netdevice.h 87.5 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 66
/* 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
67 68
#define NET_ADDR_SET		3	/* address is set using
					 * dev_set_mac_address() */
69

70 71 72 73
/* Backlog congestion levels */
#define NET_RX_SUCCESS		0	/* keep 'em coming, baby */
#define NET_RX_DROP		1	/* packet dropped */

74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92
/*
 * 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
93 94 95 96
#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
97

98 99 100
/* 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. */
101
#define net_xmit_eval(e)	((e) == NET_XMIT_CN ? 0 : (e))
L
Linus Torvalds 已提交
102 103
#define net_xmit_errno(e)	((e) != NET_XMIT_CN ? -ENOBUFS : 0)

104
/* Driver transmit return codes */
105
#define NETDEV_TX_MASK		0xf0
106

107
enum netdev_tx {
108
	__NETDEV_TX_MIN	 = INT_MIN,	/* make sure enum is signed */
109 110 111
	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 */
112 113 114
};
typedef enum netdev_tx netdev_tx_t;

115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132
/*
 * 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
133 134 135 136
/*
 *	Compute the worst case header length according to the protocols
 *	used.
 */
G
Graf Yang 已提交
137

138
#if defined(CONFIG_WLAN) || IS_ENABLED(CONFIG_AX25)
139 140 141 142 143
# if defined(CONFIG_MAC80211_MESH)
#  define LL_MAX_HEADER 128
# else
#  define LL_MAX_HEADER 96
# endif
144
#elif IS_ENABLED(CONFIG_TR)
145
# define LL_MAX_HEADER 48
L
Linus Torvalds 已提交
146
#else
147
# define LL_MAX_HEADER 32
L
Linus Torvalds 已提交
148 149
#endif

150 151
#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
152 153 154 155 156 157
#define MAX_HEADER LL_MAX_HEADER
#else
#define MAX_HEADER (LL_MAX_HEADER + 48)
#endif

/*
158 159
 *	Old network device statistics. Fields are native words
 *	(unsigned long) so they can be read and written atomically.
L
Linus Torvalds 已提交
160
 */
G
Graf Yang 已提交
161

E
Eric Dumazet 已提交
162
struct net_device_stats {
163 164 165 166 167 168 169 170 171
	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
172 173
	unsigned long	collisions;
	unsigned long	rx_length_errors;
174 175 176 177 178
	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
179 180 181 182 183 184 185 186 187 188 189 190 191
	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>

192
#ifdef CONFIG_RPS
193 194
#include <linux/static_key.h>
extern struct static_key rps_needed;
195 196
#endif

L
Linus Torvalds 已提交
197 198 199 200
struct neighbour;
struct neigh_parms;
struct sk_buff;

201 202 203 204
struct netdev_hw_addr {
	struct list_head	list;
	unsigned char		addr[MAX_ADDR_LEN];
	unsigned char		type;
J
Jiri Pirko 已提交
205 206 207 208
#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
209
#define NETDEV_HW_ADDR_T_MULTICAST	5
J
Jiri Pirko 已提交
210
	bool			synced;
211
	bool			global_use;
212
	int			refcount;
213 214 215
	struct rcu_head		rcu_head;
};

216 217 218 219 220
struct netdev_hw_addr_list {
	struct list_head	list;
	int			count;
};

221 222 223 224
#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)
225

226 227 228 229
#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)
230

231 232
#define netdev_mc_count(dev) netdev_hw_addr_list_count(&(dev)->mc)
#define netdev_mc_empty(dev) netdev_hw_addr_list_empty(&(dev)->mc)
233
#define netdev_for_each_mc_addr(ha, dev) \
234
	netdev_hw_addr_list_for_each(ha, &(dev)->mc)
235

E
Eric Dumazet 已提交
236
struct hh_cache {
237
	u16		hh_len;
238
	u16		__pad;
239
	seqlock_t	hh_lock;
L
Linus Torvalds 已提交
240 241 242 243

	/* cached hardware header; allow for machine alignment needs.        */
#define HH_DATA_MOD	16
#define HH_DATA_OFF(__len) \
J
Jiri Benc 已提交
244
	(HH_DATA_MOD - (((__len - 1) & (HH_DATA_MOD - 1)) + 1))
L
Linus Torvalds 已提交
245 246 247 248 249 250 251 252 253 254 255 256 257 258
#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) \
259
	((((dev)->hard_header_len+(dev)->needed_headroom)&~(HH_DATA_MOD - 1)) + HH_DATA_MOD)
L
Linus Torvalds 已提交
260
#define LL_RESERVED_SPACE_EXTRA(dev,extra) \
261
	((((dev)->hard_header_len+(dev)->needed_headroom+(extra))&~(HH_DATA_MOD - 1)) + HH_DATA_MOD)
L
Linus Torvalds 已提交
262

263 264 265
struct header_ops {
	int	(*create) (struct sk_buff *skb, struct net_device *dev,
			   unsigned short type, const void *daddr,
266
			   const void *saddr, unsigned int len);
267 268
	int	(*parse)(const struct sk_buff *skb, unsigned char *haddr);
	int	(*rebuild)(struct sk_buff *skb);
269
	int	(*cache)(const struct neighbour *neigh, struct hh_cache *hh, __be16 type);
270 271 272 273 274
	void	(*cache_update)(struct hh_cache *hh,
				const struct net_device *dev,
				const unsigned char *haddr);
};

L
Linus Torvalds 已提交
275 276 277 278 279
/* 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
280
enum netdev_state_t {
L
Linus Torvalds 已提交
281 282 283
	__LINK_STATE_START,
	__LINK_STATE_PRESENT,
	__LINK_STATE_NOCARRIER,
S
Stefan Rompf 已提交
284 285
	__LINK_STATE_LINKWATCH_PENDING,
	__LINK_STATE_DORMANT,
L
Linus Torvalds 已提交
286 287 288 289 290
};


/*
 * This structure holds at boot time configured netdevice settings. They
G
Graf Yang 已提交
291
 * are then used in the device probing.
L
Linus Torvalds 已提交
292 293 294 295 296 297 298
 */
struct netdev_boot_setup {
	char name[IFNAMSIZ];
	struct ifmap map;
};
#define NETDEV_BOOT_SETUP_MAX 8

299
extern int __init netdev_boot_setup(char *str);
L
Linus Torvalds 已提交
300

301 302 303 304 305 306 307 308 309 310 311 312 313 314
/*
 * 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;
315
	unsigned int		gro_count;
316 317 318 319 320
	int			(*poll)(struct napi_struct *, int);
#ifdef CONFIG_NETPOLL
	spinlock_t		poll_lock;
	int			poll_owner;
#endif
H
Herbert Xu 已提交
321
	struct net_device	*dev;
322
	struct sk_buff		*gro_list;
H
Herbert Xu 已提交
323
	struct sk_buff		*skb;
324
	struct list_head	dev_list;
325 326
};

E
Eric Dumazet 已提交
327
enum {
328
	NAPI_STATE_SCHED,	/* Poll is scheduled */
D
David S. Miller 已提交
329
	NAPI_STATE_DISABLE,	/* Disable pending */
330
	NAPI_STATE_NPSVC,	/* Netpoll - don't dequeue from poll_list */
331 332
};

333
enum gro_result {
334 335 336 337 338 339
	GRO_MERGED,
	GRO_MERGED_FREE,
	GRO_HELD,
	GRO_NORMAL,
	GRO_DROP,
};
340
typedef enum gro_result gro_result_t;
341

342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373
/*
 * 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
374
 * are registered on exact device (ptype->dev == skb->dev).
375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390
 *
 * 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);
391

392
extern void __napi_schedule(struct napi_struct *n);
393

394
static inline bool napi_disable_pending(struct napi_struct *n)
D
David S. Miller 已提交
395 396 397 398
{
	return test_bit(NAPI_STATE_DISABLE, &n->state);
}

399 400 401 402 403 404
/**
 *	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
405 406
 * insure only one NAPI poll instance runs.  We also make
 * sure there is no pending NAPI disable.
407
 */
408
static inline bool napi_schedule_prep(struct napi_struct *n)
409
{
D
David S. Miller 已提交
410 411
	return !napi_disable_pending(n) &&
		!test_and_set_bit(NAPI_STATE_SCHED, &n->state);
412 413 414 415 416 417 418 419 420 421 422 423 424 425 426
}

/**
 *	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);
}

427
/* Try to reschedule poll. Called by dev->poll() after napi_complete().  */
428
static inline bool napi_reschedule(struct napi_struct *napi)
429 430 431
{
	if (napi_schedule_prep(napi)) {
		__napi_schedule(napi);
432
		return true;
433
	}
434
	return false;
435 436
}

437 438 439 440 441 442
/**
 *	napi_complete - NAPI processing complete
 *	@n: napi context
 *
 * Mark NAPI processing as complete.
 */
443 444
extern void __napi_complete(struct napi_struct *n);
extern void napi_complete(struct napi_struct *n);
445 446 447 448 449 450 451 452 453 454

/**
 *	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
455
	set_bit(NAPI_STATE_DISABLE, &n->state);
456
	while (test_and_set_bit(NAPI_STATE_SCHED, &n->state))
457
		msleep(1);
D
David S. Miller 已提交
458
	clear_bit(NAPI_STATE_DISABLE, &n->state);
459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474
}

/**
 *	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);
}

475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492
#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
493
enum netdev_queue_state_t {
494 495
	__QUEUE_STATE_DRV_XOFF,
	__QUEUE_STATE_STACK_XOFF,
496
	__QUEUE_STATE_FROZEN,
497 498 499 500
#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))
501
};
502 503 504 505 506 507 508 509 510
/*
 * __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_*.
 */
511

512
struct netdev_queue {
513 514 515
/*
 * read mostly part
 */
516
	struct net_device	*dev;
517 518
	struct Qdisc		*qdisc;
	struct Qdisc		*qdisc_sleeping;
519
#ifdef CONFIG_SYSFS
T
Tom Herbert 已提交
520 521
	struct kobject		kobj;
#endif
522 523 524
#if defined(CONFIG_XPS) && defined(CONFIG_NUMA)
	int			numa_node;
#endif
525 526 527 528 529
/*
 * write mostly part
 */
	spinlock_t		_xmit_lock ____cacheline_aligned_in_smp;
	int			xmit_lock_owner;
530 531 532 533
	/*
	 * please use this field instead of dev->trans_start
	 */
	unsigned long		trans_start;
534 535 536 537 538 539

	/*
	 * Number of TX timeouts for this queue
	 * (/sys/class/net/DEV/Q/trans_timeout)
	 */
	unsigned long		trans_timeout;
T
Tom Herbert 已提交
540 541 542 543 544 545

	unsigned long		state;

#ifdef CONFIG_BQL
	struct dql		dql;
#endif
546
} ____cacheline_aligned_in_smp;
547

548 549 550 551 552
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
553
	return NUMA_NO_NODE;
554 555 556 557 558 559 560 561 562 563
#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
564
#ifdef CONFIG_RPS
T
Tom Herbert 已提交
565 566 567 568 569 570 571 572 573
/*
 * 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];
};
574
#define RPS_MAP_SIZE(_num) (sizeof(struct rps_map) + ((_num) * sizeof(u16)))
T
Tom Herbert 已提交
575

T
Tom Herbert 已提交
576
/*
577 578 579
 * 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
580 581 582
 */
struct rps_dev_flow {
	u16 cpu;
583
	u16 filter;
T
Tom Herbert 已提交
584 585
	unsigned int last_qtail;
};
586
#define RPS_NO_FILTER 0xffff
T
Tom Herbert 已提交
587 588 589 590 591 592 593 594 595 596 597

/*
 * 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 work_struct free_work;
	struct rps_dev_flow flows[0];
};
#define RPS_DEV_FLOW_TABLE_SIZE(_num) (sizeof(struct rps_dev_flow_table) + \
598
    ((_num) * sizeof(struct rps_dev_flow)))
T
Tom Herbert 已提交
599 600 601 602 603 604 605 606 607 608

/*
 * 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) + \
609
    ((_num) * sizeof(u16)))
T
Tom Herbert 已提交
610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633

#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
634
extern struct rps_sock_flow_table __rcu *rps_sock_flow_table;
T
Tom Herbert 已提交
635

636 637 638 639 640
#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
641 642
/* This structure contains an instance of an RX queue. */
struct netdev_rx_queue {
E
Eric Dumazet 已提交
643 644 645
	struct rps_map __rcu		*rps_map;
	struct rps_dev_flow_table __rcu	*rps_flow_table;
	struct kobject			kobj;
T
Tom Herbert 已提交
646
	struct net_device		*dev;
T
Tom Herbert 已提交
647
} ____cacheline_aligned_in_smp;
T
Tom Herbert 已提交
648
#endif /* CONFIG_RPS */
649

T
Tom Herbert 已提交
650 651 652 653 654 655 656 657 658 659 660
#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];
};
661
#define XPS_MAP_SIZE(_num) (sizeof(struct xps_map) + ((_num) * sizeof(u16)))
T
Tom Herbert 已提交
662 663 664 665 666 667 668 669
#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
670
	struct xps_map __rcu *cpu_map[0];
T
Tom Herbert 已提交
671 672 673 674 675
};
#define XPS_DEV_MAPS_SIZE (sizeof(struct xps_dev_maps) +		\
    (nr_cpu_ids * sizeof(struct xps_map *)))
#endif /* CONFIG_XPS */

676 677 678 679 680 681 682 683
#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;
};

684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700
#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

701 702
/*
 * This structure defines the management hooks for network devices.
703 704
 * The following hooks can be defined; unless noted otherwise, they are
 * optional and can be filled with a null pointer.
705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723
 *
 * 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.
 *
724 725
 * netdev_tx_t (*ndo_start_xmit)(struct sk_buff *skb,
 *                               struct net_device *dev);
726
 *	Called when a packet needs to be transmitted.
727 728
 *	Must return NETDEV_TX_OK , NETDEV_TX_BUSY.
 *        (can also return NETDEV_TX_LOCKED iff NETIF_F_LLTX)
729 730 731 732 733 734
 *	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.
 *
735 736 737 738 739 740
 * 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.
741 742
 *	If driver handles unicast address filtering, it should set
 *	IFF_UNICAST_FLT to its priv_flags.
743 744 745
 *
 * int (*ndo_set_mac_address)(struct net_device *dev, void *addr);
 *	This function  is called when the Media Access Control address
746
 *	needs to be changed. If this interface is not defined, the
747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766
 *	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.
 *
767
 * void (*ndo_tx_timeout)(struct net_device *dev);
768 769 770
 *	Callback uses when the transmitter has not made any progress
 *	for dev->watchdog ticks.
 *
771
 * struct rtnl_link_stats64* (*ndo_get_stats64)(struct net_device *dev,
772
 *                      struct rtnl_link_stats64 *storage);
773
 * struct net_device_stats* (*ndo_get_stats)(struct net_device *dev);
774
 *	Called when a user wants to get the network device usage
775
 *	statistics. Drivers must do one of the following:
776 777
 *	1. Define @ndo_get_stats64 to fill in a zero-initialised
 *	   rtnl_link_stats64 structure passed by the caller.
778
 *	2. Define @ndo_get_stats to update a net_device_stats structure
779 780 781 782 783
 *	   (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.
784
 *
785
 * int (*ndo_vlan_rx_add_vid)(struct net_device *dev, unsigned short vid);
786 787 788
 *	If device support VLAN filtering (dev->features & NETIF_F_HW_VLAN_FILTER)
 *	this function is called when a VLAN id is registered.
 *
789
 * int (*ndo_vlan_rx_kill_vid)(struct net_device *dev, unsigned short vid);
790 791 792 793
 *	If device support VLAN filtering (dev->features & NETIF_F_HW_VLAN_FILTER)
 *	this function is called when a VLAN id is unregistered.
 *
 * void (*ndo_poll_controller)(struct net_device *dev);
794 795 796 797 798
 *
 *	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);
799
 * int (*ndo_set_vf_spoofchk)(struct net_device *dev, int vf, bool setting);
800 801
 * int (*ndo_get_vf_config)(struct net_device *dev,
 *			    int vf, struct ifla_vf_info *ivf);
802 803 804
 * 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);
805 806 807 808 809
 * 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.
810
 *
811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840
 *	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.
 *
841 842 843 844 845 846 847
 * 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.
 *
848 849 850 851 852 853
 * 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.
 *
854 855 856 857 858 859
 *	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.
860 861 862 863 864 865 866 867
 *
 *	Slave management functions (for bridge, bonding, etc). User should
 *	call netdev_set_master() to set dev->master properly.
 * 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.
868 869
 *
 *      Feature/offload setting functions.
870 871
 * netdev_features_t (*ndo_fix_features)(struct net_device *dev,
 *		netdev_features_t features);
872 873 874 875
 *	Adjusts the requested feature flags according to device-specific
 *	constraints, and returns the resulting flags. Must not modify
 *	the device state.
 *
876
 * int (*ndo_set_features)(struct net_device *dev, netdev_features_t features);
877 878 879 880
 *	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.
 *
881 882
 * int (*ndo_fdb_add)(struct ndmsg *ndm, struct nlattr *tb[],
 *		      struct net_device *dev,
883
 *		      const unsigned char *addr, u16 flags)
884 885
 *	Adds an FDB entry to dev for addr.
 * int (*ndo_fdb_del)(struct ndmsg *ndm, struct net_device *dev,
886
 *		      const unsigned char *addr)
887 888 889 890 891
 *	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
892 893 894 895
 *
 * int (*ndo_bridge_setlink)(struct net_device *dev, struct nlmsghdr *nlh)
 * int (*ndo_bridge_getlink)(struct sk_buff *skb, u32 pid, u32 seq,
 *			     struct net_device *dev)
J
Jiri Pirko 已提交
896 897 898 899 900 901 902 903
 *
 * 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.
904 905 906 907 908 909
 */
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);
910
	netdev_tx_t		(*ndo_start_xmit) (struct sk_buff *skb,
911 912 913
						   struct net_device *dev);
	u16			(*ndo_select_queue)(struct net_device *dev,
						    struct sk_buff *skb);
914 915 916 917 918 919 920 921 922 923
	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);
924 925 926 927
	int			(*ndo_change_mtu)(struct net_device *dev,
						  int new_mtu);
	int			(*ndo_neigh_setup)(struct net_device *dev,
						   struct neigh_parms *);
928 929
	void			(*ndo_tx_timeout) (struct net_device *dev);

930 931
	struct rtnl_link_stats64* (*ndo_get_stats64)(struct net_device *dev,
						     struct rtnl_link_stats64 *storage);
932 933
	struct net_device_stats* (*ndo_get_stats)(struct net_device *dev);

934
	int			(*ndo_vlan_rx_add_vid)(struct net_device *dev,
935
						       unsigned short vid);
936
	int			(*ndo_vlan_rx_kill_vid)(struct net_device *dev,
937 938 939
						        unsigned short vid);
#ifdef CONFIG_NET_POLL_CONTROLLER
	void                    (*ndo_poll_controller)(struct net_device *dev);
H
Herbert Xu 已提交
940
	int			(*ndo_netpoll_setup)(struct net_device *dev,
941 942
						     struct netpoll_info *info,
						     gfp_t gfp);
943
	void			(*ndo_netpoll_cleanup)(struct net_device *dev);
944
#endif
945 946 947 948 949 950
	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);
951 952
	int			(*ndo_set_vf_spoofchk)(struct net_device *dev,
						       int vf, bool setting);
953 954 955
	int			(*ndo_get_vf_config)(struct net_device *dev,
						     int vf,
						     struct ifla_vf_info *ivf);
956 957 958 959 960
	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);
961
	int			(*ndo_setup_tc)(struct net_device *dev, u8 tc);
962
#if IS_ENABLED(CONFIG_FCOE)
963 964
	int			(*ndo_fcoe_enable)(struct net_device *dev);
	int			(*ndo_fcoe_disable)(struct net_device *dev);
965 966 967 968 969 970
	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);
971 972 973 974
	int			(*ndo_fcoe_ddp_target)(struct net_device *dev,
						       u16 xid,
						       struct scatterlist *sgl,
						       unsigned int sgc);
975 976
	int			(*ndo_fcoe_get_hbainfo)(struct net_device *dev,
							struct netdev_fcoe_hbainfo *hbainfo);
977 978
#endif

979
#if IS_ENABLED(CONFIG_LIBFCOE)
980 981 982 983
#define NETDEV_FCOE_WWNN 0
#define NETDEV_FCOE_WWPN 1
	int			(*ndo_fcoe_get_wwn)(struct net_device *dev,
						    u64 *wwn, int type);
984
#endif
985

986 987 988 989 990 991
#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
992 993 994 995
	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);
996 997
	netdev_features_t	(*ndo_fix_features)(struct net_device *dev,
						    netdev_features_t features);
998
	int			(*ndo_set_features)(struct net_device *dev,
999
						    netdev_features_t features);
1000
	int			(*ndo_neigh_construct)(struct neighbour *n);
1001
	void			(*ndo_neigh_destroy)(struct neighbour *n);
1002 1003

	int			(*ndo_fdb_add)(struct ndmsg *ndm,
1004
					       struct nlattr *tb[],
1005
					       struct net_device *dev,
1006
					       const unsigned char *addr,
1007 1008 1009
					       u16 flags);
	int			(*ndo_fdb_del)(struct ndmsg *ndm,
					       struct net_device *dev,
1010
					       const unsigned char *addr);
1011 1012 1013 1014
	int			(*ndo_fdb_dump)(struct sk_buff *skb,
						struct netlink_callback *cb,
						struct net_device *dev,
						int idx);
J
John Fastabend 已提交
1015 1016 1017 1018 1019 1020

	int			(*ndo_bridge_setlink)(struct net_device *dev,
						      struct nlmsghdr *nlh);
	int			(*ndo_bridge_getlink)(struct sk_buff *skb,
						      u32 pid, u32 seq,
						      struct net_device *dev);
J
Jiri Pirko 已提交
1021 1022
	int			(*ndo_change_carrier)(struct net_device *dev,
						      bool new_carrier);
1023 1024
};

L
Linus Torvalds 已提交
1025 1026 1027 1028 1029 1030 1031 1032 1033 1034
/*
 *	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
1035
struct net_device {
L
Linus Torvalds 已提交
1036 1037 1038 1039

	/*
	 * 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
1040
	 * of the interface.
L
Linus Torvalds 已提交
1041 1042
	 */
	char			name[IFNAMSIZ];
M
Mark Gross 已提交
1043

1044
	/* device name hash chain, please keep it close to name[] */
1045
	struct hlist_node	name_hlist;
1046

1047 1048
	/* snmp alias */
	char 			*ifalias;
L
Linus Torvalds 已提交
1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065

	/*
	 *	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;

1066
	struct list_head	dev_list;
1067
	struct list_head	napi_list;
1068
	struct list_head	unreg_list;
L
Linus Torvalds 已提交
1069

1070
	/* currently active device features */
1071
	netdev_features_t	features;
1072
	/* user-changeable features */
1073
	netdev_features_t	hw_features;
1074
	/* user-requested features */
1075
	netdev_features_t	wanted_features;
1076
	/* mask of features inheritable by VLAN devices */
1077
	netdev_features_t	vlan_features;
1078 1079 1080 1081 1082 1083
	/* 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;
1084

L
Linus Torvalds 已提交
1085 1086 1087 1088
	/* Interface index. Unique device identifier	*/
	int			ifindex;
	int			iflink;

R
Rusty Russell 已提交
1089
	struct net_device_stats	stats;
1090 1091 1092
	atomic_long_t		rx_dropped; /* dropped packets by core network
					     * Do not use this in drivers.
					     */
L
Linus Torvalds 已提交
1093

1094
#ifdef CONFIG_WIRELESS_EXT
L
Linus Torvalds 已提交
1095 1096 1097 1098 1099
	/* 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;
1100
#endif
1101 1102
	/* Management operations */
	const struct net_device_ops *netdev_ops;
1103
	const struct ethtool_ops *ethtool_ops;
L
Linus Torvalds 已提交
1104

1105 1106 1107
	/* Hardware header description */
	const struct header_ops *header_ops;

S
Stefan Rompf 已提交
1108
	unsigned int		flags;	/* interface flags (a la BSD)	*/
1109 1110
	unsigned int		priv_flags; /* Like 'flags' but invisible to userspace.
					     * See if.h for definitions. */
L
Linus Torvalds 已提交
1111 1112 1113
	unsigned short		gflags;
	unsigned short		padded;	/* How much padding added by alloc_netdev() */

S
Stefan Rompf 已提交
1114 1115 1116
	unsigned char		operstate; /* RFC2863 operstate */
	unsigned char		link_mode; /* mapping policy to operstate */

1117 1118 1119
	unsigned char		if_port;	/* Selectable AUI, TP,..*/
	unsigned char		dma;		/* DMA channel		*/

1120
	unsigned int		mtu;	/* interface MTU value		*/
L
Linus Torvalds 已提交
1121 1122 1123
	unsigned short		type;	/* interface hardware type	*/
	unsigned short		hard_header_len;	/* hardware hdr length	*/

1124 1125 1126 1127 1128 1129 1130
	/* 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
1131
	/* Interface address info. */
1132
	unsigned char		perm_addr[MAX_ADDR_LEN]; /* permanent hw address */
1133
	unsigned char		addr_assign_type; /* hw address assignment type */
L
Linus Torvalds 已提交
1134
	unsigned char		addr_len;	/* hardware address length	*/
1135
	unsigned char		neigh_priv_len;
L
Linus Torvalds 已提交
1136 1137
	unsigned short          dev_id;		/* for shared network cards */

J
Jiri Pirko 已提交
1138
	spinlock_t		addr_list_lock;
1139 1140
	struct netdev_hw_addr_list	uc;	/* Unicast mac addresses */
	struct netdev_hw_addr_list	mc;	/* Multicast mac addresses */
1141
	bool			uc_promisc;
1142 1143
	unsigned int		promiscuity;
	unsigned int		allmulti;
L
Linus Torvalds 已提交
1144 1145 1146


	/* Protocol specific pointers */
1147

1148
#if IS_ENABLED(CONFIG_VLAN_8021Q)
1149
	struct vlan_info __rcu	*vlan_info;	/* VLAN info */
1150
#endif
1151
#if IS_ENABLED(CONFIG_NET_DSA)
1152
	struct dsa_switch_tree	*dsa_ptr;	/* dsa specific data */
1153
#endif
L
Linus Torvalds 已提交
1154
	void 			*atalk_ptr;	/* AppleTalk link 	*/
E
Eric Dumazet 已提交
1155
	struct in_device __rcu	*ip_ptr;	/* IPv4 specific data	*/
1156
	struct dn_dev __rcu     *dn_ptr;        /* DECnet specific data */
E
Eric Dumazet 已提交
1157
	struct inet6_dev __rcu	*ip6_ptr;       /* IPv6 specific data */
L
Linus Torvalds 已提交
1158
	void			*ax25_ptr;	/* AX.25 specific data */
1159 1160
	struct wireless_dev	*ieee80211_ptr;	/* IEEE 802.11 specific data,
						   assign before registering */
L
Linus Torvalds 已提交
1161

1162
/*
E
Eric Dumazet 已提交
1163
 * Cache lines mostly used on receive path (including eth_type_trans())
1164
 */
1165 1166 1167 1168 1169 1170 1171 1172
	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.
						 */

E
Eric Dumazet 已提交
1173 1174 1175 1176
	struct net_device	*master; /* Pointer to master device of a group,
					  * which this device is member of.
					  */

J
Jiri Pirko 已提交
1177 1178
	struct list_head	upper_dev_list; /* List of upper devices */

1179
	/* Interface address info used in eth_type_trans() */
1180 1181 1182 1183
	unsigned char		*dev_addr;	/* hw address, (before bcast
						   because most packets are
						   unicast) */

1184 1185
	struct netdev_hw_addr_list	dev_addrs; /* list of device
						      hw addresses */
1186 1187

	unsigned char		broadcast[MAX_ADDR_LEN];	/* hw bcast add	*/
L
Linus Torvalds 已提交
1188

1189
#ifdef CONFIG_SYSFS
T
Tom Herbert 已提交
1190
	struct kset		*queues_kset;
1191
#endif
T
Tom Herbert 已提交
1192

1193
#ifdef CONFIG_RPS
T
Tom Herbert 已提交
1194 1195
	struct netdev_rx_queue	*_rx;

1196
	/* Number of RX queues allocated at register_netdev() time */
T
Tom Herbert 已提交
1197
	unsigned int		num_rx_queues;
1198 1199 1200

	/* Number of RX queues currently active in device */
	unsigned int		real_num_rx_queues;
1201 1202 1203 1204 1205 1206 1207

#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
E
Eric Dumazet 已提交
1208
#endif
T
Tom Herbert 已提交
1209

1210 1211
	rx_handler_func_t __rcu	*rx_handler;
	void __rcu		*rx_handler_data;
1212

1213
	struct netdev_queue __rcu *ingress_queue;
E
Eric Dumazet 已提交
1214 1215 1216 1217

/*
 * Cache lines mostly used on transmit path
 */
1218
	struct netdev_queue	*_tx ____cacheline_aligned_in_smp;
1219 1220

	/* Number of TX queues allocated at alloc_netdev_mq() time  */
1221
	unsigned int		num_tx_queues;
1222 1223 1224 1225

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

1226 1227 1228
	/* root qdisc from userspace point of view */
	struct Qdisc		*qdisc;

L
Linus Torvalds 已提交
1229
	unsigned long		tx_queue_len;	/* Max frames per queue allowed */
1230
	spinlock_t		tx_global_lock;
E
Eric Dumazet 已提交
1231

T
Tom Herbert 已提交
1232
#ifdef CONFIG_XPS
E
Eric Dumazet 已提交
1233
	struct xps_dev_maps __rcu *xps_maps;
T
Tom Herbert 已提交
1234
#endif
T
Tom Herbert 已提交
1235

1236
	/* These may be needed for future network-power-down code. */
1237 1238 1239 1240 1241

	/*
	 * trans_start here is expensive for high speed devices on SMP,
	 * please use netdev_queue->trans_start instead.
	 */
1242 1243 1244 1245 1246
	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
1247
	/* Number of references to this device */
E
Eric Dumazet 已提交
1248
	int __percpu		*pcpu_refcnt;
1249

L
Linus Torvalds 已提交
1250 1251 1252 1253 1254
	/* delayed register/unregister */
	struct list_head	todo_list;
	/* device index hash chain */
	struct hlist_node	index_hlist;

1255
	struct list_head	link_watch_list;
1256

L
Linus Torvalds 已提交
1257 1258
	/* register/unregister state machine */
	enum { NETREG_UNINITIALIZED=0,
1259
	       NETREG_REGISTERED,	/* completed register_netdevice */
L
Linus Torvalds 已提交
1260 1261 1262
	       NETREG_UNREGISTERING,	/* called unregister_netdevice */
	       NETREG_UNREGISTERED,	/* completed unregister todo */
	       NETREG_RELEASED,		/* called free_netdev */
1263
	       NETREG_DUMMY,		/* dummy device for NAPI poll */
1264 1265 1266
	} reg_state:8;

	bool dismantle; /* device is going do be freed */
1267 1268 1269 1270 1271

	enum {
		RTNL_LINK_INITIALIZED,
		RTNL_LINK_INITIALIZING,
	} rtnl_link_state:16;
L
Linus Torvalds 已提交
1272

1273 1274
	/* Called from unregister, can be used to call free_netdev */
	void (*destructor)(struct net_device *dev);
L
Linus Torvalds 已提交
1275 1276

#ifdef CONFIG_NETPOLL
1277
	struct netpoll_info	*npinfo;
L
Linus Torvalds 已提交
1278
#endif
1279

1280
#ifdef CONFIG_NET_NS
1281 1282
	/* Network namespace this network device is inside */
	struct net		*nd_net;
1283
#endif
1284

D
David S. Miller 已提交
1285
	/* mid-layer private */
E
Eric Dumazet 已提交
1286 1287 1288
	union {
		void				*ml_priv;
		struct pcpu_lstats __percpu	*lstats; /* loopback stats */
1289
		struct pcpu_tstats __percpu	*tstats; /* tunnel stats */
1290
		struct pcpu_dstats __percpu	*dstats; /* dummy stats */
E
Eric Dumazet 已提交
1291
		struct pcpu_vstats __percpu	*vstats; /* veth stats */
E
Eric Dumazet 已提交
1292
	};
1293
	/* GARP */
E
Eric Dumazet 已提交
1294
	struct garp_port __rcu	*garp_port;
L
Linus Torvalds 已提交
1295 1296

	/* class/net/name entry */
1297
	struct device		dev;
1298 1299
	/* space for optional device, statistics, and wireless sysfs groups */
	const struct attribute_group *sysfs_groups[4];
P
Patrick McHardy 已提交
1300 1301 1302

	/* rtnetlink link ops */
	const struct rtnl_link_ops *rtnl_link_ops;
1303

1304 1305 1306
	/* for setting kernel sock attribute on TCP connection setup */
#define GSO_MAX_SIZE		65536
	unsigned int		gso_max_size;
1307 1308
#define GSO_MAX_SEGS		65535
	u16			gso_max_segs;
1309

J
Jeff Kirsher 已提交
1310
#ifdef CONFIG_DCB
1311
	/* Data Center Bridging netlink ops */
1312
	const struct dcbnl_rtnl_ops *dcbnl_ops;
1313
#endif
1314 1315 1316
	u8 num_tc;
	struct netdev_tc_txq tc_to_txq[TC_MAX_QUEUE];
	u8 prio_tc_map[TC_BITMASK + 1];
1317

1318
#if IS_ENABLED(CONFIG_FCOE)
1319 1320
	/* max exchange id for FCoE LRO by ddp */
	unsigned int		fcoe_ddp_xid;
1321 1322 1323
#endif
#if IS_ENABLED(CONFIG_NETPRIO_CGROUP)
	struct netprio_map __rcu *priomap;
1324
#endif
1325 1326
	/* phy device may attach itself for hardware timestamping */
	struct phy_device *phydev;
1327

1328 1329
	struct lock_class_key *qdisc_tx_busylock;

1330 1331
	/* group the device belongs to */
	int group;
1332 1333

	struct pm_qos_request	pm_qos_req;
L
Linus Torvalds 已提交
1334
};
1335
#define to_net_dev(d) container_of(d, struct net_device, dev)
L
Linus Torvalds 已提交
1336 1337 1338

#define	NETDEV_ALIGN		32

1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389
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;
}

1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408
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);
}

1409 1410 1411
extern struct netdev_queue *netdev_pick_tx(struct net_device *dev,
					   struct sk_buff *skb);

1412 1413 1414 1415 1416 1417
/*
 * Net namespace inlines
 */
static inline
struct net *dev_net(const struct net_device *dev)
{
E
Eric Dumazet 已提交
1418
	return read_pnet(&dev->nd_net);
1419 1420 1421
}

static inline
1422
void dev_net_set(struct net_device *dev, struct net *net)
1423 1424
{
#ifdef CONFIG_NET_NS
1425 1426
	release_net(dev->nd_net);
	dev->nd_net = hold_net(net);
1427 1428 1429
#endif
}

1430 1431 1432 1433 1434 1435 1436 1437 1438 1439
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;
}

1440 1441 1442 1443 1444 1445 1446 1447 1448 1449
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;
}

1450 1451 1452 1453 1454 1455
/**
 *	netdev_priv - access network device private data
 *	@dev: network device
 *
 * Get network device private data
 */
1456
static inline void *netdev_priv(const struct net_device *dev)
L
Linus Torvalds 已提交
1457
{
1458
	return (char *)dev + ALIGN(sizeof(struct net_device), NETDEV_ALIGN);
L
Linus Torvalds 已提交
1459 1460 1461 1462 1463
}

/* Set the sysfs physical device reference for the network logical device
 * if set prior to registration will cause a symlink during initialization.
 */
1464
#define SET_NETDEV_DEV(net, pdev)	((net)->dev.parent = (pdev))
L
Linus Torvalds 已提交
1465

1466 1467 1468 1469 1470 1471
/* 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))

1472 1473 1474 1475 1476 1477 1478 1479 1480 1481
/**
 *	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.
 */
1482 1483
void netif_napi_add(struct net_device *dev, struct napi_struct *napi,
		    int (*poll)(struct napi_struct *, int), int weight);
1484

1485 1486 1487 1488 1489 1490
/**
 *  netif_napi_del - remove a napi context
 *  @napi: napi context
 *
 *  netif_napi_del() removes a napi context from the network device napi list
 */
1491 1492 1493
void netif_napi_del(struct napi_struct *napi);

struct napi_gro_cb {
1494 1495 1496
	/* Virtual address of skb_shinfo(skb)->frags[0].page + offset. */
	void *frag0;

1497 1498 1499
	/* Length of frag0. */
	unsigned int frag0_len;

1500 1501 1502
	/* This indicates where we are processing relative to skb->data. */
	int data_offset;

1503 1504 1505 1506
	/* This is non-zero if the packet cannot be merged with the new skb. */
	int flush;

	/* Number of segments aggregated. */
1507 1508 1509 1510
	u16	count;

	/* This is non-zero if the packet may be of the same flow. */
	u8	same_flow;
H
Herbert Xu 已提交
1511 1512

	/* Free the skb? */
1513
	u8	free;
1514 1515
#define NAPI_GRO_FREE		  1
#define NAPI_GRO_FREE_STOLEN_HEAD 2
1516 1517 1518

	/* jiffies when first packet was created/queued */
	unsigned long age;
1519 1520 1521

	/* Used in ipv6_gro_receive() */
	int	proto;
1522 1523 1524

	/* used in skb_gro_receive() slow path */
	struct sk_buff *last;
1525 1526 1527
};

#define NAPI_GRO_CB(skb) ((struct napi_gro_cb *)(skb)->cb)
1528

L
Linus Torvalds 已提交
1529
struct packet_type {
D
David S. Miller 已提交
1530 1531 1532 1533 1534 1535
	__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 *);
1536 1537
	bool			(*id_match)(struct packet_type *ptype,
					    struct sock *sk);
L
Linus Torvalds 已提交
1538 1539 1540 1541
	void			*af_packet_priv;
	struct list_head	list;
};

1542
struct offload_callbacks {
1543
	struct sk_buff		*(*gso_segment)(struct sk_buff *skb,
1544
						netdev_features_t features);
1545
	int			(*gso_send_check)(struct sk_buff *skb);
1546 1547 1548
	struct sk_buff		**(*gro_receive)(struct sk_buff **head,
					       struct sk_buff *skb);
	int			(*gro_complete)(struct sk_buff *skb);
1549 1550 1551 1552 1553 1554
};

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
1555 1556 1557 1558
};

#include <linux/notifier.h>

1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581
/* 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
1582
#define NETDEV_UNREGISTER_FINAL 0x0011
1583 1584 1585 1586 1587 1588 1589 1590 1591
#define NETDEV_RELEASE		0x0012
#define NETDEV_NOTIFY_PEERS	0x0013
#define NETDEV_JOIN		0x0014

extern int register_netdevice_notifier(struct notifier_block *nb);
extern int unregister_netdevice_notifier(struct notifier_block *nb);
extern int call_netdevice_notifiers(unsigned long val, struct net_device *dev);


L
Linus Torvalds 已提交
1592 1593
extern rwlock_t				dev_base_lock;		/* Device list lock */

1594
extern seqcount_t	devnet_rename_seq;	/* Device rename seq */
1595

1596

1597 1598
#define for_each_netdev(net, d)		\
		list_for_each_entry(d, &(net)->dev_base_head, dev_list)
1599 1600
#define for_each_netdev_reverse(net, d)	\
		list_for_each_entry_reverse(d, &(net)->dev_base_head, dev_list)
1601 1602
#define for_each_netdev_rcu(net, d)		\
		list_for_each_entry_rcu(d, &(net)->dev_base_head, dev_list)
1603 1604 1605 1606
#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)
1607 1608
#define for_each_netdev_continue_rcu(net, d)		\
	list_for_each_entry_continue_rcu(d, &(net)->dev_base_head, dev_list)
1609
#define net_device_entry(lh)	list_entry(lh, struct net_device, dev_list)
1610

1611 1612 1613 1614 1615
static inline struct net_device *next_net_device(struct net_device *dev)
{
	struct list_head *lh;
	struct net *net;

1616
	net = dev_net(dev);
1617 1618 1619 1620
	lh = dev->dev_list.next;
	return lh == &net->dev_base_head ? NULL : net_device_entry(lh);
}

1621 1622 1623 1624 1625 1626
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
1627
	lh = rcu_dereference(list_next_rcu(&dev->dev_list));
1628 1629 1630
	return lh == &net->dev_base_head ? NULL : net_device_entry(lh);
}

1631 1632 1633 1634 1635
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);
}
1636

E
Eric Dumazet 已提交
1637 1638 1639 1640 1641 1642 1643
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
1644 1645
extern int 			netdev_boot_setup_check(struct net_device *dev);
extern unsigned long		netdev_boot_base(const char *prefix, int unit);
1646 1647
extern struct net_device *dev_getbyhwaddr_rcu(struct net *net, unsigned short type,
					      const char *hwaddr);
1648 1649
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
1650 1651 1652
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);
1653 1654 1655
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
1656

E
Eric Dumazet 已提交
1657 1658
extern struct net_device	*dev_get_by_flags_rcu(struct net *net, unsigned short flags,
						      unsigned short mask);
1659
extern struct net_device	*dev_get_by_name(struct net *net, const char *name);
1660
extern struct net_device	*dev_get_by_name_rcu(struct net *net, const char *name);
1661
extern struct net_device	*__dev_get_by_name(struct net *net, const char *name);
L
Linus Torvalds 已提交
1662 1663 1664
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);
1665
extern void		dev_disable_lro(struct net_device *dev);
1666
extern int		dev_loopback_xmit(struct sk_buff *newskb);
L
Linus Torvalds 已提交
1667 1668
extern int		dev_queue_xmit(struct sk_buff *skb);
extern int		register_netdevice(struct net_device *dev);
1669 1670
extern void		unregister_netdevice_queue(struct net_device *dev,
						   struct list_head *head);
1671
extern void		unregister_netdevice_many(struct list_head *head);
1672 1673 1674 1675 1676
static inline void unregister_netdevice(struct net_device *dev)
{
	unregister_netdevice_queue(dev, NULL);
}

E
Eric Dumazet 已提交
1677
extern int 		netdev_refcnt_read(const struct net_device *dev);
L
Linus Torvalds 已提交
1678 1679
extern void		free_netdev(struct net_device *dev);
extern void		synchronize_net(void);
1680
extern int		init_dummy_netdev(struct net_device *dev);
1681
extern void		netdev_resync_ops(struct net_device *dev);
1682

1683 1684
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);
1685
extern struct net_device	*dev_get_by_index_rcu(struct net *net, int ifindex);
L
Linus Torvalds 已提交
1686 1687 1688 1689
extern int		dev_restart(struct net_device *dev);
#ifdef CONFIG_NETPOLL_TRAP
extern int		netpoll_trap(void);
#endif
1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707
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;
}

1708 1709
static inline void *skb_gro_header_fast(struct sk_buff *skb,
					unsigned int offset)
1710
{
1711 1712
	return NAPI_GRO_CB(skb)->frag0 + offset;
}
1713

1714 1715 1716 1717
static inline int skb_gro_header_hard(struct sk_buff *skb, unsigned int hlen)
{
	return NAPI_GRO_CB(skb)->frag0_len < hlen;
}
1718

1719 1720 1721
static inline void *skb_gro_header_slow(struct sk_buff *skb, unsigned int hlen,
					unsigned int offset)
{
1722 1723 1724
	if (!pskb_may_pull(skb, hlen))
		return NULL;

1725 1726
	NAPI_GRO_CB(skb)->frag0 = NULL;
	NAPI_GRO_CB(skb)->frag0_len = 0;
1727
	return skb->data + offset;
1728
}
L
Linus Torvalds 已提交
1729

1730 1731
static inline void *skb_gro_mac_header(struct sk_buff *skb)
{
1732
	return NAPI_GRO_CB(skb)->frag0 ?: skb_mac_header(skb);
1733 1734
}

H
Herbert Xu 已提交
1735 1736
static inline void *skb_gro_network_header(struct sk_buff *skb)
{
1737 1738
	return (NAPI_GRO_CB(skb)->frag0 ?: skb->data) +
	       skb_network_offset(skb);
H
Herbert Xu 已提交
1739 1740
}

1741 1742
static inline int dev_hard_header(struct sk_buff *skb, struct net_device *dev,
				  unsigned short type,
1743
				  const void *daddr, const void *saddr,
1744
				  unsigned int len)
1745
{
1746
	if (!dev->header_ops || !dev->header_ops->create)
1747
		return 0;
1748 1749

	return dev->header_ops->create(skb, dev, type, daddr, saddr, len);
1750 1751
}

S
Stephen Hemminger 已提交
1752 1753 1754 1755 1756
static inline int dev_parse_header(const struct sk_buff *skb,
				   unsigned char *haddr)
{
	const struct net_device *dev = skb->dev;

1757
	if (!dev->header_ops || !dev->header_ops->parse)
S
Stephen Hemminger 已提交
1758
		return 0;
1759
	return dev->header_ops->parse(skb, haddr);
S
Stephen Hemminger 已提交
1760 1761
}

L
Linus Torvalds 已提交
1762 1763 1764 1765 1766 1767 1768 1769
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);
}

/*
E
Eric Dumazet 已提交
1770
 * Incoming packets are placed on per-cpu queues
L
Linus Torvalds 已提交
1771
 */
E
Eric Dumazet 已提交
1772
struct softnet_data {
1773
	struct Qdisc		*output_queue;
1774
	struct Qdisc		**output_queue_tailp;
L
Linus Torvalds 已提交
1775 1776
	struct list_head	poll_list;
	struct sk_buff		*completion_queue;
1777
	struct sk_buff_head	process_queue;
L
Linus Torvalds 已提交
1778

C
Changli Gao 已提交
1779
	/* stats */
1780 1781 1782 1783
	unsigned int		processed;
	unsigned int		time_squeeze;
	unsigned int		cpu_collision;
	unsigned int		received_rps;
C
Changli Gao 已提交
1784

1785
#ifdef CONFIG_RPS
E
Eric Dumazet 已提交
1786 1787 1788
	struct softnet_data	*rps_ipi_list;

	/* Elements below can be accessed between CPUs for RPS */
T
Tom Herbert 已提交
1789
	struct call_single_data	csd ____cacheline_aligned_in_smp;
E
Eric Dumazet 已提交
1790 1791
	struct softnet_data	*rps_ipi_next;
	unsigned int		cpu;
T
Tom Herbert 已提交
1792
	unsigned int		input_queue_head;
1793
	unsigned int		input_queue_tail;
1794
#endif
1795
	unsigned int		dropped;
T
Tom Herbert 已提交
1796
	struct sk_buff_head	input_pkt_queue;
1797
	struct napi_struct	backlog;
L
Linus Torvalds 已提交
1798 1799
};

1800
static inline void input_queue_head_incr(struct softnet_data *sd)
T
Tom Herbert 已提交
1801 1802
{
#ifdef CONFIG_RPS
1803 1804 1805 1806 1807 1808 1809 1810 1811
	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
1812 1813 1814
#endif
}

T
Tom Herbert 已提交
1815
DECLARE_PER_CPU_ALIGNED(struct softnet_data, softnet_data);
L
Linus Torvalds 已提交
1816

1817
extern void __netif_schedule(struct Qdisc *q);
L
Linus Torvalds 已提交
1818

1819
static inline void netif_schedule_queue(struct netdev_queue *txq)
L
Linus Torvalds 已提交
1820
{
1821
	if (!(txq->state & QUEUE_STATE_ANY_XOFF))
1822
		__netif_schedule(txq->qdisc);
1823 1824
}

1825 1826 1827 1828 1829 1830 1831 1832
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));
}

1833 1834
static inline void netif_tx_start_queue(struct netdev_queue *dev_queue)
{
1835
	clear_bit(__QUEUE_STATE_DRV_XOFF, &dev_queue->state);
1836 1837
}

1838 1839 1840 1841 1842 1843
/**
 *	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
1844 1845
static inline void netif_start_queue(struct net_device *dev)
{
1846
	netif_tx_start_queue(netdev_get_tx_queue(dev, 0));
L
Linus Torvalds 已提交
1847 1848
}

1849 1850 1851 1852 1853 1854 1855 1856 1857 1858
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);
	}
}

1859
static inline void netif_tx_wake_queue(struct netdev_queue *dev_queue)
L
Linus Torvalds 已提交
1860 1861
{
#ifdef CONFIG_NETPOLL_TRAP
1862
	if (netpoll_trap()) {
1863
		netif_tx_start_queue(dev_queue);
L
Linus Torvalds 已提交
1864
		return;
1865
	}
L
Linus Torvalds 已提交
1866
#endif
1867
	if (test_and_clear_bit(__QUEUE_STATE_DRV_XOFF, &dev_queue->state))
1868
		__netif_schedule(dev_queue->qdisc);
1869 1870
}

1871 1872 1873 1874 1875 1876 1877
/**
 *	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.
 */
1878 1879
static inline void netif_wake_queue(struct net_device *dev)
{
1880
	netif_tx_wake_queue(netdev_get_tx_queue(dev, 0));
L
Linus Torvalds 已提交
1881 1882
}

1883 1884 1885 1886 1887 1888 1889 1890 1891 1892
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);
	}
}

1893 1894
static inline void netif_tx_stop_queue(struct netdev_queue *dev_queue)
{
1895
	if (WARN_ON(!dev_queue)) {
1896
		pr_info("netif_stop_queue() cannot be called before register_netdev()\n");
1897 1898
		return;
	}
1899
	set_bit(__QUEUE_STATE_DRV_XOFF, &dev_queue->state);
1900 1901
}

1902 1903 1904 1905 1906 1907 1908
/**
 *	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
1909 1910
static inline void netif_stop_queue(struct net_device *dev)
{
1911
	netif_tx_stop_queue(netdev_get_tx_queue(dev, 0));
L
Linus Torvalds 已提交
1912 1913
}

1914 1915 1916 1917 1918 1919 1920 1921 1922 1923
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);
	}
}

1924
static inline bool netif_tx_queue_stopped(const struct netdev_queue *dev_queue)
1925
{
1926
	return test_bit(__QUEUE_STATE_DRV_XOFF, &dev_queue->state);
1927 1928
}

1929 1930 1931 1932 1933 1934
/**
 *	netif_queue_stopped - test if transmit queue is flowblocked
 *	@dev: network device
 *
 *	Test if transmit queue on device is currently unable to send.
 */
1935
static inline bool netif_queue_stopped(const struct net_device *dev)
L
Linus Torvalds 已提交
1936
{
1937
	return netif_tx_queue_stopped(netdev_get_tx_queue(dev, 0));
L
Linus Torvalds 已提交
1938 1939
}

1940
static inline bool netif_xmit_stopped(const struct netdev_queue *dev_queue)
1941
{
1942 1943 1944
	return dev_queue->state & QUEUE_STATE_ANY_XOFF;
}

1945
static inline bool netif_xmit_frozen_or_stopped(const struct netdev_queue *dev_queue)
1946 1947 1948 1949
{
	return dev_queue->state & QUEUE_STATE_ANY_XOFF_OR_FROZEN;
}

1950 1951 1952
static inline void netdev_tx_sent_queue(struct netdev_queue *dev_queue,
					unsigned int bytes)
{
T
Tom Herbert 已提交
1953 1954
#ifdef CONFIG_BQL
	dql_queued(&dev_queue->dql, bytes);
1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970

	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
1971
#endif
1972 1973 1974 1975 1976 1977 1978 1979
}

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,
1980
					     unsigned int pkts, unsigned int bytes)
1981
{
T
Tom Herbert 已提交
1982
#ifdef CONFIG_BQL
1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999
	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
2000
#endif
2001 2002 2003
}

static inline void netdev_completed_queue(struct net_device *dev,
2004
					  unsigned int pkts, unsigned int bytes)
2005 2006 2007 2008 2009 2010
{
	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
2011
#ifdef CONFIG_BQL
2012
	clear_bit(__QUEUE_STATE_STACK_XOFF, &q->state);
T
Tom Herbert 已提交
2013 2014
	dql_reset(&q->dql);
#endif
2015 2016 2017 2018 2019
}

static inline void netdev_reset_queue(struct net_device *dev_queue)
{
	netdev_tx_reset_queue(netdev_get_tx_queue(dev_queue, 0));
2020 2021
}

2022 2023 2024 2025 2026 2027
/**
 *	netif_running - test if up
 *	@dev: network device
 *
 *	Test if the device has been brought up.
 */
2028
static inline bool netif_running(const struct net_device *dev)
L
Linus Torvalds 已提交
2029 2030 2031 2032
{
	return test_bit(__LINK_STATE_START, &dev->state);
}

2033 2034 2035 2036 2037 2038
/*
 * 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.
 */
2039 2040 2041 2042 2043 2044 2045 2046

/**
 *	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.
 */
2047 2048
static inline void netif_start_subqueue(struct net_device *dev, u16 queue_index)
{
2049
	struct netdev_queue *txq = netdev_get_tx_queue(dev, queue_index);
2050 2051

	netif_tx_start_queue(txq);
2052 2053
}

2054 2055 2056 2057 2058 2059 2060
/**
 *	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.
 */
2061 2062
static inline void netif_stop_subqueue(struct net_device *dev, u16 queue_index)
{
2063
	struct netdev_queue *txq = netdev_get_tx_queue(dev, queue_index);
2064 2065 2066 2067
#ifdef CONFIG_NETPOLL_TRAP
	if (netpoll_trap())
		return;
#endif
2068
	netif_tx_stop_queue(txq);
2069 2070
}

2071 2072 2073 2074 2075 2076 2077
/**
 *	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.
 */
2078 2079
static inline bool __netif_subqueue_stopped(const struct net_device *dev,
					    u16 queue_index)
2080
{
2081
	struct netdev_queue *txq = netdev_get_tx_queue(dev, queue_index);
2082 2083

	return netif_tx_queue_stopped(txq);
2084 2085
}

2086 2087
static inline bool netif_subqueue_stopped(const struct net_device *dev,
					  struct sk_buff *skb)
2088 2089 2090
{
	return __netif_subqueue_stopped(dev, skb_get_queue_mapping(skb));
}
2091 2092 2093 2094 2095 2096 2097 2098

/**
 *	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.
 */
2099 2100
static inline void netif_wake_subqueue(struct net_device *dev, u16 queue_index)
{
2101
	struct netdev_queue *txq = netdev_get_tx_queue(dev, queue_index);
2102 2103 2104 2105
#ifdef CONFIG_NETPOLL_TRAP
	if (netpoll_trap())
		return;
#endif
2106
	if (test_and_clear_bit(__QUEUE_STATE_DRV_XOFF, &txq->state))
2107
		__netif_schedule(txq->qdisc);
2108 2109
}

2110 2111 2112 2113 2114 2115 2116 2117 2118 2119
/*
 * 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);
}

2120 2121 2122 2123 2124 2125
/**
 *	netif_is_multiqueue - test if device has multiple transmit queues
 *	@dev: network device
 *
 * Check if device has multiple transmit queues
 */
2126
static inline bool netif_is_multiqueue(const struct net_device *dev)
2127
{
E
Eric Dumazet 已提交
2128
	return dev->num_tx_queues > 1;
2129
}
L
Linus Torvalds 已提交
2130

2131 2132
extern int netif_set_real_num_tx_queues(struct net_device *dev,
					unsigned int txq);
2133

2134 2135 2136 2137 2138 2139 2140 2141 2142 2143 2144
#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

2145 2146 2147
static inline int netif_copy_real_num_queues(struct net_device *to_dev,
					     const struct net_device *from_dev)
{
2148 2149 2150 2151 2152 2153
	int err;

	err = netif_set_real_num_tx_queues(to_dev,
					   from_dev->real_num_tx_queues);
	if (err)
		return err;
2154 2155 2156 2157 2158 2159 2160 2161
#ifdef CONFIG_RPS
	return netif_set_real_num_rx_queues(to_dev,
					    from_dev->real_num_rx_queues);
#else
	return 0;
#endif
}

2162 2163 2164
#define DEFAULT_MAX_NUM_RSS_QUEUES	(8)
extern int netif_get_num_default_rss_queues(void);

L
Linus Torvalds 已提交
2165
/* Use this variant when it is known for sure that it
2166 2167
 * is executing from hardware interrupt context or with hardware interrupts
 * disabled.
L
Linus Torvalds 已提交
2168
 */
2169
extern void dev_kfree_skb_irq(struct sk_buff *skb);
L
Linus Torvalds 已提交
2170 2171

/* Use this variant in places where it could be invoked
2172 2173
 * from either hardware interrupt or other context, with hardware interrupts
 * either disabled or enabled.
L
Linus Torvalds 已提交
2174
 */
2175
extern void dev_kfree_skb_any(struct sk_buff *skb);
L
Linus Torvalds 已提交
2176 2177 2178 2179

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);
2180
extern gro_result_t	napi_gro_receive(struct napi_struct *napi,
2181
					 struct sk_buff *skb);
2182
extern void		napi_gro_flush(struct napi_struct *napi, bool flush_old);
2183
extern struct sk_buff *	napi_get_frags(struct napi_struct *napi);
2184
extern gro_result_t	napi_gro_frags(struct napi_struct *napi);
2185 2186 2187 2188 2189 2190 2191

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

2192
extern int netdev_rx_handler_register(struct net_device *dev,
J
Jiri Pirko 已提交
2193 2194
				      rx_handler_func_t *rx_handler,
				      void *rx_handler_data);
2195 2196
extern void netdev_rx_handler_unregister(struct net_device *dev);

2197
extern bool		dev_valid_name(const char *name);
2198 2199
extern int		dev_ioctl(struct net *net, unsigned int cmd, void __user *);
extern int		dev_ethtool(struct net *net, struct ifreq *);
2200
extern unsigned int	dev_get_flags(const struct net_device *);
2201
extern int		__dev_change_flags(struct net_device *, unsigned int flags);
2202
extern int		dev_change_flags(struct net_device *, unsigned int);
2203
extern void		__dev_notify_flags(struct net_device *, unsigned int old_flags);
2204
extern int		dev_change_name(struct net_device *, const char *);
2205
extern int		dev_set_alias(struct net_device *, const char *, size_t);
2206 2207
extern int		dev_change_net_namespace(struct net_device *,
						 struct net *, const char *);
L
Linus Torvalds 已提交
2208
extern int		dev_set_mtu(struct net_device *, int);
2209
extern void		dev_set_group(struct net_device *, int);
L
Linus Torvalds 已提交
2210 2211
extern int		dev_set_mac_address(struct net_device *,
					    struct sockaddr *);
J
Jiri Pirko 已提交
2212 2213
extern int		dev_change_carrier(struct net_device *,
					   bool new_carrier);
2214
extern int		dev_hard_start_xmit(struct sk_buff *skb,
2215 2216
					    struct net_device *dev,
					    struct netdev_queue *txq);
2217 2218
extern int		dev_forward_skb(struct net_device *dev,
					struct sk_buff *skb);
L
Linus Torvalds 已提交
2219

2220
extern int		netdev_budget;
L
Linus Torvalds 已提交
2221 2222 2223 2224

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

2225 2226 2227 2228
/**
 *	dev_put - release reference to device
 *	@dev: network device
 *
2229
 * Release reference to device to allow it to be freed.
2230
 */
L
Linus Torvalds 已提交
2231 2232
static inline void dev_put(struct net_device *dev)
{
2233
	this_cpu_dec(*dev->pcpu_refcnt);
L
Linus Torvalds 已提交
2234 2235
}

2236 2237 2238 2239
/**
 *	dev_hold - get reference to device
 *	@dev: network device
 *
2240
 * Hold reference to device to keep it from being freed.
2241
 */
2242 2243
static inline void dev_hold(struct net_device *dev)
{
2244
	this_cpu_inc(*dev->pcpu_refcnt);
2245
}
L
Linus Torvalds 已提交
2246 2247 2248 2249

/* 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
2250 2251 2252 2253
 *
 * 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
2254 2255
 */

2256
extern void linkwatch_init_dev(struct net_device *dev);
L
Linus Torvalds 已提交
2257
extern void linkwatch_fire_event(struct net_device *dev);
2258
extern void linkwatch_forget_dev(struct net_device *dev);
L
Linus Torvalds 已提交
2259

2260 2261 2262 2263 2264 2265
/**
 *	netif_carrier_ok - test if carrier present
 *	@dev: network device
 *
 * Check if carrier is present on device
 */
2266
static inline bool netif_carrier_ok(const struct net_device *dev)
L
Linus Torvalds 已提交
2267 2268 2269 2270
{
	return !test_bit(__LINK_STATE_NOCARRIER, &dev->state);
}

2271 2272
extern unsigned long dev_trans_start(struct net_device *dev);

L
Linus Torvalds 已提交
2273 2274
extern void __netdev_watchdog_up(struct net_device *dev);

2275
extern void netif_carrier_on(struct net_device *dev);
L
Linus Torvalds 已提交
2276

2277
extern void netif_carrier_off(struct net_device *dev);
L
Linus Torvalds 已提交
2278

2279 2280 2281 2282 2283 2284 2285 2286 2287 2288 2289 2290 2291
/**
 *	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
2292 2293 2294 2295 2296 2297
static inline void netif_dormant_on(struct net_device *dev)
{
	if (!test_and_set_bit(__LINK_STATE_DORMANT, &dev->state))
		linkwatch_fire_event(dev);
}

2298 2299 2300 2301 2302 2303
/**
 *	netif_dormant_off - set device as not dormant.
 *	@dev: network device
 *
 * Device is not in dormant state.
 */
S
Stefan Rompf 已提交
2304 2305 2306 2307 2308 2309
static inline void netif_dormant_off(struct net_device *dev)
{
	if (test_and_clear_bit(__LINK_STATE_DORMANT, &dev->state))
		linkwatch_fire_event(dev);
}

2310 2311 2312 2313 2314 2315
/**
 *	netif_dormant - test if carrier present
 *	@dev: network device
 *
 * Check if carrier is present on device
 */
2316
static inline bool netif_dormant(const struct net_device *dev)
S
Stefan Rompf 已提交
2317 2318 2319 2320 2321
{
	return test_bit(__LINK_STATE_DORMANT, &dev->state);
}


2322 2323 2324 2325 2326 2327
/**
 *	netif_oper_up - test if device is operational
 *	@dev: network device
 *
 * Check if carrier is operational
 */
2328
static inline bool netif_oper_up(const struct net_device *dev)
E
Eric Dumazet 已提交
2329
{
S
Stefan Rompf 已提交
2330 2331 2332 2333
	return (dev->operstate == IF_OPER_UP ||
		dev->operstate == IF_OPER_UNKNOWN /* backward compat */);
}

2334 2335 2336 2337 2338 2339
/**
 *	netif_device_present - is device available or removed
 *	@dev: network device
 *
 * Check if device has not been removed from system.
 */
2340
static inline bool netif_device_present(struct net_device *dev)
L
Linus Torvalds 已提交
2341 2342 2343 2344
{
	return test_bit(__LINK_STATE_PRESENT, &dev->state);
}

2345
extern void netif_device_detach(struct net_device *dev);
L
Linus Torvalds 已提交
2346

2347
extern void netif_device_attach(struct net_device *dev);
L
Linus Torvalds 已提交
2348 2349 2350 2351 2352 2353 2354 2355 2356 2357 2358 2359 2360 2361 2362 2363 2364 2365 2366 2367 2368 2369 2370 2371 2372 2373 2374 2375 2376 2377 2378 2379 2380 2381 2382 2383 2384 2385 2386 2387 2388 2389 2390 2391 2392 2393 2394 2395 2396 2397

/*
 * 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;
}

2398
static inline void __netif_tx_lock(struct netdev_queue *txq, int cpu)
H
Herbert Xu 已提交
2399
{
2400 2401
	spin_lock(&txq->_xmit_lock);
	txq->xmit_lock_owner = cpu;
2402 2403
}

2404 2405 2406 2407 2408 2409
static inline void __netif_tx_lock_bh(struct netdev_queue *txq)
{
	spin_lock_bh(&txq->_xmit_lock);
	txq->xmit_lock_owner = smp_processor_id();
}

2410
static inline bool __netif_tx_trylock(struct netdev_queue *txq)
2411
{
2412
	bool ok = spin_trylock(&txq->_xmit_lock);
2413 2414 2415 2416 2417 2418 2419 2420 2421 2422 2423 2424 2425 2426 2427 2428 2429
	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
2430 2431 2432 2433 2434 2435
static inline void txq_trans_update(struct netdev_queue *txq)
{
	if (txq->xmit_lock_owner != -1)
		txq->trans_start = jiffies;
}

2436 2437 2438 2439 2440 2441
/**
 *	netif_tx_lock - grab network device transmit lock
 *	@dev: network device
 *
 * Get network device transmit lock
 */
2442 2443
static inline void netif_tx_lock(struct net_device *dev)
{
2444
	unsigned int i;
2445
	int cpu;
2446

2447 2448
	spin_lock(&dev->tx_global_lock);
	cpu = smp_processor_id();
2449 2450
	for (i = 0; i < dev->num_tx_queues; i++) {
		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
2451 2452 2453 2454 2455 2456 2457

		/* 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.
		 */
2458
		__netif_tx_lock(txq, cpu);
2459 2460
		set_bit(__QUEUE_STATE_FROZEN, &txq->state);
		__netif_tx_unlock(txq);
2461
	}
H
Herbert Xu 已提交
2462 2463 2464 2465
}

static inline void netif_tx_lock_bh(struct net_device *dev)
{
2466 2467
	local_bh_disable();
	netif_tx_lock(dev);
H
Herbert Xu 已提交
2468 2469 2470 2471
}

static inline void netif_tx_unlock(struct net_device *dev)
{
2472 2473 2474 2475
	unsigned int i;

	for (i = 0; i < dev->num_tx_queues; i++) {
		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
2476

2477 2478 2479 2480 2481
		/* 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);
2482
		netif_schedule_queue(txq);
2483 2484
	}
	spin_unlock(&dev->tx_global_lock);
H
Herbert Xu 已提交
2485 2486 2487 2488
}

static inline void netif_tx_unlock_bh(struct net_device *dev)
{
2489 2490
	netif_tx_unlock(dev);
	local_bh_enable();
H
Herbert Xu 已提交
2491 2492
}

2493
#define HARD_TX_LOCK(dev, txq, cpu) {			\
2494
	if ((dev->features & NETIF_F_LLTX) == 0) {	\
2495
		__netif_tx_lock(txq, cpu);		\
2496 2497 2498
	}						\
}

2499
#define HARD_TX_UNLOCK(dev, txq) {			\
2500
	if ((dev->features & NETIF_F_LLTX) == 0) {	\
2501
		__netif_tx_unlock(txq);			\
2502 2503 2504
	}						\
}

L
Linus Torvalds 已提交
2505 2506
static inline void netif_tx_disable(struct net_device *dev)
{
2507
	unsigned int i;
2508
	int cpu;
2509

2510 2511
	local_bh_disable();
	cpu = smp_processor_id();
2512 2513
	for (i = 0; i < dev->num_tx_queues; i++) {
		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
2514 2515

		__netif_tx_lock(txq, cpu);
2516
		netif_tx_stop_queue(txq);
2517
		__netif_tx_unlock(txq);
2518
	}
2519
	local_bh_enable();
L
Linus Torvalds 已提交
2520 2521
}

2522 2523 2524 2525 2526
static inline void netif_addr_lock(struct net_device *dev)
{
	spin_lock(&dev->addr_list_lock);
}

2527 2528 2529 2530 2531
static inline void netif_addr_lock_nested(struct net_device *dev)
{
	spin_lock_nested(&dev->addr_list_lock, SINGLE_DEPTH_NESTING);
}

2532 2533 2534 2535 2536 2537 2538 2539 2540 2541 2542 2543 2544 2545 2546
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);
}

2547
/*
2548
 * dev_addrs walker. Should be used only for read access. Call with
2549 2550 2551
 * rcu_read_lock held.
 */
#define for_each_dev_addr(dev, ha) \
2552
		list_for_each_entry_rcu(ha, &dev->dev_addrs.list, list)
2553

L
Linus Torvalds 已提交
2554 2555 2556 2557 2558
/* 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
2559
extern struct net_device *alloc_netdev_mqs(int sizeof_priv, const char *name,
2560
				       void (*setup)(struct net_device *),
T
Tom Herbert 已提交
2561
				       unsigned int txqs, unsigned int rxqs);
2562
#define alloc_netdev(sizeof_priv, name, setup) \
T
Tom Herbert 已提交
2563 2564 2565 2566 2567
	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
2568 2569
extern int		register_netdev(struct net_device *dev);
extern void		unregister_netdev(struct net_device *dev);
2570

2571 2572 2573 2574 2575 2576 2577 2578 2579 2580 2581 2582 2583 2584 2585 2586
/* 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);

2587
/* Functions used for device addresses handling */
2588
extern int dev_addr_add(struct net_device *dev, const unsigned char *addr,
2589
			unsigned char addr_type);
2590
extern int dev_addr_del(struct net_device *dev, const unsigned char *addr,
2591 2592 2593 2594 2595 2596 2597
			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);
2598 2599 2600 2601
extern void dev_addr_flush(struct net_device *dev);
extern int dev_addr_init(struct net_device *dev);

/* Functions used for unicast addresses handling */
2602 2603 2604
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);
2605 2606 2607 2608
extern int dev_uc_sync(struct net_device *to, struct net_device *from);
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);
2609

2610
/* Functions used for multicast addresses handling */
2611 2612 2613 2614 2615
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);
2616 2617 2618 2619
extern int dev_mc_sync(struct net_device *to, struct net_device *from);
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);
2620

2621 2622 2623
/* 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);
2624 2625
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
2626
extern void		netdev_state_change(struct net_device *dev);
2627
extern void		netdev_notify_peers(struct net_device *dev);
2628
extern void		netdev_features_change(struct net_device *dev);
L
Linus Torvalds 已提交
2629
/* Load a device via the kmod */
2630
extern void		dev_load(struct net *net, const char *name);
L
Linus Torvalds 已提交
2631
extern void		dev_mcast_init(void);
2632 2633
extern struct rtnl_link_stats64 *dev_get_stats(struct net_device *dev,
					       struct rtnl_link_stats64 *storage);
2634 2635
extern void netdev_stats_to_stats64(struct rtnl_link_stats64 *stats64,
				    const struct net_device_stats *netdev_stats);
2636

L
Linus Torvalds 已提交
2637
extern int		netdev_max_backlog;
E
Eric Dumazet 已提交
2638
extern int		netdev_tstamp_prequeue;
L
Linus Torvalds 已提交
2639
extern int		weight_p;
2640
extern int		bpf_jit_enable;
J
Jiri Pirko 已提交
2641 2642 2643 2644 2645 2646 2647 2648 2649 2650 2651 2652

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);
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);
extern void netdev_upper_dev_unlink(struct net_device *dev,
				    struct net_device *upper_dev);
L
Linus Torvalds 已提交
2653
extern int		netdev_set_master(struct net_device *dev, struct net_device *master);
J
Jiri Pirko 已提交
2654 2655
extern int netdev_set_bond_master(struct net_device *dev,
				  struct net_device *master);
2656
extern int skb_checksum_help(struct sk_buff *skb);
2657 2658
extern struct sk_buff *skb_gso_segment(struct sk_buff *skb,
	netdev_features_t features);
2659 2660 2661 2662 2663 2664 2665
#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
2666 2667 2668 2669
/* rx skb timestamps */
extern void		net_enable_timestamp(void);
extern void		net_disable_timestamp(void);

2670 2671 2672 2673 2674 2675
#ifdef CONFIG_PROC_FS
extern void *dev_seq_start(struct seq_file *seq, loff_t *pos);
extern void *dev_seq_next(struct seq_file *seq, void *v, loff_t *pos);
extern void dev_seq_stop(struct seq_file *seq, void *v);
#endif

2676 2677 2678
extern int netdev_class_create_file(struct class_attribute *class_attr);
extern void netdev_class_remove_file(struct class_attribute *class_attr);

2679 2680
extern struct kobj_ns_type_operations net_ns_type_operations;

2681
extern const char *netdev_drivername(const struct net_device *dev);
2682

2683 2684
extern void linkwatch_run_queue(void);

2685 2686
static inline netdev_features_t netdev_get_wanted_features(
	struct net_device *dev)
2687 2688 2689
{
	return (dev->features & ~dev->hw_features) | dev->wanted_features;
}
2690 2691
netdev_features_t netdev_increment_features(netdev_features_t all,
	netdev_features_t one, netdev_features_t mask);
2692
int __netdev_update_features(struct net_device *dev);
2693
void netdev_update_features(struct net_device *dev);
2694
void netdev_change_features(struct net_device *dev);
2695

2696 2697 2698
void netif_stacked_transfer_operstate(const struct net_device *rootdev,
					struct net_device *dev);

2699
netdev_features_t netif_skb_features(struct sk_buff *skb);
2700

2701
static inline bool net_gso_ok(netdev_features_t features, int gso_type)
2702
{
2703
	netdev_features_t feature = gso_type << NETIF_F_GSO_SHIFT;
2704 2705 2706 2707 2708 2709 2710 2711 2712

	/* 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));

2713
	return (features & feature) == feature;
2714 2715
}

2716
static inline bool skb_gso_ok(struct sk_buff *skb, netdev_features_t features)
2717
{
H
Herbert Xu 已提交
2718
	return net_gso_ok(features, skb_shinfo(skb)->gso_type) &&
2719
	       (!skb_has_frag_list(skb) || (features & NETIF_F_FRAGLIST));
2720 2721
}

2722 2723
static inline bool netif_needs_gso(struct sk_buff *skb,
				   netdev_features_t features)
2724
{
2725
	return skb_is_gso(skb) && (!skb_gso_ok(skb, features) ||
2726 2727
		unlikely((skb->ip_summed != CHECKSUM_PARTIAL) &&
			 (skb->ip_summed != CHECKSUM_UNNECESSARY)));
2728 2729
}

2730 2731 2732 2733 2734 2735
static inline void netif_set_gso_max_size(struct net_device *dev,
					  unsigned int size)
{
	dev->gso_max_size = size;
}

2736
static inline bool netif_is_bond_slave(struct net_device *dev)
J
Jiri Pirko 已提交
2737 2738 2739 2740
{
	return dev->flags & IFF_SLAVE && dev->priv_flags & IFF_BONDING;
}

2741 2742 2743 2744 2745
static inline bool netif_supports_nofcs(struct net_device *dev)
{
	return dev->priv_flags & IFF_SUPP_NOFCS;
}

2746
extern struct pernet_operations __net_initdata loopback_net_ops;
2747

2748 2749 2750 2751 2752 2753 2754 2755 2756 2757 2758
/* 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;
}

2759 2760 2761 2762 2763 2764 2765 2766 2767 2768 2769 2770 2771 2772 2773 2774 2775
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, ...);
2776

2777 2778 2779
#define MODULE_ALIAS_NETDEV(device) \
	MODULE_ALIAS("netdev-" device)

2780
#if defined(CONFIG_DYNAMIC_DEBUG)
2781 2782
#define netdev_dbg(__dev, format, args...)			\
do {								\
2783
	dynamic_netdev_dbg(__dev, format, ##args);		\
2784
} while (0)
2785 2786 2787
#elif defined(DEBUG)
#define netdev_dbg(__dev, format, args...)			\
	netdev_printk(KERN_DEBUG, __dev, format, ##args)
2788 2789 2790 2791 2792 2793 2794 2795 2796 2797 2798 2799 2800 2801 2802 2803 2804 2805 2806 2807 2808 2809 2810 2811 2812 2813 2814 2815 2816
#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);

2817 2818 2819 2820 2821 2822 2823 2824
/* 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)

2825 2826 2827 2828 2829 2830
#define netif_level(level, priv, type, dev, fmt, args...)	\
do {								\
	if (netif_msg_##type(priv))				\
		netdev_##level(dev, fmt, ##args);		\
} while (0)

2831
#define netif_emerg(priv, type, dev, fmt, args...)		\
2832
	netif_level(emerg, priv, type, dev, fmt, ##args)
2833
#define netif_alert(priv, type, dev, fmt, args...)		\
2834
	netif_level(alert, priv, type, dev, fmt, ##args)
2835
#define netif_crit(priv, type, dev, fmt, args...)		\
2836
	netif_level(crit, priv, type, dev, fmt, ##args)
2837
#define netif_err(priv, type, dev, fmt, args...)		\
2838
	netif_level(err, priv, type, dev, fmt, ##args)
2839
#define netif_warn(priv, type, dev, fmt, args...)		\
2840
	netif_level(warn, priv, type, dev, fmt, ##args)
2841
#define netif_notice(priv, type, dev, fmt, args...)		\
2842
	netif_level(notice, priv, type, dev, fmt, ##args)
2843
#define netif_info(priv, type, dev, fmt, args...)		\
2844
	netif_level(info, priv, type, dev, fmt, ##args)
2845

2846
#if defined(CONFIG_DYNAMIC_DEBUG)
2847 2848 2849
#define netif_dbg(priv, type, netdev, format, args...)		\
do {								\
	if (netif_msg_##type(priv))				\
2850
		dynamic_netdev_dbg(netdev, format, ##args);	\
2851
} while (0)
2852 2853 2854
#elif defined(DEBUG)
#define netif_dbg(priv, type, dev, format, args...)		\
	netif_printk(priv, type, KERN_DEBUG, dev, format, ##args)
2855 2856 2857 2858 2859 2860 2861 2862 2863 2864
#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)
2865
#define netif_vdbg	netif_dbg
2866 2867 2868 2869
#else
#define netif_vdbg(priv, type, dev, format, args...)		\
({								\
	if (0)							\
2870
		netif_printk(priv, type, KERN_DEBUG, dev, format, ##args); \
2871 2872 2873
	0;							\
})
#endif
2874

2875
#endif	/* _LINUX_NETDEVICE_H */