4 * Copyright(c) 2010-2013 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.
35 #ifndef _RTE_ETHDEV_H_
36 #define _RTE_ETHDEV_H_
41 * RTE Ethernet Device API
43 * The Ethernet Device API is composed of two parts:
45 * - The application-oriented Ethernet API that includes functions to setup
46 * an Ethernet device (configure it, setup its RX and TX queues and start it),
47 * to get its MAC address, the speed and the status of its physical link,
48 * to receive and to transmit packets, and so on.
50 * - The driver-oriented Ethernet API that exports a function allowing
51 * an Ethernet Poll Mode Driver (PMD) to simultaneously register itself as
52 * an Ethernet device driver and as a PCI driver for a set of matching PCI
53 * [Ethernet] devices classes.
55 * By default, all the functions of the Ethernet Device API exported by a PMD
56 * are lock-free functions which assume to not be invoked in parallel on
57 * different logical cores to work on the same target object. For instance,
58 * the receive function of a PMD cannot be invoked in parallel on two logical
59 * cores to poll the same RX queue [of the same port]. Of course, this function
60 * can be invoked in parallel by different logical cores on different RX queues.
61 * It is the responsibility of the upper level application to enforce this rule.
63 * If needed, parallel accesses by multiple logical cores to shared queues
64 * shall be explicitly protected by dedicated inline lock-aware functions
65 * built on top of their corresponding lock-free functions of the PMD API.
67 * In all functions of the Ethernet API, the Ethernet device is
68 * designated by an integer >= 0 named the device port identifier.
70 * At the Ethernet driver level, Ethernet devices are represented by a generic
71 * data structure of type *rte_eth_dev*.
73 * Ethernet devices are dynamically registered during the PCI probing phase
74 * performed at EAL initialization time.
75 * When an Ethernet device is being probed, an *rte_eth_dev* structure and
76 * a new port identifier are allocated for that device. Then, the eth_dev_init()
77 * function supplied by the Ethernet driver matching the probed PCI
78 * device is invoked to properly initialize the device.
80 * The role of the device init function consists of resetting the hardware,
81 * checking access to Non-volatile Memory (NVM), reading the MAC address
84 * If the device init operation is successful, the correspondence between
85 * the port identifier assigned to the new device and its associated
86 * *rte_eth_dev* structure is effectively registered.
87 * Otherwise, both the *rte_eth_dev* structure and the port identifier are
90 * The functions exported by the application Ethernet API to setup a device
91 * designated by its port identifier must be invoked in the following order:
92 * - rte_eth_dev_configure()
93 * - rte_eth_tx_queue_setup()
94 * - rte_eth_rx_queue_setup()
95 * - rte_eth_dev_start()
97 * Then, the network application can invoke, in any order, the functions
98 * exported by the Ethernet API to get the MAC address of a given device, to
99 * get the speed and the status of a device physical link, to receive/transmit
100 * [burst of] packets, and so on.
102 * If the application wants to change the configuration (i.e. call
103 * rte_eth_dev_configure(), rte_eth_tx_queue_setup(), or
104 * rte_eth_rx_queue_setup()), it must call rte_eth_dev_stop() first to stop the
105 * device and then do the reconfiguration before calling rte_eth_dev_start()
106 * again. The tramsit and receive functions should not be invoked when the
109 * Please note that some configuration is not stored between calls to
110 * rte_eth_dev_stop()/rte_eth_dev_start(). The following configuration will
113 * - flow control settings
114 * - receive mode configuration (promiscuous mode, hardware checksum mode,
115 * RSS/VMDQ settings etc.)
116 * - VLAN filtering configuration
117 * - MAC addresses supplied to MAC address array
118 * - flow director filtering mode (but not filtering rules)
119 * - NIC queue statistics mappings
121 * Any other configuration will not be stored and will need to be re-entered
122 * after a call to rte_eth_dev_start().
124 * Finally, a network application can close an Ethernet device by invoking the
125 * rte_eth_dev_close() function.
127 * Each function of the application Ethernet API invokes a specific function
128 * of the PMD that controls the target device designated by its port
130 * For this purpose, all device-specific functions of an Ethernet driver are
131 * supplied through a set of pointers contained in a generic structure of type
133 * The address of the *eth_dev_ops* structure is stored in the *rte_eth_dev*
134 * structure by the device init function of the Ethernet driver, which is
135 * invoked during the PCI probing phase, as explained earlier.
137 * In other words, each function of the Ethernet API simply retrieves the
138 * *rte_eth_dev* structure associated with the device port identifier and
139 * performs an indirect invocation of the corresponding driver function
140 * supplied in the *eth_dev_ops* structure of the *rte_eth_dev* structure.
142 * For performance reasons, the address of the burst-oriented RX and TX
143 * functions of the Ethernet driver are not contained in the *eth_dev_ops*
144 * structure. Instead, they are directly stored at the beginning of the
145 * *rte_eth_dev* structure to avoid an extra indirect memory access during
148 * RTE ethernet device drivers do not use interrupts for transmitting or
149 * receiving. Instead, Ethernet drivers export Poll-Mode receive and transmit
150 * functions to applications.
151 * Both receive and transmit functions are packet-burst oriented to minimize
152 * their cost per packet through the following optimizations:
154 * - Sharing among multiple packets the incompressible cost of the
155 * invocation of receive/transmit functions.
157 * - Enabling receive/transmit functions to take advantage of burst-oriented
158 * hardware features (L1 cache, prefetch instructions, NIC head/tail
159 * registers) to minimize the number of CPU cycles per packet, for instance,
160 * by avoiding useless read memory accesses to ring descriptors, or by
161 * systematically using arrays of pointers that exactly fit L1 cache line
162 * boundaries and sizes.
164 * The burst-oriented receive function does not provide any error notification,
165 * to avoid the corresponding overhead. As a hint, the upper-level application
166 * might check the status of the device link once being systematically returned
167 * a 0 value by the receive function of the driver for a given number of tries.
177 #include <rte_interrupts.h>
179 #include <rte_mbuf.h>
180 #include "rte_ether.h"
183 * A structure used to retrieve statistics for an Ethernet port.
185 struct rte_eth_stats {
186 uint64_t ipackets; /**< Total number of successfully received packets. */
187 uint64_t opackets; /**< Total number of successfully transmitted packets.*/
188 uint64_t ibytes; /**< Total number of successfully received bytes. */
189 uint64_t obytes; /**< Total number of successfully transmitted bytes. */
190 uint64_t ierrors; /**< Total number of erroneous received packets. */
191 uint64_t oerrors; /**< Total number of failed transmitted packets. */
192 uint64_t imcasts; /**< Total number of multicast received packets. */
193 uint64_t rx_nombuf; /**< Total number of RX mbuf allocation failures. */
194 uint64_t fdirmatch; /**< Total number of RX packets matching a filter. */
195 uint64_t fdirmiss; /**< Total number of RX packets not matching any filter. */
196 uint64_t q_ipackets[RTE_ETHDEV_QUEUE_STAT_CNTRS];
197 /**< Total number of queue RX packets. */
198 uint64_t q_opackets[RTE_ETHDEV_QUEUE_STAT_CNTRS];
199 /**< Total number of queue TX packets. */
200 uint64_t q_ibytes[RTE_ETHDEV_QUEUE_STAT_CNTRS];
201 /**< Total number of successfully received queue bytes. */
202 uint64_t q_obytes[RTE_ETHDEV_QUEUE_STAT_CNTRS];
203 /**< Total number of successfully transmitted queue bytes. */
204 uint64_t q_errors[RTE_ETHDEV_QUEUE_STAT_CNTRS];
205 /**< Total number of queue packets received that are dropped. */
207 /**< Total number of good packets received from loopback,VF Only */
209 /**< Total number of good packets transmitted to loopback,VF Only */
211 /**< Total number of good bytes received from loopback,VF Only */
213 /**< Total number of good bytes transmitted to loopback,VF Only */
217 * A structure used to retrieve link-level information of an Ethernet port.
219 struct rte_eth_link {
220 uint16_t link_speed; /**< ETH_LINK_SPEED_[10, 100, 1000, 10000] */
221 uint16_t link_duplex; /**< ETH_LINK_[HALF_DUPLEX, FULL_DUPLEX] */
222 uint8_t link_status : 1; /**< 1 -> link up, 0 -> link down */
223 }__attribute__((aligned(8))); /**< aligned for atomic64 read/write */
225 #define ETH_LINK_SPEED_AUTONEG 0 /**< Auto-negotiate link speed. */
226 #define ETH_LINK_SPEED_10 10 /**< 10 megabits/second. */
227 #define ETH_LINK_SPEED_100 100 /**< 100 megabits/second. */
228 #define ETH_LINK_SPEED_1000 1000 /**< 1 gigabits/second. */
229 #define ETH_LINK_SPEED_10000 10000 /**< 10 gigabits/second. */
231 #define ETH_LINK_AUTONEG_DUPLEX 0 /**< Auto-negotiate duplex. */
232 #define ETH_LINK_HALF_DUPLEX 1 /**< Half-duplex connection. */
233 #define ETH_LINK_FULL_DUPLEX 2 /**< Full-duplex connection. */
236 * A structure used to configure the ring threshold registers of an RX/TX
237 * queue for an Ethernet port.
239 struct rte_eth_thresh {
240 uint8_t pthresh; /**< Ring prefetch threshold. */
241 uint8_t hthresh; /**< Ring host threshold. */
242 uint8_t wthresh; /**< Ring writeback threshold. */
246 * A set of values to identify what method is to be used to route
247 * packets to multiple queues.
249 enum rte_eth_rx_mq_mode {
250 ETH_RSS = 0, /**< Default to RSS mode */
251 ETH_VMDQ_DCB, /**< Use VMDQ+DCB to route traffic to queues */
252 ETH_DCB_RX /**< For RX side,only DCB is on. */
256 * A set of values to identify what method is to be used to transmit
257 * packets using multi-TCs.
259 enum rte_eth_tx_mq_mode {
260 ETH_DCB_NONE = 0, /**< It is not in DCB mode. */
261 ETH_VMDQ_DCB_TX, /**< For TX side,both DCB and VT is on. */
262 ETH_DCB_TX /**< For TX side,only DCB is on. */
266 * A structure used to configure the RX features of an Ethernet port.
268 struct rte_eth_rxmode {
269 /** The multi-queue packet distribution mode to be used, e.g. RSS. */
270 enum rte_eth_rx_mq_mode mq_mode;
271 uint32_t max_rx_pkt_len; /**< Only used if jumbo_frame enabled. */
272 uint16_t split_hdr_size; /**< hdr buf size (header_split enabled).*/
273 uint8_t header_split : 1, /**< Header Split enable. */
274 hw_ip_checksum : 1, /**< IP/UDP/TCP checksum offload enable. */
275 hw_vlan_filter : 1, /**< VLAN filter enable. */
276 hw_vlan_strip : 1, /**< VLAN strip enable. */
277 hw_vlan_extend : 1, /**< Extended VLAN enable. */
278 jumbo_frame : 1, /**< Jumbo Frame Receipt enable. */
279 hw_strip_crc : 1; /**< Enable CRC stripping by hardware. */
283 * A structure used to configure the Receive Side Scaling (RSS) feature
284 * of an Ethernet port.
285 * If not NULL, the *rss_key* pointer of the *rss_conf* structure points
286 * to an array of 40 bytes holding the RSS key to use for hashing specific
287 * header fields of received packets.
288 * Otherwise, a default random hash key is used by the device driver.
290 * The *rss_hf* field of the *rss_conf* structure indicates the different
291 * types of IPv4/IPv6 packets to which the RSS hashing must be applied.
292 * Supplying an *rss_hf* equal to zero disables the RSS feature.
294 struct rte_eth_rss_conf {
295 uint8_t *rss_key; /**< If not NULL, 40-byte hash key. */
296 uint16_t rss_hf; /**< Hash functions to apply - see below. */
299 #define ETH_RSS_IPV4 0x0001 /**< IPv4 packet. */
300 #define ETH_RSS_IPV4_TCP 0x0002 /**< IPv4/TCP packet. */
301 #define ETH_RSS_IPV6 0x0004 /**< IPv6 packet. */
302 #define ETH_RSS_IPV6_EX 0x0008 /**< IPv6 packet with extension headers.*/
303 #define ETH_RSS_IPV6_TCP 0x0010 /**< IPv6/TCP packet. */
304 #define ETH_RSS_IPV6_TCP_EX 0x0020 /**< IPv6/TCP with extension headers. */
305 /* Intel RSS extensions to UDP packets */
306 #define ETH_RSS_IPV4_UDP 0x0040 /**< IPv4/UDP packet. */
307 #define ETH_RSS_IPV6_UDP 0x0080 /**< IPv6/UDP packet. */
308 #define ETH_RSS_IPV6_UDP_EX 0x0100 /**< IPv6/UDP with extension headers. */
310 /* Definitions used for VMDQ and DCB functionality */
311 #define ETH_VMDQ_MAX_VLAN_FILTERS 64 /**< Maximum nb. of VMDQ vlan filters. */
312 #define ETH_DCB_NUM_USER_PRIORITIES 8 /**< Maximum nb. of DCB priorities. */
313 #define ETH_VMDQ_DCB_NUM_QUEUES 128 /**< Maximum nb. of VMDQ DCB queues. */
314 #define ETH_DCB_NUM_QUEUES 128 /**< Maximum nb. of DCB queues. */
316 /* DCB capability defines */
317 #define ETH_DCB_PG_SUPPORT 0x00000001 /**< Priority Group(ETS) support. */
318 #define ETH_DCB_PFC_SUPPORT 0x00000002 /**< Priority Flow Control support. */
320 /* Definitions used for VLAN Offload functionality */
321 #define ETH_VLAN_STRIP_OFFLOAD 0x0001 /**< VLAN Strip On/Off */
322 #define ETH_VLAN_FILTER_OFFLOAD 0x0002 /**< VLAN Filter On/Off */
323 #define ETH_VLAN_EXTEND_OFFLOAD 0x0004 /**< VLAN Extend On/Off */
325 /* Definitions used for mask VLAN setting */
326 #define ETH_VLAN_STRIP_MASK 0x0001 /**< VLAN Strip setting mask */
327 #define ETH_VLAN_FILTER_MASK 0x0002 /**< VLAN Filter setting mask*/
328 #define ETH_VLAN_EXTEND_MASK 0x0004 /**< VLAN Extend setting mask*/
331 * This enum indicates the possible number of traffic classes
332 * in DCB configratioins
334 enum rte_eth_nb_tcs {
335 ETH_4_TCS = 4, /**< 4 TCs with DCB. */
336 ETH_8_TCS = 8 /**< 8 TCs with DCB. */
340 * This enum indicates the possible number of queue pools
341 * in VMDQ+DCB configurations.
343 enum rte_eth_nb_pools {
344 ETH_16_POOLS = 16, /**< 16 pools with DCB. */
345 ETH_32_POOLS = 32 /**< 32 pools with DCB. */
348 /* This structure may be extended in future. */
349 struct rte_eth_dcb_rx_conf {
350 enum rte_eth_nb_tcs nb_tcs; /**< Possible DCB TCs, 4 or 8 TCs */
351 uint8_t dcb_queue[ETH_DCB_NUM_USER_PRIORITIES];
352 /**< Possible DCB queue,4 or 8. */
355 struct rte_eth_vmdq_dcb_tx_conf {
356 enum rte_eth_nb_pools nb_queue_pools; /**< With DCB, 16 or 32 pools. */
357 uint8_t dcb_queue[ETH_DCB_NUM_USER_PRIORITIES];
358 /**< Possible DCB queue,4 or 8. */
361 struct rte_eth_dcb_tx_conf {
362 enum rte_eth_nb_tcs nb_tcs; /**< possible DCB TCs, 4 or 8 TCs. */
363 uint8_t dcb_queue[ETH_DCB_NUM_USER_PRIORITIES];
364 /**< Possible DCB queue,4 or 8. */
368 * A structure used to configure the VMDQ+DCB feature
369 * of an Ethernet port.
371 * Using this feature, packets are routed to a pool of queues, based
372 * on the vlan id in the vlan tag, and then to a specific queue within
373 * that pool, using the user priority vlan tag field.
375 * A default pool may be used, if desired, to route all traffic which
376 * does not match the vlan filter rules.
378 struct rte_eth_vmdq_dcb_conf {
379 enum rte_eth_nb_pools nb_queue_pools; /**< With DCB, 16 or 32 pools */
380 uint8_t enable_default_pool; /**< If non-zero, use a default pool */
381 uint8_t default_pool; /**< The default pool, if applicable */
382 uint8_t nb_pool_maps; /**< We can have up to 64 filters/mappings */
384 uint16_t vlan_id; /**< The vlan id of the received frame */
385 uint64_t pools; /**< Bitmask of pools for packet rx */
386 } pool_map[ETH_VMDQ_MAX_VLAN_FILTERS]; /**< VMDq vlan pool maps. */
387 uint8_t dcb_queue[ETH_DCB_NUM_USER_PRIORITIES];
388 /**< Selects a queue in a pool */
392 * A structure used to configure the TX features of an Ethernet port.
394 struct rte_eth_txmode {
395 enum rte_eth_tx_mq_mode mq_mode; /**< TX multi-queues mode. */
399 * A structure used to configure an RX ring of an Ethernet port.
401 struct rte_eth_rxconf {
402 struct rte_eth_thresh rx_thresh; /**< RX ring threshold registers. */
403 uint16_t rx_free_thresh; /**< Drives the freeing of RX descriptors. */
404 uint8_t rx_drop_en; /**< Drop packets if no descriptors are available. */
407 #define ETH_TXQ_FLAGS_NOMULTSEGS 0x0001 /**< nb_segs=1 for all mbufs */
408 #define ETH_TXQ_FLAGS_NOREFCOUNT 0x0002 /**< refcnt can be ignored */
409 #define ETH_TXQ_FLAGS_NOMULTMEMP 0x0004 /**< all bufs come from same mempool */
410 #define ETH_TXQ_FLAGS_NOVLANOFFL 0x0100 /**< disable VLAN offload */
411 #define ETH_TXQ_FLAGS_NOXSUMSCTP 0x0200 /**< disable SCTP checksum offload */
412 #define ETH_TXQ_FLAGS_NOXSUMUDP 0x0400 /**< disable UDP checksum offload */
413 #define ETH_TXQ_FLAGS_NOXSUMTCP 0x0800 /**< disable TCP checksum offload */
414 #define ETH_TXQ_FLAGS_NOOFFLOADS \
415 (ETH_TXQ_FLAGS_NOVLANOFFL | ETH_TXQ_FLAGS_NOXSUMSCTP | \
416 ETH_TXQ_FLAGS_NOXSUMUDP | ETH_TXQ_FLAGS_NOXSUMTCP)
418 * A structure used to configure a TX ring of an Ethernet port.
420 struct rte_eth_txconf {
421 struct rte_eth_thresh tx_thresh; /**< TX ring threshold registers. */
422 uint16_t tx_rs_thresh; /**< Drives the setting of RS bit on TXDs. */
423 uint16_t tx_free_thresh; /**< Drives the freeing of TX buffers. */
424 uint32_t txq_flags; /**< Set flags for the Tx queue */
428 * This enum indicates the flow control mode
430 enum rte_eth_fc_mode {
431 RTE_FC_NONE = 0, /**< Disable flow control. */
432 RTE_FC_RX_PAUSE, /**< RX pause frame, enable flowctrl on TX side. */
433 RTE_FC_TX_PAUSE, /**< TX pause frame, enable flowctrl on RX side. */
434 RTE_FC_FULL /**< Enable flow control on both side. */
438 * A structure used to configure Ethernet flow control parameter.
439 * These parameters will be configured into the register of the NIC.
440 * Please refer to the corresponding data sheet for proper value.
442 struct rte_eth_fc_conf {
443 uint32_t high_water; /**< High threshold value to trigger XOFF */
444 uint32_t low_water; /**< Low threshold value to trigger XON */
445 uint16_t pause_time; /**< Pause quota in the Pause frame */
446 uint16_t send_xon; /**< Is XON frame need be sent */
447 enum rte_eth_fc_mode mode; /**< Link flow control mode */
451 * A structure used to configure Ethernet priority flow control parameter.
452 * These parameters will be configured into the register of the NIC.
453 * Please refer to the corresponding data sheet for proper value.
455 struct rte_eth_pfc_conf {
456 struct rte_eth_fc_conf fc; /**< General flow control parameter. */
457 uint8_t priority; /**< VLAN User Priority. */
461 * Flow Director setting modes: none (default), signature or perfect.
464 RTE_FDIR_MODE_NONE = 0, /**< Disable FDIR support. */
465 RTE_FDIR_MODE_SIGNATURE, /**< Enable FDIR signature filter mode. */
466 RTE_FDIR_MODE_PERFECT, /**< Enable FDIR perfect filter mode. */
470 * Memory space that can be configured to store Flow Director filters
471 * in the board memory.
473 enum rte_fdir_pballoc_type {
474 RTE_FDIR_PBALLOC_64K = 0, /**< 64k. */
475 RTE_FDIR_PBALLOC_128K, /**< 128k. */
476 RTE_FDIR_PBALLOC_256K, /**< 256k. */
480 * Select report mode of FDIR hash information in RX descriptors.
482 enum rte_fdir_status_mode {
483 RTE_FDIR_NO_REPORT_STATUS = 0, /**< Never report FDIR hash. */
484 RTE_FDIR_REPORT_STATUS, /**< Only report FDIR hash for matching pkts. */
485 RTE_FDIR_REPORT_STATUS_ALWAYS, /**< Always report FDIR hash. */
489 * A structure used to configure the Flow Director (FDIR) feature
490 * of an Ethernet port.
492 * If mode is RTE_FDIR_DISABLE, the pballoc value is ignored.
494 struct rte_fdir_conf {
495 enum rte_fdir_mode mode; /**< Flow Director mode. */
496 enum rte_fdir_pballoc_type pballoc; /**< Space for FDIR filters. */
497 enum rte_fdir_status_mode status; /**< How to report FDIR hash. */
498 /* Offset of flexbytes field in RX packets (in 16-bit word units). */
499 uint8_t flexbytes_offset;
500 /* RX queue of packets matching a "drop" filter in perfect mode. */
505 * Possible l4type of FDIR filters.
508 RTE_FDIR_L4TYPE_NONE = 0, /**< None. */
509 RTE_FDIR_L4TYPE_UDP, /**< UDP. */
510 RTE_FDIR_L4TYPE_TCP, /**< TCP. */
511 RTE_FDIR_L4TYPE_SCTP, /**< SCTP. */
515 * Select IPv4 or IPv6 FDIR filters.
518 RTE_FDIR_IPTYPE_IPV4 = 0, /**< IPv4. */
519 RTE_FDIR_IPTYPE_IPV6 , /**< IPv6. */
523 * A structure used to define a FDIR packet filter.
525 struct rte_fdir_filter {
526 uint16_t flex_bytes; /**< Flex bytes value to match. */
527 uint16_t vlan_id; /**< VLAN ID value to match, 0 otherwise. */
528 uint16_t port_src; /**< Source port to match, 0 otherwise. */
529 uint16_t port_dst; /**< Destination port to match, 0 otherwise. */
531 uint32_t ipv4_addr; /**< IPv4 source address to match. */
532 uint32_t ipv6_addr[4]; /**< IPv6 source address to match. */
533 } ip_src; /**< IPv4/IPv6 source address to match (union of above). */
535 uint32_t ipv4_addr; /**< IPv4 destination address to match. */
536 uint32_t ipv6_addr[4]; /**< IPv6 destination address to match */
537 } ip_dst; /**< IPv4/IPv6 destination address to match (union of above). */
538 enum rte_l4type l4type; /**< l4type to match: NONE/UDP/TCP/SCTP. */
539 enum rte_iptype iptype; /**< IP packet type to match: IPv4 or IPv6. */
543 * A structure used to configure FDIR masks that are used by the device
544 * to match the various fields of RX packet headers.
545 * @note The only_ip_flow field has the opposite meaning compared to other
548 struct rte_fdir_masks {
549 /** When set to 1, packet l4type is \b NOT relevant in filters, and
550 source and destination port masks must be set to zero. */
551 uint8_t only_ip_flow;
552 /** If set to 1, vlan_id is relevant in filters. */
554 /** If set to 1, vlan_prio is relevant in filters. */
556 /** If set to 1, flexbytes is relevant in filters. */
558 /** If set to 1, set the IPv6 masks. Otherwise set the IPv4 masks. */
559 uint8_t set_ipv6_mask;
560 /** When set to 1, comparison of destination IPv6 address with IP6AT
561 registers is meaningful. */
562 uint8_t comp_ipv6_dst;
563 /** Mask of Destination IPv4 Address. All bits set to 1 define the
564 relevant bits to use in the destination address of an IPv4 packet
565 when matching it against FDIR filters. */
566 uint32_t dst_ipv4_mask;
567 /** Mask of Source IPv4 Address. All bits set to 1 define
568 the relevant bits to use in the source address of an IPv4 packet
569 when matching it against FDIR filters. */
570 uint32_t src_ipv4_mask;
571 /** Mask of Source IPv6 Address. All bits set to 1 define the
572 relevant BYTES to use in the source address of an IPv6 packet
573 when matching it against FDIR filters. */
574 uint16_t dst_ipv6_mask;
575 /** Mask of Destination IPv6 Address. All bits set to 1 define the
576 relevant BYTES to use in the destination address of an IPv6 packet
577 when matching it against FDIR filters. */
578 uint16_t src_ipv6_mask;
579 /** Mask of Source Port. All bits set to 1 define the relevant
580 bits to use in the source port of an IP packets when matching it
581 against FDIR filters. */
582 uint16_t src_port_mask;
583 /** Mask of Destination Port. All bits set to 1 define the relevant
584 bits to use in the destination port of an IP packet when matching it
585 against FDIR filters. */
586 uint16_t dst_port_mask;
590 * A structure used to report the status of the flow director filters in use.
592 struct rte_eth_fdir {
593 /** Number of filters with collision indication. */
595 /** Number of free (non programmed) filters. */
597 /** The Lookup hash value of the added filter that updated the value
598 of the MAXLEN field */
600 /** Longest linked list of filters in the table. */
602 /** Number of added filters. */
604 /** Number of removed filters. */
606 /** Number of failed added filters (no more space in device). */
608 /** Number of failed removed filters. */
613 * A structure used to enable/disable specific device interrupts.
615 struct rte_intr_conf {
616 /** enable/disable lsc interrupt. 0 (default) - disable, 1 enable */
621 * A structure used to configure an Ethernet port.
622 * Depending upon the RX multi-queue mode, extra advanced
623 * configuration settings may be needed.
625 struct rte_eth_conf {
627 /**< ETH_LINK_SPEED_10[0|00|000], or 0 for autonegotation */
628 uint16_t link_duplex;
629 /**< ETH_LINK_[HALF_DUPLEX|FULL_DUPLEX], or 0 for autonegotation */
630 struct rte_eth_rxmode rxmode; /**< Port RX configuration. */
631 struct rte_eth_txmode txmode; /**< Port TX configuration. */
633 struct rte_eth_rss_conf rss_conf; /**< Port RSS configuration */
634 struct rte_eth_vmdq_dcb_conf vmdq_dcb_conf;
635 /**< Port vmdq+dcb configuration. */
636 struct rte_eth_dcb_rx_conf dcb_rx_conf;
637 /**< Port dcb RX configuration. */
638 } rx_adv_conf; /**< Port RX filtering configuration (union). */
640 struct rte_eth_vmdq_dcb_tx_conf vmdq_dcb_tx_conf;
641 /**< Port vmdq+dcb TX configuration. */
642 struct rte_eth_dcb_tx_conf dcb_tx_conf;
643 /**< Port dcb TX configuration. */
644 } tx_adv_conf; /**< Port TX DCB configuration (union). */
645 /** Currently,Priority Flow Control(PFC) are supported,if DCB with PFC
646 is needed,and the variable must be set ETH_DCB_PFC_SUPPORT. */
647 uint32_t dcb_capability_en;
648 struct rte_fdir_conf fdir_conf; /**< FDIR configuration. */
649 struct rte_intr_conf intr_conf; /**< Interrupt mode configuration. */
653 * A structure used to retrieve the contextual information of
654 * an Ethernet device, such as the controlling driver of the device,
655 * its PCI context, etc...
657 struct rte_eth_dev_info {
658 struct rte_pci_device *pci_dev; /**< Device PCI information. */
659 const char *driver_name; /**< Device Driver name. */
660 uint32_t min_rx_bufsize; /**< Minimum size of RX buffer. */
661 uint32_t max_rx_pktlen; /**< Maximum configurable length of RX pkt. */
662 uint16_t max_rx_queues; /**< Maximum number of RX queues. */
663 uint16_t max_tx_queues; /**< Maximum number of TX queues. */
664 uint32_t max_mac_addrs; /**< Maximum number of MAC addresses. */
669 struct rte_eth_dev_callback;
670 /** @internal Structure to keep track of registered callbacks */
671 TAILQ_HEAD(rte_eth_dev_cb_list, rte_eth_dev_callback);
674 * Definitions of all functions exported by an Ethernet driver through the
675 * the generic structure of type *eth_dev_ops* supplied in the *rte_eth_dev*
676 * structure associated with an Ethernet device.
679 typedef int (*eth_dev_configure_t)(struct rte_eth_dev *dev);
680 /**< @internal Ethernet device configuration. */
682 typedef int (*eth_dev_start_t)(struct rte_eth_dev *dev);
683 /**< @internal Function used to start a configured Ethernet device. */
685 typedef void (*eth_dev_stop_t)(struct rte_eth_dev *dev);
686 /**< @internal Function used to stop a configured Ethernet device. */
688 typedef void (*eth_dev_close_t)(struct rte_eth_dev *dev);
689 /**< @internal Function used to close a configured Ethernet device. */
691 typedef void (*eth_promiscuous_enable_t)(struct rte_eth_dev *dev);
692 /**< @internal Function used to enable the RX promiscuous mode of an Ethernet device. */
694 typedef void (*eth_promiscuous_disable_t)(struct rte_eth_dev *dev);
695 /**< @internal Function used to disable the RX promiscuous mode of an Ethernet device. */
697 typedef void (*eth_allmulticast_enable_t)(struct rte_eth_dev *dev);
698 /**< @internal Enable the receipt of all multicast packets by an Ethernet device. */
700 typedef void (*eth_allmulticast_disable_t)(struct rte_eth_dev *dev);
701 /**< @internal Disable the receipt of all multicast packets by an Ethernet device. */
703 typedef int (*eth_link_update_t)(struct rte_eth_dev *dev,
704 int wait_to_complete);
705 /**< @internal Get link speed, duplex mode and state (up/down) of an Ethernet device. */
707 typedef void (*eth_stats_get_t)(struct rte_eth_dev *dev,
708 struct rte_eth_stats *igb_stats);
709 /**< @internal Get global I/O statistics of an Ethernet device. */
711 typedef void (*eth_stats_reset_t)(struct rte_eth_dev *dev);
712 /**< @internal Reset global I/O statistics of an Ethernet device to 0. */
714 typedef int (*eth_queue_stats_mapping_set_t)(struct rte_eth_dev *dev,
718 /**< @internal Set a queue statistics mapping for a tx/rx queue of an Ethernet device. */
720 typedef void (*eth_dev_infos_get_t)(struct rte_eth_dev *dev,
721 struct rte_eth_dev_info *dev_info);
722 /**< @internal Get specific informations of an Ethernet device. */
724 typedef int (*eth_rx_queue_setup_t)(struct rte_eth_dev *dev,
725 uint16_t rx_queue_id,
727 unsigned int socket_id,
728 const struct rte_eth_rxconf *rx_conf,
729 struct rte_mempool *mb_pool);
730 /**< @internal Set up a receive queue of an Ethernet device. */
732 typedef int (*eth_tx_queue_setup_t)(struct rte_eth_dev *dev,
733 uint16_t tx_queue_id,
735 unsigned int socket_id,
736 const struct rte_eth_txconf *tx_conf);
737 /**< @internal Setup a transmit queue of an Ethernet device. */
739 typedef void (*eth_queue_release_t)(void *queue);
740 /**< @internal Release memory resources allocated by given RX/TX queue. */
742 typedef int (*vlan_filter_set_t)(struct rte_eth_dev *dev,
745 /**< @internal filtering of a VLAN Tag Identifier by an Ethernet device. */
747 typedef void (*vlan_tpid_set_t)(struct rte_eth_dev *dev,
749 /**< @internal set the outer VLAN-TPID by an Ethernet device. */
751 typedef void (*vlan_offload_set_t)(struct rte_eth_dev *dev, int mask);
752 /**< @internal set VLAN offload function by an Ethernet device. */
754 typedef void (*vlan_strip_queue_set_t)(struct rte_eth_dev *dev,
755 uint16_t rx_queue_id,
757 /**< @internal VLAN stripping enable/disable by an queue of Ethernet device. */
759 typedef uint16_t (*eth_rx_burst_t)(void *rxq,
760 struct rte_mbuf **rx_pkts,
762 /**< @internal Retrieve input packets from a receive queue of an Ethernet device. */
764 typedef uint16_t (*eth_tx_burst_t)(void *txq,
765 struct rte_mbuf **tx_pkts,
767 /**< @internal Send output packets on a transmit queue of an Ethernet device. */
769 typedef int (*fdir_add_signature_filter_t)(struct rte_eth_dev *dev,
770 struct rte_fdir_filter *fdir_ftr,
772 /**< @internal Setup a new signature filter rule on an Ethernet device */
774 typedef int (*fdir_update_signature_filter_t)(struct rte_eth_dev *dev,
775 struct rte_fdir_filter *fdir_ftr,
777 /**< @internal Update a signature filter rule on an Ethernet device */
779 typedef int (*fdir_remove_signature_filter_t)(struct rte_eth_dev *dev,
780 struct rte_fdir_filter *fdir_ftr);
781 /**< @internal Remove a signature filter rule on an Ethernet device */
783 typedef void (*fdir_infos_get_t)(struct rte_eth_dev *dev,
784 struct rte_eth_fdir *fdir);
785 /**< @internal Get information about fdir status */
787 typedef int (*fdir_add_perfect_filter_t)(struct rte_eth_dev *dev,
788 struct rte_fdir_filter *fdir_ftr,
789 uint16_t soft_id, uint8_t rx_queue,
791 /**< @internal Setup a new perfect filter rule on an Ethernet device */
793 typedef int (*fdir_update_perfect_filter_t)(struct rte_eth_dev *dev,
794 struct rte_fdir_filter *fdir_ftr,
795 uint16_t soft_id, uint8_t rx_queue,
797 /**< @internal Update a perfect filter rule on an Ethernet device */
799 typedef int (*fdir_remove_perfect_filter_t)(struct rte_eth_dev *dev,
800 struct rte_fdir_filter *fdir_ftr,
802 /**< @internal Remove a perfect filter rule on an Ethernet device */
804 typedef int (*fdir_set_masks_t)(struct rte_eth_dev *dev,
805 struct rte_fdir_masks *fdir_masks);
806 /**< @internal Setup flow director masks on an Ethernet device */
808 typedef int (*flow_ctrl_set_t)(struct rte_eth_dev *dev,
809 struct rte_eth_fc_conf *fc_conf);
810 /**< @internal Setup flow control parameter on an Ethernet device */
812 typedef int (*priority_flow_ctrl_set_t)(struct rte_eth_dev *dev,
813 struct rte_eth_pfc_conf *pfc_conf);
814 /**< @internal Setup priority flow control parameter on an Ethernet device */
816 typedef int (*eth_dev_led_on_t)(struct rte_eth_dev *dev);
817 /**< @internal Turn on SW controllable LED on an Ethernet device */
819 typedef int (*eth_dev_led_off_t)(struct rte_eth_dev *dev);
820 /**< @internal Turn off SW controllable LED on an Ethernet device */
822 typedef void (*eth_mac_addr_remove_t)(struct rte_eth_dev *dev, uint32_t index);
823 /**< @internal Remove MAC address from receive address register */
825 typedef void (*eth_mac_addr_add_t)(struct rte_eth_dev *dev,
826 struct ether_addr *mac_addr,
829 /**< @internal Set a MAC address into Receive Address Address Register */
832 * @internal A structure containing the functions exported by an Ethernet driver.
835 eth_dev_configure_t dev_configure; /**< Configure device. */
836 eth_dev_start_t dev_start; /**< Start device. */
837 eth_dev_stop_t dev_stop; /**< Stop device. */
838 eth_dev_close_t dev_close; /**< Close device. */
839 eth_promiscuous_enable_t promiscuous_enable; /**< Promiscuous ON. */
840 eth_promiscuous_disable_t promiscuous_disable;/**< Promiscuous OFF. */
841 eth_allmulticast_enable_t allmulticast_enable;/**< RX multicast ON. */
842 eth_allmulticast_disable_t allmulticast_disable;/**< RX multicast OF. */
843 eth_link_update_t link_update; /**< Get device link state. */
844 eth_stats_get_t stats_get; /**< Get device statistics. */
845 eth_stats_reset_t stats_reset; /**< Reset device statistics. */
846 eth_queue_stats_mapping_set_t queue_stats_mapping_set;
847 /**< Configure per queue stat counter mapping. */
848 eth_dev_infos_get_t dev_infos_get; /**< Get device info. */
849 vlan_filter_set_t vlan_filter_set; /**< Filter VLAN Setup. */
850 vlan_tpid_set_t vlan_tpid_set; /**< Outer VLAN TPID Setup. */
851 vlan_strip_queue_set_t vlan_strip_queue_set; /**< VLAN Stripping on queue. */
852 vlan_offload_set_t vlan_offload_set; /**< Set VLAN Offload. */
853 eth_rx_queue_setup_t rx_queue_setup;/**< Set up device RX queue.*/
854 eth_queue_release_t rx_queue_release;/**< Release RX queue.*/
855 eth_tx_queue_setup_t tx_queue_setup;/**< Set up device TX queue.*/
856 eth_queue_release_t tx_queue_release;/**< Release TX queue.*/
857 eth_dev_led_on_t dev_led_on; /**< Turn on LED. */
858 eth_dev_led_off_t dev_led_off; /**< Turn off LED. */
859 flow_ctrl_set_t flow_ctrl_set; /**< Setup flow control. */
860 priority_flow_ctrl_set_t priority_flow_ctrl_set; /**< Setup priority flow control.*/
861 eth_mac_addr_remove_t mac_addr_remove; /**< Remove MAC address */
862 eth_mac_addr_add_t mac_addr_add; /**< Add a MAC address */
864 /** Add a signature filter. */
865 fdir_add_signature_filter_t fdir_add_signature_filter;
866 /** Update a signature filter. */
867 fdir_update_signature_filter_t fdir_update_signature_filter;
868 /** Remove a signature filter. */
869 fdir_remove_signature_filter_t fdir_remove_signature_filter;
870 /** Get information about FDIR status. */
871 fdir_infos_get_t fdir_infos_get;
872 /** Add a perfect filter. */
873 fdir_add_perfect_filter_t fdir_add_perfect_filter;
874 /** Update a perfect filter. */
875 fdir_update_perfect_filter_t fdir_update_perfect_filter;
876 /** Remove a perfect filter. */
877 fdir_remove_perfect_filter_t fdir_remove_perfect_filter;
878 /** Setup masks for FDIR filtering. */
879 fdir_set_masks_t fdir_set_masks;
884 * The generic data structure associated with each ethernet device.
886 * Pointers to burst-oriented packet receive and transmit functions are
887 * located at the beginning of the structure, along with the pointer to
888 * where all the data elements for the particular device are stored in shared
889 * memory. This split allows the function pointer and driver data to be per-
890 * process, while the actual configuration data for the device is shared.
893 eth_rx_burst_t rx_pkt_burst; /**< Pointer to PMD receive function. */
894 eth_tx_burst_t tx_pkt_burst; /**< Pointer to PMD transmit function. */
895 struct rte_eth_dev_data *data; /**< Pointer to device data */
896 const struct eth_driver *driver;/**< Driver for this device */
897 struct eth_dev_ops *dev_ops; /**< Functions exported by PMD */
898 struct rte_pci_device *pci_dev; /**< PCI info. supplied by probing */
899 struct rte_eth_dev_cb_list callbacks; /**< User application callbacks */
904 * The data part, with no function pointers, associated with each ethernet device.
906 * This structure is safe to place in shared memory to be common among different
907 * processes in a multi-process configuration.
909 struct rte_eth_dev_data {
910 void **rx_queues; /**< Array of pointers to RX queues. */
911 void **tx_queues; /**< Array of pointers to TX queues. */
912 uint16_t nb_rx_queues; /**< Number of RX queues. */
913 uint16_t nb_tx_queues; /**< Number of TX queues. */
915 void *dev_private; /**< PMD-specific private data */
917 struct rte_eth_link dev_link;
918 /**< Link-level information & status */
920 struct rte_eth_conf dev_conf; /**< Configuration applied to device. */
921 uint16_t max_frame_size; /**< Default is ETHER_MAX_LEN (1518). */
923 uint64_t rx_mbuf_alloc_failed; /**< RX ring mbuf allocation failures. */
924 struct ether_addr* mac_addrs;/**< Device Ethernet Link address. */
925 uint8_t port_id; /**< Device [external] port identifier. */
926 uint8_t promiscuous : 1, /**< RX promiscuous mode ON(1) / OFF(0). */
927 scattered_rx : 1, /**< RX of scattered packets is ON(1) / OFF(0) */
928 all_multicast : 1, /**< RX all multicast mode ON(1) / OFF(0). */
929 dev_started : 1; /**< Device state: STARTED(1) / STOPPED(0). */
934 * The pool of *rte_eth_dev* structures. The size of the pool
935 * is configured at compile-time in the <rte_ethdev.c> file.
937 extern struct rte_eth_dev rte_eth_devices[];
940 * Get the total number of Ethernet devices that have been successfully
941 * initialized by the [matching] Ethernet driver during the PCI probing phase.
942 * All devices whose port identifier is in the range
943 * [0, rte_eth_dev_count() - 1] can be operated on by network applications.
946 * - The total number of usable Ethernet devices.
948 extern uint8_t rte_eth_dev_count(void);
953 * Initialization function of an Ethernet driver invoked for each matching
954 * Ethernet PCI device detected during the PCI probing phase.
957 * The pointer to the [matching] Ethernet driver structure supplied by
958 * the PMD when it registered itself.
960 * The *eth_dev* pointer is the address of the *rte_eth_dev* structure
961 * associated with the matching device and which have been [automatically]
962 * allocated in the *rte_eth_devices* array.
963 * The *eth_dev* structure is supplied to the driver initialization function
964 * with the following fields already initialized:
966 * - *pci_dev*: Holds the pointers to the *rte_pci_device* structure which
967 * contains the generic PCI information of the matching device.
969 * - *dev_private*: Holds a pointer to the device private data structure.
971 * - *max_frame_size*: Contains the default Ethernet maximum frame length
974 * - *port_id*: Contains the port index of the device (actually the index
975 * of the *eth_dev* structure in the *rte_eth_devices* array).
978 * - 0: Success, the device is properly initialized by the driver.
979 * In particular, the driver MUST have set up the *dev_ops* pointer
980 * of the *eth_dev* structure.
981 * - <0: Error code of the device initialization failure.
983 typedef int (*eth_dev_init_t)(struct eth_driver *eth_drv,
984 struct rte_eth_dev *eth_dev);
988 * The structure associated with a PMD Ethernet driver.
990 * Each Ethernet driver acts as a PCI driver and is represented by a generic
991 * *eth_driver* structure that holds:
993 * - An *rte_pci_driver* structure (which must be the first field).
995 * - The *eth_dev_init* function invoked for each matching PCI device.
997 * - The size of the private data to allocate for each matching device.
1000 struct rte_pci_driver pci_drv; /**< The PMD is also a PCI driver. */
1001 eth_dev_init_t eth_dev_init; /**< Device init function. */
1002 unsigned int dev_private_size; /**< Size of device private data. */
1007 * A function invoked by the initialization function of an Ethernet driver
1008 * to simultaneously register itself as a PCI driver and as an Ethernet
1009 * Poll Mode Driver (PMD).
1012 * The pointer to the *eth_driver* structure associated with
1013 * the Ethernet driver.
1015 extern void rte_eth_driver_register(struct eth_driver *eth_drv);
1018 * The initialization function of the driver for
1019 * Intel(r) IGB Gigabit Ethernet Controller devices.
1020 * This function is invoked once at EAL start time.
1024 extern int rte_igb_pmd_init(void);
1027 * The initialization function of the driver for
1028 * Intel(r) EM Gigabit Ethernet Controller devices.
1029 * This function is invoked once at EAL start time.
1033 extern int rte_em_pmd_init(void);
1036 * The initialization function of the driver for 1Gbps Intel IGB_VF
1038 * Invoked once at EAL start time.
1042 extern int rte_igbvf_pmd_init(void);
1045 * The initialization function of the driver for 10Gbps Intel IXGBE
1047 * Invoked once at EAL start time.
1051 extern int rte_ixgbe_pmd_init(void);
1054 * The initialization function of the driver for 10Gbps Intel IXGBE_VF
1056 * Invoked once at EAL start time.
1060 extern int rte_ixgbevf_pmd_init(void);
1063 * The initialization function of *all* supported and enabled drivers.
1064 * Right now, the following PMDs are supported:
1070 * This function is invoked once at EAL start time.
1073 * Error code of the device initialization failure,
1074 * -ENODEV if there are no drivers available
1075 * (e.g. if all driver config options are = n).
1078 int rte_pmd_init_all(void)
1082 #ifdef RTE_LIBRTE_IGB_PMD
1083 if ((ret = rte_igb_pmd_init()) != 0) {
1084 RTE_LOG(ERR, PMD, "Cannot init igb PMD\n");
1087 if ((ret = rte_igbvf_pmd_init()) != 0) {
1088 RTE_LOG(ERR, PMD, "Cannot init igbvf PMD\n");
1091 #endif /* RTE_LIBRTE_IGB_PMD */
1093 #ifdef RTE_LIBRTE_EM_PMD
1094 if ((ret = rte_em_pmd_init()) != 0) {
1095 RTE_LOG(ERR, PMD, "Cannot init em PMD\n");
1098 #endif /* RTE_LIBRTE_EM_PMD */
1100 #ifdef RTE_LIBRTE_IXGBE_PMD
1101 if ((ret = rte_ixgbe_pmd_init()) != 0) {
1102 RTE_LOG(ERR, PMD, "Cannot init ixgbe PMD\n");
1105 if ((ret = rte_ixgbevf_pmd_init()) != 0) {
1106 RTE_LOG(ERR, PMD, "Cannot init ixgbevf PMD\n");
1109 #endif /* RTE_LIBRTE_IXGBE_PMD */
1112 RTE_LOG(ERR, PMD, "No PMD(s) are configured\n");
1117 * Configure an Ethernet device.
1118 * This function must be invoked first before any other function in the
1119 * Ethernet API. This function can also be re-invoked when a device is in the
1123 * The port identifier of the Ethernet device to configure.
1124 * @param nb_rx_queue
1125 * The number of receive queues to set up for the Ethernet device.
1126 * @param nb_tx_queue
1127 * The number of transmit queues to set up for the Ethernet device.
1129 * The pointer to the configuration data to be used for the Ethernet device.
1130 * The *rte_eth_conf* structure includes:
1131 * - the hardware offload features to activate, with dedicated fields for
1132 * each statically configurable offload hardware feature provided by
1133 * Ethernet devices, such as IP checksum or VLAN tag stripping for
1135 * - the Receive Side Scaling (RSS) configuration when using multiple RX
1138 * Embedding all configuration information in a single data structure
1139 * is the more flexible method that allows the addition of new features
1140 * without changing the syntax of the API.
1142 * - 0: Success, device configured.
1143 * - <0: Error code returned by the driver configuration function.
1145 extern int rte_eth_dev_configure(uint8_t port_id,
1146 uint16_t nb_rx_queue,
1147 uint16_t nb_tx_queue,
1148 const struct rte_eth_conf *eth_conf);
1151 * Allocate and set up a receive queue for an Ethernet device.
1153 * The function allocates a contiguous block of memory for *nb_rx_desc*
1154 * receive descriptors from a memory zone associated with *socket_id*
1155 * and initializes each receive descriptor with a network buffer allocated
1156 * from the memory pool *mb_pool*.
1159 * The port identifier of the Ethernet device.
1160 * @param rx_queue_id
1161 * The index of the receive queue to set up.
1162 * The value must be in the range [0, nb_rx_queue - 1] previously supplied
1163 * to rte_eth_dev_configure().
1165 * The number of receive descriptors to allocate for the receive ring.
1167 * The *socket_id* argument is the socket identifier in case of NUMA.
1168 * The value can be *SOCKET_ID_ANY* if there is no NUMA constraint for
1169 * the DMA memory allocated for the receive descriptors of the ring.
1171 * The pointer to the configuration data to be used for the receive queue.
1172 * The *rx_conf* structure contains an *rx_thresh* structure with the values
1173 * of the Prefetch, Host, and Write-Back threshold registers of the receive
1176 * The pointer to the memory pool from which to allocate *rte_mbuf* network
1177 * memory buffers to populate each descriptor of the receive ring.
1179 * - 0: Success, receive queue correctly set up.
1180 * - -EINVAL: The size of network buffers which can be allocated from the
1181 * memory pool does not fit the various buffer sizes allowed by the
1182 * device controller.
1183 * - -ENOMEM: Unable to allocate the receive ring descriptors or to
1184 * allocate network memory buffers from the memory pool when
1185 * initializing receive descriptors.
1187 extern int rte_eth_rx_queue_setup(uint8_t port_id, uint16_t rx_queue_id,
1188 uint16_t nb_rx_desc, unsigned int socket_id,
1189 const struct rte_eth_rxconf *rx_conf,
1190 struct rte_mempool *mb_pool);
1193 * Allocate and set up a transmit queue for an Ethernet device.
1196 * The port identifier of the Ethernet device.
1197 * @param tx_queue_id
1198 * The index of the transmit queue to set up.
1199 * The value must be in the range [0, nb_tx_queue - 1] previously supplied
1200 * to rte_eth_dev_configure().
1202 * The number of transmit descriptors to allocate for the transmit ring.
1204 * The *socket_id* argument is the socket identifier in case of NUMA.
1205 * Its value can be *SOCKET_ID_ANY* if there is no NUMA constraint for
1206 * the DMA memory allocated for the transmit descriptors of the ring.
1208 * The pointer to the configuration data to be used for the transmit queue.
1209 * The *tx_conf* structure contains the following data:
1210 * - The *tx_thresh* structure with the values of the Prefetch, Host, and
1211 * Write-Back threshold registers of the transmit ring.
1212 * When setting Write-Back threshold to the value greater then zero,
1213 * *tx_rs_thresh* value should be explicitly set to one.
1214 * - The *tx_free_thresh* value indicates the [minimum] number of network
1215 * buffers that must be pending in the transmit ring to trigger their
1216 * [implicit] freeing by the driver transmit function.
1217 * - The *tx_rs_thresh* value indicates the [minimum] number of transmit
1218 * descriptors that must be pending in the transmit ring before setting the
1219 * RS bit on a descriptor by the driver transmit function.
1220 * The *tx_rs_thresh* value should be less or equal then
1221 * *tx_free_thresh* value, and both of them should be less then
1223 * - The *txq_flags* member contains flags to pass to the TX queue setup
1224 * function to configure the behavior of the TX queue. This should be set
1225 * to 0 if no special configuration is required.
1227 * Note that setting *tx_free_thresh* or *tx_rs_thresh* value to 0 forces
1228 * the transmit function to use default values.
1230 * - 0: Success, the transmit queue is correctly set up.
1231 * - -ENOMEM: Unable to allocate the transmit ring descriptors.
1233 extern int rte_eth_tx_queue_setup(uint8_t port_id, uint16_t tx_queue_id,
1234 uint16_t nb_tx_desc, unsigned int socket_id,
1235 const struct rte_eth_txconf *tx_conf);
1238 * Start an Ethernet device.
1240 * The device start step is the last one and consists of setting the configured
1241 * offload features and in starting the transmit and the receive units of the
1243 * On success, all basic functions exported by the Ethernet API (link status,
1244 * receive/transmit, and so on) can be invoked.
1247 * The port identifier of the Ethernet device.
1249 * - 0: Success, Ethernet device started.
1250 * - <0: Error code of the driver device start function.
1252 extern int rte_eth_dev_start(uint8_t port_id);
1255 * Stop an Ethernet device. The device can be restarted with a call to
1256 * rte_eth_dev_start()
1259 * The port identifier of the Ethernet device.
1261 extern void rte_eth_dev_stop(uint8_t port_id);
1264 * Close an Ethernet device. The device cannot be restarted!
1267 * The port identifier of the Ethernet device.
1269 extern void rte_eth_dev_close(uint8_t port_id);
1272 * Enable receipt in promiscuous mode for an Ethernet device.
1275 * The port identifier of the Ethernet device.
1277 extern void rte_eth_promiscuous_enable(uint8_t port_id);
1280 * Disable receipt in promiscuous mode for an Ethernet device.
1283 * The port identifier of the Ethernet device.
1285 extern void rte_eth_promiscuous_disable(uint8_t port_id);
1288 * Return the value of promiscuous mode for an Ethernet device.
1291 * The port identifier of the Ethernet device.
1293 * - (1) if promiscuous is enabled
1294 * - (0) if promiscuous is disabled.
1297 extern int rte_eth_promiscuous_get(uint8_t port_id);
1300 * Enable the receipt of any multicast frame by an Ethernet device.
1303 * The port identifier of the Ethernet device.
1305 extern void rte_eth_allmulticast_enable(uint8_t port_id);
1308 * Disable the receipt of all multicast frames by an Ethernet device.
1311 * The port identifier of the Ethernet device.
1313 extern void rte_eth_allmulticast_disable(uint8_t port_id);
1316 * Return the value of allmulticast mode for an Ethernet device.
1319 * The port identifier of the Ethernet device.
1321 * - (1) if allmulticast is enabled
1322 * - (0) if allmulticast is disabled.
1325 extern int rte_eth_allmulticast_get(uint8_t port_id);
1328 * Retrieve the status (ON/OFF), the speed (in Mbps) and the mode (HALF-DUPLEX
1329 * or FULL-DUPLEX) of the physical link of an Ethernet device. It might need
1330 * to wait up to 9 seconds in it.
1333 * The port identifier of the Ethernet device.
1335 * A pointer to an *rte_eth_link* structure to be filled with
1336 * the status, the speed and the mode of the Ethernet device link.
1338 extern void rte_eth_link_get(uint8_t port_id, struct rte_eth_link *link);
1341 * Retrieve the status (ON/OFF), the speed (in Mbps) and the mode (HALF-DUPLEX
1342 * or FULL-DUPLEX) of the physical link of an Ethernet device. It is a no-wait
1343 * version of rte_eth_link_get().
1346 * The port identifier of the Ethernet device.
1348 * A pointer to an *rte_eth_link* structure to be filled with
1349 * the status, the speed and the mode of the Ethernet device link.
1351 extern void rte_eth_link_get_nowait(uint8_t port_id,
1352 struct rte_eth_link *link);
1355 * Retrieve the general I/O statistics of an Ethernet device.
1358 * The port identifier of the Ethernet device.
1360 * A pointer to a structure of type *rte_eth_stats* to be filled with
1361 * the values of device counters for the following set of statistics:
1362 * - *ipackets* with the total of successfully received packets.
1363 * - *opackets* with the total of successfully transmitted packets.
1364 * - *ibytes* with the total of successfully received bytes.
1365 * - *obytes* with the total of successfully transmitted bytes.
1366 * - *ierrors* with the total of erroneous received packets.
1367 * - *oerrors* with the total of failed transmitted packets.
1369 extern void rte_eth_stats_get(uint8_t port_id, struct rte_eth_stats *stats);
1372 * Reset the general I/O statistics of an Ethernet device.
1375 * The port identifier of the Ethernet device.
1377 extern void rte_eth_stats_reset(uint8_t port_id);
1380 * Set a mapping for the specified transmit queue to the specified per-queue
1381 * statistics counter.
1384 * The port identifier of the Ethernet device.
1385 * @param tx_queue_id
1386 * The index of the transmit queue for which a queue stats mapping is required.
1387 * The value must be in the range [0, nb_tx_queue - 1] previously supplied
1388 * to rte_eth_dev_configure().
1390 * The per-queue packet statistics functionality number that the transmit
1391 * queue is to be assigned.
1392 * The value must be in the range [0, RTE_MAX_ETHPORT_QUEUE_STATS_MAPS - 1].
1394 * Zero if successful. Non-zero otherwise.
1396 extern int rte_eth_dev_set_tx_queue_stats_mapping(uint8_t port_id,
1397 uint16_t tx_queue_id,
1401 * Set a mapping for the specified receive queue to the specified per-queue
1402 * statistics counter.
1405 * The port identifier of the Ethernet device.
1406 * @param rx_queue_id
1407 * The index of the receive queue for which a queue stats mapping is required.
1408 * The value must be in the range [0, nb_rx_queue - 1] previously supplied
1409 * to rte_eth_dev_configure().
1411 * The per-queue packet statistics functionality number that the receive
1412 * queue is to be assigned.
1413 * The value must be in the range [0, RTE_MAX_ETHPORT_QUEUE_STATS_MAPS - 1].
1415 * Zero if successful. Non-zero otherwise.
1417 extern int rte_eth_dev_set_rx_queue_stats_mapping(uint8_t port_id,
1418 uint16_t rx_queue_id,
1422 * Retrieve the Ethernet address of an Ethernet device.
1425 * The port identifier of the Ethernet device.
1427 * A pointer to a structure of type *ether_addr* to be filled with
1428 * the Ethernet address of the Ethernet device.
1430 extern void rte_eth_macaddr_get(uint8_t port_id, struct ether_addr *mac_addr);
1433 * Retrieve the contextual information of an Ethernet device.
1436 * The port identifier of the Ethernet device.
1438 * A pointer to a structure of type *rte_eth_dev_info* to be filled with
1439 * the contextual information of the Ethernet device.
1441 extern void rte_eth_dev_info_get(uint8_t port_id,
1442 struct rte_eth_dev_info *dev_info);
1445 * Enable/Disable hardware filtering by an Ethernet device of received
1446 * VLAN packets tagged with a given VLAN Tag Identifier.
1449 * The port identifier of the Ethernet device.
1451 * The VLAN Tag Identifier whose filtering must be enabled or disabled.
1453 * If > 0, enable VLAN filtering of VLAN packets tagged with *vlan_id*.
1454 * Otherwise, disable VLAN filtering of VLAN packets tagged with *vlan_id*.
1456 * - (0) if successful.
1457 * - (-ENOSUP) if hardware-assisted VLAN filtering not configured.
1458 * - (-ENODEV) if *port_id* invalid.
1459 * - (-ENOSYS) if VLAN filtering on *port_id* disabled.
1460 * - (-EINVAL) if *vlan_id* > 4095.
1462 extern int rte_eth_dev_vlan_filter(uint8_t port_id, uint16_t vlan_id , int on);
1465 * Enable/Disable hardware VLAN Strip by a rx queue of an Ethernet device.
1466 * 82599/X540 can support VLAN stripping at the rx queue level
1469 * The port identifier of the Ethernet device.
1470 * @param rx_queue_id
1471 * The index of the receive queue for which a queue stats mapping is required.
1472 * The value must be in the range [0, nb_rx_queue - 1] previously supplied
1473 * to rte_eth_dev_configure().
1475 * If 1, Enable VLAN Stripping of the receive queue of the Ethernet port.
1476 * If 0, Disable VLAN Stripping of the receive queue of the Ethernet port.
1478 * - (0) if successful.
1479 * - (-ENOSUP) if hardware-assisted VLAN stripping not configured.
1480 * - (-ENODEV) if *port_id* invalid.
1481 * - (-EINVAL) if *rx_queue_id* invalid.
1483 extern int rte_eth_dev_set_vlan_strip_on_queue(uint8_t port_id,
1484 uint16_t rx_queue_id, int on);
1487 * Set the Outer VLAN Ether Type by an Ethernet device, it can be inserted to
1488 * the VLAN Header. This is a register setup available on some Intel NIC, not
1489 * but all, please check the data sheet for availability.
1492 * The port identifier of the Ethernet device.
1494 * The Tag Protocol ID
1496 * - (0) if successful.
1497 * - (-ENOSUP) if hardware-assisted VLAN TPID setup is not supported.
1498 * - (-ENODEV) if *port_id* invalid.
1500 extern int rte_eth_dev_set_vlan_ether_type(uint8_t port_id, uint16_t tag_type);
1503 * Set VLAN offload configuration on an Ethernet device
1504 * Enable/Disable Extended VLAN by an Ethernet device, This is a register setup
1505 * available on some Intel NIC, not but all, please check the data sheet for
1507 * Enable/Disable VLAN Strip can be done on rx queue for certain NIC, but here
1508 * the configuration is applied on the port level.
1511 * The port identifier of the Ethernet device.
1512 * @param offload_mask
1513 * The VLAN Offload bit mask can be mixed use with "OR"
1514 * ETH_VLAN_STRIP_OFFLOAD
1515 * ETH_VLAN_FILTER_OFFLOAD
1516 * ETH_VLAN_EXTEND_OFFLOAD
1518 * - (0) if successful.
1519 * - (-ENOSUP) if hardware-assisted VLAN filtering not configured.
1520 * - (-ENODEV) if *port_id* invalid.
1522 extern int rte_eth_dev_set_vlan_offload(uint8_t port_id, int offload_mask);
1525 * Read VLAN Offload configuration from an Ethernet device
1528 * The port identifier of the Ethernet device.
1530 * - (>0) if successful. Bit mask to indicate
1531 * ETH_VLAN_STRIP_OFFLOAD
1532 * ETH_VLAN_FILTER_OFFLOAD
1533 * ETH_VLAN_EXTEND_OFFLOAD
1534 * - (-ENODEV) if *port_id* invalid.
1536 extern int rte_eth_dev_get_vlan_offload(uint8_t port_id);
1540 * Retrieve a burst of input packets from a receive queue of an Ethernet
1541 * device. The retrieved packets are stored in *rte_mbuf* structures whose
1542 * pointers are supplied in the *rx_pkts* array.
1544 * The rte_eth_rx_burst() function loops, parsing the RX ring of the
1545 * receive queue, up to *nb_pkts* packets, and for each completed RX
1546 * descriptor in the ring, it performs the following operations:
1548 * - Initialize the *rte_mbuf* data structure associated with the
1549 * RX descriptor according to the information provided by the NIC into
1550 * that RX descriptor.
1552 * - Store the *rte_mbuf* data structure into the next entry of the
1555 * - Replenish the RX descriptor with a new *rte_mbuf* buffer
1556 * allocated from the memory pool associated with the receive queue at
1557 * initialization time.
1559 * When retrieving an input packet that was scattered by the controller
1560 * into multiple receive descriptors, the rte_eth_rx_burst() function
1561 * appends the associated *rte_mbuf* buffers to the first buffer of the
1564 * The rte_eth_rx_burst() function returns the number of packets
1565 * actually retrieved, which is the number of *rte_mbuf* data structures
1566 * effectively supplied into the *rx_pkts* array.
1567 * A return value equal to *nb_pkts* indicates that the RX queue contained
1568 * at least *rx_pkts* packets, and this is likely to signify that other
1569 * received packets remain in the input queue. Applications implementing
1570 * a "retrieve as much received packets as possible" policy can check this
1571 * specific case and keep invoking the rte_eth_rx_burst() function until
1572 * a value less than *nb_pkts* is returned.
1574 * This receive method has the following advantages:
1576 * - It allows a run-to-completion network stack engine to retrieve and
1577 * to immediately process received packets in a fast burst-oriented
1578 * approach, avoiding the overhead of unnecessary intermediate packet
1579 * queue/dequeue operations.
1581 * - Conversely, it also allows an asynchronous-oriented processing
1582 * method to retrieve bursts of received packets and to immediately
1583 * queue them for further parallel processing by another logical core,
1584 * for instance. However, instead of having received packets being
1585 * individually queued by the driver, this approach allows the invoker
1586 * of the rte_eth_rx_burst() function to queue a burst of retrieved
1587 * packets at a time and therefore dramatically reduce the cost of
1588 * enqueue/dequeue operations per packet.
1590 * - It allows the rte_eth_rx_burst() function of the driver to take
1591 * advantage of burst-oriented hardware features (CPU cache,
1592 * prefetch instructions, and so on) to minimize the number of CPU
1593 * cycles per packet.
1595 * To summarize, the proposed receive API enables many
1596 * burst-oriented optimizations in both synchronous and asynchronous
1597 * packet processing environments with no overhead in both cases.
1599 * The rte_eth_rx_burst() function does not provide any error
1600 * notification to avoid the corresponding overhead. As a hint, the
1601 * upper-level application might check the status of the device link once
1602 * being systematically returned a 0 value for a given number of tries.
1605 * The port identifier of the Ethernet device.
1607 * The index of the receive queue from which to retrieve input packets.
1608 * The value must be in the range [0, nb_rx_queue - 1] previously supplied
1609 * to rte_eth_dev_configure().
1611 * The address of an array of pointers to *rte_mbuf* structures that
1612 * must be large enough to store *nb_pkts* pointers in it.
1614 * The maximum number of packets to retrieve.
1616 * The number of packets actually retrieved, which is the number
1617 * of pointers to *rte_mbuf* structures effectively supplied to the
1620 #ifdef RTE_LIBRTE_ETHDEV_DEBUG
1621 extern uint16_t rte_eth_rx_burst(uint8_t port_id, uint16_t queue_id,
1622 struct rte_mbuf **rx_pkts, uint16_t nb_pkts);
1624 static inline uint16_t
1625 rte_eth_rx_burst(uint8_t port_id, uint16_t queue_id,
1626 struct rte_mbuf **rx_pkts, uint16_t nb_pkts)
1628 struct rte_eth_dev *dev;
1630 dev = &rte_eth_devices[port_id];
1631 return (*dev->rx_pkt_burst)(dev->data->rx_queues[queue_id], rx_pkts, nb_pkts);
1636 * Send a burst of output packets on a transmit queue of an Ethernet device.
1638 * The rte_eth_tx_burst() function is invoked to transmit output packets
1639 * on the output queue *queue_id* of the Ethernet device designated by its
1641 * The *nb_pkts* parameter is the number of packets to send which are
1642 * supplied in the *tx_pkts* array of *rte_mbuf* structures.
1643 * The rte_eth_tx_burst() function loops, sending *nb_pkts* packets,
1644 * up to the number of transmit descriptors available in the TX ring of the
1646 * For each packet to send, the rte_eth_tx_burst() function performs
1647 * the following operations:
1649 * - Pick up the next available descriptor in the transmit ring.
1651 * - Free the network buffer previously sent with that descriptor, if any.
1653 * - Initialize the transmit descriptor with the information provided
1654 * in the *rte_mbuf data structure.
1656 * In the case of a segmented packet composed of a list of *rte_mbuf* buffers,
1657 * the rte_eth_tx_burst() function uses several transmit descriptors
1660 * The rte_eth_tx_burst() function returns the number of packets it
1661 * actually sent. A return value equal to *nb_pkts* means that all packets
1662 * have been sent, and this is likely to signify that other output packets
1663 * could be immediately transmitted again. Applications that implement a
1664 * "send as many packets to transmit as possible" policy can check this
1665 * specific case and keep invoking the rte_eth_tx_burst() function until
1666 * a value less than *nb_pkts* is returned.
1668 * It is the responsibility of the rte_eth_tx_burst() function to
1669 * transparently free the memory buffers of packets previously sent.
1670 * This feature is driven by the *tx_free_thresh* value supplied to the
1671 * rte_eth_dev_configure() function at device configuration time.
1672 * When the number of previously sent packets reached the "minimum transmit
1673 * packets to free" threshold, the rte_eth_tx_burst() function must
1674 * [attempt to] free the *rte_mbuf* buffers of those packets whose
1675 * transmission was effectively completed.
1678 * The port identifier of the Ethernet device.
1680 * The index of the transmit queue through which output packets must be
1682 * The value must be in the range [0, nb_tx_queue - 1] previously supplied
1683 * to rte_eth_dev_configure().
1685 * The address of an array of *nb_pkts* pointers to *rte_mbuf* structures
1686 * which contain the output packets.
1688 * The maximum number of packets to transmit.
1690 * The number of output packets actually stored in transmit descriptors of
1691 * the transmit ring. The return value can be less than the value of the
1692 * *tx_pkts* parameter when the transmit ring is full or has been filled up.
1694 #ifdef RTE_LIBRTE_ETHDEV_DEBUG
1695 extern uint16_t rte_eth_tx_burst(uint8_t port_id, uint16_t queue_id,
1696 struct rte_mbuf **tx_pkts, uint16_t nb_pkts);
1698 static inline uint16_t
1699 rte_eth_tx_burst(uint8_t port_id, uint16_t queue_id,
1700 struct rte_mbuf **tx_pkts, uint16_t nb_pkts)
1702 struct rte_eth_dev *dev;
1704 dev = &rte_eth_devices[port_id];
1705 return (*dev->tx_pkt_burst)(dev->data->tx_queues[queue_id], tx_pkts, nb_pkts);
1710 * Setup a new signature filter rule on an Ethernet device
1713 * The port identifier of the Ethernet device.
1714 * @param fdir_filter
1715 * The pointer to the fdir filter structure describing the signature filter
1717 * The *rte_fdir_filter* structure includes the values of the different fields
1718 * to match: source and destination IP addresses, vlan id, flexbytes, source
1719 * and destination ports, and so on.
1721 * The index of the RX queue where to store RX packets matching the added
1722 * signature filter defined in fdir_filter.
1724 * - (0) if successful.
1725 * - (-ENOTSUP) if hardware doesn't support flow director mode.
1726 * - (-ENODEV) if *port_id* invalid.
1727 * - (-ENOSYS) if the FDIR mode is not configured in signature mode
1729 * - (-EINVAL) if the fdir_filter information is not correct.
1731 int rte_eth_dev_fdir_add_signature_filter(uint8_t port_id,
1732 struct rte_fdir_filter *fdir_filter,
1736 * Update a signature filter rule on an Ethernet device.
1737 * If the rule doesn't exits, it is created.
1740 * The port identifier of the Ethernet device.
1742 * The pointer to the structure describing the signature filter rule.
1743 * The *rte_fdir_filter* structure includes the values of the different fields
1744 * to match: source and destination IP addresses, vlan id, flexbytes, source
1745 * and destination ports, and so on.
1747 * The index of the RX queue where to store RX packets matching the added
1748 * signature filter defined in fdir_ftr.
1750 * - (0) if successful.
1751 * - (-ENOTSUP) if hardware doesn't support flow director mode.
1752 * - (-ENODEV) if *port_id* invalid.
1753 * - (-ENOSYS) if the flow director mode is not configured in signature mode
1755 * - (-EINVAL) if the fdir_filter information is not correct.
1757 int rte_eth_dev_fdir_update_signature_filter(uint8_t port_id,
1758 struct rte_fdir_filter *fdir_ftr,
1762 * Remove a signature filter rule on an Ethernet device.
1765 * The port identifier of the Ethernet device.
1767 * The pointer to the structure describing the signature filter rule.
1768 * The *rte_fdir_filter* structure includes the values of the different fields
1769 * to match: source and destination IP addresses, vlan id, flexbytes, source
1770 * and destination ports, and so on.
1772 * - (0) if successful.
1773 * - (-ENOTSUP) if hardware doesn't support flow director mode.
1774 * - (-ENODEV) if *port_id* invalid.
1775 * - (-ENOSYS) if the flow director mode is not configured in signature mode
1777 * - (-EINVAL) if the fdir_filter information is not correct.
1779 int rte_eth_dev_fdir_remove_signature_filter(uint8_t port_id,
1780 struct rte_fdir_filter *fdir_ftr);
1783 * Retrieve the flow director information of an Ethernet device.
1786 * The port identifier of the Ethernet device.
1788 * A pointer to a structure of type *rte_eth_dev_fdir* to be filled with
1789 * the flow director information of the Ethernet device.
1791 * - (0) if successful.
1792 * - (-ENOTSUP) if hardware doesn't support flow director mode.
1793 * - (-ENODEV) if *port_id* invalid.
1794 * - (-ENOSYS) if the flow director mode is not configured on *port_id*.
1796 int rte_eth_dev_fdir_get_infos(uint8_t port_id, struct rte_eth_fdir *fdir);
1799 * Add a new perfect filter rule on an Ethernet device.
1802 * The port identifier of the Ethernet device.
1803 * @param fdir_filter
1804 * The pointer to the structure describing the perfect filter rule.
1805 * The *rte_fdir_filter* structure includes the values of the different fields
1806 * to match: source and destination IP addresses, vlan id, flexbytes, source
1807 * and destination ports, and so on.
1808 * IPv6 are not supported.
1810 * The 16-bit value supplied in the field hash.fdir.id of mbuf for RX
1811 * packets matching the perfect filter.
1813 * The index of the RX queue where to store RX packets matching the added
1814 * perfect filter defined in fdir_filter.
1816 * If drop is set to 1, matching RX packets are stored into the RX drop
1817 * queue defined in the rte_fdir_conf.
1819 * - (0) if successful.
1820 * - (-ENOTSUP) if hardware doesn't support flow director mode.
1821 * - (-ENODEV) if *port_id* invalid.
1822 * - (-ENOSYS) if the flow director mode is not configured in perfect mode
1824 * - (-EINVAL) if the fdir_filter information is not correct.
1826 int rte_eth_dev_fdir_add_perfect_filter(uint8_t port_id,
1827 struct rte_fdir_filter *fdir_filter,
1828 uint16_t soft_id, uint8_t rx_queue,
1832 * Update a perfect filter rule on an Ethernet device.
1833 * If the rule doesn't exits, it is created.
1836 * The port identifier of the Ethernet device.
1837 * @param fdir_filter
1838 * The pointer to the structure describing the perfect filter rule.
1839 * The *rte_fdir_filter* structure includes the values of the different fields
1840 * to match: source and destination IP addresses, vlan id, flexbytes, source
1841 * and destination ports, and so on.
1842 * IPv6 are not supported.
1844 * The 16-bit value supplied in the field hash.fdir.id of mbuf for RX
1845 * packets matching the perfect filter.
1847 * The index of the RX queue where to store RX packets matching the added
1848 * perfect filter defined in fdir_filter.
1850 * If drop is set to 1, matching RX packets are stored into the RX drop
1851 * queue defined in the rte_fdir_conf.
1853 * - (0) if successful.
1854 * - (-ENOTSUP) if hardware doesn't support flow director mode.
1855 * - (-ENODEV) if *port_id* invalid.
1856 * - (-ENOSYS) if the flow director mode is not configured in perfect mode
1858 * - (-EINVAL) if the fdir_filter information is not correct.
1860 int rte_eth_dev_fdir_update_perfect_filter(uint8_t port_id,
1861 struct rte_fdir_filter *fdir_filter,
1862 uint16_t soft_id, uint8_t rx_queue,
1866 * Remove a perfect filter rule on an Ethernet device.
1869 * The port identifier of the Ethernet device.
1870 * @param fdir_filter
1871 * The pointer to the structure describing the perfect filter rule.
1872 * The *rte_fdir_filter* structure includes the values of the different fields
1873 * to match: source and destination IP addresses, vlan id, flexbytes, source
1874 * and destination ports, and so on.
1875 * IPv6 are not supported.
1877 * The soft_id value provided when adding/updating the removed filter.
1879 * - (0) if successful.
1880 * - (-ENOTSUP) if hardware doesn't support flow director mode.
1881 * - (-ENODEV) if *port_id* invalid.
1882 * - (-ENOSYS) if the flow director mode is not configured in perfect mode
1884 * - (-EINVAL) if the fdir_filter information is not correct.
1886 int rte_eth_dev_fdir_remove_perfect_filter(uint8_t port_id,
1887 struct rte_fdir_filter *fdir_filter,
1890 * Configure globally the masks for flow director mode for an Ethernet device.
1891 * For example, the device can match packets with only the first 24 bits of
1892 * the IPv4 source address.
1894 * The following fields can be masked: IPv4 addresses and L4 port numbers.
1895 * The following fields can be either enabled or disabled completely for the
1896 * matching functionality: VLAN ID tag; VLAN Priority + CFI bit; Flexible 2-byte
1898 * IPv6 masks are not supported.
1900 * All filters must comply with the masks previously configured.
1901 * For example, with a mask equal to 255.255.255.0 for the source IPv4 address,
1902 * all IPv4 filters must be created with a source IPv4 address that fits the
1905 * This function flushes all filters that have been previously added in
1909 * The port identifier of the Ethernet device.
1911 * The pointer to the fdir mask structure describing relevant headers fields
1912 * and relevant bits to use when matching packets addresses and ports.
1913 * IPv6 masks are not supported.
1915 * - (0) if successful.
1916 * - (-ENOTSUP) if hardware doesn't support flow director mode.
1917 * - (-ENODEV) if *port_id* invalid.
1918 * - (-ENOSYS) if the flow director mode is not configured in perfect
1919 * mode on *port_id*.
1920 * - (-EINVAL) if the fdir_filter information is not correct
1922 int rte_eth_dev_fdir_set_masks(uint8_t port_id,
1923 struct rte_fdir_masks *fdir_mask);
1926 * The eth device event type for interrupt, and maybe others in the future.
1928 enum rte_eth_event_type {
1929 RTE_ETH_EVENT_UNKNOWN, /**< unknown event type */
1930 RTE_ETH_EVENT_INTR_LSC, /**< lsc interrupt event */
1931 RTE_ETH_EVENT_MAX /**< max value of this enum */
1934 typedef void (*rte_eth_dev_cb_fn)(uint8_t port_id, \
1935 enum rte_eth_event_type event, void *cb_arg);
1936 /**< user application callback to be registered for interrupts */
1941 * Register a callback function for specific port id.
1948 * User supplied callback function to be called.
1950 * Pointer to the parameters for the registered callback.
1953 * - On success, zero.
1954 * - On failure, a negative value.
1956 int rte_eth_dev_callback_register(uint8_t port_id,
1957 enum rte_eth_event_type event,
1958 rte_eth_dev_cb_fn cb_fn, void *cb_arg);
1961 * Unregister a callback function for specific port id.
1968 * User supplied callback function to be called.
1970 * Pointer to the parameters for the registered callback. -1 means to
1971 * remove all for the same callback address and same event.
1974 * - On success, zero.
1975 * - On failure, a negative value.
1977 int rte_eth_dev_callback_unregister(uint8_t port_id,
1978 enum rte_eth_event_type event,
1979 rte_eth_dev_cb_fn cb_fn, void *cb_arg);
1982 * @internal Executes all the user application registered callbacks for
1983 * the specific device. It is for DPDK internal user only. User
1984 * application should not call it directly.
1987 * Pointer to struct rte_eth_dev.
1989 * Eth device interrupt event type.
1994 void _rte_eth_dev_callback_process(struct rte_eth_dev *dev,
1995 enum rte_eth_event_type event);
1998 * Turn on the LED on the Ethernet device.
1999 * This function turns on the LED on the Ethernet device.
2002 * The port identifier of the Ethernet device.
2004 * - (0) if successful.
2005 * - (-ENOTSUP) if underlying hardware OR driver doesn't support
2007 * - (-ENODEV) if *port_id* invalid.
2009 int rte_eth_led_on(uint8_t port_id);
2012 * Turn off the LED on the Ethernet device.
2013 * This function turns off the LED on the Ethernet device.
2016 * The port identifier of the Ethernet device.
2018 * - (0) if successful.
2019 * - (-ENOTSUP) if underlying hardware OR driver doesn't support
2021 * - (-ENODEV) if *port_id* invalid.
2023 int rte_eth_led_off(uint8_t port_id);
2026 * Configure the Ethernet link flow control for Ethernet device
2029 * The port identifier of the Ethernet device.
2031 * The pointer to the structure of the flow control parameters.
2033 * - (0) if successful.
2034 * - (-ENOTSUP) if hardware doesn't support flow control mode.
2035 * - (-ENODEV) if *port_id* invalid.
2036 * - (-EINVAL) if bad parameter
2037 * - (-EIO) if flow control setup failure
2039 int rte_eth_dev_flow_ctrl_set(uint8_t port_id,
2040 struct rte_eth_fc_conf *fc_conf);
2043 * Configure the Ethernet priority flow control under DCB environment
2044 * for Ethernet device.
2047 * The port identifier of the Ethernet device.
2049 * The pointer to the structure of the priority flow control parameters.
2051 * - (0) if successful.
2052 * - (-ENOTSUP) if hardware doesn't support priority flow control mode.
2053 * - (-ENODEV) if *port_id* invalid.
2054 * - (-EINVAL) if bad parameter
2055 * - (-EIO) if flow control setup failure
2057 int rte_eth_dev_priority_flow_ctrl_set(uint8_t port_id,
2058 struct rte_eth_pfc_conf *pfc_conf);
2061 * Add a MAC address to an internal array of addresses used to enable whitelist
2062 * filtering to accept packets only if the destination MAC address matches.
2065 * The port identifier of the Ethernet device.
2067 * The MAC address to add.
2069 * VMDq pool index to associate address with (if VMDq is enabled). If VMDq is
2070 * not enabled, this should be set to 0.
2072 * - (0) if successfully added or *mac_addr" was already added.
2073 * - (-ENOTSUP) if hardware doesn't support this feature.
2074 * - (-ENODEV) if *port* is invalid.
2075 * - (-ENOSPC) if no more MAC addresses can be added.
2076 * - (-EINVAL) if MAC address is invalid.
2078 int rte_eth_dev_mac_addr_add(uint8_t port, struct ether_addr *mac_addr,
2082 * Remove a MAC address from the internal array of addresses.
2085 * The port identifier of the Ethernet device.
2087 * MAC address to remove.
2089 * - (0) if successful, or *mac_addr* didn't exist.
2090 * - (-ENOTSUP) if hardware doesn't support.
2091 * - (-ENODEV) if *port* invalid.
2092 * - (-EADDRINUSE) if attempting to remove the default MAC address
2094 int rte_eth_dev_mac_addr_remove(uint8_t port, struct ether_addr *mac_addr);
2101 #endif /* _RTE_ETHDEV_H_ */