netdevice.h 81.8 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5 6 7 8 9
/*
 * INET		An implementation of the TCP/IP protocol suite for the LINUX
 *		operating system.  INET is implemented using the  BSD Socket
 *		interface as the means of communication with the user level.
 *
 *		Definitions for the Interfaces handler.
 *
 * Version:	@(#)dev.h	1.0.10	08/12/93
 *
10
 * Authors:	Ross Biro
L
Linus Torvalds 已提交
11 12 13
 *		Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG>
 *		Corey Minyard <wf-rch!minyard@relay.EU.net>
 *		Donald J. Becker, <becker@cesdis.gsfc.nasa.gov>
14
 *		Alan Cox, <alan@lxorguk.ukuu.org.uk>
L
Linus Torvalds 已提交
15 16 17 18 19 20 21 22 23 24 25 26 27 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
#include <net/netprio_cgroup.h>
54

55 56
#include <linux/netdev_features.h>

L
Linus Torvalds 已提交
57
struct vlan_group;
58
struct netpoll_info;
59
struct phy_device;
60 61
/* 802.11 specific */
struct wireless_dev;
L
Linus Torvalds 已提交
62 63 64 65
					/* source back-compat hooks */
#define SET_ETHTOOL_OPS(netdev,ops) \
	( (netdev)->ethtool_ops = (ops) )

66 67 68 69 70
/* 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 */

71 72 73 74
/* Backlog congestion levels */
#define NET_RX_SUCCESS		0	/* keep 'em coming, baby */
#define NET_RX_DROP		1	/* packet dropped */

75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93
/*
 * 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
94 95 96 97
#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
98

99 100 101
/* 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. */
102
#define net_xmit_eval(e)	((e) == NET_XMIT_CN ? 0 : (e))
L
Linus Torvalds 已提交
103 104
#define net_xmit_errno(e)	((e) != NET_XMIT_CN ? -ENOBUFS : 0)

105
/* Driver transmit return codes */
106
#define NETDEV_TX_MASK		0xf0
107

108
enum netdev_tx {
109
	__NETDEV_TX_MIN	 = INT_MIN,	/* make sure enum is signed */
110 111 112
	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 */
113 114 115
};
typedef enum netdev_tx netdev_tx_t;

116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133
/*
 * 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
134 135 136 137
#endif

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

138 139 140
/* Initial net device group. All devices belong to group 0 by default. */
#define INIT_NETDEV_GROUP	0

141
#ifdef  __KERNEL__
L
Linus Torvalds 已提交
142 143 144 145
/*
 *	Compute the worst case header length according to the protocols
 *	used.
 */
G
Graf Yang 已提交
146

147
#if defined(CONFIG_WLAN) || IS_ENABLED(CONFIG_AX25)
148 149 150 151 152
# if defined(CONFIG_MAC80211_MESH)
#  define LL_MAX_HEADER 128
# else
#  define LL_MAX_HEADER 96
# endif
153
#elif IS_ENABLED(CONFIG_TR)
154
# define LL_MAX_HEADER 48
L
Linus Torvalds 已提交
155
#else
156
# define LL_MAX_HEADER 32
L
Linus Torvalds 已提交
157 158
#endif

159 160
#if !IS_ENABLED(CONFIG_NET_IPIP) && !IS_ENABLED(CONFIG_NET_IPGRE) && \
    !IS_ENABLED(CONFIG_IPV6_SIT) && !IS_ENABLED(CONFIG_IPV6_TUNNEL)
L
Linus Torvalds 已提交
161 162 163 164 165 166
#define MAX_HEADER LL_MAX_HEADER
#else
#define MAX_HEADER (LL_MAX_HEADER + 48)
#endif

/*
167 168
 *	Old network device statistics. Fields are native words
 *	(unsigned long) so they can be read and written atomically.
L
Linus Torvalds 已提交
169
 */
G
Graf Yang 已提交
170

E
Eric Dumazet 已提交
171
struct net_device_stats {
172 173 174 175 176 177 178 179 180
	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
181 182
	unsigned long	collisions;
	unsigned long	rx_length_errors;
183 184 185 186 187
	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
188 189 190 191 192 193 194 195 196
	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;
};

197 198
#endif  /*  __KERNEL__  */

L
Linus Torvalds 已提交
199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215

/* 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>

216 217 218 219 220
#ifdef CONFIG_RPS
#include <linux/jump_label.h>
extern struct jump_label_key rps_needed;
#endif

L
Linus Torvalds 已提交
221 222 223 224
struct neighbour;
struct neigh_parms;
struct sk_buff;

225 226 227 228
struct netdev_hw_addr {
	struct list_head	list;
	unsigned char		addr[MAX_ADDR_LEN];
	unsigned char		type;
J
Jiri Pirko 已提交
229 230 231 232
#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
233
#define NETDEV_HW_ADDR_T_MULTICAST	5
J
Jiri Pirko 已提交
234
	bool			synced;
235
	bool			global_use;
236
	int			refcount;
237 238 239
	struct rcu_head		rcu_head;
};

240 241 242 243 244
struct netdev_hw_addr_list {
	struct list_head	list;
	int			count;
};

245 246 247 248
#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)
249

250 251 252 253
#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)
254

255 256
#define netdev_mc_count(dev) netdev_hw_addr_list_count(&(dev)->mc)
#define netdev_mc_empty(dev) netdev_hw_addr_list_empty(&(dev)->mc)
257
#define netdev_for_each_mc_addr(ha, dev) \
258
	netdev_hw_addr_list_for_each(ha, &(dev)->mc)
259

E
Eric Dumazet 已提交
260
struct hh_cache {
261
	u16		hh_len;
262
	u16		__pad;
263
	seqlock_t	hh_lock;
L
Linus Torvalds 已提交
264 265 266 267

	/* cached hardware header; allow for machine alignment needs.        */
#define HH_DATA_MOD	16
#define HH_DATA_OFF(__len) \
J
Jiri Benc 已提交
268
	(HH_DATA_MOD - (((__len - 1) & (HH_DATA_MOD - 1)) + 1))
L
Linus Torvalds 已提交
269 270 271 272 273 274 275 276 277 278 279 280 281 282
#define HH_DATA_ALIGN(__len) \
	(((__len)+(HH_DATA_MOD-1))&~(HH_DATA_MOD - 1))
	unsigned long	hh_data[HH_DATA_ALIGN(LL_MAX_HEADER) / sizeof(long)];
};

/* Reserve HH_DATA_MOD byte aligned hard_header_len, but at least that much.
 * Alternative is:
 *   dev->hard_header_len ? (dev->hard_header_len +
 *                           (HH_DATA_MOD - 1)) & ~(HH_DATA_MOD - 1) : 0
 *
 * We could use other alignment values, but we must maintain the
 * relationship HH alignment <= LL alignment.
 */
#define LL_RESERVED_SPACE(dev) \
283
	((((dev)->hard_header_len+(dev)->needed_headroom)&~(HH_DATA_MOD - 1)) + HH_DATA_MOD)
L
Linus Torvalds 已提交
284
#define LL_RESERVED_SPACE_EXTRA(dev,extra) \
285
	((((dev)->hard_header_len+(dev)->needed_headroom+(extra))&~(HH_DATA_MOD - 1)) + HH_DATA_MOD)
L
Linus Torvalds 已提交
286

287 288 289 290 291 292
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);
293
	int	(*cache)(const struct neighbour *neigh, struct hh_cache *hh, __be16 type);
294 295 296 297 298
	void	(*cache_update)(struct hh_cache *hh,
				const struct net_device *dev,
				const unsigned char *haddr);
};

L
Linus Torvalds 已提交
299 300 301 302 303
/* 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
304
enum netdev_state_t {
L
Linus Torvalds 已提交
305 306 307
	__LINK_STATE_START,
	__LINK_STATE_PRESENT,
	__LINK_STATE_NOCARRIER,
S
Stefan Rompf 已提交
308 309
	__LINK_STATE_LINKWATCH_PENDING,
	__LINK_STATE_DORMANT,
L
Linus Torvalds 已提交
310 311 312 313 314
};


/*
 * This structure holds at boot time configured netdevice settings. They
G
Graf Yang 已提交
315
 * are then used in the device probing.
L
Linus Torvalds 已提交
316 317 318 319 320 321 322
 */
struct netdev_boot_setup {
	char name[IFNAMSIZ];
	struct ifmap map;
};
#define NETDEV_BOOT_SETUP_MAX 8

323
extern int __init netdev_boot_setup(char *str);
L
Linus Torvalds 已提交
324

325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343
/*
 * 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
344 345 346

	unsigned int		gro_count;

H
Herbert Xu 已提交
347
	struct net_device	*dev;
348 349
	struct list_head	dev_list;
	struct sk_buff		*gro_list;
H
Herbert Xu 已提交
350
	struct sk_buff		*skb;
351 352
};

E
Eric Dumazet 已提交
353
enum {
354
	NAPI_STATE_SCHED,	/* Poll is scheduled */
D
David S. Miller 已提交
355
	NAPI_STATE_DISABLE,	/* Disable pending */
356
	NAPI_STATE_NPSVC,	/* Netpoll - don't dequeue from poll_list */
357 358
};

359
enum gro_result {
360 361 362 363 364 365
	GRO_MERGED,
	GRO_MERGED_FREE,
	GRO_HELD,
	GRO_NORMAL,
	GRO_DROP,
};
366
typedef enum gro_result gro_result_t;
367

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
/*
 * 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);
417

418
extern void __napi_schedule(struct napi_struct *n);
419

D
David S. Miller 已提交
420 421 422 423 424
static inline int napi_disable_pending(struct napi_struct *n)
{
	return test_bit(NAPI_STATE_DISABLE, &n->state);
}

425 426 427 428 429 430
/**
 *	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
431 432
 * insure only one NAPI poll instance runs.  We also make
 * sure there is no pending NAPI disable.
433 434 435
 */
static inline int napi_schedule_prep(struct napi_struct *n)
{
D
David S. Miller 已提交
436 437
	return !napi_disable_pending(n) &&
		!test_and_set_bit(NAPI_STATE_SCHED, &n->state);
438 439 440 441 442 443 444 445 446 447 448 449 450 451 452
}

/**
 *	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);
}

453 454 455 456 457 458 459 460 461 462
/* 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;
}

463 464 465 466 467 468
/**
 *	napi_complete - NAPI processing complete
 *	@n: napi context
 *
 * Mark NAPI processing as complete.
 */
469 470
extern void __napi_complete(struct napi_struct *n);
extern void napi_complete(struct napi_struct *n);
471 472 473 474 475 476 477 478 479 480

/**
 *	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
481
	set_bit(NAPI_STATE_DISABLE, &n->state);
482
	while (test_and_set_bit(NAPI_STATE_SCHED, &n->state))
483
		msleep(1);
D
David S. Miller 已提交
484
	clear_bit(NAPI_STATE_DISABLE, &n->state);
485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500
}

/**
 *	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);
}

501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518
#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
519
enum netdev_queue_state_t {
520 521
	__QUEUE_STATE_DRV_XOFF,
	__QUEUE_STATE_STACK_XOFF,
522
	__QUEUE_STATE_FROZEN,
523 524 525 526
#define QUEUE_STATE_ANY_XOFF ((1 << __QUEUE_STATE_DRV_XOFF)		| \
			      (1 << __QUEUE_STATE_STACK_XOFF))
#define QUEUE_STATE_ANY_XOFF_OR_FROZEN (QUEUE_STATE_ANY_XOFF		| \
					(1 << __QUEUE_STATE_FROZEN))
527
};
528 529 530 531 532 533 534 535 536
/*
 * __QUEUE_STATE_DRV_XOFF is used by drivers to stop the transmit queue.  The
 * netif_tx_* functions below are used to manipulate this flag.  The
 * __QUEUE_STATE_STACK_XOFF flag is used by the stack to stop the transmit
 * queue independently.  The netif_xmit_*stopped functions below are called
 * to check if the queue has been stopped by the driver or stack (either
 * of the XOFF bits are set in the state).  Drivers should not need to call
 * netif_xmit*stopped functions, they should only be using netif_tx_*.
 */
537

538
struct netdev_queue {
539 540 541
/*
 * read mostly part
 */
542
	struct net_device	*dev;
543
	struct Qdisc		*qdisc;
544
	unsigned long		state;
545
	struct Qdisc		*qdisc_sleeping;
546
#ifdef CONFIG_SYSFS
T
Tom Herbert 已提交
547 548
	struct kobject		kobj;
#endif
549 550 551
#if defined(CONFIG_XPS) && defined(CONFIG_NUMA)
	int			numa_node;
#endif
552 553 554 555 556
/*
 * write mostly part
 */
	spinlock_t		_xmit_lock ____cacheline_aligned_in_smp;
	int			xmit_lock_owner;
557 558 559 560
	/*
	 * please use this field instead of dev->trans_start
	 */
	unsigned long		trans_start;
561 562 563 564 565 566

	/*
	 * Number of TX timeouts for this queue
	 * (/sys/class/net/DEV/Q/trans_timeout)
	 */
	unsigned long		trans_timeout;
567
} ____cacheline_aligned_in_smp;
568

569 570 571 572 573
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
574
	return NUMA_NO_NODE;
575 576 577 578 579 580 581 582 583 584
#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
585
#ifdef CONFIG_RPS
T
Tom Herbert 已提交
586 587 588 589 590 591 592 593 594 595 596
/*
 * 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
597
/*
598 599 600
 * 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
601 602 603
 */
struct rps_dev_flow {
	u16 cpu;
604
	u16 filter;
T
Tom Herbert 已提交
605 606
	unsigned int last_qtail;
};
607
#define RPS_NO_FILTER 0xffff
T
Tom Herbert 已提交
608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654

/*
 * 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
655
extern struct rps_sock_flow_table __rcu *rps_sock_flow_table;
T
Tom Herbert 已提交
656

657 658 659 660 661
#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
662 663
/* This structure contains an instance of an RX queue. */
struct netdev_rx_queue {
E
Eric Dumazet 已提交
664 665 666
	struct rps_map __rcu		*rps_map;
	struct rps_dev_flow_table __rcu	*rps_flow_table;
	struct kobject			kobj;
T
Tom Herbert 已提交
667
	struct net_device		*dev;
T
Tom Herbert 已提交
668
} ____cacheline_aligned_in_smp;
T
Tom Herbert 已提交
669
#endif /* CONFIG_RPS */
670

T
Tom Herbert 已提交
671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690
#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
691
	struct xps_map __rcu *cpu_map[0];
T
Tom Herbert 已提交
692 693 694 695 696
};
#define XPS_DEV_MAPS_SIZE (sizeof(struct xps_dev_maps) +		\
    (nr_cpu_ids * sizeof(struct xps_map *)))
#endif /* CONFIG_XPS */

697 698 699 700 701 702 703 704
#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;
};

705 706
/*
 * This structure defines the management hooks for network devices.
707 708
 * The following hooks can be defined; unless noted otherwise, they are
 * optional and can be filled with a null pointer.
709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727
 *
 * 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.
 *
728 729
 * netdev_tx_t (*ndo_start_xmit)(struct sk_buff *skb,
 *                               struct net_device *dev);
730
 *	Called when a packet needs to be transmitted.
731 732
 *	Must return NETDEV_TX_OK , NETDEV_TX_BUSY.
 *        (can also return NETDEV_TX_LOCKED iff NETIF_F_LLTX)
733 734 735 736 737 738
 *	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.
 *
739 740 741 742 743 744
 * 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.
745 746
 *	If driver handles unicast address filtering, it should set
 *	IFF_UNICAST_FLT to its priv_flags.
747 748 749
 *
 * int (*ndo_set_mac_address)(struct net_device *dev, void *addr);
 *	This function  is called when the Media Access Control address
750
 *	needs to be changed. If this interface is not defined, the
751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770
 *	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.
 *
771
 * void (*ndo_tx_timeout)(struct net_device *dev);
772 773 774
 *	Callback uses when the transmitter has not made any progress
 *	for dev->watchdog ticks.
 *
775
 * struct rtnl_link_stats64* (*ndo_get_stats64)(struct net_device *dev,
776
 *                      struct rtnl_link_stats64 *storage);
777
 * struct net_device_stats* (*ndo_get_stats)(struct net_device *dev);
778
 *	Called when a user wants to get the network device usage
779
 *	statistics. Drivers must do one of the following:
780 781
 *	1. Define @ndo_get_stats64 to fill in a zero-initialised
 *	   rtnl_link_stats64 structure passed by the caller.
782
 *	2. Define @ndo_get_stats to update a net_device_stats structure
783 784 785 786 787
 *	   (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.
788 789 790 791 792 793 794 795 796 797
 *
 * 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);
798 799 800 801 802
 *
 *	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);
803
 * int (*ndo_set_vf_spoofchk)(struct net_device *dev, int vf, bool setting);
804 805
 * int (*ndo_get_vf_config)(struct net_device *dev,
 *			    int vf, struct ifla_vf_info *ivf);
806 807 808
 * 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);
809 810 811 812 813
 * 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.
814
 *
815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850
 *	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.
 *
851 852 853 854 855 856
 *	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.
857 858 859 860 861 862 863 864
 *
 *	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.
865 866
 *
 *      Feature/offload setting functions.
867 868
 * netdev_features_t (*ndo_fix_features)(struct net_device *dev,
 *		netdev_features_t features);
869 870 871 872
 *	Adjusts the requested feature flags according to device-specific
 *	constraints, and returns the resulting flags. Must not modify
 *	the device state.
 *
873
 * int (*ndo_set_features)(struct net_device *dev, netdev_features_t features);
874 875 876 877
 *	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.
 *
878 879 880 881 882 883
 */
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);
884
	netdev_tx_t		(*ndo_start_xmit) (struct sk_buff *skb,
885 886 887
						   struct net_device *dev);
	u16			(*ndo_select_queue)(struct net_device *dev,
						    struct sk_buff *skb);
888 889 890 891 892 893 894 895 896 897
	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);
898 899 900 901
	int			(*ndo_change_mtu)(struct net_device *dev,
						  int new_mtu);
	int			(*ndo_neigh_setup)(struct net_device *dev,
						   struct neigh_parms *);
902 903
	void			(*ndo_tx_timeout) (struct net_device *dev);

904 905
	struct rtnl_link_stats64* (*ndo_get_stats64)(struct net_device *dev,
						     struct rtnl_link_stats64 *storage);
906 907 908 909 910 911 912 913
	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
914 915
	int			(*ndo_netpoll_setup)(struct net_device *dev,
						     struct netpoll_info *info);
916
	void			(*ndo_netpoll_cleanup)(struct net_device *dev);
917
#endif
918 919 920 921 922 923
	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);
924 925
	int			(*ndo_set_vf_spoofchk)(struct net_device *dev,
						       int vf, bool setting);
926 927 928
	int			(*ndo_get_vf_config)(struct net_device *dev,
						     int vf,
						     struct ifla_vf_info *ivf);
929 930 931 932 933
	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);
934
	int			(*ndo_setup_tc)(struct net_device *dev, u8 tc);
935
#if IS_ENABLED(CONFIG_FCOE)
936 937
	int			(*ndo_fcoe_enable)(struct net_device *dev);
	int			(*ndo_fcoe_disable)(struct net_device *dev);
938 939 940 941 942 943
	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);
944 945 946 947
	int			(*ndo_fcoe_ddp_target)(struct net_device *dev,
						       u16 xid,
						       struct scatterlist *sgl,
						       unsigned int sgc);
948 949
#endif

950
#if IS_ENABLED(CONFIG_LIBFCOE)
951 952 953 954
#define NETDEV_FCOE_WWNN 0
#define NETDEV_FCOE_WWPN 1
	int			(*ndo_fcoe_get_wwn)(struct net_device *dev,
						    u64 *wwn, int type);
955
#endif
956

957 958 959 960 961 962
#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
963 964 965 966
	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);
967 968
	netdev_features_t	(*ndo_fix_features)(struct net_device *dev,
						    netdev_features_t features);
969
	int			(*ndo_set_features)(struct net_device *dev,
970
						    netdev_features_t features);
971 972
};

L
Linus Torvalds 已提交
973 974 975 976 977 978 979 980 981 982
/*
 *	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
983
struct net_device {
L
Linus Torvalds 已提交
984 985 986 987

	/*
	 * 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
988
	 * of the interface.
L
Linus Torvalds 已提交
989 990
	 */
	char			name[IFNAMSIZ];
M
Mark Gross 已提交
991

J
Jean Pihet 已提交
992
	struct pm_qos_request	pm_qos_req;
M
Mark Gross 已提交
993

994 995
	/* device name hash chain */
	struct hlist_node	name_hlist;
996 997
	/* snmp alias */
	char 			*ifalias;
L
Linus Torvalds 已提交
998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014

	/*
	 *	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;

1015
	struct list_head	dev_list;
1016
	struct list_head	napi_list;
1017
	struct list_head	unreg_list;
L
Linus Torvalds 已提交
1018

1019
	/* currently active device features */
1020
	netdev_features_t	features;
1021
	/* user-changeable features */
1022
	netdev_features_t	hw_features;
1023
	/* user-requested features */
1024
	netdev_features_t	wanted_features;
1025
	/* mask of features inheritable by VLAN devices */
1026
	netdev_features_t	vlan_features;
1027

L
Linus Torvalds 已提交
1028 1029 1030 1031
	/* Interface index. Unique device identifier	*/
	int			ifindex;
	int			iflink;

R
Rusty Russell 已提交
1032
	struct net_device_stats	stats;
1033 1034 1035
	atomic_long_t		rx_dropped; /* dropped packets by core network
					     * Do not use this in drivers.
					     */
L
Linus Torvalds 已提交
1036

1037
#ifdef CONFIG_WIRELESS_EXT
L
Linus Torvalds 已提交
1038 1039 1040 1041 1042
	/* 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;
1043
#endif
1044 1045
	/* Management operations */
	const struct net_device_ops *netdev_ops;
1046
	const struct ethtool_ops *ethtool_ops;
L
Linus Torvalds 已提交
1047

1048 1049 1050
	/* Hardware header description */
	const struct header_ops *header_ops;

S
Stefan Rompf 已提交
1051
	unsigned int		flags;	/* interface flags (a la BSD)	*/
1052
	unsigned int		priv_flags; /* Like 'flags' but invisible to userspace. */
L
Linus Torvalds 已提交
1053 1054 1055
	unsigned short		gflags;
	unsigned short		padded;	/* How much padding added by alloc_netdev() */

S
Stefan Rompf 已提交
1056 1057 1058
	unsigned char		operstate; /* RFC2863 operstate */
	unsigned char		link_mode; /* mapping policy to operstate */

1059 1060 1061
	unsigned char		if_port;	/* Selectable AUI, TP,..*/
	unsigned char		dma;		/* DMA channel		*/

1062
	unsigned int		mtu;	/* interface MTU value		*/
L
Linus Torvalds 已提交
1063 1064 1065
	unsigned short		type;	/* interface hardware type	*/
	unsigned short		hard_header_len;	/* hardware hdr length	*/

1066 1067 1068 1069 1070 1071 1072
	/* 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
1073
	/* Interface address info. */
1074
	unsigned char		perm_addr[MAX_ADDR_LEN]; /* permanent hw address */
1075
	unsigned char		addr_assign_type; /* hw address assignment type */
L
Linus Torvalds 已提交
1076 1077 1078
	unsigned char		addr_len;	/* hardware address length	*/
	unsigned short          dev_id;		/* for shared network cards */

J
Jiri Pirko 已提交
1079
	spinlock_t		addr_list_lock;
1080 1081
	struct netdev_hw_addr_list	uc;	/* Unicast mac addresses */
	struct netdev_hw_addr_list	mc;	/* Multicast mac addresses */
1082
	bool			uc_promisc;
1083 1084
	unsigned int		promiscuity;
	unsigned int		allmulti;
L
Linus Torvalds 已提交
1085 1086 1087


	/* Protocol specific pointers */
1088

1089
#if IS_ENABLED(CONFIG_VLAN_8021Q)
E
Eric Dumazet 已提交
1090
	struct vlan_group __rcu	*vlgrp;		/* VLAN group */
1091
#endif
1092
#if IS_ENABLED(CONFIG_NET_DSA)
1093
	struct dsa_switch_tree	*dsa_ptr;	/* dsa specific data */
1094
#endif
L
Linus Torvalds 已提交
1095
	void 			*atalk_ptr;	/* AppleTalk link 	*/
E
Eric Dumazet 已提交
1096
	struct in_device __rcu	*ip_ptr;	/* IPv4 specific data	*/
1097
	struct dn_dev __rcu     *dn_ptr;        /* DECnet specific data */
E
Eric Dumazet 已提交
1098
	struct inet6_dev __rcu	*ip6_ptr;       /* IPv6 specific data */
L
Linus Torvalds 已提交
1099 1100
	void			*ec_ptr;	/* Econet specific data	*/
	void			*ax25_ptr;	/* AX.25 specific data */
1101 1102
	struct wireless_dev	*ieee80211_ptr;	/* IEEE 802.11 specific data,
						   assign before registering */
L
Linus Torvalds 已提交
1103

1104
/*
E
Eric Dumazet 已提交
1105
 * Cache lines mostly used on receive path (including eth_type_trans())
1106
 */
1107 1108 1109 1110 1111 1112 1113 1114
	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
1115 1116 1117 1118
	struct net_device	*master; /* Pointer to master device of a group,
					  * which this device is member of.
					  */

1119
	/* Interface address info used in eth_type_trans() */
1120 1121 1122 1123
	unsigned char		*dev_addr;	/* hw address, (before bcast
						   because most packets are
						   unicast) */

1124 1125
	struct netdev_hw_addr_list	dev_addrs; /* list of device
						      hw addresses */
1126 1127

	unsigned char		broadcast[MAX_ADDR_LEN];	/* hw bcast add	*/
L
Linus Torvalds 已提交
1128

1129
#ifdef CONFIG_SYSFS
T
Tom Herbert 已提交
1130
	struct kset		*queues_kset;
1131
#endif
T
Tom Herbert 已提交
1132

1133
#ifdef CONFIG_RPS
T
Tom Herbert 已提交
1134 1135
	struct netdev_rx_queue	*_rx;

1136
	/* Number of RX queues allocated at register_netdev() time */
T
Tom Herbert 已提交
1137
	unsigned int		num_rx_queues;
1138 1139 1140

	/* Number of RX queues currently active in device */
	unsigned int		real_num_rx_queues;
1141 1142 1143 1144 1145 1146 1147

#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
1148
#endif
T
Tom Herbert 已提交
1149

1150 1151
	rx_handler_func_t __rcu	*rx_handler;
	void __rcu		*rx_handler_data;
1152

1153
	struct netdev_queue __rcu *ingress_queue;
E
Eric Dumazet 已提交
1154 1155 1156 1157

/*
 * Cache lines mostly used on transmit path
 */
1158
	struct netdev_queue	*_tx ____cacheline_aligned_in_smp;
1159 1160

	/* Number of TX queues allocated at alloc_netdev_mq() time  */
1161
	unsigned int		num_tx_queues;
1162 1163 1164 1165

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

1166 1167 1168
	/* root qdisc from userspace point of view */
	struct Qdisc		*qdisc;

L
Linus Torvalds 已提交
1169
	unsigned long		tx_queue_len;	/* Max frames per queue allowed */
1170
	spinlock_t		tx_global_lock;
E
Eric Dumazet 已提交
1171

T
Tom Herbert 已提交
1172
#ifdef CONFIG_XPS
E
Eric Dumazet 已提交
1173
	struct xps_dev_maps __rcu *xps_maps;
T
Tom Herbert 已提交
1174
#endif
T
Tom Herbert 已提交
1175

1176
	/* These may be needed for future network-power-down code. */
1177 1178 1179 1180 1181

	/*
	 * trans_start here is expensive for high speed devices on SMP,
	 * please use netdev_queue->trans_start instead.
	 */
1182 1183 1184 1185 1186
	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
1187
	/* Number of references to this device */
E
Eric Dumazet 已提交
1188
	int __percpu		*pcpu_refcnt;
1189

L
Linus Torvalds 已提交
1190 1191 1192 1193 1194
	/* delayed register/unregister */
	struct list_head	todo_list;
	/* device index hash chain */
	struct hlist_node	index_hlist;

1195
	struct list_head	link_watch_list;
1196

L
Linus Torvalds 已提交
1197 1198
	/* register/unregister state machine */
	enum { NETREG_UNINITIALIZED=0,
1199
	       NETREG_REGISTERED,	/* completed register_netdevice */
L
Linus Torvalds 已提交
1200 1201 1202
	       NETREG_UNREGISTERING,	/* called unregister_netdevice */
	       NETREG_UNREGISTERED,	/* completed unregister todo */
	       NETREG_RELEASED,		/* called free_netdev */
1203
	       NETREG_DUMMY,		/* dummy device for NAPI poll */
1204 1205 1206
	} reg_state:8;

	bool dismantle; /* device is going do be freed */
1207 1208 1209 1210 1211

	enum {
		RTNL_LINK_INITIALIZED,
		RTNL_LINK_INITIALIZING,
	} rtnl_link_state:16;
L
Linus Torvalds 已提交
1212

1213 1214
	/* Called from unregister, can be used to call free_netdev */
	void (*destructor)(struct net_device *dev);
L
Linus Torvalds 已提交
1215 1216

#ifdef CONFIG_NETPOLL
1217
	struct netpoll_info	*npinfo;
L
Linus Torvalds 已提交
1218
#endif
1219

1220
#ifdef CONFIG_NET_NS
1221 1222
	/* Network namespace this network device is inside */
	struct net		*nd_net;
1223
#endif
1224

D
David S. Miller 已提交
1225
	/* mid-layer private */
E
Eric Dumazet 已提交
1226 1227 1228
	union {
		void				*ml_priv;
		struct pcpu_lstats __percpu	*lstats; /* loopback stats */
1229
		struct pcpu_tstats __percpu	*tstats; /* tunnel stats */
1230
		struct pcpu_dstats __percpu	*dstats; /* dummy stats */
E
Eric Dumazet 已提交
1231
	};
1232
	/* GARP */
E
Eric Dumazet 已提交
1233
	struct garp_port __rcu	*garp_port;
L
Linus Torvalds 已提交
1234 1235

	/* class/net/name entry */
1236
	struct device		dev;
1237 1238
	/* space for optional device, statistics, and wireless sysfs groups */
	const struct attribute_group *sysfs_groups[4];
P
Patrick McHardy 已提交
1239 1240 1241

	/* rtnetlink link ops */
	const struct rtnl_link_ops *rtnl_link_ops;
1242

1243 1244 1245
	/* for setting kernel sock attribute on TCP connection setup */
#define GSO_MAX_SIZE		65536
	unsigned int		gso_max_size;
1246

J
Jeff Kirsher 已提交
1247
#ifdef CONFIG_DCB
1248
	/* Data Center Bridging netlink ops */
1249
	const struct dcbnl_rtnl_ops *dcbnl_ops;
1250
#endif
1251 1252 1253
	u8 num_tc;
	struct netdev_tc_txq tc_to_txq[TC_MAX_QUEUE];
	u8 prio_tc_map[TC_BITMASK + 1];
1254

1255
#if IS_ENABLED(CONFIG_FCOE)
1256 1257
	/* max exchange id for FCoE LRO by ddp */
	unsigned int		fcoe_ddp_xid;
1258 1259 1260
#endif
#if IS_ENABLED(CONFIG_NETPRIO_CGROUP)
	struct netprio_map __rcu *priomap;
1261
#endif
1262 1263
	/* phy device may attach itself for hardware timestamping */
	struct phy_device *phydev;
1264 1265 1266

	/* group the device belongs to */
	int group;
L
Linus Torvalds 已提交
1267
};
1268
#define to_net_dev(d) container_of(d, struct net_device, dev)
L
Linus Torvalds 已提交
1269 1270 1271

#define	NETDEV_ALIGN		32

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 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322
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;
}

1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341
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);
}

1342 1343 1344 1345 1346 1347
/*
 * Net namespace inlines
 */
static inline
struct net *dev_net(const struct net_device *dev)
{
E
Eric Dumazet 已提交
1348
	return read_pnet(&dev->nd_net);
1349 1350 1351
}

static inline
1352
void dev_net_set(struct net_device *dev, struct net *net)
1353 1354
{
#ifdef CONFIG_NET_NS
1355 1356
	release_net(dev->nd_net);
	dev->nd_net = hold_net(net);
1357 1358 1359
#endif
}

1360 1361 1362 1363 1364 1365 1366 1367 1368 1369
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;
}

1370 1371 1372 1373 1374 1375 1376 1377 1378
#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

1379 1380 1381 1382 1383 1384 1385 1386 1387 1388
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;
}

1389 1390 1391 1392 1393 1394
/**
 *	netdev_priv - access network device private data
 *	@dev: network device
 *
 * Get network device private data
 */
1395
static inline void *netdev_priv(const struct net_device *dev)
L
Linus Torvalds 已提交
1396
{
1397
	return (char *)dev + ALIGN(sizeof(struct net_device), NETDEV_ALIGN);
L
Linus Torvalds 已提交
1398 1399 1400 1401 1402
}

/* Set the sysfs physical device reference for the network logical device
 * if set prior to registration will cause a symlink during initialization.
 */
1403
#define SET_NETDEV_DEV(net, pdev)	((net)->dev.parent = (pdev))
L
Linus Torvalds 已提交
1404

1405 1406 1407 1408 1409 1410
/* 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))

1411 1412 1413 1414 1415 1416 1417 1418 1419 1420
/**
 *	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.
 */
1421 1422
void netif_napi_add(struct net_device *dev, struct napi_struct *napi,
		    int (*poll)(struct napi_struct *, int), int weight);
1423

1424 1425 1426 1427 1428 1429
/**
 *  netif_napi_del - remove a napi context
 *  @napi: napi context
 *
 *  netif_napi_del() removes a napi context from the network device napi list
 */
1430 1431 1432
void netif_napi_del(struct napi_struct *napi);

struct napi_gro_cb {
1433 1434 1435
	/* Virtual address of skb_shinfo(skb)->frags[0].page + offset. */
	void *frag0;

1436 1437 1438
	/* Length of frag0. */
	unsigned int frag0_len;

1439 1440 1441
	/* This indicates where we are processing relative to skb->data. */
	int data_offset;

1442 1443 1444 1445 1446 1447 1448 1449
	/* 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
1450 1451 1452

	/* Free the skb? */
	int free;
1453 1454 1455
};

#define NAPI_GRO_CB(skb) ((struct napi_gro_cb *)(skb)->cb)
1456

L
Linus Torvalds 已提交
1457
struct packet_type {
D
David S. Miller 已提交
1458 1459 1460 1461 1462 1463
	__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 *);
1464
	struct sk_buff		*(*gso_segment)(struct sk_buff *skb,
1465
						netdev_features_t features);
1466
	int			(*gso_send_check)(struct sk_buff *skb);
1467 1468 1469
	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
1470 1471 1472 1473 1474 1475
	void			*af_packet_priv;
	struct list_head	list;
};

#include <linux/notifier.h>

1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508
/* 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
1509 1510
extern rwlock_t				dev_base_lock;		/* Device list lock */

1511

1512 1513
#define for_each_netdev(net, d)		\
		list_for_each_entry(d, &(net)->dev_base_head, dev_list)
1514 1515
#define for_each_netdev_reverse(net, d)	\
		list_for_each_entry_reverse(d, &(net)->dev_base_head, dev_list)
1516 1517
#define for_each_netdev_rcu(net, d)		\
		list_for_each_entry_rcu(d, &(net)->dev_base_head, dev_list)
1518 1519 1520 1521
#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)
1522 1523
#define for_each_netdev_continue_rcu(net, d)		\
	list_for_each_entry_continue_rcu(d, &(net)->dev_base_head, dev_list)
1524
#define net_device_entry(lh)	list_entry(lh, struct net_device, dev_list)
1525

1526 1527 1528 1529 1530
static inline struct net_device *next_net_device(struct net_device *dev)
{
	struct list_head *lh;
	struct net *net;

1531
	net = dev_net(dev);
1532 1533 1534 1535
	lh = dev->dev_list.next;
	return lh == &net->dev_base_head ? NULL : net_device_entry(lh);
}

1536 1537 1538 1539 1540 1541
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
1542
	lh = rcu_dereference(list_next_rcu(&dev->dev_list));
1543 1544 1545
	return lh == &net->dev_base_head ? NULL : net_device_entry(lh);
}

1546 1547 1548 1549 1550
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);
}
1551

E
Eric Dumazet 已提交
1552 1553 1554 1555 1556 1557 1558
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
1559 1560
extern int 			netdev_boot_setup_check(struct net_device *dev);
extern unsigned long		netdev_boot_base(const char *prefix, int unit);
1561 1562
extern struct net_device *dev_getbyhwaddr_rcu(struct net *net, unsigned short type,
					      const char *hwaddr);
1563 1564
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
1565 1566 1567 1568
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
1569 1570
extern struct net_device	*dev_get_by_flags_rcu(struct net *net, unsigned short flags,
						      unsigned short mask);
1571
extern struct net_device	*dev_get_by_name(struct net *net, const char *name);
1572
extern struct net_device	*dev_get_by_name_rcu(struct net *net, const char *name);
1573
extern struct net_device	*__dev_get_by_name(struct net *net, const char *name);
L
Linus Torvalds 已提交
1574 1575 1576
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);
1577
extern void		dev_disable_lro(struct net_device *dev);
L
Linus Torvalds 已提交
1578 1579
extern int		dev_queue_xmit(struct sk_buff *skb);
extern int		register_netdevice(struct net_device *dev);
1580 1581
extern void		unregister_netdevice_queue(struct net_device *dev,
						   struct list_head *head);
1582
extern void		unregister_netdevice_many(struct list_head *head);
1583 1584 1585 1586 1587
static inline void unregister_netdevice(struct net_device *dev)
{
	unregister_netdevice_queue(dev, NULL);
}

E
Eric Dumazet 已提交
1588
extern int 		netdev_refcnt_read(const struct net_device *dev);
L
Linus Torvalds 已提交
1589 1590
extern void		free_netdev(struct net_device *dev);
extern void		synchronize_net(void);
1591
extern int		init_dummy_netdev(struct net_device *dev);
1592
extern void		netdev_resync_ops(struct net_device *dev);
1593

1594 1595
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);
1596
extern struct net_device	*dev_get_by_index_rcu(struct net *net, int ifindex);
L
Linus Torvalds 已提交
1597 1598 1599 1600
extern int		dev_restart(struct net_device *dev);
#ifdef CONFIG_NETPOLL_TRAP
extern int		netpoll_trap(void);
#endif
1601 1602
extern int	       skb_gro_receive(struct sk_buff **head,
				       struct sk_buff *skb);
1603
extern void	       skb_gro_reset_offset(struct sk_buff *skb);
1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619

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;
}

1620 1621
static inline void *skb_gro_header_fast(struct sk_buff *skb,
					unsigned int offset)
1622
{
1623 1624
	return NAPI_GRO_CB(skb)->frag0 + offset;
}
1625

1626 1627 1628 1629
static inline int skb_gro_header_hard(struct sk_buff *skb, unsigned int hlen)
{
	return NAPI_GRO_CB(skb)->frag0_len < hlen;
}
1630

1631 1632 1633
static inline void *skb_gro_header_slow(struct sk_buff *skb, unsigned int hlen,
					unsigned int offset)
{
1634 1635 1636
	if (!pskb_may_pull(skb, hlen))
		return NULL;

1637 1638
	NAPI_GRO_CB(skb)->frag0 = NULL;
	NAPI_GRO_CB(skb)->frag0_len = 0;
1639
	return skb->data + offset;
1640
}
L
Linus Torvalds 已提交
1641

1642 1643
static inline void *skb_gro_mac_header(struct sk_buff *skb)
{
1644
	return NAPI_GRO_CB(skb)->frag0 ?: skb_mac_header(skb);
1645 1646
}

H
Herbert Xu 已提交
1647 1648
static inline void *skb_gro_network_header(struct sk_buff *skb)
{
1649 1650
	return (NAPI_GRO_CB(skb)->frag0 ?: skb->data) +
	       skb_network_offset(skb);
H
Herbert Xu 已提交
1651 1652
}

1653 1654
static inline int dev_hard_header(struct sk_buff *skb, struct net_device *dev,
				  unsigned short type,
1655 1656
				  const void *daddr, const void *saddr,
				  unsigned len)
1657
{
1658
	if (!dev->header_ops || !dev->header_ops->create)
1659
		return 0;
1660 1661

	return dev->header_ops->create(skb, dev, type, daddr, saddr, len);
1662 1663
}

S
Stephen Hemminger 已提交
1664 1665 1666 1667 1668
static inline int dev_parse_header(const struct sk_buff *skb,
				   unsigned char *haddr)
{
	const struct net_device *dev = skb->dev;

1669
	if (!dev->header_ops || !dev->header_ops->parse)
S
Stephen Hemminger 已提交
1670
		return 0;
1671
	return dev->header_ops->parse(skb, haddr);
S
Stephen Hemminger 已提交
1672 1673
}

L
Linus Torvalds 已提交
1674 1675 1676 1677 1678 1679 1680 1681
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
1682
 * Incoming packets are placed on per-cpu queues
L
Linus Torvalds 已提交
1683
 */
E
Eric Dumazet 已提交
1684
struct softnet_data {
1685
	struct Qdisc		*output_queue;
1686
	struct Qdisc		**output_queue_tailp;
L
Linus Torvalds 已提交
1687 1688
	struct list_head	poll_list;
	struct sk_buff		*completion_queue;
1689
	struct sk_buff_head	process_queue;
L
Linus Torvalds 已提交
1690

C
Changli Gao 已提交
1691
	/* stats */
1692 1693 1694 1695
	unsigned int		processed;
	unsigned int		time_squeeze;
	unsigned int		cpu_collision;
	unsigned int		received_rps;
C
Changli Gao 已提交
1696

1697
#ifdef CONFIG_RPS
E
Eric Dumazet 已提交
1698 1699 1700
	struct softnet_data	*rps_ipi_list;

	/* Elements below can be accessed between CPUs for RPS */
T
Tom Herbert 已提交
1701
	struct call_single_data	csd ____cacheline_aligned_in_smp;
E
Eric Dumazet 已提交
1702 1703
	struct softnet_data	*rps_ipi_next;
	unsigned int		cpu;
T
Tom Herbert 已提交
1704
	unsigned int		input_queue_head;
1705
	unsigned int		input_queue_tail;
1706
#endif
C
Changli Gao 已提交
1707
	unsigned		dropped;
T
Tom Herbert 已提交
1708
	struct sk_buff_head	input_pkt_queue;
1709
	struct napi_struct	backlog;
L
Linus Torvalds 已提交
1710 1711
};

1712
static inline void input_queue_head_incr(struct softnet_data *sd)
T
Tom Herbert 已提交
1713 1714
{
#ifdef CONFIG_RPS
1715 1716 1717 1718 1719 1720 1721 1722 1723
	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
1724 1725 1726
#endif
}

T
Tom Herbert 已提交
1727
DECLARE_PER_CPU_ALIGNED(struct softnet_data, softnet_data);
L
Linus Torvalds 已提交
1728

1729
extern void __netif_schedule(struct Qdisc *q);
L
Linus Torvalds 已提交
1730

1731
static inline void netif_schedule_queue(struct netdev_queue *txq)
L
Linus Torvalds 已提交
1732
{
1733
	if (!(txq->state & QUEUE_STATE_ANY_XOFF))
1734
		__netif_schedule(txq->qdisc);
1735 1736
}

1737 1738 1739 1740 1741 1742 1743 1744
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));
}

1745 1746
static inline void netif_tx_start_queue(struct netdev_queue *dev_queue)
{
1747
	clear_bit(__QUEUE_STATE_DRV_XOFF, &dev_queue->state);
1748 1749
}

1750 1751 1752 1753 1754 1755
/**
 *	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
1756 1757
static inline void netif_start_queue(struct net_device *dev)
{
1758
	netif_tx_start_queue(netdev_get_tx_queue(dev, 0));
L
Linus Torvalds 已提交
1759 1760
}

1761 1762 1763 1764 1765 1766 1767 1768 1769 1770
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);
	}
}

1771
static inline void netif_tx_wake_queue(struct netdev_queue *dev_queue)
L
Linus Torvalds 已提交
1772 1773
{
#ifdef CONFIG_NETPOLL_TRAP
1774
	if (netpoll_trap()) {
1775
		netif_tx_start_queue(dev_queue);
L
Linus Torvalds 已提交
1776
		return;
1777
	}
L
Linus Torvalds 已提交
1778
#endif
1779
	if (test_and_clear_bit(__QUEUE_STATE_DRV_XOFF, &dev_queue->state))
1780
		__netif_schedule(dev_queue->qdisc);
1781 1782
}

1783 1784 1785 1786 1787 1788 1789
/**
 *	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.
 */
1790 1791
static inline void netif_wake_queue(struct net_device *dev)
{
1792
	netif_tx_wake_queue(netdev_get_tx_queue(dev, 0));
L
Linus Torvalds 已提交
1793 1794
}

1795 1796 1797 1798 1799 1800 1801 1802 1803 1804
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);
	}
}

1805 1806
static inline void netif_tx_stop_queue(struct netdev_queue *dev_queue)
{
1807
	if (WARN_ON(!dev_queue)) {
1808
		pr_info("netif_stop_queue() cannot be called before register_netdev()\n");
1809 1810
		return;
	}
1811
	set_bit(__QUEUE_STATE_DRV_XOFF, &dev_queue->state);
1812 1813
}

1814 1815 1816 1817 1818 1819 1820
/**
 *	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
1821 1822
static inline void netif_stop_queue(struct net_device *dev)
{
1823
	netif_tx_stop_queue(netdev_get_tx_queue(dev, 0));
L
Linus Torvalds 已提交
1824 1825
}

1826 1827 1828 1829 1830 1831 1832 1833 1834 1835
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);
	}
}

1836 1837
static inline int netif_tx_queue_stopped(const struct netdev_queue *dev_queue)
{
1838
	return test_bit(__QUEUE_STATE_DRV_XOFF, &dev_queue->state);
1839 1840
}

1841 1842 1843 1844 1845 1846
/**
 *	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
1847 1848
static inline int netif_queue_stopped(const struct net_device *dev)
{
1849
	return netif_tx_queue_stopped(netdev_get_tx_queue(dev, 0));
L
Linus Torvalds 已提交
1850 1851
}

1852
static inline int netif_xmit_stopped(const struct netdev_queue *dev_queue)
1853
{
1854 1855 1856 1857 1858 1859 1860 1861
	return dev_queue->state & QUEUE_STATE_ANY_XOFF;
}

static inline int netif_xmit_frozen_or_stopped(const struct netdev_queue *dev_queue)
{
	return dev_queue->state & QUEUE_STATE_ANY_XOFF_OR_FROZEN;
}

1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889
static inline void netdev_tx_sent_queue(struct netdev_queue *dev_queue,
					unsigned int bytes)
{
}

static inline void netdev_sent_queue(struct net_device *dev, unsigned int bytes)
{
	netdev_tx_sent_queue(netdev_get_tx_queue(dev, 0), bytes);
}

static inline void netdev_tx_completed_queue(struct netdev_queue *dev_queue,
					     unsigned pkts, unsigned bytes)
{
}

static inline void netdev_completed_queue(struct net_device *dev,
					  unsigned pkts, unsigned bytes)
{
	netdev_tx_completed_queue(netdev_get_tx_queue(dev, 0), pkts, bytes);
}

static inline void netdev_tx_reset_queue(struct netdev_queue *q)
{
}

static inline void netdev_reset_queue(struct net_device *dev_queue)
{
	netdev_tx_reset_queue(netdev_get_tx_queue(dev_queue, 0));
1890 1891
}

1892 1893 1894 1895 1896 1897
/**
 *	netif_running - test if up
 *	@dev: network device
 *
 *	Test if the device has been brought up.
 */
L
Linus Torvalds 已提交
1898 1899 1900 1901 1902
static inline int netif_running(const struct net_device *dev)
{
	return test_bit(__LINK_STATE_START, &dev->state);
}

1903 1904 1905 1906 1907 1908
/*
 * 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.
 */
1909 1910 1911 1912 1913 1914 1915 1916

/**
 *	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.
 */
1917 1918
static inline void netif_start_subqueue(struct net_device *dev, u16 queue_index)
{
1919
	struct netdev_queue *txq = netdev_get_tx_queue(dev, queue_index);
1920 1921

	netif_tx_start_queue(txq);
1922 1923
}

1924 1925 1926 1927 1928 1929 1930
/**
 *	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.
 */
1931 1932
static inline void netif_stop_subqueue(struct net_device *dev, u16 queue_index)
{
1933
	struct netdev_queue *txq = netdev_get_tx_queue(dev, queue_index);
1934 1935 1936 1937
#ifdef CONFIG_NETPOLL_TRAP
	if (netpoll_trap())
		return;
#endif
1938
	netif_tx_stop_queue(txq);
1939 1940
}

1941 1942 1943 1944 1945 1946 1947
/**
 *	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.
 */
1948
static inline int __netif_subqueue_stopped(const struct net_device *dev,
1949 1950
					 u16 queue_index)
{
1951
	struct netdev_queue *txq = netdev_get_tx_queue(dev, queue_index);
1952 1953

	return netif_tx_queue_stopped(txq);
1954 1955
}

1956 1957 1958 1959 1960
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));
}
1961 1962 1963 1964 1965 1966 1967 1968

/**
 *	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.
 */
1969 1970
static inline void netif_wake_subqueue(struct net_device *dev, u16 queue_index)
{
1971
	struct netdev_queue *txq = netdev_get_tx_queue(dev, queue_index);
1972 1973 1974 1975
#ifdef CONFIG_NETPOLL_TRAP
	if (netpoll_trap())
		return;
#endif
1976
	if (test_and_clear_bit(__QUEUE_STATE_DRV_XOFF, &txq->state))
1977
		__netif_schedule(txq->qdisc);
1978 1979
}

1980 1981 1982 1983 1984 1985 1986 1987 1988 1989
/*
 * 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);
}

1990 1991 1992 1993 1994 1995
/**
 *	netif_is_multiqueue - test if device has multiple transmit queues
 *	@dev: network device
 *
 * Check if device has multiple transmit queues
 */
1996 1997
static inline int netif_is_multiqueue(const struct net_device *dev)
{
E
Eric Dumazet 已提交
1998
	return dev->num_tx_queues > 1;
1999
}
L
Linus Torvalds 已提交
2000

2001 2002
extern int netif_set_real_num_tx_queues(struct net_device *dev,
					unsigned int txq);
2003

2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014
#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

2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026
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
2027
/* Use this variant when it is known for sure that it
2028 2029
 * is executing from hardware interrupt context or with hardware interrupts
 * disabled.
L
Linus Torvalds 已提交
2030
 */
2031
extern void dev_kfree_skb_irq(struct sk_buff *skb);
L
Linus Torvalds 已提交
2032 2033

/* Use this variant in places where it could be invoked
2034 2035
 * from either hardware interrupt or other context, with hardware interrupts
 * either disabled or enabled.
L
Linus Torvalds 已提交
2036
 */
2037
extern void dev_kfree_skb_any(struct sk_buff *skb);
L
Linus Torvalds 已提交
2038 2039 2040 2041

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);
2042
extern gro_result_t	dev_gro_receive(struct napi_struct *napi,
2043
					struct sk_buff *skb);
2044 2045
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,
2046
					 struct sk_buff *skb);
E
Eric Dumazet 已提交
2047
extern void		napi_gro_flush(struct napi_struct *napi);
2048
extern struct sk_buff *	napi_get_frags(struct napi_struct *napi);
2049
extern gro_result_t	napi_frags_finish(struct napi_struct *napi,
2050 2051
					  struct sk_buff *skb,
					  gro_result_t ret);
2052
extern struct sk_buff *	napi_frags_skb(struct napi_struct *napi);
2053
extern gro_result_t	napi_gro_frags(struct napi_struct *napi);
2054 2055 2056 2057 2058 2059 2060

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

2061
extern int netdev_rx_handler_register(struct net_device *dev,
J
Jiri Pirko 已提交
2062 2063
				      rx_handler_func_t *rx_handler,
				      void *rx_handler_data);
2064 2065
extern void netdev_rx_handler_unregister(struct net_device *dev);

2066
extern int		dev_valid_name(const char *name);
2067 2068
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
2069
extern unsigned		dev_get_flags(const struct net_device *);
2070
extern int		__dev_change_flags(struct net_device *, unsigned int flags);
L
Linus Torvalds 已提交
2071
extern int		dev_change_flags(struct net_device *, unsigned);
2072
extern void		__dev_notify_flags(struct net_device *, unsigned int old_flags);
2073
extern int		dev_change_name(struct net_device *, const char *);
2074
extern int		dev_set_alias(struct net_device *, const char *, size_t);
2075 2076
extern int		dev_change_net_namespace(struct net_device *,
						 struct net *, const char *);
L
Linus Torvalds 已提交
2077
extern int		dev_set_mtu(struct net_device *, int);
2078
extern void		dev_set_group(struct net_device *, int);
L
Linus Torvalds 已提交
2079 2080
extern int		dev_set_mac_address(struct net_device *,
					    struct sockaddr *);
2081
extern int		dev_hard_start_xmit(struct sk_buff *skb,
2082 2083
					    struct net_device *dev,
					    struct netdev_queue *txq);
2084 2085
extern int		dev_forward_skb(struct net_device *dev,
					struct sk_buff *skb);
L
Linus Torvalds 已提交
2086

2087
extern int		netdev_budget;
L
Linus Torvalds 已提交
2088 2089 2090 2091

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

2092 2093 2094 2095
/**
 *	dev_put - release reference to device
 *	@dev: network device
 *
2096
 * Release reference to device to allow it to be freed.
2097
 */
L
Linus Torvalds 已提交
2098 2099
static inline void dev_put(struct net_device *dev)
{
E
Eric Dumazet 已提交
2100
	irqsafe_cpu_dec(*dev->pcpu_refcnt);
L
Linus Torvalds 已提交
2101 2102
}

2103 2104 2105 2106
/**
 *	dev_hold - get reference to device
 *	@dev: network device
 *
2107
 * Hold reference to device to keep it from being freed.
2108
 */
2109 2110
static inline void dev_hold(struct net_device *dev)
{
E
Eric Dumazet 已提交
2111
	irqsafe_cpu_inc(*dev->pcpu_refcnt);
2112
}
L
Linus Torvalds 已提交
2113 2114 2115 2116

/* 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
2117 2118 2119 2120
 *
 * 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
2121 2122 2123
 */

extern void linkwatch_fire_event(struct net_device *dev);
2124
extern void linkwatch_forget_dev(struct net_device *dev);
L
Linus Torvalds 已提交
2125

2126 2127 2128 2129 2130 2131
/**
 *	netif_carrier_ok - test if carrier present
 *	@dev: network device
 *
 * Check if carrier is present on device
 */
L
Linus Torvalds 已提交
2132 2133 2134 2135 2136
static inline int netif_carrier_ok(const struct net_device *dev)
{
	return !test_bit(__LINK_STATE_NOCARRIER, &dev->state);
}

2137 2138
extern unsigned long dev_trans_start(struct net_device *dev);

L
Linus Torvalds 已提交
2139 2140
extern void __netdev_watchdog_up(struct net_device *dev);

2141
extern void netif_carrier_on(struct net_device *dev);
L
Linus Torvalds 已提交
2142

2143
extern void netif_carrier_off(struct net_device *dev);
L
Linus Torvalds 已提交
2144

2145 2146
extern void netif_notify_peers(struct net_device *dev);

2147 2148 2149 2150 2151 2152 2153 2154 2155 2156 2157 2158 2159
/**
 *	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
2160 2161 2162 2163 2164 2165
static inline void netif_dormant_on(struct net_device *dev)
{
	if (!test_and_set_bit(__LINK_STATE_DORMANT, &dev->state))
		linkwatch_fire_event(dev);
}

2166 2167 2168 2169 2170 2171
/**
 *	netif_dormant_off - set device as not dormant.
 *	@dev: network device
 *
 * Device is not in dormant state.
 */
S
Stefan Rompf 已提交
2172 2173 2174 2175 2176 2177
static inline void netif_dormant_off(struct net_device *dev)
{
	if (test_and_clear_bit(__LINK_STATE_DORMANT, &dev->state))
		linkwatch_fire_event(dev);
}

2178 2179 2180 2181 2182 2183
/**
 *	netif_dormant - test if carrier present
 *	@dev: network device
 *
 * Check if carrier is present on device
 */
S
Stefan Rompf 已提交
2184 2185 2186 2187 2188 2189
static inline int netif_dormant(const struct net_device *dev)
{
	return test_bit(__LINK_STATE_DORMANT, &dev->state);
}


2190 2191 2192 2193 2194 2195
/**
 *	netif_oper_up - test if device is operational
 *	@dev: network device
 *
 * Check if carrier is operational
 */
E
Eric Dumazet 已提交
2196 2197
static inline int netif_oper_up(const struct net_device *dev)
{
S
Stefan Rompf 已提交
2198 2199 2200 2201
	return (dev->operstate == IF_OPER_UP ||
		dev->operstate == IF_OPER_UNKNOWN /* backward compat */);
}

2202 2203 2204 2205 2206 2207
/**
 *	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
2208 2209 2210 2211 2212
static inline int netif_device_present(struct net_device *dev)
{
	return test_bit(__LINK_STATE_PRESENT, &dev->state);
}

2213
extern void netif_device_detach(struct net_device *dev);
L
Linus Torvalds 已提交
2214

2215
extern void netif_device_attach(struct net_device *dev);
L
Linus Torvalds 已提交
2216 2217 2218 2219 2220 2221 2222 2223 2224 2225 2226 2227 2228 2229 2230 2231 2232 2233 2234 2235 2236 2237 2238 2239 2240 2241 2242 2243 2244 2245 2246 2247 2248 2249 2250 2251 2252 2253 2254 2255 2256 2257 2258 2259 2260 2261 2262 2263 2264 2265

/*
 * 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;
}

2266
static inline void __netif_tx_lock(struct netdev_queue *txq, int cpu)
H
Herbert Xu 已提交
2267
{
2268 2269
	spin_lock(&txq->_xmit_lock);
	txq->xmit_lock_owner = cpu;
2270 2271
}

2272 2273 2274 2275 2276 2277
static inline void __netif_tx_lock_bh(struct netdev_queue *txq)
{
	spin_lock_bh(&txq->_xmit_lock);
	txq->xmit_lock_owner = smp_processor_id();
}

2278 2279 2280 2281 2282 2283 2284 2285 2286 2287 2288 2289 2290 2291 2292 2293 2294 2295 2296 2297
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
2298 2299 2300 2301 2302 2303
static inline void txq_trans_update(struct netdev_queue *txq)
{
	if (txq->xmit_lock_owner != -1)
		txq->trans_start = jiffies;
}

2304 2305 2306 2307 2308 2309
/**
 *	netif_tx_lock - grab network device transmit lock
 *	@dev: network device
 *
 * Get network device transmit lock
 */
2310 2311
static inline void netif_tx_lock(struct net_device *dev)
{
2312
	unsigned int i;
2313
	int cpu;
2314

2315 2316
	spin_lock(&dev->tx_global_lock);
	cpu = smp_processor_id();
2317 2318
	for (i = 0; i < dev->num_tx_queues; i++) {
		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
2319 2320 2321 2322 2323 2324 2325

		/* 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.
		 */
2326
		__netif_tx_lock(txq, cpu);
2327 2328
		set_bit(__QUEUE_STATE_FROZEN, &txq->state);
		__netif_tx_unlock(txq);
2329
	}
H
Herbert Xu 已提交
2330 2331 2332 2333
}

static inline void netif_tx_lock_bh(struct net_device *dev)
{
2334 2335
	local_bh_disable();
	netif_tx_lock(dev);
H
Herbert Xu 已提交
2336 2337 2338 2339
}

static inline void netif_tx_unlock(struct net_device *dev)
{
2340 2341 2342 2343
	unsigned int i;

	for (i = 0; i < dev->num_tx_queues; i++) {
		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
2344

2345 2346 2347 2348 2349
		/* 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);
2350
		netif_schedule_queue(txq);
2351 2352
	}
	spin_unlock(&dev->tx_global_lock);
H
Herbert Xu 已提交
2353 2354 2355 2356
}

static inline void netif_tx_unlock_bh(struct net_device *dev)
{
2357 2358
	netif_tx_unlock(dev);
	local_bh_enable();
H
Herbert Xu 已提交
2359 2360
}

2361
#define HARD_TX_LOCK(dev, txq, cpu) {			\
2362
	if ((dev->features & NETIF_F_LLTX) == 0) {	\
2363
		__netif_tx_lock(txq, cpu);		\
2364 2365 2366
	}						\
}

2367
#define HARD_TX_UNLOCK(dev, txq) {			\
2368
	if ((dev->features & NETIF_F_LLTX) == 0) {	\
2369
		__netif_tx_unlock(txq);			\
2370 2371 2372
	}						\
}

L
Linus Torvalds 已提交
2373 2374
static inline void netif_tx_disable(struct net_device *dev)
{
2375
	unsigned int i;
2376
	int cpu;
2377

2378 2379
	local_bh_disable();
	cpu = smp_processor_id();
2380 2381
	for (i = 0; i < dev->num_tx_queues; i++) {
		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
2382 2383

		__netif_tx_lock(txq, cpu);
2384
		netif_tx_stop_queue(txq);
2385
		__netif_tx_unlock(txq);
2386
	}
2387
	local_bh_enable();
L
Linus Torvalds 已提交
2388 2389
}

2390 2391 2392 2393 2394 2395 2396 2397 2398 2399 2400 2401 2402 2403 2404 2405 2406 2407 2408 2409
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);
}

2410
/*
2411
 * dev_addrs walker. Should be used only for read access. Call with
2412 2413 2414
 * rcu_read_lock held.
 */
#define for_each_dev_addr(dev, ha) \
2415
		list_for_each_entry_rcu(ha, &dev->dev_addrs.list, list)
2416

L
Linus Torvalds 已提交
2417 2418 2419 2420 2421
/* 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
2422
extern struct net_device *alloc_netdev_mqs(int sizeof_priv, const char *name,
2423
				       void (*setup)(struct net_device *),
T
Tom Herbert 已提交
2424
				       unsigned int txqs, unsigned int rxqs);
2425
#define alloc_netdev(sizeof_priv, name, setup) \
T
Tom Herbert 已提交
2426 2427 2428 2429 2430
	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
2431 2432
extern int		register_netdev(struct net_device *dev);
extern void		unregister_netdev(struct net_device *dev);
2433

2434 2435 2436 2437 2438 2439 2440 2441 2442 2443 2444 2445 2446 2447 2448 2449
/* 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);

2450 2451 2452 2453 2454 2455 2456 2457 2458 2459 2460
/* 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);
2461 2462 2463 2464 2465 2466 2467 2468 2469 2470
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);
2471

2472 2473 2474 2475 2476 2477 2478 2479 2480
/* 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);
2481

2482 2483 2484
/* 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);
2485 2486
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
2487
extern void		netdev_state_change(struct net_device *dev);
2488
extern int		netdev_bonding_change(struct net_device *dev,
2489
					      unsigned long event);
2490
extern void		netdev_features_change(struct net_device *dev);
L
Linus Torvalds 已提交
2491
/* Load a device via the kmod */
2492
extern void		dev_load(struct net *net, const char *name);
L
Linus Torvalds 已提交
2493
extern void		dev_mcast_init(void);
2494 2495
extern struct rtnl_link_stats64 *dev_get_stats(struct net_device *dev,
					       struct rtnl_link_stats64 *storage);
2496

L
Linus Torvalds 已提交
2497
extern int		netdev_max_backlog;
E
Eric Dumazet 已提交
2498
extern int		netdev_tstamp_prequeue;
L
Linus Torvalds 已提交
2499
extern int		weight_p;
2500
extern int		bpf_jit_enable;
L
Linus Torvalds 已提交
2501
extern int		netdev_set_master(struct net_device *dev, struct net_device *master);
J
Jiri Pirko 已提交
2502 2503
extern int netdev_set_bond_master(struct net_device *dev,
				  struct net_device *master);
2504
extern int skb_checksum_help(struct sk_buff *skb);
2505 2506
extern struct sk_buff *skb_gso_segment(struct sk_buff *skb,
	netdev_features_t features);
2507 2508 2509 2510 2511 2512 2513
#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
2514 2515 2516 2517
/* rx skb timestamps */
extern void		net_enable_timestamp(void);
extern void		net_disable_timestamp(void);

2518 2519 2520 2521 2522 2523
#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

2524 2525 2526
extern int netdev_class_create_file(struct class_attribute *class_attr);
extern void netdev_class_remove_file(struct class_attribute *class_attr);

2527 2528
extern struct kobj_ns_type_operations net_ns_type_operations;

2529
extern const char *netdev_drivername(const struct net_device *dev);
2530

2531 2532
extern void linkwatch_run_queue(void);

2533 2534
static inline netdev_features_t netdev_get_wanted_features(
	struct net_device *dev)
2535 2536 2537
{
	return (dev->features & ~dev->hw_features) | dev->wanted_features;
}
2538 2539
netdev_features_t netdev_increment_features(netdev_features_t all,
	netdev_features_t one, netdev_features_t mask);
2540
int __netdev_update_features(struct net_device *dev);
2541
void netdev_update_features(struct net_device *dev);
2542
void netdev_change_features(struct net_device *dev);
2543

2544 2545 2546
void netif_stacked_transfer_operstate(const struct net_device *rootdev,
					struct net_device *dev);

2547
netdev_features_t netif_skb_features(struct sk_buff *skb);
2548

2549
static inline int net_gso_ok(netdev_features_t features, int gso_type)
2550
{
2551
	netdev_features_t feature = gso_type << NETIF_F_GSO_SHIFT;
2552 2553 2554 2555 2556 2557 2558 2559 2560

	/* 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));

2561
	return (features & feature) == feature;
2562 2563
}

2564
static inline int skb_gso_ok(struct sk_buff *skb, netdev_features_t features)
2565
{
H
Herbert Xu 已提交
2566
	return net_gso_ok(features, skb_shinfo(skb)->gso_type) &&
2567
	       (!skb_has_frag_list(skb) || (features & NETIF_F_FRAGLIST));
2568 2569
}

2570 2571
static inline int netif_needs_gso(struct sk_buff *skb,
	netdev_features_t features)
2572
{
2573 2574
	return skb_is_gso(skb) && (!skb_gso_ok(skb, features) ||
		unlikely(skb->ip_summed != CHECKSUM_PARTIAL));
2575 2576
}

2577 2578 2579 2580 2581 2582
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
2583 2584 2585 2586 2587
static inline int netif_is_bond_slave(struct net_device *dev)
{
	return dev->flags & IFF_SLAVE && dev->priv_flags & IFF_BONDING;
}

2588
extern struct pernet_operations __net_initdata loopback_net_ops;
2589

2590 2591 2592 2593 2594 2595 2596 2597 2598 2599 2600
/* 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;
}

2601 2602 2603
extern int __netdev_printk(const char *level, const struct net_device *dev,
			struct va_format *vaf);

2604 2605 2606 2607 2608 2609 2610 2611 2612 2613 2614 2615 2616 2617 2618 2619 2620
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, ...);
2621

2622 2623 2624
#define MODULE_ALIAS_NETDEV(device) \
	MODULE_ALIAS("netdev-" device)

2625 2626 2627 2628 2629 2630
#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 {								\
2631
	dynamic_netdev_dbg(__dev, format, ##args);		\
2632 2633 2634 2635 2636 2637 2638 2639 2640 2641 2642 2643 2644 2645 2646 2647 2648 2649 2650 2651 2652 2653 2654 2655 2656 2657 2658 2659 2660 2661
} 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);

2662 2663 2664 2665 2666 2667 2668 2669
/* 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)

2670 2671 2672 2673 2674 2675
#define netif_level(level, priv, type, dev, fmt, args...)	\
do {								\
	if (netif_msg_##type(priv))				\
		netdev_##level(dev, fmt, ##args);		\
} while (0)

2676
#define netif_emerg(priv, type, dev, fmt, args...)		\
2677
	netif_level(emerg, priv, type, dev, fmt, ##args)
2678
#define netif_alert(priv, type, dev, fmt, args...)		\
2679
	netif_level(alert, priv, type, dev, fmt, ##args)
2680
#define netif_crit(priv, type, dev, fmt, args...)		\
2681
	netif_level(crit, priv, type, dev, fmt, ##args)
2682
#define netif_err(priv, type, dev, fmt, args...)		\
2683
	netif_level(err, priv, type, dev, fmt, ##args)
2684
#define netif_warn(priv, type, dev, fmt, args...)		\
2685
	netif_level(warn, priv, type, dev, fmt, ##args)
2686
#define netif_notice(priv, type, dev, fmt, args...)		\
2687
	netif_level(notice, priv, type, dev, fmt, ##args)
2688
#define netif_info(priv, type, dev, fmt, args...)		\
2689
	netif_level(info, priv, type, dev, fmt, ##args)
2690 2691 2692 2693 2694 2695 2696 2697

#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))				\
2698
		dynamic_netdev_dbg(netdev, format, ##args);	\
2699 2700 2701 2702 2703 2704 2705 2706 2707 2708 2709
} 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)
2710
#define netif_vdbg	netif_dbg
2711 2712 2713 2714
#else
#define netif_vdbg(priv, type, dev, format, args...)		\
({								\
	if (0)							\
2715
		netif_printk(priv, type, KERN_DEBUG, dev, format, ##args); \
2716 2717 2718
	0;							\
})
#endif
2719

L
Linus Torvalds 已提交
2720 2721
#endif /* __KERNEL__ */

2722
#endif	/* _LINUX_NETDEVICE_H */