netdevice.h 97.1 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
void netdev_set_default_ethtool_ops(struct net_device *dev,
				    const struct ethtool_ops *ops);
65

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

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

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

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

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

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

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

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

	return false;
}

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

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

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

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

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


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

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

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

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

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

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

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

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

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

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

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

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

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

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


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

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

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

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

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

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

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

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

397
void __napi_schedule(struct napi_struct *n);
398

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

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

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

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

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

E
Eliezer Tamir 已提交
451 452 453 454 455 456 457
/**
 *	napi_by_id - lookup a NAPI by napi_id
 *	@napi_id: hashed napi_id
 *
 * lookup @napi_id in napi_hash table
 * must be called under rcu_read_lock()
 */
458
struct napi_struct *napi_by_id(unsigned int napi_id);
E
Eliezer Tamir 已提交
459 460 461 462 463 464 465

/**
 *	napi_hash_add - add a NAPI to global hashtable
 *	@napi: napi context
 *
 * generate a new napi_id and store a @napi under it in napi_hash
 */
466
void napi_hash_add(struct napi_struct *napi);
E
Eliezer Tamir 已提交
467 468 469 470 471 472 473 474

/**
 *	napi_hash_del - remove a NAPI from global table
 *	@napi: napi context
 *
 * Warning: caller must observe rcu grace period
 * before freeing memory containing @napi
 */
475
void napi_hash_del(struct napi_struct *napi);
E
Eliezer Tamir 已提交
476

477 478 479 480 481 482 483 484 485
/**
 *	napi_disable - prevent NAPI from scheduling
 *	@n: napi context
 *
 * Stop NAPI from being scheduled on this context.
 * Waits till any outstanding processing completes.
 */
static inline void napi_disable(struct napi_struct *n)
{
486
	might_sleep();
D
David S. Miller 已提交
487
	set_bit(NAPI_STATE_DISABLE, &n->state);
488
	while (test_and_set_bit(NAPI_STATE_SCHED, &n->state))
489
		msleep(1);
D
David S. Miller 已提交
490
	clear_bit(NAPI_STATE_DISABLE, &n->state);
491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506
}

/**
 *	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);
}

507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524
#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
525
enum netdev_queue_state_t {
526 527
	__QUEUE_STATE_DRV_XOFF,
	__QUEUE_STATE_STACK_XOFF,
528
	__QUEUE_STATE_FROZEN,
529 530 531 532
#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))
533
};
534 535 536 537 538 539 540 541 542
/*
 * __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_*.
 */
543

544
struct netdev_queue {
545 546 547
/*
 * read mostly part
 */
548
	struct net_device	*dev;
549 550
	struct Qdisc		*qdisc;
	struct Qdisc		*qdisc_sleeping;
551
#ifdef CONFIG_SYSFS
T
Tom Herbert 已提交
552 553
	struct kobject		kobj;
#endif
554 555 556
#if defined(CONFIG_XPS) && defined(CONFIG_NUMA)
	int			numa_node;
#endif
557 558 559 560 561
/*
 * write mostly part
 */
	spinlock_t		_xmit_lock ____cacheline_aligned_in_smp;
	int			xmit_lock_owner;
562 563 564 565
	/*
	 * please use this field instead of dev->trans_start
	 */
	unsigned long		trans_start;
566 567 568 569 570 571

	/*
	 * Number of TX timeouts for this queue
	 * (/sys/class/net/DEV/Q/trans_timeout)
	 */
	unsigned long		trans_timeout;
T
Tom Herbert 已提交
572 573 574 575 576 577

	unsigned long		state;

#ifdef CONFIG_BQL
	struct dql		dql;
#endif
578
} ____cacheline_aligned_in_smp;
579

580 581 582 583 584
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
585
	return NUMA_NO_NODE;
586 587 588 589 590 591 592 593 594 595
#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
596
#ifdef CONFIG_RPS
T
Tom Herbert 已提交
597 598 599 600 601 602 603 604 605
/*
 * 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];
};
606
#define RPS_MAP_SIZE(_num) (sizeof(struct rps_map) + ((_num) * sizeof(u16)))
T
Tom Herbert 已提交
607

T
Tom Herbert 已提交
608
/*
609 610 611
 * 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
612 613 614
 */
struct rps_dev_flow {
	u16 cpu;
615
	u16 filter;
T
Tom Herbert 已提交
616 617
	unsigned int last_qtail;
};
618
#define RPS_NO_FILTER 0xffff
T
Tom Herbert 已提交
619 620 621 622 623 624 625 626 627 628

/*
 * The rps_dev_flow_table structure contains a table of flow mappings.
 */
struct rps_dev_flow_table {
	unsigned int mask;
	struct rcu_head rcu;
	struct rps_dev_flow flows[0];
};
#define RPS_DEV_FLOW_TABLE_SIZE(_num) (sizeof(struct rps_dev_flow_table) + \
629
    ((_num) * sizeof(struct rps_dev_flow)))
T
Tom Herbert 已提交
630 631 632 633 634 635 636 637 638 639

/*
 * 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) + \
640
    ((_num) * sizeof(u16)))
T
Tom Herbert 已提交
641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664

#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
665
extern struct rps_sock_flow_table __rcu *rps_sock_flow_table;
T
Tom Herbert 已提交
666

667
#ifdef CONFIG_RFS_ACCEL
668 669
bool rps_may_expire_flow(struct net_device *dev, u16 rxq_index, u32 flow_id,
			 u16 filter_id);
670 671
#endif

T
Tom Herbert 已提交
672 673
/* This structure contains an instance of an RX queue. */
struct netdev_rx_queue {
E
Eric Dumazet 已提交
674 675 676
	struct rps_map __rcu		*rps_map;
	struct rps_dev_flow_table __rcu	*rps_flow_table;
	struct kobject			kobj;
T
Tom Herbert 已提交
677
	struct net_device		*dev;
T
Tom Herbert 已提交
678
} ____cacheline_aligned_in_smp;
T
Tom Herbert 已提交
679
#endif /* CONFIG_RPS */
680

T
Tom Herbert 已提交
681 682 683 684 685 686 687 688 689 690 691
#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];
};
692
#define XPS_MAP_SIZE(_num) (sizeof(struct xps_map) + ((_num) * sizeof(u16)))
T
Tom Herbert 已提交
693 694 695 696 697 698 699 700
#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
701
	struct xps_map __rcu *cpu_map[0];
T
Tom Herbert 已提交
702 703 704 705 706
};
#define XPS_DEV_MAPS_SIZE (sizeof(struct xps_dev_maps) +		\
    (nr_cpu_ids * sizeof(struct xps_map *)))
#endif /* CONFIG_XPS */

707 708 709 710 711 712 713 714
#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;
};

715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731
#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

732 733 734 735 736 737 738 739 740 741
#define MAX_PHYS_PORT_ID_LEN 32

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

742 743
/*
 * This structure defines the management hooks for network devices.
744 745
 * The following hooks can be defined; unless noted otherwise, they are
 * optional and can be filled with a null pointer.
746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764
 *
 * 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.
 *
765 766
 * netdev_tx_t (*ndo_start_xmit)(struct sk_buff *skb,
 *                               struct net_device *dev);
767
 *	Called when a packet needs to be transmitted.
768 769
 *	Must return NETDEV_TX_OK , NETDEV_TX_BUSY.
 *        (can also return NETDEV_TX_LOCKED iff NETIF_F_LLTX)
770 771 772 773 774 775
 *	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.
 *
776 777 778 779 780 781
 * 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.
782 783
 *	If driver handles unicast address filtering, it should set
 *	IFF_UNICAST_FLT to its priv_flags.
784 785 786
 *
 * int (*ndo_set_mac_address)(struct net_device *dev, void *addr);
 *	This function  is called when the Media Access Control address
787
 *	needs to be changed. If this interface is not defined, the
788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807
 *	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.
 *
808
 * void (*ndo_tx_timeout)(struct net_device *dev);
809 810 811
 *	Callback uses when the transmitter has not made any progress
 *	for dev->watchdog ticks.
 *
812
 * struct rtnl_link_stats64* (*ndo_get_stats64)(struct net_device *dev,
813
 *                      struct rtnl_link_stats64 *storage);
814
 * struct net_device_stats* (*ndo_get_stats)(struct net_device *dev);
815
 *	Called when a user wants to get the network device usage
816
 *	statistics. Drivers must do one of the following:
817 818
 *	1. Define @ndo_get_stats64 to fill in a zero-initialised
 *	   rtnl_link_stats64 structure passed by the caller.
819
 *	2. Define @ndo_get_stats to update a net_device_stats structure
820 821 822 823 824
 *	   (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.
825
 *
826 827 828
 * int (*ndo_vlan_rx_add_vid)(struct net_device *dev, __be16 proto, u16t vid);
 *	If device support VLAN filtering this function is called when a
 *	VLAN id is registered.
829
 *
830
 * int (*ndo_vlan_rx_kill_vid)(struct net_device *dev, unsigned short vid);
831 832
 *	If device support VLAN filtering this function is called when a
 *	VLAN id is unregistered.
833 834
 *
 * void (*ndo_poll_controller)(struct net_device *dev);
835 836 837 838 839
 *
 *	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);
840
 * int (*ndo_set_vf_spoofchk)(struct net_device *dev, int vf, bool setting);
841 842
 * int (*ndo_get_vf_config)(struct net_device *dev,
 *			    int vf, struct ifla_vf_info *ivf);
843
 * int (*ndo_set_vf_link_state)(struct net_device *dev, int vf, int link_state);
844 845 846
 * 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);
847 848 849 850 851
 * 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.
852
 *
853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882
 *	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.
 *
883 884 885 886 887 888 889
 * 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.
 *
890 891 892 893 894 895
 * 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.
 *
896 897 898 899 900 901
 *	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.
902
 *
903
 *	Slave management functions (for bridge, bonding, etc).
904 905 906 907 908
 * 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.
909 910
 *
 *      Feature/offload setting functions.
911 912
 * netdev_features_t (*ndo_fix_features)(struct net_device *dev,
 *		netdev_features_t features);
913 914 915 916
 *	Adjusts the requested feature flags according to device-specific
 *	constraints, and returns the resulting flags. Must not modify
 *	the device state.
 *
917
 * int (*ndo_set_features)(struct net_device *dev, netdev_features_t features);
918 919 920 921
 *	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.
 *
922 923
 * int (*ndo_fdb_add)(struct ndmsg *ndm, struct nlattr *tb[],
 *		      struct net_device *dev,
924
 *		      const unsigned char *addr, u16 flags)
925
 *	Adds an FDB entry to dev for addr.
926 927
 * int (*ndo_fdb_del)(struct ndmsg *ndm, struct nlattr *tb[],
 *		      struct net_device *dev,
928
 *		      const unsigned char *addr)
929 930 931 932 933
 *	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
934 935 936
 *
 * int (*ndo_bridge_setlink)(struct net_device *dev, struct nlmsghdr *nlh)
 * int (*ndo_bridge_getlink)(struct sk_buff *skb, u32 pid, u32 seq,
937
 *			     struct net_device *dev, u32 filter_mask)
J
Jiri Pirko 已提交
938 939 940 941 942 943 944 945
 *
 * 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.
946 947 948 949 950 951
 *
 * int (*ndo_get_phys_port_id)(struct net_device *dev,
 *			       struct netdev_phys_port_id *ppid);
 *	Called to get ID of physical port of this device. If driver does
 *	not implement this, it is assumed that the hw is not able to have
 *	multiple net devices on single physical port.
952 953
 *
 * void (*ndo_add_vxlan_port)(struct  net_device *dev,
J
Joseph Gasparakis 已提交
954
 *			      sa_family_t sa_family, __be16 port);
955 956 957 958 959 960
 *	Called by vxlan to notiy a driver about the UDP port and socket
 *	address family that vxlan is listnening to. It is called only when
 *	a new port starts listening. The operation is protected by the
 *	vxlan_net->sock_lock.
 *
 * void (*ndo_del_vxlan_port)(struct  net_device *dev,
J
Joseph Gasparakis 已提交
961
 *			      sa_family_t sa_family, __be16 port);
962 963 964
 *	Called by vxlan to notify the driver about a UDP port and socket
 *	address family that vxlan is not listening to anymore. The operation
 *	is protected by the vxlan_net->sock_lock.
965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983
 *
 * void* (*ndo_dfwd_add_station)(struct net_device *pdev,
 *				 struct net_device *dev)
 *	Called by upper layer devices to accelerate switching or other
 *	station functionality into hardware. 'pdev is the lowerdev
 *	to use for the offload and 'dev' is the net device that will
 *	back the offload. Returns a pointer to the private structure
 *	the upper layer will maintain.
 * void (*ndo_dfwd_del_station)(struct net_device *pdev, void *priv)
 *	Called by upper layer device to delete the station created
 *	by 'ndo_dfwd_add_station'. 'pdev' is the net device backing
 *	the station and priv is the structure returned by the add
 *	operation.
 * netdev_tx_t (*ndo_dfwd_start_xmit)(struct sk_buff *skb,
 *				      struct net_device *dev,
 *				      void *priv);
 *	Callback to use for xmit over the accelerated station. This
 *	is used in place of ndo_start_xmit on accelerated net
 *	devices.
984 985 986 987 988 989
 */
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);
990
	netdev_tx_t		(*ndo_start_xmit) (struct sk_buff *skb,
991 992 993
						   struct net_device *dev);
	u16			(*ndo_select_queue)(struct net_device *dev,
						    struct sk_buff *skb);
994 995 996 997 998 999 1000 1001 1002 1003
	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);
1004 1005 1006 1007
	int			(*ndo_change_mtu)(struct net_device *dev,
						  int new_mtu);
	int			(*ndo_neigh_setup)(struct net_device *dev,
						   struct neigh_parms *);
1008 1009
	void			(*ndo_tx_timeout) (struct net_device *dev);

1010 1011
	struct rtnl_link_stats64* (*ndo_get_stats64)(struct net_device *dev,
						     struct rtnl_link_stats64 *storage);
1012 1013
	struct net_device_stats* (*ndo_get_stats)(struct net_device *dev);

1014
	int			(*ndo_vlan_rx_add_vid)(struct net_device *dev,
1015
						       __be16 proto, u16 vid);
1016
	int			(*ndo_vlan_rx_kill_vid)(struct net_device *dev,
1017
						        __be16 proto, u16 vid);
1018 1019
#ifdef CONFIG_NET_POLL_CONTROLLER
	void                    (*ndo_poll_controller)(struct net_device *dev);
H
Herbert Xu 已提交
1020
	int			(*ndo_netpoll_setup)(struct net_device *dev,
1021 1022
						     struct netpoll_info *info,
						     gfp_t gfp);
1023
	void			(*ndo_netpoll_cleanup)(struct net_device *dev);
E
Eliezer Tamir 已提交
1024
#endif
1025
#ifdef CONFIG_NET_RX_BUSY_POLL
1026
	int			(*ndo_busy_poll)(struct napi_struct *dev);
1027
#endif
1028 1029 1030 1031 1032 1033
	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);
1034 1035
	int			(*ndo_set_vf_spoofchk)(struct net_device *dev,
						       int vf, bool setting);
1036 1037 1038
	int			(*ndo_get_vf_config)(struct net_device *dev,
						     int vf,
						     struct ifla_vf_info *ivf);
1039 1040
	int			(*ndo_set_vf_link_state)(struct net_device *dev,
							 int vf, int link_state);
1041 1042 1043 1044 1045
	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);
1046
	int			(*ndo_setup_tc)(struct net_device *dev, u8 tc);
1047
#if IS_ENABLED(CONFIG_FCOE)
1048 1049
	int			(*ndo_fcoe_enable)(struct net_device *dev);
	int			(*ndo_fcoe_disable)(struct net_device *dev);
1050 1051 1052 1053 1054 1055
	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);
1056 1057 1058 1059
	int			(*ndo_fcoe_ddp_target)(struct net_device *dev,
						       u16 xid,
						       struct scatterlist *sgl,
						       unsigned int sgc);
1060 1061
	int			(*ndo_fcoe_get_hbainfo)(struct net_device *dev,
							struct netdev_fcoe_hbainfo *hbainfo);
1062 1063
#endif

1064
#if IS_ENABLED(CONFIG_LIBFCOE)
1065 1066 1067 1068
#define NETDEV_FCOE_WWNN 0
#define NETDEV_FCOE_WWPN 1
	int			(*ndo_fcoe_get_wwn)(struct net_device *dev,
						    u64 *wwn, int type);
1069
#endif
1070

1071 1072 1073 1074 1075 1076
#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
1077 1078 1079 1080
	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);
1081 1082
	netdev_features_t	(*ndo_fix_features)(struct net_device *dev,
						    netdev_features_t features);
1083
	int			(*ndo_set_features)(struct net_device *dev,
1084
						    netdev_features_t features);
1085
	int			(*ndo_neigh_construct)(struct neighbour *n);
1086
	void			(*ndo_neigh_destroy)(struct neighbour *n);
1087 1088

	int			(*ndo_fdb_add)(struct ndmsg *ndm,
1089
					       struct nlattr *tb[],
1090
					       struct net_device *dev,
1091
					       const unsigned char *addr,
1092 1093
					       u16 flags);
	int			(*ndo_fdb_del)(struct ndmsg *ndm,
1094
					       struct nlattr *tb[],
1095
					       struct net_device *dev,
1096
					       const unsigned char *addr);
1097 1098 1099 1100
	int			(*ndo_fdb_dump)(struct sk_buff *skb,
						struct netlink_callback *cb,
						struct net_device *dev,
						int idx);
J
John Fastabend 已提交
1101 1102 1103 1104 1105

	int			(*ndo_bridge_setlink)(struct net_device *dev,
						      struct nlmsghdr *nlh);
	int			(*ndo_bridge_getlink)(struct sk_buff *skb,
						      u32 pid, u32 seq,
1106 1107
						      struct net_device *dev,
						      u32 filter_mask);
1108 1109
	int			(*ndo_bridge_dellink)(struct net_device *dev,
						      struct nlmsghdr *nlh);
J
Jiri Pirko 已提交
1110 1111
	int			(*ndo_change_carrier)(struct net_device *dev,
						      bool new_carrier);
1112 1113
	int			(*ndo_get_phys_port_id)(struct net_device *dev,
							struct netdev_phys_port_id *ppid);
1114 1115
	void			(*ndo_add_vxlan_port)(struct  net_device *dev,
						      sa_family_t sa_family,
J
Joseph Gasparakis 已提交
1116
						      __be16 port);
1117 1118
	void			(*ndo_del_vxlan_port)(struct  net_device *dev,
						      sa_family_t sa_family,
J
Joseph Gasparakis 已提交
1119
						      __be16 port);
1120 1121 1122 1123 1124 1125 1126 1127 1128

	void*			(*ndo_dfwd_add_station)(struct net_device *pdev,
							struct net_device *dev);
	void			(*ndo_dfwd_del_station)(struct net_device *pdev,
							void *priv);

	netdev_tx_t		(*ndo_dfwd_start_xmit) (struct sk_buff *skb,
							struct net_device *dev,
							void *priv);
1129 1130
};

L
Linus Torvalds 已提交
1131 1132 1133 1134 1135 1136 1137 1138 1139 1140
/*
 *	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
1141
struct net_device {
L
Linus Torvalds 已提交
1142 1143 1144 1145

	/*
	 * 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
1146
	 * of the interface.
L
Linus Torvalds 已提交
1147 1148
	 */
	char			name[IFNAMSIZ];
M
Mark Gross 已提交
1149

1150
	/* device name hash chain, please keep it close to name[] */
1151
	struct hlist_node	name_hlist;
1152

1153 1154
	/* snmp alias */
	char 			*ifalias;
L
Linus Torvalds 已提交
1155 1156 1157 1158 1159 1160 1161 1162

	/*
	 *	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	*/
1163
	int			irq;		/* device IRQ number	*/
L
Linus Torvalds 已提交
1164 1165 1166 1167 1168 1169 1170 1171

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

	unsigned long		state;

1172
	struct list_head	dev_list;
1173
	struct list_head	napi_list;
1174
	struct list_head	unreg_list;
1175
	struct list_head	close_list;
1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187

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

	/* all linked devices, *including* neighbours */
	struct {
		struct list_head upper;
		struct list_head lower;
	} all_adj_list;
1188

L
Linus Torvalds 已提交
1189

1190
	/* currently active device features */
1191
	netdev_features_t	features;
1192
	/* user-changeable features */
1193
	netdev_features_t	hw_features;
1194
	/* user-requested features */
1195
	netdev_features_t	wanted_features;
1196
	/* mask of features inheritable by VLAN devices */
1197
	netdev_features_t	vlan_features;
1198 1199 1200 1201 1202 1203
	/* mask of features inherited by encapsulating devices
	 * This field indicates what encapsulation offloads
	 * the hardware is capable of doing, and drivers will
	 * need to set them appropriately.
	 */
	netdev_features_t	hw_enc_features;
S
Simon Horman 已提交
1204 1205
	/* mask of fetures inheritable by MPLS */
	netdev_features_t	mpls_features;
1206

L
Linus Torvalds 已提交
1207 1208 1209 1210
	/* Interface index. Unique device identifier	*/
	int			ifindex;
	int			iflink;

R
Rusty Russell 已提交
1211
	struct net_device_stats	stats;
1212 1213 1214
	atomic_long_t		rx_dropped; /* dropped packets by core network
					     * Do not use this in drivers.
					     */
L
Linus Torvalds 已提交
1215

1216
#ifdef CONFIG_WIRELESS_EXT
L
Linus Torvalds 已提交
1217 1218 1219 1220 1221
	/* 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;
1222
#endif
1223 1224
	/* Management operations */
	const struct net_device_ops *netdev_ops;
1225
	const struct ethtool_ops *ethtool_ops;
1226
	const struct forwarding_accel_ops *fwd_ops;
L
Linus Torvalds 已提交
1227

1228 1229 1230
	/* Hardware header description */
	const struct header_ops *header_ops;

S
Stefan Rompf 已提交
1231
	unsigned int		flags;	/* interface flags (a la BSD)	*/
1232 1233
	unsigned int		priv_flags; /* Like 'flags' but invisible to userspace.
					     * See if.h for definitions. */
L
Linus Torvalds 已提交
1234 1235 1236
	unsigned short		gflags;
	unsigned short		padded;	/* How much padding added by alloc_netdev() */

S
Stefan Rompf 已提交
1237 1238 1239
	unsigned char		operstate; /* RFC2863 operstate */
	unsigned char		link_mode; /* mapping policy to operstate */

1240 1241 1242
	unsigned char		if_port;	/* Selectable AUI, TP,..*/
	unsigned char		dma;		/* DMA channel		*/

1243
	unsigned int		mtu;	/* interface MTU value		*/
L
Linus Torvalds 已提交
1244 1245 1246
	unsigned short		type;	/* interface hardware type	*/
	unsigned short		hard_header_len;	/* hardware hdr length	*/

1247 1248 1249 1250 1251 1252 1253
	/* 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
1254
	/* Interface address info. */
1255
	unsigned char		perm_addr[MAX_ADDR_LEN]; /* permanent hw address */
1256
	unsigned char		addr_assign_type; /* hw address assignment type */
L
Linus Torvalds 已提交
1257
	unsigned char		addr_len;	/* hardware address length	*/
1258
	unsigned short		neigh_priv_len;
1259 1260 1261 1262
	unsigned short          dev_id;		/* Used to differentiate devices
						 * that share the same link
						 * layer address
						 */
J
Jiri Pirko 已提交
1263
	spinlock_t		addr_list_lock;
1264 1265
	struct netdev_hw_addr_list	uc;	/* Unicast mac addresses */
	struct netdev_hw_addr_list	mc;	/* Multicast mac addresses */
1266 1267 1268 1269 1270 1271 1272
	struct netdev_hw_addr_list	dev_addrs; /* list of device
						    * hw addresses
						    */
#ifdef CONFIG_SYSFS
	struct kset		*queues_kset;
#endif

1273
	bool			uc_promisc;
1274 1275
	unsigned int		promiscuity;
	unsigned int		allmulti;
L
Linus Torvalds 已提交
1276 1277 1278


	/* Protocol specific pointers */
1279

1280
#if IS_ENABLED(CONFIG_VLAN_8021Q)
1281
	struct vlan_info __rcu	*vlan_info;	/* VLAN info */
1282
#endif
1283
#if IS_ENABLED(CONFIG_NET_DSA)
1284
	struct dsa_switch_tree	*dsa_ptr;	/* dsa specific data */
1285 1286 1287
#endif
#if IS_ENABLED(CONFIG_TIPC)
	struct tipc_bearer __rcu *tipc_ptr;	/* TIPC specific data */
1288
#endif
L
Linus Torvalds 已提交
1289
	void 			*atalk_ptr;	/* AppleTalk link 	*/
E
Eric Dumazet 已提交
1290
	struct in_device __rcu	*ip_ptr;	/* IPv4 specific data	*/
1291
	struct dn_dev __rcu     *dn_ptr;        /* DECnet specific data */
E
Eric Dumazet 已提交
1292
	struct inet6_dev __rcu	*ip6_ptr;       /* IPv6 specific data */
L
Linus Torvalds 已提交
1293
	void			*ax25_ptr;	/* AX.25 specific data */
1294 1295
	struct wireless_dev	*ieee80211_ptr;	/* IEEE 802.11 specific data,
						   assign before registering */
L
Linus Torvalds 已提交
1296

1297
/*
E
Eric Dumazet 已提交
1298
 * Cache lines mostly used on receive path (including eth_type_trans())
1299
 */
1300 1301 1302 1303 1304 1305 1306 1307
	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.
						 */

1308
	/* Interface address info used in eth_type_trans() */
1309 1310 1311 1312
	unsigned char		*dev_addr;	/* hw address, (before bcast
						   because most packets are
						   unicast) */

T
Tom Herbert 已提交
1313

1314
#ifdef CONFIG_RPS
T
Tom Herbert 已提交
1315 1316
	struct netdev_rx_queue	*_rx;

1317
	/* Number of RX queues allocated at register_netdev() time */
T
Tom Herbert 已提交
1318
	unsigned int		num_rx_queues;
1319 1320 1321

	/* Number of RX queues currently active in device */
	unsigned int		real_num_rx_queues;
1322

E
Eric Dumazet 已提交
1323
#endif
T
Tom Herbert 已提交
1324

1325 1326
	rx_handler_func_t __rcu	*rx_handler;
	void __rcu		*rx_handler_data;
1327

1328
	struct netdev_queue __rcu *ingress_queue;
1329 1330
	unsigned char		broadcast[MAX_ADDR_LEN];	/* hw bcast add	*/

E
Eric Dumazet 已提交
1331 1332 1333 1334

/*
 * Cache lines mostly used on transmit path
 */
1335
	struct netdev_queue	*_tx ____cacheline_aligned_in_smp;
1336 1337

	/* Number of TX queues allocated at alloc_netdev_mq() time  */
1338
	unsigned int		num_tx_queues;
1339 1340 1341 1342

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

1343 1344 1345
	/* root qdisc from userspace point of view */
	struct Qdisc		*qdisc;

L
Linus Torvalds 已提交
1346
	unsigned long		tx_queue_len;	/* Max frames per queue allowed */
1347
	spinlock_t		tx_global_lock;
E
Eric Dumazet 已提交
1348

T
Tom Herbert 已提交
1349
#ifdef CONFIG_XPS
E
Eric Dumazet 已提交
1350
	struct xps_dev_maps __rcu *xps_maps;
T
Tom Herbert 已提交
1351
#endif
1352 1353 1354 1355 1356 1357
#ifdef CONFIG_RFS_ACCEL
	/* CPU reverse-mapping for RX completion interrupts, indexed
	 * by RX queue number.  Assigned by driver.  This must only be
	 * set if the ndo_rx_flow_steer operation is defined. */
	struct cpu_rmap		*rx_cpu_rmap;
#endif
T
Tom Herbert 已提交
1358

1359
	/* These may be needed for future network-power-down code. */
1360 1361 1362 1363 1364

	/*
	 * trans_start here is expensive for high speed devices on SMP,
	 * please use netdev_queue->trans_start instead.
	 */
1365 1366 1367 1368 1369
	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
1370
	/* Number of references to this device */
E
Eric Dumazet 已提交
1371
	int __percpu		*pcpu_refcnt;
1372

L
Linus Torvalds 已提交
1373 1374 1375 1376 1377
	/* delayed register/unregister */
	struct list_head	todo_list;
	/* device index hash chain */
	struct hlist_node	index_hlist;

1378
	struct list_head	link_watch_list;
1379

L
Linus Torvalds 已提交
1380 1381
	/* register/unregister state machine */
	enum { NETREG_UNINITIALIZED=0,
1382
	       NETREG_REGISTERED,	/* completed register_netdevice */
L
Linus Torvalds 已提交
1383 1384 1385
	       NETREG_UNREGISTERING,	/* called unregister_netdevice */
	       NETREG_UNREGISTERED,	/* completed unregister todo */
	       NETREG_RELEASED,		/* called free_netdev */
1386
	       NETREG_DUMMY,		/* dummy device for NAPI poll */
1387 1388 1389
	} reg_state:8;

	bool dismantle; /* device is going do be freed */
1390 1391 1392 1393 1394

	enum {
		RTNL_LINK_INITIALIZED,
		RTNL_LINK_INITIALIZING,
	} rtnl_link_state:16;
L
Linus Torvalds 已提交
1395

1396 1397
	/* Called from unregister, can be used to call free_netdev */
	void (*destructor)(struct net_device *dev);
L
Linus Torvalds 已提交
1398 1399

#ifdef CONFIG_NETPOLL
1400
	struct netpoll_info __rcu	*npinfo;
L
Linus Torvalds 已提交
1401
#endif
1402

1403
#ifdef CONFIG_NET_NS
1404 1405
	/* Network namespace this network device is inside */
	struct net		*nd_net;
1406
#endif
1407

D
David S. Miller 已提交
1408
	/* mid-layer private */
E
Eric Dumazet 已提交
1409 1410 1411
	union {
		void				*ml_priv;
		struct pcpu_lstats __percpu	*lstats; /* loopback stats */
1412
		struct pcpu_sw_netstats __percpu	*tstats;
1413
		struct pcpu_dstats __percpu	*dstats; /* dummy stats */
E
Eric Dumazet 已提交
1414
		struct pcpu_vstats __percpu	*vstats; /* veth stats */
E
Eric Dumazet 已提交
1415
	};
1416
	/* GARP */
E
Eric Dumazet 已提交
1417
	struct garp_port __rcu	*garp_port;
1418 1419
	/* MRP */
	struct mrp_port __rcu	*mrp_port;
L
Linus Torvalds 已提交
1420 1421

	/* class/net/name entry */
1422
	struct device		dev;
1423 1424
	/* space for optional device, statistics, and wireless sysfs groups */
	const struct attribute_group *sysfs_groups[4];
P
Patrick McHardy 已提交
1425 1426 1427

	/* rtnetlink link ops */
	const struct rtnl_link_ops *rtnl_link_ops;
1428

1429 1430 1431
	/* for setting kernel sock attribute on TCP connection setup */
#define GSO_MAX_SIZE		65536
	unsigned int		gso_max_size;
1432 1433
#define GSO_MAX_SEGS		65535
	u16			gso_max_segs;
1434

J
Jeff Kirsher 已提交
1435
#ifdef CONFIG_DCB
1436
	/* Data Center Bridging netlink ops */
1437
	const struct dcbnl_rtnl_ops *dcbnl_ops;
1438
#endif
1439 1440 1441
	u8 num_tc;
	struct netdev_tc_txq tc_to_txq[TC_MAX_QUEUE];
	u8 prio_tc_map[TC_BITMASK + 1];
1442

1443
#if IS_ENABLED(CONFIG_FCOE)
1444 1445
	/* max exchange id for FCoE LRO by ddp */
	unsigned int		fcoe_ddp_xid;
1446 1447 1448
#endif
#if IS_ENABLED(CONFIG_NETPRIO_CGROUP)
	struct netprio_map __rcu *priomap;
1449
#endif
1450 1451
	/* phy device may attach itself for hardware timestamping */
	struct phy_device *phydev;
1452

1453 1454
	struct lock_class_key *qdisc_tx_busylock;

1455 1456
	/* group the device belongs to */
	int group;
1457 1458

	struct pm_qos_request	pm_qos_req;
L
Linus Torvalds 已提交
1459
};
1460
#define to_net_dev(d) container_of(d, struct net_device, dev)
L
Linus Torvalds 已提交
1461 1462 1463

#define	NETDEV_ALIGN		32

1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514
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;
}

1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533
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);
}

1534 1535 1536
struct netdev_queue *netdev_pick_tx(struct net_device *dev,
				    struct sk_buff *skb);
u16 __netdev_pick_tx(struct net_device *dev, struct sk_buff *skb);
1537

1538 1539 1540 1541 1542 1543
/*
 * Net namespace inlines
 */
static inline
struct net *dev_net(const struct net_device *dev)
{
E
Eric Dumazet 已提交
1544
	return read_pnet(&dev->nd_net);
1545 1546 1547
}

static inline
1548
void dev_net_set(struct net_device *dev, struct net *net)
1549 1550
{
#ifdef CONFIG_NET_NS
1551 1552
	release_net(dev->nd_net);
	dev->nd_net = hold_net(net);
1553 1554 1555
#endif
}

1556 1557 1558 1559 1560 1561 1562 1563 1564 1565
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;
}

1566 1567 1568 1569 1570 1571 1572 1573 1574 1575
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;
}

1576 1577 1578 1579 1580 1581
/**
 *	netdev_priv - access network device private data
 *	@dev: network device
 *
 * Get network device private data
 */
1582
static inline void *netdev_priv(const struct net_device *dev)
L
Linus Torvalds 已提交
1583
{
1584
	return (char *)dev + ALIGN(sizeof(struct net_device), NETDEV_ALIGN);
L
Linus Torvalds 已提交
1585 1586 1587 1588 1589
}

/* Set the sysfs physical device reference for the network logical device
 * if set prior to registration will cause a symlink during initialization.
 */
1590
#define SET_NETDEV_DEV(net, pdev)	((net)->dev.parent = (pdev))
L
Linus Torvalds 已提交
1591

1592
/* Set the sysfs device type for the network logical device to allow
1593
 * fine-grained identification of different network device types. For
1594 1595 1596 1597
 * example Ethernet, Wirelss LAN, Bluetooth, WiMAX etc.
 */
#define SET_NETDEV_DEVTYPE(net, devtype)	((net)->dev.type = (devtype))

E
Eric Dumazet 已提交
1598 1599 1600 1601 1602
/* Default NAPI poll() weight
 * Device drivers are strongly advised to not use bigger value
 */
#define NAPI_POLL_WEIGHT 64

1603 1604 1605 1606 1607 1608 1609 1610 1611 1612
/**
 *	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.
 */
1613 1614
void netif_napi_add(struct net_device *dev, struct napi_struct *napi,
		    int (*poll)(struct napi_struct *, int), int weight);
1615

1616 1617 1618 1619 1620 1621
/**
 *  netif_napi_del - remove a napi context
 *  @napi: napi context
 *
 *  netif_napi_del() removes a napi context from the network device napi list
 */
1622 1623 1624
void netif_napi_del(struct napi_struct *napi);

struct napi_gro_cb {
1625 1626 1627
	/* Virtual address of skb_shinfo(skb)->frags[0].page + offset. */
	void *frag0;

1628 1629 1630
	/* Length of frag0. */
	unsigned int frag0_len;

1631 1632 1633
	/* This indicates where we are processing relative to skb->data. */
	int data_offset;

1634 1635 1636 1637
	/* This is non-zero if the packet cannot be merged with the new skb. */
	int flush;

	/* Number of segments aggregated. */
1638 1639 1640 1641
	u16	count;

	/* This is non-zero if the packet may be of the same flow. */
	u8	same_flow;
H
Herbert Xu 已提交
1642 1643

	/* Free the skb? */
1644
	u8	free;
1645 1646
#define NAPI_GRO_FREE		  1
#define NAPI_GRO_FREE_STOLEN_HEAD 2
1647 1648 1649

	/* jiffies when first packet was created/queued */
	unsigned long age;
1650 1651 1652

	/* Used in ipv6_gro_receive() */
	int	proto;
1653 1654 1655

	/* used in skb_gro_receive() slow path */
	struct sk_buff *last;
1656 1657 1658
};

#define NAPI_GRO_CB(skb) ((struct napi_gro_cb *)(skb)->cb)
1659

L
Linus Torvalds 已提交
1660
struct packet_type {
D
David S. Miller 已提交
1661 1662 1663 1664 1665 1666
	__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 *);
1667 1668
	bool			(*id_match)(struct packet_type *ptype,
					    struct sock *sk);
L
Linus Torvalds 已提交
1669 1670 1671 1672
	void			*af_packet_priv;
	struct list_head	list;
};

1673
struct offload_callbacks {
1674
	struct sk_buff		*(*gso_segment)(struct sk_buff *skb,
1675
						netdev_features_t features);
1676
	int			(*gso_send_check)(struct sk_buff *skb);
1677 1678
	struct sk_buff		**(*gro_receive)(struct sk_buff **head,
					       struct sk_buff *skb);
1679
	int			(*gro_complete)(struct sk_buff *skb, int nhoff);
1680 1681 1682 1683 1684 1685
};

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
1686 1687
};

1688 1689 1690 1691 1692 1693 1694 1695 1696
/* often modified stats are per cpu, other are shared (netdev->stats) */
struct pcpu_sw_netstats {
	u64     rx_packets;
	u64     rx_bytes;
	u64     tx_packets;
	u64     tx_bytes;
	struct u64_stats_sync   syncp;
};

L
Linus Torvalds 已提交
1697 1698
#include <linux/notifier.h>

1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721
/* 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
1722
#define NETDEV_UNREGISTER_FINAL 0x0011
1723 1724 1725
#define NETDEV_RELEASE		0x0012
#define NETDEV_NOTIFY_PEERS	0x0013
#define NETDEV_JOIN		0x0014
1726
#define NETDEV_CHANGEUPPER	0x0015
1727
#define NETDEV_RESEND_IGMP	0x0016
1728

1729 1730
int register_netdevice_notifier(struct notifier_block *nb);
int unregister_netdevice_notifier(struct notifier_block *nb);
1731 1732 1733 1734 1735

struct netdev_notifier_info {
	struct net_device *dev;
};

1736 1737 1738 1739 1740
struct netdev_notifier_change_info {
	struct netdev_notifier_info info; /* must be first */
	unsigned int flags_changed;
};

1741 1742 1743 1744 1745 1746
static inline void netdev_notifier_info_init(struct netdev_notifier_info *info,
					     struct net_device *dev)
{
	info->dev = dev;
}

1747 1748 1749 1750 1751 1752
static inline struct net_device *
netdev_notifier_info_to_dev(const struct netdev_notifier_info *info)
{
	return info->dev;
}

1753
int call_netdevice_notifiers(unsigned long val, struct net_device *dev);
1754 1755


L
Linus Torvalds 已提交
1756 1757
extern rwlock_t				dev_base_lock;		/* Device list lock */

1758 1759
#define for_each_netdev(net, d)		\
		list_for_each_entry(d, &(net)->dev_base_head, dev_list)
1760 1761
#define for_each_netdev_reverse(net, d)	\
		list_for_each_entry_reverse(d, &(net)->dev_base_head, dev_list)
1762 1763
#define for_each_netdev_rcu(net, d)		\
		list_for_each_entry_rcu(d, &(net)->dev_base_head, dev_list)
1764 1765 1766 1767
#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)
1768 1769
#define for_each_netdev_continue_rcu(net, d)		\
	list_for_each_entry_continue_rcu(d, &(net)->dev_base_head, dev_list)
1770 1771 1772
#define for_each_netdev_in_bond_rcu(bond, slave)	\
		for_each_netdev_rcu(&init_net, slave)	\
			if (netdev_master_upper_dev_get_rcu(slave) == bond)
1773
#define net_device_entry(lh)	list_entry(lh, struct net_device, dev_list)
1774

1775 1776 1777 1778 1779
static inline struct net_device *next_net_device(struct net_device *dev)
{
	struct list_head *lh;
	struct net *net;

1780
	net = dev_net(dev);
1781 1782 1783 1784
	lh = dev->dev_list.next;
	return lh == &net->dev_base_head ? NULL : net_device_entry(lh);
}

1785 1786 1787 1788 1789 1790
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
1791
	lh = rcu_dereference(list_next_rcu(&dev->dev_list));
1792 1793 1794
	return lh == &net->dev_base_head ? NULL : net_device_entry(lh);
}

1795 1796 1797 1798 1799
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);
}
1800

E
Eric Dumazet 已提交
1801 1802 1803 1804 1805 1806 1807
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);
}

1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833
int netdev_boot_setup_check(struct net_device *dev);
unsigned long netdev_boot_base(const char *prefix, int unit);
struct net_device *dev_getbyhwaddr_rcu(struct net *net, unsigned short type,
				       const char *hwaddr);
struct net_device *dev_getfirstbyhwtype(struct net *net, unsigned short type);
struct net_device *__dev_getfirstbyhwtype(struct net *net, unsigned short type);
void dev_add_pack(struct packet_type *pt);
void dev_remove_pack(struct packet_type *pt);
void __dev_remove_pack(struct packet_type *pt);
void dev_add_offload(struct packet_offload *po);
void dev_remove_offload(struct packet_offload *po);

struct net_device *dev_get_by_flags_rcu(struct net *net, unsigned short flags,
					unsigned short mask);
struct net_device *dev_get_by_name(struct net *net, const char *name);
struct net_device *dev_get_by_name_rcu(struct net *net, const char *name);
struct net_device *__dev_get_by_name(struct net *net, const char *name);
int dev_alloc_name(struct net_device *dev, const char *name);
int dev_open(struct net_device *dev);
int dev_close(struct net_device *dev);
void dev_disable_lro(struct net_device *dev);
int dev_loopback_xmit(struct sk_buff *newskb);
int dev_queue_xmit(struct sk_buff *skb);
int register_netdevice(struct net_device *dev);
void unregister_netdevice_queue(struct net_device *dev, struct list_head *head);
void unregister_netdevice_many(struct list_head *head);
1834 1835 1836 1837 1838
static inline void unregister_netdevice(struct net_device *dev)
{
	unregister_netdevice_queue(dev, NULL);
}

1839 1840
int netdev_refcnt_read(const struct net_device *dev);
void free_netdev(struct net_device *dev);
1841
void netdev_freemem(struct net_device *dev);
1842 1843
void synchronize_net(void);
int init_dummy_netdev(struct net_device *dev);
1844

1845 1846 1847 1848 1849
struct net_device *dev_get_by_index(struct net *net, int ifindex);
struct net_device *__dev_get_by_index(struct net *net, int ifindex);
struct net_device *dev_get_by_index_rcu(struct net *net, int ifindex);
int netdev_get_name(struct net *net, char *name, int ifindex);
int dev_restart(struct net_device *dev);
L
Linus Torvalds 已提交
1850
#ifdef CONFIG_NETPOLL_TRAP
1851
int netpoll_trap(void);
L
Linus Torvalds 已提交
1852
#endif
1853
int skb_gro_receive(struct sk_buff **head, struct sk_buff *skb);
1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869

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;
}

1870 1871
static inline void *skb_gro_header_fast(struct sk_buff *skb,
					unsigned int offset)
1872
{
1873 1874
	return NAPI_GRO_CB(skb)->frag0 + offset;
}
1875

1876 1877 1878 1879
static inline int skb_gro_header_hard(struct sk_buff *skb, unsigned int hlen)
{
	return NAPI_GRO_CB(skb)->frag0_len < hlen;
}
1880

1881 1882 1883
static inline void *skb_gro_header_slow(struct sk_buff *skb, unsigned int hlen,
					unsigned int offset)
{
1884 1885 1886
	if (!pskb_may_pull(skb, hlen))
		return NULL;

1887 1888
	NAPI_GRO_CB(skb)->frag0 = NULL;
	NAPI_GRO_CB(skb)->frag0_len = 0;
1889
	return skb->data + offset;
1890
}
L
Linus Torvalds 已提交
1891

1892 1893
static inline void *skb_gro_mac_header(struct sk_buff *skb)
{
1894
	return NAPI_GRO_CB(skb)->frag0 ?: skb_mac_header(skb);
1895 1896
}

H
Herbert Xu 已提交
1897 1898
static inline void *skb_gro_network_header(struct sk_buff *skb)
{
1899 1900
	return (NAPI_GRO_CB(skb)->frag0 ?: skb->data) +
	       skb_network_offset(skb);
H
Herbert Xu 已提交
1901 1902
}

1903 1904
static inline int dev_hard_header(struct sk_buff *skb, struct net_device *dev,
				  unsigned short type,
1905
				  const void *daddr, const void *saddr,
1906
				  unsigned int len)
1907
{
1908
	if (!dev->header_ops || !dev->header_ops->create)
1909
		return 0;
1910 1911

	return dev->header_ops->create(skb, dev, type, daddr, saddr, len);
1912 1913
}

S
Stephen Hemminger 已提交
1914 1915 1916 1917 1918
static inline int dev_parse_header(const struct sk_buff *skb,
				   unsigned char *haddr)
{
	const struct net_device *dev = skb->dev;

1919
	if (!dev->header_ops || !dev->header_ops->parse)
S
Stephen Hemminger 已提交
1920
		return 0;
1921
	return dev->header_ops->parse(skb, haddr);
S
Stephen Hemminger 已提交
1922 1923
}

L
Linus Torvalds 已提交
1924
typedef int gifconf_func_t(struct net_device * dev, char __user * bufptr, int len);
1925
int register_gifconf(unsigned int family, gifconf_func_t *gifconf);
L
Linus Torvalds 已提交
1926 1927 1928 1929 1930
static inline int unregister_gifconf(unsigned int family)
{
	return register_gifconf(family, NULL);
}

1931
#ifdef CONFIG_NET_FLOW_LIMIT
1932
#define FLOW_LIMIT_HISTORY	(1 << 7)  /* must be ^2 and !overflow buckets */
1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943
struct sd_flow_limit {
	u64			count;
	unsigned int		num_buckets;
	unsigned int		history_head;
	u16			history[FLOW_LIMIT_HISTORY];
	u8			buckets[];
};

extern int netdev_flow_limit_table_len;
#endif /* CONFIG_NET_FLOW_LIMIT */

L
Linus Torvalds 已提交
1944
/*
E
Eric Dumazet 已提交
1945
 * Incoming packets are placed on per-cpu queues
L
Linus Torvalds 已提交
1946
 */
E
Eric Dumazet 已提交
1947
struct softnet_data {
1948
	struct Qdisc		*output_queue;
1949
	struct Qdisc		**output_queue_tailp;
L
Linus Torvalds 已提交
1950 1951
	struct list_head	poll_list;
	struct sk_buff		*completion_queue;
1952
	struct sk_buff_head	process_queue;
L
Linus Torvalds 已提交
1953

C
Changli Gao 已提交
1954
	/* stats */
1955 1956 1957 1958
	unsigned int		processed;
	unsigned int		time_squeeze;
	unsigned int		cpu_collision;
	unsigned int		received_rps;
C
Changli Gao 已提交
1959

1960
#ifdef CONFIG_RPS
E
Eric Dumazet 已提交
1961 1962 1963
	struct softnet_data	*rps_ipi_list;

	/* Elements below can be accessed between CPUs for RPS */
T
Tom Herbert 已提交
1964
	struct call_single_data	csd ____cacheline_aligned_in_smp;
E
Eric Dumazet 已提交
1965 1966
	struct softnet_data	*rps_ipi_next;
	unsigned int		cpu;
T
Tom Herbert 已提交
1967
	unsigned int		input_queue_head;
1968
	unsigned int		input_queue_tail;
1969
#endif
1970
	unsigned int		dropped;
T
Tom Herbert 已提交
1971
	struct sk_buff_head	input_pkt_queue;
1972
	struct napi_struct	backlog;
1973 1974

#ifdef CONFIG_NET_FLOW_LIMIT
1975
	struct sd_flow_limit __rcu *flow_limit;
1976
#endif
L
Linus Torvalds 已提交
1977 1978
};

1979
static inline void input_queue_head_incr(struct softnet_data *sd)
T
Tom Herbert 已提交
1980 1981
{
#ifdef CONFIG_RPS
1982 1983 1984 1985 1986 1987 1988 1989 1990
	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
1991 1992 1993
#endif
}

T
Tom Herbert 已提交
1994
DECLARE_PER_CPU_ALIGNED(struct softnet_data, softnet_data);
L
Linus Torvalds 已提交
1995

1996
void __netif_schedule(struct Qdisc *q);
L
Linus Torvalds 已提交
1997

1998
static inline void netif_schedule_queue(struct netdev_queue *txq)
L
Linus Torvalds 已提交
1999
{
2000
	if (!(txq->state & QUEUE_STATE_ANY_XOFF))
2001
		__netif_schedule(txq->qdisc);
2002 2003
}

2004 2005 2006 2007 2008 2009 2010 2011
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));
}

2012 2013
static inline void netif_tx_start_queue(struct netdev_queue *dev_queue)
{
2014
	clear_bit(__QUEUE_STATE_DRV_XOFF, &dev_queue->state);
2015 2016
}

2017 2018 2019 2020 2021 2022
/**
 *	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
2023 2024
static inline void netif_start_queue(struct net_device *dev)
{
2025
	netif_tx_start_queue(netdev_get_tx_queue(dev, 0));
L
Linus Torvalds 已提交
2026 2027
}

2028 2029 2030 2031 2032 2033 2034 2035 2036 2037
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);
	}
}

2038
static inline void netif_tx_wake_queue(struct netdev_queue *dev_queue)
L
Linus Torvalds 已提交
2039 2040
{
#ifdef CONFIG_NETPOLL_TRAP
2041
	if (netpoll_trap()) {
2042
		netif_tx_start_queue(dev_queue);
L
Linus Torvalds 已提交
2043
		return;
2044
	}
L
Linus Torvalds 已提交
2045
#endif
2046
	if (test_and_clear_bit(__QUEUE_STATE_DRV_XOFF, &dev_queue->state))
2047
		__netif_schedule(dev_queue->qdisc);
2048 2049
}

2050 2051 2052 2053 2054 2055 2056
/**
 *	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.
 */
2057 2058
static inline void netif_wake_queue(struct net_device *dev)
{
2059
	netif_tx_wake_queue(netdev_get_tx_queue(dev, 0));
L
Linus Torvalds 已提交
2060 2061
}

2062 2063 2064 2065 2066 2067 2068 2069 2070 2071
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);
	}
}

2072 2073
static inline void netif_tx_stop_queue(struct netdev_queue *dev_queue)
{
2074
	if (WARN_ON(!dev_queue)) {
2075
		pr_info("netif_stop_queue() cannot be called before register_netdev()\n");
2076 2077
		return;
	}
2078
	set_bit(__QUEUE_STATE_DRV_XOFF, &dev_queue->state);
2079 2080
}

2081 2082 2083 2084 2085 2086 2087
/**
 *	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
2088 2089
static inline void netif_stop_queue(struct net_device *dev)
{
2090
	netif_tx_stop_queue(netdev_get_tx_queue(dev, 0));
L
Linus Torvalds 已提交
2091 2092
}

2093 2094 2095 2096 2097 2098 2099 2100 2101 2102
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);
	}
}

2103
static inline bool netif_tx_queue_stopped(const struct netdev_queue *dev_queue)
2104
{
2105
	return test_bit(__QUEUE_STATE_DRV_XOFF, &dev_queue->state);
2106 2107
}

2108 2109 2110 2111 2112 2113
/**
 *	netif_queue_stopped - test if transmit queue is flowblocked
 *	@dev: network device
 *
 *	Test if transmit queue on device is currently unable to send.
 */
2114
static inline bool netif_queue_stopped(const struct net_device *dev)
L
Linus Torvalds 已提交
2115
{
2116
	return netif_tx_queue_stopped(netdev_get_tx_queue(dev, 0));
L
Linus Torvalds 已提交
2117 2118
}

2119
static inline bool netif_xmit_stopped(const struct netdev_queue *dev_queue)
2120
{
2121 2122 2123
	return dev_queue->state & QUEUE_STATE_ANY_XOFF;
}

2124
static inline bool netif_xmit_frozen_or_stopped(const struct netdev_queue *dev_queue)
2125 2126 2127 2128
{
	return dev_queue->state & QUEUE_STATE_ANY_XOFF_OR_FROZEN;
}

2129 2130 2131
static inline void netdev_tx_sent_queue(struct netdev_queue *dev_queue,
					unsigned int bytes)
{
T
Tom Herbert 已提交
2132 2133
#ifdef CONFIG_BQL
	dql_queued(&dev_queue->dql, bytes);
2134 2135 2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146 2147 2148 2149

	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
2150
#endif
2151 2152
}

2153 2154 2155 2156 2157 2158 2159 2160 2161
/**
 * 	netdev_sent_queue - report the number of bytes queued to hardware
 * 	@dev: network device
 * 	@bytes: number of bytes queued to the hardware device queue
 *
 * 	Report the number of bytes queued for sending/completion to the network
 * 	device hardware queue. @bytes should be a good approximation and should
 * 	exactly match netdev_completed_queue() @bytes
 */
2162 2163 2164 2165 2166 2167
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,
2168
					     unsigned int pkts, unsigned int bytes)
2169
{
T
Tom Herbert 已提交
2170
#ifdef CONFIG_BQL
2171 2172 2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 2183 2184 2185 2186 2187
	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
2188
#endif
2189 2190
}

2191 2192 2193 2194 2195 2196 2197 2198 2199 2200
/**
 * 	netdev_completed_queue - report bytes and packets completed by device
 * 	@dev: network device
 * 	@pkts: actual number of packets sent over the medium
 * 	@bytes: actual number of bytes sent over the medium
 *
 * 	Report the number of bytes and packets transmitted by the network device
 * 	hardware queue over the physical medium, @bytes must exactly match the
 * 	@bytes amount passed to netdev_sent_queue()
 */
2201
static inline void netdev_completed_queue(struct net_device *dev,
2202
					  unsigned int pkts, unsigned int bytes)
2203 2204 2205 2206 2207 2208
{
	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
2209
#ifdef CONFIG_BQL
2210
	clear_bit(__QUEUE_STATE_STACK_XOFF, &q->state);
T
Tom Herbert 已提交
2211 2212
	dql_reset(&q->dql);
#endif
2213 2214
}

2215 2216 2217 2218 2219 2220 2221
/**
 * 	netdev_reset_queue - reset the packets and bytes count of a network device
 * 	@dev_queue: network device
 *
 * 	Reset the bytes and packet count of a network device and clear the
 * 	software flow control OFF bit for this network device
 */
2222 2223 2224
static inline void netdev_reset_queue(struct net_device *dev_queue)
{
	netdev_tx_reset_queue(netdev_get_tx_queue(dev_queue, 0));
2225 2226
}

2227 2228 2229 2230 2231 2232
/**
 *	netif_running - test if up
 *	@dev: network device
 *
 *	Test if the device has been brought up.
 */
2233
static inline bool netif_running(const struct net_device *dev)
L
Linus Torvalds 已提交
2234 2235 2236 2237
{
	return test_bit(__LINK_STATE_START, &dev->state);
}

2238 2239 2240 2241 2242 2243
/*
 * 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.
 */
2244 2245 2246 2247 2248 2249 2250 2251

/**
 *	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.
 */
2252 2253
static inline void netif_start_subqueue(struct net_device *dev, u16 queue_index)
{
2254
	struct netdev_queue *txq = netdev_get_tx_queue(dev, queue_index);
2255 2256

	netif_tx_start_queue(txq);
2257 2258
}

2259 2260 2261 2262 2263 2264 2265
/**
 *	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.
 */
2266 2267
static inline void netif_stop_subqueue(struct net_device *dev, u16 queue_index)
{
2268
	struct netdev_queue *txq = netdev_get_tx_queue(dev, queue_index);
2269 2270 2271 2272
#ifdef CONFIG_NETPOLL_TRAP
	if (netpoll_trap())
		return;
#endif
2273
	netif_tx_stop_queue(txq);
2274 2275
}

2276 2277 2278 2279 2280 2281 2282
/**
 *	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.
 */
2283 2284
static inline bool __netif_subqueue_stopped(const struct net_device *dev,
					    u16 queue_index)
2285
{
2286
	struct netdev_queue *txq = netdev_get_tx_queue(dev, queue_index);
2287 2288

	return netif_tx_queue_stopped(txq);
2289 2290
}

2291 2292
static inline bool netif_subqueue_stopped(const struct net_device *dev,
					  struct sk_buff *skb)
2293 2294 2295
{
	return __netif_subqueue_stopped(dev, skb_get_queue_mapping(skb));
}
2296 2297 2298 2299 2300 2301 2302 2303

/**
 *	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.
 */
2304 2305
static inline void netif_wake_subqueue(struct net_device *dev, u16 queue_index)
{
2306
	struct netdev_queue *txq = netdev_get_tx_queue(dev, queue_index);
2307 2308 2309 2310
#ifdef CONFIG_NETPOLL_TRAP
	if (netpoll_trap())
		return;
#endif
2311
	if (test_and_clear_bit(__QUEUE_STATE_DRV_XOFF, &txq->state))
2312
		__netif_schedule(txq->qdisc);
2313 2314
}

2315
#ifdef CONFIG_XPS
2316
int netif_set_xps_queue(struct net_device *dev, const struct cpumask *mask,
2317
			u16 index);
2318 2319
#else
static inline int netif_set_xps_queue(struct net_device *dev,
2320
				      const struct cpumask *mask,
2321 2322 2323 2324 2325 2326
				      u16 index)
{
	return 0;
}
#endif

2327 2328 2329 2330 2331 2332 2333 2334 2335 2336
/*
 * 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);
}

2337 2338 2339 2340 2341 2342
/**
 *	netif_is_multiqueue - test if device has multiple transmit queues
 *	@dev: network device
 *
 * Check if device has multiple transmit queues
 */
2343
static inline bool netif_is_multiqueue(const struct net_device *dev)
2344
{
E
Eric Dumazet 已提交
2345
	return dev->num_tx_queues > 1;
2346
}
L
Linus Torvalds 已提交
2347

2348
int netif_set_real_num_tx_queues(struct net_device *dev, unsigned int txq);
2349

2350
#ifdef CONFIG_RPS
2351
int netif_set_real_num_rx_queues(struct net_device *dev, unsigned int rxq);
2352 2353 2354 2355 2356 2357 2358 2359
#else
static inline int netif_set_real_num_rx_queues(struct net_device *dev,
						unsigned int rxq)
{
	return 0;
}
#endif

2360 2361 2362
static inline int netif_copy_real_num_queues(struct net_device *to_dev,
					     const struct net_device *from_dev)
{
2363 2364 2365 2366 2367 2368
	int err;

	err = netif_set_real_num_tx_queues(to_dev,
					   from_dev->real_num_tx_queues);
	if (err)
		return err;
2369 2370 2371 2372 2373 2374 2375 2376
#ifdef CONFIG_RPS
	return netif_set_real_num_rx_queues(to_dev,
					    from_dev->real_num_rx_queues);
#else
	return 0;
#endif
}

2377
#define DEFAULT_MAX_NUM_RSS_QUEUES	(8)
2378
int netif_get_num_default_rss_queues(void);
2379

2380 2381 2382 2383 2384 2385 2386
enum skb_free_reason {
	SKB_REASON_CONSUMED,
	SKB_REASON_DROPPED,
};

void __dev_kfree_skb_irq(struct sk_buff *skb, enum skb_free_reason reason);
void __dev_kfree_skb_any(struct sk_buff *skb, enum skb_free_reason reason);
L
Linus Torvalds 已提交
2387

2388 2389 2390 2391 2392 2393 2394 2395 2396 2397 2398 2399 2400 2401 2402 2403 2404 2405
/*
 * It is not allowed to call kfree_skb() or consume_skb() from hardware
 * interrupt context or with hardware interrupts being disabled.
 * (in_irq() || irqs_disabled())
 *
 * We provide four helpers that can be used in following contexts :
 *
 * dev_kfree_skb_irq(skb) when caller drops a packet from irq context,
 *  replacing kfree_skb(skb)
 *
 * dev_consume_skb_irq(skb) when caller consumes a packet from irq context.
 *  Typically used in place of consume_skb(skb) in TX completion path
 *
 * dev_kfree_skb_any(skb) when caller doesn't know its current irq context,
 *  replacing kfree_skb(skb)
 *
 * dev_consume_skb_any(skb) when caller doesn't know its current irq context,
 *  and consumed a packet. Used in place of consume_skb(skb)
L
Linus Torvalds 已提交
2406
 */
2407 2408 2409 2410 2411 2412 2413 2414 2415 2416 2417 2418 2419 2420 2421 2422 2423 2424 2425
static inline void dev_kfree_skb_irq(struct sk_buff *skb)
{
	__dev_kfree_skb_irq(skb, SKB_REASON_DROPPED);
}

static inline void dev_consume_skb_irq(struct sk_buff *skb)
{
	__dev_kfree_skb_irq(skb, SKB_REASON_CONSUMED);
}

static inline void dev_kfree_skb_any(struct sk_buff *skb)
{
	__dev_kfree_skb_any(skb, SKB_REASON_DROPPED);
}

static inline void dev_consume_skb_any(struct sk_buff *skb)
{
	__dev_kfree_skb_any(skb, SKB_REASON_CONSUMED);
}
L
Linus Torvalds 已提交
2426

2427 2428 2429 2430 2431 2432 2433
int netif_rx(struct sk_buff *skb);
int netif_rx_ni(struct sk_buff *skb);
int netif_receive_skb(struct sk_buff *skb);
gro_result_t napi_gro_receive(struct napi_struct *napi, struct sk_buff *skb);
void napi_gro_flush(struct napi_struct *napi, bool flush_old);
struct sk_buff *napi_get_frags(struct napi_struct *napi);
gro_result_t napi_gro_frags(struct napi_struct *napi);
2434 2435 2436 2437 2438 2439 2440

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

2441 2442 2443 2444 2445 2446 2447 2448 2449 2450 2451
int netdev_rx_handler_register(struct net_device *dev,
			       rx_handler_func_t *rx_handler,
			       void *rx_handler_data);
void netdev_rx_handler_unregister(struct net_device *dev);

bool dev_valid_name(const char *name);
int dev_ioctl(struct net *net, unsigned int cmd, void __user *);
int dev_ethtool(struct net *net, struct ifreq *);
unsigned int dev_get_flags(const struct net_device *);
int __dev_change_flags(struct net_device *, unsigned int flags);
int dev_change_flags(struct net_device *, unsigned int);
2452 2453
void __dev_notify_flags(struct net_device *, unsigned int old_flags,
			unsigned int gchanges);
2454 2455 2456 2457 2458 2459 2460 2461 2462 2463
int dev_change_name(struct net_device *, const char *);
int dev_set_alias(struct net_device *, const char *, size_t);
int dev_change_net_namespace(struct net_device *, struct net *, const char *);
int dev_set_mtu(struct net_device *, int);
void dev_set_group(struct net_device *, int);
int dev_set_mac_address(struct net_device *, struct sockaddr *);
int dev_change_carrier(struct net_device *, bool new_carrier);
int dev_get_phys_port_id(struct net_device *dev,
			 struct netdev_phys_port_id *ppid);
int dev_hard_start_xmit(struct sk_buff *skb, struct net_device *dev,
2464
			struct netdev_queue *txq, void *accel_priv);
2465
int dev_forward_skb(struct net_device *dev, struct sk_buff *skb);
L
Linus Torvalds 已提交
2466

2467
extern int		netdev_budget;
L
Linus Torvalds 已提交
2468 2469

/* Called by rtnetlink.c:rtnl_unlock() */
2470
void netdev_run_todo(void);
L
Linus Torvalds 已提交
2471

2472 2473 2474 2475
/**
 *	dev_put - release reference to device
 *	@dev: network device
 *
2476
 * Release reference to device to allow it to be freed.
2477
 */
L
Linus Torvalds 已提交
2478 2479
static inline void dev_put(struct net_device *dev)
{
2480
	this_cpu_dec(*dev->pcpu_refcnt);
L
Linus Torvalds 已提交
2481 2482
}

2483 2484 2485 2486
/**
 *	dev_hold - get reference to device
 *	@dev: network device
 *
2487
 * Hold reference to device to keep it from being freed.
2488
 */
2489 2490
static inline void dev_hold(struct net_device *dev)
{
2491
	this_cpu_inc(*dev->pcpu_refcnt);
2492
}
L
Linus Torvalds 已提交
2493 2494 2495 2496

/* 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
2497 2498 2499 2500
 *
 * 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
2501 2502
 */

2503 2504 2505
void linkwatch_init_dev(struct net_device *dev);
void linkwatch_fire_event(struct net_device *dev);
void linkwatch_forget_dev(struct net_device *dev);
L
Linus Torvalds 已提交
2506

2507 2508 2509 2510 2511 2512
/**
 *	netif_carrier_ok - test if carrier present
 *	@dev: network device
 *
 * Check if carrier is present on device
 */
2513
static inline bool netif_carrier_ok(const struct net_device *dev)
L
Linus Torvalds 已提交
2514 2515 2516 2517
{
	return !test_bit(__LINK_STATE_NOCARRIER, &dev->state);
}

2518
unsigned long dev_trans_start(struct net_device *dev);
2519

2520
void __netdev_watchdog_up(struct net_device *dev);
L
Linus Torvalds 已提交
2521

2522
void netif_carrier_on(struct net_device *dev);
L
Linus Torvalds 已提交
2523

2524
void netif_carrier_off(struct net_device *dev);
L
Linus Torvalds 已提交
2525

2526 2527 2528 2529 2530 2531 2532 2533 2534 2535 2536 2537 2538
/**
 *	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
2539 2540 2541 2542 2543 2544
static inline void netif_dormant_on(struct net_device *dev)
{
	if (!test_and_set_bit(__LINK_STATE_DORMANT, &dev->state))
		linkwatch_fire_event(dev);
}

2545 2546 2547 2548 2549 2550
/**
 *	netif_dormant_off - set device as not dormant.
 *	@dev: network device
 *
 * Device is not in dormant state.
 */
S
Stefan Rompf 已提交
2551 2552 2553 2554 2555 2556
static inline void netif_dormant_off(struct net_device *dev)
{
	if (test_and_clear_bit(__LINK_STATE_DORMANT, &dev->state))
		linkwatch_fire_event(dev);
}

2557 2558 2559 2560 2561 2562
/**
 *	netif_dormant - test if carrier present
 *	@dev: network device
 *
 * Check if carrier is present on device
 */
2563
static inline bool netif_dormant(const struct net_device *dev)
S
Stefan Rompf 已提交
2564 2565 2566 2567 2568
{
	return test_bit(__LINK_STATE_DORMANT, &dev->state);
}


2569 2570 2571 2572 2573 2574
/**
 *	netif_oper_up - test if device is operational
 *	@dev: network device
 *
 * Check if carrier is operational
 */
2575
static inline bool netif_oper_up(const struct net_device *dev)
E
Eric Dumazet 已提交
2576
{
S
Stefan Rompf 已提交
2577 2578 2579 2580
	return (dev->operstate == IF_OPER_UP ||
		dev->operstate == IF_OPER_UNKNOWN /* backward compat */);
}

2581 2582 2583 2584 2585 2586
/**
 *	netif_device_present - is device available or removed
 *	@dev: network device
 *
 * Check if device has not been removed from system.
 */
2587
static inline bool netif_device_present(struct net_device *dev)
L
Linus Torvalds 已提交
2588 2589 2590 2591
{
	return test_bit(__LINK_STATE_PRESENT, &dev->state);
}

2592
void netif_device_detach(struct net_device *dev);
L
Linus Torvalds 已提交
2593

2594
void netif_device_attach(struct net_device *dev);
L
Linus Torvalds 已提交
2595 2596 2597 2598 2599 2600 2601 2602 2603 2604 2605 2606 2607 2608 2609 2610 2611 2612 2613 2614 2615 2616 2617 2618 2619 2620 2621 2622 2623 2624 2625 2626 2627 2628 2629 2630 2631 2632 2633 2634 2635 2636 2637 2638 2639 2640 2641 2642 2643 2644

/*
 * 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;
}

2645
static inline void __netif_tx_lock(struct netdev_queue *txq, int cpu)
H
Herbert Xu 已提交
2646
{
2647 2648
	spin_lock(&txq->_xmit_lock);
	txq->xmit_lock_owner = cpu;
2649 2650
}

2651 2652 2653 2654 2655 2656
static inline void __netif_tx_lock_bh(struct netdev_queue *txq)
{
	spin_lock_bh(&txq->_xmit_lock);
	txq->xmit_lock_owner = smp_processor_id();
}

2657
static inline bool __netif_tx_trylock(struct netdev_queue *txq)
2658
{
2659
	bool ok = spin_trylock(&txq->_xmit_lock);
2660 2661 2662 2663 2664 2665 2666 2667 2668 2669 2670 2671 2672 2673 2674 2675 2676
	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
2677 2678 2679 2680 2681 2682
static inline void txq_trans_update(struct netdev_queue *txq)
{
	if (txq->xmit_lock_owner != -1)
		txq->trans_start = jiffies;
}

2683 2684 2685 2686 2687 2688
/**
 *	netif_tx_lock - grab network device transmit lock
 *	@dev: network device
 *
 * Get network device transmit lock
 */
2689 2690
static inline void netif_tx_lock(struct net_device *dev)
{
2691
	unsigned int i;
2692
	int cpu;
2693

2694 2695
	spin_lock(&dev->tx_global_lock);
	cpu = smp_processor_id();
2696 2697
	for (i = 0; i < dev->num_tx_queues; i++) {
		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
2698 2699 2700 2701 2702 2703 2704

		/* 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.
		 */
2705
		__netif_tx_lock(txq, cpu);
2706 2707
		set_bit(__QUEUE_STATE_FROZEN, &txq->state);
		__netif_tx_unlock(txq);
2708
	}
H
Herbert Xu 已提交
2709 2710 2711 2712
}

static inline void netif_tx_lock_bh(struct net_device *dev)
{
2713 2714
	local_bh_disable();
	netif_tx_lock(dev);
H
Herbert Xu 已提交
2715 2716 2717 2718
}

static inline void netif_tx_unlock(struct net_device *dev)
{
2719 2720 2721 2722
	unsigned int i;

	for (i = 0; i < dev->num_tx_queues; i++) {
		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
2723

2724 2725 2726 2727 2728
		/* 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);
2729
		netif_schedule_queue(txq);
2730 2731
	}
	spin_unlock(&dev->tx_global_lock);
H
Herbert Xu 已提交
2732 2733 2734 2735
}

static inline void netif_tx_unlock_bh(struct net_device *dev)
{
2736 2737
	netif_tx_unlock(dev);
	local_bh_enable();
H
Herbert Xu 已提交
2738 2739
}

2740
#define HARD_TX_LOCK(dev, txq, cpu) {			\
2741
	if ((dev->features & NETIF_F_LLTX) == 0) {	\
2742
		__netif_tx_lock(txq, cpu);		\
2743 2744 2745
	}						\
}

2746
#define HARD_TX_UNLOCK(dev, txq) {			\
2747
	if ((dev->features & NETIF_F_LLTX) == 0) {	\
2748
		__netif_tx_unlock(txq);			\
2749 2750 2751
	}						\
}

L
Linus Torvalds 已提交
2752 2753
static inline void netif_tx_disable(struct net_device *dev)
{
2754
	unsigned int i;
2755
	int cpu;
2756

2757 2758
	local_bh_disable();
	cpu = smp_processor_id();
2759 2760
	for (i = 0; i < dev->num_tx_queues; i++) {
		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
2761 2762

		__netif_tx_lock(txq, cpu);
2763
		netif_tx_stop_queue(txq);
2764
		__netif_tx_unlock(txq);
2765
	}
2766
	local_bh_enable();
L
Linus Torvalds 已提交
2767 2768
}

2769 2770 2771 2772 2773
static inline void netif_addr_lock(struct net_device *dev)
{
	spin_lock(&dev->addr_list_lock);
}

2774 2775 2776 2777 2778
static inline void netif_addr_lock_nested(struct net_device *dev)
{
	spin_lock_nested(&dev->addr_list_lock, SINGLE_DEPTH_NESTING);
}

2779 2780 2781 2782 2783 2784 2785 2786 2787 2788 2789 2790 2791 2792 2793
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);
}

2794
/*
2795
 * dev_addrs walker. Should be used only for read access. Call with
2796 2797 2798
 * rcu_read_lock held.
 */
#define for_each_dev_addr(dev, ha) \
2799
		list_for_each_entry_rcu(ha, &dev->dev_addrs.list, list)
2800

L
Linus Torvalds 已提交
2801 2802
/* These functions live elsewhere (drivers/net/net_init.c, but related) */

2803
void ether_setup(struct net_device *dev);
L
Linus Torvalds 已提交
2804 2805

/* Support for loadable net-drivers */
2806 2807 2808
struct net_device *alloc_netdev_mqs(int sizeof_priv, const char *name,
				    void (*setup)(struct net_device *),
				    unsigned int txqs, unsigned int rxqs);
2809
#define alloc_netdev(sizeof_priv, name, setup) \
T
Tom Herbert 已提交
2810 2811 2812 2813 2814
	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)

2815 2816
int register_netdev(struct net_device *dev);
void unregister_netdev(struct net_device *dev);
2817

2818
/* General hardware address lists handling functions */
2819 2820 2821 2822 2823
int __hw_addr_sync(struct netdev_hw_addr_list *to_list,
		   struct netdev_hw_addr_list *from_list, int addr_len);
void __hw_addr_unsync(struct netdev_hw_addr_list *to_list,
		      struct netdev_hw_addr_list *from_list, int addr_len);
void __hw_addr_init(struct netdev_hw_addr_list *list);
2824

2825
/* Functions used for device addresses handling */
2826 2827 2828 2829 2830 2831
int dev_addr_add(struct net_device *dev, const unsigned char *addr,
		 unsigned char addr_type);
int dev_addr_del(struct net_device *dev, const unsigned char *addr,
		 unsigned char addr_type);
void dev_addr_flush(struct net_device *dev);
int dev_addr_init(struct net_device *dev);
2832 2833

/* Functions used for unicast addresses handling */
2834 2835 2836 2837 2838 2839 2840 2841
int dev_uc_add(struct net_device *dev, const unsigned char *addr);
int dev_uc_add_excl(struct net_device *dev, const unsigned char *addr);
int dev_uc_del(struct net_device *dev, const unsigned char *addr);
int dev_uc_sync(struct net_device *to, struct net_device *from);
int dev_uc_sync_multiple(struct net_device *to, struct net_device *from);
void dev_uc_unsync(struct net_device *to, struct net_device *from);
void dev_uc_flush(struct net_device *dev);
void dev_uc_init(struct net_device *dev);
2842

2843
/* Functions used for multicast addresses handling */
2844 2845 2846 2847 2848 2849 2850 2851 2852 2853
int dev_mc_add(struct net_device *dev, const unsigned char *addr);
int dev_mc_add_global(struct net_device *dev, const unsigned char *addr);
int dev_mc_add_excl(struct net_device *dev, const unsigned char *addr);
int dev_mc_del(struct net_device *dev, const unsigned char *addr);
int dev_mc_del_global(struct net_device *dev, const unsigned char *addr);
int dev_mc_sync(struct net_device *to, struct net_device *from);
int dev_mc_sync_multiple(struct net_device *to, struct net_device *from);
void dev_mc_unsync(struct net_device *to, struct net_device *from);
void dev_mc_flush(struct net_device *dev);
void dev_mc_init(struct net_device *dev);
2854

2855
/* Functions used for secondary unicast and multicast support */
2856 2857 2858 2859 2860 2861 2862
void dev_set_rx_mode(struct net_device *dev);
void __dev_set_rx_mode(struct net_device *dev);
int dev_set_promiscuity(struct net_device *dev, int inc);
int dev_set_allmulti(struct net_device *dev, int inc);
void netdev_state_change(struct net_device *dev);
void netdev_notify_peers(struct net_device *dev);
void netdev_features_change(struct net_device *dev);
L
Linus Torvalds 已提交
2863
/* Load a device via the kmod */
2864 2865 2866 2867 2868
void dev_load(struct net *net, const char *name);
struct rtnl_link_stats64 *dev_get_stats(struct net_device *dev,
					struct rtnl_link_stats64 *storage);
void netdev_stats_to_stats64(struct rtnl_link_stats64 *stats64,
			     const struct net_device_stats *netdev_stats);
2869

L
Linus Torvalds 已提交
2870
extern int		netdev_max_backlog;
E
Eric Dumazet 已提交
2871
extern int		netdev_tstamp_prequeue;
L
Linus Torvalds 已提交
2872
extern int		weight_p;
2873
extern int		bpf_jit_enable;
J
Jiri Pirko 已提交
2874

2875 2876 2877
bool netdev_has_upper_dev(struct net_device *dev, struct net_device *upper_dev);
struct net_device *netdev_all_upper_get_next_dev_rcu(struct net_device *dev,
						     struct list_head **iter);
2878 2879

/* iterate through upper list, must be called under RCU read lock */
2880 2881 2882 2883 2884
#define netdev_for_each_all_upper_dev_rcu(dev, updev, iter) \
	for (iter = &(dev)->all_adj_list.upper, \
	     updev = netdev_all_upper_get_next_dev_rcu(dev, &(iter)); \
	     updev; \
	     updev = netdev_all_upper_get_next_dev_rcu(dev, &(iter)))
2885

2886 2887 2888 2889
void *netdev_lower_get_next_private(struct net_device *dev,
				    struct list_head **iter);
void *netdev_lower_get_next_private_rcu(struct net_device *dev,
					struct list_head **iter);
2890 2891 2892 2893 2894 2895 2896 2897 2898 2899 2900 2901 2902

#define netdev_for_each_lower_private(dev, priv, iter) \
	for (iter = (dev)->adj_list.lower.next, \
	     priv = netdev_lower_get_next_private(dev, &(iter)); \
	     priv; \
	     priv = netdev_lower_get_next_private(dev, &(iter)))

#define netdev_for_each_lower_private_rcu(dev, priv, iter) \
	for (iter = &(dev)->adj_list.lower, \
	     priv = netdev_lower_get_next_private_rcu(dev, &(iter)); \
	     priv; \
	     priv = netdev_lower_get_next_private_rcu(dev, &(iter)))

2903
void *netdev_adjacent_get_private(struct list_head *adj_list);
2904
void *netdev_lower_get_first_private_rcu(struct net_device *dev);
2905 2906 2907 2908
struct net_device *netdev_master_upper_dev_get(struct net_device *dev);
struct net_device *netdev_master_upper_dev_get_rcu(struct net_device *dev);
int netdev_upper_dev_link(struct net_device *dev, struct net_device *upper_dev);
int netdev_master_upper_dev_link(struct net_device *dev,
J
Jiri Pirko 已提交
2909
				 struct net_device *upper_dev);
2910 2911 2912 2913 2914 2915 2916 2917 2918 2919 2920 2921
int netdev_master_upper_dev_link_private(struct net_device *dev,
					 struct net_device *upper_dev,
					 void *private);
void netdev_upper_dev_unlink(struct net_device *dev,
			     struct net_device *upper_dev);
void *netdev_lower_dev_get_private(struct net_device *dev,
				   struct net_device *lower_dev);
int skb_checksum_help(struct sk_buff *skb);
struct sk_buff *__skb_gso_segment(struct sk_buff *skb,
				  netdev_features_t features, bool tx_path);
struct sk_buff *skb_mac_gso_segment(struct sk_buff *skb,
				    netdev_features_t features);
2922 2923 2924 2925 2926 2927

static inline
struct sk_buff *skb_gso_segment(struct sk_buff *skb, netdev_features_t features)
{
	return __skb_gso_segment(skb, features, true);
}
2928 2929 2930 2931 2932 2933 2934 2935 2936 2937 2938 2939 2940
__be16 skb_network_protocol(struct sk_buff *skb);

static inline bool can_checksum_protocol(netdev_features_t features,
					 __be16 protocol)
{
	return ((features & NETIF_F_GEN_CSUM) ||
		((features & NETIF_F_V4_CSUM) &&
		 protocol == htons(ETH_P_IP)) ||
		((features & NETIF_F_V6_CSUM) &&
		 protocol == htons(ETH_P_IPV6)) ||
		((features & NETIF_F_FCOE_CRC) &&
		 protocol == htons(ETH_P_FCOE)));
}
2941

2942
#ifdef CONFIG_BUG
2943
void netdev_rx_csum_fault(struct net_device *dev);
2944 2945 2946 2947 2948
#else
static inline void netdev_rx_csum_fault(struct net_device *dev)
{
}
#endif
L
Linus Torvalds 已提交
2949
/* rx skb timestamps */
2950 2951
void net_enable_timestamp(void);
void net_disable_timestamp(void);
L
Linus Torvalds 已提交
2952

2953
#ifdef CONFIG_PROC_FS
2954
int __init dev_proc_init(void);
2955 2956
#else
#define dev_proc_init() 0
2957 2958
#endif

2959 2960 2961 2962
int netdev_class_create_file_ns(struct class_attribute *class_attr,
				const void *ns);
void netdev_class_remove_file_ns(struct class_attribute *class_attr,
				 const void *ns);
2963 2964 2965 2966 2967 2968 2969 2970 2971 2972

static inline int netdev_class_create_file(struct class_attribute *class_attr)
{
	return netdev_class_create_file_ns(class_attr, NULL);
}

static inline void netdev_class_remove_file(struct class_attribute *class_attr)
{
	netdev_class_remove_file_ns(class_attr, NULL);
}
2973

2974 2975
extern struct kobj_ns_type_operations net_ns_type_operations;

2976
const char *netdev_drivername(const struct net_device *dev);
2977

2978
void linkwatch_run_queue(void);
2979

2980 2981
static inline netdev_features_t netdev_get_wanted_features(
	struct net_device *dev)
2982 2983 2984
{
	return (dev->features & ~dev->hw_features) | dev->wanted_features;
}
2985 2986
netdev_features_t netdev_increment_features(netdev_features_t all,
	netdev_features_t one, netdev_features_t mask);
2987 2988 2989 2990 2991 2992 2993 2994 2995 2996 2997

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

2998
int __netdev_update_features(struct net_device *dev);
2999
void netdev_update_features(struct net_device *dev);
3000
void netdev_change_features(struct net_device *dev);
3001

3002 3003 3004
void netif_stacked_transfer_operstate(const struct net_device *rootdev,
					struct net_device *dev);

3005
netdev_features_t netif_skb_features(struct sk_buff *skb);
3006

3007
static inline bool net_gso_ok(netdev_features_t features, int gso_type)
3008
{
3009
	netdev_features_t feature = gso_type << NETIF_F_GSO_SHIFT;
3010 3011 3012 3013 3014 3015 3016 3017 3018

	/* 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));

3019
	return (features & feature) == feature;
3020 3021
}

3022
static inline bool skb_gso_ok(struct sk_buff *skb, netdev_features_t features)
3023
{
H
Herbert Xu 已提交
3024
	return net_gso_ok(features, skb_shinfo(skb)->gso_type) &&
3025
	       (!skb_has_frag_list(skb) || (features & NETIF_F_FRAGLIST));
3026 3027
}

3028 3029
static inline bool netif_needs_gso(struct sk_buff *skb,
				   netdev_features_t features)
3030
{
3031
	return skb_is_gso(skb) && (!skb_gso_ok(skb, features) ||
3032 3033
		unlikely((skb->ip_summed != CHECKSUM_PARTIAL) &&
			 (skb->ip_summed != CHECKSUM_UNNECESSARY)));
3034 3035
}

3036 3037 3038 3039 3040 3041
static inline void netif_set_gso_max_size(struct net_device *dev,
					  unsigned int size)
{
	dev->gso_max_size = size;
}

3042 3043 3044 3045 3046
static inline bool netif_is_macvlan(struct net_device *dev)
{
	return dev->priv_flags & IFF_MACVLAN;
}

3047 3048 3049 3050 3051
static inline bool netif_is_bond_master(struct net_device *dev)
{
	return dev->flags & IFF_MASTER && dev->priv_flags & IFF_BONDING;
}

3052
static inline bool netif_is_bond_slave(struct net_device *dev)
J
Jiri Pirko 已提交
3053 3054 3055 3056
{
	return dev->flags & IFF_SLAVE && dev->priv_flags & IFF_BONDING;
}

3057 3058 3059 3060 3061
static inline bool netif_supports_nofcs(struct net_device *dev)
{
	return dev->priv_flags & IFF_SUPP_NOFCS;
}

3062
extern struct pernet_operations __net_initdata loopback_net_ops;
3063

3064 3065 3066 3067 3068 3069 3070 3071 3072 3073 3074
/* 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;
}

3075
__printf(3, 4)
3076 3077
int netdev_printk(const char *level, const struct net_device *dev,
		  const char *format, ...);
3078
__printf(2, 3)
3079
int netdev_emerg(const struct net_device *dev, const char *format, ...);
3080
__printf(2, 3)
3081
int netdev_alert(const struct net_device *dev, const char *format, ...);
3082
__printf(2, 3)
3083
int netdev_crit(const struct net_device *dev, const char *format, ...);
3084
__printf(2, 3)
3085
int netdev_err(const struct net_device *dev, const char *format, ...);
3086
__printf(2, 3)
3087
int netdev_warn(const struct net_device *dev, const char *format, ...);
3088
__printf(2, 3)
3089
int netdev_notice(const struct net_device *dev, const char *format, ...);
3090
__printf(2, 3)
3091
int netdev_info(const struct net_device *dev, const char *format, ...);
3092

3093 3094 3095
#define MODULE_ALIAS_NETDEV(device) \
	MODULE_ALIAS("netdev-" device)

3096
#if defined(CONFIG_DYNAMIC_DEBUG)
3097 3098
#define netdev_dbg(__dev, format, args...)			\
do {								\
3099
	dynamic_netdev_dbg(__dev, format, ##args);		\
3100
} while (0)
3101 3102 3103
#elif defined(DEBUG)
#define netdev_dbg(__dev, format, args...)			\
	netdev_printk(KERN_DEBUG, __dev, format, ##args)
3104 3105 3106 3107 3108 3109 3110 3111 3112 3113 3114 3115 3116 3117 3118 3119 3120 3121 3122 3123 3124 3125 3126 3127 3128 3129 3130
#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...)			\
3131
	WARN(1, "netdevice: %s\n" format, netdev_name(dev), ##args)
3132

3133 3134 3135 3136 3137 3138 3139 3140
/* 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)

3141 3142 3143 3144 3145 3146
#define netif_level(level, priv, type, dev, fmt, args...)	\
do {								\
	if (netif_msg_##type(priv))				\
		netdev_##level(dev, fmt, ##args);		\
} while (0)

3147
#define netif_emerg(priv, type, dev, fmt, args...)		\
3148
	netif_level(emerg, priv, type, dev, fmt, ##args)
3149
#define netif_alert(priv, type, dev, fmt, args...)		\
3150
	netif_level(alert, priv, type, dev, fmt, ##args)
3151
#define netif_crit(priv, type, dev, fmt, args...)		\
3152
	netif_level(crit, priv, type, dev, fmt, ##args)
3153
#define netif_err(priv, type, dev, fmt, args...)		\
3154
	netif_level(err, priv, type, dev, fmt, ##args)
3155
#define netif_warn(priv, type, dev, fmt, args...)		\
3156
	netif_level(warn, priv, type, dev, fmt, ##args)
3157
#define netif_notice(priv, type, dev, fmt, args...)		\
3158
	netif_level(notice, priv, type, dev, fmt, ##args)
3159
#define netif_info(priv, type, dev, fmt, args...)		\
3160
	netif_level(info, priv, type, dev, fmt, ##args)
3161

3162
#if defined(CONFIG_DYNAMIC_DEBUG)
3163 3164 3165
#define netif_dbg(priv, type, netdev, format, args...)		\
do {								\
	if (netif_msg_##type(priv))				\
3166
		dynamic_netdev_dbg(netdev, format, ##args);	\
3167
} while (0)
3168 3169 3170
#elif defined(DEBUG)
#define netif_dbg(priv, type, dev, format, args...)		\
	netif_printk(priv, type, KERN_DEBUG, dev, format, ##args)
3171 3172 3173 3174 3175 3176 3177 3178 3179 3180
#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)
3181
#define netif_vdbg	netif_dbg
3182 3183 3184 3185
#else
#define netif_vdbg(priv, type, dev, format, args...)		\
({								\
	if (0)							\
3186
		netif_printk(priv, type, KERN_DEBUG, dev, format, ##args); \
3187 3188 3189
	0;							\
})
#endif
3190

3191 3192 3193 3194 3195 3196 3197 3198 3199 3200 3201 3202 3203 3204 3205 3206 3207 3208 3209 3210 3211 3212 3213 3214 3215 3216 3217 3218 3219 3220
/*
 *	The list of packet types we will receive (as opposed to discard)
 *	and the routines to invoke.
 *
 *	Why 16. Because with 16 the only overlap we get on a hash of the
 *	low nibble of the protocol value is RARP/SNAP/X.25.
 *
 *      NOTE:  That is no longer true with the addition of VLAN tags.  Not
 *             sure which should go first, but I bet it won't make much
 *             difference if we are running VLANs.  The good news is that
 *             this protocol won't be in the list unless compiled in, so
 *             the average user (w/out VLANs) will not be adversely affected.
 *             --BLG
 *
 *		0800	IP
 *		8100    802.1Q VLAN
 *		0001	802.3
 *		0002	AX.25
 *		0004	802.2
 *		8035	RARP
 *		0005	SNAP
 *		0805	X.25
 *		0806	ARP
 *		8137	IPX
 *		0009	Localtalk
 *		86DD	IPv6
 */
#define PTYPE_HASH_SIZE	(16)
#define PTYPE_HASH_MASK	(PTYPE_HASH_SIZE - 1)

3221
#endif	/* _LINUX_NETDEVICE_H */