netdevice.h 80.5 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5 6 7 8 9
/*
 * INET		An implementation of the TCP/IP protocol suite for the LINUX
 *		operating system.  INET is implemented using the  BSD Socket
 *		interface as the means of communication with the user level.
 *
 *		Definitions for the Interfaces handler.
 *
 * Version:	@(#)dev.h	1.0.10	08/12/93
 *
10
 * Authors:	Ross Biro
L
Linus Torvalds 已提交
11 12 13
 *		Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG>
 *		Corey Minyard <wf-rch!minyard@relay.EU.net>
 *		Donald J. Becker, <becker@cesdis.gsfc.nasa.gov>
14
 *		Alan Cox, <alan@lxorguk.ukuu.org.uk>
L
Linus Torvalds 已提交
15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
 *		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

#include <linux/if.h>
#include <linux/if_ether.h>
#include <linux/if_packet.h>
31
#include <linux/if_link.h>
L
Linus Torvalds 已提交
32 33

#ifdef __KERNEL__
34
#include <linux/pm_qos.h>
A
Al Viro 已提交
35
#include <linux/timer.h>
36
#include <linux/delay.h>
A
Arun Sharma 已提交
37
#include <linux/atomic.h>
L
Linus Torvalds 已提交
38 39 40 41 42
#include <asm/cache.h>
#include <asm/byteorder.h>

#include <linux/device.h>
#include <linux/percpu.h>
43
#include <linux/rculist.h>
44
#include <linux/dmaengine.h>
45
#include <linux/workqueue.h>
L
Linus Torvalds 已提交
46

47
#include <linux/ethtool.h>
48
#include <net/net_namespace.h>
49
#include <net/dsa.h>
J
Jeff Kirsher 已提交
50
#ifdef CONFIG_DCB
51 52
#include <net/dcbnl.h>
#endif
53

54 55
#include <linux/netdev_features.h>

L
Linus Torvalds 已提交
56
struct vlan_group;
57
struct netpoll_info;
58
struct phy_device;
59 60
/* 802.11 specific */
struct wireless_dev;
L
Linus Torvalds 已提交
61 62 63 64
					/* source back-compat hooks */
#define SET_ETHTOOL_OPS(netdev,ops) \
	( (netdev)->ethtool_ops = (ops) )

65 66 67 68 69
/* hardware address assignment types */
#define NET_ADDR_PERM		0	/* address is permanent (default) */
#define NET_ADDR_RANDOM		1	/* address is generated randomly */
#define NET_ADDR_STOLEN		2	/* address is stolen from other device */

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

74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92
/*
 * Transmit return codes: transmit return codes originate from three different
 * namespaces:
 *
 * - qdisc return codes
 * - driver transmit return codes
 * - errno values
 *
 * Drivers are allowed to return any one of those in their hard_start_xmit()
 * function. Real network devices commonly used with qdiscs should only return
 * the driver transmit return codes though - when qdiscs are used, the actual
 * transmission happens asynchronously, so the value is not propagated to
 * higher layers. Virtual network devices transmit synchronously, in this case
 * the driver transmit return codes are consumed by dev_queue_xmit(), all
 * others are propagated to higher layers.
 */

/* qdisc ->enqueue() return codes. */
#define NET_XMIT_SUCCESS	0x00
93 94 95 96
#define NET_XMIT_DROP		0x01	/* skb dropped			*/
#define NET_XMIT_CN		0x02	/* congestion notification	*/
#define NET_XMIT_POLICED	0x03	/* skb is shot by police	*/
#define NET_XMIT_MASK		0x0f	/* qdisc flags in net/sch_generic.h */
L
Linus Torvalds 已提交
97

98 99 100
/* NET_XMIT_CN is special. It does not guarantee that this packet is lost. It
 * indicates that the device will soon be dropping packets, or already drops
 * some packets of the same priority; prompting us to send less aggressively. */
101
#define net_xmit_eval(e)	((e) == NET_XMIT_CN ? 0 : (e))
L
Linus Torvalds 已提交
102 103
#define net_xmit_errno(e)	((e) != NET_XMIT_CN ? -ENOBUFS : 0)

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

107
enum netdev_tx {
108
	__NETDEV_TX_MIN	 = INT_MIN,	/* make sure enum is signed */
109 110 111
	NETDEV_TX_OK	 = 0x00,	/* driver took care of packet */
	NETDEV_TX_BUSY	 = 0x10,	/* driver tx path was busy*/
	NETDEV_TX_LOCKED = 0x20,	/* driver tx lock was already taken */
112 113 114
};
typedef enum netdev_tx netdev_tx_t;

115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132
/*
 * Current order: NETDEV_TX_MASK > NET_XMIT_MASK >= 0 is significant;
 * hard_start_xmit() return < NET_XMIT_MASK means skb was consumed.
 */
static inline bool dev_xmit_complete(int rc)
{
	/*
	 * Positive cases with an skb consumed by a driver:
	 * - successful transmission (rc == NETDEV_TX_OK)
	 * - error while transmitting (rc < 0)
	 * - error while queueing to a different device (rc & NET_XMIT_MASK)
	 */
	if (likely(rc < NET_XMIT_MASK))
		return true;

	return false;
}

L
Linus Torvalds 已提交
133 134 135 136
#endif

#define MAX_ADDR_LEN	32		/* Largest hardware address length */

137 138 139
/* Initial net device group. All devices belong to group 0 by default. */
#define INIT_NETDEV_GROUP	0

140
#ifdef  __KERNEL__
L
Linus Torvalds 已提交
141 142 143 144
/*
 *	Compute the worst case header length according to the protocols
 *	used.
 */
G
Graf Yang 已提交
145

146
#if defined(CONFIG_WLAN) || defined(CONFIG_AX25) || defined(CONFIG_AX25_MODULE)
147 148 149 150 151
# if defined(CONFIG_MAC80211_MESH)
#  define LL_MAX_HEADER 128
# else
#  define LL_MAX_HEADER 96
# endif
152
#elif defined(CONFIG_TR) || defined(CONFIG_TR_MODULE)
153
# define LL_MAX_HEADER 48
L
Linus Torvalds 已提交
154
#else
155
# define LL_MAX_HEADER 32
L
Linus Torvalds 已提交
156 157
#endif

D
David S. Miller 已提交
158 159 160 161
#if !defined(CONFIG_NET_IPIP) && !defined(CONFIG_NET_IPIP_MODULE) && \
    !defined(CONFIG_NET_IPGRE) &&  !defined(CONFIG_NET_IPGRE_MODULE) && \
    !defined(CONFIG_IPV6_SIT) && !defined(CONFIG_IPV6_SIT_MODULE) && \
    !defined(CONFIG_IPV6_TUNNEL) && !defined(CONFIG_IPV6_TUNNEL_MODULE)
L
Linus Torvalds 已提交
162 163 164 165 166 167
#define MAX_HEADER LL_MAX_HEADER
#else
#define MAX_HEADER (LL_MAX_HEADER + 48)
#endif

/*
168 169
 *	Old network device statistics. Fields are native words
 *	(unsigned long) so they can be read and written atomically.
L
Linus Torvalds 已提交
170
 */
G
Graf Yang 已提交
171

E
Eric Dumazet 已提交
172
struct net_device_stats {
173 174 175 176 177 178 179 180 181
	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
182 183
	unsigned long	collisions;
	unsigned long	rx_length_errors;
184 185 186 187 188
	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
189 190 191 192 193 194 195 196 197
	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;
};

198 199
#endif  /*  __KERNEL__  */

L
Linus Torvalds 已提交
200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220

/* Media selection options. */
enum {
        IF_PORT_UNKNOWN = 0,
        IF_PORT_10BASE2,
        IF_PORT_10BASET,
        IF_PORT_AUI,
        IF_PORT_100BASET,
        IF_PORT_100BASETX,
        IF_PORT_100BASEFX
};

#ifdef __KERNEL__

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

struct neighbour;
struct neigh_parms;
struct sk_buff;

221 222 223 224
struct netdev_hw_addr {
	struct list_head	list;
	unsigned char		addr[MAX_ADDR_LEN];
	unsigned char		type;
J
Jiri Pirko 已提交
225 226 227 228
#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
229
#define NETDEV_HW_ADDR_T_MULTICAST	5
J
Jiri Pirko 已提交
230
	bool			synced;
231
	bool			global_use;
232
	int			refcount;
233 234 235
	struct rcu_head		rcu_head;
};

236 237 238 239 240
struct netdev_hw_addr_list {
	struct list_head	list;
	int			count;
};

241 242 243 244
#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)
245

246 247 248 249
#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)
250

251 252
#define netdev_mc_count(dev) netdev_hw_addr_list_count(&(dev)->mc)
#define netdev_mc_empty(dev) netdev_hw_addr_list_empty(&(dev)->mc)
253
#define netdev_for_each_mc_addr(ha, dev) \
254
	netdev_hw_addr_list_for_each(ha, &(dev)->mc)
255

E
Eric Dumazet 已提交
256
struct hh_cache {
257
	u16		hh_len;
258
	u16		__pad;
259
	seqlock_t	hh_lock;
L
Linus Torvalds 已提交
260 261 262 263

	/* cached hardware header; allow for machine alignment needs.        */
#define HH_DATA_MOD	16
#define HH_DATA_OFF(__len) \
J
Jiri Benc 已提交
264
	(HH_DATA_MOD - (((__len - 1) & (HH_DATA_MOD - 1)) + 1))
L
Linus Torvalds 已提交
265 266 267 268 269 270 271 272 273 274 275 276
#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.
277 278 279
 *
 * LL_ALLOCATED_SPACE also takes into account the tailroom the device
 * may need.
L
Linus Torvalds 已提交
280 281
 */
#define LL_RESERVED_SPACE(dev) \
282
	((((dev)->hard_header_len+(dev)->needed_headroom)&~(HH_DATA_MOD - 1)) + HH_DATA_MOD)
L
Linus Torvalds 已提交
283
#define LL_RESERVED_SPACE_EXTRA(dev,extra) \
284 285 286
	((((dev)->hard_header_len+(dev)->needed_headroom+(extra))&~(HH_DATA_MOD - 1)) + HH_DATA_MOD)
#define LL_ALLOCATED_SPACE(dev) \
	((((dev)->hard_header_len+(dev)->needed_headroom+(dev)->needed_tailroom)&~(HH_DATA_MOD - 1)) + HH_DATA_MOD)
L
Linus Torvalds 已提交
287

288 289 290 291 292 293
struct header_ops {
	int	(*create) (struct sk_buff *skb, struct net_device *dev,
			   unsigned short type, const void *daddr,
			   const void *saddr, unsigned len);
	int	(*parse)(const struct sk_buff *skb, unsigned char *haddr);
	int	(*rebuild)(struct sk_buff *skb);
294
	int	(*cache)(const struct neighbour *neigh, struct hh_cache *hh, __be16 type);
295 296 297 298 299
	void	(*cache_update)(struct hh_cache *hh,
				const struct net_device *dev,
				const unsigned char *haddr);
};

L
Linus Torvalds 已提交
300 301 302 303 304
/* 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
305
enum netdev_state_t {
L
Linus Torvalds 已提交
306 307 308
	__LINK_STATE_START,
	__LINK_STATE_PRESENT,
	__LINK_STATE_NOCARRIER,
S
Stefan Rompf 已提交
309 310
	__LINK_STATE_LINKWATCH_PENDING,
	__LINK_STATE_DORMANT,
L
Linus Torvalds 已提交
311 312 313 314 315
};


/*
 * This structure holds at boot time configured netdevice settings. They
G
Graf Yang 已提交
316
 * are then used in the device probing.
L
Linus Torvalds 已提交
317 318 319 320 321 322 323
 */
struct netdev_boot_setup {
	char name[IFNAMSIZ];
	struct ifmap map;
};
#define NETDEV_BOOT_SETUP_MAX 8

324
extern int __init netdev_boot_setup(char *str);
L
Linus Torvalds 已提交
325

326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344
/*
 * 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;
	int			(*poll)(struct napi_struct *, int);
#ifdef CONFIG_NETPOLL
	spinlock_t		poll_lock;
	int			poll_owner;
#endif
345 346 347

	unsigned int		gro_count;

H
Herbert Xu 已提交
348
	struct net_device	*dev;
349 350
	struct list_head	dev_list;
	struct sk_buff		*gro_list;
H
Herbert Xu 已提交
351
	struct sk_buff		*skb;
352 353
};

E
Eric Dumazet 已提交
354
enum {
355
	NAPI_STATE_SCHED,	/* Poll is scheduled */
D
David S. Miller 已提交
356
	NAPI_STATE_DISABLE,	/* Disable pending */
357
	NAPI_STATE_NPSVC,	/* Netpoll - don't dequeue from poll_list */
358 359
};

360
enum gro_result {
361 362 363 364 365 366
	GRO_MERGED,
	GRO_MERGED_FREE,
	GRO_HELD,
	GRO_NORMAL,
	GRO_DROP,
};
367
typedef enum gro_result gro_result_t;
368

369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417
/*
 * 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
 * are registred on exact device (ptype->dev == skb->dev).
 *
 * 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);
418

419
extern void __napi_schedule(struct napi_struct *n);
420

D
David S. Miller 已提交
421 422 423 424 425
static inline int napi_disable_pending(struct napi_struct *n)
{
	return test_bit(NAPI_STATE_DISABLE, &n->state);
}

426 427 428 429 430 431
/**
 *	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
432 433
 * insure only one NAPI poll instance runs.  We also make
 * sure there is no pending NAPI disable.
434 435 436
 */
static inline int napi_schedule_prep(struct napi_struct *n)
{
D
David S. Miller 已提交
437 438
	return !napi_disable_pending(n) &&
		!test_and_set_bit(NAPI_STATE_SCHED, &n->state);
439 440 441 442 443 444 445 446 447 448 449 450 451 452 453
}

/**
 *	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);
}

454 455 456 457 458 459 460 461 462 463
/* Try to reschedule poll. Called by dev->poll() after napi_complete().  */
static inline int napi_reschedule(struct napi_struct *napi)
{
	if (napi_schedule_prep(napi)) {
		__napi_schedule(napi);
		return 1;
	}
	return 0;
}

464 465 466 467 468 469
/**
 *	napi_complete - NAPI processing complete
 *	@n: napi context
 *
 * Mark NAPI processing as complete.
 */
470 471
extern void __napi_complete(struct napi_struct *n);
extern void napi_complete(struct napi_struct *n);
472 473 474 475 476 477 478 479 480 481

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

/**
 *	napi_enable - enable NAPI scheduling
 *	@n: napi context
 *
 * Resume NAPI from being scheduled on this context.
 * Must be paired with napi_disable.
 */
static inline void napi_enable(struct napi_struct *n)
{
	BUG_ON(!test_bit(NAPI_STATE_SCHED, &n->state));
	smp_mb__before_clear_bit();
	clear_bit(NAPI_STATE_SCHED, &n->state);
}

502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519
#ifdef CONFIG_SMP
/**
 *	napi_synchronize - wait until NAPI is not running
 *	@n: napi context
 *
 * Wait until NAPI is done being scheduled on this context.
 * Waits till any outstanding processing completes but
 * does not disable future activations.
 */
static inline void napi_synchronize(const struct napi_struct *n)
{
	while (test_bit(NAPI_STATE_SCHED, &n->state))
		msleep(1);
}
#else
# define napi_synchronize(n)	barrier()
#endif

E
Eric Dumazet 已提交
520
enum netdev_queue_state_t {
521
	__QUEUE_STATE_XOFF,
522
	__QUEUE_STATE_FROZEN,
523 524
#define QUEUE_STATE_XOFF_OR_FROZEN ((1 << __QUEUE_STATE_XOFF)		| \
				    (1 << __QUEUE_STATE_FROZEN))
525 526
};

527
struct netdev_queue {
528 529 530
/*
 * read mostly part
 */
531
	struct net_device	*dev;
532
	struct Qdisc		*qdisc;
533
	unsigned long		state;
534
	struct Qdisc		*qdisc_sleeping;
535
#if defined(CONFIG_RPS) || defined(CONFIG_XPS)
T
Tom Herbert 已提交
536 537
	struct kobject		kobj;
#endif
538 539 540
#if defined(CONFIG_XPS) && defined(CONFIG_NUMA)
	int			numa_node;
#endif
541 542 543 544 545
/*
 * write mostly part
 */
	spinlock_t		_xmit_lock ____cacheline_aligned_in_smp;
	int			xmit_lock_owner;
546 547 548 549
	/*
	 * please use this field instead of dev->trans_start
	 */
	unsigned long		trans_start;
550
} ____cacheline_aligned_in_smp;
551

552 553 554 555 556
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
557
	return NUMA_NO_NODE;
558 559 560 561 562 563 564 565 566 567
#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
568
#ifdef CONFIG_RPS
T
Tom Herbert 已提交
569 570 571 572 573 574 575 576 577 578 579
/*
 * 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];
};
#define RPS_MAP_SIZE(_num) (sizeof(struct rps_map) + (_num * sizeof(u16)))

T
Tom Herbert 已提交
580
/*
581 582 583
 * 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
584 585 586
 */
struct rps_dev_flow {
	u16 cpu;
587
	u16 filter;
T
Tom Herbert 已提交
588 589
	unsigned int last_qtail;
};
590
#define RPS_NO_FILTER 0xffff
T
Tom Herbert 已提交
591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637

/*
 * The rps_dev_flow_table structure contains a table of flow mappings.
 */
struct rps_dev_flow_table {
	unsigned int mask;
	struct rcu_head rcu;
	struct work_struct free_work;
	struct rps_dev_flow flows[0];
};
#define RPS_DEV_FLOW_TABLE_SIZE(_num) (sizeof(struct rps_dev_flow_table) + \
    (_num * sizeof(struct rps_dev_flow)))

/*
 * The rps_sock_flow_table contains mappings of flows to the last CPU
 * on which they were processed by the application (set in recvmsg).
 */
struct rps_sock_flow_table {
	unsigned int mask;
	u16 ents[0];
};
#define	RPS_SOCK_FLOW_TABLE_SIZE(_num) (sizeof(struct rps_sock_flow_table) + \
    (_num * sizeof(u16)))

#define RPS_NO_CPU 0xffff

static inline void rps_record_sock_flow(struct rps_sock_flow_table *table,
					u32 hash)
{
	if (table && hash) {
		unsigned int cpu, index = hash & table->mask;

		/* We only give a hint, preemption can change cpu under us */
		cpu = raw_smp_processor_id();

		if (table->ents[index] != cpu)
			table->ents[index] = cpu;
	}
}

static inline void rps_reset_sock_flow(struct rps_sock_flow_table *table,
				       u32 hash)
{
	if (table && hash)
		table->ents[hash & table->mask] = RPS_NO_CPU;
}

E
Eric Dumazet 已提交
638
extern struct rps_sock_flow_table __rcu *rps_sock_flow_table;
T
Tom Herbert 已提交
639

640 641 642 643 644
#ifdef CONFIG_RFS_ACCEL
extern bool rps_may_expire_flow(struct net_device *dev, u16 rxq_index,
				u32 flow_id, u16 filter_id);
#endif

T
Tom Herbert 已提交
645 646
/* This structure contains an instance of an RX queue. */
struct netdev_rx_queue {
E
Eric Dumazet 已提交
647 648 649
	struct rps_map __rcu		*rps_map;
	struct rps_dev_flow_table __rcu	*rps_flow_table;
	struct kobject			kobj;
T
Tom Herbert 已提交
650
	struct net_device		*dev;
T
Tom Herbert 已提交
651
} ____cacheline_aligned_in_smp;
T
Tom Herbert 已提交
652
#endif /* CONFIG_RPS */
653

T
Tom Herbert 已提交
654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673
#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];
};
#define XPS_MAP_SIZE(_num) (sizeof(struct xps_map) + (_num * sizeof(u16)))
#define XPS_MIN_MAP_ALLOC ((L1_CACHE_BYTES - sizeof(struct xps_map))	\
    / sizeof(u16))

/*
 * This structure holds all XPS maps for device.  Maps are indexed by CPU.
 */
struct xps_dev_maps {
	struct rcu_head rcu;
E
Eric Dumazet 已提交
674
	struct xps_map __rcu *cpu_map[0];
T
Tom Herbert 已提交
675 676 677 678 679
};
#define XPS_DEV_MAPS_SIZE (sizeof(struct xps_dev_maps) +		\
    (nr_cpu_ids * sizeof(struct xps_map *)))
#endif /* CONFIG_XPS */

680 681 682 683 684 685 686 687
#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;
};

688 689
/*
 * This structure defines the management hooks for network devices.
690 691
 * The following hooks can be defined; unless noted otherwise, they are
 * optional and can be filled with a null pointer.
692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710
 *
 * 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.
 *
711 712
 * netdev_tx_t (*ndo_start_xmit)(struct sk_buff *skb,
 *                               struct net_device *dev);
713
 *	Called when a packet needs to be transmitted.
714 715
 *	Must return NETDEV_TX_OK , NETDEV_TX_BUSY.
 *        (can also return NETDEV_TX_LOCKED iff NETIF_F_LLTX)
716 717 718 719 720 721
 *	Required can not be NULL.
 *
 * u16 (*ndo_select_queue)(struct net_device *dev, struct sk_buff *skb);
 *	Called to decide which queue to when device supports multiple
 *	transmit queues.
 *
722 723 724 725 726 727
 * 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.
728 729
 *	If driver handles unicast address filtering, it should set
 *	IFF_UNICAST_FLT to its priv_flags.
730 731 732
 *
 * int (*ndo_set_mac_address)(struct net_device *dev, void *addr);
 *	This function  is called when the Media Access Control address
733
 *	needs to be changed. If this interface is not defined, the
734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753
 *	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.
 *
754
 * void (*ndo_tx_timeout)(struct net_device *dev);
755 756 757
 *	Callback uses when the transmitter has not made any progress
 *	for dev->watchdog ticks.
 *
758
 * struct rtnl_link_stats64* (*ndo_get_stats64)(struct net_device *dev,
759
 *                      struct rtnl_link_stats64 *storage);
760
 * struct net_device_stats* (*ndo_get_stats)(struct net_device *dev);
761
 *	Called when a user wants to get the network device usage
762
 *	statistics. Drivers must do one of the following:
763 764
 *	1. Define @ndo_get_stats64 to fill in a zero-initialised
 *	   rtnl_link_stats64 structure passed by the caller.
765
 *	2. Define @ndo_get_stats to update a net_device_stats structure
766 767 768 769 770
 *	   (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.
771 772 773 774 775 776 777 778 779 780
 *
 * void (*ndo_vlan_rx_add_vid)(struct net_device *dev, unsigned short vid);
 *	If device support VLAN filtering (dev->features & NETIF_F_HW_VLAN_FILTER)
 *	this function is called when a VLAN id is registered.
 *
 * void (*ndo_vlan_rx_kill_vid)(struct net_device *dev, unsigned short vid);
 *	If device support VLAN filtering (dev->features & NETIF_F_HW_VLAN_FILTER)
 *	this function is called when a VLAN id is unregistered.
 *
 * void (*ndo_poll_controller)(struct net_device *dev);
781 782 783 784 785
 *
 *	SR-IOV management functions.
 * int (*ndo_set_vf_mac)(struct net_device *dev, int vf, u8* mac);
 * int (*ndo_set_vf_vlan)(struct net_device *dev, int vf, u16 vlan, u8 qos);
 * int (*ndo_set_vf_tx_rate)(struct net_device *dev, int vf, int rate);
786
 * int (*ndo_set_vf_spoofchk)(struct net_device *dev, int vf, bool setting);
787 788
 * int (*ndo_get_vf_config)(struct net_device *dev,
 *			    int vf, struct ifla_vf_info *ivf);
789 790 791
 * 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);
792 793 794 795 796
 * 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.
797
 *
798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833
 *	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.
 *
 * 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.
 *
834 835 836 837 838 839
 *	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.
840 841 842 843 844 845 846 847
 *
 *	Slave management functions (for bridge, bonding, etc). User should
 *	call netdev_set_master() to set dev->master properly.
 * int (*ndo_add_slave)(struct net_device *dev, struct net_device *slave_dev);
 *	Called to make another netdev an underling.
 *
 * int (*ndo_del_slave)(struct net_device *dev, struct net_device *slave_dev);
 *	Called to release previously enslaved netdev.
848 849
 *
 *      Feature/offload setting functions.
850 851
 * netdev_features_t (*ndo_fix_features)(struct net_device *dev,
 *		netdev_features_t features);
852 853 854 855
 *	Adjusts the requested feature flags according to device-specific
 *	constraints, and returns the resulting flags. Must not modify
 *	the device state.
 *
856
 * int (*ndo_set_features)(struct net_device *dev, netdev_features_t features);
857 858 859 860
 *	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.
 *
861 862 863 864 865 866
 */
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);
867
	netdev_tx_t		(*ndo_start_xmit) (struct sk_buff *skb,
868 869 870
						   struct net_device *dev);
	u16			(*ndo_select_queue)(struct net_device *dev,
						    struct sk_buff *skb);
871 872 873 874 875 876 877 878 879 880
	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);
881 882 883 884
	int			(*ndo_change_mtu)(struct net_device *dev,
						  int new_mtu);
	int			(*ndo_neigh_setup)(struct net_device *dev,
						   struct neigh_parms *);
885 886
	void			(*ndo_tx_timeout) (struct net_device *dev);

887 888
	struct rtnl_link_stats64* (*ndo_get_stats64)(struct net_device *dev,
						     struct rtnl_link_stats64 *storage);
889 890 891 892 893 894 895 896
	struct net_device_stats* (*ndo_get_stats)(struct net_device *dev);

	void			(*ndo_vlan_rx_add_vid)(struct net_device *dev,
						       unsigned short vid);
	void			(*ndo_vlan_rx_kill_vid)(struct net_device *dev,
						        unsigned short vid);
#ifdef CONFIG_NET_POLL_CONTROLLER
	void                    (*ndo_poll_controller)(struct net_device *dev);
H
Herbert Xu 已提交
897 898
	int			(*ndo_netpoll_setup)(struct net_device *dev,
						     struct netpoll_info *info);
899
	void			(*ndo_netpoll_cleanup)(struct net_device *dev);
900
#endif
901 902 903 904 905 906
	int			(*ndo_set_vf_mac)(struct net_device *dev,
						  int queue, u8 *mac);
	int			(*ndo_set_vf_vlan)(struct net_device *dev,
						   int queue, u16 vlan, u8 qos);
	int			(*ndo_set_vf_tx_rate)(struct net_device *dev,
						      int vf, int rate);
907 908
	int			(*ndo_set_vf_spoofchk)(struct net_device *dev,
						       int vf, bool setting);
909 910 911
	int			(*ndo_get_vf_config)(struct net_device *dev,
						     int vf,
						     struct ifla_vf_info *ivf);
912 913 914 915 916
	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);
917
	int			(*ndo_setup_tc)(struct net_device *dev, u8 tc);
918
#if defined(CONFIG_FCOE) || defined(CONFIG_FCOE_MODULE)
919 920
	int			(*ndo_fcoe_enable)(struct net_device *dev);
	int			(*ndo_fcoe_disable)(struct net_device *dev);
921 922 923 924 925 926
	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);
927 928 929 930
	int			(*ndo_fcoe_ddp_target)(struct net_device *dev,
						       u16 xid,
						       struct scatterlist *sgl,
						       unsigned int sgc);
931 932 933
#endif

#if defined(CONFIG_LIBFCOE) || defined(CONFIG_LIBFCOE_MODULE)
934 935 936 937
#define NETDEV_FCOE_WWNN 0
#define NETDEV_FCOE_WWPN 1
	int			(*ndo_fcoe_get_wwn)(struct net_device *dev,
						    u64 *wwn, int type);
938
#endif
939

940 941 942 943 944 945
#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
946 947 948 949
	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);
950 951
	netdev_features_t	(*ndo_fix_features)(struct net_device *dev,
						    netdev_features_t features);
952
	int			(*ndo_set_features)(struct net_device *dev,
953
						    netdev_features_t features);
954 955
};

L
Linus Torvalds 已提交
956 957 958 959 960 961 962 963 964 965
/*
 *	The DEVICE structure.
 *	Actually, this whole structure is a big mistake.  It mixes I/O
 *	data with strictly "high-level" data, and it has to know about
 *	almost every data structure used in the INET module.
 *
 *	FIXME: cleanup struct net_device such that network protocol info
 *	moves out.
 */

E
Eric Dumazet 已提交
966
struct net_device {
L
Linus Torvalds 已提交
967 968 969 970

	/*
	 * 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
971
	 * of the interface.
L
Linus Torvalds 已提交
972 973
	 */
	char			name[IFNAMSIZ];
M
Mark Gross 已提交
974

J
Jean Pihet 已提交
975
	struct pm_qos_request	pm_qos_req;
M
Mark Gross 已提交
976

977 978
	/* device name hash chain */
	struct hlist_node	name_hlist;
979 980
	/* snmp alias */
	char 			*ifalias;
L
Linus Torvalds 已提交
981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997

	/*
	 *	I/O specific fields
	 *	FIXME: Merge these and struct ifmap into one
	 */
	unsigned long		mem_end;	/* shared mem end	*/
	unsigned long		mem_start;	/* shared mem start	*/
	unsigned long		base_addr;	/* device I/O address	*/
	unsigned int		irq;		/* device IRQ number	*/

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

	unsigned long		state;

998
	struct list_head	dev_list;
999
	struct list_head	napi_list;
1000
	struct list_head	unreg_list;
L
Linus Torvalds 已提交
1001

1002
	/* currently active device features */
1003
	netdev_features_t	features;
1004
	/* user-changeable features */
1005
	netdev_features_t	hw_features;
1006
	/* user-requested features */
1007
	netdev_features_t	wanted_features;
1008
	/* mask of features inheritable by VLAN devices */
1009
	netdev_features_t	vlan_features;
1010

L
Linus Torvalds 已提交
1011 1012 1013 1014
	/* Interface index. Unique device identifier	*/
	int			ifindex;
	int			iflink;

R
Rusty Russell 已提交
1015
	struct net_device_stats	stats;
1016 1017 1018
	atomic_long_t		rx_dropped; /* dropped packets by core network
					     * Do not use this in drivers.
					     */
L
Linus Torvalds 已提交
1019

1020
#ifdef CONFIG_WIRELESS_EXT
L
Linus Torvalds 已提交
1021 1022 1023 1024 1025
	/* List of functions to handle Wireless Extensions (instead of ioctl).
	 * See <net/iw_handler.h> for details. Jean II */
	const struct iw_handler_def *	wireless_handlers;
	/* Instance data managed by the core of Wireless Extensions. */
	struct iw_public_data *	wireless_data;
1026
#endif
1027 1028
	/* Management operations */
	const struct net_device_ops *netdev_ops;
1029
	const struct ethtool_ops *ethtool_ops;
L
Linus Torvalds 已提交
1030

1031 1032 1033
	/* Hardware header description */
	const struct header_ops *header_ops;

S
Stefan Rompf 已提交
1034
	unsigned int		flags;	/* interface flags (a la BSD)	*/
1035
	unsigned int		priv_flags; /* Like 'flags' but invisible to userspace. */
L
Linus Torvalds 已提交
1036 1037 1038
	unsigned short		gflags;
	unsigned short		padded;	/* How much padding added by alloc_netdev() */

S
Stefan Rompf 已提交
1039 1040 1041
	unsigned char		operstate; /* RFC2863 operstate */
	unsigned char		link_mode; /* mapping policy to operstate */

1042 1043 1044
	unsigned char		if_port;	/* Selectable AUI, TP,..*/
	unsigned char		dma;		/* DMA channel		*/

1045
	unsigned int		mtu;	/* interface MTU value		*/
L
Linus Torvalds 已提交
1046 1047 1048
	unsigned short		type;	/* interface hardware type	*/
	unsigned short		hard_header_len;	/* hardware hdr length	*/

1049 1050 1051 1052 1053 1054 1055
	/* extra head- and tailroom the hardware may need, but not in all cases
	 * can this be guaranteed, especially tailroom. Some cases also use
	 * LL_MAX_HEADER instead to allocate the skb.
	 */
	unsigned short		needed_headroom;
	unsigned short		needed_tailroom;

L
Linus Torvalds 已提交
1056
	/* Interface address info. */
1057
	unsigned char		perm_addr[MAX_ADDR_LEN]; /* permanent hw address */
1058
	unsigned char		addr_assign_type; /* hw address assignment type */
L
Linus Torvalds 已提交
1059 1060 1061
	unsigned char		addr_len;	/* hardware address length	*/
	unsigned short          dev_id;		/* for shared network cards */

J
Jiri Pirko 已提交
1062
	spinlock_t		addr_list_lock;
1063 1064
	struct netdev_hw_addr_list	uc;	/* Unicast mac addresses */
	struct netdev_hw_addr_list	mc;	/* Multicast mac addresses */
1065
	bool			uc_promisc;
1066 1067
	unsigned int		promiscuity;
	unsigned int		allmulti;
L
Linus Torvalds 已提交
1068 1069 1070


	/* Protocol specific pointers */
1071 1072

#if defined(CONFIG_VLAN_8021Q) || defined(CONFIG_VLAN_8021Q_MODULE)
E
Eric Dumazet 已提交
1073
	struct vlan_group __rcu	*vlgrp;		/* VLAN group */
1074
#endif
1075 1076 1077
#ifdef CONFIG_NET_DSA
	void			*dsa_ptr;	/* dsa specific data */
#endif
L
Linus Torvalds 已提交
1078
	void 			*atalk_ptr;	/* AppleTalk link 	*/
E
Eric Dumazet 已提交
1079
	struct in_device __rcu	*ip_ptr;	/* IPv4 specific data	*/
1080
	struct dn_dev __rcu     *dn_ptr;        /* DECnet specific data */
E
Eric Dumazet 已提交
1081
	struct inet6_dev __rcu	*ip6_ptr;       /* IPv6 specific data */
L
Linus Torvalds 已提交
1082 1083
	void			*ec_ptr;	/* Econet specific data	*/
	void			*ax25_ptr;	/* AX.25 specific data */
1084 1085
	struct wireless_dev	*ieee80211_ptr;	/* IEEE 802.11 specific data,
						   assign before registering */
L
Linus Torvalds 已提交
1086

1087
/*
E
Eric Dumazet 已提交
1088
 * Cache lines mostly used on receive path (including eth_type_trans())
1089
 */
1090 1091 1092 1093 1094 1095 1096 1097
	unsigned long		last_rx;	/* Time of last Rx
						 * This should not be set in
						 * drivers, unless really needed,
						 * because network stack (bonding)
						 * use it if/when necessary, to
						 * avoid dirtying this cache line.
						 */

E
Eric Dumazet 已提交
1098 1099 1100 1101
	struct net_device	*master; /* Pointer to master device of a group,
					  * which this device is member of.
					  */

1102
	/* Interface address info used in eth_type_trans() */
1103 1104 1105 1106
	unsigned char		*dev_addr;	/* hw address, (before bcast
						   because most packets are
						   unicast) */

1107 1108
	struct netdev_hw_addr_list	dev_addrs; /* list of device
						      hw addresses */
1109 1110

	unsigned char		broadcast[MAX_ADDR_LEN];	/* hw bcast add	*/
L
Linus Torvalds 已提交
1111

1112
#if defined(CONFIG_RPS) || defined(CONFIG_XPS)
T
Tom Herbert 已提交
1113 1114 1115 1116
	struct kset		*queues_kset;

	struct netdev_rx_queue	*_rx;

1117
	/* Number of RX queues allocated at register_netdev() time */
T
Tom Herbert 已提交
1118
	unsigned int		num_rx_queues;
1119 1120 1121

	/* Number of RX queues currently active in device */
	unsigned int		real_num_rx_queues;
1122 1123 1124 1125 1126 1127 1128

#ifdef CONFIG_RFS_ACCEL
	/* CPU reverse-mapping for RX completion interrupts, indexed
	 * by RX queue number.  Assigned by driver.  This must only be
	 * set if the ndo_rx_flow_steer operation is defined. */
	struct cpu_rmap		*rx_cpu_rmap;
#endif
E
Eric Dumazet 已提交
1129
#endif
T
Tom Herbert 已提交
1130

1131 1132
	rx_handler_func_t __rcu	*rx_handler;
	void __rcu		*rx_handler_data;
1133

1134
	struct netdev_queue __rcu *ingress_queue;
E
Eric Dumazet 已提交
1135 1136 1137 1138

/*
 * Cache lines mostly used on transmit path
 */
1139
	struct netdev_queue	*_tx ____cacheline_aligned_in_smp;
1140 1141

	/* Number of TX queues allocated at alloc_netdev_mq() time  */
1142
	unsigned int		num_tx_queues;
1143 1144 1145 1146

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

1147 1148 1149
	/* root qdisc from userspace point of view */
	struct Qdisc		*qdisc;

L
Linus Torvalds 已提交
1150
	unsigned long		tx_queue_len;	/* Max frames per queue allowed */
1151
	spinlock_t		tx_global_lock;
E
Eric Dumazet 已提交
1152

T
Tom Herbert 已提交
1153
#ifdef CONFIG_XPS
E
Eric Dumazet 已提交
1154
	struct xps_dev_maps __rcu *xps_maps;
T
Tom Herbert 已提交
1155
#endif
T
Tom Herbert 已提交
1156

1157
	/* These may be needed for future network-power-down code. */
1158 1159 1160 1161 1162

	/*
	 * trans_start here is expensive for high speed devices on SMP,
	 * please use netdev_queue->trans_start instead.
	 */
1163 1164 1165 1166 1167
	unsigned long		trans_start;	/* Time (in jiffies) of last Tx	*/

	int			watchdog_timeo; /* used by dev_watchdog() */
	struct timer_list	watchdog_timer;

L
Linus Torvalds 已提交
1168
	/* Number of references to this device */
E
Eric Dumazet 已提交
1169
	int __percpu		*pcpu_refcnt;
1170

L
Linus Torvalds 已提交
1171 1172 1173 1174 1175
	/* delayed register/unregister */
	struct list_head	todo_list;
	/* device index hash chain */
	struct hlist_node	index_hlist;

1176
	struct list_head	link_watch_list;
1177

L
Linus Torvalds 已提交
1178 1179
	/* register/unregister state machine */
	enum { NETREG_UNINITIALIZED=0,
1180
	       NETREG_REGISTERED,	/* completed register_netdevice */
L
Linus Torvalds 已提交
1181 1182 1183
	       NETREG_UNREGISTERING,	/* called unregister_netdevice */
	       NETREG_UNREGISTERED,	/* completed unregister todo */
	       NETREG_RELEASED,		/* called free_netdev */
1184
	       NETREG_DUMMY,		/* dummy device for NAPI poll */
1185 1186 1187
	} reg_state:8;

	bool dismantle; /* device is going do be freed */
1188 1189 1190 1191 1192

	enum {
		RTNL_LINK_INITIALIZED,
		RTNL_LINK_INITIALIZING,
	} rtnl_link_state:16;
L
Linus Torvalds 已提交
1193

1194 1195
	/* Called from unregister, can be used to call free_netdev */
	void (*destructor)(struct net_device *dev);
L
Linus Torvalds 已提交
1196 1197

#ifdef CONFIG_NETPOLL
1198
	struct netpoll_info	*npinfo;
L
Linus Torvalds 已提交
1199
#endif
1200

1201
#ifdef CONFIG_NET_NS
1202 1203
	/* Network namespace this network device is inside */
	struct net		*nd_net;
1204
#endif
1205

D
David S. Miller 已提交
1206
	/* mid-layer private */
E
Eric Dumazet 已提交
1207 1208 1209
	union {
		void				*ml_priv;
		struct pcpu_lstats __percpu	*lstats; /* loopback stats */
1210
		struct pcpu_tstats __percpu	*tstats; /* tunnel stats */
1211
		struct pcpu_dstats __percpu	*dstats; /* dummy stats */
E
Eric Dumazet 已提交
1212
	};
1213
	/* GARP */
E
Eric Dumazet 已提交
1214
	struct garp_port __rcu	*garp_port;
L
Linus Torvalds 已提交
1215 1216

	/* class/net/name entry */
1217
	struct device		dev;
1218 1219
	/* space for optional device, statistics, and wireless sysfs groups */
	const struct attribute_group *sysfs_groups[4];
P
Patrick McHardy 已提交
1220 1221 1222

	/* rtnetlink link ops */
	const struct rtnl_link_ops *rtnl_link_ops;
1223

1224 1225 1226
	/* for setting kernel sock attribute on TCP connection setup */
#define GSO_MAX_SIZE		65536
	unsigned int		gso_max_size;
1227

J
Jeff Kirsher 已提交
1228
#ifdef CONFIG_DCB
1229
	/* Data Center Bridging netlink ops */
1230
	const struct dcbnl_rtnl_ops *dcbnl_ops;
1231
#endif
1232 1233 1234
	u8 num_tc;
	struct netdev_tc_txq tc_to_txq[TC_MAX_QUEUE];
	u8 prio_tc_map[TC_BITMASK + 1];
1235

1236 1237 1238 1239
#if defined(CONFIG_FCOE) || defined(CONFIG_FCOE_MODULE)
	/* max exchange id for FCoE LRO by ddp */
	unsigned int		fcoe_ddp_xid;
#endif
1240 1241
	/* phy device may attach itself for hardware timestamping */
	struct phy_device *phydev;
1242 1243 1244

	/* group the device belongs to */
	int group;
L
Linus Torvalds 已提交
1245
};
1246
#define to_net_dev(d) container_of(d, struct net_device, dev)
L
Linus Torvalds 已提交
1247 1248 1249

#define	NETDEV_ALIGN		32

1250 1251 1252 1253 1254 1255 1256 1257 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 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300
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;
}

1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319
static inline
struct netdev_queue *netdev_get_tx_queue(const struct net_device *dev,
					 unsigned int index)
{
	return &dev->_tx[index];
}

static inline void netdev_for_each_tx_queue(struct net_device *dev,
					    void (*f)(struct net_device *,
						      struct netdev_queue *,
						      void *),
					    void *arg)
{
	unsigned int i;

	for (i = 0; i < dev->num_tx_queues; i++)
		f(dev, &dev->_tx[i], arg);
}

1320 1321 1322 1323 1324 1325
/*
 * Net namespace inlines
 */
static inline
struct net *dev_net(const struct net_device *dev)
{
E
Eric Dumazet 已提交
1326
	return read_pnet(&dev->nd_net);
1327 1328 1329
}

static inline
1330
void dev_net_set(struct net_device *dev, struct net *net)
1331 1332
{
#ifdef CONFIG_NET_NS
1333 1334
	release_net(dev->nd_net);
	dev->nd_net = hold_net(net);
1335 1336 1337
#endif
}

1338 1339 1340 1341 1342 1343 1344 1345 1346 1347
static inline bool netdev_uses_dsa_tags(struct net_device *dev)
{
#ifdef CONFIG_NET_DSA_TAG_DSA
	if (dev->dsa_ptr != NULL)
		return dsa_uses_dsa_tags(dev->dsa_ptr);
#endif

	return 0;
}

1348 1349 1350 1351 1352 1353 1354 1355 1356
#ifndef CONFIG_NET_NS
static inline void skb_set_dev(struct sk_buff *skb, struct net_device *dev)
{
	skb->dev = dev;
}
#else /* CONFIG_NET_NS */
void skb_set_dev(struct sk_buff *skb, struct net_device *dev);
#endif

1357 1358 1359 1360 1361 1362 1363 1364 1365 1366
static inline bool netdev_uses_trailer_tags(struct net_device *dev)
{
#ifdef CONFIG_NET_DSA_TAG_TRAILER
	if (dev->dsa_ptr != NULL)
		return dsa_uses_trailer_tags(dev->dsa_ptr);
#endif

	return 0;
}

1367 1368 1369 1370 1371 1372
/**
 *	netdev_priv - access network device private data
 *	@dev: network device
 *
 * Get network device private data
 */
1373
static inline void *netdev_priv(const struct net_device *dev)
L
Linus Torvalds 已提交
1374
{
1375
	return (char *)dev + ALIGN(sizeof(struct net_device), NETDEV_ALIGN);
L
Linus Torvalds 已提交
1376 1377 1378 1379 1380
}

/* Set the sysfs physical device reference for the network logical device
 * if set prior to registration will cause a symlink during initialization.
 */
1381
#define SET_NETDEV_DEV(net, pdev)	((net)->dev.parent = (pdev))
L
Linus Torvalds 已提交
1382

1383 1384 1385 1386 1387 1388
/* Set the sysfs device type for the network logical device to allow
 * fin grained indentification of different network device types. For
 * example Ethernet, Wirelss LAN, Bluetooth, WiMAX etc.
 */
#define SET_NETDEV_DEVTYPE(net, devtype)	((net)->dev.type = (devtype))

1389 1390 1391 1392 1393 1394 1395 1396 1397 1398
/**
 *	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.
 */
1399 1400
void netif_napi_add(struct net_device *dev, struct napi_struct *napi,
		    int (*poll)(struct napi_struct *, int), int weight);
1401

1402 1403 1404 1405 1406 1407
/**
 *  netif_napi_del - remove a napi context
 *  @napi: napi context
 *
 *  netif_napi_del() removes a napi context from the network device napi list
 */
1408 1409 1410
void netif_napi_del(struct napi_struct *napi);

struct napi_gro_cb {
1411 1412 1413
	/* Virtual address of skb_shinfo(skb)->frags[0].page + offset. */
	void *frag0;

1414 1415 1416
	/* Length of frag0. */
	unsigned int frag0_len;

1417 1418 1419
	/* This indicates where we are processing relative to skb->data. */
	int data_offset;

1420 1421 1422 1423 1424 1425 1426 1427
	/* This is non-zero if the packet may be of the same flow. */
	int same_flow;

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

	/* Number of segments aggregated. */
	int count;
H
Herbert Xu 已提交
1428 1429 1430

	/* Free the skb? */
	int free;
1431 1432 1433
};

#define NAPI_GRO_CB(skb) ((struct napi_gro_cb *)(skb)->cb)
1434

L
Linus Torvalds 已提交
1435
struct packet_type {
D
David S. Miller 已提交
1436 1437 1438 1439 1440 1441
	__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 *);
1442
	struct sk_buff		*(*gso_segment)(struct sk_buff *skb,
1443
						netdev_features_t features);
1444
	int			(*gso_send_check)(struct sk_buff *skb);
1445 1446 1447
	struct sk_buff		**(*gro_receive)(struct sk_buff **head,
					       struct sk_buff *skb);
	int			(*gro_complete)(struct sk_buff *skb);
L
Linus Torvalds 已提交
1448 1449 1450 1451 1452 1453
	void			*af_packet_priv;
	struct list_head	list;
};

#include <linux/notifier.h>

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 1483 1484 1485 1486
/* netdevice notifier chain. Please remember to update the rtnetlink
 * notification exclusion list in rtnetlink_event() when adding new
 * types.
 */
#define NETDEV_UP	0x0001	/* For now you can't veto a device up/down */
#define NETDEV_DOWN	0x0002
#define NETDEV_REBOOT	0x0003	/* Tell a protocol stack a network interface
				   detected a hardware crash and restarted
				   - we can use this eg to kick tcp sessions
				   once done */
#define NETDEV_CHANGE	0x0004	/* Notify device state change */
#define NETDEV_REGISTER 0x0005
#define NETDEV_UNREGISTER	0x0006
#define NETDEV_CHANGEMTU	0x0007
#define NETDEV_CHANGEADDR	0x0008
#define NETDEV_GOING_DOWN	0x0009
#define NETDEV_CHANGENAME	0x000A
#define NETDEV_FEAT_CHANGE	0x000B
#define NETDEV_BONDING_FAILOVER 0x000C
#define NETDEV_PRE_UP		0x000D
#define NETDEV_PRE_TYPE_CHANGE	0x000E
#define NETDEV_POST_TYPE_CHANGE	0x000F
#define NETDEV_POST_INIT	0x0010
#define NETDEV_UNREGISTER_BATCH 0x0011
#define NETDEV_RELEASE		0x0012
#define NETDEV_NOTIFY_PEERS	0x0013
#define NETDEV_JOIN		0x0014

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


L
Linus Torvalds 已提交
1487 1488
extern rwlock_t				dev_base_lock;		/* Device list lock */

1489

1490 1491
#define for_each_netdev(net, d)		\
		list_for_each_entry(d, &(net)->dev_base_head, dev_list)
1492 1493
#define for_each_netdev_reverse(net, d)	\
		list_for_each_entry_reverse(d, &(net)->dev_base_head, dev_list)
1494 1495
#define for_each_netdev_rcu(net, d)		\
		list_for_each_entry_rcu(d, &(net)->dev_base_head, dev_list)
1496 1497 1498 1499
#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)
1500 1501
#define for_each_netdev_continue_rcu(net, d)		\
	list_for_each_entry_continue_rcu(d, &(net)->dev_base_head, dev_list)
1502
#define net_device_entry(lh)	list_entry(lh, struct net_device, dev_list)
1503

1504 1505 1506 1507 1508
static inline struct net_device *next_net_device(struct net_device *dev)
{
	struct list_head *lh;
	struct net *net;

1509
	net = dev_net(dev);
1510 1511 1512 1513
	lh = dev->dev_list.next;
	return lh == &net->dev_base_head ? NULL : net_device_entry(lh);
}

1514 1515 1516 1517 1518 1519
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
1520
	lh = rcu_dereference(list_next_rcu(&dev->dev_list));
1521 1522 1523
	return lh == &net->dev_base_head ? NULL : net_device_entry(lh);
}

1524 1525 1526 1527 1528
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);
}
1529

E
Eric Dumazet 已提交
1530 1531 1532 1533 1534 1535 1536
static inline struct net_device *first_net_device_rcu(struct net *net)
{
	struct list_head *lh = rcu_dereference(list_next_rcu(&net->dev_base_head));

	return lh == &net->dev_base_head ? NULL : net_device_entry(lh);
}

L
Linus Torvalds 已提交
1537 1538
extern int 			netdev_boot_setup_check(struct net_device *dev);
extern unsigned long		netdev_boot_base(const char *prefix, int unit);
1539 1540
extern struct net_device *dev_getbyhwaddr_rcu(struct net *net, unsigned short type,
					      const char *hwaddr);
1541 1542
extern struct net_device *dev_getfirstbyhwtype(struct net *net, unsigned short type);
extern struct net_device *__dev_getfirstbyhwtype(struct net *net, unsigned short type);
L
Linus Torvalds 已提交
1543 1544 1545 1546
extern void		dev_add_pack(struct packet_type *pt);
extern void		dev_remove_pack(struct packet_type *pt);
extern void		__dev_remove_pack(struct packet_type *pt);

E
Eric Dumazet 已提交
1547 1548
extern struct net_device	*dev_get_by_flags_rcu(struct net *net, unsigned short flags,
						      unsigned short mask);
1549
extern struct net_device	*dev_get_by_name(struct net *net, const char *name);
1550
extern struct net_device	*dev_get_by_name_rcu(struct net *net, const char *name);
1551
extern struct net_device	*__dev_get_by_name(struct net *net, const char *name);
L
Linus Torvalds 已提交
1552 1553 1554
extern int		dev_alloc_name(struct net_device *dev, const char *name);
extern int		dev_open(struct net_device *dev);
extern int		dev_close(struct net_device *dev);
1555
extern void		dev_disable_lro(struct net_device *dev);
L
Linus Torvalds 已提交
1556 1557
extern int		dev_queue_xmit(struct sk_buff *skb);
extern int		register_netdevice(struct net_device *dev);
1558 1559
extern void		unregister_netdevice_queue(struct net_device *dev,
						   struct list_head *head);
1560
extern void		unregister_netdevice_many(struct list_head *head);
1561 1562 1563 1564 1565
static inline void unregister_netdevice(struct net_device *dev)
{
	unregister_netdevice_queue(dev, NULL);
}

E
Eric Dumazet 已提交
1566
extern int 		netdev_refcnt_read(const struct net_device *dev);
L
Linus Torvalds 已提交
1567 1568
extern void		free_netdev(struct net_device *dev);
extern void		synchronize_net(void);
1569
extern int		init_dummy_netdev(struct net_device *dev);
1570
extern void		netdev_resync_ops(struct net_device *dev);
1571

1572 1573
extern struct net_device	*dev_get_by_index(struct net *net, int ifindex);
extern struct net_device	*__dev_get_by_index(struct net *net, int ifindex);
1574
extern struct net_device	*dev_get_by_index_rcu(struct net *net, int ifindex);
L
Linus Torvalds 已提交
1575 1576 1577 1578
extern int		dev_restart(struct net_device *dev);
#ifdef CONFIG_NETPOLL_TRAP
extern int		netpoll_trap(void);
#endif
1579 1580
extern int	       skb_gro_receive(struct sk_buff **head,
				       struct sk_buff *skb);
1581
extern void	       skb_gro_reset_offset(struct sk_buff *skb);
1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597

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;
}

1598 1599
static inline void *skb_gro_header_fast(struct sk_buff *skb,
					unsigned int offset)
1600
{
1601 1602
	return NAPI_GRO_CB(skb)->frag0 + offset;
}
1603

1604 1605 1606 1607
static inline int skb_gro_header_hard(struct sk_buff *skb, unsigned int hlen)
{
	return NAPI_GRO_CB(skb)->frag0_len < hlen;
}
1608

1609 1610 1611
static inline void *skb_gro_header_slow(struct sk_buff *skb, unsigned int hlen,
					unsigned int offset)
{
1612 1613 1614
	if (!pskb_may_pull(skb, hlen))
		return NULL;

1615 1616
	NAPI_GRO_CB(skb)->frag0 = NULL;
	NAPI_GRO_CB(skb)->frag0_len = 0;
1617
	return skb->data + offset;
1618
}
L
Linus Torvalds 已提交
1619

1620 1621
static inline void *skb_gro_mac_header(struct sk_buff *skb)
{
1622
	return NAPI_GRO_CB(skb)->frag0 ?: skb_mac_header(skb);
1623 1624
}

H
Herbert Xu 已提交
1625 1626
static inline void *skb_gro_network_header(struct sk_buff *skb)
{
1627 1628
	return (NAPI_GRO_CB(skb)->frag0 ?: skb->data) +
	       skb_network_offset(skb);
H
Herbert Xu 已提交
1629 1630
}

1631 1632
static inline int dev_hard_header(struct sk_buff *skb, struct net_device *dev,
				  unsigned short type,
1633 1634
				  const void *daddr, const void *saddr,
				  unsigned len)
1635
{
1636
	if (!dev->header_ops || !dev->header_ops->create)
1637
		return 0;
1638 1639

	return dev->header_ops->create(skb, dev, type, daddr, saddr, len);
1640 1641
}

S
Stephen Hemminger 已提交
1642 1643 1644 1645 1646
static inline int dev_parse_header(const struct sk_buff *skb,
				   unsigned char *haddr)
{
	const struct net_device *dev = skb->dev;

1647
	if (!dev->header_ops || !dev->header_ops->parse)
S
Stephen Hemminger 已提交
1648
		return 0;
1649
	return dev->header_ops->parse(skb, haddr);
S
Stephen Hemminger 已提交
1650 1651
}

L
Linus Torvalds 已提交
1652 1653 1654 1655 1656 1657 1658 1659
typedef int gifconf_func_t(struct net_device * dev, char __user * bufptr, int len);
extern int		register_gifconf(unsigned int family, gifconf_func_t * gifconf);
static inline int unregister_gifconf(unsigned int family)
{
	return register_gifconf(family, NULL);
}

/*
E
Eric Dumazet 已提交
1660
 * Incoming packets are placed on per-cpu queues
L
Linus Torvalds 已提交
1661
 */
E
Eric Dumazet 已提交
1662
struct softnet_data {
1663
	struct Qdisc		*output_queue;
1664
	struct Qdisc		**output_queue_tailp;
L
Linus Torvalds 已提交
1665 1666
	struct list_head	poll_list;
	struct sk_buff		*completion_queue;
1667
	struct sk_buff_head	process_queue;
L
Linus Torvalds 已提交
1668

C
Changli Gao 已提交
1669
	/* stats */
1670 1671 1672 1673
	unsigned int		processed;
	unsigned int		time_squeeze;
	unsigned int		cpu_collision;
	unsigned int		received_rps;
C
Changli Gao 已提交
1674

1675
#ifdef CONFIG_RPS
E
Eric Dumazet 已提交
1676 1677 1678
	struct softnet_data	*rps_ipi_list;

	/* Elements below can be accessed between CPUs for RPS */
T
Tom Herbert 已提交
1679
	struct call_single_data	csd ____cacheline_aligned_in_smp;
E
Eric Dumazet 已提交
1680 1681
	struct softnet_data	*rps_ipi_next;
	unsigned int		cpu;
T
Tom Herbert 已提交
1682
	unsigned int		input_queue_head;
1683
	unsigned int		input_queue_tail;
1684
#endif
C
Changli Gao 已提交
1685
	unsigned		dropped;
T
Tom Herbert 已提交
1686
	struct sk_buff_head	input_pkt_queue;
1687
	struct napi_struct	backlog;
L
Linus Torvalds 已提交
1688 1689
};

1690
static inline void input_queue_head_incr(struct softnet_data *sd)
T
Tom Herbert 已提交
1691 1692
{
#ifdef CONFIG_RPS
1693 1694 1695 1696 1697 1698 1699 1700 1701
	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
1702 1703 1704
#endif
}

T
Tom Herbert 已提交
1705
DECLARE_PER_CPU_ALIGNED(struct softnet_data, softnet_data);
L
Linus Torvalds 已提交
1706

1707
extern void __netif_schedule(struct Qdisc *q);
L
Linus Torvalds 已提交
1708

1709
static inline void netif_schedule_queue(struct netdev_queue *txq)
L
Linus Torvalds 已提交
1710
{
1711
	if (!test_bit(__QUEUE_STATE_XOFF, &txq->state))
1712
		__netif_schedule(txq->qdisc);
1713 1714
}

1715 1716 1717 1718 1719 1720 1721 1722
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));
}

1723 1724 1725 1726 1727
static inline void netif_tx_start_queue(struct netdev_queue *dev_queue)
{
	clear_bit(__QUEUE_STATE_XOFF, &dev_queue->state);
}

1728 1729 1730 1731 1732 1733
/**
 *	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
1734 1735
static inline void netif_start_queue(struct net_device *dev)
{
1736
	netif_tx_start_queue(netdev_get_tx_queue(dev, 0));
L
Linus Torvalds 已提交
1737 1738
}

1739 1740 1741 1742 1743 1744 1745 1746 1747 1748
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);
	}
}

1749
static inline void netif_tx_wake_queue(struct netdev_queue *dev_queue)
L
Linus Torvalds 已提交
1750 1751
{
#ifdef CONFIG_NETPOLL_TRAP
1752
	if (netpoll_trap()) {
1753
		netif_tx_start_queue(dev_queue);
L
Linus Torvalds 已提交
1754
		return;
1755
	}
L
Linus Torvalds 已提交
1756
#endif
1757
	if (test_and_clear_bit(__QUEUE_STATE_XOFF, &dev_queue->state))
1758
		__netif_schedule(dev_queue->qdisc);
1759 1760
}

1761 1762 1763 1764 1765 1766 1767
/**
 *	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.
 */
1768 1769
static inline void netif_wake_queue(struct net_device *dev)
{
1770
	netif_tx_wake_queue(netdev_get_tx_queue(dev, 0));
L
Linus Torvalds 已提交
1771 1772
}

1773 1774 1775 1776 1777 1778 1779 1780 1781 1782
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);
	}
}

1783 1784
static inline void netif_tx_stop_queue(struct netdev_queue *dev_queue)
{
1785
	if (WARN_ON(!dev_queue)) {
1786
		pr_info("netif_stop_queue() cannot be called before register_netdev()\n");
1787 1788
		return;
	}
1789 1790 1791
	set_bit(__QUEUE_STATE_XOFF, &dev_queue->state);
}

1792 1793 1794 1795 1796 1797 1798
/**
 *	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
1799 1800
static inline void netif_stop_queue(struct net_device *dev)
{
1801
	netif_tx_stop_queue(netdev_get_tx_queue(dev, 0));
L
Linus Torvalds 已提交
1802 1803
}

1804 1805 1806 1807 1808 1809 1810 1811 1812 1813
static inline void netif_tx_stop_all_queues(struct net_device *dev)
{
	unsigned int i;

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

1814 1815 1816 1817 1818
static inline int netif_tx_queue_stopped(const struct netdev_queue *dev_queue)
{
	return test_bit(__QUEUE_STATE_XOFF, &dev_queue->state);
}

1819 1820 1821 1822 1823 1824
/**
 *	netif_queue_stopped - test if transmit queue is flowblocked
 *	@dev: network device
 *
 *	Test if transmit queue on device is currently unable to send.
 */
L
Linus Torvalds 已提交
1825 1826
static inline int netif_queue_stopped(const struct net_device *dev)
{
1827
	return netif_tx_queue_stopped(netdev_get_tx_queue(dev, 0));
L
Linus Torvalds 已提交
1828 1829
}

1830
static inline int netif_tx_queue_frozen_or_stopped(const struct netdev_queue *dev_queue)
1831
{
1832
	return dev_queue->state & QUEUE_STATE_XOFF_OR_FROZEN;
1833 1834
}

1835 1836 1837 1838 1839 1840
/**
 *	netif_running - test if up
 *	@dev: network device
 *
 *	Test if the device has been brought up.
 */
L
Linus Torvalds 已提交
1841 1842 1843 1844 1845
static inline int netif_running(const struct net_device *dev)
{
	return test_bit(__LINK_STATE_START, &dev->state);
}

1846 1847 1848 1849 1850 1851
/*
 * 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.
 */
1852 1853 1854 1855 1856 1857 1858 1859

/**
 *	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.
 */
1860 1861
static inline void netif_start_subqueue(struct net_device *dev, u16 queue_index)
{
1862
	struct netdev_queue *txq = netdev_get_tx_queue(dev, queue_index);
1863 1864

	netif_tx_start_queue(txq);
1865 1866
}

1867 1868 1869 1870 1871 1872 1873
/**
 *	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.
 */
1874 1875
static inline void netif_stop_subqueue(struct net_device *dev, u16 queue_index)
{
1876
	struct netdev_queue *txq = netdev_get_tx_queue(dev, queue_index);
1877 1878 1879 1880
#ifdef CONFIG_NETPOLL_TRAP
	if (netpoll_trap())
		return;
#endif
1881
	netif_tx_stop_queue(txq);
1882 1883
}

1884 1885 1886 1887 1888 1889 1890
/**
 *	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.
 */
1891
static inline int __netif_subqueue_stopped(const struct net_device *dev,
1892 1893
					 u16 queue_index)
{
1894
	struct netdev_queue *txq = netdev_get_tx_queue(dev, queue_index);
1895 1896

	return netif_tx_queue_stopped(txq);
1897 1898
}

1899 1900 1901 1902 1903
static inline int netif_subqueue_stopped(const struct net_device *dev,
					 struct sk_buff *skb)
{
	return __netif_subqueue_stopped(dev, skb_get_queue_mapping(skb));
}
1904 1905 1906 1907 1908 1909 1910 1911

/**
 *	netif_wake_subqueue - allow sending packets on subqueue
 *	@dev: network device
 *	@queue_index: sub queue index
 *
 * Resume individual transmit queue of a device with multiple transmit queues.
 */
1912 1913
static inline void netif_wake_subqueue(struct net_device *dev, u16 queue_index)
{
1914
	struct netdev_queue *txq = netdev_get_tx_queue(dev, queue_index);
1915 1916 1917 1918
#ifdef CONFIG_NETPOLL_TRAP
	if (netpoll_trap())
		return;
#endif
1919
	if (test_and_clear_bit(__QUEUE_STATE_XOFF, &txq->state))
1920
		__netif_schedule(txq->qdisc);
1921 1922
}

1923 1924 1925 1926 1927 1928 1929 1930 1931 1932
/*
 * Returns a Tx hash for the given packet when dev->real_num_tx_queues is used
 * as a distribution range limit for the returned value.
 */
static inline u16 skb_tx_hash(const struct net_device *dev,
			      const struct sk_buff *skb)
{
	return __skb_tx_hash(dev, skb, dev->real_num_tx_queues);
}

1933 1934 1935 1936 1937 1938
/**
 *	netif_is_multiqueue - test if device has multiple transmit queues
 *	@dev: network device
 *
 * Check if device has multiple transmit queues
 */
1939 1940
static inline int netif_is_multiqueue(const struct net_device *dev)
{
E
Eric Dumazet 已提交
1941
	return dev->num_tx_queues > 1;
1942
}
L
Linus Torvalds 已提交
1943

1944 1945
extern int netif_set_real_num_tx_queues(struct net_device *dev,
					unsigned int txq);
1946

1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957
#ifdef CONFIG_RPS
extern int netif_set_real_num_rx_queues(struct net_device *dev,
					unsigned int rxq);
#else
static inline int netif_set_real_num_rx_queues(struct net_device *dev,
						unsigned int rxq)
{
	return 0;
}
#endif

1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969
static inline int netif_copy_real_num_queues(struct net_device *to_dev,
					     const struct net_device *from_dev)
{
	netif_set_real_num_tx_queues(to_dev, from_dev->real_num_tx_queues);
#ifdef CONFIG_RPS
	return netif_set_real_num_rx_queues(to_dev,
					    from_dev->real_num_rx_queues);
#else
	return 0;
#endif
}

L
Linus Torvalds 已提交
1970
/* Use this variant when it is known for sure that it
1971 1972
 * is executing from hardware interrupt context or with hardware interrupts
 * disabled.
L
Linus Torvalds 已提交
1973
 */
1974
extern void dev_kfree_skb_irq(struct sk_buff *skb);
L
Linus Torvalds 已提交
1975 1976

/* Use this variant in places where it could be invoked
1977 1978
 * from either hardware interrupt or other context, with hardware interrupts
 * either disabled or enabled.
L
Linus Torvalds 已提交
1979
 */
1980
extern void dev_kfree_skb_any(struct sk_buff *skb);
L
Linus Torvalds 已提交
1981 1982 1983 1984

extern int		netif_rx(struct sk_buff *skb);
extern int		netif_rx_ni(struct sk_buff *skb);
extern int		netif_receive_skb(struct sk_buff *skb);
1985
extern gro_result_t	dev_gro_receive(struct napi_struct *napi,
1986
					struct sk_buff *skb);
1987 1988
extern gro_result_t	napi_skb_finish(gro_result_t ret, struct sk_buff *skb);
extern gro_result_t	napi_gro_receive(struct napi_struct *napi,
1989
					 struct sk_buff *skb);
E
Eric Dumazet 已提交
1990
extern void		napi_gro_flush(struct napi_struct *napi);
1991
extern struct sk_buff *	napi_get_frags(struct napi_struct *napi);
1992
extern gro_result_t	napi_frags_finish(struct napi_struct *napi,
1993 1994
					  struct sk_buff *skb,
					  gro_result_t ret);
1995
extern struct sk_buff *	napi_frags_skb(struct napi_struct *napi);
1996
extern gro_result_t	napi_gro_frags(struct napi_struct *napi);
1997 1998 1999 2000 2001 2002 2003

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

2004
extern int netdev_rx_handler_register(struct net_device *dev,
J
Jiri Pirko 已提交
2005 2006
				      rx_handler_func_t *rx_handler,
				      void *rx_handler_data);
2007 2008
extern void netdev_rx_handler_unregister(struct net_device *dev);

2009
extern int		dev_valid_name(const char *name);
2010 2011
extern int		dev_ioctl(struct net *net, unsigned int cmd, void __user *);
extern int		dev_ethtool(struct net *net, struct ifreq *);
L
Linus Torvalds 已提交
2012
extern unsigned		dev_get_flags(const struct net_device *);
2013
extern int		__dev_change_flags(struct net_device *, unsigned int flags);
L
Linus Torvalds 已提交
2014
extern int		dev_change_flags(struct net_device *, unsigned);
2015
extern void		__dev_notify_flags(struct net_device *, unsigned int old_flags);
2016
extern int		dev_change_name(struct net_device *, const char *);
2017
extern int		dev_set_alias(struct net_device *, const char *, size_t);
2018 2019
extern int		dev_change_net_namespace(struct net_device *,
						 struct net *, const char *);
L
Linus Torvalds 已提交
2020
extern int		dev_set_mtu(struct net_device *, int);
2021
extern void		dev_set_group(struct net_device *, int);
L
Linus Torvalds 已提交
2022 2023
extern int		dev_set_mac_address(struct net_device *,
					    struct sockaddr *);
2024
extern int		dev_hard_start_xmit(struct sk_buff *skb,
2025 2026
					    struct net_device *dev,
					    struct netdev_queue *txq);
2027 2028
extern int		dev_forward_skb(struct net_device *dev,
					struct sk_buff *skb);
L
Linus Torvalds 已提交
2029

2030
extern int		netdev_budget;
L
Linus Torvalds 已提交
2031 2032 2033 2034

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

2035 2036 2037 2038
/**
 *	dev_put - release reference to device
 *	@dev: network device
 *
2039
 * Release reference to device to allow it to be freed.
2040
 */
L
Linus Torvalds 已提交
2041 2042
static inline void dev_put(struct net_device *dev)
{
E
Eric Dumazet 已提交
2043
	irqsafe_cpu_dec(*dev->pcpu_refcnt);
L
Linus Torvalds 已提交
2044 2045
}

2046 2047 2048 2049
/**
 *	dev_hold - get reference to device
 *	@dev: network device
 *
2050
 * Hold reference to device to keep it from being freed.
2051
 */
2052 2053
static inline void dev_hold(struct net_device *dev)
{
E
Eric Dumazet 已提交
2054
	irqsafe_cpu_inc(*dev->pcpu_refcnt);
2055
}
L
Linus Torvalds 已提交
2056 2057 2058 2059

/* 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
2060 2061 2062 2063
 *
 * 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
2064 2065 2066
 */

extern void linkwatch_fire_event(struct net_device *dev);
2067
extern void linkwatch_forget_dev(struct net_device *dev);
L
Linus Torvalds 已提交
2068

2069 2070 2071 2072 2073 2074
/**
 *	netif_carrier_ok - test if carrier present
 *	@dev: network device
 *
 * Check if carrier is present on device
 */
L
Linus Torvalds 已提交
2075 2076 2077 2078 2079
static inline int netif_carrier_ok(const struct net_device *dev)
{
	return !test_bit(__LINK_STATE_NOCARRIER, &dev->state);
}

2080 2081
extern unsigned long dev_trans_start(struct net_device *dev);

L
Linus Torvalds 已提交
2082 2083
extern void __netdev_watchdog_up(struct net_device *dev);

2084
extern void netif_carrier_on(struct net_device *dev);
L
Linus Torvalds 已提交
2085

2086
extern void netif_carrier_off(struct net_device *dev);
L
Linus Torvalds 已提交
2087

2088 2089
extern void netif_notify_peers(struct net_device *dev);

2090 2091 2092 2093 2094 2095 2096 2097 2098 2099 2100 2101 2102
/**
 *	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
2103 2104 2105 2106 2107 2108
static inline void netif_dormant_on(struct net_device *dev)
{
	if (!test_and_set_bit(__LINK_STATE_DORMANT, &dev->state))
		linkwatch_fire_event(dev);
}

2109 2110 2111 2112 2113 2114
/**
 *	netif_dormant_off - set device as not dormant.
 *	@dev: network device
 *
 * Device is not in dormant state.
 */
S
Stefan Rompf 已提交
2115 2116 2117 2118 2119 2120
static inline void netif_dormant_off(struct net_device *dev)
{
	if (test_and_clear_bit(__LINK_STATE_DORMANT, &dev->state))
		linkwatch_fire_event(dev);
}

2121 2122 2123 2124 2125 2126
/**
 *	netif_dormant - test if carrier present
 *	@dev: network device
 *
 * Check if carrier is present on device
 */
S
Stefan Rompf 已提交
2127 2128 2129 2130 2131 2132
static inline int netif_dormant(const struct net_device *dev)
{
	return test_bit(__LINK_STATE_DORMANT, &dev->state);
}


2133 2134 2135 2136 2137 2138
/**
 *	netif_oper_up - test if device is operational
 *	@dev: network device
 *
 * Check if carrier is operational
 */
E
Eric Dumazet 已提交
2139 2140
static inline int netif_oper_up(const struct net_device *dev)
{
S
Stefan Rompf 已提交
2141 2142 2143 2144
	return (dev->operstate == IF_OPER_UP ||
		dev->operstate == IF_OPER_UNKNOWN /* backward compat */);
}

2145 2146 2147 2148 2149 2150
/**
 *	netif_device_present - is device available or removed
 *	@dev: network device
 *
 * Check if device has not been removed from system.
 */
L
Linus Torvalds 已提交
2151 2152 2153 2154 2155
static inline int netif_device_present(struct net_device *dev)
{
	return test_bit(__LINK_STATE_PRESENT, &dev->state);
}

2156
extern void netif_device_detach(struct net_device *dev);
L
Linus Torvalds 已提交
2157

2158
extern void netif_device_attach(struct net_device *dev);
L
Linus Torvalds 已提交
2159 2160 2161 2162 2163 2164 2165 2166 2167 2168 2169 2170 2171 2172 2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 2183 2184 2185 2186 2187 2188 2189 2190 2191 2192 2193 2194 2195 2196 2197 2198 2199 2200 2201 2202 2203 2204 2205 2206 2207 2208

/*
 * 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;
}

2209
static inline void __netif_tx_lock(struct netdev_queue *txq, int cpu)
H
Herbert Xu 已提交
2210
{
2211 2212
	spin_lock(&txq->_xmit_lock);
	txq->xmit_lock_owner = cpu;
2213 2214
}

2215 2216 2217 2218 2219 2220
static inline void __netif_tx_lock_bh(struct netdev_queue *txq)
{
	spin_lock_bh(&txq->_xmit_lock);
	txq->xmit_lock_owner = smp_processor_id();
}

2221 2222 2223 2224 2225 2226 2227 2228 2229 2230 2231 2232 2233 2234 2235 2236 2237 2238 2239 2240
static inline int __netif_tx_trylock(struct netdev_queue *txq)
{
	int ok = spin_trylock(&txq->_xmit_lock);
	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
2241 2242 2243 2244 2245 2246
static inline void txq_trans_update(struct netdev_queue *txq)
{
	if (txq->xmit_lock_owner != -1)
		txq->trans_start = jiffies;
}

2247 2248 2249 2250 2251 2252
/**
 *	netif_tx_lock - grab network device transmit lock
 *	@dev: network device
 *
 * Get network device transmit lock
 */
2253 2254
static inline void netif_tx_lock(struct net_device *dev)
{
2255
	unsigned int i;
2256
	int cpu;
2257

2258 2259
	spin_lock(&dev->tx_global_lock);
	cpu = smp_processor_id();
2260 2261
	for (i = 0; i < dev->num_tx_queues; i++) {
		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
2262 2263 2264 2265 2266 2267 2268

		/* 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.
		 */
2269
		__netif_tx_lock(txq, cpu);
2270 2271
		set_bit(__QUEUE_STATE_FROZEN, &txq->state);
		__netif_tx_unlock(txq);
2272
	}
H
Herbert Xu 已提交
2273 2274 2275 2276
}

static inline void netif_tx_lock_bh(struct net_device *dev)
{
2277 2278
	local_bh_disable();
	netif_tx_lock(dev);
H
Herbert Xu 已提交
2279 2280 2281 2282
}

static inline void netif_tx_unlock(struct net_device *dev)
{
2283 2284 2285 2286
	unsigned int i;

	for (i = 0; i < dev->num_tx_queues; i++) {
		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
2287

2288 2289 2290 2291 2292
		/* 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);
2293
		netif_schedule_queue(txq);
2294 2295
	}
	spin_unlock(&dev->tx_global_lock);
H
Herbert Xu 已提交
2296 2297 2298 2299
}

static inline void netif_tx_unlock_bh(struct net_device *dev)
{
2300 2301
	netif_tx_unlock(dev);
	local_bh_enable();
H
Herbert Xu 已提交
2302 2303
}

2304
#define HARD_TX_LOCK(dev, txq, cpu) {			\
2305
	if ((dev->features & NETIF_F_LLTX) == 0) {	\
2306
		__netif_tx_lock(txq, cpu);		\
2307 2308 2309
	}						\
}

2310
#define HARD_TX_UNLOCK(dev, txq) {			\
2311
	if ((dev->features & NETIF_F_LLTX) == 0) {	\
2312
		__netif_tx_unlock(txq);			\
2313 2314 2315
	}						\
}

L
Linus Torvalds 已提交
2316 2317
static inline void netif_tx_disable(struct net_device *dev)
{
2318
	unsigned int i;
2319
	int cpu;
2320

2321 2322
	local_bh_disable();
	cpu = smp_processor_id();
2323 2324
	for (i = 0; i < dev->num_tx_queues; i++) {
		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
2325 2326

		__netif_tx_lock(txq, cpu);
2327
		netif_tx_stop_queue(txq);
2328
		__netif_tx_unlock(txq);
2329
	}
2330
	local_bh_enable();
L
Linus Torvalds 已提交
2331 2332
}

2333 2334 2335 2336 2337 2338 2339 2340 2341 2342 2343 2344 2345 2346 2347 2348 2349 2350 2351 2352
static inline void netif_addr_lock(struct net_device *dev)
{
	spin_lock(&dev->addr_list_lock);
}

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);
}

2353
/*
2354
 * dev_addrs walker. Should be used only for read access. Call with
2355 2356 2357
 * rcu_read_lock held.
 */
#define for_each_dev_addr(dev, ha) \
2358
		list_for_each_entry_rcu(ha, &dev->dev_addrs.list, list)
2359

L
Linus Torvalds 已提交
2360 2361 2362 2363 2364
/* These functions live elsewhere (drivers/net/net_init.c, but related) */

extern void		ether_setup(struct net_device *dev);

/* Support for loadable net-drivers */
T
Tom Herbert 已提交
2365
extern struct net_device *alloc_netdev_mqs(int sizeof_priv, const char *name,
2366
				       void (*setup)(struct net_device *),
T
Tom Herbert 已提交
2367
				       unsigned int txqs, unsigned int rxqs);
2368
#define alloc_netdev(sizeof_priv, name, setup) \
T
Tom Herbert 已提交
2369 2370 2371 2372 2373
	alloc_netdev_mqs(sizeof_priv, name, setup, 1, 1)

#define alloc_netdev_mq(sizeof_priv, name, setup, count) \
	alloc_netdev_mqs(sizeof_priv, name, setup, count, count)

L
Linus Torvalds 已提交
2374 2375
extern int		register_netdev(struct net_device *dev);
extern void		unregister_netdev(struct net_device *dev);
2376

2377 2378 2379 2380 2381 2382 2383 2384 2385 2386 2387 2388 2389 2390 2391 2392
/* General hardware address lists handling functions */
extern int __hw_addr_add_multiple(struct netdev_hw_addr_list *to_list,
				  struct netdev_hw_addr_list *from_list,
				  int addr_len, unsigned char addr_type);
extern void __hw_addr_del_multiple(struct netdev_hw_addr_list *to_list,
				   struct netdev_hw_addr_list *from_list,
				   int addr_len, unsigned char addr_type);
extern int __hw_addr_sync(struct netdev_hw_addr_list *to_list,
			  struct netdev_hw_addr_list *from_list,
			  int addr_len);
extern void __hw_addr_unsync(struct netdev_hw_addr_list *to_list,
			     struct netdev_hw_addr_list *from_list,
			     int addr_len);
extern void __hw_addr_flush(struct netdev_hw_addr_list *list);
extern void __hw_addr_init(struct netdev_hw_addr_list *list);

2393 2394 2395 2396 2397 2398 2399 2400 2401 2402 2403
/* Functions used for device addresses handling */
extern int dev_addr_add(struct net_device *dev, unsigned char *addr,
			unsigned char addr_type);
extern int dev_addr_del(struct net_device *dev, unsigned char *addr,
			unsigned char addr_type);
extern int dev_addr_add_multiple(struct net_device *to_dev,
				 struct net_device *from_dev,
				 unsigned char addr_type);
extern int dev_addr_del_multiple(struct net_device *to_dev,
				 struct net_device *from_dev,
				 unsigned char addr_type);
2404 2405 2406 2407 2408 2409 2410 2411 2412 2413
extern void dev_addr_flush(struct net_device *dev);
extern int dev_addr_init(struct net_device *dev);

/* Functions used for unicast addresses handling */
extern int dev_uc_add(struct net_device *dev, unsigned char *addr);
extern int dev_uc_del(struct net_device *dev, unsigned char *addr);
extern int dev_uc_sync(struct net_device *to, struct net_device *from);
extern void dev_uc_unsync(struct net_device *to, struct net_device *from);
extern void dev_uc_flush(struct net_device *dev);
extern void dev_uc_init(struct net_device *dev);
2414

2415 2416 2417 2418 2419 2420 2421 2422 2423
/* Functions used for multicast addresses handling */
extern int dev_mc_add(struct net_device *dev, unsigned char *addr);
extern int dev_mc_add_global(struct net_device *dev, unsigned char *addr);
extern int dev_mc_del(struct net_device *dev, unsigned char *addr);
extern int dev_mc_del_global(struct net_device *dev, unsigned char *addr);
extern int dev_mc_sync(struct net_device *to, struct net_device *from);
extern void dev_mc_unsync(struct net_device *to, struct net_device *from);
extern void dev_mc_flush(struct net_device *dev);
extern void dev_mc_init(struct net_device *dev);
2424

2425 2426 2427
/* Functions used for secondary unicast and multicast support */
extern void		dev_set_rx_mode(struct net_device *dev);
extern void		__dev_set_rx_mode(struct net_device *dev);
2428 2429
extern int		dev_set_promiscuity(struct net_device *dev, int inc);
extern int		dev_set_allmulti(struct net_device *dev, int inc);
L
Linus Torvalds 已提交
2430
extern void		netdev_state_change(struct net_device *dev);
2431
extern int		netdev_bonding_change(struct net_device *dev,
2432
					      unsigned long event);
2433
extern void		netdev_features_change(struct net_device *dev);
L
Linus Torvalds 已提交
2434
/* Load a device via the kmod */
2435
extern void		dev_load(struct net *net, const char *name);
L
Linus Torvalds 已提交
2436
extern void		dev_mcast_init(void);
2437 2438
extern struct rtnl_link_stats64 *dev_get_stats(struct net_device *dev,
					       struct rtnl_link_stats64 *storage);
2439

L
Linus Torvalds 已提交
2440
extern int		netdev_max_backlog;
E
Eric Dumazet 已提交
2441
extern int		netdev_tstamp_prequeue;
L
Linus Torvalds 已提交
2442
extern int		weight_p;
2443
extern int		bpf_jit_enable;
L
Linus Torvalds 已提交
2444
extern int		netdev_set_master(struct net_device *dev, struct net_device *master);
J
Jiri Pirko 已提交
2445 2446
extern int netdev_set_bond_master(struct net_device *dev,
				  struct net_device *master);
2447
extern int skb_checksum_help(struct sk_buff *skb);
2448 2449
extern struct sk_buff *skb_gso_segment(struct sk_buff *skb,
	netdev_features_t features);
2450 2451 2452 2453 2454 2455 2456
#ifdef CONFIG_BUG
extern void netdev_rx_csum_fault(struct net_device *dev);
#else
static inline void netdev_rx_csum_fault(struct net_device *dev)
{
}
#endif
L
Linus Torvalds 已提交
2457 2458 2459 2460
/* rx skb timestamps */
extern void		net_enable_timestamp(void);
extern void		net_disable_timestamp(void);

2461 2462 2463 2464 2465 2466
#ifdef CONFIG_PROC_FS
extern void *dev_seq_start(struct seq_file *seq, loff_t *pos);
extern void *dev_seq_next(struct seq_file *seq, void *v, loff_t *pos);
extern void dev_seq_stop(struct seq_file *seq, void *v);
#endif

2467 2468 2469
extern int netdev_class_create_file(struct class_attribute *class_attr);
extern void netdev_class_remove_file(struct class_attribute *class_attr);

2470 2471
extern struct kobj_ns_type_operations net_ns_type_operations;

2472
extern const char *netdev_drivername(const struct net_device *dev);
2473

2474 2475
extern void linkwatch_run_queue(void);

2476 2477
static inline netdev_features_t netdev_get_wanted_features(
	struct net_device *dev)
2478 2479 2480
{
	return (dev->features & ~dev->hw_features) | dev->wanted_features;
}
2481 2482
netdev_features_t netdev_increment_features(netdev_features_t all,
	netdev_features_t one, netdev_features_t mask);
2483
int __netdev_update_features(struct net_device *dev);
2484
void netdev_update_features(struct net_device *dev);
2485
void netdev_change_features(struct net_device *dev);
2486

2487 2488 2489
void netif_stacked_transfer_operstate(const struct net_device *rootdev,
					struct net_device *dev);

2490
netdev_features_t netif_skb_features(struct sk_buff *skb);
2491

2492
static inline int net_gso_ok(netdev_features_t features, int gso_type)
2493
{
2494
	netdev_features_t feature = gso_type << NETIF_F_GSO_SHIFT;
2495 2496 2497 2498 2499 2500 2501 2502 2503

	/* 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));

2504
	return (features & feature) == feature;
2505 2506
}

2507
static inline int skb_gso_ok(struct sk_buff *skb, netdev_features_t features)
2508
{
H
Herbert Xu 已提交
2509
	return net_gso_ok(features, skb_shinfo(skb)->gso_type) &&
2510
	       (!skb_has_frag_list(skb) || (features & NETIF_F_FRAGLIST));
2511 2512
}

2513 2514
static inline int netif_needs_gso(struct sk_buff *skb,
	netdev_features_t features)
2515
{
2516 2517
	return skb_is_gso(skb) && (!skb_gso_ok(skb, features) ||
		unlikely(skb->ip_summed != CHECKSUM_PARTIAL));
2518 2519
}

2520 2521 2522 2523 2524 2525
static inline void netif_set_gso_max_size(struct net_device *dev,
					  unsigned int size)
{
	dev->gso_max_size = size;
}

J
Jiri Pirko 已提交
2526 2527 2528 2529 2530
static inline int netif_is_bond_slave(struct net_device *dev)
{
	return dev->flags & IFF_SLAVE && dev->priv_flags & IFF_BONDING;
}

2531
extern struct pernet_operations __net_initdata loopback_net_ops;
2532

2533 2534 2535 2536 2537 2538 2539 2540 2541 2542 2543
/* Logging, debugging and troubleshooting/diagnostic helpers. */

/* netdev_printk helpers, similar to dev_printk */

static inline const char *netdev_name(const struct net_device *dev)
{
	if (dev->reg_state != NETREG_REGISTERED)
		return "(unregistered net_device)";
	return dev->name;
}

2544 2545 2546
extern int __netdev_printk(const char *level, const struct net_device *dev,
			struct va_format *vaf);

2547 2548 2549 2550 2551 2552 2553 2554 2555 2556 2557 2558 2559 2560 2561 2562 2563
extern __printf(3, 4)
int netdev_printk(const char *level, const struct net_device *dev,
		  const char *format, ...);
extern __printf(2, 3)
int netdev_emerg(const struct net_device *dev, const char *format, ...);
extern __printf(2, 3)
int netdev_alert(const struct net_device *dev, const char *format, ...);
extern __printf(2, 3)
int netdev_crit(const struct net_device *dev, const char *format, ...);
extern __printf(2, 3)
int netdev_err(const struct net_device *dev, const char *format, ...);
extern __printf(2, 3)
int netdev_warn(const struct net_device *dev, const char *format, ...);
extern __printf(2, 3)
int netdev_notice(const struct net_device *dev, const char *format, ...);
extern __printf(2, 3)
int netdev_info(const struct net_device *dev, const char *format, ...);
2564

2565 2566 2567
#define MODULE_ALIAS_NETDEV(device) \
	MODULE_ALIAS("netdev-" device)

2568 2569 2570 2571 2572 2573
#if defined(DEBUG)
#define netdev_dbg(__dev, format, args...)			\
	netdev_printk(KERN_DEBUG, __dev, format, ##args)
#elif defined(CONFIG_DYNAMIC_DEBUG)
#define netdev_dbg(__dev, format, args...)			\
do {								\
2574
	dynamic_netdev_dbg(__dev, format, ##args);		\
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
} while (0)
#else
#define netdev_dbg(__dev, format, args...)			\
({								\
	if (0)							\
		netdev_printk(KERN_DEBUG, __dev, format, ##args); \
	0;							\
})
#endif

#if defined(VERBOSE_DEBUG)
#define netdev_vdbg	netdev_dbg
#else

#define netdev_vdbg(dev, format, args...)			\
({								\
	if (0)							\
		netdev_printk(KERN_DEBUG, dev, format, ##args);	\
	0;							\
})
#endif

/*
 * netdev_WARN() acts like dev_printk(), but with the key difference
 * of using a WARN/WARN_ON to get the message out, including the
 * file/line information and a backtrace.
 */
#define netdev_WARN(dev, format, args...)			\
	WARN(1, "netdevice: %s\n" format, netdev_name(dev), ##args);

2605 2606 2607 2608 2609 2610 2611 2612
/* 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)

2613 2614 2615 2616 2617 2618
#define netif_level(level, priv, type, dev, fmt, args...)	\
do {								\
	if (netif_msg_##type(priv))				\
		netdev_##level(dev, fmt, ##args);		\
} while (0)

2619
#define netif_emerg(priv, type, dev, fmt, args...)		\
2620
	netif_level(emerg, priv, type, dev, fmt, ##args)
2621
#define netif_alert(priv, type, dev, fmt, args...)		\
2622
	netif_level(alert, priv, type, dev, fmt, ##args)
2623
#define netif_crit(priv, type, dev, fmt, args...)		\
2624
	netif_level(crit, priv, type, dev, fmt, ##args)
2625
#define netif_err(priv, type, dev, fmt, args...)		\
2626
	netif_level(err, priv, type, dev, fmt, ##args)
2627
#define netif_warn(priv, type, dev, fmt, args...)		\
2628
	netif_level(warn, priv, type, dev, fmt, ##args)
2629
#define netif_notice(priv, type, dev, fmt, args...)		\
2630
	netif_level(notice, priv, type, dev, fmt, ##args)
2631
#define netif_info(priv, type, dev, fmt, args...)		\
2632
	netif_level(info, priv, type, dev, fmt, ##args)
2633 2634 2635 2636 2637 2638 2639 2640

#if defined(DEBUG)
#define netif_dbg(priv, type, dev, format, args...)		\
	netif_printk(priv, type, KERN_DEBUG, dev, format, ##args)
#elif defined(CONFIG_DYNAMIC_DEBUG)
#define netif_dbg(priv, type, netdev, format, args...)		\
do {								\
	if (netif_msg_##type(priv))				\
2641
		dynamic_netdev_dbg(netdev, format, ##args);	\
2642 2643 2644 2645 2646 2647 2648 2649 2650 2651 2652
} while (0)
#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)
2653
#define netif_vdbg	netif_dbg
2654 2655 2656 2657
#else
#define netif_vdbg(priv, type, dev, format, args...)		\
({								\
	if (0)							\
2658
		netif_printk(priv, type, KERN_DEBUG, dev, format, ##args); \
2659 2660 2661
	0;							\
})
#endif
2662

L
Linus Torvalds 已提交
2663 2664
#endif /* __KERNEL__ */

2665
#endif	/* _LINUX_NETDEVICE_H */