netdevice.h 128.8 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

55
struct netpoll_info;
56
struct device;
57
struct phy_device;
58 59
/* 802.11 specific */
struct wireless_dev;
60 61
/* 802.15.4 specific */
struct wpan_dev;
R
Robert Shearman 已提交
62
struct mpls_dev;
L
Linus Torvalds 已提交
63

64 65
void netdev_set_default_ethtool_ops(struct net_device *dev,
				    const struct ethtool_ops *ops);
66

67 68 69 70
/* Backlog congestion levels */
#define NET_RX_SUCCESS		0	/* keep 'em coming, baby */
#define NET_RX_DROP		1	/* packet dropped */

71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89
/*
 * 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
90 91 92 93
#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
94

95 96 97
/* 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. */
98
#define net_xmit_eval(e)	((e) == NET_XMIT_CN ? 0 : (e))
L
Linus Torvalds 已提交
99 100
#define net_xmit_errno(e)	((e) != NET_XMIT_CN ? -ENOBUFS : 0)

101
/* Driver transmit return codes */
102
#define NETDEV_TX_MASK		0xf0
103

104
enum netdev_tx {
105
	__NETDEV_TX_MIN	 = INT_MIN,	/* make sure enum is signed */
106 107 108
	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 */
109 110 111
};
typedef enum netdev_tx netdev_tx_t;

112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129
/*
 * 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
130 131 132 133
/*
 *	Compute the worst case header length according to the protocols
 *	used.
 */
G
Graf Yang 已提交
134

135 136 137
#if defined(CONFIG_HYPERV_NET)
# define LL_MAX_HEADER 128
#elif defined(CONFIG_WLAN) || IS_ENABLED(CONFIG_AX25)
138 139 140 141 142
# if defined(CONFIG_MAC80211_MESH)
#  define LL_MAX_HEADER 128
# else
#  define LL_MAX_HEADER 96
# endif
L
Linus Torvalds 已提交
143
#else
144
# define LL_MAX_HEADER 32
L
Linus Torvalds 已提交
145 146
#endif

147 148
#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
149 150 151 152 153 154
#define MAX_HEADER LL_MAX_HEADER
#else
#define MAX_HEADER (LL_MAX_HEADER + 48)
#endif

/*
155 156
 *	Old network device statistics. Fields are native words
 *	(unsigned long) so they can be read and written atomically.
L
Linus Torvalds 已提交
157
 */
G
Graf Yang 已提交
158

E
Eric Dumazet 已提交
159
struct net_device_stats {
160 161 162 163 164 165 166 167 168
	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
169 170
	unsigned long	collisions;
	unsigned long	rx_length_errors;
171 172 173 174 175
	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
176 177 178 179 180 181 182 183 184 185 186 187 188
	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>

189
#ifdef CONFIG_RPS
190 191
#include <linux/static_key.h>
extern struct static_key rps_needed;
192 193
#endif

L
Linus Torvalds 已提交
194 195 196 197
struct neighbour;
struct neigh_parms;
struct sk_buff;

198 199 200 201
struct netdev_hw_addr {
	struct list_head	list;
	unsigned char		addr[MAX_ADDR_LEN];
	unsigned char		type;
J
Jiri Pirko 已提交
202 203 204 205
#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
206 207
#define NETDEV_HW_ADDR_T_MULTICAST	5
	bool			global_use;
208
	int			sync_cnt;
209
	int			refcount;
210
	int			synced;
211 212 213
	struct rcu_head		rcu_head;
};

214 215 216 217 218
struct netdev_hw_addr_list {
	struct list_head	list;
	int			count;
};

219 220 221 222
#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)
223

224 225 226 227
#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)
228

229 230
#define netdev_mc_count(dev) netdev_hw_addr_list_count(&(dev)->mc)
#define netdev_mc_empty(dev) netdev_hw_addr_list_empty(&(dev)->mc)
231
#define netdev_for_each_mc_addr(ha, dev) \
232
	netdev_hw_addr_list_for_each(ha, &(dev)->mc)
233

E
Eric Dumazet 已提交
234
struct hh_cache {
235
	u16		hh_len;
236
	u16		__pad;
237
	seqlock_t	hh_lock;
L
Linus Torvalds 已提交
238 239 240 241

	/* cached hardware header; allow for machine alignment needs.        */
#define HH_DATA_MOD	16
#define HH_DATA_OFF(__len) \
J
Jiri Benc 已提交
242
	(HH_DATA_MOD - (((__len - 1) & (HH_DATA_MOD - 1)) + 1))
L
Linus Torvalds 已提交
243 244 245 246 247 248 249 250 251 252 253 254 255 256
#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) \
257
	((((dev)->hard_header_len+(dev)->needed_headroom)&~(HH_DATA_MOD - 1)) + HH_DATA_MOD)
L
Linus Torvalds 已提交
258
#define LL_RESERVED_SPACE_EXTRA(dev,extra) \
259
	((((dev)->hard_header_len+(dev)->needed_headroom+(extra))&~(HH_DATA_MOD - 1)) + HH_DATA_MOD)
L
Linus Torvalds 已提交
260

261 262 263
struct header_ops {
	int	(*create) (struct sk_buff *skb, struct net_device *dev,
			   unsigned short type, const void *daddr,
264
			   const void *saddr, unsigned int len);
265
	int	(*parse)(const struct sk_buff *skb, unsigned char *haddr);
266
	int	(*cache)(const struct neighbour *neigh, struct hh_cache *hh, __be16 type);
267 268 269 270 271
	void	(*cache_update)(struct hh_cache *hh,
				const struct net_device *dev,
				const unsigned char *haddr);
};

L
Linus Torvalds 已提交
272 273 274 275 276
/* 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
277
enum netdev_state_t {
L
Linus Torvalds 已提交
278 279 280
	__LINK_STATE_START,
	__LINK_STATE_PRESENT,
	__LINK_STATE_NOCARRIER,
S
Stefan Rompf 已提交
281 282
	__LINK_STATE_LINKWATCH_PENDING,
	__LINK_STATE_DORMANT,
L
Linus Torvalds 已提交
283 284 285 286 287
};


/*
 * This structure holds at boot time configured netdevice settings. They
G
Graf Yang 已提交
288
 * are then used in the device probing.
L
Linus Torvalds 已提交
289 290 291 292 293 294 295
 */
struct netdev_boot_setup {
	char name[IFNAMSIZ];
	struct ifmap map;
};
#define NETDEV_BOOT_SETUP_MAX 8

296
int __init netdev_boot_setup(char *str);
L
Linus Torvalds 已提交
297

298 299 300 301 302 303 304 305 306 307 308 309 310 311
/*
 * 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;
312
	unsigned int		gro_count;
313 314 315 316 317
	int			(*poll)(struct napi_struct *, int);
#ifdef CONFIG_NETPOLL
	spinlock_t		poll_lock;
	int			poll_owner;
#endif
H
Herbert Xu 已提交
318
	struct net_device	*dev;
319
	struct sk_buff		*gro_list;
H
Herbert Xu 已提交
320
	struct sk_buff		*skb;
321
	struct hrtimer		timer;
322
	struct list_head	dev_list;
E
Eliezer Tamir 已提交
323 324
	struct hlist_node	napi_hash_node;
	unsigned int		napi_id;
325 326
};

E
Eric Dumazet 已提交
327
enum {
328
	NAPI_STATE_SCHED,	/* Poll is scheduled */
D
David S. Miller 已提交
329
	NAPI_STATE_DISABLE,	/* Disable pending */
330
	NAPI_STATE_NPSVC,	/* Netpoll - don't dequeue from poll_list */
E
Eric Dumazet 已提交
331 332
	NAPI_STATE_HASHED,	/* In NAPI hash (busy polling possible) */
	NAPI_STATE_NO_BUSY_POLL,/* Do not add in napi_hash, no busy polling */
333 334
};

335
enum gro_result {
336 337 338 339 340 341
	GRO_MERGED,
	GRO_MERGED_FREE,
	GRO_HELD,
	GRO_NORMAL,
	GRO_DROP,
};
342
typedef enum gro_result gro_result_t;
343

344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375
/*
 * 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
376
 * are registered on exact device (ptype->dev == skb->dev).
377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392
 *
 * 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);
393

394
void __napi_schedule(struct napi_struct *n);
395
void __napi_schedule_irqoff(struct napi_struct *n);
396

397
static inline bool napi_disable_pending(struct napi_struct *n)
D
David S. Miller 已提交
398 399 400 401
{
	return test_bit(NAPI_STATE_DISABLE, &n->state);
}

402 403 404 405 406 407
/**
 *	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
408 409
 * insure only one NAPI poll instance runs.  We also make
 * sure there is no pending NAPI disable.
410
 */
411
static inline bool napi_schedule_prep(struct napi_struct *n)
412
{
D
David S. Miller 已提交
413 414
	return !napi_disable_pending(n) &&
		!test_and_set_bit(NAPI_STATE_SCHED, &n->state);
415 416 417 418 419 420 421 422 423 424 425 426 427 428 429
}

/**
 *	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);
}

430 431 432 433 434 435 436 437 438 439 440 441
/**
 *	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);
}

442
/* Try to reschedule poll. Called by dev->poll() after napi_complete().  */
443
static inline bool napi_reschedule(struct napi_struct *napi)
444 445 446
{
	if (napi_schedule_prep(napi)) {
		__napi_schedule(napi);
447
		return true;
448
	}
449
	return false;
450 451
}

452 453
void __napi_complete(struct napi_struct *n);
void napi_complete_done(struct napi_struct *n, int work_done);
454 455 456 457 458
/**
 *	napi_complete - NAPI processing complete
 *	@n: napi context
 *
 * Mark NAPI processing as complete.
459
 * Consider using napi_complete_done() instead.
460
 */
461 462 463 464
static inline void napi_complete(struct napi_struct *n)
{
	return napi_complete_done(n, 0);
}
465

E
Eliezer Tamir 已提交
466 467 468 469 470
/**
 *	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
471 472 473
 * 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
474
 */
475
void napi_hash_add(struct napi_struct *napi);
E
Eliezer Tamir 已提交
476 477 478 479 480 481

/**
 *	napi_hash_del - remove a NAPI from global table
 *	@napi: napi context
 *
 * Warning: caller must observe rcu grace period
482 483
 * before freeing memory containing @napi, if
 * this function returns true.
484 485 486 487
 * 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
488
 */
489
bool napi_hash_del(struct napi_struct *napi);
E
Eliezer Tamir 已提交
490

491 492 493 494 495 496 497
/**
 *	napi_disable - prevent NAPI from scheduling
 *	@n: napi context
 *
 * Stop NAPI from being scheduled on this context.
 * Waits till any outstanding processing completes.
 */
498
void napi_disable(struct napi_struct *n);
499 500 501 502 503 504 505 506 507 508 509

/**
 *	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));
510
	smp_mb__before_atomic();
511
	clear_bit(NAPI_STATE_SCHED, &n->state);
512
	clear_bit(NAPI_STATE_NPSVC, &n->state);
513 514
}

515 516 517 518 519 520 521 522 523 524
/**
 *	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)
{
525 526 527 528 529
	if (IS_ENABLED(CONFIG_SMP))
		while (test_bit(NAPI_STATE_SCHED, &n->state))
			msleep(1);
	else
		barrier();
530 531
}

E
Eric Dumazet 已提交
532
enum netdev_queue_state_t {
533 534
	__QUEUE_STATE_DRV_XOFF,
	__QUEUE_STATE_STACK_XOFF,
535
	__QUEUE_STATE_FROZEN,
536
};
537 538 539 540 541 542 543 544 545 546 547

#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)

548 549 550 551 552 553 554 555 556
/*
 * __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_*.
 */
557

558
struct netdev_queue {
559 560 561
/*
 * read mostly part
 */
562
	struct net_device	*dev;
563
	struct Qdisc __rcu	*qdisc;
564
	struct Qdisc		*qdisc_sleeping;
565
#ifdef CONFIG_SYSFS
T
Tom Herbert 已提交
566 567
	struct kobject		kobj;
#endif
568 569 570
#if defined(CONFIG_XPS) && defined(CONFIG_NUMA)
	int			numa_node;
#endif
571 572 573 574 575
/*
 * write mostly part
 */
	spinlock_t		_xmit_lock ____cacheline_aligned_in_smp;
	int			xmit_lock_owner;
576 577 578 579
	/*
	 * please use this field instead of dev->trans_start
	 */
	unsigned long		trans_start;
580 581 582 583 584 585

	/*
	 * Number of TX timeouts for this queue
	 * (/sys/class/net/DEV/Q/trans_timeout)
	 */
	unsigned long		trans_timeout;
T
Tom Herbert 已提交
586 587 588 589 590 591

	unsigned long		state;

#ifdef CONFIG_BQL
	struct dql		dql;
#endif
592
	unsigned long		tx_maxrate;
593
} ____cacheline_aligned_in_smp;
594

595 596 597 598 599
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
600
	return NUMA_NO_NODE;
601 602 603 604 605 606 607 608 609 610
#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
611
#ifdef CONFIG_RPS
T
Tom Herbert 已提交
612 613 614 615 616 617 618 619 620
/*
 * 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];
};
621
#define RPS_MAP_SIZE(_num) (sizeof(struct rps_map) + ((_num) * sizeof(u16)))
T
Tom Herbert 已提交
622

T
Tom Herbert 已提交
623
/*
624 625 626
 * 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
627 628 629
 */
struct rps_dev_flow {
	u16 cpu;
630
	u16 filter;
T
Tom Herbert 已提交
631 632
	unsigned int last_qtail;
};
633
#define RPS_NO_FILTER 0xffff
T
Tom Herbert 已提交
634 635 636 637 638 639 640 641 642 643

/*
 * 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) + \
644
    ((_num) * sizeof(struct rps_dev_flow)))
T
Tom Herbert 已提交
645 646 647 648

/*
 * The rps_sock_flow_table contains mappings of flows to the last CPU
 * on which they were processed by the application (set in recvmsg).
649 650 651 652 653 654
 * 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
655 656
 */
struct rps_sock_flow_table {
657
	u32	mask;
658 659

	u32	ents[0] ____cacheline_aligned_in_smp;
T
Tom Herbert 已提交
660
};
661
#define	RPS_SOCK_FLOW_TABLE_SIZE(_num) (offsetof(struct rps_sock_flow_table, ents[_num]))
T
Tom Herbert 已提交
662 663 664

#define RPS_NO_CPU 0xffff

665 666 667
extern u32 rps_cpu_mask;
extern struct rps_sock_flow_table __rcu *rps_sock_flow_table;

T
Tom Herbert 已提交
668 669 670 671
static inline void rps_record_sock_flow(struct rps_sock_flow_table *table,
					u32 hash)
{
	if (table && hash) {
672 673
		unsigned int index = hash & table->mask;
		u32 val = hash & ~rps_cpu_mask;
T
Tom Herbert 已提交
674 675

		/* We only give a hint, preemption can change cpu under us */
676
		val |= raw_smp_processor_id();
T
Tom Herbert 已提交
677

678 679
		if (table->ents[index] != val)
			table->ents[index] = val;
T
Tom Herbert 已提交
680 681 682
	}
}

683
#ifdef CONFIG_RFS_ACCEL
684 685
bool rps_may_expire_flow(struct net_device *dev, u16 rxq_index, u32 flow_id,
			 u16 filter_id);
686
#endif
687
#endif /* CONFIG_RPS */
688

T
Tom Herbert 已提交
689 690
/* This structure contains an instance of an RX queue. */
struct netdev_rx_queue {
691
#ifdef CONFIG_RPS
E
Eric Dumazet 已提交
692 693
	struct rps_map __rcu		*rps_map;
	struct rps_dev_flow_table __rcu	*rps_flow_table;
694
#endif
E
Eric Dumazet 已提交
695
	struct kobject			kobj;
T
Tom Herbert 已提交
696
	struct net_device		*dev;
T
Tom Herbert 已提交
697
} ____cacheline_aligned_in_smp;
698 699 700 701 702 703 704 705 706 707 708

/*
 * 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);
};
709

T
Tom Herbert 已提交
710 711 712 713 714 715 716 717 718 719 720
#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];
};
721
#define XPS_MAP_SIZE(_num) (sizeof(struct xps_map) + ((_num) * sizeof(u16)))
722 723
#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
724 725 726 727 728 729

/*
 * 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
730
	struct xps_map __rcu *cpu_map[0];
T
Tom Herbert 已提交
731 732 733 734 735
};
#define XPS_DEV_MAPS_SIZE (sizeof(struct xps_dev_maps) +		\
    (nr_cpu_ids * sizeof(struct xps_map *)))
#endif /* CONFIG_XPS */

736 737 738 739 740 741 742 743
#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;
};

744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760
#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

761
#define MAX_PHYS_ITEM_ID_LEN 32
762

763 764
/* This structure holds a unique identifier to identify some
 * physical item (port for example) used by a netdevice.
765
 */
766 767
struct netdev_phys_item_id {
	unsigned char id[MAX_PHYS_ITEM_ID_LEN];
768 769 770
	unsigned char id_len;
};

771 772 773 774 775 776 777
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;
}

778 779 780
typedef u16 (*select_queue_fallback_t)(struct net_device *dev,
				       struct sk_buff *skb);

781 782
/*
 * This structure defines the management hooks for network devices.
783 784
 * The following hooks can be defined; unless noted otherwise, they are
 * optional and can be filled with a null pointer.
785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803
 *
 * 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.
 *
804 805
 * netdev_tx_t (*ndo_start_xmit)(struct sk_buff *skb,
 *                               struct net_device *dev);
806
 *	Called when a packet needs to be transmitted.
807 808 809 810
 *	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.
811
 *        (can also return NETDEV_TX_LOCKED iff NETIF_F_LLTX)
812 813
 *	Required can not be NULL.
 *
814 815 816 817 818 819
 * 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.
 *
820
 * u16 (*ndo_select_queue)(struct net_device *dev, struct sk_buff *skb,
821
 *                         void *accel_priv, select_queue_fallback_t fallback);
822 823 824
 *	Called to decide which queue to when device supports multiple
 *	transmit queues.
 *
825 826 827 828 829 830
 * 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.
831 832
 *	If driver handles unicast address filtering, it should set
 *	IFF_UNICAST_FLT to its priv_flags.
833 834 835
 *
 * int (*ndo_set_mac_address)(struct net_device *dev, void *addr);
 *	This function  is called when the Media Access Control address
836
 *	needs to be changed. If this interface is not defined, the
837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856
 *	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.
 *
857
 * void (*ndo_tx_timeout)(struct net_device *dev);
858 859 860
 *	Callback uses when the transmitter has not made any progress
 *	for dev->watchdog ticks.
 *
861
 * struct rtnl_link_stats64* (*ndo_get_stats64)(struct net_device *dev,
862
 *                      struct rtnl_link_stats64 *storage);
863
 * struct net_device_stats* (*ndo_get_stats)(struct net_device *dev);
864
 *	Called when a user wants to get the network device usage
865
 *	statistics. Drivers must do one of the following:
866 867
 *	1. Define @ndo_get_stats64 to fill in a zero-initialised
 *	   rtnl_link_stats64 structure passed by the caller.
868
 *	2. Define @ndo_get_stats to update a net_device_stats structure
869 870 871 872 873
 *	   (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.
874
 *
875
 * int (*ndo_vlan_rx_add_vid)(struct net_device *dev, __be16 proto, u16 vid);
876 877
 *	If device support VLAN filtering this function is called when a
 *	VLAN id is registered.
878
 *
879
 * int (*ndo_vlan_rx_kill_vid)(struct net_device *dev, __be16 proto, u16 vid);
880 881
 *	If device support VLAN filtering this function is called when a
 *	VLAN id is unregistered.
882 883
 *
 * void (*ndo_poll_controller)(struct net_device *dev);
884 885 886 887
 *
 *	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);
888 889
 * int (*ndo_set_vf_rate)(struct net_device *dev, int vf, int min_tx_rate,
 *			  int max_tx_rate);
890
 * int (*ndo_set_vf_spoofchk)(struct net_device *dev, int vf, bool setting);
H
Hiroshi Shimamoto 已提交
891
 * int (*ndo_set_vf_trust)(struct net_device *dev, int vf, bool setting);
892 893
 * int (*ndo_get_vf_config)(struct net_device *dev,
 *			    int vf, struct ifla_vf_info *ivf);
894
 * int (*ndo_set_vf_link_state)(struct net_device *dev, int vf, int link_state);
895 896
 * int (*ndo_set_vf_port)(struct net_device *dev, int vf,
 *			  struct nlattr *port[]);
897 898 899 900 901
 *
 *      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);
902
 * int (*ndo_get_vf_port)(struct net_device *dev, int vf, struct sk_buff *skb);
903 904 905 906 907
 * 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.
908
 *
909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938
 *	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.
 *
939 940 941 942 943 944 945
 * 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.
 *
946 947 948 949 950 951
 * 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.
 *
952 953 954 955 956 957
 *	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.
958
 *
959
 *	Slave management functions (for bridge, bonding, etc).
960 961 962 963 964
 * 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.
965 966
 *
 *      Feature/offload setting functions.
967
 * int (*ndo_set_features)(struct net_device *dev, netdev_features_t features);
968 969 970 971
 *	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.
 *
972 973
 * int (*ndo_fdb_add)(struct ndmsg *ndm, struct nlattr *tb[],
 *		      struct net_device *dev,
974
 *		      const unsigned char *addr, u16 vid, u16 flags)
975
 *	Adds an FDB entry to dev for addr.
976 977
 * int (*ndo_fdb_del)(struct ndmsg *ndm, struct nlattr *tb[],
 *		      struct net_device *dev,
978
 *		      const unsigned char *addr, u16 vid)
979 980
 *	Deletes the FDB entry from dev coresponding to addr.
 * int (*ndo_fdb_dump)(struct sk_buff *skb, struct netlink_callback *cb,
981 982
 *		       struct net_device *dev, struct net_device *filter_dev,
 *		       int idx)
983 984
 *	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
985
 *
986 987
 * int (*ndo_bridge_setlink)(struct net_device *dev, struct nlmsghdr *nlh,
 *			     u16 flags)
J
John Fastabend 已提交
988
 * int (*ndo_bridge_getlink)(struct sk_buff *skb, u32 pid, u32 seq,
989 990
 *			     struct net_device *dev, u32 filter_mask,
 *			     int nlflags)
991 992
 * int (*ndo_bridge_dellink)(struct net_device *dev, struct nlmsghdr *nlh,
 *			     u16 flags);
J
Jiri Pirko 已提交
993 994 995 996 997 998 999 1000
 *
 * 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.
1001 1002
 *
 * int (*ndo_get_phys_port_id)(struct net_device *dev,
1003
 *			       struct netdev_phys_item_id *ppid);
1004 1005 1006
 *	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.
1007 1008
 *
 * void (*ndo_add_vxlan_port)(struct  net_device *dev,
J
Joseph Gasparakis 已提交
1009
 *			      sa_family_t sa_family, __be16 port);
1010 1011 1012 1013 1014
 *	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.
 *
1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027
 * 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.
 *
1028
 * void (*ndo_del_vxlan_port)(struct  net_device *dev,
J
Joseph Gasparakis 已提交
1029
 *			      sa_family_t sa_family, __be16 port);
1030 1031 1032
 *	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.
1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051
 *
 * 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.
1052 1053 1054
 * netdev_features_t (*ndo_features_check) (struct sk_buff *skb,
 *					    struct net_device *dev
 *					    netdev_features_t features);
T
Tom Herbert 已提交
1055
 *	Called by core transmit path to determine if device is capable of
1056 1057 1058 1059 1060
 *	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.
1061 1062 1063 1064
 * 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.
1065 1066
 * int (*ndo_get_iflink)(const struct net_device *dev);
 *	Called to get the iflink value of this device.
1067 1068 1069 1070 1071
 * 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.
1072 1073 1074 1075
 * 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.
1076
 *
1077 1078 1079 1080 1081 1082
 */
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);
1083 1084 1085 1086 1087
	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);
1088
	u16			(*ndo_select_queue)(struct net_device *dev,
1089
						    struct sk_buff *skb,
1090 1091
						    void *accel_priv,
						    select_queue_fallback_t fallback);
1092 1093 1094 1095 1096 1097 1098 1099 1100 1101
	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);
1102 1103 1104 1105
	int			(*ndo_change_mtu)(struct net_device *dev,
						  int new_mtu);
	int			(*ndo_neigh_setup)(struct net_device *dev,
						   struct neigh_parms *);
1106 1107
	void			(*ndo_tx_timeout) (struct net_device *dev);

1108 1109
	struct rtnl_link_stats64* (*ndo_get_stats64)(struct net_device *dev,
						     struct rtnl_link_stats64 *storage);
1110 1111
	struct net_device_stats* (*ndo_get_stats)(struct net_device *dev);

1112
	int			(*ndo_vlan_rx_add_vid)(struct net_device *dev,
1113
						       __be16 proto, u16 vid);
1114
	int			(*ndo_vlan_rx_kill_vid)(struct net_device *dev,
1115
						        __be16 proto, u16 vid);
1116 1117
#ifdef CONFIG_NET_POLL_CONTROLLER
	void                    (*ndo_poll_controller)(struct net_device *dev);
H
Herbert Xu 已提交
1118
	int			(*ndo_netpoll_setup)(struct net_device *dev,
1119
						     struct netpoll_info *info);
1120
	void			(*ndo_netpoll_cleanup)(struct net_device *dev);
E
Eliezer Tamir 已提交
1121
#endif
1122
#ifdef CONFIG_NET_RX_BUSY_POLL
1123
	int			(*ndo_busy_poll)(struct napi_struct *dev);
1124
#endif
1125 1126 1127 1128
	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);
1129 1130 1131
	int			(*ndo_set_vf_rate)(struct net_device *dev,
						   int vf, int min_tx_rate,
						   int max_tx_rate);
1132 1133
	int			(*ndo_set_vf_spoofchk)(struct net_device *dev,
						       int vf, bool setting);
H
Hiroshi Shimamoto 已提交
1134 1135
	int			(*ndo_set_vf_trust)(struct net_device *dev,
						    int vf, bool setting);
1136 1137 1138
	int			(*ndo_get_vf_config)(struct net_device *dev,
						     int vf,
						     struct ifla_vf_info *ivf);
1139 1140
	int			(*ndo_set_vf_link_state)(struct net_device *dev,
							 int vf, int link_state);
1141 1142 1143 1144
	int			(*ndo_get_vf_stats)(struct net_device *dev,
						    int vf,
						    struct ifla_vf_stats
						    *vf_stats);
1145 1146 1147 1148 1149
	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);
1150 1151 1152
	int			(*ndo_set_vf_rss_query_en)(
						   struct net_device *dev,
						   int vf, bool setting);
1153
	int			(*ndo_setup_tc)(struct net_device *dev, u8 tc);
1154
#if IS_ENABLED(CONFIG_FCOE)
1155 1156
	int			(*ndo_fcoe_enable)(struct net_device *dev);
	int			(*ndo_fcoe_disable)(struct net_device *dev);
1157 1158 1159 1160 1161 1162
	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);
1163 1164 1165 1166
	int			(*ndo_fcoe_ddp_target)(struct net_device *dev,
						       u16 xid,
						       struct scatterlist *sgl,
						       unsigned int sgc);
1167 1168
	int			(*ndo_fcoe_get_hbainfo)(struct net_device *dev,
							struct netdev_fcoe_hbainfo *hbainfo);
1169 1170
#endif

1171
#if IS_ENABLED(CONFIG_LIBFCOE)
1172 1173 1174 1175
#define NETDEV_FCOE_WWNN 0
#define NETDEV_FCOE_WWPN 1
	int			(*ndo_fcoe_get_wwn)(struct net_device *dev,
						    u64 *wwn, int type);
1176
#endif
1177

1178 1179 1180 1181 1182 1183
#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
1184 1185 1186 1187
	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);
1188 1189
	netdev_features_t	(*ndo_fix_features)(struct net_device *dev,
						    netdev_features_t features);
1190
	int			(*ndo_set_features)(struct net_device *dev,
1191
						    netdev_features_t features);
1192
	int			(*ndo_neigh_construct)(struct neighbour *n);
1193
	void			(*ndo_neigh_destroy)(struct neighbour *n);
1194 1195

	int			(*ndo_fdb_add)(struct ndmsg *ndm,
1196
					       struct nlattr *tb[],
1197
					       struct net_device *dev,
1198
					       const unsigned char *addr,
1199
					       u16 vid,
1200 1201
					       u16 flags);
	int			(*ndo_fdb_del)(struct ndmsg *ndm,
1202
					       struct nlattr *tb[],
1203
					       struct net_device *dev,
1204 1205
					       const unsigned char *addr,
					       u16 vid);
1206 1207 1208
	int			(*ndo_fdb_dump)(struct sk_buff *skb,
						struct netlink_callback *cb,
						struct net_device *dev,
1209
						struct net_device *filter_dev,
1210
						int idx);
J
John Fastabend 已提交
1211 1212

	int			(*ndo_bridge_setlink)(struct net_device *dev,
1213 1214
						      struct nlmsghdr *nlh,
						      u16 flags);
J
John Fastabend 已提交
1215 1216
	int			(*ndo_bridge_getlink)(struct sk_buff *skb,
						      u32 pid, u32 seq,
1217
						      struct net_device *dev,
1218 1219
						      u32 filter_mask,
						      int nlflags);
1220
	int			(*ndo_bridge_dellink)(struct net_device *dev,
1221 1222
						      struct nlmsghdr *nlh,
						      u16 flags);
J
Jiri Pirko 已提交
1223 1224
	int			(*ndo_change_carrier)(struct net_device *dev,
						      bool new_carrier);
1225
	int			(*ndo_get_phys_port_id)(struct net_device *dev,
1226
							struct netdev_phys_item_id *ppid);
1227 1228
	int			(*ndo_get_phys_port_name)(struct net_device *dev,
							  char *name, size_t len);
1229 1230
	void			(*ndo_add_vxlan_port)(struct  net_device *dev,
						      sa_family_t sa_family,
J
Joseph Gasparakis 已提交
1231
						      __be16 port);
1232 1233
	void			(*ndo_del_vxlan_port)(struct  net_device *dev,
						      sa_family_t sa_family,
J
Joseph Gasparakis 已提交
1234
						      __be16 port);
1235 1236 1237 1238 1239 1240
	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);
1241 1242 1243 1244 1245 1246 1247 1248
	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);
1249
	int			(*ndo_get_lock_subclass)(struct net_device *dev);
1250 1251 1252
	int			(*ndo_set_tx_maxrate)(struct net_device *dev,
						      int queue_index,
						      u32 maxrate);
1253
	int			(*ndo_get_iflink)(const struct net_device *dev);
1254 1255
	int			(*ndo_change_proto_down)(struct net_device *dev,
							 bool proto_down);
1256 1257
	int			(*ndo_fill_metadata_dst)(struct net_device *dev,
						       struct sk_buff *skb);
1258 1259
};

1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288
/**
 * 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
1289
 * @IFF_L3MDEV_MASTER: device is an L3 master device
1290
 * @IFF_NO_QUEUE: device can run without qdisc attached
1291
 * @IFF_OPENVSWITCH: device is a Open vSwitch master
D
David Ahern 已提交
1292
 * @IFF_L3MDEV_SLAVE: device is enslaved to an L3 master device
J
Jiri Pirko 已提交
1293
 * @IFF_TEAM: device is a team device
1294 1295 1296 1297
 */
enum netdev_priv_flags {
	IFF_802_1Q_VLAN			= 1<<0,
	IFF_EBRIDGE			= 1<<1,
1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315
	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,
1316
	IFF_L3MDEV_MASTER		= 1<<20,
1317 1318
	IFF_NO_QUEUE			= 1<<21,
	IFF_OPENVSWITCH			= 1<<22,
D
David Ahern 已提交
1319
	IFF_L3MDEV_SLAVE		= 1<<23,
J
Jiri Pirko 已提交
1320
	IFF_TEAM			= 1<<24,
1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339
};

#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
1340
#define IFF_XMIT_DST_RELEASE_PERM	IFF_XMIT_DST_RELEASE_PERM
1341 1342
#define IFF_IPVLAN_MASTER		IFF_IPVLAN_MASTER
#define IFF_IPVLAN_SLAVE		IFF_IPVLAN_SLAVE
1343
#define IFF_L3MDEV_MASTER		IFF_L3MDEV_MASTER
1344
#define IFF_NO_QUEUE			IFF_NO_QUEUE
1345
#define IFF_OPENVSWITCH			IFF_OPENVSWITCH
1346
#define IFF_L3MDEV_SLAVE		IFF_L3MDEV_SLAVE
J
Jiri Pirko 已提交
1347
#define IFF_TEAM			IFF_TEAM
1348

1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365
/**
 *	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
 *
1366 1367
 *	@carrier_changes:	Stats to monitor carrier on<->off transitions
 *
1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390
 *	@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
1391
 *	@group:		The group, that the device belongs to
1392 1393 1394 1395 1396 1397 1398 1399
 *
 *	@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
1400 1401
 *	@rx_nohandler:	nohandler dropped packets by core network on
 *			inactive devices, do not use this in drivers
1402 1403 1404 1405 1406 1407 1408 1409 1410
 *
 *	@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
1411
 *	@header_ops:	Includes callbacks for creating,parsing,caching,etc
1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424
 *			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
1425 1426
 *	@hard_header_len: Hardware header length, which means that this is the
 *			  minimum size of a packet.
1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449
 *
 *	@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()
1450 1451 1452 1453
 *	@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
1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482
 *	@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
 *
1483 1484 1485 1486 1487 1488
 *	@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
 *
1489 1490 1491 1492 1493 1494 1495 1496 1497
 *	@_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
 *
1498 1499
 *	@offload_fwd_mark:	Offload device fwding mark
 *
1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537
 *	@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
1538 1539
 *	@gso_min_segs:	Minimum number of segments that can be passed to the
 *			NIC for GSO
1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553
 *
 *	@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
 *
1554 1555 1556 1557
 *	@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
1558 1559 1560 1561
 *	FIXME: cleanup struct net_device such that network protocol info
 *	moves out.
 */

E
Eric Dumazet 已提交
1562
struct net_device {
L
Linus Torvalds 已提交
1563
	char			name[IFNAMSIZ];
1564
	struct hlist_node	name_hlist;
1565
	char 			*ifalias;
L
Linus Torvalds 已提交
1566 1567 1568 1569
	/*
	 *	I/O specific fields
	 *	FIXME: Merge these and struct ifmap into one
	 */
1570 1571 1572 1573
	unsigned long		mem_end;
	unsigned long		mem_start;
	unsigned long		base_addr;
	int			irq;
L
Linus Torvalds 已提交
1574

1575 1576
	atomic_t		carrier_changes;

L
Linus Torvalds 已提交
1577
	/*
1578 1579
	 *	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
1580 1581 1582 1583 1584
	 *	part of the usual set specified in Space.c.
	 */

	unsigned long		state;

1585
	struct list_head	dev_list;
1586
	struct list_head	napi_list;
1587
	struct list_head	unreg_list;
1588
	struct list_head	close_list;
1589 1590
	struct list_head	ptype_all;
	struct list_head	ptype_specific;
1591 1592 1593 1594 1595 1596 1597 1598 1599 1600

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

	struct {
		struct list_head upper;
		struct list_head lower;
	} all_adj_list;
1601

1602 1603 1604 1605
	netdev_features_t	features;
	netdev_features_t	hw_features;
	netdev_features_t	wanted_features;
	netdev_features_t	vlan_features;
1606
	netdev_features_t	hw_enc_features;
S
Simon Horman 已提交
1607
	netdev_features_t	mpls_features;
1608

L
Linus Torvalds 已提交
1609
	int			ifindex;
1610
	int			group;
L
Linus Torvalds 已提交
1611

R
Rusty Russell 已提交
1612
	struct net_device_stats	stats;
1613 1614 1615

	atomic_long_t		rx_dropped;
	atomic_long_t		tx_dropped;
1616
	atomic_long_t		rx_nohandler;
L
Linus Torvalds 已提交
1617

1618
#ifdef CONFIG_WIRELESS_EXT
L
Linus Torvalds 已提交
1619 1620
	const struct iw_handler_def *	wireless_handlers;
	struct iw_public_data *	wireless_data;
1621
#endif
1622
	const struct net_device_ops *netdev_ops;
1623
	const struct ethtool_ops *ethtool_ops;
S
Scott Feldman 已提交
1624
#ifdef CONFIG_NET_SWITCHDEV
J
Jiri Pirko 已提交
1625
	const struct switchdev_ops *switchdev_ops;
S
Scott Feldman 已提交
1626
#endif
1627 1628 1629
#ifdef CONFIG_NET_L3_MASTER_DEV
	const struct l3mdev_ops	*l3mdev_ops;
#endif
L
Linus Torvalds 已提交
1630

1631 1632
	const struct header_ops *header_ops;

1633 1634 1635
	unsigned int		flags;
	unsigned int		priv_flags;

L
Linus Torvalds 已提交
1636
	unsigned short		gflags;
1637
	unsigned short		padded;
L
Linus Torvalds 已提交
1638

1639 1640
	unsigned char		operstate;
	unsigned char		link_mode;
S
Stefan Rompf 已提交
1641

1642 1643
	unsigned char		if_port;
	unsigned char		dma;
1644

1645 1646 1647
	unsigned int		mtu;
	unsigned short		type;
	unsigned short		hard_header_len;
L
Linus Torvalds 已提交
1648

1649 1650 1651
	unsigned short		needed_headroom;
	unsigned short		needed_tailroom;

L
Linus Torvalds 已提交
1652
	/* Interface address info. */
1653 1654 1655
	unsigned char		perm_addr[MAX_ADDR_LEN];
	unsigned char		addr_assign_type;
	unsigned char		addr_len;
1656
	unsigned short		neigh_priv_len;
1657 1658
	unsigned short          dev_id;
	unsigned short          dev_port;
J
Jiri Pirko 已提交
1659
	spinlock_t		addr_list_lock;
1660 1661
	unsigned char		name_assign_type;
	bool			uc_promisc;
1662 1663 1664 1665
	struct netdev_hw_addr_list	uc;
	struct netdev_hw_addr_list	mc;
	struct netdev_hw_addr_list	dev_addrs;

1666 1667 1668
#ifdef CONFIG_SYSFS
	struct kset		*queues_kset;
#endif
1669 1670
	unsigned int		promiscuity;
	unsigned int		allmulti;
L
Linus Torvalds 已提交
1671 1672 1673


	/* Protocol specific pointers */
1674

1675
#if IS_ENABLED(CONFIG_VLAN_8021Q)
1676
	struct vlan_info __rcu	*vlan_info;
1677
#endif
1678
#if IS_ENABLED(CONFIG_NET_DSA)
1679
	struct dsa_switch_tree	*dsa_ptr;
1680 1681
#endif
#if IS_ENABLED(CONFIG_TIPC)
1682
	struct tipc_bearer __rcu *tipc_ptr;
1683
#endif
1684 1685 1686 1687 1688 1689
	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;
1690
	struct wpan_dev		*ieee802154_ptr;
R
Robert Shearman 已提交
1691 1692 1693
#if IS_ENABLED(CONFIG_MPLS_ROUTING)
	struct mpls_dev __rcu	*mpls_ptr;
#endif
L
Linus Torvalds 已提交
1694

1695
/*
E
Eric Dumazet 已提交
1696
 * Cache lines mostly used on receive path (including eth_type_trans())
1697
 */
1698
	unsigned long		last_rx;
1699

1700
	/* Interface address info used in eth_type_trans() */
1701
	unsigned char		*dev_addr;
1702

T
Tom Herbert 已提交
1703

1704
#ifdef CONFIG_SYSFS
T
Tom Herbert 已提交
1705 1706 1707
	struct netdev_rx_queue	*_rx;

	unsigned int		num_rx_queues;
1708
	unsigned int		real_num_rx_queues;
1709

E
Eric Dumazet 已提交
1710
#endif
T
Tom Herbert 已提交
1711

1712
	unsigned long		gro_flush_timeout;
1713 1714
	rx_handler_func_t __rcu	*rx_handler;
	void __rcu		*rx_handler_data;
1715

1716
#ifdef CONFIG_NET_CLS_ACT
1717 1718
	struct tcf_proto __rcu  *ingress_cl_list;
#endif
1719
	struct netdev_queue __rcu *ingress_queue;
1720 1721 1722
#ifdef CONFIG_NETFILTER_INGRESS
	struct list_head	nf_hooks_ingress;
#endif
1723

1724
	unsigned char		broadcast[MAX_ADDR_LEN];
1725 1726 1727 1728
#ifdef CONFIG_RFS_ACCEL
	struct cpu_rmap		*rx_cpu_rmap;
#endif
	struct hlist_node	index_hlist;
E
Eric Dumazet 已提交
1729 1730 1731 1732

/*
 * Cache lines mostly used on transmit path
 */
1733 1734
	struct netdev_queue	*_tx ____cacheline_aligned_in_smp;
	unsigned int		num_tx_queues;
1735
	unsigned int		real_num_tx_queues;
1736
	struct Qdisc		*qdisc;
1737
	unsigned long		tx_queue_len;
1738
	spinlock_t		tx_global_lock;
1739
	int			watchdog_timeo;
E
Eric Dumazet 已提交
1740

T
Tom Herbert 已提交
1741
#ifdef CONFIG_XPS
E
Eric Dumazet 已提交
1742
	struct xps_dev_maps __rcu *xps_maps;
T
Tom Herbert 已提交
1743
#endif
D
Daniel Borkmann 已提交
1744 1745 1746
#ifdef CONFIG_NET_CLS_ACT
	struct tcf_proto __rcu  *egress_cl_list;
#endif
1747 1748 1749 1750
#ifdef CONFIG_NET_SWITCHDEV
	u32			offload_fwd_mark;
#endif

1751
	/* These may be needed for future network-power-down code. */
1752 1753 1754 1755 1756

	/*
	 * trans_start here is expensive for high speed devices on SMP,
	 * please use netdev_queue->trans_start instead.
	 */
1757
	unsigned long		trans_start;
1758 1759 1760

	struct timer_list	watchdog_timer;

E
Eric Dumazet 已提交
1761
	int __percpu		*pcpu_refcnt;
L
Linus Torvalds 已提交
1762 1763
	struct list_head	todo_list;

1764
	struct list_head	link_watch_list;
1765

L
Linus Torvalds 已提交
1766
	enum { NETREG_UNINITIALIZED=0,
1767
	       NETREG_REGISTERED,	/* completed register_netdevice */
L
Linus Torvalds 已提交
1768 1769 1770
	       NETREG_UNREGISTERING,	/* called unregister_netdevice */
	       NETREG_UNREGISTERED,	/* completed unregister todo */
	       NETREG_RELEASED,		/* called free_netdev */
1771
	       NETREG_DUMMY,		/* dummy device for NAPI poll */
1772 1773
	} reg_state:8;

1774
	bool dismantle;
1775 1776 1777 1778 1779

	enum {
		RTNL_LINK_INITIALIZED,
		RTNL_LINK_INITIALIZING,
	} rtnl_link_state:16;
L
Linus Torvalds 已提交
1780

1781
	void (*destructor)(struct net_device *dev);
L
Linus Torvalds 已提交
1782 1783

#ifdef CONFIG_NETPOLL
1784
	struct netpoll_info __rcu	*npinfo;
L
Linus Torvalds 已提交
1785
#endif
1786

1787
	possible_net_t			nd_net;
1788

D
David S. Miller 已提交
1789
	/* mid-layer private */
E
Eric Dumazet 已提交
1790
	union {
1791 1792
		void					*ml_priv;
		struct pcpu_lstats __percpu		*lstats;
1793
		struct pcpu_sw_netstats __percpu	*tstats;
1794 1795
		struct pcpu_dstats __percpu		*dstats;
		struct pcpu_vstats __percpu		*vstats;
E
Eric Dumazet 已提交
1796
	};
1797

E
Eric Dumazet 已提交
1798
	struct garp_port __rcu	*garp_port;
1799
	struct mrp_port __rcu	*mrp_port;
L
Linus Torvalds 已提交
1800

1801
	struct device	dev;
1802
	const struct attribute_group *sysfs_groups[4];
1803
	const struct attribute_group *sysfs_rx_queue_group;
P
Patrick McHardy 已提交
1804 1805

	const struct rtnl_link_ops *rtnl_link_ops;
1806

1807 1808 1809
	/* for setting kernel sock attribute on TCP connection setup */
#define GSO_MAX_SIZE		65536
	unsigned int		gso_max_size;
1810 1811
#define GSO_MAX_SEGS		65535
	u16			gso_max_segs;
1812
	u16			gso_min_segs;
J
Jeff Kirsher 已提交
1813
#ifdef CONFIG_DCB
1814
	const struct dcbnl_rtnl_ops *dcbnl_ops;
1815
#endif
1816 1817 1818
	u8 num_tc;
	struct netdev_tc_txq tc_to_txq[TC_MAX_QUEUE];
	u8 prio_tc_map[TC_BITMASK + 1];
1819

1820
#if IS_ENABLED(CONFIG_FCOE)
1821
	unsigned int		fcoe_ddp_xid;
1822
#endif
1823
#if IS_ENABLED(CONFIG_CGROUP_NET_PRIO)
1824
	struct netprio_map __rcu *priomap;
1825
#endif
1826
	struct phy_device *phydev;
1827
	struct lock_class_key *qdisc_tx_busylock;
1828
	bool proto_down;
L
Linus Torvalds 已提交
1829
};
1830
#define to_net_dev(d) container_of(d, struct net_device, dev)
L
Linus Torvalds 已提交
1831 1832 1833

#define	NETDEV_ALIGN		32

1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884
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;
}

1885 1886 1887 1888 1889 1890 1891
static inline
struct netdev_queue *netdev_get_tx_queue(const struct net_device *dev,
					 unsigned int index)
{
	return &dev->_tx[index];
}

1892 1893 1894 1895 1896 1897
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));
}

1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909
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);
}

1910
struct netdev_queue *netdev_pick_tx(struct net_device *dev,
1911 1912
				    struct sk_buff *skb,
				    void *accel_priv);
1913

1914 1915 1916 1917 1918 1919
/*
 * Net namespace inlines
 */
static inline
struct net *dev_net(const struct net_device *dev)
{
E
Eric Dumazet 已提交
1920
	return read_pnet(&dev->nd_net);
1921 1922 1923
}

static inline
1924
void dev_net_set(struct net_device *dev, struct net *net)
1925
{
1926
	write_pnet(&dev->nd_net, net);
1927 1928
}

1929
static inline bool netdev_uses_dsa(struct net_device *dev)
1930
{
1931
#if IS_ENABLED(CONFIG_NET_DSA)
1932 1933
	if (dev->dsa_ptr != NULL)
		return dsa_uses_tagged_protocol(dev->dsa_ptr);
1934
#endif
1935
	return false;
1936 1937
}

1938 1939 1940 1941 1942 1943
/**
 *	netdev_priv - access network device private data
 *	@dev: network device
 *
 * Get network device private data
 */
1944
static inline void *netdev_priv(const struct net_device *dev)
L
Linus Torvalds 已提交
1945
{
1946
	return (char *)dev + ALIGN(sizeof(struct net_device), NETDEV_ALIGN);
L
Linus Torvalds 已提交
1947 1948 1949 1950 1951
}

/* Set the sysfs physical device reference for the network logical device
 * if set prior to registration will cause a symlink during initialization.
 */
1952
#define SET_NETDEV_DEV(net, pdev)	((net)->dev.parent = (pdev))
L
Linus Torvalds 已提交
1953

1954
/* Set the sysfs device type for the network logical device to allow
1955
 * fine-grained identification of different network device types. For
1956 1957 1958 1959
 * example Ethernet, Wirelss LAN, Bluetooth, WiMAX etc.
 */
#define SET_NETDEV_DEVTYPE(net, devtype)	((net)->dev.type = (devtype))

E
Eric Dumazet 已提交
1960 1961 1962 1963 1964
/* Default NAPI poll() weight
 * Device drivers are strongly advised to not use bigger value
 */
#define NAPI_POLL_WEIGHT 64

1965 1966 1967 1968 1969 1970 1971 1972 1973 1974
/**
 *	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.
 */
1975 1976
void netif_napi_add(struct net_device *dev, struct napi_struct *napi,
		    int (*poll)(struct napi_struct *, int), int weight);
1977

E
Eric Dumazet 已提交
1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997
/**
 *	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);
}

1998 1999 2000 2001 2002 2003
/**
 *  netif_napi_del - remove a napi context
 *  @napi: napi context
 *
 *  netif_napi_del() removes a napi context from the network device napi list
 */
2004 2005 2006
void netif_napi_del(struct napi_struct *napi);

struct napi_gro_cb {
2007 2008 2009
	/* Virtual address of skb_shinfo(skb)->frags[0].page + offset. */
	void *frag0;

2010 2011 2012
	/* Length of frag0. */
	unsigned int frag0_len;

2013 2014 2015
	/* This indicates where we are processing relative to skb->data. */
	int data_offset;

2016
	/* This is non-zero if the packet cannot be merged with the new skb. */
2017 2018 2019 2020
	u16	flush;

	/* Save the IP ID here and check when we get to the transport layer */
	u16	flush_id;
2021 2022

	/* Number of segments aggregated. */
2023 2024
	u16	count;

2025 2026 2027
	/* Start offset for remote checksum offload */
	u16	gro_remcsum_start;

2028 2029
	/* jiffies when first packet was created/queued */
	unsigned long age;
2030

T
Tom Herbert 已提交
2031
	/* Used in ipv6_gro_receive() and foo-over-udp */
2032 2033
	u16	proto;

2034 2035 2036
	/* This is non-zero if the packet may be of the same flow. */
	u8	same_flow:1;

2037
	/* Used in udp_gro_receive */
T
Tom Herbert 已提交
2038 2039 2040 2041 2042
	u8	udp_mark:1;

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

2043 2044
	/* Number of checksums via CHECKSUM_UNNECESSARY */
	u8	csum_cnt:3;
2045

2046 2047 2048 2049 2050
	/* Free the skb? */
	u8	free:2;
#define NAPI_GRO_FREE		  1
#define NAPI_GRO_FREE_STOLEN_HEAD 2

2051 2052 2053
	/* Used in foo-over-udp, set in udp[46]_gro_receive */
	u8	is_ipv6:1;

2054 2055
	/* 7 bit hole */

2056 2057 2058
	/* used to support CHECKSUM_COMPLETE for tunneling protocols */
	__wsum	csum;

2059 2060
	/* used in skb_gro_receive() slow path */
	struct sk_buff *last;
2061 2062 2063
};

#define NAPI_GRO_CB(skb) ((struct napi_gro_cb *)(skb)->cb)
2064

L
Linus Torvalds 已提交
2065
struct packet_type {
D
David S. Miller 已提交
2066 2067 2068 2069 2070 2071
	__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 *);
2072 2073
	bool			(*id_match)(struct packet_type *ptype,
					    struct sock *sk);
L
Linus Torvalds 已提交
2074 2075 2076 2077
	void			*af_packet_priv;
	struct list_head	list;
};

2078
struct offload_callbacks {
2079
	struct sk_buff		*(*gso_segment)(struct sk_buff *skb,
2080
						netdev_features_t features);
2081
	struct sk_buff		**(*gro_receive)(struct sk_buff **head,
2082
						 struct sk_buff *skb);
2083
	int			(*gro_complete)(struct sk_buff *skb, int nhoff);
2084 2085 2086 2087
};

struct packet_offload {
	__be16			 type;	/* This is really htons(ether_type). */
2088
	u16			 priority;
2089 2090
	struct offload_callbacks callbacks;
	struct list_head	 list;
L
Linus Torvalds 已提交
2091 2092
};

2093 2094 2095 2096 2097 2098 2099 2100 2101 2102 2103
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);
};

2104 2105
struct udp_offload {
	__be16			 port;
T
Tom Herbert 已提交
2106
	u8			 ipproto;
2107
	struct udp_offload_callbacks callbacks;
2108 2109
};

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

2119 2120 2121 2122 2123 2124 2125 2126 2127 2128 2129 2130
#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;							\
2131 2132
})

2133
#define netdev_alloc_pcpu_stats(type)					\
2134
	__netdev_alloc_pcpu_stats(type, GFP_KERNEL)
2135

2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146 2147 2148
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;
};

2149 2150 2151 2152 2153
struct netdev_lag_lower_state_info {
	u8 link_up : 1,
	   tx_enabled : 1;
};

L
Linus Torvalds 已提交
2154 2155
#include <linux/notifier.h>

2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 2166 2167 2168
/* 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
2169
#define NETDEV_CHANGEMTU	0x0007 /* notify after mtu change happened */
2170 2171 2172 2173 2174 2175 2176 2177 2178
#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
2179
#define NETDEV_UNREGISTER_FINAL 0x0011
2180 2181 2182
#define NETDEV_RELEASE		0x0012
#define NETDEV_NOTIFY_PEERS	0x0013
#define NETDEV_JOIN		0x0014
2183
#define NETDEV_CHANGEUPPER	0x0015
2184
#define NETDEV_RESEND_IGMP	0x0016
2185
#define NETDEV_PRECHANGEMTU	0x0017 /* notify before mtu change happened */
2186
#define NETDEV_CHANGEINFODATA	0x0018
2187
#define NETDEV_BONDING_INFO	0x0019
2188
#define NETDEV_PRECHANGEUPPER	0x001A
2189
#define NETDEV_CHANGELOWERSTATE	0x001B
2190

2191 2192
int register_netdevice_notifier(struct notifier_block *nb);
int unregister_netdevice_notifier(struct notifier_block *nb);
2193 2194 2195 2196 2197

struct netdev_notifier_info {
	struct net_device *dev;
};

2198 2199 2200 2201 2202
struct netdev_notifier_change_info {
	struct netdev_notifier_info info; /* must be first */
	unsigned int flags_changed;
};

2203 2204 2205 2206 2207
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 */
2208
	void *upper_info; /* upper dev info */
2209 2210
};

2211 2212 2213 2214 2215
struct netdev_notifier_changelowerstate_info {
	struct netdev_notifier_info info; /* must be first */
	void *lower_state_info; /* is lower dev state */
};

2216 2217 2218 2219 2220 2221
static inline void netdev_notifier_info_init(struct netdev_notifier_info *info,
					     struct net_device *dev)
{
	info->dev = dev;
}

2222 2223 2224 2225 2226 2227
static inline struct net_device *
netdev_notifier_info_to_dev(const struct netdev_notifier_info *info)
{
	return info->dev;
}

2228
int call_netdevice_notifiers(unsigned long val, struct net_device *dev);
2229 2230


L
Linus Torvalds 已提交
2231 2232
extern rwlock_t				dev_base_lock;		/* Device list lock */

2233 2234
#define for_each_netdev(net, d)		\
		list_for_each_entry(d, &(net)->dev_base_head, dev_list)
2235 2236
#define for_each_netdev_reverse(net, d)	\
		list_for_each_entry_reverse(d, &(net)->dev_base_head, dev_list)
2237 2238
#define for_each_netdev_rcu(net, d)		\
		list_for_each_entry_rcu(d, &(net)->dev_base_head, dev_list)
2239 2240 2241 2242
#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)
2243 2244
#define for_each_netdev_continue_rcu(net, d)		\
	list_for_each_entry_continue_rcu(d, &(net)->dev_base_head, dev_list)
2245 2246
#define for_each_netdev_in_bond_rcu(bond, slave)	\
		for_each_netdev_rcu(&init_net, slave)	\
2247
			if (netdev_master_upper_dev_get_rcu(slave) == (bond))
2248
#define net_device_entry(lh)	list_entry(lh, struct net_device, dev_list)
2249

2250 2251 2252 2253 2254
static inline struct net_device *next_net_device(struct net_device *dev)
{
	struct list_head *lh;
	struct net *net;

2255
	net = dev_net(dev);
2256 2257 2258 2259
	lh = dev->dev_list.next;
	return lh == &net->dev_base_head ? NULL : net_device_entry(lh);
}

2260 2261 2262 2263 2264 2265
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
2266
	lh = rcu_dereference(list_next_rcu(&dev->dev_list));
2267 2268 2269
	return lh == &net->dev_base_head ? NULL : net_device_entry(lh);
}

2270 2271 2272 2273 2274
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);
}
2275

E
Eric Dumazet 已提交
2276 2277 2278 2279 2280 2281 2282
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);
}

2283 2284 2285 2286 2287 2288 2289 2290 2291 2292 2293 2294
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);

2295
int dev_get_iflink(const struct net_device *dev);
2296
int dev_fill_metadata_dst(struct net_device *dev, struct sk_buff *skb);
2297 2298
struct net_device *__dev_get_by_flags(struct net *net, unsigned short flags,
				      unsigned short mask);
2299 2300 2301 2302 2303 2304
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);
2305
int dev_close_many(struct list_head *head, bool unlink);
2306
void dev_disable_lro(struct net_device *dev);
2307
int dev_loopback_xmit(struct net *net, struct sock *sk, struct sk_buff *newskb);
2308
int dev_queue_xmit(struct sk_buff *skb);
2309
int dev_queue_xmit_accel(struct sk_buff *skb, void *accel_priv);
2310 2311 2312
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);
2313 2314 2315 2316 2317
static inline void unregister_netdevice(struct net_device *dev)
{
	unregister_netdevice_queue(dev, NULL);
}

2318 2319
int netdev_refcnt_read(const struct net_device *dev);
void free_netdev(struct net_device *dev);
2320
void netdev_freemem(struct net_device *dev);
2321 2322
void synchronize_net(void);
int init_dummy_netdev(struct net_device *dev);
2323

2324 2325 2326 2327 2328 2329
DECLARE_PER_CPU(int, xmit_recursion);
static inline int dev_recursion_level(void)
{
	return this_cpu_read(xmit_recursion);
}

2330 2331 2332 2333 2334 2335
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);
2336 2337 2338 2339 2340 2341 2342 2343 2344 2345 2346 2347 2348 2349 2350 2351

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;
}

2352 2353
static inline void *skb_gro_header_fast(struct sk_buff *skb,
					unsigned int offset)
2354
{
2355 2356
	return NAPI_GRO_CB(skb)->frag0 + offset;
}
2357

2358 2359 2360 2361
static inline int skb_gro_header_hard(struct sk_buff *skb, unsigned int hlen)
{
	return NAPI_GRO_CB(skb)->frag0_len < hlen;
}
2362

2363 2364 2365
static inline void *skb_gro_header_slow(struct sk_buff *skb, unsigned int hlen,
					unsigned int offset)
{
2366 2367 2368
	if (!pskb_may_pull(skb, hlen))
		return NULL;

2369 2370
	NAPI_GRO_CB(skb)->frag0 = NULL;
	NAPI_GRO_CB(skb)->frag0_len = 0;
2371
	return skb->data + offset;
2372
}
L
Linus Torvalds 已提交
2373

H
Herbert Xu 已提交
2374 2375
static inline void *skb_gro_network_header(struct sk_buff *skb)
{
2376 2377
	return (NAPI_GRO_CB(skb)->frag0 ?: skb->data) +
	       skb_network_offset(skb);
H
Herbert Xu 已提交
2378 2379
}

2380 2381 2382
static inline void skb_gro_postpull_rcsum(struct sk_buff *skb,
					const void *start, unsigned int len)
{
T
Tom Herbert 已提交
2383
	if (NAPI_GRO_CB(skb)->csum_valid)
2384 2385 2386 2387
		NAPI_GRO_CB(skb)->csum = csum_sub(NAPI_GRO_CB(skb)->csum,
						  csum_partial(start, len, 0));
}

T
Tom Herbert 已提交
2388 2389 2390 2391 2392 2393 2394
/* 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);

2395 2396
static inline bool skb_at_gro_remcsum_start(struct sk_buff *skb)
{
2397
	return (NAPI_GRO_CB(skb)->gro_remcsum_start == skb_gro_offset(skb));
2398 2399
}

T
Tom Herbert 已提交
2400 2401 2402 2403
static inline bool __skb_gro_checksum_validate_needed(struct sk_buff *skb,
						      bool zero_okay,
						      __sum16 check)
{
2404 2405 2406
	return ((skb->ip_summed != CHECKSUM_PARTIAL ||
		skb_checksum_start_offset(skb) <
		 skb_gro_offset(skb)) &&
2407
		!skb_at_gro_remcsum_start(skb) &&
2408
		NAPI_GRO_CB(skb)->csum_cnt == 0 &&
T
Tom Herbert 已提交
2409 2410 2411 2412 2413 2414 2415 2416 2417 2418 2419 2420 2421 2422 2423 2424 2425
		(!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)
{
2426 2427 2428 2429 2430 2431 2432 2433 2434
	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
2435 2436 2437 2438 2439 2440 2441 2442 2443 2444
	}
}

#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));		\
2445 2446 2447
	if (__ret)							\
		__skb_mark_checksum_bad(skb);				\
	else								\
T
Tom Herbert 已提交
2448 2449 2450 2451 2452 2453 2454 2455 2456 2457 2458 2459 2460 2461
		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)

2462 2463 2464 2465 2466 2467 2468 2469 2470 2471 2472 2473 2474 2475 2476 2477 2478 2479 2480 2481
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)

2482 2483 2484 2485 2486 2487 2488
struct gro_remcsum {
	int offset;
	__wsum delta;
};

static inline void skb_gro_remcsum_init(struct gro_remcsum *grc)
{
2489
	grc->offset = 0;
2490 2491 2492
	grc->delta = 0;
}

2493 2494 2495 2496 2497
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)
2498 2499
{
	__wsum delta;
2500
	size_t plen = hdrlen + max_t(size_t, offset + sizeof(u16), start);
2501 2502 2503

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

2504
	if (!nopartial) {
2505 2506 2507 2508 2509 2510 2511 2512 2513
		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;
2514 2515
	}

2516 2517
	delta = remcsum_adjust(ptr + hdrlen, NAPI_GRO_CB(skb)->csum,
			       start, offset);
2518 2519 2520

	/* Adjust skb->csum since we changed the packet */
	NAPI_GRO_CB(skb)->csum = csum_add(NAPI_GRO_CB(skb)->csum, delta);
2521

2522
	grc->offset = off + hdrlen + offset;
2523
	grc->delta = delta;
2524 2525

	return ptr;
2526 2527
}

2528 2529 2530
static inline void skb_gro_remcsum_cleanup(struct sk_buff *skb,
					   struct gro_remcsum *grc)
{
2531 2532 2533
	void *ptr;
	size_t plen = grc->offset + sizeof(u16);

2534 2535 2536
	if (!grc->delta)
		return;

2537 2538 2539 2540 2541 2542 2543 2544
	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);
2545
}
2546

2547 2548 2549 2550 2551 2552 2553 2554 2555 2556 2557 2558 2559 2560 2561 2562 2563 2564 2565 2566 2567 2568 2569 2570 2571 2572 2573 2574 2575 2576 2577 2578 2579 2580 2581 2582 2583 2584 2585 2586 2587 2588 2589 2590 2591 2592 2593 2594 2595 2596 2597 2598 2599 2600 2601 2602 2603 2604 2605 2606 2607 2608 2609 2610 2611
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);
}

2612 2613
static inline int dev_hard_header(struct sk_buff *skb, struct net_device *dev,
				  unsigned short type,
2614
				  const void *daddr, const void *saddr,
2615
				  unsigned int len)
2616
{
2617
	if (!dev->header_ops || !dev->header_ops->create)
2618
		return 0;
2619 2620

	return dev->header_ops->create(skb, dev, type, daddr, saddr, len);
2621 2622
}

S
Stephen Hemminger 已提交
2623 2624 2625 2626 2627
static inline int dev_parse_header(const struct sk_buff *skb,
				   unsigned char *haddr)
{
	const struct net_device *dev = skb->dev;

2628
	if (!dev->header_ops || !dev->header_ops->parse)
S
Stephen Hemminger 已提交
2629
		return 0;
2630
	return dev->header_ops->parse(skb, haddr);
S
Stephen Hemminger 已提交
2631 2632
}

L
Linus Torvalds 已提交
2633
typedef int gifconf_func_t(struct net_device * dev, char __user * bufptr, int len);
2634
int register_gifconf(unsigned int family, gifconf_func_t *gifconf);
L
Linus Torvalds 已提交
2635 2636 2637 2638 2639
static inline int unregister_gifconf(unsigned int family)
{
	return register_gifconf(family, NULL);
}

2640
#ifdef CONFIG_NET_FLOW_LIMIT
2641
#define FLOW_LIMIT_HISTORY	(1 << 7)  /* must be ^2 and !overflow buckets */
2642 2643 2644 2645 2646 2647 2648 2649 2650 2651 2652
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
2653
/*
E
Eric Dumazet 已提交
2654
 * Incoming packets are placed on per-cpu queues
L
Linus Torvalds 已提交
2655
 */
E
Eric Dumazet 已提交
2656
struct softnet_data {
L
Linus Torvalds 已提交
2657
	struct list_head	poll_list;
2658
	struct sk_buff_head	process_queue;
L
Linus Torvalds 已提交
2659

C
Changli Gao 已提交
2660
	/* stats */
2661 2662 2663 2664
	unsigned int		processed;
	unsigned int		time_squeeze;
	unsigned int		cpu_collision;
	unsigned int		received_rps;
2665
#ifdef CONFIG_RPS
E
Eric Dumazet 已提交
2666
	struct softnet_data	*rps_ipi_list;
E
Eric Dumazet 已提交
2667 2668 2669 2670 2671 2672 2673
#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
2674

E
Eric Dumazet 已提交
2675
#ifdef CONFIG_RPS
E
Eric Dumazet 已提交
2676
	/* Elements below can be accessed between CPUs for RPS */
T
Tom Herbert 已提交
2677
	struct call_single_data	csd ____cacheline_aligned_in_smp;
E
Eric Dumazet 已提交
2678 2679
	struct softnet_data	*rps_ipi_next;
	unsigned int		cpu;
T
Tom Herbert 已提交
2680
	unsigned int		input_queue_head;
2681
	unsigned int		input_queue_tail;
2682
#endif
2683
	unsigned int		dropped;
T
Tom Herbert 已提交
2684
	struct sk_buff_head	input_pkt_queue;
2685
	struct napi_struct	backlog;
2686

L
Linus Torvalds 已提交
2687 2688
};

2689
static inline void input_queue_head_incr(struct softnet_data *sd)
T
Tom Herbert 已提交
2690 2691
{
#ifdef CONFIG_RPS
2692 2693 2694 2695 2696 2697 2698 2699 2700
	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
2701 2702 2703
#endif
}

T
Tom Herbert 已提交
2704
DECLARE_PER_CPU_ALIGNED(struct softnet_data, softnet_data);
L
Linus Torvalds 已提交
2705

2706
void __netif_schedule(struct Qdisc *q);
2707
void netif_schedule_queue(struct netdev_queue *txq);
2708

2709 2710 2711 2712 2713 2714 2715 2716
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));
}

2717 2718
static inline void netif_tx_start_queue(struct netdev_queue *dev_queue)
{
2719
	clear_bit(__QUEUE_STATE_DRV_XOFF, &dev_queue->state);
2720 2721
}

2722 2723 2724 2725 2726 2727
/**
 *	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
2728 2729
static inline void netif_start_queue(struct net_device *dev)
{
2730
	netif_tx_start_queue(netdev_get_tx_queue(dev, 0));
L
Linus Torvalds 已提交
2731 2732
}

2733 2734 2735 2736 2737 2738 2739 2740 2741 2742
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);
	}
}

2743
void netif_tx_wake_queue(struct netdev_queue *dev_queue);
2744

2745 2746 2747 2748 2749 2750 2751
/**
 *	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.
 */
2752 2753
static inline void netif_wake_queue(struct net_device *dev)
{
2754
	netif_tx_wake_queue(netdev_get_tx_queue(dev, 0));
L
Linus Torvalds 已提交
2755 2756
}

2757 2758 2759 2760 2761 2762 2763 2764 2765 2766
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);
	}
}

2767 2768
static inline void netif_tx_stop_queue(struct netdev_queue *dev_queue)
{
2769
	set_bit(__QUEUE_STATE_DRV_XOFF, &dev_queue->state);
2770 2771
}

2772 2773 2774 2775 2776 2777 2778
/**
 *	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
2779 2780
static inline void netif_stop_queue(struct net_device *dev)
{
2781
	netif_tx_stop_queue(netdev_get_tx_queue(dev, 0));
L
Linus Torvalds 已提交
2782 2783
}

2784
void netif_tx_stop_all_queues(struct net_device *dev);
2785

2786
static inline bool netif_tx_queue_stopped(const struct netdev_queue *dev_queue)
2787
{
2788
	return test_bit(__QUEUE_STATE_DRV_XOFF, &dev_queue->state);
2789 2790
}

2791 2792 2793 2794 2795 2796
/**
 *	netif_queue_stopped - test if transmit queue is flowblocked
 *	@dev: network device
 *
 *	Test if transmit queue on device is currently unable to send.
 */
2797
static inline bool netif_queue_stopped(const struct net_device *dev)
L
Linus Torvalds 已提交
2798
{
2799
	return netif_tx_queue_stopped(netdev_get_tx_queue(dev, 0));
L
Linus Torvalds 已提交
2800 2801
}

2802
static inline bool netif_xmit_stopped(const struct netdev_queue *dev_queue)
2803
{
2804 2805 2806
	return dev_queue->state & QUEUE_STATE_ANY_XOFF;
}

2807 2808
static inline bool
netif_xmit_frozen_or_stopped(const struct netdev_queue *dev_queue)
2809 2810 2811 2812
{
	return dev_queue->state & QUEUE_STATE_ANY_XOFF_OR_FROZEN;
}

2813 2814 2815 2816 2817 2818
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;
}

2819 2820 2821 2822 2823 2824 2825 2826 2827 2828 2829 2830 2831 2832 2833 2834 2835 2836 2837 2838 2839 2840 2841 2842 2843 2844 2845 2846
/**
 *	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
}

2847 2848 2849
static inline void netdev_tx_sent_queue(struct netdev_queue *dev_queue,
					unsigned int bytes)
{
T
Tom Herbert 已提交
2850 2851
#ifdef CONFIG_BQL
	dql_queued(&dev_queue->dql, bytes);
2852 2853 2854 2855 2856 2857 2858 2859 2860 2861 2862 2863 2864 2865 2866 2867

	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
2868
#endif
2869 2870
}

2871 2872 2873 2874 2875 2876 2877 2878 2879
/**
 * 	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
 */
2880 2881 2882 2883 2884 2885
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,
2886
					     unsigned int pkts, unsigned int bytes)
2887
{
T
Tom Herbert 已提交
2888
#ifdef CONFIG_BQL
2889 2890 2891 2892 2893 2894 2895 2896 2897 2898 2899 2900 2901 2902 2903 2904 2905
	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
2906
#endif
2907 2908
}

2909 2910 2911 2912 2913 2914 2915 2916 2917 2918
/**
 * 	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()
 */
2919
static inline void netdev_completed_queue(struct net_device *dev,
2920
					  unsigned int pkts, unsigned int bytes)
2921 2922 2923 2924 2925 2926
{
	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
2927
#ifdef CONFIG_BQL
2928
	clear_bit(__QUEUE_STATE_STACK_XOFF, &q->state);
T
Tom Herbert 已提交
2929 2930
	dql_reset(&q->dql);
#endif
2931 2932
}

2933 2934 2935 2936 2937 2938 2939
/**
 * 	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
 */
2940 2941 2942
static inline void netdev_reset_queue(struct net_device *dev_queue)
{
	netdev_tx_reset_queue(netdev_get_tx_queue(dev_queue, 0));
2943 2944
}

2945 2946 2947 2948 2949 2950 2951 2952 2953 2954 2955 2956 2957 2958 2959 2960 2961 2962 2963 2964
/**
 * 	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;
}

2965 2966 2967 2968 2969 2970
/**
 *	netif_running - test if up
 *	@dev: network device
 *
 *	Test if the device has been brought up.
 */
2971
static inline bool netif_running(const struct net_device *dev)
L
Linus Torvalds 已提交
2972 2973 2974 2975
{
	return test_bit(__LINK_STATE_START, &dev->state);
}

2976 2977 2978 2979 2980 2981
/*
 * 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.
 */
2982 2983 2984 2985 2986 2987 2988 2989

/**
 *	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.
 */
2990 2991
static inline void netif_start_subqueue(struct net_device *dev, u16 queue_index)
{
2992
	struct netdev_queue *txq = netdev_get_tx_queue(dev, queue_index);
2993 2994

	netif_tx_start_queue(txq);
2995 2996
}

2997 2998 2999 3000 3001 3002 3003
/**
 *	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.
 */
3004 3005
static inline void netif_stop_subqueue(struct net_device *dev, u16 queue_index)
{
3006
	struct netdev_queue *txq = netdev_get_tx_queue(dev, queue_index);
3007
	netif_tx_stop_queue(txq);
3008 3009
}

3010 3011 3012 3013 3014 3015 3016
/**
 *	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.
 */
3017 3018
static inline bool __netif_subqueue_stopped(const struct net_device *dev,
					    u16 queue_index)
3019
{
3020
	struct netdev_queue *txq = netdev_get_tx_queue(dev, queue_index);
3021 3022

	return netif_tx_queue_stopped(txq);
3023 3024
}

3025 3026
static inline bool netif_subqueue_stopped(const struct net_device *dev,
					  struct sk_buff *skb)
3027 3028 3029
{
	return __netif_subqueue_stopped(dev, skb_get_queue_mapping(skb));
}
3030

3031
void netif_wake_subqueue(struct net_device *dev, u16 queue_index);
3032

3033
#ifdef CONFIG_XPS
3034
int netif_set_xps_queue(struct net_device *dev, const struct cpumask *mask,
3035
			u16 index);
3036 3037
#else
static inline int netif_set_xps_queue(struct net_device *dev,
3038
				      const struct cpumask *mask,
3039 3040 3041 3042 3043 3044
				      u16 index)
{
	return 0;
}
#endif

J
Jiri Pirko 已提交
3045 3046 3047
u16 __skb_tx_hash(const struct net_device *dev, struct sk_buff *skb,
		  unsigned int num_tx_queues);

3048 3049 3050 3051 3052
/*
 * 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,
3053
			      struct sk_buff *skb)
3054 3055 3056 3057
{
	return __skb_tx_hash(dev, skb, dev->real_num_tx_queues);
}

3058 3059 3060 3061 3062 3063
/**
 *	netif_is_multiqueue - test if device has multiple transmit queues
 *	@dev: network device
 *
 * Check if device has multiple transmit queues
 */
3064
static inline bool netif_is_multiqueue(const struct net_device *dev)
3065
{
E
Eric Dumazet 已提交
3066
	return dev->num_tx_queues > 1;
3067
}
L
Linus Torvalds 已提交
3068

3069
int netif_set_real_num_tx_queues(struct net_device *dev, unsigned int txq);
3070

3071
#ifdef CONFIG_SYSFS
3072
int netif_set_real_num_rx_queues(struct net_device *dev, unsigned int rxq);
3073 3074 3075 3076 3077 3078 3079 3080
#else
static inline int netif_set_real_num_rx_queues(struct net_device *dev,
						unsigned int rxq)
{
	return 0;
}
#endif

3081 3082 3083 3084 3085 3086 3087 3088 3089 3090 3091 3092
#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

3093
#define DEFAULT_MAX_NUM_RSS_QUEUES	(8)
3094
int netif_get_num_default_rss_queues(void);
3095

3096 3097 3098 3099 3100 3101 3102
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
3103

3104 3105 3106 3107 3108 3109 3110 3111 3112 3113 3114 3115 3116 3117 3118 3119 3120 3121
/*
 * 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
3122
 */
3123 3124 3125 3126 3127 3128 3129 3130 3131 3132 3133 3134 3135 3136 3137 3138 3139 3140 3141
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
3142

3143 3144
int netif_rx(struct sk_buff *skb);
int netif_rx_ni(struct sk_buff *skb);
3145
int netif_receive_skb(struct sk_buff *skb);
3146 3147 3148 3149
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);
3150 3151
struct packet_offload *gro_find_receive_by_type(__be16 type);
struct packet_offload *gro_find_complete_by_type(__be16 type);
3152 3153 3154 3155 3156 3157 3158

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

3159 3160 3161 3162 3163 3164 3165 3166 3167 3168 3169
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);
3170 3171
void __dev_notify_flags(struct net_device *, unsigned int old_flags,
			unsigned int gchanges);
3172 3173 3174 3175 3176 3177 3178 3179
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,
3180
			 struct netdev_phys_item_id *ppid);
3181 3182
int dev_get_phys_port_name(struct net_device *dev,
			   char *name, size_t len);
3183
int dev_change_proto_down(struct net_device *dev, bool proto_down);
3184
struct sk_buff *validate_xmit_skb_list(struct sk_buff *skb, struct net_device *dev);
3185 3186
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
3187
int __dev_forward_skb(struct net_device *dev, struct sk_buff *skb);
3188
int dev_forward_skb(struct net_device *dev, struct sk_buff *skb);
3189
bool is_skb_forwardable(struct net_device *dev, struct sk_buff *skb);
L
Linus Torvalds 已提交
3190

3191
extern int		netdev_budget;
L
Linus Torvalds 已提交
3192 3193

/* Called by rtnetlink.c:rtnl_unlock() */
3194
void netdev_run_todo(void);
L
Linus Torvalds 已提交
3195

3196 3197 3198 3199
/**
 *	dev_put - release reference to device
 *	@dev: network device
 *
3200
 * Release reference to device to allow it to be freed.
3201
 */
L
Linus Torvalds 已提交
3202 3203
static inline void dev_put(struct net_device *dev)
{
3204
	this_cpu_dec(*dev->pcpu_refcnt);
L
Linus Torvalds 已提交
3205 3206
}

3207 3208 3209 3210
/**
 *	dev_hold - get reference to device
 *	@dev: network device
 *
3211
 * Hold reference to device to keep it from being freed.
3212
 */
3213 3214
static inline void dev_hold(struct net_device *dev)
{
3215
	this_cpu_inc(*dev->pcpu_refcnt);
3216
}
L
Linus Torvalds 已提交
3217 3218 3219 3220

/* 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
3221 3222 3223 3224
 *
 * 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
3225 3226
 */

3227 3228 3229
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
3230

3231 3232 3233 3234 3235 3236
/**
 *	netif_carrier_ok - test if carrier present
 *	@dev: network device
 *
 * Check if carrier is present on device
 */
3237
static inline bool netif_carrier_ok(const struct net_device *dev)
L
Linus Torvalds 已提交
3238 3239 3240 3241
{
	return !test_bit(__LINK_STATE_NOCARRIER, &dev->state);
}

3242
unsigned long dev_trans_start(struct net_device *dev);
3243

3244
void __netdev_watchdog_up(struct net_device *dev);
L
Linus Torvalds 已提交
3245

3246
void netif_carrier_on(struct net_device *dev);
L
Linus Torvalds 已提交
3247

3248
void netif_carrier_off(struct net_device *dev);
L
Linus Torvalds 已提交
3249

3250 3251 3252 3253 3254 3255 3256 3257 3258 3259 3260 3261 3262
/**
 *	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
3263 3264 3265 3266 3267 3268
static inline void netif_dormant_on(struct net_device *dev)
{
	if (!test_and_set_bit(__LINK_STATE_DORMANT, &dev->state))
		linkwatch_fire_event(dev);
}

3269 3270 3271 3272 3273 3274
/**
 *	netif_dormant_off - set device as not dormant.
 *	@dev: network device
 *
 * Device is not in dormant state.
 */
S
Stefan Rompf 已提交
3275 3276 3277 3278 3279 3280
static inline void netif_dormant_off(struct net_device *dev)
{
	if (test_and_clear_bit(__LINK_STATE_DORMANT, &dev->state))
		linkwatch_fire_event(dev);
}

3281 3282 3283 3284 3285 3286
/**
 *	netif_dormant - test if carrier present
 *	@dev: network device
 *
 * Check if carrier is present on device
 */
3287
static inline bool netif_dormant(const struct net_device *dev)
S
Stefan Rompf 已提交
3288 3289 3290 3291 3292
{
	return test_bit(__LINK_STATE_DORMANT, &dev->state);
}


3293 3294 3295 3296 3297 3298
/**
 *	netif_oper_up - test if device is operational
 *	@dev: network device
 *
 * Check if carrier is operational
 */
3299
static inline bool netif_oper_up(const struct net_device *dev)
E
Eric Dumazet 已提交
3300
{
S
Stefan Rompf 已提交
3301 3302 3303 3304
	return (dev->operstate == IF_OPER_UP ||
		dev->operstate == IF_OPER_UNKNOWN /* backward compat */);
}

3305 3306 3307 3308 3309 3310
/**
 *	netif_device_present - is device available or removed
 *	@dev: network device
 *
 * Check if device has not been removed from system.
 */
3311
static inline bool netif_device_present(struct net_device *dev)
L
Linus Torvalds 已提交
3312 3313 3314 3315
{
	return test_bit(__LINK_STATE_PRESENT, &dev->state);
}

3316
void netif_device_detach(struct net_device *dev);
L
Linus Torvalds 已提交
3317

3318
void netif_device_attach(struct net_device *dev);
L
Linus Torvalds 已提交
3319 3320 3321 3322 3323 3324 3325 3326 3327 3328 3329 3330 3331 3332 3333 3334 3335 3336 3337 3338 3339 3340 3341 3342 3343 3344 3345 3346 3347 3348 3349 3350 3351 3352 3353 3354 3355 3356 3357 3358 3359 3360 3361 3362 3363 3364 3365 3366 3367 3368

/*
 * 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;
}

3369
static inline void __netif_tx_lock(struct netdev_queue *txq, int cpu)
H
Herbert Xu 已提交
3370
{
3371 3372
	spin_lock(&txq->_xmit_lock);
	txq->xmit_lock_owner = cpu;
3373 3374
}

3375 3376 3377 3378 3379 3380
static inline void __netif_tx_lock_bh(struct netdev_queue *txq)
{
	spin_lock_bh(&txq->_xmit_lock);
	txq->xmit_lock_owner = smp_processor_id();
}

3381
static inline bool __netif_tx_trylock(struct netdev_queue *txq)
3382
{
3383
	bool ok = spin_trylock(&txq->_xmit_lock);
3384 3385 3386 3387 3388 3389 3390 3391 3392 3393 3394 3395 3396 3397 3398 3399 3400
	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
3401 3402 3403 3404 3405 3406
static inline void txq_trans_update(struct netdev_queue *txq)
{
	if (txq->xmit_lock_owner != -1)
		txq->trans_start = jiffies;
}

3407 3408 3409 3410 3411 3412
/**
 *	netif_tx_lock - grab network device transmit lock
 *	@dev: network device
 *
 * Get network device transmit lock
 */
3413 3414
static inline void netif_tx_lock(struct net_device *dev)
{
3415
	unsigned int i;
3416
	int cpu;
3417

3418 3419
	spin_lock(&dev->tx_global_lock);
	cpu = smp_processor_id();
3420 3421
	for (i = 0; i < dev->num_tx_queues; i++) {
		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
3422 3423 3424 3425 3426 3427 3428

		/* 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.
		 */
3429
		__netif_tx_lock(txq, cpu);
3430 3431
		set_bit(__QUEUE_STATE_FROZEN, &txq->state);
		__netif_tx_unlock(txq);
3432
	}
H
Herbert Xu 已提交
3433 3434 3435 3436
}

static inline void netif_tx_lock_bh(struct net_device *dev)
{
3437 3438
	local_bh_disable();
	netif_tx_lock(dev);
H
Herbert Xu 已提交
3439 3440 3441 3442
}

static inline void netif_tx_unlock(struct net_device *dev)
{
3443 3444 3445 3446
	unsigned int i;

	for (i = 0; i < dev->num_tx_queues; i++) {
		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
3447

3448 3449 3450 3451 3452
		/* 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);
3453
		netif_schedule_queue(txq);
3454 3455
	}
	spin_unlock(&dev->tx_global_lock);
H
Herbert Xu 已提交
3456 3457 3458 3459
}

static inline void netif_tx_unlock_bh(struct net_device *dev)
{
3460 3461
	netif_tx_unlock(dev);
	local_bh_enable();
H
Herbert Xu 已提交
3462 3463
}

3464
#define HARD_TX_LOCK(dev, txq, cpu) {			\
3465
	if ((dev->features & NETIF_F_LLTX) == 0) {	\
3466
		__netif_tx_lock(txq, cpu);		\
3467 3468 3469
	}						\
}

3470 3471 3472 3473 3474
#define HARD_TX_TRYLOCK(dev, txq)			\
	(((dev->features & NETIF_F_LLTX) == 0) ?	\
		__netif_tx_trylock(txq) :		\
		true )

3475
#define HARD_TX_UNLOCK(dev, txq) {			\
3476
	if ((dev->features & NETIF_F_LLTX) == 0) {	\
3477
		__netif_tx_unlock(txq);			\
3478 3479 3480
	}						\
}

L
Linus Torvalds 已提交
3481 3482
static inline void netif_tx_disable(struct net_device *dev)
{
3483
	unsigned int i;
3484
	int cpu;
3485

3486 3487
	local_bh_disable();
	cpu = smp_processor_id();
3488 3489
	for (i = 0; i < dev->num_tx_queues; i++) {
		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
3490 3491

		__netif_tx_lock(txq, cpu);
3492
		netif_tx_stop_queue(txq);
3493
		__netif_tx_unlock(txq);
3494
	}
3495
	local_bh_enable();
L
Linus Torvalds 已提交
3496 3497
}

3498 3499 3500 3501 3502
static inline void netif_addr_lock(struct net_device *dev)
{
	spin_lock(&dev->addr_list_lock);
}

3503 3504
static inline void netif_addr_lock_nested(struct net_device *dev)
{
3505 3506 3507 3508 3509 3510
	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);
3511 3512
}

3513 3514 3515 3516 3517 3518 3519 3520 3521 3522 3523 3524 3525 3526 3527
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);
}

3528
/*
3529
 * dev_addrs walker. Should be used only for read access. Call with
3530 3531 3532
 * rcu_read_lock held.
 */
#define for_each_dev_addr(dev, ha) \
3533
		list_for_each_entry_rcu(ha, &dev->dev_addrs.list, list)
3534

L
Linus Torvalds 已提交
3535 3536
/* These functions live elsewhere (drivers/net/net_init.c, but related) */

3537
void ether_setup(struct net_device *dev);
L
Linus Torvalds 已提交
3538 3539

/* Support for loadable net-drivers */
3540
struct net_device *alloc_netdev_mqs(int sizeof_priv, const char *name,
3541
				    unsigned char name_assign_type,
3542 3543
				    void (*setup)(struct net_device *),
				    unsigned int txqs, unsigned int rxqs);
3544 3545
#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
3546

3547 3548 3549
#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
3550

3551 3552
int register_netdev(struct net_device *dev);
void unregister_netdev(struct net_device *dev);
3553

3554
/* General hardware address lists handling functions */
3555 3556 3557 3558
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);
3559 3560 3561 3562 3563 3564 3565 3566 3567
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 *));
3568
void __hw_addr_init(struct netdev_hw_addr_list *list);
3569

3570
/* Functions used for device addresses handling */
3571 3572 3573 3574 3575 3576
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);
3577 3578

/* Functions used for unicast addresses handling */
3579 3580 3581 3582 3583 3584 3585 3586
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);
3587

3588 3589 3590 3591 3592 3593 3594 3595 3596 3597 3598 3599 3600 3601 3602 3603 3604 3605 3606
/**
 *  __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);
}

/**
3607
 *  __dev_uc_unsync - Remove synchronized addresses from device
3608 3609 3610 3611 3612 3613 3614 3615 3616 3617 3618 3619
 *  @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);
}

3620
/* Functions used for multicast addresses handling */
3621 3622 3623 3624 3625 3626 3627 3628 3629 3630
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);
3631

3632 3633 3634 3635 3636 3637 3638 3639 3640 3641 3642 3643 3644 3645 3646 3647 3648 3649 3650
/**
 *  __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);
}

/**
3651
 *  __dev_mc_unsync - Remove synchronized addresses from device
3652 3653 3654 3655 3656 3657 3658 3659 3660 3661 3662 3663
 *  @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);
}

3664
/* Functions used for secondary unicast and multicast support */
3665 3666 3667 3668 3669 3670 3671
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
3672
/* Load a device via the kmod */
3673 3674 3675 3676 3677
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);
3678

L
Linus Torvalds 已提交
3679
extern int		netdev_max_backlog;
E
Eric Dumazet 已提交
3680
extern int		netdev_tstamp_prequeue;
L
Linus Torvalds 已提交
3681
extern int		weight_p;
3682
extern int		bpf_jit_enable;
J
Jiri Pirko 已提交
3683

3684
bool netdev_has_upper_dev(struct net_device *dev, struct net_device *upper_dev);
3685 3686
struct net_device *netdev_upper_get_next_dev_rcu(struct net_device *dev,
						     struct list_head **iter);
3687 3688
struct net_device *netdev_all_upper_get_next_dev_rcu(struct net_device *dev,
						     struct list_head **iter);
3689

3690 3691 3692 3693 3694 3695 3696
/* 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)))

3697
/* iterate through upper list, must be called under RCU read lock */
3698 3699 3700 3701 3702
#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)))
3703

3704 3705 3706 3707
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);
3708 3709 3710 3711 3712 3713 3714 3715 3716 3717 3718 3719 3720

#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)))

3721 3722 3723 3724 3725 3726 3727 3728
void *netdev_lower_get_next(struct net_device *dev,
				struct list_head **iter);
#define netdev_for_each_lower_dev(dev, ldev, iter) \
	for (iter = &(dev)->adj_list.lower, \
	     ldev = netdev_lower_get_next(dev, &(iter)); \
	     ldev; \
	     ldev = netdev_lower_get_next(dev, &(iter)))

3729
void *netdev_adjacent_get_private(struct list_head *adj_list);
3730
void *netdev_lower_get_first_private_rcu(struct net_device *dev);
3731 3732 3733 3734
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,
3735
				 struct net_device *upper_dev,
3736
				 void *upper_priv, void *upper_info);
3737 3738
void netdev_upper_dev_unlink(struct net_device *dev,
			     struct net_device *upper_dev);
3739
void netdev_adjacent_rename_links(struct net_device *dev, char *oldname);
3740 3741
void *netdev_lower_dev_get_private(struct net_device *dev,
				   struct net_device *lower_dev);
3742 3743
void netdev_lower_state_changed(struct net_device *lower_dev,
				void *lower_state_info);
3744 3745 3746 3747 3748 3749

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

3750
int dev_get_nest_level(struct net_device *dev,
3751
		       bool (*type_check)(const struct net_device *dev));
3752 3753 3754 3755 3756
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);
3757

3758 3759 3760 3761 3762 3763 3764 3765 3766 3767 3768 3769 3770
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);

3771 3772 3773 3774 3775
static inline
struct sk_buff *skb_gso_segment(struct sk_buff *skb, netdev_features_t features)
{
	return __skb_gso_segment(skb, features, true);
}
3776
__be16 skb_network_protocol(struct sk_buff *skb, int *depth);
3777 3778 3779 3780

static inline bool can_checksum_protocol(netdev_features_t features,
					 __be16 protocol)
{
3781 3782 3783 3784 3785 3786 3787 3788 3789 3790 3791 3792 3793 3794 3795 3796 3797 3798
	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;
	}
3799
}
3800

3801 3802 3803 3804 3805 3806 3807 3808 3809 3810 3811 3812 3813
/* 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;
	}
}

3814
#ifdef CONFIG_BUG
3815
void netdev_rx_csum_fault(struct net_device *dev);
3816 3817 3818 3819 3820
#else
static inline void netdev_rx_csum_fault(struct net_device *dev)
{
}
#endif
L
Linus Torvalds 已提交
3821
/* rx skb timestamps */
3822 3823
void net_enable_timestamp(void);
void net_disable_timestamp(void);
L
Linus Torvalds 已提交
3824

3825
#ifdef CONFIG_PROC_FS
3826
int __init dev_proc_init(void);
3827 3828
#else
#define dev_proc_init() 0
3829 3830
#endif

D
David S. Miller 已提交
3831
static inline netdev_tx_t __netdev_start_xmit(const struct net_device_ops *ops,
3832 3833
					      struct sk_buff *skb, struct net_device *dev,
					      bool more)
D
David S. Miller 已提交
3834
{
3835
	skb->xmit_more = more ? 1 : 0;
3836
	return ops->ndo_start_xmit(skb, dev);
D
David S. Miller 已提交
3837 3838
}

3839
static inline netdev_tx_t netdev_start_xmit(struct sk_buff *skb, struct net_device *dev,
3840
					    struct netdev_queue *txq, bool more)
D
David S. Miller 已提交
3841 3842
{
	const struct net_device_ops *ops = dev->netdev_ops;
3843
	int rc;
D
David S. Miller 已提交
3844

3845
	rc = __netdev_start_xmit(ops, skb, dev, more);
3846 3847 3848 3849
	if (rc == NETDEV_TX_OK)
		txq_trans_update(txq);

	return rc;
D
David S. Miller 已提交
3850 3851
}

3852 3853 3854 3855
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);
3856 3857 3858 3859 3860 3861 3862 3863 3864 3865

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);
}
3866

3867 3868
extern struct kobj_ns_type_operations net_ns_type_operations;

3869
const char *netdev_drivername(const struct net_device *dev);
3870

3871
void linkwatch_run_queue(void);
3872

3873 3874 3875
static inline netdev_features_t netdev_intersect_features(netdev_features_t f1,
							  netdev_features_t f2)
{
3876 3877
	if ((f1 ^ f2) & NETIF_F_HW_CSUM) {
		if (f1 & NETIF_F_HW_CSUM)
3878
			f1 |= (NETIF_F_IP_CSUM|NETIF_F_IPV6_CSUM);
3879
		else
3880
			f2 |= (NETIF_F_IP_CSUM|NETIF_F_IPV6_CSUM);
3881
	}
3882

3883
	return f1 & f2;
3884 3885
}

3886 3887
static inline netdev_features_t netdev_get_wanted_features(
	struct net_device *dev)
3888 3889 3890
{
	return (dev->features & ~dev->hw_features) | dev->wanted_features;
}
3891 3892
netdev_features_t netdev_increment_features(netdev_features_t all,
	netdev_features_t one, netdev_features_t mask);
3893 3894 3895 3896 3897 3898 3899 3900 3901 3902 3903

/* 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);
}

3904
int __netdev_update_features(struct net_device *dev);
3905
void netdev_update_features(struct net_device *dev);
3906
void netdev_change_features(struct net_device *dev);
3907

3908 3909 3910
void netif_stacked_transfer_operstate(const struct net_device *rootdev,
					struct net_device *dev);

3911 3912 3913
netdev_features_t passthru_features_check(struct sk_buff *skb,
					  struct net_device *dev,
					  netdev_features_t features);
3914
netdev_features_t netif_skb_features(struct sk_buff *skb);
3915

3916
static inline bool net_gso_ok(netdev_features_t features, int gso_type)
3917
{
3918
	netdev_features_t feature = gso_type << NETIF_F_GSO_SHIFT;
3919 3920 3921 3922 3923 3924 3925 3926

	/* 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));
3927 3928 3929 3930 3931 3932
	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));
3933
	BUILD_BUG_ON(SKB_GSO_TUNNEL_REMCSUM != (NETIF_F_GSO_TUNNEL_REMCSUM >> NETIF_F_GSO_SHIFT));
3934

3935
	return (features & feature) == feature;
3936 3937
}

3938
static inline bool skb_gso_ok(struct sk_buff *skb, netdev_features_t features)
3939
{
H
Herbert Xu 已提交
3940
	return net_gso_ok(features, skb_shinfo(skb)->gso_type) &&
3941
	       (!skb_has_frag_list(skb) || (features & NETIF_F_FRAGLIST));
3942 3943
}

3944
static inline bool netif_needs_gso(struct sk_buff *skb,
3945
				   netdev_features_t features)
3946
{
3947
	return skb_is_gso(skb) && (!skb_gso_ok(skb, features) ||
3948 3949
		unlikely((skb->ip_summed != CHECKSUM_PARTIAL) &&
			 (skb->ip_summed != CHECKSUM_UNNECESSARY)));
3950 3951
}

3952 3953 3954 3955 3956 3957
static inline void netif_set_gso_max_size(struct net_device *dev,
					  unsigned int size)
{
	dev->gso_max_size = size;
}

3958 3959 3960 3961 3962 3963 3964 3965 3966 3967 3968 3969 3970
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;
}

3971
static inline bool netif_is_macvlan(const struct net_device *dev)
3972 3973 3974 3975
{
	return dev->priv_flags & IFF_MACVLAN;
}

3976
static inline bool netif_is_macvlan_port(const struct net_device *dev)
3977 3978 3979 3980
{
	return dev->priv_flags & IFF_MACVLAN_PORT;
}

3981
static inline bool netif_is_ipvlan(const struct net_device *dev)
3982 3983 3984 3985
{
	return dev->priv_flags & IFF_IPVLAN_SLAVE;
}

3986
static inline bool netif_is_ipvlan_port(const struct net_device *dev)
3987 3988 3989 3990
{
	return dev->priv_flags & IFF_IPVLAN_MASTER;
}

3991
static inline bool netif_is_bond_master(const struct net_device *dev)
3992 3993 3994 3995
{
	return dev->flags & IFF_MASTER && dev->priv_flags & IFF_BONDING;
}

3996
static inline bool netif_is_bond_slave(const struct net_device *dev)
J
Jiri Pirko 已提交
3997 3998 3999 4000
{
	return dev->flags & IFF_SLAVE && dev->priv_flags & IFF_BONDING;
}

4001 4002 4003 4004 4005
static inline bool netif_supports_nofcs(struct net_device *dev)
{
	return dev->priv_flags & IFF_SUPP_NOFCS;
}

4006
static inline bool netif_is_l3_master(const struct net_device *dev)
4007
{
4008
	return dev->priv_flags & IFF_L3MDEV_MASTER;
4009 4010
}

D
David Ahern 已提交
4011 4012 4013 4014 4015
static inline bool netif_is_l3_slave(const struct net_device *dev)
{
	return dev->priv_flags & IFF_L3MDEV_SLAVE;
}

4016 4017 4018 4019 4020
static inline bool netif_is_bridge_master(const struct net_device *dev)
{
	return dev->priv_flags & IFF_EBRIDGE;
}

4021 4022 4023 4024 4025
static inline bool netif_is_bridge_port(const struct net_device *dev)
{
	return dev->priv_flags & IFF_BRIDGE_PORT;
}

4026 4027 4028 4029 4030
static inline bool netif_is_ovs_master(const struct net_device *dev)
{
	return dev->priv_flags & IFF_OPENVSWITCH;
}

4031
static inline bool netif_is_team_master(const struct net_device *dev)
J
Jiri Pirko 已提交
4032 4033 4034 4035
{
	return dev->priv_flags & IFF_TEAM;
}

4036
static inline bool netif_is_team_port(const struct net_device *dev)
J
Jiri Pirko 已提交
4037 4038 4039 4040
{
	return dev->priv_flags & IFF_TEAM_PORT;
}

4041
static inline bool netif_is_lag_master(const struct net_device *dev)
J
Jiri Pirko 已提交
4042 4043 4044 4045
{
	return netif_is_bond_master(dev) || netif_is_team_master(dev);
}

4046
static inline bool netif_is_lag_port(const struct net_device *dev)
J
Jiri Pirko 已提交
4047 4048 4049 4050
{
	return netif_is_bond_slave(dev) || netif_is_team_port(dev);
}

4051 4052 4053 4054 4055 4056
/* 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);
}

4057
extern struct pernet_operations __net_initdata loopback_net_ops;
4058

4059 4060 4061 4062 4063 4064
/* Logging, debugging and troubleshooting/diagnostic helpers. */

/* netdev_printk helpers, similar to dev_printk */

static inline const char *netdev_name(const struct net_device *dev)
{
4065 4066
	if (!dev->name[0] || strchr(dev->name, '%'))
		return "(unnamed net_device)";
4067 4068 4069
	return dev->name;
}

4070 4071 4072 4073 4074 4075 4076 4077 4078 4079 4080 4081 4082 4083 4084
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)";
}

4085
__printf(3, 4)
4086 4087
void netdev_printk(const char *level, const struct net_device *dev,
		   const char *format, ...);
4088
__printf(2, 3)
4089
void netdev_emerg(const struct net_device *dev, const char *format, ...);
4090
__printf(2, 3)
4091
void netdev_alert(const struct net_device *dev, const char *format, ...);
4092
__printf(2, 3)
4093
void netdev_crit(const struct net_device *dev, const char *format, ...);
4094
__printf(2, 3)
4095
void netdev_err(const struct net_device *dev, const char *format, ...);
4096
__printf(2, 3)
4097
void netdev_warn(const struct net_device *dev, const char *format, ...);
4098
__printf(2, 3)
4099
void netdev_notice(const struct net_device *dev, const char *format, ...);
4100
__printf(2, 3)
4101
void netdev_info(const struct net_device *dev, const char *format, ...);
4102

4103 4104 4105
#define MODULE_ALIAS_NETDEV(device) \
	MODULE_ALIAS("netdev-" device)

4106
#if defined(CONFIG_DYNAMIC_DEBUG)
4107 4108
#define netdev_dbg(__dev, format, args...)			\
do {								\
4109
	dynamic_netdev_dbg(__dev, format, ##args);		\
4110
} while (0)
4111 4112 4113
#elif defined(DEBUG)
#define netdev_dbg(__dev, format, args...)			\
	netdev_printk(KERN_DEBUG, __dev, format, ##args)
4114 4115 4116 4117 4118 4119 4120 4121 4122 4123 4124 4125 4126 4127 4128 4129 4130 4131 4132 4133 4134 4135 4136 4137 4138 4139
#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...)			\
4140 4141
	WARN(1, "netdevice: %s%s\n" format, netdev_name(dev),	\
	     netdev_reg_state(dev), ##args)
4142

4143 4144 4145 4146 4147 4148 4149 4150
/* 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)

4151 4152 4153 4154 4155 4156
#define netif_level(level, priv, type, dev, fmt, args...)	\
do {								\
	if (netif_msg_##type(priv))				\
		netdev_##level(dev, fmt, ##args);		\
} while (0)

4157
#define netif_emerg(priv, type, dev, fmt, args...)		\
4158
	netif_level(emerg, priv, type, dev, fmt, ##args)
4159
#define netif_alert(priv, type, dev, fmt, args...)		\
4160
	netif_level(alert, priv, type, dev, fmt, ##args)
4161
#define netif_crit(priv, type, dev, fmt, args...)		\
4162
	netif_level(crit, priv, type, dev, fmt, ##args)
4163
#define netif_err(priv, type, dev, fmt, args...)		\
4164
	netif_level(err, priv, type, dev, fmt, ##args)
4165
#define netif_warn(priv, type, dev, fmt, args...)		\
4166
	netif_level(warn, priv, type, dev, fmt, ##args)
4167
#define netif_notice(priv, type, dev, fmt, args...)		\
4168
	netif_level(notice, priv, type, dev, fmt, ##args)
4169
#define netif_info(priv, type, dev, fmt, args...)		\
4170
	netif_level(info, priv, type, dev, fmt, ##args)
4171

4172
#if defined(CONFIG_DYNAMIC_DEBUG)
4173 4174 4175
#define netif_dbg(priv, type, netdev, format, args...)		\
do {								\
	if (netif_msg_##type(priv))				\
4176
		dynamic_netdev_dbg(netdev, format, ##args);	\
4177
} while (0)
4178 4179 4180
#elif defined(DEBUG)
#define netif_dbg(priv, type, dev, format, args...)		\
	netif_printk(priv, type, KERN_DEBUG, dev, format, ##args)
4181 4182 4183 4184 4185 4186 4187 4188 4189 4190
#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)
4191
#define netif_vdbg	netif_dbg
4192 4193 4194 4195
#else
#define netif_vdbg(priv, type, dev, format, args...)		\
({								\
	if (0)							\
4196
		netif_printk(priv, type, KERN_DEBUG, dev, format, ##args); \
4197 4198 4199
	0;							\
})
#endif
4200

4201 4202 4203 4204 4205 4206 4207 4208 4209 4210 4211 4212 4213 4214 4215 4216 4217 4218 4219 4220 4221 4222 4223 4224 4225 4226 4227 4228 4229 4230
/*
 *	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)

4231
#endif	/* _LINUX_NETDEVICE_H */