netdevice.h 130.7 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

A
Al Viro 已提交
28
#include <linux/timer.h>
29
#include <linux/bug.h>
30
#include <linux/delay.h>
A
Arun Sharma 已提交
31
#include <linux/atomic.h>
32
#include <linux/prefetch.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
#include <uapi/linux/if_bonding.h>
54
#include <uapi/linux/pkt_cls.h>
55

56
struct netpoll_info;
57
struct device;
58
struct phy_device;
59 60
/* 802.11 specific */
struct wireless_dev;
61 62
/* 802.15.4 specific */
struct wpan_dev;
R
Robert Shearman 已提交
63
struct mpls_dev;
L
Linus Torvalds 已提交
64

65 66
void netdev_set_default_ethtool_ops(struct net_device *dev,
				    const struct ethtool_ops *ops);
67

68 69 70 71
/* Backlog congestion levels */
#define NET_RX_SUCCESS		0	/* keep 'em coming, baby */
#define NET_RX_DROP		1	/* packet dropped */

72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90
/*
 * 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
91 92 93 94
#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
95

96 97 98
/* 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. */
99
#define net_xmit_eval(e)	((e) == NET_XMIT_CN ? 0 : (e))
L
Linus Torvalds 已提交
100 101
#define net_xmit_errno(e)	((e) != NET_XMIT_CN ? -ENOBUFS : 0)

102
/* Driver transmit return codes */
103
#define NETDEV_TX_MASK		0xf0
104

105
enum netdev_tx {
106
	__NETDEV_TX_MIN	 = INT_MIN,	/* make sure enum is signed */
107 108 109
	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 */
110 111 112
};
typedef enum netdev_tx netdev_tx_t;

113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130
/*
 * 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
131 132 133 134
/*
 *	Compute the worst case header length according to the protocols
 *	used.
 */
G
Graf Yang 已提交
135

136 137 138
#if defined(CONFIG_HYPERV_NET)
# define LL_MAX_HEADER 128
#elif defined(CONFIG_WLAN) || IS_ENABLED(CONFIG_AX25)
139 140 141 142 143
# if defined(CONFIG_MAC80211_MESH)
#  define LL_MAX_HEADER 128
# else
#  define LL_MAX_HEADER 96
# endif
L
Linus Torvalds 已提交
144
#else
145
# define LL_MAX_HEADER 32
L
Linus Torvalds 已提交
146 147
#endif

148 149
#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
150 151 152 153 154 155
#define MAX_HEADER LL_MAX_HEADER
#else
#define MAX_HEADER (LL_MAX_HEADER + 48)
#endif

/*
156 157
 *	Old network device statistics. Fields are native words
 *	(unsigned long) so they can be read and written atomically.
L
Linus Torvalds 已提交
158
 */
G
Graf Yang 已提交
159

E
Eric Dumazet 已提交
160
struct net_device_stats {
161 162 163 164 165 166 167 168 169
	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
170 171
	unsigned long	collisions;
	unsigned long	rx_length_errors;
172 173 174 175 176
	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
177 178 179 180 181 182 183 184 185 186 187 188 189
	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>

190
#ifdef CONFIG_RPS
191 192
#include <linux/static_key.h>
extern struct static_key rps_needed;
193 194
#endif

L
Linus Torvalds 已提交
195 196 197 198
struct neighbour;
struct neigh_parms;
struct sk_buff;

199 200 201 202
struct netdev_hw_addr {
	struct list_head	list;
	unsigned char		addr[MAX_ADDR_LEN];
	unsigned char		type;
J
Jiri Pirko 已提交
203 204 205 206
#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
207 208
#define NETDEV_HW_ADDR_T_MULTICAST	5
	bool			global_use;
209
	int			sync_cnt;
210
	int			refcount;
211
	int			synced;
212 213 214
	struct rcu_head		rcu_head;
};

215 216 217 218 219
struct netdev_hw_addr_list {
	struct list_head	list;
	int			count;
};

220 221 222 223
#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)
224

225 226 227 228
#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)
229

230 231
#define netdev_mc_count(dev) netdev_hw_addr_list_count(&(dev)->mc)
#define netdev_mc_empty(dev) netdev_hw_addr_list_empty(&(dev)->mc)
232
#define netdev_for_each_mc_addr(ha, dev) \
233
	netdev_hw_addr_list_for_each(ha, &(dev)->mc)
234

E
Eric Dumazet 已提交
235
struct hh_cache {
236
	u16		hh_len;
237
	u16		__pad;
238
	seqlock_t	hh_lock;
L
Linus Torvalds 已提交
239 240 241 242

	/* cached hardware header; allow for machine alignment needs.        */
#define HH_DATA_MOD	16
#define HH_DATA_OFF(__len) \
J
Jiri Benc 已提交
243
	(HH_DATA_MOD - (((__len - 1) & (HH_DATA_MOD - 1)) + 1))
L
Linus Torvalds 已提交
244 245 246 247 248 249 250 251 252 253 254 255 256 257
#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) \
258
	((((dev)->hard_header_len+(dev)->needed_headroom)&~(HH_DATA_MOD - 1)) + HH_DATA_MOD)
L
Linus Torvalds 已提交
259
#define LL_RESERVED_SPACE_EXTRA(dev,extra) \
260
	((((dev)->hard_header_len+(dev)->needed_headroom+(extra))&~(HH_DATA_MOD - 1)) + HH_DATA_MOD)
L
Linus Torvalds 已提交
261

262 263 264
struct header_ops {
	int	(*create) (struct sk_buff *skb, struct net_device *dev,
			   unsigned short type, const void *daddr,
265
			   const void *saddr, unsigned int len);
266
	int	(*parse)(const struct sk_buff *skb, unsigned char *haddr);
267
	int	(*cache)(const struct neighbour *neigh, struct hh_cache *hh, __be16 type);
268 269 270 271 272
	void	(*cache_update)(struct hh_cache *hh,
				const struct net_device *dev,
				const unsigned char *haddr);
};

L
Linus Torvalds 已提交
273 274 275 276 277
/* 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
278
enum netdev_state_t {
L
Linus Torvalds 已提交
279 280 281
	__LINK_STATE_START,
	__LINK_STATE_PRESENT,
	__LINK_STATE_NOCARRIER,
S
Stefan Rompf 已提交
282 283
	__LINK_STATE_LINKWATCH_PENDING,
	__LINK_STATE_DORMANT,
L
Linus Torvalds 已提交
284 285 286 287 288
};


/*
 * This structure holds at boot time configured netdevice settings. They
G
Graf Yang 已提交
289
 * are then used in the device probing.
L
Linus Torvalds 已提交
290 291 292 293 294 295 296
 */
struct netdev_boot_setup {
	char name[IFNAMSIZ];
	struct ifmap map;
};
#define NETDEV_BOOT_SETUP_MAX 8

297
int __init netdev_boot_setup(char *str);
L
Linus Torvalds 已提交
298

299 300 301 302 303 304 305 306 307 308 309 310 311 312
/*
 * 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;
313
	unsigned int		gro_count;
314 315 316 317 318
	int			(*poll)(struct napi_struct *, int);
#ifdef CONFIG_NETPOLL
	spinlock_t		poll_lock;
	int			poll_owner;
#endif
H
Herbert Xu 已提交
319
	struct net_device	*dev;
320
	struct sk_buff		*gro_list;
H
Herbert Xu 已提交
321
	struct sk_buff		*skb;
322
	struct hrtimer		timer;
323
	struct list_head	dev_list;
E
Eliezer Tamir 已提交
324 325
	struct hlist_node	napi_hash_node;
	unsigned int		napi_id;
326 327
};

E
Eric Dumazet 已提交
328
enum {
329
	NAPI_STATE_SCHED,	/* Poll is scheduled */
D
David S. Miller 已提交
330
	NAPI_STATE_DISABLE,	/* Disable pending */
331
	NAPI_STATE_NPSVC,	/* Netpoll - don't dequeue from poll_list */
E
Eric Dumazet 已提交
332 333
	NAPI_STATE_HASHED,	/* In NAPI hash (busy polling possible) */
	NAPI_STATE_NO_BUSY_POLL,/* Do not add in napi_hash, no busy polling */
334 335
};

336
enum gro_result {
337 338 339 340 341 342
	GRO_MERGED,
	GRO_MERGED_FREE,
	GRO_HELD,
	GRO_NORMAL,
	GRO_DROP,
};
343
typedef enum gro_result gro_result_t;
344

345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376
/*
 * 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
377
 * are registered on exact device (ptype->dev == skb->dev).
378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393
 *
 * 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);
394

395
void __napi_schedule(struct napi_struct *n);
396
void __napi_schedule_irqoff(struct napi_struct *n);
397

398
static inline bool napi_disable_pending(struct napi_struct *n)
D
David S. Miller 已提交
399 400 401 402
{
	return test_bit(NAPI_STATE_DISABLE, &n->state);
}

403 404 405 406 407 408
/**
 *	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
409 410
 * insure only one NAPI poll instance runs.  We also make
 * sure there is no pending NAPI disable.
411
 */
412
static inline bool napi_schedule_prep(struct napi_struct *n)
413
{
D
David S. Miller 已提交
414 415
	return !napi_disable_pending(n) &&
		!test_and_set_bit(NAPI_STATE_SCHED, &n->state);
416 417 418 419 420 421 422 423 424 425 426 427 428 429 430
}

/**
 *	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);
}

431 432 433 434 435 436 437 438 439 440 441 442
/**
 *	napi_schedule_irqoff - schedule NAPI poll
 *	@n: napi context
 *
 * Variant of napi_schedule(), assuming hard irqs are masked.
 */
static inline void napi_schedule_irqoff(struct napi_struct *n)
{
	if (napi_schedule_prep(n))
		__napi_schedule_irqoff(n);
}

443
/* Try to reschedule poll. Called by dev->poll() after napi_complete().  */
444
static inline bool napi_reschedule(struct napi_struct *napi)
445 446 447
{
	if (napi_schedule_prep(napi)) {
		__napi_schedule(napi);
448
		return true;
449
	}
450
	return false;
451 452
}

453 454
void __napi_complete(struct napi_struct *n);
void napi_complete_done(struct napi_struct *n, int work_done);
455 456 457 458 459
/**
 *	napi_complete - NAPI processing complete
 *	@n: napi context
 *
 * Mark NAPI processing as complete.
460
 * Consider using napi_complete_done() instead.
461
 */
462 463 464 465
static inline void napi_complete(struct napi_struct *n)
{
	return napi_complete_done(n, 0);
}
466

E
Eliezer Tamir 已提交
467 468 469 470 471
/**
 *	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
472 473 474
 * Used for busy polling (CONFIG_NET_RX_BUSY_POLL)
 * Note: This is normally automatically done from netif_napi_add(),
 * so might disappear in a future linux version.
E
Eliezer Tamir 已提交
475
 */
476
void napi_hash_add(struct napi_struct *napi);
E
Eliezer Tamir 已提交
477 478 479 480 481 482

/**
 *	napi_hash_del - remove a NAPI from global table
 *	@napi: napi context
 *
 * Warning: caller must observe rcu grace period
483 484
 * before freeing memory containing @napi, if
 * this function returns true.
485 486 487 488
 * Note: core networking stack automatically calls it
 * from netif_napi_del()
 * Drivers might want to call this helper to combine all
 * the needed rcu grace periods into a single one.
E
Eliezer Tamir 已提交
489
 */
490
bool napi_hash_del(struct napi_struct *napi);
E
Eliezer Tamir 已提交
491

492 493 494 495 496 497 498
/**
 *	napi_disable - prevent NAPI from scheduling
 *	@n: napi context
 *
 * Stop NAPI from being scheduled on this context.
 * Waits till any outstanding processing completes.
 */
499
void napi_disable(struct napi_struct *n);
500 501 502 503 504 505 506 507 508 509 510

/**
 *	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));
511
	smp_mb__before_atomic();
512
	clear_bit(NAPI_STATE_SCHED, &n->state);
513
	clear_bit(NAPI_STATE_NPSVC, &n->state);
514 515
}

516 517 518 519 520 521 522 523 524 525
/**
 *	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)
{
526 527 528 529 530
	if (IS_ENABLED(CONFIG_SMP))
		while (test_bit(NAPI_STATE_SCHED, &n->state))
			msleep(1);
	else
		barrier();
531 532
}

E
Eric Dumazet 已提交
533
enum netdev_queue_state_t {
534 535
	__QUEUE_STATE_DRV_XOFF,
	__QUEUE_STATE_STACK_XOFF,
536
	__QUEUE_STATE_FROZEN,
537
};
538 539 540 541 542 543 544 545 546 547 548

#define QUEUE_STATE_DRV_XOFF	(1 << __QUEUE_STATE_DRV_XOFF)
#define QUEUE_STATE_STACK_XOFF	(1 << __QUEUE_STATE_STACK_XOFF)
#define QUEUE_STATE_FROZEN	(1 << __QUEUE_STATE_FROZEN)

#define QUEUE_STATE_ANY_XOFF	(QUEUE_STATE_DRV_XOFF | QUEUE_STATE_STACK_XOFF)
#define QUEUE_STATE_ANY_XOFF_OR_FROZEN (QUEUE_STATE_ANY_XOFF | \
					QUEUE_STATE_FROZEN)
#define QUEUE_STATE_DRV_XOFF_OR_FROZEN (QUEUE_STATE_DRV_XOFF | \
					QUEUE_STATE_FROZEN)

549 550 551 552 553 554 555 556 557
/*
 * __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_*.
 */
558

559
struct netdev_queue {
560 561 562
/*
 * read mostly part
 */
563
	struct net_device	*dev;
564
	struct Qdisc __rcu	*qdisc;
565
	struct Qdisc		*qdisc_sleeping;
566
#ifdef CONFIG_SYSFS
T
Tom Herbert 已提交
567 568
	struct kobject		kobj;
#endif
569 570 571
#if defined(CONFIG_XPS) && defined(CONFIG_NUMA)
	int			numa_node;
#endif
572 573 574 575 576
/*
 * write mostly part
 */
	spinlock_t		_xmit_lock ____cacheline_aligned_in_smp;
	int			xmit_lock_owner;
577 578 579 580
	/*
	 * please use this field instead of dev->trans_start
	 */
	unsigned long		trans_start;
581 582 583 584 585 586

	/*
	 * Number of TX timeouts for this queue
	 * (/sys/class/net/DEV/Q/trans_timeout)
	 */
	unsigned long		trans_timeout;
T
Tom Herbert 已提交
587 588 589 590 591 592

	unsigned long		state;

#ifdef CONFIG_BQL
	struct dql		dql;
#endif
593
	unsigned long		tx_maxrate;
594
} ____cacheline_aligned_in_smp;
595

596 597 598 599 600
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
601
	return NUMA_NO_NODE;
602 603 604 605 606 607 608 609 610 611
#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
612
#ifdef CONFIG_RPS
T
Tom Herbert 已提交
613 614 615 616 617 618 619 620 621
/*
 * 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];
};
622
#define RPS_MAP_SIZE(_num) (sizeof(struct rps_map) + ((_num) * sizeof(u16)))
T
Tom Herbert 已提交
623

T
Tom Herbert 已提交
624
/*
625 626 627
 * 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
628 629 630
 */
struct rps_dev_flow {
	u16 cpu;
631
	u16 filter;
T
Tom Herbert 已提交
632 633
	unsigned int last_qtail;
};
634
#define RPS_NO_FILTER 0xffff
T
Tom Herbert 已提交
635 636 637 638 639 640 641 642 643 644

/*
 * 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) + \
645
    ((_num) * sizeof(struct rps_dev_flow)))
T
Tom Herbert 已提交
646 647 648 649

/*
 * The rps_sock_flow_table contains mappings of flows to the last CPU
 * on which they were processed by the application (set in recvmsg).
650 651 652 653 654 655
 * Each entry is a 32bit value. Upper part is the high order bits
 * of flow hash, lower part is cpu number.
 * rps_cpu_mask is used to partition the space, depending on number of
 * possible cpus : rps_cpu_mask = roundup_pow_of_two(nr_cpu_ids) - 1
 * For example, if 64 cpus are possible, rps_cpu_mask = 0x3f,
 * meaning we use 32-6=26 bits for the hash.
T
Tom Herbert 已提交
656 657
 */
struct rps_sock_flow_table {
658
	u32	mask;
659 660

	u32	ents[0] ____cacheline_aligned_in_smp;
T
Tom Herbert 已提交
661
};
662
#define	RPS_SOCK_FLOW_TABLE_SIZE(_num) (offsetof(struct rps_sock_flow_table, ents[_num]))
T
Tom Herbert 已提交
663 664 665

#define RPS_NO_CPU 0xffff

666 667 668
extern u32 rps_cpu_mask;
extern struct rps_sock_flow_table __rcu *rps_sock_flow_table;

T
Tom Herbert 已提交
669 670 671 672
static inline void rps_record_sock_flow(struct rps_sock_flow_table *table,
					u32 hash)
{
	if (table && hash) {
673 674
		unsigned int index = hash & table->mask;
		u32 val = hash & ~rps_cpu_mask;
T
Tom Herbert 已提交
675 676

		/* We only give a hint, preemption can change cpu under us */
677
		val |= raw_smp_processor_id();
T
Tom Herbert 已提交
678

679 680
		if (table->ents[index] != val)
			table->ents[index] = val;
T
Tom Herbert 已提交
681 682 683
	}
}

684
#ifdef CONFIG_RFS_ACCEL
685 686
bool rps_may_expire_flow(struct net_device *dev, u16 rxq_index, u32 flow_id,
			 u16 filter_id);
687
#endif
688
#endif /* CONFIG_RPS */
689

T
Tom Herbert 已提交
690 691
/* This structure contains an instance of an RX queue. */
struct netdev_rx_queue {
692
#ifdef CONFIG_RPS
E
Eric Dumazet 已提交
693 694
	struct rps_map __rcu		*rps_map;
	struct rps_dev_flow_table __rcu	*rps_flow_table;
695
#endif
E
Eric Dumazet 已提交
696
	struct kobject			kobj;
T
Tom Herbert 已提交
697
	struct net_device		*dev;
T
Tom Herbert 已提交
698
} ____cacheline_aligned_in_smp;
699 700 701 702 703 704 705 706 707 708 709

/*
 * 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);
};
710

T
Tom Herbert 已提交
711 712 713 714 715 716 717 718 719 720 721
#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];
};
722
#define XPS_MAP_SIZE(_num) (sizeof(struct xps_map) + ((_num) * sizeof(u16)))
723 724
#define XPS_MIN_MAP_ALLOC ((L1_CACHE_ALIGN(offsetof(struct xps_map, queues[1])) \
       - sizeof(struct xps_map)) / sizeof(u16))
T
Tom Herbert 已提交
725 726 727 728 729 730

/*
 * 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
731
	struct xps_map __rcu *cpu_map[0];
T
Tom Herbert 已提交
732 733 734 735 736
};
#define XPS_DEV_MAPS_SIZE (sizeof(struct xps_dev_maps) +		\
    (nr_cpu_ids * sizeof(struct xps_map *)))
#endif /* CONFIG_XPS */

737 738 739 740 741 742 743 744
#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;
};

745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761
#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

762
#define MAX_PHYS_ITEM_ID_LEN 32
763

764 765
/* This structure holds a unique identifier to identify some
 * physical item (port for example) used by a netdevice.
766
 */
767 768
struct netdev_phys_item_id {
	unsigned char id[MAX_PHYS_ITEM_ID_LEN];
769 770 771
	unsigned char id_len;
};

772 773 774 775 776 777 778
static inline bool netdev_phys_item_id_same(struct netdev_phys_item_id *a,
					    struct netdev_phys_item_id *b)
{
	return a->id_len == b->id_len &&
	       memcmp(a->id, b->id, a->id_len) == 0;
}

779 780 781
typedef u16 (*select_queue_fallback_t)(struct net_device *dev,
				       struct sk_buff *skb);

782
/* These structures hold the attributes of qdisc and classifiers
783 784 785 786
 * that are being passed to the netdevice through the setup_tc op.
 */
enum {
	TC_SETUP_MQPRIO,
787
	TC_SETUP_CLSU32,
788 789
};

790 791
struct tc_cls_u32_offload;

792 793 794 795
struct tc_to_netdev {
	unsigned int type;
	union {
		u8 tc;
796
		struct tc_cls_u32_offload *cls_u32;
797 798 799 800
	};
};


801 802
/*
 * This structure defines the management hooks for network devices.
803 804
 * The following hooks can be defined; unless noted otherwise, they are
 * optional and can be filled with a null pointer.
805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823
 *
 * 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.
 *
824 825
 * netdev_tx_t (*ndo_start_xmit)(struct sk_buff *skb,
 *                               struct net_device *dev);
826
 *	Called when a packet needs to be transmitted.
827 828 829 830
 *	Returns NETDEV_TX_OK.  Can return NETDEV_TX_BUSY, but you should stop
 *	the queue before that can happen; it's for obsolete devices and weird
 *	corner cases, but the stack really does a non-trivial amount
 *	of useless work if you return NETDEV_TX_BUSY.
831
 *        (can also return NETDEV_TX_LOCKED iff NETIF_F_LLTX)
832 833
 *	Required can not be NULL.
 *
834 835 836 837 838 839
 * netdev_features_t (*ndo_fix_features)(struct net_device *dev,
 *		netdev_features_t features);
 *	Adjusts the requested feature flags according to device-specific
 *	constraints, and returns the resulting flags. Must not modify
 *	the device state.
 *
840
 * u16 (*ndo_select_queue)(struct net_device *dev, struct sk_buff *skb,
841
 *                         void *accel_priv, select_queue_fallback_t fallback);
842 843 844
 *	Called to decide which queue to when device supports multiple
 *	transmit queues.
 *
845 846 847 848 849 850
 * 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.
851 852
 *	If driver handles unicast address filtering, it should set
 *	IFF_UNICAST_FLT to its priv_flags.
853 854 855
 *
 * int (*ndo_set_mac_address)(struct net_device *dev, void *addr);
 *	This function  is called when the Media Access Control address
856
 *	needs to be changed. If this interface is not defined, the
857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876
 *	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.
 *
877
 * void (*ndo_tx_timeout)(struct net_device *dev);
878 879 880
 *	Callback uses when the transmitter has not made any progress
 *	for dev->watchdog ticks.
 *
881
 * struct rtnl_link_stats64* (*ndo_get_stats64)(struct net_device *dev,
882
 *                      struct rtnl_link_stats64 *storage);
883
 * struct net_device_stats* (*ndo_get_stats)(struct net_device *dev);
884
 *	Called when a user wants to get the network device usage
885
 *	statistics. Drivers must do one of the following:
886 887
 *	1. Define @ndo_get_stats64 to fill in a zero-initialised
 *	   rtnl_link_stats64 structure passed by the caller.
888
 *	2. Define @ndo_get_stats to update a net_device_stats structure
889 890 891 892 893
 *	   (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.
894
 *
895
 * int (*ndo_vlan_rx_add_vid)(struct net_device *dev, __be16 proto, u16 vid);
896 897
 *	If device support VLAN filtering this function is called when a
 *	VLAN id is registered.
898
 *
899
 * int (*ndo_vlan_rx_kill_vid)(struct net_device *dev, __be16 proto, u16 vid);
900 901
 *	If device support VLAN filtering this function is called when a
 *	VLAN id is unregistered.
902 903
 *
 * void (*ndo_poll_controller)(struct net_device *dev);
904 905 906 907
 *
 *	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);
908 909
 * int (*ndo_set_vf_rate)(struct net_device *dev, int vf, int min_tx_rate,
 *			  int max_tx_rate);
910
 * int (*ndo_set_vf_spoofchk)(struct net_device *dev, int vf, bool setting);
H
Hiroshi Shimamoto 已提交
911
 * int (*ndo_set_vf_trust)(struct net_device *dev, int vf, bool setting);
912 913
 * int (*ndo_get_vf_config)(struct net_device *dev,
 *			    int vf, struct ifla_vf_info *ivf);
914
 * int (*ndo_set_vf_link_state)(struct net_device *dev, int vf, int link_state);
915 916
 * int (*ndo_set_vf_port)(struct net_device *dev, int vf,
 *			  struct nlattr *port[]);
917 918 919 920 921
 *
 *      Enable or disable the VF ability to query its RSS Redirection Table and
 *      Hash Key. This is needed since on some devices VF share this information
 *      with PF and querying it may adduce a theoretical security risk.
 * int (*ndo_set_vf_rss_query_en)(struct net_device *dev, int vf, bool setting);
922
 * int (*ndo_get_vf_port)(struct net_device *dev, int vf, struct sk_buff *skb);
923 924 925 926 927
 * 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.
928
 *
929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958
 *	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.
 *
959 960 961 962 963 964 965
 * 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.
 *
966 967 968 969 970 971
 * 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.
 *
972 973 974 975 976 977
 *	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.
978
 *
979
 *	Slave management functions (for bridge, bonding, etc).
980 981 982 983 984
 * 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.
985 986
 *
 *      Feature/offload setting functions.
987
 * int (*ndo_set_features)(struct net_device *dev, netdev_features_t features);
988 989 990 991
 *	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.
 *
992 993
 * int (*ndo_fdb_add)(struct ndmsg *ndm, struct nlattr *tb[],
 *		      struct net_device *dev,
994
 *		      const unsigned char *addr, u16 vid, u16 flags)
995
 *	Adds an FDB entry to dev for addr.
996 997
 * int (*ndo_fdb_del)(struct ndmsg *ndm, struct nlattr *tb[],
 *		      struct net_device *dev,
998
 *		      const unsigned char *addr, u16 vid)
999 1000
 *	Deletes the FDB entry from dev coresponding to addr.
 * int (*ndo_fdb_dump)(struct sk_buff *skb, struct netlink_callback *cb,
1001 1002
 *		       struct net_device *dev, struct net_device *filter_dev,
 *		       int idx)
1003 1004
 *	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
1005
 *
1006 1007
 * int (*ndo_bridge_setlink)(struct net_device *dev, struct nlmsghdr *nlh,
 *			     u16 flags)
J
John Fastabend 已提交
1008
 * int (*ndo_bridge_getlink)(struct sk_buff *skb, u32 pid, u32 seq,
1009 1010
 *			     struct net_device *dev, u32 filter_mask,
 *			     int nlflags)
1011 1012
 * int (*ndo_bridge_dellink)(struct net_device *dev, struct nlmsghdr *nlh,
 *			     u16 flags);
J
Jiri Pirko 已提交
1013 1014 1015 1016 1017 1018 1019 1020
 *
 * 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.
1021 1022
 *
 * int (*ndo_get_phys_port_id)(struct net_device *dev,
1023
 *			       struct netdev_phys_item_id *ppid);
1024 1025 1026
 *	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.
1027 1028
 *
 * void (*ndo_add_vxlan_port)(struct  net_device *dev,
J
Joseph Gasparakis 已提交
1029
 *			      sa_family_t sa_family, __be16 port);
1030 1031 1032 1033 1034
 *	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.
 *
1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047
 * void (*ndo_add_geneve_port)(struct net_device *dev,
 *			      sa_family_t sa_family, __be16 port);
 *	Called by geneve to notify a driver about the UDP port and socket
 *	address family that geneve is listnening to. It is called only when
 *	a new port starts listening. The operation is protected by the
 *	geneve_net->sock_lock.
 *
 * void (*ndo_del_geneve_port)(struct net_device *dev,
 *			      sa_family_t sa_family, __be16 port);
 *	Called by geneve to notify the driver about a UDP port and socket
 *	address family that geneve is not listening to anymore. The operation
 *	is protected by the geneve_net->sock_lock.
 *
1048
 * void (*ndo_del_vxlan_port)(struct  net_device *dev,
J
Joseph Gasparakis 已提交
1049
 *			      sa_family_t sa_family, __be16 port);
1050 1051 1052
 *	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.
1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071
 *
 * 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.
1072 1073 1074
 * netdev_features_t (*ndo_features_check) (struct sk_buff *skb,
 *					    struct net_device *dev
 *					    netdev_features_t features);
T
Tom Herbert 已提交
1075
 *	Called by core transmit path to determine if device is capable of
1076 1077 1078 1079 1080
 *	performing offload operations on a given packet. This is to give
 *	the device an opportunity to implement any restrictions that cannot
 *	be otherwise expressed by feature flags. The check is called with
 *	the set of features that the stack has calculated and it returns
 *	those the driver believes to be appropriate.
1081 1082 1083 1084
 * int (*ndo_set_tx_maxrate)(struct net_device *dev,
 *			     int queue_index, u32 maxrate);
 *	Called when a user wants to set a max-rate limitation of specific
 *	TX queue.
1085 1086
 * int (*ndo_get_iflink)(const struct net_device *dev);
 *	Called to get the iflink value of this device.
1087 1088 1089 1090 1091
 * void (*ndo_change_proto_down)(struct net_device *dev,
 *				  bool proto_down);
 *	This function is used to pass protocol port error state information
 *	to the switch driver. The switch driver can react to the proto_down
 *      by doing a phys down on the associated switch port.
1092 1093 1094 1095
 * int (*ndo_fill_metadata_dst)(struct net_device *dev, struct sk_buff *skb);
 *	This function is used to get egress tunnel information for given skb.
 *	This is useful for retrieving outer tunnel header parameters while
 *	sampling packet.
1096 1097 1098 1099 1100 1101
 * void (*ndo_set_rx_headroom)(struct net_device *dev, int needed_headroom);
 *	This function is used to specify the headroom that the skb must
 *	consider when allocation skb during packet reception. Setting
 *	appropriate rx headroom value allows avoiding skb head copy on
 *	forward. Setting a negative value reset the rx headroom to the
 *	default value.
1102
 *
1103 1104 1105 1106 1107 1108
 */
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);
1109 1110 1111 1112 1113
	netdev_tx_t		(*ndo_start_xmit)(struct sk_buff *skb,
						  struct net_device *dev);
	netdev_features_t	(*ndo_features_check)(struct sk_buff *skb,
						      struct net_device *dev,
						      netdev_features_t features);
1114
	u16			(*ndo_select_queue)(struct net_device *dev,
1115
						    struct sk_buff *skb,
1116 1117
						    void *accel_priv,
						    select_queue_fallback_t fallback);
1118 1119 1120 1121 1122 1123 1124 1125 1126 1127
	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);
1128 1129 1130 1131
	int			(*ndo_change_mtu)(struct net_device *dev,
						  int new_mtu);
	int			(*ndo_neigh_setup)(struct net_device *dev,
						   struct neigh_parms *);
1132 1133
	void			(*ndo_tx_timeout) (struct net_device *dev);

1134 1135
	struct rtnl_link_stats64* (*ndo_get_stats64)(struct net_device *dev,
						     struct rtnl_link_stats64 *storage);
1136 1137
	struct net_device_stats* (*ndo_get_stats)(struct net_device *dev);

1138
	int			(*ndo_vlan_rx_add_vid)(struct net_device *dev,
1139
						       __be16 proto, u16 vid);
1140
	int			(*ndo_vlan_rx_kill_vid)(struct net_device *dev,
1141
						        __be16 proto, u16 vid);
1142 1143
#ifdef CONFIG_NET_POLL_CONTROLLER
	void                    (*ndo_poll_controller)(struct net_device *dev);
H
Herbert Xu 已提交
1144
	int			(*ndo_netpoll_setup)(struct net_device *dev,
1145
						     struct netpoll_info *info);
1146
	void			(*ndo_netpoll_cleanup)(struct net_device *dev);
E
Eliezer Tamir 已提交
1147
#endif
1148
#ifdef CONFIG_NET_RX_BUSY_POLL
1149
	int			(*ndo_busy_poll)(struct napi_struct *dev);
1150
#endif
1151 1152 1153 1154
	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);
1155 1156 1157
	int			(*ndo_set_vf_rate)(struct net_device *dev,
						   int vf, int min_tx_rate,
						   int max_tx_rate);
1158 1159
	int			(*ndo_set_vf_spoofchk)(struct net_device *dev,
						       int vf, bool setting);
H
Hiroshi Shimamoto 已提交
1160 1161
	int			(*ndo_set_vf_trust)(struct net_device *dev,
						    int vf, bool setting);
1162 1163 1164
	int			(*ndo_get_vf_config)(struct net_device *dev,
						     int vf,
						     struct ifla_vf_info *ivf);
1165 1166
	int			(*ndo_set_vf_link_state)(struct net_device *dev,
							 int vf, int link_state);
1167 1168 1169 1170
	int			(*ndo_get_vf_stats)(struct net_device *dev,
						    int vf,
						    struct ifla_vf_stats
						    *vf_stats);
1171 1172 1173 1174 1175
	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);
1176 1177 1178
	int			(*ndo_set_vf_rss_query_en)(
						   struct net_device *dev,
						   int vf, bool setting);
1179 1180 1181 1182
	int			(*ndo_setup_tc)(struct net_device *dev,
						u32 handle,
						__be16 protocol,
						struct tc_to_netdev *tc);
1183
#if IS_ENABLED(CONFIG_FCOE)
1184 1185
	int			(*ndo_fcoe_enable)(struct net_device *dev);
	int			(*ndo_fcoe_disable)(struct net_device *dev);
1186 1187 1188 1189 1190 1191
	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);
1192 1193 1194 1195
	int			(*ndo_fcoe_ddp_target)(struct net_device *dev,
						       u16 xid,
						       struct scatterlist *sgl,
						       unsigned int sgc);
1196 1197
	int			(*ndo_fcoe_get_hbainfo)(struct net_device *dev,
							struct netdev_fcoe_hbainfo *hbainfo);
1198 1199
#endif

1200
#if IS_ENABLED(CONFIG_LIBFCOE)
1201 1202 1203 1204
#define NETDEV_FCOE_WWNN 0
#define NETDEV_FCOE_WWPN 1
	int			(*ndo_fcoe_get_wwn)(struct net_device *dev,
						    u64 *wwn, int type);
1205
#endif
1206

1207 1208 1209 1210 1211 1212
#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
1213 1214 1215 1216
	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);
1217 1218
	netdev_features_t	(*ndo_fix_features)(struct net_device *dev,
						    netdev_features_t features);
1219
	int			(*ndo_set_features)(struct net_device *dev,
1220
						    netdev_features_t features);
1221
	int			(*ndo_neigh_construct)(struct neighbour *n);
1222
	void			(*ndo_neigh_destroy)(struct neighbour *n);
1223 1224

	int			(*ndo_fdb_add)(struct ndmsg *ndm,
1225
					       struct nlattr *tb[],
1226
					       struct net_device *dev,
1227
					       const unsigned char *addr,
1228
					       u16 vid,
1229 1230
					       u16 flags);
	int			(*ndo_fdb_del)(struct ndmsg *ndm,
1231
					       struct nlattr *tb[],
1232
					       struct net_device *dev,
1233 1234
					       const unsigned char *addr,
					       u16 vid);
1235 1236 1237
	int			(*ndo_fdb_dump)(struct sk_buff *skb,
						struct netlink_callback *cb,
						struct net_device *dev,
1238
						struct net_device *filter_dev,
1239
						int idx);
J
John Fastabend 已提交
1240 1241

	int			(*ndo_bridge_setlink)(struct net_device *dev,
1242 1243
						      struct nlmsghdr *nlh,
						      u16 flags);
J
John Fastabend 已提交
1244 1245
	int			(*ndo_bridge_getlink)(struct sk_buff *skb,
						      u32 pid, u32 seq,
1246
						      struct net_device *dev,
1247 1248
						      u32 filter_mask,
						      int nlflags);
1249
	int			(*ndo_bridge_dellink)(struct net_device *dev,
1250 1251
						      struct nlmsghdr *nlh,
						      u16 flags);
J
Jiri Pirko 已提交
1252 1253
	int			(*ndo_change_carrier)(struct net_device *dev,
						      bool new_carrier);
1254
	int			(*ndo_get_phys_port_id)(struct net_device *dev,
1255
							struct netdev_phys_item_id *ppid);
1256 1257
	int			(*ndo_get_phys_port_name)(struct net_device *dev,
							  char *name, size_t len);
1258 1259
	void			(*ndo_add_vxlan_port)(struct  net_device *dev,
						      sa_family_t sa_family,
J
Joseph Gasparakis 已提交
1260
						      __be16 port);
1261 1262
	void			(*ndo_del_vxlan_port)(struct  net_device *dev,
						      sa_family_t sa_family,
J
Joseph Gasparakis 已提交
1263
						      __be16 port);
1264 1265 1266 1267 1268 1269
	void			(*ndo_add_geneve_port)(struct  net_device *dev,
						       sa_family_t sa_family,
						       __be16 port);
	void			(*ndo_del_geneve_port)(struct  net_device *dev,
						       sa_family_t sa_family,
						       __be16 port);
1270 1271 1272 1273 1274 1275 1276 1277
	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);
1278
	int			(*ndo_get_lock_subclass)(struct net_device *dev);
1279 1280 1281
	int			(*ndo_set_tx_maxrate)(struct net_device *dev,
						      int queue_index,
						      u32 maxrate);
1282
	int			(*ndo_get_iflink)(const struct net_device *dev);
1283 1284
	int			(*ndo_change_proto_down)(struct net_device *dev,
							 bool proto_down);
1285 1286
	int			(*ndo_fill_metadata_dst)(struct net_device *dev,
						       struct sk_buff *skb);
1287 1288
	void			(*ndo_set_rx_headroom)(struct net_device *dev,
						       int needed_headroom);
1289 1290
};

1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319
/**
 * enum net_device_priv_flags - &struct net_device priv_flags
 *
 * These are the &struct net_device, they are only set internally
 * by drivers and used in the kernel. These flags are invisible to
 * userspace, this means that the order of these flags can change
 * during any kernel release.
 *
 * You should have a pretty good reason to be extending these flags.
 *
 * @IFF_802_1Q_VLAN: 802.1Q VLAN device
 * @IFF_EBRIDGE: Ethernet bridging device
 * @IFF_BONDING: bonding master or slave
 * @IFF_ISATAP: ISATAP interface (RFC4214)
 * @IFF_WAN_HDLC: WAN HDLC device
 * @IFF_XMIT_DST_RELEASE: dev_hard_start_xmit() is allowed to
 *	release skb->dst
 * @IFF_DONT_BRIDGE: disallow bridging this ether dev
 * @IFF_DISABLE_NETPOLL: disable netpoll at run-time
 * @IFF_MACVLAN_PORT: device used as macvlan port
 * @IFF_BRIDGE_PORT: device used as bridge port
 * @IFF_OVS_DATAPATH: device used as Open vSwitch datapath port
 * @IFF_TX_SKB_SHARING: The interface supports sharing skbs on transmit
 * @IFF_UNICAST_FLT: Supports unicast filtering
 * @IFF_TEAM_PORT: device used as team port
 * @IFF_SUPP_NOFCS: device supports sending custom FCS
 * @IFF_LIVE_ADDR_CHANGE: device supports hardware address
 *	change when it's running
 * @IFF_MACVLAN: Macvlan device
1320
 * @IFF_L3MDEV_MASTER: device is an L3 master device
1321
 * @IFF_NO_QUEUE: device can run without qdisc attached
1322
 * @IFF_OPENVSWITCH: device is a Open vSwitch master
D
David Ahern 已提交
1323
 * @IFF_L3MDEV_SLAVE: device is enslaved to an L3 master device
J
Jiri Pirko 已提交
1324
 * @IFF_TEAM: device is a team device
1325
 * @IFF_RXFH_CONFIGURED: device has had Rx Flow indirection table configured
1326 1327
 * @IFF_PHONY_HEADROOM: the headroom value is controlled by an external
 *	entity (i.e. the master device for bridged veth)
1328 1329 1330 1331
 */
enum netdev_priv_flags {
	IFF_802_1Q_VLAN			= 1<<0,
	IFF_EBRIDGE			= 1<<1,
1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349
	IFF_BONDING			= 1<<2,
	IFF_ISATAP			= 1<<3,
	IFF_WAN_HDLC			= 1<<4,
	IFF_XMIT_DST_RELEASE		= 1<<5,
	IFF_DONT_BRIDGE			= 1<<6,
	IFF_DISABLE_NETPOLL		= 1<<7,
	IFF_MACVLAN_PORT		= 1<<8,
	IFF_BRIDGE_PORT			= 1<<9,
	IFF_OVS_DATAPATH		= 1<<10,
	IFF_TX_SKB_SHARING		= 1<<11,
	IFF_UNICAST_FLT			= 1<<12,
	IFF_TEAM_PORT			= 1<<13,
	IFF_SUPP_NOFCS			= 1<<14,
	IFF_LIVE_ADDR_CHANGE		= 1<<15,
	IFF_MACVLAN			= 1<<16,
	IFF_XMIT_DST_RELEASE_PERM	= 1<<17,
	IFF_IPVLAN_MASTER		= 1<<18,
	IFF_IPVLAN_SLAVE		= 1<<19,
1350
	IFF_L3MDEV_MASTER		= 1<<20,
1351 1352
	IFF_NO_QUEUE			= 1<<21,
	IFF_OPENVSWITCH			= 1<<22,
D
David Ahern 已提交
1353
	IFF_L3MDEV_SLAVE		= 1<<23,
J
Jiri Pirko 已提交
1354
	IFF_TEAM			= 1<<24,
1355
	IFF_RXFH_CONFIGURED		= 1<<25,
1356
	IFF_PHONY_HEADROOM		= 1<<26,
1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375
};

#define IFF_802_1Q_VLAN			IFF_802_1Q_VLAN
#define IFF_EBRIDGE			IFF_EBRIDGE
#define IFF_BONDING			IFF_BONDING
#define IFF_ISATAP			IFF_ISATAP
#define IFF_WAN_HDLC			IFF_WAN_HDLC
#define IFF_XMIT_DST_RELEASE		IFF_XMIT_DST_RELEASE
#define IFF_DONT_BRIDGE			IFF_DONT_BRIDGE
#define IFF_DISABLE_NETPOLL		IFF_DISABLE_NETPOLL
#define IFF_MACVLAN_PORT		IFF_MACVLAN_PORT
#define IFF_BRIDGE_PORT			IFF_BRIDGE_PORT
#define IFF_OVS_DATAPATH		IFF_OVS_DATAPATH
#define IFF_TX_SKB_SHARING		IFF_TX_SKB_SHARING
#define IFF_UNICAST_FLT			IFF_UNICAST_FLT
#define IFF_TEAM_PORT			IFF_TEAM_PORT
#define IFF_SUPP_NOFCS			IFF_SUPP_NOFCS
#define IFF_LIVE_ADDR_CHANGE		IFF_LIVE_ADDR_CHANGE
#define IFF_MACVLAN			IFF_MACVLAN
1376
#define IFF_XMIT_DST_RELEASE_PERM	IFF_XMIT_DST_RELEASE_PERM
1377 1378
#define IFF_IPVLAN_MASTER		IFF_IPVLAN_MASTER
#define IFF_IPVLAN_SLAVE		IFF_IPVLAN_SLAVE
1379
#define IFF_L3MDEV_MASTER		IFF_L3MDEV_MASTER
1380
#define IFF_NO_QUEUE			IFF_NO_QUEUE
1381
#define IFF_OPENVSWITCH			IFF_OPENVSWITCH
1382
#define IFF_L3MDEV_SLAVE		IFF_L3MDEV_SLAVE
J
Jiri Pirko 已提交
1383
#define IFF_TEAM			IFF_TEAM
1384
#define IFF_RXFH_CONFIGURED		IFF_RXFH_CONFIGURED
1385

1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402
/**
 *	struct net_device - 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.
 *
 *	@name:	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
 *	 	of the interface.
 *
 *	@name_hlist: 	Device name hash chain, please keep it close to name[]
 *	@ifalias:	SNMP alias
 *	@mem_end:	Shared memory end
 *	@mem_start:	Shared memory start
 *	@base_addr:	Device I/O address
 *	@irq:		Device IRQ number
 *
1403 1404
 *	@carrier_changes:	Stats to monitor carrier on<->off transitions
 *
1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427
 *	@state:		Generic network queuing layer state, see netdev_state_t
 *	@dev_list:	The global list of network devices
 *	@napi_list:	List entry, that is used for polling napi devices
 *	@unreg_list:	List entry, that is used, when we are unregistering the
 *			device, see the function unregister_netdev
 *	@close_list:	List entry, that is used, when we are closing the device
 *
 *	@adj_list:	Directly linked devices, like slaves for bonding
 *	@all_adj_list:	All linked devices, *including* neighbours
 *	@features:	Currently active device features
 *	@hw_features:	User-changeable features
 *
 *	@wanted_features:	User-requested features
 *	@vlan_features:		Mask of features inheritable by VLAN devices
 *
 *	@hw_enc_features:	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.
 *
 *	@mpls_features:	Mask of features inheritable by MPLS
 *
 *	@ifindex:	interface index
1428
 *	@group:		The group, that the device belongs to
1429 1430 1431 1432 1433 1434 1435 1436
 *
 *	@stats:		Statistics struct, which was left as a legacy, use
 *			rtnl_link_stats64 instead
 *
 *	@rx_dropped:	Dropped packets by core network,
 *			do not use this in drivers
 *	@tx_dropped:	Dropped packets by core network,
 *			do not use this in drivers
1437 1438
 *	@rx_nohandler:	nohandler dropped packets by core network on
 *			inactive devices, do not use this in drivers
1439 1440 1441 1442 1443 1444 1445 1446 1447
 *
 *	@wireless_handlers:	List of functions to handle Wireless Extensions,
 *				instead of ioctl,
 *				see <net/iw_handler.h> for details.
 *	@wireless_data:	Instance data managed by the core of wireless extensions
 *
 *	@netdev_ops:	Includes several pointers to callbacks,
 *			if one wants to override the ndo_*() functions
 *	@ethtool_ops:	Management operations
E
Eric W. Biederman 已提交
1448
 *	@header_ops:	Includes callbacks for creating,parsing,caching,etc
1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461
 *			of Layer 2 headers.
 *
 *	@flags:		Interface flags (a la BSD)
 *	@priv_flags:	Like 'flags' but invisible to userspace,
 *			see if.h for the definitions
 *	@gflags:	Global flags ( kept as legacy )
 *	@padded:	How much padding added by alloc_netdev()
 *	@operstate:	RFC2863 operstate
 *	@link_mode:	Mapping policy to operstate
 *	@if_port:	Selectable AUI, TP, ...
 *	@dma:		DMA channel
 *	@mtu:		Interface MTU value
 *	@type:		Interface hardware type
1462 1463
 *	@hard_header_len: Hardware header length, which means that this is the
 *			  minimum size of a packet.
1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486
 *
 *	@needed_headroom: Extra headroom the hardware may need, but not in all
 *			  cases can this be guaranteed
 *	@needed_tailroom: Extra tailroom the hardware may need, but not in all
 *			  cases can this be guaranteed. Some cases also use
 *			  LL_MAX_HEADER instead to allocate the skb
 *
 *	interface address info:
 *
 * 	@perm_addr:		Permanent hw address
 * 	@addr_assign_type:	Hw address assignment type
 * 	@addr_len:		Hardware address length
 * 	@neigh_priv_len;	Used in neigh_alloc(),
 * 				initialized only in atm/clip.c
 * 	@dev_id:		Used to differentiate devices that share
 * 				the same link layer address
 * 	@dev_port:		Used to differentiate devices that share
 * 				the same function
 *	@addr_list_lock:	XXX: need comments on this one
 *	@uc_promisc:		Counter, that indicates, that promiscuous mode
 *				has been enabled due to the need to listen to
 *				additional unicast addresses in a device that
 *				does not implement ndo_set_rx_mode()
1487 1488 1489 1490
 *	@uc:			unicast mac addresses
 *	@mc:			multicast mac addresses
 *	@dev_addrs:		list of device hw addresses
 *	@queues_kset:		Group of all Kobjects in the Tx and RX queues
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
 *	@promiscuity:		Number of times, the NIC is told to work in
 *				Promiscuous mode, if it becomes 0 the NIC will
 *				exit from working in Promiscuous mode
 *	@allmulti:		Counter, enables or disables allmulticast mode
 *
 *	@vlan_info:	VLAN info
 *	@dsa_ptr:	dsa specific data
 *	@tipc_ptr:	TIPC specific data
 *	@atalk_ptr:	AppleTalk link
 *	@ip_ptr:	IPv4 specific data
 *	@dn_ptr:	DECnet specific data
 *	@ip6_ptr:	IPv6 specific data
 *	@ax25_ptr:	AX.25 specific data
 *	@ieee80211_ptr:	IEEE 802.11 specific data, assign before registering
 *
 *	@last_rx:	Time of last Rx
 *	@dev_addr:	Hw address (before bcast,
 *			because most packets are unicast)
 *
 *	@_rx:			Array of RX queues
 *	@num_rx_queues:		Number of RX queues
 *				allocated at register_netdev() time
 *	@real_num_rx_queues: 	Number of RX queues currently active in device
 *
 *	@rx_handler:		handler for received packets
 *	@rx_handler_data: 	XXX: need comments on this one
 *	@ingress_queue:		XXX: need comments on this one
 *	@broadcast:		hw bcast address
 *
1520 1521 1522 1523 1524 1525
 *	@rx_cpu_rmap:	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
 *	@index_hlist:		Device index hash chain
 *
1526 1527 1528 1529 1530 1531 1532 1533 1534
 *	@_tx:			Array of TX queues
 *	@num_tx_queues:		Number of TX queues allocated at alloc_netdev_mq() time
 *	@real_num_tx_queues: 	Number of TX queues currently active in device
 *	@qdisc:			Root qdisc from userspace point of view
 *	@tx_queue_len:		Max frames per queue allowed
 *	@tx_global_lock: 	XXX: need comments on this one
 *
 *	@xps_maps:	XXX: need comments on this one
 *
1535 1536
 *	@offload_fwd_mark:	Offload device fwding mark
 *
1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574
 *	@trans_start:		Time (in jiffies) of last Tx
 *	@watchdog_timeo:	Represents the timeout that is used by
 *				the watchdog ( see dev_watchdog() )
 *	@watchdog_timer:	List of timers
 *
 *	@pcpu_refcnt:		Number of references to this device
 *	@todo_list:		Delayed register/unregister
 *	@link_watch_list:	XXX: need comments on this one
 *
 *	@reg_state:		Register/unregister state machine
 *	@dismantle:		Device is going to be freed
 *	@rtnl_link_state:	This enum represents the phases of creating
 *				a new link
 *
 *	@destructor:		Called from unregister,
 *				can be used to call free_netdev
 *	@npinfo:		XXX: need comments on this one
 * 	@nd_net:		Network namespace this network device is inside
 *
 * 	@ml_priv:	Mid-layer private
 * 	@lstats:	Loopback statistics
 * 	@tstats:	Tunnel statistics
 * 	@dstats:	Dummy statistics
 * 	@vstats:	Virtual ethernet statistics
 *
 *	@garp_port:	GARP
 *	@mrp_port:	MRP
 *
 *	@dev:		Class/net/name entry
 *	@sysfs_groups:	Space for optional device, statistics and wireless
 *			sysfs groups
 *
 *	@sysfs_rx_queue_group:	Space for optional per-rx queue attributes
 *	@rtnl_link_ops:	Rtnl_link_ops
 *
 *	@gso_max_size:	Maximum size of generic segmentation offload
 *	@gso_max_segs:	Maximum number of segments that can be passed to the
 *			NIC for GSO
1575 1576
 *	@gso_min_segs:	Minimum number of segments that can be passed to the
 *			NIC for GSO
1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590
 *
 *	@dcbnl_ops:	Data Center Bridging netlink ops
 *	@num_tc:	Number of traffic classes in the net device
 *	@tc_to_txq:	XXX: need comments on this one
 *	@prio_tc_map	XXX: need comments on this one
 *
 *	@fcoe_ddp_xid:	Max exchange id for FCoE LRO by ddp
 *
 *	@priomap:	XXX: need comments on this one
 *	@phydev:	Physical device may attach itself
 *			for hardware timestamping
 *
 *	@qdisc_tx_busylock:	XXX: need comments on this one
 *
1591 1592 1593 1594
 *	@proto_down:	protocol port state information can be sent to the
 *			switch driver and used to set the phys state of the
 *			switch port.
 *
L
Linus Torvalds 已提交
1595 1596 1597 1598
 *	FIXME: cleanup struct net_device such that network protocol info
 *	moves out.
 */

E
Eric Dumazet 已提交
1599
struct net_device {
L
Linus Torvalds 已提交
1600
	char			name[IFNAMSIZ];
1601
	struct hlist_node	name_hlist;
1602
	char 			*ifalias;
L
Linus Torvalds 已提交
1603 1604 1605 1606
	/*
	 *	I/O specific fields
	 *	FIXME: Merge these and struct ifmap into one
	 */
1607 1608 1609 1610
	unsigned long		mem_end;
	unsigned long		mem_start;
	unsigned long		base_addr;
	int			irq;
L
Linus Torvalds 已提交
1611

1612 1613
	atomic_t		carrier_changes;

L
Linus Torvalds 已提交
1614
	/*
1615 1616
	 *	Some hardware also needs these fields (state,dev_list,
	 *	napi_list,unreg_list,close_list) but they are not
L
Linus Torvalds 已提交
1617 1618 1619 1620 1621
	 *	part of the usual set specified in Space.c.
	 */

	unsigned long		state;

1622
	struct list_head	dev_list;
1623
	struct list_head	napi_list;
1624
	struct list_head	unreg_list;
1625
	struct list_head	close_list;
1626 1627
	struct list_head	ptype_all;
	struct list_head	ptype_specific;
1628 1629 1630 1631 1632 1633 1634 1635 1636 1637

	struct {
		struct list_head upper;
		struct list_head lower;
	} adj_list;

	struct {
		struct list_head upper;
		struct list_head lower;
	} all_adj_list;
1638

1639 1640 1641 1642
	netdev_features_t	features;
	netdev_features_t	hw_features;
	netdev_features_t	wanted_features;
	netdev_features_t	vlan_features;
1643
	netdev_features_t	hw_enc_features;
S
Simon Horman 已提交
1644
	netdev_features_t	mpls_features;
1645

L
Linus Torvalds 已提交
1646
	int			ifindex;
1647
	int			group;
L
Linus Torvalds 已提交
1648

R
Rusty Russell 已提交
1649
	struct net_device_stats	stats;
1650 1651 1652

	atomic_long_t		rx_dropped;
	atomic_long_t		tx_dropped;
1653
	atomic_long_t		rx_nohandler;
L
Linus Torvalds 已提交
1654

1655
#ifdef CONFIG_WIRELESS_EXT
L
Linus Torvalds 已提交
1656 1657
	const struct iw_handler_def *	wireless_handlers;
	struct iw_public_data *	wireless_data;
1658
#endif
1659
	const struct net_device_ops *netdev_ops;
1660
	const struct ethtool_ops *ethtool_ops;
S
Scott Feldman 已提交
1661
#ifdef CONFIG_NET_SWITCHDEV
J
Jiri Pirko 已提交
1662
	const struct switchdev_ops *switchdev_ops;
S
Scott Feldman 已提交
1663
#endif
1664 1665 1666
#ifdef CONFIG_NET_L3_MASTER_DEV
	const struct l3mdev_ops	*l3mdev_ops;
#endif
L
Linus Torvalds 已提交
1667

1668 1669
	const struct header_ops *header_ops;

1670 1671 1672
	unsigned int		flags;
	unsigned int		priv_flags;

L
Linus Torvalds 已提交
1673
	unsigned short		gflags;
1674
	unsigned short		padded;
L
Linus Torvalds 已提交
1675

1676 1677
	unsigned char		operstate;
	unsigned char		link_mode;
S
Stefan Rompf 已提交
1678

1679 1680
	unsigned char		if_port;
	unsigned char		dma;
1681

1682 1683 1684
	unsigned int		mtu;
	unsigned short		type;
	unsigned short		hard_header_len;
L
Linus Torvalds 已提交
1685

1686 1687 1688
	unsigned short		needed_headroom;
	unsigned short		needed_tailroom;

L
Linus Torvalds 已提交
1689
	/* Interface address info. */
1690 1691 1692
	unsigned char		perm_addr[MAX_ADDR_LEN];
	unsigned char		addr_assign_type;
	unsigned char		addr_len;
1693
	unsigned short		neigh_priv_len;
1694 1695
	unsigned short          dev_id;
	unsigned short          dev_port;
J
Jiri Pirko 已提交
1696
	spinlock_t		addr_list_lock;
1697 1698
	unsigned char		name_assign_type;
	bool			uc_promisc;
1699 1700 1701 1702
	struct netdev_hw_addr_list	uc;
	struct netdev_hw_addr_list	mc;
	struct netdev_hw_addr_list	dev_addrs;

1703 1704 1705
#ifdef CONFIG_SYSFS
	struct kset		*queues_kset;
#endif
1706 1707
	unsigned int		promiscuity;
	unsigned int		allmulti;
L
Linus Torvalds 已提交
1708 1709 1710


	/* Protocol specific pointers */
1711

1712
#if IS_ENABLED(CONFIG_VLAN_8021Q)
1713
	struct vlan_info __rcu	*vlan_info;
1714
#endif
1715
#if IS_ENABLED(CONFIG_NET_DSA)
1716
	struct dsa_switch_tree	*dsa_ptr;
1717 1718
#endif
#if IS_ENABLED(CONFIG_TIPC)
1719
	struct tipc_bearer __rcu *tipc_ptr;
1720
#endif
1721 1722 1723 1724 1725 1726
	void 			*atalk_ptr;
	struct in_device __rcu	*ip_ptr;
	struct dn_dev __rcu     *dn_ptr;
	struct inet6_dev __rcu	*ip6_ptr;
	void			*ax25_ptr;
	struct wireless_dev	*ieee80211_ptr;
1727
	struct wpan_dev		*ieee802154_ptr;
R
Robert Shearman 已提交
1728 1729 1730
#if IS_ENABLED(CONFIG_MPLS_ROUTING)
	struct mpls_dev __rcu	*mpls_ptr;
#endif
L
Linus Torvalds 已提交
1731

1732
/*
E
Eric Dumazet 已提交
1733
 * Cache lines mostly used on receive path (including eth_type_trans())
1734
 */
1735
	unsigned long		last_rx;
1736

1737
	/* Interface address info used in eth_type_trans() */
1738
	unsigned char		*dev_addr;
1739

T
Tom Herbert 已提交
1740

1741
#ifdef CONFIG_SYSFS
T
Tom Herbert 已提交
1742 1743 1744
	struct netdev_rx_queue	*_rx;

	unsigned int		num_rx_queues;
1745
	unsigned int		real_num_rx_queues;
1746

E
Eric Dumazet 已提交
1747
#endif
T
Tom Herbert 已提交
1748

1749
	unsigned long		gro_flush_timeout;
1750 1751
	rx_handler_func_t __rcu	*rx_handler;
	void __rcu		*rx_handler_data;
1752

1753
#ifdef CONFIG_NET_CLS_ACT
1754 1755
	struct tcf_proto __rcu  *ingress_cl_list;
#endif
1756
	struct netdev_queue __rcu *ingress_queue;
1757 1758 1759
#ifdef CONFIG_NETFILTER_INGRESS
	struct list_head	nf_hooks_ingress;
#endif
1760

1761
	unsigned char		broadcast[MAX_ADDR_LEN];
1762 1763 1764 1765
#ifdef CONFIG_RFS_ACCEL
	struct cpu_rmap		*rx_cpu_rmap;
#endif
	struct hlist_node	index_hlist;
E
Eric Dumazet 已提交
1766 1767 1768 1769

/*
 * Cache lines mostly used on transmit path
 */
1770 1771
	struct netdev_queue	*_tx ____cacheline_aligned_in_smp;
	unsigned int		num_tx_queues;
1772
	unsigned int		real_num_tx_queues;
1773
	struct Qdisc		*qdisc;
1774
	unsigned long		tx_queue_len;
1775
	spinlock_t		tx_global_lock;
1776
	int			watchdog_timeo;
E
Eric Dumazet 已提交
1777

T
Tom Herbert 已提交
1778
#ifdef CONFIG_XPS
E
Eric Dumazet 已提交
1779
	struct xps_dev_maps __rcu *xps_maps;
T
Tom Herbert 已提交
1780
#endif
D
Daniel Borkmann 已提交
1781 1782 1783
#ifdef CONFIG_NET_CLS_ACT
	struct tcf_proto __rcu  *egress_cl_list;
#endif
1784 1785 1786 1787
#ifdef CONFIG_NET_SWITCHDEV
	u32			offload_fwd_mark;
#endif

1788
	/* These may be needed for future network-power-down code. */
1789 1790 1791 1792 1793

	/*
	 * trans_start here is expensive for high speed devices on SMP,
	 * please use netdev_queue->trans_start instead.
	 */
1794
	unsigned long		trans_start;
1795 1796 1797

	struct timer_list	watchdog_timer;

E
Eric Dumazet 已提交
1798
	int __percpu		*pcpu_refcnt;
L
Linus Torvalds 已提交
1799 1800
	struct list_head	todo_list;

1801
	struct list_head	link_watch_list;
1802

L
Linus Torvalds 已提交
1803
	enum { NETREG_UNINITIALIZED=0,
1804
	       NETREG_REGISTERED,	/* completed register_netdevice */
L
Linus Torvalds 已提交
1805 1806 1807
	       NETREG_UNREGISTERING,	/* called unregister_netdevice */
	       NETREG_UNREGISTERED,	/* completed unregister todo */
	       NETREG_RELEASED,		/* called free_netdev */
1808
	       NETREG_DUMMY,		/* dummy device for NAPI poll */
1809 1810
	} reg_state:8;

1811
	bool dismantle;
1812 1813 1814 1815 1816

	enum {
		RTNL_LINK_INITIALIZED,
		RTNL_LINK_INITIALIZING,
	} rtnl_link_state:16;
L
Linus Torvalds 已提交
1817

1818
	void (*destructor)(struct net_device *dev);
L
Linus Torvalds 已提交
1819 1820

#ifdef CONFIG_NETPOLL
1821
	struct netpoll_info __rcu	*npinfo;
L
Linus Torvalds 已提交
1822
#endif
1823

1824
	possible_net_t			nd_net;
1825

D
David S. Miller 已提交
1826
	/* mid-layer private */
E
Eric Dumazet 已提交
1827
	union {
1828 1829
		void					*ml_priv;
		struct pcpu_lstats __percpu		*lstats;
1830
		struct pcpu_sw_netstats __percpu	*tstats;
1831 1832
		struct pcpu_dstats __percpu		*dstats;
		struct pcpu_vstats __percpu		*vstats;
E
Eric Dumazet 已提交
1833
	};
1834

E
Eric Dumazet 已提交
1835
	struct garp_port __rcu	*garp_port;
1836
	struct mrp_port __rcu	*mrp_port;
L
Linus Torvalds 已提交
1837

1838
	struct device	dev;
1839
	const struct attribute_group *sysfs_groups[4];
1840
	const struct attribute_group *sysfs_rx_queue_group;
P
Patrick McHardy 已提交
1841 1842

	const struct rtnl_link_ops *rtnl_link_ops;
1843

1844 1845 1846
	/* for setting kernel sock attribute on TCP connection setup */
#define GSO_MAX_SIZE		65536
	unsigned int		gso_max_size;
1847 1848
#define GSO_MAX_SEGS		65535
	u16			gso_max_segs;
1849
	u16			gso_min_segs;
J
Jeff Kirsher 已提交
1850
#ifdef CONFIG_DCB
1851
	const struct dcbnl_rtnl_ops *dcbnl_ops;
1852
#endif
1853 1854 1855
	u8 num_tc;
	struct netdev_tc_txq tc_to_txq[TC_MAX_QUEUE];
	u8 prio_tc_map[TC_BITMASK + 1];
1856

1857
#if IS_ENABLED(CONFIG_FCOE)
1858
	unsigned int		fcoe_ddp_xid;
1859
#endif
1860
#if IS_ENABLED(CONFIG_CGROUP_NET_PRIO)
1861
	struct netprio_map __rcu *priomap;
1862
#endif
1863
	struct phy_device *phydev;
1864
	struct lock_class_key *qdisc_tx_busylock;
1865
	bool proto_down;
L
Linus Torvalds 已提交
1866
};
1867
#define to_net_dev(d) container_of(d, struct net_device, dev)
L
Linus Torvalds 已提交
1868 1869 1870

#define	NETDEV_ALIGN		32

1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921
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;
}

1922 1923 1924 1925 1926 1927 1928
static inline
struct netdev_queue *netdev_get_tx_queue(const struct net_device *dev,
					 unsigned int index)
{
	return &dev->_tx[index];
}

1929 1930 1931 1932 1933 1934
static inline struct netdev_queue *skb_get_tx_queue(const struct net_device *dev,
						    const struct sk_buff *skb)
{
	return netdev_get_tx_queue(dev, skb_get_queue_mapping(skb));
}

1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946
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);
}

1947
struct netdev_queue *netdev_pick_tx(struct net_device *dev,
1948 1949
				    struct sk_buff *skb,
				    void *accel_priv);
1950

1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970
/* returns the headroom that the master device needs to take in account
 * when forwarding to this dev
 */
static inline unsigned netdev_get_fwd_headroom(struct net_device *dev)
{
	return dev->priv_flags & IFF_PHONY_HEADROOM ? 0 : dev->needed_headroom;
}

static inline void netdev_set_rx_headroom(struct net_device *dev, int new_hr)
{
	if (dev->netdev_ops->ndo_set_rx_headroom)
		dev->netdev_ops->ndo_set_rx_headroom(dev, new_hr);
}

/* set the device rx headroom to the dev's default */
static inline void netdev_reset_rx_headroom(struct net_device *dev)
{
	netdev_set_rx_headroom(dev, -1);
}

1971 1972 1973 1974 1975 1976
/*
 * Net namespace inlines
 */
static inline
struct net *dev_net(const struct net_device *dev)
{
E
Eric Dumazet 已提交
1977
	return read_pnet(&dev->nd_net);
1978 1979 1980
}

static inline
1981
void dev_net_set(struct net_device *dev, struct net *net)
1982
{
1983
	write_pnet(&dev->nd_net, net);
1984 1985
}

1986
static inline bool netdev_uses_dsa(struct net_device *dev)
1987
{
1988
#if IS_ENABLED(CONFIG_NET_DSA)
1989 1990
	if (dev->dsa_ptr != NULL)
		return dsa_uses_tagged_protocol(dev->dsa_ptr);
1991
#endif
1992
	return false;
1993 1994
}

1995 1996 1997 1998 1999 2000
/**
 *	netdev_priv - access network device private data
 *	@dev: network device
 *
 * Get network device private data
 */
2001
static inline void *netdev_priv(const struct net_device *dev)
L
Linus Torvalds 已提交
2002
{
2003
	return (char *)dev + ALIGN(sizeof(struct net_device), NETDEV_ALIGN);
L
Linus Torvalds 已提交
2004 2005 2006 2007 2008
}

/* Set the sysfs physical device reference for the network logical device
 * if set prior to registration will cause a symlink during initialization.
 */
2009
#define SET_NETDEV_DEV(net, pdev)	((net)->dev.parent = (pdev))
L
Linus Torvalds 已提交
2010

2011
/* Set the sysfs device type for the network logical device to allow
2012
 * fine-grained identification of different network device types. For
2013 2014 2015 2016
 * example Ethernet, Wirelss LAN, Bluetooth, WiMAX etc.
 */
#define SET_NETDEV_DEVTYPE(net, devtype)	((net)->dev.type = (devtype))

E
Eric Dumazet 已提交
2017 2018 2019 2020 2021
/* Default NAPI poll() weight
 * Device drivers are strongly advised to not use bigger value
 */
#define NAPI_POLL_WEIGHT 64

2022 2023 2024 2025 2026 2027 2028 2029 2030 2031
/**
 *	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.
 */
2032 2033
void netif_napi_add(struct net_device *dev, struct napi_struct *napi,
		    int (*poll)(struct napi_struct *, int), int weight);
2034

E
Eric Dumazet 已提交
2035 2036 2037 2038 2039 2040 2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 2051 2052 2053 2054
/**
 *	netif_tx_napi_add - initialize a napi context
 *	@dev:  network device
 *	@napi: napi context
 *	@poll: polling function
 *	@weight: default weight
 *
 * This variant of netif_napi_add() should be used from drivers using NAPI
 * to exclusively poll a TX queue.
 * This will avoid we add it into napi_hash[], thus polluting this hash table.
 */
static inline void netif_tx_napi_add(struct net_device *dev,
				     struct napi_struct *napi,
				     int (*poll)(struct napi_struct *, int),
				     int weight)
{
	set_bit(NAPI_STATE_NO_BUSY_POLL, &napi->state);
	netif_napi_add(dev, napi, poll, weight);
}

2055 2056 2057 2058 2059 2060
/**
 *  netif_napi_del - remove a napi context
 *  @napi: napi context
 *
 *  netif_napi_del() removes a napi context from the network device napi list
 */
2061 2062 2063
void netif_napi_del(struct napi_struct *napi);

struct napi_gro_cb {
2064 2065 2066
	/* Virtual address of skb_shinfo(skb)->frags[0].page + offset. */
	void *frag0;

2067 2068 2069
	/* Length of frag0. */
	unsigned int frag0_len;

2070 2071 2072
	/* This indicates where we are processing relative to skb->data. */
	int data_offset;

2073
	/* This is non-zero if the packet cannot be merged with the new skb. */
2074 2075 2076 2077
	u16	flush;

	/* Save the IP ID here and check when we get to the transport layer */
	u16	flush_id;
2078 2079

	/* Number of segments aggregated. */
2080 2081
	u16	count;

2082 2083 2084
	/* Start offset for remote checksum offload */
	u16	gro_remcsum_start;

2085 2086
	/* jiffies when first packet was created/queued */
	unsigned long age;
2087

T
Tom Herbert 已提交
2088
	/* Used in ipv6_gro_receive() and foo-over-udp */
2089 2090
	u16	proto;

2091 2092 2093
	/* This is non-zero if the packet may be of the same flow. */
	u8	same_flow:1;

2094
	/* Used in udp_gro_receive */
T
Tom Herbert 已提交
2095 2096 2097 2098 2099
	u8	udp_mark:1;

	/* GRO checksum is valid */
	u8	csum_valid:1;

2100 2101
	/* Number of checksums via CHECKSUM_UNNECESSARY */
	u8	csum_cnt:3;
2102

2103 2104 2105 2106 2107
	/* Free the skb? */
	u8	free:2;
#define NAPI_GRO_FREE		  1
#define NAPI_GRO_FREE_STOLEN_HEAD 2

2108 2109 2110
	/* Used in foo-over-udp, set in udp[46]_gro_receive */
	u8	is_ipv6:1;

2111 2112
	/* 7 bit hole */

2113 2114 2115
	/* used to support CHECKSUM_COMPLETE for tunneling protocols */
	__wsum	csum;

2116 2117
	/* used in skb_gro_receive() slow path */
	struct sk_buff *last;
2118 2119 2120
};

#define NAPI_GRO_CB(skb) ((struct napi_gro_cb *)(skb)->cb)
2121

L
Linus Torvalds 已提交
2122
struct packet_type {
D
David S. Miller 已提交
2123 2124 2125 2126 2127 2128
	__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 *);
2129 2130
	bool			(*id_match)(struct packet_type *ptype,
					    struct sock *sk);
L
Linus Torvalds 已提交
2131 2132 2133 2134
	void			*af_packet_priv;
	struct list_head	list;
};

2135
struct offload_callbacks {
2136
	struct sk_buff		*(*gso_segment)(struct sk_buff *skb,
2137
						netdev_features_t features);
2138
	struct sk_buff		**(*gro_receive)(struct sk_buff **head,
2139
						 struct sk_buff *skb);
2140
	int			(*gro_complete)(struct sk_buff *skb, int nhoff);
2141 2142 2143 2144
};

struct packet_offload {
	__be16			 type;	/* This is really htons(ether_type). */
2145
	u16			 priority;
2146 2147
	struct offload_callbacks callbacks;
	struct list_head	 list;
L
Linus Torvalds 已提交
2148 2149
};

2150 2151 2152 2153 2154 2155 2156 2157 2158 2159 2160
struct udp_offload;

struct udp_offload_callbacks {
	struct sk_buff		**(*gro_receive)(struct sk_buff **head,
						 struct sk_buff *skb,
						 struct udp_offload *uoff);
	int			(*gro_complete)(struct sk_buff *skb,
						int nhoff,
						struct udp_offload *uoff);
};

2161 2162
struct udp_offload {
	__be16			 port;
T
Tom Herbert 已提交
2163
	u8			 ipproto;
2164
	struct udp_offload_callbacks callbacks;
2165 2166
};

2167 2168 2169 2170 2171 2172 2173 2174 2175
/* 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;
};

2176 2177 2178 2179 2180 2181 2182 2183 2184 2185 2186 2187
#define __netdev_alloc_pcpu_stats(type, gfp)				\
({									\
	typeof(type) __percpu *pcpu_stats = alloc_percpu_gfp(type, gfp);\
	if (pcpu_stats)	{						\
		int __cpu;						\
		for_each_possible_cpu(__cpu) {				\
			typeof(type) *stat;				\
			stat = per_cpu_ptr(pcpu_stats, __cpu);		\
			u64_stats_init(&stat->syncp);			\
		}							\
	}								\
	pcpu_stats;							\
2188 2189
})

2190
#define netdev_alloc_pcpu_stats(type)					\
2191
	__netdev_alloc_pcpu_stats(type, GFP_KERNEL)
2192

2193 2194 2195 2196 2197 2198 2199 2200 2201 2202 2203 2204 2205
enum netdev_lag_tx_type {
	NETDEV_LAG_TX_TYPE_UNKNOWN,
	NETDEV_LAG_TX_TYPE_RANDOM,
	NETDEV_LAG_TX_TYPE_BROADCAST,
	NETDEV_LAG_TX_TYPE_ROUNDROBIN,
	NETDEV_LAG_TX_TYPE_ACTIVEBACKUP,
	NETDEV_LAG_TX_TYPE_HASH,
};

struct netdev_lag_upper_info {
	enum netdev_lag_tx_type tx_type;
};

2206 2207 2208 2209 2210
struct netdev_lag_lower_state_info {
	u8 link_up : 1,
	   tx_enabled : 1;
};

L
Linus Torvalds 已提交
2211 2212
#include <linux/notifier.h>

2213 2214 2215 2216 2217 2218 2219 2220 2221 2222 2223 2224 2225
/* 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
2226
#define NETDEV_CHANGEMTU	0x0007 /* notify after mtu change happened */
2227 2228 2229 2230 2231 2232 2233 2234 2235
#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
2236
#define NETDEV_UNREGISTER_FINAL 0x0011
2237 2238 2239
#define NETDEV_RELEASE		0x0012
#define NETDEV_NOTIFY_PEERS	0x0013
#define NETDEV_JOIN		0x0014
2240
#define NETDEV_CHANGEUPPER	0x0015
2241
#define NETDEV_RESEND_IGMP	0x0016
2242
#define NETDEV_PRECHANGEMTU	0x0017 /* notify before mtu change happened */
2243
#define NETDEV_CHANGEINFODATA	0x0018
2244
#define NETDEV_BONDING_INFO	0x0019
2245
#define NETDEV_PRECHANGEUPPER	0x001A
2246
#define NETDEV_CHANGELOWERSTATE	0x001B
2247

2248 2249
int register_netdevice_notifier(struct notifier_block *nb);
int unregister_netdevice_notifier(struct notifier_block *nb);
2250 2251 2252 2253 2254

struct netdev_notifier_info {
	struct net_device *dev;
};

2255 2256 2257 2258 2259
struct netdev_notifier_change_info {
	struct netdev_notifier_info info; /* must be first */
	unsigned int flags_changed;
};

2260 2261 2262 2263 2264
struct netdev_notifier_changeupper_info {
	struct netdev_notifier_info info; /* must be first */
	struct net_device *upper_dev; /* new upper dev */
	bool master; /* is upper dev master */
	bool linking; /* is the nofication for link or unlink */
2265
	void *upper_info; /* upper dev info */
2266 2267
};

2268 2269 2270 2271 2272
struct netdev_notifier_changelowerstate_info {
	struct netdev_notifier_info info; /* must be first */
	void *lower_state_info; /* is lower dev state */
};

2273 2274 2275 2276 2277 2278
static inline void netdev_notifier_info_init(struct netdev_notifier_info *info,
					     struct net_device *dev)
{
	info->dev = dev;
}

2279 2280 2281 2282 2283 2284
static inline struct net_device *
netdev_notifier_info_to_dev(const struct netdev_notifier_info *info)
{
	return info->dev;
}

2285
int call_netdevice_notifiers(unsigned long val, struct net_device *dev);
2286 2287


L
Linus Torvalds 已提交
2288 2289
extern rwlock_t				dev_base_lock;		/* Device list lock */

2290 2291
#define for_each_netdev(net, d)		\
		list_for_each_entry(d, &(net)->dev_base_head, dev_list)
2292 2293
#define for_each_netdev_reverse(net, d)	\
		list_for_each_entry_reverse(d, &(net)->dev_base_head, dev_list)
2294 2295
#define for_each_netdev_rcu(net, d)		\
		list_for_each_entry_rcu(d, &(net)->dev_base_head, dev_list)
2296 2297 2298 2299
#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)
2300 2301
#define for_each_netdev_continue_rcu(net, d)		\
	list_for_each_entry_continue_rcu(d, &(net)->dev_base_head, dev_list)
2302 2303
#define for_each_netdev_in_bond_rcu(bond, slave)	\
		for_each_netdev_rcu(&init_net, slave)	\
2304
			if (netdev_master_upper_dev_get_rcu(slave) == (bond))
2305
#define net_device_entry(lh)	list_entry(lh, struct net_device, dev_list)
2306

2307 2308 2309 2310 2311
static inline struct net_device *next_net_device(struct net_device *dev)
{
	struct list_head *lh;
	struct net *net;

2312
	net = dev_net(dev);
2313 2314 2315 2316
	lh = dev->dev_list.next;
	return lh == &net->dev_base_head ? NULL : net_device_entry(lh);
}

2317 2318 2319 2320 2321 2322
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
2323
	lh = rcu_dereference(list_next_rcu(&dev->dev_list));
2324 2325 2326
	return lh == &net->dev_base_head ? NULL : net_device_entry(lh);
}

2327 2328 2329 2330 2331
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);
}
2332

E
Eric Dumazet 已提交
2333 2334 2335 2336 2337 2338 2339
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);
}

2340 2341 2342 2343 2344 2345 2346 2347 2348 2349 2350 2351
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);

2352
int dev_get_iflink(const struct net_device *dev);
2353
int dev_fill_metadata_dst(struct net_device *dev, struct sk_buff *skb);
2354 2355
struct net_device *__dev_get_by_flags(struct net *net, unsigned short flags,
				      unsigned short mask);
2356 2357 2358 2359 2360 2361
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);
2362
int dev_close_many(struct list_head *head, bool unlink);
2363
void dev_disable_lro(struct net_device *dev);
2364
int dev_loopback_xmit(struct net *net, struct sock *sk, struct sk_buff *newskb);
2365
int dev_queue_xmit(struct sk_buff *skb);
2366
int dev_queue_xmit_accel(struct sk_buff *skb, void *accel_priv);
2367 2368 2369
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);
2370 2371 2372 2373 2374
static inline void unregister_netdevice(struct net_device *dev)
{
	unregister_netdevice_queue(dev, NULL);
}

2375 2376
int netdev_refcnt_read(const struct net_device *dev);
void free_netdev(struct net_device *dev);
2377
void netdev_freemem(struct net_device *dev);
2378 2379
void synchronize_net(void);
int init_dummy_netdev(struct net_device *dev);
2380

2381 2382 2383 2384 2385 2386
DECLARE_PER_CPU(int, xmit_recursion);
static inline int dev_recursion_level(void)
{
	return this_cpu_read(xmit_recursion);
}

2387 2388 2389 2390 2391 2392
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);
int skb_gro_receive(struct sk_buff **head, struct sk_buff *skb);
2393 2394 2395 2396 2397 2398 2399 2400 2401 2402 2403 2404 2405 2406 2407 2408

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;
}

2409 2410
static inline void *skb_gro_header_fast(struct sk_buff *skb,
					unsigned int offset)
2411
{
2412 2413
	return NAPI_GRO_CB(skb)->frag0 + offset;
}
2414

2415 2416 2417 2418
static inline int skb_gro_header_hard(struct sk_buff *skb, unsigned int hlen)
{
	return NAPI_GRO_CB(skb)->frag0_len < hlen;
}
2419

2420 2421 2422
static inline void *skb_gro_header_slow(struct sk_buff *skb, unsigned int hlen,
					unsigned int offset)
{
2423 2424 2425
	if (!pskb_may_pull(skb, hlen))
		return NULL;

2426 2427
	NAPI_GRO_CB(skb)->frag0 = NULL;
	NAPI_GRO_CB(skb)->frag0_len = 0;
2428
	return skb->data + offset;
2429
}
L
Linus Torvalds 已提交
2430

H
Herbert Xu 已提交
2431 2432
static inline void *skb_gro_network_header(struct sk_buff *skb)
{
2433 2434
	return (NAPI_GRO_CB(skb)->frag0 ?: skb->data) +
	       skb_network_offset(skb);
H
Herbert Xu 已提交
2435 2436
}

2437 2438 2439
static inline void skb_gro_postpull_rcsum(struct sk_buff *skb,
					const void *start, unsigned int len)
{
T
Tom Herbert 已提交
2440
	if (NAPI_GRO_CB(skb)->csum_valid)
2441 2442 2443 2444
		NAPI_GRO_CB(skb)->csum = csum_sub(NAPI_GRO_CB(skb)->csum,
						  csum_partial(start, len, 0));
}

T
Tom Herbert 已提交
2445 2446 2447 2448 2449 2450 2451
/* GRO checksum functions. These are logical equivalents of the normal
 * checksum functions (in skbuff.h) except that they operate on the GRO
 * offsets and fields in sk_buff.
 */

__sum16 __skb_gro_checksum_complete(struct sk_buff *skb);

2452 2453
static inline bool skb_at_gro_remcsum_start(struct sk_buff *skb)
{
2454
	return (NAPI_GRO_CB(skb)->gro_remcsum_start == skb_gro_offset(skb));
2455 2456
}

T
Tom Herbert 已提交
2457 2458 2459 2460
static inline bool __skb_gro_checksum_validate_needed(struct sk_buff *skb,
						      bool zero_okay,
						      __sum16 check)
{
2461 2462 2463
	return ((skb->ip_summed != CHECKSUM_PARTIAL ||
		skb_checksum_start_offset(skb) <
		 skb_gro_offset(skb)) &&
2464
		!skb_at_gro_remcsum_start(skb) &&
2465
		NAPI_GRO_CB(skb)->csum_cnt == 0 &&
T
Tom Herbert 已提交
2466 2467 2468 2469 2470 2471 2472 2473 2474 2475 2476 2477 2478 2479 2480 2481 2482
		(!zero_okay || check));
}

static inline __sum16 __skb_gro_checksum_validate_complete(struct sk_buff *skb,
							   __wsum psum)
{
	if (NAPI_GRO_CB(skb)->csum_valid &&
	    !csum_fold(csum_add(psum, NAPI_GRO_CB(skb)->csum)))
		return 0;

	NAPI_GRO_CB(skb)->csum = psum;

	return __skb_gro_checksum_complete(skb);
}

static inline void skb_gro_incr_csum_unnecessary(struct sk_buff *skb)
{
2483 2484 2485 2486 2487 2488 2489 2490 2491
	if (NAPI_GRO_CB(skb)->csum_cnt > 0) {
		/* Consume a checksum from CHECKSUM_UNNECESSARY */
		NAPI_GRO_CB(skb)->csum_cnt--;
	} else {
		/* Update skb for CHECKSUM_UNNECESSARY and csum_level when we
		 * verified a new top level checksum or an encapsulated one
		 * during GRO. This saves work if we fallback to normal path.
		 */
		__skb_incr_checksum_unnecessary(skb);
T
Tom Herbert 已提交
2492 2493 2494 2495 2496 2497 2498 2499 2500 2501
	}
}

#define __skb_gro_checksum_validate(skb, proto, zero_okay, check,	\
				    compute_pseudo)			\
({									\
	__sum16 __ret = 0;						\
	if (__skb_gro_checksum_validate_needed(skb, zero_okay, check))	\
		__ret = __skb_gro_checksum_validate_complete(skb,	\
				compute_pseudo(skb, proto));		\
2502 2503 2504
	if (__ret)							\
		__skb_mark_checksum_bad(skb);				\
	else								\
T
Tom Herbert 已提交
2505 2506 2507 2508 2509 2510 2511 2512 2513 2514 2515 2516 2517 2518
		skb_gro_incr_csum_unnecessary(skb);			\
	__ret;								\
})

#define skb_gro_checksum_validate(skb, proto, compute_pseudo)		\
	__skb_gro_checksum_validate(skb, proto, false, 0, compute_pseudo)

#define skb_gro_checksum_validate_zero_check(skb, proto, check,		\
					     compute_pseudo)		\
	__skb_gro_checksum_validate(skb, proto, true, check, compute_pseudo)

#define skb_gro_checksum_simple_validate(skb)				\
	__skb_gro_checksum_validate(skb, 0, false, 0, null_compute_pseudo)

2519 2520 2521 2522 2523 2524 2525 2526 2527 2528 2529 2530 2531 2532 2533 2534 2535 2536 2537 2538
static inline bool __skb_gro_checksum_convert_check(struct sk_buff *skb)
{
	return (NAPI_GRO_CB(skb)->csum_cnt == 0 &&
		!NAPI_GRO_CB(skb)->csum_valid);
}

static inline void __skb_gro_checksum_convert(struct sk_buff *skb,
					      __sum16 check, __wsum pseudo)
{
	NAPI_GRO_CB(skb)->csum = ~pseudo;
	NAPI_GRO_CB(skb)->csum_valid = 1;
}

#define skb_gro_checksum_try_convert(skb, proto, check, compute_pseudo)	\
do {									\
	if (__skb_gro_checksum_convert_check(skb))			\
		__skb_gro_checksum_convert(skb, check,			\
					   compute_pseudo(skb, proto));	\
} while (0)

2539 2540 2541 2542 2543 2544 2545
struct gro_remcsum {
	int offset;
	__wsum delta;
};

static inline void skb_gro_remcsum_init(struct gro_remcsum *grc)
{
2546
	grc->offset = 0;
2547 2548 2549
	grc->delta = 0;
}

2550 2551 2552 2553 2554
static inline void *skb_gro_remcsum_process(struct sk_buff *skb, void *ptr,
					    unsigned int off, size_t hdrlen,
					    int start, int offset,
					    struct gro_remcsum *grc,
					    bool nopartial)
2555 2556
{
	__wsum delta;
2557
	size_t plen = hdrlen + max_t(size_t, offset + sizeof(u16), start);
2558 2559 2560

	BUG_ON(!NAPI_GRO_CB(skb)->csum_valid);

2561
	if (!nopartial) {
2562 2563 2564 2565 2566 2567 2568 2569 2570
		NAPI_GRO_CB(skb)->gro_remcsum_start = off + hdrlen + start;
		return ptr;
	}

	ptr = skb_gro_header_fast(skb, off);
	if (skb_gro_header_hard(skb, off + plen)) {
		ptr = skb_gro_header_slow(skb, off + plen, off);
		if (!ptr)
			return NULL;
2571 2572
	}

2573 2574
	delta = remcsum_adjust(ptr + hdrlen, NAPI_GRO_CB(skb)->csum,
			       start, offset);
2575 2576 2577

	/* Adjust skb->csum since we changed the packet */
	NAPI_GRO_CB(skb)->csum = csum_add(NAPI_GRO_CB(skb)->csum, delta);
2578

2579
	grc->offset = off + hdrlen + offset;
2580
	grc->delta = delta;
2581 2582

	return ptr;
2583 2584
}

2585 2586 2587
static inline void skb_gro_remcsum_cleanup(struct sk_buff *skb,
					   struct gro_remcsum *grc)
{
2588 2589 2590
	void *ptr;
	size_t plen = grc->offset + sizeof(u16);

2591 2592 2593
	if (!grc->delta)
		return;

2594 2595 2596 2597 2598 2599 2600 2601
	ptr = skb_gro_header_fast(skb, grc->offset);
	if (skb_gro_header_hard(skb, grc->offset + sizeof(u16))) {
		ptr = skb_gro_header_slow(skb, plen, grc->offset);
		if (!ptr)
			return;
	}

	remcsum_unadjust((__sum16 *)ptr, grc->delta);
2602
}
2603

2604 2605 2606 2607 2608 2609 2610 2611 2612 2613 2614 2615 2616 2617 2618 2619 2620 2621 2622 2623 2624 2625 2626 2627 2628 2629 2630 2631 2632 2633 2634 2635 2636 2637 2638 2639 2640 2641 2642 2643 2644 2645 2646 2647 2648 2649 2650 2651 2652 2653 2654 2655 2656 2657 2658 2659 2660 2661 2662 2663 2664 2665 2666 2667 2668
struct skb_csum_offl_spec {
	__u16		ipv4_okay:1,
			ipv6_okay:1,
			encap_okay:1,
			ip_options_okay:1,
			ext_hdrs_okay:1,
			tcp_okay:1,
			udp_okay:1,
			sctp_okay:1,
			vlan_okay:1,
			no_encapped_ipv6:1,
			no_not_encapped:1;
};

bool __skb_csum_offload_chk(struct sk_buff *skb,
			    const struct skb_csum_offl_spec *spec,
			    bool *csum_encapped,
			    bool csum_help);

static inline bool skb_csum_offload_chk(struct sk_buff *skb,
					const struct skb_csum_offl_spec *spec,
					bool *csum_encapped,
					bool csum_help)
{
	if (skb->ip_summed != CHECKSUM_PARTIAL)
		return false;

	return __skb_csum_offload_chk(skb, spec, csum_encapped, csum_help);
}

static inline bool skb_csum_offload_chk_help(struct sk_buff *skb,
					     const struct skb_csum_offl_spec *spec)
{
	bool csum_encapped;

	return skb_csum_offload_chk(skb, spec, &csum_encapped, true);
}

static inline bool skb_csum_off_chk_help_cmn(struct sk_buff *skb)
{
	static const struct skb_csum_offl_spec csum_offl_spec = {
		.ipv4_okay = 1,
		.ip_options_okay = 1,
		.ipv6_okay = 1,
		.vlan_okay = 1,
		.tcp_okay = 1,
		.udp_okay = 1,
	};

	return skb_csum_offload_chk_help(skb, &csum_offl_spec);
}

static inline bool skb_csum_off_chk_help_cmn_v4_only(struct sk_buff *skb)
{
	static const struct skb_csum_offl_spec csum_offl_spec = {
		.ipv4_okay = 1,
		.ip_options_okay = 1,
		.tcp_okay = 1,
		.udp_okay = 1,
		.vlan_okay = 1,
	};

	return skb_csum_offload_chk_help(skb, &csum_offl_spec);
}

2669 2670
static inline int dev_hard_header(struct sk_buff *skb, struct net_device *dev,
				  unsigned short type,
2671
				  const void *daddr, const void *saddr,
2672
				  unsigned int len)
2673
{
2674
	if (!dev->header_ops || !dev->header_ops->create)
2675
		return 0;
2676 2677

	return dev->header_ops->create(skb, dev, type, daddr, saddr, len);
2678 2679
}

S
Stephen Hemminger 已提交
2680 2681 2682 2683 2684
static inline int dev_parse_header(const struct sk_buff *skb,
				   unsigned char *haddr)
{
	const struct net_device *dev = skb->dev;

2685
	if (!dev->header_ops || !dev->header_ops->parse)
S
Stephen Hemminger 已提交
2686
		return 0;
2687
	return dev->header_ops->parse(skb, haddr);
S
Stephen Hemminger 已提交
2688 2689
}

L
Linus Torvalds 已提交
2690
typedef int gifconf_func_t(struct net_device * dev, char __user * bufptr, int len);
2691
int register_gifconf(unsigned int family, gifconf_func_t *gifconf);
L
Linus Torvalds 已提交
2692 2693 2694 2695 2696
static inline int unregister_gifconf(unsigned int family)
{
	return register_gifconf(family, NULL);
}

2697
#ifdef CONFIG_NET_FLOW_LIMIT
2698
#define FLOW_LIMIT_HISTORY	(1 << 7)  /* must be ^2 and !overflow buckets */
2699 2700 2701 2702 2703 2704 2705 2706 2707 2708 2709
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
2710
/*
E
Eric Dumazet 已提交
2711
 * Incoming packets are placed on per-cpu queues
L
Linus Torvalds 已提交
2712
 */
E
Eric Dumazet 已提交
2713
struct softnet_data {
L
Linus Torvalds 已提交
2714
	struct list_head	poll_list;
2715
	struct sk_buff_head	process_queue;
L
Linus Torvalds 已提交
2716

C
Changli Gao 已提交
2717
	/* stats */
2718 2719 2720 2721
	unsigned int		processed;
	unsigned int		time_squeeze;
	unsigned int		cpu_collision;
	unsigned int		received_rps;
2722
#ifdef CONFIG_RPS
E
Eric Dumazet 已提交
2723
	struct softnet_data	*rps_ipi_list;
E
Eric Dumazet 已提交
2724 2725 2726 2727 2728 2729 2730
#endif
#ifdef CONFIG_NET_FLOW_LIMIT
	struct sd_flow_limit __rcu *flow_limit;
#endif
	struct Qdisc		*output_queue;
	struct Qdisc		**output_queue_tailp;
	struct sk_buff		*completion_queue;
E
Eric Dumazet 已提交
2731

E
Eric Dumazet 已提交
2732
#ifdef CONFIG_RPS
E
Eric Dumazet 已提交
2733
	/* Elements below can be accessed between CPUs for RPS */
T
Tom Herbert 已提交
2734
	struct call_single_data	csd ____cacheline_aligned_in_smp;
E
Eric Dumazet 已提交
2735 2736
	struct softnet_data	*rps_ipi_next;
	unsigned int		cpu;
T
Tom Herbert 已提交
2737
	unsigned int		input_queue_head;
2738
	unsigned int		input_queue_tail;
2739
#endif
2740
	unsigned int		dropped;
T
Tom Herbert 已提交
2741
	struct sk_buff_head	input_pkt_queue;
2742
	struct napi_struct	backlog;
2743

L
Linus Torvalds 已提交
2744 2745
};

2746
static inline void input_queue_head_incr(struct softnet_data *sd)
T
Tom Herbert 已提交
2747 2748
{
#ifdef CONFIG_RPS
2749 2750 2751 2752 2753 2754 2755 2756 2757
	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
2758 2759 2760
#endif
}

T
Tom Herbert 已提交
2761
DECLARE_PER_CPU_ALIGNED(struct softnet_data, softnet_data);
L
Linus Torvalds 已提交
2762

2763
void __netif_schedule(struct Qdisc *q);
2764
void netif_schedule_queue(struct netdev_queue *txq);
2765

2766 2767 2768 2769 2770 2771 2772 2773
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));
}

2774 2775
static inline void netif_tx_start_queue(struct netdev_queue *dev_queue)
{
2776
	clear_bit(__QUEUE_STATE_DRV_XOFF, &dev_queue->state);
2777 2778
}

2779 2780 2781 2782 2783 2784
/**
 *	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
2785 2786
static inline void netif_start_queue(struct net_device *dev)
{
2787
	netif_tx_start_queue(netdev_get_tx_queue(dev, 0));
L
Linus Torvalds 已提交
2788 2789
}

2790 2791 2792 2793 2794 2795 2796 2797 2798 2799
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);
	}
}

2800
void netif_tx_wake_queue(struct netdev_queue *dev_queue);
2801

2802 2803 2804 2805 2806 2807 2808
/**
 *	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.
 */
2809 2810
static inline void netif_wake_queue(struct net_device *dev)
{
2811
	netif_tx_wake_queue(netdev_get_tx_queue(dev, 0));
L
Linus Torvalds 已提交
2812 2813
}

2814 2815 2816 2817 2818 2819 2820 2821 2822 2823
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);
	}
}

2824 2825
static inline void netif_tx_stop_queue(struct netdev_queue *dev_queue)
{
2826
	set_bit(__QUEUE_STATE_DRV_XOFF, &dev_queue->state);
2827 2828
}

2829 2830 2831 2832 2833 2834 2835
/**
 *	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
2836 2837
static inline void netif_stop_queue(struct net_device *dev)
{
2838
	netif_tx_stop_queue(netdev_get_tx_queue(dev, 0));
L
Linus Torvalds 已提交
2839 2840
}

2841
void netif_tx_stop_all_queues(struct net_device *dev);
2842

2843
static inline bool netif_tx_queue_stopped(const struct netdev_queue *dev_queue)
2844
{
2845
	return test_bit(__QUEUE_STATE_DRV_XOFF, &dev_queue->state);
2846 2847
}

2848 2849 2850 2851 2852 2853
/**
 *	netif_queue_stopped - test if transmit queue is flowblocked
 *	@dev: network device
 *
 *	Test if transmit queue on device is currently unable to send.
 */
2854
static inline bool netif_queue_stopped(const struct net_device *dev)
L
Linus Torvalds 已提交
2855
{
2856
	return netif_tx_queue_stopped(netdev_get_tx_queue(dev, 0));
L
Linus Torvalds 已提交
2857 2858
}

2859
static inline bool netif_xmit_stopped(const struct netdev_queue *dev_queue)
2860
{
2861 2862 2863
	return dev_queue->state & QUEUE_STATE_ANY_XOFF;
}

2864 2865
static inline bool
netif_xmit_frozen_or_stopped(const struct netdev_queue *dev_queue)
2866 2867 2868 2869
{
	return dev_queue->state & QUEUE_STATE_ANY_XOFF_OR_FROZEN;
}

2870 2871 2872 2873 2874 2875
static inline bool
netif_xmit_frozen_or_drv_stopped(const struct netdev_queue *dev_queue)
{
	return dev_queue->state & QUEUE_STATE_DRV_XOFF_OR_FROZEN;
}

2876 2877 2878 2879 2880 2881 2882 2883 2884 2885 2886 2887 2888 2889 2890 2891 2892 2893 2894 2895 2896 2897 2898 2899 2900 2901 2902 2903
/**
 *	netdev_txq_bql_enqueue_prefetchw - prefetch bql data for write
 *	@dev_queue: pointer to transmit queue
 *
 * BQL enabled drivers might use this helper in their ndo_start_xmit(),
 * to give appropriate hint to the cpu.
 */
static inline void netdev_txq_bql_enqueue_prefetchw(struct netdev_queue *dev_queue)
{
#ifdef CONFIG_BQL
	prefetchw(&dev_queue->dql.num_queued);
#endif
}

/**
 *	netdev_txq_bql_complete_prefetchw - prefetch bql data for write
 *	@dev_queue: pointer to transmit queue
 *
 * BQL enabled drivers might use this helper in their TX completion path,
 * to give appropriate hint to the cpu.
 */
static inline void netdev_txq_bql_complete_prefetchw(struct netdev_queue *dev_queue)
{
#ifdef CONFIG_BQL
	prefetchw(&dev_queue->dql.limit);
#endif
}

2904 2905 2906
static inline void netdev_tx_sent_queue(struct netdev_queue *dev_queue,
					unsigned int bytes)
{
T
Tom Herbert 已提交
2907 2908
#ifdef CONFIG_BQL
	dql_queued(&dev_queue->dql, bytes);
2909 2910 2911 2912 2913 2914 2915 2916 2917 2918 2919 2920 2921 2922 2923 2924

	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
2925
#endif
2926 2927
}

2928 2929 2930 2931 2932 2933 2934 2935 2936
/**
 * 	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
 */
2937 2938 2939 2940 2941 2942
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,
2943
					     unsigned int pkts, unsigned int bytes)
2944
{
T
Tom Herbert 已提交
2945
#ifdef CONFIG_BQL
2946 2947 2948 2949 2950 2951 2952 2953 2954 2955 2956 2957 2958 2959 2960 2961 2962
	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
2963
#endif
2964 2965
}

2966 2967 2968 2969 2970 2971 2972 2973 2974 2975
/**
 * 	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()
 */
2976
static inline void netdev_completed_queue(struct net_device *dev,
2977
					  unsigned int pkts, unsigned int bytes)
2978 2979 2980 2981 2982 2983
{
	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
2984
#ifdef CONFIG_BQL
2985
	clear_bit(__QUEUE_STATE_STACK_XOFF, &q->state);
T
Tom Herbert 已提交
2986 2987
	dql_reset(&q->dql);
#endif
2988 2989
}

2990 2991 2992 2993 2994 2995 2996
/**
 * 	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
 */
2997 2998 2999
static inline void netdev_reset_queue(struct net_device *dev_queue)
{
	netdev_tx_reset_queue(netdev_get_tx_queue(dev_queue, 0));
3000 3001
}

3002 3003 3004 3005 3006 3007 3008 3009 3010 3011 3012 3013 3014 3015 3016 3017 3018 3019 3020 3021
/**
 * 	netdev_cap_txqueue - check if selected tx queue exceeds device queues
 * 	@dev: network device
 * 	@queue_index: given tx queue index
 *
 * 	Returns 0 if given tx queue index >= number of device tx queues,
 * 	otherwise returns the originally passed tx queue index.
 */
static inline u16 netdev_cap_txqueue(struct net_device *dev, u16 queue_index)
{
	if (unlikely(queue_index >= dev->real_num_tx_queues)) {
		net_warn_ratelimited("%s selects TX queue %d, but real number of TX queues is %d\n",
				     dev->name, queue_index,
				     dev->real_num_tx_queues);
		return 0;
	}

	return queue_index;
}

3022 3023 3024 3025 3026 3027
/**
 *	netif_running - test if up
 *	@dev: network device
 *
 *	Test if the device has been brought up.
 */
3028
static inline bool netif_running(const struct net_device *dev)
L
Linus Torvalds 已提交
3029 3030 3031 3032
{
	return test_bit(__LINK_STATE_START, &dev->state);
}

3033 3034 3035 3036 3037 3038
/*
 * 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.
 */
3039 3040 3041 3042 3043 3044 3045 3046

/**
 *	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.
 */
3047 3048
static inline void netif_start_subqueue(struct net_device *dev, u16 queue_index)
{
3049
	struct netdev_queue *txq = netdev_get_tx_queue(dev, queue_index);
3050 3051

	netif_tx_start_queue(txq);
3052 3053
}

3054 3055 3056 3057 3058 3059 3060
/**
 *	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.
 */
3061 3062
static inline void netif_stop_subqueue(struct net_device *dev, u16 queue_index)
{
3063
	struct netdev_queue *txq = netdev_get_tx_queue(dev, queue_index);
3064
	netif_tx_stop_queue(txq);
3065 3066
}

3067 3068 3069 3070 3071 3072 3073
/**
 *	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.
 */
3074 3075
static inline bool __netif_subqueue_stopped(const struct net_device *dev,
					    u16 queue_index)
3076
{
3077
	struct netdev_queue *txq = netdev_get_tx_queue(dev, queue_index);
3078 3079

	return netif_tx_queue_stopped(txq);
3080 3081
}

3082 3083
static inline bool netif_subqueue_stopped(const struct net_device *dev,
					  struct sk_buff *skb)
3084 3085 3086
{
	return __netif_subqueue_stopped(dev, skb_get_queue_mapping(skb));
}
3087

3088
void netif_wake_subqueue(struct net_device *dev, u16 queue_index);
3089

3090
#ifdef CONFIG_XPS
3091
int netif_set_xps_queue(struct net_device *dev, const struct cpumask *mask,
3092
			u16 index);
3093 3094
#else
static inline int netif_set_xps_queue(struct net_device *dev,
3095
				      const struct cpumask *mask,
3096 3097 3098 3099 3100 3101
				      u16 index)
{
	return 0;
}
#endif

J
Jiri Pirko 已提交
3102 3103 3104
u16 __skb_tx_hash(const struct net_device *dev, struct sk_buff *skb,
		  unsigned int num_tx_queues);

3105 3106 3107 3108 3109
/*
 * 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,
3110
			      struct sk_buff *skb)
3111 3112 3113 3114
{
	return __skb_tx_hash(dev, skb, dev->real_num_tx_queues);
}

3115 3116 3117 3118 3119 3120
/**
 *	netif_is_multiqueue - test if device has multiple transmit queues
 *	@dev: network device
 *
 * Check if device has multiple transmit queues
 */
3121
static inline bool netif_is_multiqueue(const struct net_device *dev)
3122
{
E
Eric Dumazet 已提交
3123
	return dev->num_tx_queues > 1;
3124
}
L
Linus Torvalds 已提交
3125

3126
int netif_set_real_num_tx_queues(struct net_device *dev, unsigned int txq);
3127

3128
#ifdef CONFIG_SYSFS
3129
int netif_set_real_num_rx_queues(struct net_device *dev, unsigned int rxq);
3130 3131 3132 3133 3134 3135 3136 3137
#else
static inline int netif_set_real_num_rx_queues(struct net_device *dev,
						unsigned int rxq)
{
	return 0;
}
#endif

3138 3139 3140 3141 3142 3143 3144 3145 3146 3147 3148 3149
#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

3150
#define DEFAULT_MAX_NUM_RSS_QUEUES	(8)
3151
int netif_get_num_default_rss_queues(void);
3152

3153 3154 3155 3156 3157 3158 3159
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
3160

3161 3162 3163 3164 3165 3166 3167 3168 3169 3170 3171 3172 3173 3174 3175 3176 3177 3178
/*
 * 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
3179
 */
3180 3181 3182 3183 3184 3185 3186 3187 3188 3189 3190 3191 3192 3193 3194 3195 3196 3197 3198
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
3199

3200 3201
int netif_rx(struct sk_buff *skb);
int netif_rx_ni(struct sk_buff *skb);
3202
int netif_receive_skb(struct sk_buff *skb);
3203 3204 3205 3206
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);
3207 3208
struct packet_offload *gro_find_receive_by_type(__be16 type);
struct packet_offload *gro_find_complete_by_type(__be16 type);
3209 3210 3211 3212 3213 3214 3215

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

3216 3217 3218 3219 3220 3221 3222 3223 3224 3225 3226
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);
3227 3228
void __dev_notify_flags(struct net_device *, unsigned int old_flags,
			unsigned int gchanges);
3229 3230 3231 3232 3233 3234 3235 3236
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,
3237
			 struct netdev_phys_item_id *ppid);
3238 3239
int dev_get_phys_port_name(struct net_device *dev,
			   char *name, size_t len);
3240
int dev_change_proto_down(struct net_device *dev, bool proto_down);
3241
struct sk_buff *validate_xmit_skb_list(struct sk_buff *skb, struct net_device *dev);
3242 3243
struct sk_buff *dev_hard_start_xmit(struct sk_buff *skb, struct net_device *dev,
				    struct netdev_queue *txq, int *ret);
H
Herbert Xu 已提交
3244
int __dev_forward_skb(struct net_device *dev, struct sk_buff *skb);
3245
int dev_forward_skb(struct net_device *dev, struct sk_buff *skb);
3246
bool is_skb_forwardable(struct net_device *dev, struct sk_buff *skb);
L
Linus Torvalds 已提交
3247

3248
extern int		netdev_budget;
L
Linus Torvalds 已提交
3249 3250

/* Called by rtnetlink.c:rtnl_unlock() */
3251
void netdev_run_todo(void);
L
Linus Torvalds 已提交
3252

3253 3254 3255 3256
/**
 *	dev_put - release reference to device
 *	@dev: network device
 *
3257
 * Release reference to device to allow it to be freed.
3258
 */
L
Linus Torvalds 已提交
3259 3260
static inline void dev_put(struct net_device *dev)
{
3261
	this_cpu_dec(*dev->pcpu_refcnt);
L
Linus Torvalds 已提交
3262 3263
}

3264 3265 3266 3267
/**
 *	dev_hold - get reference to device
 *	@dev: network device
 *
3268
 * Hold reference to device to keep it from being freed.
3269
 */
3270 3271
static inline void dev_hold(struct net_device *dev)
{
3272
	this_cpu_inc(*dev->pcpu_refcnt);
3273
}
L
Linus Torvalds 已提交
3274 3275 3276 3277

/* 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
3278 3279 3280 3281
 *
 * 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
3282 3283
 */

3284 3285 3286
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
3287

3288 3289 3290 3291 3292 3293
/**
 *	netif_carrier_ok - test if carrier present
 *	@dev: network device
 *
 * Check if carrier is present on device
 */
3294
static inline bool netif_carrier_ok(const struct net_device *dev)
L
Linus Torvalds 已提交
3295 3296 3297 3298
{
	return !test_bit(__LINK_STATE_NOCARRIER, &dev->state);
}

3299
unsigned long dev_trans_start(struct net_device *dev);
3300

3301
void __netdev_watchdog_up(struct net_device *dev);
L
Linus Torvalds 已提交
3302

3303
void netif_carrier_on(struct net_device *dev);
L
Linus Torvalds 已提交
3304

3305
void netif_carrier_off(struct net_device *dev);
L
Linus Torvalds 已提交
3306

3307 3308 3309 3310 3311 3312 3313 3314 3315 3316 3317 3318 3319
/**
 *	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
3320 3321 3322 3323 3324 3325
static inline void netif_dormant_on(struct net_device *dev)
{
	if (!test_and_set_bit(__LINK_STATE_DORMANT, &dev->state))
		linkwatch_fire_event(dev);
}

3326 3327 3328 3329 3330 3331
/**
 *	netif_dormant_off - set device as not dormant.
 *	@dev: network device
 *
 * Device is not in dormant state.
 */
S
Stefan Rompf 已提交
3332 3333 3334 3335 3336 3337
static inline void netif_dormant_off(struct net_device *dev)
{
	if (test_and_clear_bit(__LINK_STATE_DORMANT, &dev->state))
		linkwatch_fire_event(dev);
}

3338 3339 3340 3341 3342 3343
/**
 *	netif_dormant - test if carrier present
 *	@dev: network device
 *
 * Check if carrier is present on device
 */
3344
static inline bool netif_dormant(const struct net_device *dev)
S
Stefan Rompf 已提交
3345 3346 3347 3348 3349
{
	return test_bit(__LINK_STATE_DORMANT, &dev->state);
}


3350 3351 3352 3353 3354 3355
/**
 *	netif_oper_up - test if device is operational
 *	@dev: network device
 *
 * Check if carrier is operational
 */
3356
static inline bool netif_oper_up(const struct net_device *dev)
E
Eric Dumazet 已提交
3357
{
S
Stefan Rompf 已提交
3358 3359 3360 3361
	return (dev->operstate == IF_OPER_UP ||
		dev->operstate == IF_OPER_UNKNOWN /* backward compat */);
}

3362 3363 3364 3365 3366 3367
/**
 *	netif_device_present - is device available or removed
 *	@dev: network device
 *
 * Check if device has not been removed from system.
 */
3368
static inline bool netif_device_present(struct net_device *dev)
L
Linus Torvalds 已提交
3369 3370 3371 3372
{
	return test_bit(__LINK_STATE_PRESENT, &dev->state);
}

3373
void netif_device_detach(struct net_device *dev);
L
Linus Torvalds 已提交
3374

3375
void netif_device_attach(struct net_device *dev);
L
Linus Torvalds 已提交
3376 3377 3378 3379 3380 3381 3382 3383 3384 3385 3386 3387 3388 3389 3390 3391 3392 3393 3394 3395 3396 3397 3398 3399 3400 3401 3402 3403 3404 3405 3406 3407 3408 3409 3410 3411 3412 3413 3414 3415 3416 3417 3418 3419 3420 3421 3422 3423 3424 3425

/*
 * 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;
}

3426
static inline void __netif_tx_lock(struct netdev_queue *txq, int cpu)
H
Herbert Xu 已提交
3427
{
3428 3429
	spin_lock(&txq->_xmit_lock);
	txq->xmit_lock_owner = cpu;
3430 3431
}

3432 3433 3434 3435 3436 3437
static inline void __netif_tx_lock_bh(struct netdev_queue *txq)
{
	spin_lock_bh(&txq->_xmit_lock);
	txq->xmit_lock_owner = smp_processor_id();
}

3438
static inline bool __netif_tx_trylock(struct netdev_queue *txq)
3439
{
3440
	bool ok = spin_trylock(&txq->_xmit_lock);
3441 3442 3443 3444 3445 3446 3447 3448 3449 3450 3451 3452 3453 3454 3455 3456 3457
	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
3458 3459 3460 3461 3462 3463
static inline void txq_trans_update(struct netdev_queue *txq)
{
	if (txq->xmit_lock_owner != -1)
		txq->trans_start = jiffies;
}

3464 3465 3466 3467 3468 3469
/**
 *	netif_tx_lock - grab network device transmit lock
 *	@dev: network device
 *
 * Get network device transmit lock
 */
3470 3471
static inline void netif_tx_lock(struct net_device *dev)
{
3472
	unsigned int i;
3473
	int cpu;
3474

3475 3476
	spin_lock(&dev->tx_global_lock);
	cpu = smp_processor_id();
3477 3478
	for (i = 0; i < dev->num_tx_queues; i++) {
		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
3479 3480 3481 3482 3483 3484 3485

		/* 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.
		 */
3486
		__netif_tx_lock(txq, cpu);
3487 3488
		set_bit(__QUEUE_STATE_FROZEN, &txq->state);
		__netif_tx_unlock(txq);
3489
	}
H
Herbert Xu 已提交
3490 3491 3492 3493
}

static inline void netif_tx_lock_bh(struct net_device *dev)
{
3494 3495
	local_bh_disable();
	netif_tx_lock(dev);
H
Herbert Xu 已提交
3496 3497 3498 3499
}

static inline void netif_tx_unlock(struct net_device *dev)
{
3500 3501 3502 3503
	unsigned int i;

	for (i = 0; i < dev->num_tx_queues; i++) {
		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
3504

3505 3506 3507 3508 3509
		/* 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);
3510
		netif_schedule_queue(txq);
3511 3512
	}
	spin_unlock(&dev->tx_global_lock);
H
Herbert Xu 已提交
3513 3514 3515 3516
}

static inline void netif_tx_unlock_bh(struct net_device *dev)
{
3517 3518
	netif_tx_unlock(dev);
	local_bh_enable();
H
Herbert Xu 已提交
3519 3520
}

3521
#define HARD_TX_LOCK(dev, txq, cpu) {			\
3522
	if ((dev->features & NETIF_F_LLTX) == 0) {	\
3523
		__netif_tx_lock(txq, cpu);		\
3524 3525 3526
	}						\
}

3527 3528 3529 3530 3531
#define HARD_TX_TRYLOCK(dev, txq)			\
	(((dev->features & NETIF_F_LLTX) == 0) ?	\
		__netif_tx_trylock(txq) :		\
		true )

3532
#define HARD_TX_UNLOCK(dev, txq) {			\
3533
	if ((dev->features & NETIF_F_LLTX) == 0) {	\
3534
		__netif_tx_unlock(txq);			\
3535 3536 3537
	}						\
}

L
Linus Torvalds 已提交
3538 3539
static inline void netif_tx_disable(struct net_device *dev)
{
3540
	unsigned int i;
3541
	int cpu;
3542

3543 3544
	local_bh_disable();
	cpu = smp_processor_id();
3545 3546
	for (i = 0; i < dev->num_tx_queues; i++) {
		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
3547 3548

		__netif_tx_lock(txq, cpu);
3549
		netif_tx_stop_queue(txq);
3550
		__netif_tx_unlock(txq);
3551
	}
3552
	local_bh_enable();
L
Linus Torvalds 已提交
3553 3554
}

3555 3556 3557 3558 3559
static inline void netif_addr_lock(struct net_device *dev)
{
	spin_lock(&dev->addr_list_lock);
}

3560 3561
static inline void netif_addr_lock_nested(struct net_device *dev)
{
3562 3563 3564 3565 3566 3567
	int subclass = SINGLE_DEPTH_NESTING;

	if (dev->netdev_ops->ndo_get_lock_subclass)
		subclass = dev->netdev_ops->ndo_get_lock_subclass(dev);

	spin_lock_nested(&dev->addr_list_lock, subclass);
3568 3569
}

3570 3571 3572 3573 3574 3575 3576 3577 3578 3579 3580 3581 3582 3583 3584
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);
}

3585
/*
3586
 * dev_addrs walker. Should be used only for read access. Call with
3587 3588 3589
 * rcu_read_lock held.
 */
#define for_each_dev_addr(dev, ha) \
3590
		list_for_each_entry_rcu(ha, &dev->dev_addrs.list, list)
3591

L
Linus Torvalds 已提交
3592 3593
/* These functions live elsewhere (drivers/net/net_init.c, but related) */

3594
void ether_setup(struct net_device *dev);
L
Linus Torvalds 已提交
3595 3596

/* Support for loadable net-drivers */
3597
struct net_device *alloc_netdev_mqs(int sizeof_priv, const char *name,
3598
				    unsigned char name_assign_type,
3599 3600
				    void (*setup)(struct net_device *),
				    unsigned int txqs, unsigned int rxqs);
3601 3602
#define alloc_netdev(sizeof_priv, name, name_assign_type, setup) \
	alloc_netdev_mqs(sizeof_priv, name, name_assign_type, setup, 1, 1)
T
Tom Herbert 已提交
3603

3604 3605 3606
#define alloc_netdev_mq(sizeof_priv, name, name_assign_type, setup, count) \
	alloc_netdev_mqs(sizeof_priv, name, name_assign_type, setup, count, \
			 count)
T
Tom Herbert 已提交
3607

3608 3609
int register_netdev(struct net_device *dev);
void unregister_netdev(struct net_device *dev);
3610

3611
/* General hardware address lists handling functions */
3612 3613 3614 3615
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);
3616 3617 3618 3619 3620 3621 3622 3623 3624
int __hw_addr_sync_dev(struct netdev_hw_addr_list *list,
		       struct net_device *dev,
		       int (*sync)(struct net_device *, const unsigned char *),
		       int (*unsync)(struct net_device *,
				     const unsigned char *));
void __hw_addr_unsync_dev(struct netdev_hw_addr_list *list,
			  struct net_device *dev,
			  int (*unsync)(struct net_device *,
					const unsigned char *));
3625
void __hw_addr_init(struct netdev_hw_addr_list *list);
3626

3627
/* Functions used for device addresses handling */
3628 3629 3630 3631 3632 3633
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);
3634 3635

/* Functions used for unicast addresses handling */
3636 3637 3638 3639 3640 3641 3642 3643
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);
3644

3645 3646 3647 3648 3649 3650 3651 3652 3653 3654 3655 3656 3657 3658 3659 3660 3661 3662 3663
/**
 *  __dev_uc_sync - Synchonize device's unicast list
 *  @dev:  device to sync
 *  @sync: function to call if address should be added
 *  @unsync: function to call if address should be removed
 *
 *  Add newly added addresses to the interface, and release
 *  addresses that have been deleted.
 **/
static inline int __dev_uc_sync(struct net_device *dev,
				int (*sync)(struct net_device *,
					    const unsigned char *),
				int (*unsync)(struct net_device *,
					      const unsigned char *))
{
	return __hw_addr_sync_dev(&dev->uc, dev, sync, unsync);
}

/**
3664
 *  __dev_uc_unsync - Remove synchronized addresses from device
3665 3666 3667 3668 3669 3670 3671 3672 3673 3674 3675 3676
 *  @dev:  device to sync
 *  @unsync: function to call if address should be removed
 *
 *  Remove all addresses that were added to the device by dev_uc_sync().
 **/
static inline void __dev_uc_unsync(struct net_device *dev,
				   int (*unsync)(struct net_device *,
						 const unsigned char *))
{
	__hw_addr_unsync_dev(&dev->uc, dev, unsync);
}

3677
/* Functions used for multicast addresses handling */
3678 3679 3680 3681 3682 3683 3684 3685 3686 3687
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);
3688

3689 3690 3691 3692 3693 3694 3695 3696 3697 3698 3699 3700 3701 3702 3703 3704 3705 3706 3707
/**
 *  __dev_mc_sync - Synchonize device's multicast list
 *  @dev:  device to sync
 *  @sync: function to call if address should be added
 *  @unsync: function to call if address should be removed
 *
 *  Add newly added addresses to the interface, and release
 *  addresses that have been deleted.
 **/
static inline int __dev_mc_sync(struct net_device *dev,
				int (*sync)(struct net_device *,
					    const unsigned char *),
				int (*unsync)(struct net_device *,
					      const unsigned char *))
{
	return __hw_addr_sync_dev(&dev->mc, dev, sync, unsync);
}

/**
3708
 *  __dev_mc_unsync - Remove synchronized addresses from device
3709 3710 3711 3712 3713 3714 3715 3716 3717 3718 3719 3720
 *  @dev:  device to sync
 *  @unsync: function to call if address should be removed
 *
 *  Remove all addresses that were added to the device by dev_mc_sync().
 **/
static inline void __dev_mc_unsync(struct net_device *dev,
				   int (*unsync)(struct net_device *,
						 const unsigned char *))
{
	__hw_addr_unsync_dev(&dev->mc, dev, unsync);
}

3721
/* Functions used for secondary unicast and multicast support */
3722 3723 3724 3725 3726 3727 3728
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
3729
/* Load a device via the kmod */
3730 3731 3732 3733 3734
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);
3735

L
Linus Torvalds 已提交
3736
extern int		netdev_max_backlog;
E
Eric Dumazet 已提交
3737
extern int		netdev_tstamp_prequeue;
L
Linus Torvalds 已提交
3738
extern int		weight_p;
3739
extern int		bpf_jit_enable;
J
Jiri Pirko 已提交
3740

3741
bool netdev_has_upper_dev(struct net_device *dev, struct net_device *upper_dev);
3742 3743
struct net_device *netdev_upper_get_next_dev_rcu(struct net_device *dev,
						     struct list_head **iter);
3744 3745
struct net_device *netdev_all_upper_get_next_dev_rcu(struct net_device *dev,
						     struct list_head **iter);
3746

3747 3748 3749 3750 3751 3752 3753
/* iterate through upper list, must be called under RCU read lock */
#define netdev_for_each_upper_dev_rcu(dev, updev, iter) \
	for (iter = &(dev)->adj_list.upper, \
	     updev = netdev_upper_get_next_dev_rcu(dev, &(iter)); \
	     updev; \
	     updev = netdev_upper_get_next_dev_rcu(dev, &(iter)))

3754
/* iterate through upper list, must be called under RCU read lock */
3755 3756 3757 3758 3759
#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)))
3760

3761 3762 3763 3764
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);
3765 3766 3767 3768 3769 3770 3771 3772 3773 3774 3775 3776 3777

#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)))

3778 3779 3780
void *netdev_lower_get_next(struct net_device *dev,
				struct list_head **iter);
#define netdev_for_each_lower_dev(dev, ldev, iter) \
3781
	for (iter = (dev)->adj_list.lower.next, \
3782 3783 3784 3785
	     ldev = netdev_lower_get_next(dev, &(iter)); \
	     ldev; \
	     ldev = netdev_lower_get_next(dev, &(iter)))

3786
void *netdev_adjacent_get_private(struct list_head *adj_list);
3787
void *netdev_lower_get_first_private_rcu(struct net_device *dev);
3788 3789 3790 3791
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,
3792
				 struct net_device *upper_dev,
3793
				 void *upper_priv, void *upper_info);
3794 3795
void netdev_upper_dev_unlink(struct net_device *dev,
			     struct net_device *upper_dev);
3796
void netdev_adjacent_rename_links(struct net_device *dev, char *oldname);
3797 3798
void *netdev_lower_dev_get_private(struct net_device *dev,
				   struct net_device *lower_dev);
3799 3800
void netdev_lower_state_changed(struct net_device *lower_dev,
				void *lower_state_info);
3801 3802 3803

/* RSS keys are 40 or 52 bytes long */
#define NETDEV_RSS_KEY_LEN 52
3804
extern u8 netdev_rss_key[NETDEV_RSS_KEY_LEN] __read_mostly;
3805 3806
void netdev_rss_key_fill(void *buffer, size_t len);

3807
int dev_get_nest_level(struct net_device *dev,
3808
		       bool (*type_check)(const struct net_device *dev));
3809 3810 3811 3812 3813
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);
3814

3815 3816 3817 3818 3819 3820 3821 3822 3823 3824 3825 3826 3827
struct netdev_bonding_info {
	ifslave	slave;
	ifbond	master;
};

struct netdev_notifier_bonding_info {
	struct netdev_notifier_info info; /* must be first */
	struct netdev_bonding_info  bonding_info;
};

void netdev_bonding_info_change(struct net_device *dev,
				struct netdev_bonding_info *bonding_info);

3828 3829 3830 3831 3832
static inline
struct sk_buff *skb_gso_segment(struct sk_buff *skb, netdev_features_t features)
{
	return __skb_gso_segment(skb, features, true);
}
3833
__be16 skb_network_protocol(struct sk_buff *skb, int *depth);
3834 3835 3836 3837

static inline bool can_checksum_protocol(netdev_features_t features,
					 __be16 protocol)
{
3838 3839 3840 3841 3842 3843 3844 3845 3846 3847 3848 3849 3850 3851 3852 3853 3854 3855
	if (protocol == htons(ETH_P_FCOE))
		return !!(features & NETIF_F_FCOE_CRC);

	/* Assume this is an IP checksum (not SCTP CRC) */

	if (features & NETIF_F_HW_CSUM) {
		/* Can checksum everything */
		return true;
	}

	switch (protocol) {
	case htons(ETH_P_IP):
		return !!(features & NETIF_F_IP_CSUM);
	case htons(ETH_P_IPV6):
		return !!(features & NETIF_F_IPV6_CSUM);
	default:
		return false;
	}
3856
}
3857

3858 3859 3860 3861 3862 3863 3864 3865 3866 3867 3868 3869 3870
/* Map an ethertype into IP protocol if possible */
static inline int eproto_to_ipproto(int eproto)
{
	switch (eproto) {
	case htons(ETH_P_IP):
		return IPPROTO_IP;
	case htons(ETH_P_IPV6):
		return IPPROTO_IPV6;
	default:
		return -1;
	}
}

3871
#ifdef CONFIG_BUG
3872
void netdev_rx_csum_fault(struct net_device *dev);
3873 3874 3875 3876 3877
#else
static inline void netdev_rx_csum_fault(struct net_device *dev)
{
}
#endif
L
Linus Torvalds 已提交
3878
/* rx skb timestamps */
3879 3880
void net_enable_timestamp(void);
void net_disable_timestamp(void);
L
Linus Torvalds 已提交
3881

3882
#ifdef CONFIG_PROC_FS
3883
int __init dev_proc_init(void);
3884 3885
#else
#define dev_proc_init() 0
3886 3887
#endif

D
David S. Miller 已提交
3888
static inline netdev_tx_t __netdev_start_xmit(const struct net_device_ops *ops,
3889 3890
					      struct sk_buff *skb, struct net_device *dev,
					      bool more)
D
David S. Miller 已提交
3891
{
3892
	skb->xmit_more = more ? 1 : 0;
3893
	return ops->ndo_start_xmit(skb, dev);
D
David S. Miller 已提交
3894 3895
}

3896
static inline netdev_tx_t netdev_start_xmit(struct sk_buff *skb, struct net_device *dev,
3897
					    struct netdev_queue *txq, bool more)
D
David S. Miller 已提交
3898 3899
{
	const struct net_device_ops *ops = dev->netdev_ops;
3900
	int rc;
D
David S. Miller 已提交
3901

3902
	rc = __netdev_start_xmit(ops, skb, dev, more);
3903 3904 3905 3906
	if (rc == NETDEV_TX_OK)
		txq_trans_update(txq);

	return rc;
D
David S. Miller 已提交
3907 3908
}

3909 3910 3911 3912
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);
3913 3914 3915 3916 3917 3918 3919 3920 3921 3922

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);
}
3923

3924 3925
extern struct kobj_ns_type_operations net_ns_type_operations;

3926
const char *netdev_drivername(const struct net_device *dev);
3927

3928
void linkwatch_run_queue(void);
3929

3930 3931 3932
static inline netdev_features_t netdev_intersect_features(netdev_features_t f1,
							  netdev_features_t f2)
{
3933 3934
	if ((f1 ^ f2) & NETIF_F_HW_CSUM) {
		if (f1 & NETIF_F_HW_CSUM)
3935
			f1 |= (NETIF_F_IP_CSUM|NETIF_F_IPV6_CSUM);
3936
		else
3937
			f2 |= (NETIF_F_IP_CSUM|NETIF_F_IPV6_CSUM);
3938
	}
3939

3940
	return f1 & f2;
3941 3942
}

3943 3944
static inline netdev_features_t netdev_get_wanted_features(
	struct net_device *dev)
3945 3946 3947
{
	return (dev->features & ~dev->hw_features) | dev->wanted_features;
}
3948 3949
netdev_features_t netdev_increment_features(netdev_features_t all,
	netdev_features_t one, netdev_features_t mask);
3950 3951 3952 3953 3954 3955 3956 3957 3958 3959 3960

/* 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);
}

3961
int __netdev_update_features(struct net_device *dev);
3962
void netdev_update_features(struct net_device *dev);
3963
void netdev_change_features(struct net_device *dev);
3964

3965 3966 3967
void netif_stacked_transfer_operstate(const struct net_device *rootdev,
					struct net_device *dev);

3968 3969 3970
netdev_features_t passthru_features_check(struct sk_buff *skb,
					  struct net_device *dev,
					  netdev_features_t features);
3971
netdev_features_t netif_skb_features(struct sk_buff *skb);
3972

3973
static inline bool net_gso_ok(netdev_features_t features, int gso_type)
3974
{
3975
	netdev_features_t feature = gso_type << NETIF_F_GSO_SHIFT;
3976 3977 3978 3979 3980 3981 3982 3983

	/* 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));
3984 3985 3986 3987 3988 3989
	BUILD_BUG_ON(SKB_GSO_GRE     != (NETIF_F_GSO_GRE >> NETIF_F_GSO_SHIFT));
	BUILD_BUG_ON(SKB_GSO_GRE_CSUM != (NETIF_F_GSO_GRE_CSUM >> NETIF_F_GSO_SHIFT));
	BUILD_BUG_ON(SKB_GSO_IPIP    != (NETIF_F_GSO_IPIP >> NETIF_F_GSO_SHIFT));
	BUILD_BUG_ON(SKB_GSO_SIT     != (NETIF_F_GSO_SIT >> NETIF_F_GSO_SHIFT));
	BUILD_BUG_ON(SKB_GSO_UDP_TUNNEL != (NETIF_F_GSO_UDP_TUNNEL >> NETIF_F_GSO_SHIFT));
	BUILD_BUG_ON(SKB_GSO_UDP_TUNNEL_CSUM != (NETIF_F_GSO_UDP_TUNNEL_CSUM >> NETIF_F_GSO_SHIFT));
3990
	BUILD_BUG_ON(SKB_GSO_TUNNEL_REMCSUM != (NETIF_F_GSO_TUNNEL_REMCSUM >> NETIF_F_GSO_SHIFT));
3991

3992
	return (features & feature) == feature;
3993 3994
}

3995
static inline bool skb_gso_ok(struct sk_buff *skb, netdev_features_t features)
3996
{
H
Herbert Xu 已提交
3997
	return net_gso_ok(features, skb_shinfo(skb)->gso_type) &&
3998
	       (!skb_has_frag_list(skb) || (features & NETIF_F_FRAGLIST));
3999 4000
}

4001
static inline bool netif_needs_gso(struct sk_buff *skb,
4002
				   netdev_features_t features)
4003
{
4004
	return skb_is_gso(skb) && (!skb_gso_ok(skb, features) ||
4005 4006
		unlikely((skb->ip_summed != CHECKSUM_PARTIAL) &&
			 (skb->ip_summed != CHECKSUM_UNNECESSARY)));
4007 4008
}

4009 4010 4011 4012 4013 4014
static inline void netif_set_gso_max_size(struct net_device *dev,
					  unsigned int size)
{
	dev->gso_max_size = size;
}

4015 4016 4017 4018 4019 4020 4021 4022 4023 4024 4025 4026 4027
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;
}

4028
static inline bool netif_is_macvlan(const struct net_device *dev)
4029 4030 4031 4032
{
	return dev->priv_flags & IFF_MACVLAN;
}

4033
static inline bool netif_is_macvlan_port(const struct net_device *dev)
4034 4035 4036 4037
{
	return dev->priv_flags & IFF_MACVLAN_PORT;
}

4038
static inline bool netif_is_ipvlan(const struct net_device *dev)
4039 4040 4041 4042
{
	return dev->priv_flags & IFF_IPVLAN_SLAVE;
}

4043
static inline bool netif_is_ipvlan_port(const struct net_device *dev)
4044 4045 4046 4047
{
	return dev->priv_flags & IFF_IPVLAN_MASTER;
}

4048
static inline bool netif_is_bond_master(const struct net_device *dev)
4049 4050 4051 4052
{
	return dev->flags & IFF_MASTER && dev->priv_flags & IFF_BONDING;
}

4053
static inline bool netif_is_bond_slave(const struct net_device *dev)
J
Jiri Pirko 已提交
4054 4055 4056 4057
{
	return dev->flags & IFF_SLAVE && dev->priv_flags & IFF_BONDING;
}

4058 4059 4060 4061 4062
static inline bool netif_supports_nofcs(struct net_device *dev)
{
	return dev->priv_flags & IFF_SUPP_NOFCS;
}

4063
static inline bool netif_is_l3_master(const struct net_device *dev)
4064
{
4065
	return dev->priv_flags & IFF_L3MDEV_MASTER;
4066 4067
}

D
David Ahern 已提交
4068 4069 4070 4071 4072
static inline bool netif_is_l3_slave(const struct net_device *dev)
{
	return dev->priv_flags & IFF_L3MDEV_SLAVE;
}

4073 4074 4075 4076 4077
static inline bool netif_is_bridge_master(const struct net_device *dev)
{
	return dev->priv_flags & IFF_EBRIDGE;
}

4078 4079 4080 4081 4082
static inline bool netif_is_bridge_port(const struct net_device *dev)
{
	return dev->priv_flags & IFF_BRIDGE_PORT;
}

4083 4084 4085 4086 4087
static inline bool netif_is_ovs_master(const struct net_device *dev)
{
	return dev->priv_flags & IFF_OPENVSWITCH;
}

4088
static inline bool netif_is_team_master(const struct net_device *dev)
J
Jiri Pirko 已提交
4089 4090 4091 4092
{
	return dev->priv_flags & IFF_TEAM;
}

4093
static inline bool netif_is_team_port(const struct net_device *dev)
J
Jiri Pirko 已提交
4094 4095 4096 4097
{
	return dev->priv_flags & IFF_TEAM_PORT;
}

4098
static inline bool netif_is_lag_master(const struct net_device *dev)
J
Jiri Pirko 已提交
4099 4100 4101 4102
{
	return netif_is_bond_master(dev) || netif_is_team_master(dev);
}

4103
static inline bool netif_is_lag_port(const struct net_device *dev)
J
Jiri Pirko 已提交
4104 4105 4106 4107
{
	return netif_is_bond_slave(dev) || netif_is_team_port(dev);
}

4108 4109 4110 4111 4112
static inline bool netif_is_rxfh_configured(const struct net_device *dev)
{
	return dev->priv_flags & IFF_RXFH_CONFIGURED;
}

4113 4114 4115 4116 4117 4118
/* This device needs to keep skb dst for qdisc enqueue or ndo_start_xmit() */
static inline void netif_keep_dst(struct net_device *dev)
{
	dev->priv_flags &= ~(IFF_XMIT_DST_RELEASE | IFF_XMIT_DST_RELEASE_PERM);
}

4119
extern struct pernet_operations __net_initdata loopback_net_ops;
4120

4121 4122 4123 4124 4125 4126
/* Logging, debugging and troubleshooting/diagnostic helpers. */

/* netdev_printk helpers, similar to dev_printk */

static inline const char *netdev_name(const struct net_device *dev)
{
4127 4128
	if (!dev->name[0] || strchr(dev->name, '%'))
		return "(unnamed net_device)";
4129 4130 4131
	return dev->name;
}

4132 4133 4134 4135 4136 4137 4138 4139 4140 4141 4142 4143 4144 4145 4146
static inline const char *netdev_reg_state(const struct net_device *dev)
{
	switch (dev->reg_state) {
	case NETREG_UNINITIALIZED: return " (uninitialized)";
	case NETREG_REGISTERED: return "";
	case NETREG_UNREGISTERING: return " (unregistering)";
	case NETREG_UNREGISTERED: return " (unregistered)";
	case NETREG_RELEASED: return " (released)";
	case NETREG_DUMMY: return " (dummy)";
	}

	WARN_ONCE(1, "%s: unknown reg_state %d\n", dev->name, dev->reg_state);
	return " (unknown)";
}

4147
__printf(3, 4)
4148 4149
void netdev_printk(const char *level, const struct net_device *dev,
		   const char *format, ...);
4150
__printf(2, 3)
4151
void netdev_emerg(const struct net_device *dev, const char *format, ...);
4152
__printf(2, 3)
4153
void netdev_alert(const struct net_device *dev, const char *format, ...);
4154
__printf(2, 3)
4155
void netdev_crit(const struct net_device *dev, const char *format, ...);
4156
__printf(2, 3)
4157
void netdev_err(const struct net_device *dev, const char *format, ...);
4158
__printf(2, 3)
4159
void netdev_warn(const struct net_device *dev, const char *format, ...);
4160
__printf(2, 3)
4161
void netdev_notice(const struct net_device *dev, const char *format, ...);
4162
__printf(2, 3)
4163
void netdev_info(const struct net_device *dev, const char *format, ...);
4164

4165 4166 4167
#define MODULE_ALIAS_NETDEV(device) \
	MODULE_ALIAS("netdev-" device)

4168
#if defined(CONFIG_DYNAMIC_DEBUG)
4169 4170
#define netdev_dbg(__dev, format, args...)			\
do {								\
4171
	dynamic_netdev_dbg(__dev, format, ##args);		\
4172
} while (0)
4173 4174 4175
#elif defined(DEBUG)
#define netdev_dbg(__dev, format, args...)			\
	netdev_printk(KERN_DEBUG, __dev, format, ##args)
4176 4177 4178 4179 4180 4181 4182 4183 4184 4185 4186 4187 4188 4189 4190 4191 4192 4193 4194 4195 4196 4197 4198 4199 4200 4201
#else
#define netdev_dbg(__dev, format, args...)			\
({								\
	if (0)							\
		netdev_printk(KERN_DEBUG, __dev, format, ##args); \
})
#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...)			\
4202 4203
	WARN(1, "netdevice: %s%s\n" format, netdev_name(dev),	\
	     netdev_reg_state(dev), ##args)
4204

4205 4206 4207 4208 4209 4210 4211 4212
/* 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)

4213 4214 4215 4216 4217 4218
#define netif_level(level, priv, type, dev, fmt, args...)	\
do {								\
	if (netif_msg_##type(priv))				\
		netdev_##level(dev, fmt, ##args);		\
} while (0)

4219
#define netif_emerg(priv, type, dev, fmt, args...)		\
4220
	netif_level(emerg, priv, type, dev, fmt, ##args)
4221
#define netif_alert(priv, type, dev, fmt, args...)		\
4222
	netif_level(alert, priv, type, dev, fmt, ##args)
4223
#define netif_crit(priv, type, dev, fmt, args...)		\
4224
	netif_level(crit, priv, type, dev, fmt, ##args)
4225
#define netif_err(priv, type, dev, fmt, args...)		\
4226
	netif_level(err, priv, type, dev, fmt, ##args)
4227
#define netif_warn(priv, type, dev, fmt, args...)		\
4228
	netif_level(warn, priv, type, dev, fmt, ##args)
4229
#define netif_notice(priv, type, dev, fmt, args...)		\
4230
	netif_level(notice, priv, type, dev, fmt, ##args)
4231
#define netif_info(priv, type, dev, fmt, args...)		\
4232
	netif_level(info, priv, type, dev, fmt, ##args)
4233

4234
#if defined(CONFIG_DYNAMIC_DEBUG)
4235 4236 4237
#define netif_dbg(priv, type, netdev, format, args...)		\
do {								\
	if (netif_msg_##type(priv))				\
4238
		dynamic_netdev_dbg(netdev, format, ##args);	\
4239
} while (0)
4240 4241 4242
#elif defined(DEBUG)
#define netif_dbg(priv, type, dev, format, args...)		\
	netif_printk(priv, type, KERN_DEBUG, dev, format, ##args)
4243 4244 4245 4246 4247 4248 4249 4250 4251 4252
#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)
4253
#define netif_vdbg	netif_dbg
4254 4255 4256 4257
#else
#define netif_vdbg(priv, type, dev, format, args...)		\
({								\
	if (0)							\
4258
		netif_printk(priv, type, KERN_DEBUG, dev, format, ##args); \
4259 4260 4261
	0;							\
})
#endif
4262

4263 4264 4265 4266 4267 4268 4269 4270 4271 4272 4273 4274 4275 4276 4277 4278 4279 4280 4281 4282 4283 4284 4285 4286 4287 4288 4289 4290 4291 4292
/*
 *	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)

4293
#endif	/* _LINUX_NETDEVICE_H */