netdevice.h 99.2 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
#endif
671
#endif /* CONFIG_RPS */
672

T
Tom Herbert 已提交
673 674
/* This structure contains an instance of an RX queue. */
struct netdev_rx_queue {
675
#ifdef CONFIG_RPS
E
Eric Dumazet 已提交
676 677
	struct rps_map __rcu		*rps_map;
	struct rps_dev_flow_table __rcu	*rps_flow_table;
678
#endif
E
Eric Dumazet 已提交
679
	struct kobject			kobj;
T
Tom Herbert 已提交
680
	struct net_device		*dev;
T
Tom Herbert 已提交
681
} ____cacheline_aligned_in_smp;
682 683 684 685 686 687 688 689 690 691 692

/*
 * RX queue sysfs structures and functions.
 */
struct rx_queue_attribute {
	struct attribute attr;
	ssize_t (*show)(struct netdev_rx_queue *queue,
	    struct rx_queue_attribute *attr, char *buf);
	ssize_t (*store)(struct netdev_rx_queue *queue,
	    struct rx_queue_attribute *attr, const char *buf, size_t len);
};
693

T
Tom Herbert 已提交
694 695 696 697 698 699 700 701 702 703 704
#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];
};
705
#define XPS_MAP_SIZE(_num) (sizeof(struct xps_map) + ((_num) * sizeof(u16)))
T
Tom Herbert 已提交
706 707 708 709 710 711 712 713
#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
714
	struct xps_map __rcu *cpu_map[0];
T
Tom Herbert 已提交
715 716 717 718 719
};
#define XPS_DEV_MAPS_SIZE (sizeof(struct xps_dev_maps) +		\
    (nr_cpu_ids * sizeof(struct xps_map *)))
#endif /* CONFIG_XPS */

720 721 722 723 724 725 726 727
#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;
};

728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744
#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

745 746 747 748 749 750 751 752 753 754
#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;
};

755 756
/*
 * This structure defines the management hooks for network devices.
757 758
 * The following hooks can be defined; unless noted otherwise, they are
 * optional and can be filled with a null pointer.
759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777
 *
 * 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.
 *
778 779
 * netdev_tx_t (*ndo_start_xmit)(struct sk_buff *skb,
 *                               struct net_device *dev);
780
 *	Called when a packet needs to be transmitted.
781 782
 *	Must return NETDEV_TX_OK , NETDEV_TX_BUSY.
 *        (can also return NETDEV_TX_LOCKED iff NETIF_F_LLTX)
783 784
 *	Required can not be NULL.
 *
785 786
 * u16 (*ndo_select_queue)(struct net_device *dev, struct sk_buff *skb,
 *                         void *accel_priv);
787 788 789
 *	Called to decide which queue to when device supports multiple
 *	transmit queues.
 *
790 791 792 793 794 795
 * 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.
796 797
 *	If driver handles unicast address filtering, it should set
 *	IFF_UNICAST_FLT to its priv_flags.
798 799 800
 *
 * int (*ndo_set_mac_address)(struct net_device *dev, void *addr);
 *	This function  is called when the Media Access Control address
801
 *	needs to be changed. If this interface is not defined, the
802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821
 *	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.
 *
822
 * void (*ndo_tx_timeout)(struct net_device *dev);
823 824 825
 *	Callback uses when the transmitter has not made any progress
 *	for dev->watchdog ticks.
 *
826
 * struct rtnl_link_stats64* (*ndo_get_stats64)(struct net_device *dev,
827
 *                      struct rtnl_link_stats64 *storage);
828
 * struct net_device_stats* (*ndo_get_stats)(struct net_device *dev);
829
 *	Called when a user wants to get the network device usage
830
 *	statistics. Drivers must do one of the following:
831 832
 *	1. Define @ndo_get_stats64 to fill in a zero-initialised
 *	   rtnl_link_stats64 structure passed by the caller.
833
 *	2. Define @ndo_get_stats to update a net_device_stats structure
834 835 836 837 838
 *	   (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.
839
 *
840 841 842
 * 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.
843
 *
844
 * int (*ndo_vlan_rx_kill_vid)(struct net_device *dev, unsigned short vid);
845 846
 *	If device support VLAN filtering this function is called when a
 *	VLAN id is unregistered.
847 848
 *
 * void (*ndo_poll_controller)(struct net_device *dev);
849 850 851 852 853
 *
 *	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);
854
 * int (*ndo_set_vf_spoofchk)(struct net_device *dev, int vf, bool setting);
855 856
 * int (*ndo_get_vf_config)(struct net_device *dev,
 *			    int vf, struct ifla_vf_info *ivf);
857
 * int (*ndo_set_vf_link_state)(struct net_device *dev, int vf, int link_state);
858 859 860
 * 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);
861 862 863 864 865
 * 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.
866
 *
867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896
 *	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.
 *
897 898 899 900 901 902 903
 * 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.
 *
904 905 906 907 908 909
 * 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.
 *
910 911 912 913 914 915
 *	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.
916
 *
917
 *	Slave management functions (for bridge, bonding, etc).
918 919 920 921 922
 * 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.
923 924
 *
 *      Feature/offload setting functions.
925 926
 * netdev_features_t (*ndo_fix_features)(struct net_device *dev,
 *		netdev_features_t features);
927 928 929 930
 *	Adjusts the requested feature flags according to device-specific
 *	constraints, and returns the resulting flags. Must not modify
 *	the device state.
 *
931
 * int (*ndo_set_features)(struct net_device *dev, netdev_features_t features);
932 933 934 935
 *	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.
 *
936 937
 * int (*ndo_fdb_add)(struct ndmsg *ndm, struct nlattr *tb[],
 *		      struct net_device *dev,
938
 *		      const unsigned char *addr, u16 flags)
939
 *	Adds an FDB entry to dev for addr.
940 941
 * int (*ndo_fdb_del)(struct ndmsg *ndm, struct nlattr *tb[],
 *		      struct net_device *dev,
942
 *		      const unsigned char *addr)
943 944 945 946 947
 *	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
948 949 950
 *
 * int (*ndo_bridge_setlink)(struct net_device *dev, struct nlmsghdr *nlh)
 * int (*ndo_bridge_getlink)(struct sk_buff *skb, u32 pid, u32 seq,
951
 *			     struct net_device *dev, u32 filter_mask)
J
Jiri Pirko 已提交
952 953 954 955 956 957 958 959
 *
 * 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.
960 961 962 963 964 965
 *
 * 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.
966 967
 *
 * void (*ndo_add_vxlan_port)(struct  net_device *dev,
J
Joseph Gasparakis 已提交
968
 *			      sa_family_t sa_family, __be16 port);
969 970 971 972 973 974
 *	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
975
 *			      sa_family_t sa_family, __be16 port);
976 977 978
 *	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.
979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997
 *
 * 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.
998 999 1000 1001 1002 1003
 */
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);
1004
	netdev_tx_t		(*ndo_start_xmit) (struct sk_buff *skb,
1005 1006
						   struct net_device *dev);
	u16			(*ndo_select_queue)(struct net_device *dev,
1007 1008
						    struct sk_buff *skb,
						    void *accel_priv);
1009 1010 1011 1012 1013 1014 1015 1016 1017 1018
	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);
1019 1020 1021 1022
	int			(*ndo_change_mtu)(struct net_device *dev,
						  int new_mtu);
	int			(*ndo_neigh_setup)(struct net_device *dev,
						   struct neigh_parms *);
1023 1024
	void			(*ndo_tx_timeout) (struct net_device *dev);

1025 1026
	struct rtnl_link_stats64* (*ndo_get_stats64)(struct net_device *dev,
						     struct rtnl_link_stats64 *storage);
1027 1028
	struct net_device_stats* (*ndo_get_stats)(struct net_device *dev);

1029
	int			(*ndo_vlan_rx_add_vid)(struct net_device *dev,
1030
						       __be16 proto, u16 vid);
1031
	int			(*ndo_vlan_rx_kill_vid)(struct net_device *dev,
1032
						        __be16 proto, u16 vid);
1033 1034
#ifdef CONFIG_NET_POLL_CONTROLLER
	void                    (*ndo_poll_controller)(struct net_device *dev);
H
Herbert Xu 已提交
1035
	int			(*ndo_netpoll_setup)(struct net_device *dev,
1036 1037
						     struct netpoll_info *info,
						     gfp_t gfp);
1038
	void			(*ndo_netpoll_cleanup)(struct net_device *dev);
E
Eliezer Tamir 已提交
1039
#endif
1040
#ifdef CONFIG_NET_RX_BUSY_POLL
1041
	int			(*ndo_busy_poll)(struct napi_struct *dev);
1042
#endif
1043 1044 1045 1046 1047 1048
	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);
1049 1050
	int			(*ndo_set_vf_spoofchk)(struct net_device *dev,
						       int vf, bool setting);
1051 1052 1053
	int			(*ndo_get_vf_config)(struct net_device *dev,
						     int vf,
						     struct ifla_vf_info *ivf);
1054 1055
	int			(*ndo_set_vf_link_state)(struct net_device *dev,
							 int vf, int link_state);
1056 1057 1058 1059 1060
	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);
1061
	int			(*ndo_setup_tc)(struct net_device *dev, u8 tc);
1062
#if IS_ENABLED(CONFIG_FCOE)
1063 1064
	int			(*ndo_fcoe_enable)(struct net_device *dev);
	int			(*ndo_fcoe_disable)(struct net_device *dev);
1065 1066 1067 1068 1069 1070
	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);
1071 1072 1073 1074
	int			(*ndo_fcoe_ddp_target)(struct net_device *dev,
						       u16 xid,
						       struct scatterlist *sgl,
						       unsigned int sgc);
1075 1076
	int			(*ndo_fcoe_get_hbainfo)(struct net_device *dev,
							struct netdev_fcoe_hbainfo *hbainfo);
1077 1078
#endif

1079
#if IS_ENABLED(CONFIG_LIBFCOE)
1080 1081 1082 1083
#define NETDEV_FCOE_WWNN 0
#define NETDEV_FCOE_WWPN 1
	int			(*ndo_fcoe_get_wwn)(struct net_device *dev,
						    u64 *wwn, int type);
1084
#endif
1085

1086 1087 1088 1089 1090 1091
#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
1092 1093 1094 1095
	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);
1096 1097
	netdev_features_t	(*ndo_fix_features)(struct net_device *dev,
						    netdev_features_t features);
1098
	int			(*ndo_set_features)(struct net_device *dev,
1099
						    netdev_features_t features);
1100
	int			(*ndo_neigh_construct)(struct neighbour *n);
1101
	void			(*ndo_neigh_destroy)(struct neighbour *n);
1102 1103

	int			(*ndo_fdb_add)(struct ndmsg *ndm,
1104
					       struct nlattr *tb[],
1105
					       struct net_device *dev,
1106
					       const unsigned char *addr,
1107 1108
					       u16 flags);
	int			(*ndo_fdb_del)(struct ndmsg *ndm,
1109
					       struct nlattr *tb[],
1110
					       struct net_device *dev,
1111
					       const unsigned char *addr);
1112 1113 1114 1115
	int			(*ndo_fdb_dump)(struct sk_buff *skb,
						struct netlink_callback *cb,
						struct net_device *dev,
						int idx);
J
John Fastabend 已提交
1116 1117 1118 1119 1120

	int			(*ndo_bridge_setlink)(struct net_device *dev,
						      struct nlmsghdr *nlh);
	int			(*ndo_bridge_getlink)(struct sk_buff *skb,
						      u32 pid, u32 seq,
1121 1122
						      struct net_device *dev,
						      u32 filter_mask);
1123 1124
	int			(*ndo_bridge_dellink)(struct net_device *dev,
						      struct nlmsghdr *nlh);
J
Jiri Pirko 已提交
1125 1126
	int			(*ndo_change_carrier)(struct net_device *dev,
						      bool new_carrier);
1127 1128
	int			(*ndo_get_phys_port_id)(struct net_device *dev,
							struct netdev_phys_port_id *ppid);
1129 1130
	void			(*ndo_add_vxlan_port)(struct  net_device *dev,
						      sa_family_t sa_family,
J
Joseph Gasparakis 已提交
1131
						      __be16 port);
1132 1133
	void			(*ndo_del_vxlan_port)(struct  net_device *dev,
						      sa_family_t sa_family,
J
Joseph Gasparakis 已提交
1134
						      __be16 port);
1135 1136 1137 1138 1139 1140 1141 1142 1143

	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);
1144 1145
};

L
Linus Torvalds 已提交
1146 1147 1148 1149 1150 1151 1152 1153 1154 1155
/*
 *	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
1156
struct net_device {
L
Linus Torvalds 已提交
1157 1158 1159 1160

	/*
	 * 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
1161
	 * of the interface.
L
Linus Torvalds 已提交
1162 1163
	 */
	char			name[IFNAMSIZ];
M
Mark Gross 已提交
1164

1165
	/* device name hash chain, please keep it close to name[] */
1166
	struct hlist_node	name_hlist;
1167

1168 1169
	/* snmp alias */
	char 			*ifalias;
L
Linus Torvalds 已提交
1170 1171 1172 1173 1174 1175 1176 1177

	/*
	 *	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	*/
1178
	int			irq;		/* device IRQ number	*/
L
Linus Torvalds 已提交
1179 1180 1181 1182 1183 1184 1185 1186

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

	unsigned long		state;

1187
	struct list_head	dev_list;
1188
	struct list_head	napi_list;
1189
	struct list_head	unreg_list;
1190
	struct list_head	close_list;
1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202

	/* 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;
1203

L
Linus Torvalds 已提交
1204

1205
	/* currently active device features */
1206
	netdev_features_t	features;
1207
	/* user-changeable features */
1208
	netdev_features_t	hw_features;
1209
	/* user-requested features */
1210
	netdev_features_t	wanted_features;
1211
	/* mask of features inheritable by VLAN devices */
1212
	netdev_features_t	vlan_features;
1213 1214 1215 1216 1217 1218
	/* 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
1219 1220
	/* mask of fetures inheritable by MPLS */
	netdev_features_t	mpls_features;
1221

L
Linus Torvalds 已提交
1222 1223 1224 1225
	/* Interface index. Unique device identifier	*/
	int			ifindex;
	int			iflink;

R
Rusty Russell 已提交
1226
	struct net_device_stats	stats;
1227 1228 1229
	atomic_long_t		rx_dropped; /* dropped packets by core network
					     * Do not use this in drivers.
					     */
L
Linus Torvalds 已提交
1230

1231
#ifdef CONFIG_WIRELESS_EXT
L
Linus Torvalds 已提交
1232 1233 1234 1235 1236
	/* 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;
1237
#endif
1238 1239
	/* Management operations */
	const struct net_device_ops *netdev_ops;
1240
	const struct ethtool_ops *ethtool_ops;
1241
	const struct forwarding_accel_ops *fwd_ops;
L
Linus Torvalds 已提交
1242

1243 1244 1245
	/* Hardware header description */
	const struct header_ops *header_ops;

S
Stefan Rompf 已提交
1246
	unsigned int		flags;	/* interface flags (a la BSD)	*/
1247 1248
	unsigned int		priv_flags; /* Like 'flags' but invisible to userspace.
					     * See if.h for definitions. */
L
Linus Torvalds 已提交
1249 1250 1251
	unsigned short		gflags;
	unsigned short		padded;	/* How much padding added by alloc_netdev() */

S
Stefan Rompf 已提交
1252 1253 1254
	unsigned char		operstate; /* RFC2863 operstate */
	unsigned char		link_mode; /* mapping policy to operstate */

1255 1256 1257
	unsigned char		if_port;	/* Selectable AUI, TP,..*/
	unsigned char		dma;		/* DMA channel		*/

1258
	unsigned int		mtu;	/* interface MTU value		*/
L
Linus Torvalds 已提交
1259 1260 1261
	unsigned short		type;	/* interface hardware type	*/
	unsigned short		hard_header_len;	/* hardware hdr length	*/

1262 1263 1264 1265 1266 1267 1268
	/* 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
1269
	/* Interface address info. */
1270
	unsigned char		perm_addr[MAX_ADDR_LEN]; /* permanent hw address */
1271
	unsigned char		addr_assign_type; /* hw address assignment type */
L
Linus Torvalds 已提交
1272
	unsigned char		addr_len;	/* hardware address length	*/
1273
	unsigned short		neigh_priv_len;
1274 1275 1276 1277
	unsigned short          dev_id;		/* Used to differentiate devices
						 * that share the same link
						 * layer address
						 */
J
Jiri Pirko 已提交
1278
	spinlock_t		addr_list_lock;
1279 1280
	struct netdev_hw_addr_list	uc;	/* Unicast mac addresses */
	struct netdev_hw_addr_list	mc;	/* Multicast mac addresses */
1281 1282 1283 1284 1285 1286 1287
	struct netdev_hw_addr_list	dev_addrs; /* list of device
						    * hw addresses
						    */
#ifdef CONFIG_SYSFS
	struct kset		*queues_kset;
#endif

1288
	bool			uc_promisc;
1289 1290
	unsigned int		promiscuity;
	unsigned int		allmulti;
L
Linus Torvalds 已提交
1291 1292 1293


	/* Protocol specific pointers */
1294

1295
#if IS_ENABLED(CONFIG_VLAN_8021Q)
1296
	struct vlan_info __rcu	*vlan_info;	/* VLAN info */
1297
#endif
1298
#if IS_ENABLED(CONFIG_NET_DSA)
1299
	struct dsa_switch_tree	*dsa_ptr;	/* dsa specific data */
1300 1301 1302
#endif
#if IS_ENABLED(CONFIG_TIPC)
	struct tipc_bearer __rcu *tipc_ptr;	/* TIPC specific data */
1303
#endif
L
Linus Torvalds 已提交
1304
	void 			*atalk_ptr;	/* AppleTalk link 	*/
E
Eric Dumazet 已提交
1305
	struct in_device __rcu	*ip_ptr;	/* IPv4 specific data	*/
1306
	struct dn_dev __rcu     *dn_ptr;        /* DECnet specific data */
E
Eric Dumazet 已提交
1307
	struct inet6_dev __rcu	*ip6_ptr;       /* IPv6 specific data */
L
Linus Torvalds 已提交
1308
	void			*ax25_ptr;	/* AX.25 specific data */
1309 1310
	struct wireless_dev	*ieee80211_ptr;	/* IEEE 802.11 specific data,
						   assign before registering */
L
Linus Torvalds 已提交
1311

1312
/*
E
Eric Dumazet 已提交
1313
 * Cache lines mostly used on receive path (including eth_type_trans())
1314
 */
1315 1316 1317 1318 1319 1320 1321 1322
	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.
						 */

1323
	/* Interface address info used in eth_type_trans() */
1324 1325 1326 1327
	unsigned char		*dev_addr;	/* hw address, (before bcast
						   because most packets are
						   unicast) */

T
Tom Herbert 已提交
1328

1329
#ifdef CONFIG_SYSFS
T
Tom Herbert 已提交
1330 1331
	struct netdev_rx_queue	*_rx;

1332
	/* Number of RX queues allocated at register_netdev() time */
T
Tom Herbert 已提交
1333
	unsigned int		num_rx_queues;
1334 1335 1336

	/* Number of RX queues currently active in device */
	unsigned int		real_num_rx_queues;
1337

E
Eric Dumazet 已提交
1338
#endif
T
Tom Herbert 已提交
1339

1340 1341
	rx_handler_func_t __rcu	*rx_handler;
	void __rcu		*rx_handler_data;
1342

1343
	struct netdev_queue __rcu *ingress_queue;
1344 1345
	unsigned char		broadcast[MAX_ADDR_LEN];	/* hw bcast add	*/

E
Eric Dumazet 已提交
1346 1347 1348 1349

/*
 * Cache lines mostly used on transmit path
 */
1350
	struct netdev_queue	*_tx ____cacheline_aligned_in_smp;
1351 1352

	/* Number of TX queues allocated at alloc_netdev_mq() time  */
1353
	unsigned int		num_tx_queues;
1354 1355 1356 1357

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

1358 1359 1360
	/* root qdisc from userspace point of view */
	struct Qdisc		*qdisc;

L
Linus Torvalds 已提交
1361
	unsigned long		tx_queue_len;	/* Max frames per queue allowed */
1362
	spinlock_t		tx_global_lock;
E
Eric Dumazet 已提交
1363

T
Tom Herbert 已提交
1364
#ifdef CONFIG_XPS
E
Eric Dumazet 已提交
1365
	struct xps_dev_maps __rcu *xps_maps;
T
Tom Herbert 已提交
1366
#endif
1367 1368 1369 1370 1371 1372
#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
1373

1374
	/* These may be needed for future network-power-down code. */
1375 1376 1377 1378 1379

	/*
	 * trans_start here is expensive for high speed devices on SMP,
	 * please use netdev_queue->trans_start instead.
	 */
1380 1381 1382 1383 1384
	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
1385
	/* Number of references to this device */
E
Eric Dumazet 已提交
1386
	int __percpu		*pcpu_refcnt;
1387

L
Linus Torvalds 已提交
1388 1389 1390 1391 1392
	/* delayed register/unregister */
	struct list_head	todo_list;
	/* device index hash chain */
	struct hlist_node	index_hlist;

1393
	struct list_head	link_watch_list;
1394

L
Linus Torvalds 已提交
1395 1396
	/* register/unregister state machine */
	enum { NETREG_UNINITIALIZED=0,
1397
	       NETREG_REGISTERED,	/* completed register_netdevice */
L
Linus Torvalds 已提交
1398 1399 1400
	       NETREG_UNREGISTERING,	/* called unregister_netdevice */
	       NETREG_UNREGISTERED,	/* completed unregister todo */
	       NETREG_RELEASED,		/* called free_netdev */
1401
	       NETREG_DUMMY,		/* dummy device for NAPI poll */
1402 1403 1404
	} reg_state:8;

	bool dismantle; /* device is going do be freed */
1405 1406 1407 1408 1409

	enum {
		RTNL_LINK_INITIALIZED,
		RTNL_LINK_INITIALIZING,
	} rtnl_link_state:16;
L
Linus Torvalds 已提交
1410

1411 1412
	/* Called from unregister, can be used to call free_netdev */
	void (*destructor)(struct net_device *dev);
L
Linus Torvalds 已提交
1413 1414

#ifdef CONFIG_NETPOLL
1415
	struct netpoll_info __rcu	*npinfo;
L
Linus Torvalds 已提交
1416
#endif
1417

1418
#ifdef CONFIG_NET_NS
1419 1420
	/* Network namespace this network device is inside */
	struct net		*nd_net;
1421
#endif
1422

D
David S. Miller 已提交
1423
	/* mid-layer private */
E
Eric Dumazet 已提交
1424 1425 1426
	union {
		void				*ml_priv;
		struct pcpu_lstats __percpu	*lstats; /* loopback stats */
1427
		struct pcpu_sw_netstats __percpu	*tstats;
1428
		struct pcpu_dstats __percpu	*dstats; /* dummy stats */
E
Eric Dumazet 已提交
1429
		struct pcpu_vstats __percpu	*vstats; /* veth stats */
E
Eric Dumazet 已提交
1430
	};
1431
	/* GARP */
E
Eric Dumazet 已提交
1432
	struct garp_port __rcu	*garp_port;
1433 1434
	/* MRP */
	struct mrp_port __rcu	*mrp_port;
L
Linus Torvalds 已提交
1435 1436

	/* class/net/name entry */
1437
	struct device		dev;
1438 1439
	/* space for optional device, statistics, and wireless sysfs groups */
	const struct attribute_group *sysfs_groups[4];
1440 1441
	/* space for optional per-rx queue attributes */
	const struct attribute_group *sysfs_rx_queue_group;
P
Patrick McHardy 已提交
1442 1443 1444

	/* rtnetlink link ops */
	const struct rtnl_link_ops *rtnl_link_ops;
1445

1446 1447 1448
	/* for setting kernel sock attribute on TCP connection setup */
#define GSO_MAX_SIZE		65536
	unsigned int		gso_max_size;
1449 1450
#define GSO_MAX_SEGS		65535
	u16			gso_max_segs;
1451

J
Jeff Kirsher 已提交
1452
#ifdef CONFIG_DCB
1453
	/* Data Center Bridging netlink ops */
1454
	const struct dcbnl_rtnl_ops *dcbnl_ops;
1455
#endif
1456 1457 1458
	u8 num_tc;
	struct netdev_tc_txq tc_to_txq[TC_MAX_QUEUE];
	u8 prio_tc_map[TC_BITMASK + 1];
1459

1460
#if IS_ENABLED(CONFIG_FCOE)
1461 1462
	/* max exchange id for FCoE LRO by ddp */
	unsigned int		fcoe_ddp_xid;
1463
#endif
1464
#if IS_ENABLED(CONFIG_CGROUP_NET_PRIO)
1465
	struct netprio_map __rcu *priomap;
1466
#endif
1467 1468
	/* phy device may attach itself for hardware timestamping */
	struct phy_device *phydev;
1469

1470 1471
	struct lock_class_key *qdisc_tx_busylock;

1472 1473
	/* group the device belongs to */
	int group;
1474 1475

	struct pm_qos_request	pm_qos_req;
L
Linus Torvalds 已提交
1476
};
1477
#define to_net_dev(d) container_of(d, struct net_device, dev)
L
Linus Torvalds 已提交
1478 1479 1480

#define	NETDEV_ALIGN		32

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

1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550
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);
}

1551
struct netdev_queue *netdev_pick_tx(struct net_device *dev,
1552 1553
				    struct sk_buff *skb,
				    void *accel_priv);
1554
u16 __netdev_pick_tx(struct net_device *dev, struct sk_buff *skb);
1555

1556 1557 1558 1559 1560 1561
/*
 * Net namespace inlines
 */
static inline
struct net *dev_net(const struct net_device *dev)
{
E
Eric Dumazet 已提交
1562
	return read_pnet(&dev->nd_net);
1563 1564 1565
}

static inline
1566
void dev_net_set(struct net_device *dev, struct net *net)
1567 1568
{
#ifdef CONFIG_NET_NS
1569 1570
	release_net(dev->nd_net);
	dev->nd_net = hold_net(net);
1571 1572 1573
#endif
}

1574 1575 1576 1577 1578 1579 1580 1581 1582 1583
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;
}

1584 1585 1586 1587 1588 1589 1590 1591 1592 1593
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;
}

1594 1595 1596 1597 1598 1599
/**
 *	netdev_priv - access network device private data
 *	@dev: network device
 *
 * Get network device private data
 */
1600
static inline void *netdev_priv(const struct net_device *dev)
L
Linus Torvalds 已提交
1601
{
1602
	return (char *)dev + ALIGN(sizeof(struct net_device), NETDEV_ALIGN);
L
Linus Torvalds 已提交
1603 1604 1605 1606 1607
}

/* Set the sysfs physical device reference for the network logical device
 * if set prior to registration will cause a symlink during initialization.
 */
1608
#define SET_NETDEV_DEV(net, pdev)	((net)->dev.parent = (pdev))
L
Linus Torvalds 已提交
1609

1610
/* Set the sysfs device type for the network logical device to allow
1611
 * fine-grained identification of different network device types. For
1612 1613 1614 1615
 * example Ethernet, Wirelss LAN, Bluetooth, WiMAX etc.
 */
#define SET_NETDEV_DEVTYPE(net, devtype)	((net)->dev.type = (devtype))

E
Eric Dumazet 已提交
1616 1617 1618 1619 1620
/* Default NAPI poll() weight
 * Device drivers are strongly advised to not use bigger value
 */
#define NAPI_POLL_WEIGHT 64

1621 1622 1623 1624 1625 1626 1627 1628 1629 1630
/**
 *	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.
 */
1631 1632
void netif_napi_add(struct net_device *dev, struct napi_struct *napi,
		    int (*poll)(struct napi_struct *, int), int weight);
1633

1634 1635 1636 1637 1638 1639
/**
 *  netif_napi_del - remove a napi context
 *  @napi: napi context
 *
 *  netif_napi_del() removes a napi context from the network device napi list
 */
1640 1641 1642
void netif_napi_del(struct napi_struct *napi);

struct napi_gro_cb {
1643 1644 1645
	/* Virtual address of skb_shinfo(skb)->frags[0].page + offset. */
	void *frag0;

1646 1647 1648
	/* Length of frag0. */
	unsigned int frag0_len;

1649 1650 1651
	/* This indicates where we are processing relative to skb->data. */
	int data_offset;

1652
	/* This is non-zero if the packet cannot be merged with the new skb. */
1653 1654 1655 1656
	u16	flush;

	/* Save the IP ID here and check when we get to the transport layer */
	u16	flush_id;
1657 1658

	/* Number of segments aggregated. */
1659 1660 1661 1662
	u16	count;

	/* This is non-zero if the packet may be of the same flow. */
	u8	same_flow;
H
Herbert Xu 已提交
1663 1664

	/* Free the skb? */
1665
	u8	free;
1666 1667
#define NAPI_GRO_FREE		  1
#define NAPI_GRO_FREE_STOLEN_HEAD 2
1668 1669 1670

	/* jiffies when first packet was created/queued */
	unsigned long age;
1671 1672 1673

	/* Used in ipv6_gro_receive() */
	int	proto;
1674

1675 1676 1677
	/* used to support CHECKSUM_COMPLETE for tunneling protocols */
	__wsum	csum;

1678 1679
	/* used in skb_gro_receive() slow path */
	struct sk_buff *last;
1680 1681 1682
};

#define NAPI_GRO_CB(skb) ((struct napi_gro_cb *)(skb)->cb)
1683

L
Linus Torvalds 已提交
1684
struct packet_type {
D
David S. Miller 已提交
1685 1686 1687 1688 1689 1690
	__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 *);
1691 1692
	bool			(*id_match)(struct packet_type *ptype,
					    struct sock *sk);
L
Linus Torvalds 已提交
1693 1694 1695 1696
	void			*af_packet_priv;
	struct list_head	list;
};

1697
struct offload_callbacks {
1698
	struct sk_buff		*(*gso_segment)(struct sk_buff *skb,
1699
						netdev_features_t features);
1700
	int			(*gso_send_check)(struct sk_buff *skb);
1701 1702
	struct sk_buff		**(*gro_receive)(struct sk_buff **head,
					       struct sk_buff *skb);
1703
	int			(*gro_complete)(struct sk_buff *skb, int nhoff);
1704 1705 1706 1707 1708 1709
};

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
1710 1711
};

1712 1713 1714 1715 1716 1717 1718 1719 1720
/* 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
1721 1722
#include <linux/notifier.h>

1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735
/* 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
1736
#define NETDEV_CHANGEMTU	0x0007 /* notify after mtu change happened */
1737 1738 1739 1740 1741 1742 1743 1744 1745
#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
1746
#define NETDEV_UNREGISTER_FINAL 0x0011
1747 1748 1749
#define NETDEV_RELEASE		0x0012
#define NETDEV_NOTIFY_PEERS	0x0013
#define NETDEV_JOIN		0x0014
1750
#define NETDEV_CHANGEUPPER	0x0015
1751
#define NETDEV_RESEND_IGMP	0x0016
1752
#define NETDEV_PRECHANGEMTU	0x0017 /* notify before mtu change happened */
1753

1754 1755
int register_netdevice_notifier(struct notifier_block *nb);
int unregister_netdevice_notifier(struct notifier_block *nb);
1756 1757 1758 1759 1760

struct netdev_notifier_info {
	struct net_device *dev;
};

1761 1762 1763 1764 1765
struct netdev_notifier_change_info {
	struct netdev_notifier_info info; /* must be first */
	unsigned int flags_changed;
};

1766 1767 1768 1769 1770 1771
static inline void netdev_notifier_info_init(struct netdev_notifier_info *info,
					     struct net_device *dev)
{
	info->dev = dev;
}

1772 1773 1774 1775 1776 1777
static inline struct net_device *
netdev_notifier_info_to_dev(const struct netdev_notifier_info *info)
{
	return info->dev;
}

1778
int call_netdevice_notifiers(unsigned long val, struct net_device *dev);
1779 1780


L
Linus Torvalds 已提交
1781 1782
extern rwlock_t				dev_base_lock;		/* Device list lock */

1783 1784
#define for_each_netdev(net, d)		\
		list_for_each_entry(d, &(net)->dev_base_head, dev_list)
1785 1786
#define for_each_netdev_reverse(net, d)	\
		list_for_each_entry_reverse(d, &(net)->dev_base_head, dev_list)
1787 1788
#define for_each_netdev_rcu(net, d)		\
		list_for_each_entry_rcu(d, &(net)->dev_base_head, dev_list)
1789 1790 1791 1792
#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)
1793 1794
#define for_each_netdev_continue_rcu(net, d)		\
	list_for_each_entry_continue_rcu(d, &(net)->dev_base_head, dev_list)
1795 1796 1797
#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)
1798
#define net_device_entry(lh)	list_entry(lh, struct net_device, dev_list)
1799

1800 1801 1802 1803 1804
static inline struct net_device *next_net_device(struct net_device *dev)
{
	struct list_head *lh;
	struct net *net;

1805
	net = dev_net(dev);
1806 1807 1808 1809
	lh = dev->dev_list.next;
	return lh == &net->dev_base_head ? NULL : net_device_entry(lh);
}

1810 1811 1812 1813 1814 1815
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
1816
	lh = rcu_dereference(list_next_rcu(&dev->dev_list));
1817 1818 1819
	return lh == &net->dev_base_head ? NULL : net_device_entry(lh);
}

1820 1821 1822 1823 1824
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);
}
1825

E
Eric Dumazet 已提交
1826 1827 1828 1829 1830 1831 1832
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);
}

1833 1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855
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);
1856
int dev_queue_xmit_accel(struct sk_buff *skb, void *accel_priv);
1857 1858 1859
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);
1860 1861 1862 1863 1864
static inline void unregister_netdevice(struct net_device *dev)
{
	unregister_netdevice_queue(dev, NULL);
}

1865 1866
int netdev_refcnt_read(const struct net_device *dev);
void free_netdev(struct net_device *dev);
1867
void netdev_freemem(struct net_device *dev);
1868 1869
void synchronize_net(void);
int init_dummy_netdev(struct net_device *dev);
1870

1871 1872 1873 1874 1875
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
1876
#ifdef CONFIG_NETPOLL_TRAP
1877
int netpoll_trap(void);
L
Linus Torvalds 已提交
1878
#endif
1879
int skb_gro_receive(struct sk_buff **head, struct sk_buff *skb);
1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895

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;
}

1896 1897
static inline void *skb_gro_header_fast(struct sk_buff *skb,
					unsigned int offset)
1898
{
1899 1900
	return NAPI_GRO_CB(skb)->frag0 + offset;
}
1901

1902 1903 1904 1905
static inline int skb_gro_header_hard(struct sk_buff *skb, unsigned int hlen)
{
	return NAPI_GRO_CB(skb)->frag0_len < hlen;
}
1906

1907 1908 1909
static inline void *skb_gro_header_slow(struct sk_buff *skb, unsigned int hlen,
					unsigned int offset)
{
1910 1911 1912
	if (!pskb_may_pull(skb, hlen))
		return NULL;

1913 1914
	NAPI_GRO_CB(skb)->frag0 = NULL;
	NAPI_GRO_CB(skb)->frag0_len = 0;
1915
	return skb->data + offset;
1916
}
L
Linus Torvalds 已提交
1917

1918 1919
static inline void *skb_gro_mac_header(struct sk_buff *skb)
{
1920
	return NAPI_GRO_CB(skb)->frag0 ?: skb_mac_header(skb);
1921 1922
}

H
Herbert Xu 已提交
1923 1924
static inline void *skb_gro_network_header(struct sk_buff *skb)
{
1925 1926
	return (NAPI_GRO_CB(skb)->frag0 ?: skb->data) +
	       skb_network_offset(skb);
H
Herbert Xu 已提交
1927 1928
}

1929 1930 1931 1932 1933 1934 1935 1936
static inline void skb_gro_postpull_rcsum(struct sk_buff *skb,
					const void *start, unsigned int len)
{
	if (skb->ip_summed == CHECKSUM_COMPLETE)
		NAPI_GRO_CB(skb)->csum = csum_sub(NAPI_GRO_CB(skb)->csum,
						  csum_partial(start, len, 0));
}

1937 1938
static inline int dev_hard_header(struct sk_buff *skb, struct net_device *dev,
				  unsigned short type,
1939
				  const void *daddr, const void *saddr,
1940
				  unsigned int len)
1941
{
1942
	if (!dev->header_ops || !dev->header_ops->create)
1943
		return 0;
1944 1945

	return dev->header_ops->create(skb, dev, type, daddr, saddr, len);
1946 1947
}

S
Stephen Hemminger 已提交
1948 1949 1950 1951 1952
static inline int dev_parse_header(const struct sk_buff *skb,
				   unsigned char *haddr)
{
	const struct net_device *dev = skb->dev;

1953
	if (!dev->header_ops || !dev->header_ops->parse)
S
Stephen Hemminger 已提交
1954
		return 0;
1955
	return dev->header_ops->parse(skb, haddr);
S
Stephen Hemminger 已提交
1956 1957
}

1958 1959 1960 1961 1962 1963 1964 1965 1966
static inline int dev_rebuild_header(struct sk_buff *skb)
{
	const struct net_device *dev = skb->dev;

	if (!dev->header_ops || !dev->header_ops->rebuild)
		return 0;
	return dev->header_ops->rebuild(skb);
}

L
Linus Torvalds 已提交
1967
typedef int gifconf_func_t(struct net_device * dev, char __user * bufptr, int len);
1968
int register_gifconf(unsigned int family, gifconf_func_t *gifconf);
L
Linus Torvalds 已提交
1969 1970 1971 1972 1973
static inline int unregister_gifconf(unsigned int family)
{
	return register_gifconf(family, NULL);
}

1974
#ifdef CONFIG_NET_FLOW_LIMIT
1975
#define FLOW_LIMIT_HISTORY	(1 << 7)  /* must be ^2 and !overflow buckets */
1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986
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
1987
/*
E
Eric Dumazet 已提交
1988
 * Incoming packets are placed on per-cpu queues
L
Linus Torvalds 已提交
1989
 */
E
Eric Dumazet 已提交
1990
struct softnet_data {
1991
	struct Qdisc		*output_queue;
1992
	struct Qdisc		**output_queue_tailp;
L
Linus Torvalds 已提交
1993 1994
	struct list_head	poll_list;
	struct sk_buff		*completion_queue;
1995
	struct sk_buff_head	process_queue;
L
Linus Torvalds 已提交
1996

C
Changli Gao 已提交
1997
	/* stats */
1998 1999 2000 2001
	unsigned int		processed;
	unsigned int		time_squeeze;
	unsigned int		cpu_collision;
	unsigned int		received_rps;
C
Changli Gao 已提交
2002

2003
#ifdef CONFIG_RPS
E
Eric Dumazet 已提交
2004 2005 2006
	struct softnet_data	*rps_ipi_list;

	/* Elements below can be accessed between CPUs for RPS */
T
Tom Herbert 已提交
2007
	struct call_single_data	csd ____cacheline_aligned_in_smp;
E
Eric Dumazet 已提交
2008 2009
	struct softnet_data	*rps_ipi_next;
	unsigned int		cpu;
T
Tom Herbert 已提交
2010
	unsigned int		input_queue_head;
2011
	unsigned int		input_queue_tail;
2012
#endif
2013
	unsigned int		dropped;
T
Tom Herbert 已提交
2014
	struct sk_buff_head	input_pkt_queue;
2015
	struct napi_struct	backlog;
2016 2017

#ifdef CONFIG_NET_FLOW_LIMIT
2018
	struct sd_flow_limit __rcu *flow_limit;
2019
#endif
L
Linus Torvalds 已提交
2020 2021
};

2022
static inline void input_queue_head_incr(struct softnet_data *sd)
T
Tom Herbert 已提交
2023 2024
{
#ifdef CONFIG_RPS
2025 2026 2027 2028 2029 2030 2031 2032 2033
	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
2034 2035 2036
#endif
}

T
Tom Herbert 已提交
2037
DECLARE_PER_CPU_ALIGNED(struct softnet_data, softnet_data);
L
Linus Torvalds 已提交
2038

2039
void __netif_schedule(struct Qdisc *q);
L
Linus Torvalds 已提交
2040

2041
static inline void netif_schedule_queue(struct netdev_queue *txq)
L
Linus Torvalds 已提交
2042
{
2043
	if (!(txq->state & QUEUE_STATE_ANY_XOFF))
2044
		__netif_schedule(txq->qdisc);
2045 2046
}

2047 2048 2049 2050 2051 2052 2053 2054
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));
}

2055 2056
static inline void netif_tx_start_queue(struct netdev_queue *dev_queue)
{
2057
	clear_bit(__QUEUE_STATE_DRV_XOFF, &dev_queue->state);
2058 2059
}

2060 2061 2062 2063 2064 2065
/**
 *	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
2066 2067
static inline void netif_start_queue(struct net_device *dev)
{
2068
	netif_tx_start_queue(netdev_get_tx_queue(dev, 0));
L
Linus Torvalds 已提交
2069 2070
}

2071 2072 2073 2074 2075 2076 2077 2078 2079 2080
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);
	}
}

2081
static inline void netif_tx_wake_queue(struct netdev_queue *dev_queue)
L
Linus Torvalds 已提交
2082 2083
{
#ifdef CONFIG_NETPOLL_TRAP
2084
	if (netpoll_trap()) {
2085
		netif_tx_start_queue(dev_queue);
L
Linus Torvalds 已提交
2086
		return;
2087
	}
L
Linus Torvalds 已提交
2088
#endif
2089
	if (test_and_clear_bit(__QUEUE_STATE_DRV_XOFF, &dev_queue->state))
2090
		__netif_schedule(dev_queue->qdisc);
2091 2092
}

2093 2094 2095 2096 2097 2098 2099
/**
 *	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.
 */
2100 2101
static inline void netif_wake_queue(struct net_device *dev)
{
2102
	netif_tx_wake_queue(netdev_get_tx_queue(dev, 0));
L
Linus Torvalds 已提交
2103 2104
}

2105 2106 2107 2108 2109 2110 2111 2112 2113 2114
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);
	}
}

2115 2116
static inline void netif_tx_stop_queue(struct netdev_queue *dev_queue)
{
2117
	if (WARN_ON(!dev_queue)) {
2118
		pr_info("netif_stop_queue() cannot be called before register_netdev()\n");
2119 2120
		return;
	}
2121
	set_bit(__QUEUE_STATE_DRV_XOFF, &dev_queue->state);
2122 2123
}

2124 2125 2126 2127 2128 2129 2130
/**
 *	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
2131 2132
static inline void netif_stop_queue(struct net_device *dev)
{
2133
	netif_tx_stop_queue(netdev_get_tx_queue(dev, 0));
L
Linus Torvalds 已提交
2134 2135
}

2136 2137 2138 2139 2140 2141 2142 2143 2144 2145
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);
	}
}

2146
static inline bool netif_tx_queue_stopped(const struct netdev_queue *dev_queue)
2147
{
2148
	return test_bit(__QUEUE_STATE_DRV_XOFF, &dev_queue->state);
2149 2150
}

2151 2152 2153 2154 2155 2156
/**
 *	netif_queue_stopped - test if transmit queue is flowblocked
 *	@dev: network device
 *
 *	Test if transmit queue on device is currently unable to send.
 */
2157
static inline bool netif_queue_stopped(const struct net_device *dev)
L
Linus Torvalds 已提交
2158
{
2159
	return netif_tx_queue_stopped(netdev_get_tx_queue(dev, 0));
L
Linus Torvalds 已提交
2160 2161
}

2162
static inline bool netif_xmit_stopped(const struct netdev_queue *dev_queue)
2163
{
2164 2165 2166
	return dev_queue->state & QUEUE_STATE_ANY_XOFF;
}

2167
static inline bool netif_xmit_frozen_or_stopped(const struct netdev_queue *dev_queue)
2168 2169 2170 2171
{
	return dev_queue->state & QUEUE_STATE_ANY_XOFF_OR_FROZEN;
}

2172 2173 2174
static inline void netdev_tx_sent_queue(struct netdev_queue *dev_queue,
					unsigned int bytes)
{
T
Tom Herbert 已提交
2175 2176
#ifdef CONFIG_BQL
	dql_queued(&dev_queue->dql, bytes);
2177 2178 2179 2180 2181 2182 2183 2184 2185 2186 2187 2188 2189 2190 2191 2192

	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
2193
#endif
2194 2195
}

2196 2197 2198 2199 2200 2201 2202 2203 2204
/**
 * 	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
 */
2205 2206 2207 2208 2209 2210
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,
2211
					     unsigned int pkts, unsigned int bytes)
2212
{
T
Tom Herbert 已提交
2213
#ifdef CONFIG_BQL
2214 2215 2216 2217 2218 2219 2220 2221 2222 2223 2224 2225 2226 2227 2228 2229 2230
	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
2231
#endif
2232 2233
}

2234 2235 2236 2237 2238 2239 2240 2241 2242 2243
/**
 * 	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()
 */
2244
static inline void netdev_completed_queue(struct net_device *dev,
2245
					  unsigned int pkts, unsigned int bytes)
2246 2247 2248 2249 2250 2251
{
	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
2252
#ifdef CONFIG_BQL
2253
	clear_bit(__QUEUE_STATE_STACK_XOFF, &q->state);
T
Tom Herbert 已提交
2254 2255
	dql_reset(&q->dql);
#endif
2256 2257
}

2258 2259 2260 2261 2262 2263 2264
/**
 * 	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
 */
2265 2266 2267
static inline void netdev_reset_queue(struct net_device *dev_queue)
{
	netdev_tx_reset_queue(netdev_get_tx_queue(dev_queue, 0));
2268 2269
}

2270 2271 2272 2273 2274 2275
/**
 *	netif_running - test if up
 *	@dev: network device
 *
 *	Test if the device has been brought up.
 */
2276
static inline bool netif_running(const struct net_device *dev)
L
Linus Torvalds 已提交
2277 2278 2279 2280
{
	return test_bit(__LINK_STATE_START, &dev->state);
}

2281 2282 2283 2284 2285 2286
/*
 * 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.
 */
2287 2288 2289 2290 2291 2292 2293 2294

/**
 *	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.
 */
2295 2296
static inline void netif_start_subqueue(struct net_device *dev, u16 queue_index)
{
2297
	struct netdev_queue *txq = netdev_get_tx_queue(dev, queue_index);
2298 2299

	netif_tx_start_queue(txq);
2300 2301
}

2302 2303 2304 2305 2306 2307 2308
/**
 *	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.
 */
2309 2310
static inline void netif_stop_subqueue(struct net_device *dev, u16 queue_index)
{
2311
	struct netdev_queue *txq = netdev_get_tx_queue(dev, queue_index);
2312 2313 2314 2315
#ifdef CONFIG_NETPOLL_TRAP
	if (netpoll_trap())
		return;
#endif
2316
	netif_tx_stop_queue(txq);
2317 2318
}

2319 2320 2321 2322 2323 2324 2325
/**
 *	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.
 */
2326 2327
static inline bool __netif_subqueue_stopped(const struct net_device *dev,
					    u16 queue_index)
2328
{
2329
	struct netdev_queue *txq = netdev_get_tx_queue(dev, queue_index);
2330 2331

	return netif_tx_queue_stopped(txq);
2332 2333
}

2334 2335
static inline bool netif_subqueue_stopped(const struct net_device *dev,
					  struct sk_buff *skb)
2336 2337 2338
{
	return __netif_subqueue_stopped(dev, skb_get_queue_mapping(skb));
}
2339 2340 2341 2342 2343 2344 2345 2346

/**
 *	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.
 */
2347 2348
static inline void netif_wake_subqueue(struct net_device *dev, u16 queue_index)
{
2349
	struct netdev_queue *txq = netdev_get_tx_queue(dev, queue_index);
2350 2351 2352 2353
#ifdef CONFIG_NETPOLL_TRAP
	if (netpoll_trap())
		return;
#endif
2354
	if (test_and_clear_bit(__QUEUE_STATE_DRV_XOFF, &txq->state))
2355
		__netif_schedule(txq->qdisc);
2356 2357
}

2358
#ifdef CONFIG_XPS
2359
int netif_set_xps_queue(struct net_device *dev, const struct cpumask *mask,
2360
			u16 index);
2361 2362
#else
static inline int netif_set_xps_queue(struct net_device *dev,
2363
				      const struct cpumask *mask,
2364 2365 2366 2367 2368 2369
				      u16 index)
{
	return 0;
}
#endif

2370 2371 2372 2373 2374 2375 2376 2377 2378 2379
/*
 * 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);
}

2380 2381 2382 2383 2384 2385
/**
 *	netif_is_multiqueue - test if device has multiple transmit queues
 *	@dev: network device
 *
 * Check if device has multiple transmit queues
 */
2386
static inline bool netif_is_multiqueue(const struct net_device *dev)
2387
{
E
Eric Dumazet 已提交
2388
	return dev->num_tx_queues > 1;
2389
}
L
Linus Torvalds 已提交
2390

2391
int netif_set_real_num_tx_queues(struct net_device *dev, unsigned int txq);
2392

2393
#ifdef CONFIG_SYSFS
2394
int netif_set_real_num_rx_queues(struct net_device *dev, unsigned int rxq);
2395 2396 2397 2398 2399 2400 2401 2402
#else
static inline int netif_set_real_num_rx_queues(struct net_device *dev,
						unsigned int rxq)
{
	return 0;
}
#endif

2403 2404 2405
static inline int netif_copy_real_num_queues(struct net_device *to_dev,
					     const struct net_device *from_dev)
{
2406 2407 2408 2409 2410 2411
	int err;

	err = netif_set_real_num_tx_queues(to_dev,
					   from_dev->real_num_tx_queues);
	if (err)
		return err;
2412
#ifdef CONFIG_SYSFS
2413 2414 2415 2416 2417 2418 2419
	return netif_set_real_num_rx_queues(to_dev,
					    from_dev->real_num_rx_queues);
#else
	return 0;
#endif
}

2420 2421 2422 2423 2424 2425 2426 2427 2428 2429 2430 2431
#ifdef CONFIG_SYSFS
static inline unsigned int get_netdev_rx_queue_index(
		struct netdev_rx_queue *queue)
{
	struct net_device *dev = queue->dev;
	int index = queue - dev->_rx;

	BUG_ON(index >= dev->num_rx_queues);
	return index;
}
#endif

2432
#define DEFAULT_MAX_NUM_RSS_QUEUES	(8)
2433
int netif_get_num_default_rss_queues(void);
2434

2435 2436 2437 2438 2439 2440 2441
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
2442

2443 2444 2445 2446 2447 2448 2449 2450 2451 2452 2453 2454 2455 2456 2457 2458 2459 2460
/*
 * 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
2461
 */
2462 2463 2464 2465 2466 2467 2468 2469 2470 2471 2472 2473 2474 2475 2476 2477 2478 2479 2480
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
2481

2482 2483 2484 2485 2486 2487 2488
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);
2489 2490
struct packet_offload *gro_find_receive_by_type(__be16 type);
struct packet_offload *gro_find_complete_by_type(__be16 type);
2491 2492 2493 2494 2495 2496 2497

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

2498 2499 2500 2501 2502 2503 2504 2505 2506 2507 2508
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);
2509 2510
void __dev_notify_flags(struct net_device *, unsigned int old_flags,
			unsigned int gchanges);
2511 2512 2513 2514 2515 2516 2517 2518 2519 2520
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,
2521
			struct netdev_queue *txq);
2522
int dev_forward_skb(struct net_device *dev, struct sk_buff *skb);
L
Linus Torvalds 已提交
2523

2524
extern int		netdev_budget;
L
Linus Torvalds 已提交
2525 2526

/* Called by rtnetlink.c:rtnl_unlock() */
2527
void netdev_run_todo(void);
L
Linus Torvalds 已提交
2528

2529 2530 2531 2532
/**
 *	dev_put - release reference to device
 *	@dev: network device
 *
2533
 * Release reference to device to allow it to be freed.
2534
 */
L
Linus Torvalds 已提交
2535 2536
static inline void dev_put(struct net_device *dev)
{
2537
	this_cpu_dec(*dev->pcpu_refcnt);
L
Linus Torvalds 已提交
2538 2539
}

2540 2541 2542 2543
/**
 *	dev_hold - get reference to device
 *	@dev: network device
 *
2544
 * Hold reference to device to keep it from being freed.
2545
 */
2546 2547
static inline void dev_hold(struct net_device *dev)
{
2548
	this_cpu_inc(*dev->pcpu_refcnt);
2549
}
L
Linus Torvalds 已提交
2550 2551 2552 2553

/* 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
2554 2555 2556 2557
 *
 * 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
2558 2559
 */

2560 2561 2562
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
2563

2564 2565 2566 2567 2568 2569
/**
 *	netif_carrier_ok - test if carrier present
 *	@dev: network device
 *
 * Check if carrier is present on device
 */
2570
static inline bool netif_carrier_ok(const struct net_device *dev)
L
Linus Torvalds 已提交
2571 2572 2573 2574
{
	return !test_bit(__LINK_STATE_NOCARRIER, &dev->state);
}

2575
unsigned long dev_trans_start(struct net_device *dev);
2576

2577
void __netdev_watchdog_up(struct net_device *dev);
L
Linus Torvalds 已提交
2578

2579
void netif_carrier_on(struct net_device *dev);
L
Linus Torvalds 已提交
2580

2581
void netif_carrier_off(struct net_device *dev);
L
Linus Torvalds 已提交
2582

2583 2584 2585 2586 2587 2588 2589 2590 2591 2592 2593 2594 2595
/**
 *	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
2596 2597 2598 2599 2600 2601
static inline void netif_dormant_on(struct net_device *dev)
{
	if (!test_and_set_bit(__LINK_STATE_DORMANT, &dev->state))
		linkwatch_fire_event(dev);
}

2602 2603 2604 2605 2606 2607
/**
 *	netif_dormant_off - set device as not dormant.
 *	@dev: network device
 *
 * Device is not in dormant state.
 */
S
Stefan Rompf 已提交
2608 2609 2610 2611 2612 2613
static inline void netif_dormant_off(struct net_device *dev)
{
	if (test_and_clear_bit(__LINK_STATE_DORMANT, &dev->state))
		linkwatch_fire_event(dev);
}

2614 2615 2616 2617 2618 2619
/**
 *	netif_dormant - test if carrier present
 *	@dev: network device
 *
 * Check if carrier is present on device
 */
2620
static inline bool netif_dormant(const struct net_device *dev)
S
Stefan Rompf 已提交
2621 2622 2623 2624 2625
{
	return test_bit(__LINK_STATE_DORMANT, &dev->state);
}


2626 2627 2628 2629 2630 2631
/**
 *	netif_oper_up - test if device is operational
 *	@dev: network device
 *
 * Check if carrier is operational
 */
2632
static inline bool netif_oper_up(const struct net_device *dev)
E
Eric Dumazet 已提交
2633
{
S
Stefan Rompf 已提交
2634 2635 2636 2637
	return (dev->operstate == IF_OPER_UP ||
		dev->operstate == IF_OPER_UNKNOWN /* backward compat */);
}

2638 2639 2640 2641 2642 2643
/**
 *	netif_device_present - is device available or removed
 *	@dev: network device
 *
 * Check if device has not been removed from system.
 */
2644
static inline bool netif_device_present(struct net_device *dev)
L
Linus Torvalds 已提交
2645 2646 2647 2648
{
	return test_bit(__LINK_STATE_PRESENT, &dev->state);
}

2649
void netif_device_detach(struct net_device *dev);
L
Linus Torvalds 已提交
2650

2651
void netif_device_attach(struct net_device *dev);
L
Linus Torvalds 已提交
2652 2653 2654 2655 2656 2657 2658 2659 2660 2661 2662 2663 2664 2665 2666 2667 2668 2669 2670 2671 2672 2673 2674 2675 2676 2677 2678 2679 2680 2681 2682 2683 2684 2685 2686 2687 2688 2689 2690 2691 2692 2693 2694 2695 2696 2697 2698 2699 2700 2701

/*
 * 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;
}

2702
static inline void __netif_tx_lock(struct netdev_queue *txq, int cpu)
H
Herbert Xu 已提交
2703
{
2704 2705
	spin_lock(&txq->_xmit_lock);
	txq->xmit_lock_owner = cpu;
2706 2707
}

2708 2709 2710 2711 2712 2713
static inline void __netif_tx_lock_bh(struct netdev_queue *txq)
{
	spin_lock_bh(&txq->_xmit_lock);
	txq->xmit_lock_owner = smp_processor_id();
}

2714
static inline bool __netif_tx_trylock(struct netdev_queue *txq)
2715
{
2716
	bool ok = spin_trylock(&txq->_xmit_lock);
2717 2718 2719 2720 2721 2722 2723 2724 2725 2726 2727 2728 2729 2730 2731 2732 2733
	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
2734 2735 2736 2737 2738 2739
static inline void txq_trans_update(struct netdev_queue *txq)
{
	if (txq->xmit_lock_owner != -1)
		txq->trans_start = jiffies;
}

2740 2741 2742 2743 2744 2745
/**
 *	netif_tx_lock - grab network device transmit lock
 *	@dev: network device
 *
 * Get network device transmit lock
 */
2746 2747
static inline void netif_tx_lock(struct net_device *dev)
{
2748
	unsigned int i;
2749
	int cpu;
2750

2751 2752
	spin_lock(&dev->tx_global_lock);
	cpu = smp_processor_id();
2753 2754
	for (i = 0; i < dev->num_tx_queues; i++) {
		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
2755 2756 2757 2758 2759 2760 2761

		/* 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.
		 */
2762
		__netif_tx_lock(txq, cpu);
2763 2764
		set_bit(__QUEUE_STATE_FROZEN, &txq->state);
		__netif_tx_unlock(txq);
2765
	}
H
Herbert Xu 已提交
2766 2767 2768 2769
}

static inline void netif_tx_lock_bh(struct net_device *dev)
{
2770 2771
	local_bh_disable();
	netif_tx_lock(dev);
H
Herbert Xu 已提交
2772 2773 2774 2775
}

static inline void netif_tx_unlock(struct net_device *dev)
{
2776 2777 2778 2779
	unsigned int i;

	for (i = 0; i < dev->num_tx_queues; i++) {
		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
2780

2781 2782 2783 2784 2785
		/* 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);
2786
		netif_schedule_queue(txq);
2787 2788
	}
	spin_unlock(&dev->tx_global_lock);
H
Herbert Xu 已提交
2789 2790 2791 2792
}

static inline void netif_tx_unlock_bh(struct net_device *dev)
{
2793 2794
	netif_tx_unlock(dev);
	local_bh_enable();
H
Herbert Xu 已提交
2795 2796
}

2797
#define HARD_TX_LOCK(dev, txq, cpu) {			\
2798
	if ((dev->features & NETIF_F_LLTX) == 0) {	\
2799
		__netif_tx_lock(txq, cpu);		\
2800 2801 2802
	}						\
}

2803
#define HARD_TX_UNLOCK(dev, txq) {			\
2804
	if ((dev->features & NETIF_F_LLTX) == 0) {	\
2805
		__netif_tx_unlock(txq);			\
2806 2807 2808
	}						\
}

L
Linus Torvalds 已提交
2809 2810
static inline void netif_tx_disable(struct net_device *dev)
{
2811
	unsigned int i;
2812
	int cpu;
2813

2814 2815
	local_bh_disable();
	cpu = smp_processor_id();
2816 2817
	for (i = 0; i < dev->num_tx_queues; i++) {
		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
2818 2819

		__netif_tx_lock(txq, cpu);
2820
		netif_tx_stop_queue(txq);
2821
		__netif_tx_unlock(txq);
2822
	}
2823
	local_bh_enable();
L
Linus Torvalds 已提交
2824 2825
}

2826 2827 2828 2829 2830
static inline void netif_addr_lock(struct net_device *dev)
{
	spin_lock(&dev->addr_list_lock);
}

2831 2832 2833 2834 2835
static inline void netif_addr_lock_nested(struct net_device *dev)
{
	spin_lock_nested(&dev->addr_list_lock, SINGLE_DEPTH_NESTING);
}

2836 2837 2838 2839 2840 2841 2842 2843 2844 2845 2846 2847 2848 2849 2850
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);
}

2851
/*
2852
 * dev_addrs walker. Should be used only for read access. Call with
2853 2854 2855
 * rcu_read_lock held.
 */
#define for_each_dev_addr(dev, ha) \
2856
		list_for_each_entry_rcu(ha, &dev->dev_addrs.list, list)
2857

L
Linus Torvalds 已提交
2858 2859
/* These functions live elsewhere (drivers/net/net_init.c, but related) */

2860
void ether_setup(struct net_device *dev);
L
Linus Torvalds 已提交
2861 2862

/* Support for loadable net-drivers */
2863 2864 2865
struct net_device *alloc_netdev_mqs(int sizeof_priv, const char *name,
				    void (*setup)(struct net_device *),
				    unsigned int txqs, unsigned int rxqs);
2866
#define alloc_netdev(sizeof_priv, name, setup) \
T
Tom Herbert 已提交
2867 2868 2869 2870 2871
	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)

2872 2873
int register_netdev(struct net_device *dev);
void unregister_netdev(struct net_device *dev);
2874

2875
/* General hardware address lists handling functions */
2876 2877 2878 2879 2880
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);
2881

2882
/* Functions used for device addresses handling */
2883 2884 2885 2886 2887 2888
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);
2889 2890

/* Functions used for unicast addresses handling */
2891 2892 2893 2894 2895 2896 2897 2898
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);
2899

2900
/* Functions used for multicast addresses handling */
2901 2902 2903 2904 2905 2906 2907 2908 2909 2910
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);
2911

2912
/* Functions used for secondary unicast and multicast support */
2913 2914 2915 2916 2917 2918 2919
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
2920
/* Load a device via the kmod */
2921 2922 2923 2924 2925
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);
2926

L
Linus Torvalds 已提交
2927
extern int		netdev_max_backlog;
E
Eric Dumazet 已提交
2928
extern int		netdev_tstamp_prequeue;
L
Linus Torvalds 已提交
2929
extern int		weight_p;
2930
extern int		bpf_jit_enable;
J
Jiri Pirko 已提交
2931

2932 2933 2934
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);
2935 2936

/* iterate through upper list, must be called under RCU read lock */
2937 2938 2939 2940 2941
#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)))
2942

2943 2944 2945 2946
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);
2947 2948 2949 2950 2951 2952 2953 2954 2955 2956 2957 2958 2959

#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)))

2960
void *netdev_adjacent_get_private(struct list_head *adj_list);
2961
void *netdev_lower_get_first_private_rcu(struct net_device *dev);
2962 2963 2964 2965
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
2966
				 struct net_device *upper_dev);
2967 2968 2969 2970 2971
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);
2972
void netdev_adjacent_rename_links(struct net_device *dev, char *oldname);
2973 2974 2975 2976 2977 2978 2979
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);
2980 2981 2982 2983 2984 2985

static inline
struct sk_buff *skb_gso_segment(struct sk_buff *skb, netdev_features_t features)
{
	return __skb_gso_segment(skb, features, true);
}
2986 2987 2988 2989 2990 2991 2992 2993 2994 2995 2996 2997 2998
__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)));
}
2999

3000
#ifdef CONFIG_BUG
3001
void netdev_rx_csum_fault(struct net_device *dev);
3002 3003 3004 3005 3006
#else
static inline void netdev_rx_csum_fault(struct net_device *dev)
{
}
#endif
L
Linus Torvalds 已提交
3007
/* rx skb timestamps */
3008 3009
void net_enable_timestamp(void);
void net_disable_timestamp(void);
L
Linus Torvalds 已提交
3010

3011
#ifdef CONFIG_PROC_FS
3012
int __init dev_proc_init(void);
3013 3014
#else
#define dev_proc_init() 0
3015 3016
#endif

3017 3018 3019 3020
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);
3021 3022 3023 3024 3025 3026 3027 3028 3029 3030

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);
}
3031

3032 3033
extern struct kobj_ns_type_operations net_ns_type_operations;

3034
const char *netdev_drivername(const struct net_device *dev);
3035

3036
void linkwatch_run_queue(void);
3037

3038 3039
static inline netdev_features_t netdev_get_wanted_features(
	struct net_device *dev)
3040 3041 3042
{
	return (dev->features & ~dev->hw_features) | dev->wanted_features;
}
3043 3044
netdev_features_t netdev_increment_features(netdev_features_t all,
	netdev_features_t one, netdev_features_t mask);
3045 3046 3047 3048 3049 3050 3051 3052 3053 3054 3055

/* 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);
}

3056
int __netdev_update_features(struct net_device *dev);
3057
void netdev_update_features(struct net_device *dev);
3058
void netdev_change_features(struct net_device *dev);
3059

3060 3061 3062
void netif_stacked_transfer_operstate(const struct net_device *rootdev,
					struct net_device *dev);

3063
netdev_features_t netif_skb_features(struct sk_buff *skb);
3064

3065
static inline bool net_gso_ok(netdev_features_t features, int gso_type)
3066
{
3067
	netdev_features_t feature = gso_type << NETIF_F_GSO_SHIFT;
3068 3069 3070 3071 3072 3073 3074 3075 3076

	/* 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));

3077
	return (features & feature) == feature;
3078 3079
}

3080
static inline bool skb_gso_ok(struct sk_buff *skb, netdev_features_t features)
3081
{
H
Herbert Xu 已提交
3082
	return net_gso_ok(features, skb_shinfo(skb)->gso_type) &&
3083
	       (!skb_has_frag_list(skb) || (features & NETIF_F_FRAGLIST));
3084 3085
}

3086 3087
static inline bool netif_needs_gso(struct sk_buff *skb,
				   netdev_features_t features)
3088
{
3089
	return skb_is_gso(skb) && (!skb_gso_ok(skb, features) ||
3090 3091
		unlikely((skb->ip_summed != CHECKSUM_PARTIAL) &&
			 (skb->ip_summed != CHECKSUM_UNNECESSARY)));
3092 3093
}

3094 3095 3096 3097 3098 3099
static inline void netif_set_gso_max_size(struct net_device *dev,
					  unsigned int size)
{
	dev->gso_max_size = size;
}

3100 3101 3102 3103 3104 3105 3106 3107 3108 3109 3110 3111 3112
static inline void skb_gso_error_unwind(struct sk_buff *skb, __be16 protocol,
					int pulled_hlen, u16 mac_offset,
					int mac_len)
{
	skb->protocol = protocol;
	skb->encapsulation = 1;
	skb_push(skb, pulled_hlen);
	skb_reset_transport_header(skb);
	skb->mac_header = mac_offset;
	skb->network_header = skb->mac_header + mac_len;
	skb->mac_len = mac_len;
}

3113 3114 3115 3116 3117
static inline bool netif_is_macvlan(struct net_device *dev)
{
	return dev->priv_flags & IFF_MACVLAN;
}

3118 3119 3120 3121 3122
static inline bool netif_is_bond_master(struct net_device *dev)
{
	return dev->flags & IFF_MASTER && dev->priv_flags & IFF_BONDING;
}

3123
static inline bool netif_is_bond_slave(struct net_device *dev)
J
Jiri Pirko 已提交
3124 3125 3126 3127
{
	return dev->flags & IFF_SLAVE && dev->priv_flags & IFF_BONDING;
}

3128 3129 3130 3131 3132
static inline bool netif_supports_nofcs(struct net_device *dev)
{
	return dev->priv_flags & IFF_SUPP_NOFCS;
}

3133
extern struct pernet_operations __net_initdata loopback_net_ops;
3134

3135 3136 3137 3138 3139 3140 3141 3142 3143 3144 3145
/* 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;
}

3146
__printf(3, 4)
3147 3148
int netdev_printk(const char *level, const struct net_device *dev,
		  const char *format, ...);
3149
__printf(2, 3)
3150
int netdev_emerg(const struct net_device *dev, const char *format, ...);
3151
__printf(2, 3)
3152
int netdev_alert(const struct net_device *dev, const char *format, ...);
3153
__printf(2, 3)
3154
int netdev_crit(const struct net_device *dev, const char *format, ...);
3155
__printf(2, 3)
3156
int netdev_err(const struct net_device *dev, const char *format, ...);
3157
__printf(2, 3)
3158
int netdev_warn(const struct net_device *dev, const char *format, ...);
3159
__printf(2, 3)
3160
int netdev_notice(const struct net_device *dev, const char *format, ...);
3161
__printf(2, 3)
3162
int netdev_info(const struct net_device *dev, const char *format, ...);
3163

3164 3165 3166
#define MODULE_ALIAS_NETDEV(device) \
	MODULE_ALIAS("netdev-" device)

3167
#if defined(CONFIG_DYNAMIC_DEBUG)
3168 3169
#define netdev_dbg(__dev, format, args...)			\
do {								\
3170
	dynamic_netdev_dbg(__dev, format, ##args);		\
3171
} while (0)
3172 3173 3174
#elif defined(DEBUG)
#define netdev_dbg(__dev, format, args...)			\
	netdev_printk(KERN_DEBUG, __dev, format, ##args)
3175 3176 3177 3178 3179 3180 3181 3182 3183 3184 3185 3186 3187 3188 3189 3190 3191 3192 3193 3194 3195 3196 3197 3198 3199 3200 3201
#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...)			\
3202
	WARN(1, "netdevice: %s\n" format, netdev_name(dev), ##args)
3203

3204 3205 3206 3207 3208 3209 3210 3211
/* 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)

3212 3213 3214 3215 3216 3217
#define netif_level(level, priv, type, dev, fmt, args...)	\
do {								\
	if (netif_msg_##type(priv))				\
		netdev_##level(dev, fmt, ##args);		\
} while (0)

3218
#define netif_emerg(priv, type, dev, fmt, args...)		\
3219
	netif_level(emerg, priv, type, dev, fmt, ##args)
3220
#define netif_alert(priv, type, dev, fmt, args...)		\
3221
	netif_level(alert, priv, type, dev, fmt, ##args)
3222
#define netif_crit(priv, type, dev, fmt, args...)		\
3223
	netif_level(crit, priv, type, dev, fmt, ##args)
3224
#define netif_err(priv, type, dev, fmt, args...)		\
3225
	netif_level(err, priv, type, dev, fmt, ##args)
3226
#define netif_warn(priv, type, dev, fmt, args...)		\
3227
	netif_level(warn, priv, type, dev, fmt, ##args)
3228
#define netif_notice(priv, type, dev, fmt, args...)		\
3229
	netif_level(notice, priv, type, dev, fmt, ##args)
3230
#define netif_info(priv, type, dev, fmt, args...)		\
3231
	netif_level(info, priv, type, dev, fmt, ##args)
3232

3233
#if defined(CONFIG_DYNAMIC_DEBUG)
3234 3235 3236
#define netif_dbg(priv, type, netdev, format, args...)		\
do {								\
	if (netif_msg_##type(priv))				\
3237
		dynamic_netdev_dbg(netdev, format, ##args);	\
3238
} while (0)
3239 3240 3241
#elif defined(DEBUG)
#define netif_dbg(priv, type, dev, format, args...)		\
	netif_printk(priv, type, KERN_DEBUG, dev, format, ##args)
3242 3243 3244 3245 3246 3247 3248 3249 3250 3251
#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)
3252
#define netif_vdbg	netif_dbg
3253 3254 3255 3256
#else
#define netif_vdbg(priv, type, dev, format, args...)		\
({								\
	if (0)							\
3257
		netif_printk(priv, type, KERN_DEBUG, dev, format, ##args); \
3258 3259 3260
	0;							\
})
#endif
3261

3262 3263 3264 3265 3266 3267 3268 3269 3270 3271 3272 3273 3274 3275 3276 3277 3278 3279 3280 3281 3282 3283 3284 3285 3286 3287 3288 3289 3290 3291
/*
 *	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)

3292
#endif	/* _LINUX_NETDEVICE_H */