netdevice.h 95.6 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5 6 7 8 9
/*
 * INET		An implementation of the TCP/IP protocol suite for the LINUX
 *		operating system.  INET is implemented using the  BSD Socket
 *		interface as the means of communication with the user level.
 *
 *		Definitions for the Interfaces handler.
 *
 * Version:	@(#)dev.h	1.0.10	08/12/93
 *
10
 * Authors:	Ross Biro
L
Linus Torvalds 已提交
11 12 13
 *		Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG>
 *		Corey Minyard <wf-rch!minyard@relay.EU.net>
 *		Donald J. Becker, <becker@cesdis.gsfc.nasa.gov>
14
 *		Alan Cox, <alan@lxorguk.ukuu.org.uk>
L
Linus Torvalds 已提交
15 16 17 18 19 20 21 22 23 24 25 26 27
 *		Bjorn Ekwall. <bj0rn@blox.se>
 *              Pekka Riikonen <priikone@poseidon.pspt.fi>
 *
 *		This program is free software; you can redistribute it and/or
 *		modify it under the terms of the GNU General Public License
 *		as published by the Free Software Foundation; either version
 *		2 of the License, or (at your option) any later version.
 *
 *		Moved to /usr/include/linux for NET3
 */
#ifndef _LINUX_NETDEVICE_H
#define _LINUX_NETDEVICE_H

28
#include <linux/pm_qos.h>
A
Al Viro 已提交
29
#include <linux/timer.h>
30
#include <linux/bug.h>
31
#include <linux/delay.h>
A
Arun Sharma 已提交
32
#include <linux/atomic.h>
L
Linus Torvalds 已提交
33 34 35 36
#include <asm/cache.h>
#include <asm/byteorder.h>

#include <linux/percpu.h>
37
#include <linux/rculist.h>
38
#include <linux/dmaengine.h>
39
#include <linux/workqueue.h>
T
Tom Herbert 已提交
40
#include <linux/dynamic_queue_limits.h>
L
Linus Torvalds 已提交
41

42
#include <linux/ethtool.h>
43
#include <net/net_namespace.h>
44
#include <net/dsa.h>
J
Jeff Kirsher 已提交
45
#ifdef CONFIG_DCB
46 47
#include <net/dcbnl.h>
#endif
48
#include <net/netprio_cgroup.h>
49

50
#include <linux/netdev_features.h>
51
#include <linux/neighbour.h>
52
#include <uapi/linux/netdevice.h>
53

54
struct netpoll_info;
55
struct device;
56
struct phy_device;
57 58
/* 802.11 specific */
struct wireless_dev;
L
Linus Torvalds 已提交
59 60 61 62
					/* source back-compat hooks */
#define SET_ETHTOOL_OPS(netdev,ops) \
	( (netdev)->ethtool_ops = (ops) )

63 64 65
extern void netdev_set_default_ethtool_ops(struct net_device *dev,
					   const struct ethtool_ops *ops);

66 67 68 69
/* hardware address assignment types */
#define NET_ADDR_PERM		0	/* address is permanent (default) */
#define NET_ADDR_RANDOM		1	/* address is generated randomly */
#define NET_ADDR_STOLEN		2	/* address is stolen from other device */
J
Jiri Pirko 已提交
70 71
#define NET_ADDR_SET		3	/* address is set using
					 * dev_set_mac_address() */
72

73 74 75 76
/* Backlog congestion levels */
#define NET_RX_SUCCESS		0	/* keep 'em coming, baby */
#define NET_RX_DROP		1	/* packet dropped */

77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95
/*
 * Transmit return codes: transmit return codes originate from three different
 * namespaces:
 *
 * - qdisc return codes
 * - driver transmit return codes
 * - errno values
 *
 * Drivers are allowed to return any one of those in their hard_start_xmit()
 * function. Real network devices commonly used with qdiscs should only return
 * the driver transmit return codes though - when qdiscs are used, the actual
 * transmission happens asynchronously, so the value is not propagated to
 * higher layers. Virtual network devices transmit synchronously, in this case
 * the driver transmit return codes are consumed by dev_queue_xmit(), all
 * others are propagated to higher layers.
 */

/* qdisc ->enqueue() return codes. */
#define NET_XMIT_SUCCESS	0x00
96 97 98 99
#define NET_XMIT_DROP		0x01	/* skb dropped			*/
#define NET_XMIT_CN		0x02	/* congestion notification	*/
#define NET_XMIT_POLICED	0x03	/* skb is shot by police	*/
#define NET_XMIT_MASK		0x0f	/* qdisc flags in net/sch_generic.h */
L
Linus Torvalds 已提交
100

101 102 103
/* NET_XMIT_CN is special. It does not guarantee that this packet is lost. It
 * indicates that the device will soon be dropping packets, or already drops
 * some packets of the same priority; prompting us to send less aggressively. */
104
#define net_xmit_eval(e)	((e) == NET_XMIT_CN ? 0 : (e))
L
Linus Torvalds 已提交
105 106
#define net_xmit_errno(e)	((e) != NET_XMIT_CN ? -ENOBUFS : 0)

107
/* Driver transmit return codes */
108
#define NETDEV_TX_MASK		0xf0
109

110
enum netdev_tx {
111
	__NETDEV_TX_MIN	 = INT_MIN,	/* make sure enum is signed */
112 113 114
	NETDEV_TX_OK	 = 0x00,	/* driver took care of packet */
	NETDEV_TX_BUSY	 = 0x10,	/* driver tx path was busy*/
	NETDEV_TX_LOCKED = 0x20,	/* driver tx lock was already taken */
115 116 117
};
typedef enum netdev_tx netdev_tx_t;

118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135
/*
 * Current order: NETDEV_TX_MASK > NET_XMIT_MASK >= 0 is significant;
 * hard_start_xmit() return < NET_XMIT_MASK means skb was consumed.
 */
static inline bool dev_xmit_complete(int rc)
{
	/*
	 * Positive cases with an skb consumed by a driver:
	 * - successful transmission (rc == NETDEV_TX_OK)
	 * - error while transmitting (rc < 0)
	 * - error while queueing to a different device (rc & NET_XMIT_MASK)
	 */
	if (likely(rc < NET_XMIT_MASK))
		return true;

	return false;
}

L
Linus Torvalds 已提交
136 137 138 139
/*
 *	Compute the worst case header length according to the protocols
 *	used.
 */
G
Graf Yang 已提交
140

141
#if defined(CONFIG_WLAN) || IS_ENABLED(CONFIG_AX25)
142 143 144 145 146
# if defined(CONFIG_MAC80211_MESH)
#  define LL_MAX_HEADER 128
# else
#  define LL_MAX_HEADER 96
# endif
L
Linus Torvalds 已提交
147
#else
148
# define LL_MAX_HEADER 32
L
Linus Torvalds 已提交
149 150
#endif

151 152
#if !IS_ENABLED(CONFIG_NET_IPIP) && !IS_ENABLED(CONFIG_NET_IPGRE) && \
    !IS_ENABLED(CONFIG_IPV6_SIT) && !IS_ENABLED(CONFIG_IPV6_TUNNEL)
L
Linus Torvalds 已提交
153 154 155 156 157 158
#define MAX_HEADER LL_MAX_HEADER
#else
#define MAX_HEADER (LL_MAX_HEADER + 48)
#endif

/*
159 160
 *	Old network device statistics. Fields are native words
 *	(unsigned long) so they can be read and written atomically.
L
Linus Torvalds 已提交
161
 */
G
Graf Yang 已提交
162

E
Eric Dumazet 已提交
163
struct net_device_stats {
164 165 166 167 168 169 170 171 172
	unsigned long	rx_packets;
	unsigned long	tx_packets;
	unsigned long	rx_bytes;
	unsigned long	tx_bytes;
	unsigned long	rx_errors;
	unsigned long	tx_errors;
	unsigned long	rx_dropped;
	unsigned long	tx_dropped;
	unsigned long	multicast;
L
Linus Torvalds 已提交
173 174
	unsigned long	collisions;
	unsigned long	rx_length_errors;
175 176 177 178 179
	unsigned long	rx_over_errors;
	unsigned long	rx_crc_errors;
	unsigned long	rx_frame_errors;
	unsigned long	rx_fifo_errors;
	unsigned long	rx_missed_errors;
L
Linus Torvalds 已提交
180 181 182 183 184 185 186 187 188 189 190 191 192
	unsigned long	tx_aborted_errors;
	unsigned long	tx_carrier_errors;
	unsigned long	tx_fifo_errors;
	unsigned long	tx_heartbeat_errors;
	unsigned long	tx_window_errors;
	unsigned long	rx_compressed;
	unsigned long	tx_compressed;
};


#include <linux/cache.h>
#include <linux/skbuff.h>

193
#ifdef CONFIG_RPS
194 195
#include <linux/static_key.h>
extern struct static_key rps_needed;
196 197
#endif

L
Linus Torvalds 已提交
198 199 200 201
struct neighbour;
struct neigh_parms;
struct sk_buff;

202 203 204 205
struct netdev_hw_addr {
	struct list_head	list;
	unsigned char		addr[MAX_ADDR_LEN];
	unsigned char		type;
J
Jiri Pirko 已提交
206 207 208 209
#define NETDEV_HW_ADDR_T_LAN		1
#define NETDEV_HW_ADDR_T_SAN		2
#define NETDEV_HW_ADDR_T_SLAVE		3
#define NETDEV_HW_ADDR_T_UNICAST	4
210 211
#define NETDEV_HW_ADDR_T_MULTICAST	5
	bool			global_use;
212
	int			sync_cnt;
213
	int			refcount;
214
	int			synced;
215 216 217
	struct rcu_head		rcu_head;
};

218 219 220 221 222
struct netdev_hw_addr_list {
	struct list_head	list;
	int			count;
};

223 224 225 226
#define netdev_hw_addr_list_count(l) ((l)->count)
#define netdev_hw_addr_list_empty(l) (netdev_hw_addr_list_count(l) == 0)
#define netdev_hw_addr_list_for_each(ha, l) \
	list_for_each_entry(ha, &(l)->list, list)
227

228 229 230 231
#define netdev_uc_count(dev) netdev_hw_addr_list_count(&(dev)->uc)
#define netdev_uc_empty(dev) netdev_hw_addr_list_empty(&(dev)->uc)
#define netdev_for_each_uc_addr(ha, dev) \
	netdev_hw_addr_list_for_each(ha, &(dev)->uc)
232

233 234
#define netdev_mc_count(dev) netdev_hw_addr_list_count(&(dev)->mc)
#define netdev_mc_empty(dev) netdev_hw_addr_list_empty(&(dev)->mc)
235
#define netdev_for_each_mc_addr(ha, dev) \
236
	netdev_hw_addr_list_for_each(ha, &(dev)->mc)
237

E
Eric Dumazet 已提交
238
struct hh_cache {
239
	u16		hh_len;
240
	u16		__pad;
241
	seqlock_t	hh_lock;
L
Linus Torvalds 已提交
242 243 244 245

	/* cached hardware header; allow for machine alignment needs.        */
#define HH_DATA_MOD	16
#define HH_DATA_OFF(__len) \
J
Jiri Benc 已提交
246
	(HH_DATA_MOD - (((__len - 1) & (HH_DATA_MOD - 1)) + 1))
L
Linus Torvalds 已提交
247 248 249 250 251 252 253 254 255 256 257 258 259 260
#define HH_DATA_ALIGN(__len) \
	(((__len)+(HH_DATA_MOD-1))&~(HH_DATA_MOD - 1))
	unsigned long	hh_data[HH_DATA_ALIGN(LL_MAX_HEADER) / sizeof(long)];
};

/* Reserve HH_DATA_MOD byte aligned hard_header_len, but at least that much.
 * Alternative is:
 *   dev->hard_header_len ? (dev->hard_header_len +
 *                           (HH_DATA_MOD - 1)) & ~(HH_DATA_MOD - 1) : 0
 *
 * We could use other alignment values, but we must maintain the
 * relationship HH alignment <= LL alignment.
 */
#define LL_RESERVED_SPACE(dev) \
261
	((((dev)->hard_header_len+(dev)->needed_headroom)&~(HH_DATA_MOD - 1)) + HH_DATA_MOD)
L
Linus Torvalds 已提交
262
#define LL_RESERVED_SPACE_EXTRA(dev,extra) \
263
	((((dev)->hard_header_len+(dev)->needed_headroom+(extra))&~(HH_DATA_MOD - 1)) + HH_DATA_MOD)
L
Linus Torvalds 已提交
264

265 266 267
struct header_ops {
	int	(*create) (struct sk_buff *skb, struct net_device *dev,
			   unsigned short type, const void *daddr,
268
			   const void *saddr, unsigned int len);
269 270
	int	(*parse)(const struct sk_buff *skb, unsigned char *haddr);
	int	(*rebuild)(struct sk_buff *skb);
271
	int	(*cache)(const struct neighbour *neigh, struct hh_cache *hh, __be16 type);
272 273 274 275 276
	void	(*cache_update)(struct hh_cache *hh,
				const struct net_device *dev,
				const unsigned char *haddr);
};

L
Linus Torvalds 已提交
277 278 279 280 281
/* These flag bits are private to the generic network queueing
 * layer, they may not be explicitly referenced by any other
 * code.
 */

E
Eric Dumazet 已提交
282
enum netdev_state_t {
L
Linus Torvalds 已提交
283 284 285
	__LINK_STATE_START,
	__LINK_STATE_PRESENT,
	__LINK_STATE_NOCARRIER,
S
Stefan Rompf 已提交
286 287
	__LINK_STATE_LINKWATCH_PENDING,
	__LINK_STATE_DORMANT,
L
Linus Torvalds 已提交
288 289 290 291 292
};


/*
 * This structure holds at boot time configured netdevice settings. They
G
Graf Yang 已提交
293
 * are then used in the device probing.
L
Linus Torvalds 已提交
294 295 296 297 298 299 300
 */
struct netdev_boot_setup {
	char name[IFNAMSIZ];
	struct ifmap map;
};
#define NETDEV_BOOT_SETUP_MAX 8

301
extern int __init netdev_boot_setup(char *str);
L
Linus Torvalds 已提交
302

303 304 305 306 307 308 309 310 311 312 313 314 315 316
/*
 * Structure for NAPI scheduling similar to tasklet but with weighting
 */
struct napi_struct {
	/* The poll_list must only be managed by the entity which
	 * changes the state of the NAPI_STATE_SCHED bit.  This means
	 * whoever atomically sets that bit can add this napi_struct
	 * to the per-cpu poll_list, and whoever clears that bit
	 * can remove from the list right before clearing the bit.
	 */
	struct list_head	poll_list;

	unsigned long		state;
	int			weight;
317
	unsigned int		gro_count;
318 319 320 321 322
	int			(*poll)(struct napi_struct *, int);
#ifdef CONFIG_NETPOLL
	spinlock_t		poll_lock;
	int			poll_owner;
#endif
H
Herbert Xu 已提交
323
	struct net_device	*dev;
324
	struct sk_buff		*gro_list;
H
Herbert Xu 已提交
325
	struct sk_buff		*skb;
326
	struct list_head	dev_list;
E
Eliezer Tamir 已提交
327 328
	struct hlist_node	napi_hash_node;
	unsigned int		napi_id;
329 330
};

E
Eric Dumazet 已提交
331
enum {
332
	NAPI_STATE_SCHED,	/* Poll is scheduled */
D
David S. Miller 已提交
333
	NAPI_STATE_DISABLE,	/* Disable pending */
334
	NAPI_STATE_NPSVC,	/* Netpoll - don't dequeue from poll_list */
E
Eliezer Tamir 已提交
335
	NAPI_STATE_HASHED,	/* In NAPI hash */
336 337
};

338
enum gro_result {
339 340 341 342 343 344
	GRO_MERGED,
	GRO_MERGED_FREE,
	GRO_HELD,
	GRO_NORMAL,
	GRO_DROP,
};
345
typedef enum gro_result gro_result_t;
346

347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378
/*
 * enum rx_handler_result - Possible return values for rx_handlers.
 * @RX_HANDLER_CONSUMED: skb was consumed by rx_handler, do not process it
 * further.
 * @RX_HANDLER_ANOTHER: Do another round in receive path. This is indicated in
 * case skb->dev was changed by rx_handler.
 * @RX_HANDLER_EXACT: Force exact delivery, no wildcard.
 * @RX_HANDLER_PASS: Do nothing, passe the skb as if no rx_handler was called.
 *
 * rx_handlers are functions called from inside __netif_receive_skb(), to do
 * special processing of the skb, prior to delivery to protocol handlers.
 *
 * Currently, a net_device can only have a single rx_handler registered. Trying
 * to register a second rx_handler will return -EBUSY.
 *
 * To register a rx_handler on a net_device, use netdev_rx_handler_register().
 * To unregister a rx_handler on a net_device, use
 * netdev_rx_handler_unregister().
 *
 * Upon return, rx_handler is expected to tell __netif_receive_skb() what to
 * do with the skb.
 *
 * If the rx_handler consumed to skb in some way, it should return
 * RX_HANDLER_CONSUMED. This is appropriate when the rx_handler arranged for
 * the skb to be delivered in some other ways.
 *
 * If the rx_handler changed skb->dev, to divert the skb to another
 * net_device, it should return RX_HANDLER_ANOTHER. The rx_handler for the
 * new device will be called if it exists.
 *
 * If the rx_handler consider the skb should be ignored, it should return
 * RX_HANDLER_EXACT. The skb will only be delivered to protocol handlers that
379
 * are registered on exact device (ptype->dev == skb->dev).
380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395
 *
 * If the rx_handler didn't changed skb->dev, but want the skb to be normally
 * delivered, it should return RX_HANDLER_PASS.
 *
 * A device without a registered rx_handler will behave as if rx_handler
 * returned RX_HANDLER_PASS.
 */

enum rx_handler_result {
	RX_HANDLER_CONSUMED,
	RX_HANDLER_ANOTHER,
	RX_HANDLER_EXACT,
	RX_HANDLER_PASS,
};
typedef enum rx_handler_result rx_handler_result_t;
typedef rx_handler_result_t rx_handler_func_t(struct sk_buff **pskb);
396

397
extern void __napi_schedule(struct napi_struct *n);
398

399
static inline bool napi_disable_pending(struct napi_struct *n)
D
David S. Miller 已提交
400 401 402 403
{
	return test_bit(NAPI_STATE_DISABLE, &n->state);
}

404 405 406 407 408 409
/**
 *	napi_schedule_prep - check if napi can be scheduled
 *	@n: napi context
 *
 * Test if NAPI routine is already running, and if not mark
 * it as running.  This is used as a condition variable
D
David S. Miller 已提交
410 411
 * insure only one NAPI poll instance runs.  We also make
 * sure there is no pending NAPI disable.
412
 */
413
static inline bool napi_schedule_prep(struct napi_struct *n)
414
{
D
David S. Miller 已提交
415 416
	return !napi_disable_pending(n) &&
		!test_and_set_bit(NAPI_STATE_SCHED, &n->state);
417 418 419 420 421 422 423 424 425 426 427 428 429 430 431
}

/**
 *	napi_schedule - schedule NAPI poll
 *	@n: napi context
 *
 * Schedule NAPI poll routine to be called if it is not already
 * running.
 */
static inline void napi_schedule(struct napi_struct *n)
{
	if (napi_schedule_prep(n))
		__napi_schedule(n);
}

432
/* Try to reschedule poll. Called by dev->poll() after napi_complete().  */
433
static inline bool napi_reschedule(struct napi_struct *napi)
434 435 436
{
	if (napi_schedule_prep(napi)) {
		__napi_schedule(napi);
437
		return true;
438
	}
439
	return false;
440 441
}

442 443 444 445 446 447
/**
 *	napi_complete - NAPI processing complete
 *	@n: napi context
 *
 * Mark NAPI processing as complete.
 */
448 449
extern void __napi_complete(struct napi_struct *n);
extern void napi_complete(struct napi_struct *n);
450

E
Eliezer Tamir 已提交
451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476
/**
 *	napi_by_id - lookup a NAPI by napi_id
 *	@napi_id: hashed napi_id
 *
 * lookup @napi_id in napi_hash table
 * must be called under rcu_read_lock()
 */
extern struct napi_struct *napi_by_id(unsigned int napi_id);

/**
 *	napi_hash_add - add a NAPI to global hashtable
 *	@napi: napi context
 *
 * generate a new napi_id and store a @napi under it in napi_hash
 */
extern void napi_hash_add(struct napi_struct *napi);

/**
 *	napi_hash_del - remove a NAPI from global table
 *	@napi: napi context
 *
 * Warning: caller must observe rcu grace period
 * before freeing memory containing @napi
 */
extern void napi_hash_del(struct napi_struct *napi);

477 478 479 480 481 482 483 484 485
/**
 *	napi_disable - prevent NAPI from scheduling
 *	@n: napi context
 *
 * Stop NAPI from being scheduled on this context.
 * Waits till any outstanding processing completes.
 */
static inline void napi_disable(struct napi_struct *n)
{
D
David S. Miller 已提交
486
	set_bit(NAPI_STATE_DISABLE, &n->state);
487
	while (test_and_set_bit(NAPI_STATE_SCHED, &n->state))
488
		msleep(1);
D
David S. Miller 已提交
489
	clear_bit(NAPI_STATE_DISABLE, &n->state);
490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505
}

/**
 *	napi_enable - enable NAPI scheduling
 *	@n: napi context
 *
 * Resume NAPI from being scheduled on this context.
 * Must be paired with napi_disable.
 */
static inline void napi_enable(struct napi_struct *n)
{
	BUG_ON(!test_bit(NAPI_STATE_SCHED, &n->state));
	smp_mb__before_clear_bit();
	clear_bit(NAPI_STATE_SCHED, &n->state);
}

506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523
#ifdef CONFIG_SMP
/**
 *	napi_synchronize - wait until NAPI is not running
 *	@n: napi context
 *
 * Wait until NAPI is done being scheduled on this context.
 * Waits till any outstanding processing completes but
 * does not disable future activations.
 */
static inline void napi_synchronize(const struct napi_struct *n)
{
	while (test_bit(NAPI_STATE_SCHED, &n->state))
		msleep(1);
}
#else
# define napi_synchronize(n)	barrier()
#endif

E
Eric Dumazet 已提交
524
enum netdev_queue_state_t {
525 526
	__QUEUE_STATE_DRV_XOFF,
	__QUEUE_STATE_STACK_XOFF,
527
	__QUEUE_STATE_FROZEN,
528 529 530 531
#define QUEUE_STATE_ANY_XOFF ((1 << __QUEUE_STATE_DRV_XOFF)		| \
			      (1 << __QUEUE_STATE_STACK_XOFF))
#define QUEUE_STATE_ANY_XOFF_OR_FROZEN (QUEUE_STATE_ANY_XOFF		| \
					(1 << __QUEUE_STATE_FROZEN))
532
};
533 534 535 536 537 538 539 540 541
/*
 * __QUEUE_STATE_DRV_XOFF is used by drivers to stop the transmit queue.  The
 * netif_tx_* functions below are used to manipulate this flag.  The
 * __QUEUE_STATE_STACK_XOFF flag is used by the stack to stop the transmit
 * queue independently.  The netif_xmit_*stopped functions below are called
 * to check if the queue has been stopped by the driver or stack (either
 * of the XOFF bits are set in the state).  Drivers should not need to call
 * netif_xmit*stopped functions, they should only be using netif_tx_*.
 */
542

543
struct netdev_queue {
544 545 546
/*
 * read mostly part
 */
547
	struct net_device	*dev;
548 549
	struct Qdisc		*qdisc;
	struct Qdisc		*qdisc_sleeping;
550
#ifdef CONFIG_SYSFS
T
Tom Herbert 已提交
551 552
	struct kobject		kobj;
#endif
553 554 555
#if defined(CONFIG_XPS) && defined(CONFIG_NUMA)
	int			numa_node;
#endif
556 557 558 559 560
/*
 * write mostly part
 */
	spinlock_t		_xmit_lock ____cacheline_aligned_in_smp;
	int			xmit_lock_owner;
561 562 563 564
	/*
	 * please use this field instead of dev->trans_start
	 */
	unsigned long		trans_start;
565 566 567 568 569 570

	/*
	 * Number of TX timeouts for this queue
	 * (/sys/class/net/DEV/Q/trans_timeout)
	 */
	unsigned long		trans_timeout;
T
Tom Herbert 已提交
571 572 573 574 575 576

	unsigned long		state;

#ifdef CONFIG_BQL
	struct dql		dql;
#endif
577
} ____cacheline_aligned_in_smp;
578

579 580 581 582 583
static inline int netdev_queue_numa_node_read(const struct netdev_queue *q)
{
#if defined(CONFIG_XPS) && defined(CONFIG_NUMA)
	return q->numa_node;
#else
584
	return NUMA_NO_NODE;
585 586 587 588 589 590 591 592 593 594
#endif
}

static inline void netdev_queue_numa_node_write(struct netdev_queue *q, int node)
{
#if defined(CONFIG_XPS) && defined(CONFIG_NUMA)
	q->numa_node = node;
#endif
}

E
Eric Dumazet 已提交
595
#ifdef CONFIG_RPS
T
Tom Herbert 已提交
596 597 598 599 600 601 602 603 604
/*
 * This structure holds an RPS map which can be of variable length.  The
 * map is an array of CPUs.
 */
struct rps_map {
	unsigned int len;
	struct rcu_head rcu;
	u16 cpus[0];
};
605
#define RPS_MAP_SIZE(_num) (sizeof(struct rps_map) + ((_num) * sizeof(u16)))
T
Tom Herbert 已提交
606

T
Tom Herbert 已提交
607
/*
608 609 610
 * The rps_dev_flow structure contains the mapping of a flow to a CPU, the
 * tail pointer for that CPU's input queue at the time of last enqueue, and
 * a hardware filter index.
T
Tom Herbert 已提交
611 612 613
 */
struct rps_dev_flow {
	u16 cpu;
614
	u16 filter;
T
Tom Herbert 已提交
615 616
	unsigned int last_qtail;
};
617
#define RPS_NO_FILTER 0xffff
T
Tom Herbert 已提交
618 619 620 621 622 623 624 625 626 627

/*
 * The rps_dev_flow_table structure contains a table of flow mappings.
 */
struct rps_dev_flow_table {
	unsigned int mask;
	struct rcu_head rcu;
	struct rps_dev_flow flows[0];
};
#define RPS_DEV_FLOW_TABLE_SIZE(_num) (sizeof(struct rps_dev_flow_table) + \
628
    ((_num) * sizeof(struct rps_dev_flow)))
T
Tom Herbert 已提交
629 630 631 632 633 634 635 636 637 638

/*
 * The rps_sock_flow_table contains mappings of flows to the last CPU
 * on which they were processed by the application (set in recvmsg).
 */
struct rps_sock_flow_table {
	unsigned int mask;
	u16 ents[0];
};
#define	RPS_SOCK_FLOW_TABLE_SIZE(_num) (sizeof(struct rps_sock_flow_table) + \
639
    ((_num) * sizeof(u16)))
T
Tom Herbert 已提交
640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663

#define RPS_NO_CPU 0xffff

static inline void rps_record_sock_flow(struct rps_sock_flow_table *table,
					u32 hash)
{
	if (table && hash) {
		unsigned int cpu, index = hash & table->mask;

		/* We only give a hint, preemption can change cpu under us */
		cpu = raw_smp_processor_id();

		if (table->ents[index] != cpu)
			table->ents[index] = cpu;
	}
}

static inline void rps_reset_sock_flow(struct rps_sock_flow_table *table,
				       u32 hash)
{
	if (table && hash)
		table->ents[hash & table->mask] = RPS_NO_CPU;
}

E
Eric Dumazet 已提交
664
extern struct rps_sock_flow_table __rcu *rps_sock_flow_table;
T
Tom Herbert 已提交
665

666 667 668 669 670
#ifdef CONFIG_RFS_ACCEL
extern bool rps_may_expire_flow(struct net_device *dev, u16 rxq_index,
				u32 flow_id, u16 filter_id);
#endif

T
Tom Herbert 已提交
671 672
/* This structure contains an instance of an RX queue. */
struct netdev_rx_queue {
E
Eric Dumazet 已提交
673 674 675
	struct rps_map __rcu		*rps_map;
	struct rps_dev_flow_table __rcu	*rps_flow_table;
	struct kobject			kobj;
T
Tom Herbert 已提交
676
	struct net_device		*dev;
T
Tom Herbert 已提交
677
} ____cacheline_aligned_in_smp;
T
Tom Herbert 已提交
678
#endif /* CONFIG_RPS */
679

T
Tom Herbert 已提交
680 681 682 683 684 685 686 687 688 689 690
#ifdef CONFIG_XPS
/*
 * This structure holds an XPS map which can be of variable length.  The
 * map is an array of queues.
 */
struct xps_map {
	unsigned int len;
	unsigned int alloc_len;
	struct rcu_head rcu;
	u16 queues[0];
};
691
#define XPS_MAP_SIZE(_num) (sizeof(struct xps_map) + ((_num) * sizeof(u16)))
T
Tom Herbert 已提交
692 693 694 695 696 697 698 699
#define XPS_MIN_MAP_ALLOC ((L1_CACHE_BYTES - sizeof(struct xps_map))	\
    / sizeof(u16))

/*
 * This structure holds all XPS maps for device.  Maps are indexed by CPU.
 */
struct xps_dev_maps {
	struct rcu_head rcu;
E
Eric Dumazet 已提交
700
	struct xps_map __rcu *cpu_map[0];
T
Tom Herbert 已提交
701 702 703 704 705
};
#define XPS_DEV_MAPS_SIZE (sizeof(struct xps_dev_maps) +		\
    (nr_cpu_ids * sizeof(struct xps_map *)))
#endif /* CONFIG_XPS */

706 707 708 709 710 711 712 713
#define TC_MAX_QUEUE	16
#define TC_BITMASK	15
/* HW offloaded queuing disciplines txq count and offset maps */
struct netdev_tc_txq {
	u16 count;
	u16 offset;
};

714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730
#if defined(CONFIG_FCOE) || defined(CONFIG_FCOE_MODULE)
/*
 * This structure is to hold information about the device
 * configured to run FCoE protocol stack.
 */
struct netdev_fcoe_hbainfo {
	char	manufacturer[64];
	char	serial_number[64];
	char	hardware_version[64];
	char	driver_version[64];
	char	optionrom_version[64];
	char	firmware_version[64];
	char	model[256];
	char	model_description[256];
};
#endif

731 732 733 734 735 736 737 738 739 740
#define MAX_PHYS_PORT_ID_LEN 32

/* This structure holds a unique identifier to identify the
 * physical port used by a netdevice.
 */
struct netdev_phys_port_id {
	unsigned char id[MAX_PHYS_PORT_ID_LEN];
	unsigned char id_len;
};

741 742
/*
 * This structure defines the management hooks for network devices.
743 744
 * The following hooks can be defined; unless noted otherwise, they are
 * optional and can be filled with a null pointer.
745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763
 *
 * int (*ndo_init)(struct net_device *dev);
 *     This function is called once when network device is registered.
 *     The network device can use this to any late stage initializaton
 *     or semantic validattion. It can fail with an error code which will
 *     be propogated back to register_netdev
 *
 * void (*ndo_uninit)(struct net_device *dev);
 *     This function is called when device is unregistered or when registration
 *     fails. It is not called if init fails.
 *
 * int (*ndo_open)(struct net_device *dev);
 *     This function is called when network device transistions to the up
 *     state.
 *
 * int (*ndo_stop)(struct net_device *dev);
 *     This function is called when network device transistions to the down
 *     state.
 *
764 765
 * netdev_tx_t (*ndo_start_xmit)(struct sk_buff *skb,
 *                               struct net_device *dev);
766
 *	Called when a packet needs to be transmitted.
767 768
 *	Must return NETDEV_TX_OK , NETDEV_TX_BUSY.
 *        (can also return NETDEV_TX_LOCKED iff NETIF_F_LLTX)
769 770 771 772 773 774
 *	Required can not be NULL.
 *
 * u16 (*ndo_select_queue)(struct net_device *dev, struct sk_buff *skb);
 *	Called to decide which queue to when device supports multiple
 *	transmit queues.
 *
775 776 777 778 779 780
 * void (*ndo_change_rx_flags)(struct net_device *dev, int flags);
 *	This function is called to allow device receiver to make
 *	changes to configuration when multicast or promiscious is enabled.
 *
 * void (*ndo_set_rx_mode)(struct net_device *dev);
 *	This function is called device changes address list filtering.
781 782
 *	If driver handles unicast address filtering, it should set
 *	IFF_UNICAST_FLT to its priv_flags.
783 784 785
 *
 * int (*ndo_set_mac_address)(struct net_device *dev, void *addr);
 *	This function  is called when the Media Access Control address
786
 *	needs to be changed. If this interface is not defined, the
787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806
 *	mac address can not be changed.
 *
 * int (*ndo_validate_addr)(struct net_device *dev);
 *	Test if Media Access Control address is valid for the device.
 *
 * int (*ndo_do_ioctl)(struct net_device *dev, struct ifreq *ifr, int cmd);
 *	Called when a user request an ioctl which can't be handled by
 *	the generic interface code. If not defined ioctl's return
 *	not supported error code.
 *
 * int (*ndo_set_config)(struct net_device *dev, struct ifmap *map);
 *	Used to set network devices bus interface parameters. This interface
 *	is retained for legacy reason, new devices should use the bus
 *	interface (PCI) for low level management.
 *
 * int (*ndo_change_mtu)(struct net_device *dev, int new_mtu);
 *	Called when a user wants to change the Maximum Transfer Unit
 *	of a device. If not defined, any request to change MTU will
 *	will return an error.
 *
807
 * void (*ndo_tx_timeout)(struct net_device *dev);
808 809 810
 *	Callback uses when the transmitter has not made any progress
 *	for dev->watchdog ticks.
 *
811
 * struct rtnl_link_stats64* (*ndo_get_stats64)(struct net_device *dev,
812
 *                      struct rtnl_link_stats64 *storage);
813
 * struct net_device_stats* (*ndo_get_stats)(struct net_device *dev);
814
 *	Called when a user wants to get the network device usage
815
 *	statistics. Drivers must do one of the following:
816 817
 *	1. Define @ndo_get_stats64 to fill in a zero-initialised
 *	   rtnl_link_stats64 structure passed by the caller.
818
 *	2. Define @ndo_get_stats to update a net_device_stats structure
819 820 821 822 823
 *	   (which should normally be dev->stats) and return a pointer to
 *	   it. The structure may be changed asynchronously only if each
 *	   field is written atomically.
 *	3. Update dev->stats asynchronously and atomically, and define
 *	   neither operation.
824
 *
825 826 827
 * int (*ndo_vlan_rx_add_vid)(struct net_device *dev, __be16 proto, u16t vid);
 *	If device support VLAN filtering this function is called when a
 *	VLAN id is registered.
828
 *
829
 * int (*ndo_vlan_rx_kill_vid)(struct net_device *dev, unsigned short vid);
830 831
 *	If device support VLAN filtering this function is called when a
 *	VLAN id is unregistered.
832 833
 *
 * void (*ndo_poll_controller)(struct net_device *dev);
834 835 836 837 838
 *
 *	SR-IOV management functions.
 * int (*ndo_set_vf_mac)(struct net_device *dev, int vf, u8* mac);
 * int (*ndo_set_vf_vlan)(struct net_device *dev, int vf, u16 vlan, u8 qos);
 * int (*ndo_set_vf_tx_rate)(struct net_device *dev, int vf, int rate);
839
 * int (*ndo_set_vf_spoofchk)(struct net_device *dev, int vf, bool setting);
840 841
 * int (*ndo_get_vf_config)(struct net_device *dev,
 *			    int vf, struct ifla_vf_info *ivf);
842
 * int (*ndo_set_vf_link_state)(struct net_device *dev, int vf, int link_state);
843 844 845
 * int (*ndo_set_vf_port)(struct net_device *dev, int vf,
 *			  struct nlattr *port[]);
 * int (*ndo_get_vf_port)(struct net_device *dev, int vf, struct sk_buff *skb);
846 847 848 849 850
 * int (*ndo_setup_tc)(struct net_device *dev, u8 tc)
 * 	Called to setup 'tc' number of traffic classes in the net device. This
 * 	is always called from the stack with the rtnl lock held and netif tx
 * 	queues stopped. This allows the netdevice to perform queue management
 * 	safely.
851
 *
852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881
 *	Fiber Channel over Ethernet (FCoE) offload functions.
 * int (*ndo_fcoe_enable)(struct net_device *dev);
 *	Called when the FCoE protocol stack wants to start using LLD for FCoE
 *	so the underlying device can perform whatever needed configuration or
 *	initialization to support acceleration of FCoE traffic.
 *
 * int (*ndo_fcoe_disable)(struct net_device *dev);
 *	Called when the FCoE protocol stack wants to stop using LLD for FCoE
 *	so the underlying device can perform whatever needed clean-ups to
 *	stop supporting acceleration of FCoE traffic.
 *
 * int (*ndo_fcoe_ddp_setup)(struct net_device *dev, u16 xid,
 *			     struct scatterlist *sgl, unsigned int sgc);
 *	Called when the FCoE Initiator wants to initialize an I/O that
 *	is a possible candidate for Direct Data Placement (DDP). The LLD can
 *	perform necessary setup and returns 1 to indicate the device is set up
 *	successfully to perform DDP on this I/O, otherwise this returns 0.
 *
 * int (*ndo_fcoe_ddp_done)(struct net_device *dev,  u16 xid);
 *	Called when the FCoE Initiator/Target is done with the DDPed I/O as
 *	indicated by the FC exchange id 'xid', so the underlying device can
 *	clean up and reuse resources for later DDP requests.
 *
 * int (*ndo_fcoe_ddp_target)(struct net_device *dev, u16 xid,
 *			      struct scatterlist *sgl, unsigned int sgc);
 *	Called when the FCoE Target wants to initialize an I/O that
 *	is a possible candidate for Direct Data Placement (DDP). The LLD can
 *	perform necessary setup and returns 1 to indicate the device is set up
 *	successfully to perform DDP on this I/O, otherwise this returns 0.
 *
882 883 884 885 886 887 888
 * int (*ndo_fcoe_get_hbainfo)(struct net_device *dev,
 *			       struct netdev_fcoe_hbainfo *hbainfo);
 *	Called when the FCoE Protocol stack wants information on the underlying
 *	device. This information is utilized by the FCoE protocol stack to
 *	register attributes with Fiber Channel management service as per the
 *	FC-GS Fabric Device Management Information(FDMI) specification.
 *
889 890 891 892 893 894
 * int (*ndo_fcoe_get_wwn)(struct net_device *dev, u64 *wwn, int type);
 *	Called when the underlying device wants to override default World Wide
 *	Name (WWN) generation mechanism in FCoE protocol stack to pass its own
 *	World Wide Port Name (WWPN) or World Wide Node Name (WWNN) to the FCoE
 *	protocol stack to use.
 *
895 896 897 898 899 900
 *	RFS acceleration.
 * int (*ndo_rx_flow_steer)(struct net_device *dev, const struct sk_buff *skb,
 *			    u16 rxq_index, u32 flow_id);
 *	Set hardware filter for RFS.  rxq_index is the target queue index;
 *	flow_id is a flow ID to be passed to rps_may_expire_flow() later.
 *	Return the filter ID on success, or a negative error code.
901
 *
902
 *	Slave management functions (for bridge, bonding, etc).
903 904 905 906 907
 * int (*ndo_add_slave)(struct net_device *dev, struct net_device *slave_dev);
 *	Called to make another netdev an underling.
 *
 * int (*ndo_del_slave)(struct net_device *dev, struct net_device *slave_dev);
 *	Called to release previously enslaved netdev.
908 909
 *
 *      Feature/offload setting functions.
910 911
 * netdev_features_t (*ndo_fix_features)(struct net_device *dev,
 *		netdev_features_t features);
912 913 914 915
 *	Adjusts the requested feature flags according to device-specific
 *	constraints, and returns the resulting flags. Must not modify
 *	the device state.
 *
916
 * int (*ndo_set_features)(struct net_device *dev, netdev_features_t features);
917 918 919 920
 *	Called to update device configuration to new features. Passed
 *	feature set might be less than what was returned by ndo_fix_features()).
 *	Must return >0 or -errno if it changed dev->features itself.
 *
921 922
 * int (*ndo_fdb_add)(struct ndmsg *ndm, struct nlattr *tb[],
 *		      struct net_device *dev,
923
 *		      const unsigned char *addr, u16 flags)
924
 *	Adds an FDB entry to dev for addr.
925 926
 * int (*ndo_fdb_del)(struct ndmsg *ndm, struct nlattr *tb[],
 *		      struct net_device *dev,
927
 *		      const unsigned char *addr)
928 929 930 931 932
 *	Deletes the FDB entry from dev coresponding to addr.
 * int (*ndo_fdb_dump)(struct sk_buff *skb, struct netlink_callback *cb,
 *		       struct net_device *dev, int idx)
 *	Used to add FDB entries to dump requests. Implementers should add
 *	entries to skb and update idx with the number of entries.
J
John Fastabend 已提交
933 934 935
 *
 * int (*ndo_bridge_setlink)(struct net_device *dev, struct nlmsghdr *nlh)
 * int (*ndo_bridge_getlink)(struct sk_buff *skb, u32 pid, u32 seq,
936
 *			     struct net_device *dev, u32 filter_mask)
J
Jiri Pirko 已提交
937 938 939 940 941 942 943 944
 *
 * int (*ndo_change_carrier)(struct net_device *dev, bool new_carrier);
 *	Called to change device carrier. Soft-devices (like dummy, team, etc)
 *	which do not represent real hardware may define this to allow their
 *	userspace components to manage their virtual carrier state. Devices
 *	that determine carrier state from physical hardware properties (eg
 *	network cables) or protocol-dependent mechanisms (eg
 *	USB_CDC_NOTIFY_NETWORK_CONNECTION) should NOT implement this function.
945 946 947 948 949 950
 *
 * int (*ndo_get_phys_port_id)(struct net_device *dev,
 *			       struct netdev_phys_port_id *ppid);
 *	Called to get ID of physical port of this device. If driver does
 *	not implement this, it is assumed that the hw is not able to have
 *	multiple net devices on single physical port.
951 952
 *
 * void (*ndo_add_vxlan_port)(struct  net_device *dev,
J
Joseph Gasparakis 已提交
953
 *			      sa_family_t sa_family, __be16 port);
954 955 956 957 958 959
 *	Called by vxlan to notiy a driver about the UDP port and socket
 *	address family that vxlan is listnening to. It is called only when
 *	a new port starts listening. The operation is protected by the
 *	vxlan_net->sock_lock.
 *
 * void (*ndo_del_vxlan_port)(struct  net_device *dev,
J
Joseph Gasparakis 已提交
960
 *			      sa_family_t sa_family, __be16 port);
961 962 963
 *	Called by vxlan to notify the driver about a UDP port and socket
 *	address family that vxlan is not listening to anymore. The operation
 *	is protected by the vxlan_net->sock_lock.
964 965 966 967 968 969
 */
struct net_device_ops {
	int			(*ndo_init)(struct net_device *dev);
	void			(*ndo_uninit)(struct net_device *dev);
	int			(*ndo_open)(struct net_device *dev);
	int			(*ndo_stop)(struct net_device *dev);
970
	netdev_tx_t		(*ndo_start_xmit) (struct sk_buff *skb,
971 972 973
						   struct net_device *dev);
	u16			(*ndo_select_queue)(struct net_device *dev,
						    struct sk_buff *skb);
974 975 976 977 978 979 980 981 982 983
	void			(*ndo_change_rx_flags)(struct net_device *dev,
						       int flags);
	void			(*ndo_set_rx_mode)(struct net_device *dev);
	int			(*ndo_set_mac_address)(struct net_device *dev,
						       void *addr);
	int			(*ndo_validate_addr)(struct net_device *dev);
	int			(*ndo_do_ioctl)(struct net_device *dev,
					        struct ifreq *ifr, int cmd);
	int			(*ndo_set_config)(struct net_device *dev,
					          struct ifmap *map);
984 985 986 987
	int			(*ndo_change_mtu)(struct net_device *dev,
						  int new_mtu);
	int			(*ndo_neigh_setup)(struct net_device *dev,
						   struct neigh_parms *);
988 989
	void			(*ndo_tx_timeout) (struct net_device *dev);

990 991
	struct rtnl_link_stats64* (*ndo_get_stats64)(struct net_device *dev,
						     struct rtnl_link_stats64 *storage);
992 993
	struct net_device_stats* (*ndo_get_stats)(struct net_device *dev);

994
	int			(*ndo_vlan_rx_add_vid)(struct net_device *dev,
995
						       __be16 proto, u16 vid);
996
	int			(*ndo_vlan_rx_kill_vid)(struct net_device *dev,
997
						        __be16 proto, u16 vid);
998 999
#ifdef CONFIG_NET_POLL_CONTROLLER
	void                    (*ndo_poll_controller)(struct net_device *dev);
H
Herbert Xu 已提交
1000
	int			(*ndo_netpoll_setup)(struct net_device *dev,
1001 1002
						     struct netpoll_info *info,
						     gfp_t gfp);
1003
	void			(*ndo_netpoll_cleanup)(struct net_device *dev);
E
Eliezer Tamir 已提交
1004
#endif
1005
#ifdef CONFIG_NET_RX_BUSY_POLL
1006
	int			(*ndo_busy_poll)(struct napi_struct *dev);
1007
#endif
1008 1009 1010 1011 1012 1013
	int			(*ndo_set_vf_mac)(struct net_device *dev,
						  int queue, u8 *mac);
	int			(*ndo_set_vf_vlan)(struct net_device *dev,
						   int queue, u16 vlan, u8 qos);
	int			(*ndo_set_vf_tx_rate)(struct net_device *dev,
						      int vf, int rate);
1014 1015
	int			(*ndo_set_vf_spoofchk)(struct net_device *dev,
						       int vf, bool setting);
1016 1017 1018
	int			(*ndo_get_vf_config)(struct net_device *dev,
						     int vf,
						     struct ifla_vf_info *ivf);
1019 1020
	int			(*ndo_set_vf_link_state)(struct net_device *dev,
							 int vf, int link_state);
1021 1022 1023 1024 1025
	int			(*ndo_set_vf_port)(struct net_device *dev,
						   int vf,
						   struct nlattr *port[]);
	int			(*ndo_get_vf_port)(struct net_device *dev,
						   int vf, struct sk_buff *skb);
1026
	int			(*ndo_setup_tc)(struct net_device *dev, u8 tc);
1027
#if IS_ENABLED(CONFIG_FCOE)
1028 1029
	int			(*ndo_fcoe_enable)(struct net_device *dev);
	int			(*ndo_fcoe_disable)(struct net_device *dev);
1030 1031 1032 1033 1034 1035
	int			(*ndo_fcoe_ddp_setup)(struct net_device *dev,
						      u16 xid,
						      struct scatterlist *sgl,
						      unsigned int sgc);
	int			(*ndo_fcoe_ddp_done)(struct net_device *dev,
						     u16 xid);
1036 1037 1038 1039
	int			(*ndo_fcoe_ddp_target)(struct net_device *dev,
						       u16 xid,
						       struct scatterlist *sgl,
						       unsigned int sgc);
1040 1041
	int			(*ndo_fcoe_get_hbainfo)(struct net_device *dev,
							struct netdev_fcoe_hbainfo *hbainfo);
1042 1043
#endif

1044
#if IS_ENABLED(CONFIG_LIBFCOE)
1045 1046 1047 1048
#define NETDEV_FCOE_WWNN 0
#define NETDEV_FCOE_WWPN 1
	int			(*ndo_fcoe_get_wwn)(struct net_device *dev,
						    u64 *wwn, int type);
1049
#endif
1050

1051 1052 1053 1054 1055 1056
#ifdef CONFIG_RFS_ACCEL
	int			(*ndo_rx_flow_steer)(struct net_device *dev,
						     const struct sk_buff *skb,
						     u16 rxq_index,
						     u32 flow_id);
#endif
1057 1058 1059 1060
	int			(*ndo_add_slave)(struct net_device *dev,
						 struct net_device *slave_dev);
	int			(*ndo_del_slave)(struct net_device *dev,
						 struct net_device *slave_dev);
1061 1062
	netdev_features_t	(*ndo_fix_features)(struct net_device *dev,
						    netdev_features_t features);
1063
	int			(*ndo_set_features)(struct net_device *dev,
1064
						    netdev_features_t features);
1065
	int			(*ndo_neigh_construct)(struct neighbour *n);
1066
	void			(*ndo_neigh_destroy)(struct neighbour *n);
1067 1068

	int			(*ndo_fdb_add)(struct ndmsg *ndm,
1069
					       struct nlattr *tb[],
1070
					       struct net_device *dev,
1071
					       const unsigned char *addr,
1072 1073
					       u16 flags);
	int			(*ndo_fdb_del)(struct ndmsg *ndm,
1074
					       struct nlattr *tb[],
1075
					       struct net_device *dev,
1076
					       const unsigned char *addr);
1077 1078 1079 1080
	int			(*ndo_fdb_dump)(struct sk_buff *skb,
						struct netlink_callback *cb,
						struct net_device *dev,
						int idx);
J
John Fastabend 已提交
1081 1082 1083 1084 1085

	int			(*ndo_bridge_setlink)(struct net_device *dev,
						      struct nlmsghdr *nlh);
	int			(*ndo_bridge_getlink)(struct sk_buff *skb,
						      u32 pid, u32 seq,
1086 1087
						      struct net_device *dev,
						      u32 filter_mask);
1088 1089
	int			(*ndo_bridge_dellink)(struct net_device *dev,
						      struct nlmsghdr *nlh);
J
Jiri Pirko 已提交
1090 1091
	int			(*ndo_change_carrier)(struct net_device *dev,
						      bool new_carrier);
1092 1093
	int			(*ndo_get_phys_port_id)(struct net_device *dev,
							struct netdev_phys_port_id *ppid);
1094 1095
	void			(*ndo_add_vxlan_port)(struct  net_device *dev,
						      sa_family_t sa_family,
J
Joseph Gasparakis 已提交
1096
						      __be16 port);
1097 1098
	void			(*ndo_del_vxlan_port)(struct  net_device *dev,
						      sa_family_t sa_family,
J
Joseph Gasparakis 已提交
1099
						      __be16 port);
1100 1101
};

L
Linus Torvalds 已提交
1102 1103 1104 1105 1106 1107 1108 1109 1110 1111
/*
 *	The DEVICE structure.
 *	Actually, this whole structure is a big mistake.  It mixes I/O
 *	data with strictly "high-level" data, and it has to know about
 *	almost every data structure used in the INET module.
 *
 *	FIXME: cleanup struct net_device such that network protocol info
 *	moves out.
 */

E
Eric Dumazet 已提交
1112
struct net_device {
L
Linus Torvalds 已提交
1113 1114 1115 1116

	/*
	 * This is the first field of the "visible" part of this structure
	 * (i.e. as seen by users in the "Space.c" file).  It is the name
1117
	 * of the interface.
L
Linus Torvalds 已提交
1118 1119
	 */
	char			name[IFNAMSIZ];
M
Mark Gross 已提交
1120

1121
	/* device name hash chain, please keep it close to name[] */
1122
	struct hlist_node	name_hlist;
1123

1124 1125
	/* snmp alias */
	char 			*ifalias;
L
Linus Torvalds 已提交
1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142

	/*
	 *	I/O specific fields
	 *	FIXME: Merge these and struct ifmap into one
	 */
	unsigned long		mem_end;	/* shared mem end	*/
	unsigned long		mem_start;	/* shared mem start	*/
	unsigned long		base_addr;	/* device I/O address	*/
	unsigned int		irq;		/* device IRQ number	*/

	/*
	 *	Some hardware also needs these fields, but they are not
	 *	part of the usual set specified in Space.c.
	 */

	unsigned long		state;

1143
	struct list_head	dev_list;
1144
	struct list_head	napi_list;
1145
	struct list_head	unreg_list;
1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157

	/* directly linked devices, like slaves for bonding */
	struct {
		struct list_head upper;
		struct list_head lower;
	} adj_list;

	/* all linked devices, *including* neighbours */
	struct {
		struct list_head upper;
		struct list_head lower;
	} all_adj_list;
1158

L
Linus Torvalds 已提交
1159

1160
	/* currently active device features */
1161
	netdev_features_t	features;
1162
	/* user-changeable features */
1163
	netdev_features_t	hw_features;
1164
	/* user-requested features */
1165
	netdev_features_t	wanted_features;
1166
	/* mask of features inheritable by VLAN devices */
1167
	netdev_features_t	vlan_features;
1168 1169 1170 1171 1172 1173
	/* mask of features inherited by encapsulating devices
	 * This field indicates what encapsulation offloads
	 * the hardware is capable of doing, and drivers will
	 * need to set them appropriately.
	 */
	netdev_features_t	hw_enc_features;
S
Simon Horman 已提交
1174 1175
	/* mask of fetures inheritable by MPLS */
	netdev_features_t	mpls_features;
1176

L
Linus Torvalds 已提交
1177 1178 1179 1180
	/* Interface index. Unique device identifier	*/
	int			ifindex;
	int			iflink;

R
Rusty Russell 已提交
1181
	struct net_device_stats	stats;
1182 1183 1184
	atomic_long_t		rx_dropped; /* dropped packets by core network
					     * Do not use this in drivers.
					     */
L
Linus Torvalds 已提交
1185

1186
#ifdef CONFIG_WIRELESS_EXT
L
Linus Torvalds 已提交
1187 1188 1189 1190 1191
	/* List of functions to handle Wireless Extensions (instead of ioctl).
	 * See <net/iw_handler.h> for details. Jean II */
	const struct iw_handler_def *	wireless_handlers;
	/* Instance data managed by the core of Wireless Extensions. */
	struct iw_public_data *	wireless_data;
1192
#endif
1193 1194
	/* Management operations */
	const struct net_device_ops *netdev_ops;
1195
	const struct ethtool_ops *ethtool_ops;
L
Linus Torvalds 已提交
1196

1197 1198 1199
	/* Hardware header description */
	const struct header_ops *header_ops;

S
Stefan Rompf 已提交
1200
	unsigned int		flags;	/* interface flags (a la BSD)	*/
1201 1202
	unsigned int		priv_flags; /* Like 'flags' but invisible to userspace.
					     * See if.h for definitions. */
L
Linus Torvalds 已提交
1203 1204 1205
	unsigned short		gflags;
	unsigned short		padded;	/* How much padding added by alloc_netdev() */

S
Stefan Rompf 已提交
1206 1207 1208
	unsigned char		operstate; /* RFC2863 operstate */
	unsigned char		link_mode; /* mapping policy to operstate */

1209 1210 1211
	unsigned char		if_port;	/* Selectable AUI, TP,..*/
	unsigned char		dma;		/* DMA channel		*/

1212
	unsigned int		mtu;	/* interface MTU value		*/
L
Linus Torvalds 已提交
1213 1214 1215
	unsigned short		type;	/* interface hardware type	*/
	unsigned short		hard_header_len;	/* hardware hdr length	*/

1216 1217 1218 1219 1220 1221 1222
	/* extra head- and tailroom the hardware may need, but not in all cases
	 * can this be guaranteed, especially tailroom. Some cases also use
	 * LL_MAX_HEADER instead to allocate the skb.
	 */
	unsigned short		needed_headroom;
	unsigned short		needed_tailroom;

L
Linus Torvalds 已提交
1223
	/* Interface address info. */
1224
	unsigned char		perm_addr[MAX_ADDR_LEN]; /* permanent hw address */
1225
	unsigned char		addr_assign_type; /* hw address assignment type */
L
Linus Torvalds 已提交
1226
	unsigned char		addr_len;	/* hardware address length	*/
1227
	unsigned char		neigh_priv_len;
1228 1229 1230 1231
	unsigned short          dev_id;		/* Used to differentiate devices
						 * that share the same link
						 * layer address
						 */
J
Jiri Pirko 已提交
1232
	spinlock_t		addr_list_lock;
1233 1234
	struct netdev_hw_addr_list	uc;	/* Unicast mac addresses */
	struct netdev_hw_addr_list	mc;	/* Multicast mac addresses */
1235 1236 1237 1238 1239 1240 1241
	struct netdev_hw_addr_list	dev_addrs; /* list of device
						    * hw addresses
						    */
#ifdef CONFIG_SYSFS
	struct kset		*queues_kset;
#endif

1242
	bool			uc_promisc;
1243 1244
	unsigned int		promiscuity;
	unsigned int		allmulti;
L
Linus Torvalds 已提交
1245 1246 1247


	/* Protocol specific pointers */
1248

1249
#if IS_ENABLED(CONFIG_VLAN_8021Q)
1250
	struct vlan_info __rcu	*vlan_info;	/* VLAN info */
1251
#endif
1252
#if IS_ENABLED(CONFIG_NET_DSA)
1253
	struct dsa_switch_tree	*dsa_ptr;	/* dsa specific data */
1254
#endif
L
Linus Torvalds 已提交
1255
	void 			*atalk_ptr;	/* AppleTalk link 	*/
E
Eric Dumazet 已提交
1256
	struct in_device __rcu	*ip_ptr;	/* IPv4 specific data	*/
1257
	struct dn_dev __rcu     *dn_ptr;        /* DECnet specific data */
E
Eric Dumazet 已提交
1258
	struct inet6_dev __rcu	*ip6_ptr;       /* IPv6 specific data */
L
Linus Torvalds 已提交
1259
	void			*ax25_ptr;	/* AX.25 specific data */
1260 1261
	struct wireless_dev	*ieee80211_ptr;	/* IEEE 802.11 specific data,
						   assign before registering */
L
Linus Torvalds 已提交
1262

1263
/*
E
Eric Dumazet 已提交
1264
 * Cache lines mostly used on receive path (including eth_type_trans())
1265
 */
1266 1267 1268 1269 1270 1271 1272 1273
	unsigned long		last_rx;	/* Time of last Rx
						 * This should not be set in
						 * drivers, unless really needed,
						 * because network stack (bonding)
						 * use it if/when necessary, to
						 * avoid dirtying this cache line.
						 */

1274
	/* Interface address info used in eth_type_trans() */
1275 1276 1277 1278
	unsigned char		*dev_addr;	/* hw address, (before bcast
						   because most packets are
						   unicast) */

T
Tom Herbert 已提交
1279

1280
#ifdef CONFIG_RPS
T
Tom Herbert 已提交
1281 1282
	struct netdev_rx_queue	*_rx;

1283
	/* Number of RX queues allocated at register_netdev() time */
T
Tom Herbert 已提交
1284
	unsigned int		num_rx_queues;
1285 1286 1287

	/* Number of RX queues currently active in device */
	unsigned int		real_num_rx_queues;
1288

E
Eric Dumazet 已提交
1289
#endif
T
Tom Herbert 已提交
1290

1291 1292
	rx_handler_func_t __rcu	*rx_handler;
	void __rcu		*rx_handler_data;
1293

1294
	struct netdev_queue __rcu *ingress_queue;
1295 1296
	unsigned char		broadcast[MAX_ADDR_LEN];	/* hw bcast add	*/

E
Eric Dumazet 已提交
1297 1298 1299 1300

/*
 * Cache lines mostly used on transmit path
 */
1301
	struct netdev_queue	*_tx ____cacheline_aligned_in_smp;
1302 1303

	/* Number of TX queues allocated at alloc_netdev_mq() time  */
1304
	unsigned int		num_tx_queues;
1305 1306 1307 1308

	/* Number of TX queues currently active in device  */
	unsigned int		real_num_tx_queues;

1309 1310 1311
	/* root qdisc from userspace point of view */
	struct Qdisc		*qdisc;

L
Linus Torvalds 已提交
1312
	unsigned long		tx_queue_len;	/* Max frames per queue allowed */
1313
	spinlock_t		tx_global_lock;
E
Eric Dumazet 已提交
1314

T
Tom Herbert 已提交
1315
#ifdef CONFIG_XPS
E
Eric Dumazet 已提交
1316
	struct xps_dev_maps __rcu *xps_maps;
T
Tom Herbert 已提交
1317
#endif
1318 1319 1320 1321 1322 1323
#ifdef CONFIG_RFS_ACCEL
	/* CPU reverse-mapping for RX completion interrupts, indexed
	 * by RX queue number.  Assigned by driver.  This must only be
	 * set if the ndo_rx_flow_steer operation is defined. */
	struct cpu_rmap		*rx_cpu_rmap;
#endif
T
Tom Herbert 已提交
1324

1325
	/* These may be needed for future network-power-down code. */
1326 1327 1328 1329 1330

	/*
	 * trans_start here is expensive for high speed devices on SMP,
	 * please use netdev_queue->trans_start instead.
	 */
1331 1332 1333 1334 1335
	unsigned long		trans_start;	/* Time (in jiffies) of last Tx	*/

	int			watchdog_timeo; /* used by dev_watchdog() */
	struct timer_list	watchdog_timer;

L
Linus Torvalds 已提交
1336
	/* Number of references to this device */
E
Eric Dumazet 已提交
1337
	int __percpu		*pcpu_refcnt;
1338

L
Linus Torvalds 已提交
1339 1340 1341 1342 1343
	/* delayed register/unregister */
	struct list_head	todo_list;
	/* device index hash chain */
	struct hlist_node	index_hlist;

1344
	struct list_head	link_watch_list;
1345

L
Linus Torvalds 已提交
1346 1347
	/* register/unregister state machine */
	enum { NETREG_UNINITIALIZED=0,
1348
	       NETREG_REGISTERED,	/* completed register_netdevice */
L
Linus Torvalds 已提交
1349 1350 1351
	       NETREG_UNREGISTERING,	/* called unregister_netdevice */
	       NETREG_UNREGISTERED,	/* completed unregister todo */
	       NETREG_RELEASED,		/* called free_netdev */
1352
	       NETREG_DUMMY,		/* dummy device for NAPI poll */
1353 1354 1355
	} reg_state:8;

	bool dismantle; /* device is going do be freed */
1356 1357 1358 1359 1360

	enum {
		RTNL_LINK_INITIALIZED,
		RTNL_LINK_INITIALIZING,
	} rtnl_link_state:16;
L
Linus Torvalds 已提交
1361

1362 1363
	/* Called from unregister, can be used to call free_netdev */
	void (*destructor)(struct net_device *dev);
L
Linus Torvalds 已提交
1364 1365

#ifdef CONFIG_NETPOLL
1366
	struct netpoll_info __rcu	*npinfo;
L
Linus Torvalds 已提交
1367
#endif
1368

1369
#ifdef CONFIG_NET_NS
1370 1371
	/* Network namespace this network device is inside */
	struct net		*nd_net;
1372
#endif
1373

D
David S. Miller 已提交
1374
	/* mid-layer private */
E
Eric Dumazet 已提交
1375 1376 1377
	union {
		void				*ml_priv;
		struct pcpu_lstats __percpu	*lstats; /* loopback stats */
1378
		struct pcpu_tstats __percpu	*tstats; /* tunnel stats */
1379
		struct pcpu_dstats __percpu	*dstats; /* dummy stats */
E
Eric Dumazet 已提交
1380
		struct pcpu_vstats __percpu	*vstats; /* veth stats */
E
Eric Dumazet 已提交
1381
	};
1382
	/* GARP */
E
Eric Dumazet 已提交
1383
	struct garp_port __rcu	*garp_port;
1384 1385
	/* MRP */
	struct mrp_port __rcu	*mrp_port;
L
Linus Torvalds 已提交
1386 1387

	/* class/net/name entry */
1388
	struct device		dev;
1389 1390
	/* space for optional device, statistics, and wireless sysfs groups */
	const struct attribute_group *sysfs_groups[4];
P
Patrick McHardy 已提交
1391 1392 1393

	/* rtnetlink link ops */
	const struct rtnl_link_ops *rtnl_link_ops;
1394

1395 1396 1397
	/* for setting kernel sock attribute on TCP connection setup */
#define GSO_MAX_SIZE		65536
	unsigned int		gso_max_size;
1398 1399
#define GSO_MAX_SEGS		65535
	u16			gso_max_segs;
1400

J
Jeff Kirsher 已提交
1401
#ifdef CONFIG_DCB
1402
	/* Data Center Bridging netlink ops */
1403
	const struct dcbnl_rtnl_ops *dcbnl_ops;
1404
#endif
1405 1406 1407
	u8 num_tc;
	struct netdev_tc_txq tc_to_txq[TC_MAX_QUEUE];
	u8 prio_tc_map[TC_BITMASK + 1];
1408

1409
#if IS_ENABLED(CONFIG_FCOE)
1410 1411
	/* max exchange id for FCoE LRO by ddp */
	unsigned int		fcoe_ddp_xid;
1412 1413 1414
#endif
#if IS_ENABLED(CONFIG_NETPRIO_CGROUP)
	struct netprio_map __rcu *priomap;
1415
#endif
1416 1417
	/* phy device may attach itself for hardware timestamping */
	struct phy_device *phydev;
1418

1419 1420
	struct lock_class_key *qdisc_tx_busylock;

1421 1422
	/* group the device belongs to */
	int group;
1423 1424

	struct pm_qos_request	pm_qos_req;
L
Linus Torvalds 已提交
1425
};
1426
#define to_net_dev(d) container_of(d, struct net_device, dev)
L
Linus Torvalds 已提交
1427 1428 1429

#define	NETDEV_ALIGN		32

1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480
static inline
int netdev_get_prio_tc_map(const struct net_device *dev, u32 prio)
{
	return dev->prio_tc_map[prio & TC_BITMASK];
}

static inline
int netdev_set_prio_tc_map(struct net_device *dev, u8 prio, u8 tc)
{
	if (tc >= dev->num_tc)
		return -EINVAL;

	dev->prio_tc_map[prio & TC_BITMASK] = tc & TC_BITMASK;
	return 0;
}

static inline
void netdev_reset_tc(struct net_device *dev)
{
	dev->num_tc = 0;
	memset(dev->tc_to_txq, 0, sizeof(dev->tc_to_txq));
	memset(dev->prio_tc_map, 0, sizeof(dev->prio_tc_map));
}

static inline
int netdev_set_tc_queue(struct net_device *dev, u8 tc, u16 count, u16 offset)
{
	if (tc >= dev->num_tc)
		return -EINVAL;

	dev->tc_to_txq[tc].count = count;
	dev->tc_to_txq[tc].offset = offset;
	return 0;
}

static inline
int netdev_set_num_tc(struct net_device *dev, u8 num_tc)
{
	if (num_tc > TC_MAX_QUEUE)
		return -EINVAL;

	dev->num_tc = num_tc;
	return 0;
}

static inline
int netdev_get_num_tc(struct net_device *dev)
{
	return dev->num_tc;
}

1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499
static inline
struct netdev_queue *netdev_get_tx_queue(const struct net_device *dev,
					 unsigned int index)
{
	return &dev->_tx[index];
}

static inline void netdev_for_each_tx_queue(struct net_device *dev,
					    void (*f)(struct net_device *,
						      struct netdev_queue *,
						      void *),
					    void *arg)
{
	unsigned int i;

	for (i = 0; i < dev->num_tx_queues; i++)
		f(dev, &dev->_tx[i], arg);
}

1500 1501
extern struct netdev_queue *netdev_pick_tx(struct net_device *dev,
					   struct sk_buff *skb);
1502
extern u16 __netdev_pick_tx(struct net_device *dev, struct sk_buff *skb);
1503

1504 1505 1506 1507 1508 1509
/*
 * Net namespace inlines
 */
static inline
struct net *dev_net(const struct net_device *dev)
{
E
Eric Dumazet 已提交
1510
	return read_pnet(&dev->nd_net);
1511 1512 1513
}

static inline
1514
void dev_net_set(struct net_device *dev, struct net *net)
1515 1516
{
#ifdef CONFIG_NET_NS
1517 1518
	release_net(dev->nd_net);
	dev->nd_net = hold_net(net);
1519 1520 1521
#endif
}

1522 1523 1524 1525 1526 1527 1528 1529 1530 1531
static inline bool netdev_uses_dsa_tags(struct net_device *dev)
{
#ifdef CONFIG_NET_DSA_TAG_DSA
	if (dev->dsa_ptr != NULL)
		return dsa_uses_dsa_tags(dev->dsa_ptr);
#endif

	return 0;
}

1532 1533 1534 1535 1536 1537 1538 1539 1540 1541
static inline bool netdev_uses_trailer_tags(struct net_device *dev)
{
#ifdef CONFIG_NET_DSA_TAG_TRAILER
	if (dev->dsa_ptr != NULL)
		return dsa_uses_trailer_tags(dev->dsa_ptr);
#endif

	return 0;
}

1542 1543 1544 1545 1546 1547
/**
 *	netdev_priv - access network device private data
 *	@dev: network device
 *
 * Get network device private data
 */
1548
static inline void *netdev_priv(const struct net_device *dev)
L
Linus Torvalds 已提交
1549
{
1550
	return (char *)dev + ALIGN(sizeof(struct net_device), NETDEV_ALIGN);
L
Linus Torvalds 已提交
1551 1552 1553 1554 1555
}

/* Set the sysfs physical device reference for the network logical device
 * if set prior to registration will cause a symlink during initialization.
 */
1556
#define SET_NETDEV_DEV(net, pdev)	((net)->dev.parent = (pdev))
L
Linus Torvalds 已提交
1557

1558 1559 1560 1561 1562 1563
/* Set the sysfs device type for the network logical device to allow
 * fin grained indentification of different network device types. For
 * example Ethernet, Wirelss LAN, Bluetooth, WiMAX etc.
 */
#define SET_NETDEV_DEVTYPE(net, devtype)	((net)->dev.type = (devtype))

E
Eric Dumazet 已提交
1564 1565 1566 1567 1568
/* Default NAPI poll() weight
 * Device drivers are strongly advised to not use bigger value
 */
#define NAPI_POLL_WEIGHT 64

1569 1570 1571 1572 1573 1574 1575 1576 1577 1578
/**
 *	netif_napi_add - initialize a napi context
 *	@dev:  network device
 *	@napi: napi context
 *	@poll: polling function
 *	@weight: default weight
 *
 * netif_napi_add() must be used to initialize a napi context prior to calling
 * *any* of the other napi related functions.
 */
1579 1580
void netif_napi_add(struct net_device *dev, struct napi_struct *napi,
		    int (*poll)(struct napi_struct *, int), int weight);
1581

1582 1583 1584 1585 1586 1587
/**
 *  netif_napi_del - remove a napi context
 *  @napi: napi context
 *
 *  netif_napi_del() removes a napi context from the network device napi list
 */
1588 1589 1590
void netif_napi_del(struct napi_struct *napi);

struct napi_gro_cb {
1591 1592 1593
	/* Virtual address of skb_shinfo(skb)->frags[0].page + offset. */
	void *frag0;

1594 1595 1596
	/* Length of frag0. */
	unsigned int frag0_len;

1597 1598 1599
	/* This indicates where we are processing relative to skb->data. */
	int data_offset;

1600 1601 1602 1603
	/* This is non-zero if the packet cannot be merged with the new skb. */
	int flush;

	/* Number of segments aggregated. */
1604 1605 1606 1607
	u16	count;

	/* This is non-zero if the packet may be of the same flow. */
	u8	same_flow;
H
Herbert Xu 已提交
1608 1609

	/* Free the skb? */
1610
	u8	free;
1611 1612
#define NAPI_GRO_FREE		  1
#define NAPI_GRO_FREE_STOLEN_HEAD 2
1613 1614 1615

	/* jiffies when first packet was created/queued */
	unsigned long age;
1616 1617 1618

	/* Used in ipv6_gro_receive() */
	int	proto;
1619 1620 1621

	/* used in skb_gro_receive() slow path */
	struct sk_buff *last;
1622 1623 1624
};

#define NAPI_GRO_CB(skb) ((struct napi_gro_cb *)(skb)->cb)
1625

L
Linus Torvalds 已提交
1626
struct packet_type {
D
David S. Miller 已提交
1627 1628 1629 1630 1631 1632
	__be16			type;	/* This is really htons(ether_type). */
	struct net_device	*dev;	/* NULL is wildcarded here	     */
	int			(*func) (struct sk_buff *,
					 struct net_device *,
					 struct packet_type *,
					 struct net_device *);
1633 1634
	bool			(*id_match)(struct packet_type *ptype,
					    struct sock *sk);
L
Linus Torvalds 已提交
1635 1636 1637 1638
	void			*af_packet_priv;
	struct list_head	list;
};

1639
struct offload_callbacks {
1640
	struct sk_buff		*(*gso_segment)(struct sk_buff *skb,
1641
						netdev_features_t features);
1642
	int			(*gso_send_check)(struct sk_buff *skb);
1643 1644 1645
	struct sk_buff		**(*gro_receive)(struct sk_buff **head,
					       struct sk_buff *skb);
	int			(*gro_complete)(struct sk_buff *skb);
1646 1647 1648 1649 1650 1651
};

struct packet_offload {
	__be16			 type;	/* This is really htons(ether_type). */
	struct offload_callbacks callbacks;
	struct list_head	 list;
L
Linus Torvalds 已提交
1652 1653 1654 1655
};

#include <linux/notifier.h>

1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678
/* netdevice notifier chain. Please remember to update the rtnetlink
 * notification exclusion list in rtnetlink_event() when adding new
 * types.
 */
#define NETDEV_UP	0x0001	/* For now you can't veto a device up/down */
#define NETDEV_DOWN	0x0002
#define NETDEV_REBOOT	0x0003	/* Tell a protocol stack a network interface
				   detected a hardware crash and restarted
				   - we can use this eg to kick tcp sessions
				   once done */
#define NETDEV_CHANGE	0x0004	/* Notify device state change */
#define NETDEV_REGISTER 0x0005
#define NETDEV_UNREGISTER	0x0006
#define NETDEV_CHANGEMTU	0x0007
#define NETDEV_CHANGEADDR	0x0008
#define NETDEV_GOING_DOWN	0x0009
#define NETDEV_CHANGENAME	0x000A
#define NETDEV_FEAT_CHANGE	0x000B
#define NETDEV_BONDING_FAILOVER 0x000C
#define NETDEV_PRE_UP		0x000D
#define NETDEV_PRE_TYPE_CHANGE	0x000E
#define NETDEV_POST_TYPE_CHANGE	0x000F
#define NETDEV_POST_INIT	0x0010
1679
#define NETDEV_UNREGISTER_FINAL 0x0011
1680 1681 1682
#define NETDEV_RELEASE		0x0012
#define NETDEV_NOTIFY_PEERS	0x0013
#define NETDEV_JOIN		0x0014
1683
#define NETDEV_CHANGEUPPER	0x0015
1684
#define NETDEV_RESEND_IGMP	0x0016
1685 1686 1687

extern int register_netdevice_notifier(struct notifier_block *nb);
extern int unregister_netdevice_notifier(struct notifier_block *nb);
1688 1689 1690 1691 1692

struct netdev_notifier_info {
	struct net_device *dev;
};

1693 1694 1695 1696 1697
struct netdev_notifier_change_info {
	struct netdev_notifier_info info; /* must be first */
	unsigned int flags_changed;
};

1698 1699 1700 1701 1702 1703
static inline void netdev_notifier_info_init(struct netdev_notifier_info *info,
					     struct net_device *dev)
{
	info->dev = dev;
}

1704 1705 1706 1707 1708 1709 1710 1711
static inline struct net_device *
netdev_notifier_info_to_dev(const struct netdev_notifier_info *info)
{
	return info->dev;
}

extern int call_netdevice_notifiers_info(unsigned long val, struct net_device *dev,
					 struct netdev_notifier_info *info);
1712 1713 1714
extern int call_netdevice_notifiers(unsigned long val, struct net_device *dev);


L
Linus Torvalds 已提交
1715 1716
extern rwlock_t				dev_base_lock;		/* Device list lock */

1717 1718
#define for_each_netdev(net, d)		\
		list_for_each_entry(d, &(net)->dev_base_head, dev_list)
1719 1720
#define for_each_netdev_reverse(net, d)	\
		list_for_each_entry_reverse(d, &(net)->dev_base_head, dev_list)
1721 1722
#define for_each_netdev_rcu(net, d)		\
		list_for_each_entry_rcu(d, &(net)->dev_base_head, dev_list)
1723 1724 1725 1726
#define for_each_netdev_safe(net, d, n)	\
		list_for_each_entry_safe(d, n, &(net)->dev_base_head, dev_list)
#define for_each_netdev_continue(net, d)		\
		list_for_each_entry_continue(d, &(net)->dev_base_head, dev_list)
1727 1728
#define for_each_netdev_continue_rcu(net, d)		\
	list_for_each_entry_continue_rcu(d, &(net)->dev_base_head, dev_list)
1729 1730 1731
#define for_each_netdev_in_bond_rcu(bond, slave)	\
		for_each_netdev_rcu(&init_net, slave)	\
			if (netdev_master_upper_dev_get_rcu(slave) == bond)
1732
#define net_device_entry(lh)	list_entry(lh, struct net_device, dev_list)
1733

1734 1735 1736 1737 1738
static inline struct net_device *next_net_device(struct net_device *dev)
{
	struct list_head *lh;
	struct net *net;

1739
	net = dev_net(dev);
1740 1741 1742 1743
	lh = dev->dev_list.next;
	return lh == &net->dev_base_head ? NULL : net_device_entry(lh);
}

1744 1745 1746 1747 1748 1749
static inline struct net_device *next_net_device_rcu(struct net_device *dev)
{
	struct list_head *lh;
	struct net *net;

	net = dev_net(dev);
E
Eric Dumazet 已提交
1750
	lh = rcu_dereference(list_next_rcu(&dev->dev_list));
1751 1752 1753
	return lh == &net->dev_base_head ? NULL : net_device_entry(lh);
}

1754 1755 1756 1757 1758
static inline struct net_device *first_net_device(struct net *net)
{
	return list_empty(&net->dev_base_head) ? NULL :
		net_device_entry(net->dev_base_head.next);
}
1759

E
Eric Dumazet 已提交
1760 1761 1762 1763 1764 1765 1766
static inline struct net_device *first_net_device_rcu(struct net *net)
{
	struct list_head *lh = rcu_dereference(list_next_rcu(&net->dev_base_head));

	return lh == &net->dev_base_head ? NULL : net_device_entry(lh);
}

L
Linus Torvalds 已提交
1767 1768
extern int 			netdev_boot_setup_check(struct net_device *dev);
extern unsigned long		netdev_boot_base(const char *prefix, int unit);
1769 1770
extern struct net_device *dev_getbyhwaddr_rcu(struct net *net, unsigned short type,
					      const char *hwaddr);
1771 1772
extern struct net_device *dev_getfirstbyhwtype(struct net *net, unsigned short type);
extern struct net_device *__dev_getfirstbyhwtype(struct net *net, unsigned short type);
L
Linus Torvalds 已提交
1773 1774 1775
extern void		dev_add_pack(struct packet_type *pt);
extern void		dev_remove_pack(struct packet_type *pt);
extern void		__dev_remove_pack(struct packet_type *pt);
1776 1777 1778
extern void		dev_add_offload(struct packet_offload *po);
extern void		dev_remove_offload(struct packet_offload *po);
extern void		__dev_remove_offload(struct packet_offload *po);
L
Linus Torvalds 已提交
1779

E
Eric Dumazet 已提交
1780 1781
extern struct net_device	*dev_get_by_flags_rcu(struct net *net, unsigned short flags,
						      unsigned short mask);
1782
extern struct net_device	*dev_get_by_name(struct net *net, const char *name);
1783
extern struct net_device	*dev_get_by_name_rcu(struct net *net, const char *name);
1784
extern struct net_device	*__dev_get_by_name(struct net *net, const char *name);
L
Linus Torvalds 已提交
1785 1786 1787
extern int		dev_alloc_name(struct net_device *dev, const char *name);
extern int		dev_open(struct net_device *dev);
extern int		dev_close(struct net_device *dev);
1788
extern void		dev_disable_lro(struct net_device *dev);
1789
extern int		dev_loopback_xmit(struct sk_buff *newskb);
L
Linus Torvalds 已提交
1790 1791
extern int		dev_queue_xmit(struct sk_buff *skb);
extern int		register_netdevice(struct net_device *dev);
1792 1793
extern void		unregister_netdevice_queue(struct net_device *dev,
						   struct list_head *head);
1794
extern void		unregister_netdevice_many(struct list_head *head);
1795 1796 1797 1798 1799
static inline void unregister_netdevice(struct net_device *dev)
{
	unregister_netdevice_queue(dev, NULL);
}

E
Eric Dumazet 已提交
1800
extern int 		netdev_refcnt_read(const struct net_device *dev);
L
Linus Torvalds 已提交
1801 1802
extern void		free_netdev(struct net_device *dev);
extern void		synchronize_net(void);
1803 1804
extern int		init_dummy_netdev(struct net_device *dev);

1805 1806
extern struct net_device	*dev_get_by_index(struct net *net, int ifindex);
extern struct net_device	*__dev_get_by_index(struct net *net, int ifindex);
1807
extern struct net_device	*dev_get_by_index_rcu(struct net *net, int ifindex);
1808
extern int		netdev_get_name(struct net *net, char *name, int ifindex);
L
Linus Torvalds 已提交
1809 1810 1811 1812
extern int		dev_restart(struct net_device *dev);
#ifdef CONFIG_NETPOLL_TRAP
extern int		netpoll_trap(void);
#endif
1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830
extern int	       skb_gro_receive(struct sk_buff **head,
				       struct sk_buff *skb);

static inline unsigned int skb_gro_offset(const struct sk_buff *skb)
{
	return NAPI_GRO_CB(skb)->data_offset;
}

static inline unsigned int skb_gro_len(const struct sk_buff *skb)
{
	return skb->len - NAPI_GRO_CB(skb)->data_offset;
}

static inline void skb_gro_pull(struct sk_buff *skb, unsigned int len)
{
	NAPI_GRO_CB(skb)->data_offset += len;
}

1831 1832
static inline void *skb_gro_header_fast(struct sk_buff *skb,
					unsigned int offset)
1833
{
1834 1835
	return NAPI_GRO_CB(skb)->frag0 + offset;
}
1836

1837 1838 1839 1840
static inline int skb_gro_header_hard(struct sk_buff *skb, unsigned int hlen)
{
	return NAPI_GRO_CB(skb)->frag0_len < hlen;
}
1841

1842 1843 1844
static inline void *skb_gro_header_slow(struct sk_buff *skb, unsigned int hlen,
					unsigned int offset)
{
1845 1846 1847
	if (!pskb_may_pull(skb, hlen))
		return NULL;

1848 1849
	NAPI_GRO_CB(skb)->frag0 = NULL;
	NAPI_GRO_CB(skb)->frag0_len = 0;
1850
	return skb->data + offset;
1851
}
L
Linus Torvalds 已提交
1852

1853 1854
static inline void *skb_gro_mac_header(struct sk_buff *skb)
{
1855
	return NAPI_GRO_CB(skb)->frag0 ?: skb_mac_header(skb);
1856 1857
}

H
Herbert Xu 已提交
1858 1859
static inline void *skb_gro_network_header(struct sk_buff *skb)
{
1860 1861
	return (NAPI_GRO_CB(skb)->frag0 ?: skb->data) +
	       skb_network_offset(skb);
H
Herbert Xu 已提交
1862 1863
}

1864 1865
static inline int dev_hard_header(struct sk_buff *skb, struct net_device *dev,
				  unsigned short type,
1866
				  const void *daddr, const void *saddr,
1867
				  unsigned int len)
1868
{
1869
	if (!dev->header_ops || !dev->header_ops->create)
1870
		return 0;
1871 1872

	return dev->header_ops->create(skb, dev, type, daddr, saddr, len);
1873 1874
}

S
Stephen Hemminger 已提交
1875 1876 1877 1878 1879
static inline int dev_parse_header(const struct sk_buff *skb,
				   unsigned char *haddr)
{
	const struct net_device *dev = skb->dev;

1880
	if (!dev->header_ops || !dev->header_ops->parse)
S
Stephen Hemminger 已提交
1881
		return 0;
1882
	return dev->header_ops->parse(skb, haddr);
S
Stephen Hemminger 已提交
1883 1884
}

L
Linus Torvalds 已提交
1885 1886 1887 1888 1889 1890 1891
typedef int gifconf_func_t(struct net_device * dev, char __user * bufptr, int len);
extern int		register_gifconf(unsigned int family, gifconf_func_t * gifconf);
static inline int unregister_gifconf(unsigned int family)
{
	return register_gifconf(family, NULL);
}

1892
#ifdef CONFIG_NET_FLOW_LIMIT
1893
#define FLOW_LIMIT_HISTORY	(1 << 7)  /* must be ^2 and !overflow buckets */
1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904
struct sd_flow_limit {
	u64			count;
	unsigned int		num_buckets;
	unsigned int		history_head;
	u16			history[FLOW_LIMIT_HISTORY];
	u8			buckets[];
};

extern int netdev_flow_limit_table_len;
#endif /* CONFIG_NET_FLOW_LIMIT */

L
Linus Torvalds 已提交
1905
/*
E
Eric Dumazet 已提交
1906
 * Incoming packets are placed on per-cpu queues
L
Linus Torvalds 已提交
1907
 */
E
Eric Dumazet 已提交
1908
struct softnet_data {
1909
	struct Qdisc		*output_queue;
1910
	struct Qdisc		**output_queue_tailp;
L
Linus Torvalds 已提交
1911 1912
	struct list_head	poll_list;
	struct sk_buff		*completion_queue;
1913
	struct sk_buff_head	process_queue;
L
Linus Torvalds 已提交
1914

C
Changli Gao 已提交
1915
	/* stats */
1916 1917 1918 1919
	unsigned int		processed;
	unsigned int		time_squeeze;
	unsigned int		cpu_collision;
	unsigned int		received_rps;
C
Changli Gao 已提交
1920

1921
#ifdef CONFIG_RPS
E
Eric Dumazet 已提交
1922 1923 1924
	struct softnet_data	*rps_ipi_list;

	/* Elements below can be accessed between CPUs for RPS */
T
Tom Herbert 已提交
1925
	struct call_single_data	csd ____cacheline_aligned_in_smp;
E
Eric Dumazet 已提交
1926 1927
	struct softnet_data	*rps_ipi_next;
	unsigned int		cpu;
T
Tom Herbert 已提交
1928
	unsigned int		input_queue_head;
1929
	unsigned int		input_queue_tail;
1930
#endif
1931
	unsigned int		dropped;
T
Tom Herbert 已提交
1932
	struct sk_buff_head	input_pkt_queue;
1933
	struct napi_struct	backlog;
1934 1935

#ifdef CONFIG_NET_FLOW_LIMIT
1936
	struct sd_flow_limit __rcu *flow_limit;
1937
#endif
L
Linus Torvalds 已提交
1938 1939
};

1940
static inline void input_queue_head_incr(struct softnet_data *sd)
T
Tom Herbert 已提交
1941 1942
{
#ifdef CONFIG_RPS
1943 1944 1945 1946 1947 1948 1949 1950 1951
	sd->input_queue_head++;
#endif
}

static inline void input_queue_tail_incr_save(struct softnet_data *sd,
					      unsigned int *qtail)
{
#ifdef CONFIG_RPS
	*qtail = ++sd->input_queue_tail;
T
Tom Herbert 已提交
1952 1953 1954
#endif
}

T
Tom Herbert 已提交
1955
DECLARE_PER_CPU_ALIGNED(struct softnet_data, softnet_data);
L
Linus Torvalds 已提交
1956

1957
extern void __netif_schedule(struct Qdisc *q);
L
Linus Torvalds 已提交
1958

1959
static inline void netif_schedule_queue(struct netdev_queue *txq)
L
Linus Torvalds 已提交
1960
{
1961
	if (!(txq->state & QUEUE_STATE_ANY_XOFF))
1962
		__netif_schedule(txq->qdisc);
1963 1964
}

1965 1966 1967 1968 1969 1970 1971 1972
static inline void netif_tx_schedule_all(struct net_device *dev)
{
	unsigned int i;

	for (i = 0; i < dev->num_tx_queues; i++)
		netif_schedule_queue(netdev_get_tx_queue(dev, i));
}

1973 1974
static inline void netif_tx_start_queue(struct netdev_queue *dev_queue)
{
1975
	clear_bit(__QUEUE_STATE_DRV_XOFF, &dev_queue->state);
1976 1977
}

1978 1979 1980 1981 1982 1983
/**
 *	netif_start_queue - allow transmit
 *	@dev: network device
 *
 *	Allow upper layers to call the device hard_start_xmit routine.
 */
L
Linus Torvalds 已提交
1984 1985
static inline void netif_start_queue(struct net_device *dev)
{
1986
	netif_tx_start_queue(netdev_get_tx_queue(dev, 0));
L
Linus Torvalds 已提交
1987 1988
}

1989 1990 1991 1992 1993 1994 1995 1996 1997 1998
static inline void netif_tx_start_all_queues(struct net_device *dev)
{
	unsigned int i;

	for (i = 0; i < dev->num_tx_queues; i++) {
		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
		netif_tx_start_queue(txq);
	}
}

1999
static inline void netif_tx_wake_queue(struct netdev_queue *dev_queue)
L
Linus Torvalds 已提交
2000 2001
{
#ifdef CONFIG_NETPOLL_TRAP
2002
	if (netpoll_trap()) {
2003
		netif_tx_start_queue(dev_queue);
L
Linus Torvalds 已提交
2004
		return;
2005
	}
L
Linus Torvalds 已提交
2006
#endif
2007
	if (test_and_clear_bit(__QUEUE_STATE_DRV_XOFF, &dev_queue->state))
2008
		__netif_schedule(dev_queue->qdisc);
2009 2010
}

2011 2012 2013 2014 2015 2016 2017
/**
 *	netif_wake_queue - restart transmit
 *	@dev: network device
 *
 *	Allow upper layers to call the device hard_start_xmit routine.
 *	Used for flow control when transmit resources are available.
 */
2018 2019
static inline void netif_wake_queue(struct net_device *dev)
{
2020
	netif_tx_wake_queue(netdev_get_tx_queue(dev, 0));
L
Linus Torvalds 已提交
2021 2022
}

2023 2024 2025 2026 2027 2028 2029 2030 2031 2032
static inline void netif_tx_wake_all_queues(struct net_device *dev)
{
	unsigned int i;

	for (i = 0; i < dev->num_tx_queues; i++) {
		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
		netif_tx_wake_queue(txq);
	}
}

2033 2034
static inline void netif_tx_stop_queue(struct netdev_queue *dev_queue)
{
2035
	if (WARN_ON(!dev_queue)) {
2036
		pr_info("netif_stop_queue() cannot be called before register_netdev()\n");
2037 2038
		return;
	}
2039
	set_bit(__QUEUE_STATE_DRV_XOFF, &dev_queue->state);
2040 2041
}

2042 2043 2044 2045 2046 2047 2048
/**
 *	netif_stop_queue - stop transmitted packets
 *	@dev: network device
 *
 *	Stop upper layers calling the device hard_start_xmit routine.
 *	Used for flow control when transmit resources are unavailable.
 */
L
Linus Torvalds 已提交
2049 2050
static inline void netif_stop_queue(struct net_device *dev)
{
2051
	netif_tx_stop_queue(netdev_get_tx_queue(dev, 0));
L
Linus Torvalds 已提交
2052 2053
}

2054 2055 2056 2057 2058 2059 2060 2061 2062 2063
static inline void netif_tx_stop_all_queues(struct net_device *dev)
{
	unsigned int i;

	for (i = 0; i < dev->num_tx_queues; i++) {
		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
		netif_tx_stop_queue(txq);
	}
}

2064
static inline bool netif_tx_queue_stopped(const struct netdev_queue *dev_queue)
2065
{
2066
	return test_bit(__QUEUE_STATE_DRV_XOFF, &dev_queue->state);
2067 2068
}

2069 2070 2071 2072 2073 2074
/**
 *	netif_queue_stopped - test if transmit queue is flowblocked
 *	@dev: network device
 *
 *	Test if transmit queue on device is currently unable to send.
 */
2075
static inline bool netif_queue_stopped(const struct net_device *dev)
L
Linus Torvalds 已提交
2076
{
2077
	return netif_tx_queue_stopped(netdev_get_tx_queue(dev, 0));
L
Linus Torvalds 已提交
2078 2079
}

2080
static inline bool netif_xmit_stopped(const struct netdev_queue *dev_queue)
2081
{
2082 2083 2084
	return dev_queue->state & QUEUE_STATE_ANY_XOFF;
}

2085
static inline bool netif_xmit_frozen_or_stopped(const struct netdev_queue *dev_queue)
2086 2087 2088 2089
{
	return dev_queue->state & QUEUE_STATE_ANY_XOFF_OR_FROZEN;
}

2090 2091 2092
static inline void netdev_tx_sent_queue(struct netdev_queue *dev_queue,
					unsigned int bytes)
{
T
Tom Herbert 已提交
2093 2094
#ifdef CONFIG_BQL
	dql_queued(&dev_queue->dql, bytes);
2095 2096 2097 2098 2099 2100 2101 2102 2103 2104 2105 2106 2107 2108 2109 2110

	if (likely(dql_avail(&dev_queue->dql) >= 0))
		return;

	set_bit(__QUEUE_STATE_STACK_XOFF, &dev_queue->state);

	/*
	 * The XOFF flag must be set before checking the dql_avail below,
	 * because in netdev_tx_completed_queue we update the dql_completed
	 * before checking the XOFF flag.
	 */
	smp_mb();

	/* check again in case another CPU has just made room avail */
	if (unlikely(dql_avail(&dev_queue->dql) >= 0))
		clear_bit(__QUEUE_STATE_STACK_XOFF, &dev_queue->state);
T
Tom Herbert 已提交
2111
#endif
2112 2113
}

2114 2115 2116 2117 2118 2119 2120 2121 2122
/**
 * 	netdev_sent_queue - report the number of bytes queued to hardware
 * 	@dev: network device
 * 	@bytes: number of bytes queued to the hardware device queue
 *
 * 	Report the number of bytes queued for sending/completion to the network
 * 	device hardware queue. @bytes should be a good approximation and should
 * 	exactly match netdev_completed_queue() @bytes
 */
2123 2124 2125 2126 2127 2128
static inline void netdev_sent_queue(struct net_device *dev, unsigned int bytes)
{
	netdev_tx_sent_queue(netdev_get_tx_queue(dev, 0), bytes);
}

static inline void netdev_tx_completed_queue(struct netdev_queue *dev_queue,
2129
					     unsigned int pkts, unsigned int bytes)
2130
{
T
Tom Herbert 已提交
2131
#ifdef CONFIG_BQL
2132 2133 2134 2135 2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146 2147 2148
	if (unlikely(!bytes))
		return;

	dql_completed(&dev_queue->dql, bytes);

	/*
	 * Without the memory barrier there is a small possiblity that
	 * netdev_tx_sent_queue will miss the update and cause the queue to
	 * be stopped forever
	 */
	smp_mb();

	if (dql_avail(&dev_queue->dql) < 0)
		return;

	if (test_and_clear_bit(__QUEUE_STATE_STACK_XOFF, &dev_queue->state))
		netif_schedule_queue(dev_queue);
T
Tom Herbert 已提交
2149
#endif
2150 2151
}

2152 2153 2154 2155 2156 2157 2158 2159 2160 2161
/**
 * 	netdev_completed_queue - report bytes and packets completed by device
 * 	@dev: network device
 * 	@pkts: actual number of packets sent over the medium
 * 	@bytes: actual number of bytes sent over the medium
 *
 * 	Report the number of bytes and packets transmitted by the network device
 * 	hardware queue over the physical medium, @bytes must exactly match the
 * 	@bytes amount passed to netdev_sent_queue()
 */
2162
static inline void netdev_completed_queue(struct net_device *dev,
2163
					  unsigned int pkts, unsigned int bytes)
2164 2165 2166 2167 2168 2169
{
	netdev_tx_completed_queue(netdev_get_tx_queue(dev, 0), pkts, bytes);
}

static inline void netdev_tx_reset_queue(struct netdev_queue *q)
{
T
Tom Herbert 已提交
2170
#ifdef CONFIG_BQL
2171
	clear_bit(__QUEUE_STATE_STACK_XOFF, &q->state);
T
Tom Herbert 已提交
2172 2173
	dql_reset(&q->dql);
#endif
2174 2175
}

2176 2177 2178 2179 2180 2181 2182
/**
 * 	netdev_reset_queue - reset the packets and bytes count of a network device
 * 	@dev_queue: network device
 *
 * 	Reset the bytes and packet count of a network device and clear the
 * 	software flow control OFF bit for this network device
 */
2183 2184 2185
static inline void netdev_reset_queue(struct net_device *dev_queue)
{
	netdev_tx_reset_queue(netdev_get_tx_queue(dev_queue, 0));
2186 2187
}

2188 2189 2190 2191 2192 2193
/**
 *	netif_running - test if up
 *	@dev: network device
 *
 *	Test if the device has been brought up.
 */
2194
static inline bool netif_running(const struct net_device *dev)
L
Linus Torvalds 已提交
2195 2196 2197 2198
{
	return test_bit(__LINK_STATE_START, &dev->state);
}

2199 2200 2201 2202 2203 2204
/*
 * Routines to manage the subqueues on a device.  We only need start
 * stop, and a check if it's stopped.  All other device management is
 * done at the overall netdevice level.
 * Also test the device if we're multiqueue.
 */
2205 2206 2207 2208 2209 2210 2211 2212

/**
 *	netif_start_subqueue - allow sending packets on subqueue
 *	@dev: network device
 *	@queue_index: sub queue index
 *
 * Start individual transmit queue of a device with multiple transmit queues.
 */
2213 2214
static inline void netif_start_subqueue(struct net_device *dev, u16 queue_index)
{
2215
	struct netdev_queue *txq = netdev_get_tx_queue(dev, queue_index);
2216 2217

	netif_tx_start_queue(txq);
2218 2219
}

2220 2221 2222 2223 2224 2225 2226
/**
 *	netif_stop_subqueue - stop sending packets on subqueue
 *	@dev: network device
 *	@queue_index: sub queue index
 *
 * Stop individual transmit queue of a device with multiple transmit queues.
 */
2227 2228
static inline void netif_stop_subqueue(struct net_device *dev, u16 queue_index)
{
2229
	struct netdev_queue *txq = netdev_get_tx_queue(dev, queue_index);
2230 2231 2232 2233
#ifdef CONFIG_NETPOLL_TRAP
	if (netpoll_trap())
		return;
#endif
2234
	netif_tx_stop_queue(txq);
2235 2236
}

2237 2238 2239 2240 2241 2242 2243
/**
 *	netif_subqueue_stopped - test status of subqueue
 *	@dev: network device
 *	@queue_index: sub queue index
 *
 * Check individual transmit queue of a device with multiple transmit queues.
 */
2244 2245
static inline bool __netif_subqueue_stopped(const struct net_device *dev,
					    u16 queue_index)
2246
{
2247
	struct netdev_queue *txq = netdev_get_tx_queue(dev, queue_index);
2248 2249

	return netif_tx_queue_stopped(txq);
2250 2251
}

2252 2253
static inline bool netif_subqueue_stopped(const struct net_device *dev,
					  struct sk_buff *skb)
2254 2255 2256
{
	return __netif_subqueue_stopped(dev, skb_get_queue_mapping(skb));
}
2257 2258 2259 2260 2261 2262 2263 2264

/**
 *	netif_wake_subqueue - allow sending packets on subqueue
 *	@dev: network device
 *	@queue_index: sub queue index
 *
 * Resume individual transmit queue of a device with multiple transmit queues.
 */
2265 2266
static inline void netif_wake_subqueue(struct net_device *dev, u16 queue_index)
{
2267
	struct netdev_queue *txq = netdev_get_tx_queue(dev, queue_index);
2268 2269 2270 2271
#ifdef CONFIG_NETPOLL_TRAP
	if (netpoll_trap())
		return;
#endif
2272
	if (test_and_clear_bit(__QUEUE_STATE_DRV_XOFF, &txq->state))
2273
		__netif_schedule(txq->qdisc);
2274 2275
}

2276 2277 2278 2279 2280 2281 2282 2283 2284 2285 2286 2287
#ifdef CONFIG_XPS
extern int netif_set_xps_queue(struct net_device *dev, struct cpumask *mask,
			       u16 index);
#else
static inline int netif_set_xps_queue(struct net_device *dev,
				      struct cpumask *mask,
				      u16 index)
{
	return 0;
}
#endif

2288 2289 2290 2291 2292 2293 2294 2295 2296 2297
/*
 * Returns a Tx hash for the given packet when dev->real_num_tx_queues is used
 * as a distribution range limit for the returned value.
 */
static inline u16 skb_tx_hash(const struct net_device *dev,
			      const struct sk_buff *skb)
{
	return __skb_tx_hash(dev, skb, dev->real_num_tx_queues);
}

2298 2299 2300 2301 2302 2303
/**
 *	netif_is_multiqueue - test if device has multiple transmit queues
 *	@dev: network device
 *
 * Check if device has multiple transmit queues
 */
2304
static inline bool netif_is_multiqueue(const struct net_device *dev)
2305
{
E
Eric Dumazet 已提交
2306
	return dev->num_tx_queues > 1;
2307
}
L
Linus Torvalds 已提交
2308

2309 2310
extern int netif_set_real_num_tx_queues(struct net_device *dev,
					unsigned int txq);
2311

2312 2313 2314 2315 2316 2317 2318 2319 2320 2321 2322
#ifdef CONFIG_RPS
extern int netif_set_real_num_rx_queues(struct net_device *dev,
					unsigned int rxq);
#else
static inline int netif_set_real_num_rx_queues(struct net_device *dev,
						unsigned int rxq)
{
	return 0;
}
#endif

2323 2324 2325
static inline int netif_copy_real_num_queues(struct net_device *to_dev,
					     const struct net_device *from_dev)
{
2326 2327 2328 2329 2330 2331
	int err;

	err = netif_set_real_num_tx_queues(to_dev,
					   from_dev->real_num_tx_queues);
	if (err)
		return err;
2332 2333 2334 2335 2336 2337 2338 2339
#ifdef CONFIG_RPS
	return netif_set_real_num_rx_queues(to_dev,
					    from_dev->real_num_rx_queues);
#else
	return 0;
#endif
}

2340 2341 2342
#define DEFAULT_MAX_NUM_RSS_QUEUES	(8)
extern int netif_get_num_default_rss_queues(void);

L
Linus Torvalds 已提交
2343
/* Use this variant when it is known for sure that it
2344 2345
 * is executing from hardware interrupt context or with hardware interrupts
 * disabled.
L
Linus Torvalds 已提交
2346
 */
2347
extern void dev_kfree_skb_irq(struct sk_buff *skb);
L
Linus Torvalds 已提交
2348 2349

/* Use this variant in places where it could be invoked
2350 2351
 * from either hardware interrupt or other context, with hardware interrupts
 * either disabled or enabled.
L
Linus Torvalds 已提交
2352
 */
2353
extern void dev_kfree_skb_any(struct sk_buff *skb);
L
Linus Torvalds 已提交
2354 2355 2356 2357

extern int		netif_rx(struct sk_buff *skb);
extern int		netif_rx_ni(struct sk_buff *skb);
extern int		netif_receive_skb(struct sk_buff *skb);
2358
extern gro_result_t	napi_gro_receive(struct napi_struct *napi,
2359
					 struct sk_buff *skb);
2360
extern void		napi_gro_flush(struct napi_struct *napi, bool flush_old);
2361
extern struct sk_buff *	napi_get_frags(struct napi_struct *napi);
2362
extern gro_result_t	napi_gro_frags(struct napi_struct *napi);
2363 2364 2365 2366 2367 2368 2369

static inline void napi_free_frags(struct napi_struct *napi)
{
	kfree_skb(napi->skb);
	napi->skb = NULL;
}

2370
extern int netdev_rx_handler_register(struct net_device *dev,
J
Jiri Pirko 已提交
2371 2372
				      rx_handler_func_t *rx_handler,
				      void *rx_handler_data);
2373 2374
extern void netdev_rx_handler_unregister(struct net_device *dev);

2375
extern bool		dev_valid_name(const char *name);
2376 2377
extern int		dev_ioctl(struct net *net, unsigned int cmd, void __user *);
extern int		dev_ethtool(struct net *net, struct ifreq *);
2378
extern unsigned int	dev_get_flags(const struct net_device *);
2379
extern int		__dev_change_flags(struct net_device *, unsigned int flags);
2380
extern int		dev_change_flags(struct net_device *, unsigned int);
2381
extern void		__dev_notify_flags(struct net_device *, unsigned int old_flags);
2382
extern int		dev_change_name(struct net_device *, const char *);
2383
extern int		dev_set_alias(struct net_device *, const char *, size_t);
2384 2385
extern int		dev_change_net_namespace(struct net_device *,
						 struct net *, const char *);
L
Linus Torvalds 已提交
2386
extern int		dev_set_mtu(struct net_device *, int);
2387
extern void		dev_set_group(struct net_device *, int);
L
Linus Torvalds 已提交
2388 2389
extern int		dev_set_mac_address(struct net_device *,
					    struct sockaddr *);
J
Jiri Pirko 已提交
2390 2391
extern int		dev_change_carrier(struct net_device *,
					   bool new_carrier);
2392 2393
extern int		dev_get_phys_port_id(struct net_device *dev,
					     struct netdev_phys_port_id *ppid);
2394
extern int		dev_hard_start_xmit(struct sk_buff *skb,
2395 2396
					    struct net_device *dev,
					    struct netdev_queue *txq);
2397 2398
extern int		dev_forward_skb(struct net_device *dev,
					struct sk_buff *skb);
L
Linus Torvalds 已提交
2399

2400
extern int		netdev_budget;
L
Linus Torvalds 已提交
2401 2402 2403 2404

/* Called by rtnetlink.c:rtnl_unlock() */
extern void netdev_run_todo(void);

2405 2406 2407 2408
/**
 *	dev_put - release reference to device
 *	@dev: network device
 *
2409
 * Release reference to device to allow it to be freed.
2410
 */
L
Linus Torvalds 已提交
2411 2412
static inline void dev_put(struct net_device *dev)
{
2413
	this_cpu_dec(*dev->pcpu_refcnt);
L
Linus Torvalds 已提交
2414 2415
}

2416 2417 2418 2419
/**
 *	dev_hold - get reference to device
 *	@dev: network device
 *
2420
 * Hold reference to device to keep it from being freed.
2421
 */
2422 2423
static inline void dev_hold(struct net_device *dev)
{
2424
	this_cpu_inc(*dev->pcpu_refcnt);
2425
}
L
Linus Torvalds 已提交
2426 2427 2428 2429

/* Carrier loss detection, dial on demand. The functions netif_carrier_on
 * and _off may be called from IRQ context, but it is caller
 * who is responsible for serialization of these calls.
S
Stefan Rompf 已提交
2430 2431 2432 2433
 *
 * The name carrier is inappropriate, these functions should really be
 * called netif_lowerlayer_*() because they represent the state of any
 * kind of lower layer not just hardware media.
L
Linus Torvalds 已提交
2434 2435
 */

2436
extern void linkwatch_init_dev(struct net_device *dev);
L
Linus Torvalds 已提交
2437
extern void linkwatch_fire_event(struct net_device *dev);
2438
extern void linkwatch_forget_dev(struct net_device *dev);
L
Linus Torvalds 已提交
2439

2440 2441 2442 2443 2444 2445
/**
 *	netif_carrier_ok - test if carrier present
 *	@dev: network device
 *
 * Check if carrier is present on device
 */
2446
static inline bool netif_carrier_ok(const struct net_device *dev)
L
Linus Torvalds 已提交
2447 2448 2449 2450
{
	return !test_bit(__LINK_STATE_NOCARRIER, &dev->state);
}

2451 2452
extern unsigned long dev_trans_start(struct net_device *dev);

L
Linus Torvalds 已提交
2453 2454
extern void __netdev_watchdog_up(struct net_device *dev);

2455
extern void netif_carrier_on(struct net_device *dev);
L
Linus Torvalds 已提交
2456

2457
extern void netif_carrier_off(struct net_device *dev);
L
Linus Torvalds 已提交
2458

2459 2460 2461 2462 2463 2464 2465 2466 2467 2468 2469 2470 2471
/**
 *	netif_dormant_on - mark device as dormant.
 *	@dev: network device
 *
 * Mark device as dormant (as per RFC2863).
 *
 * The dormant state indicates that the relevant interface is not
 * actually in a condition to pass packets (i.e., it is not 'up') but is
 * in a "pending" state, waiting for some external event.  For "on-
 * demand" interfaces, this new state identifies the situation where the
 * interface is waiting for events to place it in the up state.
 *
 */
S
Stefan Rompf 已提交
2472 2473 2474 2475 2476 2477
static inline void netif_dormant_on(struct net_device *dev)
{
	if (!test_and_set_bit(__LINK_STATE_DORMANT, &dev->state))
		linkwatch_fire_event(dev);
}

2478 2479 2480 2481 2482 2483
/**
 *	netif_dormant_off - set device as not dormant.
 *	@dev: network device
 *
 * Device is not in dormant state.
 */
S
Stefan Rompf 已提交
2484 2485 2486 2487 2488 2489
static inline void netif_dormant_off(struct net_device *dev)
{
	if (test_and_clear_bit(__LINK_STATE_DORMANT, &dev->state))
		linkwatch_fire_event(dev);
}

2490 2491 2492 2493 2494 2495
/**
 *	netif_dormant - test if carrier present
 *	@dev: network device
 *
 * Check if carrier is present on device
 */
2496
static inline bool netif_dormant(const struct net_device *dev)
S
Stefan Rompf 已提交
2497 2498 2499 2500 2501
{
	return test_bit(__LINK_STATE_DORMANT, &dev->state);
}


2502 2503 2504 2505 2506 2507
/**
 *	netif_oper_up - test if device is operational
 *	@dev: network device
 *
 * Check if carrier is operational
 */
2508
static inline bool netif_oper_up(const struct net_device *dev)
E
Eric Dumazet 已提交
2509
{
S
Stefan Rompf 已提交
2510 2511 2512 2513
	return (dev->operstate == IF_OPER_UP ||
		dev->operstate == IF_OPER_UNKNOWN /* backward compat */);
}

2514 2515 2516 2517 2518 2519
/**
 *	netif_device_present - is device available or removed
 *	@dev: network device
 *
 * Check if device has not been removed from system.
 */
2520
static inline bool netif_device_present(struct net_device *dev)
L
Linus Torvalds 已提交
2521 2522 2523 2524
{
	return test_bit(__LINK_STATE_PRESENT, &dev->state);
}

2525
extern void netif_device_detach(struct net_device *dev);
L
Linus Torvalds 已提交
2526

2527
extern void netif_device_attach(struct net_device *dev);
L
Linus Torvalds 已提交
2528 2529 2530 2531 2532 2533 2534 2535 2536 2537 2538 2539 2540 2541 2542 2543 2544 2545 2546 2547 2548 2549 2550 2551 2552 2553 2554 2555 2556 2557 2558 2559 2560 2561 2562 2563 2564 2565 2566 2567 2568 2569 2570 2571 2572 2573 2574 2575 2576 2577

/*
 * Network interface message level settings
 */

enum {
	NETIF_MSG_DRV		= 0x0001,
	NETIF_MSG_PROBE		= 0x0002,
	NETIF_MSG_LINK		= 0x0004,
	NETIF_MSG_TIMER		= 0x0008,
	NETIF_MSG_IFDOWN	= 0x0010,
	NETIF_MSG_IFUP		= 0x0020,
	NETIF_MSG_RX_ERR	= 0x0040,
	NETIF_MSG_TX_ERR	= 0x0080,
	NETIF_MSG_TX_QUEUED	= 0x0100,
	NETIF_MSG_INTR		= 0x0200,
	NETIF_MSG_TX_DONE	= 0x0400,
	NETIF_MSG_RX_STATUS	= 0x0800,
	NETIF_MSG_PKTDATA	= 0x1000,
	NETIF_MSG_HW		= 0x2000,
	NETIF_MSG_WOL		= 0x4000,
};

#define netif_msg_drv(p)	((p)->msg_enable & NETIF_MSG_DRV)
#define netif_msg_probe(p)	((p)->msg_enable & NETIF_MSG_PROBE)
#define netif_msg_link(p)	((p)->msg_enable & NETIF_MSG_LINK)
#define netif_msg_timer(p)	((p)->msg_enable & NETIF_MSG_TIMER)
#define netif_msg_ifdown(p)	((p)->msg_enable & NETIF_MSG_IFDOWN)
#define netif_msg_ifup(p)	((p)->msg_enable & NETIF_MSG_IFUP)
#define netif_msg_rx_err(p)	((p)->msg_enable & NETIF_MSG_RX_ERR)
#define netif_msg_tx_err(p)	((p)->msg_enable & NETIF_MSG_TX_ERR)
#define netif_msg_tx_queued(p)	((p)->msg_enable & NETIF_MSG_TX_QUEUED)
#define netif_msg_intr(p)	((p)->msg_enable & NETIF_MSG_INTR)
#define netif_msg_tx_done(p)	((p)->msg_enable & NETIF_MSG_TX_DONE)
#define netif_msg_rx_status(p)	((p)->msg_enable & NETIF_MSG_RX_STATUS)
#define netif_msg_pktdata(p)	((p)->msg_enable & NETIF_MSG_PKTDATA)
#define netif_msg_hw(p)		((p)->msg_enable & NETIF_MSG_HW)
#define netif_msg_wol(p)	((p)->msg_enable & NETIF_MSG_WOL)

static inline u32 netif_msg_init(int debug_value, int default_msg_enable_bits)
{
	/* use default */
	if (debug_value < 0 || debug_value >= (sizeof(u32) * 8))
		return default_msg_enable_bits;
	if (debug_value == 0)	/* no output */
		return 0;
	/* set low N bits */
	return (1 << debug_value) - 1;
}

2578
static inline void __netif_tx_lock(struct netdev_queue *txq, int cpu)
H
Herbert Xu 已提交
2579
{
2580 2581
	spin_lock(&txq->_xmit_lock);
	txq->xmit_lock_owner = cpu;
2582 2583
}

2584 2585 2586 2587 2588 2589
static inline void __netif_tx_lock_bh(struct netdev_queue *txq)
{
	spin_lock_bh(&txq->_xmit_lock);
	txq->xmit_lock_owner = smp_processor_id();
}

2590
static inline bool __netif_tx_trylock(struct netdev_queue *txq)
2591
{
2592
	bool ok = spin_trylock(&txq->_xmit_lock);
2593 2594 2595 2596 2597 2598 2599 2600 2601 2602 2603 2604 2605 2606 2607 2608 2609
	if (likely(ok))
		txq->xmit_lock_owner = smp_processor_id();
	return ok;
}

static inline void __netif_tx_unlock(struct netdev_queue *txq)
{
	txq->xmit_lock_owner = -1;
	spin_unlock(&txq->_xmit_lock);
}

static inline void __netif_tx_unlock_bh(struct netdev_queue *txq)
{
	txq->xmit_lock_owner = -1;
	spin_unlock_bh(&txq->_xmit_lock);
}

E
Eric Dumazet 已提交
2610 2611 2612 2613 2614 2615
static inline void txq_trans_update(struct netdev_queue *txq)
{
	if (txq->xmit_lock_owner != -1)
		txq->trans_start = jiffies;
}

2616 2617 2618 2619 2620 2621
/**
 *	netif_tx_lock - grab network device transmit lock
 *	@dev: network device
 *
 * Get network device transmit lock
 */
2622 2623
static inline void netif_tx_lock(struct net_device *dev)
{
2624
	unsigned int i;
2625
	int cpu;
2626

2627 2628
	spin_lock(&dev->tx_global_lock);
	cpu = smp_processor_id();
2629 2630
	for (i = 0; i < dev->num_tx_queues; i++) {
		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
2631 2632 2633 2634 2635 2636 2637

		/* We are the only thread of execution doing a
		 * freeze, but we have to grab the _xmit_lock in
		 * order to synchronize with threads which are in
		 * the ->hard_start_xmit() handler and already
		 * checked the frozen bit.
		 */
2638
		__netif_tx_lock(txq, cpu);
2639 2640
		set_bit(__QUEUE_STATE_FROZEN, &txq->state);
		__netif_tx_unlock(txq);
2641
	}
H
Herbert Xu 已提交
2642 2643 2644 2645
}

static inline void netif_tx_lock_bh(struct net_device *dev)
{
2646 2647
	local_bh_disable();
	netif_tx_lock(dev);
H
Herbert Xu 已提交
2648 2649 2650 2651
}

static inline void netif_tx_unlock(struct net_device *dev)
{
2652 2653 2654 2655
	unsigned int i;

	for (i = 0; i < dev->num_tx_queues; i++) {
		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
2656

2657 2658 2659 2660 2661
		/* No need to grab the _xmit_lock here.  If the
		 * queue is not stopped for another reason, we
		 * force a schedule.
		 */
		clear_bit(__QUEUE_STATE_FROZEN, &txq->state);
2662
		netif_schedule_queue(txq);
2663 2664
	}
	spin_unlock(&dev->tx_global_lock);
H
Herbert Xu 已提交
2665 2666 2667 2668
}

static inline void netif_tx_unlock_bh(struct net_device *dev)
{
2669 2670
	netif_tx_unlock(dev);
	local_bh_enable();
H
Herbert Xu 已提交
2671 2672
}

2673
#define HARD_TX_LOCK(dev, txq, cpu) {			\
2674
	if ((dev->features & NETIF_F_LLTX) == 0) {	\
2675
		__netif_tx_lock(txq, cpu);		\
2676 2677 2678
	}						\
}

2679
#define HARD_TX_UNLOCK(dev, txq) {			\
2680
	if ((dev->features & NETIF_F_LLTX) == 0) {	\
2681
		__netif_tx_unlock(txq);			\
2682 2683 2684
	}						\
}

L
Linus Torvalds 已提交
2685 2686
static inline void netif_tx_disable(struct net_device *dev)
{
2687
	unsigned int i;
2688
	int cpu;
2689

2690 2691
	local_bh_disable();
	cpu = smp_processor_id();
2692 2693
	for (i = 0; i < dev->num_tx_queues; i++) {
		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
2694 2695

		__netif_tx_lock(txq, cpu);
2696
		netif_tx_stop_queue(txq);
2697
		__netif_tx_unlock(txq);
2698
	}
2699
	local_bh_enable();
L
Linus Torvalds 已提交
2700 2701
}

2702 2703 2704 2705 2706
static inline void netif_addr_lock(struct net_device *dev)
{
	spin_lock(&dev->addr_list_lock);
}

2707 2708 2709 2710 2711
static inline void netif_addr_lock_nested(struct net_device *dev)
{
	spin_lock_nested(&dev->addr_list_lock, SINGLE_DEPTH_NESTING);
}

2712 2713 2714 2715 2716 2717 2718 2719 2720 2721 2722 2723 2724 2725 2726
static inline void netif_addr_lock_bh(struct net_device *dev)
{
	spin_lock_bh(&dev->addr_list_lock);
}

static inline void netif_addr_unlock(struct net_device *dev)
{
	spin_unlock(&dev->addr_list_lock);
}

static inline void netif_addr_unlock_bh(struct net_device *dev)
{
	spin_unlock_bh(&dev->addr_list_lock);
}

2727
/*
2728
 * dev_addrs walker. Should be used only for read access. Call with
2729 2730 2731
 * rcu_read_lock held.
 */
#define for_each_dev_addr(dev, ha) \
2732
		list_for_each_entry_rcu(ha, &dev->dev_addrs.list, list)
2733

L
Linus Torvalds 已提交
2734 2735 2736 2737 2738
/* These functions live elsewhere (drivers/net/net_init.c, but related) */

extern void		ether_setup(struct net_device *dev);

/* Support for loadable net-drivers */
T
Tom Herbert 已提交
2739
extern struct net_device *alloc_netdev_mqs(int sizeof_priv, const char *name,
2740
				       void (*setup)(struct net_device *),
T
Tom Herbert 已提交
2741
				       unsigned int txqs, unsigned int rxqs);
2742
#define alloc_netdev(sizeof_priv, name, setup) \
T
Tom Herbert 已提交
2743 2744 2745 2746 2747
	alloc_netdev_mqs(sizeof_priv, name, setup, 1, 1)

#define alloc_netdev_mq(sizeof_priv, name, setup, count) \
	alloc_netdev_mqs(sizeof_priv, name, setup, count, count)

L
Linus Torvalds 已提交
2748 2749
extern int		register_netdev(struct net_device *dev);
extern void		unregister_netdev(struct net_device *dev);
2750

2751 2752 2753 2754 2755 2756 2757 2758 2759 2760 2761 2762 2763 2764 2765 2766
/* General hardware address lists handling functions */
extern int __hw_addr_add_multiple(struct netdev_hw_addr_list *to_list,
				  struct netdev_hw_addr_list *from_list,
				  int addr_len, unsigned char addr_type);
extern void __hw_addr_del_multiple(struct netdev_hw_addr_list *to_list,
				   struct netdev_hw_addr_list *from_list,
				   int addr_len, unsigned char addr_type);
extern int __hw_addr_sync(struct netdev_hw_addr_list *to_list,
			  struct netdev_hw_addr_list *from_list,
			  int addr_len);
extern void __hw_addr_unsync(struct netdev_hw_addr_list *to_list,
			     struct netdev_hw_addr_list *from_list,
			     int addr_len);
extern void __hw_addr_flush(struct netdev_hw_addr_list *list);
extern void __hw_addr_init(struct netdev_hw_addr_list *list);

2767
/* Functions used for device addresses handling */
2768
extern int dev_addr_add(struct net_device *dev, const unsigned char *addr,
2769
			unsigned char addr_type);
2770
extern int dev_addr_del(struct net_device *dev, const unsigned char *addr,
2771 2772 2773 2774 2775 2776 2777
			unsigned char addr_type);
extern int dev_addr_add_multiple(struct net_device *to_dev,
				 struct net_device *from_dev,
				 unsigned char addr_type);
extern int dev_addr_del_multiple(struct net_device *to_dev,
				 struct net_device *from_dev,
				 unsigned char addr_type);
2778 2779 2780 2781
extern void dev_addr_flush(struct net_device *dev);
extern int dev_addr_init(struct net_device *dev);

/* Functions used for unicast addresses handling */
2782 2783 2784
extern int dev_uc_add(struct net_device *dev, const unsigned char *addr);
extern int dev_uc_add_excl(struct net_device *dev, const unsigned char *addr);
extern int dev_uc_del(struct net_device *dev, const unsigned char *addr);
2785
extern int dev_uc_sync(struct net_device *to, struct net_device *from);
2786
extern int dev_uc_sync_multiple(struct net_device *to, struct net_device *from);
2787 2788 2789
extern void dev_uc_unsync(struct net_device *to, struct net_device *from);
extern void dev_uc_flush(struct net_device *dev);
extern void dev_uc_init(struct net_device *dev);
2790

2791
/* Functions used for multicast addresses handling */
2792 2793 2794 2795 2796
extern int dev_mc_add(struct net_device *dev, const unsigned char *addr);
extern int dev_mc_add_global(struct net_device *dev, const unsigned char *addr);
extern int dev_mc_add_excl(struct net_device *dev, const unsigned char *addr);
extern int dev_mc_del(struct net_device *dev, const unsigned char *addr);
extern int dev_mc_del_global(struct net_device *dev, const unsigned char *addr);
2797
extern int dev_mc_sync(struct net_device *to, struct net_device *from);
2798
extern int dev_mc_sync_multiple(struct net_device *to, struct net_device *from);
2799 2800 2801
extern void dev_mc_unsync(struct net_device *to, struct net_device *from);
extern void dev_mc_flush(struct net_device *dev);
extern void dev_mc_init(struct net_device *dev);
2802

2803 2804 2805
/* Functions used for secondary unicast and multicast support */
extern void		dev_set_rx_mode(struct net_device *dev);
extern void		__dev_set_rx_mode(struct net_device *dev);
2806 2807
extern int		dev_set_promiscuity(struct net_device *dev, int inc);
extern int		dev_set_allmulti(struct net_device *dev, int inc);
L
Linus Torvalds 已提交
2808
extern void		netdev_state_change(struct net_device *dev);
2809
extern void		netdev_notify_peers(struct net_device *dev);
2810
extern void		netdev_features_change(struct net_device *dev);
L
Linus Torvalds 已提交
2811
/* Load a device via the kmod */
2812
extern void		dev_load(struct net *net, const char *name);
2813 2814
extern struct rtnl_link_stats64 *dev_get_stats(struct net_device *dev,
					       struct rtnl_link_stats64 *storage);
2815 2816
extern void netdev_stats_to_stats64(struct rtnl_link_stats64 *stats64,
				    const struct net_device_stats *netdev_stats);
2817

L
Linus Torvalds 已提交
2818
extern int		netdev_max_backlog;
E
Eric Dumazet 已提交
2819
extern int		netdev_tstamp_prequeue;
L
Linus Torvalds 已提交
2820
extern int		weight_p;
2821
extern int		bpf_jit_enable;
J
Jiri Pirko 已提交
2822 2823 2824 2825

extern bool netdev_has_upper_dev(struct net_device *dev,
				 struct net_device *upper_dev);
extern bool netdev_has_any_upper_dev(struct net_device *dev);
2826 2827
extern struct net_device *netdev_all_upper_get_next_dev_rcu(struct net_device *dev,
							    struct list_head **iter);
2828 2829

/* iterate through upper list, must be called under RCU read lock */
2830 2831 2832 2833 2834
#define netdev_for_each_all_upper_dev_rcu(dev, updev, iter) \
	for (iter = &(dev)->all_adj_list.upper, \
	     updev = netdev_all_upper_get_next_dev_rcu(dev, &(iter)); \
	     updev; \
	     updev = netdev_all_upper_get_next_dev_rcu(dev, &(iter)))
2835

J
Jiri Pirko 已提交
2836 2837 2838 2839 2840 2841
extern struct net_device *netdev_master_upper_dev_get(struct net_device *dev);
extern struct net_device *netdev_master_upper_dev_get_rcu(struct net_device *dev);
extern int netdev_upper_dev_link(struct net_device *dev,
				 struct net_device *upper_dev);
extern int netdev_master_upper_dev_link(struct net_device *dev,
					struct net_device *upper_dev);
2842 2843 2844
extern int netdev_master_upper_dev_link_private(struct net_device *dev,
						struct net_device *upper_dev,
						void *private);
J
Jiri Pirko 已提交
2845 2846
extern void netdev_upper_dev_unlink(struct net_device *dev,
				    struct net_device *upper_dev);
2847 2848 2849 2850
extern void *netdev_lower_dev_get_private_rcu(struct net_device *dev,
					      struct net_device *lower_dev);
extern void *netdev_lower_dev_get_private(struct net_device *dev,
					  struct net_device *lower_dev);
2851
extern int skb_checksum_help(struct sk_buff *skb);
2852 2853
extern struct sk_buff *__skb_gso_segment(struct sk_buff *skb,
	netdev_features_t features, bool tx_path);
2854 2855
extern struct sk_buff *skb_mac_gso_segment(struct sk_buff *skb,
					  netdev_features_t features);
2856 2857 2858 2859 2860 2861

static inline
struct sk_buff *skb_gso_segment(struct sk_buff *skb, netdev_features_t features)
{
	return __skb_gso_segment(skb, features, true);
}
2862 2863 2864 2865 2866 2867 2868 2869 2870 2871 2872 2873 2874
__be16 skb_network_protocol(struct sk_buff *skb);

static inline bool can_checksum_protocol(netdev_features_t features,
					 __be16 protocol)
{
	return ((features & NETIF_F_GEN_CSUM) ||
		((features & NETIF_F_V4_CSUM) &&
		 protocol == htons(ETH_P_IP)) ||
		((features & NETIF_F_V6_CSUM) &&
		 protocol == htons(ETH_P_IPV6)) ||
		((features & NETIF_F_FCOE_CRC) &&
		 protocol == htons(ETH_P_FCOE)));
}
2875

2876 2877 2878 2879 2880 2881 2882
#ifdef CONFIG_BUG
extern void netdev_rx_csum_fault(struct net_device *dev);
#else
static inline void netdev_rx_csum_fault(struct net_device *dev)
{
}
#endif
L
Linus Torvalds 已提交
2883 2884 2885 2886
/* rx skb timestamps */
extern void		net_enable_timestamp(void);
extern void		net_disable_timestamp(void);

2887
#ifdef CONFIG_PROC_FS
2888 2889 2890
extern int __init dev_proc_init(void);
#else
#define dev_proc_init() 0
2891 2892
#endif

2893 2894 2895
extern int netdev_class_create_file(struct class_attribute *class_attr);
extern void netdev_class_remove_file(struct class_attribute *class_attr);

2896 2897
extern struct kobj_ns_type_operations net_ns_type_operations;

2898
extern const char *netdev_drivername(const struct net_device *dev);
2899

2900 2901
extern void linkwatch_run_queue(void);

2902 2903
static inline netdev_features_t netdev_get_wanted_features(
	struct net_device *dev)
2904 2905 2906
{
	return (dev->features & ~dev->hw_features) | dev->wanted_features;
}
2907 2908
netdev_features_t netdev_increment_features(netdev_features_t all,
	netdev_features_t one, netdev_features_t mask);
2909 2910 2911 2912 2913 2914 2915 2916 2917 2918 2919

/* Allow TSO being used on stacked device :
 * Performing the GSO segmentation before last device
 * is a performance improvement.
 */
static inline netdev_features_t netdev_add_tso_features(netdev_features_t features,
							netdev_features_t mask)
{
	return netdev_increment_features(features, NETIF_F_ALL_TSO, mask);
}

2920
int __netdev_update_features(struct net_device *dev);
2921
void netdev_update_features(struct net_device *dev);
2922
void netdev_change_features(struct net_device *dev);
2923

2924 2925 2926
void netif_stacked_transfer_operstate(const struct net_device *rootdev,
					struct net_device *dev);

2927
netdev_features_t netif_skb_features(struct sk_buff *skb);
2928

2929
static inline bool net_gso_ok(netdev_features_t features, int gso_type)
2930
{
2931
	netdev_features_t feature = gso_type << NETIF_F_GSO_SHIFT;
2932 2933 2934 2935 2936 2937 2938 2939 2940

	/* check flags correspondence */
	BUILD_BUG_ON(SKB_GSO_TCPV4   != (NETIF_F_TSO >> NETIF_F_GSO_SHIFT));
	BUILD_BUG_ON(SKB_GSO_UDP     != (NETIF_F_UFO >> NETIF_F_GSO_SHIFT));
	BUILD_BUG_ON(SKB_GSO_DODGY   != (NETIF_F_GSO_ROBUST >> NETIF_F_GSO_SHIFT));
	BUILD_BUG_ON(SKB_GSO_TCP_ECN != (NETIF_F_TSO_ECN >> NETIF_F_GSO_SHIFT));
	BUILD_BUG_ON(SKB_GSO_TCPV6   != (NETIF_F_TSO6 >> NETIF_F_GSO_SHIFT));
	BUILD_BUG_ON(SKB_GSO_FCOE    != (NETIF_F_FSO >> NETIF_F_GSO_SHIFT));

2941
	return (features & feature) == feature;
2942 2943
}

2944
static inline bool skb_gso_ok(struct sk_buff *skb, netdev_features_t features)
2945
{
H
Herbert Xu 已提交
2946
	return net_gso_ok(features, skb_shinfo(skb)->gso_type) &&
2947
	       (!skb_has_frag_list(skb) || (features & NETIF_F_FRAGLIST));
2948 2949
}

2950 2951
static inline bool netif_needs_gso(struct sk_buff *skb,
				   netdev_features_t features)
2952
{
2953
	return skb_is_gso(skb) && (!skb_gso_ok(skb, features) ||
2954 2955
		unlikely((skb->ip_summed != CHECKSUM_PARTIAL) &&
			 (skb->ip_summed != CHECKSUM_UNNECESSARY)));
2956 2957
}

2958 2959 2960 2961 2962 2963
static inline void netif_set_gso_max_size(struct net_device *dev,
					  unsigned int size)
{
	dev->gso_max_size = size;
}

2964 2965 2966 2967 2968
static inline bool netif_is_bond_master(struct net_device *dev)
{
	return dev->flags & IFF_MASTER && dev->priv_flags & IFF_BONDING;
}

2969
static inline bool netif_is_bond_slave(struct net_device *dev)
J
Jiri Pirko 已提交
2970 2971 2972 2973
{
	return dev->flags & IFF_SLAVE && dev->priv_flags & IFF_BONDING;
}

2974 2975 2976 2977 2978
static inline bool netif_supports_nofcs(struct net_device *dev)
{
	return dev->priv_flags & IFF_SUPP_NOFCS;
}

2979
extern struct pernet_operations __net_initdata loopback_net_ops;
2980

2981 2982 2983 2984 2985 2986 2987 2988 2989 2990 2991
/* Logging, debugging and troubleshooting/diagnostic helpers. */

/* netdev_printk helpers, similar to dev_printk */

static inline const char *netdev_name(const struct net_device *dev)
{
	if (dev->reg_state != NETREG_REGISTERED)
		return "(unregistered net_device)";
	return dev->name;
}

2992 2993 2994 2995 2996 2997 2998 2999 3000 3001 3002 3003 3004 3005 3006 3007 3008
extern __printf(3, 4)
int netdev_printk(const char *level, const struct net_device *dev,
		  const char *format, ...);
extern __printf(2, 3)
int netdev_emerg(const struct net_device *dev, const char *format, ...);
extern __printf(2, 3)
int netdev_alert(const struct net_device *dev, const char *format, ...);
extern __printf(2, 3)
int netdev_crit(const struct net_device *dev, const char *format, ...);
extern __printf(2, 3)
int netdev_err(const struct net_device *dev, const char *format, ...);
extern __printf(2, 3)
int netdev_warn(const struct net_device *dev, const char *format, ...);
extern __printf(2, 3)
int netdev_notice(const struct net_device *dev, const char *format, ...);
extern __printf(2, 3)
int netdev_info(const struct net_device *dev, const char *format, ...);
3009

3010 3011 3012
#define MODULE_ALIAS_NETDEV(device) \
	MODULE_ALIAS("netdev-" device)

3013
#if defined(CONFIG_DYNAMIC_DEBUG)
3014 3015
#define netdev_dbg(__dev, format, args...)			\
do {								\
3016
	dynamic_netdev_dbg(__dev, format, ##args);		\
3017
} while (0)
3018 3019 3020
#elif defined(DEBUG)
#define netdev_dbg(__dev, format, args...)			\
	netdev_printk(KERN_DEBUG, __dev, format, ##args)
3021 3022 3023 3024 3025 3026 3027 3028 3029 3030 3031 3032 3033 3034 3035 3036 3037 3038 3039 3040 3041 3042 3043 3044 3045 3046 3047 3048 3049
#else
#define netdev_dbg(__dev, format, args...)			\
({								\
	if (0)							\
		netdev_printk(KERN_DEBUG, __dev, format, ##args); \
	0;							\
})
#endif

#if defined(VERBOSE_DEBUG)
#define netdev_vdbg	netdev_dbg
#else

#define netdev_vdbg(dev, format, args...)			\
({								\
	if (0)							\
		netdev_printk(KERN_DEBUG, dev, format, ##args);	\
	0;							\
})
#endif

/*
 * netdev_WARN() acts like dev_printk(), but with the key difference
 * of using a WARN/WARN_ON to get the message out, including the
 * file/line information and a backtrace.
 */
#define netdev_WARN(dev, format, args...)			\
	WARN(1, "netdevice: %s\n" format, netdev_name(dev), ##args);

3050 3051 3052 3053 3054 3055 3056 3057
/* netif printk helpers, similar to netdev_printk */

#define netif_printk(priv, type, level, dev, fmt, args...)	\
do {					  			\
	if (netif_msg_##type(priv))				\
		netdev_printk(level, (dev), fmt, ##args);	\
} while (0)

3058 3059 3060 3061 3062 3063
#define netif_level(level, priv, type, dev, fmt, args...)	\
do {								\
	if (netif_msg_##type(priv))				\
		netdev_##level(dev, fmt, ##args);		\
} while (0)

3064
#define netif_emerg(priv, type, dev, fmt, args...)		\
3065
	netif_level(emerg, priv, type, dev, fmt, ##args)
3066
#define netif_alert(priv, type, dev, fmt, args...)		\
3067
	netif_level(alert, priv, type, dev, fmt, ##args)
3068
#define netif_crit(priv, type, dev, fmt, args...)		\
3069
	netif_level(crit, priv, type, dev, fmt, ##args)
3070
#define netif_err(priv, type, dev, fmt, args...)		\
3071
	netif_level(err, priv, type, dev, fmt, ##args)
3072
#define netif_warn(priv, type, dev, fmt, args...)		\
3073
	netif_level(warn, priv, type, dev, fmt, ##args)
3074
#define netif_notice(priv, type, dev, fmt, args...)		\
3075
	netif_level(notice, priv, type, dev, fmt, ##args)
3076
#define netif_info(priv, type, dev, fmt, args...)		\
3077
	netif_level(info, priv, type, dev, fmt, ##args)
3078

3079
#if defined(CONFIG_DYNAMIC_DEBUG)
3080 3081 3082
#define netif_dbg(priv, type, netdev, format, args...)		\
do {								\
	if (netif_msg_##type(priv))				\
3083
		dynamic_netdev_dbg(netdev, format, ##args);	\
3084
} while (0)
3085 3086 3087
#elif defined(DEBUG)
#define netif_dbg(priv, type, dev, format, args...)		\
	netif_printk(priv, type, KERN_DEBUG, dev, format, ##args)
3088 3089 3090 3091 3092 3093 3094 3095 3096 3097
#else
#define netif_dbg(priv, type, dev, format, args...)			\
({									\
	if (0)								\
		netif_printk(priv, type, KERN_DEBUG, dev, format, ##args); \
	0;								\
})
#endif

#if defined(VERBOSE_DEBUG)
3098
#define netif_vdbg	netif_dbg
3099 3100 3101 3102
#else
#define netif_vdbg(priv, type, dev, format, args...)		\
({								\
	if (0)							\
3103
		netif_printk(priv, type, KERN_DEBUG, dev, format, ##args); \
3104 3105 3106
	0;							\
})
#endif
3107

3108 3109 3110 3111 3112 3113 3114 3115 3116 3117 3118 3119 3120 3121 3122 3123 3124 3125 3126 3127 3128 3129 3130 3131 3132 3133 3134 3135 3136 3137
/*
 *	The list of packet types we will receive (as opposed to discard)
 *	and the routines to invoke.
 *
 *	Why 16. Because with 16 the only overlap we get on a hash of the
 *	low nibble of the protocol value is RARP/SNAP/X.25.
 *
 *      NOTE:  That is no longer true with the addition of VLAN tags.  Not
 *             sure which should go first, but I bet it won't make much
 *             difference if we are running VLANs.  The good news is that
 *             this protocol won't be in the list unless compiled in, so
 *             the average user (w/out VLANs) will not be adversely affected.
 *             --BLG
 *
 *		0800	IP
 *		8100    802.1Q VLAN
 *		0001	802.3
 *		0002	AX.25
 *		0004	802.2
 *		8035	RARP
 *		0005	SNAP
 *		0805	X.25
 *		0806	ARP
 *		8137	IPX
 *		0009	Localtalk
 *		86DD	IPv6
 */
#define PTYPE_HASH_SIZE	(16)
#define PTYPE_HASH_MASK	(PTYPE_HASH_SIZE - 1)

3138
#endif	/* _LINUX_NETDEVICE_H */