash_update_t rss_hash_update;
1337 /** Get current RSS hash configuration. */
1338 rss_hash_conf_get_t rss_hash_conf_get;
1339 eth_add_syn_filter_t add_syn_filter; /**< add syn filter. */
1340 eth_remove_syn_filter_t remove_syn_filter; /**< remove syn filter. */
1341 eth_get_syn_filter_t get_syn_filter; /**< get syn filter. */
1342 eth_add_ethertype_filter_t add_ethertype_filter; /**< add ethertype filter. */
1343 eth_remove_ethertype_filter_t remove_ethertype_filter; /**< remove ethertype filter. */
1344 eth_get_ethertype_filter_t get_ethertype_filter; /**< get ethertype filter. */
1345 eth_add_2tuple_filter_t add_2tuple_filter; /**< add 2tuple filter. */
1346 eth_remove_2tuple_filter_t remove_2tuple_filter; /**< remove 2tuple filter. */
1347 eth_get_2tuple_filter_t get_2tuple_filter; /**< get 2tuple filter. */
1348 eth_add_5tuple_filter_t add_5tuple_filter; /**< add 5tuple filter. */
1349 eth_remove_5tuple_filter_t remove_5tuple_filter; /**< remove 5tuple filter. */
1350 eth_get_5tuple_filter_t get_5tuple_filter; /**< get 5tuple filter. */
1351 eth_add_flex_filter_t add_flex_filter; /**< add flex filter. */
1352 eth_remove_flex_filter_t remove_flex_filter; /**< remove flex filter. */
1353 eth_get_flex_filter_t get_flex_filter; /**< get flex filter. */
1358 * The generic data structure associated with each ethernet device.
1360 * Pointers to burst-oriented packet receive and transmit functions are
1361 * located at the beginning of the structure, along with the pointer to
1362 * where all the data elements for the particular device are stored in shared
1363 * memory. This split allows the function pointer and driver data to be per-
1364 * process, while the actual configuration data for the device is shared.
1366 struct rte_eth_dev {
1367 eth_rx_burst_t rx_pkt_burst; /**< Pointer to PMD receive function. */
1368 eth_tx_burst_t tx_pkt_burst; /**< Pointer to PMD transmit function. */
1369 struct rte_eth_dev_data *data; /**< Pointer to device data */
1370 const struct eth_driver *driver;/**< Driver for this device */
1371 struct eth_dev_ops *dev_ops; /**< Functions exported by PMD */
1372 struct rte_pci_device *pci_dev; /**< PCI info. supplied by probing */
1373 struct rte_eth_dev_cb_list callbacks; /**< User application callbacks */
1376 struct rte_eth_dev_sriov {
1377 uint8_t active; /**< SRIOV is active with 16, 32 or 64 pools */
1378 uint8_t nb_q_per_pool; /**< rx queue number per pool */
1379 uint16_t def_vmdq_idx; /**< Default pool num used for PF */
1380 uint16_t def_pool_q_idx; /**< Default pool queue start reg index */
1382 #define RTE_ETH_DEV_SRIOV(dev) ((dev)->data->sriov)
1386 * The data part, with no function pointers, associated with each ethernet device.
1388 * This structure is safe to place in shared memory to be common among different
1389 * processes in a multi-process configuration.
1391 struct rte_eth_dev_data {
1392 void **rx_queues; /**< Array of pointers to RX queues. */
1393 void **tx_queues; /**< Array of pointers to TX queues. */
1394 uint16_t nb_rx_queues; /**< Number of RX queues. */
1395 uint16_t nb_tx_queues; /**< Number of TX queues. */
1397 struct rte_eth_dev_sriov sriov; /**< SRIOV data */
1399 void *dev_private; /**< PMD-specific private data */
1401 struct rte_eth_link dev_link;
1402 /**< Link-level information & status */
1404 struct rte_eth_conf dev_conf; /**< Configuration applied to device. */
1405 uint16_t max_frame_size; /**< Default is ETHER_MAX_LEN (1518). */
1407 uint64_t rx_mbuf_alloc_failed; /**< RX ring mbuf allocation failures. */
1408 struct ether_addr* mac_addrs;/**< Device Ethernet Link address. */
1409 uint64_t mac_pool_sel[ETH_NUM_RECEIVE_MAC_ADDR];
1410 /** bitmap array of associating Ethernet MAC addresses to pools */
1411 struct ether_addr* hash_mac_addrs;
1412 /** Device Ethernet MAC addresses of hash filtering. */
1413 uint8_t port_id; /**< Device [external] port identifier. */
1414 uint8_t promiscuous : 1, /**< RX promiscuous mode ON(1) / OFF(0). */
1415 scattered_rx : 1, /**< RX of scattered packets is ON(1) / OFF(0) */
1416 all_multicast : 1, /**< RX all multicast mode ON(1) / OFF(0). */
1417 dev_started : 1; /**< Device state: STARTED(1) / STOPPED(0). */
1422 * The pool of *rte_eth_dev* structures. The size of the pool
1423 * is configured at compile-time in the <rte_ethdev.c> file.
1425 extern struct rte_eth_dev rte_eth_devices[];
1428 * Get the total number of Ethernet devices that have been successfully
1429 * initialized by the [matching] Ethernet driver during the PCI probing phase.
1430 * All devices whose port identifier is in the range
1431 * [0, rte_eth_dev_count() - 1] can be operated on by network applications.
1434 * - The total number of usable Ethernet devices.
1436 extern uint8_t rte_eth_dev_count(void);
1439 * Function for internal use by dummy drivers primarily, e.g. ring-based
1441 * Allocates a new ethdev slot for an ethernet device and returns the pointer
1442 * to that slot for the driver to use.
1445 * - Slot in the rte_dev_devices array for a new device;
1447 struct rte_eth_dev *rte_eth_dev_allocate(void);
1452 * Initialization function of an Ethernet driver invoked for each matching
1453 * Ethernet PCI device detected during the PCI probing phase.
1456 * The pointer to the [matching] Ethernet driver structure supplied by
1457 * the PMD when it registered itself.
1459 * The *eth_dev* pointer is the address of the *rte_eth_dev* structure
1460 * associated with the matching device and which have been [automatically]
1461 * allocated in the *rte_eth_devices* array.
1462 * The *eth_dev* structure is supplied to the driver initialization function
1463 * with the following fields already initialized:
1465 * - *pci_dev*: Holds the pointers to the *rte_pci_device* structure which
1466 * contains the generic PCI information of the matching device.
1468 * - *dev_private*: Holds a pointer to the device private data structure.
1470 * - *max_frame_size*: Contains the default Ethernet maximum frame length
1473 * - *port_id*: Contains the port index of the device (actually the index
1474 * of the *eth_dev* structure in the *rte_eth_devices* array).
1477 * - 0: Success, the device is properly initialized by the driver.
1478 * In particular, the driver MUST have set up the *dev_ops* pointer
1479 * of the *eth_dev* structure.
1480 * - <0: Error code of the device initialization failure.
1482 typedef int (*eth_dev_init_t)(struct eth_driver *eth_drv,
1483 struct rte_eth_dev *eth_dev);
1487 * The structure associated with a PMD Ethernet driver.
1489 * Each Ethernet driver acts as a PCI driver and is represented by a generic
1490 * *eth_driver* structure that holds:
1492 * - An *rte_pci_driver* structure (which must be the first field).
1494 * - The *eth_dev_init* function invoked for each matching PCI device.
1496 * - The size of the private data to allocate for each matching device.
1499 struct rte_pci_driver pci_drv; /**< The PMD is also a PCI driver. */
1500 eth_dev_init_t eth_dev_init; /**< Device init function. */
1501 unsigned int dev_private_size; /**< Size of device private data. */
1506 * A function invoked by the initialization function of an Ethernet driver
1507 * to simultaneously register itself as a PCI driver and as an Ethernet
1508 * Poll Mode Driver (PMD).
1511 * The pointer to the *eth_driver* structure associated with
1512 * the Ethernet driver.
1514 extern void rte_eth_driver_register(struct eth_driver *eth_drv);
1517 * Configure an Ethernet device.
1518 * This function must be invoked first before any other function in the
1519 * Ethernet API. This function can also be re-invoked when a device is in the
1523 * The port identifier of the Ethernet device to configure.
1524 * @param nb_rx_queue
1525 * The number of receive queues to set up for the Ethernet device.
1526 * @param nb_tx_queue
1527 * The number of transmit queues to set up for the Ethernet device.
1529 * The pointer to the configuration data to be used for the Ethernet device.
1530 * The *rte_eth_conf* structure includes:
1531 * - the hardware offload features to activate, with dedicated fields for
1532 * each statically configurable offload hardware feature provided by
1533 * Ethernet devices, such as IP checksum or VLAN tag stripping for
1535 * - the Receive Side Scaling (RSS) configuration when using multiple RX
1538 * Embedding all configuration information in a single data structure
1539 * is the more flexible method that allows the addition of new features
1540 * without changing the syntax of the API.
1542 * - 0: Success, device configured.
1543 * - <0: Error code returned by the driver configuration function.
1545 extern int rte_eth_dev_configure(uint8_t port_id,
1546 uint16_t nb_rx_queue,
1547 uint16_t nb_tx_queue,
1548 const struct rte_eth_conf *eth_conf);
1551 * Allocate and set up a receive queue for an Ethernet device.
1553 * The function allocates a contiguous block of memory for *nb_rx_desc*
1554 * receive descriptors from a memory zone associated with *socket_id*
1555 * and initializes each receive descriptor with a network buffer allocated
1556 * from the memory pool *mb_pool*.
1559 * The port identifier of the Ethernet device.
1560 * @param rx_queue_id
1561 * The index of the receive queue to set up.
1562 * The value must be in the range [0, nb_rx_queue - 1] previously supplied
1563 * to rte_eth_dev_configure().
1565 * The number of receive descriptors to allocate for the receive ring.
1567 * The *socket_id* argument is the socket identifier in case of NUMA.
1568 * The value can be *SOCKET_ID_ANY* if there is no NUMA constraint for
1569 * the DMA memory allocated for the receive descriptors of the ring.
1571 * The pointer to the configuration data to be used for the receive queue.
1572 * The *rx_conf* structure contains an *rx_thresh* structure with the values
1573 * of the Prefetch, Host, and Write-Back threshold registers of the receive
1576 * The pointer to the memory pool from which to allocate *rte_mbuf* network
1577 * memory buffers to populate each descriptor of the receive ring.
1579 * - 0: Success, receive queue correctly set up.
1580 * - -EINVAL: The size of network buffers which can be allocated from the
1581 * memory pool does not fit the various buffer sizes allowed by the
1582 * device controller.
1583 * - -ENOMEM: Unable to allocate the receive ring descriptors or to
1584 * allocate network memory buffers from the memory pool when
1585 * initializing receive descriptors.
1587 extern int rte_eth_rx_queue_setup(uint8_t port_id, uint16_t rx_queue_id,
1588 uint16_t nb_rx_desc, unsigned int socket_id,
1589 const struct rte_eth_rxconf *rx_conf,
1590 struct rte_mempool *mb_pool);
1593 * Allocate and set up a transmit queue for an Ethernet device.
1596 * The port identifier of the Ethernet device.
1597 * @param tx_queue_id
1598 * The index of the transmit queue to set up.
1599 * The value must be in the range [0, nb_tx_queue - 1] previously supplied
1600 * to rte_eth_dev_configure().
1602 * The number of transmit descriptors to allocate for the transmit ring.
1604 * The *socket_id* argument is the socket identifier in case of NUMA.
1605 * Its value can be *SOCKET_ID_ANY* if there is no NUMA constraint for
1606 * the DMA memory allocated for the transmit descriptors of the ring.
1608 * The pointer to the configuration data to be used for the transmit queue.
1609 * The *tx_conf* structure contains the following data:
1610 * - The *tx_thresh* structure with the values of the Prefetch, Host, and
1611 * Write-Back threshold registers of the transmit ring.
1612 * When setting Write-Back threshold to the value greater then zero,
1613 * *tx_rs_thresh* value should be explicitly set to one.
1614 * - The *tx_free_thresh* value indicates the [minimum] number of network
1615 * buffers that must be pending in the transmit ring to trigger their
1616 * [implicit] freeing by the driver transmit function.
1617 * - The *tx_rs_thresh* value indicates the [minimum] number of transmit
1618 * descriptors that must be pending in the transmit ring before setting the
1619 * RS bit on a descriptor by the driver transmit function.
1620 * The *tx_rs_thresh* value should be less or equal then
1621 * *tx_free_thresh* value, and both of them should be less then
1623 * - The *txq_flags* member contains flags to pass to the TX queue setup
1624 * function to configure the behavior of the TX queue. This should be set
1625 * to 0 if no special configuration is required.
1627 * Note that setting *tx_free_thresh* or *tx_rs_thresh* value to 0 forces
1628 * the transmit function to use default values.
1630 * - 0: Success, the transmit queue is correctly set up.
1631 * - -ENOMEM: Unable to allocate the transmit ring descriptors.
1633 extern int rte_eth_tx_queue_setup(uint8_t port_id, uint16_t tx_queue_id,
1634 uint16_t nb_tx_desc, unsigned int socket_id,
1635 const struct rte_eth_txconf *tx_conf);
1638 * Return the NUMA socket to which an Ethernet device is connected
1641 * The port identifier of the Ethernet device
1643 * The NUMA socket id to which the Ethernet device is connected or
1644 * a default of zero if the socket could not be determined.
1645 * -1 is returned is the port_id value is out of range.
1647 extern int rte_eth_dev_socket_id(uint8_t port_id);
1650 * Start specified RX queue of a port
1653 * The port identifier of the Ethernet device
1654 * @param rx_queue_id
1655 * The index of the rx queue to update the ring.
1656 * The value must be in the range [0, nb_rx_queue - 1] previously supplied
1657 * to rte_eth_dev_configure().
1659 * - 0: Success, the transmit queue is correctly set up.
1660 * - -EINVAL: The port_id or the queue_id out of range.
1661 * - -ENOTSUP: The function not supported in PMD driver.
1663 extern int rte_eth_dev_rx_queue_start(uint8_t port_id, uint16_t rx_queue_id);
1666 * Stop specified RX queue of a port
1669 * The port identifier of the Ethernet device
1670 * @param rx_queue_id
1671 * The index of the rx queue to update the ring.
1672 * The value must be in the range [0, nb_rx_queue - 1] previously supplied
1673 * to rte_eth_dev_configure().
1675 * - 0: Success, the transmit queue is correctly set up.
1676 * - -EINVAL: The port_id or the queue_id out of range.
1677 * - -ENOTSUP: The function not supported in PMD driver.
1679 extern int rte_eth_dev_rx_queue_stop(uint8_t port_id, uint16_t rx_queue_id);
1682 * Start specified TX queue of a port
1685 * The port identifier of the Ethernet device
1686 * @param tx_queue_id
1687 * The index of the tx queue to update the ring.
1688 * The value must be in the range [0, nb_tx_queue - 1] previously supplied
1689 * to rte_eth_dev_configure().
1691 * - 0: Success, the transmit queue is correctly set up.
1692 * - -EINVAL: The port_id or the queue_id out of range.
1693 * - -ENOTSUP: The function not supported in PMD driver.
1695 extern int rte_eth_dev_tx_queue_start(uint8_t port_id, uint16_t tx_queue_id);
1698 * Stop specified TX queue of a port
1701 * The port identifier of the Ethernet device
1702 * @param tx_queue_id
1703 * The index of the tx queue to update the ring.
1704 * The value must be in the range [0, nb_tx_queue - 1] previously supplied
1705 * to rte_eth_dev_configure().
1707 * - 0: Success, the transmit queue is correctly set up.
1708 * - -EINVAL: The port_id or the queue_id out of range.
1709 * - -ENOTSUP: The function not supported in PMD driver.
1711 extern int rte_eth_dev_tx_queue_stop(uint8_t port_id, uint16_t tx_queue_id);
1716 * Start an Ethernet device.
1718 * The device start step is the last one and consists of setting the configured
1719 * offload features and in starting the transmit and the receive units of the
1721 * On success, all basic functions exported by the Ethernet API (link status,
1722 * receive/transmit, and so on) can be invoked.
1725 * The port identifier of the Ethernet device.
1727 * - 0: Success, Ethernet device started.
1728 * - <0: Error code of the driver device start function.
1730 extern int rte_eth_dev_start(uint8_t port_id);
1733 * Stop an Ethernet device. The device can be restarted with a call to
1734 * rte_eth_dev_start()
1737 * The port identifier of the Ethernet device.
1739 extern void rte_eth_dev_stop(uint8_t port_id);
1743 * Link up an Ethernet device.
1745 * Set device link up will re-enable the device rx/tx
1746 * functionality after it is previously set device linked down.
1749 * The port identifier of the Ethernet device.
1751 * - 0: Success, Ethernet device linked up.
1752 * - <0: Error code of the driver device link up function.
1754 extern int rte_eth_dev_set_link_up(uint8_t port_id);
1757 * Link down an Ethernet device.
1758 * The device rx/tx functionality will be disabled if success,
1759 * and it can be re-enabled with a call to
1760 * rte_eth_dev_set_link_up()
1763 * The port identifier of the Ethernet device.
1765 extern int rte_eth_dev_set_link_down(uint8_t port_id);
1768 * Close an Ethernet device. The device cannot be restarted!
1771 * The port identifier of the Ethernet device.
1773 extern void rte_eth_dev_close(uint8_t port_id);
1776 * Enable receipt in promiscuous mode for an Ethernet device.
1779 * The port identifier of the Ethernet device.
1781 extern void rte_eth_promiscuous_enable(uint8_t port_id);
1784 * Disable receipt in promiscuous mode for an Ethernet device.
1787 * The port identifier of the Ethernet device.
1789 extern void rte_eth_promiscuous_disable(uint8_t port_id);
1792 * Return the value of promiscuous mode for an Ethernet device.
1795 * The port identifier of the Ethernet device.
1797 * - (1) if promiscuous is enabled
1798 * - (0) if promiscuous is disabled.
1801 extern int rte_eth_promiscuous_get(uint8_t port_id);
1804 * Enable the receipt of any multicast frame by an Ethernet device.
1807 * The port identifier of the Ethernet device.
1809 extern void rte_eth_allmulticast_enable(uint8_t port_id);
1812 * Disable the receipt of all multicast frames by an Ethernet device.
1815 * The port identifier of the Ethernet device.
1817 extern void rte_eth_allmulticast_disable(uint8_t port_id);
1820 * Return the value of allmulticast mode for an Ethernet device.
1823 * The port identifier of the Ethernet device.
1825 * - (1) if allmulticast is enabled
1826 * - (0) if allmulticast is disabled.
1829 extern int rte_eth_allmulticast_get(uint8_t port_id);
1832 * Retrieve the status (ON/OFF), the speed (in Mbps) and the mode (HALF-DUPLEX
1833 * or FULL-DUPLEX) of the physical link of an Ethernet device. It might need
1834 * to wait up to 9 seconds in it.
1837 * The port identifier of the Ethernet device.
1839 * A pointer to an *rte_eth_link* structure to be filled with
1840 * the status, the speed and the mode of the Ethernet device link.
1842 extern void rte_eth_link_get(uint8_t port_id, struct rte_eth_link *link);
1845 * Retrieve the status (ON/OFF), the speed (in Mbps) and the mode (HALF-DUPLEX
1846 * or FULL-DUPLEX) of the physical link of an Ethernet device. It is a no-wait
1847 * version of rte_eth_link_get().
1850 * The port identifier of the Ethernet device.
1852 * A pointer to an *rte_eth_link* structure to be filled with
1853 * the status, the speed and the mode of the Ethernet device link.
1855 extern void rte_eth_link_get_nowait(uint8_t port_id,
1856 struct rte_eth_link *link);
1859 * Retrieve the general I/O statistics of an Ethernet device.
1862 * The port identifier of the Ethernet device.
1864 * A pointer to a structure of type *rte_eth_stats* to be filled with
1865 * the values of device counters for the following set of statistics:
1866 * - *ipackets* with the total of successfully received packets.
1867 * - *opackets* with the total of successfully transmitted packets.
1868 * - *ibytes* with the total of successfully received bytes.
1869 * - *obytes* with the total of successfully transmitted bytes.
1870 * - *ierrors* with the total of erroneous received packets.
1871 * - *oerrors* with the total of failed transmitted packets.
1873 extern void rte_eth_stats_get(uint8_t port_id, struct rte_eth_stats *stats);
1876 * Reset the general I/O statistics of an Ethernet device.
1879 * The port identifier of the Ethernet device.
1881 extern void rte_eth_stats_reset(uint8_t port_id);
1884 * Set a mapping for the specified transmit queue to the specified per-queue
1885 * statistics counter.
1888 * The port identifier of the Ethernet device.
1889 * @param tx_queue_id
1890 * The index of the transmit queue for which a queue stats mapping is required.
1891 * The value must be in the range [0, nb_tx_queue - 1] previously supplied
1892 * to rte_eth_dev_configure().
1894 * The per-queue packet statistics functionality number that the transmit
1895 * queue is to be assigned.
1896 * The value must be in the range [0, RTE_MAX_ETHPORT_QUEUE_STATS_MAPS - 1].
1898 * Zero if successful. Non-zero otherwise.
1900 extern int rte_eth_dev_set_tx_queue_stats_mapping(uint8_t port_id,
1901 uint16_t tx_queue_id,
1905 * Set a mapping for the specified receive queue to the specified per-queue
1906 * statistics counter.
1909 * The port identifier of the Ethernet device.
1910 * @param rx_queue_id
1911 * The index of the receive queue for which a queue stats mapping is required.
1912 * The value must be in the range [0, nb_rx_queue - 1] previously supplied
1913 * to rte_eth_dev_configure().
1915 * The per-queue packet statistics functionality number that the receive
1916 * queue is to be assigned.
1917 * The value must be in the range [0, RTE_MAX_ETHPORT_QUEUE_STATS_MAPS - 1].
1919 * Zero if successful. Non-zero otherwise.
1921 extern int rte_eth_dev_set_rx_queue_stats_mapping(uint8_t port_id,
1922 uint16_t rx_queue_id,
1926 * Retrieve the Ethernet address of an Ethernet device.
1929 * The port identifier of the Ethernet device.
1931 * A pointer to a structure of type *ether_addr* to be filled with
1932 * the Ethernet address of the Ethernet device.
1934 extern void rte_eth_macaddr_get(uint8_t port_id, struct ether_addr *mac_addr);
1937 * Retrieve the contextual information of an Ethernet device.
1940 * The port identifier of the Ethernet device.
1942 * A pointer to a structure of type *rte_eth_dev_info* to be filled with
1943 * the contextual information of the Ethernet device.
1945 extern void rte_eth_dev_info_get(uint8_t port_id,
1946 struct rte_eth_dev_info *dev_info);
1949 * Enable/Disable hardware filtering by an Ethernet device of received
1950 * VLAN packets tagged with a given VLAN Tag Identifier.
1953 * The port identifier of the Ethernet device.
1955 * The VLAN Tag Identifier whose filtering must be enabled or disabled.
1957 * If > 0, enable VLAN filtering of VLAN packets tagged with *vlan_id*.
1958 * Otherwise, disable VLAN filtering of VLAN packets tagged with *vlan_id*.
1960 * - (0) if successful.
1961 * - (-ENOSUP) if hardware-assisted VLAN filtering not configured.
1962 * - (-ENODEV) if *port_id* invalid.
1963 * - (-ENOSYS) if VLAN filtering on *port_id* disabled.
1964 * - (-EINVAL) if *vlan_id* > 4095.
1966 extern int rte_eth_dev_vlan_filter(uint8_t port_id, uint16_t vlan_id , int on);
1969 * Enable/Disable hardware VLAN Strip by a rx queue of an Ethernet device.
1970 * 82599/X540 can support VLAN stripping at the rx queue level
1973 * The port identifier of the Ethernet device.
1974 * @param rx_queue_id
1975 * The index of the receive queue for which a queue stats mapping is required.
1976 * The value must be in the range [0, nb_rx_queue - 1] previously supplied
1977 * to rte_eth_dev_configure().
1979 * If 1, Enable VLAN Stripping of the receive queue of the Ethernet port.
1980 * If 0, Disable VLAN Stripping of the receive queue of the Ethernet port.
1982 * - (0) if successful.
1983 * - (-ENOSUP) if hardware-assisted VLAN stripping not configured.
1984 * - (-ENODEV) if *port_id* invalid.
1985 * - (-EINVAL) if *rx_queue_id* invalid.
1987 extern int rte_eth_dev_set_vlan_strip_on_queue(uint8_t port_id,
1988 uint16_t rx_queue_id, int on);
1991 * Set the Outer VLAN Ether Type by an Ethernet device, it can be inserted to
1992 * the VLAN Header. This is a register setup available on some Intel NIC, not
1993 * but all, please check the data sheet for availability.
1996 * The port identifier of the Ethernet device.
1998 * The Tag Protocol ID
2000 * - (0) if successful.
2001 * - (-ENOSUP) if hardware-assisted VLAN TPID setup is not supported.
2002 * - (-ENODEV) if *port_id* invalid.
2004 extern int rte_eth_dev_set_vlan_ether_type(uint8_t port_id, uint16_t tag_type);
2007 * Set VLAN offload configuration on an Ethernet device
2008 * Enable/Disable Extended VLAN by an Ethernet device, This is a register setup
2009 * available on some Intel NIC, not but all, please check the data sheet for
2011 * Enable/Disable VLAN Strip can be done on rx queue for certain NIC, but here
2012 * the configuration is applied on the port level.
2015 * The port identifier of the Ethernet device.
2016 * @param offload_mask
2017 * The VLAN Offload bit mask can be mixed use with "OR"
2018 * ETH_VLAN_STRIP_OFFLOAD
2019 * ETH_VLAN_FILTER_OFFLOAD
2020 * ETH_VLAN_EXTEND_OFFLOAD
2022 * - (0) if successful.
2023 * - (-ENOSUP) if hardware-assisted VLAN filtering not configured.
2024 * - (-ENODEV) if *port_id* invalid.
2026 extern int rte_eth_dev_set_vlan_offload(uint8_t port_id, int offload_mask);
2029 * Read VLAN Offload configuration from an Ethernet device
2032 * The port identifier of the Ethernet device.
2034 * - (>0) if successful. Bit mask to indicate
2035 * ETH_VLAN_STRIP_OFFLOAD
2036 * ETH_VLAN_FILTER_OFFLOAD
2037 * ETH_VLAN_EXTEND_OFFLOAD
2038 * - (-ENODEV) if *port_id* invalid.
2040 extern int rte_eth_dev_get_vlan_offload(uint8_t port_id);
2044 * Retrieve a burst of input packets from a receive queue of an Ethernet
2045 * device. The retrieved packets are stored in *rte_mbuf* structures whose
2046 * pointers are supplied in the *rx_pkts* array.
2048 * The rte_eth_rx_burst() function loops, parsing the RX ring of the
2049 * receive queue, up to *nb_pkts* packets, and for each completed RX
2050 * descriptor in the ring, it performs the following operations:
2052 * - Initialize the *rte_mbuf* data structure associated with the
2053 * RX descriptor according to the information provided by the NIC into
2054 * that RX descriptor.
2056 * - Store the *rte_mbuf* data structure into the next entry of the
2059 * - Replenish the RX descriptor with a new *rte_mbuf* buffer
2060 * allocated from the memory pool associated with the receive queue at
2061 * initialization time.
2063 * When retrieving an input packet that was scattered by the controller
2064 * into multiple receive descriptors, the rte_eth_rx_burst() function
2065 * appends the associated *rte_mbuf* buffers to the first buffer of the
2068 * The rte_eth_rx_burst() function returns the number of packets
2069 * actually retrieved, which is the number of *rte_mbuf* data structures
2070 * effectively supplied into the *rx_pkts* array.
2071 * A return value equal to *nb_pkts* indicates that the RX queue contained
2072 * at least *rx_pkts* packets, and this is likely to signify that other
2073 * received packets remain in the input queue. Applications implementing
2074 * a "retrieve as much received packets as possible" policy can check this
2075 * specific case and keep invoking the rte_eth_rx_burst() function until
2076 * a value less than *nb_pkts* is returned.
2078 * This receive method has the following advantages:
2080 * - It allows a run-to-completion network stack engine to retrieve and
2081 * to immediately process received packets in a fast burst-oriented
2082 * approach, avoiding the overhead of unnecessary intermediate packet
2083 * queue/dequeue operations.
2085 * - Conversely, it also allows an asynchronous-oriented processing
2086 * method to retrieve bursts of received packets and to immediately
2087 * queue them for further parallel processing by another logical core,
2088 * for instance. However, instead of having received packets being
2089 * individually queued by the driver, this approach allows the invoker
2090 * of the rte_eth_rx_burst() function to queue a burst of retrieved
2091 * packets at a time and therefore dramatically reduce the cost of
2092 * enqueue/dequeue operations per packet.
2094 * - It allows the rte_eth_rx_burst() function of the driver to take
2095 * advantage of burst-oriented hardware features (CPU cache,
2096 * prefetch instructions, and so on) to minimize the number of CPU
2097 * cycles per packet.
2099 * To summarize, the proposed receive API enables many
2100 * burst-oriented optimizations in both synchronous and asynchronous
2101 * packet processing environments with no overhead in both cases.
2103 * The rte_eth_rx_burst() function does not provide any error
2104 * notification to avoid the corresponding overhead. As a hint, the
2105 * upper-level application might check the status of the device link once
2106 * being systematically returned a 0 value for a given number of tries.
2109 * The port identifier of the Ethernet device.
2111 * The index of the receive queue from which to retrieve input packets.
2112 * The value must be in the range [0, nb_rx_queue - 1] previously supplied
2113 * to rte_eth_dev_configure().
2115 * The address of an array of pointers to *rte_mbuf* structures that
2116 * must be large enough to store *nb_pkts* pointers in it.
2118 * The maximum number of packets to retrieve.
2120 * The number of packets actually retrieved, which is the number
2121 * of pointers to *rte_mbuf* structures effectively supplied to the
2124 #ifdef RTE_LIBRTE_ETHDEV_DEBUG
2125 extern uint16_t rte_eth_rx_burst(uint8_t port_id, uint16_t queue_id,
2126 struct rte_mbuf **rx_pkts, uint16_t nb_pkts);
2128 static inline uint16_t
2129 rte_eth_rx_burst(uint8_t port_id, uint16_t queue_id,
2130 struct rte_mbuf **rx_pkts, uint16_t nb_pkts)
2132 struct rte_eth_dev *dev;
2134 dev = &rte_eth_devices[port_id];
2135 return (*dev->rx_pkt_burst)(dev->data->rx_queues[queue_id], rx_pkts, nb_pkts);
2140 * Get the number of used descriptors in a specific queue
2143 * The port identifier of the Ethernet device.
2145 * The queue id on the specific port.
2147 * The number of used descriptors in the specific queue.
2149 #ifdef RTE_LIBRTE_ETHDEV_DEBUG
2150 extern uint32_t rte_eth_rx_queue_count(uint8_t port_id, uint16_t queue_id);
2152 static inline uint32_t
2153 rte_eth_rx_queue_count(uint8_t port_id, uint16_t queue_id)
2155 struct rte_eth_dev *dev;
2157 dev = &rte_eth_devices[port_id];
2158 return (*dev->dev_ops->rx_queue_count)(dev, queue_id);
2163 * Check if the DD bit of the specific RX descriptor in the queue has been set
2166 * The port identifier of the Ethernet device.
2168 * The queue id on the specific port.
2170 * The offset of the descriptor ID from tail.
2172 * - (1) if the specific DD bit is set.
2173 * - (0) if the specific DD bit is not set.
2174 * - (-ENODEV) if *port_id* invalid.
2176 #ifdef RTE_LIBRTE_ETHDEV_DEBUG
2177 extern int rte_eth_rx_descriptor_done(uint8_t port_id,
2182 rte_eth_rx_descriptor_done(uint8_t port_id, uint16_t queue_id, uint16_t offset)
2184 struct rte_eth_dev *dev;
2186 dev = &rte_eth_devices[port_id];
2187 return (*dev->dev_ops->rx_descriptor_done)( \
2188 dev->data->rx_queues[queue_id], offset);
2193 * Send a burst of output packets on a transmit queue of an Ethernet device.
2195 * The rte_eth_tx_burst() function is invoked to transmit output packets
2196 * on the output queue *queue_id* of the Ethernet device designated by its
2198 * The *nb_pkts* parameter is the number of packets to send which are
2199 * supplied in the *tx_pkts* array of *rte_mbuf* structures.
2200 * The rte_eth_tx_burst() function loops, sending *nb_pkts* packets,
2201 * up to the number of transmit descriptors available in the TX ring of the
2203 * For each packet to send, the rte_eth_tx_burst() function performs
2204 * the following operations:
2206 * - Pick up the next available descriptor in the transmit ring.
2208 * - Free the network buffer previously sent with that descriptor, if any.
2210 * - Initialize the transmit descriptor with the information provided
2211 * in the *rte_mbuf data structure.
2213 * In the case of a segmented packet composed of a list of *rte_mbuf* buffers,
2214 * the rte_eth_tx_burst() function uses several transmit descriptors
2217 * The rte_eth_tx_burst() function returns the number of packets it
2218 * actually sent. A return value equal to *nb_pkts* means that all packets
2219 * have been sent, and this is likely to signify that other output packets
2220 * could be immediately transmitted again. Applications that implement a
2221 * "send as many packets to transmit as possible" policy can check this
2222 * specific case and keep invoking the rte_eth_tx_burst() function until
2223 * a value less than *nb_pkts* is returned.
2225 * It is the responsibility of the rte_eth_tx_burst() function to
2226 * transparently free the memory buffers of packets previously sent.
2227 * This feature is driven by the *tx_free_thresh* value supplied to the
2228 * rte_eth_dev_configure() function at device configuration time.
2229 * When the number of previously sent packets reached the "minimum transmit
2230 * packets to free" threshold, the rte_eth_tx_burst() function must
2231 * [attempt to] free the *rte_mbuf* buffers of those packets whose
2232 * transmission was effectively completed.
2235 * The port identifier of the Ethernet device.
2237 * The index of the transmit queue through which output packets must be
2239 * The value must be in the range [0, nb_tx_queue - 1] previously supplied
2240 * to rte_eth_dev_configure().
2242 * The address of an array of *nb_pkts* pointers to *rte_mbuf* structures
2243 * which contain the output packets.
2245 * The maximum number of packets to transmit.
2247 * The number of output packets actually stored in transmit descriptors of
2248 * the transmit ring. The return value can be less than the value of the
2249 * *tx_pkts* parameter when the transmit ring is full or has been filled up.
2251 #ifdef RTE_LIBRTE_ETHDEV_DEBUG
2252 extern uint16_t rte_eth_tx_burst(uint8_t port_id, uint16_t queue_id,
2253 struct rte_mbuf **tx_pkts, uint16_t nb_pkts);
2255 static inline uint16_t
2256 rte_eth_tx_burst(uint8_t port_id, uint16_t queue_id,
2257 struct rte_mbuf **tx_pkts, uint16_t nb_pkts)
2259 struct rte_eth_dev *dev;
2261 dev = &rte_eth_devices[port_id];
2262 return (*dev->tx_pkt_burst)(dev->data->tx_queues[queue_id], tx_pkts, nb_pkts);
2267 * Setup a new signature filter rule on an Ethernet device
2270 * The port identifier of the Ethernet device.
2271 * @param fdir_filter
2272 * The pointer to the fdir filter structure describing the signature filter
2274 * The *rte_fdir_filter* structure includes the values of the different fields
2275 * to match: source and destination IP addresses, vlan id, flexbytes, source
2276 * and destination ports, and so on.
2278 * The index of the RX queue where to store RX packets matching the added
2279 * signature filter defined in fdir_filter.
2281 * - (0) if successful.
2282 * - (-ENOTSUP) if hardware doesn't support flow director mode.
2283 * - (-ENODEV) if *port_id* invalid.
2284 * - (-ENOSYS) if the FDIR mode is not configured in signature mode
2286 * - (-EINVAL) if the fdir_filter information is not correct.
2288 int rte_eth_dev_fdir_add_signature_filter(uint8_t port_id,
2289 struct rte_fdir_filter *fdir_filter,
2293 * Update a signature filter rule on an Ethernet device.
2294 * If the rule doesn't exits, it is created.
2297 * The port identifier of the Ethernet device.
2299 * The pointer to the structure describing the signature filter rule.
2300 * The *rte_fdir_filter* structure includes the values of the different fields
2301 * to match: source and destination IP addresses, vlan id, flexbytes, source
2302 * and destination ports, and so on.
2304 * The index of the RX queue where to store RX packets matching the added
2305 * signature filter defined in fdir_ftr.
2307 * - (0) if successful.
2308 * - (-ENOTSUP) if hardware doesn't support flow director mode.
2309 * - (-ENODEV) if *port_id* invalid.
2310 * - (-ENOSYS) if the flow director mode is not configured in signature mode
2312 * - (-EINVAL) if the fdir_filter information is not correct.
2314 int rte_eth_dev_fdir_update_signature_filter(uint8_t port_id,
2315 struct rte_fdir_filter *fdir_ftr,
2319 * Remove a signature filter rule on an Ethernet device.
2322 * The port identifier of the Ethernet device.
2324 * The pointer to the structure describing the signature filter rule.
2325 * The *rte_fdir_filter* structure includes the values of the different fields
2326 * to match: source and destination IP addresses, vlan id, flexbytes, source
2327 * and destination ports, and so on.
2329 * - (0) if successful.
2330 * - (-ENOTSUP) if hardware doesn't support flow director mode.
2331 * - (-ENODEV) if *port_id* invalid.
2332 * - (-ENOSYS) if the flow director mode is not configured in signature mode
2334 * - (-EINVAL) if the fdir_filter information is not correct.
2336 int rte_eth_dev_fdir_remove_signature_filter(uint8_t port_id,
2337 struct rte_fdir_filter *fdir_ftr);
2340 * Retrieve the flow director information of an Ethernet device.
2343 * The port identifier of the Ethernet device.
2345 * A pointer to a structure of type *rte_eth_dev_fdir* to be filled with
2346 * the flow director information of the Ethernet device.
2348 * - (0) if successful.
2349 * - (-ENOTSUP) if hardware doesn't support flow director mode.
2350 * - (-ENODEV) if *port_id* invalid.
2351 * - (-ENOSYS) if the flow director mode is not configured on *port_id*.
2353 int rte_eth_dev_fdir_get_infos(uint8_t port_id, struct rte_eth_fdir *fdir);
2356 * Add a new perfect filter rule on an Ethernet device.
2359 * The port identifier of the Ethernet device.
2360 * @param fdir_filter
2361 * The pointer to the structure describing the perfect filter rule.
2362 * The *rte_fdir_filter* structure includes the values of the different fields
2363 * to match: source and destination IP addresses, vlan id, flexbytes, source
2364 * and destination ports, and so on.
2365 * IPv6 are not supported.
2367 * The 16-bit value supplied in the field hash.fdir.id of mbuf for RX
2368 * packets matching the perfect filter.
2370 * The index of the RX queue where to store RX packets matching the added
2371 * perfect filter defined in fdir_filter.
2373 * If drop is set to 1, matching RX packets are stored into the RX drop
2374 * queue defined in the rte_fdir_conf.
2376 * - (0) if successful.
2377 * - (-ENOTSUP) if hardware doesn't support flow director mode.
2378 * - (-ENODEV) if *port_id* invalid.
2379 * - (-ENOSYS) if the flow director mode is not configured in perfect mode
2381 * - (-EINVAL) if the fdir_filter information is not correct.
2383 int rte_eth_dev_fdir_add_perfect_filter(uint8_t port_id,
2384 struct rte_fdir_filter *fdir_filter,
2385 uint16_t soft_id, uint8_t rx_queue,
2389 * Update a perfect filter rule on an Ethernet device.
2390 * If the rule doesn't exits, it is created.
2393 * The port identifier of the Ethernet device.
2394 * @param fdir_filter
2395 * The pointer to the structure describing the perfect filter rule.
2396 * The *rte_fdir_filter* structure includes the values of the different fields
2397 * to match: source and destination IP addresses, vlan id, flexbytes, source
2398 * and destination ports, and so on.
2399 * IPv6 are not supported.
2401 * The 16-bit value supplied in the field hash.fdir.id of mbuf for RX
2402 * packets matching the perfect filter.
2404 * The index of the RX queue where to store RX packets matching the added
2405 * perfect filter defined in fdir_filter.
2407 * If drop is set to 1, matching RX packets are stored into the RX drop
2408 * queue defined in the rte_fdir_conf.
2410 * - (0) if successful.
2411 * - (-ENOTSUP) if hardware doesn't support flow director mode.
2412 * - (-ENODEV) if *port_id* invalid.
2413 * - (-ENOSYS) if the flow director mode is not configured in perfect mode
2415 * - (-EINVAL) if the fdir_filter information is not correct.
2417 int rte_eth_dev_fdir_update_perfect_filter(uint8_t port_id,
2418 struct rte_fdir_filter *fdir_filter,
2419 uint16_t soft_id, uint8_t rx_queue,
2423 * Remove a perfect filter rule on an Ethernet device.
2426 * The port identifier of the Ethernet device.
2427 * @param fdir_filter
2428 * The pointer to the structure describing the perfect filter rule.
2429 * The *rte_fdir_filter* structure includes the values of the different fields
2430 * to match: source and destination IP addresses, vlan id, flexbytes, source
2431 * and destination ports, and so on.
2432 * IPv6 are not supported.
2434 * The soft_id value provided when adding/updating the removed filter.
2436 * - (0) if successful.
2437 * - (-ENOTSUP) if hardware doesn't support flow director mode.
2438 * - (-ENODEV) if *port_id* invalid.
2439 * - (-ENOSYS) if the flow director mode is not configured in perfect mode
2441 * - (-EINVAL) if the fdir_filter information is not correct.
2443 int rte_eth_dev_fdir_remove_perfect_filter(uint8_t port_id,
2444 struct rte_fdir_filter *fdir_filter,
2447 * Configure globally the masks for flow director mode for an Ethernet device.
2448 * For example, the device can match packets with only the first 24 bits of
2449 * the IPv4 source address.
2451 * The following fields can be masked: IPv4 addresses and L4 port numbers.
2452 * The following fields can be either enabled or disabled completely for the
2453 * matching functionality: VLAN ID tag; VLAN Priority + CFI bit; Flexible 2-byte
2455 * IPv6 masks are not supported.
2457 * All filters must comply with the masks previously configured.
2458 * For example, with a mask equal to 255.255.255.0 for the source IPv4 address,
2459 * all IPv4 filters must be created with a source IPv4 address that fits the
2462 * This function flushes all filters that have been previously added in
2466 * The port identifier of the Ethernet device.
2468 * The pointer to the fdir mask structure describing relevant headers fields
2469 * and relevant bits to use when matching packets addresses and ports.
2470 * IPv6 masks are not supported.
2472 * - (0) if successful.
2473 * - (-ENOTSUP) if hardware doesn't support flow director mode.
2474 * - (-ENODEV) if *port_id* invalid.
2475 * - (-ENOSYS) if the flow director mode is not configured in perfect
2476 * mode on *port_id*.
2477 * - (-EINVAL) if the fdir_filter information is not correct
2479 int rte_eth_dev_fdir_set_masks(uint8_t port_id,
2480 struct rte_fdir_masks *fdir_mask);
2483 * The eth device event type for interrupt, and maybe others in the future.
2485 enum rte_eth_event_type {
2486 RTE_ETH_EVENT_UNKNOWN, /**< unknown event type */
2487 RTE_ETH_EVENT_INTR_LSC, /**< lsc interrupt event */
2488 RTE_ETH_EVENT_MAX /**< max value of this enum */
2491 typedef void (*rte_eth_dev_cb_fn)(uint8_t port_id, \
2492 enum rte_eth_event_type event, void *cb_arg);
2493 /**< user application callback to be registered for interrupts */
2498 * Register a callback function for specific port id.
2505 * User supplied callback function to be called.
2507 * Pointer to the parameters for the registered callback.
2510 * - On success, zero.
2511 * - On failure, a negative value.
2513 int rte_eth_dev_callback_register(uint8_t port_id,
2514 enum rte_eth_event_type event,
2515 rte_eth_dev_cb_fn cb_fn, void *cb_arg);
2518 * Unregister a callback function for specific port id.
2525 * User supplied callback function to be called.
2527 * Pointer to the parameters for the registered callback. -1 means to
2528 * remove all for the same callback address and same event.
2531 * - On success, zero.
2532 * - On failure, a negative value.
2534 int rte_eth_dev_callback_unregister(uint8_t port_id,
2535 enum rte_eth_event_type event,
2536 rte_eth_dev_cb_fn cb_fn, void *cb_arg);
2539 * @internal Executes all the user application registered callbacks for
2540 * the specific device. It is for DPDK internal user only. User
2541 * application should not call it directly.
2544 * Pointer to struct rte_eth_dev.
2546 * Eth device interrupt event type.
2551 void _rte_eth_dev_callback_process(struct rte_eth_dev *dev,
2552 enum rte_eth_event_type event);
2555 * Turn on the LED on the Ethernet device.
2556 * This function turns on the LED on the Ethernet device.
2559 * The port identifier of the Ethernet device.
2561 * - (0) if successful.
2562 * - (-ENOTSUP) if underlying hardware OR driver doesn't support
2564 * - (-ENODEV) if *port_id* invalid.
2566 int rte_eth_led_on(uint8_t port_id);
2569 * Turn off the LED on the Ethernet device.
2570 * This function turns off the LED on the Ethernet device.
2573 * The port identifier of the Ethernet device.
2575 * - (0) if successful.
2576 * - (-ENOTSUP) if underlying hardware OR driver doesn't support
2578 * - (-ENODEV) if *port_id* invalid.
2580 int rte_eth_led_off(uint8_t port_id);
2583 * Configure the Ethernet link flow control for Ethernet device
2586 * The port identifier of the Ethernet device.
2588 * The pointer to the structure of the flow control parameters.
2590 * - (0) if successful.
2591 * - (-ENOTSUP) if hardware doesn't support flow control mode.
2592 * - (-ENODEV) if *port_id* invalid.
2593 * - (-EINVAL) if bad parameter
2594 * - (-EIO) if flow control setup failure
2596 int rte_eth_dev_flow_ctrl_set(uint8_t port_id,
2597 struct rte_eth_fc_conf *fc_conf);
2600 * Configure the Ethernet priority flow control under DCB environment
2601 * for Ethernet device.
2604 * The port identifier of the Ethernet device.
2606 * The pointer to the structure of the priority flow control parameters.
2608 * - (0) if successful.
2609 * - (-ENOTSUP) if hardware doesn't support priority flow control mode.
2610 * - (-ENODEV) if *port_id* invalid.
2611 * - (-EINVAL) if bad parameter
2612 * - (-EIO) if flow control setup failure
2614 int rte_eth_dev_priority_flow_ctrl_set(uint8_t port_id,
2615 struct rte_eth_pfc_conf *pfc_conf);
2618 * Add a MAC address to an internal array of addresses used to enable whitelist
2619 * filtering to accept packets only if the destination MAC address matches.
2622 * The port identifier of the Ethernet device.
2624 * The MAC address to add.
2626 * VMDq pool index to associate address with (if VMDq is enabled). If VMDq is
2627 * not enabled, this should be set to 0.
2629 * - (0) if successfully added or *mac_addr" was already added.
2630 * - (-ENOTSUP) if hardware doesn't support this feature.
2631 * - (-ENODEV) if *port* is invalid.
2632 * - (-ENOSPC) if no more MAC addresses can be added.
2633 * - (-EINVAL) if MAC address is invalid.
2635 int rte_eth_dev_mac_addr_add(uint8_t port, struct ether_addr *mac_addr,
2639 * Remove a MAC address from the internal array of addresses.
2642 * The port identifier of the Ethernet device.
2644 * MAC address to remove.
2646 * - (0) if successful, or *mac_addr* didn't exist.
2647 * - (-ENOTSUP) if hardware doesn't support.
2648 * - (-ENODEV) if *port* invalid.
2649 * - (-EADDRINUSE) if attempting to remove the default MAC address
2651 int rte_eth_dev_mac_addr_remove(uint8_t port, struct ether_addr *mac_addr);
2654 * Update Redirection Table(RETA) of Receive Side Scaling of Ethernet device.
2657 * The port identifier of the Ethernet device.
2661 * - (0) if successful.
2662 * - (-ENOTSUP) if hardware doesn't support.
2663 * - (-EINVAL) if bad parameter.
2665 int rte_eth_dev_rss_reta_update(uint8_t port,
2666 struct rte_eth_rss_reta *reta_conf);
2669 * Query Redirection Table(RETA) of Receive Side Scaling of Ethernet device.
2672 * The port identifier of the Ethernet device.
2676 * - (0) if successful.
2677 * - (-ENOTSUP) if hardware doesn't support.
2678 * - (-EINVAL) if bad parameter.
2680 int rte_eth_dev_rss_reta_query(uint8_t port,
2681 struct rte_eth_rss_reta *reta_conf);
2684 * Updates unicast hash table for receiving packet with the given destination
2685 * MAC address, and the packet is routed to all VFs for which the RX mode is
2686 * accept packets that match the unicast hash table.
2689 * The port identifier of the Ethernet device.
2691 * Unicast MAC address.
2693 * 1 - Set an unicast hash bit for receiving packets with the MAC address.
2694 * 0 - Clear an unicast hash bit.
2696 * - (0) if successful.
2697 * - (-ENOTSUP) if hardware doesn't support.
2698 * - (-ENODEV) if *port_id* invalid.
2699 * - (-EINVAL) if bad parameter.
2701 int rte_eth_dev_uc_hash_table_set(uint8_t port,struct ether_addr *addr,
2705 * Updates all unicast hash bitmaps for receiving packet with any Unicast
2706 * Ethernet MAC addresses,the packet is routed to all VFs for which the RX
2707 * mode is accept packets that match the unicast hash table.
2710 * The port identifier of the Ethernet device.
2712 * 1 - Set all unicast hash bitmaps for receiving all the Ethernet
2714 * 0 - Clear all unicast hash bitmaps
2716 * - (0) if successful.
2717 * - (-ENOTSUP) if hardware doesn't support.
2718 * - (-ENODEV) if *port_id* invalid.
2719 * - (-EINVAL) if bad parameter.
2721 int rte_eth_dev_uc_all_hash_table_set(uint8_t port,uint8_t on);
2724 * Set RX L2 Filtering mode of a VF of an Ethernet device.
2727 * The port identifier of the Ethernet device.
2731 * The RX mode mask, which is one or more of accepting Untagged Packets,
2732 * packets that match the PFUTA table, Broadcast and Multicast Promiscuous.
2733 * ETH_VMDQ_ACCEPT_UNTAG,ETH_VMDQ_ACCEPT_HASH_UC,
2734 * ETH_VMDQ_ACCEPT_BROADCAST and ETH_VMDQ_ACCEPT_MULTICAST will be used
2737 * 1 - Enable a VF RX mode.
2738 * 0 - Disable a VF RX mode.
2740 * - (0) if successful.
2741 * - (-ENOTSUP) if hardware doesn't support.
2742 * - (-ENOTSUP) if hardware doesn't support.
2743 * - (-EINVAL) if bad parameter.
2745 int rte_eth_dev_set_vf_rxmode(uint8_t port, uint16_t vf, uint16_t rx_mode,
2749 * Enable or disable a VF traffic transmit of the Ethernet device.
2752 * The port identifier of the Ethernet device.
2756 * 1 - Enable a VF traffic transmit.
2757 * 0 - Disable a VF traffic transmit.
2759 * - (0) if successful.
2760 * - (-ENODEV) if *port_id* invalid.
2761 * - (-ENOTSUP) if hardware doesn't support.
2762 * - (-EINVAL) if bad parameter.
2765 rte_eth_dev_set_vf_tx(uint8_t port,uint16_t vf, uint8_t on);
2768 * Enable or disable a VF traffic receive of an Ethernet device.
2771 * The port identifier of the Ethernet device.
2775 * 1 - Enable a VF traffic receive.
2776 * 0 - Disable a VF traffic receive.
2778 * - (0) if successful.
2779 * - (-ENOTSUP) if hardware doesn't support.
2780 * - (-ENODEV) if *port_id* invalid.
2781 * - (-EINVAL) if bad parameter.
2784 rte_eth_dev_set_vf_rx(uint8_t port,uint16_t vf, uint8_t on);
2787 * Enable/Disable hardware VF VLAN filtering by an Ethernet device of
2788 * received VLAN packets tagged with a given VLAN Tag Identifier.
2791 * The port identifier of the Ethernet device.
2793 * The VLAN Tag Identifier whose filtering must be enabled or disabled.
2795 * Bitmap listing which VFs participate in the VLAN filtering.
2797 * 1 - Enable VFs VLAN filtering.
2798 * 0 - Disable VFs VLAN filtering.
2800 * - (0) if successful.
2801 * - (-ENOTSUP) if hardware doesn't support.
2802 * - (-ENODEV) if *port_id* invalid.
2803 * - (-EINVAL) if bad parameter.
2806 rte_eth_dev_set_vf_vlan_filter(uint8_t port, uint16_t vlan_id,
2811 * Set a traffic mirroring rule on an Ethernet device
2814 * The port identifier of the Ethernet device.
2815 * @param mirror_conf
2816 * The pointer to the traffic mirroring structure describing the mirroring rule.
2817 * The *rte_eth_vm_mirror_conf* structure includes the type of mirroring rule,
2818 * destination pool and the value of rule if enable vlan or pool mirroring.
2821 * The index of traffic mirroring rule, we support four separated rules.
2823 * 1 - Enable a mirroring rule.
2824 * 0 - Disable a mirroring rule.
2826 * - (0) if successful.
2827 * - (-ENOTSUP) if hardware doesn't support this feature.
2828 * - (-ENODEV) if *port_id* invalid.
2829 * - (-EINVAL) if the mr_conf information is not correct.
2831 int rte_eth_mirror_rule_set(uint8_t port_id,
2832 struct rte_eth_vmdq_mirror_conf *mirror_conf,
2837 * Reset a traffic mirroring rule on an Ethernet device.
2840 * The port identifier of the Ethernet device.
2842 * The index of traffic mirroring rule, we support four separated rules.
2844 * - (0) if successful.
2845 * - (-ENOTSUP) if hardware doesn't support this feature.
2846 * - (-ENODEV) if *port_id* invalid.
2847 * - (-EINVAL) if bad parameter.
2849 int rte_eth_mirror_rule_reset(uint8_t port_id,
2853 * Set the rate limitation for a queue on an Ethernet device.
2856 * The port identifier of the Ethernet device.
2860 * The tx rate allocated from the total link speed for this queue.
2862 * - (0) if successful.
2863 * - (-ENOTSUP) if hardware doesn't support this feature.
2864 * - (-ENODEV) if *port_id* invalid.
2865 * - (-EINVAL) if bad parameter.
2867 int rte_eth_set_queue_rate_limit(uint8_t port_id, uint16_t queue_idx,
2871 * Set the rate limitation for a vf on an Ethernet device.
2874 * The port identifier of the Ethernet device.
2878 * The tx rate allocated from the total link speed for this VF id.
2880 * The queue mask which need to set the rate.
2882 * - (0) if successful.
2883 * - (-ENOTSUP) if hardware doesn't support this feature.
2884 * - (-ENODEV) if *port_id* invalid.
2885 * - (-EINVAL) if bad parameter.
2887 int rte_eth_set_vf_rate_limit(uint8_t port_id, uint16_t vf,
2888 uint16_t tx_rate, uint64_t q_msk);
2891 * Initialize bypass logic. This function needs to be called before
2892 * executing any other bypass API.
2895 * The port identifier of the Ethernet device.
2897 * - (0) if successful.
2898 * - (-ENOTSUP) if hardware doesn't support.
2899 * - (-EINVAL) if bad parameter.
2901 int rte_eth_dev_bypass_init(uint8_t port);
2904 * Return bypass state.
2907 * The port identifier of the Ethernet device.
2909 * The return bypass state.
2912 * - (3) Isolate mode
2914 * - (0) if successful.
2915 * - (-ENOTSUP) if hardware doesn't support.
2916 * - (-EINVAL) if bad parameter.
2918 int rte_eth_dev_bypass_state_show(uint8_t port, uint32_t *state);
2924 * The port identifier of the Ethernet device.
2926 * The current bypass state.
2929 * - (3) Isolate mode
2931 * - (0) if successful.
2932 * - (-ENOTSUP) if hardware doesn't support.
2933 * - (-EINVAL) if bad parameter.
2935 int rte_eth_dev_bypass_state_set(uint8_t port, uint32_t *new_state);
2938 * Return bypass state when given event occurs.
2941 * The port identifier of the Ethernet device.
2944 * - (1) Main power on (power button is pushed)
2945 * - (2) Auxiliary power on (power supply is being plugged)
2946 * - (3) Main power off (system shutdown and power supply is left plugged in)
2947 * - (4) Auxiliary power off (power supply is being unplugged)
2948 * - (5) Display or set the watchdog timer
2950 * The bypass state when given event occurred.
2953 * - (3) Isolate mode
2955 * - (0) if successful.
2956 * - (-ENOTSUP) if hardware doesn't support.
2957 * - (-EINVAL) if bad parameter.
2959 int rte_eth_dev_bypass_event_show(uint8_t port, uint32_t event, uint32_t *state);
2962 * Set bypass state when given event occurs.
2965 * The port identifier of the Ethernet device.
2968 * - (1) Main power on (power button is pushed)
2969 * - (2) Auxiliary power on (power supply is being plugged)
2970 * - (3) Main power off (system shutdown and power supply is left plugged in)
2971 * - (4) Auxiliary power off (power supply is being unplugged)
2972 * - (5) Display or set the watchdog timer
2974 * The assigned state when given event occurs.
2977 * - (3) Isolate mode
2979 * - (0) if successful.
2980 * - (-ENOTSUP) if hardware doesn't support.
2981 * - (-EINVAL) if bad parameter.
2983 int rte_eth_dev_bypass_event_store(uint8_t port, uint32_t event, uint32_t state);
2986 * Set bypass watchdog timeout count.
2989 * The port identifier of the Ethernet device.
2991 * The timeout to be set.
2992 * - (0) 0 seconds (timer is off)
3001 * - (0) if successful.
3002 * - (-ENOTSUP) if hardware doesn't support.
3003 * - (-EINVAL) if bad parameter.
3005 int rte_eth_dev_wd_timeout_store(uint8_t port, uint32_t timeout);
3008 * Get bypass firmware version.
3011 * The port identifier of the Ethernet device.
3013 * The firmware version
3015 * - (0) if successful.
3016 * - (-ENOTSUP) if hardware doesn't support.
3017 * - (-EINVAL) if bad parameter.
3019 int rte_eth_dev_bypass_ver_show(uint8_t port, uint32_t *ver);
3022 * Return bypass watchdog timeout in seconds
3025 * The port identifier of the Ethernet device.
3027 * The return watchdog timeout. "0" represents timer expired
3028 * - (0) 0 seconds (timer is off)
3037 * - (0) if successful.
3038 * - (-ENOTSUP) if hardware doesn't support.
3039 * - (-EINVAL) if bad parameter.
3041 int rte_eth_dev_bypass_wd_timeout_show(uint8_t port, uint32_t *wd_timeout);
3044 * Reset bypass watchdog timer
3047 * The port identifier of the Ethernet device.
3049 * - (0) if successful.
3050 * - (-ENOTSUP) if hardware doesn't support.
3051 * - (-EINVAL) if bad parameter.
3053 int rte_eth_dev_bypass_wd_reset(uint8_t port);
3056 * Configuration of Receive Side Scaling hash computation of Ethernet device.
3059 * The port identifier of the Ethernet device.
3061 * The new configuration to use for RSS hash computation on the port.
3063 * - (0) if successful.
3064 * - (-ENODEV) if port identifier is invalid.
3065 * - (-ENOTSUP) if hardware doesn't support.
3066 * - (-EINVAL) if bad parameter.
3068 int rte_eth_dev_rss_hash_update(uint8_t port_id,
3069 struct rte_eth_rss_conf *rss_conf);
3072 * Retrieve current configuration of Receive Side Scaling hash computation
3073 * of Ethernet device.
3076 * The port identifier of the Ethernet device.
3078 * Where to store the current RSS hash configuration of the Ethernet device.
3080 * - (0) if successful.
3081 * - (-ENODEV) if port identifier is invalid.
3082 * - (-ENOTSUP) if hardware doesn't support RSS.
3085 rte_eth_dev_rss_hash_conf_get(uint8_t port_id,
3086 struct rte_eth_rss_conf *rss_conf);
3092 * The port identifier of the Ethernet device.
3094 * The index of RX queue where to store RX packets matching the syn filter.
3096 * The pointer to the structure describing the syn filter rule.
3098 * - (0) if successful.
3099 * - (-ENOTSUP) if hardware doesn't support.
3100 * - (-EINVAL) if bad parameter.
3102 int rte_eth_dev_add_syn_filter(uint8_t port_id,
3103 struct rte_syn_filter *filter, uint16_t rx_queue);
3109 * The port identifier of the Ethernet device.
3111 * - (0) if successful.
3112 * - (-ENOTSUP) if hardware doesn't support.
3113 * - (-EINVAL) if bad parameter.
3115 int rte_eth_dev_remove_syn_filter(uint8_t port_id);
3121 * The port identifier of the Ethernet device.
3123 * The pointer to the structure describing the syn filter.
3125 * A pointer to get the queue index of syn filter.
3127 * - (0) if successful.
3128 * - (-ENOTSUP) if hardware doesn't support.
3129 * - (-EINVAL) if bad parameter.
3131 int rte_eth_dev_get_syn_filter(uint8_t port_id,
3132 struct rte_syn_filter *filter, uint16_t *rx_queue);
3135 * Add a new ethertype filter rule on an Ethernet device.
3138 * The port identifier of the Ethernet device.
3140 * The identifier of ethertype filter.
3142 * The pointer to the structure describing the ethertype filter rule.
3143 * The *rte_ethertype_filter* structure includes the values of the different
3144 * fields to match: ethertype and priority in vlan tag.
3145 * priority in vlan tag is not supported for E1000 dev.
3147 * The index of the RX queue where to store RX packets matching the added
3150 * - (0) if successful.
3151 * - (-ENOTSUP) if hardware doesn't support ethertype filter.
3152 * - (-ENODEV) if *port_id* invalid.
3153 * - (-EINVAL) if the filter information is not correct.
3155 int rte_eth_dev_add_ethertype_filter(uint8_t port_id, uint16_t index,
3156 struct rte_ethertype_filter *filter, uint16_t rx_queue);
3159 * remove an ethertype filter rule on an Ethernet device.
3162 * The port identifier of the Ethernet device.
3164 * The identifier of ethertype filter.
3166 * - (0) if successful.
3167 * - (-ENOTSUP) if hardware doesn't support ethertype filter.
3168 * - (-ENODEV) if *port_id* invalid.
3169 * - (-EINVAL) if the filter information is not correct.
3171 int rte_eth_dev_remove_ethertype_filter(uint8_t port_id,
3175 * Get an ethertype filter rule on an Ethernet device.
3178 * The port identifier of the Ethernet device.
3180 * The identifier of ethertype filter.
3182 * A pointer to a structure of type *rte_ethertype_filter* to be filled with
3183 * the information of the Ethertype filter.
3185 * A pointer to get the queue index.
3187 * - (0) if successful.
3188 * - (-ENOTSUP) if hardware doesn't support ethertype filter.
3189 * - (-ENODEV) if *port_id* invalid.
3190 * - (-EINVAL) if the filter information is not correct.
3191 * - (-ENOENT) if no enabled filter in this index.
3193 int rte_eth_dev_get_ethertype_filter(uint8_t port_id, uint16_t index,
3194 struct rte_ethertype_filter *filter, uint16_t *rx_queue);
3197 * Add a new 2tuple filter rule on an Ethernet device.
3200 * The port identifier of the Ethernet device.
3202 * The identifier of 2tuple filter.
3204 * The pointer to the structure describing the 2tuple filter rule.
3205 * The *rte_2tuple_filter* structure includes the values of the different
3206 * fields to match: protocol, dst_port and
3207 * tcp_flags if the protocol is tcp type.
3209 * The index of the RX queue where to store RX packets matching the added
3212 * - (0) if successful.
3213 * - (-ENOTSUP) if hardware doesn't support 2tuple filter.
3214 * - (-ENODEV) if *port_id* invalid.
3215 * - (-EINVAL) if the filter information is not correct.
3217 int rte_eth_dev_add_2tuple_filter(uint8_t port_id, uint16_t index,
3218 struct rte_2tuple_filter *filter, uint16_t rx_queue);
3221 * remove a 2tuple filter rule on an Ethernet device.
3224 * The port identifier of the Ethernet device.
3226 * The identifier of 2tuple filter.
3228 * - (0) if successful.
3229 * - (-ENOTSUP) if hardware doesn't support 2tuple filter.
3230 * - (-ENODEV) if *port_id* invalid.
3231 * - (-EINVAL) if the filter information is not correct.
3233 int rte_eth_dev_remove_2tuple_filter(uint8_t port_id, uint16_t index);
3236 * Get an 2tuple filter rule on an Ethernet device.
3239 * The port identifier of the Ethernet device.
3241 * The identifier of 2tuple filter.
3243 * A pointer to a structure of type *rte_2tuple_filter* to be filled with
3244 * the information of the 2tuple filter.
3246 * A pointer to get the queue index.
3248 * - (0) if successful.
3249 * - (-ENOTSUP) if hardware doesn't support 2tuple filter.
3250 * - (-ENODEV) if *port_id* invalid.
3251 * - (-EINVAL) if the filter information is not correct.
3252 * - (-ENOENT) if no enabled filter in this index.
3254 int rte_eth_dev_get_2tuple_filter(uint8_t port_id, uint16_t index,
3255 struct rte_2tuple_filter *filter, uint16_t *rx_queue);
3258 * Add a new 5tuple filter rule on an Ethernet device.
3261 * The port identifier of the Ethernet device.
3263 * The identifier of 5tuple filter.
3265 * The pointer to the structure describing the 5tuple filter rule.
3266 * The *rte_5tuple_filter* structure includes the values of the different
3267 * fields to match: dst src IP, dst src port, protocol and relative masks
3269 * The index of the RX queue where to store RX packets matching the added
3272 * - (0) if successful.
3273 * - (-ENOTSUP) if hardware doesn't support 5tuple filter.
3274 * - (-ENODEV) if *port_id* invalid.
3275 * - (-EINVAL) if the filter information is not correct.
3277 int rte_eth_dev_add_5tuple_filter(uint8_t port_id, uint16_t index,
3278 struct rte_5tuple_filter *filter, uint16_t rx_queue);
3281 * remove a 5tuple filter rule on an Ethernet device.
3284 * The port identifier of the Ethernet device.
3286 * The identifier of 5tuple filter.
3288 * - (0) if successful.
3289 * - (-ENOTSUP) if hardware doesn't support 5tuple filter.
3290 * - (-ENODEV) if *port_id* invalid.
3291 * - (-EINVAL) if the filter information is not correct.
3293 int rte_eth_dev_remove_5tuple_filter(uint8_t port_id, uint16_t index);
3296 * Get an 5tuple filter rule on an Ethernet device.
3299 * The port identifier of the Ethernet device.
3301 * The identifier of 5tuple filter.
3303 * A pointer to a structure of type *rte_5tuple_filter* to be filled with
3304 * the information of the 5tuple filter.
3306 * A pointer to get the queue index.
3308 * - (0) if successful.
3309 * - (-ENOTSUP) if hardware doesn't support 5tuple filter.
3310 * - (-ENODEV) if *port_id* invalid.
3311 * - (-EINVAL) if the filter information is not correct.
3313 int rte_eth_dev_get_5tuple_filter(uint8_t port_id, uint16_t index,
3314 struct rte_5tuple_filter *filter, uint16_t *rx_queue);
3317 * Add a new flex filter rule on an Ethernet device.
3320 * The port identifier of the Ethernet device.
3322 * The identifier of flex filter.
3324 * The pointer to the structure describing the flex filter rule.
3325 * The *rte_flex_filter* structure includes the values of the different fields
3326 * to match: the dwords (first len bytes of packet ) and relative masks.
3328 * The index of the RX queue where to store RX packets matching the added
3331 * - (0) if successful.
3332 * - (-ENOTSUP) if hardware doesn't support flex filter.
3333 * - (-ENODEV) if *port_id* invalid.
3334 * - (-EINVAL) if the filter information is not correct.
3335 * - (-ENOENT) if no enabled filter in this index.
3337 int rte_eth_dev_add_flex_filter(uint8_t port_id, uint16_t index,
3338 struct rte_flex_filter *filter, uint16_t rx_queue);
3341 * remove a flex filter rule on an Ethernet device.
3344 * The port identifier of the Ethernet device.
3346 * The identifier of flex filter.
3348 * - (0) if successful.
3349 * - (-ENOTSUP) if hardware doesn't support flex filter.
3350 * - (-ENODEV) if *port_id* invalid.
3351 * - (-EINVAL) if the filter information is not correct.
3353 int rte_eth_dev_remove_flex_filter(uint8_t port_id, uint16_t index);
3356 * Get an flex filter rule on an Ethernet device.
3359 * The port identifier of the Ethernet device.
3361 * The identifier of flex filter.
3363 * A pointer to a structure of type *rte_flex_filter* to be filled with
3364 * the information of the flex filter.
3366 * A pointer to get the queue index.
3368 * - (0) if successful.
3369 * - (-ENOTSUP) if hardware doesn't support flex filter.
3370 * - (-ENODEV) if *port_id* invalid.
3371 * - (-EINVAL) if the filter information is not correct.
3372 * - (-ENOENT) if no enabled filter in this index.
3374 int rte_eth_dev_get_flex_filter(uint8_t port_id, uint16_t index,
3375 struct rte_flex_filter *filter, uint16_t *rx_queue);
3381 #endif /* _RTE_ETHDEV_H_ */